-mm1
--- diff/CREDITS 2004-03-16 11:24:01.000000000 +0000
+++ source/CREDITS 2004-03-16 11:27:47.976894064 +0000
@@ -82,13 +82,13 @@ S: Howell, NJ 07731
S: USA
N: Erik Andersen
-E: andersee@debian.org
-W: http://www.xmission.com/~andersen
-P: 1024/FC4CFFED 78 3C 6A 19 FA 5D 92 5A FB AC 7B A5 A5 E1 FF 8E
+E: andersen@codepoet.org
+W: http://www.codepoet.org/
+P: 1024D/30D39057 1BC4 2742 E885 E4DE 9301 0C82 5F9B 643E 30D3 9057
D: Maintainer of ide-cd and Uniform CD-ROM driver,
D: ATAPI CD-Changer support, Major 2.1.x CD-ROM update.
-S: 4538 South Carnegie Tech Street
-S: Salt Lake City, Utah 84120
+S: 352 North 525 East
+S: Springville, Utah 84663
S: USA
N: Michael Ang
@@ -289,6 +289,15 @@ S: Via Delle Palme, 9
S: Terni 05100
S: Italy
+N: Krzysztof Benedyczak
+E: golbi@mat.uni.torun.pl
+W: http://www.mat.uni.torun.pl/~golbi
+D: POSIX message queues fs (with M. Wronski)
+S: ul. Podmiejska 52
+S: Radunica
+S: 83-000 Pruszcz Gdanski
+S: Poland
+
N: Randolph Bentson
E: bentson@grieg.seaslug.org
W: http://www.aa.net/~bentson/
@@ -673,11 +682,6 @@ S: Northampton
S: NN1 3QT
S: United Kingdom
-N: Stephane Dalton
-E: sdalton@videotron.ca
-D: Tieman Voyager USB Braille display driver.
-S: Qu�bec, Canada
-
N: Uwe Dannowski
E: Uwe.Dannowski@ira.uka.de
W: http://i30www.ira.uka.de/~dannowsk/
@@ -797,11 +801,6 @@ E: cort@fsmlabs.com
W: http://www.fsmlabs.com/linuxppcbk.html
D: PowerPC
-N: St�phane Doyon
-E: s.doyon@videotron.ca
-D: Tieman Voyager USB Braille display driver.
-S: Qu�bec, Canada
-
N: Oleg Drokin
E: green@ccssu.crimea.ua
W: http://www.ccssu.crimea.ua/~green
@@ -3488,6 +3487,14 @@ S: 12725 SW Millikan Way, Suite 400
S: Beaverton, OR 97005
S: USA
+N: Michal Wronski
+E: wrona@mat.uni.torun.pl
+W: http://www.mat.uni.torun.pl/~wrona
+D: POSIX message queues fs (with K. Benedyczak)
+S: ul. Teczowa 23/12
+S: 80-680 Gdansk-Sobieszewo
+S: Poland
+
N: Frank Xia
E: qx@math.columbia.edu
D: Xiafs filesystem [defunct]
--- diff/Documentation/DocBook/Makefile 2003-10-27 09:20:43.000000000 +0000
+++ source/Documentation/DocBook/Makefile 2004-03-16 11:27:47.978893760 +0000
@@ -47,7 +47,7 @@ installmandocs: mandocs
###
#External programs used
KERNELDOC = scripts/kernel-doc
-DOCPROC = scripts/docproc
+DOCPROC = scripts/basic/docproc
SPLITMAN = $(PERL) $(srctree)/scripts/split-man
MAKEMAN = $(PERL) $(srctree)/scripts/makeman
@@ -149,12 +149,18 @@ quiet_cmd_fig2eps = FIG2EPS $@
cmd_fig2eps = fig2dev -Leps $< $@
%.eps: %.fig
+ @(which fig2dev > /dev/null 2>&1) || \
+ (echo "*** You need to install transfig ***"; \
+ exit 1)
$(call cmd,fig2eps)
quiet_cmd_fig2png = FIG2PNG $@
cmd_fig2png = fig2dev -Lpng $< $@
%.png: %.fig
+ @(which fig2dev > /dev/null 2>&1) || \
+ (echo "*** You need to install transfig ***"; \
+ exit 1)
$(call cmd,fig2png)
###
--- diff/Documentation/DocBook/gadget.tmpl 2003-09-30 14:46:10.000000000 +0000
+++ source/Documentation/DocBook/gadget.tmpl 2004-03-16 11:27:47.977893912 +0000
@@ -454,6 +454,7 @@ but some optional utilities are provided
</para>
!Edrivers/usb/gadget/usbstring.c
+!Edrivers/usb/gadget/config.c
</sect1>
</chapter>
--- diff/Documentation/DocBook/parportbook.tmpl 2003-05-21 10:49:45.000000000 +0000
+++ source/Documentation/DocBook/parportbook.tmpl 2004-03-16 11:27:47.980893456 +0000
@@ -2730,7 +2730,7 @@ to permit their use in free software.
</book>
<!-- Additional function to be documented:
-!Ddrivers/parport/init.c
+--! Ddrivers/parport/init.c (this file doesn't exist any more)
-->
<!-- Local Variables: -->
<!-- sgml-indent-step: 1 -->
--- diff/Documentation/IPMI.txt 2003-05-21 10:49:59.000000000 +0000
+++ source/Documentation/IPMI.txt 2004-03-16 11:27:47.993891480 +0000
@@ -22,6 +22,58 @@ are not familiar with IPMI itself, see t
http://www.intel.com/design/servers/ipmi/index.htm. IPMI is a big
subject and I can't cover it all here!
+Configuration
+-------------
+
+The LinuxIPMI driver is modular, which means you have to pick several
+things to have it work right depending on your hardware. Most of
+these are available in the 'Character Devices' menu.
+
+No matter what, you must pick 'IPMI top-level message handler' to use
+IPMI. What you do beyond that depends on your needs and hardware.
+
+The message handler does not provide any user-level interfaces.
+Kernel code (like the watchdog) can still use it. If you need access
+from userland, you need to select 'Device interface for IPMI' if you
+want access through a device driver. Another interface is also
+available, you may select 'IPMI sockets' in the 'Networking Support'
+main menu. This provides a socket interface to IPMI. You may select
+both of these at the same time, they will both work together.
+
+The driver interface depends on your hardware. If you have a board
+with a standard interface (These will generally be either "KCS",
+"SMIC", or "BT", consult your hardware manual), choose the 'IPMI SI
+handler' option. A driver also exists for direct I2C access to the
+IPMI management controller. Some boards support this, but it is
+unknown if it will work on every board. For this, choose 'IPMI SMBus
+handler', but be ready to try to do some figuring to see if it will
+work.
+
+There is also a KCS-only driver interface supplied, but it is
+depracated in favor of the SI interface.
+
+You should generally enable ACPI on your system, as systems with IPMI
+should have ACPI tables describing them.
+
+If you have a standard interface and the board manufacturer has done
+their job correctly, the IPMI controller should be automatically
+detect (via ACPI or SMBIOS tables) and should just work. Sadly, many
+boards do not have this information. The driver attempts standard
+defaults, but they may not work. If you fall into this situation, you
+need to read the section below named 'The SI Driver' on how to
+hand-configure your system.
+
+IPMI defines a standard watchdog timer. You can enable this with the
+'IPMI Watchdog Timer' config option. If you compile the driver into
+the kernel, then via a kernel command-line option you can have the
+watchdog timer start as soon as it intitializes. It also have a lot
+of other options, see the 'Watchdog' section below for more details.
+Note that you can also have the watchdog continue to run if it is
+closed (by default it is disabled on close). Go into the 'Watchdog
+Cards' menu, enable 'Watchdog Timer Support', and enable the option
+'Disable watchdog shutdown on close'.
+
+
Basic Design
------------
@@ -41,18 +93,30 @@ ipmi_devintf - This provides a userland
driver, each open file for this device ties in to the message handler
as an IPMI user.
-ipmi_kcs_drv - A driver for the KCS SMI. Most system have a KCS
-interface for IPMI.
+ipmi_si - A driver for various system interfaces. This supports
+KCS, SMIC, and may support BT in the future. Unless you have your own
+custom interface, you probably need to use this.
+
+ipmi_smb - A driver for accessing BMCs on the SMBus. It uses the
+I2C kernel driver's SMBus interfaces to send and receive IPMI messages
+over the SMBus.
+af_ipmi - A network socket interface to IPMI. This doesn't take up
+a character device in your system.
+
+Note that the KCS-only interface ahs been removed.
Much documentation for the interface is in the include files. The
IPMI include files are:
-ipmi.h - Contains the user interface and IOCTL interface for IPMI.
+net/af_ipmi.h - Contains the socket interface.
+
+linux/ipmi.h - Contains the user interface and IOCTL interface for IPMI.
-ipmi_smi.h - Contains the interface for SMI drivers to use.
+linux/ipmi_smi.h - Contains the interface for system management interfaces
+(things that interface to IPMI controllers) to use.
-ipmi_msgdefs.h - General definitions for base IPMI messaging.
+linux/ipmi_msgdefs.h - General definitions for base IPMI messaging.
Addressing
@@ -260,70 +324,131 @@ they register with the message handler.
in the order they register, although if an SMI unregisters and then
another one registers, all bets are off.
-The ipmi_smi.h defines the interface for SMIs, see that for more
-details.
+The ipmi_smi.h defines the interface for management interfaces, see
+that for more details.
-The KCS Driver
---------------
+The SI Driver
+-------------
-The KCS driver allows up to 4 KCS interfaces to be configured in the
-system. By default, the driver will register one KCS interface at the
-spec-specified I/O port 0xca2 without interrupts. You can change this
-at module load time (for a module) with:
+The SI driver allows up to 4 KCS or SMIC interfaces to be configured
+in the system. By default, scan the ACPI tables for interfaces, and
+if it doesn't find any the driver will attempt to register one KCS
+interface at the spec-specified I/O port 0xca2 without interrupts.
+You can change this at module load time (for a module) with:
- insmod ipmi_kcs_drv.o kcs_ports=<port1>,<port2>... kcs_addrs=<addr1>,<addr2>
- kcs_irqs=<irq1>,<irq2>... kcs_trydefaults=[0|1]
+ modprobe ipmi_si.o type=<type1>,<type2>....
+ ports=<port1>,<port2>... addrs=<addr1>,<addr2>...
+ irqs=<irq1>,<irq2>... trydefaults=[0|1]
-The KCS driver supports two types of interfaces, ports (for I/O port
-based KCS interfaces) and memory addresses (for KCS interfaces in
-memory). The driver will support both of them simultaneously, setting
-the port to zero (or just not specifying it) will allow the memory
-address to be used. The port will override the memory address if it
-is specified and non-zero. kcs_trydefaults sets whether the standard
-IPMI interface at 0xca2 and any interfaces specified by ACPE are
-tried. By default, the driver tries it, set this value to zero to
-turn this off.
+Each of these except si_trydefaults is a list, the first item for the
+first interface, second item for the second interface, etc.
+
+The si_type may be either "kcs", "smic", or "bt". If you leave it blank, it
+defaults to "kcs".
+
+If you specify si_addrs as non-zero for an interface, the driver will
+use the memory address given as the address of the device. This
+overrides si_ports.
+
+If you specify si_ports as non-zero for an interface, the driver will
+use the I/O port given as the device address.
+
+If you specify si_irqs as non-zero for an interface, the driver will
+attempt to use the given interrupt for the device.
+
+si_trydefaults sets whether the standard IPMI interface at 0xca2 and
+any interfaces specified by ACPE are tried. By default, the driver
+tries it, set this value to zero to turn this off.
When compiled into the kernel, the addresses can be specified on the
kernel command line as:
- ipmi_kcs=<bmc1>:<irq1>,<bmc2>:<irq2>....,[nodefault]
-
-The <bmcx> values is either "p<port>" or "m<addr>" for port or memory
-addresses. So for instance, a KCS interface at port 0xca2 using
-interrupt 9 and a memory interface at address 0xf9827341 with no
-interrupt would be specified "ipmi_kcs=p0xca2:9,m0xf9827341".
-If you specify zero for in irq or don't specify it, the driver will
-run polled unless the software can detect the interrupt to use in the
-ACPI tables.
-
-By default, the driver will attempt to detect a KCS device at the
-spec-specified 0xca2 address and any address specified by ACPI. If
-you want to turn this off, use the "nodefault" option.
+ ipmi_si.type=<type1>,<type2>...
+ ipmi_si.ports=<port1>,<port2>... ipmi_si.addrs=<addr1>,<addr2>...
+ ipmi_si.irqs=<irq1>,<irq2>... ipmi_si.trydefaults=[0|1]
+
+It works the same as the module parameters of the same names.
+
+By default, the driver will attempt to detect any device specified by
+ACPI, and if none of those then a KCS device at the spec-specified
+0xca2. If you want to turn this off, set the "trydefaults" option to
+false.
If you have high-res timers compiled into the kernel, the driver will
use them to provide much better performance. Note that if you do not
have high-res timers enabled in the kernel and you don't have
interrupts enabled, the driver will run VERY slowly. Don't blame me,
-the KCS interface sucks.
+these interfaces suck.
+
+
+The SMBus Driver
+----------------
+
+The SMBus driver allows up to 4 SMBus devices to be configured in the
+system. By default, the driver will register any SMBus interfaces it finds
+in the I2C address range of 0x20 to 0x4f on any adapter. You can change this
+at module load time (for a module) with:
+
+ modprobe ipmi_smb.o
+ addr=<adapter1>,<i2caddr1>[,<adapter2>,<i2caddr2>[,...]]
+ dbg=<flags1>,<flags2>...
+ [defaultprobe=0] [dbg_probe=1]
+
+The addresses are specified in pairs, the first is the adapter ID and the
+second is the I2C address on that adapter.
+
+The debug flags are bit flags for each BMC found, they are:
+IPMI messages: 1, driver state: 2, timing: 4, I2C probe: 8
+
+Setting smb_defaultprobe to zero disabled the default probing of SMBus
+interfaces at address range 0x20 to 0x4f. This means that only the
+BMCs specified on the smb_addr line will be detected.
+
+Setting smb_dbg_probe to 1 will enable debugging of the probing and
+detection process for BMCs on the SMBusses.
+
+Discovering the IPMI compilant BMC on the SMBus can cause devices
+on the I2C bus to fail. The SMBus driver writes a "Get Device ID" IPMI
+message as a block write to the I2C bus and waits for a response.
+This action can be detrimental to some I2C devices. It is highly recommended
+that the known I2c address be given to the SMBus driver in the smb_addr
+parameter. The default adrress range will not be used when a smb_addr
+parameter is provided.
+
+When compiled into the kernel, the addresses can be specified on the
+kernel command line as:
+
+ ipmb_smb.addr=<adapter1>,<i2caddr1>[,<adapter2>,<i2caddr2>[,...]]
+ ipmi_smb.dbg=<flags1>,<flags2>...
+ ipmi_smb.defaultprobe=0 ipmi_smb.dbg_probe=1
+
+These are the same options as on the module command line.
+
+Note that you might need some I2C changes if CONFIG_IPMI_PANIC_EVENT
+is enabled along with this, so the I2C driver knows to run to
+completion during sending a panic event.
Other Pieces
------------
Watchdog
+--------
A watchdog timer is provided that implements the Linux-standard
watchdog timer interface. It has three module parameters that can be
used to control it:
- insmod ipmi_watchdog timeout=<t> pretimeout=<t> action=<action type>
- preaction=<preaction type> preop=<preop type>
+ modprobe ipmi_watchdog timeout=<t> pretimeout=<t> action=<action type>
+ preaction=<preaction type> preop=<preop type> start_now=x
The timeout is the number of seconds to the action, and the pretimeout
is the amount of seconds before the reset that the pre-timeout panic will
-occur (if pretimeout is zero, then pretimeout will not be enabled).
+occur (if pretimeout is zero, then pretimeout will not be enabled). Note
+that the pretimeout is the time before the final timeout. So if the
+timeout is 50 seconds and the pretimeout is 10 seconds, then the pretimeout
+will occur in 40 second (10 seconds before the timeout).
The action may be "reset", "power_cycle", or "power_off", and
specifies what to do when the timer times out, and defaults to
@@ -344,16 +469,19 @@ When preop is set to "preop_give_data",
on the device when the pretimeout occurs. Select and fasync work on
the device, as well.
+If start_now is set to 1, the watchdog timer will start running as
+soon as the driver is loaded.
+
When compiled into the kernel, the kernel command line is available
for configuring the watchdog:
- ipmi_wdog=<timeout>[,<pretimeout>[,<option>[,<options>....]]]
+ ipmi_watchdog.timeout=<t> ipmi_watchdog.pretimeout=<t>
+ ipmi_watchdog.action=<action type>
+ ipmi_watchdog.preaction=<preaction type>
+ ipmi_watchdog.preop=<preop type>
+ ipmi_watchdog.start_now=x
-The options are the actions and preaction above (if an option
-controlling the same thing is specified twice, the last is taken). An
-options "start_now" is also there, if included, the watchdog will
-start running immediately when all the drivers are ready, it doesn't
-have to have a user hooked up to start it.
+The options are the same as the module parameter options.
The watchdog will panic and start a 120 second reset timeout if it
gets a pre-action. During a panic or a reboot, the watchdog will
--- diff/Documentation/binfmt_misc.txt 2003-10-09 08:47:33.000000000 +0000
+++ source/Documentation/binfmt_misc.txt 2004-03-16 11:27:47.976894064 +0000
@@ -15,7 +15,7 @@ First you must mount binfmt_misc:
mount binfmt_misc -t binfmt_misc /proc/sys/fs/binfmt_misc
To actually register a new binary type, you have to set up a string looking like
-:name:type:offset:magic:mask:interpreter: (where you can choose the ':' upon
+:name:type:offset:magic:mask:interpreter:flags (where you can choose the ':' upon
your needs) and echo it to /proc/sys/fs/binfmt_misc/register.
Here is what the fields mean:
- 'name' is an identifier string. A new /proc file will be created with this
@@ -34,6 +34,28 @@ Here is what the fields mean:
The mask is anded with the byte sequence of the file.
- 'interpreter' is the program that should be invoked with the binary as first
argument (specify the full path)
+ - 'flags' is an optional field that controls several aspects of the invocation
+ of the interpreter. It is a string of capital letters, each controls a certain
+ aspect. The following flags are supported -
+ 'P' - preserve-argv[0]. Legacy behavior of binfmt_misc is to overwrite the
+ original argv[0] with the full path to the binary. When this flag is
+ included, binfmt_misc will add an argument to the argument vector for
+ this purpose, thus preserving the original argv[0].
+ 'O' - open-binary. Legacy behavior of binfmt_misc is to pass the full path
+ of the binary to the interpreter as an argument. When this flag is
+ included, binfmt_misc will open the file for reading and pass its
+ descriptor as an argument, instead of the full path, thus allowing
+ the interpreter to execute non-readable binaries. This feature should
+ be used with care - the interpreter has to be trusted not to emit
+ the contents of the non-readable binary.
+ 'C' - credentials. Currently, the behavior of binfmt_misc is to calculate
+ the credentials and security token of the new process according to
+ the interpreter. When this flag is included, these attributes are
+ calculated according to the binary. It also implies the 'O' flag.
+ This feature should be used with care as the interpreter
+ will run with root permissions when a setuid binary owned by root
+ is run with binfmt_misc.
+
There are some restrictions:
- the whole register string may not exceed 255 characters
@@ -83,9 +105,9 @@ If you want to pass special arguments to
write a wrapper script for it. See Documentation/java.txt for an
example.
-Your interpreter should NOT look in the PATH for the filename; the
-kernel passes it the full filename to use. Using the PATH can cause
-unexpected behaviour and be a security hazard.
+Your interpreter should NOT look in the PATH for the filename; the kernel
+passes it the full filename (or the file descriptor) to use. Using $PATH can
+cause unexpected behaviour and can be a security hazard.
There is a web page about binfmt_misc at
--- diff/Documentation/early-userspace/README 2003-08-26 09:00:51.000000000 +0000
+++ source/Documentation/early-userspace/README 2004-03-16 11:27:47.980893456 +0000
@@ -71,5 +71,31 @@ custom initramfs images that meet your n
For questions and help, you can sign up for the early userspace
mailing list at http://www.zytor.com/mailman/listinfo/klibc
+How does it work?
+=================
+
+The kernel has currently 3 ways to mount the root filesystem:
+
+a) all required device and filesystem drivers compiled into the kernel, no
+ initrd. init/main.c:init() will call prepare_namespace() to mount the
+ final root filesystem, based on the root= option and optional init= to run
+ some other init binary than listed at the end of init/main.c:init().
+
+b) some device and filesystem drivers built as modules and stored in an
+ initrd. The initrd must contain a binary '/linuxrc' which is supposed to
+ load these driver modules. It is also possible to mount the final root
+ filesystem via linuxrc and use the pivot_root syscall. The initrd is
+ mounted and executed via prepare_namespace().
+
+c) using initramfs. The call to prepare_namespace() must be skipped.
+ This means that a binary must do all the work. Said binary can be stored
+ into initramfs either via modifying usr/gen_init_cpio.c or via the new
+ initrd format, an cpio archive. It must be called "/init". This binary
+ is responsible to do all the things prepare_namespace() would do.
+
+ To remain backwards compatibility, the /init binary will only run if it
+ comes via an initramfs cpio archive. If this is not the case,
+ init/main.c:init() will run prepare_namespace() to mount the final root
+ and exec one of the predefined init binaries.
Bryan O'Sullivan <bos@serpentine.com>
--- diff/Documentation/filesystems/proc.txt 2004-03-11 10:20:19.000000000 +0000
+++ source/Documentation/filesystems/proc.txt 2004-03-16 11:27:47.982893152 +0000
@@ -38,6 +38,7 @@ Table of Contents
2.8 /proc/sys/net/ipv4 - IPV4 settings
2.9 Appletalk
2.10 IPX
+ 2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem
------------------------------------------------------------------------------
Preface
@@ -1814,6 +1815,30 @@ The /proc/net/ipx_route table holds a
gives the destination network, the router node (or Directly) and the network
address of the router (or Connected) for internal networks.
+2.11 /proc/sys/fs/mqueue - POSIX message queues filesystem
+----------------------------------------------------------
+
+The "mqueue" filesystem provides the necessary kernel features to enable the
+creation of a user space library that implements the POSIX message queues
+API (as noted by the MSG tag in the POSIX 1003.1-2001 version of the System
+Interfaces specification.)
+
+The "mqueue" filesystem contains values for determining/setting the amount of
+resources used by the file system.
+
+/proc/sys/fs/mqueue/queues_max is a read/write file for setting/getting the
+maximum number of message queues allowed on the system.
+
+/proc/sys/fs/mqueue/msg_max is a read/write file for setting/getting the
+maximum number of messages in a queue value. In fact it is the limiting value
+for another (user) limit which is set in mq_open invocation. This attribute of
+a queue must be less or equal then msg_max.
+
+/proc/sys/fs/mqueue/msgsize_max is a read/write file for setting/getting the
+maximum message size value (it is every message queue's attribute set during
+its creation).
+
+
------------------------------------------------------------------------------
Summary
------------------------------------------------------------------------------
--- diff/Documentation/i386/zero-page.txt 2004-03-16 11:24:01.000000000 +0000
+++ source/Documentation/i386/zero-page.txt 2004-03-16 11:27:47.989892088 +0000
@@ -1,7 +1,7 @@
-Summary of empty_zero_page layout (kernel point of view)
+Summary of boot_params layout (kernel point of view)
( collected by Hans Lermen and Martin Mares )
-The contents of empty_zero_page are used to pass parameters from the
+The contents of boot_params are used to pass parameters from the
16-bit realmode code of the kernel to the 32-bit part. References/settings
to it mainly are in:
@@ -76,9 +76,3 @@ Offset Type Description
0x2d0 - 0x600 E820MAP
0x600 - 0x7ff EDDBUF (setup.S) for disk signature read sector
0x600 - 0x7eb EDDBUF (setup.S) for edd data
-
-0x800 string, 2K max COMMAND_LINE, the kernel commandline as
- copied using CL_OFFSET.
- Note: this will be copied once more by setup.c
- into a local buffer which is only 256 bytes long.
- ( #define COMMAND_LINE_SIZE 256 )
--- diff/Documentation/input/joystick-parport.txt 2002-10-16 03:29:06.000000000 +0000
+++ source/Documentation/input/joystick-parport.txt 2004-03-16 11:27:47.990891936 +0000
@@ -434,7 +434,7 @@ Here are described their command lines:
Using gamecon.c you can connect up to five devices to one parallel port. It
uses the following kernel/module command line:
- gc=port,pad1,pad2,pad3,pad4,pad5
+ gamecon.map=port,pad1,pad2,pad3,pad4,pad5
Where 'port' the number of the parport interface (eg. 0 for parport0).
@@ -457,15 +457,15 @@ uses the following kernel/module command
your controller plugged in before initializing.
Should you want to use more than one of parallel ports at once, you can use
-gc_2 and gc_3 as additional command line parameters for two more parallel
-ports.
+gamecon.map2 and gamecon.map3 as additional command line parameters for two
+more parallel ports.
3.2 db9.c
~~~~~~~~~
Apart from making an interface, there is nothing difficult on using the
db9.c driver. It uses the following kernel/module command line:
- db9=port,type
+ db9.dev=port,type
Where 'port' is the number of the parport interface (eg. 0 for parport0).
@@ -489,14 +489,14 @@ Old parallel ports may not have this fea
10 | Amiga CD32 pad
Should you want to use more than one of these joysticks/pads at once, you
-can use db9_2 and db9_3 as additional command line parameters for two
+can use db9.dev2 and db9.dev3 as additional command line parameters for two
more joysticks/pads.
3.3 turbografx.c
~~~~~~~~~~~~~~~~
The turbografx.c driver uses a very simple kernel/module command line:
- tgfx=port,js1,js2,js3,js4,js5,js6,js7
+ turbografx.map=port,js1,js2,js3,js4,js5,js6,js7
Where 'port' is the number of the parport interface (eg. 0 for parport0).
@@ -504,8 +504,8 @@ more joysticks/pads.
interface ports 1-7 have. For a standard multisystem joystick, this is 1.
Should you want to use more than one of these interfaces at once, you can
-use tgfx_2 and tgfx_3 as additional command line parameters for two more
-interfaces.
+use turbografx.map2 and turbografx.map3 as additional command line parameters
+for two more interfaces.
3.4 PC parallel port pinout
~~~~~~~~~~~~~~~~~~~~~~~~~~~
--- diff/Documentation/input/joystick.txt 2002-10-16 03:29:05.000000000 +0000
+++ source/Documentation/input/joystick.txt 2004-03-16 11:27:47.991891784 +0000
@@ -111,7 +111,7 @@ your needs:
alias tty-ldisc-2 serport
alias char-major-13 input
above input joydev ns558 analog
- options analog js=gamepad
+ options analog map=gamepad,none,2btn
2.5 Verifying that it works
~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -185,7 +185,7 @@ you'll need to specify the types either
module command line, when inserting analog.o into the kernel. The
parameters are:
- js=type,type,type,....
+ analog.map=<type1>,<type2>,<type3>,....
'type' is type of the joystick from the table below, defining joysticks
present on gameports in the system, starting with gameport0, second 'type'
@@ -419,7 +419,7 @@ card.
Amiga joysticks, connected to an Amiga, are supported by the amijoy.c
driver. Since they can't be autodetected, the driver has a command line.
- amijoy=a,b
+ amijoy.map=<a>,<b>
a and b define the joysticks connected to the JOY0DAT and JOY1DAT ports of
the Amiga.
--- diff/Documentation/kernel-parameters.txt 2004-03-16 11:24:01.000000000 +0000
+++ source/Documentation/kernel-parameters.txt 2004-03-16 11:27:47.994891328 +0000
@@ -98,6 +98,10 @@ running once the system is up.
strictly ACPI specification compliant.
See also Documentation/pm.txt.
+
+ acpi_sleep= [HW,ACPI] Sleep options
+ Format: { s3_bios, s3_mode }
+ See Documentation/power/video.txt
acpi_pic_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode
Format: { level | edge }
@@ -158,7 +162,15 @@ running once the system is up.
Format: <host-scsi-id>,<target-scsi-id>,<max-rate>,<max-offset>
See also header of drivers/scsi/AM53C974.c.
- amijoy= [HW,JOY] Amiga joystick support
+ amijoy.map= [HW,JOY] Amiga joystick support
+ Map of devices attached to JOY0DAT and JOY1DAT
+ Format: <a>,<b>
+ See also Documentation/kernel/input/joystick.txt
+
+ analog.map= [HW,JOY] Analog joystick and gamepad support
+ Specifies type or capabilities of an analog joystick
+ connected to one of 16 gameports
+ Format: <type1>,<type2>,..<type16>
apc= [HW,SPARC] Power management functions (SPARCstation-4/5 + deriv.)
Format: noidle
@@ -181,11 +193,18 @@ running once the system is up.
atascsi= [HW,SCSI] Atari SCSI
- atkbd.set= [HW] Select keyboard code set
- Format: <int>
+ atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess, EzKey
+ and similar keyboards
+
+ atkbd.reset= [HW] Reset keyboard during initialization
+
+ atkbd.set= [HW] Select keyboard code set
+ Format: <int> (2 = AT (default) 3 = PS/2)
+
+ atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar keyboards
+
atkbd.softrepeat=
[HW] Use software keyboard repeat
- atkbd.reset= [HW] Reset keyboard during initialization
autotest [IA64]
@@ -287,10 +306,11 @@ running once the system is up.
dasd= [HW,NET]
See header of drivers/s390/block/dasd_devmap.c.
- db9= [HW,JOY]
- db9_2=
- db9_3=
-
+ db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port
+ (one device per port)
+ Format: <port#>,<type>
+ See also Documentation/input/joystick-parport.txt
+
debug [KNL] Enable kernel debugging (events log level).
decnet= [HW,NET]
@@ -384,12 +404,14 @@ running once the system is up.
ftape= [HW] Floppy Tape subsystem debugging options.
See Documentation/ftape.txt.
+ gamecon.map[2|3]=
+ [HW,JOY] Multisystem joystick and NES/SNES/PSX pad
+ support via parallel port (up to 5 devices per port)
+ Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5>
+ See also Documentation/input/joystick-parport.txt
+
gamma= [HW,DRM]
- gc= [HW,JOY]
- gc_2= See Documentation/input/joystick-parport.txt.
- gc_3=
-
gdth= [HW,SCSI]
See header of drivers/scsi/gdth.c.
@@ -616,9 +638,9 @@ running once the system is up.
mga= [HW,DRM]
- mousedev.xres [MOUSE] Horizontal screen resolution, used for devices
+ mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices
reporting absolute coordinates, such as tablets
- mousedev.yres [MOUSE] Vertical screen resolution, used for devices
+ mousedev.yres= [MOUSE] Vertical screen resolution, used for devices
reporting absolute coordinates, such as tablets
mpu401= [HW,OSS]
@@ -1163,10 +1185,6 @@ running once the system is up.
See header of drivers/scsi/t128.c.
tdfx= [HW,DRM]
-
- tgfx= [HW,JOY] TurboGraFX parallel port interface
- tgfx_2= See Documentation/input/joystick-parport.txt.
- tgfx_3=
thash_entries= [KNL,NET]
Set number of hash buckets for TCP connection
@@ -1189,8 +1207,13 @@ running once the system is up.
trix= [HW,OSS] MediaTrix AudioTrix Pro
Format: <io>,<irq>,<dma>,<dma2>,<sb_io>,<sb_irq>,<sb_dma>,<mpu_io>,<mpu_irq>
- tsdev.xres [TS] Horizontal screen resolution.
- tsdev.yres [TS] Vertical screen resolution.
+ tsdev.xres= [TS] Horizontal screen resolution.
+ tsdev.yres= [TS] Vertical screen resolution.
+
+ turbografx.map[2|3]=
+ [HW,JOY] TurboGraFX parallel port interface
+ Format: <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7>
+ See also Documentation/input/joystick-parport.txt
u14-34f= [HW,SCSI] UltraStor 14F/34F SCSI host adapter
See header of drivers/scsi/u14-34f.c.
--- diff/Documentation/usb/acm.txt 2002-10-16 03:27:22.000000000 +0000
+++ source/Documentation/usb/acm.txt 2004-03-16 11:27:48.000890416 +0000
@@ -28,7 +28,7 @@ in the package: See the file COPYING.
1. Usage
~~~~~~~~
- The drivers/usb/acm.c drivers works with USB modems and USB ISDN terminal
+ The drivers/usb/class/cdc-acm.c drivers works with USB modems and USB ISDN terminal
adapters that conform to the Universal Serial Bus Communication Device Class
Abstract Control Model (USB CDC ACM) specification.
@@ -65,9 +65,9 @@ minor/major numbers anywhere you want, b
To use the modems you need these modules loaded:
- usbcore.o
- usb-[uo]hci.o or uhci.o
- acm.o
+ usbcore.ko
+ uhci-hcd.ko ohci-hcd.ko or ehci-hcd.ko
+ cdc-acm.ko
After that, the modem[s] should be accessible. You should be able to use
minicom, ppp and mgetty with them.
--- diff/Documentation/usb/error-codes.txt 2002-10-16 03:27:08.000000000 +0000
+++ source/Documentation/usb/error-codes.txt 2004-03-16 11:27:48.001890264 +0000
@@ -70,7 +70,9 @@ one or more packets could finish before
(That is, if drivers see this it's a bug.)
-EPROTO (*) a) bitstuff error
- b) unknown USB error
+ b) no response packet received within the
+ prescribed bus turn-around time
+ c) unknown USB error
-EILSEQ (*) CRC mismatch
--- diff/MAINTAINERS 2004-03-16 11:24:01.000000000 +0000
+++ source/MAINTAINERS 2004-03-16 11:27:48.672788272 +0000
@@ -1186,6 +1186,12 @@ W: http://sf.net/projects/kernel-janitor
W: http://developer.osdl.org/rddunlap/kj-patches/
S: Maintained
+KGDB FOR I386 PLATFORM
+P: George Anzinger
+M: george@mvista.com
+L: linux-net@vger.kernel.org
+S: Supported
+
KERNEL NFSD
P: Neil Brown
M: neilb@cse.unsw.edu.au
@@ -1966,13 +1972,6 @@ P: Romain Lievin
M: roms@lpg.ticalc.org
S: Maintained
-TIEMAN VOYAGER USB BRAILLE DISPLAY DRIVER
-P: Stephane Dalton
-M: sdalton@videotron.ca
-P: St�phane Doyon
-M: s.doyon@videotron.ca
-S: Maintained
-
TLAN NETWORK DRIVER
P: Samuel Chessman
M: chessman@tux.org
@@ -2152,14 +2151,6 @@ L: linux-usb-devel@lists.sourceforge.net
W: http://pegasus2.sourceforge.net/
S: Maintained
-USB SCANNER DRIVER
-P: Henning Meier-Geinitz
-M: henning@meier-geinitz.de
-L: linux-usb-users@lists.sourceforge.net
-L: linux-usb-devel@lists.sourceforge.net
-W: http://www.meier-geinitz.de/kernel/
-S: Maintained
-
USB SE401 DRIVER
P: Jeroen Vreeken
M: pe1rxq@amsat.org
--- diff/Makefile 2004-03-16 11:24:01.000000000 +0000
+++ source/Makefile 2004-03-16 11:27:48.673788120 +0000
@@ -1,7 +1,7 @@
VERSION = 2
PATCHLEVEL = 6
SUBLEVEL = 5
-EXTRAVERSION =-rc1
+EXTRAVERSION =-rc1-mm1
NAME=Feisty Dunnart
# *DOCUMENTATION*
@@ -304,17 +304,10 @@ RCS_TAR_IGNORE := --exclude SCCS --exclu
# ===========================================================================
# Rules shared between *config targets and build targets
-# Helpers built in scripts/
-
-scripts/docproc scripts/split-include : scripts ;
-
-.PHONY: scripts scripts/fixdep
-scripts:
- $(Q)$(MAKE) $(build)=scripts
-
-scripts/fixdep:
- $(Q)$(MAKE) $(build)=scripts $@
-
+# Basic helpers built in scripts/
+.PHONY: scripts_basic
+scripts_basic:
+ $(Q)$(MAKE) $(build)=scripts/basic
# To make sure we do not include .config for any of the *config targets
# catch them early, and hand them over to scripts/kconfig/Makefile
@@ -358,9 +351,9 @@ ifeq ($(config-targets),1)
# *config targets only - make sure prerequisites are updated, and descend
# in scripts/kconfig to make the *config target
-%config: scripts/fixdep FORCE
+config: scripts_basic FORCE
$(Q)$(MAKE) $(build)=scripts/kconfig $@
-config : scripts/fixdep FORCE
+%config: scripts_basic FORCE
$(Q)$(MAKE) $(build)=scripts/kconfig $@
else
@@ -368,6 +361,16 @@ else
# Build targets only - this includes vmlinux, arch specific targets, clean
# targets and others. In general all targets except *config targets.
+# Additional helpers built in scripts/
+# Carefully list dependencies so we do not try to build scripts twice
+# in parrallel
+.PHONY: scripts
+scripts: scripts_basic include/config/MARKER
+ $(Q)$(MAKE) $(build)=$(@)
+
+scripts_basic: include/linux/autoconf.h
+
+
# That's our default target when none is given on the command line
# Note that 'modules' will be added as a prerequisite as well,
# in the CONFIG_MODULES part below
@@ -400,7 +403,9 @@ include .config
# with it and forgot to run make oldconfig
include/linux/autoconf.h: .config
$(Q)$(MAKE) -f $(srctree)/Makefile silentoldconfig
-
+else
+# Dummy target needed, because used as prerequisite
+include/linux/autoconf.h: ;
endif
include $(srctree)/arch/$(ARCH)/Makefile
@@ -444,6 +449,7 @@ endif
ifdef CONFIG_DEBUG_INFO
CFLAGS += -g
+AFLAGS += -g
endif
# warn about C99 declaration after statement
@@ -633,9 +639,9 @@ include/asm:
# Split autoconf.h into include/linux/config/*
-include/config/MARKER: scripts/split-include include/linux/autoconf.h
+include/config/MARKER: include/linux/autoconf.h
@echo ' SPLIT include/linux/autoconf.h -> include/config/*'
- @scripts/split-include include/linux/autoconf.h include/config
+ @scripts/basic/split-include include/linux/autoconf.h include/config
@touch $@
# Generate some files
@@ -929,7 +935,7 @@ help:
# Documentation targets
# ---------------------------------------------------------------------------
-%docs: scripts/docproc FORCE
+%docs: scripts FORCE
$(Q)$(MAKE) $(build)=Documentation/DocBook $@
# Scripts to check various things for consistency
@@ -982,7 +988,7 @@ if_changed_dep = $(if $(strip $? $(filte
@set -e; \
$(if $($(quiet)cmd_$(1)),echo ' $(subst ','\'',$($(quiet)cmd_$(1)))';) \
$(cmd_$(1)); \
- scripts/fixdep $(depfile) $@ '$(subst $$,$$$$,$(subst ','\'',$(cmd_$(1))))' > $(@D)/.$(@F).tmp; \
+ scripts/basic/fixdep $(depfile) $@ '$(subst $$,$$$$,$(subst ','\'',$(cmd_$(1))))' > $(@D)/.$(@F).tmp; \
rm -f $(depfile); \
mv -f $(@D)/.$(@F).tmp $(@D)/.$(@F).cmd)
--- diff/arch/arm26/Kconfig 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/arm26/Kconfig 2004-03-16 11:27:47.859911848 +0000
@@ -198,11 +198,6 @@ source "drivers/input/Kconfig"
source "drivers/char/Kconfig"
-config KBDMOUSE
- bool
- depends on ARCH_ACORN && BUSMOUSE=y
- default y
-
source "drivers/media/Kconfig"
source "fs/Kconfig"
--- diff/arch/i386/Kconfig 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/i386/Kconfig 2004-03-16 11:27:47.861911544 +0000
@@ -418,6 +418,54 @@ config X86_OOSTORE
depends on (MWINCHIP3D || MWINCHIP2 || MWINCHIPC6) && MTRR
default y
+config X86_4G
+ bool "4 GB kernel-space and 4 GB user-space virtual memory support"
+ help
+ This option is only useful for systems that have more than 1 GB
+ of RAM.
+
+ The default kernel VM layout leaves 1 GB of virtual memory for
+ kernel-space mappings, and 3 GB of VM for user-space applications.
+ This option ups both the kernel-space VM and the user-space VM to
+ 4 GB.
+
+ The cost of this option is additional TLB flushes done at
+ system-entry points that transition from user-mode into kernel-mode.
+ I.e. system calls and page faults, and IRQs that interrupt user-mode
+ code. There's also additional overhead to kernel operations that copy
+ memory to/from user-space. The overhead from this is hard to tell and
+ depends on the workload - it can be anything from no visible overhead
+ to 20-30% overhead. A good rule of thumb is to count with a runtime
+ overhead of 20%.
+
+ The upside is the much increased kernel-space VM, which more than
+ quadruples the maximum amount of RAM supported. Kernels compiled with
+ this option boot on 64GB of RAM and still have more than 3.1 GB of
+ 'lowmem' left. Another bonus is that highmem IO bouncing decreases,
+ if used with drivers that still use bounce-buffers.
+
+ There's also a 33% increase in user-space VM size - database
+ applications might see a boost from this.
+
+ But the cost of the TLB flushes and the runtime overhead has to be
+ weighed against the bonuses offered by the larger VM spaces. The
+ dividing line depends on the actual workload - there might be 4 GB
+ systems that benefit from this option. Systems with less than 4 GB
+ of RAM will rarely see a benefit from this option - but it's not
+ out of question, the exact circumstances have to be considered.
+
+config X86_SWITCH_PAGETABLES
+ def_bool X86_4G
+
+config X86_4G_VM_LAYOUT
+ def_bool X86_4G
+
+config X86_UACCESS_INDIRECT
+ def_bool X86_4G
+
+config X86_HIGH_ENTRY
+ def_bool X86_4G
+
config HPET_TIMER
bool "HPET Timer Support"
help
@@ -475,6 +523,16 @@ config NR_CPUS
This is purely to save memory - each supported CPU adds
approximately eight kilobytes to the kernel image.
+config SCHED_SMT
+ bool "SMT (Hyperthreading) scheduler support"
+ depends on SMP
+ default off
+ help
+ SMT scheduler support improves the CPU scheduler's decision making
+ when dealing with Intel Pentium 4 chips with HyperThreading at a
+ cost of slightly increased overhead in some places. If unsure say
+ N here.
+
config PREEMPT
bool "Preemptible Kernel"
help
@@ -1055,9 +1113,6 @@ choice
prompt "PCI access mode"
depends on PCI && !X86_VISWS
default PCI_GOANY
-
-config PCI_GOBIOS
- bool "BIOS"
---help---
On PCI systems, the BIOS can be used to detect the PCI devices and
determine their configuration. However, some old PCI motherboards
@@ -1073,6 +1128,9 @@ config PCI_GOBIOS
direct access method and falls back to the BIOS if that doesn't
work. If unsure, go with the default, which is "Any".
+config PCI_GOBIOS
+ bool "BIOS"
+
config PCI_GOMMCONFIG
bool "MMConfig"
@@ -1269,6 +1327,15 @@ config DEBUG_PAGEALLOC
This results in a large slowdown, but helps to find certain types
of memory corruptions.
+config SPINLINE
+ bool "Spinlock inlining"
+ depends on DEBUG_KERNEL
+ help
+ This will change spinlocks from out of line to inline, making them
+ account cost to the callers in readprofile, rather than the lock
+ itself (as ".text.lock.filename"). This can be helpful for finding
+ the callers of locks.
+
config DEBUG_HIGHMEM
bool "Highmem debugging"
depends on DEBUG_KERNEL && HIGHMEM
@@ -1285,20 +1352,211 @@ config DEBUG_INFO
Say Y here only if you plan to use gdb to debug the kernel.
If you don't debug the kernel, you can say N.
+config LOCKMETER
+ bool "Kernel lock metering"
+ depends on SMP
+ help
+ Say Y to enable kernel lock metering, which adds overhead to SMP locks,
+ but allows you to see various statistics using the lockstat command.
+
config DEBUG_SPINLOCK_SLEEP
bool "Sleep-inside-spinlock checking"
help
If you say Y here, various routines which may sleep will become very
noisy if they are called with a spinlock held.
+config KGDB
+ bool "Include kgdb kernel debugger"
+ depends on DEBUG_KERNEL
+ help
+ If you say Y here, the system will be compiled with the debug
+ option (-g) and a debugging stub will be included in the
+ kernel. This stub communicates with gdb on another (host)
+ computer via a serial port. The host computer should have
+ access to the kernel binary file (vmlinux) and a serial port
+ that is connected to the target machine. Gdb can be made to
+ configure the serial port or you can use stty and setserial to
+ do this. See the 'target' command in gdb. This option also
+ configures in the ability to request a breakpoint early in the
+ boot process. To request the breakpoint just include 'kgdb'
+ as a boot option when booting the target machine. The system
+ will then break as soon as it looks at the boot options. This
+ option also installs a breakpoint in panic and sends any
+ kernel faults to the debugger. For more information see the
+ Documentation/i386/kgdb/kgdb.txt file.
+
+choice
+ depends on KGDB
+ prompt "Debug serial port BAUD"
+ default KGDB_115200BAUD
+ help
+ Gdb and the kernel stub need to agree on the baud rate to be
+ used. Some systems (x86 family at this writing) allow this to
+ be configured.
+
+config KGDB_9600BAUD
+ bool "9600"
+
+config KGDB_19200BAUD
+ bool "19200"
+
+config KGDB_38400BAUD
+ bool "38400"
+
+config KGDB_57600BAUD
+ bool "57600"
+
+config KGDB_115200BAUD
+ bool "115200"
+endchoice
+
+config KGDB_PORT
+ hex "hex I/O port address of the debug serial port"
+ depends on KGDB
+ default 3f8
+ help
+ Some systems (x86 family at this writing) allow the port
+ address to be configured. The number entered is assumed to be
+ hex, don't put 0x in front of it. The standard address are:
+ COM1 3f8 , irq 4 and COM2 2f8 irq 3. Setserial /dev/ttySx
+ will tell you what you have. It is good to test the serial
+ connection with a live system before trying to debug.
+
+config KGDB_IRQ
+ int "IRQ of the debug serial port"
+ depends on KGDB
+ default 4
+ help
+ This is the irq for the debug port. If everything is working
+ correctly and the kernel has interrupts on a control C to the
+ port should cause a break into the kernel debug stub.
+
+config DEBUG_INFO
+ bool
+ depends on KGDB
+ default y
+
+config KGDB_MORE
+ bool "Add any additional compile options"
+ depends on KGDB
+ default n
+ help
+ Saying yes here turns on the ability to enter additional
+ compile options.
+
+
+config KGDB_OPTIONS
+ depends on KGDB_MORE
+ string "Additional compile arguments"
+ default "-O1"
+ help
+ This option allows you enter additional compile options for
+ the whole kernel compile. Each platform will have a default
+ that seems right for it. For example on PPC "-ggdb -O1", and
+ for i386 "-O1". Note that by configuring KGDB "-g" is already
+ turned on. In addition, on i386 platforms
+ "-fomit-frame-pointer" is deleted from the standard compile
+ options.
+
+config NO_KGDB_CPUS
+ int "Number of CPUs"
+ depends on KGDB && SMP
+ default NR_CPUS
+ help
+
+ This option sets the number of cpus for kgdb ONLY. It is used
+ to prune some internal structures so they look "nice" when
+ displayed with gdb. This is to overcome possibly larger
+ numbers that may have been entered above. Enter the real
+ number to get nice clean kgdb_info displays.
+
+config KGDB_TS
+ bool "Enable kgdb time stamp macros?"
+ depends on KGDB
+ default n
+ help
+ Kgdb event macros allow you to instrument your code with calls
+ to the kgdb event recording function. The event log may be
+ examined with gdb at a break point. Turning on this
+ capability also allows you to choose how many events to
+ keep. Kgdb always keeps the lastest events.
+
+choice
+ depends on KGDB_TS
+ prompt "Max number of time stamps to save?"
+ default KGDB_TS_128
+
+config KGDB_TS_64
+ bool "64"
+
+config KGDB_TS_128
+ bool "128"
+
+config KGDB_TS_256
+ bool "256"
+
+config KGDB_TS_512
+ bool "512"
+
+config KGDB_TS_1024
+ bool "1024"
+
+endchoice
+
+config STACK_OVERFLOW_TEST
+ bool "Turn on kernel stack overflow testing?"
+ depends on KGDB
+ default n
+ help
+ This option enables code in the front line interrupt handlers
+ to check for kernel stack overflow on interrupts and system
+ calls. This is part of the kgdb code on x86 systems.
+
+config KGDB_CONSOLE
+ bool "Enable serial console thru kgdb port"
+ depends on KGDB
+ default n
+ help
+ This option enables the command line "console=kgdb" option.
+ When the system is booted with this option in the command line
+ all kernel printk output is sent to gdb (as well as to other
+ consoles). For this to work gdb must be connected. For this
+ reason, this command line option will generate a breakpoint if
+ gdb has not yet connected. After the gdb continue command is
+ given all pent up console output will be printed by gdb on the
+ host machine. Neither this option, nor KGDB require the
+ serial driver to be configured.
+
+config KGDB_SYSRQ
+ bool "Turn on SysRq 'G' command to do a break?"
+ depends on KGDB
+ default y
+ help
+ This option includes an option in the SysRq code that allows
+ you to enter SysRq G which generates a breakpoint to the KGDB
+ stub. This will work if the keyboard is alive and can
+ interrupt the system. Because of constraints on when the
+ serial port interrupt can be enabled, this code may allow you
+ to interrupt the system before the serial port control C is
+ available. Just say yes here.
+
config FRAME_POINTER
bool "Compile the kernel with frame pointers"
+ default KGDB
help
If you say Y here the resulting kernel image will be slightly larger
and slower, but it will give very useful debugging information.
If you don't debug the kernel, you can say N, but we may not be able
to solve problems without frame pointers.
+config MAGIC_SYSRQ
+ bool
+ depends on KGDB_SYSRQ
+ default y
+
+config 4KSTACKS
+ def_bool y
+
config X86_FIND_SMP_CONFIG
bool
depends on X86_LOCAL_APIC || X86_VOYAGER
--- diff/arch/i386/Makefile 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/i386/Makefile 2004-03-16 11:27:47.900905616 +0000
@@ -19,7 +19,7 @@ LDFLAGS := -m elf_i386
OBJCOPYFLAGS := -O binary -R .note -R .comment -S
LDFLAGS_vmlinux :=
-CFLAGS += -pipe
+CFLAGS += -pipe -msoft-float
# prevent gcc from keeping the stack 16 byte aligned
CFLAGS += $(call check_gcc,-mpreferred-stack-boundary=2,)
@@ -56,9 +56,9 @@ cflags-$(CONFIG_X86_ELAN) += -march=i486
GCC_VERSION := $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-version.sh $(CC))
cflags-$(CONFIG_REGPARM) += $(shell if [ $(GCC_VERSION) -ge 0300 ] ; then echo "-mregparm=3"; fi ;)
-# Enable unit-at-a-time mode when possible. It shrinks the
-# kernel considerably.
-CFLAGS += $(call check_gcc,-funit-at-a-time,)
+# Disable unit-at-a-time mode, it makes gcc use a lot more stack
+# due to the lack of sharing of stacklots.
+CFLAGS += $(call check_gcc,-fno-unit-at-a-time,)
CFLAGS += $(cflags-y)
@@ -97,6 +97,9 @@ mcore-$(CONFIG_X86_ES7000) := mach-es700
# default subarch .h files
mflags-y += -Iinclude/asm-i386/mach-default
+mflags-$(CONFIG_KGDB) += -gdwarf-2
+mflags-$(CONFIG_KGDB_MORE) += $(shell echo $(CONFIG_KGDB_OPTIONS) | sed -e 's/"//g')
+
head-y := arch/i386/kernel/head.o arch/i386/kernel/init_task.o
libs-y += arch/i386/lib/
--- diff/arch/i386/boot/setup.S 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/i386/boot/setup.S 2004-03-16 11:27:47.859911848 +0000
@@ -166,7 +166,7 @@ cmd_line_ptr: .long 0 # (Header versio
# can be located anywhere in
# low memory 0x10000 or higher.
-ramdisk_max: .long MAXMEM-1 # (Header version 0x0203 or later)
+ramdisk_max: .long __MAXMEM-1 # (Header version 0x0203 or later)
# The highest safe address for
# the contents of an initrd
@@ -612,7 +612,7 @@ disk_sig_done:
# int 13h ah=08h "Legacy Get Device Parameters"
#
# A buffer of size EDDMAXNR*(EDDEXTSIZE+EDDPARMSIZE) is reserved for our use
-# in the empty_zero_page at EDDBUF. The first four bytes of which are
+# in the boot_params at EDDBUF. The first four bytes of which are
# used to store the device number, interface support map and version
# results from fn41. The next four bytes are used to store the legacy
# cylinders, heads, and sectors from fn08. The following 74 bytes are used to
@@ -625,9 +625,9 @@ disk_sig_done:
# the structure, and the fn41 and fn08 results are stored at offsets
# from there. This removes the need to increment the pointer for
# every store, and leaves it ready for the fn48 call.
-# A second one-byte buffer, EDDNR, in the empty_zero_page stores
+# A second one-byte buffer, EDDNR, in boot_params stores
# the number of BIOS devices which exist, up to EDDMAXNR.
-# In setup.c, copy_edd() stores both empty_zero_page buffers away
+# In setup.c, copy_edd() stores both boot_params buffers away
# for later use, as they would get overwritten otherwise.
# This code is sensitive to the size of the structs in edd.h
edd_start:
--- diff/arch/i386/kernel/Makefile 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/i386/kernel/Makefile 2004-03-16 11:27:47.883908200 +0000
@@ -7,13 +7,14 @@ extra-y := head.o init_task.o vmlinux.ld
obj-y := process.o semaphore.o signal.o entry.o traps.o irq.o vm86.o \
ptrace.o i8259.o ioport.o ldt.o setup.o time.o sys_i386.o \
pci-dma.o i386_ksyms.o i387.o dmi_scan.o bootflag.o \
- doublefault.o
+ doublefault.o entry_trampoline.o
obj-y += cpu/
obj-y += timers/
obj-$(CONFIG_ACPI_BOOT) += acpi/
obj-$(CONFIG_X86_BIOS_REBOOT) += reboot.o
obj-$(CONFIG_MCA) += mca.o
+obj-$(CONFIG_KGDB) += kgdb_stub.o
obj-$(CONFIG_X86_MSR) += msr.o
obj-$(CONFIG_X86_CPUID) += cpuid.o
obj-$(CONFIG_MICROCODE) += microcode.o
--- diff/arch/i386/kernel/acpi/boot.c 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/i386/kernel/acpi/boot.c 2004-03-16 11:27:47.861911544 +0000
@@ -372,7 +372,7 @@ acpi_scan_rsdp (
* RSDP signature.
*/
for (offset = 0; offset < length; offset += 16) {
- if (strncmp((char *) (start + offset), "RSD PTR ", sig_len))
+ if (strncmp((char *) __va(start + offset), "RSD PTR ", sig_len))
continue;
return (start + offset);
}
--- diff/arch/i386/kernel/acpi/sleep.c 2003-08-20 13:16:24.000000000 +0000
+++ source/arch/i386/kernel/acpi/sleep.c 2004-03-16 11:27:47.862911392 +0000
@@ -19,13 +19,29 @@ extern void zap_low_mappings(void);
extern unsigned long FASTCALL(acpi_copy_wakeup_routine(unsigned long));
-static void init_low_mapping(pgd_t *pgd, int pgd_limit)
+static void map_low(pgd_t *pgd_base, unsigned long start, unsigned long end)
{
- int pgd_ofs = 0;
-
- while ((pgd_ofs < pgd_limit) && (pgd_ofs + USER_PTRS_PER_PGD < PTRS_PER_PGD)) {
- set_pgd(pgd, *(pgd+USER_PTRS_PER_PGD));
- pgd_ofs++, pgd++;
+ unsigned long vaddr;
+ pmd_t *pmd;
+ pgd_t *pgd;
+ int i, j;
+
+ pgd = pgd_base;
+
+ for (i = 0; i < PTRS_PER_PGD; pgd++, i++) {
+ vaddr = i*PGDIR_SIZE;
+ if (end && (vaddr >= end))
+ break;
+ pmd = pmd_offset(pgd, 0);
+ for (j = 0; j < PTRS_PER_PMD; pmd++, j++) {
+ vaddr = i*PGDIR_SIZE + j*PMD_SIZE;
+ if (end && (vaddr >= end))
+ break;
+ if (vaddr < start)
+ continue;
+ set_pmd(pmd, __pmd(_KERNPG_TABLE + _PAGE_PSE +
+ vaddr - start));
+ }
}
}
@@ -39,7 +55,9 @@ int acpi_save_state_mem (void)
{
if (!acpi_wakeup_address)
return 1;
- init_low_mapping(swapper_pg_dir, USER_PTRS_PER_PGD);
+ if (!cpu_has_pse)
+ return 1;
+ map_low(swapper_pg_dir, 0, LOW_MAPPINGS_SIZE);
memcpy((void *) acpi_wakeup_address, &wakeup_start, &wakeup_end - &wakeup_start);
acpi_copy_wakeup_routine(acpi_wakeup_address);
--- diff/arch/i386/kernel/apic.c 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/i386/kernel/apic.c 2004-03-16 11:27:47.863911240 +0000
@@ -1182,9 +1182,6 @@ int __init APIC_init_uniprocessor (void)
phys_cpu_present_map = physid_mask_of_physid(boot_cpu_physical_apicid);
setup_local_APIC();
-
- if (nmi_watchdog == NMI_LOCAL_APIC)
- check_nmi_watchdog();
#ifdef CONFIG_X86_IO_APIC
if (smp_found_config)
if (!skip_ioapic_setup && nr_ioapics)
--- diff/arch/i386/kernel/asm-offsets.c 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/i386/kernel/asm-offsets.c 2004-03-16 11:27:47.863911240 +0000
@@ -31,5 +31,19 @@ void foo(void)
DEFINE(RT_SIGFRAME_sigcontext,
offsetof (struct rt_sigframe, uc.uc_mcontext));
+ DEFINE(TI_task, offsetof (struct thread_info, task));
+ DEFINE(TI_exec_domain, offsetof (struct thread_info, exec_domain));
+ DEFINE(TI_flags, offsetof (struct thread_info, flags));
+ DEFINE(TI_preempt_count, offsetof (struct thread_info, preempt_count));
+ DEFINE(TI_addr_limit, offsetof (struct thread_info, addr_limit));
+ DEFINE(TI_real_stack, offsetof (struct thread_info, real_stack));
+ DEFINE(TI_virtual_stack, offsetof (struct thread_info, virtual_stack));
+ DEFINE(TI_user_pgd, offsetof (struct thread_info, user_pgd));
+
+ DEFINE(FIX_ENTRY_TRAMPOLINE_0_addr,
+ __fix_to_virt(FIX_ENTRY_TRAMPOLINE_0));
+ DEFINE(FIX_VSYSCALL_addr, __fix_to_virt(FIX_VSYSCALL));
DEFINE(PAGE_SIZE_asm, PAGE_SIZE);
+ DEFINE(task_thread_db7,
+ offsetof (struct task_struct, thread.debugreg[7]));
}
--- diff/arch/i386/kernel/cpu/common.c 2004-01-19 10:22:55.000000000 +0000
+++ source/arch/i386/kernel/cpu/common.c 2004-03-16 11:27:47.864911088 +0000
@@ -196,7 +196,6 @@ int __init have_cpuid_p(void)
void __init generic_identify(struct cpuinfo_x86 * c)
{
u32 tfms, xlvl;
- int junk;
if (have_cpuid_p()) {
/* Get vendor name */
@@ -211,8 +210,8 @@ void __init generic_identify(struct cpui
/* Intel-defined flags: level 0x00000001 */
if ( c->cpuid_level >= 0x00000001 ) {
- u32 capability, excap;
- cpuid(0x00000001, &tfms, &junk, &excap, &capability);
+ u32 capability, excap, misc;
+ cpuid(0x00000001, &tfms, &misc, &excap, &capability);
c->x86_capability[0] = capability;
c->x86_capability[4] = excap;
c->x86 = (tfms >> 8) & 15;
@@ -222,6 +221,9 @@ void __init generic_identify(struct cpui
c->x86_model += ((tfms >> 16) & 0xF) << 4;
}
c->x86_mask = tfms & 15;
+
+ if (c->x86_capability[0] & (1<<19))
+ c->x86_clflush_size = ((misc >> 8) & 0xff) * 8;
} else {
/* Have CPUID level 0 only - unheard of */
c->x86 = 4;
@@ -261,8 +263,6 @@ static int __init x86_serial_nr_setup(ch
}
__setup("serialnumber", x86_serial_nr_setup);
-
-
/*
* This does the hard work of actually picking apart the CPU stuff...
*/
@@ -270,7 +270,6 @@ void __init identify_cpu(struct cpuinfo_
{
int i;
- c->loops_per_jiffy = loops_per_jiffy;
c->x86_cache_size = -1;
c->x86_vendor = X86_VENDOR_UNKNOWN;
c->cpuid_level = -1; /* CPUID not detected */
@@ -279,6 +278,7 @@ void __init identify_cpu(struct cpuinfo_
c->x86_model_id[0] = '\0'; /* Unset */
memset(&c->x86_capability, 0, sizeof c->x86_capability);
+ c->x86_clflush_size = 0;
if (!have_cpuid_p()) {
/* First of all, decide if this is a 486 or higher */
/* It's a 486 if we can modify the AC flag */
@@ -290,6 +290,14 @@ void __init identify_cpu(struct cpuinfo_
generic_identify(c);
+ if (!c->x86_clflush_size) {
+ /* No cache line size autodetected - manual estimate: */
+ if (c->x86 <= 4)
+ c->x86_clflush_size = 16;
+ else
+ c->x86_clflush_size = 32;
+ }
+
printk(KERN_DEBUG "CPU: After generic identify, caps: %08lx %08lx %08lx %08lx\n",
c->x86_capability[0],
c->x86_capability[1],
@@ -376,7 +384,19 @@ void __init identify_cpu(struct cpuinfo_
#ifdef CONFIG_X86_MCE
mcheck_init(c);
#endif
+ late_identify_cpu(c);
}
+
+void __init late_identify_cpu(struct cpuinfo_x86 *c)
+{
+ /*
+ * The timer is not yet running when identify cpu is called for the
+ * first cpu - check_bugs() calls late_identify_cpu and transfers
+ * loops_per_jiffy from calibrate_delay into the cpu data area.
+ */
+ c->loops_per_jiffy = loops_per_jiffy;
+}
+
/*
* Perform early boot up checks for a valid TSC. See arch/i386/kernel/time.c
*/
@@ -514,12 +534,16 @@ void __init cpu_init (void)
set_tss_desc(cpu,t);
cpu_gdt_table[cpu][GDT_ENTRY_TSS].b &= 0xfffffdff;
load_TR_desc();
- load_LDT(&init_mm.context);
+ if (cpu)
+ load_LDT(&init_mm.context);
/* Set up doublefault TSS pointer in the GDT */
__set_tss_desc(cpu, GDT_ENTRY_DOUBLEFAULT_TSS, &doublefault_tss);
cpu_gdt_table[cpu][GDT_ENTRY_DOUBLEFAULT_TSS].b &= 0xfffffdff;
+ if (cpu)
+ trap_init_virtual_GDT();
+
/* Clear %fs and %gs. */
asm volatile ("xorl %eax, %eax; movl %eax, %fs; movl %eax, %gs");
--- diff/arch/i386/kernel/cpu/cpufreq/p4-clockmod.c 2004-02-18 08:54:07.000000000 +0000
+++ source/arch/i386/kernel/cpu/cpufreq/p4-clockmod.c 2004-03-16 11:27:47.865910936 +0000
@@ -57,8 +57,7 @@ static int cpufreq_p4_setdc(unsigned int
u32 l, h;
cpumask_t cpus_allowed, affected_cpu_map;
struct cpufreq_freqs freqs;
- int hyperthreading = 0;
- int sibling = 0;
+ int j;
if (!cpu_online(cpu) || (newstate > DC_DISABLE) ||
(newstate == DC_RESV))
@@ -68,13 +67,10 @@ static int cpufreq_p4_setdc(unsigned int
cpus_allowed = current->cpus_allowed;
/* only run on CPU to be set, or on its sibling */
- affected_cpu_map = cpumask_of_cpu(cpu);
-#ifdef CONFIG_X86_HT
- hyperthreading = ((cpu_has_ht) && (smp_num_siblings == 2));
- if (hyperthreading) {
- sibling = cpu_sibling_map[cpu];
- cpu_set(sibling, affected_cpu_map);
- }
+#ifdef CONFIG_SMP
+ affected_cpu_map = cpu_sibling_map[cpu];
+#else
+ affected_cpu_map = cpumask_of_cpu(cpu);
#endif
set_cpus_allowed(current, affected_cpu_map);
BUG_ON(!cpu_isset(smp_processor_id(), affected_cpu_map));
@@ -97,11 +93,11 @@ static int cpufreq_p4_setdc(unsigned int
/* notifiers */
freqs.old = stock_freq * l / 8;
freqs.new = stock_freq * newstate / 8;
- freqs.cpu = cpu;
- cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
- if (hyperthreading) {
- freqs.cpu = sibling;
- cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
+ for_each_cpu(j) {
+ if (cpu_isset(j, affected_cpu_map)) {
+ freqs.cpu = j;
+ cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
+ }
}
rdmsr(MSR_IA32_THERM_STATUS, l, h);
@@ -132,10 +128,11 @@ static int cpufreq_p4_setdc(unsigned int
set_cpus_allowed(current, cpus_allowed);
/* notifiers */
- cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
- if (hyperthreading) {
- freqs.cpu = cpu;
- cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
+ for_each_cpu(j) {
+ if (cpu_isset(j, affected_cpu_map)) {
+ freqs.cpu = j;
+ cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
+ }
}
return 0;
--- diff/arch/i386/kernel/cpu/intel.c 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/i386/kernel/cpu/intel.c 2004-03-16 11:27:47.866910784 +0000
@@ -10,6 +10,7 @@
#include <asm/processor.h>
#include <asm/msr.h>
#include <asm/uaccess.h>
+#include <asm/desc.h>
#include "cpu.h"
@@ -19,8 +20,6 @@
#include <mach_apic.h>
#endif
-extern int trap_init_f00f_bug(void);
-
#ifdef CONFIG_X86_INTEL_USERCOPY
/*
* Alignment at which movsl is preferred for bulk memory copies.
@@ -165,7 +164,7 @@ static void __init init_intel(struct cpu
c->f00f_bug = 1;
if ( !f00f_workaround_enabled ) {
- trap_init_f00f_bug();
+ trap_init_virtual_IDT();
printk(KERN_NOTICE "Intel Pentium with F0 0F bug - workaround enabled.\n");
f00f_workaround_enabled = 1;
}
@@ -250,6 +249,12 @@ static void __init init_intel(struct cpu
/* SEP CPUID bug: Pentium Pro reports SEP but doesn't have it until model 3 mask 3 */
if ((c->x86<<8 | c->x86_model<<4 | c->x86_mask) < 0x633)
clear_bit(X86_FEATURE_SEP, c->x86_capability);
+ /*
+ * FIXME: SEP is disabled for 4G/4G for now:
+ */
+#ifdef CONFIG_X86_HIGH_ENTRY
+ clear_bit(X86_FEATURE_SEP, c->x86_capability);
+#endif
/* Names for the Pentium II/Celeron processors
detectable only by also checking the cache size.
--- diff/arch/i386/kernel/dmi_scan.c 2004-02-09 10:36:07.000000000 +0000
+++ source/arch/i386/kernel/dmi_scan.c 2004-03-16 11:27:47.867910632 +0000
@@ -525,7 +525,7 @@ static __init int print_if_true(struct d
#ifdef CONFIG_ACPI_BOOT
extern int acpi_disabled, acpi_force;
-static __init __attribute__((unused)) int acpi_disable(struct dmi_blacklist *d)
+static __init __attribute__((unused)) int disable_acpi(struct dmi_blacklist *d)
{
if (!acpi_force) {
printk(KERN_NOTICE "%s detected: acpi off\n",d->ident);
@@ -933,7 +933,7 @@ static __initdata struct dmi_blacklist d
* Boxes that need ACPI disabled
*/
- { acpi_disable, "IBM Thinkpad", {
+ { disable_acpi, "IBM Thinkpad", {
MATCH(DMI_BOARD_VENDOR, "IBM"),
MATCH(DMI_BOARD_NAME, "2629H1G"),
NO_MATCH, NO_MATCH }},
--- diff/arch/i386/kernel/doublefault.c 2003-10-09 08:47:16.000000000 +0000
+++ source/arch/i386/kernel/doublefault.c 2004-03-16 11:27:47.867910632 +0000
@@ -7,12 +7,13 @@
#include <asm/uaccess.h>
#include <asm/pgtable.h>
#include <asm/desc.h>
+#include <asm/fixmap.h>
#define DOUBLEFAULT_STACKSIZE (1024)
static unsigned long doublefault_stack[DOUBLEFAULT_STACKSIZE];
#define STACK_START (unsigned long)(doublefault_stack+DOUBLEFAULT_STACKSIZE)
-#define ptr_ok(x) ((x) > 0xc0000000 && (x) < 0xc1000000)
+#define ptr_ok(x) (((x) > __PAGE_OFFSET && (x) < (__PAGE_OFFSET + 0x01000000)) || ((x) >= FIXADDR_START))
static void doublefault_fn(void)
{
@@ -38,8 +39,8 @@ static void doublefault_fn(void)
printk("eax = %08lx, ebx = %08lx, ecx = %08lx, edx = %08lx\n",
t->eax, t->ebx, t->ecx, t->edx);
- printk("esi = %08lx, edi = %08lx\n",
- t->esi, t->edi);
+ printk("esi = %08lx, edi = %08lx, ebp = %08lx\n",
+ t->esi, t->edi, t->ebp);
}
}
--- diff/arch/i386/kernel/entry.S 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/i386/kernel/entry.S 2004-03-16 11:27:47.870910176 +0000
@@ -43,11 +43,25 @@
#include <linux/config.h>
#include <linux/linkage.h>
#include <asm/thread_info.h>
+#include <asm/asm_offsets.h>
#include <asm/errno.h>
#include <asm/segment.h>
+#include <asm/page.h>
#include <asm/smp.h>
#include <asm/page.h>
#include "irq_vectors.h"
+ /* We do not recover from a stack overflow, but at least
+ * we know it happened and should be able to track it down.
+ */
+#ifdef CONFIG_STACK_OVERFLOW_TEST
+#define STACK_OVERFLOW_TEST \
+ testl $(THREAD_SIZE - 512),%esp; \
+ jnz 10f; \
+ call stack_overflow; \
+10:
+#else
+#define STACK_OVERFLOW_TEST
+#endif
#define nr_syscalls ((syscall_table_size)/4)
@@ -87,7 +101,102 @@ TSS_ESP0_OFFSET = (4 - 0x200)
#define resume_kernel restore_all
#endif
-#define SAVE_ALL \
+#ifdef CONFIG_X86_HIGH_ENTRY
+
+#ifdef CONFIG_X86_SWITCH_PAGETABLES
+
+#if defined(CONFIG_PREEMPT) && defined(CONFIG_SMP)
+/*
+ * If task is preempted in __SWITCH_KERNELSPACE, and moved to another cpu,
+ * __switch_to repoints %esp to the appropriate virtual stack; but %ebp is
+ * left stale, so we must check whether to repeat the real stack calculation.
+ */
+#define repeat_if_esp_changed \
+ xorl %esp, %ebp; \
+ testl $-THREAD_SIZE, %ebp; \
+ jnz 0b
+#else
+#define repeat_if_esp_changed
+#endif
+
+/* clobbers ebx, edx and ebp */
+
+#define __SWITCH_KERNELSPACE \
+ cmpl $0xff000000, %esp; \
+ jb 1f; \
+ \
+ /* \
+ * switch pagetables and load the real stack, \
+ * keep the stack offset: \
+ */ \
+ \
+ movl $swapper_pg_dir-__PAGE_OFFSET, %edx; \
+ \
+ /* GET_THREAD_INFO(%ebp) intermixed */ \
+0: \
+ movl %esp, %ebp; \
+ movl %esp, %ebx; \
+ andl $(-THREAD_SIZE), %ebp; \
+ andl $(THREAD_SIZE-1), %ebx; \
+ orl TI_real_stack(%ebp), %ebx; \
+ repeat_if_esp_changed; \
+ \
+ movl %edx, %cr3; \
+ movl %ebx, %esp; \
+1:
+
+#endif
+
+
+#define __SWITCH_USERSPACE \
+ /* interrupted any of the user return paths? */ \
+ \
+ movl EIP(%esp), %eax; \
+ \
+ cmpl $int80_ret_start_marker, %eax; \
+ jb 33f; /* nope - continue with sysexit check */\
+ cmpl $int80_ret_end_marker, %eax; \
+ jb 22f; /* yes - switch to virtual stack */ \
+33: \
+ cmpl $sysexit_ret_start_marker, %eax; \
+ jb 44f; /* nope - continue with user check */ \
+ cmpl $sysexit_ret_end_marker, %eax; \
+ jb 22f; /* yes - switch to virtual stack */ \
+ /* return to userspace? */ \
+44: \
+ movl EFLAGS(%esp),%ecx; \
+ movb CS(%esp),%cl; \
+ testl $(VM_MASK | 3),%ecx; \
+ jz 2f; \
+22: \
+ /* \
+ * switch to the virtual stack, then switch to \
+ * the userspace pagetables. \
+ */ \
+ \
+ GET_THREAD_INFO(%ebp); \
+ movl TI_virtual_stack(%ebp), %edx; \
+ movl TI_user_pgd(%ebp), %ecx; \
+ \
+ movl %esp, %ebx; \
+ andl $(THREAD_SIZE-1), %ebx; \
+ orl %ebx, %edx; \
+int80_ret_start_marker: \
+ movl %edx, %esp; \
+ movl %ecx, %cr3; \
+ \
+ __RESTORE_ALL; \
+int80_ret_end_marker: \
+2:
+
+#else /* !CONFIG_X86_HIGH_ENTRY */
+
+#define __SWITCH_KERNELSPACE
+#define __SWITCH_USERSPACE
+
+#endif
+
+#define __SAVE_ALL \
cld; \
pushl %es; \
pushl %ds; \
@@ -102,7 +211,7 @@ TSS_ESP0_OFFSET = (4 - 0x200)
movl %edx, %ds; \
movl %edx, %es;
-#define RESTORE_INT_REGS \
+#define __RESTORE_INT_REGS \
popl %ebx; \
popl %ecx; \
popl %edx; \
@@ -111,29 +220,28 @@ TSS_ESP0_OFFSET = (4 - 0x200)
popl %ebp; \
popl %eax
-#define RESTORE_REGS \
- RESTORE_INT_REGS; \
-1: popl %ds; \
-2: popl %es; \
+#define __RESTORE_REGS \
+ __RESTORE_INT_REGS; \
+111: popl %ds; \
+222: popl %es; \
.section .fixup,"ax"; \
-3: movl $0,(%esp); \
- jmp 1b; \
-4: movl $0,(%esp); \
- jmp 2b; \
+444: movl $0,(%esp); \
+ jmp 111b; \
+555: movl $0,(%esp); \
+ jmp 222b; \
.previous; \
.section __ex_table,"a";\
.align 4; \
- .long 1b,3b; \
- .long 2b,4b; \
+ .long 111b,444b;\
+ .long 222b,555b;\
.previous
-
-#define RESTORE_ALL \
- RESTORE_REGS \
+#define __RESTORE_ALL \
+ __RESTORE_REGS \
addl $4, %esp; \
-1: iret; \
+333: iret; \
.section .fixup,"ax"; \
-2: sti; \
+666: sti; \
movl $(__USER_DS), %edx; \
movl %edx, %ds; \
movl %edx, %es; \
@@ -142,10 +250,19 @@ TSS_ESP0_OFFSET = (4 - 0x200)
.previous; \
.section __ex_table,"a";\
.align 4; \
- .long 1b,2b; \
+ .long 333b,666b;\
.previous
+#define SAVE_ALL \
+ __SAVE_ALL; \
+ __SWITCH_KERNELSPACE; \
+ STACK_OVERFLOW_TEST;
+
+#define RESTORE_ALL \
+ __SWITCH_USERSPACE; \
+ __RESTORE_ALL;
+.section .entry.text,"ax"
ENTRY(lcall7)
pushfl # We get a different stack layout with call
@@ -163,7 +280,7 @@ do_lcall:
movl %edx,EIP(%ebp) # Now we move them to their "normal" places
movl %ecx,CS(%ebp) #
GET_THREAD_INFO_WITH_ESP(%ebp) # GET_THREAD_INFO
- movl TI_EXEC_DOMAIN(%ebp), %edx # Get the execution domain
+ movl TI_exec_domain(%ebp), %edx # Get the execution domain
call *4(%edx) # Call the lcall7 handler for the domain
addl $4, %esp
popl %eax
@@ -208,7 +325,7 @@ ENTRY(resume_userspace)
cli # make sure we don't miss an interrupt
# setting need_resched or sigpending
# between sampling and the iret
- movl TI_FLAGS(%ebp), %ecx
+ movl TI_flags(%ebp), %ecx
andl $_TIF_WORK_MASK, %ecx # is there any work to be done on
# int/exception return?
jne work_pending
@@ -216,18 +333,18 @@ ENTRY(resume_userspace)
#ifdef CONFIG_PREEMPT
ENTRY(resume_kernel)
- cmpl $0,TI_PRE_COUNT(%ebp) # non-zero preempt_count ?
+ cmpl $0,TI_preempt_count(%ebp) # non-zero preempt_count ?
jnz restore_all
need_resched:
- movl TI_FLAGS(%ebp), %ecx # need_resched set ?
+ movl TI_flags(%ebp), %ecx # need_resched set ?
testb $_TIF_NEED_RESCHED, %cl
jz restore_all
testl $IF_MASK,EFLAGS(%esp) # interrupts off (exception path) ?
jz restore_all
- movl $PREEMPT_ACTIVE,TI_PRE_COUNT(%ebp)
+ movl $PREEMPT_ACTIVE,TI_preempt_count(%ebp)
sti
call schedule
- movl $0,TI_PRE_COUNT(%ebp)
+ movl $0,TI_preempt_count(%ebp)
cli
jmp need_resched
#endif
@@ -246,38 +363,50 @@ sysenter_past_esp:
pushl $(__USER_CS)
pushl $SYSENTER_RETURN
-/*
- * Load the potential sixth argument from user stack.
- * Careful about security.
- */
- cmpl $__PAGE_OFFSET-3,%ebp
- jae syscall_fault
-1: movl (%ebp),%ebp
-.section __ex_table,"a"
- .align 4
- .long 1b,syscall_fault
-.previous
-
pushl %eax
SAVE_ALL
GET_THREAD_INFO(%ebp)
cmpl $(nr_syscalls), %eax
jae syscall_badsys
- testb $_TIF_SYSCALL_TRACE,TI_FLAGS(%ebp)
+ testb $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),TI_flags(%ebp)
jnz syscall_trace_entry
call *sys_call_table(,%eax,4)
movl %eax,EAX(%esp)
cli
- movl TI_FLAGS(%ebp), %ecx
+ movl TI_flags(%ebp), %ecx
testw $_TIF_ALLWORK_MASK, %cx
jne syscall_exit_work
+
+#ifdef CONFIG_X86_SWITCH_PAGETABLES
+
+ GET_THREAD_INFO(%ebp)
+ movl TI_virtual_stack(%ebp), %edx
+ movl TI_user_pgd(%ebp), %ecx
+ movl %esp, %ebx
+ andl $(THREAD_SIZE-1), %ebx
+ orl %ebx, %edx
+sysexit_ret_start_marker:
+ movl %edx, %esp
+ movl %ecx, %cr3
+#endif
+ /*
+ * only ebx is not restored by the userspace sysenter vsyscall
+ * code, it assumes it to be callee-saved.
+ */
+ movl EBX(%esp), %ebx
+
/* if something modifies registers it must also disable sysexit */
+
movl EIP(%esp), %edx
movl OLDESP(%esp), %ecx
+
sti
sysexit
-
+#ifdef CONFIG_X86_SWITCH_PAGETABLES
+sysexit_ret_end_marker:
+ nop
+#endif
# system call handler stub
ENTRY(system_call)
@@ -287,7 +416,7 @@ ENTRY(system_call)
cmpl $(nr_syscalls), %eax
jae syscall_badsys
# system call tracing in operation
- testb $_TIF_SYSCALL_TRACE,TI_FLAGS(%ebp)
+ testb $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),TI_flags(%ebp)
jnz syscall_trace_entry
syscall_call:
call *sys_call_table(,%eax,4)
@@ -296,10 +425,23 @@ syscall_exit:
cli # make sure we don't miss an interrupt
# setting need_resched or sigpending
# between sampling and the iret
- movl TI_FLAGS(%ebp), %ecx
+ movl TI_flags(%ebp), %ecx
testw $_TIF_ALLWORK_MASK, %cx # current->work
jne syscall_exit_work
restore_all:
+#ifdef CONFIG_TRAP_BAD_SYSCALL_EXITS
+ movl EFLAGS(%esp), %eax # mix EFLAGS and CS
+ movb CS(%esp), %al
+ testl $(VM_MASK | 3), %eax
+ jz resume_kernelX # returning to kernel or vm86-space
+
+ cmpl $0,TI_preempt_count(%ebp) # non-zero preempt_count ?
+ jz resume_kernelX
+
+ int $3
+
+resume_kernelX:
+#endif
RESTORE_ALL
# perform work that needs to be done immediately before resumption
@@ -312,7 +454,7 @@ work_resched:
cli # make sure we don't miss an interrupt
# setting need_resched or sigpending
# between sampling and the iret
- movl TI_FLAGS(%ebp), %ecx
+ movl TI_flags(%ebp), %ecx
andl $_TIF_WORK_MASK, %ecx # is there any work to be done other
# than syscall tracing?
jz restore_all
@@ -327,6 +469,22 @@ work_notifysig: # deal with pending s
# vm86-space
xorl %edx, %edx
call do_notify_resume
+
+#if CONFIG_X86_HIGH_ENTRY
+ /*
+ * Reload db7 if necessary:
+ */
+ movl TI_flags(%ebp), %ecx
+ testb $_TIF_DB7, %cl
+ jnz work_db7
+
+ jmp restore_all
+
+work_db7:
+ movl TI_task(%ebp), %edx;
+ movl task_thread_db7(%edx), %edx;
+ movl %edx, %db7;
+#endif
jmp restore_all
ALIGN
@@ -354,7 +512,7 @@ syscall_trace_entry:
# perform syscall exit tracing
ALIGN
syscall_exit_work:
- testb $_TIF_SYSCALL_TRACE, %cl
+ testb $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT), %cl
jz work_pending
sti # could let do_syscall_trace() call
# schedule() instead
@@ -382,7 +540,7 @@ syscall_badsys:
*/
.data
ENTRY(interrupt)
-.text
+.previous
vector=0
ENTRY(irq_entries_start)
@@ -392,7 +550,7 @@ ENTRY(irq_entries_start)
jmp common_interrupt
.data
.long 1b
-.text
+.previous
vector=vector+1
.endr
@@ -433,12 +591,17 @@ error_code:
movl ES(%esp), %edi # get the function address
movl %eax, ORIG_EAX(%esp)
movl %ecx, ES(%esp)
- movl %esp, %edx
pushl %esi # push the error code
- pushl %edx # push the pt_regs pointer
movl $(__USER_DS), %edx
movl %edx, %ds
movl %edx, %es
+
+/* clobbers edx, ebx and ebp */
+ __SWITCH_KERNELSPACE
+
+ leal 4(%esp), %edx # prepare pt_regs
+ pushl %edx # push pt_regs
+
call *%edi
addl $8, %esp
jmp ret_from_exception
@@ -529,7 +692,7 @@ nmi_stack_correct:
pushl %edx
call do_nmi
addl $8, %esp
- RESTORE_ALL
+ jmp restore_all
nmi_stack_fixup:
FIX_STACK(12,nmi_stack_correct, 1)
@@ -606,6 +769,8 @@ ENTRY(spurious_interrupt_bug)
pushl $do_spurious_interrupt_bug
jmp error_code
+.previous
+
.data
ENTRY(sys_call_table)
.long sys_restart_syscall /* 0 - old "setup()" system call, used for restarting */
@@ -865,7 +1030,7 @@ ENTRY(sys_call_table)
.long sys_epoll_create
.long sys_epoll_ctl /* 255 */
.long sys_epoll_wait
- .long sys_remap_file_pages
+ .long old_remap_file_pages
.long sys_set_tid_address
.long sys_timer_create
.long sys_timer_settime /* 260 */
@@ -882,5 +1047,12 @@ ENTRY(sys_call_table)
.long sys_utimes
.long sys_fadvise64_64
.long sys_ni_syscall /* sys_vserver */
+ .long sys_mq_open
+ .long sys_mq_unlink /* 275 */
+ .long sys_mq_timedsend
+ .long sys_mq_timedreceive
+ .long sys_mq_notify
+ .long sys_mq_getsetattr
+ .long sys_remap_file_pages /* 280 */
syscall_table_size=(.-sys_call_table)
--- diff/arch/i386/kernel/head.S 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/i386/kernel/head.S 2004-03-16 11:27:47.871910024 +0000
@@ -18,12 +18,7 @@
#include <asm/cache.h>
#include <asm/thread_info.h>
#include <asm/asm_offsets.h>
-
-#define OLD_CL_MAGIC_ADDR 0x90020
-#define OLD_CL_MAGIC 0xA33F
-#define OLD_CL_BASE_ADDR 0x90000
-#define OLD_CL_OFFSET 0x90022
-#define NEW_CL_POINTER 0x228 /* Relative to real mode data */
+#include <asm/setup.h>
/*
* References to members of the new_cpu_data structure.
@@ -195,22 +190,15 @@ ENTRY(startup_32_smp)
call setup_idt
/*
- * Copy bootup parameters out of the way. First 2kB of
- * _empty_zero_page is for boot parameters, second 2kB
- * is for the command line.
- *
+ * Copy bootup parameters out of the way.
* Note: %esi still has the pointer to the real-mode data.
*/
- movl $empty_zero_page,%edi
- movl $512,%ecx
+ movl $boot_params,%edi
+ movl $(PARAM_SIZE/4),%ecx
cld
rep
movsl
- xorl %eax,%eax
- movl $512,%ecx
- rep
- stosl
- movl empty_zero_page+NEW_CL_POINTER,%esi
+ movl boot_params+NEW_CL_POINTER,%esi
andl %esi,%esi
jnz 2f # New command line protocol
cmpw $(OLD_CL_MAGIC),OLD_CL_MAGIC_ADDR
@@ -218,8 +206,8 @@ ENTRY(startup_32_smp)
movzwl OLD_CL_OFFSET,%esi
addl $(OLD_CL_BASE_ADDR),%esi
2:
- movl $empty_zero_page+2048,%edi
- movl $512,%ecx
+ movl $saved_command_line,%edi
+ movl $(COMMAND_LINE_SIZE/4),%ecx
rep
movsl
1:
--- diff/arch/i386/kernel/i386_ksyms.c 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/i386/kernel/i386_ksyms.c 2004-03-16 11:27:47.872909872 +0000
@@ -93,7 +93,6 @@ EXPORT_SYMBOL_NOVERS(__down_failed_inter
EXPORT_SYMBOL_NOVERS(__down_failed_trylock);
EXPORT_SYMBOL_NOVERS(__up_wakeup);
/* Networking helper routines. */
-EXPORT_SYMBOL(csum_partial_copy_generic);
/* Delay loops */
EXPORT_SYMBOL(__ndelay);
EXPORT_SYMBOL(__udelay);
@@ -107,13 +106,17 @@ EXPORT_SYMBOL_NOVERS(__get_user_4);
EXPORT_SYMBOL(strpbrk);
EXPORT_SYMBOL(strstr);
+#if !defined(CONFIG_X86_UACCESS_INDIRECT)
EXPORT_SYMBOL(strncpy_from_user);
-EXPORT_SYMBOL(__strncpy_from_user);
+EXPORT_SYMBOL(__direct_strncpy_from_user);
EXPORT_SYMBOL(clear_user);
EXPORT_SYMBOL(__clear_user);
EXPORT_SYMBOL(__copy_from_user_ll);
EXPORT_SYMBOL(__copy_to_user_ll);
EXPORT_SYMBOL(strnlen_user);
+#else /* CONFIG_X86_UACCESS_INDIRECT */
+EXPORT_SYMBOL(direct_csum_partial_copy_generic);
+#endif
EXPORT_SYMBOL(dma_alloc_coherent);
EXPORT_SYMBOL(dma_free_coherent);
--- diff/arch/i386/kernel/i387.c 2004-02-18 08:54:07.000000000 +0000
+++ source/arch/i386/kernel/i387.c 2004-03-16 11:27:47.872909872 +0000
@@ -218,6 +218,7 @@ void set_fpu_mxcsr( struct task_struct *
static int convert_fxsr_to_user( struct _fpstate __user *buf,
struct i387_fxsave_struct *fxsave )
{
+ struct _fpreg tmp[8]; /* 80 bytes scratch area */
unsigned long env[7];
struct _fpreg __user *to;
struct _fpxreg *from;
@@ -234,23 +235,25 @@ static int convert_fxsr_to_user( struct
if ( __copy_to_user( buf, env, 7 * sizeof(unsigned long) ) )
return 1;
- to = &buf->_st[0];
+ to = tmp;
from = (struct _fpxreg *) &fxsave->st_space[0];
for ( i = 0 ; i < 8 ; i++, to++, from++ ) {
unsigned long *t = (unsigned long *)to;
unsigned long *f = (unsigned long *)from;
- if (__put_user(*f, t) ||
- __put_user(*(f + 1), t + 1) ||
- __put_user(from->exponent, &to->exponent))
- return 1;
+ *t = *f;
+ *(t + 1) = *(f+1);
+ to->exponent = from->exponent;
}
+ if (copy_to_user(buf->_st, tmp, sizeof(struct _fpreg [8])))
+ return 1;
return 0;
}
static int convert_fxsr_from_user( struct i387_fxsave_struct *fxsave,
struct _fpstate __user *buf )
{
+ struct _fpreg tmp[8]; /* 80 bytes scratch area */
unsigned long env[7];
struct _fpxreg *to;
struct _fpreg __user *from;
@@ -258,6 +261,8 @@ static int convert_fxsr_from_user( struc
if ( __copy_from_user( env, buf, 7 * sizeof(long) ) )
return 1;
+ if (copy_from_user(tmp, buf->_st, sizeof(struct _fpreg [8])))
+ return 1;
fxsave->cwd = (unsigned short)(env[0] & 0xffff);
fxsave->swd = (unsigned short)(env[1] & 0xffff);
@@ -269,15 +274,14 @@ static int convert_fxsr_from_user( struc
fxsave->fos = env[6];
to = (struct _fpxreg *) &fxsave->st_space[0];
- from = &buf->_st[0];
+ from = tmp;
for ( i = 0 ; i < 8 ; i++, to++, from++ ) {
unsigned long *t = (unsigned long *)to;
unsigned long *f = (unsigned long *)from;
- if (__get_user(*t, f) ||
- __get_user(*(t + 1), f + 1) ||
- __get_user(to->exponent, &from->exponent))
- return 1;
+ *t = *f;
+ *(t + 1) = *(f + 1);
+ to->exponent = from->exponent;
}
return 0;
}
--- diff/arch/i386/kernel/i8259.c 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/i386/kernel/i8259.c 2004-03-16 11:27:47.873909720 +0000
@@ -444,4 +444,7 @@ void __init init_IRQ(void)
*/
if (boot_cpu_data.hard_math && !cpu_has_fpu)
setup_irq(FPU_IRQ, &fpu_irq);
+
+ current_thread_info()->cpu = 0;
+ irq_ctx_init(0);
}
--- diff/arch/i386/kernel/init_task.c 2003-10-09 08:47:33.000000000 +0000
+++ source/arch/i386/kernel/init_task.c 2004-03-16 11:27:47.873909720 +0000
@@ -20,13 +20,13 @@ EXPORT_SYMBOL(init_mm);
/*
* Initial thread structure.
*
- * We need to make sure that this is 8192-byte aligned due to the
+ * We need to make sure that this is THREAD_SIZE aligned due to the
* way process stacks are handled. This is done by having a special
* "init_task" linker map entry..
*/
union thread_union init_thread_union
__attribute__((__section__(".data.init_task"))) =
- { INIT_THREAD_INFO(init_task) };
+ { INIT_THREAD_INFO(init_task, init_thread_union) };
/*
* Initial task structure.
@@ -44,5 +44,5 @@ EXPORT_SYMBOL(init_task);
* section. Since TSS's are completely CPU-local, we want them
* on exact cacheline boundaries, to eliminate cacheline ping-pong.
*/
-struct tss_struct init_tss[NR_CPUS] __cacheline_aligned = { [0 ... NR_CPUS-1] = INIT_TSS };
+struct tss_struct init_tss[NR_CPUS] __attribute__((__section__(".data.tss"))) = { [0 ... NR_CPUS-1] = INIT_TSS };
--- diff/arch/i386/kernel/io_apic.c 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/i386/kernel/io_apic.c 2004-03-16 11:27:47.875909416 +0000
@@ -317,8 +317,7 @@ struct irq_cpu_info {
#define IRQ_ALLOWED(cpu, allowed_mask) cpu_isset(cpu, allowed_mask)
-#define CPU_TO_PACKAGEINDEX(i) \
- ((physical_balance && i > cpu_sibling_map[i]) ? cpu_sibling_map[i] : i)
+#define CPU_TO_PACKAGEINDEX(i) (first_cpu(cpu_sibling_map[i]))
#define MAX_BALANCED_IRQ_INTERVAL (5*HZ)
#define MIN_BALANCED_IRQ_INTERVAL (HZ/2)
@@ -401,6 +400,7 @@ static void do_irq_balance(void)
unsigned long max_cpu_irq = 0, min_cpu_irq = (~0);
unsigned long move_this_load = 0;
int max_loaded = 0, min_loaded = 0;
+ int load;
unsigned long useful_load_threshold = balanced_irq_interval + 10;
int selected_irq;
int tmp_loaded, first_attempt = 1;
@@ -452,7 +452,7 @@ static void do_irq_balance(void)
for (i = 0; i < NR_CPUS; i++) {
if (!cpu_online(i))
continue;
- if (physical_balance && i > cpu_sibling_map[i])
+ if (i != CPU_TO_PACKAGEINDEX(i))
continue;
if (min_cpu_irq > CPU_IRQ(i)) {
min_cpu_irq = CPU_IRQ(i);
@@ -471,7 +471,7 @@ tryanothercpu:
for (i = 0; i < NR_CPUS; i++) {
if (!cpu_online(i))
continue;
- if (physical_balance && i > cpu_sibling_map[i])
+ if (i != CPU_TO_PACKAGEINDEX(i))
continue;
if (max_cpu_irq <= CPU_IRQ(i))
continue;
@@ -551,9 +551,14 @@ tryanotherirq:
* We seek the least loaded sibling by making the comparison
* (A+B)/2 vs B
*/
- if (physical_balance && (CPU_IRQ(min_loaded) >> 1) >
- CPU_IRQ(cpu_sibling_map[min_loaded]))
- min_loaded = cpu_sibling_map[min_loaded];
+ load = CPU_IRQ(min_loaded) >> 1;
+ for_each_cpu_mask(j, cpu_sibling_map[min_loaded]) {
+ if (load > CPU_IRQ(j)) {
+ /* This won't change cpu_sibling_map[min_loaded] */
+ load = CPU_IRQ(j);
+ min_loaded = j;
+ }
+ }
cpus_and(allowed_mask, cpu_online_map, irq_affinity[selected_irq]);
target_cpu_mask = cpumask_of_cpu(min_loaded);
--- diff/arch/i386/kernel/irq.c 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/i386/kernel/irq.c 2004-03-16 11:27:47.876909264 +0000
@@ -75,6 +75,14 @@ irq_desc_t irq_desc[NR_IRQS] __cacheline
static void register_irq_proc (unsigned int irq);
/*
+ * per-CPU IRQ handling stacks
+ */
+#ifdef CONFIG_4KSTACKS
+union irq_ctx *hardirq_ctx[NR_CPUS];
+union irq_ctx *softirq_ctx[NR_CPUS];
+#endif
+
+/*
* Special irq handlers.
*/
@@ -213,7 +221,7 @@ inline void synchronize_irq(unsigned int
* waste of time and is not what some drivers would
* prefer.
*/
-int handle_IRQ_event(unsigned int irq,
+asmlinkage int handle_IRQ_event(unsigned int irq,
struct pt_regs *regs, struct irqaction *action)
{
int status = 1; /* Force the "do bottom halves" bit */
@@ -436,7 +444,7 @@ asmlinkage unsigned int do_IRQ(struct pt
__asm__ __volatile__("andl %%esp,%0" :
"=r" (esp) : "0" (THREAD_SIZE - 1));
- if (unlikely(esp < (sizeof(struct thread_info) + 1024))) {
+ if (unlikely(esp < (sizeof(struct thread_info) + STACK_WARN))) {
printk("do_IRQ: stack overflow: %ld\n",
esp - sizeof(struct thread_info));
dump_stack();
@@ -484,11 +492,70 @@ asmlinkage unsigned int do_IRQ(struct pt
* useful for irq hardware that does not mask cleanly in an
* SMP environment.
*/
+#ifdef CONFIG_4KSTACKS
+
+ for (;;) {
+ irqreturn_t action_ret;
+ u32 *isp;
+ union irq_ctx * curctx;
+ union irq_ctx * irqctx;
+
+ curctx = (union irq_ctx *) current_thread_info();
+ irqctx = hardirq_ctx[smp_processor_id()];
+
+ spin_unlock(&desc->lock);
+
+ /*
+ * this is where we switch to the IRQ stack. However, if we are already using
+ * the IRQ stack (because we interrupted a hardirq handler) we can't do that
+ * and just have to keep using the current stack (which is the irq stack already
+ * after all)
+ */
+
+ if (curctx == irqctx)
+ action_ret = handle_IRQ_event(irq, ®s, action);
+ else {
+ /* build the stack frame on the IRQ stack */
+ isp = (u32*) ((char*)irqctx + sizeof(*irqctx));
+ irqctx->tinfo.task = curctx->tinfo.task;
+ irqctx->tinfo.real_stack = curctx->tinfo.real_stack;
+ irqctx->tinfo.virtual_stack = curctx->tinfo.virtual_stack;
+ irqctx->tinfo.previous_esp = current_stack_pointer();
+
+ *--isp = (u32) action;
+ *--isp = (u32) ®s;
+ *--isp = (u32) irq;
+
+ asm volatile(
+ " xchgl %%ebx,%%esp \n"
+ " call handle_IRQ_event \n"
+ " xchgl %%ebx,%%esp \n"
+ : "=a"(action_ret)
+ : "b"(isp)
+ : "memory", "cc", "edx", "ecx"
+ );
+
+
+ }
+ spin_lock(&desc->lock);
+ if (!noirqdebug)
+ note_interrupt(irq, desc, action_ret);
+ if (curctx != irqctx)
+ irqctx->tinfo.task = NULL;
+ if (likely(!(desc->status & IRQ_PENDING)))
+ break;
+ desc->status &= ~IRQ_PENDING;
+ }
+
+#else
+
for (;;) {
irqreturn_t action_ret;
spin_unlock(&desc->lock);
+
action_ret = handle_IRQ_event(irq, ®s, action);
+
spin_lock(&desc->lock);
if (!noirqdebug)
note_interrupt(irq, desc, action_ret);
@@ -496,6 +563,7 @@ asmlinkage unsigned int do_IRQ(struct pt
break;
desc->status &= ~IRQ_PENDING;
}
+#endif
desc->status &= ~IRQ_INPROGRESS;
out:
@@ -508,6 +576,8 @@ out:
irq_exit();
+ kgdb_process_breakpoint();
+
return 1;
}
@@ -1053,3 +1123,81 @@ void init_irq_proc (void)
register_irq_proc(i);
}
+
+#ifdef CONFIG_4KSTACKS
+static char softirq_stack[NR_CPUS * THREAD_SIZE] __attribute__((__aligned__(THREAD_SIZE)));
+static char hardirq_stack[NR_CPUS * THREAD_SIZE] __attribute__((__aligned__(THREAD_SIZE)));
+
+/*
+ * allocate per-cpu stacks for hardirq and for softirq processing
+ */
+void irq_ctx_init(int cpu)
+{
+ union irq_ctx *irqctx;
+
+ if (hardirq_ctx[cpu])
+ return;
+
+ irqctx = (union irq_ctx*) &hardirq_stack[cpu*THREAD_SIZE];
+ irqctx->tinfo.task = NULL;
+ irqctx->tinfo.exec_domain = NULL;
+ irqctx->tinfo.cpu = cpu;
+ irqctx->tinfo.preempt_count = HARDIRQ_OFFSET;
+ irqctx->tinfo.addr_limit = MAKE_MM_SEG(0);
+
+ hardirq_ctx[cpu] = irqctx;
+
+ irqctx = (union irq_ctx*) &softirq_stack[cpu*THREAD_SIZE];
+ irqctx->tinfo.task = NULL;
+ irqctx->tinfo.exec_domain = NULL;
+ irqctx->tinfo.cpu = cpu;
+ irqctx->tinfo.preempt_count = SOFTIRQ_OFFSET;
+ irqctx->tinfo.addr_limit = MAKE_MM_SEG(0);
+
+ softirq_ctx[cpu] = irqctx;
+
+ printk("CPU %u irqstacks, hard=%p soft=%p\n",
+ cpu,hardirq_ctx[cpu],softirq_ctx[cpu]);
+}
+
+extern asmlinkage void __do_softirq(void);
+
+asmlinkage void do_softirq(void)
+{
+ unsigned long flags;
+ struct thread_info *curctx;
+ union irq_ctx *irqctx;
+ u32 *isp;
+
+ if (in_interrupt())
+ return;
+
+ local_irq_save(flags);
+
+ if (local_softirq_pending()) {
+ curctx = current_thread_info();
+ irqctx = softirq_ctx[smp_processor_id()];
+ irqctx->tinfo.task = curctx->task;
+ irqctx->tinfo.real_stack = curctx->real_stack;
+ irqctx->tinfo.virtual_stack = curctx->virtual_stack;
+ irqctx->tinfo.previous_esp = current_stack_pointer();
+
+ /* build the stack frame on the softirq stack */
+ isp = (u32*) ((char*)irqctx + sizeof(*irqctx));
+
+
+ asm volatile(
+ " xchgl %%ebx,%%esp \n"
+ " call __do_softirq \n"
+ " movl %%ebx,%%esp \n"
+ : "=b"(isp)
+ : "0"(isp)
+ : "memory", "cc", "edx", "ecx", "eax"
+ );
+ }
+
+ local_irq_restore(flags);
+}
+
+EXPORT_SYMBOL(do_softirq);
+#endif
--- diff/arch/i386/kernel/ldt.c 2003-10-09 08:47:16.000000000 +0000
+++ source/arch/i386/kernel/ldt.c 2004-03-16 11:27:47.882908352 +0000
@@ -2,7 +2,7 @@
* linux/kernel/ldt.c
*
* Copyright (C) 1992 Krishna Balasubramanian and Linus Torvalds
- * Copyright (C) 1999 Ingo Molnar <mingo@redhat.com>
+ * Copyright (C) 1999, 2003 Ingo Molnar <mingo@redhat.com>
*/
#include <linux/errno.h>
@@ -18,6 +18,8 @@
#include <asm/system.h>
#include <asm/ldt.h>
#include <asm/desc.h>
+#include <linux/highmem.h>
+#include <asm/atomic_kmap.h>
#ifdef CONFIG_SMP /* avoids "defined but not used" warnig */
static void flush_ldt(void *null)
@@ -29,34 +31,31 @@ static void flush_ldt(void *null)
static int alloc_ldt(mm_context_t *pc, int mincount, int reload)
{
- void *oldldt;
- void *newldt;
- int oldsize;
+ int oldsize, newsize, i;
if (mincount <= pc->size)
return 0;
+ /*
+ * LDT got larger - reallocate if necessary.
+ */
oldsize = pc->size;
mincount = (mincount+511)&(~511);
- if (mincount*LDT_ENTRY_SIZE > PAGE_SIZE)
- newldt = vmalloc(mincount*LDT_ENTRY_SIZE);
- else
- newldt = kmalloc(mincount*LDT_ENTRY_SIZE, GFP_KERNEL);
-
- if (!newldt)
- return -ENOMEM;
-
- if (oldsize)
- memcpy(newldt, pc->ldt, oldsize*LDT_ENTRY_SIZE);
- oldldt = pc->ldt;
- memset(newldt+oldsize*LDT_ENTRY_SIZE, 0, (mincount-oldsize)*LDT_ENTRY_SIZE);
- pc->ldt = newldt;
- wmb();
+ newsize = mincount*LDT_ENTRY_SIZE;
+ for (i = 0; i < newsize; i += PAGE_SIZE) {
+ int nr = i/PAGE_SIZE;
+ BUG_ON(i >= 64*1024);
+ if (!pc->ldt_pages[nr]) {
+ pc->ldt_pages[nr] = alloc_page(GFP_HIGHUSER);
+ if (!pc->ldt_pages[nr])
+ return -ENOMEM;
+ clear_highpage(pc->ldt_pages[nr]);
+ }
+ }
pc->size = mincount;
- wmb();
-
if (reload) {
#ifdef CONFIG_SMP
cpumask_t mask;
+
preempt_disable();
load_LDT(pc);
mask = cpumask_of_cpu(smp_processor_id());
@@ -67,21 +66,20 @@ static int alloc_ldt(mm_context_t *pc, i
load_LDT(pc);
#endif
}
- if (oldsize) {
- if (oldsize*LDT_ENTRY_SIZE > PAGE_SIZE)
- vfree(oldldt);
- else
- kfree(oldldt);
- }
return 0;
}
static inline int copy_ldt(mm_context_t *new, mm_context_t *old)
{
- int err = alloc_ldt(new, old->size, 0);
- if (err < 0)
+ int i, err, size = old->size, nr_pages = (size*LDT_ENTRY_SIZE + PAGE_SIZE-1)/PAGE_SIZE;
+
+ err = alloc_ldt(new, size, 0);
+ if (err < 0) {
+ new->size = 0;
return err;
- memcpy(new->ldt, old->ldt, old->size*LDT_ENTRY_SIZE);
+ }
+ for (i = 0; i < nr_pages; i++)
+ copy_user_highpage(new->ldt_pages[i], old->ldt_pages[i], 0);
return 0;
}
@@ -96,6 +94,7 @@ int init_new_context(struct task_struct
init_MUTEX(&mm->context.sem);
mm->context.size = 0;
+ memset(mm->context.ldt_pages, 0, sizeof(struct page *) * MAX_LDT_PAGES);
old_mm = current->mm;
if (old_mm && old_mm->context.size > 0) {
down(&old_mm->context.sem);
@@ -107,23 +106,21 @@ int init_new_context(struct task_struct
/*
* No need to lock the MM as we are the last user
+ * Do not touch the ldt register, we are already
+ * in the next thread.
*/
void destroy_context(struct mm_struct *mm)
{
- if (mm->context.size) {
- if (mm == current->active_mm)
- clear_LDT();
- if (mm->context.size*LDT_ENTRY_SIZE > PAGE_SIZE)
- vfree(mm->context.ldt);
- else
- kfree(mm->context.ldt);
- mm->context.size = 0;
- }
+ int i, nr_pages = (mm->context.size*LDT_ENTRY_SIZE + PAGE_SIZE-1) / PAGE_SIZE;
+
+ for (i = 0; i < nr_pages; i++)
+ __free_page(mm->context.ldt_pages[i]);
+ mm->context.size = 0;
}
static int read_ldt(void __user * ptr, unsigned long bytecount)
{
- int err;
+ int err, i;
unsigned long size;
struct mm_struct * mm = current->mm;
@@ -138,8 +135,25 @@ static int read_ldt(void __user * ptr, u
size = bytecount;
err = 0;
- if (copy_to_user(ptr, mm->context.ldt, size))
- err = -EFAULT;
+ /*
+ * This is necessary just in case we got here straight from a
+ * context-switch where the ptes were set but no tlb flush
+ * was done yet. We rather avoid doing a TLB flush in the
+ * context-switch path and do it here instead.
+ */
+ __flush_tlb_global();
+
+ for (i = 0; i < size; i += PAGE_SIZE) {
+ int nr = i / PAGE_SIZE, bytes;
+ char *kaddr = kmap(mm->context.ldt_pages[nr]);
+
+ bytes = size - i;
+ if (bytes > PAGE_SIZE)
+ bytes = PAGE_SIZE;
+ if (copy_to_user(ptr + i, kaddr, size - i))
+ err = -EFAULT;
+ kunmap(mm->context.ldt_pages[nr]);
+ }
up(&mm->context.sem);
if (err < 0)
return err;
@@ -158,7 +172,7 @@ static int read_default_ldt(void __user
err = 0;
address = &default_ldt[0];
- size = 5*sizeof(struct desc_struct);
+ size = 5*LDT_ENTRY_SIZE;
if (size > bytecount)
size = bytecount;
@@ -200,7 +214,15 @@ static int write_ldt(void __user * ptr,
goto out_unlock;
}
- lp = (__u32 *) ((ldt_info.entry_number << 3) + (char *) mm->context.ldt);
+ /*
+ * No rescheduling allowed from this point to the install.
+ *
+ * We do a TLB flush for the same reason as in the read_ldt() path.
+ */
+ preempt_disable();
+ __flush_tlb_global();
+ lp = (__u32 *) ((ldt_info.entry_number << 3) +
+ (char *) __kmap_atomic_vaddr(KM_LDT_PAGE0));
/* Allow LDTs to be cleared by the user. */
if (ldt_info.base_addr == 0 && ldt_info.limit == 0) {
@@ -221,6 +243,7 @@ install:
*lp = entry_1;
*(lp+1) = entry_2;
error = 0;
+ preempt_enable();
out_unlock:
up(&mm->context.sem);
@@ -248,3 +271,26 @@ asmlinkage int sys_modify_ldt(int func,
}
return ret;
}
+
+/*
+ * load one particular LDT into the current CPU
+ */
+void load_LDT_nolock(mm_context_t *pc, int cpu)
+{
+ struct page **pages = pc->ldt_pages;
+ int count = pc->size;
+ int nr_pages, i;
+
+ if (likely(!count)) {
+ pages = &default_ldt_page;
+ count = 5;
+ }
+ nr_pages = (count*LDT_ENTRY_SIZE + PAGE_SIZE-1) / PAGE_SIZE;
+
+ for (i = 0; i < nr_pages; i++) {
+ __kunmap_atomic_type(KM_LDT_PAGE0 - i);
+ __kmap_atomic(pages[i], KM_LDT_PAGE0 - i);
+ }
+ set_ldt_desc(cpu, (void *)__kmap_atomic_vaddr(KM_LDT_PAGE0), count);
+ load_LDT_desc();
+}
--- diff/arch/i386/kernel/mpparse.c 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/i386/kernel/mpparse.c 2004-03-16 11:27:47.884908048 +0000
@@ -668,7 +668,7 @@ void __init get_smp_config (void)
* Read the physical hardware table. Anything here will
* override the defaults.
*/
- if (!smp_read_mpc((void *)mpf->mpf_physptr)) {
+ if (!smp_read_mpc((void *)phys_to_virt(mpf->mpf_physptr))) {
smp_found_config = 0;
printk(KERN_ERR "BIOS bug, MP table errors detected!...\n");
printk(KERN_ERR "... disabling SMP support. (tell your hw vendor)\n");
--- diff/arch/i386/kernel/nmi.c 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/i386/kernel/nmi.c 2004-03-16 11:27:47.885907896 +0000
@@ -31,9 +31,19 @@
#include <asm/mpspec.h>
#include <asm/nmi.h>
+#ifdef CONFIG_KGDB
+#include <asm/kgdb.h>
+#ifdef CONFIG_SMP
+unsigned int nmi_watchdog = NMI_IO_APIC;
+#else
+unsigned int nmi_watchdog = NMI_LOCAL_APIC;
+#endif
+#else
unsigned int nmi_watchdog = NMI_NONE;
+#endif
static unsigned int nmi_hz = HZ;
-unsigned int nmi_perfctr_msr; /* the MSR to reset in NMI handler */
+static unsigned int nmi_perfctr_msr; /* the MSR to reset in NMI handler */
+static unsigned int nmi_p4_cccr_val;
extern void show_registers(struct pt_regs *regs);
/* nmi_active:
@@ -66,7 +76,8 @@ int nmi_active;
#define P4_ESCR_EVENT_SELECT(N) ((N)<<25)
#define P4_ESCR_OS (1<<3)
#define P4_ESCR_USR (1<<2)
-#define P4_CCCR_OVF_PMI (1<<26)
+#define P4_CCCR_OVF_PMI0 (1<<26)
+#define P4_CCCR_OVF_PMI1 (1<<27)
#define P4_CCCR_THRESHOLD(N) ((N)<<20)
#define P4_CCCR_COMPLEMENT (1<<19)
#define P4_CCCR_COMPARE (1<<18)
@@ -79,7 +90,7 @@ int nmi_active;
#define MSR_P4_IQ_COUNTER0 0x30C
#define P4_NMI_CRU_ESCR0 (P4_ESCR_EVENT_SELECT(0x3F)|P4_ESCR_OS|P4_ESCR_USR)
#define P4_NMI_IQ_CCCR0 \
- (P4_CCCR_OVF_PMI|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT| \
+ (P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT| \
P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE)
int __init check_nmi_watchdog (void)
@@ -107,11 +118,6 @@ int __init check_nmi_watchdog (void)
}
printk("OK.\n");
- /* now that we know it works we can reduce NMI frequency to
- something more reasonable; makes a difference in some configs */
- if (nmi_watchdog == NMI_LOCAL_APIC)
- nmi_hz = 1;
-
return 0;
}
@@ -322,6 +328,11 @@ static int setup_p4_watchdog(void)
return 0;
nmi_perfctr_msr = MSR_P4_IQ_COUNTER0;
+ nmi_p4_cccr_val = P4_NMI_IQ_CCCR0;
+#ifdef CONFIG_SMP
+ if (smp_num_siblings == 2)
+ nmi_p4_cccr_val |= P4_CCCR_OVF_PMI1;
+#endif
if (!(misc_enable & MSR_P4_MISC_ENABLE_PEBS_UNAVAIL))
clear_msr_range(0x3F1, 2);
@@ -339,12 +350,14 @@ static int setup_p4_watchdog(void)
Dprintk("setting P4_IQ_COUNTER0 to 0x%08lx\n", -(cpu_khz/nmi_hz*1000));
wrmsr(MSR_P4_IQ_COUNTER0, -(cpu_khz/nmi_hz*1000), -1);
apic_write(APIC_LVTPC, APIC_DM_NMI);
- wrmsr(MSR_P4_IQ_CCCR0, P4_NMI_IQ_CCCR0, 0);
+ wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
return 1;
}
void setup_apic_nmi_watchdog (void)
{
+ nmi_hz = 1;
+
switch (boot_cpu_data.x86_vendor) {
case X86_VENDOR_AMD:
if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15)
@@ -408,6 +421,9 @@ void touch_nmi_watchdog (void)
for (i = 0; i < NR_CPUS; i++)
alert_counter[i] = 0;
}
+#ifdef CONFIG_KGDB
+int tune_watchdog = 5*HZ;
+#endif
void nmi_watchdog_tick (struct pt_regs * regs)
{
@@ -421,12 +437,24 @@ void nmi_watchdog_tick (struct pt_regs *
sum = irq_stat[cpu].apic_timer_irqs;
+#ifdef CONFIG_KGDB
+ if (! in_kgdb(regs) && last_irq_sums[cpu] == sum ) {
+
+#else
if (last_irq_sums[cpu] == sum) {
+#endif
/*
* Ayiee, looks like this CPU is stuck ...
* wait a few IRQs (5 seconds) before doing the oops ...
*/
alert_counter[cpu]++;
+#ifdef CONFIG_KGDB
+ if (alert_counter[cpu] == tune_watchdog) {
+ kgdb_handle_exception(2, SIGPWR, 0, regs);
+ last_irq_sums[cpu] = sum;
+ alert_counter[cpu] = 0;
+ }
+#endif
if (alert_counter[cpu] == 5*nmi_hz) {
spin_lock(&nmi_print_lock);
/*
@@ -455,7 +483,7 @@ void nmi_watchdog_tick (struct pt_regs *
* - LVTPC is masked on interrupt and must be
* unmasked by the LVTPC handler.
*/
- wrmsr(MSR_P4_IQ_CCCR0, P4_NMI_IQ_CCCR0, 0);
+ wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
apic_write(APIC_LVTPC, APIC_DM_NMI);
}
wrmsr(nmi_perfctr_msr, -(cpu_khz/nmi_hz*1000), -1);
--- diff/arch/i386/kernel/process.c 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/i386/kernel/process.c 2004-03-16 11:27:47.886907744 +0000
@@ -46,6 +46,7 @@
#include <asm/i387.h>
#include <asm/irq.h>
#include <asm/desc.h>
+#include <asm/atomic_kmap.h>
#ifdef CONFIG_MATH_EMULATION
#include <asm/math_emu.h>
#endif
@@ -303,6 +304,9 @@ void flush_thread(void)
struct task_struct *tsk = current;
memset(tsk->thread.debugreg, 0, sizeof(unsigned long)*8);
+#ifdef CONFIG_X86_HIGH_ENTRY
+ clear_thread_flag(TIF_DB7);
+#endif
memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array));
/*
* Forget coprocessor state..
@@ -316,9 +320,8 @@ void release_thread(struct task_struct *
if (dead_task->mm) {
// temporary debugging check
if (dead_task->mm->context.size) {
- printk("WARNING: dead process %8s still has LDT? <%p/%d>\n",
+ printk("WARNING: dead process %8s still has LDT? <%d>\n",
dead_task->comm,
- dead_task->mm->context.ldt,
dead_task->mm->context.size);
BUG();
}
@@ -342,7 +345,7 @@ int copy_thread(int nr, unsigned long cl
{
struct pt_regs * childregs;
struct task_struct *tsk;
- int err;
+ int err, i;
childregs = ((struct pt_regs *) (THREAD_SIZE + (unsigned long) p->thread_info)) - 1;
struct_cpy(childregs, regs);
@@ -353,7 +356,18 @@ int copy_thread(int nr, unsigned long cl
p->thread.esp = (unsigned long) childregs;
p->thread.esp0 = (unsigned long) (childregs+1);
+ /*
+ * get the two stack pages, for the virtual stack.
+ *
+ * IMPORTANT: this code relies on the fact that the task
+ * structure is an THREAD_SIZE aligned piece of physical memory.
+ */
+ for (i = 0; i < ARRAY_SIZE(p->thread.stack_page); i++)
+ p->thread.stack_page[i] =
+ virt_to_page((unsigned long)p->thread_info + (i*PAGE_SIZE));
+
p->thread.eip = (unsigned long) ret_from_fork;
+ p->thread_info->real_stack = p->thread_info;
savesegment(fs,p->thread.fs);
savesegment(gs,p->thread.gs);
@@ -505,10 +519,42 @@ struct task_struct fastcall * __switch_t
__unlazy_fpu(prev_p);
+#ifdef CONFIG_X86_HIGH_ENTRY
+{
+ int i;
+ /*
+ * Set the ptes of the virtual stack. (NOTE: a one-page TLB flush is
+ * needed because otherwise NMIs could interrupt the
+ * user-return code with a virtual stack and stale TLBs.)
+ */
+ for (i = 0; i < ARRAY_SIZE(next->stack_page); i++) {
+ __kunmap_atomic_type(KM_VSTACK_TOP-i);
+ __kmap_atomic(next->stack_page[i], KM_VSTACK_TOP-i);
+ }
+ /*
+ * NOTE: here we rely on the task being the stack as well
+ */
+ next_p->thread_info->virtual_stack =
+ (void *)__kmap_atomic_vaddr(KM_VSTACK_TOP);
+}
+#if defined(CONFIG_PREEMPT) && defined(CONFIG_SMP)
+ /*
+ * If next was preempted on entry from userspace to kernel,
+ * and now it's on a different cpu, we need to adjust %esp.
+ * This assumes that entry.S does not copy %esp while on the
+ * virtual stack (with interrupts enabled): which is so,
+ * except within __SWITCH_KERNELSPACE itself.
+ */
+ if (unlikely(next->esp >= TASK_SIZE)) {
+ next->esp &= THREAD_SIZE - 1;
+ next->esp |= (unsigned long) next_p->thread_info->virtual_stack;
+ }
+#endif
+#endif
/*
* Reload esp0, LDT and the page table pointer:
*/
- load_esp0(tss, next);
+ load_virtual_esp0(tss, next_p);
/*
* Load the per-thread Thread-Local Storage descriptor.
--- diff/arch/i386/kernel/ptrace.c 2003-06-09 13:18:17.000000000 +0000
+++ source/arch/i386/kernel/ptrace.c 2004-03-16 11:27:47.887907592 +0000
@@ -14,6 +14,7 @@
#include <linux/ptrace.h>
#include <linux/user.h>
#include <linux/security.h>
+#include <linux/audit.h>
#include <asm/uaccess.h>
#include <asm/pgtable.h>
@@ -524,6 +525,14 @@ out:
__attribute__((regparm(3)))
void do_syscall_trace(struct pt_regs *regs, int entryexit)
{
+ if (unlikely(current->audit_context)) {
+ if (!entryexit)
+ audit_syscall_entry(current, regs->orig_eax,
+ regs->ebx);
+ else
+ audit_syscall_exit(current);
+ }
+
if (!test_thread_flag(TIF_SYSCALL_TRACE))
return;
if (!(current->ptrace & PT_PTRACED))
--- diff/arch/i386/kernel/reboot.c 2004-01-19 10:22:55.000000000 +0000
+++ source/arch/i386/kernel/reboot.c 2004-03-16 11:27:47.887907592 +0000
@@ -155,12 +155,11 @@ void machine_real_restart(unsigned char
CMOS_WRITE(0x00, 0x8f);
spin_unlock_irqrestore(&rtc_lock, flags);
- /* Remap the kernel at virtual address zero, as well as offset zero
- from the kernel segment. This assumes the kernel segment starts at
- virtual address PAGE_OFFSET. */
-
- memcpy (swapper_pg_dir, swapper_pg_dir + USER_PGD_PTRS,
- sizeof (swapper_pg_dir [0]) * KERNEL_PGD_PTRS);
+ /*
+ * Remap the first 16 MB of RAM (which includes the kernel image)
+ * at virtual address zero:
+ */
+ setup_identity_mappings(swapper_pg_dir, 0, LOW_MAPPINGS_SIZE);
/*
* Use `swapper_pg_dir' as our page directory.
--- diff/arch/i386/kernel/setup.c 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/i386/kernel/setup.c 2004-03-16 11:27:47.889907288 +0000
@@ -130,6 +130,8 @@ unsigned long saved_videomode;
static char command_line[COMMAND_LINE_SIZE];
char saved_command_line[COMMAND_LINE_SIZE];
+unsigned char __initdata boot_params[PARAM_SIZE];
+
static struct resource code_resource = { "Kernel code", 0x100000, 0 };
static struct resource data_resource = { "Kernel data", 0, 0 };
@@ -456,7 +458,7 @@ EXPORT_SYMBOL(edd_disk80_sig);
#endif
/**
* copy_edd() - Copy the BIOS EDD information
- * from empty_zero_page into a safe place.
+ * from boot_params into a safe place.
*
*/
static inline void copy_edd(void)
@@ -485,12 +487,11 @@ static void __init setup_memory_region(v
static void __init parse_cmdline_early (char ** cmdline_p)
{
- char c = ' ', *to = command_line, *from = COMMAND_LINE;
+ char c = ' ', *to = command_line, *from = saved_command_line;
int len = 0;
int userdef = 0;
/* Save unparsed command line copy for /proc/cmdline */
- memcpy(saved_command_line, COMMAND_LINE, COMMAND_LINE_SIZE);
saved_command_line[COMMAND_LINE_SIZE-1] = '\0';
for (;;) {
@@ -1124,6 +1125,7 @@ void __init setup_arch(char **cmdline_p)
max_low_pfn = setup_memory();
+ identify_cpu(&boot_cpu_data);
/*
* NOTE: before this point _nobody_ is allowed to allocate
* any memory using the bootmem allocator.
--- diff/arch/i386/kernel/signal.c 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/i386/kernel/signal.c 2004-03-16 11:27:47.890907136 +0000
@@ -128,28 +128,29 @@ sys_sigaltstack(const stack_t __user *us
*/
static int
-restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *peax)
+restore_sigcontext(struct pt_regs *regs,
+ struct sigcontext __user *__sc, int *peax)
{
- unsigned int err = 0;
+ struct sigcontext scratch; /* 88 bytes of scratch area */
/* Always make any pending restarted system calls return -EINTR */
current_thread_info()->restart_block.fn = do_no_restart_syscall;
-#define COPY(x) err |= __get_user(regs->x, &sc->x)
+ if (copy_from_user(&scratch, __sc, sizeof(scratch)))
+ return -EFAULT;
+
+#define COPY(x) regs->x = scratch.x
#define COPY_SEG(seg) \
- { unsigned short tmp; \
- err |= __get_user(tmp, &sc->seg); \
+ { unsigned short tmp = scratch.seg; \
regs->x##seg = tmp; }
#define COPY_SEG_STRICT(seg) \
- { unsigned short tmp; \
- err |= __get_user(tmp, &sc->seg); \
+ { unsigned short tmp = scratch.seg; \
regs->x##seg = tmp|3; }
#define GET_SEG(seg) \
- { unsigned short tmp; \
- err |= __get_user(tmp, &sc->seg); \
+ { unsigned short tmp = scratch.seg; \
loadsegment(seg,tmp); }
GET_SEG(gs);
@@ -168,27 +169,23 @@ restore_sigcontext(struct pt_regs *regs,
COPY_SEG_STRICT(ss);
{
- unsigned int tmpflags;
- err |= __get_user(tmpflags, &sc->eflags);
+ unsigned int tmpflags = scratch.eflags;
regs->eflags = (regs->eflags & ~0x40DD5) | (tmpflags & 0x40DD5);
regs->orig_eax = -1; /* disable syscall checks */
}
{
- struct _fpstate __user * buf;
- err |= __get_user(buf, &sc->fpstate);
+ struct _fpstate * buf = scratch.fpstate;
if (buf) {
if (verify_area(VERIFY_READ, buf, sizeof(*buf)))
- goto badframe;
- err |= restore_i387(buf);
+ return -EFAULT;
+ if (restore_i387(buf))
+ return -EFAULT;
}
}
- err |= __get_user(*peax, &sc->eax);
- return err;
-
-badframe:
- return 1;
+ *peax = scratch.eax;
+ return 0;
}
asmlinkage int sys_sigreturn(unsigned long __unused)
@@ -266,46 +263,47 @@ badframe:
*/
static int
-setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate,
+setup_sigcontext(struct sigcontext __user *__sc, struct _fpstate __user *fpstate,
struct pt_regs *regs, unsigned long mask)
{
- int tmp, err = 0;
+ struct sigcontext sc; /* 88 bytes of scratch area */
+ int tmp;
tmp = 0;
__asm__("movl %%gs,%0" : "=r"(tmp): "0"(tmp));
- err |= __put_user(tmp, (unsigned int *)&sc->gs);
+ *(unsigned int *)&sc.gs = tmp;
__asm__("movl %%fs,%0" : "=r"(tmp): "0"(tmp));
- err |= __put_user(tmp, (unsigned int *)&sc->fs);
-
- err |= __put_user(regs->xes, (unsigned int *)&sc->es);
- err |= __put_user(regs->xds, (unsigned int *)&sc->ds);
- err |= __put_user(regs->edi, &sc->edi);
- err |= __put_user(regs->esi, &sc->esi);
- err |= __put_user(regs->ebp, &sc->ebp);
- err |= __put_user(regs->esp, &sc->esp);
- err |= __put_user(regs->ebx, &sc->ebx);
- err |= __put_user(regs->edx, &sc->edx);
- err |= __put_user(regs->ecx, &sc->ecx);
- err |= __put_user(regs->eax, &sc->eax);
- err |= __put_user(current->thread.trap_no, &sc->trapno);
- err |= __put_user(current->thread.error_code, &sc->err);
- err |= __put_user(regs->eip, &sc->eip);
- err |= __put_user(regs->xcs, (unsigned int *)&sc->cs);
- err |= __put_user(regs->eflags, &sc->eflags);
- err |= __put_user(regs->esp, &sc->esp_at_signal);
- err |= __put_user(regs->xss, (unsigned int *)&sc->ss);
+ *(unsigned int *)&sc.fs = tmp;
+ *(unsigned int *)&sc.es = regs->xes;
+ *(unsigned int *)&sc.ds = regs->xds;
+ sc.edi = regs->edi;
+ sc.esi = regs->esi;
+ sc.ebp = regs->ebp;
+ sc.esp = regs->esp;
+ sc.ebx = regs->ebx;
+ sc.edx = regs->edx;
+ sc.ecx = regs->ecx;
+ sc.eax = regs->eax;
+ sc.trapno = current->thread.trap_no;
+ sc.err = current->thread.error_code;
+ sc.eip = regs->eip;
+ *(unsigned int *)&sc.cs = regs->xcs;
+ sc.eflags = regs->eflags;
+ sc.esp_at_signal = regs->esp;
+ *(unsigned int *)&sc.ss = regs->xss;
tmp = save_i387(fpstate);
if (tmp < 0)
- err = 1;
- else
- err |= __put_user(tmp ? fpstate : NULL, &sc->fpstate);
+ return 1;
+ sc.fpstate = tmp ? fpstate : NULL;
/* non-iBCS2 extensions.. */
- err |= __put_user(mask, &sc->oldmask);
- err |= __put_user(current->thread.cr2, &sc->cr2);
+ sc.oldmask = mask;
+ sc.cr2 = current->thread.cr2;
- return err;
+ if (copy_to_user(__sc, &sc, sizeof(sc)))
+ return 1;
+ return 0;
}
/*
@@ -443,7 +441,7 @@ static void setup_rt_frame(int sig, stru
/* Create the ucontext. */
err |= __put_user(0, &frame->uc.uc_flags);
err |= __put_user(0, &frame->uc.uc_link);
- err |= __put_user(current->sas_ss_sp, &frame->uc.uc_stack.ss_sp);
+ err |= __put_user(current->sas_ss_sp, (unsigned long *)&frame->uc.uc_stack.ss_sp);
err |= __put_user(sas_ss_flags(regs->esp),
&frame->uc.uc_stack.ss_flags);
err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size);
--- diff/arch/i386/kernel/smp.c 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/i386/kernel/smp.c 2004-03-16 11:27:47.892906832 +0000
@@ -327,10 +327,12 @@ asmlinkage void smp_invalidate_interrupt
if (flush_mm == cpu_tlbstate[cpu].active_mm) {
if (cpu_tlbstate[cpu].state == TLBSTATE_OK) {
+#ifndef CONFIG_X86_SWITCH_PAGETABLES
if (flush_va == FLUSH_ALL)
local_flush_tlb();
else
__flush_tlb_one(flush_va);
+#endif
} else
leave_mm(cpu);
}
@@ -396,21 +398,6 @@ static void flush_tlb_others(cpumask_t c
spin_unlock(&tlbstate_lock);
}
-void flush_tlb_current_task(void)
-{
- struct mm_struct *mm = current->mm;
- cpumask_t cpu_mask;
-
- preempt_disable();
- cpu_mask = mm->cpu_vm_mask;
- cpu_clear(smp_processor_id(), cpu_mask);
-
- local_flush_tlb();
- if (!cpus_empty(cpu_mask))
- flush_tlb_others(cpu_mask, mm, FLUSH_ALL);
- preempt_enable();
-}
-
void flush_tlb_mm (struct mm_struct * mm)
{
cpumask_t cpu_mask;
@@ -442,7 +429,10 @@ void flush_tlb_page(struct vm_area_struc
if (current->active_mm == mm) {
if(current->mm)
- __flush_tlb_one(va);
+#ifndef CONFIG_X86_SWITCH_PAGETABLES
+ __flush_tlb_one(va)
+#endif
+ ;
else
leave_mm(smp_processor_id());
}
@@ -466,7 +456,17 @@ void flush_tlb_all(void)
{
on_each_cpu(do_flush_tlb_all, 0, 1, 1);
}
-
+#ifdef CONFIG_KGDB
+/*
+ * By using the NMI code instead of a vector we just sneak thru the
+ * word generator coming out with just what we want. AND it does
+ * not matter if clustered_apic_mode is set or not.
+ */
+void smp_send_nmi_allbutself(void)
+{
+ send_IPI_allbutself(APIC_DM_NMI);
+}
+#endif
/*
* this function sends a 'reschedule' IPI to another CPU.
* it goes straight through and wastes no time serializing
--- diff/arch/i386/kernel/smpboot.c 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/i386/kernel/smpboot.c 2004-03-16 11:27:47.891906984 +0000
@@ -39,6 +39,7 @@
#include <linux/kernel.h>
#include <linux/mm.h>
+#include <linux/sched.h>
#include <linux/kernel_stat.h>
#include <linux/smp_lock.h>
#include <linux/irq.h>
@@ -815,6 +816,8 @@ static int __init do_boot_cpu(int apicid
/* Stack for startup_32 can be just as for start_secondary onwards */
stack_start.esp = (void *) idle->thread.esp;
+ irq_ctx_init(cpu);
+
/*
* This grunge runs the startup process for
* the targeted processor.
@@ -934,7 +937,7 @@ static int boot_cpu_logical_apicid;
/* Where the IO area was mapped on multiquad, always 0 otherwise */
void *xquad_portio;
-int cpu_sibling_map[NR_CPUS] __cacheline_aligned;
+cpumask_t cpu_sibling_map[NR_CPUS] __cacheline_aligned;
static void __init smp_boot_cpus(unsigned int max_cpus)
{
@@ -953,6 +956,8 @@ static void __init smp_boot_cpus(unsigne
current_thread_info()->cpu = 0;
smp_tune_scheduling();
+ cpus_clear(cpu_sibling_map[0]);
+ cpu_set(0, cpu_sibling_map[0]);
/*
* If we couldn't find an SMP configuration at boot time,
@@ -1079,32 +1084,34 @@ static void __init smp_boot_cpus(unsigne
Dprintk("Boot done.\n");
/*
- * If Hyper-Threading is avaialble, construct cpu_sibling_map[], so
- * that we can tell the sibling CPU efficiently.
+ * construct cpu_sibling_map[], so that we can tell sibling CPUs
+ * efficiently.
*/
- if (cpu_has_ht && smp_num_siblings > 1) {
- for (cpu = 0; cpu < NR_CPUS; cpu++)
- cpu_sibling_map[cpu] = NO_PROC_ID;
-
- for (cpu = 0; cpu < NR_CPUS; cpu++) {
- int i;
- if (!cpu_isset(cpu, cpu_callout_map))
- continue;
+ for (cpu = 0; cpu < NR_CPUS; cpu++)
+ cpus_clear(cpu_sibling_map[cpu]);
+ for (cpu = 0; cpu < NR_CPUS; cpu++) {
+ int siblings = 0;
+ int i;
+ if (!cpu_isset(cpu, cpu_callout_map))
+ continue;
+
+ if (smp_num_siblings > 1) {
for (i = 0; i < NR_CPUS; i++) {
- if (i == cpu || !cpu_isset(i, cpu_callout_map))
+ if (!cpu_isset(i, cpu_callout_map))
continue;
if (phys_proc_id[cpu] == phys_proc_id[i]) {
- cpu_sibling_map[cpu] = i;
- printk("cpu_sibling_map[%d] = %d\n", cpu, cpu_sibling_map[cpu]);
- break;
+ siblings++;
+ cpu_set(i, cpu_sibling_map[cpu]);
}
}
- if (cpu_sibling_map[cpu] == NO_PROC_ID) {
- smp_num_siblings = 1;
- printk(KERN_WARNING "WARNING: No sibling found for CPU %d.\n", cpu);
- }
+ } else {
+ siblings++;
+ cpu_set(cpu, cpu_sibling_map[cpu]);
}
+
+ if (siblings != smp_num_siblings)
+ printk(KERN_WARNING "WARNING: %d siblings found for CPU%d, should be %d\n", siblings, cpu, smp_num_siblings);
}
smpboot_setup_io_apic();
@@ -1118,6 +1125,224 @@ static void __init smp_boot_cpus(unsigne
synchronize_tsc_bp();
}
+#ifdef CONFIG_SCHED_SMT
+#ifdef CONFIG_NUMA
+static struct sched_group sched_group_cpus[NR_CPUS];
+static struct sched_group sched_group_phys[NR_CPUS];
+static struct sched_group sched_group_nodes[MAX_NUMNODES];
+static DEFINE_PER_CPU(struct sched_domain, phys_domains);
+static DEFINE_PER_CPU(struct sched_domain, node_domains);
+__init void arch_init_sched_domains(void)
+{
+ int i;
+ struct sched_group *first_cpu = NULL, *last_cpu = NULL;
+
+ /* Set up domains */
+ for_each_cpu(i) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ struct sched_domain *phys_domain = &per_cpu(phys_domains, i);
+ struct sched_domain *node_domain = &per_cpu(node_domains, i);
+ int node = cpu_to_node(i);
+ cpumask_t nodemask = node_to_cpumask(node);
+
+ *cpu_domain = SD_SIBLING_INIT;
+ cpu_domain->span = cpu_sibling_map[i];
+
+ *phys_domain = SD_CPU_INIT;
+ phys_domain->span = nodemask;
+
+ *node_domain = SD_NODE_INIT;
+ node_domain->span = cpu_possible_map;
+ }
+
+ /* Set up CPU (sibling) groups */
+ for_each_cpu(i) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ int j;
+ first_cpu = last_cpu = NULL;
+
+ if (i != first_cpu(cpu_domain->span)) {
+ cpu_sched_domain(i)->flags |= SD_FLAG_SHARE_CPUPOWER;
+ cpu_sched_domain(first_cpu(cpu_domain->span))->flags |=
+ SD_FLAG_SHARE_CPUPOWER;
+ continue;
+ }
+
+ for_each_cpu_mask(j, cpu_domain->span) {
+ struct sched_group *cpu = &sched_group_cpus[j];
+
+ cpu->cpumask = CPU_MASK_NONE;
+ cpu_set(j, cpu->cpumask);
+ cpu->cpu_power = SCHED_LOAD_SCALE;
+
+ if (!first_cpu)
+ first_cpu = cpu;
+ if (last_cpu)
+ last_cpu->next = cpu;
+ last_cpu = cpu;
+ }
+ last_cpu->next = first_cpu;
+ }
+
+ for (i = 0; i < MAX_NUMNODES; i++) {
+ int j;
+ cpumask_t nodemask;
+ struct sched_group *node = &sched_group_nodes[i];
+ cpus_and(nodemask, node_to_cpumask(i), cpu_possible_map);
+
+ if (cpus_empty(nodemask))
+ continue;
+
+ first_cpu = last_cpu = NULL;
+ /* Set up physical groups */
+ for_each_cpu_mask(j, nodemask) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(j);
+ struct sched_group *cpu = &sched_group_phys[j];
+
+ if (j != first_cpu(cpu_domain->span))
+ continue;
+
+ cpu->cpumask = cpu_domain->span;
+ /*
+ * Make each extra sibling increase power by 10% of
+ * the basic CPU. This is very arbitrary.
+ */
+ cpu->cpu_power = SCHED_LOAD_SCALE + SCHED_LOAD_SCALE*(cpus_weight(cpu->cpumask)-1) / 10;
+ node->cpu_power += cpu->cpu_power;
+
+ if (!first_cpu)
+ first_cpu = cpu;
+ if (last_cpu)
+ last_cpu->next = cpu;
+ last_cpu = cpu;
+ }
+ last_cpu->next = first_cpu;
+ }
+
+ /* Set up nodes */
+ first_cpu = last_cpu = NULL;
+ for (i = 0; i < MAX_NUMNODES; i++) {
+ struct sched_group *cpu = &sched_group_nodes[i];
+ cpumask_t nodemask;
+ cpus_and(nodemask, node_to_cpumask(i), cpu_possible_map);
+
+ if (cpus_empty(nodemask))
+ continue;
+
+ cpu->cpumask = nodemask;
+ /* ->cpu_power already setup */
+
+ if (!first_cpu)
+ first_cpu = cpu;
+ if (last_cpu)
+ last_cpu->next = cpu;
+ last_cpu = cpu;
+ }
+ last_cpu->next = first_cpu;
+
+ mb();
+ for_each_cpu(i) {
+ int node = cpu_to_node(i);
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ struct sched_domain *phys_domain = &per_cpu(phys_domains, i);
+ struct sched_domain *node_domain = &per_cpu(node_domains, i);
+ struct sched_group *cpu_group = &sched_group_cpus[i];
+ struct sched_group *phys_group = &sched_group_phys[first_cpu(cpu_domain->span)];
+ struct sched_group *node_group = &sched_group_nodes[node];
+
+ cpu_domain->parent = phys_domain;
+ phys_domain->parent = node_domain;
+
+ node_domain->groups = node_group;
+ phys_domain->groups = phys_group;
+ cpu_domain->groups = cpu_group;
+ }
+}
+#else /* CONFIG_NUMA */
+static struct sched_group sched_group_cpus[NR_CPUS];
+static struct sched_group sched_group_phys[NR_CPUS];
+static DEFINE_PER_CPU(struct sched_domain, phys_domains);
+__init void arch_init_sched_domains(void)
+{
+ int i;
+ struct sched_group *first_cpu = NULL, *last_cpu = NULL;
+
+ /* Set up domains */
+ for_each_cpu(i) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ struct sched_domain *phys_domain = &per_cpu(phys_domains, i);
+
+ *cpu_domain = SD_SIBLING_INIT;
+ cpu_domain->span = cpu_sibling_map[i];
+
+ *phys_domain = SD_CPU_INIT;
+ phys_domain->span = cpu_possible_map;
+ }
+
+ /* Set up CPU (sibling) groups */
+ for_each_cpu(i) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ int j;
+ first_cpu = last_cpu = NULL;
+
+ if (i != first_cpu(cpu_domain->span)) {
+ cpu_sched_domain(i)->flags |= SD_FLAG_SHARE_CPUPOWER;
+ cpu_sched_domain(first_cpu(cpu_domain->span))->flags |=
+ SD_FLAG_SHARE_CPUPOWER;
+ continue;
+ }
+
+ for_each_cpu_mask(j, cpu_domain->span) {
+ struct sched_group *cpu = &sched_group_cpus[j];
+
+ cpus_clear(cpu->cpumask);
+ cpu_set(j, cpu->cpumask);
+ cpu->cpu_power = SCHED_LOAD_SCALE;
+
+ if (!first_cpu)
+ first_cpu = cpu;
+ if (last_cpu)
+ last_cpu->next = cpu;
+ last_cpu = cpu;
+ }
+ last_cpu->next = first_cpu;
+ }
+
+ first_cpu = last_cpu = NULL;
+ /* Set up physical groups */
+ for_each_cpu(i) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ struct sched_group *cpu = &sched_group_phys[i];
+
+ if (i != first_cpu(cpu_domain->span))
+ continue;
+
+ cpu->cpumask = cpu_domain->span;
+ /* See SMT+NUMA setup for comment */
+ cpu->cpu_power = SCHED_LOAD_SCALE + SCHED_LOAD_SCALE*(cpus_weight(cpu->cpumask)-1) / 10;
+
+ if (!first_cpu)
+ first_cpu = cpu;
+ if (last_cpu)
+ last_cpu->next = cpu;
+ last_cpu = cpu;
+ }
+ last_cpu->next = first_cpu;
+
+ mb();
+ for_each_cpu(i) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ struct sched_domain *phys_domain = &per_cpu(phys_domains, i);
+ struct sched_group *cpu_group = &sched_group_cpus[i];
+ struct sched_group *phys_group = &sched_group_phys[first_cpu(cpu_domain->span)];
+ cpu_domain->parent = phys_domain;
+ phys_domain->groups = phys_group;
+ cpu_domain->groups = cpu_group;
+ }
+}
+#endif /* CONFIG_NUMA */
+#endif /* CONFIG_SCHED_SMT */
+
/* These are wrappers to interface to the new boot process. Someone
who understands all this stuff should rewrite it properly. --RR 15/Jul/02 */
void __init smp_prepare_cpus(unsigned int max_cpus)
--- diff/arch/i386/kernel/sysenter.c 2003-10-09 08:47:16.000000000 +0000
+++ source/arch/i386/kernel/sysenter.c 2004-03-16 11:27:47.892906832 +0000
@@ -18,13 +18,18 @@
#include <asm/msr.h>
#include <asm/pgtable.h>
#include <asm/unistd.h>
+#include <linux/highmem.h>
extern asmlinkage void sysenter_entry(void);
void enable_sep_cpu(void *info)
{
int cpu = get_cpu();
+#ifdef CONFIG_X86_HIGH_ENTRY
+ struct tss_struct *tss = (struct tss_struct *) __fix_to_virt(FIX_TSS_0) + cpu;
+#else
struct tss_struct *tss = init_tss + cpu;
+#endif
tss->ss1 = __KERNEL_CS;
tss->esp1 = sizeof(struct tss_struct) + (unsigned long) tss;
--- diff/arch/i386/kernel/traps.c 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/i386/kernel/traps.c 2004-03-16 11:27:47.894906528 +0000
@@ -55,12 +55,8 @@
#include "mach_traps.h"
-asmlinkage int system_call(void);
-asmlinkage void lcall7(void);
-asmlinkage void lcall27(void);
-
-struct desc_struct default_ldt[] = { { 0, 0 }, { 0, 0 }, { 0, 0 },
- { 0, 0 }, { 0, 0 } };
+struct desc_struct default_ldt[] __attribute__((__section__(".data.default_ldt"))) = { { 0, 0 }, { 0, 0 }, { 0, 0 }, { 0, 0 }, { 0, 0 } };
+struct page *default_ldt_page;
/* Do we ignore FPU interrupts ? */
char ignore_fpu_irq = 0;
@@ -92,6 +88,41 @@ asmlinkage void alignment_check(void);
asmlinkage void spurious_interrupt_bug(void);
asmlinkage void machine_check(void);
+#ifdef CONFIG_KGDB
+extern void sysenter_entry(void);
+#include <asm/kgdb.h>
+#include <linux/init.h>
+void set_intr_gate(unsigned int n, void *addr);
+static void set_intr_usr_gate(unsigned int n, void *addr);
+/*
+ * Should be able to call this breakpoint() very early in
+ * bring up. Just hard code the call where needed.
+ * The breakpoint() code is here because set_?_gate() functions
+ * are local (static) to trap.c. They need be done only once,
+ * but it does not hurt to do them over.
+ */
+void breakpoint(void)
+{
+ init_entry_mappings();
+ set_intr_usr_gate(3,&int3); /* disable ints on trap */
+ set_intr_gate(1,&debug);
+ set_intr_gate(14,&page_fault);
+
+ BREAKPOINT;
+}
+#define CHK_REMOTE_DEBUG(trapnr,signr,error_code,regs,after) \
+ { \
+ if (!user_mode(regs) ) \
+ { \
+ kgdb_handle_exception(trapnr, signr, error_code, regs); \
+ after; \
+ } else if ((trapnr == 3) && (regs->eflags &0x200)) local_irq_enable(); \
+ }
+#else
+#define CHK_REMOTE_DEBUG(trapnr,signr,error_code,regs,after)
+#endif
+
+
static int kstack_depth_to_print = 24;
void show_trace(struct task_struct *task, unsigned long * stack)
@@ -105,12 +136,20 @@ void show_trace(struct task_struct *task
#ifdef CONFIG_KALLSYMS
printk("\n");
#endif
- while (!kstack_end(stack)) {
- addr = *stack++;
- if (kernel_text_address(addr)) {
- printk(" [<%08lx>] ", addr);
- print_symbol("%s\n", addr);
+ while (1) {
+ struct thread_info *context;
+ context = (struct thread_info*) ((unsigned long)stack & (~(THREAD_SIZE - 1)));
+ while (!kstack_end(stack)) {
+ addr = *stack++;
+ if (kernel_text_address(addr)) {
+ printk(" [<%08lx>] ", addr);
+ print_symbol("%s\n", addr);
+ }
}
+ stack = (unsigned long*)context->previous_esp;
+ if (!stack)
+ break;
+ printk(" =======================\n");
}
printk("\n");
}
@@ -176,7 +215,7 @@ void show_registers(struct pt_regs *regs
ss = regs->xss & 0xffff;
}
print_modules();
- printk("CPU: %d\nEIP: %04x:[<%08lx>] %s\nEFLAGS: %08lx"
+ printk("CPU: %d\nEIP: %04x:[<%08lx>] %s VLI\nEFLAGS: %08lx"
" (%s) \n",
smp_processor_id(), 0xffff & regs->xcs, regs->eip,
print_tainted(), regs->eflags, UTS_RELEASE);
@@ -194,23 +233,27 @@ void show_registers(struct pt_regs *regs
* time of the fault..
*/
if (in_kernel) {
+ u8 *eip;
printk("\nStack: ");
show_stack(NULL, (unsigned long*)esp);
printk("Code: ");
- if(regs->eip < PAGE_OFFSET)
- goto bad;
- for(i=0;i<20;i++)
- {
- unsigned char c;
- if(__get_user(c, &((unsigned char*)regs->eip)[i])) {
-bad:
+ eip = (u8 *)regs->eip - 43;
+ for (i = 0; i < 64; i++, eip++) {
+ unsigned char c = 0xff;
+
+ if ((user_mode(regs) && get_user(c, eip)) ||
+ (!user_mode(regs) && __direct_get_user(c, eip))) {
+
printk(" Bad EIP value.");
break;
}
- printk("%02x ", c);
+ if (eip == (u8 *)regs->eip)
+ printk("<%02x> ", c);
+ else
+ printk("%02x ", c);
}
}
printk("\n");
@@ -229,16 +272,14 @@ static void handle_BUG(struct pt_regs *r
eip = regs->eip;
- if (eip < PAGE_OFFSET)
- goto no_bug;
- if (__get_user(ud2, (unsigned short *)eip))
+ if (__direct_get_user(ud2, (unsigned short *)eip))
goto no_bug;
if (ud2 != 0x0b0f)
goto no_bug;
- if (__get_user(line, (unsigned short *)(eip + 2)))
+ if (__direct_get_user(line, (unsigned short *)(eip + 2)))
goto bug;
- if (__get_user(file, (char **)(eip + 4)) ||
- (unsigned long)file < PAGE_OFFSET || __get_user(c, file))
+ if (__direct_get_user(file, (char **)(eip + 4)) ||
+ __direct_get_user(c, file))
file = "<bad filename>";
printk("------------[ cut here ]------------\n");
@@ -278,6 +319,15 @@ void die(const char * str, struct pt_reg
#endif
if (nl)
printk("\n");
+#ifdef CONFIG_KGDB
+ /* This is about the only place we want to go to kgdb even if in
+ * user mode. But we must go in via a trap so within kgdb we will
+ * always be in kernel mode.
+ */
+ if (user_mode(regs))
+ BREAKPOINT;
+#endif
+ CHK_REMOTE_DEBUG(0,SIGTRAP,err,regs,)
show_registers(regs);
bust_spinlocks(0);
spin_unlock_irq(&die_lock);
@@ -347,6 +397,7 @@ static inline void do_trap(int trapnr, i
#define DO_ERROR(trapnr, signr, str, name) \
asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
{ \
+ CHK_REMOTE_DEBUG(trapnr,signr,error_code,regs,)\
do_trap(trapnr, signr, str, 0, regs, error_code, NULL); \
}
@@ -364,7 +415,9 @@ asmlinkage void do_##name(struct pt_regs
#define DO_VM86_ERROR(trapnr, signr, str, name) \
asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
{ \
+ CHK_REMOTE_DEBUG(trapnr, signr, error_code,regs, return)\
do_trap(trapnr, signr, str, 1, regs, error_code, NULL); \
+ return; \
}
#define DO_VM86_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \
@@ -411,8 +464,10 @@ gp_in_vm86:
return;
gp_in_kernel:
- if (!fixup_exception(regs))
+ if (!fixup_exception(regs)){
+ CHK_REMOTE_DEBUG(13,SIGSEGV,error_code,regs,)
die("general protection fault", regs, error_code);
+ }
}
static void mem_parity_error(unsigned char reason, struct pt_regs * regs)
@@ -551,10 +606,18 @@ asmlinkage void do_debug(struct pt_regs
if (regs->eflags & X86_EFLAGS_IF)
local_irq_enable();
- /* Mask out spurious debug traps due to lazy DR7 setting */
+ /*
+ * Mask out spurious debug traps due to lazy DR7 setting or
+ * due to 4G/4G kernel mode:
+ */
if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) {
if (!tsk->thread.debugreg[7])
goto clear_dr7;
+ if (!user_mode(regs)) {
+ // restore upon return-to-userspace:
+ set_thread_flag(TIF_DB7);
+ goto clear_dr7;
+ }
}
if (regs->eflags & VM_MASK)
@@ -574,8 +637,18 @@ asmlinkage void do_debug(struct pt_regs
* allowing programs to debug themselves without the ptrace()
* interface.
*/
+#ifdef CONFIG_KGDB
+ /*
+ * I think this is the only "real" case of a TF in the kernel
+ * that really belongs to user space. Others are
+ * "Ours all ours!"
+ */
+ if (((regs->xcs & 3) == 0) && ((void *)regs->eip == sysenter_entry))
+ goto clear_TF_reenable;
+#else
if ((regs->xcs & 3) == 0)
goto clear_TF_reenable;
+#endif
if ((tsk->ptrace & (PT_DTRACE|PT_PTRACED)) == PT_DTRACE)
goto clear_TF;
}
@@ -587,6 +660,17 @@ asmlinkage void do_debug(struct pt_regs
info.si_errno = 0;
info.si_code = TRAP_BRKPT;
+#ifdef CONFIG_KGDB
+ /*
+ * If this is a kernel mode trap, we need to reset db7 to allow us
+ * to continue sanely ALSO skip the signal delivery
+ */
+ if ((regs->xcs & 3) == 0)
+ goto clear_dr7;
+
+ /* if not kernel, allow ints but only if they were on */
+ if ( regs->eflags & 0x200) local_irq_enable();
+#endif
/* If this is a kernel mode trap, save the user PC on entry to
* the kernel, that's what the debugger can make sense of.
*/
@@ -601,6 +685,7 @@ clear_dr7:
__asm__("movl %0,%%db7"
: /* no output */
: "r" (0));
+ CHK_REMOTE_DEBUG(1,SIGTRAP,error_code,regs,)
return;
debug_vm86:
@@ -796,19 +881,53 @@ asmlinkage void math_emulate(long arg)
#endif /* CONFIG_MATH_EMULATION */
-#ifdef CONFIG_X86_F00F_BUG
-void __init trap_init_f00f_bug(void)
+void __init trap_init_virtual_IDT(void)
{
- __set_fixmap(FIX_F00F_IDT, __pa(&idt_table), PAGE_KERNEL_RO);
-
/*
- * Update the IDT descriptor and reload the IDT so that
- * it uses the read-only mapped virtual address.
+ * "idt" is magic - it overlaps the idt_descr
+ * variable so that updating idt will automatically
+ * update the idt descriptor..
*/
- idt_descr.address = fix_to_virt(FIX_F00F_IDT);
+ __set_fixmap(FIX_IDT, __pa(&idt_table), PAGE_KERNEL_RO);
+ idt_descr.address = __fix_to_virt(FIX_IDT);
+
__asm__ __volatile__("lidt %0" : : "m" (idt_descr));
}
-#endif
+
+void __init trap_init_virtual_GDT(void)
+{
+ int cpu = smp_processor_id();
+ struct Xgt_desc_struct *gdt_desc = cpu_gdt_descr + cpu;
+ struct Xgt_desc_struct tmp_desc = {0, 0};
+ struct tss_struct * t;
+
+ __asm__ __volatile__("sgdt %0": "=m" (tmp_desc): :"memory");
+
+#ifdef CONFIG_X86_HIGH_ENTRY
+ if (!cpu) {
+ __set_fixmap(FIX_GDT_0, __pa(cpu_gdt_table), PAGE_KERNEL);
+ __set_fixmap(FIX_GDT_1, __pa(cpu_gdt_table) + PAGE_SIZE, PAGE_KERNEL);
+ __set_fixmap(FIX_TSS_0, __pa(init_tss), PAGE_KERNEL);
+ __set_fixmap(FIX_TSS_1, __pa(init_tss) + 1*PAGE_SIZE, PAGE_KERNEL);
+ __set_fixmap(FIX_TSS_2, __pa(init_tss) + 2*PAGE_SIZE, PAGE_KERNEL);
+ __set_fixmap(FIX_TSS_3, __pa(init_tss) + 3*PAGE_SIZE, PAGE_KERNEL);
+ }
+
+ gdt_desc->address = __fix_to_virt(FIX_GDT_0) + sizeof(cpu_gdt_table[0]) * cpu;
+#else
+ gdt_desc->address = (unsigned long)cpu_gdt_table[cpu];
+#endif
+ __asm__ __volatile__("lgdt %0": "=m" (*gdt_desc));
+
+#ifdef CONFIG_X86_HIGH_ENTRY
+ t = (struct tss_struct *) __fix_to_virt(FIX_TSS_0) + cpu;
+#else
+ t = init_tss + cpu;
+#endif
+ set_tss_desc(cpu, t);
+ cpu_gdt_table[cpu][GDT_ENTRY_TSS].b &= 0xfffffdff;
+ load_TR_desc();
+}
#define _set_gate(gate_addr,type,dpl,addr,seg) \
do { \
@@ -835,20 +954,26 @@ void set_intr_gate(unsigned int n, void
_set_gate(idt_table+n,14,0,addr,__KERNEL_CS);
}
-static void __init set_trap_gate(unsigned int n, void *addr)
+void __init set_trap_gate(unsigned int n, void *addr)
{
_set_gate(idt_table+n,15,0,addr,__KERNEL_CS);
}
-static void __init set_system_gate(unsigned int n, void *addr)
+void __init set_system_gate(unsigned int n, void *addr)
{
_set_gate(idt_table+n,15,3,addr,__KERNEL_CS);
}
-static void __init set_call_gate(void *a, void *addr)
+void __init set_call_gate(void *a, void *addr)
{
_set_gate(a,12,3,addr,__KERNEL_CS);
}
+#ifdef CONFIG_KGDB
+void set_intr_usr_gate(unsigned int n, void *addr)
+{
+ _set_gate(idt_table+n,14,3,addr,__KERNEL_CS);
+}
+#endif
static void __init set_task_gate(unsigned int n, unsigned int gdt_entry)
{
@@ -867,11 +992,16 @@ void __init trap_init(void)
#ifdef CONFIG_X86_LOCAL_APIC
init_apic_mappings();
#endif
+ init_entry_mappings();
set_trap_gate(0,÷_error);
set_intr_gate(1,&debug);
set_intr_gate(2,&nmi);
+#ifndef CONFIG_KGDB
set_system_gate(3,&int3); /* int3-5 can be called from all */
+#else
+ set_intr_usr_gate(3,&int3); /* int3-5 can be called from all */
+#endif
set_system_gate(4,&overflow);
set_system_gate(5,&bounds);
set_trap_gate(6,&invalid_op);
--- diff/arch/i386/kernel/vm86.c 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/i386/kernel/vm86.c 2004-03-16 11:27:47.894906528 +0000
@@ -125,7 +125,7 @@ struct pt_regs * fastcall save_v86_state
tss = init_tss + get_cpu();
current->thread.esp0 = current->thread.saved_esp0;
current->thread.sysenter_cs = __KERNEL_CS;
- load_esp0(tss, ¤t->thread);
+ load_virtual_esp0(tss, current);
current->thread.saved_esp0 = 0;
put_cpu();
@@ -305,7 +305,7 @@ static void do_sys_vm86(struct kernel_vm
tsk->thread.esp0 = (unsigned long) &info->VM86_TSS_ESP0;
if (cpu_has_sep)
tsk->thread.sysenter_cs = 0;
- load_esp0(tss, &tsk->thread);
+ load_virtual_esp0(tss, tsk);
put_cpu();
tsk->thread.screen_bitmap = info->screen_bitmap;
--- diff/arch/i386/kernel/vmlinux.lds.S 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/i386/kernel/vmlinux.lds.S 2004-03-16 11:27:47.895906376 +0000
@@ -3,14 +3,19 @@
*/
#include <asm-generic/vmlinux.lds.h>
-
+#include <asm/thread_info.h>
+
+#include <linux/config.h>
+#include <asm/page.h>
+#include <asm/asm_offsets.h>
+
OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386")
OUTPUT_ARCH(i386)
ENTRY(startup_32)
jiffies = jiffies_64;
SECTIONS
{
- . = 0xC0000000 + 0x100000;
+ . = __PAGE_OFFSET + 0x100000;
/* read-only */
_text = .; /* Text and read-only data */
.text : {
@@ -19,6 +24,19 @@ SECTIONS
*(.gnu.warning)
} = 0x9090
+#ifdef CONFIG_X86_4G
+ . = ALIGN(PAGE_SIZE_asm);
+ __entry_tramp_start = .;
+ . = FIX_ENTRY_TRAMPOLINE_0_addr;
+ __start___entry_text = .;
+ .entry.text : AT (__entry_tramp_start) { *(.entry.text) }
+ __entry_tramp_end = __entry_tramp_start + SIZEOF(.entry.text);
+ . = __entry_tramp_end;
+ . = ALIGN(PAGE_SIZE_asm);
+#else
+ .entry.text : { *(.entry.text) }
+#endif
+
_etext = .; /* End of text section */
. = ALIGN(16); /* Exception table */
@@ -34,25 +52,22 @@ SECTIONS
CONSTRUCTORS
}
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE_asm);
__nosave_begin = .;
.data_nosave : { *(.data.nosave) }
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE_asm);
__nosave_end = .;
- . = ALIGN(4096);
- .data.page_aligned : { *(.data.idt) }
-
. = ALIGN(32);
.data.cacheline_aligned : { *(.data.cacheline_aligned) }
_edata = .; /* End of data section */
- . = ALIGN(8192); /* init_task */
+ . = ALIGN(THREAD_SIZE); /* init_task */
.data.init_task : { *(.data.init_task) }
/* will be freed after init */
- . = ALIGN(4096); /* Init code and data */
+ . = ALIGN(PAGE_SIZE_asm); /* Init code and data */
__init_begin = .;
.init.text : {
_sinittext = .;
@@ -91,7 +106,7 @@ SECTIONS
from .altinstructions and .eh_frame */
.exit.text : { *(.exit.text) }
.exit.data : { *(.exit.data) }
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE_asm);
__initramfs_start = .;
.init.ramfs : { *(.init.ramfs) }
__initramfs_end = .;
@@ -99,10 +114,22 @@ SECTIONS
__per_cpu_start = .;
.data.percpu : { *(.data.percpu) }
__per_cpu_end = .;
- . = ALIGN(4096);
+ . = ALIGN(PAGE_SIZE_asm);
__init_end = .;
/* freed after init ends here */
-
+
+ . = ALIGN(PAGE_SIZE_asm);
+ .data.page_aligned_tss : { *(.data.tss) }
+
+ . = ALIGN(PAGE_SIZE_asm);
+ .data.page_aligned_default_ldt : { *(.data.default_ldt) }
+
+ . = ALIGN(PAGE_SIZE_asm);
+ .data.page_aligned_idt : { *(.data.idt) }
+
+ . = ALIGN(PAGE_SIZE_asm);
+ .data.page_aligned_gdt : { *(.data.gdt) }
+
__bss_start = .; /* BSS */
.bss : { *(.bss) }
. = ALIGN(4);
@@ -127,4 +154,6 @@ SECTIONS
.stab.index 0 : { *(.stab.index) }
.stab.indexstr 0 : { *(.stab.indexstr) }
.comment 0 : { *(.comment) }
+
+
}
--- diff/arch/i386/kernel/vsyscall-int80.S 2003-05-21 10:50:14.000000000 +0000
+++ source/arch/i386/kernel/vsyscall-int80.S 2004-03-16 11:27:47.895906376 +0000
@@ -1,5 +1,10 @@
/*
* Code for the vsyscall page. This version uses the old int $0x80 method.
+ *
+ * NOTE:
+ * 1) __kernel_vsyscall _must_ be first in this page.
+ * 2) there are alignment constraints on this stub, see vsyscall-sigreturn.S
+ * for details.
*/
.text
--- diff/arch/i386/kernel/vsyscall-sigreturn.S 2003-05-21 10:50:14.000000000 +0000
+++ source/arch/i386/kernel/vsyscall-sigreturn.S 2004-03-16 11:27:47.896906224 +0000
@@ -2,8 +2,8 @@
* Common code for the sigreturn entry points on the vsyscall page.
* So far this code is the same for both int80 and sysenter versions.
* This file is #include'd by vsyscall-*.S to define them after the
- * vsyscall entry point. The addresses we get for these entry points
- * by doing ".balign 32" must match in both versions of the page.
+ * vsyscall entry point. The kernel assumes that the addresses of these
+ * routines are constant for all vsyscall implementations.
*/
#include <asm/unistd.h>
@@ -15,7 +15,7 @@
*/
.text
- .balign 32
+ .org __kernel_vsyscall+32
.globl __kernel_sigreturn
.type __kernel_sigreturn,@function
__kernel_sigreturn:
--- diff/arch/i386/kernel/vsyscall-sysenter.S 2003-10-09 08:47:16.000000000 +0000
+++ source/arch/i386/kernel/vsyscall-sysenter.S 2004-03-16 11:27:47.896906224 +0000
@@ -1,5 +1,10 @@
/*
* Code for the vsyscall page. This version uses the sysenter instruction.
+ *
+ * NOTE:
+ * 1) __kernel_vsyscall _must_ be first in this page.
+ * 2) there are alignment constraints on this stub, see vsyscall-sigreturn.S
+ * for details.
*/
.text
@@ -7,6 +12,11 @@
.type __kernel_vsyscall,@function
__kernel_vsyscall:
.LSTART_vsyscall:
+ cmpl $192, %eax
+ jne 1f
+ int $0x80
+ ret
+1:
push %ecx
.Lpush_ecx:
push %edx
--- diff/arch/i386/kernel/vsyscall.lds 2003-10-09 08:47:16.000000000 +0000
+++ source/arch/i386/kernel/vsyscall.lds 2004-03-16 11:27:47.896906224 +0000
@@ -5,7 +5,7 @@
*/
/* This must match <asm/fixmap.h>. */
-VSYSCALL_BASE = 0xffffe000;
+VSYSCALL_BASE = 0xffffd000;
SECTIONS
{
--- diff/arch/i386/lib/Makefile 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/i386/lib/Makefile 2004-03-16 11:27:47.899905768 +0000
@@ -9,3 +9,4 @@ lib-y = checksum.o delay.o \
lib-$(CONFIG_X86_USE_3DNOW) += mmx.o
lib-$(CONFIG_HAVE_DEC_LOCK) += dec_and_lock.o
+lib-$(CONFIG_KGDB) += kgdb_serial.o
--- diff/arch/i386/lib/checksum.S 2003-10-09 08:47:16.000000000 +0000
+++ source/arch/i386/lib/checksum.S 2004-03-16 11:27:47.897906072 +0000
@@ -280,14 +280,14 @@ unsigned int csum_partial_copy_generic (
.previous
.align 4
-.globl csum_partial_copy_generic
+.globl direct_csum_partial_copy_generic
#ifndef CONFIG_X86_USE_PPRO_CHECKSUM
#define ARGBASE 16
#define FP 12
-csum_partial_copy_generic:
+direct_csum_partial_copy_generic:
subl $4,%esp
pushl %edi
pushl %esi
@@ -422,7 +422,7 @@ DST( movb %cl, (%edi) )
#define ARGBASE 12
-csum_partial_copy_generic:
+direct_csum_partial_copy_generic:
pushl %ebx
pushl %edi
pushl %esi
--- diff/arch/i386/lib/dec_and_lock.c 2003-10-09 08:47:16.000000000 +0000
+++ source/arch/i386/lib/dec_and_lock.c 2004-03-16 11:27:47.897906072 +0000
@@ -10,6 +10,7 @@
#include <linux/spinlock.h>
#include <asm/atomic.h>
+#ifndef ATOMIC_DEC_AND_LOCK
int atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock)
{
int counter;
@@ -38,3 +39,5 @@ slow_path:
spin_unlock(lock);
return 0;
}
+#endif
+
--- diff/arch/i386/lib/getuser.S 2003-10-09 08:47:16.000000000 +0000
+++ source/arch/i386/lib/getuser.S 2004-03-16 11:27:47.898905920 +0000
@@ -9,6 +9,7 @@
* return value.
*/
#include <asm/thread_info.h>
+#include <asm/asm_offsets.h>
/*
@@ -28,7 +29,7 @@
.globl __get_user_1
__get_user_1:
GET_THREAD_INFO(%edx)
- cmpl TI_ADDR_LIMIT(%edx),%eax
+ cmpl TI_addr_limit(%edx),%eax
jae bad_get_user
1: movzbl (%eax),%edx
xorl %eax,%eax
@@ -40,7 +41,7 @@ __get_user_2:
addl $1,%eax
jc bad_get_user
GET_THREAD_INFO(%edx)
- cmpl TI_ADDR_LIMIT(%edx),%eax
+ cmpl TI_addr_limit(%edx),%eax
jae bad_get_user
2: movzwl -1(%eax),%edx
xorl %eax,%eax
@@ -52,7 +53,7 @@ __get_user_4:
addl $3,%eax
jc bad_get_user
GET_THREAD_INFO(%edx)
- cmpl TI_ADDR_LIMIT(%edx),%eax
+ cmpl TI_addr_limit(%edx),%eax
jae bad_get_user
3: movl -3(%eax),%edx
xorl %eax,%eax
--- diff/arch/i386/lib/usercopy.c 2004-01-19 10:22:55.000000000 +0000
+++ source/arch/i386/lib/usercopy.c 2004-03-16 11:27:47.900905616 +0000
@@ -76,7 +76,7 @@ do { \
* and returns @count.
*/
long
-__strncpy_from_user(char *dst, const char __user *src, long count)
+__direct_strncpy_from_user(char *dst, const char __user *src, long count)
{
long res;
__do_strncpy_from_user(dst, src, count, res);
@@ -102,7 +102,7 @@ __strncpy_from_user(char *dst, const cha
* and returns @count.
*/
long
-strncpy_from_user(char *dst, const char __user *src, long count)
+direct_strncpy_from_user(char *dst, const char __user *src, long count)
{
long res = -EFAULT;
if (access_ok(VERIFY_READ, src, 1))
@@ -147,7 +147,7 @@ do { \
* On success, this will be zero.
*/
unsigned long
-clear_user(void __user *to, unsigned long n)
+direct_clear_user(void __user *to, unsigned long n)
{
might_sleep();
if (access_ok(VERIFY_WRITE, to, n))
@@ -167,7 +167,7 @@ clear_user(void __user *to, unsigned lon
* On success, this will be zero.
*/
unsigned long
-__clear_user(void __user *to, unsigned long n)
+__direct_clear_user(void __user *to, unsigned long n)
{
__do_clear_user(to, n);
return n;
@@ -184,7 +184,7 @@ __clear_user(void __user *to, unsigned l
* On exception, returns 0.
* If the string is too long, returns a value greater than @n.
*/
-long strnlen_user(const char __user *s, long n)
+long direct_strnlen_user(const char __user *s, long n)
{
unsigned long mask = -__addr_ok(s);
unsigned long res, tmp;
@@ -575,3 +575,4 @@ unsigned long __copy_from_user_ll(void *
n = __copy_user_zeroing_intel(to, (const void *) from, n);
return n;
}
+
--- diff/arch/i386/math-emu/fpu_system.h 2003-10-09 08:47:16.000000000 +0000
+++ source/arch/i386/math-emu/fpu_system.h 2004-03-16 11:27:47.901905464 +0000
@@ -15,6 +15,7 @@
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/mm.h>
+#include <asm/atomic_kmap.h>
/* This sets the pointer FPU_info to point to the argument part
of the stack frame of math_emulate() */
@@ -22,7 +23,7 @@
/* s is always from a cpu register, and the cpu does bounds checking
* during register load --> no further bounds checks needed */
-#define LDT_DESCRIPTOR(s) (((struct desc_struct *)current->mm->context.ldt)[(s) >> 3])
+#define LDT_DESCRIPTOR(s) (((struct desc_struct *)__kmap_atomic_vaddr(KM_LDT_PAGE0))[(s) >> 3])
#define SEG_D_SIZE(x) ((x).b & (3 << 21))
#define SEG_G_BIT(x) ((x).b & (1 << 23))
#define SEG_GRANULARITY(x) (((x).b & (1 << 23)) ? 4096 : 1)
--- diff/arch/i386/mm/fault.c 2003-12-19 09:51:11.000000000 +0000
+++ source/arch/i386/mm/fault.c 2004-03-16 11:27:47.902905312 +0000
@@ -27,6 +27,7 @@
#include <asm/pgalloc.h>
#include <asm/hardirq.h>
#include <asm/desc.h>
+#include <asm/tlbflush.h>
extern void die(const char *,struct pt_regs *,long);
@@ -104,8 +105,17 @@ static inline unsigned long get_segment_
if (seg & (1<<2)) {
/* Must lock the LDT while reading it. */
down(¤t->mm->context.sem);
+#if 1
+ /* horrible hack for 4/4 disabled kernels.
+ I'm not quite sure what the TLB flush is good for,
+ it's mindlessly copied from the read_ldt code */
+ __flush_tlb_global();
+ desc = kmap(current->mm->context.ldt_pages[(seg&~7)/PAGE_SIZE]);
+ desc = (void *)desc + ((seg & ~7) % PAGE_SIZE);
+#else
desc = current->mm->context.ldt;
desc = (void *)desc + (seg & ~7);
+#endif
} else {
/* Must disable preemption while reading the GDT. */
desc = (u32 *)&cpu_gdt_table[get_cpu()];
@@ -118,6 +128,9 @@ static inline unsigned long get_segment_
(desc[1] & 0xff000000);
if (seg & (1<<2)) {
+#if 1
+ kunmap((void *)((unsigned long)desc & PAGE_MASK));
+#endif
up(¤t->mm->context.sem);
} else
put_cpu();
@@ -243,6 +256,19 @@ asmlinkage void do_page_fault(struct pt_
* (error_code & 4) == 0, and that the fault was not a
* protection error (error_code & 1) == 0.
*/
+#ifdef CONFIG_X86_4G
+ /*
+ * On 4/4 all kernels faults are either bugs, vmalloc or prefetch
+ */
+ if (unlikely((regs->xcs & 3) == 0)) {
+ if (error_code & 3)
+ goto bad_area_nosemaphore;
+
+ /* If it's vm86 fall through */
+ if (!(regs->eflags & VM_MASK))
+ goto vmalloc_fault;
+ }
+#else
if (unlikely(address >= TASK_SIZE)) {
if (!(error_code & 5))
goto vmalloc_fault;
@@ -252,6 +278,7 @@ asmlinkage void do_page_fault(struct pt_
*/
goto bad_area_nosemaphore;
}
+#endif
mm = tsk->mm;
@@ -326,6 +353,8 @@ good_area:
goto do_sigbus;
case VM_FAULT_OOM:
goto out_of_memory;
+ case VM_FAULT_SIGSEGV:
+ goto bad_area;
default:
BUG();
}
@@ -403,6 +432,12 @@ no_context:
* Oops. The kernel tried to access some bad page. We'll have to
* terminate things with extreme prejudice.
*/
+#ifdef CONFIG_KGDB
+ if (!user_mode(regs)){
+ kgdb_handle_exception(14,SIGBUS, error_code, regs);
+ return;
+ }
+#endif
bust_spinlocks(1);
--- diff/arch/i386/mm/hugetlbpage.c 2004-01-19 10:22:55.000000000 +0000
+++ source/arch/i386/mm/hugetlbpage.c 2004-03-16 11:27:47.902905312 +0000
@@ -29,7 +29,7 @@ static spinlock_t htlbpage_lock = SPIN_L
static void enqueue_huge_page(struct page *page)
{
- list_add(&page->list,
+ list_add(&page->lru,
&hugepage_freelists[page_zone(page)->zone_pgdat->node_id]);
}
@@ -44,8 +44,8 @@ static struct page *dequeue_huge_page(vo
break;
}
if (nid >= 0 && nid < MAX_NUMNODES && !list_empty(&hugepage_freelists[nid])) {
- page = list_entry(hugepage_freelists[nid].next, struct page, list);
- list_del(&page->list);
+ page = list_entry(hugepage_freelists[nid].next, struct page, lru);
+ list_del(&page->lru);
}
return page;
}
@@ -61,6 +61,27 @@ static struct page *alloc_fresh_huge_pag
static void free_huge_page(struct page *page);
+#ifdef CONFIG_NUMA
+
+static inline void huge_inc_rss(struct mm_struct *mm, struct page *page)
+{
+ mm->rss += (HPAGE_SIZE / PAGE_SIZE);
+ mm->pernode_rss[page_nodenum(page)] += (HPAGE_SIZE / PAGE_SIZE);
+}
+
+static inline void huge_dec_rss(struct mm_struct *mm, struct page *page)
+{
+ mm->rss -= (HPAGE_SIZE / PAGE_SIZE);
+ mm->pernode_rss[page_nodenum(page)] -= (HPAGE_SIZE / PAGE_SIZE);
+}
+
+#else /* !CONFIG_NUMA */
+
+#define huge_inc_rss(mm, page) ((mm)->rss += (HPAGE_SIZE / PAGE_SIZE))
+#define huge_dec_rss(mm, page) ((mm)->rss -= (HPAGE_SIZE / PAGE_SIZE))
+
+#endif /* CONFIG_NUMA */
+
static struct page *alloc_hugetlb_page(void)
{
int i;
@@ -105,7 +126,7 @@ static void set_huge_pte(struct mm_struc
{
pte_t entry;
- mm->rss += (HPAGE_SIZE / PAGE_SIZE);
+ huge_inc_rss(mm, page);
if (write_access) {
entry =
pte_mkwrite(pte_mkdirty(mk_pte(page, vma->vm_page_prot)));
@@ -145,7 +166,7 @@ int copy_hugetlb_page_range(struct mm_st
ptepage = pte_page(entry);
get_page(ptepage);
set_pte(dst_pte, entry);
- dst->rss += (HPAGE_SIZE / PAGE_SIZE);
+ huge_inc_rss(dst, ptepage);
addr += HPAGE_SIZE;
}
return 0;
@@ -278,9 +299,8 @@ follow_huge_pmd(struct mm_struct *mm, un
static void free_huge_page(struct page *page)
{
BUG_ON(page_count(page));
- BUG_ON(page->mapping);
- INIT_LIST_HEAD(&page->list);
+ INIT_LIST_HEAD(&page->lru);
spin_lock(&htlbpage_lock);
enqueue_huge_page(page);
@@ -314,8 +334,8 @@ void unmap_hugepage_range(struct vm_area
page = pte_page(*pte);
huge_page_release(page);
pte_clear(pte);
+ huge_dec_rss(mm, page);
}
- mm->rss -= (end - start) >> PAGE_SHIFT;
flush_tlb_range(vma, start, end);
}
@@ -409,19 +429,19 @@ static int try_to_free_low(int count)
/* all lowmem is on node 0 */
list_for_each(p, &hugepage_freelists[0]) {
if (map) {
- list_del(&map->list);
+ list_del(&map->lru);
update_and_free_page(map);
htlbpagemem--;
map = NULL;
if (++count == 0)
break;
}
- page = list_entry(p, struct page, list);
+ page = list_entry(p, struct page, lru);
if (!PageHighMem(page))
map = page;
}
if (map) {
- list_del(&map->list);
+ list_del(&map->lru);
update_and_free_page(map);
htlbpagemem--;
count++;
--- diff/arch/i386/mm/init.c 2004-01-19 10:22:55.000000000 +0000
+++ source/arch/i386/mm/init.c 2004-03-16 11:27:47.904905008 +0000
@@ -40,125 +40,13 @@
#include <asm/tlb.h>
#include <asm/tlbflush.h>
#include <asm/sections.h>
+#include <asm/desc.h>
DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
unsigned long highstart_pfn, highend_pfn;
static int do_test_wp_bit(void);
-/*
- * Creates a middle page table and puts a pointer to it in the
- * given global directory entry. This only returns the gd entry
- * in non-PAE compilation mode, since the middle layer is folded.
- */
-static pmd_t * __init one_md_table_init(pgd_t *pgd)
-{
- pmd_t *pmd_table;
-
-#ifdef CONFIG_X86_PAE
- pmd_table = (pmd_t *) alloc_bootmem_low_pages(PAGE_SIZE);
- set_pgd(pgd, __pgd(__pa(pmd_table) | _PAGE_PRESENT));
- if (pmd_table != pmd_offset(pgd, 0))
- BUG();
-#else
- pmd_table = pmd_offset(pgd, 0);
-#endif
-
- return pmd_table;
-}
-
-/*
- * Create a page table and place a pointer to it in a middle page
- * directory entry.
- */
-static pte_t * __init one_page_table_init(pmd_t *pmd)
-{
- if (pmd_none(*pmd)) {
- pte_t *page_table = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE);
- set_pmd(pmd, __pmd(__pa(page_table) | _PAGE_TABLE));
- if (page_table != pte_offset_kernel(pmd, 0))
- BUG();
-
- return page_table;
- }
-
- return pte_offset_kernel(pmd, 0);
-}
-
-/*
- * This function initializes a certain range of kernel virtual memory
- * with new bootmem page tables, everywhere page tables are missing in
- * the given range.
- */
-
-/*
- * NOTE: The pagetables are allocated contiguous on the physical space
- * so we can cache the place of the first one and move around without
- * checking the pgd every time.
- */
-static void __init page_table_range_init (unsigned long start, unsigned long end, pgd_t *pgd_base)
-{
- pgd_t *pgd;
- pmd_t *pmd;
- int pgd_idx, pmd_idx;
- unsigned long vaddr;
-
- vaddr = start;
- pgd_idx = pgd_index(vaddr);
- pmd_idx = pmd_index(vaddr);
- pgd = pgd_base + pgd_idx;
-
- for ( ; (pgd_idx < PTRS_PER_PGD) && (vaddr != end); pgd++, pgd_idx++) {
- if (pgd_none(*pgd))
- one_md_table_init(pgd);
-
- pmd = pmd_offset(pgd, vaddr);
- for (; (pmd_idx < PTRS_PER_PMD) && (vaddr != end); pmd++, pmd_idx++) {
- if (pmd_none(*pmd))
- one_page_table_init(pmd);
-
- vaddr += PMD_SIZE;
- }
- pmd_idx = 0;
- }
-}
-
-/*
- * This maps the physical memory to kernel virtual address space, a total
- * of max_low_pfn pages, by creating page tables starting from address
- * PAGE_OFFSET.
- */
-static void __init kernel_physical_mapping_init(pgd_t *pgd_base)
-{
- unsigned long pfn;
- pgd_t *pgd;
- pmd_t *pmd;
- pte_t *pte;
- int pgd_idx, pmd_idx, pte_ofs;
-
- pgd_idx = pgd_index(PAGE_OFFSET);
- pgd = pgd_base + pgd_idx;
- pfn = 0;
-
- for (; pgd_idx < PTRS_PER_PGD; pgd++, pgd_idx++) {
- pmd = one_md_table_init(pgd);
- if (pfn >= max_low_pfn)
- continue;
- for (pmd_idx = 0; pmd_idx < PTRS_PER_PMD && pfn < max_low_pfn; pmd++, pmd_idx++) {
- /* Map with big pages if possible, otherwise create normal page tables. */
- if (cpu_has_pse) {
- set_pmd(pmd, pfn_pmd(pfn, PAGE_KERNEL_LARGE));
- pfn += PTRS_PER_PTE;
- } else {
- pte = one_page_table_init(pmd);
-
- for (pte_ofs = 0; pte_ofs < PTRS_PER_PTE && pfn < max_low_pfn; pte++, pfn++, pte_ofs++)
- set_pte(pte, pfn_pte(pfn, PAGE_KERNEL));
- }
- }
- }
-}
-
static inline int page_kills_ppro(unsigned long pagenr)
{
if (pagenr >= 0x70000 && pagenr <= 0x7003F)
@@ -206,11 +94,8 @@ static inline int page_is_ram(unsigned l
return 0;
}
-#ifdef CONFIG_HIGHMEM
pte_t *kmap_pte;
-pgprot_t kmap_prot;
-EXPORT_SYMBOL(kmap_prot);
EXPORT_SYMBOL(kmap_pte);
#define kmap_get_fixmap_pte(vaddr) \
@@ -218,29 +103,7 @@ EXPORT_SYMBOL(kmap_pte);
void __init kmap_init(void)
{
- unsigned long kmap_vstart;
-
- /* cache the first kmap pte */
- kmap_vstart = __fix_to_virt(FIX_KMAP_BEGIN);
- kmap_pte = kmap_get_fixmap_pte(kmap_vstart);
-
- kmap_prot = PAGE_KERNEL;
-}
-
-void __init permanent_kmaps_init(pgd_t *pgd_base)
-{
- pgd_t *pgd;
- pmd_t *pmd;
- pte_t *pte;
- unsigned long vaddr;
-
- vaddr = PKMAP_BASE;
- page_table_range_init(vaddr, vaddr + PAGE_SIZE*LAST_PKMAP, pgd_base);
-
- pgd = swapper_pg_dir + pgd_index(vaddr);
- pmd = pmd_offset(pgd, vaddr);
- pte = pte_offset_kernel(pmd, vaddr);
- pkmap_page_table = pte;
+ kmap_pte = kmap_get_fixmap_pte(__fix_to_virt(FIX_KMAP_BEGIN));
}
void __init one_highpage_init(struct page *page, int pfn, int bad_ppro)
@@ -255,6 +118,8 @@ void __init one_highpage_init(struct pag
SetPageReserved(page);
}
+#ifdef CONFIG_HIGHMEM
+
#ifndef CONFIG_DISCONTIGMEM
void __init set_highmem_pages_init(int bad_ppro)
{
@@ -266,12 +131,9 @@ void __init set_highmem_pages_init(int b
#else
extern void set_highmem_pages_init(int);
#endif /* !CONFIG_DISCONTIGMEM */
-
#else
-#define kmap_init() do { } while (0)
-#define permanent_kmaps_init(pgd_base) do { } while (0)
-#define set_highmem_pages_init(bad_ppro) do { } while (0)
-#endif /* CONFIG_HIGHMEM */
+# define set_highmem_pages_init(bad_ppro) do { } while (0)
+#endif
unsigned long __PAGE_KERNEL = _PAGE_KERNEL;
@@ -281,30 +143,125 @@ unsigned long __PAGE_KERNEL = _PAGE_KERN
extern void __init remap_numa_kva(void);
#endif
-static void __init pagetable_init (void)
+static __init void prepare_pagetables(pgd_t *pgd_base, unsigned long address)
+{
+ pgd_t *pgd;
+ pmd_t *pmd;
+ pte_t *pte;
+
+ pgd = pgd_base + pgd_index(address);
+ pmd = pmd_offset(pgd, address);
+ if (!pmd_present(*pmd)) {
+ pte = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE);
+ set_pmd(pmd, __pmd(_PAGE_TABLE + __pa(pte)));
+ }
+}
+
+static void __init fixrange_init (unsigned long start, unsigned long end, pgd_t *pgd_base)
+{
+ unsigned long vaddr;
+
+ for (vaddr = start; vaddr != end; vaddr += PAGE_SIZE)
+ prepare_pagetables(pgd_base, vaddr);
+}
+
+void setup_identity_mappings(pgd_t *pgd_base, unsigned long start, unsigned long end)
{
unsigned long vaddr;
- pgd_t *pgd_base = swapper_pg_dir;
+ pgd_t *pgd;
+ int i, j, k;
+ pmd_t *pmd;
+ pte_t *pte, *pte_base;
+ pgd = pgd_base;
+
+ for (i = 0; i < PTRS_PER_PGD; pgd++, i++) {
+ vaddr = i*PGDIR_SIZE;
+ if (end && (vaddr >= end))
+ break;
+ pmd = pmd_offset(pgd, 0);
+ for (j = 0; j < PTRS_PER_PMD; pmd++, j++) {
+ vaddr = i*PGDIR_SIZE + j*PMD_SIZE;
+ if (end && (vaddr >= end))
+ break;
+ if (vaddr < start)
+ continue;
+ if (cpu_has_pse) {
+ unsigned long __pe;
+
+ set_in_cr4(X86_CR4_PSE);
+ boot_cpu_data.wp_works_ok = 1;
+ __pe = _KERNPG_TABLE + _PAGE_PSE + vaddr - start;
+ /* Make it "global" too if supported */
+ if (cpu_has_pge) {
+ set_in_cr4(X86_CR4_PGE);
+#if !defined(CONFIG_X86_SWITCH_PAGETABLES)
+ __pe += _PAGE_GLOBAL;
+ __PAGE_KERNEL |= _PAGE_GLOBAL;
+#endif
+ }
+ set_pmd(pmd, __pmd(__pe));
+ continue;
+ }
+ if (!pmd_present(*pmd))
+ pte_base = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE);
+ else
+ pte_base = (pte_t *) page_address(pmd_page(*pmd));
+ pte = pte_base;
+ for (k = 0; k < PTRS_PER_PTE; pte++, k++) {
+ vaddr = i*PGDIR_SIZE + j*PMD_SIZE + k*PAGE_SIZE;
+ if (end && (vaddr >= end))
+ break;
+ if (vaddr < start)
+ continue;
+ *pte = mk_pte_phys(vaddr-start, PAGE_KERNEL);
+ }
+ set_pmd(pmd, __pmd(_KERNPG_TABLE + __pa(pte_base)));
+ }
+ }
+}
+
+static void __init pagetable_init (void)
+{
+ unsigned long vaddr, end;
+ pgd_t *pgd_base;
#ifdef CONFIG_X86_PAE
int i;
- /* Init entries of the first-level page table to the zero page */
- for (i = 0; i < PTRS_PER_PGD; i++)
- set_pgd(pgd_base + i, __pgd(__pa(empty_zero_page) | _PAGE_PRESENT));
#endif
- /* Enable PSE if available */
- if (cpu_has_pse) {
- set_in_cr4(X86_CR4_PSE);
- }
+ /*
+ * This can be zero as well - no problem, in that case we exit
+ * the loops anyway due to the PTRS_PER_* conditions.
+ */
+ end = (unsigned long)__va(max_low_pfn*PAGE_SIZE);
- /* Enable PGE if available */
- if (cpu_has_pge) {
- set_in_cr4(X86_CR4_PGE);
- __PAGE_KERNEL |= _PAGE_GLOBAL;
+ pgd_base = swapper_pg_dir;
+#ifdef CONFIG_X86_PAE
+ /*
+ * It causes too many problems if there's no proper pmd set up
+ * for all 4 entries of the PGD - so we allocate all of them.
+ * PAE systems will not miss this extra 4-8K anyway ...
+ */
+ for (i = 0; i < PTRS_PER_PGD; i++) {
+ pmd_t *pmd = (pmd_t *) alloc_bootmem_low_pages(PAGE_SIZE);
+ set_pgd(pgd_base + i, __pgd(__pa(pmd) + 0x1));
}
+#endif
+ /*
+ * Set up lowmem-sized identity mappings at PAGE_OFFSET:
+ */
+ setup_identity_mappings(pgd_base, PAGE_OFFSET, end);
- kernel_physical_mapping_init(pgd_base);
+ /*
+ * Add flat-mode identity-mappings - SMP needs it when
+ * starting up on an AP from real-mode. (In the non-PAE
+ * case we already have these mappings through head.S.)
+ * All user-space mappings are explicitly cleared after
+ * SMP startup.
+ */
+#if defined(CONFIG_SMP) && defined(CONFIG_X86_PAE)
+ setup_identity_mappings(pgd_base, 0, 16*1024*1024);
+#endif
remap_numa_kva();
/*
@@ -312,38 +269,64 @@ static void __init pagetable_init (void)
* created - mappings will be set by set_fixmap():
*/
vaddr = __fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK;
- page_table_range_init(vaddr, 0, pgd_base);
+ fixrange_init(vaddr, 0, pgd_base);
- permanent_kmaps_init(pgd_base);
+#ifdef CONFIG_HIGHMEM
+ {
+ pgd_t *pgd;
+ pmd_t *pmd;
+ pte_t *pte;
-#ifdef CONFIG_X86_PAE
- /*
- * Add low memory identity-mappings - SMP needs it when
- * starting up on an AP from real-mode. In the non-PAE
- * case we already have these mappings through head.S.
- * All user-space mappings are explicitly cleared after
- * SMP startup.
- */
- pgd_base[0] = pgd_base[USER_PTRS_PER_PGD];
+ /*
+ * Permanent kmaps:
+ */
+ vaddr = PKMAP_BASE;
+ fixrange_init(vaddr, vaddr + PAGE_SIZE*LAST_PKMAP, pgd_base);
+
+ pgd = swapper_pg_dir + pgd_index(vaddr);
+ pmd = pmd_offset(pgd, vaddr);
+ pte = pte_offset_kernel(pmd, vaddr);
+ pkmap_page_table = pte;
+ }
#endif
}
-void zap_low_mappings (void)
+/*
+ * Clear kernel pagetables in a PMD_SIZE-aligned range.
+ */
+static void clear_mappings(pgd_t *pgd_base, unsigned long start, unsigned long end)
{
- int i;
+ unsigned long vaddr;
+ pgd_t *pgd;
+ pmd_t *pmd;
+ int i, j;
+
+ pgd = pgd_base;
+
+ for (i = 0; i < PTRS_PER_PGD; pgd++, i++) {
+ vaddr = i*PGDIR_SIZE;
+ if (end && (vaddr >= end))
+ break;
+ pmd = pmd_offset(pgd, 0);
+ for (j = 0; j < PTRS_PER_PMD; pmd++, j++) {
+ vaddr = i*PGDIR_SIZE + j*PMD_SIZE;
+ if (end && (vaddr >= end))
+ break;
+ if (vaddr < start)
+ continue;
+ pmd_clear(pmd);
+ }
+ }
+ flush_tlb_all();
+}
+
+void zap_low_mappings(void)
+{
+ printk("zapping low mappings.\n");
/*
* Zap initial low-memory mappings.
- *
- * Note that "pgd_clear()" doesn't do it for
- * us, because pgd_clear() is a no-op on i386.
*/
- for (i = 0; i < USER_PTRS_PER_PGD; i++)
-#ifdef CONFIG_X86_PAE
- set_pgd(swapper_pg_dir+i, __pgd(1 + __pa(empty_zero_page)));
-#else
- set_pgd(swapper_pg_dir+i, __pgd(0));
-#endif
- flush_tlb_all();
+ clear_mappings(swapper_pg_dir, 0, 16*1024*1024);
}
#ifndef CONFIG_DISCONTIGMEM
@@ -393,7 +376,15 @@ void __init paging_init(void)
set_in_cr4(X86_CR4_PAE);
#endif
__flush_tlb_all();
-
+ /*
+ * Subtle. SMP is doing it's boot stuff late (because it has to
+ * fork idle threads) - but it also needs low mappings for the
+ * protected-mode entry to work. We zap these entries only after
+ * the WP-bit has been tested.
+ */
+#ifndef CONFIG_SMP
+ zap_low_mappings();
+#endif
kmap_init();
zone_sizes_init();
}
@@ -474,9 +465,6 @@ void __init mem_init(void)
high_memory = (void *) __va(max_low_pfn * PAGE_SIZE);
#endif
- /* clear the zero-page */
- memset(empty_zero_page, 0, PAGE_SIZE);
-
/* this will put all low memory onto the freelists */
totalram_pages += __free_all_bootmem();
@@ -515,38 +503,43 @@ void __init mem_init(void)
if (boot_cpu_data.wp_works_ok < 0)
test_wp_bit();
- /*
- * Subtle. SMP is doing it's boot stuff late (because it has to
- * fork idle threads) - but it also needs low mappings for the
- * protected-mode entry to work. We zap these entries only after
- * the WP-bit has been tested.
- */
-#ifndef CONFIG_SMP
- zap_low_mappings();
-#endif
+ entry_trampoline_setup();
+ default_ldt_page = virt_to_page(default_ldt);
+ load_LDT(&init_mm.context);
}
-kmem_cache_t *pgd_cache;
-kmem_cache_t *pmd_cache;
+kmem_cache_t *pgd_cache, *pmd_cache, *kpmd_cache;
void __init pgtable_cache_init(void)
{
if (PTRS_PER_PMD > 1) {
pmd_cache = kmem_cache_create("pmd",
- PTRS_PER_PMD*sizeof(pmd_t),
- 0,
- SLAB_HWCACHE_ALIGN | SLAB_MUST_HWCACHE_ALIGN,
- pmd_ctor,
- NULL);
+ PTRS_PER_PMD*sizeof(pmd_t),
+ PTRS_PER_PMD*sizeof(pmd_t),
+ 0,
+ pmd_ctor,
+ NULL);
if (!pmd_cache)
panic("pgtable_cache_init(): cannot create pmd cache");
+
+ if (TASK_SIZE > PAGE_OFFSET) {
+ kpmd_cache = kmem_cache_create("kpmd",
+ PTRS_PER_PMD*sizeof(pmd_t),
+ PTRS_PER_PMD*sizeof(pmd_t),
+ 0,
+ kpmd_ctor,
+ NULL);
+ if (!kpmd_cache)
+ panic("pgtable_cache_init(): "
+ "cannot create kpmd cache");
+ }
}
+
pgd_cache = kmem_cache_create("pgd",
PTRS_PER_PGD*sizeof(pgd_t),
+ PTRS_PER_PGD*sizeof(pgd_t),
0,
- SLAB_HWCACHE_ALIGN | SLAB_MUST_HWCACHE_ALIGN,
- pgd_ctor,
- PTRS_PER_PMD == 1 ? pgd_dtor : NULL);
+ NULL, NULL);
if (!pgd_cache)
panic("pgtable_cache_init(): Cannot create pgd cache");
}
--- diff/arch/i386/mm/pageattr.c 2003-08-20 13:16:24.000000000 +0000
+++ source/arch/i386/mm/pageattr.c 2004-03-16 11:27:47.904905008 +0000
@@ -67,22 +67,19 @@ static void flush_kernel_map(void *dummy
static void set_pmd_pte(pte_t *kpte, unsigned long address, pte_t pte)
{
- struct page *page;
- unsigned long flags;
-
set_pte_atomic(kpte, pte); /* change init_mm */
- if (PTRS_PER_PMD > 1)
- return;
-
- spin_lock_irqsave(&pgd_lock, flags);
- list_for_each_entry(page, &pgd_list, lru) {
- pgd_t *pgd;
- pmd_t *pmd;
- pgd = (pgd_t *)page_address(page) + pgd_index(address);
- pmd = pmd_offset(pgd, address);
- set_pte_atomic((pte_t *)pmd, pte);
+#ifndef CONFIG_X86_PAE
+ {
+ struct list_head *l;
+ spin_lock(&mmlist_lock);
+ list_for_each(l, &init_mm.mmlist) {
+ struct mm_struct *mm = list_entry(l, struct mm_struct, mmlist);
+ pmd_t *pmd = pmd_offset(pgd_offset(mm, address), address);
+ set_pte_atomic((pte_t *)pmd, pte);
+ }
+ spin_unlock(&mmlist_lock);
}
- spin_unlock_irqrestore(&pgd_lock, flags);
+#endif
}
/*
@@ -135,7 +132,7 @@ __change_page_attr(struct page *page, pg
}
if (cpu_has_pse && (atomic_read(&kpte_page->count) == 1)) {
- list_add(&kpte_page->list, &df_list);
+ list_add(&kpte_page->lru, &df_list);
revert_page(kpte_page, address);
}
return 0;
@@ -188,7 +185,7 @@ void global_flush_tlb(void)
flush_map();
n = l.next;
while (n != &l) {
- struct page *pg = list_entry(n, struct page, list);
+ struct page *pg = list_entry(n, struct page, lru);
n = n->next;
__free_page(pg);
}
--- diff/arch/i386/mm/pgtable.c 2003-10-09 08:47:16.000000000 +0000
+++ source/arch/i386/mm/pgtable.c 2004-03-16 11:27:47.905904856 +0000
@@ -21,6 +21,7 @@
#include <asm/e820.h>
#include <asm/tlb.h>
#include <asm/tlbflush.h>
+#include <asm/atomic_kmap.h>
void show_mem(void)
{
@@ -157,50 +158,50 @@ void pmd_ctor(void *pmd, kmem_cache_t *c
memset(pmd, 0, PTRS_PER_PMD*sizeof(pmd_t));
}
+void kpmd_ctor(void *__pmd, kmem_cache_t *cache, unsigned long flags)
+{
+ pmd_t *kpmd, *pmd;
+ kpmd = pmd_offset(&swapper_pg_dir[PTRS_PER_PGD-1],
+ (PTRS_PER_PMD - NR_SHARED_PMDS)*PMD_SIZE);
+ pmd = (pmd_t *)__pmd + (PTRS_PER_PMD - NR_SHARED_PMDS);
+
+ memset(__pmd, 0, (PTRS_PER_PMD - NR_SHARED_PMDS)*sizeof(pmd_t));
+ memcpy(pmd, kpmd, NR_SHARED_PMDS*sizeof(pmd_t));
+}
+
/*
- * List of all pgd's needed for non-PAE so it can invalidate entries
- * in both cached and uncached pgd's; not needed for PAE since the
- * kernel pmd is shared. If PAE were not to share the pmd a similar
- * tactic would be needed. This is essentially codepath-based locking
- * against pageattr.c; it is the unique case in which a valid change
- * of kernel pagetables can't be lazily synchronized by vmalloc faults.
- * vmalloc faults work because attached pagetables are never freed.
- * If the locking proves to be non-performant, a ticketing scheme with
- * checks at dup_mmap(), exec(), and other mmlist addition points
- * could be used. The locking scheme was chosen on the basis of
- * manfred's recommendations and having no core impact whatsoever.
+ * This is not that hard to figure out.
+ * (a) PTRS_PER_PMD == 1 means non-PAE.
+ * (b) PTRS_PER_PMD > 1 means PAE.
+ * (c) TASK_SIZE > PAGE_OFFSET means 4:4.
+ * (d) TASK_SIZE <= PAGE_OFFSET means non-4:4.
+ *
+ * Do *NOT* back out the preconstruction like the patch I'm cleaning
+ * up after this very instant did, or at all, for that matter.
+ * This is never called when PTRS_PER_PMD > 1 && TASK_SIZE > PAGE_OFFSET.
* -- wli
*/
-spinlock_t pgd_lock = SPIN_LOCK_UNLOCKED;
-LIST_HEAD(pgd_list);
-
-void pgd_ctor(void *pgd, kmem_cache_t *cache, unsigned long unused)
+void pgd_ctor(void *__pgd, kmem_cache_t *cache, unsigned long unused)
{
- unsigned long flags;
+ pgd_t *pgd = (pgd_t *)__pgd;
- if (PTRS_PER_PMD == 1)
- spin_lock_irqsave(&pgd_lock, flags);
+ if ((PTRS_PER_PMD == 1) && (TASK_SIZE > PAGE_OFFSET))
+ memcpy(&pgd[PTRS_PER_PGD - NR_SHARED_PMDS],
+ &swapper_pg_dir[PTRS_PER_PGD - NR_SHARED_PMDS],
+ NR_SHARED_PMDS * sizeof(pgd_t));
- memcpy((pgd_t *)pgd + USER_PTRS_PER_PGD,
+ if (TASK_SIZE <= PAGE_OFFSET)
+ memcpy(pgd + USER_PTRS_PER_PGD,
swapper_pg_dir + USER_PTRS_PER_PGD,
(PTRS_PER_PGD - USER_PTRS_PER_PGD) * sizeof(pgd_t));
if (PTRS_PER_PMD > 1)
return;
- list_add(&virt_to_page(pgd)->lru, &pgd_list);
- spin_unlock_irqrestore(&pgd_lock, flags);
- memset(pgd, 0, USER_PTRS_PER_PGD*sizeof(pgd_t));
-}
-
-/* never called when PTRS_PER_PMD > 1 */
-void pgd_dtor(void *pgd, kmem_cache_t *cache, unsigned long unused)
-{
- unsigned long flags; /* can be called from interrupt context */
-
- spin_lock_irqsave(&pgd_lock, flags);
- list_del(&virt_to_page(pgd)->lru);
- spin_unlock_irqrestore(&pgd_lock, flags);
+ if (TASK_SIZE > PAGE_OFFSET)
+ memset(pgd, 0, (PTRS_PER_PGD - NR_SHARED_PMDS)*sizeof(pgd_t));
+ else
+ memset(pgd, 0, USER_PTRS_PER_PGD*sizeof(pgd_t));
}
pgd_t *pgd_alloc(struct mm_struct *mm)
@@ -208,18 +209,39 @@ pgd_t *pgd_alloc(struct mm_struct *mm)
int i;
pgd_t *pgd = kmem_cache_alloc(pgd_cache, GFP_KERNEL);
+ if (pgd)
+ pgd_ctor(pgd, NULL, 0);
+
if (PTRS_PER_PMD == 1 || !pgd)
return pgd;
+ /*
+ * In the 4G userspace case alias the top 16 MB virtual
+ * memory range into the user mappings as well (these
+ * include the trampoline and CPU data structures).
+ */
for (i = 0; i < USER_PTRS_PER_PGD; ++i) {
- pmd_t *pmd = kmem_cache_alloc(pmd_cache, GFP_KERNEL);
+ kmem_cache_t *cache;
+ pmd_t *pmd;
+
+ if (TASK_SIZE > PAGE_OFFSET && i == USER_PTRS_PER_PGD - 1)
+ cache = kpmd_cache;
+ else
+ cache = pmd_cache;
+
+ pmd = kmem_cache_alloc(cache, GFP_KERNEL);
if (!pmd)
goto out_oom;
set_pgd(&pgd[i], __pgd(1 + __pa((u64)((u32)pmd))));
}
- return pgd;
+ return pgd;
out_oom:
+ /*
+ * we don't have to handle the kpmd_cache here, since it's the
+ * last allocation, and has either nothing to free or when it
+ * succeeds the whole operation succeeds.
+ */
for (i--; i >= 0; i--)
kmem_cache_free(pmd_cache, (void *)__va(pgd_val(pgd[i])-1));
kmem_cache_free(pgd_cache, pgd);
@@ -230,10 +252,29 @@ void pgd_free(pgd_t *pgd)
{
int i;
- /* in the PAE case user pgd entries are overwritten before usage */
- if (PTRS_PER_PMD > 1)
- for (i = 0; i < USER_PTRS_PER_PGD; ++i)
- kmem_cache_free(pmd_cache, (void *)__va(pgd_val(pgd[i])-1));
/* in the non-PAE case, clear_page_tables() clears user pgd entries */
+ if (PTRS_PER_PMD == 1)
+ goto out_free;
+
+ /* in the PAE case user pgd entries are overwritten before usage */
+ for (i = 0; i < USER_PTRS_PER_PGD; ++i) {
+ kmem_cache_t *cache;
+ pmd_t *pmd = __va(pgd_val(pgd[i]) - 1);
+
+ /*
+ * only userspace pmd's are cleared for us
+ * by mm/memory.c; it's a slab cache invariant
+ * that we must separate the kernel pmd slab
+ * all times, else we'll have bad pmd's.
+ */
+ if (TASK_SIZE > PAGE_OFFSET && i == USER_PTRS_PER_PGD - 1)
+ cache = kpmd_cache;
+ else
+ cache = pmd_cache;
+
+ kmem_cache_free(cache, pmd);
+ }
+out_free:
kmem_cache_free(pgd_cache, pgd);
}
+
--- diff/arch/i386/oprofile/op_model_p4.c 2003-08-26 09:00:51.000000000 +0000
+++ source/arch/i386/oprofile/op_model_p4.c 2004-03-16 11:27:47.905904856 +0000
@@ -382,11 +382,8 @@ static struct p4_event_binding p4_events
static unsigned int get_stagger(void)
{
#ifdef CONFIG_SMP
- int cpu;
- if (smp_num_siblings > 1) {
- cpu = smp_processor_id();
- return (cpu_sibling_map[cpu] > cpu) ? 0 : 1;
- }
+ int cpu = smp_processor_id();
+ return (cpu != first_cpu(cpu_sibling_map[cpu]));
#endif
return 0;
}
--- diff/arch/ia64/Kconfig 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/ia64/Kconfig 2004-03-16 11:27:47.907904552 +0000
@@ -538,6 +538,13 @@ config DEBUG_INFO
Say Y here only if you plan to use gdb to debug the kernel.
If you don't debug the kernel, you can say N.
+config LOCKMETER
+ bool "Kernel lock metering"
+ depends on SMP
+ help
+ Say Y to enable kernel lock metering, which adds overhead to SMP locks,
+ but allows you to see various statistics using the lockstat command.
+
config SYSVIPC_COMPAT
bool
depends on COMPAT && SYSVIPC
--- diff/arch/ia64/ia32/ia32_entry.S 2004-01-19 10:22:55.000000000 +0000
+++ source/arch/ia64/ia32/ia32_entry.S 2004-03-16 11:27:47.906904704 +0000
@@ -465,7 +465,7 @@ ia32_syscall_table:
data8 sys_epoll_create
data8 sys32_epoll_ctl /* 255 */
data8 sys32_epoll_wait
- data8 sys_remap_file_pages
+ data8 old_remap_file_pages
data8 sys_set_tid_address
data8 sys32_timer_create
data8 compat_timer_settime /* 260 */
@@ -484,5 +484,12 @@ ia32_syscall_table:
data8 sys_ni_syscall
data8 sys_ni_syscall
+ data8 sys_ni_syscall /* 275 */
+ data8 sys_ni_syscall
+ data8 sys_ni_syscall
+ data8 sys_ni_syscall
+ data8 sys_ni_syscall
+ data8 sys_remap_file_pages /* 280 */
+
// guard against failures to increase IA32_NR_syscalls
.org ia32_syscall_table + 8*IA32_NR_syscalls
--- diff/arch/ia64/kernel/entry.S 2004-02-09 10:36:07.000000000 +0000
+++ source/arch/ia64/kernel/entry.S 2004-03-16 11:27:47.908904400 +0000
@@ -1367,7 +1367,7 @@ sys_call_table:
data8 sys_ni_syscall /* was: ia64_oldfstat */
data8 sys_vhangup
data8 sys_lchown
- data8 sys_remap_file_pages // 1125
+ data8 old_remap_file_pages // 1125
data8 sys_wait4
data8 sys_sysinfo
data8 sys_clone
@@ -1501,7 +1501,7 @@ sys_call_table:
data8 sys_clock_nanosleep
data8 sys_fstatfs64
data8 sys_statfs64
- data8 sys_ni_syscall
+ data8 sys_remap_file_pages
data8 sys_ni_syscall // 1260
data8 sys_ni_syscall
data8 sys_ni_syscall
--- diff/arch/ia64/kernel/unaligned.c 2004-02-18 08:54:07.000000000 +0000
+++ source/arch/ia64/kernel/unaligned.c 2004-03-16 11:27:47.909904248 +0000
@@ -1337,7 +1337,7 @@ ia64_handle_unaligned (unsigned long ifa
* be holding locks...
*/
if (user_mode(regs))
- tty_write_message(current->tty, buf);
+ tty_write_message(current->signal->tty, buf);
buf[len-1] = '\0'; /* drop '\r' */
printk(KERN_WARNING "%s", buf); /* watch for command names containing %s */
}
--- diff/arch/ia64/lib/dec_and_lock.c 2004-01-19 10:22:55.000000000 +0000
+++ source/arch/ia64/lib/dec_and_lock.c 2004-03-16 11:27:47.909904248 +0000
@@ -13,6 +13,7 @@
#include <linux/spinlock.h>
#include <asm/atomic.h>
+#ifndef CONFIG_LOCKMETER
/*
* Decrement REFCOUNT and if the count reaches zero, acquire the spinlock. Both of these
* operations have to be done atomically, so that the count doesn't drop to zero without
@@ -40,3 +41,4 @@ atomic_dec_and_lock (atomic_t *refcount,
}
EXPORT_SYMBOL(atomic_dec_and_lock);
+#endif
--- diff/arch/ia64/mm/hugetlbpage.c 2004-03-11 10:20:20.000000000 +0000
+++ source/arch/ia64/mm/hugetlbpage.c 2004-03-16 11:27:47.910904096 +0000
@@ -32,7 +32,7 @@ static spinlock_t htlbpage_lock = SPIN_L
static void enqueue_huge_page(struct page *page)
{
- list_add(&page->list,
+ list_add(&page->lru,
&hugepage_freelists[page_zone(page)->zone_pgdat->node_id]);
}
@@ -48,8 +48,8 @@ static struct page *dequeue_huge_page(vo
}
if (nid >= 0 && nid < MAX_NUMNODES &&
!list_empty(&hugepage_freelists[nid])) {
- page = list_entry(hugepage_freelists[nid].next, struct page, list);
- list_del(&page->list);
+ page = list_entry(hugepage_freelists[nid].next, struct page, lru);
+ list_del(&page->lru);
}
return page;
}
@@ -246,9 +246,8 @@ follow_huge_pmd(struct mm_struct *mm, un
void free_huge_page(struct page *page)
{
BUG_ON(page_count(page));
- BUG_ON(page->mapping);
- INIT_LIST_HEAD(&page->list);
+ INIT_LIST_HEAD(&page->lru);
spin_lock(&htlbpage_lock);
enqueue_huge_page(page);
@@ -449,19 +448,19 @@ int try_to_free_low(int count)
spin_lock(&htlbpage_lock);
list_for_each(p, &hugepage_freelists[0]) {
if (map) {
- list_del(&map->list);
+ list_del(&map->lru);
update_and_free_page(map);
htlbpagemem--;
map = NULL;
if (++count == 0)
break;
}
- page = list_entry(p, struct page, list);
+ page = list_entry(p, struct page, lru);
if (!PageHighMem(page))
map = page;
}
if (map) {
- list_del(&map->list);
+ list_del(&map->lru);
update_and_free_page(map);
htlbpagemem--;
count++;
--- diff/arch/parisc/kernel/process.c 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/parisc/kernel/process.c 2004-03-16 11:27:47.910904096 +0000
@@ -325,7 +325,7 @@ copy_thread(int nr, unsigned long clone_
/* Use same stack depth as parent */
cregs->ksp = ((unsigned long)(ti))
- + (pregs->gr[21] & (INIT_THREAD_SIZE - 1));
+ + (pregs->gr[21] & (THREAD_SIZE - 1));
cregs->gr[30] = usp;
if (p->personality == PER_HPUX) {
#ifdef CONFIG_HPUX
--- diff/arch/ppc/boot/ld.script 2003-10-09 08:47:16.000000000 +0000
+++ source/arch/ppc/boot/ld.script 2004-03-16 11:27:47.951897864 +0000
@@ -82,6 +82,7 @@ SECTIONS
*(__ksymtab)
*(__ksymtab_strings)
*(__bug_table)
+ *(__kcrctab)
}
}
--- diff/arch/ppc/boot/openfirmware/Makefile 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/ppc/boot/openfirmware/Makefile 2004-03-16 11:27:47.952897712 +0000
@@ -104,10 +104,10 @@ quiet_cmd_gen-coff = COFF $@
$(HACKCOFF) $@ && \
ln -sf $(notdir $@) $(images)/zImage$(initrd).pmac
-$(images)/vmlinux.coff: $(obj)/coffboot
+$(images)/vmlinux.coff: $(obj)/coffboot $(boot)/ld.script
$(call cmd,gen-coff)
-$(images)/vmlinux.initrd.coff: $(obj)/coffboot.initrd
+$(images)/vmlinux.initrd.coff: $(obj)/coffboot.initrd $(boot)/ld.script
$(call cmd,gen-coff)
quiet_cmd_gen-elf-pmac = ELF $@
@@ -116,19 +116,19 @@ quiet_cmd_gen-elf-pmac = ELF $@
$(OBJCOPY) $@ $@ --add-section=.note=$(obj)/note \
-R .comment $(del-ramdisk-sec)
-$(images)/vmlinux.elf-pmac: $(obj)/image.o $(NEWWORLDOBJS) $(LIBS) $(obj)/note
+$(images)/vmlinux.elf-pmac: $(obj)/image.o $(NEWWORLDOBJS) $(LIBS) $(obj)/note $(boot)/ld.script
$(call cmd,gen-elf-pmac)
$(images)/vmlinux.initrd.elf-pmac: $(obj)/image.initrd.o $(NEWWORLDOBJS) \
- $(LIBS) $(obj)/note
+ $(LIBS) $(obj)/note $(boot)/ld.script
$(call cmd,gen-elf-pmac)
quiet_cmd_gen-chrp = CHRP $@
cmd_gen-chrp = $(LD) $(CHRP_LD_ARGS) -o $@ $^ && \
$(OBJCOPY) $@ $@ -R .comment $(del-ramdisk-sec)
-$(images)/zImage.chrp: $(CHRPOBJS) $(obj)/image.o $(LIBS)
+$(images)/zImage.chrp: $(CHRPOBJS) $(obj)/image.o $(LIBS) $(boot)/ld.script
$(call cmd,gen-chrp)
-$(images)/zImage.initrd.chrp: $(CHRPOBJS) $(obj)/image.initrd.o $(LIBS)
+$(images)/zImage.initrd.chrp: $(CHRPOBJS) $(obj)/image.initrd.o $(LIBS) $(boot)/ld.script
$(call cmd,gen-chrp)
quiet_cmd_addnote = ADDNOTE $@
--- diff/arch/ppc/platforms/proc_rtas.c 2003-09-30 14:46:12.000000000 +0000
+++ source/arch/ppc/platforms/proc_rtas.c 2004-03-16 11:27:47.953897560 +0000
@@ -20,6 +20,7 @@
#include <linux/ctype.h>
#include <linux/time.h>
#include <linux/string.h>
+#include <linux/init.h>
#include <asm/uaccess.h>
#include <asm/bitops.h>
@@ -194,18 +195,18 @@ int check_location (char *c, int idx, ch
/* ****************************************************************** */
/* MAIN */
/* ****************************************************************** */
-void proc_rtas_init(void)
+static int __init proc_rtas_init(void)
{
struct proc_dir_entry *entry;
rtas = find_devices("rtas");
if ((rtas == 0) || (_machine != _MACH_chrp)) {
- return;
+ return 1;
}
proc_rtas = proc_mkdir("rtas", 0);
if (proc_rtas == 0)
- return;
+ return 1;
/* /proc/rtas entries */
@@ -226,7 +227,10 @@ void proc_rtas_init(void)
entry = create_proc_entry("volume", S_IWUSR|S_IRUGO, proc_rtas);
if (entry) entry->proc_fops = &ppc_rtas_tone_volume_operations;
+
+ return 0;
}
+__initcall(proc_rtas_init);
/* ****************************************************************** */
/* POWER-ON-TIME */
--- diff/arch/ppc64/Kconfig 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/Kconfig 2004-03-16 11:27:47.914903488 +0000
@@ -173,6 +173,16 @@ config NUMA
bool "NUMA support"
depends on DISCONTIGMEM
+config SCHED_SMT
+ bool "SMT (Hyperthreading) scheduler support"
+ depends on SMP
+ default off
+ help
+ SMT scheduler support improves the CPU scheduler's decision making
+ when dealing with Intel Pentium 4 chips with HyperThreading at a
+ cost of slightly increased overhead in some places. If unsure say
+ N here.
+
config PREEMPT
bool
help
--- diff/arch/ppc64/Makefile 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/Makefile 2004-03-16 11:27:47.948898320 +0000
@@ -53,7 +53,7 @@ $(boottarget-y): vmlinux
$(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
bootimage-$(CONFIG_PPC_PSERIES) := zImage
-bootimage-$(CONFIG_PPC_ISERIES) := vmlinux.sm
+bootimage-$(CONFIG_PPC_ISERIES) := vmlinux
BOOTIMAGE := $(bootimage-y)
install: vmlinux
$(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(BOOTIMAGE) $@
--- diff/arch/ppc64/configs/iSeries_defconfig 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/configs/iSeries_defconfig 2004-03-16 11:27:47.911903944 +0000
@@ -47,6 +47,7 @@ CONFIG_MODULE_UNLOAD=y
CONFIG_OBSOLETE_MODPARM=y
# CONFIG_MODVERSIONS is not set
# CONFIG_KMOD is not set
+CONFIG_STOP_MACHINE=y
#
# Platform support
@@ -56,6 +57,7 @@ CONFIG_PPC_ISERIES=y
CONFIG_PPC=y
CONFIG_PPC64=y
# CONFIG_POWER4_ONLY is not set
+# CONFIG_IOMMU_VMERGE is not set
CONFIG_SMP=y
CONFIG_NR_CPUS=32
CONFIG_MSCHUNKS=y
@@ -89,6 +91,7 @@ CONFIG_PCI_NAMES=y
# Generic Driver Options
#
CONFIG_FW_LOADER=m
+# CONFIG_DEBUG_DRIVER is not set
#
# Memory Technology Devices (MTD)
@@ -115,10 +118,10 @@ CONFIG_FW_LOADER=m
CONFIG_BLK_DEV_LOOP=y
# CONFIG_BLK_DEV_CRYPTOLOOP is not set
CONFIG_BLK_DEV_NBD=m
+# CONFIG_BLK_DEV_CARMEL is not set
CONFIG_BLK_DEV_RAM=y
CONFIG_BLK_DEV_RAM_SIZE=4096
CONFIG_BLK_DEV_INITRD=y
-# CONFIG_DCSSBLK is not set
#
# ATA/ATAPI/MFM/RLL support
@@ -150,6 +153,12 @@ CONFIG_SCSI_CONSTANTS=y
# CONFIG_SCSI_LOGGING is not set
#
+# SCSI Transport Attributes
+#
+CONFIG_SCSI_SPI_ATTRS=y
+CONFIG_SCSI_FC_ATTRS=y
+
+#
# SCSI low-level drivers
#
# CONFIG_BLK_DEV_3W_XXXX_RAID is not set
@@ -197,7 +206,6 @@ CONFIG_MD_RAID5=y
CONFIG_MD_RAID6=y
# CONFIG_MD_MULTIPATH is not set
CONFIG_BLK_DEV_DM=y
-CONFIG_DM_IOCTL_V4=y
CONFIG_DM_CRYPT=m
#
@@ -437,6 +445,7 @@ CONFIG_IBMOL=y
# CONFIG_TMS380TR is not set
# CONFIG_NET_FC is not set
# CONFIG_SHAPER is not set
+CONFIG_NETCONSOLE=y
#
# Wan interfaces
@@ -457,6 +466,10 @@ CONFIG_IBMOL=y
# Bluetooth support
#
# CONFIG_BT is not set
+CONFIG_NETPOLL=y
+CONFIG_NETPOLL_RX=y
+CONFIG_NETPOLL_TRAP=y
+CONFIG_NET_POLL_CONTROLLER=y
#
# ISDN subsystem
@@ -535,7 +548,6 @@ CONFIG_LEGACY_PTY_COUNT=256
# Watchdog Cards
#
# CONFIG_WATCHDOG is not set
-# CONFIG_NVRAM is not set
# CONFIG_RTC is not set
# CONFIG_GEN_RTC is not set
# CONFIG_DTLK is not set
@@ -556,6 +568,10 @@ CONFIG_MAX_RAW_DEVS=256
# CONFIG_I2C is not set
#
+# Misc devices
+#
+
+#
# Multimedia devices
#
# CONFIG_VIDEO_DEV is not set
@@ -742,7 +758,6 @@ CONFIG_NLS_DEFAULT="iso8859-1"
CONFIG_VIOCONS=y
CONFIG_VIODASD=y
CONFIG_VIOCD=y
-# CONFIG_VIOCD_AZTECH is not set
# CONFIG_VIOTAPE is not set
CONFIG_VIOPATH=y
@@ -787,6 +802,7 @@ CONFIG_CRYPTO_SERPENT=m
CONFIG_CRYPTO_AES=m
CONFIG_CRYPTO_CAST5=m
CONFIG_CRYPTO_CAST6=m
+CONFIG_CRYPTO_ARC4=m
CONFIG_CRYPTO_DEFLATE=m
CONFIG_CRYPTO_TEST=m
--- diff/arch/ppc64/configs/pSeries_defconfig 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/configs/pSeries_defconfig 2004-03-16 11:27:47.912903792 +0000
@@ -47,6 +47,7 @@ CONFIG_MODULE_UNLOAD=y
CONFIG_OBSOLETE_MODPARM=y
# CONFIG_MODVERSIONS is not set
# CONFIG_KMOD is not set
+CONFIG_STOP_MACHINE=y
#
# Platform support
@@ -60,6 +61,7 @@ CONFIG_ALTIVEC=y
# CONFIG_PPC_PMAC is not set
# CONFIG_BOOTX_TEXT is not set
# CONFIG_POWER4_ONLY is not set
+# CONFIG_IOMMU_VMERGE is not set
CONFIG_SMP=y
CONFIG_IRQ_ALL_CPUS=y
CONFIG_NR_CPUS=32
@@ -106,6 +108,7 @@ CONFIG_PROC_DEVICETREE=y
# Generic Driver Options
#
CONFIG_FW_LOADER=m
+# CONFIG_DEBUG_DRIVER is not set
#
# Memory Technology Devices (MTD)
@@ -132,10 +135,10 @@ CONFIG_BLK_DEV_FD=y
CONFIG_BLK_DEV_LOOP=y
# CONFIG_BLK_DEV_CRYPTOLOOP is not set
CONFIG_BLK_DEV_NBD=m
+# CONFIG_BLK_DEV_CARMEL is not set
CONFIG_BLK_DEV_RAM=y
CONFIG_BLK_DEV_RAM_SIZE=4096
CONFIG_BLK_DEV_INITRD=y
-# CONFIG_DCSSBLK is not set
#
# ATA/ATAPI/MFM/RLL support
@@ -173,7 +176,7 @@ CONFIG_IDEDMA_PCI_AUTO=y
CONFIG_BLK_DEV_ADMA=y
# CONFIG_BLK_DEV_AEC62XX is not set
# CONFIG_BLK_DEV_ALI15X3 is not set
-# CONFIG_BLK_DEV_AMD74XX is not set
+CONFIG_BLK_DEV_AMD74XX=y
# CONFIG_BLK_DEV_CMD64X is not set
# CONFIG_BLK_DEV_TRIFLEX is not set
# CONFIG_BLK_DEV_CY82C693 is not set
@@ -222,6 +225,12 @@ CONFIG_SCSI_CONSTANTS=y
# CONFIG_SCSI_LOGGING is not set
#
+# SCSI Transport Attributes
+#
+CONFIG_SCSI_SPI_ATTRS=y
+CONFIG_SCSI_FC_ATTRS=y
+
+#
# SCSI low-level drivers
#
# CONFIG_BLK_DEV_3W_XXXX_RAID is not set
@@ -273,7 +282,6 @@ CONFIG_MD_RAID5=y
CONFIG_MD_RAID6=y
# CONFIG_MD_MULTIPATH is not set
CONFIG_BLK_DEV_DM=y
-CONFIG_DM_IOCTL_V4=y
CONFIG_DM_CRYPT=m
#
@@ -515,6 +523,7 @@ CONFIG_IBMOL=y
# CONFIG_TMS380TR is not set
# CONFIG_NET_FC is not set
# CONFIG_SHAPER is not set
+CONFIG_NETCONSOLE=y
#
# Wan interfaces
@@ -535,6 +544,10 @@ CONFIG_IBMOL=y
# Bluetooth support
#
# CONFIG_BT is not set
+CONFIG_NETPOLL=y
+CONFIG_NETPOLL_RX=y
+CONFIG_NETPOLL_TRAP=y
+CONFIG_NET_POLL_CONTROLLER=y
#
# ISDN subsystem
@@ -633,7 +646,6 @@ CONFIG_HVC_CONSOLE=y
# Watchdog Cards
#
# CONFIG_WATCHDOG is not set
-# CONFIG_NVRAM is not set
# CONFIG_RTC is not set
# CONFIG_GEN_RTC is not set
# CONFIG_DTLK is not set
@@ -707,6 +719,10 @@ CONFIG_I2C_ALGOBIT=y
# CONFIG_I2C_DEBUG_CHIP is not set
#
+# Misc devices
+#
+
+#
# Multimedia devices
#
# CONFIG_VIDEO_DEV is not set
@@ -720,6 +736,7 @@ CONFIG_I2C_ALGOBIT=y
# Graphics support
#
CONFIG_FB=y
+# CONFIG_FB_PM2 is not set
# CONFIG_FB_CYBER2000 is not set
CONFIG_FB_OF=y
# CONFIG_FB_CT65550 is not set
@@ -776,7 +793,7 @@ CONFIG_LOGO_LINUX_CLUT224=y
#
# USB support
#
-CONFIG_USB=m
+CONFIG_USB=y
# CONFIG_USB_DEBUG is not set
#
@@ -789,8 +806,8 @@ CONFIG_USB_DEVICEFS=y
#
# USB Host Controller Drivers
#
-CONFIG_USB_EHCI_HCD=m
-CONFIG_USB_OHCI_HCD=m
+CONFIG_USB_EHCI_HCD=y
+CONFIG_USB_OHCI_HCD=y
# CONFIG_USB_UHCI_HCD is not set
#
@@ -799,7 +816,7 @@ CONFIG_USB_OHCI_HCD=m
# CONFIG_USB_BLUETOOTH_TTY is not set
# CONFIG_USB_ACM is not set
# CONFIG_USB_PRINTER is not set
-CONFIG_USB_STORAGE=m
+CONFIG_USB_STORAGE=y
# CONFIG_USB_STORAGE_DEBUG is not set
# CONFIG_USB_STORAGE_DATAFAB is not set
# CONFIG_USB_STORAGE_FREECOM is not set
@@ -813,16 +830,10 @@ CONFIG_USB_STORAGE=m
#
# USB Human Interface Devices (HID)
#
-CONFIG_USB_HID=m
+CONFIG_USB_HID=y
CONFIG_USB_HIDINPUT=y
# CONFIG_HID_FF is not set
CONFIG_USB_HIDDEV=y
-
-#
-# USB HID Boot Protocol drivers
-#
-# CONFIG_USB_KBD is not set
-# CONFIG_USB_MOUSE is not set
# CONFIG_USB_AIPTEK is not set
# CONFIG_USB_WACOM is not set
# CONFIG_USB_KBTAB is not set
@@ -1076,6 +1087,7 @@ CONFIG_CRYPTO_SERPENT=m
CONFIG_CRYPTO_AES=m
CONFIG_CRYPTO_CAST5=m
CONFIG_CRYPTO_CAST6=m
+CONFIG_CRYPTO_ARC4=m
CONFIG_CRYPTO_DEFLATE=m
CONFIG_CRYPTO_TEST=m
--- diff/arch/ppc64/defconfig 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/defconfig 2004-03-16 11:27:47.913903640 +0000
@@ -47,6 +47,7 @@ CONFIG_MODULE_UNLOAD=y
CONFIG_OBSOLETE_MODPARM=y
# CONFIG_MODVERSIONS is not set
# CONFIG_KMOD is not set
+CONFIG_STOP_MACHINE=y
#
# Platform support
@@ -60,6 +61,7 @@ CONFIG_ALTIVEC=y
# CONFIG_PPC_PMAC is not set
# CONFIG_BOOTX_TEXT is not set
# CONFIG_POWER4_ONLY is not set
+# CONFIG_IOMMU_VMERGE is not set
CONFIG_SMP=y
CONFIG_IRQ_ALL_CPUS=y
CONFIG_NR_CPUS=32
@@ -106,6 +108,7 @@ CONFIG_PROC_DEVICETREE=y
# Generic Driver Options
#
CONFIG_FW_LOADER=m
+# CONFIG_DEBUG_DRIVER is not set
#
# Memory Technology Devices (MTD)
@@ -132,10 +135,10 @@ CONFIG_BLK_DEV_FD=y
CONFIG_BLK_DEV_LOOP=y
# CONFIG_BLK_DEV_CRYPTOLOOP is not set
CONFIG_BLK_DEV_NBD=m
+# CONFIG_BLK_DEV_CARMEL is not set
CONFIG_BLK_DEV_RAM=y
CONFIG_BLK_DEV_RAM_SIZE=4096
CONFIG_BLK_DEV_INITRD=y
-# CONFIG_DCSSBLK is not set
#
# ATA/ATAPI/MFM/RLL support
@@ -173,7 +176,7 @@ CONFIG_IDEDMA_PCI_AUTO=y
CONFIG_BLK_DEV_ADMA=y
# CONFIG_BLK_DEV_AEC62XX is not set
# CONFIG_BLK_DEV_ALI15X3 is not set
-# CONFIG_BLK_DEV_AMD74XX is not set
+CONFIG_BLK_DEV_AMD74XX=y
# CONFIG_BLK_DEV_CMD64X is not set
# CONFIG_BLK_DEV_TRIFLEX is not set
# CONFIG_BLK_DEV_CY82C693 is not set
@@ -222,6 +225,12 @@ CONFIG_SCSI_CONSTANTS=y
# CONFIG_SCSI_LOGGING is not set
#
+# SCSI Transport Attributes
+#
+CONFIG_SCSI_SPI_ATTRS=y
+CONFIG_SCSI_FC_ATTRS=y
+
+#
# SCSI low-level drivers
#
# CONFIG_BLK_DEV_3W_XXXX_RAID is not set
@@ -273,7 +282,6 @@ CONFIG_MD_RAID5=y
CONFIG_MD_RAID6=y
# CONFIG_MD_MULTIPATH is not set
CONFIG_BLK_DEV_DM=y
-CONFIG_DM_IOCTL_V4=y
CONFIG_DM_CRYPT=m
#
@@ -515,6 +523,7 @@ CONFIG_IBMOL=y
# CONFIG_TMS380TR is not set
# CONFIG_NET_FC is not set
# CONFIG_SHAPER is not set
+CONFIG_NETCONSOLE=y
#
# Wan interfaces
@@ -535,6 +544,10 @@ CONFIG_IBMOL=y
# Bluetooth support
#
# CONFIG_BT is not set
+CONFIG_NETPOLL=y
+CONFIG_NETPOLL_RX=y
+CONFIG_NETPOLL_TRAP=y
+CONFIG_NET_POLL_CONTROLLER=y
#
# ISDN subsystem
@@ -633,7 +646,6 @@ CONFIG_HVC_CONSOLE=y
# Watchdog Cards
#
# CONFIG_WATCHDOG is not set
-# CONFIG_NVRAM is not set
# CONFIG_RTC is not set
# CONFIG_GEN_RTC is not set
# CONFIG_DTLK is not set
@@ -707,6 +719,10 @@ CONFIG_I2C_ALGOBIT=y
# CONFIG_I2C_DEBUG_CHIP is not set
#
+# Misc devices
+#
+
+#
# Multimedia devices
#
# CONFIG_VIDEO_DEV is not set
@@ -720,6 +736,7 @@ CONFIG_I2C_ALGOBIT=y
# Graphics support
#
CONFIG_FB=y
+# CONFIG_FB_PM2 is not set
# CONFIG_FB_CYBER2000 is not set
CONFIG_FB_OF=y
# CONFIG_FB_CT65550 is not set
@@ -776,7 +793,7 @@ CONFIG_LOGO_LINUX_CLUT224=y
#
# USB support
#
-CONFIG_USB=m
+CONFIG_USB=y
# CONFIG_USB_DEBUG is not set
#
@@ -789,8 +806,8 @@ CONFIG_USB_DEVICEFS=y
#
# USB Host Controller Drivers
#
-CONFIG_USB_EHCI_HCD=m
-CONFIG_USB_OHCI_HCD=m
+CONFIG_USB_EHCI_HCD=y
+CONFIG_USB_OHCI_HCD=y
# CONFIG_USB_UHCI_HCD is not set
#
@@ -799,7 +816,7 @@ CONFIG_USB_OHCI_HCD=m
# CONFIG_USB_BLUETOOTH_TTY is not set
# CONFIG_USB_ACM is not set
# CONFIG_USB_PRINTER is not set
-CONFIG_USB_STORAGE=m
+CONFIG_USB_STORAGE=y
# CONFIG_USB_STORAGE_DEBUG is not set
# CONFIG_USB_STORAGE_DATAFAB is not set
# CONFIG_USB_STORAGE_FREECOM is not set
@@ -813,16 +830,10 @@ CONFIG_USB_STORAGE=m
#
# USB Human Interface Devices (HID)
#
-CONFIG_USB_HID=m
+CONFIG_USB_HID=y
CONFIG_USB_HIDINPUT=y
# CONFIG_HID_FF is not set
CONFIG_USB_HIDDEV=y
-
-#
-# USB HID Boot Protocol drivers
-#
-# CONFIG_USB_KBD is not set
-# CONFIG_USB_MOUSE is not set
# CONFIG_USB_AIPTEK is not set
# CONFIG_USB_WACOM is not set
# CONFIG_USB_KBTAB is not set
@@ -1076,6 +1087,7 @@ CONFIG_CRYPTO_SERPENT=m
CONFIG_CRYPTO_AES=m
CONFIG_CRYPTO_CAST5=m
CONFIG_CRYPTO_CAST6=m
+CONFIG_CRYPTO_ARC4=m
CONFIG_CRYPTO_DEFLATE=m
CONFIG_CRYPTO_TEST=m
--- diff/arch/ppc64/kernel/HvLpConfig.c 2002-10-16 03:27:53.000000000 +0000
+++ source/arch/ppc64/kernel/HvLpConfig.c 2004-03-16 11:27:47.920902576 +0000
@@ -17,12 +17,11 @@
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
-#ifndef _HVLPCONFIG_H
+#include <linux/module.h>
#include <asm/iSeries/HvLpConfig.h>
-#endif
HvLpIndex HvLpConfig_getLpIndex_outline(void)
{
return HvLpConfig_getLpIndex();
}
-
+EXPORT_SYMBOL(HvLpConfig_getLpIndex_outline);
--- diff/arch/ppc64/kernel/Makefile 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/Makefile 2004-03-16 11:27:47.925901816 +0000
@@ -5,11 +5,11 @@
EXTRA_CFLAGS += -mno-minimal-toc
extra-y := head.o vmlinux.lds.s
-obj-y := setup.o entry.o traps.o irq.o idle.o \
+obj-y := setup.o entry.o traps.o irq.o idle.o dma.o \
time.o process.o signal.o syscalls.o misc.o ptrace.o \
align.o semaphore.o bitops.o stab.o pacaData.o \
udbg.o binfmt_elf32.o sys_ppc32.o ioctl32.o \
- ptrace32.o signal32.o pmc.o rtc.o init_task.o \
+ ptrace32.o signal32.o rtc.o init_task.o \
lmb.o cputable.o cpu_setup_power4.o idle_power4.o \
iommu.o
@@ -29,7 +29,7 @@ obj-$(CONFIG_PPC_ISERIES) += iSeries_irq
HvCall.o HvLpConfig.o LparData.o mf_proc.o \
iSeries_setup.o ItLpQueue.o hvCall.o \
mf.o HvLpEvent.o iSeries_proc.o iSeries_htab.o \
- proc_pmc.o iSeries_iommu.o
+ iSeries_iommu.o
obj-$(CONFIG_PPC_PSERIES) += pSeries_pci.o pSeries_lpar.o pSeries_hvCall.o \
eeh.o nvram.o pSeries_nvram.o rtasd.o ras.o \
--- diff/arch/ppc64/kernel/bitops.c 2002-10-16 03:28:27.000000000 +0000
+++ source/arch/ppc64/kernel/bitops.c 2004-03-16 11:27:47.914903488 +0000
@@ -3,6 +3,7 @@
*/
#include <linux/kernel.h>
+#include <linux/module.h>
#include <asm/bitops.h>
#include <asm/byteorder.h>
@@ -44,8 +45,10 @@ found_first:
found_middle:
return result + ffz(tmp);
}
+EXPORT_SYMBOL(find_next_zero_bit);
-unsigned long find_next_bit(unsigned long *addr, unsigned long size, unsigned long offset)
+unsigned long find_next_bit(unsigned long *addr, unsigned long size,
+ unsigned long offset)
{
unsigned long *p = addr + (offset >> 6);
unsigned long result = offset & ~63UL;
@@ -82,8 +85,9 @@ found_first:
found_middle:
return result + __ffs(tmp);
}
+EXPORT_SYMBOL(find_next_bit);
-static __inline__ unsigned int ext2_ilog2(unsigned int x)
+static inline unsigned int ext2_ilog2(unsigned int x)
{
int lz;
@@ -91,16 +95,17 @@ static __inline__ unsigned int ext2_ilog
return 31 - lz;
}
-static __inline__ unsigned int ext2_ffz(unsigned int x)
+static inline unsigned int ext2_ffz(unsigned int x)
{
- u32 tempRC;
+ u32 rc;
if ((x = ~x) == 0)
return 32;
- tempRC = ext2_ilog2(x & -x);
- return tempRC;
+ rc = ext2_ilog2(x & -x);
+ return rc;
}
-unsigned long find_next_zero_le_bit(unsigned long *addr, unsigned long size, unsigned long offset)
+unsigned long find_next_zero_le_bit(unsigned long *addr, unsigned long size,
+ unsigned long offset)
{
unsigned int *p = ((unsigned int *) addr) + (offset >> 5);
unsigned int result = offset & ~31;
@@ -136,3 +141,4 @@ found_first:
found_middle:
return result + ext2_ffz(tmp);
}
+EXPORT_SYMBOL(find_next_zero_le_bit);
--- diff/arch/ppc64/kernel/eeh.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/eeh.c 2004-03-16 11:27:47.919902728 +0000
@@ -17,63 +17,348 @@
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
-/* Change Activity:
- * 2001/10/27 : engebret : Created.
- * End Change Activity
- */
-
#include <linux/init.h>
#include <linux/pci.h>
#include <linux/proc_fs.h>
#include <linux/bootmem.h>
#include <linux/mm.h>
+#include <linux/rbtree.h>
+#include <linux/spinlock.h>
+#include <linux/seq_file.h>
#include <asm/paca.h>
#include <asm/processor.h>
#include <asm/naca.h>
#include <asm/io.h>
#include <asm/machdep.h>
+#include <asm/pgtable.h>
#include "pci.h"
+#undef DEBUG
+
#define BUID_HI(buid) ((buid) >> 32)
#define BUID_LO(buid) ((buid) & 0xffffffff)
-#define CONFIG_ADDR(busno, devfn) (((((busno) & 0xff) << 8) | ((devfn) & 0xf8)) << 8)
+#define CONFIG_ADDR(busno, devfn) \
+ (((((busno) & 0xff) << 8) | ((devfn) & 0xf8)) << 8)
-unsigned long eeh_total_mmio_ffs;
-unsigned long eeh_false_positives;
/* RTAS tokens */
static int ibm_set_eeh_option;
static int ibm_set_slot_reset;
static int ibm_read_slot_reset_state;
-static int eeh_implemented;
+static int eeh_subsystem_enabled;
#define EEH_MAX_OPTS 4096
static char *eeh_opts;
static int eeh_opts_last;
-unsigned char slot_err_buf[RTAS_ERROR_LOG_MAX];
+/* System monitoring statistics */
+static DEFINE_PER_CPU(unsigned long, total_mmio_ffs);
+static DEFINE_PER_CPU(unsigned long, false_positives);
+static DEFINE_PER_CPU(unsigned long, ignored_failures);
-pte_t *find_linux_pte(pgd_t *pgdir, unsigned long va); /* from htab.c */
-static int eeh_check_opts_config(struct device_node *dn,
- int class_code, int vendor_id, int device_id,
+static int eeh_check_opts_config(struct device_node *dn, int class_code,
+ int vendor_id, int device_id,
int default_state);
-unsigned long eeh_token_to_phys(unsigned long token)
+/**
+ * The pci address cache subsystem. This subsystem places
+ * PCI device address resources into a red-black tree, sorted
+ * according to the address range, so that given only an i/o
+ * address, the corresponding PCI device can be **quickly**
+ * found.
+ *
+ * Currently, the only customer of this code is the EEH subsystem;
+ * thus, this code has been somewhat tailored to suit EEH better.
+ * In particular, the cache does *not* hold the addresses of devices
+ * for which EEH is not enabled.
+ *
+ * (Implementation Note: The RB tree seems to be better/faster
+ * than any hash algo I could think of for this problem, even
+ * with the penalty of slow pointer chases for d-cache misses).
+ */
+struct pci_io_addr_range
+{
+ struct rb_node rb_node;
+ unsigned long addr_lo;
+ unsigned long addr_hi;
+ struct pci_dev *pcidev;
+ unsigned int flags;
+};
+
+static struct pci_io_addr_cache
+{
+ struct rb_root rb_root;
+ spinlock_t piar_lock;
+} pci_io_addr_cache_root;
+
+static inline struct pci_dev *__pci_get_device_by_addr(unsigned long addr)
+{
+ struct rb_node *n = pci_io_addr_cache_root.rb_root.rb_node;
+
+ while (n) {
+ struct pci_io_addr_range *piar;
+ piar = rb_entry(n, struct pci_io_addr_range, rb_node);
+
+ if (addr < piar->addr_lo) {
+ n = n->rb_left;
+ } else {
+ if (addr > piar->addr_hi) {
+ n = n->rb_right;
+ } else {
+ pci_dev_get(piar->pcidev);
+ return piar->pcidev;
+ }
+ }
+ }
+
+ return NULL;
+}
+
+/**
+ * pci_get_device_by_addr - Get device, given only address
+ * @addr: mmio (PIO) phys address or i/o port number
+ *
+ * Given an mmio phys address, or a port number, find a pci device
+ * that implements this address. Be sure to pci_dev_put the device
+ * when finished. I/O port numbers are assumed to be offset
+ * from zero (that is, they do *not* have pci_io_addr added in).
+ * It is safe to call this function within an interrupt.
+ */
+static struct pci_dev *pci_get_device_by_addr(unsigned long addr)
+{
+ struct pci_dev *dev;
+ unsigned long flags;
+
+ spin_lock_irqsave(&pci_io_addr_cache_root.piar_lock, flags);
+ dev = __pci_get_device_by_addr(addr);
+ spin_unlock_irqrestore(&pci_io_addr_cache_root.piar_lock, flags);
+ return dev;
+}
+
+#ifdef DEBUG
+/*
+ * Handy-dandy debug print routine, does nothing more
+ * than print out the contents of our addr cache.
+ */
+static void pci_addr_cache_print(struct pci_io_addr_cache *cache)
+{
+ struct rb_node *n;
+ int cnt = 0;
+
+ n = rb_first(&cache->rb_root);
+ while (n) {
+ struct pci_io_addr_range *piar;
+ piar = rb_entry(n, struct pci_io_addr_range, rb_node);
+ printk(KERN_DEBUG "PCI: %s addr range %d [%lx-%lx]: %s %s\n",
+ (piar->flags & IORESOURCE_IO) ? "i/o" : "mem", cnt,
+ piar->addr_lo, piar->addr_hi, pci_name(piar->pcidev),
+ pci_pretty_name(piar->pcidev));
+ cnt++;
+ n = rb_next(n);
+ }
+}
+#endif
+
+/* Insert address range into the rb tree. */
+static struct pci_io_addr_range *
+pci_addr_cache_insert(struct pci_dev *dev, unsigned long alo,
+ unsigned long ahi, unsigned int flags)
+{
+ struct rb_node **p = &pci_io_addr_cache_root.rb_root.rb_node;
+ struct rb_node *parent = NULL;
+ struct pci_io_addr_range *piar;
+
+ /* Walk tree, find a place to insert into tree */
+ while (*p) {
+ parent = *p;
+ piar = rb_entry(parent, struct pci_io_addr_range, rb_node);
+ if (alo < piar->addr_lo) {
+ p = &parent->rb_left;
+ } else if (ahi > piar->addr_hi) {
+ p = &parent->rb_right;
+ } else {
+ if (dev != piar->pcidev ||
+ alo != piar->addr_lo || ahi != piar->addr_hi) {
+ printk(KERN_WARNING "PIAR: overlapping address range\n");
+ }
+ return piar;
+ }
+ }
+ piar = (struct pci_io_addr_range *)kmalloc(sizeof(struct pci_io_addr_range), GFP_ATOMIC);
+ if (!piar)
+ return NULL;
+
+ piar->addr_lo = alo;
+ piar->addr_hi = ahi;
+ piar->pcidev = dev;
+ piar->flags = flags;
+
+ rb_link_node(&piar->rb_node, parent, p);
+ rb_insert_color(&piar->rb_node, &pci_io_addr_cache_root.rb_root);
+
+ return piar;
+}
+
+static void __pci_addr_cache_insert_device(struct pci_dev *dev)
+{
+ struct device_node *dn;
+ int i;
+
+ dn = pci_device_to_OF_node(dev);
+ if (!dn) {
+ printk(KERN_WARNING "PCI: no pci dn found for dev=%s %s\n",
+ pci_name(dev), pci_pretty_name(dev));
+ pci_dev_put(dev);
+ return;
+ }
+
+ /* Skip any devices for which EEH is not enabled. */
+ if (!(dn->eeh_mode & EEH_MODE_SUPPORTED) ||
+ dn->eeh_mode & EEH_MODE_NOCHECK) {
+#ifdef DEBUG
+ printk(KERN_INFO "PCI: skip building address cache for=%s %s\n",
+ pci_name(dev), pci_pretty_name(dev));
+#endif
+ pci_dev_put(dev);
+ return;
+ }
+
+ /* Walk resources on this device, poke them into the tree */
+ for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) {
+ unsigned long start = pci_resource_start(dev,i);
+ unsigned long end = pci_resource_end(dev,i);
+ unsigned int flags = pci_resource_flags(dev,i);
+
+ /* We are interested only bus addresses, not dma or other stuff */
+ if (0 == (flags & (IORESOURCE_IO | IORESOURCE_MEM)))
+ continue;
+ if (start == 0 || ~start == 0 || end == 0 || ~end == 0)
+ continue;
+ pci_addr_cache_insert(dev, start, end, flags);
+ }
+}
+
+/**
+ * pci_addr_cache_insert_device - Add a device to the address cache
+ * @dev: PCI device whose I/O addresses we are interested in.
+ *
+ * In order to support the fast lookup of devices based on addresses,
+ * we maintain a cache of devices that can be quickly searched.
+ * This routine adds a device to that cache.
+ */
+void pci_addr_cache_insert_device(struct pci_dev *dev)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&pci_io_addr_cache_root.piar_lock, flags);
+ __pci_addr_cache_insert_device(dev);
+ spin_unlock_irqrestore(&pci_io_addr_cache_root.piar_lock, flags);
+}
+
+static inline void __pci_addr_cache_remove_device(struct pci_dev *dev)
+{
+ struct rb_node *n;
+
+restart:
+ n = rb_first(&pci_io_addr_cache_root.rb_root);
+ while (n) {
+ struct pci_io_addr_range *piar;
+ piar = rb_entry(n, struct pci_io_addr_range, rb_node);
+
+ if (piar->pcidev == dev) {
+ rb_erase(n, &pci_io_addr_cache_root.rb_root);
+ kfree(piar);
+ goto restart;
+ }
+ n = rb_next(n);
+ }
+ pci_dev_put(dev);
+}
+
+/**
+ * pci_addr_cache_remove_device - remove pci device from addr cache
+ * @dev: device to remove
+ *
+ * Remove a device from the addr-cache tree.
+ * This is potentially expensive, since it will walk
+ * the tree multiple times (once per resource).
+ * But so what; device removal doesn't need to be that fast.
+ */
+void pci_addr_cache_remove_device(struct pci_dev *dev)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&pci_io_addr_cache_root.piar_lock, flags);
+ __pci_addr_cache_remove_device(dev);
+ spin_unlock_irqrestore(&pci_io_addr_cache_root.piar_lock, flags);
+}
+
+/**
+ * pci_addr_cache_build - Build a cache of I/O addresses
+ *
+ * Build a cache of pci i/o addresses. This cache will be used to
+ * find the pci device that corresponds to a given address.
+ * This routine scans all pci busses to build the cache.
+ * Must be run late in boot process, after the pci controllers
+ * have been scaned for devices (after all device resources are known).
+ */
+void __init pci_addr_cache_build(void)
+{
+ struct pci_dev *dev = NULL;
+
+ spin_lock_init(&pci_io_addr_cache_root.piar_lock);
+
+ while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
+ /* Ignore PCI bridges ( XXX why ??) */
+ if ((dev->class >> 16) == PCI_BASE_CLASS_BRIDGE) {
+ pci_dev_put(dev);
+ continue;
+ }
+ pci_addr_cache_insert_device(dev);
+ }
+
+#ifdef DEBUG
+ /* Verify tree built up above, echo back the list of addrs. */
+ pci_addr_cache_print(&pci_io_addr_cache_root);
+#endif
+}
+
+/**
+ * eeh_token_to_phys - convert EEH address token to phys address
+ * @token i/o token, should be address in the form 0xA....
+ *
+ * Converts EEH address tokens into physical addresses. Note that
+ * ths routine does *not* convert I/O BAR addresses (which start
+ * with 0xE...) to phys addresses!
+ */
+static unsigned long eeh_token_to_phys(unsigned long token)
{
- if (REGION_ID(token) == EEH_REGION_ID) {
- unsigned long vaddr = IO_TOKEN_TO_ADDR(token);
- pte_t *ptep = find_linux_pte(ioremap_mm.pgd, vaddr);
- unsigned long pa = pte_pfn(*ptep) << PAGE_SHIFT;
- return pa | (vaddr & (PAGE_SIZE-1));
- } else
+ pte_t *ptep;
+ unsigned long pa, vaddr;
+
+ if (REGION_ID(token) == EEH_REGION_ID)
+ vaddr = IO_TOKEN_TO_ADDR(token);
+ else
return token;
+
+ ptep = find_linux_pte(ioremap_mm.pgd, vaddr);
+ pa = pte_pfn(*ptep) << PAGE_SHIFT;
+
+ return pa | (vaddr & (PAGE_SIZE-1));
}
-/* Check for an eeh failure at the given token address.
+/**
+ * eeh_check_failure - check if all 1's data is due to EEH slot freeze
+ * @token i/o token, should be address in the form 0xA....
+ * @val value, should be all 1's (XXX why do we need this arg??)
+ *
+ * Check for an eeh failure at the given token address.
* The given value has been read and it should be 1's (0xff, 0xffff or
* 0xffffffff).
*
* Probe to determine if an error actually occurred. If not return val.
* Otherwise panic.
+ *
+ * Note this routine might be called in an interrupt context ...
*/
unsigned long eeh_check_failure(void *token, unsigned long val)
{
@@ -81,81 +366,97 @@ unsigned long eeh_check_failure(void *to
struct pci_dev *dev;
struct device_node *dn;
unsigned long ret, rets[2];
+ static spinlock_t lock = SPIN_LOCK_UNLOCKED;
+ /* dont want this on the stack */
+ static unsigned char slot_err_buf[RTAS_ERROR_LOG_MAX];
+ unsigned long flags;
- /* IO BAR access could get us here...or if we manually force EEH
- * operation on even if the hardware won't support it.
- */
- if (!eeh_implemented || ibm_read_slot_reset_state == RTAS_UNKNOWN_SERVICE)
+ __get_cpu_var(total_mmio_ffs)++;
+
+ if (!eeh_subsystem_enabled)
return val;
- /* Finding the phys addr + pci device is quite expensive.
- * However, the RTAS call is MUCH slower.... :(
- */
+ /* Finding the phys addr + pci device; this is pretty quick. */
addr = eeh_token_to_phys((unsigned long)token);
- dev = pci_find_dev_by_addr(addr);
- if (!dev) {
- printk("EEH: no pci dev found for addr=0x%lx\n", addr);
+ dev = pci_get_device_by_addr(addr);
+ if (!dev)
return val;
- }
+
dn = pci_device_to_OF_node(dev);
if (!dn) {
- printk("EEH: no pci dn found for addr=0x%lx\n", addr);
+ pci_dev_put(dev);
return val;
}
/* Access to IO BARs might get this far and still not want checking. */
- if (!(dn->eeh_mode & EEH_MODE_SUPPORTED) || dn->eeh_mode & EEH_MODE_NOCHECK)
+ if (!(dn->eeh_mode & EEH_MODE_SUPPORTED) ||
+ dn->eeh_mode & EEH_MODE_NOCHECK) {
+ pci_dev_put(dev);
return val;
+ }
+ if (!dn->eeh_config_addr) {
+ pci_dev_put(dev);
+ return val;
+ }
- /* Now test for an EEH failure. This is VERY expensive.
+ /*
+ * Now test for an EEH failure. This is VERY expensive.
* Note that the eeh_config_addr may be a parent device
* in the case of a device behind a bridge, or it may be
* function zero of a multi-function device.
* In any case they must share a common PHB.
*/
- if (dn->eeh_config_addr) {
- ret = rtas_call(ibm_read_slot_reset_state, 3, 3, rets,
- dn->eeh_config_addr, BUID_HI(dn->phb->buid),
- BUID_LO(dn->phb->buid));
- if (ret == 0 && rets[1] == 1 && rets[0] >= 2) {
- unsigned long slot_err_ret;
-
- memset(slot_err_buf, 0, RTAS_ERROR_LOG_MAX);
- slot_err_ret = rtas_call(rtas_token("ibm,slot-error-detail"),
- 8, 1, NULL, dn->eeh_config_addr,
- BUID_HI(dn->phb->buid),
- BUID_LO(dn->phb->buid), NULL, 0,
- __pa(slot_err_buf), RTAS_ERROR_LOG_MAX,
- 2 /* Permanent Error */);
-
- if (slot_err_ret == 0)
- log_error(slot_err_buf, ERR_TYPE_RTAS_LOG, 1 /* Fatal */);
-
- /*
- * XXX We should create a separate sysctl for this.
- *
- * Since the panic_on_oops sysctl is used to halt
- * the system in light of potential corruption, we
- * can use it here.
- */
- if (panic_on_oops)
- panic("EEH: MMIO failure (%ld) on device:\n%s\n",
- rets[0], pci_name(dev));
- else
- printk("EEH: MMIO failure (%ld) on device:\n%s\n",
- rets[0], pci_name(dev));
+ ret = rtas_call(ibm_read_slot_reset_state, 3, 3, rets,
+ dn->eeh_config_addr, BUID_HI(dn->phb->buid),
+ BUID_LO(dn->phb->buid));
+
+ if (ret == 0 && rets[1] == 1 && rets[0] >= 2) {
+ unsigned long slot_err_ret;
+
+ spin_lock_irqsave(&lock, flags);
+ memset(slot_err_buf, 0, RTAS_ERROR_LOG_MAX);
+ slot_err_ret = rtas_call(rtas_token("ibm,slot-error-detail"),
+ 8, 1, NULL, dn->eeh_config_addr,
+ BUID_HI(dn->phb->buid),
+ BUID_LO(dn->phb->buid), NULL, 0,
+ __pa(slot_err_buf),
+ RTAS_ERROR_LOG_MAX,
+ 2 /* Permanent Error */);
+
+ if (slot_err_ret == 0)
+ log_error(slot_err_buf, ERR_TYPE_RTAS_LOG,
+ 1 /* Fatal */);
+
+ spin_unlock_irqrestore(&lock, flags);
+
+ /*
+ * XXX We should create a separate sysctl for this.
+ *
+ * Since the panic_on_oops sysctl is used to halt
+ * the system in light of potential corruption, we
+ * can use it here.
+ */
+ if (panic_on_oops) {
+ panic("EEH: MMIO failure (%ld) on device:%s %s\n",
+ rets[0], pci_name(dev), pci_pretty_name(dev));
+ } else {
+ __get_cpu_var(ignored_failures)++;
+ printk(KERN_INFO "EEH: MMIO failure (%ld) on device:%s %s\n",
+ rets[0], pci_name(dev), pci_pretty_name(dev));
}
+ } else {
+ __get_cpu_var(false_positives)++;
}
- eeh_false_positives++;
- return val; /* good case */
+ pci_dev_put(dev);
+ return val;
}
+EXPORT_SYMBOL(eeh_check_failure);
struct eeh_early_enable_info {
unsigned int buid_hi;
unsigned int buid_lo;
- int adapters_enabled;
};
/* Enable eeh for the given device node. */
@@ -165,7 +466,7 @@ static void *early_enable_eeh(struct dev
long ret;
char *status = get_property(dn, "status", 0);
u32 *class_code = (u32 *)get_property(dn, "class-code", 0);
- u32 *vendor_id =(u32 *) get_property(dn, "vendor-id", 0);
+ u32 *vendor_id = (u32 *)get_property(dn, "vendor-id", 0);
u32 *device_id = (u32 *)get_property(dn, "device-id", 0);
u32 *regs;
int enable;
@@ -183,7 +484,8 @@ static void *early_enable_eeh(struct dev
*device_id == 0x0188 || *device_id == 0x0302))
return NULL;
- /* Now decide if we are going to "Disable" EEH checking
+ /*
+ * Now decide if we are going to "Disable" EEH checking
* for this device. We still run with the EEH hardware active,
* but we won't be checking for ff's. This means a driver
* could return bad data (very bad!), an interrupt handler could
@@ -194,15 +496,19 @@ static void *early_enable_eeh(struct dev
if ((*class_code >> 16) == PCI_BASE_CLASS_DISPLAY)
enable = 0;
- if (!eeh_check_opts_config(dn, *class_code, *vendor_id, *device_id, enable)) {
+ if (!eeh_check_opts_config(dn, *class_code, *vendor_id, *device_id,
+ enable)) {
if (enable) {
- printk(KERN_INFO "EEH: %s user requested to run without EEH.\n", dn->full_name);
+ printk(KERN_WARNING "EEH: %s user requested to run "
+ "without EEH.\n", dn->full_name);
enable = 0;
}
}
- if (!enable)
+ if (!enable) {
dn->eeh_mode = EEH_MODE_NOCHECK;
+ return NULL;
+ }
/* This device may already have an EEH parent. */
if (dn->parent && (dn->parent->eeh_mode & EEH_MODE_SUPPORTED)) {
@@ -212,7 +518,7 @@ static void *early_enable_eeh(struct dev
return NULL;
}
- /* Ok..see if this device supports EEH. */
+ /* Ok... see if this device supports EEH. */
regs = (u32 *)get_property(dn, "reg", 0);
if (regs) {
/* First register entry is addr (00BBSS00) */
@@ -221,16 +527,27 @@ static void *early_enable_eeh(struct dev
regs[0], info->buid_hi, info->buid_lo,
EEH_ENABLE);
if (ret == 0) {
- info->adapters_enabled++;
+ eeh_subsystem_enabled = 1;
dn->eeh_mode |= EEH_MODE_SUPPORTED;
dn->eeh_config_addr = regs[0];
+#ifdef DEBUG
+ printk(KERN_DEBUG "EEH: %s: eeh enabled\n",
+ dn->full_name);
+#endif
+ } else {
+ printk(KERN_WARNING "EEH: %s: rtas_call failed.\n",
+ dn->full_name);
}
+ } else {
+ printk(KERN_WARNING "EEH: %s: unable to get reg property.\n",
+ dn->full_name);
}
+
return NULL;
}
/*
- * Initialize eeh by trying to enable it for all of the adapters in the system.
+ * Initialize EEH by trying to enable it for all of the adapters in the system.
* As a side effect we can determine here if eeh is supported at all.
* Note that we leave EEH on so failed config cycles won't cause a machine
* check. If a user turns off EEH for a particular adapter they are really
@@ -240,43 +557,35 @@ static void *early_enable_eeh(struct dev
* but for now disabling EEH for adapters is mostly to work around drivers that
* directly access mmio space (without using the macros).
*
- * The eeh-force-off/on option does literally what it says, so if Linux must
+ * The eeh-force-off option does literally what it says, so if Linux must
* avoid enabling EEH this must be done.
*/
-void eeh_init(void)
+void __init eeh_init(void)
{
struct device_node *phb;
struct eeh_early_enable_info info;
char *eeh_force_off = strstr(saved_command_line, "eeh-force-off");
- char *eeh_force_on = strstr(saved_command_line, "eeh-force-on");
ibm_set_eeh_option = rtas_token("ibm,set-eeh-option");
ibm_set_slot_reset = rtas_token("ibm,set-slot-reset");
ibm_read_slot_reset_state = rtas_token("ibm,read-slot-reset-state");
- /* Allow user to force eeh mode on or off -- even if the hardware
- * doesn't exist. This allows driver writers to at least test use
- * of I/O macros even if we can't actually test for EEH failure.
- */
- if (eeh_force_on > eeh_force_off)
- eeh_implemented = 1;
- else if (ibm_set_eeh_option == RTAS_UNKNOWN_SERVICE)
+ if (ibm_set_eeh_option == RTAS_UNKNOWN_SERVICE)
return;
- if (eeh_force_off > eeh_force_on) {
- /* User is forcing EEH off. Be noisy if it is implemented. */
- if (eeh_implemented)
- printk(KERN_WARNING "EEH: WARNING: PCI Enhanced I/O Error Handling is user disabled\n");
- eeh_implemented = 0;
+ if (eeh_force_off) {
+ printk(KERN_WARNING "EEH: WARNING: PCI Enhanced I/O Error "
+ "Handling is user disabled\n");
return;
}
-
/* Enable EEH for all adapters. Note that eeh requires buid's */
- info.adapters_enabled = 0;
- for (phb = of_find_node_by_name(NULL, "pci"); phb; phb = of_find_node_by_name(phb, "pci")) {
+ for (phb = of_find_node_by_name(NULL, "pci"); phb;
+ phb = of_find_node_by_name(phb, "pci")) {
int len;
- int *buid_vals = (int *) get_property(phb, "ibm,fw-phb-id", &len);
+ int *buid_vals;
+
+ buid_vals = (int *)get_property(phb, "ibm,fw-phb-id", &len);
if (!buid_vals)
continue;
if (len == sizeof(int)) {
@@ -286,35 +595,82 @@ void eeh_init(void)
info.buid_hi = buid_vals[0];
info.buid_lo = buid_vals[1];
} else {
- printk("EEH: odd ibm,fw-phb-id len returned: %d\n", len);
+ printk(KERN_INFO "EEH: odd ibm,fw-phb-id len returned: %d\n", len);
continue;
}
traverse_pci_devices(phb, early_enable_eeh, NULL, &info);
}
- if (info.adapters_enabled) {
+
+ if (eeh_subsystem_enabled)
printk(KERN_INFO "EEH: PCI Enhanced I/O Error Handling Enabled\n");
- eeh_implemented = 1;
- }
}
-
-int eeh_set_option(struct pci_dev *dev, int option)
+/**
+ * eeh_add_device - perform EEH initialization for the indicated pci device
+ * @dev: pci device for which to set up EEH
+ *
+ * This routine can be used to perform EEH initialization for PCI
+ * devices that were added after system boot (e.g. hotplug, dlpar).
+ * Whether this actually enables EEH or not for this device depends
+ * on the type of the device, on earlier boot command-line
+ * arguments & etc.
+ */
+void eeh_add_device(struct pci_dev *dev)
{
- struct device_node *dn = pci_device_to_OF_node(dev);
- struct pci_controller *phb = PCI_GET_PHB_PTR(dev);
+ struct device_node *dn;
+ struct pci_controller *phb;
+ struct eeh_early_enable_info info;
- if (dn == NULL || phb == NULL || phb->buid == 0 || !eeh_implemented)
- return -2;
+ if (!dev || !eeh_subsystem_enabled)
+ return;
- return rtas_call(ibm_set_eeh_option, 4, 1, NULL,
- CONFIG_ADDR(dn->busno, dn->devfn),
- BUID_HI(phb->buid), BUID_LO(phb->buid), option);
+#ifdef DEBUG
+ printk(KERN_DEBUG "EEH: adding device %s %s\n", pci_name(dev),
+ pci_pretty_name(dev));
+#endif
+ dn = pci_device_to_OF_node(dev);
+ if (NULL == dn)
+ return;
+
+ phb = PCI_GET_PHB_PTR(dev);
+ if (NULL == phb || 0 == phb->buid) {
+ printk(KERN_WARNING "EEH: Expected buid but found none\n");
+ return;
+ }
+
+ info.buid_hi = BUID_HI(phb->buid);
+ info.buid_lo = BUID_LO(phb->buid);
+
+ early_enable_eeh(dn, &info);
+ pci_addr_cache_insert_device (dev);
}
+EXPORT_SYMBOL(eeh_add_device);
+
+/**
+ * eeh_remove_device - undo EEH setup for the indicated pci device
+ * @dev: pci device to be removed
+ *
+ * This routine should be when a device is removed from a running
+ * system (e.g. by hotplug or dlpar).
+ */
+void eeh_remove_device(struct pci_dev *dev)
+{
+ if (!dev || !eeh_subsystem_enabled)
+ return;
+ /* Unregister the device with the EEH/PCI address search system */
+#ifdef DEBUG
+ printk(KERN_DEBUG "EEH: remove device %s %s\n", pci_name(dev),
+ pci_pretty_name(dev));
+#endif
+ pci_addr_cache_remove_device(dev);
+}
+EXPORT_SYMBOL(eeh_remove_device);
-/* If EEH is implemented, find the PCI device using given phys addr
+/*
+ * If EEH is implemented, find the PCI device using given phys addr
* and check to see if eeh failure checking is disabled.
- * Remap the addr (trivially) to the EEH region if not.
+ * Remap the addr (trivially) to the EEH region if EEH checking enabled.
* For addresses not known to PCI the vaddr is simply returned unchanged.
*/
void *eeh_ioremap(unsigned long addr, void *vaddr)
@@ -322,43 +678,78 @@ void *eeh_ioremap(unsigned long addr, vo
struct pci_dev *dev;
struct device_node *dn;
- if (!eeh_implemented)
+ if (!eeh_subsystem_enabled)
return vaddr;
- dev = pci_find_dev_by_addr(addr);
+
+ dev = pci_get_device_by_addr(addr);
if (!dev)
return vaddr;
+
dn = pci_device_to_OF_node(dev);
- if (!dn)
+ if (!dn) {
+ pci_dev_put(dev);
return vaddr;
- if (dn->eeh_mode & EEH_MODE_NOCHECK)
+ }
+
+ if (dn->eeh_mode & EEH_MODE_NOCHECK) {
+ pci_dev_put(dev);
return vaddr;
+ }
+ pci_dev_put(dev);
return (void *)IO_ADDR_TO_TOKEN(vaddr);
}
-static int eeh_proc_falsepositive_read(char *page, char **start, off_t off,
- int count, int *eof, void *data)
+static int proc_eeh_show(struct seq_file *m, void *v)
{
- int len;
- len = sprintf(page, "eeh_false_positives=%ld\n"
- "eeh_total_mmio_ffs=%ld\n",
- eeh_false_positives, eeh_total_mmio_ffs);
- return len;
+ unsigned int cpu;
+ unsigned long ffs = 0, positives = 0, failures = 0;
+
+ for_each_cpu(cpu) {
+ ffs += per_cpu(total_mmio_ffs, cpu);
+ positives += per_cpu(false_positives, cpu);
+ failures += per_cpu(ignored_failures, cpu);
+ }
+
+ if (0 == eeh_subsystem_enabled) {
+ seq_printf(m, "EEH Subsystem is globally disabled\n");
+ seq_printf(m, "eeh_total_mmio_ffs=%ld\n", ffs);
+ } else {
+ seq_printf(m, "EEH Subsystem is enabled\n");
+ seq_printf(m, "eeh_total_mmio_ffs=%ld\n"
+ "eeh_false_positives=%ld\n"
+ "eeh_ignored_failures=%ld\n",
+ ffs, positives, failures);
+ }
+
+ return 0;
}
-/* Implementation of /proc/ppc64/eeh
- * For now it is one file showing false positives.
- */
+static int proc_eeh_open(struct inode *inode, struct file *file)
+{
+ return single_open(file, proc_eeh_show, NULL);
+}
+
+static struct file_operations proc_eeh_operations = {
+ .open = proc_eeh_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
+
static int __init eeh_init_proc(void)
{
- struct proc_dir_entry *ent = create_proc_entry("ppc64/eeh", S_IRUGO, 0);
- if (ent) {
- ent->nlink = 1;
- ent->data = NULL;
- ent->read_proc = (void *)eeh_proc_falsepositive_read;
+ struct proc_dir_entry *e;
+
+ if (systemcfg->platform & PLATFORM_PSERIES) {
+ e = create_proc_entry("ppc64/eeh", 0, NULL);
+ if (e)
+ e->proc_fops = &proc_eeh_operations;
}
- return 0;
+
+ return 0;
}
+__initcall(eeh_init_proc);
/*
* Test if "dev" should be configured on or off.
@@ -386,10 +777,12 @@ static int eeh_check_opts_config(struct
strs[nstrs++] = classname;
strs[nstrs++] = ""; /* yes, this matches the empty string */
- /* Now see if any string matches the eeh_opts list.
+ /*
+ * Now see if any string matches the eeh_opts list.
* The eeh_opts list entries start with + or -.
*/
- for (s = eeh_opts; s && (s < (eeh_opts + eeh_opts_last)); s += strlen(s)+1) {
+ for (s = eeh_opts; s && (s < (eeh_opts + eeh_opts_last));
+ s += strlen(s)+1) {
for (i = 0; i < nstrs; i++) {
if (strcasecmp(strs[i], s+1) == 0) {
ret = (strs[i][0] == '+') ? 1 : 0;
@@ -399,7 +792,8 @@ static int eeh_check_opts_config(struct
return ret;
}
-/* Handle kernel eeh-on & eeh-off cmd line options for eeh.
+/*
+ * Handle kernel eeh-on & eeh-off cmd line options for eeh.
*
* We support:
* eeh-off=loc1,loc2,loc3...
@@ -420,7 +814,8 @@ static int eeh_check_opts_config(struct
* so eeh-off means eeh by default is off.
*/
-/* This is implemented as a null separated list of strings.
+/*
+ * This is implemented as a null separated list of strings.
* Each string looks like this: "+X" or "-X"
* where X is a loc code, vendor:device, class (as shown above)
* or empty which is used to indicate all.
@@ -428,10 +823,10 @@ static int eeh_check_opts_config(struct
* We interpret this option string list so that it will literally
* behave left-to-right even if some combinations don't make sense.
*/
-
static int __init eeh_parm(char *str, int state)
{
char *s, *cur, *curend;
+
if (!eeh_opts) {
eeh_opts = alloc_bootmem(EEH_MAX_OPTS);
eeh_opts[eeh_opts_last++] = '+'; /* default */
@@ -446,15 +841,17 @@ static int __init eeh_parm(char *str, in
str++;
for (s = str; s && *s != '\0'; s = curend) {
cur = s;
+ /* ignore empties. Don't treat as "all-on" or "all-off" */
while (*cur == ',')
- cur++; /* ignore empties. Don't treat as "all-on" or "all-off" */
+ cur++;
curend = strchr(cur, ',');
if (!curend)
curend = cur + strlen(cur);
if (*cur) {
int curlen = curend-cur;
if (eeh_opts_last + curlen > EEH_MAX_OPTS-2) {
- printk(KERN_INFO "EEH: sorry...too many eeh cmd line options\n");
+ printk(KERN_WARNING "EEH: sorry...too many "
+ "eeh cmd line options\n");
return 1;
}
eeh_opts[eeh_opts_last++] = state ? '+' : '-';
@@ -463,6 +860,7 @@ static int __init eeh_parm(char *str, in
eeh_opts[eeh_opts_last++] = '\0';
}
}
+
return 1;
}
@@ -476,6 +874,5 @@ static int __init eehon_parm(char *str)
return eeh_parm(str, 1);
}
-__initcall(eeh_init_proc);
__setup("eeh-off", eehoff_parm);
__setup("eeh-on", eehon_parm);
--- diff/arch/ppc64/kernel/entry.S 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/entry.S 2004-03-16 11:27:47.920902576 +0000
@@ -95,7 +95,7 @@ _GLOBAL(DoSyscall)
#endif /* SHOW_SYSCALLS */
clrrdi r10,r1,THREAD_SHIFT
ld r10,TI_FLAGS(r10)
- andi. r11,r10,_TIF_SYSCALL_TRACE
+ andi. r11,r10,_TIF_SYSCALL_T_OR_A
bne- 50f
cmpli 0,r0,NR_syscalls
bge- 66f
@@ -151,7 +151,8 @@ _GLOBAL(ret_from_syscall_1)
b 22b
/* Traced system call support */
-50: bl .do_syscall_trace
+50: addi r3,r1,STACK_FRAME_OVERHEAD
+ bl .do_syscall_trace_enter
ld r0,GPR0(r1) /* Restore original registers */
ld r3,GPR3(r1)
ld r4,GPR4(r1)
@@ -201,7 +202,7 @@ _GLOBAL(ret_from_syscall_2)
oris r10,r10,0x1000
std r10,_CCR(r1)
60: std r3,GPR3(r1) /* Update return value */
- bl .do_syscall_trace
+ bl .do_syscall_trace_leave
b .ret_from_except
66: li r3,ENOSYS
b 57b
@@ -234,14 +235,14 @@ _GLOBAL(ppc64_rt_sigreturn)
80: clrrdi r4,r1,THREAD_SHIFT
ld r4,TI_FLAGS(r4)
- andi. r4,r4,_TIF_SYSCALL_TRACE
+ andi. r4,r4,_TIF_SYSCALL_T_OR_A
bne- 81f
cmpi 0,r3,0
bge .ret_from_except
b .ret_from_syscall_1
81: cmpi 0,r3,0
blt .ret_from_syscall_2
- bl .do_syscall_trace
+ bl .do_syscall_trace_leave
b .ret_from_except
/*
@@ -327,9 +328,9 @@ _GLOBAL(ret_from_fork)
bl .schedule_tail
clrrdi r4,r1,THREAD_SHIFT
ld r4,TI_FLAGS(r4)
- andi. r4,r4,_TIF_SYSCALL_TRACE
+ andi. r4,r4,_TIF_SYSCALL_T_OR_A
beq+ .ret_from_except
- bl .do_syscall_trace
+ bl .do_syscall_trace_leave
b .ret_from_except
_GLOBAL(ret_from_except)
--- diff/arch/ppc64/kernel/iSeries_VpdInfo.c 2004-02-09 10:36:08.000000000 +0000
+++ source/arch/ppc64/kernel/iSeries_VpdInfo.c 2004-03-16 11:27:47.925901816 +0000
@@ -27,6 +27,7 @@
/************************************************************************/
#include <linux/config.h>
#include <linux/init.h>
+#include <linux/module.h>
#include <linux/pci.h>
#include <asm/types.h>
#include <asm/resource.h>
@@ -105,6 +106,7 @@ LocationData* iSeries_GetLocationData(st
strcpy(&LocationPtr->CardLocation[0], &DevNode->CardLocation[0]);
return LocationPtr;
}
+EXPORT_SYMBOL(iSeries_GetLocationData);
/*
* Formats the device information.
--- diff/arch/ppc64/kernel/iSeries_pci.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/iSeries_pci.c 2004-03-16 11:27:47.922902272 +0000
@@ -26,6 +26,7 @@
#include <linux/list.h>
#include <linux/string.h>
#include <linux/init.h>
+#include <linux/module.h>
#include <linux/ide.h>
#include <linux/pci.h>
@@ -472,6 +473,7 @@ void *iSeries_memset_io(void *dest, char
}
return dest;
}
+EXPORT_SYMBOL(iSeries_memset_io);
void *iSeries_memcpy_toio(void *dest, void *source, size_t count)
{
@@ -485,6 +487,7 @@ void *iSeries_memcpy_toio(void *dest, vo
}
return dest;
}
+EXPORT_SYMBOL(iSeries_memcpy_toio);
void *iSeries_memcpy_fromio(void *dest, void *source, size_t count)
{
@@ -498,6 +501,7 @@ void *iSeries_memcpy_fromio(void *dest,
}
return dest;
}
+EXPORT_SYMBOL(iSeries_memcpy_fromio);
/*
* Look down the chain to find the matching Device Device
@@ -708,6 +712,7 @@ u8 iSeries_Read_Byte(void *IoAddress)
return (u8)ret.value;
}
+EXPORT_SYMBOL(iSeries_Read_Byte);
u16 iSeries_Read_Word(void *IoAddress)
{
@@ -737,6 +742,7 @@ u16 iSeries_Read_Word(void *IoAddress)
return swab16((u16)ret.value);
}
+EXPORT_SYMBOL(iSeries_Read_Word);
u32 iSeries_Read_Long(void *IoAddress)
{
@@ -766,6 +772,7 @@ u32 iSeries_Read_Long(void *IoAddress)
return swab32((u32)ret.value);
}
+EXPORT_SYMBOL(iSeries_Read_Long);
/*
* Write MM I/O Instructions for the iSeries
@@ -799,6 +806,7 @@ void iSeries_Write_Byte(u8 data, void *I
rc = HvCall4(HvCallPciBarStore8, dsa, BarOffset, data, 0);
} while (CheckReturnCode("WWB", DevNode, rc) != 0);
}
+EXPORT_SYMBOL(iSeries_Write_Byte);
void iSeries_Write_Word(u16 data, void *IoAddress)
{
@@ -825,6 +833,7 @@ void iSeries_Write_Word(u16 data, void *
rc = HvCall4(HvCallPciBarStore16, dsa, BarOffset, swab16(data), 0);
} while (CheckReturnCode("WWW", DevNode, rc) != 0);
}
+EXPORT_SYMBOL(iSeries_Write_Word);
void iSeries_Write_Long(u32 data, void *IoAddress)
{
@@ -851,6 +860,7 @@ void iSeries_Write_Long(u32 data, void *
rc = HvCall4(HvCallPciBarStore32, dsa, BarOffset, swab32(data), 0);
} while (CheckReturnCode("WWL", DevNode, rc) != 0);
}
+EXPORT_SYMBOL(iSeries_Write_Long);
void pcibios_name_device(struct pci_dev *dev)
{
--- diff/arch/ppc64/kernel/iSeries_pci_reset.c 2004-02-09 10:36:08.000000000 +0000
+++ source/arch/ppc64/kernel/iSeries_pci_reset.c 2004-03-16 11:27:47.922902272 +0000
@@ -29,6 +29,7 @@
/************************************************************************/
#include <linux/kernel.h>
#include <linux/init.h>
+#include <linux/module.h>
#include <linux/pci.h>
#include <linux/irq.h>
@@ -100,3 +101,4 @@ int iSeries_Device_ToggleReset(struct pc
}
return DeviceNode->ReturnCode;
}
+EXPORT_SYMBOL(iSeries_Device_ToggleReset);
--- diff/arch/ppc64/kernel/iSeries_proc.c 2004-02-09 10:36:08.000000000 +0000
+++ source/arch/ppc64/kernel/iSeries_proc.c 2004-03-16 11:27:47.923902120 +0000
@@ -1,131 +1,162 @@
/*
- * iSeries_proc.c
- * Copyright (C) 2001 Kyle A. Lucke IBM Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- */
-#include <linux/proc_fs.h>
-#include <linux/spinlock.h>
+ * iSeries_proc.c
+ * Copyright (C) 2001 Kyle A. Lucke IBM Corporation
+ * Copyright (C) 2001 Mike Corrigan & Dave Engebretsen IBM Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
#include <linux/init.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <asm/paca.h>
+#include <asm/processor.h>
+#include <asm/time.h>
+#include <asm/naca.h>
+#include <asm/iSeries/ItLpPaca.h>
+#include <asm/iSeries/ItLpQueue.h>
+#include <asm/iSeries/HvCallXm.h>
+#include <asm/iSeries/IoHriMainStore.h>
+#include <asm/iSeries/LparData.h>
#include <asm/iSeries/iSeries_proc.h>
-static struct proc_dir_entry *iSeries_proc_root;
-static int iSeries_proc_initializationDone;
-static spinlock_t iSeries_proc_lock;
-
-struct iSeries_proc_registration {
- struct iSeries_proc_registration *next;
- iSeriesProcFunction functionMember;
+static int __init iseries_proc_create(void)
+{
+ struct proc_dir_entry *e = proc_mkdir("iSeries", 0);
+ if (!e)
+ return 1;
+
+ return 0;
+}
+core_initcall(iseries_proc_create);
+
+static char *event_types[9] = {
+ "Hypervisor\t\t",
+ "Machine Facilities\t",
+ "Session Manager\t",
+ "SPD I/O\t\t",
+ "Virtual Bus\t\t",
+ "PCI I/O\t\t",
+ "RIO I/O\t\t",
+ "Virtual Lan\t\t",
+ "Virtual I/O\t\t"
};
-struct iSeries_proc_registration preallocated[16];
+static int proc_lpevents_show(struct seq_file *m, void *v)
+{
+ unsigned int i;
+
+ seq_printf(m, "LpEventQueue 0\n");
+ seq_printf(m, " events processed:\t%lu\n",
+ (unsigned long)xItLpQueue.xLpIntCount);
-#define MYQUEUETYPE(T) struct MYQueue##T
-#define MYQUEUE(T) \
-MYQUEUETYPE(T) \
-{ \
- struct T *head; \
- struct T *tail; \
-}
-#define MYQUEUECTOR(q) do { (q)->head = NULL; (q)->tail = NULL; } while(0)
-#define MYQUEUEENQ(q, p) \
-do { \
- (p)->next = NULL; \
- if ((q)->head != NULL) { \
- (q)->head->next = (p); \
- (q)->head = (p); \
- } else { \
- (q)->tail = (q)->head = (p); \
- } \
-} while(0)
-
-#define MYQUEUEDEQ(q,p) \
-do { \
- (p) = (q)->tail; \
- if ((p) != NULL) { \
- (q)->tail = (p)->next; \
- (p)->next = NULL; \
- } \
- if ((q)->tail == NULL) \
- (q)->head = NULL; \
-} while(0)
-
-MYQUEUE(iSeries_proc_registration);
-typedef MYQUEUETYPE(iSeries_proc_registration) aQueue;
-
-static aQueue iSeries_free;
-static aQueue iSeries_queued;
-
-void iSeries_proc_early_init(void)
-{
- int i = 0;
- unsigned long flags;
-
- iSeries_proc_initializationDone = 0;
- spin_lock_init(&iSeries_proc_lock);
- MYQUEUECTOR(&iSeries_free);
- MYQUEUECTOR(&iSeries_queued);
-
- spin_lock_irqsave(&iSeries_proc_lock, flags);
- for (i = 0; i < 16; ++i)
- MYQUEUEENQ(&iSeries_free, preallocated + i);
- spin_unlock_irqrestore(&iSeries_proc_lock, flags);
-}
-
-static int iSeries_proc_create(void)
-{
- unsigned long flags;
- struct iSeries_proc_registration *reg;
-
- printk("iSeries_proc: Creating /proc/iSeries\n");
-
- spin_lock_irqsave(&iSeries_proc_lock, flags);
- iSeries_proc_root = proc_mkdir("iSeries", 0);
- if (!iSeries_proc_root)
- goto out;
-
- MYQUEUEDEQ(&iSeries_queued, reg);
- while (reg != NULL) {
- (*(reg->functionMember))(iSeries_proc_root);
- MYQUEUEDEQ(&iSeries_queued, reg);
- }
+ for (i = 0; i < 9; ++i)
+ seq_printf(m, " %s %10lu\n", event_types[i],
+ (unsigned long)xItLpQueue.xLpIntCountByType[i]);
+
+ seq_printf(m, "\n events processed by processor:\n");
+
+ for_each_online_cpu(i)
+ seq_printf(m, " CPU%02d %10u\n", i, paca[i].lpEvent_count);
- iSeries_proc_initializationDone = 1;
-out:
- spin_unlock_irqrestore(&iSeries_proc_lock, flags);
return 0;
}
-arch_initcall(iSeries_proc_create);
+static int proc_lpevents_open(struct inode *inode, struct file *file)
+{
+ return single_open(file, proc_lpevents_show, NULL);
+}
+
+static struct file_operations proc_lpevents_operations = {
+ .open = proc_lpevents_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
-void iSeries_proc_callback(iSeriesProcFunction initFunction)
+static unsigned long startTitan = 0;
+static unsigned long startTb = 0;
+
+static int proc_titantod_show(struct seq_file *m, void *v)
{
- unsigned long flags;
+ unsigned long tb0, titan_tod;
+
+ tb0 = get_tb();
+ titan_tod = HvCallXm_loadTod();
- spin_lock_irqsave(&iSeries_proc_lock, flags);
- if (iSeries_proc_initializationDone)
- (*initFunction)(iSeries_proc_root);
- else {
- struct iSeries_proc_registration *reg = NULL;
-
- MYQUEUEDEQ(&iSeries_free, reg);
- if (reg != NULL) {
- reg->functionMember = initFunction;
- MYQUEUEENQ(&iSeries_queued, reg);
- } else
- printk("Couldn't get a queue entry\n");
+ seq_printf(m, "Titan\n" );
+ seq_printf(m, " time base = %016lx\n", tb0);
+ seq_printf(m, " titan tod = %016lx\n", titan_tod);
+ seq_printf(m, " xProcFreq = %016x\n",
+ xIoHriProcessorVpd[0].xProcFreq);
+ seq_printf(m, " xTimeBaseFreq = %016x\n",
+ xIoHriProcessorVpd[0].xTimeBaseFreq);
+ seq_printf(m, " tb_ticks_per_jiffy = %lu\n", tb_ticks_per_jiffy);
+ seq_printf(m, " tb_ticks_per_usec = %lu\n", tb_ticks_per_usec);
+
+ if (!startTitan) {
+ startTitan = titan_tod;
+ startTb = tb0;
+ } else {
+ unsigned long titan_usec = (titan_tod - startTitan) >> 12;
+ unsigned long tb_ticks = (tb0 - startTb);
+ unsigned long titan_jiffies = titan_usec / (1000000/HZ);
+ unsigned long titan_jiff_usec = titan_jiffies * (1000000/HZ);
+ unsigned long titan_jiff_rem_usec = titan_usec - titan_jiff_usec;
+ unsigned long tb_jiffies = tb_ticks / tb_ticks_per_jiffy;
+ unsigned long tb_jiff_ticks = tb_jiffies * tb_ticks_per_jiffy;
+ unsigned long tb_jiff_rem_ticks = tb_ticks - tb_jiff_ticks;
+ unsigned long tb_jiff_rem_usec = tb_jiff_rem_ticks / tb_ticks_per_usec;
+ unsigned long new_tb_ticks_per_jiffy = (tb_ticks * (1000000/HZ))/titan_usec;
+
+ seq_printf(m, " titan elapsed = %lu uSec\n", titan_usec);
+ seq_printf(m, " tb elapsed = %lu ticks\n", tb_ticks);
+ seq_printf(m, " titan jiffies = %lu.%04lu \n", titan_jiffies,
+ titan_jiff_rem_usec);
+ seq_printf(m, " tb jiffies = %lu.%04lu\n", tb_jiffies,
+ tb_jiff_rem_usec);
+ seq_printf(m, " new tb_ticks_per_jiffy = %lu\n",
+ new_tb_ticks_per_jiffy);
}
- spin_unlock_irqrestore(&iSeries_proc_lock, flags);
+
+ return 0;
+}
+
+static int proc_titantod_open(struct inode *inode, struct file *file)
+{
+ return single_open(file, proc_titantod_show, NULL);
+}
+
+static struct file_operations proc_titantod_operations = {
+ .open = proc_titantod_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
+
+static int __init iseries_proc_init(void)
+{
+ struct proc_dir_entry *e;
+
+ e = create_proc_entry("iSeries/lpevents", S_IFREG|S_IRUGO, NULL);
+ if (e)
+ e->proc_fops = &proc_lpevents_operations;
+
+ e = create_proc_entry("iSeries/titanTod", S_IFREG|S_IRUGO, NULL);
+ if (e)
+ e->proc_fops = &proc_titantod_operations;
+
+ return 0;
}
+__initcall(iseries_proc_init);
--- diff/arch/ppc64/kernel/iSeries_setup.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/iSeries_setup.c 2004-03-16 11:27:47.924901968 +0000
@@ -51,7 +51,6 @@
#include <asm/iSeries/ItLpQueue.h>
#include <asm/iSeries/IoHriMainStore.h>
#include <asm/iSeries/iSeries_proc.h>
-#include <asm/proc_pmc.h>
#include <asm/iSeries/mf.h>
/* Function Prototypes */
@@ -393,12 +392,9 @@ void __init iSeries_init(unsigned long r
iSeries_setup_dprofile();
- iSeries_proc_early_init();
mf_init();
mf_initialized = 1;
mb();
-
- iSeries_proc_callback(&pmc_proc_init);
}
/*
--- diff/arch/ppc64/kernel/irq.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/irq.c 2004-03-16 11:27:47.921902424 +0000
@@ -66,7 +66,7 @@ irq_desc_t irq_desc[NR_IRQS] __cacheline
.lock = SPIN_LOCK_UNLOCKED
}
};
-
+
int ppc_spurious_interrupts = 0;
unsigned long lpEvent_count = 0;
@@ -183,6 +183,7 @@ do_free_irq(int irq, void* dev_id)
return -ENOENT;
}
+
int request_irq(unsigned int irq,
irqreturn_t (*handler)(int, void *, struct pt_regs *),
unsigned long irqflags, const char * devname, void *dev_id)
@@ -195,25 +196,25 @@ int request_irq(unsigned int irq,
if (!handler)
/* We could implement really free_irq() instead of that... */
return do_free_irq(irq, dev_id);
-
+
action = (struct irqaction *)
kmalloc(sizeof(struct irqaction), GFP_KERNEL);
if (!action) {
printk(KERN_ERR "kmalloc() failed for irq %d !\n", irq);
return -ENOMEM;
}
-
+
action->handler = handler;
- action->flags = irqflags;
+ action->flags = irqflags;
action->mask = 0;
action->name = devname;
action->dev_id = dev_id;
action->next = NULL;
-
+
retval = setup_irq(irq, action);
if (retval)
kfree(action);
-
+
return 0;
}
@@ -342,13 +343,13 @@ int show_interrupts(struct seq_file *p,
action = irq_desc[i].action;
if (!action || !action->handler)
goto skip;
- seq_printf(p, "%3d: ", i);
+ seq_printf(p, "%3d: ", i);
#ifdef CONFIG_SMP
for (j = 0; j < NR_CPUS; j++) {
if (cpu_online(j))
seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
}
-#else
+#else
seq_printf(p, "%10u ", kstat_irqs(i));
#endif /* CONFIG_SMP */
if (irq_desc[i].handler)
@@ -373,8 +374,10 @@ static inline int handle_irq_event(int i
int status = 0;
int retval = 0;
+#ifndef CONFIG_PPC_ISERIES
if (!(action->flags & SA_INTERRUPT))
local_irq_enable();
+#endif
do {
status |= action->flags;
@@ -383,7 +386,9 @@ static inline int handle_irq_event(int i
} while (action);
if (status & SA_SAMPLE_RANDOM)
add_interrupt_randomness(irq);
+#ifndef CONFIG_PPC_ISERIES
local_irq_disable();
+#endif
return retval;
}
@@ -577,13 +582,13 @@ int do_IRQ(struct pt_regs *regs)
irq_enter();
#ifdef CONFIG_DEBUG_STACKOVERFLOW
- /* Debugging check for stack overflow: is there less than 8KB free? */
+ /* Debugging check for stack overflow: is there less than 4KB free? */
{
long sp;
sp = __get_SP() & (THREAD_SIZE-1);
- if (unlikely(sp < (sizeof(struct thread_info) + 8192))) {
+ if (unlikely(sp < (sizeof(struct thread_info) + 4096))) {
printk("do_IRQ: stack overflow: %ld\n",
sp - sizeof(struct thread_info));
dump_stack();
@@ -670,7 +675,7 @@ void __init init_IRQ(void)
return;
once++;
-
+
ppc_md.init_IRQ();
}
@@ -699,6 +704,7 @@ static int irq_affinity_write_proc (stru
{
int irq = (long)data, full_count = count, err;
cpumask_t new_value, tmp;
+ cpumask_t allcpus = CPU_MASK_ALL;
if (!irq_desc[irq].handler->set_affinity)
return -EIO;
@@ -708,6 +714,14 @@ static int irq_affinity_write_proc (stru
return err;
/*
+ * We check for CPU_MASK_ALL in xics to send irqs to all cpus.
+ * In some cases CPU_MASK_ALL is smaller than the cpumask (eg
+ * NR_CPUS == 32 and cpumask is a long), so we mask it here to
+ * be consistent.
+ */
+ cpus_and(new_value, new_value, allcpus);
+
+ /*
* Do not allow disabling IRQs completely - it's a too easy
* way to make the system unusable accidentally :-) At least
* one online CPU still has to be targeted.
@@ -900,4 +914,37 @@ int virt_irq_create_mapping(unsigned int
return NO_IRQ;
}
+/*
+ * In most cases will get a hit on the very first slot checked in the
+ * virt_irq_to_real_map. Only when there are a large number of
+ * IRQs will this be expensive.
+ */
+unsigned int real_irq_to_virt_slowpath(unsigned int real_irq)
+{
+ unsigned int virq;
+ unsigned int first_virq;
+
+ virq = real_irq;
+
+ if (virq > MAX_VIRT_IRQ)
+ virq = (virq % NR_VIRT_IRQS) + MIN_VIRT_IRQ;
+
+ first_virq = virq;
+
+ do {
+ if (virt_irq_to_real_map[virq] == real_irq)
+ return virq;
+
+ virq++;
+
+ if (virq >= MAX_VIRT_IRQ)
+ virq = 0;
+
+ } while (first_virq != virq);
+
+ return NO_IRQ;
+
+}
+
+
#endif
--- diff/arch/ppc64/kernel/mf.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/mf.c 2004-03-16 11:27:47.927901512 +0000
@@ -30,13 +30,13 @@
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/mm.h>
+#include <linux/module.h>
#include <asm/iSeries/HvLpConfig.h>
#include <linux/slab.h>
#include <linux/delay.h>
#include <asm/nvram.h>
#include <asm/time.h>
#include <asm/iSeries/ItSpCommArea.h>
-#include <asm/iSeries/iSeries_proc.h>
#include <asm/uaccess.h>
#include <linux/dma-mapping.h>
#include <linux/bcd.h>
@@ -560,6 +560,7 @@ void mf_allocateLpEvents(HvLpIndex targe
if ((rc != 0) && (hdlr != NULL))
(*hdlr)(userToken, rc);
}
+EXPORT_SYMBOL(mf_allocateLpEvents);
/*
* Global kernel interface to unseed and deallocate events already in
@@ -590,6 +591,7 @@ void mf_deallocateLpEvents(HvLpIndex tar
if ((rc != 0) && (hdlr != NULL))
(*hdlr)(userToken, rc);
}
+EXPORT_SYMBOL(mf_deallocateLpEvents);
/*
* Global kernel interface to tell the VSP object in the primary
@@ -680,8 +682,6 @@ void mf_init(void)
/* initialization complete */
printk(KERN_NOTICE "mf.c: iSeries Linux LPAR Machine Facilities initialized\n");
-
- iSeries_proc_callback(&mf_proc_init);
}
void mf_setSide(char side)
--- diff/arch/ppc64/kernel/mf_proc.c 2004-02-09 10:36:08.000000000 +0000
+++ source/arch/ppc64/kernel/mf_proc.c 2004-03-16 11:27:47.927901512 +0000
@@ -16,11 +16,10 @@
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
+#include <linux/init.h>
#include <asm/uaccess.h>
#include <asm/iSeries/mf.h>
-static struct proc_dir_entry *mf_proc_root = NULL;
-
static int proc_mf_dump_cmdline(char *page, char **start, off_t off,
int count, int *eof, void *data)
{
@@ -187,27 +186,28 @@ static int proc_mf_change_vmlinux(struct
return count;
}
-void mf_proc_init(struct proc_dir_entry *iSeries_proc)
+static int __init mf_proc_init(void)
{
+ struct proc_dir_entry *mf_proc_root;
struct proc_dir_entry *ent;
struct proc_dir_entry *mf;
char name[2];
int i;
- mf_proc_root = proc_mkdir("mf", iSeries_proc);
+ mf_proc_root = proc_mkdir("iSeries/mf", NULL);
if (!mf_proc_root)
- return;
+ return 1;
name[1] = '\0';
for (i = 0; i < 4; i++) {
name[0] = 'A' + i;
mf = proc_mkdir(name, mf_proc_root);
if (!mf)
- return;
+ return 1;
ent = create_proc_entry("cmdline", S_IFREG|S_IRUSR|S_IWUSR, mf);
if (!ent)
- return;
+ return 1;
ent->nlink = 1;
ent->data = (void *)(long)i;
ent->read_proc = proc_mf_dump_cmdline;
@@ -218,7 +218,7 @@ void mf_proc_init(struct proc_dir_entry
ent = create_proc_entry("vmlinux", S_IFREG|S_IWUSR, mf);
if (!ent)
- return;
+ return 1;
ent->nlink = 1;
ent->data = (void *)(long)i;
#if 0
@@ -239,7 +239,7 @@ void mf_proc_init(struct proc_dir_entry
ent = create_proc_entry("side", S_IFREG|S_IRUSR|S_IWUSR, mf_proc_root);
if (!ent)
- return;
+ return 1;
ent->nlink = 1;
ent->data = (void *)0;
ent->read_proc = proc_mf_dump_side;
@@ -247,9 +247,13 @@ void mf_proc_init(struct proc_dir_entry
ent = create_proc_entry("src", S_IFREG|S_IRUSR|S_IWUSR, mf_proc_root);
if (!ent)
- return;
+ return 1;
ent->nlink = 1;
ent->data = (void *)0;
ent->read_proc = proc_mf_dump_src;
ent->write_proc = proc_mf_change_src;
+
+ return 0;
}
+
+__initcall(mf_proc_init);
--- diff/arch/ppc64/kernel/misc.S 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/ppc64/kernel/misc.S 2004-03-16 11:27:47.928901360 +0000
@@ -624,7 +624,7 @@ _GLOBAL(sys_call_table32)
.llong .sys_geteuid
.llong .sys_getegid /* 50 */
.llong .sys_acct
- .llong .sys32_umount
+ .llong .sys_umount
.llong .sys_ni_syscall /* old lock syscall */
.llong .compat_sys_ioctl
.llong .compat_sys_fcntl /* 55 */
--- diff/arch/ppc64/kernel/pSeries_lpar.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/pSeries_lpar.c 2004-03-16 11:27:47.937899992 +0000
@@ -37,6 +37,12 @@
#include <asm/hvcall.h>
#include <asm/prom.h>
+/* in pSeries_hvCall.S */
+EXPORT_SYMBOL(plpar_hcall);
+EXPORT_SYMBOL(plpar_hcall_4out);
+EXPORT_SYMBOL(plpar_hcall_norets);
+EXPORT_SYMBOL(plpar_hcall_8arg_2ret);
+
long poll_pending(void)
{
unsigned long dummy;
--- diff/arch/ppc64/kernel/pci.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/pci.c 2004-03-16 11:27:47.929901208 +0000
@@ -119,43 +119,6 @@ static void fixup_windbond_82c105(struct
}
}
-/* Given an mmio phys address, find a pci device that implements
- * this address. This is of course expensive, but only used
- * for device initialization or error paths.
- * For io BARs it is assumed the pci_io_base has already been added
- * into addr.
- *
- * Bridges are ignored although they could be used to optimize the search.
- */
-struct pci_dev *pci_find_dev_by_addr(unsigned long addr)
-{
- struct pci_dev *dev = NULL;
- int i;
- unsigned long ioaddr;
-
- ioaddr = (addr > isa_io_base) ? addr - isa_io_base : 0;
-
- while ((dev = pci_find_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
- if ((dev->class >> 16) == PCI_BASE_CLASS_BRIDGE)
- continue;
- for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) {
- unsigned long start = pci_resource_start(dev,i);
- unsigned long end = pci_resource_end(dev,i);
- unsigned int flags = pci_resource_flags(dev,i);
- if (start == 0 || ~start == 0 ||
- end == 0 || ~end == 0)
- continue;
- if ((flags & IORESOURCE_IO) &&
- (ioaddr >= start && ioaddr <= end))
- return dev;
- else if ((flags & IORESOURCE_MEM) &&
- (addr >= start && addr <= end))
- return dev;
- }
- }
- return NULL;
-}
-
void
pcibios_resource_to_bus(struct pci_dev *dev, struct pci_bus_region *region,
struct resource *res)
@@ -359,6 +322,10 @@ static int __init pcibios_init(void)
printk("PCI: Probing PCI hardware done\n");
//ppc64_boot_msg(0x41, "PCI Done");
+#ifdef CONFIG_PPC_PSERIES
+ pci_addr_cache_build();
+#endif
+
return 0;
}
--- diff/arch/ppc64/kernel/pci.h 2003-07-08 08:55:18.000000000 +0000
+++ source/arch/ppc64/kernel/pci.h 2004-03-16 11:27:47.929901208 +0000
@@ -37,11 +37,14 @@ typedef void *(*traverse_func)(struct de
void *traverse_pci_devices(struct device_node *start, traverse_func pre, traverse_func post, void *data);
void *traverse_all_pci_devices(traverse_func pre);
-struct pci_dev *pci_find_dev_by_addr(unsigned long addr);
void pci_devs_phb_init(void);
void pci_fix_bus_sysdata(void);
struct device_node *fetch_dev_dn(struct pci_dev *dev);
#define PCI_GET_PHB_PTR(dev) (((struct device_node *)(dev)->sysdata)->phb)
+/* PCI address cache management routines */
+void pci_addr_cache_insert_device(struct pci_dev *dev);
+void pci_addr_cache_remove_device(struct pci_dev *dev);
+
#endif /* __PPC_KERNEL_PCI_H__ */
--- diff/arch/ppc64/kernel/pmac_iommu.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/pmac_iommu.c 2004-03-16 11:27:47.930901056 +0000
@@ -289,8 +289,11 @@ void iommu_setup_pmac(void)
* things simple. Setup all PCI devices to point to this table
*/
while ((dev = pci_find_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
- dn = PCI_GET_DN(dev);
-
+ /* We must use pci_device_to_OF_node() to make sure that
+ * we get the real "final" pointer to the device in the
+ * pci_dev sysdata and not the temporary PHB one
+ */
+ struct device_node *dn = pci_device_to_OF_node(dev);
if (dn)
dn->iommu_table = &iommu_table_pmac;
}
--- diff/arch/ppc64/kernel/pmc.c 2002-10-16 03:28:29.000000000 +0000
+++ source/arch/ppc64/kernel/pmc.c 2004-03-16 11:27:47.930901056 +0000
@@ -1,167 +0,0 @@
-/*
- * pmc.c
- * Copyright (C) 2001 Dave Engebretsen & Mike Corrigan IBM Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- */
-
-/* Change Activity:
- * 2001/06/05 : engebret : Created.
- * End Change Activity
- */
-
-#include <asm/proc_fs.h>
-#include <asm/paca.h>
-#include <asm/iSeries/ItLpPaca.h>
-#include <asm/iSeries/ItLpQueue.h>
-#include <asm/processor.h>
-
-#include <linux/proc_fs.h>
-#include <linux/spinlock.h>
-#include <asm/pmc.h>
-#include <asm/uaccess.h>
-#include <asm/naca.h>
-
-struct _pmc_sw pmc_sw_system = {
- 0
-};
-
-struct _pmc_sw pmc_sw_cpu[NR_CPUS] = {
- {0 },
-};
-
-/*
- * Provide enough storage for either system level counters or
- * one cpu's counters.
- */
-struct _pmc_sw_text pmc_sw_text;
-struct _pmc_hw_text pmc_hw_text;
-
-char *
-ppc64_pmc_stab(int file)
-{
- int n;
- unsigned long stab_faults, stab_capacity_castouts, stab_invalidations;
- unsigned long i;
-
- stab_faults = stab_capacity_castouts = stab_invalidations = n = 0;
-
- if (file == -1) {
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_online(i))
- continue;
- stab_faults += pmc_sw_cpu[i].stab_faults;
- stab_capacity_castouts += pmc_sw_cpu[i].stab_capacity_castouts;
- stab_invalidations += pmc_sw_cpu[i].stab_invalidations;
- }
- n += sprintf(pmc_sw_text.buffer + n,
- "Faults 0x%lx\n", stab_faults);
- n += sprintf(pmc_sw_text.buffer + n,
- "Castouts 0x%lx\n", stab_capacity_castouts);
- n += sprintf(pmc_sw_text.buffer + n,
- "Invalidations 0x%lx\n", stab_invalidations);
- } else {
- n += sprintf(pmc_sw_text.buffer + n,
- "Faults 0x%lx\n",
- pmc_sw_cpu[file].stab_faults);
-
- n += sprintf(pmc_sw_text.buffer + n,
- "Castouts 0x%lx\n",
- pmc_sw_cpu[file].stab_capacity_castouts);
-
- n += sprintf(pmc_sw_text.buffer + n,
- "Invalidations 0x%lx\n",
- pmc_sw_cpu[file].stab_invalidations);
-
- for (i = 0; i < STAB_ENTRY_MAX; i++) {
- if (pmc_sw_cpu[file].stab_entry_use[i]) {
- n += sprintf(pmc_sw_text.buffer + n,
- "Entry %02ld 0x%lx\n", i,
- pmc_sw_cpu[file].stab_entry_use[i]);
- }
- }
-
- }
-
- return(pmc_sw_text.buffer);
-}
-
-char *
-ppc64_pmc_htab(int file)
-{
- int n;
- unsigned long htab_primary_overflows, htab_capacity_castouts;
- unsigned long htab_read_to_write_faults;
-
- htab_primary_overflows = htab_capacity_castouts = 0;
- htab_read_to_write_faults = n = 0;
-
- if (file == -1) {
- n += sprintf(pmc_sw_text.buffer + n,
- "Primary Overflows 0x%lx\n",
- pmc_sw_system.htab_primary_overflows);
- n += sprintf(pmc_sw_text.buffer + n,
- "Castouts 0x%lx\n",
- pmc_sw_system.htab_capacity_castouts);
- } else {
- n += sprintf(pmc_sw_text.buffer + n,
- "Primary Overflows N/A\n");
-
- n += sprintf(pmc_sw_text.buffer + n,
- "Castouts N/A\n\n");
-
- }
-
- return(pmc_sw_text.buffer);
-}
-
-char *
-ppc64_pmc_hw(int file)
-{
- int n;
-
- n = 0;
- if (file == -1) {
- n += sprintf(pmc_hw_text.buffer + n, "Not Implemented\n");
- } else {
- n += sprintf(pmc_hw_text.buffer + n,
- "MMCR0 0x%lx\n", mfspr(MMCR0));
- n += sprintf(pmc_hw_text.buffer + n,
- "MMCR1 0x%lx\n", mfspr(MMCR1));
-#if 0
- n += sprintf(pmc_hw_text.buffer + n,
- "MMCRA 0x%lx\n", mfspr(MMCRA));
-#endif
-
- n += sprintf(pmc_hw_text.buffer + n,
- "PMC1 0x%lx\n", mfspr(PMC1));
- n += sprintf(pmc_hw_text.buffer + n,
- "PMC2 0x%lx\n", mfspr(PMC2));
- n += sprintf(pmc_hw_text.buffer + n,
- "PMC3 0x%lx\n", mfspr(PMC3));
- n += sprintf(pmc_hw_text.buffer + n,
- "PMC4 0x%lx\n", mfspr(PMC4));
- n += sprintf(pmc_hw_text.buffer + n,
- "PMC5 0x%lx\n", mfspr(PMC5));
- n += sprintf(pmc_hw_text.buffer + n,
- "PMC6 0x%lx\n", mfspr(PMC6));
- n += sprintf(pmc_hw_text.buffer + n,
- "PMC7 0x%lx\n", mfspr(PMC7));
- n += sprintf(pmc_hw_text.buffer + n,
- "PMC8 0x%lx\n", mfspr(PMC8));
- }
-
- return(pmc_hw_text.buffer);
-}
--- diff/arch/ppc64/kernel/ppc_ksyms.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/ppc_ksyms.c 2004-03-16 11:27:47.931900904 +0000
@@ -40,13 +40,8 @@
#include <asm/hw_irq.h>
#include <asm/abs_addr.h>
#include <asm/cacheflush.h>
-#include <asm/proc_fs.h>
#ifdef CONFIG_PPC_ISERIES
-#include <asm/iSeries/iSeries_pci.h>
-#include <asm/iSeries/iSeries_proc.h>
-#include <asm/iSeries/mf.h>
-#include <asm/iSeries/HvLpEvent.h>
-#include <asm/iSeries/HvLpConfig.h>
+#include <asm/iSeries/HvCallSc.h>
#endif
extern int do_signal(sigset_t *, struct pt_regs *);
@@ -59,9 +54,6 @@ EXPORT_SYMBOL(sys_ioctl);
EXPORT_SYMBOL(isa_io_base);
EXPORT_SYMBOL(pci_io_base);
-EXPORT_SYMBOL(find_next_zero_bit);
-EXPORT_SYMBOL(find_next_zero_le_bit);
-
EXPORT_SYMBOL(strcpy);
EXPORT_SYMBOL(strncpy);
EXPORT_SYMBOL(strcat);
@@ -98,7 +90,6 @@ EXPORT_SYMBOL(msChunks);
EXPORT_SYMBOL(reloc_offset);
#ifdef CONFIG_PPC_ISERIES
-EXPORT_SYMBOL(iSeries_proc_callback);
EXPORT_SYMBOL(HvCall0);
EXPORT_SYMBOL(HvCall1);
EXPORT_SYMBOL(HvCall2);
@@ -107,11 +98,6 @@ EXPORT_SYMBOL(HvCall4);
EXPORT_SYMBOL(HvCall5);
EXPORT_SYMBOL(HvCall6);
EXPORT_SYMBOL(HvCall7);
-EXPORT_SYMBOL(HvLpEvent_unregisterHandler);
-EXPORT_SYMBOL(HvLpEvent_registerHandler);
-EXPORT_SYMBOL(mf_allocateLpEvents);
-EXPORT_SYMBOL(mf_deallocateLpEvents);
-EXPORT_SYMBOL(HvLpConfig_getLpIndex_outline);
#endif
EXPORT_SYMBOL(_insb);
@@ -128,32 +114,6 @@ EXPORT_SYMBOL(ioremap);
EXPORT_SYMBOL(__ioremap);
EXPORT_SYMBOL(iounmap);
-#ifdef CONFIG_PCI
-EXPORT_SYMBOL(pci_alloc_consistent);
-EXPORT_SYMBOL(pci_free_consistent);
-EXPORT_SYMBOL(pci_map_single);
-EXPORT_SYMBOL(pci_unmap_single);
-EXPORT_SYMBOL(pci_map_sg);
-EXPORT_SYMBOL(pci_unmap_sg);
-#ifdef CONFIG_PPC_ISERIES
-EXPORT_SYMBOL(iSeries_GetLocationData);
-EXPORT_SYMBOL(iSeries_Device_ToggleReset);
-EXPORT_SYMBOL(iSeries_memset_io);
-EXPORT_SYMBOL(iSeries_memcpy_toio);
-EXPORT_SYMBOL(iSeries_memcpy_fromio);
-EXPORT_SYMBOL(iSeries_Read_Byte);
-EXPORT_SYMBOL(iSeries_Read_Word);
-EXPORT_SYMBOL(iSeries_Read_Long);
-EXPORT_SYMBOL(iSeries_Write_Byte);
-EXPORT_SYMBOL(iSeries_Write_Word);
-EXPORT_SYMBOL(iSeries_Write_Long);
-#endif /* CONFIG_PPC_ISERIES */
-#ifndef CONFIG_PPC_ISERIES
-EXPORT_SYMBOL(eeh_check_failure);
-EXPORT_SYMBOL(eeh_total_mmio_ffs);
-#endif /* CONFIG_PPC_ISERIES */
-#endif /* CONFIG_PCI */
-
EXPORT_SYMBOL(start_thread);
EXPORT_SYMBOL(kernel_thread);
@@ -203,4 +163,5 @@ EXPORT_SYMBOL(console_drivers);
EXPORT_SYMBOL(tb_ticks_per_usec);
EXPORT_SYMBOL(paca);
-EXPORT_SYMBOL(proc_ppc64);
+EXPORT_SYMBOL(cur_cpu_spec);
+EXPORT_SYMBOL(systemcfg);
--- diff/arch/ppc64/kernel/proc_pmc.c 2004-02-09 10:36:09.000000000 +0000
+++ source/arch/ppc64/kernel/proc_pmc.c 2004-03-16 11:27:47.933900600 +0000
@@ -1,810 +0,0 @@
-/*
- * proc_pmc.c
- * Copyright (C) 2001 Mike Corrigan & Dave Engebretsen IBM Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- */
-
-
-/* Change Activity:
- * 2001 : mikec : Created
- * 2001/06/05 : engebret : Software event count support.
- * End Change Activity
- */
-
-#include <linux/proc_fs.h>
-#include <linux/spinlock.h>
-
-#include <asm/proc_fs.h>
-#include <asm/paca.h>
-#include <asm/iSeries/ItLpPaca.h>
-#include <asm/iSeries/ItLpQueue.h>
-#include <asm/iSeries/HvCallXm.h>
-#include <asm/iSeries/IoHriMainStore.h>
-#include <asm/processor.h>
-#include <asm/time.h>
-#include <asm/iSeries/LparData.h>
-#include <asm/pmc.h>
-#include <asm/uaccess.h>
-#include <asm/naca.h>
-#include <asm/rtas.h>
-
-
-static int proc_pmc_control_mode = 0;
-
-static struct proc_dir_entry *proc_ppc64_pmc_system_root = NULL;
-static struct proc_dir_entry *proc_ppc64_pmc_cpu_root[NR_CPUS] = {NULL, };
-
-static spinlock_t proc_ppc64_lock;
-int proc_ppc64_pmc_find_file(void *data);
-int proc_ppc64_pmc_read(char *page, char **start, off_t off,
- int count, int *eof, char *buffer);
-int proc_ppc64_pmc_stab_read(char *page, char **start, off_t off,
- int count, int *eof, void *data);
-int proc_ppc64_pmc_htab_read(char *page, char **start, off_t off,
- int count, int *eof, void *data);
-int proc_ppc64_pmc_hw_read(char *page, char **start, off_t off,
- int count, int *eof, void *data);
-
-static struct proc_dir_entry *pmc_proc_root = NULL;
-
-int proc_get_lpevents( char *page, char **start, off_t off, int count, int *eof, void *data);
-int proc_reset_lpevents( struct file *file, const char *buffer, unsigned long count, void *data);
-
-int proc_get_titanTod( char *page, char **start, off_t off, int count, int *eof, void *data);
-
-int proc_pmc_get_control( char *page, char **start, off_t off, int count, int *eof, void *data);
-
-int proc_pmc_set_control( struct file *file, const char *buffer, unsigned long count, void *data);
-int proc_pmc_set_mmcr0( struct file *file, const char *buffer, unsigned long count, void *data);
-int proc_pmc_set_mmcr1( struct file *file, const char *buffer, unsigned long count, void *data);
-int proc_pmc_set_mmcra( struct file *file, const char *buffer, unsigned long count, void *data);
-int proc_pmc_set_pmc1( struct file *file, const char *buffer, unsigned long count, void *data);
-int proc_pmc_set_pmc2( struct file *file, const char *buffer, unsigned long count, void *data);
-int proc_pmc_set_pmc3( struct file *file, const char *buffer, unsigned long count, void *data);
-int proc_pmc_set_pmc4( struct file *file, const char *buffer, unsigned long count, void *data);
-int proc_pmc_set_pmc5( struct file *file, const char *buffer, unsigned long count, void *data);
-int proc_pmc_set_pmc6( struct file *file, const char *buffer, unsigned long count, void *data);
-int proc_pmc_set_pmc7( struct file *file, const char *buffer, unsigned long count, void *data);
-int proc_pmc_set_pmc8( struct file *file, const char *buffer, unsigned long count, void *data);
-
-#if 0
-int proc_ppc64_init(void)
-{
- unsigned long i;
- struct proc_dir_entry *ent = NULL;
- char buf[256];
-
- printk("proc_ppc64: Creating /proc/ppc64/pmc\n");
-
- /*
- * Create the root, system, and cpu directories as follows:
- * /proc/ppc64/pmc/system
- * /proc/ppc64/pmc/cpu0
- */
- spin_lock(&proc_ppc64_lock);
- if (proc_ppc64.root == NULL) {
- proc_ppc64_init();
- if (!proc_ppc64.root) {
- spin_unlock(&proc_ppc64_lock);
- return;
- }
- }
- spin_unlock(&proc_ppc64_lock);
-
- /* Placeholder for rtas interfaces. */
- if (proc_ppc64.rtas == NULL) {
- return;
- }
-
- proc_ppc64_pmc_root = proc_mkdir("pmc", proc_ppc64.root);
-
- proc_ppc64_pmc_system_root = proc_mkdir("system", proc_ppc64_pmc_root);
- for (i = 0; i < NR_CPUS; i++) {
- if (cpu_online(i)) {
- sprintf(buf, "cpu%ld", i);
- proc_ppc64_pmc_cpu_root[i] =
- proc_mkdir(buf, proc_ppc64_pmc_root);
- }
- }
-
- /* Create directories for the software counters. */
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_online(i))
- continue;
- ent = create_proc_entry("stab", S_IRUGO | S_IWUSR,
- proc_ppc64_pmc_cpu_root[i]);
- if (ent) {
- ent->nlink = 1;
- ent->data = (void *)proc_ppc64_pmc_cpu_root[i];
- ent->read_proc = (void *)proc_ppc64_pmc_stab_read;
- ent->write_proc = NULL;
- }
-
- ent = create_proc_entry("htab", S_IRUGO | S_IWUSR,
- proc_ppc64_pmc_cpu_root[i]);
- if (ent) {
- ent->nlink = 1;
- ent->data = (void *)proc_ppc64_pmc_cpu_root[i];
- ent->read_proc = (void *)proc_ppc64_pmc_htab_read;
- ent->write_proc = NULL;
- }
- }
-
- ent = create_proc_entry("stab", S_IRUGO | S_IWUSR,
- proc_ppc64_pmc_system_root);
- if (ent) {
- ent->nlink = 1;
- ent->data = (void *)proc_ppc64_pmc_system_root;
- ent->read_proc = (void *)proc_ppc64_pmc_stab_read;
- ent->write_proc = NULL;
- }
-
- ent = create_proc_entry("htab", S_IRUGO | S_IWUSR,
- proc_ppc64_pmc_system_root);
- if (ent) {
- ent->nlink = 1;
- ent->data = (void *)proc_ppc64_pmc_system_root;
- ent->read_proc = (void *)proc_ppc64_pmc_htab_read;
- ent->write_proc = NULL;
- }
-
- /* Create directories for the hardware counters. */
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_online(i))
- continue;
- ent = create_proc_entry("hardware", S_IRUGO | S_IWUSR,
- proc_ppc64_pmc_cpu_root[i]);
- if (ent) {
- ent->nlink = 1;
- ent->data = (void *)proc_ppc64_pmc_cpu_root[i];
- ent->read_proc = (void *)proc_ppc64_pmc_hw_read;
- ent->write_proc = NULL;
- }
- }
-
- ent = create_proc_entry("hardware", S_IRUGO | S_IWUSR,
- proc_ppc64_pmc_system_root);
- if (ent) {
- ent->nlink = 1;
- ent->data = (void *)proc_ppc64_pmc_system_root;
- ent->read_proc = (void *)proc_ppc64_pmc_hw_read;
- ent->write_proc = NULL;
- }
-}
-#endif
-
-/*
- * Find the requested 'file' given a proc token.
- *
- * Inputs: void * data: proc token
- * Output: int : (0, ..., +N) = CPU number.
- * -1 = System.
- */
-int proc_ppc64_pmc_find_file(void *data)
-{
- int i;
-
- if ((unsigned long)data ==
- (unsigned long) proc_ppc64_pmc_system_root) {
- return(-1);
- } else {
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_online(i))
- continue;
- if ((unsigned long)data ==
- (unsigned long)proc_ppc64_pmc_cpu_root[i]) {
- return(i);
- }
- }
- }
-
- /* On error, just default to a type of system. */
- printk("proc_ppc64_pmc_find_file: failed to find file token.\n");
- return(-1);
-}
-
-int
-proc_ppc64_pmc_read(char *page, char **start, off_t off,
- int count, int *eof, char *buffer)
-{
- int buffer_size, n;
-
- if (count < 0) return 0;
-
- if (buffer == NULL) {
- *eof = 1;
- return 0;
- }
-
- /* Check for read beyond EOF */
- buffer_size = n = strlen(buffer);
- if (off >= buffer_size) {
- *eof = 1;
- return 0;
- }
- if (n > (buffer_size - off)) n = buffer_size - off;
-
- /* Never return more than was requested */
- if (n > count) {
- n = count;
- } else {
- *eof = 1;
- }
-
- memcpy(page, buffer + off, n);
-
- *start = page;
-
- return n;
-}
-
-int
-proc_ppc64_pmc_stab_read(char *page, char **start, off_t off,
- int count, int *eof, void *data)
-{
- int n, file;
- char *buffer = NULL;
-
- if (count < 0) return 0;
- spin_lock(&proc_ppc64_lock);
-
- /* Figure out which file is being request. */
- file = proc_ppc64_pmc_find_file(data);
-
- /* Update the counters and the text buffer representation. */
- buffer = ppc64_pmc_stab(file);
-
- /* Put the data into the requestor's buffer. */
- n = proc_ppc64_pmc_read(page, start, off, count, eof, buffer);
-
- spin_unlock(&proc_ppc64_lock);
- return n;
-}
-
-int
-proc_ppc64_pmc_htab_read(char *page, char **start, off_t off,
- int count, int *eof, void *data)
-{
- int n, file;
- char *buffer = NULL;
-
- if (count < 0) return 0;
- spin_lock(&proc_ppc64_lock);
-
- /* Figure out which file is being request. */
- file = proc_ppc64_pmc_find_file(data);
-
- /* Update the counters and the text buffer representation. */
- buffer = ppc64_pmc_htab(file);
-
- /* Put the data into the requestor's buffer. */
- n = proc_ppc64_pmc_read(page, start, off, count, eof, buffer);
-
- spin_unlock(&proc_ppc64_lock);
- return n;
-}
-
-int
-proc_ppc64_pmc_hw_read(char *page, char **start, off_t off,
- int count, int *eof, void *data)
-{
- int n, file;
- char *buffer = NULL;
-
- if (count < 0) return 0;
- spin_lock(&proc_ppc64_lock);
-
- /* Figure out which file is being request. */
- file = proc_ppc64_pmc_find_file(data);
-
- /* Update the counters and the text buffer representation. */
- buffer = ppc64_pmc_hw(file);
-
- /* Put the data into the requestor's buffer. */
- n = proc_ppc64_pmc_read(page, start, off, count, eof, buffer);
-
- spin_unlock(&proc_ppc64_lock);
- return n;
-}
-
-/*
- * DRENG the remainder of these functions still need work ...
- */
-void pmc_proc_init(struct proc_dir_entry *iSeries_proc)
-{
- struct proc_dir_entry *ent = NULL;
-
- ent = create_proc_entry("lpevents", S_IFREG|S_IRUGO, iSeries_proc);
- if (!ent) return;
- ent->nlink = 1;
- ent->data = (void *)0;
- ent->read_proc = proc_get_lpevents;
- ent->write_proc = proc_reset_lpevents;
-
- ent = create_proc_entry("titanTod", S_IFREG|S_IRUGO, iSeries_proc);
- if (!ent) return;
- ent->nlink = 1;
- ent->data = (void *)0;
- ent->size = 0;
- ent->read_proc = proc_get_titanTod;
- ent->write_proc = NULL;
-
- pmc_proc_root = proc_mkdir("pmc", iSeries_proc);
- if (!pmc_proc_root) return;
-
- ent = create_proc_entry("control", S_IFREG|S_IRUSR|S_IWUSR, pmc_proc_root);
- if (!ent) return;
- ent->nlink = 1;
- ent->data = (void *)0;
- ent->read_proc = proc_pmc_get_control;
- ent->write_proc = proc_pmc_set_control;
-
-}
-
-static int pmc_calc_metrics( char *page, char **start, off_t off, int count, int *eof, int len)
-{
- if ( len <= off+count)
- *eof = 1;
- *start = page+off;
- len -= off;
- if ( len > count )
- len = count;
- if ( len < 0 )
- len = 0;
- return len;
-}
-
-static char * lpEventTypes[9] = {
- "Hypervisor\t\t",
- "Machine Facilities\t",
- "Session Manager\t",
- "SPD I/O\t\t",
- "Virtual Bus\t\t",
- "PCI I/O\t\t",
- "RIO I/O\t\t",
- "Virtual Lan\t\t",
- "Virtual I/O\t\t"
- };
-
-
-int proc_get_lpevents
-(char *page, char **start, off_t off, int count, int *eof, void *data)
-{
- unsigned i;
- int len = 0;
-
- len += sprintf( page+len, "LpEventQueue 0\n" );
- len += sprintf( page+len, " events processed:\t%lu\n",
- (unsigned long)xItLpQueue.xLpIntCount );
- for (i=0; i<9; ++i) {
- len += sprintf( page+len, " %s %10lu\n",
- lpEventTypes[i],
- (unsigned long)xItLpQueue.xLpIntCountByType[i] );
- }
- len += sprintf( page+len, "\n events processed by processor:\n" );
- for (i = 0; i < NR_CPUS; ++i) {
- if (cpu_online(i))
- len += sprintf( page+len, " CPU%02d %10u\n",
- i, paca[i].lpEvent_count );
- }
-
- return pmc_calc_metrics( page, start, off, count, eof, len );
-
-}
-
-int proc_reset_lpevents( struct file *file, const char *buffer, unsigned long count, void *data )
-{
- return count;
-}
-
-static unsigned long startTitan = 0;
-static unsigned long startTb = 0;
-
-
-int proc_get_titanTod
-(char *page, char **start, off_t off, int count, int *eof, void *data)
-{
- int len = 0;
- unsigned long tb0, titan_tod;
-
- tb0 = get_tb();
- titan_tod = HvCallXm_loadTod();
-
- len += sprintf( page+len, "Titan\n" );
- len += sprintf( page+len, " time base = %016lx\n", tb0 );
- len += sprintf( page+len, " titan tod = %016lx\n", titan_tod );
- len += sprintf( page+len, " xProcFreq = %016x\n", xIoHriProcessorVpd[0].xProcFreq );
- len += sprintf( page+len, " xTimeBaseFreq = %016x\n", xIoHriProcessorVpd[0].xTimeBaseFreq );
- len += sprintf( page+len, " tb_ticks_per_jiffy = %lu\n", tb_ticks_per_jiffy );
- len += sprintf( page+len, " tb_ticks_per_usec = %lu\n", tb_ticks_per_usec );
-
- if ( !startTitan ) {
- startTitan = titan_tod;
- startTb = tb0;
- }
- else {
- unsigned long titan_usec = (titan_tod - startTitan) >> 12;
- unsigned long tb_ticks = (tb0 - startTb);
- unsigned long titan_jiffies = titan_usec / (1000000/HZ);
- unsigned long titan_jiff_usec = titan_jiffies * (1000000/HZ);
- unsigned long titan_jiff_rem_usec = titan_usec - titan_jiff_usec;
- unsigned long tb_jiffies = tb_ticks / tb_ticks_per_jiffy;
- unsigned long tb_jiff_ticks = tb_jiffies * tb_ticks_per_jiffy;
- unsigned long tb_jiff_rem_ticks = tb_ticks - tb_jiff_ticks;
- unsigned long tb_jiff_rem_usec = tb_jiff_rem_ticks / tb_ticks_per_usec;
- unsigned long new_tb_ticks_per_jiffy = (tb_ticks * (1000000/HZ))/titan_usec;
-
- len += sprintf( page+len, " titan elapsed = %lu uSec\n", titan_usec);
- len += sprintf( page+len, " tb elapsed = %lu ticks\n", tb_ticks);
- len += sprintf( page+len, " titan jiffies = %lu.%04lu \n", titan_jiffies, titan_jiff_rem_usec );
- len += sprintf( page+len, " tb jiffies = %lu.%04lu\n", tb_jiffies, tb_jiff_rem_usec );
- len += sprintf( page+len, " new tb_ticks_per_jiffy = %lu\n", new_tb_ticks_per_jiffy );
-
- }
-
- return pmc_calc_metrics( page, start, off, count, eof, len );
-}
-
-int proc_pmc_get_control
-(char *page, char **start, off_t off, int count, int *eof, void *data)
-{
- int len = 0;
-
- if ( proc_pmc_control_mode == PMC_CONTROL_CPI ) {
- unsigned long mach_cycles = mfspr( PMC5 );
- unsigned long inst_complete = mfspr( PMC4 );
- unsigned long inst_dispatch = mfspr( PMC3 );
- unsigned long thread_active_run = mfspr( PMC1 );
- unsigned long thread_active = mfspr( PMC2 );
- unsigned long cpi = 0;
- unsigned long cpithou = 0;
- unsigned long remain;
-
- if ( inst_complete ) {
- cpi = thread_active_run / inst_complete;
- remain = thread_active_run % inst_complete;
- if ( inst_complete > 1000000 )
- cpithou = remain / ( inst_complete / 1000 );
- else
- cpithou = ( remain * 1000 ) / inst_complete;
- }
- len += sprintf( page+len, "PMC CPI Mode\nRaw Counts\n" );
- len += sprintf( page+len, "machine cycles : %12lu\n", mach_cycles );
- len += sprintf( page+len, "thread active cycles : %12lu\n\n", thread_active );
-
- len += sprintf( page+len, "instructions completed : %12lu\n", inst_complete );
- len += sprintf( page+len, "instructions dispatched : %12lu\n", inst_dispatch );
- len += sprintf( page+len, "thread active run cycles : %12lu\n", thread_active_run );
-
- len += sprintf( page+len, "thread active run cycles/instructions completed\n" );
- len += sprintf( page+len, "CPI = %lu.%03lu\n", cpi, cpithou );
-
- }
- else if ( proc_pmc_control_mode == PMC_CONTROL_TLB ) {
- len += sprintf( page+len, "PMC TLB Mode\n" );
- len += sprintf( page+len, "I-miss count : %12lu\n", mfspr( PMC1 ) );
- len += sprintf( page+len, "I-miss latency : %12lu\n", mfspr( PMC2 ) );
- len += sprintf( page+len, "D-miss count : %12lu\n", mfspr( PMC3 ) );
- len += sprintf( page+len, "D-miss latency : %12lu\n", mfspr( PMC4 ) );
- len += sprintf( page+len, "IERAT miss count : %12lu\n", mfspr( PMC5 ) );
- len += sprintf( page+len, "D-reference count : %12lu\n", mfspr( PMC6 ) );
- len += sprintf( page+len, "miss PTEs searched : %12lu\n", mfspr( PMC7 ) );
- len += sprintf( page+len, "miss >8 PTEs searched : %12lu\n", mfspr( PMC8 ) );
- }
- /* IMPLEMENT ME */
- return pmc_calc_metrics( page, start, off, count, eof, len );
-}
-
-unsigned long proc_pmc_conv_int( const char *buf, unsigned count )
-{
- const char * p;
- char b0, b1;
- unsigned v, multiplier, mult, i;
- unsigned long val;
- multiplier = 10;
- p = buf;
- if ( count >= 3 ) {
- b0 = buf[0];
- b1 = buf[1];
- if ( ( b0 == '0' ) &&
- ( ( b1 == 'x' ) || ( b1 == 'X' ) ) ) {
- p = buf + 2;
- count -= 2;
- multiplier = 16;
- }
-
- }
- val = 0;
- for ( i=0; i<count; ++i ) {
- b0 = *p++;
- v = 0;
- mult = multiplier;
- if ( ( b0 >= '0' ) && ( b0 <= '9' ) )
- v = b0 - '0';
- else if ( multiplier == 16 ) {
- if ( ( b0 >= 'a' ) && ( b0 <= 'f' ) )
- v = b0 - 'a' + 10;
- else if ( ( b0 >= 'A' ) && ( b0 <= 'F' ) )
- v = b0 - 'A' + 10;
- else
- mult = 1;
- }
- else
- mult = 1;
- val *= mult;
- val += v;
- }
-
- return val;
-
-}
-
-static inline void proc_pmc_stop(void)
-{
- /* Freeze all counters, leave everything else alone */
- mtspr( MMCR0, mfspr( MMCR0 ) | 0x80000000 );
-}
-
-static inline void proc_pmc_start(void)
-{
- /* Unfreeze all counters, leave everything else alone */
- mtspr( MMCR0, mfspr( MMCR0 ) & ~0x80000000 );
-
-}
-
-static inline void proc_pmc_reset(void)
-{
- /* Clear all the PMCs to zeros
- * Assume a "stop" has already frozen the counters
- * Clear all the PMCs
- */
- mtspr( PMC1, 0 );
- mtspr( PMC2, 0 );
- mtspr( PMC3, 0 );
- mtspr( PMC4, 0 );
- mtspr( PMC5, 0 );
- mtspr( PMC6, 0 );
- mtspr( PMC7, 0 );
- mtspr( PMC8, 0 );
-
-}
-
-static inline void proc_pmc_cpi(void)
-{
- /* Configure the PMC registers to count cycles and instructions */
- /* so we can compute cpi */
- /*
- * MMCRA[30] = 1 Don't count in wait state (CTRL[31]=0)
- * MMCR0[6] = 1 Freeze counters when any overflow
- * MMCR0[19:25] = 0x01 PMC1 counts Thread Active Run Cycles
- * MMCR0[26:31] = 0x05 PMC2 counts Thread Active Cycles
- * MMCR1[0:4] = 0x07 PMC3 counts Instructions Dispatched
- * MMCR1[5:9] = 0x03 PMC4 counts Instructions Completed
- * MMCR1[10:14] = 0x06 PMC5 counts Machine Cycles
- *
- */
-
- proc_pmc_control_mode = PMC_CONTROL_CPI;
-
- /* Indicate to hypervisor that we are using the PMCs */
- get_paca()->xLpPacaPtr->xPMCRegsInUse = 1;
-
- /* Freeze all counters */
- mtspr( MMCR0, 0x80000000 );
- mtspr( MMCR1, 0x00000000 );
-
- /* Clear all the PMCs */
- mtspr( PMC1, 0 );
- mtspr( PMC2, 0 );
- mtspr( PMC3, 0 );
- mtspr( PMC4, 0 );
- mtspr( PMC5, 0 );
- mtspr( PMC6, 0 );
- mtspr( PMC7, 0 );
- mtspr( PMC8, 0 );
-
- /* Freeze counters in Wait State (CTRL[31]=0) */
- mtspr( MMCRA, 0x00000002 );
-
- /* PMC3<-0x07, PMC4<-0x03, PMC5<-0x06 */
- mtspr( MMCR1, 0x38cc0000 );
-
- mb();
-
- /* PMC1<-0x01, PMC2<-0x05
- * Start all counters
- */
- mtspr( MMCR0, 0x02000045 );
-
-}
-
-static inline void proc_pmc_tlb(void)
-{
- /* Configure the PMC registers to count tlb misses */
- /*
- * MMCR0[6] = 1 Freeze counters when any overflow
- * MMCR0[19:25] = 0x55 Group count
- * PMC1 counts I misses
- * PMC2 counts I miss duration (latency)
- * PMC3 counts D misses
- * PMC4 counts D miss duration (latency)
- * PMC5 counts IERAT misses
- * PMC6 counts D references (including PMC7)
- * PMC7 counts miss PTEs searched
- * PMC8 counts miss >8 PTEs searched
- *
- */
-
- proc_pmc_control_mode = PMC_CONTROL_TLB;
-
- /* Indicate to hypervisor that we are using the PMCs */
- get_paca()->xLpPacaPtr->xPMCRegsInUse = 1;
-
- /* Freeze all counters */
- mtspr( MMCR0, 0x80000000 );
- mtspr( MMCR1, 0x00000000 );
-
- /* Clear all the PMCs */
- mtspr( PMC1, 0 );
- mtspr( PMC2, 0 );
- mtspr( PMC3, 0 );
- mtspr( PMC4, 0 );
- mtspr( PMC5, 0 );
- mtspr( PMC6, 0 );
- mtspr( PMC7, 0 );
- mtspr( PMC8, 0 );
-
- mtspr( MMCRA, 0x00000000 );
-
- mb();
-
- /* PMC1<-0x55
- * Start all counters
- */
- mtspr( MMCR0, 0x02001540 );
-
-}
-
-int proc_pmc_set_control( struct file *file, const char *buffer, unsigned long count, void *data )
-{
- char stkbuf[10];
- if (count > 9) count = 9;
- if (copy_from_user (stkbuf, buffer, count)) {
- return -EFAULT;
- }
- stkbuf[count] = 0;
-
- if ( ! strncmp( stkbuf, "stop", 4 ) )
- proc_pmc_stop();
- else if ( ! strncmp( stkbuf, "start", 5 ) )
- proc_pmc_start();
- else if ( ! strncmp( stkbuf, "reset", 5 ) )
- proc_pmc_reset();
- else if ( ! strncmp( stkbuf, "cpi", 3 ) )
- proc_pmc_cpi();
- else if ( ! strncmp( stkbuf, "tlb", 3 ) )
- proc_pmc_tlb();
-
- /* IMPLEMENT ME */
- return count;
-}
-
-int proc_pmc_set_mmcr0( struct file *file, const char *buffer, unsigned long count, void *data )
-{
- unsigned long v;
- v = proc_pmc_conv_int( buffer, count );
- v = v & ~0x04000000; /* Don't allow interrupts for now */
- if ( v & ~0x80000000 ) /* Inform hypervisor we are using PMCs */
- get_paca()->xLpPacaPtr->xPMCRegsInUse = 1;
- else
- get_paca()->xLpPacaPtr->xPMCRegsInUse = 0;
- mtspr( MMCR0, v );
-
- return count;
-}
-
-int proc_pmc_set_mmcr1( struct file *file, const char *buffer, unsigned long count, void *data )
-{
- unsigned long v;
- v = proc_pmc_conv_int( buffer, count );
- mtspr( MMCR1, v );
-
- return count;
-}
-
-int proc_pmc_set_mmcra( struct file *file, const char *buffer, unsigned long count, void *data )
-{
- unsigned long v;
- v = proc_pmc_conv_int( buffer, count );
- v = v & ~0x00008000; /* Don't allow interrupts for now */
- mtspr( MMCRA, v );
-
- return count;
-}
-
-
-int proc_pmc_set_pmc1( struct file *file, const char *buffer, unsigned long count, void *data )
-{
- unsigned long v;
- v = proc_pmc_conv_int( buffer, count );
- mtspr( PMC1, v );
-
- return count;
-}
-
-int proc_pmc_set_pmc2( struct file *file, const char *buffer, unsigned long count, void *data )
-{
- unsigned long v;
- v = proc_pmc_conv_int( buffer, count );
- mtspr( PMC2, v );
-
- return count;
-}
-
-int proc_pmc_set_pmc3( struct file *file, const char *buffer, unsigned long count, void *data )
-{
- unsigned long v;
- v = proc_pmc_conv_int( buffer, count );
- mtspr( PMC3, v );
-
- return count;
-}
-
-int proc_pmc_set_pmc4( struct file *file, const char *buffer, unsigned long count, void *data )
-{
- unsigned long v;
- v = proc_pmc_conv_int( buffer, count );
- mtspr( PMC4, v );
-
- return count;
-}
-
-int proc_pmc_set_pmc5( struct file *file, const char *buffer, unsigned long count, void *data )
-{
- unsigned long v;
- v = proc_pmc_conv_int( buffer, count );
- mtspr( PMC5, v );
-
- return count;
-}
-
-int proc_pmc_set_pmc6( struct file *file, const char *buffer, unsigned long count, void *data )
-{
- unsigned long v;
- v = proc_pmc_conv_int( buffer, count );
- mtspr( PMC6, v );
-
- return count;
-}
-
-int proc_pmc_set_pmc7( struct file *file, const char *buffer, unsigned long count, void *data )
-{
- unsigned long v;
- v = proc_pmc_conv_int( buffer, count );
- mtspr( PMC7, v );
-
- return count;
-}
-
-int proc_pmc_set_pmc8( struct file *file, const char *buffer, unsigned long count, void *data )
-{
- unsigned long v;
- v = proc_pmc_conv_int( buffer, count );
- mtspr( PMC8, v );
-
- return count;
-}
-
--- diff/arch/ppc64/kernel/proc_ppc64.c 2004-02-18 08:54:08.000000000 +0000
+++ source/arch/ppc64/kernel/proc_ppc64.c 2004-03-16 11:27:47.934900448 +0000
@@ -18,15 +18,6 @@
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
-
-/*
- * Change Activity:
- * 2001 : mikec : Created
- * 2001/06/05 : engebret : Software event count support.
- * 2003/02/13 : bergner : Move PMC code to pmc.c
- * End Change Activity
- */
-
#include <linux/config.h>
#include <linux/init.h>
#include <linux/mm.h>
@@ -34,7 +25,6 @@
#include <linux/slab.h>
#include <linux/kernel.h>
-#include <asm/proc_fs.h>
#include <asm/naca.h>
#include <asm/paca.h>
#include <asm/systemcfg.h>
@@ -42,10 +32,6 @@
#include <asm/uaccess.h>
#include <asm/prom.h>
-struct proc_ppc64_t proc_ppc64;
-
-void proc_ppc64_create_paca(int num);
-
static loff_t page_map_seek( struct file *file, loff_t off, int whence);
static ssize_t page_map_read( struct file *file, char *buf, size_t nbytes, loff_t *ppos);
static int page_map_mmap( struct file *file, struct vm_area_struct *vma );
@@ -59,7 +45,7 @@ static struct file_operations page_map_f
#ifdef CONFIG_PPC_PSERIES
/* routines for /proc/ppc64/ofdt */
static ssize_t ofdt_write(struct file *, const char __user *, size_t, loff_t *);
-static void proc_ppc64_create_ofdt(struct proc_dir_entry *);
+static void proc_ppc64_create_ofdt(void);
static int do_remove_node(char *);
static int do_add_node(char *, size_t);
static void release_prop_list(const struct property *);
@@ -70,77 +56,19 @@ static struct file_operations ofdt_fops
};
#endif
-int __init proc_ppc64_init(void)
-{
-
-
- if (proc_ppc64.root == NULL) {
- printk(KERN_INFO "proc_ppc64: Creating /proc/ppc64/\n");
- proc_ppc64.root = proc_mkdir("ppc64", 0);
- if (!proc_ppc64.root)
- return 0;
- } else {
- return 0;
- }
-
- proc_ppc64.naca = create_proc_entry("naca", S_IRUSR, proc_ppc64.root);
- if ( proc_ppc64.naca ) {
- proc_ppc64.naca->nlink = 1;
- proc_ppc64.naca->data = naca;
- proc_ppc64.naca->size = 4096;
- proc_ppc64.naca->proc_fops = &page_map_fops;
- }
-
- proc_ppc64.systemcfg = create_proc_entry("systemcfg", S_IFREG|S_IRUGO, proc_ppc64.root);
- if ( proc_ppc64.systemcfg ) {
- proc_ppc64.systemcfg->nlink = 1;
- proc_ppc64.systemcfg->data = systemcfg;
- proc_ppc64.systemcfg->size = 4096;
- proc_ppc64.systemcfg->proc_fops = &page_map_fops;
- }
-
- /* /proc/ppc64/paca/XX -- raw paca contents. Only readable to root */
- proc_ppc64.paca = proc_mkdir("paca", proc_ppc64.root);
- if (proc_ppc64.paca) {
- unsigned long i;
-
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_online(i))
- continue;
- proc_ppc64_create_paca(i);
- }
- }
-
-#ifdef CONFIG_PPC_PSERIES
- /* Placeholder for rtas interfaces. */
- if (proc_ppc64.rtas == NULL)
- proc_ppc64.rtas = proc_mkdir("rtas", proc_ppc64.root);
-
- if (proc_ppc64.rtas)
- proc_symlink("rtas", 0, "ppc64/rtas");
-
- proc_ppc64_create_ofdt(proc_ppc64.root);
-#endif
-
- return 0;
-}
-
-
/*
- * NOTE: since paca data is always in flux the values will never be a consistant set.
- * In theory it could be made consistent if we made the corresponding cpu
- * copy the page for us (via an IPI). Probably not worth it.
- *
+ * NOTE: since paca data is always in flux the values will never be a
+ * consistant set.
*/
-void proc_ppc64_create_paca(int num)
+static void __init proc_create_paca(struct proc_dir_entry *dir, int num)
{
struct proc_dir_entry *ent;
struct paca_struct *lpaca = paca + num;
char buf[16];
sprintf(buf, "%02x", num);
- ent = create_proc_entry(buf, S_IRUSR, proc_ppc64.paca);
- if ( ent ) {
+ ent = create_proc_entry(buf, S_IRUSR, dir);
+ if (ent) {
ent->nlink = 1;
ent->data = lpaca;
ent->size = 4096;
@@ -148,6 +76,67 @@ void proc_ppc64_create_paca(int num)
}
}
+/*
+ * Create the ppc64 and ppc64/rtas directories early. This allows us to
+ * assume that they have been previously created in drivers.
+ */
+static int __init proc_ppc64_create(void)
+{
+ struct proc_dir_entry *root;
+
+ root = proc_mkdir("ppc64", 0);
+ if (!root)
+ return 1;
+
+ if (!(systemcfg->platform & PLATFORM_PSERIES))
+ return 0;
+
+ if (!proc_mkdir("rtas", root))
+ return 1;
+
+ if (!proc_symlink("rtas", 0, "ppc64/rtas"))
+ return 1;
+
+ return 0;
+}
+core_initcall(proc_ppc64_create);
+
+static int __init proc_ppc64_init(void)
+{
+ unsigned long i;
+ struct proc_dir_entry *pde;
+
+ pde = create_proc_entry("ppc64/naca", S_IRUSR, NULL);
+ if (!pde)
+ return 1;
+ pde->nlink = 1;
+ pde->data = naca;
+ pde->size = 4096;
+ pde->proc_fops = &page_map_fops;
+
+ pde = create_proc_entry("ppc64/systemcfg", S_IFREG|S_IRUGO, NULL);
+ if (!pde)
+ return 1;
+ pde->nlink = 1;
+ pde->data = systemcfg;
+ pde->size = 4096;
+ pde->proc_fops = &page_map_fops;
+
+ /* /proc/ppc64/paca/XX -- raw paca contents. Only readable to root */
+ pde = proc_mkdir("ppc64/paca", NULL);
+ if (!pde)
+ return 1;
+ for_each_cpu(i)
+ proc_create_paca(pde, i);
+
+#ifdef CONFIG_PPC_PSERIES
+ if ((systemcfg->platform & PLATFORM_PSERIES))
+ proc_ppc64_create_ofdt();
+#endif
+
+ return 0;
+}
+__initcall(proc_ppc64_init);
static loff_t page_map_seek( struct file *file, loff_t off, int whence)
{
@@ -204,11 +193,11 @@ static int page_map_mmap( struct file *f
#ifdef CONFIG_PPC_PSERIES
/* create /proc/ppc64/ofdt write-only by root */
-static void proc_ppc64_create_ofdt(struct proc_dir_entry *parent)
+static void proc_ppc64_create_ofdt(void)
{
struct proc_dir_entry *ent;
- ent = create_proc_entry("ofdt", S_IWUSR, parent);
+ ent = create_proc_entry("ppc64/ofdt", S_IWUSR, NULL);
if (ent) {
ent->nlink = 1;
ent->data = NULL;
@@ -423,5 +412,3 @@ static void release_prop_list(const stru
}
#endif /* defined(CONFIG_PPC_PSERIES) */
-
-fs_initcall(proc_ppc64_init);
--- diff/arch/ppc64/kernel/process.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/process.c 2004-03-16 11:27:47.932900752 +0000
@@ -34,6 +34,7 @@
#include <linux/prctl.h>
#include <linux/ptrace.h>
#include <linux/kallsyms.h>
+#include <linux/version.h>
#include <asm/pgtable.h>
#include <asm/uaccess.h>
@@ -193,8 +194,8 @@ void show_regs(struct pt_regs * regs)
printk("NIP: %016lX XER: %016lX LR: %016lX\n",
regs->nip, regs->xer, regs->link);
- printk("REGS: %p TRAP: %04lx %s\n",
- regs, regs->trap, print_tainted());
+ printk("REGS: %p TRAP: %04lx %s (%s)\n",
+ regs, regs->trap, print_tainted(), UTS_RELEASE);
printk("MSR: %016lx EE: %01x PR: %01x FP: %01x ME: %01x IR/DR: %01x%01x\n",
regs->msr, regs->msr&MSR_EE ? 1 : 0, regs->msr&MSR_PR ? 1 : 0,
regs->msr & MSR_FP ? 1 : 0,regs->msr&MSR_ME ? 1 : 0,
--- diff/arch/ppc64/kernel/prom.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/prom.c 2004-03-16 11:27:47.937899992 +0000
@@ -61,6 +61,14 @@ extern const struct linux_logo logo_linu
#endif
/*
+ * Properties whose value is longer than this get excluded from our
+ * copy of the device tree. This value does need to be big enough to
+ * ensure that we don't lose things like the interrupt-map property
+ * on a PCI-PCI bridge.
+ */
+#define MAX_PROPERTY_LENGTH (1UL * 1024 * 1024)
+
+/*
* prom_init() is called very early on, before the kernel text
* and data have been mapped to KERNELBASE. At this point the code
* is running at whatever address it has been loaded at, so
@@ -908,9 +916,11 @@ prom_initialize_tce_table(void)
*tce_entryp = tce_entry;
}
+ /* It seems OF doesn't null-terminate the path :-( */
+ memset(path, 0, sizeof(path));
/* Call OF to setup the TCE hardware */
if (call_prom(RELOC("package-to-path"), 3, 1, node,
- path, 255) <= 0) {
+ path, sizeof(path)-1) <= 0) {
prom_print(RELOC("package-to-path failed\n"));
} else {
prom_print(RELOC("opened "));
@@ -1687,6 +1697,11 @@ check_display(unsigned long mem)
/* It seems OF doesn't null-terminate the path :-( */
path = (char *) mem;
memset(path, 0, 256);
+
+ /*
+ * leave some room at the end of the path for appending extra
+ * arguments
+ */
if ((long) call_prom(RELOC("package-to-path"), 3, 1,
node, path, 250) < 0)
continue;
@@ -1794,8 +1809,7 @@ copy_device_tree(unsigned long mem_start
return new_start;
}
-__init
-static unsigned long
+static unsigned long __init
inspect_node(phandle node, struct device_node *dad,
unsigned long mem_start, unsigned long mem_end,
struct device_node ***allnextpp)
@@ -1843,6 +1857,22 @@ inspect_node(phandle node, struct device
valp, mem_end - mem_start);
if (pp->length < 0)
continue;
+ if (pp->length > MAX_PROPERTY_LENGTH) {
+ char path[128];
+
+ prom_print(RELOC("WARNING: ignoring large property "));
+ /* It seems OF doesn't null-terminate the path :-( */
+ memset(path, 0, sizeof(path));
+ if (call_prom(RELOC("package-to-path"), 3, 1, node,
+ path, sizeof(path)-1) > 0)
+ prom_print(path);
+ prom_print(namep);
+ prom_print(RELOC(" length 0x"));
+ prom_print_hex(pp->length);
+ prom_print_nl();
+
+ continue;
+ }
mem_start = DOUBLEWORD_ALIGN(mem_start + pp->length);
*prev_propp = PTRUNRELOC(pp);
prev_propp = &pp->next;
@@ -2853,40 +2883,12 @@ int of_remove_node(struct device_node *n
struct device_node *parent, *child;
parent = of_get_parent(np);
-
if (!parent)
return -EINVAL;
- /* Make sure we are not recursively removing
- * more than one level of nodes. We need to
- * allow this so we can remove a slot containing
- * an IOA.
- */
- for (child = of_get_next_child(np, NULL);
- child != NULL;
- child = of_get_next_child(np, child)) {
- struct device_node *grandchild;
-
- if ((grandchild = of_get_next_child(child, NULL))) {
- /* Too deep */
- of_node_put(grandchild);
- of_node_put(child);
- return -EBUSY;
- }
- }
-
- /* Now that we're reasonably sure that we won't
- * overflow our stack, remove any children of np.
- */
- for (child = of_get_next_child(np, NULL);
- child != NULL;
- child = of_get_next_child(np, child)) {
- int rc;
-
- if ((rc = of_remove_node(child))) {
- of_node_put(child);
- return rc;
- }
+ if ((child = of_get_next_child(np, NULL))) {
+ of_node_put(child);
+ return -EBUSY;
}
write_lock(&devtree_lock);
@@ -2964,6 +2966,53 @@ static void remove_node_proc_entries(str
#endif /* CONFIG_PROC_DEVICETREE */
/*
+ * Fix up n_intrs and intrs fields in a new device node
+ *
+ */
+static int of_finish_dynamic_node_interrupts(struct device_node *node)
+{
+ int intrcells, intlen, i;
+ unsigned *irq, *ints, virq;
+ struct device_node *ic;
+
+ ints = (unsigned int *)get_property(node, "interrupts", &intlen);
+ intrcells = prom_n_intr_cells(node);
+ intlen /= intrcells * sizeof(unsigned int);
+ node->n_intrs = intlen;
+ node->intrs = kmalloc(sizeof(struct interrupt_info) * intlen,
+ GFP_KERNEL);
+ if (!node->intrs)
+ return -ENOMEM;
+
+ for (i = 0; i < intlen; ++i) {
+ int n, j;
+ node->intrs[i].line = 0;
+ node->intrs[i].sense = 1;
+ n = map_interrupt(&irq, &ic, node, ints, intrcells);
+ if (n <= 0)
+ continue;
+ virq = virt_irq_create_mapping(irq[0]);
+ if (virq == NO_IRQ) {
+ printk(KERN_CRIT "Could not allocate interrupt "
+ "number for %s\n", node->full_name);
+ return -ENOMEM;
+ }
+ node->intrs[i].line = openpic_to_irq(virq);
+ if (n > 1)
+ node->intrs[i].sense = irq[1];
+ if (n > 2) {
+ printk(KERN_DEBUG "hmmm, got %d intr cells for %s:", n,
+ node->full_name);
+ for (j = 0; j < n; ++j)
+ printk(" %d", irq[j]);
+ printk("\n");
+ }
+ ints += intrcells;
+ }
+ return 0;
+}
+
+/*
* Fix up the uninitialized fields in a new device node:
* name, type, n_addrs, addrs, n_intrs, intrs, and pci-specific fields
*
@@ -2978,11 +3027,8 @@ static int of_finish_dynamic_node(struct
{
struct device_node *parent = of_get_parent(node);
u32 *regs;
- unsigned int *ints;
- int intlen, intrcells;
- int i, j, n, err = 0;
- unsigned int *irq, virq;
- struct device_node *ic;
+ int err = 0;
+ phandle *ibm_phandle;
node->name = get_property(node, "name", 0);
node->type = get_property(node, "device_type", 0);
@@ -2998,6 +3044,10 @@ static int of_finish_dynamic_node(struct
if (systemcfg->platform == PLATFORM_POWERMAC)
return -ENODEV;
+ /* fix up new node's linux_phandle field */
+ if ((ibm_phandle = (unsigned int *)get_property(node, "ibm,phandle", NULL)))
+ node->linux_phandle = *ibm_phandle;
+
/* do the work of interpret_pci_props */
if (parent->type && !strcmp(parent->type, "pci")) {
struct address_range *adr;
@@ -3027,45 +3077,9 @@ static int of_finish_dynamic_node(struct
}
/* now do the work of finish_node_interrupts */
-
- ints = (unsigned int *) get_property(node, "interrupts", &intlen);
- if (!ints) {
- err = -ENODEV;
- goto out;
- }
-
- intrcells = prom_n_intr_cells(node);
- intlen /= intrcells * sizeof(unsigned int);
- node->n_intrs = intlen;
- node->intrs = kmalloc(sizeof(struct interrupt_info) * intlen,
- GFP_KERNEL);
- if (!node->intrs) {
- err = -ENOMEM;
- goto out;
- }
-
- for (i = 0; i < intlen; ++i) {
- node->intrs[i].line = 0;
- node->intrs[i].sense = 1;
- n = map_interrupt(&irq, &ic, node, ints, intrcells);
- if (n <= 0)
- continue;
- virq = virt_irq_create_mapping(irq[0]);
- if (virq == NO_IRQ) {
- printk(KERN_CRIT "Could not allocate interrupt "
- "number for %s\n", node->full_name);
- } else
- node->intrs[i].line = openpic_to_irq(virq);
- if (n > 1)
- node->intrs[i].sense = irq[1];
- if (n > 2) {
- printk(KERN_DEBUG "hmmm, got %d intr cells for %s:", n,
- node->full_name);
- for (j = 0; j < n; ++j)
- printk(" %d", irq[j]);
- printk("\n");
- }
- ints += intrcells;
+ if (get_property(node, "interrupts", 0)) {
+ err = of_finish_dynamic_node_interrupts(node);
+ if (err) goto out;
}
/* now do the rough equivalent of update_dn_pci_info, this
--- diff/arch/ppc64/kernel/ptrace.c 2003-02-26 16:01:01.000000000 +0000
+++ source/arch/ppc64/kernel/ptrace.c 2004-03-16 11:27:47.938899840 +0000
@@ -26,6 +26,7 @@
#include <linux/ptrace.h>
#include <linux/user.h>
#include <linux/security.h>
+#include <linux/audit.h>
#include <asm/uaccess.h>
#include <asm/page.h>
@@ -286,12 +287,8 @@ out:
return ret;
}
-void do_syscall_trace(void)
+static void do_syscall_trace(void)
{
- if (!test_thread_flag(TIF_SYSCALL_TRACE))
- return;
- if (!(current->ptrace & PT_PTRACED))
- return;
/* the 0x80 provides a way for the tracing parent to distinguish
between a syscall stop and SIGTRAP delivery */
ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD)
@@ -307,3 +304,23 @@ void do_syscall_trace(void)
current->exit_code = 0;
}
}
+
+void do_syscall_trace_enter(struct pt_regs *regs)
+{
+ if (unlikely(current->audit_context))
+ audit_syscall_entry(current, regs->gpr[0], regs->gpr[3]);
+
+ if (test_thread_flag(TIF_SYSCALL_TRACE)
+ && (current->ptrace & PT_PTRACED))
+ do_syscall_trace();
+}
+
+void do_syscall_trace_leave(void)
+{
+ if (unlikely(current->audit_context))
+ audit_syscall_exit(current);
+
+ if (test_thread_flag(TIF_SYSCALL_TRACE)
+ && (current->ptrace & PT_PTRACED))
+ do_syscall_trace();
+}
--- diff/arch/ppc64/kernel/ras.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/ras.c 2004-03-16 11:27:47.938899840 +0000
@@ -36,7 +36,6 @@
#include <linux/pci.h>
#include <linux/delay.h>
#include <linux/irq.h>
-#include <linux/proc_fs.h>
#include <linux/random.h>
#include <linux/sysrq.h>
--- diff/arch/ppc64/kernel/rtas-proc.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/rtas-proc.c 2004-03-16 11:27:47.942899232 +0000
@@ -28,7 +28,6 @@
#include <asm/io.h>
#include <asm/prom.h>
#include <asm/rtas.h>
-#include <asm/proc_fs.h>
#include <asm/machdep.h> /* for ppc_md */
#include <asm/time.h>
@@ -117,8 +116,6 @@
/* Globals */
-extern struct proc_dir_entry *proc_rtas;
-
static struct rtas_sensors sensors;
static struct device_node *rtas_node = NULL;
static unsigned long power_on_time = 0; /* Save the time the user set */
@@ -201,51 +198,50 @@ int get_location_code(struct individual_
int check_location_string (char *c, char * buf);
int check_location (char *c, int idx, char * buf);
-/* ****************************************************************** */
-/* MAIN */
-/* ****************************************************************** */
-void proc_rtas_init(void)
+static int __init proc_rtas_init(void)
{
struct proc_dir_entry *entry;
- rtas_node = of_find_node_by_name(NULL, "rtas");
- if ((rtas_node == NULL) || (systemcfg->platform == PLATFORM_ISERIES_LPAR)) {
- return;
- }
-
- if (proc_ppc64.rtas == NULL) {
- proc_ppc64_init();
- }
-
- if (proc_ppc64.rtas == NULL) {
- printk(KERN_ERR "Failed to create /proc/rtas in proc_rtas_init\n");
- return;
- }
-
- /* /proc/rtas entries */
-
- entry = create_proc_entry("progress", S_IRUGO|S_IWUSR, proc_ppc64.rtas);
- if (entry) entry->proc_fops = &ppc_rtas_progress_operations;
-
- entry = create_proc_entry("clock", S_IRUGO|S_IWUSR, proc_ppc64.rtas);
- if (entry) entry->proc_fops = &ppc_rtas_clock_operations;
-
- entry = create_proc_entry("poweron", S_IWUSR|S_IRUGO, proc_ppc64.rtas);
- if (entry) entry->proc_fops = &ppc_rtas_poweron_operations;
+ if (!(systemcfg->platform & PLATFORM_PSERIES))
+ return 1;
- create_proc_read_entry("sensors", S_IRUGO, proc_ppc64.rtas,
- ppc_rtas_sensor_read, NULL);
-
- entry = create_proc_entry("frequency", S_IWUSR|S_IRUGO, proc_ppc64.rtas);
- if (entry) entry->proc_fops = &ppc_rtas_tone_freq_operations;
+ rtas_node = of_find_node_by_name(NULL, "rtas");
+ if (rtas_node == NULL)
+ return 1;
- entry = create_proc_entry("volume", S_IWUSR|S_IRUGO, proc_ppc64.rtas);
- if (entry) entry->proc_fops = &ppc_rtas_tone_volume_operations;
+ entry = create_proc_entry("ppc64/rtas/progress", S_IRUGO|S_IWUSR, NULL);
+ if (entry)
+ entry->proc_fops = &ppc_rtas_progress_operations;
+
+ entry = create_proc_entry("ppc64/rtas/clock", S_IRUGO|S_IWUSR, NULL);
+ if (entry)
+ entry->proc_fops = &ppc_rtas_clock_operations;
+
+ entry = create_proc_entry("ppc64/rtas/poweron", S_IWUSR|S_IRUGO, NULL);
+ if (entry)
+ entry->proc_fops = &ppc_rtas_poweron_operations;
+
+ create_proc_read_entry("ppc64/rtas/sensors", S_IRUGO, NULL,
+ ppc_rtas_sensor_read, NULL);
+
+ entry = create_proc_entry("ppc64/rtas/frequency", S_IWUSR|S_IRUGO,
+ NULL);
+ if (entry)
+ entry->proc_fops = &ppc_rtas_tone_freq_operations;
+
+ entry = create_proc_entry("ppc64/rtas/volume", S_IWUSR|S_IRUGO, NULL);
+ if (entry)
+ entry->proc_fops = &ppc_rtas_tone_volume_operations;
+
+ entry = create_proc_entry("ppc64/rtas/rmo_buffer", S_IRUSR, NULL);
+ if (entry)
+ entry->proc_fops = &ppc_rtas_rmo_buf_ops;
- entry = create_proc_entry("rmo_buffer", S_IRUSR, proc_ppc64.rtas);
- if (entry) entry->proc_fops = &ppc_rtas_rmo_buf_ops;
+ return 0;
}
+__initcall(proc_rtas_init);
+
/* ****************************************************************** */
/* POWER-ON-TIME */
/* ****************************************************************** */
--- diff/arch/ppc64/kernel/rtas.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/rtas.c 2004-03-16 11:27:47.939899688 +0000
@@ -19,7 +19,6 @@
#include <linux/init.h>
#include <asm/prom.h>
-#include <asm/proc_fs.h>
#include <asm/rtas.h>
#include <asm/semaphore.h>
#include <asm/machdep.h>
@@ -259,6 +258,33 @@ rtas_get_power_level(int powerdomain, in
}
int
+rtas_set_power_level(int powerdomain, int level, int *setlevel)
+{
+ int token = rtas_token("set-power-level");
+ unsigned int wait_time;
+ long returned_level;
+ int rc;
+
+ if (token == RTAS_UNKNOWN_SERVICE)
+ return RTAS_UNKNOWN_OP;
+
+ while (1) {
+ rc = (int) rtas_call(token, 2, 2, &returned_level, powerdomain,
+ level);
+ if (rc == RTAS_BUSY)
+ udelay(1);
+ else if (rtas_is_extended_busy(rc)) {
+ wait_time = rtas_extended_busy_delay_time(rc);
+ udelay(wait_time * 1000);
+ }
+ else
+ break;
+ }
+ *setlevel = (int) returned_level;
+ return rc;
+}
+
+int
rtas_get_sensor(int sensor, int index, int *state)
{
int token = rtas_token("get-sensor-state");
@@ -426,6 +452,7 @@ asmlinkage int ppc_rtas(struct rtas_args
{
struct rtas_args args;
unsigned long flags;
+ int nargs;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
@@ -433,14 +460,15 @@ asmlinkage int ppc_rtas(struct rtas_args
if (copy_from_user(&args, uargs, 3 * sizeof(u32)) != 0)
return -EFAULT;
- if (args.nargs > ARRAY_SIZE(args.args)
+ nargs = args.nargs;
+ if (nargs > ARRAY_SIZE(args.args)
|| args.nret > ARRAY_SIZE(args.args)
- || args.nargs + args.nret > ARRAY_SIZE(args.args))
+ || nargs + args.nret > ARRAY_SIZE(args.args))
return -EINVAL;
/* Copy in args. */
if (copy_from_user(args.args, uargs->args,
- args.nargs * sizeof(rtas_arg_t)) != 0)
+ nargs * sizeof(rtas_arg_t)) != 0)
return -EFAULT;
spin_lock_irqsave(&rtas.lock, flags);
@@ -449,14 +477,15 @@ asmlinkage int ppc_rtas(struct rtas_args
enter_rtas((void *)__pa((unsigned long)&get_paca()->xRtas));
args = get_paca()->xRtas;
+ args.rets = (rtas_arg_t *)&(args.args[nargs]);
if (args.rets[0] == -1)
log_rtas_error(&args);
spin_unlock_irqrestore(&rtas.lock, flags);
/* Copy out args. */
- if (copy_to_user(uargs->args + args.nargs,
- args.args + args.nargs,
+ if (copy_to_user(uargs->args + nargs,
+ args.args + nargs,
args.nret * sizeof(rtas_arg_t)) != 0)
return -EFAULT;
@@ -472,4 +501,5 @@ EXPORT_SYMBOL(rtas_data_buf_lock);
EXPORT_SYMBOL(rtas_extended_busy_delay_time);
EXPORT_SYMBOL(rtas_get_sensor);
EXPORT_SYMBOL(rtas_get_power_level);
+EXPORT_SYMBOL(rtas_set_power_level);
EXPORT_SYMBOL(rtas_set_indicator);
--- diff/arch/ppc64/kernel/rtas_flash.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/rtas_flash.c 2004-03-16 11:27:47.940899536 +0000
@@ -15,7 +15,7 @@
#include <linux/module.h>
#include <linux/init.h>
-#include <asm/proc_fs.h>
+#include <linux/proc_fs.h>
#include <asm/delay.h>
#include <asm/uaccess.h>
#include <asm/rtas.h>
@@ -106,10 +106,10 @@ struct rtas_validate_flash_t
};
static spinlock_t flash_file_open_lock = SPIN_LOCK_UNLOCKED;
-static struct proc_dir_entry *firmware_flash_pde = NULL;
-static struct proc_dir_entry *firmware_update_pde = NULL;
-static struct proc_dir_entry *validate_pde = NULL;
-static struct proc_dir_entry *manage_pde = NULL;
+static struct proc_dir_entry *firmware_flash_pde;
+static struct proc_dir_entry *firmware_update_pde;
+static struct proc_dir_entry *validate_pde;
+static struct proc_dir_entry *manage_pde;
/* Do simple sanity checks on the flash image. */
static int flash_list_valid(struct flash_block_list *flist)
@@ -501,7 +501,7 @@ static ssize_t validate_flash_read(struc
}
static ssize_t validate_flash_write(struct file *file, const char *buf,
- size_t count, loff_t *off)
+ size_t count, loff_t *off)
{
struct proc_dir_entry *dp = PDE(file->f_dentry->d_inode);
struct rtas_validate_flash_t *args_buf;
@@ -567,18 +567,18 @@ static int validate_flash_release(struct
return 0;
}
-static inline void remove_flash_pde(struct proc_dir_entry *dp)
+static void remove_flash_pde(struct proc_dir_entry *dp)
{
if (dp) {
if (dp->data != NULL)
kfree(dp->data);
- remove_proc_entry(dp->name, proc_ppc64.rtas);
+ remove_proc_entry(dp->name, NULL);
}
}
-static inline int initialize_flash_pde_data(const char *rtas_call_name,
- size_t buf_size,
- struct proc_dir_entry *dp)
+static int initialize_flash_pde_data(const char *rtas_call_name,
+ size_t buf_size,
+ struct proc_dir_entry *dp)
{
int *status;
int token;
@@ -591,7 +591,8 @@ static inline int initialize_flash_pde_d
memset(dp->data, 0, buf_size);
- /* This code assumes that the status int is the first member of the
+ /*
+ * This code assumes that the status int is the first member of the
* struct
*/
status = (int *) dp->data;
@@ -604,12 +605,12 @@ static inline int initialize_flash_pde_d
return 0;
}
-static inline struct proc_dir_entry * create_flash_pde(const char *filename,
- struct file_operations *fops)
+static struct proc_dir_entry *create_flash_pde(const char *filename,
+ struct file_operations *fops)
{
struct proc_dir_entry *ent = NULL;
- ent = create_proc_entry(filename, S_IRUSR | S_IWUSR, proc_ppc64.rtas);
+ ent = create_proc_entry(filename, S_IRUSR | S_IWUSR, NULL);
if (ent != NULL) {
ent->nlink = 1;
ent->proc_fops = fops;
@@ -644,50 +645,79 @@ int __init rtas_flash_init(void)
{
int rc;
- if (!proc_ppc64.rtas) {
- printk(KERN_WARNING "rtas proc dir does not already exist");
- return -ENOENT;
+ if (rtas_token("ibm,update-flash-64-and-reboot") ==
+ RTAS_UNKNOWN_SERVICE) {
+ printk(KERN_ERR "rtas_flash: no firmware flash support\n");
+ return 1;
}
- firmware_flash_pde = create_flash_pde(FIRMWARE_FLASH_NAME,
+ firmware_flash_pde = create_flash_pde("ppc64/rtas/"
+ FIRMWARE_FLASH_NAME,
&rtas_flash_operations);
+ if (firmware_flash_pde == NULL) {
+ rc = -ENOMEM;
+ goto cleanup;
+ }
+
rc = initialize_flash_pde_data("ibm,update-flash-64-and-reboot",
sizeof(struct rtas_update_flash_t),
firmware_flash_pde);
if (rc != 0)
- return rc;
-
- firmware_update_pde = create_flash_pde(FIRMWARE_UPDATE_NAME,
+ goto cleanup;
+
+ firmware_update_pde = create_flash_pde("ppc64/rtas/"
+ FIRMWARE_UPDATE_NAME,
&rtas_flash_operations);
+ if (firmware_update_pde == NULL) {
+ rc = -ENOMEM;
+ goto cleanup;
+ }
+
rc = initialize_flash_pde_data("ibm,update-flash-64-and-reboot",
sizeof(struct rtas_update_flash_t),
firmware_update_pde);
if (rc != 0)
- return rc;
-
- validate_pde = create_flash_pde(VALIDATE_FLASH_NAME,
+ goto cleanup;
+
+ validate_pde = create_flash_pde("ppc64/rtas/" VALIDATE_FLASH_NAME,
&validate_flash_operations);
+ if (validate_pde == NULL) {
+ rc = -ENOMEM;
+ goto cleanup;
+ }
+
rc = initialize_flash_pde_data("ibm,validate-flash-image",
sizeof(struct rtas_validate_flash_t),
validate_pde);
if (rc != 0)
- return rc;
-
- manage_pde = create_flash_pde(MANAGE_FLASH_NAME,
+ goto cleanup;
+
+ manage_pde = create_flash_pde("ppc64/rtas" MANAGE_FLASH_NAME,
&manage_flash_operations);
+ if (manage_pde == NULL) {
+ rc = -ENOMEM;
+ goto cleanup;
+ }
+
rc = initialize_flash_pde_data("ibm,manage-flash-image",
sizeof(struct rtas_manage_flash_t),
manage_pde);
if (rc != 0)
- return rc;
-
+ goto cleanup;
+
return 0;
+
+cleanup:
+ remove_flash_pde(firmware_flash_pde);
+ remove_flash_pde(firmware_update_pde);
+ remove_flash_pde(validate_pde);
+ remove_flash_pde(manage_pde);
+
+ return rc;
}
void __exit rtas_flash_cleanup(void)
{
- if (!proc_ppc64.rtas)
- return;
remove_flash_pde(firmware_flash_pde);
remove_flash_pde(firmware_update_pde);
remove_flash_pde(validate_pde);
--- diff/arch/ppc64/kernel/rtasd.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/rtasd.c 2004-03-16 11:27:47.939899688 +0000
@@ -26,7 +26,6 @@
#include <asm/prom.h>
#include <asm/nvram.h>
#include <asm/atomic.h>
-#include <asm/proc_fs.h>
#if 0
#define DEBUG(A...) printk(KERN_ERR A)
@@ -47,7 +46,6 @@ static unsigned int rtas_event_scan_rate
static unsigned int rtas_error_log_max;
static unsigned int rtas_error_log_buffer_max;
-extern spinlock_t proc_ppc64_lock;
extern volatile int no_more_logging;
volatile int error_log_cnt = 0;
@@ -445,20 +443,18 @@ static int __init rtas_init(void)
{
struct proc_dir_entry *entry;
- if (proc_ppc64.rtas == NULL) {
- proc_ppc64_init();
+ /* No RTAS, only warn if we are on a pSeries box */
+ if (rtas_token("event-scan") == RTAS_UNKNOWN_SERVICE) {
+ if (systemcfg->platform & PLATFORM_PSERIES);
+ printk(KERN_ERR "rtasd: no RTAS on system\n");
+ return 1;
}
- if (proc_ppc64.rtas == NULL) {
- printk(KERN_ERR "rtas_init: /proc/ppc64/rtas does not exist.");
- return -EIO;
- }
-
- entry = create_proc_entry("error_log", S_IRUSR, proc_ppc64.rtas);
+ entry = create_proc_entry("ppc64/error_log", S_IRUSR, NULL);
if (entry)
entry->proc_fops = &proc_rtas_log_operations;
else
- printk(KERN_ERR "Failed to create rtas/error_log proc entry\n");
+ printk(KERN_ERR "Failed to create error_log proc entry\n");
if (kernel_thread(rtasd, 0, CLONE_FS) < 0)
printk(KERN_ERR "Failed to start RTAS daemon\n");
--- diff/arch/ppc64/kernel/scanlog.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/scanlog.c 2004-03-16 11:27:47.942899232 +0000
@@ -28,7 +28,6 @@
#include <asm/uaccess.h>
#include <asm/rtas.h>
#include <asm/prom.h>
-#include <asm/proc_fs.h>
#define MODULE_VERS "1.0"
#define MODULE_NAME "scanlog"
@@ -212,16 +211,7 @@ int __init scanlog_init(void)
return -EIO;
}
- if (proc_ppc64.rtas == NULL) {
- proc_ppc64_init();
- }
-
- if (proc_ppc64.rtas == NULL) {
- printk(KERN_ERR "Failed to create /proc/rtas in scanlog_init\n");
- return -EIO;
- }
-
- ent = create_proc_entry("scan-log-dump", S_IRUSR, proc_ppc64.rtas);
+ ent = create_proc_entry("ppc64/rtas/scan-log-dump", S_IRUSR, NULL);
if (ent) {
ent->proc_fops = &scanlog_fops;
/* Ideally we could allocate a buffer < 4G */
--- diff/arch/ppc64/kernel/smp.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/smp.c 2004-03-16 11:27:47.943899080 +0000
@@ -796,3 +796,225 @@ static int __init topology_init(void)
return 0;
}
__initcall(topology_init);
+
+#ifdef CONFIG_SCHED_SMT
+#ifdef CONFIG_NUMA
+static struct sched_group sched_group_cpus[NR_CPUS];
+static struct sched_group sched_group_phys[NR_CPUS];
+static struct sched_group sched_group_nodes[MAX_NUMNODES];
+static DEFINE_PER_CPU(struct sched_domain, phys_domains);
+static DEFINE_PER_CPU(struct sched_domain, node_domains);
+__init void arch_init_sched_domains(void)
+{
+ int i;
+ struct sched_group *first_cpu = NULL, *last_cpu = NULL;
+
+ /* Set up domains */
+ for_each_cpu(i) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ struct sched_domain *phys_domain = &per_cpu(phys_domains, i);
+ struct sched_domain *node_domain = &per_cpu(node_domains, i);
+ int node = cpu_to_node(i);
+ cpumask_t nodemask = node_to_cpumask(node);
+ cpumask_t tmp1 = cpumask_of_cpu(i ^ 0x1);
+ cpumask_t tmp2 = cpumask_of_cpu(i);
+
+ *cpu_domain = SD_SIBLING_INIT;
+ cpus_or(cpu_domain->span, tmp1, tmp2);
+
+ *phys_domain = SD_CPU_INIT;
+ phys_domain->span = nodemask;
+
+ *node_domain = SD_NODE_INIT;
+ node_domain->span = cpu_possible_map;
+ }
+
+ /* Set up CPU (sibling) groups */
+ for_each_cpu(i) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ int j;
+ first_cpu = last_cpu = NULL;
+
+ if (i != first_cpu(cpu_domain->span)) {
+ cpu_sched_domain(i)->flags |= SD_FLAG_SHARE_CPUPOWER;
+ cpu_sched_domain(first_cpu(cpu_domain->span))->flags |=
+ SD_FLAG_SHARE_CPUPOWER;
+ continue;
+ }
+
+ for_each_cpu_mask(j, cpu_domain->span) {
+ struct sched_group *cpu = &sched_group_cpus[j];
+
+ cpus_clear(cpu->cpumask);
+ cpu_set(j, cpu->cpumask);
+ cpu->cpu_power = SCHED_LOAD_SCALE;
+
+ if (!first_cpu)
+ first_cpu = cpu;
+ if (last_cpu)
+ last_cpu->next = cpu;
+ last_cpu = cpu;
+ }
+ last_cpu->next = first_cpu;
+ }
+
+ for (i = 0; i < MAX_NUMNODES; i++) {
+ int j;
+ cpumask_t nodemask;
+ struct sched_group *node = &sched_group_nodes[i];
+ cpumask_t node_cpumask = node_to_cpumask(i);
+ cpus_and(nodemask, node_cpumask, cpu_online_map);
+
+ if (cpus_empty(nodemask))
+ continue;
+
+ first_cpu = last_cpu = NULL;
+ /* Set up physical groups */
+ for_each_cpu_mask(j, nodemask) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(j);
+ struct sched_group *cpu = &sched_group_phys[j];
+
+ if (j != first_cpu(cpu_domain->span))
+ continue;
+
+ cpu->cpumask = cpu_domain->span;
+ /*
+ * Make each extra sibling increase power by 10% of
+ * the basic CPU. This is very arbitrary.
+ */
+ cpu->cpu_power = SCHED_LOAD_SCALE + SCHED_LOAD_SCALE*(cpus_weight(cpu->cpumask)-1) / 10;
+ node->cpu_power += cpu->cpu_power;
+
+ if (!first_cpu)
+ first_cpu = cpu;
+ if (last_cpu)
+ last_cpu->next = cpu;
+ last_cpu = cpu;
+ }
+ last_cpu->next = first_cpu;
+ }
+
+ /* Set up nodes */
+ first_cpu = last_cpu = NULL;
+ for (i = 0; i < MAX_NUMNODES; i++) {
+ struct sched_group *cpu = &sched_group_nodes[i];
+ cpumask_t nodemask;
+ cpumask_t node_cpumask = node_to_cpumask(i);
+ cpus_and(nodemask, node_cpumask, cpu_possible_map);
+
+ if (cpus_empty(nodemask))
+ continue;
+
+ cpu->cpumask = nodemask;
+ /* ->cpu_power already setup */
+
+ if (!first_cpu)
+ first_cpu = cpu;
+ if (last_cpu)
+ last_cpu->next = cpu;
+ last_cpu = cpu;
+ }
+ last_cpu->next = first_cpu;
+
+ mb();
+ for_each_cpu(i) {
+ int node = cpu_to_node(i);
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ struct sched_domain *phys_domain = &per_cpu(phys_domains, i);
+ struct sched_domain *node_domain = &per_cpu(node_domains, i);
+ struct sched_group *cpu_group = &sched_group_cpus[i];
+ struct sched_group *phys_group = &sched_group_phys[first_cpu(cpu_domain->span)];
+ struct sched_group *node_group = &sched_group_nodes[node];
+
+ cpu_domain->parent = phys_domain;
+ phys_domain->parent = node_domain;
+
+ node_domain->groups = node_group;
+ phys_domain->groups = phys_group;
+ cpu_domain->groups = cpu_group;
+ }
+}
+#else /* CONFIG_NUMA */
+static struct sched_group sched_group_cpus[NR_CPUS];
+static struct sched_group sched_group_phys[NR_CPUS];
+static DEFINE_PER_CPU(struct sched_domain, phys_domains);
+__init void arch_init_sched_domains(void)
+{
+ int i;
+ struct sched_group *first_cpu = NULL, *last_cpu = NULL;
+
+ /* Set up domains */
+ for_each_cpu(i) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ struct sched_domain *phys_domain = &per_cpu(phys_domains, i);
+
+ *cpu_domain = SD_SIBLING_INIT;
+ cpu_domain->span = cpu_sibling_map[i];
+
+ *phys_domain = SD_CPU_INIT;
+ phys_domain->span = cpu_possible_map;
+ }
+
+ /* Set up CPU (sibling) groups */
+ for_each_cpu(i) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ int j;
+ first_cpu = last_cpu = NULL;
+
+ if (i != first_cpu(cpu_domain->span)) {
+ cpu_sched_domain(i)->flags |= SD_FLAG_SHARE_CPUPOWER;
+ cpu_sched_domain(first_cpu(cpu_domain->span))->flags |=
+ SD_FLAG_SHARE_CPUPOWER;
+ continue;
+ }
+
+ for_each_cpu_mask(j, cpu_domain->span) {
+ struct sched_group *cpu = &sched_group_cpus[j];
+
+ cpus_clear(cpu->cpumask);
+ cpu_set(j, cpu->cpumask);
+ cpu->cpu_power = SCHED_LOAD_SCALE;
+
+ if (!first_cpu)
+ first_cpu = cpu;
+ if (last_cpu)
+ last_cpu->next = cpu;
+ last_cpu = cpu;
+ }
+ last_cpu->next = first_cpu;
+ }
+
+ first_cpu = last_cpu = NULL;
+ /* Set up physical groups */
+ for_each_cpu(i) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ struct sched_group *cpu = &sched_group_phys[i];
+
+ if (i != first_cpu(cpu_domain->span))
+ continue;
+
+ cpu->cpumask = cpu_domain->span;
+ /* See SMT+NUMA setup for comment */
+ cpu->cpu_power = SCHED_LOAD_SCALE + SCHED_LOAD_SCALE*(cpus_weight(cpu->cpumask)-1) / 10;
+
+ if (!first_cpu)
+ first_cpu = cpu;
+ if (last_cpu)
+ last_cpu->next = cpu;
+ last_cpu = cpu;
+ }
+ last_cpu->next = first_cpu;
+
+ mb();
+ for_each_cpu(i) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ struct sched_domain *phys_domain = &per_cpu(phys_domains, i);
+ struct sched_group *cpu_group = &sched_group_cpus[i];
+ struct sched_group *phys_group = &sched_group_phys[first_cpu(cpu_domain->span)];
+ cpu_domain->parent = phys_domain;
+ phys_domain->groups = phys_group;
+ cpu_domain->groups = cpu_group;
+ }
+}
+#endif /* CONFIG_NUMA */
+#endif /* CONFIG_SCHED_SMT */
--- diff/arch/ppc64/kernel/sys_ppc32.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/sys_ppc32.c 2004-03-16 11:27:47.945898776 +0000
@@ -2577,17 +2577,6 @@ asmlinkage long sys32_umask(u32 mask)
return sys_umask((int)mask);
}
-
-/* Note: it is necessary to treat flags as an unsigned int,
- * with the corresponding cast to a signed int to insure that the
- * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode)
- * and the register representation of a signed int (msr in 64-bit mode) is performed.
- */
-asmlinkage long sys32_umount(char * name, u32 flags)
-{
- return sys_umount(name, (int)flags);
-}
-
struct __sysctl_args32 {
u32 name;
int nlen;
--- diff/arch/ppc64/kernel/vio.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/vio.c 2004-03-16 11:27:47.946898624 +0000
@@ -580,5 +580,4 @@ struct bus_type vio_bus_type = {
.match = vio_bus_match,
};
-EXPORT_SYMBOL(plpar_hcall_norets);
-EXPORT_SYMBOL(plpar_hcall_8arg_2ret);
+EXPORT_SYMBOL(vio_bus_type);
--- diff/arch/ppc64/kernel/viopath.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/viopath.c 2004-03-16 11:27:47.947898472 +0000
@@ -28,7 +28,6 @@
* Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*
*/
-#include <asm/uaccess.h>
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/errno.h>
@@ -37,9 +36,10 @@
#include <linux/proc_fs.h>
#include <linux/dma-mapping.h>
#include <linux/wait.h>
+#include <linux/seq_file.h>
-#include <asm/hardirq.h> /* for is_atomic */
-
+#include <asm/hardirq.h>
+#include <asm/uaccess.h>
#include <asm/iSeries/LparData.h>
#include <asm/iSeries/HvLpEvent.h>
#include <asm/iSeries/HvLpConfig.h>
@@ -184,21 +184,21 @@ static unsigned char e2a(unsigned char x
return ' ';
}
-/* Handle reads from the proc file system
- */
-static int proc_read(char *buf, char **start, off_t offset,
- int blen, int *eof, void *data)
+static int proc_viopath_show(struct seq_file *m, void *v)
{
+ char *buf;
+ dma_addr_t handle;
HvLpEvent_Rc hvrc;
DECLARE_MUTEX_LOCKED(Semaphore);
- dma_addr_t dmaa =
- dma_map_single(iSeries_vio_dev, buf, PAGE_SIZE, DMA_FROM_DEVICE);
- int len = PAGE_SIZE;
- if (len > blen)
- len = blen;
+ buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
+ if (!buf)
+ return 0;
+ memset(buf, 0, PAGE_SIZE);
+
+ handle = dma_map_single(iSeries_vio_dev, buf, PAGE_SIZE,
+ DMA_FROM_DEVICE);
- memset(buf, 0x00, len);
hvrc = HvCallEvent_signalLpEventFast(viopath_hostLp,
HvLpEvent_Type_VirtualIo,
viomajorsubtype_config | vioconfigget,
@@ -206,50 +206,54 @@ static int proc_read(char *buf, char **s
viopath_sourceinst(viopath_hostLp),
viopath_targetinst(viopath_hostLp),
(u64)(unsigned long)&Semaphore, VIOVERSION << 16,
- ((u64)dmaa) << 32, len, 0, 0);
+ ((u64)handle) << 32, PAGE_SIZE, 0, 0);
+
if (hvrc != HvLpEvent_Rc_Good)
- printk("viopath hv error on op %d\n", (int) hvrc);
+ printk("viopath hv error on op %d\n", (int)hvrc);
down(&Semaphore);
- dma_unmap_single(iSeries_vio_dev, dmaa, PAGE_SIZE, DMA_FROM_DEVICE);
+ dma_unmap_single(iSeries_vio_dev, handle, PAGE_SIZE, DMA_FROM_DEVICE);
+ kfree(buf);
+
+ buf[PAGE_SIZE] = '\0';
+ seq_printf(m, "%s", buf);
+
+ seq_printf(m, "SRLNBR=%c%c%c%c%c%c%c\n",
+ e2a(xItExtVpdPanel.mfgID[2]),
+ e2a(xItExtVpdPanel.mfgID[3]),
+ e2a(xItExtVpdPanel.systemSerial[1]),
+ e2a(xItExtVpdPanel.systemSerial[2]),
+ e2a(xItExtVpdPanel.systemSerial[3]),
+ e2a(xItExtVpdPanel.systemSerial[4]),
+ e2a(xItExtVpdPanel.systemSerial[5]));
- sprintf(buf + strlen(buf), "SRLNBR=");
- buf[strlen(buf)] = e2a(xItExtVpdPanel.mfgID[2]);
- buf[strlen(buf)] = e2a(xItExtVpdPanel.mfgID[3]);
- buf[strlen(buf)] = e2a(xItExtVpdPanel.systemSerial[1]);
- buf[strlen(buf)] = e2a(xItExtVpdPanel.systemSerial[2]);
- buf[strlen(buf)] = e2a(xItExtVpdPanel.systemSerial[3]);
- buf[strlen(buf)] = e2a(xItExtVpdPanel.systemSerial[4]);
- buf[strlen(buf)] = e2a(xItExtVpdPanel.systemSerial[5]);
- buf[strlen(buf)] = '\n';
- *eof = 1;
- return strlen(buf);
+ return 0;
}
-/* Handle writes to our proc file system
- */
-static int proc_write(struct file *file, const char *buffer,
- unsigned long count, void *data)
+static int proc_viopath_open(struct inode *inode, struct file *file)
{
- /* Doesn't do anything today!!!
- */
- return count;
+ return single_open(file, proc_viopath_show, NULL);
}
-/* setup our proc file system entries
- */
-static void vio_proc_init(struct proc_dir_entry *iSeries_proc)
+static struct file_operations proc_viopath_operations = {
+ .open = proc_viopath_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
+
+static int __init vio_proc_init(void)
{
- struct proc_dir_entry *ent;
- ent = create_proc_entry("config", S_IFREG | S_IRUSR, iSeries_proc);
- if (!ent)
- return;
- ent->nlink = 1;
- ent->data = NULL;
- ent->read_proc = proc_read;
- ent->write_proc = proc_write;
+ struct proc_dir_entry *e;
+
+ e = create_proc_entry("iSeries/config", 0, NULL);
+ if (e)
+ e->proc_fops = &proc_viopath_operations;
+
+ return 0;
}
+__initcall(vio_proc_init);
/* See if a given LP is active. Allow for invalid lps to be passed in
* and just return invalid
@@ -433,13 +437,8 @@ void vio_set_hostlp(void)
viopath_ourLp = HvLpConfig_getLpIndex();
viopath_hostLp = HvCallCfg_getHostingLpIndex(viopath_ourLp);
- /* If we have a valid hosting LP, create a proc file system entry
- * for config information
- */
- if (viopath_hostLp != HvLpIndexInvalid) {
- iSeries_proc_callback(&vio_proc_init);
+ if (viopath_hostLp != HvLpIndexInvalid)
vio_setHandler(viomajorsubtype_config, handleConfig);
- }
}
EXPORT_SYMBOL(vio_set_hostlp);
--- diff/arch/ppc64/kernel/xics.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/kernel/xics.c 2004-03-16 11:27:47.947898472 +0000
@@ -327,6 +327,8 @@ static void xics_mask_and_ack_irq(unsign
}
}
+extern unsigned int real_irq_to_virt_slowpath(unsigned int real_irq);
+
int xics_get_irq(struct pt_regs *regs)
{
unsigned int cpu = smp_processor_id();
@@ -349,6 +351,8 @@ int xics_get_irq(struct pt_regs *regs)
irq = -1;
} else {
irq = real_irq_to_virt(vec);
+ if (irq == NO_IRQ)
+ irq = real_irq_to_virt_slowpath(vec);
if (irq == NO_IRQ) {
printk(KERN_ERR "Interrupt 0x%x (real) is invalid,"
" disabling it.\n", vec);
--- diff/arch/ppc64/mm/hash_utils.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/mm/hash_utils.c 2004-03-16 11:27:47.948898320 +0000
@@ -39,7 +39,6 @@
#include <asm/system.h>
#include <asm/uaccess.h>
#include <asm/naca.h>
-#include <asm/pmc.h>
#include <asm/machdep.h>
#include <asm/lmb.h>
#include <asm/abs_addr.h>
--- diff/arch/ppc64/mm/hugetlbpage.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/mm/hugetlbpage.c 2004-03-16 11:27:47.949898168 +0000
@@ -40,7 +40,7 @@ static struct list_head hugepage_freelis
static void enqueue_huge_page(struct page *page)
{
- list_add(&page->list,
+ list_add(&page->lru,
&hugepage_freelists[page_zone(page)->zone_pgdat->node_id]);
}
@@ -63,8 +63,8 @@ static struct page *dequeue_huge_page(vo
}
if (!list_empty(&hugepage_freelists[nid])) {
- page = list_entry(hugepage_freelists[nid].next, struct page, list);
- list_del(&page->list);
+ page = list_entry(hugepage_freelists[nid].next, struct page, lru);
+ list_del(&page->lru);
}
if (largepage_roundrobin)
@@ -407,9 +407,8 @@ follow_huge_pmd(struct mm_struct *mm, un
static void free_huge_page(struct page *page)
{
BUG_ON(page_count(page));
- BUG_ON(page->mapping);
- INIT_LIST_HEAD(&page->list);
+ INIT_LIST_HEAD(&page->lru);
spin_lock(&htlbpage_lock);
enqueue_huge_page(page);
--- diff/arch/ppc64/mm/init.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/mm/init.c 2004-03-16 11:27:47.950898016 +0000
@@ -36,7 +36,6 @@
#include <linux/delay.h>
#include <linux/bootmem.h>
#include <linux/highmem.h>
-#include <linux/proc_fs.h>
#include <asm/pgalloc.h>
#include <asm/page.h>
@@ -76,8 +75,6 @@ pgd_t * ioremap_pgd = (pgd_t *)&ioremap_
unsigned long klimit = (unsigned long)_end;
-HPTE *Hash=0;
-unsigned long Hash_size=0;
unsigned long _SDR1=0;
unsigned long _ASR=0;
--- diff/arch/ppc64/xmon/xmon.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/ppc64/xmon/xmon.c 2004-03-16 11:27:47.951897864 +0000
@@ -56,9 +56,6 @@ static u_int bus_error_jmp[100];
#define setjmp xmon_setjmp
#define longjmp xmon_longjmp
-/* Max number of stack frames we are willing to produce on a backtrace. */
-#define MAXFRAMECOUNT 50
-
/* Breakpoint stuff */
struct bpt {
unsigned long address;
@@ -85,7 +82,6 @@ static void dump(void);
static void prdump(unsigned long, long);
static int ppc_inst_dump(unsigned long, long);
void print_address(unsigned long);
-static int getsp(void);
static void backtrace(struct pt_regs *);
static void excprint(struct pt_regs *);
static void prregs(struct pt_regs *);
@@ -897,93 +893,75 @@ const char *getvecname(unsigned long vec
return ret;
}
-static void
-backtrace(struct pt_regs *excp)
+/*
+ * Most of our exceptions are in the form:
+ * bl handler
+ * b .ret_from_exception
+ * and this currently fails to catch them.
+ */
+static inline int exception_frame(unsigned long ip)
{
- unsigned long sp;
- unsigned long lr;
- unsigned long stack[3];
+ extern void *ret_from_syscall_1, *ret_from_syscall_2, *ret_from_except;
+
+ if ((ip == (unsigned long)ret_from_syscall_1) ||
+ (ip == (unsigned long)ret_from_syscall_2) ||
+ (ip == (unsigned long)ret_from_except))
+ return 1;
+
+ return 0;
+}
+
+static int xmon_depth_to_print = 64;
+
+static void xmon_show_stack(unsigned long sp)
+{
+ unsigned long ip;
+ unsigned long newsp;
+ int count = 0;
struct pt_regs regs;
- int framecount;
- char *funcname;
- /* declare these as raw ptrs so we don't get func descriptors */
- extern void *ret_from_except, *ret_from_syscall_1;
- if (excp != NULL) {
- lr = excp->link;
- sp = excp->gpr[1];
- } else {
- /* Use care not to call any function before this point
- so the saved lr has a chance of being good. */
- asm volatile ("mflr %0" : "=r" (lr) :);
- sp = getsp();
- }
- scanhex(&sp);
- scannl();
- for (framecount = 0;
- sp != 0 && framecount < MAXFRAMECOUNT;
- sp = stack[0], framecount++) {
- if (mread(sp, stack, sizeof(stack)) != sizeof(stack))
- break;
-#if 0
- if (lr != 0) {
- stack[2] = lr; /* fake out the first saved lr. It may not be saved yet. */
- lr = 0;
+ do {
+ if (sp < PAGE_OFFSET) {
+ printf("SP in userspace\n");
+ break;
}
-#endif
- printf("%.16lx %.16lx", sp, stack[2]);
- /* TAI -- for now only the ones cast to unsigned long will match.
- * Need to test the rest...
- */
- if ((stack[2] == (unsigned long)ret_from_except &&
- (funcname = "ret_from_except"))
- || (stack[2] == (unsigned long)ret_from_syscall_1 &&
- (funcname = "ret_from_syscall_1"))
-#if 0
- || stack[2] == (unsigned) &ret_from_syscall_2
- || stack[2] == (unsigned) &do_signal_ret
-#endif
- ) {
- printf(" %s\n", funcname);
+
+ if (!mread((sp + 16), &ip, sizeof(unsigned long)))
+ break;
+
+ printf("[%016lx] [%016lx] ", sp, ip);
+ xmon_print_symbol("%s\n", ip);
+
+ if (exception_frame(ip)) {
if (mread(sp+112, ®s, sizeof(regs)) != sizeof(regs))
break;
- printf("exception: %lx %s regs %lx\n", regs.trap, getvecname(regs.trap), sp+112);
- printf(" %.16lx", regs.nip);
- if (regs.nip & 0xffffffff00000000UL)
- xmon_print_symbol(" %s", regs.nip);
- printf("\n");
- if (regs.gpr[1] < sp) {
- printf("<Stack drops into userspace %.16lx>\n", regs.gpr[1]);
- break;
- }
- sp = regs.gpr[1];
- if (mread(sp, stack, sizeof(stack)) != sizeof(stack))
- break;
- } else {
- if (stack[2])
- xmon_print_symbol(" %s", stack[2]);
- printf("\n");
- }
- if (stack[0] && stack[0] <= sp) {
- if ((stack[0] & 0xffffffff00000000UL) == 0)
- printf("<Stack drops into 32-bit userspace %.16lx>\n", stack[0]);
- else
- printf("<Corrupt stack. Next backchain is %.16lx>\n", stack[0]);
- break;
+ printf(" exception: %lx %s regs %lx\n", regs.trap,
+ getvecname(regs.trap), sp+112);
}
- }
- if (framecount >= MAXFRAMECOUNT)
- printf("<Punt. Too many stack frames>\n");
+
+ if (!mread(sp, &newsp, sizeof(unsigned long)))
+ break;
+ if (newsp < sp)
+ break;
+
+ sp = newsp;
+ } while (count++ < xmon_depth_to_print);
}
-int
-getsp()
+static void backtrace(struct pt_regs *excp)
{
- int x;
+ unsigned long sp;
+
+ if (excp == NULL)
+ sp = __get_SP();
+ else
+ sp = excp->gpr[1];
+
+ scanhex(&sp);
+ scannl();
- asm("mr %0,1" : "=r" (x) :);
- return x;
+ xmon_show_stack(sp);
}
spinlock_t exception_print_lock = SPIN_LOCK_UNLOCKED;
--- diff/arch/s390/appldata/appldata_mem.c 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/s390/appldata/appldata_mem.c 2004-03-16 11:27:47.954897408 +0000
@@ -27,6 +27,15 @@
/*
* Memory data
+ *
+ * This is accessed as binary data by z/VM. If changes to it can't be avoided,
+ * the structure version (product ID, see appldata_base.c) needs to be changed
+ * as well and all documentation and z/VM applications using it must be
+ * updated.
+ *
+ * The record layout is documented in the Linux for zSeries Device Drivers
+ * book:
+ * http://oss.software.ibm.com/developerworks/opensource/linux390/index.shtml
*/
struct appldata_mem_data {
u64 timestamp;
@@ -54,9 +63,7 @@ struct appldata_mem_data {
u64 freeswap; /* free swap space */
// New in 2.6 -->
- u64 pgalloc_high; /* page allocations */
- u64 pgalloc_normal;
- u64 pgalloc_dma;
+ u64 pgalloc; /* page allocations */
u64 pgfault; /* page faults (major+minor) */
u64 pgmajfault; /* page faults (major only) */
// <-- New in 2.6
@@ -71,9 +78,7 @@ static inline void appldata_debug_print(
P_DEBUG("pgpgout = %8lu KB\n", mem_data->pgpgout);
P_DEBUG("pswpin = %8lu Pages\n", mem_data->pswpin);
P_DEBUG("pswpout = %8lu Pages\n", mem_data->pswpout);
- P_DEBUG("pgalloc_high = %8lu \n", mem_data->pgalloc_high);
- P_DEBUG("pgalloc_normal = %8lu \n", mem_data->pgalloc_normal);
- P_DEBUG("pgalloc_dma = %8lu \n", mem_data->pgalloc_dma);
+ P_DEBUG("pgalloc = %8lu \n", mem_data->pgalloc);
P_DEBUG("pgfault = %8lu \n", mem_data->pgfault);
P_DEBUG("pgmajfault = %8lu \n", mem_data->pgmajfault);
P_DEBUG("sharedram = %8lu KB\n", mem_data->sharedram);
@@ -109,15 +114,11 @@ static void appldata_get_mem_data(void *
mem_data->pgpgout = ps.pgpgout >> 1;
mem_data->pswpin = ps.pswpin;
mem_data->pswpout = ps.pswpout;
- mem_data->pgalloc_high = ps.pgalloc_high;
- mem_data->pgalloc_normal = ps.pgalloc_normal;
- mem_data->pgalloc_dma = ps.pgalloc_dma;
+ mem_data->pgalloc = ps.pgalloc_high + ps.pgalloc_normal +
+ ps.pgalloc_dma;
mem_data->pgfault = ps.pgfault;
mem_data->pgmajfault = ps.pgmajfault;
-P_DEBUG("pgalloc_high = %lu, pgalloc_normal = %lu, pgalloc_dma = %lu, pgfree = %lu\n",
- ps.pgalloc_high, ps.pgalloc_normal, ps.pgalloc_dma, ps.pgfree);
-
si_meminfo(&val);
mem_data->sharedram = val.sharedram;
mem_data->totalram = P2K(val.totalram);
--- diff/arch/s390/appldata/appldata_net_sum.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/s390/appldata/appldata_net_sum.c 2004-03-16 11:27:47.954897408 +0000
@@ -26,6 +26,14 @@
/*
* Network data
+ *
+ * This is accessed as binary data by z/VM. If changes to it can't be avoided,
+ * the structure version (product ID, see appldata_base.c) needs to be changed
+ * as well and all documentation and z/VM applications using it must be updated.
+ *
+ * The record layout is documented in the Linux for zSeries Device Drivers
+ * book:
+ * http://oss.software.ibm.com/developerworks/opensource/linux390/index.shtml
*/
struct appldata_net_sum_data {
u64 timestamp;
--- diff/arch/s390/appldata/appldata_os.c 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/s390/appldata/appldata_os.c 2004-03-16 11:27:47.955897256 +0000
@@ -28,6 +28,15 @@
/*
* OS data
+ *
+ * This is accessed as binary data by z/VM. If changes to it can't be avoided,
+ * the structure version (product ID, see appldata_base.c) needs to be changed
+ * as well and all documentation and z/VM applications using it must be
+ * updated.
+ *
+ * The record layout is documented in the Linux for zSeries Device Drivers
+ * book:
+ * http://oss.software.ibm.com/developerworks/opensource/linux390/index.shtml
*/
struct appldata_os_per_cpu {
u32 per_cpu_user; /* timer ticks spent in user mode */
--- diff/arch/s390/kernel/compat_wrapper.S 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/s390/kernel/compat_wrapper.S 2004-03-16 11:27:47.956897104 +0000
@@ -1338,3 +1338,13 @@ sys32_io_cancel_wrapper:
llgtr %r3,%r3 # struct iocb *
llgtr %r4,%r4 # struct io_event *
jg sys_io_cancel
+
+ .globl sys32_remap_file_pages_wrapper
+sys32_remap_file_pages_wrapper:
+ llgfr %r2,%r2 # unsigned long
+ llgfr %r3,%r3 # unsigned long
+ llgfr %r4,%r4 # unsigned long
+ llgfr %r5,%r5 # unsigned long
+ llgfr %r6,%r6 # unsigned long
+ jg sys_remap_file_pages
+
--- diff/arch/s390/kernel/syscalls.S 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/s390/kernel/syscalls.S 2004-03-16 11:27:47.957896952 +0000
@@ -273,3 +273,4 @@ SYSCALL(sys_clock_getres,sys_clock_getre
SYSCALL(sys_clock_nanosleep,sys_clock_nanosleep,sys32_clock_nanosleep_wrapper)
NI_SYSCALL /* reserved for vserver */
SYSCALL(s390_fadvise64_64,sys_ni_syscall,sys32_fadvise64_64_wrapper)
+SYSCALL(sys_remap_file_pages,sys_remap_file_pages,sys32_remap_file_pages_wrapper)
--- diff/arch/s390/mm/fault.c 2004-02-09 10:36:09.000000000 +0000
+++ source/arch/s390/mm/fault.c 2004-03-16 11:27:47.958896800 +0000
@@ -256,6 +256,8 @@ survive:
goto do_sigbus;
case VM_FAULT_OOM:
goto out_of_memory;
+ case VM_FAULT_SIGSEGV:
+ goto bad_area;
default:
BUG();
}
--- diff/arch/sparc64/Kconfig 2004-03-11 10:20:22.000000000 +0000
+++ source/arch/sparc64/Kconfig 2004-03-16 11:27:47.959896648 +0000
@@ -687,12 +687,19 @@ config DEBUG_BOOTMEM
depends on DEBUG_KERNEL
bool "Debug BOOTMEM initialization"
+config LOCKMETER
+ bool "Kernel lock metering"
+ depends on SMP && !PREEMPT
+ help
+ Say Y to enable kernel lock metering, which adds overhead to SMP locks,
+ but allows you to see various statistics using the lockstat command.
+
# We have a custom atomic_dec_and_lock() implementation but it's not
# compatible with spinlock debugging so we need to fall back on
# the generic version in that case.
config HAVE_DEC_LOCK
bool
- depends on SMP && !DEBUG_SPINLOCK
+ depends on SMP && !DEBUG_SPINLOCK && !LOCKMETER
default y
config MCOUNT
--- diff/arch/sparc64/kernel/init_task.c 2003-10-09 08:47:33.000000000 +0000
+++ source/arch/sparc64/kernel/init_task.c 2004-03-16 11:27:47.959896648 +0000
@@ -24,14 +24,6 @@ __asm__ (".text");
union thread_union init_thread_union = { INIT_THREAD_INFO(init_task) };
/*
- * This is to make the init_thread+stack be the right size for >8k pagesize.
- * The definition of thread_union in sched.h makes it 16k wide.
- */
-#if PAGE_SHIFT != 13
-char init_task_stack[THREAD_SIZE - INIT_THREAD_SIZE] = { 0 };
-#endif
-
-/*
* Initial task structure.
*
* All other task structs will be allocated on slabs in fork.c
--- diff/arch/sparc64/lib/rwlock.S 2003-11-25 15:24:57.000000000 +0000
+++ source/arch/sparc64/lib/rwlock.S 2004-03-16 11:27:47.960896496 +0000
@@ -85,5 +85,20 @@ __write_trylock_succeed:
__write_trylock_fail:
retl
mov 0, %o0
+
+ .globl __read_trylock
+__read_trylock: /* %o0 = lock_ptr */
+ ldsw [%o0], %g5
+ brlz,pn %g5, 100f
+ add %g5, 1, %g7
+ cas [%o0], %g5, %g7
+ cmp %g5, %g7
+ bne,pn %icc, __read_trylock
+ membar #StoreLoad | #StoreStore
+ retl
+ mov 1, %o0
+100: retl
+ mov 0, %o0
+
rwlock_impl_end:
--- diff/arch/sparc64/mm/hugetlbpage.c 2004-01-19 10:22:55.000000000 +0000
+++ source/arch/sparc64/mm/hugetlbpage.c 2004-03-16 11:27:47.960896496 +0000
@@ -29,7 +29,7 @@ static spinlock_t htlbpage_lock = SPIN_L
static void enqueue_huge_page(struct page *page)
{
- list_add(&page->list,
+ list_add(&page->lru,
&hugepage_freelists[page_zone(page)->zone_pgdat->node_id]);
}
@@ -46,8 +46,8 @@ static struct page *dequeue_huge_page(vo
if (nid >= 0 && nid < MAX_NUMNODES &&
!list_empty(&hugepage_freelists[nid])) {
page = list_entry(hugepage_freelists[nid].next,
- struct page, list);
- list_del(&page->list);
+ struct page, lru);
+ list_del(&page->lru);
}
return page;
}
@@ -248,9 +248,8 @@ struct page *follow_huge_pmd(struct mm_s
static void free_huge_page(struct page *page)
{
BUG_ON(page_count(page));
- BUG_ON(page->mapping);
- INIT_LIST_HEAD(&page->list);
+ INIT_LIST_HEAD(&page->lru);
spin_lock(&htlbpage_lock);
enqueue_huge_page(page);
@@ -384,19 +383,19 @@ static int try_to_free_low(int count)
/* all lowmem is on node 0 */
list_for_each(p, &hugepage_freelists[0]) {
if (map) {
- list_del(&map->list);
+ list_del(&map->lru);
update_and_free_page(map);
htlbpagemem--;
map = NULL;
if (++count == 0)
break;
}
- page = list_entry(p, struct page, list);
+ page = list_entry(p, struct page, lru);
if (!PageHighMem(page))
map = page;
}
if (map) {
- list_del(&map->list);
+ list_del(&map->lru);
update_and_free_page(map);
htlbpagemem--;
count++;
--- diff/arch/x86_64/Kconfig 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/x86_64/Kconfig 2004-03-16 11:27:47.962896192 +0000
@@ -462,6 +462,7 @@ config INIT_DEBUG
config DEBUG_INFO
bool "Compile the kernel with debug info"
depends on DEBUG_KERNEL
+ default n
help
If you say Y here the resulting kernel image will include
debugging info resulting in a larger kernel image.
@@ -493,9 +494,8 @@ config IOMMU_LEAK
help
Add a simple leak tracer to the IOMMU code. This is useful when you
are debugging a buggy device driver that leaks IOMMU mappings.
-
-#config X86_REMOTE_DEBUG
-# bool "kgdb debugging stub"
+
+source "arch/x86_64/Kconfig.kgdb"
endmenu
--- diff/arch/x86_64/ia32/ia32entry.S 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/x86_64/ia32/ia32entry.S 2004-03-16 11:27:47.961896344 +0000
@@ -78,8 +78,8 @@ ENTRY(ia32_sysenter_target)
.quad 1b,ia32_badarg
.previous
GET_THREAD_INFO(%r10)
- bt $TIF_SYSCALL_TRACE,threadinfo_flags(%r10)
- jc sysenter_tracesys
+ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),threadinfo_flags(%r10)
+ jnz sysenter_tracesys
sysenter_do_call:
cmpl $(IA32_NR_syscalls),%eax
jae ia32_badsys
@@ -106,7 +106,7 @@ sysenter_tracesys:
CLEAR_RREGS
movq $-ENOSYS,RAX(%rsp) /* really needed? */
movq %rsp,%rdi /* &pt_regs -> arg1 */
- call syscall_trace
+ call syscall_trace_enter
LOAD_ARGS ARGOFFSET /* reload args from stack in case ptrace changed it */
RESTORE_REST
movl %ebp, %ebp
@@ -163,8 +163,8 @@ ENTRY(ia32_cstar_target)
.quad 1b,ia32_badarg
.previous
GET_THREAD_INFO(%r10)
- bt $TIF_SYSCALL_TRACE,threadinfo_flags(%r10)
- jc cstar_tracesys
+ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),threadinfo_flags(%r10)
+ jnz cstar_tracesys
cstar_do_call:
cmpl $IA32_NR_syscalls,%eax
jae ia32_badsys
@@ -187,7 +187,7 @@ cstar_tracesys:
CLEAR_RREGS
movq $-ENOSYS,RAX(%rsp) /* really needed? */
movq %rsp,%rdi /* &pt_regs -> arg1 */
- call syscall_trace
+ call syscall_trace_enter
LOAD_ARGS ARGOFFSET /* reload args from stack in case ptrace changed it */
RESTORE_REST
movl RSP-ARGOFFSET(%rsp), %r8d
@@ -236,8 +236,8 @@ ENTRY(ia32_syscall)
this could be a problem. */
SAVE_ARGS 0,0,1
GET_THREAD_INFO(%r10)
- bt $TIF_SYSCALL_TRACE,threadinfo_flags(%r10)
- jc ia32_tracesys
+ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),threadinfo_flags(%r10)
+ jnz ia32_tracesys
ia32_do_syscall:
cmpl $(IA32_NR_syscalls),%eax
jae ia32_badsys
@@ -251,7 +251,7 @@ ia32_tracesys:
SAVE_REST
movq $-ENOSYS,RAX(%rsp) /* really needed? */
movq %rsp,%rdi /* &pt_regs -> arg1 */
- call syscall_trace
+ call syscall_trace_enter
LOAD_ARGS ARGOFFSET /* reload args from stack in case ptrace changed it */
RESTORE_REST
jmp ia32_do_syscall
@@ -562,7 +562,7 @@ ia32_sys_call_table:
.quad sys_epoll_create
.quad sys_epoll_ctl
.quad sys_epoll_wait
- .quad sys_remap_file_pages
+ .quad old_remap_file_pages
.quad sys_set_tid_address
.quad sys32_timer_create
.quad compat_timer_settime
--- diff/arch/x86_64/kernel/Makefile 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/x86_64/kernel/Makefile 2004-03-16 11:27:47.970894976 +0000
@@ -26,6 +26,7 @@ obj-$(CONFIG_DUMMY_IOMMU) += pci-nommu.o
obj-$(CONFIG_SWIOTLB) += swiotlb.o
obj-$(CONFIG_MODULES) += module.o
+obj-$(CONFIG_KGDB) += kgdb_stub.o
obj-y += topology.o
--- diff/arch/x86_64/kernel/entry.S 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/x86_64/kernel/entry.S 2004-03-16 11:27:47.963896040 +0000
@@ -131,8 +131,8 @@ ENTRY(ret_from_fork)
CFI_DEFAULT_STACK
call schedule_tail
GET_THREAD_INFO(%rcx)
- bt $TIF_SYSCALL_TRACE,threadinfo_flags(%rcx)
- jc rff_trace
+ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),threadinfo_flags(%rcx)
+ jnz rff_trace
rff_action:
RESTORE_REST
testl $3,CS-ARGOFFSET(%rsp) # from kernel_thread?
@@ -143,7 +143,7 @@ rff_action:
jmp ret_from_sys_call
rff_trace:
movq %rsp,%rdi
- call syscall_trace
+ call syscall_trace_leave
GET_THREAD_INFO(%rcx)
jmp rff_action
CFI_ENDPROC
@@ -185,8 +185,8 @@ ENTRY(system_call)
movq %rax,ORIG_RAX-ARGOFFSET(%rsp)
movq %rcx,RIP-ARGOFFSET(%rsp)
GET_THREAD_INFO(%rcx)
- bt $TIF_SYSCALL_TRACE,threadinfo_flags(%rcx)
- jc tracesys
+ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),threadinfo_flags(%rcx)
+ jnz tracesys
cmpq $__NR_syscall_max,%rax
ja badsys
movq %r10,%rcx
@@ -244,7 +244,7 @@ tracesys:
movq $-ENOSYS,RAX(%rsp)
FIXUP_TOP_OF_STACK %rdi
movq %rsp,%rdi
- call syscall_trace
+ call syscall_trace_enter
LOAD_ARGS ARGOFFSET /* reload args from stack in case ptrace changed it */
RESTORE_REST
cmpq $__NR_syscall_max,%rax
@@ -254,7 +254,7 @@ tracesys:
movq %rax,RAX-ARGOFFSET(%rsp)
1: SAVE_REST
movq %rsp,%rdi
- call syscall_trace
+ call syscall_trace_leave
RESTORE_TOP_OF_STACK %rbx
RESTORE_REST
jmp ret_from_sys_call
@@ -297,13 +297,14 @@ int_very_careful:
sti
SAVE_REST
/* Check for syscall exit trace */
- bt $TIF_SYSCALL_TRACE,%edx
- jnc int_signal
+ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),%edx
+ jz int_signal
pushq %rdi
leaq 8(%rsp),%rdi # &ptregs -> arg1
- call syscall_trace
+ call syscall_trace_leave
popq %rdi
btr $TIF_SYSCALL_TRACE,%edi
+ btr $TIF_SYSCALL_AUDIT,%edi
jmp int_restore_rest
int_signal:
--- diff/arch/x86_64/kernel/irq.c 2004-03-11 10:20:23.000000000 +0000
+++ source/arch/x86_64/kernel/irq.c 2004-03-16 11:27:47.964895888 +0000
@@ -405,6 +405,9 @@ out:
spin_unlock(&desc->lock);
irq_exit();
+
+ kgdb_process_breakpoint();
+
return 1;
}
--- diff/arch/x86_64/kernel/ptrace.c 2003-06-09 13:18:18.000000000 +0000
+++ source/arch/x86_64/kernel/ptrace.c 2004-03-16 11:27:47.970894976 +0000
@@ -16,6 +16,7 @@
#include <linux/ptrace.h>
#include <linux/user.h>
#include <linux/security.h>
+#include <linux/audit.h>
#include <asm/uaccess.h>
#include <asm/pgtable.h>
@@ -486,7 +487,7 @@ out:
return ret;
}
-asmlinkage void syscall_trace(struct pt_regs *regs)
+static void syscall_trace(struct pt_regs *regs)
{
#if 0
@@ -513,3 +514,23 @@ asmlinkage void syscall_trace(struct pt_
current->exit_code = 0;
}
}
+
+asmlinkage void syscall_trace_enter(struct pt_regs *regs)
+{
+ if (unlikely(current->audit_context))
+ audit_syscall_entry(current, regs->orig_rax, regs->rdi);
+
+ if (test_thread_flag(TIF_SYSCALL_TRACE)
+ && (current->ptrace & PT_PTRACED))
+ syscall_trace(regs);
+}
+
+asmlinkage void syscall_trace_leave(struct pt_regs *regs)
+{
+ if (unlikely(current->audit_context))
+ audit_syscall_exit(current);
+
+ if (test_thread_flag(TIF_SYSCALL_TRACE)
+ && (current->ptrace & PT_PTRACED))
+ syscall_trace(regs);
+}
--- diff/arch/x86_64/kernel/smp.c 2003-11-25 15:24:57.000000000 +0000
+++ source/arch/x86_64/kernel/smp.c 2004-03-16 11:27:47.971894824 +0000
@@ -362,6 +362,18 @@ void smp_send_reschedule(int cpu)
send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR);
}
+#ifdef CONFIG_KGDB
+/*
+ * By using the NMI code instead of a vector we just sneak thru the
+ * word generator coming out with just what we want. AND it does
+ * not matter if clustered_apic_mode is set or not.
+ */
+void smp_send_nmi_allbutself(void)
+{
+ send_IPI_allbutself(APIC_DM_NMI);
+}
+#endif
+
/*
* Structure and data for smp_call_function(). This is designed to minimise
* static memory requirements. It also looks cleaner.
--- diff/arch/x86_64/kernel/traps.c 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/x86_64/kernel/traps.c 2004-03-16 11:27:47.971894824 +0000
@@ -45,6 +45,9 @@
#include <asm/proto.h>
#include <linux/irq.h>
+#ifdef CONFIG_KGDB
+#include <asm/kgdb.h>
+#endif
extern struct gate_struct idt_table[256];
--- diff/arch/x86_64/lib/Makefile 2004-03-16 11:24:01.000000000 +0000
+++ source/arch/x86_64/lib/Makefile 2004-03-16 11:27:47.973894520 +0000
@@ -10,3 +10,4 @@ lib-y := csum-partial.o csum-copy.o csum
lib-y += memcpy.o memmove.o memset.o copy_user.o
lib-$(CONFIG_HAVE_DEC_LOCK) += dec_and_lock.o
+lib-$(CONFIG_KGDB) += kgdb_serial.o
--- diff/drivers/block/Kconfig.iosched 2003-10-09 08:47:16.000000000 +0000
+++ source/drivers/block/Kconfig.iosched 2004-03-16 11:27:48.005889656 +0000
@@ -27,3 +27,10 @@ config IOSCHED_DEADLINE
a disk at any one time, its behaviour is almost identical to the
anticipatory I/O scheduler and so is a good choice.
+config IOSCHED_CFQ
+ bool "CFQ I/O scheduler" if EMBEDDED
+ default y
+ ---help---
+ The CFQ I/O scheduler tries to distribute bandwidth equally
+ among all processes in the system. It should provide a fair
+ working environment, suitable for desktop systems.
--- diff/drivers/block/Makefile 2004-03-16 11:24:01.000000000 +0000
+++ source/drivers/block/Makefile 2004-03-16 11:27:48.009889048 +0000
@@ -18,6 +18,7 @@ obj-y := elevator.o ll_rw_blk.o ioctl.o
obj-$(CONFIG_IOSCHED_NOOP) += noop-iosched.o
obj-$(CONFIG_IOSCHED_AS) += as-iosched.o
obj-$(CONFIG_IOSCHED_DEADLINE) += deadline-iosched.o
+obj-$(CONFIG_IOSCHED_CFQ) += cfq-iosched.o
obj-$(CONFIG_MAC_FLOPPY) += swim3.o
obj-$(CONFIG_BLK_DEV_FD) += floppy.o
obj-$(CONFIG_BLK_DEV_FD98) += floppy98.o
--- diff/drivers/block/genhd.c 2004-03-11 10:20:23.000000000 +0000
+++ source/drivers/block/genhd.c 2004-03-16 11:27:48.005889656 +0000
@@ -184,7 +184,7 @@ static int exact_lock(dev_t dev, void *d
}
/**
- * add_gendisk - add partitioning information to kernel list
+ * add_disk - add partitioning information to kernel list
* @disk: per-device partitioning information
*
* This function registers the partitioning information in @disk
--- diff/drivers/block/ll_rw_blk.c 2004-03-16 11:24:01.000000000 +0000
+++ source/drivers/block/ll_rw_blk.c 2004-03-16 11:27:48.007889352 +0000
@@ -27,6 +27,7 @@
#include <linux/completion.h>
#include <linux/slab.h>
#include <linux/swap.h>
+#include <linux/writeback.h>
/*
* for max sense size
@@ -41,13 +42,10 @@ static void blk_unplug_timeout(unsigned
*/
static kmem_cache_t *request_cachep;
-/*
- * plug management
- */
-static LIST_HEAD(blk_plug_list);
-static spinlock_t blk_plug_lock __cacheline_aligned_in_smp = SPIN_LOCK_UNLOCKED;
-
-static wait_queue_head_t congestion_wqh[2];
+static wait_queue_head_t congestion_wqh[2] = {
+ __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]),
+ __WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1])
+ };
/*
* Controlling structure to kblockd
@@ -233,9 +231,13 @@ void blk_queue_make_request(request_queu
blk_queue_dma_alignment(q, 511);
q->unplug_thresh = 4; /* hmm */
+#if 0
q->unplug_delay = (3 * HZ) / 1000; /* 3 milliseconds */
if (q->unplug_delay == 0)
q->unplug_delay = 1;
+#else
+ q->unplug_delay = HZ;
+#endif
INIT_WORK(&q->unplug_work, blk_unplug_work, q);
@@ -247,8 +249,6 @@ void blk_queue_make_request(request_queu
*/
blk_queue_bounce_limit(q, BLK_BOUNCE_HIGH);
- INIT_LIST_HEAD(&q->plug_list);
-
blk_queue_activity_fn(q, NULL, NULL);
}
@@ -527,10 +527,10 @@ init_tag_map(request_queue_t *q, struct
{
int bits, i;
- if (depth > q->nr_requests * 2) {
- depth = q->nr_requests * 2;
- printk(KERN_ERR "%s: adjusted depth to %d\n",
- __FUNCTION__, depth);
+ if (depth > q->nr_requests / 2) {
+ q->nr_requests = depth * 2;
+ printk(KERN_INFO "%s: large TCQ depth: adjusted nr_requests "
+ "to %lu\n", __FUNCTION__, q->nr_requests);
}
tags->tag_index = kmalloc(depth * sizeof(struct request *), GFP_ATOMIC);
@@ -1100,13 +1100,11 @@ void blk_plug_device(request_queue_t *q)
* don't plug a stopped queue, it must be paired with blk_start_queue()
* which will restart the queueing
*/
- if (!blk_queue_plugged(q)
- && !test_bit(QUEUE_FLAG_STOPPED, &q->queue_flags)) {
- spin_lock(&blk_plug_lock);
- list_add_tail(&q->plug_list, &blk_plug_list);
+ if (test_bit(QUEUE_FLAG_STOPPED, &q->queue_flags))
+ return;
+
+ if (!test_and_set_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags))
mod_timer(&q->unplug_timer, jiffies + q->unplug_delay);
- spin_unlock(&blk_plug_lock);
- }
}
EXPORT_SYMBOL(blk_plug_device);
@@ -1118,15 +1116,12 @@ EXPORT_SYMBOL(blk_plug_device);
int blk_remove_plug(request_queue_t *q)
{
WARN_ON(!irqs_disabled());
- if (blk_queue_plugged(q)) {
- spin_lock(&blk_plug_lock);
- list_del_init(&q->plug_list);
- del_timer(&q->unplug_timer);
- spin_unlock(&blk_plug_lock);
- return 1;
- }
- return 0;
+ if (!test_and_clear_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags))
+ return 0;
+
+ del_timer(&q->unplug_timer);
+ return 1;
}
EXPORT_SYMBOL(blk_remove_plug);
@@ -1157,14 +1152,11 @@ static inline void __generic_unplug_devi
* Linux uses plugging to build bigger requests queues before letting
* the device have at them. If a queue is plugged, the I/O scheduler
* is still adding and merging requests on the queue. Once the queue
- * gets unplugged (either by manually calling this function, or by
- * calling blk_run_queues()), the request_fn defined for the
- * queue is invoked and transfers started.
+ * gets unplugged, the request_fn defined for the queue is invoked and
+ * transfers started.
**/
-void generic_unplug_device(void *data)
+void generic_unplug_device(request_queue_t *q)
{
- request_queue_t *q = data;
-
spin_lock_irq(q->queue_lock);
__generic_unplug_device(q);
spin_unlock_irq(q->queue_lock);
@@ -1172,9 +1164,23 @@ void generic_unplug_device(void *data)
EXPORT_SYMBOL(generic_unplug_device);
+static inline void blk_backing_dev_unplug(struct backing_dev_info *bdi)
+{
+ request_queue_t *q = bdi->unplug_io_data;
+
+ /*
+ * devices don't necessarily have an ->unplug_fn defined
+ */
+ if (q->unplug_fn)
+ q->unplug_fn(q);
+}
+
+EXPORT_SYMBOL(blk_backing_dev_unplug);
+
static void blk_unplug_work(void *data)
{
request_queue_t *q = data;
+
q->unplug_fn(q);
}
@@ -1252,42 +1258,6 @@ void blk_run_queue(struct request_queue
EXPORT_SYMBOL(blk_run_queue);
/**
- * blk_run_queues - fire all plugged queues
- *
- * Description:
- * Start I/O on all plugged queues known to the block layer. Queues that
- * are currently stopped are ignored. This is equivalent to the older
- * tq_disk task queue run.
- **/
-#define blk_plug_entry(entry) list_entry((entry), request_queue_t, plug_list)
-void blk_run_queues(void)
-{
- LIST_HEAD(local_plug_list);
-
- spin_lock_irq(&blk_plug_lock);
-
- /*
- * this will happen fairly often
- */
- if (list_empty(&blk_plug_list))
- goto out;
-
- list_splice_init(&blk_plug_list, &local_plug_list);
-
- while (!list_empty(&local_plug_list)) {
- request_queue_t *q = blk_plug_entry(local_plug_list.next);
-
- spin_unlock_irq(&blk_plug_lock);
- q->unplug_fn(q);
- spin_lock_irq(&blk_plug_lock);
- }
-out:
- spin_unlock_irq(&blk_plug_lock);
-}
-
-EXPORT_SYMBOL(blk_run_queues);
-
-/**
* blk_cleanup_queue: - release a &request_queue_t when it is no longer needed
* @q: the request queue to be released
*
@@ -1348,6 +1318,8 @@ static elevator_t *chosen_elevator =
&iosched_as;
#elif defined(CONFIG_IOSCHED_DEADLINE)
&iosched_deadline;
+#elif defined(CONFIG_IOSCHED_CFQ)
+ &iosched_cfq;
#elif defined(CONFIG_IOSCHED_NOOP)
&elevator_noop;
#else
@@ -1366,6 +1338,10 @@ static int __init elevator_setup(char *s
if (!strcmp(str, "as"))
chosen_elevator = &iosched_as;
#endif
+#ifdef CONFIG_IOSCHED_CFQ
+ if (!strcmp(str, "cfq"))
+ chosen_elevator = &iosched_cfq;
+#endif
#ifdef CONFIG_IOSCHED_NOOP
if (!strcmp(str, "noop"))
chosen_elevator = &elevator_noop;
@@ -1386,6 +1362,10 @@ request_queue_t *blk_alloc_queue(int gfp
memset(q, 0, sizeof(*q));
init_timer(&q->unplug_timer);
atomic_set(&q->refcnt, 1);
+
+ q->backing_dev_info.unplug_io_fn = blk_backing_dev_unplug;
+ q->backing_dev_info.unplug_io_data = q;
+
return q;
}
@@ -2046,7 +2026,6 @@ long blk_congestion_wait(int rw, long ti
DEFINE_WAIT(wait);
wait_queue_head_t *wqh = &congestion_wqh[rw];
- blk_run_queues();
prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE);
ret = io_schedule_timeout(timeout);
finish_wait(wqh, &wait);
@@ -2084,6 +2063,15 @@ static int attempt_merge(request_queue_t
if (!q->merge_requests_fn(q, req, next))
return 0;
+ /*
+ * At this point we have either done a back merge
+ * or front merge. We need the smaller start_time of
+ * the merged requests to be the current request
+ * for accounting purposes.
+ */
+ if (time_after(req->start_time, next->start_time))
+ req->start_time = next->start_time;
+
req->biotail->bi_next = next->bio;
req->biotail = next->biotail;
@@ -2302,7 +2290,7 @@ out:
if (blk_queue_plugged(q)) {
int nr_queued = q->rq.count[READ] + q->rq.count[WRITE];
- if (nr_queued == q->unplug_thresh)
+ if (nr_queued == q->unplug_thresh || bio_sync(bio))
__generic_unplug_device(q);
}
spin_unlock_irq(q->queue_lock);
@@ -2459,6 +2447,16 @@ int submit_bio(int rw, struct bio *bio)
mod_page_state(pgpgout, count);
else
mod_page_state(pgpgin, count);
+
+ if (unlikely(block_dump)) {
+ char b[BDEVNAME_SIZE];
+ printk("%s(%d): %s block %Lu on %s\n",
+ current->comm, current->pid,
+ (rw & WRITE) ? "WRITE" : "READ",
+ (unsigned long long)bio->bi_sector,
+ bdevname(bio->bi_bdev,b));
+ }
+
generic_make_request(bio);
return 1;
}
@@ -2746,10 +2744,24 @@ void end_that_request_last(struct reques
unsigned long duration = jiffies - req->start_time;
switch (rq_data_dir(req)) {
case WRITE:
+ /*
+ * schedule the writeout of pending dirty data when the
+ * disk is idle. (Writeback is not postponed by
+ * writes, only by reads.)
+ */
+ if (unlikely(laptop_mode))
+ disk_is_spun_up(0);
disk_stat_inc(disk, writes);
disk_stat_add(disk, write_ticks, duration);
break;
case READ:
+ /*
+ * schedule the writeout of pending dirty data when the
+ * disk is idle. (postpone writeback until system is
+ * quiescent again.)
+ */
+ if (unlikely(laptop_mode))
+ disk_is_spun_up(1);
disk_stat_inc(disk, reads);
disk_stat_add(disk, read_ticks, duration);
break;
@@ -2825,8 +2837,6 @@ void kblockd_flush(void)
int __init blk_dev_init(void)
{
- int i;
-
kblockd_workqueue = create_workqueue("kblockd");
if (!kblockd_workqueue)
panic("Failed to create kblockd\n");
@@ -2838,9 +2848,6 @@ int __init blk_dev_init(void)
blk_max_low_pfn = max_low_pfn;
blk_max_pfn = max_pfn;
-
- for (i = 0; i < ARRAY_SIZE(congestion_wqh); i++)
- init_waitqueue_head(&congestion_wqh[i]);
return 0;
}
--- diff/drivers/block/loop.c 2004-03-16 11:24:01.000000000 +0000
+++ source/drivers/block/loop.c 2004-03-16 11:27:48.008889200 +0000
@@ -434,6 +434,17 @@ inactive:
goto out;
}
+/*
+ * kick off io on the underlying address space
+ */
+static void loop_unplug(request_queue_t *q)
+{
+ struct loop_device *lo = q->queuedata;
+
+ clear_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags);
+ blk_run_address_space(lo->lo_backing_file->f_mapping);
+}
+
struct switch_request {
struct file *file;
struct completion wait;
@@ -614,7 +625,6 @@ static int loop_set_fd(struct loop_devic
{
struct file *file;
struct inode *inode;
- struct block_device *lo_device = NULL;
struct address_space *mapping;
unsigned lo_blocksize;
int lo_flags = 0;
@@ -671,7 +681,7 @@ static int loop_set_fd(struct loop_devic
set_device_ro(bdev, (lo_flags & LO_FLAGS_READ_ONLY) != 0);
lo->lo_blocksize = lo_blocksize;
- lo->lo_device = lo_device;
+ lo->lo_device = bdev;
lo->lo_flags = lo_flags;
lo->lo_backing_file = file;
lo->transfer = NULL;
@@ -689,6 +699,7 @@ static int loop_set_fd(struct loop_devic
*/
blk_queue_make_request(lo->lo_queue, loop_make_request);
lo->lo_queue->queuedata = lo;
+ lo->lo_queue->unplug_fn = loop_unplug;
set_capacity(disks[lo->lo_number], size);
--- diff/drivers/block/rd.c 2004-03-11 10:20:23.000000000 +0000
+++ source/drivers/block/rd.c 2004-03-16 11:27:48.009889048 +0000
@@ -271,6 +271,7 @@ static int rd_ioctl(struct inode *inode,
static struct backing_dev_info rd_backing_dev_info = {
.ra_pages = 0, /* No readahead */
.memory_backed = 1, /* Does not contribute to dirty memory */
+ .unplug_io_fn = default_unplug_io_fn,
};
static int rd_open(struct inode *inode, struct file *filp)
--- diff/drivers/block/scsi_ioctl.c 2004-03-16 11:24:01.000000000 +0000
+++ source/drivers/block/scsi_ioctl.c 2004-03-16 11:27:48.010888896 +0000
@@ -230,7 +230,7 @@ static int sg_scsi_ioctl(request_queue_t
return -EFAULT;
if (in_len > PAGE_SIZE || out_len > PAGE_SIZE)
return -EINVAL;
- if (get_user(opcode, sic->data))
+ if (get_user(opcode, (int *)sic->data))
return -EFAULT;
bytes = max(in_len, out_len);
--- diff/drivers/block/umem.c 2003-09-30 14:46:12.000000000 +0000
+++ source/drivers/block/umem.c 2004-03-16 11:27:48.011888744 +0000
@@ -368,9 +368,8 @@ static inline void reset_page(struct mm_
page->biotail = & page->bio;
}
-static void mm_unplug_device(void *data)
+static void mm_unplug_device(request_queue_t *q)
{
- request_queue_t *q = data;
struct cardinfo *card = q->queuedata;
unsigned long flags;
--- diff/drivers/cdrom/cdu31a.c 2003-09-17 11:28:04.000000000 +0000
+++ source/drivers/cdrom/cdu31a.c 2004-03-16 11:27:48.013888440 +0000
@@ -3203,13 +3203,16 @@ static struct gendisk *scd_gendisk;
static char *load_mech[] __initdata =
{ "caddy", "tray", "pop-up", "unknown" };
-static void __init
+static int __init
get_drive_configuration(unsigned short base_io,
unsigned char res_reg[], unsigned int *res_size)
{
unsigned long retry_count;
+ if (!request_region(base_io, 4, "cdu31a"))
+ return 0;
+
/* Set the base address */
cdu31a_port = base_io;
@@ -3244,7 +3247,7 @@ get_drive_configuration(unsigned short b
/* If attention is never seen probably not a CDU31a present */
if (!is_attention()) {
res_reg[0] = 0x20;
- return;
+ goto out_err;
}
#endif
@@ -3254,11 +3257,17 @@ get_drive_configuration(unsigned short b
do_sony_cd_cmd(SONY_REQ_DRIVE_CONFIG_CMD,
NULL,
0, (unsigned char *) res_reg, res_size);
- return;
+ if (*res_size <= 2 || (res_reg[0] & 0xf0) != 0)
+ goto out_err;
+ return 1;
}
/* Return an error */
res_reg[0] = 0x20;
+out_err:
+ release_region(cdu31a_port, 4);
+ cdu31a_port = 0;
+ return 0;
}
#ifndef MODULE
@@ -3307,9 +3316,6 @@ int __init cdu31a_init(void)
char msg[255];
char buf[40];
int i;
- int drive_found;
- int tmp_irq;
-
/*
* According to Alex Freed (freed@europa.orion.adobe.com), this is
@@ -3323,52 +3329,33 @@ int __init cdu31a_init(void)
outb(0xe2, 0x9a01);
}
- drive_found = 0;
-
/* Setting the base I/O address to 0xffff will disable it. */
- if (cdu31a_port == 0xffff) {
- } else if (cdu31a_port != 0) {
- tmp_irq = cdu31a_irq; /* Need IRQ 0 because we can't sleep here. */
- cdu31a_irq = 0;
-
- get_drive_configuration(cdu31a_port,
- drive_config.exec_status,
- &res_size);
- if ((res_size > 2)
- && ((drive_config.exec_status[0] & 0xf0) == 0x00)) {
- drive_found = 1;
- }
+ if (cdu31a_port == 0xffff)
+ goto errout3;
+ if (cdu31a_port != 0) {
+ /* Need IRQ 0 because we can't sleep here. */
+ int tmp_irq = cdu31a_irq;
+ cdu31a_irq = 0;
+ if (!get_drive_configuration(cdu31a_port,
+ drive_config.exec_status,
+ &res_size))
+ goto errout3;
cdu31a_irq = tmp_irq;
} else {
cdu31a_irq = 0;
- i = 0;
- while ((cdu31a_addresses[i].base != 0)
- && (!drive_found)) {
- if (check_region(cdu31a_addresses[i].base, 4)) {
- i++;
- continue;
- }
- get_drive_configuration(cdu31a_addresses[i].base,
- drive_config.exec_status,
- &res_size);
- if ((res_size > 2)
- && ((drive_config.exec_status[0] & 0xf0) ==
- 0x00)) {
- drive_found = 1;
+ for (i = 0; cdu31a_addresses[i].base; i++) {
+ if (get_drive_configuration(cdu31a_addresses[i].base,
+ drive_config.exec_status,
+ &res_size)) {
cdu31a_irq = cdu31a_addresses[i].int_num;
- } else {
- i++;
+ break;
}
}
+ if (!cdu31a_port)
+ goto errout3;
}
- if (!drive_found)
- goto errout3;
-
- if (!request_region(cdu31a_port, 4, "cdu31a"))
- goto errout3;
-
if (register_blkdev(MAJOR_NR, "cdu31a"))
goto errout2;
--- diff/drivers/cdrom/cm206.c 2003-09-17 11:28:04.000000000 +0000
+++ source/drivers/cdrom/cm206.c 2004-03-16 11:27:48.014888288 +0000
@@ -1389,7 +1389,7 @@ static struct gendisk *cm206_gendisk;
Linus says it is too dangerous to use writes for probing, so we
stick with pure reads for a while. Hope that 8 possible ranges,
- check_region, 15 bits of one port and 6 of another make things
+ request_region, 15 bits of one port and 6 of another make things
likely enough to accept the region on the first hit...
*/
int __init probe_base_port(int base)
@@ -1400,13 +1400,15 @@ int __init probe_base_port(int base)
if (base)
b = e = base;
for (base = b; base <= e; base += 0x10) {
- if (check_region(base, 0x10))
+ if (!request_region(base, 0x10,"cm206"))
continue;
for (i = 0; i < 3; i++)
fool = inw(base + 2); /* empty possibly uart_receive_buffer */
if ((inw(base + 6) & 0xffef) != 0x0001 || /* line_status */
- (inw(base) & 0xad00) != 0) /* data status */
+ (inw(base) & 0xad00) != 0) { /* data status */
+ release_region(base,0x10);
continue;
+ }
return (base);
}
return 0;
@@ -1444,7 +1446,6 @@ int __init cm206_init(void)
return -EIO;
}
printk(" adapter at 0x%x", cm206_base);
- request_region(cm206_base, 16, "cm206");
cd = (struct cm206_struct *) kmalloc(size, GFP_KERNEL);
if (!cd)
goto out_base;
--- diff/drivers/cdrom/sjcd.c 2003-10-27 09:20:37.000000000 +0000
+++ source/drivers/cdrom/sjcd.c 2004-03-16 11:27:48.016887984 +0000
@@ -1700,7 +1700,7 @@ static int __init sjcd_init(void)
sprintf(sjcd_disk->disk_name, "sjcd");
sprintf(sjcd_disk->devfs_name, "sjcd");
- if (check_region(sjcd_base, 4)) {
+ if (!request_region(sjcd_base, 4,"sjcd")) {
printk
("SJCD: Init failed, I/O port (%X) is already in use\n",
sjcd_base);
--- diff/drivers/char/Kconfig 2004-03-16 11:24:01.000000000 +0000
+++ source/drivers/char/Kconfig 2004-03-16 11:27:48.104874608 +0000
@@ -599,30 +599,6 @@ config PC9800_OLDLP_CONSOLE
bool "Support for console on line printer"
depends on PC9800_OLDLP
-
-menu "Mice"
-
-config BUSMOUSE
- tristate "Bus Mouse Support"
- ---help---
- Say Y here if your machine has a bus mouse as opposed to a serial
- mouse. Most people have a regular serial MouseSystem or
- Microsoft mouse (made by Logitech) that plugs into a COM port
- (rectangular with 9 or 25 pins). These people say N here.
-
- If you have a laptop, you either have to check the documentation or
- experiment a bit to find out whether the trackball is a serial mouse
- or not; it's best to say Y here for you.
-
- This is the generic bus mouse driver code. If you have a bus mouse,
- you will have to say Y here and also to the specific driver for your
- mouse below.
-
- To compile this driver as a module, choose M here: the
- module will be called busmouse.
-
-endmenu
-
config QIC02_TAPE
tristate "QIC-02 tape support"
help
--- diff/drivers/char/Makefile 2004-02-18 08:54:09.000000000 +0000
+++ source/drivers/char/Makefile 2004-03-16 11:27:48.106874304 +0000
@@ -49,7 +49,6 @@ obj-$(CONFIG_PRINTER) += lp.o
obj-$(CONFIG_TIPAR) += tipar.o
obj-$(CONFIG_PC9800_OLDLP) += lp_old98.o
-obj-$(CONFIG_BUSMOUSE) += busmouse.o
obj-$(CONFIG_DTLK) += dtlk.o
obj-$(CONFIG_R3964) += n_r3964.o
obj-$(CONFIG_APPLICOM) += applicom.o
--- diff/drivers/char/drm/Kconfig 2004-02-09 10:36:10.000000000 +0000
+++ source/drivers/char/drm/Kconfig 2004-03-16 11:27:48.048883120 +0000
@@ -76,7 +76,7 @@ config DRM_SIS
tristate "SiS video cards"
depends on DRM && AGP
help
- Choose this option if you have a SiS 630 or compatibel video
+ Choose this option if you have a SiS 630 or compatible video
chipset. If M is selected the module will be called sis. AGP
support is required for this driver to work.
--- diff/drivers/char/drm/drm.h 2003-07-22 17:54:27.000000000 +0000
+++ source/drivers/char/drm/drm.h 2004-03-16 11:27:48.021887224 +0000
@@ -580,6 +580,16 @@ typedef struct drm_scatter_gather {
unsigned long handle; /**< Used for mapping / unmapping */
} drm_scatter_gather_t;
+/**
+ * DRM_IOCTL_SET_VERSION ioctl argument type.
+ */
+typedef struct drm_set_version {
+ int drm_di_major;
+ int drm_di_minor;
+ int drm_dd_major;
+ int drm_dd_minor;
+} drm_set_version_t;
+
#define DRM_IOCTL_BASE 'd'
#define DRM_IO(nr) _IO(DRM_IOCTL_BASE,nr)
@@ -594,6 +604,7 @@ typedef struct drm_scatter_gather {
#define DRM_IOCTL_GET_MAP DRM_IOWR(0x04, drm_map_t)
#define DRM_IOCTL_GET_CLIENT DRM_IOWR(0x05, drm_client_t)
#define DRM_IOCTL_GET_STATS DRM_IOR( 0x06, drm_stats_t)
+#define DRM_IOCTL_SET_VERSION DRM_IOWR(0x07, drm_set_version_t)
#define DRM_IOCTL_SET_UNIQUE DRM_IOW( 0x10, drm_unique_t)
#define DRM_IOCTL_AUTH_MAGIC DRM_IOW( 0x11, drm_auth_t)
--- diff/drivers/char/drm/drmP.h 2004-01-19 10:22:56.000000000 +0000
+++ source/drivers/char/drm/drmP.h 2004-03-16 11:27:48.026886464 +0000
@@ -92,8 +92,8 @@
#ifndef __HAVE_DMA
#define __HAVE_DMA 0
#endif
-#ifndef __HAVE_DMA_IRQ
-#define __HAVE_DMA_IRQ 0
+#ifndef __HAVE_IRQ
+#define __HAVE_IRQ 0
#endif
#ifndef __HAVE_DMA_WAITLIST
#define __HAVE_DMA_WAITLIST 0
@@ -324,6 +324,7 @@ do { \
#define DRM_BUFCOUNT(x) ((x)->count - DRM_LEFTCOUNT(x))
#define DRM_WAITCOUNT(dev,idx) DRM_BUFCOUNT(&dev->queuelist[idx]->waitlist)
+#define DRM_IF_VERSION(maj, min) ((maj) << 16 | (min))
/**
* Get the private SAREA mapping.
*
@@ -362,10 +363,13 @@ do { \
typedef int drm_ioctl_t( struct inode *inode, struct file *filp,
unsigned int cmd, unsigned long arg );
-typedef struct drm_pci_list {
- u16 vendor;
- u16 device;
-} drm_pci_list_t;
+typedef struct drm_pci_id_list
+{
+ int vendor;
+ int device;
+ long driver_private;
+ char *name;
+} drm_pci_id_list_t;
typedef struct drm_ioctl_desc {
drm_ioctl_t *func;
@@ -488,6 +492,9 @@ typedef struct drm_file {
struct drm_device *dev;
int remove_auth_on_close;
unsigned long lock_count;
+#ifdef DRIVER_FILE_FIELDS
+ DRIVER_FILE_FIELDS;
+#endif
} drm_file_t;
/** Wait queue */
@@ -622,6 +629,8 @@ typedef struct drm_device {
int unique_len; /**< Length of unique field */
dev_t device; /**< Device number for mknod */
char *devname; /**< For /proc/interrupts */
+ int minor; /**< Minor device number */
+ int if_version; /**< Highest interface version set */
int blocked; /**< Blocked due to VC switch? */
struct proc_dir_entry *root; /**< Root for this device's entries */
@@ -679,6 +688,7 @@ typedef struct drm_device {
/** \name Context support */
/*@{*/
int irq; /**< Interrupt used by board */
+ int irq_enabled; /**< True if irq handler is enabled */
__volatile__ long context_flag; /**< Context swapping flag */
__volatile__ long interrupt_flag; /**< Interruption handler flag */
__volatile__ long dma_flag; /**< DMA dispatch flag */
@@ -714,7 +724,12 @@ typedef struct drm_device {
#if __REALLY_HAVE_AGP
drm_agp_head_t *agp; /**< AGP data */
#endif
- struct pci_dev *pdev; /**< PCI device structure */
+
+ struct pci_dev *pdev; /**< PCI device structure */
+ int pci_domain; /**< PCI bus domain number */
+ int pci_bus; /**< PCI bus number */
+ int pci_slot; /**< PCI slot number */
+ int pci_func; /**< PCI function number */
#ifdef __alpha__
#if LINUX_VERSION_CODE < KERNEL_VERSION(2,4,3)
struct pci_controler *hose;
@@ -804,8 +819,8 @@ extern int DRM(unbind_agp)(DRM
#endif
/* Misc. IOCTL support (drm_ioctl.h) */
-extern int DRM(irq_busid)(struct inode *inode, struct file *filp,
- unsigned int cmd, unsigned long arg);
+extern int DRM(irq_by_busid)(struct inode *inode, struct file *filp,
+ unsigned int cmd, unsigned long arg);
extern int DRM(getunique)(struct inode *inode, struct file *filp,
unsigned int cmd, unsigned long arg);
extern int DRM(setunique)(struct inode *inode, struct file *filp,
@@ -816,6 +831,8 @@ extern int DRM(getclient)(struct in
unsigned int cmd, unsigned long arg);
extern int DRM(getstats)(struct inode *inode, struct file *filp,
unsigned int cmd, unsigned long arg);
+extern int DRM(setversion)(struct inode *inode, struct file *filp,
+ unsigned int cmd, unsigned long arg);
/* Context IOCTL support (drm_context.h) */
extern int DRM(resctx)( struct inode *inode, struct file *filp,
@@ -900,12 +917,17 @@ extern int DRM(dma_setup)(drm_devic
extern void DRM(dma_takedown)(drm_device_t *dev);
extern void DRM(free_buffer)(drm_device_t *dev, drm_buf_t *buf);
extern void DRM(reclaim_buffers)( struct file *filp );
-#if __HAVE_DMA_IRQ
+#endif /* __HAVE_DMA */
+
+ /* IRQ support (drm_irq.h) */
+#if __HAVE_IRQ || __HAVE_DMA
extern int DRM(control)( struct inode *inode, struct file *filp,
unsigned int cmd, unsigned long arg );
-extern int DRM(irq_install)( drm_device_t *dev, int irq );
+#endif
+#if __HAVE_IRQ
+extern int DRM(irq_install)( drm_device_t *dev );
extern int DRM(irq_uninstall)( drm_device_t *dev );
-extern irqreturn_t DRM(dma_service)( DRM_IRQ_ARGS );
+extern irqreturn_t DRM(irq_handler)( DRM_IRQ_ARGS );
extern void DRM(driver_irq_preinstall)( drm_device_t *dev );
extern void DRM(driver_irq_postinstall)( drm_device_t *dev );
extern void DRM(driver_irq_uninstall)( drm_device_t *dev );
@@ -915,12 +937,11 @@ extern int DRM(wait_vblank)(st
extern int DRM(vblank_wait)(drm_device_t *dev, unsigned int *vbl_seq);
extern void DRM(vbl_send_signals)( drm_device_t *dev );
#endif
-#if __HAVE_DMA_IRQ_BH
-extern void DRM(dma_immediate_bh)( void *dev );
+#if __HAVE_IRQ_BH
+extern void DRM(irq_immediate_bh)( void *dev );
#endif
#endif
-#endif /* __HAVE_DMA */
#if __REALLY_HAVE_AGP
/* AGP/GART support (drm_agpsupport.h) */
--- diff/drivers/char/drm/drm_bufs.h 2003-07-22 17:54:27.000000000 +0000
+++ source/drivers/char/drm/drm_bufs.h 2004-03-16 11:27:48.017887832 +0000
@@ -147,7 +147,9 @@ int DRM(addmap)( struct inode *inode, st
MTRR_TYPE_WRCOMB, 1 );
}
#endif
- map->handle = DRM(ioremap)( map->offset, map->size, dev );
+ if (map->type == _DRM_REGISTERS)
+ map->handle = DRM(ioremap)( map->offset, map->size,
+ dev );
break;
case _DRM_SHM:
@@ -160,6 +162,12 @@ int DRM(addmap)( struct inode *inode, st
}
map->offset = (unsigned long)map->handle;
if ( map->flags & _DRM_CONTAINS_LOCK ) {
+ /* Prevent a 2nd X Server from creating a 2nd lock */
+ if (dev->lock.hw_lock != NULL) {
+ vfree( map->handle );
+ DRM(free)( map, sizeof(*map), DRM_MEM_MAPS );
+ return -EBUSY;
+ }
dev->sigdata.lock =
dev->lock.hw_lock = map->handle; /* Pointer to lock */
}
--- diff/drivers/char/drm/drm_dma.h 2003-07-22 17:54:27.000000000 +0000
+++ source/drivers/char/drm/drm_dma.h 2004-03-16 11:27:48.018887680 +0000
@@ -43,15 +43,6 @@
#ifndef __HAVE_DMA_RECLAIM
#define __HAVE_DMA_RECLAIM 0
#endif
-#ifndef __HAVE_SHARED_IRQ
-#define __HAVE_SHARED_IRQ 0
-#endif
-
-#if __HAVE_SHARED_IRQ
-#define DRM_IRQ_TYPE SA_SHIRQ
-#else
-#define DRM_IRQ_TYPE 0
-#endif
#if __HAVE_DMA
@@ -214,293 +205,11 @@ void DRM(reclaim_buffers)( struct file *
}
#endif
-
-
-
-#if __HAVE_DMA_IRQ
-
-/**
- * Install IRQ handler.
- *
- * \param dev DRM device.
- * \param irq IRQ number.
- *
- * Initializes the IRQ related data, and setups drm_device::vbl_queue. Installs the handler, calling the driver
- * \c DRM(driver_irq_preinstall)() and \c DRM(driver_irq_postinstall)() functions
- * before and after the installation.
- */
-int DRM(irq_install)( drm_device_t *dev, int irq )
-{
- int ret;
-
- if ( !irq )
- return -EINVAL;
-
- down( &dev->struct_sem );
-
- /* Driver must have been initialized */
- if ( !dev->dev_private ) {
- up( &dev->struct_sem );
- return -EINVAL;
- }
-
- if ( dev->irq ) {
- up( &dev->struct_sem );
- return -EBUSY;
- }
- dev->irq = irq;
- up( &dev->struct_sem );
-
- DRM_DEBUG( "%s: irq=%d\n", __FUNCTION__, irq );
-
- dev->context_flag = 0;
- dev->interrupt_flag = 0;
- dev->dma_flag = 0;
-
- dev->dma->next_buffer = NULL;
- dev->dma->next_queue = NULL;
- dev->dma->this_buffer = NULL;
-
-#if __HAVE_DMA_IRQ_BH
- INIT_WORK(&dev->work, DRM(dma_immediate_bh), dev);
-#endif
-
-#if __HAVE_VBL_IRQ
- init_waitqueue_head(&dev->vbl_queue);
-
- spin_lock_init( &dev->vbl_lock );
-
- INIT_LIST_HEAD( &dev->vbl_sigs.head );
-
- dev->vbl_pending = 0;
-#endif
-
- /* Before installing handler */
- DRM(driver_irq_preinstall)(dev);
-
- /* Install handler */
- ret = request_irq( dev->irq, DRM(dma_service),
- DRM_IRQ_TYPE, dev->devname, dev );
- if ( ret < 0 ) {
- down( &dev->struct_sem );
- dev->irq = 0;
- up( &dev->struct_sem );
- return ret;
- }
-
- /* After installing handler */
- DRM(driver_irq_postinstall)(dev);
-
- return 0;
-}
-
-/**
- * Uninstall the IRQ handler.
- *
- * \param dev DRM device.
- *
- * Calls the driver's \c DRM(driver_irq_uninstall)() function, and stops the irq.
- */
-int DRM(irq_uninstall)( drm_device_t *dev )
-{
- int irq;
-
- down( &dev->struct_sem );
- irq = dev->irq;
- dev->irq = 0;
- up( &dev->struct_sem );
-
- if ( !irq )
- return -EINVAL;
-
- DRM_DEBUG( "%s: irq=%d\n", __FUNCTION__, irq );
-
- DRM(driver_irq_uninstall)( dev );
-
- free_irq( irq, dev );
-
- return 0;
-}
-
-/**
- * IRQ control ioctl.
- *
- * \param inode device inode.
- * \param filp file pointer.
- * \param cmd command.
- * \param arg user argument, pointing to a drm_control structure.
- * \return zero on success or a negative number on failure.
- *
- * Calls irq_install() or irq_uninstall() according to \p arg.
- */
-int DRM(control)( struct inode *inode, struct file *filp,
- unsigned int cmd, unsigned long arg )
-{
- drm_file_t *priv = filp->private_data;
- drm_device_t *dev = priv->dev;
- drm_control_t ctl;
-
- if ( copy_from_user( &ctl, (drm_control_t *)arg, sizeof(ctl) ) )
- return -EFAULT;
-
- switch ( ctl.func ) {
- case DRM_INST_HANDLER:
- return DRM(irq_install)( dev, ctl.irq );
- case DRM_UNINST_HANDLER:
- return DRM(irq_uninstall)( dev );
- default:
- return -EINVAL;
- }
-}
-
-#if __HAVE_VBL_IRQ
-
-/**
- * Wait for VBLANK.
- *
- * \param inode device inode.
- * \param filp file pointer.
- * \param cmd command.
- * \param data user argument, pointing to a drm_wait_vblank structure.
- * \return zero on success or a negative number on failure.
- *
- * Verifies the IRQ is installed.
- *
- * If a signal is requested checks if this task has already scheduled the same signal
- * for the same vblank sequence number - nothing to be done in
- * that case. If the number of tasks waiting for the interrupt exceeds 100 the
- * function fails. Otherwise adds a new entry to drm_device::vbl_sigs for this
- * task.
- *
- * If a signal is not requested, then calls vblank_wait().
- */
-int DRM(wait_vblank)( DRM_IOCTL_ARGS )
-{
- drm_file_t *priv = filp->private_data;
- drm_device_t *dev = priv->dev;
- drm_wait_vblank_t vblwait;
- struct timeval now;
- int ret = 0;
- unsigned int flags;
-
- if (!dev->irq)
- return -EINVAL;
-
- DRM_COPY_FROM_USER_IOCTL( vblwait, (drm_wait_vblank_t *)data,
- sizeof(vblwait) );
-
- switch ( vblwait.request.type & ~_DRM_VBLANK_FLAGS_MASK ) {
- case _DRM_VBLANK_RELATIVE:
- vblwait.request.sequence += atomic_read( &dev->vbl_received );
- vblwait.request.type &= ~_DRM_VBLANK_RELATIVE;
- case _DRM_VBLANK_ABSOLUTE:
- break;
- default:
- return -EINVAL;
- }
-
- flags = vblwait.request.type & _DRM_VBLANK_FLAGS_MASK;
-
- if ( flags & _DRM_VBLANK_SIGNAL ) {
- unsigned long irqflags;
- drm_vbl_sig_t *vbl_sig;
-
- vblwait.reply.sequence = atomic_read( &dev->vbl_received );
-
- spin_lock_irqsave( &dev->vbl_lock, irqflags );
-
- /* Check if this task has already scheduled the same signal
- * for the same vblank sequence number; nothing to be done in
- * that case
- */
- list_for_each_entry( vbl_sig, &dev->vbl_sigs.head, head ) {
- if (vbl_sig->sequence == vblwait.request.sequence
- && vbl_sig->info.si_signo == vblwait.request.signal
- && vbl_sig->task == current)
- {
- spin_unlock_irqrestore( &dev->vbl_lock, irqflags );
- goto done;
- }
- }
-
- if ( dev->vbl_pending >= 100 ) {
- spin_unlock_irqrestore( &dev->vbl_lock, irqflags );
- return -EBUSY;
- }
-
- dev->vbl_pending++;
-
- spin_unlock_irqrestore( &dev->vbl_lock, irqflags );
-
- if ( !( vbl_sig = DRM_MALLOC( sizeof( drm_vbl_sig_t ) ) ) ) {
- return -ENOMEM;
- }
-
- memset( (void *)vbl_sig, 0, sizeof(*vbl_sig) );
-
- vbl_sig->sequence = vblwait.request.sequence;
- vbl_sig->info.si_signo = vblwait.request.signal;
- vbl_sig->task = current;
-
- spin_lock_irqsave( &dev->vbl_lock, irqflags );
-
- list_add_tail( (struct list_head *) vbl_sig, &dev->vbl_sigs.head );
-
- spin_unlock_irqrestore( &dev->vbl_lock, irqflags );
- } else {
- ret = DRM(vblank_wait)( dev, &vblwait.request.sequence );
-
- do_gettimeofday( &now );
- vblwait.reply.tval_sec = now.tv_sec;
- vblwait.reply.tval_usec = now.tv_usec;
- }
-
-done:
- DRM_COPY_TO_USER_IOCTL( (drm_wait_vblank_t *)data, vblwait,
- sizeof(vblwait) );
-
- return ret;
-}
-
-/**
- * Send the VBLANK signals.
- *
- * \param dev DRM device.
- *
- * Sends a signal for each task in drm_device::vbl_sigs and empties the list.
- *
- * If a signal is not requested, then calls vblank_wait().
+#if !__HAVE_IRQ
+/* This stub DRM_IOCTL_CONTROL handler is for the drivers that used to require
+ * IRQs for DMA but no longer do. It maintains compatibility with the X Servers
+ * that try to use the control ioctl by simply returning success.
*/
-void DRM(vbl_send_signals)( drm_device_t *dev )
-{
- struct list_head *list, *tmp;
- drm_vbl_sig_t *vbl_sig;
- unsigned int vbl_seq = atomic_read( &dev->vbl_received );
- unsigned long flags;
-
- spin_lock_irqsave( &dev->vbl_lock, flags );
-
- list_for_each_safe( list, tmp, &dev->vbl_sigs.head ) {
- vbl_sig = list_entry( list, drm_vbl_sig_t, head );
- if ( ( vbl_seq - vbl_sig->sequence ) <= (1<<23) ) {
- vbl_sig->info.si_code = vbl_seq;
- send_sig_info( vbl_sig->info.si_signo, &vbl_sig->info, vbl_sig->task );
-
- list_del( list );
-
- DRM_FREE( vbl_sig, sizeof(*vbl_sig) );
-
- dev->vbl_pending--;
- }
- }
-
- spin_unlock_irqrestore( &dev->vbl_lock, flags );
-}
-
-#endif /* __HAVE_VBL_IRQ */
-
-#else
-
int DRM(control)( struct inode *inode, struct file *filp,
unsigned int cmd, unsigned long arg )
{
@@ -517,7 +226,6 @@ int DRM(control)( struct inode *inode, s
return -EINVAL;
}
}
-
-#endif /* __HAVE_DMA_IRQ */
+#endif
#endif /* __HAVE_DMA */
--- diff/drivers/char/drm/drm_drv.h 2003-10-27 09:20:43.000000000 +0000
+++ source/drivers/char/drm/drm_drv.h 2004-03-16 11:27:48.020887376 +0000
@@ -58,8 +58,8 @@
#ifndef __HAVE_CTX_BITMAP
#define __HAVE_CTX_BITMAP 0
#endif
-#ifndef __HAVE_DMA_IRQ
-#define __HAVE_DMA_IRQ 0
+#ifndef __HAVE_IRQ
+#define __HAVE_IRQ 0
#endif
#ifndef __HAVE_DMA_QUEUE
#define __HAVE_DMA_QUEUE 0
@@ -126,6 +126,9 @@
#ifndef DRIVER_IOCTLS
#define DRIVER_IOCTLS
#endif
+#ifndef DRIVER_OPEN_HELPER
+#define DRIVER_OPEN_HELPER( priv, dev )
+#endif
#ifndef DRIVER_FOPS
#define DRIVER_FOPS \
static struct file_operations DRM(fops) = { \
@@ -159,15 +162,8 @@ __setup( DRIVER_NAME "=", DRM_OPTIONS_FU
#undef DRM_OPTIONS_FUNC
#endif
-/**
- * The default number of instances (minor numbers) to initialize.
- */
-#ifndef DRIVER_NUM_CARDS
-#define DRIVER_NUM_CARDS 1
-#endif
-
-static drm_device_t *DRM(device);
-static int *DRM(minor);
+#define MAX_DEVICES 4
+static drm_device_t DRM(device)[MAX_DEVICES];
static int DRM(numdevs) = 0;
DRIVER_FOPS;
@@ -177,10 +173,13 @@ static drm_ioctl_desc_t DRM(ioctls)[]
[DRM_IOCTL_NR(DRM_IOCTL_VERSION)] = { DRM(version), 0, 0 },
[DRM_IOCTL_NR(DRM_IOCTL_GET_UNIQUE)] = { DRM(getunique), 0, 0 },
[DRM_IOCTL_NR(DRM_IOCTL_GET_MAGIC)] = { DRM(getmagic), 0, 0 },
- [DRM_IOCTL_NR(DRM_IOCTL_IRQ_BUSID)] = { DRM(irq_busid), 0, 1 },
+#if __HAVE_IRQ
+ [DRM_IOCTL_NR(DRM_IOCTL_IRQ_BUSID)] = { DRM(irq_by_busid), 0, 1 },
+#endif
[DRM_IOCTL_NR(DRM_IOCTL_GET_MAP)] = { DRM(getmap), 0, 0 },
[DRM_IOCTL_NR(DRM_IOCTL_GET_CLIENT)] = { DRM(getclient), 0, 0 },
[DRM_IOCTL_NR(DRM_IOCTL_GET_STATS)] = { DRM(getstats), 0, 0 },
+ [DRM_IOCTL_NR(DRM_IOCTL_SET_VERSION)] = { DRM(setversion), 0, 1 },
[DRM_IOCTL_NR(DRM_IOCTL_SET_UNIQUE)] = { DRM(setunique), 1, 1 },
[DRM_IOCTL_NR(DRM_IOCTL_BLOCK)] = { DRM(noop), 1, 1 },
@@ -222,9 +221,9 @@ static drm_ioctl_desc_t DRM(ioctls)[]
[DRM_IOCTL_NR(DRM_IOCTL_INFO_BUFS)] = { DRM(infobufs), 1, 0 },
[DRM_IOCTL_NR(DRM_IOCTL_MAP_BUFS)] = { DRM(mapbufs), 1, 0 },
[DRM_IOCTL_NR(DRM_IOCTL_FREE_BUFS)] = { DRM(freebufs), 1, 0 },
-
- /* The DRM_IOCTL_DMA ioctl should be defined by the driver.
- */
+ /* The DRM_IOCTL_DMA ioctl should be defined by the driver. */
+#endif
+#if __HAVE_IRQ || __HAVE_DMA
[DRM_IOCTL_NR(DRM_IOCTL_CONTROL)] = { DRM(control), 1, 1 },
#endif
@@ -337,7 +336,7 @@ static int DRM(setup)( drm_device_t *dev
dev->queue_reserved = 0;
dev->queue_slots = 0;
dev->queuelist = NULL;
- dev->irq = 0;
+ dev->irq_enabled = 0;
dev->context_flag = 0;
dev->interrupt_flag = 0;
dev->dma_flag = 0;
@@ -345,6 +344,7 @@ static int DRM(setup)( drm_device_t *dev
dev->last_switch = 0;
dev->last_checked = 0;
init_waitqueue_head( &dev->context_wait );
+ dev->if_version = 0;
dev->ctx_start = 0;
dev->lck_start = 0;
@@ -391,8 +391,8 @@ static int DRM(takedown)( drm_device_t *
DRM_DEBUG( "\n" );
DRIVER_PRETAKEDOWN();
-#if __HAVE_DMA_IRQ
- if ( dev->irq ) DRM(irq_uninstall)( dev );
+#if __HAVE_IRQ
+ if ( dev->irq_enabled ) DRM(irq_uninstall)( dev );
#endif
down( &dev->struct_sem );
@@ -534,52 +534,111 @@ static int DRM(takedown)( drm_device_t *
return 0;
}
-/**
- * Figure out how many instances to initialize.
- *
- * \return number of cards found.
- *
- * Searches for every PCI card in \c DRIVER_CARD_LIST with matching vendor and device ids.
- */
-static int drm_count_cards(void)
+static drm_pci_id_list_t DRM(pciidlist)[] = {
+ DRIVER_PCI_IDS
+};
+
+static int DRM(probe)(struct pci_dev *pdev)
{
- int num = 0;
-#if defined(DRIVER_CARD_LIST)
- int i;
- drm_pci_list_t *l;
- u16 device, vendor;
- struct pci_dev *pdev = NULL;
+ drm_device_t *dev;
+#if __HAVE_CTX_BITMAP
+ int retcode;
#endif
+ int i;
+ char *desc = NULL;
DRM_DEBUG( "\n" );
-#if defined(DRIVER_COUNT_CARDS)
- num = DRIVER_COUNT_CARDS();
-#elif defined(DRIVER_CARD_LIST)
- for (i = 0, l = DRIVER_CARD_LIST; l[i].vendor != 0; i++) {
- pdev = NULL;
- vendor = l[i].vendor;
- device = l[i].device;
- if(device == 0xffff) device = PCI_ANY_ID;
- if(vendor == 0xffff) vendor = PCI_ANY_ID;
- while ((pdev = pci_find_device(vendor, device, pdev))) {
- num++;
+ for (i = 0; DRM(pciidlist)[i].vendor != 0; i++) {
+ if ((DRM(pciidlist)[i].vendor == pdev->vendor) &&
+ (DRM(pciidlist)[i].device == pdev->device)) {
+ desc = DRM(pciidlist)[i].name;
}
}
+ if (desc == NULL)
+ return -ENODEV;
+
+ if (DRM(numdevs) >= MAX_DEVICES)
+ return -ENODEV;
+
+ dev = &(DRM(device)[DRM(numdevs)]);
+
+ memset( (void *)dev, 0, sizeof(*dev) );
+ dev->count_lock = SPIN_LOCK_UNLOCKED;
+ init_timer( &dev->timer );
+ sema_init( &dev->struct_sem, 1 );
+
+ if ((dev->minor = DRM(stub_register)(DRIVER_NAME, &DRM(fops),dev)) < 0)
+ return -EPERM;
+ dev->device = MKDEV(DRM_MAJOR, dev->minor );
+ dev->name = DRIVER_NAME;
+
+ dev->pdev = pdev;
+#ifdef __alpha__
+ dev->hose = pdev->sysdata;
+ dev->pci_domain = dev->hose->bus->number;
#else
- num = DRIVER_NUM_CARDS;
+ dev->pci_domain = 0;
+#endif
+ dev->pci_bus = pdev->bus->number;
+ dev->pci_slot = PCI_SLOT(pdev->devfn);
+ dev->pci_func = PCI_FUNC(pdev->devfn);
+ dev->irq = pdev->irq;
+
+ DRIVER_PREINIT();
+
+#if __REALLY_HAVE_AGP
+ dev->agp = DRM(agp_init)();
+#if __MUST_HAVE_AGP
+ if ( dev->agp == NULL ) {
+ DRM_ERROR( "Cannot initialize the agpgart module.\n" );
+ DRM(stub_unregister)(dev->minor);
+ DRM(takedown)( dev );
+ return -ENOMEM;
+ }
+#endif
+#if __REALLY_HAVE_MTRR
+ if (dev->agp)
+ dev->agp->agp_mtrr = mtrr_add( dev->agp->agp_info.aper_base,
+ dev->agp->agp_info.aper_size*1024*1024,
+ MTRR_TYPE_WRCOMB,
+ 1 );
+#endif
+#endif
+
+#if __HAVE_CTX_BITMAP
+ retcode = DRM(ctxbitmap_init)( dev );
+ if( retcode ) {
+ DRM_ERROR( "Cannot allocate memory for context bitmap.\n" );
+ DRM(stub_unregister)(dev->minor);
+ DRM(takedown)( dev );
+ return retcode;
+ }
#endif
- DRM_DEBUG("numdevs = %d\n", num);
- return num;
+ DRM(numdevs)++; /* no errors, mark it reserved */
+
+ DRM_INFO( "Initialized %s %d.%d.%d %s on minor %d: %s\n",
+ DRIVER_NAME,
+ DRIVER_MAJOR,
+ DRIVER_MINOR,
+ DRIVER_PATCHLEVEL,
+ DRIVER_DATE,
+ dev->minor,
+ desc );
+
+ DRIVER_POSTINIT();
+
+ return 0;
}
+
/**
* Module initialization. Called via init_module at module load time, or via
* linux/init/main.c (this is not currently supported).
*
* \return zero on success or a negative number on failure.
*
- * Allocates and initialize an array of drm_device structures, and attempts to
+ * Initializes an array of drm_device structures, and attempts to
* initialize all available devices, using consecutive minors, registering the
* stubs and initializing the AGP device.
*
@@ -588,88 +647,19 @@ static int drm_count_cards(void)
*/
static int __init drm_init( void )
{
+ struct pci_dev *pdev = NULL;
- drm_device_t *dev;
- int i;
-#if __HAVE_CTX_BITMAP
- int retcode;
-#endif
DRM_DEBUG( "\n" );
#ifdef MODULE
DRM(parse_options)( drm_opts );
#endif
- DRM(numdevs) = drm_count_cards();
- /* Force at least one instance. */
- if (DRM(numdevs) <= 0)
- DRM(numdevs) = 1;
-
- DRM(device) = kmalloc(sizeof(*DRM(device)) * DRM(numdevs), GFP_KERNEL);
- if (!DRM(device)) {
- return -ENOMEM;
- }
- DRM(minor) = kmalloc(sizeof(*DRM(minor)) * DRM(numdevs), GFP_KERNEL);
- if (!DRM(minor)) {
- kfree(DRM(device));
- return -ENOMEM;
- }
-
- DRIVER_PREINIT();
-
DRM(mem_init)();
- for (i = 0; i < DRM(numdevs); i++) {
- dev = &(DRM(device)[i]);
- memset( (void *)dev, 0, sizeof(*dev) );
- dev->count_lock = SPIN_LOCK_UNLOCKED;
- init_timer( &dev->timer );
- sema_init( &dev->struct_sem, 1 );
-
- if ((DRM(minor)[i] = DRM(stub_register)(DRIVER_NAME, &DRM(fops),dev)) < 0)
- return -EPERM;
- dev->device = MKDEV(DRM_MAJOR, DRM(minor)[i] );
- dev->name = DRIVER_NAME;
-
-#if __REALLY_HAVE_AGP
- dev->agp = DRM(agp_init)();
-#if __MUST_HAVE_AGP
- if ( dev->agp == NULL ) {
- DRM_ERROR( "Cannot initialize the agpgart module.\n" );
- DRM(stub_unregister)(DRM(minor)[i]);
- DRM(takedown)( dev );
- return -EINVAL;
- }
-#endif
-#if __REALLY_HAVE_MTRR
- if (dev->agp)
- dev->agp->agp_mtrr = mtrr_add( dev->agp->agp_info.aper_base,
- dev->agp->agp_info.aper_size*1024*1024,
- MTRR_TYPE_WRCOMB,
- 1 );
-#endif
-#endif
-
-#if __HAVE_CTX_BITMAP
- retcode = DRM(ctxbitmap_init)( dev );
- if( retcode ) {
- DRM_ERROR( "Cannot allocate memory for context bitmap.\n" );
- DRM(stub_unregister)(DRM(minor)[i]);
- DRM(takedown)( dev );
- return retcode;
- }
-#endif
- DRM_INFO( "Initialized %s %d.%d.%d %s on minor %d\n",
- DRIVER_NAME,
- DRIVER_MAJOR,
- DRIVER_MINOR,
- DRIVER_PATCHLEVEL,
- DRIVER_DATE,
- DRM(minor)[i] );
+ while ((pdev = pci_find_device(PCI_ANY_ID, PCI_ANY_ID, pdev)) != NULL) {
+ DRM(probe)(pdev);
}
-
- DRIVER_POSTINIT();
-
return 0;
}
@@ -689,10 +679,10 @@ static void __exit drm_cleanup( void )
for (i = DRM(numdevs) - 1; i >= 0; i--) {
dev = &(DRM(device)[i]);
- if ( DRM(stub_unregister)(DRM(minor)[i]) ) {
+ if ( DRM(stub_unregister)(dev->minor) ) {
DRM_ERROR( "Cannot unload module\n" );
} else {
- DRM_DEBUG("minor %d unregistered\n", DRM(minor)[i]);
+ DRM_DEBUG("minor %d unregistered\n", dev->minor);
if (i == 0) {
DRM_INFO( "Module unloaded\n" );
}
@@ -722,8 +712,6 @@ static void __exit drm_cleanup( void )
#endif
}
DRIVER_POSTCLEANUP();
- kfree(DRM(minor));
- kfree(DRM(device));
DRM(numdevs) = 0;
}
@@ -795,7 +783,7 @@ int DRM(open)( struct inode *inode, stru
int i;
for (i = 0; i < DRM(numdevs); i++) {
- if (iminor(inode) == DRM(minor)[i]) {
+ if (iminor(inode) == DRM(device)[i].minor) {
dev = &(DRM(device)[i]);
break;
}
--- diff/drivers/char/drm/drm_fops.h 2003-10-09 08:47:16.000000000 +0000
+++ source/drivers/char/drm/drm_fops.h 2004-03-16 11:27:48.020887376 +0000
@@ -72,6 +72,8 @@ int DRM(open_helper)(struct inode *inode
priv->authenticated = capable(CAP_SYS_ADMIN);
priv->lock_count = 0;
+ DRIVER_OPEN_HELPER( priv, dev );
+
down(&dev->struct_sem);
if (!dev->file_last) {
priv->next = NULL;
--- diff/drivers/char/drm/drm_ioctl.h 2003-07-22 17:54:27.000000000 +0000
+++ source/drivers/char/drm/drm_ioctl.h 2004-03-16 11:27:48.022887072 +0000
@@ -35,69 +35,7 @@
#include "drmP.h"
-
-/**
- * Get interrupt from bus id.
- *
- * \param inode device inode.
- * \param filp file pointer.
- * \param cmd command.
- * \param arg user argument, pointing to a drm_irq_busid structure.
- * \return zero on success or a negative number on failure.
- *
- * Finds the PCI device with the specified bus id and gets its IRQ number.
- */
-int DRM(irq_busid)(struct inode *inode, struct file *filp,
- unsigned int cmd, unsigned long arg)
-{
- drm_irq_busid_t p;
- struct pci_dev *dev;
-
- if (copy_from_user(&p, (drm_irq_busid_t *)arg, sizeof(p)))
- return -EFAULT;
-#ifdef __alpha__
- {
- int domain = p.busnum >> 8;
- p.busnum &= 0xff;
-
- /*
- * Find the hose the device is on (the domain number is the
- * hose index) and offset the bus by the root bus of that
- * hose.
- */
- for(dev = pci_find_device(PCI_ANY_ID,PCI_ANY_ID,NULL);
- dev;
- dev = pci_find_device(PCI_ANY_ID,PCI_ANY_ID,dev)) {
- struct pci_controller *hose = dev->sysdata;
-
- if (hose->index == domain) {
- p.busnum += hose->bus->number;
- break;
- }
- }
- }
-#endif
- dev = pci_find_slot(p.busnum, PCI_DEVFN(p.devnum, p.funcnum));
- if (!dev) {
- DRM_ERROR("pci_find_slot failed for %d:%d:%d\n",
- p.busnum, p.devnum, p.funcnum);
- p.irq = 0;
- goto out;
- }
- if (pci_enable_device(dev) != 0) {
- DRM_ERROR("pci_enable_device failed for %d:%d:%d\n",
- p.busnum, p.devnum, p.funcnum);
- p.irq = 0;
- goto out;
- }
- p.irq = dev->irq;
- out:
- DRM_DEBUG("%d:%d:%d => IRQ %d\n",
- p.busnum, p.devnum, p.funcnum, p.irq);
- if (copy_to_user((drm_irq_busid_t *)arg, &p, sizeof(p)))
- return -EFAULT;
- return 0;
-}
+#include <linux/pci.h>
/**
* Get the bus id.
@@ -138,8 +76,10 @@ int DRM(getunique)(struct inode *inode,
* \param arg user argument, pointing to a drm_unique structure.
* \return zero on success or a negative number on failure.
*
- * Copies the bus id from userspace into drm_device::unique, and searches for
- * the respective PCI device, updating drm_device::pdev.
+ * Copies the bus id from userspace into drm_device::unique, and verifies that
+ * it matches the device this DRM is attached to (EINVAL otherwise). Deprecated
+ * in interface version 1.1 and will return EBUSY when setversion has requested
+ * version 1.1 or greater.
*/
int DRM(setunique)(struct inode *inode, struct file *filp,
unsigned int cmd, unsigned long arg)
@@ -147,6 +87,7 @@ int DRM(setunique)(struct inode *inode,
drm_file_t *priv = filp->private_data;
drm_device_t *dev = priv->dev;
drm_unique_t u;
+ int domain, bus, slot, func, ret;
if (dev->unique_len || dev->unique) return -EBUSY;
@@ -164,55 +105,42 @@ int DRM(setunique)(struct inode *inode,
dev->devname = DRM(alloc)(strlen(dev->name) + strlen(dev->unique) + 2,
DRM_MEM_DRIVER);
- if(!dev->devname) {
- DRM(free)(dev->devname, sizeof(*dev->devname), DRM_MEM_DRIVER);
+ if (!dev->devname)
return -ENOMEM;
- }
+
sprintf(dev->devname, "%s@%s", dev->name, dev->unique);
- do {
- struct pci_dev *pci_dev;
- int domain, b, d, f;
- char *p;
-
- for(p = dev->unique; p && *p && *p != ':'; p++);
- if (!p || !*p) break;
- b = (int)simple_strtoul(p+1, &p, 10);
- if (*p != ':') break;
- d = (int)simple_strtoul(p+1, &p, 10);
- if (*p != ':') break;
- f = (int)simple_strtoul(p+1, &p, 10);
- if (*p) break;
-
- domain = b >> 8;
- b &= 0xff;
-
-#ifdef __alpha__
- /*
- * Find the hose the device is on (the domain number is the
- * hose index) and offset the bus by the root bus of that
- * hose.
- */
- for(pci_dev = pci_find_device(PCI_ANY_ID,PCI_ANY_ID,NULL);
- pci_dev;
- pci_dev = pci_find_device(PCI_ANY_ID,PCI_ANY_ID,pci_dev)) {
- struct pci_controller *hose = pci_dev->sysdata;
-
- if (hose->index == domain) {
- b += hose->bus->number;
- break;
- }
- }
-#endif
+ /* Return error if the busid submitted doesn't match the device's actual
+ * busid.
+ */
+ ret = sscanf(dev->unique, "PCI:%d:%d:%d", &bus, &slot, &func);
+ if (ret != 3)
+ return DRM_ERR(EINVAL);
+ domain = bus >> 8;
+ bus &= 0xff;
+
+ if ((domain != dev->pci_domain) ||
+ (bus != dev->pci_bus) ||
+ (slot != dev->pci_slot) ||
+ (func != dev->pci_func))
+ return -EINVAL;
- pci_dev = pci_find_slot(b, PCI_DEVFN(d,f));
- if (pci_dev) {
- dev->pdev = pci_dev;
-#ifdef __alpha__
- dev->hose = pci_dev->sysdata;
-#endif
- }
- } while(0);
+ return 0;
+}
+
+static int
+DRM(set_busid)(drm_device_t *dev)
+{
+ if (dev->unique != NULL)
+ return EBUSY;
+
+ dev->unique_len = 20;
+ dev->unique = DRM(alloc)(dev->unique_len + 1, DRM_MEM_DRIVER);
+ if (dev->unique == NULL)
+ return ENOMEM;
+
+ snprintf(dev->unique, dev->unique_len, "pci:%04x:%02x:%02x.%d",
+ dev->pci_domain, dev->pci_bus, dev->pci_slot, dev->pci_func);
return 0;
}
@@ -363,3 +291,48 @@ int DRM(getstats)( struct inode *inode,
return -EFAULT;
return 0;
}
+
+#define DRM_IF_MAJOR 1
+#define DRM_IF_MINOR 2
+
+int DRM(setversion)(DRM_IOCTL_ARGS)
+{
+ DRM_DEVICE;
+ drm_set_version_t sv;
+ drm_set_version_t retv;
+ int if_version;
+
+ DRM_COPY_FROM_USER_IOCTL(sv, (drm_set_version_t *)data, sizeof(sv));
+
+ retv.drm_di_major = DRM_IF_MAJOR;
+ retv.drm_di_minor = DRM_IF_MINOR;
+ retv.drm_dd_major = DRIVER_MAJOR;
+ retv.drm_dd_minor = DRIVER_MINOR;
+
+ DRM_COPY_TO_USER_IOCTL((drm_set_version_t *)data, retv, sizeof(sv));
+
+ if (sv.drm_di_major != -1) {
+ if (sv.drm_di_major != DRM_IF_MAJOR ||
+ sv.drm_di_minor < 0 || sv.drm_di_minor > DRM_IF_MINOR)
+ return EINVAL;
+ if_version = DRM_IF_VERSION(sv.drm_di_major, sv.drm_dd_minor);
+ dev->if_version = DRM_MAX(if_version, dev->if_version);
+ if (sv.drm_di_minor >= 1) {
+ /*
+ * Version 1.1 includes tying of DRM to specific device
+ */
+ DRM(set_busid)(dev);
+ }
+ }
+
+ if (sv.drm_dd_major != -1) {
+ if (sv.drm_dd_major != DRIVER_MAJOR ||
+ sv.drm_dd_minor < 0 || sv.drm_dd_minor > DRIVER_MINOR)
+ return EINVAL;
+#ifdef DRIVER_SETVERSION
+ DRIVER_SETVERSION(dev, &sv);
+#endif
+ }
+ return 0;
+}
+
--- diff/drivers/char/drm/drm_os_linux.h 2003-09-30 14:46:12.000000000 +0000
+++ source/drivers/char/drm/drm_os_linux.h 2004-03-16 11:27:48.024886768 +0000
@@ -62,8 +62,12 @@
verify_area( VERIFY_READ, uaddr, size )
#define DRM_COPY_FROM_USER_UNCHECKED(arg1, arg2, arg3) \
__copy_from_user(arg1, arg2, arg3)
+#define DRM_COPY_TO_USER_UNCHECKED(arg1, arg2, arg3) \
+ __copy_to_user(arg1, arg2, arg3)
#define DRM_GET_USER_UNCHECKED(val, uaddr) \
__get_user(val, uaddr)
+#define DRM_PUT_USER_UNCHECKED(uaddr, val) \
+ __put_user(val, uaddr)
/** 'malloc' without the overhead of DRM(alloc)() */
@@ -71,6 +75,8 @@
/** 'free' without the overhead of DRM(free)() */
#define DRM_FREE(x,size) kfree(x)
+#define DRM_GET_PRIV_WITH_RETURN(_priv, _filp) _priv = _filp->private_data
+
/**
* Get the pointer to the SAREA.
*
--- diff/drivers/char/drm/gamma.h 2003-05-21 10:50:14.000000000 +0000
+++ source/drivers/char/drm/gamma.h 2004-03-16 11:27:48.028886160 +0000
@@ -53,6 +53,10 @@
[DRM_IOCTL_NR(DRM_IOCTL_GAMMA_INIT)] = { gamma_dma_init, 1, 1 }, \
[DRM_IOCTL_NR(DRM_IOCTL_GAMMA_COPY)] = { gamma_dma_copy, 1, 1 }
+#define DRIVER_PCI_IDS \
+ {0x3d3d, 0x0008, 0, "3DLabs GLINT Gamma G1"}, \
+ {0, 0, 0, NULL}
+
#define IOCTL_TABLE_NAME DRM(ioctls)
#define IOCTL_FUNC_NAME DRM(ioctl)
@@ -104,8 +108,8 @@
return 0; \
} while (0)
-#define __HAVE_DMA_IRQ 1
-#define __HAVE_DMA_IRQ_BH 1
+#define __HAVE_IRQ 1
+#define __HAVE_IRQ_BH 1
#define DRIVER_AGP_BUFFERS_MAP( dev ) \
((drm_gamma_private_t *)((dev)->dev_private))->buffers
--- diff/drivers/char/drm/gamma_dma.c 2003-09-30 14:46:12.000000000 +0000
+++ source/drivers/char/drm/gamma_dma.c 2004-03-16 11:27:48.027886312 +0000
@@ -116,7 +116,7 @@ static inline int gamma_dma_is_ready(drm
return (!GAMMA_READ(GAMMA_DMACOUNT));
}
-irqreturn_t gamma_dma_service( DRM_IRQ_ARGS )
+irqreturn_t gamma_irq_handler( DRM_IRQ_ARGS )
{
drm_device_t *dev = (drm_device_t *)arg;
drm_device_dma_t *dma = dev->dma;
@@ -262,7 +262,7 @@ static void gamma_dma_timer_bh(unsigned
gamma_dma_schedule((drm_device_t *)dev, 0);
}
-void gamma_dma_immediate_bh(void *dev)
+void gamma_irq_immediate_bh(void *dev)
{
gamma_dma_schedule(dev, 0);
}
@@ -656,12 +656,12 @@ int gamma_do_cleanup_dma( drm_device_t *
{
DRM_DEBUG( "%s\n", __FUNCTION__ );
-#if _HAVE_DMA_IRQ
+#if __HAVE_IRQ
/* Make sure interrupts are disabled here because the uninstall ioctl
* may not have been called from userspace and after dev_private
* is freed, it's too late.
*/
- if ( dev->irq ) DRM(irq_uninstall)(dev);
+ if ( dev->irq_enabled ) DRM(irq_uninstall)(dev);
#endif
if ( dev->dev_private ) {
--- diff/drivers/char/drm/gamma_drv.c 2003-05-21 10:50:14.000000000 +0000
+++ source/drivers/char/drm/gamma_drv.c 2004-03-16 11:27:48.027886312 +0000
@@ -48,6 +48,7 @@
#include "drm_fops.h"
#include "drm_init.h"
#include "drm_ioctl.h"
+#include "drm_irq.h"
#include "gamma_lists.h" /* NOTE */
#include "drm_lock.h"
#include "gamma_lock.h" /* NOTE */
--- diff/drivers/char/drm/i810.h 2003-08-26 09:00:52.000000000 +0000
+++ source/drivers/char/drm/i810.h 2004-03-16 11:27:48.032885552 +0000
@@ -77,7 +77,14 @@
[DRM_IOCTL_NR(DRM_IOCTL_I810_MC)] = { i810_dma_mc, 1, 1 }, \
[DRM_IOCTL_NR(DRM_IOCTL_I810_RSTATUS)] = { i810_rstatus, 1, 0 }, \
[DRM_IOCTL_NR(DRM_IOCTL_I810_FLIP)] = { i810_flip_bufs, 1, 0 }
-
+
+#define DRIVER_PCI_IDS \
+ {0x8086, 0x7121, 0, "Intel i810 GMCH"}, \
+ {0x8086, 0x7123, 0, "Intel i810-DC100 GMCH"}, \
+ {0x8086, 0x7125, 0, "Intel i810E GMCH"}, \
+ {0x8086, 0x1132, 0, "Intel i815 GMCH"}, \
+ {0, 0, 0, NULL}
+
#define __HAVE_COUNTERS 4
#define __HAVE_COUNTER6 _DRM_STAT_IRQ
@@ -112,7 +119,7 @@
* a noop stub is generated for compatibility.
*/
/* XXX: Add vblank support? */
-#define __HAVE_DMA_IRQ 0
+#define __HAVE_IRQ 0
/* Buffer customization:
*/
--- diff/drivers/char/drm/i810_dma.c 2003-09-30 14:46:12.000000000 +0000
+++ source/drivers/char/drm/i810_dma.c 2004-03-16 11:27:48.031885704 +0000
@@ -53,41 +53,41 @@
static inline void i810_print_status_page(drm_device_t *dev)
{
- drm_device_dma_t *dma = dev->dma;
- drm_i810_private_t *dev_priv = dev->dev_private;
+ drm_device_dma_t *dma = dev->dma;
+ drm_i810_private_t *dev_priv = dev->dev_private;
u32 *temp = dev_priv->hw_status_page;
- int i;
+ int i;
- DRM_DEBUG( "hw_status: Interrupt Status : %x\n", temp[0]);
- DRM_DEBUG( "hw_status: LpRing Head ptr : %x\n", temp[1]);
- DRM_DEBUG( "hw_status: IRing Head ptr : %x\n", temp[2]);
- DRM_DEBUG( "hw_status: Reserved : %x\n", temp[3]);
+ DRM_DEBUG( "hw_status: Interrupt Status : %x\n", temp[0]);
+ DRM_DEBUG( "hw_status: LpRing Head ptr : %x\n", temp[1]);
+ DRM_DEBUG( "hw_status: IRing Head ptr : %x\n", temp[2]);
+ DRM_DEBUG( "hw_status: Reserved : %x\n", temp[3]);
DRM_DEBUG( "hw_status: Last Render: %x\n", temp[4]);
- DRM_DEBUG( "hw_status: Driver Counter : %d\n", temp[5]);
- for(i = 6; i < dma->buf_count + 6; i++) {
- DRM_DEBUG( "buffer status idx : %d used: %d\n", i - 6, temp[i]);
+ DRM_DEBUG( "hw_status: Driver Counter : %d\n", temp[5]);
+ for(i = 6; i < dma->buf_count + 6; i++) {
+ DRM_DEBUG( "buffer status idx : %d used: %d\n", i - 6, temp[i]);
}
}
static drm_buf_t *i810_freelist_get(drm_device_t *dev)
{
- drm_device_dma_t *dma = dev->dma;
+ drm_device_dma_t *dma = dev->dma;
int i;
- int used;
+ int used;
/* Linear search might not be the best solution */
- for (i = 0; i < dma->buf_count; i++) {
- drm_buf_t *buf = dma->buflist[ i ];
- drm_i810_buf_priv_t *buf_priv = buf->dev_private;
+ for (i = 0; i < dma->buf_count; i++) {
+ drm_buf_t *buf = dma->buflist[ i ];
+ drm_i810_buf_priv_t *buf_priv = buf->dev_private;
/* In use is already a pointer */
- used = cmpxchg(buf_priv->in_use, I810_BUF_FREE,
+ used = cmpxchg(buf_priv->in_use, I810_BUF_FREE,
I810_BUF_CLIENT);
if (used == I810_BUF_FREE) {
return buf;
}
}
- return NULL;
+ return NULL;
}
/* This should only be called if the buffer is not sent to the hardware
@@ -96,17 +96,17 @@ static drm_buf_t *i810_freelist_get(drm_
static int i810_freelist_put(drm_device_t *dev, drm_buf_t *buf)
{
- drm_i810_buf_priv_t *buf_priv = buf->dev_private;
- int used;
+ drm_i810_buf_priv_t *buf_priv = buf->dev_private;
+ int used;
- /* In use is already a pointer */
- used = cmpxchg(buf_priv->in_use, I810_BUF_CLIENT, I810_BUF_FREE);
+ /* In use is already a pointer */
+ used = cmpxchg(buf_priv->in_use, I810_BUF_CLIENT, I810_BUF_FREE);
if (used != I810_BUF_CLIENT) {
- DRM_ERROR("Freeing buffer thats not in use : %d\n", buf->idx);
- return -EINVAL;
+ DRM_ERROR("Freeing buffer thats not in use : %d\n", buf->idx);
+ return -EINVAL;
}
- return 0;
+ return 0;
}
static struct file_operations i810_buffer_fops = {
@@ -135,7 +135,7 @@ int i810_mmap_buffers(struct file *filp,
vma->vm_flags |= (VM_IO | VM_DONTCOPY);
vma->vm_file = filp;
- buf_priv->currently_mapped = I810_BUF_MAPPED;
+ buf_priv->currently_mapped = I810_BUF_MAPPED;
unlock_kernel();
if (remap_page_range(DRM_RPR_ARG(vma) vma->vm_start,
@@ -150,8 +150,8 @@ static int i810_map_buffer(drm_buf_t *bu
drm_file_t *priv = filp->private_data;
drm_device_t *dev = priv->dev;
drm_i810_buf_priv_t *buf_priv = buf->dev_private;
- drm_i810_private_t *dev_priv = dev->dev_private;
- struct file_operations *old_fops;
+ drm_i810_private_t *dev_priv = dev->dev_private;
+ struct file_operations *old_fops;
int retcode = 0;
if (buf_priv->currently_mapped == I810_BUF_MAPPED)
@@ -192,8 +192,8 @@ static int i810_unmap_buffer(drm_buf_t *
(size_t) buf->total);
up_write(¤t->mm->mmap_sem);
- buf_priv->currently_mapped = I810_BUF_UNMAPPED;
- buf_priv->virtual = 0;
+ buf_priv->currently_mapped = I810_BUF_UNMAPPED;
+ buf_priv->virtual = 0;
return retcode;
}
@@ -208,22 +208,22 @@ static int i810_dma_get_buffer(drm_devic
buf = i810_freelist_get(dev);
if (!buf) {
retcode = -ENOMEM;
- DRM_DEBUG("retcode=%d\n", retcode);
+ DRM_DEBUG("retcode=%d\n", retcode);
return retcode;
}
retcode = i810_map_buffer(buf, filp);
if (retcode) {
i810_freelist_put(dev, buf);
- DRM_ERROR("mapbuf failed, retcode %d\n", retcode);
+ DRM_ERROR("mapbuf failed, retcode %d\n", retcode);
return retcode;
}
buf->filp = filp;
buf_priv = buf->dev_private;
d->granted = 1;
- d->request_idx = buf->idx;
- d->request_size = buf->total;
- d->virtual = buf_priv->virtual;
+ d->request_idx = buf->idx;
+ d->request_size = buf->total;
+ d->virtual = buf_priv->virtual;
return retcode;
}
@@ -232,33 +232,33 @@ int i810_dma_cleanup(drm_device_t *dev)
{
drm_device_dma_t *dma = dev->dma;
-#if _HAVE_DMA_IRQ
+#if __HAVE_IRQ
/* Make sure interrupts are disabled here because the uninstall ioctl
* may not have been called from userspace and after dev_private
* is freed, it's too late.
*/
- if (dev->irq) DRM(irq_uninstall)(dev);
+ if (dev->irq_enabled) DRM(irq_uninstall)(dev);
#endif
if (dev->dev_private) {
int i;
- drm_i810_private_t *dev_priv =
- (drm_i810_private_t *) dev->dev_private;
+ drm_i810_private_t *dev_priv =
+ (drm_i810_private_t *) dev->dev_private;
if (dev_priv->ring.virtual_start) {
- DRM(ioremapfree)((void *) dev_priv->ring.virtual_start,
+ DRM(ioremapfree)((void *) dev_priv->ring.virtual_start,
dev_priv->ring.Size, dev);
}
- if (dev_priv->hw_status_page) {
- pci_free_consistent(dev->pdev, PAGE_SIZE,
+ if (dev_priv->hw_status_page) {
+ pci_free_consistent(dev->pdev, PAGE_SIZE,
dev_priv->hw_status_page,
dev_priv->dma_status_page);
- /* Need to rewrite hardware status page */
- I810_WRITE(0x02080, 0x1ffff000);
+ /* Need to rewrite hardware status page */
+ I810_WRITE(0x02080, 0x1ffff000);
}
- DRM(free)(dev->dev_private, sizeof(drm_i810_private_t),
+ DRM(free)(dev->dev_private, sizeof(drm_i810_private_t),
DRM_MEM_DRIVER);
- dev->dev_private = NULL;
+ dev->dev_private = NULL;
for (i = 0; i < dma->buf_count; i++) {
drm_buf_t *buf = dma->buflist[ i ];
@@ -267,73 +267,73 @@ int i810_dma_cleanup(drm_device_t *dev)
DRM(ioremapfree)(buf_priv->kernel_virtual, buf->total, dev);
}
}
- return 0;
+ return 0;
}
static int i810_wait_ring(drm_device_t *dev, int n)
{
- drm_i810_private_t *dev_priv = dev->dev_private;
- drm_i810_ring_buffer_t *ring = &(dev_priv->ring);
- int iters = 0;
- unsigned long end;
+ drm_i810_private_t *dev_priv = dev->dev_private;
+ drm_i810_ring_buffer_t *ring = &(dev_priv->ring);
+ int iters = 0;
+ unsigned long end;
unsigned int last_head = I810_READ(LP_RING + RING_HEAD) & HEAD_ADDR;
end = jiffies + (HZ*3);
- while (ring->space < n) {
- ring->head = I810_READ(LP_RING + RING_HEAD) & HEAD_ADDR;
- ring->space = ring->head - (ring->tail+8);
+ while (ring->space < n) {
+ ring->head = I810_READ(LP_RING + RING_HEAD) & HEAD_ADDR;
+ ring->space = ring->head - (ring->tail+8);
if (ring->space < 0) ring->space += ring->Size;
-
+
if (ring->head != last_head) {
end = jiffies + (HZ*3);
last_head = ring->head;
}
- iters++;
+ iters++;
if (time_before(end, jiffies)) {
- DRM_ERROR("space: %d wanted %d\n", ring->space, n);
- DRM_ERROR("lockup\n");
- goto out_wait_ring;
+ DRM_ERROR("space: %d wanted %d\n", ring->space, n);
+ DRM_ERROR("lockup\n");
+ goto out_wait_ring;
}
udelay(1);
}
out_wait_ring:
- return iters;
+ return iters;
}
static void i810_kernel_lost_context(drm_device_t *dev)
{
- drm_i810_private_t *dev_priv = dev->dev_private;
- drm_i810_ring_buffer_t *ring = &(dev_priv->ring);
+ drm_i810_private_t *dev_priv = dev->dev_private;
+ drm_i810_ring_buffer_t *ring = &(dev_priv->ring);
- ring->head = I810_READ(LP_RING + RING_HEAD) & HEAD_ADDR;
- ring->tail = I810_READ(LP_RING + RING_TAIL);
- ring->space = ring->head - (ring->tail+8);
- if (ring->space < 0) ring->space += ring->Size;
+ ring->head = I810_READ(LP_RING + RING_HEAD) & HEAD_ADDR;
+ ring->tail = I810_READ(LP_RING + RING_TAIL);
+ ring->space = ring->head - (ring->tail+8);
+ if (ring->space < 0) ring->space += ring->Size;
}
static int i810_freelist_init(drm_device_t *dev, drm_i810_private_t *dev_priv)
{
- drm_device_dma_t *dma = dev->dma;
- int my_idx = 24;
- u32 *hw_status = (u32 *)(dev_priv->hw_status_page + my_idx);
- int i;
+ drm_device_dma_t *dma = dev->dma;
+ int my_idx = 24;
+ u32 *hw_status = (u32 *)(dev_priv->hw_status_page + my_idx);
+ int i;
if (dma->buf_count > 1019) {
- /* Not enough space in the status page for the freelist */
- return -EINVAL;
+ /* Not enough space in the status page for the freelist */
+ return -EINVAL;
}
- for (i = 0; i < dma->buf_count; i++) {
- drm_buf_t *buf = dma->buflist[ i ];
- drm_i810_buf_priv_t *buf_priv = buf->dev_private;
+ for (i = 0; i < dma->buf_count; i++) {
+ drm_buf_t *buf = dma->buflist[ i ];
+ drm_i810_buf_priv_t *buf_priv = buf->dev_private;
- buf_priv->in_use = hw_status++;
- buf_priv->my_use_idx = my_idx;
- my_idx += 4;
+ buf_priv->in_use = hw_status++;
+ buf_priv->my_use_idx = my_idx;
+ my_idx += 4;
- *buf_priv->in_use = I810_BUF_FREE;
+ *buf_priv->in_use = I810_BUF_FREE;
buf_priv->kernel_virtual = DRM(ioremap)(buf->bus_address,
buf->total, dev);
@@ -347,7 +347,7 @@ static int i810_dma_initialize(drm_devic
{
struct list_head *list;
- memset(dev_priv, 0, sizeof(drm_i810_private_t));
+ memset(dev_priv, 0, sizeof(drm_i810_private_t));
list_for_each(list, &dev->maplist->head) {
drm_map_list_t *r_list = list_entry(list, drm_map_list_t, head);
@@ -355,51 +355,51 @@ static int i810_dma_initialize(drm_devic
r_list->map->type == _DRM_SHM &&
r_list->map->flags & _DRM_CONTAINS_LOCK ) {
dev_priv->sarea_map = r_list->map;
- break;
- }
- }
+ break;
+ }
+ }
if (!dev_priv->sarea_map) {
dev->dev_private = (void *)dev_priv;
- i810_dma_cleanup(dev);
- DRM_ERROR("can not find sarea!\n");
- return -EINVAL;
+ i810_dma_cleanup(dev);
+ DRM_ERROR("can not find sarea!\n");
+ return -EINVAL;
}
DRM_FIND_MAP( dev_priv->mmio_map, init->mmio_offset );
if (!dev_priv->mmio_map) {
dev->dev_private = (void *)dev_priv;
- i810_dma_cleanup(dev);
- DRM_ERROR("can not find mmio map!\n");
- return -EINVAL;
+ i810_dma_cleanup(dev);
+ DRM_ERROR("can not find mmio map!\n");
+ return -EINVAL;
}
DRM_FIND_MAP( dev_priv->buffer_map, init->buffers_offset );
if (!dev_priv->buffer_map) {
dev->dev_private = (void *)dev_priv;
- i810_dma_cleanup(dev);
- DRM_ERROR("can not find dma buffer map!\n");
- return -EINVAL;
+ i810_dma_cleanup(dev);
+ DRM_ERROR("can not find dma buffer map!\n");
+ return -EINVAL;
}
dev_priv->sarea_priv = (drm_i810_sarea_t *)
((u8 *)dev_priv->sarea_map->handle +
init->sarea_priv_offset);
- dev_priv->ring.Start = init->ring_start;
- dev_priv->ring.End = init->ring_end;
- dev_priv->ring.Size = init->ring_size;
+ dev_priv->ring.Start = init->ring_start;
+ dev_priv->ring.End = init->ring_end;
+ dev_priv->ring.Size = init->ring_size;
- dev_priv->ring.virtual_start = DRM(ioremap)(dev->agp->base +
+ dev_priv->ring.virtual_start = DRM(ioremap)(dev->agp->base +
init->ring_start,
init->ring_size, dev);
- if (dev_priv->ring.virtual_start == NULL) {
+ if (dev_priv->ring.virtual_start == NULL) {
dev->dev_private = (void *) dev_priv;
- i810_dma_cleanup(dev);
- DRM_ERROR("can not ioremap virtual address for"
+ i810_dma_cleanup(dev);
+ DRM_ERROR("can not ioremap virtual address for"
" ring buffer\n");
- return -ENOMEM;
+ return -ENOMEM;
}
- dev_priv->ring.tail_mask = dev_priv->ring.Size - 1;
+ dev_priv->ring.tail_mask = dev_priv->ring.Size - 1;
dev_priv->w = init->w;
dev_priv->h = init->h;
@@ -415,33 +415,33 @@ static int i810_dma_initialize(drm_devic
dev_priv->back_di1 = init->back_offset | init->pitch_bits;
dev_priv->zi1 = init->depth_offset | init->pitch_bits;
- /* Program Hardware Status Page */
- dev_priv->hw_status_page =
+ /* Program Hardware Status Page */
+ dev_priv->hw_status_page =
pci_alloc_consistent(dev->pdev, PAGE_SIZE,
&dev_priv->dma_status_page);
- if (!dev_priv->hw_status_page) {
+ if (!dev_priv->hw_status_page) {
dev->dev_private = (void *)dev_priv;
i810_dma_cleanup(dev);
DRM_ERROR("Can not allocate hardware status page\n");
return -ENOMEM;
}
- memset(dev_priv->hw_status_page, 0, PAGE_SIZE);
- DRM_DEBUG("hw status page @ %p\n", dev_priv->hw_status_page);
+ memset(dev_priv->hw_status_page, 0, PAGE_SIZE);
+ DRM_DEBUG("hw status page @ %p\n", dev_priv->hw_status_page);
I810_WRITE(0x02080, dev_priv->dma_status_page);
- DRM_DEBUG("Enabled hardware status page\n");
+ DRM_DEBUG("Enabled hardware status page\n");
- /* Now we need to init our freelist */
+ /* Now we need to init our freelist */
if (i810_freelist_init(dev, dev_priv) != 0) {
dev->dev_private = (void *)dev_priv;
- i810_dma_cleanup(dev);
- DRM_ERROR("Not enough space in the status page for"
+ i810_dma_cleanup(dev);
+ DRM_ERROR("Not enough space in the status page for"
" the freelist\n");
- return -ENOMEM;
+ return -ENOMEM;
}
dev->dev_private = (void *)dev_priv;
- return 0;
+ return 0;
}
/* i810 DRM version 1.1 used a smaller init structure with different
@@ -476,12 +476,12 @@ int i810_dma_init_compat(drm_i810_init_t
/* This is a v1.1 client, fix the params */
DRM_INFO("Using PRE v1.2 init.\n");
- init->pitch_bits = init->h;
- init->pitch = init->w;
- init->h = init->overlay_physical;
- init->w = init->overlay_offset;
- init->overlay_physical = 0;
- init->overlay_offset = 0;
+ init->pitch_bits = init->h;
+ init->pitch = init->w;
+ init->h = init->overlay_physical;
+ init->w = init->overlay_offset;
+ init->overlay_physical = 0;
+ init->overlay_offset = 0;
}
return 0;
@@ -490,55 +490,55 @@ int i810_dma_init_compat(drm_i810_init_t
int i810_dma_init(struct inode *inode, struct file *filp,
unsigned int cmd, unsigned long arg)
{
- drm_file_t *priv = filp->private_data;
- drm_device_t *dev = priv->dev;
- drm_i810_private_t *dev_priv;
- drm_i810_init_t init;
- int retcode = 0;
+ drm_file_t *priv = filp->private_data;
+ drm_device_t *dev = priv->dev;
+ drm_i810_private_t *dev_priv;
+ drm_i810_init_t init;
+ int retcode = 0;
/* Get only the init func */
if (copy_from_user(&init, (void *)arg, sizeof(drm_i810_init_func_t)))
return -EFAULT;
- switch(init.func) {
- case I810_INIT_DMA:
- /* This case is for backward compatibility. It
+ switch(init.func) {
+ case I810_INIT_DMA:
+ /* This case is for backward compatibility. It
* handles XFree 4.1.0 and 4.2.0, and has to
* do some parameter checking as described below.
* It will someday go away.
*/
retcode = i810_dma_init_compat(&init, arg);
- if (retcode)
+ if (retcode)
return retcode;
- dev_priv = DRM(alloc)(sizeof(drm_i810_private_t),
+ dev_priv = DRM(alloc)(sizeof(drm_i810_private_t),
DRM_MEM_DRIVER);
- if (dev_priv == NULL)
- return -ENOMEM;
- retcode = i810_dma_initialize(dev, dev_priv, &init);
+ if (dev_priv == NULL)
+ return -ENOMEM;
+ retcode = i810_dma_initialize(dev, dev_priv, &init);
break;
default:
- case I810_INIT_DMA_1_4:
+ case I810_INIT_DMA_1_4:
DRM_INFO("Using v1.4 init.\n");
- if (copy_from_user(&init, (drm_i810_init_t *)arg,
+ if (copy_from_user(&init, (drm_i810_init_t *)arg,
sizeof(drm_i810_init_t))) {
return -EFAULT;
}
- dev_priv = DRM(alloc)(sizeof(drm_i810_private_t),
+ dev_priv = DRM(alloc)(sizeof(drm_i810_private_t),
DRM_MEM_DRIVER);
if (dev_priv == NULL)
return -ENOMEM;
- retcode = i810_dma_initialize(dev, dev_priv, &init);
+ retcode = i810_dma_initialize(dev, dev_priv, &init);
break;
- case I810_CLEANUP_DMA:
- DRM_INFO("DMA Cleanup\n");
- retcode = i810_dma_cleanup(dev);
- break;
+ case I810_CLEANUP_DMA:
+ DRM_INFO("DMA Cleanup\n");
+ retcode = i810_dma_cleanup(dev);
+ break;
}
- return retcode;
+ return retcode;
}
@@ -552,7 +552,7 @@ int i810_dma_init(struct inode *inode, s
static void i810EmitContextVerified( drm_device_t *dev,
volatile unsigned int *code )
{
- drm_i810_private_t *dev_priv = dev->dev_private;
+ drm_i810_private_t *dev_priv = dev->dev_private;
int i, j = 0;
unsigned int tmp;
RING_LOCALS;
@@ -586,7 +586,7 @@ static void i810EmitContextVerified( drm
static void i810EmitTexVerified( drm_device_t *dev,
volatile unsigned int *code )
{
- drm_i810_private_t *dev_priv = dev->dev_private;
+ drm_i810_private_t *dev_priv = dev->dev_private;
int i, j = 0;
unsigned int tmp;
RING_LOCALS;
@@ -622,7 +622,7 @@ static void i810EmitTexVerified( drm_dev
static void i810EmitDestVerified( drm_device_t *dev,
volatile unsigned int *code )
{
- drm_i810_private_t *dev_priv = dev->dev_private;
+ drm_i810_private_t *dev_priv = dev->dev_private;
unsigned int tmp;
RING_LOCALS;
@@ -658,8 +658,8 @@ static void i810EmitDestVerified( drm_de
static void i810EmitState( drm_device_t *dev )
{
- drm_i810_private_t *dev_priv = dev->dev_private;
- drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv;
+ drm_i810_private_t *dev_priv = dev->dev_private;
+ drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv;
unsigned int dirty = sarea_priv->dirty;
DRM_DEBUG("%s %x\n", __FUNCTION__, dirty);
@@ -693,8 +693,8 @@ static void i810_dma_dispatch_clear( drm
unsigned int clear_color,
unsigned int clear_zval )
{
- drm_i810_private_t *dev_priv = dev->dev_private;
- drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv;
+ drm_i810_private_t *dev_priv = dev->dev_private;
+ drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv;
int nbox = sarea_priv->nbox;
drm_clip_rect_t *pbox = sarea_priv->boxes;
int pitch = dev_priv->pitch;
@@ -703,17 +703,17 @@ static void i810_dma_dispatch_clear( drm
RING_LOCALS;
if ( dev_priv->current_page == 1 ) {
- unsigned int tmp = flags;
-
+ unsigned int tmp = flags;
+
flags &= ~(I810_FRONT | I810_BACK);
if (tmp & I810_FRONT) flags |= I810_BACK;
if (tmp & I810_BACK) flags |= I810_FRONT;
}
- i810_kernel_lost_context(dev);
+ i810_kernel_lost_context(dev);
- if (nbox > I810_NR_SAREA_CLIPRECTS)
- nbox = I810_NR_SAREA_CLIPRECTS;
+ if (nbox > I810_NR_SAREA_CLIPRECTS)
+ nbox = I810_NR_SAREA_CLIPRECTS;
for (i = 0 ; i < nbox ; i++, pbox++) {
unsigned int x = pbox->x1;
@@ -728,7 +728,7 @@ static void i810_dma_dispatch_clear( drm
pbox->y2 > dev_priv->h)
continue;
- if ( flags & I810_FRONT ) {
+ if ( flags & I810_FRONT ) {
BEGIN_LP_RING( 6 );
OUT_RING( BR00_BITBLT_CLIENT |
BR00_OP_COLOR_BLT | 0x3 );
@@ -768,8 +768,8 @@ static void i810_dma_dispatch_clear( drm
static void i810_dma_dispatch_swap( drm_device_t *dev )
{
- drm_i810_private_t *dev_priv = dev->dev_private;
- drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv;
+ drm_i810_private_t *dev_priv = dev->dev_private;
+ drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv;
int nbox = sarea_priv->nbox;
drm_clip_rect_t *pbox = sarea_priv->boxes;
int pitch = dev_priv->pitch;
@@ -779,10 +779,10 @@ static void i810_dma_dispatch_swap( drm_
DRM_DEBUG("swapbuffers\n");
- i810_kernel_lost_context(dev);
+ i810_kernel_lost_context(dev);
- if (nbox > I810_NR_SAREA_CLIPRECTS)
- nbox = I810_NR_SAREA_CLIPRECTS;
+ if (nbox > I810_NR_SAREA_CLIPRECTS)
+ nbox = I810_NR_SAREA_CLIPRECTS;
for (i = 0 ; i < nbox; i++, pbox++)
{
@@ -820,19 +820,19 @@ static void i810_dma_dispatch_vertex(drm
int discard,
int used)
{
- drm_i810_private_t *dev_priv = dev->dev_private;
+ drm_i810_private_t *dev_priv = dev->dev_private;
drm_i810_buf_priv_t *buf_priv = buf->dev_private;
- drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv;
- drm_clip_rect_t *box = sarea_priv->boxes;
- int nbox = sarea_priv->nbox;
+ drm_i810_sarea_t *sarea_priv = dev_priv->sarea_priv;
+ drm_clip_rect_t *box = sarea_priv->boxes;
+ int nbox = sarea_priv->nbox;
unsigned long address = (unsigned long)buf->bus_address;
unsigned long start = address - dev->agp->base;
int i = 0;
- RING_LOCALS;
+ RING_LOCALS;
- i810_kernel_lost_context(dev);
+ i810_kernel_lost_context(dev);
- if (nbox > I810_NR_SAREA_CLIPRECTS)
+ if (nbox > I810_NR_SAREA_CLIPRECTS)
nbox = I810_NR_SAREA_CLIPRECTS;
if (used > 4*1024)
@@ -898,7 +898,7 @@ static void i810_dma_dispatch_vertex(drm
static void i810_dma_dispatch_flip( drm_device_t *dev )
{
- drm_i810_private_t *dev_priv = dev->dev_private;
+ drm_i810_private_t *dev_priv = dev->dev_private;
int pitch = dev_priv->pitch;
RING_LOCALS;
@@ -907,10 +907,10 @@ static void i810_dma_dispatch_flip( drm_
dev_priv->current_page,
dev_priv->sarea_priv->pf_current_page);
- i810_kernel_lost_context(dev);
+ i810_kernel_lost_context(dev);
BEGIN_LP_RING( 2 );
- OUT_RING( INST_PARSER_CLIENT | INST_OP_FLUSH | INST_FLUSH_MAP_CACHE );
+ OUT_RING( INST_PARSER_CLIENT | INST_OP_FLUSH | INST_FLUSH_MAP_CACHE );
OUT_RING( 0 );
ADVANCE_LP_RING();
@@ -945,44 +945,44 @@ static void i810_dma_dispatch_flip( drm_
void i810_dma_quiescent(drm_device_t *dev)
{
- drm_i810_private_t *dev_priv = dev->dev_private;
- RING_LOCALS;
+ drm_i810_private_t *dev_priv = dev->dev_private;
+ RING_LOCALS;
-/* printk("%s\n", __FUNCTION__); */
+/* printk("%s\n", __FUNCTION__); */
- i810_kernel_lost_context(dev);
+ i810_kernel_lost_context(dev);
- BEGIN_LP_RING(4);
- OUT_RING( INST_PARSER_CLIENT | INST_OP_FLUSH | INST_FLUSH_MAP_CACHE );
- OUT_RING( CMD_REPORT_HEAD );
- OUT_RING( 0 );
- OUT_RING( 0 );
- ADVANCE_LP_RING();
+ BEGIN_LP_RING(4);
+ OUT_RING( INST_PARSER_CLIENT | INST_OP_FLUSH | INST_FLUSH_MAP_CACHE );
+ OUT_RING( CMD_REPORT_HEAD );
+ OUT_RING( 0 );
+ OUT_RING( 0 );
+ ADVANCE_LP_RING();
i810_wait_ring( dev, dev_priv->ring.Size - 8 );
}
static int i810_flush_queue(drm_device_t *dev)
{
- drm_i810_private_t *dev_priv = dev->dev_private;
+ drm_i810_private_t *dev_priv = dev->dev_private;
drm_device_dma_t *dma = dev->dma;
- int i, ret = 0;
- RING_LOCALS;
+ int i, ret = 0;
+ RING_LOCALS;
-/* printk("%s\n", __FUNCTION__); */
+/* printk("%s\n", __FUNCTION__); */
- i810_kernel_lost_context(dev);
+ i810_kernel_lost_context(dev);
- BEGIN_LP_RING(2);
- OUT_RING( CMD_REPORT_HEAD );
- OUT_RING( 0 );
- ADVANCE_LP_RING();
+ BEGIN_LP_RING(2);
+ OUT_RING( CMD_REPORT_HEAD );
+ OUT_RING( 0 );
+ ADVANCE_LP_RING();
i810_wait_ring( dev, dev_priv->ring.Size - 8 );
- for (i = 0; i < dma->buf_count; i++) {
- drm_buf_t *buf = dma->buflist[ i ];
- drm_i810_buf_priv_t *buf_priv = buf->dev_private;
+ for (i = 0; i < dma->buf_count; i++) {
+ drm_buf_t *buf = dma->buflist[ i ];
+ drm_i810_buf_priv_t *buf_priv = buf->dev_private;
int used = cmpxchg(buf_priv->in_use, I810_BUF_HARDWARE,
I810_BUF_FREE);
@@ -993,7 +993,7 @@ static int i810_flush_queue(drm_device_t
DRM_DEBUG("still on client\n");
}
- return ret;
+ return ret;
}
/* Must be called with the lock held */
@@ -1005,14 +1005,14 @@ void i810_reclaim_buffers(struct file *f
int i;
if (!dma) return;
- if (!dev->dev_private) return;
+ if (!dev->dev_private) return;
if (!dma->buflist) return;
- i810_flush_queue(dev);
+ i810_flush_queue(dev);
for (i = 0; i < dma->buf_count; i++) {
- drm_buf_t *buf = dma->buflist[ i ];
- drm_i810_buf_priv_t *buf_priv = buf->dev_private;
+ drm_buf_t *buf = dma->buflist[ i ];
+ drm_i810_buf_priv_t *buf_priv = buf->dev_private;
if (buf->filp == filp && buf_priv) {
int used = cmpxchg(buf_priv->in_use, I810_BUF_CLIENT,
@@ -1021,7 +1021,7 @@ void i810_reclaim_buffers(struct file *f
if (used == I810_BUF_CLIENT)
DRM_DEBUG("reclaimed from client\n");
if (buf_priv->currently_mapped == I810_BUF_MAPPED)
- buf_priv->currently_mapped = I810_BUF_UNMAPPED;
+ buf_priv->currently_mapped = I810_BUF_UNMAPPED;
}
}
}
@@ -1029,16 +1029,16 @@ void i810_reclaim_buffers(struct file *f
int i810_flush_ioctl(struct inode *inode, struct file *filp,
unsigned int cmd, unsigned long arg)
{
- drm_file_t *priv = filp->private_data;
- drm_device_t *dev = priv->dev;
+ drm_file_t *priv = filp->private_data;
+ drm_device_t *dev = priv->dev;
if (!_DRM_LOCK_IS_HELD(dev->lock.hw_lock->lock)) {
DRM_ERROR("i810_flush_ioctl called without lock held\n");
return -EINVAL;
}
- i810_flush_queue(dev);
- return 0;
+ i810_flush_queue(dev);
+ return 0;
}
@@ -1048,10 +1048,10 @@ int i810_dma_vertex(struct inode *inode,
drm_file_t *priv = filp->private_data;
drm_device_t *dev = priv->dev;
drm_device_dma_t *dma = dev->dma;
- drm_i810_private_t *dev_priv = (drm_i810_private_t *)dev->dev_private;
- u32 *hw_status = dev_priv->hw_status_page;
- drm_i810_sarea_t *sarea_priv = (drm_i810_sarea_t *)
- dev_priv->sarea_priv;
+ drm_i810_private_t *dev_priv = (drm_i810_private_t *)dev->dev_private;
+ u32 *hw_status = dev_priv->hw_status_page;
+ drm_i810_sarea_t *sarea_priv = (drm_i810_sarea_t *)
+ dev_priv->sarea_priv;
drm_i810_vertex_t vertex;
if (copy_from_user(&vertex, (drm_i810_vertex_t *)arg, sizeof(vertex)))
@@ -1072,10 +1072,10 @@ int i810_dma_vertex(struct inode *inode,
dma->buflist[ vertex.idx ],
vertex.discard, vertex.used );
- atomic_add(vertex.used, &dev->counts[_DRM_STAT_SECONDARY]);
+ atomic_add(vertex.used, &dev->counts[_DRM_STAT_SECONDARY]);
atomic_inc(&dev->counts[_DRM_STAT_DMA]);
sarea_priv->last_enqueue = dev_priv->counter-1;
- sarea_priv->last_dispatch = (int) hw_status[5];
+ sarea_priv->last_dispatch = (int) hw_status[5];
return 0;
}
@@ -1089,7 +1089,7 @@ int i810_clear_bufs(struct inode *inode,
drm_device_t *dev = priv->dev;
drm_i810_clear_t clear;
- if (copy_from_user(&clear, (drm_i810_clear_t *)arg, sizeof(clear)))
+ if (copy_from_user(&clear, (drm_i810_clear_t *)arg, sizeof(clear)))
return -EFAULT;
if (!_DRM_LOCK_IS_HELD(dev->lock.hw_lock->lock)) {
@@ -1097,15 +1097,15 @@ int i810_clear_bufs(struct inode *inode,
return -EINVAL;
}
- /* GH: Someone's doing nasty things... */
- if (!dev->dev_private) {
- return -EINVAL;
- }
+ /* GH: Someone's doing nasty things... */
+ if (!dev->dev_private) {
+ return -EINVAL;
+ }
i810_dma_dispatch_clear( dev, clear.flags,
clear.clear_color,
clear.clear_depth );
- return 0;
+ return 0;
}
int i810_swap_bufs(struct inode *inode, struct file *filp,
@@ -1122,20 +1122,20 @@ int i810_swap_bufs(struct inode *inode,
}
i810_dma_dispatch_swap( dev );
- return 0;
+ return 0;
}
int i810_getage(struct inode *inode, struct file *filp, unsigned int cmd,
unsigned long arg)
{
- drm_file_t *priv = filp->private_data;
+ drm_file_t *priv = filp->private_data;
drm_device_t *dev = priv->dev;
- drm_i810_private_t *dev_priv = (drm_i810_private_t *)dev->dev_private;
- u32 *hw_status = dev_priv->hw_status_page;
- drm_i810_sarea_t *sarea_priv = (drm_i810_sarea_t *)
- dev_priv->sarea_priv;
+ drm_i810_private_t *dev_priv = (drm_i810_private_t *)dev->dev_private;
+ u32 *hw_status = dev_priv->hw_status_page;
+ drm_i810_sarea_t *sarea_priv = (drm_i810_sarea_t *)
+ dev_priv->sarea_priv;
- sarea_priv->last_dispatch = (int) hw_status[5];
+ sarea_priv->last_dispatch = (int) hw_status[5];
return 0;
}
@@ -1146,12 +1146,12 @@ int i810_getbuf(struct inode *inode, str
drm_device_t *dev = priv->dev;
int retcode = 0;
drm_i810_dma_t d;
- drm_i810_private_t *dev_priv = (drm_i810_private_t *)dev->dev_private;
- u32 *hw_status = dev_priv->hw_status_page;
- drm_i810_sarea_t *sarea_priv = (drm_i810_sarea_t *)
- dev_priv->sarea_priv;
+ drm_i810_private_t *dev_priv = (drm_i810_private_t *)dev->dev_private;
+ u32 *hw_status = dev_priv->hw_status_page;
+ drm_i810_sarea_t *sarea_priv = (drm_i810_sarea_t *)
+ dev_priv->sarea_priv;
- if (copy_from_user(&d, (drm_i810_dma_t *)arg, sizeof(d)))
+ if (copy_from_user(&d, (drm_i810_dma_t *)arg, sizeof(d)))
return -EFAULT;
if (!_DRM_LOCK_IS_HELD(dev->lock.hw_lock->lock)) {
@@ -1168,7 +1168,7 @@ int i810_getbuf(struct inode *inode, str
if (copy_to_user((drm_dma_t *)arg, &d, sizeof(d)))
return -EFAULT;
- sarea_priv->last_dispatch = (int) hw_status[5];
+ sarea_priv->last_dispatch = (int) hw_status[5];
return retcode;
}
@@ -1384,5 +1384,5 @@ int i810_flip_bufs(struct inode *inode,
i810_do_init_pageflip( dev );
i810_dma_dispatch_flip( dev );
- return 0;
+ return 0;
}
--- diff/drivers/char/drm/i830.h 2003-05-21 10:50:14.000000000 +0000
+++ source/drivers/char/drm/i830.h 2004-03-16 11:27:48.033885400 +0000
@@ -77,6 +77,13 @@
[DRM_IOCTL_NR(DRM_IOCTL_I830_GETPARAM)] = { i830_getparam, 1, 0 }, \
[DRM_IOCTL_NR(DRM_IOCTL_I830_SETPARAM)] = { i830_setparam, 1, 0 }
+#define DRIVER_PCI_IDS \
+ {0x8086, 0x3577, 0, "Intel i830M GMCH"}, \
+ {0x8086, 0x2562, 0, "Intel i845G GMCH"}, \
+ {0x8086, 0x3582, 0, "Intel i852GM/i855GM GMCH"}, \
+ {0x8086, 0x2572, 0, "Intel i865G GMCH"}, \
+ {0, 0, 0, NULL}
+
#define __HAVE_COUNTERS 4
#define __HAVE_COUNTER6 _DRM_STAT_IRQ
#define __HAVE_COUNTER7 _DRM_STAT_PRIMARY
@@ -115,10 +122,10 @@
#define USE_IRQS 0
#if USE_IRQS
-#define __HAVE_DMA_IRQ 1
+#define __HAVE_IRQ 1
#define __HAVE_SHARED_IRQ 1
#else
-#define __HAVE_DMA_IRQ 0
+#define __HAVE_IRQ 0
#endif
--- diff/drivers/char/drm/i830_dma.c 2003-06-09 13:18:18.000000000 +0000
+++ source/drivers/char/drm/i830_dma.c 2004-03-16 11:27:48.033885400 +0000
@@ -231,12 +231,12 @@ int i830_dma_cleanup(drm_device_t *dev)
{
drm_device_dma_t *dma = dev->dma;
-#if _HAVE_DMA_IRQ
+#if __HAVE_IRQ
/* Make sure interrupts are disabled here because the uninstall ioctl
* may not have been called from userspace and after dev_private
* is freed, it's too late.
*/
- if (dev->irq) DRM(irq_uninstall)(dev);
+ if (dev->irq_enabled) DRM(irq_uninstall)(dev);
#endif
if (dev->dev_private) {
@@ -1539,7 +1539,7 @@ int i830_getparam( struct inode *inode,
switch( param.param ) {
case I830_PARAM_IRQ_ACTIVE:
- value = dev->irq ? 1 : 0;
+ value = dev->irq_enabled;
break;
default:
return -EINVAL;
--- diff/drivers/char/drm/i830_drv.c 2003-05-21 10:50:14.000000000 +0000
+++ source/drivers/char/drm/i830_drv.c 2004-03-16 11:27:48.033885400 +0000
@@ -50,6 +50,7 @@
#include "drm_fops.h"
#include "drm_init.h"
#include "drm_ioctl.h"
+#include "drm_irq.h"
#include "drm_lock.h"
#include "drm_memory.h"
#include "drm_proc.h"
--- diff/drivers/char/drm/i830_irq.c 2003-09-30 14:46:12.000000000 +0000
+++ source/drivers/char/drm/i830_irq.c 2004-03-16 11:27:48.034885248 +0000
@@ -35,7 +35,7 @@
#include <linux/delay.h>
-irqreturn_t DRM(dma_service)( DRM_IRQ_ARGS )
+irqreturn_t DRM(irq_handler)( DRM_IRQ_ARGS )
{
drm_device_t *dev = (drm_device_t *)arg;
drm_i830_private_t *dev_priv = (drm_i830_private_t *)dev->dev_private;
--- diff/drivers/char/drm/mga.h 2003-05-21 10:50:14.000000000 +0000
+++ source/drivers/char/drm/mga.h 2004-03-16 11:27:48.050882816 +0000
@@ -64,6 +64,12 @@
[DRM_IOCTL_NR(DRM_IOCTL_MGA_BLIT)] = { mga_dma_blit, 1, 0 }, \
[DRM_IOCTL_NR(DRM_IOCTL_MGA_GETPARAM)]= { mga_getparam, 1, 0 },
+#define DRIVER_PCI_IDS \
+ {0x102b, 0x0521, 0, "Matrox G200 (AGP)"}, \
+ {0x102b, 0x0525, 0, "Matrox G400/G450 (AGP)"}, \
+ {0x102b, 0x2527, 0, "Matrox G550 (AGP)"}, \
+ {0, 0, 0, NULL}
+
#define __HAVE_COUNTERS 3
#define __HAVE_COUNTER6 _DRM_STAT_IRQ
#define __HAVE_COUNTER7 _DRM_STAT_PRIMARY
@@ -78,7 +84,7 @@
/* DMA customization:
*/
#define __HAVE_DMA 1
-#define __HAVE_DMA_IRQ 1
+#define __HAVE_IRQ 1
#define __HAVE_VBL_IRQ 1
#define __HAVE_SHARED_IRQ 1
--- diff/drivers/char/drm/mga_dma.c 2003-06-09 13:18:18.000000000 +0000
+++ source/drivers/char/drm/mga_dma.c 2004-03-16 11:27:48.049882968 +0000
@@ -500,14 +500,6 @@ static int mga_do_init_dma( drm_device_t
return DRM_ERR(EINVAL);
}
- DRM_FIND_MAP( dev_priv->fb, init->fb_offset );
- if(!dev_priv->fb) {
- DRM_ERROR( "failed to find framebuffer!\n" );
- /* Assign dev_private so we can do cleanup. */
- dev->dev_private = (void *)dev_priv;
- mga_do_cleanup_dma( dev );
- return DRM_ERR(EINVAL);
- }
DRM_FIND_MAP( dev_priv->mmio, init->mmio_offset );
if(!dev_priv->mmio) {
DRM_ERROR( "failed to find mmio region!\n" );
@@ -639,12 +631,12 @@ int mga_do_cleanup_dma( drm_device_t *de
{
DRM_DEBUG( "\n" );
-#if _HAVE_DMA_IRQ
+#if __HAVE_IRQ
/* Make sure interrupts are disabled here because the uninstall ioctl
* may not have been called from userspace and after dev_private
* is freed, it's too late.
*/
- if ( dev->irq ) DRM(irq_uninstall)(dev);
+ if ( dev->irq_enabled ) DRM(irq_uninstall)(dev);
#endif
if ( dev->dev_private ) {
--- diff/drivers/char/drm/mga_drv.c 2002-10-16 03:27:56.000000000 +0000
+++ source/drivers/char/drm/mga_drv.c 2004-03-16 11:27:48.049882968 +0000
@@ -45,6 +45,7 @@
#include "drm_fops.h"
#include "drm_init.h"
#include "drm_ioctl.h"
+#include "drm_irq.h"
#include "drm_lock.h"
#include "drm_memory.h"
#include "drm_proc.h"
--- diff/drivers/char/drm/mga_drv.h 2003-06-09 13:18:18.000000000 +0000
+++ source/drivers/char/drm/mga_drv.h 2004-03-16 11:27:48.050882816 +0000
@@ -91,7 +91,6 @@ typedef struct drm_mga_private {
unsigned int texture_size;
drm_local_map_t *sarea;
- drm_local_map_t *fb;
drm_local_map_t *mmio;
drm_local_map_t *status;
drm_local_map_t *warp;
--- diff/drivers/char/drm/mga_irq.c 2003-05-21 10:50:14.000000000 +0000
+++ source/drivers/char/drm/mga_irq.c 2004-03-16 11:27:48.051882664 +0000
@@ -36,7 +36,7 @@
#include "mga_drm.h"
#include "mga_drv.h"
-irqreturn_t mga_dma_service( DRM_IRQ_ARGS )
+irqreturn_t mga_irq_handler( DRM_IRQ_ARGS )
{
drm_device_t *dev = (drm_device_t *) arg;
drm_mga_private_t *dev_priv =
--- diff/drivers/char/drm/r128.h 2003-08-26 09:00:52.000000000 +0000
+++ source/drivers/char/drm/r128.h 2004-03-16 11:27:48.054882208 +0000
@@ -79,6 +79,46 @@
[DRM_IOCTL_NR(DRM_IOCTL_R128_INDIRECT)] = { r128_cce_indirect, 1, 1 }, \
[DRM_IOCTL_NR(DRM_IOCTL_R128_GETPARAM)] = { r128_getparam, 1, 0 },
+#define DRIVER_PCI_IDS \
+ {0x1002, 0x4c45, 0, "ATI Rage 128 Mobility LE (PCI)"}, \
+ {0x1002, 0x4c46, 0, "ATI Rage 128 Mobility LF (AGP)"}, \
+ {0x1002, 0x4d46, 0, "ATI Rage 128 Mobility MF (AGP)"}, \
+ {0x1002, 0x4d4c, 0, "ATI Rage 128 Mobility ML (AGP)"}, \
+ {0x1002, 0x5041, 0, "ATI Rage 128 Pro PA (PCI)"}, \
+ {0x1002, 0x5042, 0, "ATI Rage 128 Pro PB (AGP)"}, \
+ {0x1002, 0x5043, 0, "ATI Rage 128 Pro PC (AGP)"}, \
+ {0x1002, 0x5044, 0, "ATI Rage 128 Pro PD (PCI)"}, \
+ {0x1002, 0x5045, 0, "ATI Rage 128 Pro PE (AGP)"}, \
+ {0x1002, 0x5046, 0, "ATI Rage 128 Pro PF (AGP)"}, \
+ {0x1002, 0x5047, 0, "ATI Rage 128 Pro PG (PCI)"}, \
+ {0x1002, 0x5048, 0, "ATI Rage 128 Pro PH (AGP)"}, \
+ {0x1002, 0x5049, 0, "ATI Rage 128 Pro PI (AGP)"}, \
+ {0x1002, 0x504A, 0, "ATI Rage 128 Pro PJ (PCI)"}, \
+ {0x1002, 0x504B, 0, "ATI Rage 128 Pro PK (AGP)"}, \
+ {0x1002, 0x504C, 0, "ATI Rage 128 Pro PL (AGP)"}, \
+ {0x1002, 0x504D, 0, "ATI Rage 128 Pro PM (PCI)"}, \
+ {0x1002, 0x504E, 0, "ATI Rage 128 Pro PN (AGP)"}, \
+ {0x1002, 0x504F, 0, "ATI Rage 128 Pro PO (AGP)"}, \
+ {0x1002, 0x5050, 0, "ATI Rage 128 Pro PP (PCI)"}, \
+ {0x1002, 0x5051, 0, "ATI Rage 128 Pro PQ (AGP)"}, \
+ {0x1002, 0x5052, 0, "ATI Rage 128 Pro PR (PCI)"}, \
+ {0x1002, 0x5053, 0, "ATI Rage 128 Pro PS (PCI)"}, \
+ {0x1002, 0x5054, 0, "ATI Rage 128 Pro PT (AGP)"}, \
+ {0x1002, 0x5055, 0, "ATI Rage 128 Pro PU (AGP)"}, \
+ {0x1002, 0x5056, 0, "ATI Rage 128 Pro PV (PCI)"}, \
+ {0x1002, 0x5057, 0, "ATI Rage 128 Pro PW (AGP)"}, \
+ {0x1002, 0x5058, 0, "ATI Rage 128 Pro PX (AGP)"}, \
+ {0x1002, 0x5245, 0, "ATI Rage 128 RE (PCI)"}, \
+ {0x1002, 0x5246, 0, "ATI Rage 128 RF (AGP)"}, \
+ {0x1002, 0x5247, 0, "ATI Rage 128 RG (AGP)"}, \
+ {0x1002, 0x524b, 0, "ATI Rage 128 RK (PCI)"}, \
+ {0x1002, 0x524c, 0, "ATI Rage 128 RL (AGP)"}, \
+ {0x1002, 0x534d, 0, "ATI Rage 128 SM (AGP)"}, \
+ {0x1002, 0x5446, 0, "ATI Rage 128 Pro Ultra TF (AGP)"}, \
+ {0x1002, 0x544C, 0, "ATI Rage 128 Pro Ultra TL (AGP)"}, \
+ {0x1002, 0x5452, 0, "ATI Rage 128 Pro Ultra TR (AGP)"}, \
+ {0, 0, 0, NULL}
+
/* Driver customization:
*/
#define DRIVER_PRERELEASE() do { \
@@ -97,7 +137,7 @@
/* DMA customization:
*/
#define __HAVE_DMA 1
-#define __HAVE_DMA_IRQ 1
+#define __HAVE_IRQ 1
#define __HAVE_VBL_IRQ 1
#define __HAVE_SHARED_IRQ 1
--- diff/drivers/char/drm/r128_cce.c 2003-09-30 14:46:12.000000000 +0000
+++ source/drivers/char/drm/r128_cce.c 2004-03-16 11:27:48.052882512 +0000
@@ -212,7 +212,7 @@ int r128_do_cce_idle( drm_r128_private_t
int i;
for ( i = 0 ; i < dev_priv->usec_timeout ; i++ ) {
- if ( GET_RING_HEAD( &dev_priv->ring ) == dev_priv->ring.tail ) {
+ if ( GET_RING_HEAD( dev_priv ) == dev_priv->ring.tail ) {
int pm4stat = R128_READ( R128_PM4_STAT );
if ( ( (pm4stat & R128_PM4_FIFOCNT_MASK) >=
dev_priv->cce_fifo_size ) &&
@@ -238,7 +238,8 @@ static void r128_do_cce_start( drm_r128_
r128_do_wait_for_idle( dev_priv );
R128_WRITE( R128_PM4_BUFFER_CNTL,
- dev_priv->cce_mode | dev_priv->ring.size_l2qw );
+ dev_priv->cce_mode | dev_priv->ring.size_l2qw
+ | R128_PM4_BUFFER_CNTL_NOUPDATE );
R128_READ( R128_PM4_BUFFER_ADDR ); /* as per the sample code */
R128_WRITE( R128_PM4_MICRO_CNTL, R128_PM4_MICRO_FREERUN );
@@ -253,7 +254,6 @@ static void r128_do_cce_reset( drm_r128_
{
R128_WRITE( R128_PM4_BUFFER_DL_WPTR, 0 );
R128_WRITE( R128_PM4_BUFFER_DL_RPTR, 0 );
- SET_RING_HEAD( &dev_priv->ring, 0 );
dev_priv->ring.tail = 0;
}
@@ -264,7 +264,8 @@ static void r128_do_cce_reset( drm_r128_
static void r128_do_cce_stop( drm_r128_private_t *dev_priv )
{
R128_WRITE( R128_PM4_MICRO_CNTL, 0 );
- R128_WRITE( R128_PM4_BUFFER_CNTL, R128_PM4_NONPM4 );
+ R128_WRITE( R128_PM4_BUFFER_CNTL,
+ R128_PM4_NONPM4 | R128_PM4_BUFFER_CNTL_NOUPDATE );
dev_priv->cce_running = 0;
}
@@ -333,26 +334,6 @@ static void r128_cce_init_ring_buffer( d
R128_WRITE( R128_PM4_BUFFER_DL_WPTR, 0 );
R128_WRITE( R128_PM4_BUFFER_DL_RPTR, 0 );
- /* DL_RPTR_ADDR is a physical address in AGP space. */
- SET_RING_HEAD( &dev_priv->ring, 0 );
-
- if ( !dev_priv->is_pci ) {
- R128_WRITE( R128_PM4_BUFFER_DL_RPTR_ADDR,
- dev_priv->ring_rptr->offset );
- } else {
- drm_sg_mem_t *entry = dev->sg;
- unsigned long tmp_ofs, page_ofs;
-
- tmp_ofs = dev_priv->ring_rptr->offset - dev->sg->handle;
- page_ofs = tmp_ofs >> PAGE_SHIFT;
-
- R128_WRITE( R128_PM4_BUFFER_DL_RPTR_ADDR,
- entry->busaddr[page_ofs]);
- DRM_DEBUG( "ring rptr: offset=0x%08lx handle=0x%08lx\n",
- (unsigned long) entry->busaddr[page_ofs],
- entry->handle + tmp_ofs );
- }
-
/* Set watermark control */
R128_WRITE( R128_PM4_BUFFER_WM_CNTL,
((R128_WATERMARK_L/4) << R128_WMA_SHIFT)
@@ -486,13 +467,6 @@ static int r128_do_init_cce( drm_device_
return DRM_ERR(EINVAL);
}
- DRM_FIND_MAP( dev_priv->fb, init->fb_offset );
- if(!dev_priv->fb) {
- DRM_ERROR("could not find framebuffer!\n");
- dev->dev_private = (void *)dev_priv;
- r128_do_cleanup_cce( dev );
- return DRM_ERR(EINVAL);
- }
DRM_FIND_MAP( dev_priv->mmio, init->mmio_offset );
if(!dev_priv->mmio) {
DRM_ERROR("could not find mmio region!\n");
@@ -567,9 +541,6 @@ static int r128_do_init_cce( drm_device_
#endif
dev_priv->cce_buffers_offset = dev->sg->handle;
- dev_priv->ring.head = ((__volatile__ u32 *)
- dev_priv->ring_rptr->handle);
-
dev_priv->ring.start = (u32 *)dev_priv->cce_ring->handle;
dev_priv->ring.end = ((u32 *)dev_priv->cce_ring->handle
+ init->ring_size / sizeof(u32));
@@ -580,7 +551,6 @@ static int r128_do_init_cce( drm_device_
(dev_priv->ring.size / sizeof(u32)) - 1;
dev_priv->ring.high_mark = 128;
- dev_priv->ring.ring_rptr = dev_priv->ring_rptr;
dev_priv->sarea_priv->last_frame = 0;
R128_WRITE( R128_LAST_FRAME_REG, dev_priv->sarea_priv->last_frame );
@@ -589,8 +559,9 @@ static int r128_do_init_cce( drm_device_
R128_WRITE( R128_LAST_DISPATCH_REG,
dev_priv->sarea_priv->last_dispatch );
-#if __REALLY_HAVE_SG
+#if __REALLY_HAVE_AGP
if ( dev_priv->is_pci ) {
+#endif
if (!DRM(ati_pcigart_init)( dev, &dev_priv->phys_pci_gart,
&dev_priv->bus_pci_gart) ) {
DRM_ERROR( "failed to init PCI GART!\n" );
@@ -599,6 +570,7 @@ static int r128_do_init_cce( drm_device_
return DRM_ERR(ENOMEM);
}
R128_WRITE( R128_PCI_GART_PAGE, dev_priv->bus_pci_gart );
+#if __REALLY_HAVE_AGP
}
#endif
@@ -615,12 +587,12 @@ static int r128_do_init_cce( drm_device_
int r128_do_cleanup_cce( drm_device_t *dev )
{
-#if _HAVE_DMA_IRQ
+#if __HAVE_IRQ
/* Make sure interrupts are disabled here because the uninstall ioctl
* may not have been called from userspace and after dev_private
* is freed, it's too late.
*/
- if ( dev->irq ) DRM(irq_uninstall)(dev);
+ if ( dev->irq_enabled ) DRM(irq_uninstall)(dev);
#endif
if ( dev->dev_private ) {
@@ -901,7 +873,7 @@ int r128_wait_ring( drm_r128_private_t *
int i;
for ( i = 0 ; i < dev_priv->usec_timeout ; i++ ) {
- r128_update_ring_snapshot( ring );
+ r128_update_ring_snapshot( dev_priv );
if ( ring->space >= n )
return 0;
DRM_UDELAY( 1 );
--- diff/drivers/char/drm/r128_drv.c 2002-10-16 03:27:48.000000000 +0000
+++ source/drivers/char/drm/r128_drv.c 2004-03-16 11:27:48.052882512 +0000
@@ -47,6 +47,7 @@
#include "drm_fops.h"
#include "drm_init.h"
#include "drm_ioctl.h"
+#include "drm_irq.h"
#include "drm_lock.h"
#include "drm_memory.h"
#include "drm_proc.h"
--- diff/drivers/char/drm/r128_drv.h 2003-08-20 13:16:27.000000000 +0000
+++ source/drivers/char/drm/r128_drv.h 2004-03-16 11:27:48.053882360 +0000
@@ -34,8 +34,7 @@
#ifndef __R128_DRV_H__
#define __R128_DRV_H__
-#define GET_RING_HEAD(ring) DRM_READ32( (ring)->ring_rptr, 0 ) /* (ring)->head */
-#define SET_RING_HEAD(ring,val) DRM_WRITE32( (ring)->ring_rptr, 0, (val) ) /* (ring)->head */
+#define GET_RING_HEAD(dev_priv) R128_READ( R128_PM4_BUFFER_DL_RPTR )
typedef struct drm_r128_freelist {
unsigned int age;
@@ -50,13 +49,11 @@ typedef struct drm_r128_ring_buffer {
int size;
int size_l2qw;
- volatile u32 *head;
u32 tail;
u32 tail_mask;
int space;
int high_mark;
- drm_local_map_t *ring_rptr;
} drm_r128_ring_buffer_t;
typedef struct drm_r128_private {
@@ -100,7 +97,6 @@ typedef struct drm_r128_private {
u32 span_pitch_offset_c;
drm_local_map_t *sarea;
- drm_local_map_t *fb;
drm_local_map_t *mmio;
drm_local_map_t *cce_ring;
drm_local_map_t *ring_rptr;
@@ -132,14 +128,6 @@ extern drm_buf_t *r128_freelist_get( drm
extern int r128_wait_ring( drm_r128_private_t *dev_priv, int n );
-static __inline__ void
-r128_update_ring_snapshot( drm_r128_ring_buffer_t *ring )
-{
- ring->space = (GET_RING_HEAD( ring ) - ring->tail) * sizeof(u32);
- if ( ring->space <= 0 )
- ring->space += ring->size;
-}
-
extern int r128_do_cce_idle( drm_r128_private_t *dev_priv );
extern int r128_do_cleanup_cce( drm_device_t *dev );
extern int r128_do_cleanup_pageflip( drm_device_t *dev );
@@ -279,6 +267,7 @@ extern int r128_cce_indirect( DRM_IOCTL_
# define R128_PM4_64PIO_64VCBM_64INDBM (7 << 28)
# define R128_PM4_64BM_64VCBM_64INDBM (8 << 28)
# define R128_PM4_64PIO_64VCPIO_64INDPIO (15 << 28)
+# define R128_PM4_BUFFER_CNTL_NOUPDATE (1 << 27)
#define R128_PM4_BUFFER_WM_CNTL 0x0708
# define R128_WMA_SHIFT 0
@@ -403,6 +392,15 @@ extern int R128_READ_PLL(drm_device_t *d
(pkt) | ((n) << 16))
+static __inline__ void
+r128_update_ring_snapshot( drm_r128_private_t *dev_priv )
+{
+ drm_r128_ring_buffer_t *ring = &dev_priv->ring;
+ ring->space = (GET_RING_HEAD( dev_priv ) - ring->tail) * sizeof(u32);
+ if ( ring->space <= 0 )
+ ring->space += ring->size;
+}
+
/* ================================================================
* Misc helper macros
*/
@@ -412,7 +410,7 @@ do { \
drm_r128_ring_buffer_t *ring = &dev_priv->ring; int i; \
if ( ring->space < ring->high_mark ) { \
for ( i = 0 ; i < dev_priv->usec_timeout ; i++ ) { \
- r128_update_ring_snapshot( ring ); \
+ r128_update_ring_snapshot( dev_priv ); \
if ( ring->space >= ring->high_mark ) \
goto __ring_space_done; \
DRM_UDELAY(1); \
@@ -445,17 +443,10 @@ do { \
* Ring control
*/
-#if defined(__powerpc__)
-#define r128_flush_write_combine() (void) GET_RING_HEAD( &dev_priv->ring )
-#else
-#define r128_flush_write_combine() DRM_WRITEMEMORYBARRIER()
-#endif
-
-
#define R128_VERBOSE 0
#define RING_LOCALS \
- int write; unsigned int tail_mask; volatile u32 *ring;
+ int write, _nr; unsigned int tail_mask; volatile u32 *ring;
#define BEGIN_RING( n ) do { \
if ( R128_VERBOSE ) { \
@@ -463,9 +454,10 @@ do { \
(n), __FUNCTION__ ); \
} \
if ( dev_priv->ring.space <= (n) * sizeof(u32) ) { \
+ COMMIT_RING(); \
r128_wait_ring( dev_priv, (n) * sizeof(u32) ); \
} \
- dev_priv->ring.space -= (n) * sizeof(u32); \
+ _nr = n; dev_priv->ring.space -= (n) * sizeof(u32); \
ring = dev_priv->ring.start; \
write = dev_priv->ring.tail; \
tail_mask = dev_priv->ring.tail_mask; \
@@ -488,9 +480,23 @@ do { \
dev_priv->ring.start, \
write * sizeof(u32) ); \
} \
- r128_flush_write_combine(); \
- dev_priv->ring.tail = write; \
- R128_WRITE( R128_PM4_BUFFER_DL_WPTR, write ); \
+ if (((dev_priv->ring.tail + _nr) & tail_mask) != write) { \
+ DRM_ERROR( \
+ "ADVANCE_RING(): mismatch: nr: %x write: %x line: %d\n", \
+ ((dev_priv->ring.tail + _nr) & tail_mask), \
+ write, __LINE__); \
+ } else \
+ dev_priv->ring.tail = write; \
+} while (0)
+
+#define COMMIT_RING() do { \
+ if ( R128_VERBOSE ) { \
+ DRM_INFO( "COMMIT_RING() tail=0x%06x\n", \
+ dev_priv->ring.tail ); \
+ } \
+ DRM_MEMORYBARRIER(); \
+ R128_WRITE( R128_PM4_BUFFER_DL_WPTR, dev_priv->ring.tail ); \
+ R128_READ( R128_PM4_BUFFER_DL_WPTR ); \
} while (0)
#define OUT_RING( x ) do { \
--- diff/drivers/char/drm/r128_irq.c 2003-05-21 10:50:14.000000000 +0000
+++ source/drivers/char/drm/r128_irq.c 2004-03-16 11:27:48.054882208 +0000
@@ -36,7 +36,7 @@
#include "r128_drm.h"
#include "r128_drv.h"
-irqreturn_t r128_dma_service( DRM_IRQ_ARGS )
+irqreturn_t r128_irq_handler( DRM_IRQ_ARGS )
{
drm_device_t *dev = (drm_device_t *) arg;
drm_r128_private_t *dev_priv =
--- diff/drivers/char/drm/r128_state.c 2004-03-11 10:20:23.000000000 +0000
+++ source/drivers/char/drm/r128_state.c 2004-03-16 11:27:48.056881904 +0000
@@ -45,7 +45,7 @@ static void r128_emit_clip_rects( drm_r1
RING_LOCALS;
DRM_DEBUG( " %s\n", __FUNCTION__ );
- BEGIN_RING( 17 );
+ BEGIN_RING( (count < 3? count: 3) * 5 + 2 );
if ( count >= 1 ) {
OUT_RING( CCE_PACKET0( R128_AUX1_SC_LEFT, 3 ) );
@@ -1280,6 +1280,7 @@ int r128_cce_clear( DRM_IOCTL_ARGS )
sarea_priv->nbox = R128_NR_SAREA_CLIPRECTS;
r128_cce_dispatch_clear( dev, &clear );
+ COMMIT_RING();
/* Make sure we restore the 3D state next time.
*/
@@ -1315,8 +1316,10 @@ int r128_do_cleanup_pageflip( drm_device
R128_WRITE( R128_CRTC_OFFSET, dev_priv->crtc_offset );
R128_WRITE( R128_CRTC_OFFSET_CNTL, dev_priv->crtc_offset_cntl );
- if (dev_priv->current_page != 0)
+ if (dev_priv->current_page != 0) {
r128_cce_dispatch_flip( dev );
+ COMMIT_RING();
+ }
dev_priv->page_flipping = 0;
return 0;
@@ -1341,6 +1344,7 @@ int r128_cce_flip( DRM_IOCTL_ARGS )
r128_cce_dispatch_flip( dev );
+ COMMIT_RING();
return 0;
}
@@ -1362,6 +1366,7 @@ int r128_cce_swap( DRM_IOCTL_ARGS )
dev_priv->sarea_priv->dirty |= (R128_UPLOAD_CONTEXT |
R128_UPLOAD_MASKS);
+ COMMIT_RING();
return 0;
}
@@ -1421,6 +1426,7 @@ int r128_cce_vertex( DRM_IOCTL_ARGS )
r128_cce_dispatch_vertex( dev, buf );
+ COMMIT_RING();
return 0;
}
@@ -1492,6 +1498,7 @@ int r128_cce_indices( DRM_IOCTL_ARGS )
r128_cce_dispatch_indices( dev, buf, elts.start, elts.end, count );
+ COMMIT_RING();
return 0;
}
@@ -1501,6 +1508,7 @@ int r128_cce_blit( DRM_IOCTL_ARGS )
drm_device_dma_t *dma = dev->dma;
drm_r128_private_t *dev_priv = dev->dev_private;
drm_r128_blit_t blit;
+ int ret;
LOCK_TEST_WITH_RETURN( dev, filp );
@@ -1518,7 +1526,10 @@ int r128_cce_blit( DRM_IOCTL_ARGS )
RING_SPACE_TEST_WITH_RETURN( dev_priv );
VB_AGE_TEST_WITH_RETURN( dev_priv );
- return r128_cce_dispatch_blit( filp, dev, &blit );
+ ret = r128_cce_dispatch_blit( filp, dev, &blit );
+
+ COMMIT_RING();
+ return ret;
}
int r128_cce_depth( DRM_IOCTL_ARGS )
@@ -1526,6 +1537,7 @@ int r128_cce_depth( DRM_IOCTL_ARGS )
DRM_DEVICE;
drm_r128_private_t *dev_priv = dev->dev_private;
drm_r128_depth_t depth;
+ int ret;
LOCK_TEST_WITH_RETURN( dev, filp );
@@ -1534,18 +1546,20 @@ int r128_cce_depth( DRM_IOCTL_ARGS )
RING_SPACE_TEST_WITH_RETURN( dev_priv );
+ ret = DRM_ERR(EINVAL);
switch ( depth.func ) {
case R128_WRITE_SPAN:
- return r128_cce_dispatch_write_span( dev, &depth );
+ ret = r128_cce_dispatch_write_span( dev, &depth );
case R128_WRITE_PIXELS:
- return r128_cce_dispatch_write_pixels( dev, &depth );
+ ret = r128_cce_dispatch_write_pixels( dev, &depth );
case R128_READ_SPAN:
- return r128_cce_dispatch_read_span( dev, &depth );
+ ret = r128_cce_dispatch_read_span( dev, &depth );
case R128_READ_PIXELS:
- return r128_cce_dispatch_read_pixels( dev, &depth );
+ ret = r128_cce_dispatch_read_pixels( dev, &depth );
}
- return DRM_ERR(EINVAL);
+ COMMIT_RING();
+ return ret;
}
int r128_cce_stipple( DRM_IOCTL_ARGS )
@@ -1568,6 +1582,7 @@ int r128_cce_stipple( DRM_IOCTL_ARGS )
r128_cce_dispatch_stipple( dev, mask );
+ COMMIT_RING();
return 0;
}
@@ -1643,6 +1658,7 @@ int r128_cce_indirect( DRM_IOCTL_ARGS )
*/
r128_cce_dispatch_indirect( dev, buf, indirect.start, indirect.end );
+ COMMIT_RING();
return 0;
}
--- diff/drivers/char/drm/radeon.h 2003-09-30 14:46:12.000000000 +0000
+++ source/drivers/char/drm/radeon.h 2004-03-16 11:27:48.061881144 +0000
@@ -51,7 +51,7 @@
#define DRIVER_DATE "20020828"
#define DRIVER_MAJOR 1
-#define DRIVER_MINOR 9
+#define DRIVER_MINOR 10
#define DRIVER_PATCHLEVEL 0
/* Interface history:
@@ -81,6 +81,9 @@
* Add 'GET' queries for starting additional clients on different VT's.
* 1.9 - Add DRM_IOCTL_RADEON_CP_RESUME ioctl.
* Add texture rectangle support for r100.
+ * 1.10- Add SETPARAM ioctl; first parameter to set is FB_LOCATION, which
+ * clients use to tell the DRM where they think the framebuffer is
+ * located in the card's address space
*/
#define DRIVER_IOCTLS \
[DRM_IOCTL_NR(DRM_IOCTL_DMA)] = { radeon_cp_buffers, 1, 0 }, \
@@ -106,10 +109,82 @@
[DRM_IOCTL_NR(DRM_IOCTL_RADEON_ALLOC)] = { radeon_mem_alloc, 1, 0 }, \
[DRM_IOCTL_NR(DRM_IOCTL_RADEON_FREE)] = { radeon_mem_free, 1, 0 }, \
[DRM_IOCTL_NR(DRM_IOCTL_RADEON_INIT_HEAP)] = { radeon_mem_init_heap, 1, 1 }, \
- [DRM_IOCTL_NR(DRM_IOCTL_RADEON_IRQ_EMIT)] = { radeon_irq_emit, 1, 0 }, \
- [DRM_IOCTL_NR(DRM_IOCTL_RADEON_IRQ_WAIT)] = { radeon_irq_wait, 1, 0 },
+ [DRM_IOCTL_NR(DRM_IOCTL_RADEON_IRQ_EMIT)] = { radeon_irq_emit, 1, 0 }, \
+ [DRM_IOCTL_NR(DRM_IOCTL_RADEON_IRQ_WAIT)] = { radeon_irq_wait, 1, 0 }, \
+ [DRM_IOCTL_NR(DRM_IOCTL_RADEON_SETPARAM)] = { radeon_cp_setparam, 1, 0 }, \
+
+#define DRIVER_PCI_IDS \
+ {0x1002, 0x4136, 0, "ATI Radeon RS100 IGP 320M"}, \
+ {0x1002, 0x4137, 0, "ATI Radeon RS200 IGP"}, \
+ {0x1002, 0x4237, 0, "ATI Radeon RS250 IGP"}, \
+ {0x1002, 0x4242, 0, "ATI Radeon BB R200 AIW 8500DV"}, \
+ {0x1002, 0x4242, 0, "ATI Radeon BC R200"}, \
+ {0x1002, 0x4336, 0, "ATI Radeon RS100 Mobility U1"}, \
+ {0x1002, 0x4337, 0, "ATI Radeon RS200 Mobility IGP 340M"}, \
+ {0x1002, 0x4437, 0, "ATI Radeon RS250 Mobility IGP"}, \
+ {0x1002, 0x4964, 0, "ATI Radeon Id R250 9000"}, \
+ {0x1002, 0x4965, 0, "ATI Radeon Ie R250 9000"}, \
+ {0x1002, 0x4966, 0, "ATI Radeon If R250 9000"}, \
+ {0x1002, 0x4967, 0, "ATI Radeon Ig R250 9000"}, \
+ {0x1002, 0x4C57, 0, "ATI Radeon LW Mobility 7500 M7"}, \
+ {0x1002, 0x4C58, 0, "ATI Radeon LX RV200 Mobility FireGL 7800 M7"}, \
+ {0x1002, 0x4C59, 0, "ATI Radeon LY Mobility M6"}, \
+ {0x1002, 0x4C5A, 0, "ATI Radeon LZ Mobility M6"}, \
+ {0x1002, 0x4C64, 0, "ATI Radeon Ld R250 Mobility 9000 M9"}, \
+ {0x1002, 0x4C65, 0, "ATI Radeon Le R250 Mobility 9000 M9"}, \
+ {0x1002, 0x4C66, 0, "ATI Radeon Lf R250 Mobility 9000 M9"}, \
+ {0x1002, 0x4C67, 0, "ATI Radeon Lg R250 Mobility 9000 M9"}, \
+ {0x1002, 0x5144, 0, "ATI Radeon QD R100"}, \
+ {0x1002, 0x5145, 0, "ATI Radeon QE R100"}, \
+ {0x1002, 0x5146, 0, "ATI Radeon QF R100"}, \
+ {0x1002, 0x5147, 0, "ATI Radeon QG R100"}, \
+ {0x1002, 0x5148, 0, "ATI Radeon QH R200 8500"}, \
+ {0x1002, 0x5149, 0, "ATI Radeon QI R200"}, \
+ {0x1002, 0x514A, 0, "ATI Radeon QJ R200"}, \
+ {0x1002, 0x514B, 0, "ATI Radeon QK R200"}, \
+ {0x1002, 0x514C, 0, "ATI Radeon QL R200 8500 LE"}, \
+ {0x1002, 0x514D, 0, "ATI Radeon QM R200 9100"}, \
+ {0x1002, 0x514E, 0, "ATI Radeon QN R200 8500 LE"}, \
+ {0x1002, 0x514F, 0, "ATI Radeon QO R200 8500 LE"}, \
+ {0x1002, 0x5157, 0, "ATI Radeon QW RV200 7500"}, \
+ {0x1002, 0x5158, 0, "ATI Radeon QX RV200 7500"}, \
+ {0x1002, 0x5159, 0, "ATI Radeon QY RV100 7000/VE"}, \
+ {0x1002, 0x515A, 0, "ATI Radeon QZ RV100 7000/VE"}, \
+ {0x1002, 0x5168, 0, "ATI Radeon Qh R200"}, \
+ {0x1002, 0x5169, 0, "ATI Radeon Qi R200"}, \
+ {0x1002, 0x516A, 0, "ATI Radeon Qj R200"}, \
+ {0x1002, 0x516B, 0, "ATI Radeon Qk R200"}, \
+ {0x1002, 0x516C, 0, "ATI Radeon Ql R200"}, \
+ {0x1002, 0x5834, 0, "ATI Radeon RS300 IGP"}, \
+ {0x1002, 0x5835, 0, "ATI Radeon RS300 Mobility IGP"}, \
+ {0x1002, 0x5836, 0, "ATI Radeon RS300 IGP"}, \
+ {0x1002, 0x5837, 0, "ATI Radeon RS300 IGP"}, \
+ {0x1002, 0x5960, 0, "ATI Radeon RV280 9200"}, \
+ {0x1002, 0x5961, 0, "ATI Radeon RV280 9200 SE"}, \
+ {0x1002, 0x5962, 0, "ATI Radeon RV280 9200"}, \
+ {0x1002, 0x5963, 0, "ATI Radeon RV280 9200"}, \
+ {0x1002, 0x5964, 0, "ATI Radeon RV280 9200 SE"}, \
+ {0x1002, 0x5968, 0, "ATI Radeon RV280 9200"}, \
+ {0x1002, 0x5969, 0, "ATI Radeon RV280 9200"}, \
+ {0x1002, 0x596A, 0, "ATI Radeon RV280 9200"}, \
+ {0x1002, 0x596B, 0, "ATI Radeon RV280 9200"}, \
+ {0x1002, 0x5c61, 0, "ATI Radeon RV280 Mobility"}, \
+ {0x1002, 0x5c62, 0, "ATI Radeon RV280"}, \
+ {0x1002, 0x5c63, 0, "ATI Radeon RV280 Mobility"}, \
+ {0x1002, 0x5c64, 0, "ATI Radeon RV280"}, \
+ {0, 0, 0, NULL}
+#define DRIVER_FILE_FIELDS \
+ int64_t radeon_fb_delta; \
+#define DRIVER_OPEN_HELPER( filp_priv, dev ) \
+do { \
+ drm_radeon_private_t *dev_priv = dev->dev_private; \
+ if ( dev_priv ) \
+ filp_priv->radeon_fb_delta = dev_priv->fb_location; \
+ else \
+ filp_priv->radeon_fb_delta = 0; \
+} while( 0 )
/* When a client dies:
* - Check for and clean up flipped page state
@@ -125,7 +200,7 @@ do { \
radeon_do_cleanup_pageflip( dev ); \
} \
radeon_mem_release( filp, dev_priv->gart_heap ); \
- radeon_mem_release( filp, dev_priv->fb_heap ); \
+ radeon_mem_release( filp, dev_priv->fb_heap ); \
} \
} while (0)
@@ -142,7 +217,7 @@ do { \
/* DMA customization:
*/
#define __HAVE_DMA 1
-#define __HAVE_DMA_IRQ 1
+#define __HAVE_IRQ 1
#define __HAVE_VBL_IRQ 1
#define __HAVE_SHARED_IRQ 1
--- diff/drivers/char/drm/radeon_cp.c 2003-09-30 14:46:12.000000000 +0000
+++ source/drivers/char/drm/radeon_cp.c 2004-03-16 11:27:48.057881752 +0000
@@ -855,7 +855,8 @@ static void radeon_cp_init_ring_buffer(
/* Initialize the memory controller */
RADEON_WRITE( RADEON_MC_FB_LOCATION,
- (dev_priv->gart_vm_start - 1) & 0xffff0000 );
+ ( ( dev_priv->gart_vm_start - 1 ) & 0xffff0000 )
+ | ( dev_priv->fb_location >> 16 ) );
#if __REALLY_HAVE_AGP
if ( !dev_priv->is_pci ) {
@@ -1071,13 +1072,6 @@ static int radeon_do_init_cp( drm_device
dev_priv->depth_offset = init->depth_offset;
dev_priv->depth_pitch = init->depth_pitch;
- dev_priv->front_pitch_offset = (((dev_priv->front_pitch/64) << 22) |
- (dev_priv->front_offset >> 10));
- dev_priv->back_pitch_offset = (((dev_priv->back_pitch/64) << 22) |
- (dev_priv->back_offset >> 10));
- dev_priv->depth_pitch_offset = (((dev_priv->depth_pitch/64) << 22) |
- (dev_priv->depth_offset >> 10));
-
/* Hardware state for depth clears. Remove this if/when we no
* longer clear the depth buffer with a 3D rectangle. Hard-code
* all values to prevent unwanted 3D state from slipping through
@@ -1124,13 +1118,6 @@ static int radeon_do_init_cp( drm_device
return DRM_ERR(EINVAL);
}
- DRM_FIND_MAP( dev_priv->fb, init->fb_offset );
- if(!dev_priv->fb) {
- DRM_ERROR("could not find framebuffer!\n");
- dev->dev_private = (void *)dev_priv;
- radeon_do_cleanup_cp(dev);
- return DRM_ERR(EINVAL);
- }
DRM_FIND_MAP( dev_priv->mmio, init->mmio_offset );
if(!dev_priv->mmio) {
DRM_ERROR("could not find mmio region!\n");
@@ -1204,9 +1191,26 @@ static int radeon_do_init_cp( drm_device
dev_priv->buffers->handle );
}
+ dev_priv->fb_location = ( RADEON_READ( RADEON_MC_FB_LOCATION )
+ & 0xffff ) << 16;
+
+ dev_priv->front_pitch_offset = (((dev_priv->front_pitch/64) << 22) |
+ ( ( dev_priv->front_offset
+ + dev_priv->fb_location ) >> 10 ) );
+
+ dev_priv->back_pitch_offset = (((dev_priv->back_pitch/64) << 22) |
+ ( ( dev_priv->back_offset
+ + dev_priv->fb_location ) >> 10 ) );
+
+ dev_priv->depth_pitch_offset = (((dev_priv->depth_pitch/64) << 22) |
+ ( ( dev_priv->depth_offset
+ + dev_priv->fb_location ) >> 10 ) );
+
dev_priv->gart_size = init->gart_size;
- dev_priv->gart_vm_start = RADEON_READ( RADEON_CONFIG_APER_SIZE );
+ dev_priv->gart_vm_start = dev_priv->fb_location
+ + RADEON_READ( RADEON_CONFIG_APER_SIZE );
+
#if __REALLY_HAVE_AGP
if ( !dev_priv->is_pci )
dev_priv->gart_buffers_offset = (dev_priv->buffers->offset
@@ -1271,12 +1275,12 @@ int radeon_do_cleanup_cp( drm_device_t *
{
DRM_DEBUG( "\n" );
-#if _HAVE_DMA_IRQ
+#if __HAVE_IRQ
/* Make sure interrupts are disabled here because the uninstall ioctl
* may not have been called from userspace and after dev_private
* is freed, it's too late.
*/
- if ( dev->irq ) DRM(irq_uninstall)(dev);
+ if ( dev->irq_enabled ) DRM(irq_uninstall)(dev);
#endif
if ( dev->dev_private ) {
--- diff/drivers/char/drm/radeon_drm.h 2003-09-30 14:46:12.000000000 +0000
+++ source/drivers/char/drm/radeon_drm.h 2004-03-16 11:27:48.058881600 +0000
@@ -390,6 +390,7 @@ typedef struct {
#define DRM_IOCTL_RADEON_IRQ_WAIT DRM_IOW( 0x57, drm_radeon_irq_wait_t)
/* added by Charl P. Botha - see radeon_cp.c for details */
#define DRM_IOCTL_RADEON_CP_RESUME DRM_IO(0x58)
+#define DRM_IOCTL_RADEON_SETPARAM DRM_IOW(0x59, drm_radeon_setparam_t)
typedef struct drm_radeon_init {
enum {
@@ -502,7 +503,7 @@ typedef struct drm_radeon_tex_image {
} drm_radeon_tex_image_t;
typedef struct drm_radeon_texture {
- int offset;
+ unsigned int offset;
int pitch;
int format;
int width; /* Texture image coordinates */
@@ -537,6 +538,7 @@ typedef struct drm_radeon_indirect {
#define RADEON_PARAM_STATUS_HANDLE 8
#define RADEON_PARAM_SAREA_HANDLE 9
#define RADEON_PARAM_GART_TEX_HANDLE 10
+#define RADEON_PARAM_SCRATCH_OFFSET 11
typedef struct drm_radeon_getparam {
int param;
@@ -578,4 +580,16 @@ typedef struct drm_radeon_irq_wait {
} drm_radeon_irq_wait_t;
+/* 1.10: Clients tell the DRM where they think the framebuffer is located in
+ * the card's address space, via a new generic ioctl to set parameters
+ */
+
+typedef struct drm_radeon_setparam {
+ unsigned int param;
+ int64_t value;
+} drm_radeon_setparam_t;
+
+#define RADEON_SETPARAM_FB_LOCATION 1 /* determined framebuffer location */
+
+
#endif
--- diff/drivers/char/drm/radeon_drv.c 2003-07-22 17:54:27.000000000 +0000
+++ source/drivers/char/drm/radeon_drv.c 2004-03-16 11:27:48.059881448 +0000
@@ -48,6 +48,7 @@
#include "drm_fops.h"
#include "drm_init.h"
#include "drm_ioctl.h"
+#include "drm_irq.h"
#include "drm_lock.h"
#include "drm_memory.h"
#include "drm_proc.h"
--- diff/drivers/char/drm/radeon_drv.h 2003-09-30 14:46:12.000000000 +0000
+++ source/drivers/char/drm/radeon_drv.h 2004-03-16 11:27:48.060881296 +0000
@@ -73,6 +73,8 @@ typedef struct drm_radeon_private {
drm_radeon_ring_buffer_t ring;
drm_radeon_sarea_t *sarea_priv;
+ u32 fb_location;
+
int gart_size;
u32 gart_vm_start;
unsigned long gart_buffers_offset;
@@ -133,7 +135,6 @@ typedef struct drm_radeon_private {
unsigned long gart_textures_offset;
drm_local_map_t *sarea;
- drm_local_map_t *fb;
drm_local_map_t *mmio;
drm_local_map_t *cp_ring;
drm_local_map_t *ring_rptr;
@@ -184,6 +185,7 @@ extern int radeon_cp_indirect( DRM_IOCTL
extern int radeon_cp_vertex2( DRM_IOCTL_ARGS );
extern int radeon_cp_cmdbuf( DRM_IOCTL_ARGS );
extern int radeon_cp_getparam( DRM_IOCTL_ARGS );
+extern int radeon_cp_setparam( DRM_IOCTL_ARGS );
extern int radeon_cp_flip( DRM_IOCTL_ARGS );
extern int radeon_mem_alloc( DRM_IOCTL_ARGS );
@@ -239,6 +241,7 @@ extern void radeon_do_release(drm_device
#define RADEON_CRTC2_OFFSET 0x0324
#define RADEON_CRTC2_OFFSET_CNTL 0x0328
+#define RADEON_RB3D_COLOROFFSET 0x1c40
#define RADEON_RB3D_COLORPITCH 0x1c48
#define RADEON_DP_GUI_MASTER_CNTL 0x146c
@@ -332,6 +335,7 @@ extern void radeon_do_release(drm_device
#define RADEON_PP_MISC 0x1c14
#define RADEON_PP_ROT_MATRIX_0 0x1d58
#define RADEON_PP_TXFILTER_0 0x1c54
+#define RADEON_PP_TXOFFSET_0 0x1c5c
#define RADEON_PP_TXFILTER_1 0x1c6c
#define RADEON_PP_TXFILTER_2 0x1c84
--- diff/drivers/char/drm/radeon_irq.c 2003-05-21 10:50:14.000000000 +0000
+++ source/drivers/char/drm/radeon_irq.c 2004-03-16 11:27:48.062880992 +0000
@@ -54,7 +54,7 @@
* tied to dma at all, this is just a hangover from dri prehistory.
*/
-irqreturn_t DRM(dma_service)( DRM_IRQ_ARGS )
+irqreturn_t DRM(irq_handler)( DRM_IRQ_ARGS )
{
drm_device_t *dev = (drm_device_t *) arg;
drm_radeon_private_t *dev_priv =
--- diff/drivers/char/drm/radeon_state.c 2004-02-18 08:54:09.000000000 +0000
+++ source/drivers/char/drm/radeon_state.c 2004-03-16 11:27:48.065880536 +0000
@@ -36,6 +36,151 @@
/* ================================================================
+ * Helper functions for client state checking and fixup
+ */
+
+static __inline__ int radeon_check_and_fixup_offset( drm_radeon_private_t *dev_priv,
+ drm_file_t *filp_priv,
+ u32 *offset ) {
+ u32 off = *offset;
+
+ if ( off >= dev_priv->fb_location &&
+ off < ( dev_priv->gart_vm_start + dev_priv->gart_size ) )
+ return 0;
+
+ off += filp_priv->radeon_fb_delta;
+
+ DRM_DEBUG( "offset fixed up to 0x%x\n", off );
+
+ if ( off < dev_priv->fb_location ||
+ off >= ( dev_priv->gart_vm_start + dev_priv->gart_size ) )
+ return DRM_ERR( EINVAL );
+
+ *offset = off;
+
+ return 0;
+}
+
+static __inline__ int radeon_check_and_fixup_offset_user( drm_radeon_private_t *dev_priv,
+ drm_file_t *filp_priv,
+ u32 *offset ) {
+ u32 off;
+
+ DRM_GET_USER_UNCHECKED( off, offset );
+
+ if ( radeon_check_and_fixup_offset( dev_priv, filp_priv, &off ) )
+ return DRM_ERR( EINVAL );
+
+ DRM_PUT_USER_UNCHECKED( offset, off );
+
+ return 0;
+}
+
+static __inline__ int radeon_check_and_fixup_packets( drm_radeon_private_t *dev_priv,
+ drm_file_t *filp_priv,
+ int id,
+ u32 *data ) {
+ if ( id == RADEON_EMIT_PP_MISC &&
+ radeon_check_and_fixup_offset_user( dev_priv, filp_priv,
+ &data[( RADEON_RB3D_DEPTHOFFSET
+ - RADEON_PP_MISC ) / 4] ) ) {
+ DRM_ERROR( "Invalid depth buffer offset\n" );
+ return DRM_ERR( EINVAL );
+ } else if ( id == RADEON_EMIT_PP_CNTL &&
+ radeon_check_and_fixup_offset_user( dev_priv, filp_priv,
+ &data[( RADEON_RB3D_COLOROFFSET
+ - RADEON_PP_CNTL ) / 4] ) ) {
+ DRM_ERROR( "Invalid colour buffer offset\n" );
+ return DRM_ERR( EINVAL );
+ } else if ( id >= R200_EMIT_PP_TXOFFSET_0 &&
+ id <= R200_EMIT_PP_TXOFFSET_5 &&
+ radeon_check_and_fixup_offset_user( dev_priv, filp_priv,
+ &data[0] ) ) {
+ DRM_ERROR( "Invalid R200 texture offset\n" );
+ return DRM_ERR( EINVAL );
+ } else if ( ( id == RADEON_EMIT_PP_TXFILTER_0 || id == RADEON_EMIT_PP_TXFILTER_1 ||
+ id == RADEON_EMIT_PP_TXFILTER_2 /*|| id == RADEON_EMIT_PP_TXFILTER_3 ||
+ id == RADEON_EMIT_PP_TXFILTER_4 || id == RADEON_EMIT_PP_TXFILTER_5*/ ) &&
+ radeon_check_and_fixup_offset_user( dev_priv, filp_priv,
+ &data[( RADEON_PP_TXOFFSET_0
+ - RADEON_PP_TXFILTER_0 ) / 4] ) ) {
+ DRM_ERROR( "Invalid R100 texture offset\n" );
+ return DRM_ERR( EINVAL );
+ } else if ( id == R200_PP_CUBIC_OFFSET_F1_0 || id == R200_PP_CUBIC_OFFSET_F1_1 ||
+ id == R200_PP_CUBIC_OFFSET_F1_2 || id == R200_PP_CUBIC_OFFSET_F1_3 ||
+ id == R200_PP_CUBIC_OFFSET_F1_4 || id == R200_PP_CUBIC_OFFSET_F1_5 ) {
+ int i;
+ for ( i = 0; i < 6; i++ ) {
+ if ( radeon_check_and_fixup_offset_user( dev_priv,
+ filp_priv,
+ &data[i] ) ) {
+ DRM_ERROR( "Invalid R200 cubic texture offset\n" );
+ return DRM_ERR( EINVAL );
+ }
+ }
+ }
+
+ return 0;
+}
+
+static __inline__ int radeon_check_and_fixup_packet3( drm_radeon_private_t *dev_priv,
+ drm_file_t *filp_priv,
+ drm_radeon_cmd_buffer_t *cmdbuf,
+ unsigned int *cmdsz ) {
+ u32 tmp[4], *cmd = ( u32* )cmdbuf->buf;
+
+ if ( DRM_COPY_FROM_USER_UNCHECKED( tmp, cmd, sizeof( tmp ) ) ) {
+ DRM_ERROR( "Failed to copy data from user space\n" );
+ return DRM_ERR( EFAULT );
+ }
+
+ *cmdsz = 2 + ( ( tmp[0] & RADEON_CP_PACKET_COUNT_MASK ) >> 16 );
+
+ if ( ( tmp[0] & 0xc0000000 ) != RADEON_CP_PACKET3 ) {
+ DRM_ERROR( "Not a type 3 packet\n" );
+ return DRM_ERR( EINVAL );
+ }
+
+ if ( 4 * *cmdsz > cmdbuf->bufsz ) {
+ DRM_ERROR( "Packet size larger than size of data provided\n" );
+ return DRM_ERR( EINVAL );
+ }
+
+ /* Check client state and fix it up if necessary */
+ if ( tmp[0] & 0x8000 ) { /* MSB of opcode: next DWORD GUI_CNTL */
+ u32 offset;
+
+ if ( tmp[1] & ( RADEON_GMC_SRC_PITCH_OFFSET_CNTL
+ | RADEON_GMC_DST_PITCH_OFFSET_CNTL ) ) {
+ offset = tmp[2] << 10;
+ if ( radeon_check_and_fixup_offset( dev_priv, filp_priv, &offset ) ) {
+ DRM_ERROR( "Invalid first packet offset\n" );
+ return DRM_ERR( EINVAL );
+ }
+ tmp[2] = ( tmp[2] & 0xffc00000 ) | offset >> 10;
+ }
+
+ if ( ( tmp[1] & RADEON_GMC_SRC_PITCH_OFFSET_CNTL ) &&
+ ( tmp[1] & RADEON_GMC_DST_PITCH_OFFSET_CNTL ) ) {
+ offset = tmp[3] << 10;
+ if ( radeon_check_and_fixup_offset( dev_priv, filp_priv, &offset ) ) {
+ DRM_ERROR( "Invalid second packet offset\n" );
+ return DRM_ERR( EINVAL );
+ }
+ tmp[3] = ( tmp[3] & 0xffc00000 ) | offset >> 10;
+ }
+
+ if ( DRM_COPY_TO_USER_UNCHECKED( cmd, tmp, sizeof( tmp ) ) ) {
+ DRM_ERROR( "Failed to copy data to user space\n" );
+ return DRM_ERR( EFAULT );
+ }
+ }
+
+ return 0;
+}
+
+
+/* ================================================================
* CP hardware state programming functions
*/
@@ -57,15 +202,28 @@ static __inline__ void radeon_emit_clip_
/* Emit 1.1 state
*/
-static void radeon_emit_state( drm_radeon_private_t *dev_priv,
- drm_radeon_context_regs_t *ctx,
- drm_radeon_texture_regs_t *tex,
- unsigned int dirty )
+static int radeon_emit_state( drm_radeon_private_t *dev_priv,
+ drm_file_t *filp_priv,
+ drm_radeon_context_regs_t *ctx,
+ drm_radeon_texture_regs_t *tex,
+ unsigned int dirty )
{
RING_LOCALS;
DRM_DEBUG( "dirty=0x%08x\n", dirty );
if ( dirty & RADEON_UPLOAD_CONTEXT ) {
+ if ( radeon_check_and_fixup_offset( dev_priv, filp_priv,
+ &ctx->rb3d_depthoffset ) ) {
+ DRM_ERROR( "Invalid depth buffer offset\n" );
+ return DRM_ERR( EINVAL );
+ }
+
+ if ( radeon_check_and_fixup_offset( dev_priv, filp_priv,
+ &ctx->rb3d_coloroffset ) ) {
+ DRM_ERROR( "Invalid depth buffer offset\n" );
+ return DRM_ERR( EINVAL );
+ }
+
BEGIN_RING( 14 );
OUT_RING( CP_PACKET0( RADEON_PP_MISC, 6 ) );
OUT_RING( ctx->pp_misc );
@@ -149,6 +307,12 @@ static void radeon_emit_state( drm_radeo
}
if ( dirty & RADEON_UPLOAD_TEX0 ) {
+ if ( radeon_check_and_fixup_offset( dev_priv, filp_priv,
+ &tex[0].pp_txoffset ) ) {
+ DRM_ERROR( "Invalid texture offset for unit 0\n" );
+ return DRM_ERR( EINVAL );
+ }
+
BEGIN_RING( 9 );
OUT_RING( CP_PACKET0( RADEON_PP_TXFILTER_0, 5 ) );
OUT_RING( tex[0].pp_txfilter );
@@ -163,6 +327,12 @@ static void radeon_emit_state( drm_radeo
}
if ( dirty & RADEON_UPLOAD_TEX1 ) {
+ if ( radeon_check_and_fixup_offset( dev_priv, filp_priv,
+ &tex[1].pp_txoffset ) ) {
+ DRM_ERROR( "Invalid texture offset for unit 1\n" );
+ return DRM_ERR( EINVAL );
+ }
+
BEGIN_RING( 9 );
OUT_RING( CP_PACKET0( RADEON_PP_TXFILTER_1, 5 ) );
OUT_RING( tex[1].pp_txfilter );
@@ -177,6 +347,12 @@ static void radeon_emit_state( drm_radeo
}
if ( dirty & RADEON_UPLOAD_TEX2 ) {
+ if ( radeon_check_and_fixup_offset( dev_priv, filp_priv,
+ &tex[2].pp_txoffset ) ) {
+ DRM_ERROR( "Invalid texture offset for unit 2\n" );
+ return DRM_ERR( EINVAL );
+ }
+
BEGIN_RING( 9 );
OUT_RING( CP_PACKET0( RADEON_PP_TXFILTER_2, 5 ) );
OUT_RING( tex[2].pp_txfilter );
@@ -189,12 +365,15 @@ static void radeon_emit_state( drm_radeo
OUT_RING( tex[2].pp_border_color );
ADVANCE_RING();
}
+
+ return 0;
}
/* Emit 1.2 state
*/
-static void radeon_emit_state2( drm_radeon_private_t *dev_priv,
- drm_radeon_state_t *state )
+static int radeon_emit_state2( drm_radeon_private_t *dev_priv,
+ drm_file_t *filp_priv,
+ drm_radeon_state_t *state )
{
RING_LOCALS;
@@ -206,7 +385,7 @@ static void radeon_emit_state2( drm_rade
ADVANCE_RING();
}
- radeon_emit_state( dev_priv, &state->context,
+ return radeon_emit_state( dev_priv, filp_priv, &state->context,
state->tex, state->dirty );
}
@@ -1065,6 +1244,7 @@ static int radeon_cp_dispatch_texture( D
drm_radeon_tex_image_t *image )
{
drm_radeon_private_t *dev_priv = dev->dev_private;
+ drm_file_t *filp_priv;
drm_buf_t *buf;
u32 format;
u32 *buffer;
@@ -1074,6 +1254,13 @@ static int radeon_cp_dispatch_texture( D
int i;
RING_LOCALS;
+ DRM_GET_PRIV_WITH_RETURN( filp_priv, filp );
+
+ if ( radeon_check_and_fixup_offset( dev_priv, filp_priv, &tex->offset ) ) {
+ DRM_ERROR( "Invalid destination offset\n" );
+ return DRM_ERR( EINVAL );
+ }
+
dev_priv->stats.boxes |= RADEON_BOX_TEXTURE_LOAD;
/* Flush the pixel cache. This ensures no pixel data gets mixed
@@ -1377,6 +1564,7 @@ int radeon_cp_vertex( DRM_IOCTL_ARGS )
{
DRM_DEVICE;
drm_radeon_private_t *dev_priv = dev->dev_private;
+ drm_file_t *filp_priv;
drm_radeon_sarea_t *sarea_priv = dev_priv->sarea_priv;
drm_device_dma_t *dma = dev->dma;
drm_buf_t *buf;
@@ -1390,6 +1578,8 @@ int radeon_cp_vertex( DRM_IOCTL_ARGS )
return DRM_ERR(EINVAL);
}
+ DRM_GET_PRIV_WITH_RETURN( filp_priv, filp );
+
DRM_COPY_FROM_USER_IOCTL( vertex, (drm_radeon_vertex_t *)data,
sizeof(vertex) );
@@ -1429,11 +1619,14 @@ int radeon_cp_vertex( DRM_IOCTL_ARGS )
buf->used = vertex.count; /* not used? */
if ( sarea_priv->dirty & ~RADEON_UPLOAD_CLIPRECTS ) {
- radeon_emit_state( dev_priv,
- &sarea_priv->context_state,
- sarea_priv->tex_state,
- sarea_priv->dirty );
-
+ if ( radeon_emit_state( dev_priv, filp_priv,
+ &sarea_priv->context_state,
+ sarea_priv->tex_state,
+ sarea_priv->dirty ) ) {
+ DRM_ERROR( "radeon_emit_state failed\n" );
+ return DRM_ERR( EINVAL );
+ }
+
sarea_priv->dirty &= ~(RADEON_UPLOAD_TEX0IMAGES |
RADEON_UPLOAD_TEX1IMAGES |
RADEON_UPLOAD_TEX2IMAGES |
@@ -1461,6 +1654,7 @@ int radeon_cp_indices( DRM_IOCTL_ARGS )
{
DRM_DEVICE;
drm_radeon_private_t *dev_priv = dev->dev_private;
+ drm_file_t *filp_priv;
drm_radeon_sarea_t *sarea_priv = dev_priv->sarea_priv;
drm_device_dma_t *dma = dev->dma;
drm_buf_t *buf;
@@ -1475,6 +1669,8 @@ int radeon_cp_indices( DRM_IOCTL_ARGS )
return DRM_ERR(EINVAL);
}
+ DRM_GET_PRIV_WITH_RETURN( filp_priv, filp );
+
DRM_COPY_FROM_USER_IOCTL( elts, (drm_radeon_indices_t *)data,
sizeof(elts) );
@@ -1523,10 +1719,13 @@ int radeon_cp_indices( DRM_IOCTL_ARGS )
buf->used = elts.end;
if ( sarea_priv->dirty & ~RADEON_UPLOAD_CLIPRECTS ) {
- radeon_emit_state( dev_priv,
- &sarea_priv->context_state,
- sarea_priv->tex_state,
- sarea_priv->dirty );
+ if ( radeon_emit_state( dev_priv, filp_priv,
+ &sarea_priv->context_state,
+ sarea_priv->tex_state,
+ sarea_priv->dirty ) ) {
+ DRM_ERROR( "radeon_emit_state failed\n" );
+ return DRM_ERR( EINVAL );
+ }
sarea_priv->dirty &= ~(RADEON_UPLOAD_TEX0IMAGES |
RADEON_UPLOAD_TEX1IMAGES |
@@ -1686,6 +1885,7 @@ int radeon_cp_vertex2( DRM_IOCTL_ARGS )
{
DRM_DEVICE;
drm_radeon_private_t *dev_priv = dev->dev_private;
+ drm_file_t *filp_priv;
drm_radeon_sarea_t *sarea_priv = dev_priv->sarea_priv;
drm_device_dma_t *dma = dev->dma;
drm_buf_t *buf;
@@ -1700,6 +1900,8 @@ int radeon_cp_vertex2( DRM_IOCTL_ARGS )
return DRM_ERR(EINVAL);
}
+ DRM_GET_PRIV_WITH_RETURN( filp_priv, filp );
+
DRM_COPY_FROM_USER_IOCTL( vertex, (drm_radeon_vertex2_t *)data,
sizeof(vertex) );
@@ -1747,7 +1949,10 @@ int radeon_cp_vertex2( DRM_IOCTL_ARGS )
sizeof(state) ) )
return DRM_ERR(EFAULT);
- radeon_emit_state2( dev_priv, &state );
+ if ( radeon_emit_state2( dev_priv, filp_priv, &state ) ) {
+ DRM_ERROR( "radeon_emit_state2 failed\n" );
+ return DRM_ERR( EINVAL );
+ }
laststate = prim.stateidx;
}
@@ -1784,6 +1989,7 @@ int radeon_cp_vertex2( DRM_IOCTL_ARGS )
static int radeon_emit_packets(
drm_radeon_private_t *dev_priv,
+ drm_file_t *filp_priv,
drm_radeon_cmd_header_t header,
drm_radeon_cmd_buffer_t *cmdbuf )
{
@@ -1798,8 +2004,15 @@ static int radeon_emit_packets(
sz = packet[id].len;
reg = packet[id].start;
- if (sz * sizeof(int) > cmdbuf->bufsz)
+ if (sz * sizeof(int) > cmdbuf->bufsz) {
+ DRM_ERROR( "Packet size provided larger than data provided\n" );
return DRM_ERR(EINVAL);
+ }
+
+ if ( radeon_check_and_fixup_packets( dev_priv, filp_priv, id, data ) ) {
+ DRM_ERROR( "Packet verification failed\n" );
+ return DRM_ERR( EINVAL );
+ }
BEGIN_RING(sz+1);
OUT_RING( CP_PACKET0( reg, (sz-1) ) );
@@ -1882,24 +2095,21 @@ static __inline__ int radeon_emit_vector
static int radeon_emit_packet3( drm_device_t *dev,
+ drm_file_t *filp_priv,
drm_radeon_cmd_buffer_t *cmdbuf )
{
drm_radeon_private_t *dev_priv = dev->dev_private;
- int cmdsz, tmp;
- int *cmd = (int *)cmdbuf->buf;
+ unsigned int cmdsz;
+ int *cmd = (int *)cmdbuf->buf, ret;
RING_LOCALS;
-
DRM_DEBUG("\n");
- if (DRM_GET_USER_UNCHECKED( tmp, &cmd[0]))
- return DRM_ERR(EFAULT);
-
- cmdsz = 2 + ((tmp & RADEON_CP_PACKET_COUNT_MASK) >> 16);
-
- if ((tmp & 0xc0000000) != RADEON_CP_PACKET3 ||
- cmdsz * 4 > cmdbuf->bufsz)
- return DRM_ERR(EINVAL);
+ if ( ( ret = radeon_check_and_fixup_packet3( dev_priv, filp_priv,
+ cmdbuf, &cmdsz ) ) ) {
+ DRM_ERROR( "Packet verification failed\n" );
+ return ret;
+ }
BEGIN_RING( cmdsz );
OUT_RING_USER_TABLE( cmd, cmdsz );
@@ -1912,27 +2122,25 @@ static int radeon_emit_packet3( drm_devi
static int radeon_emit_packet3_cliprect( drm_device_t *dev,
+ drm_file_t *filp_priv,
drm_radeon_cmd_buffer_t *cmdbuf,
int orig_nbox )
{
drm_radeon_private_t *dev_priv = dev->dev_private;
drm_clip_rect_t box;
- int cmdsz, tmp;
- int *cmd = (int *)cmdbuf->buf;
+ unsigned int cmdsz;
+ int *cmd = (int *)cmdbuf->buf, ret;
drm_clip_rect_t *boxes = cmdbuf->boxes;
int i = 0;
RING_LOCALS;
DRM_DEBUG("\n");
- if (DRM_GET_USER_UNCHECKED( tmp, &cmd[0]))
- return DRM_ERR(EFAULT);
-
- cmdsz = 2 + ((tmp & RADEON_CP_PACKET_COUNT_MASK) >> 16);
-
- if ((tmp & 0xc0000000) != RADEON_CP_PACKET3 ||
- cmdsz * 4 > cmdbuf->bufsz)
- return DRM_ERR(EINVAL);
+ if ( ( ret = radeon_check_and_fixup_packet3( dev_priv, filp_priv,
+ cmdbuf, &cmdsz ) ) ) {
+ DRM_ERROR( "Packet verification failed\n" );
+ return ret;
+ }
if (!orig_nbox)
goto out;
@@ -2009,6 +2217,7 @@ int radeon_cp_cmdbuf( DRM_IOCTL_ARGS )
{
DRM_DEVICE;
drm_radeon_private_t *dev_priv = dev->dev_private;
+ drm_file_t *filp_priv;
drm_device_dma_t *dma = dev->dma;
drm_buf_t *buf = 0;
int idx;
@@ -2023,6 +2232,8 @@ int radeon_cp_cmdbuf( DRM_IOCTL_ARGS )
return DRM_ERR(EINVAL);
}
+ DRM_GET_PRIV_WITH_RETURN( filp_priv, filp );
+
DRM_COPY_FROM_USER_IOCTL( cmdbuf, (drm_radeon_cmd_buffer_t *)data,
sizeof(cmdbuf) );
@@ -2053,7 +2264,7 @@ int radeon_cp_cmdbuf( DRM_IOCTL_ARGS )
switch (header.header.cmd_type) {
case RADEON_CMD_PACKET:
DRM_DEBUG("RADEON_CMD_PACKET\n");
- if (radeon_emit_packets( dev_priv, header, &cmdbuf )) {
+ if (radeon_emit_packets( dev_priv, filp_priv, header, &cmdbuf )) {
DRM_ERROR("radeon_emit_packets failed\n");
return DRM_ERR(EINVAL);
}
@@ -2096,7 +2307,7 @@ int radeon_cp_cmdbuf( DRM_IOCTL_ARGS )
case RADEON_CMD_PACKET3:
DRM_DEBUG("RADEON_CMD_PACKET3\n");
- if (radeon_emit_packet3( dev, &cmdbuf )) {
+ if (radeon_emit_packet3( dev, filp_priv, &cmdbuf )) {
DRM_ERROR("radeon_emit_packet3 failed\n");
return DRM_ERR(EINVAL);
}
@@ -2104,7 +2315,7 @@ int radeon_cp_cmdbuf( DRM_IOCTL_ARGS )
case RADEON_CMD_PACKET3_CLIP:
DRM_DEBUG("RADEON_CMD_PACKET3_CLIP\n");
- if (radeon_emit_packet3_cliprect( dev, &cmdbuf, orig_nbox )) {
+ if (radeon_emit_packet3_cliprect( dev, filp_priv, &cmdbuf, orig_nbox )) {
DRM_ERROR("radeon_emit_packet3_clip failed\n");
return DRM_ERR(EINVAL);
}
@@ -2214,3 +2425,31 @@ int radeon_cp_getparam( DRM_IOCTL_ARGS )
return 0;
}
+
+int radeon_cp_setparam( DRM_IOCTL_ARGS ) {
+ DRM_DEVICE;
+ drm_radeon_private_t *dev_priv = dev->dev_private;
+ drm_file_t *filp_priv;
+ drm_radeon_setparam_t sp;
+
+ if ( !dev_priv ) {
+ DRM_ERROR( "%s called with no initialization\n", __FUNCTION__ );
+ return DRM_ERR( EINVAL );
+ }
+
+ DRM_GET_PRIV_WITH_RETURN( filp_priv, filp );
+
+ DRM_COPY_FROM_USER_IOCTL( sp, ( drm_radeon_setparam_t* )data,
+ sizeof( sp ) );
+
+ switch( sp.param ) {
+ case RADEON_SETPARAM_FB_LOCATION:
+ filp_priv->radeon_fb_delta = dev_priv->fb_location - sp.value;
+ break;
+ default:
+ DRM_DEBUG( "Invalid parameter %d\n", sp.param );
+ return DRM_ERR( EINVAL );
+ }
+
+ return 0;
+}
--- diff/drivers/char/drm/sis.h 2003-09-30 14:46:12.000000000 +0000
+++ source/drivers/char/drm/sis.h 2004-03-16 11:27:48.066880384 +0000
@@ -62,6 +62,13 @@
[DRM_IOCTL_NR(DRM_IOCTL_SIS_AGP_FREE)] = { sis_ioctl_agp_free, 1, 0 }, \
[DRM_IOCTL_NR(DRM_IOCTL_SIS_FB_INIT)] = { sis_fb_init, 1, 1 }
+#define DRIVER_PCI_IDS \
+ {0x1039, 0x0300, 0, "SiS 300/305"}, \
+ {0x1039, 0x5300, 0, "SiS 540"}, \
+ {0x1039, 0x6300, 0, "SiS 630"}, \
+ {0x1039, 0x7300, 0, "SiS 730"}, \
+ {0, 0, 0, NULL}
+
#define __HAVE_COUNTERS 5
/* Buffer customization:
--- diff/drivers/char/drm/sis_mm.c 2003-09-30 14:46:12.000000000 +0000
+++ source/drivers/char/drm/sis_mm.c 2004-03-16 11:27:48.067880232 +0000
@@ -34,7 +34,11 @@
#include "sis_drv.h"
#include "sis_ds.h"
#if defined(__linux__) && defined(CONFIG_FB_SIS)
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,0)
#include <video/sisfb.h>
+#else
+#include <linux/sisfb.h>
+#endif
#endif
#define MAX_CONTEXT 100
@@ -132,7 +136,7 @@ int sis_fb_free( DRM_IOCTL_ARGS )
retval = DRM_ERR(EINVAL);
sis_free(fb.free);
- DRM_DEBUG("free fb, offset = %lu\n", fb.free);
+ DRM_DEBUG("free fb, offset = 0x%lx\n", fb.free);
return retval;
}
--- diff/drivers/char/drm/tdfx.h 2002-10-16 03:28:22.000000000 +0000
+++ source/drivers/char/drm/tdfx.h 2004-03-16 11:27:48.068880080 +0000
@@ -39,4 +39,22 @@
#define __HAVE_MTRR 1
#define __HAVE_CTX_BITMAP 1
+#define DRIVER_AUTHOR "VA Linux Systems Inc."
+
+#define DRIVER_NAME "tdfx"
+#define DRIVER_DESC "3dfx Banshee/Voodoo3+"
+#define DRIVER_DATE "20010216"
+
+#define DRIVER_MAJOR 1
+#define DRIVER_MINOR 0
+#define DRIVER_PATCHLEVEL 0
+
+#define DRIVER_PCI_IDS \
+ {0x121a, 0x0003, 0, "3dfx Voodoo Banshee"}, \
+ {0x121a, 0x0004, 0, "3dfx Voodoo3 2000"}, \
+ {0x121a, 0x0005, 0, "3dfx Voodoo3 3000"}, \
+ {0x121a, 0x0007, 0, "3dfx Voodoo4"}, \
+ {0x121a, 0x0009, 0, "3dfx Voodoo5"}, \
+ {0, 0, 0, NULL}
+
#endif
--- diff/drivers/char/drm/tdfx_drv.c 2002-10-16 03:28:22.000000000 +0000
+++ source/drivers/char/drm/tdfx_drv.c 2004-03-16 11:27:48.067880232 +0000
@@ -34,47 +34,6 @@
#include "tdfx.h"
#include "drmP.h"
-#define DRIVER_AUTHOR "VA Linux Systems Inc."
-
-#define DRIVER_NAME "tdfx"
-#define DRIVER_DESC "3dfx Banshee/Voodoo3+"
-#define DRIVER_DATE "20010216"
-
-#define DRIVER_MAJOR 1
-#define DRIVER_MINOR 0
-#define DRIVER_PATCHLEVEL 0
-
-#ifndef PCI_VENDOR_ID_3DFX
-#define PCI_VENDOR_ID_3DFX 0x121A
-#endif
-#ifndef PCI_DEVICE_ID_3DFX_VOODOO5
-#define PCI_DEVICE_ID_3DFX_VOODOO5 0x0009
-#endif
-#ifndef PCI_DEVICE_ID_3DFX_VOODOO4
-#define PCI_DEVICE_ID_3DFX_VOODOO4 0x0007
-#endif
-#ifndef PCI_DEVICE_ID_3DFX_VOODOO3_3000 /* Voodoo3 3000 */
-#define PCI_DEVICE_ID_3DFX_VOODOO3_3000 0x0005
-#endif
-#ifndef PCI_DEVICE_ID_3DFX_VOODOO3_2000 /* Voodoo3 3000 */
-#define PCI_DEVICE_ID_3DFX_VOODOO3_2000 0x0004
-#endif
-#ifndef PCI_DEVICE_ID_3DFX_BANSHEE
-#define PCI_DEVICE_ID_3DFX_BANSHEE 0x0003
-#endif
-
-static drm_pci_list_t DRM(idlist)[] = {
- { PCI_VENDOR_ID_3DFX, PCI_DEVICE_ID_3DFX_BANSHEE },
- { PCI_VENDOR_ID_3DFX, PCI_DEVICE_ID_3DFX_VOODOO3_2000 },
- { PCI_VENDOR_ID_3DFX, PCI_DEVICE_ID_3DFX_VOODOO3_3000 },
- { PCI_VENDOR_ID_3DFX, PCI_DEVICE_ID_3DFX_VOODOO4 },
- { PCI_VENDOR_ID_3DFX, PCI_DEVICE_ID_3DFX_VOODOO5 },
- { 0, 0 }
-};
-
-#define DRIVER_CARD_LIST DRM(idlist)
-
-
#include "drm_auth.h"
#include "drm_bufs.h"
#include "drm_context.h"
--- diff/drivers/char/hvc_console.c 2004-02-18 08:54:09.000000000 +0000
+++ source/drivers/char/hvc_console.c 2004-03-16 11:27:48.068880080 +0000
@@ -131,31 +131,65 @@ static int hvc_write(struct tty_struct *
const unsigned char *buf, int count)
{
struct hvc_struct *hp = tty->driver_data;
- char *p;
- int todo, written = 0;
+ char *tbuf, *p;
+ int tbsize, rsize, written = 0;
unsigned long flags;
- spin_lock_irqsave(&hp->lock, flags);
- while (count > 0 && (todo = N_OUTBUF - hp->n_outbuf) > 0) {
- if (todo > count)
- todo = count;
- p = hp->outbuf + hp->n_outbuf;
- if (from_user) {
- todo -= copy_from_user(p, buf, todo);
- if (todo == 0) {
+ if (from_user) {
+ tbsize = min(count, (int)PAGE_SIZE);
+ if (!(tbuf = kmalloc(tbsize, GFP_KERNEL)))
+ return -ENOMEM;
+
+ while ((rsize = count - written) > 0) {
+ int wsize;
+ if (rsize > tbsize)
+ rsize = tbsize;
+
+ p = tbuf;
+ rsize -= copy_from_user(p, buf, rsize);
+ if (!rsize) {
if (written == 0)
written = -EFAULT;
break;
}
- } else
- memcpy(p, buf, todo);
- count -= todo;
- buf += todo;
- hp->n_outbuf += todo;
- written += todo;
- hvc_push(hp);
+ buf += rsize;
+ written += rsize;
+
+ spin_lock_irqsave(&hp->lock, flags);
+ for (wsize = N_OUTBUF - hp->n_outbuf; rsize && wsize;
+ wsize = N_OUTBUF - hp->n_outbuf) {
+ if (wsize > rsize)
+ wsize = rsize;
+ memcpy(hp->outbuf + hp->n_outbuf, p, wsize);
+ hp->n_outbuf += wsize;
+ hvc_push(hp);
+ rsize -= wsize;
+ p += wsize;
+ }
+ spin_unlock_irqrestore(&hp->lock, flags);
+
+ if (rsize)
+ break;
+
+ if (count < tbsize)
+ tbsize = count;
+ }
+
+ kfree(tbuf);
+ } else {
+ spin_lock_irqsave(&hp->lock, flags);
+ while (count > 0 && (rsize = N_OUTBUF - hp->n_outbuf) > 0) {
+ if (rsize > count)
+ rsize = count;
+ memcpy(hp->outbuf + hp->n_outbuf, buf, rsize);
+ count -= rsize;
+ buf += rsize;
+ hp->n_outbuf += rsize;
+ written += rsize;
+ hvc_push(hp);
+ }
+ spin_unlock_irqrestore(&hp->lock, flags);
}
- spin_unlock_irqrestore(&hp->lock, flags);
return written;
}
--- diff/drivers/char/ipmi/Kconfig 2003-10-27 09:20:37.000000000 +0000
+++ source/drivers/char/ipmi/Kconfig 2004-03-16 11:27:48.102874912 +0000
@@ -43,11 +43,24 @@ config IPMI_DEVICE_INTERFACE
This provides an IOCTL interface to the IPMI message handler so
userland processes may use IPMI. It supports poll() and select().
-config IPMI_KCS
- tristate 'IPMI KCS handler'
+config IPMI_SI
+ tristate 'IPMI System Interface handler'
depends on IPMI_HANDLER
help
- Provides a driver for a KCS-style interface to a BMC.
+ Provides a driver for System Interfaces (KCS, SMIC, BT).
+ Currently, only KCS and SMIC are supported. If
+ you are using IPMI, you should probably say "y" here.
+
+config IPMI_SMB
+ tristate 'IPMI SMBus handler'
+ depends on IPMI_HANDLER && I2C
+ help
+ Provides a driver for a SMBus interface to a BMC, meaning that you
+ have a driver that must be accessed over an I2C bus instead of a
+ standard interface. This module requires I2C support. Note that
+ you might need some I2C changes if CONFIG_IPMI_PANIC_EVENT is
+ enabled along with this, so the I2C driver knows to run to
+ completion during sending a panic event.
config IPMI_WATCHDOG
tristate 'IPMI Watchdog Timer'
--- diff/drivers/char/ipmi/Makefile 2003-02-13 11:46:51.000000000 +0000
+++ source/drivers/char/ipmi/Makefile 2004-03-16 11:27:48.103874760 +0000
@@ -2,12 +2,14 @@
# Makefile for the ipmi drivers.
#
-ipmi_kcs_drv-objs := ipmi_kcs_sm.o ipmi_kcs_intf.o
+ipmi_si-objs := ipmi_si_intf.o ipmi_kcs_sm.o ipmi_smic_sm.o ipmi_bt_sm.o
obj-$(CONFIG_IPMI_HANDLER) += ipmi_msghandler.o
obj-$(CONFIG_IPMI_DEVICE_INTERFACE) += ipmi_devintf.o
-obj-$(CONFIG_IPMI_KCS) += ipmi_kcs_drv.o
+obj-$(CONFIG_IPMI_SI) += ipmi_si.o
+obj-$(CONFIG_IPMI_SMB) += ipmi_smb.o
obj-$(CONFIG_IPMI_WATCHDOG) += ipmi_watchdog.o
-ipmi_kcs_drv.o: $(ipmi_kcs_drv-objs)
- $(LD) -r -o $@ $(ipmi_kcs_drv-objs)
+ipmi_si.o: $(ipmi_si-objs)
+ $(LD) -r -o $@ $(ipmi_si-objs)
+
--- diff/drivers/char/ipmi/ipmi_devintf.c 2003-09-17 11:28:05.000000000 +0000
+++ source/drivers/char/ipmi/ipmi_devintf.c 2004-03-16 11:27:48.071879624 +0000
@@ -33,6 +33,7 @@
#include <linux/config.h>
#include <linux/module.h>
+#include <linux/moduleparam.h>
#include <linux/errno.h>
#include <asm/system.h>
#include <linux/sched.h>
@@ -44,6 +45,8 @@
#include <asm/semaphore.h>
#include <linux/init.h>
+#define IPMI_DEVINTF_VERSION "v31"
+
struct ipmi_file_private
{
ipmi_user_t user;
@@ -53,6 +56,8 @@ struct ipmi_file_private
struct fasync_struct *fasync_queue;
wait_queue_head_t wait;
struct semaphore recv_sem;
+ int default_retries;
+ unsigned int default_retry_time_ms;
};
static void file_receive_handler(struct ipmi_recv_msg *msg,
@@ -138,6 +143,10 @@ static int ipmi_open(struct inode *inode
priv->fasync_queue = NULL;
sema_init(&(priv->recv_sem), 1);
+ /* Use the low-level defaults. */
+ priv->default_retries = -1;
+ priv->default_retry_time_ms = 0;
+
return 0;
}
@@ -158,6 +167,63 @@ static int ipmi_release(struct inode *in
return 0;
}
+static int handle_send_req(ipmi_user_t user,
+ struct ipmi_req *req,
+ int retries,
+ unsigned int retry_time_ms)
+{
+ int rv;
+ struct ipmi_addr addr;
+ unsigned char *msgdata;
+
+ if (req->addr_len > sizeof(struct ipmi_addr))
+ return -EINVAL;
+
+ if (copy_from_user(&addr, req->addr, req->addr_len))
+ return -EFAULT;
+
+ msgdata = kmalloc(IPMI_MAX_MSG_LENGTH, GFP_KERNEL);
+ if (!msgdata)
+ return -ENOMEM;
+
+ /* From here out we cannot return, we must jump to "out" for
+ error exits to free msgdata. */
+
+ rv = ipmi_validate_addr(&addr, req->addr_len);
+ if (rv)
+ goto out;
+
+ if (req->msg.data != NULL) {
+ if (req->msg.data_len > IPMI_MAX_MSG_LENGTH) {
+ rv = -EMSGSIZE;
+ goto out;
+ }
+
+ if (copy_from_user(&msgdata,
+ req->msg.data,
+ req->msg.data_len))
+ {
+ rv = -EFAULT;
+ goto out;
+ }
+ } else {
+ req->msg.data_len = 0;
+ }
+ req->msg.data = msgdata;
+
+ rv = ipmi_request_settime(user,
+ &addr,
+ req->msgid,
+ &(req->msg),
+ NULL,
+ 0,
+ retries,
+ retry_time_ms);
+ out:
+ kfree(msgdata);
+ return rv;
+}
+
static int ipmi_ioctl(struct inode *inode,
struct file *file,
unsigned int cmd,
@@ -170,54 +236,33 @@ static int ipmi_ioctl(struct inode *ino
{
case IPMICTL_SEND_COMMAND:
{
- struct ipmi_req req;
- struct ipmi_addr addr;
- unsigned char msgdata[IPMI_MAX_MSG_LENGTH];
+ struct ipmi_req req;
if (copy_from_user(&req, (void *) data, sizeof(req))) {
rv = -EFAULT;
break;
}
- if (req.addr_len > sizeof(struct ipmi_addr))
- {
- rv = -EINVAL;
- break;
- }
+ rv = handle_send_req(priv->user,
+ &req,
+ priv->default_retries,
+ priv->default_retry_time_ms);
+ break;
+ }
- if (copy_from_user(&addr, req.addr, req.addr_len)) {
- rv = -EFAULT;
- break;
- }
+ case IPMICTL_SEND_COMMAND_SETTIME:
+ {
+ struct ipmi_req_settime req;
- rv = ipmi_validate_addr(&addr, req.addr_len);
- if (rv)
+ if (copy_from_user(&req, (void *) data, sizeof(req))) {
+ rv = -EFAULT;
break;
-
- if (req.msg.data != NULL) {
- if (req.msg.data_len > IPMI_MAX_MSG_LENGTH) {
- rv = -EMSGSIZE;
- break;
- }
-
- if (copy_from_user(&msgdata,
- req.msg.data,
- req.msg.data_len))
- {
- rv = -EFAULT;
- break;
- }
- } else {
- req.msg.data_len = 0;
}
- req.msg.data = msgdata;
-
- rv = ipmi_request(priv->user,
- &addr,
- req.msgid,
- &(req.msg),
- 0);
+ rv = handle_send_req(priv->user,
+ &req.req,
+ req.retries,
+ req.retry_time_ms);
break;
}
@@ -416,7 +461,36 @@ static int ipmi_ioctl(struct inode *ino
rv = 0;
break;
}
+ case IPMICTL_SET_TIMING_PARMS_CMD:
+ {
+ struct ipmi_timing_parms parms;
+
+ if (copy_from_user(&parms, (void *) data, sizeof(parms))) {
+ rv = -EFAULT;
+ break;
+ }
+
+ priv->default_retries = parms.retries;
+ priv->default_retry_time_ms = parms.retry_time_ms;
+ rv = 0;
+ break;
+ }
+
+ case IPMICTL_GET_TIMING_PARMS_CMD:
+ {
+ struct ipmi_timing_parms parms;
+
+ parms.retries = priv->default_retries;
+ parms.retry_time_ms = priv->default_retry_time_ms;
+ if (copy_to_user((void *) data, &parms, sizeof(parms))) {
+ rv = -EFAULT;
+ break;
+ }
+
+ rv = 0;
+ break;
+ }
}
return rv;
@@ -435,29 +509,30 @@ static struct file_operations ipmi_fops
#define DEVICE_NAME "ipmidev"
static int ipmi_major = 0;
-MODULE_PARM(ipmi_major, "i");
-
-#define MAX_DEVICES 10
+module_param(ipmi_major, int, 0);
+MODULE_PARM_DESC(ipmi_major, "Sets the major number of the IPMI device. By"
+ " default, or if you set it to zero, it will choose the next"
+ " available device. Setting it to -1 will disable the"
+ " interface. Other values will set the major device number"
+ " to that value.");
static void ipmi_new_smi(int if_num)
{
- if (if_num <= MAX_DEVICES) {
- devfs_mk_cdev(MKDEV(ipmi_major, if_num),
- S_IFCHR | S_IRUSR | S_IWUSR,
- "ipmidev/%d", if_num);
- }
+ devfs_mk_cdev(MKDEV(ipmi_major, if_num),
+ S_IFCHR | S_IRUSR | S_IWUSR,
+ "ipmidev/%d", if_num);
}
static void ipmi_smi_gone(int if_num)
{
- if (if_num <= MAX_DEVICES)
- devfs_remove("ipmidev/%d", if_num);
+ devfs_remove("ipmidev/%d", if_num);
}
static struct ipmi_smi_watcher smi_watcher =
{
- .new_smi = ipmi_new_smi,
- .smi_gone = ipmi_smi_gone,
+ .owner = THIS_MODULE,
+ .new_smi = ipmi_new_smi,
+ .smi_gone = ipmi_smi_gone,
};
static __init int init_ipmi_devintf(void)
@@ -467,6 +542,9 @@ static __init int init_ipmi_devintf(void
if (ipmi_major < 0)
return -EINVAL;
+ printk(KERN_INFO "ipmi device interface version "
+ IPMI_DEVINTF_VERSION "\n");
+
rv = register_chrdev(ipmi_major, DEVICE_NAME, &ipmi_fops);
if (rv < 0) {
printk(KERN_ERR "ipmi: can't get major %d\n", ipmi_major);
@@ -482,13 +560,10 @@ static __init int init_ipmi_devintf(void
rv = ipmi_smi_watcher_register(&smi_watcher);
if (rv) {
unregister_chrdev(ipmi_major, DEVICE_NAME);
- printk(KERN_WARNING "ipmi: can't register smi watcher");
+ printk(KERN_WARNING "ipmi: can't register smi watcher\n");
return rv;
}
- printk(KERN_INFO "ipmi: device interface at char major %d\n",
- ipmi_major);
-
return 0;
}
module_init(init_ipmi_devintf);
@@ -500,21 +575,5 @@ static __exit void cleanup_ipmi(void)
unregister_chrdev(ipmi_major, DEVICE_NAME);
}
module_exit(cleanup_ipmi);
-#ifndef MODULE
-static __init int ipmi_setup (char *str)
-{
- int x;
-
- if (get_option (&str, &x)) {
- /* ipmi=x sets the major number to x. */
- ipmi_major = x;
- } else if (!strcmp(str, "off")) {
- ipmi_major = -1;
- }
-
- return 1;
-}
-#endif
-__setup("ipmi=", ipmi_setup);
MODULE_LICENSE("GPL");
--- diff/drivers/char/ipmi/ipmi_kcs_sm.c 2004-03-11 10:20:23.000000000 +0000
+++ source/drivers/char/ipmi/ipmi_kcs_sm.c 2004-03-16 11:27:48.080878256 +0000
@@ -39,11 +39,13 @@
#include <linux/types.h>
-#include <asm/io.h>
+#include <linux/kernel.h> /* For printk. */
#include <asm/string.h> /* Gets rid of memcpy warning */
-#include <asm/system.h>
+#include <linux/ipmi_msgdefs.h> /* for completion codes */
-#include "ipmi_kcs_sm.h"
+#include "ipmi_si_sm.h"
+
+#define IPMI_KCS_VERSION "v31"
/* Set this if you want a printout of why the state machine was hosed
when it gets hosed. */
@@ -95,32 +97,28 @@ enum kcs_states {
#define OBF_RETRY_TIMEOUT 1000000
#define MAX_ERROR_RETRIES 10
-#define IPMI_ERR_MSG_TRUNCATED 0xc6
-#define IPMI_ERR_UNSPECIFIED 0xff
-
-struct kcs_data
+struct si_sm_data
{
- enum kcs_states state;
- unsigned int port;
- unsigned char *addr;
- unsigned char write_data[MAX_KCS_WRITE_SIZE];
- int write_pos;
- int write_count;
- int orig_write_count;
- unsigned char read_data[MAX_KCS_READ_SIZE];
- int read_pos;
- int truncated;
+ enum kcs_states state;
+ struct si_sm_io *io;
+ unsigned char write_data[MAX_KCS_WRITE_SIZE];
+ int write_pos;
+ int write_count;
+ int orig_write_count;
+ unsigned char read_data[MAX_KCS_READ_SIZE];
+ int read_pos;
+ int truncated;
unsigned int error_retries;
long ibf_timeout;
long obf_timeout;
};
-void init_kcs_data(struct kcs_data *kcs, unsigned int port, unsigned char *addr)
+static unsigned int init_kcs_data(struct si_sm_data *kcs,
+ struct si_sm_io *io)
{
kcs->state = KCS_IDLE;
- kcs->port = port;
- kcs->addr = addr;
+ kcs->io = io;
kcs->write_pos = 0;
kcs->write_count = 0;
kcs->orig_write_count = 0;
@@ -129,40 +127,29 @@ void init_kcs_data(struct kcs_data *kcs,
kcs->truncated = 0;
kcs->ibf_timeout = IBF_RETRY_TIMEOUT;
kcs->obf_timeout = OBF_RETRY_TIMEOUT;
-}
-/* Remember, init_one_kcs() insured port and addr can't both be set */
+ /* Reserve 2 I/O bytes. */
+ return 2;
+}
-static inline unsigned char read_status(struct kcs_data *kcs)
+static inline unsigned char read_status(struct si_sm_data *kcs)
{
- if (kcs->port)
- return inb(kcs->port + 1);
- else
- return readb(kcs->addr + 1);
+ return kcs->io->inputb(kcs->io, 1);
}
-static inline unsigned char read_data(struct kcs_data *kcs)
+static inline unsigned char read_data(struct si_sm_data *kcs)
{
- if (kcs->port)
- return inb(kcs->port + 0);
- else
- return readb(kcs->addr + 0);
+ return kcs->io->inputb(kcs->io, 0);
}
-static inline void write_cmd(struct kcs_data *kcs, unsigned char data)
+static inline void write_cmd(struct si_sm_data *kcs, unsigned char data)
{
- if (kcs->port)
- outb(data, kcs->port + 1);
- else
- writeb(data, kcs->addr + 1);
+ kcs->io->outputb(kcs->io, 1, data);
}
-static inline void write_data(struct kcs_data *kcs, unsigned char data)
+static inline void write_data(struct si_sm_data *kcs, unsigned char data)
{
- if (kcs->port)
- outb(data, kcs->port + 0);
- else
- writeb(data, kcs->addr + 0);
+ kcs->io->outputb(kcs->io, 0, data);
}
/* Control codes. */
@@ -182,14 +169,14 @@ static inline void write_data(struct kcs
#define GET_STATUS_OBF(status) ((status) & 0x01)
-static inline void write_next_byte(struct kcs_data *kcs)
+static inline void write_next_byte(struct si_sm_data *kcs)
{
write_data(kcs, kcs->write_data[kcs->write_pos]);
(kcs->write_pos)++;
(kcs->write_count)--;
}
-static inline void start_error_recovery(struct kcs_data *kcs, char *reason)
+static inline void start_error_recovery(struct si_sm_data *kcs, char *reason)
{
(kcs->error_retries)++;
if (kcs->error_retries > MAX_ERROR_RETRIES) {
@@ -202,7 +189,7 @@ static inline void start_error_recovery(
}
}
-static inline void read_next_byte(struct kcs_data *kcs)
+static inline void read_next_byte(struct si_sm_data *kcs)
{
if (kcs->read_pos >= MAX_KCS_READ_SIZE) {
/* Throw the data away and mark it truncated. */
@@ -215,9 +202,8 @@ static inline void read_next_byte(struct
write_data(kcs, KCS_READ_BYTE);
}
-static inline int check_ibf(struct kcs_data *kcs,
- unsigned char status,
- long time)
+static inline int check_ibf(struct si_sm_data *kcs, unsigned char status,
+ long time)
{
if (GET_STATUS_IBF(status)) {
kcs->ibf_timeout -= time;
@@ -232,9 +218,8 @@ static inline int check_ibf(struct kcs_d
return 1;
}
-static inline int check_obf(struct kcs_data *kcs,
- unsigned char status,
- long time)
+static inline int check_obf(struct si_sm_data *kcs, unsigned char status,
+ long time)
{
if (! GET_STATUS_OBF(status)) {
kcs->obf_timeout -= time;
@@ -248,13 +233,13 @@ static inline int check_obf(struct kcs_d
return 1;
}
-static void clear_obf(struct kcs_data *kcs, unsigned char status)
+static void clear_obf(struct si_sm_data *kcs, unsigned char status)
{
if (GET_STATUS_OBF(status))
read_data(kcs);
}
-static void restart_kcs_transaction(struct kcs_data *kcs)
+static void restart_kcs_transaction(struct si_sm_data *kcs)
{
kcs->write_count = kcs->orig_write_count;
kcs->write_pos = 0;
@@ -265,7 +250,8 @@ static void restart_kcs_transaction(stru
write_cmd(kcs, KCS_WRITE_START);
}
-int start_kcs_transaction(struct kcs_data *kcs, char *data, unsigned int size)
+static int start_kcs_transaction(struct si_sm_data *kcs, unsigned char *data,
+ unsigned int size)
{
if ((size < 2) || (size > MAX_KCS_WRITE_SIZE)) {
return -1;
@@ -287,7 +273,8 @@ int start_kcs_transaction(struct kcs_dat
return 0;
}
-int kcs_get_result(struct kcs_data *kcs, unsigned char *data, int length)
+static int get_kcs_result(struct si_sm_data *kcs, unsigned char *data,
+ unsigned int length)
{
if (length < kcs->read_pos) {
kcs->read_pos = length;
@@ -316,7 +303,7 @@ int kcs_get_result(struct kcs_data *kcs,
/* This implements the state machine defined in the IPMI manual, see
that for details on how this works. Divide that flowchart into
sections delimited by "Wait for IBF" and this will become clear. */
-enum kcs_result kcs_event(struct kcs_data *kcs, long time)
+static enum si_sm_result kcs_event(struct si_sm_data *kcs, long time)
{
unsigned char status;
unsigned char state;
@@ -328,7 +315,7 @@ enum kcs_result kcs_event(struct kcs_dat
#endif
/* All states wait for ibf, so just do it here. */
if (!check_ibf(kcs, status, time))
- return KCS_CALL_WITH_DELAY;
+ return SI_SM_CALL_WITH_DELAY;
/* Just about everything looks at the KCS state, so grab that, too. */
state = GET_STATUS_STATE(status);
@@ -339,9 +326,9 @@ enum kcs_result kcs_event(struct kcs_dat
clear_obf(kcs, status);
if (GET_STATUS_ATN(status))
- return KCS_ATTN;
+ return SI_SM_ATTN;
else
- return KCS_SM_IDLE;
+ return SI_SM_IDLE;
case KCS_START_OP:
if (state != KCS_IDLE) {
@@ -408,7 +395,7 @@ enum kcs_result kcs_event(struct kcs_dat
if (state == KCS_READ_STATE) {
if (! check_obf(kcs, status, time))
- return KCS_CALL_WITH_DELAY;
+ return SI_SM_CALL_WITH_DELAY;
read_next_byte(kcs);
} else {
/* We don't implement this exactly like the state
@@ -421,7 +408,7 @@ enum kcs_result kcs_event(struct kcs_dat
clear_obf(kcs, status);
kcs->orig_write_count = 0;
kcs->state = KCS_IDLE;
- return KCS_TRANSACTION_COMPLETE;
+ return SI_SM_TRANSACTION_COMPLETE;
}
break;
@@ -444,7 +431,7 @@ enum kcs_result kcs_event(struct kcs_dat
break;
}
if (! check_obf(kcs, status, time))
- return KCS_CALL_WITH_DELAY;
+ return SI_SM_CALL_WITH_DELAY;
clear_obf(kcs, status);
write_data(kcs, KCS_READ_BYTE);
@@ -459,14 +446,14 @@ enum kcs_result kcs_event(struct kcs_dat
}
if (! check_obf(kcs, status, time))
- return KCS_CALL_WITH_DELAY;
+ return SI_SM_CALL_WITH_DELAY;
clear_obf(kcs, status);
if (kcs->orig_write_count) {
restart_kcs_transaction(kcs);
} else {
kcs->state = KCS_IDLE;
- return KCS_TRANSACTION_COMPLETE;
+ return SI_SM_TRANSACTION_COMPLETE;
}
break;
@@ -475,14 +462,42 @@ enum kcs_result kcs_event(struct kcs_dat
}
if (kcs->state == KCS_HOSED) {
- init_kcs_data(kcs, kcs->port, kcs->addr);
- return KCS_SM_HOSED;
+ init_kcs_data(kcs, kcs->io);
+ return SI_SM_HOSED;
}
- return KCS_CALL_WITHOUT_DELAY;
+ return SI_SM_CALL_WITHOUT_DELAY;
}
-int kcs_size(void)
+static int kcs_size(void)
{
- return sizeof(struct kcs_data);
+ return sizeof(struct si_sm_data);
}
+
+static int kcs_detect(struct si_sm_data *kcs)
+{
+ /* It's impossible for the KCS status register to be all 1's,
+ (assuming a properly functioning, self-initialized BMC)
+ but that's what you get from reading a bogus address, so we
+ test that first. */
+ if (read_status(kcs) == 0xff)
+ return 1;
+
+ return 0;
+}
+
+static void kcs_cleanup(struct si_sm_data *kcs)
+{
+}
+
+struct si_sm_handlers kcs_smi_handlers =
+{
+ .version = IPMI_KCS_VERSION,
+ .init_data = init_kcs_data,
+ .start_transaction = start_kcs_transaction,
+ .get_result = get_kcs_result,
+ .event = kcs_event,
+ .detect = kcs_detect,
+ .cleanup = kcs_cleanup,
+ .size = kcs_size,
+};
--- diff/drivers/char/ipmi/ipmi_msghandler.c 2003-10-27 09:20:37.000000000 +0000
+++ source/drivers/char/ipmi/ipmi_msghandler.c 2004-03-16 11:27:48.090876736 +0000
@@ -44,16 +44,21 @@
#include <linux/ipmi_smi.h>
#include <linux/notifier.h>
#include <linux/init.h>
+#include <linux/proc_fs.h>
+
+#define IPMI_MSGHANDLER_VERSION "v31"
struct ipmi_recv_msg *ipmi_alloc_recv_msg(void);
static int ipmi_init_msghandler(void);
static int initialized = 0;
+static struct proc_dir_entry *proc_ipmi_root = NULL;
+
#define MAX_EVENTS_IN_QUEUE 25
/* Don't let a message sit in a queue forever, always time it with at lest
- the max message timer. */
+ the max message timer. This is in milliseconds. */
#define MAX_MSG_TIMEOUT 60000
struct ipmi_user
@@ -82,7 +87,8 @@ struct cmd_rcvr
struct seq_table
{
- int inuse : 1;
+ unsigned int inuse : 1;
+ unsigned int broadcast : 1;
unsigned long timeout;
unsigned long orig_timeout;
@@ -111,10 +117,19 @@ struct seq_table
#define NEXT_SEQID(seqid) (((seqid) + 1) & 0x3fffff)
+struct ipmi_channel
+{
+ unsigned char medium;
+ unsigned char protocol;
+};
#define IPMI_IPMB_NUM_SEQ 64
+#define IPMI_MAX_CHANNELS 8
struct ipmi_smi
{
+ /* What interface number are we? */
+ int intf_num;
+
/* The list of upper layers that are using me. We read-lock
this when delivering messages to the upper layer to keep
the user from going away while we are processing the
@@ -123,6 +138,9 @@ struct ipmi_smi
rwlock_t users_lock;
struct list_head users;
+ /* Used for wake ups at startup. */
+ wait_queue_head_t waitq;
+
/* The IPMI version of the BMC on the other end. */
unsigned char version_major;
unsigned char version_minor;
@@ -182,6 +200,86 @@ struct ipmi_smi
it. Note that the message will still be freed by the
caller. This only works on the system interface. */
void (*null_user_handler)(ipmi_smi_t intf, struct ipmi_smi_msg *msg);
+
+ /* When we are scanning the channels for an SMI, this will
+ tell which channel we are scanning. */
+ int curr_channel;
+
+ /* Channel information */
+ struct ipmi_channel channels[IPMI_MAX_CHANNELS];
+
+ /* Proc FS stuff. */
+ struct proc_dir_entry *proc_dir;
+ char proc_dir_name[10];
+
+ spinlock_t counter_lock; /* For making counters atomic. */
+
+ /* Commands we got that were invalid. */
+ unsigned int sent_invalid_commands;
+
+ /* Commands we sent to the MC. */
+ unsigned int sent_local_commands;
+ /* Responses from the MC that were delivered to a user. */
+ unsigned int handled_local_responses;
+ /* Responses from the MC that were not delivered to a user. */
+ unsigned int unhandled_local_responses;
+
+ /* Commands we sent out to the IPMB bus. */
+ unsigned int sent_ipmb_commands;
+ /* Commands sent on the IPMB that had errors on the SEND CMD */
+ unsigned int sent_ipmb_command_errs;
+ /* Each retransmit increments this count. */
+ unsigned int retransmitted_ipmb_commands;
+ /* When a message times out (runs out of retransmits) this is
+ incremented. */
+ unsigned int timed_out_ipmb_commands;
+
+ /* This is like above, but for broadcasts. Broadcasts are
+ *not* included in the above count (they are expected to
+ time out). */
+ unsigned int timed_out_ipmb_broadcasts;
+
+ /* Responses I have sent to the IPMB bus. */
+ unsigned int sent_ipmb_responses;
+
+ /* The response was delivered to the user. */
+ unsigned int handled_ipmb_responses;
+ /* The response had invalid data in it. */
+ unsigned int invalid_ipmb_responses;
+ /* The response didn't have anyone waiting for it. */
+ unsigned int unhandled_ipmb_responses;
+
+ /* Commands we sent out to the IPMB bus. */
+ unsigned int sent_lan_commands;
+ /* Commands sent on the IPMB that had errors on the SEND CMD */
+ unsigned int sent_lan_command_errs;
+ /* Each retransmit increments this count. */
+ unsigned int retransmitted_lan_commands;
+ /* When a message times out (runs out of retransmits) this is
+ incremented. */
+ unsigned int timed_out_lan_commands;
+
+ /* Responses I have sent to the IPMB bus. */
+ unsigned int sent_lan_responses;
+
+ /* The response was delivered to the user. */
+ unsigned int handled_lan_responses;
+ /* The response had invalid data in it. */
+ unsigned int invalid_lan_responses;
+ /* The response didn't have anyone waiting for it. */
+ unsigned int unhandled_lan_responses;
+
+ /* The command was delivered to the user. */
+ unsigned int handled_commands;
+ /* The command had invalid data in it. */
+ unsigned int invalid_commands;
+ /* The command didn't have anyone waiting for it. */
+ unsigned int unhandled_commands;
+
+ /* Invalid data in an event. */
+ unsigned int invalid_events;
+ /* Events that were received with the proper format. */
+ unsigned int events;
};
int
@@ -264,6 +362,21 @@ int ipmi_smi_watcher_unregister(struct i
return 0;
}
+static void
+call_smi_watchers(int i)
+{
+ struct ipmi_smi_watcher *w;
+
+ down_read(&smi_watchers_sem);
+ list_for_each_entry(w, &smi_watchers, link) {
+ if (try_module_get(w->owner)) {
+ w->new_smi(i);
+ module_put(w->owner);
+ }
+ }
+ up_read(&smi_watchers_sem);
+}
+
int
ipmi_addr_equal(struct ipmi_addr *addr1, struct ipmi_addr *addr2)
{
@@ -293,6 +406,19 @@ ipmi_addr_equal(struct ipmi_addr *addr1,
&& (ipmb_addr1->lun == ipmb_addr2->lun));
}
+ if (addr1->addr_type == IPMI_LAN_ADDR_TYPE) {
+ struct ipmi_lan_addr *lan_addr1
+ = (struct ipmi_lan_addr *) addr1;
+ struct ipmi_lan_addr *lan_addr2
+ = (struct ipmi_lan_addr *) addr2;
+
+ return ((lan_addr1->remote_SWID == lan_addr2->remote_SWID)
+ && (lan_addr1->local_SWID == lan_addr2->local_SWID)
+ && (lan_addr1->session_handle
+ == lan_addr2->session_handle)
+ && (lan_addr1->lun == lan_addr2->lun));
+ }
+
return 1;
}
@@ -322,6 +448,13 @@ int ipmi_validate_addr(struct ipmi_addr
return 0;
}
+ if (addr->addr_type == IPMI_LAN_ADDR_TYPE) {
+ if (len < sizeof(struct ipmi_lan_addr)) {
+ return -EINVAL;
+ }
+ return 0;
+ }
+
return -EINVAL;
}
@@ -341,7 +474,7 @@ unsigned int ipmi_addr_length(int addr_t
static void deliver_response(struct ipmi_recv_msg *msg)
{
- msg->user->handler->ipmi_recv_hndl(msg, msg->user->handler_data);
+ msg->user->handler->ipmi_recv_hndl(msg, msg->user->handler_data);
}
/* Find the next sequence number not being used and add the given
@@ -351,6 +484,7 @@ static int intf_next_seq(ipmi_smi_t
struct ipmi_recv_msg *recv_msg,
unsigned long timeout,
int retries,
+ int broadcast,
unsigned char *seq,
long *seqid)
{
@@ -373,6 +507,7 @@ static int intf_next_seq(ipmi_smi_t
intf->seq_table[i].timeout = MAX_MSG_TIMEOUT;
intf->seq_table[i].orig_timeout = timeout;
intf->seq_table[i].retries_left = retries;
+ intf->seq_table[i].broadcast = broadcast;
intf->seq_table[i].inuse = 1;
intf->seq_table[i].seqid = NEXT_SEQID(intf->seq_table[i].seqid);
*seq = i;
@@ -425,8 +560,8 @@ static int intf_find_seq(ipmi_smi_t
/* Start the timer for a specific sequence table entry. */
-static int intf_start_seq_timer(ipmi_smi_t intf,
- long msgid)
+static int intf_start_seq_timer(ipmi_smi_t intf,
+ long msgid)
{
int rv = -ENODEV;
unsigned long flags;
@@ -451,6 +586,46 @@ static int intf_start_seq_timer(ipmi_smi
return rv;
}
+/* Got an error for the send message for a specific sequence number. */
+static int intf_err_seq(ipmi_smi_t intf,
+ long msgid,
+ unsigned int err)
+{
+ int rv = -ENODEV;
+ unsigned long flags;
+ unsigned char seq;
+ unsigned long seqid;
+ struct ipmi_recv_msg *msg = NULL;
+
+
+ GET_SEQ_FROM_MSGID(msgid, seq, seqid);
+
+ spin_lock_irqsave(&(intf->seq_lock), flags);
+ /* We do this verification because the user can be deleted
+ while a message is outstanding. */
+ if ((intf->seq_table[seq].inuse)
+ && (intf->seq_table[seq].seqid == seqid))
+ {
+ struct seq_table *ent = &(intf->seq_table[seq]);
+
+ ent->inuse = 0;
+ msg = ent->recv_msg;
+ rv = 0;
+ }
+ spin_unlock_irqrestore(&(intf->seq_lock), flags);
+
+ if (msg) {
+ msg->recv_type = IPMI_RESPONSE_RECV_TYPE;
+ msg->msg_data[0] = err;
+ msg->msg.netfn |= 1; /* Convert to a response. */
+ msg->msg.data_len = 1;
+ msg->msg.data = msg->msg_data;
+ deliver_response(msg);
+ }
+
+ return rv;
+}
+
int ipmi_create_user(unsigned int if_num,
struct ipmi_user_hndl *handler,
@@ -523,15 +698,14 @@ static int ipmi_destroy_user_nolock(ipmi
{
int rv = -ENODEV;
ipmi_user_t t_user;
- struct list_head *entry, *entry2;
+ struct cmd_rcvr *rcvr, *rcvr2;
int i;
unsigned long flags;
/* Find the user and delete them from the list. */
- list_for_each(entry, &(user->intf->users)) {
- t_user = list_entry(entry, struct ipmi_user, link);
+ list_for_each_entry(t_user, &(user->intf->users), link) {
if (t_user == user) {
- list_del(entry);
+ list_del(&t_user->link);
rv = 0;
break;
}
@@ -554,11 +728,9 @@ static int ipmi_destroy_user_nolock(ipmi
/* Remove the user from the command receiver's table. */
write_lock_irqsave(&(user->intf->cmd_rcvr_lock), flags);
- list_for_each_safe(entry, entry2, &(user->intf->cmd_rcvrs)) {
- struct cmd_rcvr *rcvr;
- rcvr = list_entry(entry, struct cmd_rcvr, link);
+ list_for_each_entry_safe(rcvr, rcvr2, &(user->intf->cmd_rcvrs), link) {
if (rcvr->user == user) {
- list_del(entry);
+ list_del(&rcvr->link);
kfree(rcvr);
}
}
@@ -621,8 +793,7 @@ unsigned char ipmi_get_my_LUN(ipmi_user_
int ipmi_set_gets_events(ipmi_user_t user, int val)
{
unsigned long flags;
- struct list_head *e, *e2;
- struct ipmi_recv_msg *msg;
+ struct ipmi_recv_msg *msg, *msg2;
read_lock(&(user->intf->users_lock));
spin_lock_irqsave(&(user->intf->events_lock), flags);
@@ -630,9 +801,8 @@ int ipmi_set_gets_events(ipmi_user_t use
if (val) {
/* Deliver any queued events. */
- list_for_each_safe(e, e2, &(user->intf->waiting_events)) {
- msg = list_entry(e, struct ipmi_recv_msg, link);
- list_del(e);
+ list_for_each_entry_safe(msg, msg2, &(user->intf->waiting_events), link) {
+ list_del(&msg->link);
msg->user = user;
deliver_response(msg);
}
@@ -648,7 +818,7 @@ int ipmi_register_for_cmd(ipmi_user_t
unsigned char netfn,
unsigned char cmd)
{
- struct list_head *entry;
+ struct cmd_rcvr *cmp;
unsigned long flags;
struct cmd_rcvr *rcvr;
int rv = 0;
@@ -666,9 +836,7 @@ int ipmi_register_for_cmd(ipmi_user_t
}
/* Make sure the command/netfn is not already registered. */
- list_for_each(entry, &(user->intf->cmd_rcvrs)) {
- struct cmd_rcvr *cmp;
- cmp = list_entry(entry, struct cmd_rcvr, link);
+ list_for_each_entry(cmp, &(user->intf->cmd_rcvrs), link) {
if ((cmp->netfn == netfn) && (cmp->cmd == cmd)) {
rv = -EBUSY;
break;
@@ -695,7 +863,6 @@ int ipmi_unregister_for_cmd(ipmi_user_t
unsigned char netfn,
unsigned char cmd)
{
- struct list_head *entry;
unsigned long flags;
struct cmd_rcvr *rcvr;
int rv = -ENOENT;
@@ -703,11 +870,10 @@ int ipmi_unregister_for_cmd(ipmi_user_t
read_lock(&(user->intf->users_lock));
write_lock_irqsave(&(user->intf->cmd_rcvr_lock), flags);
/* Make sure the command/netfn is not already registered. */
- list_for_each(entry, &(user->intf->cmd_rcvrs)) {
- rcvr = list_entry(entry, struct cmd_rcvr, link);
+ list_for_each_entry(rcvr, &(user->intf->cmd_rcvrs), link) {
if ((rcvr->netfn == netfn) && (rcvr->cmd == cmd)) {
rv = 0;
- list_del(entry);
+ list_del(&rcvr->link);
kfree(rcvr);
break;
}
@@ -771,6 +937,43 @@ static inline void format_ipmb_msg(struc
smi_msg->msgid = msgid;
}
+static inline void format_lan_msg(struct ipmi_smi_msg *smi_msg,
+ struct ipmi_msg *msg,
+ struct ipmi_lan_addr *lan_addr,
+ long msgid,
+ unsigned char ipmb_seq,
+ unsigned char source_lun)
+{
+ /* Format the IPMB header data. */
+ smi_msg->data[0] = (IPMI_NETFN_APP_REQUEST << 2);
+ smi_msg->data[1] = IPMI_SEND_MSG_CMD;
+ smi_msg->data[2] = lan_addr->channel;
+ smi_msg->data[3] = lan_addr->session_handle;
+ smi_msg->data[4] = lan_addr->remote_SWID;
+ smi_msg->data[5] = (msg->netfn << 2) | (lan_addr->lun & 0x3);
+ smi_msg->data[6] = ipmb_checksum(&(smi_msg->data[4]), 2);
+ smi_msg->data[7] = lan_addr->local_SWID;
+ smi_msg->data[8] = (ipmb_seq << 2) | source_lun;
+ smi_msg->data[9] = msg->cmd;
+
+ /* Now tack on the data to the message. */
+ if (msg->data_len > 0)
+ memcpy(&(smi_msg->data[10]), msg->data,
+ msg->data_len);
+ smi_msg->data_size = msg->data_len + 10;
+
+ /* Now calculate the checksum and tack it on. */
+ smi_msg->data[smi_msg->data_size]
+ = ipmb_checksum(&(smi_msg->data[7]),
+ smi_msg->data_size-7);
+
+ /* Add on the checksum size and the offset from the
+ broadcast. */
+ smi_msg->data_size += 1;
+
+ smi_msg->msgid = msgid;
+}
+
/* Separate from ipmi_request so that the user does not have to be
supplied in certain circumstances (mainly at panic time). If
messages are supplied, they will be freed, even if an error
@@ -780,11 +983,14 @@ static inline int i_ipmi_request(ipmi_us
struct ipmi_addr *addr,
long msgid,
struct ipmi_msg *msg,
+ void *user_msg_data,
void *supplied_smi,
struct ipmi_recv_msg *supplied_recv,
int priority,
unsigned char source_address,
- unsigned char source_lun)
+ unsigned char source_lun,
+ int retries,
+ unsigned int retry_time_ms)
{
int rv = 0;
struct ipmi_smi_msg *smi_msg;
@@ -800,6 +1006,7 @@ static inline int i_ipmi_request(ipmi_us
return -ENOMEM;
}
}
+ recv_msg->user_msg_data = user_msg_data;
if (supplied_smi) {
smi_msg = (struct ipmi_smi_msg *) supplied_smi;
@@ -811,11 +1018,6 @@ static inline int i_ipmi_request(ipmi_us
}
}
- if (addr->channel > IPMI_NUM_CHANNELS) {
- rv = -EINVAL;
- goto out_err;
- }
-
recv_msg->user = user;
recv_msg->msgid = msgid;
/* Store the message to send in the receive message so timeout
@@ -825,10 +1027,20 @@ static inline int i_ipmi_request(ipmi_us
if (addr->addr_type == IPMI_SYSTEM_INTERFACE_ADDR_TYPE) {
struct ipmi_system_interface_addr *smi_addr;
+ if (msg->netfn & 1) {
+ /* Responses are not allowed to the SMI. */
+ rv = -EINVAL;
+ goto out_err;
+ }
smi_addr = (struct ipmi_system_interface_addr *) addr;
- if (smi_addr->lun > 3)
- return -EINVAL;
+ if (smi_addr->lun > 3) {
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_invalid_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+ rv = -EINVAL;
+ goto out_err;
+ }
memcpy(&recv_msg->addr, smi_addr, sizeof(*smi_addr));
@@ -839,11 +1051,17 @@ static inline int i_ipmi_request(ipmi_us
{
/* We don't let the user do these, since we manage
the sequence numbers. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_invalid_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
rv = -EINVAL;
goto out_err;
}
if ((msg->data_len + 2) > IPMI_MAX_MSG_LENGTH) {
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_invalid_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
rv = -EMSGSIZE;
goto out_err;
}
@@ -855,41 +1073,69 @@ static inline int i_ipmi_request(ipmi_us
if (msg->data_len > 0)
memcpy(&(smi_msg->data[2]), msg->data, msg->data_len);
smi_msg->data_size = msg->data_len + 2;
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_local_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
} else if ((addr->addr_type == IPMI_IPMB_ADDR_TYPE)
|| (addr->addr_type == IPMI_IPMB_BROADCAST_ADDR_TYPE))
{
struct ipmi_ipmb_addr *ipmb_addr;
unsigned char ipmb_seq;
long seqid;
- int broadcast;
- int retries;
+ int broadcast = 0;
+
+ if (addr->channel > IPMI_NUM_CHANNELS) {
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_invalid_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+ rv = -EINVAL;
+ goto out_err;
+ }
- if (addr == NULL) {
+ if (intf->channels[addr->channel].medium
+ != IPMI_CHANNEL_MEDIUM_IPMB)
+ {
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_invalid_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
rv = -EINVAL;
goto out_err;
}
+ if (retries < 0) {
+ if (addr->addr_type == IPMI_IPMB_BROADCAST_ADDR_TYPE)
+ retries = 0; /* Don't retry broadcasts. */
+ else
+ retries = 4;
+ }
if (addr->addr_type == IPMI_IPMB_BROADCAST_ADDR_TYPE) {
/* Broadcasts add a zero at the beginning of the
message, but otherwise is the same as an IPMB
address. */
addr->addr_type = IPMI_IPMB_ADDR_TYPE;
broadcast = 1;
- retries = 0; /* Don't retry broadcasts. */
- } else {
- broadcast = 0;
- retries = 4;
}
+
+ /* Default to 1 second retries. */
+ if (retry_time_ms == 0)
+ retry_time_ms = 1000;
+
/* 9 for the header and 1 for the checksum, plus
possibly one for the broadcast. */
if ((msg->data_len + 10 + broadcast) > IPMI_MAX_MSG_LENGTH) {
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_invalid_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
rv = -EMSGSIZE;
goto out_err;
}
ipmb_addr = (struct ipmi_ipmb_addr *) addr;
if (ipmb_addr->lun > 3) {
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_invalid_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
rv = -EINVAL;
goto out_err;
}
@@ -899,21 +1145,32 @@ static inline int i_ipmi_request(ipmi_us
if (recv_msg->msg.netfn & 0x1) {
/* It's a response, so use the user's sequence
from msgid. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_ipmb_responses++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
format_ipmb_msg(smi_msg, msg, ipmb_addr, msgid,
msgid, broadcast,
source_address, source_lun);
+
+ /* Save the receive message so we can use it
+ to deliver the response. */
+ smi_msg->user_data = recv_msg;
} else {
/* It's a command, so get a sequence for it. */
spin_lock_irqsave(&(intf->seq_lock), flags);
+ spin_lock(&intf->counter_lock);
+ intf->sent_ipmb_commands++;
+ spin_unlock(&intf->counter_lock);
+
/* Create a sequence number with a 1 second
timeout and 4 retries. */
- /* FIXME - magic number for the timeout. */
rv = intf_next_seq(intf,
recv_msg,
- 1000,
+ retry_time_ms,
retries,
+ broadcast,
&ipmb_seq,
&seqid);
if (rv) {
@@ -947,18 +1204,132 @@ static inline int i_ipmi_request(ipmi_us
to be correct. */
spin_unlock_irqrestore(&(intf->seq_lock), flags);
}
+ } else if (addr->addr_type == IPMI_LAN_ADDR_TYPE) {
+ struct ipmi_lan_addr *lan_addr;
+ unsigned char ipmb_seq;
+ long seqid;
+
+ if (addr->channel > IPMI_NUM_CHANNELS) {
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_invalid_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+ rv = -EINVAL;
+ goto out_err;
+ }
+
+ if ((intf->channels[addr->channel].medium
+ != IPMI_CHANNEL_MEDIUM_8023LAN)
+ && (intf->channels[addr->channel].medium
+ != IPMI_CHANNEL_MEDIUM_ASYNC))
+ {
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_invalid_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+ rv = -EINVAL;
+ goto out_err;
+ }
+
+ retries = 4;
+
+ /* Default to 1 second retries. */
+ if (retry_time_ms == 0)
+ retry_time_ms = 1000;
+
+ /* 11 for the header and 1 for the checksum. */
+ if ((msg->data_len + 12) > IPMI_MAX_MSG_LENGTH) {
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_invalid_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+ rv = -EMSGSIZE;
+ goto out_err;
+ }
+
+ lan_addr = (struct ipmi_lan_addr *) addr;
+ if (lan_addr->lun > 3) {
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_invalid_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+ rv = -EINVAL;
+ goto out_err;
+ }
+
+ memcpy(&recv_msg->addr, lan_addr, sizeof(*lan_addr));
+
+ if (recv_msg->msg.netfn & 0x1) {
+ /* It's a response, so use the user's sequence
+ from msgid. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_lan_responses++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+ format_lan_msg(smi_msg, msg, lan_addr, msgid,
+ msgid, source_lun);
+
+ /* Save the receive message so we can use it
+ to deliver the response. */
+ smi_msg->user_data = recv_msg;
+ } else {
+ /* It's a command, so get a sequence for it. */
+
+ spin_lock_irqsave(&(intf->seq_lock), flags);
+
+ spin_lock(&intf->counter_lock);
+ intf->sent_lan_commands++;
+ spin_unlock(&intf->counter_lock);
+
+ /* Create a sequence number with a 1 second
+ timeout and 4 retries. */
+ rv = intf_next_seq(intf,
+ recv_msg,
+ retry_time_ms,
+ retries,
+ 0,
+ &ipmb_seq,
+ &seqid);
+ if (rv) {
+ /* We have used up all the sequence numbers,
+ probably, so abort. */
+ spin_unlock_irqrestore(&(intf->seq_lock),
+ flags);
+ goto out_err;
+ }
+
+ /* Store the sequence number in the message,
+ so that when the send message response
+ comes back we can start the timer. */
+ format_lan_msg(smi_msg, msg, lan_addr,
+ STORE_SEQ_IN_MSGID(ipmb_seq, seqid),
+ ipmb_seq, source_lun);
+
+ /* Copy the message into the recv message data, so we
+ can retransmit it later if necessary. */
+ memcpy(recv_msg->msg_data, smi_msg->data,
+ smi_msg->data_size);
+ recv_msg->msg.data = recv_msg->msg_data;
+ recv_msg->msg.data_len = smi_msg->data_size;
+
+ /* We don't unlock until here, because we need
+ to copy the completed message into the
+ recv_msg before we release the lock.
+ Otherwise, race conditions may bite us. I
+ know that's pretty paranoid, but I prefer
+ to be correct. */
+ spin_unlock_irqrestore(&(intf->seq_lock), flags);
+ }
} else {
/* Unknown address type. */
- rv = -EINVAL;
- goto out_err;
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->sent_invalid_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+ rv = -EINVAL;
+ goto out_err;
}
#if DEBUG_MSGING
{
- int m;
- for (m=0; m<smi_msg->data_size; m++)
- printk(" %2.2x", smi_msg->data[m]);
- printk("\n");
+ int m;
+ for (m=0; m<smi_msg->data_size; m++)
+ printk(" %2.2x", smi_msg->data[m]);
+ printk("\n");
}
#endif
intf->handlers->sender(intf->send_info, smi_msg, priority);
@@ -975,6 +1346,7 @@ int ipmi_request(ipmi_user_t user,
struct ipmi_addr *addr,
long msgid,
struct ipmi_msg *msg,
+ void *user_msg_data,
int priority)
{
return i_ipmi_request(user,
@@ -982,16 +1354,42 @@ int ipmi_request(ipmi_user_t user,
addr,
msgid,
msg,
+ user_msg_data,
NULL, NULL,
priority,
user->intf->my_address,
- user->intf->my_lun);
+ user->intf->my_lun,
+ -1, 0);
+}
+
+int ipmi_request_settime(ipmi_user_t user,
+ struct ipmi_addr *addr,
+ long msgid,
+ struct ipmi_msg *msg,
+ void *user_msg_data,
+ int priority,
+ int retries,
+ unsigned int retry_time_ms)
+{
+ return i_ipmi_request(user,
+ user->intf,
+ addr,
+ msgid,
+ msg,
+ user_msg_data,
+ NULL, NULL,
+ priority,
+ user->intf->my_address,
+ user->intf->my_lun,
+ retries,
+ retry_time_ms);
}
int ipmi_request_supply_msgs(ipmi_user_t user,
struct ipmi_addr *addr,
long msgid,
struct ipmi_msg *msg,
+ void *user_msg_data,
void *supplied_smi,
struct ipmi_recv_msg *supplied_recv,
int priority)
@@ -1001,17 +1399,20 @@ int ipmi_request_supply_msgs(ipmi_user_t
addr,
msgid,
msg,
+ user_msg_data,
supplied_smi,
supplied_recv,
priority,
user->intf->my_address,
- user->intf->my_lun);
+ user->intf->my_lun,
+ -1, 0);
}
int ipmi_request_with_source(ipmi_user_t user,
struct ipmi_addr *addr,
long msgid,
struct ipmi_msg *msg,
+ void *user_msg_data,
int priority,
unsigned char source_address,
unsigned char source_lun)
@@ -1021,10 +1422,215 @@ int ipmi_request_with_source(ipmi_user_t
addr,
msgid,
msg,
+ user_msg_data,
NULL, NULL,
priority,
source_address,
- source_lun);
+ source_lun,
+ -1, 0);
+}
+
+static int ipmb_file_read_proc(char *page, char **start, off_t off,
+ int count, int *eof, void *data)
+{
+ char *out = (char *) page;
+ ipmi_smi_t intf = data;
+
+ return sprintf(out, "%x\n", intf->my_address);
+}
+
+static int version_file_read_proc(char *page, char **start, off_t off,
+ int count, int *eof, void *data)
+{
+ char *out = (char *) page;
+ ipmi_smi_t intf = data;
+
+ return sprintf(out, "%d.%d\n",
+ intf->version_major, intf->version_minor);
+}
+
+static int stat_file_read_proc(char *page, char **start, off_t off,
+ int count, int *eof, void *data)
+{
+ char *out = (char *) page;
+ ipmi_smi_t intf = data;
+
+ out += sprintf(out, "sent_invalid_commands: %d\n",
+ intf->sent_invalid_commands);
+ out += sprintf(out, "sent_local_commands: %d\n",
+ intf->sent_local_commands);
+ out += sprintf(out, "handled_local_responses: %d\n",
+ intf->handled_local_responses);
+ out += sprintf(out, "unhandled_local_responses: %d\n",
+ intf->unhandled_local_responses);
+ out += sprintf(out, "sent_ipmb_commands: %d\n",
+ intf->sent_ipmb_commands);
+ out += sprintf(out, "sent_ipmb_command_errs: %d\n",
+ intf->sent_ipmb_command_errs);
+ out += sprintf(out, "retransmitted_ipmb_commands: %d\n",
+ intf->retransmitted_ipmb_commands);
+ out += sprintf(out, "timed_out_ipmb_commands: %d\n",
+ intf->timed_out_ipmb_commands);
+ out += sprintf(out, "timed_out_ipmb_broadcasts: %d\n",
+ intf->timed_out_ipmb_broadcasts);
+ out += sprintf(out, "sent_ipmb_responses: %d\n",
+ intf->sent_ipmb_responses);
+ out += sprintf(out, "handled_ipmb_responses: %d\n",
+ intf->handled_ipmb_responses);
+ out += sprintf(out, "invalid_ipmb_responses: %d\n",
+ intf->invalid_ipmb_responses);
+ out += sprintf(out, "unhandled_ipmb_responses: %d\n",
+ intf->unhandled_ipmb_responses);
+ out += sprintf(out, "sent_lan_commands: %d\n",
+ intf->sent_lan_commands);
+ out += sprintf(out, "sent_lan_command_errs: %d\n",
+ intf->sent_lan_command_errs);
+ out += sprintf(out, "retransmitted_lan_commands: %d\n",
+ intf->retransmitted_lan_commands);
+ out += sprintf(out, "timed_out_lan_commands: %d\n",
+ intf->timed_out_lan_commands);
+ out += sprintf(out, "sent_lan_responses: %d\n",
+ intf->sent_lan_responses);
+ out += sprintf(out, "handled_lan_responses: %d\n",
+ intf->handled_lan_responses);
+ out += sprintf(out, "invalid_lan_responses: %d\n",
+ intf->invalid_lan_responses);
+ out += sprintf(out, "unhandled_lan_responses: %d\n",
+ intf->unhandled_lan_responses);
+ out += sprintf(out, "handled_commands: %d\n",
+ intf->handled_commands);
+ out += sprintf(out, "invalid_commands: %d\n",
+ intf->invalid_commands);
+ out += sprintf(out, "unhandled_commands: %d\n",
+ intf->unhandled_commands);
+ out += sprintf(out, "invalid_events: %d\n",
+ intf->invalid_events);
+ out += sprintf(out, "events: %d\n",
+ intf->events);
+
+ return (out - ((char *) page));
+}
+
+int ipmi_smi_add_proc_entry(ipmi_smi_t smi, char *name,
+ read_proc_t *read_proc, write_proc_t *write_proc,
+ void *data, struct module *owner)
+{
+ struct proc_dir_entry *file;
+ int rv = 0;
+
+ file = create_proc_entry(name, 0, smi->proc_dir);
+ if (!file)
+ rv = -ENOMEM;
+ else {
+ file->nlink = 1;
+ file->data = data;
+ file->read_proc = read_proc;
+ file->write_proc = write_proc;
+ file->owner = owner;
+ }
+
+ return rv;
+}
+
+static int add_proc_entries(ipmi_smi_t smi, int num)
+{
+ int rv = 0;
+
+ sprintf(smi->proc_dir_name, "%d", num);
+ smi->proc_dir = proc_mkdir(smi->proc_dir_name, proc_ipmi_root);
+ if (!smi->proc_dir)
+ rv = -ENOMEM;
+ else {
+ smi->proc_dir->owner = THIS_MODULE;
+ }
+
+ if (rv == 0)
+ rv = ipmi_smi_add_proc_entry(smi, "stats",
+ stat_file_read_proc, NULL,
+ smi, THIS_MODULE);
+
+ if (rv == 0)
+ rv = ipmi_smi_add_proc_entry(smi, "ipmb",
+ ipmb_file_read_proc, NULL,
+ smi, THIS_MODULE);
+
+ if (rv == 0)
+ rv = ipmi_smi_add_proc_entry(smi, "version",
+ version_file_read_proc, NULL,
+ smi, THIS_MODULE);
+
+ return rv;
+}
+
+static int
+send_channel_info_cmd(ipmi_smi_t intf, int chan)
+{
+ struct ipmi_msg msg;
+ unsigned char data[1];
+ struct ipmi_system_interface_addr si;
+
+ si.addr_type = IPMI_SYSTEM_INTERFACE_ADDR_TYPE;
+ si.channel = IPMI_BMC_CHANNEL;
+ si.lun = 0;
+
+ msg.netfn = IPMI_NETFN_APP_REQUEST;
+ msg.cmd = IPMI_GET_CHANNEL_INFO_CMD;
+ msg.data = data;
+ msg.data_len = 1;
+ data[0] = chan;
+ return i_ipmi_request(NULL,
+ intf,
+ (struct ipmi_addr *) &si,
+ 0,
+ &msg,
+ NULL,
+ NULL,
+ NULL,
+ 0,
+ intf->my_address,
+ intf->my_lun,
+ -1, 0);
+}
+
+static void
+channel_handler(ipmi_smi_t intf, struct ipmi_smi_msg *msg)
+{
+ int rv = 0;
+ int chan;
+
+ if ((msg->rsp[0] == (IPMI_NETFN_APP_RESPONSE << 2))
+ && (msg->rsp[1] == IPMI_GET_CHANNEL_INFO_CMD))
+ {
+ /* It's the one we want */
+ if (msg->rsp[2] != 0) {
+ /* Got an error from the channel, just go on. */
+ goto next_channel;
+ }
+ if (msg->rsp_size < 6) {
+ /* Message not big enough, just go on. */
+ goto next_channel;
+ }
+ chan = intf->curr_channel;
+ intf->channels[chan].medium = msg->rsp[4] & 0x7f;
+ intf->channels[chan].protocol = msg->rsp[5] & 0x1f;
+
+ next_channel:
+ intf->curr_channel++;
+ if (intf->curr_channel >= IPMI_MAX_CHANNELS)
+ wake_up(&intf->waitq);
+ else
+ rv = send_channel_info_cmd(intf, intf->curr_channel);
+
+ if (rv) {
+ /* Got an error somehow, just give up. */
+ intf->curr_channel = IPMI_MAX_CHANNELS;
+ wake_up(&intf->waitq);
+
+ printk(KERN_WARNING "ipmi_msghandler: Error sending"
+ "channel information: 0x%x\n",
+ rv);
+ }
+ }
}
int ipmi_register_smi(struct ipmi_smi_handlers *handlers,
@@ -1036,7 +1642,6 @@ int ipmi_register_smi(struct ipmi_smi_ha
int i, j;
int rv;
ipmi_smi_t new_intf;
- struct list_head *entry;
unsigned long flags;
@@ -1055,12 +1660,16 @@ int ipmi_register_smi(struct ipmi_smi_ha
new_intf = kmalloc(sizeof(*new_intf), GFP_KERNEL);
if (!new_intf)
return -ENOMEM;
+ memset(new_intf, 0, sizeof(*new_intf));
+
+ new_intf->proc_dir = NULL;
rv = -ENOMEM;
down_write(&interfaces_sem);
for (i=0; i<MAX_IPMI_INTERFACES; i++) {
if (ipmi_interfaces[i] == NULL) {
+ new_intf->intf_num = i;
new_intf->version_major = version_major;
new_intf->version_minor = version_minor;
new_intf->my_address = IPMI_BMC_SLAVE_ADDR;
@@ -1081,9 +1690,12 @@ int ipmi_register_smi(struct ipmi_smi_ha
INIT_LIST_HEAD(&(new_intf->waiting_events));
new_intf->waiting_events_count = 0;
rwlock_init(&(new_intf->cmd_rcvr_lock));
+ init_waitqueue_head(&new_intf->waitq);
INIT_LIST_HEAD(&(new_intf->cmd_rcvrs));
new_intf->all_cmd_rcvr = NULL;
+ spin_lock_init(&(new_intf->counter_lock));
+
spin_lock_irqsave(&interfaces_lock, flags);
ipmi_interfaces[i] = new_intf;
spin_unlock_irqrestore(&interfaces_lock, flags);
@@ -1096,46 +1708,71 @@ int ipmi_register_smi(struct ipmi_smi_ha
downgrade_write(&interfaces_sem);
+ if (rv == 0)
+ rv = add_proc_entries(*intf, i);
+
if (rv == 0) {
- /* Call all the watcher interfaces to tell them that a
- new interface is available. */
- down_read(&smi_watchers_sem);
- list_for_each(entry, &smi_watchers) {
- struct ipmi_smi_watcher *w;
- w = list_entry(entry, struct ipmi_smi_watcher, link);
- w->new_smi(i);
- }
- up_read(&smi_watchers_sem);
+ if ((version_major > 1)
+ || ((version_major == 1) && (version_minor >= 5)))
+ {
+ /* Start scanning the channels to see what is
+ available. */
+ (*intf)->null_user_handler = channel_handler;
+ (*intf)->curr_channel = 0;
+ rv = send_channel_info_cmd(*intf, 0);
+ if (rv)
+ goto out;
+
+ /* Wait for the channel info to be read. */
+ up_read(&interfaces_sem);
+ wait_event((*intf)->waitq,
+ ((*intf)->curr_channel>=IPMI_MAX_CHANNELS));
+ down_read(&interfaces_sem);
+
+ if (ipmi_interfaces[i] != new_intf)
+ /* Well, it went away. Just return. */
+ goto out;
+ } else {
+ /* Assume a single IPMB channel at zero. */
+ (*intf)->channels[0].medium = IPMI_CHANNEL_MEDIUM_IPMB;
+ (*intf)->channels[0].protocol
+ = IPMI_CHANNEL_PROTOCOL_IPMB;
+ }
+
+ /* Call all the watcher interfaces to tell
+ them that a new interface is available. */
+ call_smi_watchers(i);
}
+ out:
up_read(&interfaces_sem);
- if (rv)
+ if (rv) {
+ if (new_intf->proc_dir)
+ remove_proc_entry(new_intf->proc_dir_name,
+ proc_ipmi_root);
kfree(new_intf);
+ }
return rv;
}
static void free_recv_msg_list(struct list_head *q)
{
- struct list_head *entry, *entry2;
- struct ipmi_recv_msg *msg;
+ struct ipmi_recv_msg *msg, *msg2;
- list_for_each_safe(entry, entry2, q) {
- msg = list_entry(entry, struct ipmi_recv_msg, link);
- list_del(entry);
+ list_for_each_entry_safe(msg, msg2, q, link) {
+ list_del(&msg->link);
ipmi_free_recv_msg(msg);
}
}
static void free_cmd_rcvr_list(struct list_head *q)
{
- struct list_head *entry, *entry2;
- struct cmd_rcvr *rcvr;
+ struct cmd_rcvr *rcvr, *rcvr2;
- list_for_each_safe(entry, entry2, q) {
- rcvr = list_entry(entry, struct cmd_rcvr, link);
- list_del(entry);
+ list_for_each_entry_safe(rcvr, rcvr2, q, link) {
+ list_del(&rcvr->link);
kfree(rcvr);
}
}
@@ -1159,16 +1796,18 @@ static void clean_up_interface_data(ipmi
int ipmi_unregister_smi(ipmi_smi_t intf)
{
- int rv = -ENODEV;
- int i;
- struct list_head *entry;
- unsigned long flags;
+ int rv = -ENODEV;
+ int i;
+ struct ipmi_smi_watcher *w;
+ unsigned long flags;
down_write(&interfaces_sem);
if (list_empty(&(intf->users)))
{
for (i=0; i<MAX_IPMI_INTERFACES; i++) {
if (ipmi_interfaces[i] == intf) {
+ remove_proc_entry(intf->proc_dir_name,
+ proc_ipmi_root);
spin_lock_irqsave(&interfaces_lock, flags);
ipmi_interfaces[i] = NULL;
clean_up_interface_data(intf);
@@ -1191,11 +1830,7 @@ int ipmi_unregister_smi(ipmi_smi_t intf)
/* Call all the watcher interfaces to tell them that
an interface is gone. */
down_read(&smi_watchers_sem);
- list_for_each(entry, &smi_watchers) {
- struct ipmi_smi_watcher *w;
- w = list_entry(entry,
- struct ipmi_smi_watcher,
- link);
+ list_for_each_entry(w, &smi_watchers, link) {
w->smi_gone(i);
}
up_read(&smi_watchers_sem);
@@ -1203,20 +1838,28 @@ int ipmi_unregister_smi(ipmi_smi_t intf)
return 0;
}
-static int handle_get_msg_rsp(ipmi_smi_t intf,
- struct ipmi_smi_msg *msg)
+static int handle_ipmb_get_msg_rsp(ipmi_smi_t intf,
+ struct ipmi_smi_msg *msg)
{
struct ipmi_ipmb_addr ipmb_addr;
struct ipmi_recv_msg *recv_msg;
+ unsigned long flags;
- if (msg->rsp_size < 11)
+ /* This is 11, not 10, because the response must contain a
+ * completion code. */
+ if (msg->rsp_size < 11) {
/* Message not big enough, just ignore it. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->invalid_ipmb_responses++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
return 0;
+ }
- if (msg->rsp[2] != 0)
+ if (msg->rsp[2] != 0) {
/* An error getting the response, just ignore it. */
return 0;
+ }
ipmb_addr.addr_type = IPMI_IPMB_ADDR_TYPE;
ipmb_addr.slave_addr = msg->rsp[6];
@@ -1235,6 +1878,9 @@ static int handle_get_msg_rsp(ipmi_smi_t
{
/* We were unable to find the sequence number,
so just nuke the message. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->unhandled_ipmb_responses++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
return 0;
}
@@ -1248,26 +1894,33 @@ static int handle_get_msg_rsp(ipmi_smi_t
recv_msg->msg.data = recv_msg->msg_data;
recv_msg->msg.data_len = msg->rsp_size - 10;
recv_msg->recv_type = IPMI_RESPONSE_RECV_TYPE;
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->handled_ipmb_responses++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
deliver_response(recv_msg);
return 0;
}
-static int handle_get_msg_cmd(ipmi_smi_t intf,
- struct ipmi_smi_msg *msg)
+static int handle_ipmb_get_msg_cmd(ipmi_smi_t intf,
+ struct ipmi_smi_msg *msg)
{
- struct list_head *entry;
struct cmd_rcvr *rcvr;
- int rv = 0;
- unsigned char netfn;
- unsigned char cmd;
- ipmi_user_t user = NULL;
+ int rv = 0;
+ unsigned char netfn;
+ unsigned char cmd;
+ ipmi_user_t user = NULL;
struct ipmi_ipmb_addr *ipmb_addr;
struct ipmi_recv_msg *recv_msg;
+ unsigned long flags;
- if (msg->rsp_size < 10)
+ if (msg->rsp_size < 10) {
/* Message not big enough, just ignore it. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->invalid_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
return 0;
+ }
if (msg->rsp[2] != 0) {
/* An error getting the response, just ignore it. */
@@ -1283,8 +1936,7 @@ static int handle_get_msg_cmd(ipmi_smi_t
user = intf->all_cmd_rcvr;
} else {
/* Find the command/netfn. */
- list_for_each(entry, &(intf->cmd_rcvrs)) {
- rcvr = list_entry(entry, struct cmd_rcvr, link);
+ list_for_each_entry(rcvr, &(intf->cmd_rcvrs), link) {
if ((rcvr->netfn == netfn) && (rcvr->cmd == cmd)) {
user = rcvr->user;
break;
@@ -1295,6 +1947,10 @@ static int handle_get_msg_cmd(ipmi_smi_t
if (user == NULL) {
/* We didn't find a user, deliver an error response. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->unhandled_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+
msg->data[0] = (IPMI_NETFN_APP_REQUEST << 2);
msg->data[1] = IPMI_SEND_MSG_CMD;
msg->data[2] = msg->rsp[3];
@@ -1309,12 +1965,25 @@ static int handle_get_msg_cmd(ipmi_smi_t
msg->data[10] = ipmb_checksum(&(msg->data[6]), 4);
msg->data_size = 11;
+#if DEBUG_MSGING
+ {
+ int m;
+ printk("Invalid command:");
+ for (m=0; m<msg->data_size; m++)
+ printk(" %2.2x", msg->data[m]);
+ printk("\n");
+ }
+#endif
intf->handlers->sender(intf->send_info, msg, 0);
rv = -1; /* We used the message, so return the value that
causes it to not be freed or queued. */
} else {
/* Deliver the message to the user. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->handled_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+
recv_msg = ipmi_alloc_recv_msg();
if (! recv_msg) {
/* We couldn't allocate memory for the
@@ -1322,18 +1991,24 @@ static int handle_get_msg_cmd(ipmi_smi_t
later. */
rv = 1;
} else {
+ /* Extract the source address from the data. */
ipmb_addr = (struct ipmi_ipmb_addr *) &recv_msg->addr;
ipmb_addr->addr_type = IPMI_IPMB_ADDR_TYPE;
ipmb_addr->slave_addr = msg->rsp[6];
ipmb_addr->lun = msg->rsp[7] & 3;
- ipmb_addr->channel = msg->rsp[3];
+ ipmb_addr->channel = msg->rsp[3] & 0xf;
+ /* Extract the rest of the message information
+ from the IPMB header.*/
recv_msg->user = user;
recv_msg->recv_type = IPMI_CMD_RECV_TYPE;
recv_msg->msgid = msg->rsp[7] >> 2;
recv_msg->msg.netfn = msg->rsp[4] >> 2;
recv_msg->msg.cmd = msg->rsp[8];
recv_msg->msg.data = recv_msg->msg_data;
+
+ /* We chop off 10, not 9 bytes because the checksum
+ at the end also needs to be removed. */
recv_msg->msg.data_len = msg->rsp_size - 10;
memcpy(recv_msg->msg_data,
&(msg->rsp[9]),
@@ -1345,6 +2020,169 @@ static int handle_get_msg_cmd(ipmi_smi_t
return rv;
}
+static int handle_lan_get_msg_rsp(ipmi_smi_t intf,
+ struct ipmi_smi_msg *msg)
+{
+ struct ipmi_lan_addr lan_addr;
+ struct ipmi_recv_msg *recv_msg;
+ unsigned long flags;
+
+
+ /* This is 13, not 12, because the response must contain a
+ * completion code. */
+ if (msg->rsp_size < 13) {
+ /* Message not big enough, just ignore it. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->invalid_lan_responses++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+ return 0;
+ }
+
+ if (msg->rsp[2] != 0) {
+ /* An error getting the response, just ignore it. */
+ return 0;
+ }
+
+ lan_addr.addr_type = IPMI_LAN_ADDR_TYPE;
+ lan_addr.session_handle = msg->rsp[4];
+ lan_addr.remote_SWID = msg->rsp[8];
+ lan_addr.local_SWID = msg->rsp[5];
+ lan_addr.channel = msg->rsp[3] & 0x0f;
+ lan_addr.privilege = msg->rsp[3] >> 4;
+ lan_addr.lun = msg->rsp[9] & 3;
+
+ /* It's a response from a remote entity. Look up the sequence
+ number and handle the response. */
+ if (intf_find_seq(intf,
+ msg->rsp[9] >> 2,
+ msg->rsp[3] & 0x0f,
+ msg->rsp[10],
+ (msg->rsp[6] >> 2) & (~1),
+ (struct ipmi_addr *) &(lan_addr),
+ &recv_msg))
+ {
+ /* We were unable to find the sequence number,
+ so just nuke the message. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->unhandled_lan_responses++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+ return 0;
+ }
+
+ memcpy(recv_msg->msg_data,
+ &(msg->rsp[11]),
+ msg->rsp_size - 11);
+ /* The other fields matched, so no need to set them, except
+ for netfn, which needs to be the response that was
+ returned, not the request value. */
+ recv_msg->msg.netfn = msg->rsp[6] >> 2;
+ recv_msg->msg.data = recv_msg->msg_data;
+ recv_msg->msg.data_len = msg->rsp_size - 12;
+ recv_msg->recv_type = IPMI_RESPONSE_RECV_TYPE;
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->handled_lan_responses++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+ deliver_response(recv_msg);
+
+ return 0;
+}
+
+static int handle_lan_get_msg_cmd(ipmi_smi_t intf,
+ struct ipmi_smi_msg *msg)
+{
+ struct cmd_rcvr *rcvr;
+ int rv = 0;
+ unsigned char netfn;
+ unsigned char cmd;
+ ipmi_user_t user = NULL;
+ struct ipmi_lan_addr *lan_addr;
+ struct ipmi_recv_msg *recv_msg;
+ unsigned long flags;
+
+ if (msg->rsp_size < 12) {
+ /* Message not big enough, just ignore it. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->invalid_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+ return 0;
+ }
+
+ if (msg->rsp[2] != 0) {
+ /* An error getting the response, just ignore it. */
+ return 0;
+ }
+
+ netfn = msg->rsp[6] >> 2;
+ cmd = msg->rsp[10];
+
+ read_lock(&(intf->cmd_rcvr_lock));
+
+ if (intf->all_cmd_rcvr) {
+ user = intf->all_cmd_rcvr;
+ } else {
+ /* Find the command/netfn. */
+ list_for_each_entry(rcvr, &(intf->cmd_rcvrs), link) {
+ if ((rcvr->netfn == netfn) && (rcvr->cmd == cmd)) {
+ user = rcvr->user;
+ break;
+ }
+ }
+ }
+ read_unlock(&(intf->cmd_rcvr_lock));
+
+ if (user == NULL) {
+ /* We didn't find a user, deliver an error response. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->unhandled_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+
+ rv = 0; /* Don't do anything with these messages, just
+ allow them to be freed. */
+ } else {
+ /* Deliver the message to the user. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->handled_commands++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+
+ recv_msg = ipmi_alloc_recv_msg();
+ if (! recv_msg) {
+ /* We couldn't allocate memory for the
+ message, so requeue it for handling
+ later. */
+ rv = 1;
+ } else {
+ /* Extract the source address from the data. */
+ lan_addr = (struct ipmi_lan_addr *) &recv_msg->addr;
+ lan_addr->addr_type = IPMI_LAN_ADDR_TYPE;
+ lan_addr->session_handle = msg->rsp[4];
+ lan_addr->remote_SWID = msg->rsp[8];
+ lan_addr->local_SWID = msg->rsp[5];
+ lan_addr->lun = msg->rsp[9] & 3;
+ lan_addr->channel = msg->rsp[3] & 0xf;
+ lan_addr->privilege = msg->rsp[3] >> 4;
+
+ /* Extract the rest of the message information
+ from the IPMB header.*/
+ recv_msg->user = user;
+ recv_msg->recv_type = IPMI_CMD_RECV_TYPE;
+ recv_msg->msgid = msg->rsp[9] >> 2;
+ recv_msg->msg.netfn = msg->rsp[6] >> 2;
+ recv_msg->msg.cmd = msg->rsp[10];
+ recv_msg->msg.data = recv_msg->msg_data;
+
+ /* We chop off 12, not 11 bytes because the checksum
+ at the end also needs to be removed. */
+ recv_msg->msg.data_len = msg->rsp_size - 12;
+ memcpy(recv_msg->msg_data,
+ &(msg->rsp[11]),
+ msg->rsp_size - 12);
+ deliver_response(recv_msg);
+ }
+ }
+
+ return rv;
+}
+
static void copy_event_into_recv_msg(struct ipmi_recv_msg *recv_msg,
struct ipmi_smi_msg *msg)
{
@@ -1368,9 +2206,8 @@ static void copy_event_into_recv_msg(str
static int handle_read_event_rsp(ipmi_smi_t intf,
struct ipmi_smi_msg *msg)
{
- struct ipmi_recv_msg *recv_msg;
+ struct ipmi_recv_msg *recv_msg, *recv_msg2;
struct list_head msgs;
- struct list_head *entry, *entry2;
ipmi_user_t user;
int rv = 0;
int deliver_count = 0;
@@ -1378,6 +2215,9 @@ static int handle_read_event_rsp(ipmi_sm
if (msg->rsp_size < 19) {
/* Message is too small to be an IPMB event. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->invalid_events++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
return 0;
}
@@ -1390,21 +2230,20 @@ static int handle_read_event_rsp(ipmi_sm
spin_lock_irqsave(&(intf->events_lock), flags);
+ spin_lock(&intf->counter_lock);
+ intf->events++;
+ spin_unlock(&intf->counter_lock);
+
/* Allocate and fill in one message for every user that is getting
events. */
- list_for_each(entry, &(intf->users)) {
- user = list_entry(entry, struct ipmi_user, link);
-
+ list_for_each_entry(user, &(intf->users), link) {
if (! user->gets_events)
continue;
recv_msg = ipmi_alloc_recv_msg();
if (! recv_msg) {
- list_for_each_safe(entry, entry2, &msgs) {
- recv_msg = list_entry(entry,
- struct ipmi_recv_msg,
- link);
- list_del(entry);
+ list_for_each_entry_safe(recv_msg, recv_msg2, &msgs, link) {
+ list_del(&recv_msg->link);
ipmi_free_recv_msg(recv_msg);
}
/* We couldn't allocate memory for the
@@ -1423,11 +2262,8 @@ static int handle_read_event_rsp(ipmi_sm
if (deliver_count) {
/* Now deliver all the messages. */
- list_for_each_safe(entry, entry2, &msgs) {
- recv_msg = list_entry(entry,
- struct ipmi_recv_msg,
- link);
- list_del(entry);
+ list_for_each_entry_safe(recv_msg, recv_msg2, &msgs, link) {
+ list_del(&recv_msg->link);
deliver_response(recv_msg);
}
} else if (intf->waiting_events_count < MAX_EVENTS_IN_QUEUE) {
@@ -1462,15 +2298,14 @@ static int handle_bmc_rsp(ipmi_smi_t
{
struct ipmi_recv_msg *recv_msg;
int found = 0;
- struct list_head *entry;
+ struct ipmi_user *user;
+ unsigned long flags;
recv_msg = (struct ipmi_recv_msg *) msg->user_data;
/* Make sure the user still exists. */
- list_for_each(entry, &(intf->users)) {
- if (list_entry(entry, struct ipmi_user, link)
- == recv_msg->user)
- {
+ list_for_each_entry(user, &(intf->users), link) {
+ if (user == recv_msg->user) {
/* Found it, so we can deliver it */
found = 1;
break;
@@ -1482,10 +2317,16 @@ static int handle_bmc_rsp(ipmi_smi_t
if (!recv_msg->user && intf->null_user_handler)
intf->null_user_handler(intf, msg);
/* The user for the message went away, so give up. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->unhandled_local_responses++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
ipmi_free_recv_msg(recv_msg);
} else {
struct ipmi_system_interface_addr *smi_addr;
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ intf->handled_local_responses++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
recv_msg->recv_type = IPMI_RESPONSE_RECV_TYPE;
recv_msg->msgid = msg->msgid;
smi_addr = ((struct ipmi_system_interface_addr *)
@@ -1513,28 +2354,86 @@ static int handle_new_recv_msg(ipmi_smi_
struct ipmi_smi_msg *msg)
{
int requeue;
+ int chan;
+#if DEBUG_MSGING
+ int m;
+ printk("Recv:");
+ for (m=0; m<msg->rsp_size; m++)
+ printk(" %2.2x", msg->rsp[m]);
+ printk("\n");
+#endif
if (msg->rsp_size < 2) {
/* Message is too small to be correct. */
requeue = 0;
- } else if (msg->rsp[1] == IPMI_GET_MSG_CMD) {
-#if DEBUG_MSGING
- int m;
- printk("Response:");
- for (m=0; m<msg->rsp_size; m++)
- printk(" %2.2x", msg->rsp[m]);
- printk("\n");
-#endif
+ } else if ((msg->rsp[0] == ((IPMI_NETFN_APP_REQUEST|1) << 2))
+ && (msg->rsp[1] == IPMI_SEND_MSG_CMD)
+ && (msg->user_data != NULL))
+ {
+ /* It's a response to a response we sent. For this we
+ deliver a send message response to the user. */
+ struct ipmi_recv_msg *recv_msg = msg->user_data;
+
+ requeue = 0;
+ if (msg->rsp_size < 2)
+ /* Message is too small to be correct. */
+ goto out;
+
+ chan = msg->data[2] & 0x0f;
+ if (chan >= IPMI_MAX_CHANNELS)
+ /* Invalid channel number */
+ goto out;
+
+ if (recv_msg) {
+ recv_msg->recv_type = IPMI_RESPONSE_RESPONSE_TYPE;
+ recv_msg->msg.data = recv_msg->msg_data;
+ recv_msg->msg.data_len = 1;
+ recv_msg->msg_data[0] = msg->rsp[2];
+ deliver_response(recv_msg);
+ }
+ } else if ((msg->rsp[0] == ((IPMI_NETFN_APP_REQUEST|1) << 2))
+ && (msg->rsp[1] == IPMI_GET_MSG_CMD))
+ {
/* It's from the receive queue. */
- if (msg->rsp[4] & 0x04) {
- /* It's a response, so find the
- requesting message and send it up. */
- requeue = handle_get_msg_rsp(intf, msg);
- } else {
- /* It's a command to the SMS from some other
- entity. Handle that. */
- requeue = handle_get_msg_cmd(intf, msg);
+ chan = msg->rsp[3] & 0xf;
+ if (chan >= IPMI_MAX_CHANNELS) {
+ /* Invalid channel number */
+ requeue = 0;
+ goto out;
}
+
+ switch (intf->channels[chan].medium) {
+ case IPMI_CHANNEL_MEDIUM_IPMB:
+ if (msg->rsp[4] & 0x04) {
+ /* It's a response, so find the
+ requesting message and send it up. */
+ requeue = handle_ipmb_get_msg_rsp(intf, msg);
+ } else {
+ /* It's a command to the SMS from some other
+ entity. Handle that. */
+ requeue = handle_ipmb_get_msg_cmd(intf, msg);
+ }
+ break;
+
+ case IPMI_CHANNEL_MEDIUM_8023LAN:
+ case IPMI_CHANNEL_MEDIUM_ASYNC:
+ if (msg->rsp[6] & 0x04) {
+ /* It's a response, so find the
+ requesting message and send it up. */
+ requeue = handle_lan_get_msg_rsp(intf, msg);
+ } else {
+ /* It's a command to the SMS from some other
+ entity. Handle that. */
+ requeue = handle_lan_get_msg_cmd(intf, msg);
+ }
+ break;
+
+ default:
+ /* We don't handle the channel type, so just
+ * free the message. */
+ requeue = 0;
+ }
+
} else if (msg->rsp[1] == IPMI_READ_EVENT_MSG_BUFFER_CMD) {
/* It's an asyncronous event. */
requeue = handle_read_event_rsp(intf, msg);
@@ -1543,6 +2442,7 @@ static int handle_new_recv_msg(ipmi_smi_
requeue = handle_bmc_rsp(intf, msg);
}
+ out:
return requeue;
}
@@ -1558,10 +2458,43 @@ void ipmi_smi_msg_received(ipmi_smi_t
working on it. */
read_lock(&(intf->users_lock));
- if ((msg->data_size >= 2) && (msg->data[1] == IPMI_SEND_MSG_CMD)) {
- /* This is the local response to a send, start the
- timer for these. */
- intf_start_seq_timer(intf, msg->msgid);
+ if ((msg->data_size >= 2)
+ && (msg->data[0] == (IPMI_NETFN_APP_REQUEST << 2))
+ && (msg->data[1] == IPMI_SEND_MSG_CMD)
+ && (msg->user_data == NULL)) {
+ /* This is the local response to a command send, start
+ the timer for these. The user_data will not be
+ NULL if this is a response send, and we will let
+ response sends just go through. */
+
+ /* Check for errors, if we get certain errors (ones
+ that mean basically we can try again later), we
+ ignore them and start the timer. Otherwise we
+ report the error immediately. */
+ if ((msg->rsp_size >= 3) && (msg->rsp[2] != 0)
+ && (msg->rsp[2] != IPMI_NODE_BUSY_ERR)
+ && (msg->rsp[2] != IPMI_LOST_ARBITRATION_ERR))
+ {
+ int chan = msg->rsp[3] & 0xf;
+
+ /* Got an error sending the message, handle it. */
+ spin_lock_irqsave(&intf->counter_lock, flags);
+ if (chan >= IPMI_MAX_CHANNELS)
+ ; /* This shouldn't happen */
+ else if ((intf->channels[chan].medium
+ == IPMI_CHANNEL_MEDIUM_8023LAN)
+ || (intf->channels[chan].medium
+ == IPMI_CHANNEL_MEDIUM_ASYNC))
+ intf->sent_lan_command_errs++;
+ else
+ intf->sent_ipmb_command_errs++;
+ spin_unlock_irqrestore(&intf->counter_lock, flags);
+ intf_err_seq(intf, msg->msgid, msg->rsp[2]);
+ } else {
+ /* The message was sent, start the timer. */
+ intf_start_seq_timer(intf, msg->msgid);
+ }
+
ipmi_free_smi_msg(msg);
goto out_unlock;
}
@@ -1593,13 +2526,10 @@ void ipmi_smi_msg_received(ipmi_smi_t
void ipmi_smi_watchdog_pretimeout(ipmi_smi_t intf)
{
- struct list_head *entry;
- ipmi_user_t user;
+ ipmi_user_t user;
read_lock(&(intf->users_lock));
- list_for_each(entry, &(intf->users)) {
- user = list_entry(entry, struct ipmi_user, link);
-
+ list_for_each_entry(user, &(intf->users), link) {
if (! user->handler->ipmi_watchdog_pretimeout)
continue;
@@ -1657,10 +2587,9 @@ ipmi_timeout_handler(long timeout_period
{
ipmi_smi_t intf;
struct list_head timeouts;
- struct ipmi_recv_msg *msg;
- struct ipmi_smi_msg *smi_msg;
+ struct ipmi_recv_msg *msg, *msg2;
+ struct ipmi_smi_msg *smi_msg, *smi_msg2;
unsigned long flags;
- struct list_head *entry, *entry2;
int i, j;
INIT_LIST_HEAD(&timeouts);
@@ -1675,10 +2604,9 @@ ipmi_timeout_handler(long timeout_period
/* See if any waiting messages need to be processed. */
spin_lock_irqsave(&(intf->waiting_msgs_lock), flags);
- list_for_each_safe(entry, entry2, &(intf->waiting_msgs)) {
- smi_msg = list_entry(entry, struct ipmi_smi_msg, link);
+ list_for_each_entry_safe(smi_msg, smi_msg2, &(intf->waiting_msgs), link) {
if (! handle_new_recv_msg(intf, smi_msg)) {
- list_del(entry);
+ list_del(&smi_msg->link);
ipmi_free_smi_msg(smi_msg);
} else {
/* To preserve message order, quit if we
@@ -1706,6 +2634,15 @@ ipmi_timeout_handler(long timeout_period
ent->inuse = 0;
msg = ent->recv_msg;
list_add_tail(&(msg->link), &timeouts);
+ spin_lock(&intf->counter_lock);
+ if (ent->broadcast)
+ intf->timed_out_ipmb_broadcasts++;
+ else if (ent->recv_msg->addr.addr_type
+ == IPMI_LAN_ADDR_TYPE)
+ intf->timed_out_lan_commands++;
+ else
+ intf->timed_out_ipmb_commands++;
+ spin_unlock(&intf->counter_lock);
} else {
/* More retries, send again. */
@@ -1715,12 +2652,18 @@ ipmi_timeout_handler(long timeout_period
ent->retries_left--;
send_from_recv_msg(intf, ent->recv_msg, NULL,
j, ent->seqid);
+ spin_lock(&intf->counter_lock);
+ if (ent->recv_msg->addr.addr_type
+ == IPMI_LAN_ADDR_TYPE)
+ intf->retransmitted_lan_commands++;
+ else
+ intf->retransmitted_ipmb_commands++;
+ spin_unlock(&intf->counter_lock);
}
}
spin_unlock_irqrestore(&(intf->seq_lock), flags);
- list_for_each_safe(entry, entry2, &timeouts) {
- msg = list_entry(entry, struct ipmi_recv_msg, link);
+ list_for_each_entry_safe(msg, msg2, &timeouts, link) {
handle_msg_timeout(msg);
}
@@ -1747,13 +2690,16 @@ static void ipmi_request_event(void)
static struct timer_list ipmi_timer;
-/* Call every 100 ms. */
+/* Call every ~100 ms. */
#define IPMI_TIMEOUT_TIME 100
-#define IPMI_TIMEOUT_JIFFIES ((IPMI_TIMEOUT_TIME * HZ)/1000)
-/* Request events from the queue every second. Hopefully, in the
- future, IPMI will add a way to know immediately if an event is
- in the queue. */
+/* How many jiffies does it take to get to the timeout time. */
+#define IPMI_TIMEOUT_JIFFIES ((IPMI_TIMEOUT_TIME * HZ) / 1000)
+
+/* Request events from the queue every second (this is the number of
+ IPMI_TIMEOUT_TIMES between event requests). Hopefully, in the
+ future, IPMI will add a way to know immediately if an event is in
+ the queue and this silliness can go away. */
#define IPMI_REQUEST_EV_TIME (1000 / (IPMI_TIMEOUT_TIME))
static volatile int stop_operation = 0;
@@ -1796,6 +2742,7 @@ struct ipmi_smi_msg *ipmi_alloc_smi_msg(
rv = kmalloc(sizeof(struct ipmi_smi_msg), GFP_ATOMIC);
if (rv) {
rv->done = free_smi_msg;
+ rv->user_data = NULL;
atomic_inc(&smi_msg_inuse_count);
}
return rv;
@@ -1907,11 +2854,13 @@ static void send_panic_events(char *str)
&addr,
0,
&msg,
+ NULL,
&smi_msg,
&recv_msg,
0,
intf->my_address,
- intf->my_lun);
+ intf->my_lun,
+ 0, 1); /* Don't retry, and don't wait. */
}
#ifdef CONFIG_IPMI_PANIC_STRING
@@ -1951,11 +2900,13 @@ static void send_panic_events(char *str)
&addr,
0,
&msg,
+ NULL,
&smi_msg,
&recv_msg,
0,
intf->my_address,
- intf->my_lun);
+ intf->my_lun,
+ 0, 1); /* Don't retry, and don't wait. */
if (intf->local_event_generator) {
/* Request the event receiver from the local MC. */
@@ -1969,11 +2920,13 @@ static void send_panic_events(char *str)
&addr,
0,
&msg,
+ NULL,
&smi_msg,
&recv_msg,
0,
intf->my_address,
- intf->my_lun);
+ intf->my_lun,
+ 0, 1); /* no retry, and no wait. */
}
intf->null_user_handler = NULL;
@@ -2029,11 +2982,13 @@ static void send_panic_events(char *str)
&addr,
0,
&msg,
+ NULL,
&smi_msg,
&recv_msg,
0,
intf->my_address,
- intf->my_lun);
+ intf->my_lun,
+ 0, 1); /* no retry, and no wait. */
}
}
#endif /* CONFIG_IPMI_PANIC_STRING */
@@ -2075,7 +3030,6 @@ static struct notifier_block panic_block
200 /* priority: INT_MAX >= x >= 0 */
};
-
static __init int ipmi_init_msghandler(void)
{
int i;
@@ -2083,10 +3037,21 @@ static __init int ipmi_init_msghandler(v
if (initialized)
return 0;
+ printk(KERN_INFO "ipmi message handler version "
+ IPMI_MSGHANDLER_VERSION "\n");
+
for (i=0; i<MAX_IPMI_INTERFACES; i++) {
ipmi_interfaces[i] = NULL;
}
+ proc_ipmi_root = proc_mkdir("ipmi", 0);
+ if (!proc_ipmi_root) {
+ printk("Unable to create IPMI proc dir");
+ return -ENOMEM;
+ }
+
+ proc_ipmi_root->owner = THIS_MODULE;
+
init_timer(&ipmi_timer);
ipmi_timer.data = 0;
ipmi_timer.function = ipmi_timeout;
@@ -2097,8 +3062,6 @@ static __init int ipmi_init_msghandler(v
initialized = 1;
- printk(KERN_INFO "ipmi: message handler initialized\n");
-
return 0;
}
@@ -2118,9 +3081,12 @@ static __exit void cleanup_ipmi(void)
problems with race conditions removing the timer here. */
stop_operation = 1;
while (!timer_stopped) {
+ set_current_state(TASK_UNINTERRUPTIBLE);
schedule_timeout(1);
}
+ remove_proc_entry(proc_ipmi_root->name, &proc_root);
+
initialized = 0;
/* Check for buffer leaks. */
@@ -2143,6 +3109,7 @@ EXPORT_SYMBOL(ipmi_create_user);
EXPORT_SYMBOL(ipmi_destroy_user);
EXPORT_SYMBOL(ipmi_get_version);
EXPORT_SYMBOL(ipmi_request);
+EXPORT_SYMBOL(ipmi_request_settime);
EXPORT_SYMBOL(ipmi_request_supply_msgs);
EXPORT_SYMBOL(ipmi_request_with_source);
EXPORT_SYMBOL(ipmi_register_smi);
@@ -2164,3 +3131,4 @@ EXPORT_SYMBOL(ipmi_set_my_address);
EXPORT_SYMBOL(ipmi_get_my_address);
EXPORT_SYMBOL(ipmi_set_my_LUN);
EXPORT_SYMBOL(ipmi_get_my_LUN);
+EXPORT_SYMBOL(ipmi_smi_add_proc_entry);
--- diff/drivers/char/ipmi/ipmi_watchdog.c 2003-09-17 11:28:05.000000000 +0000
+++ source/drivers/char/ipmi/ipmi_watchdog.c 2004-03-16 11:27:48.102874912 +0000
@@ -33,6 +33,7 @@
#include <linux/config.h>
#include <linux/module.h>
+#include <linux/moduleparam.h>
#include <linux/ipmi.h>
#include <linux/ipmi_smi.h>
#include <linux/watchdog.h>
@@ -50,6 +51,8 @@
#include <asm/apic.h>
#endif
+#define IPMI_WATCHDOG_VERSION "v31"
+
/*
* The IPMI command/response information for the watchdog timer.
*/
@@ -137,26 +140,41 @@ static int pretimeout = 0;
/* Default action is to reset the board on a timeout. */
static unsigned char action_val = WDOG_TIMEOUT_RESET;
-static char *action = "reset";
+static char action[16] = "reset";
static unsigned char preaction_val = WDOG_PRETIMEOUT_NONE;
-static char *preaction = "pre_none";
+static char preaction[16] = "pre_none";
static unsigned char preop_val = WDOG_PREOP_NONE;
-static char *preop = "preop_none";
+static char preop[16] = "preop_none";
static spinlock_t ipmi_read_lock = SPIN_LOCK_UNLOCKED;
static char data_to_read = 0;
static DECLARE_WAIT_QUEUE_HEAD(read_q);
static struct fasync_struct *fasync_q = NULL;
static char pretimeout_since_last_heartbeat = 0;
-MODULE_PARM(timeout, "i");
-MODULE_PARM(pretimeout, "i");
-MODULE_PARM(action, "s");
-MODULE_PARM(preaction, "s");
-MODULE_PARM(preop, "s");
+/* If true, the driver will start running as soon as it is configured
+ and ready. */
+static int start_now = 0;
+
+module_param(timeout, int, 0);
+MODULE_PARM_DESC(timeout, "Timeout value in seconds.");
+module_param(pretimeout, int, 0);
+MODULE_PARM_DESC(pretimeout, "Pretimeout value in seconds.");
+module_param_string(action, action, sizeof(action), 0);
+MODULE_PARM_DESC(action, "Timeout action. One of: "
+ "reset, none, power_cycle, power_off.");
+module_param_string(preaction, preaction, sizeof(preaction), 0);
+MODULE_PARM_DESC(preaction, "Pretimeout action. One of: "
+ "pre_none, pre_smi, pre_nmi, pre_int.");
+module_param_string(preop, preop, sizeof(preop), 0);
+MODULE_PARM_DESC(preop, "Pretimeout driver operation. One of: "
+ "preop_none, preop_panic, preop_give_data.");
+module_param(start_now, int, 0);
+MODULE_PARM_DESC(start_now, "Set to 1 to start the watchdog as"
+ "soon as the driver is loaded.");
/* Default state of the timer. */
static unsigned char ipmi_watchdog_state = WDOG_TIMEOUT_NONE;
@@ -167,10 +185,6 @@ static int ipmi_ignore_heartbeat = 0;
/* Is someone using the watchdog? Only one user is allowed. */
static int ipmi_wdog_open = 0;
-/* If true, the driver will start running as soon as it is configured
- and ready. */
-static int start_now = 0;
-
/* If set to 1, the heartbeat command will set the state to reset and
start the timer. The timer doesn't normally run when the driver is
first opened until the heartbeat is set the first time, this
@@ -260,6 +274,7 @@ static int i_ipmi_set_timeout(struct ipm
(struct ipmi_addr *) &addr,
0,
&msg,
+ NULL,
smi_msg,
recv_msg,
1);
@@ -435,6 +450,7 @@ static int ipmi_heartbeat(void)
(struct ipmi_addr *) &addr,
0,
&msg,
+ NULL,
&heartbeat_smi_msg,
&heartbeat_recv_msg,
1);
@@ -483,6 +499,7 @@ static void panic_halt_ipmi_heartbeat(vo
(struct ipmi_addr *) &addr,
0,
&msg,
+ NULL,
&panic_halt_heartbeat_smi_msg,
&panic_halt_heartbeat_recv_msg,
1);
@@ -903,6 +920,7 @@ static void ipmi_smi_gone(int if_num)
static struct ipmi_smi_watcher smi_watcher =
{
+ .owner = THIS_MODULE,
.new_smi = ipmi_new_smi,
.smi_gone = ipmi_smi_gone
};
@@ -911,6 +929,9 @@ static int __init ipmi_wdog_init(void)
{
int rv;
+ printk(KERN_INFO "IPMI watchdog driver version "
+ IPMI_WATCHDOG_VERSION "\n");
+
if (strcmp(action, "reset") == 0) {
action_val = WDOG_TIMEOUT_RESET;
} else if (strcmp(action, "none") == 0) {
@@ -999,14 +1020,10 @@ static int __init ipmi_wdog_init(void)
register_reboot_notifier(&wdog_reboot_notifier);
notifier_chain_register(&panic_notifier_list, &wdog_panic_notifier);
- printk(KERN_INFO "IPMI watchdog by "
- "Corey Minyard (minyard@mvista.com)\n");
-
return 0;
}
-#ifdef MODULE
-static void ipmi_unregister_watchdog(void)
+static __exit void ipmi_unregister_watchdog(void)
{
int rv;
@@ -1034,6 +1051,7 @@ static void ipmi_unregister_watchdog(voi
pointers to our buffers, we want to make sure they are done before
we release our memory. */
while (atomic_read(&set_timeout_tofree)) {
+ set_current_state(TASK_UNINTERRUPTIBLE);
schedule_timeout(1);
}
@@ -1056,76 +1074,6 @@ static void __exit ipmi_wdog_exit(void)
ipmi_unregister_watchdog();
}
module_exit(ipmi_wdog_exit);
-#else
-static int __init ipmi_wdog_setup(char *str)
-{
- int val;
- int rv;
- char *option;
-
- rv = get_option(&str, &val);
- if (rv == 0)
- return 1;
- if (val > 0)
- timeout = val;
- if (rv == 1)
- return 1;
-
- rv = get_option(&str, &val);
- if (rv == 0)
- return 1;
- if (val >= 0)
- pretimeout = val;
- if (rv == 1)
- return 1;
-
- while ((option = strsep(&str, ",")) != NULL) {
- if (strcmp(option, "reset") == 0) {
- action = "reset";
- }
- else if (strcmp(option, "none") == 0) {
- action = "none";
- }
- else if (strcmp(option, "power_cycle") == 0) {
- action = "power_cycle";
- }
- else if (strcmp(option, "power_off") == 0) {
- action = "power_off";
- }
- else if (strcmp(option, "pre_none") == 0) {
- preaction = "pre_none";
- }
- else if (strcmp(option, "pre_smi") == 0) {
- preaction = "pre_smi";
- }
-#ifdef HAVE_NMI_HANDLER
- else if (strcmp(option, "pre_nmi") == 0) {
- preaction = "pre_nmi";
- }
-#endif
- else if (strcmp(option, "pre_int") == 0) {
- preaction = "pre_int";
- }
- else if (strcmp(option, "start_now") == 0) {
- start_now = 1;
- }
- else if (strcmp(option, "preop_none") == 0) {
- preop = "preop_none";
- }
- else if (strcmp(option, "preop_panic") == 0) {
- preop = "preop_panic";
- }
- else if (strcmp(option, "preop_give_data") == 0) {
- preop = "preop_give_data";
- } else {
- printk("Unknown IPMI watchdog option: '%s'\n", option);
- }
- }
-
- return 1;
-}
-__setup("ipmi_wdog=", ipmi_wdog_setup);
-#endif
EXPORT_SYMBOL(ipmi_delayed_shutdown);
--- diff/drivers/char/keyboard.c 2004-02-18 08:54:09.000000000 +0000
+++ source/drivers/char/keyboard.c 2004-03-16 11:27:48.106874304 +0000
@@ -1066,6 +1066,9 @@ void kbd_keycode(unsigned int keycode, i
}
if (sysrq_down && down && !rep) {
handle_sysrq(kbd_sysrq_xlate[keycode], regs, tty);
+#ifdef CONFIG_KGDB_SYSRQ
+ sysrq_down = 0; /* in case we miss the "up" event */
+#endif
return;
}
#endif
--- diff/drivers/char/mem.c 2004-03-11 10:20:23.000000000 +0000
+++ source/drivers/char/mem.c 2004-03-16 11:27:48.107874152 +0000
@@ -39,6 +39,7 @@ extern void fbmem_init(void);
extern void tapechar_init(void);
#endif
+#ifdef pgprot_noncached
/*
* Architectures vary in how they handle caching for addresses
* outside of main memory.
@@ -64,7 +65,8 @@ static inline int uncached_access(struct
&& addr >= __pa(high_memory);
#elif defined(CONFIG_IA64)
/*
- * On ia64, we ignore O_SYNC because we cannot tolerate memory attribute aliases.
+ * On ia64, we ignore O_SYNC because we cannot tolerate memory
+ * attribute aliases.
*/
return !(efi_mem_attributes(addr) & EFI_MEMORY_WB);
#elif defined(CONFIG_PPC64)
@@ -77,14 +79,15 @@ static inline int uncached_access(struct
return !page_is_ram(addr);
#else
/*
- * Accessing memory above the top the kernel knows about or through a file pointer
- * that was marked O_SYNC will be done non-cached.
+ * Accessing memory above the top the kernel knows about or through a
+ * file pointer that was marked O_SYNC will be done non-cached.
*/
if (file->f_flags & O_SYNC)
return 1;
return addr >= __pa(high_memory);
#endif
}
+#endif
#ifndef ARCH_HAS_VALID_PHYS_ADDR_RANGE
static inline int valid_phys_addr_range(unsigned long addr, size_t *count)
@@ -102,10 +105,11 @@ static inline int valid_phys_addr_range(
}
#endif
-static ssize_t do_write_mem(struct file * file, void *p, unsigned long realp,
+static ssize_t do_write_mem(void *p, unsigned long realp,
const char * buf, size_t count, loff_t *ppos)
{
ssize_t written;
+ unsigned long copied;
written = 0;
#if defined(__sparc__) || (defined(__mc68000__) && defined(CONFIG_MMU))
@@ -120,8 +124,14 @@ static ssize_t do_write_mem(struct file
written+=sz;
}
#endif
- if (copy_from_user(p, buf, count))
+ copied = copy_from_user(p, buf, count);
+ if (copied) {
+ ssize_t ret = written + (count - copied);
+
+ if (ret)
+ return ret;
return -EFAULT;
+ }
written += count;
*ppos += written;
return written;
@@ -171,31 +181,27 @@ static ssize_t write_mem(struct file * f
if (!valid_phys_addr_range(p, &count))
return -EFAULT;
- return do_write_mem(file, __va(p), p, buf, count, ppos);
+ return do_write_mem(__va(p), p, buf, count, ppos);
}
-static int mmap_mem(struct file * file, struct vm_area_struct * vma)
+static int mmap_mem(struct file *file, struct vm_area_struct *vma)
{
unsigned long offset = vma->vm_pgoff << PAGE_SHIFT;
- int uncached;
- uncached = uncached_access(file, offset);
#ifdef pgprot_noncached
- if (uncached)
+ if (uncached_access(file, offset))
vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
#endif
- /* Don't try to swap out physical pages.. */
- vma->vm_flags |= VM_RESERVED;
-
/*
- * Don't dump addresses that are not real memory to a core file.
+ * Don't try to swap out physical pages..
+ * And treat /dev/mem mappings as "IO" regions: they may not
+ * describe valid pageframes.
*/
- if (uncached)
- vma->vm_flags |= VM_IO;
+ vma->vm_flags |= VM_RESERVED|VM_IO;
- if (remap_page_range(vma, vma->vm_start, offset, vma->vm_end-vma->vm_start,
- vma->vm_page_prot))
+ if (remap_page_range(vma, vma->vm_start, offset,
+ vma->vm_end-vma->vm_start, vma->vm_page_prot))
return -EAGAIN;
return 0;
}
@@ -275,15 +281,19 @@ static ssize_t write_kmem(struct file *
unsigned long p = *ppos;
ssize_t wrote = 0;
ssize_t virtr = 0;
+ ssize_t written;
char * kbuf; /* k-addr because vwrite() takes vmlist_lock rwlock */
if (p < (unsigned long) high_memory) {
+
wrote = count;
if (count > (unsigned long) high_memory - p)
wrote = (unsigned long) high_memory - p;
- wrote = do_write_mem(file, (void*)p, p, buf, wrote, ppos);
-
+ written = do_write_mem((void*)p, p, buf, wrote, ppos);
+ if (written != wrote)
+ return written;
+ wrote = written;
p += wrote;
buf += wrote;
count -= wrote;
@@ -292,15 +302,21 @@ static ssize_t write_kmem(struct file *
if (count > 0) {
kbuf = (char *)__get_free_page(GFP_KERNEL);
if (!kbuf)
- return -ENOMEM;
+ return wrote ? wrote : -ENOMEM;
while (count > 0) {
int len = count;
if (len > PAGE_SIZE)
len = PAGE_SIZE;
- if (len && copy_from_user(kbuf, buf, len)) {
- free_page((unsigned long)kbuf);
- return -EFAULT;
+ if (len) {
+ written = copy_from_user(kbuf, buf, len);
+ if (written != len) {
+ ssize_t ret;
+
+ free_page((unsigned long)kbuf);
+ ret = wrote + virtr + (len - written);
+ return ret ? ret : -EFAULT;
+ }
}
len = vwrite(kbuf, (char *)p, len);
count -= len;
--- diff/drivers/char/n_tty.c 2004-03-11 10:20:23.000000000 +0000
+++ source/drivers/char/n_tty.c 2004-03-16 11:27:48.108874000 +0000
@@ -999,7 +999,8 @@ do_it_again:
/* NOTE: not yet done after every sleep pending a thorough
check of the logic of this change. -- jlc */
/* don't stop on /dev/console */
- if (file->f_op->write != redirected_tty_write && current->tty == tty) {
+ if (file->f_op->write != redirected_tty_write &&
+ current->signal->tty == tty) {
if (tty->pgrp <= 0)
printk("read_chan: tty->pgrp <= 0!\n");
else if (process_group(current) != tty->pgrp) {
--- diff/drivers/char/raw.c 2004-03-11 10:20:23.000000000 +0000
+++ source/drivers/char/raw.c 2004-03-16 11:27:48.109873848 +0000
@@ -17,6 +17,7 @@
#include <linux/raw.h>
#include <linux/capability.h>
#include <linux/uio.h>
+#include <linux/cdev.h>
#include <asm/uaccess.h>
@@ -260,11 +261,26 @@ static struct file_operations raw_ctl_fo
.owner = THIS_MODULE,
};
+static struct cdev raw_cdev = {
+ .kobj = {.name = "raw", },
+ .owner = THIS_MODULE,
+};
+
static int __init raw_init(void)
{
int i;
+ dev_t dev = MKDEV(RAW_MAJOR, 0);
+
+ if (register_chrdev_region(dev, MAX_RAW_MINORS, "raw"))
+ goto error;
+
+ cdev_init(&raw_cdev, &raw_fops);
+ if (cdev_add(&raw_cdev, dev, MAX_RAW_MINORS)) {
+ kobject_put(&raw_cdev.kobj);
+ unregister_chrdev_region(dev, MAX_RAW_MINORS);
+ goto error;
+ }
- register_chrdev(RAW_MAJOR, "raw", &raw_fops);
devfs_mk_cdev(MKDEV(RAW_MAJOR, 0),
S_IFCHR | S_IRUGO | S_IWUGO,
"raw/rawctl");
@@ -273,6 +289,10 @@ static int __init raw_init(void)
S_IFCHR | S_IRUGO | S_IWUGO,
"raw/raw%d", i);
return 0;
+
+error:
+ printk(KERN_ERR "error register raw device\n");
+ return 1;
}
static void __exit raw_exit(void)
@@ -283,7 +303,8 @@ static void __exit raw_exit(void)
devfs_remove("raw/raw%d", i);
devfs_remove("raw/rawctl");
devfs_remove("raw");
- unregister_chrdev(RAW_MAJOR, "raw");
+ cdev_del(&raw_cdev);
+ unregister_chrdev_region(MKDEV(RAW_MAJOR, 0), MAX_RAW_MINORS);
}
module_init(raw_init);
--- diff/drivers/char/rocket.c 2004-03-11 10:20:23.000000000 +0000
+++ source/drivers/char/rocket.c 2004-03-16 11:27:48.111873544 +0000
@@ -953,7 +953,7 @@ static int rp_open(struct tty_struct *tt
/*
* Info->count is now 1; so it's safe to sleep now.
*/
- info->session = current->session;
+ info->session = current->signal->session;
info->pgrp = process_group(current);
if ((info->flags & ROCKET_INITIALIZED) == 0) {
--- diff/drivers/char/sx.c 2004-03-11 10:20:23.000000000 +0000
+++ source/drivers/char/sx.c 2004-03-16 11:27:48.112873392 +0000
@@ -1420,7 +1420,7 @@ static int sx_open (struct tty_struct *
line = tty->index;
sx_dprintk (SX_DEBUG_OPEN, "%d: opening line %d. tty=%p ctty=%p, np=%d)\n",
- current->pid, line, tty, current->tty, sx_nports);
+ current->pid, line, tty, current->signal->tty, sx_nports);
if ((line < 0) || (line >= SX_NPORTS) || (line >= sx_nports))
return -ENODEV;
--- diff/drivers/char/sysrq.c 2004-02-09 10:36:10.000000000 +0000
+++ source/drivers/char/sysrq.c 2004-03-16 11:27:48.113873240 +0000
@@ -35,6 +35,25 @@
#include <linux/spinlock.h>
#include <asm/ptrace.h>
+#ifdef CONFIG_KGDB_SYSRQ
+
+#define GDB_OP &kgdb_op
+static void kgdb_sysrq(int key, struct pt_regs *pt_regs, struct tty_struct *tty)
+{
+ printk("kgdb sysrq\n");
+ breakpoint();
+}
+
+static struct sysrq_key_op kgdb_op = {
+ .handler = kgdb_sysrq,
+ .help_msg = "kGdb|Fgdb",
+ .action_msg = "Debug breakpoint\n",
+};
+
+#else
+#define GDB_OP NULL
+#endif
+
extern void reset_vc(unsigned int);
@@ -238,8 +257,8 @@ static struct sysrq_key_op *sysrq_key_ta
/* c */ NULL,
/* d */ NULL,
/* e */ &sysrq_term_op,
-/* f */ NULL,
-/* g */ NULL,
+/* f */ GDB_OP,
+/* g */ GDB_OP,
/* h */ NULL,
/* i */ &sysrq_kill_op,
/* j */ NULL,
--- diff/drivers/char/tty_io.c 2004-03-16 11:24:01.000000000 +0000
+++ source/drivers/char/tty_io.c 2004-03-16 11:27:48.115872936 +0000
@@ -316,7 +316,7 @@ struct tty_driver *get_tty_driver(dev_t
*/
int tty_check_change(struct tty_struct * tty)
{
- if (current->tty != tty)
+ if (current->signal->tty != tty)
return 0;
if (tty->pgrp <= 0) {
printk(KERN_WARNING "tty_check_change: tty->pgrp <= 0!\n");
@@ -481,17 +481,14 @@ void do_tty_hangup(void *data)
if (tty->session > 0) {
struct list_head *l;
for_each_task_pid(tty->session, PIDTYPE_SID, p, l, pid) {
- task_t *task = p;
- do {
- if (task->tty == tty)
- task->tty = NULL;
- if (task->leader) {
- send_group_sig_info(SIGHUP, SEND_SIG_PRIV, task);
- send_group_sig_info(SIGCONT, SEND_SIG_PRIV, task);
- }
- } while_each_thread(p, task);
+ if (p->signal->tty == tty)
+ p->signal->tty = NULL;
+ if (!p->signal->leader)
+ continue;
+ send_group_sig_info(SIGHUP, SEND_SIG_PRIV, p);
+ send_group_sig_info(SIGCONT, SEND_SIG_PRIV, p);
if (tty->pgrp > 0)
- p->tty_old_pgrp = tty->pgrp;
+ p->signal->tty_old_pgrp = tty->pgrp;
}
}
read_unlock(&tasklist_lock);
@@ -570,15 +567,15 @@ void disassociate_ctty(int on_exit)
lock_kernel();
- tty = current->tty;
+ tty = current->signal->tty;
if (tty) {
tty_pgrp = tty->pgrp;
if (on_exit && tty->driver->type != TTY_DRIVER_TYPE_PTY)
tty_vhangup(tty);
} else {
- if (current->tty_old_pgrp) {
- kill_pg(current->tty_old_pgrp, SIGHUP, on_exit);
- kill_pg(current->tty_old_pgrp, SIGCONT, on_exit);
+ if (current->signal->tty_old_pgrp) {
+ kill_pg(current->signal->tty_old_pgrp, SIGHUP, on_exit);
+ kill_pg(current->signal->tty_old_pgrp, SIGCONT, on_exit);
}
unlock_kernel();
return;
@@ -589,17 +586,13 @@ void disassociate_ctty(int on_exit)
kill_pg(tty_pgrp, SIGCONT, on_exit);
}
- current->tty_old_pgrp = 0;
+ current->signal->tty_old_pgrp = 0;
tty->session = 0;
tty->pgrp = -1;
read_lock(&tasklist_lock);
- for_each_task_pid(current->session, PIDTYPE_SID, p, l, pid) {
- task_t *task = p;
- do {
- task->tty = NULL;
- } while_each_thread(p, task);
- }
+ for_each_task_pid(current->signal->session, PIDTYPE_SID, p, l, pid)
+ p->signal->tty = NULL;
read_unlock(&tasklist_lock);
unlock_kernel();
}
@@ -1267,20 +1260,11 @@ static void release_dev(struct file * fi
struct pid *pid;
read_lock(&tasklist_lock);
- for_each_task_pid(tty->session, PIDTYPE_SID, p, l, pid) {
- task_t *task = p;
- do {
- task->tty = NULL;
- } while_each_thread(p, task);
- }
- if (o_tty) {
- for_each_task_pid(o_tty->session, PIDTYPE_SID, p,l, pid) {
- task_t *task = p;
- do {
- task->tty = NULL;
- } while_each_thread(p, task);
- }
- }
+ for_each_task_pid(tty->session, PIDTYPE_SID, p, l, pid)
+ p->signal->tty = NULL;
+ if (o_tty)
+ for_each_task_pid(o_tty->session, PIDTYPE_SID, p,l, pid)
+ p->signal->tty = NULL;
read_unlock(&tasklist_lock);
}
@@ -1351,10 +1335,10 @@ static int tty_open(struct inode * inode
retry_open:
noctty = filp->f_flags & O_NOCTTY;
if (device == MKDEV(TTYAUX_MAJOR,0)) {
- if (!current->tty)
+ if (!current->signal->tty)
return -ENXIO;
- driver = current->tty->driver;
- index = current->tty->index;
+ driver = current->signal->tty->driver;
+ index = current->signal->tty->index;
filp->f_flags |= O_NONBLOCK; /* Don't let /dev/tty block */
/* noctty = 1; */
goto got_driver;
@@ -1455,14 +1439,14 @@ got_driver:
goto retry_open;
}
if (!noctty &&
- current->leader &&
- !current->tty &&
+ current->signal->leader &&
+ !current->signal->tty &&
tty->session == 0) {
task_lock(current);
- current->tty = tty;
+ current->signal->tty = tty;
task_unlock(current);
- current->tty_old_pgrp = 0;
- tty->session = current->session;
+ current->signal->tty_old_pgrp = 0;
+ tty->session = current->signal->session;
tty->pgrp = process_group(current);
}
return 0;
@@ -1520,7 +1504,7 @@ static int tiocsti(struct tty_struct *tt
{
char ch, mbz = 0;
- if ((current->tty != tty) && !capable(CAP_SYS_ADMIN))
+ if ((current->signal->tty != tty) && !capable(CAP_SYS_ADMIN))
return -EPERM;
if (get_user(ch, arg))
return -EFAULT;
@@ -1611,14 +1595,14 @@ static int tiocsctty(struct tty_struct *
struct pid *pid;
task_t *p;
- if (current->leader &&
- (current->session == tty->session))
+ if (current->signal->leader &&
+ (current->signal->session == tty->session))
return 0;
/*
* The process must be a session leader and
* not have a controlling tty already.
*/
- if (!current->leader || current->tty)
+ if (!current->signal->leader || current->signal->tty)
return -EPERM;
if (tty->session > 0) {
/*
@@ -1631,21 +1615,17 @@ static int tiocsctty(struct tty_struct *
*/
read_lock(&tasklist_lock);
- for_each_task_pid(tty->session, PIDTYPE_SID, p, l, pid) {
- task_t *task = p;
- do {
- task->tty = NULL;
- } while_each_thread(p, task);
- }
+ for_each_task_pid(tty->session, PIDTYPE_SID, p, l, pid)
+ p->signal->tty = NULL;
read_unlock(&tasklist_lock);
} else
return -EPERM;
}
task_lock(current);
- current->tty = tty;
+ current->signal->tty = tty;
task_unlock(current);
- current->tty_old_pgrp = 0;
- tty->session = current->session;
+ current->signal->tty_old_pgrp = 0;
+ tty->session = current->signal->session;
tty->pgrp = process_group(current);
return 0;
}
@@ -1656,7 +1636,7 @@ static int tiocgpgrp(struct tty_struct *
* (tty == real_tty) is a cheap way of
* testing if the tty is NOT a master pty.
*/
- if (tty == real_tty && current->tty != real_tty)
+ if (tty == real_tty && current->signal->tty != real_tty)
return -ENOTTY;
return put_user(real_tty->pgrp, arg);
}
@@ -1670,15 +1650,15 @@ static int tiocspgrp(struct tty_struct *
return -ENOTTY;
if (retval)
return retval;
- if (!current->tty ||
- (current->tty != real_tty) ||
- (real_tty->session != current->session))
+ if (!current->signal->tty ||
+ (current->signal->tty != real_tty) ||
+ (real_tty->session != current->signal->session))
return -ENOTTY;
if (get_user(pgrp, (pid_t *) arg))
return -EFAULT;
if (pgrp < 0)
return -EINVAL;
- if (session_of_pgrp(pgrp) != current->session)
+ if (session_of_pgrp(pgrp) != current->signal->session)
return -EPERM;
real_tty->pgrp = pgrp;
return 0;
@@ -1690,7 +1670,7 @@ static int tiocgsid(struct tty_struct *t
* (tty == real_tty) is a cheap way of
* testing if the tty is NOT a master pty.
*/
- if (tty == real_tty && current->tty != real_tty)
+ if (tty == real_tty && current->signal->tty != real_tty)
return -ENOTTY;
if (real_tty->session <= 0)
return -ENOTTY;
@@ -1848,12 +1828,12 @@ int tty_ioctl(struct inode * inode, stru
clear_bit(TTY_EXCLUSIVE, &tty->flags);
return 0;
case TIOCNOTTY:
- if (current->tty != tty)
+ if (current->signal->tty != tty)
return -ENOTTY;
- if (current->leader)
+ if (current->signal->leader)
disassociate_ctty(0);
task_lock(current);
- current->tty = NULL;
+ current->signal->tty = NULL;
task_unlock(current);
return 0;
case TIOCSCTTY:
@@ -1957,9 +1937,9 @@ static void __do_SAK(void *arg)
tty->driver->flush_buffer(tty);
read_lock(&tasklist_lock);
for_each_task_pid(session, PIDTYPE_SID, p, l, pid) {
- if (p->tty == tty || session > 0) {
+ if (p->signal->tty == tty || session > 0) {
printk(KERN_NOTICE "SAK: killed process %d"
- " (%s): p->session==tty->session\n",
+ " (%s): p->signal->session==tty->session\n",
p->pid, p->comm);
send_sig(SIGKILL, p, 1);
continue;
--- diff/drivers/char/vt.c 2004-03-11 10:20:23.000000000 +0000
+++ source/drivers/char/vt.c 2004-03-16 11:27:48.117872632 +0000
@@ -2277,7 +2277,7 @@ int tioclinux(struct tty_struct *tty, un
if (tty->driver->type != TTY_DRIVER_TYPE_CONSOLE)
return -EINVAL;
- if (current->tty != tty && !capable(CAP_SYS_ADMIN))
+ if (current->signal->tty != tty && !capable(CAP_SYS_ADMIN))
return -EPERM;
if (get_user(type, (char *)arg))
return -EFAULT;
--- diff/drivers/char/vt_ioctl.c 2004-03-11 10:20:23.000000000 +0000
+++ source/drivers/char/vt_ioctl.c 2004-03-16 11:27:48.118872480 +0000
@@ -382,7 +382,7 @@ int vt_ioctl(struct tty_struct *tty, str
* to be the owner of the tty, or have CAP_SYS_TTY_CONFIG.
*/
perm = 0;
- if (current->tty == tty || capable(CAP_SYS_TTY_CONFIG))
+ if (current->signal->tty == tty || capable(CAP_SYS_TTY_CONFIG))
perm = 1;
kbd = kbd_table + console;
@@ -1221,4 +1221,3 @@ void change_console(unsigned int new_con
complete_change_console(new_console);
}
-
--- diff/drivers/char/watchdog/pcwd_usb.c 2004-03-11 10:20:23.000000000 +0000
+++ source/drivers/char/watchdog/pcwd_usb.c 2004-03-16 11:27:48.119872328 +0000
@@ -589,7 +589,7 @@ static int usb_pcwd_probe(struct usb_int
}
/* get the active interface descriptor */
- iface_desc = &interface->altsetting[interface->act_altsetting];
+ iface_desc = interface->cur_altsetting;
/* check out that we have a HID device */
if (!(iface_desc->desc.bInterfaceClass == USB_CLASS_HID)) {
--- diff/drivers/ide/Kconfig 2004-03-11 10:20:23.000000000 +0000
+++ source/drivers/ide/Kconfig 2004-03-16 11:27:48.124871568 +0000
@@ -559,6 +559,16 @@ config BLK_DEV_AMD74XX
change PIO, DMA and UDMA speeds and to configure the chip to
optimum performance.
+config BLK_DEV_ATIIXP
+ tristate "ATI IXP chipset IDE support"
+ depends on X86
+ help
+ This driver adds explicit support for ATI IXP chipset.
+ This allows the kernel to change PIO, DMA and UDMA speeds
+ and to configure the chip to optimum performance.
+
+ Say Y here if you have an ATI IXP chipset IDE controller.
+
config BLK_DEV_CMD64X
tristate "CMD64{3|6|8|9} chipset support"
help
--- diff/drivers/ide/ide-cd.c 2004-03-16 11:24:01.000000000 +0000
+++ source/drivers/ide/ide-cd.c 2004-03-16 11:27:48.121872024 +0000
@@ -554,53 +554,6 @@ static void cdrom_queue_request_sense(id
(void) ide_do_drive_cmd(drive, rq, ide_preempt);
}
-
-/*
- * Error reporting, in human readable form (luxurious, but a memory hog).
- */
-byte ide_cdrom_dump_status (ide_drive_t *drive, const char *msg, byte stat)
-{
- unsigned long flags;
-
- atapi_status_t status;
- atapi_error_t error;
-
- status.all = stat;
- local_irq_set(flags);
- printk("%s: %s: status=0x%02x", drive->name, msg, stat);
-#if FANCY_STATUS_DUMPS
- printk(" { ");
- if (status.b.bsy)
- printk("Busy ");
- else {
- if (status.b.drdy) printk("DriveReady ");
- if (status.b.df) printk("DeviceFault ");
- if (status.b.dsc) printk("SeekComplete ");
- if (status.b.drq) printk("DataRequest ");
- if (status.b.corr) printk("CorrectedError ");
- if (status.b.idx) printk("Index ");
- if (status.b.check) printk("Error ");
- }
- printk("}");
-#endif /* FANCY_STATUS_DUMPS */
- printk("\n");
- if ((status.all & (status.b.bsy|status.b.check)) == status.b.check) {
- error.all = HWIF(drive)->INB(IDE_ERROR_REG);
- printk("%s: %s: error=0x%02x", drive->name, msg, error.all);
-#if FANCY_STATUS_DUMPS
- if (error.b.ili) printk("IllegalLengthIndication ");
- if (error.b.eom) printk("EndOfMedia ");
- if (error.b.abrt) printk("Aborted Command ");
- if (error.b.mcr) printk("MediaChangeRequested ");
- if (error.b.sense_key) printk("LastFailedSense 0x%02x ",
- error.b.sense_key);
-#endif /* FANCY_STATUS_DUMPS */
- printk("\n");
- }
- local_irq_restore(flags);
- return error.all;
-}
-
/*
* ide_error() takes action based on the error returned by the drive.
*/
@@ -609,7 +562,7 @@ ide_startstop_t ide_cdrom_error (ide_dri
struct request *rq;
byte err;
- err = ide_cdrom_dump_status(drive, msg, stat);
+ err = ide_dump_atapi_status(drive, msg, stat);
if (drive == NULL || (rq = HWGROUP(drive)->rq) == NULL)
return ide_stopped;
/* retry only "normal" I/O: */
@@ -3412,7 +3365,7 @@ static ide_driver_t ide_cdrom_driver = {
.supports_dsc_overlap = 1,
.cleanup = ide_cdrom_cleanup,
.do_request = ide_do_rw_cdrom,
- .sense = ide_cdrom_dump_status,
+ .sense = ide_dump_atapi_status,
.error = ide_cdrom_error,
.abort = ide_cdrom_abort,
.capacity = ide_cdrom_capacity,
--- diff/drivers/ide/ide-lib.c 2003-09-17 11:28:05.000000000 +0000
+++ source/drivers/ide/ide-lib.c 2004-03-16 11:27:48.122871872 +0000
@@ -447,3 +447,55 @@ int ide_set_xfer_rate(ide_drive_t *drive
EXPORT_SYMBOL_GPL(ide_set_xfer_rate);
+/**
+ * ide_dump_atapi_status - print human readable atapi status
+ * @drive: drive that status applies to
+ * @msg: text message to print
+ * @stat: status byte to decode
+ *
+ * Error reporting, in human readable form (luxurious, but a memory hog).
+ */
+byte ide_dump_atapi_status (ide_drive_t *drive, const char *msg, byte stat)
+{
+ unsigned long flags;
+
+ atapi_status_t status;
+ atapi_error_t error;
+
+ status.all = stat;
+ local_irq_set(flags);
+ printk("%s: %s: status=0x%02x", drive->name, msg, stat);
+#if FANCY_STATUS_DUMPS
+ printk(" { ");
+ if (status.b.bsy)
+ printk("Busy ");
+ else {
+ if (status.b.drdy) printk("DriveReady ");
+ if (status.b.df) printk("DeviceFault ");
+ if (status.b.dsc) printk("SeekComplete ");
+ if (status.b.drq) printk("DataRequest ");
+ if (status.b.corr) printk("CorrectedError ");
+ if (status.b.idx) printk("Index ");
+ if (status.b.check) printk("Error ");
+ }
+ printk("}");
+#endif /* FANCY_STATUS_DUMPS */
+ printk("\n");
+ if ((status.all & (status.b.bsy|status.b.check)) == status.b.check) {
+ error.all = HWIF(drive)->INB(IDE_ERROR_REG);
+ printk("%s: %s: error=0x%02x", drive->name, msg, error.all);
+#if FANCY_STATUS_DUMPS
+ if (error.b.ili) printk("IllegalLengthIndication ");
+ if (error.b.eom) printk("EndOfMedia ");
+ if (error.b.abrt) printk("Aborted Command ");
+ if (error.b.mcr) printk("MediaChangeRequested ");
+ if (error.b.sense_key) printk("LastFailedSense 0x%02x ",
+ error.b.sense_key);
+#endif /* FANCY_STATUS_DUMPS */
+ printk("\n");
+ }
+ local_irq_restore(flags);
+ return error.all;
+}
+
+EXPORT_SYMBOL(ide_dump_atapi_status);
--- diff/drivers/ide/ide-probe.c 2004-03-11 10:20:24.000000000 +0000
+++ source/drivers/ide/ide-probe.c 2004-03-16 11:27:48.123871720 +0000
@@ -922,8 +922,12 @@ static int ide_init_queue(ide_drive_t *d
q->queuedata = HWGROUP(drive);
blk_queue_segment_boundary(q, 0xffff);
- if (!hwif->rqsize)
- hwif->rqsize = hwif->no_lba48 ? 256 : 65536;
+ if (!hwif->rqsize) {
+ if (hwif->max_rqsize)
+ hwif->rqsize = hwif->max_rqsize(drive);
+ else
+ hwif->rqsize = hwif->no_lba48 ? 256 : 65536;
+ }
if (hwif->rqsize < max_sectors)
max_sectors = hwif->rqsize;
blk_queue_max_sectors(q, max_sectors);
--- diff/drivers/ide/pci/Makefile 2004-03-11 10:20:24.000000000 +0000
+++ source/drivers/ide/pci/Makefile 2004-03-16 11:27:48.126871264 +0000
@@ -3,6 +3,7 @@ obj-$(CONFIG_BLK_DEV_ADMA100) += adma10
obj-$(CONFIG_BLK_DEV_AEC62XX) += aec62xx.o
obj-$(CONFIG_BLK_DEV_ALI15X3) += alim15x3.o
obj-$(CONFIG_BLK_DEV_AMD74XX) += amd74xx.o
+obj-$(CONFIG_BLK_DEV_ATIIXP) += atiixp.o
obj-$(CONFIG_BLK_DEV_CMD64X) += cmd64x.o
obj-$(CONFIG_BLK_DEV_CS5520) += cs5520.o
obj-$(CONFIG_BLK_DEV_CS5530) += cs5530.o
--- diff/drivers/ide/pci/siimage.c 2004-03-11 10:20:24.000000000 +0000
+++ source/drivers/ide/pci/siimage.c 2004-03-16 11:27:48.127871112 +0000
@@ -203,13 +203,12 @@ static byte siimage_ratemask (ide_drive_
else
pci_read_config_byte(hwif->pci_dev, 0x8A, &scsc);
- if(is_sata(hwif))
- {
- if(strstr(drive->id->model, "Maxtor"))
+ if (is_sata(hwif)) {
+ if (strstr(drive->id->model, "Maxtor 4D060H3"))
return 3;
return 4;
}
-
+
if ((scsc & 0x30) == 0x10) /* 133 */
mode = 4;
else if ((scsc & 0x30) == 0x20) /* 2xPCI */
@@ -1046,25 +1045,34 @@ static void __init init_mmio_iops_siimag
hwif->mmio = 2;
}
-static int is_dev_seagate_sata(ide_drive_t *drive)
+/* TODO firmware versions should be added - eric */
+static const char * sil_blacklist [] = {
+ "ST320012AS",
+ "ST330013AS",
+ "ST340017AS",
+ "ST360015AS",
+ "ST380023AS",
+ "ST3120023AS",
+ "ST340014ASL",
+ "ST360014ASL",
+ "ST380011ASL",
+ "ST3120022ASL",
+ "ST3160021ASL",
+};
+
+static unsigned int siimage_sata_max_rqsize(ide_drive_t *drive)
{
const char *s = &drive->id->model[0];
- unsigned len;
+ unsigned int n;
- if (!drive->present)
- return 0;
-
- len = strnlen(s, sizeof(drive->id->model));
-
- if ((len > 4) && (!memcmp(s, "ST", 2))) {
- if ((!memcmp(s + len - 2, "AS", 2)) ||
- (!memcmp(s + len - 3, "ASL", 3))) {
- printk(KERN_INFO "%s: applying pessimistic Seagate "
- "errata fix\n", drive->name);
- return 1;
+ for (n = 0; n < ARRAY_SIZE(sil_blacklist); n++)
+ if (!memcmp(sil_blacklist[n], s, strlen(sil_blacklist[n]))) {
+ printk(KERN_INFO "%s: applying Seagate errata fix\n",
+ drive->name);
+ return 15;
}
- }
- return 0;
+
+ return 128;
}
/**
@@ -1087,9 +1095,10 @@ static void __init init_iops_siimage (id
hwif->hwif_data = 0;
- hwif->rqsize = 128;
- if (is_sata(hwif) && is_dev_seagate_sata(&hwif->drives[0]))
- hwif->rqsize = 15;
+ if (is_sata(hwif) && (class_rev <= 0x01))
+ hwif->max_rqsize = siimage_sata_max_rqsize;
+ else
+ hwif->rqsize = 128;
if (pci_get_drvdata(dev) == NULL)
return;
--- diff/drivers/ieee1394/ohci1394.c 2004-03-16 11:24:01.000000000 +0000
+++ source/drivers/ieee1394/ohci1394.c 2004-03-16 11:27:48.129870808 +0000
@@ -620,6 +620,39 @@ static void ohci_initialize(struct ti_oh
if (status & 0x20)
set_phy_reg(ohci, 8, status & ~1);
}
+
+ /* Serial EEPROM Sanity check. */
+ if ((ohci->max_packet_size < 512) ||
+ (ohci->max_packet_size > 4096)) {
+ /* Serial EEPROM contents are suspect, set a sane max packet
+ * size and print the raw contents for bug reports if verbose
+ * debug is enabled. */
+#ifdef CONFIG_IEEE1394_VERBOSEDEBUG
+ int i;
+#endif
+
+ PRINT(KERN_DEBUG, "Serial EEPROM has suspicious values, "
+ "attempting to setting max_packet_size to 512 bytes");
+ reg_write(ohci, OHCI1394_BusOptions,
+ (reg_read(ohci, OHCI1394_BusOptions) & 0xf007) | 0x8002);
+ ohci->max_packet_size = 512;
+#ifdef CONFIG_IEEE1394_VERBOSEDEBUG
+ PRINT(KERN_DEBUG, " EEPROM Present: %d",
+ (reg_read(ohci, OHCI1394_Version) >> 24) & 0x1);
+ reg_write(ohci, OHCI1394_GUID_ROM, 0x80000000);
+
+ for (i = 0;
+ ((i < 1000) &&
+ (reg_read(ohci, OHCI1394_GUID_ROM) & 0x80000000)); i++)
+ udelay(10);
+
+ for (i = 0; i < 0x20; i++) {
+ reg_write(ohci, OHCI1394_GUID_ROM, 0x02000000);
+ PRINT(KERN_DEBUG, " EEPROM %02x: %02x", i,
+ (reg_read(ohci, OHCI1394_GUID_ROM) >> 16) & 0xff);
+ }
+#endif
+ }
}
/*
--- diff/drivers/ieee1394/pcilynx.c 2004-03-11 10:20:24.000000000 +0000
+++ source/drivers/ieee1394/pcilynx.c 2004-03-16 11:27:48.131870504 +0000
@@ -1460,7 +1460,7 @@ static void remove_card(struct pci_dev *
reg_write(lynx, PCI_INT_ENABLE, 0);
free_irq(lynx->dev->irq, lynx);
- /* Disable IRM Contender */
+ /* Disable IRM Contender and LCtrl */
if (lynx->phyic.reg_1394a)
set_phy_reg(lynx, 4, ~0xc0 & get_phy_reg(lynx, 4));
@@ -1788,12 +1788,12 @@ static int __devinit add_card(struct pci
reg_set_bits(lynx, GPIO_CTRL_A, 0x1);
reg_write(lynx, GPIO_DATA_BASE + 0x3c, 0x1);
} else {
- /* set the contender bit in the extended PHY register
+ /* set the contender and LCtrl bit in the extended PHY register
* set. (Should check that bis 0,1,2 (=0xE0) is set
* in register 2?)
*/
i = get_phy_reg(lynx, 4);
- if (i != -1) set_phy_reg(lynx, 4, i | 0x40);
+ if (i != -1) set_phy_reg(lynx, 4, i | 0xc0);
}
--- diff/drivers/input/evdev.c 2004-02-18 08:54:09.000000000 +0000
+++ source/drivers/input/evdev.c 2004-03-16 11:27:48.132870352 +0000
@@ -209,7 +209,7 @@ static int evdev_ioctl(struct inode *ino
struct evdev *evdev = list->evdev;
struct input_dev *dev = evdev->handle.dev;
struct input_absinfo abs;
- int t, u, v;
+ int i, t, u, v;
if (!evdev->exist) return -ENODEV;
@@ -234,6 +234,9 @@ static int evdev_ioctl(struct inode *ino
u = SET_INPUT_KEYCODE(dev, t, v);
clear_bit(u, dev->keybit);
set_bit(v, dev->keybit);
+ for (i = 0; i < dev->keycodemax; i++)
+ if (INPUT_KEYCODE(dev,i) == u)
+ set_bit(u, dev->keybit);
return 0;
case EVIOCSFF:
--- diff/drivers/input/gameport/ns558.c 2003-08-26 09:00:52.000000000 +0000
+++ source/drivers/input/gameport/ns558.c 2004-03-16 11:27:48.133870200 +0000
@@ -77,7 +77,7 @@ static void ns558_isa_probe(int io)
* No one should be using this address.
*/
- if (check_region(io, 1))
+ if (!request_region(io, 1, "ns558-isa"))
return;
/*
@@ -89,7 +89,8 @@ static void ns558_isa_probe(int io)
outb(~c & ~3, io);
if (~(u = v = inb(io)) & 3) {
outb(c, io);
- return;
+ i = 0;
+ goto out;
}
/*
* After a trigger, there must be at least some bits changing.
@@ -99,7 +100,8 @@ static void ns558_isa_probe(int io)
if (u == v) {
outb(c, io);
- return;
+ i = 0;
+ goto out;
}
wait_ms(3);
/*
@@ -110,7 +112,8 @@ static void ns558_isa_probe(int io)
for (i = 0; i < 1000; i++)
if ((u ^ inb(io)) & 0xf) {
outb(c, io);
- return;
+ i = 0;
+ goto out;
}
/*
* And now find the number of mirrors of the port.
@@ -118,7 +121,9 @@ static void ns558_isa_probe(int io)
for (i = 1; i < 5; i++) {
- if (check_region(io & (-1 << i), (1 << i))) /* Don't disturb anyone */
+ release_region(io & (-1 << (i-1)), (1 << (i-1)));
+
+ if (!request_region(io & (-1 << i), (1 << i), "ns558-isa")) /* Don't disturb anyone */
break;
outb(0xff, io & (-1 << i));
@@ -126,18 +131,25 @@ static void ns558_isa_probe(int io)
if (inb(io & (-1 << i)) != inb((io & (-1 << i)) + (1 << i) - 1)) b++;
wait_ms(3);
- if (b > 300) /* We allow 30% difference */
+ if (b > 300) { /* We allow 30% difference */
+ release_region(io & (-1 << i), (1 << i));
break;
+ }
}
i--;
+ if (i != 4) {
+ if (!request_region(io & (-1 << i), (1 << i), "ns558-isa"))
+ return;
+ }
+
if (!(port = kmalloc(sizeof(struct ns558), GFP_KERNEL))) {
printk(KERN_ERR "ns558: Memory allocation failed.\n");
- return;
+ goto out;
}
- memset(port, 0, sizeof(struct ns558));
-
+ memset(port, 0, sizeof(struct ns558));
+
port->type = NS558_ISA;
port->size = (1 << i);
port->gameport.io = io;
@@ -148,8 +160,6 @@ static void ns558_isa_probe(int io)
sprintf(port->phys, "isa%04x/gameport0", io & (-1 << i));
sprintf(port->name, "NS558 ISA");
- request_region(io & (-1 << i), (1 << i), "ns558-isa");
-
gameport_register_port(&port->gameport);
printk(KERN_INFO "gameport: NS558 ISA at %#x", port->gameport.io);
@@ -157,6 +167,9 @@ static void ns558_isa_probe(int io)
printk(" speed %d kHz\n", port->gameport.speed);
list_add(&port->node, &ns558_list);
+ return;
+out:
+ release_region(io & (-1 << i), (1 << i));
}
#ifdef CONFIG_PNP
@@ -276,6 +289,7 @@ void __exit ns558_exit(void)
#endif
case NS558_ISA:
release_region(port->gameport.io & ~(port->size - 1), port->size);
+ kfree(port);
break;
default:
--- diff/drivers/input/joystick/amijoy.c 2003-06-09 13:18:18.000000000 +0000
+++ source/drivers/input/joystick/amijoy.c 2004-03-16 11:27:48.134870048 +0000
@@ -32,6 +32,7 @@
#include <linux/errno.h>
#include <linux/kernel.h>
#include <linux/module.h>
+#include <linux/moduleparam.h>
#include <linux/init.h>
#include <linux/input.h>
#include <linux/interrupt.h>
@@ -42,10 +43,15 @@
MODULE_AUTHOR("Vojtech Pavlik <vojtech@ucw.cz>");
MODULE_DESCRIPTION("Driver for Amiga joysticks");
-MODULE_PARM(amijoy, "1-2i");
MODULE_LICENSE("GPL");
static int amijoy[2] = { 0, 1 };
+static int amijoy_nargs;
+module_param_array_named(map, amijoy, uint, amijoy_nargs, 0);
+MODULE_PARM_DESC(map, "Map of attached joysticks in form of <a>,<b> (default is 0,1)");
+
+__obsolete_setup("amijoy=");
+
static int amijoy_used[2] = { 0, 0 };
static struct input_dev amijoy_dev[2];
static char *amijoy_phys[2] = { "amijoy/input0", "amijoy/input1" };
@@ -101,17 +107,6 @@ static void amijoy_close(struct input_de
free_irq(IRQ_AMIGA_VERTB, amijoy_interrupt);
}
-static int __init amijoy_setup(char *str)
-{
- int i;
- int ints[4];
-
- str = get_options(str, ARRAY_SIZE(ints), ints);
- for (i = 0; i <= ints[0] && i < 2; i++) amijoy[i] = ints[i+1];
- return 1;
-}
-__setup("amijoy=", amijoy_setup);
-
static int __init amijoy_init(void)
{
int i, j;
--- diff/drivers/input/joystick/analog.c 2003-10-09 08:47:34.000000000 +0000
+++ source/drivers/input/joystick/analog.c 2004-03-16 11:27:48.135869896 +0000
@@ -32,6 +32,7 @@
#include <linux/delay.h>
#include <linux/kernel.h>
#include <linux/module.h>
+#include <linux/moduleparam.h>
#include <linux/slab.h>
#include <linux/bitops.h>
#include <linux/init.h>
@@ -50,9 +51,12 @@ MODULE_LICENSE("GPL");
#define ANALOG_PORTS 16
static char *js[ANALOG_PORTS];
+static int js_nargs;
static int analog_options[ANALOG_PORTS];
-MODULE_PARM(js, "1-" __MODULE_STRING(ANALOG_PORTS) "s");
-MODULE_PARM_DESC(js, "Analog joystick options");
+module_param_array_named(map, js, charp, js_nargs, 0);
+MODULE_PARM_DESC(map, "Describes analog joysticks type/capabilities");
+
+__obsolete_setup("js=");
/*
* Times, feature definitions.
@@ -711,7 +715,7 @@ static void analog_parse_options(void)
int i, j;
char *end;
- for (i = 0; i < ANALOG_PORTS && js[i]; i++) {
+ for (i = 0; i < js_nargs; i++) {
for (j = 0; analog_types[j].name; j++)
if (!strcmp(analog_types[j].name, js[i])) {
@@ -742,24 +746,6 @@ static struct gameport_dev analog_dev =
.disconnect = analog_disconnect,
};
-#ifndef MODULE
-static int __init analog_setup(char *str)
-{
- char *s = str;
- int i = 0;
-
- if (!str || !*str) return 0;
-
- while ((str = s) && (i < ANALOG_PORTS)) {
- if ((s = strchr(str,','))) *s++ = 0;
- js[i++] = str;
- }
-
- return 1;
-}
-__setup("js=", analog_setup);
-#endif
-
int __init analog_init(void)
{
analog_parse_options();
--- diff/drivers/input/joystick/db9.c 2004-02-18 08:54:09.000000000 +0000
+++ source/drivers/input/joystick/db9.c 2004-03-16 11:27:48.136869744 +0000
@@ -33,6 +33,7 @@
#include <linux/kernel.h>
#include <linux/module.h>
+#include <linux/moduleparam.h>
#include <linux/delay.h>
#include <linux/init.h>
#include <linux/parport.h>
@@ -42,9 +43,24 @@ MODULE_AUTHOR("Vojtech Pavlik <vojtech@u
MODULE_DESCRIPTION("Atari, Amstrad, Commodore, Amiga, Sega, etc. joystick driver");
MODULE_LICENSE("GPL");
-MODULE_PARM(db9, "2i");
-MODULE_PARM(db9_2, "2i");
-MODULE_PARM(db9_3, "2i");
+static int db9[] __initdata = { -1, 0 };
+static int db9_nargs __initdata = 0;
+module_param_array_named(dev, db9, int, db9_nargs, 0);
+MODULE_PARM_DESC(dev, "Describes first attached device (<parport#>,<type>)");
+
+static int db9_2[] __initdata = { -1, 0 };
+static int db9_nargs_2 __initdata = 0;
+module_param_array_named(dev2, db9_2, int, db9_nargs_2, 0);
+MODULE_PARM_DESC(dev2, "Describes second attached device (<parport#>,<type>)");
+
+static int db9_3[] __initdata = { -1, 0 };
+static int db9_nargs_3 __initdata = 0;
+module_param_array_named(dev3, db9_3, int, db9_nargs_3, 0);
+MODULE_PARM_DESC(dev3, "Describes third attached device (<parport#>,<type>)");
+
+__obsolete_setup("db9=");
+__obsolete_setup("db9_2=");
+__obsolete_setup("db9_3=");
#define DB9_MULTI_STICK 0x01
#define DB9_MULTI2_STICK 0x02
@@ -76,10 +92,6 @@ MODULE_PARM(db9_3, "2i");
#define DB9_GENESIS6_DELAY 14
#define DB9_REFRESH_TIME HZ/100
-static int db9[] __initdata = { -1, 0 };
-static int db9_2[] __initdata = { -1, 0 };
-static int db9_3[] __initdata = { -1, 0 };
-
struct db9 {
struct input_dev dev[DB9_MAX_DEVICES];
struct timer_list timer;
@@ -518,7 +530,7 @@ static void db9_close(struct input_dev *
}
}
-static struct db9 __init *db9_probe(int *config)
+static struct db9 __init *db9_probe(int *config, int nargs)
{
struct db9 *db9;
struct parport *pp;
@@ -526,6 +538,12 @@ static struct db9 __init *db9_probe(int
if (config[0] < 0)
return NULL;
+
+ if (nargs < 2) {
+ printk(KERN_ERR "db9.c: Device type must be specified.\n");
+ return NULL;
+ }
+
if (config[1] < 1 || config[1] >= DB9_MAX_PAD || !db9_buttons[config[1]]) {
printk(KERN_ERR "db9.c: bad config\n");
return NULL;
@@ -601,38 +619,11 @@ static struct db9 __init *db9_probe(int
return db9;
}
-#ifndef MODULE
-static int __init db9_setup(char *str)
-{
- int i, ints[3];
- get_options(str, ARRAY_SIZE(ints), ints);
- for (i = 0; i <= ints[0] && i < 2; i++) db9[i] = ints[i + 1];
- return 1;
-}
-static int __init db9_setup_2(char *str)
-{
- int i, ints[3];
- get_options(str, ARRAY_SIZE(ints), ints);
- for (i = 0; i <= ints[0] && i < 2; i++) db9_2[i] = ints[i + 1];
- return 1;
-}
-static int __init db9_setup_3(char *str)
-{
- int i, ints[3];
- get_options(str, ARRAY_SIZE(ints), ints);
- for (i = 0; i <= ints[0] && i < 2; i++) db9_3[i] = ints[i + 1];
- return 1;
-}
-__setup("db9=", db9_setup);
-__setup("db9_2=", db9_setup_2);
-__setup("db9_3=", db9_setup_3);
-#endif
-
int __init db9_init(void)
{
- db9_base[0] = db9_probe(db9);
- db9_base[1] = db9_probe(db9_2);
- db9_base[2] = db9_probe(db9_3);
+ db9_base[0] = db9_probe(db9, db9_nargs);
+ db9_base[1] = db9_probe(db9_2, db9_nargs_2);
+ db9_base[2] = db9_probe(db9_3, db9_nargs_3);
if (db9_base[0] || db9_base[1] || db9_base[2])
return 0;
--- diff/drivers/input/joystick/gamecon.c 2004-02-18 08:54:09.000000000 +0000
+++ source/drivers/input/joystick/gamecon.c 2004-03-16 11:27:48.137869592 +0000
@@ -35,6 +35,7 @@
#include <linux/kernel.h>
#include <linux/delay.h>
#include <linux/module.h>
+#include <linux/moduleparam.h>
#include <linux/init.h>
#include <linux/parport.h>
#include <linux/input.h>
@@ -43,10 +44,26 @@ MODULE_AUTHOR("Vojtech Pavlik <vojtech@u
MODULE_DESCRIPTION("NES, SNES, N64, MultiSystem, PSX gamepad driver");
MODULE_LICENSE("GPL");
-MODULE_PARM(gc, "2-6i");
-MODULE_PARM(gc_2,"2-6i");
-MODULE_PARM(gc_3,"2-6i");
-MODULE_PARM(gc_psx_delay, "i");
+static int gc[] __initdata = { -1, 0, 0, 0, 0, 0 };
+static int gc_nargs __initdata = 0;
+module_param_array_named(map, gc, int, gc_nargs, 0);
+MODULE_PARM_DESC(map, "Describers first set of devices (<parport#>,<pad1>,<pad2>,..<pad5>)");
+
+static int gc_2[] __initdata = { -1, 0, 0, 0, 0, 0 };
+static int gc_nargs_2 __initdata = 0;
+module_param_array_named(map2, gc_2, int, gc_nargs_2, 0);
+MODULE_PARM_DESC(map2, "Describers second set of devices");
+
+static int gc_3[] __initdata = { -1, 0, 0, 0, 0, 0 };
+static int gc_nargs_3 __initdata = 0;
+module_param_array_named(map3, gc_3, int, gc_nargs_3, 0);
+MODULE_PARM_DESC(map3, "Describers third set of devices");
+
+__obsolete_setup("gc=");
+__obsolete_setup("gc_2=");
+__obsolete_setup("gc_3=");
+
+/* see also gs_psx_delay parameter in PSX support section */
#define GC_SNES 1
#define GC_NES 2
@@ -71,10 +88,6 @@ struct gc {
static struct gc *gc_base[3];
-static int gc[] __initdata = { -1, 0, 0, 0, 0, 0 };
-static int gc_2[] __initdata = { -1, 0, 0, 0, 0, 0 };
-static int gc_3[] __initdata = { -1, 0, 0, 0, 0, 0 };
-
static int gc_status_bit[] = { 0x40, 0x80, 0x20, 0x10, 0x08 };
static char *gc_names[] = { NULL, "SNES pad", "NES pad", "NES FourPort", "Multisystem joystick",
@@ -232,6 +245,11 @@ static void gc_multi_read_packet(struct
#define GC_PSX_LEN(x) ((x) & 0xf) /* Low nibble is length in words */
static int gc_psx_delay = GC_PSX_DELAY;
+module_param_named(psx_delay, gc_psx_delay, uint, 0);
+MODULE_PARM_DESC(psx_delay, "Delay when accessing Sony PSX controller (usecs)");
+
+__obsolete_setup("gc_psx_delay=");
+
static short gc_psx_abs[] = { ABS_X, ABS_Y, ABS_RX, ABS_RY, ABS_HAT0X, ABS_HAT0Y };
static short gc_psx_btn[] = { BTN_TL, BTN_TR, BTN_TL2, BTN_TR2, BTN_A, BTN_B, BTN_X, BTN_Y,
BTN_START, BTN_SELECT, BTN_THUMBL, BTN_THUMBR };
@@ -468,7 +486,7 @@ static void gc_close(struct input_dev *d
}
}
-static struct gc __init *gc_probe(int *config)
+static struct gc __init *gc_probe(int *config, int nargs)
{
struct gc *gc;
struct parport *pp;
@@ -478,6 +496,11 @@ static struct gc __init *gc_probe(int *c
if (config[0] < 0)
return NULL;
+ if (nargs < 2) {
+ printk(KERN_ERR "gamecon.c: at least one device must be specified\n");
+ return NULL;
+ }
+
pp = parport_find_number(config[0]);
if (!pp) {
@@ -507,7 +530,7 @@ static struct gc __init *gc_probe(int *c
gc->timer.data = (long) gc;
gc->timer.function = gc_timer;
- for (i = 0; i < 5; i++) {
+ for (i = 0; i < nargs - 1; i++) {
if (!config[i + 1])
continue;
@@ -632,44 +655,11 @@ static struct gc __init *gc_probe(int *c
return gc;
}
-#ifndef MODULE
-static int __init gc_setup(char *str)
-{
- int i, ints[7];
- get_options(str, ARRAY_SIZE(ints), ints);
- for (i = 0; i <= ints[0] && i < 6; i++) gc[i] = ints[i + 1];
- return 1;
-}
-static int __init gc_setup_2(char *str)
-{
- int i, ints[7];
- get_options(str, ARRAY_SIZE(ints), ints);
- for (i = 0; i <= ints[0] && i < 6; i++) gc_2[i] = ints[i + 1];
- return 1;
-}
-static int __init gc_setup_3(char *str)
-{
- int i, ints[7];
- get_options(str, ARRAY_SIZE(ints), ints);
- for (i = 0; i <= ints[0] && i < 6; i++) gc_3[i] = ints[i + 1];
- return 1;
-}
-static int __init gc_psx_setup(char *str)
-{
- get_option(&str, &gc_psx_delay);
- return 1;
-}
-__setup("gc=", gc_setup);
-__setup("gc_2=", gc_setup_2);
-__setup("gc_3=", gc_setup_3);
-__setup("gc_psx_delay=", gc_psx_setup);
-#endif
-
int __init gc_init(void)
{
- gc_base[0] = gc_probe(gc);
- gc_base[1] = gc_probe(gc_2);
- gc_base[2] = gc_probe(gc_3);
+ gc_base[0] = gc_probe(gc, gc_nargs);
+ gc_base[1] = gc_probe(gc_2, gc_nargs_2);
+ gc_base[2] = gc_probe(gc_3, gc_nargs_3);
if (gc_base[0] || gc_base[1] || gc_base[2])
return 0;
--- diff/drivers/input/joystick/iforce/iforce-usb.c 2003-09-30 14:46:14.000000000 +0000
+++ source/drivers/input/joystick/iforce/iforce-usb.c 2004-03-16 11:27:48.138869440 +0000
@@ -135,7 +135,7 @@ static int iforce_usb_probe(struct usb_i
struct usb_endpoint_descriptor *epirq, *epout;
struct iforce *iforce;
- interface = &intf->altsetting[intf->act_altsetting];
+ interface = intf->cur_altsetting;
epirq = &interface->endpoint[0].desc;
epout = &interface->endpoint[1].desc;
--- diff/drivers/input/joystick/turbografx.c 2004-02-18 08:54:09.000000000 +0000
+++ source/drivers/input/joystick/turbografx.c 2004-03-16 11:27:48.138869440 +0000
@@ -35,15 +35,31 @@
#include <linux/parport.h>
#include <linux/input.h>
#include <linux/module.h>
+#include <linux/moduleparam.h>
#include <linux/init.h>
MODULE_AUTHOR("Vojtech Pavlik <vojtech@ucw.cz>");
MODULE_DESCRIPTION("TurboGraFX parallel port interface driver");
MODULE_LICENSE("GPL");
-MODULE_PARM(tgfx, "2-8i");
-MODULE_PARM(tgfx_2, "2-8i");
-MODULE_PARM(tgfx_3, "2-8i");
+static int tgfx[] __initdata = { -1, 0, 0, 0, 0, 0, 0, 0 };
+static int tgfx_nargs __initdata = 0;
+module_param_array_named(map, tgfx, int, tgfx_nargs, 0);
+MODULE_PARM_DESC(map, "Describes first set of devices (<parport#>,<js1>,<js2>,..<js7>");
+
+static int tgfx_2[] __initdata = { -1, 0, 0, 0, 0, 0, 0, 0 };
+static int tgfx_nargs_2 __initdata = 0;
+module_param_array_named(map2, tgfx_2, int, tgfx_nargs_2, 0);
+MODULE_PARM_DESC(map2, "Describes second set of devices");
+
+static int tgfx_3[] __initdata = { -1, 0, 0, 0, 0, 0, 0, 0 };
+static int tgfx_nargs_3 __initdata = 0;
+module_param_array_named(map3, tgfx_3, int, tgfx_nargs_3, 0);
+MODULE_PARM_DESC(map3, "Describes third set of devices");
+
+__obsolete_setup("tgfx=");
+__obsolete_setup("tgfx_2=");
+__obsolete_setup("tgfx_3=");
#define TGFX_REFRESH_TIME HZ/100 /* 10 ms */
@@ -58,10 +74,6 @@ MODULE_PARM(tgfx_3, "2-8i");
#define TGFX_TOP 0x01
#define TGFX_TOP2 0x08
-static int tgfx[] __initdata = { -1, 0, 0, 0, 0, 0, 0, 0 };
-static int tgfx_2[] __initdata = { -1, 0, 0, 0, 0, 0, 0, 0 };
-static int tgfx_3[] __initdata = { -1, 0, 0, 0, 0, 0, 0, 0 };
-
static int tgfx_buttons[] = { BTN_TRIGGER, BTN_THUMB, BTN_THUMB2, BTN_TOP, BTN_TOP2 };
static char *tgfx_name = "TurboGraFX Multisystem joystick";
@@ -133,7 +145,7 @@ static void tgfx_close(struct input_dev
* tgfx_probe() probes for tg gamepads.
*/
-static struct tgfx __init *tgfx_probe(int *config)
+static struct tgfx __init *tgfx_probe(int *config, int nargs)
{
struct tgfx *tgfx;
struct parport *pp;
@@ -142,6 +154,11 @@ static struct tgfx __init *tgfx_probe(in
if (config[0] < 0)
return NULL;
+ if (nargs < 2) {
+ printk(KERN_ERR "turbografx.c: at least one joystick must be specified\n");
+ return NULL;
+ }
+
pp = parport_find_number(config[0]);
if (!pp) {
@@ -171,7 +188,7 @@ static struct tgfx __init *tgfx_probe(in
tgfx->sticks = 0;
- for (i = 0; i < 7; i++)
+ for (i = 0; i < nargs - 1; i++)
if (config[i+1] > 0 && config[i+1] < 6) {
tgfx->sticks |= (1 << i);
@@ -212,38 +229,11 @@ static struct tgfx __init *tgfx_probe(in
return tgfx;
}
-#ifndef MODULE
-static int __init tgfx_setup(char *str)
-{
- int i, ints[9];
- get_options(str, ARRAY_SIZE(ints), ints);
- for (i = 0; i <= ints[0] && i < 8; i++) tgfx[i] = ints[i + 1];
- return 1;
-}
-static int __init tgfx_setup_2(char *str)
-{
- int i, ints[9];
- get_options(str, ARRAY_SIZE(ints), ints);
- for (i = 0; i <= ints[0] && i < 8; i++) tgfx_2[i] = ints[i + 1];
- return 1;
-}
-static int __init tgfx_setup_3(char *str)
-{
- int i, ints[9];
- get_options(str, ARRAY_SIZE(ints), ints);
- for (i = 0; i <= ints[0] && i < 8; i++) tgfx_3[i] = ints[i + 1];
- return 1;
-}
-__setup("tgfx=", tgfx_setup);
-__setup("tgfx_2=", tgfx_setup_2);
-__setup("tgfx_3=", tgfx_setup_3);
-#endif
-
int __init tgfx_init(void)
{
- tgfx_base[0] = tgfx_probe(tgfx);
- tgfx_base[1] = tgfx_probe(tgfx_2);
- tgfx_base[2] = tgfx_probe(tgfx_3);
+ tgfx_base[0] = tgfx_probe(tgfx, tgfx_nargs);
+ tgfx_base[1] = tgfx_probe(tgfx_2, tgfx_nargs_2);
+ tgfx_base[2] = tgfx_probe(tgfx_3, tgfx_nargs_3);
if (tgfx_base[0] || tgfx_base[1] || tgfx_base[2])
return 0;
--- diff/drivers/input/keyboard/Kconfig 2003-10-09 08:47:34.000000000 +0000
+++ source/drivers/input/keyboard/Kconfig 2004-03-16 11:27:48.142868832 +0000
@@ -17,6 +17,7 @@ config KEYBOARD_ATKBD
depends on INPUT && INPUT_KEYBOARD
select SERIO
select SERIO_I8042 if PC
+ select SERIO_GSCPS2 if GSC
help
Say Y here if you want to use a standard AT or PS/2 keyboard. Usually
you'll need this, unless you have a different type keyboard (USB, ADB
@@ -40,6 +41,19 @@ config KEYBOARD_SUNKBD
To compile this driver as a module, choose M here: the
module will be called sunkbd.
+config KEYBOARD_LKKBD
+ tristate "DECstation/VAXstation LK201/LK401 keyboard support"
+ depends on INPUT && INPUT_KEYBOARD
+ select SERIO
+ help
+ Say Y here if you want to use a LK201 or LK401 style serial
+ keyboard. This keyboard is also useable on PCs if you attach
+ it with the inputattach program. The connector pinout is
+ described within lkkbd.c.
+
+ To compile this driver as a module, choose M here: the
+ module will be called lkkbd.
+
config KEYBOARD_XTKBD
tristate "XT Keyboard support"
depends on INPUT && INPUT_KEYBOARD
--- diff/drivers/input/keyboard/Makefile 2003-05-21 10:49:55.000000000 +0000
+++ source/drivers/input/keyboard/Makefile 2004-03-16 11:27:48.144868528 +0000
@@ -7,6 +7,7 @@
obj-$(CONFIG_KEYBOARD_ATKBD) += atkbd.o
obj-$(CONFIG_KEYBOARD_MAPLE) += maple_keyb.o
obj-$(CONFIG_KEYBOARD_SUNKBD) += sunkbd.o
+obj-$(CONFIG_KEYBOARD_LKKBD) += lkkbd.o
obj-$(CONFIG_KEYBOARD_XTKBD) += xtkbd.o
obj-$(CONFIG_KEYBOARD_AMIGA) += amikbd.o
obj-$(CONFIG_KEYBOARD_NEWTON) += newtonkbd.o
--- diff/drivers/input/keyboard/atkbd.c 2004-02-09 10:36:10.000000000 +0000
+++ source/drivers/input/keyboard/atkbd.c 2004-03-16 11:27:48.140869136 +0000
@@ -30,21 +30,17 @@
MODULE_AUTHOR("Vojtech Pavlik <vojtech@suse.cz>");
MODULE_DESCRIPTION("AT and PS/2 keyboard driver");
-MODULE_PARM(atkbd_set, "1i");
-MODULE_PARM(atkbd_reset, "1i");
-MODULE_PARM(atkbd_softrepeat, "1i");
MODULE_LICENSE("GPL");
static int atkbd_set = 2;
module_param_named(set, atkbd_set, int, 0);
-MODULE_PARM_DESC(set, "Select keyboard code set (2 = default, 3, 4)");
+MODULE_PARM_DESC(set, "Select keyboard code set (2 = default, 3 = PS/2 native)");
+
#if defined(__i386__) || defined(__x86_64__) || defined(__hppa__)
static int atkbd_reset;
#else
static int atkbd_reset = 1;
#endif
-static int atkbd_softrepeat;
-
module_param_named(reset, atkbd_reset, bool, 0);
MODULE_PARM_DESC(reset, "Reset keyboard during initialization");
@@ -52,6 +48,18 @@ static int atkbd_softrepeat;
module_param_named(softrepeat, atkbd_softrepeat, bool, 0);
MODULE_PARM_DESC(softrepeat, "Use software keyboard repeat");
+static int atkbd_scroll;
+module_param_named(scroll, atkbd_scroll, bool, 0);
+MODULE_PARM_DESC(scroll, "Enable scroll-wheel on MS Office and similar keyboards");
+
+static int atkbd_extra;
+module_param_named(extra, atkbd_extra, bool, 0);
+MODULE_PARM_DESC(extra, "Enable extra LEDs and keys on IBM RapidAcces, EzKey and similar keyboards");
+
+__obsolete_setup("atkbd_set=");
+__obsolete_setup("atkbd_reset");
+__obsolete_setup("atkbd_softrepeat=");
+
/*
* Scancode to keycode tables. These are just the default setting, and
* are loadable via an userland utility.
@@ -127,11 +135,11 @@ static unsigned char atkbd_unxlate_table
#define ATKBD_CMD_EX_SETLEDS 0x20eb
#define ATKBD_CMD_OK_GETID 0x02e8
+
#define ATKBD_RET_ACK 0xfa
#define ATKBD_RET_NAK 0xfe
#define ATKBD_RET_BAT 0xaa
#define ATKBD_RET_EMUL0 0xe0
-#define ATKBD_RET_EMULX 0x80
#define ATKBD_RET_EMUL1 0xe1
#define ATKBD_RET_RELEASE 0xf0
#define ATKBD_RET_HANGUEL 0xf1
@@ -141,6 +149,22 @@ static unsigned char atkbd_unxlate_table
#define ATKBD_KEY_UNKNOWN 0
#define ATKBD_KEY_NULL 255
+#define ATKBD_SCR_1 254
+#define ATKBD_SCR_2 253
+#define ATKBD_SCR_4 252
+#define ATKBD_SCR_8 251
+#define ATKBD_SCR_CLICK 250
+
+#define ATKBD_SPECIAL 250
+
+static unsigned char atkbd_scroll_keys[5][2] = {
+ { ATKBD_SCR_1, 0x45 },
+ { ATKBD_SCR_2, 0x29 },
+ { ATKBD_SCR_4, 0x36 },
+ { ATKBD_SCR_8, 0x27 },
+ { ATKBD_SCR_CLICK, 0x60 },
+};
+
/*
* The atkbd control structure
*/
@@ -155,6 +179,7 @@ struct atkbd {
unsigned char cmdbuf[4];
unsigned char cmdcnt;
unsigned char set;
+ unsigned char extra;
unsigned char release;
int lastkey;
volatile signed char ack;
@@ -189,6 +214,7 @@ static irqreturn_t atkbd_interrupt(struc
{
struct atkbd *atkbd = serio->private;
unsigned int code = data;
+ int scroll = 0, click = -1;
int value;
#ifdef ATKBD_DEBUG
@@ -202,7 +228,7 @@ static irqreturn_t atkbd_interrupt(struc
atkbd->resend = 1;
goto out;
}
-
+
if (!flags && data == ATKBD_RET_ACK)
atkbd->resend = 0;
#endif
@@ -276,7 +302,7 @@ static irqreturn_t atkbd_interrupt(struc
case ATKBD_KEY_UNKNOWN:
printk(KERN_WARNING "atkbd.c: Unknown key %s (%s set %d, code %#x on %s).\n",
atkbd->release ? "released" : "pressed",
- atkbd->translated ? "translated" : "raw",
+ atkbd->translated ? "translated" : "raw",
atkbd->set, code, serio->phys);
if (atkbd->translated && atkbd->set == 2 && code == 0x7a)
printk(KERN_WARNING "atkbd.c: This is an XFree86 bug. It shouldn't access"
@@ -284,6 +310,21 @@ static irqreturn_t atkbd_interrupt(struc
else
printk(KERN_WARNING "atkbd.c: Use 'setkeycodes %s%02x <keycode>' to make it known.\n", code & 0x80 ? "e0" : "", code & 0x7f);
break;
+ case ATKBD_SCR_1:
+ scroll = 1 - atkbd->release * 2;
+ break;
+ case ATKBD_SCR_2:
+ scroll = 2 - atkbd->release * 4;
+ break;
+ case ATKBD_SCR_4:
+ scroll = 4 - atkbd->release * 8;
+ break;
+ case ATKBD_SCR_8:
+ scroll = 8 - atkbd->release * 16;
+ break;
+ case ATKBD_SCR_CLICK:
+ click = !atkbd->release;
+ break;
default:
value = atkbd->release ? 0 :
(1 + (!atkbd_softrepeat && test_bit(atkbd->keycode[code], atkbd->dev.key)));
@@ -305,6 +346,13 @@ static irqreturn_t atkbd_interrupt(struc
atkbd_report_key(&atkbd->dev, regs, atkbd->keycode[code], value);
}
+ if (scroll || click != -1) {
+ input_regs(&atkbd->dev, regs);
+ input_report_key(&atkbd->dev, BTN_MIDDLE, click);
+ input_report_rel(&atkbd->dev, REL_WHEEL, scroll);
+ input_sync(&atkbd->dev);
+ }
+
atkbd->release = 0;
out:
return IRQ_HANDLED;
@@ -353,7 +401,7 @@ static int atkbd_command(struct atkbd *a
if (receive && param)
for (i = 0; i < receive; i++)
atkbd->cmdbuf[(receive - 1) - i] = param[i];
-
+
if (command & 0xff)
if (atkbd_sendbyte(atkbd, command & 0xff))
return (atkbd->cmdcnt = 0) - 1;
@@ -373,7 +421,7 @@ static int atkbd_command(struct atkbd *a
atkbd->cmdcnt = 0;
break;
}
-
+
udelay(1);
}
@@ -420,7 +468,7 @@ static int atkbd_event(struct input_dev
| (test_bit(LED_CAPSL, dev->led) ? 4 : 0);
atkbd_command(atkbd, param, ATKBD_CMD_SETLEDS);
- if (atkbd->set == 4) {
+ if (atkbd->extra) {
param[0] = 0;
param[1] = (test_bit(LED_COMPOSE, dev->led) ? 0x01 : 0)
| (test_bit(LED_SLEEP, dev->led) ? 0x02 : 0)
@@ -466,7 +514,7 @@ static int atkbd_probe(struct atkbd *atk
*/
if (atkbd_reset)
- if (atkbd_command(atkbd, NULL, ATKBD_CMD_RESET_BAT))
+ if (atkbd_command(atkbd, NULL, ATKBD_CMD_RESET_BAT))
printk(KERN_WARNING "atkbd.c: keyboard reset failed on %s\n", atkbd->serio->phys);
/*
@@ -529,20 +577,21 @@ static int atkbd_set_3(struct atkbd *atk
return 3;
}
- if (atkbd_set != 2)
- if (!atkbd_command(atkbd, param, ATKBD_CMD_OK_GETID)) {
- atkbd->id = param[0] << 8 | param[1];
+ if (atkbd_extra) {
+ param[0] = 0x71;
+ if (!atkbd_command(atkbd, param, ATKBD_CMD_EX_ENABLE)) {
+ atkbd->extra = 1;
return 2;
}
-
- if (atkbd_set == 4) {
- param[0] = 0x71;
- if (!atkbd_command(atkbd, param, ATKBD_CMD_EX_ENABLE))
- return 4;
}
- if (atkbd_set != 3)
+ if (atkbd_set != 3)
+ return 2;
+
+ if (!atkbd_command(atkbd, param, ATKBD_CMD_OK_GETID)) {
+ atkbd->id = param[0] << 8 | param[1];
return 2;
+ }
param[0] = 3;
if (atkbd_command(atkbd, param, ATKBD_CMD_SSCANSET))
@@ -637,7 +686,7 @@ static void atkbd_connect(struct serio *
switch (serio->type & SERIO_TYPE) {
- case SERIO_8042_XL:
+ case SERIO_8042_XL:
atkbd->translated = 1;
case SERIO_8042:
if (serio->write)
@@ -650,7 +699,7 @@ static void atkbd_connect(struct serio *
kfree(atkbd);
return;
}
-
+
if (atkbd->write) {
atkbd->dev.evbit[0] = BIT(EV_KEY) | BIT(EV_LED) | BIT(EV_REP);
atkbd->dev.ledbit[0] = BIT(LED_NUML) | BIT(LED_CAPSL) | BIT(LED_SCROLLL);
@@ -687,7 +736,7 @@ static void atkbd_connect(struct serio *
kfree(atkbd);
return;
}
-
+
atkbd->set = atkbd_set_3(atkbd);
atkbd_enable(atkbd);
@@ -696,24 +745,32 @@ static void atkbd_connect(struct serio *
atkbd->id = 0xab00;
}
- if (atkbd->set == 4) {
+ if (atkbd->extra) {
atkbd->dev.ledbit[0] |= BIT(LED_COMPOSE) | BIT(LED_SUSPEND) | BIT(LED_SLEEP) | BIT(LED_MUTE) | BIT(LED_MISC);
- sprintf(atkbd->name, "AT Set 2 Extended keyboard");
+ sprintf(atkbd->name, "AT Set 2 Extra keyboard");
} else
sprintf(atkbd->name, "AT %s Set %d keyboard",
atkbd->translated ? "Translated" : "Raw", atkbd->set);
sprintf(atkbd->phys, "%s/input0", serio->phys);
+ if (atkbd_scroll) {
+ for (i = 0; i < 5; i++)
+ atkbd_set2_keycode[atkbd_scroll_keys[i][1]] = atkbd_scroll_keys[i][0];
+ atkbd->dev.evbit[0] |= BIT(EV_REL);
+ atkbd->dev.relbit[0] = BIT(REL_WHEEL);
+ set_bit(BTN_MIDDLE, atkbd->dev.keybit);
+ }
+
if (atkbd->translated) {
for (i = 0; i < 128; i++) {
atkbd->keycode[i] = atkbd_set2_keycode[atkbd_unxlate_table[i]];
atkbd->keycode[i | 0x80] = atkbd_set2_keycode[atkbd_unxlate_table[i] | 0x80];
}
- } else if (atkbd->set == 2) {
- memcpy(atkbd->keycode, atkbd_set2_keycode, sizeof(atkbd->keycode));
- } else {
+ } else if (atkbd->set == 3) {
memcpy(atkbd->keycode, atkbd_set3_keycode, sizeof(atkbd->keycode));
+ } else {
+ memcpy(atkbd->keycode, atkbd_set2_keycode, sizeof(atkbd->keycode));
}
atkbd->dev.name = atkbd->name;
@@ -724,7 +781,7 @@ static void atkbd_connect(struct serio *
atkbd->dev.id.version = atkbd->id;
for (i = 0; i < 512; i++)
- if (atkbd->keycode[i] && atkbd->keycode[i] < 255)
+ if (atkbd->keycode[i] && atkbd->keycode[i] < ATKBD_SPECIAL)
set_bit(atkbd->keycode[i], atkbd->dev.keybit);
input_register_device(&atkbd->dev);
@@ -741,46 +798,29 @@ static int atkbd_reconnect(struct serio
{
struct atkbd *atkbd = serio->private;
struct serio_dev *dev = serio->dev;
- int i;
+ unsigned char param[1];
- if (!dev) {
- printk(KERN_DEBUG "atkbd: reconnect request, but serio is disconnected, ignoring...\n");
- return -1;
- }
+ if (!dev) {
+ printk(KERN_DEBUG "atkbd: reconnect request, but serio is disconnected, ignoring...\n");
+ return -1;
+ }
if (atkbd->write) {
+ param[0] = (test_bit(LED_SCROLLL, atkbd->dev.led) ? 1 : 0)
+ | (test_bit(LED_NUML, atkbd->dev.led) ? 2 : 0)
+ | (test_bit(LED_CAPSL, atkbd->dev.led) ? 4 : 0);
+
if (atkbd_probe(atkbd))
return -1;
-
- atkbd->set = atkbd_set_3(atkbd);
+ if (atkbd->set != atkbd_set_3(atkbd))
+ return -1;
+
atkbd_enable(atkbd);
- } else {
- atkbd->set = 2;
- atkbd->id = 0xab00;
- }
- /*
- * Here we probably should check if the keyboard has the same set that
- * it had before and bail out if it's different. But this will most likely
- * cause new keyboard device be created... and for the user it will look
- * like keyboard is lost
- */
-
- if (atkbd->translated) {
- for (i = 0; i < 128; i++) {
- atkbd->keycode[i] = atkbd_set2_keycode[atkbd_unxlate_table[i]];
- atkbd->keycode[i | 0x80] = atkbd_set2_keycode[atkbd_unxlate_table[i] | 0x80];
- }
- } else if (atkbd->set == 2) {
- memcpy(atkbd->keycode, atkbd_set2_keycode, sizeof(atkbd->keycode));
- } else {
- memcpy(atkbd->keycode, atkbd_set3_keycode, sizeof(atkbd->keycode));
+ if (atkbd_command(atkbd, param, ATKBD_CMD_SETLEDS))
+ return -1;
}
- for (i = 0; i < 512; i++)
- if (atkbd->keycode[i] && atkbd->keycode[i] < 255)
- set_bit(atkbd->keycode[i], atkbd->dev.keybit);
-
return 0;
}
--- diff/drivers/input/keyboard/hpps2atkbd.h 2004-02-09 10:36:10.000000000 +0000
+++ source/drivers/input/keyboard/hpps2atkbd.h 2004-03-16 11:27:48.141868984 +0000
@@ -4,14 +4,9 @@
* Copyright (c) 2004 Helge Deller <deller@gmx.de>
* Copyright (c) 2002 Laurent Canet <canetl@esiee.fr>
* Copyright (c) 2002 Thibaut Varene <varenet@esiee.fr>
+ * Copyright (c) 2000 Xavier Debacker <debackex@esiee.fr>
*
- * based on linux-2.4's hp_mouse.c & hp_keyb.c
- * Copyright (c) 1999 Alex deVries <adevries@thepuffingroup.com>
- * Copyright (c) 1999-2000 Philipp Rumpf <prumpf@tux.org>
- * Copyright (c) 2000 Xavier Debacker <debackex@esiee.fr>
- * Copyright (c) 2000-2001 Thomas Marteau <marteaut@esiee.fr>
- *
- * HP PS/2 AT-compatible Keyboard, found in PA/RISC Workstations
+ * HP PS/2 AT-compatible Keyboard, found in PA/RISC Workstations & Laptops
*
* This file is subject to the terms and conditions of the GNU General Public
* License. See the file "COPYING" in the main directory of this archive
@@ -19,87 +14,100 @@
*/
-#define KBD_UNKNOWN 0
-
-/* Raw SET 2 scancode table */
+/* undefine if you have a RDI PRECISIONBOOK */
+#define STANDARD_KEYBOARD
-#if 0
- /* conflicting keys between a RDI Precisionbook keyboard and a normal HP keyboard */
- keytable[0x07] = KEY_F1; /* KEY_F12 */
- keytable[0x11] = KEY_LEFTCTRL; /* KEY_LEFTALT */
- keytable[0x14] = KEY_CAPSLOCK; /* KEY_LEFTCTRL */
- keytable[0x61] = KEY_LEFT; /* KEY_102ND */
+#if defined(STANDARD_KEYBOARD)
+# define CONFLICT(x,y) x
+#else
+# define CONFLICT(x,y) y
#endif
+/* sadly RDI (Tadpole) decided to ship a different keyboard layout
+ than HP for their PS/2 laptop keyboard which leads to conflicting
+ keycodes between a normal HP PS/2 keyboard and a RDI Precisionbook.
+ HP: RDI: */
+#define C_07 CONFLICT( KEY_F12, KEY_F1 )
+#define C_11 CONFLICT( KEY_LEFTALT, KEY_LEFTCTRL )
+#define C_14 CONFLICT( KEY_LEFTCTRL, KEY_CAPSLOCK )
+#define C_58 CONFLICT( KEY_CAPSLOCK, KEY_RIGHTCTRL )
+#define C_61 CONFLICT( KEY_102ND, KEY_LEFT )
-static unsigned char atkbd_set2_keycode[512] = {
+/* Raw SET 2 scancode table */
- /* 00 */ KBD_UNKNOWN, KEY_F9, KBD_UNKNOWN, KEY_F5, KEY_F3, KEY_F1, KEY_F2, KEY_F1,
- /* 08 */ KEY_ESC, KEY_F10, KEY_F8, KEY_F6, KEY_F4, KEY_TAB, KEY_GRAVE, KEY_F2,
- /* 10 */ KBD_UNKNOWN, KEY_LEFTCTRL, KEY_LEFTSHIFT, KBD_UNKNOWN, KEY_CAPSLOCK, KEY_Q, KEY_1, KEY_F3,
- /* 18 */ KBD_UNKNOWN, KEY_LEFTALT, KEY_Z, KEY_S, KEY_A, KEY_W, KEY_2, KEY_F4,
- /* 20 */ KBD_UNKNOWN, KEY_C, KEY_X, KEY_D, KEY_E, KEY_4, KEY_3, KEY_F5,
- /* 28 */ KBD_UNKNOWN, KEY_SPACE, KEY_V, KEY_F, KEY_T, KEY_R, KEY_5, KEY_F6,
- /* 30 */ KBD_UNKNOWN, KEY_N, KEY_B, KEY_H, KEY_G, KEY_Y, KEY_6, KEY_F7,
- /* 38 */ KBD_UNKNOWN, KEY_RIGHTALT, KEY_M, KEY_J, KEY_U, KEY_7, KEY_8, KEY_F8,
- /* 40 */ KBD_UNKNOWN, KEY_COMMA, KEY_K, KEY_I, KEY_O, KEY_0, KEY_9, KEY_F9,
- /* 48 */ KBD_UNKNOWN, KEY_DOT, KEY_SLASH, KEY_L, KEY_SEMICOLON, KEY_P, KEY_MINUS, KEY_F10,
- /* 50 */ KBD_UNKNOWN, KBD_UNKNOWN, KEY_APOSTROPHE,KBD_UNKNOWN, KEY_LEFTBRACE, KEY_EQUAL, KEY_F11, KEY_SYSRQ,
- /* 58 */ KEY_CAPSLOCK, KEY_RIGHTSHIFT,KEY_ENTER, KEY_RIGHTBRACE,KEY_BACKSLASH, KEY_BACKSLASH,KEY_F12, KEY_SCROLLLOCK,
- /* 60 */ KEY_DOWN, KEY_LEFT, KEY_PAUSE, KEY_UP, KEY_DELETE, KEY_END, KEY_BACKSPACE, KEY_INSERT,
- /* 68 */ KBD_UNKNOWN, KEY_KP1, KEY_RIGHT, KEY_KP4, KEY_KP7, KEY_PAGEDOWN, KEY_HOME, KEY_PAGEUP,
- /* 70 */ KEY_KP0, KEY_KPDOT, KEY_KP2, KEY_KP5, KEY_KP6, KEY_KP8, KEY_ESC, KEY_NUMLOCK,
- /* 78 */ KEY_F11, KEY_KPPLUS, KEY_KP3, KEY_KPMINUS, KEY_KPASTERISK,KEY_KP9, KEY_SCROLLLOCK,KEY_103RD,
- /* 80 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 88 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 90 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 98 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* a0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* a8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* b0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* b8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* c0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* c8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* d0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* d8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* e0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* e8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* f0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* f8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
-
- /* These are offset for escaped keycodes: */
-
- /* 00 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KEY_F7, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 08 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KEY_LEFTMETA, KEY_RIGHTMETA, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 10 */ KBD_UNKNOWN, KEY_RIGHTALT, KBD_UNKNOWN, KBD_UNKNOWN, KEY_RIGHTCTRL, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 18 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 20 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 28 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 30 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 38 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 40 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 48 */ KBD_UNKNOWN, KBD_UNKNOWN, KEY_KPSLASH, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 50 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 58 */ KBD_UNKNOWN, KBD_UNKNOWN, KEY_KPENTER, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 60 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 68 */ KBD_UNKNOWN, KEY_END, KBD_UNKNOWN, KEY_LEFT, KEY_HOME, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 70 */ KEY_INSERT, KEY_DELETE, KEY_DOWN, KBD_UNKNOWN, KEY_RIGHT, KEY_UP, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 78 */ KBD_UNKNOWN, KBD_UNKNOWN, KEY_PAGEDOWN, KBD_UNKNOWN, KEY_SYSRQ, KEY_PAGEUP, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 80 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 88 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 90 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 98 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* a0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* a8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* b0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* b8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* c0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* c8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* d0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* d8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* e0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* e8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* f0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* f8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN
+/* 00 */ KEY_RESERVED, KEY_F9, KEY_RESERVED, KEY_F5, KEY_F3, KEY_F1, KEY_F2, C_07,
+/* 08 */ KEY_ESC, KEY_F10, KEY_F8, KEY_F6, KEY_F4, KEY_TAB, KEY_GRAVE, KEY_F2,
+/* 10 */ KEY_RESERVED, C_11, KEY_LEFTSHIFT, KEY_RESERVED, C_14, KEY_Q, KEY_1, KEY_F3,
+/* 18 */ KEY_RESERVED, KEY_LEFTALT, KEY_Z, KEY_S, KEY_A, KEY_W, KEY_2, KEY_F4,
+/* 20 */ KEY_RESERVED, KEY_C, KEY_X, KEY_D, KEY_E, KEY_4, KEY_3, KEY_F5,
+/* 28 */ KEY_RESERVED, KEY_SPACE, KEY_V, KEY_F, KEY_T, KEY_R, KEY_5, KEY_F6,
+/* 30 */ KEY_RESERVED, KEY_N, KEY_B, KEY_H, KEY_G, KEY_Y, KEY_6, KEY_F7,
+/* 38 */ KEY_RESERVED, KEY_RIGHTALT, KEY_M, KEY_J, KEY_U, KEY_7, KEY_8, KEY_F8,
+/* 40 */ KEY_RESERVED, KEY_COMMA, KEY_K, KEY_I, KEY_O, KEY_0, KEY_9, KEY_F9,
+/* 48 */ KEY_RESERVED, KEY_DOT, KEY_SLASH, KEY_L, KEY_SEMICOLON, KEY_P, KEY_MINUS, KEY_F10,
+/* 50 */ KEY_RESERVED, KEY_RESERVED, KEY_APOSTROPHE,KEY_RESERVED, KEY_LEFTBRACE, KEY_EQUAL, KEY_F11, KEY_SYSRQ,
+/* 58 */ C_58, KEY_RIGHTSHIFT,KEY_ENTER, KEY_RIGHTBRACE,KEY_BACKSLASH, KEY_BACKSLASH,KEY_F12, KEY_SCROLLLOCK,
+/* 60 */ KEY_DOWN, C_61, KEY_PAUSE, KEY_UP, KEY_DELETE, KEY_END, KEY_BACKSPACE, KEY_INSERT,
+/* 68 */ KEY_RESERVED, KEY_KP1, KEY_RIGHT, KEY_KP4, KEY_KP7, KEY_PAGEDOWN, KEY_HOME, KEY_PAGEUP,
+/* 70 */ KEY_KP0, KEY_KPDOT, KEY_KP2, KEY_KP5, KEY_KP6, KEY_KP8, KEY_ESC, KEY_NUMLOCK,
+/* 78 */ KEY_F11, KEY_KPPLUS, KEY_KP3, KEY_KPMINUS, KEY_KPASTERISK,KEY_KP9, KEY_SCROLLLOCK,KEY_103RD,
+/* 80 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 88 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 90 */ KEY_RESERVED, KEY_RIGHTALT, KEY_SYSRQ, KEY_RESERVED, KEY_RIGHTCTRL, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 98 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_CAPSLOCK, KEY_RESERVED, KEY_LEFTMETA,
+/* a0 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RIGHTMETA,
+/* a8 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_COMPOSE,
+/* b0 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* b8 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* c0 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* c8 */ KEY_RESERVED, KEY_RESERVED, KEY_KPSLASH, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* d0 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* d8 */ KEY_RESERVED, KEY_RESERVED, KEY_KPENTER, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* e0 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* e8 */ KEY_RESERVED, KEY_END, KEY_RESERVED, KEY_LEFT, KEY_HOME, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* f0 */ KEY_INSERT, KEY_DELETE, KEY_DOWN, KEY_RESERVED, KEY_RIGHT, KEY_UP, KEY_RESERVED, KEY_PAUSE,
+/* f8 */ KEY_RESERVED, KEY_RESERVED, KEY_PAGEDOWN, KEY_RESERVED, KEY_SYSRQ, KEY_PAGEUP, KEY_RESERVED, KEY_RESERVED,
+
+/* These are offset for escaped keycodes: */
+
+/* 00 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_F7, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 08 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_LEFTMETA, KEY_RIGHTMETA, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 10 */ KEY_RESERVED, KEY_RIGHTALT, KEY_RESERVED, KEY_RESERVED, KEY_RIGHTCTRL, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 18 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 20 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 28 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 30 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 38 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 40 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 48 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 50 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 58 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 60 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 68 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 70 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 78 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 80 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 88 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 90 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* 98 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* a0 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* a8 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* b0 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* b8 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* c0 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* c8 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* d0 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* d8 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* e0 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* e8 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* f0 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED,
+/* f8 */ KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED, KEY_RESERVED
+
+#undef STANDARD_KEYBOARD
+#undef CONFLICT
+#undef C_07
+#undef C_11
+#undef C_14
+#undef C_58
+#undef C_61
-};
--- diff/drivers/input/keyboard/sunkbd.c 2003-08-20 13:16:09.000000000 +0000
+++ source/drivers/input/keyboard/sunkbd.c 2004-03-16 11:27:48.144868528 +0000
@@ -77,6 +77,7 @@ struct sunkbd {
struct input_dev dev;
struct serio *serio;
struct work_struct tq;
+ wait_queue_head_t wait;
char name[64];
char phys[32];
char type;
@@ -96,11 +97,13 @@ static irqreturn_t sunkbd_interrupt(stru
if (sunkbd->reset <= -1) { /* If cp[i] is 0xff, sunkbd->reset will stay -1. */
sunkbd->reset = data; /* The keyboard sends 0xff 0xff 0xID on powerup */
+ wake_up_interruptible(&sunkbd->wait);
goto out;
}
if (sunkbd->layout == -1) {
sunkbd->layout = data;
+ wake_up_interruptible(&sunkbd->wait);
goto out;
}
@@ -176,22 +179,19 @@ static int sunkbd_event(struct input_dev
static int sunkbd_initialize(struct sunkbd *sunkbd)
{
- int t;
-
- t = 1000;
sunkbd->reset = -2;
sunkbd->serio->write(sunkbd->serio, SUNKBD_CMD_RESET);
- while (sunkbd->reset < 0 && --t) mdelay(1);
- if (!t) return -1;
+ wait_event_interruptible_timeout(sunkbd->wait, sunkbd->reset >= 0, HZ);
+ if (sunkbd->reset <0)
+ return -1;
sunkbd->type = sunkbd->reset;
if (sunkbd->type == 4) { /* Type 4 keyboard */
- t = 250;
sunkbd->layout = -2;
sunkbd->serio->write(sunkbd->serio, SUNKBD_CMD_LAYOUT);
- while (sunkbd->layout < 0 && --t) mdelay(1);
- if (!t) return -1;
+ wait_event_interruptible_timeout(sunkbd->wait, sunkbd->layout >= 0, HZ/4);
+ if (sunkbd->layout < 0) return -1;
if (sunkbd->layout & SUNKBD_LAYOUT_5_MASK) sunkbd->type = 5;
}
@@ -206,9 +206,8 @@ static int sunkbd_initialize(struct sunk
static void sunkbd_reinit(void *data)
{
struct sunkbd *sunkbd = data;
- int t = 1000;
- while (sunkbd->reset < 0 && --t) mdelay(1);
+ wait_event_interruptible_timeout(sunkbd->wait, sunkbd->reset >= 0, HZ);
sunkbd->serio->write(sunkbd->serio, SUNKBD_CMD_SETLED);
sunkbd->serio->write(sunkbd->serio,
@@ -239,6 +238,7 @@ static void sunkbd_connect(struct serio
memset(sunkbd, 0, sizeof(struct sunkbd));
init_input_dev(&sunkbd->dev);
+ init_waitqueue_head(&sunkbd->wait);
sunkbd->dev.evbit[0] = BIT(EV_KEY) | BIT(EV_LED) | BIT(EV_SND) | BIT(EV_REP);
sunkbd->dev.ledbit[0] = BIT(LED_CAPSL) | BIT(LED_COMPOSE) | BIT(LED_SCROLLL) | BIT(LED_NUML);
@@ -275,7 +275,7 @@ static void sunkbd_connect(struct serio
set_bit(sunkbd->keycode[i], sunkbd->dev.keybit);
clear_bit(0, sunkbd->dev.keybit);
- sprintf(sunkbd->name, "%s/input", serio->phys);
+ sprintf(sunkbd->phys, "%s/input0", serio->phys);
sunkbd->dev.name = sunkbd->name;
sunkbd->dev.phys = sunkbd->phys;
--- diff/drivers/input/misc/Kconfig 2003-09-30 14:46:14.000000000 +0000
+++ source/drivers/input/misc/Kconfig 2004-03-16 11:27:48.148867920 +0000
@@ -54,12 +54,3 @@ config INPUT_UINPUT
To compile this driver as a module, choose M here: the
module will be called uinput.
-config INPUT_GSC
- tristate "PA-RISC GSC PS/2 keyboard/mouse support"
- depends on GSC && INPUT && INPUT_MISC
- help
- Say Y here if you have a PS/2 keyboard and/or mouse attached
- to your PA-RISC box. HP run the keyboard in AT mode rather than
- XT mode like everyone else, so we need our own driver.
- Furthermore, the GSC PS/2 controller shares IRQ between mouse and
- keyboard.
--- diff/drivers/input/misc/Makefile 2003-02-26 16:00:54.000000000 +0000
+++ source/drivers/input/misc/Makefile 2004-03-16 11:27:48.149867768 +0000
@@ -9,4 +9,3 @@ obj-$(CONFIG_INPUT_PCSPKR) += pcspkr.o
obj-$(CONFIG_INPUT_M68K_BEEP) += m68kspkr.o
obj-$(CONFIG_INPUT_98SPKR) += 98spkr.o
obj-$(CONFIG_INPUT_UINPUT) += uinput.o
-obj-$(CONFIG_INPUT_GSC) += gsc_ps2.o
--- diff/drivers/input/mouse/98busmouse.c 2004-02-09 10:36:10.000000000 +0000
+++ source/drivers/input/mouse/98busmouse.c 2004-03-16 11:27:48.149867768 +0000
@@ -74,6 +74,8 @@ static int pc98bm_irq = PC98BM_IRQ;
module_param_named(irq, pc98bm_irq, uint, 0);
MODULE_PARM_DESC(irq, "IRQ number (13=default)");
+__obsolete_setup("pc98bm_irq=");
+
static int pc98bm_used = 0;
static irqreturn_t pc98bm_interrupt(int irq, void *dev_id, struct pt_regs *regs);
--- diff/drivers/input/mouse/Kconfig 2004-02-09 10:36:10.000000000 +0000
+++ source/drivers/input/mouse/Kconfig 2004-03-16 11:27:48.150867616 +0000
@@ -17,6 +17,7 @@ config MOUSE_PS2
depends on INPUT && INPUT_MOUSE
select SERIO
select SERIO_I8042 if PC
+ select SERIO_GSCPS2 if GSC
---help---
Say Y here if you have a PS/2 mouse connected to your system. This
includes the standard 2 or 3-button PS/2 mouse, as well as PS/2
@@ -117,6 +118,19 @@ config MOUSE_RISCPC
To compile this driver as a module, choose M here: the
module will be called rpcmouse.
+config MOUSE_VSXXXAA
+ tristate "DEC VSXXX-AA/GA mouse and tablet"
+ depends on INPUT && INPUT_MOUSE
+ select SERIO
+ help
+ Say Y (or M) if you want to use a DEC VSXXX-AA (hockey
+ puck) or a VSXXX-GA (rectangular) mouse. Theses mice are
+ typically used on DECstations or VAXstations, but can also
+ be used on any box capable of RS232 (with some adaptor
+ described in the source file). This driver should, in theory,
+ also work with the digitizer DEC produced, but it isn't tested
+ with that (I don't have the hardware yet).
+
config MOUSE_PC9800
tristate "NEC PC-9800 busmouse"
depends on X86_PC9800 && INPUT && INPUT_MOUSE && ISA
--- diff/drivers/input/mouse/Makefile 2003-06-30 09:07:33.000000000 +0000
+++ source/drivers/input/mouse/Makefile 2004-03-16 11:27:48.151867464 +0000
@@ -13,5 +13,6 @@ obj-$(CONFIG_MOUSE_PC110PAD) += pc110pad
obj-$(CONFIG_MOUSE_PC9800) += 98busmouse.o
obj-$(CONFIG_MOUSE_PS2) += psmouse.o
obj-$(CONFIG_MOUSE_SERIAL) += sermouse.o
+obj-$(CONFIG_MOUSE_VSXXXAA) += vsxxxaa.o
psmouse-objs := psmouse-base.o logips2pp.o synaptics.o
--- diff/drivers/input/mouse/inport.c 2004-02-09 10:36:10.000000000 +0000
+++ source/drivers/input/mouse/inport.c 2004-03-16 11:27:48.150867616 +0000
@@ -85,6 +85,8 @@ static int inport_irq = INPORT_IRQ;
module_param_named(irq, inport_irq, uint, 0);
MODULE_PARM_DESC(irq, "IRQ number (5=default)");
+__obsolete_setup("inport_irq=");
+
static int inport_used;
static irqreturn_t inport_interrupt(int irq, void *dev_id, struct pt_regs *regs);
--- diff/drivers/input/mouse/logibm.c 2004-02-09 10:36:10.000000000 +0000
+++ source/drivers/input/mouse/logibm.c 2004-03-16 11:27:48.151867464 +0000
@@ -75,6 +75,8 @@ static int logibm_irq = LOGIBM_IRQ;
module_param_named(irq, logibm_irq, uint, 0);
MODULE_PARM_DESC(irq, "IRQ number (5=default)");
+__obsolete_setup("logibm_irq=");
+
static int logibm_used = 0;
static irqreturn_t logibm_interrupt(int irq, void *dev_id, struct pt_regs *regs);
--- diff/drivers/input/mouse/psmouse-base.c 2004-02-09 10:36:10.000000000 +0000
+++ source/drivers/input/mouse/psmouse-base.c 2004-03-16 11:27:48.152867312 +0000
@@ -47,6 +47,12 @@ unsigned int psmouse_resetafter;
module_param_named(resetafter, psmouse_resetafter, uint, 0);
MODULE_PARM_DESC(resetafter, "Reset Synaptics Touchpad after so many bad packets (0 = never).");
+__obsolete_setup("psmouse_noext");
+__obsolete_setup("psmouse_resolution=");
+__obsolete_setup("psmouse_smartscroll=");
+__obsolete_setup("psmouse_resetafter=");
+__obsolete_setup("psmouse_rate=");
+
static char *psmouse_protocols[] = { "None", "PS/2", "PS2++", "PS2T++", "GenPS/2", "ImPS/2", "ImExPS/2", "SynPS/2"};
/*
@@ -163,14 +169,14 @@ static irqreturn_t psmouse_interrupt(str
psmouse->name, psmouse->phys, psmouse->pktcnt);
psmouse->pktcnt = 0;
}
-
+
psmouse->last = jiffies;
psmouse->packet[psmouse->pktcnt++] = data;
if (psmouse->packet[0] == PSMOUSE_RET_BAT) {
if (psmouse->pktcnt == 1)
goto out;
-
+
if (psmouse->pktcnt == 2) {
if (psmouse->packet[1] == PSMOUSE_RET_ID) {
psmouse->state = PSMOUSE_IGNORE;
@@ -258,7 +264,7 @@ int psmouse_command(struct psmouse *psmo
return (psmouse->cmdcnt = 0) - 1;
while (psmouse->cmdcnt && timeout--) {
-
+
if (psmouse->cmdcnt == 1 && command == PSMOUSE_CMD_RESET_BAT &&
timeout > 100000) /* do not run in a endless loop */
timeout = 100000; /* 1 sec */
@@ -442,7 +448,7 @@ static int psmouse_probe(struct psmouse
*/
if (psmouse_command(psmouse, NULL, PSMOUSE_CMD_RESET_DIS))
- return -1;
+ printk(KERN_WARNING "psmouse.c: Failed to reset mouse on %s\n", psmouse->serio->phys);
/*
* And here we try to determine if it has any extensions over the
@@ -497,7 +503,7 @@ static void psmouse_set_rate(struct psmo
static void psmouse_initialize(struct psmouse *psmouse)
{
unsigned char param[2];
-
+
/*
* We set the mouse report rate, resolution and scaling.
*/
@@ -571,7 +577,7 @@ static void psmouse_disconnect(struct se
static void psmouse_connect(struct serio *serio, struct serio_dev *dev)
{
struct psmouse *psmouse;
-
+
if ((serio->type & SERIO_TYPE) != SERIO_8042 &&
(serio->type & SERIO_TYPE) != SERIO_PS_PSTHRU)
return;
@@ -603,7 +609,7 @@ static void psmouse_connect(struct serio
serio->private = NULL;
return;
}
-
+
sprintf(psmouse->devname, "%s %s %s",
psmouse_protocols[psmouse->type], psmouse->vendor, psmouse->name);
sprintf(psmouse->phys, "%s/input0",
@@ -617,7 +623,7 @@ static void psmouse_connect(struct serio
psmouse->dev.id.version = psmouse->model;
input_register_device(&psmouse->dev);
-
+
printk(KERN_INFO "input: %s on %s\n", psmouse->devname, serio->phys);
psmouse_initialize(psmouse);
@@ -637,13 +643,15 @@ static int psmouse_reconnect(struct seri
{
struct psmouse *psmouse = serio->private;
struct serio_dev *dev = serio->dev;
- int old_type = psmouse->type;
+ int old_type;
- if (!dev) {
+ if (!dev || !psmouse) {
printk(KERN_DEBUG "psmouse: reconnect request, but serio is disconnected, ignoring...\n");
return -1;
}
+ old_type = psmouse->type;
+
psmouse->state = PSMOUSE_NEW_DEVICE;
psmouse->type = psmouse->acking = psmouse->cmdcnt = psmouse->pktcnt = 0;
if (psmouse->reconnect) {
--- diff/drivers/input/mouse/synaptics.c 2004-02-09 10:36:10.000000000 +0000
+++ source/drivers/input/mouse/synaptics.c 2004-03-16 11:27:48.153867160 +0000
@@ -435,6 +435,8 @@ int synaptics_init(struct psmouse *psmou
goto init_fail;
}
+ priv->pkt_type = SYN_MODEL_NEWABS(priv->model_id) ? SYN_NEWABS : SYN_OLDABS;
+
if (SYN_CAP_EXTENDED(priv->capabilities) && SYN_CAP_PASS_THROUGH(priv->capabilities))
synaptics_pt_create(psmouse);
@@ -602,19 +604,42 @@ static void synaptics_process_packet(str
input_sync(dev);
}
-static int synaptics_validate_byte(struct psmouse *psmouse)
+static int synaptics_validate_byte(unsigned char packet[], int idx, unsigned char pkt_type)
{
- static unsigned char newabs_mask[] = { 0xC0, 0x00, 0x00, 0xC0, 0x00 };
- static unsigned char newabs_rslt[] = { 0x80, 0x00, 0x00, 0xC0, 0x00 };
- static unsigned char oldabs_mask[] = { 0xC0, 0x60, 0x00, 0xC0, 0x60 };
- static unsigned char oldabs_rslt[] = { 0xC0, 0x00, 0x00, 0x80, 0x00 };
- struct synaptics_data *priv = psmouse->private;
- int idx = psmouse->pktcnt - 1;
+ static unsigned char newabs_mask[] = { 0xC8, 0x00, 0x00, 0xC8, 0x00 };
+ static unsigned char newabs_rel_mask[] = { 0xC0, 0x00, 0x00, 0xC0, 0x00 };
+ static unsigned char newabs_rslt[] = { 0x80, 0x00, 0x00, 0xC0, 0x00 };
+ static unsigned char oldabs_mask[] = { 0xC0, 0x60, 0x00, 0xC0, 0x60 };
+ static unsigned char oldabs_rslt[] = { 0xC0, 0x00, 0x00, 0x80, 0x00 };
+
+ switch (pkt_type) {
+ case SYN_NEWABS:
+ case SYN_NEWABS_RELAXED:
+ return (packet[idx] & newabs_rel_mask[idx]) == newabs_rslt[idx];
+
+ case SYN_NEWABS_STRICT:
+ return (packet[idx] & newabs_mask[idx]) == newabs_rslt[idx];
+
+ case SYN_OLDABS:
+ return (packet[idx] & oldabs_mask[idx]) == oldabs_rslt[idx];
- if (SYN_MODEL_NEWABS(priv->model_id))
- return (psmouse->packet[idx] & newabs_mask[idx]) == newabs_rslt[idx];
- else
- return (psmouse->packet[idx] & oldabs_mask[idx]) == oldabs_rslt[idx];
+ default:
+ printk(KERN_ERR "synaptics: unknown packet type %d\n", pkt_type);
+ return 0;
+ }
+}
+
+static unsigned char synaptics_detect_pkt_type(struct psmouse *psmouse)
+{
+ int i;
+
+ for (i = 0; i < 5; i++)
+ if (!synaptics_validate_byte(psmouse->packet, i, SYN_NEWABS_STRICT)) {
+ printk(KERN_INFO "synaptics: using relaxed packet validation\n");
+ return SYN_NEWABS_RELAXED;
+ }
+
+ return SYN_NEWABS_STRICT;
}
void synaptics_process_byte(struct psmouse *psmouse, struct pt_regs *regs)
@@ -630,13 +655,17 @@ void synaptics_process_byte(struct psmou
printk(KERN_NOTICE "Synaptics driver resynced.\n");
}
+ if (unlikely(priv->pkt_type == SYN_NEWABS))
+ priv->pkt_type = synaptics_detect_pkt_type(psmouse);
+
if (psmouse->ptport && psmouse->ptport->serio.dev && synaptics_is_pt_packet(psmouse->packet))
synaptics_pass_pt_packet(&psmouse->ptport->serio, psmouse->packet);
else
synaptics_process_packet(psmouse);
psmouse->pktcnt = 0;
- } else if (psmouse->pktcnt && !synaptics_validate_byte(psmouse)) {
+ } else if (psmouse->pktcnt &&
+ !synaptics_validate_byte(psmouse->packet, psmouse->pktcnt - 1, priv->pkt_type)) {
printk(KERN_WARNING "Synaptics driver lost sync at byte %d\n", psmouse->pktcnt);
psmouse->pktcnt = 0;
if (++priv->out_of_sync == psmouse_resetafter) {
--- diff/drivers/input/mouse/synaptics.h 2004-01-19 10:22:56.000000000 +0000
+++ source/drivers/input/mouse/synaptics.h 2004-03-16 11:27:48.153867160 +0000
@@ -70,6 +70,12 @@ extern int synaptics_init(struct psmouse
#define SYN_PS_SET_MODE2 0x14
#define SYN_PS_CLIENT_CMD 0x28
+/* synaptics packet types */
+#define SYN_NEWABS 0
+#define SYN_NEWABS_STRICT 1
+#define SYN_NEWABS_RELAXED 2
+#define SYN_OLDABS 3
+
/*
* A structure to describe the state of the touchpad hardware (buttons and pad)
*/
@@ -103,6 +109,7 @@ struct synaptics_data {
/* Data for normal processing */
unsigned int out_of_sync; /* # of packets out of sync */
int old_w; /* Previous w value */
+ unsigned char pkt_type; /* packet type - old, new, etc */
};
#endif /* _SYNAPTICS_H */
--- diff/drivers/input/serio/Kconfig 2003-09-30 14:46:14.000000000 +0000
+++ source/drivers/input/serio/Kconfig 2004-03-16 11:27:48.157866552 +0000
@@ -20,6 +20,7 @@ config SERIO_I8042
tristate "i8042 PC Keyboard controller" if EMBEDDED || !X86
default y
select SERIO
+ depends on !PARISC
---help---
i8042 is the chip over which the standard AT keyboard and PS/2
mouse are connected to the computer. If you use these devices,
@@ -48,6 +49,7 @@ config SERIO_SERPORT
config SERIO_CT82C710
tristate "ct82c710 Aux port controller"
depends on SERIO
+ depends on !PARISC
---help---
Say Y here if you have a Texas Instruments TravelMate notebook
equipped with the ct82c710 chip and want to use a mouse connected
@@ -105,6 +107,20 @@ config SERIO_98KBD
To compile this driver as a module, choose M here: the
module will be called 98kbd-io.
+config SERIO_GSCPS2
+ tristate "HP GSC PS/2 keyboard and PS/2 mouse controller"
+ depends on GSC && SERIO
+ default y
+ help
+ This driver provides support for the PS/2 ports on PA-RISC machines
+ over which HP PS/2 keyboards and PS/2 mice may be connected.
+ If you use these devices, you'll need to say Y here.
+
+ It's safe to enable this driver, so if unsure, say Y.
+
+ To compile this driver as a module, choose M here: the
+ module will be called gscps2.
+
config SERIO_PCIPS2
tristate "PCI PS/2 keyboard and PS/2 mouse controller"
depends on PCI && SERIO
--- diff/drivers/input/serio/Makefile 2003-06-30 09:07:29.000000000 +0000
+++ source/drivers/input/serio/Makefile 2004-03-16 11:27:48.158866400 +0000
@@ -14,4 +14,5 @@ obj-$(CONFIG_SERIO_SA1111) += sa1111ps2.
obj-$(CONFIG_SERIO_AMBAKMI) += ambakmi.o
obj-$(CONFIG_SERIO_Q40KBD) += q40kbd.o
obj-$(CONFIG_SERIO_98KBD) += 98kbd-io.o
+obj-$(CONFIG_SERIO_GSCPS2) += gscps2.o
obj-$(CONFIG_SERIO_PCIPS2) += pcips2.o
--- diff/drivers/input/serio/i8042-io.h 2003-06-30 09:07:29.000000000 +0000
+++ source/drivers/input/serio/i8042-io.h 2004-03-16 11:27:48.157866552 +0000
@@ -25,6 +25,9 @@
#elif defined(__ia64__)
# define I8042_KBD_IRQ isa_irq_to_vector(1)
# define I8042_AUX_IRQ isa_irq_to_vector(12)
+#elif defined(__arm__)
+/* defined in include/asm-arm/arch-xxx/irqs.h */
+#include <asm/irq.h>
#else
# define I8042_KBD_IRQ 1
# define I8042_AUX_IRQ 12
--- diff/drivers/input/serio/i8042.c 2004-03-11 10:20:24.000000000 +0000
+++ source/drivers/input/serio/i8042.c 2004-03-16 11:27:48.157866552 +0000
@@ -52,6 +52,13 @@ static unsigned int i8042_dumbkbd;
module_param_named(dumbkbd, i8042_dumbkbd, bool, 0);
MODULE_PARM_DESC(dumbkbd, "Pretend that controller can only read data from keyboard");
+__obsolete_setup("i8042_noaux");
+__obsolete_setup("i8042_nomux");
+__obsolete_setup("i8042_unlock");
+__obsolete_setup("i8042_reset");
+__obsolete_setup("i8042_direct");
+__obsolete_setup("i8042_dumbkbd");
+
#undef DEBUG
#include "i8042.h"
@@ -379,6 +386,8 @@ static irqreturn_t i8042_interrupt(int i
unsigned int dfl;
int ret;
+ mod_timer(&i8042_timer, jiffies + I8042_POLL_PERIOD);
+
spin_lock_irqsave(&i8042_lock, flags);
str = i8042_read_status();
if (str & I8042_STR_OBF)
@@ -433,7 +442,6 @@ static irqreturn_t i8042_interrupt(int i
irq_ret:
ret = 1;
out:
- mod_timer(&i8042_timer, jiffies + I8042_POLL_PERIOD);
return IRQ_RETVAL(ret);
}
@@ -522,6 +530,11 @@ static int __init i8042_check_mux(struct
if (i8042_enable_mux_mode(values, &mux_version))
return -1;
+
+ /* Workaround for broken chips which seem to support MUX, but in reality don't. */
+ /* They all report version 12.10 */
+ if (mux_version == 0xCA)
+ return -1;
printk(KERN_INFO "i8042.c: Detected active multiplexing controller, rev %d.%d.\n",
(mux_version >> 4) & 0xf, mux_version & 0xf);
@@ -709,14 +722,6 @@ static int i8042_controller_init(void)
}
/*
- * If the chip is configured into nontranslated mode by the BIOS, don't
- * bother enabling translating and be happy.
- */
-
- if (~i8042_ctr & I8042_CTR_XLATE)
- i8042_direct = 1;
-
-/*
* Set nontranslated mode for the kbd interface if requested by an option.
* After this the kbd interface becomes a simple serial in/out, like the aux
* interface is. We don't do this by default, since it can confuse notebook
--- diff/drivers/input/serio/serio.c 2004-01-19 10:22:56.000000000 +0000
+++ source/drivers/input/serio/serio.c 2004-03-16 11:27:48.158866400 +0000
@@ -195,6 +195,9 @@ irqreturn_t serio_interrupt(struct serio
ret = serio->dev->interrupt(serio, data, flags, regs);
} else {
if (!flags) {
+ if ((serio->type == SERIO_8042 ||
+ serio->type == SERIO_8042_XL) && (data != 0xaa))
+ return ret;
serio_rescan(serio);
ret = IRQ_HANDLED;
}
--- diff/drivers/isdn/hisax/hfc_usb.c 2004-03-11 10:20:24.000000000 +0000
+++ source/drivers/isdn/hisax/hfc_usb.c 2004-03-16 11:27:48.160866096 +0000
@@ -1349,9 +1349,11 @@ static int __devinit hfc_usb_probe(struc
{
struct usb_device *dev= interface_to_usbdev(intf);
hfcusb_data *context;
- struct usb_host_interface *iface = intf->altsetting + intf->act_altsetting;
+ struct usb_host_interface *iface = intf->cur_altsetting;
+ struct usb_host_interface *iface_used = NULL;
struct usb_host_endpoint *ep;
- int i, idx, probe_alt_setting,vend_idx, cfg_used, *vcf, attr, cfg_found, cidx, ep_addr;
+ int ifnum = iface->desc.bInterfaceNumber;
+ int i, idx, alt_idx, probe_alt_setting, vend_idx, cfg_used, *vcf, attr, cfg_found, cidx, ep_addr;
int cmptbl[16],small_match,iso_packet_size,packet_size,alt_used=0;
// usb_show_device(dev);
@@ -1366,7 +1368,7 @@ static int __devinit hfc_usb_probe(struc
#ifdef VERBOSE_USB_DEBUG
printk(KERN_INFO "HFC-USB: probing interface(%d) actalt(%d) minor(%d)\n",
- intf->altsetting->desc.bInterfaceNumber, intf->act_altsetting, intf->minor);
+ ifnum, iface->desc.bAlternateSetting, intf->minor);
#endif
if (vend_idx != 0xffff) {
@@ -1374,14 +1376,15 @@ static int __devinit hfc_usb_probe(struc
printk(KERN_INFO "HFC-USB: found vendor idx:%d name:%s\n",vend_idx,vdata[vend_idx].vend_name);
#endif
/* if vendor and product ID is OK, start probing a matching alternate setting ... */
- probe_alt_setting = 0;
+ alt_idx = 0;
small_match=0xffff;
// default settings
iso_packet_size=16;
packet_size=64;
- while(probe_alt_setting < intf->num_altsetting) {
- iface = intf->altsetting + probe_alt_setting;
+ while (alt_idx < intf->num_altsetting) {
+ iface = intf->altsetting + alt_idx;
+ probe_alt_setting = iface->desc.bAlternateSetting;
cfg_used=0;
#ifdef VERBOSE_USB_DEBUG
@@ -1395,7 +1398,7 @@ static int __devinit hfc_usb_probe(struc
#ifdef VERBOSE_USB_DEBUG
printk(KERN_INFO "HFC-USB: (if=%d alt=%d cfg_used=%d)\n",
- probe_alt_setting, intf->act_altsetting,cfg_used);
+ ifnum, probe_alt_setting, cfg_used);
#endif
// copy table
memcpy(cmptbl,vcf,16*sizeof(int));
@@ -1448,6 +1451,7 @@ static int __devinit hfc_usb_probe(struc
if (cfg_used < small_match) {
small_match = cfg_used;
alt_used = probe_alt_setting;
+ iface_used = iface;
}
#ifdef VERBOSE_USB_DEBUG
printk(KERN_INFO "HFC-USB: small_match=%x %x\n", small_match, alt_used);
@@ -1457,15 +1461,14 @@ static int __devinit hfc_usb_probe(struc
cfg_used++;
}
- probe_alt_setting++;
- } /* (probe_alt_setting < intf->num_altsetting) */
+ alt_idx++;
+ } /* (alt_idx < intf->num_altsetting) */
#ifdef VERBOSE_USB_DEBUG
printk(KERN_INFO "HFC-USB: final small_match=%x alt_used=%x\n",small_match, alt_used);
#endif
// yiipiee, we found a valid config
if (small_match != 0xffff) {
- intf->act_altsetting = alt_used;
- iface = intf->altsetting + intf->act_altsetting;
+ iface = iface_used;
if (!(context = kmalloc(sizeof(hfcusb_data), GFP_KERNEL)))
return(-ENOMEM); /* got no mem */
@@ -1542,8 +1545,8 @@ static int __devinit hfc_usb_probe(struc
// now share our luck
context->dev = dev; /* save device */
- context->if_used = intf->altsetting->desc.bInterfaceNumber; /* save used interface */
- context->alt_used = intf->act_altsetting; /* and alternate config */
+ context->if_used = ifnum; /* save used interface */
+ context->alt_used = alt_used; /* and alternate config */
context->ctrl_paksize = dev->descriptor.bMaxPacketSize0; /* control size */
context->cfg_used=vcf[16]; // store used config
context->vend_idx=vend_idx; // store found vendor
--- diff/drivers/md/dm-crypt.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/md/dm-crypt.c 2004-03-16 11:27:48.162865792 +0000
@@ -668,7 +668,7 @@ static int crypt_map(struct dm_target *t
/* out of memory -> run queues */
if (remaining)
- blk_run_queues();
+ blk_congestion_wait(bio_data_dir(clone), HZ/100);
}
/* drop reference, clones could have returned before we reach this */
--- diff/drivers/md/dm-table.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/md/dm-table.c 2004-03-16 11:27:48.163865640 +0000
@@ -279,6 +279,9 @@ void dm_table_get(struct dm_table *t)
void dm_table_put(struct dm_table *t)
{
+ if (!t)
+ return;
+
if (atomic_dec_and_test(&t->holders))
table_destroy(t);
}
@@ -867,9 +870,41 @@ void dm_table_resume_targets(struct dm_t
}
}
+int dm_table_any_congested(struct dm_table *t, int bdi_bits)
+{
+ struct list_head *d, *devices;
+ int r = 0;
+
+ devices = dm_table_get_devices(t);
+ for (d = devices->next; d != devices; d = d->next) {
+ struct dm_dev *dd = list_entry(d, struct dm_dev, list);
+ request_queue_t *q = bdev_get_queue(dd->bdev);
+ r |= bdi_congested(&q->backing_dev_info, bdi_bits);
+ }
+
+ return r;
+}
+
+void dm_table_unplug_all(struct dm_table *t)
+{
+ struct list_head *d, *devices = dm_table_get_devices(t);
+
+ for (d = devices->next; d != devices; d = d->next) {
+ struct dm_dev *dd = list_entry(d, struct dm_dev, list);
+ request_queue_t *q = bdev_get_queue(dd->bdev);
+
+ if (q->unplug_fn) {
+ set_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags);
+ q->unplug_fn(q);
+ }
+ }
+}
EXPORT_SYMBOL(dm_vcalloc);
EXPORT_SYMBOL(dm_get_device);
EXPORT_SYMBOL(dm_put_device);
EXPORT_SYMBOL(dm_table_event);
EXPORT_SYMBOL(dm_table_get_mode);
+EXPORT_SYMBOL(dm_table_put);
+EXPORT_SYMBOL(dm_table_get);
+EXPORT_SYMBOL(dm_table_unplug_all);
--- diff/drivers/md/dm.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/md/dm.c 2004-03-16 11:27:48.161865944 +0000
@@ -49,6 +49,7 @@ struct target_io {
struct mapped_device {
struct rw_semaphore lock;
+ rwlock_t map_lock;
atomic_t holders;
unsigned long flags;
@@ -237,6 +238,24 @@ static int queue_io(struct mapped_device
return 0; /* deferred successfully */
}
+/*
+ * Everyone (including functions in this file), should use this
+ * function to access the md->map field, and make sure they call
+ * dm_table_put() when finished.
+ */
+struct dm_table *dm_get_table(struct mapped_device *md)
+{
+ struct dm_table *t;
+
+ read_lock(&md->map_lock);
+ t = md->map;
+ if (t)
+ dm_table_get(t);
+ read_unlock(&md->map_lock);
+
+ return t;
+}
+
/*-----------------------------------------------------------------
* CRUD START:
* A more elegant soln is in the works that uses the queue
@@ -345,6 +364,7 @@ static void __map_bio(struct dm_target *
struct clone_info {
struct mapped_device *md;
+ struct dm_table *map;
struct bio *bio;
struct dm_io *io;
sector_t sector;
@@ -398,7 +418,7 @@ static struct bio *clone_bio(struct bio
static void __clone_and_map(struct clone_info *ci)
{
struct bio *clone, *bio = ci->bio;
- struct dm_target *ti = dm_table_find_target(ci->md->map, ci->sector);
+ struct dm_target *ti = dm_table_find_target(ci->map, ci->sector);
sector_t len = 0, max = max_io_len(ci->md, ci->sector, ti);
struct target_io *tio;
@@ -459,7 +479,7 @@ static void __clone_and_map(struct clone
ci->sector += max;
ci->sector_count -= max;
- ti = dm_table_find_target(ci->md->map, ci->sector);
+ ti = dm_table_find_target(ci->map, ci->sector);
len = to_sector(bv->bv_len) - max;
clone = split_bvec(bio, ci->sector, ci->idx,
@@ -483,6 +503,12 @@ static void __split_bio(struct mapped_de
{
struct clone_info ci;
+ ci.map = dm_get_table(md);
+ if (!ci.map) {
+ bio_io_error(bio, bio->bi_size);
+ return;
+ }
+
ci.md = md;
ci.bio = bio;
ci.io = alloc_io(md);
@@ -500,22 +526,12 @@ static void __split_bio(struct mapped_de
/* drop the extra reference count */
dec_pending(ci.io, 0);
+ dm_table_put(ci.map);
}
/*-----------------------------------------------------------------
* CRUD END
*---------------------------------------------------------------*/
-
-static inline void __dm_request(struct mapped_device *md, struct bio *bio)
-{
- if (!md->map) {
- bio_io_error(bio, bio->bi_size);
- return;
- }
-
- __split_bio(md, bio);
-}
-
/*
* The request function that just remaps the bio built up by
* dm_merge_bvec.
@@ -554,11 +570,38 @@ static int dm_request(request_queue_t *q
down_read(&md->lock);
}
- __dm_request(md, bio);
+ __split_bio(md, bio);
up_read(&md->lock);
return 0;
}
+static void dm_unplug_all(request_queue_t *q)
+{
+ struct mapped_device *md = q->queuedata;
+ struct dm_table *map = dm_get_table(md);
+
+ clear_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags);
+ if (map) {
+ dm_table_unplug_all(map);
+ dm_table_put(map);
+ }
+}
+
+static int dm_any_congested(void *congested_data, int bdi_bits)
+{
+ int r;
+ struct mapped_device *md = (struct mapped_device *) congested_data;
+ struct dm_table *map = dm_get_table(md);
+
+ if (!map || test_bit(DMF_BLOCK_IO, &md->flags))
+ r = bdi_bits;
+ else
+ r = dm_table_any_congested(map, bdi_bits);
+
+ dm_table_put(map);
+ return r;
+}
+
/*-----------------------------------------------------------------
* A bitset is used to keep track of allocated minor numbers.
*---------------------------------------------------------------*/
@@ -630,6 +673,7 @@ static struct mapped_device *alloc_dev(u
memset(md, 0, sizeof(*md));
init_rwsem(&md->lock);
+ rwlock_init(&md->map_lock);
atomic_set(&md->holders, 1);
md->queue = blk_alloc_queue(GFP_KERNEL);
@@ -637,7 +681,10 @@ static struct mapped_device *alloc_dev(u
goto bad1;
md->queue->queuedata = md;
+ md->queue->backing_dev_info.congested_fn = dm_any_congested;
+ md->queue->backing_dev_info.congested_data = md;
blk_queue_make_request(md->queue, dm_request);
+ md->queue->unplug_fn = dm_unplug_all;
md->io_pool = mempool_create(MIN_IOS, mempool_alloc_slab,
mempool_free_slab, _io_cache);
@@ -727,22 +774,28 @@ static int __bind(struct mapped_device *
if (size == 0)
return 0;
+ write_lock(&md->map_lock);
md->map = t;
- dm_table_event_callback(md->map, event_callback, md);
+ write_unlock(&md->map_lock);
dm_table_get(t);
+ dm_table_event_callback(md->map, event_callback, md);
dm_table_set_restrictions(t, q);
return 0;
}
static void __unbind(struct mapped_device *md)
{
- if (!md->map)
+ struct dm_table *map = md->map;
+
+ if (!map)
return;
- dm_table_event_callback(md->map, NULL, NULL);
- dm_table_put(md->map);
+ dm_table_event_callback(map, NULL, NULL);
+ write_lock(&md->map_lock);
md->map = NULL;
+ write_unlock(&md->map_lock);
+ dm_table_put(map);
}
/*
@@ -778,12 +831,16 @@ void dm_get(struct mapped_device *md)
void dm_put(struct mapped_device *md)
{
+ struct dm_table *map = dm_get_table(md);
+
if (atomic_dec_and_test(&md->holders)) {
- if (!test_bit(DMF_SUSPENDED, &md->flags) && md->map)
- dm_table_suspend_targets(md->map);
+ if (!test_bit(DMF_SUSPENDED, &md->flags) && map)
+ dm_table_suspend_targets(map);
__unbind(md);
free_dev(md);
}
+
+ dm_table_put(map);
}
/*
@@ -796,7 +853,7 @@ static void __flush_deferred_io(struct m
while (c) {
n = c->bi_next;
c->bi_next = NULL;
- __dm_request(md, c);
+ __split_bio(md, c);
c = n;
}
}
@@ -834,6 +891,7 @@ int dm_swap_table(struct mapped_device *
*/
int dm_suspend(struct mapped_device *md)
{
+ struct dm_table *map;
DECLARE_WAITQUEUE(wait, current);
down_write(&md->lock);
@@ -851,11 +909,17 @@ int dm_suspend(struct mapped_device *md)
add_wait_queue(&md->wait, &wait);
up_write(&md->lock);
+ /* unplug */
+ map = dm_get_table(md);
+ if (map) {
+ dm_table_unplug_all(map);
+ dm_table_put(map);
+ }
+
/*
* Then we wait for the already mapped ios to
* complete.
*/
- blk_run_queues();
while (1) {
set_current_state(TASK_INTERRUPTIBLE);
@@ -869,8 +933,11 @@ int dm_suspend(struct mapped_device *md)
down_write(&md->lock);
remove_wait_queue(&md->wait, &wait);
set_bit(DMF_SUSPENDED, &md->flags);
- if (md->map)
- dm_table_suspend_targets(md->map);
+
+ map = dm_get_table(md);
+ if (map)
+ dm_table_suspend_targets(map);
+ dm_table_put(map);
up_write(&md->lock);
return 0;
@@ -879,24 +946,26 @@ int dm_suspend(struct mapped_device *md)
int dm_resume(struct mapped_device *md)
{
struct bio *def;
+ struct dm_table *map = dm_get_table(md);
down_write(&md->lock);
- if (!md->map ||
+ if (!map ||
!test_bit(DMF_SUSPENDED, &md->flags) ||
- !dm_table_get_size(md->map)) {
+ !dm_table_get_size(map)) {
up_write(&md->lock);
+ dm_table_put(map);
return -EINVAL;
}
- dm_table_resume_targets(md->map);
+ dm_table_resume_targets(map);
clear_bit(DMF_SUSPENDED, &md->flags);
clear_bit(DMF_BLOCK_IO, &md->flags);
def = bio_list_get(&md->deferred);
__flush_deferred_io(md, def);
up_write(&md->lock);
-
- blk_run_queues();
+ dm_table_unplug_all(md->map);
+ dm_table_put(map);
return 0;
}
@@ -946,19 +1015,6 @@ struct gendisk *dm_disk(struct mapped_de
return md->disk;
}
-struct dm_table *dm_get_table(struct mapped_device *md)
-{
- struct dm_table *t;
-
- down_read(&md->lock);
- t = md->map;
- if (t)
- dm_table_get(t);
- up_read(&md->lock);
-
- return t;
-}
-
int dm_suspended(struct mapped_device *md)
{
return test_bit(DMF_SUSPENDED, &md->flags);
--- diff/drivers/md/dm.h 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/md/dm.h 2004-03-16 11:27:48.162865792 +0000
@@ -115,6 +115,8 @@ struct list_head *dm_table_get_devices(s
int dm_table_get_mode(struct dm_table *t);
void dm_table_suspend_targets(struct dm_table *t);
void dm_table_resume_targets(struct dm_table *t);
+int dm_table_any_congested(struct dm_table *t, int bdi_bits);
+void dm_table_unplug_all(struct dm_table *t);
/*-----------------------------------------------------------------
* A registry of target types.
--- diff/drivers/md/md.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/md/md.c 2004-03-16 11:27:48.178863360 +0000
@@ -160,6 +160,33 @@ static int md_fail_request (request_queu
return 0;
}
+void md_unplug_mddev(mddev_t *mddev)
+{
+ struct list_head *tmp;
+ mdk_rdev_t *rdev;
+
+ /*
+ * this list iteration is done without any locking in md?!
+ */
+ ITERATE_RDEV(mddev, rdev, tmp) {
+ request_queue_t *r_queue = bdev_get_queue(rdev->bdev);
+
+ if (r_queue->unplug_fn) {
+ set_bit(QUEUE_FLAG_PLUGGED, &r_queue->queue_flags);
+ r_queue->unplug_fn(r_queue);
+ }
+ }
+}
+EXPORT_SYMBOL(md_unplug_mddev);
+
+static void md_unplug_all(request_queue_t *q)
+{
+ mddev_t *mddev = q->queuedata;
+
+ clear_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags);
+ md_unplug_mddev(mddev);
+}
+
static inline mddev_t *mddev_get(mddev_t *mddev)
{
atomic_inc(&mddev->active);
@@ -335,6 +362,8 @@ static int sync_page_io(struct block_dev
struct bio_vec vec;
struct completion event;
+ rw |= (1 << BIO_RW_SYNC);
+
bio_init(&bio);
bio.bi_io_vec = &vec;
vec.bv_page = page;
@@ -349,7 +378,6 @@ static int sync_page_io(struct block_dev
bio.bi_private = &event;
bio.bi_end_io = bi_complete;
submit_bio(rw, &bio);
- blk_run_queues();
wait_for_completion(&event);
return test_bit(BIO_UPTODATE, &bio.bi_flags);
@@ -1644,6 +1672,7 @@ static int do_md_run(mddev_t * mddev)
*/
mddev->queue->queuedata = mddev;
mddev->queue->make_request_fn = mddev->pers->make_request;
+ mddev->queue->unplug_fn = md_unplug_all;
mddev->changed = 1;
return 0;
@@ -2718,7 +2747,7 @@ int md_thread(void * arg)
run = thread->run;
if (run) {
run(thread->mddev);
- blk_run_queues();
+ md_unplug_mddev(thread->mddev);
}
if (signal_pending(current))
flush_signals(current);
@@ -3286,7 +3315,7 @@ static void md_do_sync(mddev_t *mddev)
test_bit(MD_RECOVERY_ERR, &mddev->recovery))
break;
- blk_run_queues();
+ md_unplug_mddev(mddev);
repeat:
if (jiffies >= mark[last_mark] + SYNC_MARK_STEP ) {
--- diff/drivers/md/raid1.c 2004-03-11 10:20:25.000000000 +0000
+++ source/drivers/md/raid1.c 2004-03-16 11:27:48.179863208 +0000
@@ -451,6 +451,7 @@ rb_out:
static void device_barrier(conf_t *conf, sector_t sect)
{
+ md_unplug_mddev(conf->mddev);
spin_lock_irq(&conf->resync_lock);
wait_event_lock_irq(conf->wait_idle, !waitqueue_active(&conf->wait_resume), conf->resync_lock);
@@ -478,6 +479,7 @@ static int make_request(request_queue_t
* thread has put up a bar for new requests.
* Continue immediately if no resync is active currently.
*/
+ md_unplug_mddev(conf->mddev);
spin_lock_irq(&conf->resync_lock);
wait_event_lock_irq(conf->wait_resume, !conf->barrier, conf->resync_lock);
conf->nr_pending++;
@@ -644,6 +646,7 @@ static void print_conf(conf_t *conf)
static void close_sync(conf_t *conf)
{
+ md_unplug_mddev(conf->mddev);
spin_lock_irq(&conf->resync_lock);
wait_event_lock_irq(conf->wait_resume, !conf->barrier, conf->resync_lock);
spin_unlock_irq(&conf->resync_lock);
--- diff/drivers/md/raid5.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/md/raid5.c 2004-03-16 11:27:48.181862904 +0000
@@ -249,6 +249,7 @@ static struct stripe_head *get_active_st
break;
if (!sh) {
conf->inactive_blocked = 1;
+ md_unplug_mddev(conf->mddev);
wait_event_lock_irq(conf->wait_for_stripe,
!list_empty(&conf->inactive_list) &&
(atomic_read(&conf->active_stripes) < (NR_STRIPES *3/4)
@@ -1292,9 +1293,8 @@ static inline void raid5_activate_delaye
}
}
}
-static void raid5_unplug_device(void *data)
+static void raid5_unplug_device(request_queue_t *q)
{
- request_queue_t *q = data;
mddev_t *mddev = q->queuedata;
raid5_conf_t *conf = mddev_to_conf(mddev);
unsigned long flags;
--- diff/drivers/md/raid6main.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/md/raid6main.c 2004-03-16 11:27:48.182862752 +0000
@@ -1454,9 +1454,8 @@ static inline void raid6_activate_delaye
}
}
}
-static void raid6_unplug_device(void *data)
+static void raid6_unplug_device(request_queue_t *q)
{
- request_queue_t *q = data;
mddev_t *mddev = q->queuedata;
raid6_conf_t *conf = mddev_to_conf(mddev);
unsigned long flags;
--- diff/drivers/mtd/devices/blkmtd.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/mtd/devices/blkmtd.c 2004-03-16 11:27:48.183862600 +0000
@@ -147,8 +147,7 @@ static int blkmtd_readpage(struct blkmtd
bio->bi_private = &event;
bio->bi_end_io = bi_read_complete;
if(bio_add_page(bio, page, PAGE_SIZE, 0) == PAGE_SIZE) {
- submit_bio(READ, bio);
- blk_run_queues();
+ submit_bio(READ_SYNC, bio);
wait_for_completion(&event);
err = test_bit(BIO_UPTODATE, &bio->bi_flags) ? 0 : -EIO;
bio_put(bio);
@@ -179,8 +178,7 @@ static int blkmtd_write_out(struct bio *
init_completion(&event);
bio->bi_private = &event;
bio->bi_end_io = bi_write_complete;
- submit_bio(WRITE, bio);
- blk_run_queues();
+ submit_bio(WRITE_SYNC, bio);
wait_for_completion(&event);
DEBUG(3, "submit_bio completed, bi_vcnt = %d\n", bio->bi_vcnt);
err = test_bit(BIO_UPTODATE, &bio->bi_flags) ? 0 : -EIO;
--- diff/drivers/net/Makefile 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/net/Makefile 2004-03-16 11:27:48.190861536 +0000
@@ -188,4 +188,6 @@ obj-$(CONFIG_NET_TULIP) += tulip/
obj-$(CONFIG_HAMRADIO) += hamradio/
obj-$(CONFIG_IRDA) += irda/
+# Must come after all NICs that might use them
obj-$(CONFIG_NETCONSOLE) += netconsole.o
+obj-$(CONFIG_KGDB) += kgdb_eth.o
--- diff/drivers/net/acenic.c 2004-03-11 10:20:25.000000000 +0000
+++ source/drivers/net/acenic.c 2004-03-16 11:27:48.187861992 +0000
@@ -131,7 +131,6 @@
#define PCI_DEVICE_ID_SGI_ACENIC 0x0009
#endif
-#if LINUX_VERSION_CODE >= 0x20400
static struct pci_device_id acenic_pci_tbl[] = {
{ PCI_VENDOR_ID_ALTEON, PCI_DEVICE_ID_ALTEON_ACENIC_FIBRE,
PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_NETWORK_ETHERNET << 8, 0xffff00, },
@@ -156,37 +155,6 @@ static struct pci_device_id acenic_pci_t
{ }
};
MODULE_DEVICE_TABLE(pci, acenic_pci_tbl);
-#endif
-
-
-#ifndef MODULE_LICENSE
-#define MODULE_LICENSE(a)
-#endif
-
-#ifndef wmb
-#define wmb() mb()
-#endif
-
-#ifndef __exit
-#define __exit
-#endif
-
-#ifndef __devinit
-#define __devinit __init
-#endif
-
-#ifndef SMP_CACHE_BYTES
-#define SMP_CACHE_BYTES L1_CACHE_BYTES
-#endif
-
-#ifndef SET_MODULE_OWNER
-#define SET_MODULE_OWNER(dev) do{} while(0)
-#define ACE_MOD_INC_USE_COUNT MOD_INC_USE_COUNT
-#define ACE_MOD_DEC_USE_COUNT MOD_DEC_USE_COUNT
-#else
-#define ACE_MOD_INC_USE_COUNT do{} while(0)
-#define ACE_MOD_DEC_USE_COUNT do{} while(0)
-#endif
#ifndef SET_NETDEV_DEV
#define SET_NETDEV_DEV(net, pdev) do{} while(0)
@@ -198,151 +166,8 @@ MODULE_DEVICE_TABLE(pci, acenic_pci_tbl)
#define ace_sync_irq(irq) synchronize_irq()
#endif
-#if LINUX_VERSION_CODE < 0x2051e
-#define local_irq_save(flags) do{__save_flags(flags) ; \
- __cli();} while(0)
-#define local_irq_restore(flags) __restore_flags(flags)
-#endif
-
-#if (LINUX_VERSION_CODE < 0x02030d)
-#define pci_resource_start(dev, bar) dev->base_address[bar]
-#elif (LINUX_VERSION_CODE < 0x02032c)
-#define pci_resource_start(dev, bar) dev->resource[bar].start
-#endif
-
-#if (LINUX_VERSION_CODE < 0x02030e)
-#define net_device device
-#endif
-
-
-#if (LINUX_VERSION_CODE < 0x02032a)
-typedef u32 dma_addr_t;
-
-static inline void *pci_alloc_consistent(struct pci_dev *hwdev, size_t size,
- dma_addr_t *dma_handle)
-{
- void *virt_ptr;
-
- virt_ptr = kmalloc(size, GFP_KERNEL);
- if (!virt_ptr)
- return NULL;
- *dma_handle = virt_to_bus(virt_ptr);
- return virt_ptr;
-}
-
-#define pci_free_consistent(cookie, size, ptr, dma_ptr) kfree(ptr)
-#define pci_map_page(cookie, page, off, size, dir) \
- virt_to_bus(page_address(page)+(off))
-#define pci_unmap_page(cookie, address, size, dir)
-#define pci_set_dma_mask(dev, mask) \
- (((u64)(mask) & 0xffffffff00000000) == 0 ? 0 : -EIO)
-#define pci_dma_supported(dev, mask) \
- (((u64)(mask) & 0xffffffff00000000) == 0 ? 1 : 0)
-
-#elif (LINUX_VERSION_CODE < 0x02040d)
-
-/*
- * 2.4.13 introduced pci_map_page()/pci_unmap_page() - for 2.4.12 and prior,
- * fall back on pci_map_single()/pci_unnmap_single().
- *
- * We are guaranteed that the page is mapped at this point since
- * pci_map_page() is only used upon valid struct skb's.
- */
-static inline dma_addr_t
-pci_map_page(struct pci_dev *cookie, struct page *page, unsigned long off,
- size_t size, int dir)
-{
- void *page_virt;
-
- page_virt = page_address(page);
- if (!page_virt)
- BUG();
- return pci_map_single(cookie, (page_virt + off), size, dir);
-}
-#define pci_unmap_page(cookie, dma_addr, size, dir) \
- pci_unmap_single(cookie, dma_addr, size, dir)
-#endif
-
-#if (LINUX_VERSION_CODE < 0x020412)
-#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME)
-#define DECLARE_PCI_UNMAP_LEN(LEN_NAME)
-#define pci_unmap_addr(PTR, ADDR_NAME) 0
-#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL) do{} while(0)
-#define pci_unmap_len(PTR, LEN_NAME) 0
-#define pci_unmap_len_set(PTR, LEN_NAME, VAL) do{} while(0)
-#endif
-
-
-#if (LINUX_VERSION_CODE < 0x02032b)
-/*
- * SoftNet
- *
- * For pre-softnet kernels we need to tell the upper layer not to
- * re-enter start_xmit() while we are in there. However softnet
- * guarantees not to enter while we are in there so there is no need
- * to do the netif_stop_queue() dance unless the transmit queue really
- * gets stuck. This should also improve performance according to tests
- * done by Aman Singla.
- */
-#define dev_kfree_skb_irq(a) dev_kfree_skb(a)
-#define netif_wake_queue(dev) clear_bit(0, &dev->tbusy)
-#define netif_stop_queue(dev) set_bit(0, &dev->tbusy)
-#define late_stop_netif_stop_queue(dev) do{} while(0)
-#define early_stop_netif_stop_queue(dev) test_and_set_bit(0,&dev->tbusy)
-#define early_stop_netif_wake_queue(dev) netif_wake_queue(dev)
-
-static inline void netif_start_queue(struct net_device *dev)
-{
- dev->tbusy = 0;
- dev->interrupt = 0;
- dev->start = 1;
-}
-
-#define ace_mark_net_bh() mark_bh(NET_BH)
-#define netif_queue_stopped(dev) dev->tbusy
-#define netif_running(dev) dev->start
-#define ace_if_down(dev) do{dev->start = 0;} while(0)
-
-#define tasklet_struct tq_struct
-static inline void tasklet_schedule(struct tasklet_struct *tasklet)
-{
- queue_task(tasklet, &tq_immediate);
- mark_bh(IMMEDIATE_BH);
-}
-
-static inline void tasklet_init(struct tasklet_struct *tasklet,
- void (*func)(unsigned long),
- unsigned long data)
-{
- tasklet->next = NULL;
- tasklet->sync = 0;
- tasklet->routine = (void (*)(void *))func;
- tasklet->data = (void *)data;
-}
-#define tasklet_kill(tasklet) do{} while(0)
-#else
-#define late_stop_netif_stop_queue(dev) netif_stop_queue(dev)
-#define early_stop_netif_stop_queue(dev) 0
-#define early_stop_netif_wake_queue(dev) do{} while(0)
-#define ace_mark_net_bh() do{} while(0)
-#define ace_if_down(dev) do{} while(0)
-#endif
-
-#if (LINUX_VERSION_CODE >= 0x02031b)
-#define NEW_NETINIT
-#define ACE_PROBE_ARG void
-#else
-#define ACE_PROBE_ARG struct net_device *dev
-#endif
-
-#ifndef min_t
-#define min_t(type,a,b) (((a)<(b))?(a):(b))
-#endif
-
-#ifndef ARCH_HAS_PREFETCHW
-#ifndef prefetchw
-#define prefetchw(x) do{} while(0)
-#endif
+#ifndef offset_in_page
+#define offset_in_page(ptr) ((unsigned long)(ptr) & ~PAGE_MASK)
#endif
#define ACE_MAX_MOD_PARMS 8
@@ -595,407 +420,323 @@ static int max_rx_desc[ACE_MAX_MOD_PARMS
static int tx_ratio[ACE_MAX_MOD_PARMS];
static int dis_pci_mem_inval[ACE_MAX_MOD_PARMS] = {1, 1, 1, 1, 1, 1, 1, 1};
+MODULE_AUTHOR("Jes Sorensen <jes@trained-monkey.org>");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("AceNIC/3C985/GA620 Gigabit Ethernet driver");
+MODULE_PARM(link, "1-" __MODULE_STRING(8) "i");
+MODULE_PARM(trace, "1-" __MODULE_STRING(8) "i");
+MODULE_PARM(tx_coal_tick, "1-" __MODULE_STRING(8) "i");
+MODULE_PARM(max_tx_desc, "1-" __MODULE_STRING(8) "i");
+MODULE_PARM(rx_coal_tick, "1-" __MODULE_STRING(8) "i");
+MODULE_PARM(max_rx_desc, "1-" __MODULE_STRING(8) "i");
+MODULE_PARM(tx_ratio, "1-" __MODULE_STRING(8) "i");
+MODULE_PARM_DESC(link, "AceNIC/3C985/NetGear link state");
+MODULE_PARM_DESC(trace, "AceNIC/3C985/NetGear firmware trace level");
+MODULE_PARM_DESC(tx_coal_tick, "AceNIC/3C985/GA620 max clock ticks to wait from first tx descriptor arrives");
+MODULE_PARM_DESC(max_tx_desc, "AceNIC/3C985/GA620 max number of transmit descriptors to wait");
+MODULE_PARM_DESC(rx_coal_tick, "AceNIC/3C985/GA620 max clock ticks to wait from first rx descriptor arrives");
+MODULE_PARM_DESC(max_rx_desc, "AceNIC/3C985/GA620 max number of receive descriptors to wait");
+MODULE_PARM_DESC(tx_ratio, "AceNIC/3C985/GA620 ratio of NIC memory used for TX/RX descriptors (range 0-63)");
+
+
static char version[] __initdata =
"acenic.c: v0.92 08/05/2002 Jes Sorensen, linux-acenic@SunSITE.dk\n"
" http://home.cern.ch/~jes/gige/acenic.html\n";
-static struct net_device *root_dev;
-
-static int probed __initdata = 0;
-
-
-int __devinit acenic_probe (ACE_PROBE_ARG)
+static int __devinit acenic_probe_one(struct pci_dev *pdev,
+ const struct pci_device_id *id)
{
-#ifdef NEW_NETINIT
struct net_device *dev;
-#endif
struct ace_private *ap;
- struct pci_dev *pdev = NULL;
- int boards_found = 0;
- int version_disp;
-
- if (probed)
- return -ENODEV;
- probed++;
-
- version_disp = 0;
-
- while ((pdev = pci_find_class(PCI_CLASS_NETWORK_ETHERNET<<8, pdev))) {
-
- if (!((pdev->vendor == PCI_VENDOR_ID_ALTEON) &&
- ((pdev->device == PCI_DEVICE_ID_ALTEON_ACENIC_FIBRE) ||
- (pdev->device == PCI_DEVICE_ID_ALTEON_ACENIC_COPPER)))&&
- !((pdev->vendor == PCI_VENDOR_ID_3COM) &&
- (pdev->device == PCI_DEVICE_ID_3COM_3C985)) &&
- !((pdev->vendor == PCI_VENDOR_ID_NETGEAR) &&
- ((pdev->device == PCI_DEVICE_ID_NETGEAR_GA620) ||
- (pdev->device == PCI_DEVICE_ID_NETGEAR_GA620T))) &&
- /*
- * Farallon used the DEC vendor ID on their cards by
- * mistake for a while
- */
- !((pdev->vendor == PCI_VENDOR_ID_DEC) &&
- (pdev->device == PCI_DEVICE_ID_FARALLON_PN9000SX)) &&
- !((pdev->vendor == PCI_VENDOR_ID_ALTEON) &&
- (pdev->device == PCI_DEVICE_ID_FARALLON_PN9100T)) &&
- !((pdev->vendor == PCI_VENDOR_ID_SGI) &&
- (pdev->device == PCI_DEVICE_ID_SGI_ACENIC)))
- continue;
-
- dev = alloc_etherdev(sizeof(struct ace_private));
- if (dev == NULL) {
- printk(KERN_ERR "acenic: Unable to allocate "
- "net_device structure!\n");
- break;
- }
+ static int boards_found;
- SET_MODULE_OWNER(dev);
- SET_NETDEV_DEV(dev, &pdev->dev);
+ dev = alloc_etherdev(sizeof(struct ace_private));
+ if (dev == NULL) {
+ printk(KERN_ERR "acenic: Unable to allocate "
+ "net_device structure!\n");
+ return -ENOMEM;
+ }
- ap = dev->priv;
- ap->pdev = pdev;
+ SET_MODULE_OWNER(dev);
+ SET_NETDEV_DEV(dev, &pdev->dev);
- dev->open = &ace_open;
- dev->hard_start_xmit = &ace_start_xmit;
- dev->features |= NETIF_F_SG | NETIF_F_IP_CSUM;
+ ap = dev->priv;
+ ap->pdev = pdev;
+
+ dev->features |= NETIF_F_SG | NETIF_F_IP_CSUM;
#if ACENIC_DO_VLAN
- dev->features |= NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX;
- dev->vlan_rx_register = ace_vlan_rx_register;
- dev->vlan_rx_kill_vid = ace_vlan_rx_kill_vid;
-#endif
- if (1) {
- static void ace_watchdog(struct net_device *dev);
- dev->tx_timeout = &ace_watchdog;
- dev->watchdog_timeo = 5*HZ;
- }
- dev->stop = &ace_close;
- dev->get_stats = &ace_get_stats;
- dev->set_multicast_list = &ace_set_multicast_list;
- dev->do_ioctl = &ace_ioctl;
- dev->set_mac_address = &ace_set_mac_addr;
- dev->change_mtu = &ace_change_mtu;
+ dev->features |= NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX;
+ dev->vlan_rx_register = ace_vlan_rx_register;
+ dev->vlan_rx_kill_vid = ace_vlan_rx_kill_vid;
+#endif
+ if (1) {
+ static void ace_watchdog(struct net_device *dev);
+ dev->tx_timeout = &ace_watchdog;
+ dev->watchdog_timeo = 5*HZ;
+ }
- /* display version info if adapter is found */
- if (!version_disp)
- {
- /* set display flag to TRUE so that */
- /* we only display this string ONCE */
- version_disp = 1;
- printk(version);
- }
+ dev->open = &ace_open;
+ dev->stop = &ace_close;
+ dev->hard_start_xmit = &ace_start_xmit;
+ dev->get_stats = &ace_get_stats;
+ dev->set_multicast_list = &ace_set_multicast_list;
+ dev->do_ioctl = &ace_ioctl;
+ dev->set_mac_address = &ace_set_mac_addr;
+ dev->change_mtu = &ace_change_mtu;
- if (pci_enable_device(pdev)) {
- free_netdev(dev);
- continue;
- }
+ /* we only display this string ONCE */
+ if (!boards_found)
+ printk(version);
- /*
- * Enable master mode before we start playing with the
- * pci_command word since pci_set_master() will modify
- * it.
- */
- pci_set_master(pdev);
+ if (pci_enable_device(pdev))
+ goto fail_free_netdev;
- pci_read_config_word(pdev, PCI_COMMAND, &ap->pci_command);
+ /*
+ * Enable master mode before we start playing with the
+ * pci_command word since pci_set_master() will modify
+ * it.
+ */
+ pci_set_master(pdev);
- /* OpenFirmware on Mac's does not set this - DOH.. */
- if (!(ap->pci_command & PCI_COMMAND_MEMORY)) {
- printk(KERN_INFO "%s: Enabling PCI Memory Mapped "
- "access - was not enabled by BIOS/Firmware\n",
- dev->name);
- ap->pci_command = ap->pci_command | PCI_COMMAND_MEMORY;
- pci_write_config_word(ap->pdev, PCI_COMMAND,
- ap->pci_command);
- wmb();
- }
+ pci_read_config_word(pdev, PCI_COMMAND, &ap->pci_command);
- pci_read_config_byte(pdev, PCI_LATENCY_TIMER,
- &ap->pci_latency);
- if (ap->pci_latency <= 0x40) {
- ap->pci_latency = 0x40;
- pci_write_config_byte(pdev, PCI_LATENCY_TIMER,
- ap->pci_latency);
- }
+ /* OpenFirmware on Mac's does not set this - DOH.. */
+ if (!(ap->pci_command & PCI_COMMAND_MEMORY)) {
+ printk(KERN_INFO "%s: Enabling PCI Memory Mapped "
+ "access - was not enabled by BIOS/Firmware\n",
+ dev->name);
+ ap->pci_command = ap->pci_command | PCI_COMMAND_MEMORY;
+ pci_write_config_word(ap->pdev, PCI_COMMAND,
+ ap->pci_command);
+ wmb();
+ }
- /*
- * Remap the regs into kernel space - this is abuse of
- * dev->base_addr since it was means for I/O port
- * addresses but who gives a damn.
- */
- dev->base_addr = pci_resource_start(pdev, 0);
- ap->regs = (struct ace_regs *)ioremap(dev->base_addr, 0x4000);
- if (!ap->regs) {
- printk(KERN_ERR "%s: Unable to map I/O register, "
- "AceNIC %i will be disabled.\n",
- dev->name, boards_found);
- break;
- }
+ pci_read_config_byte(pdev, PCI_LATENCY_TIMER, &ap->pci_latency);
+ if (ap->pci_latency <= 0x40) {
+ ap->pci_latency = 0x40;
+ pci_write_config_byte(pdev, PCI_LATENCY_TIMER, ap->pci_latency);
+ }
- switch(pdev->vendor) {
- case PCI_VENDOR_ID_ALTEON:
- if (pdev->device == PCI_DEVICE_ID_FARALLON_PN9100T) {
- strncpy(ap->name, "Farallon PN9100-T "
- "Gigabit Ethernet", sizeof (ap->name));
- printk(KERN_INFO "%s: Farallon PN9100-T ",
- dev->name);
- } else {
- strncpy(ap->name, "AceNIC Gigabit Ethernet",
- sizeof (ap->name));
- printk(KERN_INFO "%s: Alteon AceNIC ",
- dev->name);
- }
- break;
- case PCI_VENDOR_ID_3COM:
- strncpy(ap->name, "3Com 3C985 Gigabit Ethernet",
- sizeof (ap->name));
- printk(KERN_INFO "%s: 3Com 3C985 ", dev->name);
- break;
- case PCI_VENDOR_ID_NETGEAR:
- strncpy(ap->name, "NetGear GA620 Gigabit Ethernet",
- sizeof (ap->name));
- printk(KERN_INFO "%s: NetGear GA620 ", dev->name);
- break;
- case PCI_VENDOR_ID_DEC:
- if (pdev->device == PCI_DEVICE_ID_FARALLON_PN9000SX) {
- strncpy(ap->name, "Farallon PN9000-SX "
- "Gigabit Ethernet", sizeof (ap->name));
- printk(KERN_INFO "%s: Farallon PN9000-SX ",
- dev->name);
- break;
- }
- case PCI_VENDOR_ID_SGI:
- strncpy(ap->name, "SGI AceNIC Gigabit Ethernet",
+ /*
+ * Remap the regs into kernel space - this is abuse of
+ * dev->base_addr since it was means for I/O port
+ * addresses but who gives a damn.
+ */
+ dev->base_addr = pci_resource_start(pdev, 0);
+ ap->regs = (struct ace_regs *)ioremap(dev->base_addr, 0x4000);
+ if (!ap->regs) {
+ printk(KERN_ERR "%s: Unable to map I/O register, "
+ "AceNIC %i will be disabled.\n",
+ dev->name, boards_found);
+ goto fail_free_netdev;
+ }
+
+ switch(pdev->vendor) {
+ case PCI_VENDOR_ID_ALTEON:
+ if (pdev->device == PCI_DEVICE_ID_FARALLON_PN9100T) {
+ strncpy(ap->name, "Farallon PN9100-T "
+ "Gigabit Ethernet", sizeof (ap->name));
+ printk(KERN_INFO "%s: Farallon PN9100-T ",
+ dev->name);
+ } else {
+ strncpy(ap->name, "AceNIC Gigabit Ethernet",
sizeof (ap->name));
- printk(KERN_INFO "%s: SGI AceNIC ", dev->name);
- break;
- default:
- strncpy(ap->name, "Unknown AceNIC based Gigabit "
- "Ethernet", sizeof (ap->name));
- printk(KERN_INFO "%s: Unknown AceNIC ", dev->name);
+ printk(KERN_INFO "%s: Alteon AceNIC ",
+ dev->name);
+ }
+ break;
+ case PCI_VENDOR_ID_3COM:
+ strncpy(ap->name, "3Com 3C985 Gigabit Ethernet",
+ sizeof (ap->name));
+ printk(KERN_INFO "%s: 3Com 3C985 ", dev->name);
+ break;
+ case PCI_VENDOR_ID_NETGEAR:
+ strncpy(ap->name, "NetGear GA620 Gigabit Ethernet",
+ sizeof (ap->name));
+ printk(KERN_INFO "%s: NetGear GA620 ", dev->name);
+ break;
+ case PCI_VENDOR_ID_DEC:
+ if (pdev->device == PCI_DEVICE_ID_FARALLON_PN9000SX) {
+ strncpy(ap->name, "Farallon PN9000-SX "
+ "Gigabit Ethernet", sizeof (ap->name));
+ printk(KERN_INFO "%s: Farallon PN9000-SX ",
+ dev->name);
break;
}
- ap->name [sizeof (ap->name) - 1] = '\0';
- printk("Gigabit Ethernet at 0x%08lx, ", dev->base_addr);
+ case PCI_VENDOR_ID_SGI:
+ strncpy(ap->name, "SGI AceNIC Gigabit Ethernet",
+ sizeof (ap->name));
+ printk(KERN_INFO "%s: SGI AceNIC ", dev->name);
+ break;
+ default:
+ strncpy(ap->name, "Unknown AceNIC based Gigabit "
+ "Ethernet", sizeof (ap->name));
+ printk(KERN_INFO "%s: Unknown AceNIC ", dev->name);
+ break;
+ }
+
+ ap->name [sizeof (ap->name) - 1] = '\0';
+ printk("Gigabit Ethernet at 0x%08lx, ", dev->base_addr);
#ifdef __sparc__
- printk("irq %s\n", __irq_itoa(pdev->irq));
+ printk("irq %s\n", __irq_itoa(pdev->irq));
#else
- printk("irq %i\n", pdev->irq);
+ printk("irq %i\n", pdev->irq);
#endif
#ifdef CONFIG_ACENIC_OMIT_TIGON_I
- if ((readl(&ap->regs->HostCtrl) >> 28) == 4) {
- printk(KERN_ERR "%s: Driver compiled without Tigon I"
- " support - NIC disabled\n", dev->name);
- ace_init_cleanup(dev);
- free_netdev(dev);
- continue;
- }
+ if ((readl(&ap->regs->HostCtrl) >> 28) == 4) {
+ printk(KERN_ERR "%s: Driver compiled without Tigon I"
+ " support - NIC disabled\n", dev->name);
+ goto fail_uninit;
+ }
#endif
- if (ace_allocate_descriptors(dev)) {
- /*
- * ace_allocate_descriptors() calls
- * ace_init_cleanup() on error.
- */
- free_netdev(dev);
- continue;
- }
+ if (ace_allocate_descriptors(dev))
+ goto fail_free_netdev;
#ifdef MODULE
- if (boards_found >= ACE_MAX_MOD_PARMS)
- ap->board_idx = BOARD_IDX_OVERFLOW;
- else
- ap->board_idx = boards_found;
+ if (boards_found >= ACE_MAX_MOD_PARMS)
+ ap->board_idx = BOARD_IDX_OVERFLOW;
+ else
+ ap->board_idx = boards_found;
#else
- ap->board_idx = BOARD_IDX_STATIC;
+ ap->board_idx = BOARD_IDX_STATIC;
#endif
- if (ace_init(dev)) {
- /*
- * ace_init() calls ace_init_cleanup() on error.
- */
- free_netdev(dev);
- continue;
- }
+ if (ace_init(dev))
+ goto fail_free_netdev;
- if (register_netdev(dev)) {
- printk(KERN_ERR "acenic: device registration failed\n");
- ace_init_cleanup(dev);
- free_netdev(dev);
- continue;
- }
-
- if (ap->pci_using_dac)
- dev->features |= NETIF_F_HIGHDMA;
-
- boards_found++;
+ if (register_netdev(dev)) {
+ printk(KERN_ERR "acenic: device registration failed\n");
+ goto fail_uninit;
}
- /*
- * If we're at this point we're going through ace_probe() for
- * the first time. Return success (0) if we've initialized 1
- * or more boards. Otherwise, return failure (-ENODEV).
- */
-
- if (boards_found > 0)
- return 0;
- else
- return -ENODEV;
-}
+ if (ap->pci_using_dac)
+ dev->features |= NETIF_F_HIGHDMA;
+ pci_set_drvdata(pdev, dev);
-#ifdef MODULE
-MODULE_AUTHOR("Jes Sorensen <jes@trained-monkey.org>");
-MODULE_LICENSE("GPL");
-MODULE_DESCRIPTION("AceNIC/3C985/GA620 Gigabit Ethernet driver");
-MODULE_PARM(link, "1-" __MODULE_STRING(8) "i");
-MODULE_PARM(trace, "1-" __MODULE_STRING(8) "i");
-MODULE_PARM(tx_coal_tick, "1-" __MODULE_STRING(8) "i");
-MODULE_PARM(max_tx_desc, "1-" __MODULE_STRING(8) "i");
-MODULE_PARM(rx_coal_tick, "1-" __MODULE_STRING(8) "i");
-MODULE_PARM(max_rx_desc, "1-" __MODULE_STRING(8) "i");
-MODULE_PARM(tx_ratio, "1-" __MODULE_STRING(8) "i");
-MODULE_PARM_DESC(link, "AceNIC/3C985/NetGear link state");
-MODULE_PARM_DESC(trace, "AceNIC/3C985/NetGear firmware trace level");
-MODULE_PARM_DESC(tx_coal_tick, "AceNIC/3C985/GA620 max clock ticks to wait from first tx descriptor arrives");
-MODULE_PARM_DESC(max_tx_desc, "AceNIC/3C985/GA620 max number of transmit descriptors to wait");
-MODULE_PARM_DESC(rx_coal_tick, "AceNIC/3C985/GA620 max clock ticks to wait from first rx descriptor arrives");
-MODULE_PARM_DESC(max_rx_desc, "AceNIC/3C985/GA620 max number of receive descriptors to wait");
-MODULE_PARM_DESC(tx_ratio, "AceNIC/3C985/GA620 ratio of NIC memory used for TX/RX descriptors (range 0-63)");
-#endif
+ boards_found++;
+ return 0;
+ fail_uninit:
+ ace_init_cleanup(dev);
+ fail_free_netdev:
+ free_netdev(dev);
+ return -ENODEV;
+}
-static void __exit ace_module_cleanup(void)
+static void __devexit acenic_remove_one(struct pci_dev *pdev)
{
- struct ace_private *ap;
- struct ace_regs *regs;
- struct net_device *next;
+ struct net_device *dev = pci_get_drvdata(pdev);
+ struct ace_private *ap = dev->priv;
+ struct ace_regs *regs = ap->regs;
short i;
- while (root_dev) {
- ap = root_dev->priv;
- next = ap->next;
- unregister_netdev(root_dev);
+ unregister_netdev(dev);
- regs = ap->regs;
-
- writel(readl(®s->CpuCtrl) | CPU_HALT, ®s->CpuCtrl);
- if (ap->version >= 2)
- writel(readl(®s->CpuBCtrl) | CPU_HALT,
- ®s->CpuBCtrl);
- /*
- * This clears any pending interrupts
- */
- writel(1, ®s->Mb0Lo);
- readl(®s->CpuCtrl); /* flush */
+ writel(readl(®s->CpuCtrl) | CPU_HALT, ®s->CpuCtrl);
+ if (ap->version >= 2)
+ writel(readl(®s->CpuBCtrl) | CPU_HALT, ®s->CpuBCtrl);
+
+ /*
+ * This clears any pending interrupts
+ */
+ writel(1, ®s->Mb0Lo);
+ readl(®s->CpuCtrl); /* flush */
- /*
- * Make sure no other CPUs are processing interrupts
- * on the card before the buffers are being released.
- * Otherwise one might experience some `interesting'
- * effects.
- *
- * Then release the RX buffers - jumbo buffers were
- * already released in ace_close().
- */
- ace_sync_irq(root_dev->irq);
+ /*
+ * Make sure no other CPUs are processing interrupts
+ * on the card before the buffers are being released.
+ * Otherwise one might experience some `interesting'
+ * effects.
+ *
+ * Then release the RX buffers - jumbo buffers were
+ * already released in ace_close().
+ */
+ ace_sync_irq(dev->irq);
- for (i = 0; i < RX_STD_RING_ENTRIES; i++) {
- struct sk_buff *skb = ap->skb->rx_std_skbuff[i].skb;
+ for (i = 0; i < RX_STD_RING_ENTRIES; i++) {
+ struct sk_buff *skb = ap->skb->rx_std_skbuff[i].skb;
- if (skb) {
- struct ring_info *ringp;
- dma_addr_t mapping;
+ if (skb) {
+ struct ring_info *ringp;
+ dma_addr_t mapping;
- ringp = &ap->skb->rx_std_skbuff[i];
- mapping = pci_unmap_addr(ringp, mapping);
- pci_unmap_page(ap->pdev, mapping,
- ACE_STD_BUFSIZE - (2 + 16),
- PCI_DMA_FROMDEVICE);
+ ringp = &ap->skb->rx_std_skbuff[i];
+ mapping = pci_unmap_addr(ringp, mapping);
+ pci_unmap_page(ap->pdev, mapping,
+ ACE_STD_BUFSIZE - (2 + 16),
+ PCI_DMA_FROMDEVICE);
- ap->rx_std_ring[i].size = 0;
- ap->skb->rx_std_skbuff[i].skb = NULL;
- dev_kfree_skb(skb);
- }
- }
- if (ap->version >= 2) {
- for (i = 0; i < RX_MINI_RING_ENTRIES; i++) {
- struct sk_buff *skb = ap->skb->rx_mini_skbuff[i].skb;
-
- if (skb) {
- struct ring_info *ringp;
- dma_addr_t mapping;
-
- ringp = &ap->skb->rx_mini_skbuff[i];
- mapping = pci_unmap_addr(ringp,mapping);
- pci_unmap_page(ap->pdev, mapping,
- ACE_MINI_BUFSIZE - (2 + 16),
- PCI_DMA_FROMDEVICE);
-
- ap->rx_mini_ring[i].size = 0;
- ap->skb->rx_mini_skbuff[i].skb = NULL;
- dev_kfree_skb(skb);
- }
- }
+ ap->rx_std_ring[i].size = 0;
+ ap->skb->rx_std_skbuff[i].skb = NULL;
+ dev_kfree_skb(skb);
}
- for (i = 0; i < RX_JUMBO_RING_ENTRIES; i++) {
- struct sk_buff *skb = ap->skb->rx_jumbo_skbuff[i].skb;
+ }
+
+ if (ap->version >= 2) {
+ for (i = 0; i < RX_MINI_RING_ENTRIES; i++) {
+ struct sk_buff *skb = ap->skb->rx_mini_skbuff[i].skb;
+
if (skb) {
struct ring_info *ringp;
dma_addr_t mapping;
- ringp = &ap->skb->rx_jumbo_skbuff[i];
- mapping = pci_unmap_addr(ringp, mapping);
+ ringp = &ap->skb->rx_mini_skbuff[i];
+ mapping = pci_unmap_addr(ringp,mapping);
pci_unmap_page(ap->pdev, mapping,
- ACE_JUMBO_BUFSIZE - (2 + 16),
+ ACE_MINI_BUFSIZE - (2 + 16),
PCI_DMA_FROMDEVICE);
- ap->rx_jumbo_ring[i].size = 0;
- ap->skb->rx_jumbo_skbuff[i].skb = NULL;
+ ap->rx_mini_ring[i].size = 0;
+ ap->skb->rx_mini_skbuff[i].skb = NULL;
dev_kfree_skb(skb);
}
}
-
- ace_init_cleanup(root_dev);
- free_netdev(root_dev);
- root_dev = next;
}
-}
+ for (i = 0; i < RX_JUMBO_RING_ENTRIES; i++) {
+ struct sk_buff *skb = ap->skb->rx_jumbo_skbuff[i].skb;
+ if (skb) {
+ struct ring_info *ringp;
+ dma_addr_t mapping;
-int __init ace_module_init(void)
-{
- int status;
+ ringp = &ap->skb->rx_jumbo_skbuff[i];
+ mapping = pci_unmap_addr(ringp, mapping);
+ pci_unmap_page(ap->pdev, mapping,
+ ACE_JUMBO_BUFSIZE - (2 + 16),
+ PCI_DMA_FROMDEVICE);
- root_dev = NULL;
+ ap->rx_jumbo_ring[i].size = 0;
+ ap->skb->rx_jumbo_skbuff[i].skb = NULL;
+ dev_kfree_skb(skb);
+ }
+ }
-#ifdef NEW_NETINIT
- status = acenic_probe();
-#else
- status = acenic_probe(NULL);
-#endif
- return status;
+ ace_init_cleanup(dev);
+ free_netdev(dev);
}
+static struct pci_driver acenic_pci_driver = {
+ .name = "acenic",
+ .id_table = acenic_pci_tbl,
+ .probe = acenic_probe_one,
+ .remove = __devexit_p(acenic_remove_one),
+};
-#if (LINUX_VERSION_CODE < 0x02032a)
-#ifdef MODULE
-int init_module(void)
+static int __init acenic_init(void)
{
- return ace_module_init();
+ return pci_module_init(&acenic_pci_driver);
}
-
-void cleanup_module(void)
+static void __exit acenic_exit(void)
{
- ace_module_cleanup();
+ pci_unregister_driver(&acenic_pci_driver);
}
-#endif
-#else
-module_init(ace_module_init);
-module_exit(ace_module_cleanup);
-#endif
+module_init(acenic_init);
+module_exit(acenic_exit);
static void ace_free_descriptors(struct net_device *dev)
{
@@ -1462,13 +1203,6 @@ static int __init ace_init(struct net_de
} else
dev->irq = pdev->irq;
- /*
- * Register the device here to be able to catch allocated
- * interrupt handlers in case the firmware doesn't come up.
- */
- ap->next = root_dev;
- root_dev = dev;
-
#ifdef INDEX_DEBUG
spin_lock_init(&ap->debug_lock);
ap->last_tx = ACE_TX_RING_ENTRIES(ap) - 1;
@@ -2642,8 +2376,6 @@ static int ace_open(struct net_device *d
netif_start_queue(dev);
- ACE_MOD_INC_USE_COUNT;
-
/*
* Setup the bottom half rx ring refill handler
*/
@@ -2660,8 +2392,6 @@ static int ace_close(struct net_device *
unsigned long flags;
short i;
- ace_if_down(dev);
-
/*
* Without (or before) releasing irq and stopping hardware, this
* is an absolute non-sense, by the way. It will be reset instantly
@@ -2733,7 +2463,6 @@ static int ace_close(struct net_device *
ace_unmask_irq(dev);
local_irq_restore(flags);
- ACE_MOD_DEC_USE_COUNT;
return 0;
}
@@ -2790,12 +2519,6 @@ static int ace_start_xmit(struct sk_buff
struct tx_desc *desc;
u32 idx, flagsize;
- /*
- * This only happens with pre-softnet, ie. 2.2.x kernels.
- */
- if (early_stop_netif_stop_queue(dev))
- return 1;
-
restart:
idx = ap->tx_prd;
--- diff/drivers/net/ibmveth.c 2004-03-11 10:20:25.000000000 +0000
+++ source/drivers/net/ibmveth.c 2004-03-16 11:27:48.188861840 +0000
@@ -527,7 +527,10 @@ static int ibmveth_open(struct net_devic
ibmveth_debug_printk("registering irq 0x%x\n", netdev->irq);
if((rc = request_irq(netdev->irq, &ibmveth_interrupt, 0, netdev->name, netdev)) != 0) {
ibmveth_error_printk("unable to request irq 0x%x, rc %d\n", netdev->irq, rc);
- h_free_logical_lan(adapter->vdev->unit_address);
+ do {
+ rc = h_free_logical_lan(adapter->vdev->unit_address);
+ } while H_isLongBusy(rc);
+
ibmveth_cleanup(adapter);
return rc;
}
@@ -556,7 +559,9 @@ static int ibmveth_close(struct net_devi
cancel_delayed_work(&adapter->replenish_task);
flush_scheduled_work();
- lpar_rc = h_free_logical_lan(adapter->vdev->unit_address);
+ do {
+ lpar_rc = h_free_logical_lan(adapter->vdev->unit_address);
+ } while H_isLongBusy(lpar_rc);
if(lpar_rc != H_Success)
{
@@ -617,6 +622,8 @@ static int ibmveth_start_xmit(struct sk_
union ibmveth_buf_desc desc[IbmVethMaxSendFrags];
unsigned long lpar_rc;
int nfrags = 0, curfrag;
+ unsigned long correlator;
+ unsigned int retry_count;
if ((skb_shinfo(skb)->nr_frags + 1) > IbmVethMaxSendFrags) {
adapter->stats.tx_dropped++;
@@ -674,8 +681,8 @@ static int ibmveth_start_xmit(struct sk_
}
/* send the frame. Arbitrarily set retrycount to 1024 */
- unsigned long correlator = 0;
- unsigned int retry_count = 1024;
+ correlator = 0;
+ retry_count = 1024;
do {
lpar_rc = h_send_logical_lan(adapter->vdev->unit_address,
desc[0].desc,
--- diff/drivers/net/pcmcia/3c574_cs.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/net/pcmcia/3c574_cs.c 2004-03-16 11:27:48.191861384 +0000
@@ -384,6 +384,8 @@ static void tc574_detach(dev_link_t *lin
#define CS_CHECK(fn, ret) \
do { last_fn = (fn); if ((last_ret = (ret)) != 0) goto cs_failed; } while (0)
+static char *ram_split[] = {"5:3", "3:1", "1:1", "3:5"};
+
static void tc574_config(dev_link_t *link)
{
client_handle_t handle = link->handle;
@@ -396,6 +398,7 @@ static void tc574_config(dev_link_t *lin
ioaddr_t ioaddr;
u16 *phys_addr;
char *cardname;
+ union wn3_config config;
phys_addr = (u16 *)dev->dev_addr;
@@ -431,15 +434,7 @@ static void tc574_config(dev_link_t *lin
dev->irq = link->irq.AssignedIRQ;
dev->base_addr = link->io.BasePort1;
- if (register_netdev(dev) != 0) {
- printk(KERN_NOTICE "3c574_cs: register_netdev() failed\n");
- goto failed;
- }
-
ioaddr = dev->base_addr;
- strcpy(lp->node.dev_name, dev->name);
- link->dev = &lp->node;
- link->state &= ~DEV_CONFIG_PENDING;
/* The 3c574 normally uses an EEPROM for configuration info, including
the hardware address. The future products may include a modem chip
@@ -467,24 +462,14 @@ static void tc574_config(dev_link_t *lin
} else
cardname = "3Com 3c574";
- printk(KERN_INFO "%s: %s at io %#3lx, irq %d, hw_addr ",
- dev->name, cardname, dev->base_addr, dev->irq);
-
- for (i = 0; i < 6; i++)
- printk("%02X%s", dev->dev_addr[i], ((i<5) ? ":" : ".\n"));
-
{
- u_char mcr, *ram_split[] = {"5:3", "3:1", "1:1", "3:5"};
- union wn3_config config;
+ u_char mcr;
outw(2<<11, ioaddr + RunnerRdCtrl);
mcr = inb(ioaddr + 2);
outw(0<<11, ioaddr + RunnerRdCtrl);
printk(KERN_INFO " ASIC rev %d,", mcr>>3);
EL3WINDOW(3);
config.i = inl(ioaddr + Wn3_Config);
- printk(" %dK FIFO split %s Rx:Tx, %sMII interface.\n",
- 8 << config.u.ram_size, ram_split[config.u.ram_split],
- config.u.autoselect ? "autoselect " : "");
lp->default_media = config.u.xcvr;
lp->autoselect = config.u.autoselect;
}
@@ -531,6 +516,25 @@ static void tc574_config(dev_link_t *lin
}
}
+ link->state &= ~DEV_CONFIG_PENDING;
+ link->dev = &lp->node;
+
+ if (register_netdev(dev) != 0) {
+ printk(KERN_NOTICE "3c574_cs: register_netdev() failed\n");
+ link->dev = NULL;
+ goto failed;
+ }
+
+ strcpy(lp->node.dev_name, dev->name);
+
+ printk(KERN_INFO "%s: %s at io %#3lx, irq %d, hw_addr ",
+ dev->name, cardname, dev->base_addr, dev->irq);
+ for (i = 0; i < 6; i++)
+ printk("%02X%s", dev->dev_addr[i], ((i<5) ? ":" : ".\n"));
+ printk(" %dK FIFO split %s Rx:Tx, %sMII interface.\n",
+ 8 << config.u.ram_size, ram_split[config.u.ram_split],
+ config.u.autoselect ? "autoselect " : "");
+
return;
cs_failed:
--- diff/drivers/net/pcmcia/3c589_cs.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/net/pcmcia/3c589_cs.c 2004-03-16 11:27:48.193861080 +0000
@@ -308,7 +308,7 @@ static void tc589_config(dev_link_t *lin
tuple_t tuple;
cisparse_t parse;
u16 buf[32], *phys_addr;
- int last_fn, last_ret, i, j, multi = 0;
+ int last_fn, last_ret, i, j, multi = 0, fifo;
ioaddr_t ioaddr;
char *ram_split[] = {"5:3", "3:1", "1:1", "3:5"};
@@ -357,11 +357,6 @@ static void tc589_config(dev_link_t *lin
dev->irq = link->irq.AssignedIRQ;
dev->base_addr = link->io.BasePort1;
- if (register_netdev(dev) != 0) {
- printk(KERN_ERR "3c589_cs: register_netdev() failed\n");
- goto failed;
- }
-
ioaddr = dev->base_addr;
EL3WINDOW(0);
@@ -382,13 +377,10 @@ static void tc589_config(dev_link_t *lin
}
}
- strcpy(lp->node.dev_name, dev->name);
- link->dev = &lp->node;
- link->state &= ~DEV_CONFIG_PENDING;
-
/* The address and resource configuration register aren't loaded from
the EEPROM and *must* be set to 0 and IRQ3 for the PCMCIA version. */
outw(0x3f00, ioaddr + 8);
+ fifo = inl(ioaddr);
/* The if_port symbol can be set when the module is loaded */
if ((if_port >= 0) && (if_port <= 3))
@@ -396,14 +388,24 @@ static void tc589_config(dev_link_t *lin
else
printk(KERN_ERR "3c589_cs: invalid if_port requested\n");
+ link->dev = &lp->node;
+ link->state &= ~DEV_CONFIG_PENDING;
+
+ if (register_netdev(dev) != 0) {
+ printk(KERN_ERR "3c589_cs: register_netdev() failed\n");
+ link->dev = NULL;
+ goto failed;
+ }
+
+ strcpy(lp->node.dev_name, dev->name);
+
printk(KERN_INFO "%s: 3Com 3c%s, io %#3lx, irq %d, hw_addr ",
dev->name, (multi ? "562" : "589"), dev->base_addr,
dev->irq);
for (i = 0; i < 6; i++)
printk("%02X%s", dev->dev_addr[i], ((i<5) ? ":" : "\n"));
- i = inl(ioaddr);
printk(KERN_INFO " %dK FIFO split %s Rx:Tx, %s xcvr\n",
- (i & 7) ? 32 : 8, ram_split[(i >> 16) & 3],
+ (fifo & 7) ? 32 : 8, ram_split[(fifo >> 16) & 3],
if_names[dev->if_port]);
return;
--- diff/drivers/net/pcmcia/axnet_cs.c 2004-02-18 08:54:10.000000000 +0000
+++ source/drivers/net/pcmcia/axnet_cs.c 2004-03-16 11:27:48.194860928 +0000
@@ -430,19 +430,11 @@ static void axnet_config(dev_link_t *lin
ei_status.block_input = &block_input;
ei_status.block_output = &block_output;
- strcpy(info->node.dev_name, dev->name);
-
if (inb(dev->base_addr + AXNET_TEST) != 0)
info->flags |= IS_AX88790;
else
info->flags |= IS_AX88190;
- printk(KERN_INFO "%s: Asix AX88%d90: io %#3lx, irq %d, hw_addr ",
- dev->name, ((info->flags & IS_AX88790) ? 7 : 1),
- dev->base_addr, dev->irq);
- for (i = 0; i < 6; i++)
- printk("%02X%s", dev->dev_addr[i], ((i<5) ? ":" : "\n"));
-
if (info->flags & IS_AX88790)
outb(0x10, dev->base_addr + AXNET_GPIO); /* select Internal PHY */
@@ -463,19 +455,27 @@ static void axnet_config(dev_link_t *lin
}
info->phy_id = (i < 32) ? i : -1;
- if (i < 32) {
- DEBUG(0, " MII transceiver at index %d, status %x.\n", i, j);
- } else {
- printk(KERN_NOTICE " No MII transceivers found!\n");
- }
+ link->dev = &info->node;
+ link->state &= ~DEV_CONFIG_PENDING;
if (register_netdev(dev) != 0) {
printk(KERN_NOTICE "axnet_cs: register_netdev() failed\n");
+ link->dev = NULL;
goto failed;
}
- link->dev = &info->node;
- link->state &= ~DEV_CONFIG_PENDING;
+ strcpy(info->node.dev_name, dev->name);
+
+ printk(KERN_INFO "%s: Asix AX88%d90: io %#3lx, irq %d, hw_addr ",
+ dev->name, ((info->flags & IS_AX88790) ? 7 : 1),
+ dev->base_addr, dev->irq);
+ for (i = 0; i < 6; i++)
+ printk("%02X%s", dev->dev_addr[i], ((i<5) ? ":" : "\n"));
+ if (info->phy_id != -1) {
+ DEBUG(0, " MII transceiver at index %d, status %x.\n", info->phy_id, j);
+ } else {
+ printk(KERN_NOTICE " No MII transceivers found!\n");
+ }
return;
cs_failed:
--- diff/drivers/net/pcmcia/com20020_cs.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/net/pcmcia/com20020_cs.c 2004-03-16 11:27:48.195860776 +0000
@@ -141,7 +141,6 @@ static dev_link_t *dev_list;
typedef struct com20020_dev_t {
struct net_device *dev;
- int dev_configured;
dev_node_t node;
} com20020_dev_t;
@@ -277,13 +276,10 @@ static void com20020_detach(dev_link_t *
dev = info->dev;
if (dev)
{
- if (info->dev_configured)
+ if (link->dev)
{
DEBUG(1,"unregister...\n");
- if (netif_running(dev))
- dev->stop(dev);
-
unregister_netdev(dev);
/*
@@ -398,17 +394,18 @@ static void com20020_config(dev_link_t *
lp->card_name = "PCMCIA COM20020";
lp->card_flags = ARC_CAN_10MBIT; /* pretend all of them can 10Mbit */
+ link->dev = &info->node;
+ link->state &= ~DEV_CONFIG_PENDING;
+
i = com20020_found(dev, 0); /* calls register_netdev */
if (i != 0) {
DEBUG(1,KERN_NOTICE "com20020_cs: com20020_found() failed\n");
+ link->dev = NULL;
goto failed;
}
- info->dev_configured = 1;
strcpy(info->node.dev_name, dev->name);
- link->dev = &info->node;
- link->state &= ~DEV_CONFIG_PENDING;
DEBUG(1,KERN_INFO "%s: port %#3lx, irq %d\n",
dev->name, dev->base_addr, dev->irq);
--- diff/drivers/net/pcmcia/fmvj18x_cs.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/net/pcmcia/fmvj18x_cs.c 2004-03-16 11:27:48.197860472 +0000
@@ -510,10 +510,6 @@ static void fmvj18x_config(dev_link_t *l
CS_CHECK(RequestConfiguration, pcmcia_request_configuration(link->handle, &link->conf));
dev->irq = link->irq.AssignedIRQ;
dev->base_addr = link->io.BasePort1;
- if (register_netdev(dev) != 0) {
- printk(KERN_NOTICE "fmvj18x_cs: register_netdev() failed\n");
- goto failed;
- }
if (link->io.BasePort2 != 0)
fmvj18x_setup_mfc(link);
@@ -575,7 +571,6 @@ static void fmvj18x_config(dev_link_t *l
/* Read MACID from Buggy CIS */
if (fmvj18x_get_hwinfo(link, tuple.TupleData) == -1) {
printk(KERN_NOTICE "fmvj18x_cs: unable to read hardware net address.\n");
- unregister_netdev(dev);
goto failed;
}
for (i = 0 ; i < 6; i++) {
@@ -592,10 +587,18 @@ static void fmvj18x_config(dev_link_t *l
break;
}
- strcpy(lp->node.dev_name, dev->name);
+ lp->cardtype = cardtype;
link->dev = &lp->node;
+ link->state &= ~DEV_CONFIG_PENDING;
+
+ if (register_netdev(dev) != 0) {
+ printk(KERN_NOTICE "fmvj18x_cs: register_netdev() failed\n");
+ link->dev = NULL;
+ goto failed;
+ }
+
+ strcpy(lp->node.dev_name, dev->name);
- lp->cardtype = cardtype;
/* print current configuration */
printk(KERN_INFO "%s: %s, sram %s, port %#3lx, irq %d, hw_addr ",
dev->name, card_name, sram_config == 0 ? "4K TX*2" : "8K TX*2",
@@ -603,7 +606,6 @@ static void fmvj18x_config(dev_link_t *l
for (i = 0; i < 6; i++)
printk("%02X%s", dev->dev_addr[i], ((i<5) ? ":" : "\n"));
- link->state &= ~DEV_CONFIG_PENDING;
return;
cs_failed:
--- diff/drivers/net/pcmcia/ibmtr_cs.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/net/pcmcia/ibmtr_cs.c 2004-03-16 11:27:48.198860320 +0000
@@ -317,13 +317,10 @@ static void ibmtr_config(dev_link_t *lin
/* Try PRIMARY card at 0xA20-0xA23 */
link->io.BasePort1 = 0xA20;
i = pcmcia_request_io(link->handle, &link->io);
- if (i == CS_SUCCESS) {
- memcpy(info->node.dev_name, "tr0\0", 4);
- } else {
+ if (i != CS_SUCCESS) {
/* Couldn't get 0xA20-0xA23. Try ALTERNATE at 0xA24-0xA27. */
link->io.BasePort1 = 0xA24;
CS_CHECK(RequestIO, pcmcia_request_io(link->handle, &link->io));
- memcpy(info->node.dev_name, "tr1\0", 4);
}
dev->base_addr = link->io.BasePort1;
@@ -367,15 +364,17 @@ static void ibmtr_config(dev_link_t *lin
Adapters Technical Reference" SC30-3585 for this info. */
ibmtr_hw_setup(dev, mmiobase);
+ link->dev = &info->node;
+ link->state &= ~DEV_CONFIG_PENDING;
+
i = ibmtr_probe_card(dev);
-
if (i != 0) {
printk(KERN_NOTICE "ibmtr_cs: register_netdev() failed\n");
+ link->dev = NULL;
goto failed;
}
- link->dev = &info->node;
- link->state &= ~DEV_CONFIG_PENDING;
+ strcpy(info->node.dev_name, dev->name);
printk(KERN_INFO "%s: port %#3lx, irq %d,",
dev->name, dev->base_addr, dev->irq);
--- diff/drivers/net/pcmcia/nmclan_cs.c 2004-02-18 08:54:10.000000000 +0000
+++ source/drivers/net/pcmcia/nmclan_cs.c 2004-03-16 11:27:48.199860168 +0000
@@ -734,11 +734,6 @@ static void nmclan_config(dev_link_t *li
CS_CHECK(RequestConfiguration, pcmcia_request_configuration(handle, &link->conf));
dev->irq = link->irq.AssignedIRQ;
dev->base_addr = link->io.BasePort1;
- i = register_netdev(dev);
- if (i != 0) {
- printk(KERN_NOTICE "nmclan_cs: register_netdev() failed\n");
- goto failed;
- }
ioaddr = dev->base_addr;
@@ -777,10 +772,18 @@ static void nmclan_config(dev_link_t *li
else
printk(KERN_NOTICE "nmclan_cs: invalid if_port requested\n");
- strcpy(lp->node.dev_name, dev->name);
link->dev = &lp->node;
link->state &= ~DEV_CONFIG_PENDING;
+ i = register_netdev(dev);
+ if (i != 0) {
+ printk(KERN_NOTICE "nmclan_cs: register_netdev() failed\n");
+ link->dev = NULL;
+ goto failed;
+ }
+
+ strcpy(lp->node.dev_name, dev->name);
+
printk(KERN_INFO "%s: nmclan: port %#3lx, irq %d, %s port, hw_addr ",
dev->name, dev->base_addr, dev->irq, if_names[dev->if_port]);
for (i = 0; i < 6; i++)
--- diff/drivers/net/pcmcia/smc91c92_cs.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/net/pcmcia/smc91c92_cs.c 2004-03-16 11:27:48.201859864 +0000
@@ -901,6 +901,7 @@ static void smc91c92_config(dev_link_t *
char *name;
int i, j, rev;
ioaddr_t ioaddr;
+ u_long mir;
DEBUG(0, "smc91c92_config(0x%p)\n", link);
@@ -952,11 +953,6 @@ static void smc91c92_config(dev_link_t *
else
printk(KERN_NOTICE "smc91c92_cs: invalid if_port requested\n");
- if (register_netdev(dev) != 0) {
- printk(KERN_ERR "smc91c92_cs: register_netdev() failed\n");
- goto config_undo;
- }
-
switch (smc->manfid) {
case MANFID_OSITECH:
case MANFID_PSION:
@@ -977,8 +973,6 @@ static void smc91c92_config(dev_link_t *
goto config_undo;
}
- strcpy(smc->node.dev_name, dev->name);
- link->dev = &smc->node;
smc->duplex = 0;
smc->rx_ovrn = 0;
@@ -993,25 +987,16 @@ static void smc91c92_config(dev_link_t *
case 8: name = "100-FD"; break;
case 9: name = "110"; break;
}
- printk(KERN_INFO "%s: smc91c%s rev %d: io %#3lx, irq %d, "
- "hw_addr ", dev->name, name, (rev & 0x0f), dev->base_addr,
- dev->irq);
- for (i = 0; i < 6; i++)
- printk("%02X%s", dev->dev_addr[i], ((i<5) ? ":" : "\n"));
ioaddr = dev->base_addr;
if (rev > 0) {
- u_long mir, mcr;
+ u_long mcr;
SMC_SELECT_BANK(0);
mir = inw(ioaddr + MEMINFO) & 0xff;
if (mir == 0xff) mir++;
/* Get scale factor for memory size */
mcr = ((rev >> 4) > 3) ? inw(ioaddr + MEMCFG) : 0x0200;
mir *= 128 * (1<<((mcr >> 9) & 7));
- if (mir & 0x3ff)
- printk(KERN_INFO " %lu byte", mir);
- else
- printk(KERN_INFO " %lu kb", mir>>10);
SMC_SELECT_BANK(1);
smc->cfg = inw(ioaddr + CONFIG) & ~CFG_AUI_SELECT;
smc->cfg |= CFG_NO_WAIT | CFG_16BIT | CFG_STATIC;
@@ -1019,9 +1004,8 @@ static void smc91c92_config(dev_link_t *
smc->cfg |= CFG_IRQ_SEL_1 | CFG_IRQ_SEL_0;
if ((rev >> 4) >= 7)
smc->cfg |= CFG_MII_SELECT;
- printk(" buffer, %s xcvr\n", (smc->cfg & CFG_MII_SELECT) ?
- "MII" : if_names[dev->if_port]);
- }
+ } else
+ mir = 0;
if (smc->cfg & CFG_MII_SELECT) {
SMC_SELECT_BANK(3);
@@ -1031,16 +1015,45 @@ static void smc91c92_config(dev_link_t *
if ((j != 0) && (j != 0xffff)) break;
}
smc->mii_if.phy_id = (i < 32) ? i : -1;
- if (i < 32) {
- DEBUG(0, " MII transceiver at index %d, status %x.\n", i, j);
- } else {
- printk(KERN_NOTICE " No MII transceivers found!\n");
- }
SMC_SELECT_BANK(0);
}
+ link->dev = &smc->node;
link->state &= ~DEV_CONFIG_PENDING;
+
+ if (register_netdev(dev) != 0) {
+ printk(KERN_ERR "smc91c92_cs: register_netdev() failed\n");
+ link->dev = NULL;
+ goto config_undo;
+ }
+
+ strcpy(smc->node.dev_name, dev->name);
+
+ printk(KERN_INFO "%s: smc91c%s rev %d: io %#3lx, irq %d, "
+ "hw_addr ", dev->name, name, (rev & 0x0f), dev->base_addr,
+ dev->irq);
+ for (i = 0; i < 6; i++)
+ printk("%02X%s", dev->dev_addr[i], ((i<5) ? ":" : "\n"));
+
+ if (rev > 0) {
+ if (mir & 0x3ff)
+ printk(KERN_INFO " %lu byte", mir);
+ else
+ printk(KERN_INFO " %lu kb", mir>>10);
+ printk(" buffer, %s xcvr\n", (smc->cfg & CFG_MII_SELECT) ?
+ "MII" : if_names[dev->if_port]);
+ }
+
+ if (smc->cfg & CFG_MII_SELECT) {
+ if (smc->mii_if.phy_id != -1) {
+ DEBUG(0, " MII transceiver at index %d, status %x.\n",
+ smc->mii_if.phy_id, j);
+ } else {
+ printk(KERN_NOTICE " No MII transceivers found!\n");
+ }
+ }
+
return;
config_undo:
--- diff/drivers/net/pcmcia/xirc2ps_cs.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/net/pcmcia/xirc2ps_cs.c 2004-03-16 11:27:48.202859712 +0000
@@ -1114,17 +1114,20 @@ xirc2ps_config(dev_link_t * link)
/* we can now register the device with the net subsystem */
dev->irq = link->irq.AssignedIRQ;
dev->base_addr = link->io.BasePort1;
+
+ if (local->dingo)
+ do_reset(dev, 1); /* a kludge to make the cem56 work */
+
+ link->dev = &local->node;
+ link->state &= ~DEV_CONFIG_PENDING;
+
if ((err=register_netdev(dev))) {
printk(KNOT_XIRC "register_netdev() failed\n");
+ link->dev = NULL;
goto config_error;
}
strcpy(local->node.dev_name, dev->name);
- link->dev = &local->node;
- link->state &= ~DEV_CONFIG_PENDING;
-
- if (local->dingo)
- do_reset(dev, 1); /* a kludge to make the cem56 work */
/* give some infos about the hardware */
printk(KERN_INFO "%s: %s: port %#3lx, irq %d, hwaddr",
--- diff/drivers/net/r8169.c 2004-03-11 10:20:26.000000000 +0000
+++ source/drivers/net/r8169.c 2004-03-16 11:27:48.206859104 +0000
@@ -56,9 +56,11 @@ VERSION 1.2 <2002/11/30>
printk( "Assertion failed! %s,%s,%s,line=%d\n", \
#expr,__FILE__,__FUNCTION__,__LINE__); \
}
+#define dprintk(fmt, args...) do { printk(PFX fmt, ## args) } while (0)
#else
#define assert(expr) do {} while (0)
-#endif
+#define dprintk(fmt, args...) do {} while (0)
+#endif /* RTL8169_DEBUG */
/* media options */
#define MAX_UNITS 8
@@ -89,9 +91,12 @@ static int multicast_filter_limit = 32;
#define NUM_TX_DESC 64 /* Number of Tx descriptor registers */
#define NUM_RX_DESC 64 /* Number of Rx descriptor registers */
#define RX_BUF_SIZE 1536 /* Rx Buffer size */
+#define R8169_TX_RING_BYTES (NUM_TX_DESC * sizeof(struct TxDesc))
+#define R8169_RX_RING_BYTES (NUM_RX_DESC * sizeof(struct RxDesc))
#define RTL_MIN_IO_SIZE 0x80
-#define TX_TIMEOUT (6*HZ)
+#define RTL8169_TX_TIMEOUT (6*HZ)
+#define RTL8169_PHY_TIMEOUT (HZ)
/* write/read MMIO register */
#define RTL_W8(reg, val8) writeb ((val8), ioaddr + (reg))
@@ -101,11 +106,35 @@ static int multicast_filter_limit = 32;
#define RTL_R16(reg) readw (ioaddr + (reg))
#define RTL_R32(reg) ((unsigned long) readl (ioaddr + (reg)))
-static struct {
+enum mac_version {
+ RTL_GIGA_MAC_VER_B = 0x00,
+ /* RTL_GIGA_MAC_VER_C = 0x03, */
+ RTL_GIGA_MAC_VER_D = 0x01,
+ RTL_GIGA_MAC_VER_E = 0x02
+};
+
+enum phy_version {
+ RTL_GIGA_PHY_VER_C = 0x03, /* PHY Reg 0x03 bit0-3 == 0x0000 */
+ RTL_GIGA_PHY_VER_D = 0x04, /* PHY Reg 0x03 bit0-3 == 0x0000 */
+ RTL_GIGA_PHY_VER_E = 0x05, /* PHY Reg 0x03 bit0-3 == 0x0000 */
+ RTL_GIGA_PHY_VER_F = 0x06, /* PHY Reg 0x03 bit0-3 == 0x0001 */
+ RTL_GIGA_PHY_VER_G = 0x07, /* PHY Reg 0x03 bit0-3 == 0x0002 */
+};
+
+
+#define _R(NAME,MAC,MASK) \
+ { .name = NAME, .mac_version = MAC, .RxConfigMask = MASK }
+
+const static struct {
const char *name;
-} board_info[] __devinitdata = {
- {
-"RealTek RTL8169 Gigabit Ethernet"},};
+ u8 mac_version;
+ u32 RxConfigMask; /* Clears the bits supported by this chip */
+} rtl_chip_info[] = {
+ _R("RTL8169", RTL_GIGA_MAC_VER_B, 0xff7e1880),
+ _R("RTL8169s/8110s", RTL_GIGA_MAC_VER_D, 0xff7e1880),
+ _R("RTL8169s/8110s", RTL_GIGA_MAC_VER_E, 0xff7e1880)
+};
+#undef _R
static struct pci_device_id rtl8169_pci_tbl[] = {
{0x10ec, 0x8169, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
@@ -114,6 +143,8 @@ static struct pci_device_id rtl8169_pci_
MODULE_DEVICE_TABLE(pci, rtl8169_pci_tbl);
+static int rx_copybreak = 200;
+
enum RTL8169_registers {
MAC0 = 0, /* Ethernet hardware address. */
MAR0 = 8, /* Multicast filter. */
@@ -242,14 +273,6 @@ enum RTL8169_register_content {
TBILinkOK = 0x02000000,
};
-const static struct {
- const char *name;
- u8 version; /* depend on RTL8169 docs */
- u32 RxConfigMask; /* should clear the bits supported by this chip */
-} rtl_chip_info[] = {
- {
-"RTL-8169", 0x00, 0xff7e1880,},};
-
enum _DescStatusBit {
OWNbit = 0x80000000,
EORbit = 0x40000000,
@@ -257,6 +280,8 @@ enum _DescStatusBit {
LSbit = 0x10000000,
};
+#define RsvdMask 0x3fffc000
+
struct TxDesc {
u32 status;
u32 vlan_tag;
@@ -277,28 +302,33 @@ struct rtl8169_private {
struct net_device_stats stats; /* statistics of net device */
spinlock_t lock; /* spin lock flag */
int chipset;
- unsigned long cur_rx; /* Index into the Rx descriptor buffer of next Rx pkt. */
- unsigned long cur_tx; /* Index into the Tx descriptor buffer of next Rx pkt. */
- unsigned long dirty_tx;
- unsigned char *TxDescArrays; /* Index of Tx Descriptor buffer */
- unsigned char *RxDescArrays; /* Index of Rx Descriptor buffer */
+ int mac_version;
+ int phy_version;
+ u32 cur_rx; /* Index into the Rx descriptor buffer of next Rx pkt. */
+ u32 cur_tx; /* Index into the Tx descriptor buffer of next Rx pkt. */
+ u32 dirty_rx;
+ u32 dirty_tx;
struct TxDesc *TxDescArray; /* Index of 256-alignment Tx Descriptor buffer */
struct RxDesc *RxDescArray; /* Index of 256-alignment Rx Descriptor buffer */
- unsigned char *RxBufferRings; /* Index of Rx Buffer */
- unsigned char *RxBufferRing[NUM_RX_DESC]; /* Index of Rx Buffer array */
+ dma_addr_t TxPhyAddr;
+ dma_addr_t RxPhyAddr;
+ struct sk_buff *Rx_skbuff[NUM_RX_DESC]; /* Rx data buffers */
struct sk_buff *Tx_skbuff[NUM_TX_DESC]; /* Index of Transmit data buffer */
+ struct timer_list timer;
+ unsigned long phy_link_down_cnt;
};
MODULE_AUTHOR("Realtek");
MODULE_DESCRIPTION("RealTek RTL-8169 Gigabit Ethernet driver");
MODULE_PARM(media, "1-" __MODULE_STRING(MAX_UNITS) "i");
+MODULE_PARM(rx_copybreak, "i");
MODULE_LICENSE("GPL");
static int rtl8169_open(struct net_device *dev);
static int rtl8169_start_xmit(struct sk_buff *skb, struct net_device *dev);
static irqreturn_t rtl8169_interrupt(int irq, void *dev_instance,
struct pt_regs *regs);
-static void rtl8169_init_ring(struct net_device *dev);
+static int rtl8169_init_ring(struct net_device *dev);
static void rtl8169_hw_start(struct net_device *dev);
static int rtl8169_close(struct net_device *dev);
static void rtl8169_set_rx_mode(struct net_device *dev);
@@ -306,11 +336,15 @@ static void rtl8169_tx_timeout(struct ne
static struct net_device_stats *rtl8169_get_stats(struct net_device *netdev);
static const u16 rtl8169_intr_mask =
- SYSErr | PCSTimeout | RxUnderrun | RxOverflow | RxFIFOOver | TxErr | TxOK |
- RxErr | RxOK;
+ RxUnderrun | RxOverflow | RxFIFOOver | TxErr | TxOK | RxErr | RxOK;
static const unsigned int rtl8169_rx_config =
(RX_FIFO_THRESH << RxCfgFIFOShift) | (RX_DMA_BURST << RxCfgDMAShift);
+#define PHY_Cap_10_Half_Or_Less PHY_Cap_10_Half
+#define PHY_Cap_10_Full_Or_Less PHY_Cap_10_Full | PHY_Cap_10_Half_Or_Less
+#define PHY_Cap_100_Half_Or_Less PHY_Cap_100_Half | PHY_Cap_10_Full_Or_Less
+#define PHY_Cap_100_Full_Or_Less PHY_Cap_100_Full | PHY_Cap_100_Half_Or_Less
+
void
mdio_write(void *ioaddr, int RegAddr, int value)
{
@@ -342,13 +376,258 @@ mdio_read(void *ioaddr, int RegAddr)
if (RTL_R32(PHYAR) & 0x80000000) {
value = (int) (RTL_R32(PHYAR) & 0xFFFF);
break;
- } else {
- udelay(100);
}
+ udelay(100);
}
return value;
}
+static void rtl8169_write_gmii_reg_bit(void *ioaddr, int reg, int bitnum,
+ int bitval)
+{
+ int val;
+
+ val = mdio_read(ioaddr, reg);
+ val = (bitval == 1) ?
+ val | (bitval << bitnum) : val & ~(0x0001 << bitnum);
+ mdio_write(ioaddr, reg, val & 0xffff);
+}
+
+static void rtl8169_get_mac_version(struct rtl8169_private *tp, void *ioaddr)
+{
+ const struct {
+ u32 mask;
+ int mac_version;
+ } mac_info[] = {
+ { 0x1 << 26, RTL_GIGA_MAC_VER_E },
+ { 0x1 << 23, RTL_GIGA_MAC_VER_D },
+ { 0x00000000, RTL_GIGA_MAC_VER_B } /* Catch-all */
+ }, *p = mac_info;
+ u32 reg;
+
+ reg = RTL_R32(TxConfig) & 0x7c800000;
+ while ((reg & p->mask) != p->mask)
+ p++;
+ tp->mac_version = p->mac_version;
+}
+
+static void rtl8169_print_mac_version(struct rtl8169_private *tp)
+{
+ struct {
+ int version;
+ char *msg;
+ } mac_print[] = {
+ { RTL_GIGA_MAC_VER_E, "RTL_GIGA_MAC_VER_E" },
+ { RTL_GIGA_MAC_VER_D, "RTL_GIGA_MAC_VER_D" },
+ { RTL_GIGA_MAC_VER_B, "RTL_GIGA_MAC_VER_B" },
+ { 0, NULL }
+ }, *p;
+
+ for (p = mac_print; p->msg; p++) {
+ if (tp->mac_version == p->version) {
+ dprintk("mac_version == %s (%04d)\n", p->msg,
+ p->version);
+ return;
+ }
+ }
+ dprintk("mac_version == Unknown\n");
+}
+
+static void rtl8169_get_phy_version(struct rtl8169_private *tp, void *ioaddr)
+{
+ const struct {
+ u16 mask;
+ u16 set;
+ int phy_version;
+ } phy_info[] = {
+ { 0x000f, 0x0002, RTL_GIGA_PHY_VER_G },
+ { 0x000f, 0x0001, RTL_GIGA_PHY_VER_F },
+ { 0x000f, 0x0000, RTL_GIGA_PHY_VER_E },
+ { 0x0000, 0x0000, RTL_GIGA_PHY_VER_D } /* Catch-all */
+ }, *p = phy_info;
+ u16 reg;
+
+ reg = mdio_read(ioaddr, 3) & 0xffff;
+ while ((reg & p->mask) != p->set)
+ p++;
+ tp->phy_version = p->phy_version;
+}
+
+static void rtl8169_print_phy_version(struct rtl8169_private *tp)
+{
+ struct {
+ int version;
+ char *msg;
+ u32 reg;
+ } phy_print[] = {
+ { RTL_GIGA_PHY_VER_G, "RTL_GIGA_PHY_VER_G", 0x0002 },
+ { RTL_GIGA_PHY_VER_F, "RTL_GIGA_PHY_VER_F", 0x0001 },
+ { RTL_GIGA_PHY_VER_E, "RTL_GIGA_PHY_VER_E", 0x0000 },
+ { RTL_GIGA_PHY_VER_D, "RTL_GIGA_PHY_VER_D", 0x0000 },
+ { 0, NULL, 0x0000 }
+ }, *p;
+
+ for (p = phy_print; p->msg; p++) {
+ if (tp->phy_version == p->version) {
+ dprintk("phy_version == %s (%04x)\n", p->msg, p->reg);
+ return;
+ }
+ }
+ dprintk("phy_version == Unknown\n");
+}
+
+static void rtl8169_hw_phy_config(struct net_device *dev)
+{
+ struct rtl8169_private *tp = dev->priv;
+ void *ioaddr = tp->mmio_addr;
+ struct {
+ u16 regs[5]; /* Beware of bit-sign propagation */
+ } phy_magic[5] = { {
+ { 0x0000, //w 4 15 12 0
+ 0x00a1, //w 3 15 0 00a1
+ 0x0008, //w 2 15 0 0008
+ 0x1020, //w 1 15 0 1020
+ 0x1000 } },{ //w 0 15 0 1000
+ { 0x7000, //w 4 15 12 7
+ 0xff41, //w 3 15 0 ff41
+ 0xde60, //w 2 15 0 de60
+ 0x0140, //w 1 15 0 0140
+ 0x0077 } },{ //w 0 15 0 0077
+ { 0xa000, //w 4 15 12 a
+ 0xdf01, //w 3 15 0 df01
+ 0xdf20, //w 2 15 0 df20
+ 0xff95, //w 1 15 0 ff95
+ 0xfa00 } },{ //w 0 15 0 fa00
+ { 0xb000, //w 4 15 12 b
+ 0xff41, //w 3 15 0 ff41
+ 0xde20, //w 2 15 0 de20
+ 0x0140, //w 1 15 0 0140
+ 0x00bb } },{ //w 0 15 0 00bb
+ { 0xf000, //w 4 15 12 f
+ 0xdf01, //w 3 15 0 df01
+ 0xdf20, //w 2 15 0 df20
+ 0xff95, //w 1 15 0 ff95
+ 0xbf00 } //w 0 15 0 bf00
+ }
+ }, *p = phy_magic;
+ int i;
+
+ rtl8169_print_mac_version(tp);
+ rtl8169_print_phy_version(tp);
+
+ if (tp->mac_version <= RTL_GIGA_MAC_VER_B)
+ return;
+ if (tp->phy_version >= RTL_GIGA_PHY_VER_F)
+ return;
+
+ dprintk("MAC version != 0 && PHY version == 0 or 1\n");
+ dprintk("Do final_reg2.cfg\n");
+
+ /* Shazam ! */
+
+ // phy config for RTL8169s mac_version C chip
+ mdio_write(ioaddr, 31, 0x0001); //w 31 2 0 1
+ mdio_write(ioaddr, 21, 0x1000); //w 21 15 0 1000
+ mdio_write(ioaddr, 24, 0x65c7); //w 24 15 0 65c7
+ rtl8169_write_gmii_reg_bit(ioaddr, 4, 11, 0); //w 4 11 11 0
+
+ for (i = 0; i < ARRAY_SIZE(phy_magic); i++, p++) {
+ int val, pos = 4;
+
+ val = (mdio_read(ioaddr, pos) & 0x0fff) | (p->regs[0] & 0xffff);
+ mdio_write(ioaddr, pos, val);
+ while (--pos >= 0)
+ mdio_write(ioaddr, pos, p->regs[4 - pos] & 0xffff);
+ rtl8169_write_gmii_reg_bit(ioaddr, 4, 11, 1); //w 4 11 11 1
+ rtl8169_write_gmii_reg_bit(ioaddr, 4, 11, 0); //w 4 11 11 0
+ }
+ mdio_write(ioaddr, 31, 0x0000); //w 31 2 0 0
+}
+
+static void rtl8169_hw_phy_reset(struct net_device *dev)
+{
+ struct rtl8169_private *tp = dev->priv;
+ void *ioaddr = tp->mmio_addr;
+ int i, val;
+
+ printk(KERN_WARNING PFX "%s: Reset RTL8169s PHY\n", dev->name);
+
+ val = (mdio_read(ioaddr, 0) | 0x8000) & 0xffff;
+ mdio_write(ioaddr, 0, val);
+
+ for (i = 50; i >= 0; i--) {
+ if (!(mdio_read(ioaddr, 0) & 0x8000))
+ break;
+ udelay(100); /* Gross */
+ }
+
+ if (i < 0) {
+ printk(KERN_WARNING PFX "%s: no PHY Reset ack. Giving up.\n",
+ dev->name);
+ }
+}
+
+static void rtl8169_phy_timer(unsigned long __opaque)
+{
+ struct net_device *dev = (struct net_device *)__opaque;
+ struct rtl8169_private *tp = dev->priv;
+ struct timer_list *timer = &tp->timer;
+ void *ioaddr = tp->mmio_addr;
+
+ assert(tp->mac_version > RTL_GIGA_MAC_VER_B);
+ assert(tp->phy_version < RTL_GIGA_PHY_VER_G);
+
+ if (RTL_R8(PHYstatus) & LinkStatus)
+ tp->phy_link_down_cnt = 0;
+ else {
+ tp->phy_link_down_cnt++;
+ if (tp->phy_link_down_cnt >= 12) {
+ int reg;
+
+ // If link on 1000, perform phy reset.
+ reg = mdio_read(ioaddr, PHY_1000_CTRL_REG);
+ if (reg & PHY_Cap_1000_Full)
+ rtl8169_hw_phy_reset(dev);
+
+ tp->phy_link_down_cnt = 0;
+ }
+ }
+
+ mod_timer(timer, RTL8169_PHY_TIMEOUT);
+}
+
+static inline void rtl8169_delete_timer(struct net_device *dev)
+{
+ struct rtl8169_private *tp = dev->priv;
+ struct timer_list *timer = &tp->timer;
+
+ if ((tp->mac_version <= RTL_GIGA_MAC_VER_B) ||
+ (tp->phy_version >= RTL_GIGA_PHY_VER_G))
+ return;
+
+ del_timer_sync(timer);
+
+ tp->phy_link_down_cnt = 0;
+}
+
+static inline void rtl8169_request_timer(struct net_device *dev)
+{
+ struct rtl8169_private *tp = dev->priv;
+ struct timer_list *timer = &tp->timer;
+
+ if ((tp->mac_version <= RTL_GIGA_MAC_VER_B) ||
+ (tp->phy_version >= RTL_GIGA_PHY_VER_G))
+ return;
+
+ tp->phy_link_down_cnt = 0;
+
+ init_timer(timer);
+ timer->expires = jiffies + RTL8169_PHY_TIMEOUT;
+ timer->data = (unsigned long)(dev);
+ timer->function = rtl8169_phy_timer;
+ add_timer(timer);
+}
+
static int __devinit
rtl8169_init_board(struct pci_dev *pdev, struct net_device **dev_out,
void **ioaddr_out)
@@ -356,9 +635,9 @@ rtl8169_init_board(struct pci_dev *pdev,
void *ioaddr = NULL;
struct net_device *dev;
struct rtl8169_private *tp;
- int rc, i;
unsigned long mmio_start, mmio_end, mmio_flags, mmio_len;
- u32 tmp;
+ int rc, i, acpi_idle_state = 0, pm_cap;
+
assert(pdev != NULL);
assert(ioaddr_out != NULL);
@@ -379,8 +658,22 @@ rtl8169_init_board(struct pci_dev *pdev,
// enable device (incl. PCI PM wakeup and hotplug setup)
rc = pci_enable_device(pdev);
- if (rc)
+ if (rc) {
+ printk(KERN_ERR PFX "%s: unable to enable device\n", pdev->slot_name);
goto err_out;
+ }
+
+ /* save power state before pci_enable_device overwrites it */
+ pm_cap = pci_find_capability(pdev, PCI_CAP_ID_PM);
+ if (pm_cap) {
+ u16 pwr_command;
+
+ pci_read_config_word(pdev, pm_cap + PCI_PM_CTRL, &pwr_command);
+ acpi_idle_state = pwr_command & PCI_PM_CTRL_STATE_MASK;
+ } else {
+ printk(KERN_ERR PFX "Cannot find PowerManagement capability, aborting.\n");
+ goto err_out_free_res;
+ }
mmio_start = pci_resource_start(pdev, 1);
mmio_end = pci_resource_end(pdev, 1);
@@ -402,8 +695,10 @@ rtl8169_init_board(struct pci_dev *pdev,
}
rc = pci_request_regions(pdev, dev->name);
- if (rc)
+ if (rc) {
+ printk(KERN_ERR PFX "%s: Could not request regions.\n", pdev->slot_name);
goto err_out_disable;
+ }
// enable PCI bus-mastering
pci_set_master(pdev);
@@ -420,30 +715,32 @@ rtl8169_init_board(struct pci_dev *pdev,
RTL_W8(ChipCmd, CmdReset);
// Check that the chip has finished the reset.
- for (i = 1000; i > 0; i--)
+ for (i = 1000; i > 0; i--) {
if ((RTL_R8(ChipCmd) & CmdReset) == 0)
break;
- else
- udelay(10);
+ udelay(10);
+ }
- // identify chip attached to board
- tmp = RTL_R32(TxConfig);
- tmp = ((tmp & 0x7c000000) + ((tmp & 0x00800000) << 2)) >> 24;
-
- for (i = ARRAY_SIZE(rtl_chip_info) - 1; i >= 0; i--)
- if (tmp == rtl_chip_info[i].version) {
- tp->chipset = i;
- goto match;
- }
- //if unknown chip, assume array element #0, original RTL-8169 in this case
- printk(KERN_DEBUG PFX
- "PCI device %s: unknown chip version, assuming RTL-8169\n",
- pci_name(pdev));
- printk(KERN_DEBUG PFX "PCI device %s: TxConfig = 0x%lx\n",
- pci_name(pdev), (unsigned long) RTL_R32(TxConfig));
- tp->chipset = 0;
+ // Identify chip attached to board
+ rtl8169_get_mac_version(tp, ioaddr);
+ rtl8169_get_phy_version(tp, ioaddr);
+
+ rtl8169_print_mac_version(tp);
+ rtl8169_print_phy_version(tp);
+
+ for (i = ARRAY_SIZE(rtl_chip_info) - 1; i >= 0; i--) {
+ if (tp->mac_version == rtl_chip_info[i].mac_version)
+ break;
+ }
+ if (i < 0) {
+ /* Unknown chip: assume array element #0, original RTL-8169 */
+ printk(KERN_DEBUG PFX
+ "PCI device %s: unknown chip version, assuming %s\n",
+ pci_name(pdev), rtl_chip_info[0].name);
+ i++;
+ }
+ tp->chipset = i;
-match:
*ioaddr_out = ioaddr;
*dev_out = dev;
return 0;
@@ -499,7 +796,7 @@ rtl8169_init_one(struct pci_dev *pdev, c
dev->stop = rtl8169_close;
dev->tx_timeout = rtl8169_tx_timeout;
dev->set_multicast_list = rtl8169_set_rx_mode;
- dev->watchdog_timeo = TX_TIMEOUT;
+ dev->watchdog_timeo = RTL8169_TX_TIMEOUT;
dev->irq = pdev->irq;
dev->base_addr = (unsigned long) ioaddr;
// dev->do_ioctl = mii_ioctl;
@@ -528,12 +825,29 @@ rtl8169_init_one(struct pci_dev *pdev, c
"%2.2x:%2.2x:%2.2x:%2.2x:%2.2x:%2.2x, "
"IRQ %d\n",
dev->name,
- board_info[ent->driver_data].name,
+ rtl_chip_info[ent->driver_data].name,
dev->base_addr,
dev->dev_addr[0], dev->dev_addr[1],
dev->dev_addr[2], dev->dev_addr[3],
dev->dev_addr[4], dev->dev_addr[5], dev->irq);
+ rtl8169_hw_phy_config(dev);
+
+ dprintk("Set MAC Reg C+CR Offset 0x82h = 0x01h\n");
+ RTL_W8(0x82, 0x01);
+
+ if (tp->mac_version < RTL_GIGA_MAC_VER_E) {
+ dprintk("Set PCI Latency=0x40\n");
+ pci_write_config_byte(pdev, PCI_LATENCY_TIMER, 0x40);
+ }
+
+ if (tp->mac_version == RTL_GIGA_MAC_VER_D) {
+ dprintk("Set MAC Reg C+CR Offset 0x82h = 0x01h\n");
+ RTL_W8(0x82, 0x01);
+ dprintk("Set PHY Reg 0x0bh = 0x00h\n");
+ mdio_write(ioaddr, 0x0b, 0x0000); //w 0x0b 15 0 0
+ }
+
// if TBI is not endbled
if (!(RTL_R8(PHYstatus) & TBI_Enable)) {
int val = mdio_read(ioaddr, PHY_AUTO_NEGO_REG);
@@ -546,23 +860,23 @@ rtl8169_init_one(struct pci_dev *pdev, c
Cap10_100 = 0, Cap1000 = 0;
switch (option) {
case _10_Half:
- Cap10_100 = PHY_Cap_10_Half;
+ Cap10_100 = PHY_Cap_10_Half_Or_Less;
Cap1000 = PHY_Cap_Null;
break;
case _10_Full:
- Cap10_100 = PHY_Cap_10_Full;
+ Cap10_100 = PHY_Cap_10_Full_Or_Less;
Cap1000 = PHY_Cap_Null;
break;
case _100_Half:
- Cap10_100 = PHY_Cap_100_Half;
+ Cap10_100 = PHY_Cap_100_Half_Or_Less;
Cap1000 = PHY_Cap_Null;
break;
case _100_Full:
- Cap10_100 = PHY_Cap_100_Full;
+ Cap10_100 = PHY_Cap_100_Full_Or_Less;
Cap1000 = PHY_Cap_Null;
break;
case _1000_Full:
- Cap10_100 = PHY_Cap_Null;
+ Cap10_100 = PHY_Cap_100_Full_Or_Less;
Cap1000 = PHY_Cap_1000_Full;
break;
default:
@@ -576,9 +890,7 @@ rtl8169_init_one(struct pci_dev *pdev, c
// enable 10/100 Full/Half Mode, leave PHY_AUTO_NEGO_REG bit4:0 unchanged
mdio_write(ioaddr, PHY_AUTO_NEGO_REG,
- PHY_Cap_10_Half | PHY_Cap_10_Full |
- PHY_Cap_100_Half | PHY_Cap_100_Full | (val &
- 0x1F));
+ PHY_Cap_100_Full_Or_Less | (val & 0x1f));
// enable 1000 Full Mode
mdio_write(ioaddr, PHY_1000_CTRL_REG,
@@ -647,56 +959,96 @@ rtl8169_remove_one(struct pci_dev *pdev)
pci_set_drvdata(pdev, NULL);
}
+#ifdef CONFIG_PM
+
+static int rtl8169_suspend(struct pci_dev *pdev, u32 state)
+{
+ struct net_device *dev = pci_get_drvdata(pdev);
+ struct rtl8169_private *tp = dev->priv;
+ void *ioaddr = tp->mmio_addr;
+ unsigned long flags;
+
+ if (!netif_running(dev))
+ return 0;
+
+ netif_device_detach(dev);
+ netif_stop_queue(dev);
+ spin_lock_irqsave(&tp->lock, flags);
+
+ /* Disable interrupts, stop Rx and Tx */
+ RTL_W16(IntrMask, 0);
+ RTL_W8(ChipCmd, 0);
+
+ /* Update the error counts. */
+ tp->stats.rx_missed_errors += RTL_R32(RxMissed);
+ RTL_W32(RxMissed, 0);
+ spin_unlock_irqrestore(&tp->lock, flags);
+
+ return 0;
+}
+
+static int rtl8169_resume(struct pci_dev *pdev)
+{
+ struct net_device *dev = pci_get_drvdata(pdev);
+
+ if (!netif_running(dev))
+ return 0;
+
+ netif_device_attach(dev);
+ rtl8169_hw_start(dev);
+
+ return 0;
+}
+
+#endif /* CONFIG_PM */
+
static int
rtl8169_open(struct net_device *dev)
{
struct rtl8169_private *tp = dev->priv;
+ struct pci_dev *pdev = tp->pci_dev;
int retval;
- u8 diff;
- u32 TxPhyAddr, RxPhyAddr;
retval =
request_irq(dev->irq, rtl8169_interrupt, SA_SHIRQ, dev->name, dev);
- if (retval) {
- return retval;
- }
+ if (retval < 0)
+ goto out;
- tp->TxDescArrays =
- kmalloc(NUM_TX_DESC * sizeof (struct TxDesc) + 256, GFP_KERNEL);
- // Tx Desscriptor needs 256 bytes alignment;
- TxPhyAddr = virt_to_bus(tp->TxDescArrays);
- diff = 256 - (TxPhyAddr - ((TxPhyAddr >> 8) << 8));
- TxPhyAddr += diff;
- tp->TxDescArray = (struct TxDesc *) (tp->TxDescArrays + diff);
-
- tp->RxDescArrays =
- kmalloc(NUM_RX_DESC * sizeof (struct RxDesc) + 256, GFP_KERNEL);
- // Rx Desscriptor needs 256 bytes alignment;
- RxPhyAddr = virt_to_bus(tp->RxDescArrays);
- diff = 256 - (RxPhyAddr - ((RxPhyAddr >> 8) << 8));
- RxPhyAddr += diff;
- tp->RxDescArray = (struct RxDesc *) (tp->RxDescArrays + diff);
+ retval = -ENOMEM;
- if (tp->TxDescArrays == NULL || tp->RxDescArrays == NULL) {
- printk(KERN_INFO
- "Allocate RxDescArray or TxDescArray failed\n");
- free_irq(dev->irq, dev);
- if (tp->TxDescArrays)
- kfree(tp->TxDescArrays);
- if (tp->RxDescArrays)
- kfree(tp->RxDescArrays);
- return -ENOMEM;
- }
- tp->RxBufferRings = kmalloc(RX_BUF_SIZE * NUM_RX_DESC, GFP_KERNEL);
- if (tp->RxBufferRings == NULL) {
- printk(KERN_INFO "Allocate RxBufferRing failed\n");
- }
+ /*
+ * Rx and Tx desscriptors needs 256 bytes alignment.
+ * pci_alloc_consistent provides more.
+ */
+ tp->TxDescArray = pci_alloc_consistent(pdev, R8169_TX_RING_BYTES,
+ &tp->TxPhyAddr);
+ if (!tp->TxDescArray)
+ goto err_free_irq;
+
+ tp->RxDescArray = pci_alloc_consistent(pdev, R8169_RX_RING_BYTES,
+ &tp->RxPhyAddr);
+ if (!tp->RxDescArray)
+ goto err_free_tx;
+
+ retval = rtl8169_init_ring(dev);
+ if (retval < 0)
+ goto err_free_rx;
- rtl8169_init_ring(dev);
rtl8169_hw_start(dev);
- return 0;
-
+ rtl8169_request_timer(dev);
+out:
+ return retval;
+
+err_free_rx:
+ pci_free_consistent(pdev, R8169_RX_RING_BYTES, tp->RxDescArray,
+ tp->RxPhyAddr);
+err_free_tx:
+ pci_free_consistent(pdev, R8169_TX_RING_BYTES, tp->TxDescArray,
+ tp->TxPhyAddr);
+err_free_irq:
+ free_irq(dev->irq, dev);
+ goto out;
}
static void
@@ -733,11 +1085,17 @@ rtl8169_hw_start(struct net_device *dev)
RTL_W32(TxConfig,
(TX_DMA_BURST << TxDMAShift) | (InterFrameGap <<
TxInterFrameGapShift));
+ RTL_W16(CPlusCmd, RTL_R16(CPlusCmd));
+
+ if (tp->mac_version == RTL_GIGA_MAC_VER_D) {
+ dprintk(KERN_INFO PFX "Set MAC Reg C+CR Offset 0xE0: bit-3 and bit-14 MUST be 1\n");
+ RTL_W16(CPlusCmd, RTL_R16(CPlusCmd) | (1 << 14) | (1 << 3));
+ }
tp->cur_rx = 0;
- RTL_W32(TxDescStartAddr, virt_to_bus(tp->TxDescArray));
- RTL_W32(RxDescStartAddr, virt_to_bus(tp->RxDescArray));
+ RTL_W32(TxDescStartAddr, tp->TxPhyAddr);
+ RTL_W32(RxDescStartAddr, tp->RxPhyAddr);
RTL_W8(Cfg9346, Cfg9346_Lock);
udelay(10);
@@ -755,31 +1113,131 @@ rtl8169_hw_start(struct net_device *dev)
}
-static void
-rtl8169_init_ring(struct net_device *dev)
+static inline void rtl8169_make_unusable_by_asic(struct RxDesc *desc)
+{
+ desc->buf_addr = 0xdeadbeef;
+ desc->status &= ~cpu_to_le32(OWNbit | RsvdMask);
+}
+
+static void rtl8169_free_rx_skb(struct pci_dev *pdev, struct sk_buff **sk_buff,
+ struct RxDesc *desc)
+{
+ pci_unmap_single(pdev, le32_to_cpu(desc->buf_addr), RX_BUF_SIZE,
+ PCI_DMA_FROMDEVICE);
+ dev_kfree_skb(*sk_buff);
+ *sk_buff = NULL;
+ rtl8169_make_unusable_by_asic(desc);
+}
+
+static inline void rtl8169_return_to_asic(struct RxDesc *desc)
+{
+ desc->status |= cpu_to_le32(OWNbit + RX_BUF_SIZE);
+}
+
+static inline void rtl8169_give_to_asic(struct RxDesc *desc, dma_addr_t mapping)
+{
+ desc->buf_addr = cpu_to_le32(mapping);
+ desc->status |= cpu_to_le32(OWNbit + RX_BUF_SIZE);
+}
+
+static int rtl8169_alloc_rx_skb(struct pci_dev *pdev, struct net_device *dev,
+ struct sk_buff **sk_buff, struct RxDesc *desc)
+{
+ struct sk_buff *skb;
+ dma_addr_t mapping;
+ int ret = 0;
+
+ skb = dev_alloc_skb(RX_BUF_SIZE);
+ if (!skb)
+ goto err_out;
+
+ skb->dev = dev;
+ skb_reserve(skb, 2);
+ *sk_buff = skb;
+
+ mapping = pci_map_single(pdev, skb->tail, RX_BUF_SIZE,
+ PCI_DMA_FROMDEVICE);
+
+ rtl8169_give_to_asic(desc, mapping);
+
+out:
+ return ret;
+
+err_out:
+ ret = -ENOMEM;
+ rtl8169_make_unusable_by_asic(desc);
+ goto out;
+}
+
+static void rtl8169_rx_clear(struct rtl8169_private *tp)
{
- struct rtl8169_private *tp = dev->priv;
int i;
- tp->cur_rx = 0;
- tp->cur_tx = 0;
- tp->dirty_tx = 0;
+ for (i = 0; i < NUM_RX_DESC; i++) {
+ if (tp->Rx_skbuff[i]) {
+ rtl8169_free_rx_skb(tp->pci_dev, tp->Rx_skbuff + i,
+ tp->RxDescArray + i);
+ }
+ }
+}
+
+static u32 rtl8169_rx_fill(struct rtl8169_private *tp, struct net_device *dev,
+ u32 start, u32 end)
+{
+ u32 cur;
+
+ for (cur = start; end - cur > 0; cur++) {
+ int ret, i = cur % NUM_RX_DESC;
+
+ if (tp->Rx_skbuff[i])
+ continue;
+
+ ret = rtl8169_alloc_rx_skb(tp->pci_dev, dev, tp->Rx_skbuff + i,
+ tp->RxDescArray + i);
+ if (ret < 0)
+ break;
+ }
+ return cur - start;
+}
+
+static inline void rtl8169_mark_as_last_descriptor(struct RxDesc *desc)
+{
+ desc->status |= cpu_to_le32(EORbit);
+}
+
+static int rtl8169_init_ring(struct net_device *dev)
+{
+ struct rtl8169_private *tp = dev->priv;
+
+ tp->cur_rx = tp->dirty_rx = 0;
+ tp->cur_tx = tp->dirty_tx = 0;
memset(tp->TxDescArray, 0x0, NUM_TX_DESC * sizeof (struct TxDesc));
memset(tp->RxDescArray, 0x0, NUM_RX_DESC * sizeof (struct RxDesc));
- for (i = 0; i < NUM_TX_DESC; i++) {
- tp->Tx_skbuff[i] = NULL;
- }
- for (i = 0; i < NUM_RX_DESC; i++) {
- if (i == (NUM_RX_DESC - 1))
- tp->RxDescArray[i].status =
- (OWNbit | EORbit) + RX_BUF_SIZE;
- else
- tp->RxDescArray[i].status = OWNbit + RX_BUF_SIZE;
+ memset(tp->Tx_skbuff, 0x0, NUM_TX_DESC * sizeof(struct sk_buff *));
+ memset(tp->Rx_skbuff, 0x0, NUM_RX_DESC * sizeof(struct sk_buff *));
- tp->RxBufferRing[i] = &(tp->RxBufferRings[i * RX_BUF_SIZE]);
- tp->RxDescArray[i].buf_addr = virt_to_bus(tp->RxBufferRing[i]);
- }
+ if (rtl8169_rx_fill(tp, dev, 0, NUM_RX_DESC) != NUM_RX_DESC)
+ goto err_out;
+
+ rtl8169_mark_as_last_descriptor(tp->RxDescArray + NUM_RX_DESC - 1);
+
+ return 0;
+
+err_out:
+ rtl8169_rx_clear(tp);
+ return -ENOMEM;
+}
+
+static void rtl8169_unmap_tx_skb(struct pci_dev *pdev, struct sk_buff **sk_buff,
+ struct TxDesc *desc)
+{
+ u32 len = sk_buff[0]->len;
+
+ pci_unmap_single(pdev, le32_to_cpu(desc->buf_addr),
+ len < ETH_ZLEN ? ETH_ZLEN : len, PCI_DMA_TODEVICE);
+ desc->buf_addr = 0x00;
+ *sk_buff = NULL;
}
static void
@@ -789,9 +1247,12 @@ rtl8169_tx_clear(struct rtl8169_private
tp->cur_tx = 0;
for (i = 0; i < NUM_TX_DESC; i++) {
- if (tp->Tx_skbuff[i] != NULL) {
- dev_kfree_skb(tp->Tx_skbuff[i]);
- tp->Tx_skbuff[i] = NULL;
+ struct sk_buff *skb = tp->Tx_skbuff[i];
+
+ if (skb) {
+ rtl8169_unmap_tx_skb(tp->pci_dev, tp->Tx_skbuff + i,
+ tp->TxDescArray + i);
+ dev_kfree_skb(skb);
tp->stats.tx_dropped++;
}
}
@@ -829,48 +1290,58 @@ rtl8169_start_xmit(struct sk_buff *skb,
struct rtl8169_private *tp = dev->priv;
void *ioaddr = tp->mmio_addr;
int entry = tp->cur_tx % NUM_TX_DESC;
+ u32 len = skb->len;
- if (skb->len < ETH_ZLEN) {
+ if (unlikely(skb->len < ETH_ZLEN)) {
skb = skb_padto(skb, ETH_ZLEN);
- if (skb == NULL)
- return 0;
+ if (!skb)
+ goto err_update_stats;
+ len = ETH_ZLEN;
}
spin_lock_irq(&tp->lock);
- if ((tp->TxDescArray[entry].status & OWNbit) == 0) {
+ if (!(le32_to_cpu(tp->TxDescArray[entry].status) & OWNbit)) {
+ dma_addr_t mapping;
+
+ mapping = pci_map_single(tp->pci_dev, skb->data, len,
+ PCI_DMA_TODEVICE);
+
tp->Tx_skbuff[entry] = skb;
- tp->TxDescArray[entry].buf_addr = virt_to_bus(skb->data);
- if (entry != (NUM_TX_DESC - 1))
- tp->TxDescArray[entry].status =
- (OWNbit | FSbit | LSbit) | ((skb->len > ETH_ZLEN) ?
- skb->len : ETH_ZLEN);
- else
- tp->TxDescArray[entry].status =
- (OWNbit | EORbit | FSbit | LSbit) |
- ((skb->len > ETH_ZLEN) ? skb->len : ETH_ZLEN);
+ tp->TxDescArray[entry].buf_addr = cpu_to_le32(mapping);
+ tp->TxDescArray[entry].status = cpu_to_le32(OWNbit | FSbit |
+ LSbit | len | (EORbit * !((entry + 1) % NUM_TX_DESC)));
+
RTL_W8(TxPoll, 0x40); //set polling bit
dev->trans_start = jiffies;
tp->cur_tx++;
- }
+ } else
+ goto err_drop;
- spin_unlock_irq(&tp->lock);
if ((tp->cur_tx - NUM_TX_DESC) == tp->dirty_tx) {
netif_stop_queue(dev);
}
+out:
+ spin_unlock_irq(&tp->lock);
return 0;
+
+err_drop:
+ dev_kfree_skb(skb);
+err_update_stats:
+ tp->stats.tx_dropped++;
+ goto out;
}
static void
rtl8169_tx_interrupt(struct net_device *dev, struct rtl8169_private *tp,
void *ioaddr)
{
- unsigned long dirty_tx, tx_left = 0;
+ unsigned long dirty_tx, tx_left;
assert(dev != NULL);
assert(tp != NULL);
@@ -882,12 +1353,15 @@ rtl8169_tx_interrupt(struct net_device *
while (tx_left > 0) {
int entry = dirty_tx % NUM_TX_DESC;
- if ((tp->TxDescArray[entry].status & OWNbit) == 0) {
+ if (!(le32_to_cpu(tp->TxDescArray[entry].status) & OWNbit)) {
struct sk_buff *skb = tp->Tx_skbuff[entry];
+ /* FIXME: is it really accurate for TxErr ? */
tp->stats.tx_bytes += skb->len >= ETH_ZLEN ?
skb->len : ETH_ZLEN;
tp->stats.tx_packets++;
+ rtl8169_unmap_tx_skb(tp->pci_dev, tp->Tx_skbuff + entry,
+ tp->TxDescArray + entry);
dev_kfree_skb_irq(skb);
tp->Tx_skbuff[entry] = NULL;
dirty_tx++;
@@ -902,70 +1376,102 @@ rtl8169_tx_interrupt(struct net_device *
}
}
+static inline int rtl8169_try_rx_copy(struct sk_buff **sk_buff, int pkt_size,
+ struct RxDesc *desc,
+ struct net_device *dev)
+{
+ int ret = -1;
+
+ if (pkt_size < rx_copybreak) {
+ struct sk_buff *skb;
+
+ skb = dev_alloc_skb(pkt_size + 2);
+ if (skb) {
+ skb->dev = dev;
+ skb_reserve(skb, 2);
+ eth_copy_and_sum(skb, sk_buff[0]->tail, pkt_size, 0);
+ *sk_buff = skb;
+ rtl8169_return_to_asic(desc);
+ ret = 0;
+ }
+ }
+ return ret;
+}
+
static void
rtl8169_rx_interrupt(struct net_device *dev, struct rtl8169_private *tp,
void *ioaddr)
{
- int cur_rx;
- struct sk_buff *skb;
- int pkt_size = 0;
+ unsigned long cur_rx, rx_left;
+ int delta;
assert(dev != NULL);
assert(tp != NULL);
assert(ioaddr != NULL);
cur_rx = tp->cur_rx;
+ rx_left = NUM_RX_DESC + tp->dirty_rx - cur_rx;
- while ((tp->RxDescArray[cur_rx].status & OWNbit) == 0) {
+ while (rx_left > 0) {
+ int entry = cur_rx % NUM_RX_DESC;
+ u32 status = le32_to_cpu(tp->RxDescArray[entry].status);
- if (tp->RxDescArray[cur_rx].status & RxRES) {
+ if (status & OWNbit)
+ break;
+ if (status & RxRES) {
printk(KERN_INFO "%s: Rx ERROR!!!\n", dev->name);
tp->stats.rx_errors++;
- if (tp->RxDescArray[cur_rx].status & (RxRWT | RxRUNT))
+ if (status & (RxRWT | RxRUNT))
tp->stats.rx_length_errors++;
- if (tp->RxDescArray[cur_rx].status & RxCRC)
+ if (status & RxCRC)
tp->stats.rx_crc_errors++;
} else {
- pkt_size =
- (int) (tp->RxDescArray[cur_rx].
- status & 0x00001FFF) - 4;
- skb = dev_alloc_skb(pkt_size + 2);
- if (skb != NULL) {
- skb->dev = dev;
- skb_reserve(skb, 2); // 16 byte align the IP fields. //
- eth_copy_and_sum(skb, tp->RxBufferRing[cur_rx],
- pkt_size, 0);
- skb_put(skb, pkt_size);
- skb->protocol = eth_type_trans(skb, dev);
- netif_rx(skb);
-
- if (cur_rx == (NUM_RX_DESC - 1))
- tp->RxDescArray[cur_rx].status =
- (OWNbit | EORbit) + RX_BUF_SIZE;
- else
- tp->RxDescArray[cur_rx].status =
- OWNbit + RX_BUF_SIZE;
-
- tp->RxDescArray[cur_rx].buf_addr =
- virt_to_bus(tp->RxBufferRing[cur_rx]);
- dev->last_rx = jiffies;
- tp->stats.rx_bytes += pkt_size;
- tp->stats.rx_packets++;
- } else {
- printk(KERN_WARNING
- "%s: Memory squeeze, deferring packet.\n",
- dev->name);
- /* We should check that some rx space is free.
- If not, free one and mark stats->rx_dropped++. */
- tp->stats.rx_dropped++;
+ struct RxDesc *desc = tp->RxDescArray + entry;
+ struct sk_buff *skb = tp->Rx_skbuff[entry];
+ int pkt_size = (status & 0x00001FFF) - 4;
+
+ pci_dma_sync_single(tp->pci_dev,
+ le32_to_cpu(desc->buf_addr),
+ RX_BUF_SIZE, PCI_DMA_FROMDEVICE);
+
+ if (rtl8169_try_rx_copy(&skb, pkt_size, desc, dev)) {
+ pci_unmap_single(tp->pci_dev,
+ le32_to_cpu(desc->buf_addr),
+ RX_BUF_SIZE,
+ PCI_DMA_FROMDEVICE);
+ tp->Rx_skbuff[entry] = NULL;
}
- }
-
- cur_rx = (cur_rx + 1) % NUM_RX_DESC;
+ skb_put(skb, pkt_size);
+ skb->protocol = eth_type_trans(skb, dev);
+ netif_rx(skb);
+
+ dev->last_rx = jiffies;
+ tp->stats.rx_bytes += pkt_size;
+ tp->stats.rx_packets++;
+ }
+
+ cur_rx++;
+ rx_left--;
}
tp->cur_rx = cur_rx;
+
+ delta = rtl8169_rx_fill(tp, dev, tp->dirty_rx, tp->cur_rx);
+ if (delta > 0)
+ tp->dirty_rx += delta;
+ else if (delta < 0)
+ printk(KERN_INFO "%s: no Rx buffer allocated\n", dev->name);
+
+ /*
+ * FIXME: until there is periodic timer to try and refill the ring,
+ * a temporary shortage may definitely kill the Rx process.
+ * - disable the asic to try and avoid an overflow and kick it again
+ * after refill ?
+ * - how do others driver handle this condition (Uh oh...).
+ */
+ if (tp->dirty_rx + NUM_RX_DESC == tp->cur_rx)
+ printk(KERN_EMERG "%s: Rx buffers exhausted\n", dev->name);
}
/* The interrupt handler does all of the Rx thread work and cleans up after the Tx thread. */
@@ -994,9 +1500,7 @@ rtl8169_interrupt(int irq, void *dev_ins
RTL_W16(IntrStatus,
(status & RxFIFOOver) ? (status | RxOverflow) : status);
- if ((status &
- (SYSErr | PCSTimeout | RxUnderrun | RxOverflow | RxFIFOOver
- | TxErr | TxOK | RxErr | RxOK)) == 0)
+ if (!(status & rtl8169_intr_mask))
break;
// Rx interrupt
@@ -1026,11 +1530,13 @@ static int
rtl8169_close(struct net_device *dev)
{
struct rtl8169_private *tp = dev->priv;
+ struct pci_dev *pdev = tp->pci_dev;
void *ioaddr = tp->mmio_addr;
- int i;
netif_stop_queue(dev);
+ rtl8169_delete_timer(dev);
+
spin_lock_irq(&tp->lock);
/* Stop the chip's Tx and Rx DMA processes. */
@@ -1049,16 +1555,15 @@ rtl8169_close(struct net_device *dev)
free_irq(dev->irq, dev);
rtl8169_tx_clear(tp);
- kfree(tp->TxDescArrays);
- kfree(tp->RxDescArrays);
- tp->TxDescArrays = NULL;
- tp->RxDescArrays = NULL;
+
+ rtl8169_rx_clear(tp);
+
+ pci_free_consistent(pdev, R8169_RX_RING_BYTES, tp->RxDescArray,
+ tp->RxPhyAddr);
+ pci_free_consistent(pdev, R8169_TX_RING_BYTES, tp->TxDescArray,
+ tp->TxPhyAddr);
tp->TxDescArray = NULL;
tp->RxDescArray = NULL;
- kfree(tp->RxBufferRings);
- for (i = 0; i < NUM_RX_DESC; i++) {
- tp->RxBufferRing[i] = NULL;
- }
return 0;
}
@@ -1112,11 +1617,26 @@ rtl8169_set_rx_mode(struct net_device *d
spin_unlock_irqrestore(&tp->lock, flags);
}
+/**
+ * rtl8169_get_stats - Get rtl8169 read/write statistics
+ * @dev: The Ethernet Device to get statistics for
+ *
+ * Get TX/RX statistics for rtl8169
+ */
struct net_device_stats *
rtl8169_get_stats(struct net_device *dev)
{
struct rtl8169_private *tp = dev->priv;
+ void *ioaddr = tp->mmio_addr;
+ unsigned long flags;
+ if (netif_running(dev)) {
+ spin_lock_irqsave(&tp->lock, flags);
+ tp->stats.rx_missed_errors += RTL_R32(RxMissed);
+ RTL_W32(RxMissed, 0);
+ spin_unlock_irqrestore(&tp->lock, flags);
+ }
+
return &tp->stats;
}
@@ -1125,8 +1645,10 @@ static struct pci_driver rtl8169_pci_dri
.id_table = rtl8169_pci_tbl,
.probe = rtl8169_init_one,
.remove = __devexit_p(rtl8169_remove_one),
- .suspend = NULL,
- .resume = NULL,
+#ifdef CONFIG_PM
+ .suspend = rtl8169_suspend,
+ .resume = rtl8169_resume,
+#endif
};
static int __init
--- diff/drivers/net/slip.c 2003-10-27 09:20:38.000000000 +0000
+++ source/drivers/net/slip.c 2004-03-16 11:27:48.207858952 +0000
@@ -1307,7 +1307,7 @@ static int sl_ioctl(struct net_device *d
/* Resolve race condition, when ioctl'ing hanged up
and opened by another process device.
*/
- if (sl->tty != current->tty && sl->pid != current->pid) {
+ if (sl->tty != current->signal->tty && sl->pid != current->pid) {
spin_unlock_bh(&sl->lock);
return -EPERM;
}
--- diff/drivers/net/tulip/tulip_core.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/net/tulip/tulip_core.c 2004-03-16 11:27:48.208858800 +0000
@@ -1512,7 +1512,7 @@ static int __devinit tulip_init_one (str
}
}
/* Lite-On boards have the address byte-swapped. */
- if ((dev->dev_addr[0] == 0xA0 || dev->dev_addr[0] == 0xC0)
+ if ((dev->dev_addr[0] == 0xA0 || dev->dev_addr[0] == 0xC0 || dev->dev_addr[0] == 0x02)
&& dev->dev_addr[1] == 0x00)
for (i = 0; i < 6; i+=2) {
char tmp = dev->dev_addr[i];
--- diff/drivers/parport/parport_pc.c 2004-03-11 10:20:26.000000000 +0000
+++ source/drivers/parport/parport_pc.c 2004-03-16 11:27:48.212858192 +0000
@@ -13,6 +13,7 @@
* Many ECP bugs fixed. Fred Barnes & Jamie Lokier, 1999
* More PCI support now conditional on CONFIG_PCI, 03/2001, Paul G.
* Various hacks, Fred Barnes, 04/2001
+ * Updated probing logic - Adam Belay <ambx1@neo.rr.com>
*/
/* This driver should work with any hardware that is broadly compatible
@@ -98,7 +99,8 @@ static int user_specified;
(defined(CONFIG_PARPORT_1284) && defined(CONFIG_PARPORT_PC_FIFO))
static int verbose_probing;
#endif
-static int registered_parport;
+static int pci_registered_parport;
+static int pnp_registered_parport;
/* frob_control, but for ECR */
static void frob_econtrol (struct parport *pb, unsigned char m,
@@ -2771,10 +2773,11 @@ static struct pci_device_id parport_pc_p
};
MODULE_DEVICE_TABLE(pci,parport_pc_pci_tbl);
-static int __devinit parport_pc_pci_probe (struct pci_dev *dev,
+static int parport_pc_pci_probe (struct pci_dev *dev,
const struct pci_device_id *id)
{
int err, count, n, i = id->driver_data;
+
if (i < last_sio)
/* This is an onboard Super-IO and has already been probed */
return 0;
@@ -2847,23 +2850,72 @@ static struct pci_driver parport_pc_pci_
static int __init parport_pc_init_superio(int autoirq, int autodma) {return 0;}
#endif /* CONFIG_PCI */
-#ifdef CONFIG_PNP
-static const struct pnp_device_id pnp_dev_table[] = {
+
+static const struct pnp_device_id parport_pc_pnp_tbl[] = {
/* Standard LPT Printer Port */
{.id = "PNP0400", .driver_data = 0},
/* ECP Printer Port */
{.id = "PNP0401", .driver_data = 0},
- {.id = ""}
+ { }
};
+MODULE_DEVICE_TABLE(pnp,parport_pc_pnp_tbl);
+
+static int parport_pc_pnp_probe(struct pnp_dev *dev, const struct pnp_device_id *id)
+{
+ struct parport *pdata;
+ unsigned long io_lo, io_hi;
+ int dma, irq;
+
+ if (pnp_port_valid(dev,0) &&
+ !(pnp_port_flags(dev,0) & IORESOURCE_DISABLED)) {
+ io_lo = pnp_port_start(dev,0);
+ } else
+ return -EINVAL;
+
+ if (pnp_port_valid(dev,1) &&
+ !(pnp_port_flags(dev,1) & IORESOURCE_DISABLED)) {
+ io_hi = pnp_port_start(dev,1);
+ } else
+ io_hi = 0;
+
+ if (pnp_irq_valid(dev,0) &&
+ !(pnp_irq_flags(dev,0) & IORESOURCE_DISABLED)) {
+ irq = pnp_irq(dev,0);
+ } else
+ irq = PARPORT_IRQ_NONE;
+
+ if (pnp_dma_valid(dev,0) &&
+ !(pnp_dma_flags(dev,0) & IORESOURCE_DISABLED)) {
+ dma = pnp_dma(dev,0);
+ } else
+ dma = PARPORT_DMA_NONE;
+
+ printk(KERN_INFO "parport: PnPBIOS parport detected.\n");
+ if (!(pdata = parport_pc_probe_port (io_lo, io_hi, irq, dma, NULL)))
+ return -ENODEV;
+
+ pnp_set_drvdata(dev,pdata);
+ return 0;
+}
+
+static void parport_pc_pnp_remove(struct pnp_dev *dev)
+{
+ struct parport *pdata = (struct parport *)pnp_get_drvdata(dev);
+ if (!pdata)
+ return;
+
+ parport_pc_unregister_port(pdata);
+}
+
/* we only need the pnp layer to activate the device, at least for now */
static struct pnp_driver parport_pc_pnp_driver = {
.name = "parport_pc",
- .id_table = pnp_dev_table,
+ .id_table = parport_pc_pnp_tbl,
+ .probe = parport_pc_pnp_probe,
+ .remove = parport_pc_pnp_remove,
};
-#else
-static struct pnp_driver parport_pc_pnp_driver;
-#endif
+
/* This is called by parport_pc_find_nonpci_ports (in asm/parport.h) */
static int __init __attribute__((unused))
@@ -2903,12 +2955,18 @@ static int __init parport_pc_find_ports
/* Onboard SuperIO chipsets that show themselves on the PCI bus. */
count += parport_pc_init_superio (autoirq, autodma);
+ r = pnp_register_driver (&parport_pc_pnp_driver);
+ if (r >= 0) {
+ pnp_registered_parport = 1;
+ count += r;
+ }
+
/* ISA ports and whatever (see asm/parport.h). */
count += parport_pc_find_nonpci_ports (autoirq, autodma);
r = pci_register_driver (&parport_pc_pci_driver);
if (r >= 0) {
- registered_parport = 1;
+ pci_registered_parport = 1;
count += r;
}
@@ -3104,9 +3162,6 @@ static int __init parport_pc_init(void)
if (parse_parport_params())
return -EINVAL;
- /* try to activate any PnP parports first */
- pnp_register_driver(&parport_pc_pnp_driver);
-
if (io[0]) {
int i;
/* Only probe the ports we were given. */
@@ -3120,24 +3175,18 @@ static int __init parport_pc_init(void)
irqval[i], dmaval[i], NULL))
count++;
}
- } else {
+ } else
count += parport_pc_find_ports (irqval[0], dmaval[0]);
- if (!count && registered_parport)
- pci_unregister_driver (&parport_pc_pci_driver);
- }
-
- if (!count) {
- pnp_unregister_driver (&parport_pc_pnp_driver);
- return -ENODEV;
- }
return 0;
}
static void __exit parport_pc_exit(void)
{
- if (registered_parport)
+ if (pci_registered_parport)
pci_unregister_driver (&parport_pc_pci_driver);
+ if (pnp_registered_parport)
+ pnp_unregister_driver (&parport_pc_pnp_driver);
spin_lock(&ports_lock);
while (!list_empty(&ports_list)) {
@@ -3151,13 +3200,8 @@ static void __exit parport_pc_exit(void)
spin_lock(&ports_lock);
}
spin_unlock(&ports_lock);
- pnp_unregister_driver (&parport_pc_pnp_driver);
}
-
-MODULE_AUTHOR("Phil Blundell, Tim Waugh, others");
-MODULE_DESCRIPTION("PC-style parallel port driver");
-MODULE_LICENSE("GPL");
MODULE_AUTHOR("Phil Blundell, Tim Waugh, others");
MODULE_DESCRIPTION("PC-style parallel port driver");
MODULE_LICENSE("GPL");
--- diff/drivers/pci/hotplug/Makefile 2004-03-11 10:20:26.000000000 +0000
+++ source/drivers/pci/hotplug/Makefile 2004-03-16 11:27:48.216857584 +0000
@@ -40,7 +40,9 @@ acpiphp-objs := acpiphp_core.o \
acpiphp_res.o
rpaphp-objs := rpaphp_core.o \
- rpaphp_pci.o
+ rpaphp_pci.o \
+ rpaphp_slot.o \
+ rpaphp_vio.o
rpadlpar_io-objs := rpadlpar_core.o \
rpadlpar_sysfs.o
--- diff/drivers/pci/hotplug/acpiphp.h 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/pci/hotplug/acpiphp.h 2004-03-16 11:27:48.214857888 +0000
@@ -235,7 +235,7 @@ extern u32 acpiphp_get_address (struct a
extern struct pci_dev *acpiphp_allocate_pcidev (struct pci_bus *pbus, int dev, int fn);
extern int acpiphp_configure_slot (struct acpiphp_slot *slot);
extern int acpiphp_configure_function (struct acpiphp_func *func);
-extern int acpiphp_unconfigure_function (struct acpiphp_func *func);
+extern void acpiphp_unconfigure_function (struct acpiphp_func *func);
extern int acpiphp_detect_pci_resource (struct acpiphp_bridge *bridge);
extern int acpiphp_init_func_resource (struct acpiphp_func *func);
--- diff/drivers/pci/hotplug/acpiphp_glue.c 2004-02-18 08:54:10.000000000 +0000
+++ source/drivers/pci/hotplug/acpiphp_glue.c 2004-03-16 11:27:48.214857888 +0000
@@ -694,14 +694,14 @@ static int power_on_slot(struct acpiphp_
func = list_entry(l, struct acpiphp_func, sibling);
if (func->flags & FUNC_HAS_PS0) {
- dbg("%s: executing _PS0 on %s\n", __FUNCTION__,
- pci_name(func->pci_dev));
+ dbg("%s: executing _PS0\n", __FUNCTION__);
status = acpi_evaluate_object(func->handle, "_PS0", NULL, NULL);
if (ACPI_FAILURE(status)) {
warn("%s: _PS0 failed\n", __FUNCTION__);
retval = -1;
goto err_exit;
- }
+ } else
+ break;
}
}
@@ -737,7 +737,8 @@ static int power_off_slot(struct acpiphp
warn("%s: _PS3 failed\n", __FUNCTION__);
retval = -1;
goto err_exit;
- }
+ } else
+ break;
}
}
@@ -757,7 +758,8 @@ static int power_off_slot(struct acpiphp
warn("%s: _EJ0 failed\n", __FUNCTION__);
retval = -1;
goto err_exit;
- }
+ } else
+ break;
}
}
@@ -865,15 +867,8 @@ static int disable_device(struct acpiphp
list_for_each (l, &slot->funcs) {
func = list_entry(l, struct acpiphp_func, sibling);
- if (func->pci_dev) {
- if (acpiphp_unconfigure_function(func) == 0) {
- func->pci_dev = NULL;
- } else {
- err("failed to unconfigure device\n");
- retval = -1;
- goto err_exit;
- }
- }
+ if (func->pci_dev)
+ acpiphp_unconfigure_function(func);
}
slot->flags &= (~SLOT_ENABLED);
@@ -1269,7 +1264,7 @@ int acpiphp_check_bridge(struct acpiphp_
up(&slot->crit_sect);
goto err_exit;
}
- enabled++;
+ disabled++;
}
} else {
/* if disabled but present, enable */
@@ -1280,7 +1275,7 @@ int acpiphp_check_bridge(struct acpiphp_
up(&slot->crit_sect);
goto err_exit;
}
- disabled++;
+ enabled++;
}
}
}
--- diff/drivers/pci/hotplug/acpiphp_pci.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/pci/hotplug/acpiphp_pci.c 2004-03-16 11:27:48.215857736 +0000
@@ -83,8 +83,8 @@ static int init_config_space (struct acp
if (bar & PCI_BASE_ADDRESS_SPACE_IO) {
/* This is IO */
- len = bar & 0xFFFFFFFC;
- len = ~len + 1;
+ len = bar & (PCI_BASE_ADDRESS_IO_MASK & 0xFFFF);
+ len = len & ~(len - 1);
dbg("len in IO %x, BAR %d\n", len, count);
@@ -226,8 +226,8 @@ static int detect_used_resource (struct
if (len & PCI_BASE_ADDRESS_SPACE_IO) {
/* This is IO */
base = bar & 0xFFFFFFFC;
- len &= 0xFFFFFFFC;
- len = ~len + 1;
+ len = len & (PCI_BASE_ADDRESS_IO_MASK & 0xFFFF);
+ len = len & ~(len - 1);
dbg("BAR[%d] %08x - %08x (IO)\n", count, (u32)base, (u32)base + len - 1);
@@ -351,8 +351,8 @@ int acpiphp_init_func_resource (struct a
if (len & PCI_BASE_ADDRESS_SPACE_IO) {
/* This is IO */
base = bar & 0xFFFFFFFC;
- len &= 0xFFFFFFFC;
- len = ~len + 1;
+ len = len & (PCI_BASE_ADDRESS_IO_MASK & 0xFFFF);
+ len = len & ~(len - 1);
dbg("BAR[%d] %08x - %08x (IO)\n", count, (u32)base, (u32)base + len - 1);
@@ -485,14 +485,13 @@ int acpiphp_configure_function (struct a
* @func: function to be unconfigured
*
*/
-int acpiphp_unconfigure_function (struct acpiphp_func *func)
+void acpiphp_unconfigure_function (struct acpiphp_func *func)
{
struct acpiphp_bridge *bridge;
- int retval = 0;
/* if pci_dev is NULL, ignore it */
if (!func->pci_dev)
- goto err_exit;
+ return;
pci_remove_bus_device(func->pci_dev);
@@ -505,7 +504,4 @@ int acpiphp_unconfigure_function (struct
acpiphp_move_resource(&func->p_mem_head, &bridge->p_mem_head);
acpiphp_move_resource(&func->bus_head, &bridge->bus_head);
spin_unlock(&bridge->res_lock);
-
- err_exit:
- return retval;
}
--- diff/drivers/pci/hotplug/acpiphp_res.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/pci/hotplug/acpiphp_res.c 2004-03-16 11:27:48.216857584 +0000
@@ -224,7 +224,7 @@ struct pci_resource *acpiphp_get_io_reso
} /* End of too big on top end */
/* For IO make sure it's not in the ISA aliasing space */
- if (node->base & 0x300L)
+ if ((node->base & 0x300L) && !(node->base & 0xfffff000))
continue;
/* If we got here, then it is the right size
--- diff/drivers/pci/hotplug/pciehp_pci.c 2004-03-11 10:20:26.000000000 +0000
+++ source/drivers/pci/hotplug/pciehp_pci.c 2004-03-16 11:27:48.217857432 +0000
@@ -103,7 +103,7 @@ int pciehp_unconfigure_device(struct pci
*/
int pciehp_set_irq (u8 bus_num, u8 dev_num, u8 int_pin, u8 irq_num)
{
-#if !defined(CONFIG_X86_IO_APIC) && !defined(CONFIG_X86_64)
+#if defined(CONFIG_X86) && !defined(CONFIG_X86_IO_APIC) && !defined(CONFIG_X86_64)
int rc;
u16 temp_word;
struct pci_dev fakedev;
--- diff/drivers/pci/hotplug/pciehprm_acpi.c 2004-03-11 10:20:26.000000000 +0000
+++ source/drivers/pci/hotplug/pciehprm_acpi.c 2004-03-16 11:27:48.218857280 +0000
@@ -1268,7 +1268,8 @@ static int print_acpi_resources (struct
int pciehprm_print_pirt(void)
{
dbg("PCIEHPRM ACPI Slots\n");
- print_acpi_resources (acpi_bridges_head);
+ if (acpi_bridges_head)
+ print_acpi_resources (acpi_bridges_head);
return 0;
}
--- diff/drivers/pci/hotplug/rpadlpar_core.c 2004-03-11 10:20:26.000000000 +0000
+++ source/drivers/pci/hotplug/rpadlpar_core.c 2004-03-16 11:27:48.219857128 +0000
@@ -43,7 +43,28 @@ static char *get_node_drc_name(struct de
return ptr;
}
-static struct device_node *find_php_slot_node(char *drc_name)
+static struct device_node *find_php_slot_vio_node(char *drc_name)
+{
+ struct device_node *child;
+ struct device_node *parent = of_find_node_by_name(NULL, "vdevice");
+
+ if (!parent)
+ return NULL;
+
+ for (child = of_get_next_child(parent, NULL);
+ child; child = of_get_next_child(parent, child)) {
+
+ char *loc_code;
+
+ loc_code = get_property(child, "ibm,loc-code", NULL);
+ if (loc_code && !strcmp(loc_code, drc_name))
+ return child;
+ }
+
+ return NULL;
+}
+
+static struct device_node *find_php_slot_pci_node(char *drc_name)
{
struct device_node *np = NULL;
char *name;
@@ -72,7 +93,7 @@ static inline struct hotplug_slot *find_
static struct slot *find_slot(char *drc_name)
{
struct hotplug_slot *php_slot = find_php_slot(drc_name);
-
+
if (!php_slot)
return NULL;
@@ -127,14 +148,14 @@ static int pci_add_secondary_bus(struct
rpadlpar_claim_one_bus(bridge_dev->bus);
if (hose->last_busno < child->number)
- hose->last_busno = child->number;
+ hose->last_busno = child->number;
dn->bussubno = child->number;
/* ioremap() for child bus */
if (remap_bus_range(child)) {
printk(KERN_ERR "%s: could not ioremap() child bus\n",
- __FUNCTION__);
+ __FUNCTION__);
return 1;
}
@@ -162,9 +183,9 @@ static struct pci_dev *dlpar_pci_add_bus
return NULL;
}
- if (dev->hdr_type != PCI_HEADER_TYPE_BRIDGE) {
+ if (dev->hdr_type != PCI_HEADER_TYPE_BRIDGE) {
printk(KERN_ERR "%s: unexpected header type %d\n",
- __FUNCTION__, dev->hdr_type);
+ __FUNCTION__, dev->hdr_type);
return NULL;
}
@@ -180,7 +201,7 @@ static int dlpar_pci_remove_bus(struct p
if (!bridge_dev) {
printk(KERN_ERR "%s: unexpected null device\n",
- __FUNCTION__);
+ __FUNCTION__);
return 1;
}
@@ -188,11 +209,25 @@ static int dlpar_pci_remove_bus(struct p
if (unmap_bus_range(secondary_bus)) {
printk(KERN_ERR "%s: failed to unmap bus range\n",
- __FUNCTION__);
+ __FUNCTION__);
return 1;
}
pci_remove_bus_device(bridge_dev);
+ return 0;
+}
+
+static inline int dlpar_add_pci_slot(char *drc_name, struct device_node *dn)
+{
+ struct pci_dev *dev;
+
+ /* Add pci bus */
+ dev = dlpar_pci_add_bus(dn);
+ if (!dev) {
+ printk(KERN_ERR "%s: unable to add bus %s\n", __FUNCTION__,
+ drc_name);
+ return -EIO;
+ }
return 0;
}
@@ -212,37 +247,33 @@ static int dlpar_pci_remove_bus(struct p
*/
int dlpar_add_slot(char *drc_name)
{
- struct device_node *dn = find_php_slot_node(drc_name);
- struct pci_dev *dev;
+ struct device_node *dn;
int rc = 0;
if (down_interruptible(&rpadlpar_sem))
return -ERESTARTSYS;
- if (!dn) {
- rc = -ENODEV;
- goto exit;
- }
-
/* Check for existing hotplug slot */
if (find_slot(drc_name)) {
rc = -EINVAL;
goto exit;
}
- /* Add pci bus */
- dev = dlpar_pci_add_bus(dn);
- if (!dev) {
- printk(KERN_ERR "%s: unable to add bus %s\n", __FUNCTION__,
- drc_name);
- rc = -EIO;
- goto exit;
+ dn = find_php_slot_vio_node(drc_name);
+ if (!dn) {
+ dn = find_php_slot_pci_node(drc_name);
+ if (dn)
+ rc = dlpar_add_pci_slot(drc_name, dn);
+ else {
+ rc = -ENODEV;
+ goto exit;
+ }
}
- /* Add hotplug slot for new bus */
- if (rpaphp_add_slot(drc_name)) {
+ /* Add hotplug slot for new VIOA or PCI */
+ if (!rc && rpaphp_add_slot(dn)) {
printk(KERN_ERR "%s: unable to add hotplug slot %s\n",
- __FUNCTION__, drc_name);
+ __FUNCTION__, drc_name);
rc = -EIO;
}
exit:
@@ -251,60 +282,107 @@ exit:
}
/**
- * dlpar_remove_slot - DLPAR remove an I/O Slot
+ * dlpar_remove_vio_slot - DLPAR remove a virtual I/O Slot
* @drc_name: drc-name of newly added slot
*
* Remove the kernel and hotplug representations
* of an I/O Slot.
* Return Codes:
* 0 Success
+ * -EIO Internal Error
+ */
+int dlpar_remove_vio_slot(struct slot *slot, char *drc_name)
+{
+ /* Remove hotplug slot */
+
+ if (rpaphp_remove_slot(slot)) {
+ printk(KERN_ERR "%s: unable to remove hotplug slot %s\n",
+ __FUNCTION__, drc_name);
+ return -EIO;
+ }
+ return 0;
+}
+
+/**
+ * dlpar_remove_slot - DLPAR remove a PCI I/O Slot
+ * @drc_name: drc-name of newly added slot
+ *
+ * Remove the kernel and hotplug representations
+ * of a PCI I/O Slot.
+ * Return Codes:
+ * 0 Success
* -ENODEV Not a valid drc_name
- * -EINVAL Slot already removed
- * -ERESTARTSYS Signalled before obtaining lock
* -EIO Internal PCI Error
*/
-int dlpar_remove_slot(char *drc_name)
+int dlpar_remove_pci_slot(struct slot *slot, char *drc_name)
{
- struct device_node *dn = find_php_slot_node(drc_name);
- struct slot *slot;
+ struct device_node *dn = find_php_slot_pci_node(drc_name);
struct pci_dev *bridge_dev;
- int rc = 0;
-
- if (down_interruptible(&rpadlpar_sem))
- return -ERESTARTSYS;
- if (!dn) {
- rc = -ENODEV;
- goto exit;
- }
-
- slot = find_slot(drc_name);
- if (!slot) {
- rc = -EINVAL;
- goto exit;
- }
+ if (!dn)
+ return -ENODEV;
bridge_dev = slot->bridge;
if (!bridge_dev) {
printk(KERN_ERR "%s: unexpected null bridge device\n",
- __FUNCTION__);
- rc = -EIO;
- goto exit;
+ __FUNCTION__);
+ return -EIO;
}
/* Remove hotplug slot */
if (rpaphp_remove_slot(slot)) {
printk(KERN_ERR "%s: unable to remove hotplug slot %s\n",
- __FUNCTION__, drc_name);
- rc = -EIO;
- goto exit;
+ __FUNCTION__, drc_name);
+ return -EIO;
}
/* Remove pci bus */
if (dlpar_pci_remove_bus(bridge_dev)) {
printk(KERN_ERR "%s: unable to remove pci bus %s\n",
- __FUNCTION__, drc_name);
- rc = -EIO;
+ __FUNCTION__, drc_name);
+ return -EIO;
+ }
+ return 0;
+}
+
+/**
+ * dlpar_remove_slot - DLPAR remove an I/O Slot
+ * @drc_name: drc-name of newly added slot
+ *
+ * Remove the kernel and hotplug representations
+ * of an I/O Slot.
+ * Return Codes:
+ * 0 Success
+ * -ENODEV Not a valid drc_name
+ * -EINVAL Slot already removed
+ * -ERESTARTSYS Signalled before obtaining lock
+ * -EIO Internal Error
+ */
+int dlpar_remove_slot(char *drc_name)
+{
+ struct slot *slot;
+ int rc = 0;
+
+ if (down_interruptible(&rpadlpar_sem))
+ return -ERESTARTSYS;
+
+ slot = find_slot(drc_name);
+ if (!slot) {
+ rc = -EINVAL;
+ goto exit;
+ }
+
+ switch (slot->dev_type) {
+ case PCI_DEV:
+ rc = dlpar_remove_pci_slot(slot, drc_name);
+ break;
+
+ case VIO_DEV:
+ rc = dlpar_remove_vio_slot(slot, drc_name);
+ break;
+
+ default:
+ rc = -EIO;
}
exit:
up(&rpadlpar_sem);
@@ -324,7 +402,7 @@ int __init rpadlpar_io_init(void)
if (!is_dlpar_capable()) {
printk(KERN_WARNING "%s: partition not DLPAR capable\n",
- __FUNCTION__);
+ __FUNCTION__);
return -EPERM;
}
--- diff/drivers/pci/hotplug/rpaphp.h 2004-03-11 10:20:26.000000000 +0000
+++ source/drivers/pci/hotplug/rpaphp.h 2004-03-16 11:27:48.222856672 +0000
@@ -26,6 +26,8 @@
#ifndef _PPC64PHP_H
#define _PPC64PHP_H
+
+#include <linux/pci.h>
#include "pci_hotplug.h"
#define DR_INDICATOR 9002
@@ -34,24 +36,22 @@
#define POWER_ON 100
#define POWER_OFF 0
-#define LED_OFF 0
-#define LED_ON 1 /* continuous on */
+#define LED_OFF 0
+#define LED_ON 1 /* continuous on */
#define LED_ID 2 /* slow blinking */
#define LED_ACTION 3 /* fast blinking */
-#define SLOT_NAME_SIZE 12
-
/* Error status from rtas_get-sensor */
-#define NEED_POWER -9000 /* slot must be power up and unisolated to get state */
-#define PWR_ONLY -9001 /* slot must be powerd up to get state, leave isolated */
-#define ERR_SENSE_USE -9002 /* No DR operation will succeed, slot is unusable */
+#define NEED_POWER -9000 /* slot must be power up and unisolated to get state */
+#define PWR_ONLY -9001 /* slot must be powerd up to get state, leave isolated */
+#define ERR_SENSE_USE -9002 /* No DR operation will succeed, slot is unusable */
/* Sensor values from rtas_get-sensor */
-#define EMPTY 0 /* No card in slot */
-#define PRESENT 1 /* Card in slot */
+#define EMPTY 0 /* No card in slot */
+#define PRESENT 1 /* Card in slot */
#define MY_NAME "rpaphp"
-
+extern int debug;
#define dbg(format, arg...) \
do { \
if (debug) \
@@ -64,6 +64,10 @@
#define SLOT_MAGIC 0x67267322
+/* slot types */
+#define VIO_DEV 1
+#define PCI_DEV 2
+
/* slot states */
#define NOT_VALID 3
@@ -75,27 +79,55 @@
* struct slot - slot information for each *physical* slot
*/
struct slot {
- u32 magic;
- int state;
- u32 index;
- u32 type;
- u32 power_domain;
- char *name;
- struct device_node *dn;/* slot's device_node in OFDT */
- /* dn has phb info */
- struct pci_dev *bridge;/* slot's pci_dev in pci_devices */
-
- struct pci_dev *dev; /* pci_dev of device in this slot */
- /* it will be used for unconfig */
- /* NULL if slot is empty */
-
- struct hotplug_slot *hotplug_slot;
- struct list_head rpaphp_slot_list;
+ u32 magic;
+ int state;
+ u32 index;
+ u32 type;
+ u32 power_domain;
+ char *name;
+ struct device_node *dn; /* slot's device_node in OFDT */
+ /* dn has phb info */
+ struct pci_dev *bridge; /* slot's pci_dev in pci_devices */
+ union {
+ struct pci_dev *pci_dev; /* pci_dev of device in this slot */
+ /* it will be used for unconfig */
+ /* NULL if slot is empty */
+ struct vio_dev *vio_dev; /* vio_dev of the device in this slot */
+ } dev;
+ u8 dev_type; /* VIO or PCI */
+ struct hotplug_slot *hotplug_slot;
+ struct list_head rpaphp_slot_list;
};
+extern struct hotplug_slot_ops rpaphp_hotplug_slot_ops;
+extern struct list_head rpaphp_slot_head;
+extern int num_slots;
+
+/* function prototypes */
+
+/* rpaphp_pci.c */
extern struct pci_dev *rpaphp_find_pci_dev(struct device_node *dn);
-extern int rpaphp_add_slot(char *slot_name);
-extern int rpaphp_remove_slot(struct slot *slot);
extern int rpaphp_claim_resource(struct pci_dev *dev, int resource);
+extern int rpaphp_enable_pci_slot(struct slot *slot);
+extern int register_pci_slot(struct slot *slot);
+extern int rpaphp_unconfig_pci_adapter(struct slot *slot);
+extern int rpaphp_get_pci_adapter_status(struct slot *slot, int is_init, u8 * value);
+
+/* rpaphp_core.c */
+extern int rpaphp_add_slot(struct device_node *dn);
+extern int rpaphp_remove_slot(struct slot *slot);
-#endif /* _PPC64PHP_H */
+/* rpaphp_vio.c */
+extern int rpaphp_get_vio_adapter_status(struct slot *slot, int is_init, u8 * value);
+extern int rpaphp_unconfig_vio_adapter(struct slot *slot);
+extern int register_vio_slot(struct device_node *dn);
+extern int rpaphp_enable_vio_slot(struct slot *slot);
+
+/* rpaphp_slot.c */
+extern void dealloc_slot_struct(struct slot *slot);
+extern struct slot *alloc_slot_struct(struct device_node *dn, int drc_index, char *drc_name, int power_domain);
+extern int register_slot(struct slot *slot);
+extern int rpaphp_get_power_status(struct slot *slot, u8 * value);
+extern int rpaphp_set_attention_status(struct slot *slot, u8 status);
+
+#endif /* _PPC64PHP_H */
--- diff/drivers/pci/hotplug/rpaphp_core.c 2004-03-11 10:20:26.000000000 +0000
+++ source/drivers/pci/hotplug/rpaphp_core.c 2004-03-16 11:27:48.221856824 +0000
@@ -31,18 +31,18 @@
#include <linux/smp.h>
#include <linux/smp_lock.h>
#include <linux/init.h>
+#include <asm/eeh.h> /* for eeh_add_device() */
#include <asm/rtas.h> /* rtas_call */
#include <asm/pci-bridge.h> /* for pci_controller */
-#include "../pci.h" /* for pci_add_new_bus*/
- /* and pci_do_scan_bus*/
+#include "../pci.h" /* for pci_add_new_bus */
+ /* and pci_do_scan_bus */
#include "rpaphp.h"
#include "pci_hotplug.h"
-
-static int debug;
+int debug;
static struct semaphore rpaphp_sem;
-static LIST_HEAD (rpaphp_slot_head);
-static int num_slots;
+LIST_HEAD(rpaphp_slot_head);
+int num_slots;
#define DRIVER_VERSION "0.1"
#define DRIVER_AUTHOR "Linda Xie <lxie@us.ibm.com>"
@@ -59,109 +59,35 @@ module_param(debug, int, 0644);
static int enable_slot(struct hotplug_slot *slot);
static int disable_slot(struct hotplug_slot *slot);
static int set_attention_status(struct hotplug_slot *slot, u8 value);
-static int get_power_status(struct hotplug_slot *slot, u8 *value);
-static int get_attention_status(struct hotplug_slot *slot, u8 *value);
-static int get_adapter_status(struct hotplug_slot *slot, u8 *value);
+static int get_power_status(struct hotplug_slot *slot, u8 * value);
+static int get_attention_status(struct hotplug_slot *slot, u8 * value);
+static int get_adapter_status(struct hotplug_slot *slot, u8 * value);
static int get_max_bus_speed(struct hotplug_slot *hotplug_slot, enum pci_bus_speed *value);
static int get_cur_bus_speed(struct hotplug_slot *hotplug_slot, enum pci_bus_speed *value);
-static struct hotplug_slot_ops rpaphp_hotplug_slot_ops = {
- .owner = THIS_MODULE,
- .enable_slot = enable_slot,
- .disable_slot = disable_slot,
- .set_attention_status = set_attention_status,
- .get_power_status = get_power_status,
- .get_attention_status = get_attention_status,
- .get_adapter_status = get_adapter_status,
- .get_max_bus_speed = get_max_bus_speed,
- .get_cur_bus_speed = get_cur_bus_speed,
+struct hotplug_slot_ops rpaphp_hotplug_slot_ops = {
+ .owner = THIS_MODULE,
+ .enable_slot = enable_slot,
+ .disable_slot = disable_slot,
+ .set_attention_status = set_attention_status,
+ .get_power_status = get_power_status,
+ .get_attention_status = get_attention_status,
+ .get_adapter_status = get_adapter_status,
+ .get_max_bus_speed = get_max_bus_speed,
+ .get_cur_bus_speed = get_cur_bus_speed,
};
-static int rpaphp_get_sensor_state(int index, int *state)
-{
- int rc;
-
- rc = rtas_get_sensor(DR_ENTITY_SENSE, index, state);
-
- if (rc) {
- if (rc == NEED_POWER || rc == PWR_ONLY) {
- dbg("%s: slot must be power up to get sensor-state\n",
- __FUNCTION__);
- } else if (rc == ERR_SENSE_USE)
- info("%s: slot is unusable\n", __FUNCTION__);
- else err("%s failed to get sensor state\n", __FUNCTION__);
- }
- return rc;
-}
-
-static struct pci_dev *rpaphp_find_bridge_pdev(struct slot *slot)
-{
- return rpaphp_find_pci_dev(slot->dn);
-}
-
-static struct pci_dev *rpaphp_find_adapter_pdev(struct slot *slot)
-{
- return rpaphp_find_pci_dev(slot->dn->child);
-}
-
-/* Inline functions to check the sanity of a pointer that is passed to us */
-static inline int slot_paranoia_check(struct slot *slot, const char *function)
+static inline struct slot *get_slot (struct hotplug_slot *hotplug_slot, const char *function)
{
- if (!slot) {
- dbg("%s - slot == NULL\n", function);
- return -1;
- }
-
- if (!slot->hotplug_slot) {
- dbg("%s - slot->hotplug_slot == NULL!\n", function);
- return -1;
- }
- return 0;
-}
-
-static inline struct slot *get_slot(struct hotplug_slot *hotplug_slot, const char *function)
-{
- struct slot *slot;
-
if (!hotplug_slot) {
dbg("%s - hotplug_slot == NULL\n", function);
return NULL;
}
-
- slot = (struct slot *)hotplug_slot->private;
- if (slot_paranoia_check(slot, function))
- return NULL;
- return slot;
-}
-
-static inline int rpaphp_set_attention_status(struct slot *slot, u8 status)
-{
- int rc;
-
- /* status: LED_OFF or LED_ON */
- rc = rtas_set_indicator(DR_INDICATOR, slot->index, status);
- if (rc)
- err("slot(%s) set attention-status(%d) failed! rc=0x%x\n",
- slot->name, status, rc);
-
- return rc;
-}
-
-static int rpaphp_get_power_status(struct slot *slot, u8 *value)
-{
- int rc;
-
- rc = rtas_get_power_level(slot->power_domain, (int *)value);
- if (rc)
- err("failed to get power-level for slot(%s), rc=0x%x\n",
- slot->name, rc);
-
- return rc;
+ return (struct slot *)hotplug_slot->private;
}
static int rpaphp_get_attention_status(struct slot *slot)
{
-
return slot->hotplug_slot->info->attention_status;
}
@@ -172,7 +98,7 @@ static int rpaphp_get_attention_status(s
* echo 2 > attention -- set LED ID(identify, light is blinking)
*
*/
-static int set_attention_status (struct hotplug_slot *hotplug_slot, u8 value)
+static int set_attention_status(struct hotplug_slot *hotplug_slot, u8 value)
{
int retval = 0;
struct slot *slot = get_slot(hotplug_slot, __FUNCTION__);
@@ -182,25 +108,21 @@ static int set_attention_status (struct
down(&rpaphp_sem);
switch (value) {
- case 0:
- retval = rpaphp_set_attention_status(slot, LED_OFF);
- hotplug_slot->info->attention_status = 0;
- break;
-
- case 1:
- default:
- retval = rpaphp_set_attention_status(slot, LED_ON);
- hotplug_slot->info->attention_status = 1;
- break;
-
- case 2:
- retval = rpaphp_set_attention_status(slot, LED_ID);
- hotplug_slot->info->attention_status = 2;
- break;
-
+ case 0:
+ retval = rpaphp_set_attention_status(slot, LED_OFF);
+ hotplug_slot->info->attention_status = 0;
+ break;
+ case 1:
+ default:
+ retval = rpaphp_set_attention_status(slot, LED_ON);
+ hotplug_slot->info->attention_status = 1;
+ break;
+ case 2:
+ retval = rpaphp_set_attention_status(slot, LED_ID);
+ hotplug_slot->info->attention_status = 2;
+ break;
}
up(&rpaphp_sem);
-
return retval;
}
@@ -211,7 +133,7 @@ static int set_attention_status (struct
*
*
*/
-static int get_power_status (struct hotplug_slot *hotplug_slot, u8 *value)
+static int get_power_status(struct hotplug_slot *hotplug_slot, u8 * value)
{
int retval;
struct slot *slot = get_slot(hotplug_slot, __FUNCTION__);
@@ -222,7 +144,6 @@ static int get_power_status (struct hotp
down(&rpaphp_sem);
retval = rpaphp_get_power_status(slot, value);
up(&rpaphp_sem);
-
return retval;
}
@@ -231,7 +152,7 @@ static int get_power_status (struct hotp
*
*
*/
-static int get_attention_status (struct hotplug_slot *hotplug_slot, u8 *value)
+static int get_attention_status(struct hotplug_slot *hotplug_slot, u8 * value)
{
int retval = 0;
struct slot *slot = get_slot(hotplug_slot, __FUNCTION__);
@@ -239,81 +160,36 @@ static int get_attention_status (struct
if (slot == NULL)
return -ENODEV;
-
down(&rpaphp_sem);
*value = rpaphp_get_attention_status(slot);
up(&rpaphp_sem);
-
return retval;
}
-/*
- * get_adapter_status - get the status of a slot
- *
- * 0-- slot is empty
- * 1-- adapter is configured
- * 2-- adapter is not configured
- * 3-- not valid
- */
-static int rpaphp_get_adapter_status(struct slot *slot, int is_init, u8 *value)
-{
- int state, rc;
-
- *value = NOT_VALID;
-
- rc = rpaphp_get_sensor_state(slot->index, &state);
-
- if (rc)
- return rc;
-
- if (state == PRESENT) {
- dbg("slot is occupied\n");
-
- if (!is_init) /* at run-time slot->state can be changed by */
- /* config/unconfig adapter */
- *value = slot->state;
- else {
- if (!slot->dn->child)
- dbg("%s: %s is not valid OFDT node\n",
- __FUNCTION__, slot->dn->full_name);
- else
- if (rpaphp_find_pci_dev(slot->dn->child))
- *value = CONFIGURED;
- else {
- dbg("%s: can't find pdev of adapter in slot[%s]\n",
- __FUNCTION__, slot->name);
- *value = NOT_CONFIGURED;
- }
- }
- } else
- if (state == EMPTY) {
- dbg("slot is empty\n");
- *value = state;
- }
-
- return 0;
-}
-
-static int get_adapter_status (struct hotplug_slot *hotplug_slot, u8 *value)
+static int get_adapter_status(struct hotplug_slot *hotplug_slot, u8 * value)
{
struct slot *slot = get_slot(hotplug_slot, __FUNCTION__);
int retval = 0;
if (slot == NULL)
return -ENODEV;
-
down(&rpaphp_sem);
-
/* have to go through this */
- retval = rpaphp_get_adapter_status(slot, 0, value);
-
+ switch (slot->dev_type) {
+ case PCI_DEV:
+ retval = rpaphp_get_pci_adapter_status(slot, 0, value);
+ break;
+ case VIO_DEV:
+ retval = rpaphp_get_vio_adapter_status(slot, 0, value);
+ break;
+ default:
+ retval = -EINVAL;
+ }
up(&rpaphp_sem);
-
return retval;
}
-
-static int get_max_bus_speed (struct hotplug_slot *hotplug_slot, enum pci_bus_speed *value)
+static int get_max_bus_speed(struct hotplug_slot *hotplug_slot, enum pci_bus_speed *value)
{
struct slot *slot = get_slot(hotplug_slot, __FUNCTION__);
@@ -321,46 +197,42 @@ static int get_max_bus_speed (struct hot
return -ENODEV;
down(&rpaphp_sem);
-
switch (slot->type) {
- case 1:
- case 2:
- case 3:
- case 4:
- case 5:
- case 6:
- *value = PCI_SPEED_33MHz; /* speed for case 1-6 */
- break;
- case 7:
- case 8:
- *value = PCI_SPEED_66MHz;
- break;
- case 11:
- case 14:
- *value = PCI_SPEED_66MHz_PCIX;
- break;
- case 12:
- case 15:
- *value = PCI_SPEED_100MHz_PCIX;
- break;
- case 13:
- case 16:
- *value = PCI_SPEED_133MHz_PCIX;
- break;
- default:
- *value = PCI_SPEED_UNKNOWN;
- break;
+ case 1:
+ case 2:
+ case 3:
+ case 4:
+ case 5:
+ case 6:
+ *value = PCI_SPEED_33MHz; /* speed for case 1-6 */
+ break;
+ case 7:
+ case 8:
+ *value = PCI_SPEED_66MHz;
+ break;
+ case 11:
+ case 14:
+ *value = PCI_SPEED_66MHz_PCIX;
+ break;
+ case 12:
+ case 15:
+ *value = PCI_SPEED_100MHz_PCIX;
+ break;
+ case 13:
+ case 16:
+ *value = PCI_SPEED_133MHz_PCIX;
+ break;
+ default:
+ *value = PCI_SPEED_UNKNOWN;
+ break;
}
-
up(&rpaphp_sem);
-
return 0;
}
-
/* return dummy value because not sure if PRA provides any method... */
-static int get_cur_bus_speed (struct hotplug_slot *hotplug_slot, enum pci_bus_speed *value)
+static int get_cur_bus_speed(struct hotplug_slot *hotplug_slot, enum pci_bus_speed *value)
{
struct slot *slot = get_slot(hotplug_slot, __FUNCTION__);
@@ -368,416 +240,106 @@ static int get_cur_bus_speed (struct hot
return -ENODEV;
*value = PCI_SPEED_UNKNOWN;
-
return 0;
}
-/*
- * rpaphp_validate_slot - make sure the name of the slot matches
- * the location code , if the slots is not
- * empty.
- */
-static int rpaphp_validate_slot(const char *slot_name, const int slot_index)
-{
- struct device_node *dn;
-
- for(dn = find_all_nodes(); dn; dn = dn->next) {
-
- int *index;
- unsigned char *loc_code;
-
- index = (int *)get_property(dn, "ibm,my-drc-index", NULL);
-
- if (index && *index == slot_index) {
- char *slash, *tmp_str;
-
- loc_code = get_property(dn, "ibm,loc-code", NULL);
- if (!loc_code) {
- return -1;
- }
-
- tmp_str = kmalloc(MAX_LOC_CODE, GFP_KERNEL);
- if (!tmp_str) {
- err("%s: out of memory\n", __FUNCTION__);
- return -1;
- }
-
- strcpy(tmp_str, loc_code);
- slash = strrchr(tmp_str, '/');
- if (slash)
- *slash = '\0';
-
- if (strcmp(slot_name, tmp_str)) {
- kfree(tmp_str);
- return -1;
- }
-
- kfree(tmp_str);
- break;
- }
- }
-
- return 0;
-}
-
-/* Must be called before pci_bus_add_devices */
-static void rpaphp_fixup_new_devices(struct pci_bus *bus)
-{
- struct pci_dev *dev;
-
- list_for_each_entry(dev, &bus->devices, bus_list) {
- /*
- * Skip already-present devices (which are on the
- * global device list.)
- */
- if (list_empty(&dev->global_list)) {
- int i;
- pcibios_fixup_device_resources(dev, bus);
- pci_read_irq_line(dev);
- for (i = 0; i < PCI_NUM_RESOURCES; i++) {
- struct resource *r = &dev->resource[i];
- if (r->parent || !r->start || !r->flags)
- continue;
- rpaphp_claim_resource(dev, i);
- }
- }
- }
-}
-
-static struct pci_dev *rpaphp_config_adapter(struct slot *slot)
-{
- struct pci_bus *pci_bus;
- struct device_node *dn;
- int num;
- struct pci_dev *dev = NULL;
-
- if (slot->bridge) {
-
- pci_bus = slot->bridge->subordinate;
-
- if (!pci_bus) {
- err("%s: can't find bus structure\n", __FUNCTION__);
- goto exit;
- }
-
- for (dn = slot->dn->child; dn; dn = dn->sibling) {
- dbg("child dn's devfn=[%x]\n", dn->devfn);
- num = pci_scan_slot(pci_bus,
- PCI_DEVFN(PCI_SLOT(dn->devfn), 0));
-
- dbg("pci_scan_slot return num=%d\n", num);
-
- if (num) {
- rpaphp_fixup_new_devices(pci_bus);
- pci_bus_add_devices(pci_bus);
- }
- }
-
- dev = rpaphp_find_pci_dev(slot->dn->child);
- } else {
- /* slot is not enabled */
- err("slot doesn't have pci_dev structure\n");
- dev = NULL;
- goto exit;
- }
-
-exit:
- dbg("Exit %s: pci_dev %s\n", __FUNCTION__, dev? "found":"not found");
-
- return dev;
-}
-
-static int rpaphp_unconfig_adapter(struct slot *slot)
-{
- if (!slot->dev) {
- info("%s: no card in slot[%s]\n",
- __FUNCTION__, slot->name);
-
- return -EINVAL;
- }
-
- /* remove the device from the pci core */
- pci_remove_bus_device(slot->dev);
-
- pci_dev_put(slot->dev);
- slot->state = NOT_CONFIGURED;
-
- dbg("%s: adapter in slot[%s] unconfigured.\n", __FUNCTION__, slot->name);
-
- return 0;
-}
-
-/* free up the memory user be a slot */
-
-static void rpaphp_release_slot(struct hotplug_slot *hotplug_slot)
-{
- struct slot *slot = get_slot(hotplug_slot, __FUNCTION__);
-
- if (slot == NULL)
- return;
-
- kfree(slot->hotplug_slot->info);
- kfree(slot->hotplug_slot->name);
- kfree(slot->hotplug_slot);
- pci_dev_put(slot->bridge);
- pci_dev_put(slot->dev);
- kfree(slot);
-}
-
int rpaphp_remove_slot(struct slot *slot)
{
int retval = 0;
+ char *rm_link;
- sysfs_remove_link(slot->hotplug_slot->kobj.parent,
- slot->bridge->slot_name);
+ dbg("%s - Entry: slot[%s]\n", __FUNCTION__, slot->name);
+ if (slot->dev_type == PCI_DEV)
+ rm_link = pci_name(slot->bridge);
+ else
+ rm_link = strstr(slot->dn->full_name, "@");
+ sysfs_remove_link(slot->hotplug_slot->kobj.parent, rm_link);
list_del(&slot->rpaphp_slot_list);
retval = pci_hp_deregister(slot->hotplug_slot);
if (retval)
err("Problem unregistering a slot %s\n", slot->name);
+
num_slots--;
+ dbg("%s - Exit: rc[%d]\n", __FUNCTION__, retval);
return retval;
}
-static int is_php_dn(struct device_node *dn, int **indexes, int **names, int **types, int **power_domains)
+static int is_php_dn(struct device_node *dn, int **indexes, int **names, int **types,
+ int **power_domains)
{
- *indexes = (int *)get_property(dn, "ibm,drc-indexes", NULL);
+ *indexes = (int *) get_property(dn, "ibm,drc-indexes", NULL);
if (!*indexes)
- return(0);
-
+ return (0);
/* &names[1] contains NULL terminated slot names */
- *names = (int *)get_property(dn, "ibm,drc-names", NULL);
+ *names = (int *) get_property(dn, "ibm,drc-names", NULL);
if (!*names)
- return(0);
-
+ return (0);
/* &types[1] contains NULL terminated slot types */
- *types = (int *)get_property(dn, "ibm,drc-types", NULL);
+ *types = (int *) get_property(dn, "ibm,drc-types", NULL);
if (!*types)
- return(0);
-
+ return (0);
/* power_domains[1...n] are the slot power domains */
- *power_domains = (int *)get_property(dn,
- "ibm,drc-power-domains", NULL);
+ *power_domains = (int *) get_property(dn,
+ "ibm,drc-power-domains", NULL);
if (!*power_domains)
- return(0);
-
- if (!get_property(dn, "ibm,fw-pci-hot-plug-ctrl", NULL))
- return(0);
-
- return(1);
-}
-
-static struct slot *alloc_slot_struct(void)
-{
- struct slot *slot;
-
- slot = kmalloc(sizeof(struct slot), GFP_KERNEL);
- if (!slot)
- return (NULL);
- memset(slot, 0, sizeof(struct slot));
- slot->hotplug_slot = kmalloc(sizeof(struct hotplug_slot),
- GFP_KERNEL);
- if (!slot->hotplug_slot) {
- kfree(slot);
- return (NULL);
- }
- memset(slot->hotplug_slot, 0, sizeof(struct hotplug_slot));
- slot->hotplug_slot->info = kmalloc(sizeof(struct hotplug_slot_info),
- GFP_KERNEL);
- if (!slot->hotplug_slot->info) {
- kfree(slot->hotplug_slot);
- kfree(slot);
- return (NULL);
- }
- memset(slot->hotplug_slot->info, 0, sizeof(struct hotplug_slot_info));
- slot->hotplug_slot->name = kmalloc(SLOT_NAME_SIZE, GFP_KERNEL);
- if (!slot->hotplug_slot->name) {
- kfree(slot->hotplug_slot->info);
- kfree(slot->hotplug_slot);
- kfree(slot);
- return (NULL);
- }
- return (slot);
+ return (0);
+ if (strcmp(dn->name, "pci") == 0 &&
+ !get_property(dn, "ibm,fw-pci-hot-plug-ctrl", NULL))
+ return (0);
+ return (1);
}
-static int setup_hotplug_slot_info(struct slot *slot)
+static inline int is_vdevice_root(struct device_node *dn)
{
- rpaphp_get_power_status(slot,
- &slot->hotplug_slot->info->power_status);
-
- rpaphp_get_adapter_status(slot, 1,
- &slot->hotplug_slot->info->adapter_status);
-
- if (slot->hotplug_slot->info->adapter_status == NOT_VALID) {
- dbg("%s: NOT_VALID: skip dn->full_name=%s\n",
- __FUNCTION__, slot->dn->full_name);
- kfree(slot->hotplug_slot->info);
- kfree(slot->hotplug_slot->name);
- kfree(slot->hotplug_slot);
- kfree(slot);
- return (-1);
- }
- return (0);
-}
-
-static int register_slot(struct slot *slot)
-{
- int retval;
-
- retval = pci_hp_register(slot->hotplug_slot);
- if (retval) {
- err("pci_hp_register failed with error %d\n", retval);
- rpaphp_release_slot(slot->hotplug_slot);
- return (retval);
- }
- /* create symlink between slot->name and it's bus_id */
- dbg("%s: sysfs_create_link: %s --> %s\n", __FUNCTION__,
- slot->bridge->slot_name, slot->name);
- retval = sysfs_create_link(slot->hotplug_slot->kobj.parent,
- &slot->hotplug_slot->kobj,
- slot->bridge->slot_name);
- if (retval) {
- err("sysfs_create_link failed with error %d\n", retval);
- rpaphp_release_slot(slot->hotplug_slot);
- return (retval);
- }
- /* add slot to our internal list */
- dbg("%s adding slot[%s] to rpaphp_slot_list\n",
- __FUNCTION__, slot->name);
-
- list_add(&slot->rpaphp_slot_list, &rpaphp_slot_head);
-
- info("Slot [%s] (bus_id=%s) registered\n",
- slot->name, slot->bridge->slot_name);
- return (0);
+ return !strcmp(dn->name, "vdevice");
}
/*************************************
* Add Hot Plug slot(s) to sysfs
*
************************************/
-int rpaphp_add_slot(char *slot_name)
+int rpaphp_add_slot(struct device_node *dn)
{
- struct slot *slot;
- int retval = 0;
- int i;
- struct device_node *dn;
- int *indexes, *names, *types, *power_domains;
- char *name, *type;
-
- for (dn = find_all_nodes(); dn; dn = dn->next) {
-
- if (dn->name != 0 && strcmp(dn->name, "pci") == 0) {
- if (!is_php_dn(dn, &indexes, &names, &types, &power_domains))
- continue;
-
- dbg("%s : found device_node in OFDT full_name=%s, name=%s\n",
- __FUNCTION__, dn->full_name, dn->name);
-
- name = (char *)&names[1];
- type = (char *)&types[1];
-
- for (i = 0; i < indexes[0];
- i++,
- name += (strlen(name) + 1),
- type += (strlen(type) + 1)) {
-
- dbg("%s: name[%s] index[%x]\n",
- __FUNCTION__, name, indexes[i+1]);
-
- if (slot_name && strcmp(slot_name, name))
- continue;
-
- if (rpaphp_validate_slot(name, indexes[i + 1])) {
- dbg("%s: slot(%s, 0x%x) is invalid.\n",
- __FUNCTION__, name, indexes[i+ 1]);
- continue;
- }
-
- slot = alloc_slot_struct();
- if (!slot) {
- retval = -ENOMEM;
- goto exit;
- }
-
- slot->name = slot->hotplug_slot->name;
- slot->index = indexes[i + 1];
- strcpy(slot->name, name);
- slot->type = simple_strtoul(type, NULL, 10);
- if (slot->type < 1 || slot->type > 16)
- slot->type = 0;
-
- slot->power_domain = power_domains[i + 1];
- slot->magic = SLOT_MAGIC;
- slot->hotplug_slot->private = slot;
- slot->hotplug_slot->ops = &rpaphp_hotplug_slot_ops;
- slot->hotplug_slot->release = &rpaphp_release_slot;
- slot->dn = dn;
-
- /*
- * Initilize the slot info structure with some known
- * good values.
- */
- if (setup_hotplug_slot_info(slot))
- continue;
-
- slot->bridge = rpaphp_find_bridge_pdev(slot);
- if (!slot->bridge && slot_name) { /* slot being added doesn't have pci_dev yet*/
- dbg("%s: no pci_dev for bridge dn %s\n",
- __FUNCTION__, slot_name);
- kfree(slot->hotplug_slot->info);
- kfree(slot->hotplug_slot->name);
- kfree(slot->hotplug_slot);
- kfree(slot);
- continue;
- }
-
- /* find slot's pci_dev if it's not empty*/
- if (slot->hotplug_slot->info->adapter_status == EMPTY) {
- slot->state = EMPTY; /* slot is empty */
- slot->dev = NULL;
- } else { /* slot is occupied */
- if(!(slot->dn->child)) { /* non-empty slot has to have child */
- err("%s: slot[%s]'s device_node doesn't have child for adapter\n",
- __FUNCTION__, slot->name);
- kfree(slot->hotplug_slot->info);
- kfree(slot->hotplug_slot->name);
- kfree(slot->hotplug_slot);
- kfree(slot);
- continue;
-
- }
-
- slot->dev = rpaphp_find_adapter_pdev(slot);
- if(slot->dev) {
- slot->state = CONFIGURED;
- pci_dev_get(slot->dev);
- } else {
- /* DLPAR add as opposed to
- * boot time */
- slot->state = NOT_CONFIGURED;
- }
- }
- dbg("%s registering slot:path[%s] index[%x], name[%s] pdomain[%x] type[%d]\n",
- __FUNCTION__, dn->full_name, slot->index, slot->name,
- slot->power_domain, slot->type);
-
- retval = register_slot(slot);
- if (retval)
- goto exit;
-
- num_slots++;
-
- if (slot_name)
- goto exit;
-
- }/* for indexes */
- }/* "pci" */
- }/* find_all_nodes */
-exit:
+ struct slot *slot;
+ int retval = 0;
+ int i;
+ int *indexes, *names, *types, *power_domains;
+ char *name, *type;
+
+ dbg("Entry %s: dn->full_name=%s\n", __FUNCTION__, dn->full_name);
+
+ if (dn->parent && is_vdevice_root(dn->parent)) {
+ /* register a VIO device */
+ retval = register_vio_slot(dn);
+ goto exit;
+ }
+
+ /* register PCI devices */
+ if (dn->name != 0 && strcmp(dn->name, "pci") == 0 &&
+ is_php_dn(dn, &indexes, &names, &types, &power_domains)) {
+
+ name = (char *) &names[1];
+ type = (char *) &types[1];
+ for (i = 0; i < indexes[0];
+ i++,
+ name += (strlen(name) + 1), type += (strlen(type) + 1)) {
+ if (!(slot = alloc_slot_struct(dn, indexes[i + 1], name,
+ power_domains[i + 1]))) {
+ retval = -ENOMEM;
+ goto exit;
+ }
+ slot->type = simple_strtoul(type, NULL, 10);
+ if (slot->type < 1 || slot->type > 16)
+ slot->type = 0;
+ retval = register_pci_slot(slot);
+
+ } /* for indexes */
+ } /* end of PCI device_node */
+ exit:
dbg("%s - Exit: num_slots=%d rc[%d]\n",
- __FUNCTION__, num_slots, retval);
+ __FUNCTION__, num_slots, retval);
return retval;
}
@@ -785,31 +347,28 @@ exit:
* init_slots - initialize 'struct slot' structures for each slot
*
*/
-static int init_slots (void)
+static void init_slots(void)
{
- int retval = 0;
+ struct device_node *dn;
- retval = rpaphp_add_slot(NULL);
-
- return retval;
+ for (dn = find_all_nodes(); dn; dn = dn->next)
+ rpaphp_add_slot(dn);
}
-
-static int init_rpa (void)
+static int init_rpa(void)
{
- int retval = 0;
init_MUTEX(&rpaphp_sem);
/* initialize internal data structure etc. */
- retval = init_slots();
+ init_slots();
if (!num_slots)
- retval = -ENODEV;
+ return -ENODEV;
- return retval;
+ return 0;
}
-static void cleanup_slots (void)
+static void cleanup_slots(void)
{
struct list_head *tmp, *n;
struct slot *slot;
@@ -817,125 +376,100 @@ static void cleanup_slots (void)
/*
* Unregister all of our slots with the pci_hotplug subsystem,
* and free up all memory that we had allocated.
- * memory will be freed in release_slot callback.
+ * memory will be freed in release_slot callback.
*/
- list_for_each_safe (tmp, n, &rpaphp_slot_head) {
+ list_for_each_safe(tmp, n, &rpaphp_slot_head) {
+ char *rm_link;
+
slot = list_entry(tmp, struct slot, rpaphp_slot_list);
- sysfs_remove_link(slot->hotplug_slot->kobj.parent,
- slot->bridge->slot_name);
+ if (slot->dev_type == PCI_DEV)
+ rm_link = pci_name(slot->bridge);
+ else
+ rm_link = strstr(slot->dn->full_name, "@");
+ sysfs_remove_link(slot->hotplug_slot->kobj.parent, rm_link);
list_del(&slot->rpaphp_slot_list);
pci_hp_deregister(slot->hotplug_slot);
}
-
return;
}
-
static int __init rpaphp_init(void)
{
- int retval = 0;
-
info(DRIVER_DESC " version: " DRIVER_VERSION "\n");
/* read all the PRA info from the system */
- retval = init_rpa();
-
- return retval;
+ return init_rpa();
}
-
static void __exit rpaphp_exit(void)
{
cleanup_slots();
}
-
static int enable_slot(struct hotplug_slot *hotplug_slot)
{
- int retval = 0, state;
-
+ int retval = 0;
struct slot *slot = get_slot(hotplug_slot, __FUNCTION__);
if (slot == NULL)
return -ENODEV;
if (slot->state == CONFIGURED) {
- dbg("%s: %s is already enabled\n",
- __FUNCTION__, slot->name);
+ dbg("%s: %s is already enabled\n", __FUNCTION__, slot->name);
goto exit;
}
dbg("ENABLING SLOT %s\n", slot->name);
-
down(&rpaphp_sem);
-
- retval = rpaphp_get_sensor_state(slot->index, &state);
-
- if (retval)
- goto exit;
-
- dbg("%s: sensor state[%d]\n", __FUNCTION__, state);
-
- /* if slot is not empty, enable the adapter */
- if (state == PRESENT) {
- dbg("%s : slot[%s] is occupid.\n", __FUNCTION__, slot->name);
-
-
- slot->dev = rpaphp_config_adapter(slot);
- if (slot->dev != NULL) {
- slot->state = CONFIGURED;
-
- dbg("%s: adapter %s in slot[%s] has been configured\n",
- __FUNCTION__, slot->dev->slot_name,
- slot->name);
- } else {
- slot->state = NOT_CONFIGURED;
-
- dbg("%s: no pci_dev struct for adapter in slot[%s]\n",
- __FUNCTION__, slot->name);
- }
-
- } else if (state == EMPTY) {
- dbg("%s : slot[%s] is empty\n", __FUNCTION__, slot->name);
- slot->state = EMPTY;
- } else {
- err("%s: slot[%s] is in invalid state\n", __FUNCTION__, slot->name);
- slot->state = NOT_VALID;
+ switch (slot->dev_type) {
+ case PCI_DEV:
+ retval = rpaphp_enable_pci_slot(slot);
+ break;
+ case VIO_DEV:
+ retval = rpaphp_enable_vio_slot(slot);
+ break;
+ default:
retval = -EINVAL;
}
-
-exit:
- if (slot->state != NOT_VALID)
- rpaphp_set_attention_status(slot, LED_ON);
- else
- rpaphp_set_attention_status(slot, LED_ID);
-
up(&rpaphp_sem);
-
+ exit:
+ dbg("%s - Exit: rc[%d]\n", __FUNCTION__, retval);
return retval;
}
static int disable_slot(struct hotplug_slot *hotplug_slot)
{
- int retval;
+ int retval;
struct slot *slot = get_slot(hotplug_slot, __FUNCTION__);
if (slot == NULL)
return -ENODEV;
- dbg("DISABLING SLOT %s\n", slot->name);
-
- down(&rpaphp_sem);
-
- rpaphp_set_attention_status(slot, LED_ID);
+ dbg("%s - Entry: slot[%s]\n", __FUNCTION__, slot->name);
- retval = rpaphp_unconfig_adapter(slot);
-
- rpaphp_set_attention_status(slot, LED_OFF);
+ if (slot->state == NOT_CONFIGURED) {
+ dbg("%s: %s is already disabled\n", __FUNCTION__, slot->name);
+ goto exit;
+ }
+ dbg("DISABLING SLOT %s\n", slot->name);
+ down(&rpaphp_sem);
+ switch (slot->dev_type) {
+ case PCI_DEV:
+ rpaphp_set_attention_status(slot, LED_ID);
+ retval = rpaphp_unconfig_pci_adapter(slot);
+ rpaphp_set_attention_status(slot, LED_OFF);
+ break;
+ case VIO_DEV:
+ retval = rpaphp_unconfig_vio_adapter(slot);
+ break;
+ default:
+ retval = -ENODEV;
+ }
up(&rpaphp_sem);
-
+ exit:
+ dbg("%s - Exit: rc[%d]\n", __FUNCTION__, retval);
return retval;
}
--- diff/drivers/pci/hotplug/rpaphp_pci.c 2004-03-11 10:20:26.000000000 +0000
+++ source/drivers/pci/hotplug/rpaphp_pci.c 2004-03-16 11:27:48.223856520 +0000
@@ -23,32 +23,35 @@
*
*/
#include <linux/pci.h>
-#include <asm/pci-bridge.h> /* for pci_controller */
-#include "rpaphp.h"
+#include <asm/pci-bridge.h>
+#include "../pci.h" /* for pci_add_new_bus */
+#include "rpaphp.h"
struct pci_dev *rpaphp_find_pci_dev(struct device_node *dn)
{
- struct pci_dev *retval_dev = NULL, *dev = NULL;
-
- while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
- if(!dev->bus)
- continue;
+ struct pci_dev *retval_dev = NULL, *dev = NULL;
+ char bus_id[BUS_ID_SIZE];
- if (dev->devfn != dn->devfn)
- continue;
+ sprintf(bus_id, "%04x:%02x:%02x.%d",dn->phb->global_number,
+ dn->busno, PCI_SLOT(dn->devfn), PCI_FUNC(dn->devfn));
- if (dn->phb->global_number == pci_domain_nr(dev->bus) &&
- dn->busno == dev->bus->number) {
+ dbg("Enter rpaphp_find_pci_dev() full_name=%s bus_id=%s\n",
+ dn->full_name, bus_id);
+
+ while ((dev = pci_find_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
+ if (!strcmp(pci_name(dev), bus_id)) {
retval_dev = dev;
+ dbg("rpaphp_find_pci_dev(): found dev=%p\n\n", dev);
break;
}
}
-
return retval_dev;
}
+EXPORT_SYMBOL_GPL(rpaphp_find_pci_dev);
+
int rpaphp_claim_resource(struct pci_dev *dev, int resource)
{
struct resource *res = &dev->resource[resource];
@@ -63,13 +66,333 @@ int rpaphp_claim_resource(struct pci_dev
if (err) {
err("PCI: %s region %d of %s %s [%lx:%lx]\n",
- root ? "Address space collision on" :
- "No parent found for",
- resource, dtype, pci_name(dev), res->start, res->end);
+ root ? "Address space collision on" :
+ "No parent found for",
+ resource, dtype, pci_name(dev), res->start, res->end);
}
-
return err;
}
-EXPORT_SYMBOL_GPL(rpaphp_find_pci_dev);
EXPORT_SYMBOL_GPL(rpaphp_claim_resource);
+
+static struct pci_dev *rpaphp_find_bridge_pdev(struct slot *slot)
+{
+ return rpaphp_find_pci_dev(slot->dn);
+}
+
+static struct pci_dev *rpaphp_find_adapter_pdev(struct slot *slot)
+{
+ return rpaphp_find_pci_dev(slot->dn->child);
+}
+
+static int rpaphp_get_sensor_state(struct slot *slot, int *state)
+{
+ int rc;
+ int setlevel;
+
+ rc = rtas_get_sensor(DR_ENTITY_SENSE, slot->index, state);
+
+ if (rc) {
+ if (rc == NEED_POWER || rc == PWR_ONLY) {
+ dbg("%s: slot must be power up to get sensor-state\n",
+ __FUNCTION__);
+
+ /* some slots have to be powered up
+ * before get-sensor will succeed.
+ */
+ rc = rtas_set_power_level(slot->power_domain, POWER_ON,
+ &setlevel);
+ if (rc) {
+ dbg("%s: power on slot[%s] failed rc=%d.\n",
+ __FUNCTION__, slot->name, rc);
+ } else {
+ rc = rtas_get_sensor(DR_ENTITY_SENSE,
+ slot->index, state);
+ }
+ } else if (rc == ERR_SENSE_USE)
+ info("%s: slot is unusable\n", __FUNCTION__);
+ else
+ err("%s failed to get sensor state\n", __FUNCTION__);
+ }
+ return rc;
+}
+
+/*
+ * get_pci_adapter_status - get the status of a slot
+ *
+ * 0-- slot is empty
+ * 1-- adapter is configured
+ * 2-- adapter is not configured
+ * 3-- not valid
+ */
+int rpaphp_get_pci_adapter_status(struct slot *slot, int is_init, u8 * value)
+{
+ int state, rc;
+ *value = NOT_VALID;
+
+ rc = rpaphp_get_sensor_state(slot, &state);
+ if (rc)
+ goto exit;
+ if (state == PRESENT) {
+ if (!is_init)
+ /* at run-time slot->state can be changed by */
+ /* config/unconfig adapter */
+ *value = slot->state;
+ else {
+ if (!slot->dn->child)
+ dbg("%s: %s is not valid OFDT node\n",
+ __FUNCTION__, slot->dn->full_name);
+ else if (rpaphp_find_pci_dev(slot->dn->child))
+ *value = CONFIGURED;
+ else {
+ dbg("%s: can't find pdev of adapter in slot[%s]\n", __FUNCTION__, slot->name);
+ *value = NOT_CONFIGURED;
+ }
+ }
+ } else if (state == EMPTY) {
+ dbg("slot is empty\n");
+ *value = state;
+ }
+
+ exit:
+ return rc;
+}
+
+/* Must be called before pci_bus_add_devices */
+static void rpaphp_fixup_new_pci_devices(struct pci_bus *bus)
+{
+ struct pci_dev *dev;
+
+ list_for_each_entry(dev, &bus->devices, bus_list) {
+ /*
+ * Skip already-present devices (which are on the
+ * global device list.)
+ */
+ if (list_empty(&dev->global_list)) {
+ int i;
+
+ pcibios_fixup_device_resources(dev, bus);
+ pci_read_irq_line(dev);
+ for (i = 0; i < PCI_NUM_RESOURCES; i++) {
+ struct resource *r = &dev->resource[i];
+ if (r->parent || !r->start || !r->flags)
+ continue;
+ rpaphp_claim_resource(dev, i);
+ }
+ }
+ }
+}
+
+static void rpaphp_pci_config_device(struct pci_bus *pci_bus, struct device_node *dn)
+{
+ int num;
+
+ num = pci_scan_slot(pci_bus, PCI_DEVFN(PCI_SLOT(dn->devfn), 0));
+ if (num) {
+ rpaphp_fixup_new_pci_devices(pci_bus);
+ pci_bus_add_devices(pci_bus);
+ }
+ return;
+}
+
+static int rpaphp_pci_config_bridge(struct pci_dev *dev, struct device_node *dn);
+
+/*****************************************************************************
+ rpaphp_pci_config_dn() will recursively configure all devices under the
+ given slot->dn and return the dn's pci_dev.
+ *****************************************************************************/
+static struct pci_dev *rpaphp_pci_config_dn(struct device_node *dn, struct pci_bus *bus)
+{
+ struct device_node *local;
+ struct pci_dev *dev;
+
+ for (local = dn->child; local; local = local->sibling) {
+ rpaphp_pci_config_device(bus, local);
+ dev = rpaphp_find_pci_dev(local);
+ if (!rpaphp_pci_config_bridge(dev, local))
+ return NULL;
+ }
+
+ return dev;
+}
+
+static int rpaphp_pci_config_bridge(struct pci_dev *dev, struct device_node *dn)
+{
+ if (dev && dn->child) { /* dn is a PCI bridge node */
+ struct pci_bus *child;
+ u8 sec_busno;
+
+ /* get busno of downstream bus */
+ pci_read_config_byte(dev, PCI_SECONDARY_BUS, &sec_busno);
+
+ /* add to children of PCI bridge dev->bus */
+ child = pci_add_new_bus(dev->bus, dev, sec_busno);
+ if (!child) {
+ err("%s: could not add second bus\n", __FUNCTION__);
+ return 0;
+ }
+ sprintf(child->name, "PCI Bus #%02x", child->number);
+ /* Fixup subordinate bridge bases and resureces */
+ pcibios_fixup_bus(child);
+
+ /* may need do more stuff here */
+ rpaphp_pci_config_dn(dn, dev->subordinate);
+ }
+ return 1;
+}
+
+static struct pci_dev *rpaphp_config_pci_adapter(struct slot *slot)
+{
+ struct pci_bus *pci_bus;
+ struct pci_dev *dev = NULL;
+
+ dbg("Entry %s: slot[%s]\n", __FUNCTION__, slot->name);
+
+ if (slot->bridge) {
+
+ pci_bus = slot->bridge->subordinate;
+ if (!pci_bus) {
+ err("%s: can't find bus structure\n", __FUNCTION__);
+ goto exit;
+ }
+
+ dev = rpaphp_pci_config_dn(slot->dn, pci_bus);
+ } else {
+ /* slot is not enabled */
+ err("slot doesn't have pci_dev structure\n");
+ dev = NULL;
+ goto exit;
+ }
+
+ exit:
+ dbg("Exit %s: pci_dev %s\n", __FUNCTION__, dev ? "found" : "not found");
+ return dev;
+}
+
+int rpaphp_unconfig_pci_adapter(struct slot *slot)
+{
+ int retval = 0;
+
+ dbg("Entry %s: slot[%s]\n", __FUNCTION__, slot->name);
+ if (!slot->dev.pci_dev) {
+ info("%s: no card in slot[%s]\n", __FUNCTION__, slot->name);
+
+ retval = -EINVAL;
+ goto exit;
+ }
+ /* remove the device from the pci core */
+ pci_remove_bus_device(slot->dev.pci_dev);
+
+ slot->state = NOT_CONFIGURED;
+ info("%s: adapter in slot[%s] unconfigured.\n", __FUNCTION__,
+ slot->name);
+exit:
+ dbg("Exit %s, rc=0x%x\n", __FUNCTION__, retval);
+ return retval;
+}
+
+static int setup_pci_hotplug_slot_info(struct slot *slot)
+{
+ dbg("%s Initilize the PCI slot's hotplug->info structure ...\n",
+ __FUNCTION__);
+ rpaphp_get_power_status(slot, &slot->hotplug_slot->info->power_status);
+ rpaphp_get_pci_adapter_status(slot, 1,
+ &slot->hotplug_slot->info->
+ adapter_status);
+ if (slot->hotplug_slot->info->adapter_status == NOT_VALID) {
+ dbg("%s: NOT_VALID: skip dn->full_name=%s\n",
+ __FUNCTION__, slot->dn->full_name);
+ dealloc_slot_struct(slot);
+ return (-1);
+ }
+ return (0);
+}
+
+static int setup_pci_slot(struct slot *slot)
+{
+ slot->bridge = rpaphp_find_bridge_pdev(slot);
+ if (!slot->bridge) { /* slot being added doesn't have pci_dev yet */
+ dbg("%s: no pci_dev for bridge dn %s\n", __FUNCTION__, slot->name);
+ dealloc_slot_struct(slot);
+ return 1;
+ }
+
+ /* find slot's pci_dev if it's not empty */
+ if (slot->hotplug_slot->info->adapter_status == EMPTY) {
+ slot->state = EMPTY; /* slot is empty */
+ slot->dev.pci_dev = NULL;
+ } else {
+ /* slot is occupied */
+ if (!(slot->dn->child)) {
+ /* non-empty slot has to have child */
+ err("%s: slot[%s]'s device_node doesn't have child for adapter\n", __FUNCTION__, slot->name);
+ dealloc_slot_struct(slot);
+ return 1;
+ }
+ slot->dev.pci_dev = rpaphp_find_adapter_pdev(slot);
+ if (slot->dev.pci_dev) {
+ slot->state = CONFIGURED;
+
+ } else {
+ /* DLPAR add as opposed to
+ * boot time */
+ slot->state = NOT_CONFIGURED;
+ }
+ }
+ return 0;
+}
+
+int register_pci_slot(struct slot *slot)
+{
+ int rc = 1;
+
+ slot->dev_type = PCI_DEV;
+ if (setup_pci_hotplug_slot_info(slot))
+ goto exit_rc;
+ if (setup_pci_slot(slot))
+ goto exit_rc;
+ rc = register_slot(slot);
+ exit_rc:
+ if (rc)
+ dealloc_slot_struct(slot);
+ return rc;
+}
+
+int rpaphp_enable_pci_slot(struct slot *slot)
+{
+ int retval = 0, state;
+
+ retval = rpaphp_get_sensor_state(slot, &state);
+ if (retval)
+ goto exit;
+ dbg("%s: sensor state[%d]\n", __FUNCTION__, state);
+ /* if slot is not empty, enable the adapter */
+ if (state == PRESENT) {
+ dbg("%s : slot[%s] is occupid.\n", __FUNCTION__, slot->name);
+ if ((slot->dev.pci_dev =
+ rpaphp_config_pci_adapter(slot)) != NULL) {
+ slot->state = CONFIGURED;
+ dbg("%s: PCI adapter %s in slot[%s] has been configured\n",
+ __FUNCTION__, pci_name(slot->dev.pci_dev), slot->name);
+ } else {
+ slot->state = NOT_CONFIGURED;
+ dbg("%s: no pci_dev struct for adapter in slot[%s]\n",
+ __FUNCTION__, slot->name);
+ }
+ } else if (state == EMPTY) {
+ dbg("%s : slot[%s] is empty\n", __FUNCTION__, slot->name);
+ slot->state = EMPTY;
+ } else {
+ err("%s: slot[%s] is in invalid state\n", __FUNCTION__,
+ slot->name);
+ slot->state = NOT_VALID;
+ retval = -EINVAL;
+ }
+ exit:
+ if (slot->state != NOT_VALID)
+ rpaphp_set_attention_status(slot, LED_ON);
+ else
+ rpaphp_set_attention_status(slot, LED_ID);
+ dbg("%s - Exit: rc[%d]\n", __FUNCTION__, retval);
+ return retval;
+}
--- diff/drivers/pci/hotplug/shpchp_pci.c 2004-03-11 10:20:26.000000000 +0000
+++ source/drivers/pci/hotplug/shpchp_pci.c 2004-03-16 11:27:48.225856216 +0000
@@ -101,7 +101,7 @@ int shpchp_unconfigure_device(struct pci
*/
int shpchp_set_irq (u8 bus_num, u8 dev_num, u8 int_pin, u8 irq_num)
{
-#if !defined(CONFIG_X86_IO_APIC) && !defined(CONFIG_X86_64)
+#if defined(CONFIG_X86) && !defined(CONFIG_X86_IO_APIC) && !defined(CONFIG_X86_64)
int rc;
u16 temp_word;
struct pci_dev fakedev;
--- diff/drivers/pci/pci.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/pci/pci.c 2004-03-16 11:27:48.225856216 +0000
@@ -205,6 +205,8 @@ pci_set_power_state(struct pci_dev *dev,
int pm;
u16 pmcsr;
+ might_sleep();
+
/* bound the state we're entering */
if (state > 3) state = 3;
--- diff/drivers/pnp/isapnp/Kconfig 2004-02-18 08:54:10.000000000 +0000
+++ source/drivers/pnp/isapnp/Kconfig 2004-03-16 11:27:48.226856064 +0000
@@ -2,8 +2,8 @@
# ISA Plug and Play configuration
#
config ISAPNP
- bool "ISA Plug and Play support (EXPERIMENTAL)"
- depends on PNP && EXPERIMENTAL
+ bool "ISA Plug and Play support"
+ depends on PNP
help
Say Y here if you would like support for ISA Plug and Play devices.
Some information is in <file:Documentation/isapnp.txt>.
--- diff/drivers/pnp/isapnp/core.c 2004-02-18 08:54:10.000000000 +0000
+++ source/drivers/pnp/isapnp/core.c 2004-03-16 11:27:48.226856064 +0000
@@ -99,6 +99,7 @@ MODULE_LICENSE("GPL");
static unsigned char isapnp_checksum_value;
static DECLARE_MUTEX(isapnp_cfg_mutex);
static int isapnp_detected;
+static int isapnp_csn_count;
/* some prototypes */
@@ -371,11 +372,14 @@ static int __init isapnp_isolate(void)
break;
}
__next:
+ if (csn == 255)
+ break;
checksum = 0x6a;
chksum = 0x00;
bit = 0x00;
}
isapnp_wait();
+ isapnp_csn_count = csn;
return csn;
}
@@ -880,7 +884,7 @@ static int __init isapnp_build_device_li
isapnp_wait();
isapnp_key();
- for (csn = 1; csn <= 10; csn++) {
+ for (csn = 1; csn <= isapnp_csn_count; csn++) {
isapnp_wake(csn);
isapnp_peek(header, 9);
checksum = isapnp_checksum(header);
@@ -890,12 +894,6 @@ static int __init isapnp_build_device_li
header[4], header[5], header[6], header[7], header[8]);
printk(KERN_DEBUG "checksum = 0x%x\n", checksum);
#endif
- /* Don't be strict on the checksum, here !
- e.g. 'SCM SwapBox Plug and Play' has header[8]==0 (should be: b7)*/
- if (header[8] == 0)
- ;
- else if (checksum == 0x00 || checksum != header[8]) /* not valid CSN */
- continue;
if ((card = isapnp_alloc(sizeof(struct pnp_card))) == NULL)
continue;
@@ -932,9 +930,8 @@ int isapnp_present(void)
int isapnp_cfg_begin(int csn, int logdev)
{
- if (csn < 1 || csn > 10 || logdev > 10)
+ if (csn < 1 || csn > isapnp_csn_count || logdev > 10)
return -EINVAL;
- MOD_INC_USE_COUNT;
down(&isapnp_cfg_mutex);
isapnp_wait();
isapnp_key();
@@ -962,7 +959,6 @@ int isapnp_cfg_end(void)
{
isapnp_wait();
up(&isapnp_cfg_mutex);
- MOD_DEC_USE_COUNT;
return 0;
}
@@ -1050,7 +1046,7 @@ static int isapnp_set_resources(struct p
for (tmp = 0; tmp < PNP_MAX_DMA && (res->dma_resource[tmp].flags & (IORESOURCE_DMA | IORESOURCE_UNSET)) == IORESOURCE_DMA; tmp++)
isapnp_write_byte(ISAPNP_CFG_DMA+tmp, res->dma_resource[tmp].start);
for (tmp = 0; tmp < PNP_MAX_MEM && (res->mem_resource[tmp].flags & (IORESOURCE_MEM | IORESOURCE_UNSET)) == IORESOURCE_MEM; tmp++)
- isapnp_write_word(ISAPNP_CFG_MEM+(tmp<<2), (res->mem_resource[tmp].start >> 8) & 0xffff);
+ isapnp_write_word(ISAPNP_CFG_MEM+(tmp<<3), (res->mem_resource[tmp].start >> 8) & 0xffff);
/* FIXME: We aren't handling 32bit mems properly here */
isapnp_activate(dev->number);
isapnp_cfg_end();
--- diff/drivers/pnp/resource.c 2004-02-18 08:54:10.000000000 +0000
+++ source/drivers/pnp/resource.c 2004-03-16 11:27:48.227855912 +0000
@@ -231,15 +231,9 @@ void pnp_free_option(struct pnp_option *
#define length(start, end) (*(end) - *(start) + 1)
-/* ranged_conflict - used to determine if two resource ranges conflict
- * condition 1: check if the start of a is within b
- * condition 2: check if the end of a is within b
- * condition 3: check if b is engulfed by a */
-
+/* Two ranges conflict if one doesn't end before the other starts */
#define ranged_conflict(starta, enda, startb, endb) \
-((*(starta) >= *(startb) && *(starta) <= *(endb)) || \
- (*(enda) >= *(startb) && *(enda) <= *(endb)) || \
- (*(starta) < *(startb) && *(enda) > *(endb)))
+ !((*(enda) < *(startb)) || (*(endb) < *(starta)))
#define cannot_compare(flags) \
((flags) & (IORESOURCE_UNSET | IORESOURCE_DISABLED))
--- diff/drivers/pnp/system.c 2003-11-25 15:24:58.000000000 +0000
+++ source/drivers/pnp/system.c 2004-03-16 11:27:48.227855912 +0000
@@ -21,7 +21,7 @@ static const struct pnp_device_id pnp_de
{ "", 0 }
};
-static void __init reserve_ioport_range(char *pnpid, int start, int end)
+static void reserve_ioport_range(char *pnpid, int start, int end)
{
struct resource *res;
char *regionid;
@@ -49,7 +49,7 @@ static void __init reserve_ioport_range(
return;
}
-static void __init reserve_resources_of_dev( struct pnp_dev *dev )
+static void reserve_resources_of_dev( struct pnp_dev *dev )
{
int i;
--- diff/drivers/s390/char/keyboard.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/s390/char/keyboard.c 2004-03-16 11:27:48.228855760 +0000
@@ -471,7 +471,7 @@ kbd_ioctl(struct kbd_data *kbd, struct f
* To have permissions to do most of the vt ioctls, we either have
* to be the owner of the tty, or have CAP_SYS_TTY_CONFIG.
*/
- perm = current->tty == kbd->tty || capable(CAP_SYS_TTY_CONFIG);
+ perm = current->signal->tty == kbd->tty || capable(CAP_SYS_TTY_CONFIG);
switch (cmd) {
case KDGKBTYPE:
return put_user(KB_101, (char*) arg);
--- diff/drivers/scsi/3w-xxxx.c 2003-09-30 14:46:16.000000000 +0000
+++ source/drivers/scsi/3w-xxxx.c 2004-03-16 11:27:48.231855304 +0000
@@ -6,7 +6,7 @@
Arnaldo Carvalho de Melo <acme@conectiva.com.br>
Brad Strand <linux@3ware.com>
- Copyright (C) 1999-2003 3ware Inc.
+ Copyright (C) 1999-2004 3ware Inc.
Kernel compatiblity By: Andre Hedrick <andre@suse.com>
Non-Copyright (C) 2000 Andre Hedrick <andre@suse.com>
@@ -179,6 +179,8 @@
1.02.00.036 - Increase character ioctl timeout to 60 seconds.
1.02.00.037 - Fix tw_ioctl() to handle all non-data ATA passthru cmds
for 'smartmontools' support.
+ 1.26.00.038 - Roll driver minor version to 26 to denote kernel 2.6.
+ Add support for cmds_per_lun module parameter.
*/
#include <linux/module.h>
@@ -205,6 +207,7 @@ MODULE_LICENSE("GPL");
#include <linux/reboot.h>
#include <linux/spinlock.h>
#include <linux/interrupt.h>
+#include <linux/moduleparam.h>
#include <asm/errno.h>
#include <asm/io.h>
@@ -242,10 +245,15 @@ static struct file_operations tw_fops =
};
/* Globals */
-char *tw_driver_version="1.02.00.037";
+char *tw_driver_version="1.26.00.038";
TW_Device_Extension *tw_device_extension_list[TW_MAX_SLOT];
int tw_device_extension_count = 0;
static int twe_major = -1;
+static int cmds_per_lun;
+
+/* Module parameters */
+module_param(cmds_per_lun, int, 0);
+MODULE_PARM_DESC(cmds_per_lun, "Maximum commands per LUN");
/* Functions */
@@ -1141,14 +1149,6 @@ int tw_findcards(Scsi_Host_Template *tw_
/* Set card status as online */
tw_dev->online = 1;
-#ifdef CONFIG_3W_XXXX_CMD_PER_LUN
- tw_host->cmd_per_lun = CONFIG_3W_XXXX_CMD_PER_LUN;
- if (tw_host->cmd_per_lun > TW_MAX_CMDS_PER_LUN)
- tw_host->cmd_per_lun = TW_MAX_CMDS_PER_LUN;
-#else
- /* Use SHT cmd_per_lun here */
- tw_host->cmd_per_lun = TW_MAX_CMDS_PER_LUN;
-#endif
tw_dev->free_head = TW_Q_START;
tw_dev->free_tail = TW_Q_START;
tw_dev->free_wrap = TW_Q_LENGTH - 1;
@@ -3386,13 +3386,13 @@ int tw_slave_configure(Scsi_Device *SDpt
dprintk(KERN_WARNING "3w-xxxx: tw_slave_configure()\n");
-#ifdef CONFIG_3W_XXXX_CMD_PER_LUN
- max_cmds = CONFIG_3W_XXXX_CMD_PER_LUN;
- if (max_cmds > TW_MAX_CMDS_PER_LUN)
+ if (cmds_per_lun) {
+ max_cmds = cmds_per_lun;
+ if (max_cmds > TW_MAX_CMDS_PER_LUN)
+ max_cmds = TW_MAX_CMDS_PER_LUN;
+ } else {
max_cmds = TW_MAX_CMDS_PER_LUN;
-#else
- max_cmds = TW_MAX_CMDS_PER_LUN;
-#endif
+ }
scsi_adjust_queue_depth(SDptr, MSG_ORDERED_TAG, max_cmds);
return 0;
@@ -3488,6 +3488,7 @@ static Scsi_Host_Template driver_templat
.eh_abort_handler = tw_scsi_eh_abort,
.eh_host_reset_handler = tw_scsi_eh_reset,
.bios_param = tw_scsi_biosparam,
+ .slave_configure = tw_slave_configure,
.can_queue = TW_Q_LENGTH-2,
.this_id = -1,
.sg_tablesize = TW_MAX_SGL_LENGTH,
--- diff/drivers/scsi/3w-xxxx.h 2003-09-30 14:46:16.000000000 +0000
+++ source/drivers/scsi/3w-xxxx.h 2004-03-16 11:27:48.231855304 +0000
@@ -6,7 +6,7 @@
Arnaldo Carvalho de Melo <acme@conectiva.com.br>
Brad Strand <linux@3ware.com>
- Copyright (C) 1999-2003 3ware Inc.
+ Copyright (C) 1999-2004 3ware Inc.
Kernel compatiblity By: Andre Hedrick <andre@suse.com>
Non-Copyright (C) 2000 Andre Hedrick <andre@suse.com>
--- diff/drivers/scsi/aic7xxx/aic7xxx_osm.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/scsi/aic7xxx/aic7xxx_osm.c 2004-03-16 11:27:48.235854696 +0000
@@ -3969,11 +3969,10 @@ ahc_linux_alloc_device(struct ahc_softc
}
static void
-ahc_linux_free_device(struct ahc_softc *ahc, struct ahc_linux_device *dev)
+__ahc_linux_free_device(struct ahc_softc *ahc, struct ahc_linux_device *dev)
{
struct ahc_linux_target *targ;
- del_timer_sync(&dev->timer);
targ = dev->target;
targ->devices[dev->lun] = NULL;
free(dev, M_DEVBUF);
@@ -3983,6 +3982,13 @@ ahc_linux_free_device(struct ahc_softc *
ahc_linux_free_target(ahc, targ);
}
+static void
+ahc_linux_free_device(struct ahc_softc *ahc, struct ahc_linux_device *dev)
+{
+ del_timer_sync(&dev->timer);
+ __ahc_linux_free_device(ahc, dev);
+}
+
void
ahc_send_async(struct ahc_softc *ahc, char channel,
u_int target, u_int lun, ac_code code, void *arg)
@@ -4693,7 +4699,7 @@ ahc_linux_dev_timed_unfreeze(u_long arg)
ahc_linux_run_device_queue(ahc, dev);
if (TAILQ_EMPTY(&dev->busyq)
&& dev->active == 0)
- ahc_linux_free_device(ahc, dev);
+ __ahc_linux_free_device(ahc, dev);
ahc_unlock(ahc, &s);
}
--- diff/drivers/scsi/ide-scsi.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/scsi/ide-scsi.c 2004-03-16 11:27:48.237854392 +0000
@@ -78,6 +78,7 @@ typedef struct idescsi_pc_s {
#define PC_DMA_IN_PROGRESS 0 /* 1 while DMA in progress */
#define PC_WRITING 1 /* Data direction */
#define PC_TRANSFORM 2 /* transform SCSI commands */
+#define PC_TIMEDOUT 3 /* command timed out */
#define PC_DMA_OK 4 /* Use DMA */
/*
@@ -307,6 +308,41 @@ static int idescsi_check_condition(ide_d
return ide_do_drive_cmd(drive, rq, ide_preempt);
}
+ide_startstop_t idescsi_atapi_error (ide_drive_t *drive, const char *msg, byte stat)
+{
+ struct request *rq;
+ byte err;
+
+ err = ide_dump_atapi_status(drive, msg, stat);
+
+ if (drive == NULL || (rq = HWGROUP(drive)->rq) == NULL)
+ return ide_stopped;
+
+ if (HWIF(drive)->INB(IDE_STATUS_REG) & (BUSY_STAT|DRQ_STAT))
+ /* force an abort */
+ HWIF(drive)->OUTB(WIN_IDLEIMMEDIATE,IDE_COMMAND_REG);
+
+ rq->errors++;
+ DRIVER(drive)->end_request(drive, 0, 0);
+ return ide_stopped;
+}
+
+ide_startstop_t idescsi_atapi_abort (ide_drive_t *drive, const char *msg)
+{
+ struct request *rq;
+
+ if (drive == NULL || (rq = HWGROUP(drive)->rq) == NULL)
+ return ide_stopped;
+
+#if IDESCSI_DEBUG_LOG
+ printk(KERN_WARNING "idescsi_atapi_abort called for %lu\n",
+ ((idescsi_pc_t *) rq->special)->scsi_cmd->serial_number);
+#endif
+ rq->errors |= ERROR_MAX;
+ DRIVER(drive)->end_request(drive, 0, 0);
+ return ide_stopped;
+}
+
static int idescsi_end_request (ide_drive_t *drive, int uptodate, int nrsecs)
{
idescsi_scsi_t *scsi = drive_to_idescsi(drive);
@@ -334,7 +370,13 @@ static int idescsi_end_request (ide_driv
kfree(rq);
pc = opc;
rq = pc->rq;
- pc->scsi_cmd->result = (CHECK_CONDITION << 1) | (DID_OK << 16);
+ pc->scsi_cmd->result = (CHECK_CONDITION << 1) |
+ ((test_bit(PC_TIMEDOUT, &pc->flags)?DID_TIME_OUT:DID_OK) << 16);
+ } else if (test_bit(PC_TIMEDOUT, &pc->flags)) {
+ if (log)
+ printk (KERN_WARNING "ide-scsi: %s: timed out for %lu\n",
+ drive->name, pc->scsi_cmd->serial_number);
+ pc->scsi_cmd->result = DID_TIME_OUT << 16;
} else if (rq->errors >= ERROR_MAX) {
pc->scsi_cmd->result = DID_ERROR << 16;
if (log)
@@ -374,6 +416,19 @@ static inline unsigned long get_timeout(
return IDE_MAX(WAIT_CMD, pc->timeout - jiffies);
}
+static int idescsi_expiry(ide_drive_t *drive)
+{
+ idescsi_scsi_t *scsi = drive->driver_data;
+ idescsi_pc_t *pc = scsi->pc;
+
+#if IDESCSI_DEBUG_LOG
+ printk(KERN_WARNING "idescsi_expiry called for %lu at %lu\n", pc->scsi_cmd->serial_number, jiffies);
+#endif
+ set_bit(PC_TIMEDOUT, &pc->flags);
+
+ return 0; /* we do not want the ide subsystem to retry */
+}
+
/*
* Our interrupt handler.
*/
@@ -393,6 +448,15 @@ static ide_startstop_t idescsi_pc_intr (
printk (KERN_INFO "ide-scsi: Reached idescsi_pc_intr interrupt handler\n");
#endif /* IDESCSI_DEBUG_LOG */
+ if (test_bit(PC_TIMEDOUT, &pc->flags)){
+#if IDESCSI_DEBUG_LOG
+ printk(KERN_WARNING "idescsi_pc_intr: got timed out packet %lu at %lu\n",
+ pc->scsi_cmd->serial_number, jiffies);
+#endif
+ /* end this request now - scsi should retry it*/
+ idescsi_end_request (drive, 1, 0);
+ return ide_stopped;
+ }
if (test_and_clear_bit (PC_DMA_IN_PROGRESS, &pc->flags)) {
#if IDESCSI_DEBUG_LOG
printk ("ide-scsi: %s: DMA complete\n", drive->name);
@@ -442,7 +506,7 @@ static ide_startstop_t idescsi_pc_intr (
pc->actually_transferred += temp;
pc->current_position += temp;
idescsi_discard_data(drive, bcount.all - temp);
- ide_set_handler(drive, &idescsi_pc_intr, get_timeout(pc), NULL);
+ ide_set_handler(drive, &idescsi_pc_intr, get_timeout(pc), idescsi_expiry);
return ide_started;
}
#if IDESCSI_DEBUG_LOG
@@ -467,7 +531,8 @@ static ide_startstop_t idescsi_pc_intr (
pc->actually_transferred += bcount.all;
pc->current_position += bcount.all;
- ide_set_handler(drive, &idescsi_pc_intr, get_timeout(pc), NULL); /* And set the interrupt handler again */
+ /* And set the interrupt handler again */
+ ide_set_handler(drive, &idescsi_pc_intr, get_timeout(pc), idescsi_expiry);
return ide_started;
}
@@ -492,7 +557,7 @@ static ide_startstop_t idescsi_transfer_
if (HWGROUP(drive)->handler != NULL)
BUG();
/* Set the interrupt routine */
- ide_set_handler(drive, &idescsi_pc_intr, get_timeout(pc), NULL);
+ ide_set_handler(drive, &idescsi_pc_intr, get_timeout(pc), idescsi_expiry);
/* Send the actual packet */
atapi_output_bytes(drive, scsi->pc->c, 12);
if (test_bit (PC_DMA_OK, &pc->flags)) {
@@ -540,7 +605,7 @@ static ide_startstop_t idescsi_issue_pc
if (HWGROUP(drive)->handler != NULL)
BUG();
ide_set_handler(drive, &idescsi_transfer_pc,
- get_timeout(pc), NULL);
+ get_timeout(pc), idescsi_expiry);
/* Issue the packet command */
HWIF(drive)->OUTB(WIN_PACKETCMD, IDE_COMMAND_REG);
return ide_started;
@@ -633,6 +698,8 @@ static ide_driver_t idescsi_driver = {
.cleanup = idescsi_cleanup,
.do_request = idescsi_do_request,
.end_request = idescsi_end_request,
+ .error = idescsi_atapi_error,
+ .abort = idescsi_atapi_abort,
.drives = LIST_HEAD_INIT(idescsi_driver.drives),
};
@@ -852,66 +919,132 @@ abort:
return 1;
}
-static int idescsi_abort (Scsi_Cmnd *cmd)
+static int idescsi_eh_abort (Scsi_Cmnd *cmd)
{
- int countdown = 8;
- unsigned long flags;
- idescsi_scsi_t *scsi = scsihost_to_idescsi(cmd->device->host);
- ide_drive_t *drive = scsi->drive;
+ idescsi_scsi_t *scsi = scsihost_to_idescsi(cmd->device->host);
+ ide_drive_t *drive = scsi->drive;
+ int busy;
+ int ret = FAILED;
- printk (KERN_ERR "ide-scsi: abort called for %lu\n", cmd->serial_number);
- while (countdown--) {
- /* is cmd active?
- * need to lock so this stuff doesn't change under us */
- spin_lock_irqsave(&ide_lock, flags);
- if (scsi->pc && scsi->pc->scsi_cmd &&
- scsi->pc->scsi_cmd->serial_number == cmd->serial_number) {
- /* yep - let's give it some more time -
- * we can do that, we're in _our_ error kernel thread */
- spin_unlock_irqrestore(&ide_lock, flags);
- scsi_sleep(HZ);
- continue;
- }
- /* no, but is it queued in the ide subsystem? */
- if (elv_queue_empty(drive->queue)) {
- spin_unlock_irqrestore(&ide_lock, flags);
- return SUCCESS;
- }
- spin_unlock_irqrestore(&ide_lock, flags);
- schedule_timeout(HZ/10);
+ /* In idescsi_eh_abort we try to gently pry our command from the ide subsystem */
+
+ if (test_bit(IDESCSI_LOG_CMD, &scsi->log))
+ printk (KERN_WARNING "ide-scsi: abort called for %lu\n", cmd->serial_number);
+
+ if (!drive) {
+ printk (KERN_WARNING "ide-scsi: Drive not set in idescsi_eh_abort\n");
+ WARN_ON(1);
+ goto no_drive;
+ }
+
+ /* First give it some more time, how much is "right" is hard to say :-( */
+
+ busy = ide_wait_not_busy(HWIF(drive), 100); /* FIXME - uses mdelay which causes latency? */
+ if (test_bit(IDESCSI_LOG_CMD, &scsi->log))
+ printk (KERN_WARNING "ide-scsi: drive did%s become ready\n", busy?" not":"");
+
+ spin_lock_irq(&ide_lock);
+
+ /* If there is no pc running we're done (our interrupt took care of it) */
+ if (!scsi->pc) {
+ ret = SUCCESS;
+ goto ide_unlock;
+ }
+
+ /* It's somewhere in flight. Does ide subsystem agree? */
+ if (scsi->pc->scsi_cmd->serial_number == cmd->serial_number && !busy &&
+ elv_queue_empty(drive->queue) && HWGROUP(drive)->rq != scsi->pc->rq) {
+ /*
+ * FIXME - not sure this condition can ever occur
+ */
+ printk (KERN_ERR "ide-scsi: cmd aborted!\n");
+
+ idescsi_free_bio(scsi->pc->rq->bio);
+ if (scsi->pc->rq->flags & REQ_SENSE)
+ kfree(scsi->pc->buffer);
+ kfree(scsi->pc->rq);
+ kfree(scsi->pc);
+ scsi->pc = NULL;
+
+ ret = SUCCESS;
}
- return FAILED;
+
+ide_unlock:
+ spin_unlock_irq(&ide_lock);
+no_drive:
+ if (test_bit(IDESCSI_LOG_CMD, &scsi->log))
+ printk (KERN_WARNING "ide-scsi: abort returns %s\n", ret == SUCCESS?"success":"failed");
+
+ return ret;
}
-static int idescsi_reset (Scsi_Cmnd *cmd)
+static int idescsi_eh_reset (Scsi_Cmnd *cmd)
{
- unsigned long flags;
struct request *req;
- idescsi_scsi_t *idescsi = scsihost_to_idescsi(cmd->device->host);
- ide_drive_t *drive = idescsi->drive;
+ idescsi_scsi_t *scsi = scsihost_to_idescsi(cmd->device->host);
+ ide_drive_t *drive = scsi->drive;
+ int ready = 0;
+ int ret = SUCCESS;
+
+ /* In idescsi_eh_reset we forcefully remove the command from the ide subsystem and reset the device. */
+
+ if (test_bit(IDESCSI_LOG_CMD, &scsi->log))
+ printk (KERN_WARNING "ide-scsi: reset called for %lu\n", cmd->serial_number);
+
+ if (!drive) {
+ printk (KERN_WARNING "ide-scsi: Drive not set in idescsi_eh_reset\n");
+ WARN_ON(1);
+ return FAILED;
+ }
+
+ spin_lock_irq(&ide_lock);
+
+ if (!scsi->pc || (req = scsi->pc->rq) != HWGROUP(drive)->rq || !HWGROUP(drive)->handler) {
+ printk (KERN_WARNING "ide-scsi: No active request in idescsi_eh_reset\n");
+ spin_unlock(&ide_lock);
+ return FAILED;
+ }
+
+ /* kill current request */
+ blkdev_dequeue_request(req);
+ end_that_request_last(req);
+ idescsi_free_bio(req->bio);
+ if (req->flags & REQ_SENSE)
+ kfree(scsi->pc->buffer);
+ kfree(scsi->pc);
+ scsi->pc = NULL;
+ kfree(req);
- printk (KERN_ERR "ide-scsi: reset called for %lu\n", cmd->serial_number);
- /* first null the handler for the drive and let any process
- * doing IO (on another CPU) run to (partial) completion
- * the lock prevents processing new requests */
- spin_lock_irqsave(&ide_lock, flags);
- while (HWGROUP(drive)->handler) {
- HWGROUP(drive)->handler = NULL;
- schedule_timeout(1);
- }
/* now nuke the drive queue */
while ((req = elv_next_request(drive->queue))) {
blkdev_dequeue_request(req);
end_that_request_last(req);
}
- /* FIXME - this will probably leak memory */
+
HWGROUP(drive)->rq = NULL;
- if (drive_to_idescsi(drive))
- drive_to_idescsi(drive)->pc = NULL;
- spin_unlock_irqrestore(&ide_lock, flags);
- /* finally, reset the drive (and its partner on the bus...) */
- ide_do_reset (drive);
- return SUCCESS;
+ HWGROUP(drive)->handler = NULL;
+ HWGROUP(drive)->busy = 1; /* will set this to zero when ide reset finished */
+ spin_unlock_irq(&ide_lock);
+
+ ide_do_reset(drive);
+
+ /* ide_do_reset starts a polling handler which restarts itself every 50ms until the reset finishes */
+
+ do {
+ set_current_state(TASK_UNINTERRUPTIBLE);
+ spin_unlock_irq(cmd->device->host->host_lock);
+ schedule_timeout(HZ/20);
+ spin_lock_irq(cmd->device->host->host_lock);
+ } while ( HWGROUP(drive)->handler );
+
+ ready = drive_is_ready(drive);
+ HWGROUP(drive)->busy--;
+ if (!ready) {
+ printk (KERN_ERR "ide-scsi: reset failed!\n");
+ ret = FAILED;
+ }
+
+ return ret;
}
static int idescsi_bios(struct scsi_device *sdev, struct block_device *bdev,
@@ -935,8 +1068,8 @@ static Scsi_Host_Template idescsi_templa
.slave_configure = idescsi_slave_configure,
.ioctl = idescsi_ioctl,
.queuecommand = idescsi_queue,
- .eh_abort_handler = idescsi_abort,
- .eh_device_reset_handler = idescsi_reset,
+ .eh_abort_handler = idescsi_eh_abort,
+ .eh_host_reset_handler = idescsi_eh_reset,
.bios_param = idescsi_bios,
.can_queue = 40,
.this_id = -1,
--- diff/drivers/scsi/scsi_transport_spi.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/scsi/scsi_transport_spi.c 2004-03-16 11:27:48.238854240 +0000
@@ -33,7 +33,7 @@
#include <scsi/scsi_transport.h>
#include <scsi/scsi_transport_spi.h>
-#define SPI_PRINTK(x, l, f, a...) printk(l "scsi(%d:%d:%d:%d): " f, (x)->host->host_no, (x)->channel, (x)->id, (x)->lun, ##a)
+#define SPI_PRINTK(x, l, f, a...) printk(l "scsi(%d:%d:%d:%d): " f, (x)->host->host_no, (x)->channel, (x)->id, (x)->lun , ##a)
static void transport_class_release(struct class_device *class_dev);
--- diff/drivers/scsi/st.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/scsi/st.c 2004-03-16 11:27:48.241853784 +0000
@@ -3896,6 +3896,11 @@ static int st_probe(struct device *dev)
dev_num);
goto out_free_tape;
}
+ /* Make sure that the minor numbers corresponding to the four
+ first modes always get the same names */
+ i = mode << (4 - ST_NBR_MODE_BITS);
+ snprintf(cdev->kobj.name, KOBJ_NAME_LEN, "%s%s%s", j ? "n" : "",
+ disk->disk_name, st_formats[i]);
cdev->owner = THIS_MODULE;
cdev->ops = &st_fops;
--- diff/drivers/serial/21285.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/serial/21285.c 2004-03-16 11:27:48.242853632 +0000
@@ -15,6 +15,7 @@
#include <linux/console.h>
#include <linux/serial_core.h>
#include <linux/serial.h>
+#include <linux/device.h>
#include <asm/io.h>
#include <asm/irq.h>
@@ -543,3 +544,4 @@ module_exit(serial21285_exit);
MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("Intel Footbridge (21285) serial driver $Revision: 1.37 $");
+MODULE_ALIAS_CHARDEV(SERIAL_21285_MAJOR, SERIAL_21285_MINOR);
--- diff/drivers/serial/8250.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/serial/8250.c 2004-03-16 11:27:48.244853328 +0000
@@ -834,7 +834,7 @@ receive_chars(struct uart_8250_port *up,
if (unlikely(tty->flip.count >= TTY_FLIPBUF_SIZE)) {
tty->flip.work.func((void *)tty);
if (tty->flip.count >= TTY_FLIPBUF_SIZE)
- return; // if TTY_DONT_FLIP is set
+ return; /* if TTY_DONT_FLIP is set */
}
ch = serial_inp(up, UART_RX);
*tty->flip.char_buf_ptr = ch;
@@ -1195,12 +1195,21 @@ static void serial8250_break_ctl(struct
spin_unlock_irqrestore(&up->port.lock, flags);
}
+#ifdef CONFIG_KGDB
+static int kgdb_irq = -1;
+#endif
+
static int serial8250_startup(struct uart_port *port)
{
struct uart_8250_port *up = (struct uart_8250_port *)port;
unsigned long flags;
int retval;
+#ifdef CONFIG_KGDB
+ if (up->port.irq == kgdb_irq)
+ return -EBUSY;
+#endif
+
up->capabilities = uart_config[up->port.type].flags;
if (up->port.type == PORT_16C950) {
@@ -1866,6 +1875,10 @@ static void __init serial8250_register_p
for (i = 0; i < UART_NR; i++) {
struct uart_8250_port *up = &serial8250_ports[i];
+#ifdef CONFIG_KGDB
+ if (up->port.irq == kgdb_irq)
+ up->port.kgdb = 1;
+#endif
up->port.line = i;
up->port.ops = &serial8250_pops;
init_timer(&up->timer);
@@ -2145,6 +2158,31 @@ void serial8250_resume_port(int line)
uart_resume_port(&serial8250_reg, &serial8250_ports[line].port);
}
+#ifdef CONFIG_KGDB
+/*
+ * Find all the ports using the given irq and shut them down.
+ * Result should be that the irq will be released.
+ */
+void shutdown_for_kgdb(struct async_struct * info)
+{
+ int irq = info->state->irq;
+ struct uart_8250_port *up;
+ int ttyS;
+
+ kgdb_irq = irq; /* save for later init */
+ for (ttyS = 0; ttyS < UART_NR; ttyS++){
+ up = &serial8250_ports[ttyS];
+ if (up->port.irq == irq && (irq_lists + irq)->head) {
+#ifdef CONFIG_DEBUG_SPINLOCK /* ugly business... */
+ if(up->port.lock.magic != SPINLOCK_MAGIC)
+ spin_lock_init(&up->port.lock);
+#endif
+ serial8250_shutdown(&up->port);
+ }
+ }
+}
+#endif /* CONFIG_KGDB */
+
static int __init serial8250_init(void)
{
int ret, i;
--- diff/drivers/serial/8250_pnp.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/serial/8250_pnp.c 2004-03-16 11:27:48.245853176 +0000
@@ -266,6 +266,8 @@ static const struct pnp_device_id pnp_de
{ "RSS00A0", 0 },
/* Viking 56K FAX INT */
{ "RSS0262", 0 },
+ /* K56 par,VV,Voice,Speakphone,AudioSpan,PnP */
+ { "RSS0250", 0 },
/* SupraExpress 28.8 Data/Fax PnP modem */
{ "SUP1310", 0 },
/* SupraExpress 33.6 Data/Fax PnP modem */
@@ -283,6 +285,8 @@ static const struct pnp_device_id pnp_de
/* 3Com Corp. */
/* Gateway Telepath IIvi 33.6 */
{ "USR0000", 0 },
+ /* U.S. Robotics Sporster 33.6K Fax INT PnP */
+ { "USR0002", 0 },
/* Sportster Vi 14.4 PnP FAX Voicemail */
{ "USR0004", 0 },
/* U.S. Robotics 33.6K Voice INT PnP */
@@ -315,6 +319,8 @@ static const struct pnp_device_id pnp_de
{ "USR9180", 0 },
/* U.S. Robotics 56K Voice INT PnP*/
{ "USR9190", 0 },
+ /* Rockwell's (PORALiNK) 33600 INT PNP */
+ { "WCI0003", 0 },
/* Unkown PnP modems */
{ "PNPCXXX", UNKNOWN_DEV },
/* More unkown PnP modems */
--- diff/drivers/serial/amba.c 2003-06-09 13:18:19.000000000 +0000
+++ source/drivers/serial/amba.c 2004-03-16 11:27:48.245853176 +0000
@@ -39,6 +39,7 @@
#include <linux/serial.h>
#include <linux/console.h>
#include <linux/sysrq.h>
+#include <linux/device.h>
#include <asm/io.h>
#include <asm/irq.h>
@@ -758,3 +759,4 @@ module_exit(ambauart_exit);
MODULE_AUTHOR("ARM Ltd/Deep Blue Solutions Ltd");
MODULE_DESCRIPTION("ARM AMBA serial port driver $Revision: 1.41 $");
MODULE_LICENSE("GPL");
+MODULE_ALIAS_CHARDEV(SERIAL_AMBA_MAJOR, SERIAL_AMBA_MINOR);
--- diff/drivers/serial/anakin.c 2003-05-21 10:50:16.000000000 +0000
+++ source/drivers/serial/anakin.c 2004-03-16 11:27:48.246853024 +0000
@@ -30,6 +30,7 @@
#include <linux/serial.h>
#include <linux/console.h>
#include <linux/sysrq.h>
+#include <linux/device.h>
#include <asm/io.h>
#include <asm/irq.h>
@@ -541,3 +542,4 @@ MODULE_DESCRIPTION("Anakin serial driver
MODULE_AUTHOR("Tak-Shing Chan <chan@aleph1.co.uk>");
MODULE_SUPPORTED_DEVICE("ttyAN");
MODULE_LICENSE("GPL");
+MODULE_ALIAS_CHARDEV(SERIAL_ANAKIN_MAJOR, SERIAL_ANAKIN_MINOR);
--- diff/drivers/serial/clps711x.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/serial/clps711x.c 2004-03-16 11:27:48.246853024 +0000
@@ -34,6 +34,7 @@
#include <linux/console.h>
#include <linux/sysrq.h>
#include <linux/spinlock.h>
+#include <linux/device.h>
#include <asm/hardware.h>
#include <asm/io.h>
@@ -611,3 +612,4 @@ module_exit(clps711xuart_exit);
MODULE_AUTHOR("Deep Blue Solutions Ltd");
MODULE_DESCRIPTION("CLPS-711x generic serial driver $Revision: 1.42 $");
MODULE_LICENSE("GPL");
+MODULE_ALIAS_CHARDEV(SERIAL_CLPS711X_MAJOR, SERIAL_CLPS711X_MINOR);
--- diff/drivers/serial/mux.c 2004-02-18 08:54:11.000000000 +0000
+++ source/drivers/serial/mux.c 2004-03-16 11:27:48.247852872 +0000
@@ -25,6 +25,7 @@
#include <linux/console.h>
#include <linux/slab.h>
#include <linux/delay.h> /* for udelay */
+#include <linux/device.h>
#include <asm/io.h>
#include <asm/parisc-device.h>
@@ -535,3 +536,4 @@ module_exit(mux_exit);
MODULE_AUTHOR("Ryan Bradetich");
MODULE_DESCRIPTION("Serial MUX driver");
MODULE_LICENSE("GPL");
+MODULE_ALIAS_CHARDEV_MAJOR(MUX_MAJOR);
--- diff/drivers/serial/sa1100.c 2004-01-19 10:22:58.000000000 +0000
+++ source/drivers/serial/sa1100.c 2004-03-16 11:27:48.248852720 +0000
@@ -953,3 +953,4 @@ module_exit(sa1100_serial_exit);
MODULE_AUTHOR("Deep Blue Solutions Ltd");
MODULE_DESCRIPTION("SA1100 generic serial port driver $Revision: 1.50 $");
MODULE_LICENSE("GPL");
+MODULE_ALIAS_CHARDEV_MAJOR(SERIAL_SA1100_MAJOR);
--- diff/drivers/serial/serial_core.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/serial/serial_core.c 2004-03-16 11:27:48.249852568 +0000
@@ -1985,6 +1985,11 @@ uart_configure_port(struct uart_driver *
{
unsigned int flags;
+#ifdef CONFIG_KGDB
+ if (port->kgdb)
+ return;
+#endif
+
/*
* If there isn't a port here, don't do anything further.
*/
--- diff/drivers/usb/Kconfig 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/Kconfig 2004-03-16 11:27:48.355836456 +0000
@@ -6,32 +6,36 @@ menu "USB support"
# ARM SA1111 chips have a non-PCI based "OHCI-compatible" USB host interface.
config USB
- tristate "Support for USB"
+ tristate "Support for Host-side USB"
depends on PCI || SA1111 || ARCH_OMAP1510 || ARCH_OMAP1610
---help---
Universal Serial Bus (USB) is a specification for a serial bus
subsystem which offers higher speeds and more features than the
traditional PC serial port. The bus supplies power to peripherals
and allows for hot swapping. Up to 127 USB peripherals can be
- connected to a single USB port in a tree structure. The USB port is
- the root of the tree, the peripherals are the leaves and the inner
- nodes are special USB devices called hubs. Many newer PC's have USB
- ports and newer peripherals such as scanners, keyboards, mice,
- modems, and printers support the USB protocol and can be connected
- to the PC via those ports.
-
- Say Y here if your computer has a USB port and you want to use USB
- devices. You then need to say Y to at least one of "UHCI HCD support"
- or "OHCI HCD support" below (the type of interface that the USB hardware
- in your computer provides to the operating system) and then choose
- from amongst the drivers for USB peripherals. You may want to check
- out the information provided in <file:Documentation/usb/> and
- especially the links given in <file:Documentation/usb/usb-help.txt>.
-
- If you have a new USB 2.0 High Speed system, you should also choose
- "EHCI HCD (USB 2.0) support" as well as at least one of UHCI or OHCI.
-
- It doesn't normally hurt to select them all if you are not certain.
+ connected to a single USB host in a tree structure.
+
+ The USB host is the root of the tree, the peripherals are the
+ leaves and the inner nodes are special USB devices called hubs.
+ Most PCs now have USB host ports, used to connect peripherals
+ such as scanners, keyboards, mice, modems, cameras, disks,
+ flash memory, network links, and printers to the PC.
+
+ Say Y here if your computer has a host-side USB port and you want
+ to use USB devices. You then need to say Y to at least one of the
+ Host Controller Driver (HCD) options below. Choose a USB 1.1
+ controller, such as "UHCI HCD support" or "OHCI HCD support",
+ and "EHCI HCD (USB 2.0) support" except for older systems that
+ do not have USB 2.0 support. It doesn't normally hurt to select
+ them all if you are not certain.
+
+ If your system has a device-side USB port, used in the peripheral
+ side of the USB protocol, see the "USB Gadget" framework instead.
+
+ After choosing your HCD, then select drivers for the USB peripherals
+ you'll be using. You may want to check out the information provided
+ in <file:Documentation/usb/> and especially the links given in
+ <file:Documentation/usb/usb-help.txt>.
To compile this driver as a module, choose M here: the
module will be called usbcore.
--- diff/drivers/usb/Makefile 2004-02-18 08:54:11.000000000 +0000
+++ source/drivers/usb/Makefile 2004-03-16 11:27:48.356836304 +0000
@@ -20,10 +20,15 @@ obj-$(CONFIG_USB_PRINTER) += class/
obj-$(CONFIG_USB_STORAGE) += storage/
obj-$(CONFIG_USB_AIPTEK) += input/
+obj-$(CONFIG_USB_ATI_REMOTE) += input/
obj-$(CONFIG_USB_HID) += input/
obj-$(CONFIG_USB_KBD) += input/
+obj-$(CONFIG_USB_KBTAB) += input/
obj-$(CONFIG_USB_MOUSE) += input/
+obj-$(CONFIG_USB_MTOUCH) += input/
+obj-$(CONFIG_USB_POWERMATE) += input/
obj-$(CONFIG_USB_WACOM) += input/
+obj-$(CONFIG_USB_XPAD) += input/
obj-$(CONFIG_USB_DABUSB) += media/
obj-$(CONFIG_USB_DSBR) += media/
@@ -50,7 +55,6 @@ obj-$(CONFIG_USB_MICROTEK) += image/
obj-$(CONFIG_USB_SERIAL) += serial/
obj-$(CONFIG_USB_AUERSWALD) += misc/
-obj-$(CONFIG_USB_BRLVGER) += misc/
obj-$(CONFIG_USB_EMI26) += misc/
obj-$(CONFIG_USB_LCD) += misc/
obj-$(CONFIG_USB_LEGOTOWER) += misc/
--- diff/drivers/usb/class/audio.c 2004-02-18 08:54:11.000000000 +0000
+++ source/drivers/usb/class/audio.c 2004-03-16 11:27:48.253851960 +0000
@@ -3,7 +3,7 @@
/*
* audio.c -- USB Audio Class driver
*
- * Copyright (C) 1999, 2000, 2001
+ * Copyright (C) 1999, 2000, 2001, 2003, 2004
* Alan Cox (alan@lxorguk.ukuu.org.uk)
* Thomas Sailer (sailer@ife.ee.ethz.ch)
*
@@ -101,6 +101,11 @@
* Fix SNDCTL_DSP_STEREO API violation
* 2003-04-08: Oliver Neukum (oliver@neukum.name):
* Setting a configuration is done by usbcore and must not be overridden
+ * 2004-02-27: Workaround for broken synch descriptors
+ * 2004-03-07: Alan Stern <stern@rowland.harvard.edu>
+ * Add usb_ifnum_to_if() and usb_altnum_to_altsetting() support.
+ * Use the in-memory descriptors instead of reading them from the device.
+ *
*/
/*
@@ -141,8 +146,8 @@
*
* How does the parsing work? First, all interfaces are searched
* for an AudioControl class interface. If found, the config descriptor
- * that belongs to the current configuration is fetched from the device.
- * Then the HEADER descriptor is fetched. It contains a list of
+ * that belongs to the current configuration is searched and
+ * the HEADER descriptor is found. It contains a list of
* all AudioStreaming and MIDIStreaming devices. This list is then walked,
* and all AudioStreaming interfaces are classified into input and output
* interfaces (according to the endpoint0 direction in altsetting1) (MIDIStreaming
@@ -1512,7 +1517,6 @@ static int find_format(struct audioforma
static int set_format_in(struct usb_audiodev *as)
{
struct usb_device *dev = as->state->usbdev;
- struct usb_host_config *config = dev->actconfig;
struct usb_host_interface *alts;
struct usb_interface *iface;
struct usbin *u = &as->usbin;
@@ -1522,9 +1526,9 @@ static int set_format_in(struct usb_audi
unsigned char data[3];
int fmtnr, ret;
- if (u->interface < 0 || u->interface >= config->desc.bNumInterfaces)
+ iface = usb_ifnum_to_if(dev, u->interface);
+ if (!iface)
return 0;
- iface = config->interface[u->interface];
fmtnr = find_format(as->fmtin, as->numfmtin, d->format, d->srate);
if (fmtnr < 0) {
@@ -1533,7 +1537,7 @@ static int set_format_in(struct usb_audi
}
fmt = as->fmtin + fmtnr;
- alts = &iface->altsetting[fmt->altsetting];
+ alts = usb_altnum_to_altsetting(iface, fmt->altsetting);
u->format = fmt->format;
u->datapipe = usb_rcvisocpipe(dev, alts->endpoint[0].desc.bEndpointAddress & 0xf);
u->syncpipe = u->syncinterval = 0;
@@ -1542,18 +1546,20 @@ static int set_format_in(struct usb_audi
alts->endpoint[1].desc.bmAttributes != 0x01 ||
alts->endpoint[1].desc.bSynchAddress != 0 ||
alts->endpoint[1].desc.bEndpointAddress != (alts->endpoint[0].desc.bSynchAddress & 0x7f)) {
- printk(KERN_ERR "usbaudio: device %d interface %d altsetting %d invalid synch pipe\n",
+ printk(KERN_WARNING "usbaudio: device %d interface %d altsetting %d claims adaptive in "
+ "but has invalid synch pipe; treating as asynchronous in\n",
dev->devnum, u->interface, fmt->altsetting);
- return -1;
+ } else {
+ u->syncpipe = usb_sndisocpipe(dev, alts->endpoint[1].desc.bEndpointAddress & 0xf);
+ u->syncinterval = alts->endpoint[1].desc.bRefresh;
}
- u->syncpipe = usb_sndisocpipe(dev, alts->endpoint[1].desc.bEndpointAddress & 0xf);
- u->syncinterval = alts->endpoint[1].desc.bRefresh;
}
if (d->srate < fmt->sratelo)
d->srate = fmt->sratelo;
if (d->srate > fmt->sratehi)
d->srate = fmt->sratehi;
- dprintk((KERN_DEBUG "usbaudio: set_format_in: usb_set_interface %u %u\n", alts->desc.bInterfaceNumber, fmt->altsetting));
+ dprintk((KERN_DEBUG "usbaudio: set_format_in: usb_set_interface %u %u\n",
+ u->interface, fmt->altsetting));
if (usb_set_interface(dev, alts->desc.bInterfaceNumber, fmt->altsetting) < 0) {
printk(KERN_WARNING "usbaudio: usb_set_interface failed, device %d interface %d altsetting %d\n",
dev->devnum, u->interface, fmt->altsetting);
@@ -1600,7 +1606,6 @@ static int set_format_in(struct usb_audi
static int set_format_out(struct usb_audiodev *as)
{
struct usb_device *dev = as->state->usbdev;
- struct usb_host_config *config = dev->actconfig;
struct usb_host_interface *alts;
struct usb_interface *iface;
struct usbout *u = &as->usbout;
@@ -1610,9 +1615,9 @@ static int set_format_out(struct usb_aud
unsigned char data[3];
int fmtnr, ret;
- if (u->interface < 0 || u->interface >= config->desc.bNumInterfaces)
+ iface = usb_ifnum_to_if(dev, u->interface);
+ if (!iface)
return 0;
- iface = config->interface[u->interface];
fmtnr = find_format(as->fmtout, as->numfmtout, d->format, d->srate);
if (fmtnr < 0) {
@@ -1622,7 +1627,7 @@ static int set_format_out(struct usb_aud
fmt = as->fmtout + fmtnr;
u->format = fmt->format;
- alts = &iface->altsetting[fmt->altsetting];
+ alts = usb_altnum_to_altsetting(iface, fmt->altsetting);
u->datapipe = usb_sndisocpipe(dev, alts->endpoint[0].desc.bEndpointAddress & 0xf);
u->syncpipe = u->syncinterval = 0;
if ((alts->endpoint[0].desc.bmAttributes & 0x0c) == 0x04) {
@@ -1637,18 +1642,20 @@ static int set_format_out(struct usb_aud
alts->endpoint[1].desc.bmAttributes != 0x01 ||
alts->endpoint[1].desc.bSynchAddress != 0 ||
alts->endpoint[1].desc.bEndpointAddress != (alts->endpoint[0].desc.bSynchAddress | 0x80)) {
- printk(KERN_ERR "usbaudio: device %d interface %d altsetting %d invalid synch pipe\n",
+ printk(KERN_WARNING "usbaudio: device %d interface %d altsetting %d claims asynch out "
+ "but has invalid synch pipe; treating as adaptive out\n",
dev->devnum, u->interface, fmt->altsetting);
- return -1;
+ } else {
+ u->syncpipe = usb_rcvisocpipe(dev, alts->endpoint[1].desc.bEndpointAddress & 0xf);
+ u->syncinterval = alts->endpoint[1].desc.bRefresh;
}
- u->syncpipe = usb_rcvisocpipe(dev, alts->endpoint[1].desc.bEndpointAddress & 0xf);
- u->syncinterval = alts->endpoint[1].desc.bRefresh;
}
if (d->srate < fmt->sratelo)
d->srate = fmt->sratelo;
if (d->srate > fmt->sratehi)
d->srate = fmt->sratehi;
- dprintk((KERN_DEBUG "usbaudio: set_format_out: usb_set_interface %u %u\n", alts->desc.bInterfaceNumber, fmt->altsetting));
+ dprintk((KERN_DEBUG "usbaudio: set_format_out: usb_set_interface %u %u\n",
+ u->interface, fmt->altsetting));
if (usb_set_interface(dev, u->interface, fmt->altsetting) < 0) {
printk(KERN_WARNING "usbaudio: usb_set_interface failed, device %d interface %d altsetting %d\n",
dev->devnum, u->interface, fmt->altsetting);
@@ -2697,7 +2704,6 @@ static int usb_audio_release(struct inod
struct usb_audiodev *as = (struct usb_audiodev *)file->private_data;
struct usb_audio_state *s;
struct usb_device *dev;
- struct usb_interface *iface;
lock_kernel();
s = as->state;
@@ -2707,19 +2713,15 @@ static int usb_audio_release(struct inod
down(&open_sem);
if (file->f_mode & FMODE_WRITE) {
usbout_stop(as);
- if (dev && as->usbout.interface >= 0) {
- iface = dev->actconfig->interface[as->usbout.interface];
- usb_set_interface(dev, iface->altsetting->desc.bInterfaceNumber, 0);
- }
+ if (dev && as->usbout.interface >= 0)
+ usb_set_interface(dev, as->usbout.interface, 0);
dmabuf_release(&as->usbout.dma);
usbout_release(as);
}
if (file->f_mode & FMODE_READ) {
usbin_stop(as);
- if (dev && as->usbin.interface >= 0) {
- iface = dev->actconfig->interface[as->usbin.interface];
- usb_set_interface(dev, iface->altsetting->desc.bInterfaceNumber, 0);
- }
+ if (dev && as->usbin.interface >= 0)
+ usb_set_interface(dev, as->usbin.interface, 0);
dmabuf_release(&as->usbin.dma);
usbin_release(as);
}
@@ -2824,12 +2826,11 @@ static void usb_audio_parsestreaming(str
{
struct usb_device *dev = s->usbdev;
struct usb_audiodev *as;
- struct usb_host_config *config = dev->actconfig;
struct usb_host_interface *alts;
struct usb_interface *iface;
struct audioformat *fp;
unsigned char *fmt, *csep;
- unsigned int i, j, k, format;
+ unsigned int i, j, k, format, idx;
if (!(as = kmalloc(sizeof(struct usb_audiodev), GFP_KERNEL)))
return;
@@ -2870,9 +2871,10 @@ static void usb_audio_parsestreaming(str
/* search for input formats */
if (asifin >= 0) {
as->usbin.flags = FLG_CONNECTED;
- iface = config->interface[asifin];
- for (i = 0; i < iface->num_altsetting; i++) {
- alts = &iface->altsetting[i];
+ iface = usb_ifnum_to_if(dev, asifin);
+ for (idx = 0; idx < iface->num_altsetting; idx++) {
+ alts = &iface->altsetting[idx];
+ i = alts->desc.bAlternateSetting;
if (alts->desc.bInterfaceClass != USB_CLASS_AUDIO || alts->desc.bInterfaceSubClass != 2)
continue;
if (alts->desc.bNumEndpoints < 1) {
@@ -2951,14 +2953,15 @@ static void usb_audio_parsestreaming(str
/* search for output formats */
if (asifout >= 0) {
as->usbout.flags = FLG_CONNECTED;
- iface = config->interface[asifout];
- for (i = 0; i < iface->num_altsetting; i++) {
- alts = &iface->altsetting[i];
+ iface = usb_ifnum_to_if(dev, asifout);
+ for (idx = 0; idx < iface->num_altsetting; idx++) {
+ alts = &iface->altsetting[idx];
+ i = alts->desc.bAlternateSetting;
if (alts->desc.bInterfaceClass != USB_CLASS_AUDIO || alts->desc.bInterfaceSubClass != 2)
continue;
if (alts->desc.bNumEndpoints < 1) {
/* altsetting 0 should never have iso EPs */
- if (alts->desc.bAlternateSetting != 0)
+ if (i != 0)
printk(KERN_ERR "usbaudio: device %u interface %u altsetting %u does not have an endpoint\n",
dev->devnum, asifout, i);
continue;
@@ -3655,8 +3658,8 @@ static void usb_audio_constructmixer(str
static struct usb_audio_state *usb_audio_parsecontrol(struct usb_device *dev, unsigned char *buffer, unsigned int buflen, unsigned int ctrlif)
{
struct usb_audio_state *s;
- struct usb_host_config *config = dev->actconfig;
struct usb_interface *iface;
+ struct usb_host_interface *alt;
unsigned char ifin[USB_MAXINTERFACES], ifout[USB_MAXINTERFACES];
unsigned char *p1;
unsigned int i, j, k, numifin = 0, numifout = 0;
@@ -3685,54 +3688,63 @@ static struct usb_audio_state *usb_audio
dev->devnum, ctrlif);
for (i = 0; i < p1[7]; i++) {
j = p1[8+i];
- if (j >= config->desc.bNumInterfaces) {
+ iface = usb_ifnum_to_if(dev, j);
+ if (!iface) {
printk(KERN_ERR "usbaudio: device %d audiocontrol interface %u interface %u does not exist\n",
dev->devnum, ctrlif, j);
continue;
}
- iface = config->interface[j];
- if (iface->altsetting[0].desc.bInterfaceClass != USB_CLASS_AUDIO) {
- printk(KERN_ERR "usbaudio: device %d audiocontrol interface %u interface %u is not an AudioClass interface\n",
- dev->devnum, ctrlif, j);
+ if (iface->num_altsetting == 1) {
+ printk(KERN_ERR "usbaudio: device %d audiocontrol interface %u has only 1 altsetting.\n", dev->devnum, ctrlif);
continue;
}
- if (iface->altsetting[0].desc.bInterfaceSubClass == 3) {
- printk(KERN_INFO "usbaudio: device %d audiocontrol interface %u interface %u MIDIStreaming not supported\n",
+ alt = usb_altnum_to_altsetting(iface, 0);
+ if (!alt) {
+ printk(KERN_ERR "usbaudio: device %d audiocontrol interface %u interface %u has no altsetting 0\n",
dev->devnum, ctrlif, j);
continue;
}
- if (iface->altsetting[0].desc.bInterfaceSubClass != 2) {
- printk(KERN_ERR "usbaudio: device %d audiocontrol interface %u interface %u invalid AudioClass subtype\n",
+ if (alt->desc.bInterfaceClass != USB_CLASS_AUDIO) {
+ printk(KERN_ERR "usbaudio: device %d audiocontrol interface %u interface %u is not an AudioClass interface\n",
dev->devnum, ctrlif, j);
continue;
}
- if (iface->num_altsetting == 0) {
- printk(KERN_ERR "usbaudio: device %d audiocontrol interface %u has no working interface.\n", dev->devnum, ctrlif);
+ if (alt->desc.bInterfaceSubClass == 3) {
+ printk(KERN_INFO "usbaudio: device %d audiocontrol interface %u interface %u MIDIStreaming not supported\n",
+ dev->devnum, ctrlif, j);
continue;
}
- if (iface->num_altsetting == 1) {
- printk(KERN_ERR "usbaudio: device %d audiocontrol interface %u has only 1 altsetting.\n", dev->devnum, ctrlif);
+ if (alt->desc.bInterfaceSubClass != 2) {
+ printk(KERN_ERR "usbaudio: device %d audiocontrol interface %u interface %u invalid AudioClass subtype\n",
+ dev->devnum, ctrlif, j);
continue;
}
- if (iface->altsetting[0].desc.bNumEndpoints > 0) {
+ if (alt->desc.bNumEndpoints > 0) {
/* Check all endpoints; should they all have a bandwidth of 0 ? */
- for (k = 0; k < iface->altsetting[0].desc.bNumEndpoints; k++) {
- if (iface->altsetting[0].endpoint[k].desc.wMaxPacketSize > 0) {
+ for (k = 0; k < alt->desc.bNumEndpoints; k++) {
+ if (alt->endpoint[k].desc.wMaxPacketSize > 0) {
printk(KERN_ERR "usbaudio: device %d audiocontrol interface %u endpoint %d does not have 0 bandwidth at alt[0]\n", dev->devnum, ctrlif, k);
break;
}
}
- if (k < iface->altsetting[0].desc.bNumEndpoints)
+ if (k < alt->desc.bNumEndpoints)
continue;
}
- if (iface->altsetting[1].desc.bNumEndpoints < 1) {
+
+ alt = usb_altnum_to_altsetting(iface, 1);
+ if (!alt) {
+ printk(KERN_ERR "usbaudio: device %d audiocontrol interface %u interface %u has no altsetting 1\n",
+ dev->devnum, ctrlif, j);
+ continue;
+ }
+ if (alt->desc.bNumEndpoints < 1) {
printk(KERN_ERR "usbaudio: device %d audiocontrol interface %u interface %u has no endpoint\n",
dev->devnum, ctrlif, j);
continue;
}
/* note: this requires the data endpoint to be ep0 and the optional sync
ep to be ep1, which seems to be the case */
- if (iface->altsetting[1].endpoint[0].desc.bEndpointAddress & USB_DIR_IN) {
+ if (alt->endpoint[0].desc.bEndpointAddress & USB_DIR_IN) {
if (numifin < USB_MAXINTERFACES) {
ifin[numifin++] = j;
usb_driver_claim_interface(&usb_audio_driver, iface, (void *)-1);
@@ -3779,12 +3791,9 @@ static int usb_audio_probe(struct usb_in
const struct usb_device_id *id)
{
struct usb_device *dev = interface_to_usbdev (intf);
- struct usb_host_config *config = dev->actconfig;
struct usb_audio_state *s;
unsigned char *buffer;
- unsigned char buf[8];
- unsigned int i, buflen;
- int ret;
+ unsigned int buflen;
#if 0
printk(KERN_DEBUG "usbaudio: Probing if %i: IC %x, ISC %x\n", ifnum,
@@ -3796,26 +3805,8 @@ static int usb_audio_probe(struct usb_in
* audiocontrol interface found
* find which configuration number is active
*/
- i = dev->actconfig - config;
-
- ret = usb_get_descriptor(dev, USB_DT_CONFIG, i, buf, 8);
- if (ret < 0) {
- printk(KERN_ERR "usbaudio: cannot get first 8 bytes of config descriptor %d of device %d (error %d)\n", i, dev->devnum, ret);
- return -EIO;
- }
- if (buf[1] != USB_DT_CONFIG || buf[0] < 9) {
- printk(KERN_ERR "usbaudio: invalid config descriptor %d of device %d\n", i, dev->devnum);
- return -EIO;
- }
- buflen = buf[2] | (buf[3] << 8);
- if (!(buffer = kmalloc(buflen, GFP_KERNEL)))
- return -ENOMEM;
- ret = usb_get_descriptor(dev, USB_DT_CONFIG, i, buffer, buflen);
- if (ret < 0) {
- kfree(buffer);
- printk(KERN_ERR "usbaudio: cannot get config descriptor %d of device %d (error %d)\n", i, dev->devnum, ret);
- return -EIO;
- }
+ buffer = dev->rawdescriptors[dev->actconfig - dev->config];
+ buflen = dev->actconfig->desc.wTotalLength;
s = usb_audio_parsecontrol(dev, buffer, buflen, intf->altsetting->desc.bInterfaceNumber);
if (s) {
usb_set_intfdata (intf, s);
--- diff/drivers/usb/class/bluetty.c 2003-09-17 11:28:10.000000000 +0000
+++ source/drivers/usb/class/bluetty.c 2004-03-16 11:27:48.254851808 +0000
@@ -1025,7 +1025,7 @@ static int usb_bluetooth_probe (struct u
int num_bulk_in = 0;
int num_bulk_out = 0;
- interface = &intf->altsetting[0];
+ interface = intf->cur_altsetting;
control_out_endpoint = interface->desc.bInterfaceNumber;
/* find the endpoints that we need */
--- diff/drivers/usb/class/cdc-acm.c 2004-02-18 08:54:11.000000000 +0000
+++ source/drivers/usb/class/cdc-acm.c 2004-03-16 11:27:48.254851808 +0000
@@ -595,12 +595,12 @@ static int acm_probe (struct usb_interfa
* is there it's not for call management ... so use
* the cdc union descriptor whenever there is one.
*/
- ifcom = intf->altsetting + 0;
+ ifcom = intf->cur_altsetting;
if (intf == cfacm->interface[j]) {
- ifdata = cfacm->interface[j + 1]->altsetting + 0;
+ ifdata = cfacm->interface[j + 1]->cur_altsetting;
data = cfacm->interface[j + 1];
} else if (intf == cfacm->interface[j + 1]) {
- ifdata = cfacm->interface[j]->altsetting + 0;
+ ifdata = cfacm->interface[j]->cur_altsetting;
data = cfacm->interface[j];
} else
continue;
--- diff/drivers/usb/class/usb-midi.c 2003-09-30 14:46:17.000000000 +0000
+++ source/drivers/usb/class/usb-midi.c 2004-03-16 11:27:48.258851200 +0000
@@ -39,9 +39,6 @@
#include <linux/init.h>
#include <asm/semaphore.h>
-/** This declaration is missing from linux/usb.h **/
-extern int usb_get_string(struct usb_device *dev, unsigned short langid, unsigned char index, void *buf, int size);
-
#include "usb-midi.h"
/* ------------------------------------------------------------------------- */
@@ -1519,15 +1516,17 @@ static int on_bits( unsigned short v )
static int get_alt_setting( struct usb_device *d, int ifnum )
{
int alts, alt=0;
+ struct usb_interface *iface;
struct usb_host_interface *interface;
struct usb_endpoint_descriptor *ep;
int epin, epout;
int i;
- alts = d->actconfig->interface[ifnum]->num_altsetting;
+ iface = usb_ifnum_to_if( d, ifnum );
+ alts = iface->num_altsetting;
for ( alt=0 ; alt<alts ; alt++ ) {
- interface = &d->actconfig->interface[ifnum]->altsetting[alt];
+ interface = &iface->altsetting[alt];
epin = -1;
epout = -1;
@@ -1542,7 +1541,7 @@ static int get_alt_setting( struct usb_d
epout = i;
}
if ( epin >= 0 && epout >= 0 ) {
- return alt;
+ return interface->desc.bAlternateSetting;
}
}
}
@@ -1780,12 +1779,13 @@ static int alloc_usb_midi_device( struct
* Called by usb_midi_probe();
**/
-static int detect_yamaha_device( struct usb_device *d, unsigned int ifnum, struct usb_midi_state *s)
+static int detect_yamaha_device( struct usb_device *d,
+ struct usb_interface *iface, unsigned int ifnum,
+ struct usb_midi_state *s)
{
- struct usb_host_config *c = d->actconfig;
struct usb_host_interface *interface;
struct usb_midi_device *u;
- unsigned char buf[USB_DT_CONFIG_SIZE], *buffer;
+ unsigned char *buffer;
int bufSize;
int i;
int alts=-1;
@@ -1795,13 +1795,13 @@ static int detect_yamaha_device( struct
return -EINVAL;
}
- for ( i=0 ; i < c->interface[ifnum]->num_altsetting; i++ ) {
- interface = c->interface[ifnum]->altsetting + i;
+ for ( i=0 ; i < iface->num_altsetting; i++ ) {
+ interface = iface->altsetting + i;
if ( interface->desc.bInterfaceClass != 255 ||
interface->desc.bInterfaceSubClass != 0 )
continue;
- alts = i;
+ alts = interface->desc.bAlternateSetting;
}
if ( alts == -1 ) {
return -EINVAL;
@@ -1810,30 +1810,11 @@ static int detect_yamaha_device( struct
printk(KERN_INFO "usb-midi: Found YAMAHA USB-MIDI device on dev %04x:%04x, iface %d\n",
d->descriptor.idVendor, d->descriptor.idProduct, ifnum);
- ret = usb_get_descriptor( d, USB_DT_CONFIG, i, buf, USB_DT_CONFIG_SIZE );
- if ( ret < 0 ) {
- printk(KERN_INFO "usb-midi: Could not get config (error=%d).\n", ret);
- return -EINVAL;
- }
- if ( buf[1] != USB_DT_CONFIG || buf[0] < USB_DT_CONFIG_SIZE ) {
- printk(KERN_INFO "usb-midi: config not as expected.\n");
- return -EINVAL;
- }
- bufSize = buf[2] | buf[3]<<8;
- buffer = (unsigned char *)kmalloc(sizeof(unsigned char)*bufSize, GFP_KERNEL);
- if ( !buffer ) {
- printk(KERN_INFO "usb-midi: Could not allocate memory.\n");
- return -EINVAL;
- }
- ret = usb_get_descriptor( d, USB_DT_CONFIG, i, buffer, bufSize );
- if ( ret < 0 ) {
- printk(KERN_INFO "usb-midi: Could not get full config (error=%d).\n", ret);
- kfree(buffer);
- return -EINVAL;
- }
+ i = d->actconfig - d->config;
+ buffer = d->rawdescriptors[i];
+ bufSize = d->actconfig->desc.wTotalLength;
u = parse_descriptor( d, buffer, bufSize, ifnum, alts, 1);
- kfree(buffer);
if ( u == NULL ) {
return -EINVAL;
}
@@ -1878,24 +1859,25 @@ static int detect_vendor_specific_device
* Returns 0 on success, negative on failure.
* Called by usb_midi_probe();
**/
-static int detect_midi_subclass(struct usb_device *d, unsigned int ifnum, struct usb_midi_state *s)
+static int detect_midi_subclass(struct usb_device *d,
+ struct usb_interface *iface, unsigned int ifnum,
+ struct usb_midi_state *s)
{
- struct usb_host_config *c = d->actconfig;
struct usb_host_interface *interface;
struct usb_midi_device *u;
- unsigned char buf[USB_DT_CONFIG_SIZE], *buffer;
+ unsigned char *buffer;
int bufSize;
int i;
int alts=-1;
int ret;
- for ( i=0 ; i < c->interface[ifnum]->num_altsetting; i++ ) {
- interface = c->interface[ifnum]->altsetting + i;
+ for ( i=0 ; i < iface->num_altsetting; i++ ) {
+ interface = iface->altsetting + i;
if ( interface->desc.bInterfaceClass != USB_CLASS_AUDIO ||
interface->desc.bInterfaceSubClass != USB_SUBCLASS_MIDISTREAMING )
continue;
- alts = i;
+ alts = interface->desc.bAlternateSetting;
}
if ( alts == -1 ) {
return -EINVAL;
@@ -1915,30 +1897,11 @@ static int detect_midi_subclass(struct u
descriptor they modify or extend.
*/
- ret = usb_get_descriptor( d, USB_DT_CONFIG, i, buf, USB_DT_CONFIG_SIZE );
- if ( ret < 0 ) {
- printk(KERN_INFO "usb-midi: Could not get config (error=%d).\n", ret);
- return -EINVAL;
- }
- if ( buf[1] != USB_DT_CONFIG || buf[0] < USB_DT_CONFIG_SIZE ) {
- printk(KERN_INFO "usb-midi: config not as expected.\n");
- return -EINVAL;
- }
- bufSize = buf[2] | buf[3]<<8;
- buffer = (unsigned char *)kmalloc(sizeof(unsigned char)*bufSize, GFP_KERNEL);
- if ( !buffer ) {
- printk(KERN_INFO "usb-midi: Could not allocate memory.\n");
- return -EINVAL;
- }
- ret = usb_get_descriptor( d, USB_DT_CONFIG, i, buffer, bufSize );
- if ( ret < 0 ) {
- printk(KERN_INFO "usb-midi: Could not get full config (error=%d).\n", ret);
- kfree(buffer);
- return -EINVAL;
- }
+ i = d->actconfig - d->config;
+ buffer = d->rawdescriptors[i];
+ bufSize = d->actconfig->desc.wTotalLength;
u = parse_descriptor( d, buffer, bufSize, ifnum, alts, 0);
- kfree(buffer);
if ( u == NULL ) {
return -EINVAL;
}
@@ -2002,7 +1965,7 @@ static int usb_midi_probe(struct usb_int
{
struct usb_midi_state *s;
struct usb_device *dev = interface_to_usbdev(intf);
- int ifnum = intf->altsetting->desc.bInterfaceNumber;
+ int ifnum = intf->cur_altsetting->desc.bInterfaceNumber;
s = (struct usb_midi_state *)kmalloc(sizeof(struct usb_midi_state), GFP_KERNEL);
if ( !s )
@@ -2018,9 +1981,9 @@ static int usb_midi_probe(struct usb_int
if (
detect_by_hand( dev, ifnum, s ) &&
- detect_midi_subclass( dev, ifnum, s ) &&
+ detect_midi_subclass( dev, intf, ifnum, s ) &&
detect_vendor_specific_device( dev, ifnum, s ) &&
- detect_yamaha_device( dev, ifnum, s) ) {
+ detect_yamaha_device( dev, intf, ifnum, s) ) {
kfree(s);
return -EIO;
}
--- diff/drivers/usb/class/usblp.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/usb/class/usblp.c 2004-03-16 11:27:48.255851656 +0000
@@ -133,6 +133,7 @@ struct usblp {
wait_queue_head_t wait; /* Zzzzz ... */
int readcount; /* Counter for reads */
int ifnum; /* Interface number */
+ struct usb_interface *intf; /* The interface */
/* Alternate-setting numbers and endpoints for each protocol
* (7/1/{index=1,2,3}) that the device supports: */
struct {
@@ -609,8 +610,10 @@ static ssize_t usblp_write(struct file *
while (writecount < count) {
if (!usblp->wcomplete) {
barrier();
- if (file->f_flags & O_NONBLOCK)
+ if (file->f_flags & O_NONBLOCK) {
+ writecount += transfer_length;
return writecount ? writecount : -EAGAIN;
+ }
timeout = USBLP_WRITE_TIMEOUT;
add_wait_queue(&usblp->wait, &wait);
@@ -670,7 +673,8 @@ static ssize_t usblp_write(struct file *
usblp->writeurb->transfer_buffer_length = transfer_length;
- if (copy_from_user(usblp->writeurb->transfer_buffer, buffer + writecount, transfer_length)) {
+ if (copy_from_user(usblp->writeurb->transfer_buffer,
+ buffer + writecount, transfer_length)) {
up(&usblp->sem);
return writecount ? writecount : -EFAULT;
}
@@ -837,7 +841,8 @@ static int usblp_probe(struct usb_interf
usblp->dev = dev;
init_MUTEX (&usblp->sem);
init_waitqueue_head(&usblp->wait);
- usblp->ifnum = intf->altsetting->desc.bInterfaceNumber;
+ usblp->ifnum = intf->cur_altsetting->desc.bInterfaceNumber;
+ usblp->intf = intf;
usblp->writeurb = usb_alloc_urb(0, GFP_KERNEL);
if (!usblp->writeurb) {
@@ -973,7 +978,7 @@ static int usblp_select_alts(struct usbl
struct usb_endpoint_descriptor *epd, *epwrite, *epread;
int p, i, e;
- if_alt = usblp->dev->actconfig->interface[usblp->ifnum];
+ if_alt = usblp->intf;
for (p = 0; p < USBLP_MAX_PROTOCOLS; p++)
usblp->protocol[p].alt_setting = -1;
@@ -1022,7 +1027,8 @@ static int usblp_select_alts(struct usbl
epread = NULL;
}
- usblp->protocol[ifd->desc.bInterfaceProtocol].alt_setting = i;
+ usblp->protocol[ifd->desc.bInterfaceProtocol].alt_setting =
+ ifd->desc.bAlternateSetting;
usblp->protocol[ifd->desc.bInterfaceProtocol].epwrite = epwrite;
usblp->protocol[ifd->desc.bInterfaceProtocol].epread = epread;
}
--- diff/drivers/usb/core/buffer.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/usb/core/buffer.c 2004-03-16 11:27:48.259851048 +0000
@@ -51,8 +51,8 @@ static const size_t pool_max [HCD_BUFFER
* @hcd: the bus whose buffer pools are to be initialized
* Context: !in_interrupt()
*
- * Call this as part of initializing a host controller that uses the pci dma
- * memory allocators. It initializes some pools of dma-consistent memory that
+ * Call this as part of initializing a host controller that uses the dma
+ * memory allocators. It initializes some pools of dma-coherent memory that
* will be shared by all drivers using that controller, or returns a negative
* errno value on error.
*
@@ -115,6 +115,12 @@ void *hcd_buffer_alloc (
struct usb_hcd *hcd = bus->hcpriv;
int i;
+ /* some USB hosts just use PIO */
+ if (!bus->controller->dma_mask) {
+ *dma = ~(dma_addr_t) 0;
+ return kmalloc (size, mem_flags);
+ }
+
for (i = 0; i < HCD_BUFFER_POOLS; i++) {
if (size <= pool_max [i])
return dma_pool_alloc (hcd->pool [i], mem_flags, dma);
@@ -134,6 +140,12 @@ void hcd_buffer_free (
if (!addr)
return;
+
+ if (!bus->controller->dma_mask) {
+ kfree (addr);
+ return;
+ }
+
for (i = 0; i < HCD_BUFFER_POOLS; i++) {
if (size <= pool_max [i]) {
dma_pool_free (hcd->pool [i], addr, dma);
--- diff/drivers/usb/core/config.c 2003-09-30 14:46:17.000000000 +0000
+++ source/drivers/usb/core/config.c 2004-03-16 11:27:48.260850896 +0000
@@ -72,13 +72,10 @@ static int usb_parse_endpoint(struct usb
return buffer - buffer0;
}
-static void usb_release_intf(struct device *dev)
+static void usb_free_intf(struct usb_interface *intf)
{
- struct usb_interface *intf;
int j;
- intf = to_usb_interface(dev);
-
if (intf->altsetting) {
for (j = 0; j < intf->num_altsetting; j++) {
struct usb_host_interface *as = &intf->altsetting[j];
@@ -235,8 +232,6 @@ int usb_parse_configuration(struct usb_h
return -ENOMEM;
}
memset(interface, 0, sizeof(struct usb_interface));
- interface->dev.release = usb_release_intf;
- device_initialize(&interface->dev);
}
/* Go through the descriptors, checking their length and counting the
@@ -374,7 +369,7 @@ void usb_destroy_configuration(struct us
struct usb_interface *ifp = cf->interface[i];
if (ifp)
- put_device(&ifp->dev);
+ usb_free_intf(ifp);
}
}
kfree(dev->config);
--- diff/drivers/usb/core/devio.c 2003-12-19 09:51:11.000000000 +0000
+++ source/drivers/usb/core/devio.c 2004-03-16 11:27:48.261850744 +0000
@@ -430,19 +430,14 @@ static int findintfep(struct usb_device
static int findintfif(struct usb_device *dev, unsigned int ifn)
{
- unsigned int i, j;
- struct usb_interface *iface;
- struct usb_host_interface *alts;
+ unsigned int i;
if (ifn & ~0xff)
return -EINVAL;
for (i = 0; i < dev->actconfig->desc.bNumInterfaces; i++) {
- iface = dev->actconfig->interface[i];
- for (j = 0; j < iface->num_altsetting; j++) {
- alts = &iface->altsetting[j];
- if (alts->desc.bInterfaceNumber == ifn)
- return i;
- }
+ if (dev->actconfig->interface[i]->
+ altsetting[0].desc.bInterfaceNumber == ifn)
+ return i;
}
return -ENOENT;
}
@@ -688,9 +683,7 @@ static int proc_getdriver(struct dev_sta
return -EFAULT;
if ((ret = findintfif(ps->dev, gd.interface)) < 0)
return ret;
- interface = usb_ifnum_to_if(ps->dev, gd.interface);
- if (!interface)
- return -EINVAL;
+ interface = ps->dev->actconfig->interface[ret];
if (!interface->driver)
return -ENODATA;
strcpy(gd.driver, interface->driver->name);
@@ -744,9 +737,7 @@ static int proc_setintf(struct dev_state
return -EFAULT;
if ((ret = findintfif(ps->dev, setintf.interface)) < 0)
return ret;
- interface = usb_ifnum_to_if(ps->dev, setintf.interface);
- if (!interface)
- return -EINVAL;
+ interface = ps->dev->actconfig->interface[ret];
if (interface->driver) {
if ((ret = checkintf(ps, ret)))
return ret;
--- diff/drivers/usb/core/driverfs.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/core/driverfs.c 2004-03-16 11:27:48.261850744 +0000
@@ -166,13 +166,9 @@ void usb_create_driverfs_dev_files (stru
static ssize_t \
show_##field (struct device *dev, char *buf) \
{ \
- struct usb_interface *intf; \
- int alt; \
+ struct usb_interface *intf = to_usb_interface (dev); \
\
- intf = to_usb_interface (dev); \
- alt = intf->act_altsetting; \
- \
- return sprintf (buf, format_string, intf->altsetting[alt].desc.field); \
+ return sprintf (buf, format_string, intf->cur_altsetting->desc.field); \
} \
static DEVICE_ATTR(field, S_IRUGO, show_##field, NULL);
--- diff/drivers/usb/core/hcd-pci.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/usb/core/hcd-pci.c 2004-03-16 11:27:48.263850440 +0000
@@ -147,8 +147,12 @@ clean_2:
hcd->driver = driver;
hcd->description = driver->description;
hcd->self.bus_name = pci_name(dev);
+#ifdef CONFIG_PCI_NAMES
+ hcd->product_desc = dev->pretty_name;
+#else
if (hcd->product_desc == NULL)
hcd->product_desc = "USB Host Controller";
+#endif
hcd->self.controller = &dev->dev;
if ((retval = hcd_buffer_create (hcd)) != 0) {
--- diff/drivers/usb/core/hcd.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/usb/core/hcd.c 2004-03-16 11:27:48.263850440 +0000
@@ -1213,7 +1213,7 @@ static int hcd_unlink_urb (struct urb *u
break;
}
if (tmp != &urb->urb_list) {
- retval = -EINVAL;
+ retval = -EIDRM;
goto done;
}
@@ -1294,7 +1294,7 @@ done:
spin_unlock (&hcd_data_lock);
spin_unlock_irqrestore (&urb->lock, flags);
bye:
- if (retval && sys && sys->driver)
+ if (retval != -EIDRM && sys && sys->driver)
dev_dbg (sys, "hcd_unlink_urb %p fail %d\n", urb, retval);
return retval;
}
--- diff/drivers/usb/core/hub.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/usb/core/hub.c 2004-03-16 11:27:48.264850288 +0000
@@ -560,7 +560,7 @@ static int hub_probe(struct usb_interfac
struct usb_hub *hub;
unsigned long flags;
- desc = intf->altsetting + intf->act_altsetting;
+ desc = intf->cur_altsetting;
dev = interface_to_usbdev(intf);
/* Some hubs have a subclass of 1, which AFAICT according to the */
@@ -1344,15 +1344,15 @@ int usb_physical_reset_device(struct usb
for (i = 0; i < dev->actconfig->desc.bNumInterfaces; i++) {
struct usb_interface *intf = dev->actconfig->interface[i];
- struct usb_interface_descriptor *as;
+ struct usb_interface_descriptor *desc;
- as = &intf->altsetting[intf->act_altsetting].desc;
- ret = usb_set_interface(dev, as->bInterfaceNumber,
- as->bAlternateSetting);
+ desc = &intf->cur_altsetting->desc;
+ ret = usb_set_interface(dev, desc->bInterfaceNumber,
+ desc->bAlternateSetting);
if (ret < 0) {
err("failed to set active alternate setting "
"for dev %s interface %d (error=%d)",
- dev->devpath, i, ret);
+ dev->devpath, desc->bInterfaceNumber, ret);
return ret;
}
}
--- diff/drivers/usb/core/message.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/usb/core/message.c 2004-03-16 11:27:48.266849984 +0000
@@ -783,16 +783,22 @@ void usb_disable_endpoint(struct usb_dev
*/
void usb_disable_interface(struct usb_device *dev, struct usb_interface *intf)
{
- struct usb_host_interface *hintf =
- &intf->altsetting[intf->act_altsetting];
+ struct usb_host_interface *alt = intf->cur_altsetting;
int i;
- for (i = 0; i < hintf->desc.bNumEndpoints; ++i) {
+ for (i = 0; i < alt->desc.bNumEndpoints; ++i) {
usb_disable_endpoint(dev,
- hintf->endpoint[i].desc.bEndpointAddress);
+ alt->endpoint[i].desc.bEndpointAddress);
}
}
+static void release_interface(struct device *dev)
+{
+ struct usb_interface *interface = to_usb_interface(dev);
+
+ complete(interface->released);
+}
+
/*
* usb_disable_device - Disable all the endpoints for a USB device
* @dev: the device whose endpoints are being disabled
@@ -822,12 +828,16 @@ void usb_disable_device(struct usb_devic
if (dev->actconfig) {
for (i = 0; i < dev->actconfig->desc.bNumInterfaces; i++) {
struct usb_interface *interface;
+ struct completion intf_completion;
/* remove this interface */
interface = dev->actconfig->interface[i];
dev_dbg (&dev->dev, "unregistering interface %s\n",
interface->dev.bus_id);
- device_del(&interface->dev);
+ init_completion (&intf_completion);
+ interface->released = &intf_completion;
+ device_unregister (&interface->dev);
+ wait_for_completion (&intf_completion);
}
dev->actconfig = 0;
if (dev->state == USB_STATE_CONFIGURED)
@@ -876,12 +886,11 @@ void usb_enable_endpoint(struct usb_devi
void usb_enable_interface(struct usb_device *dev,
struct usb_interface *intf)
{
- struct usb_host_interface *hintf =
- &intf->altsetting[intf->act_altsetting];
+ struct usb_host_interface *alt = intf->cur_altsetting;
int i;
- for (i = 0; i < hintf->desc.bNumEndpoints; ++i)
- usb_enable_endpoint(dev, &hintf->endpoint[i].desc);
+ for (i = 0; i < alt->desc.bNumEndpoints; ++i)
+ usb_enable_endpoint(dev, &alt->endpoint[i].desc);
}
/**
@@ -920,6 +929,7 @@ void usb_enable_interface(struct usb_dev
int usb_set_interface(struct usb_device *dev, int interface, int alternate)
{
struct usb_interface *iface;
+ struct usb_host_interface *alt;
int ret;
int manual = 0;
@@ -929,14 +939,15 @@ int usb_set_interface(struct usb_device
return -EINVAL;
}
- if (alternate < 0 || alternate >= iface->num_altsetting)
+ alt = usb_altnum_to_altsetting(iface, alternate);
+ if (!alt) {
+ warn("selecting invalid altsetting %d", alternate);
return -EINVAL;
+ }
ret = usb_control_msg(dev, usb_sndctrlpipe(dev, 0),
USB_REQ_SET_INTERFACE, USB_RECIP_INTERFACE,
- iface->altsetting[alternate]
- .desc.bAlternateSetting,
- interface, NULL, 0, HZ * 5);
+ alternate, interface, NULL, 0, HZ * 5);
/* 9.4.10 says devices don't need this and are free to STALL the
* request if the interface only has one alternate setting.
@@ -957,7 +968,7 @@ int usb_set_interface(struct usb_device
/* prevent submissions using previous endpoint settings */
usb_disable_interface(dev, iface);
- iface->act_altsetting = alternate;
+ iface->cur_altsetting = alt;
/* If the interface only has one altsetting and the device didn't
* accept the request, we attempt to carry out the equivalent action
@@ -965,13 +976,11 @@ int usb_set_interface(struct usb_device
* new altsetting.
*/
if (manual) {
- struct usb_host_interface *iface_as =
- &iface->altsetting[alternate];
int i;
- for (i = 0; i < iface_as->desc.bNumEndpoints; i++) {
+ for (i = 0; i < alt->desc.bNumEndpoints; i++) {
unsigned int epaddr =
- iface_as->endpoint[i].desc.bEndpointAddress;
+ alt->endpoint[i].desc.bEndpointAddress;
unsigned int pipe =
__create_pipe(dev, USB_ENDPOINT_NUMBER_MASK & epaddr)
| (usb_endpoint_out(epaddr) ? USB_DIR_OUT : USB_DIR_IN);
@@ -1045,8 +1054,19 @@ int usb_reset_configuration(struct usb_d
/* re-init hc/hcd interface/endpoint state */
for (i = 0; i < config->desc.bNumInterfaces; i++) {
struct usb_interface *intf = config->interface[i];
+ struct usb_host_interface *alt;
+
+ alt = usb_altnum_to_altsetting(intf, 0);
- intf->act_altsetting = 0;
+ /* No altsetting 0? We'll assume the first altsetting.
+ * We could use a GetInterface call, but if a device is
+ * so non-compliant that it doesn't have altsetting 0
+ * then I wouldn't trust its reply anyway.
+ */
+ if (!alt)
+ alt = &intf->altsetting[0];
+
+ intf->cur_altsetting = alt;
usb_enable_interface(dev, intf);
}
return 0;
@@ -1135,25 +1155,34 @@ int usb_set_configuration(struct usb_dev
*/
for (i = 0; i < cp->desc.bNumInterfaces; ++i) {
struct usb_interface *intf = cp->interface[i];
- struct usb_interface_descriptor *desc;
+ struct usb_host_interface *alt;
- intf->act_altsetting = 0;
- desc = &intf->altsetting [0].desc;
- usb_enable_interface(dev, intf);
+ alt = usb_altnum_to_altsetting(intf, 0);
+ /* No altsetting 0? We'll assume the first altsetting.
+ * We could use a GetInterface call, but if a device is
+ * so non-compliant that it doesn't have altsetting 0
+ * then I wouldn't trust its reply anyway.
+ */
+ if (!alt)
+ alt = &intf->altsetting[0];
+
+ intf->cur_altsetting = alt;
+ usb_enable_interface(dev, intf);
intf->dev.parent = &dev->dev;
intf->dev.driver = NULL;
intf->dev.bus = &usb_bus_type;
intf->dev.dma_mask = dev->dev.dma_mask;
+ intf->dev.release = release_interface;
sprintf (&intf->dev.bus_id[0], "%d-%s:%d.%d",
dev->bus->busnum, dev->devpath,
configuration,
- desc->bInterfaceNumber);
+ alt->desc.bInterfaceNumber);
dev_dbg (&dev->dev,
"registering %s (config #%d, interface %d)\n",
intf->dev.bus_id, configuration,
- desc->bInterfaceNumber);
- device_add (&intf->dev);
+ alt->desc.bInterfaceNumber);
+ device_register (&intf->dev);
usb_create_driverfs_intf_files (intf);
}
}
--- diff/drivers/usb/core/urb.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/core/urb.c 2004-03-16 11:27:48.267849832 +0000
@@ -116,7 +116,8 @@ struct urb * usb_get_urb(struct urb *urb
* describing that request to the USB subsystem. Request completion will
* be indicated later, asynchronously, by calling the completion handler.
* The three types of completion are success, error, and unlink
- * (also called "request cancellation").
+ * (a software-induced fault, also called "request cancelation").
+ *
* URBs may be submitted in interrupt context.
*
* The caller must have correctly initialized the URB before submitting
@@ -127,12 +128,23 @@ struct urb * usb_get_urb(struct urb *urb
*
* Successful submissions return 0; otherwise this routine returns a
* negative error number. If the submission is successful, the complete()
- * callback from the urb will be called exactly once, when the USB core and
- * host controller driver are finished with the urb. When the completion
+ * callback from the URB will be called exactly once, when the USB core and
+ * Host Controller Driver (HCD) are finished with the URB. When the completion
* function is called, control of the URB is returned to the device
* driver which issued the request. The completion handler may then
* immediately free or reuse that URB.
*
+ * With few exceptions, USB device drivers should never access URB fields
+ * provided by usbcore or the HCD until its complete() is called.
+ * The exceptions relate to periodic transfer scheduling. For both
+ * interrupt and isochronous urbs, as part of successful URB submission
+ * urb->interval is modified to reflect the actual transfer period used
+ * (normally some power of two units). And for isochronous urbs,
+ * urb->start_frame is modified to reflect when the URB's transfers were
+ * scheduled to start. Not all isochronous transfer scheduling policies
+ * will work, but most host controller drivers should easily handle ISO
+ * queues going from now until 10-200 msec into the future.
+ *
* For control endpoints, the synchronous usb_control_msg() call is
* often used (in non-interrupt context) instead of this call.
* That is often used through convenience wrappers, for the requests
@@ -143,15 +155,17 @@ struct urb * usb_get_urb(struct urb *urb
*
* URBs may be submitted to endpoints before previous ones complete, to
* minimize the impact of interrupt latencies and system overhead on data
- * throughput. This is required for continuous isochronous data streams,
+ * throughput. With that queuing policy, an endpoint's queue would never
+ * be empty. This is required for continuous isochronous data streams,
* and may also be required for some kinds of interrupt transfers. Such
- * queueing also maximizes bandwidth utilization by letting USB controllers
+ * queuing also maximizes bandwidth utilization by letting USB controllers
* start work on later requests before driver software has finished the
- * completion processing for earlier requests.
+ * completion processing for earlier (successful) requests.
*
- * Bulk and Isochronous URBs may always be queued. At this writing, all
- * mainstream host controller drivers support queueing for control and
- * interrupt transfer requests.
+ * As of Linux 2.6, all USB endpoint transfer queues support depths greater
+ * than one. This was previously a HCD-specific behavior, except for ISO
+ * transfers. Non-isochronous endpoint queues are inactive during cleanup
+ * after faults (transfer errors or cancelation).
*
* Reserved Bandwidth Transfers:
*
@@ -389,7 +403,7 @@ int usb_submit_urb(struct urb *urb, int
* When the URB_ASYNC_UNLINK transfer flag for the URB is clear, this
* request is synchronous. Success is indicated by returning zero,
* at which time the urb will have been unlinked and its completion
- * handler will have been called with urb->status -ENOENT. Failure is
+ * handler will have been called with urb->status == -ENOENT. Failure is
* indicated by any other return value.
*
* The synchronous cancelation mode may not be used
@@ -400,8 +414,37 @@ int usb_submit_urb(struct urb *urb, int
* When the URB_ASYNC_UNLINK transfer flag for the URB is set, this
* request is asynchronous. Success is indicated by returning -EINPROGRESS,
* at which time the urb will normally not have been unlinked.
- * The completion function will see urb->status -ECONNRESET. Failure
+ * The completion function will see urb->status == -ECONNRESET. Failure
* is indicated by any other return value.
+ *
+ * Unlinking and Endpoint Queues:
+ *
+ * Host Controller Driver (HCDs) place all the URBs for a particular
+ * endpoint in a queue. Normally the queue advances as the controller
+ * hardware processes each request. But when an URB terminates with any
+ * fault (such as an error, or being unlinked) its queue stops, at least
+ * until that URB's completion routine returns. It is guaranteed that
+ * the queue will not restart until all its unlinked URBs have been fully
+ * retired, with their completion routines run, even if that's not until
+ * some time after the original completion handler returns.
+ *
+ * This means that USB device drivers can safely build deep queues for
+ * large or complex transfers, and clean them up reliably after any sort
+ * of aborted transfer by unlinking all pending URBs at the first fault.
+ *
+ * Note that an URB terminating early because a short packet was received
+ * will count as an error if and only if the URB_SHORT_NOT_OK flag is set.
+ * Also, that all unlinks performed in any URB completion handler must
+ * be asynchronous.
+ *
+ * Queues for isochronous endpoints are treated differently, because they
+ * advance at fixed rates. Such queues do not stop when an URB is unlinked.
+ * An unlinked URB may leave a gap in the stream of packets. It is undefined
+ * whether such gaps can be filled in.
+ *
+ * When control URBs terminates with an error, it is likely that the
+ * status stage of the transfer will not take place, even if it is merely
+ * a soft error resulting from a short-packet with URB_SHORT_NOT_OK set.
*/
int usb_unlink_urb(struct urb *urb)
{
--- diff/drivers/usb/core/usb.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/usb/core/usb.c 2004-03-16 11:27:48.268849680 +0000
@@ -189,7 +189,7 @@ void usb_deregister(struct usb_driver *d
}
/**
- * usb_ifnum_to_if - get the interface object with a given interface number (usbcore-internal)
+ * usb_ifnum_to_if - get the interface object with a given interface number
* @dev: the device whose current configuration is considered
* @ifnum: the desired interface
*
@@ -220,6 +220,33 @@ struct usb_interface *usb_ifnum_to_if(st
}
/**
+ * usb_altnum_to_altsetting - get the altsetting structure with a given
+ * alternate setting number.
+ * @intf: the interface containing the altsetting in question
+ * @altnum: the desired alternate setting number
+ *
+ * This searches the altsetting array of the specified interface for
+ * an entry with the correct bAlternateSetting value and returns a pointer
+ * to that entry, or null.
+ *
+ * Note that altsettings need not be stored sequentially by number, so
+ * it would be incorrect to assume that the first altsetting entry in
+ * the array corresponds to altsetting zero. This routine helps device
+ * drivers avoid such mistakes.
+ */
+struct usb_host_interface *usb_altnum_to_altsetting(struct usb_interface *intf,
+ unsigned int altnum)
+{
+ int i;
+
+ for (i = 0; i < intf->num_altsetting; i++) {
+ if (intf->altsetting[i].desc.bAlternateSetting == altnum)
+ return &intf->altsetting[i];
+ }
+ return NULL;
+}
+
+/**
* usb_epnum_to_ep_desc - get the endpoint object with a given endpoint number
* @dev: the device whose current configuration+altsettings is considered
* @epnum: the desired endpoint, masked with USB_DIR_IN as appropriate.
@@ -247,7 +274,7 @@ usb_epnum_to_ep_desc(struct usb_device *
/* only endpoints in current altsetting are active */
intf = config->interface[i];
- alt = intf->altsetting + intf->act_altsetting;
+ alt = intf->cur_altsetting;
for (k = 0; k < alt->desc.bNumEndpoints; k++)
if (epnum == alt->endpoint[k].desc.bEndpointAddress)
@@ -421,7 +448,7 @@ usb_match_id(struct usb_interface *inter
if (id == NULL)
return NULL;
- intf = &interface->altsetting [interface->act_altsetting];
+ intf = interface->cur_altsetting;
dev = interface_to_usbdev(interface);
/* It is important to check that id->driver_info is nonzero,
@@ -624,7 +651,7 @@ static int usb_hotplug (struct device *d
scratch += length;
if (usb_dev->descriptor.bDeviceClass == 0) {
- int alt = intf->act_altsetting;
+ struct usb_host_interface *alt = intf->cur_altsetting;
/* 2.4 only exposed interface zero. in 2.5, hotplug
* agents are called for all interfaces, and can use
@@ -633,9 +660,9 @@ static int usb_hotplug (struct device *d
envp [i++] = scratch;
length += snprintf (scratch, buffer_size - length,
"INTERFACE=%d/%d/%d",
- intf->altsetting[alt].desc.bInterfaceClass,
- intf->altsetting[alt].desc.bInterfaceSubClass,
- intf->altsetting[alt].desc.bInterfaceProtocol);
+ alt->desc.bInterfaceClass,
+ alt->desc.bInterfaceSubClass,
+ alt->desc.bInterfaceProtocol);
if ((buffer_size - length <= 0) || (i >= num_envp))
return -ENOMEM;
++length;
@@ -1598,6 +1625,7 @@ EXPORT_SYMBOL(usb_driver_release_interfa
EXPORT_SYMBOL(usb_match_id);
EXPORT_SYMBOL(usb_find_interface);
EXPORT_SYMBOL(usb_ifnum_to_if);
+EXPORT_SYMBOL(usb_altnum_to_altsetting);
EXPORT_SYMBOL(usb_reset_device);
EXPORT_SYMBOL(usb_disconnect);
--- diff/drivers/usb/gadget/Kconfig 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/gadget/Kconfig 2004-03-16 11:27:48.272849072 +0000
@@ -3,6 +3,15 @@
# (a) a peripheral controller, and
# (b) the gadget driver using it.
#
+# NOTE: Gadget support ** DOES NOT ** depend on host-side CONFIG_USB !!
+#
+# - Host systems (like PCs) need CONFIG_USB (with "A" jacks).
+# - Peripherals (like PDAs) need CONFIG_USB_GADGET (with "B" jacks).
+# - Some systems have both kinds of of controller.
+#
+# With help from a special transceiver and a "Mini-AB" jack, systems with
+# both kinds of controller can also support "USB On-the-Go" (CONFIG_USB_OTG).
+#
menu "USB Gadget Support"
config USB_GADGET
@@ -11,7 +20,7 @@ config USB_GADGET
USB is a master/slave protocol, organized with one master
host (such as a PC) controlling up to 127 peripheral devices.
The USB hardware is asymmetric, which makes it easier to set up:
- you can't connect two "to-the-host" connectors to each other.
+ you can't connect a "to-the-host" connector to a peripheral.
Linux can run in the host, or in the peripheral. In both cases
you need a low level bus controller driver, and some software
@@ -43,6 +52,7 @@ choice
config USB_GADGET_NET2280
boolean "NetChip 2280"
depends on PCI
+ select USB_GADGET_DUALSPEED
help
NetChip 2280 is a PCI based USB peripheral controller which
supports both full and high speed USB 2.0 data transfers.
@@ -126,6 +136,13 @@ config USB_SA1100
endchoice
+config USB_GADGET_DUALSPEED
+ bool
+ depends on USB_GADGET
+ default n
+ help
+ Means that gadget drivers should include extra descriptors
+ and code to handle dual-speed controllers.
#
# USB Gadget Drivers
--- diff/drivers/usb/gadget/Makefile 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/gadget/Makefile 2004-03-16 11:27:48.273848920 +0000
@@ -8,8 +8,8 @@ obj-$(CONFIG_USB_GOKU) += goku_udc.o
#
# USB gadget drivers
#
-g_zero-objs := zero.o usbstring.o
-g_ether-objs := ether.o usbstring.o
+g_zero-objs := zero.o usbstring.o config.o
+g_ether-objs := ether.o usbstring.o config.o
g_serial-objs := serial.o usbstring.o
gadgetfs-objs := inode.o usbstring.o
g_file_storage-objs := file_storage.o usbstring.o
--- diff/drivers/usb/gadget/ether.c 2004-02-18 08:54:11.000000000 +0000
+++ source/drivers/usb/gadget/ether.c 2004-03-16 11:27:48.271849224 +0000
@@ -124,7 +124,6 @@ struct eth_dev {
* DRIVER_VERSION_NUM ... alerts the host side driver to differences
* EP_*_NAME ... which endpoints do we use for which purpose?
* EP_*_NUM ... numbers for them (often limited by hardware)
- * HIGHSPEED ... define if ep0 and descriptors need high speed support
* WAKEUP ... if hardware supports remote wakeup AND we will issue the
* usb_gadget_wakeup() call to initiate it, USB_CONFIG_ATT_WAKEUP
*
@@ -162,7 +161,6 @@ static const char EP_IN_NAME [] = "ep-b"
#define EP_IN_NUM 2
static const char EP_STATUS_NAME [] = "ep-f";
#define EP_STATUS_NUM 3
-#define HIGHSPEED
/* supports remote wakeup, but this driver doesn't */
extern int net2280_set_fifo_mode (struct usb_gadget *gadget, int mode);
@@ -311,7 +309,7 @@ static const char EP_IN_NAME[] = "ep2in-
#define DEFAULT_QLEN 2 /* double buffering by default */
#endif
-#ifdef HIGHSPEED
+#ifdef CONFIG_USB_GADGET_DUALSPEED
static unsigned qmult = 5;
module_param (qmult, uint, S_IRUGO|S_IWUSR);
@@ -324,7 +322,7 @@ module_param (qmult, uint, S_IRUGO|S_IWU
/* also defer IRQs on highspeed TX */
#define TX_DELAY DEFAULT_QLEN
-#else /* !HIGHSPEED ... full speed: */
+#else /* full speed (low speed doesn't do bulk) */
#define qlen(gadget) DEFAULT_QLEN
#endif
@@ -607,7 +605,26 @@ fs_sink_desc = {
.wMaxPacketSize = __constant_cpu_to_le16 (64),
};
-#ifdef HIGHSPEED
+static const struct usb_descriptor_header *fs_function [] = {
+#ifdef DEV_CONFIG_CDC
+ /* "cdc" mode descriptors */
+ (struct usb_descriptor_header *) &control_intf,
+ (struct usb_descriptor_header *) &header_desc,
+ (struct usb_descriptor_header *) &union_desc,
+ (struct usb_descriptor_header *) ðer_desc,
+#ifdef EP_STATUS_NUM
+ (struct usb_descriptor_header *) &fs_status_desc,
+#endif
+ (struct usb_descriptor_header *) &data_nop_intf,
+#endif /* DEV_CONFIG_CDC */
+ /* minimalist core */
+ (struct usb_descriptor_header *) &data_intf,
+ (struct usb_descriptor_header *) &fs_source_desc,
+ (struct usb_descriptor_header *) &fs_sink_desc,
+ 0,
+};
+
+#ifdef CONFIG_USB_GADGET_DUALSPEED
/*
* usb 2.0 devices need to expose both high speed and full speed
@@ -660,6 +677,25 @@ dev_qualifier = {
.bNumConfigurations = 1,
};
+static const struct usb_descriptor_header *hs_function [] = {
+#ifdef DEV_CONFIG_CDC
+ /* "cdc" mode descriptors */
+ (struct usb_descriptor_header *) &control_intf,
+ (struct usb_descriptor_header *) &header_desc,
+ (struct usb_descriptor_header *) &union_desc,
+ (struct usb_descriptor_header *) ðer_desc,
+#ifdef EP_STATUS_NUM
+ (struct usb_descriptor_header *) &hs_status_desc,
+#endif
+ (struct usb_descriptor_header *) &data_nop_intf,
+#endif /* DEV_CONFIG_CDC */
+ /* minimalist core */
+ (struct usb_descriptor_header *) &data_intf,
+ (struct usb_descriptor_header *) &hs_source_desc,
+ (struct usb_descriptor_header *) &hs_sink_desc,
+ 0,
+};
+
/* maxpacket and other transfer characteristics vary by speed. */
#define ep_desc(g,hs,fs) (((g)->speed==USB_SPEED_HIGH)?(hs):(fs))
@@ -669,7 +705,7 @@ dev_qualifier = {
/* if there's no high speed support, maxpacket doesn't change. */
#define ep_desc(g,hs,fs) fs
-#endif /* !HIGHSPEED */
+#endif /* !CONFIG_USB_GADGET_DUALSPEED */
/*-------------------------------------------------------------------------*/
@@ -704,86 +740,25 @@ static struct usb_gadget_strings stringt
static int
config_buf (enum usb_device_speed speed, u8 *buf, u8 type, unsigned index)
{
- const unsigned config_len = USB_DT_CONFIG_SIZE
-#ifdef DEV_CONFIG_CDC
- + 2 * USB_DT_INTERFACE_SIZE
- + sizeof header_desc
- + sizeof union_desc
- + sizeof ether_desc
-#ifdef EP_STATUS_NUM
- + USB_DT_ENDPOINT_SIZE
-#endif
-#endif /* DEV_CONFIG_CDC */
- + USB_DT_INTERFACE_SIZE
- + 2 * USB_DT_ENDPOINT_SIZE;
+ int len;
+ const struct usb_descriptor_header **function = fs_function;
+#ifdef CONFIG_USB_GADGET_DUALSPEED
+ int hs = (speed == USB_SPEED_HIGH);
-#ifdef HIGHSPEED
- int hs;
-#endif
- /* a single configuration must always be index 0 */
- if (index > 0)
- return -EINVAL;
- if (config_len > USB_BUFSIZ)
- return -EDOM;
-
- /* config (or other speed config) */
- memcpy (buf, ð_config, USB_DT_CONFIG_SIZE);
- buf [1] = type;
- ((struct usb_config_descriptor *) buf)->wTotalLength
- = __constant_cpu_to_le16 (config_len);
- buf += USB_DT_CONFIG_SIZE;
-#ifdef HIGHSPEED
- hs = (speed == USB_SPEED_HIGH);
if (type == USB_DT_OTHER_SPEED_CONFIG)
hs = !hs;
-#endif
-
-#ifdef DEV_CONFIG_CDC
- /* control interface, class descriptors, optional status endpoint */
- memcpy (buf, &control_intf, USB_DT_INTERFACE_SIZE);
- buf += USB_DT_INTERFACE_SIZE;
-
- memcpy (buf, &header_desc, sizeof header_desc);
- buf += sizeof header_desc;
- memcpy (buf, &union_desc, sizeof union_desc);
- buf += sizeof union_desc;
- memcpy (buf, ðer_desc, sizeof ether_desc);
- buf += sizeof ether_desc;
-
-#ifdef EP_STATUS_NUM
-#ifdef HIGHSPEED
if (hs)
- memcpy (buf, &hs_status_desc, USB_DT_ENDPOINT_SIZE);
- else
-#endif /* HIGHSPEED */
- memcpy (buf, &fs_status_desc, USB_DT_ENDPOINT_SIZE);
- buf += USB_DT_ENDPOINT_SIZE;
-#endif /* EP_STATUS_NUM */
-
- /* default data altsetting has no endpoints */
- memcpy (buf, &data_nop_intf, USB_DT_INTERFACE_SIZE);
- buf += USB_DT_INTERFACE_SIZE;
-#endif /* DEV_CONFIG_CDC */
-
- /* the "real" data interface has two endpoints */
- memcpy (buf, &data_intf, USB_DT_INTERFACE_SIZE);
- buf += USB_DT_INTERFACE_SIZE;
-#ifdef HIGHSPEED
- if (hs) {
- memcpy (buf, &hs_source_desc, USB_DT_ENDPOINT_SIZE);
- buf += USB_DT_ENDPOINT_SIZE;
- memcpy (buf, &hs_sink_desc, USB_DT_ENDPOINT_SIZE);
- buf += USB_DT_ENDPOINT_SIZE;
- } else
+ function = hs_function;
#endif
- {
- memcpy (buf, &fs_source_desc, USB_DT_ENDPOINT_SIZE);
- buf += USB_DT_ENDPOINT_SIZE;
- memcpy (buf, &fs_sink_desc, USB_DT_ENDPOINT_SIZE);
- buf += USB_DT_ENDPOINT_SIZE;
- }
- return config_len;
+ /* a single configuration must always be index 0 */
+ if (index > 0)
+ return -EINVAL;
+ len = usb_gadget_config_buf (ð_config, buf, USB_BUFSIZ, function);
+ if (len < 0)
+ return len;
+ ((struct usb_config_descriptor *) buf)->bDescriptorType = type;
+ return len;
}
/*-------------------------------------------------------------------------*/
@@ -992,7 +967,7 @@ eth_set_config (struct eth_dev *dev, uns
switch (gadget->speed) {
case USB_SPEED_FULL: speed = "full"; break;
-#ifdef HIGHSPEED
+#ifdef CONFIG_USB_GADGET_DUALSPEED
case USB_SPEED_HIGH: speed = "high"; break;
#endif
default: speed = "?"; break;
@@ -1163,15 +1138,19 @@ eth_setup (struct usb_gadget *gadget, co
value = min (ctrl->wLength, (u16) sizeof device_desc);
memcpy (req->buf, &device_desc, value);
break;
-#ifdef HIGHSPEED
+#ifdef CONFIG_USB_GADGET_DUALSPEED
case USB_DT_DEVICE_QUALIFIER:
+ if (!gadget->is_dualspeed)
+ break;
value = min (ctrl->wLength, (u16) sizeof dev_qualifier);
memcpy (req->buf, &dev_qualifier, value);
break;
case USB_DT_OTHER_SPEED_CONFIG:
+ if (!gadget->is_dualspeed)
+ break;
// FALLTHROUGH
-#endif /* HIGHSPEED */
+#endif /* CONFIG_USB_GADGET_DUALSPEED */
case USB_DT_CONFIG:
value = config_buf (gadget->speed, req->buf,
ctrl->wValue >> 8,
@@ -1675,7 +1654,7 @@ static int eth_start_xmit (struct sk_buf
#endif
req->length = length;
-#ifdef HIGHSPEED
+#ifdef CONFIG_USB_GADGET_DUALSPEED
/* throttle highspeed IRQ rate back slightly */
req->no_interrupt = (dev->gadget->speed == USB_SPEED_HIGH)
? ((atomic_read (&dev->tx_qlen) % TX_DELAY) != 0)
@@ -1798,7 +1777,7 @@ eth_bind (struct usb_gadget *gadget)
#endif
device_desc.bMaxPacketSize0 = gadget->ep0->maxpacket;
-#ifdef HIGHSPEED
+#ifdef CONFIG_USB_GADGET_DUALSPEED
/* assumes ep0 uses the same value for both speeds ... */
dev_qualifier.bMaxPacketSize0 = device_desc.bMaxPacketSize0;
#endif
@@ -1894,7 +1873,7 @@ fail:
/*-------------------------------------------------------------------------*/
static struct usb_gadget_driver eth_driver = {
-#ifdef HIGHSPEED
+#ifdef CONFIG_USB_GADGET_DUALSPEED
.speed = USB_SPEED_HIGH,
#else
.speed = USB_SPEED_FULL,
--- diff/drivers/usb/gadget/net2280.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/usb/gadget/net2280.c 2004-03-16 11:27:48.275848616 +0000
@@ -2663,7 +2663,7 @@ static void gadget_release (struct devic
/* tear down the binding between this driver and the pci device */
-static void __exit net2280_remove (struct pci_dev *pdev)
+static void net2280_remove (struct pci_dev *pdev)
{
struct net2280 *dev = pci_get_drvdata (pdev);
@@ -2736,6 +2736,7 @@ static int net2280_probe (struct pci_dev
spin_lock_init (&dev->lock);
dev->pdev = pdev;
dev->gadget.ops = &net2280_ops;
+ dev->gadget.is_dualspeed = 1;
/* the "gadget" abstracts/virtualizes the controller */
strcpy (dev->gadget.dev.bus_id, "gadget");
@@ -2884,7 +2885,7 @@ static struct pci_driver net2280_pci_dri
.id_table = pci_ids,
.probe = net2280_probe,
- .remove = __exit_p(net2280_remove),
+ .remove = net2280_remove,
/* FIXME add power management support */
};
--- diff/drivers/usb/gadget/usbstring.c 2003-06-09 13:18:19.000000000 +0000
+++ source/drivers/usb/gadget/usbstring.c 2004-03-16 11:27:48.276848464 +0000
@@ -16,24 +16,89 @@
#include <linux/usb_ch9.h>
#include <linux/usb_gadget.h>
+#include <asm/unaligned.h>
+
+
+static int utf8_to_utf16le(const char *s, u16 *cp, unsigned len)
+{
+ int count = 0;
+ u8 c;
+ u16 uchar;
+
+ /* this insists on correct encodings, though not minimal ones.
+ * BUT it currently rejects legit 4-byte UTF-8 code points,
+ * which need surrogate pairs. (Unicode 3.1 can use them.)
+ */
+ while (len != 0 && (c = (u8) *s++) != 0) {
+ if (unlikely(c & 0x80)) {
+ // 2-byte sequence:
+ // 00000yyyyyxxxxxx = 110yyyyy 10xxxxxx
+ if ((c & 0xe0) == 0xc0) {
+ uchar = (c & 0x1f) << 6;
+
+ c = (u8) *s++;
+ if ((c & 0xc0) != 0xc0)
+ goto fail;
+ c &= 0x3f;
+ uchar |= c;
+
+ // 3-byte sequence (most CJKV characters):
+ // zzzzyyyyyyxxxxxx = 1110zzzz 10yyyyyy 10xxxxxx
+ } else if ((c & 0xf0) == 0xe0) {
+ uchar = (c & 0x0f) << 12;
+
+ c = (u8) *s++;
+ if ((c & 0xc0) != 0xc0)
+ goto fail;
+ c &= 0x3f;
+ uchar |= c << 6;
+
+ c = (u8) *s++;
+ if ((c & 0xc0) != 0xc0)
+ goto fail;
+ c &= 0x3f;
+ uchar |= c;
+
+ /* no bogus surrogates */
+ if (0xd800 <= uchar && uchar <= 0xdfff)
+ goto fail;
+
+ // 4-byte sequence (surrogate pairs, currently rare):
+ // 11101110wwwwzzzzyy + 110111yyyyxxxxxx
+ // = 11110uuu 10uuzzzz 10yyyyyy 10xxxxxx
+ // (uuuuu = wwww + 1)
+ // FIXME accept the surrogate code points (only)
+
+ } else
+ goto fail;
+ } else
+ uchar = c;
+ put_unaligned (cpu_to_le16 (uchar), cp++);
+ count++;
+ len--;
+ }
+ return count;
+fail:
+ return -1;
+}
+
/**
* usb_gadget_get_string - fill out a string descriptor
- * @table: of c strings using iso latin/1 characters
+ * @table: of c strings encoded using UTF-8
* @id: string id, from low byte of wValue in get string descriptor
* @buf: at least 256 bytes
*
- * Finds the iso latin/1 string matching the ID, and converts it into a
+ * Finds the UTF-8 string matching the ID, and converts it into a
* string descriptor in utf16-le.
* Returns length of descriptor (always even) or negative errno
*
- * If your driver needs stings in multiple languages, you'll need to
- * to use some alternate solution for languages where the ISO 8859/1
- * (latin/1) character set can't be used. For example, they can't be
- * used with Chinese (Big5, GB2312, etc), Japanese, Korean, or many other
- * languages. You'd likely "switch (wIndex) { ... }" in your ep0
- * string descriptor logic, using this routine in cases where "western
- * european" characters suffice for the strings being returned.
+ * If your driver needs stings in multiple languages, you'll probably
+ * "switch (wIndex) { ... }" in your ep0 string descriptor logic,
+ * using this routine after choosing which set of UTF-8 strings to use.
+ * Note that US-ASCII is a strict subset of UTF-8; any string bytes with
+ * the eighth bit set will be multibyte UTF-8 characters, not ISO-8859/1
+ * characters (which are also widely used in C strings).
*/
int
usb_gadget_get_string (struct usb_gadget_strings *table, int id, u8 *buf)
@@ -59,13 +124,12 @@ usb_gadget_get_string (struct usb_gadget
/* string descriptors have length, tag, then UTF16-LE text */
len = min ((size_t) 126, strlen (s->s));
+ memset (buf + 2, 0, 2 * len); /* zero all the bytes */
+ len = utf8_to_utf16le(s->s, (u16 *)&buf[2], len);
+ if (len < 0)
+ return -EINVAL;
buf [0] = (len + 1) * 2;
buf [1] = USB_DT_STRING;
- memset (buf + 2, 0, 2 * len); /* zero all the high bytes */
- while (len) {
- buf [2 * len] = s->s [len - 1];
- len--;
- }
return buf [0];
}
--- diff/drivers/usb/gadget/zero.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/gadget/zero.c 2004-03-16 11:27:48.278848160 +0000
@@ -1,7 +1,7 @@
/*
* zero.c -- Gadget Zero, for USB development
*
- * Copyright (C) 2003 David Brownell
+ * Copyright (C) 2003-2004 David Brownell
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
@@ -103,6 +103,11 @@ static const char loopback [] = "loop in
/*-------------------------------------------------------------------------*/
/*
+ * driver assumes self-powered hardware, and
+ * has no way for users to trigger remote wakeup.
+ */
+
+/*
* hardware-specific configuration, controlled by which device
* controller driver was configured.
*
@@ -110,11 +115,6 @@ static const char loopback [] = "loop in
* DRIVER_VERSION_NUM ... alerts the host side driver to differences
* EP_*_NAME ... which endpoints do we use for which purpose?
* EP_*_NUM ... numbers for them (often limited by hardware)
- * HIGHSPEED ... define if ep0 and descriptors need high speed support
- * MAX_USB_POWER ... define if we use other than 100 mA bus current
- * SELFPOWER ... if we can run on bus power, zero
- * WAKEUP ... if hardware supports remote wakeup AND we will issue the
- * usb_gadget_wakeup() call to initiate it, USB_CONFIG_ATT_WAKEUP
*
* add other defines for other portability issues, like hardware that
* for some reason doesn't handle full speed bulk maxpacket of 64.
@@ -138,9 +138,6 @@ static const char EP_OUT_NAME [] = "ep-a
#define EP_OUT_NUM 2
static const char EP_IN_NAME [] = "ep-b";
#define EP_IN_NUM 2
-#define HIGHSPEED
-/* specific hardware configs could be bus-powered */
-/* supports remote wakeup, but this driver doesn't */
#endif
/*
@@ -161,8 +158,6 @@ static const char EP_OUT_NAME [] = "ep12
#define EP_OUT_NUM 12
static const char EP_IN_NAME [] = "ep11in-bulk";
#define EP_IN_NUM 11
-/* doesn't support bus-powered operation */
-/* supports remote wakeup, but this driver doesn't */
#endif
/*
@@ -183,8 +178,6 @@ static const char EP_OUT_NAME [] = "ep1o
#define EP_OUT_NUM 1
static const char EP_IN_NAME [] = "ep2in-bulk";
#define EP_IN_NUM 2
-/* doesn't support bus-powered operation */
-/* doesn't support remote wakeup? */
#endif
/*
@@ -199,7 +192,6 @@ static const char EP_OUT_NAME [] = "ep1-
#define EP_OUT_NUM 1
static const char EP_IN_NAME [] = "ep2-bulk";
#define EP_IN_NUM 2
-/* doesn't support remote wakeup */
#endif
/*-------------------------------------------------------------------------*/
@@ -208,30 +200,6 @@ static const char EP_IN_NAME [] = "ep2-b
# error Configure some USB peripheral controller driver!
#endif
-/* power usage is config specific.
- * hardware that supports remote wakeup defaults to disabling it.
- */
-
-#ifndef SELFPOWER
-/* default: say we're self-powered */
-#define SELFPOWER USB_CONFIG_ATT_SELFPOWER
-/* else:
- * - SELFPOWER value must be zero
- * - MAX_USB_POWER may be nonzero.
- */
-#endif
-
-#ifndef MAX_USB_POWER
-/* any hub supports this steady state bus power consumption */
-#define MAX_USB_POWER 100 /* mA */
-#endif
-
-#ifndef WAKEUP
-/* default: this driver won't do remote wakeup */
-#define WAKEUP 0
-/* else value must be USB_CONFIG_ATT_WAKEUP */
-#endif
-
/*-------------------------------------------------------------------------*/
/* big enough to hold our biggest descriptor */
@@ -290,8 +258,8 @@ module_param (pattern, uint, S_IRUGO|S_I
/*
* Normally the "loopback" configuration is second (index 1) so
* it's not the default. Here's where to change that order, to
- * work better with hosts (like Linux ... for now!) where config
- * changes are problematic.
+ * work better with hosts where config changes are problematic.
+ * Or controllers (like superh) that only support one config.
*/
static int loopdefault = 0;
@@ -301,7 +269,7 @@ module_param (loopdefault, bool, S_IRUGO
/* Thanks to NetChip Technologies for donating this product ID.
*
- * DO NOT REUSE THESE IDs with any other driver!! Ever!!
+ * DO NOT REUSE THESE IDs with a protocol-incompatible driver!! Ever!!
* Instead: allocate your own, using normal USB-IF procedures.
*/
#define DRIVER_VENDOR_NUM 0x0525 /* NetChip */
@@ -353,8 +321,8 @@ source_sink_config = {
.bNumInterfaces = 1,
.bConfigurationValue = CONFIG_SOURCE_SINK,
.iConfiguration = STRING_SOURCE_SINK,
- .bmAttributes = USB_CONFIG_ATT_ONE | SELFPOWER | WAKEUP,
- .bMaxPower = (MAX_USB_POWER + 1) / 2,
+ .bmAttributes = USB_CONFIG_ATT_ONE | USB_CONFIG_ATT_SELFPOWER,
+ .bMaxPower = 1, /* self-powered */
};
static const struct usb_config_descriptor
@@ -366,8 +334,8 @@ loopback_config = {
.bNumInterfaces = 1,
.bConfigurationValue = CONFIG_LOOPBACK,
.iConfiguration = STRING_LOOPBACK,
- .bmAttributes = USB_CONFIG_ATT_ONE | SELFPOWER | WAKEUP,
- .bMaxPower = (MAX_USB_POWER + 1) / 2,
+ .bmAttributes = USB_CONFIG_ATT_ONE | USB_CONFIG_ATT_SELFPOWER,
+ .bMaxPower = 1, /* self-powered */
};
/* one interface in each configuration */
@@ -414,7 +382,21 @@ fs_sink_desc = {
.wMaxPacketSize = __constant_cpu_to_le16 (64),
};
-#ifdef HIGHSPEED
+static const struct usb_descriptor_header *fs_source_sink_function [] = {
+ (struct usb_descriptor_header *) &source_sink_intf,
+ (struct usb_descriptor_header *) &fs_sink_desc,
+ (struct usb_descriptor_header *) &fs_source_desc,
+ 0,
+};
+
+static const struct usb_descriptor_header *fs_loopback_function [] = {
+ (struct usb_descriptor_header *) &loopback_intf,
+ (struct usb_descriptor_header *) &fs_sink_desc,
+ (struct usb_descriptor_header *) &fs_source_desc,
+ 0,
+};
+
+#ifdef CONFIG_USB_GADGET_DUALSPEED
/*
* usb 2.0 devices need to expose both high speed and full speed
@@ -425,22 +407,20 @@ fs_sink_desc = {
* for the config descriptor.
*/
-static const struct usb_endpoint_descriptor
+static struct usb_endpoint_descriptor
hs_source_desc = {
.bLength = USB_DT_ENDPOINT_SIZE,
.bDescriptorType = USB_DT_ENDPOINT,
- .bEndpointAddress = EP_IN_NUM | USB_DIR_IN,
.bmAttributes = USB_ENDPOINT_XFER_BULK,
.wMaxPacketSize = __constant_cpu_to_le16 (512),
};
-static const struct usb_endpoint_descriptor
+static struct usb_endpoint_descriptor
hs_sink_desc = {
.bLength = USB_DT_ENDPOINT_SIZE,
.bDescriptorType = USB_DT_ENDPOINT,
- .bEndpointAddress = EP_OUT_NUM,
.bmAttributes = USB_ENDPOINT_XFER_BULK,
.wMaxPacketSize = __constant_cpu_to_le16 (512),
};
@@ -456,6 +436,20 @@ dev_qualifier = {
.bNumConfigurations = 2,
};
+static const struct usb_descriptor_header *hs_source_sink_function [] = {
+ (struct usb_descriptor_header *) &source_sink_intf,
+ (struct usb_descriptor_header *) &hs_source_desc,
+ (struct usb_descriptor_header *) &hs_sink_desc,
+ 0,
+};
+
+static const struct usb_descriptor_header *hs_loopback_function [] = {
+ (struct usb_descriptor_header *) &loopback_intf,
+ (struct usb_descriptor_header *) &hs_source_desc,
+ (struct usb_descriptor_header *) &hs_sink_desc,
+ 0,
+};
+
/* maxpacket and other transfer characteristics vary by speed. */
#define ep_desc(g,hs,fs) (((g)->speed==USB_SPEED_HIGH)?(hs):(fs))
@@ -464,13 +458,14 @@ dev_qualifier = {
/* if there's no high speed support, maxpacket doesn't change. */
#define ep_desc(g,hs,fs) fs
-#endif /* !HIGHSPEED */
+#endif /* !CONFIG_USB_GADGET_DUALSPEED */
+static char manufacturer [40];
static char serial [40];
/* static strings, in iso 8859/1 */
static struct usb_string strings [] = {
- { STRING_MANUFACTURER, UTS_SYSNAME " " UTS_RELEASE " with " CHIP, },
+ { STRING_MANUFACTURER, manufacturer, },
{ STRING_PRODUCT, longname, },
{ STRING_SERIAL, serial, },
{ STRING_LOOPBACK, loopback, },
@@ -502,60 +497,42 @@ static struct usb_gadget_strings stringt
* device?)
*/
static int
-config_buf (enum usb_device_speed speed,
+config_buf (struct usb_gadget *gadget,
u8 *buf, u8 type, unsigned index)
{
- int is_source_sink;
- const unsigned config_len = USB_DT_CONFIG_SIZE
- + USB_DT_INTERFACE_SIZE
- + 2 * USB_DT_ENDPOINT_SIZE;
-#ifdef HIGHSPEED
- int hs;
+ int is_source_sink;
+ int len;
+ const struct usb_descriptor_header **function;
+#ifdef CONFIG_USB_GADGET_DUALSPEED
+ int hs = (gadget->speed == USB_SPEED_HIGH);
#endif
+
/* two configurations will always be index 0 and index 1 */
if (index > 1)
return -EINVAL;
- if (config_len > USB_BUFSIZ)
- return -EDOM;
is_source_sink = loopdefault ? (index == 1) : (index == 0);
- /* config (or other speed config) */
- if (is_source_sink)
- memcpy (buf, &source_sink_config, USB_DT_CONFIG_SIZE);
- else
- memcpy (buf, &loopback_config, USB_DT_CONFIG_SIZE);
- buf [1] = type;
- ((struct usb_config_descriptor *) buf)->wTotalLength
- = __constant_cpu_to_le16 (config_len);
- buf += USB_DT_CONFIG_SIZE;
-
- /* one interface */
- if (is_source_sink)
- memcpy (buf, &source_sink_intf, USB_DT_INTERFACE_SIZE);
- else
- memcpy (buf, &loopback_intf, USB_DT_INTERFACE_SIZE);
- buf += USB_DT_INTERFACE_SIZE;
-
- /* the endpoints in that interface (at that speed) */
-#ifdef HIGHSPEED
- hs = (speed == USB_SPEED_HIGH);
+#ifdef CONFIG_USB_GADGET_DUALSPEED
if (type == USB_DT_OTHER_SPEED_CONFIG)
hs = !hs;
- if (hs) {
- memcpy (buf, &hs_source_desc, USB_DT_ENDPOINT_SIZE);
- buf += USB_DT_ENDPOINT_SIZE;
- memcpy (buf, &hs_sink_desc, USB_DT_ENDPOINT_SIZE);
- buf += USB_DT_ENDPOINT_SIZE;
- } else
+ if (hs)
+ function = is_source_sink
+ ? hs_source_sink_function
+ : hs_loopback_function;
+ else
#endif
- {
- memcpy (buf, &fs_source_desc, USB_DT_ENDPOINT_SIZE);
- buf += USB_DT_ENDPOINT_SIZE;
- memcpy (buf, &fs_sink_desc, USB_DT_ENDPOINT_SIZE);
- buf += USB_DT_ENDPOINT_SIZE;
- }
-
- return config_len;
+ function = is_source_sink
+ ? fs_source_sink_function
+ : fs_loopback_function;
+
+ len = usb_gadget_config_buf (is_source_sink
+ ? &source_sink_config
+ : &loopback_config,
+ buf, USB_BUFSIZ, function);
+ if (len < 0)
+ return len;
+ ((struct usb_config_descriptor *) buf)->bDescriptorType = type;
+ return len;
}
/*-------------------------------------------------------------------------*/
@@ -1019,17 +996,21 @@ zero_setup (struct usb_gadget *gadget, c
value = min (ctrl->wLength, (u16) sizeof device_desc);
memcpy (req->buf, &device_desc, value);
break;
-#ifdef HIGHSPEED
+#ifdef CONFIG_USB_GADGET_DUALSPEED
case USB_DT_DEVICE_QUALIFIER:
+ if (!gadget->is_dualspeed)
+ break;
value = min (ctrl->wLength, (u16) sizeof dev_qualifier);
memcpy (req->buf, &dev_qualifier, value);
break;
case USB_DT_OTHER_SPEED_CONFIG:
+ if (!gadget->is_dualspeed)
+ break;
// FALLTHROUGH
-#endif /* HIGHSPEED */
+#endif /* CONFIG_USB_GADGET_DUALSPEED */
case USB_DT_CONFIG:
- value = config_buf (gadget->speed, req->buf,
+ value = config_buf (gadget, req->buf,
ctrl->wValue >> 8,
ctrl->wValue & 0xff);
if (value >= 0)
@@ -1212,14 +1193,26 @@ zero_bind (struct usb_gadget *gadget)
dev->req->complete = zero_setup_complete;
device_desc.bMaxPacketSize0 = gadget->ep0->maxpacket;
-#ifdef HIGHSPEED
+
+#ifdef CONFIG_USB_GADGET_DUALSPEED
/* assume ep0 uses the same value for both speeds ... */
dev_qualifier.bMaxPacketSize0 = device_desc.bMaxPacketSize0;
+
+ /* and that all endpoints are dual-speed */
+ hs_source_desc.bEndpointAddress = fs_source_desc.bEndpointAddress;
+ hs_sink_desc.bEndpointAddress = fs_sink_desc.bEndpointAddress;
#endif
gadget->ep0->driver_data = dev;
INFO (dev, "%s, version: " DRIVER_VERSION "\n", longname);
+ INFO (dev, "using %s, OUT %s IN %s\n", gadget->name,
+ EP_OUT_NAME, EP_IN_NAME);
+
+ snprintf (manufacturer, sizeof manufacturer,
+ UTS_SYSNAME " " UTS_RELEASE " with %s",
+ gadget->name);
+
return 0;
enomem:
@@ -1230,7 +1223,7 @@ enomem:
/*-------------------------------------------------------------------------*/
static struct usb_gadget_driver zero_driver = {
-#ifdef HIGHSPEED
+#ifdef CONFIG_USB_GADGET_DUALSPEED
.speed = USB_SPEED_HIGH,
#else
.speed = USB_SPEED_FULL,
--- diff/drivers/usb/host/Kconfig 2003-09-30 14:46:17.000000000 +0000
+++ source/drivers/usb/host/Kconfig 2004-03-16 11:27:48.284847248 +0000
@@ -29,6 +29,15 @@ config USB_EHCI_HCD
To compile this driver as a module, choose M here: the
module will be called ehci-hcd.
+config USB_EHCI_SPLIT_ISO
+ bool "Full speed ISO transactions (EXPERIMENTAL)"
+ depends on USB_EHCI_HCD && EXPERIMENTAL
+ default n
+ ---help---
+ This code is new and hasn't been used with many different
+ EHCI or USB 2.0 transaction translator implementations.
+ It should work for ISO-OUT transfers, like audio.
+
config USB_OHCI_HCD
tristate "OHCI HCD support"
depends on USB
--- diff/drivers/usb/host/ehci-dbg.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/usb/host/ehci-dbg.c 2004-03-16 11:27:48.279848008 +0000
@@ -579,7 +579,11 @@ show_periodic (struct class_device *clas
break;
case Q_TYPE_SITD:
temp = scnprintf (next, size,
- " sitd/%p", p.sitd);
+ " sitd%d-%04x/%p",
+ p.sitd->stream->interval,
+ le32_to_cpup (&p.sitd->hw_uframe)
+ & 0x0000ffff,
+ p.sitd);
tag = Q_NEXT_TYPE (p.sitd->hw_next);
p = p.sitd->sitd_next;
break;
--- diff/drivers/usb/host/ehci-hcd.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/usb/host/ehci-hcd.c 2004-03-16 11:27:48.281847704 +0000
@@ -106,8 +106,6 @@ static const char hcd_name [] = "ehci_hc
#undef EHCI_VERBOSE_DEBUG
#undef EHCI_URB_TRACE
-// #define have_split_iso
-
#ifdef DEBUG
#define EHCI_STATS
#endif
@@ -676,6 +674,7 @@ static void ehci_work (struct ehci_hcd *
/* the IO watchdog guards against hardware or driver bugs that
* misplace IRQs, and should let us run completely without IRQs.
+ * such lossage has been observed on both VT6202 and VT8235.
*/
if ((ehci->async->qh_next.ptr != 0) || (ehci->periodic_sched != 0))
timer_action (ehci, TIMER_IO_WATCHDOG);
@@ -796,13 +795,8 @@ static int ehci_urb_enqueue (
case PIPE_ISOCHRONOUS:
if (urb->dev->speed == USB_SPEED_HIGH)
return itd_submit (ehci, urb, mem_flags);
-#ifdef have_split_iso
else
return sitd_submit (ehci, urb, mem_flags);
-#else
- dbg ("no split iso support yet");
- return -ENOSYS;
-#endif /* have_split_iso */
}
}
--- diff/drivers/usb/host/ehci-sched.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/usb/host/ehci-sched.c 2004-03-16 11:27:48.284847248 +0000
@@ -53,14 +53,10 @@ periodic_next_shadow (union ehci_shadow
return &periodic->fstn->fstn_next;
case Q_TYPE_ITD:
return &periodic->itd->itd_next;
-#ifdef have_split_iso
- case Q_TYPE_SITD:
+ // case Q_TYPE_SITD:
+ default:
return &periodic->sitd->sitd_next;
-#endif /* have_split_iso */
}
- dbg ("BAD shadow %p tag %d", periodic->ptr, tag);
- // BUG ();
- return 0;
}
/* returns true after successful unlink */
@@ -133,7 +129,6 @@ periodic_usecs (struct ehci_hcd *ehci, u
hw_p = &q->itd->hw_next;
q = &q->itd->itd_next;
break;
-#ifdef have_split_iso
case Q_TYPE_SITD:
/* is it in the S-mask? (count SPLIT, DATA) */
if (q->sitd->hw_uframe & cpu_to_le32 (1 << uframe)) {
@@ -154,7 +149,6 @@ periodic_usecs (struct ehci_hcd *ehci, u
hw_p = &q->sitd->hw_next;
q = &q->sitd->sitd_next;
break;
-#endif /* have_split_iso */
default:
BUG ();
}
@@ -229,7 +223,8 @@ static int tt_no_collision (
if (same_tt (dev, here.itd->urb->dev)) {
u16 mask;
- mask = le32_to_cpu (here.sitd->hw_uframe);
+ mask = le32_to_cpu (here.sitd
+ ->hw_uframe);
/* FIXME assumes no gap for IN! */
mask |= mask >> 8;
if (mask & uf_mask)
@@ -237,7 +232,7 @@ static int tt_no_collision (
}
type = Q_NEXT_TYPE (here.qh->hw_next);
here = here.sitd->sitd_next;
- break;
+ continue;
// case Q_TYPE_FSTN:
default:
ehci_dbg (ehci,
@@ -698,12 +693,27 @@ iso_stream_put(struct ehci_hcd *ehci, st
// BUG_ON (!list_empty(&stream->td_list));
while (!list_empty (&stream->free_list)) {
- struct ehci_itd *itd;
+ struct list_head *entry;
- itd = list_entry (stream->free_list.next,
- struct ehci_itd, itd_list);
- list_del (&itd->itd_list);
- dma_pool_free (ehci->itd_pool, itd, itd->itd_dma);
+ entry = stream->free_list.next;
+ list_del (entry);
+
+ /* knows about ITD vs SITD */
+ if (stream->highspeed) {
+ struct ehci_itd *itd;
+
+ itd = list_entry (entry, struct ehci_itd,
+ itd_list);
+ dma_pool_free (ehci->itd_pool, itd,
+ itd->itd_dma);
+ } else {
+ struct ehci_sitd *sitd;
+
+ sitd = list_entry (entry, struct ehci_sitd,
+ sitd_list);
+ dma_pool_free (ehci->sitd_pool, sitd,
+ sitd->sitd_dma);
+ }
}
is_in = (stream->bEndpointAddress & USB_DIR_IN) ? 0x10 : 0;
@@ -858,6 +868,7 @@ itd_urb_transaction (
int i;
unsigned num_itds;
struct ehci_iso_sched *sched;
+ unsigned long flags;
sched = iso_sched_alloc (urb->number_of_packets, mem_flags);
if (unlikely (sched == 0))
@@ -871,6 +882,7 @@ itd_urb_transaction (
num_itds = urb->number_of_packets;
/* allocate/init ITDs */
+ spin_lock_irqsave (&ehci->lock, flags);
for (i = 0; i < num_itds; i++) {
/* free_list.next might be cache-hot ... but maybe
@@ -884,8 +896,14 @@ itd_urb_transaction (
list_del (&itd->itd_list);
itd_dma = itd->itd_dma;
} else
+ itd = 0;
+
+ if (!itd) {
+ spin_unlock_irqrestore (&ehci->lock, flags);
itd = dma_pool_alloc (ehci->itd_pool, mem_flags,
&itd_dma);
+ spin_lock_irqsave (&ehci->lock, flags);
+ }
if (unlikely (0 == itd)) {
iso_sched_free (stream, sched);
@@ -895,6 +913,7 @@ itd_urb_transaction (
itd->itd_dma = itd_dma;
list_add (&itd->itd_list, &sched->td_list);
}
+ spin_unlock_irqrestore (&ehci->lock, flags);
/* temporarily store schedule info in hcpriv */
urb->hcpriv = sched;
@@ -909,11 +928,11 @@ itd_slot_ok (
struct ehci_hcd *ehci,
u32 mod,
u32 uframe,
- u32 end,
u8 usecs,
u32 period
)
{
+ uframe %= period;
do {
/* can't commit more than 80% periodic == 100 usec */
if (periodic_usecs (ehci, uframe >> 3, uframe & 0x7)
@@ -922,8 +941,7 @@ itd_slot_ok (
/* we know urb->interval is 2^N uframes */
uframe += period;
- uframe %= mod;
- } while (uframe != end);
+ } while (uframe < mod);
return 1;
}
@@ -933,7 +951,6 @@ sitd_slot_ok (
u32 mod,
struct ehci_iso_stream *stream,
u32 uframe,
- u32 end,
struct ehci_iso_sched *sched,
u32 period_uframes
)
@@ -952,12 +969,20 @@ sitd_slot_ok (
*/
/* check bandwidth */
+ uframe %= period_uframes;
do {
u32 max_used;
frame = uframe >> 3;
uf = uframe & 7;
+ /* tt must be idle for start(s), any gap, and csplit.
+ * assume scheduling slop leaves 10+% for control/bulk.
+ */
+ if (!tt_no_collision (ehci, period_uframes << 3,
+ stream->udev, frame, mask))
+ return 0;
+
/* check starts (OUT uses more than one) */
max_used = 100 - stream->usecs;
for (tmp = stream->raw_mask & 0xff; tmp; tmp >>= 1, uf++) {
@@ -969,25 +994,19 @@ sitd_slot_ok (
if (stream->c_usecs) {
max_used = 100 - stream->c_usecs;
do {
- /* tt is busy in the gap before CSPLIT */
tmp = 1 << uf;
- mask |= tmp;
tmp <<= 8;
- if (stream->raw_mask & tmp)
- break;
+ if ((stream->raw_mask & tmp) == 0)
+ continue;
+ if (periodic_usecs (ehci, frame, uf)
+ > max_used)
+ return 0;
} while (++uf < 8);
- if (periodic_usecs (ehci, frame, uf) > max_used)
- return 0;
}
/* we know urb->interval is 2^N uframes */
uframe += period_uframes;
- uframe %= mod;
- } while (uframe != end);
-
- /* tt must be idle for start(s), any gap, and csplit */
- if (!tt_no_collision (ehci, period_uframes, stream->udev, frame, mask))
- return 0;
+ } while (uframe < mod);
stream->splits = stream->raw_mask << (uframe & 7);
cpu_to_le32s (&stream->splits);
@@ -1014,7 +1033,7 @@ iso_stream_schedule (
struct ehci_iso_stream *stream
)
{
- u32 now, start, end, max, period;
+ u32 now, start, max, period;
int status;
unsigned mod = ehci->periodic_size << 3;
struct ehci_iso_sched *sched = urb->hcpriv;
@@ -1036,8 +1055,6 @@ iso_stream_schedule (
/* when's the last uframe this urb could start? */
max = now + mod;
- max -= sched->span;
- max -= 8 * SCHEDULE_SLOP;
/* typical case: reuse current schedule. stream is still active,
* and no gaps from host falling behind (irq delays etc)
@@ -1046,9 +1063,11 @@ iso_stream_schedule (
start = stream->next_uframe;
if (start < now)
start += mod;
- if (likely (start < max))
+ if (likely ((start + sched->span) < max))
goto ready;
- /* else fell behind; try to reschedule */
+ /* else fell behind; someday, try to reschedule */
+ status = -EL2NSYNC;
+ goto fail;
}
/* need to schedule; when's the next (u)frame we could start?
@@ -1059,63 +1078,40 @@ iso_stream_schedule (
*/
start = SCHEDULE_SLOP * 8 + (now & ~0x07);
start %= mod;
- end = start;
+ stream->next_uframe = start;
/* NOTE: assumes URB_ISO_ASAP, to limit complexity/bugs */
period = urb->interval;
if (!stream->highspeed)
period <<= 3;
- if (max > (start + period))
- max = start + period;
- /* hack: account for itds already scheduled to this endpoint */
- if (list_empty (&stream->td_list))
- end = max;
-
- /* within [start..max] find a uframe slot with enough bandwidth */
- end %= mod;
- do {
+ /* find a uframe slot with enough bandwidth */
+ for (; start < (stream->next_uframe + period); start++) {
int enough_space;
/* check schedule: enough space? */
if (stream->highspeed)
- enough_space = itd_slot_ok (ehci, mod, start, end,
+ enough_space = itd_slot_ok (ehci, mod, start,
stream->usecs, period);
else {
if ((start % 8) >= 6)
continue;
enough_space = sitd_slot_ok (ehci, mod, stream,
- start, end, sched, period);
+ start, sched, period);
}
- /* (re)schedule it here if there's enough bandwidth */
+ /* schedule it here if there's enough bandwidth */
if (enough_space) {
- start %= mod;
- if (unlikely (!list_empty (&stream->td_list))) {
- /* host fell behind ... maybe irq latencies
- * delayed this request queue for too long.
- */
- stream->rescheduled++;
- dev_dbg (&urb->dev->dev,
- "iso%d%s %d.%d skip %d.%d\n",
- stream->bEndpointAddress & 0x0f,
- (stream->bEndpointAddress & USB_DIR_IN)
- ? "in" : "out",
- stream->next_uframe >> 3,
- stream->next_uframe & 0x7,
- start >> 3, start & 0x7);
- }
- stream->next_uframe = start;
+ stream->next_uframe = start % mod;
goto ready;
}
-
- } while (++start < max);
+ }
/* no room in the schedule */
- ehci_dbg (ehci, "iso %ssched full %p (now %d end %d max %d)\n",
+ ehci_dbg (ehci, "iso %ssched full %p (now %d max %d)\n",
list_empty (&stream->td_list) ? "" : "re",
- urb, now, end, max);
+ urb, now, max);
status = -ENOSPC;
fail:
@@ -1260,6 +1256,7 @@ itd_link_urb (
iso_sched_free (stream, iso_sched);
urb->hcpriv = 0;
+ timer_action (ehci, TIMER_IO_WATCHDOG);
if (unlikely (!ehci->periodic_sched++))
return enable_periodic (ehci);
return 0;
@@ -1404,18 +1401,392 @@ done:
return status;
}
-#ifdef have_split_iso
+#ifdef CONFIG_USB_EHCI_SPLIT_ISO
/*-------------------------------------------------------------------------*/
/*
- * "Split ISO TDs" ... used for USB 1.1 devices going through
- * the TTs in USB 2.0 hubs.
- *
- * FIXME not yet implemented
+ * "Split ISO TDs" ... used for USB 1.1 devices going through the
+ * TTs in USB 2.0 hubs. These need microframe scheduling.
*/
-#endif /* have_split_iso */
+static inline void
+sitd_sched_init (
+ struct ehci_iso_sched *iso_sched,
+ struct ehci_iso_stream *stream,
+ struct urb *urb
+)
+{
+ unsigned i;
+ dma_addr_t dma = urb->transfer_dma;
+
+ /* how many frames are needed for these transfers */
+ iso_sched->span = urb->number_of_packets * stream->interval;
+
+ /* figure out per-frame sitd fields that we'll need later
+ * when we fit new sitds into the schedule.
+ */
+ for (i = 0; i < urb->number_of_packets; i++) {
+ struct ehci_iso_packet *packet = &iso_sched->packet [i];
+ unsigned length;
+ dma_addr_t buf;
+ u32 trans;
+
+ length = urb->iso_frame_desc [i].length & 0x03ff;
+ buf = dma + urb->iso_frame_desc [i].offset;
+
+ trans = SITD_STS_ACTIVE;
+ if (((i + 1) == urb->number_of_packets)
+ && !(urb->transfer_flags & URB_NO_INTERRUPT))
+ trans |= SITD_IOC;
+ trans |= length << 16;
+ packet->transaction = cpu_to_le32 (trans);
+
+ /* might need to cross a buffer page within a td */
+ packet->bufp = buf;
+ buf += length;
+ packet->buf1 = buf & ~0x0fff;
+ if (packet->buf1 != (buf & ~(u64)0x0fff))
+ packet->cross = 1;
+
+ /* OUT uses multiple start-splits */
+ if (stream->bEndpointAddress & USB_DIR_IN)
+ continue;
+ length = 1 + (length / 188);
+ packet->buf1 |= length;
+ if (length > 1) /* BEGIN vs ALL */
+ packet->buf1 |= 1 << 3;
+ }
+}
+
+static int
+sitd_urb_transaction (
+ struct ehci_iso_stream *stream,
+ struct ehci_hcd *ehci,
+ struct urb *urb,
+ int mem_flags
+)
+{
+ struct ehci_sitd *sitd;
+ dma_addr_t sitd_dma;
+ int i;
+ struct ehci_iso_sched *iso_sched;
+ unsigned long flags;
+
+ iso_sched = iso_sched_alloc (urb->number_of_packets, mem_flags);
+ if (iso_sched == 0)
+ return -ENOMEM;
+
+ sitd_sched_init (iso_sched, stream, urb);
+
+ /* allocate/init sITDs */
+ spin_lock_irqsave (&ehci->lock, flags);
+ for (i = 0; i < urb->number_of_packets; i++) {
+
+ /* NOTE: for now, we don't try to handle wraparound cases
+ * for IN (using sitd->hw_backpointer, like a FSTN), which
+ * means we never need two sitds for full speed packets.
+ */
+
+ /* free_list.next might be cache-hot ... but maybe
+ * the HC caches it too. avoid that issue for now.
+ */
+
+ /* prefer previously-allocated sitds */
+ if (!list_empty(&stream->free_list)) {
+ sitd = list_entry (stream->free_list.prev,
+ struct ehci_sitd, sitd_list);
+ list_del (&sitd->sitd_list);
+ sitd_dma = sitd->sitd_dma;
+ } else
+ sitd = 0;
+
+ if (!sitd) {
+ spin_unlock_irqrestore (&ehci->lock, flags);
+ sitd = dma_pool_alloc (ehci->sitd_pool, mem_flags,
+ &sitd_dma);
+ spin_lock_irqsave (&ehci->lock, flags);
+ }
+
+ if (!sitd) {
+ iso_sched_free (stream, iso_sched);
+ spin_unlock_irqrestore (&ehci->lock, flags);
+ return -ENOMEM;
+ }
+ memset (sitd, 0, sizeof *sitd);
+ sitd->sitd_dma = sitd_dma;
+ list_add (&sitd->sitd_list, &iso_sched->td_list);
+ }
+
+ /* temporarily store schedule info in hcpriv */
+ urb->hcpriv = iso_sched;
+ urb->error_count = 0;
+
+ spin_unlock_irqrestore (&ehci->lock, flags);
+ return 0;
+}
+
+/*-------------------------------------------------------------------------*/
+
+static inline void
+sitd_patch (
+ struct ehci_iso_stream *stream,
+ struct ehci_sitd *sitd,
+ struct ehci_iso_sched *iso_sched,
+ unsigned index
+)
+{
+ struct ehci_iso_packet *uf = &iso_sched->packet [index];
+ u64 bufp = uf->bufp;
+
+ sitd->hw_next = EHCI_LIST_END;
+ sitd->hw_fullspeed_ep = stream->address;
+ sitd->hw_uframe = stream->splits;
+ sitd->hw_results = uf->transaction;
+ sitd->hw_backpointer = EHCI_LIST_END;
+
+ bufp = uf->bufp;
+ sitd->hw_buf [0] = cpu_to_le32 (bufp);
+ sitd->hw_buf_hi [0] = cpu_to_le32 (bufp >> 32);
+
+ sitd->hw_buf [1] = cpu_to_le32 (uf->buf1);
+ if (uf->cross) {
+ bufp += 4096;
+ sitd->hw_buf_hi [1] = cpu_to_le32 (bufp >> 32);
+ }
+ sitd->index = index;
+}
+
+static inline void
+sitd_link (struct ehci_hcd *ehci, unsigned frame, struct ehci_sitd *sitd)
+{
+ /* note: sitd ordering could matter (CSPLIT then SSPLIT) */
+ sitd->sitd_next = ehci->pshadow [frame];
+ sitd->hw_next = ehci->periodic [frame];
+ ehci->pshadow [frame].sitd = sitd;
+ sitd->frame = frame;
+ wmb ();
+ ehci->periodic [frame] = cpu_to_le32 (sitd->sitd_dma) | Q_TYPE_SITD;
+}
+
+/* fit urb's sitds into the selected schedule slot; activate as needed */
+static int
+sitd_link_urb (
+ struct ehci_hcd *ehci,
+ struct urb *urb,
+ unsigned mod,
+ struct ehci_iso_stream *stream
+)
+{
+ int packet;
+ unsigned next_uframe;
+ struct ehci_iso_sched *sched = urb->hcpriv;
+ struct ehci_sitd *sitd;
+
+ next_uframe = stream->next_uframe;
+
+ if (list_empty(&stream->td_list)) {
+ /* usbfs ignores TT bandwidth */
+ hcd_to_bus (&ehci->hcd)->bandwidth_allocated
+ += stream->bandwidth;
+ ehci_vdbg (ehci,
+ "sched dev%s ep%d%s-iso [%d] %dms/%04x\n",
+ urb->dev->devpath, stream->bEndpointAddress & 0x0f,
+ (stream->bEndpointAddress & USB_DIR_IN) ? "in" : "out",
+ (next_uframe >> 3) % ehci->periodic_size,
+ stream->interval, le32_to_cpu (stream->splits));
+ stream->start = jiffies;
+ }
+ hcd_to_bus (&ehci->hcd)->bandwidth_isoc_reqs++;
+
+ /* fill sITDs frame by frame */
+ for (packet = 0, sitd = 0;
+ packet < urb->number_of_packets;
+ packet++) {
+
+ /* ASSERT: we have all necessary sitds */
+ BUG_ON (list_empty (&sched->td_list));
+
+ /* ASSERT: no itds for this endpoint in this frame */
+
+ sitd = list_entry (sched->td_list.next,
+ struct ehci_sitd, sitd_list);
+ list_move_tail (&sitd->sitd_list, &stream->td_list);
+ sitd->stream = iso_stream_get (stream);
+ sitd->urb = usb_get_urb (urb);
+
+ sitd_patch (stream, sitd, sched, packet);
+ sitd_link (ehci, (next_uframe >> 3) % ehci->periodic_size,
+ sitd);
+
+ next_uframe += stream->interval << 3;
+ stream->depth += stream->interval << 3;
+ }
+ stream->next_uframe = next_uframe % mod;
+
+ /* don't need that schedule data any more */
+ iso_sched_free (stream, sched);
+ urb->hcpriv = 0;
+
+ timer_action (ehci, TIMER_IO_WATCHDOG);
+ if (!ehci->periodic_sched++)
+ return enable_periodic (ehci);
+ return 0;
+}
+
+/*-------------------------------------------------------------------------*/
+
+#define SITD_ERRS (SITD_STS_ERR | SITD_STS_DBE | SITD_STS_BABBLE \
+ | SITD_STS_XACT | SITD_STS_MMF | SITD_STS_STS)
+
+static unsigned
+sitd_complete (
+ struct ehci_hcd *ehci,
+ struct ehci_sitd *sitd,
+ struct pt_regs *regs
+) {
+ struct urb *urb = sitd->urb;
+ struct usb_iso_packet_descriptor *desc;
+ u32 t;
+ int urb_index = -1;
+ struct ehci_iso_stream *stream = sitd->stream;
+ struct usb_device *dev;
+
+ urb_index = sitd->index;
+ desc = &urb->iso_frame_desc [urb_index];
+ t = le32_to_cpup (&sitd->hw_results);
+
+ /* report transfer status */
+ if (t & SITD_ERRS) {
+ urb->error_count++;
+ if (t & SITD_STS_DBE)
+ desc->status = usb_pipein (urb->pipe)
+ ? -ENOSR /* hc couldn't read */
+ : -ECOMM; /* hc couldn't write */
+ else if (t & SITD_STS_BABBLE)
+ desc->status = -EOVERFLOW;
+ else /* XACT, MMF, etc */
+ desc->status = -EPROTO;
+ } else {
+ desc->status = 0;
+ desc->actual_length = desc->length - SITD_LENGTH (t);
+ }
+
+ usb_put_urb (urb);
+ sitd->urb = 0;
+ sitd->stream = 0;
+ list_move (&sitd->sitd_list, &stream->free_list);
+ stream->depth -= stream->interval << 3;
+ iso_stream_put (ehci, stream);
+
+ /* handle completion now? */
+ if ((urb_index + 1) != urb->number_of_packets)
+ return 0;
+
+ /* ASSERT: it's really the last sitd for this urb
+ list_for_each_entry (sitd, &stream->td_list, sitd_list)
+ BUG_ON (sitd->urb == urb);
+ */
+
+ /* give urb back to the driver */
+ dev = usb_get_dev (urb->dev);
+ ehci_urb_done (ehci, urb, regs);
+ urb = 0;
+
+ /* defer stopping schedule; completion can submit */
+ ehci->periodic_sched--;
+ if (!ehci->periodic_sched)
+ (void) disable_periodic (ehci);
+ hcd_to_bus (&ehci->hcd)->bandwidth_isoc_reqs--;
+
+ if (list_empty (&stream->td_list)) {
+ hcd_to_bus (&ehci->hcd)->bandwidth_allocated
+ -= stream->bandwidth;
+ ehci_vdbg (ehci,
+ "deschedule devp %s ep%d%s-iso\n",
+ dev->devpath, stream->bEndpointAddress & 0x0f,
+ (stream->bEndpointAddress & USB_DIR_IN) ? "in" : "out");
+ }
+ iso_stream_put (ehci, stream);
+ usb_put_dev (dev);
+
+ return 1;
+}
+
+
+static int sitd_submit (struct ehci_hcd *ehci, struct urb *urb, int mem_flags)
+{
+ int status = -EINVAL;
+ unsigned long flags;
+ struct ehci_iso_stream *stream;
+
+ // FIXME remove when csplits behave
+ if (usb_pipein(urb->pipe)) {
+ ehci_dbg (ehci, "no iso-IN split transactions yet\n");
+ return -ENOMEM;
+ }
+
+ /* Get iso_stream head */
+ stream = iso_stream_find (ehci, urb);
+ if (stream == 0) {
+ ehci_dbg (ehci, "can't get iso stream\n");
+ return -ENOMEM;
+ }
+ if (urb->interval != stream->interval) {
+ ehci_dbg (ehci, "can't change iso interval %d --> %d\n",
+ stream->interval, urb->interval);
+ goto done;
+ }
+
+#ifdef EHCI_URB_TRACE
+ ehci_dbg (ehci,
+ "submit %p dev%s ep%d%s-iso len %d\n",
+ urb, urb->dev->devpath,
+ usb_pipeendpoint (urb->pipe),
+ usb_pipein (urb->pipe) ? "in" : "out",
+ urb->transfer_buffer_length);
+#endif
+
+ /* allocate SITDs */
+ status = sitd_urb_transaction (stream, ehci, urb, mem_flags);
+ if (status < 0) {
+ ehci_dbg (ehci, "can't init sitds\n");
+ goto done;
+ }
+
+ /* schedule ... need to lock */
+ spin_lock_irqsave (&ehci->lock, flags);
+ status = iso_stream_schedule (ehci, urb, stream);
+ if (status == 0)
+ sitd_link_urb (ehci, urb, ehci->periodic_size << 3, stream);
+ spin_unlock_irqrestore (&ehci->lock, flags);
+
+done:
+ if (status < 0)
+ iso_stream_put (ehci, stream);
+ return status;
+}
+
+#else
+
+static inline int
+sitd_submit (struct ehci_hcd *ehci, struct urb *urb, int mem_flags)
+{
+ ehci_dbg (ehci, "split iso support is disabled\n");
+ return -ENOSYS;
+}
+
+static inline unsigned
+sitd_complete (
+ struct ehci_hcd *ehci,
+ struct ehci_sitd *sitd,
+ struct pt_regs *regs
+) {
+ ehci_err (ehci, "sitd_complete %p?\n", sitd);
+ return 0;
+}
+
+#endif /* USB_EHCI_SPLIT_ISO */
/*-------------------------------------------------------------------------*/
@@ -1513,7 +1884,6 @@ restart:
modified = itd_complete (ehci, q.itd, regs);
q = *q_p;
break;
-#ifdef have_split_iso
case Q_TYPE_SITD:
if (q.sitd->hw_results & SITD_ACTIVE) {
q_p = &q.sitd->sitd_next;
@@ -1529,7 +1899,6 @@ restart:
modified = sitd_complete (ehci, q.sitd, regs);
q = *q_p;
break;
-#endif /* have_split_iso */
default:
dbg ("corrupt type %d frame %d shadow %p",
type, frame, q.ptr);
--- diff/drivers/usb/host/ehci.h 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/usb/host/ehci.h 2004-03-16 11:27:48.280847856 +0000
@@ -492,16 +492,16 @@ struct ehci_itd {
/*
* EHCI Specification 0.95 Section 3.4
* siTD, aka split-transaction isochronous Transfer Descriptor
- * ... describe low/full speed iso xfers through TT in hubs
+ * ... describe full speed iso xfers through TT in hubs
* see Figure 3-5 "Split-transaction Isochronous Transaction Descriptor (siTD)
*/
struct ehci_sitd {
/* first part defined by EHCI spec */
u32 hw_next;
/* uses bit field macros above - see EHCI 0.95 Table 3-8 */
- u32 hw_fullspeed_ep; /* see EHCI table 3-9 */
- u32 hw_uframe; /* see EHCI table 3-10 */
- u32 hw_results; /* see EHCI table 3-11 */
+ u32 hw_fullspeed_ep; /* EHCI table 3-9 */
+ u32 hw_uframe; /* EHCI table 3-10 */
+ u32 hw_results; /* EHCI table 3-11 */
#define SITD_IOC (1 << 31) /* interrupt on completion */
#define SITD_PAGE (1 << 30) /* buffer 0/1 */
#define SITD_LENGTH(x) (0x3ff & ((x)>>16))
@@ -515,8 +515,8 @@ struct ehci_sitd {
#define SITD_ACTIVE __constant_cpu_to_le32(SITD_STS_ACTIVE)
- u32 hw_buf [2]; /* see EHCI table 3-12 */
- u32 hw_backpointer; /* see EHCI table 3-13 */
+ u32 hw_buf [2]; /* EHCI table 3-12 */
+ u32 hw_backpointer; /* EHCI table 3-13 */
u32 hw_buf_hi [2]; /* Appendix B */
/* the rest is HCD-private */
@@ -552,8 +552,6 @@ struct ehci_fstn {
/*-------------------------------------------------------------------------*/
-#define SUBMIT_URB(urb,mem_flags) usb_submit_urb(urb,mem_flags)
-
#ifndef DEBUG
#define STUB_DEBUG_FILES
#endif /* DEBUG */
--- diff/drivers/usb/host/uhci-hcd.c 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/usb/host/uhci-hcd.c 2004-03-16 11:27:48.286846944 +0000
@@ -781,7 +781,8 @@ static void uhci_dec_fsbr(struct uhci_hc
/*
* Map status to standard result codes
*
- * <status> is (td->status & 0xFE0000) [a.k.a. uhci_status_bits(td->status)]
+ * <status> is (td->status & 0xF60000) [a.k.a. uhci_status_bits(td->status)]
+ * Note: status does not include the TD_CTRL_NAK bit.
* <dir_out> is True for output TDs and False for input TDs.
*/
static int uhci_map_status(int status, int dir_out)
@@ -792,22 +793,18 @@ static int uhci_map_status(int status, i
return -EPROTO;
if (status & TD_CTRL_CRCTIMEO) { /* CRC/Timeout */
if (dir_out)
- return -ETIMEDOUT;
+ return -EPROTO;
else
return -EILSEQ;
}
- if (status & TD_CTRL_NAK) /* NAK */
- return -ETIMEDOUT;
if (status & TD_CTRL_BABBLE) /* Babble */
return -EOVERFLOW;
if (status & TD_CTRL_DBUFERR) /* Buffer error */
return -ENOSR;
if (status & TD_CTRL_STALLED) /* Stalled */
return -EPIPE;
- if (status & TD_CTRL_ACTIVE) /* Active */
- return 0;
-
- return -EINVAL;
+ WARN_ON(status & TD_CTRL_ACTIVE); /* Active */
+ return 0;
}
/*
@@ -832,7 +829,7 @@ static int uhci_submit_control(struct uh
status |= TD_CTRL_LS;
/*
- * Build the TD for the control request
+ * Build the TD for the control request setup packet
*/
td = uhci_alloc_td(uhci, urb->dev);
if (!td)
@@ -990,13 +987,13 @@ static int uhci_result_control(struct uh
if (urbp->short_control_packet) {
tmp = head->prev;
- goto status_phase;
+ goto status_stage;
}
tmp = head->next;
td = list_entry(tmp, struct uhci_td, list);
- /* The first TD is the SETUP phase, check the status, but skip */
+ /* The first TD is the SETUP stage, check the status, but skip */
/* the count */
status = uhci_status_bits(td_status(td));
if (status & TD_CTRL_ACTIVE)
@@ -1037,10 +1034,10 @@ static int uhci_result_control(struct uh
}
}
-status_phase:
+status_stage:
td = list_entry(tmp, struct uhci_td, list);
- /* Control status phase */
+ /* Control status stage */
status = td_status(td);
#ifdef I_HAVE_BUGGY_APC_BACKUPS
@@ -1053,10 +1050,11 @@ status_phase:
return 0;
#endif
+ status = uhci_status_bits(status);
if (status & TD_CTRL_ACTIVE)
return -EINPROGRESS;
- if (uhci_status_bits(status))
+ if (status)
goto td_error;
return 0;
@@ -1273,12 +1271,6 @@ static inline int uhci_submit_interrupt(
}
/*
- * Bulk and interrupt use common result
- */
-#define uhci_result_bulk uhci_result_common
-#define uhci_result_interrupt uhci_result_common
-
-/*
* Isochronous transfers
*/
static int isochronous_find_limits(struct uhci_hcd *uhci, struct urb *urb, unsigned int *start, unsigned int *end)
@@ -1403,7 +1395,8 @@ static int uhci_result_isochronous(struc
urb->iso_frame_desc[i].actual_length = actlength;
urb->actual_length += actlength;
- status = uhci_map_status(uhci_status_bits(td_status(td)), usb_pipeout(urb->pipe));
+ status = uhci_map_status(uhci_status_bits(td_status(td)),
+ usb_pipeout(urb->pipe));
urb->iso_frame_desc[i].status = status;
if (status) {
urb->error_count++;
@@ -1508,12 +1501,9 @@ static int uhci_urb_enqueue(struct usb_h
struct urb_priv *urbp = urb->hcpriv;
list_del_init(&urbp->urb_list);
- spin_unlock_irqrestore(&uhci->urb_list_lock, flags);
- uhci_destroy_urb_priv (uhci, urb);
-
- return ret;
- }
- ret = 0;
+ uhci_destroy_urb_priv(uhci, urb);
+ } else
+ ret = 0;
out:
spin_unlock_irqrestore(&uhci->urb_list_lock, flags);
@@ -1541,11 +1531,9 @@ static void uhci_transfer_result(struct
case PIPE_CONTROL:
ret = uhci_result_control(uhci, urb);
break;
- case PIPE_INTERRUPT:
- ret = uhci_result_interrupt(uhci, urb);
- break;
case PIPE_BULK:
- ret = uhci_result_bulk(uhci, urb);
+ case PIPE_INTERRUPT:
+ ret = uhci_result_common(uhci, urb);
break;
case PIPE_ISOCHRONOUS:
ret = uhci_result_isochronous(uhci, urb);
@@ -1649,10 +1637,12 @@ static int uhci_urb_dequeue(struct usb_h
{
struct uhci_hcd *uhci = hcd_to_uhci(hcd);
unsigned long flags;
- struct urb_priv *urbp = urb->hcpriv;
+ struct urb_priv *urbp;
spin_lock_irqsave(&uhci->urb_list_lock, flags);
-
+ urbp = urb->hcpriv;
+ if (!urbp) /* URB was never linked! */
+ goto done;
list_del_init(&urbp->urb_list);
uhci_unlink_generic(uhci, urb);
@@ -1665,6 +1655,7 @@ static int uhci_urb_dequeue(struct usb_h
list_add_tail(&urbp->urb_list, &uhci->urb_remove_list);
spin_unlock(&uhci->urb_remove_list_lock);
+done:
spin_unlock_irqrestore(&uhci->urb_list_lock, flags);
return 0;
}
@@ -1861,17 +1852,12 @@ static void uhci_finish_completion(struc
static void uhci_remove_pending_urbps(struct uhci_hcd *uhci)
{
- struct list_head *tmp, *head;
-
spin_lock(&uhci->urb_remove_list_lock);
- head = &uhci->urb_remove_list;
- tmp = head->next;
- while (tmp != head) {
- struct urb_priv *urbp = list_entry(tmp, struct urb_priv, urb_list);
+ spin_lock(&uhci->complete_list_lock);
- tmp = tmp->next;
- uhci_moveto_complete(uhci, urbp);
- }
+ /* Splice the urb_remove_list onto the end of the complete_list */
+ list_splice_init(&uhci->urb_remove_list, uhci->complete_list.prev);
+ spin_unlock(&uhci->complete_list_lock);
spin_unlock(&uhci->urb_remove_list_lock);
}
@@ -2458,9 +2444,11 @@ static int uhci_suspend(struct usb_hcd *
struct uhci_hcd *uhci = hcd_to_uhci(hcd);
/* Don't try to suspend broken motherboards, reset instead */
- if (suspend_allowed(uhci))
+ if (suspend_allowed(uhci)) {
suspend_hc(uhci);
- else
+ uhci->saved_framenumber =
+ inw(uhci->io_addr + USBFRNUM) & 0x3ff;
+ } else
reset_hc(uhci);
return 0;
}
@@ -2471,9 +2459,20 @@ static int uhci_resume(struct usb_hcd *h
pci_set_master(to_pci_dev(uhci_dev(uhci)));
- if (uhci->state == UHCI_SUSPENDED)
+ if (uhci->state == UHCI_SUSPENDED) {
+
+ /*
+ * Some systems don't maintain the UHCI register values
+ * during a PM suspend/resume cycle, so reinitialize
+ * the Frame Number, the Framelist Base Address, and the
+ * Interrupt Enable registers.
+ */
+ outw(uhci->saved_framenumber, uhci->io_addr + USBFRNUM);
+ outl(uhci->fl->dma_handle, uhci->io_addr + USBFLBASEADD);
+ outw(USBINTR_TIMEOUT | USBINTR_RESUME | USBINTR_IOC |
+ USBINTR_SP, uhci->io_addr + USBINTR);
uhci->resume_detect = 1;
- else {
+ } else {
reset_hc(uhci);
start_hc(uhci);
}
--- diff/drivers/usb/host/uhci-hcd.h 2004-03-11 10:20:27.000000000 +0000
+++ source/drivers/usb/host/uhci-hcd.h 2004-03-16 11:27:48.287846792 +0000
@@ -141,7 +141,7 @@ struct uhci_qh {
TD_CTRL_BABBLE | TD_CTRL_CRCTIME | TD_CTRL_BITSTUFF)
#define uhci_maxerr(err) ((err) << TD_CTRL_C_ERR_SHIFT)
-#define uhci_status_bits(ctrl_sts) ((ctrl_sts) & 0xFE0000)
+#define uhci_status_bits(ctrl_sts) ((ctrl_sts) & 0xF60000)
#define uhci_actual_length(ctrl_sts) (((ctrl_sts) + 1) & TD_CTRL_ACTLEN_MASK) /* 1-based */
/*
@@ -350,6 +350,7 @@ struct uhci_hcd {
enum uhci_state state; /* FIXME: needs a spinlock */
unsigned long state_end; /* Time of next transition */
int resume_detect; /* Need a Global Resume */
+ unsigned int saved_framenumber; /* Save during PM suspend */
/* Main list of URB's currently controlled by this HC */
spinlock_t urb_list_lock;
--- diff/drivers/usb/image/hpusbscsi.c 2003-08-20 13:16:13.000000000 +0000
+++ source/drivers/usb/image/hpusbscsi.c 2004-03-16 11:27:48.287846792 +0000
@@ -42,7 +42,7 @@ hpusbscsi_usb_probe(struct usb_interface
const struct usb_device_id *id)
{
struct usb_device *dev = interface_to_usbdev(intf);
- struct usb_host_interface *altsetting = intf->altsetting;
+ struct usb_host_interface *altsetting = intf->cur_altsetting;
struct hpusbscsi *new;
int error = -ENOMEM;
int i;
--- diff/drivers/usb/image/mdc800.c 2003-09-17 11:28:10.000000000 +0000
+++ source/drivers/usb/image/mdc800.c 2004-03-16 11:27:48.288846640 +0000
@@ -431,7 +431,7 @@ static int mdc800_usb_probe (struct usb_
err ("probe fails -> wrong Number of Configuration");
return -ENODEV;
}
- intf_desc = &intf->altsetting[0];
+ intf_desc = intf->cur_altsetting;
if (
( intf_desc->desc.bInterfaceClass != 0xff )
@@ -469,13 +469,6 @@ static int mdc800_usb_probe (struct usb_
}
- usb_driver_claim_interface (&mdc800_usb_driver, intf, mdc800);
- if (usb_set_interface (dev, intf_desc->desc.bInterfaceNumber, 0) < 0)
- {
- err ("MDC800 Configuration fails.");
- return -ENODEV;
- }
-
info ("Found Mustek MDC800 on USB.");
down (&mdc800->io_lock);
@@ -551,8 +544,6 @@ static void mdc800_usb_disconnect (struc
usb_unlink_urb (mdc800->write_urb);
usb_unlink_urb (mdc800->download_urb);
- usb_driver_release_interface (&mdc800_usb_driver, intf);
-
mdc800->dev=0;
usb_set_intfdata(intf, NULL);
}
--- diff/drivers/usb/image/microtek.c 2003-08-20 13:16:13.000000000 +0000
+++ source/drivers/usb/image/microtek.c 2004-03-16 11:27:48.289846488 +0000
@@ -693,7 +693,6 @@ static int mts_usb_probe(struct usb_inte
const struct usb_device_id *id)
{
int i;
- int result;
int ep_out = -1;
int ep_in_set[3]; /* this will break if we have more than three endpoints
which is why we check */
@@ -703,7 +702,7 @@ static int mts_usb_probe(struct usb_inte
struct vendor_product const* p;
struct usb_device *dev = interface_to_usbdev (intf);
- /* the altsettting 0 on the interface we're probing */
+ /* the current altsetting on the interface we're probing */
struct usb_host_interface *altsetting;
MTS_DEBUG_GOT_HERE();
@@ -724,8 +723,8 @@ static int mts_usb_probe(struct usb_inte
MTS_MESSAGE( "model %s is not known to be fully supported, reports welcome!\n",
p->name );
- /* the altsettting 0 on the interface we're probing */
- altsetting = &(intf->altsetting[0]);
+ /* the current altsetting on the interface we're probing */
+ altsetting = intf->cur_altsetting;
/* Check if the config is sane */
@@ -766,20 +765,6 @@ static int mts_usb_probe(struct usb_inte
MTS_WARNING( "couldn't find an output bulk endpoint. Bailing out.\n" );
return -ENODEV;
}
-
- result = usb_set_interface(dev, altsetting->desc.bInterfaceNumber, 0);
-
- MTS_DEBUG("usb_set_interface returned %d.\n",result);
- switch( result )
- {
- case 0: /* no error */
- break;
-
- default:
- MTS_DEBUG( "unknown error %d from usb_set_interface\n",
- (int)result );
- return -ENODEV;
- }
new_desc = kmalloc(sizeof(struct mts_desc), GFP_KERNEL);
--- diff/drivers/usb/input/Kconfig 2004-02-18 08:54:12.000000000 +0000
+++ source/drivers/usb/input/Kconfig 2004-03-16 11:27:48.343838280 +0000
@@ -179,6 +179,18 @@ config USB_POWERMATE
To compile this driver as a module, choose M here: the
module will be called powermate.
+config USB_MTOUCH
+ tristate "MicroTouch USB Touchscreen Driver"
+ depends on USB && INPUT
+ ---help---
+ Say Y here if you want to use a MicroTouch (Now 3M) USB
+ Touchscreen controller.
+
+ See <file:Documentation/usb/mtouch.txt> for additional information.
+
+ To compile this driver as a module, choose M here: the
+ module will be called mtouchusb.
+
config USB_XPAD
tristate "X-Box gamepad support"
depends on USB && INPUT
@@ -192,3 +204,17 @@ config USB_XPAD
To compile this driver as a module, choose M here: the
module will be called xpad.
+
+config USB_ATI_REMOTE
+ tristate "ATI USB RF remote control"
+ depends on USB && INPUT
+ ---help---
+ Say Y here if you want to use one of ATI's USB remote controls.
+ These are RF remotes with USB receivers. They come with many of ATI's
+ All-In-Wonder video cards. This driver provides mouse pointer, left
+ and right mouse buttons, and maps all the other remote buttons to
+ keypress events.
+
+ To compile this driver as a module, choose M here: the module will be
+ called ati_remote.
+
--- diff/drivers/usb/input/Makefile 2003-05-21 10:49:46.000000000 +0000
+++ source/drivers/usb/input/Makefile 2004-03-16 11:27:48.343838280 +0000
@@ -27,10 +27,12 @@ ifeq ($(CONFIG_HID_FF),y)
endif
obj-$(CONFIG_USB_AIPTEK) += aiptek.o
+obj-$(CONFIG_USB_ATI_REMOTE) += ati_remote.o
obj-$(CONFIG_USB_HID) += hid.o
obj-$(CONFIG_USB_KBD) += usbkbd.o
-obj-$(CONFIG_USB_MOUSE) += usbmouse.o
-obj-$(CONFIG_USB_WACOM) += wacom.o
obj-$(CONFIG_USB_KBTAB) += kbtab.o
+obj-$(CONFIG_USB_MOUSE) += usbmouse.o
+obj-$(CONFIG_USB_MTOUCH) += mtouchusb.o
obj-$(CONFIG_USB_POWERMATE) += powermate.o
+obj-$(CONFIG_USB_WACOM) += wacom.o
obj-$(CONFIG_USB_XPAD) += xpad.o
--- diff/drivers/usb/input/aiptek.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/input/aiptek.c 2004-03-16 11:27:48.290846336 +0000
@@ -43,7 +43,8 @@
#include <linux/module.h>
#include <linux/init.h>
#include <linux/usb.h>
-
+#include <asm/unaligned.h>
+#include <asm/byteorder.h>
/*
* Version Information
*/
@@ -160,9 +161,9 @@ aiptek_irq(struct urb *urb, struct pt_re
proximity = data[5] & 0x01;
input_report_key(dev, BTN_TOOL_PEN, proximity);
- x = ((__u32) data[1]) | ((__u32) data[2] << 8);
- y = ((__u32) data[3]) | ((__u32) data[4] << 8);
- pressure = ((__u32) data[6]) | ((__u32) data[7] << 8);
+ x = le16_to_cpu(get_unaligned((u16 *) &data[1]));
+ y = le16_to_cpu(get_unaligned((u16 *) &data[3]));
+ pressure = le16_to_cpu(*(u16 *) &data[6]);
pressure -= aiptek->features->pressure_min;
if (pressure < 0) {
@@ -209,8 +210,10 @@ aiptek_open(struct input_dev *dev)
return 0;
aiptek->irq->dev = aiptek->usbdev;
- if (usb_submit_urb(aiptek->irq, GFP_KERNEL))
+ if (usb_submit_urb(aiptek->irq, GFP_KERNEL)) {
+ aiptek->open--;
return -EIO;
+ }
return 0;
}
@@ -234,19 +237,27 @@ usb_set_report(struct usb_device *dev, s
(type << 8) + id, inter->desc.bInterfaceNumber, buf, size, HZ);
}
-static void
+static int
aiptek_command(struct usb_device *dev, struct usb_host_interface *inter,
unsigned char command, unsigned char data)
{
- __u8 buf[3];
+ u8 *buf;
+ int err;
+
+ buf = kmalloc(3, GFP_KERNEL);
+ if (!buf)
+ return -ENOMEM;
buf[0] = 4;
buf[1] = command;
buf[2] = data;
- if (usb_set_report(dev, inter, 3, 2, buf, 3) != 3) {
+ if ((err = usb_set_report(dev, inter, 3, 2, buf, 3)) != 3) {
dbg("aiptek_command: 0x%x 0x%x\n", command, data);
}
+
+ kfree(buf);
+ return err < 0 ? err : 0;
}
static int
@@ -257,30 +268,32 @@ aiptek_probe(struct usb_interface *intf,
struct usb_host_interface *interface = intf->altsetting + 0;
struct usb_endpoint_descriptor *endpoint;
struct aiptek *aiptek;
+ int err = -ENOMEM;
if (!(aiptek = kmalloc(sizeof (struct aiptek), GFP_KERNEL)))
- return -ENOMEM;
+ goto error_out_noalloc;
memset(aiptek, 0, sizeof (struct aiptek));
- aiptek->data = usb_buffer_alloc(dev, 10, SLAB_ATOMIC, &aiptek->data_dma);
+ aiptek->data = usb_buffer_alloc(dev, 10, GFP_KERNEL, &aiptek->data_dma);
if (!aiptek->data) {
- kfree(aiptek);
- return -ENOMEM;
+ goto error_out_nobuf;
}
aiptek->irq = usb_alloc_urb(0, GFP_KERNEL);
if (!aiptek->irq) {
- usb_buffer_free(dev, 10, aiptek->data, aiptek->data_dma);
- kfree(aiptek);
- return -ENOMEM;
+ goto error_out_nourb;
}
/* Resolution500LPI */
- aiptek_command(dev, interface, 0x18, 0x04);
+ err = aiptek_command(dev, interface, 0x18, 0x04);
+ if (err)
+ goto error_out;
/* SwitchToTablet */
- aiptek_command(dev, interface, 0x10, 0x01);
+ err = aiptek_command(dev, interface, 0x10, 0x01);
+ if (err)
+ goto error_out;
aiptek->features = aiptek_features + id->driver_info;
@@ -340,6 +353,16 @@ aiptek_probe(struct usb_interface *intf,
usb_set_intfdata(intf, aiptek);
return 0;
+
+error_out:
+ usb_free_urb(aiptek->irq);
+error_out_nourb:
+ usb_buffer_free(dev, 10, aiptek->data, aiptek->data_dma);
+error_out_nobuf:
+ kfree(aiptek);
+error_out_noalloc:
+ return err;
+
}
static void
--- diff/drivers/usb/input/hid-core.c 2004-02-18 08:54:12.000000000 +0000
+++ source/drivers/usb/input/hid-core.c 2004-03-16 11:27:48.295845576 +0000
@@ -224,6 +224,9 @@ static int hid_add_field(struct hid_pars
offset = report->size;
report->size += parser->global.report_size * parser->global.report_count;
+ if (usages < parser->global.report_count)
+ usages = parser->global.report_count;
+
if (usages == 0)
return 0; /* ignore padding fields */
@@ -235,9 +238,13 @@ static int hid_add_field(struct hid_pars
field->application = hid_lookup_collection(parser, HID_COLLECTION_APPLICATION);
for (i = 0; i < usages; i++) {
- field->usage[i].hid = parser->local.usage[i];
+ int j = i;
+ /* Duplicate the last usage we parsed if we have excess values */
+ if (i >= parser->local.usage_index)
+ j = parser->local.usage_index - 1;
+ field->usage[i].hid = parser->local.usage[j];
field->usage[i].collection_index =
- parser->local.collection_index[i];
+ parser->local.collection_index[j];
}
field->maxusage = usages;
@@ -1069,24 +1076,41 @@ static int hid_submit_ctrl(struct hid_de
{
struct hid_report *report;
unsigned char dir;
+ int len;
report = hid->ctrl[hid->ctrltail].report;
dir = hid->ctrl[hid->ctrltail].dir;
- if (dir == USB_DIR_OUT)
+ len = ((report->size - 1) >> 3) + 1 + (report->id > 0);
+ if (dir == USB_DIR_OUT) {
hid_output_report(report, hid->ctrlbuf);
-
- hid->urbctrl->transfer_buffer_length = ((report->size - 1) >> 3) + 1 + (report->id > 0);
- hid->urbctrl->pipe = (dir == USB_DIR_OUT) ? usb_sndctrlpipe(hid->dev, 0) : usb_rcvctrlpipe(hid->dev, 0);
+ hid->urbctrl->pipe = usb_sndctrlpipe(hid->dev, 0);
+ hid->urbctrl->transfer_buffer_length = len;
+ } else {
+ int maxpacket, padlen;
+
+ hid->urbctrl->pipe = usb_rcvctrlpipe(hid->dev, 0);
+ maxpacket = usb_maxpacket(hid->dev, hid->urbctrl->pipe, 0);
+ if (maxpacket > 0) {
+ padlen = (len + maxpacket - 1) / maxpacket;
+ padlen *= maxpacket;
+ if (padlen > HID_BUFFER_SIZE)
+ padlen = HID_BUFFER_SIZE;
+ } else
+ padlen = 0;
+ hid->urbctrl->transfer_buffer_length = padlen;
+ }
hid->urbctrl->dev = hid->dev;
hid->cr->bRequestType = USB_TYPE_CLASS | USB_RECIP_INTERFACE | dir;
hid->cr->bRequest = (dir == USB_DIR_OUT) ? HID_REQ_SET_REPORT : HID_REQ_GET_REPORT;
hid->cr->wValue = cpu_to_le16(((report->type + 1) << 8) | report->id);
hid->cr->wIndex = cpu_to_le16(hid->ifnum);
- hid->cr->wLength = cpu_to_le16(hid->urbctrl->transfer_buffer_length);
+ hid->cr->wLength = cpu_to_le16(len);
- dbg("submitting ctrl urb");
+ dbg("submitting ctrl urb: %s wValue=0x%04x wIndex=0x%04x wLength=%u",
+ hid->cr->bRequest == HID_REQ_SET_REPORT ? "Set_Report" : "Get_Report",
+ hid->cr->wValue, hid->cr->wIndex, hid->cr->wLength);
if (usb_submit_urb(hid->urbctrl, GFP_ATOMIC)) {
err("usb_submit_urb(ctrl) failed");
@@ -1262,9 +1286,25 @@ void hid_init_reports(struct hid_device
struct hid_report_enum *report_enum;
struct hid_report *report;
struct list_head *list;
- int len;
int err, ret;
+ /*
+ * The Set_Idle request is supposed to affect only the
+ * "Interrupt In" pipe. Unfortunately, buggy devices such as
+ * the BTC keyboard (ID 046e:5303) the request also affects
+ * Get_Report requests on the control pipe. In the worst
+ * case, if the device was put on idle for an indefinite
+ * amount of time (as we do below) and there are no input
+ * events to report, the Get_Report requests will just hang
+ * until we get a USB timeout. To avoid this, we temporarily
+ * establish a minimal idle time of 1ms. This shouldn't hurt
+ * bugfree devices and will cause a worst-case extra delay of
+ * 1ms for buggy ones.
+ */
+ usb_control_msg(hid->dev, usb_sndctrlpipe(hid->dev, 0),
+ HID_REQ_SET_IDLE, USB_TYPE_CLASS | USB_RECIP_INTERFACE, (1 << 8),
+ hid->ifnum, NULL, 0, HZ * USB_CTRL_SET_TIMEOUT);
+
report_enum = hid->report_enum + HID_INPUT_REPORT;
list = report_enum->report_list.next;
while (list != &report_enum->report_list) {
@@ -1297,11 +1337,8 @@ void hid_init_reports(struct hid_device
list = report_enum->report_list.next;
while (list != &report_enum->report_list) {
report = (struct hid_report *) list;
- len = ((report->size - 1) >> 3) + 1 + report_enum->numbered;
- if (len > hid->urbin->transfer_buffer_length)
- hid->urbin->transfer_buffer_length = len < HID_BUFFER_SIZE ? len : HID_BUFFER_SIZE;
usb_control_msg(hid->dev, usb_sndctrlpipe(hid->dev, 0),
- 0x0a, USB_TYPE_CLASS | USB_RECIP_INTERFACE, report->id,
+ HID_REQ_SET_IDLE, USB_TYPE_CLASS | USB_RECIP_INTERFACE, report->id,
hid->ifnum, NULL, 0, HZ * USB_CTRL_SET_TIMEOUT);
list = list->next;
}
@@ -1313,11 +1350,12 @@ void hid_init_reports(struct hid_device
#define USB_DEVICE_ID_WACOM_INTUOS 0x0020
#define USB_DEVICE_ID_WACOM_PL 0x0030
#define USB_DEVICE_ID_WACOM_INTUOS2 0x0040
+#define USB_DEVICE_ID_WACOM_VOLITO 0x0060
+#define USB_DEVICE_ID_WACOM_PTU 0x0003
#define USB_VENDOR_ID_KBGEAR 0x084e
#define USB_DEVICE_ID_KBGEAR_JAMSTUDIO 0x1001
-
#define USB_VENDOR_ID_AIPTEK 0x08ca
#define USB_DEVICE_ID_AIPTEK_6000 0x0020
@@ -1356,22 +1394,52 @@ void hid_init_reports(struct hid_device
#define USB_VENDOR_ID_A4TECH 0x09DA
#define USB_DEVICE_ID_A4TECH_WCP32PU 0x0006
+#define USB_VENDOR_ID_CYPRESS 0x04b4
+#define USB_DEVICE_ID_CYPRESS_MOUSE 0x0001
+
#define USB_VENDOR_ID_BERKSHIRE 0x0c98
#define USB_DEVICE_ID_BERKSHIRE_PCWD 0x1140
#define USB_VENDOR_ID_ALPS 0x0433
#define USB_DEVICE_ID_IBM_GAMEPAD 0x1101
+#define USB_VENDOR_ID_SAITEK 0x06a3
+#define USB_DEVICE_ID_SAITEK_RUMBLEPAD 0xff17
+
+#define USB_VENDOR_ID_NEC 0x073e
+#define USB_DEVICE_ID_NEC_USB_GAME_PAD 0x0301
+
+#define USB_VENDOR_ID_CHIC 0x05fe
+#define USB_DEVICE_ID_CHIC_GAMEPAD 0x0014
+
struct hid_blacklist {
__u16 idVendor;
__u16 idProduct;
unsigned quirks;
} hid_blacklist[] = {
+
+ { USB_VENDOR_ID_AIPTEK, USB_DEVICE_ID_AIPTEK_6000, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_BERKSHIRE, USB_DEVICE_ID_BERKSHIRE_PCWD, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_ESSENTIAL_REALITY, USB_DEVICE_ID_ESSENTIAL_REALITY_P5, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_KBGEAR, USB_DEVICE_ID_KBGEAR_JAMSTUDIO, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_GRIFFIN, USB_DEVICE_ID_POWERMATE, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_GRIFFIN, USB_DEVICE_ID_SOUNDKNOB, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_MGE, USB_DEVICE_ID_MGE_UPS, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_MGE, USB_DEVICE_ID_MGE_UPS1, HID_QUIRK_IGNORE },
+
+ { USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 100, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 200, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 300, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 400, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 500, HID_QUIRK_IGNORE },
+
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PENPARTNER, HID_QUIRK_IGNORE },
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_GRAPHIRE, HID_QUIRK_IGNORE },
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_GRAPHIRE + 1, HID_QUIRK_IGNORE },
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_GRAPHIRE + 2, HID_QUIRK_IGNORE },
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_GRAPHIRE + 3, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_GRAPHIRE + 4, HID_QUIRK_IGNORE },
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS, HID_QUIRK_IGNORE },
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS + 1, HID_QUIRK_IGNORE },
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS + 2, HID_QUIRK_IGNORE },
@@ -1383,37 +1451,34 @@ struct hid_blacklist {
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PL + 3, HID_QUIRK_IGNORE },
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PL + 4, HID_QUIRK_IGNORE },
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PL + 5, HID_QUIRK_IGNORE },
- { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS2, HID_QUIRK_IGNORE },
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS2 + 1, HID_QUIRK_IGNORE },
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS2 + 2, HID_QUIRK_IGNORE },
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS2 + 3, HID_QUIRK_IGNORE },
{ USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS2 + 4, HID_QUIRK_IGNORE },
- { USB_VENDOR_ID_KBGEAR, USB_DEVICE_ID_KBGEAR_JAMSTUDIO, HID_QUIRK_IGNORE },
- { USB_VENDOR_ID_AIPTEK, USB_DEVICE_ID_AIPTEK_6000, HID_QUIRK_IGNORE },
- { USB_VENDOR_ID_GRIFFIN, USB_DEVICE_ID_POWERMATE, HID_QUIRK_IGNORE },
- { USB_VENDOR_ID_GRIFFIN, USB_DEVICE_ID_SOUNDKNOB, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS2 + 5, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_INTUOS2 + 7, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_VOLITO, HID_QUIRK_IGNORE },
+ { USB_VENDOR_ID_WACOM, USB_DEVICE_ID_WACOM_PTU, HID_QUIRK_IGNORE },
+
{ USB_VENDOR_ID_ATEN, USB_DEVICE_ID_ATEN_UC100KM, HID_QUIRK_NOGET },
{ USB_VENDOR_ID_ATEN, USB_DEVICE_ID_ATEN_CS124U, HID_QUIRK_NOGET },
{ USB_VENDOR_ID_ATEN, USB_DEVICE_ID_ATEN_2PORTKVM, HID_QUIRK_NOGET },
{ USB_VENDOR_ID_ATEN, USB_DEVICE_ID_ATEN_4PORTKVM, HID_QUIRK_NOGET },
{ USB_VENDOR_ID_ATEN, USB_DEVICE_ID_ATEN_4PORTKVMC, HID_QUIRK_NOGET },
- { USB_VENDOR_ID_MGE, USB_DEVICE_ID_MGE_UPS, HID_QUIRK_IGNORE },
- { USB_VENDOR_ID_MGE, USB_DEVICE_ID_MGE_UPS1, HID_QUIRK_IGNORE },
- { USB_VENDOR_ID_TOPMAX, USB_DEVICE_ID_TOPMAX_COBRAPAD, HID_QUIRK_BADPAD },
- { USB_VENDOR_ID_HAPP, USB_DEVICE_ID_UGCI_DRIVING, HID_QUIRK_BADPAD|HID_QUIRK_MULTI_INPUT },
- { USB_VENDOR_ID_HAPP, USB_DEVICE_ID_UGCI_FLYING, HID_QUIRK_BADPAD|HID_QUIRK_MULTI_INPUT },
- { USB_VENDOR_ID_HAPP, USB_DEVICE_ID_UGCI_FIGHTING, HID_QUIRK_BADPAD|HID_QUIRK_MULTI_INPUT },
- { USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100, HID_QUIRK_IGNORE },
- { USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 100, HID_QUIRK_IGNORE },
- { USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 200, HID_QUIRK_IGNORE },
- { USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 300, HID_QUIRK_IGNORE },
- { USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 400, HID_QUIRK_IGNORE },
- { USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 500, HID_QUIRK_IGNORE },
{ USB_VENDOR_ID_TANGTOP, USB_DEVICE_ID_TANGTOP_USBPS2, HID_QUIRK_NOGET },
- { USB_VENDOR_ID_ESSENTIAL_REALITY, USB_DEVICE_ID_ESSENTIAL_REALITY_P5, HID_QUIRK_IGNORE },
- { USB_VENDOR_ID_A4TECH, USB_DEVICE_ID_A4TECH_WCP32PU, HID_QUIRK_2WHEEL_MOUSE_HACK },
- { USB_VENDOR_ID_BERKSHIRE, USB_DEVICE_ID_BERKSHIRE_PCWD, HID_QUIRK_IGNORE },
+
+ { USB_VENDOR_ID_A4TECH, USB_DEVICE_ID_A4TECH_WCP32PU, HID_QUIRK_2WHEEL_MOUSE_HACK_BACK },
+ { USB_VENDOR_ID_CYPRESS, USB_DEVICE_ID_CYPRESS_MOUSE, HID_QUIRK_2WHEEL_MOUSE_HACK_EXTRA },
+
{ USB_VENDOR_ID_ALPS, USB_DEVICE_ID_IBM_GAMEPAD, HID_QUIRK_BADPAD },
+ { USB_VENDOR_ID_CHIC, USB_DEVICE_ID_CHIC_GAMEPAD, HID_QUIRK_BADPAD },
+ { USB_VENDOR_ID_HAPP, USB_DEVICE_ID_UGCI_DRIVING, HID_QUIRK_BADPAD | HID_QUIRK_MULTI_INPUT },
+ { USB_VENDOR_ID_HAPP, USB_DEVICE_ID_UGCI_FLYING, HID_QUIRK_BADPAD | HID_QUIRK_MULTI_INPUT },
+ { USB_VENDOR_ID_HAPP, USB_DEVICE_ID_UGCI_FIGHTING, HID_QUIRK_BADPAD | HID_QUIRK_MULTI_INPUT },
+ { USB_VENDOR_ID_NEC, USB_DEVICE_ID_NEC_USB_GAME_PAD, HID_QUIRK_BADPAD },
+ { USB_VENDOR_ID_SAITEK, USB_DEVICE_ID_SAITEK_RUMBLEPAD, HID_QUIRK_BADPAD },
+ { USB_VENDOR_ID_TOPMAX, USB_DEVICE_ID_TOPMAX_COBRAPAD, HID_QUIRK_BADPAD },
+
{ 0, 0 }
};
@@ -1445,7 +1510,7 @@ static void hid_free_buffers(struct usb_
static struct hid_device *usb_hid_configure(struct usb_interface *intf)
{
- struct usb_host_interface *interface = intf->altsetting + intf->act_altsetting;
+ struct usb_host_interface *interface = intf->cur_altsetting;
struct usb_device *dev = interface_to_usbdev (intf);
struct hid_descriptor *hdesc;
struct hid_device *hid;
@@ -1518,12 +1583,17 @@ static struct hid_device *usb_hid_config
continue;
if (endpoint->bEndpointAddress & USB_DIR_IN) {
+ int len;
+
if (hid->urbin)
continue;
if (!(hid->urbin = usb_alloc_urb(0, GFP_KERNEL)))
goto fail;
pipe = usb_rcvintpipe(dev, endpoint->bEndpointAddress);
- usb_fill_int_urb(hid->urbin, dev, pipe, hid->inbuf, 0,
+ len = usb_maxpacket(dev, pipe, 0);
+ if (len > HID_BUFFER_SIZE)
+ len = HID_BUFFER_SIZE;
+ usb_fill_int_urb(hid->urbin, dev, pipe, hid->inbuf, len,
hid_irq_in, hid, endpoint->bInterval);
hid->urbin->transfer_dma = hid->inbuf_dma;
hid->urbin->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
--- diff/drivers/usb/input/hid-ff.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/input/hid-ff.c 2004-03-16 11:27:48.296845424 +0000
@@ -29,7 +29,7 @@
#include <linux/input.h>
-#define DEBUG
+#undef DEBUG
#include <linux/usb.h>
#include "hid.h"
--- diff/drivers/usb/input/hid-input.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/input/hid-input.c 2004-03-16 11:27:48.342838432 +0000
@@ -377,7 +377,8 @@ static void hidinput_configure_usage(str
set_bit(usage->type, input->evbit);
if ((usage->type == EV_REL)
- && (device->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK)
+ && (device->quirks & (HID_QUIRK_2WHEEL_MOUSE_HACK_BACK
+ | HID_QUIRK_2WHEEL_MOUSE_HACK_EXTRA))
&& (usage->code == REL_WHEEL)) {
set_bit(REL_HWHEEL, bit);
}
@@ -431,21 +432,22 @@ void hidinput_hid_event(struct hid_devic
input_regs(input, regs);
- if ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK)
- && (usage->code == BTN_BACK)) {
+ if (((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_EXTRA) && (usage->code == BTN_EXTRA))
+ || ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_BACK) && (usage->code == BTN_BACK))) {
if (value)
hid->quirks |= HID_QUIRK_2WHEEL_MOUSE_HACK_ON;
else
hid->quirks &= ~HID_QUIRK_2WHEEL_MOUSE_HACK_ON;
return;
}
+
if ((hid->quirks & HID_QUIRK_2WHEEL_MOUSE_HACK_ON)
&& (usage->code == REL_WHEEL)) {
input_event(input, usage->type, REL_HWHEEL, value);
return;
}
- if (usage->hat_min != usage->hat_max) {
+ if (usage->hat_min != usage->hat_max ) { /* FIXME: hat_max can be 0 and hat_min 1 */
value = (value - usage->hat_min) * 8 / (usage->hat_max - usage->hat_min + 1) + 1;
if (value < 0 || value > 8) value = 0;
input_event(input, usage->type, usage->code , hid_hat_to_axis[value].x);
@@ -484,7 +486,7 @@ void hidinput_hid_event(struct hid_devic
return;
}
- if((usage->type == EV_KEY) && (usage->code == 0)) /* Key 0 is "unassigned", not KEY_UKNOWN */
+ if((usage->type == EV_KEY) && (usage->code == 0)) /* Key 0 is "unassigned", not KEY_UNKNOWN */
return;
input_event(input, usage->type, usage->code, value);
@@ -567,8 +569,10 @@ int hidinput_connect(struct hid_device *
while (list != &report_enum->report_list) {
report = (struct hid_report *) list;
- if (!report->maxfield)
+ if (!report->maxfield) {
+ list = list->next;
continue;
+ }
if (!hidinput) {
hidinput = kmalloc(sizeof(*hidinput), GFP_KERNEL);
--- diff/drivers/usb/input/hid.h 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/input/hid.h 2004-03-16 11:27:48.341838584 +0000
@@ -201,15 +201,16 @@ struct hid_item {
* HID device quirks.
*/
-#define HID_QUIRK_INVERT 0x001
-#define HID_QUIRK_NOTOUCH 0x002
-#define HID_QUIRK_IGNORE 0x004
-#define HID_QUIRK_NOGET 0x008
-#define HID_QUIRK_HIDDEV 0x010
-#define HID_QUIRK_BADPAD 0x020
-#define HID_QUIRK_MULTI_INPUT 0x040
-#define HID_QUIRK_2WHEEL_MOUSE_HACK 0x080
-#define HID_QUIRK_2WHEEL_MOUSE_HACK_ON 0x100
+#define HID_QUIRK_INVERT 0x001
+#define HID_QUIRK_NOTOUCH 0x002
+#define HID_QUIRK_IGNORE 0x004
+#define HID_QUIRK_NOGET 0x008
+#define HID_QUIRK_HIDDEV 0x010
+#define HID_QUIRK_BADPAD 0x020
+#define HID_QUIRK_MULTI_INPUT 0x040
+#define HID_QUIRK_2WHEEL_MOUSE_HACK_BACK 0x080
+#define HID_QUIRK_2WHEEL_MOUSE_HACK_EXTRA 0x100
+#define HID_QUIRK_2WHEEL_MOUSE_HACK_ON 0x200
/*
* This is the global environment of the parser. This information is
@@ -309,7 +310,7 @@ struct hid_report_enum {
#define HID_REPORT_TYPES 3
-#define HID_BUFFER_SIZE 32
+#define HID_BUFFER_SIZE 64 /* use 64 for compatibility with all possible packetlen */
#define HID_CONTROL_FIFO_SIZE 256 /* to init devices with >100 reports */
#define HID_OUTPUT_FIFO_SIZE 64
--- diff/drivers/usb/input/hiddev.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/input/hiddev.c 2004-03-16 11:27:48.296845424 +0000
@@ -403,7 +403,8 @@ static int hiddev_ioctl(struct inode *in
struct hiddev_collection_info cinfo;
struct hiddev_report_info rinfo;
struct hiddev_field_info finfo;
- struct hiddev_usage_ref uref;
+ struct hiddev_usage_ref_multi uref_multi;
+ struct hiddev_usage_ref *uref = &uref_multi.uref;
struct hiddev_devinfo dinfo;
struct hid_report *report;
struct hid_field *field;
@@ -575,68 +576,98 @@ static int hiddev_ioctl(struct inode *in
return 0;
case HIDIOCGUCODE:
- if (copy_from_user(&uref, (void *) arg, sizeof(uref)))
+ if (copy_from_user(uref, (void *) arg, sizeof(*uref)))
return -EFAULT;
- rinfo.report_type = uref.report_type;
- rinfo.report_id = uref.report_id;
+ rinfo.report_type = uref->report_type;
+ rinfo.report_id = uref->report_id;
if ((report = hiddev_lookup_report(hid, &rinfo)) == NULL)
return -EINVAL;
- if (uref.field_index >= report->maxfield)
+ if (uref->field_index >= report->maxfield)
return -EINVAL;
- field = report->field[uref.field_index];
- if (uref.usage_index >= field->maxusage)
+ field = report->field[uref->field_index];
+ if (uref->usage_index >= field->maxusage)
return -EINVAL;
- uref.usage_code = field->usage[uref.usage_index].hid;
+ uref->usage_code = field->usage[uref->usage_index].hid;
- if (copy_to_user((void *) arg, &uref, sizeof(uref)))
+ if (copy_to_user((void *) arg, uref, sizeof(*uref)))
return -EFAULT;
return 0;
case HIDIOCGUSAGE:
case HIDIOCSUSAGE:
+ case HIDIOCGUSAGES:
+ case HIDIOCSUSAGES:
case HIDIOCGCOLLECTIONINDEX:
- if (copy_from_user(&uref, (void *) arg, sizeof(uref)))
- return -EFAULT;
+ if (cmd == HIDIOCGUSAGES || cmd == HIDIOCSUSAGES) {
+ if (copy_from_user(&uref_multi, (void *) arg,
+ sizeof(uref_multi)))
+ return -EFAULT;
+ } else {
+ if (copy_from_user(uref, (void *) arg, sizeof(*uref)))
+ return -EFAULT;
+ }
- if (cmd != HIDIOCGUSAGE && uref.report_type == HID_REPORT_TYPE_INPUT)
- return -EINVAL;
+ if (cmd != HIDIOCGUSAGE &&
+ cmd != HIDIOCGUSAGES &&
+ uref->report_type == HID_REPORT_TYPE_INPUT)
+ return -EINVAL;
- if (uref.report_id == HID_REPORT_ID_UNKNOWN) {
- field = hiddev_lookup_usage(hid, &uref);
+ if (uref->report_id == HID_REPORT_ID_UNKNOWN) {
+ field = hiddev_lookup_usage(hid, uref);
if (field == NULL)
return -EINVAL;
} else {
- rinfo.report_type = uref.report_type;
- rinfo.report_id = uref.report_id;
+ rinfo.report_type = uref->report_type;
+ rinfo.report_id = uref->report_id;
if ((report = hiddev_lookup_report(hid, &rinfo)) == NULL)
return -EINVAL;
- if (uref.field_index >= report->maxfield)
+ if (uref->field_index >= report->maxfield)
return -EINVAL;
- field = report->field[uref.field_index];
- if (uref.usage_index >= field->maxusage)
+ field = report->field[uref->field_index];
+ if (uref->usage_index >= field->maxusage)
return -EINVAL;
+
+ if (cmd == HIDIOCGUSAGES || cmd == HIDIOCSUSAGES) {
+ if (uref_multi.num_values >= HID_MAX_USAGES ||
+ uref->usage_index >= field->maxusage ||
+ (uref->usage_index + uref_multi.num_values) >= field->maxusage)
+ return -EINVAL;
+ }
}
switch (cmd) {
case HIDIOCGUSAGE:
- uref.value = field->value[uref.usage_index];
- if (copy_to_user((void *) arg, &uref, sizeof(uref)))
+ uref->value = field->value[uref->usage_index];
+ if (copy_to_user((void *) arg, uref, sizeof(*uref)))
return -EFAULT;
return 0;
case HIDIOCSUSAGE:
- field->value[uref.usage_index] = uref.value;
+ field->value[uref->usage_index] = uref->value;
return 0;
case HIDIOCGCOLLECTIONINDEX:
- return field->usage[uref.usage_index].collection_index;
+ return field->usage[uref->usage_index].collection_index;
+ case HIDIOCGUSAGES:
+ for (i = 0; i < uref_multi.num_values; i++)
+ uref_multi.values[i] =
+ field->value[uref->usage_index + i];
+ if (copy_to_user((void *) arg, &uref_multi,
+ sizeof(uref_multi)))
+ return -EFAULT;
+ return 0;
+ case HIDIOCSUSAGES:
+ for (i = 0; i < uref_multi.num_values; i++)
+ field->value[uref->usage_index + i] =
+ uref_multi.values[i];
+ return 0;
}
return 0;
--- diff/drivers/usb/input/kbtab.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/input/kbtab.c 2004-03-16 11:27:48.342838432 +0000
@@ -4,6 +4,8 @@
#include <linux/module.h>
#include <linux/init.h>
#include <linux/usb.h>
+#include <asm/unaligned.h>
+#include <asm/byteorder.h>
/*
* Version Information
@@ -65,8 +67,8 @@ static void kbtab_irq(struct urb *urb, s
goto exit;
}
- kbtab->x = (data[2] << 8) + data[1];
- kbtab->y = (data[4] << 8) + data[3];
+ kbtab->x = le16_to_cpu(get_unaligned((u16 *) &data[1]));
+ kbtab->y = le16_to_cpu(get_unaligned((u16 *) &data[3]));
kbtab->pressure = (data[5]);
@@ -74,12 +76,15 @@ static void kbtab_irq(struct urb *urb, s
input_report_abs(dev, ABS_X, kbtab->x);
input_report_abs(dev, ABS_Y, kbtab->y);
- /*input_report_abs(dev, ABS_PRESSURE, kbtab->pressure);*/
/*input_report_key(dev, BTN_TOUCH , data[0] & 0x01);*/
input_report_key(dev, BTN_RIGHT, data[0] & 0x02);
-
- input_report_key(dev, BTN_LEFT, (kbtab->pressure > kb_pressure_click) ? 1 : 0);
+
+ if( -1 == kb_pressure_click){
+ input_report_abs(dev, ABS_PRESSURE, kbtab->pressure);
+ } else {
+ input_report_key(dev, BTN_LEFT, (kbtab->pressure > kb_pressure_click) ? 1 : 0);
+ };
input_sync(dev);
@@ -105,8 +110,10 @@ static int kbtab_open(struct input_dev *
return 0;
kbtab->irq->dev = kbtab->usbdev;
- if (usb_submit_urb(kbtab->irq, GFP_KERNEL))
+ if (usb_submit_urb(kbtab->irq, GFP_KERNEL)) {
+ kbtab->open--;
return -EIO;
+ }
return 0;
}
@@ -130,7 +137,7 @@ static int kbtab_probe(struct usb_interf
return -ENOMEM;
memset(kbtab, 0, sizeof(struct kbtab));
- kbtab->data = usb_buffer_alloc(dev, 8, SLAB_ATOMIC, &kbtab->data_dma);
+ kbtab->data = usb_buffer_alloc(dev, 8, GFP_KERNEL, &kbtab->data_dma);
if (!kbtab->data) {
kfree(kbtab);
return -ENOMEM;
--- diff/drivers/usb/input/pid.c 2003-06-30 09:07:23.000000000 +0000
+++ source/drivers/usb/input/pid.c 2004-03-16 11:27:48.351837064 +0000
@@ -200,6 +200,7 @@ static int hid_pid_upload_effect(struct
break;
if ( id == FF_EFFECTS_MAX) {
+ spin_unlock_irqrestore(&pid_private->lock,flags);
// TEMP - We need to get ff_effects_max correctly first: || id >= dev->ff_effects_max) {
dev_dbg(&pid_private->hid->dev->dev, "Not enough device memory\n");
return -ENOMEM;
--- diff/drivers/usb/input/usbkbd.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/input/usbkbd.c 2004-03-16 11:27:48.352836912 +0000
@@ -240,7 +240,7 @@ static int usb_kbd_probe(struct usb_inte
char path[64];
char *buf;
- interface = &iface->altsetting[iface->act_altsetting];
+ interface = iface->cur_altsetting;
if (interface->desc.bNumEndpoints != 1)
return -ENODEV;
--- diff/drivers/usb/input/usbmouse.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/input/usbmouse.c 2004-03-16 11:27:48.353836760 +0000
@@ -131,7 +131,7 @@ static int usb_mouse_probe(struct usb_in
char path[64];
char *buf;
- interface = &intf->altsetting[intf->act_altsetting];
+ interface = intf->cur_altsetting;
if (interface->desc.bNumEndpoints != 1)
return -ENODEV;
--- diff/drivers/usb/input/wacom.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/input/wacom.c 2004-03-16 11:27:48.355836456 +0000
@@ -1,14 +1,15 @@
/*
* USB Wacom Graphire and Wacom Intuos tablet support
*
- * Copyright (c) 2000-2002 Vojtech Pavlik <vojtech@ucw.cz>
+ * Copyright (c) 2000-2004 Vojtech Pavlik <vojtech@ucw.cz>
* Copyright (c) 2000 Andreas Bach Aaen <abach@stofanet.dk>
* Copyright (c) 2000 Clifford Wolf <clifford@clifford.at>
* Copyright (c) 2000 Sam Mosel <sam.mosel@computer.org>
* Copyright (c) 2000 James E. Blair <corvus@gnu.org>
* Copyright (c) 2000 Daniel Egger <egger@suse.de>
* Copyright (c) 2001 Frederic Lepied <flepied@mandrakesoft.com>
- * Copyright (c) 2002 Ping Cheng <pingc@wacom.com>
+ * Copyright (c) 2004 Panagiotis Issaris <panagiotis.issaris@mech.kuleuven.ac.be>
+ * Copyright (c) 2002-2004 Ping Cheng <pingc@wacom.com>
*
* ChangeLog:
* v0.1 (vp) - Initial release
@@ -48,6 +49,8 @@
* v1.30 (vp) - Merge 2.4 and 2.5 drivers
* - Since 2.5 now has input_sync(), remove MSC_SERIAL abuse
* - Cleanups here and there
+ * v1.30.1 (pi) - Added Graphire3 support
+ * v1.40 (pc) - Add support for several new devices, fix eraser reporting, ...
*/
/*
@@ -63,6 +66,8 @@
#include <linux/module.h>
#include <linux/init.h>
#include <linux/usb.h>
+#include <asm/unaligned.h>
+#include <asm/byteorder.h>
/*
* Version Information
@@ -106,7 +111,7 @@ struct wacom {
static int usb_set_report(struct usb_interface *intf, unsigned char type,
unsigned char id, void *buf, int size)
{
- return usb_control_msg(interface_to_usbdev(intf),
+ return usb_control_msg(interface_to_usbdev(intf),
usb_sndctrlpipe(interface_to_usbdev(intf), 0),
USB_REQ_SET_REPORT, USB_TYPE_CLASS | USB_RECIP_INTERFACE,
(type << 8) + id, intf->altsetting[0].desc.bInterfaceNumber,
@@ -137,14 +142,12 @@ static void wacom_pl_irq(struct urb *urb
}
if (data[0] != 2)
- dbg("received unknown report #%d", data[0]);
+ dbg("wacom_pl_irq: received unknown report #%d", data[0]);
prox = data[1] & 0x40;
input_regs(dev, regs);
- input_report_key(dev, BTN_TOOL_PEN, prox);
-
if (prox) {
pressure = (signed char)((data[7] << 1) | ((data[4] >> 2) & 1));
@@ -152,15 +155,103 @@ static void wacom_pl_irq(struct urb *urb
pressure = (pressure << 1) | ((data[4] >> 6) & 1);
pressure += (wacom->features->pressure_max + 1) / 2;
+ /*
+ * if going from out of proximity into proximity select between the eraser
+ * and the pen based on the state of the stylus2 button, choose eraser if
+ * pressed else choose pen. if not a proximity change from out to in, send
+ * an out of proximity for previous tool then a in for new tool.
+ */
+ if (!wacom->tool[0]) {
+ /* Going into proximity select tool */
+ wacom->tool[1] = (data[4] & 0x20)? BTN_TOOL_RUBBER : BTN_TOOL_PEN;
+ }
+ else {
+ /* was entered with stylus2 pressed */
+ if (wacom->tool[1] == BTN_TOOL_RUBBER && !(data[4] & 0x20) ) {
+ /* report out proximity for previous tool */
+ input_report_key(dev, wacom->tool[1], 0);
+ input_sync(dev);
+ wacom->tool[1] = BTN_TOOL_PEN;
+ goto exit;
+ }
+ }
+ if (wacom->tool[1] != BTN_TOOL_RUBBER) {
+ /* Unknown tool selected default to pen tool */
+ wacom->tool[1] = BTN_TOOL_PEN;
+ }
+ input_report_key(dev, wacom->tool[1], prox); /* report in proximity for tool */
input_report_abs(dev, ABS_X, data[3] | ((__u32)data[2] << 7) | ((__u32)(data[1] & 0x03) << 14));
input_report_abs(dev, ABS_Y, data[6] | ((__u32)data[5] << 7) | ((__u32)(data[4] & 0x03) << 14));
input_report_abs(dev, ABS_PRESSURE, pressure);
input_report_key(dev, BTN_TOUCH, data[4] & 0x08);
input_report_key(dev, BTN_STYLUS, data[4] & 0x10);
- input_report_key(dev, BTN_STYLUS2, data[4] & 0x20);
+ /* Only allow the stylus2 button to be reported for the pen tool. */
+ input_report_key(dev, BTN_STYLUS2, (wacom->tool[1] == BTN_TOOL_PEN) && (data[4] & 0x20));
}
-
+ else {
+ /* report proximity-out of a (valid) tool */
+ if (wacom->tool[1] != BTN_TOOL_RUBBER) {
+ /* Unknown tool selected default to pen tool */
+ wacom->tool[1] = BTN_TOOL_PEN;
+ }
+ input_report_key(dev, wacom->tool[1], prox);
+ }
+
+ wacom->tool[0] = prox; /* Save proximity state */
+ input_sync(dev);
+
+exit:
+ retval = usb_submit_urb (urb, GFP_ATOMIC);
+ if (retval)
+ err ("%s - usb_submit_urb failed with result %d",
+ __FUNCTION__, retval);
+}
+
+static void wacom_ptu_irq(struct urb *urb, struct pt_regs *regs)
+{
+ struct wacom *wacom = urb->context;
+ unsigned char *data = wacom->data;
+ struct input_dev *dev = &wacom->dev;
+ int retval;
+
+ switch (urb->status) {
+ case 0:
+ /* success */
+ break;
+ case -ECONNRESET:
+ case -ENOENT:
+ case -ESHUTDOWN:
+ /* this urb is terminated, clean up */
+ dbg("%s - urb shutting down with status: %d", __FUNCTION__, urb->status);
+ return;
+ default:
+ dbg("%s - nonzero urb status received: %d", __FUNCTION__, urb->status);
+ goto exit;
+ }
+
+ if (data[0] != 2)
+ {
+ printk(KERN_INFO "wacom_ptu_irq: received unknown report #%d\n", data[0]);
+ }
+
+ input_regs(dev, regs);
+ if (data[1] & 0x04)
+ {
+ input_report_key(dev, BTN_TOOL_RUBBER, data[1] & 0x20);
+ input_report_key(dev, BTN_TOUCH, data[1] & 0x08);
+ }
+ else
+ {
+ input_report_key(dev, BTN_TOOL_PEN, data[1] & 0x20);
+ input_report_key(dev, BTN_TOUCH, data[1] & 0x01);
+ }
+ input_report_abs(dev, ABS_X, data[3] << 8 | data[2]);
+ input_report_abs(dev, ABS_Y, data[5] << 8 | data[4]);
+ input_report_abs(dev, ABS_PRESSURE, (data[6]|data[7] << 8));
+ input_report_key(dev, BTN_STYLUS, data[1] & 0x02);
+ input_report_key(dev, BTN_STYLUS2, data[1] & 0x10);
+
input_sync(dev);
exit:
@@ -194,8 +285,8 @@ static void wacom_penpartner_irq(struct
input_regs(dev, regs);
input_report_key(dev, BTN_TOOL_PEN, 1);
- input_report_abs(dev, ABS_X, data[2] << 8 | data[1]);
- input_report_abs(dev, ABS_Y, data[4] << 8 | data[3]);
+ input_report_abs(dev, ABS_X, le16_to_cpu(get_unaligned((u16 *) &data[1])));
+ input_report_abs(dev, ABS_Y, le16_to_cpu(get_unaligned((u16 *) &data[3])));
input_report_abs(dev, ABS_PRESSURE, (signed char)data[6] + 127);
input_report_key(dev, BTN_TOUCH, ((signed char)data[6] > -80) && !(data[5] & 0x20));
input_report_key(dev, BTN_STYLUS, (data[5] & 0x40));
@@ -231,11 +322,15 @@ static void wacom_graphire_irq(struct ur
goto exit;
}
+ /* check if we can handle the data */
+ if (data[0] == 99)
+ goto exit;
+
if (data[0] != 2)
- dbg("received unknown report #%d", data[0]);
+ dbg("wacom_graphire_irq: received unknown report #%d", data[0]);
- x = data[2] | ((__u32)data[3] << 8);
- y = data[4] | ((__u32)data[5] << 8);
+ x = le16_to_cpu(*(u16 *) &data[2]);
+ y = le16_to_cpu(*(u16 *) &data[4]);
input_regs(dev, regs);
@@ -249,13 +344,16 @@ static void wacom_graphire_irq(struct ur
input_report_key(dev, BTN_TOOL_RUBBER, data[1] & 0x80);
break;
- case 2: /* Mouse */
+ case 2: /* Mouse with wheel */
+ input_report_key(dev, BTN_MIDDLE, data[1] & 0x04);
+ input_report_rel(dev, REL_WHEEL, (signed char) data[6]);
+ /* fall through */
+
+ case 3: /* Mouse without wheel */
input_report_key(dev, BTN_TOOL_MOUSE, data[7] > 24);
input_report_key(dev, BTN_LEFT, data[1] & 0x01);
input_report_key(dev, BTN_RIGHT, data[1] & 0x02);
- input_report_key(dev, BTN_MIDDLE, data[1] & 0x04);
input_report_abs(dev, ABS_DISTANCE, data[7]);
- input_report_rel(dev, REL_WHEEL, (signed char) data[6]);
input_report_abs(dev, ABS_X, x);
input_report_abs(dev, ABS_Y, y);
@@ -269,7 +367,7 @@ static void wacom_graphire_irq(struct ur
input_report_abs(dev, ABS_Y, y);
}
- input_report_abs(dev, ABS_PRESSURE, data[6] | ((__u32)data[7] << 8));
+ input_report_abs(dev, ABS_PRESSURE, le16_to_cpu(*(u16 *) &data[6]));
input_report_key(dev, BTN_TOUCH, data[1] & 0x01);
input_report_key(dev, BTN_STYLUS, data[1] & 0x02);
input_report_key(dev, BTN_STYLUS2, data[1] & 0x04);
@@ -308,7 +406,7 @@ static void wacom_intuos_irq(struct urb
}
if (data[0] != 2)
- dbg("received unknown report #%d", data[0]);
+ dbg("wacom_intuos_irq: received unknown report #%d", data[0]);
input_regs(dev, regs);
@@ -317,18 +415,18 @@ static void wacom_intuos_irq(struct urb
if ((data[1] & 0xfc) == 0xc0) { /* Enter report */
- wacom->serial[idx] = ((__u32)(data[3] & 0x0f) << 4) + /* serial number of the tool */
- ((__u32)data[4] << 16) + ((__u32)data[5] << 12) +
+ wacom->serial[idx] = ((__u32)(data[3] & 0x0f) << 28) + /* serial number of the tool */
+ ((__u32)data[4] << 20) + ((__u32)data[5] << 12) +
((__u32)data[6] << 4) + (data[7] >> 4);
switch (((__u32)data[2] << 4) | (data[3] >> 4)) {
- case 0x832:
+ case 0x812:
case 0x012: wacom->tool[idx] = BTN_TOOL_PENCIL; break; /* Inking pen */
case 0x822:
case 0x842:
case 0x852:
case 0x022: wacom->tool[idx] = BTN_TOOL_PEN; break; /* Pen */
- case 0x812:
+ case 0x832:
case 0x032: wacom->tool[idx] = BTN_TOOL_BRUSH; break; /* Stroke pen */
case 0x007:
case 0x09c:
@@ -337,7 +435,10 @@ static void wacom_intuos_irq(struct urb
case 0x82a:
case 0x85a:
case 0x91a:
+ case 0xd1a:
case 0x0fa: wacom->tool[idx] = BTN_TOOL_RUBBER; break; /* Eraser */
+ case 0xd12:
+ case 0x912:
case 0x112: wacom->tool[idx] = BTN_TOOL_AIRBRUSH; break; /* Airbrush */
default: wacom->tool[idx] = BTN_TOOL_PEN; break; /* Unknown tool */
}
@@ -350,13 +451,14 @@ static void wacom_intuos_irq(struct urb
if ((data[1] & 0xfe) == 0x80) { /* Exit report */
input_report_key(dev, wacom->tool[idx], 0);
+ input_event(dev, EV_MSC, MSC_SERIAL, wacom->serial[idx]);
input_sync(dev);
goto exit;
}
- input_report_abs(dev, ABS_X, ((__u32)data[2] << 8) | data[3]);
- input_report_abs(dev, ABS_Y, ((__u32)data[4] << 8) | data[5]);
- input_report_abs(dev, ABS_DISTANCE, data[9] >> 4);
+ input_report_abs(dev, ABS_X, be16_to_cpu(*(u16 *) &data[2]));
+ input_report_abs(dev, ABS_Y, be16_to_cpu(*(u16 *) &data[4]));
+ input_report_abs(dev, ABS_DISTANCE, data[9]);
if ((data[1] & 0xb8) == 0xa0) { /* general pen packet */
input_report_abs(dev, ABS_PRESSURE, t = ((__u32)data[6] << 2) | ((data[7] >> 6) & 3));
@@ -378,8 +480,8 @@ static void wacom_intuos_irq(struct urb
if (data[1] & 0x02) { /* Rotation packet */
input_report_abs(dev, ABS_RZ, (data[7] & 0x20) ?
- ((__u32)data[6] << 2) | ((data[7] >> 6) & 3):
- (-(((__u32)data[6] << 2) | ((data[7] >> 6) & 3))) - 1);
+ ((__u32)data[6] << 3) | ((data[7] >> 5) & 7):
+ (-(((__u32)data[6] << 3) | ((data[7] >> 5) & 7))) - 1);
} else {
@@ -391,17 +493,17 @@ static void wacom_intuos_irq(struct urb
input_report_key(dev, BTN_SIDE, data[8] & 0x20);
input_report_key(dev, BTN_EXTRA, data[8] & 0x10);
- input_report_abs(dev, ABS_THROTTLE, (data[8] & 0x08) ?
+ input_report_abs(dev, ABS_THROTTLE, -((data[8] & 0x08) ?
((__u32)data[6] << 2) | ((data[7] >> 6) & 3) :
- -((__u32)data[6] << 2) | ((data[7] >> 6) & 3));
+ -((__u32)data[6] << 2) | ((data[7] >> 6) & 3)));
} else {
if (wacom->tool[idx] == BTN_TOOL_MOUSE) { /* 2D mouse packets */
input_report_key(dev, BTN_LEFT, data[8] & 0x04);
input_report_key(dev, BTN_MIDDLE, data[8] & 0x08);
input_report_key(dev, BTN_RIGHT, data[8] & 0x10);
- input_report_abs(dev, REL_WHEEL,
- ((__u32)(data[8] & 0x01) - (__u32)((data[8] & 0x02) >> 1)));
+ input_report_rel(dev, REL_WHEEL,
+ (-(__u32)(data[8] & 0x01) + (__u32)((data[8] & 0x02) >> 1)));
}
else { /* Lens cursor packets */
input_report_key(dev, BTN_LEFT, data[8] & 0x01);
@@ -414,6 +516,8 @@ static void wacom_intuos_irq(struct urb
}
}
+ input_report_key(dev, wacom->tool[idx], 1);
+ input_event(dev, EV_MSC, MSC_SERIAL, wacom->serial[idx]);
input_sync(dev);
exit:
@@ -429,22 +533,26 @@ struct wacom_features wacom_features[] =
{ "Wacom Graphire2 4x5", 8, 10206, 7422, 511, 32, 1, wacom_graphire_irq },
{ "Wacom Graphire2 5x7", 8, 13918, 10206, 511, 32, 1, wacom_graphire_irq },
{ "Wacom Graphire3", 8, 10208, 7424, 511, 32, 1, wacom_graphire_irq },
- { "Wacom Intuos 4x5", 10, 12700, 10360, 1023, 15, 2, wacom_intuos_irq },
- { "Wacom Intuos 6x8", 10, 20600, 16450, 1023, 15, 2, wacom_intuos_irq },
- { "Wacom Intuos 9x12", 10, 30670, 24130, 1023, 15, 2, wacom_intuos_irq },
- { "Wacom Intuos 12x12", 10, 30670, 31040, 1023, 15, 2, wacom_intuos_irq },
- { "Wacom Intuos 12x18", 10, 45860, 31040, 1023, 15, 2, wacom_intuos_irq },
+ { "Wacom Graphire3 6x8", 8, 16704, 12064, 511, 32, 1, wacom_graphire_irq },
+ { "Wacom Intuos 4x5", 10, 12700, 10600, 1023, 15, 2, wacom_intuos_irq },
+ { "Wacom Intuos 6x8", 10, 20320, 16240, 1023, 15, 2, wacom_intuos_irq },
+ { "Wacom Intuos 9x12", 10, 30480, 24060, 1023, 15, 2, wacom_intuos_irq },
+ { "Wacom Intuos 12x12", 10, 30480, 31680, 1023, 15, 2, wacom_intuos_irq },
+ { "Wacom Intuos 12x18", 10, 45720, 31680, 1023, 15, 2, wacom_intuos_irq },
{ "Wacom PL400", 8, 5408, 4056, 255, 32, 3, wacom_pl_irq },
{ "Wacom PL500", 8, 6144, 4608, 255, 32, 3, wacom_pl_irq },
{ "Wacom PL600", 8, 6126, 4604, 255, 32, 3, wacom_pl_irq },
{ "Wacom PL600SX", 8, 6260, 5016, 255, 32, 3, wacom_pl_irq },
{ "Wacom PL550", 8, 6144, 4608, 511, 32, 3, wacom_pl_irq },
{ "Wacom PL800", 8, 7220, 5780, 511, 32, 3, wacom_pl_irq },
- { "Wacom Intuos2 4x5", 10, 12700, 10360, 1023, 15, 2, wacom_intuos_irq },
- { "Wacom Intuos2 6x8", 10, 20600, 16450, 1023, 15, 2, wacom_intuos_irq },
- { "Wacom Intuos2 9x12", 10, 30670, 24130, 1023, 15, 2, wacom_intuos_irq },
- { "Wacom Intuos2 12x12", 10, 30670, 31040, 1023, 15, 2, wacom_intuos_irq },
- { "Wacom Intuos2 12x18", 10, 45860, 31040, 1023, 15, 2, wacom_intuos_irq },
+ { "Wacom Intuos2 4x5", 10, 12700, 10600, 1023, 15, 2, wacom_intuos_irq },
+ { "Wacom Intuos2 6x8", 10, 20320, 16240, 1023, 15, 2, wacom_intuos_irq },
+ { "Wacom Intuos2 9x12", 10, 30480, 24060, 1023, 15, 2, wacom_intuos_irq },
+ { "Wacom Intuos2 12x12", 10, 30480, 31680, 1023, 15, 2, wacom_intuos_irq },
+ { "Wacom Intuos2 12x18", 10, 45720, 31680, 1023, 15, 2, wacom_intuos_irq },
+ { "Wacom Volito", 8, 5104, 3712, 511, 32, 1, wacom_graphire_irq },
+ { "Wacom Cintiq Partner",8, 20480, 15360, 511, 32, 3, wacom_ptu_irq },
+ { "Wacom Intuos2 6x8", 10, 20320, 16240, 1023, 15, 2, wacom_intuos_irq },
{ }
};
@@ -454,6 +562,7 @@ struct usb_device_id wacom_ids[] = {
{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x11) },
{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x12) },
{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x13) },
+ { USB_DEVICE(USB_VENDOR_ID_WACOM, 0x14) },
{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x20) },
{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x21) },
{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x22) },
@@ -470,6 +579,9 @@ struct usb_device_id wacom_ids[] = {
{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x43) },
{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x44) },
{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x45) },
+ { USB_DEVICE(USB_VENDOR_ID_WACOM, 0x60) },
+ { USB_DEVICE(USB_VENDOR_ID_WACOM, 0x03) },
+ { USB_DEVICE(USB_VENDOR_ID_WACOM, 0x47) },
{ }
};
@@ -483,8 +595,10 @@ static int wacom_open(struct input_dev *
return 0;
wacom->irq->dev = wacom->usbdev;
- if (usb_submit_urb(wacom->irq, GFP_KERNEL))
+ if (usb_submit_urb(wacom->irq, GFP_KERNEL)) {
+ wacom->open--;
return -EIO;
+ }
return 0;
}
@@ -509,7 +623,7 @@ static int wacom_probe(struct usb_interf
return -ENOMEM;
memset(wacom, 0, sizeof(struct wacom));
- wacom->data = usb_buffer_alloc(dev, 10, SLAB_ATOMIC, &wacom->data_dma);
+ wacom->data = usb_buffer_alloc(dev, 10, GFP_KERNEL, &wacom->data_dma);
if (!wacom->data) {
kfree(wacom);
return -ENOMEM;
@@ -538,8 +652,9 @@ static int wacom_probe(struct usb_interf
break;
case 2:
- wacom->dev.evbit[0] |= BIT(EV_MSC);
+ wacom->dev.evbit[0] |= BIT(EV_MSC) | BIT(EV_REL);
wacom->dev.mscbit[0] |= BIT(MSC_SERIAL);
+ wacom->dev.relbit[0] |= BIT(REL_WHEEL);
wacom->dev.keybit[LONG(BTN_LEFT)] |= BIT(BTN_LEFT) | BIT(BTN_RIGHT) | BIT(BTN_MIDDLE) | BIT(BTN_SIDE) | BIT(BTN_EXTRA);
wacom->dev.keybit[LONG(BTN_DIGI)] |= BIT(BTN_TOOL_RUBBER) | BIT(BTN_TOOL_MOUSE) | BIT(BTN_TOOL_BRUSH)
| BIT(BTN_TOOL_PENCIL) | BIT(BTN_TOOL_AIRBRUSH) | BIT(BTN_TOOL_LENS) | BIT(BTN_STYLUS2);
--- diff/drivers/usb/misc/Kconfig 2004-02-18 08:54:12.000000000 +0000
+++ source/drivers/usb/misc/Kconfig 2004-03-16 11:27:48.358836000 +0000
@@ -86,17 +86,6 @@ config USB_LEGOTOWER
a module, say M here and read
<file:Documentation/kbuild/modules.txt>.
-config USB_BRLVGER
- tristate "Tieman Voyager USB Braille display support (EXPERIMENTAL)"
- depends on USB && EXPERIMENTAL
- help
- Say Y here if you want to use the Voyager USB Braille display from
- Tieman. See <file:Documentation/usb/brlvger.txt> for more
- information.
-
- To compile this driver as a module, choose M here: the
- module will be called brlvger.
-
config USB_LCD
tristate "USB LCD driver support"
depends on USB
--- diff/drivers/usb/misc/Makefile 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/misc/Makefile 2004-03-16 11:27:48.359835848 +0000
@@ -4,7 +4,6 @@
#
obj-$(CONFIG_USB_AUERSWALD) += auerswald.o
-obj-$(CONFIG_USB_BRLVGER) += brlvger.o
obj-$(CONFIG_USB_EMI62) += emi62.o
obj-$(CONFIG_USB_EMI26) += emi26.o
obj-$(CONFIG_USB_LCD) += usblcd.o
--- diff/drivers/usb/misc/usbtest.c 2004-03-11 10:20:28.000000000 +0000
+++ source/drivers/usb/misc/usbtest.c 2004-03-16 11:27:48.361835544 +0000
@@ -149,8 +149,6 @@ get_endpoints (struct usbtest_dev *dev,
if (!out)
out = e;
}
- if (in && out)
- goto found;
continue;
try_iso:
if (e->desc.bEndpointAddress & USB_DIR_IN) {
@@ -160,9 +158,9 @@ try_iso:
if (!iso_out)
iso_out = e;
}
- if (iso_in && iso_out)
- goto found;
}
+ if ((in && out) || (iso_in && iso_out))
+ goto found;
}
return -EINVAL;
@@ -181,7 +179,8 @@ found:
in->desc.bEndpointAddress & USB_ENDPOINT_NUMBER_MASK);
dev->out_pipe = usb_sndbulkpipe (udev,
out->desc.bEndpointAddress & USB_ENDPOINT_NUMBER_MASK);
- } else if (iso_in) {
+ }
+ if (iso_in) {
dev->iso_in = &iso_in->desc;
dev->in_iso_pipe = usb_rcvisocpipe (udev,
iso_in->desc.bEndpointAddress
@@ -211,7 +210,7 @@ static void simple_callback (struct urb
static struct urb *simple_alloc_urb (
struct usb_device *udev,
int pipe,
- long bytes
+ unsigned long bytes
)
{
struct urb *urb;
@@ -490,7 +489,7 @@ static int set_altsetting (struct usbtes
struct usb_interface *iface = dev->intf;
struct usb_device *udev;
- if (alternate < 0 || alternate >= iface->num_altsetting)
+ if (alternate < 0 || alternate >= 256)
return -EINVAL;
udev = interface_to_usbdev (iface);
@@ -556,23 +555,19 @@ static int ch9_postconfig (struct usbtes
{
struct usb_interface *iface = dev->intf;
struct usb_device *udev = interface_to_usbdev (iface);
- int i, retval;
+ int i, alt, retval;
/* [9.2.3] if there's more than one altsetting, we need to be able to
* set and get each one. mostly trusts the descriptors from usbcore.
*/
for (i = 0; i < iface->num_altsetting; i++) {
- /* 9.2.3 constrains the range here, and Linux ensures
- * they're ordered meaningfully in this array
- */
- if (iface->altsetting [i].desc.bAlternateSetting != i) {
+ /* 9.2.3 constrains the range here */
+ alt = iface->altsetting [i].desc.bAlternateSetting;
+ if (alt < 0 || alt >= iface->num_altsetting) {
dev_dbg (&iface->dev,
"invalid alt [%d].bAltSetting = %d\n",
- i,
- iface->altsetting [i].desc
- .bAlternateSetting);
- return -EDOM;
+ i, alt);
}
/* [real world] get/set unimplemented if there's only one */
@@ -580,18 +575,18 @@ static int ch9_postconfig (struct usbtes
continue;
/* [9.4.10] set_interface */
- retval = set_altsetting (dev, i);
+ retval = set_altsetting (dev, alt);
if (retval) {
dev_dbg (&iface->dev, "can't set_interface = %d, %d\n",
- i, retval);
+ alt, retval);
return retval;
}
/* [9.4.4] get_interface always works */
retval = get_altsetting (dev);
- if (retval != i) {
+ if (retval != alt) {
dev_dbg (&iface->dev, "get alt should be %d, was %d\n",
- i, retval);
+ alt, retval);
return (retval < 0) ? retval : -EDOM;
}
@@ -916,7 +911,7 @@ test_ctrl_queue (struct usbtest_dev *dev
req.wValue = cpu_to_le16 (USB_DT_INTERFACE << 8);
// interface == 0
len = sizeof (struct usb_interface_descriptor);
- expected = -EPIPE;
+ expected = EPIPE;
break;
// NOTE: two consecutive stalls in the queue here.
// that tests fault recovery a bit more aggressively.
@@ -945,7 +940,7 @@ test_ctrl_queue (struct usbtest_dev *dev
req.wValue = cpu_to_le16 (USB_DT_ENDPOINT << 8);
// endpoint == 0
len = sizeof (struct usb_interface_descriptor);
- expected = -EPIPE;
+ expected = EPIPE;
break;
// NOTE: sometimes even a third fault in the queue!
case 12: // get string 0 descriptor (MAY STALL)
@@ -1072,7 +1067,7 @@ static int unlink1 (struct usbtest_dev *
* due to errors, or is just NAKing requests.
*/
if ((retval = usb_submit_urb (urb, SLAB_KERNEL)) != 0) {
- dbg ("submit/unlink fail %d", retval);
+ dev_dbg (&dev->intf->dev, "submit fail %d\n", retval);
return retval;
}
@@ -1087,18 +1082,22 @@ retry:
* "normal" drivers would prevent resubmission, but
* since we're testing unlink paths, we can't.
*/
- dbg ("unlink retry");
+ dev_dbg (&dev->intf->dev, "unlink retry\n");
goto retry;
}
if (!(retval == 0 || retval == -EINPROGRESS)) {
- dbg ("submit/unlink fail %d", retval);
+ dev_dbg (&dev->intf->dev, "unlink fail %d\n", retval);
return retval;
}
wait_for_completion (&completion);
retval = urb->status;
simple_free_urb (urb);
- return retval;
+
+ if (async)
+ return (retval != -ECONNRESET) ? -ECONNRESET : 0;
+ else
+ return (retval != -ENOENT) ? -ENOENT : 0;
}
static int unlink_simple (struct usbtest_dev *dev, int pipe, int len)
@@ -1723,7 +1722,8 @@ usbtest_ioctl (struct usb_interface *int
retval = unlink_simple (dev, dev->in_pipe,
param->length);
if (retval)
- dbg ("unlink reads failed, iterations left %d", i);
+ dev_dbg (&intf->dev, "unlink reads failed %d, "
+ "iterations left %d\n", retval, i);
break;
case 12:
if (dev->out_pipe == 0 || !param->length)
@@ -1735,7 +1735,8 @@ usbtest_ioctl (struct usb_interface *int
retval = unlink_simple (dev, dev->out_pipe,
param->length);
if (retval)
- dbg ("unlink writes failed, iterations left %d", i);
+ dev_dbg (&intf->dev, "unlink writes failed %d, "
+ "iterations left %d\n", retval, i);
break;
/* ep halt tests */
@@ -1965,7 +1966,10 @@ static struct usbtest_info fw_info = {
.name = "usb test device",
.ep_in = 2,
.ep_out = 2,
- .alt = 0,
+ .alt = 1,
+ .autoconf = 1, // iso and ctrl_out need autoconf
+ .ctrl_out = 1,
+ .iso = 1, // iso_ep's are #8 in/out
};
/* peripheral running Linux and 'zero.c' test firmware, or
--- diff/drivers/usb/net/Kconfig 2004-02-18 08:54:12.000000000 +0000
+++ source/drivers/usb/net/Kconfig 2004-03-16 11:27:48.361835544 +0000
@@ -131,6 +131,14 @@ config USB_USBNET
comment "USB Host-to-Host Cables"
depends on USB_USBNET
+config USB_ALI_M5632
+ boolean "ALi M5632 based 'USB 2.0 Data Link' cables"
+ depends on USB_USBNET
+ default y
+ help
+ Choose this option if you're using a host-to-host cable
+ based on this design, which supports USB 2.0 high speed.
+
config USB_AN2720
boolean "AnchorChips 2720 based cables (Xircom PGUNET, ...)"
depends on USB_USBNET
--- diff/drivers/usb/net/pegasus.h 2004-01-19 10:22:59.000000000 +0000
+++ source/drivers/usb/net/pegasus.h 2004-03-16 11:27:48.362835392 +0000
@@ -128,6 +128,7 @@ struct usb_eth_dev {
#define VENDOR_DLINK 0x2001
#define VENDOR_ELCON 0x0db7
#define VENDOR_ELSA 0x05cc
+#define VENDOR_GIGABYTE 0x1044
#define VENDOR_HAWKING 0x0e66
#define VENDOR_HP 0x03f0
#define VENDOR_IODATA 0x04bb
@@ -178,6 +179,9 @@ PEGASUS_DEV( "ADMtek ADM8511 \"Pegasus I
PEGASUS_DEV( "ADMtek ADM8513 \"Pegasus II\" USB Ethernet",
VENDOR_ADMTEK, 0x8513,
DEFAULT_GPIO_RESET | PEGASUS_II )
+PEGASUS_DEV( "ADMtek ADM8515 \"Pegasus II\" USB-2.0 Ethernet",
+ VENDOR_ADMTEK, 0x8515,
+ DEFAULT_GPIO_RESET | PEGASUS_II )
PEGASUS_DEV( "ADMtek AN986 \"Pegasus\" USB Ethernet (evaluation board)",
VENDOR_ADMTEK, 0x0986,
DEFAULT_GPIO_RESET | HAS_HOME_PNA )
@@ -223,6 +227,8 @@ PEGASUS_DEV( "EasiDock Ethernet", VENDOR
DEFAULT_GPIO_RESET )
PEGASUS_DEV( "Elsa Micolink USB2Ethernet", VENDOR_ELSA, 0x3000,
DEFAULT_GPIO_RESET )
+PEGASUS_DEV( "GIGABYTE GN-BR402W Wireless Router", VENDOR_GIGABYTE, 0x8002,
+ DEFAULT_GPIO_RESET )
PEGASUS_DEV( "Hawking UF100 10/100 Ethernet", VENDOR_HAWKING, 0x400c,
DEFAULT_GPIO_RESET | PEGASUS_II )
PEGASUS_DEV( "HP hn210c Ethernet USB", VENDOR_HP, 0x811c,
@@ -263,6 +269,8 @@ PEGASUS_DEV( "NETGEAR FA101", VENDOR_NET
DEFAULT_GPIO_RESET | PEGASUS_II )
PEGASUS_DEV( "OCT Inc.", VENDOR_OCT, 0x0109,
DEFAULT_GPIO_RESET | PEGASUS_II )
+PEGASUS_DEV( "OCT USB TO Ethernet", VENDOR_OCT, 0x0901,
+ DEFAULT_GPIO_RESET | PEGASUS_II )
PEGASUS_DEV( "smartNIC 2 PnP Adapter", VENDOR_SMARTBRIDGES, 0x0003,
DEFAULT_GPIO_RESET | PEGASUS_II )
PEGASUS_DEV( "SMC 202 USB Ethernet", VENDOR_SMC, 0x0200,
--- diff/drivers/usb/net/usbnet.c 2004-03-11 10:20:28.000000000 +0000
+++ source/drivers/usb/net/usbnet.c 2004-03-16 11:27:48.365834936 +0000
@@ -384,6 +384,23 @@ static void skb_return (struct usbnet *d
}
+#ifdef CONFIG_USB_ALI_M5632
+#define HAVE_HARDWARE
+
+/*-------------------------------------------------------------------------
+ *
+ * ALi M5632 driver ... does high speed
+ *
+ *-------------------------------------------------------------------------*/
+
+static const struct driver_info ali_m5632_info = {
+ .description = "ALi M5632",
+};
+
+
+#endif
+
+
#ifdef CONFIG_USB_AN2720
#define HAVE_HARDWARE
@@ -3009,7 +3026,7 @@ usbnet_probe (struct usb_interface *udev
return -ENODEV;
}
xdev = interface_to_usbdev (udev);
- interface = &udev->altsetting [udev->act_altsetting];
+ interface = udev->cur_altsetting;
usb_get_dev (xdev);
@@ -3133,6 +3150,13 @@ out:
static const struct usb_device_id products [] = {
+#ifdef CONFIG_USB_ALI_M5632
+{
+ USB_DEVICE (0x0402, 0x5632), // ALi defaults
+ .driver_info = (unsigned long) &ali_m5632_info,
+},
+#endif
+
#ifdef CONFIG_USB_AN2720
{
USB_DEVICE (0x0547, 0x2720), // AnchorChips defaults
@@ -3314,6 +3338,15 @@ static const struct usb_device_id produc
.bInterfaceSubClass = 0x0a,
.bInterfaceProtocol = 0x00,
.driver_info = (unsigned long) &zaurus_pxa_info,
+}, {
+ .match_flags = USB_DEVICE_ID_MATCH_INT_INFO
+ | USB_DEVICE_ID_MATCH_DEVICE,
+ .idVendor = 0x04DD,
+ .idProduct = 0x9050, /* C-860 */
+ .bInterfaceClass = 0x02,
+ .bInterfaceSubClass = 0x0a,
+ .bInterfaceProtocol = 0x00,
+ .driver_info = (unsigned long) &zaurus_pxa_info,
},
#endif
--- diff/drivers/usb/serial/ftdi_sio.c 2004-03-11 10:20:28.000000000 +0000
+++ source/drivers/usb/serial/ftdi_sio.c 2004-03-16 11:27:48.367834632 +0000
@@ -286,6 +286,7 @@ static struct usb_device_id id_table_sio
static struct usb_device_id id_table_8U232AM [] = {
+ { USB_DEVICE_VER(FTDI_VID, FTDI_IRTRANS_PID, 0, 0x3ff) },
{ USB_DEVICE_VER(FTDI_VID, FTDI_8U232AM_PID, 0, 0x3ff) },
{ USB_DEVICE_VER(FTDI_VID, FTDI_8U232AM_ALT_PID, 0, 0x3ff) },
{ USB_DEVICE_VER(FTDI_VID, FTDI_RELAIS_PID, 0, 0x3ff) },
@@ -358,6 +359,7 @@ static struct usb_device_id id_table_8U2
static struct usb_device_id id_table_FT232BM [] = {
+ { USB_DEVICE_VER(FTDI_VID, FTDI_IRTRANS_PID, 0x400, 0xffff) },
{ USB_DEVICE_VER(FTDI_VID, FTDI_8U232AM_PID, 0x400, 0xffff) },
{ USB_DEVICE_VER(FTDI_VID, FTDI_8U232AM_ALT_PID, 0x400, 0xffff) },
{ USB_DEVICE_VER(FTDI_VID, FTDI_RELAIS_PID, 0x400, 0xffff) },
@@ -451,6 +453,7 @@ static struct usb_device_id id_table_HE_
static struct usb_device_id id_table_combined [] = {
+ { USB_DEVICE(FTDI_VID, FTDI_IRTRANS_PID) },
{ USB_DEVICE(FTDI_VID, FTDI_SIO_PID) },
{ USB_DEVICE(FTDI_VID, FTDI_8U232AM_PID) },
{ USB_DEVICE(FTDI_VID, FTDI_8U232AM_ALT_PID) },
--- diff/drivers/usb/serial/ftdi_sio.h 2004-03-11 10:20:28.000000000 +0000
+++ source/drivers/usb/serial/ftdi_sio.h 2004-03-16 11:27:48.367834632 +0000
@@ -30,6 +30,8 @@
#define FTDI_NF_RIC_VID 0x0DCD /* Vendor Id */
#define FTDI_NF_RIC_PID 0x0001 /* Product Id */
+/* www.irtrans.de device */
+#define FTDI_IRTRANS_PID 0xFC60 /* Product Id */
/* www.crystalfontz.com devices - thanx for providing free devices for evaluation ! */
/* they use the ftdi chipset for the USB interface and the vendor id is the same */
--- diff/drivers/usb/serial/kl5kusb105.c 2003-09-17 11:28:11.000000000 +0000
+++ source/drivers/usb/serial/kl5kusb105.c 2004-03-16 11:27:48.368834480 +0000
@@ -273,6 +273,7 @@ static int klsi_105_startup (struct usb_
/* allocate the private data structure */
for (i=0; i<serial->num_ports; i++) {
+ int j;
priv = kmalloc(sizeof(struct klsi_105_private),
GFP_KERNEL);
if (!priv) {
@@ -293,10 +294,10 @@ static int klsi_105_startup (struct usb_
usb_set_serial_port_data(serial->port[i], priv);
spin_lock_init (&priv->lock);
- for (i=0; i<NUM_URBS; i++) {
+ for (j=0; j<NUM_URBS; j++) {
struct urb* urb = usb_alloc_urb(0, GFP_KERNEL);
- priv->write_urb_pool[i] = urb;
+ priv->write_urb_pool[j] = urb;
if (urb == NULL) {
err("No more urbs???");
continue;
--- diff/drivers/usb/serial/pl2303.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/serial/pl2303.c 2004-03-16 11:27:48.369834328 +0000
@@ -403,7 +403,7 @@ static int pl2303_open (struct usb_seria
{
struct termios tmp_termios;
struct usb_serial *serial = port->serial;
- unsigned char buf[10];
+ unsigned char *buf;
int result;
if (port_paranoia_check (port, __FUNCTION__))
@@ -414,6 +414,10 @@ static int pl2303_open (struct usb_seria
usb_clear_halt(serial->dev, port->write_urb->pipe);
usb_clear_halt(serial->dev, port->read_urb->pipe);
+ buf = kmalloc(10, GFP_KERNEL);
+ if (buf==NULL)
+ return -ENOMEM;
+
#define FISH(a,b,c,d) \
result=usb_control_msg(serial->dev, usb_rcvctrlpipe(serial->dev,0), \
b, a, c, d, buf, 1, 100); \
@@ -433,6 +437,8 @@ static int pl2303_open (struct usb_seria
FISH (VENDOR_READ_REQUEST_TYPE, VENDOR_READ_REQUEST, 0x8484, 0);
FISH (VENDOR_READ_REQUEST_TYPE, VENDOR_READ_REQUEST, 0x8383, 0);
+ kfree(buf);
+
/* Setup termios */
if (port->tty) {
pl2303_set_termios (port, &tmp_termios);
--- diff/drivers/usb/serial/usb-serial.c 2003-12-19 09:51:11.000000000 +0000
+++ source/drivers/usb/serial/usb-serial.c 2004-03-16 11:27:48.370834176 +0000
@@ -1006,7 +1006,7 @@ int usb_serial_probe(struct usb_interfac
/* descriptor matches, let's find the endpoints needed */
/* check out the endpoints */
- iface_desc = &interface->altsetting[0];
+ iface_desc = interface->cur_altsetting;
for (i = 0; i < iface_desc->desc.bNumEndpoints; ++i) {
endpoint = &iface_desc->endpoint[i].desc;
--- diff/drivers/usb/serial/visor.c 2004-02-18 08:54:12.000000000 +0000
+++ source/drivers/usb/serial/visor.c 2004-03-16 11:27:48.371834024 +0000
@@ -239,6 +239,8 @@ static struct usb_device_id id_table []
.driver_info = (kernel_ulong_t)&palm_os_4_probe },
{ USB_DEVICE(SAMSUNG_VENDOR_ID, SAMSUNG_SCH_I330_ID),
.driver_info = (kernel_ulong_t)&palm_os_4_probe },
+ { USB_DEVICE(SAMSUNG_VENDOR_ID, SAMSUNG_SPH_I500_ID),
+ .driver_info = (kernel_ulong_t)&palm_os_4_probe },
{ USB_DEVICE(GARMIN_VENDOR_ID, GARMIN_IQUE_3600_ID),
.driver_info = (kernel_ulong_t)&palm_os_4_probe },
{ USB_DEVICE(ACEECA_VENDOR_ID, ACEECA_MEZ1000_ID),
@@ -275,6 +277,7 @@ static struct usb_device_id id_table_com
{ USB_DEVICE(SONY_VENDOR_ID, SONY_CLIE_UX50_ID) },
{ USB_DEVICE(SONY_VENDOR_ID, SONY_CLIE_TJ25_ID) },
{ USB_DEVICE(SAMSUNG_VENDOR_ID, SAMSUNG_SCH_I330_ID) },
+ { USB_DEVICE(SAMSUNG_VENDOR_ID, SAMSUNG_SPH_I500_ID) },
{ USB_DEVICE(GARMIN_VENDOR_ID, GARMIN_IQUE_3600_ID) },
{ USB_DEVICE(ACEECA_VENDOR_ID, ACEECA_MEZ1000_ID) },
{ }, /* optional parameter entry */
--- diff/drivers/usb/serial/visor.h 2004-02-18 08:54:12.000000000 +0000
+++ source/drivers/usb/serial/visor.h 2004-03-16 11:27:48.371834024 +0000
@@ -46,6 +46,7 @@
#define SAMSUNG_VENDOR_ID 0x04E8
#define SAMSUNG_SCH_I330_ID 0x8001
+#define SAMSUNG_SPH_I500_ID 0x6601
#define GARMIN_VENDOR_ID 0x091E
#define GARMIN_IQUE_3600_ID 0x0004
--- diff/drivers/usb/storage/Kconfig 2004-02-18 08:54:12.000000000 +0000
+++ source/drivers/usb/storage/Kconfig 2004-03-16 11:27:48.371834024 +0000
@@ -1,8 +1,6 @@
#
# USB Storage driver configuration
#
-comment "SCSI support is needed for USB Storage"
- depends on USB && SCSI=n
config USB_STORAGE
tristate "USB Mass Storage support"
--- diff/drivers/usb/storage/scsiglue.c 2004-02-09 10:36:11.000000000 +0000
+++ source/drivers/usb/storage/scsiglue.c 2004-03-16 11:27:48.372833872 +0000
@@ -64,8 +64,10 @@ static const char* host_info(struct Scsi
return "SCSI emulation for USB Mass Storage devices";
}
-static int slave_configure (struct scsi_device *sdev)
+static int slave_configure(struct scsi_device *sdev)
{
+ struct us_data *us = (struct us_data *) sdev->host->hostdata[0];
+
/* Scatter-gather buffers (all but the last) must have a length
* divisible by the bulk maxpacket size. Otherwise a data packet
* would end up being short, causing a premature end to the data
@@ -76,6 +78,16 @@ static int slave_configure (struct scsi_
* the end, scatter-gather buffers follow page boundaries. */
blk_queue_dma_alignment(sdev->request_queue, (512 - 1));
+ /* Devices using Genesys Logic chips cause a lot of trouble for
+ * high-speed transfers; they die unpredictably when given more
+ * than 64 KB of data at a time. If we detect such a device,
+ * reduce the maximum transfer size to 64 KB = 128 sectors. */
+
+#define USB_VENDOR_ID_GENESYS 0x05e3 // Needs a standard location
+ if (us->pusb_dev->descriptor.idVendor == USB_VENDOR_ID_GENESYS &&
+ us->pusb_dev->speed == USB_SPEED_HIGH)
+ blk_queue_max_sectors(sdev->request_queue, 128);
+
/* this is to satisify the compiler, tho I don't think the
* return code is ever checked anywhere. */
return 0;
--- diff/drivers/usb/storage/transport.c 2004-03-11 10:20:28.000000000 +0000
+++ source/drivers/usb/storage/transport.c 2004-03-16 11:27:48.373833720 +0000
@@ -563,9 +563,9 @@ void usb_stor_invoke_transport(Scsi_Cmnd
/*
* If we're running the CB transport, which is incapable
- * of determining status on its own, we need to auto-sense
+ * of determining status on its own, we will auto-sense
* unless the operation involved a data-in transfer. Devices
- * can signal data-in errors by stalling the bulk-in pipe.
+ * can signal most data-in errors by stalling the bulk-in pipe.
*/
if ((us->protocol == US_PR_CB || us->protocol == US_PR_DPCM_USB) &&
srb->sc_data_direction != SCSI_DATA_READ) {
@@ -698,7 +698,11 @@ void usb_stor_invoke_transport(Scsi_Cmnd
* out the sense buffer so the higher layers won't realize
* we did an unsolicited auto-sense. */
if (result == USB_STOR_TRANSPORT_GOOD &&
- (srb->sense_buffer[2] & 0xf) == 0x0) {
+ /* Filemark 0, ignore EOM, ILI 0, no sense */
+ (srb->sense_buffer[2] & 0xaf) == 0 &&
+ /* No ASC or ASCQ */
+ srb->sense_buffer[12] == 0 &&
+ srb->sense_buffer[13] == 0) {
srb->result = SAM_STAT_GOOD;
srb->sense_buffer[0] = 0x0;
}
@@ -809,15 +813,19 @@ int usb_stor_CBI_transport(Scsi_Cmnd *sr
}
/* If not UFI, we interpret the data as a result code
- * The first byte should always be a 0x0
- * The second byte & 0x0F should be 0x0 for good, otherwise error
+ * The first byte should always be a 0x0.
+ *
+ * Some bogus devices don't follow that rule. They stuff the ASC
+ * into the first byte -- so if it's non-zero, call it a failure.
*/
if (us->iobuf[0]) {
- US_DEBUGP("CBI IRQ data showed reserved bType %d\n",
+ US_DEBUGP("CBI IRQ data showed reserved bType 0x%x\n",
us->iobuf[0]);
- return USB_STOR_TRANSPORT_ERROR;
+ goto Failed;
+
}
+ /* The second byte & 0x0F should be 0x0 for good, otherwise error */
switch (us->iobuf[1] & 0x0F) {
case 0x00:
return USB_STOR_TRANSPORT_GOOD;
--- diff/drivers/usb/storage/unusual_devs.h 2004-03-11 10:20:28.000000000 +0000
+++ source/drivers/usb/storage/unusual_devs.h 2004-03-16 11:27:48.374833568 +0000
@@ -261,6 +261,14 @@ UNUSUAL_DEV( 0x054c, 0x0010, 0x0106, 0x
US_SC_SCSI, US_PR_DEVICE, NULL,
US_FL_SINGLE_LUN | US_FL_MODE_XLATE ),
+/* This entry is needed because the device reports Sub=ff */
+UNUSUAL_DEV( 0x054c, 0x0010, 0x0500, 0x0500,
+ "Sony",
+ "DSC-T1",
+ US_SC_8070, US_PR_DEVICE, NULL,
+ US_FL_SINGLE_LUN | US_FL_MODE_XLATE ),
+
+
/* Reported by wim@geeks.nl */
UNUSUAL_DEV( 0x054c, 0x0025, 0x0100, 0x0100,
"Sony",
@@ -368,7 +376,7 @@ UNUSUAL_DEV( 0x05ab, 0x5701, 0x0100, 0x
UNUSUAL_DEV( 0x05dc, 0x0001, 0x0000, 0x0001,
"Lexar",
"Jumpshot USB CF Reader",
- US_SC_SCSI, US_PR_JUMPSHOT, NULL,
+ US_SC_DEVICE, US_PR_JUMPSHOT, NULL,
US_FL_MODE_XLATE ),
#endif
@@ -409,6 +417,13 @@ UNUSUAL_DEV( 0x05e3, 0x0702, 0x0000, 0x
US_SC_DEVICE, US_PR_DEVICE, NULL,
US_FL_FIX_INQUIRY ),
+/* Reported by Henning Schild <henning@wh9.tu-dresden.de> */
+UNUSUAL_DEV( 0x05e3, 0x0702, 0x0113, 0x0113,
+ "EagleTec",
+ "External Hard Disk",
+ US_SC_DEVICE, US_PR_DEVICE, NULL,
+ US_FL_FIX_INQUIRY ),
+
/* Reported by Hanno Boeck <hanno@gmx.de>
* Taken from the Lycoris Kernel */
UNUSUAL_DEV( 0x0636, 0x0003, 0x0000, 0x9999,
@@ -440,12 +455,6 @@ UNUSUAL_DEV( 0x0686, 0x4006, 0x0001, 0x0
US_SC_SCSI, US_PR_DEVICE, NULL,
0 ),
-UNUSUAL_DEV( 0x0686, 0x400b, 0x0001, 0x0001,
- "Minolta",
- "DiMAGE 7i",
- US_SC_SCSI, US_PR_DEVICE, NULL,
- 0 ),
-
UNUSUAL_DEV( 0x0686, 0x400f, 0x0001, 0x0001,
"Minolta",
"DiMAGE 7Hi",
@@ -619,6 +628,9 @@ UNUSUAL_DEV( 0x07c4, 0xa400, 0x0000, 0x
* are using transport protocol CB.
* - They don't like the INQUIRY command. So we must handle this command
* of the SCSI layer ourselves.
+ * - Some cameras with idProduct=0x1001 and bcdDevice=0x1000 have
+ * bInterfaceProtocol=0x00 (US_PR_CBI) while others have 0x01 (US_PR_CB).
+ * So don't remove the US_PR_CB override!
*/
UNUSUAL_DEV( 0x07cf, 0x1001, 0x1000, 0x9009,
"Casio",
@@ -649,6 +661,17 @@ UNUSUAL_DEV( 0x08ca, 0x2011, 0x0000, 0x
US_SC_DEVICE, US_PR_DEVICE, NULL,
US_FL_MODE_XLATE ),
+/* Entry needed for flags. Moreover, all devices with this ID use
+ * bulk-only transport, but _some_ falsely report Control/Bulk instead.
+ * One example is "Trumpion Digital Research MYMP3".
+ * Submitted by Bjoern Brill <brill(at)fs.math.uni-frankfurt.de>
+ */
+UNUSUAL_DEV( 0x090a, 0x1001, 0x0100, 0x0100,
+ "Trumpion",
+ "t33520 USB Flash Card Controller",
+ US_SC_DEVICE, US_PR_BULK, NULL,
+ US_FL_MODE_XLATE),
+
/* Trumpion Microelectronics MP3 player (felipe_alfaro@linuxmail.org) */
UNUSUAL_DEV( 0x090a, 0x1200, 0x0000, 0x9999,
"Trumpion",
@@ -688,15 +711,9 @@ UNUSUAL_DEV( 0x0a17, 0x0004, 0x1000, 0x1
US_SC_DEVICE, US_PR_DEVICE, NULL,
US_FL_FIX_INQUIRY ),
-/* This entry from <matthias@ma-c.de> in the Debian mailing list */
-UNUSUAL_DEV( 0x0a17, 0x0006, 0x0000, 0xffff,
- "Pentax",
- "Optio 330GS",
- US_SC_8070, US_PR_CB, NULL,
- US_FL_MODE_XLATE | US_FL_FIX_INQUIRY ),
/* Submitted by Per Winkvist <per.winkvist@uk.com> */
-UNUSUAL_DEV( 0x0a17, 0x006, 0x1000, 0x9009,
+UNUSUAL_DEV( 0x0a17, 0x006, 0x0000, 0xffff,
"Pentax",
"Optio S/S4",
US_SC_DEVICE, US_PR_DEVICE, NULL,
--- diff/drivers/usb/storage/usb.c 2004-02-18 08:54:12.000000000 +0000
+++ source/drivers/usb/storage/usb.c 2004-03-16 11:27:48.375833416 +0000
@@ -423,7 +423,7 @@ static int associate_dev(struct us_data
/* Fill in the device-related fields */
us->pusb_dev = interface_to_usbdev(intf);
us->pusb_intf = intf;
- us->ifnum = intf->altsetting->desc.bInterfaceNumber;
+ us->ifnum = intf->cur_altsetting->desc.bInterfaceNumber;
/* Store our private data in the interface and increment the
* device's reference count */
@@ -452,7 +452,7 @@ static void get_device_info(struct us_da
{
struct usb_device *dev = us->pusb_dev;
struct usb_interface_descriptor *idesc =
- &us->pusb_intf->altsetting[us->pusb_intf->act_altsetting].desc;
+ &us->pusb_intf->cur_altsetting->desc;
struct us_unusual_dev *unusual_dev = &us_unusual_dev_list[id_index];
struct usb_device_id *id = &storage_usb_ids[id_index];
@@ -686,7 +686,7 @@ static int get_protocol(struct us_data *
static int get_pipes(struct us_data *us)
{
struct usb_host_interface *altsetting =
- &us->pusb_intf->altsetting[us->pusb_intf->act_altsetting];
+ us->pusb_intf->cur_altsetting;
int i;
struct usb_endpoint_descriptor *ep;
struct usb_endpoint_descriptor *ep_in = NULL;
@@ -877,8 +877,9 @@ static int storage_probe(struct usb_inte
int result;
US_DEBUGP("USB Mass Storage device detected\n");
- US_DEBUGP("act_altsetting is %d, id_index is %d\n",
- intf->act_altsetting, id_index);
+ US_DEBUGP("altsetting is %d, id_index is %d\n",
+ intf->cur_altsetting->desc.bAlternateSetting,
+ id_index);
/* Allocate the us_data structure and initialize the mutexes */
us = (struct us_data *) kmalloc(sizeof(*us), GFP_KERNEL);
@@ -954,8 +955,6 @@ static int storage_probe(struct usb_inte
scsi_scan_host(us->host);
printk(KERN_DEBUG
- "WARNING: USB Mass Storage data integrity not assured\n");
- printk(KERN_DEBUG
"USB Mass Storage device found at %d\n", us->pusb_dev->devnum);
return 0;
--- diff/drivers/usb/storage/usb.h 2003-09-17 11:28:11.000000000 +0000
+++ source/drivers/usb/storage/usb.h 2004-03-16 11:27:48.375833416 +0000
@@ -176,6 +176,5 @@ extern void fill_inquiry_response(struct
* single queue element srb for write access */
#define scsi_unlock(host) spin_unlock_irq(host->host_lock)
#define scsi_lock(host) spin_lock_irq(host->host_lock)
-#define sg_address(psg) (page_address((psg).page) + (psg).offset)
#endif
--- diff/drivers/video/console/fbcon.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/video/console/fbcon.c 2004-03-16 11:27:48.377833112 +0000
@@ -309,94 +309,6 @@ int set_con2fb_map(int unit, int newidx)
}
/*
- * drawing helpers
- */
-static void putcs_unaligned(struct vc_data *vc, struct fb_info *info,
- struct fb_image *image, int count,
- const unsigned short *s)
-{
- unsigned short charmask = vc->vc_hi_font_mask ? 0x1ff : 0xff;
- unsigned int width = (vc->vc_font.width + 7) >> 3;
- unsigned int cellsize = vc->vc_font.height * width;
- unsigned int maxcnt = info->pixmap.size/cellsize;
- unsigned int shift_low = 0, mod = vc->vc_font.width % 8;
- unsigned int shift_high = 8, size, pitch, cnt, k;
- unsigned int buf_align = info->pixmap.buf_align - 1;
- unsigned int scan_align = info->pixmap.scan_align - 1;
- unsigned int idx = vc->vc_font.width >> 3;
- u8 *src, *dst, *dst0;
-
- while (count) {
- if (count > maxcnt)
- cnt = k = maxcnt;
- else
- cnt = k = count;
-
- image->width = vc->vc_font.width * cnt;
- pitch = ((image->width + 7) >> 3) + scan_align;
- pitch &= ~scan_align;
- size = pitch * vc->vc_font.height + buf_align;
- size &= ~buf_align;
- dst0 = fb_get_buffer_offset(info, &info->pixmap, size);
- image->data = dst0;
- while (k--) {
- src = vc->vc_font.data + (scr_readw(s++) & charmask)*
- cellsize;
- dst = dst0;
- fb_move_buf_unaligned(info, &info->pixmap, dst, pitch, src,
- idx, image->height, shift_high,
- shift_low, mod);
- shift_low += mod;
- dst0 += (shift_low >= 8) ? width : width - 1;
- shift_low &= 7;
- shift_high = 8 - shift_low;
- }
- info->fbops->fb_imageblit(info, image);
- image->dx += cnt * vc->vc_font.width;
- count -= cnt;
- }
-}
-
-static void putcs_aligned(struct vc_data *vc, struct fb_info *info,
- struct fb_image *image, int count,
- const unsigned short *s)
-{
- unsigned short charmask = vc->vc_hi_font_mask ? 0x1ff : 0xff;
- unsigned int width = vc->vc_font.width >> 3;
- unsigned int cellsize = vc->vc_font.height * width;
- unsigned int maxcnt = info->pixmap.size/cellsize;
- unsigned int scan_align = info->pixmap.scan_align - 1;
- unsigned int buf_align = info->pixmap.buf_align - 1;
- unsigned int pitch, cnt, size, k;
- u8 *src, *dst, *dst0;
-
- while (count) {
- if (count > maxcnt)
- cnt = k = maxcnt;
- else
- cnt = k = count;
-
- pitch = width * cnt + scan_align;
- pitch &= ~scan_align;
- size = pitch * vc->vc_font.height + buf_align;
- size &= ~buf_align;
- image->width = vc->vc_font.width * cnt;
- dst0 = fb_get_buffer_offset(info, &info->pixmap, size);
- image->data = dst0;
- while (k--) {
- src = vc->vc_font.data + (scr_readw(s++)&charmask)*cellsize;
- dst = dst0;
- fb_move_buf_aligned(info, &info->pixmap, dst, pitch, src,
- width, image->height);
- dst0 += width;
- }
- info->fbops->fb_imageblit(info, image);
- image->dx += cnt * vc->vc_font.width;
- count -= cnt;
- }
-}
-
-/*
* Accelerated handlers.
*/
void accel_bmove(struct vc_data *vc, struct fb_info *info, int sy,
@@ -430,48 +342,23 @@ void accel_clear(struct vc_data *vc, str
info->fbops->fb_fillrect(info, ®ion);
}
-static void accel_putc(struct vc_data *vc, struct fb_info *info,
- int c, int ypos, int xpos)
+void accel_putcs(struct vc_data *vc, struct fb_info *info,
+ const unsigned short *s, int count, int yy, int xx)
{
unsigned short charmask = vc->vc_hi_font_mask ? 0x1ff : 0xff;
unsigned int width = (vc->vc_font.width + 7) >> 3;
+ unsigned int cellsize = vc->vc_font.height * width;
+ unsigned int maxcnt = info->pixmap.size/cellsize;
unsigned int scan_align = info->pixmap.scan_align - 1;
unsigned int buf_align = info->pixmap.buf_align - 1;
+ unsigned int shift_low = 0, mod = vc->vc_font.width % 8;
+ unsigned int shift_high = 8, pitch, cnt, size, k;
int bgshift = (vc->vc_hi_font_mask) ? 13 : 12;
int fgshift = (vc->vc_hi_font_mask) ? 9 : 8;
- unsigned int size, pitch;
- struct fb_image image;
- u8 *src, *dst;
-
- image.dx = xpos * vc->vc_font.width;
- image.dy = ypos * vc->vc_font.height;
- image.width = vc->vc_font.width;
- image.height = vc->vc_font.height;
- image.fg_color = attr_fgcol(fgshift, c);
- image.bg_color = attr_bgcol(bgshift, c);
- image.depth = 1;
-
- pitch = width + scan_align;
- pitch &= ~scan_align;
- size = pitch * vc->vc_font.height;
- size += buf_align;
- size &= ~buf_align;
- dst = fb_get_buffer_offset(info, &info->pixmap, size);
- image.data = dst;
- src = vc->vc_font.data + (c & charmask) * vc->vc_font.height * width;
-
- fb_move_buf_aligned(info, &info->pixmap, dst, pitch, src, width, image.height);
-
- info->fbops->fb_imageblit(info, &image);
-}
-
-void accel_putcs(struct vc_data *vc, struct fb_info *info,
- const unsigned short *s, int count, int yy, int xx)
-{
- int bgshift = (vc->vc_hi_font_mask) ? 13 : 12;
- int fgshift = (vc->vc_hi_font_mask) ? 9 : 8;
+ unsigned int idx = vc->vc_font.width >> 3;
struct fb_image image;
u16 c = scr_readw(s);
+ u8 *src, *dst, *dst0;
image.fg_color = attr_fgcol(fgshift, c);
image.bg_color = attr_bgcol(bgshift, c);
@@ -480,10 +367,41 @@ void accel_putcs(struct vc_data *vc, str
image.height = vc->vc_font.height;
image.depth = 1;
- if (!(vc->vc_font.width & 7))
- putcs_aligned(vc, info, &image, count, s);
- else
- putcs_unaligned(vc, info, &image, count, s);
+ while (count) {
+ if (count > maxcnt)
+ cnt = k = maxcnt;
+ else
+ cnt = k = count;
+
+ image.width = vc->vc_font.width * cnt;
+ pitch = ((image.width + 7) >> 3) + scan_align;
+ pitch &= ~scan_align;
+ size = pitch * image.height + buf_align;
+ size &= ~buf_align;
+ dst0 = fb_get_buffer_offset(info, &info->pixmap, size);
+ image.data = dst0;
+ while (k--) {
+ src = vc->vc_font.data + (scr_readw(s++) & charmask)*cellsize;
+ dst = dst0;
+
+ if (mod) {
+ fb_move_buf_unaligned(info, &info->pixmap, dst, pitch,
+ src, idx, image.height, shift_high,
+ shift_low, mod);
+ shift_low += mod;
+ dst0 += (shift_low >= 8) ? width : width - 1;
+ shift_low &= 7;
+ shift_high = 8 - shift_low;
+ } else {
+ fb_move_buf_aligned(info, &info->pixmap, dst, pitch,
+ src, idx, image.height);
+ dst0 += width;
+ }
+ }
+ info->fbops->fb_imageblit(info, &image);
+ image.dx += cnt * vc->vc_font.width;
+ count -= cnt;
+ }
}
void accel_clear_margins(struct vc_data *vc, struct fb_info *info,
@@ -724,15 +642,13 @@ static __inline__ void updatescrollmode(
static void fbcon_set_display(struct vc_data *vc, int init, int logo)
{
struct fb_info *info = registered_fb[(int) con2fb_map[vc->vc_num]];
+ int nr_rows, nr_cols, old_rows, old_cols, i, charcnt = 256;
struct display *p = &fb_display[vc->vc_num];
- int nr_rows, nr_cols;
- int old_rows, old_cols;
unsigned short *save = NULL, *r, *q;
- int i, charcnt = 256;
struct font_desc *font;
if (vc->vc_num != fg_console || (info->flags & FBINFO_FLAG_MODULE) ||
- info->fix.type == FB_TYPE_TEXT)
+ (info->fix.type == FB_TYPE_TEXT))
logo = 0;
info->var.xoffset = info->var.yoffset = p->yscroll = 0; /* reset wrap/pan */
@@ -958,11 +874,19 @@ static void fbcon_clear(struct vc_data *
accel_clear(vc, info, real_y(p, sy), sx, height, width);
}
-
static void fbcon_putc(struct vc_data *vc, int c, int ypos, int xpos)
{
struct fb_info *info = registered_fb[(int) con2fb_map[vc->vc_num]];
+ unsigned short charmask = vc->vc_hi_font_mask ? 0x1ff : 0xff;
+ unsigned int scan_align = info->pixmap.scan_align - 1;
+ unsigned int buf_align = info->pixmap.buf_align - 1;
+ unsigned int width = (vc->vc_font.width + 7) >> 3;
+ int bgshift = (vc->vc_hi_font_mask) ? 13 : 12;
+ int fgshift = (vc->vc_hi_font_mask) ? 9 : 8;
struct display *p = &fb_display[vc->vc_num];
+ unsigned int size, pitch;
+ struct fb_image image;
+ u8 *src, *dst;
if (!info->fbops->fb_blank && console_blanked)
return;
@@ -972,7 +896,28 @@ static void fbcon_putc(struct vc_data *v
if (vt_cons[vc->vc_num]->vc_mode != KD_TEXT)
return;
- accel_putc(vc, info, c, real_y(p, ypos), xpos);
+ image.dx = xpos * vc->vc_font.width;
+ image.dy = real_y(p, ypos) * vc->vc_font.height;
+ image.width = vc->vc_font.width;
+ image.height = vc->vc_font.height;
+ image.fg_color = attr_fgcol(fgshift, c);
+ image.bg_color = attr_bgcol(bgshift, c);
+ image.depth = 1;
+
+ src = vc->vc_font.data + (c & charmask) * vc->vc_font.height * width;
+
+ pitch = width + scan_align;
+ pitch &= ~scan_align;
+ size = pitch * vc->vc_font.height;
+ size += buf_align;
+ size &= ~buf_align;
+
+ dst = fb_get_buffer_offset(info, &info->pixmap, size);
+ image.data = dst;
+
+ fb_move_buf_aligned(info, &info->pixmap, dst, pitch, src, width, image.height);
+
+ info->fbops->fb_imageblit(info, &image);
}
static void fbcon_putcs(struct vc_data *vc, const unsigned short *s,
--- diff/drivers/video/modedb.c 2004-03-16 11:24:02.000000000 +0000
+++ source/drivers/video/modedb.c 2004-03-16 11:27:48.378832960 +0000
@@ -554,5 +554,5 @@ done:
return 0;
}
-EXPORT_SYMBOL(__fb_try_mode);
EXPORT_SYMBOL(vesa_modes);
+EXPORT_SYMBOL(fb_find_mode);
--- diff/fs/Kconfig 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/Kconfig 2004-03-16 11:27:48.406828704 +0000
@@ -780,6 +780,30 @@ config PROC_KCORE
bool
default y if !ARM
+config SYSFS
+ bool "sysfs file system support" if EMBEDDED
+ default y
+ help
+ The sysfs filesystem is a virtual filesystem that the kernel uses to export
+ internal kernel objects, their attributes, and their relationships to one
+ another.
+
+ Users can use sysfs to ascertain useful information about the running kernel,
+ such as the devices the kernel has discovered on each bus and which driver
+ each is bound to. sysfs can also be used to tune devices and other kernel
+ subsystems.
+
+ Some system agents rely on the information in sysfs to operate. /sbin/hotplug
+ uses device and object attributes in sysfs to assist in delegating policy
+ decisions, like persistantly naming devices.
+
+ sysfs is currently needed by the block subsystem to mount the root partition.
+ Therefore, you MUST say Y if you're booting from a hard drive. If you use any
+ type of hotpluggable device, you'll also need sysfs for /sbin/hotplug support.
+
+ However, designers of embedded systems may want to say N here to conserve
+ space.
+
config DEVFS_FS
bool "/dev file system support (OBSOLETE)"
depends on EXPERIMENTAL
--- diff/fs/Makefile 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/Makefile 2004-03-16 11:27:48.407828552 +0000
@@ -39,7 +39,7 @@ obj-$(CONFIG_QUOTACTL) += quota.o
obj-$(CONFIG_PROC_FS) += proc/
obj-y += partitions/
-obj-y += sysfs/
+obj-$(CONFIG_SYSFS) += sysfs/
obj-y += devpts/
obj-$(CONFIG_PROFILING) += dcookies.o
--- diff/fs/binfmt_aout.c 2003-09-30 14:46:18.000000000 +0000
+++ source/fs/binfmt_aout.c 2004-03-16 11:27:48.378832960 +0000
@@ -309,7 +309,7 @@ static int load_aout_binary(struct linux
(current->mm->start_brk = N_BSSADDR(ex));
current->mm->free_area_cache = TASK_UNMAPPED_BASE;
- current->mm->rss = 0;
+ zero_rss(current->mm);
current->mm->mmap = NULL;
compute_creds(bprm);
current->flags &= ~PF_FORKNOEXEC;
--- diff/fs/binfmt_elf.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/binfmt_elf.c 2004-03-16 11:27:48.379832808 +0000
@@ -672,7 +672,7 @@ static int load_elf_binary(struct linux_
/* Do this so that we can load the interpreter, if need be. We will
change some of these later */
- current->mm->rss = 0;
+ zero_rss(current->mm);
current->mm->free_area_cache = TASK_UNMAPPED_BASE;
retval = setup_arg_pages(bprm);
if (retval < 0) {
@@ -1129,7 +1129,7 @@ static void fill_prstatus(struct elf_prs
prstatus->pr_pid = p->pid;
prstatus->pr_ppid = p->parent->pid;
prstatus->pr_pgrp = process_group(p);
- prstatus->pr_sid = p->session;
+ prstatus->pr_sid = p->signal->session;
jiffies_to_timeval(p->utime, &prstatus->pr_utime);
jiffies_to_timeval(p->stime, &prstatus->pr_stime);
jiffies_to_timeval(p->cutime, &prstatus->pr_cutime);
@@ -1157,7 +1157,7 @@ static void fill_psinfo(struct elf_prpsi
psinfo->pr_pid = p->pid;
psinfo->pr_ppid = p->parent->pid;
psinfo->pr_pgrp = process_group(p);
- psinfo->pr_sid = p->session;
+ psinfo->pr_sid = p->signal->session;
i = p->state ? ffz(~p->state) + 1 : 0;
psinfo->pr_state = i;
--- diff/fs/binfmt_flat.c 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/binfmt_flat.c 2004-03-16 11:27:48.380832656 +0000
@@ -651,7 +651,7 @@ static int load_flat_file(struct linux_b
current->mm->start_brk = datapos + data_len + bss_len;
current->mm->brk = (current->mm->start_brk + 3) & ~3;
current->mm->context.end_brk = memp + ksize((void *) memp) - stack_len;
- current->mm->rss = 0;
+ zero_rss(current->mm);
}
if (flags & FLAT_FLAG_KTRACE)
--- diff/fs/binfmt_misc.c 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/binfmt_misc.c 2004-03-16 11:27:48.381832504 +0000
@@ -39,6 +39,8 @@ static int enabled = 1;
enum {Enabled, Magic};
#define MISC_FMT_PRESERVE_ARGV0 (1<<31)
+#define MISC_FMT_OPEN_BINARY (1<<30)
+#define MISC_FMT_CREDENTIALS (1<<29)
typedef struct {
struct list_head list;
@@ -102,10 +104,15 @@ static Node *check_file(struct linux_bin
static int load_misc_binary(struct linux_binprm *bprm, struct pt_regs *regs)
{
Node *fmt;
- struct file * file;
+ struct file * interp_file = NULL;
+ struct file * binary_file = NULL;
char iname[BINPRM_BUF_SIZE];
char *iname_addr = iname;
int retval;
+ int fd_binary = -1;
+ char fd_str[32];
+ char * fdsp = fd_str;
+ int is_open_bin;
retval = -ENOEXEC;
if (!enabled)
@@ -120,33 +127,105 @@ static int load_misc_binary(struct linux
if (!fmt)
goto _ret;
- allow_write_access(bprm->file);
- fput(bprm->file);
- bprm->file = NULL;
+ is_open_bin = (fmt->flags & MISC_FMT_OPEN_BINARY) ? 1 : 0;
+
+ if (is_open_bin) {
+ /* if the binary should be opened on behalf of the
+ * interpreter than keep it open and assign descriptor
+ * to it */
+ fd_binary = get_unused_fd ();
+ if (fd_binary < 0) {
+ retval = fd_binary;
+ goto _ret;
+ }
+ snprintf (fd_str, sizeof(fd_str) - 1, "%d", fd_binary);
+ } else {
+ allow_write_access (bprm->file);
+ fput (bprm->file);
+ bprm->file = NULL;
+ }
/* Build args for interpreter */
if (!(fmt->flags & MISC_FMT_PRESERVE_ARGV0)) {
remove_arg_zero(bprm);
}
- retval = copy_strings_kernel(1, &bprm->interp, bprm);
- if (retval < 0) goto _ret;
- bprm->argc++;
- retval = copy_strings_kernel(1, &iname_addr, bprm);
- if (retval < 0) goto _ret;
- bprm->argc++;
+
+ if (is_open_bin) {
+ /* make argv[1] be the file descriptor of the binary */
+ retval = copy_strings_kernel (1, &fdsp, bprm);
+ } else {
+ /* make argv[1] be the path to the binary */
+ retval = copy_strings_kernel (1, &bprm->interp, bprm);
+ }
+ if (retval < 0)
+ goto _error;
+ bprm->argc ++;
+ retval = copy_strings_kernel (1, &iname_addr, bprm);
+ if (retval < 0)
+ goto _error;
+ bprm->argc ++;
bprm->interp = iname; /* for binfmt_script */
- file = open_exec(iname);
- retval = PTR_ERR(file);
- if (IS_ERR(file))
- goto _ret;
- bprm->file = file;
+ interp_file = open_exec (iname);
+ retval = PTR_ERR (interp_file);
+ if (IS_ERR (interp_file))
+ goto _error;
+
+
+ binary_file = bprm->file;
+ if (fmt->flags & MISC_FMT_CREDENTIALS) {
+ /*
+ * Call prepare_binprm before switching to interpreter's file
+ * so that all security calculation will be done according to
+ * binary and not interpreter
+ */
+ retval = prepare_binprm(bprm);
+ if (retval < 0)
+ goto _error;
+ bprm->file = interp_file;
+ memset(bprm->buf, 0, BINPRM_BUF_SIZE);
+ retval = kernel_read(bprm->file, 0, bprm->buf, BINPRM_BUF_SIZE);
+ } else {
+ bprm->file = interp_file;
+ retval = prepare_binprm (bprm);
+ }
+
+ if (retval < 0)
+ goto _error;
+
+ if (is_open_bin) {
+ /* if the binary is not readable than enforce mm->dumpable=0
+ regardless of the interpreter's permissions */
+ if (permission (binary_file->f_dentry->d_inode, MAY_READ, NULL)) {
+ bprm->interp_flags |= BINPRM_FLAGS_ENFORCE_NONDUMP;
+ }
+ /* install the binary's fd. it is done at the latest possible point
+ * because once it is installed it will need to be sys_close()ed
+ * in case of error.
+ */
+ fd_install (fd_binary, binary_file);
+ }
+
+ retval = search_binary_handler (bprm, regs);
+
+ if (retval < 0)
+ goto _error_close_file;
- retval = prepare_binprm(bprm);
- if (retval >= 0)
- retval = search_binary_handler(bprm, regs);
_ret:
return retval;
+
+_error_close_file:
+ if (fd_binary > 0) {
+ sys_close (fd_binary);
+ fd_binary = -1;
+ bprm->file = NULL;
+ }
+_error:
+ if (fd_binary > 0)
+ put_unused_fd (fd_binary);
+ bprm->interp_flags = 0;
+ goto _ret;
+
}
/* Command parsers */
@@ -191,6 +270,36 @@ static int unquote(char *from)
return p - from;
}
+static inline char * check_special_flags (char * sfs, Node * e)
+{
+ char * p = sfs;
+ int cont = 1;
+
+ /* special flags */
+ while (cont) {
+ switch (*p) {
+ case 'P':
+ p++;
+ e->flags |= MISC_FMT_PRESERVE_ARGV0;
+ break;
+ case 'O':
+ p++;
+ e->flags |= MISC_FMT_OPEN_BINARY;
+ break;
+ case 'C':
+ p++;
+ /* this flags also implies the
+ open-binary flag */
+ e->flags |= (MISC_FMT_CREDENTIALS |
+ MISC_FMT_OPEN_BINARY);
+ break;
+ default:
+ cont = 0;
+ }
+ }
+
+ return p;
+}
/*
* This registers a new binary format, it recognises the syntax
* ':name:type:offset:magic:mask:interpreter:'
@@ -293,10 +402,8 @@ static Node *create_entry(const char *bu
if (!e->interpreter[0])
goto Einval;
- if (*p == 'P') {
- p++;
- e->flags |= MISC_FMT_PRESERVE_ARGV0;
- }
+
+ p = check_special_flags (p, e);
if (*p == '\n')
p++;
@@ -346,6 +453,7 @@ static void entry_status(Node *e, char *
{
char *dp;
char *status = "disabled";
+ const char * flags = "flags: ";
if (test_bit(Enabled, &e->flags))
status = "enabled";
@@ -357,6 +465,22 @@ static void entry_status(Node *e, char *
sprintf(page, "%s\ninterpreter %s\n", status, e->interpreter);
dp = page + strlen(page);
+
+ /* print the special flags */
+ sprintf (dp, "%s", flags);
+ dp += strlen (flags);
+ if (e->flags & MISC_FMT_PRESERVE_ARGV0) {
+ *dp ++ = 'P';
+ }
+ if (e->flags & MISC_FMT_OPEN_BINARY) {
+ *dp ++ = 'O';
+ }
+ if (e->flags & MISC_FMT_CREDENTIALS) {
+ *dp ++ = 'C';
+ }
+ *dp ++ = '\n';
+
+
if (!test_bit(Magic, &e->flags)) {
sprintf(dp, "extension .%s\n", e->magic);
} else {
--- diff/fs/binfmt_som.c 2004-01-19 10:22:59.000000000 +0000
+++ source/fs/binfmt_som.c 2004-03-16 11:27:48.381832504 +0000
@@ -259,7 +259,7 @@ load_som_binary(struct linux_binprm * bp
create_som_tables(bprm);
current->mm->start_stack = bprm->p;
- current->mm->rss = 0;
+ zero_rss(current->mm);
#if 0
printk("(start_brk) %08lx\n" , (unsigned long) current->mm->start_brk);
--- diff/fs/block_dev.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/block_dev.c 2004-03-16 11:27:48.382832352 +0000
@@ -155,8 +155,8 @@ blkdev_direct_IO(int rw, struct kiocb *i
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
- return blockdev_direct_IO(rw, iocb, inode, I_BDEV(inode), iov, offset,
- nr_segs, blkdev_get_blocks, NULL);
+ return blockdev_direct_IO_no_locking(rw, iocb, inode, I_BDEV(inode),
+ iov, offset, nr_segs, blkdev_get_blocks, NULL);
}
static int blkdev_writepage(struct page *page, struct writeback_control *wbc)
@@ -796,7 +796,7 @@ struct file_operations def_blk_fops = {
.fsync = block_fsync,
.ioctl = block_ioctl,
.readv = generic_file_readv,
- .writev = generic_file_writev,
+ .writev = generic_file_write_nolock,
.sendfile = generic_file_sendfile,
};
--- diff/fs/buffer.c 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/buffer.c 2004-03-16 11:27:48.384832048 +0000
@@ -132,7 +132,11 @@ void __wait_on_buffer(struct buffer_head
do {
prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE);
if (buffer_locked(bh)) {
- blk_run_queues();
+ struct block_device *bd;
+ smp_mb();
+ bd = bh->b_bdev;
+ if (bd)
+ blk_run_address_space(bd->bd_inode->i_mapping);
io_schedule();
}
} while (buffer_locked(bh));
@@ -396,7 +400,7 @@ out:
* Hack idea: for the blockdev mapping, i_bufferlist_lock contention
* may be quite high. This code could TryLock the page, and if that
* succeeds, there is no need to take private_lock. (But if
- * private_lock is contended then so is mapping->page_lock).
+ * private_lock is contended then so is mapping->tree_lock).
*/
static struct buffer_head *
__find_get_block_slow(struct block_device *bdev, sector_t block, int unused)
@@ -431,6 +435,7 @@ __find_get_block_slow(struct block_devic
printk("block=%llu, b_blocknr=%llu\n",
(unsigned long long)block, (unsigned long long)bh->b_blocknr);
printk("b_state=0x%08lx, b_size=%u\n", bh->b_state, bh->b_size);
+ printk("device blocksize: %d\n", 1 << bd_inode->i_blkbits);
out_unlock:
spin_unlock(&bd_mapping->private_lock);
page_cache_release(page);
@@ -490,7 +495,6 @@ static void free_more_memory(void)
pg_data_t *pgdat;
wakeup_bdflush(1024);
- blk_run_queues();
yield();
for_each_pgdat(pgdat) {
@@ -825,12 +829,6 @@ EXPORT_SYMBOL(mark_buffer_dirty_inode);
* bit, see a bunch of clean buffers and we'd end up with dirty buffers/clean
* page on the dirty page list.
*
- * There is also a small window where the page is dirty, and not on dirty_pages.
- * Also a possibility that by the time the page is added to dirty_pages, it has
- * been set clean. The page lists are somewhat approximate in this regard.
- * It's better to have clean pages accidentally attached to dirty_pages than to
- * leave dirty pages attached to clean_pages.
- *
* We use private_lock to lock against try_to_free_buffers while using the
* page's buffer list. Also use this to protect against clean buffers being
* added to the page after it was set dirty.
@@ -857,24 +855,28 @@ int __set_page_dirty_buffers(struct page
struct buffer_head *bh = head;
do {
- if (buffer_uptodate(bh))
+ if (buffer_uptodate(bh)) {
set_buffer_dirty(bh);
- else
+ if (unlikely(block_dump))
+ printk("%s(%d): dirtied buffer\n",
+ current->comm, current->pid);
+ } else {
buffer_error();
+ }
bh = bh->b_this_page;
} while (bh != head);
}
spin_unlock(&mapping->private_lock);
if (!TestSetPageDirty(page)) {
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&mapping->tree_lock);
if (page->mapping) { /* Race with truncate? */
if (!mapping->backing_dev_info->memory_backed)
inc_page_state(nr_dirty);
- list_del(&page->list);
- list_add(&page->list, &mapping->dirty_pages);
+ radix_tree_tag_set(&mapping->page_tree, page->index,
+ PAGECACHE_TAG_DIRTY);
}
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
}
@@ -1226,7 +1228,7 @@ __getblk_slow(struct block_device *bdev,
* The relationship between dirty buffers and dirty pages:
*
* Whenever a page has any dirty buffers, the page's dirty bit is set, and
- * the page appears on its address_space.dirty_pages list.
+ * the page is tagged dirty in its radix tree.
*
* At all times, the dirtiness of the buffers represents the dirtiness of
* subsections of the page. If the page has buffers, the page dirty bit is
@@ -1248,13 +1250,13 @@ __getblk_slow(struct block_device *bdev,
/**
* mark_buffer_dirty - mark a buffer_head as needing writeout
*
- * mark_buffer_dirty() will set the dirty bit against the buffer,
- * then set its backing page dirty, then attach the page to its
- * address_space's dirty_pages list and then attach the address_space's
- * inode to its superblock's dirty inode list.
+ * mark_buffer_dirty() will set the dirty bit against the buffer, then set its
+ * backing page dirty, then tag the page as dirty in its address_space's radix
+ * tree and then attach the address_space's inode to its superblock's dirty
+ * inode list.
*
* mark_buffer_dirty() is atomic. It takes bh->b_page->mapping->private_lock,
- * mapping->page_lock and the global inode_lock.
+ * mapping->tree_lock and the global inode_lock.
*/
void fastcall mark_buffer_dirty(struct buffer_head *bh)
{
@@ -1806,27 +1808,28 @@ static int __block_write_full_page(struc
do {
get_bh(bh);
- if (buffer_mapped(bh) && buffer_dirty(bh)) {
- if (wbc->sync_mode != WB_SYNC_NONE) {
- lock_buffer(bh);
- } else {
- if (test_set_buffer_locked(bh)) {
+ if (!buffer_mapped(bh))
+ continue;
+ if (wbc->sync_mode != WB_SYNC_NONE) {
+ lock_buffer(bh);
+ } else {
+ if (test_set_buffer_locked(bh)) {
+ if (buffer_dirty(bh))
__set_page_dirty_nobuffers(page);
- continue;
- }
- }
- if (test_clear_buffer_dirty(bh)) {
- if (!buffer_uptodate(bh))
- buffer_error();
- mark_buffer_async_write(bh);
- } else {
- unlock_buffer(bh);
+ continue;
}
}
+ if (test_clear_buffer_dirty(bh)) {
+ if (!buffer_uptodate(bh))
+ buffer_error();
+ mark_buffer_async_write(bh);
+ } else {
+ unlock_buffer(bh);
+ }
} while ((bh = bh->b_this_page) != head);
BUG_ON(PageWriteback(page));
- SetPageWriteback(page); /* Keeps try_to_free_buffers() away */
+ set_page_writeback(page); /* Keeps try_to_free_buffers() away */
unlock_page(page);
/*
@@ -1889,7 +1892,7 @@ recover:
} while ((bh = bh->b_this_page) != head);
SetPageError(page);
BUG_ON(PageWriteback(page));
- SetPageWriteback(page);
+ set_page_writeback(page);
unlock_page(page);
do {
struct buffer_head *next = bh->b_this_page;
@@ -2924,7 +2927,10 @@ EXPORT_SYMBOL(try_to_free_buffers);
int block_sync_page(struct page *page)
{
- blk_run_queues();
+ struct address_space *mapping;
+ smp_mb();
+ mapping = page->mapping;
+ blk_run_address_space(mapping);
return 0;
}
--- diff/fs/cifs/file.c 2004-02-18 08:54:12.000000000 +0000
+++ source/fs/cifs/file.c 2004-03-16 11:27:48.385831896 +0000
@@ -898,11 +898,9 @@ static void cifs_copy_cache_pages(struct
if(list_empty(pages))
break;
- spin_lock(&mapping->page_lock);
- page = list_entry(pages->prev, struct page, list);
+ page = list_entry(pages->prev, struct page, lru);
- list_del(&page->list);
- spin_unlock(&mapping->page_lock);
+ list_del(&page->lru);
if (add_to_page_cache(page, mapping, page->index, GFP_KERNEL)) {
page_cache_release(page);
@@ -962,14 +960,10 @@ cifs_readpages(struct file *file, struct
pagevec_init(&lru_pvec, 0);
for(i = 0;i<num_pages;) {
- spin_lock(&mapping->page_lock);
- if(list_empty(page_list)) {
- spin_unlock(&mapping->page_lock);
+ if(list_empty(page_list))
break;
- }
- page = list_entry(page_list->prev, struct page, list);
+ page = list_entry(page_list->prev, struct page, lru);
offset = (loff_t)page->index << PAGE_CACHE_SHIFT;
- spin_unlock(&mapping->page_lock);
/* for reads over a certain size could initiate async read ahead */
@@ -989,12 +983,11 @@ cifs_readpages(struct file *file, struct
cFYI(1,("Read error in readpages: %d",rc));
/* clean up remaing pages off list */
- spin_lock(&mapping->page_lock);
while (!list_empty(page_list) && (i < num_pages)) {
- page = list_entry(page_list->prev, struct page, list);
- list_del(&page->list);
+ page = list_entry(page_list->prev,
+ struct page, lru);
+ list_del(&page->lru);
}
- spin_unlock(&mapping->page_lock);
break;
} else if (bytes_read > 0) {
pSMBr = (struct smb_com_read_rsp *)smb_read_data;
@@ -1010,8 +1003,9 @@ cifs_readpages(struct file *file, struct
cFYI(1,("No bytes read cleaning remaining pages off readahead list"));
/* BB turn off caching and do new lookup on file size at server? */
while (!list_empty(page_list) && (i < num_pages)) {
- page = list_entry(page_list->prev, struct page, list);
- list_del(&page->list);
+ page = list_entry(page_list->prev,
+ struct page, lru);
+ list_del(&page->lru);
}
break;
--- diff/fs/compat_ioctl.c 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/compat_ioctl.c 2004-03-16 11:27:48.387831592 +0000
@@ -1604,7 +1604,7 @@ static int vt_check(struct file *file)
* To have permissions to do most of the vt ioctls, we either have
* to be the owner of the tty, or super-user.
*/
- if (current->tty == tty || capable(CAP_SYS_ADMIN))
+ if (current->signal->tty == tty || capable(CAP_SYS_ADMIN))
return 1;
return 0;
}
--- diff/fs/direct-io.c 2004-01-19 10:22:59.000000000 +0000
+++ source/fs/direct-io.c 2004-03-16 11:27:48.389831288 +0000
@@ -52,6 +52,10 @@
*
* If blkfactor is zero then the user's request was aligned to the filesystem's
* blocksize.
+ *
+ * needs_locking is set for regular files on direct-IO-naive filesystems. It
+ * determines whether we need to do the fancy locking which prevents direct-IO
+ * from being able to read uninitialised disk blocks.
*/
struct dio {
@@ -59,6 +63,7 @@ struct dio {
struct bio *bio; /* bio under assembly */
struct inode *inode;
int rw;
+ int needs_locking; /* doesn't change */
unsigned blkbits; /* doesn't change */
unsigned blkfactor; /* When we're using an alignment which
is finer than the filesystem's soft
@@ -69,6 +74,7 @@ struct dio {
been performed at the start of a
write */
int pages_in_io; /* approximate total IO pages */
+ size_t size; /* total request size (doesn't change)*/
sector_t block_in_file; /* Current offset into the underlying
file in dio_block units. */
unsigned blocks_available; /* At block_in_file. changes */
@@ -110,9 +116,9 @@ struct dio {
int page_errors; /* errno from get_user_pages() */
/* BIO completion state */
- atomic_t bio_count; /* nr bios to be completed */
- atomic_t bios_in_flight; /* nr bios in flight */
- spinlock_t bio_list_lock; /* protects bio_list */
+ spinlock_t bio_lock; /* protects BIO fields below */
+ int bio_count; /* nr bios to be completed */
+ int bios_in_flight; /* nr bios in flight */
struct bio *bio_list; /* singly linked via bi_private */
struct task_struct *waiter; /* waiting task (NULL if none) */
@@ -204,8 +210,10 @@ static struct page *dio_get_page(struct
*/
static void dio_complete(struct dio *dio, loff_t offset, ssize_t bytes)
{
- if (dio->end_io)
+ if (dio->end_io && dio->result)
dio->end_io(dio->inode, offset, bytes, dio->map_bh.b_private);
+ if (dio->needs_locking)
+ up_read(&dio->inode->i_alloc_sem);
}
/*
@@ -214,14 +222,38 @@ static void dio_complete(struct dio *dio
*/
static void finished_one_bio(struct dio *dio)
{
- if (atomic_dec_and_test(&dio->bio_count)) {
+ unsigned long flags;
+
+ spin_lock_irqsave(&dio->bio_lock, flags);
+ if (dio->bio_count == 1) {
if (dio->is_async) {
+ /*
+ * Last reference to the dio is going away.
+ * Drop spinlock and complete the DIO.
+ */
+ spin_unlock_irqrestore(&dio->bio_lock, flags);
dio_complete(dio, dio->block_in_file << dio->blkbits,
dio->result);
- aio_complete(dio->iocb, dio->result, 0);
- kfree(dio);
+ /* Complete AIO later if falling back to buffered i/o */
+ if (dio->result == dio->size || dio->rw == READ) {
+ aio_complete(dio->iocb, dio->result, 0);
+ kfree(dio);
+ return;
+ } else {
+ /*
+ * Falling back to buffered
+ */
+ spin_lock_irqsave(&dio->bio_lock, flags);
+ dio->bio_count--;
+ if (dio->waiter)
+ wake_up_process(dio->waiter);
+ spin_unlock_irqrestore(&dio->bio_lock, flags);
+ return;
+ }
}
}
+ dio->bio_count--;
+ spin_unlock_irqrestore(&dio->bio_lock, flags);
}
static int dio_bio_complete(struct dio *dio, struct bio *bio);
@@ -255,13 +287,13 @@ static int dio_bio_end_io(struct bio *bi
if (bio->bi_size)
return 1;
- spin_lock_irqsave(&dio->bio_list_lock, flags);
+ spin_lock_irqsave(&dio->bio_lock, flags);
bio->bi_private = dio->bio_list;
dio->bio_list = bio;
- atomic_dec(&dio->bios_in_flight);
- if (dio->waiter && atomic_read(&dio->bios_in_flight) == 0)
+ dio->bios_in_flight--;
+ if (dio->waiter && dio->bios_in_flight == 0)
wake_up_process(dio->waiter);
- spin_unlock_irqrestore(&dio->bio_list_lock, flags);
+ spin_unlock_irqrestore(&dio->bio_lock, flags);
return 0;
}
@@ -294,10 +326,13 @@ dio_bio_alloc(struct dio *dio, struct bl
static void dio_bio_submit(struct dio *dio)
{
struct bio *bio = dio->bio;
+ unsigned long flags;
bio->bi_private = dio;
- atomic_inc(&dio->bio_count);
- atomic_inc(&dio->bios_in_flight);
+ spin_lock_irqsave(&dio->bio_lock, flags);
+ dio->bio_count++;
+ dio->bios_in_flight++;
+ spin_unlock_irqrestore(&dio->bio_lock, flags);
if (dio->is_async && dio->rw == READ)
bio_set_pages_dirty(bio);
submit_bio(dio->rw, bio);
@@ -323,22 +358,22 @@ static struct bio *dio_await_one(struct
unsigned long flags;
struct bio *bio;
- spin_lock_irqsave(&dio->bio_list_lock, flags);
+ spin_lock_irqsave(&dio->bio_lock, flags);
while (dio->bio_list == NULL) {
set_current_state(TASK_UNINTERRUPTIBLE);
if (dio->bio_list == NULL) {
dio->waiter = current;
- spin_unlock_irqrestore(&dio->bio_list_lock, flags);
- blk_run_queues();
+ spin_unlock_irqrestore(&dio->bio_lock, flags);
+ blk_run_address_space(dio->inode->i_mapping);
io_schedule();
- spin_lock_irqsave(&dio->bio_list_lock, flags);
+ spin_lock_irqsave(&dio->bio_lock, flags);
dio->waiter = NULL;
}
set_current_state(TASK_RUNNING);
}
bio = dio->bio_list;
dio->bio_list = bio->bi_private;
- spin_unlock_irqrestore(&dio->bio_list_lock, flags);
+ spin_unlock_irqrestore(&dio->bio_lock, flags);
return bio;
}
@@ -380,7 +415,12 @@ static int dio_await_completion(struct d
if (dio->bio)
dio_bio_submit(dio);
- while (atomic_read(&dio->bio_count)) {
+ /*
+ * The bio_lock is not held for the read of bio_count.
+ * This is ok since it is the dio_bio_complete() that changes
+ * bio_count.
+ */
+ while (dio->bio_count) {
struct bio *bio = dio_await_one(dio);
int ret2;
@@ -407,10 +447,10 @@ static int dio_bio_reap(struct dio *dio)
unsigned long flags;
struct bio *bio;
- spin_lock_irqsave(&dio->bio_list_lock, flags);
+ spin_lock_irqsave(&dio->bio_lock, flags);
bio = dio->bio_list;
dio->bio_list = bio->bi_private;
- spin_unlock_irqrestore(&dio->bio_list_lock, flags);
+ spin_unlock_irqrestore(&dio->bio_lock, flags);
ret = dio_bio_complete(dio, bio);
}
dio->reap_counter = 0;
@@ -449,6 +489,7 @@ static int get_more_blocks(struct dio *d
unsigned long fs_count; /* Number of filesystem-sized blocks */
unsigned long dio_count;/* Number of dio_block-sized blocks */
unsigned long blkmask;
+ int beyond_eof = 0;
/*
* If there was a memory error and we've overwritten all the
@@ -466,8 +507,19 @@ static int get_more_blocks(struct dio *d
if (dio_count & blkmask)
fs_count++;
+ if (dio->needs_locking) {
+ if (dio->block_in_file >= (i_size_read(dio->inode) >>
+ dio->blkbits))
+ beyond_eof = 1;
+ }
+ /*
+ * For writes inside i_size we forbid block creations: only
+ * overwrites are permitted. We fall back to buffered writes
+ * at a higher level for inside-i_size block-instantiating
+ * writes.
+ */
ret = (*dio->get_blocks)(dio->inode, fs_startblk, fs_count,
- map_bh, dio->rw == WRITE);
+ map_bh, (dio->rw == WRITE) && beyond_eof);
}
return ret;
}
@@ -774,6 +826,10 @@ do_holes:
if (!buffer_mapped(map_bh)) {
char *kaddr;
+ /* AKPM: eargh, -ENOTBLK is a hack */
+ if (dio->rw == WRITE)
+ return -ENOTBLK;
+
if (dio->block_in_file >=
i_size_read(dio->inode)>>blkbits) {
/* We hit eof */
@@ -839,32 +895,30 @@ out:
return ret;
}
+/*
+ * Releases both i_sem and i_alloc_sem
+ */
static int
direct_io_worker(int rw, struct kiocb *iocb, struct inode *inode,
const struct iovec *iov, loff_t offset, unsigned long nr_segs,
- unsigned blkbits, get_blocks_t get_blocks, dio_iodone_t end_io)
+ unsigned blkbits, get_blocks_t get_blocks, dio_iodone_t end_io,
+ struct dio *dio)
{
unsigned long user_addr;
int seg;
int ret = 0;
int ret2;
- struct dio *dio;
size_t bytes;
- dio = kmalloc(sizeof(*dio), GFP_KERNEL);
- if (!dio)
- return -ENOMEM;
- dio->is_async = !is_sync_kiocb(iocb);
-
dio->bio = NULL;
dio->inode = inode;
dio->rw = rw;
dio->blkbits = blkbits;
dio->blkfactor = inode->i_blkbits - blkbits;
dio->start_zero_done = 0;
+ dio->size = 0;
dio->block_in_file = offset >> blkbits;
dio->blocks_available = 0;
-
dio->cur_page = NULL;
dio->boundary = 0;
@@ -887,9 +941,9 @@ direct_io_worker(int rw, struct kiocb *i
* (or synchronous) device could take the count to zero while we're
* still submitting BIOs.
*/
- atomic_set(&dio->bio_count, 1);
- atomic_set(&dio->bios_in_flight, 0);
- spin_lock_init(&dio->bio_list_lock);
+ dio->bio_count = 1;
+ dio->bios_in_flight = 0;
+ spin_lock_init(&dio->bio_lock);
dio->bio_list = NULL;
dio->waiter = NULL;
@@ -899,7 +953,7 @@ direct_io_worker(int rw, struct kiocb *i
for (seg = 0; seg < nr_segs; seg++) {
user_addr = (unsigned long)iov[seg].iov_base;
- bytes = iov[seg].iov_len;
+ dio->size += bytes = iov[seg].iov_len;
/* Index into the first page of the first block */
dio->first_block_in_page = (user_addr & ~PAGE_MASK) >> blkbits;
@@ -930,6 +984,13 @@ direct_io_worker(int rw, struct kiocb *i
}
} /* end iovec loop */
+ if (ret == -ENOTBLK && rw == WRITE) {
+ /*
+ * The remaining part of the request will be
+ * be handled by buffered I/O when we return
+ */
+ ret = 0;
+ }
/*
* There may be some unwritten disk at the end of a part-written
* fs-block-sized block. Go zero that now.
@@ -953,14 +1014,48 @@ direct_io_worker(int rw, struct kiocb *i
dio_cleanup(dio);
/*
+ * All block lookups have been performed. For READ requests
+ * we can let i_sem go now that its achieved its purpose
+ * of protecting us from looking up uninitialized blocks.
+ */
+ if ((rw == READ) && dio->needs_locking)
+ up(&dio->inode->i_sem);
+
+ /*
* OK, all BIOs are submitted, so we can decrement bio_count to truly
* reflect the number of to-be-processed BIOs.
*/
if (dio->is_async) {
+ int should_wait = 0;
+
+ if (dio->result < dio->size && rw == WRITE) {
+ dio->waiter = current;
+ should_wait = 1;
+ }
if (ret == 0)
- ret = dio->result; /* Bytes written */
+ ret = dio->result;
finished_one_bio(dio); /* This can free the dio */
- blk_run_queues();
+ blk_run_address_space(inode->i_mapping);
+ if (should_wait) {
+ unsigned long flags;
+ /*
+ * Wait for already issued I/O to drain out and
+ * release its references to user-space pages
+ * before returning to fallback on buffered I/O
+ */
+
+ spin_lock_irqsave(&dio->bio_lock, flags);
+ set_current_state(TASK_UNINTERRUPTIBLE);
+ while (dio->bio_count) {
+ spin_unlock_irqrestore(&dio->bio_lock, flags);
+ io_schedule();
+ spin_lock_irqsave(&dio->bio_lock, flags);
+ set_current_state(TASK_UNINTERRUPTIBLE);
+ }
+ spin_unlock_irqrestore(&dio->bio_lock, flags);
+ set_current_state(TASK_RUNNING);
+ kfree(dio);
+ }
} else {
finished_one_bio(dio);
ret2 = dio_await_completion(dio);
@@ -980,6 +1075,14 @@ direct_io_worker(int rw, struct kiocb *i
ret = i_size - offset;
}
dio_complete(dio, offset, ret);
+ /* We could have also come here on an AIO file extend */
+ if (!is_sync_kiocb(iocb) && rw == WRITE &&
+ ret >= 0 && dio->result == dio->size)
+ /*
+ * For AIO writes where we have completed the
+ * i/o, we have to mark the the aio complete.
+ */
+ aio_complete(iocb, ret, 0);
kfree(dio);
}
return ret;
@@ -987,11 +1090,17 @@ direct_io_worker(int rw, struct kiocb *i
/*
* This is a library function for use by filesystem drivers.
+ *
+ * For writes to S_ISREG files, we are called under i_sem and return with i_sem
+ * held, even though it is internally dropped.
+ *
+ * For writes to S_ISBLK files, i_sem is not held on entry; it is never taken.
*/
int
-blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
+__blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
struct block_device *bdev, const struct iovec *iov, loff_t offset,
- unsigned long nr_segs, get_blocks_t get_blocks, dio_iodone_t end_io)
+ unsigned long nr_segs, get_blocks_t get_blocks, dio_iodone_t end_io,
+ int needs_special_locking)
{
int seg;
size_t size;
@@ -1000,6 +1109,9 @@ blockdev_direct_IO(int rw, struct kiocb
unsigned bdev_blkbits = 0;
unsigned blocksize_mask = (1 << blkbits) - 1;
ssize_t retval = -EINVAL;
+ loff_t end = offset;
+ struct dio *dio;
+ int needs_locking;
if (bdev)
bdev_blkbits = blksize_bits(bdev_hardsect_size(bdev));
@@ -1016,6 +1128,7 @@ blockdev_direct_IO(int rw, struct kiocb
for (seg = 0; seg < nr_segs; seg++) {
addr = (unsigned long)iov[seg].iov_base;
size = iov[seg].iov_len;
+ end += size;
if ((addr & blocksize_mask) || (size & blocksize_mask)) {
if (bdev)
blkbits = bdev_blkbits;
@@ -1025,10 +1138,46 @@ blockdev_direct_IO(int rw, struct kiocb
}
}
- retval = direct_io_worker(rw, iocb, inode, iov, offset,
- nr_segs, blkbits, get_blocks, end_io);
+ dio = kmalloc(sizeof(*dio), GFP_KERNEL);
+ retval = -ENOMEM;
+ if (!dio)
+ goto out;
+
+ /*
+ * For regular files,
+ * readers need to grab i_sem and i_alloc_sem
+ * writers need to grab i_alloc_sem only (i_sem is already held)
+ */
+ needs_locking = 0;
+ if (S_ISREG(inode->i_mode) && needs_special_locking) {
+ needs_locking = 1;
+ if (rw == READ) {
+ struct address_space *mapping;
+
+ mapping = iocb->ki_filp->f_mapping;
+ down(&inode->i_sem);
+ retval = filemap_write_and_wait(mapping);
+ if (retval) {
+ up(&inode->i_sem);
+ kfree(dio);
+ goto out;
+ }
+ }
+ down_read(&inode->i_alloc_sem);
+ }
+ dio->needs_locking = needs_locking;
+ /*
+ * For file extending writes updating i_size before data
+ * writeouts complete can expose uninitialized blocks. So
+ * even for AIO, we need to wait for i/o to complete before
+ * returning in this case.
+ */
+ dio->is_async = !is_sync_kiocb(iocb) && !((rw == WRITE) &&
+ (end > i_size_read(inode)));
+
+ retval = direct_io_worker(rw, iocb, inode, iov, offset,
+ nr_segs, blkbits, get_blocks, end_io, dio);
out:
return retval;
}
-
-EXPORT_SYMBOL(blockdev_direct_IO);
+EXPORT_SYMBOL(__blockdev_direct_IO);
--- diff/fs/dquot.c 2004-02-09 10:36:12.000000000 +0000
+++ source/fs/dquot.c 2004-03-16 11:27:48.392830832 +0000
@@ -1,16 +1,13 @@
/*
- * Implementation of the diskquota system for the LINUX operating
- * system. QUOTA is implemented using the BSD system call interface as
- * the means of communication with the user level. Currently only the
- * ext2 filesystem has support for disk quotas. Other filesystems may
- * be added in the future. This file contains the generic routines
- * called by the different filesystems on allocation of an inode or
- * block. These routines take care of the administration needed to
- * have a consistent diskquota tracking system. The ideas of both
- * user and group quotas are based on the Melbourne quota system as
- * used on BSD derived systems. The internal implementation is
- * based on one of the several variants of the LINUX inode-subsystem
- * with added complexity of the diskquota system.
+ * Implementation of the diskquota system for the LINUX operating system. QUOTA
+ * is implemented using the BSD system call interface as the means of
+ * communication with the user level. This file contains the generic routines
+ * called by the different filesystems on allocation of an inode or block.
+ * These routines take care of the administration needed to have a consistent
+ * diskquota tracking system. The ideas of both user and group quotas are based
+ * on the Melbourne quota system as used on BSD derived systems. The internal
+ * implementation is based on one of the several variants of the LINUX
+ * inode-subsystem with added complexity of the diskquota system.
*
* Version: $Id: dquot.c,v 6.3 1996/11/17 18:35:34 mvw Exp mvw $
*
@@ -52,6 +49,9 @@
* New SMP locking.
* Jan Kara, <jack@suse.cz>, 10/2002
*
+ * Added journalled quota support
+ * Jan Kara, <jack@suse.cz>, 2003,2004
+ *
* (C) Copyright 1994 - 1997 Marco van Wieringen
*/
@@ -85,13 +85,36 @@
* and quota formats and also dqstats structure containing statistics about the
* lists. dq_data_lock protects data from dq_dqb and also mem_dqinfo structures
* and also guards consistency of dquot->dq_dqb with inode->i_blocks, i_bytes.
- * Note that we don't have to do the locking of i_blocks and i_bytes when the
- * quota is disabled - i_sem should serialize the access. dq_data_lock should
- * be always grabbed before dq_list_lock.
+ * i_blocks and i_bytes updates itself are guarded by i_lock acquired directly
+ * in inode_add_bytes() and inode_sub_bytes().
+ *
+ * The spinlock ordering is hence: dq_data_lock > dq_list_lock > i_lock
*
* Note that some things (eg. sb pointer, type, id) doesn't change during
* the life of the dquot structure and so needn't to be protected by a lock
+ *
+ * Any operation working on dquots via inode pointers must hold dqptr_sem. If
+ * operation is just reading pointers from inode (or not using them at all) the
+ * read lock is enough. If pointers are altered function must hold write lock.
+ * If operation is holding reference to dquot in other way (e.g. quotactl ops)
+ * it must be guarded by dqonoff_sem.
+ * This locking assures that:
+ * a) update/access to dquot pointers in inode is serialized
+ * b) everyone is guarded against invalidate_dquots()
+ *
+ * Each dquot has its dq_lock semaphore. Locked dquots might not be referenced
+ * from inodes (dquot_alloc_space() and such don't check the dq_lock).
+ * Currently dquot is locked only when it is being read to memory (or space for
+ * it is being allocated) on the first dqget() and when it is being released on
+ * the last dqput(). The allocation and release oparations are serialized by
+ * the dq_lock and by checking the use count in dquot_release(). Write
+ * operations on dquots don't hold dq_lock as they copy data under dq_data_lock
+ * spinlock to internal buffers before writing.
+ *
+ * Lock ordering (including journal_lock) is following:
+ * dqonoff_sem > journal_lock > dqptr_sem > dquot->dq_lock > dqio_sem
*/
+
spinlock_t dq_list_lock = SPIN_LOCK_UNLOCKED;
spinlock_t dq_data_lock = SPIN_LOCK_UNLOCKED;
@@ -169,23 +192,6 @@ static void put_quota_format(struct quot
* mechanism to locate a specific dquot.
*/
-/*
- * Note that any operation which operates on dquot data (ie. dq_dqb) must
- * hold dq_data_lock.
- *
- * Any operation working with dquots must hold dqptr_sem. If operation is
- * just reading pointers from inodes than read lock is enough. If pointers
- * are altered function must hold write lock.
- *
- * Locked dquots might not be referenced in inodes. Currently dquot it locked
- * only once in its existence - when it's being read to memory on first dqget()
- * and at that time it can't be referenced from inode. Write operations on
- * dquots don't hold dquot lock as they copy data to internal buffers before
- * writing anyway and copying as well as any data update should be atomic. Also
- * nobody can change used entries in dquot structure as this is done only when
- * quota is destroyed and invalidate_dquots() is called only when dq_count == 0.
- */
-
static LIST_HEAD(inuse_list);
static LIST_HEAD(free_dquots);
static struct list_head dquot_hash[NR_DQHASH];
@@ -254,6 +260,9 @@ static inline void remove_inuse(struct d
dqstats.allocated_dquots--;
list_del(&dquot->dq_inuse);
}
+/*
+ * End of list functions needing dq_list_lock
+ */
static void wait_on_dquot(struct dquot *dquot)
{
@@ -261,34 +270,98 @@ static void wait_on_dquot(struct dquot *
up(&dquot->dq_lock);
}
-static int read_dqblk(struct dquot *dquot)
+#define mark_dquot_dirty(dquot) ((dquot)->dq_sb->dq_op->mark_dirty(dquot))
+
+/* No locks needed here as ANY_DQUOT_DIRTY is used just by sync and so the
+ * worst what can happen is that dquot is not written by concurrent sync... */
+int dquot_mark_dquot_dirty(struct dquot *dquot)
+{
+ set_bit(DQ_MOD_B, &(dquot)->dq_flags);
+ set_bit(DQF_ANY_DQUOT_DIRTY_B, &(sb_dqopt((dquot)->dq_sb)->
+ info[(dquot)->dq_type].dqi_flags));
+ return 0;
+}
+
+void mark_info_dirty(struct super_block *sb, int type)
{
- int ret;
+ set_bit(DQF_INFO_DIRTY_B, &sb_dqopt(sb)->info[type].dqi_flags);
+}
+EXPORT_SYMBOL(mark_info_dirty);
+
+/*
+ * Read dquot from disk and alloc space for it
+ */
+
+int dquot_acquire(struct dquot *dquot)
+{
+ int ret = 0;
struct quota_info *dqopt = sb_dqopt(dquot->dq_sb);
down(&dquot->dq_lock);
down(&dqopt->dqio_sem);
- ret = dqopt->ops[dquot->dq_type]->read_dqblk(dquot);
+ if (!test_bit(DQ_READ_B, &dquot->dq_flags))
+ ret = dqopt->ops[dquot->dq_type]->read_dqblk(dquot);
+ if (ret < 0)
+ goto out_iolock;
+ set_bit(DQ_READ_B, &dquot->dq_flags);
+ /* Instantiate dquot if needed */
+ if (!test_bit(DQ_ACTIVE_B, &dquot->dq_flags) && !dquot->dq_off) {
+ ret = dqopt->ops[dquot->dq_type]->commit_dqblk(dquot);
+ if (ret < 0)
+ goto out_iolock;
+ }
+ set_bit(DQ_ACTIVE_B, &dquot->dq_flags);
+out_iolock:
up(&dqopt->dqio_sem);
up(&dquot->dq_lock);
return ret;
}
-static int commit_dqblk(struct dquot *dquot)
+/*
+ * Write dquot to disk
+ */
+int dquot_commit(struct dquot *dquot)
{
- int ret;
+ int ret = 0;
struct quota_info *dqopt = sb_dqopt(dquot->dq_sb);
down(&dqopt->dqio_sem);
- ret = dqopt->ops[dquot->dq_type]->commit_dqblk(dquot);
+ clear_bit(DQ_MOD_B, &dquot->dq_flags);
+ /* Inactive dquot can be only if there was error during read/init
+ * => we have better not writing it */
+ if (test_bit(DQ_ACTIVE_B, &dquot->dq_flags))
+ ret = dqopt->ops[dquot->dq_type]->commit_dqblk(dquot);
up(&dqopt->dqio_sem);
+ if (info_dirty(&dqopt->info[dquot->dq_type]))
+ dquot->dq_sb->dq_op->write_info(dquot->dq_sb, dquot->dq_type);
+ return ret;
+}
+
+/*
+ * Release dquot
+ */
+int dquot_release(struct dquot *dquot)
+{
+ int ret = 0;
+ struct quota_info *dqopt = sb_dqopt(dquot->dq_sb);
+
+ down(&dquot->dq_lock);
+ /* Check whether we are not racing with some other dqget() */
+ if (atomic_read(&dquot->dq_count) > 1)
+ goto out_dqlock;
+ down(&dqopt->dqio_sem);
+ ret = dqopt->ops[dquot->dq_type]->release_dqblk(dquot);
+ clear_bit(DQ_ACTIVE_B, &dquot->dq_flags);
+ up(&dqopt->dqio_sem);
+out_dqlock:
+ up(&dquot->dq_lock);
return ret;
}
/* Invalidate all dquots on the list. Note that this function is called after
- * quota is disabled so no new quota might be created. Because we hold dqptr_sem
- * for writing and pointers were already removed from inodes we actually know that
- * no quota for this sb+type should be held. */
+ * quota is disabled and pointers from inodes removed so there cannot be new
+ * quota users. Also because we hold dqonoff_sem there can be no quota users
+ * for this sb+type at all. */
static void invalidate_dquots(struct super_block *sb, int type)
{
struct dquot *dquot;
@@ -302,12 +375,11 @@ static void invalidate_dquots(struct sup
continue;
if (dquot->dq_type != type)
continue;
-#ifdef __DQUOT_PARANOIA
- /* There should be no users of quota - we hold dqptr_sem for writing */
+#ifdef __DQUOT_PARANOIA
if (atomic_read(&dquot->dq_count))
BUG();
#endif
- /* Quota now have no users and it has been written on last dqput() */
+ /* Quota now has no users and it has been written on last dqput() */
remove_dquot_hash(dquot);
remove_free_dquot(dquot);
remove_inuse(dquot);
@@ -316,20 +388,22 @@ static void invalidate_dquots(struct sup
spin_unlock(&dq_list_lock);
}
-static int vfs_quota_sync(struct super_block *sb, int type)
+int vfs_quota_sync(struct super_block *sb, int type)
{
struct list_head *head;
struct dquot *dquot;
struct quota_info *dqopt = sb_dqopt(sb);
int cnt;
- down_read(&dqopt->dqptr_sem);
+ down(&dqopt->dqonoff_sem);
restart:
/* At this point any dirty dquot will definitely be written so we can clear
dirty flag from info */
+ spin_lock(&dq_data_lock);
for (cnt = 0; cnt < MAXQUOTAS; cnt++)
if ((cnt == type || type == -1) && sb_has_quota_enabled(sb, cnt))
clear_bit(DQF_ANY_DQUOT_DIRTY_B, &dqopt->info[cnt].dqi_flags);
+ spin_unlock(&dq_data_lock);
spin_lock(&dq_list_lock);
list_for_each(head, &inuse_list) {
dquot = list_entry(head, struct dquot, dq_inuse);
@@ -337,10 +411,13 @@ restart:
continue;
if (type != -1 && dquot->dq_type != type)
continue;
- if (!dquot->dq_sb) /* Invalidated? */
- continue;
if (!dquot_dirty(dquot))
continue;
+ /* Dirty and inactive can be only bad dquot... */
+ if (!test_bit(DQ_ACTIVE_B, &dquot->dq_flags))
+ continue;
+ /* Now we have active dquot from which someone is holding reference so we
+ * can safely just increase use count */
atomic_inc(&dquot->dq_count);
dqstats.lookups++;
spin_unlock(&dq_list_lock);
@@ -351,15 +428,13 @@ restart:
spin_unlock(&dq_list_lock);
for (cnt = 0; cnt < MAXQUOTAS; cnt++)
- if ((cnt == type || type == -1) && sb_has_quota_enabled(sb, cnt) && info_dirty(&dqopt->info[cnt])) {
- down(&dqopt->dqio_sem);
- dqopt->ops[cnt]->write_file_info(sb, cnt);
- up(&dqopt->dqio_sem);
- }
+ if ((cnt == type || type == -1) && sb_has_quota_enabled(sb, cnt)
+ && info_dirty(&dqopt->info[cnt]))
+ sb->dq_op->write_info(sb, cnt);
spin_lock(&dq_list_lock);
dqstats.syncs++;
spin_unlock(&dq_list_lock);
- up_read(&dqopt->dqptr_sem);
+ up(&dqopt->dqonoff_sem);
return 0;
}
@@ -402,7 +477,7 @@ static int shrink_dqcache_memory(int nr,
/*
* Put reference to dquot
* NOTE: If you change this function please check whether dqput_blocks() works right...
- * MUST be called with dqptr_sem held
+ * MUST be called with either dqptr_sem or dqonoff_sem
*/
static void dqput(struct dquot *dquot)
{
@@ -430,11 +505,20 @@ we_slept:
spin_unlock(&dq_list_lock);
return;
}
- if (dquot_dirty(dquot)) {
+ /* Need to release dquot? */
+ if (test_bit(DQ_ACTIVE_B, &dquot->dq_flags) && dquot_dirty(dquot)) {
spin_unlock(&dq_list_lock);
+ /* Commit dquot before releasing */
dquot->dq_sb->dq_op->write_dquot(dquot);
goto we_slept;
}
+ /* Clear flag in case dquot was inactive (something bad happened) */
+ clear_bit(DQ_MOD_B, &dquot->dq_flags);
+ if (test_bit(DQ_ACTIVE_B, &dquot->dq_flags)) {
+ spin_unlock(&dq_list_lock);
+ dquot_release(dquot);
+ goto we_slept;
+ }
atomic_dec(&dquot->dq_count);
#ifdef __DQUOT_PARANOIA
/* sanity check */
@@ -467,7 +551,7 @@ static struct dquot *get_empty_dquot(str
/*
* Get reference to dquot
- * MUST be called with dqptr_sem held
+ * MUST be called with dqptr_sem or dqonoff_sem held
*/
static struct dquot *dqget(struct super_block *sb, unsigned int id, int type)
{
@@ -493,7 +577,6 @@ we_slept:
insert_dquot_hash(dquot);
dqstats.lookups++;
spin_unlock(&dq_list_lock);
- read_dqblk(dquot);
} else {
if (!atomic_read(&dquot->dq_count))
remove_free_dquot(dquot);
@@ -501,11 +584,17 @@ we_slept:
dqstats.cache_hits++;
dqstats.lookups++;
spin_unlock(&dq_list_lock);
- wait_on_dquot(dquot);
if (empty)
kmem_cache_free(dquot_cachep, empty);
}
-
+ /* Wait for dq_lock - after this we know that either dquot_release() is already
+ * finished or it will be canceled due to dq_count > 1 test */
+ wait_on_dquot(dquot);
+ /* Read the dquot and instantiate it (everything done only if needed) */
+ if (!test_bit(DQ_ACTIVE_B, &dquot->dq_flags) && dquot_acquire(dquot) < 0) {
+ dqput(dquot);
+ return NODQUOT;
+ }
#ifdef __DQUOT_PARANOIA
if (!dquot->dq_sb) /* Has somebody invalidated entry under us? */
BUG();
@@ -528,7 +617,7 @@ static int dqinit_needed(struct inode *i
return 0;
}
-/* This routine is guarded by dqptr_sem semaphore */
+/* This routine is guarded by dqonoff_sem semaphore */
static void add_dquot_ref(struct super_block *sb, int type)
{
struct list_head *p;
@@ -539,12 +628,10 @@ restart:
struct file *filp = list_entry(p, struct file, f_list);
struct inode *inode = filp->f_dentry->d_inode;
if (filp->f_mode & FMODE_WRITE && dqinit_needed(inode, type)) {
- struct vfsmount *mnt = mntget(filp->f_vfsmnt);
struct dentry *dentry = dget(filp->f_dentry);
file_list_unlock();
sb->dq_op->initialize(inode, type);
dput(dentry);
- mntput(mnt);
/* As we may have blocked we had better restart... */
goto restart;
}
@@ -594,7 +681,7 @@ put_it:
/* Free list of dquots - called from inode.c */
/* dquots are removed from inodes, no new references can be got so we are the only ones holding reference */
-void put_dquot_list(struct list_head *tofree_head)
+static void put_dquot_list(struct list_head *tofree_head)
{
struct list_head *act_head;
struct dquot *dquot;
@@ -609,16 +696,28 @@ void put_dquot_list(struct list_head *to
}
}
+/* Function in inode.c - remove pointers to dquots in icache */
+extern void remove_dquot_ref(struct super_block *, int, struct list_head *);
+
+/* Gather all references from inodes and drop them */
+static void drop_dquot_ref(struct super_block *sb, int type)
+{
+ LIST_HEAD(tofree_head);
+
+ down_write(&sb_dqopt(sb)->dqptr_sem);
+ remove_dquot_ref(sb, type, &tofree_head);
+ up_write(&sb_dqopt(sb)->dqptr_sem);
+ put_dquot_list(&tofree_head);
+}
+
static inline void dquot_incr_inodes(struct dquot *dquot, unsigned long number)
{
dquot->dq_dqb.dqb_curinodes += number;
- mark_dquot_dirty(dquot);
}
static inline void dquot_incr_space(struct dquot *dquot, qsize_t number)
{
dquot->dq_dqb.dqb_curspace += number;
- mark_dquot_dirty(dquot);
}
static inline void dquot_decr_inodes(struct dquot *dquot, unsigned long number)
@@ -630,7 +729,6 @@ static inline void dquot_decr_inodes(str
if (dquot->dq_dqb.dqb_curinodes < dquot->dq_dqb.dqb_isoftlimit)
dquot->dq_dqb.dqb_itime = (time_t) 0;
clear_bit(DQ_INODES_B, &dquot->dq_flags);
- mark_dquot_dirty(dquot);
}
static inline void dquot_decr_space(struct dquot *dquot, qsize_t number)
@@ -642,7 +740,6 @@ static inline void dquot_decr_space(stru
if (toqb(dquot->dq_dqb.dqb_curspace) < dquot->dq_dqb.dqb_bsoftlimit)
dquot->dq_dqb.dqb_btime = (time_t) 0;
clear_bit(DQ_BLKS_B, &dquot->dq_flags);
- mark_dquot_dirty(dquot);
}
static inline int need_print_warning(struct dquot *dquot)
@@ -674,12 +771,12 @@ static void print_warning(struct dquot *
if (!need_print_warning(dquot) || (flag && test_and_set_bit(flag, &dquot->dq_flags)))
return;
- tty_write_message(current->tty, dquot->dq_sb->s_id);
+ tty_write_message(current->signal->tty, dquot->dq_sb->s_id);
if (warntype == ISOFTWARN || warntype == BSOFTWARN)
- tty_write_message(current->tty, ": warning, ");
+ tty_write_message(current->signal->tty, ": warning, ");
else
- tty_write_message(current->tty, ": write failed, ");
- tty_write_message(current->tty, quotatypes[dquot->dq_type]);
+ tty_write_message(current->signal->tty, ": write failed, ");
+ tty_write_message(current->signal->tty, quotatypes[dquot->dq_type]);
switch (warntype) {
case IHARDWARN:
msg = " file limit reached.\n";
@@ -700,7 +797,7 @@ static void print_warning(struct dquot *
msg = " block quota exceeded.\n";
break;
}
- tty_write_message(current->tty, msg);
+ tty_write_message(current->signal->tty, msg);
}
static inline void flush_warnings(struct dquot **dquots, char *warntype)
@@ -795,22 +892,22 @@ static int check_bdq(struct dquot *dquot
}
/*
- * Externally referenced functions through dquot_operations in inode.
- *
- * Note: this is a blocking operation.
+ * Initialize quota pointers in inode
+ * Transaction must be started at entry
*/
-void dquot_initialize(struct inode *inode, int type)
+int dquot_initialize(struct inode *inode, int type)
{
unsigned int id = 0;
- int cnt;
+ int cnt, ret = 0;
+ /* First test before acquiring semaphore - solves deadlocks when we
+ * re-enter the quota code and are already holding the semaphore */
+ if (IS_NOQUOTA(inode))
+ return 0;
down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
/* Having dqptr_sem we know NOQUOTA flags can't be altered... */
- if (IS_NOQUOTA(inode)) {
- up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
- return;
- }
- /* Build list of quotas to initialize... */
+ if (IS_NOQUOTA(inode))
+ goto out_err;
for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
if (type != -1 && cnt != type)
continue;
@@ -828,54 +925,39 @@ void dquot_initialize(struct inode *inod
inode->i_flags |= S_QUOTA;
}
}
+out_err:
up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
+ return ret;
}
/*
- * Remove references to quota from inode
- * This function needs dqptr_sem for writing
+ * Release all quotas referenced by inode
+ * Transaction must be started at entry
*/
-static void dquot_drop_iupdate(struct inode *inode, struct dquot **to_drop)
+int dquot_drop(struct inode *inode)
{
int cnt;
+ down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
inode->i_flags &= ~S_QUOTA;
for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
- to_drop[cnt] = inode->i_dquot[cnt];
- inode->i_dquot[cnt] = NODQUOT;
+ if (inode->i_dquot[cnt]) {
+ dqput(inode->i_dquot[cnt]);
+ inode->i_dquot[cnt] = NODQUOT;
+ }
}
-}
-
-/*
- * Release all quotas referenced by inode
- */
-void dquot_drop(struct inode *inode)
-{
- struct dquot *to_drop[MAXQUOTAS];
- int cnt;
-
- down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
- dquot_drop_iupdate(inode, to_drop);
up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
- for (cnt = 0; cnt < MAXQUOTAS; cnt++)
- if (to_drop[cnt] != NODQUOT)
- dqput(to_drop[cnt]);
+ return 0;
}
/*
- * Release all quotas referenced by inode.
- * This function assumes dqptr_sem for writing
+ * Following four functions update i_blocks+i_bytes fields and
+ * quota information (together with appropriate checks)
+ * NOTE: We absolutely rely on the fact that caller dirties
+ * the inode (usually macros in quotaops.h care about this) and
+ * holds a handle for the current transaction so that dquot write and
+ * inode write go into the same transaction.
*/
-void dquot_drop_nolock(struct inode *inode)
-{
- struct dquot *to_drop[MAXQUOTAS];
- int cnt;
-
- dquot_drop_iupdate(inode, to_drop);
- for (cnt = 0; cnt < MAXQUOTAS; cnt++)
- if (to_drop[cnt] != NODQUOT)
- dqput(to_drop[cnt]);
-}
/*
* This operation can block, but only after everything is updated
@@ -885,13 +967,22 @@ int dquot_alloc_space(struct inode *inod
int cnt, ret = NO_QUOTA;
char warntype[MAXQUOTAS];
+ /* First test before acquiring semaphore - solves deadlocks when we
+ * re-enter the quota code and are already holding the semaphore */
+ if (IS_NOQUOTA(inode)) {
+out_add:
+ inode_add_bytes(inode, number);
+ return QUOTA_OK;
+ }
for (cnt = 0; cnt < MAXQUOTAS; cnt++)
warntype[cnt] = NOWARN;
down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+ if (IS_NOQUOTA(inode)) { /* Now we can do reliable test... */
+ up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+ goto out_add;
+ }
spin_lock(&dq_data_lock);
- if (IS_NOQUOTA(inode))
- goto add_bytes;
for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
if (inode->i_dquot[cnt] == NODQUOT)
continue;
@@ -903,11 +994,15 @@ int dquot_alloc_space(struct inode *inod
continue;
dquot_incr_space(inode->i_dquot[cnt], number);
}
-add_bytes:
inode_add_bytes(inode, number);
ret = QUOTA_OK;
warn_put_all:
spin_unlock(&dq_data_lock);
+ if (ret == QUOTA_OK)
+ /* Dirtify all the dquots - this can block when journalling */
+ for (cnt = 0; cnt < MAXQUOTAS; cnt++)
+ if (inode->i_dquot[cnt])
+ mark_dquot_dirty(inode->i_dquot[cnt]);
flush_warnings(inode->i_dquot, warntype);
up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
return ret;
@@ -921,6 +1016,10 @@ int dquot_alloc_inode(const struct inode
int cnt, ret = NO_QUOTA;
char warntype[MAXQUOTAS];
+ /* First test before acquiring semaphore - solves deadlocks when we
+ * re-enter the quota code and are already holding the semaphore */
+ if (IS_NOQUOTA(inode))
+ return QUOTA_OK;
for (cnt = 0; cnt < MAXQUOTAS; cnt++)
warntype[cnt] = NOWARN;
down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
@@ -944,6 +1043,11 @@ int dquot_alloc_inode(const struct inode
ret = QUOTA_OK;
warn_put_all:
spin_unlock(&dq_data_lock);
+ if (ret == QUOTA_OK)
+ /* Dirtify all the dquots - this can block when journalling */
+ for (cnt = 0; cnt < MAXQUOTAS; cnt++)
+ if (inode->i_dquot[cnt])
+ mark_dquot_dirty(inode->i_dquot[cnt]);
flush_warnings((struct dquot **)inode->i_dquot, warntype);
up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
return ret;
@@ -952,36 +1056,53 @@ warn_put_all:
/*
* This is a non-blocking operation.
*/
-void dquot_free_space(struct inode *inode, qsize_t number)
+int dquot_free_space(struct inode *inode, qsize_t number)
{
unsigned int cnt;
+ /* First test before acquiring semaphore - solves deadlocks when we
+ * re-enter the quota code and are already holding the semaphore */
+ if (IS_NOQUOTA(inode)) {
+out_sub:
+ inode_sub_bytes(inode, number);
+ return QUOTA_OK;
+ }
down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+ if (IS_NOQUOTA(inode)) {
+ up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+ goto out_sub;
+ }
spin_lock(&dq_data_lock);
- if (IS_NOQUOTA(inode))
- goto sub_bytes;
for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
if (inode->i_dquot[cnt] == NODQUOT)
continue;
dquot_decr_space(inode->i_dquot[cnt], number);
}
-sub_bytes:
inode_sub_bytes(inode, number);
spin_unlock(&dq_data_lock);
+ /* Dirtify all the dquots - this can block when journalling */
+ for (cnt = 0; cnt < MAXQUOTAS; cnt++)
+ if (inode->i_dquot[cnt])
+ mark_dquot_dirty(inode->i_dquot[cnt]);
up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+ return QUOTA_OK;
}
/*
* This is a non-blocking operation.
*/
-void dquot_free_inode(const struct inode *inode, unsigned long number)
+int dquot_free_inode(const struct inode *inode, unsigned long number)
{
unsigned int cnt;
+ /* First test before acquiring semaphore - solves deadlocks when we
+ * re-enter the quota code and are already holding the semaphore */
+ if (IS_NOQUOTA(inode))
+ return QUOTA_OK;
down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
if (IS_NOQUOTA(inode)) {
up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
- return;
+ return QUOTA_OK;
}
spin_lock(&dq_data_lock);
for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
@@ -990,7 +1111,12 @@ void dquot_free_inode(const struct inode
dquot_decr_inodes(inode->i_dquot[cnt], number);
}
spin_unlock(&dq_data_lock);
+ /* Dirtify all the dquots - this can block when journalling */
+ for (cnt = 0; cnt < MAXQUOTAS; cnt++)
+ if (inode->i_dquot[cnt])
+ mark_dquot_dirty(inode->i_dquot[cnt]);
up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+ return QUOTA_OK;
}
/*
@@ -1007,6 +1133,10 @@ int dquot_transfer(struct inode *inode,
chgid = (iattr->ia_valid & ATTR_GID) && inode->i_gid != iattr->ia_gid;
char warntype[MAXQUOTAS];
+ /* First test before acquiring semaphore - solves deadlocks when we
+ * re-enter the quota code and are already holding the semaphore */
+ if (IS_NOQUOTA(inode))
+ return QUOTA_OK;
/* Clear the arrays */
for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
transfer_to[cnt] = transfer_from[cnt] = NODQUOT;
@@ -1017,7 +1147,9 @@ int dquot_transfer(struct inode *inode,
up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
return QUOTA_OK;
}
- /* First build the transfer_to list - here we can block on reading of dquots... */
+ /* First build the transfer_to list - here we can block on
+ * reading/instantiating of dquots. We know that the transaction for
+ * us was already started so we don't violate lock ranking here */
for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
switch (cnt) {
case USRQUOTA:
@@ -1065,6 +1197,13 @@ int dquot_transfer(struct inode *inode,
ret = QUOTA_OK;
warn_put_all:
spin_unlock(&dq_data_lock);
+ /* Dirtify all the dquots - this can block when journalling */
+ for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
+ if (transfer_from[cnt])
+ mark_dquot_dirty(transfer_from[cnt]);
+ if (transfer_to[cnt])
+ mark_dquot_dirty(transfer_to[cnt]);
+ }
flush_warnings(transfer_to, warntype);
for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
@@ -1078,25 +1217,35 @@ warn_put_all:
}
/*
+ * Write info of quota file to disk
+ */
+int dquot_commit_info(struct super_block *sb, int type)
+{
+ int ret;
+ struct quota_info *dqopt = sb_dqopt(sb);
+
+ down(&dqopt->dqio_sem);
+ ret = dqopt->ops[type]->write_file_info(sb, type);
+ up(&dqopt->dqio_sem);
+ return ret;
+}
+
+/*
* Definitions of diskquota operations.
*/
struct dquot_operations dquot_operations = {
- .initialize = dquot_initialize, /* mandatory */
- .drop = dquot_drop, /* mandatory */
+ .initialize = dquot_initialize,
+ .drop = dquot_drop,
.alloc_space = dquot_alloc_space,
.alloc_inode = dquot_alloc_inode,
.free_space = dquot_free_space,
.free_inode = dquot_free_inode,
.transfer = dquot_transfer,
- .write_dquot = commit_dqblk
+ .write_dquot = dquot_commit,
+ .mark_dirty = dquot_mark_dquot_dirty,
+ .write_info = dquot_commit_info
};
-/* Function used by filesystems for initializing the dquot_operations structure */
-void init_dquot_operations(struct dquot_operations *fsdqops)
-{
- memcpy(fsdqops, &dquot_operations, sizeof(dquot_operations));
-}
-
static inline void set_enable_flags(struct quota_info *dqopt, int type)
{
switch (type) {
@@ -1121,9 +1270,6 @@ static inline void reset_enable_flags(st
}
}
-/* Function in inode.c - remove pointers to dquots in icache */
-extern void remove_dquot_ref(struct super_block *, int);
-
/*
* Turn quota off on a device. type == -1 ==> quotaoff for all types (umount)
*/
@@ -1137,7 +1283,6 @@ int vfs_quota_off(struct super_block *sb
/* We need to serialize quota_off() for device */
down(&dqopt->dqonoff_sem);
- down_write(&dqopt->dqptr_sem);
for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
if (type != -1 && cnt != type)
continue;
@@ -1146,63 +1291,56 @@ int vfs_quota_off(struct super_block *sb
reset_enable_flags(dqopt, cnt);
/* Note: these are blocking operations */
- remove_dquot_ref(sb, cnt);
+ drop_dquot_ref(sb, cnt);
invalidate_dquots(sb, cnt);
/*
* Now all dquots should be invalidated, all writes done so we should be only
* users of the info. No locks needed.
*/
- if (info_dirty(&dqopt->info[cnt])) {
- down(&dqopt->dqio_sem);
- dqopt->ops[cnt]->write_file_info(sb, cnt);
- up(&dqopt->dqio_sem);
- }
+ if (info_dirty(&dqopt->info[cnt]))
+ sb->dq_op->write_info(sb, cnt);
if (dqopt->ops[cnt]->free_file_info)
dqopt->ops[cnt]->free_file_info(sb, cnt);
put_quota_format(dqopt->info[cnt].dqi_format);
fput(dqopt->files[cnt]);
- dqopt->files[cnt] = (struct file *)NULL;
+ dqopt->files[cnt] = NULL;
dqopt->info[cnt].dqi_flags = 0;
dqopt->info[cnt].dqi_igrace = 0;
dqopt->info[cnt].dqi_bgrace = 0;
dqopt->ops[cnt] = NULL;
}
- up_write(&dqopt->dqptr_sem);
up(&dqopt->dqonoff_sem);
out:
return 0;
}
-int vfs_quota_on(struct super_block *sb, int type, int format_id, char *path)
+/*
+ * Turn quotas on on a device
+ */
+
+/* Helper function when we already have file open */
+static int vfs_quota_on_file(struct file *f, int type, int format_id)
{
- struct file *f;
+ struct quota_format_type *fmt = find_quota_format(format_id);
struct inode *inode;
+ struct super_block *sb = f->f_dentry->d_sb;
struct quota_info *dqopt = sb_dqopt(sb);
- struct quota_format_type *fmt = find_quota_format(format_id);
- int error;
+ struct dquot *to_drop[MAXQUOTAS];
+ int error, cnt;
unsigned int oldflags;
if (!fmt)
return -ESRCH;
- f = filp_open(path, O_RDWR, 0600);
- if (IS_ERR(f)) {
- error = PTR_ERR(f);
- goto out_fmt;
- }
error = -EIO;
if (!f->f_op || !f->f_op->read || !f->f_op->write)
- goto out_f;
- error = security_quota_on(f);
- if (error)
- goto out_f;
+ goto out_fmt;
inode = f->f_dentry->d_inode;
error = -EACCES;
if (!S_ISREG(inode->i_mode))
- goto out_f;
+ goto out_fmt;
down(&dqopt->dqonoff_sem);
- down_write(&dqopt->dqptr_sem);
if (sb_has_quota_enabled(sb, type)) {
error = -EBUSY;
goto out_lock;
@@ -1213,8 +1351,20 @@ int vfs_quota_on(struct super_block *sb,
if (!fmt->qf_ops->check_quota_file(sb, type))
goto out_file_init;
/* We don't want quota and atime on quota files (deadlocks possible) */
- dquot_drop_nolock(inode);
+ down_write(&dqopt->dqptr_sem);
inode->i_flags |= S_NOQUOTA | S_NOATIME;
+ for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
+ to_drop[cnt] = inode->i_dquot[cnt];
+ inode->i_dquot[cnt] = NODQUOT;
+ }
+ inode->i_flags &= ~S_QUOTA;
+ up_write(&dqopt->dqptr_sem);
+ /* We must put dquots outside of dqptr_sem because we may need to
+ * start transaction for dquot_release() */
+ for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
+ if (to_drop[cnt])
+ dqput(to_drop[cnt]);
+ }
dqopt->ops[type] = fmt->qf_ops;
dqopt->info[type].dqi_format = fmt;
@@ -1225,7 +1375,6 @@ int vfs_quota_on(struct super_block *sb,
}
up(&dqopt->dqio_sem);
set_enable_flags(dqopt, type);
- up_write(&dqopt->dqptr_sem);
add_dquot_ref(sb, type);
up(&dqopt->dqonoff_sem);
@@ -1238,14 +1387,58 @@ out_file_init:
out_lock:
up_write(&dqopt->dqptr_sem);
up(&dqopt->dqonoff_sem);
-out_f:
- filp_close(f, NULL);
out_fmt:
put_quota_format(fmt);
return error;
}
+/* Actual function called from quotactl() */
+int vfs_quota_on(struct super_block *sb, int type, int format_id, char *path)
+{
+ struct file *f;
+ int error;
+
+ f = filp_open(path, O_RDWR, 0600);
+ if (IS_ERR(f))
+ return PTR_ERR(f);
+ error = security_quota_on(f);
+ if (error)
+ goto out_f;
+ error = vfs_quota_on_file(f, type, format_id);
+ if (!error)
+ return 0;
+out_f:
+ filp_close(f, NULL);
+ return error;
+}
+
+/*
+ * Function used by filesystems when filp_open() would fail (filesystem is
+ * being mounted now). We will use a private file structure. Caller is
+ * responsible that it's IO functions won't need vfsmnt structure or
+ * some dentry tricks...
+ */
+int vfs_quota_on_mount(int type, int format_id, struct dentry *dentry)
+{
+ struct file *f;
+ int error;
+
+ dget(dentry); /* Get a reference for struct file */
+ f = dentry_open(dentry, NULL, O_RDWR);
+ if (IS_ERR(f)) {
+ error = PTR_ERR(f);
+ goto out_dentry;
+ }
+ error = vfs_quota_on_file(f, type, format_id);
+ if (!error)
+ return 0;
+ fput(f);
+out_dentry:
+ dput(dentry);
+ return error;
+}
+
/* Generic routine for getting common part of quota structure */
static void do_get_dqblk(struct dquot *dquot, struct if_dqblk *di)
{
@@ -1268,14 +1461,14 @@ int vfs_get_dqblk(struct super_block *sb
{
struct dquot *dquot;
- down_read(&sb_dqopt(sb)->dqptr_sem);
+ down(&sb_dqopt(sb)->dqonoff_sem);
if (!(dquot = dqget(sb, id, type))) {
- up_read(&sb_dqopt(sb)->dqptr_sem);
+ up(&sb_dqopt(sb)->dqonoff_sem);
return -ESRCH;
}
do_get_dqblk(dquot, di);
dqput(dquot);
- up_read(&sb_dqopt(sb)->dqptr_sem);
+ up(&sb_dqopt(sb)->dqonoff_sem);
return 0;
}
@@ -1329,22 +1522,22 @@ static void do_set_dqblk(struct dquot *d
clear_bit(DQ_FAKE_B, &dquot->dq_flags);
else
set_bit(DQ_FAKE_B, &dquot->dq_flags);
- mark_dquot_dirty(dquot);
spin_unlock(&dq_data_lock);
+ mark_dquot_dirty(dquot);
}
int vfs_set_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *di)
{
struct dquot *dquot;
- down_read(&sb_dqopt(sb)->dqptr_sem);
+ down(&sb_dqopt(sb)->dqonoff_sem);
if (!(dquot = dqget(sb, id, type))) {
- up_read(&sb_dqopt(sb)->dqptr_sem);
+ up(&sb_dqopt(sb)->dqonoff_sem);
return -ESRCH;
}
do_set_dqblk(dquot, di);
dqput(dquot);
- up_read(&sb_dqopt(sb)->dqptr_sem);
+ up(&sb_dqopt(sb)->dqonoff_sem);
return 0;
}
@@ -1353,9 +1546,9 @@ int vfs_get_dqinfo(struct super_block *s
{
struct mem_dqinfo *mi;
- down_read(&sb_dqopt(sb)->dqptr_sem);
+ down(&sb_dqopt(sb)->dqonoff_sem);
if (!sb_has_quota_enabled(sb, type)) {
- up_read(&sb_dqopt(sb)->dqptr_sem);
+ up(&sb_dqopt(sb)->dqonoff_sem);
return -ESRCH;
}
mi = sb_dqopt(sb)->info + type;
@@ -1365,7 +1558,7 @@ int vfs_get_dqinfo(struct super_block *s
ii->dqi_flags = mi->dqi_flags & DQF_MASK;
ii->dqi_valid = IIF_ALL;
spin_unlock(&dq_data_lock);
- up_read(&sb_dqopt(sb)->dqptr_sem);
+ up(&sb_dqopt(sb)->dqonoff_sem);
return 0;
}
@@ -1374,9 +1567,9 @@ int vfs_set_dqinfo(struct super_block *s
{
struct mem_dqinfo *mi;
- down_read(&sb_dqopt(sb)->dqptr_sem);
+ down(&sb_dqopt(sb)->dqonoff_sem);
if (!sb_has_quota_enabled(sb, type)) {
- up_read(&sb_dqopt(sb)->dqptr_sem);
+ up(&sb_dqopt(sb)->dqonoff_sem);
return -ESRCH;
}
mi = sb_dqopt(sb)->info + type;
@@ -1387,9 +1580,11 @@ int vfs_set_dqinfo(struct super_block *s
mi->dqi_igrace = ii->dqi_igrace;
if (ii->dqi_valid & IIF_FLAGS)
mi->dqi_flags = (mi->dqi_flags & ~DQF_MASK) | (ii->dqi_flags & DQF_MASK);
- mark_info_dirty(mi);
spin_unlock(&dq_data_lock);
- up_read(&sb_dqopt(sb)->dqptr_sem);
+ mark_info_dirty(sb, type);
+ /* Force write to disk */
+ sb->dq_op->write_info(sb, type);
+ up(&sb_dqopt(sb)->dqonoff_sem);
return 0;
}
@@ -1520,4 +1715,21 @@ EXPORT_SYMBOL(unregister_quota_format);
EXPORT_SYMBOL(dqstats);
EXPORT_SYMBOL(dq_list_lock);
EXPORT_SYMBOL(dq_data_lock);
-EXPORT_SYMBOL(init_dquot_operations);
+EXPORT_SYMBOL(vfs_quota_on);
+EXPORT_SYMBOL(vfs_quota_on_mount);
+EXPORT_SYMBOL(vfs_quota_off);
+EXPORT_SYMBOL(vfs_quota_sync);
+EXPORT_SYMBOL(vfs_get_dqinfo);
+EXPORT_SYMBOL(vfs_set_dqinfo);
+EXPORT_SYMBOL(vfs_get_dqblk);
+EXPORT_SYMBOL(vfs_set_dqblk);
+EXPORT_SYMBOL(dquot_commit);
+EXPORT_SYMBOL(dquot_commit_info);
+EXPORT_SYMBOL(dquot_mark_dquot_dirty);
+EXPORT_SYMBOL(dquot_initialize);
+EXPORT_SYMBOL(dquot_drop);
+EXPORT_SYMBOL(dquot_alloc_space);
+EXPORT_SYMBOL(dquot_alloc_inode);
+EXPORT_SYMBOL(dquot_free_space);
+EXPORT_SYMBOL(dquot_free_inode);
+EXPORT_SYMBOL(dquot_transfer);
--- diff/fs/exec.c 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/exec.c 2004-03-16 11:27:48.393830680 +0000
@@ -327,7 +327,7 @@ void put_dirty_page(struct task_struct *
set_pte(pte, pte_mkdirty(pte_mkwrite(mk_pte(page, prot))));
pte_chain = page_add_rmap(page, pte, pte_chain);
pte_unmap(pte);
- tsk->mm->rss++;
+ inc_rss(tsk->mm, page);
spin_unlock(&tsk->mm->page_table_lock);
/* no need for flush_tlb */
@@ -601,6 +601,11 @@ static inline int de_thread(struct task_
newsig->group_stop_count = 0;
newsig->curr_target = NULL;
init_sigpending(&newsig->shared_pending);
+
+ newsig->pgrp = oldsig->pgrp;
+ newsig->session = oldsig->session;
+ newsig->leader = oldsig->leader;
+ newsig->tty_old_pgrp = oldsig->tty_old_pgrp;
}
if (thread_group_empty(current))
@@ -833,7 +838,8 @@ int flush_old_exec(struct linux_binprm *
flush_thread();
if (bprm->e_uid != current->euid || bprm->e_gid != current->egid ||
- permission(bprm->file->f_dentry->d_inode,MAY_READ, NULL))
+ permission(bprm->file->f_dentry->d_inode,MAY_READ, NULL) ||
+ (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP))
current->mm->dumpable = 0;
/* An exec changes our domain. We are no longer part of the thread
@@ -1106,6 +1112,7 @@ int do_execve(char * filename,
bprm.file = file;
bprm.filename = filename;
bprm.interp = filename;
+ bprm.interp_flags = 0;
bprm.sh_bang = 0;
bprm.loader = 0;
bprm.exec = 0;
--- diff/fs/ext3/inode.c 2004-02-09 10:36:12.000000000 +0000
+++ source/fs/ext3/inode.c 2004-03-16 11:27:48.395830376 +0000
@@ -2772,9 +2772,28 @@ int ext3_setattr(struct dentry *dentry,
if ((ia_valid & ATTR_UID && attr->ia_uid != inode->i_uid) ||
(ia_valid & ATTR_GID && attr->ia_gid != inode->i_gid)) {
+ handle_t *handle;
+
+ /* (user+group)*(old+new) structure, inode write (sb,
+ * inode block, ? - but truncate inode update has it) */
+ handle = ext3_journal_start(inode, 4*EXT3_QUOTA_INIT_BLOCKS+3);
+ if (IS_ERR(handle)) {
+ error = PTR_ERR(handle);
+ goto err_out;
+ }
error = DQUOT_TRANSFER(inode, attr) ? -EDQUOT : 0;
- if (error)
+ if (error) {
+ ext3_journal_stop(handle);
return error;
+ }
+ /* Update corresponding info in inode so that everything is in
+ * one transaction */
+ if (attr->ia_valid & ATTR_UID)
+ inode->i_uid = attr->ia_uid;
+ if (attr->ia_valid & ATTR_GID)
+ inode->i_gid = attr->ia_gid;
+ error = ext3_mark_inode_dirty(handle, inode);
+ ext3_journal_stop(handle);
}
if (S_ISREG(inode->i_mode) &&
@@ -2853,7 +2872,9 @@ int ext3_writepage_trans_blocks(struct i
ret = 2 * (bpp + indirects) + 2;
#ifdef CONFIG_QUOTA
- ret += 2 * EXT3_SINGLEDATA_TRANS_BLOCKS;
+ /* We know that structure was already allocated during DQUOT_INIT so
+ * we will be updating only the data blocks + inodes */
+ ret += 2*EXT3_QUOTA_TRANS_BLOCKS;
#endif
return ret;
--- diff/fs/ext3/namei.c 2004-02-18 08:54:12.000000000 +0000
+++ source/fs/ext3/namei.c 2004-03-16 11:27:48.396830224 +0000
@@ -1633,7 +1633,8 @@ static int ext3_create (struct inode * d
int err;
handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS +
- EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3);
+ EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 +
+ 2*EXT3_QUOTA_INIT_BLOCKS);
if (IS_ERR(handle))
return PTR_ERR(handle);
@@ -1663,7 +1664,8 @@ static int ext3_mknod (struct inode * di
return -EINVAL;
handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS +
- EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3);
+ EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 +
+ 2*EXT3_QUOTA_INIT_BLOCKS);
if (IS_ERR(handle))
return PTR_ERR(handle);
@@ -1695,7 +1697,8 @@ static int ext3_mkdir(struct inode * dir
return -EMLINK;
handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS +
- EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3);
+ EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 +
+ 2*EXT3_QUOTA_INIT_BLOCKS);
if (IS_ERR(handle))
return PTR_ERR(handle);
@@ -1974,6 +1977,9 @@ static int ext3_rmdir (struct inode * di
struct ext3_dir_entry_2 * de;
handle_t *handle;
+ /* Initialize quotas before so that eventual writes go in
+ * separate transaction */
+ DQUOT_INIT(dentry->d_inode);
handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS);
if (IS_ERR(handle))
return PTR_ERR(handle);
@@ -1987,7 +1993,6 @@ static int ext3_rmdir (struct inode * di
handle->h_sync = 1;
inode = dentry->d_inode;
- DQUOT_INIT(inode);
retval = -EIO;
if (le32_to_cpu(de->inode) != inode->i_ino)
@@ -2031,6 +2036,9 @@ static int ext3_unlink(struct inode * di
struct ext3_dir_entry_2 * de;
handle_t *handle;
+ /* Initialize quotas before so that eventual writes go
+ * in separate transaction */
+ DQUOT_INIT(dentry->d_inode);
handle = ext3_journal_start(dir, EXT3_DELETE_TRANS_BLOCKS);
if (IS_ERR(handle))
return PTR_ERR(handle);
@@ -2044,7 +2052,6 @@ static int ext3_unlink(struct inode * di
goto end_unlink;
inode = dentry->d_inode;
- DQUOT_INIT(inode);
retval = -EIO;
if (le32_to_cpu(de->inode) != inode->i_ino)
@@ -2087,7 +2094,8 @@ static int ext3_symlink (struct inode *
return -ENAMETOOLONG;
handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS +
- EXT3_INDEX_EXTRA_TRANS_BLOCKS + 5);
+ EXT3_INDEX_EXTRA_TRANS_BLOCKS + 5 +
+ 2*EXT3_QUOTA_INIT_BLOCKS);
if (IS_ERR(handle))
return PTR_ERR(handle);
@@ -2172,6 +2180,10 @@ static int ext3_rename (struct inode * o
old_bh = new_bh = dir_bh = NULL;
+ /* Initialize quotas before so that eventual writes go
+ * in separate transaction */
+ if (new_dentry->d_inode)
+ DQUOT_INIT(new_dentry->d_inode);
handle = ext3_journal_start(old_dir, 2 * EXT3_DATA_TRANS_BLOCKS +
EXT3_INDEX_EXTRA_TRANS_BLOCKS + 2);
if (IS_ERR(handle))
@@ -2198,8 +2210,6 @@ static int ext3_rename (struct inode * o
if (!new_inode) {
brelse (new_bh);
new_bh = NULL;
- } else {
- DQUOT_INIT(new_inode);
}
}
if (S_ISDIR(old_inode->i_mode)) {
--- diff/fs/ext3/super.c 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/ext3/super.c 2004-03-16 11:27:48.398829920 +0000
@@ -32,6 +32,9 @@
#include <linux/buffer_head.h>
#include <linux/vfs.h>
#include <linux/random.h>
+#include <linux/mount.h>
+#include <linux/namei.h>
+#include <linux/quotaops.h>
#include <asm/uaccess.h>
#include "xattr.h"
#include "acl.h"
@@ -504,7 +507,43 @@ static void ext3_clear_inode(struct inod
# define ext3_clear_inode NULL
#endif
-static struct dquot_operations ext3_qops;
+#ifdef CONFIG_QUOTA
+
+#define QTYPE2NAME(t) ((t)==USRQUOTA?"user":"group")
+#define QTYPE2MOPT(on, t) ((t)==USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA))
+
+static int ext3_dquot_initialize(struct inode *inode, int type);
+static int ext3_dquot_drop(struct inode *inode);
+static int ext3_write_dquot(struct dquot *dquot);
+static int ext3_mark_dquot_dirty(struct dquot *dquot);
+static int ext3_write_info(struct super_block *sb, int type);
+static int ext3_quota_on(struct super_block *sb, int type, int format_id, char *path);
+static int ext3_quota_on_mount(struct super_block *sb, int type);
+static int ext3_quota_off_mount(struct super_block *sb, int type);
+
+static struct dquot_operations ext3_quota_operations = {
+ .initialize = ext3_dquot_initialize,
+ .drop = ext3_dquot_drop,
+ .alloc_space = dquot_alloc_space,
+ .alloc_inode = dquot_alloc_inode,
+ .free_space = dquot_free_space,
+ .free_inode = dquot_free_inode,
+ .transfer = dquot_transfer,
+ .write_dquot = ext3_write_dquot,
+ .mark_dirty = ext3_mark_dquot_dirty,
+ .write_info = ext3_write_info
+};
+
+static struct quotactl_ops ext3_qctl_operations = {
+ .quota_on = ext3_quota_on,
+ .quota_off = vfs_quota_off,
+ .quota_sync = vfs_quota_sync,
+ .get_info = vfs_get_dqinfo,
+ .set_info = vfs_set_dqinfo,
+ .get_dqblk = vfs_get_dqblk,
+ .set_dqblk = vfs_set_dqblk
+};
+#endif
static struct super_operations ext3_sops = {
.alloc_inode = ext3_alloc_inode,
@@ -536,6 +575,8 @@ enum {
Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl, Opt_noload,
Opt_commit, Opt_journal_update, Opt_journal_inum,
Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
+ Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
+ Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0,
Opt_ignore, Opt_err,
};
@@ -571,6 +612,12 @@ static match_table_t tokens = {
{Opt_data_journal, "data=journal"},
{Opt_data_ordered, "data=ordered"},
{Opt_data_writeback, "data=writeback"},
+ {Opt_offusrjquota, "usrjquota="},
+ {Opt_usrjquota, "usrjquota=%s"},
+ {Opt_offgrpjquota, "grpjquota="},
+ {Opt_grpjquota, "grpjquota=%s"},
+ {Opt_jqfmt_vfsold, "jqfmt=vfsold"},
+ {Opt_jqfmt_vfsv0, "jqfmt=vfsv0"},
{Opt_ignore, "grpquota"},
{Opt_ignore, "noquota"},
{Opt_ignore, "quota"},
@@ -598,13 +645,17 @@ static unsigned long get_sb_block(void *
return sb_block;
}
-static int parse_options (char * options, struct ext3_sb_info *sbi,
+static int parse_options (char * options, struct super_block *sb,
unsigned long * inum, int is_remount)
{
+ struct ext3_sb_info *sbi = EXT3_SB(sb);
char * p;
substring_t args[MAX_OPT_ARGS];
int data_opt = 0;
int option;
+#ifdef CONFIG_QUOTA
+ int qtype;
+#endif
if (!options)
return 1;
@@ -759,6 +810,76 @@ static int parse_options (char * options
sbi->s_mount_opt |= data_opt;
}
break;
+#ifdef CONFIG_QUOTA
+ case Opt_usrjquota:
+ qtype = USRQUOTA;
+ goto set_qf_name;
+ case Opt_grpjquota:
+ qtype = GRPQUOTA;
+set_qf_name:
+ if (sb_any_quota_enabled(sb)) {
+ printk(KERN_ERR
+ "EXT3-fs: Cannot change journalled "
+ "quota options when quota turned on.\n");
+ return 0;
+ }
+ if (sbi->s_qf_names[qtype]) {
+ printk(KERN_ERR
+ "EXT3-fs: %s quota file already "
+ "specified.\n", QTYPE2NAME(qtype));
+ return 0;
+ }
+ sbi->s_qf_names[qtype] = match_strdup(&args[0]);
+ if (!sbi->s_qf_names[qtype]) {
+ printk(KERN_ERR
+ "EXT3-fs: not enough memory for "
+ "storing quotafile name.\n");
+ return 0;
+ }
+ if (strchr(sbi->s_qf_names[qtype], '/')) {
+ printk(KERN_ERR
+ "EXT3-fs: quotafile must be on "
+ "filesystem root.\n");
+ kfree(sbi->s_qf_names[qtype]);
+ sbi->s_qf_names[qtype] = NULL;
+ return 0;
+ }
+ break;
+ case Opt_offusrjquota:
+ qtype = USRQUOTA;
+ goto clear_qf_name;
+ case Opt_offgrpjquota:
+ qtype = GRPQUOTA;
+clear_qf_name:
+ if (sb_any_quota_enabled(sb)) {
+ printk(KERN_ERR "EXT3-fs: Cannot change "
+ "journalled quota options when "
+ "quota turned on.\n");
+ return 0;
+ }
+ if (sbi->s_qf_names[qtype]) {
+ kfree(sbi->s_qf_names[qtype]);
+ sbi->s_qf_names[qtype] = NULL;
+ }
+ break;
+ case Opt_jqfmt_vfsold:
+ sbi->s_jquota_fmt = QFMT_VFS_OLD;
+ break;
+ case Opt_jqfmt_vfsv0:
+ sbi->s_jquota_fmt = QFMT_VFS_V0;
+ break;
+#else
+ case Opt_usrjquota:
+ case Opt_grpjquota:
+ case Opt_offusrjquota:
+ case Opt_offgrpjquota:
+ case Opt_jqfmt_vfsold:
+ case Opt_jqfmt_vfsv0:
+ printk(KERN_ERR
+ "EXT3-fs: journalled quota options not "
+ "supported.\n");
+ break;
+#endif
case Opt_abort:
set_opt(sbi->s_mount_opt, ABORT);
break;
@@ -771,6 +892,13 @@ static int parse_options (char * options
return 0;
}
}
+#ifdef CONFIG_QUOTA
+ if (!sbi->s_jquota_fmt && (sbi->s_qf_names[0] || sbi->s_qf_names[1])) {
+ printk(KERN_ERR
+ "EXT3-fs: journalled quota format not specified.\n");
+ return 0;
+ }
+#endif
return 1;
}
@@ -930,6 +1058,9 @@ static void ext3_orphan_cleanup (struct
{
unsigned int s_flags = sb->s_flags;
int nr_orphans = 0, nr_truncates = 0;
+#ifdef CONFIG_QUOTA
+ int i;
+#endif
if (!es->s_last_orphan) {
jbd_debug(4, "no orphan inodes to clean up\n");
return;
@@ -949,6 +1080,20 @@ static void ext3_orphan_cleanup (struct
sb->s_id);
sb->s_flags &= ~MS_RDONLY;
}
+#ifdef CONFIG_QUOTA
+ /* Needed for iput() to work correctly and not trash data */
+ sb->s_flags |= MS_ACTIVE;
+ /* Turn on quotas so that they are updated correctly */
+ for (i = 0; i < MAXQUOTAS; i++) {
+ if (EXT3_SB(sb)->s_qf_names[i]) {
+ int ret = ext3_quota_on_mount(sb, i);
+ if (ret < 0)
+ printk(KERN_ERR
+ "EXT3-fs: Cannot turn on journalled "
+ "quota: error %d\n", ret);
+ }
+ }
+#endif
while (es->s_last_orphan) {
struct inode *inode;
@@ -960,6 +1105,7 @@ static void ext3_orphan_cleanup (struct
}
list_add(&EXT3_I(inode)->i_orphan, &EXT3_SB(sb)->s_orphan);
+ DQUOT_INIT(inode);
if (inode->i_nlink) {
printk(KERN_DEBUG
"%s: truncating inode %ld to %Ld bytes\n",
@@ -987,6 +1133,13 @@ static void ext3_orphan_cleanup (struct
if (nr_truncates)
printk(KERN_INFO "EXT3-fs: %s: %d truncate%s cleaned up\n",
sb->s_id, PLURAL(nr_truncates));
+#ifdef CONFIG_QUOTA
+ /* Turn quotas off */
+ for (i = 0; i < MAXQUOTAS; i++) {
+ if (sb_dqopt(sb)->files[i])
+ ext3_quota_off_mount(sb, i);
+ }
+#endif
sb->s_flags = s_flags; /* Restore MS_RDONLY status */
}
@@ -1117,7 +1270,7 @@ static int ext3_fill_super (struct super
sbi->s_resuid = le16_to_cpu(es->s_def_resuid);
sbi->s_resgid = le16_to_cpu(es->s_def_resgid);
- if (!parse_options ((char *) data, sbi, &journal_inum, 0))
+ if (!parse_options ((char *) data, sb, &journal_inum, 0))
goto failed_mount;
sb->s_flags |= MS_ONE_SECOND;
@@ -1296,7 +1449,10 @@ static int ext3_fill_super (struct super
*/
sb->s_op = &ext3_sops;
sb->s_export_op = &ext3_export_ops;
- sb->dq_op = &ext3_qops;
+#ifdef CONFIG_QUOTA
+ sb->s_qcop = &ext3_qctl_operations;
+ sb->dq_op = &ext3_quota_operations;
+#endif
INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */
sb->s_root = 0;
@@ -1406,6 +1562,12 @@ failed_mount2:
brelse(sbi->s_group_desc[i]);
kfree(sbi->s_group_desc);
failed_mount:
+#ifdef CONFIG_QUOTA
+ for (i = 0; i < MAXQUOTAS; i++) {
+ if (sbi->s_qf_names[i])
+ kfree(sbi->s_qf_names[i]);
+ }
+#endif
ext3_blkdev_remove(sbi);
brelse(bh);
out_fail:
@@ -1832,7 +1994,7 @@ int ext3_remount (struct super_block * s
/*
* Allow the "check" option to be passed as a remount option.
*/
- if (!parse_options(data, sbi, &tmp, 1))
+ if (!parse_options(data, sb, &tmp, 1))
return -EINVAL;
if (sbi->s_mount_opt & EXT3_MOUNT_ABORT)
@@ -1952,45 +2114,152 @@ int ext3_statfs (struct super_block * sb
#ifdef CONFIG_QUOTA
-/* Blocks: (2 data blocks) * (3 indirect + 1 descriptor + 1 bitmap) + superblock */
-#define EXT3_OLD_QFMT_BLOCKS 11
-/* Blocks: quota info + (4 pointer blocks + 1 entry block) * (3 indirect + 1 descriptor + 1 bitmap) + superblock */
-#define EXT3_V0_QFMT_BLOCKS 27
+static inline struct inode *dquot_to_inode(struct dquot *dquot)
+{
+ return sb_dqopt(dquot->dq_sb)->files[dquot->dq_type]->f_dentry->d_inode;
+}
-static int (*old_write_dquot)(struct dquot *dquot);
+static int ext3_dquot_initialize(struct inode *inode, int type)
+{
+ handle_t *handle;
+ int ret, err;
+
+ /* We may create quota structure so we need to reserve enough blocks */
+ handle = ext3_journal_start(inode, 2*EXT3_QUOTA_INIT_BLOCKS);
+ if (IS_ERR(handle))
+ return PTR_ERR(handle);
+ ret = dquot_initialize(inode, type);
+ err = ext3_journal_stop(handle);
+ if (!ret)
+ ret = err;
+ return ret;
+}
+
+static int ext3_dquot_drop(struct inode *inode)
+{
+ handle_t *handle;
+ int ret, err;
+
+ /* We may delete quota structure so we need to reserve enough blocks */
+ handle = ext3_journal_start(inode, 2*EXT3_QUOTA_INIT_BLOCKS);
+ if (IS_ERR(handle))
+ return PTR_ERR(handle);
+ ret = dquot_drop(inode);
+ err = ext3_journal_stop(handle);
+ if (!ret)
+ ret = err;
+ return ret;
+}
static int ext3_write_dquot(struct dquot *dquot)
{
- int nblocks;
- int ret;
- int err;
+ int ret, err;
handle_t *handle;
- struct quota_info *dqops = sb_dqopt(dquot->dq_sb);
- struct inode *qinode;
- switch (dqops->info[dquot->dq_type].dqi_format->qf_fmt_id) {
- case QFMT_VFS_OLD:
- nblocks = EXT3_OLD_QFMT_BLOCKS;
- break;
- case QFMT_VFS_V0:
- nblocks = EXT3_V0_QFMT_BLOCKS;
- break;
- default:
- nblocks = EXT3_MAX_TRANS_DATA;
- }
- qinode = dqops->files[dquot->dq_type]->f_dentry->d_inode;
- handle = ext3_journal_start(qinode, nblocks);
- if (IS_ERR(handle)) {
- ret = PTR_ERR(handle);
- goto out;
- }
- ret = old_write_dquot(dquot);
+ handle = ext3_journal_start(dquot_to_inode(dquot),
+ EXT3_QUOTA_TRANS_BLOCKS);
+ if (IS_ERR(handle))
+ return PTR_ERR(handle);
+ ret = dquot_commit(dquot);
err = ext3_journal_stop(handle);
- if (ret == 0)
+ if (!ret)
+ ret = err;
+ return ret;
+}
+
+static int ext3_mark_dquot_dirty(struct dquot * dquot)
+{
+ /* Are we journalling quotas? */
+ if (EXT3_SB(dquot->dq_sb)->s_qf_names[0] ||
+ EXT3_SB(dquot->dq_sb)->s_qf_names[1])
+ return ext3_write_dquot(dquot);
+ else
+ return dquot_mark_dquot_dirty(dquot);
+}
+
+static int ext3_write_info(struct super_block *sb, int type)
+{
+ int ret, err;
+ handle_t *handle;
+
+ /* Data block + inode block */
+ handle = ext3_journal_start(sb->s_root->d_inode, 2);
+ if (IS_ERR(handle))
+ return PTR_ERR(handle);
+ ret = dquot_commit_info(sb, type);
+ err = ext3_journal_stop(handle);
+ if (!ret)
ret = err;
-out:
return ret;
}
+
+/*
+ * Turn on quotas during mount time - we need to find
+ * the quota file and such...
+ */
+static int ext3_quota_on_mount(struct super_block *sb, int type)
+{
+ int err;
+ struct dentry *dentry;
+ struct qstr name = { .name = EXT3_SB(sb)->s_qf_names[type],
+ .hash = 0,
+ .len = strlen(EXT3_SB(sb)->s_qf_names[type])};
+
+ dentry = lookup_hash(&name, sb->s_root);
+ if (IS_ERR(dentry))
+ return PTR_ERR(dentry);
+ err = vfs_quota_on_mount(type, EXT3_SB(sb)->s_jquota_fmt, dentry);
+ if (err)
+ dput(dentry);
+ /* We keep the dentry reference if everything went ok - we drop it
+ * on quota_off time */
+ return err;
+}
+
+/* Turn quotas off during mount time */
+static int ext3_quota_off_mount(struct super_block *sb, int type)
+{
+ int err;
+ struct dentry *dentry;
+
+ dentry = sb_dqopt(sb)->files[type]->f_dentry;
+ err = vfs_quota_off_mount(sb, type);
+ /* We invalidate dentry - it has at least wrong hash... */
+ d_invalidate(dentry);
+ dput(dentry);
+ return err;
+}
+
+/*
+ * Standard function to be called on quota_on
+ */
+static int ext3_quota_on(struct super_block *sb, int type, int format_id,
+ char *path)
+{
+ int err;
+ struct nameidata nd;
+
+ /* Not journalling quota? */
+ if (!EXT3_SB(sb)->s_qf_names[0] && !EXT3_SB(sb)->s_qf_names[1])
+ return vfs_quota_on(sb, type, format_id, path);
+ err = path_lookup(path, LOOKUP_FOLLOW, &nd);
+ if (err)
+ return err;
+ /* Quotafile not on the same filesystem? */
+ if (nd.mnt->mnt_sb != sb)
+ return -EXDEV;
+ /* Quotafile not of fs root? */
+ if (nd.dentry->d_parent->d_inode != sb->s_root->d_inode)
+ printk(KERN_WARNING
+ "EXT3-fs: Quota file not on filesystem root. "
+ "Journalled quota will not work.\n");
+ if (!ext3_should_journal_data(nd.dentry->d_inode))
+ printk(KERN_WARNING "EXT3-fs: Quota file does not have "
+ "data-journalling. Journalled quota will not work.\n");
+ path_release(&nd);
+ return vfs_quota_on(sb, type, format_id, path);
+}
+
#endif
static struct super_block *ext3_get_sb(struct file_system_type *fs_type,
@@ -2015,11 +2284,6 @@ static int __init init_ext3_fs(void)
err = init_inodecache();
if (err)
goto out1;
-#ifdef CONFIG_QUOTA
- init_dquot_operations(&ext3_qops);
- old_write_dquot = ext3_qops.write_dquot;
- ext3_qops.write_dquot = ext3_write_dquot;
-#endif
err = register_filesystem(&ext3_fs_type);
if (err)
goto out;
--- diff/fs/fs-writeback.c 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/fs-writeback.c 2004-03-16 11:27:48.399829768 +0000
@@ -123,12 +123,6 @@ static void write_inode(struct inode *in
* starvation of particular inodes when others are being redirtied, prevent
* livelocks, etc.
*
- * So what we do is to move all pages which are to be written from dirty_pages
- * onto io_pages. And keep on writing io_pages until it's empty. Refusing to
- * move more pages onto io_pages until io_pages is empty. Once that point has
- * been reached, we are ready to take another pass across the inode's dirty
- * pages.
- *
* Called under inode_lock.
*/
static void
@@ -152,10 +146,6 @@ __sync_single_inode(struct inode *inode,
* read speculatively by this cpu before &= ~I_DIRTY -- mikulas
*/
- spin_lock(&mapping->page_lock);
- if (wait || !wbc->for_kupdate || list_empty(&mapping->io_pages))
- list_splice_init(&mapping->dirty_pages, &mapping->io_pages);
- spin_unlock(&mapping->page_lock);
spin_unlock(&inode_lock);
do_writepages(mapping, wbc);
@@ -170,10 +160,7 @@ __sync_single_inode(struct inode *inode,
spin_lock(&inode_lock);
inode->i_state &= ~I_LOCK;
if (!(inode->i_state & I_FREEING)) {
- if (!list_empty(&mapping->io_pages)) {
- /* Needs more writeback */
- inode->i_state |= I_DIRTY_PAGES;
- } else if (!list_empty(&mapping->dirty_pages)) {
+ if (mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) {
/* Redirtied */
inode->i_state |= I_DIRTY_PAGES;
mapping->dirtied_when = jiffies;
--- diff/fs/hpfs/buffer.c 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/hpfs/buffer.c 2004-03-16 11:27:48.400829616 +0000
@@ -60,74 +60,6 @@ void hpfs_unlock_inode(struct inode *i)
}
}
-void hpfs_lock_2inodes(struct inode *i1, struct inode *i2)
-{
- if (!i2 || i1 == i2) {
- hpfs_lock_inode(i1);
- } else if (!i1) {
- hpfs_lock_inode(i2);
- } else {
- struct hpfs_inode_info *hpfs_i1 = hpfs_i(i1);
- struct hpfs_inode_info *hpfs_i2 = hpfs_i(i2);
- if (i1->i_ino < i2->i_ino) {
- down(&hpfs_i1->i_sem);
- down(&hpfs_i2->i_sem);
- } else {
- down(&hpfs_i2->i_sem);
- down(&hpfs_i1->i_sem);
- }
- }
-}
-
-void hpfs_unlock_2inodes(struct inode *i1, struct inode *i2)
-{
- /* order of up() doesn't matter here */
- hpfs_unlock_inode(i1);
- hpfs_unlock_inode(i2);
-}
-
-void hpfs_lock_3inodes(struct inode *i1, struct inode *i2, struct inode *i3)
-{
- if (!i1) { hpfs_lock_2inodes(i2, i3); return; }
- if (!i2) { hpfs_lock_2inodes(i1, i3); return; }
- if (!i3) { hpfs_lock_2inodes(i1, i2); return; }
- if (i1->i_ino < i2->i_ino && i1->i_ino < i3->i_ino) {
- struct hpfs_inode_info *hpfs_i1 = hpfs_i(i1);
- down(&hpfs_i1->i_sem);
- hpfs_lock_2inodes(i2, i3);
- } else if (i2->i_ino < i1->i_ino && i2->i_ino < i3->i_ino) {
- struct hpfs_inode_info *hpfs_i2 = hpfs_i(i2);
- down(&hpfs_i2->i_sem);
- hpfs_lock_2inodes(i1, i3);
- } else if (i3->i_ino < i1->i_ino && i3->i_ino < i2->i_ino) {
- struct hpfs_inode_info *hpfs_i3 = hpfs_i(i3);
- down(&hpfs_i3->i_sem);
- hpfs_lock_2inodes(i1, i2);
- } else if (i1->i_ino != i2->i_ino) hpfs_lock_2inodes(i1, i2);
- else hpfs_lock_2inodes(i1, i3);
-}
-
-void hpfs_unlock_3inodes(struct inode *i1, struct inode *i2, struct inode *i3)
-{
- if (!i1) { hpfs_unlock_2inodes(i2, i3); return; }
- if (!i2) { hpfs_unlock_2inodes(i1, i3); return; }
- if (!i3) { hpfs_unlock_2inodes(i1, i2); return; }
- if (i1->i_ino < i2->i_ino && i1->i_ino < i3->i_ino) {
- struct hpfs_inode_info *hpfs_i1 = hpfs_i(i1);
- hpfs_unlock_2inodes(i2, i3);
- up(&hpfs_i1->i_sem);
- } else if (i2->i_ino < i1->i_ino && i2->i_ino < i3->i_ino) {
- struct hpfs_inode_info *hpfs_i2 = hpfs_i(i2);
- hpfs_unlock_2inodes(i1, i3);
- up(&hpfs_i2->i_sem);
- } else if (i3->i_ino < i1->i_ino && i3->i_ino < i2->i_ino) {
- struct hpfs_inode_info *hpfs_i3 = hpfs_i(i3);
- hpfs_unlock_2inodes(i1, i2);
- up(&hpfs_i3->i_sem);
- } else if (i1->i_ino != i2->i_ino) hpfs_unlock_2inodes(i1, i2);
- else hpfs_unlock_2inodes(i1, i3);
-}
-
/* Map a sector into a buffer and return pointers to it and to the buffer. */
void *hpfs_map_sector(struct super_block *s, unsigned secno, struct buffer_head **bhp,
--- diff/fs/hpfs/hpfs_fn.h 2003-10-09 08:47:34.000000000 +0000
+++ source/fs/hpfs/hpfs_fn.h 2004-03-16 11:27:48.400829616 +0000
@@ -196,10 +196,6 @@ void hpfs_lock_iget(struct super_block *
void hpfs_unlock_iget(struct super_block *);
void hpfs_lock_inode(struct inode *);
void hpfs_unlock_inode(struct inode *);
-void hpfs_lock_2inodes(struct inode *, struct inode *);
-void hpfs_unlock_2inodes(struct inode *, struct inode *);
-void hpfs_lock_3inodes(struct inode *, struct inode *, struct inode *);
-void hpfs_unlock_3inodes(struct inode *, struct inode *, struct inode *);
void *hpfs_map_sector(struct super_block *, unsigned, struct buffer_head **, int);
void *hpfs_get_sector(struct super_block *, unsigned, struct buffer_head **);
void *hpfs_map_4sectors(struct super_block *, unsigned, struct quad_buffer_head *, int);
--- diff/fs/hpfs/namei.c 2003-10-09 08:47:17.000000000 +0000
+++ source/fs/hpfs/namei.c 2004-03-16 11:27:48.401829464 +0000
@@ -28,8 +28,13 @@ int hpfs_mkdir(struct inode *dir, struct
int err;
if ((err = hpfs_chk_name((char *)name, &len))) return err==-ENOENT ? -EINVAL : err;
lock_kernel();
- if (!(fnode = hpfs_alloc_fnode(dir->i_sb, hpfs_i(dir)->i_dno, &fno, &bh))) goto bail;
- if (!(dnode = hpfs_alloc_dnode(dir->i_sb, fno, &dno, &qbh0, 1))) goto bail1;
+ err = -ENOSPC;
+ fnode = hpfs_alloc_fnode(dir->i_sb, hpfs_i(dir)->i_dno, &fno, &bh);
+ if (!fnode)
+ goto bail;
+ dnode = hpfs_alloc_dnode(dir->i_sb, fno, &dno, &qbh0, 1);
+ if (!dnode)
+ goto bail1;
memset(&dee, 0, sizeof dee);
dee.directory = 1;
if (!(mode & 0222)) dee.read_only = 1;
@@ -39,15 +44,11 @@ int hpfs_mkdir(struct inode *dir, struct
dee.creation_date = dee.write_date = dee.read_date = gmt_to_local(dir->i_sb, get_seconds());
hpfs_lock_inode(dir);
r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
- if (r == 1) goto bail2;
+ if (r == 1)
+ goto bail2;
if (r == -1) {
- brelse(bh);
- hpfs_brelse4(&qbh0);
- hpfs_free_sectors(dir->i_sb, fno, 1);
- hpfs_free_dnode(dir->i_sb, dno);
- hpfs_unlock_inode(dir);
- unlock_kernel();
- return -EEXIST;
+ err = -EEXIST;
+ goto bail2;
}
fnode->len = len;
memcpy(fnode->name, name, len > 15 ? 15 : len);
@@ -103,7 +104,7 @@ bail1:
hpfs_free_sectors(dir->i_sb, fno, 1);
bail:
unlock_kernel();
- return -ENOSPC;
+ return err;
}
int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struct nameidata *nd)
@@ -120,7 +121,10 @@ int hpfs_create(struct inode *dir, struc
if ((err = hpfs_chk_name((char *)name, &len)))
return err==-ENOENT ? -EINVAL : err;
lock_kernel();
- if (!(fnode = hpfs_alloc_fnode(dir->i_sb, hpfs_i(dir)->i_dno, &fno, &bh))) goto bail;
+ err = -ENOSPC;
+ fnode = hpfs_alloc_fnode(dir->i_sb, hpfs_i(dir)->i_dno, &fno, &bh);
+ if (!fnode)
+ goto bail;
memset(&dee, 0, sizeof dee);
if (!(mode & 0222)) dee.read_only = 1;
dee.archive = 1;
@@ -129,13 +133,11 @@ int hpfs_create(struct inode *dir, struc
dee.creation_date = dee.write_date = dee.read_date = gmt_to_local(dir->i_sb, get_seconds());
hpfs_lock_inode(dir);
r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
- if (r == 1) goto bail1;
+ if (r == 1)
+ goto bail1;
if (r == -1) {
- brelse(bh);
- hpfs_free_sectors(dir->i_sb, fno, 1);
- hpfs_unlock_inode(dir);
- unlock_kernel();
- return -EEXIST;
+ err = -EEXIST;
+ goto bail1;
}
fnode->len = len;
memcpy(fnode->name, name, len > 15 ? 15 : len);
@@ -178,7 +180,7 @@ bail1:
hpfs_unlock_inode(dir);
bail:
unlock_kernel();
- return -ENOSPC;
+ return err;
}
int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev)
@@ -197,7 +199,10 @@ int hpfs_mknod(struct inode *dir, struct
if (!new_valid_dev(rdev))
return -EINVAL;
lock_kernel();
- if (!(fnode = hpfs_alloc_fnode(dir->i_sb, hpfs_i(dir)->i_dno, &fno, &bh))) goto bail;
+ err = -ENOSPC;
+ fnode = hpfs_alloc_fnode(dir->i_sb, hpfs_i(dir)->i_dno, &fno, &bh);
+ if (!fnode)
+ goto bail;
memset(&dee, 0, sizeof dee);
if (!(mode & 0222)) dee.read_only = 1;
dee.archive = 1;
@@ -206,13 +211,11 @@ int hpfs_mknod(struct inode *dir, struct
dee.creation_date = dee.write_date = dee.read_date = gmt_to_local(dir->i_sb, get_seconds());
hpfs_lock_inode(dir);
r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
- if (r == 1) goto bail1;
+ if (r == 1)
+ goto bail1;
if (r == -1) {
- brelse(bh);
- hpfs_free_sectors(dir->i_sb, fno, 1);
- hpfs_unlock_inode(dir);
- unlock_kernel();
- return -EEXIST;
+ err = -EEXIST;
+ goto bail1;
}
fnode->len = len;
memcpy(fnode->name, name, len > 15 ? 15 : len);
@@ -248,7 +251,7 @@ bail1:
hpfs_unlock_inode(dir);
bail:
unlock_kernel();
- return -ENOSPC;
+ return err;
}
extern struct address_space_operations hpfs_symlink_aops;
@@ -270,7 +273,10 @@ int hpfs_symlink(struct inode *dir, stru
unlock_kernel();
return -EPERM;
}
- if (!(fnode = hpfs_alloc_fnode(dir->i_sb, hpfs_i(dir)->i_dno, &fno, &bh))) goto bail;
+ err = -ENOSPC;
+ fnode = hpfs_alloc_fnode(dir->i_sb, hpfs_i(dir)->i_dno, &fno, &bh);
+ if (!fnode)
+ goto bail;
memset(&dee, 0, sizeof dee);
dee.archive = 1;
dee.hidden = name[0] == '.';
@@ -278,13 +284,11 @@ int hpfs_symlink(struct inode *dir, stru
dee.creation_date = dee.write_date = dee.read_date = gmt_to_local(dir->i_sb, get_seconds());
hpfs_lock_inode(dir);
r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
- if (r == 1) goto bail1;
+ if (r == 1)
+ goto bail1;
if (r == -1) {
- brelse(bh);
- hpfs_free_sectors(dir->i_sb, fno, 1);
- hpfs_unlock_inode(dir);
- unlock_kernel();
- return -EEXIST;
+ err = -EEXIST;
+ goto bail1;
}
fnode->len = len;
memcpy(fnode->name, name, len > 15 ? 15 : len);
@@ -326,7 +330,7 @@ bail1:
hpfs_unlock_inode(dir);
bail:
unlock_kernel();
- return -ENOSPC;
+ return err;
}
int hpfs_unlink(struct inode *dir, struct dentry *dentry)
@@ -340,39 +344,41 @@ int hpfs_unlink(struct inode *dir, struc
fnode_secno fno;
int r;
int rep = 0;
+ int err;
lock_kernel();
hpfs_adjust_length((char *)name, &len);
again:
- hpfs_lock_2inodes(dir, inode);
- if (!(de = map_dirent(dir, hpfs_i(dir)->i_dno, (char *)name, len, &dno, &qbh))) {
- hpfs_unlock_2inodes(dir, inode);
- unlock_kernel();
- return -ENOENT;
- }
- if (de->first) {
- hpfs_brelse4(&qbh);
- hpfs_unlock_2inodes(dir, inode);
- unlock_kernel();
- return -EPERM;
- }
- if (de->directory) {
- hpfs_brelse4(&qbh);
- hpfs_unlock_2inodes(dir, inode);
- unlock_kernel();
- return -EISDIR;
- }
+ hpfs_lock_inode(dir);
+ hpfs_lock_inode(inode);
+ de = map_dirent(dir, hpfs_i(dir)->i_dno, (char *)name, len, &dno, &qbh);
+ err = -ENOENT;
+ if (!de)
+ goto out;
+
+ err = -EPERM;
+ if (de->first)
+ goto out1;
+
+ err = -EISDIR;
+ if (de->directory)
+ goto out1;
+
fno = de->fnode;
- if ((r = hpfs_remove_dirent(dir, dno, de, &qbh, 1)) == 1) hpfs_error(dir->i_sb, "there was error when removing dirent");
- if (r != 2) {
- inode->i_nlink--;
- hpfs_unlock_2inodes(dir, inode);
- } else { /* no space for deleting, try to truncate file */
- struct iattr newattrs;
- int err;
- hpfs_unlock_2inodes(dir, inode);
- if (rep)
- goto ret;
+ r = hpfs_remove_dirent(dir, dno, de, &qbh, 1);
+ switch (r) {
+ case 1:
+ hpfs_error(dir->i_sb, "there was error when removing dirent");
+ err = -EFSERROR;
+ break;
+ case 2: /* no space for deleting, try to truncate file */
+
+ err = -ENOSPC;
+ if (rep++)
+ break;
+
+ hpfs_unlock_inode(inode);
+ hpfs_unlock_inode(dir);
d_drop(dentry);
spin_lock(&dentry->d_lock);
if (atomic_read(&dentry->d_count) > 1 ||
@@ -381,22 +387,32 @@ again:
get_write_access(inode)) {
spin_unlock(&dentry->d_lock);
d_rehash(dentry);
- goto ret;
+ } else {
+ struct iattr newattrs;
+ spin_unlock(&dentry->d_lock);
+ /*printk("HPFS: truncating file before delete.\n");*/
+ newattrs.ia_size = 0;
+ newattrs.ia_valid = ATTR_SIZE | ATTR_CTIME;
+ err = notify_change(dentry, &newattrs);
+ put_write_access(inode);
+ if (!err)
+ goto again;
}
- spin_unlock(&dentry->d_lock);
- /*printk("HPFS: truncating file before delete.\n");*/
- newattrs.ia_size = 0;
- newattrs.ia_valid = ATTR_SIZE | ATTR_CTIME;
- err = notify_change(dentry, &newattrs);
- put_write_access(inode);
- if (err)
- goto ret;
- rep = 1;
- goto again;
+ unlock_kernel();
+ return -ENOSPC;
+ default:
+ inode->i_nlink--;
+ err = 0;
}
-ret:
+ goto out;
+
+out1:
+ hpfs_brelse4(&qbh);
+out:
+ hpfs_unlock_inode(inode);
+ hpfs_unlock_inode(dir);
unlock_kernel();
- return r == 2 ? -ENOSPC : r == 1 ? -EFSERROR : 0;
+ return err;
}
int hpfs_rmdir(struct inode *dir, struct dentry *dentry)
@@ -409,44 +425,54 @@ int hpfs_rmdir(struct inode *dir, struct
dnode_secno dno;
fnode_secno fno;
int n_items = 0;
+ int err;
int r;
+
hpfs_adjust_length((char *)name, &len);
lock_kernel();
- hpfs_lock_2inodes(dir, inode);
- if (!(de = map_dirent(dir, hpfs_i(dir)->i_dno, (char *)name, len, &dno, &qbh))) {
- hpfs_unlock_2inodes(dir, inode);
- unlock_kernel();
- return -ENOENT;
- }
- if (de->first) {
- hpfs_brelse4(&qbh);
- hpfs_unlock_2inodes(dir, inode);
- unlock_kernel();
- return -EPERM;
- }
- if (!de->directory) {
- hpfs_brelse4(&qbh);
- hpfs_unlock_2inodes(dir, inode);
- unlock_kernel();
- return -ENOTDIR;
- }
+ hpfs_lock_inode(dir);
+ hpfs_lock_inode(inode);
+ err = -ENOENT;
+ de = map_dirent(dir, hpfs_i(dir)->i_dno, (char *)name, len, &dno, &qbh);
+ if (!de)
+ goto out;
+
+ err = -EPERM;
+ if (de->first)
+ goto out1;
+
+ err = -ENOTDIR;
+ if (!de->directory)
+ goto out1;
+
hpfs_count_dnodes(dir->i_sb, hpfs_i(inode)->i_dno, NULL, NULL, &n_items);
- if (n_items) {
- hpfs_brelse4(&qbh);
- hpfs_unlock_2inodes(dir, inode);
- unlock_kernel();
- return -ENOTEMPTY;
- }
+ err = -ENOTEMPTY;
+ if (n_items)
+ goto out1;
+
fno = de->fnode;
- if ((r = hpfs_remove_dirent(dir, dno, de, &qbh, 1)) == 1)
+ r = hpfs_remove_dirent(dir, dno, de, &qbh, 1);
+ switch (r) {
+ case 1:
hpfs_error(dir->i_sb, "there was error when removing dirent");
- if (r != 2) {
+ err = -EFSERROR;
+ break;
+ case 2:
+ err = -ENOSPC;
+ break;
+ default:
dir->i_nlink--;
inode->i_nlink = 0;
- hpfs_unlock_2inodes(dir, inode);
- } else hpfs_unlock_2inodes(dir, inode);
+ err = 0;
+ }
+ goto out;
+out1:
+ hpfs_brelse4(&qbh);
+out:
+ hpfs_unlock_inode(inode);
+ hpfs_unlock_inode(dir);
unlock_kernel();
- return r == 2 ? -ENOSPC : r == 1 ? -EFSERROR : 0;
+ return err;
}
int hpfs_symlink_readpage(struct file *file, struct page *page)
@@ -501,7 +527,10 @@ int hpfs_rename(struct inode *old_dir, s
hpfs_adjust_length((char *)old_name, &old_len);
lock_kernel();
- hpfs_lock_3inodes(old_dir, new_dir, i);
+ hpfs_lock_inode(old_dir);
+ if (new_dir != old_dir)
+ hpfs_lock_inode(new_dir);
+ hpfs_lock_inode(i);
/* Erm? Moving over the empty non-busy directory is perfectly legal */
if (new_inode && S_ISDIR(new_inode->i_mode)) {
@@ -580,7 +609,10 @@ int hpfs_rename(struct inode *old_dir, s
hpfs_i(i)->i_conv = hpfs_sb(i->i_sb)->sb_conv;
hpfs_decide_conv(i, (char *)new_name, new_len);
end1:
- hpfs_unlock_3inodes(old_dir, new_dir, i);
+ hpfs_unlock_inode(i);
+ if (old_dir != new_dir)
+ hpfs_unlock_inode(new_dir);
+ hpfs_unlock_inode(old_dir);
unlock_kernel();
return err;
}
--- diff/fs/hugetlbfs/inode.c 2004-02-18 08:54:12.000000000 +0000
+++ source/fs/hugetlbfs/inode.c 2004-03-16 11:27:48.402829312 +0000
@@ -194,6 +194,7 @@ static void hugetlbfs_delete_inode(struc
hlist_del_init(&inode->i_hash);
list_del_init(&inode->i_list);
+ list_del_init(&inode->i_sb_list);
inode->i_state |= I_FREEING;
inodes_stat.nr_inodes--;
spin_unlock(&inode_lock);
@@ -236,6 +237,7 @@ static void hugetlbfs_forget_inode(struc
hlist_del_init(&inode->i_hash);
out_truncate:
list_del_init(&inode->i_list);
+ list_del_init(&inode->i_sb_list);
inode->i_state |= I_FREEING;
inodes_stat.nr_inodes--;
spin_unlock(&inode_lock);
--- diff/fs/inode.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/inode.c 2004-03-16 11:27:48.404829008 +0000
@@ -176,15 +176,12 @@ void inode_init_once(struct inode *inode
{
memset(inode, 0, sizeof(*inode));
INIT_HLIST_NODE(&inode->i_hash);
- INIT_LIST_HEAD(&inode->i_data.clean_pages);
- INIT_LIST_HEAD(&inode->i_data.dirty_pages);
- INIT_LIST_HEAD(&inode->i_data.locked_pages);
- INIT_LIST_HEAD(&inode->i_data.io_pages);
INIT_LIST_HEAD(&inode->i_dentry);
INIT_LIST_HEAD(&inode->i_devices);
sema_init(&inode->i_sem, 1);
+ init_rwsem(&inode->i_alloc_sem);
INIT_RADIX_TREE(&inode->i_data.page_tree, GFP_ATOMIC);
- spin_lock_init(&inode->i_data.page_lock);
+ spin_lock_init(&inode->i_data.tree_lock);
init_MUTEX(&inode->i_data.i_shared_sem);
atomic_set(&inode->i_data.truncate_count, 0);
INIT_LIST_HEAD(&inode->i_data.private_list);
@@ -285,7 +282,7 @@ static void dispose_list(struct list_hea
/*
* Invalidate all inodes for a device.
*/
-static int invalidate_list(struct list_head *head, struct super_block * sb, struct list_head * dispose)
+static int invalidate_list(struct list_head *head, struct list_head *dispose)
{
struct list_head *next;
int busy = 0, count = 0;
@@ -298,13 +295,12 @@ static int invalidate_list(struct list_h
next = next->next;
if (tmp == head)
break;
- inode = list_entry(tmp, struct inode, i_list);
- if (inode->i_sb != sb)
- continue;
+ inode = list_entry(tmp, struct inode, i_sb_list);
invalidate_inode_buffers(inode);
if (!atomic_read(&inode->i_count)) {
hlist_del_init(&inode->i_hash);
list_del(&inode->i_list);
+ list_del(&inode->i_sb_list);
list_add(&inode->i_list, dispose);
inode->i_state |= I_FREEING;
count++;
@@ -340,10 +336,7 @@ int invalidate_inodes(struct super_block
down(&iprune_sem);
spin_lock(&inode_lock);
- busy = invalidate_list(&inode_in_use, sb, &throw_away);
- busy |= invalidate_list(&inode_unused, sb, &throw_away);
- busy |= invalidate_list(&sb->s_dirty, sb, &throw_away);
- busy |= invalidate_list(&sb->s_io, sb, &throw_away);
+ busy = invalidate_list(&sb->s_inodes, &throw_away);
spin_unlock(&inode_lock);
dispose_list(&throw_away);
@@ -443,6 +436,7 @@ static void prune_icache(int nr_to_scan)
continue;
}
hlist_del_init(&inode->i_hash);
+ list_del_init(&inode->i_sb_list);
list_move(&inode->i_list, &freeable);
inode->i_state |= I_FREEING;
nr_pruned++;
@@ -553,6 +547,7 @@ struct inode *new_inode(struct super_blo
spin_lock(&inode_lock);
inodes_stat.nr_inodes++;
list_add(&inode->i_list, &inode_in_use);
+ list_add(&inode->i_sb_list, &sb->s_inodes);
inode->i_ino = ++last_ino;
inode->i_state = 0;
spin_unlock(&inode_lock);
@@ -601,6 +596,7 @@ static struct inode * get_new_inode(stru
inodes_stat.nr_inodes++;
list_add(&inode->i_list, &inode_in_use);
+ list_add(&inode->i_sb_list, &sb->s_inodes);
hlist_add_head(&inode->i_hash, head);
inode->i_state = I_LOCK|I_NEW;
spin_unlock(&inode_lock);
@@ -649,6 +645,7 @@ static struct inode * get_new_inode_fast
inode->i_ino = ino;
inodes_stat.nr_inodes++;
list_add(&inode->i_list, &inode_in_use);
+ list_add(&inode->i_sb_list, &sb->s_inodes);
hlist_add_head(&inode->i_hash, head);
inode->i_state = I_LOCK|I_NEW;
spin_unlock(&inode_lock);
@@ -984,6 +981,7 @@ void generic_delete_inode(struct inode *
struct super_operations *op = inode->i_sb->s_op;
list_del_init(&inode->i_list);
+ list_del_init(&inode->i_sb_list);
inode->i_state|=I_FREEING;
inodes_stat.nr_inodes--;
spin_unlock(&inode_lock);
@@ -1031,6 +1029,7 @@ static void generic_forget_inode(struct
hlist_del_init(&inode->i_hash);
}
list_del_init(&inode->i_list);
+ list_del_init(&inode->i_sb_list);
inode->i_state|=I_FREEING;
inodes_stat.nr_inodes--;
spin_unlock(&inode_lock);
@@ -1216,44 +1215,27 @@ EXPORT_SYMBOL(inode_needs_sync);
*/
#ifdef CONFIG_QUOTA
-/* Functions back in dquot.c */
-void put_dquot_list(struct list_head *);
+/* Function back in dquot.c */
int remove_inode_dquot_ref(struct inode *, int, struct list_head *);
-void remove_dquot_ref(struct super_block *sb, int type)
+void remove_dquot_ref(struct super_block *sb, int type,
+ struct list_head *tofree_head)
{
struct inode *inode;
- struct list_head *act_head;
- LIST_HEAD(tofree_head);
if (!sb->dq_op)
return; /* nothing to do */
spin_lock(&inode_lock); /* This lock is for inodes code */
- /* We don't have to lock against quota code - test IS_QUOTAINIT is just for speedup... */
-
- list_for_each(act_head, &inode_in_use) {
- inode = list_entry(act_head, struct inode, i_list);
- if (inode->i_sb == sb && IS_QUOTAINIT(inode))
- remove_inode_dquot_ref(inode, type, &tofree_head);
- }
- list_for_each(act_head, &inode_unused) {
- inode = list_entry(act_head, struct inode, i_list);
- if (inode->i_sb == sb && IS_QUOTAINIT(inode))
- remove_inode_dquot_ref(inode, type, &tofree_head);
- }
- list_for_each(act_head, &sb->s_dirty) {
- inode = list_entry(act_head, struct inode, i_list);
- if (IS_QUOTAINIT(inode))
- remove_inode_dquot_ref(inode, type, &tofree_head);
- }
- list_for_each(act_head, &sb->s_io) {
- inode = list_entry(act_head, struct inode, i_list);
+ /*
+ * We don't have to lock against quota code - test IS_QUOTAINIT is
+ * just for speedup...
+ */
+
+ list_for_each_entry(inode, &sb->s_inodes, i_sb_list)
if (IS_QUOTAINIT(inode))
- remove_inode_dquot_ref(inode, type, &tofree_head);
- }
- spin_unlock(&inode_lock);
+ remove_inode_dquot_ref(inode, type, tofree_head);
- put_dquot_list(&tofree_head);
+ spin_unlock(&inode_lock);
}
#endif
--- diff/fs/jfs/jfs_logmgr.c 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/jfs/jfs_logmgr.c 2004-03-16 11:27:48.405828856 +0000
@@ -1972,8 +1972,7 @@ static int lbmRead(struct jfs_log * log,
bio->bi_end_io = lbmIODone;
bio->bi_private = bp;
- submit_bio(READ, bio);
- blk_run_queues();
+ submit_bio(READ_SYNC, bio);
wait_event(bp->l_ioevent, (bp->l_flag != lbmREAD));
@@ -2117,9 +2116,8 @@ static void lbmStartIO(struct lbuf * bp)
/* check if journaling to disk has been disabled */
if (!log->no_integrity) {
- submit_bio(WRITE, bio);
+ submit_bio(WRITE_SYNC, bio);
INCREMENT(lmStat.submitted);
- blk_run_queues();
}
else {
bio->bi_size = 0;
--- diff/fs/mpage.c 2003-08-26 09:00:54.000000000 +0000
+++ source/fs/mpage.c 2004-03-16 11:27:48.408828400 +0000
@@ -329,10 +329,10 @@ mpage_readpages(struct address_space *ma
pagevec_init(&lru_pvec, 0);
for (page_idx = 0; page_idx < nr_pages; page_idx++) {
- struct page *page = list_entry(pages->prev, struct page, list);
+ struct page *page = list_entry(pages->prev, struct page, lru);
prefetchw(&page->flags);
- list_del(&page->list);
+ list_del(&page->lru);
if (!add_to_page_cache(page, mapping,
page->index, GFP_KERNEL)) {
bio = do_mpage_readpage(bio, page,
@@ -546,7 +546,7 @@ alloc_new:
}
BUG_ON(PageWriteback(page));
- SetPageWriteback(page);
+ set_page_writeback(page);
unlock_page(page);
if (boundary || (first_unmapped != blocks_per_page)) {
bio = mpage_bio_submit(WRITE, bio);
@@ -589,31 +589,13 @@ out:
* This is a library function, which implements the writepages()
* address_space_operation.
*
- * (The next two paragraphs refer to code which isn't here yet, but they
- * explain the presence of address_space.io_pages)
- *
- * Pages can be moved from clean_pages or locked_pages onto dirty_pages
- * at any time - it's not possible to lock against that. So pages which
- * have already been added to a BIO may magically reappear on the dirty_pages
- * list. And mpage_writepages() will again try to lock those pages.
- * But I/O has not yet been started against the page. Thus deadlock.
- *
- * To avoid this, mpage_writepages() will only write pages from io_pages. The
- * caller must place them there. We walk io_pages, locking the pages and
- * submitting them for I/O, moving them to locked_pages.
- *
- * This has the added benefit of preventing a livelock which would otherwise
- * occur if pages are being dirtied faster than we can write them out.
- *
* If a page is already under I/O, generic_writepages() skips it, even
* if it's dirty. This is desirable behaviour for memory-cleaning writeback,
* but it is INCORRECT for data-integrity system calls such as fsync(). fsync()
* and msync() need to guarantee that all the data which was dirty at the time
- * the call was made get new I/O started against them. So if called_for_sync()
- * is true, we must wait for existing IO to complete.
- *
- * It's fairly rare for PageWriteback pages to be on ->dirty_pages. It
- * means that someone redirtied the page while it was under I/O.
+ * the call was made get new I/O started against them. If wbc->sync_mode is
+ * WB_SYNC_ALL then we were called for data integrity and we must wait for
+ * existing IO to complete.
*/
int
mpage_writepages(struct address_space *mapping,
@@ -625,6 +607,9 @@ mpage_writepages(struct address_space *m
int ret = 0;
int done = 0;
int (*writepage)(struct page *page, struct writeback_control *wbc);
+ struct pagevec pvec;
+ int nr_pages;
+ pgoff_t index;
if (wbc->nonblocking && bdi_write_congested(bdi)) {
wbc->encountered_congestion = 1;
@@ -635,72 +620,58 @@ mpage_writepages(struct address_space *m
if (get_block == NULL)
writepage = mapping->a_ops->writepage;
- spin_lock(&mapping->page_lock);
- while (!list_empty(&mapping->io_pages) && !done) {
- struct page *page = list_entry(mapping->io_pages.prev,
- struct page, list);
- list_del(&page->list);
- if (PageWriteback(page) && wbc->sync_mode == WB_SYNC_NONE) {
- if (PageDirty(page)) {
- list_add(&page->list, &mapping->dirty_pages);
- continue;
- }
- list_add(&page->list, &mapping->locked_pages);
- continue;
- }
- if (!PageDirty(page)) {
- list_add(&page->list, &mapping->clean_pages);
- continue;
- }
- list_add(&page->list, &mapping->locked_pages);
-
- page_cache_get(page);
- spin_unlock(&mapping->page_lock);
-
- /*
- * At this point we hold neither mapping->page_lock nor
- * lock on the page itself: the page may be truncated or
- * invalidated (changing page->mapping to NULL), or even
- * swizzled back from swapper_space to tmpfs file mapping.
- */
-
- lock_page(page);
-
- if (wbc->sync_mode != WB_SYNC_NONE)
- wait_on_page_writeback(page);
-
- if (page->mapping == mapping && !PageWriteback(page) &&
- test_clear_page_dirty(page)) {
- if (writepage) {
- ret = (*writepage)(page, wbc);
- if (ret) {
- if (ret == -ENOSPC)
- set_bit(AS_ENOSPC,
- &mapping->flags);
- else
- set_bit(AS_EIO,
- &mapping->flags);
+ pagevec_init(&pvec, 0);
+ index = 0;
+ while (!done && (nr_pages = pagevec_lookup_tag(&pvec, mapping, &index,
+ PAGECACHE_TAG_DIRTY, PAGEVEC_SIZE))) {
+ unsigned i;
+
+ for (i = 0; i < nr_pages; i++) {
+ struct page *page = pvec.pages[i];
+
+ /*
+ * At this point we hold neither mapping->tree_lock nor
+ * lock on the page itself: the page may be truncated or
+ * invalidated (changing page->mapping to NULL), or even
+ * swizzled back from swapper_space to tmpfs file
+ * mapping
+ */
+
+ lock_page(page);
+
+ if (wbc->sync_mode != WB_SYNC_NONE)
+ wait_on_page_writeback(page);
+
+ if (page->mapping == mapping && !PageWriteback(page) &&
+ test_clear_page_dirty(page)) {
+ if (writepage) {
+ ret = (*writepage)(page, wbc);
+ if (ret) {
+ if (ret == -ENOSPC)
+ set_bit(AS_ENOSPC,
+ &mapping->flags);
+ else
+ set_bit(AS_EIO,
+ &mapping->flags);
+ }
+ } else {
+ bio = mpage_writepage(bio, page,
+ get_block, &last_block_in_bio,
+ &ret, wbc);
+ }
+ if (ret || (--(wbc->nr_to_write) <= 0))
+ done = 1;
+ if (wbc->nonblocking &&
+ bdi_write_congested(bdi)) {
+ wbc->encountered_congestion = 1;
+ done = 1;
}
} else {
- bio = mpage_writepage(bio, page, get_block,
- &last_block_in_bio, &ret, wbc);
- }
- if (ret || (--(wbc->nr_to_write) <= 0))
- done = 1;
- if (wbc->nonblocking && bdi_write_congested(bdi)) {
- wbc->encountered_congestion = 1;
- done = 1;
+ unlock_page(page);
}
- } else {
- unlock_page(page);
}
- page_cache_release(page);
- spin_lock(&mapping->page_lock);
+ pagevec_release(&pvec);
}
- /*
- * Leave any remaining dirty pages on ->io_pages
- */
- spin_unlock(&mapping->page_lock);
if (bio)
mpage_bio_submit(WRITE, bio);
return ret;
--- diff/fs/namei.c 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/namei.c 2004-03-16 11:27:48.410828096 +0000
@@ -26,6 +26,7 @@
#include <linux/personality.h>
#include <linux/security.h>
#include <linux/mount.h>
+#include <linux/audit.h>
#include <asm/namei.h>
#include <asm/uaccess.h>
@@ -141,10 +142,12 @@ char * getname(const char __user * filen
result = tmp;
if (retval < 0) {
- putname(tmp);
+ __putname(tmp);
result = ERR_PTR(retval);
}
}
+ if (unlikely(current->audit_context) && !IS_ERR(result) && result)
+ audit_getname(result);
return result;
}
@@ -860,6 +863,8 @@ walk_init_root(const char *name, struct
int fastcall path_lookup(const char *name, unsigned int flags, struct nameidata *nd)
{
+ int retval;
+
nd->last_type = LAST_ROOT; /* if there are only slashes... */
nd->flags = flags;
@@ -882,7 +887,13 @@ int fastcall path_lookup(const char *nam
}
read_unlock(¤t->fs->lock);
current->total_link_count = 0;
- return link_path_walk(name, nd);
+ retval = link_path_walk(name, nd);
+ if (unlikely(current->audit_context
+ && nd && nd->dentry && nd->dentry->d_inode))
+ audit_inode(name,
+ nd->dentry->d_inode->i_ino,
+ nd->dentry->d_inode->i_rdev);
+ return retval;
}
/*
--- diff/fs/namespace.c 2004-02-18 08:54:12.000000000 +0000
+++ source/fs/namespace.c 2004-03-16 11:27:48.411827944 +0000
@@ -24,7 +24,15 @@
#include <asm/uaccess.h>
extern int __init init_rootfs(void);
+
+#ifdef CONFIG_SYSFS
extern int __init sysfs_init(void);
+#else
+static inline int sysfs_init(void)
+{
+ return 0;
+}
+#endif
/* spinlock for vfsmount related operations, inplace of dcache_lock */
spinlock_t vfsmount_lock __cacheline_aligned_in_smp = SPIN_LOCK_UNLOCKED;
--- diff/fs/nfs/direct.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/nfs/direct.c 2004-03-16 11:27:48.411827944 +0000
@@ -128,6 +128,7 @@ nfs_direct_read_seg(struct inode *inode,
.inode = inode,
.args = {
.fh = NFS_FH(inode),
+ .lockowner = current->files,
},
.res = {
.fattr = &rdata.fattr,
@@ -258,6 +259,7 @@ nfs_direct_write_seg(struct inode *inode
.inode = inode,
.args = {
.fh = NFS_FH(inode),
+ .lockowner = current->files,
},
.res = {
.fattr = &wdata.fattr,
--- diff/fs/nfs/nfs3proc.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/nfs/nfs3proc.c 2004-03-16 11:27:48.413827640 +0000
@@ -729,11 +729,10 @@ nfs3_read_done(struct rpc_task *task)
}
static void
-nfs3_proc_read_setup(struct nfs_read_data *data, unsigned int count)
+nfs3_proc_read_setup(struct nfs_read_data *data)
{
struct rpc_task *task = &data->task;
struct inode *inode = data->inode;
- struct nfs_page *req;
int flags;
struct rpc_message msg = {
.rpc_proc = &nfs3_procedures[NFS3PROC_READ],
@@ -741,27 +740,13 @@ nfs3_proc_read_setup(struct nfs_read_dat
.rpc_resp = &data->res,
.rpc_cred = data->cred,
};
-
- req = nfs_list_entry(data->pages.next);
- data->args.fh = NFS_FH(inode);
- data->args.offset = req_offset(req);
- data->args.pgbase = req->wb_pgbase;
- data->args.pages = data->pagevec;
- data->args.count = count;
- data->res.fattr = &data->fattr;
- data->res.count = count;
- data->res.eof = 0;
-
+
/* N.B. Do we need to test? Never called for swapfile inode */
flags = RPC_TASK_ASYNC | (IS_SWAPFILE(inode)? NFS_RPC_SWAPFLAGS : 0);
/* Finalize the task. */
rpc_init_task(task, NFS_CLIENT(inode), nfs3_read_done, flags);
- task->tk_calldata = data;
- /* Release requests */
- task->tk_release = nfs_readdata_release;
-
- rpc_call_setup(&data->task, &msg, 0);
+ rpc_call_setup(task, &msg, 0);
}
static void
@@ -778,11 +763,10 @@ nfs3_write_done(struct rpc_task *task)
}
static void
-nfs3_proc_write_setup(struct nfs_write_data *data, unsigned int count, int how)
+nfs3_proc_write_setup(struct nfs_write_data *data, int how)
{
struct rpc_task *task = &data->task;
struct inode *inode = data->inode;
- struct nfs_page *req;
int stable;
int flags;
struct rpc_message msg = {
@@ -799,28 +783,14 @@ nfs3_proc_write_setup(struct nfs_write_d
stable = NFS_DATA_SYNC;
} else
stable = NFS_UNSTABLE;
-
- req = nfs_list_entry(data->pages.next);
- data->args.fh = NFS_FH(inode);
- data->args.offset = req_offset(req);
- data->args.pgbase = req->wb_pgbase;
- data->args.count = count;
data->args.stable = stable;
- data->args.pages = data->pagevec;
- data->res.fattr = &data->fattr;
- data->res.count = count;
- data->res.verf = &data->verf;
/* Set the initial flags for the task. */
flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
/* Finalize the task. */
rpc_init_task(task, NFS_CLIENT(inode), nfs3_write_done, flags);
- task->tk_calldata = data;
- /* Release requests */
- task->tk_release = nfs_writedata_release;
-
- rpc_call_setup(&data->task, &msg, 0);
+ rpc_call_setup(task, &msg, 0);
}
static void
@@ -837,7 +807,7 @@ nfs3_commit_done(struct rpc_task *task)
}
static void
-nfs3_proc_commit_setup(struct nfs_write_data *data, u64 start, u32 len, int how)
+nfs3_proc_commit_setup(struct nfs_write_data *data, int how)
{
struct rpc_task *task = &data->task;
struct inode *inode = data->inode;
@@ -849,23 +819,12 @@ nfs3_proc_commit_setup(struct nfs_write_
.rpc_cred = data->cred,
};
- data->args.fh = NFS_FH(data->inode);
- data->args.offset = start;
- data->args.count = len;
- data->res.count = len;
- data->res.fattr = &data->fattr;
- data->res.verf = &data->verf;
-
/* Set the initial flags for the task. */
flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
/* Finalize the task. */
rpc_init_task(task, NFS_CLIENT(inode), nfs3_commit_done, flags);
- task->tk_calldata = data;
- /* Release requests */
- task->tk_release = nfs_commit_release;
-
- rpc_call_setup(&data->task, &msg, 0);
+ rpc_call_setup(task, &msg, 0);
}
/*
--- diff/fs/nfs/nfs4proc.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/nfs/nfs4proc.c 2004-03-16 11:27:48.415827336 +0000
@@ -1096,10 +1096,10 @@ nfs4_proc_read(struct nfs_read_data *rda
if (filp) {
struct nfs4_state *state;
state = (struct nfs4_state *)filp->private_data;
- nfs4_copy_stateid(&rdata->args.stateid, state, rdata->lockowner);
+ rdata->args.state = state;
msg.rpc_cred = state->owner->so_cred;
} else {
- memcpy(&rdata->args.stateid, &zero_stateid, sizeof(rdata->args.stateid));
+ rdata->args.state = NULL;
msg.rpc_cred = NFS_I(inode)->mm_cred;
}
@@ -1134,10 +1134,10 @@ nfs4_proc_write(struct nfs_write_data *w
if (filp) {
struct nfs4_state *state;
state = (struct nfs4_state *)filp->private_data;
- nfs4_copy_stateid(&wdata->args.stateid, state, wdata->lockowner);
+ wdata->args.state = state;
msg.rpc_cred = state->owner->so_cred;
} else {
- memcpy(&wdata->args.stateid, &zero_stateid, sizeof(wdata->args.stateid));
+ wdata->args.state = NULL;
msg.rpc_cred = NFS_I(inode)->mm_cred;
}
@@ -1166,15 +1166,10 @@ nfs4_proc_commit(struct nfs_write_data *
/*
* Try first to use O_WRONLY, then O_RDWR stateid.
*/
- if (filp) {
- struct nfs4_state *state;
- state = (struct nfs4_state *)filp->private_data;
- nfs4_copy_stateid(&cdata->args.stateid, state, cdata->lockowner);
- msg.rpc_cred = state->owner->so_cred;
- } else {
- memcpy(&cdata->args.stateid, &zero_stateid, sizeof(cdata->args.stateid));
+ if (filp)
+ msg.rpc_cred = ((struct nfs4_state *)filp->private_data)->owner->so_cred;
+ else
msg.rpc_cred = NFS_I(inode)->mm_cred;
- }
fattr->valid = 0;
status = rpc_call_sync(server->client, &msg, 0);
@@ -1510,27 +1505,13 @@ nfs4_proc_pathconf(struct nfs_server *se
}
static void
-nfs4_restart_read(struct rpc_task *task)
-{
- struct nfs_read_data *data = (struct nfs_read_data *)task->tk_calldata;
- struct nfs_page *req;
-
- rpc_restart_call(task);
- req = nfs_list_entry(data->pages.next);
- if (req->wb_state)
- nfs4_copy_stateid(&data->args.stateid, req->wb_state, req->wb_lockowner);
- else
- memcpy(&data->args.stateid, &zero_stateid, sizeof(data->args.stateid));
-}
-
-static void
nfs4_read_done(struct rpc_task *task)
{
struct nfs_read_data *data = (struct nfs_read_data *) task->tk_calldata;
struct inode *inode = data->inode;
if (nfs4_async_handle_error(task, NFS_SERVER(inode)) == -EAGAIN) {
- task->tk_action = nfs4_restart_read;
+ rpc_restart_call(task);
return;
}
if (task->tk_status > 0)
@@ -1540,7 +1521,7 @@ nfs4_read_done(struct rpc_task *task)
}
static void
-nfs4_proc_read_setup(struct nfs_read_data *data, unsigned int count)
+nfs4_proc_read_setup(struct nfs_read_data *data)
{
struct rpc_task *task = &data->task;
struct rpc_message msg = {
@@ -1550,59 +1531,26 @@ nfs4_proc_read_setup(struct nfs_read_dat
.rpc_cred = data->cred,
};
struct inode *inode = data->inode;
- struct nfs_page *req = nfs_list_entry(data->pages.next);
int flags;
- data->args.fh = NFS_FH(inode);
- data->args.offset = req_offset(req);
- data->args.pgbase = req->wb_pgbase;
- data->args.pages = data->pagevec;
- data->args.count = count;
- data->res.fattr = &data->fattr;
- data->res.count = count;
- data->res.eof = 0;
data->timestamp = jiffies;
- data->lockowner = req->wb_lockowner;
- if (req->wb_state)
- nfs4_copy_stateid(&data->args.stateid, req->wb_state, req->wb_lockowner);
- else
- memcpy(&data->args.stateid, &zero_stateid, sizeof(data->args.stateid));
-
/* N.B. Do we need to test? Never called for swapfile inode */
flags = RPC_TASK_ASYNC | (IS_SWAPFILE(inode)? NFS_RPC_SWAPFLAGS : 0);
/* Finalize the task. */
rpc_init_task(task, NFS_CLIENT(inode), nfs4_read_done, flags);
- task->tk_calldata = data;
- /* Release requests */
- task->tk_release = nfs_readdata_release;
-
rpc_call_setup(task, &msg, 0);
}
static void
-nfs4_restart_write(struct rpc_task *task)
-{
- struct nfs_write_data *data = (struct nfs_write_data *)task->tk_calldata;
- struct nfs_page *req;
-
- rpc_restart_call(task);
- req = nfs_list_entry(data->pages.next);
- if (req->wb_state)
- nfs4_copy_stateid(&data->args.stateid, req->wb_state, req->wb_lockowner);
- else
- memcpy(&data->args.stateid, &zero_stateid, sizeof(data->args.stateid));
-}
-
-static void
nfs4_write_done(struct rpc_task *task)
{
struct nfs_write_data *data = (struct nfs_write_data *) task->tk_calldata;
struct inode *inode = data->inode;
if (nfs4_async_handle_error(task, NFS_SERVER(inode)) == -EAGAIN) {
- task->tk_action = nfs4_restart_write;
+ rpc_restart_call(task);
return;
}
if (task->tk_status >= 0)
@@ -1612,7 +1560,7 @@ nfs4_write_done(struct rpc_task *task)
}
static void
-nfs4_proc_write_setup(struct nfs_write_data *data, unsigned int count, int how)
+nfs4_proc_write_setup(struct nfs_write_data *data, int how)
{
struct rpc_task *task = &data->task;
struct rpc_message msg = {
@@ -1622,7 +1570,6 @@ nfs4_proc_write_setup(struct nfs_write_d
.rpc_cred = data->cred,
};
struct inode *inode = data->inode;
- struct nfs_page *req = nfs_list_entry(data->pages.next);
int stable;
int flags;
@@ -1633,33 +1580,15 @@ nfs4_proc_write_setup(struct nfs_write_d
stable = NFS_DATA_SYNC;
} else
stable = NFS_UNSTABLE;
-
- data->args.fh = NFS_FH(inode);
- data->args.offset = req_offset(req);
- data->args.pgbase = req->wb_pgbase;
- data->args.count = count;
data->args.stable = stable;
- data->args.pages = data->pagevec;
- data->res.fattr = &data->fattr;
- data->res.count = count;
- data->res.verf = &data->verf;
- data->timestamp = jiffies;
- data->lockowner = req->wb_lockowner;
- if (req->wb_state)
- nfs4_copy_stateid(&data->args.stateid, req->wb_state, req->wb_lockowner);
- else
- memcpy(&data->args.stateid, &zero_stateid, sizeof(data->args.stateid));
+ data->timestamp = jiffies;
/* Set the initial flags for the task. */
flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
/* Finalize the task. */
rpc_init_task(task, NFS_CLIENT(inode), nfs4_write_done, flags);
- task->tk_calldata = data;
- /* Release requests */
- task->tk_release = nfs_writedata_release;
-
rpc_call_setup(task, &msg, 0);
}
@@ -1670,7 +1599,7 @@ nfs4_commit_done(struct rpc_task *task)
struct inode *inode = data->inode;
if (nfs4_async_handle_error(task, NFS_SERVER(inode)) == -EAGAIN) {
- task->tk_action = nfs4_restart_write;
+ rpc_restart_call(task);
return;
}
/* Call back common NFS writeback processing */
@@ -1678,7 +1607,7 @@ nfs4_commit_done(struct rpc_task *task)
}
static void
-nfs4_proc_commit_setup(struct nfs_write_data *data, u64 start, u32 len, int how)
+nfs4_proc_commit_setup(struct nfs_write_data *data, int how)
{
struct rpc_task *task = &data->task;
struct rpc_message msg = {
@@ -1690,22 +1619,11 @@ nfs4_proc_commit_setup(struct nfs_write_
struct inode *inode = data->inode;
int flags;
- data->args.fh = NFS_FH(data->inode);
- data->args.offset = start;
- data->args.count = len;
- data->res.count = len;
- data->res.fattr = &data->fattr;
- data->res.verf = &data->verf;
-
/* Set the initial flags for the task. */
flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
/* Finalize the task. */
rpc_init_task(task, NFS_CLIENT(inode), nfs4_commit_done, flags);
- task->tk_calldata = data;
- /* Release requests */
- task->tk_release = nfs_commit_release;
-
rpc_call_setup(task, &msg, 0);
}
--- diff/fs/nfs/nfs4state.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/nfs/nfs4state.c 2004-03-16 11:27:48.416827184 +0000
@@ -105,7 +105,7 @@ nfs4_alloc_client(struct in_addr *addr)
INIT_WORK(&clp->cl_renewd, nfs4_renew_state, clp);
INIT_LIST_HEAD(&clp->cl_superblocks);
init_waitqueue_head(&clp->cl_waitq);
- INIT_RPC_WAITQ(&clp->cl_rpcwaitq, "NFS4 client");
+ rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS4 client");
clp->cl_state = 1 << NFS4CLNT_NEW;
}
return clp;
--- diff/fs/nfs/nfs4xdr.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/nfs/nfs4xdr.c 2004-03-16 11:27:48.418826880 +0000
@@ -868,14 +868,32 @@ encode_putrootfh(struct xdr_stream *xdr)
return 0;
}
+static void
+encode_stateid(struct xdr_stream *xdr, struct nfs4_state *state, fl_owner_t lockowner)
+{
+ extern nfs4_stateid zero_stateid;
+ nfs4_stateid stateid;
+ uint32_t *p;
+
+ RESERVE_SPACE(16);
+ if (state != NULL) {
+ nfs4_copy_stateid(&stateid, state, lockowner);
+ WRITEMEM(stateid.data, sizeof(stateid.data));
+ } else
+ WRITEMEM(zero_stateid.data, sizeof(zero_stateid.data));
+}
+
static int
encode_read(struct xdr_stream *xdr, struct nfs_readargs *args)
{
uint32_t *p;
- RESERVE_SPACE(32);
+ RESERVE_SPACE(4);
WRITE32(OP_READ);
- WRITEMEM(args->stateid.data, sizeof(args->stateid.data));
+
+ encode_stateid(xdr, args->state, args->lockowner);
+
+ RESERVE_SPACE(12);
WRITE64(args->offset);
WRITE32(args->count);
@@ -1057,9 +1075,12 @@ encode_write(struct xdr_stream *xdr, str
{
uint32_t *p;
- RESERVE_SPACE(36);
+ RESERVE_SPACE(4);
WRITE32(OP_WRITE);
- WRITEMEM(args->stateid.data, sizeof(args->stateid.data));
+
+ encode_stateid(xdr, args->state, args->lockowner);
+
+ RESERVE_SPACE(16);
WRITE64(args->offset);
WRITE32(args->stable);
WRITE32(args->count);
--- diff/fs/nfs/pagelist.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/nfs/pagelist.c 2004-03-16 11:27:48.419826728 +0000
@@ -32,7 +32,7 @@ static inline struct nfs_page *
nfs_page_alloc(void)
{
struct nfs_page *p;
- p = kmem_cache_alloc(nfs_page_cachep, SLAB_NOFS);
+ p = kmem_cache_alloc(nfs_page_cachep, SLAB_KERNEL);
if (p) {
memset(p, 0, sizeof(*p));
INIT_LIST_HEAD(&p->wb_list);
@@ -88,6 +88,7 @@ nfs_create_request(struct file *file, st
* long write-back delay. This will be adjusted in
* update_nfs_request below if the region is not locked. */
req->wb_page = page;
+ atomic_set(&req->wb_complete, 0);
req->wb_index = page->index;
page_cache_get(page);
req->wb_offset = offset;
--- diff/fs/nfs/proc.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/nfs/proc.c 2004-03-16 11:27:48.420826576 +0000
@@ -559,11 +559,10 @@ nfs_read_done(struct rpc_task *task)
}
static void
-nfs_proc_read_setup(struct nfs_read_data *data, unsigned int count)
+nfs_proc_read_setup(struct nfs_read_data *data)
{
struct rpc_task *task = &data->task;
struct inode *inode = data->inode;
- struct nfs_page *req;
int flags;
struct rpc_message msg = {
.rpc_proc = &nfs_procedures[NFSPROC_READ],
@@ -571,27 +570,13 @@ nfs_proc_read_setup(struct nfs_read_data
.rpc_resp = &data->res,
.rpc_cred = data->cred,
};
-
- req = nfs_list_entry(data->pages.next);
- data->args.fh = NFS_FH(inode);
- data->args.offset = req_offset(req);
- data->args.pgbase = req->wb_pgbase;
- data->args.pages = data->pagevec;
- data->args.count = count;
- data->res.fattr = &data->fattr;
- data->res.count = count;
- data->res.eof = 0;
-
+
/* N.B. Do we need to test? Never called for swapfile inode */
flags = RPC_TASK_ASYNC | (IS_SWAPFILE(inode)? NFS_RPC_SWAPFLAGS : 0);
/* Finalize the task. */
rpc_init_task(task, NFS_CLIENT(inode), nfs_read_done, flags);
- task->tk_calldata = data;
- /* Release requests */
- task->tk_release = nfs_readdata_release;
-
- rpc_call_setup(&data->task, &msg, 0);
+ rpc_call_setup(task, &msg, 0);
}
static void
@@ -605,11 +590,10 @@ nfs_write_done(struct rpc_task *task)
}
static void
-nfs_proc_write_setup(struct nfs_write_data *data, unsigned int count, int how)
+nfs_proc_write_setup(struct nfs_write_data *data, int how)
{
struct rpc_task *task = &data->task;
struct inode *inode = data->inode;
- struct nfs_page *req;
int flags;
struct rpc_message msg = {
.rpc_proc = &nfs_procedures[NFSPROC_WRITE],
@@ -619,32 +603,18 @@ nfs_proc_write_setup(struct nfs_write_da
};
/* Note: NFSv2 ignores @stable and always uses NFS_FILE_SYNC */
-
- req = nfs_list_entry(data->pages.next);
- data->args.fh = NFS_FH(inode);
- data->args.offset = req_offset(req);
- data->args.pgbase = req->wb_pgbase;
- data->args.count = count;
data->args.stable = NFS_FILE_SYNC;
- data->args.pages = data->pagevec;
- data->res.fattr = &data->fattr;
- data->res.count = count;
- data->res.verf = &data->verf;
/* Set the initial flags for the task. */
flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
/* Finalize the task. */
rpc_init_task(task, NFS_CLIENT(inode), nfs_write_done, flags);
- task->tk_calldata = data;
- /* Release requests */
- task->tk_release = nfs_writedata_release;
-
- rpc_call_setup(&data->task, &msg, 0);
+ rpc_call_setup(task, &msg, 0);
}
static void
-nfs_proc_commit_setup(struct nfs_write_data *data, u64 start, u32 len, int how)
+nfs_proc_commit_setup(struct nfs_write_data *data, int how)
{
BUG();
}
--- diff/fs/nfs/read.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/nfs/read.c 2004-03-16 11:27:48.421826424 +0000
@@ -35,6 +35,8 @@
#define NFSDBG_FACILITY NFSDBG_PAGECACHE
static int nfs_pagein_one(struct list_head *, struct inode *);
+static void nfs_readpage_result_partial(struct nfs_read_data *, int);
+static void nfs_readpage_result_full(struct nfs_read_data *, int);
static kmem_cache_t *nfs_rdata_cachep;
static mempool_t *nfs_rdata_mempool;
@@ -57,12 +59,37 @@ static __inline__ void nfs_readdata_free
mempool_free(p, nfs_rdata_mempool);
}
-void nfs_readdata_release(struct rpc_task *task)
+static void nfs_readdata_release(struct rpc_task *task)
{
struct nfs_read_data *data = (struct nfs_read_data *)task->tk_calldata;
nfs_readdata_free(data);
}
+static
+unsigned int nfs_page_length(struct inode *inode, struct page *page)
+{
+ loff_t i_size = i_size_read(inode);
+ unsigned long idx;
+
+ if (i_size <= 0)
+ return 0;
+ idx = (i_size - 1) >> PAGE_CACHE_SHIFT;
+ if (page->index > idx)
+ return 0;
+ if (page->index != idx)
+ return PAGE_CACHE_SIZE;
+ return 1 + ((i_size - 1) & (PAGE_CACHE_SIZE - 1));
+}
+
+static
+int nfs_return_empty_page(struct page *page)
+{
+ memclear_highpage_flush(page, 0, PAGE_CACHE_SIZE);
+ SetPageUptodate(page);
+ unlock_page(page);
+ return 0;
+}
+
/*
* Read a page synchronously.
*/
@@ -78,6 +105,7 @@ nfs_readpage_sync(struct file *file, str
.inode = inode,
.args = {
.fh = NFS_FH(inode),
+ .lockowner = current->files,
.pages = &page,
.pgbase = 0UL,
.count = rsize,
@@ -146,89 +174,209 @@ nfs_readpage_async(struct file *file, st
{
LIST_HEAD(one_request);
struct nfs_page *new;
+ unsigned int len;
- new = nfs_create_request(file, inode, page, 0, PAGE_CACHE_SIZE);
+ len = nfs_page_length(inode, page);
+ if (len == 0)
+ return nfs_return_empty_page(page);
+ new = nfs_create_request(file, inode, page, 0, len);
if (IS_ERR(new)) {
unlock_page(page);
return PTR_ERR(new);
}
+ if (len < PAGE_CACHE_SIZE)
+ memclear_highpage_flush(page, len, PAGE_CACHE_SIZE - len);
+
nfs_lock_request(new);
nfs_list_add_request(new, &one_request);
nfs_pagein_one(&one_request, inode);
return 0;
}
+static void nfs_readpage_release(struct nfs_page *req)
+{
+ unlock_page(req->wb_page);
+
+ nfs_clear_request(req);
+ nfs_release_request(req);
+ nfs_unlock_request(req);
+
+ dprintk("NFS: read done (%s/%Ld %d@%Ld)\n",
+ req->wb_inode->i_sb->s_id,
+ (long long)NFS_FILEID(req->wb_inode),
+ req->wb_bytes,
+ (long long)req_offset(req));
+}
+
/*
* Set up the NFS read request struct
*/
-static void
-nfs_read_rpcsetup(struct list_head *head, struct nfs_read_data *data)
+static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data,
+ unsigned int count, unsigned int offset)
{
struct inode *inode;
- struct nfs_page *req;
- struct page **pages;
- unsigned int count;
- pages = data->pagevec;
- count = 0;
- while (!list_empty(head)) {
- req = nfs_list_entry(head->next);
- nfs_list_remove_request(req);
- nfs_list_add_request(req, &data->pages);
- *pages++ = req->wb_page;
- count += req->wb_bytes;
- }
- req = nfs_list_entry(data->pages.next);
+ data->req = req;
data->inode = inode = req->wb_inode;
data->cred = req->wb_cred;
- NFS_PROTO(inode)->read_setup(data, count);
+ data->args.fh = NFS_FH(inode);
+ data->args.offset = req_offset(req) + offset;
+ data->args.pgbase = req->wb_pgbase + offset;
+ data->args.pages = data->pagevec;
+ data->args.count = count;
+ data->args.lockowner = req->wb_lockowner;
+ data->args.state = req->wb_state;
+
+ data->res.fattr = &data->fattr;
+ data->res.count = count;
+ data->res.eof = 0;
+
+ NFS_PROTO(inode)->read_setup(data);
+
+ data->task.tk_cookie = (unsigned long)inode;
+ data->task.tk_calldata = data;
+ /* Release requests */
+ data->task.tk_release = nfs_readdata_release;
- dprintk("NFS: %4d initiated read call (req %s/%Ld, %u bytes @ offset %Lu.\n",
+ dprintk("NFS: %4d initiated read call (req %s/%Ld, %u bytes @ offset %Lu)\n",
data->task.tk_pid,
inode->i_sb->s_id,
(long long)NFS_FILEID(inode),
count,
- (unsigned long long)req_offset(req));
+ data->args.offset);
}
static void
nfs_async_read_error(struct list_head *head)
{
struct nfs_page *req;
- struct page *page;
while (!list_empty(head)) {
req = nfs_list_entry(head->next);
- page = req->wb_page;
nfs_list_remove_request(req);
- SetPageError(page);
- unlock_page(page);
- nfs_clear_request(req);
- nfs_release_request(req);
- nfs_unlock_request(req);
+ SetPageError(req->wb_page);
+ nfs_readpage_release(req);
}
}
-static int
-nfs_pagein_one(struct list_head *head, struct inode *inode)
+/*
+ * Start an async read operation
+ */
+static void nfs_execute_read(struct nfs_read_data *data)
+{
+ struct rpc_clnt *clnt = NFS_CLIENT(data->inode);
+ sigset_t oldset;
+
+ rpc_clnt_sigmask(clnt, &oldset);
+ lock_kernel();
+ rpc_execute(&data->task);
+ unlock_kernel();
+ rpc_clnt_sigunmask(clnt, &oldset);
+}
+
+/*
+ * Generate multiple requests to fill a single page.
+ *
+ * We optimize to reduce the number of read operations on the wire. If we
+ * detect that we're reading a page, or an area of a page, that is past the
+ * end of file, we do not generate NFS read operations but just clear the
+ * parts of the page that would have come back zero from the server anyway.
+ *
+ * We rely on the cached value of i_size to make this determination; another
+ * client can fill pages on the server past our cached end-of-file, but we
+ * won't see the new data until our attribute cache is updated. This is more
+ * or less conventional NFS client behavior.
+ */
+static int nfs_pagein_multi(struct list_head *head, struct inode *inode)
{
- struct rpc_clnt *clnt = NFS_CLIENT(inode);
+ struct nfs_page *req = nfs_list_entry(head->next);
+ struct page *page = req->wb_page;
+ struct nfs_read_data *data;
+ unsigned int rsize = NFS_SERVER(inode)->rsize;
+ unsigned int nbytes, offset;
+ int requests = 0;
+ LIST_HEAD(list);
+
+ nfs_list_remove_request(req);
+
+ nbytes = req->wb_bytes;
+ for(;;) {
+ data = nfs_readdata_alloc();
+ if (!data)
+ goto out_bad;
+ list_add(&data->pages, &list);
+ requests++;
+ if (nbytes <= rsize)
+ break;
+ nbytes -= rsize;
+ }
+ atomic_set(&req->wb_complete, requests);
+
+ ClearPageError(page);
+ offset = 0;
+ nbytes = req->wb_bytes;
+ do {
+ data = list_entry(list.next, struct nfs_read_data, pages);
+ list_del_init(&data->pages);
+
+ data->pagevec[0] = page;
+ data->complete = nfs_readpage_result_partial;
+
+ if (nbytes > rsize) {
+ nfs_read_rpcsetup(req, data, rsize, offset);
+ offset += rsize;
+ nbytes -= rsize;
+ } else {
+ nfs_read_rpcsetup(req, data, nbytes, offset);
+ nbytes = 0;
+ }
+ nfs_execute_read(data);
+ } while (nbytes != 0);
+
+ return 0;
+
+out_bad:
+ while (!list_empty(&list)) {
+ data = list_entry(list.next, struct nfs_read_data, pages);
+ list_del(&data->pages);
+ nfs_readdata_free(data);
+ }
+ SetPageError(page);
+ nfs_readpage_release(req);
+ return -ENOMEM;
+}
+
+static int nfs_pagein_one(struct list_head *head, struct inode *inode)
+{
+ struct nfs_page *req;
+ struct page **pages;
struct nfs_read_data *data;
- sigset_t oldset;
+ unsigned int count;
+
+ if (NFS_SERVER(inode)->rsize < PAGE_CACHE_SIZE)
+ return nfs_pagein_multi(head, inode);
data = nfs_readdata_alloc();
if (!data)
goto out_bad;
- nfs_read_rpcsetup(head, data);
+ pages = data->pagevec;
+ count = 0;
+ while (!list_empty(head)) {
+ req = nfs_list_entry(head->next);
+ nfs_list_remove_request(req);
+ nfs_list_add_request(req, &data->pages);
+ ClearPageError(req->wb_page);
+ *pages++ = req->wb_page;
+ count += req->wb_bytes;
+ }
+ req = nfs_list_entry(data->pages.next);
+
+ data->complete = nfs_readpage_result_full;
+ nfs_read_rpcsetup(req, data, count, 0);
- /* Start the async call */
- rpc_clnt_sigmask(clnt, &oldset);
- lock_kernel();
- rpc_execute(&data->task);
- unlock_kernel();
- rpc_clnt_sigunmask(clnt, &oldset);
+ nfs_execute_read(data);
return 0;
out_bad:
nfs_async_read_error(head);
@@ -258,55 +406,95 @@ nfs_pagein_list(struct list_head *head,
}
/*
+ * Handle a read reply that fills part of a page.
+ */
+static void nfs_readpage_result_partial(struct nfs_read_data *data, int status)
+{
+ struct nfs_page *req = data->req;
+ struct page *page = req->wb_page;
+
+ if (status >= 0) {
+ unsigned int request = data->args.count;
+ unsigned int result = data->res.count;
+
+ if (result < request) {
+ memclear_highpage_flush(page,
+ data->args.pgbase + result,
+ request - result);
+ }
+ } else
+ SetPageError(page);
+
+ if (atomic_dec_and_test(&req->wb_complete)) {
+ if (!PageError(page))
+ SetPageUptodate(page);
+ nfs_readpage_release(req);
+ }
+}
+
+/*
* This is the callback from RPC telling us whether a reply was
* received or some error occurred (timeout or socket shutdown).
*/
-void
-nfs_readpage_result(struct rpc_task *task)
+static void nfs_readpage_result_full(struct nfs_read_data *data, int status)
{
- struct nfs_read_data *data = (struct nfs_read_data *)task->tk_calldata;
unsigned int count = data->res.count;
- dprintk("NFS: %4d nfs_readpage_result, (status %d)\n",
- task->tk_pid, task->tk_status);
-
- NFS_FLAGS(data->inode) |= NFS_INO_INVALID_ATIME;
while (!list_empty(&data->pages)) {
struct nfs_page *req = nfs_list_entry(data->pages.next);
struct page *page = req->wb_page;
nfs_list_remove_request(req);
- if (task->tk_status >= 0) {
+ if (status >= 0) {
if (count < PAGE_CACHE_SIZE) {
- memclear_highpage_flush(page,
+ if (count < req->wb_bytes)
+ memclear_highpage_flush(page,
req->wb_pgbase + count,
req->wb_bytes - count);
-
count = 0;
} else
count -= PAGE_CACHE_SIZE;
SetPageUptodate(page);
} else
SetPageError(page);
- unlock_page(page);
+ nfs_readpage_release(req);
+ }
+}
- dprintk("NFS: read (%s/%Ld %d@%Ld)\n",
- req->wb_inode->i_sb->s_id,
- (long long)NFS_FILEID(req->wb_inode),
- req->wb_bytes,
- (long long)req_offset(req));
- nfs_clear_request(req);
- nfs_release_request(req);
- nfs_unlock_request(req);
+/*
+ * This is the callback from RPC telling us whether a reply was
+ * received or some error occurred (timeout or socket shutdown).
+ */
+void nfs_readpage_result(struct rpc_task *task)
+{
+ struct nfs_read_data *data = (struct nfs_read_data *)task->tk_calldata;
+ struct nfs_readargs *argp = &data->args;
+ struct nfs_readres *resp = &data->res;
+ int status = task->tk_status;
+
+ dprintk("NFS: %4d nfs_readpage_result, (status %d)\n",
+ task->tk_pid, status);
+
+ /* Is this a short read? */
+ if (task->tk_status >= 0 && resp->count < argp->count && !resp->eof) {
+ /* Has the server at least made some progress? */
+ if (resp->count != 0) {
+ /* Yes, so retry the read at the end of the data */
+ argp->offset += resp->count;
+ argp->pgbase += resp->count;
+ argp->count -= resp->count;
+ rpc_restart_call(task);
+ return;
+ }
+ task->tk_status = -EIO;
}
+ NFS_FLAGS(data->inode) |= NFS_INO_INVALID_ATIME;
+ data->complete(data, status);
}
/*
* Read a page over NFS.
- * We read the page synchronously in the following cases:
- * - The NFS rsize is smaller than PAGE_CACHE_SIZE. We could kludge our way
- * around this by creating several consecutive read requests, but
- * that's hardly worth it.
+ * We read the page synchronously in the following case:
* - The error flag is set for this page. This happens only when a
* previous async read operation failed.
*/
@@ -329,7 +517,7 @@ nfs_readpage(struct file *file, struct p
if (error)
goto out_error;
- if (!PageError(page) && NFS_SERVER(inode)->rsize >= PAGE_CACHE_SIZE) {
+ if (!IS_SYNC(inode)) {
error = nfs_readpage_async(file, inode, page);
goto out;
}
@@ -351,26 +539,25 @@ struct nfs_readdesc {
};
static int
-readpage_sync_filler(void *data, struct page *page)
-{
- struct nfs_readdesc *desc = (struct nfs_readdesc *)data;
- return nfs_readpage_sync(desc->filp, page->mapping->host, page);
-}
-
-static int
readpage_async_filler(void *data, struct page *page)
{
struct nfs_readdesc *desc = (struct nfs_readdesc *)data;
struct inode *inode = page->mapping->host;
struct nfs_page *new;
+ unsigned int len;
nfs_wb_page(inode, page);
- new = nfs_create_request(desc->filp, inode, page, 0, PAGE_CACHE_SIZE);
+ len = nfs_page_length(inode, page);
+ if (len == 0)
+ return nfs_return_empty_page(page);
+ new = nfs_create_request(desc->filp, inode, page, 0, len);
if (IS_ERR(new)) {
SetPageError(page);
unlock_page(page);
return PTR_ERR(new);
}
+ if (len < PAGE_CACHE_SIZE)
+ memclear_highpage_flush(page, len, PAGE_CACHE_SIZE - len);
nfs_lock_request(new);
nfs_list_add_request(new, desc->head);
return 0;
@@ -385,14 +572,16 @@ nfs_readpages(struct file *filp, struct
.filp = filp,
.head = &head,
};
- struct nfs_server *server = NFS_SERVER(mapping->host);
- int is_sync = server->rsize < PAGE_CACHE_SIZE;
+ struct inode *inode = mapping->host;
+ struct nfs_server *server = NFS_SERVER(inode);
int ret;
- ret = read_cache_pages(mapping, pages,
- is_sync ? readpage_sync_filler :
- readpage_async_filler,
- &desc);
+ dprintk("NFS: nfs_readpages (%s/%Ld %d)\n",
+ inode->i_sb->s_id,
+ (long long)NFS_FILEID(inode),
+ nr_pages);
+
+ ret = read_cache_pages(mapping, pages, readpage_async_filler, &desc);
if (!list_empty(&head)) {
int err = nfs_pagein_list(&head, server->rpages);
if (!ret)
--- diff/fs/nfs/write.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/nfs/write.c 2004-03-16 11:27:48.424825968 +0000
@@ -74,11 +74,17 @@
static struct nfs_page * nfs_update_request(struct file*, struct inode *,
struct page *,
unsigned int, unsigned int);
+static void nfs_writeback_done_partial(struct nfs_write_data *, int);
+static void nfs_writeback_done_full(struct nfs_write_data *, int);
+static int nfs_wait_on_write_congestion(struct address_space *, int);
+static int nfs_wait_on_requests(struct inode *, unsigned long, unsigned int);
static kmem_cache_t *nfs_wdata_cachep;
static mempool_t *nfs_wdata_mempool;
static mempool_t *nfs_commit_mempool;
+static DECLARE_WAIT_QUEUE_HEAD(nfs_write_congestion);
+
static __inline__ struct nfs_write_data *nfs_writedata_alloc(void)
{
struct nfs_write_data *p;
@@ -95,7 +101,7 @@ static __inline__ void nfs_writedata_fre
mempool_free(p, nfs_wdata_mempool);
}
-void nfs_writedata_release(struct rpc_task *task)
+static void nfs_writedata_release(struct rpc_task *task)
{
struct nfs_write_data *wdata = (struct nfs_write_data *)task->tk_calldata;
nfs_writedata_free(wdata);
@@ -117,12 +123,6 @@ static __inline__ void nfs_commit_free(s
mempool_free(p, nfs_commit_mempool);
}
-void nfs_commit_release(struct rpc_task *task)
-{
- struct nfs_write_data *wdata = (struct nfs_write_data *)task->tk_calldata;
- nfs_commit_free(wdata);
-}
-
/* Adjust the file length if we're writing beyond the end */
static void nfs_grow_file(struct page *page, unsigned int offset, unsigned int count)
{
@@ -173,19 +173,19 @@ static void nfs_mark_uptodate(struct pag
* Write a page synchronously.
* Offset is the data offset within the page.
*/
-static int
-nfs_writepage_sync(struct file *file, struct inode *inode, struct page *page,
- unsigned int offset, unsigned int count)
+static int nfs_writepage_sync(struct file *file, struct inode *inode,
+ struct page *page, unsigned int offset, unsigned int count,
+ int how)
{
unsigned int wsize = NFS_SERVER(inode)->wsize;
int result, written = 0;
- int swapfile = IS_SWAPFILE(inode);
struct nfs_write_data wdata = {
- .flags = swapfile ? NFS_RPC_SWAPFLAGS : 0,
+ .flags = how,
.cred = NULL,
.inode = inode,
.args = {
.fh = NFS_FH(inode),
+ .lockowner = current->files,
.pages = &page,
.stable = NFS_FILE_SYNC,
.pgbase = offset,
@@ -204,7 +204,7 @@ nfs_writepage_sync(struct file *file, st
nfs_begin_data_update(inode);
do {
- if (count < wsize && !swapfile)
+ if (count < wsize)
wdata.args.count = count;
wdata.args.offset = page_offset(page) + wdata.args.pgbase;
@@ -259,17 +259,26 @@ static int nfs_writepage_async(struct fi
return status;
}
+static int wb_priority(struct writeback_control *wbc)
+{
+ if (wbc->for_reclaim)
+ return FLUSH_HIGHPRI;
+ if (wbc->for_kupdate)
+ return FLUSH_LOWPRI;
+ return 0;
+}
+
/*
* Write an mmapped page to the server.
*/
-int
-nfs_writepage(struct page *page, struct writeback_control *wbc)
+int nfs_writepage(struct page *page, struct writeback_control *wbc)
{
struct inode *inode = page->mapping->host;
unsigned long end_index;
unsigned offset = PAGE_CACHE_SIZE;
loff_t i_size = i_size_read(inode);
int inode_referenced = 0;
+ int priority = wb_priority(wbc);
int err;
/*
@@ -285,7 +294,7 @@ nfs_writepage(struct page *page, struct
end_index = i_size >> PAGE_CACHE_SHIFT;
/* Ensure we've flushed out any previous writes */
- nfs_wb_page(inode,page);
+ nfs_wb_page_priority(inode, page, priority);
/* easy case */
if (page->index < end_index)
@@ -299,44 +308,60 @@ nfs_writepage(struct page *page, struct
goto out;
do_it:
lock_kernel();
- if (NFS_SERVER(inode)->wsize >= PAGE_CACHE_SIZE && !IS_SYNC(inode) &&
- inode_referenced) {
+ if (!IS_SYNC(inode) && inode_referenced) {
err = nfs_writepage_async(NULL, inode, page, 0, offset);
- if (err >= 0)
+ if (err >= 0) {
err = 0;
+ if (wbc->for_reclaim)
+ err = WRITEPAGE_ACTIVATE;
+ }
} else {
- err = nfs_writepage_sync(NULL, inode, page, 0, offset);
+ err = nfs_writepage_sync(NULL, inode, page, 0, offset, priority);
if (err == offset)
err = 0;
}
unlock_kernel();
out:
- unlock_page(page);
+ if (err != WRITEPAGE_ACTIVATE)
+ unlock_page(page);
if (inode_referenced)
iput(inode);
return err;
}
-int
-nfs_writepages(struct address_space *mapping, struct writeback_control *wbc)
+/*
+ * Note: causes nfs_update_request() to block on the assumption
+ * that the writeback is generated due to memory pressure.
+ */
+int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc)
{
+ struct backing_dev_info *bdi = mapping->backing_dev_info;
struct inode *inode = mapping->host;
- int is_sync = !wbc->nonblocking;
int err;
err = generic_writepages(mapping, wbc);
if (err)
- goto out;
- err = nfs_flush_inode(inode, 0, 0, 0);
+ return err;
+ while (test_and_set_bit(BDI_write_congested, &bdi->state) != 0) {
+ if (wbc->nonblocking)
+ return 0;
+ nfs_wait_on_write_congestion(mapping, 0);
+ }
+ err = nfs_flush_inode(inode, 0, 0, wb_priority(wbc));
if (err < 0)
goto out;
- if (wbc->sync_mode == WB_SYNC_HOLD)
- goto out;
- if (is_sync && wbc->sync_mode == WB_SYNC_ALL) {
- err = nfs_wb_all(inode);
- } else
- nfs_commit_inode(inode, 0, 0, 0);
+ wbc->nr_to_write -= err;
+ if (!wbc->nonblocking && wbc->sync_mode == WB_SYNC_ALL) {
+ err = nfs_wait_on_requests(inode, 0, 0);
+ if (err < 0)
+ goto out;
+ }
+ err = nfs_commit_inode(inode, 0, 0, wb_priority(wbc));
+ if (err > 0)
+ wbc->nr_to_write -= err;
out:
+ clear_bit(BDI_write_congested, &bdi->state);
+ wake_up_all(&nfs_write_congestion);
return err;
}
@@ -365,7 +390,7 @@ nfs_inode_add_request(struct inode *inod
/*
* Insert a write request into an inode
*/
-static inline void
+static void
nfs_inode_remove_request(struct nfs_page *req)
{
struct nfs_inode *nfsi;
@@ -416,7 +441,7 @@ nfs_find_request(struct inode *inode, un
/*
* Add a request to the inode's dirty list.
*/
-static inline void
+static void
nfs_mark_request_dirty(struct nfs_page *req)
{
struct inode *inode = req->wb_inode;
@@ -444,7 +469,7 @@ nfs_dirty_request(struct nfs_page *req)
/*
* Add a request to the inode's commit list.
*/
-static inline void
+static void
nfs_mark_request_commit(struct nfs_page *req)
{
struct inode *inode = req->wb_inode;
@@ -548,6 +573,38 @@ nfs_scan_commit(struct inode *inode, str
}
#endif
+static int nfs_wait_on_write_congestion(struct address_space *mapping, int intr)
+{
+ struct backing_dev_info *bdi = mapping->backing_dev_info;
+ DEFINE_WAIT(wait);
+ int ret = 0;
+
+ might_sleep();
+
+ if (!bdi_write_congested(bdi))
+ return 0;
+ if (intr) {
+ struct rpc_clnt *clnt = NFS_CLIENT(mapping->host);
+ sigset_t oldset;
+
+ rpc_clnt_sigmask(clnt, &oldset);
+ prepare_to_wait(&nfs_write_congestion, &wait, TASK_INTERRUPTIBLE);
+ if (bdi_write_congested(bdi)) {
+ if (signalled())
+ ret = -ERESTARTSYS;
+ else
+ schedule();
+ }
+ rpc_clnt_sigunmask(clnt, &oldset);
+ } else {
+ prepare_to_wait(&nfs_write_congestion, &wait, TASK_UNINTERRUPTIBLE);
+ if (bdi_write_congested(bdi))
+ schedule();
+ }
+ finish_wait(&nfs_write_congestion, &wait);
+ return ret;
+}
+
/*
* Try to update any existing write request, or create one if there is none.
@@ -560,11 +617,14 @@ static struct nfs_page *
nfs_update_request(struct file* file, struct inode *inode, struct page *page,
unsigned int offset, unsigned int bytes)
{
+ struct nfs_server *server = NFS_SERVER(inode);
struct nfs_page *req, *new = NULL;
unsigned long rqend, end;
end = offset + bytes;
+ if (nfs_wait_on_write_congestion(page->mapping, server->flags & NFS_MOUNT_INTR))
+ return ERR_PTR(-ERESTARTSYS);
for (;;) {
/* Loop over all inode entries and see if we find
* A request for the page we wish to update
@@ -668,8 +728,8 @@ nfs_flush_incompatible(struct file *file
* XXX: Keep an eye on generic_file_read to make sure it doesn't do bad
* things with a page scheduled for an RPC call (e.g. invalidate it).
*/
-int
-nfs_updatepage(struct file *file, struct page *page, unsigned int offset, unsigned int count)
+int nfs_updatepage(struct file *file, struct page *page,
+ unsigned int offset, unsigned int count)
{
struct dentry *dentry = file->f_dentry;
struct inode *inode = page->mapping->host;
@@ -680,12 +740,8 @@ nfs_updatepage(struct file *file, struct
dentry->d_parent->d_name.name, dentry->d_name.name,
count, (long long)(page_offset(page) +offset));
- /*
- * If wsize is smaller than page size, update and write
- * page synchronously.
- */
- if (NFS_SERVER(inode)->wsize < PAGE_CACHE_SIZE || IS_SYNC(inode)) {
- status = nfs_writepage_sync(file, inode, page, offset, count);
+ if (IS_SYNC(inode)) {
+ status = nfs_writepage_sync(file, inode, page, offset, count, 0);
if (status > 0) {
if (offset == 0 && status == PAGE_CACHE_SIZE)
SetPageUptodate(page);
@@ -747,43 +803,159 @@ done:
return status;
}
+static void nfs_writepage_release(struct nfs_page *req)
+{
+ end_page_writeback(req->wb_page);
+
+#if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
+ if (!PageError(req->wb_page)) {
+ if (NFS_NEED_RESCHED(req)) {
+ nfs_mark_request_dirty(req);
+ goto out;
+ } else if (NFS_NEED_COMMIT(req)) {
+ nfs_mark_request_commit(req);
+ goto out;
+ }
+ }
+ nfs_inode_remove_request(req);
+
+out:
+ nfs_clear_commit(req);
+ nfs_clear_reschedule(req);
+#else
+ nfs_inode_remove_request(req);
+#endif
+ nfs_unlock_request(req);
+}
+
+static inline int flush_task_priority(int how)
+{
+ switch (how & (FLUSH_HIGHPRI|FLUSH_LOWPRI)) {
+ case FLUSH_HIGHPRI:
+ return RPC_PRIORITY_HIGH;
+ case FLUSH_LOWPRI:
+ return RPC_PRIORITY_LOW;
+ }
+ return RPC_PRIORITY_NORMAL;
+}
+
/*
* Set up the argument/result storage required for the RPC call.
*/
-static void
-nfs_write_rpcsetup(struct list_head *head, struct nfs_write_data *data, int how)
+static void nfs_write_rpcsetup(struct nfs_page *req,
+ struct nfs_write_data *data,
+ unsigned int count, unsigned int offset,
+ int how)
{
struct rpc_task *task = &data->task;
struct inode *inode;
- struct nfs_page *req;
- struct page **pages;
- unsigned int count;
/* Set up the RPC argument and reply structs
* NB: take care not to mess about with data->commit et al. */
- pages = data->pagevec;
- count = 0;
- while (!list_empty(head)) {
- req = nfs_list_entry(head->next);
- nfs_list_remove_request(req);
- nfs_list_add_request(req, &data->pages);
- SetPageWriteback(req->wb_page);
- *pages++ = req->wb_page;
- count += req->wb_bytes;
- }
- req = nfs_list_entry(data->pages.next);
+ data->req = req;
data->inode = inode = req->wb_inode;
data->cred = req->wb_cred;
- NFS_PROTO(inode)->write_setup(data, count, how);
+ data->args.fh = NFS_FH(inode);
+ data->args.offset = req_offset(req) + offset;
+ data->args.pgbase = req->wb_pgbase + offset;
+ data->args.count = count;
+ data->args.pages = data->pagevec;
+ data->res.fattr = &data->fattr;
+ data->res.count = count;
+ data->res.verf = &data->verf;
+
+ NFS_PROTO(inode)->write_setup(data, how);
+
+ data->task.tk_priority = flush_task_priority(how);
+ data->task.tk_cookie = (unsigned long)inode;
+ data->task.tk_calldata = data;
+ /* Release requests */
+ data->task.tk_release = nfs_writedata_release;
dprintk("NFS: %4d initiated write call (req %s/%Ld, %u bytes @ offset %Lu)\n",
task->tk_pid,
inode->i_sb->s_id,
(long long)NFS_FILEID(inode),
count,
- (unsigned long long)req_offset(req));
+ data->args.offset);
+}
+
+static void nfs_execute_write(struct nfs_write_data *data)
+{
+ struct rpc_clnt *clnt = NFS_CLIENT(data->inode);
+ sigset_t oldset;
+
+ rpc_clnt_sigmask(clnt, &oldset);
+ lock_kernel();
+ rpc_execute(&data->task);
+ unlock_kernel();
+ rpc_clnt_sigunmask(clnt, &oldset);
+}
+
+/*
+ * Generate multiple small requests to write out a single
+ * contiguous dirty area on one page.
+ */
+static int nfs_flush_multi(struct list_head *head, struct inode *inode, int how)
+{
+ struct nfs_page *req = nfs_list_entry(head->next);
+ struct page *page = req->wb_page;
+ struct nfs_write_data *data;
+ unsigned int wsize = NFS_SERVER(inode)->wsize;
+ unsigned int nbytes, offset;
+ int requests = 0;
+ LIST_HEAD(list);
+
+ nfs_list_remove_request(req);
+
+ nbytes = req->wb_bytes;
+ for (;;) {
+ data = nfs_writedata_alloc();
+ if (!data)
+ goto out_bad;
+ list_add(&data->pages, &list);
+ requests++;
+ if (nbytes <= wsize)
+ break;
+ nbytes -= wsize;
+ }
+ atomic_set(&req->wb_complete, requests);
+
+ ClearPageError(page);
+ SetPageWriteback(page);
+ offset = 0;
+ nbytes = req->wb_bytes;
+ do {
+ data = list_entry(list.next, struct nfs_write_data, pages);
+ list_del_init(&data->pages);
+
+ data->pagevec[0] = page;
+ data->complete = nfs_writeback_done_partial;
+
+ if (nbytes > wsize) {
+ nfs_write_rpcsetup(req, data, wsize, offset, how);
+ offset += wsize;
+ nbytes -= wsize;
+ } else {
+ nfs_write_rpcsetup(req, data, nbytes, offset, how);
+ nbytes = 0;
+ }
+ nfs_execute_write(data);
+ } while (nbytes != 0);
+
+ return 0;
+
+out_bad:
+ while (!list_empty(&list)) {
+ data = list_entry(list.next, struct nfs_write_data, pages);
+ list_del(&data->pages);
+ nfs_writedata_free(data);
+ }
+ nfs_mark_request_dirty(req);
+ nfs_unlock_request(req);
+ return -ENOMEM;
}
/*
@@ -794,25 +966,38 @@ nfs_write_rpcsetup(struct list_head *hea
* This is the case if nfs_updatepage detects a conflicting request
* that has been written but not committed.
*/
-static int
-nfs_flush_one(struct list_head *head, struct inode *inode, int how)
+static int nfs_flush_one(struct list_head *head, struct inode *inode, int how)
{
- struct rpc_clnt *clnt = NFS_CLIENT(inode);
+ struct nfs_page *req;
+ struct page **pages;
struct nfs_write_data *data;
- sigset_t oldset;
+ unsigned int count;
+
+ if (NFS_SERVER(inode)->wsize < PAGE_CACHE_SIZE)
+ return nfs_flush_multi(head, inode, how);
data = nfs_writedata_alloc();
if (!data)
goto out_bad;
+ pages = data->pagevec;
+ count = 0;
+ while (!list_empty(head)) {
+ req = nfs_list_entry(head->next);
+ nfs_list_remove_request(req);
+ nfs_list_add_request(req, &data->pages);
+ ClearPageError(req->wb_page);
+ SetPageWriteback(req->wb_page);
+ *pages++ = req->wb_page;
+ count += req->wb_bytes;
+ }
+ req = nfs_list_entry(data->pages.next);
+
+ data->complete = nfs_writeback_done_full;
/* Set up the argument struct */
- nfs_write_rpcsetup(head, data, how);
+ nfs_write_rpcsetup(req, data, count, 0, how);
- rpc_clnt_sigmask(clnt, &oldset);
- lock_kernel();
- rpc_execute(&data->task);
- unlock_kernel();
- rpc_clnt_sigunmask(clnt, &oldset);
+ nfs_execute_write(data);
return 0;
out_bad:
while (!list_empty(head)) {
@@ -851,59 +1036,59 @@ nfs_flush_list(struct list_head *head, i
return error;
}
-
/*
- * This function is called when the WRITE call is complete.
+ * Handle a write reply that flushed part of a page.
*/
-void
-nfs_writeback_done(struct rpc_task *task)
+static void nfs_writeback_done_partial(struct nfs_write_data *data, int status)
{
- struct nfs_write_data *data = (struct nfs_write_data *) task->tk_calldata;
- struct nfs_writeargs *argp = &data->args;
- struct nfs_writeres *resp = &data->res;
- struct nfs_page *req;
- struct page *page;
+ struct nfs_page *req = data->req;
+ struct page *page = req->wb_page;
- dprintk("NFS: %4d nfs_writeback_done (status %d)\n",
- task->tk_pid, task->tk_status);
+ dprintk("NFS: write (%s/%Ld %d@%Ld)",
+ req->wb_inode->i_sb->s_id,
+ (long long)NFS_FILEID(req->wb_inode),
+ req->wb_bytes,
+ (long long)req_offset(req));
- /* We can't handle that yet but we check for it nevertheless */
- if (resp->count < argp->count && task->tk_status >= 0) {
- static unsigned long complain;
- if (time_before(complain, jiffies)) {
- printk(KERN_WARNING
- "NFS: Server wrote less than requested.\n");
- complain = jiffies + 300 * HZ;
- }
- /* Can't do anything about it right now except throw
- * an error. */
- task->tk_status = -EIO;
- }
+ if (status < 0) {
+ ClearPageUptodate(page);
+ SetPageError(page);
+ if (req->wb_file)
+ req->wb_file->f_error = status;
+ dprintk(", error = %d\n", status);
+ } else {
#if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
- if (data->verf.committed < argp->stable && task->tk_status >= 0) {
- /* We tried a write call, but the server did not
- * commit data to stable storage even though we
- * requested it.
- * Note: There is a known bug in Tru64 < 5.0 in which
- * the server reports NFS_DATA_SYNC, but performs
- * NFS_FILE_SYNC. We therefore implement this checking
- * as a dprintk() in order to avoid filling syslog.
- */
- static unsigned long complain;
-
- if (time_before(complain, jiffies)) {
- dprintk("NFS: faulty NFS server %s:"
- " (committed = %d) != (stable = %d)\n",
- NFS_SERVER(data->inode)->hostname,
- data->verf.committed, argp->stable);
- complain = jiffies + 300 * HZ;
- }
- }
+ if (data->verf.committed < NFS_FILE_SYNC) {
+ if (!NFS_NEED_COMMIT(req)) {
+ nfs_defer_commit(req);
+ memcpy(&req->wb_verf, &data->verf, sizeof(req->wb_verf));
+ dprintk(" defer commit\n");
+ } else if (memcmp(&req->wb_verf, &data->verf, sizeof(req->wb_verf))) {
+ nfs_defer_reschedule(req);
+ dprintk(" server reboot detected\n");
+ }
+ } else
#endif
+ dprintk(" OK\n");
+ }
- /*
- * Process the nfs_page list
- */
+ if (atomic_dec_and_test(&req->wb_complete))
+ nfs_writepage_release(req);
+}
+
+/*
+ * Handle a write reply that flushes a whole page.
+ *
+ * FIXME: There is an inherent race with invalidate_inode_pages and
+ * writebacks since the page->count is kept > 1 for as long
+ * as the page has a write request pending.
+ */
+static void nfs_writeback_done_full(struct nfs_write_data *data, int status)
+{
+ struct nfs_page *req;
+ struct page *page;
+
+ /* Update attributes as result of writeback. */
while (!list_empty(&data->pages)) {
req = nfs_list_entry(data->pages.next);
nfs_list_remove_request(req);
@@ -915,20 +1100,20 @@ nfs_writeback_done(struct rpc_task *task
req->wb_bytes,
(long long)req_offset(req));
- if (task->tk_status < 0) {
+ if (status < 0) {
ClearPageUptodate(page);
SetPageError(page);
if (req->wb_file)
- req->wb_file->f_error = task->tk_status;
+ req->wb_file->f_error = status;
end_page_writeback(page);
nfs_inode_remove_request(req);
- dprintk(", error = %d\n", task->tk_status);
+ dprintk(", error = %d\n", status);
goto next;
}
end_page_writeback(page);
#if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
- if (argp->stable != NFS_UNSTABLE || data->verf.committed == NFS_FILE_SYNC) {
+ if (data->args.stable != NFS_UNSTABLE || data->verf.committed == NFS_FILE_SYNC) {
nfs_inode_remove_request(req);
dprintk(" OK\n");
goto next;
@@ -944,13 +1129,88 @@ nfs_writeback_done(struct rpc_task *task
}
}
+/*
+ * This function is called when the WRITE call is complete.
+ */
+void nfs_writeback_done(struct rpc_task *task)
+{
+ struct nfs_write_data *data = (struct nfs_write_data *) task->tk_calldata;
+ struct nfs_writeargs *argp = &data->args;
+ struct nfs_writeres *resp = &data->res;
+
+ dprintk("NFS: %4d nfs_writeback_done (status %d)\n",
+ task->tk_pid, task->tk_status);
+
+#if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
+ if (resp->verf->committed < argp->stable && task->tk_status >= 0) {
+ /* We tried a write call, but the server did not
+ * commit data to stable storage even though we
+ * requested it.
+ * Note: There is a known bug in Tru64 < 5.0 in which
+ * the server reports NFS_DATA_SYNC, but performs
+ * NFS_FILE_SYNC. We therefore implement this checking
+ * as a dprintk() in order to avoid filling syslog.
+ */
+ static unsigned long complain;
+
+ if (time_before(complain, jiffies)) {
+ dprintk("NFS: faulty NFS server %s:"
+ " (committed = %d) != (stable = %d)\n",
+ NFS_SERVER(data->inode)->hostname,
+ resp->verf->committed, argp->stable);
+ complain = jiffies + 300 * HZ;
+ }
+ }
+#endif
+ /* Is this a short write? */
+ if (task->tk_status >= 0 && resp->count < argp->count) {
+ static unsigned long complain;
+
+ /* Has the server at least made some progress? */
+ if (resp->count != 0) {
+ /* Was this an NFSv2 write or an NFSv3 stable write? */
+ if (resp->verf->committed != NFS_UNSTABLE) {
+ /* Resend from where the server left off */
+ argp->offset += resp->count;
+ argp->pgbase += resp->count;
+ argp->count -= resp->count;
+ } else {
+ /* Resend as a stable write in order to avoid
+ * headaches in the case of a server crash.
+ */
+ argp->stable = NFS_FILE_SYNC;
+ }
+ rpc_restart_call(task);
+ return;
+ }
+ if (time_before(complain, jiffies)) {
+ printk(KERN_WARNING
+ "NFS: Server wrote less than requested.\n");
+ complain = jiffies + 300 * HZ;
+ }
+ /* Can't do anything about it except throw an error. */
+ task->tk_status = -EIO;
+ }
+
+ /*
+ * Process the nfs_page list
+ */
+ data->complete(data, task->tk_status);
+}
+
#if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
+static void nfs_commit_release(struct rpc_task *task)
+{
+ struct nfs_write_data *wdata = (struct nfs_write_data *)task->tk_calldata;
+ nfs_commit_free(wdata);
+}
+
/*
* Set up the argument/result storage required for the RPC call.
*/
-static void
-nfs_commit_rpcsetup(struct list_head *head, struct nfs_write_data *data, int how)
+static void nfs_commit_rpcsetup(struct list_head *head,
+ struct nfs_write_data *data, int how)
{
struct rpc_task *task = &data->task;
struct nfs_page *first, *last;
@@ -979,7 +1239,20 @@ nfs_commit_rpcsetup(struct list_head *he
data->inode = inode;
data->cred = first->wb_cred;
- NFS_PROTO(inode)->commit_setup(data, start, len, how);
+ data->args.fh = NFS_FH(data->inode);
+ data->args.offset = start;
+ data->args.count = len;
+ data->res.count = len;
+ data->res.fattr = &data->fattr;
+ data->res.verf = &data->verf;
+
+ NFS_PROTO(inode)->commit_setup(data, how);
+
+ data->task.tk_priority = flush_task_priority(how);
+ data->task.tk_cookie = (unsigned long)inode;
+ data->task.tk_calldata = data;
+ /* Release requests */
+ data->task.tk_release = nfs_commit_release;
dprintk("NFS: %4d initiated commit call\n", task->tk_pid);
}
@@ -990,10 +1263,8 @@ nfs_commit_rpcsetup(struct list_head *he
int
nfs_commit_list(struct list_head *head, int how)
{
- struct rpc_clnt *clnt;
struct nfs_write_data *data;
struct nfs_page *req;
- sigset_t oldset;
data = nfs_commit_alloc();
@@ -1002,13 +1273,8 @@ nfs_commit_list(struct list_head *head,
/* Set up the argument struct */
nfs_commit_rpcsetup(head, data, how);
- clnt = NFS_CLIENT(data->inode);
- rpc_clnt_sigmask(clnt, &oldset);
- lock_kernel();
- rpc_execute(&data->task);
- unlock_kernel();
- rpc_clnt_sigunmask(clnt, &oldset);
+ nfs_execute_write(data);
return 0;
out_bad:
while (!list_empty(head)) {
--- diff/fs/ntfs/compress.c 2003-08-26 09:00:54.000000000 +0000
+++ source/fs/ntfs/compress.c 2004-03-16 11:27:48.425825816 +0000
@@ -23,6 +23,7 @@
#include <linux/fs.h>
#include <linux/buffer_head.h>
+#include <linux/blkdev.h>
#include "ntfs.h"
@@ -668,7 +669,7 @@ lock_retry_remap:
"uptodate! Unplugging the disk queue "
"and rescheduling.");
get_bh(tbh);
- blk_run_queues();
+ blk_run_address_space(mapping);
schedule();
put_bh(tbh);
if (unlikely(!buffer_uptodate(tbh)))
--- diff/fs/open.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/open.c 2004-03-16 11:27:48.426825664 +0000
@@ -192,7 +192,9 @@ int do_truncate(struct dentry *dentry, l
newattrs.ia_size = length;
newattrs.ia_valid = ATTR_SIZE | ATTR_CTIME;
down(&dentry->d_inode->i_sem);
+ down_write(&dentry->d_inode->i_alloc_sem);
err = notify_change(dentry, &newattrs);
+ up_write(&dentry->d_inode->i_alloc_sem);
up(&dentry->d_inode->i_sem);
return err;
}
@@ -1037,7 +1039,7 @@ EXPORT_SYMBOL(sys_close);
asmlinkage long sys_vhangup(void)
{
if (capable(CAP_SYS_TTY_CONFIG)) {
- tty_vhangup(current->tty);
+ tty_vhangup(current->signal->tty);
return 0;
}
return -EPERM;
--- diff/fs/proc/array.c 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/proc/array.c 2004-03-16 11:27:48.427825512 +0000
@@ -168,7 +168,7 @@ static inline char * task_state(struct t
p->pid && p->ptrace ? p->parent->pid : 0,
p->uid, p->euid, p->suid, p->fsuid,
p->gid, p->egid, p->sgid, p->fsgid);
- read_unlock(&tasklist_lock);
+ read_unlock(&tasklist_lock);
task_lock(p);
buffer += sprintf(buffer,
"FDSize:\t%d\n"
@@ -301,7 +301,7 @@ int proc_pid_stat(struct task_struct *ta
sigset_t sigign, sigcatch;
char state;
int res;
- pid_t ppid;
+ pid_t ppid, pgid = -1, sid = -1;
int num_threads = 0;
struct mm_struct *mm;
@@ -311,10 +311,6 @@ int proc_pid_stat(struct task_struct *ta
mm = task->mm;
if(mm)
mm = mmgrab(mm);
- if (task->tty) {
- tty_pgrp = task->tty->pgrp;
- tty_nr = new_encode_dev(tty_devnum(task->tty));
- }
task_unlock(task);
if (mm) {
down_read(&mm->mmap_sem);
@@ -335,7 +331,15 @@ int proc_pid_stat(struct task_struct *ta
collect_sigign_sigcatch(task, &sigign, &sigcatch);
spin_unlock_irq(&task->sighand->siglock);
}
- read_unlock(&tasklist_lock);
+ if (task->signal) {
+ if (task->signal->tty) {
+ tty_pgrp = task->signal->tty->pgrp;
+ tty_nr = new_encode_dev(tty_devnum(task->signal->tty));
+ }
+ pgid = process_group(task);
+ sid = task->signal->session;
+ }
+ read_unlock(&tasklist_lock);
/* scale priority and nice values from timeslices to -20..20 */
/* to make it look like a "normal" Unix priority/nice value */
@@ -352,8 +356,8 @@ int proc_pid_stat(struct task_struct *ta
task->comm,
state,
ppid,
- process_group(task),
- task->session,
+ pgid,
+ sid,
tty_nr,
tty_pgrp,
task->flags,
--- diff/fs/proc/proc_misc.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/proc/proc_misc.c 2004-03-16 11:27:48.428825360 +0000
@@ -651,6 +651,36 @@ static void create_seq_entry(char *name,
entry->proc_fops = f;
}
+#ifdef CONFIG_LOCKMETER
+extern ssize_t get_lockmeter_info(char *, size_t, loff_t *);
+extern ssize_t put_lockmeter_info(const char *, size_t);
+extern int get_lockmeter_info_size(void);
+
+/*
+ * This function accesses lock metering information.
+ */
+static ssize_t read_lockmeter(struct file *file, char *buf,
+ size_t count, loff_t *ppos)
+{
+ return get_lockmeter_info(buf, count, ppos);
+}
+
+/*
+ * Writing to /proc/lockmeter resets the counters
+ */
+static ssize_t write_lockmeter(struct file * file, const char * buf,
+ size_t count, loff_t *ppos)
+{
+ return put_lockmeter_info(buf, count);
+}
+
+static struct file_operations proc_lockmeter_operations = {
+ NULL, /* lseek */
+ read: read_lockmeter,
+ write: write_lockmeter,
+};
+#endif /* CONFIG_LOCKMETER */
+
void __init proc_misc_init(void)
{
struct proc_dir_entry *entry;
@@ -718,6 +748,13 @@ void __init proc_misc_init(void)
if (entry)
entry->proc_fops = &proc_sysrq_trigger_operations;
#endif
+#ifdef CONFIG_LOCKMETER
+ entry = create_proc_entry("lockmeter", S_IWUSR | S_IRUGO, NULL);
+ if (entry) {
+ entry->proc_fops = &proc_lockmeter_operations;
+ entry->size = get_lockmeter_info_size();
+ }
+#endif
#ifdef CONFIG_PPC32
{
extern struct file_operations ppc_htab_operations;
--- diff/fs/proc/root.c 2003-08-20 13:16:33.000000000 +0000
+++ source/fs/proc/root.c 2004-03-16 11:27:48.429825208 +0000
@@ -74,9 +74,6 @@ void __init proc_root_init(void)
#ifdef CONFIG_PROC_DEVICETREE
proc_device_tree_init();
#endif
-#ifdef CONFIG_PPC_RTAS
- proc_rtas_init();
-#endif
proc_bus = proc_mkdir("bus", 0);
}
--- diff/fs/proc/task_mmu.c 2004-02-09 10:36:12.000000000 +0000
+++ source/fs/proc/task_mmu.c 2004-03-16 11:27:48.429825208 +0000
@@ -4,6 +4,22 @@
#include <asm/elf.h>
#include <asm/uaccess.h>
+#ifdef CONFIG_NUMA
+char *task_mem_pernode(struct mm_struct *mm, char *buffer)
+{
+ int nid;
+
+ for (nid = 0; nid < MAX_NUMNODES; nid++){
+ buffer += sprintf(buffer, "VmRSS-node_%d:\t%8lu kb\n",
+ nid, mm->pernode_rss[nid] << (PAGE_SHIFT-10));
+ }
+
+ return buffer;
+}
+#else /* !CONFIG_NUMA */
+#define task_mem_pernode(mm, buffer) (buffer)
+#endif /* CONFIG_NUMA */
+
char *task_mem(struct mm_struct *mm, char *buffer)
{
unsigned long data = 0, stack = 0, exec = 0, lib = 0;
@@ -40,6 +56,7 @@ char *task_mem(struct mm_struct *mm, cha
mm->rss << (PAGE_SHIFT-10),
data - stack, stack,
exec - lib, lib);
+ buffer = task_mem_pernode(mm, buffer);
up_read(&mm->mmap_sem);
return buffer;
}
--- diff/fs/quota_v1.c 2003-10-09 08:47:34.000000000 +0000
+++ source/fs/quota_v1.c 2004-03-16 11:27:48.430825056 +0000
@@ -60,7 +60,7 @@ static int v1_read_dqblk(struct dquot *d
v1_disk2mem_dqblk(&dquot->dq_dqb, &dqblk);
if (dquot->dq_dqb.dqb_bhardlimit == 0 && dquot->dq_dqb.dqb_bsoftlimit == 0 &&
dquot->dq_dqb.dqb_ihardlimit == 0 && dquot->dq_dqb.dqb_isoftlimit == 0)
- dquot->dq_flags |= DQ_FAKE;
+ set_bit(DQ_FAKE_B, &dquot->dq_flags);
dqstats.reads++;
return 0;
@@ -80,12 +80,7 @@ static int v1_commit_dqblk(struct dquot
fs = get_fs();
set_fs(KERNEL_DS);
- /*
- * Note: clear the DQ_MOD flag unconditionally,
- * so we don't loop forever on failure.
- */
v1_mem2disk_dqblk(&dqblk, &dquot->dq_dqb);
- dquot->dq_flags &= ~DQ_MOD;
if (dquot->dq_id == 0) {
dqblk.dqb_btime = sb_dqopt(dquot->dq_sb)->info[type].dqi_bgrace;
dqblk.dqb_itime = sb_dqopt(dquot->dq_sb)->info[type].dqi_igrace;
--- diff/fs/quota_v2.c 2003-05-21 10:50:00.000000000 +0000
+++ source/fs/quota_v2.c 2004-03-16 11:27:48.431824904 +0000
@@ -65,7 +65,7 @@ static int v2_read_file_info(struct supe
set_fs(fs);
if (size != sizeof(struct v2_disk_dqinfo)) {
printk(KERN_WARNING "Can't read info structure on device %s.\n",
- f->f_vfsmnt->mnt_sb->s_id);
+ f->f_dentry->d_sb->s_id);
return -1;
}
info->dqi_bgrace = le32_to_cpu(dinfo.dqi_bgrace);
@@ -87,10 +87,12 @@ static int v2_write_file_info(struct sup
ssize_t size;
loff_t offset = V2_DQINFOOFF;
+ spin_lock(&dq_data_lock);
info->dqi_flags &= ~DQF_INFO_DIRTY;
dinfo.dqi_bgrace = cpu_to_le32(info->dqi_bgrace);
dinfo.dqi_igrace = cpu_to_le32(info->dqi_igrace);
dinfo.dqi_flags = cpu_to_le32(info->dqi_flags & DQF_MASK);
+ spin_unlock(&dq_data_lock);
dinfo.dqi_blocks = cpu_to_le32(info->u.v2_i.dqi_blocks);
dinfo.dqi_free_blk = cpu_to_le32(info->u.v2_i.dqi_free_blk);
dinfo.dqi_free_entry = cpu_to_le32(info->u.v2_i.dqi_free_entry);
@@ -100,7 +102,7 @@ static int v2_write_file_info(struct sup
set_fs(fs);
if (size != sizeof(struct v2_disk_dqinfo)) {
printk(KERN_WARNING "Can't write info structure on device %s.\n",
- f->f_vfsmnt->mnt_sb->s_id);
+ f->f_dentry->d_sb->s_id);
return -1;
}
return 0;
@@ -173,9 +175,10 @@ static ssize_t write_blk(struct file *fi
}
/* Remove empty block from list and return it */
-static int get_free_dqblk(struct file *filp, struct mem_dqinfo *info)
+static int get_free_dqblk(struct file *filp, int type)
{
dqbuf_t buf = getdqbuf();
+ struct mem_dqinfo *info = sb_dqinfo(filp->f_dentry->d_sb, type);
struct v2_disk_dqdbheader *dh = (struct v2_disk_dqdbheader *)buf;
int ret, blk;
@@ -193,7 +196,7 @@ static int get_free_dqblk(struct file *f
goto out_buf;
blk = info->u.v2_i.dqi_blocks++;
}
- mark_info_dirty(info);
+ mark_info_dirty(filp->f_dentry->d_sb, type);
ret = blk;
out_buf:
freedqbuf(buf);
@@ -201,8 +204,9 @@ out_buf:
}
/* Insert empty block to the list */
-static int put_free_dqblk(struct file *filp, struct mem_dqinfo *info, dqbuf_t buf, uint blk)
+static int put_free_dqblk(struct file *filp, int type, dqbuf_t buf, uint blk)
{
+ struct mem_dqinfo *info = sb_dqinfo(filp->f_dentry->d_sb, type);
struct v2_disk_dqdbheader *dh = (struct v2_disk_dqdbheader *)buf;
int err;
@@ -210,16 +214,17 @@ static int put_free_dqblk(struct file *f
dh->dqdh_prev_free = cpu_to_le32(0);
dh->dqdh_entries = cpu_to_le16(0);
info->u.v2_i.dqi_free_blk = blk;
- mark_info_dirty(info);
+ mark_info_dirty(filp->f_dentry->d_sb, type);
if ((err = write_blk(filp, blk, buf)) < 0) /* Some strange block. We had better leave it... */
return err;
return 0;
}
/* Remove given block from the list of blocks with free entries */
-static int remove_free_dqentry(struct file *filp, struct mem_dqinfo *info, dqbuf_t buf, uint blk)
+static int remove_free_dqentry(struct file *filp, int type, dqbuf_t buf, uint blk)
{
dqbuf_t tmpbuf = getdqbuf();
+ struct mem_dqinfo *info = sb_dqinfo(filp->f_dentry->d_sb, type);
struct v2_disk_dqdbheader *dh = (struct v2_disk_dqdbheader *)buf;
uint nextblk = le32_to_cpu(dh->dqdh_next_free), prevblk = le32_to_cpu(dh->dqdh_prev_free);
int err;
@@ -242,7 +247,7 @@ static int remove_free_dqentry(struct fi
}
else {
info->u.v2_i.dqi_free_entry = nextblk;
- mark_info_dirty(info);
+ mark_info_dirty(filp->f_dentry->d_sb, type);
}
freedqbuf(tmpbuf);
dh->dqdh_next_free = dh->dqdh_prev_free = cpu_to_le32(0);
@@ -255,9 +260,10 @@ out_buf:
}
/* Insert given block to the beginning of list with free entries */
-static int insert_free_dqentry(struct file *filp, struct mem_dqinfo *info, dqbuf_t buf, uint blk)
+static int insert_free_dqentry(struct file *filp, int type, dqbuf_t buf, uint blk)
{
dqbuf_t tmpbuf = getdqbuf();
+ struct mem_dqinfo *info = sb_dqinfo(filp->f_dentry->d_sb, type);
struct v2_disk_dqdbheader *dh = (struct v2_disk_dqdbheader *)buf;
int err;
@@ -276,7 +282,7 @@ static int insert_free_dqentry(struct fi
}
freedqbuf(tmpbuf);
info->u.v2_i.dqi_free_entry = blk;
- mark_info_dirty(info);
+ mark_info_dirty(filp->f_dentry->d_sb, type);
return 0;
out_buf:
freedqbuf(tmpbuf);
@@ -307,7 +313,7 @@ static uint find_free_dqentry(struct dqu
goto out_buf;
}
else {
- blk = get_free_dqblk(filp, info);
+ blk = get_free_dqblk(filp, dquot->dq_type);
if ((int)blk < 0) {
*err = blk;
freedqbuf(buf);
@@ -315,10 +321,10 @@ static uint find_free_dqentry(struct dqu
}
memset(buf, 0, V2_DQBLKSIZE);
info->u.v2_i.dqi_free_entry = blk; /* This is enough as block is already zeroed and entry list is empty... */
- mark_info_dirty(info);
+ mark_info_dirty(dquot->dq_sb, dquot->dq_type);
}
if (le16_to_cpu(dh->dqdh_entries)+1 >= V2_DQSTRINBLK) /* Block will be full? */
- if ((*err = remove_free_dqentry(filp, info, buf, blk)) < 0) {
+ if ((*err = remove_free_dqentry(filp, dquot->dq_type, buf, blk)) < 0) {
printk(KERN_ERR "VFS: find_free_dqentry(): Can't remove block (%u) from entry free list.\n", blk);
goto out_buf;
}
@@ -349,7 +355,6 @@ out_buf:
static int do_insert_tree(struct dquot *dquot, uint *treeblk, int depth)
{
struct file *filp = sb_dqopt(dquot->dq_sb)->files[dquot->dq_type];
- struct mem_dqinfo *info = sb_dqopt(dquot->dq_sb)->info + dquot->dq_type;
dqbuf_t buf;
int ret = 0, newson = 0, newact = 0;
u32 *ref;
@@ -358,7 +363,7 @@ static int do_insert_tree(struct dquot *
if (!(buf = getdqbuf()))
return -ENOMEM;
if (!*treeblk) {
- ret = get_free_dqblk(filp, info);
+ ret = get_free_dqblk(filp, dquot->dq_type);
if (ret < 0)
goto out_buf;
*treeblk = ret;
@@ -392,7 +397,7 @@ static int do_insert_tree(struct dquot *
ret = write_blk(filp, *treeblk, buf);
}
else if (newact && ret < 0)
- put_free_dqblk(filp, info, buf, *treeblk);
+ put_free_dqblk(filp, dquot->dq_type, buf, *treeblk);
out_buf:
freedqbuf(buf);
return ret;
@@ -417,6 +422,7 @@ static int v2_write_dquot(struct dquot *
ssize_t ret;
struct v2_disk_dqblk ddquot;
+ /* dq_off is guarded by dqio_sem */
if (!dquot->dq_off)
if ((ret = dq_insert_tree(dquot)) < 0) {
printk(KERN_ERR "VFS: Error %Zd occurred while creating quota.\n", ret);
@@ -424,7 +430,9 @@ static int v2_write_dquot(struct dquot *
}
filp = sb_dqopt(dquot->dq_sb)->files[type];
offset = dquot->dq_off;
+ spin_lock(&dq_data_lock);
mem2diskdqb(&ddquot, &dquot->dq_dqb, dquot->dq_id);
+ spin_unlock(&dq_data_lock);
fs = get_fs();
set_fs(KERNEL_DS);
ret = filp->f_op->write(filp, (char *)&ddquot, sizeof(struct v2_disk_dqblk), &offset);
@@ -445,7 +453,6 @@ static int v2_write_dquot(struct dquot *
static int free_dqentry(struct dquot *dquot, uint blk)
{
struct file *filp = sb_dqopt(dquot->dq_sb)->files[dquot->dq_type];
- struct mem_dqinfo *info = sb_dqopt(dquot->dq_sb)->info + dquot->dq_type;
struct v2_disk_dqdbheader *dh;
dqbuf_t buf = getdqbuf();
int ret = 0;
@@ -463,8 +470,8 @@ static int free_dqentry(struct dquot *dq
dh = (struct v2_disk_dqdbheader *)buf;
dh->dqdh_entries = cpu_to_le16(le16_to_cpu(dh->dqdh_entries)-1);
if (!le16_to_cpu(dh->dqdh_entries)) { /* Block got free? */
- if ((ret = remove_free_dqentry(filp, info, buf, blk)) < 0 ||
- (ret = put_free_dqblk(filp, info, buf, blk)) < 0) {
+ if ((ret = remove_free_dqentry(filp, dquot->dq_type, buf, blk)) < 0 ||
+ (ret = put_free_dqblk(filp, dquot->dq_type, buf, blk)) < 0) {
printk(KERN_ERR "VFS: Can't move quota data block (%u) to free list.\n", blk);
goto out_buf;
}
@@ -473,7 +480,7 @@ static int free_dqentry(struct dquot *dq
memset(buf+(dquot->dq_off & ((1 << V2_DQBLKSIZE_BITS)-1)), 0, sizeof(struct v2_disk_dqblk));
if (le16_to_cpu(dh->dqdh_entries) == V2_DQSTRINBLK-1) {
/* Insert will write block itself */
- if ((ret = insert_free_dqentry(filp, info, buf, blk)) < 0) {
+ if ((ret = insert_free_dqentry(filp, dquot->dq_type, buf, blk)) < 0) {
printk(KERN_ERR "VFS: Can't insert quota data block (%u) to free entry list.\n", blk);
goto out_buf;
}
@@ -494,7 +501,6 @@ out_buf:
static int remove_tree(struct dquot *dquot, uint *blk, int depth)
{
struct file *filp = sb_dqopt(dquot->dq_sb)->files[dquot->dq_type];
- struct mem_dqinfo *info = sb_dqopt(dquot->dq_sb)->info + dquot->dq_type;
dqbuf_t buf = getdqbuf();
int ret = 0;
uint newblk;
@@ -518,7 +524,7 @@ static int remove_tree(struct dquot *dqu
ref[GETIDINDEX(dquot->dq_id, depth)] = cpu_to_le32(0);
for (i = 0; i < V2_DQBLKSIZE && !buf[i]; i++); /* Block got empty? */
if (i == V2_DQBLKSIZE) {
- put_free_dqblk(filp, info, buf, *blk);
+ put_free_dqblk(filp, dquot->dq_type, buf, *blk);
*blk = 0;
}
else
@@ -632,7 +638,7 @@ static int v2_read_dquot(struct dquot *d
if (offset < 0)
printk(KERN_ERR "VFS: Can't read quota structure for id %u.\n", dquot->dq_id);
dquot->dq_off = 0;
- dquot->dq_flags |= DQ_FAKE;
+ set_bit(DQ_FAKE_B, &dquot->dq_flags);
memset(&dquot->dq_dqb, 0, sizeof(struct mem_dqblk));
ret = offset;
}
@@ -650,21 +656,24 @@ static int v2_read_dquot(struct dquot *d
ret = 0;
set_fs(fs);
disk2memdqb(&dquot->dq_dqb, &ddquot);
+ if (!dquot->dq_dqb.dqb_bhardlimit &&
+ !dquot->dq_dqb.dqb_bsoftlimit &&
+ !dquot->dq_dqb.dqb_ihardlimit &&
+ !dquot->dq_dqb.dqb_isoftlimit)
+ set_bit(DQ_FAKE_B, &dquot->dq_flags);
}
dqstats.reads++;
return ret;
}
-/* Commit changes of dquot to disk - it might also mean deleting it when quota became fake one and user has no blocks... */
-static int v2_commit_dquot(struct dquot *dquot)
+/* Check whether dquot should not be deleted. We know we are
+ * the only one operating on dquot (thanks to dq_lock) */
+static int v2_release_dquot(struct dquot *dquot)
{
- /* We clear the flag everytime so we don't loop when there was an IO error... */
- dquot->dq_flags &= ~DQ_MOD;
- if (dquot->dq_flags & DQ_FAKE && !(dquot->dq_dqb.dqb_curinodes | dquot->dq_dqb.dqb_curspace))
+ if (test_bit(DQ_FAKE_B, &dquot->dq_flags) && !(dquot->dq_dqb.dqb_curinodes | dquot->dq_dqb.dqb_curspace))
return v2_delete_dquot(dquot);
- else
- return v2_write_dquot(dquot);
+ return 0;
}
static struct quota_format_ops v2_format_ops = {
@@ -673,7 +682,8 @@ static struct quota_format_ops v2_format
.write_file_info = v2_write_file_info,
.free_file_info = NULL,
.read_dqblk = v2_read_dquot,
- .commit_dqblk = v2_commit_dquot,
+ .commit_dqblk = v2_write_dquot,
+ .release_dqblk = v2_release_dquot,
};
static struct quota_format_type v2_quota_format = {
--- diff/fs/reiserfs/file.c 2004-02-09 10:36:12.000000000 +0000
+++ source/fs/reiserfs/file.c 2004-03-16 11:27:48.432824752 +0000
@@ -365,7 +365,7 @@ int reiserfs_allocate_blocks_for_region(
// it means there are no existing in-tree representation for file area
// we are going to overwrite, so there is nothing to scan through for holes.
for ( curr_block = 0, itempos = path.pos_in_item ; curr_block < blocks_to_allocate && res == POSITION_FOUND ; ) {
-
+retry:
if ( itempos >= ih_item_len(ih)/UNFM_P_SIZE ) {
/* We run out of data in this indirect item, let's look for another
one. */
@@ -422,8 +422,8 @@ int reiserfs_allocate_blocks_for_region(
bh=get_last_bh(&path);
ih=get_ih(&path);
item = get_item(&path);
- // Itempos is still the same
- continue;
+ itempos = path.pos_in_item;
+ goto retry;
}
modifying_this_item = 1;
}
@@ -856,8 +856,12 @@ int reiserfs_prepare_file_region_for_wri
/* Try to find next item */
res = search_for_position_by_key(inode->i_sb, &key, &path);
/* Abort if no more items */
- if ( res != POSITION_FOUND )
+ if ( res != POSITION_FOUND ) {
+ /* make sure later loops don't use this item */
+ itembuf = NULL;
+ item = NULL;
break;
+ }
/* Update information about current indirect item */
itembuf = get_last_bh( &path );
@@ -1191,6 +1195,14 @@ out:
return res;
}
+static ssize_t reiserfs_aio_write(struct kiocb *iocb, const char __user *buf,
+ size_t count, loff_t pos)
+{
+ return generic_file_aio_write(iocb, buf, count, pos);
+}
+
+
+
struct file_operations reiserfs_file_operations = {
.read = generic_file_read,
.write = reiserfs_file_write,
@@ -1199,6 +1211,8 @@ struct file_operations reiserfs_file_ope
.release = reiserfs_file_release,
.fsync = reiserfs_sync_file,
.sendfile = generic_file_sendfile,
+ .aio_read = generic_file_aio_read,
+ .aio_write = reiserfs_aio_write,
};
--- diff/fs/reiserfs/inode.c 2004-02-09 10:36:12.000000000 +0000
+++ source/fs/reiserfs/inode.c 2004-03-16 11:27:48.434824448 +0000
@@ -209,7 +209,7 @@ static int file_capable (struct inode *
pathrelse(path) ;
reiserfs_update_sd(th, inode) ;
journal_end(th, s, len) ;
- journal_begin(th, s, len) ;
+ journal_begin(th, s, JOURNAL_PER_BALANCE_CNT * 6) ;
reiserfs_update_inode_transaction(inode) ;
}
@@ -444,7 +444,7 @@ static int reiserfs_get_blocks_direct_io
/* make sure future calls to the direct io funcs for this offset
** in the file fail by unmapping the buffer
*/
- reiserfs_unmap_buffer(bh_result);
+ clear_buffer_mapped(bh_result);
ret = -EINVAL ;
}
/* Possible unpacked tail. Flush the data before pages have
--- diff/fs/reiserfs/stree.c 2003-08-20 13:16:33.000000000 +0000
+++ source/fs/reiserfs/stree.c 2004-03-16 11:27:48.435824296 +0000
@@ -652,8 +652,8 @@ int search_by_key (struct super_block *
stop at leaf level - set to
DISK_LEAF_NODE_LEVEL */
) {
- int n_block_number = SB_ROOT_BLOCK (p_s_sb),
- expected_level = SB_TREE_HEIGHT (p_s_sb);
+ int n_block_number;
+ int expected_level;
struct buffer_head * p_s_bh;
struct path_element * p_s_last_element;
int n_node_level, n_retval;
@@ -677,6 +677,8 @@ int search_by_key (struct super_block *
/* With each iteration of this loop we search through the items in the
current node, and calculate the next current node(next path element)
for the next iteration of this loop.. */
+ n_block_number = SB_ROOT_BLOCK (p_s_sb);
+ expected_level = -1;
while ( 1 ) {
#ifdef CONFIG_REISERFS_CHECK
@@ -690,7 +692,6 @@ int search_by_key (struct super_block *
/* prep path to have another element added to it. */
p_s_last_element = PATH_OFFSET_PELEMENT(p_s_search_path, ++p_s_search_path->path_length);
fs_gen = get_generation (p_s_sb);
- expected_level --;
#ifdef SEARCH_BY_KEY_READA
/* schedule read of right neighbor */
@@ -705,25 +706,26 @@ int search_by_key (struct super_block *
pathrelse(p_s_search_path);
return IO_ERROR;
}
-
- if( fs_changed (fs_gen, p_s_sb) ) {
- PROC_INFO_INC( p_s_sb, search_by_key_fs_changed );
- PROC_INFO_INC( p_s_sb, sbk_fs_changed[ expected_level - 1 ] );
- }
+ if (expected_level == -1)
+ expected_level = SB_TREE_HEIGHT (p_s_sb);
+ expected_level --;
/* It is possible that schedule occurred. We must check whether the key
to search is still in the tree rooted from the current buffer. If
not then repeat search from the root. */
if ( fs_changed (fs_gen, p_s_sb) &&
- (!B_IS_IN_TREE (p_s_bh) || !key_in_buffer(p_s_search_path, p_s_key, p_s_sb)) ) {
- PROC_INFO_INC( p_s_sb, search_by_key_restarted );
+ (!B_IS_IN_TREE (p_s_bh) ||
+ B_LEVEL(p_s_bh) != expected_level ||
+ !key_in_buffer(p_s_search_path, p_s_key, p_s_sb))) {
+ PROC_INFO_INC( p_s_sb, search_by_key_fs_changed );
+ PROC_INFO_INC( p_s_sb, search_by_key_restarted );
PROC_INFO_INC( p_s_sb, sbk_restarted[ expected_level - 1 ] );
decrement_counters_in_path(p_s_search_path);
/* Get the root block number so that we can repeat the search
- starting from the root. */
+ starting from the root. */
n_block_number = SB_ROOT_BLOCK (p_s_sb);
- expected_level = SB_TREE_HEIGHT (p_s_sb);
+ expected_level = -1;
right_neighbor_of_leaf_node = 0;
/* repeat search from the root */
@@ -1103,6 +1105,7 @@ static char prepare_for_delete_or_cut(
for (n_counter = *p_n_removed;
n_counter < n_unfm_number; n_counter++, p_n_unfm_pointer-- ) {
+ cond_resched();
if (item_moved (&s_ih, p_s_path)) {
need_research = 1 ;
break;
@@ -1754,7 +1757,7 @@ void reiserfs_do_truncate (struct reiser
reiserfs_update_sd(th, p_s_inode) ;
journal_end(th, p_s_inode->i_sb, orig_len_alloc) ;
- journal_begin(th, p_s_inode->i_sb, orig_len_alloc) ;
+ journal_begin(th, p_s_inode->i_sb, JOURNAL_PER_BALANCE_CNT * 6) ;
reiserfs_update_inode_transaction(p_s_inode) ;
}
} while ( n_file_size > ROUND_UP (n_new_file_size) &&
--- diff/fs/select.c 2003-10-09 08:47:34.000000000 +0000
+++ source/fs/select.c 2004-03-16 11:27:48.436824144 +0000
@@ -291,8 +291,6 @@ static void select_bits_free(void *bits,
* Update: ERESTARTSYS breaks at least the xview clock binary, so
* I'm trying ERESTARTNOHAND which restart only when you want to.
*/
-#define MAX_SELECT_SECONDS \
- ((unsigned long) (MAX_SCHEDULE_TIMEOUT / HZ)-1)
asmlinkage long
sys_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, struct timeval __user *tvp)
@@ -315,9 +313,11 @@ sys_select(int n, fd_set __user *inp, fd
if (sec < 0 || usec < 0)
goto out_nofds;
- if ((unsigned long) sec < MAX_SELECT_SECONDS) {
+ if ((unsigned long) sec < (MAX_SCHEDULE_TIMEOUT-1) / HZ - 1) {
timeout = ROUND_UP(usec, 1000000/HZ);
timeout += sec * (unsigned long) HZ;
+ } else {
+ timeout = MAX_SCHEDULE_TIMEOUT-1;
}
}
@@ -469,11 +469,17 @@ asmlinkage long sys_poll(struct pollfd _
return -EINVAL;
if (timeout) {
- /* Careful about overflow in the intermediate values */
- if ((unsigned long) timeout < MAX_SCHEDULE_TIMEOUT / HZ)
- timeout = (unsigned long)(timeout*HZ+999)/1000+1;
- else /* Negative or overflow */
+ if (timeout < 0) {
timeout = MAX_SCHEDULE_TIMEOUT;
+ } else {
+ /* Careful about overflow in the intermediate values */
+ long seconds = timeout/1000;
+ timeout = ((timeout - 1000*seconds)*HZ + 999)/1000 + 1;
+ if (seconds <= (MAX_SCHEDULE_TIMEOUT-2) / HZ - 1)
+ timeout += seconds*HZ;
+ else
+ timeout = MAX_SCHEDULE_TIMEOUT-1;
+ }
}
poll_initwait(&table);
--- diff/fs/stat.c 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/stat.c 2004-03-16 11:27:48.436824144 +0000
@@ -397,6 +397,8 @@ EXPORT_SYMBOL(inode_get_bytes);
void inode_set_bytes(struct inode *inode, loff_t bytes)
{
+ /* Caller is here responsible for sufficient locking
+ * (ie. inode->i_lock) */
inode->i_blocks = bytes >> 9;
inode->i_bytes = bytes & 511;
}
--- diff/fs/super.c 2004-03-16 11:24:02.000000000 +0000
+++ source/fs/super.c 2004-03-16 11:27:48.437823992 +0000
@@ -68,6 +68,7 @@ static struct super_block *alloc_super(v
INIT_LIST_HEAD(&s->s_files);
INIT_LIST_HEAD(&s->s_instances);
INIT_HLIST_HEAD(&s->s_anon);
+ INIT_LIST_HEAD(&s->s_inodes);
init_rwsem(&s->s_umount);
sema_init(&s->s_lock, 1);
down_write(&s->s_umount);
--- diff/fs/sysfs/dir.c 2004-01-19 10:22:59.000000000 +0000
+++ source/fs/sysfs/dir.c 2004-03-16 11:27:48.438823840 +0000
@@ -20,6 +20,18 @@ static int init_dir(struct inode * inode
return 0;
}
+static void sysfs_d_iput(struct dentry * dentry, struct inode * inode)
+{
+ struct kobject * kobj = dentry->d_fsdata;
+
+ if (kobj)
+ kobject_put(kobj);
+ iput(inode);
+}
+
+static struct dentry_operations sysfs_dentry_operations = {
+ .d_iput = &sysfs_d_iput,
+};
static int create_dir(struct kobject * k, struct dentry * p,
const char * n, struct dentry ** d)
@@ -33,7 +45,8 @@ static int create_dir(struct kobject * k
S_IFDIR| S_IRWXU | S_IRUGO | S_IXUGO,
init_dir);
if (!error) {
- (*d)->d_fsdata = k;
+ (*d)->d_op = &sysfs_dentry_operations;
+ (*d)->d_fsdata = kobject_get(k);
p->d_inode->i_nlink++;
}
dput(*d);
@@ -120,13 +133,14 @@ void sysfs_remove_dir(struct kobject * k
down(&dentry->d_inode->i_sem);
spin_lock(&dcache_lock);
+restart:
node = dentry->d_subdirs.next;
while (node != &dentry->d_subdirs) {
struct dentry * d = list_entry(node,struct dentry,d_child);
- list_del_init(node);
+ node = node->next;
pr_debug(" o %s (%d): ",d->d_name.name,atomic_read(&d->d_count));
- if (d->d_inode) {
+ if (!d_unhashed(d) && (d->d_inode)) {
d = dget_locked(d);
pr_debug("removing");
@@ -137,12 +151,12 @@ void sysfs_remove_dir(struct kobject * k
d_delete(d);
simple_unlink(dentry->d_inode,d);
dput(d);
+ pr_debug(" done\n");
spin_lock(&dcache_lock);
+ /* re-acquired dcache_lock, need to restart */
+ goto restart;
}
- pr_debug(" done\n");
- node = dentry->d_subdirs.next;
}
- list_del_init(&dentry->d_child);
spin_unlock(&dcache_lock);
up(&dentry->d_inode->i_sem);
--- diff/fs/sysfs/group.c 2003-09-30 14:46:19.000000000 +0000
+++ source/fs/sysfs/group.c 2004-03-16 11:27:48.438823840 +0000
@@ -55,8 +55,8 @@ int sysfs_create_group(struct kobject *
if ((error = create_files(dir,grp))) {
if (grp->name)
sysfs_remove_subdir(dir);
- dput(dir);
}
+ dput(dir);
return error;
}
@@ -68,12 +68,13 @@ void sysfs_remove_group(struct kobject *
if (grp->name)
dir = sysfs_get_dentry(kobj->dentry,grp->name);
else
- dir = kobj->dentry;
+ dir = dget(kobj->dentry);
remove_files(dir,grp);
- dput(dir);
if (grp->name)
sysfs_remove_subdir(dir);
+ /* release the ref. taken in this routine */
+ dput(dir);
}
--- diff/fs/ufs/truncate.c 2003-02-26 16:00:55.000000000 +0000
+++ source/fs/ufs/truncate.c 2004-03-16 11:27:48.438823840 +0000
@@ -38,6 +38,7 @@
#include <linux/string.h>
#include <linux/smp_lock.h>
#include <linux/buffer_head.h>
+#include <linux/blkdev.h>
#include <linux/sched.h>
#include "swab.h"
@@ -456,7 +457,7 @@ void ufs_truncate (struct inode * inode)
break;
if (IS_SYNC(inode) && (inode->i_state & I_DIRTY))
ufs_sync_inode (inode);
- blk_run_queues();
+ blk_run_address_space(inode->i_mapping);
yield();
}
offset = inode->i_size & uspi->s_fshift;
--- diff/fs/xfs/linux/xfs_aops.c 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/xfs/linux/xfs_aops.c 2004-03-16 11:27:48.439823688 +0000
@@ -1032,7 +1032,8 @@ linvfs_direct_IO(
if (error)
return -error;
- return blockdev_direct_IO(rw, iocb, inode, iomap.iomap_target->pbr_bdev,
+ return blockdev_direct_IO_no_locking(rw, iocb, inode,
+ iomap.iomap_target->pbr_bdev,
iov, offset, nr_segs,
linvfs_get_blocks_direct,
linvfs_unwritten_convert_direct);
--- diff/fs/xfs/linux/xfs_buf.c 2004-03-11 10:20:28.000000000 +0000
+++ source/fs/xfs/linux/xfs_buf.c 2004-03-16 11:27:48.441823384 +0000
@@ -1013,7 +1013,7 @@ pagebuf_lock(
{
PB_TRACE(pb, "lock", 0);
if (atomic_read(&pb->pb_io_remaining))
- blk_run_queues();
+ blk_run_address_space(pb->pb_target->pbr_mapping);
down(&pb->pb_sema);
PB_SET_OWNER(pb);
PB_TRACE(pb, "locked", 0);
@@ -1109,7 +1109,7 @@ _pagebuf_wait_unpin(
if (atomic_read(&pb->pb_pin_count) == 0)
break;
if (atomic_read(&pb->pb_io_remaining))
- blk_run_queues();
+ blk_run_address_space(pb->pb_target->pbr_mapping);
schedule();
}
remove_wait_queue(&pb->pb_waiters, &wait);
@@ -1407,7 +1407,7 @@ submit_io:
if (pb->pb_flags & PBF_RUN_QUEUES) {
pb->pb_flags &= ~PBF_RUN_QUEUES;
if (atomic_read(&pb->pb_io_remaining) > 1)
- blk_run_queues();
+ blk_run_address_space(pb->pb_target->pbr_mapping);
}
}
@@ -1471,7 +1471,7 @@ pagebuf_iowait(
{
PB_TRACE(pb, "iowait", 0);
if (atomic_read(&pb->pb_io_remaining))
- blk_run_queues();
+ blk_run_address_space(pb->pb_target->pbr_mapping);
down(&pb->pb_iodonesema);
PB_TRACE(pb, "iowaited", (long)pb->pb_error);
return pb->pb_error;
@@ -1617,7 +1617,6 @@ STATIC int
pagebuf_daemon(
void *data)
{
- int count;
page_buf_t *pb;
struct list_head *curr, *next, tmp;
@@ -1640,7 +1639,6 @@ pagebuf_daemon(
spin_lock(&pbd_delwrite_lock);
- count = 0;
list_for_each_safe(curr, next, &pbd_delwrite_queue) {
pb = list_entry(curr, page_buf_t, pb_list);
@@ -1657,7 +1655,6 @@ pagebuf_daemon(
pb->pb_flags &= ~PBF_DELWRI;
pb->pb_flags |= PBF_WRITE;
list_move(&pb->pb_list, &tmp);
- count++;
}
}
@@ -1667,12 +1664,11 @@ pagebuf_daemon(
list_del_init(&pb->pb_list);
pagebuf_iostrategy(pb);
+ blk_run_address_space(pb->pb_target->pbr_mapping);
}
if (as_list_len > 0)
purge_addresses();
- if (count)
- blk_run_queues();
force_flush = 0;
} while (pagebuf_daemon_active);
@@ -1689,7 +1685,6 @@ pagebuf_delwri_flush(
page_buf_t *pb;
struct list_head *curr, *next, tmp;
int pincount = 0;
- int flush_cnt = 0;
pagebuf_runall_queues(pagebuf_dataio_workqueue);
pagebuf_runall_queues(pagebuf_logio_workqueue);
@@ -1733,14 +1728,8 @@ pagebuf_delwri_flush(
pagebuf_lock(pb);
pagebuf_iostrategy(pb);
- if (++flush_cnt > 32) {
- blk_run_queues();
- flush_cnt = 0;
- }
}
- blk_run_queues();
-
while (!list_empty(&tmp)) {
pb = list_entry(tmp.next, page_buf_t, pb_list);
@@ -1751,6 +1740,9 @@ pagebuf_delwri_flush(
pagebuf_rele(pb);
}
+ if (flags & PBDF_WAIT)
+ blk_run_address_space(target->pbr_mapping);
+
if (pinptr)
*pinptr = pincount;
}
--- diff/fs/xfs/linux/xfs_vnode.h 2004-02-09 10:36:12.000000000 +0000
+++ source/fs/xfs/linux/xfs_vnode.h 2004-03-16 11:27:48.441823384 +0000
@@ -600,7 +600,8 @@ static __inline__ void vn_flagclr(struct
(!list_empty(&(LINVFS_GET_IP(vp)->i_mapping->i_mmap)) || \
(!list_empty(&(LINVFS_GET_IP(vp)->i_mapping->i_mmap_shared))))
#define VN_CACHED(vp) (LINVFS_GET_IP(vp)->i_mapping->nrpages)
-#define VN_DIRTY(vp) (!list_empty(&(LINVFS_GET_IP(vp)->i_mapping->dirty_pages)))
+#define VN_DIRTY(vp) mapping_tagged(LINVFS_GET_IP(vp)->i_mapping, \
+ PAGECACHE_TAG_DIRTY)
#define VMODIFY(vp) VN_FLAGSET(vp, VMODIFIED)
#define VUNMODIFY(vp) VN_FLAGCLR(vp, VMODIFIED)
--- diff/include/asm-alpha/irq.h 2003-08-20 13:16:33.000000000 +0000
+++ source/include/asm-alpha/irq.h 2004-03-16 11:27:48.442823232 +0000
@@ -93,5 +93,8 @@ extern void enable_irq(unsigned int);
struct pt_regs;
extern void (*perf_irq)(unsigned long, struct pt_regs *);
+struct irqaction;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif /* _ALPHA_IRQ_H */
--- diff/include/asm-alpha/spinlock.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/asm-alpha/spinlock.h 2004-03-16 11:27:48.443823080 +0000
@@ -6,6 +6,10 @@
#include <linux/kernel.h>
#include <asm/current.h>
+#ifdef CONFIG_LOCKMETER
+#undef DEBUG_SPINLOCK
+#undef DEBUG_RWLOCK
+#endif
/*
* Simple spin lock operations. There are two variants, one clears IRQ's
@@ -95,9 +99,18 @@ static inline int _raw_spin_trylock(spin
typedef struct {
volatile int write_lock:1, read_counter:31;
+#ifdef CONFIG_LOCKMETER
+ /* required for LOCKMETER since all bits in lock are used */
+ /* need this storage for CPU and lock INDEX ............. */
+ unsigned magic;
+#endif
} /*__attribute__((aligned(32)))*/ rwlock_t;
+#ifdef CONFIG_LOCKMETER
+#define RW_LOCK_UNLOCKED (rwlock_t) { 0, 0, 0 }
+#else
#define RW_LOCK_UNLOCKED (rwlock_t) { 0, 0 }
+#endif
#define rwlock_init(x) do { *(x) = RW_LOCK_UNLOCKED; } while(0)
#define rwlock_is_locked(x) (*(volatile int *)(x) != 0)
@@ -169,4 +182,41 @@ static inline void _raw_read_unlock(rwlo
: "m" (*lock) : "memory");
}
+#ifdef CONFIG_LOCKMETER
+static inline int _raw_write_trylock(rwlock_t *lock)
+{
+ long temp,result;
+
+ __asm__ __volatile__(
+ " ldl_l %1,%0\n"
+ " mov $31,%2\n"
+ " bne %1,1f\n"
+ " or $31,1,%2\n"
+ " stl_c %2,%0\n"
+ "1: mb\n"
+ : "=m" (*(volatile int *)lock), "=&r" (temp), "=&r" (result)
+ : "m" (*(volatile int *)lock)
+ );
+
+ return (result);
+}
+
+static inline int _raw_read_trylock(rwlock_t *lock)
+{
+ unsigned long temp,result;
+
+ __asm__ __volatile__(
+ " ldl_l %1,%0\n"
+ " mov $31,%2\n"
+ " blbs %1,1f\n"
+ " subl %1,2,%2\n"
+ " stl_c %2,%0\n"
+ "1: mb\n"
+ : "=m" (*(volatile int *)lock), "=&r" (temp), "=&r" (result)
+ : "m" (*(volatile int *)lock)
+ );
+ return (result);
+}
+#endif /* CONFIG_LOCKMETER */
+
#endif /* _ALPHA_SPINLOCK_H */
--- diff/include/asm-arm/arch-ebsa285/irqs.h 2003-05-21 10:50:16.000000000 +0000
+++ source/include/asm-arm/arch-ebsa285/irqs.h 2004-03-16 11:27:48.444822928 +0000
@@ -91,8 +91,8 @@
#undef RTC_IRQ
#define RTC_IRQ IRQ_ISA_RTC_ALARM
-#undef AUX_IRQ
-#define AUX_IRQ (machine_is_netwinder() ? IRQ_NETWINDER_PS2MOUSE : IRQ_ISA_PS2MOUSE)
+#define I8042_KBD_IRQ IRQ_ISA_KEYBOARD
+#define I8042_AUX_IRQ (machine_is_netwinder() ? IRQ_NETWINDER_PS2MOUSE : IRQ_ISA_PS2MOUSE)
#define IRQ_FLOPPYDISK IRQ_ISA_FLOPPY
#define irq_canonicalize(_i) (((_i) == IRQ_ISA_CASCADE) ? IRQ_ISA_2 : _i)
--- diff/include/asm-arm/arch-shark/irqs.h 2002-11-28 11:30:39.000000000 +0000
+++ source/include/asm-arm/arch-shark/irqs.h 2004-03-16 11:27:48.444822928 +0000
@@ -8,5 +8,6 @@
#define IRQ_ISA_KEYBOARD 1
#define RTC_IRQ 8
-#define AUX_IRQ 12
+#define I8042_KBD_IRQ 1
+#define I8042_AUX_IRQ 12
#define IRQ_HARDDISK 14
--- diff/include/asm-arm/irq.h 2003-05-21 10:50:16.000000000 +0000
+++ source/include/asm-arm/irq.h 2004-03-16 11:27:48.445822776 +0000
@@ -44,5 +44,9 @@ void disable_irq_wake(unsigned int irq);
void enable_irq_wake(unsigned int irq);
int setup_irq(unsigned int, struct irqaction *);
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif
--- diff/include/asm-arm26/irq.h 2003-06-30 09:07:24.000000000 +0000
+++ source/include/asm-arm26/irq.h 2004-03-16 11:27:48.443823080 +0000
@@ -45,6 +45,8 @@ extern void enable_irq(unsigned int);
int set_irq_type(unsigned int irq, unsigned int type);
int setup_irq(unsigned int, struct irqaction *);
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
#endif
--- diff/include/asm-arm26/thread_info.h 2004-02-18 08:54:12.000000000 +0000
+++ source/include/asm-arm26/thread_info.h 2004-03-16 11:27:48.444822928 +0000
@@ -81,8 +81,6 @@ static inline struct thread_info *curren
/* FIXME - PAGE_SIZE < 32K */
#define THREAD_SIZE (8192)
-/*FIXME INIT_THREAD_SIZE - how big? */
-//#define INIT_THREAD_SIZE (65536)
#define __get_user_regs(x) (((struct pt_regs *)((unsigned long)(x) + THREAD_SIZE - 8)) - 1)
extern struct thread_info *alloc_thread_info(struct task_struct *task);
--- diff/include/asm-cris/irq.h 2003-07-11 08:39:50.000000000 +0000
+++ source/include/asm-cris/irq.h 2004-03-16 11:27:48.445822776 +0000
@@ -14,6 +14,10 @@ extern void enable_irq(unsigned int);
#define disable_irq_nosync disable_irq
#define enable_irq_nosync enable_irq
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif /* _ASM_IRQ_H */
--- diff/include/asm-generic/siginfo.h 2003-08-26 09:00:54.000000000 +0000
+++ source/include/asm-generic/siginfo.h 2004-03-16 11:27:48.446822624 +0000
@@ -118,6 +118,7 @@ typedef struct siginfo {
#define __SI_FAULT (3 << 16)
#define __SI_CHLD (4 << 16)
#define __SI_RT (5 << 16)
+#define __SI_MESGQ (6 << 16)
#define __SI_CODE(T,N) ((T) | ((N) & 0xffff))
#else
#define __SI_KILL 0
@@ -126,6 +127,7 @@ typedef struct siginfo {
#define __SI_FAULT 0
#define __SI_CHLD 0
#define __SI_RT 0
+#define __SI_MESGQ 0
#define __SI_CODE(T,N) (N)
#endif
@@ -137,7 +139,7 @@ typedef struct siginfo {
#define SI_KERNEL 0x80 /* sent by the kernel from somewhere */
#define SI_QUEUE -1 /* sent by sigqueue */
#define SI_TIMER __SI_CODE(__SI_TIMER,-2) /* sent by timer expiration */
-#define SI_MESGQ -3 /* sent by real time mesq state change */
+#define SI_MESGQ __SI_CODE(__SI_MESGQ,-3) /* sent by real time mesq state change */
#define SI_ASYNCIO -4 /* sent by AIO completion */
#define SI_SIGIO -5 /* sent by queued SIGIO */
#define SI_TKILL -6 /* sent by tkill system call */
--- diff/include/asm-generic/tlb.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-generic/tlb.h 2004-03-16 11:27:48.446822624 +0000
@@ -39,7 +39,6 @@ struct mmu_gather {
unsigned int nr; /* set to ~0U means fast mode */
unsigned int need_flush;/* Really unmapped some ptes? */
unsigned int fullmm; /* non-zero means full mm flush */
- unsigned long freed;
struct page * pages[FREE_PTE_NR];
};
@@ -60,7 +59,6 @@ tlb_gather_mmu(struct mm_struct *mm, uns
tlb->nr = num_online_cpus() > 1 ? 0U : ~0U;
tlb->fullmm = full_mm_flush;
- tlb->freed = 0;
return tlb;
}
@@ -85,13 +83,6 @@ tlb_flush_mmu(struct mmu_gather *tlb, un
static inline void
tlb_finish_mmu(struct mmu_gather *tlb, unsigned long start, unsigned long end)
{
- int freed = tlb->freed;
- struct mm_struct *mm = tlb->mm;
- int rss = mm->rss;
-
- if (rss < freed)
- freed = rss;
- mm->rss = rss - freed;
tlb_flush_mmu(tlb, start, end);
/* keep the page table cache within bounds */
@@ -146,4 +137,6 @@ static inline void tlb_remove_page(struc
__pmd_free_tlb(tlb, pmdp); \
} while (0)
+#define tlb_migrate_prepare(mm) do { } while(0)
+
#endif /* _ASM_GENERIC__TLB_H */
--- diff/include/asm-h8300/irq.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/asm-h8300/irq.h 2004-03-16 11:27:48.447822472 +0000
@@ -68,4 +68,8 @@ extern void disable_irq(unsigned int);
#define enable_irq_nosync(x) enable_irq(x)
#define disable_irq_nosync(x) disable_irq(x)
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif /* _H8300_IRQ_H_ */
--- diff/include/asm-i386/bugs.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/asm-i386/bugs.h 2004-03-16 11:27:48.448822320 +0000
@@ -1,11 +1,11 @@
/*
* include/asm-i386/bugs.h
*
- * Copyright (C) 1994 Linus Torvalds
+ * Copyright (C) 1994 Linus Torvalds
*
* Cyrix stuff, June 1998 by:
* - Rafael R. Reilova (moved everything from head.S),
- * <rreilova@ececs.uc.edu>
+ * <rreilova@ececs.uc.edu>
* - Channing Corn (tests & fixes),
* - Andrew D. Balsa (code cleanup).
*
@@ -25,7 +25,20 @@
#include <asm/processor.h>
#include <asm/i387.h>
#include <asm/msr.h>
-
+#ifdef CONFIG_KGDB
+/*
+ * Provied the command line "gdb" initial break
+ */
+int __init kgdb_initial_break(char * str)
+{
+ if (*str == '\0'){
+ breakpoint();
+ return 1;
+ }
+ return 0;
+}
+__setup("gdb",kgdb_initial_break);
+#endif
static int __init no_halt(char *s)
{
boot_cpu_data.hlt_works_ok = 0;
@@ -140,7 +153,7 @@ static void __init check_popad(void)
: "ecx", "edi" );
/* If this fails, it means that any user program may lock the CPU hard. Too bad. */
if (res != 12345678) printk( "Buggy.\n" );
- else printk( "OK.\n" );
+ else printk( "OK.\n" );
#endif
}
--- diff/include/asm-i386/checksum.h 2003-11-25 15:24:59.000000000 +0000
+++ source/include/asm-i386/checksum.h 2004-03-16 11:27:48.449822168 +0000
@@ -25,7 +25,7 @@ asmlinkage unsigned int csum_partial(con
* better 64-bit) boundary
*/
-asmlinkage unsigned int csum_partial_copy_generic( const char *src, char *dst, int len, int sum,
+asmlinkage unsigned int direct_csum_partial_copy_generic( const char *src, char *dst, int len, int sum,
int *src_err_ptr, int *dst_err_ptr);
/*
@@ -39,14 +39,19 @@ static __inline__
unsigned int csum_partial_copy_nocheck ( const char *src, char *dst,
int len, int sum)
{
- return csum_partial_copy_generic ( src, dst, len, sum, NULL, NULL);
+ /*
+ * The direct function is OK for kernel-space => kernel-space copies:
+ */
+ return direct_csum_partial_copy_generic ( src, dst, len, sum, NULL, NULL);
}
static __inline__
unsigned int csum_partial_copy_from_user ( const char *src, char *dst,
int len, int sum, int *err_ptr)
{
- return csum_partial_copy_generic ( src, dst, len, sum, err_ptr, NULL);
+ if (copy_from_user(dst, src, len))
+ *err_ptr = -EFAULT;
+ return csum_partial(dst, len, sum);
}
/*
@@ -172,11 +177,26 @@ static __inline__ unsigned short int csu
* Copy and checksum to user
*/
#define HAVE_CSUM_COPY_USER
-static __inline__ unsigned int csum_and_copy_to_user(const char *src, char *dst,
+static __inline__ unsigned int direct_csum_and_copy_to_user(const char *src, char *dst,
int len, int sum, int *err_ptr)
{
if (access_ok(VERIFY_WRITE, dst, len))
- return csum_partial_copy_generic(src, dst, len, sum, NULL, err_ptr);
+ return direct_csum_partial_copy_generic(src, dst, len, sum, NULL, err_ptr);
+
+ if (len)
+ *err_ptr = -EFAULT;
+
+ return -1; /* invalid checksum */
+}
+
+static __inline__ unsigned int csum_and_copy_to_user(const char *src, char *dst,
+ int len, int sum, int *err_ptr)
+{
+ if (access_ok(VERIFY_WRITE, dst, len)) {
+ if (copy_to_user(dst, src, len))
+ *err_ptr = -EFAULT;
+ return csum_partial(src, len, sum);
+ }
if (len)
*err_ptr = -EFAULT;
--- diff/include/asm-i386/desc.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/asm-i386/desc.h 2004-03-16 11:27:48.449822168 +0000
@@ -21,6 +21,13 @@ struct Xgt_desc_struct {
extern struct Xgt_desc_struct idt_descr, cpu_gdt_descr[NR_CPUS];
+extern void trap_init_virtual_IDT(void);
+extern void trap_init_virtual_GDT(void);
+
+asmlinkage int system_call(void);
+asmlinkage void lcall7(void);
+asmlinkage void lcall27(void);
+
#define load_TR_desc() __asm__ __volatile__("ltr %%ax"::"a" (GDT_ENTRY_TSS*8))
#define load_LDT_desc() __asm__ __volatile__("lldt %%ax"::"a" (GDT_ENTRY_LDT*8))
@@ -30,6 +37,7 @@ extern struct Xgt_desc_struct idt_descr,
*/
extern struct desc_struct default_ldt[];
extern void set_intr_gate(unsigned int irq, void * addr);
+extern void set_trap_gate(unsigned int n, void *addr);
#define _set_tssldt_desc(n,addr,limit,type) \
__asm__ __volatile__ ("movw %w3,0(%2)\n\t" \
@@ -90,31 +98,8 @@ static inline void load_TLS(struct threa
#undef C
}
-static inline void clear_LDT(void)
-{
- int cpu = get_cpu();
-
- set_ldt_desc(cpu, &default_ldt[0], 5);
- load_LDT_desc();
- put_cpu();
-}
-
-/*
- * load one particular LDT into the current CPU
- */
-static inline void load_LDT_nolock(mm_context_t *pc, int cpu)
-{
- void *segments = pc->ldt;
- int count = pc->size;
-
- if (likely(!count)) {
- segments = &default_ldt[0];
- count = 5;
- }
-
- set_ldt_desc(cpu, segments, count);
- load_LDT_desc();
-}
+extern struct page *default_ldt_page;
+extern void load_LDT_nolock(mm_context_t *pc, int cpu);
static inline void load_LDT(mm_context_t *pc)
{
@@ -123,6 +108,6 @@ static inline void load_LDT(mm_context_t
put_cpu();
}
-#endif /* !__ASSEMBLY__ */
+#endif /* !__ASSEMBLY__ */
#endif
--- diff/include/asm-i386/fixmap.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-i386/fixmap.h 2004-03-16 11:27:48.450822016 +0000
@@ -18,17 +18,15 @@
#include <asm/acpi.h>
#include <asm/apicdef.h>
#include <asm/page.h>
-#ifdef CONFIG_HIGHMEM
#include <linux/threads.h>
#include <asm/kmap_types.h>
-#endif
/*
* Here we define all the compile-time 'special' virtual
* addresses. The point is to have a constant address at
* compile time, but to set the physical address only
- * in the boot process. We allocate these special addresses
- * from the end of virtual memory (0xfffff000) backwards.
+ * in the boot process. We allocate these special addresses
+ * from the end of virtual memory (0xffffe000) backwards.
* Also this lets us do fail-safe vmalloc(), we
* can guarantee that these special addresses and
* vmalloc()-ed addresses never overlap.
@@ -41,11 +39,20 @@
* TLB entries of such buffers will not be flushed across
* task switches.
*/
+
+/*
+ * on UP currently we will have no trace of the fixmap mechanizm,
+ * no page table allocations, etc. This might change in the
+ * future, say framebuffers for the console driver(s) could be
+ * fix-mapped?
+ */
enum fixed_addresses {
FIX_HOLE,
FIX_VSYSCALL,
#ifdef CONFIG_X86_LOCAL_APIC
FIX_APIC_BASE, /* local (CPU) APIC) -- required for SMP or not */
+#else
+ FIX_VSTACK_HOLE_1,
#endif
#ifdef CONFIG_X86_IO_APIC
FIX_IO_APIC_BASE_0,
@@ -57,16 +64,21 @@ enum fixed_addresses {
FIX_LI_PCIA, /* Lithium PCI Bridge A */
FIX_LI_PCIB, /* Lithium PCI Bridge B */
#endif
-#ifdef CONFIG_X86_F00F_BUG
- FIX_F00F_IDT, /* Virtual mapping for IDT */
-#endif
+ FIX_IDT,
+ FIX_GDT_1,
+ FIX_GDT_0,
+ FIX_TSS_3,
+ FIX_TSS_2,
+ FIX_TSS_1,
+ FIX_TSS_0,
+ FIX_ENTRY_TRAMPOLINE_1,
+ FIX_ENTRY_TRAMPOLINE_0,
#ifdef CONFIG_X86_CYCLONE_TIMER
FIX_CYCLONE_TIMER, /*cyclone timer register*/
+ FIX_VSTACK_HOLE_2,
#endif
-#ifdef CONFIG_HIGHMEM
FIX_KMAP_BEGIN, /* reserved pte's for temporary kernel mappings */
FIX_KMAP_END = FIX_KMAP_BEGIN+(KM_TYPE_NR*NR_CPUS)-1,
-#endif
#ifdef CONFIG_ACPI_BOOT
FIX_ACPI_BEGIN,
FIX_ACPI_END = FIX_ACPI_BEGIN + FIX_ACPI_PAGES - 1,
@@ -98,12 +110,15 @@ extern void __set_fixmap (enum fixed_add
__set_fixmap(idx, 0, __pgprot(0))
/*
- * used by vmalloc.c.
+ * used by vmalloc.c and various other places.
*
* Leave one empty page between vmalloc'ed areas and
* the start of the fixmap.
+ *
+ * IMPORTANT: we have to align FIXADDR_TOP so that the virtual stack
+ * is THREAD_SIZE aligned.
*/
-#define FIXADDR_TOP (0xfffff000UL)
+#define FIXADDR_TOP (0xffffe000UL & ~(THREAD_SIZE-1))
#define __FIXADDR_SIZE (__end_of_permanent_fixed_addresses << PAGE_SHIFT)
#define FIXADDR_START (FIXADDR_TOP - __FIXADDR_SIZE)
--- diff/include/asm-i386/highmem.h 2003-10-09 08:47:34.000000000 +0000
+++ source/include/asm-i386/highmem.h 2004-03-16 11:27:48.450822016 +0000
@@ -25,26 +25,19 @@
#include <linux/threads.h>
#include <asm/kmap_types.h>
#include <asm/tlbflush.h>
+#include <asm/atomic_kmap.h>
/* declarations for highmem.c */
extern unsigned long highstart_pfn, highend_pfn;
-extern pte_t *kmap_pte;
-extern pgprot_t kmap_prot;
extern pte_t *pkmap_page_table;
-
-extern void kmap_init(void);
+extern void kmap_init(void) __init;
/*
* Right now we initialize only a single pte table. It can be extended
* easily, subsequent pte tables have to be allocated in one physical
* chunk of RAM.
*/
-#if NR_CPUS <= 32
-#define PKMAP_BASE (0xff800000UL)
-#else
-#define PKMAP_BASE (0xff600000UL)
-#endif
#ifdef CONFIG_X86_PAE
#define LAST_PKMAP 512
#else
--- diff/include/asm-i386/irq.h 2004-01-19 10:22:59.000000000 +0000
+++ source/include/asm-i386/irq.h 2004-03-16 11:27:48.451821864 +0000
@@ -14,6 +14,7 @@
#include <linux/sched.h>
/* include comes from machine specific directory */
#include "irq_vectors.h"
+#include <asm/thread_info.h>
static __inline__ int irq_canonicalize(int irq)
{
@@ -30,4 +31,28 @@ extern int can_request_irq(unsigned int,
#define ARCH_HAS_NMI_WATCHDOG /* See include/linux/nmi.h */
#endif
+#ifdef CONFIG_4KSTACKS
+/*
+ * per-CPU IRQ handling contexts (thread information and stack)
+ */
+union irq_ctx {
+ struct thread_info tinfo;
+ u32 stack[THREAD_SIZE/sizeof(u32)];
+};
+
+extern union irq_ctx *hardirq_ctx[NR_CPUS];
+extern union irq_ctx *softirq_ctx[NR_CPUS];
+
+extern void irq_ctx_init(int cpu);
+
+#define __ARCH_HAS_DO_SOFTIRQ
+#else
+#define irq_ctx_init(cpu) do { ; } while (0)
+#endif
+
+struct irqaction;
+struct pt_regs;
+asmlinkage int handle_IRQ_event(unsigned int, struct pt_regs *,
+ struct irqaction *);
+
#endif /* _ASM_IRQ_H */
--- diff/include/asm-i386/kmap_types.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/asm-i386/kmap_types.h 2004-03-16 11:27:48.484816848 +0000
@@ -2,31 +2,38 @@
#define _ASM_KMAP_TYPES_H
#include <linux/config.h>
-
-#ifdef CONFIG_DEBUG_HIGHMEM
-# define D(n) __KM_FENCE_##n ,
-#else
-# define D(n)
-#endif
+#include <linux/thread_info.h>
enum km_type {
-D(0) KM_BOUNCE_READ,
-D(1) KM_SKB_SUNRPC_DATA,
-D(2) KM_SKB_DATA_SOFTIRQ,
-D(3) KM_USER0,
-D(4) KM_USER1,
-D(5) KM_BIO_SRC_IRQ,
-D(6) KM_BIO_DST_IRQ,
-D(7) KM_PTE0,
-D(8) KM_PTE1,
-D(9) KM_PTE2,
-D(10) KM_IRQ0,
-D(11) KM_IRQ1,
-D(12) KM_SOFTIRQ0,
-D(13) KM_SOFTIRQ1,
-D(14) KM_TYPE_NR
-};
-
-#undef D
+ /*
+ * IMPORTANT: don't move these 3 entries, be wary when adding entries,
+ * the 4G/4G virtual stack must be THREAD_SIZE aligned on each cpu.
+ */
+ KM_BOUNCE_READ,
+ KM_VSTACK_BASE,
+ KM_VSTACK_TOP = KM_VSTACK_BASE + STACK_PAGE_COUNT-1,
+ KM_LDT_PAGE15,
+ KM_LDT_PAGE0 = KM_LDT_PAGE15 + 16-1,
+ KM_USER_COPY,
+ KM_VSTACK_HOLE,
+ KM_SKB_SUNRPC_DATA,
+ KM_SKB_DATA_SOFTIRQ,
+ KM_USER0,
+ KM_USER1,
+ KM_BIO_SRC_IRQ,
+ KM_BIO_DST_IRQ,
+ KM_PTE0,
+ KM_PTE1,
+ KM_PTE2,
+ KM_IRQ0,
+ KM_IRQ1,
+ KM_SOFTIRQ0,
+ KM_SOFTIRQ1,
+ /*
+ * Be wary when adding entries:
+ * the 4G/4G virtual stack must be THREAD_SIZE aligned on each cpu.
+ */
+ KM_TYPE_NR
+};
#endif
--- diff/include/asm-i386/mman.h 2003-10-09 08:47:34.000000000 +0000
+++ source/include/asm-i386/mman.h 2004-03-16 11:27:48.485816696 +0000
@@ -22,6 +22,7 @@
#define MAP_NORESERVE 0x4000 /* don't check for reservations */
#define MAP_POPULATE 0x8000 /* populate (prefault) pagetables */
#define MAP_NONBLOCK 0x10000 /* do not block on IO */
+#define MAP_INHERIT 0x20000 /* inherit the protection bits of the underlying vma*/
#define MS_ASYNC 1 /* sync memory asynchronously */
#define MS_INVALIDATE 2 /* invalidate the caches */
--- diff/include/asm-i386/mmu.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/asm-i386/mmu.h 2004-03-16 11:27:48.486816544 +0000
@@ -8,10 +8,13 @@
*
* cpu_vm_mask is used to optimize ldt flushing.
*/
+
+#define MAX_LDT_PAGES 16
+
typedef struct {
int size;
struct semaphore sem;
- void *ldt;
+ struct page *ldt_pages[MAX_LDT_PAGES];
} mm_context_t;
#endif
--- diff/include/asm-i386/mmu_context.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/asm-i386/mmu_context.h 2004-03-16 11:27:48.486816544 +0000
@@ -29,6 +29,10 @@ static inline void switch_mm(struct mm_s
{
int cpu = smp_processor_id();
+#ifdef CONFIG_X86_SWITCH_PAGETABLES
+ if (tsk->mm)
+ tsk->thread_info->user_pgd = (void *)__pa(tsk->mm->pgd);
+#endif
if (likely(prev != next)) {
/* stop flush ipis for the previous mm */
cpu_clear(cpu, prev->cpu_vm_mask);
@@ -39,12 +43,14 @@ static inline void switch_mm(struct mm_s
cpu_set(cpu, next->cpu_vm_mask);
/* Re-load page tables */
+#if !defined(CONFIG_X86_SWITCH_PAGETABLES)
load_cr3(next->pgd);
+#endif
/*
* load the LDT, if the LDT is different:
*/
- if (unlikely(prev->context.ldt != next->context.ldt))
+ if (unlikely(prev->context.size + next->context.size))
load_LDT_nolock(&next->context, cpu);
}
#ifdef CONFIG_SMP
@@ -56,7 +62,9 @@ static inline void switch_mm(struct mm_s
/* We were in lazy tlb mode and leave_mm disabled
* tlb flush IPI delivery. We must reload %cr3.
*/
+#if !defined(CONFIG_X86_SWITCH_PAGETABLES)
load_cr3(next->pgd);
+#endif
load_LDT_nolock(&next->context, cpu);
}
}
@@ -67,6 +75,6 @@ static inline void switch_mm(struct mm_s
asm("movl %0,%%fs ; movl %0,%%gs": :"r" (0))
#define activate_mm(prev, next) \
- switch_mm((prev),(next),NULL)
+ switch_mm((prev),(next),current)
#endif
--- diff/include/asm-i386/module.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/asm-i386/module.h 2004-03-16 11:27:48.486816544 +0000
@@ -60,6 +60,12 @@ struct mod_arch_specific
#define MODULE_REGPARM ""
#endif
-#define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_REGPARM
+#ifdef CONFIG_4KSTACKS
+#define MODULE_STACKSIZE "4KSTACKS "
+#else
+#define MODULE_STACKSIZE ""
+#endif
+
+#define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_REGPARM MODULE_STACKSIZE
#endif /* _ASM_I386_MODULE_H */
--- diff/include/asm-i386/page.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/asm-i386/page.h 2004-03-16 11:27:48.487816392 +0000
@@ -1,6 +1,8 @@
#ifndef _I386_PAGE_H
#define _I386_PAGE_H
+#include <linux/config.h>
+
/* PAGE_SHIFT determines the page size */
#define PAGE_SHIFT 12
#define PAGE_SIZE (1UL << PAGE_SHIFT)
@@ -9,11 +11,10 @@
#define LARGE_PAGE_MASK (~(LARGE_PAGE_SIZE-1))
#define LARGE_PAGE_SIZE (1UL << PMD_SHIFT)
-#ifdef __KERNEL__
-#ifndef __ASSEMBLY__
-
#include <linux/config.h>
+#ifdef __KERNEL__
+#ifndef __ASSEMBLY__
#ifdef CONFIG_X86_USE_3DNOW
#include <asm/mmx.h>
@@ -88,8 +89,19 @@ typedef struct { unsigned long pgprot; }
*
* If you want more physical memory than this then see the CONFIG_HIGHMEM4G
* and CONFIG_HIGHMEM64G options in the kernel configuration.
+ *
+ * Note: on PAE the kernel must never go below 32 MB, we use the
+ * first 8 entries of the 2-level boot pgd for PAE magic.
*/
+#ifdef CONFIG_X86_4G_VM_LAYOUT
+#define __PAGE_OFFSET (0x02000000)
+#define TASK_SIZE (0xff000000)
+#else
+#define __PAGE_OFFSET (0xc0000000)
+#define TASK_SIZE (0xc0000000)
+#endif
+
/*
* This much address space is reserved for vmalloc() and iomap()
* as well as fixmap mappings.
@@ -114,16 +126,10 @@ static __inline__ int get_order(unsigned
#endif /* __ASSEMBLY__ */
-#ifdef __ASSEMBLY__
-#define __PAGE_OFFSET (0xC0000000)
-#else
-#define __PAGE_OFFSET (0xC0000000UL)
-#endif
-
-
#define PAGE_OFFSET ((unsigned long)__PAGE_OFFSET)
#define VMALLOC_RESERVE ((unsigned long)__VMALLOC_RESERVE)
-#define MAXMEM (-__PAGE_OFFSET-__VMALLOC_RESERVE)
+#define __MAXMEM (-__PAGE_OFFSET-__VMALLOC_RESERVE)
+#define MAXMEM ((unsigned long)(-PAGE_OFFSET-VMALLOC_RESERVE))
#define __pa(x) ((unsigned long)(x)-PAGE_OFFSET)
#define __va(x) ((void *)((unsigned long)(x)+PAGE_OFFSET))
#define pfn_to_kaddr(pfn) __va((pfn) << PAGE_SHIFT)
--- diff/include/asm-i386/pgtable-2level.h 2003-05-21 10:49:56.000000000 +0000
+++ source/include/asm-i386/pgtable-2level.h 2004-03-16 11:27:48.487816392 +0000
@@ -64,15 +64,20 @@ static inline pmd_t * pmd_offset(pgd_t *
#define pfn_pmd(pfn, prot) __pmd(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
/*
- * Bits 0, 6 and 7 are taken, split up the 29 bits of offset
+ * Bits 0, 1, 6 and 7 are taken, split up the 28 bits of offset
* into this range:
*/
-#define PTE_FILE_MAX_BITS 29
+#define PTE_FILE_MAX_BITS 28
#define pte_to_pgoff(pte) \
- ((((pte).pte_low >> 1) & 0x1f ) + (((pte).pte_low >> 8) << 5 ))
+ ((((pte).pte_low >> 2) & 0xf ) + (((pte).pte_low >> 8) << 4 ))
+#define pte_to_pgprot(pte) \
+ __pgprot(((pte).pte_low & (_PAGE_RW | _PAGE_PROTNONE)) \
+ | (((pte).pte_low & _PAGE_PROTNONE) ? 0 : \
+ (_PAGE_USER | _PAGE_PRESENT)) | _PAGE_ACCESSED)
-#define pgoff_to_pte(off) \
- ((pte_t) { (((off) & 0x1f) << 1) + (((off) >> 5) << 8) + _PAGE_FILE })
+#define pgoff_prot_to_pte(off, prot) \
+ ((pte_t) { (((off) & 0xf) << 2) + (((off) >> 4) << 8) + \
+ (pgprot_val(prot) & (_PAGE_RW | _PAGE_PROTNONE)) + _PAGE_FILE })
#endif /* _I386_PGTABLE_2LEVEL_H */
--- diff/include/asm-i386/pgtable-3level.h 2003-08-20 13:16:33.000000000 +0000
+++ source/include/asm-i386/pgtable-3level.h 2004-03-16 11:27:48.487816392 +0000
@@ -120,7 +120,18 @@ static inline pmd_t pfn_pmd(unsigned lon
* put the 32 bits of offset into the high part.
*/
#define pte_to_pgoff(pte) ((pte).pte_high)
-#define pgoff_to_pte(off) ((pte_t) { _PAGE_FILE, (off) })
+
+#define pte_to_pgprot(pte) \
+ __pgprot(((pte).pte_low & (_PAGE_RW | _PAGE_PROTNONE)) \
+ | (((pte).pte_low & _PAGE_PROTNONE) ? 0 : \
+ (_PAGE_USER | _PAGE_PRESENT)) | _PAGE_ACCESSED)
+
+#define pgoff_prot_to_pte(off, prot) \
+ ((pte_t) { _PAGE_FILE + \
+ (pgprot_val(prot) & (_PAGE_RW | _PAGE_PROTNONE)) , (off) })
+
#define PTE_FILE_MAX_BITS 32
+extern struct kmem_cache_s *pae_pgd_cachep;
+
#endif /* _I386_PGTABLE_3LEVEL_H */
--- diff/include/asm-i386/pgtable.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/asm-i386/pgtable.h 2004-03-16 11:27:48.488816240 +0000
@@ -32,16 +32,20 @@
#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
extern unsigned long empty_zero_page[1024];
extern pgd_t swapper_pg_dir[1024];
-extern kmem_cache_t *pgd_cache;
-extern kmem_cache_t *pmd_cache;
-extern spinlock_t pgd_lock;
-extern struct list_head pgd_list;
+extern kmem_cache_t *pgd_cache, *pmd_cache, *kpmd_cache;
void pmd_ctor(void *, kmem_cache_t *, unsigned long);
+void kpmd_ctor(void *, kmem_cache_t *, unsigned long);
void pgd_ctor(void *, kmem_cache_t *, unsigned long);
-void pgd_dtor(void *, kmem_cache_t *, unsigned long);
void pgtable_cache_init(void);
void paging_init(void);
+void setup_identity_mappings(pgd_t *pgd_base, unsigned long start, unsigned long end);
+
+/*
+ * The size of the low 1:1 mappings that we use during bootup,
+ * SMP-boot and ACPI-sleep:
+ */
+#define LOW_MAPPINGS_SIZE (16*1024*1024)
#endif /* !__ASSEMBLY__ */
@@ -51,6 +55,11 @@ void paging_init(void);
* newer 3-level PAE-mode page tables.
*/
#ifndef __ASSEMBLY__
+
+extern void set_system_gate(unsigned int n, void *addr);
+extern void init_entry_mappings(void);
+extern void entry_trampoline_setup(void);
+
#ifdef CONFIG_X86_PAE
# include <asm/pgtable-3level.h>
#else
@@ -63,7 +72,12 @@ void paging_init(void);
#define PGDIR_SIZE (1UL << PGDIR_SHIFT)
#define PGDIR_MASK (~(PGDIR_SIZE-1))
-#define USER_PTRS_PER_PGD (TASK_SIZE/PGDIR_SIZE)
+#if defined(CONFIG_X86_PAE) && defined(CONFIG_X86_4G_VM_LAYOUT)
+# define USER_PTRS_PER_PGD 4
+#else
+# define USER_PTRS_PER_PGD ((TASK_SIZE/PGDIR_SIZE) + ((TASK_SIZE % PGDIR_SIZE) + PGDIR_SIZE-1)/PGDIR_SIZE)
+#endif
+
#define FIRST_USER_PGD_NR 0
#define USER_PGD_PTRS (PAGE_OFFSET >> PGDIR_SHIFT)
@@ -233,6 +247,7 @@ static inline void ptep_mkdirty(pte_t *p
#define mk_pte(page, pgprot) pfn_pte(page_to_pfn(page), (pgprot))
#define mk_pte_huge(entry) ((entry).pte_low |= _PAGE_PRESENT | _PAGE_PSE)
+#define mk_pte_phys(physpage, pgprot) pfn_pte((physpage) >> PAGE_SHIFT, pgprot)
static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
{
--- diff/include/asm-i386/processor.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-i386/processor.h 2004-03-16 11:27:48.489816088 +0000
@@ -63,6 +63,7 @@ struct cpuinfo_x86 {
int f00f_bug;
int coma_bug;
unsigned long loops_per_jiffy;
+ int x86_clflush_size; /* cache line size of L2 */
} __attribute__((__aligned__(SMP_CACHE_BYTES)));
#define X86_VENDOR_INTEL 0
@@ -97,6 +98,7 @@ extern struct cpuinfo_x86 cpu_data[];
extern char ignore_fpu_irq;
extern void identify_cpu(struct cpuinfo_x86 *);
+extern void late_identify_cpu(struct cpuinfo_x86 *c);
extern void print_cpu_info(struct cpuinfo_x86 *);
extern void dodgy_tsc(void);
@@ -291,11 +293,6 @@ extern unsigned int machine_submodel_id;
extern unsigned int BIOS_revision;
extern unsigned int mca_pentium_flag;
-/*
- * User space process size: 3GB (default).
- */
-#define TASK_SIZE (PAGE_OFFSET)
-
/* This decides where the kernel will search for a free chunk of vm
* space during mmap's.
*/
@@ -403,9 +400,18 @@ struct tss_struct {
unsigned long stack[64];
} __attribute__((packed));
+#define ARCH_MIN_TASKALIGN 16
+
+#ifdef CONFIG_4KSTACKS
+#define STACK_PAGE_COUNT (4096/PAGE_SIZE)
+#else
+#define STACK_PAGE_COUNT (8192/PAGE_SIZE) /* THREAD_SIZE/PAGE_SIZE */
+#endif
+
struct thread_struct {
/* cached TLS descriptors. */
struct desc_struct tls_array[GDT_ENTRY_TLS_ENTRIES];
+ void *stack_page[STACK_PAGE_COUNT];
unsigned long esp0;
unsigned long sysenter_cs;
unsigned long eip;
@@ -449,7 +455,8 @@ struct thread_struct {
.io_bitmap = { [ 0 ... IO_BITMAP_LONGS] = ~0 }, \
}
-static inline void load_esp0(struct tss_struct *tss, struct thread_struct *thread)
+static inline void
+load_esp0(struct tss_struct *tss, struct thread_struct *thread)
{
tss->esp0 = thread->esp0;
/* This can only happen when SEP is enabled, no need to test "SEP"arately */
@@ -485,6 +492,23 @@ extern void prepare_to_copy(struct task_
*/
extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
+#ifdef CONFIG_X86_HIGH_ENTRY
+#define virtual_esp0(tsk) \
+ ((unsigned long)(tsk)->thread_info->virtual_stack + ((tsk)->thread.esp0 - (unsigned long)(tsk)->thread_info->real_stack))
+#else
+# define virtual_esp0(tsk) ((tsk)->thread.esp0)
+#endif
+
+#define load_virtual_esp0(tss, task) \
+ do { \
+ tss->esp0 = virtual_esp0(task); \
+ if (likely(cpu_has_sep) && unlikely(tss->ss1 != task->thread.sysenter_cs)) { \
+ tss->ss1 = task->thread.sysenter_cs; \
+ wrmsr(MSR_IA32_SYSENTER_CS, \
+ task->thread.sysenter_cs, 0); \
+ } \
+ } while (0)
+
extern unsigned long thread_saved_pc(struct task_struct *tsk);
void show_trace(struct task_struct *task, unsigned long *stack);
@@ -646,4 +670,11 @@ extern inline void prefetchw(const void
extern void select_idle_routine(const struct cpuinfo_x86 *c);
+#ifdef CONFIG_SCHED_SMT
+#define ARCH_HAS_SCHED_DOMAIN
+#define ARCH_HAS_SCHED_WAKE_BALANCE
+#endif
+
+#define cache_line_size() (boot_cpu_data.x86_clflush_size)
+
#endif /* __ASM_I386_PROCESSOR_H */
--- diff/include/asm-i386/rwlock.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/asm-i386/rwlock.h 2004-03-16 11:27:48.490815936 +0000
@@ -20,28 +20,52 @@
#define RW_LOCK_BIAS 0x01000000
#define RW_LOCK_BIAS_STR "0x01000000"
-#define __build_read_lock_ptr(rw, helper) \
- asm volatile(LOCK "subl $1,(%0)\n\t" \
- "js 2f\n" \
- "1:\n" \
- LOCK_SECTION_START("") \
- "2:\tcall " helper "\n\t" \
- "jmp 1b\n" \
- LOCK_SECTION_END \
- ::"a" (rw) : "memory")
-
-#define __build_read_lock_const(rw, helper) \
- asm volatile(LOCK "subl $1,%0\n\t" \
- "js 2f\n" \
- "1:\n" \
- LOCK_SECTION_START("") \
- "2:\tpushl %%eax\n\t" \
- "leal %0,%%eax\n\t" \
- "call " helper "\n\t" \
- "popl %%eax\n\t" \
- "jmp 1b\n" \
- LOCK_SECTION_END \
- :"=m" (*(volatile int *)rw) : : "memory")
+#ifdef CONFIG_SPINLINE
+
+ #define __build_read_lock_ptr(rw, helper) \
+ asm volatile(LOCK "subl $1,(%0)\n\t" \
+ "jns 1f\n\t" \
+ "call " helper "\n\t" \
+ "1:\t" \
+ ::"a" (rw) : "memory")
+
+ #define __build_read_lock_const(rw, helper) \
+ asm volatile(LOCK "subl $1,%0\n\t" \
+ "jns 1f\n\t" \
+ "pushl %%eax\n\t" \
+ "leal %0,%%eax\n\t" \
+ "call " helper "\n\t" \
+ "popl %%eax\n\t" \
+ "1:\t" \
+ :"=m" (*(volatile int *)rw) : : "memory")
+
+#else /* !CONFIG_SPINLINE */
+
+ #define __build_read_lock_ptr(rw, helper) \
+ asm volatile(LOCK "subl $1,(%0)\n\t" \
+ "js 2f\n" \
+ "1:\n" \
+ LOCK_SECTION_START("") \
+ "2:\tcall " helper "\n\t" \
+ "jmp 1b\n" \
+ LOCK_SECTION_END \
+ ::"a" (rw) : "memory")
+
+ #define __build_read_lock_const(rw, helper) \
+ asm volatile(LOCK "subl $1,%0\n\t" \
+ "js 2f\n" \
+ "1:\n" \
+ LOCK_SECTION_START("") \
+ "2:\tpushl %%eax\n\t" \
+ "leal %0,%%eax\n\t" \
+ "call " helper "\n\t" \
+ "popl %%eax\n\t" \
+ "jmp 1b\n" \
+ LOCK_SECTION_END \
+ :"=m" (*(volatile int *)rw) : : "memory")
+
+#endif /* CONFIG_SPINLINE */
+
#define __build_read_lock(rw, helper) do { \
if (__builtin_constant_p(rw)) \
@@ -50,28 +74,51 @@
__build_read_lock_ptr(rw, helper); \
} while (0)
-#define __build_write_lock_ptr(rw, helper) \
- asm volatile(LOCK "subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \
- "jnz 2f\n" \
- "1:\n" \
- LOCK_SECTION_START("") \
- "2:\tcall " helper "\n\t" \
- "jmp 1b\n" \
- LOCK_SECTION_END \
- ::"a" (rw) : "memory")
-
-#define __build_write_lock_const(rw, helper) \
- asm volatile(LOCK "subl $" RW_LOCK_BIAS_STR ",%0\n\t" \
- "jnz 2f\n" \
- "1:\n" \
- LOCK_SECTION_START("") \
- "2:\tpushl %%eax\n\t" \
- "leal %0,%%eax\n\t" \
- "call " helper "\n\t" \
- "popl %%eax\n\t" \
- "jmp 1b\n" \
- LOCK_SECTION_END \
- :"=m" (*(volatile int *)rw) : : "memory")
+#ifdef CONFIG_SPINLINE
+
+ #define __build_write_lock_ptr(rw, helper) \
+ asm volatile(LOCK "subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \
+ "jz 1f\n\t" \
+ "call " helper "\n\t" \
+ "1:\n" \
+ ::"a" (rw) : "memory")
+
+ #define __build_write_lock_const(rw, helper) \
+ asm volatile(LOCK "subl $" RW_LOCK_BIAS_STR ",%0\n\t" \
+ "jz 1f\n\t" \
+ "pushl %%eax\n\t" \
+ "leal %0,%%eax\n\t" \
+ "call " helper "\n\t" \
+ "popl %%eax\n\t" \
+ "1:\n" \
+ :"=m" (*(volatile int *)rw) : : "memory")
+
+#else /* !CONFIG_SPINLINE */
+
+ #define __build_write_lock_ptr(rw, helper) \
+ asm volatile(LOCK "subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \
+ "jnz 2f\n" \
+ "1:\n" \
+ LOCK_SECTION_START("") \
+ "2:\tcall " helper "\n\t" \
+ "jmp 1b\n" \
+ LOCK_SECTION_END \
+ ::"a" (rw) : "memory")
+
+ #define __build_write_lock_const(rw, helper) \
+ asm volatile(LOCK "subl $" RW_LOCK_BIAS_STR ",%0\n\t" \
+ "jnz 2f\n" \
+ "1:\n" \
+ LOCK_SECTION_START("") \
+ "2:\tpushl %%eax\n\t" \
+ "leal %0,%%eax\n\t" \
+ "call " helper "\n\t" \
+ "popl %%eax\n\t" \
+ "jmp 1b\n" \
+ LOCK_SECTION_END \
+ :"=m" (*(volatile int *)rw) : : "memory")
+
+#endif /* CONFIG_SPINLINE */
#define __build_write_lock(rw, helper) do { \
if (__builtin_constant_p(rw)) \
--- diff/include/asm-i386/setup.h 2004-02-18 08:54:12.000000000 +0000
+++ source/include/asm-i386/setup.h 2004-03-16 11:27:48.490815936 +0000
@@ -16,10 +16,22 @@
#define MAXMEM_PFN PFN_DOWN(MAXMEM)
#define MAX_NONPAE_PFN (1 << 20)
+#define PARAM_SIZE 2048
+#define COMMAND_LINE_SIZE 256
+
+#define OLD_CL_MAGIC_ADDR 0x90020
+#define OLD_CL_MAGIC 0xA33F
+#define OLD_CL_BASE_ADDR 0x90000
+#define OLD_CL_OFFSET 0x90022
+#define NEW_CL_POINTER 0x228 /* Relative to real mode data */
+
+#ifndef __ASSEMBLY__
/*
* This is set up by the setup-routine at boot-time
*/
-#define PARAM ((unsigned char *)empty_zero_page)
+extern unsigned char boot_params[PARAM_SIZE];
+
+#define PARAM (boot_params)
#define SCREEN_INFO (*(struct screen_info *) (PARAM+0))
#define EXT_MEM_K (*(unsigned short *) (PARAM+2))
#define ALT_MEM_K (*(unsigned long *) (PARAM+0x1e0))
@@ -47,7 +59,7 @@
#define DISK80_SIGNATURE (*(unsigned int*) (PARAM+DISK80_SIG_BUFFER))
#define EDD_NR (*(unsigned char *) (PARAM+EDDNR))
#define EDD_BUF ((struct edd_info *) (PARAM+EDDBUF))
-#define COMMAND_LINE ((char *) (PARAM+2048))
-#define COMMAND_LINE_SIZE 256
+
+#endif /* __ASSEMBLY__ */
#endif /* _i386_SETUP_H */
--- diff/include/asm-i386/smp.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-i386/smp.h 2004-03-16 11:27:48.491815784 +0000
@@ -34,7 +34,7 @@
extern void smp_alloc_memory(void);
extern int pic_mode;
extern int smp_num_siblings;
-extern int cpu_sibling_map[];
+extern cpumask_t cpu_sibling_map[];
extern void smp_flush_tlb(void);
extern void smp_message_irq(int cpl, void *dev_id, struct pt_regs *regs);
--- diff/include/asm-i386/spinlock.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-i386/spinlock.h 2004-03-16 11:27:48.491815784 +0000
@@ -43,18 +43,35 @@ typedef struct {
#define spin_is_locked(x) (*(volatile signed char *)(&(x)->lock) <= 0)
#define spin_unlock_wait(x) do { barrier(); } while(spin_is_locked(x))
-#define spin_lock_string \
- "\n1:\t" \
- "lock ; decb %0\n\t" \
- "js 2f\n" \
- LOCK_SECTION_START("") \
- "2:\t" \
- "rep;nop\n\t" \
- "cmpb $0,%0\n\t" \
- "jle 2b\n\t" \
- "jmp 1b\n" \
- LOCK_SECTION_END
+#ifdef CONFIG_SPINLINE
+ #define spin_lock_string \
+ "\n1:\t" \
+ "lock ; decb %0\n\t" \
+ "js 2f\n" \
+ "jmp 3f\n" \
+ "2:\t" \
+ "rep;nop\n\t" \
+ "cmpb $0,%0\n\t" \
+ "jle 2b\n\t" \
+ "jmp 1b\n" \
+ "3:\t"
+
+#else /* !CONFIG_SPINLINE */
+
+ #define spin_lock_string \
+ "\n1:\t" \
+ "lock ; decb %0\n\t" \
+ "js 2f\n" \
+ LOCK_SECTION_START("") \
+ "2:\t" \
+ "rep;nop\n\t" \
+ "cmpb $0,%0\n\t" \
+ "jle 2b\n\t" \
+ "jmp 1b\n" \
+ LOCK_SECTION_END
+
+#endif /* CONFIG_SPINLINE */
/*
* This works. Despite all the confusion.
* (except on PPro SMP or if we are using OOSTORE)
@@ -138,6 +155,11 @@ here:
*/
typedef struct {
volatile unsigned int lock;
+#ifdef CONFIG_LOCKMETER
+ /* required for LOCKMETER since all bits in lock are used */
+ /* and we need this storage for CPU and lock INDEX */
+ unsigned lockmeter_magic;
+#endif
#ifdef CONFIG_DEBUG_SPINLOCK
unsigned magic;
#endif
@@ -145,11 +167,19 @@ typedef struct {
#define RWLOCK_MAGIC 0xdeaf1eed
+#ifdef CONFIG_LOCKMETER
+#ifdef CONFIG_DEBUG_SPINLOCK
+#define RWLOCK_MAGIC_INIT , 0, RWLOCK_MAGIC
+#else
+#define RWLOCK_MAGIC_INIT , 0
+#endif
+#else /* !CONFIG_LOCKMETER */
#ifdef CONFIG_DEBUG_SPINLOCK
#define RWLOCK_MAGIC_INIT , RWLOCK_MAGIC
#else
#define RWLOCK_MAGIC_INIT /* */
#endif
+#endif /* !CONFIG_LOCKMETER */
#define RW_LOCK_UNLOCKED (rwlock_t) { RW_LOCK_BIAS RWLOCK_MAGIC_INIT }
@@ -196,4 +226,60 @@ static inline int _raw_write_trylock(rwl
return 0;
}
+#ifdef CONFIG_LOCKMETER
+static inline int _raw_read_trylock(rwlock_t *lock)
+{
+/* FIXME -- replace with assembler */
+ atomic_t *count = (atomic_t *)lock;
+ atomic_dec(count);
+ if (count->counter > 0)
+ return 1;
+ atomic_inc(count);
+ return 0;
+}
+#endif
+
+#if defined(CONFIG_LOCKMETER) && defined(CONFIG_HAVE_DEC_LOCK)
+extern void _metered_spin_lock (spinlock_t *lock);
+extern void _metered_spin_unlock(spinlock_t *lock);
+
+/*
+ * Matches what is in arch/i386/lib/dec_and_lock.c, except this one is
+ * "static inline" so that the spin_lock(), if actually invoked, is charged
+ * against the real caller, not against the catch-all atomic_dec_and_lock
+ */
+static inline int atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock)
+{
+ int counter;
+ int newcount;
+
+repeat:
+ counter = atomic_read(atomic);
+ newcount = counter-1;
+
+ if (!newcount)
+ goto slow_path;
+
+ asm volatile("lock; cmpxchgl %1,%2"
+ :"=a" (newcount)
+ :"r" (newcount), "m" (atomic->counter), "0" (counter));
+
+ /* If the above failed, "eax" will have changed */
+ if (newcount != counter)
+ goto repeat;
+ return 0;
+
+slow_path:
+ preempt_disable();
+ _metered_spin_lock(lock);
+ if (atomic_dec_and_test(atomic))
+ return 1;
+ _metered_spin_unlock(lock);
+ preempt_enable();
+ return 0;
+}
+
+#define ATOMIC_DEC_AND_LOCK
+#endif
+
#endif /* __ASM_SPINLOCK_H */
--- diff/include/asm-i386/string.h 2004-02-18 08:54:12.000000000 +0000
+++ source/include/asm-i386/string.h 2004-03-16 11:27:48.492815632 +0000
@@ -58,6 +58,29 @@ __asm__ __volatile__(
return dest;
}
+/*
+ * This is a more generic variant of strncpy_count() suitable for
+ * implementing string-access routines with all sorts of return
+ * code semantics. It's used by mm/usercopy.c.
+ */
+static inline size_t strncpy_count(char * dest,const char *src,size_t count)
+{
+ __asm__ __volatile__(
+
+ "1:\tdecl %0\n\t"
+ "js 2f\n\t"
+ "lodsb\n\t"
+ "stosb\n\t"
+ "testb %%al,%%al\n\t"
+ "jne 1b\n\t"
+ "2:"
+ "incl %0"
+ : "=c" (count)
+ :"S" (src),"D" (dest),"0" (count) : "memory");
+
+ return count;
+}
+
static inline char * strcat(char * dest,const char * src)
{
int d0, d1, d2, d3;
--- diff/include/asm-i386/thread_info.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-i386/thread_info.h 2004-03-16 11:27:48.493815480 +0000
@@ -9,6 +9,9 @@
#ifdef __KERNEL__
+#include <linux/config.h>
+#include <asm/page.h>
+
#ifndef __ASSEMBLY__
#include <asm/processor.h>
#endif
@@ -29,31 +32,32 @@ struct thread_info {
__u32 cpu; /* current CPU */
__s32 preempt_count; /* 0 => preemptable, <0 => BUG */
+ unsigned long previous_esp; /* ESP of the previous stack in case
+ of nested (IRQ) stacks
+ */
+
mm_segment_t addr_limit; /* thread address space:
0-0xBFFFFFFF for user-thead
0-0xFFFFFFFF for kernel-thread
*/
- struct restart_block restart_block;
+ void *real_stack, *virtual_stack, *user_pgd;
+ struct restart_block restart_block;
__u8 supervisor_stack[0];
};
-#else /* !__ASSEMBLY__ */
-
-/* offsets into the thread_info struct for assembly code access */
-#define TI_TASK 0x00000000
-#define TI_EXEC_DOMAIN 0x00000004
-#define TI_FLAGS 0x00000008
-#define TI_STATUS 0x0000000C
-#define TI_CPU 0x00000010
-#define TI_PRE_COUNT 0x00000014
-#define TI_ADDR_LIMIT 0x00000018
-#define TI_RESTART_BLOCK 0x000001C
-
#endif
#define PREEMPT_ACTIVE 0x4000000
+/* if you change THREAD_SIZE here, don't forget to change STACK_PAGE_COUNT */
+#ifdef CONFIG_4KSTACKS
+#define THREAD_SIZE (4096)
+#else
+#define THREAD_SIZE (8192)
+#endif
+
+#define STACK_WARN (THREAD_SIZE/8)
/*
* macros/functions for gaining access to the thread information structure
*
@@ -61,7 +65,7 @@ struct thread_info {
*/
#ifndef __ASSEMBLY__
-#define INIT_THREAD_INFO(tsk) \
+#define INIT_THREAD_INFO(tsk, thread_info) \
{ \
.task = &tsk, \
.exec_domain = &default_exec_domain, \
@@ -72,12 +76,12 @@ struct thread_info {
.restart_block = { \
.fn = do_no_restart_syscall, \
}, \
+ .real_stack = &thread_info, \
}
#define init_thread_info (init_thread_union.thread_info)
#define init_stack (init_thread_union.stack)
-#define THREAD_SIZE (2*PAGE_SIZE)
/* how to get the thread information struct from C */
static inline struct thread_info *current_thread_info(void)
@@ -87,6 +91,14 @@ static inline struct thread_info *curren
return ti;
}
+/* how to get the current stack pointer from C */
+static inline unsigned long current_stack_pointer(void)
+{
+ unsigned long ti;
+ __asm__("movl %%esp,%0; ":"=r" (ti) : );
+ return ti;
+}
+
/* thread information allocation */
#ifdef CONFIG_DEBUG_STACK_USAGE
#define alloc_thread_info(tsk) \
@@ -108,8 +120,6 @@ static inline struct thread_info *curren
#else /* !__ASSEMBLY__ */
-#define THREAD_SIZE 8192
-
/* how to get the thread information struct from ASM */
#define GET_THREAD_INFO(reg) \
movl $-THREAD_SIZE, reg; \
@@ -133,6 +143,8 @@ static inline struct thread_info *curren
#define TIF_NEED_RESCHED 3 /* rescheduling necessary */
#define TIF_SINGLESTEP 4 /* restore singlestep on return to user mode */
#define TIF_IRET 5 /* return with iret */
+#define TIF_DB7 6 /* has debug registers */
+#define TIF_SYSCALL_AUDIT 7 /* syscall auditing active */
#define TIF_POLLING_NRFLAG 16 /* true if poll_idle() is polling TIF_NEED_RESCHED */
#define _TIF_SYSCALL_TRACE (1<<TIF_SYSCALL_TRACE)
@@ -141,9 +153,13 @@ static inline struct thread_info *curren
#define _TIF_NEED_RESCHED (1<<TIF_NEED_RESCHED)
#define _TIF_SINGLESTEP (1<<TIF_SINGLESTEP)
#define _TIF_IRET (1<<TIF_IRET)
+#define _TIF_SYSCALL_AUDIT (1<<TIF_SYSCALL_AUDIT)
+#define _TIF_DB7 (1<<TIF_DB7)
#define _TIF_POLLING_NRFLAG (1<<TIF_POLLING_NRFLAG)
-#define _TIF_WORK_MASK 0x0000FFFE /* work to do on interrupt/exception return */
+/* work to do on interrupt/exception return */
+#define _TIF_WORK_MASK \
+ (0x0000FFFF & ~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT))
#define _TIF_ALLWORK_MASK 0x0000FFFF /* work to do on any return to u-space */
/*
--- diff/include/asm-i386/tlbflush.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/asm-i386/tlbflush.h 2004-03-16 11:27:48.494815328 +0000
@@ -85,22 +85,28 @@ extern unsigned long pgkern_mask;
static inline void flush_tlb_mm(struct mm_struct *mm)
{
+#ifndef CONFIG_X86_SWITCH_PAGETABLES
if (mm == current->active_mm)
__flush_tlb();
+#endif
}
static inline void flush_tlb_page(struct vm_area_struct *vma,
unsigned long addr)
{
+#ifndef CONFIG_X86_SWITCH_PAGETABLES
if (vma->vm_mm == current->active_mm)
__flush_tlb_one(addr);
+#endif
}
static inline void flush_tlb_range(struct vm_area_struct *vma,
unsigned long start, unsigned long end)
{
+#ifndef CONFIG_X86_SWITCH_PAGETABLES
if (vma->vm_mm == current->active_mm)
__flush_tlb();
+#endif
}
#else
@@ -111,11 +117,10 @@ static inline void flush_tlb_range(struc
__flush_tlb()
extern void flush_tlb_all(void);
-extern void flush_tlb_current_task(void);
extern void flush_tlb_mm(struct mm_struct *);
extern void flush_tlb_page(struct vm_area_struct *, unsigned long);
-#define flush_tlb() flush_tlb_current_task()
+#define flush_tlb() flush_tlb_all()
static inline void flush_tlb_range(struct vm_area_struct * vma, unsigned long start, unsigned long end)
{
--- diff/include/asm-i386/uaccess.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/asm-i386/uaccess.h 2004-03-16 11:27:48.495815176 +0000
@@ -26,7 +26,7 @@
#define KERNEL_DS MAKE_MM_SEG(0xFFFFFFFFUL)
-#define USER_DS MAKE_MM_SEG(PAGE_OFFSET)
+#define USER_DS MAKE_MM_SEG(TASK_SIZE)
#define get_ds() (KERNEL_DS)
#define get_fs() (current_thread_info()->addr_limit)
@@ -149,6 +149,45 @@ extern void __get_user_4(void);
:"=a" (ret),"=d" (x) \
:"0" (ptr))
+extern int get_user_size(unsigned int size, void *val, const void *ptr);
+extern int put_user_size(unsigned int size, const void *val, void *ptr);
+extern int zero_user_size(unsigned int size, void *ptr);
+extern int copy_str_fromuser_size(unsigned int size, void *val, const void *ptr);
+extern int strlen_fromuser_size(unsigned int size, const void *ptr);
+
+
+# define indirect_get_user(x,ptr) \
+({ int __ret_gu,__val_gu; \
+ __typeof__(ptr) __ptr_gu = (ptr); \
+ __ret_gu = get_user_size(sizeof(*__ptr_gu), &__val_gu,__ptr_gu) ? -EFAULT : 0;\
+ (x) = (__typeof__(*__ptr_gu))__val_gu; \
+ __ret_gu; \
+})
+#define indirect_put_user(x,ptr) \
+({ \
+ __typeof__(*(ptr)) *__ptr_pu = (ptr), __x_pu = (x); \
+ put_user_size(sizeof(*__ptr_pu), &__x_pu, __ptr_pu) ? -EFAULT : 0; \
+})
+#define __indirect_put_user indirect_put_user
+#define __indirect_get_user indirect_get_user
+
+#define indirect_copy_from_user(to,from,n) get_user_size(n,to,from)
+#define indirect_copy_to_user(to,from,n) put_user_size(n,from,to)
+
+#define __indirect_copy_from_user indirect_copy_from_user
+#define __indirect_copy_to_user indirect_copy_to_user
+
+#define indirect_strncpy_from_user(dst, src, count) \
+ copy_str_fromuser_size(count, dst, src)
+
+extern int strlen_fromuser_size(unsigned int size, const void *ptr);
+#define indirect_strnlen_user(str, n) strlen_fromuser_size(n, str)
+#define indirect_strlen_user(str) indirect_strnlen_user(str, ~0UL >> 1)
+
+extern int zero_user_size(unsigned int size, void *ptr);
+
+#define indirect_clear_user(mem, len) zero_user_size(len, mem)
+#define __indirect_clear_user clear_user
/* Careful: we have to cast the result to the type of the pointer for sign reasons */
/**
@@ -168,7 +207,7 @@ extern void __get_user_4(void);
* Returns zero on success, or -EFAULT on error.
* On error, the variable @x is set to zero.
*/
-#define get_user(x,ptr) \
+#define direct_get_user(x,ptr) \
({ int __ret_gu,__val_gu; \
switch(sizeof (*(ptr))) { \
case 1: __get_user_x(1,__ret_gu,__val_gu,ptr); break; \
@@ -198,7 +237,7 @@ extern void __put_user_bad(void);
*
* Returns zero on success, or -EFAULT on error.
*/
-#define put_user(x,ptr) \
+#define direct_put_user(x,ptr) \
__put_user_check((__typeof__(*(ptr)))(x),(ptr),sizeof(*(ptr)))
@@ -222,7 +261,7 @@ extern void __put_user_bad(void);
* Returns zero on success, or -EFAULT on error.
* On error, the variable @x is set to zero.
*/
-#define __get_user(x,ptr) \
+#define __direct_get_user(x,ptr) \
__get_user_nocheck((x),(ptr),sizeof(*(ptr)))
@@ -245,7 +284,7 @@ extern void __put_user_bad(void);
*
* Returns zero on success, or -EFAULT on error.
*/
-#define __put_user(x,ptr) \
+#define __direct_put_user(x,ptr) \
__put_user_nocheck((__typeof__(*(ptr)))(x),(ptr),sizeof(*(ptr)))
#define __put_user_nocheck(x,ptr,size) \
@@ -396,7 +435,7 @@ unsigned long __copy_from_user_ll(void *
* On success, this will be zero.
*/
static inline unsigned long
-__copy_to_user(void __user *to, const void *from, unsigned long n)
+__direct_copy_to_user(void __user *to, const void *from, unsigned long n)
{
if (__builtin_constant_p(n)) {
unsigned long ret;
@@ -434,7 +473,7 @@ __copy_to_user(void __user *to, const vo
* data to the requested size using zero bytes.
*/
static inline unsigned long
-__copy_from_user(void *to, const void __user *from, unsigned long n)
+__direct_copy_from_user(void *to, const void __user *from, unsigned long n)
{
if (__builtin_constant_p(n)) {
unsigned long ret;
@@ -468,11 +507,11 @@ __copy_from_user(void *to, const void __
* On success, this will be zero.
*/
static inline unsigned long
-copy_to_user(void __user *to, const void *from, unsigned long n)
+direct_copy_to_user(void __user *to, const void *from, unsigned long n)
{
might_sleep();
if (access_ok(VERIFY_WRITE, to, n))
- n = __copy_to_user(to, from, n);
+ n = __direct_copy_to_user(to, from, n);
return n;
}
@@ -493,11 +532,11 @@ copy_to_user(void __user *to, const void
* data to the requested size using zero bytes.
*/
static inline unsigned long
-copy_from_user(void *to, const void __user *from, unsigned long n)
+direct_copy_from_user(void *to, const void __user *from, unsigned long n)
{
might_sleep();
if (access_ok(VERIFY_READ, from, n))
- n = __copy_from_user(to, from, n);
+ n = __direct_copy_from_user(to, from, n);
else
memset(to, 0, n);
return n;
@@ -520,10 +559,68 @@ long __strncpy_from_user(char *dst, cons
* If there is a limit on the length of a valid string, you may wish to
* consider using strnlen_user() instead.
*/
-#define strlen_user(str) strnlen_user(str, ~0UL >> 1)
-long strnlen_user(const char __user *str, long n);
-unsigned long clear_user(void __user *mem, unsigned long len);
-unsigned long __clear_user(void __user *mem, unsigned long len);
+long direct_strncpy_from_user(char *dst, const char *src, long count);
+long __direct_strncpy_from_user(char *dst, const char *src, long count);
+#define direct_strlen_user(str) direct_strnlen_user(str, ~0UL >> 1)
+long direct_strnlen_user(const char *str, long n);
+unsigned long direct_clear_user(void *mem, unsigned long len);
+unsigned long __direct_clear_user(void *mem, unsigned long len);
+
+extern int indirect_uaccess;
+
+#ifdef CONFIG_X86_UACCESS_INDIRECT
+
+/*
+ * Return code and zeroing semantics:
+
+ __clear_user 0 <-> bytes not done
+ clear_user 0 <-> bytes not done
+ __copy_to_user 0 <-> bytes not done
+ copy_to_user 0 <-> bytes not done
+ __copy_from_user 0 <-> bytes not done, zero rest
+ copy_from_user 0 <-> bytes not done, zero rest
+ __get_user 0 <-> -EFAULT
+ get_user 0 <-> -EFAULT
+ __put_user 0 <-> -EFAULT
+ put_user 0 <-> -EFAULT
+ strlen_user strlen + 1 <-> 0
+ strnlen_user strlen + 1 (or n+1) <-> 0
+ strncpy_from_user strlen (or n) <-> -EFAULT
+
+ */
+
+#define __clear_user(mem,len) __indirect_clear_user(mem,len)
+#define clear_user(mem,len) indirect_clear_user(mem,len)
+#define __copy_to_user(to,from,n) __indirect_copy_to_user(to,from,n)
+#define copy_to_user(to,from,n) indirect_copy_to_user(to,from,n)
+#define __copy_from_user(to,from,n) __indirect_copy_from_user(to,from,n)
+#define copy_from_user(to,from,n) indirect_copy_from_user(to,from,n)
+#define __get_user(val,ptr) __indirect_get_user(val,ptr)
+#define get_user(val,ptr) indirect_get_user(val,ptr)
+#define __put_user(val,ptr) __indirect_put_user(val,ptr)
+#define put_user(val,ptr) indirect_put_user(val,ptr)
+#define strlen_user(str) indirect_strlen_user(str)
+#define strnlen_user(src,count) indirect_strnlen_user(src,count)
+#define strncpy_from_user(dst,src,count) \
+ indirect_strncpy_from_user(dst,src,count)
+
+#else
+
+#define __clear_user __direct_clear_user
+#define clear_user direct_clear_user
+#define __copy_to_user __direct_copy_to_user
+#define copy_to_user direct_copy_to_user
+#define __copy_from_user __direct_copy_from_user
+#define copy_from_user direct_copy_from_user
+#define __get_user __direct_get_user
+#define get_user direct_get_user
+#define __put_user __direct_put_user
+#define put_user direct_put_user
+#define strlen_user direct_strlen_user
+#define strnlen_user direct_strnlen_user
+#define strncpy_from_user direct_strncpy_from_user
+
+#endif /* CONFIG_X86_UACCESS_INDIRECT */
#endif /* __i386_UACCESS_H */
--- diff/include/asm-i386/unistd.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/asm-i386/unistd.h 2004-03-16 11:27:48.496815024 +0000
@@ -262,7 +262,7 @@
#define __NR_epoll_create 254
#define __NR_epoll_ctl 255
#define __NR_epoll_wait 256
-#define __NR_remap_file_pages 257
+#define __NR_old_remap_file_pages 257
#define __NR_set_tid_address 258
#define __NR_timer_create 259
#define __NR_timer_settime (__NR_timer_create+1)
@@ -279,8 +279,15 @@
#define __NR_utimes 271
#define __NR_fadvise64_64 272
#define __NR_vserver 273
+#define __NR_mq_open 274
+#define __NR_mq_unlink (__NR_mq_open+1)
+#define __NR_mq_timedsend (__NR_mq_open+2)
+#define __NR_mq_timedreceive (__NR_mq_open+3)
+#define __NR_mq_notify (__NR_mq_open+4)
+#define __NR_mq_getsetattr (__NR_mq_open+5)
+#define __NR_remap_file_pages 280
-#define NR_syscalls 274
+#define NR_syscalls 281
/* user-visible error numbers are in the range -1 - -124: see <asm-i386/errno.h> */
--- diff/include/asm-ia64/compat.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/asm-ia64/compat.h 2004-03-16 11:27:48.497814872 +0000
@@ -27,6 +27,7 @@ typedef u16 compat_ipc_pid_t;
typedef s32 compat_daddr_t;
typedef u32 compat_caddr_t;
typedef __kernel_fsid_t compat_fsid_t;
+typedef u32 compat_timer_t;
typedef s32 compat_int_t;
typedef s32 compat_long_t;
--- diff/include/asm-ia64/ia32.h 2004-01-19 10:22:59.000000000 +0000
+++ source/include/asm-ia64/ia32.h 2004-03-16 11:27:48.497814872 +0000
@@ -6,7 +6,7 @@
#include <asm/ptrace.h>
#include <asm/signal.h>
-#define IA32_NR_syscalls 275 /* length of syscall table */
+#define IA32_NR_syscalls 281 /* length of syscall table */
#ifndef __ASSEMBLY__
--- diff/include/asm-ia64/irq.h 2003-05-21 10:50:16.000000000 +0000
+++ source/include/asm-ia64/irq.h 2004-03-16 11:27:48.498814720 +0000
@@ -29,4 +29,8 @@ extern void disable_irq_nosync (unsigned
extern void enable_irq (unsigned int);
extern void set_irq_affinity_info (unsigned int irq, int dest, int redir);
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif /* _ASM_IA64_IRQ_H */
--- diff/include/asm-ia64/mman.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/asm-ia64/mman.h 2004-03-16 11:27:48.498814720 +0000
@@ -30,6 +30,7 @@
#define MAP_NORESERVE 0x04000 /* don't check for reservations */
#define MAP_POPULATE 0x08000 /* populate (prefault) pagetables */
#define MAP_NONBLOCK 0x10000 /* do not block on IO */
+#define MAP_INHERIT 0x20000 /* inherit prot of underlying vma */
#define MS_ASYNC 1 /* sync memory asynchronously */
#define MS_INVALIDATE 2 /* invalidate the caches */
--- diff/include/asm-ia64/pgtable.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/asm-ia64/pgtable.h 2004-03-16 11:27:48.499814568 +0000
@@ -433,7 +433,8 @@ extern void paging_init (void);
* Format of file pte:
* bit 0 : present bit (must be zero)
* bit 1 : _PAGE_FILE (must be one)
- * bits 2-62: file_offset/PAGE_SIZE
+ * bit 2 : _PAGE_AR_RW
+ * bits 3-62: file_offset/PAGE_SIZE
* bit 63 : _PAGE_PROTNONE bit
*/
#define __swp_type(entry) (((entry).val >> 2) & 0x7f)
@@ -442,9 +443,17 @@ extern void paging_init (void);
#define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) })
#define __swp_entry_to_pte(x) ((pte_t) { (x).val })
-#define PTE_FILE_MAX_BITS 61
-#define pte_to_pgoff(pte) ((pte_val(pte) << 1) >> 3)
-#define pgoff_to_pte(off) ((pte_t) { ((off) << 2) | _PAGE_FILE })
+#define PTE_FILE_MAX_BITS 59
+#define pte_to_pgoff(pte) ((pte_val(pte) << 1) >> 4)
+
+#define pte_to_pgprot(pte) \
+ __pgprot((pte_val(pte) & (_PAGE_AR_RW | _PAGE_PROTNONE)) \
+ | ((pte_val(pte) & _PAGE_PROTNONE) ? 0 : \
+ (__ACCESS_BITS | _PAGE_PL_3)))
+
+#define pgoff_prot_to_pte(off, prot) \
+ ((pte_t) { _PAGE_FILE + \
+ (pgprot_val(prot) & (_PAGE_AR_RW | _PAGE_PROTNONE)) + (off) })
#define io_remap_page_range remap_page_range /* XXX is this right? */
--- diff/include/asm-ia64/spinlock.h 2004-01-19 10:22:59.000000000 +0000
+++ source/include/asm-ia64/spinlock.h 2004-03-16 11:27:48.500814416 +0000
@@ -110,8 +110,18 @@ do { \
typedef struct {
volatile int read_counter : 31;
volatile int write_lock : 1;
+#ifdef CONFIG_LOCKMETER
+ /* required for LOCKMETER since all bits in lock are used */
+ /* and we need this storage for CPU and lock INDEX */
+ unsigned lockmeter_magic;
+#endif
} rwlock_t;
+
+#ifdef CONFIG_LOCKMETER
+#define RW_LOCK_UNLOCKED (rwlock_t) { 0, 0, 0 }
+#else
#define RW_LOCK_UNLOCKED (rwlock_t) { 0, 0 }
+#endif
#define rwlock_init(x) do { *(x) = RW_LOCK_UNLOCKED; } while(0)
#define rwlock_is_locked(x) (*(volatile int *) (x) != 0)
@@ -127,6 +137,48 @@ do { \
} \
} while (0)
+#ifdef CONFIG_LOCKMETER
+/*
+ * HACK: This works, but still have a timing window that affects performance:
+ * we see that no one owns the Write lock, then someone * else grabs for Write
+ * lock before we do a read_lock().
+ * This means that on rare occasions our read_lock() will stall and spin-wait
+ * until we acquire for Read, instead of simply returning a trylock failure.
+ */
+static inline int _raw_read_trylock(rwlock_t *rw)
+{
+ if (rw->write_lock) {
+ return 0;
+ } else {
+ _raw_read_lock(rw);
+ return 1;
+ }
+}
+
+static inline int _raw_write_trylock(rwlock_t *rw)
+{
+ if (!(rw->write_lock)) {
+ /* isn't currently write-locked... that looks promising... */
+ if (test_and_set_bit(31, rw) == 0) {
+ /* now it is write-locked by me... */
+ if (rw->read_counter) {
+ /* really read-locked, so release write-lock and fail */
+ clear_bit(31, rw);
+ } else {
+ /* we've the the write-lock, no read-lockers... success! */
+ barrier();
+ return 1;
+ }
+
+ }
+ }
+
+ /* falls through ... fails to write-lock */
+ barrier();
+ return 0;
+}
+#endif
+
#define _raw_read_unlock(rw) \
do { \
rwlock_t *__read_lock_ptr = (rw); \
@@ -190,4 +242,25 @@ do { \
clear_bit(31, (x)); \
})
+#ifdef CONFIG_LOCKMETER
+extern void _metered_spin_lock (spinlock_t *lock);
+extern void _metered_spin_unlock(spinlock_t *lock);
+
+/*
+ * Use a less efficient, and inline, atomic_dec_and_lock() if lockmetering
+ * so we can see the callerPC of who is actually doing the spin_lock().
+ * Otherwise, all we see is the generic rollup of all locks done by
+ * atomic_dec_and_lock().
+ */
+static inline int atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock)
+{
+ _metered_spin_lock(lock);
+ if (atomic_dec_and_test(atomic))
+ return 1;
+ _metered_spin_unlock(lock);
+ return 0;
+}
+#define ATOMIC_DEC_AND_LOCK
+#endif
+
#endif /* _ASM_IA64_SPINLOCK_H */
--- diff/include/asm-ia64/tlb.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-ia64/tlb.h 2004-03-16 11:27:48.500814416 +0000
@@ -211,6 +211,8 @@ __tlb_remove_tlb_entry (struct mmu_gathe
tlb->end_addr = address + PAGE_SIZE;
}
+#define tlb_migrate_prepare(mm) flush_tlb_mm(mm)
+
#define tlb_start_vma(tlb, vma) do { } while (0)
#define tlb_end_vma(tlb, vma) do { } while (0)
--- diff/include/asm-m68k/irq.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/asm-m68k/irq.h 2004-03-16 11:27:48.500814416 +0000
@@ -124,4 +124,8 @@ extern volatile unsigned int num_spuriou
*/
extern irq_node_t *new_irq_node(void);
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif /* _M68K_IRQ_H_ */
--- diff/include/asm-m68knommu/irq.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-m68knommu/irq.h 2004-03-16 11:27:48.501814264 +0000
@@ -121,4 +121,8 @@ extern irq_node_t *new_irq_node(void);
#define enable_irq_nosync(x) enable_irq(x)
#define disable_irq_nosync(x) disable_irq(x)
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif /* _M68K_IRQ_H_ */
--- diff/include/asm-mips/irq.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-mips/irq.h 2004-03-16 11:27:48.501814264 +0000
@@ -31,4 +31,7 @@ extern asmlinkage unsigned int do_IRQ(in
extern void init_generic_irq(void);
+struct irqaction;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif /* _ASM_IRQ_H */
--- diff/include/asm-mips/spinlock.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-mips/spinlock.h 2004-03-16 11:27:48.502814112 +0000
@@ -91,9 +91,18 @@ static inline unsigned int _raw_spin_try
typedef struct {
volatile unsigned int lock;
+#ifdef CONFIG_LOCKMETER
+ /* required for LOCKMETER since all bits in lock are used */
+ /* and we need this storage for CPU and lock INDEX */
+ unsigned lockmeter_magic;
+#endif
} rwlock_t;
+#ifdef CONFIG_LOCKMETER
+#define RW_LOCK_UNLOCKED (rwlock_t) { 0, 0 }
+#else
#define RW_LOCK_UNLOCKED (rwlock_t) { 0 }
+#endif
#define rwlock_init(x) do { *(x) = RW_LOCK_UNLOCKED; } while(0)
--- diff/include/asm-parisc/irq.h 2003-10-09 08:47:34.000000000 +0000
+++ source/include/asm-parisc/irq.h 2004-03-16 11:27:48.503813960 +0000
@@ -96,4 +96,7 @@ extern unsigned long txn_alloc_addr(int)
/* soft power switch support (power.c) */
extern struct tasklet_struct power_tasklet;
+struct irqaction;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif /* _ASM_PARISC_IRQ_H */
--- diff/include/asm-ppc/irq.h 2003-09-30 14:46:19.000000000 +0000
+++ source/include/asm-ppc/irq.h 2004-03-16 11:27:48.511812744 +0000
@@ -211,5 +211,9 @@ extern unsigned long ppc_cached_irq_mask
extern unsigned long ppc_lost_interrupts[NR_MASK_WORDS];
extern atomic_t ppc_n_lost_interrupts;
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif /* _ASM_IRQ_H */
#endif /* __KERNEL__ */
--- diff/include/asm-ppc/mman.h 2003-09-30 14:46:19.000000000 +0000
+++ source/include/asm-ppc/mman.h 2004-03-16 11:27:48.511812744 +0000
@@ -23,6 +23,7 @@
#define MAP_EXECUTABLE 0x1000 /* mark it as an executable */
#define MAP_POPULATE 0x8000 /* populate (prefault) pagetables */
#define MAP_NONBLOCK 0x10000 /* do not block on IO */
+#define MAP_INHERIT 0x20000 /* inherit prot of underlying vma */
#define MS_ASYNC 1 /* sync memory asynchronously */
#define MS_INVALIDATE 2 /* invalidate the caches */
--- diff/include/asm-ppc/pgtable.h 2004-02-18 08:54:12.000000000 +0000
+++ source/include/asm-ppc/pgtable.h 2004-03-16 11:27:48.512812592 +0000
@@ -264,8 +264,8 @@ extern unsigned long ioremap_bot, iorema
/* Definitions for 60x, 740/750, etc. */
#define _PAGE_PRESENT 0x001 /* software: pte contains a translation */
#define _PAGE_HASHPTE 0x002 /* hash_page has made an HPTE for this pte */
-#define _PAGE_FILE 0x004 /* when !present: nonlinear file mapping */
#define _PAGE_USER 0x004 /* usermode access allowed */
+#define _PAGE_FILE 0x008 /* when !present: nonlinear file mapping */
#define _PAGE_GUARDED 0x008 /* G: prohibit speculative access */
#define _PAGE_COHERENT 0x010 /* M: enforce memory coherence (SMP systems) */
#define _PAGE_NO_CACHE 0x020 /* I: cache inhibit */
@@ -628,9 +628,16 @@ extern void add_hash_page(unsigned conte
#define __swp_entry_to_pte(x) ((pte_t) { (x).val << 3 })
/* Encode and decode a nonlinear file mapping entry */
-#define PTE_FILE_MAX_BITS 29
-#define pte_to_pgoff(pte) (pte_val(pte) >> 3)
-#define pgoff_to_pte(off) ((pte_t) { ((off) << 3) | _PAGE_FILE })
+#define PTE_FILE_MAX_BITS 27
+#define pte_to_pgoff(pte) (((pte_val(pte) & ~0x7ff) >> 5) \
+ | ((pte_val(pte) & 0x3f0) >> 4))
+#define pte_to_pgprot(pte) \
+__pgprot((pte_val(pte) & (_PAGE_USER|_PAGE_RW|_PAGE_PRESENT)) | _PAGE_ACCESSED)
+
+#define pgoff_prot_to_pte(off, prot) \
+ ((pte_t) { (((off) << 5) & ~0x7ff) | (((off) << 4) & 0x3f0) \
+ | (pgprot_val(prot) & (_PAGE_USER|_PAGE_RW)) \
+ | _PAGE_FILE })
/* CONFIG_APUS */
/* For virtual address to physical address conversion */
--- diff/include/asm-ppc/unistd.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-ppc/unistd.h 2004-03-16 11:27:48.513812440 +0000
@@ -415,10 +415,10 @@ int sys_vfork(int p1, int p2, int p3, in
int sys_pipe(int __user *fildes);
int sys_ptrace(long request, long pid, long addr, long data);
struct sigaction;
-asmlinkage long sys_rt_sigaction(int sig,
- const struct sigaction __user *act,
- struct sigaction __user *oact,
- size_t sigsetsize);
+long sys_rt_sigaction(int sig,
+ const struct sigaction __user *act,
+ struct sigaction __user *oact,
+ size_t sigsetsize);
#endif /* __KERNEL_SYSCALLS__ */
--- diff/include/asm-ppc64/compat.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/asm-ppc64/compat.h 2004-03-16 11:27:48.503813960 +0000
@@ -25,6 +25,7 @@ typedef u16 compat_ipc_pid_t;
typedef s32 compat_daddr_t;
typedef u32 compat_caddr_t;
typedef __kernel_fsid_t compat_fsid_t;
+typedef u32 compat_timer_t;
typedef s32 compat_int_t;
typedef s32 compat_long_t;
--- diff/include/asm-ppc64/dma-mapping.h 2002-12-30 10:17:13.000000000 +0000
+++ source/include/asm-ppc64/dma-mapping.h 2004-03-16 11:27:48.504813808 +0000
@@ -1 +1,75 @@
-#include <asm-generic/dma-mapping.h>
+/* Copyright (C) 2004 IBM
+ *
+ * Implements the generic device dma API for ppc64. Handles
+ * the pci and vio busses
+ */
+
+#ifndef _ASM_DMA_MAPPING_H
+#define _ASM_DMA_MAPPING_H
+
+#include <linux/types.h>
+#include <linux/device.h>
+#include <linux/cache.h>
+/* need struct page definitions */
+#include <linux/mm.h>
+#include <asm/scatterlist.h>
+#include <asm/bug.h>
+
+extern int dma_supported(struct device *dev, u64 mask);
+extern int dma_set_mask(struct device *dev, u64 dma_mask);
+extern void *dma_alloc_coherent(struct device *dev, size_t size,
+ dma_addr_t *dma_handle, int flag);
+extern void dma_free_coherent(struct device *dev, size_t size, void *cpu_addr,
+ dma_addr_t dma_handle);
+extern dma_addr_t dma_map_single(struct device *dev, void *cpu_addr,
+ size_t size, enum dma_data_direction direction);
+extern void dma_unmap_single(struct device *dev, dma_addr_t dma_addr,
+ size_t size, enum dma_data_direction direction);
+extern dma_addr_t dma_map_page(struct device *dev, struct page *page,
+ unsigned long offset, size_t size,
+ enum dma_data_direction direction);
+extern void dma_unmap_page(struct device *dev, dma_addr_t dma_address,
+ size_t size, enum dma_data_direction direction);
+extern int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
+ enum dma_data_direction direction);
+extern void dma_unmap_sg(struct device *dev, struct scatterlist *sg,
+ int nhwentries, enum dma_data_direction direction);
+extern void dma_sync_single(struct device *dev, dma_addr_t dma_handle,
+ size_t size, enum dma_data_direction direction);
+extern void dma_sync_sg(struct device *dev, struct scatterlist *sg, int nelems,
+ enum dma_data_direction direction);
+
+/* Now for the API extensions over the pci_ one */
+
+#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
+#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
+#define dma_is_consistent(d) (1)
+
+static inline int
+dma_get_cache_alignment(void)
+{
+ /* no easy way to get cache size on all processors, so return
+ * the maximum possible, to be safe */
+ return (1 << L1_CACHE_SHIFT_MAX);
+}
+
+static inline void
+dma_sync_single_range(struct device *dev, dma_addr_t dma_handle,
+ unsigned long offset, size_t size,
+ enum dma_data_direction direction)
+{
+ /* just sync everything, that's all the pci API can do */
+ dma_sync_single(dev, dma_handle, offset+size, direction);
+}
+
+static inline void
+dma_cache_sync(void *vaddr, size_t size,
+ enum dma_data_direction direction)
+{
+ /* could define this in terms of the dma_cache ... operations,
+ * but if you get this on a platform, you should convert the platform
+ * to using the generic device DMA API */
+ BUG();
+}
+
+#endif /* _ASM_DMA_MAPPING_H */
--- diff/include/asm-ppc64/eeh.h 2004-02-18 08:54:12.000000000 +0000
+++ source/include/asm-ppc64/eeh.h 2004-03-16 11:27:48.505813656 +0000
@@ -17,15 +17,11 @@
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
-/* Start Change Log
- * 2001/10/27 : engebret : Created.
- * End Change Log
- */
-
-#ifndef _EEH_H
-#define _EEH_H
+#ifndef _PPC64_EEH_H
+#define _PPC64_EEH_H
#include <linux/string.h>
+#include <linux/init.h>
struct pci_dev;
@@ -33,22 +29,43 @@ struct pci_dev;
* a bad page fault if the address is used directly (i.e. these addresses are
* never actually mapped. Translation between IO <-> EEH region is 1 to 1.
*/
-#define IO_TOKEN_TO_ADDR(token) (((unsigned long)(token) & ~(0xfUL << REGION_SHIFT)) | \
- (IO_REGION_ID << REGION_SHIFT))
-#define IO_ADDR_TO_TOKEN(addr) (((unsigned long)(addr) & ~(0xfUL << REGION_SHIFT)) | \
- (EEH_REGION_ID << REGION_SHIFT))
+#define IO_TOKEN_TO_ADDR(token) \
+ (((unsigned long)(token) & ~(0xfUL << REGION_SHIFT)) | \
+ (IO_REGION_ID << REGION_SHIFT))
+
+#define IO_ADDR_TO_TOKEN(addr) \
+ (((unsigned long)(addr) & ~(0xfUL << REGION_SHIFT)) | \
+ (EEH_REGION_ID << REGION_SHIFT))
/* Values for eeh_mode bits in device_node */
#define EEH_MODE_SUPPORTED (1<<0)
#define EEH_MODE_NOCHECK (1<<1)
-/* This is for profiling only */
-extern unsigned long eeh_total_mmio_ffs;
-
-void eeh_init(void);
-int eeh_get_state(unsigned long ea);
+extern void __init eeh_init(void);
unsigned long eeh_check_failure(void *token, unsigned long val);
void *eeh_ioremap(unsigned long addr, void *vaddr);
+void __init pci_addr_cache_build(void);
+
+/**
+ * eeh_add_device - perform EEH initialization for the indicated pci device
+ * @dev: pci device for which to set up EEH
+ *
+ * This routine can be used to perform EEH initialization for PCI
+ * devices that were added after system boot (e.g. hotplug, dlpar).
+ * Whether this actually enables EEH or not for this device depends
+ * on the type of the device, on earlier boot command-line
+ * arguments & etc.
+ */
+void eeh_add_device(struct pci_dev *);
+
+/**
+ * eeh_remove_device - undo EEH setup for the indicated pci device
+ * @dev: pci device to be removed
+ *
+ * This routine should be when a device is removed from a running
+ * system (e.g. by hotplug or dlpar).
+ */
+void eeh_remove_device(struct pci_dev *);
#define EEH_DISABLE 0
#define EEH_ENABLE 1
@@ -56,18 +73,8 @@ void *eeh_ioremap(unsigned long addr, vo
#define EEH_RELEASE_DMA 3
int eeh_set_option(struct pci_dev *dev, int options);
-/* Given a PCI device check if eeh should be configured or not.
- * This may look at firmware properties and/or kernel cmdline options.
- */
-int is_eeh_configured(struct pci_dev *dev);
-
-/* Translate a (possible) eeh token to a physical addr.
- * If "token" is not an eeh token it is simply returned under
- * the assumption that it is already a physical addr.
- */
-unsigned long eeh_token_to_phys(unsigned long token);
-
-/* EEH_POSSIBLE_ERROR() -- test for possible MMIO failure.
+/*
+ * EEH_POSSIBLE_ERROR() -- test for possible MMIO failure.
*
* Order this macro for performance.
* If EEH is off for a device and it is a memory BAR, ioremap will
@@ -78,30 +85,22 @@ unsigned long eeh_token_to_phys(unsigned
* If this macro yields TRUE, the caller relays to eeh_check_failure()
* which does further tests out of line.
*/
-/* #define EEH_POSSIBLE_IO_ERROR(val) (~(val) == 0) */
-/* #define EEH_POSSIBLE_ERROR(addr, vaddr, val) ((vaddr) != (addr) && EEH_POSSIBLE_IO_ERROR(val) */
-/* This version is rearranged to collect some profiling data */
-#define EEH_POSSIBLE_IO_ERROR(val) (~(val) == 0 && ++eeh_total_mmio_ffs)
-#define EEH_POSSIBLE_ERROR(addr, vaddr, val) (EEH_POSSIBLE_IO_ERROR(val) && (vaddr) != (addr))
+#define EEH_POSSIBLE_IO_ERROR(val, type) ((val) == (type)~0)
+
+/* The vaddr will equal the addr if EEH checking is disabled for
+ * this device. This is because eeh_ioremap() will not have
+ * remapped to 0xA0, and thus both vaddr and addr will be 0xE0...
+ */
+#define EEH_POSSIBLE_ERROR(addr, vaddr, val, type) \
+ ((vaddr) != (addr) && EEH_POSSIBLE_IO_ERROR(val, type))
/*
* MMIO read/write operations with EEH support.
- *
- * addr: 64b token of the form 0xA0PPBBDDyyyyyyyy
- * 0xA0 : Unmapped MMIO region
- * PP : PHB index (starting at zero)
- * BB : PCI Bus number under given PHB
- * DD : PCI devfn under given bus
- * yyyyyyyy : Virtual address offset
- *
- * An actual virtual address is produced from this token
- * by masking into the form:
- * 0xE0000000yyyyyyyy
*/
static inline u8 eeh_readb(void *addr) {
volatile u8 *vaddr = (volatile u8 *)IO_TOKEN_TO_ADDR(addr);
u8 val = in_8(vaddr);
- if (EEH_POSSIBLE_ERROR(addr, vaddr, val))
+ if (EEH_POSSIBLE_ERROR(addr, vaddr, val, u8))
return eeh_check_failure(addr, val);
return val;
}
@@ -109,10 +108,11 @@ static inline void eeh_writeb(u8 val, vo
volatile u8 *vaddr = (volatile u8 *)IO_TOKEN_TO_ADDR(addr);
out_8(vaddr, val);
}
+
static inline u16 eeh_readw(void *addr) {
volatile u16 *vaddr = (volatile u16 *)IO_TOKEN_TO_ADDR(addr);
u16 val = in_le16(vaddr);
- if (EEH_POSSIBLE_ERROR(addr, vaddr, val))
+ if (EEH_POSSIBLE_ERROR(addr, vaddr, val, u16))
return eeh_check_failure(addr, val);
return val;
}
@@ -120,10 +120,22 @@ static inline void eeh_writew(u16 val, v
volatile u16 *vaddr = (volatile u16 *)IO_TOKEN_TO_ADDR(addr);
out_le16(vaddr, val);
}
+static inline u16 eeh_raw_readw(void *addr) {
+ volatile u16 *vaddr = (volatile u16 *)IO_TOKEN_TO_ADDR(addr);
+ u16 val = in_be16(vaddr);
+ if (EEH_POSSIBLE_ERROR(addr, vaddr, val, u16))
+ return eeh_check_failure(addr, val);
+ return val;
+}
+static inline void eeh_raw_writew(u16 val, void *addr) {
+ volatile u16 *vaddr = (volatile u16 *)IO_TOKEN_TO_ADDR(addr);
+ out_be16(vaddr, val);
+}
+
static inline u32 eeh_readl(void *addr) {
volatile u32 *vaddr = (volatile u32 *)IO_TOKEN_TO_ADDR(addr);
u32 val = in_le32(vaddr);
- if (EEH_POSSIBLE_ERROR(addr, vaddr, val))
+ if (EEH_POSSIBLE_ERROR(addr, vaddr, val, u32))
return eeh_check_failure(addr, val);
return val;
}
@@ -131,10 +143,22 @@ static inline void eeh_writel(u32 val, v
volatile u32 *vaddr = (volatile u32 *)IO_TOKEN_TO_ADDR(addr);
out_le32(vaddr, val);
}
+static inline u32 eeh_raw_readl(void *addr) {
+ volatile u32 *vaddr = (volatile u32 *)IO_TOKEN_TO_ADDR(addr);
+ u32 val = in_be32(vaddr);
+ if (EEH_POSSIBLE_ERROR(addr, vaddr, val, u32))
+ return eeh_check_failure(addr, val);
+ return val;
+}
+static inline void eeh_raw_writel(u32 val, void *addr) {
+ volatile u32 *vaddr = (volatile u32 *)IO_TOKEN_TO_ADDR(addr);
+ out_be32(vaddr, val);
+}
+
static inline u64 eeh_readq(void *addr) {
volatile u64 *vaddr = (volatile u64 *)IO_TOKEN_TO_ADDR(addr);
u64 val = in_le64(vaddr);
- if (EEH_POSSIBLE_ERROR(addr, vaddr, val))
+ if (EEH_POSSIBLE_ERROR(addr, vaddr, val, u64))
return eeh_check_failure(addr, val);
return val;
}
@@ -142,6 +166,17 @@ static inline void eeh_writeq(u64 val, v
volatile u64 *vaddr = (volatile u64 *)IO_TOKEN_TO_ADDR(addr);
out_le64(vaddr, val);
}
+static inline u64 eeh_raw_readq(void *addr) {
+ volatile u64 *vaddr = (volatile u64 *)IO_TOKEN_TO_ADDR(addr);
+ u64 val = in_be64(vaddr);
+ if (EEH_POSSIBLE_ERROR(addr, vaddr, val, u64))
+ return eeh_check_failure(addr, val);
+ return val;
+}
+static inline void eeh_raw_writeq(u64 val, void *addr) {
+ volatile u64 *vaddr = (volatile u64 *)IO_TOKEN_TO_ADDR(addr);
+ out_be64(vaddr, val);
+}
static inline void eeh_memset_io(void *addr, int c, unsigned long n) {
void *vaddr = (void *)IO_TOKEN_TO_ADDR(addr);
@@ -150,8 +185,15 @@ static inline void eeh_memset_io(void *a
static inline void eeh_memcpy_fromio(void *dest, void *src, unsigned long n) {
void *vsrc = (void *)IO_TOKEN_TO_ADDR(src);
memcpy(dest, vsrc, n);
- /* look for ffff's here at dest[n] */
+ /* Look for ffff's here at dest[n]. Assume that at least 4 bytes
+ * were copied. Check all four bytes.
+ */
+ if ((n >= 4) &&
+ (EEH_POSSIBLE_ERROR(src, vsrc, (*((u32 *) dest+n-4)), u32))) {
+ eeh_check_failure(src, (*((u32 *) dest+n-4)));
+ }
}
+
static inline void eeh_memcpy_toio(void *dest, void *src, unsigned long n) {
void *vdest = (void *)IO_TOKEN_TO_ADDR(dest);
memcpy(vdest, src, n);
@@ -169,8 +211,8 @@ static inline u8 eeh_inb(unsigned long p
if (_IO_IS_ISA(port) && !_IO_HAS_ISA_BUS)
return ~0;
val = in_8((u8 *)(port+pci_io_base));
- if (!_IO_IS_ISA(port) && EEH_POSSIBLE_IO_ERROR(val))
- return eeh_check_failure((void*)(port+pci_io_base), val);
+ if (!_IO_IS_ISA(port) && EEH_POSSIBLE_IO_ERROR(val, u8))
+ return eeh_check_failure((void*)(port), val);
return val;
}
@@ -184,8 +226,8 @@ static inline u16 eeh_inw(unsigned long
if (_IO_IS_ISA(port) && !_IO_HAS_ISA_BUS)
return ~0;
val = in_le16((u16 *)(port+pci_io_base));
- if (!_IO_IS_ISA(port) && EEH_POSSIBLE_IO_ERROR(val))
- return eeh_check_failure((void*)(port+pci_io_base), val);
+ if (!_IO_IS_ISA(port) && EEH_POSSIBLE_IO_ERROR(val, u16))
+ return eeh_check_failure((void*)(port), val);
return val;
}
@@ -199,8 +241,8 @@ static inline u32 eeh_inl(unsigned long
if (_IO_IS_ISA(port) && !_IO_HAS_ISA_BUS)
return ~0;
val = in_le32((u32 *)(port+pci_io_base));
- if (!_IO_IS_ISA(port) && EEH_POSSIBLE_IO_ERROR(val))
- return eeh_check_failure((void*)(port+pci_io_base), val);
+ if (!_IO_IS_ISA(port) && EEH_POSSIBLE_IO_ERROR(val, u32))
+ return eeh_check_failure((void*)(port), val);
return val;
}
@@ -209,4 +251,23 @@ static inline void eeh_outl(u32 val, uns
return out_le32((u32 *)(port+pci_io_base), val);
}
-#endif /* _EEH_H */
+/* in-string eeh macros */
+static inline void eeh_insb(unsigned long port, void * buf, int ns) {
+ _insb((u8 *)(port+pci_io_base), buf, ns);
+ if (!_IO_IS_ISA(port) && EEH_POSSIBLE_IO_ERROR((*(((u8*)buf)+ns-1)), u8))
+ eeh_check_failure((void*)(port), *(u8*)buf);
+}
+
+static inline void eeh_insw_ns(unsigned long port, void * buf, int ns) {
+ _insw_ns((u16 *)(port+pci_io_base), buf, ns);
+ if (!_IO_IS_ISA(port) && EEH_POSSIBLE_IO_ERROR((*(((u16*)buf)+ns-1)), u16))
+ eeh_check_failure((void*)(port), *(u16*)buf);
+}
+
+static inline void eeh_insl_ns(unsigned long port, void * buf, int nl) {
+ _insl_ns((u32 *)(port+pci_io_base), buf, nl);
+ if (!_IO_IS_ISA(port) && EEH_POSSIBLE_IO_ERROR((*(((u32*)buf)+nl-1)), u32))
+ eeh_check_failure((void*)(port), *(u32*)buf);
+}
+
+#endif /* _PPC64_EEH_H */
--- diff/include/asm-ppc64/hvcall.h 2004-02-18 08:54:12.000000000 +0000
+++ source/include/asm-ppc64/hvcall.h 2004-03-16 11:27:48.505813656 +0000
@@ -4,6 +4,14 @@
#define H_Success 0
#define H_Busy 1 /* Hardware busy -- retry later */
#define H_Constrained 4 /* Resource request constrained to max allowed */
+#define H_LongBusyStartRange 9900 /* Start of long busy range */
+#define H_LongBusyOrder1msec 9900 /* Long busy, hint that 1msec is a good time to retry */
+#define H_LongBusyOrder10msec 9901 /* Long busy, hint that 10msec is a good time to retry */
+#define H_LongBusyOrder100msec 9902 /* Long busy, hint that 100msec is a good time to retry */
+#define H_LongBusyOrder1sec 9903 /* Long busy, hint that 1sec is a good time to retry */
+#define H_LongBusyOrder10sec 9904 /* Long busy, hint that 10sec is a good time to retry */
+#define H_LongBusyOrder100sec 9905 /* Long busy, hint that 100sec is a good time to retry */
+#define H_LongBusyEndRange 9905 /* End of long busy range */
#define H_Hardware -1 /* Hardware error */
#define H_Function -2 /* Function not supported */
#define H_Privilege -3 /* Caller not privileged */
@@ -21,6 +29,16 @@
#define H_RemoteParm -15
#define H_Resource -16
+/* Long Busy is a condition that can be returned by the firmware
+ * when a call cannot be completed now, but the identical call
+ * should be retried later. This prevents calls blocking in the
+ * firmware for long periods of time. Annoyingly the firmware can return
+ * a range of return codes, hinting at how long we should wait before
+ * retrying. If you don't care for the hint, the macro below is a good
+ * way to check for the long_busy return codes
+ */
+#define H_isLongBusy(x) ((x >= H_LongBusyStartRange) && (x <= H_LongBusyEndRange))
+
/* Flags */
#define H_LARGE_PAGE (1UL<<(63-16))
#define H_EXACT (1UL<<(63-24)) /* Use exact PTE or return H_PTEG_FULL */
@@ -76,7 +94,7 @@
#define H_PROD 0xE8
#define H_GET_PPP 0xEC
#define H_SET_PPP 0xF0
-#define H_SET_PURR 0xF4
+#define H_PURR 0xF4
#define H_PIC 0xF8
#define H_REG_CRQ 0xFC
#define H_FREE_CRQ 0x100
@@ -84,6 +102,9 @@
#define H_SEND_CRQ 0x108
#define H_COPY_RDMA 0x110
#define H_POLL_PENDING 0x1D8
+#define H_VTERM_PARTNER_INFO 0x150
+#define H_REGISTER_VTERM 0x154
+#define H_FREE_VTERM 0x158
/* plpar_hcall() -- Generic call interface using above opcodes
*
--- diff/include/asm-ppc64/iSeries/iSeries_proc.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/asm-ppc64/iSeries/iSeries_proc.h 2004-03-16 11:27:48.507813352 +0000
@@ -19,12 +19,6 @@
#ifndef _ISERIES_PROC_H
#define _ISERIES_PROC_H
-#include <linux/proc_fs.h>
-
extern void iSeries_proc_early_init(void);
-typedef void (*iSeriesProcFunction)(struct proc_dir_entry *iSeries_proc);
-
-extern void iSeries_proc_callback(iSeriesProcFunction initFunction);
-
#endif /* _iSeries_PROC_H */
--- diff/include/asm-ppc64/iSeries/mf.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/asm-ppc64/iSeries/mf.h 2004-03-16 11:27:48.507813352 +0000
@@ -67,6 +67,4 @@ extern int mf_getRtcTime(unsigned long *
extern int mf_getRtc( struct rtc_time * tm );
extern int mf_setRtc( struct rtc_time * tm );
-extern void mf_proc_init(struct proc_dir_entry *iSeries_proc);
-
#endif /* MF_H_INCLUDED */
--- diff/include/asm-ppc64/io.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-ppc64/io.h 2004-03-16 11:27:48.506813504 +0000
@@ -58,6 +58,13 @@ extern unsigned long pci_io_base;
#define outb(data,addr) writeb(data,((unsigned long)(addr)))
#define outw(data,addr) writew(data,((unsigned long)(addr)))
#define outl(data,addr) writel(data,((unsigned long)(addr)))
+/*
+ * The *_ns versions below don't do byte-swapping.
+ * Neither do the standard versions now, these are just here
+ * for older code.
+ */
+#define insw_ns(port, buf, ns) _insw_ns((u16 *)((port)+pci_io_base), (buf), (ns))
+#define insl_ns(port, buf, nl) _insl_ns((u32 *)((port)+pci_io_base), (buf), (nl))
#else
#define __raw_readb(addr) (*(volatile unsigned char *)(addr))
#define __raw_readw(addr) (*(volatile unsigned short *)(addr))
@@ -90,12 +97,16 @@ extern unsigned long pci_io_base;
* They are only used in practice for transferring buffers which
* are arrays of bytes, and byte-swapping is not appropriate in
* that case. - paulus */
-#define insb(port, buf, ns) _insb((u8 *)((port)+pci_io_base), (buf), (ns))
-#define outsb(port, buf, ns) _outsb((u8 *)((port)+pci_io_base), (buf), (ns))
-#define insw(port, buf, ns) _insw_ns((u16 *)((port)+pci_io_base), (buf), (ns))
-#define outsw(port, buf, ns) _outsw_ns((u16 *)((port)+pci_io_base), (buf), (ns))
-#define insl(port, buf, nl) _insl_ns((u32 *)((port)+pci_io_base), (buf), (nl))
-#define outsl(port, buf, nl) _outsl_ns((u32 *)((port)+pci_io_base), (buf), (nl))
+#define insb(port, buf, ns) eeh_insb((port), (buf), (ns))
+#define insw(port, buf, ns) eeh_insw_ns((port), (buf), (ns))
+#define insl(port, buf, nl) eeh_insl_ns((port), (buf), (nl))
+#define insw_ns(port, buf, ns) eeh_insw_ns((port), (buf), (ns))
+#define insl_ns(port, buf, nl) eeh_insl_ns((port), (buf), (nl))
+
+#define outsb(port, buf, ns) _outsb((u8 *)((port)+pci_io_base), (buf), (ns))
+#define outsw(port, buf, ns) _outsw_ns((u16 *)((port)+pci_io_base), (buf), (ns))
+#define outsl(port, buf, nl) _outsl_ns((u32 *)((port)+pci_io_base), (buf), (nl))
+
#endif
#define readb_relaxed(addr) readb(addr)
@@ -130,9 +141,7 @@ extern void _outsl_ns(volatile u32 *port
* Neither do the standard versions now, these are just here
* for older code.
*/
-#define insw_ns(port, buf, ns) _insw_ns((u16 *)((port)+pci_io_base), (buf), (ns))
#define outsw_ns(port, buf, ns) _outsw_ns((u16 *)((port)+pci_io_base), (buf), (ns))
-#define insl_ns(port, buf, nl) _insl_ns((u32 *)((port)+pci_io_base), (buf), (nl))
#define outsl_ns(port, buf, nl) _outsl_ns((u32 *)((port)+pci_io_base), (buf), (nl))
@@ -204,6 +213,9 @@ static inline void iosync(void)
/*
* 8, 16 and 32 bit, big and little endian I/O operations, with barrier.
+ * These routines do not perform EEH-related I/O address translation,
+ * and should not be used directly by device drivers. Use inb/readb
+ * instead.
*/
static inline int in_8(volatile unsigned char *addr)
{
--- diff/include/asm-ppc64/irq.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-ppc64/irq.h 2004-03-16 11:27:48.506813504 +0000
@@ -48,5 +48,9 @@ static __inline__ int irq_canonicalize(i
#define NR_MASK_WORDS ((NR_IRQS + 63) / 64)
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif /* _ASM_IRQ_H */
#endif /* __KERNEL__ */
--- diff/include/asm-ppc64/pmc.h 2002-10-16 03:28:24.000000000 +0000
+++ source/include/asm-ppc64/pmc.h 2004-03-16 11:27:48.507813352 +0000
@@ -1,113 +0,0 @@
-/*
- * pmc.h
- * Copyright (C) 2001 Dave Engebretsen & Mike Corrigan IBM Corporation.
- *
- * The PPC64 PMC subsystem encompases both the hardware PMC registers and
- * a set of software event counters. An interface is provided via the
- * proc filesystem which can be used to access this subsystem.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- */
-
-/* Start Change Log
- * 2001/06/05 : engebret : Created.
- * End Change Log
- */
-
-#ifndef _PPC64_TYPES_H
-#include <asm/types.h>
-#endif
-
-#ifndef _PMC_H
-#define _PMC_H
-
-#define STAB_ENTRY_MAX 64
-
-struct _pmc_hw
-{
- u64 mmcr0;
- u64 mmcr1;
- u64 mmcra;
-
- u64 pmc1;
- u64 pmc2;
- u64 pmc3;
- u64 pmc4;
- u64 pmc5;
- u64 pmc6;
- u64 pmc7;
- u64 pmc8;
-};
-
-struct _pmc_sw
-{
- u64 stab_faults; /* Count of faults on the stab */
- u64 stab_capacity_castouts;/* Count of castouts from the stab */
- u64 stab_invalidations; /* Count of invalidations from the */
- /* stab, not including castouts */
- u64 stab_entry_use[STAB_ENTRY_MAX];
-
- u64 htab_primary_overflows;
- u64 htab_capacity_castouts;
- u64 htab_read_to_write_fault;
-};
-
-#define PMC_HW_TEXT_ENTRY_COUNT (sizeof(struct _pmc_hw) / sizeof(u64))
-#define PMC_SW_TEXT_ENTRY_COUNT (sizeof(struct _pmc_sw) / sizeof(u64))
-#define PMC_TEXT_ENTRY_SIZE 64
-
-struct _pmc_sw_text {
- char buffer[PMC_SW_TEXT_ENTRY_COUNT * PMC_TEXT_ENTRY_SIZE];
-};
-
-struct _pmc_hw_text {
- char buffer[PMC_HW_TEXT_ENTRY_COUNT * PMC_TEXT_ENTRY_SIZE];
-};
-
-extern struct _pmc_sw pmc_sw_system;
-extern struct _pmc_sw pmc_sw_cpu[];
-
-extern struct _pmc_sw_text pmc_sw_text;
-extern struct _pmc_hw_text pmc_hw_text;
-extern char *ppc64_pmc_stab(int file);
-extern char *ppc64_pmc_htab(int file);
-extern char *ppc64_pmc_hw(int file);
-
-#if 1
-#define PMC_SW_PROCESSOR(F) pmc_sw_cpu[smp_processor_id()].F++
-#define PMC_SW_PROCESSOR_A(F, E) (pmc_sw_cpu[smp_processor_id()].F[(E)])++
-#define PMC_SW_SYSTEM(F) pmc_sw_system.F++
-#else
-#define PMC_SW_PROCESSOR(F) do {;} while (0)
-#define PMC_SW_PROCESSOR_A(F) do {;} while (0)
-#define PMC_SW_SYSTEM(F) do {;} while (0)
-#endif
-
-#define MMCR0 795
-#define MMCR1 798
-#define MMCRA 786
-#define PMC1 787
-#define PMC2 788
-#define PMC3 789
-#define PMC4 790
-#define PMC5 791
-#define PMC6 792
-#define PMC7 793
-#define PMC8 794
-
-#define PMC_CONTROL_CPI 1
-#define PMC_CONTROL_TLB 2
-
-#endif /* _PMC_H */
--- diff/include/asm-ppc64/ppc32.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/asm-ppc64/ppc32.h 2004-03-16 11:27:48.508813200 +0000
@@ -2,6 +2,7 @@
#define _PPC64_PPC32_H
#include <linux/compat.h>
+#include <linux/compat_siginfo.h>
#include <asm/siginfo.h>
#include <asm/signal.h>
@@ -40,55 +41,6 @@
/* These are here to support 32-bit syscalls on a 64-bit kernel. */
-typedef struct compat_siginfo {
- int si_signo;
- int si_errno;
- int si_code;
-
- union {
- int _pad[SI_PAD_SIZE32];
-
- /* kill() */
- struct {
- compat_pid_t _pid; /* sender's pid */
- compat_uid_t _uid; /* sender's uid */
- } _kill;
-
- /* POSIX.1b timers */
- struct {
- unsigned int _timer1;
- unsigned int _timer2;
- } _timer;
-
- /* POSIX.1b signals */
- struct {
- compat_pid_t _pid; /* sender's pid */
- compat_uid_t _uid; /* sender's uid */
- compat_sigval_t _sigval;
- } _rt;
-
- /* SIGCHLD */
- struct {
- compat_pid_t _pid; /* which child */
- compat_uid_t _uid; /* sender's uid */
- int _status; /* exit code */
- compat_clock_t _utime;
- compat_clock_t _stime;
- } _sigchld;
-
- /* SIGILL, SIGFPE, SIGSEGV, SIGBUS, SIGEMT */
- struct {
- unsigned int _addr; /* faulting insn/memory ref. */
- } _sigfault;
-
- /* SIGPOLL */
- struct {
- int _band; /* POLL_IN, POLL_OUT, POLL_MSG */
- int _fd;
- } _sigpoll;
- } _sifields;
-} compat_siginfo_t;
-
#define __old_sigaction32 old_sigaction32
struct __old_sigaction32 {
@@ -141,20 +93,6 @@ struct ucontext32 {
struct mcontext32 uc_mcontext;
};
-typedef struct compat_sigevent {
- compat_sigval_t sigev_value;
- int sigev_signo;
- int sigev_notify;
- union {
- int _pad[SIGEV_PAD_SIZE];
- int _tid;
- struct {
- compat_uptr_t _function;
- compat_uptr_t _attribute;
- } _sigev_thread;
- } _sigev_un;
-} compat_sigevent_t;
-
struct ipc_kludge_32 {
unsigned int msgp;
int msgtyp;
--- diff/include/asm-ppc64/proc_fs.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/asm-ppc64/proc_fs.h 2004-03-16 11:27:48.509813048 +0000
@@ -1,39 +0,0 @@
-#ifndef _PPC64_PROC_FS_H
-#define _PPC64_PROC_FS_H
-/*
- * proc_fs.h
- * Copyright (C) 2001 Mike Corrigan IBM Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- */
-
-/* Change Activity: */
-/* tgall -- merge of iSeries/iSeries_proc.h and proc_pmc.h */
-/* End Change Activity */
-
-#include <linux/proc_fs.h>
-
-struct proc_ppc64_t {
- struct proc_dir_entry *root;
- struct proc_dir_entry *naca;
- struct proc_dir_entry *paca;
- struct proc_dir_entry *systemcfg;
- struct proc_dir_entry *rtas;
-};
-
-extern struct proc_ppc64_t proc_ppc64;
-extern int proc_ppc64_init(void);
-
-#endif /* _PPC64_PROC_FS_H */
--- diff/include/asm-ppc64/proc_pmc.h 2002-10-16 03:28:30.000000000 +0000
+++ source/include/asm-ppc64/proc_pmc.h 2004-03-16 11:27:48.509813048 +0000
@@ -1,33 +0,0 @@
-/*
- * pmc_proc.h
- * Copyright (C) 2001 Mike Corrigan IBM Corporation
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- */
-
-
-/* Change Activity: */
-/* End Change Activity */
-
-#ifndef _PMC_PROC_H
-#define _PMC_PROC_H
-
-#include <linux/proc_fs.h>
-
-void pmc_proc_init(struct proc_dir_entry *iSeries_proc);
-void proc_ppc64_init(void);
-
-#endif /* _PMC_PROC_H */
-
--- diff/include/asm-ppc64/processor.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-ppc64/processor.h 2004-03-16 11:27:48.509813048 +0000
@@ -618,6 +618,11 @@ static inline void prefetchw(const void
#define spin_lock_prefetch(x) prefetchw(x)
+#ifdef CONFIG_SCHED_SMT
+#define ARCH_HAS_SCHED_DOMAIN
+#define ARCH_HAS_SCHED_WAKE_BALANCE
+#endif
+
#endif /* ASSEMBLY */
#endif /* __ASM_PPC64_PROCESSOR_H */
--- diff/include/asm-ppc64/rtas.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/asm-ppc64/rtas.h 2004-03-16 11:27:48.509813048 +0000
@@ -177,6 +177,7 @@ extern void rtas_power_off(void);
extern void rtas_halt(void);
extern int rtas_get_sensor(int sensor, int index, int *state);
extern int rtas_get_power_level(int powerdomain, int *level);
+extern int rtas_set_power_level(int powerdomain, int level, int *setlevel);
extern int rtas_set_indicator(int indicator, int index, int new_value);
/* Given an RTAS status code of 9900..9905 compute the hinted delay */
--- diff/include/asm-ppc64/thread_info.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-ppc64/thread_info.h 2004-03-16 11:27:48.510812896 +0000
@@ -97,6 +97,7 @@ static inline struct thread_info *curren
#define TIF_32BIT 5 /* 32 bit binary */
#define TIF_RUN_LIGHT 6 /* iSeries run light */
#define TIF_ABI_PENDING 7 /* 32/64 bit switch needed */
+#define TIF_SYSCALL_AUDIT 8 /* syscall auditing active */
/* as above, but as bit values */
#define _TIF_SYSCALL_TRACE (1<<TIF_SYSCALL_TRACE)
@@ -107,6 +108,8 @@ static inline struct thread_info *curren
#define _TIF_32BIT (1<<TIF_32BIT)
#define _TIF_RUN_LIGHT (1<<TIF_RUN_LIGHT)
#define _TIF_ABI_PENDING (1<<TIF_ABI_PENDING)
+#define _TIF_SYSCALL_AUDIT (1<<TIF_SYSCALL_AUDIT)
+#define _TIF_SYSCALL_T_OR_A (_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT)
#define _TIF_USER_WORK_MASK (_TIF_NOTIFY_RESUME | _TIF_SIGPENDING | \
_TIF_NEED_RESCHED)
--- diff/include/asm-ppc64/vio.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-ppc64/vio.h 2004-03-16 11:27:48.511812744 +0000
@@ -17,6 +17,7 @@
#include <linux/init.h>
#include <linux/errno.h>
#include <linux/device.h>
+#include <linux/pci.h>
#include <asm/hvcall.h>
#include <asm/prom.h>
#include <asm/scatterlist.h>
@@ -44,8 +45,11 @@ int vio_register_driver(struct vio_drive
void vio_unregister_driver(struct vio_driver *drv);
const struct vio_device_id * vio_match_device(const struct vio_device_id *ids,
const struct vio_dev *dev);
+
struct vio_dev * __devinit vio_register_device(struct device_node *node_vdev);
void __devinit vio_unregister_device(struct vio_dev *dev);
+struct vio_dev *vio_find_node(struct device_node *vnode);
+
const void * vio_get_attribute(struct vio_dev *vdev, void* which, int* length);
int vio_get_irq(struct vio_dev *dev);
struct iommu_table * vio_build_iommu_table(struct vio_dev *dev);
@@ -65,6 +69,33 @@ void *vio_alloc_consistent(struct vio_de
void vio_free_consistent(struct vio_dev *dev, size_t size, void *vaddr,
dma_addr_t dma_handle);
+static inline int vio_dma_supported(struct vio_dev *hwdev, u64 mask)
+{
+ return 1;
+}
+
+#define vio_map_page(dev, page, off, size, dir) \
+ vio_map_single(dev, (page_address(page) + (off)), size, dir)
+#define vio_unmap_page(dev,addr,sz,dir) vio_unmap_single(dev,addr,sz,dir)
+
+
+static inline void vio_dma_sync_single(struct vio_dev *hwdev,
+ dma_addr_t dma_handle,
+ size_t size, int direction)
+{
+ BUG_ON(direction == PCI_DMA_NONE);
+ /* nothing to do */
+}
+
+static inline void vio_dma_sync_sg(struct vio_dev *hwdev,
+ struct scatterlist *sg,
+ int nelems, int direction)
+{
+ BUG_ON(direction == PCI_DMA_NONE);
+ /* nothing to do */
+}
+static inline int vio_set_dma_mask(struct vio_dev *dev, u64 mask) { return -EIO; }
+
extern struct bus_type vio_bus_type;
struct vio_device_id {
--- diff/include/asm-s390/irq.h 2003-08-20 13:16:14.000000000 +0000
+++ source/include/asm-s390/irq.h 2004-03-16 11:27:48.514812288 +0000
@@ -21,6 +21,10 @@ enum interruption_class {
#define touch_nmi_watchdog() do { } while(0)
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif /* __KERNEL__ */
#endif
--- diff/include/asm-s390/mman.h 2003-09-30 14:46:20.000000000 +0000
+++ source/include/asm-s390/mman.h 2004-03-16 11:27:48.514812288 +0000
@@ -30,6 +30,7 @@
#define MAP_NORESERVE 0x4000 /* don't check for reservations */
#define MAP_POPULATE 0x8000 /* populate (prefault) pagetables */
#define MAP_NONBLOCK 0x10000 /* do not block on IO */
+#define MAP_INHERIT 0x20000 /* inherit the protection bits of the underlying vma*/
#define MS_ASYNC 1 /* sync memory asynchronously */
#define MS_INVALIDATE 2 /* invalidate the caches */
--- diff/include/asm-s390/pgtable.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/asm-s390/pgtable.h 2004-03-16 11:27:48.515812136 +0000
@@ -209,16 +209,41 @@ extern char empty_zero_page[PAGE_SIZE];
* C : changed bit
*/
-/* Hardware bits in the page table entry */
+/* Hardware bits in the page table entry. */
#define _PAGE_RO 0x200 /* HW read-only */
#define _PAGE_INVALID 0x400 /* HW invalid */
-/* Mask and four different kinds of invalid pages. */
-#define _PAGE_INVALID_MASK 0x601
+/* Software bits in the page table entry. */
+#define _PAGE_FILE 0x001
+#define _PAGE_PROTNONE 0x002
+
+/*
+ * We have 8 different page "types", two valid types and 6 invalid types
+ * (p = page address, o = swap offset, t = swap type, f = file offset):
+ * 0 xxxxxxxxxxxxxxxxxxx 0IP0 yyyyyy NF
+ * valid rw: 0 <--------p--------> 0000 <--0-> 00
+ * valid ro: 0 <--------p--------> 0010 <--0-> 00
+ * invalid none: 0 <--------p--------> 0100 <--0-> 10
+ * invalid empty: 0 <--------0--------> 0100 <--0-> 00
+ * invalid swap: 0 <--------o--------> 0110 <--t-> 00
+ * invalid file rw: 0 <--------f--------> 0100 <--f-> 01
+ * invalid file ro: 0 <--------f--------> 0110 <--f-> 01
+ * invaild file none: 0 <--------f--------> 0100 <--f-> 11
+ *
+ * The format for 64 bit is almost identical, there isn't a leading zero
+ * and the number of bits in the page address part of the pte is 52 bits
+ * instead of 19.
+ */
+
#define _PAGE_INVALID_EMPTY 0x400
-#define _PAGE_INVALID_NONE 0x401
#define _PAGE_INVALID_SWAP 0x600
-#define _PAGE_INVALID_FILE 0x601
+#define _PAGE_INVALID_FILE 0x401
+
+#define _PTE_IS_VALID(__pte) (!(pte_val(__pte) & _PAGE_INVALID))
+#define _PTE_IS_NONE(__pte) ((pte_val(__pte) & 0x603) == 0x402)
+#define _PTE_IS_EMPTY(__pte) ((pte_val(__pte) & 0x603) == 0x400)
+#define _PTE_IS_SWAP(__pte) ((pte_val(__pte) & 0x603) == 0x600)
+#define _PTE_IS_FILE(__pte) ((pte_val(__pte) & 0x401) == 0x401)
#ifndef __s390x__
@@ -279,13 +304,11 @@ extern char empty_zero_page[PAGE_SIZE];
/*
* No mapping available
*/
-#define PAGE_NONE_SHARED __pgprot(_PAGE_INVALID_NONE)
-#define PAGE_NONE_PRIVATE __pgprot(_PAGE_INVALID_NONE)
-#define PAGE_RO_SHARED __pgprot(_PAGE_RO)
-#define PAGE_RO_PRIVATE __pgprot(_PAGE_RO)
-#define PAGE_COPY __pgprot(_PAGE_RO)
-#define PAGE_SHARED __pgprot(0)
-#define PAGE_KERNEL __pgprot(0)
+#define PAGE_NONE __pgprot(_PAGE_INVALID | _PAGE_PROTNONE)
+#define PAGE_READONLY __pgprot(_PAGE_RO)
+#define PAGE_COPY __pgprot(_PAGE_RO)
+#define PAGE_SHARED __pgprot(0)
+#define PAGE_KERNEL __pgprot(0)
/*
* The S390 can't do page protection for execute, and considers that the
@@ -293,21 +316,21 @@ extern char empty_zero_page[PAGE_SIZE];
* the closest we can get..
*/
/*xwr*/
-#define __P000 PAGE_NONE_PRIVATE
-#define __P001 PAGE_RO_PRIVATE
+#define __P000 PAGE_NONE
+#define __P001 PAGE_READONLY
#define __P010 PAGE_COPY
#define __P011 PAGE_COPY
-#define __P100 PAGE_RO_PRIVATE
-#define __P101 PAGE_RO_PRIVATE
+#define __P100 PAGE_READONLY
+#define __P101 PAGE_READONLY
#define __P110 PAGE_COPY
#define __P111 PAGE_COPY
-#define __S000 PAGE_NONE_SHARED
-#define __S001 PAGE_RO_SHARED
+#define __S000 PAGE_NONE
+#define __S001 PAGE_READONLY
#define __S010 PAGE_SHARED
#define __S011 PAGE_SHARED
-#define __S100 PAGE_RO_SHARED
-#define __S101 PAGE_RO_SHARED
+#define __S100 PAGE_READONLY
+#define __S101 PAGE_READONLY
#define __S110 PAGE_SHARED
#define __S111 PAGE_SHARED
@@ -373,18 +396,17 @@ extern inline int pmd_bad(pmd_t pmd)
extern inline int pte_none(pte_t pte)
{
- return (pte_val(pte) & _PAGE_INVALID_MASK) == _PAGE_INVALID_EMPTY;
+ return _PTE_IS_EMPTY(pte);
}
extern inline int pte_present(pte_t pte)
{
- return !(pte_val(pte) & _PAGE_INVALID) ||
- (pte_val(pte) & _PAGE_INVALID_MASK) == _PAGE_INVALID_NONE;
+ return _PTE_IS_VALID(pte) || _PTE_IS_NONE(pte);
}
extern inline int pte_file(pte_t pte)
{
- return (pte_val(pte) & _PAGE_INVALID_MASK) == _PAGE_INVALID_FILE;
+ return _PTE_IS_FILE(pte);
}
#define pte_same(a,b) (pte_val(a) == pte_val(b))
@@ -465,8 +487,8 @@ extern inline pte_t pte_modify(pte_t pte
extern inline pte_t pte_wrprotect(pte_t pte)
{
- /* Do not clobber _PAGE_INVALID_NONE pages! */
- if (!(pte_val(pte) & _PAGE_INVALID))
+ /* Do not clobber PROT_NONE pages! */
+ if (!_PTE_IS_NONE(pte))
pte_val(pte) |= _PAGE_RO;
return pte;
}
@@ -719,14 +741,14 @@ extern inline pmd_t * pmd_offset(pgd_t *
* information in the lowcore.
* Bit 21 and bit 22 are the page invalid bit and the page protection
* bit. We set both to indicate a swapped page.
- * Bit 31 is used as the software page present bit. If a page is
- * swapped this obviously has to be zero.
- * This leaves the bits 1-19 and bits 24-30 to store type and offset.
- * We use the 7 bits from 24-30 for the type and the 19 bits from 1-19
+ * Bit 30 and 31 are used to distinguish the different page types. For
+ * a swapped page these bits need to be zero.
+ * This leaves the bits 1-19 and bits 24-29 to store type and offset.
+ * We use the 6 bits from 24-29 for the type and the 19 bits from 1-19
* for the offset.
- * 0| offset |0110|type |0
- * 00000000001111111111222222222233
- * 01234567890123456789012345678901
+ * 0| offset |0110| type |00|
+ * 0 0000000001111111111 2222 222222 33
+ * 0 1234567890123456789 0123 456789 01
*
* 64 bit swap entry format:
* A page-table entry has some bits we have to treat in a special way.
@@ -736,19 +758,19 @@ extern inline pmd_t * pmd_offset(pgd_t *
* information in the lowcore.
* Bit 53 and bit 54 are the page invalid bit and the page protection
* bit. We set both to indicate a swapped page.
- * Bit 63 is used as the software page present bit. If a page is
- * swapped this obviously has to be zero.
- * This leaves the bits 0-51 and bits 56-62 to store type and offset.
- * We use the 7 bits from 56-62 for the type and the 52 bits from 0-51
+ * Bit 62 and 63 are used to distinguish the different page types. For
+ * a swapped page these bits need to be zero.
+ * This leaves the bits 0-51 and bits 56-61 to store type and offset.
+ * We use the 6 bits from 56-61 for the type and the 52 bits from 0-51
* for the offset.
- * | offset |0110|type |0
- * 0000000000111111111122222222223333333333444444444455555555556666
- * 0123456789012345678901234567890123456789012345678901234567890123
+ * | offset |0110| type |00|
+ * 0000000000111111111122222222223333333333444444444455 5555 555566 66
+ * 0123456789012345678901234567890123456789012345678901 2345 678901 23
*/
extern inline pte_t mk_swap_pte(unsigned long type, unsigned long offset)
{
pte_t pte;
- pte_val(pte) = (type << 1) | (offset << 12) | _PAGE_INVALID_SWAP;
+ pte_val(pte) = (type << 2) | (offset << 12) | _PAGE_INVALID_SWAP;
#ifndef __s390x__
BUG_ON((pte_val(pte) & 0x80000901) != 0);
#else /* __s390x__ */
@@ -757,7 +779,7 @@ extern inline pte_t mk_swap_pte(unsigned
return pte;
}
-#define __swp_type(entry) (((entry).val >> 1) & 0x3f)
+#define __swp_type(entry) (((entry).val >> 2) & 0x3f)
#define __swp_offset(entry) ((entry).val >> 12)
#define __swp_entry(type,offset) ((swp_entry_t) { pte_val(mk_swap_pte((type),(offset))) })
@@ -767,17 +789,21 @@ extern inline pte_t mk_swap_pte(unsigned
typedef pte_t *pte_addr_t;
#ifndef __s390x__
-# define PTE_FILE_MAX_BITS 26
+# define PTE_FILE_MAX_BITS 25
#else /* __s390x__ */
-# define PTE_FILE_MAX_BITS 59
+# define PTE_FILE_MAX_BITS 58
#endif /* __s390x__ */
#define pte_to_pgoff(__pte) \
- ((((__pte).pte >> 12) << 7) + (((__pte).pte >> 1) & 0x7f))
+ ((((__pte).pte >> 12) << 6) + (((__pte).pte >> 2) & 0x3f))
+
+#define pte_to_pgprot(pte) \
+ __pgprot(pte_val(pte) & (_PAGE_RO | _PAGE_PROTNONE))
-#define pgoff_to_pte(__off) \
- ((pte_t) { ((((__off) & 0x7f) << 1) + (((__off) >> 7) << 12)) \
- | _PAGE_INVALID_FILE })
+#define pgoff_prot_to_pte(__off, __prot) \
+ ((pte_t) { _PAGE_INVALID_FILE | \
+ ((((__off) & 0x3f) << 2) + (((__off) >> 6) << 12)) | \
+ (pgprot_val(__prot) & (_PAGE_RO | _PAGE_PROTNONE)) })
#endif /* !__ASSEMBLY__ */
--- diff/include/asm-s390/unistd.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/asm-s390/unistd.h 2004-03-16 11:27:48.516811984 +0000
@@ -260,8 +260,9 @@
* Number 263 is reserved for vserver
*/
#define __NR_fadvise64_64 264
+#define __NR_remap_file_pages 265
-#define NR_syscalls 265
+#define NR_syscalls 266
/*
* There are some system calls that are not present on 64 bit, some
--- diff/include/asm-sh/irq.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/asm-sh/irq.h 2004-03-16 11:27:48.516811984 +0000
@@ -324,4 +324,8 @@ static inline int generic_irq_demux(int
#define irq_canonicalize(irq) (irq)
#define irq_demux(irq) __irq_demux(sh_mv.mv_irq_demux(irq))
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif /* __ASM_SH_IRQ_H */
--- diff/include/asm-sparc/irq.h 2003-10-27 09:20:39.000000000 +0000
+++ source/include/asm-sparc/irq.h 2004-03-16 11:27:48.519811528 +0000
@@ -184,4 +184,8 @@ extern struct sun4m_intregs *sun4m_inter
#define SUN4M_INT_SBUS(x) (1 << (x+7))
#define SUN4M_INT_VME(x) (1 << (x))
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif
--- diff/include/asm-sparc/thread_info.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-sparc/thread_info.h 2004-03-16 11:27:48.519811528 +0000
@@ -100,9 +100,6 @@ BTFIXUPDEF_CALL(void, free_thread_info,
* Size of kernel stack for each process.
* Observe the order of get_free_pages() in alloc_thread_info().
* The sun4 has 8K stack too, because it's short on memory, and 16K is a waste.
- *
- * XXX Watch how INIT_THREAD_SIZE evolves in linux/sched.h and elsewhere.
- * On 2.5.24 it happens to match 8192 magically.
*/
#define THREAD_SIZE 8192
--- diff/include/asm-sparc64/compat.h 2003-07-08 08:55:19.000000000 +0000
+++ source/include/asm-sparc64/compat.h 2004-03-16 11:27:48.517811832 +0000
@@ -29,6 +29,7 @@ typedef s32 compat_int_t;
typedef s32 compat_long_t;
typedef u32 compat_uint_t;
typedef u32 compat_ulong_t;
+typedef u32 compat_timer_t;
struct compat_timespec {
compat_time_t tv_sec;
--- diff/include/asm-sparc64/irq.h 2003-08-26 09:00:54.000000000 +0000
+++ source/include/asm-sparc64/irq.h 2004-03-16 11:27:48.517811832 +0000
@@ -150,4 +150,8 @@ static __inline__ unsigned long get_soft
return retval;
}
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif
--- diff/include/asm-sparc64/pgtable.h 2004-01-19 10:22:59.000000000 +0000
+++ source/include/asm-sparc64/pgtable.h 2004-03-16 11:27:48.518811680 +0000
@@ -322,9 +322,16 @@ static inline pte_t mk_pte_io(unsigned l
/* File offset in PTE support. */
#define pte_file(pte) (pte_val(pte) & _PAGE_FILE)
-#define pte_to_pgoff(pte) (pte_val(pte) >> PAGE_SHIFT)
-#define pgoff_to_pte(off) (__pte(((off) << PAGE_SHIFT) | _PAGE_FILE))
-#define PTE_FILE_MAX_BITS (64UL - PAGE_SHIFT - 1UL)
+#define __pte_to_pgprot(pte) \
+ __pgprot(pte_val(pte) & (_PAGE_READ|_PAGE_WRITE))
+#define __file_pte_to_pgprot(pte) \
+ __pgprot(((pte_val(pte) >> PAGE_SHIFT) & 0x3UL) << 8)
+#define pte_to_pgprot(pte) \
+ (pte_file(pte) ? __file_pte_to_pgprot(pte) : __pte_to_pgprot(pte))
+#define pte_to_pgoff(pte) (pte_val(pte) >> (PAGE_SHIFT+2))
+#define pgoff_prot_to_pte(off, prot) \
+ ((__pte(((off) | ((pgprot_val(prot) >> 8) & 0x3UL)))) << (PAGE_SHIFT+2) | _PAGE_FILE)
+#define PTE_FILE_MAX_BITS (64UL - PAGE_SHIFT - 3UL)
extern unsigned long prom_virt_to_phys(unsigned long, int *);
--- diff/include/asm-sparc64/spinlock.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-sparc64/spinlock.h 2004-03-16 11:27:48.518811680 +0000
@@ -31,15 +31,23 @@
#ifndef CONFIG_DEBUG_SPINLOCK
-typedef unsigned char spinlock_t;
-#define SPIN_LOCK_UNLOCKED 0
+typedef struct {
+ unsigned char lock;
+ unsigned int index;
+} spinlock_t;
-#define spin_lock_init(lock) (*((unsigned char *)(lock)) = 0)
-#define spin_is_locked(lock) (*((volatile unsigned char *)(lock)) != 0)
+#ifdef CONFIG_LOCKMETER
+#define SPIN_LOCK_UNLOCKED (spinlock_t) {0, 0}
+#else
+#define SPIN_LOCK_UNLOCKED (spinlock_t) { 0 }
+#endif
-#define spin_unlock_wait(lock) \
+#define spin_lock_init(__lock) do { *(__lock) = SPIN_LOCK_UNLOCKED; } while(0)
+#define spin_is_locked(__lock) (*((volatile unsigned char *)(&((__lock)->lock))) != 0)
+
+#define spin_unlock_wait(__lock) \
do { membar("#LoadLoad"); \
-} while(*((volatile unsigned char *)lock))
+} while(*((volatile unsigned char *)(&(((spinlock_t *)__lock)->lock))))
static __inline__ void _raw_spin_lock(spinlock_t *lock)
{
@@ -110,17 +118,31 @@ extern int _spin_trylock (spinlock_t *lo
#ifndef CONFIG_DEBUG_SPINLOCK
-typedef unsigned int rwlock_t;
-#define RW_LOCK_UNLOCKED 0
-#define rwlock_init(lp) do { *(lp) = RW_LOCK_UNLOCKED; } while(0)
-#define rwlock_is_locked(x) (*(x) != RW_LOCK_UNLOCKED)
+#ifdef CONFIG_LOCKMETER
+typedef struct {
+ unsigned int lock;
+ unsigned int index;
+ unsigned int cpu;
+} rwlock_t;
+#define RW_LOCK_UNLOCKED (rwlock_t) { 0, 0, 0xff }
+#else
+typedef struct {
+ unsigned int lock;
+} rwlock_t;
+#define RW_LOCK_UNLOCKED (rwlock_t) { 0 }
+#endif
+
+#define rwlock_init(lp) do { *(lp) = RW_LOCK_UNLOCKED; } while(0)
+#define rwlock_is_locked(x) ((x)->lock != 0)
+extern int __read_trylock(rwlock_t *);
extern void __read_lock(rwlock_t *);
extern void __read_unlock(rwlock_t *);
extern void __write_lock(rwlock_t *);
extern void __write_unlock(rwlock_t *);
extern int __write_trylock(rwlock_t *);
+#define _raw_read_trylock(p) __read_trylock(p)
#define _raw_read_lock(p) __read_lock(p)
#define _raw_read_unlock(p) __read_unlock(p)
#define _raw_write_lock(p) __write_lock(p)
--- diff/include/asm-um/irq.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/asm-um/irq.h 2004-03-16 11:27:48.520811376 +0000
@@ -32,4 +32,9 @@ extern int um_request_irq(unsigned int i
void (*handler)(int, void *, struct pt_regs *),
unsigned long irqflags, const char * devname,
void *dev_id);
+
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif
--- diff/include/asm-um/processor-generic.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/asm-um/processor-generic.h 2004-03-16 11:27:48.520811376 +0000
@@ -94,8 +94,6 @@ struct thread_struct {
.request = { 0 } \
}
-#define INIT_THREAD_SIZE ((1 << CONFIG_KERNEL_STACK_ORDER) * PAGE_SIZE)
-
typedef struct {
unsigned long seg;
} mm_segment_t;
--- diff/include/asm-v850/irq.h 2003-05-21 10:50:16.000000000 +0000
+++ source/include/asm-v850/irq.h 2004-03-16 11:27:48.521811224 +0000
@@ -65,4 +65,8 @@ extern void disable_irq_nosync (unsigned
#endif /* !__ASSEMBLY__ */
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif /* __V850_IRQ_H__ */
--- diff/include/asm-x86_64/bitops.h 2004-01-19 10:22:59.000000000 +0000
+++ source/include/asm-x86_64/bitops.h 2004-03-16 11:27:48.521811224 +0000
@@ -503,6 +503,8 @@ static __inline__ int ffs(int x)
/* find last set bit */
#define fls(x) generic_fls(x)
+#define ARCH_HAS_ATOMIC_UNSIGNED 1
+
#endif /* __KERNEL__ */
#endif /* _X86_64_BITOPS_H */
--- diff/include/asm-x86_64/ia32_unistd.h 2003-09-17 11:28:12.000000000 +0000
+++ source/include/asm-x86_64/ia32_unistd.h 2004-03-16 11:27:48.522811072 +0000
@@ -262,7 +262,7 @@
#define __NR_ia32_sys_epoll_create 254
#define __NR_ia32_sys_epoll_ctl 255
#define __NR_ia32_sys_epoll_wait 256
-#define __NR_ia32_remap_file_pages 257
+#define __NR_ia32_old_remap_file_pages 257
#define __NR_ia32_set_tid_address 258
#define __NR_ia32_timer_create 259
#define __NR_ia32_timer_settime (__NR_ia32_timer_create+1)
--- diff/include/asm-x86_64/irq.h 2004-01-19 10:22:59.000000000 +0000
+++ source/include/asm-x86_64/irq.h 2004-03-16 11:27:48.522811072 +0000
@@ -53,4 +53,8 @@ extern int can_request_irq(unsigned int,
#define ARCH_HAS_NMI_WATCHDOG /* See include/linux/nmi.h */
#endif
+struct irqaction;
+struct pt_regs;
+int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
+
#endif /* _ASM_IRQ_H */
--- diff/include/asm-x86_64/mman.h 2003-09-30 14:46:20.000000000 +0000
+++ source/include/asm-x86_64/mman.h 2004-03-16 11:27:48.524810768 +0000
@@ -23,6 +23,7 @@
#define MAP_NORESERVE 0x4000 /* don't check for reservations */
#define MAP_POPULATE 0x8000 /* populate (prefault) pagetables */
#define MAP_NONBLOCK 0x10000 /* do not block on IO */
+#define MAP_INHERIT 0x20000 /* inherit the protection bits of the underlying vma*/
#define MS_ASYNC 1 /* sync memory asynchronously */
#define MS_INVALIDATE 2 /* invalidate the caches */
--- diff/include/asm-x86_64/pgtable.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-x86_64/pgtable.h 2004-03-16 11:27:48.525810616 +0000
@@ -344,9 +344,19 @@ static inline pgd_t *current_pgd_offset_
#define pmd_pfn(x) ((pmd_val(x) >> PAGE_SHIFT) & __PHYSICAL_MASK)
#define pte_to_pgoff(pte) ((pte_val(pte) & PHYSICAL_PAGE_MASK) >> PAGE_SHIFT)
-#define pgoff_to_pte(off) ((pte_t) { ((off) << PAGE_SHIFT) | _PAGE_FILE })
#define PTE_FILE_MAX_BITS __PHYSICAL_MASK_SHIFT
+#define pte_to_pgprot(pte) \
+ __pgprot((pte_val(pte) & (_PAGE_RW | _PAGE_PROTNONE)) \
+ | ((pte_val(pte) & _PAGE_PROTNONE) ? 0 : \
+ (_PAGE_USER | _PAGE_PRESENT)) | _PAGE_ACCESSED)
+
+#define pgoff_prot_to_pte(off, prot) \
+ ((pte_t) { _PAGE_FILE + \
+ (pgprot_val(prot) & (_PAGE_RW | _PAGE_PROTNONE)) + \
+ ((off) << PAGE_SHIFT) })
+
+
/* PTE - Level 1 access. */
/* page, protection -> pte */
--- diff/include/asm-x86_64/thread_info.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/asm-x86_64/thread_info.h 2004-03-16 11:27:48.526810464 +0000
@@ -101,6 +101,7 @@ static inline struct thread_info *stack_
#define TIF_NEED_RESCHED 3 /* rescheduling necessary */
#define TIF_SINGLESTEP 4 /* reenable singlestep on user return*/
#define TIF_IRET 5 /* force IRET */
+#define TIF_SYSCALL_AUDIT 7 /* syscall auditing active */
#define TIF_POLLING_NRFLAG 16 /* true if poll_idle() is polling TIF_NEED_RESCHED */
#define TIF_IA32 17 /* 32bit process */
#define TIF_FORK 18 /* ret_from_fork */
@@ -112,13 +113,15 @@ static inline struct thread_info *stack_
#define _TIF_SINGLESTEP (1<<TIF_SINGLESTEP)
#define _TIF_NEED_RESCHED (1<<TIF_NEED_RESCHED)
#define _TIF_IRET (1<<TIF_IRET)
+#define _TIF_SYSCALL_AUDIT (1<<TIF_SYSCALL_AUDIT)
#define _TIF_POLLING_NRFLAG (1<<TIF_POLLING_NRFLAG)
#define _TIF_IA32 (1<<TIF_IA32)
#define _TIF_FORK (1<<TIF_FORK)
#define _TIF_ABI_PENDING (1<<TIF_ABI_PENDING)
/* work to do on interrupt/exception return */
-#define _TIF_WORK_MASK (0x0000FFFF & ~(_TIF_SYSCALL_TRACE|_TIF_SINGLESTEP))
+#define _TIF_WORK_MASK \
+ (0x0000FFFF & ~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP))
/* work to do on any return to user space */
#define _TIF_ALLWORK_MASK 0x0000FFFF
--- diff/include/asm-x86_64/unistd.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/asm-x86_64/unistd.h 2004-03-16 11:27:48.527810312 +0000
@@ -490,8 +490,8 @@ __SYSCALL(__NR_epoll_create, sys_epoll_c
__SYSCALL(__NR_epoll_ctl_old, sys_ni_syscall)
#define __NR_epoll_wait_old 215
__SYSCALL(__NR_epoll_wait_old, sys_ni_syscall)
-#define __NR_remap_file_pages 216
-__SYSCALL(__NR_remap_file_pages, sys_remap_file_pages)
+#define __NR_old_remap_file_pages 216
+__SYSCALL(__NR_old_remap_file_pages, old_remap_file_pages)
#define __NR_getdents64 217
__SYSCALL(__NR_getdents64, sys_getdents64)
#define __NR_set_tid_address 218
--- diff/include/linux/backing-dev.h 2003-05-21 10:49:56.000000000 +0000
+++ source/include/linux/backing-dev.h 2004-03-16 11:27:48.528810160 +0000
@@ -20,26 +20,46 @@ enum bdi_state {
BDI_unused, /* Available bits start here */
};
+typedef int (congested_fn)(void *, int);
+
struct backing_dev_info {
unsigned long ra_pages; /* max readahead in PAGE_CACHE_SIZE units */
unsigned long state; /* Always use atomic bitops on this */
int memory_backed; /* Cannot clean pages with writepage */
+ congested_fn *congested_fn; /* Function pointer if device is md/dm */
+ void *congested_data; /* Pointer to aux data for congested func */
+ void (*unplug_io_fn)(struct backing_dev_info *);
+ void *unplug_io_data;
};
extern struct backing_dev_info default_backing_dev_info;
+void default_unplug_io_fn(struct backing_dev_info *bdi);
int writeback_acquire(struct backing_dev_info *bdi);
int writeback_in_progress(struct backing_dev_info *bdi);
void writeback_release(struct backing_dev_info *bdi);
+static inline int bdi_congested(struct backing_dev_info *bdi, int bdi_bits)
+{
+ if (bdi->congested_fn)
+ return bdi->congested_fn(bdi->congested_data, bdi_bits);
+ return (bdi->state & bdi_bits);
+}
+
static inline int bdi_read_congested(struct backing_dev_info *bdi)
{
- return test_bit(BDI_read_congested, &bdi->state);
+ return bdi_congested(bdi, 1 << BDI_read_congested);
}
static inline int bdi_write_congested(struct backing_dev_info *bdi)
{
- return test_bit(BDI_write_congested, &bdi->state);
+ return bdi_congested(bdi, 1 << BDI_write_congested);
+}
+
+static inline int bdi_rw_congested(struct backing_dev_info *bdi)
+{
+ return bdi_congested(bdi, (1 << BDI_read_congested)|
+ (1 << BDI_write_congested));
}
#endif /* _LINUX_BACKING_DEV_H */
--- diff/include/linux/binfmts.h 2003-08-20 13:16:34.000000000 +0000
+++ source/include/linux/binfmts.h 2004-03-16 11:27:48.529810008 +0000
@@ -35,9 +35,13 @@ struct linux_binprm{
char * interp; /* Name of the binary really executed. Most
of the time same as filename, but could be
different for binfmt_{misc,script} */
+ unsigned long interp_flags;
unsigned long loader, exec;
};
+#define BINPRM_FLAGS_ENFORCE_NONDUMP_BIT 0
+#define BINPRM_FLAGS_ENFORCE_NONDUMP (1 << BINPRM_FLAGS_ENFORCE_NONDUMP_BIT)
+
/*
* This structure defines the functions that are used to load the binary formats that
* linux accepts.
--- diff/include/linux/bio.h 2004-02-18 08:54:13.000000000 +0000
+++ source/include/linux/bio.h 2004-03-16 11:27:48.529810008 +0000
@@ -119,11 +119,13 @@ struct bio {
* bit 1 -- rw-ahead when set
* bit 2 -- barrier
* bit 3 -- fail fast, don't want low level driver retries
+ * bit 4 -- synchronous I/O hint: the block layer will unplug immediately
*/
#define BIO_RW 0
#define BIO_RW_AHEAD 1
#define BIO_RW_BARRIER 2
#define BIO_RW_FAILFAST 3
+#define BIO_RW_SYNC 4
/*
* various member access, note that bio_data should of course not be used
@@ -138,6 +140,7 @@ struct bio {
#define bio_cur_sectors(bio) (bio_iovec(bio)->bv_len >> 9)
#define bio_data(bio) (page_address(bio_page((bio))) + bio_offset((bio)))
#define bio_barrier(bio) ((bio)->bi_rw & (1 << BIO_RW_BARRIER))
+#define bio_sync(bio) ((bio)->bi_rw & (1 << BIO_RW_SYNC))
/*
* will die
--- diff/include/linux/blkdev.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/linux/blkdev.h 2004-03-16 11:27:48.530809856 +0000
@@ -243,7 +243,7 @@ typedef int (merge_requests_fn) (request
typedef void (request_fn_proc) (request_queue_t *q);
typedef int (make_request_fn) (request_queue_t *q, struct bio *bio);
typedef int (prep_rq_fn) (request_queue_t *, struct request *);
-typedef void (unplug_fn) (void *q);
+typedef void (unplug_fn) (request_queue_t *);
struct bio_vec;
typedef int (merge_bvec_fn) (request_queue_t *, struct bio *, struct bio_vec *);
@@ -315,8 +315,6 @@ struct request_queue
unsigned long bounce_pfn;
int bounce_gfp;
- struct list_head plug_list;
-
/*
* various queue flags, see QUEUE_* below
*/
@@ -370,8 +368,9 @@ struct request_queue
#define QUEUE_FLAG_WRITEFULL 4 /* read queue has been filled */
#define QUEUE_FLAG_DEAD 5 /* queue being torn down */
#define QUEUE_FLAG_REENTER 6 /* Re-entrancy avoidance */
+#define QUEUE_FLAG_PLUGGED 7 /* queue is plugged */
-#define blk_queue_plugged(q) !list_empty(&(q)->plug_list)
+#define blk_queue_plugged(q) test_bit(QUEUE_FLAG_PLUGGED, &(q)->queue_flags)
#define blk_queue_tagged(q) test_bit(QUEUE_FLAG_QUEUED, &(q)->queue_flags)
#define blk_queue_stopped(q) test_bit(QUEUE_FLAG_STOPPED, &(q)->queue_flags)
@@ -515,7 +514,7 @@ extern int scsi_cmd_ioctl(struct gendisk
extern void blk_start_queue(request_queue_t *q);
extern void blk_stop_queue(request_queue_t *q);
extern void __blk_stop_queue(request_queue_t *q);
-extern void blk_run_queue(request_queue_t *q);
+extern void blk_run_queue(request_queue_t *);
extern void blk_queue_activity_fn(request_queue_t *, activity_fn *, void *);
extern struct request *blk_rq_map_user(request_queue_t *, int, void __user *, unsigned int);
extern int blk_rq_unmap_user(struct request *, void __user *, unsigned int);
@@ -526,6 +525,18 @@ static inline request_queue_t *bdev_get_
return bdev->bd_disk->queue;
}
+static inline void blk_run_backing_dev(struct backing_dev_info *bdi)
+{
+ if (bdi && bdi->unplug_io_fn)
+ bdi->unplug_io_fn(bdi);
+}
+
+static inline void blk_run_address_space(struct address_space *mapping)
+{
+ if (mapping)
+ blk_run_backing_dev(mapping->backing_dev_info);
+}
+
/*
* end_request() and friends. Must be called with the request queue spinlock
* acquired. All functions called within end_request() _must_be_ atomic.
@@ -572,7 +583,7 @@ extern struct backing_dev_info *blk_get_
extern int blk_rq_map_sg(request_queue_t *, struct request *, struct scatterlist *);
extern void blk_dump_rq_flags(struct request *, char *);
-extern void generic_unplug_device(void *);
+extern void generic_unplug_device(request_queue_t *);
extern long nr_blockdev_pages(void);
int blk_get_queue(request_queue_t *);
--- diff/include/linux/compat.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/linux/compat.h 2004-03-16 11:27:48.536808944 +0000
@@ -6,10 +6,16 @@
*/
#include <linux/config.h>
-#ifdef CONFIG_COMPAT
+#ifndef CONFIG_COMPAT
+
+/* Non-native task requiring compat... doesn't exist */
+#define is_compat_task(x) 0
+
+#else
#include <linux/stat.h>
#include <linux/param.h> /* for HZ */
+#include <linux/personality.h> /* Conditional process compat */
#include <linux/sem.h>
#include <asm/compat.h>
@@ -17,6 +23,11 @@
#define compat_jiffies_to_clock_t(x) \
(((unsigned long)(x) * COMPAT_USER_HZ) / HZ)
+/* Non-native task requiring compat */
+#ifndef HAVE_ARCH_IS_COMPAT_TASK
+#define is_compat_task(x) (x->personality == PER_LINUX32)
+#endif
+
struct compat_itimerspec {
struct compat_timespec it_interval;
struct compat_timespec it_value;
@@ -85,11 +96,6 @@ struct compat_dirent {
char d_name[256];
};
-typedef union compat_sigval {
- compat_int_t sival_int;
- compat_uptr_t sival_ptr;
-} compat_sigval_t;
-
long compat_sys_semctl(int first, int second, int third, void __user *uptr);
long compat_sys_msgsnd(int first, int second, int third, void __user *uptr);
long compat_sys_msgrcv(int first, int second, int msgtyp, int third,
--- diff/include/linux/compiler-gcc.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/linux/compiler-gcc.h 2004-03-16 11:27:48.537808792 +0000
@@ -13,5 +13,5 @@
shouldn't recognize the original var, and make assumptions about it */
#define RELOC_HIDE(ptr, off) \
({ unsigned long __ptr; \
- __asm__ ("" : "=g"(__ptr) : "0"(ptr)); \
+ __asm__ ("" : "=r"(__ptr) : "0"(ptr)); \
(typeof(ptr)) (__ptr + (off)); })
--- diff/include/linux/compiler-gcc3.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/linux/compiler-gcc3.h 2004-03-16 11:27:48.537808792 +0000
@@ -3,7 +3,7 @@
/* These definitions are for GCC v3.x. */
#include <linux/compiler-gcc.h>
-#if __GNUC_MINOR__ >= 1
+#if __GNUC_MINOR__ >= 1 && __GNUC_MINOR__ < 4
# define inline __inline__ __attribute__((always_inline))
# define __inline__ __inline__ __attribute__((always_inline))
# define __inline __inline__ __attribute__((always_inline))
--- diff/include/linux/config.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/linux/config.h 2004-03-16 11:27:48.538808640 +0000
@@ -2,5 +2,8 @@
#define _LINUX_CONFIG_H
#include <linux/autoconf.h>
+#ifdef CONFIG_X86
+#include <asm/kgdb.h>
+#endif
#endif
--- diff/include/linux/elevator.h 2004-02-18 08:54:13.000000000 +0000
+++ source/include/linux/elevator.h 2004-03-16 11:27:48.541808184 +0000
@@ -94,6 +94,11 @@ extern elevator_t iosched_deadline;
*/
extern elevator_t iosched_as;
+/*
+ * completely fair queueing I/O scheduler
+ */
+extern elevator_t iosched_cfq;
+
extern int elevator_init(request_queue_t *, elevator_t *);
extern void elevator_exit(request_queue_t *);
extern int elv_rq_merge_ok(struct request *, struct bio *);
--- diff/include/linux/ext3_fs_sb.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/linux/ext3_fs_sb.h 2004-03-16 11:27:48.541808184 +0000
@@ -69,6 +69,10 @@ struct ext3_sb_info {
struct timer_list turn_ro_timer; /* For turning read-only (crash simulation) */
wait_queue_head_t ro_wait_queue; /* For people waiting for the fs to go read-only */
#endif
+#ifdef CONFIG_QUOTA
+ char *s_qf_names[MAXQUOTAS]; /* Names of quota files with journalled quota */
+ int s_jquota_fmt; /* Format of quota to use */
+#endif
};
#endif /* _LINUX_EXT3_FS_SB */
--- diff/include/linux/ext3_jbd.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/linux/ext3_jbd.h 2004-03-16 11:27:48.541808184 +0000
@@ -42,8 +42,9 @@
* superblock only gets updated once, of course, so don't bother
* counting that again for the quota updates. */
-#define EXT3_DATA_TRANS_BLOCKS (3 * EXT3_SINGLEDATA_TRANS_BLOCKS + \
- EXT3_XATTR_TRANS_BLOCKS - 2)
+#define EXT3_DATA_TRANS_BLOCKS (EXT3_SINGLEDATA_TRANS_BLOCKS + \
+ EXT3_XATTR_TRANS_BLOCKS - 2 + \
+ 2*EXT3_QUOTA_TRANS_BLOCKS)
extern int ext3_writepage_trans_blocks(struct inode *inode);
@@ -72,6 +73,19 @@ extern int ext3_writepage_trans_blocks(s
#define EXT3_INDEX_EXTRA_TRANS_BLOCKS 8
+#ifdef CONFIG_QUOTA
+/* Amount of blocks needed for quota update - we know that the structure was
+ * allocated so we need to update only inode+data */
+#define EXT3_QUOTA_TRANS_BLOCKS 2
+/* Amount of blocks needed for quota insert/delete - we do some block writes
+ * but inode, sb and group updates are done only once */
+#define EXT3_QUOTA_INIT_BLOCKS (DQUOT_MAX_WRITES*\
+ (EXT3_SINGLEDATA_TRANS_BLOCKS-3)+3)
+#else
+#define EXT3_QUOTA_TRANS_BLOCKS 0
+#define EXT3_QUOTA_INIT_BLOCKS 0
+#endif
+
int
ext3_mark_iloc_dirty(handle_t *handle,
struct inode *inode,
--- diff/include/linux/fs.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/linux/fs.h 2004-03-16 11:27:48.544807728 +0000
@@ -20,6 +20,7 @@
#include <linux/radix-tree.h>
#include <linux/kobject.h>
#include <asm/atomic.h>
+#include <linux/audit.h>
struct iovec;
struct nameidata;
@@ -82,6 +83,8 @@ extern int leases_enable, dir_notify_ena
#define WRITE 1
#define READA 2 /* read-ahead - don't block if no resources */
#define SPECIAL 4 /* For non-blockdevice requests in request queue */
+#define READ_SYNC (READ | (1 << BIO_RW_SYNC))
+#define WRITE_SYNC (WRITE | (1 << BIO_RW_SYNC))
#define SEL_IN 1
#define SEL_OUT 2
@@ -322,11 +325,7 @@ struct backing_dev_info;
struct address_space {
struct inode *host; /* owner: inode, block_device */
struct radix_tree_root page_tree; /* radix tree of all pages */
- spinlock_t page_lock; /* and spinlock protecting it */
- struct list_head clean_pages; /* list of clean pages */
- struct list_head dirty_pages; /* list of dirty pages */
- struct list_head locked_pages; /* list of locked pages */
- struct list_head io_pages; /* being prepared for I/O */
+ spinlock_t tree_lock; /* and spinlock protecting it */
unsigned long nrpages; /* number of total pages */
struct address_space_operations *a_ops; /* methods */
struct list_head i_mmap; /* list of private mappings */
@@ -366,6 +365,15 @@ struct block_device {
};
/*
+ * Radix-tre tags, for tagging dirty and writeback pages within the pagecache
+ * radix trees
+ */
+#define PAGECACHE_TAG_DIRTY 0
+#define PAGECACHE_TAG_WRITEBACK 1
+
+int mapping_tagged(struct address_space *mapping, int tag);
+
+/*
* Use sequence counter to get consistent i_size on 32-bit processors.
*/
#if BITS_PER_LONG==32 && defined(CONFIG_SMP)
@@ -379,6 +387,7 @@ struct block_device {
struct inode {
struct hlist_node i_hash;
struct list_head i_list;
+ struct list_head i_sb_list;
struct list_head i_dentry;
unsigned long i_ino;
atomic_t i_count;
@@ -398,6 +407,7 @@ struct inode {
unsigned short i_bytes;
spinlock_t i_lock; /* i_blocks, i_bytes, maybe i_size */
struct semaphore i_sem;
+ struct rw_semaphore i_alloc_sem;
struct inode_operations *i_op;
struct file_operations *i_fop; /* former ->i_op->default_file_ops */
struct super_block *i_sb;
@@ -705,6 +715,7 @@ struct super_block {
atomic_t s_active;
void *s_security;
+ struct list_head s_inodes; /* all inodes */
struct list_head s_dirty; /* dirty inodes */
struct list_head s_io; /* parked for writeback */
struct hlist_head s_anon; /* anonymous dentries for (nfs) exporting */
@@ -1133,7 +1144,18 @@ extern char * getname(const char __user
extern void vfs_caches_init(unsigned long);
#define __getname() kmem_cache_alloc(names_cachep, SLAB_KERNEL)
-#define putname(name) kmem_cache_free(names_cachep, (void *)(name))
+#define __putname(name) kmem_cache_free(names_cachep, (void *)(name))
+#ifndef CONFIG_AUDITSYSCALL
+#define putname(name) __putname(name)
+#else
+#define putname(name) \
+ do { \
+ if (unlikely(current->audit_context)) \
+ audit_putname(name); \
+ else \
+ __putname(name); \
+ } while (0)
+#endif
extern int register_blkdev(unsigned int, const char *);
extern int unregister_blkdev(unsigned int, const char *);
@@ -1216,6 +1238,7 @@ extern void write_inode_now(struct inode
extern int filemap_fdatawrite(struct address_space *);
extern int filemap_flush(struct address_space *);
extern int filemap_fdatawait(struct address_space *);
+extern int filemap_write_and_wait(struct address_space *mapping);
extern void sync_supers(void);
extern void sync_filesystems(int wait);
extern void emergency_sync(void);
@@ -1327,9 +1350,6 @@ extern void
file_ra_state_init(struct file_ra_state *ra, struct address_space *mapping);
extern ssize_t generic_file_direct_IO(int rw, struct kiocb *iocb,
const struct iovec *iov, loff_t offset, unsigned long nr_segs);
-extern int blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
- struct block_device *bdev, const struct iovec *iov, loff_t offset,
- unsigned long nr_segs, get_blocks_t *get_blocks, dio_iodone_t *end_io);
extern ssize_t generic_file_readv(struct file *filp, const struct iovec *iov,
unsigned long nr_segs, loff_t *ppos);
ssize_t generic_file_writev(struct file *filp, const struct iovec *iov,
@@ -1351,6 +1371,32 @@ static inline void do_generic_file_read(
actor);
}
+int __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
+ struct block_device *bdev, const struct iovec *iov, loff_t offset,
+ unsigned long nr_segs, get_blocks_t get_blocks, dio_iodone_t end_io,
+ int needs_special_locking);
+
+/*
+ * For filesystems which need locking between buffered and direct access
+ */
+static inline int blockdev_direct_IO(int rw, struct kiocb *iocb,
+ struct inode *inode, struct block_device *bdev, const struct iovec *iov,
+ loff_t offset, unsigned long nr_segs, get_blocks_t get_blocks,
+ dio_iodone_t end_io)
+{
+ return __blockdev_direct_IO(rw, iocb, inode, bdev, iov, offset,
+ nr_segs, get_blocks, end_io, 1);
+}
+
+static inline int blockdev_direct_IO_no_locking(int rw, struct kiocb *iocb,
+ struct inode *inode, struct block_device *bdev, const struct iovec *iov,
+ loff_t offset, unsigned long nr_segs, get_blocks_t get_blocks,
+ dio_iodone_t end_io)
+{
+ return __blockdev_direct_IO(rw, iocb, inode, bdev, iov, offset,
+ nr_segs, get_blocks, end_io, 0);
+}
+
extern struct file_operations generic_ro_fops;
#define special_file(m) (S_ISCHR(m)||S_ISBLK(m)||S_ISFIFO(m)||S_ISSOCK(m))
--- diff/include/linux/hiddev.h 2003-09-17 11:28:12.000000000 +0000
+++ source/include/linux/hiddev.h 2004-03-16 11:27:48.546807424 +0000
@@ -39,33 +39,33 @@ struct hiddev_event {
};
struct hiddev_devinfo {
- unsigned int bustype;
- unsigned int busnum;
- unsigned int devnum;
- unsigned int ifnum;
- short vendor;
- short product;
- short version;
- unsigned num_applications;
+ __u32 bustype;
+ __u32 busnum;
+ __u32 devnum;
+ __u32 ifnum;
+ __s16 vendor;
+ __s16 product;
+ __s16 version;
+ __u32 num_applications;
};
struct hiddev_collection_info {
- unsigned index;
- unsigned type;
- unsigned usage;
- unsigned level;
+ __u32 index;
+ __u32 type;
+ __u32 usage;
+ __u32 level;
};
#define HID_STRING_SIZE 256
struct hiddev_string_descriptor {
- int index;
+ __s32 index;
char value[HID_STRING_SIZE];
};
struct hiddev_report_info {
- unsigned report_type;
- unsigned report_id;
- unsigned num_fields;
+ __u32 report_type;
+ __u32 report_id;
+ __u32 num_fields;
};
/* To do a GUSAGE/SUSAGE, fill in at least usage_code, report_type and
@@ -88,20 +88,20 @@ struct hiddev_report_info {
#define HID_REPORT_TYPE_MAX 3
struct hiddev_field_info {
- unsigned report_type;
- unsigned report_id;
- unsigned field_index;
- unsigned maxusage;
- unsigned flags;
- unsigned physical; /* physical usage for this field */
- unsigned logical; /* logical usage for this field */
- unsigned application; /* application usage for this field */
+ __u32 report_type;
+ __u32 report_id;
+ __u32 field_index;
+ __u32 maxusage;
+ __u32 flags;
+ __u32 physical; /* physical usage for this field */
+ __u32 logical; /* logical usage for this field */
+ __u32 application; /* application usage for this field */
__s32 logical_minimum;
__s32 logical_maximum;
__s32 physical_minimum;
__s32 physical_maximum;
- unsigned unit_exponent;
- unsigned unit;
+ __u32 unit_exponent;
+ __u32 unit;
};
/* Fill in report_type, report_id and field_index to get the information on a
@@ -118,14 +118,22 @@ struct hiddev_field_info {
#define HID_FIELD_BUFFERED_BYTE 0x100
struct hiddev_usage_ref {
- unsigned report_type;
- unsigned report_id;
- unsigned field_index;
- unsigned usage_index;
- unsigned usage_code;
+ __u32 report_type;
+ __u32 report_id;
+ __u32 field_index;
+ __u32 usage_index;
+ __u32 usage_code;
__s32 value;
};
+/* hiddev_usage_ref_multi is used for sending multiple bytes to a control.
+ * It really manifests itself as setting the value of consecutive usages */
+struct hiddev_usage_ref_multi {
+ struct hiddev_usage_ref uref;
+ __u32 num_values;
+ __s32 values[HID_MAX_USAGES];
+};
+
/* FIELD_INDEX_NONE is returned in read() data from the kernel when flags
* is set to (HIDDEV_FLAG_UREF | HIDDEV_FLAG_REPORT) and a new report has
* been sent by the device
@@ -161,6 +169,10 @@ struct hiddev_usage_ref {
#define HIDIOCGCOLLECTIONINFO _IOWR('H', 0x11, struct hiddev_collection_info)
#define HIDIOCGPHYS(len) _IOC(_IOC_READ, 'H', 0x12, len)
+/* For writing/reading to multiple/consecutive usages */
+#define HIDIOCGUSAGES _IOWR('H', 0x13, struct hiddev_usage_ref_multi)
+#define HIDIOCSUSAGES _IOW('H', 0x14, struct hiddev_usage_ref_multi)
+
/*
* Flags to be used in HIDIOCSFLAG
*/
--- diff/include/linux/ide.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/linux/ide.h 2004-03-16 11:27:48.547807272 +0000
@@ -972,6 +972,7 @@ typedef struct hwif_s {
unsigned dma;
void (*led_act)(void *data, int rw);
+ unsigned int (*max_rqsize)(ide_drive_t *);
} ide_hwif_t;
/*
@@ -1612,6 +1613,7 @@ extern int ide_dma_enable(ide_drive_t *d
extern char *ide_xfer_verbose(u8 xfer_rate);
extern void ide_toggle_bounce(ide_drive_t *drive, int on);
extern int ide_set_xfer_rate(ide_drive_t *drive, u8 rate);
+extern byte ide_dump_atapi_status(ide_drive_t *drive, const char *msg, byte stat);
typedef struct ide_pio_timings_s {
int setup_time; /* Address setup (ns) minimum */
--- diff/include/linux/idr.h 2004-03-11 10:20:28.000000000 +0000
+++ source/include/linux/idr.h 2004-03-16 11:27:48.548807120 +0000
@@ -23,6 +23,7 @@
# error "BITS_PER_LONG is not 32 or 64"
#endif
+#define IDR_SIZE (1 << IDR_BITS)
#define IDR_MASK ((1 << IDR_BITS)-1)
/* Define the size of the id's */
@@ -53,6 +54,17 @@ struct idr {
spinlock_t lock;
};
+#define IDR_INIT(name) \
+{ \
+ .top = NULL, \
+ .id_free = NULL, \
+ .count = 0, \
+ .layers = 0, \
+ .id_free_cnt = 0, \
+ .lock = SPIN_LOCK_UNLOCKED, \
+}
+#define DEFINE_IDR(name) struct idr name = IDR_INIT(name)
+
/*
* This is what we export.
*/
--- diff/include/linux/init.h 2003-11-25 15:24:59.000000000 +0000
+++ source/include/linux/init.h 2004-03-16 11:27:48.593800280 +0000
@@ -110,12 +110,21 @@ struct obs_kernel_param {
};
/* OBSOLETE: see moduleparam.h for the right way. */
-#define __setup(str, fn) \
- static char __setup_str_##fn[] __initdata = str; \
- static struct obs_kernel_param __setup_##fn \
+#define __setup_param(str, unique_id, fn) \
+ static char __setup_str_##unique_id[] __initdata = str; \
+ static struct obs_kernel_param __setup_##unique_id \
__attribute_used__ \
__attribute__((__section__(".init.setup"))) \
- = { __setup_str_##fn, fn }
+ = { __setup_str_##unique_id, fn }
+
+#define __setup_null_param(str, unique_id) \
+ __setup_param(str, unique_id, NULL)
+
+#define __setup(str, fn) \
+ __setup_param(str, fn, fn)
+
+#define __obsolete_setup(str) \
+ __setup_null_param(str, __LINE__)
#endif /* __ASSEMBLY__ */
@@ -172,7 +181,10 @@ struct obs_kernel_param {
{ return exitfn; } \
void cleanup_module(void) __attribute__((alias(#exitfn)));
-#define __setup(str,func) /* nothing */
+#define __setup_param(str, unique_id, fn) /* nothing */
+#define __setup_null_param(str, unique_id) /* nothing */
+#define __setup(str, func) /* nothing */
+#define __obsolete_setup(str) /* nothing */
#endif
/* Data marked not to be saved by software_suspend() */
--- diff/include/linux/ipmi.h 2003-05-21 10:49:46.000000000 +0000
+++ source/include/linux/ipmi.h 2004-03-16 11:27:48.594800128 +0000
@@ -109,6 +109,35 @@ struct ipmi_ipmb_addr
unsigned char lun;
};
+/*
+ * A LAN Address. This is an address to/from a LAN interface bridged
+ * by the BMC, not an address actually out on the LAN.
+ *
+ * A concious decision was made here to deviate slightly from the IPMI
+ * spec. We do not use rqSWID and rsSWID like it shows in the
+ * message. Instead, we use remote_SWID and local_SWID. This means
+ * that any message (a request or response) from another device will
+ * always have exactly the same address. If you didn't do this,
+ * requests and responses from the same device would have different
+ * addresses, and that's not too cool.
+ *
+ * In this address, the remote_SWID is always the SWID the remote
+ * message came from, or the SWID we are sending the message to.
+ * local_SWID is always our SWID. Note that having our SWID in the
+ * message is a little wierd, but this is required.
+ */
+#define IPMI_LAN_ADDR_TYPE 0x04
+struct ipmi_lan_addr
+{
+ int addr_type;
+ short channel;
+ unsigned char privilege;
+ unsigned char session_handle;
+ unsigned char remote_SWID;
+ unsigned char local_SWID;
+ unsigned char lun;
+};
+
/*
* Channel for talking directly with the BMC. When using this
@@ -145,10 +174,20 @@ struct ipmi_msg
* Receive types for messages coming from the receive interface. This
* is used for the receive in-kernel interface and in the receive
* IOCTL.
+ *
+ * The "IPMI_RESPONSE_RESPNOSE_TYPE" is a little strange sounding, but
+ * it allows you to get the message results when you send a response
+ * message.
*/
#define IPMI_RESPONSE_RECV_TYPE 1 /* A response to a command */
#define IPMI_ASYNC_EVENT_RECV_TYPE 2 /* Something from the event queue */
#define IPMI_CMD_RECV_TYPE 3 /* A command from somewhere else */
+#define IPMI_RESPONSE_RESPONSE_TYPE 4 /* The response for
+ a sent response, giving any
+ error status for sending the
+ response. When you send a
+ response message, this will
+ be returned. */
/* Note that async events and received commands do not have a completion
code as the first byte of the incoming data, unlike a response. */
@@ -160,6 +199,7 @@ struct ipmi_msg
* The in-kernel interface.
*/
#include <linux/list.h>
+#include <linux/module.h>
/* Opaque type for a IPMI message user. One of these is needed to
send and receive messages. */
@@ -185,6 +225,12 @@ struct ipmi_recv_msg
long msgid;
struct ipmi_msg msg;
+ /* The user_msg_data is the data supplied when a message was
+ sent, if this is a response to a sent message. If this is
+ not a response to a sent message, then user_msg_data will
+ be NULL. */
+ void *user_msg_data;
+
/* Call this when done with the message. It will presumably free
the message and do any other necessary cleanup. */
void (*done)(struct ipmi_recv_msg *msg);
@@ -206,9 +252,10 @@ struct ipmi_user_hndl
/* Routine type to call when a message needs to be routed to
the upper layer. This will be called with some locks held,
the only IPMI routines that can be called are ipmi_request
- and the alloc/free operations. */
+ and the alloc/free operations. The handler_data is the
+ variable supplied when the receive handler was registered. */
void (*ipmi_recv_hndl)(struct ipmi_recv_msg *msg,
- void *handler_data);
+ void *user_msg_data);
/* Called when the interface detects a watchdog pre-timeout. If
this is NULL, it will be ignored for the user. */
@@ -221,7 +268,12 @@ int ipmi_create_user(unsigned int
void *handler_data,
ipmi_user_t *user);
-/* Destroy the given user of the IPMI layer. */
+/* Destroy the given user of the IPMI layer. Note that after this
+ function returns, the system is guaranteed to not call any
+ callbacks for the user. Thus as long as you destroy all the users
+ before you unload a module, you will be safe. And if you destroy
+ the users before you destroy the callback structures, it should be
+ safe, too. */
int ipmi_destroy_user(ipmi_user_t user);
/* Get the IPMI version of the BMC we are talking to. */
@@ -253,20 +305,51 @@ unsigned char ipmi_get_my_LUN(ipmi_user_
* in the msgid field of the received command. If the priority is >
* 0, the message will go into a high-priority queue and be sent
* first. Otherwise, it goes into a normal-priority queue.
+ * The user_msg_data field will be returned in any response to this
+ * message.
+ *
+ * Note that if you send a response (with the netfn lower bit set),
+ * you *will* get back a SEND_MSG response telling you what happened
+ * when the response was sent. You will not get back a response to
+ * the message itself.
*/
int ipmi_request(ipmi_user_t user,
struct ipmi_addr *addr,
long msgid,
struct ipmi_msg *msg,
+ void *user_msg_data,
int priority);
/*
+ * Like ipmi_request, but lets you specify the number of retries and
+ * the retry time. The retries is the number of times the message
+ * will be resent if no reply is received. If set to -1, the default
+ * value will be used. The retry time is the time in milliseconds
+ * between retries. If set to zero, the default value will be
+ * used.
+ *
+ * Don't use this unless you *really* have to. It's primarily for the
+ * IPMI over LAN converter; since the LAN stuff does its own retries,
+ * it makes no sense to do it here. However, this can be used if you
+ * have unusual requirements.
+ */
+int ipmi_request_settime(ipmi_user_t user,
+ struct ipmi_addr *addr,
+ long msgid,
+ struct ipmi_msg *msg,
+ void *user_msg_data,
+ int priority,
+ int max_retries,
+ unsigned int retry_time_ms);
+
+/*
* Like ipmi_request, but lets you specify the slave return address.
*/
int ipmi_request_with_source(ipmi_user_t user,
struct ipmi_addr *addr,
long msgid,
struct ipmi_msg *msg,
+ void *user_msg_data,
int priority,
unsigned char source_address,
unsigned char source_lun);
@@ -284,6 +367,7 @@ int ipmi_request_supply_msgs(ipmi_user_t
struct ipmi_addr *addr,
long msgid,
struct ipmi_msg *msg,
+ void *user_msg_data,
void *supplied_smi,
struct ipmi_recv_msg *supplied_recv,
int priority);
@@ -331,6 +415,10 @@ struct ipmi_smi_watcher
{
struct list_head link;
+ /* You must set the owner to the current module, if you are in
+ a module (generally just set it to "THIS_MODULE"). */
+ struct module *owner;
+
/* These two are called with read locks held for the interface
the watcher list. So you can add and remove users from the
IPMI interface, send messages, etc., but you cannot add
@@ -422,6 +510,29 @@ struct ipmi_req
#define IPMICTL_SEND_COMMAND _IOR(IPMI_IOC_MAGIC, 13, \
struct ipmi_req)
+/* Messages sent to the interface with timing parameters are this
+ format. */
+struct ipmi_req_settime
+{
+ struct ipmi_req req;
+
+ /* See ipmi_request_settime() above for details on these
+ values. */
+ int retries;
+ unsigned int retry_time_ms;
+};
+/*
+ * Send a message to the interfaces with timing parameters. error values
+ * are:
+ * - EFAULT - an address supplied was invalid.
+ * - EINVAL - The address supplied was not valid, or the command
+ * was not allowed.
+ * - EMSGSIZE - The message to was too large.
+ * - ENOMEM - Buffers could not be allocated for the command.
+ */
+#define IPMICTL_SEND_COMMAND_SETTIME _IOR(IPMI_IOC_MAGIC, 21, \
+ struct ipmi_req_settime)
+
/* Messages received from the interface are this format. */
struct ipmi_recv
{
@@ -513,4 +624,18 @@ struct ipmi_cmdspec
#define IPMICTL_SET_MY_LUN_CMD _IOR(IPMI_IOC_MAGIC, 19, unsigned int)
#define IPMICTL_GET_MY_LUN_CMD _IOR(IPMI_IOC_MAGIC, 20, unsigned int)
+/*
+ * Get/set the default timing values for an interface. You shouldn't
+ * generally mess with these.
+ */
+struct ipmi_timing_parms
+{
+ int retries;
+ unsigned int retry_time_ms;
+};
+#define IPMICTL_SET_TIMING_PARMS_CMD _IOR(IPMI_IOC_MAGIC, 22, \
+ struct ipmi_timing_parms)
+#define IPMICTL_GET_TIMING_PARMS_CMD _IOR(IPMI_IOC_MAGIC, 23, \
+ struct ipmi_timing_parms)
+
#endif /* __LINUX_IPMI_H */
--- diff/include/linux/ipmi_msgdefs.h 2003-10-27 09:20:39.000000000 +0000
+++ source/include/linux/ipmi_msgdefs.h 2004-03-16 11:27:48.595799976 +0000
@@ -53,6 +53,7 @@
#define IPMI_SET_BMC_GLOBAL_ENABLES_CMD 0x2e
#define IPMI_GET_BMC_GLOBAL_ENABLES_CMD 0x2f
#define IPMI_READ_EVENT_MSG_BUFFER_CMD 0x35
+#define IPMI_GET_CHANNEL_INFO_CMD 0x42
#define IPMI_NETFN_STORAGE_REQUEST 0x0a
#define IPMI_NETFN_STORAGE_RESPONSE 0x0b
@@ -61,8 +62,39 @@
/* The default slave address */
#define IPMI_BMC_SLAVE_ADDR 0x20
-#define IPMI_MAX_MSG_LENGTH 80
+/* The BT interface on high-end HP systems supports up to 255 bytes in
+ * one transfer. Its "virtual" BMC supports some commands that are longer
+ * than 128 bytes. Use the full 256, plus NetFn/LUN, Cmd, cCode, plus
+ * some overhead. It would be nice to base this on the "BT Capabilities"
+ * but that's too hard to propogate to the rest of the driver. */
+#define IPMI_MAX_MSG_LENGTH 272 /* multiple of 16 */
-#define IPMI_CC_NO_ERROR 0
+#define IPMI_CC_NO_ERROR 0x00
+#define IPMI_NODE_BUSY_ERR 0xc0
+#define IPMI_ERR_MSG_TRUNCATED 0xc6
+#define IPMI_LOST_ARBITRATION_ERR 0x81
+#define IPMI_ERR_UNSPECIFIED 0xff
+
+#define IPMI_CHANNEL_PROTOCOL_IPMB 1
+#define IPMI_CHANNEL_PROTOCOL_ICMB 2
+#define IPMI_CHANNEL_PROTOCOL_SMBUS 4
+#define IPMI_CHANNEL_PROTOCOL_KCS 5
+#define IPMI_CHANNEL_PROTOCOL_SMIC 6
+#define IPMI_CHANNEL_PROTOCOL_BT10 7
+#define IPMI_CHANNEL_PROTOCOL_BT15 8
+#define IPMI_CHANNEL_PROTOCOL_TMODE 9
+
+#define IPMI_CHANNEL_MEDIUM_IPMB 1
+#define IPMI_CHANNEL_MEDIUM_ICMB10 2
+#define IPMI_CHANNEL_MEDIUM_ICMB09 3
+#define IPMI_CHANNEL_MEDIUM_8023LAN 4
+#define IPMI_CHANNEL_MEDIUM_ASYNC 5
+#define IPMI_CHANNEL_MEDIUM_OTHER_LAN 6
+#define IPMI_CHANNEL_MEDIUM_PCI_SMBUS 7
+#define IPMI_CHANNEL_MEDIUM_SMBUS1 8
+#define IPMI_CHANNEL_MEDIUM_SMBUS2 9
+#define IPMI_CHANNEL_MEDIUM_USB1 10
+#define IPMI_CHANNEL_MEDIUM_USB2 11
+#define IPMI_CHANNEL_MEDIUM_SYSINTF 12
#endif /* __LINUX_IPMI_MSGDEFS_H */
--- diff/include/linux/ipmi_smi.h 2003-02-26 16:01:09.000000000 +0000
+++ source/include/linux/ipmi_smi.h 2004-03-16 11:27:48.595799976 +0000
@@ -35,6 +35,8 @@
#define __LINUX_IPMI_SMI_H
#include <linux/ipmi_msgdefs.h>
+#include <linux/proc_fs.h>
+#include <linux/module.h>
/* This files describes the interface for IPMI system management interface
drivers to bind into the IPMI message handler. */
@@ -48,7 +50,7 @@ typedef struct ipmi_smi *ipmi_smi_t;
* been received, it will report this same data structure back up to
* the upper layer. If an error occurs, it should fill in the
* response with an error code in the completion code location. When
- * asyncronous data is received, one of these is allocated, the
+ * asynchronous data is received, one of these is allocated, the
* data_size is set to zero and the response holds the data from the
* get message or get event command that the interface initiated.
* Note that it is the interfaces responsibility to detect
@@ -62,9 +64,6 @@ struct ipmi_smi_msg
long msgid;
void *user_data;
- /* If 0, add to the end of the queue. If 1, add to the beginning. */
- int prio;
-
int data_size;
unsigned char data[IPMI_MAX_MSG_LENGTH];
@@ -134,4 +133,11 @@ static inline void ipmi_free_smi_msg(str
msg->done(msg);
}
+/* Allow the lower layer to add things to the proc filesystem
+ directory for this interface. Note that the entry will
+ automatically be dstroyed when the interface is destroyed. */
+int ipmi_smi_add_proc_entry(ipmi_smi_t smi, char *name,
+ read_proc_t *read_proc, write_proc_t *write_proc,
+ void *data, struct module *owner);
+
#endif /* __LINUX_IPMI_SMI_H */
--- diff/include/linux/irq.h 2003-08-26 09:00:54.000000000 +0000
+++ source/include/linux/irq.h 2004-03-16 11:27:48.596799824 +0000
@@ -71,7 +71,6 @@ extern irq_desc_t irq_desc [NR_IRQS];
#include <asm/hw_irq.h> /* the arch dependent stuff */
-extern int handle_IRQ_event(unsigned int, struct pt_regs *, struct irqaction *);
extern int setup_irq(unsigned int , struct irqaction * );
extern hw_irq_controller no_irq_type; /* needed in every arch ? */
--- diff/include/linux/kmalloc_sizes.h 2003-05-21 10:50:00.000000000 +0000
+++ source/include/linux/kmalloc_sizes.h 2004-03-16 11:27:48.596799824 +0000
@@ -12,6 +12,9 @@
CACHE(256)
CACHE(512)
CACHE(1024)
+#if (PAGE_SIZE != 4096) /* special cache for eth skbs - 5 fit into one 8 kB page */
+ CACHE(1620)
+#endif
CACHE(2048)
CACHE(4096)
CACHE(8192)
--- diff/include/linux/list.h 2004-02-18 08:54:13.000000000 +0000
+++ source/include/linux/list.h 2004-03-16 11:27:48.597799672 +0000
@@ -142,8 +142,11 @@ static inline void __list_del(struct lis
* Note: list_empty on entry does not return true after this, the entry is
* in an undefined state.
*/
+#include <linux/kernel.h> /* BUG_ON */
static inline void list_del(struct list_head *entry)
{
+ BUG_ON(entry->prev->next != entry);
+ BUG_ON(entry->next->prev != entry);
__list_del(entry->prev, entry->next);
entry->next = LIST_POISON1;
entry->prev = LIST_POISON2;
--- diff/include/linux/miscdevice.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/linux/miscdevice.h 2004-03-16 11:27:48.598799520 +0000
@@ -3,7 +3,6 @@
#include <linux/module.h>
#include <linux/major.h>
-#define BUSMOUSE_MINOR 0
#define PSMOUSE_MINOR 1
#define MS_BUSMOUSE_MINOR 2
#define ATIXL_BUSMOUSE_MINOR 3
--- diff/include/linux/mm.h 2004-03-11 10:20:29.000000000 +0000
+++ source/include/linux/mm.h 2004-03-16 11:27:48.599799368 +0000
@@ -152,6 +152,12 @@ struct pte_chain;
struct mmu_gather;
struct inode;
+#ifdef ARCH_HAS_ATOMIC_UNSIGNED
+typedef unsigned page_flags_t;
+#else
+typedef unsigned long page_flags_t;
+#endif
+
/*
* Each physical page in the system has a struct page associated with
* it to keep track of whatever it is we are using the page for at the
@@ -168,12 +174,11 @@ struct inode;
* TODO: make this structure smaller, it could be as small as 32 bytes.
*/
struct page {
- unsigned long flags; /* atomic flags, some possibly
+ page_flags_t flags; /* atomic flags, some possibly
updated asynchronously */
atomic_t count; /* Usage count, see below. */
- struct list_head list; /* ->mapping has some page lists. */
struct address_space *mapping; /* The inode (or ...) we belong to. */
- unsigned long index; /* Our offset within mapping. */
+ pgoff_t index; /* Our offset within mapping. */
struct list_head lru; /* Pageout list, eg. active_list;
protected by zone->lru_lock !! */
union {
@@ -233,24 +238,24 @@ extern void FASTCALL(__page_cache_releas
static inline int page_count(struct page *p)
{
if (PageCompound(p))
- p = (struct page *)p->lru.next;
+ p = (struct page *)p->private;
return atomic_read(&(p)->count);
}
static inline void get_page(struct page *page)
{
if (PageCompound(page))
- page = (struct page *)page->lru.next;
+ page = (struct page *)page->private;
atomic_inc(&page->count);
}
static inline void put_page(struct page *page)
{
if (PageCompound(page)) {
- page = (struct page *)page->lru.next;
+ page = (struct page *)page->private;
if (put_page_testzero(page)) {
- if (page->lru.prev) { /* destructor? */
- (*(void (*)(struct page *))page->lru.prev)(page);
+ if (page[1].mapping) { /* destructor? */
+ (*(void (*)(struct page *))page[1].mapping)(page);
} else {
__page_cache_release(page);
}
@@ -333,7 +338,7 @@ static inline void put_page(struct page
* We'll have up to (MAX_NUMNODES * MAX_NR_ZONES) zones total,
* so we use (MAX_NODES_SHIFT + MAX_ZONES_SHIFT) here to get enough bits.
*/
-#define NODEZONE_SHIFT (BITS_PER_LONG - MAX_NODES_SHIFT - MAX_ZONES_SHIFT)
+#define NODEZONE_SHIFT (sizeof(page_flags_t)*8 - MAX_NODES_SHIFT - MAX_ZONES_SHIFT)
#define NODEZONE(node, zone) ((node << ZONES_SHIFT) | zone)
static inline unsigned long page_zonenum(struct page *page)
@@ -415,10 +420,11 @@ static inline int page_mapped(struct pag
* Used to decide whether a process gets delivered SIGBUS or
* just gets major/minor fault counters bumped up.
*/
-#define VM_FAULT_OOM (-1)
-#define VM_FAULT_SIGBUS 0
-#define VM_FAULT_MINOR 1
-#define VM_FAULT_MAJOR 2
+#define VM_FAULT_OOM (-1)
+#define VM_FAULT_SIGBUS 0
+#define VM_FAULT_MINOR 1
+#define VM_FAULT_MAJOR 2
+#define VM_FAULT_SIGSEGV 3
#define offset_in_page(p) ((unsigned long)(p) & ~PAGE_MASK)
@@ -451,9 +457,10 @@ extern pmd_t *FASTCALL(__pmd_alloc(struc
extern pte_t *FASTCALL(pte_alloc_kernel(struct mm_struct *mm, pmd_t *pmd, unsigned long address));
extern pte_t *FASTCALL(pte_alloc_map(struct mm_struct *mm, pmd_t *pmd, unsigned long address));
extern int install_page(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long addr, struct page *page, pgprot_t prot);
-extern int install_file_pte(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long addr, unsigned long pgoff, pgprot_t prot);
+extern int install_file_pte(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long addr, unsigned long pgoff, pgprot_t prot, int linear);
extern int handle_mm_fault(struct mm_struct *mm,struct vm_area_struct *vma, unsigned long address, int write_access);
extern int make_pages_present(unsigned long addr, unsigned long end);
+extern long __remap_file_pages(struct mm_struct *mm, unsigned long start, unsigned long size, unsigned long prot, unsigned long pgoff, unsigned long flags);
extern int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write);
void put_dirty_page(struct task_struct *tsk, struct page *page,
unsigned long address, pgprot_t prot);
@@ -634,6 +641,30 @@ extern struct page * follow_page(struct
extern int remap_page_range(struct vm_area_struct *vma, unsigned long from,
unsigned long to, unsigned long size, pgprot_t prot);
+#ifdef CONFIG_NUMA
+static inline void zero_rss(struct mm_struct *mm)
+{
+ mm->rss = 0;
+ memset(mm->pernode_rss, 0, MAX_NUMNODES * sizeof(*mm->pernode_rss));
+}
+
+static inline void inc_rss(struct mm_struct *mm, struct page *page)
+{
+ mm->rss++;
+ mm->pernode_rss[page_nodenum(page)]++;
+}
+
+static inline void dec_rss(struct mm_struct *mm, struct page *page)
+{
+ mm->rss--;
+ mm->pernode_rss[page_nodenum(page)]--;
+}
+#else /* !CONFIG_NUMA */
+#define zero_rss(mm) ((mm)->rss = 0)
+#define inc_rss(mm, page) ((mm)->rss++)
+#define dec_rss(mm, page) ((mm)->rss--)
+#endif /* CONFIG_NUMA */
+
#ifndef CONFIG_DEBUG_PAGEALLOC
static inline void
kernel_map_pages(struct page *page, int numpages, int enable)
--- diff/include/linux/mmzone.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/linux/mmzone.h 2004-03-16 11:27:48.600799216 +0000
@@ -316,7 +316,7 @@ extern struct pglist_data contig_page_da
#include <asm/mmzone.h>
-#if BITS_PER_LONG == 32
+#if BITS_PER_LONG == 32 || defined(ARCH_HAS_ATOMIC_UNSIGNED)
/*
* with 32 bit page->flags field, we reserve 8 bits for node/zone info.
* there are 3 zones (2 bits) and this leaves 8-2=6 bits for nodes.
--- diff/include/linux/module.h 2004-03-11 10:20:29.000000000 +0000
+++ source/include/linux/module.h 2004-03-16 11:27:48.601799064 +0000
@@ -64,11 +64,12 @@ void sort_main_extable(void);
#define __MODULE_INFO(tag, name, info) \
static const char __module_cat(name,__LINE__)[] \
__attribute_used__ \
- __attribute__((section(".modinfo"),unused)) = __stringify(tag) "=" info
+ __attribute__((section(".modinfo"))) = __stringify(tag) "=" info
#define MODULE_GENERIC_TABLE(gtype,name) \
extern const struct gtype##_id __mod_##gtype##_table \
- __attribute__ ((unused, alias(__stringify(name))))
+ __attribute_used__ \
+ __attribute__ ((alias(__stringify(name))))
#define THIS_MODULE (&__this_module)
@@ -165,7 +166,7 @@ void *__symbol_get_gpl(const char *symbo
extern void *__crc_##sym __attribute__((weak)); \
static const unsigned long __kcrctab_##sym \
__attribute_used__ \
- __attribute__((section("__kcrctab" sec), unused)) \
+ __attribute__((section("__kcrctab" sec))) \
= (unsigned long) &__crc_##sym;
#else
#define __CRC_SYMBOL(sym, sec)
@@ -179,7 +180,7 @@ void *__symbol_get_gpl(const char *symbo
= MODULE_SYMBOL_PREFIX #sym; \
static const struct kernel_symbol __ksymtab_##sym \
__attribute_used__ \
- __attribute__((section("__ksymtab" sec), unused)) \
+ __attribute__((section("__ksymtab" sec))) \
= { (unsigned long)&sym, __kstrtab_##sym }
#define EXPORT_SYMBOL(sym) \
--- diff/include/linux/moduleparam.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/linux/moduleparam.h 2004-03-16 11:27:48.601799064 +0000
@@ -126,13 +126,16 @@ extern int param_get_invbool(char *buffe
#define param_check_invbool(name, p) __param_check(name, p, int)
/* Comma-separated array: num is set to number they actually specified. */
-#define module_param_array(name, type, num, perm) \
+#define module_param_array_named(name, array, type, num, perm) \
static struct kparam_array __param_arr_##name \
- = { ARRAY_SIZE(name), &num, param_set_##type, param_get_##type, \
- sizeof(name[0]), name }; \
+ = { ARRAY_SIZE(array), &num, param_set_##type, param_get_##type,\
+ sizeof(array[0]), array }; \
module_param_call(name, param_array_set, param_array_get, \
&__param_arr_##name, perm)
+#define module_param_array(name, type, num, perm) \
+ module_param_array_named(name, name, type, num, perm)
+
extern int param_array_set(const char *val, struct kernel_param *kp);
extern int param_array_get(char *buffer, struct kernel_param *kp);
--- diff/include/linux/msg.h 2004-03-11 10:20:29.000000000 +0000
+++ source/include/linux/msg.h 2004-03-16 11:27:48.603798760 +0000
@@ -74,9 +74,6 @@ struct msg_msg {
/* the actual message follows immediately */
};
-#define DATALEN_MSG (PAGE_SIZE-sizeof(struct msg_msg))
-#define DATALEN_SEG (PAGE_SIZE-sizeof(struct msg_msgseg))
-
/* one msq_queue structure for each present queue on the system */
struct msg_queue {
struct kern_ipc_perm q_perm;
--- diff/include/linux/net.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/linux/net.h 2004-03-16 11:27:48.603798760 +0000
@@ -25,7 +25,7 @@
struct poll_table_struct;
struct inode;
-#define NPROTO 32 /* should be enough for now.. */
+#define NPROTO 64 /* should be enough for now.. */
#define SYS_SOCKET 1 /* sys_socket(2) */
#define SYS_BIND 2 /* sys_bind(2) */
--- diff/include/linux/netlink.h 2004-03-11 10:20:29.000000000 +0000
+++ source/include/linux/netlink.h 2004-03-16 11:27:48.604798608 +0000
@@ -13,6 +13,7 @@
#define NETLINK_XFRM 6 /* ipsec */
#define NETLINK_SELINUX 7 /* SELinux event notifications */
#define NETLINK_ARPD 8
+#define NETLINK_AUDIT 9 /* auditing */
#define NETLINK_ROUTE6 11 /* af_inet6 route comm channel */
#define NETLINK_IP6_FW 13
#define NETLINK_DNRTMSG 14 /* DECnet routing messages */
--- diff/include/linux/nfs_fs.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/linux/nfs_fs.h 2004-03-16 11:27:48.605798456 +0000
@@ -69,6 +69,8 @@
#define FLUSH_SYNC 1 /* file being synced, or contention */
#define FLUSH_WAIT 2 /* wait for completion */
#define FLUSH_STABLE 4 /* commit to stable storage */
+#define FLUSH_LOWPRI 8 /* low priority background flush */
+#define FLUSH_HIGHPRI 16 /* high priority memory reclaim flush */
#ifdef __KERNEL__
@@ -335,10 +337,8 @@ extern int nfs_writepages(struct addres
extern int nfs_flush_incompatible(struct file *file, struct page *page);
extern int nfs_updatepage(struct file *, struct page *, unsigned int, unsigned int);
extern void nfs_writeback_done(struct rpc_task *task);
-extern void nfs_writedata_release(struct rpc_task *task);
#if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
-extern void nfs_commit_release(struct rpc_task *task);
extern void nfs_commit_done(struct rpc_task *);
#endif
@@ -376,14 +376,18 @@ nfs_wb_all(struct inode *inode)
/*
* Write back all requests on one page - we do this before reading it.
*/
-static inline int
-nfs_wb_page(struct inode *inode, struct page* page)
+static inline int nfs_wb_page_priority(struct inode *inode, struct page* page, int how)
{
int error = nfs_sync_inode(inode, page->index, 1,
- FLUSH_WAIT | FLUSH_STABLE);
+ how | FLUSH_WAIT | FLUSH_STABLE);
return (error < 0) ? error : 0;
}
+static inline int nfs_wb_page(struct inode *inode, struct page* page)
+{
+ return nfs_wb_page_priority(inode, page, 0);
+}
+
/* Hack for future NFS swap support */
#ifndef IS_SWAPFILE
# define IS_SWAPFILE(inode) (0)
@@ -397,7 +401,6 @@ extern int nfs_readpages(struct file *,
struct list_head *, unsigned);
extern int nfs_pagein_list(struct list_head *, int);
extern void nfs_readpage_result(struct rpc_task *);
-extern void nfs_readdata_release(struct rpc_task *);
/*
* linux/fs/mount_clnt.c
--- diff/include/linux/nfs_page.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/linux/nfs_page.h 2004-03-16 11:27:48.606798304 +0000
@@ -17,10 +17,14 @@
#include <linux/sunrpc/auth.h>
#include <linux/nfs_xdr.h>
+#include <asm/atomic.h>
+
/*
* Valid flags for a dirty buffer
*/
#define PG_BUSY 0
+#define PG_NEED_COMMIT 1
+#define PG_NEED_RESCHED 2
struct nfs_page {
struct list_head wb_list, /* Defines state of page: */
@@ -31,6 +35,7 @@ struct nfs_page {
struct rpc_cred *wb_cred;
struct nfs4_state *wb_state;
struct page *wb_page; /* page to read in/write out */
+ atomic_t wb_complete; /* i/os we're waiting for */
wait_queue_head_t wb_wait; /* wait queue */
unsigned long wb_index; /* Offset >> PAGE_CACHE_SHIFT */
unsigned int wb_offset, /* Offset & ~PAGE_CACHE_MASK */
@@ -42,6 +47,8 @@ struct nfs_page {
};
#define NFS_WBACK_BUSY(req) (test_bit(PG_BUSY,&(req)->wb_flags))
+#define NFS_NEED_COMMIT(req) (test_bit(PG_NEED_COMMIT,&(req)->wb_flags))
+#define NFS_NEED_RESCHED(req) (test_bit(PG_NEED_RESCHED,&(req)->wb_flags))
extern struct nfs_page *nfs_create_request(struct file *, struct inode *,
struct page *,
@@ -93,8 +100,7 @@ nfs_unlock_request(struct nfs_page *req)
smp_mb__before_clear_bit();
clear_bit(PG_BUSY, &req->wb_flags);
smp_mb__after_clear_bit();
- if (waitqueue_active(&req->wb_wait))
- wake_up_all(&req->wb_wait);
+ wake_up_all(&req->wb_wait);
nfs_release_request(req);
}
@@ -115,6 +121,38 @@ nfs_list_remove_request(struct nfs_page
req->wb_list_head = NULL;
}
+static inline int
+nfs_defer_commit(struct nfs_page *req)
+{
+ if (test_and_set_bit(PG_NEED_COMMIT, &req->wb_flags))
+ return 0;
+ return 1;
+}
+
+static inline void
+nfs_clear_commit(struct nfs_page *req)
+{
+ smp_mb__before_clear_bit();
+ clear_bit(PG_NEED_COMMIT, &req->wb_flags);
+ smp_mb__after_clear_bit();
+}
+
+static inline int
+nfs_defer_reschedule(struct nfs_page *req)
+{
+ if (test_and_set_bit(PG_NEED_RESCHED, &req->wb_flags))
+ return 0;
+ return 1;
+}
+
+static inline void
+nfs_clear_reschedule(struct nfs_page *req)
+{
+ smp_mb__before_clear_bit();
+ clear_bit(PG_NEED_RESCHED, &req->wb_flags);
+ smp_mb__after_clear_bit();
+}
+
static inline struct nfs_page *
nfs_list_entry(struct list_head *head)
{
--- diff/include/linux/nfs_xdr.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/linux/nfs_xdr.h 2004-03-16 11:27:48.607798152 +0000
@@ -229,7 +229,8 @@ struct nfs_lockres {
struct nfs_readargs {
struct nfs_fh * fh;
- nfs4_stateid stateid;
+ fl_owner_t lockowner;
+ struct nfs4_state * state;
__u64 offset;
__u32 count;
unsigned int pgbase;
@@ -252,7 +253,8 @@ struct nfs_readres {
struct nfs_writeargs {
struct nfs_fh * fh;
- nfs4_stateid stateid;
+ fl_owner_t lockowner;
+ struct nfs4_state * state;
__u64 offset;
__u32 count;
enum nfs3_stable_how stable;
@@ -656,20 +658,23 @@ struct nfs4_compound {
#endif /* CONFIG_NFS_V4 */
+struct nfs_page;
+
struct nfs_read_data {
int flags;
struct rpc_task task;
struct inode *inode;
struct rpc_cred *cred;
- fl_owner_t lockowner;
struct nfs_fattr fattr; /* fattr storage */
struct list_head pages; /* Coalesced read requests */
+ struct nfs_page *req; /* multi ops per nfs_page */
struct page *pagevec[NFS_READ_MAXIOV];
struct nfs_readargs args;
struct nfs_readres res;
#ifdef CONFIG_NFS_V4
unsigned long timestamp; /* For lease renewal */
#endif
+ void (*complete) (struct nfs_read_data *, int);
};
struct nfs_write_data {
@@ -677,20 +682,19 @@ struct nfs_write_data {
struct rpc_task task;
struct inode *inode;
struct rpc_cred *cred;
- fl_owner_t lockowner;
struct nfs_fattr fattr;
struct nfs_writeverf verf;
struct list_head pages; /* Coalesced requests we wish to flush */
+ struct nfs_page *req; /* multi ops per nfs_page */
struct page *pagevec[NFS_WRITE_MAXIOV];
struct nfs_writeargs args; /* argument struct */
struct nfs_writeres res; /* result struct */
#ifdef CONFIG_NFS_V4
unsigned long timestamp; /* For lease renewal */
#endif
+ void (*complete) (struct nfs_write_data *, int);
};
-struct nfs_page;
-
/*
* RPC procedure vector for NFSv2/NFSv3 demuxing
*/
@@ -737,9 +741,9 @@ struct nfs_rpc_ops {
int (*pathconf) (struct nfs_server *, struct nfs_fh *,
struct nfs_pathconf *);
u32 * (*decode_dirent)(u32 *, struct nfs_entry *, int plus);
- void (*read_setup) (struct nfs_read_data *, unsigned int count);
- void (*write_setup) (struct nfs_write_data *, unsigned int count, int how);
- void (*commit_setup) (struct nfs_write_data *, u64 start, u32 len, int how);
+ void (*read_setup) (struct nfs_read_data *);
+ void (*write_setup) (struct nfs_write_data *, int how);
+ void (*commit_setup) (struct nfs_write_data *, int how);
int (*file_open) (struct inode *, struct file *);
int (*file_release) (struct inode *, struct file *);
void (*request_init)(struct nfs_page *, struct file *);
--- diff/include/linux/page-flags.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/linux/page-flags.h 2004-03-16 11:27:48.607798152 +0000
@@ -80,9 +80,6 @@
/*
* Global page accounting. One instance per CPU. Only unsigned longs are
* allowed.
- *
- * NOTE: if this structure is changed then mm/page_alloc.c and
- * arch/s390/appldata/appldata_mem.c must be updated accordingly
*/
struct page_state {
unsigned long nr_dirty; /* Dirty writeable pages */
@@ -315,10 +312,18 @@ extern struct address_space swapper_spac
struct page; /* forward declaration */
int test_clear_page_dirty(struct page *page);
+int __clear_page_dirty(struct page *page);
+int test_clear_page_writeback(struct page *page);
+int test_set_page_writeback(struct page *page);
static inline void clear_page_dirty(struct page *page)
{
test_clear_page_dirty(page);
}
+static inline void set_page_writeback(struct page *page)
+{
+ test_set_page_writeback(page);
+}
+
#endif /* PAGE_FLAGS_H */
--- diff/include/linux/pagemap.h 2004-01-19 10:22:59.000000000 +0000
+++ source/include/linux/pagemap.h 2004-03-16 11:27:48.608798000 +0000
@@ -69,9 +69,10 @@ extern struct page * find_trylock_page(s
unsigned long index);
extern struct page * find_or_create_page(struct address_space *mapping,
unsigned long index, unsigned int gfp_mask);
-extern unsigned int find_get_pages(struct address_space *mapping,
- pgoff_t start, unsigned int nr_pages,
- struct page **pages);
+unsigned find_get_pages(struct address_space *mapping, pgoff_t start,
+ unsigned int nr_pages, struct page **pages);
+unsigned find_get_pages_tag(struct address_space *mapping, pgoff_t *index,
+ int tag, unsigned int nr_pages, struct page **pages);
/*
* Returns locked page at given index in given cache, creating it if needed.
@@ -141,7 +142,6 @@ static inline unsigned long get_page_cac
static inline void ___add_to_page_cache(struct page *page,
struct address_space *mapping, unsigned long index)
{
- list_add(&page->list, &mapping->clean_pages);
page->mapping = mapping;
page->index = index;
--- diff/include/linux/pagevec.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/linux/pagevec.h 2004-03-16 11:27:48.608798000 +0000
@@ -22,8 +22,11 @@ void __pagevec_free(struct pagevec *pvec
void __pagevec_lru_add(struct pagevec *pvec);
void __pagevec_lru_add_active(struct pagevec *pvec);
void pagevec_strip(struct pagevec *pvec);
-unsigned int pagevec_lookup(struct pagevec *pvec, struct address_space *mapping,
- pgoff_t start, unsigned int nr_pages);
+unsigned pagevec_lookup(struct pagevec *pvec, struct address_space *mapping,
+ pgoff_t start, unsigned nr_pages);
+unsigned pagevec_lookup_tag(struct pagevec *pvec,
+ struct address_space *mapping, pgoff_t *index, int tag,
+ unsigned nr_pages);
static inline void pagevec_init(struct pagevec *pvec, int cold)
{
--- diff/include/linux/pci_ids.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/linux/pci_ids.h 2004-03-16 11:27:48.610797696 +0000
@@ -342,6 +342,8 @@
#define PCI_DEVICE_ID_ATI_RS300_133 0x5831
#define PCI_DEVICE_ID_ATI_RS300_166 0x5832
#define PCI_DEVICE_ID_ATI_RS300_200 0x5833
+/* ATI IXP Chipset */
+#define PCI_DEVICE_ID_ATI_IXP_IDE 0x4349
#define PCI_VENDOR_ID_VLSI 0x1004
#define PCI_DEVICE_ID_VLSI_82C592 0x0005
--- diff/include/linux/quota.h 2004-01-19 10:22:59.000000000 +0000
+++ source/include/linux/quota.h 2004-03-16 11:27:48.611797544 +0000
@@ -138,6 +138,10 @@ struct if_dqinfo {
#include <linux/dqblk_v1.h>
#include <linux/dqblk_v2.h>
+/* Maximal numbers of writes for quota operation (insert/delete/update)
+ * (over all formats) - info block, 4 pointer blocks, data block */
+#define DQUOT_MAX_WRITES 6
+
/*
* Data for one user/group kept in memory
*/
@@ -168,22 +172,21 @@ struct mem_dqinfo {
} u;
};
+struct super_block;
+
#define DQF_MASK 0xffff /* Mask for format specific flags */
#define DQF_INFO_DIRTY_B 16
#define DQF_ANY_DQUOT_DIRTY_B 17
#define DQF_INFO_DIRTY (1 << DQF_INFO_DIRTY_B) /* Is info dirty? */
#define DQF_ANY_DQUOT_DIRTY (1 << DQF_ANY_DQUOT_DIRTY_B) /* Is any dquot dirty? */
-extern inline void mark_info_dirty(struct mem_dqinfo *info)
-{
- set_bit(DQF_INFO_DIRTY_B, &info->dqi_flags);
-}
-
+extern void mark_info_dirty(struct super_block *sb, int type);
#define info_dirty(info) test_bit(DQF_INFO_DIRTY_B, &(info)->dqi_flags)
#define info_any_dquot_dirty(info) test_bit(DQF_ANY_DQUOT_DIRTY_B, &(info)->dqi_flags)
#define info_any_dirty(info) (info_dirty(info) || info_any_dquot_dirty(info))
#define sb_dqopt(sb) (&(sb)->s_dquot)
+#define sb_dqinfo(sb, type) (sb_dqopt(sb)->info+(type))
struct dqstats {
int lookups;
@@ -200,15 +203,13 @@ extern struct dqstats dqstats;
#define NR_DQHASH 43 /* Just an arbitrary number */
-#define DQ_MOD_B 0
-#define DQ_BLKS_B 1
-#define DQ_INODES_B 2
-#define DQ_FAKE_B 3
-
-#define DQ_MOD (1 << DQ_MOD_B) /* dquot modified since read */
-#define DQ_BLKS (1 << DQ_BLKS_B) /* uid/gid has been warned about blk limit */
-#define DQ_INODES (1 << DQ_INODES_B) /* uid/gid has been warned about inode limit */
-#define DQ_FAKE (1 << DQ_FAKE_B) /* no limits only usage */
+#define DQ_MOD_B 0 /* dquot modified since read */
+#define DQ_BLKS_B 1 /* uid/gid has been warned about blk limit */
+#define DQ_INODES_B 2 /* uid/gid has been warned about inode limit */
+#define DQ_FAKE_B 3 /* no limits only usage */
+#define DQ_READ_B 4 /* dquot was read into memory */
+#define DQ_ACTIVE_B 5 /* dquot is active (dquot_release not called) */
+#define DQ_WAITFREE_B 6 /* dquot being waited (by invalidate_dquots) */
struct dquot {
struct list_head dq_hash; /* Hash list in memory */
@@ -216,8 +217,7 @@ struct dquot {
struct list_head dq_free; /* Free list element */
struct semaphore dq_lock; /* dquot IO lock */
atomic_t dq_count; /* Use count */
-
- /* fields after this point are cleared when invalidating */
+ wait_queue_head_t dq_wait_unused; /* Wait queue for dquot to become unused */
struct super_block *dq_sb; /* superblock this applies to */
unsigned int dq_id; /* ID this applies to (uid, gid) */
loff_t dq_off; /* Offset of dquot on disk */
@@ -238,19 +238,22 @@ struct quota_format_ops {
int (*write_file_info)(struct super_block *sb, int type); /* Write main info about file */
int (*free_file_info)(struct super_block *sb, int type); /* Called on quotaoff() */
int (*read_dqblk)(struct dquot *dquot); /* Read structure for one user */
- int (*commit_dqblk)(struct dquot *dquot); /* Write (or delete) structure for one user */
+ int (*commit_dqblk)(struct dquot *dquot); /* Write structure for one user */
+ int (*release_dqblk)(struct dquot *dquot); /* Called when last reference to dquot is being dropped */
};
/* Operations working with dquots */
struct dquot_operations {
- void (*initialize) (struct inode *, int);
- void (*drop) (struct inode *);
+ int (*initialize) (struct inode *, int);
+ int (*drop) (struct inode *);
int (*alloc_space) (struct inode *, qsize_t, int);
int (*alloc_inode) (const struct inode *, unsigned long);
- void (*free_space) (struct inode *, qsize_t);
- void (*free_inode) (const struct inode *, unsigned long);
+ int (*free_space) (struct inode *, qsize_t);
+ int (*free_inode) (const struct inode *, unsigned long);
int (*transfer) (struct inode *, struct iattr *);
- int (*write_dquot) (struct dquot *);
+ int (*write_dquot) (struct dquot *); /* Ordinary dquot write */
+ int (*mark_dirty) (struct dquot *); /* Dquot is marked dirty */
+ int (*write_info) (struct super_block *, int); /* Write of quota "superblock" */
};
/* Operations handling requests from userspace */
@@ -289,10 +292,7 @@ struct quota_info {
};
/* Inline would be better but we need to dereference super_block which is not defined yet */
-#define mark_dquot_dirty(dquot) do {\
- set_bit(DQF_ANY_DQUOT_DIRTY_B, &(sb_dqopt((dquot)->dq_sb)->info[(dquot)->dq_type].dqi_flags));\
- set_bit(DQ_MOD_B, &(dquot)->dq_flags);\
-} while (0)
+int mark_dquot_dirty(struct dquot *dquot);
#define dquot_dirty(dquot) test_bit(DQ_MOD_B, &(dquot)->dq_flags)
@@ -304,7 +304,6 @@ struct quota_info {
int register_quota_format(struct quota_format_type *fmt);
void unregister_quota_format(struct quota_format_type *fmt);
-void init_dquot_operations(struct dquot_operations *fsdqops);
struct quota_module_name {
int qm_fmt_id;
--- diff/include/linux/quotaops.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/linux/quotaops.h 2004-03-16 11:27:48.611797544 +0000
@@ -22,16 +22,31 @@
*/
extern void sync_dquots(struct super_block *sb, int type);
-extern void dquot_initialize(struct inode *inode, int type);
-extern void dquot_drop(struct inode *inode);
+extern int dquot_initialize(struct inode *inode, int type);
+extern int dquot_drop(struct inode *inode);
-extern int dquot_alloc_space(struct inode *inode, qsize_t number, int prealloc);
-extern int dquot_alloc_inode(const struct inode *inode, unsigned long number);
+extern int dquot_alloc_space(struct inode *inode, qsize_t number, int prealloc);
+extern int dquot_alloc_inode(const struct inode *inode, unsigned long number);
-extern void dquot_free_space(struct inode *inode, qsize_t number);
-extern void dquot_free_inode(const struct inode *inode, unsigned long number);
+extern int dquot_free_space(struct inode *inode, qsize_t number);
+extern int dquot_free_inode(const struct inode *inode, unsigned long number);
-extern int dquot_transfer(struct inode *inode, struct iattr *iattr);
+extern int dquot_transfer(struct inode *inode, struct iattr *iattr);
+extern int dquot_commit(struct dquot *dquot);
+extern int dquot_acquire(struct dquot *dquot);
+extern int dquot_release(struct dquot *dquot);
+extern int dquot_commit_info(struct super_block *sb, int type);
+extern int dquot_mark_dquot_dirty(struct dquot *dquot);
+
+extern int vfs_quota_on(struct super_block *sb, int type, int format_id, char *path);
+extern int vfs_quota_on_mount(int type, int format_id, struct dentry *dentry);
+extern int vfs_quota_off(struct super_block *sb, int type);
+#define vfs_quota_off_mount(sb, type) vfs_quota_off(sb, type)
+extern int vfs_quota_sync(struct super_block *sb, int type);
+extern int vfs_get_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii);
+extern int vfs_set_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii);
+extern int vfs_get_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *di);
+extern int vfs_set_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *di);
/*
* Operations supported for diskquotas.
@@ -42,6 +57,8 @@ extern struct quotactl_ops vfs_quotactl_
#define sb_dquot_ops (&dquot_operations)
#define sb_quotactl_ops (&vfs_quotactl_ops)
+/* It is better to call this function outside of any transaction as it might
+ * need a lot of space in journal for dquot structure allocation. */
static __inline__ void DQUOT_INIT(struct inode *inode)
{
BUG_ON(!inode->i_sb);
@@ -49,6 +66,7 @@ static __inline__ void DQUOT_INIT(struct
inode->i_sb->dq_op->initialize(inode, -1);
}
+/* The same as with DQUOT_INIT */
static __inline__ void DQUOT_DROP(struct inode *inode)
{
if (IS_QUOTAINIT(inode)) {
@@ -57,6 +75,8 @@ static __inline__ void DQUOT_DROP(struct
}
}
+/* The following allocation/freeing/transfer functions *must* be called inside
+ * a transaction (deadlocks possible otherwise) */
static __inline__ int DQUOT_PREALLOC_SPACE_NODIRTY(struct inode *inode, qsize_t nr)
{
if (sb_any_quota_enabled(inode->i_sb)) {
@@ -64,11 +84,8 @@ static __inline__ int DQUOT_PREALLOC_SPA
if (inode->i_sb->dq_op->alloc_space(inode, nr, 1) == NO_QUOTA)
return 1;
}
- else {
- spin_lock(&dq_data_lock);
+ else
inode_add_bytes(inode, nr);
- spin_unlock(&dq_data_lock);
- }
return 0;
}
@@ -87,11 +104,8 @@ static __inline__ int DQUOT_ALLOC_SPACE_
if (inode->i_sb->dq_op->alloc_space(inode, nr, 0) == NO_QUOTA)
return 1;
}
- else {
- spin_lock(&dq_data_lock);
+ else
inode_add_bytes(inode, nr);
- spin_unlock(&dq_data_lock);
- }
return 0;
}
@@ -117,11 +131,8 @@ static __inline__ void DQUOT_FREE_SPACE_
{
if (sb_any_quota_enabled(inode->i_sb))
inode->i_sb->dq_op->free_space(inode, nr);
- else {
- spin_lock(&dq_data_lock);
+ else
inode_sub_bytes(inode, nr);
- spin_unlock(&dq_data_lock);
- }
}
static __inline__ void DQUOT_FREE_SPACE(struct inode *inode, qsize_t nr)
@@ -146,6 +157,7 @@ static __inline__ int DQUOT_TRANSFER(str
return 0;
}
+/* The following two functions cannot be called inside a transaction */
#define DQUOT_SYNC(sb) sync_dquots(sb, -1)
static __inline__ int DQUOT_OFF(struct super_block *sb)
--- diff/include/linux/radix-tree.h 2003-05-21 10:50:10.000000000 +0000
+++ source/include/linux/radix-tree.h 2004-03-16 11:27:48.612797392 +0000
@@ -20,8 +20,7 @@
#define _LINUX_RADIX_TREE_H
#include <linux/preempt.h>
-
-struct radix_tree_node;
+#include <linux/types.h>
struct radix_tree_root {
unsigned int height;
@@ -29,25 +28,40 @@ struct radix_tree_root {
struct radix_tree_node *rnode;
};
-#define RADIX_TREE_INIT(mask) {0, (mask), NULL}
+#define RADIX_TREE_INIT(mask) { \
+ .height = 0, \
+ .gfp_mask = (mask), \
+ .rnode = NULL, \
+}
#define RADIX_TREE(name, mask) \
struct radix_tree_root name = RADIX_TREE_INIT(mask)
-#define INIT_RADIX_TREE(root, mask) \
-do { \
- (root)->height = 0; \
- (root)->gfp_mask = (mask); \
- (root)->rnode = NULL; \
+#define INIT_RADIX_TREE(root, mask) \
+do { \
+ (root)->height = 0; \
+ (root)->gfp_mask = (mask); \
+ (root)->rnode = NULL; \
} while (0)
-extern int radix_tree_insert(struct radix_tree_root *, unsigned long, void *);
-extern void *radix_tree_lookup(struct radix_tree_root *, unsigned long);
-extern void *radix_tree_delete(struct radix_tree_root *, unsigned long);
-extern unsigned int
+int radix_tree_insert(struct radix_tree_root *, unsigned long, void *);
+void *radix_tree_lookup(struct radix_tree_root *, unsigned long);
+void *radix_tree_delete(struct radix_tree_root *, unsigned long);
+unsigned int
radix_tree_gang_lookup(struct radix_tree_root *root, void **results,
unsigned long first_index, unsigned int max_items);
int radix_tree_preload(int gfp_mask);
+void radix_tree_init(void);
+void *radix_tree_tag_set(struct radix_tree_root *root,
+ unsigned long index, int tag);
+void *radix_tree_tag_clear(struct radix_tree_root *root,
+ unsigned long index, int tag);
+int radix_tree_tag_get(struct radix_tree_root *root,
+ unsigned long index, int tag);
+unsigned int
+radix_tree_gang_lookup_tag(struct radix_tree_root *root, void **results,
+ unsigned long first_index, unsigned int max_items, int tag);
+int radix_tree_tagged(struct radix_tree_root *root, int tag);
static inline void radix_tree_preload_end(void)
{
--- diff/include/linux/raid/md.h 2003-06-30 09:07:24.000000000 +0000
+++ source/include/linux/raid/md.h 2004-03-16 11:27:48.612797392 +0000
@@ -76,6 +76,7 @@ extern void md_handle_safemode(mddev_t *
extern void md_done_sync(mddev_t *mddev, int blocks, int ok);
extern void md_sync_acct(mdk_rdev_t *rdev, unsigned long nr_sectors);
extern void md_error (mddev_t *mddev, mdk_rdev_t *rdev);
+extern void md_unplug_mddev(mddev_t *mddev);
extern void md_print_devices (void);
--- diff/include/linux/raid/md_k.h 2004-03-11 10:20:29.000000000 +0000
+++ source/include/linux/raid/md_k.h 2004-03-16 11:27:48.613797240 +0000
@@ -326,7 +326,6 @@ do { \
if (condition) \
break; \
spin_unlock_irq(&lock); \
- blk_run_queues(); \
schedule(); \
spin_lock_irq(&lock); \
} \
@@ -341,30 +340,5 @@ do { \
__wait_event_lock_irq(wq, condition, lock); \
} while (0)
-
-#define __wait_disk_event(wq, condition) \
-do { \
- wait_queue_t __wait; \
- init_waitqueue_entry(&__wait, current); \
- \
- add_wait_queue(&wq, &__wait); \
- for (;;) { \
- set_current_state(TASK_UNINTERRUPTIBLE); \
- if (condition) \
- break; \
- blk_run_queues(); \
- schedule(); \
- } \
- current->state = TASK_RUNNING; \
- remove_wait_queue(&wq, &__wait); \
-} while (0)
-
-#define wait_disk_event(wq, condition) \
-do { \
- if (condition) \
- break; \
- __wait_disk_event(wq, condition); \
-} while (0)
-
#endif
--- diff/include/linux/reiserfs_fs.h 2004-02-18 08:54:13.000000000 +0000
+++ source/include/linux/reiserfs_fs.h 2004-03-16 11:27:48.615796936 +0000
@@ -1335,7 +1335,8 @@ static inline loff_t max_reiserfs_offset
#define fs_generation(s) (REISERFS_SB(s)->s_generation_counter)
#define get_generation(s) atomic_read (&fs_generation(s))
#define FILESYSTEM_CHANGED_TB(tb) (get_generation((tb)->tb_sb) != (tb)->fs_gen)
-#define fs_changed(gen,s) (gen != get_generation (s))
+#define __fs_changed(gen,s) (gen != get_generation (s))
+#define fs_changed(gen,s) ({cond_resched(); __fs_changed(gen, s);})
/***************************************************************************/
--- diff/include/linux/rmap-locking.h 2003-06-30 09:07:34.000000000 +0000
+++ source/include/linux/rmap-locking.h 2004-03-16 11:27:48.615796936 +0000
@@ -10,8 +10,8 @@
struct pte_chain;
extern kmem_cache_t *pte_chain_cache;
-#define pte_chain_lock(page) bit_spin_lock(PG_chainlock, &page->flags)
-#define pte_chain_unlock(page) bit_spin_unlock(PG_chainlock, &page->flags)
+#define pte_chain_lock(page) bit_spin_lock(PG_chainlock, (unsigned long *)&page->flags)
+#define pte_chain_unlock(page) bit_spin_unlock(PG_chainlock, (unsigned long *)&page->flags)
struct pte_chain *pte_chain_alloc(int gfp_flags);
void __pte_chain_free(struct pte_chain *pte_chain);
--- diff/include/linux/sched.h 2004-03-11 10:20:29.000000000 +0000
+++ source/include/linux/sched.h 2004-03-16 11:27:48.616796784 +0000
@@ -147,6 +147,7 @@ extern spinlock_t mmlist_lock;
typedef struct task_struct task_t;
extern void sched_init(void);
+extern void sched_init_smp(void);
extern void init_idle(task_t *idle, int cpu);
extern void show_state(void);
@@ -193,7 +194,7 @@ struct mm_struct {
atomic_t mm_count; /* How many references to "struct mm_struct" (users count as 1) */
int map_count; /* number of VMAs */
struct rw_semaphore mmap_sem;
- spinlock_t page_table_lock; /* Protects task page tables and mm->rss */
+ spinlock_t page_table_lock; /* Protects task page tables and RSS data */
struct list_head mmlist; /* List of all active mm's. These are globally strung
* together off init_mm.mmlist, and are protected
@@ -203,7 +204,11 @@ struct mm_struct {
unsigned long start_code, end_code, start_data, end_data;
unsigned long start_brk, brk, start_stack;
unsigned long arg_start, arg_end, env_start, env_end;
- unsigned long rss, total_vm, locked_vm;
+ unsigned long total_vm, locked_vm;
+ unsigned long rss;
+#ifdef CONFIG_NUMA
+ unsigned long pernode_rss[MAX_NUMNODES];
+#endif
unsigned long def_flags;
unsigned long saved_auxv[40]; /* for /proc/PID/auxv */
@@ -265,6 +270,15 @@ struct signal_struct {
/* thread group stop support, overloads group_exit_code too */
int group_stop_count;
+
+ /* job control IDs */
+ pid_t pgrp;
+ pid_t tty_old_pgrp;
+ pid_t session;
+ /* boolean value for session group leader */
+ int leader;
+
+ struct tty_struct *tty; /* NULL if no tty */
};
/*
@@ -358,6 +372,8 @@ int set_current_groups(struct group_info
((gi)->blocks[(i)/NGROUPS_PER_BLOCK][(i)%NGROUPS_PER_BLOCK])
+struct audit_context; /* See audit.c */
+
struct task_struct {
volatile long state; /* -1 unrunnable, 0 runnable, >0 stopped */
struct thread_info *thread_info;
@@ -394,12 +410,7 @@ struct task_struct {
unsigned long personality;
int did_exec:1;
pid_t pid;
- pid_t __pgrp; /* Accessed via process_group() */
- pid_t tty_old_pgrp;
- pid_t session;
pid_t tgid;
- /* boolean value for session group leader */
- int leader;
/*
* pointers to (original) parent process, youngest child, younger sibling,
* older sibling, respectively. (p->father can be replaced with
@@ -442,7 +453,6 @@ struct task_struct {
char comm[16];
/* file system info */
int link_count, total_link_count;
- struct tty_struct *tty; /* NULL if no tty */
/* ipc stuff */
struct sysv_sem sysvsem;
/* CPU-specific state of this task */
@@ -467,6 +477,7 @@ struct task_struct {
sigset_t *notifier_mask;
void *security;
+ struct audit_context *audit_context;
/* Thread group tracking */
u32 parent_exec_id;
@@ -495,7 +506,7 @@ struct task_struct {
static inline pid_t process_group(struct task_struct *tsk)
{
- return tsk->group_leader->__pgrp;
+ return tsk->signal->pgrp;
}
extern void __put_task_struct(struct task_struct *tsk);
@@ -527,8 +538,111 @@ do { if (atomic_dec_and_test(&(tsk)->usa
#define PF_SWAPOFF 0x00080000 /* I am in swapoff */
#define PF_LESS_THROTTLE 0x00100000 /* Throttle me less: I clean memory */
#define PF_SYNCWRITE 0x00200000 /* I am doing a sync write */
+#define PF_FUTEX_DEBUG 0x00400000
#ifdef CONFIG_SMP
+#define SCHED_LOAD_SHIFT 7 /* increase resolution of load calculations */
+#define SCHED_LOAD_SCALE (1UL << SCHED_LOAD_SHIFT)
+
+#define SD_FLAG_NEWIDLE 1 /* Balance when about to become idle */
+#define SD_FLAG_EXEC 2 /* Balance on exec */
+#define SD_FLAG_WAKE 4 /* Balance on task wakeup */
+#define SD_FLAG_FASTMIGRATE 8 /* Sync wakes put task on waking CPU */
+#define SD_FLAG_SHARE_CPUPOWER 16 /* Domain members share cpu power */
+
+struct sched_group {
+ struct sched_group *next; /* Must be a circular list */
+ cpumask_t cpumask;
+
+ /*
+ * CPU power of this group, SCHED_LOAD_SCALE being max power for a
+ * single CPU. This should be read only (except for setup). Although
+ * it will need to be written to at cpu hot(un)plug time, perhaps the
+ * cpucontrol semaphore will provide enough exclusion?
+ */
+ unsigned long cpu_power;
+};
+
+struct sched_domain {
+ /* These fields must be setup */
+ struct sched_domain *parent; /* top domain must be null terminated */
+ struct sched_group *groups; /* the balancing groups of the domain */
+ cpumask_t span; /* span of all CPUs in this domain */
+ unsigned long min_interval; /* Minimum balance interval ms */
+ unsigned long max_interval; /* Maximum balance interval ms */
+ unsigned int busy_factor; /* less balancing by factor if busy */
+ unsigned int imbalance_pct; /* No balance until over watermark */
+ unsigned long long cache_hot_time; /* Task considered cache hot (ns) */
+ unsigned int cache_nice_tries; /* Leave cache hot tasks for # tries */
+ unsigned int per_cpu_gain; /* CPU % gained by adding domain cpus */
+ int flags; /* See SD_FLAG_* */
+
+ /* Runtime fields. */
+ unsigned long last_balance; /* init to jiffies. units in jiffies */
+ unsigned int balance_interval; /* initialise to 1. units in ms. */
+ unsigned int nr_balance_failed; /* initialise to 0 */
+};
+
+/* Common values for SMT siblings */
+#define SD_SIBLING_INIT (struct sched_domain) { \
+ .span = CPU_MASK_NONE, \
+ .parent = NULL, \
+ .groups = NULL, \
+ .min_interval = 1, \
+ .max_interval = 2, \
+ .busy_factor = 8, \
+ .imbalance_pct = 110, \
+ .cache_hot_time = 0, \
+ .cache_nice_tries = 0, \
+ .per_cpu_gain = 15, \
+ .flags = SD_FLAG_FASTMIGRATE | SD_FLAG_NEWIDLE | SD_FLAG_WAKE,\
+ .last_balance = jiffies, \
+ .balance_interval = 1, \
+ .nr_balance_failed = 0, \
+}
+
+/* Common values for CPUs */
+#define SD_CPU_INIT (struct sched_domain) { \
+ .span = CPU_MASK_NONE, \
+ .parent = NULL, \
+ .groups = NULL, \
+ .min_interval = 1, \
+ .max_interval = 4, \
+ .busy_factor = 64, \
+ .imbalance_pct = 125, \
+ .cache_hot_time = (5*1000000/2), \
+ .cache_nice_tries = 1, \
+ .per_cpu_gain = 100, \
+ .flags = SD_FLAG_FASTMIGRATE | SD_FLAG_NEWIDLE,\
+ .last_balance = jiffies, \
+ .balance_interval = 1, \
+ .nr_balance_failed = 0, \
+}
+
+#ifdef CONFIG_NUMA
+/* Common values for NUMA nodes */
+#define SD_NODE_INIT (struct sched_domain) { \
+ .span = CPU_MASK_NONE, \
+ .parent = NULL, \
+ .groups = NULL, \
+ .min_interval = 8, \
+ .max_interval = 256*fls(num_online_cpus()),\
+ .busy_factor = 8, \
+ .imbalance_pct = 125, \
+ .cache_hot_time = (10*1000000), \
+ .cache_nice_tries = 1, \
+ .per_cpu_gain = 100, \
+ .flags = SD_FLAG_EXEC, \
+ .last_balance = jiffies, \
+ .balance_interval = 1, \
+ .nr_balance_failed = 0, \
+}
+#endif
+
+DECLARE_PER_CPU(struct sched_domain, base_domains);
+#define cpu_sched_domain(cpu) (&per_cpu(base_domains, (cpu)))
+#define this_sched_domain() (&__get_cpu_var(base_domains))
+
extern int set_cpus_allowed(task_t *p, cpumask_t new_mask);
#else
static inline int set_cpus_allowed(task_t *p, cpumask_t new_mask)
@@ -541,10 +655,8 @@ extern unsigned long long sched_clock(vo
#ifdef CONFIG_NUMA
extern void sched_balance_exec(void);
-extern void node_nr_running_init(void);
#else
#define sched_balance_exec() {}
-#define node_nr_running_init() {}
#endif
extern void set_user_nice(task_t *p, long nice);
@@ -560,13 +672,9 @@ void yield(void);
*/
extern struct exec_domain default_exec_domain;
-#ifndef INIT_THREAD_SIZE
-# define INIT_THREAD_SIZE 2048*sizeof(long)
-#endif
-
union thread_union {
struct thread_info thread_info;
- unsigned long stack[INIT_THREAD_SIZE/sizeof(long)];
+ unsigned long stack[THREAD_SIZE/sizeof(long)];
};
#ifndef __HAVE_ARCH_KSTACK_END
--- diff/include/linux/sem.h 2004-03-11 10:20:29.000000000 +0000
+++ source/include/linux/sem.h 2004-03-16 11:27:48.617796632 +0000
@@ -134,7 +134,22 @@ struct sysv_sem {
struct sem_undo_list *undo_list;
};
-void exit_sem(struct task_struct *p);
+#ifdef CONFIG_SYSVIPC
+
+extern int copy_semundo(unsigned long clone_flags, struct task_struct *tsk);
+extern void exit_sem(struct task_struct *tsk);
+
+#else
+static inline int copy_semundo(unsigned long clone_flags, struct task_struct *tsk)
+{
+ return 0;
+}
+
+static inline void exit_sem(struct task_struct *tsk)
+{
+ return;
+}
+#endif
#endif /* __KERNEL__ */
--- diff/include/linux/serial_core.h 2004-03-11 10:20:29.000000000 +0000
+++ source/include/linux/serial_core.h 2004-03-16 11:27:48.617796632 +0000
@@ -160,7 +160,9 @@ struct uart_port {
unsigned char x_char; /* xon/xoff char */
unsigned char regshift; /* reg offset shift */
unsigned char iotype; /* io access style */
-
+#ifdef CONFIG_KGDB
+ int kgdb; /* in use by kgdb */
+#endif
#define UPIO_PORT (0)
#define UPIO_HUB6 (1)
#define UPIO_MEM (2)
--- diff/include/linux/serio.h 2004-01-19 10:22:59.000000000 +0000
+++ source/include/linux/serio.h 2004-03-16 11:27:48.618796480 +0000
@@ -117,6 +117,7 @@ static __inline__ void serio_cleanup(str
#define SERIO_MZ 0x05
#define SERIO_MZP 0x06
#define SERIO_MZPP 0x07
+#define SERIO_VSXXXAA 0x08
#define SERIO_SUNKBD 0x10
#define SERIO_WARRIOR 0x18
#define SERIO_SPACEORB 0x19
@@ -134,6 +135,7 @@ static __inline__ void serio_cleanup(str
#define SERIO_HIL 0x25
#define SERIO_SNES232 0x26
#define SERIO_SEMTECH 0x27
+#define SERIO_LKKBD 0x28
#define SERIO_ID 0xff00UL
#define SERIO_EXTRA 0xff0000UL
--- diff/include/linux/socket.h 2004-03-11 10:20:29.000000000 +0000
+++ source/include/linux/socket.h 2004-03-16 11:27:48.618796480 +0000
@@ -178,7 +178,8 @@ struct ucred {
#define AF_WANPIPE 25 /* Wanpipe API Sockets */
#define AF_LLC 26 /* Linux LLC */
#define AF_BLUETOOTH 31 /* Bluetooth sockets */
-#define AF_MAX 32 /* For now.. */
+#define AF_IPMI 32 /* IPMI sockers */
+#define AF_MAX 33 /* For now.. */
/* Protocol families, same as address families. */
#define PF_UNSPEC AF_UNSPEC
@@ -210,6 +211,7 @@ struct ucred {
#define PF_WANPIPE AF_WANPIPE
#define PF_LLC AF_LLC
#define PF_BLUETOOTH AF_BLUETOOTH
+#define PF_IPMI AF_IPMI
#define PF_MAX AF_MAX
/* Maximum queue length specifiable by listen. */
--- diff/include/linux/spinlock.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/linux/spinlock.h 2004-03-16 11:27:48.620796176 +0000
@@ -15,6 +15,12 @@
#include <asm/processor.h> /* for cpu relax */
#include <asm/system.h>
+#ifdef CONFIG_KGDB
+#include <asm/current.h>
+#define SET_WHO(x, him) (x)->who = him;
+#else
+#define SET_WHO(x, him)
+#endif
/*
* Must define these before including other files, inline functions need them
@@ -55,6 +61,9 @@ typedef struct {
const char *module;
char *owner;
int oline;
+#ifdef CONFIG_KGDB
+ struct task_struct *who;
+#endif
} spinlock_t;
#define SPIN_LOCK_UNLOCKED (spinlock_t) { SPINLOCK_MAGIC, 0, 10, __FILE__ , NULL, 0}
@@ -66,6 +75,7 @@ typedef struct {
(x)->module = __FILE__; \
(x)->owner = NULL; \
(x)->oline = 0; \
+ SET_WHO(x, NULL) \
} while (0)
#define CHECK_LOCK(x) \
@@ -88,6 +98,7 @@ typedef struct {
(x)->lock = 1; \
(x)->owner = __FILE__; \
(x)->oline = __LINE__; \
+ SET_WHO(x, current) \
} while (0)
/* without debugging, spin_is_locked on UP always says
@@ -118,6 +129,7 @@ typedef struct {
(x)->lock = 1; \
(x)->owner = __FILE__; \
(x)->oline = __LINE__; \
+ SET_WHO(x, current) \
1; \
})
@@ -184,6 +196,17 @@ typedef struct {
#endif /* !SMP */
+#ifdef CONFIG_LOCKMETER
+extern void _metered_spin_lock (spinlock_t *lock);
+extern void _metered_spin_unlock (spinlock_t *lock);
+extern int _metered_spin_trylock(spinlock_t *lock);
+extern void _metered_read_lock (rwlock_t *lock);
+extern void _metered_read_unlock (rwlock_t *lock);
+extern void _metered_write_lock (rwlock_t *lock);
+extern void _metered_write_unlock (rwlock_t *lock);
+extern int _metered_write_trylock(rwlock_t *lock);
+#endif
+
/*
* Define the various spin_lock and rw_lock methods. Note we define these
* regardless of whether CONFIG_SMP or CONFIG_PREEMPT are set. The various
@@ -389,6 +412,141 @@ do { \
_raw_spin_trylock(lock) ? 1 : \
({preempt_enable(); local_bh_enable(); 0;});})
+#ifdef CONFIG_LOCKMETER
+#undef spin_lock
+#undef spin_trylock
+#undef spin_unlock
+#undef spin_lock_irqsave
+#undef spin_lock_irq
+#undef spin_lock_bh
+#undef read_lock
+#undef read_unlock
+#undef write_lock
+#undef write_unlock
+#undef write_trylock
+#undef spin_unlock_bh
+#undef read_lock_irqsave
+#undef read_lock_irq
+#undef read_lock_bh
+#undef read_unlock_bh
+#undef write_lock_irqsave
+#undef write_lock_irq
+#undef write_lock_bh
+#undef write_unlock_bh
+
+#define spin_lock(lock) \
+do { \
+ preempt_disable(); \
+ _metered_spin_lock(lock); \
+} while(0)
+
+#define spin_trylock(lock) ({preempt_disable(); _metered_spin_trylock(lock) ? \
+ 1 : ({preempt_enable(); 0;});})
+#define spin_unlock(lock) \
+do { \
+ _metered_spin_unlock(lock); \
+ preempt_enable(); \
+} while (0)
+
+#define spin_lock_irqsave(lock, flags) \
+do { \
+ local_irq_save(flags); \
+ preempt_disable(); \
+ _metered_spin_lock(lock); \
+} while (0)
+
+#define spin_lock_irq(lock) \
+do { \
+ local_irq_disable(); \
+ preempt_disable(); \
+ _metered_spin_lock(lock); \
+} while (0)
+
+#define spin_lock_bh(lock) \
+do { \
+ local_bh_disable(); \
+ preempt_disable(); \
+ _metered_spin_lock(lock); \
+} while (0)
+
+#define spin_unlock_bh(lock) \
+do { \
+ _metered_spin_unlock(lock); \
+ preempt_enable(); \
+ local_bh_enable(); \
+} while (0)
+
+
+#define read_lock(lock) ({preempt_disable(); _metered_read_lock(lock);})
+#define read_unlock(lock) ({_metered_read_unlock(lock); preempt_enable();})
+#define write_lock(lock) ({preempt_disable(); _metered_write_lock(lock);})
+#define write_unlock(lock) ({_metered_write_unlock(lock); preempt_enable();})
+#define write_trylock(lock) ({preempt_disable();_metered_write_trylock(lock) ? \
+ 1 : ({preempt_enable(); 0;});})
+#define spin_unlock_no_resched(lock) \
+do { \
+ _metered_spin_unlock(lock); \
+ preempt_enable_no_resched(); \
+} while (0)
+
+#define read_lock_irqsave(lock, flags) \
+do { \
+ local_irq_save(flags); \
+ preempt_disable(); \
+ _metered_read_lock(lock); \
+} while (0)
+
+#define read_lock_irq(lock) \
+do { \
+ local_irq_disable(); \
+ preempt_disable(); \
+ _metered_read_lock(lock); \
+} while (0)
+
+#define read_lock_bh(lock) \
+do { \
+ local_bh_disable(); \
+ preempt_disable(); \
+ _metered_read_lock(lock); \
+} while (0)
+
+#define read_unlock_bh(lock) \
+do { \
+ _metered_read_unlock(lock); \
+ preempt_enable(); \
+ local_bh_enable(); \
+} while (0)
+
+#define write_lock_irqsave(lock, flags) \
+do { \
+ local_irq_save(flags); \
+ preempt_disable(); \
+ _metered_write_lock(lock); \
+} while (0)
+
+#define write_lock_irq(lock) \
+do { \
+ local_irq_disable(); \
+ preempt_disable(); \
+ _metered_write_lock(lock); \
+} while (0)
+
+#define write_lock_bh(lock) \
+do { \
+ local_bh_disable(); \
+ preempt_disable(); \
+ _metered_write_lock(lock); \
+} while (0)
+
+#define write_unlock_bh(lock) \
+do { \
+ _metered_write_unlock(lock); \
+ preempt_enable(); \
+ local_bh_enable(); \
+} while (0)
+
+#endif /* !CONFIG_LOCKMETER */
+
/* "lock on reference count zero" */
#ifndef ATOMIC_DEC_AND_LOCK
#include <asm/atomic.h>
--- diff/include/linux/sunrpc/sched.h 2004-02-18 08:54:13.000000000 +0000
+++ source/include/linux/sunrpc/sched.h 2004-03-16 11:27:48.620796176 +0000
@@ -49,6 +49,8 @@ struct rpc_task {
tk_cred_retry,
tk_suid_retry;
+ unsigned long tk_cookie; /* Cookie for batching tasks */
+
/*
* timeout_fn to be executed by timer bottom half
* callback to be executed after waking up
@@ -72,7 +74,9 @@ struct rpc_task {
unsigned long tk_timeout; /* timeout for rpc_sleep() */
unsigned short tk_flags; /* misc flags */
unsigned char tk_active : 1;/* Task has been activated */
+ unsigned char tk_priority : 2;/* Task priority */
unsigned long tk_runstate; /* Task run status */
+ struct list_head tk_links; /* links to related tasks */
#ifdef RPC_DEBUG
unsigned short tk_pid; /* debugging aid */
#endif
@@ -138,28 +142,58 @@ typedef void (*rpc_action)(struct rpc_
} while(0)
/*
+ * Task priorities.
+ * Note: if you change these, you must also change
+ * the task initialization definitions below.
+ */
+#define RPC_PRIORITY_LOW 0
+#define RPC_PRIORITY_NORMAL 1
+#define RPC_PRIORITY_HIGH 2
+#define RPC_NR_PRIORITY (RPC_PRIORITY_HIGH+1)
+
+/*
* RPC synchronization objects
*/
struct rpc_wait_queue {
- struct list_head tasks;
+ struct list_head tasks[RPC_NR_PRIORITY]; /* task queue for each priority level */
+ unsigned long cookie; /* cookie of last task serviced */
+ unsigned char maxpriority; /* maximum priority (0 if queue is not a priority queue) */
+ unsigned char priority; /* current priority */
+ unsigned char count; /* # task groups remaining serviced so far */
+ unsigned char nr; /* # tasks remaining for cookie */
#ifdef RPC_DEBUG
- char * name;
+ const char * name;
#endif
};
+/*
+ * This is the # requests to send consecutively
+ * from a single cookie. The aim is to improve
+ * performance of NFS operations such as read/write.
+ */
+#define RPC_BATCH_COUNT 16
+
#ifndef RPC_DEBUG
-# define RPC_WAITQ_INIT(var,qname) ((struct rpc_wait_queue) {LIST_HEAD_INIT(var)})
-# define RPC_WAITQ(var,qname) struct rpc_wait_queue var = RPC_WAITQ_INIT(var.tasks,qname)
-# define INIT_RPC_WAITQ(ptr,qname) do { \
- INIT_LIST_HEAD(&(ptr)->tasks); \
- } while(0)
+# define RPC_WAITQ_INIT(var,qname) { \
+ .tasks = { \
+ [0] = LIST_HEAD_INIT(var.tasks[0]), \
+ [1] = LIST_HEAD_INIT(var.tasks[1]), \
+ [2] = LIST_HEAD_INIT(var.tasks[2]), \
+ }, \
+ }
#else
-# define RPC_WAITQ_INIT(var,qname) ((struct rpc_wait_queue) {LIST_HEAD_INIT(var.tasks), qname})
-# define RPC_WAITQ(var,qname) struct rpc_wait_queue var = RPC_WAITQ_INIT(var,qname)
-# define INIT_RPC_WAITQ(ptr,qname) do { \
- INIT_LIST_HEAD(&(ptr)->tasks); (ptr)->name = qname; \
- } while(0)
+# define RPC_WAITQ_INIT(var,qname) { \
+ .tasks = { \
+ [0] = LIST_HEAD_INIT(var.tasks[0]), \
+ [1] = LIST_HEAD_INIT(var.tasks[1]), \
+ [2] = LIST_HEAD_INIT(var.tasks[2]), \
+ }, \
+ .name = qname, \
+ }
#endif
+# define RPC_WAITQ(var,qname) struct rpc_wait_queue var = RPC_WAITQ_INIT(var,qname)
+
+#define RPC_IS_PRIORITY(q) ((q)->maxpriority > 0)
/*
* Function prototypes
@@ -175,6 +209,8 @@ void rpc_run_child(struct rpc_task *par
rpc_action action);
int rpc_add_wait_queue(struct rpc_wait_queue *, struct rpc_task *);
void rpc_remove_wait_queue(struct rpc_task *);
+void rpc_init_priority_wait_queue(struct rpc_wait_queue *, const char *);
+void rpc_init_wait_queue(struct rpc_wait_queue *, const char *);
void rpc_sleep_on(struct rpc_wait_queue *, struct rpc_task *,
rpc_action action, rpc_action timer);
void rpc_add_timer(struct rpc_task *, rpc_action);
@@ -194,16 +230,14 @@ void rpc_show_tasks(void);
int rpc_init_mempool(void);
void rpc_destroy_mempool(void);
-static __inline__ void
-rpc_exit(struct rpc_task *task, int status)
+static inline void rpc_exit(struct rpc_task *task, int status)
{
task->tk_status = status;
task->tk_action = NULL;
}
#ifdef RPC_DEBUG
-static __inline__ char *
-rpc_qname(struct rpc_wait_queue *q)
+static inline const char * rpc_qname(struct rpc_wait_queue *q)
{
return ((q && q->name) ? q->name : "unknown");
}
--- diff/include/linux/swap.h 2004-02-09 10:36:12.000000000 +0000
+++ source/include/linux/swap.h 2004-03-16 11:27:48.621796024 +0000
@@ -232,6 +232,8 @@ extern sector_t map_swap_page(struct swa
extern struct swap_info_struct *get_swap_info_struct(unsigned);
extern int can_share_swap_page(struct page *);
extern int remove_exclusive_swap_page(struct page *);
+struct backing_dev_info;
+extern void swap_unplug_io_fn(struct backing_dev_info *);
extern struct swap_list_t swap_list;
extern spinlock_t swaplock;
--- diff/include/linux/syscalls.h 2004-03-11 10:20:29.000000000 +0000
+++ source/include/linux/syscalls.h 2004-03-16 11:27:48.622795872 +0000
@@ -48,6 +48,8 @@ struct timex;
struct timezone;
struct tms;
struct utimbuf;
+typedef int mqd_t;
+struct mq_attr;
#include <linux/config.h>
#include <linux/types.h>
@@ -252,7 +254,10 @@ asmlinkage long sys_mprotect(unsigned lo
asmlinkage unsigned long sys_mremap(unsigned long addr,
unsigned long old_len, unsigned long new_len,
unsigned long flags, unsigned long new_addr);
-long sys_remap_file_pages(unsigned long start, unsigned long size,
+asmlinkage long old_remap_file_pages(unsigned long start, unsigned long size,
+ unsigned long __prot, unsigned long pgoff,
+ unsigned long flags);
+asmlinkage long sys_remap_file_pages(unsigned long start, unsigned long size,
unsigned long prot, unsigned long pgoff,
unsigned long flags);
asmlinkage long sys_msync(unsigned long start, size_t len, int flags);
@@ -450,6 +455,13 @@ asmlinkage long sys_shmget(key_t key, si
asmlinkage long sys_shmdt(char __user *shmaddr);
asmlinkage long sys_shmctl(int shmid, int cmd, struct shmid_ds __user *buf);
+asmlinkage long sys_mq_open(const char __user *name, int oflag, mode_t mode, struct mq_attr __user *attr);
+asmlinkage long sys_mq_unlink(const char __user *name);
+asmlinkage long sys_mq_timedsend(mqd_t mqdes, const char __user *msg_ptr, size_t msg_len, unsigned int msg_prio, const struct timespec __user *abs_timeout);
+asmlinkage ssize_t sys_mq_timedreceive(mqd_t mqdes, char __user *msg_ptr, size_t msg_len, unsigned int __user *msg_prio, const struct timespec __user *abs_timeout);
+asmlinkage long sys_mq_notify(mqd_t mqdes, const struct sigevent __user *notification);
+asmlinkage long sys_mq_getsetattr(mqd_t mqdes, const struct mq_attr __user *mqstat, struct mq_attr __user *omqstat);
+
asmlinkage long sys_pciconfig_iobase(long which, unsigned long bus, unsigned long devfn);
asmlinkage long sys_pciconfig_read(unsigned long bus, unsigned long dfn,
unsigned long off, unsigned long len,
--- diff/include/linux/sysctl.h 2004-03-11 10:20:29.000000000 +0000
+++ source/include/linux/sysctl.h 2004-03-16 11:27:48.623795720 +0000
@@ -158,6 +158,8 @@ enum
VM_SWAPPINESS=19, /* Tendency to steal mapped memory */
VM_LOWER_ZONE_PROTECTION=20,/* Amount of protection of lower zones */
VM_MIN_FREE_KBYTES=21, /* Minimum free kilobytes to maintain */
+ VM_LAPTOP_MODE=22, /* vm laptop mode */
+ VM_BLOCK_DUMP=23, /* block dump mode */
};
--- diff/include/linux/sysfs.h 2003-08-26 09:00:54.000000000 +0000
+++ source/include/linux/sysfs.h 2004-03-16 11:27:48.623795720 +0000
@@ -18,6 +18,12 @@ struct attribute {
mode_t mode;
};
+struct attribute_group {
+ char * name;
+ struct attribute ** attrs;
+};
+
+
struct bin_attribute {
struct attribute attr;
size_t size;
@@ -25,14 +31,13 @@ struct bin_attribute {
ssize_t (*write)(struct kobject *, char *, loff_t, size_t);
};
-int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr);
-int sysfs_remove_bin_file(struct kobject * kobj, struct bin_attribute * attr);
-
struct sysfs_ops {
ssize_t (*show)(struct kobject *, struct attribute *,char *);
ssize_t (*store)(struct kobject *,struct attribute *,const char *, size_t);
};
+#ifdef CONFIG_SYSFS
+
extern int
sysfs_create_dir(struct kobject *);
@@ -57,13 +62,75 @@ sysfs_create_link(struct kobject * kobj,
extern void
sysfs_remove_link(struct kobject *, char * name);
-
-struct attribute_group {
- char * name;
- struct attribute ** attrs;
-};
+int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr);
+int sysfs_remove_bin_file(struct kobject * kobj, struct bin_attribute * attr);
int sysfs_create_group(struct kobject *, const struct attribute_group *);
void sysfs_remove_group(struct kobject *, const struct attribute_group *);
+#else /* CONFIG_SYSFS */
+
+static inline int sysfs_create_dir(struct kobject * k)
+{
+ return 0;
+}
+
+static inline void sysfs_remove_dir(struct kobject * k)
+{
+ ;
+}
+
+static inline void sysfs_rename_dir(struct kobject * k, const char *new_name)
+{
+ ;
+}
+
+static inline int sysfs_create_file(struct kobject * k, const struct attribute * a)
+{
+ return 0;
+}
+
+static inline int sysfs_update_file(struct kobject * k, const struct attribute * a)
+{
+ return 0;
+}
+
+static inline void sysfs_remove_file(struct kobject * k, const struct attribute * a)
+{
+ ;
+}
+
+static inline int sysfs_create_link(struct kobject * k, struct kobject * t, char * n)
+{
+ return 0;
+}
+
+static inline void sysfs_remove_link(struct kobject * k, char * name)
+{
+ ;
+}
+
+
+static inline int sysfs_create_bin_file(struct kobject * k, struct bin_attribute * a)
+{
+ return 0;
+}
+
+static inline int sysfs_remove_bin_file(struct kobject * k, struct bin_attribute * a)
+{
+ return 0;
+}
+
+static inline int sysfs_create_group(struct kobject * k, const struct attribute_group *g)
+{
+ return 0;
+}
+
+static inline void sysfs_remove_group(struct kobject * k, const struct attribute_group * g)
+{
+ ;
+}
+
+#endif /* CONFIG_SYSFS */
+
#endif /* _SYSFS_H_ */
--- diff/include/linux/usb.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/linux/usb.h 2004-03-16 11:27:48.626795264 +0000
@@ -72,14 +72,11 @@ struct usb_host_interface {
/**
* struct usb_interface - what usb device drivers talk to
- * @altsetting: array of interface descriptors, one for each alternate
+ * @altsetting: array of interface structures, one for each alternate
* setting that may be selected. Each one includes a set of
- * endpoint configurations and will be in numberic order,
- * 0..num_altsetting.
+ * endpoint configurations. They will be in no particular order.
* @num_altsetting: number of altsettings defined.
- * @act_altsetting: index of current altsetting. this number is always
- * less than num_altsetting. after the device is configured, each
- * interface uses its default setting of zero.
+ * @cur_altsetting: the current altsetting.
* @driver: the USB driver that is bound to this interface.
* @minor: the minor number assigned to this interface, if this
* interface is bound to a driver that uses the USB major number.
@@ -89,6 +86,8 @@ struct usb_host_interface {
* number from the USB core by calling usb_register_dev().
* @dev: driver model's view of this device
* @class_dev: driver model's class view of this device.
+ * @released: wait for the interface to be released when changing
+ * configurations.
*
* USB device drivers attach to interfaces on a physical device. Each
* interface encapsulates a single high level function, such as feeding
@@ -102,26 +101,33 @@ struct usb_host_interface {
* calls such as dev_get_drvdata() on the dev member of this structure.
*
* Each interface may have alternate settings. The initial configuration
- * of a device sets the first of these, but the device driver can change
+ * of a device sets altsetting 0, but the device driver can change
* that setting using usb_set_interface(). Alternate settings are often
* used to control the the use of periodic endpoints, such as by having
* different endpoints use different amounts of reserved USB bandwidth.
* All standards-conformant USB devices that use isochronous endpoints
* will use them in non-default settings.
+ *
+ * The USB specification says that alternate setting numbers must run from
+ * 0 to one less than the total number of alternate settings. But some
+ * devices manage to mess this up, and the structures aren't necessarily
+ * stored in numerical order anyhow. Use usb_altnum_to_altsetting() to
+ * look up an alternate setting in the altsetting array based on its number.
*/
struct usb_interface {
- /* array of alternate settings for this interface.
- * these will be in numeric order, 0..num_altsettting
- */
+ /* array of alternate settings for this interface,
+ * stored in no particular order */
struct usb_host_interface *altsetting;
- unsigned act_altsetting; /* active alternate setting */
+ struct usb_host_interface *cur_altsetting; /* the currently
+ * active alternate setting */
unsigned num_altsetting; /* number of alternate settings */
struct usb_driver *driver; /* driver */
int minor; /* minor number this interface is bound to */
struct device dev; /* interface specific device info */
struct class_device *class_dev;
+ struct completion *released; /* wait for release */
};
#define to_usb_interface(d) container_of(d, struct usb_interface, dev)
#define interface_to_usbdev(intf) \
@@ -140,19 +146,43 @@ static inline void usb_set_intfdata (str
/* this maximum is arbitrary */
#define USB_MAXINTERFACES 32
-/* USB_DT_CONFIG: Configuration descriptor information.
- *
- * USB_DT_OTHER_SPEED_CONFIG is the same descriptor, except that the
- * descriptor type is different. Highspeed-capable devices can look
- * different depending on what speed they're currently running. Only
- * devices with a USB_DT_DEVICE_QUALIFIER have an OTHER_SPEED_CONFIG.
+/**
+ * struct usb_host_config - representation of a device's configuration
+ * @desc: the device's configuration descriptor.
+ * @interface: array of usb_interface structures, one for each interface
+ * in the configuration. The number of interfaces is stored in
+ * desc.bNumInterfaces.
+ * @extra: pointer to buffer containing all extra descriptors associated
+ * with this configuration (those preceding the first interface
+ * descriptor).
+ * @extralen: length of the extra descriptors buffer.
+ *
+ * USB devices may have multiple configurations, but only one can be active
+ * at any time. Each encapsulates a different operational environment;
+ * for example, a dual-speed device would have separate configurations for
+ * full-speed and high-speed operation. The number of configurations
+ * available is stored in the device descriptor as bNumConfigurations.
+ *
+ * A configuration can contain multiple interfaces. Each corresponds to
+ * a different function of the USB device, and all are available whenever
+ * the configuration is active. The USB standard says that interfaces
+ * are supposed to be numbered from 0 to desc.bNumInterfaces-1, but a lot
+ * of devices get this wrong. In addition, the interface array is not
+ * guaranteed to be sorted in numerical order. Use usb_ifnum_to_if() to
+ * look up an interface entry based on its number.
+ *
+ * Device drivers should not attempt to activate configurations. The choice
+ * of which configuration to install is a policy decision based on such
+ * considerations as available power, functionality provided, and the user's
+ * desires (expressed through hotplug scripts). However, drivers can call
+ * usb_reset_configuration() to reinitialize the current configuration and
+ * all its interfaces.
*/
struct usb_host_config {
struct usb_config_descriptor desc;
- /* the interfaces associated with this configuration
- * these will be in numeric order, 0..desc.bNumInterfaces
- */
+ /* the interfaces associated with this configuration,
+ * stored in no particular order */
struct usb_interface *interface[USB_MAXINTERFACES];
unsigned char *extra; /* Extra descriptors */
@@ -294,8 +324,12 @@ extern void usb_driver_release_interface
const struct usb_device_id *usb_match_id(struct usb_interface *interface,
const struct usb_device_id *id);
-extern struct usb_interface *usb_find_interface(struct usb_driver *drv, int minor);
-extern struct usb_interface *usb_ifnum_to_if(struct usb_device *dev, unsigned ifnum);
+extern struct usb_interface *usb_find_interface(struct usb_driver *drv,
+ int minor);
+extern struct usb_interface *usb_ifnum_to_if(struct usb_device *dev,
+ unsigned ifnum);
+extern struct usb_host_interface *usb_altnum_to_altsetting(
+ struct usb_interface *intf, unsigned int altnum);
/**
@@ -461,8 +495,8 @@ extern struct bus_type usb_bus_type;
* @minor_base: the start of the minor range for this driver.
*
* This structure is used for the usb_register_dev() and
- * usb_unregister_dev() functions, to consolodate a number of the
- * paramaters used for them.
+ * usb_unregister_dev() functions, to consolidate a number of the
+ * parameters used for them.
*/
struct usb_class_driver {
char *name;
@@ -520,7 +554,7 @@ typedef void (*usb_complete_t)(struct ur
* @urb_list: For use by current owner of the URB.
* @pipe: Holds endpoint number, direction, type, and more.
* Create these values with the eight macros available;
- * usb_{snd,rcv}TYPEpipe(dev,endpoint), where the type is "ctrl"
+ * usb_{snd,rcv}TYPEpipe(dev,endpoint), where the TYPE is "ctrl"
* (control), "bulk", "int" (interrupt), or "iso" (isochronous).
* For example usb_sndbulkpipe() or usb_rcvintpipe(). Endpoint
* numbers range from zero to fifteen. Note that "in" endpoint two
@@ -539,8 +573,8 @@ typedef void (*usb_complete_t)(struct ur
* the I/O request will be performed (unless URB_NO_TRANSFER_DMA_MAP
* is set). This buffer must be suitable for DMA; allocate it with
* kmalloc() or equivalent. For transfers to "in" endpoints, contents
- * of this buffer will be modified. This buffer is used for data
- * phases of control transfers.
+ * of this buffer will be modified. This buffer is used for the data
+ * stage of control transfers.
* @transfer_dma: When transfer_flags includes URB_NO_TRANSFER_DMA_MAP,
* the device driver is saying that it provided this DMA address,
* which the host controller driver should use in preference to the
@@ -563,8 +597,7 @@ typedef void (*usb_complete_t)(struct ur
* device driver has provided this DMA address for the setup packet.
* The host controller driver should use this in preference to
* setup_packet.
- * @start_frame: Returns the initial frame for interrupt or isochronous
- * transfers.
+ * @start_frame: Returns the initial frame for isochronous transfers.
* @number_of_packets: Lists the number of ISO transfer buffers.
* @interval: Specifies the polling interval for interrupt or isochronous
* transfers. The units are frames (milliseconds) for for full and low
@@ -632,13 +665,14 @@ typedef void (*usb_complete_t)(struct ur
* Interrupt UBS must provide an interval, saying how often (in milliseconds
* or, for highspeed devices, 125 microsecond units)
* to poll for transfers. After the URB has been submitted, the interval
- * and start_frame fields reflect how the transfer was actually scheduled.
+ * field reflects how the transfer was actually scheduled.
* The polling interval may be more frequent than requested.
* For example, some controllers have a maximum interval of 32 microseconds,
* while others support intervals of up to 1024 microseconds.
* Isochronous URBs also have transfer intervals. (Note that for isochronous
* endpoints, as well as high speed interrupt endpoints, the encoding of
- * the transfer interval in the endpoint descriptor is logarithmic.)
+ * the transfer interval in the endpoint descriptor is logarithmic.
+ * Device drivers must convert that value to linear units themselves.)
*
* Isochronous URBs normally use the URB_ISO_ASAP transfer flag, telling
* the host controller to schedule the transfer as soon as bandwidth
@@ -671,8 +705,9 @@ typedef void (*usb_complete_t)(struct ur
* The context field is normally used to link URBs back to the relevant
* driver or request state.
*
- * When completion callback is invoked for non-isochronous URBs, the
- * actual_length field tells how many bytes were transferred.
+ * When the completion callback is invoked for non-isochronous URBs, the
+ * actual_length field tells how many bytes were transferred. This field
+ * is updated even when the URB terminated with an error or was unlinked.
*
* ISO transfer status is reported in the status and actual_length fields
* of the iso_frame_desc array, and the number of errors is reported in
@@ -699,9 +734,9 @@ struct urb
int actual_length; /* (return) actual transfer length */
unsigned char *setup_packet; /* (in) setup packet (control only) */
dma_addr_t setup_dma; /* (in) dma addr for setup_packet */
- int start_frame; /* (modify) start frame (INT/ISO) */
+ int start_frame; /* (modify) start frame (ISO) */
int number_of_packets; /* (in) number of ISO packets */
- int interval; /* (in) transfer interval (INT/ISO) */
+ int interval; /* (modify) transfer interval (INT/ISO) */
int error_count; /* (return) number of ISO errors */
int timeout; /* (in) timeout, in jiffies */
void *context; /* (in) context for completion */
--- diff/include/linux/usb_gadget.h 2004-03-16 11:24:02.000000000 +0000
+++ source/include/linux/usb_gadget.h 2004-03-16 11:27:48.624795568 +0000
@@ -465,6 +465,8 @@ struct usb_gadget_ops {
* driver setup() requests
* @ep_list: List of other endpoints supported by the device.
* @speed: Speed of current connection to USB host.
+ * @is_dualspeed: True if the controller supports both high and full speed
+ * operation. If it does, the gadget driver must also support both.
* @name: Identifies the controller hardware type. Used in diagnostics
* and sometimes configuration.
* @dev: Driver model state for this abstract device.
@@ -488,6 +490,7 @@ struct usb_gadget {
struct usb_ep *ep0;
struct list_head ep_list; /* of usb_ep */
enum usb_device_speed speed;
+ unsigned is_dualspeed:1;
const char *name;
struct device dev;
};
@@ -690,7 +693,7 @@ int usb_gadget_unregister_driver (struct
/**
* struct usb_string - wraps a C string and its USB id
* @id:the (nonzero) ID for this string
- * @s:the string, in ISO-8859/1 characters
+ * @s:the string, in UTF-8 encoding
*
* If you're using usb_gadget_get_string(), use this to wrap a string
* together with its ID.
@@ -716,6 +719,17 @@ struct usb_gadget_strings {
/* put descriptor for string with that id into buf (buflen >= 256) */
int usb_gadget_get_string (struct usb_gadget_strings *table, int id, u8 *buf);
+/*-------------------------------------------------------------------------*/
+
+/* utility to simplify managing config descriptors */
+
+/* write vector of descriptors into buffer */
+int usb_descriptor_fillbuf(void *, unsigned,
+ const struct usb_descriptor_header **);
+
+/* build config descriptor from single descriptor vector */
+int usb_gadget_config_buf(const struct usb_config_descriptor *config,
+ void *buf, unsigned buflen, const struct usb_descriptor_header **desc);
#endif /* __KERNEL__ */
--- diff/include/linux/workqueue.h 2004-03-11 10:20:29.000000000 +0000
+++ source/include/linux/workqueue.h 2004-03-16 11:27:48.626795264 +0000
@@ -52,7 +52,9 @@ struct work_struct {
extern struct workqueue_struct *create_workqueue(const char *name);
extern void destroy_workqueue(struct workqueue_struct *wq);
-extern int FASTCALL(queue_work(struct workqueue_struct *wq, struct work_struct *work));
+int queue_work(struct workqueue_struct *wq, struct work_struct *work);
+int queue_work_on_cpu(struct workqueue_struct *wq,
+ struct work_struct *work, int cpu);
extern int FASTCALL(queue_delayed_work(struct workqueue_struct *wq, struct work_struct *work, unsigned long delay));
extern void FASTCALL(flush_workqueue(struct workqueue_struct *wq));
--- diff/include/linux/writeback.h 2003-10-09 08:47:17.000000000 +0000
+++ source/include/linux/writeback.h 2004-03-16 11:27:48.627795112 +0000
@@ -71,12 +71,15 @@ static inline void wait_on_inode(struct
* mm/page-writeback.c
*/
int wakeup_bdflush(long nr_pages);
+void disk_is_spun_up(int postpone_writeback);
-/* These 5 are exported to sysctl. */
+/* These are exported to sysctl. */
extern int dirty_background_ratio;
extern int vm_dirty_ratio;
extern int dirty_writeback_centisecs;
extern int dirty_expire_centisecs;
+extern int block_dump;
+extern int laptop_mode;
struct ctl_table;
struct file;
--- diff/init/Kconfig 2004-03-11 10:20:29.000000000 +0000
+++ source/init/Kconfig 2004-03-16 11:27:48.629794808 +0000
@@ -43,7 +43,7 @@ config CLEAN_COMPILE
config STANDALONE
bool "Select only drivers that don't need compile-time external firmware" if EXPERIMENTAL
- default y
+ default n
help
Select this option if you don't have magic firmware for drivers that
need it.
@@ -91,6 +91,24 @@ config SYSVIPC
section 6.4 of the Linux Programmer's Guide, available from
<http://www.tldp.org/docs.html#guide>.
+config POSIX_MQUEUE
+ bool "POSIX Message Queues"
+ depends on EXPERIMENTAL
+ ---help---
+ POSIX variant of message queues is a part of IPC. In POSIX message
+ queues every message has a priority which decides about succession
+ of receiving it by a process. If you want to compile and run
+ programs written e.g. for Solaris with use of its POSIX message
+ queues (functions mq_*) say Y here. To use this feature you will
+ also need mqueue library, available from
+ <http://www.mat.uni.torun.pl/~wrona/posix_ipc/>
+
+ POSIX message queues are visible as a filesystem called 'mqueue'
+ and can be mounted somewhere if you want to do filesystem
+ operations on message queues.
+
+ If unsure, say Y.
+
config BSD_PROCESS_ACCT
bool "BSD Process Accounting"
help
@@ -120,6 +138,26 @@ config SYSCTL
building a kernel for install/rescue disks or your system is very
limited in memory.
+config AUDIT
+ bool "Auditing support"
+ default y if SECURITY_SELINUX
+ default n
+ help
+ Enable auditing infrastructure that can be used with another
+ kernel subsystem, such as SELinux (which requires this for
+ logging of avc messages output). Does not do system-call
+ auditing without CONFIG_AUDITSYSCALL.
+
+config AUDITSYSCALL
+ bool "Enable system-call auditing support"
+ depends on AUDIT && (X86 || PPC64)
+ default y if SECURITY_SELINUX
+ default n
+ help
+ Enable low-overhead system-call auditing infrastructure that
+ can be used independently or with another kernel subsystem,
+ such as SELinux.
+
config LOG_BUF_SHIFT
int "Kernel log buffer size (16 => 64KB, 17 => 128KB)" if DEBUG_KERNEL
range 12 20
--- diff/init/do_mounts.c 2004-03-16 11:24:02.000000000 +0000
+++ source/init/do_mounts.c 2004-03-16 11:27:48.628794960 +0000
@@ -141,9 +141,11 @@ dev_t __init name_to_dev_t(char *name)
dev_t res = 0;
int part;
+#ifdef CONFIG_SYSFS
sys_mkdir("/sys", 0700);
if (sys_mount("sysfs", "/sys", "sysfs", 0, NULL) < 0)
goto out;
+#endif
if (strncmp(name, "/dev/", 5) != 0) {
unsigned maj, min;
--- diff/init/main.c 2004-03-16 11:24:02.000000000 +0000
+++ source/init/main.c 2004-03-16 11:27:48.629794808 +0000
@@ -156,8 +156,11 @@ static int __init obsolete_checksetup(ch
p = &__setup_start;
do {
int n = strlen(p->str);
- if (!strncmp(line,p->str,n)) {
- if (p->setup_func(line+n))
+ if (!strncmp(line, p->str, n)) {
+ if (!p->setup_func) {
+ printk(KERN_WARNING "Parameter %s is obsolete, ignored\n", p->str);
+ return 1;
+ } else if (p->setup_func(line + n))
return 1;
}
p++;
@@ -418,12 +421,12 @@ asmlinkage void __init start_kernel(void
build_all_zonelists();
page_alloc_init();
+ trap_init();
printk("Kernel command line: %s\n", saved_command_line);
parse_args("Booting kernel", command_line, __start___param,
__stop___param - __start___param,
&unknown_bootoption);
sort_main_extable();
- trap_init();
rcu_init();
init_IRQ();
pidhash_init();
@@ -568,7 +571,6 @@ static void do_pre_smp_initcalls(void)
migration_init();
#endif
- node_nr_running_init();
spawn_ksoftirqd();
}
@@ -599,8 +601,16 @@ static int init(void * unused)
do_pre_smp_initcalls();
smp_init();
+ sched_init_smp();
do_basic_setup();
+ /*
+ * check if there is an early userspace init, if yes
+ * let it do all the work
+ */
+ if (sys_access("/init", 0) == 0)
+ execute_command = "/init";
+ else
prepare_namespace();
/*
--- diff/ipc/Makefile 2004-03-16 11:24:02.000000000 +0000
+++ source/ipc/Makefile 2004-03-16 11:27:48.630794656 +0000
@@ -2,7 +2,7 @@
# Makefile for the linux ipc.
#
-obj-y := util.o
-
obj-$(CONFIG_SYSVIPC_COMPAT) += compat.o
-obj-$(CONFIG_SYSVIPC) += msg.o sem.o shm.o
+obj-$(CONFIG_SYSVIPC) += util.o msgutil.o msg.o sem.o shm.o
+obj-$(CONFIG_POSIX_MQUEUE) += mqueue.o msgutil.o
+
--- diff/ipc/msg.c 2004-02-09 10:36:12.000000000 +0000
+++ source/ipc/msg.c 2004-03-16 11:27:48.635793896 +0000
@@ -51,11 +51,6 @@ struct msg_sender {
struct task_struct* tsk;
};
-struct msg_msgseg {
- struct msg_msgseg* next;
- /* the next part of the message follows immediately */
-};
-
#define SEARCH_ANY 1
#define SEARCH_EQUAL 2
#define SEARCH_NOTEQUAL 3
@@ -129,106 +124,6 @@ static int newque (key_t key, int msgflg
return msg_buildid(id,msq->q_perm.seq);
}
-static void free_msg(struct msg_msg* msg)
-{
- struct msg_msgseg* seg;
-
- security_msg_msg_free(msg);
-
- seg = msg->next;
- kfree(msg);
- while(seg != NULL) {
- struct msg_msgseg* tmp = seg->next;
- kfree(seg);
- seg = tmp;
- }
-}
-
-static struct msg_msg* load_msg(void* src, int len)
-{
- struct msg_msg* msg;
- struct msg_msgseg** pseg;
- int err;
- int alen;
-
- alen = len;
- if(alen > DATALEN_MSG)
- alen = DATALEN_MSG;
-
- msg = (struct msg_msg *) kmalloc (sizeof(*msg) + alen, GFP_KERNEL);
- if(msg==NULL)
- return ERR_PTR(-ENOMEM);
-
- msg->next = NULL;
- msg->security = NULL;
-
- if (copy_from_user(msg+1, src, alen)) {
- err = -EFAULT;
- goto out_err;
- }
-
- len -= alen;
- src = ((char*)src)+alen;
- pseg = &msg->next;
- while(len > 0) {
- struct msg_msgseg* seg;
- alen = len;
- if(alen > DATALEN_SEG)
- alen = DATALEN_SEG;
- seg = (struct msg_msgseg *) kmalloc (sizeof(*seg) + alen, GFP_KERNEL);
- if(seg==NULL) {
- err=-ENOMEM;
- goto out_err;
- }
- *pseg = seg;
- seg->next = NULL;
- if(copy_from_user (seg+1, src, alen)) {
- err = -EFAULT;
- goto out_err;
- }
- pseg = &seg->next;
- len -= alen;
- src = ((char*)src)+alen;
- }
-
- err = security_msg_msg_alloc(msg);
- if (err)
- goto out_err;
-
- return msg;
-
-out_err:
- free_msg(msg);
- return ERR_PTR(err);
-}
-
-static int store_msg(void* dest, struct msg_msg* msg, int len)
-{
- int alen;
- struct msg_msgseg *seg;
-
- alen = len;
- if(alen > DATALEN_MSG)
- alen = DATALEN_MSG;
- if(copy_to_user (dest, msg+1, alen))
- return -1;
-
- len -= alen;
- dest = ((char*)dest)+alen;
- seg = msg->next;
- while(len > 0) {
- alen = len;
- if(alen > DATALEN_SEG)
- alen = DATALEN_SEG;
- if(copy_to_user (dest, seg+1, alen))
- return -1;
- len -= alen;
- dest = ((char*)dest)+alen;
- seg=seg->next;
- }
- return 0;
-}
-
static inline void ss_add(struct msg_queue* msq, struct msg_sender* mss)
{
mss->tsk=current;
--- diff/ipc/shm.c 2004-03-11 10:20:29.000000000 +0000
+++ source/ipc/shm.c 2004-03-16 11:27:48.636793744 +0000
@@ -380,9 +380,7 @@ static void shm_get_stat(unsigned long *
if (is_file_hugepages(shp->shm_file)) {
struct address_space *mapping = inode->i_mapping;
- spin_lock(&mapping->page_lock);
*rss += (HPAGE_SIZE/PAGE_SIZE)*mapping->nrpages;
- spin_unlock(&mapping->page_lock);
} else {
struct shmem_inode_info *info = SHMEM_I(inode);
spin_lock(&info->lock);
@@ -760,6 +758,21 @@ asmlinkage long sys_shmdt(char __user *s
down_write(&mm->mmap_sem);
/*
+ * This function tries to be smart and unmap shm segments that
+ * were modified by partial mlock or munmap calls:
+ * - It first determines the size of the shm segment that should be
+ * unmapped: It searches for a vma that is backed by shm and that
+ * started at address shmaddr. It records it's size and then unmaps
+ * it.
+ * - Then it unmaps all shm vmas that started at shmaddr and that
+ * are within the initially determined size.
+ * Errors from do_munmap are ignored: the function only fails if
+ * it's called with invalid parameters or if it's called to unmap
+ * a part of a vma. Both calls in this function are for full vmas,
+ * the parameters are directly copied from the vma itself and always
+ * valid - therefore do_munmap cannot fail. (famous last words?)
+ */
+ /*
* If it had been mremap()'d, the starting address would not
* match the usual checks anyway. So assume all vma's are
* above the starting address given.
@@ -776,10 +789,12 @@ asmlinkage long sys_shmdt(char __user *s
*/
if ((vma->vm_ops == &shm_vm_ops || is_vm_hugetlb_page(vma)) &&
(vma->vm_start - addr)/PAGE_SIZE == vma->vm_pgoff) {
-
+ int ret;
size = vma->vm_file->f_dentry->d_inode->i_size;
- do_munmap(mm, vma->vm_start, vma->vm_end - vma->vm_start);
+ ret = do_munmap(mm, vma->vm_start,
+ vma->vm_end - vma->vm_start);
+ WARN_ON(ret);
/*
* We discovered the size of the shm segment, so
* break out of here and fall through to the next
@@ -803,9 +818,13 @@ asmlinkage long sys_shmdt(char __user *s
/* finding a matching vma now does not alter retval */
if ((vma->vm_ops == &shm_vm_ops || is_vm_hugetlb_page(vma)) &&
- (vma->vm_start - addr)/PAGE_SIZE == vma->vm_pgoff)
+ (vma->vm_start - addr)/PAGE_SIZE == vma->vm_pgoff) {
+ int ret;
- do_munmap(mm, vma->vm_start, vma->vm_end - vma->vm_start);
+ ret = do_munmap(mm, vma->vm_start,
+ vma->vm_end - vma->vm_start);
+ WARN_ON(ret);
+ }
vma = next;
}
--- diff/ipc/util.c 2004-03-11 10:20:29.000000000 +0000
+++ source/ipc/util.c 2004-03-16 11:27:48.637793592 +0000
@@ -25,8 +25,6 @@
#include <linux/rcupdate.h>
#include <linux/workqueue.h>
-#if defined(CONFIG_SYSVIPC)
-
#include "util.h"
/**
@@ -531,20 +529,3 @@ int ipc_parse_version (int *cmd)
}
#endif /* __ia64__ */
-
-#else
-/*
- * Dummy functions when SYSV IPC isn't configured
- */
-
-int copy_semundo(unsigned long clone_flags, struct task_struct *tsk)
-{
- return 0;
-}
-
-void exit_sem(struct task_struct *tsk)
-{
- return;
-}
-
-#endif /* CONFIG_SYSVIPC */
--- diff/ipc/util.h 2004-02-18 08:54:13.000000000 +0000
+++ source/ipc/util.h 2004-03-16 11:27:48.638793440 +0000
@@ -4,6 +4,10 @@
*
* ipc helper functions (c) 1999 Manfred Spraul <manfreds@colorfullife.com>
*/
+
+#ifndef _IPC_UTIL_H
+#define _IPC_UTIL_H
+
#define USHRT_MAX 0xffff
#define SEQ_MULTIPLIER (IPCMNI)
@@ -62,3 +66,9 @@ void ipc64_perm_to_ipc_perm(struct ipc64
#else
int ipc_parse_version (int *cmd);
#endif
+
+extern void free_msg(struct msg_msg *msg);
+extern struct msg_msg *load_msg(void __user *src, int len);
+extern int store_msg(void __user *dest, struct msg_msg *msg, int len);
+
+#endif
--- diff/kernel/Makefile 2004-03-11 10:20:29.000000000 +0000
+++ source/kernel/Makefile 2004-03-16 11:27:48.649791768 +0000
@@ -12,15 +12,18 @@ obj-y = sched.o fork.o exec_domain.o
obj-$(CONFIG_FUTEX) += futex.o
obj-$(CONFIG_GENERIC_ISA_DMA) += dma.o
obj-$(CONFIG_SMP) += cpu.o
+obj-$(CONFIG_LOCKMETER) += lockmeter.o
obj-$(CONFIG_UID16) += uid16.o
obj-$(CONFIG_MODULES) += module.o
obj-$(CONFIG_KALLSYMS) += kallsyms.o
obj-$(CONFIG_PM) += power/
obj-$(CONFIG_BSD_PROCESS_ACCT) += acct.o
-obj-$(CONFIG_COMPAT) += compat.o
+obj-$(CONFIG_COMPAT) += compat.o compat_signal.o
obj-$(CONFIG_IKCONFIG) += configs.o
obj-$(CONFIG_IKCONFIG_PROC) += configs.o
obj-$(CONFIG_STOP_MACHINE) += stop_machine.o
+obj-$(CONFIG_AUDIT) += audit.o
+obj-$(CONFIG_AUDITSYSCALL) += auditsc.o
ifneq ($(CONFIG_IA64),y)
# According to Alan Modra <alan@linuxcare.com.au>, the -fno-omit-frame-pointer is
--- diff/kernel/acct.c 2004-03-11 10:20:29.000000000 +0000
+++ source/kernel/acct.c 2004-03-16 11:27:48.638793440 +0000
@@ -346,7 +346,7 @@ static void do_acct_process(long exitcod
/* we really need to bite the bullet and change layout */
ac.ac_uid = current->uid;
ac.ac_gid = current->gid;
- ac.ac_tty = current->tty ? old_encode_dev(tty_devnum(current->tty)) : 0;
+ ac.ac_tty = current->signal->tty ? old_encode_dev(tty_devnum(current->signal->tty)) : 0;
ac.ac_flag = 0;
if (current->flags & PF_FORKNOEXEC)
--- diff/kernel/exit.c 2004-03-11 10:20:29.000000000 +0000
+++ source/kernel/exit.c 2004-03-16 11:27:48.644792528 +0000
@@ -136,13 +136,13 @@ int session_of_pgrp(int pgrp)
read_lock(&tasklist_lock);
for_each_task_pid(pgrp, PIDTYPE_PGID, p, l, pid)
- if (p->session > 0) {
- sid = p->session;
+ if (p->signal->session > 0) {
+ sid = p->signal->session;
goto out;
}
p = find_task_by_pid(pgrp);
if (p)
- sid = p->session;
+ sid = p->signal->session;
out:
read_unlock(&tasklist_lock);
@@ -170,7 +170,7 @@ static int will_become_orphaned_pgrp(int
|| p->real_parent->pid == 1)
continue;
if (process_group(p->real_parent) != pgrp
- && p->real_parent->session == p->session) {
+ && p->real_parent->signal->session == p->signal->session) {
ret = 0;
break;
}
@@ -259,14 +259,14 @@ void __set_special_pids(pid_t session, p
{
struct task_struct *curr = current;
- if (curr->session != session) {
+ if (curr->signal->session != session) {
detach_pid(curr, PIDTYPE_SID);
- curr->session = session;
+ curr->signal->session = session;
attach_pid(curr, PIDTYPE_SID, session);
}
if (process_group(curr) != pgrp) {
detach_pid(curr, PIDTYPE_PGID);
- curr->group_leader->__pgrp = pgrp;
+ curr->signal->pgrp = pgrp;
attach_pid(curr, PIDTYPE_PGID, pgrp);
}
}
@@ -341,7 +341,7 @@ void daemonize(const char *name, ...)
exit_mm(current);
set_special_pids(1, 1);
- current->tty = NULL;
+ current->signal->tty = NULL;
/* Block and flush all signals */
sigfillset(&blocked);
@@ -564,7 +564,7 @@ static inline void reparent_thread(task_
* outside, so the child pgrp is now orphaned.
*/
if ((process_group(p) != process_group(father)) &&
- (p->session == father->session)) {
+ (p->signal->session == father->signal->session)) {
int pgrp = process_group(p);
if (will_become_orphaned_pgrp(pgrp, NULL) && has_stopped_jobs(pgrp)) {
@@ -675,7 +675,7 @@ static void exit_notify(struct task_stru
t = tsk->real_parent;
if ((process_group(t) != process_group(tsk)) &&
- (t->session == tsk->session) &&
+ (t->signal->session == tsk->signal->session) &&
will_become_orphaned_pgrp(process_group(tsk), tsk) &&
has_stopped_jobs(process_group(tsk))) {
__kill_pg_info(SIGHUP, (void *)1, process_group(tsk));
@@ -780,7 +780,7 @@ asmlinkage NORET_TYPE void do_exit(long
exit_itimers(tsk);
exit_thread();
- if (tsk->leader)
+ if (tsk->signal->leader)
disassociate_ctty(1);
module_put(tsk->thread_info->exec_domain->module);
--- diff/kernel/fork.c 2004-03-11 10:20:29.000000000 +0000
+++ source/kernel/fork.c 2004-03-16 11:27:48.645792376 +0000
@@ -21,6 +21,7 @@
#include <linux/completion.h>
#include <linux/namespace.h>
#include <linux/personality.h>
+#include <linux/sem.h>
#include <linux/file.h>
#include <linux/binfmts.h>
#include <linux/mman.h>
@@ -31,6 +32,7 @@
#include <linux/futex.h>
#include <linux/ptrace.h>
#include <linux/mount.h>
+#include <linux/audit.h>
#include <asm/pgtable.h>
#include <asm/pgalloc.h>
@@ -39,9 +41,6 @@
#include <asm/cacheflush.h>
#include <asm/tlbflush.h>
-extern int copy_semundo(unsigned long clone_flags, struct task_struct *tsk);
-extern void exit_sem(struct task_struct *tsk);
-
/* The idle threads do not count..
* Protected by write_lock_irq(&tasklist_lock)
*/
@@ -85,6 +84,8 @@ void __put_task_struct(struct task_struc
WARN_ON(atomic_read(&tsk->usage));
WARN_ON(tsk == current);
+ if (unlikely(tsk->audit_context))
+ audit_free(tsk);
security_task_free(tsk);
free_uid(tsk->user);
put_group_info(tsk->group_info);
@@ -209,11 +210,14 @@ EXPORT_SYMBOL(autoremove_wake_function);
void __init fork_init(unsigned long mempages)
{
#ifndef __HAVE_ARCH_TASK_STRUCT_ALLOCATOR
+#ifndef ARCH_MIN_TASKALIGN
+#define ARCH_MIN_TASKALIGN 0
+#endif
/* create a slab on which task_structs can be allocated */
task_struct_cachep =
kmem_cache_create("task_struct",
- sizeof(struct task_struct),0,
- SLAB_MUST_HWCACHE_ALIGN, NULL, NULL);
+ sizeof(struct task_struct),ARCH_MIN_TASKALIGN,
+ 0, NULL, NULL);
if (!task_struct_cachep)
panic("fork_init(): cannot create task_struct SLAB cache");
#endif
@@ -276,7 +280,7 @@ static inline int dup_mmap(struct mm_str
mm->mmap_cache = NULL;
mm->free_area_cache = TASK_UNMAPPED_BASE;
mm->map_count = 0;
- mm->rss = 0;
+ zero_rss(mm);
cpus_clear(mm->cpu_vm_mask);
mm->mm_rb = RB_ROOT;
rb_link = &mm->mm_rb.rb_node;
@@ -813,6 +817,12 @@ static inline int copy_signal(unsigned l
sig->curr_target = NULL;
init_sigpending(&sig->shared_pending);
+ sig->tty = current->signal->tty;
+ sig->pgrp = process_group(current);
+ sig->session = current->signal->session;
+ sig->leader = 0; /* session leadership doesn't inherit */
+ sig->tty_old_pgrp = 0;
+
return 0;
}
@@ -937,21 +947,22 @@ struct task_struct *copy_process(unsigne
init_timer(&p->real_timer);
p->real_timer.data = (unsigned long) p;
- p->leader = 0; /* session leadership doesn't inherit */
- p->tty_old_pgrp = 0;
p->utime = p->stime = 0;
p->cutime = p->cstime = 0;
p->lock_depth = -1; /* -1 = no lock */
p->start_time = get_jiffies_64();
p->security = NULL;
p->io_context = NULL;
+ p->audit_context = NULL;
retval = -ENOMEM;
if ((retval = security_task_alloc(p)))
goto bad_fork_cleanup;
+ if ((retval = audit_alloc(p)))
+ goto bad_fork_cleanup_security;
/* copy all the process information */
if ((retval = copy_semundo(clone_flags, p)))
- goto bad_fork_cleanup_security;
+ goto bad_fork_cleanup_audit;
if ((retval = copy_files(clone_flags, p)))
goto bad_fork_cleanup_semundo;
if ((retval = copy_fs(clone_flags, p)))
@@ -1057,7 +1068,7 @@ struct task_struct *copy_process(unsigne
if (thread_group_leader(p)) {
attach_pid(p, PIDTYPE_TGID, p->tgid);
attach_pid(p, PIDTYPE_PGID, process_group(p));
- attach_pid(p, PIDTYPE_SID, p->session);
+ attach_pid(p, PIDTYPE_SID, p->signal->session);
if (p->pid)
__get_cpu_var(process_counts)++;
} else
@@ -1086,6 +1097,8 @@ bad_fork_cleanup_files:
exit_files(p); /* blocking */
bad_fork_cleanup_semundo:
exit_sem(p);
+bad_fork_cleanup_audit:
+ audit_free(p);
bad_fork_cleanup_security:
security_task_free(p);
bad_fork_cleanup:
--- diff/kernel/futex.c 2004-02-18 08:54:13.000000000 +0000
+++ source/kernel/futex.c 2004-03-16 11:27:48.646792224 +0000
@@ -269,7 +269,11 @@ static void wake_futex(struct futex_q *q
* The lock in wake_up_all() is a crucial memory barrier after the
* list_del_init() and also before assigning to q->lock_ptr.
*/
+
+ current->flags |= PF_FUTEX_DEBUG;
wake_up_all(&q->waiters);
+ current->flags &= ~PF_FUTEX_DEBUG;
+
/*
* The waiting task can free the futex_q as soon as this is written,
* without taking any locks. This must come last.
@@ -490,8 +494,11 @@ static int futex_wait(unsigned long uadd
* !list_empty() is safe here without any lock.
* q.lock_ptr != 0 is not safe, because of ordering against wakeup.
*/
- if (likely(!list_empty(&q.list)))
+ if (likely(!list_empty(&q.list))) {
+ current->flags |= PF_FUTEX_DEBUG;
time = schedule_timeout(time);
+ current->flags &= ~PF_FUTEX_DEBUG;
+ }
__set_current_state(TASK_RUNNING);
/*
@@ -505,7 +512,11 @@ static int futex_wait(unsigned long uadd
if (time == 0)
return -ETIMEDOUT;
/* A spurious wakeup should never happen. */
- WARN_ON(!signal_pending(current));
+ if (!signal_pending(current)) {
+ printk("futex_wait woken: %lu %i %lu\n",
+ uaddr, val, time);
+ WARN_ON(1);
+ }
return -EINTR;
out_unqueue:
--- diff/kernel/kthread.c 2004-03-16 11:24:02.000000000 +0000
+++ source/kernel/kthread.c 2004-03-16 11:27:48.646792224 +0000
@@ -10,6 +10,7 @@
#include <linux/completion.h>
#include <linux/err.h>
#include <linux/unistd.h>
+#include <linux/file.h>
#include <asm/semaphore.h>
struct kthread_create_info
@@ -41,6 +42,21 @@ int kthread_should_stop(void)
return (kthread_stop_info.k == current);
}
+
+static void kthread_exit_files(void)
+{
+ struct fs_struct *fs;
+ struct task_struct *tsk = current;
+
+ exit_fs(tsk); /* current->fs->count--; */
+ fs = init_task.fs;
+ tsk->fs = fs;
+ atomic_inc(&fs->count);
+ exit_files(tsk);
+ current->files = init_task.files;
+ atomic_inc(&tsk->files->count);
+}
+
static int kthread(void *_create)
{
struct kthread_create_info *create = _create;
@@ -50,6 +66,8 @@ static int kthread(void *_create)
int ret = -EINTR;
cpumask_t mask = CPU_MASK_ALL;
+ kthread_exit_files();
+
/* Copy data: it's on keventd's stack */
threadfn = create->threadfn;
data = create->data;
--- diff/kernel/module.c 2004-03-16 11:24:02.000000000 +0000
+++ source/kernel/module.c 2004-03-16 11:27:48.651791464 +0000
@@ -1243,7 +1243,15 @@ static void add_kallsyms(struct module *
mod->symtab[i].st_info
= elf_type(&mod->symtab[i], sechdrs, secstrings, mod);
}
-#endif
+#else
+static inline void add_kallsyms(struct module *mod,
+ Elf_Shdr *sechdrs,
+ unsigned int symindex,
+ unsigned int strindex,
+ const char *secstrings)
+{
+}
+#endif /* CONFIG_KALLSYMS */
/* Allocate and load the module: note that size of section 0 is always
zero, and we rely on this for optional sections. */
@@ -1516,14 +1524,12 @@ static struct module *load_module(void _
percpu_modcopy(mod->percpu, (void *)sechdrs[pcpuindex].sh_addr,
sechdrs[pcpuindex].sh_size);
+ add_kallsyms(mod, sechdrs, symindex, strindex, secstrings);
+
err = module_finalize(hdr, sechdrs, mod);
if (err < 0)
goto cleanup;
-#ifdef CONFIG_KALLSYMS
- add_kallsyms(mod, sechdrs, symindex, strindex, secstrings);
-#endif
-
mod->args = args;
if (obsparmindex) {
err = obsolete_params(mod->name, mod->args,
--- diff/kernel/pid.c 2004-03-11 10:20:29.000000000 +0000
+++ source/kernel/pid.c 2004-03-16 11:27:48.652791312 +0000
@@ -122,6 +122,8 @@ return_pid:
}
if (!offset || !atomic_read(&map->nr_free)) {
+ if (!offset)
+ map--;
next_map:
map = next_free_map(map, &max_steps);
if (!map)
@@ -253,14 +255,14 @@ void switch_exec_pids(task_t *leader, ta
attach_pid(thread, PIDTYPE_PID, thread->pid);
attach_pid(thread, PIDTYPE_TGID, thread->tgid);
- attach_pid(thread, PIDTYPE_PGID, leader->__pgrp);
- attach_pid(thread, PIDTYPE_SID, thread->session);
+ attach_pid(thread, PIDTYPE_PGID, thread->signal->pgrp);
+ attach_pid(thread, PIDTYPE_SID, thread->signal->session);
list_add_tail(&thread->tasks, &init_task.tasks);
attach_pid(leader, PIDTYPE_PID, leader->pid);
attach_pid(leader, PIDTYPE_TGID, leader->tgid);
- attach_pid(leader, PIDTYPE_PGID, leader->__pgrp);
- attach_pid(leader, PIDTYPE_SID, leader->session);
+ attach_pid(leader, PIDTYPE_PGID, leader->signal->pgrp);
+ attach_pid(leader, PIDTYPE_SID, leader->signal->session);
}
/*
@@ -268,6 +270,9 @@ void switch_exec_pids(task_t *leader, ta
* machine. From a minimum of 16 slots up to 4096 slots at one gigabyte or
* more.
*/
+#ifdef CONFIG_KGDB
+int kgdb_pid_init_done; /* so we don't call prior to... */
+#endif
void __init pidhash_init(void)
{
int i, j, pidhash_size;
@@ -289,6 +294,9 @@ void __init pidhash_init(void)
for (j = 0; j < pidhash_size; j++)
INIT_LIST_HEAD(&pid_hash[i][j]);
}
+#ifdef CONFIG_KGDB
+ kgdb_pid_init_done++;
+#endif
}
void __init pidmap_init(void)
--- diff/kernel/power/disk.c 2004-03-11 10:20:29.000000000 +0000
+++ source/kernel/power/disk.c 2004-03-16 11:27:48.652791312 +0000
@@ -84,7 +84,6 @@ static void free_some_memory(void)
while (shrink_all_memory(10000))
printk(".");
printk("|\n");
- blk_run_queues();
}
--- diff/kernel/power/pmdisk.c 2004-03-11 10:20:29.000000000 +0000
+++ source/kernel/power/pmdisk.c 2004-03-16 11:27:48.653791160 +0000
@@ -859,7 +859,6 @@ static int end_io(struct bio * bio, unsi
static void wait_io(void)
{
- blk_run_queues();
while(atomic_read(&io_done))
io_schedule();
}
@@ -898,7 +897,7 @@ static int submit(int rw, pgoff_t page_o
if (rw == WRITE)
bio_set_pages_dirty(bio);
start_io();
- submit_bio(rw,bio);
+ submit_bio(rw | (1 << BIO_RW_SYNC), bio);
wait_io();
Done:
bio_put(bio);
--- diff/kernel/power/swsusp.c 2004-03-11 10:20:29.000000000 +0000
+++ source/kernel/power/swsusp.c 2004-03-16 11:27:48.654791008 +0000
@@ -707,11 +707,6 @@ int software_suspend(void)
free_some_memory();
- /* No need to invalidate any vfsmnt list --
- * they will be valid after resume, anyway.
- */
- blk_run_queues();
-
/* Save state of all device drivers, and stop them. */
if ((res = device_suspend(4))==0)
/* If stopping device drivers worked, we proceed basically into
--- diff/kernel/sched.c 2004-03-16 11:24:02.000000000 +0000
+++ source/kernel/sched.c 2004-03-16 11:27:48.659790248 +0000
@@ -25,6 +25,7 @@
#include <linux/highmem.h>
#include <linux/smp_lock.h>
#include <asm/mmu_context.h>
+#include <asm/tlb.h>
#include <linux/interrupt.h>
#include <linux/completion.h>
#include <linux/kernel_stat.h>
@@ -91,7 +92,6 @@
#define MAX_SLEEP_AVG (AVG_TIMESLICE * MAX_BONUS)
#define STARVATION_LIMIT (MAX_SLEEP_AVG)
#define NS_MAX_SLEEP_AVG (JIFFIES_TO_NS(MAX_SLEEP_AVG))
-#define NODE_THRESHOLD 125
#define CREDIT_LIMIT 100
/*
@@ -187,7 +187,7 @@ static inline unsigned int task_timeslic
typedef struct runqueue runqueue_t;
struct prio_array {
- int nr_active;
+ unsigned int nr_active;
unsigned long bitmap[BITMAP_SIZE];
struct list_head queue[MAX_PRIO];
};
@@ -201,25 +201,39 @@ struct prio_array {
*/
struct runqueue {
spinlock_t lock;
+
+ /*
+ * nr_running and cpu_load should be in the same cacheline because
+ * remote CPUs use both these fields when doing load calculation.
+ */
+ unsigned long nr_running;
+#ifdef CONFIG_SMP
+ unsigned long cpu_load;
+#endif
unsigned long long nr_switches;
- unsigned long nr_running, expired_timestamp, nr_uninterruptible,
- timestamp_last_tick;
+ unsigned long expired_timestamp, nr_uninterruptible;
+ unsigned long long timestamp_last_tick;
task_t *curr, *idle;
struct mm_struct *prev_mm;
prio_array_t *active, *expired, arrays[2];
- int best_expired_prio, prev_cpu_load[NR_CPUS];
-#ifdef CONFIG_NUMA
- atomic_t *node_nr_running;
- int prev_node_load[MAX_NUMNODES];
-#endif
+ int best_expired_prio;
+ atomic_t nr_iowait;
+
+ /* For active balancing */
+ int active_balance;
+ int push_cpu;
+
task_t *migration_thread;
struct list_head migration_queue;
-
- atomic_t nr_iowait;
};
static DEFINE_PER_CPU(struct runqueue, runqueues);
+#ifdef CONFIG_SMP
+/* Mandatory scheduling domains */
+DEFINE_PER_CPU(struct sched_domain, base_domains);
+#endif
+
#define cpu_rq(cpu) (&per_cpu(runqueues, (cpu)))
#define this_rq() (&__get_cpu_var(runqueues))
#define task_rq(p) cpu_rq(task_cpu(p))
@@ -234,51 +248,16 @@ static DEFINE_PER_CPU(struct runqueue, r
# define task_running(rq, p) ((rq)->curr == (p))
#endif
-#ifdef CONFIG_NUMA
-
-/*
- * Keep track of running tasks.
- */
-
-static atomic_t node_nr_running[MAX_NUMNODES] ____cacheline_maxaligned_in_smp =
- {[0 ...MAX_NUMNODES-1] = ATOMIC_INIT(0)};
-
-static inline void nr_running_init(struct runqueue *rq)
-{
- rq->node_nr_running = &node_nr_running[0];
-}
-
static inline void nr_running_inc(runqueue_t *rq)
{
- atomic_inc(rq->node_nr_running);
rq->nr_running++;
}
static inline void nr_running_dec(runqueue_t *rq)
{
- atomic_dec(rq->node_nr_running);
rq->nr_running--;
}
-__init void node_nr_running_init(void)
-{
- int i;
-
- for (i = 0; i < NR_CPUS; i++) {
- if (cpu_possible(i))
- cpu_rq(i)->node_nr_running =
- &node_nr_running[cpu_to_node(i)];
- }
-}
-
-#else /* !CONFIG_NUMA */
-
-# define nr_running_init(rq) do { } while (0)
-# define nr_running_inc(rq) do { (rq)->nr_running++; } while (0)
-# define nr_running_dec(rq) do { (rq)->nr_running--; } while (0)
-
-#endif /* CONFIG_NUMA */
-
/*
* task_rq_lock - lock the runqueue a given task resides on and disable
* interrupts. Note the ordering: we can safely lookup the task_rq without
@@ -546,37 +525,30 @@ inline int task_curr(task_t *p)
typedef struct {
struct list_head list;
task_t *task;
+ int dest_cpu;
struct completion done;
} migration_req_t;
/*
- * The task's runqueue lock must be held, and the new mask must be valid.
+ * The task's runqueue lock must be held.
* Returns true if you have to wait for migration thread.
*/
-static int __set_cpus_allowed(task_t *p, cpumask_t new_mask,
- migration_req_t *req)
+static int migrate_task(task_t *p, int dest_cpu, migration_req_t *req)
{
runqueue_t *rq = task_rq(p);
- p->cpus_allowed = new_mask;
- /*
- * Can the task run on the task's current CPU? If not then
- * migrate the thread off to a proper CPU.
- */
- if (cpu_isset(task_cpu(p), new_mask))
- return 0;
-
/*
* If the task is not on a runqueue (and not running), then
* it is sufficient to simply update the task's cpu field.
*/
if (!p->array && !task_running(rq, p)) {
- set_task_cpu(p, any_online_cpu(p->cpus_allowed));
+ set_task_cpu(p, dest_cpu);
return 0;
}
init_completion(&req->done);
req->task = p;
+ req->dest_cpu = dest_cpu;
list_add(&req->list, &rq->migration_queue);
return 1;
}
@@ -637,7 +609,63 @@ void kick_process(task_t *p)
}
EXPORT_SYMBOL_GPL(kick_process);
+/*
+ * Return a low guess at the load of cpu.
+ */
+static inline unsigned long get_low_cpu_load(int cpu)
+{
+ runqueue_t *rq = cpu_rq(cpu);
+ unsigned long load_now = rq->nr_running << SCHED_LOAD_SHIFT;
+
+ return min(rq->cpu_load, load_now);
+}
+
+static inline unsigned long get_high_cpu_load(int cpu)
+{
+ runqueue_t *rq = cpu_rq(cpu);
+ unsigned long load_now = rq->nr_running << SCHED_LOAD_SHIFT;
+
+ return max(rq->cpu_load, load_now);
+}
+
+#endif
+
+/*
+ * sched_balance_wake can be used with SMT architectures to wake a
+ * task onto an idle sibling if cpu is not idle. Returns cpu if
+ * cpu is idle or no siblings are idle, otherwise returns an idle
+ * sibling.
+ */
+#if defined(CONFIG_SMP) && defined(ARCH_HAS_SCHED_WAKE_BALANCE)
+static int sched_balance_wake(int cpu, task_t *p)
+{
+ cpumask_t tmp;
+ struct sched_domain *domain;
+ int i;
+
+ if (idle_cpu(cpu))
+ return cpu;
+
+ domain = cpu_sched_domain(cpu);
+ if (!(domain->flags & SD_FLAG_WAKE))
+ return cpu;
+
+ cpus_and(tmp, domain->span, cpu_online_map);
+ for_each_cpu_mask(i, tmp) {
+ if (!cpu_isset(i, p->cpus_allowed))
+ continue;
+ if (idle_cpu(i))
+ return i;
+ }
+
+ return cpu;
+}
+#else
+static inline int sched_balance_wake(int cpu, task_t *p)
+{
+ return cpu;
+}
#endif
/***
@@ -660,44 +688,113 @@ static int try_to_wake_up(task_t * p, un
int success = 0;
long old_state;
runqueue_t *rq;
+ int cpu, this_cpu;
+#ifdef CONFIG_SMP
+ unsigned long long now;
+ unsigned long load, this_load;
+ int new_cpu;
+ struct sched_domain *sd;
+#endif
-repeat_lock_task:
rq = task_rq_lock(p, &flags);
old_state = p->state;
- if (old_state & state) {
- if (!p->array) {
- /*
- * Fast-migrate the task if it's not running or runnable
- * currently. Do not violate hard affinity.
- */
- if (unlikely(sync && !task_running(rq, p) &&
- (task_cpu(p) != smp_processor_id()) &&
- cpu_isset(smp_processor_id(),
- p->cpus_allowed))) {
-
- set_task_cpu(p, smp_processor_id());
- task_rq_unlock(rq, &flags);
- goto repeat_lock_task;
- }
- if (old_state == TASK_UNINTERRUPTIBLE) {
- rq->nr_uninterruptible--;
- /*
- * Tasks on involuntary sleep don't earn
- * sleep_avg beyond just interactive state.
- */
- p->activated = -1;
- }
- if (sync && (task_cpu(p) == smp_processor_id()))
- __activate_task(p, rq);
- else {
- activate_task(p, rq);
- if (TASK_PREEMPTS_CURR(p, rq))
- resched_task(rq->curr);
- }
- success = 1;
+ if (!(old_state & state))
+ goto out;
+
+ if (p->array)
+ goto out_running;
+
+ this_cpu = smp_processor_id();
+ cpu = task_cpu(p);
+
+#ifdef CONFIG_SMP
+ if (cpu == this_cpu)
+ goto out_activate;
+
+ if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed)
+ || task_running(rq, p)))
+ goto out_activate;
+
+ /* Passive load balancing */
+ load = get_low_cpu_load(cpu);
+ this_load = get_high_cpu_load(this_cpu) + SCHED_LOAD_SCALE;
+ if (load > this_load) {
+ new_cpu = sched_balance_wake(this_cpu, p);
+ set_task_cpu(p, new_cpu);
+ goto repeat_lock_task;
+ }
+
+ now = sched_clock();
+ sd = cpu_sched_domain(this_cpu);
+
+ /*
+ * Fast-migrate the task if it's not running or
+ * runnable currently. Do not violate hard affinity.
+ */
+ do {
+ if (!(sd->flags & SD_FLAG_FASTMIGRATE))
+ break;
+ if (now - p->timestamp < sd->cache_hot_time)
+ break;
+
+ if (cpu_isset(cpu, sd->span)) {
+ new_cpu = sched_balance_wake(this_cpu, p);
+ set_task_cpu(p, new_cpu);
+ goto repeat_lock_task;
}
- p->state = TASK_RUNNING;
+ sd = sd->parent;
+ } while (sd);
+
+ new_cpu = sched_balance_wake(cpu, p);
+ if (new_cpu != cpu) {
+ set_task_cpu(p, new_cpu);
+ goto repeat_lock_task;
+ }
+ if ((p->flags & PF_FUTEX_DEBUG)
+ && !(current->flags & PF_FUTEX_DEBUG)) {
+ printk("%s %i waking %s: %i %i\n",
+ current->comm, (int)in_interrupt(),
+ p->comm, p->tgid, p->pid);
+ WARN_ON(1);
+ }
+ goto out_activate;
+
+repeat_lock_task:
+ task_rq_unlock(rq, &flags);
+ rq = task_rq_lock(p, &flags);
+ old_state = p->state;
+ if (!(old_state & state))
+ goto out;
+
+ if (p->array)
+ goto out_running;
+
+ this_cpu = smp_processor_id();
+ cpu = task_cpu(p);
+
+out_activate:
+#endif /* CONFIG_SMP */
+ if (old_state == TASK_UNINTERRUPTIBLE) {
+ rq->nr_uninterruptible--;
+ /*
+ * Tasks on involuntary sleep don't earn
+ * sleep_avg beyond just interactive state.
+ */
+ p->activated = -1;
}
+
+ if (sync && cpu == this_cpu) {
+ __activate_task(p, rq);
+ } else {
+ activate_task(p, rq);
+ if (TASK_PREEMPTS_CURR(p, rq))
+ resched_task(rq->curr);
+ }
+ success = 1;
+
+out_running:
+ p->state = TASK_RUNNING;
+out:
task_rq_unlock(rq, &flags);
return success;
@@ -756,8 +853,8 @@ void fastcall sched_fork(task_t *p)
p->timestamp = sched_clock();
if (!current->time_slice) {
/*
- * This case is rare, it happens when the parent has only
- * a single jiffy left from its timeslice. Taking the
+ * This case is rare, it happens when the parent has only
+ * a single jiffy left from its timeslice. Taking the
* runqueue lock is not a problem.
*/
current->time_slice = 1;
@@ -873,7 +970,7 @@ static inline void finish_task_switch(ta
* still held, otherwise prev could be scheduled on another cpu, die
* there before we look at prev->state, and then the reference would
* be dropped twice.
- * Manfred Spraul <manfred@colorfullife.com>
+ * Manfred Spraul <manfred@colorfullife.com>
*/
prev_task_flags = prev->flags;
finish_arch_switch(rq, prev);
@@ -935,7 +1032,7 @@ unsigned long nr_running(void)
{
unsigned long i, sum = 0;
- for (i = 0; i < NR_CPUS; i++)
+ for_each_cpu(i)
sum += cpu_rq(i)->nr_running;
return sum;
@@ -1005,6 +1102,14 @@ static inline void double_rq_unlock(runq
spin_unlock(&rq2->lock);
}
+enum idle_type
+{
+ IDLE,
+ NOT_IDLE,
+ NEWLY_IDLE,
+};
+
+#ifdef CONFIG_SMP
#ifdef CONFIG_NUMA
/*
* If dest_cpu is allowed for this process, migrate the task to it.
@@ -1017,28 +1122,26 @@ static void sched_migrate_task(task_t *p
runqueue_t *rq;
migration_req_t req;
unsigned long flags;
- cpumask_t old_mask, new_mask = cpumask_of_cpu(dest_cpu);
rq = task_rq_lock(p, &flags);
- old_mask = p->cpus_allowed;
- if (!cpu_isset(dest_cpu, old_mask) || !cpu_online(dest_cpu))
+ if (!cpu_isset(dest_cpu, p->cpus_allowed))
goto out;
/* force the process onto the specified CPU */
- if (__set_cpus_allowed(p, new_mask, &req)) {
+ if (migrate_task(p, dest_cpu, &req)) {
/* Need to wait for migration thread. */
task_rq_unlock(rq, &flags);
wake_up_process(rq->migration_thread);
wait_for_completion(&req.done);
- /* If we raced with sys_sched_setaffinity, don't
- * restore mask. */
- rq = task_rq_lock(p, &flags);
- if (likely(cpus_equal(p->cpus_allowed, new_mask))) {
- /* Restore old mask: won't need migration
- * thread, since current cpu is allowed. */
- BUG_ON(__set_cpus_allowed(p, old_mask, NULL));
- }
+ /*
+ * we want a new context here. This eliminates TLB
+ * flushes on the cpus where the process executed prior to
+ * the migration.
+ */
+ tlb_migrate_prepare(current->mm);
+
+ return;
}
out:
task_rq_unlock(rq, &flags);
@@ -1048,215 +1151,81 @@ out:
* Find the least loaded CPU. Slightly favor the current CPU by
* setting its runqueue length as the minimum to start.
*/
-static int sched_best_cpu(struct task_struct *p)
+static int sched_best_cpu(struct task_struct *p, struct sched_domain *domain)
{
- int i, minload, load, best_cpu, node = 0;
- cpumask_t cpumask;
+ cpumask_t tmp;
+ int i, min_load, this_cpu, best_cpu;
- best_cpu = task_cpu(p);
- if (cpu_rq(best_cpu)->nr_running <= 2)
- return best_cpu;
+ best_cpu = this_cpu = task_cpu(p);
+ min_load = INT_MAX;
- minload = 10000000;
- for_each_node_with_cpus(i) {
- /*
- * Node load is always divided by nr_cpus_node to normalise
- * load values in case cpu count differs from node to node.
- * We first multiply node_nr_running by 10 to get a little
- * better resolution.
- */
- load = 10 * atomic_read(&node_nr_running[i]) / nr_cpus_node(i);
- if (load < minload) {
- minload = load;
- node = i;
- }
- }
+ cpus_and(tmp, domain->span, cpu_online_map);
+ for_each_cpu_mask(i, tmp) {
+ unsigned long load;
+ if (i == this_cpu)
+ load = get_low_cpu_load(i);
+ else
+ load = get_high_cpu_load(i) + SCHED_LOAD_SCALE;
- minload = 10000000;
- cpumask = node_to_cpumask(node);
- for (i = 0; i < NR_CPUS; ++i) {
- if (!cpu_isset(i, cpumask))
- continue;
- if (cpu_rq(i)->nr_running < minload) {
+ if (min_load > load) {
best_cpu = i;
- minload = cpu_rq(i)->nr_running;
+ min_load = load;
}
+
}
return best_cpu;
}
void sched_balance_exec(void)
{
+ struct sched_domain *domain = this_sched_domain();
int new_cpu;
+ int this_cpu = smp_processor_id();
+ if (numnodes == 1)
+ return;
- if (numnodes > 1) {
- new_cpu = sched_best_cpu(current);
- if (new_cpu != smp_processor_id())
- sched_migrate_task(current, new_cpu);
- }
-}
+ if (this_rq()->nr_running <= 1)
+ return;
-/*
- * Find the busiest node. All previous node loads contribute with a
- * geometrically deccaying weight to the load measure:
- * load_{t} = load_{t-1}/2 + nr_node_running_{t}
- * This way sudden load peaks are flattened out a bit.
- * Node load is divided by nr_cpus_node() in order to compare nodes
- * of different cpu count but also [first] multiplied by 10 to
- * provide better resolution.
- */
-static int find_busiest_node(int this_node)
-{
- int i, node = -1, load, this_load, maxload;
-
- if (!nr_cpus_node(this_node))
- return node;
- this_load = maxload = (this_rq()->prev_node_load[this_node] >> 1)
- + (10 * atomic_read(&node_nr_running[this_node])
- / nr_cpus_node(this_node));
- this_rq()->prev_node_load[this_node] = this_load;
- for_each_node_with_cpus(i) {
- if (i == this_node)
- continue;
- load = (this_rq()->prev_node_load[i] >> 1)
- + (10 * atomic_read(&node_nr_running[i])
- / nr_cpus_node(i));
- this_rq()->prev_node_load[i] = load;
- if (load > maxload && (100*load > NODE_THRESHOLD*this_load)) {
- maxload = load;
- node = i;
- }
+ while (domain->parent && !(domain->flags & SD_FLAG_EXEC))
+ domain = domain->parent;
+
+ if (domain->flags & SD_FLAG_EXEC) {
+ new_cpu = sched_best_cpu(current, domain);
+ if (new_cpu != this_cpu)
+ sched_migrate_task(current, new_cpu);
}
- return node;
}
-
#endif /* CONFIG_NUMA */
-#ifdef CONFIG_SMP
-
/*
- * double_lock_balance - lock the busiest runqueue
- *
- * this_rq is locked already. Recalculate nr_running if we have to
- * drop the runqueue lock.
+ * double_lock_balance - lock the busiest runqueue, this_rq is locked already.
*/
-static inline
-unsigned int double_lock_balance(runqueue_t *this_rq, runqueue_t *busiest,
- int this_cpu, int idle,
- unsigned int nr_running)
+static inline void double_lock_balance(runqueue_t *this_rq, runqueue_t *busiest)
{
if (unlikely(!spin_trylock(&busiest->lock))) {
if (busiest < this_rq) {
spin_unlock(&this_rq->lock);
spin_lock(&busiest->lock);
spin_lock(&this_rq->lock);
- /* Need to recalculate nr_running */
- if (idle || (this_rq->nr_running >
- this_rq->prev_cpu_load[this_cpu]))
- nr_running = this_rq->nr_running;
- else
- nr_running = this_rq->prev_cpu_load[this_cpu];
} else
spin_lock(&busiest->lock);
}
- return nr_running;
-}
-
-/*
- * find_busiest_queue - find the busiest runqueue among the cpus in cpumask.
- */
-static inline
-runqueue_t *find_busiest_queue(runqueue_t *this_rq, int this_cpu, int idle,
- int *imbalance, cpumask_t cpumask)
-{
- int nr_running, load, max_load, i;
- runqueue_t *busiest, *rq_src;
-
- /*
- * We search all runqueues to find the most busy one.
- * We do this lockless to reduce cache-bouncing overhead,
- * we re-check the 'best' source CPU later on again, with
- * the lock held.
- *
- * We fend off statistical fluctuations in runqueue lengths by
- * saving the runqueue length (as seen by the balancing CPU) during
- * the previous load-balancing operation and using the smaller one
- * of the current and saved lengths. If a runqueue is long enough
- * for a longer amount of time then we recognize it and pull tasks
- * from it.
- *
- * The 'current runqueue length' is a statistical maximum variable,
- * for that one we take the longer one - to avoid fluctuations in
- * the other direction. So for a load-balance to happen it needs
- * stable long runqueue on the target CPU and stable short runqueue
- * on the local runqueue.
- *
- * We make an exception if this CPU is about to become idle - in
- * that case we are less picky about moving a task across CPUs and
- * take what can be taken.
- */
- if (idle || (this_rq->nr_running > this_rq->prev_cpu_load[this_cpu]))
- nr_running = this_rq->nr_running;
- else
- nr_running = this_rq->prev_cpu_load[this_cpu];
-
- busiest = NULL;
- max_load = 1;
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_isset(i, cpumask))
- continue;
-
- rq_src = cpu_rq(i);
- if (idle || (rq_src->nr_running < this_rq->prev_cpu_load[i]))
- load = rq_src->nr_running;
- else
- load = this_rq->prev_cpu_load[i];
- this_rq->prev_cpu_load[i] = rq_src->nr_running;
-
- if ((load > max_load) && (rq_src != this_rq)) {
- busiest = rq_src;
- max_load = load;
- }
- }
-
- if (likely(!busiest))
- goto out;
-
- *imbalance = max_load - nr_running;
-
- /* It needs an at least ~25% imbalance to trigger balancing. */
- if (!idle && ((*imbalance)*4 < max_load)) {
- busiest = NULL;
- goto out;
- }
-
- nr_running = double_lock_balance(this_rq, busiest, this_cpu,
- idle, nr_running);
- /*
- * Make sure nothing changed since we checked the
- * runqueue length.
- */
- if (busiest->nr_running <= nr_running) {
- spin_unlock(&busiest->lock);
- busiest = NULL;
- }
-out:
- return busiest;
}
/*
* pull_task - move a task from a remote runqueue to the local runqueue.
* Both runqueues must be locked.
*/
-static inline
-void pull_task(runqueue_t *src_rq, prio_array_t *src_array, task_t *p,
- runqueue_t *this_rq, int this_cpu)
+static inline void pull_task(runqueue_t *src_rq, prio_array_t *src_array,
+ task_t *p, runqueue_t *this_rq, prio_array_t *this_array,
+ int this_cpu)
{
dequeue_task(p, src_array);
nr_running_dec(src_rq);
set_task_cpu(p, this_cpu);
nr_running_inc(this_rq);
- enqueue_task(p, this_rq->active);
+ enqueue_task(p, this_array);
p->timestamp = sched_clock() -
(src_rq->timestamp_last_tick - p->timestamp);
/*
@@ -1264,69 +1233,71 @@ void pull_task(runqueue_t *src_rq, prio_
* to be always true for them.
*/
if (TASK_PREEMPTS_CURR(p, this_rq))
- set_need_resched();
+ resched_task(this_rq->curr);
}
/*
* can_migrate_task - may task p from runqueue rq be migrated to this_cpu?
*/
static inline
-int can_migrate_task(task_t *tsk, runqueue_t *rq, int this_cpu, int idle)
+int can_migrate_task(task_t *p, runqueue_t *rq, int this_cpu,
+ struct sched_domain *domain, enum idle_type idle)
{
- unsigned long delta = rq->timestamp_last_tick - tsk->timestamp;
-
/*
* We do not migrate tasks that are:
* 1) running (obviously), or
* 2) cannot be migrated to this CPU due to cpus_allowed, or
* 3) are cache-hot on their current CPU.
*/
- if (task_running(rq, tsk))
+ if (task_running(rq, p))
return 0;
- if (!cpu_isset(this_cpu, tsk->cpus_allowed))
- return 0;
- if (!idle && (delta <= JIFFIES_TO_NS(cache_decay_ticks)))
+ if (!cpu_isset(this_cpu, p->cpus_allowed))
return 0;
+
+ /* Aggressive migration if we've failed balancing */
+ if (idle == NEWLY_IDLE ||
+ domain->nr_balance_failed < domain->cache_nice_tries) {
+ if ((rq->timestamp_last_tick - p->timestamp)
+ < domain->cache_hot_time)
+ return 0;
+ }
+
return 1;
}
/*
- * Current runqueue is empty, or rebalance tick: if there is an
- * inbalance (current runqueue is too short) then pull from
- * busiest runqueue(s).
- *
- * We call this with the current runqueue locked,
- * irqs disabled.
- */
-static void load_balance(runqueue_t *this_rq, int idle, cpumask_t cpumask)
+ * move_tasks tries to move up to max_nr_move tasks from busiest to this_rq,
+ * as part of a balancing operation within "domain". Returns the number of
+ * tasks moved.
+ *
+ * Called with both runqueues locked.
+ */
+static int move_tasks(runqueue_t *this_rq, int this_cpu, runqueue_t *busiest,
+ unsigned long max_nr_move, struct sched_domain *domain,
+ enum idle_type idle)
{
- int imbalance, idx, this_cpu = smp_processor_id();
- runqueue_t *busiest;
- prio_array_t *array;
+ int idx;
+ int pulled = 0;
+ prio_array_t *array, *dst_array;
struct list_head *head, *curr;
task_t *tmp;
- busiest = find_busiest_queue(this_rq, this_cpu, idle,
- &imbalance, cpumask);
- if (!busiest)
+ if (max_nr_move <= 0 || busiest->nr_running <= 1)
goto out;
/*
- * We only want to steal a number of tasks equal to 1/2 the imbalance,
- * otherwise we'll just shift the imbalance to the new queue:
- */
- imbalance /= 2;
-
- /*
* We first consider expired tasks. Those will likely not be
* executed in the near future, and they are most likely to
* be cache-cold, thus switching CPUs has the least effect
* on them.
*/
- if (busiest->expired->nr_active)
+ if (busiest->expired->nr_active) {
array = busiest->expired;
- else
+ dst_array = this_rq->expired;
+ } else {
array = busiest->active;
+ dst_array = this_rq->active;
+ }
new_array:
/* Start searching at priority 0: */
@@ -1339,9 +1310,10 @@ skip_bitmap:
if (idx >= MAX_PRIO) {
if (array == busiest->expired) {
array = busiest->active;
+ dst_array = this_rq->active;
goto new_array;
}
- goto out_unlock;
+ goto out;
}
head = array->queue + idx;
@@ -1351,104 +1323,468 @@ skip_queue:
curr = curr->prev;
- if (!can_migrate_task(tmp, busiest, this_cpu, idle)) {
+ if (!can_migrate_task(tmp, busiest, this_cpu, domain, idle)) {
if (curr != head)
goto skip_queue;
idx++;
goto skip_bitmap;
}
- pull_task(busiest, array, tmp, this_rq, this_cpu);
+ pull_task(busiest, array, tmp, this_rq, dst_array, this_cpu);
+ pulled++;
- /* Only migrate one task if we are idle */
- if (!idle && --imbalance) {
+ /* We only want to steal up to the prescribed number of tasks. */
+ if (pulled < max_nr_move) {
if (curr != head)
goto skip_queue;
idx++;
goto skip_bitmap;
}
-out_unlock:
- spin_unlock(&busiest->lock);
out:
- ;
+ return pulled;
+}
+
+/*
+ * find_busiest_group finds and returns the busiest CPU group within the
+ * domain. It calculates and returns the number of tasks which should be
+ * moved to restore balance via the imbalance parameter.
+ */
+static struct sched_group *
+find_busiest_group(struct sched_domain *domain, int this_cpu,
+ unsigned long *imbalance, enum idle_type idle)
+{
+ unsigned long max_load, avg_load, total_load, this_load;
+ unsigned int total_pwr;
+ struct sched_group *busiest = NULL, *this = NULL, *group = domain->groups;
+
+ max_load = 0;
+ this_load = 0;
+ total_load = 0;
+ total_pwr = 0;
+
+ if (group == NULL)
+ goto out_balanced;
+
+ do {
+ cpumask_t tmp;
+ unsigned long load;
+ int local_group;
+ int i, nr_cpus = 0;
+
+ local_group = cpu_isset(this_cpu, group->cpumask);
+
+ /* Tally up the load of all CPUs in the group */
+ avg_load = 0;
+ cpus_and(tmp, group->cpumask, cpu_online_map);
+ for_each_cpu_mask(i, tmp) {
+ /* Bias balancing toward cpus of our domain */
+ if (local_group) {
+ load = get_high_cpu_load(i);
+ } else
+ load = get_low_cpu_load(i);
+
+ nr_cpus++;
+ avg_load += load;
+ }
+
+ if (!nr_cpus)
+ goto nextgroup;
+
+ total_load += avg_load;
+ total_pwr += group->cpu_power;
+
+ /* Adjust by relative CPU power of the group */
+ avg_load = (avg_load << SCHED_LOAD_SHIFT) / group->cpu_power;
+
+ if (local_group) {
+ this_load = avg_load;
+ this = group;
+ goto nextgroup;
+ }
+ if (avg_load > max_load) {
+ max_load = avg_load;
+ busiest = group;
+ }
+nextgroup:
+ group = group->next;
+ } while (group != domain->groups);
+
+ if (!busiest || this_load >= max_load)
+ goto out_balanced;
+
+ avg_load = (SCHED_LOAD_SCALE * total_load) / total_pwr;
+
+ if (idle == NOT_IDLE) {
+ if (this_load >= avg_load ||
+ 100*max_load <= domain->imbalance_pct*this_load)
+ goto out_balanced;
+ }
+
+ /*
+ * We're trying to get all the cpus to the average_load, so we don't
+ * want to push ourselves above the average load, nor do we wish to
+ * reduce the max loaded cpu below the average load, as either of these
+ * actions would just result in more rebalancing later, and ping-pong
+ * tasks around. Thus we look for the minimum possible imbalance.
+ * Negative imbalances (*we* are more loaded than anyone else) will
+ * be counted as no imbalance for these purposes -- we can't fix that
+ * by pulling tasks to us. Be careful of negative numbers as they'll
+ * appear as very large values with unsigned longs.
+ */
+ *imbalance = (min(max_load - avg_load, avg_load - this_load) + 1) / 2;
+
+ if (*imbalance <= SCHED_LOAD_SCALE/2) {
+ unsigned long pwr_now = 0, pwr_move = 0;
+ unsigned long tmp;
+
+ /*
+ * OK, we don't have enough imbalance to justify moving tasks,
+ * however we may be able to increase total CPU power used by
+ * moving them.
+ */
+
+ pwr_now += busiest->cpu_power*min(SCHED_LOAD_SCALE, max_load);
+ pwr_now += this->cpu_power*min(SCHED_LOAD_SCALE, this_load);
+ pwr_now >>= SCHED_LOAD_SHIFT;
+
+ /* Amount of load we'd subtract */
+ tmp = SCHED_LOAD_SCALE*SCHED_LOAD_SCALE/busiest->cpu_power;
+ if (max_load > tmp)
+ pwr_move += busiest->cpu_power*min(SCHED_LOAD_SCALE,
+ max_load - tmp);
+
+ /* Amount of load we'd add */
+ tmp = SCHED_LOAD_SCALE*SCHED_LOAD_SCALE/this->cpu_power;
+ pwr_move += this->cpu_power*min(this->cpu_power, this_load + tmp);
+ pwr_move >>= SCHED_LOAD_SHIFT;
+
+ /* Move if we gain another 8th of a CPU worth of throughput */
+ if (pwr_move < pwr_now + SCHED_LOAD_SCALE / 8)
+ goto out_balanced;
+ *imbalance = 1;
+ return busiest;
+ }
+
+ /* How many tasks to actually move to equalise the imbalance */
+ *imbalance = (*imbalance * min(busiest->cpu_power, this->cpu_power))
+ >> SCHED_LOAD_SHIFT;
+ /* Get rid of the scaling factor, rounding *up* as we divide */
+ *imbalance = (*imbalance + SCHED_LOAD_SCALE/2) >> SCHED_LOAD_SHIFT;
+
+ return busiest;
+
+out_balanced:
+ if (busiest && idle == NEWLY_IDLE) {
+ *imbalance = 1;
+ return busiest;
+ }
+
+ *imbalance = 0;
+ return NULL;
}
/*
- * One of the idle_cpu_tick() and busy_cpu_tick() functions will
- * get called every timer tick, on every CPU. Our balancing action
- * frequency and balancing agressivity depends on whether the CPU is
- * idle or not.
+ * find_busiest_queue - find the busiest runqueue among the cpus in group.
+ */
+static runqueue_t *find_busiest_queue(struct sched_group *group)
+{
+ cpumask_t tmp;
+ int i;
+ unsigned long max_load = 0;
+ runqueue_t *busiest = NULL;
+
+ cpus_and(tmp, group->cpumask, cpu_online_map);
+ for_each_cpu_mask(i, tmp) {
+ unsigned long load;
+
+ load = get_low_cpu_load(i);
+
+ if (load >= max_load) {
+ max_load = load;
+ busiest = cpu_rq(i);
+ }
+ }
+
+ return busiest;
+}
+
+/*
+ * Check this_cpu to ensure it is balanced within domain. Attempt to move
+ * tasks if there is an imbalance.
*
- * busy-rebalance every 200 msecs. idle-rebalance every 1 msec. (or on
- * systems with HZ=100, every 10 msecs.)
+ * Called with this_rq unlocked.
+ */
+static int load_balance(int this_cpu, runqueue_t *this_rq,
+ struct sched_domain *domain, enum idle_type idle)
+{
+ struct sched_group *group;
+ runqueue_t *busiest = NULL;
+ unsigned long imbalance;
+ int balanced = 0, failed = 0;
+ int nr_moved = 0;
+
+ spin_lock(&this_rq->lock);
+
+ group = find_busiest_group(domain, this_cpu, &imbalance, idle);
+ if (!group) {
+ balanced = 1;
+ goto out;
+ }
+
+ busiest = find_busiest_queue(group);
+ if (!busiest || busiest == this_rq) {
+ balanced = 1;
+ goto out;
+ }
+
+ /* Attempt to move tasks */
+ double_lock_balance(this_rq, busiest);
+
+ nr_moved = move_tasks(this_rq, this_cpu, busiest,
+ imbalance, domain, idle);
+ spin_unlock(&busiest->lock);
+out:
+ spin_unlock(&this_rq->lock);
+
+ if (!balanced && nr_moved == 0)
+ failed = 1;
+
+ if (failed && busiest &&
+ domain->nr_balance_failed > domain->cache_nice_tries) {
+ int wake = 0;
+
+ spin_lock(&busiest->lock);
+ if (!busiest->active_balance) {
+ busiest->active_balance = 1;
+ busiest->push_cpu = this_cpu;
+ wake = 1;
+ }
+ spin_unlock(&busiest->lock);
+ if (wake)
+ wake_up_process(busiest->migration_thread);
+ }
+
+ if (failed)
+ domain->nr_balance_failed++;
+ else
+ domain->nr_balance_failed = 0;
+
+ if (balanced) {
+ if (domain->balance_interval < domain->max_interval)
+ domain->balance_interval *= 2;
+ } else {
+ domain->balance_interval = domain->min_interval;
+ }
+
+ return nr_moved;
+}
+
+/*
+ * Check this_cpu to ensure it is balanced within domain. Attempt to move
+ * tasks if there is an imbalance.
*
- * On NUMA, do a node-rebalance every 400 msecs.
+ * Called from schedule when this_rq is about to become idle (NEWLY_IDLE).
+ * this_rq is locked.
*/
-#define IDLE_REBALANCE_TICK (HZ/1000 ?: 1)
-#define BUSY_REBALANCE_TICK (HZ/5 ?: 1)
-#define IDLE_NODE_REBALANCE_TICK (IDLE_REBALANCE_TICK * 5)
-#define BUSY_NODE_REBALANCE_TICK (BUSY_REBALANCE_TICK * 2)
+static int load_balance_newidle(int this_cpu, runqueue_t *this_rq,
+ struct sched_domain *domain)
+{
+ struct sched_group *group;
+ runqueue_t *busiest = NULL;
+ unsigned long imbalance;
+ int nr_moved = 0;
-#ifdef CONFIG_NUMA
-static void balance_node(runqueue_t *this_rq, int idle, int this_cpu)
+ group = find_busiest_group(domain, this_cpu, &imbalance, NEWLY_IDLE);
+ if (!group)
+ goto out;
+
+ busiest = find_busiest_queue(group);
+ if (!busiest || busiest == this_rq)
+ goto out;
+
+ /* Attempt to move tasks */
+ double_lock_balance(this_rq, busiest);
+
+ nr_moved = move_tasks(this_rq, this_cpu, busiest,
+ imbalance, domain, NEWLY_IDLE);
+
+ spin_unlock(&busiest->lock);
+
+out:
+ return nr_moved;
+}
+
+/*
+ * idle_balance is called by schedule() if this_cpu is about to become
+ * idle. Attempts to pull tasks from other CPUs.
+ */
+static inline void idle_balance(int this_cpu, runqueue_t *this_rq)
{
- int node = find_busiest_node(cpu_to_node(this_cpu));
+ struct sched_domain *domain = this_sched_domain();
- if (node >= 0) {
- cpumask_t cpumask = node_to_cpumask(node);
- cpu_set(this_cpu, cpumask);
- spin_lock(&this_rq->lock);
- load_balance(this_rq, idle, cpumask);
- spin_unlock(&this_rq->lock);
- }
+ do {
+ if (unlikely(!domain->groups))
+ /* hasn't been setup yet */
+ break;
+
+ if (domain->flags & SD_FLAG_NEWIDLE) {
+ if (load_balance_newidle(this_cpu, this_rq, domain)) {
+ /* We've pulled tasks over so stop searching */
+ break;
+ }
+ }
+
+ domain = domain->parent;
+ } while (domain);
}
-#endif
-static void rebalance_tick(runqueue_t *this_rq, int idle)
+/*
+ * active_load_balance is run by migration threads. It pushes a running
+ * task off the cpu. It can be required to correctly have at least 1 task
+ * running on each physical CPU where possible, and not have a physical /
+ * logical imbalance.
+ *
+ * Called with busiest locked.
+ */
+static void active_load_balance(runqueue_t *busiest, int busiest_cpu)
{
-#ifdef CONFIG_NUMA
- int this_cpu = smp_processor_id();
-#endif
- unsigned long j = jiffies;
+ int i;
+ struct sched_domain *sd = cpu_sched_domain(busiest_cpu);
+ struct sched_group *group, *busy_group;
- /*
- * First do inter-node rebalancing, then intra-node rebalancing,
- * if both events happen in the same tick. The inter-node
- * rebalancing does not necessarily have to create a perfect
- * balance within the node, since we load-balance the most loaded
- * node with the current CPU. (ie. other CPUs in the local node
- * are not balanced.)
- */
- if (idle) {
-#ifdef CONFIG_NUMA
- if (!(j % IDLE_NODE_REBALANCE_TICK))
- balance_node(this_rq, idle, this_cpu);
-#endif
- if (!(j % IDLE_REBALANCE_TICK)) {
- spin_lock(&this_rq->lock);
- load_balance(this_rq, idle, cpu_to_node_mask(this_cpu));
- spin_unlock(&this_rq->lock);
+ if (busiest->nr_running <= 1)
+ return;
+
+ /* sd->parent should never cause a NULL dereference, if it did so,
+ * then push_cpu was set to a buggy value */
+ while (!cpu_isset(busiest->push_cpu, sd->span)) {
+ sd = sd->parent;
+ if (!sd->parent && !cpu_isset(busiest->push_cpu, sd->span)) {
+ WARN_ON(1);
+ return;
}
+ }
+
+ if (!sd->groups) {
+ WARN_ON(1);
return;
}
-#ifdef CONFIG_NUMA
- if (!(j % BUSY_NODE_REBALANCE_TICK))
- balance_node(this_rq, idle, this_cpu);
-#endif
- if (!(j % BUSY_REBALANCE_TICK)) {
- spin_lock(&this_rq->lock);
- load_balance(this_rq, idle, cpu_to_node_mask(this_cpu));
- spin_unlock(&this_rq->lock);
+
+ group = sd->groups;
+ while (!cpu_isset(busiest_cpu, group->cpumask)) {
+ group = group->next;
+ if (group == sd->groups) {
+ WARN_ON(1);
+ return;
+ }
}
+ busy_group = group;
+
+ group = sd->groups;
+ do {
+ cpumask_t tmp;
+ runqueue_t *rq;
+ int push_cpu = 0, nr = 0;
+
+ if (group == busy_group)
+ goto next_group;
+
+ cpus_and(tmp, group->cpumask, cpu_online_map);
+ for_each_cpu_mask(i, tmp) {
+ if (!idle_cpu(i))
+ goto next_group;
+ push_cpu = i;
+ nr++;
+ }
+ if (nr == 0)
+ goto next_group;
+
+ rq = cpu_rq(push_cpu);
+ double_lock_balance(busiest, rq);
+ move_tasks(rq, push_cpu, busiest, 1, sd, IDLE);
+ spin_unlock(&rq->lock);
+next_group:
+ group = group->next;
+ } while (group != sd->groups);
+}
+
+/*
+ * rebalance_tick will get called every timer tick, on every CPU.
+ *
+ * It checks each scheduling domain to see if it is due to be balanced,
+ * and initiates a balancing operation if so.
+ *
+ * Balancing parameters are set up in arch_init_sched_domains.
+ */
+
+/* Don't have all balancing operations going off at once */
+#define CPU_OFFSET(cpu) (HZ * cpu / NR_CPUS)
+
+static void rebalance_tick(int this_cpu, runqueue_t *this_rq, enum idle_type idle)
+{
+ unsigned long old_load, this_load;
+ unsigned long j = jiffies + CPU_OFFSET(this_cpu);
+ struct sched_domain *domain = this_sched_domain();
+
+ /* Update our load */
+ old_load = this_rq->cpu_load;
+ this_load = this_rq->nr_running << SCHED_LOAD_SHIFT;
+ this_rq->cpu_load = (old_load + this_load) / 2;
+
+ /* Run through all this CPU's domains */
+ do {
+ unsigned long interval;
+
+ if (unlikely(!domain->groups))
+ break;
+
+ interval = domain->balance_interval;
+ if (idle != IDLE)
+ interval *= domain->busy_factor;
+
+ /* scale ms to jiffies */
+ interval = interval * HZ / 1000;
+ if (unlikely(interval == 0))
+ interval = 1;
+
+ if (j - domain->last_balance >= interval) {
+ if (load_balance(this_cpu, this_rq, domain, idle)) {
+ /* We've pulled tasks over so no longer idle */
+ idle = NOT_IDLE;
+ }
+ domain->last_balance += interval;
+ }
+
+ domain = domain->parent;
+ } while (domain);
}
#else
/*
* on UP we do not need to balance between CPUs:
*/
-static inline void rebalance_tick(runqueue_t *this_rq, int idle)
+static inline void rebalance_tick(int this_cpu, runqueue_t *this_rq, enum idle_type idle)
{
}
#endif
+#ifdef CONFIG_SCHED_SMT
+static inline int wake_priority_sleeper(runqueue_t *rq)
+{ /*
+ * If an SMT sibling task has been put to sleep for priority
+ * reasons reschedule the idle task to see if it can now run.
+ */
+ if (rq->nr_running) {
+ resched_task(rq->idle);
+ return 1;
+ }
+ return 0;
+}
+#else
+static inline int wake_priority_sleeper(runqueue_t *rq)
+{
+ return 0;
+}
+#endif
+
DEFINE_PER_CPU(struct kernel_stat, kstat);
EXPORT_PER_CPU_SYMBOL(kstat);
@@ -1502,7 +1838,9 @@ void scheduler_tick(int user_ticks, int
cpustat->iowait += sys_ticks;
else
cpustat->idle += sys_ticks;
- rebalance_tick(rq, 1);
+ if (wake_priority_sleeper(rq))
+ goto out;
+ rebalance_tick(cpu, rq, IDLE);
return;
}
if (TASK_NICE(p) > 0)
@@ -1586,9 +1924,94 @@ void scheduler_tick(int user_ticks, int
out_unlock:
spin_unlock(&rq->lock);
out:
- rebalance_tick(rq, 0);
+ rebalance_tick(cpu, rq, NOT_IDLE);
}
+#ifdef CONFIG_SCHED_SMT
+static inline void wake_sleeping_dependent(int cpu, runqueue_t *rq)
+{
+ int i;
+ struct sched_domain *sd = cpu_sched_domain(cpu);
+ cpumask_t sibling_map;
+
+ if (!(sd->flags & SD_FLAG_SHARE_CPUPOWER)) {
+ /* Not SMT */
+ return;
+ }
+
+ cpus_and(sibling_map, sd->span, cpu_online_map);
+ cpu_clear(cpu, sibling_map);
+ for_each_cpu_mask(i, sibling_map) {
+ runqueue_t *smt_rq;
+
+ smt_rq = cpu_rq(i);
+
+ /*
+ * If an SMT sibling task is sleeping due to priority
+ * reasons wake it up now.
+ */
+ if (smt_rq->curr == smt_rq->idle && smt_rq->nr_running)
+ resched_task(smt_rq->idle);
+ }
+}
+
+static inline int dependent_sleeper(int cpu, runqueue_t *rq, task_t *p)
+{
+ int ret = 0, i;
+ struct sched_domain *sd = cpu_sched_domain(cpu);
+ cpumask_t sibling_map;
+
+ if (!(sd->flags & SD_FLAG_SHARE_CPUPOWER)) {
+ /* Not SMT */
+ return 0;
+ }
+
+ cpus_and(sibling_map, sd->span, cpu_online_map);
+ cpu_clear(cpu, sibling_map);
+ for_each_cpu_mask(i, sibling_map) {
+ runqueue_t *smt_rq;
+ task_t *smt_curr;
+
+ smt_rq = cpu_rq(i);
+ smt_curr = smt_rq->curr;
+
+ /*
+ * If a user task with lower static priority than the
+ * running task on the SMT sibling is trying to schedule,
+ * delay it till there is proportionately less timeslice
+ * left of the sibling task to prevent a lower priority
+ * task from using an unfair proportion of the
+ * physical cpu's resources. -ck
+ */
+ if (((smt_curr->time_slice * (100 - sd->per_cpu_gain) / 100) >
+ task_timeslice(p) || rt_task(smt_curr)) &&
+ p->mm && smt_curr->mm && !rt_task(p))
+ ret |= 1;
+
+ /*
+ * Reschedule a lower priority task on the SMT sibling,
+ * or wake it up if it has been put to sleep for priority
+ * reasons.
+ */
+ if ((((p->time_slice * (100 - sd->per_cpu_gain) / 100) >
+ task_timeslice(smt_curr) || rt_task(p)) &&
+ smt_curr->mm && p->mm && !rt_task(smt_curr)) ||
+ (smt_curr == smt_rq->idle && smt_rq->nr_running))
+ resched_task(smt_curr);
+ }
+ return ret;
+}
+#else
+static inline void wake_sleeping_dependent(int cpu, runqueue_t *rq)
+{
+}
+
+static inline int dependent_sleeper(int cpu, runqueue_t *rq, task_t *p)
+{
+ return 0;
+}
+#endif
+
void scheduling_functions_start_here(void) { }
/*
@@ -1603,7 +2026,7 @@ asmlinkage void schedule(void)
struct list_head *queue;
unsigned long long now;
unsigned long run_time;
- int idx;
+ int cpu, idx;
/*
* Test if we are atomic. Since do_exit() needs to call into
@@ -1653,13 +2076,15 @@ need_resched:
deactivate_task(prev, rq);
}
+ cpu = smp_processor_id();
if (unlikely(!rq->nr_running)) {
#ifdef CONFIG_SMP
- load_balance(rq, 1, cpu_to_node_mask(smp_processor_id()));
+ idle_balance(cpu, rq);
#endif
if (!rq->nr_running) {
next = rq->idle;
rq->expired_timestamp = 0;
+ wake_sleeping_dependent(cpu, rq);
goto switch_tasks;
}
}
@@ -1680,6 +2105,11 @@ need_resched:
queue = array->queue + idx;
next = list_entry(queue->next, task_t, run_list);
+ if (dependent_sleeper(cpu, rq, next)) {
+ next = rq->idle;
+ goto switch_tasks;
+ }
+
if (next->activated > 0) {
unsigned long long delta = now - next->timestamp;
@@ -2016,6 +2446,13 @@ out_unlock:
EXPORT_SYMBOL(set_user_nice);
+#if defined( CONFIG_KGDB)
+struct task_struct * kgdb_get_idle(int this_cpu)
+{
+ return cpu_rq(this_cpu)->idle;
+}
+#endif
+
#ifndef __alpha__
/*
@@ -2714,7 +3151,12 @@ int set_cpus_allowed(task_t *p, cpumask_
goto out;
}
- if (__set_cpus_allowed(p, new_mask, &req)) {
+ p->cpus_allowed = new_mask;
+ /* Can the task run on the task's current CPU? If so, we're done */
+ if (cpu_isset(task_cpu(p), new_mask))
+ goto out;
+
+ if (migrate_task(p, any_online_cpu(new_mask), &req)) {
/* Need help from migration thread: drop lock and wait. */
task_rq_unlock(rq, &flags);
wake_up_process(rq->migration_thread);
@@ -2728,8 +3170,16 @@ out:
EXPORT_SYMBOL_GPL(set_cpus_allowed);
-/* Move (not current) task off this cpu, onto dest cpu. */
-static void move_task_away(struct task_struct *p, int dest_cpu)
+/*
+ * Move (not current) task off this cpu, onto dest cpu. We're doing
+ * this because either it can't run here any more (set_cpus_allowed()
+ * away from this CPU, or CPU going down), or because we're
+ * attempting to rebalance this task on exec (sched_balance_exec).
+ *
+ * So we race with normal scheduler movements, but that's OK, as long
+ * as the task is no longer on this CPU.
+ */
+static void __migrate_task(struct task_struct *p, int dest_cpu)
{
runqueue_t *rq_dest;
unsigned long flags;
@@ -2738,14 +3188,18 @@ static void move_task_away(struct task_s
local_irq_save(flags);
double_rq_lock(this_rq(), rq_dest);
+ /* Already moved. */
if (task_cpu(p) != smp_processor_id())
- goto out; /* Already moved */
+ goto out;
+ /* Affinity changed (again). */
+ if (!cpu_isset(dest_cpu, p->cpus_allowed))
+ goto out;
set_task_cpu(p, dest_cpu);
if (p->array) {
deactivate_task(p, this_rq());
activate_task(p, rq_dest);
- if (p->prio < rq_dest->curr->prio)
+ if (TASK_PREEMPTS_CURR(p, rq_dest))
resched_task(rq_dest->curr);
}
p->timestamp = rq_dest->timestamp_last_tick;
@@ -2782,7 +3236,13 @@ static int migration_thread(void * data)
refrigerator(PF_IOTHREAD);
spin_lock_irq(&rq->lock);
+ if (rq->active_balance) {
+ active_load_balance(rq, cpu);
+ rq->active_balance = 0;
+ }
+
head = &rq->migration_queue;
+
current->state = TASK_INTERRUPTIBLE;
if (list_empty(head)) {
spin_unlock_irq(&rq->lock);
@@ -2793,8 +3253,7 @@ static int migration_thread(void * data)
list_del_init(head->next);
spin_unlock_irq(&rq->lock);
- move_task_away(req->task,
- any_online_cpu(req->task->cpus_allowed));
+ __migrate_task(req->task, req->dest_cpu);
complete(&req->done);
}
return 0;
@@ -2861,6 +3320,210 @@ int __init migration_init(void)
spinlock_t kernel_flag __cacheline_aligned_in_smp = SPIN_LOCK_UNLOCKED;
EXPORT_SYMBOL(kernel_flag);
+#ifdef CONFIG_SMP
+#ifdef ARCH_HAS_SCHED_DOMAIN
+extern void __init arch_init_sched_domains(void);
+#else
+static struct sched_group sched_group_cpus[NR_CPUS];
+#ifdef CONFIG_NUMA
+static struct sched_group sched_group_nodes[MAX_NUMNODES];
+DEFINE_PER_CPU(struct sched_domain, node_domains);
+static void __init arch_init_sched_domains(void)
+{
+ int i;
+ struct sched_group *first_node = NULL, *last_node = NULL;
+
+ /* Set up domains */
+ for_each_cpu(i) {
+ int node = cpu_to_node(i);
+ cpumask_t nodemask = node_to_cpumask(node);
+ struct sched_domain *node_domain = &per_cpu(node_domains, i);
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+
+ *node_domain = SD_NODE_INIT;
+ node_domain->span = cpu_possible_map;
+
+ *cpu_domain = SD_CPU_INIT;
+ cpus_and(cpu_domain->span, nodemask, cpu_possible_map);
+ cpu_domain->parent = node_domain;
+ }
+
+ /* Set up groups */
+ for (i = 0; i < MAX_NUMNODES; i++) {
+ struct sched_group *first_cpu = NULL, *last_cpu = NULL;
+ int j;
+ cpumask_t nodemask;
+ struct sched_group *node = &sched_group_nodes[i];
+ cpumask_t tmp = node_to_cpumask(i);
+
+ cpus_and(nodemask, tmp, cpu_possible_map);
+
+ if (cpus_empty(nodemask))
+ continue;
+
+ node->cpumask = nodemask;
+ node->cpu_power = SCHED_LOAD_SCALE * cpus_weight(node->cpumask);
+
+ for_each_cpu_mask(j, node->cpumask) {
+ struct sched_group *cpu = &sched_group_cpus[j];
+
+ cpus_clear(cpu->cpumask);
+ cpu_set(j, cpu->cpumask);
+ cpu->cpu_power = SCHED_LOAD_SCALE;
+
+ if (!first_cpu)
+ first_cpu = cpu;
+ if (last_cpu)
+ last_cpu->next = cpu;
+ last_cpu = cpu;
+ }
+ last_cpu->next = first_cpu;
+
+ if (!first_node)
+ first_node = node;
+ if (last_node)
+ last_node->next = node;
+ last_node = node;
+ }
+ last_node->next = first_node;
+
+ mb();
+ for_each_cpu(i) {
+ struct sched_domain *node_domain = &per_cpu(node_domains, i);
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ node_domain->groups = &sched_group_nodes[cpu_to_node(i)];
+ cpu_domain->groups = &sched_group_cpus[i];
+ }
+}
+
+#else /* CONFIG_NUMA */
+static void __init arch_init_sched_domains(void)
+{
+ int i;
+ struct sched_group *first_cpu = NULL, *last_cpu = NULL;
+
+ /* Set up domains */
+ for_each_cpu(i) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+
+ *cpu_domain = SD_CPU_INIT;
+ cpu_domain->span = cpu_possible_map;
+ }
+
+ /* Set up CPU groups */
+ for_each_cpu_mask(i, cpu_possible_map) {
+ struct sched_group *cpu = &sched_group_cpus[i];
+
+ cpus_clear(cpu->cpumask);
+ cpu_set(i, cpu->cpumask);
+ cpu->cpu_power = SCHED_LOAD_SCALE;
+
+ if (!first_cpu)
+ first_cpu = cpu;
+ if (last_cpu)
+ last_cpu->next = cpu;
+ last_cpu = cpu;
+ }
+ last_cpu->next = first_cpu;
+
+ mb();
+ for_each_cpu(i) {
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+ cpu_domain->groups = &sched_group_cpus[i];
+ }
+}
+
+#endif /* CONFIG_NUMA */
+#endif /* ARCH_HAS_SCHED_DOMAIN */
+
+#undef SCHED_DOMAIN_DEBUG
+#ifdef SCHED_DOMAIN_DEBUG
+void sched_domain_debug(void)
+{
+ int i;
+
+ for_each_cpu(i) {
+ int level = 0;
+ struct sched_domain *cpu_domain = cpu_sched_domain(i);
+
+ printk(KERN_DEBUG "CPU%d: %s\n",
+ i, (cpu_online(i) ? " online" : "offline"));
+
+ do {
+ int j;
+ char str[NR_CPUS];
+ struct sched_group *group = cpu_domain->groups;
+ cpumask_t groupmask, tmp;
+
+ cpumask_snprintf(str, NR_CPUS, cpu_domain->span);
+ cpus_clear(groupmask);
+
+ printk(KERN_DEBUG);
+ for (j = 0; j < level + 1; j++)
+ printk(" ");
+ printk("domain %d: span %s\n", level, str);
+
+ if (!cpu_isset(i, cpu_domain->span))
+ printk(KERN_DEBUG "ERROR domain->span does not contain CPU%d\n", i);
+ if (!cpu_isset(i, group->cpumask))
+ printk(KERN_DEBUG "ERROR domain->groups does not contain CPU%d\n", i);
+
+ printk(KERN_DEBUG);
+ for (j = 0; j < level + 2; j++)
+ printk(" ");
+ printk("groups:");
+ do {
+ if (group == NULL) {
+ printk(" ERROR: NULL");
+ break;
+ }
+
+ if (cpus_weight(group->cpumask) == 0)
+ printk(" ERROR empty group:");
+
+ cpus_and(tmp, groupmask, group->cpumask);
+ if (cpus_weight(tmp) > 0)
+ printk(" ERROR repeated CPUs:");
+
+ cpus_or(groupmask, groupmask, group->cpumask);
+
+ cpumask_snprintf(str, NR_CPUS, group->cpumask);
+ printk(" %s", str);
+
+ group = group->next;
+ } while (group != cpu_domain->groups);
+ printk("\n");
+
+ if (!cpus_equal(cpu_domain->span, groupmask))
+ printk(KERN_DEBUG "ERROR groups don't span domain->span\n");
+
+ level++;
+ cpu_domain = cpu_domain->parent;
+
+ if (cpu_domain) {
+ cpus_and(tmp, groupmask, cpu_domain->span);
+ if (!cpus_equal(tmp, groupmask))
+ printk(KERN_DEBUG "ERROR parent span is not a superset of domain->span\n");
+ }
+
+ } while (cpu_domain);
+ }
+}
+#else
+#define sched_domain_debug() {}
+#endif
+
+void __init sched_init_smp(void)
+{
+ arch_init_sched_domains();
+ sched_domain_debug();
+}
+#else
+void __init sched_init_smp(void)
+{
+}
+#endif /* CONFIG_SMP */
+
void __init sched_init(void)
{
runqueue_t *rq;
@@ -2868,6 +3531,11 @@ void __init sched_init(void)
for (i = 0; i < NR_CPUS; i++) {
prio_array_t *array;
+#ifdef CONFIG_SMP
+ struct sched_domain *domain;
+ domain = cpu_sched_domain(i);
+ memset(domain, 0, sizeof(struct sched_domain));
+#endif
rq = cpu_rq(i);
rq->active = rq->arrays;
@@ -2877,7 +3545,6 @@ void __init sched_init(void)
spin_lock_init(&rq->lock);
INIT_LIST_HEAD(&rq->migration_queue);
atomic_set(&rq->nr_iowait, 0);
- nr_running_init(rq);
for (j = 0; j < 2; j++) {
array = rq->arrays + j;
--- diff/kernel/signal.c 2004-03-16 11:24:02.000000000 +0000
+++ source/kernel/signal.c 2004-03-16 11:27:48.662789792 +0000
@@ -21,6 +21,7 @@
#include <linux/binfmts.h>
#include <linux/security.h>
#include <linux/ptrace.h>
+#include <linux/compat_siginfo.h>
#include <asm/param.h>
#include <asm/uaccess.h>
#include <asm/siginfo.h>
@@ -588,7 +589,8 @@ static int check_kill_permission(int sig
error = -EPERM;
if ((!info || ((unsigned long)info != 1 &&
(unsigned long)info != 2 && SI_FROMUSER(info)))
- && ((sig != SIGCONT) || (current->session != t->session))
+ && ((sig != SIGCONT) ||
+ (current->signal->session != t->signal->session))
&& (current->euid ^ t->suid) && (current->euid ^ t->uid)
&& (current->uid ^ t->suid) && (current->uid ^ t->uid)
&& !capable(CAP_KILL))
@@ -1103,7 +1105,7 @@ kill_sl_info(int sig, struct siginfo *in
retval = -ESRCH;
read_lock(&tasklist_lock);
for_each_task_pid(sid, PIDTYPE_SID, p, l, pid) {
- if (!p->leader)
+ if (!p->signal->leader)
continue;
err = group_send_sig_info(sig, info, p);
if (retval)
@@ -2009,6 +2011,12 @@ int copy_siginfo_to_user(siginfo_t __use
if (from->si_code < 0)
return __copy_to_user(to, from, sizeof(siginfo_t))
? -EFAULT : 0;
+
+ /* Use compat_siginfo_t with 32-bit signals */
+ if(is_compat_task(current)){
+ return compat_copy_siginfo_to_user((compat_siginfo_t __user *)to,from);
+ }
+
/*
* If you change siginfo_t structure, please be sure
* this code is fixed accordingly.
@@ -2047,6 +2055,7 @@ int copy_siginfo_to_user(siginfo_t __use
err |= __put_user(from->si_stime, &to->si_stime);
break;
case __SI_RT: /* This is not generated by the kernel as of now. */
+ case __SI_MESGQ: /* But this is */
err |= __put_user(from->si_pid, &to->si_pid);
err |= __put_user(from->si_uid, &to->si_uid);
err |= __put_user(from->si_int, &to->si_int);
--- diff/kernel/softirq.c 2004-03-11 10:20:29.000000000 +0000
+++ source/kernel/softirq.c 2004-03-16 11:27:48.662789792 +0000
@@ -16,6 +16,7 @@
#include <linux/cpu.h>
#include <linux/kthread.h>
+#include <asm/irq.h>
/*
- No shared variables, all the data are CPU local.
- If a softirq needs serialization, let it serialize itself
@@ -69,60 +70,84 @@ static inline void wakeup_softirqd(void)
*/
#define MAX_SOFTIRQ_RESTART 10
-asmlinkage void do_softirq(void)
+asmlinkage void __do_softirq(void)
{
- int max_restart = MAX_SOFTIRQ_RESTART;
+ struct softirq_action *h;
__u32 pending;
- unsigned long flags;
+ int max_restart = MAX_SOFTIRQ_RESTART;
- if (in_interrupt())
- return;
+ pending = local_softirq_pending();
- local_irq_save(flags);
+ local_bh_disable();
+restart:
+ /* Reset the pending bitmask before enabling irqs */
+ local_softirq_pending() = 0;
+
+ local_irq_enable();
+
+ h = softirq_vec;
+
+ do {
+ if (pending & 1)
+ h->action(h);
+ h++;
+ pending >>= 1;
+ } while (pending);
+
+ local_irq_disable();
pending = local_softirq_pending();
+ if (pending && --max_restart)
+ goto restart;
- if (pending) {
- struct softirq_action *h;
+ if (pending)
+ wakeup_softirqd();
- local_bh_disable();
-restart:
- /* Reset the pending bitmask before enabling irqs */
- local_softirq_pending() = 0;
+ __local_bh_enable();
+}
- local_irq_enable();
+#ifndef __ARCH_HAS_DO_SOFTIRQ
- h = softirq_vec;
+asmlinkage void do_softirq(void)
+{
+ __u32 pending;
+ unsigned long flags;
- do {
- if (pending & 1)
- h->action(h);
- h++;
- pending >>= 1;
- } while (pending);
+ if (in_interrupt())
+ return;
- local_irq_disable();
+ local_irq_save(flags);
- pending = local_softirq_pending();
- if (pending && --max_restart)
- goto restart;
- if (pending)
- wakeup_softirqd();
- __local_bh_enable();
- }
+ pending = local_softirq_pending();
+
+ if (pending)
+ __do_softirq();
local_irq_restore(flags);
}
EXPORT_SYMBOL(do_softirq);
+#endif
+
void local_bh_enable(void)
{
+ if (in_irq()) {
+ printk("local_bh_enable() was called in hard irq context. "
+ "This is probably a bug\n");
+ dump_stack();
+ }
+
__local_bh_enable();
- WARN_ON(irqs_disabled());
- if (unlikely(!in_interrupt() &&
- local_softirq_pending()))
+ if (unlikely(!in_interrupt() && local_softirq_pending())) {
+ if (irqs_disabled()) {
+ printk("local_bh_enable() was called with local "
+ "interrupts disabled. This is probably a"
+ " bug\n");
+ dump_stack();
+ }
invoke_softirq();
+ }
preempt_check_resched();
}
EXPORT_SYMBOL(local_bh_enable);
--- diff/kernel/sys.c 2004-03-11 10:20:29.000000000 +0000
+++ source/kernel/sys.c 2004-03-16 11:27:48.664789488 +0000
@@ -260,6 +260,12 @@ cond_syscall(sys_msgctl)
cond_syscall(sys_shmget)
cond_syscall(sys_shmdt)
cond_syscall(sys_shmctl)
+cond_syscall(sys_mq_open)
+cond_syscall(sys_mq_unlink)
+cond_syscall(sys_mq_timedsend)
+cond_syscall(sys_mq_timedreceive)
+cond_syscall(sys_mq_notify)
+cond_syscall(sys_mq_getsetattr)
/* arch-specific weak syscall entries */
cond_syscall(sys_pciconfig_read)
@@ -979,7 +985,7 @@ asmlinkage long sys_setpgid(pid_t pid, p
if (p->parent == current || p->real_parent == current) {
err = -EPERM;
- if (p->session != current->session)
+ if (p->signal->session != current->signal->session)
goto out;
err = -EACCES;
if (p->did_exec)
@@ -991,7 +997,7 @@ asmlinkage long sys_setpgid(pid_t pid, p
}
err = -EPERM;
- if (p->leader)
+ if (p->signal->leader)
goto out;
if (pgid != pid) {
@@ -1000,7 +1006,7 @@ asmlinkage long sys_setpgid(pid_t pid, p
struct list_head *l;
for_each_task_pid(pgid, PIDTYPE_PGID, p, l, pid)
- if (p->session == current->session)
+ if (p->signal->session == current->signal->session)
goto ok_pgid;
goto out;
}
@@ -1012,7 +1018,7 @@ ok_pgid:
if (process_group(p) != pgid) {
detach_pid(p, PIDTYPE_PGID);
- p->group_leader->__pgrp = pgid;
+ p->signal->pgrp = pgid;
attach_pid(p, PIDTYPE_PGID, pgid);
}
@@ -1054,7 +1060,7 @@ asmlinkage long sys_getpgrp(void)
asmlinkage long sys_getsid(pid_t pid)
{
if (!pid) {
- return current->session;
+ return current->signal->session;
} else {
int retval;
struct task_struct *p;
@@ -1066,7 +1072,7 @@ asmlinkage long sys_getsid(pid_t pid)
if(p) {
retval = security_task_getsid(p);
if (!retval)
- retval = p->session;
+ retval = p->signal->session;
}
read_unlock(&tasklist_lock);
return retval;
@@ -1087,10 +1093,10 @@ asmlinkage long sys_setsid(void)
if (pid)
goto out;
- current->leader = 1;
+ current->signal->leader = 1;
__set_special_pids(current->pid, current->pid);
- current->tty = NULL;
- current->tty_old_pgrp = 0;
+ current->signal->tty = NULL;
+ current->signal->tty_old_pgrp = 0;
err = process_group(current);
out:
write_unlock_irq(&tasklist_lock);
--- diff/kernel/sysctl.c 2004-03-11 10:20:29.000000000 +0000
+++ source/kernel/sysctl.c 2004-03-16 11:27:48.666789184 +0000
@@ -736,6 +736,26 @@ static ctl_table vm_table[] = {
.strategy = &sysctl_intvec,
.extra1 = &zero,
},
+ {
+ .ctl_name = VM_LAPTOP_MODE,
+ .procname = "laptop_mode",
+ .data = &laptop_mode,
+ .maxlen = sizeof(laptop_mode),
+ .mode = 0644,
+ .proc_handler = &proc_dointvec,
+ .strategy = &sysctl_intvec,
+ .extra1 = &zero,
+ },
+ {
+ .ctl_name = VM_BLOCK_DUMP,
+ .procname = "block_dump",
+ .data = &block_dump,
+ .maxlen = sizeof(block_dump),
+ .mode = 0644,
+ .proc_handler = &proc_dointvec,
+ .strategy = &sysctl_intvec,
+ .extra1 = &zero,
+ },
{ .ctl_name = 0 }
};
--- diff/kernel/timer.c 2004-03-16 11:24:02.000000000 +0000
+++ source/kernel/timer.c 2004-03-16 11:27:48.667789032 +0000
@@ -970,6 +970,13 @@ static void process_timeout(unsigned lon
wake_up_process((task_t *)__data);
}
+static void futex_timeout(unsigned long __data)
+{
+ current->flags |= PF_FUTEX_DEBUG;
+ wake_up_process((task_t *)__data);
+ current->flags &= ~PF_FUTEX_DEBUG;
+}
+
/**
* schedule_timeout - sleep until timeout
* @timeout: timeout value in jiffies
@@ -1036,7 +1043,10 @@ fastcall signed long schedule_timeout(si
init_timer(&timer);
timer.expires = expire;
timer.data = (unsigned long) current;
- timer.function = process_timeout;
+ if (current->flags & PF_FUTEX_DEBUG)
+ timer.function = futex_timeout;
+ else
+ timer.function = process_timeout;
add_timer(&timer);
schedule();
--- diff/kernel/workqueue.c 2004-03-16 11:24:02.000000000 +0000
+++ source/kernel/workqueue.c 2004-03-16 11:27:48.668788880 +0000
@@ -47,6 +47,7 @@ struct cpu_workqueue_struct {
struct workqueue_struct *wq;
task_t *thread;
+ int run_depth; /* Detect run_workqueue() recursion depth */
} ____cacheline_aligned;
/*
@@ -78,18 +79,29 @@ static void __queue_work(struct cpu_work
* We queue the work to the CPU it was submitted, but there is no
* guarantee that it will be processed by that CPU.
*/
-int fastcall queue_work(struct workqueue_struct *wq, struct work_struct *work)
+int queue_work_on_cpu(struct workqueue_struct *wq,
+ struct work_struct *work, int cpu)
{
- int ret = 0, cpu = get_cpu();
+ int ret = 0;
if (!test_and_set_bit(0, &work->pending)) {
BUG_ON(!list_empty(&work->entry));
__queue_work(wq->cpu_wq + cpu, work);
ret = 1;
}
+ return ret;
+}
+EXPORT_SYMBOL_GPL(queue_work);
+
+int queue_work(struct workqueue_struct *wq, struct work_struct *work)
+{
+ int ret;
+
+ ret = queue_work_on_cpu(wq, work, get_cpu());
put_cpu();
return ret;
}
+EXPORT_SYMBOL_GPL(queue_work_on_cpu);
static void delayed_work_timer_fn(unsigned long __data)
{
@@ -129,6 +141,13 @@ static inline void run_workqueue(struct
* done.
*/
spin_lock_irqsave(&cwq->lock, flags);
+ cwq->run_depth++;
+ if (cwq->run_depth > 3) {
+ /* morton gets to eat his hat */
+ printk("%s: recursion depth exceeded: %d\n",
+ __FUNCTION__, cwq->run_depth);
+ dump_stack();
+ }
while (!list_empty(&cwq->worklist)) {
struct work_struct *work = list_entry(cwq->worklist.next,
struct work_struct, entry);
@@ -146,6 +165,7 @@ static inline void run_workqueue(struct
cwq->remove_sequence++;
wake_up(&cwq->work_done);
}
+ cwq->run_depth--;
spin_unlock_irqrestore(&cwq->lock, flags);
}
@@ -218,6 +238,14 @@ void fastcall flush_workqueue(struct wor
continue;
cwq = wq->cpu_wq + cpu;
+ if (cwq->thread == current) {
+ /*
+ * Probably keventd trying to flush its own queue.
+ * So simply run it by hand rather than deadlocking.
+ */
+ run_workqueue(cwq);
+ continue;
+ }
spin_lock_irq(&cwq->lock);
sequence_needed = cwq->insert_sequence;
@@ -267,6 +295,7 @@ struct workqueue_struct *create_workqueu
wq = kmalloc(sizeof(*wq), GFP_KERNEL);
if (!wq)
return NULL;
+ memset(wq, 0, sizeof(*wq));
for (cpu = 0; cpu < NR_CPUS; cpu++) {
if (!cpu_online(cpu))
@@ -353,7 +382,6 @@ void init_workqueues(void)
}
EXPORT_SYMBOL_GPL(create_workqueue);
-EXPORT_SYMBOL_GPL(queue_work);
EXPORT_SYMBOL_GPL(queue_delayed_work);
EXPORT_SYMBOL_GPL(flush_workqueue);
EXPORT_SYMBOL_GPL(destroy_workqueue);
--- diff/lib/idr.c 2004-03-11 10:20:29.000000000 +0000
+++ source/lib/idr.c 2004-03-16 11:27:48.669788728 +0000
@@ -109,7 +109,7 @@ static kmem_cache_t *idr_layer_cache;
-static inline struct idr_layer *alloc_layer(struct idr *idp)
+static struct idr_layer *alloc_layer(struct idr *idp)
{
struct idr_layer *p;
@@ -123,7 +123,7 @@ static inline struct idr_layer *alloc_la
return(p);
}
-static inline void free_layer(struct idr *idp, struct idr_layer *p)
+static void free_layer(struct idr *idp, struct idr_layer *p)
{
/*
* Depends on the return element being zeroed.
@@ -137,7 +137,7 @@ static inline void free_layer(struct idr
int idr_pre_get(struct idr *idp, unsigned gfp_mask)
{
- while (idp->id_free_cnt < idp->layers + 1) {
+ while (idp->id_free_cnt < IDR_FREE_MAX) {
struct idr_layer *new;
new = kmem_cache_alloc(idr_layer_cache, gfp_mask);
if(new == NULL)
@@ -148,91 +148,125 @@ int idr_pre_get(struct idr *idp, unsigne
}
EXPORT_SYMBOL(idr_pre_get);
-static inline int sub_alloc(struct idr *idp, int shift, void *ptr)
+static int sub_alloc(struct idr *idp, void *ptr, int *starting_id)
{
- int n, v = 0;
- struct idr_layer *p;
- struct idr_layer **pa[MAX_LEVEL];
- struct idr_layer ***paa = &pa[0];
-
- *paa = NULL;
- *++paa = &idp->top;
+ int n, m, sh;
+ struct idr_layer *p, *new;
+ struct idr_layer *pa[MAX_LEVEL];
+ int l, id;
+ long bm;
- /*
- * By keeping each pointer in an array we can do the
- * "after" recursion processing. In this case, that means
- * we can update the upper level bit map.
- */
-
- while (1){
- p = **paa;
- n = ffz(p->bitmap);
- if (shift){
- /*
- * We run around this while until we
- * reach the leaf node...
- */
- if (!p->ary[n]){
- /*
- * If no node, allocate one, AFTER
- * we insure that we will not
- * intrude on the reserved bit field.
- */
- if ((n << shift) >= MAX_ID_BIT)
- return -1;
- p->ary[n] = alloc_layer(idp);
- p->count++;
+ id = *starting_id;
+ p = idp->top;
+ l = idp->layers;
+ pa[l--] = NULL;
+ while (1) {
+ /*
+ * We run around this while until we reach the leaf node...
+ */
+ n = (id >> (IDR_BITS*l)) & IDR_MASK;
+ bm = ~p->bitmap;
+ m = find_next_bit(&bm, IDR_SIZE, n);
+ if (m == IDR_SIZE) {
+ /* no space available go back to previous layer. */
+ l++;
+ id = (id | ((1 << (IDR_BITS*l))-1)) + 1;
+ if (!(p = pa[l])) {
+ *starting_id = id;
+ return -2;
}
- *++paa = &p->ary[n];
- v += (n << shift);
- shift -= IDR_BITS;
- } else {
- /*
- * We have reached the leaf node, plant the
- * users pointer and return the raw id.
- */
- p->ary[n] = (struct idr_layer *)ptr;
- __set_bit(n, &p->bitmap);
- v += n;
+ continue;
+ }
+ if (m != n) {
+ sh = IDR_BITS*l;
+ id = ((id >> sh) ^ n ^ m) << sh;
+ }
+ if (id >= MAX_ID_BIT)
+ return -1;
+ if (l == 0)
+ break;
+ /*
+ * Create the layer below if it is missing.
+ */
+ if (!p->ary[m]) {
+ if (!(new = alloc_layer(idp)))
+ return -1;
+ p->ary[m] = new;
p->count++;
- /*
- * This is the post recursion processing. Once
- * we find a bitmap that is not full we are
- * done
- */
- while (*(paa-1) && (**paa)->bitmap == IDR_FULL){
- n = *paa - &(**(paa-1))->ary[0];
- __set_bit(n, &(**--paa)->bitmap);
- }
- return(v);
}
+ pa[l--] = p;
+ p = p->ary[m];
+ }
+ /*
+ * We have reached the leaf node, plant the
+ * users pointer and return the raw id.
+ */
+ p->ary[m] = (struct idr_layer *)ptr;
+ __set_bit(m, &p->bitmap);
+ p->count++;
+ /*
+ * If this layer is full mark the bit in the layer above
+ * to show that this part of the radix tree is full.
+ * This may complete the layer above and require walking
+ * up the radix tree.
+ */
+ n = id;
+ while (p->bitmap == IDR_FULL) {
+ if (!(p = pa[++l]))
+ break;
+ n = n >> IDR_BITS;
+ __set_bit((n & IDR_MASK), &p->bitmap);
}
+ return(id);
}
-int idr_get_new(struct idr *idp, void *ptr)
+int idr_get_new_above(struct idr *idp, void *ptr, int starting_id)
{
- int v;
+ struct idr_layer *p, *new;
+ int layers, v, id;
- if (idp->id_free_cnt < idp->layers + 1)
- return (-1);
+ id = starting_id;
+build_up:
+ p = idp->top;
+ layers = idp->layers;
+ if (unlikely(!p)) {
+ if (!(p = alloc_layer(idp)))
+ return -1;
+ layers = 1;
+ }
/*
- * Add a new layer if the array is full
+ * Add a new layer to the top of the tree if the requested
+ * id is larger than the currently allocated space.
*/
- if (unlikely(!idp->top || idp->top->bitmap == IDR_FULL)){
- /*
- * This is a bit different than the lower layers because
- * we have one branch already allocated and full.
- */
- struct idr_layer *new = alloc_layer(idp);
- new->ary[0] = idp->top;
- if ( idp->top)
- ++new->count;
- idp->top = new;
- if ( idp->layers++ )
+ while (id >= (1 << (layers*IDR_BITS))) {
+ layers++;
+ if (!p->count)
+ continue;
+ if (!(new = alloc_layer(idp))) {
+ /*
+ * The allocation failed. If we built part of
+ * the structure tear it down.
+ */
+ for (new = p; p && p != idp->top; new = p) {
+ p = p->ary[0];
+ new->ary[0] = 0;
+ new->bitmap = new->count = 0;
+ free_layer(idp, new);
+ }
+ return -1;
+ }
+ new->ary[0] = p;
+ new->count = 1;
+ if (p->bitmap == IDR_FULL)
__set_bit(0, &new->bitmap);
+ p = new;
}
- v = sub_alloc(idp, (idp->layers - 1) * IDR_BITS, ptr);
- if ( likely(v >= 0 )){
+ idp->top = p;
+ idp->layers = layers;
+ v = sub_alloc(idp, ptr, &id);
+ if (v == -2)
+ goto build_up;
+ if ( likely(v >= 0 )) {
idp->count++;
v += (idp->count << MAX_ID_SHIFT);
if ( unlikely( v == -1 ))
@@ -240,10 +274,16 @@ int idr_get_new(struct idr *idp, void *p
}
return(v);
}
+EXPORT_SYMBOL(idr_get_new_above);
+
+int idr_get_new(struct idr *idp, void *ptr)
+{
+ return idr_get_new_above(idp, ptr, 0);
+}
EXPORT_SYMBOL(idr_get_new);
-static inline void sub_remove(struct idr *idp, int shift, int id)
+static void sub_remove(struct idr *idp, int shift, int id)
{
struct idr_layer *p = idp->top;
struct idr_layer **pa[MAX_LEVEL];
--- diff/lib/radix-tree.c 2003-05-21 10:50:10.000000000 +0000
+++ source/lib/radix-tree.c 2004-03-16 11:27:48.671788424 +0000
@@ -6,12 +6,12 @@
* modify it under the terms of the GNU General Public License as
* published by the Free Software Foundation; either version 2, or (at
* your option) any later version.
- *
+ *
* This program is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
- *
+ *
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
@@ -26,21 +26,36 @@
#include <linux/slab.h>
#include <linux/gfp.h>
#include <linux/string.h>
+#include <linux/bitops.h>
/*
* Radix tree node definition.
+ *
+ * RADIX_TREE_MAP_SHIFT must be >= log2(BITS_PER_LONG). Otherwise the tags
+ * array will have zero size and the set_tag() arithmetic will go wrong.
*/
-#define RADIX_TREE_MAP_SHIFT 6
-#define RADIX_TREE_MAP_SIZE (1UL << RADIX_TREE_MAP_SHIFT)
-#define RADIX_TREE_MAP_MASK (RADIX_TREE_MAP_SIZE-1)
+#ifdef __KERNEL__
+#define RADIX_TREE_MAP_SHIFT 6
+#else
+#define RADIX_TREE_MAP_SHIFT 3 /* For more stressful testing */
+#endif
+#define RADIX_TREE_TAGS 2
+
+#define RADIX_TREE_MAP_SIZE (1UL << RADIX_TREE_MAP_SHIFT)
+#define RADIX_TREE_MAP_MASK (RADIX_TREE_MAP_SIZE-1)
+
+#define RADIX_TREE_TAG_LONGS \
+ ((RADIX_TREE_MAP_SIZE + BITS_PER_LONG - 1) / BITS_PER_LONG)
struct radix_tree_node {
unsigned int count;
void *slots[RADIX_TREE_MAP_SIZE];
+ unsigned long tags[RADIX_TREE_TAGS][RADIX_TREE_TAG_LONGS];
};
struct radix_tree_path {
struct radix_tree_node *node, **slot;
+ int offset;
};
#define RADIX_TREE_INDEX_BITS (8 /* CHAR_BIT */ * sizeof(unsigned long))
@@ -122,6 +137,22 @@ out:
return ret;
}
+static inline void tag_set(struct radix_tree_node *node, int tag, int offset)
+{
+ if (!test_bit(offset, &node->tags[tag][0]))
+ __set_bit(offset, &node->tags[tag][0]);
+}
+
+static inline void tag_clear(struct radix_tree_node *node, int tag, int offset)
+{
+ __clear_bit(offset, &node->tags[tag][0]);
+}
+
+static inline int tag_get(struct radix_tree_node *node, int tag, int offset)
+{
+ return test_bit(offset, &node->tags[tag][0]);
+}
+
/*
* Return the maximum key which can be store into a
* radix tree with height HEIGHT.
@@ -138,26 +169,53 @@ static int radix_tree_extend(struct radi
{
struct radix_tree_node *node;
unsigned int height;
+ char tags[RADIX_TREE_TAGS];
+ int tag;
/* Figure out what the height should be. */
height = root->height + 1;
while (index > radix_tree_maxindex(height))
height++;
- if (root->rnode) {
- do {
- if (!(node = radix_tree_node_alloc(root)))
- return -ENOMEM;
-
- /* Increase the height. */
- node->slots[0] = root->rnode;
- node->count = 1;
- root->rnode = node;
- root->height++;
- } while (height > root->height);
- } else
+ if (root->rnode == NULL) {
root->height = height;
+ goto out;
+ }
+ /*
+ * Prepare the tag status of the top-level node for propagation
+ * into the newly-pushed top-level node(s)
+ */
+ for (tag = 0; tag < RADIX_TREE_TAGS; tag++) {
+ int idx;
+
+ tags[tag] = 0;
+ for (idx = 0; idx < RADIX_TREE_TAG_LONGS; idx++) {
+ if (root->rnode->tags[tag][idx]) {
+ tags[tag] = 1;
+ break;
+ }
+ }
+ }
+
+ do {
+ if (!(node = radix_tree_node_alloc(root)))
+ return -ENOMEM;
+
+ /* Increase the height. */
+ node->slots[0] = root->rnode;
+
+ /* Propagate the aggregated tag info into the new root */
+ for (tag = 0; tag < RADIX_TREE_TAGS; tag++) {
+ if (tags[tag])
+ tag_set(node, tag, 0);
+ }
+
+ node->count = 1;
+ root->rnode = node;
+ root->height++;
+ } while (height > root->height);
+out:
return 0;
}
@@ -169,23 +227,27 @@ static int radix_tree_extend(struct radi
*
* Insert an item into the radix tree at position @index.
*/
-int radix_tree_insert(struct radix_tree_root *root, unsigned long index, void *item)
+int radix_tree_insert(struct radix_tree_root *root,
+ unsigned long index, void *item)
{
struct radix_tree_node *node = NULL, *tmp, **slot;
unsigned int height, shift;
+ int offset;
int error;
/* Make sure the tree is high enough. */
- if (index > radix_tree_maxindex(root->height)) {
+ if ((!index && !root->rnode) ||
+ index > radix_tree_maxindex(root->height)) {
error = radix_tree_extend(root, index);
if (error)
return error;
}
-
+
slot = &root->rnode;
height = root->height;
shift = (height-1) * RADIX_TREE_MAP_SHIFT;
+ offset = 0; /* uninitialised var warning */
while (height > 0) {
if (*slot == NULL) {
/* Have to add a child node. */
@@ -196,18 +258,21 @@ int radix_tree_insert(struct radix_tree_
node->count++;
}
- /* Go a level down. */
+ /* Go a level down */
+ offset = (index >> shift) & RADIX_TREE_MAP_MASK;
node = *slot;
- slot = (struct radix_tree_node **)
- (node->slots + ((index >> shift) & RADIX_TREE_MAP_MASK));
+ slot = (struct radix_tree_node **)(node->slots + offset);
shift -= RADIX_TREE_MAP_SHIFT;
height--;
}
if (*slot != NULL)
return -EEXIST;
- if (node)
+ if (node) {
node->count++;
+ BUG_ON(tag_get(node, 0, offset));
+ BUG_ON(tag_get(node, 1, offset));
+ }
*slot = item;
return 0;
@@ -219,7 +284,7 @@ EXPORT_SYMBOL(radix_tree_insert);
* @root: radix tree root
* @index: index key
*
- * Lookup them item at the position @index in the radix tree @root.
+ * Lookup the item at the position @index in the radix tree @root.
*/
void *radix_tree_lookup(struct radix_tree_root *root, unsigned long index)
{
@@ -238,16 +303,174 @@ void *radix_tree_lookup(struct radix_tre
return NULL;
slot = (struct radix_tree_node **)
- ((*slot)->slots + ((index >> shift) & RADIX_TREE_MAP_MASK));
+ ((*slot)->slots +
+ ((index >> shift) & RADIX_TREE_MAP_MASK));
shift -= RADIX_TREE_MAP_SHIFT;
height--;
}
- return (void *) *slot;
+ return *slot;
}
EXPORT_SYMBOL(radix_tree_lookup);
-static /* inline */ unsigned int
+/**
+ * radix_tree_tag_set - set a tag on a radix tree node
+ * @root: radix tree root
+ * @index: index key
+ * @tag: tag index
+ *
+ * Set the search tag corresponging to @index in the radix tree. From
+ * the root all the way down to the leaf node.
+ *
+ * Returns the address of the tagged item. Setting a tag on a not-present
+ * item is a bug.
+ */
+void *radix_tree_tag_set(struct radix_tree_root *root,
+ unsigned long index, int tag)
+{
+ unsigned int height, shift;
+ struct radix_tree_node **slot;
+
+ height = root->height;
+ if (index > radix_tree_maxindex(height))
+ return NULL;
+
+ shift = (height - 1) * RADIX_TREE_MAP_SHIFT;
+ slot = &root->rnode;
+
+ while (height > 0) {
+ int offset;
+
+ offset = (index >> shift) & RADIX_TREE_MAP_MASK;
+ tag_set(*slot, tag, offset);
+ slot = (struct radix_tree_node **)((*slot)->slots + offset);
+ BUG_ON(*slot == NULL);
+ shift -= RADIX_TREE_MAP_SHIFT;
+ height--;
+ }
+
+ return *slot;
+}
+EXPORT_SYMBOL(radix_tree_tag_set);
+
+/**
+ * radix_tree_tag_clear - clear a tag on a radix tree node
+ * @root: radix tree root
+ * @index: index key
+ * @tag: tag index
+ *
+ * Clear the search tag corresponging to @index in the radix tree. If
+ * this causes the leaf node to have no tags set then clear the tag in the
+ * next-to-leaf node, etc.
+ *
+ * Returns the address of the tagged item on success, else NULL. ie:
+ * has the same return value and semantics as radix_tree_lookup().
+ */
+void *radix_tree_tag_clear(struct radix_tree_root *root,
+ unsigned long index, int tag)
+{
+ struct radix_tree_path path[RADIX_TREE_MAX_PATH], *pathp = path;
+ unsigned int height, shift;
+ void *ret = NULL;
+
+ height = root->height;
+ if (index > radix_tree_maxindex(height))
+ goto out;
+
+ shift = (height - 1) * RADIX_TREE_MAP_SHIFT;
+ pathp->node = NULL;
+ pathp->slot = &root->rnode;
+
+ while (height > 0) {
+ int offset;
+
+ if (*pathp->slot == NULL)
+ goto out;
+
+ offset = (index >> shift) & RADIX_TREE_MAP_MASK;
+ pathp[1].offset = offset;
+ pathp[1].node = *pathp[0].slot;
+ pathp[1].slot = (struct radix_tree_node **)
+ (pathp[1].node->slots + offset);
+ pathp++;
+ shift -= RADIX_TREE_MAP_SHIFT;
+ height--;
+ }
+
+ ret = *pathp[0].slot;
+ if (ret == NULL)
+ goto out;
+
+ do {
+ int idx;
+
+ tag_clear(pathp[0].node, tag, pathp[0].offset);
+ for (idx = 0; idx < RADIX_TREE_TAG_LONGS; idx++) {
+ if (pathp[0].node->tags[tag][idx])
+ goto out;
+ }
+ pathp--;
+ } while (pathp[0].node);
+out:
+ return ret;
+}
+EXPORT_SYMBOL(radix_tree_tag_clear);
+
+#ifndef __KERNEL__ /* Only the test harness uses this at present */
+/**
+ * radix_tree_tag_get - get a tag on a radix tree node
+ * @root: radix tree root
+ * @index: index key
+ * @tag: tag index
+ *
+ * Return the search tag corresponging to @index in the radix tree.
+ *
+ * Returns zero if the tag is unset, or if there is no corresponding item
+ * in the tree.
+ */
+int radix_tree_tag_get(struct radix_tree_root *root,
+ unsigned long index, int tag)
+{
+ unsigned int height, shift;
+ struct radix_tree_node **slot;
+ int saw_unset_tag = 0;
+
+ height = root->height;
+ if (index > radix_tree_maxindex(height))
+ return 0;
+
+ shift = (height - 1) * RADIX_TREE_MAP_SHIFT;
+ slot = &root->rnode;
+
+ for ( ; ; ) {
+ int offset;
+
+ if (*slot == NULL)
+ return 0;
+
+ offset = (index >> shift) & RADIX_TREE_MAP_MASK;
+
+ /*
+ * This is just a debug check. Later, we can bale as soon as
+ * we see an unset tag.
+ */
+ if (!tag_get(*slot, tag, offset))
+ saw_unset_tag = 1;
+ if (height == 1) {
+ int ret = tag_get(*slot, tag, offset);
+
+ BUG_ON(ret && saw_unset_tag);
+ return ret;
+ }
+ slot = (struct radix_tree_node **)((*slot)->slots + offset);
+ shift -= RADIX_TREE_MAP_SHIFT;
+ height--;
+ }
+}
+EXPORT_SYMBOL(radix_tree_tag_get);
+#endif
+
+static unsigned int
__lookup(struct radix_tree_root *root, void **results, unsigned long index,
unsigned int max_items, unsigned long *next_index)
{
@@ -314,17 +537,6 @@ radix_tree_gang_lookup(struct radix_tree
unsigned long cur_index = first_index;
unsigned int ret = 0;
- if (root->rnode == NULL)
- goto out;
- if (max_index == 0) { /* Bah. Special case */
- if (first_index == 0) {
- if (max_items > 0) {
- *results = root->rnode;
- ret = 1;
- }
- }
- goto out;
- }
while (ret < max_items) {
unsigned int nr_found;
unsigned long next_index; /* Index of next search */
@@ -338,11 +550,101 @@ radix_tree_gang_lookup(struct radix_tree
break;
cur_index = next_index;
}
-out:
return ret;
}
EXPORT_SYMBOL(radix_tree_gang_lookup);
+/*
+ * FIXME: the two tag_get()s here should use find_next_bit() instead of
+ * open-coding the search.
+ */
+static unsigned int
+__lookup_tag(struct radix_tree_root *root, void **results, unsigned long index,
+ unsigned int max_items, unsigned long *next_index, int tag)
+{
+ unsigned int nr_found = 0;
+ unsigned int shift;
+ unsigned int height = root->height;
+ struct radix_tree_node *slot;
+
+ shift = (height - 1) * RADIX_TREE_MAP_SHIFT;
+ slot = root->rnode;
+
+ while (height > 0) {
+ unsigned long i = (index >> shift) & RADIX_TREE_MAP_MASK;
+
+ for ( ; i < RADIX_TREE_MAP_SIZE; i++) {
+ if (tag_get(slot, tag, i)) {
+ BUG_ON(slot->slots[i] == NULL);
+ break;
+ }
+ index &= ~((1 << shift) - 1);
+ index += 1 << shift;
+ if (index == 0)
+ goto out; /* 32-bit wraparound */
+ }
+ if (i == RADIX_TREE_MAP_SIZE)
+ goto out;
+ height--;
+ if (height == 0) { /* Bottom level: grab some items */
+ unsigned long j = index & RADIX_TREE_MAP_MASK;
+
+ for ( ; j < RADIX_TREE_MAP_SIZE; j++) {
+ index++;
+ if (tag_get(slot, tag, j)) {
+ BUG_ON(slot->slots[j] == NULL);
+ results[nr_found++] = slot->slots[j];
+ if (nr_found == max_items)
+ goto out;
+ }
+ }
+ }
+ shift -= RADIX_TREE_MAP_SHIFT;
+ slot = slot->slots[i];
+ }
+out:
+ *next_index = index;
+ return nr_found;
+}
+
+/**
+ * radix_tree_gang_lookup_tag - perform multiple lookup on a radix tree
+ * based on a tag
+ * @root: radix tree root
+ * @results: where the results of the lookup are placed
+ * @first_index: start the lookup from this key
+ * @max_items: place up to this many items at *results
+ * @tag: the tag index
+ *
+ * Performs an index-ascending scan of the tree for present items which
+ * have the tag indexed by @tag set. Places the items at *@results and
+ * returns the number of items which were placed at *@results.
+ */
+unsigned int
+radix_tree_gang_lookup_tag(struct radix_tree_root *root, void **results,
+ unsigned long first_index, unsigned int max_items, int tag)
+{
+ const unsigned long max_index = radix_tree_maxindex(root->height);
+ unsigned long cur_index = first_index;
+ unsigned int ret = 0;
+
+ while (ret < max_items) {
+ unsigned int nr_found;
+ unsigned long next_index; /* Index of next search */
+
+ if (cur_index > max_index)
+ break;
+ nr_found = __lookup_tag(root, results + ret, cur_index,
+ max_items - ret, &next_index, tag);
+ ret += nr_found;
+ if (next_index == 0)
+ break;
+ cur_index = next_index;
+ }
+ return ret;
+}
+EXPORT_SYMBOL(radix_tree_gang_lookup_tag);
+
/**
* radix_tree_delete - delete an item from a radix tree
* @root: radix tree root
@@ -355,24 +657,31 @@ EXPORT_SYMBOL(radix_tree_gang_lookup);
void *radix_tree_delete(struct radix_tree_root *root, unsigned long index)
{
struct radix_tree_path path[RADIX_TREE_MAX_PATH], *pathp = path;
+ struct radix_tree_path *orig_pathp;
unsigned int height, shift;
void *ret = NULL;
+ char tags[RADIX_TREE_TAGS];
+ int nr_cleared_tags;
height = root->height;
if (index > radix_tree_maxindex(height))
goto out;
- shift = (height-1) * RADIX_TREE_MAP_SHIFT;
+ shift = (height - 1) * RADIX_TREE_MAP_SHIFT;
pathp->node = NULL;
pathp->slot = &root->rnode;
while (height > 0) {
+ int offset;
+
if (*pathp->slot == NULL)
goto out;
+ offset = (index >> shift) & RADIX_TREE_MAP_MASK;
+ pathp[1].offset = offset;
pathp[1].node = *pathp[0].slot;
pathp[1].slot = (struct radix_tree_node **)
- (pathp[1].node->slots + ((index >> shift) & RADIX_TREE_MAP_MASK));
+ (pathp[1].node->slots + offset);
pathp++;
shift -= RADIX_TREE_MAP_SHIFT;
height--;
@@ -382,20 +691,67 @@ void *radix_tree_delete(struct radix_tre
if (ret == NULL)
goto out;
+ orig_pathp = pathp;
+
+ /*
+ * Clear all tags associated with the just-deleted item
+ */
+ memset(tags, 0, sizeof(tags));
+ do {
+ int tag;
+
+ nr_cleared_tags = RADIX_TREE_TAGS;
+ for (tag = 0; tag < RADIX_TREE_TAGS; tag++) {
+ int idx;
+
+ if (!tags[tag])
+ tag_clear(pathp[0].node, tag, pathp[0].offset);
+
+ for (idx = 0; idx < RADIX_TREE_TAG_LONGS; idx++) {
+ if (pathp[0].node->tags[tag][idx]) {
+ tags[tag] = 1;
+ nr_cleared_tags--;
+ break;
+ }
+ }
+ }
+ pathp--;
+ } while (pathp[0].node && nr_cleared_tags);
+
+ pathp = orig_pathp;
*pathp[0].slot = NULL;
while (pathp[0].node && --pathp[0].node->count == 0) {
pathp--;
+ BUG_ON(*pathp[0].slot == NULL);
*pathp[0].slot = NULL;
radix_tree_node_free(pathp[1].node);
}
-
if (root->rnode == NULL)
- root->height = 0; /* Empty tree, we can reset the height */
+ root->height = 0;
out:
return ret;
}
EXPORT_SYMBOL(radix_tree_delete);
+/**
+ * radix_tree_tagged - test whether any items in the tree are tagged
+ * @root: radix tree root
+ * @tag: tag to test
+ */
+int radix_tree_tagged(struct radix_tree_root *root, int tag)
+{
+ int idx;
+
+ if (!root->rnode)
+ return 0;
+ for (idx = 0; idx < RADIX_TREE_TAG_LONGS; idx++) {
+ if (root->rnode->tags[tag][idx])
+ return 1;
+ }
+ return 0;
+}
+EXPORT_SYMBOL(radix_tree_tagged);
+
static void
radix_tree_node_ctor(void *node, kmem_cache_t *cachep, unsigned long flags)
{
--- diff/mm/Makefile 2003-10-09 08:47:17.000000000 +0000
+++ source/mm/Makefile 2004-03-16 11:27:48.677787512 +0000
@@ -12,3 +12,4 @@ obj-y := bootmem.o filemap.o mempool.o
slab.o swap.o truncate.o vmscan.o $(mmu-y)
obj-$(CONFIG_SWAP) += page_io.o swap_state.o swapfile.o
+obj-$(CONFIG_X86_4G) += usercopy.o
--- diff/mm/filemap.c 2004-03-11 10:20:29.000000000 +0000
+++ source/mm/filemap.c 2004-03-16 11:27:48.676787664 +0000
@@ -39,6 +39,7 @@
#include <asm/uaccess.h>
#include <asm/mman.h>
+#include <asm/tlbflush.h>
/*
* Shared mappings implemented 30.11.1994. It's not fully working yet,
@@ -59,7 +60,7 @@
* ->private_lock (__free_pte->__set_page_dirty_buffers)
* ->swap_list_lock
* ->swap_device_lock (exclusive_swap_page, others)
- * ->mapping->page_lock
+ * ->mapping->tree_lock
*
* ->i_sem
* ->i_shared_sem (truncate->invalidate_mmap_range)
@@ -73,14 +74,17 @@
* ->mmap_sem
* ->i_sem (msync)
*
+ * ->i_sem
+ * ->i_alloc_sem (various)
+ *
* ->inode_lock
* ->sb_lock (fs/fs-writeback.c)
- * ->mapping->page_lock (__sync_single_inode)
+ * ->mapping->tree_lock (__sync_single_inode)
*
* ->page_table_lock
* ->swap_device_lock (try_to_unmap_one)
* ->private_lock (try_to_unmap_one)
- * ->page_lock (try_to_unmap_one)
+ * ->tree_lock (try_to_unmap_one)
* ->zone.lru_lock (follow_page->mark_page_accessed)
*
* ->task->proc_lock
@@ -90,16 +94,14 @@
/*
* Remove a page from the page cache and free it. Caller has to make
* sure the page is locked and that nobody else uses it - or that usage
- * is safe. The caller must hold a write_lock on the mapping's page_lock.
+ * is safe. The caller must hold a write_lock on the mapping's tree_lock.
*/
void __remove_from_page_cache(struct page *page)
{
struct address_space *mapping = page->mapping;
radix_tree_delete(&mapping->page_tree, page->index);
- list_del(&page->list);
page->mapping = NULL;
-
mapping->nrpages--;
pagecache_acct(-1);
}
@@ -111,15 +113,17 @@ void remove_from_page_cache(struct page
if (unlikely(!PageLocked(page)))
PAGE_BUG(page);
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&mapping->tree_lock);
__remove_from_page_cache(page);
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
}
static inline int sync_page(struct page *page)
{
- struct address_space *mapping = page->mapping;
+ struct address_space *mapping;
+ smp_mb();
+ mapping = page->mapping;
if (mapping && mapping->a_ops && mapping->a_ops->sync_page)
return mapping->a_ops->sync_page(page);
return 0;
@@ -145,9 +149,6 @@ static int __filemap_fdatawrite(struct a
if (mapping->backing_dev_info->memory_backed)
return 0;
- spin_lock(&mapping->page_lock);
- list_splice_init(&mapping->dirty_pages, &mapping->io_pages);
- spin_unlock(&mapping->page_lock);
ret = do_writepages(mapping, &wbc);
return ret;
}
@@ -156,7 +157,6 @@ int filemap_fdatawrite(struct address_sp
{
return __filemap_fdatawrite(mapping, WB_SYNC_ALL);
}
-
EXPORT_SYMBOL(filemap_fdatawrite);
/*
@@ -167,55 +167,40 @@ int filemap_flush(struct address_space *
{
return __filemap_fdatawrite(mapping, WB_SYNC_NONE);
}
-
EXPORT_SYMBOL(filemap_flush);
-/**
- * filemap_fdatawait - walk the list of locked pages of the given address
- * space and wait for all of them.
- * @mapping: address space structure to wait for
+/*
+ * Wait for writeback to complete against pages indexed by start->end
+ * inclusive
*/
-int filemap_fdatawait(struct address_space * mapping)
+static int wait_on_page_writeback_range(struct address_space *mapping,
+ pgoff_t start, pgoff_t end)
{
+ struct pagevec pvec;
+ int nr_pages;
int ret = 0;
- int progress;
+ pgoff_t index;
-restart:
- progress = 0;
- spin_lock(&mapping->page_lock);
- while (!list_empty(&mapping->locked_pages)) {
- struct page *page;
-
- page = list_entry(mapping->locked_pages.next,struct page,list);
- list_del(&page->list);
- if (PageDirty(page))
- list_add(&page->list, &mapping->dirty_pages);
- else
- list_add(&page->list, &mapping->clean_pages);
+ if (end > start)
+ return 0;
- if (!PageWriteback(page)) {
- if (++progress > 32) {
- if (need_resched()) {
- spin_unlock(&mapping->page_lock);
- __cond_resched();
- goto restart;
- }
- }
- continue;
+ pagevec_init(&pvec, 0);
+ index = start;
+ while ((nr_pages = pagevec_lookup_tag(&pvec, mapping, &index,
+ PAGECACHE_TAG_WRITEBACK,
+ min(end - index + 1, (pgoff_t)PAGEVEC_SIZE)))) {
+ unsigned i;
+
+ for (i = 0; i < nr_pages; i++) {
+ struct page *page = pvec.pages[i];
+
+ wait_on_page_writeback(page);
+ if (PageError(page))
+ ret = -EIO;
}
-
- progress = 0;
- page_cache_get(page);
- spin_unlock(&mapping->page_lock);
-
- wait_on_page_writeback(page);
- if (PageError(page))
- ret = -EIO;
-
- page_cache_release(page);
- spin_lock(&mapping->page_lock);
+ pagevec_release(&pvec);
+ cond_resched();
}
- spin_unlock(&mapping->page_lock);
/* Check for outstanding write errors */
if (test_and_clear_bit(AS_ENOSPC, &mapping->flags))
@@ -226,8 +211,31 @@ restart:
return ret;
}
+/**
+ * filemap_fdatawait - walk the list of under-writeback pages of the given
+ * address space and wait for all of them.
+ *
+ * @mapping: address space structure to wait for
+ */
+int filemap_fdatawait(struct address_space *mapping)
+{
+ return wait_on_page_writeback_range(mapping, 0, -1);
+}
+
EXPORT_SYMBOL(filemap_fdatawait);
+int filemap_write_and_wait(struct address_space *mapping)
+{
+ int retval = 0;
+
+ if (mapping->nrpages) {
+ retval = filemap_fdatawrite(mapping);
+ if (retval == 0)
+ retval = filemap_fdatawait(mapping);
+ }
+ return retval;
+}
+
/*
* This adds a page to the page cache, starting out as locked, unreferenced,
* not uptodate and with no errors.
@@ -252,7 +260,7 @@ int add_to_page_cache(struct page *page,
if (error == 0) {
page_cache_get(page);
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&mapping->tree_lock);
error = radix_tree_insert(&mapping->page_tree, offset, page);
if (!error) {
SetPageLocked(page);
@@ -260,7 +268,7 @@ int add_to_page_cache(struct page *page,
} else {
page_cache_release(page);
}
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
radix_tree_preload_end();
}
return error;
@@ -348,8 +356,7 @@ void end_page_writeback(struct page *pag
wait_queue_head_t *waitqueue = page_waitqueue(page);
if (!TestClearPageReclaim(page) || rotate_reclaimable_page(page)) {
- smp_mb__before_clear_bit();
- if (!TestClearPageWriteback(page))
+ if (!test_clear_page_writeback(page))
BUG();
smp_mb__after_clear_bit();
}
@@ -396,11 +403,11 @@ struct page * find_get_page(struct addre
* We scan the hash list read-only. Addition to and removal from
* the hash-list needs a held write-lock.
*/
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&mapping->tree_lock);
page = radix_tree_lookup(&mapping->page_tree, offset);
if (page)
page_cache_get(page);
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
return page;
}
@@ -413,11 +420,11 @@ struct page *find_trylock_page(struct ad
{
struct page *page;
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&mapping->tree_lock);
page = radix_tree_lookup(&mapping->page_tree, offset);
if (page && TestSetPageLocked(page))
page = NULL;
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
return page;
}
@@ -439,15 +446,15 @@ struct page *find_lock_page(struct addre
{
struct page *page;
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&mapping->tree_lock);
repeat:
page = radix_tree_lookup(&mapping->page_tree, offset);
if (page) {
page_cache_get(page);
if (TestSetPageLocked(page)) {
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
lock_page(page);
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&mapping->tree_lock);
/* Has the page been truncated while we slept? */
if (page->mapping != mapping || page->index != offset) {
@@ -457,7 +464,7 @@ repeat:
}
}
}
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
return page;
}
@@ -525,18 +532,39 @@ EXPORT_SYMBOL(find_or_create_page);
*
* find_get_pages() returns the number of pages which were found.
*/
-unsigned int find_get_pages(struct address_space *mapping, pgoff_t start,
+unsigned find_get_pages(struct address_space *mapping, pgoff_t start,
unsigned int nr_pages, struct page **pages)
{
unsigned int i;
unsigned int ret;
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&mapping->tree_lock);
ret = radix_tree_gang_lookup(&mapping->page_tree,
(void **)pages, start, nr_pages);
for (i = 0; i < ret; i++)
page_cache_get(pages[i]);
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
+ return ret;
+}
+
+/*
+ * Like find_get_pages, except we only return pages which are tagged with
+ * `tag'. We update *start to index the next page for the traversal.
+ */
+unsigned find_get_pages_tag(struct address_space *mapping, pgoff_t *index,
+ int tag, unsigned int nr_pages, struct page **pages)
+{
+ unsigned int i;
+ unsigned int ret;
+
+ spin_lock_irq(&mapping->tree_lock);
+ ret = radix_tree_gang_lookup_tag(&mapping->page_tree,
+ (void **)pages, *index, nr_pages, tag);
+ for (i = 0; i < ret; i++)
+ page_cache_get(pages[i]);
+ if (ret)
+ *index = pages[ret - 1]->index + 1;
+ spin_unlock_irq(&mapping->tree_lock);
return ret;
}
@@ -574,7 +602,7 @@ EXPORT_SYMBOL(grab_cache_page_nowait);
/*
* This is a generic file read routine, and uses the
- * inode->i_op->readpage() function for the actual low-level
+ * mapping->a_ops->readpage() function for the actual low-level
* stuff.
*
* This is really ugly. But the goto's actually try to clarify some
@@ -1200,8 +1228,13 @@ retry_find:
* Ok, found a page in the page cache, now we need to check
* that it's up-to-date.
*/
- if (!PageUptodate(page))
+ if (!PageUptodate(page)) {
+ if (nonblock) {
+ page_cache_release(page);
+ return NULL;
+ }
goto page_not_uptodate;
+ }
success:
/*
@@ -1294,12 +1327,22 @@ static int filemap_populate(struct vm_ar
{
struct file *file = vma->vm_file;
struct address_space *mapping = file->f_mapping;
+ int linear = !(vma->vm_flags & VM_NONLINEAR);
struct inode *inode = mapping->host;
unsigned long size;
struct mm_struct *mm = vma->vm_mm;
struct page *page;
int err;
+ /*
+ * mapping-removal fastpath:
+ */
+ if ((vma->vm_flags & VM_SHARED) &&
+ (pgprot_val(prot) == pgprot_val(PAGE_NONE))) {
+ zap_page_range(vma, addr, len);
+ return 0;
+ }
+
if (!nonblock)
force_page_cache_readahead(mapping, vma->vm_file,
pgoff, len >> PAGE_CACHE_SHIFT);
@@ -1319,19 +1362,9 @@ repeat:
return err;
}
} else {
- /*
- * If a nonlinear mapping then store the file page offset
- * in the pte.
- */
- unsigned long pgidx;
- pgidx = (addr - vma->vm_start) >> PAGE_SHIFT;
- pgidx += vma->vm_pgoff;
- pgidx >>= PAGE_CACHE_SHIFT - PAGE_SHIFT;
- if (pgoff != pgidx) {
- err = install_file_pte(mm, vma, addr, pgoff, prot);
- if (err)
- return err;
- }
+ err = install_file_pte(mm, vma, addr, pgoff, prot, linear);
+ if (err)
+ return err;
}
len -= PAGE_SIZE;
@@ -1715,6 +1748,7 @@ EXPORT_SYMBOL(generic_write_checks);
/*
* Write to a file through the page cache.
+ * Called under i_sem for S_ISREG files.
*
* We put everything into the page cache prior to writing it. This is not a
* problem when writing full pages. With partial pages, however, we first have
@@ -1803,12 +1837,21 @@ generic_file_aio_write_nolock(struct kio
/*
* Sync the fs metadata but not the minor inode changes and
* of course not the data as we did direct DMA for the IO.
+ * i_sem is held, which protects generic_osync_inode() from
+ * livelocking.
*/
if (written >= 0 && file->f_flags & O_SYNC)
status = generic_osync_inode(inode, mapping, OSYNC_METADATA);
- if (written >= 0 && !is_sync_kiocb(iocb))
+ if (written == count && !is_sync_kiocb(iocb))
written = -EIOCBQUEUED;
- goto out_status;
+ if (written < 0 || written == count)
+ goto out_status;
+ /*
+ * direct-io write to a hole: fall through to buffered I/O
+ * for completing the rest of the request.
+ */
+ pos += written;
+ count -= written;
}
buf = iov->iov_base;
@@ -1897,6 +1940,14 @@ generic_file_aio_write_nolock(struct kio
OSYNC_METADATA|OSYNC_DATA);
}
+ /*
+ * If we get here for O_DIRECT writes then we must have fallen through
+ * to buffered writes (block instantiation inside i_size). So we sync
+ * the file data here, to try to honour O_DIRECT expectations.
+ */
+ if (unlikely(file->f_flags & O_DIRECT) && written)
+ status = filemap_write_and_wait(mapping);
+
out_status:
err = written ? written : status;
out:
@@ -1988,6 +2039,9 @@ ssize_t generic_file_writev(struct file
EXPORT_SYMBOL(generic_file_writev);
+/*
+ * Called under i_sem for writes to S_ISREG files
+ */
ssize_t
generic_file_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov,
loff_t offset, unsigned long nr_segs)
@@ -1996,18 +2050,13 @@ generic_file_direct_IO(int rw, struct ki
struct address_space *mapping = file->f_mapping;
ssize_t retval;
- if (mapping->nrpages) {
- retval = filemap_fdatawrite(mapping);
- if (retval == 0)
- retval = filemap_fdatawait(mapping);
- if (retval)
- goto out;
+ retval = filemap_write_and_wait(mapping);
+ if (retval == 0) {
+ retval = mapping->a_ops->direct_IO(rw, iocb, iov,
+ offset, nr_segs);
+ if (rw == WRITE && mapping->nrpages)
+ invalidate_inode_pages2(mapping);
}
-
- retval = mapping->a_ops->direct_IO(rw, iocb, iov, offset, nr_segs);
- if (rw == WRITE && mapping->nrpages)
- invalidate_inode_pages2(mapping);
-out:
return retval;
}
--- diff/mm/fremap.c 2004-03-11 10:20:29.000000000 +0000
+++ source/mm/fremap.c 2004-03-16 11:27:48.677787512 +0000
@@ -38,7 +38,7 @@ static inline void zap_pte(struct mm_str
set_page_dirty(page);
page_remove_rmap(page, ptep);
page_cache_release(page);
- mm->rss--;
+ dec_rss(mm, page);
}
}
} else {
@@ -53,7 +53,7 @@ static inline void zap_pte(struct mm_str
* previously existing mapping.
*/
int install_page(struct mm_struct *mm, struct vm_area_struct *vma,
- unsigned long addr, struct page *page, pgprot_t prot)
+ unsigned long addr, struct page *page, pgprot_t pgprot)
{
int err = -ENOMEM;
pte_t *pte;
@@ -75,12 +75,19 @@ int install_page(struct mm_struct *mm, s
pte = pte_alloc_map(mm, pmd, addr);
if (!pte)
goto err_unlock;
+ /*
+ * Only install a new page for a non-shared mapping if it's
+ * not existent yet:
+ */
+ err = -EEXIST;
+ if (!pte_none(*pte) && !(vma->vm_flags & VM_SHARED))
+ goto err_unlock;
zap_pte(mm, vma, addr, pte);
- mm->rss++;
+ inc_rss(mm, page);
flush_icache_page(vma, page);
- set_pte(pte, mk_pte(page, prot));
+ set_pte(pte, mk_pte(page, pgprot));
pte_chain = page_add_rmap(page, pte, pte_chain);
pte_val = *pte;
pte_unmap(pte);
@@ -103,7 +110,7 @@ EXPORT_SYMBOL(install_page);
* previously existing mapping.
*/
int install_file_pte(struct mm_struct *mm, struct vm_area_struct *vma,
- unsigned long addr, unsigned long pgoff, pgprot_t prot)
+ unsigned long addr, unsigned long pgoff, pgprot_t pgprot, int linear)
{
int err = -ENOMEM;
pte_t *pte;
@@ -111,6 +118,8 @@ int install_file_pte(struct mm_struct *m
pmd_t *pmd;
pte_t pte_val;
+ BUG_ON(!linear && !(vma->vm_flags & VM_SHARED));
+
pgd = pgd_offset(mm, addr);
spin_lock(&mm->page_table_lock);
@@ -121,10 +130,23 @@ int install_file_pte(struct mm_struct *m
pte = pte_alloc_map(mm, pmd, addr);
if (!pte)
goto err_unlock;
+ /*
+ * Skip linear non-existent ptes:
+ */
+ err = 0;
+ if (linear && pte_none(*pte))
+ goto err_unlock;
+ /*
+ * Only install a new page for a non-shared mapping if it's
+ * not existent yet:
+ */
+ err = -EEXIST;
+ if (!pte_none(*pte) && !(vma->vm_flags & VM_SHARED))
+ goto err_unlock;
zap_pte(mm, vma, addr, pte);
- set_pte(pte, pgoff_to_pte(pgoff));
+ set_pte(pte, pgoff_prot_to_pte(pgoff, pgprot));
pte_val = *pte;
pte_unmap(pte);
update_mmu_cache(vma, addr, pte_val);
@@ -144,27 +166,22 @@ err_unlock:
* @size: size of the remapped virtual memory range
* @prot: new protection bits of the range
* @pgoff: to be mapped page of the backing store file
- * @flags: 0 or MAP_NONBLOCKED - the later will cause no IO.
+ * @flags: bits MAP_INHERIT or MAP_NONBLOCKED - the later will cause no IO.
*
* this syscall works purely via pagetables, so it's the most efficient
* way to map the same (large) file into a given virtual window. Unlike
* mmap()/mremap() it does not create any new vmas. The new mappings are
* also safe across swapout.
- *
- * NOTE: the 'prot' parameter right now is ignored, and the vma's default
- * protection is used. Arbitrary protections might be implemented in the
- * future.
*/
-asmlinkage long sys_remap_file_pages(unsigned long start, unsigned long size,
- unsigned long __prot, unsigned long pgoff, unsigned long flags)
+long __remap_file_pages(struct mm_struct *mm, unsigned long start,
+ unsigned long size, unsigned long prot,
+ unsigned long pgoff, unsigned long flags)
{
- struct mm_struct *mm = current->mm;
+ pgprot_t pgprot = protection_map[calc_vm_prot_bits(prot) | VM_SHARED];
unsigned long end = start + size;
struct vm_area_struct *vma;
int err = -EINVAL;
- if (__prot)
- return err;
/*
* Sanitize the syscall parameters:
*/
@@ -184,37 +201,71 @@ asmlinkage long sys_remap_file_pages(uns
/* We need down_write() to change vma->vm_flags. */
down_write(&mm->mmap_sem);
vma = find_vma(mm, start);
-
/*
- * Make sure the vma is shared, that it supports prefaulting,
- * and that the remapped range is valid and fully within
- * the single existing vma:
- */
- if (vma && (vma->vm_flags & VM_SHARED) &&
- vma->vm_ops && vma->vm_ops->populate &&
- end > start && start >= vma->vm_start &&
- end <= vma->vm_end) {
-
- /* Must set VM_NONLINEAR before any pages are populated. */
- if (pgoff != ((start - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff)
- vma->vm_flags |= VM_NONLINEAR;
-
- /* ->populate can take a long time, so downgrade the lock. */
- downgrade_write(&mm->mmap_sem);
- err = vma->vm_ops->populate(vma, start, size,
- vma->vm_page_prot,
- pgoff, flags & MAP_NONBLOCK);
-
- /*
- * We can't clear VM_NONLINEAR because we'd have to do
- * it after ->populate completes, and that would prevent
- * downgrading the lock. (Locks can't be upgraded).
- */
- up_read(&mm->mmap_sem);
- } else {
- up_write(&mm->mmap_sem);
+ * Make sure the permissions are right, the vma is shared
+ * (or linearly remapped - ie. prefaulted), that it supports
+ * prefaulting, and that the remapped range is valid and fully
+ * within the single existing vma:
+ */
+ if (!vma)
+ goto out_unlock;
+ if (unlikely(flags & MAP_INHERIT))
+ pgprot = vma->vm_page_prot;
+ else {
+ err = -EPERM;
+ if (((prot & PROT_READ) && !(vma->vm_flags & VM_MAYREAD)))
+ goto out_unlock;
+ if (((prot & PROT_WRITE) && !(vma->vm_flags & VM_MAYWRITE)))
+ goto out_unlock;
+ if (((prot & PROT_EXEC) && !(vma->vm_flags & VM_MAYEXEC)))
+ goto out_unlock;
+ }
+
+ if (!vma->vm_ops || !vma->vm_ops->populate || end <= start ||
+ start < vma->vm_start || end > vma->vm_end)
+ goto out_unlock;
+
+ if (pgoff != ((start - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff) {
+ if (!(vma->vm_flags & VM_SHARED))
+ goto out_unlock;
+ vma->vm_flags |= VM_NONLINEAR;
}
+ /*
+ * ->populate can take a long time, so downgrade the lock:
+ */
+ downgrade_write(&mm->mmap_sem);
+ err = vma->vm_ops->populate(vma, start, size,
+ pgprot, pgoff, flags & MAP_NONBLOCK);
+
+ /*
+ * We can't clear VM_NONLINEAR because we'd have to do
+ * it after ->populate completes, and that would prevent
+ * downgrading the lock. (Locks can't be upgraded).
+ */
+ up_read(&mm->mmap_sem);
+ return err;
+
+out_unlock:
+ up_write(&mm->mmap_sem);
return err;
}
+asmlinkage long sys_remap_file_pages(unsigned long start, unsigned long size,
+ unsigned long prot, unsigned long pgoff, unsigned long flags)
+{
+ return __remap_file_pages(current->mm, start, size, prot, pgoff, flags);
+}
+
+/*
+ * sys_remap_file_pages - the old API. Implies MAP_INHERIT.
+ */
+asmlinkage long old_remap_file_pages(unsigned long start, unsigned long size,
+ unsigned long __prot, unsigned long pgoff, unsigned long flags)
+{
+ if (__prot)
+ return -EINVAL;
+
+ return __remap_file_pages(current->mm, start, size, PROT_NONE,
+ pgoff, flags | MAP_INHERIT);
+}
--- diff/mm/memory.c 2004-03-16 11:24:02.000000000 +0000
+++ source/mm/memory.c 2004-03-16 11:27:48.679787208 +0000
@@ -109,7 +109,8 @@ static inline void free_one_pmd(struct m
pte_free_tlb(tlb, page);
}
-static inline void free_one_pgd(struct mmu_gather *tlb, pgd_t * dir)
+static inline void free_one_pgd(struct mmu_gather *tlb, pgd_t * dir,
+ int pgd_idx)
{
int j;
pmd_t * pmd;
@@ -123,8 +124,11 @@ static inline void free_one_pgd(struct m
}
pmd = pmd_offset(dir, 0);
pgd_clear(dir);
- for (j = 0; j < PTRS_PER_PMD ; j++)
+ for (j = 0; j < PTRS_PER_PMD ; j++) {
+ if (pgd_idx * PGDIR_SIZE + j * PMD_SIZE >= TASK_SIZE)
+ break;
free_one_pmd(tlb, pmd+j);
+ }
pmd_free_tlb(tlb, pmd);
}
@@ -137,11 +141,13 @@ static inline void free_one_pgd(struct m
void clear_page_tables(struct mmu_gather *tlb, unsigned long first, int nr)
{
pgd_t * page_dir = tlb->mm->pgd;
+ int pgd_idx = first;
page_dir += first;
do {
- free_one_pgd(tlb, page_dir);
+ free_one_pgd(tlb, page_dir, pgd_idx);
page_dir++;
+ pgd_idx++;
} while (--nr);
}
@@ -328,7 +334,7 @@ skip_copy_pte_range:
pte = pte_mkclean(pte);
pte = pte_mkold(pte);
get_page(page);
- dst->rss++;
+ inc_rss(dst, page);
set_pte(dst_pte, pte);
pte_chain = page_add_rmap(page, dst_pte,
@@ -420,7 +426,14 @@ zap_pte_range(struct mmu_gather *tlb, pm
if (page->mapping && pte_young(pte) &&
!PageSwapCache(page))
mark_page_accessed(page);
- tlb->freed++;
+ /*
+ * While we have the page that is being
+ * freed handy, make sure we decrement
+ * the mm's RSS accordingly. This is
+ * only important for NUMA per-node
+ * RSS accounting.
+ */
+ dec_rss(tlb->mm, page);
page_remove_rmap(page, ptep);
tlb_remove_page(tlb, page);
}
@@ -439,7 +452,7 @@ zap_pmd_range(struct mmu_gather *tlb, pg
unsigned long address, unsigned long size)
{
pmd_t * pmd;
- unsigned long end;
+ unsigned long end, pgd_boundary;
if (pgd_none(*dir))
return;
@@ -450,8 +463,9 @@ zap_pmd_range(struct mmu_gather *tlb, pg
}
pmd = pmd_offset(dir, address);
end = address + size;
- if (end > ((address + PGDIR_SIZE) & PGDIR_MASK))
- end = ((address + PGDIR_SIZE) & PGDIR_MASK);
+ pgd_boundary = ((address + PGDIR_SIZE) & PGDIR_MASK);
+ if (pgd_boundary && (end > pgd_boundary))
+ end = pgd_boundary;
do {
zap_pte_range(tlb, pmd, address, end - address);
address = (address + PMD_SIZE) & PMD_MASK;
@@ -693,6 +707,7 @@ int get_user_pages(struct task_struct *t
struct page **pages, struct vm_area_struct **vmas)
{
int i;
+ int vm_io;
unsigned int flags;
/*
@@ -736,8 +751,10 @@ int get_user_pages(struct task_struct *t
continue;
}
- if (!vma || (pages && (vma->vm_flags & VM_IO))
- || !(flags & vma->vm_flags))
+ if (!vma)
+ return i ? : -EFAULT;
+ vm_io = vma->vm_flags & VM_IO;
+ if ((pages && vm_io) || !(flags & vma->vm_flags))
return i ? : -EFAULT;
if (is_vm_hugetlb_page(vma)) {
@@ -747,8 +764,15 @@ int get_user_pages(struct task_struct *t
}
spin_lock(&mm->page_table_lock);
do {
- struct page *map;
+ struct page *map = NULL;
int lookup_write = write;
+
+ /*
+ * We don't follow pagetables for VM_IO regions - they
+ * may have no pageframes.
+ */
+ if (vm_io)
+ goto no_follow;
while (!(map = follow_page(mm, start, lookup_write))) {
spin_unlock(&mm->page_table_lock);
switch (handle_mm_fault(mm,vma,start,write)) {
@@ -788,6 +812,7 @@ int get_user_pages(struct task_struct *t
if (!PageReserved(pages[i]))
page_cache_get(pages[i]);
}
+no_follow:
if (vmas)
vmas[i] = vma;
i++;
@@ -1068,7 +1093,7 @@ static int do_wp_page(struct mm_struct *
page_table = pte_offset_map(pmd, address);
if (pte_same(*page_table, pte)) {
if (PageReserved(old_page))
- ++mm->rss;
+ inc_rss(mm, new_page);
page_remove_rmap(old_page, page_table);
break_cow(vma, new_page, address, page_table);
pte_chain = page_add_rmap(new_page, page_table, pte_chain);
@@ -1302,7 +1327,7 @@ static int do_swap_page(struct mm_struct
if (vm_swap_full())
remove_exclusive_swap_page(page);
- mm->rss++;
+ inc_rss(mm, page);
pte = mk_pte(page, vma->vm_page_prot);
if (write_access && can_share_swap_page(page))
pte = maybe_mkwrite(pte_mkdirty(pte), vma);
@@ -1371,7 +1396,7 @@ do_anonymous_page(struct mm_struct *mm,
ret = VM_FAULT_MINOR;
goto out;
}
- mm->rss++;
+ inc_rss(mm, page);
entry = maybe_mkwrite(pte_mkdirty(mk_pte(page,
vma->vm_page_prot)),
vma);
@@ -1486,7 +1511,7 @@ retry:
/* Only go through if we didn't race with anybody else... */
if (pte_none(*page_table)) {
if (!PageReserved(new_page))
- ++mm->rss;
+ inc_rss(mm, new_page);
flush_icache_page(vma, new_page);
entry = mk_pte(new_page, vma->vm_page_prot);
if (write_access)
@@ -1523,6 +1548,7 @@ static int do_file_page(struct mm_struct
unsigned long address, int write_access, pte_t *pte, pmd_t *pmd)
{
unsigned long pgoff;
+ pgprot_t pgprot;
int err;
BUG_ON(!vma->vm_ops || !vma->vm_ops->nopage);
@@ -1537,11 +1563,12 @@ static int do_file_page(struct mm_struct
}
pgoff = pte_to_pgoff(*pte);
+ pgprot = pte_to_pgprot(*pte);
pte_unmap(pte);
spin_unlock(&mm->page_table_lock);
- err = vma->vm_ops->populate(vma, address & PAGE_MASK, PAGE_SIZE, vma->vm_page_prot, pgoff, 0);
+ err = vma->vm_ops->populate(vma, address & PAGE_MASK, PAGE_SIZE, pgprot, pgoff, 0);
if (err == -ENOMEM)
return VM_FAULT_OOM;
if (err)
@@ -1590,6 +1617,15 @@ static inline int handle_pte_fault(struc
return do_swap_page(mm, vma, address, pte, pmd, entry, write_access);
}
+ /*
+ * Generate a SIGSEGV if a PROT_NONE page is accessed:
+ */
+ if (pgprot_val(pte_to_pgprot(entry)) == pgprot_val(__P000)) {
+ pte_unmap(pte);
+ spin_unlock(&mm->page_table_lock);
+ return VM_FAULT_SIGSEGV;
+ }
+
if (write_access) {
if (!pte_write(entry))
return do_wp_page(mm, vma, address, pte, pmd, entry);
--- diff/mm/mempool.c 2004-03-16 11:24:02.000000000 +0000
+++ source/mm/mempool.c 2004-03-16 11:27:48.679787208 +0000
@@ -234,8 +234,6 @@ repeat_alloc:
if (!(gfp_mask & __GFP_WAIT))
return NULL;
- blk_run_queues();
-
prepare_to_wait(&pool->wait, &wait, TASK_UNINTERRUPTIBLE);
mb();
if (!pool->curr_nr)
--- diff/mm/mmap.c 2004-03-16 11:24:02.000000000 +0000
+++ source/mm/mmap.c 2004-03-16 11:27:48.680787056 +0000
@@ -708,8 +708,12 @@ out:
}
if (flags & MAP_POPULATE) {
up_write(&mm->mmap_sem);
- sys_remap_file_pages(addr, len, prot,
- pgoff, flags & MAP_NONBLOCK);
+ /*
+ * remap_file_pages() works even if the mapping is private,
+ * in the linearly-mapped case:
+ */
+ __remap_file_pages(mm, addr, len, PROT_NONE, pgoff,
+ MAP_INHERIT | (flags & MAP_NONBLOCK));
down_write(&mm->mmap_sem);
}
return addr;
@@ -1447,7 +1451,7 @@ void exit_mmap(struct mm_struct *mm)
vma = mm->mmap;
mm->mmap = mm->mmap_cache = NULL;
mm->mm_rb = RB_ROOT;
- mm->rss = 0;
+ zero_rss(mm);
mm->total_vm = 0;
mm->locked_vm = 0;
--- diff/mm/page-writeback.c 2004-02-09 10:36:12.000000000 +0000
+++ source/mm/page-writeback.c 2004-03-16 11:27:48.684786448 +0000
@@ -28,6 +28,7 @@
#include <linux/smp.h>
#include <linux/sysctl.h>
#include <linux/cpu.h>
+#include <linux/quotaops.h>
/*
* The maximum number of pages to writeout in a single bdflush/kupdate
@@ -81,6 +82,16 @@ int dirty_writeback_centisecs = 5 * 100;
*/
int dirty_expire_centisecs = 30 * 100;
+/*
+ * Flag that makes the machine dump writes/reads and block dirtyings.
+ */
+int block_dump;
+
+/*
+ * Flag that puts the machine in "laptop mode".
+ */
+int laptop_mode;
+
/* End of sysctl-exported parameters */
@@ -195,7 +206,8 @@ static void balance_dirty_pages(struct a
if (nr_reclaimable + ps.nr_writeback <= dirty_thresh)
dirty_exceeded = 0;
- if (!writeback_in_progress(bdi) && nr_reclaimable > background_thresh)
+ if (!unlikely(laptop_mode) && !writeback_in_progress(bdi) &&
+ nr_reclaimable > background_thresh)
pdflush_operation(background_writeout, 0);
}
@@ -289,7 +301,17 @@ int wakeup_bdflush(long nr_pages)
return pdflush_operation(background_writeout, nr_pages);
}
-static struct timer_list wb_timer;
+
+static void wb_timer_fn(unsigned long unused);
+
+/*
+ * Both timers share the same handler
+ */
+static struct timer_list wb_timer =
+ TIMER_INITIALIZER(wb_timer_fn, 0, 0);
+static struct timer_list laptop_mode_wb_timer =
+ TIMER_INITIALIZER(wb_timer_fn, 0, 0);
+
/*
* Periodic writeback of "old" data.
@@ -328,6 +350,8 @@ static void wb_kupdate(unsigned long arg
oldest_jif = jiffies - (dirty_expire_centisecs * HZ) / 100;
start_jif = jiffies;
next_jif = start_jif + (dirty_writeback_centisecs * HZ) / 100;
+ if (laptop_mode)
+ wbc.older_than_this = NULL;
nr_to_write = ps.nr_dirty + ps.nr_unstable +
(inodes_stat.nr_inodes - inodes_stat.nr_unused);
while (nr_to_write > 0) {
@@ -346,6 +370,11 @@ static void wb_kupdate(unsigned long arg
next_jif = jiffies + HZ;
if (dirty_writeback_centisecs)
mod_timer(&wb_timer, next_jif);
+
+ /*
+ * The timer may have been set as a result of our writes.
+ */
+ del_timer(&laptop_mode_wb_timer);
}
/*
@@ -364,13 +393,31 @@ int dirty_writeback_centisecs_handler(ct
return 0;
}
+/*
+ * We've spun up the disk and we're in laptop mode: schedule writeback
+ * of all dirty data in 5 seconds.
+ *
+ * Laptop mode writeback will be delayed if it has previously been
+ * scheduled to occur within 5 seconds. That way, the writeback will
+ * only be triggered if the system is truly quiet again.
+ */
+void disk_is_spun_up(int postpone_writeback)
+{
+ if (postpone_writeback || !timer_pending(&laptop_mode_wb_timer))
+ mod_timer(&laptop_mode_wb_timer, jiffies + 5 * HZ);
+}
+
+
+/*
+ * Handler for wb_timer and laptop_mode_wb_timer.
+ */
static void wb_timer_fn(unsigned long unused)
{
if (pdflush_operation(wb_kupdate, 0) < 0)
mod_timer(&wb_timer, jiffies + HZ); /* delay 1 second */
-
}
+
/*
* If ratelimit_pages is too high then we can get into dirty-data overload
* if a large number of processes all perform writes at the same time.
@@ -430,11 +477,9 @@ void __init page_writeback_init(void)
vm_dirty_ratio /= 100;
}
- init_timer(&wb_timer);
wb_timer.expires = jiffies + (dirty_writeback_centisecs * HZ) / 100;
- wb_timer.data = 0;
- wb_timer.function = wb_timer_fn;
add_timer(&wb_timer);
+
set_ratelimit();
register_cpu_notifier(&ratelimit_nb);
}
@@ -470,12 +515,8 @@ int write_one_page(struct page *page, in
if (wait)
wait_on_page_writeback(page);
- spin_lock(&mapping->page_lock);
- list_del(&page->list);
if (test_clear_page_dirty(page)) {
- list_add(&page->list, &mapping->locked_pages);
page_cache_get(page);
- spin_unlock(&mapping->page_lock);
ret = mapping->a_ops->writepage(page, &wbc);
if (ret == 0 && wait) {
wait_on_page_writeback(page);
@@ -484,8 +525,6 @@ int write_one_page(struct page *page, in
}
page_cache_release(page);
} else {
- list_add(&page->list, &mapping->clean_pages);
- spin_unlock(&mapping->page_lock);
unlock_page(page);
}
return ret;
@@ -493,9 +532,8 @@ int write_one_page(struct page *page, in
EXPORT_SYMBOL(write_one_page);
/*
- * For address_spaces which do not use buffers. Just set the page's dirty bit
- * and move it to the dirty_pages list. Also perform space reservation if
- * required.
+ * For address_spaces which do not use buffers. Just tag the page as dirty in
+ * its radix tree.
*
* __set_page_dirty_nobuffers() may return -ENOSPC. But if it does, the page
* is still safe, as long as it actually manages to find some blocks at
@@ -513,19 +551,22 @@ int __set_page_dirty_nobuffers(struct pa
struct address_space *mapping = page->mapping;
if (mapping) {
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&mapping->tree_lock);
if (page->mapping) { /* Race with truncate? */
BUG_ON(page->mapping != mapping);
if (!mapping->backing_dev_info->memory_backed)
inc_page_state(nr_dirty);
- list_del(&page->list);
- list_add(&page->list, &mapping->dirty_pages);
+ radix_tree_tag_set(&mapping->page_tree,
+ page->index, PAGECACHE_TAG_DIRTY);
}
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
if (!PageSwapCache(page))
__mark_inode_dirty(mapping->host,
I_DIRTY_PAGES);
}
+ if (unlikely(block_dump))
+ printk("%s(%d): dirtied page\n",
+ current->comm, current->pid);
}
return ret;
}
@@ -558,13 +599,102 @@ EXPORT_SYMBOL(set_page_dirty_lock);
*/
int test_clear_page_dirty(struct page *page)
{
- if (TestClearPageDirty(page)) {
- struct address_space *mapping = page->mapping;
+ struct address_space *mapping = page->mapping;
+ unsigned long flags;
- if (mapping && !mapping->backing_dev_info->memory_backed)
- dec_page_state(nr_dirty);
- return 1;
+ if (mapping) {
+ spin_lock_irqsave(&mapping->tree_lock, flags);
+ if (TestClearPageDirty(page)) {
+ radix_tree_tag_clear(&mapping->page_tree, page->index,
+ PAGECACHE_TAG_DIRTY);
+ spin_unlock_irqrestore(&mapping->tree_lock, flags);
+ if (!mapping->backing_dev_info->memory_backed)
+ dec_page_state(nr_dirty);
+ return 1;
+ }
+ spin_unlock_irqrestore(&mapping->tree_lock, flags);
+ return 0;
}
- return 0;
+ return TestClearPageDirty(page);
}
EXPORT_SYMBOL(test_clear_page_dirty);
+
+/*
+ * Clear a page's dirty flag while ignoring dirty memory accounting
+ */
+int __clear_page_dirty(struct page *page)
+{
+ struct address_space *mapping = page->mapping;
+
+ if (mapping) {
+ unsigned long flags;
+
+ spin_lock_irqsave(&mapping->tree_lock, flags);
+ if (TestClearPageDirty(page)) {
+ radix_tree_tag_clear(&mapping->page_tree, page->index,
+ PAGECACHE_TAG_DIRTY);
+ spin_unlock_irqrestore(&mapping->tree_lock, flags);
+ return 1;
+ }
+ spin_unlock_irqrestore(&mapping->tree_lock, flags);
+ return 0;
+ }
+ return TestClearPageDirty(page);
+}
+
+int test_clear_page_writeback(struct page *page)
+{
+ struct address_space *mapping = page->mapping;
+ int ret;
+
+ if (mapping) {
+ unsigned long flags;
+
+ spin_lock_irqsave(&mapping->tree_lock, flags);
+ ret = TestClearPageWriteback(page);
+ if (ret)
+ radix_tree_tag_clear(&mapping->page_tree, page->index,
+ PAGECACHE_TAG_WRITEBACK);
+ spin_unlock_irqrestore(&mapping->tree_lock, flags);
+ } else {
+ ret = TestClearPageWriteback(page);
+ }
+ return ret;
+}
+
+int test_set_page_writeback(struct page *page)
+{
+ struct address_space *mapping = page->mapping;
+ int ret;
+
+ if (mapping) {
+ unsigned long flags;
+
+ spin_lock_irqsave(&mapping->tree_lock, flags);
+ ret = TestSetPageWriteback(page);
+ if (ret)
+ radix_tree_tag_set(&mapping->page_tree, page->index,
+ PAGECACHE_TAG_WRITEBACK);
+ spin_unlock_irqrestore(&mapping->tree_lock, flags);
+ } else {
+ ret = TestSetPageWriteback(page);
+ }
+ return ret;
+
+}
+
+/*
+ * Return true if any of the pages in the mapping are marged with the
+ * passed tag.
+ */
+int mapping_tagged(struct address_space *mapping, int tag)
+{
+ unsigned long flags;
+ int ret;
+
+ spin_lock_irqsave(&mapping->tree_lock, flags);
+ ret = radix_tree_tagged(&mapping->page_tree, tag);
+ spin_unlock_irqrestore(&mapping->tree_lock, flags);
+ return ret;
+}
+EXPORT_SYMBOL(mapping_tagged);
--- diff/mm/page_alloc.c 2004-03-16 11:24:02.000000000 +0000
+++ source/mm/page_alloc.c 2004-03-16 11:27:48.682786752 +0000
@@ -71,13 +71,14 @@ static int bad_range(struct zone *zone,
static void bad_page(const char *function, struct page *page)
{
- printk("Bad page state at %s (in process '%s', page %p)\n", function, current->comm, page);
- printk("flags:0x%08lx mapping:%p mapped:%d count:%d\n",
+ printk(KERN_EMERG "Bad page state at %s (in process '%s', page %p)\n",
+ function, current->comm, page);
+ printk(KERN_EMERG "flags:0x%08lx mapping:%p mapped:%d count:%d\n",
page->flags, page->mapping,
page_mapped(page), page_count(page));
- printk("Backtrace:\n");
+ printk(KERN_EMERG "Backtrace:\n");
dump_stack();
- printk("Trying to fix it up, but a reboot is needed\n");
+ printk(KERN_EMERG "Trying to fix it up, but a reboot is needed\n");
page->flags &= ~(1 << PG_private |
1 << PG_locked |
1 << PG_lru |
@@ -99,13 +100,13 @@ static void bad_page(const char *functio
*
* The remaining PAGE_SIZE pages are called "tail pages".
*
- * All pages have PG_compound set. All pages have their lru.next pointing at
+ * All pages have PG_compound set. All pages have their ->private pointing at
* the head page (even the head page has this).
*
- * The head page's lru.prev, if non-zero, holds the address of the compound
- * page's put_page() function.
+ * The first tail page's ->mapping, if non-zero, holds the address of the
+ * compound page's put_page() function.
*
- * The order of the allocation is stored in the first tail page's lru.prev.
+ * The order of the allocation is stored in the first tail page's ->index
* This is only for debug at present. This usage means that zero-order pages
* may not be compound.
*/
@@ -114,13 +115,13 @@ static void prep_compound_page(struct pa
int i;
int nr_pages = 1 << order;
- page->lru.prev = NULL;
- page[1].lru.prev = (void *)order;
+ page[1].mapping = 0;
+ page[1].index = order;
for (i = 0; i < nr_pages; i++) {
struct page *p = page + i;
SetPageCompound(p);
- p->lru.next = (void *)page;
+ p->private = (unsigned long)page;
}
}
@@ -129,7 +130,7 @@ static void destroy_compound_page(struct
int i;
int nr_pages = 1 << order;
- if (page[1].lru.prev != (void *)order)
+ if (page[1].index != order)
bad_page(__FUNCTION__, page);
for (i = 0; i < nr_pages; i++) {
@@ -137,7 +138,7 @@ static void destroy_compound_page(struct
if (!PageCompound(p))
bad_page(__FUNCTION__, page);
- if (p->lru.next != (void *)page)
+ if (p->private != (unsigned long)page)
bad_page(__FUNCTION__, page);
ClearPageCompound(p);
}
@@ -199,13 +200,13 @@ static inline void __free_pages_bulk (st
buddy2 = base + page_idx;
BUG_ON(bad_range(zone, buddy1));
BUG_ON(bad_range(zone, buddy2));
- list_del(&buddy1->list);
+ list_del(&buddy1->lru);
mask <<= 1;
area++;
index >>= 1;
page_idx &= mask;
}
- list_add(&(base + page_idx)->list, &area->free_list);
+ list_add(&(base + page_idx)->lru, &area->free_list);
}
static inline void free_pages_check(const char *function, struct page *page)
@@ -253,9 +254,9 @@ free_pages_bulk(struct zone *zone, int c
zone->all_unreclaimable = 0;
zone->pages_scanned = 0;
while (!list_empty(list) && count--) {
- page = list_entry(list->prev, struct page, list);
+ page = list_entry(list->prev, struct page, lru);
/* have to delete it as __free_pages_bulk list manipulates */
- list_del(&page->list);
+ list_del(&page->lru);
__free_pages_bulk(page, base, zone, area, mask, order);
ret++;
}
@@ -271,7 +272,7 @@ void __free_pages_ok(struct page *page,
mod_page_state(pgfree, 1 << order);
for (i = 0 ; i < (1 << order) ; ++i)
free_pages_check(__FUNCTION__, page + i);
- list_add(&page->list, &list);
+ list_add(&page->lru, &list);
kernel_map_pages(page, 1<<order, 0);
free_pages_bulk(page_zone(page), 1, &list, order);
}
@@ -290,7 +291,7 @@ expand(struct zone *zone, struct page *p
area--;
high--;
size >>= 1;
- list_add(&page->list, &area->free_list);
+ list_add(&page->lru, &area->free_list);
MARK_USED(index, high, area);
index += size;
page += size;
@@ -353,8 +354,8 @@ static struct page *__rmqueue(struct zon
if (list_empty(&area->free_list))
continue;
- page = list_entry(area->free_list.next, struct page, list);
- list_del(&page->list);
+ page = list_entry(area->free_list.next, struct page, lru);
+ list_del(&page->lru);
index = page - zone->zone_mem_map;
if (current_order != MAX_ORDER-1)
MARK_USED(index, current_order, area);
@@ -384,7 +385,7 @@ static int rmqueue_bulk(struct zone *zon
if (page == NULL)
break;
allocated++;
- list_add_tail(&page->list, list);
+ list_add_tail(&page->lru, list);
}
spin_unlock_irqrestore(&zone->lock, flags);
return allocated;
@@ -426,7 +427,7 @@ int is_head_of_free_region(struct page *
spin_lock_irqsave(&zone->lock, flags);
for (order = MAX_ORDER - 1; order >= 0; --order)
list_for_each(curr, &zone->free_area[order].free_list)
- if (page == list_entry(curr, struct page, list)) {
+ if (page == list_entry(curr, struct page, lru)) {
spin_unlock_irqrestore(&zone->lock, flags);
return 1 << order;
}
@@ -464,7 +465,7 @@ static void fastcall free_hot_cold_page(
local_irq_save(flags);
if (pcp->count >= pcp->high)
pcp->count -= free_pages_bulk(zone, pcp->batch, &pcp->list, 0);
- list_add(&page->list, &pcp->list);
+ list_add(&page->lru, &pcp->list);
pcp->count++;
local_irq_restore(flags);
put_cpu();
@@ -500,8 +501,8 @@ static struct page *buffered_rmqueue(str
pcp->count += rmqueue_bulk(zone, 0,
pcp->batch, &pcp->list);
if (pcp->count) {
- page = list_entry(pcp->list.next, struct page, list);
- list_del(&page->list);
+ page = list_entry(pcp->list.next, struct page, lru);
+ list_del(&page->lru);
pcp->count--;
}
local_irq_restore(flags);
@@ -512,14 +513,14 @@ static struct page *buffered_rmqueue(str
spin_lock_irqsave(&zone->lock, flags);
page = __rmqueue(zone, order);
spin_unlock_irqrestore(&zone->lock, flags);
- if (order && page)
- prep_compound_page(page, order);
}
if (page != NULL) {
BUG_ON(bad_range(zone, page));
mod_page_state_zone(zone, pgalloc, 1 << order);
prep_new_page(page, order);
+ if (order)
+ prep_compound_page(page, order);
}
return page;
}
@@ -1360,7 +1361,7 @@ void __init memmap_init_zone(struct page
set_page_zone(page, NODEZONE(nid, zone));
set_page_count(page, 0);
SetPageReserved(page);
- INIT_LIST_HEAD(&page->list);
+ INIT_LIST_HEAD(&page->lru);
#ifdef WANT_PAGE_VIRTUAL
/* The shift won't overflow because ZONE_NORMAL is below 4G. */
if (zone != ZONE_HIGHMEM)
--- diff/mm/page_io.c 2002-12-16 09:26:07.000000000 +0000
+++ source/mm/page_io.c 2004-03-16 11:27:48.683786600 +0000
@@ -104,7 +104,7 @@ int swap_writepage(struct page *page, st
goto out;
}
inc_page_state(pswpout);
- SetPageWriteback(page);
+ set_page_writeback(page);
unlock_page(page);
submit_bio(WRITE, bio);
out:
--- diff/mm/pdflush.c 2004-03-11 10:20:29.000000000 +0000
+++ source/mm/pdflush.c 2004-03-16 11:27:48.685786296 +0000
@@ -88,6 +88,8 @@ struct pdflush_work {
unsigned long when_i_went_to_sleep;
};
+static int wakeup_count = 100;
+
static int __pdflush(struct pdflush_work *my_work)
{
current->flags |= PF_FLUSHER;
@@ -114,7 +116,10 @@ static int __pdflush(struct pdflush_work
spin_lock_irq(&pdflush_lock);
if (!list_empty(&my_work->list)) {
- printk("pdflush: bogus wakeup!\n");
+ if (wakeup_count > 0) {
+ wakeup_count--;
+ printk("pdflush: bogus wakeup!\n");
+ }
my_work->fn = NULL;
continue;
}
@@ -184,6 +189,7 @@ int pdflush_operation(void (*fn)(unsigne
{
unsigned long flags;
int ret = 0;
+ static int poke_count = 0;
if (fn == NULL)
BUG(); /* Hard to diagnose if it's deferred */
@@ -192,9 +198,19 @@ int pdflush_operation(void (*fn)(unsigne
if (list_empty(&pdflush_list)) {
spin_unlock_irqrestore(&pdflush_lock, flags);
ret = -1;
+ if (wakeup_count < 100 && poke_count < 10) {
+ printk("%s: no threads\n", __FUNCTION__);
+ dump_stack();
+ poke_count++;
+ }
} else {
struct pdflush_work *pdf;
+ if (wakeup_count < 100 && poke_count < 10) {
+ printk("%s: found a thread\n", __FUNCTION__);
+ dump_stack();
+ poke_count++;
+ }
pdf = list_entry(pdflush_list.next, struct pdflush_work, list);
list_del_init(&pdf->list);
if (list_empty(&pdflush_list))
--- diff/mm/readahead.c 2004-03-16 11:24:02.000000000 +0000
+++ source/mm/readahead.c 2004-03-16 11:27:48.686786144 +0000
@@ -15,11 +15,16 @@
#include <linux/backing-dev.h>
#include <linux/pagevec.h>
+void default_unplug_io_fn(struct backing_dev_info *bdi)
+{
+}
+EXPORT_SYMBOL(default_unplug_io_fn);
+
struct backing_dev_info default_backing_dev_info = {
.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE,
.state = 0,
+ .unplug_io_fn = default_unplug_io_fn,
};
-
EXPORT_SYMBOL_GPL(default_backing_dev_info);
/*
@@ -32,7 +37,6 @@ file_ra_state_init(struct file_ra_state
ra->ra_pages = mapping->backing_dev_info->ra_pages;
ra->average = ra->ra_pages / 2;
}
-
EXPORT_SYMBOL(file_ra_state_init);
/*
@@ -48,7 +52,7 @@ static inline unsigned long get_min_read
return (VM_MIN_READAHEAD * 1024) / PAGE_CACHE_SIZE;
}
-#define list_to_page(head) (list_entry((head)->prev, struct page, list))
+#define list_to_page(head) (list_entry((head)->prev, struct page, lru))
/**
* read_cache_pages - populate an address space with some pages, and
@@ -72,7 +76,7 @@ int read_cache_pages(struct address_spac
while (!list_empty(pages)) {
page = list_to_page(pages);
- list_del(&page->list);
+ list_del(&page->lru);
if (add_to_page_cache(page, mapping, page->index, GFP_KERNEL)) {
page_cache_release(page);
continue;
@@ -85,7 +89,7 @@ int read_cache_pages(struct address_spac
struct page *victim;
victim = list_to_page(pages);
- list_del(&victim->list);
+ list_del(&victim->lru);
page_cache_release(victim);
}
break;
@@ -112,7 +116,7 @@ static int read_pages(struct address_spa
pagevec_init(&lru_pvec, 0);
for (page_idx = 0; page_idx < nr_pages; page_idx++) {
struct page *page = list_to_page(pages);
- list_del(&page->list);
+ list_del(&page->lru);
if (!add_to_page_cache(page, mapping,
page->index, GFP_KERNEL)) {
mapping->a_ops->readpage(filp, page);
@@ -230,7 +234,7 @@ __do_page_cache_readahead(struct address
/*
* Preallocate as many pages as we will need.
*/
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&mapping->tree_lock);
for (page_idx = 0; page_idx < nr_to_read; page_idx++) {
unsigned long page_offset = offset + page_idx;
@@ -241,16 +245,16 @@ __do_page_cache_readahead(struct address
if (page)
continue;
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
page = page_cache_alloc_cold(mapping);
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&mapping->tree_lock);
if (!page)
break;
page->index = page_offset;
- list_add(&page->list, &page_pool);
+ list_add(&page->lru, &page_pool);
ret++;
}
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
/*
* Now start the IO. We ignore I/O errors - if the page is not
--- diff/mm/rmap.c 2004-03-11 10:20:29.000000000 +0000
+++ source/mm/rmap.c 2004-03-16 11:27:48.686786144 +0000
@@ -118,7 +118,7 @@ int fastcall page_referenced(struct page
int referenced = 0;
if (page_test_and_clear_young(page))
- mark_page_accessed(page);
+ referenced++;
if (TestClearPageReferenced(page))
referenced++;
@@ -343,6 +343,7 @@ static int fastcall try_to_unmap_one(str
BUG_ON(pte_file(*ptep));
} else {
unsigned long pgidx;
+ pgprot_t pgprot = pte_to_pgprot(pte);
/*
* If a nonlinear mapping then store the file page offset
* in the pte.
@@ -350,8 +351,10 @@ static int fastcall try_to_unmap_one(str
pgidx = (address - vma->vm_start) >> PAGE_SHIFT;
pgidx += vma->vm_pgoff;
pgidx >>= PAGE_CACHE_SHIFT - PAGE_SHIFT;
- if (page->index != pgidx) {
- set_pte(ptep, pgoff_to_pte(page->index));
+ if (page->index != pgidx ||
+ pgprot_val(pgprot) != pgprot_val(vma->vm_page_prot)) {
+
+ set_pte(ptep, pgoff_prot_to_pte(page->index, pgprot));
BUG_ON(!pte_file(*ptep));
}
}
@@ -360,7 +363,7 @@ static int fastcall try_to_unmap_one(str
if (pte_dirty(pte))
set_page_dirty(page);
- mm->rss--;
+ dec_rss(mm, page);
page_cache_release(page);
ret = SWAP_SUCCESS;
@@ -522,8 +525,8 @@ void __init pte_chain_init(void)
{
pte_chain_cache = kmem_cache_create( "pte_chain",
sizeof(struct pte_chain),
+ sizeof(struct pte_chain),
0,
- SLAB_MUST_HWCACHE_ALIGN,
pte_chain_ctor,
NULL);
--- diff/mm/shmem.c 2004-02-09 10:36:12.000000000 +0000
+++ source/mm/shmem.c 2004-03-16 11:27:48.688785840 +0000
@@ -133,6 +133,7 @@ static struct vm_operations_struct shmem
static struct backing_dev_info shmem_backing_dev_info = {
.ra_pages = 0, /* No readahead */
.memory_backed = 1, /* Does not contribute to dirty memory */
+ .unplug_io_fn = default_unplug_io_fn,
};
LIST_HEAD(shmem_inodes);
@@ -1002,6 +1003,7 @@ static int shmem_populate(struct vm_area
struct mm_struct *mm = vma->vm_mm;
enum sgp_type sgp = nonblock? SGP_QUICK: SGP_CACHE;
unsigned long size;
+ int linear = !(vma->vm_flags & VM_NONLINEAR);
size = (i_size_read(inode) + PAGE_SIZE - 1) >> PAGE_SHIFT;
if (pgoff >= size || pgoff + (len >> PAGE_SHIFT) > size)
@@ -1023,20 +1025,14 @@ static int shmem_populate(struct vm_area
page_cache_release(page);
return err;
}
- } else if (nonblock) {
+ } else {
/*
- * If a nonlinear mapping then store the file page
- * offset in the pte.
+ * Store the file page offset in the pte:
*/
- unsigned long pgidx;
- pgidx = (addr - vma->vm_start) >> PAGE_SHIFT;
- pgidx += vma->vm_pgoff;
- pgidx >>= PAGE_CACHE_SHIFT - PAGE_SHIFT;
- if (pgoff != pgidx) {
- err = install_file_pte(mm, vma, addr, pgoff, prot);
- if (err)
- return err;
- }
+ err = install_file_pte(mm, vma, addr,
+ pgoff, prot, linear);
+ if (err)
+ return err;
}
len -= PAGE_SIZE;
--- diff/mm/slab.c 2004-03-16 11:24:02.000000000 +0000
+++ source/mm/slab.c 2004-03-16 11:27:48.691785384 +0000
@@ -121,6 +121,14 @@
/* Shouldn't this be in a header file somewhere? */
#define BYTES_PER_WORD sizeof(void *)
+#ifndef cache_line_size
+#define cache_line_size() L1_CACHE_BYTES
+#endif
+
+#ifndef ARCH_KMALLOC_MINALIGN
+#define ARCH_KMALLOC_MINALIGN 0
+#endif
+
/* Legal flag mask for kmem_cache_create(). */
#if DEBUG
# define CREATE_MASK (SLAB_DEBUG_INITIAL | SLAB_RED_ZONE | \
@@ -268,6 +276,7 @@ struct kmem_cache_s {
unsigned int colour_off; /* colour offset */
unsigned int colour_next; /* cache colouring */
kmem_cache_t *slabp_cache;
+ unsigned int slab_size;
unsigned int dflags; /* dynamic flags */
/* constructor func */
@@ -453,10 +462,10 @@ static int slab_break_gfp_order = BREAK_
* global 'mem_map'. These are used to find the slab an obj belongs to.
* With kfree(), these are used to find the cache which an obj belongs to.
*/
-#define SET_PAGE_CACHE(pg,x) ((pg)->list.next = (struct list_head *)(x))
-#define GET_PAGE_CACHE(pg) ((kmem_cache_t *)(pg)->list.next)
-#define SET_PAGE_SLAB(pg,x) ((pg)->list.prev = (struct list_head *)(x))
-#define GET_PAGE_SLAB(pg) ((struct slab *)(pg)->list.prev)
+#define SET_PAGE_CACHE(pg,x) ((pg)->lru.next = (struct list_head *)(x))
+#define GET_PAGE_CACHE(pg) ((kmem_cache_t *)(pg)->lru.next)
+#define SET_PAGE_SLAB(pg,x) ((pg)->lru.prev = (struct list_head *)(x))
+#define GET_PAGE_SLAB(pg) ((struct slab *)(pg)->lru.prev)
/* These are the default caches for kmalloc. Custom caches can have other sizes. */
struct cache_sizes malloc_sizes[] = {
@@ -490,8 +499,10 @@ static kmem_cache_t cache_cache = {
.objsize = sizeof(kmem_cache_t),
.flags = SLAB_NO_REAP,
.spinlock = SPIN_LOCK_UNLOCKED,
- .colour_off = L1_CACHE_BYTES,
.name = "kmem_cache",
+#if DEBUG
+ .reallen = sizeof(kmem_cache_t),
+#endif
};
/* Guard access to the cache-chain. */
@@ -535,7 +546,7 @@ static inline struct array_cache *ac_dat
}
/* Cal the num objs, wastage, and bytes left over for a given slab size. */
-static void cache_estimate (unsigned long gfporder, size_t size,
+static void cache_estimate (unsigned long gfporder, size_t size, size_t align,
int flags, size_t *left_over, unsigned int *num)
{
int i;
@@ -548,7 +559,7 @@ static void cache_estimate (unsigned lon
extra = sizeof(kmem_bufctl_t);
}
i = 0;
- while (i*size + L1_CACHE_ALIGN(base+i*extra) <= wastage)
+ while (i*size + ALIGN(base+i*extra, align) <= wastage)
i++;
if (i > 0)
i--;
@@ -558,7 +569,7 @@ static void cache_estimate (unsigned lon
*num = i;
wastage -= i*size;
- wastage -= L1_CACHE_ALIGN(base+i*extra);
+ wastage -= ALIGN(base+i*extra, align);
*left_over = wastage;
}
@@ -688,16 +699,20 @@ void __init kmem_cache_init(void)
init_MUTEX(&cache_chain_sem);
INIT_LIST_HEAD(&cache_chain);
list_add(&cache_cache.next, &cache_chain);
+ cache_cache.colour_off = cache_line_size();
cache_cache.array[smp_processor_id()] = &initarray_cache.cache;
- cache_estimate(0, cache_cache.objsize, 0,
- &left_over, &cache_cache.num);
+ cache_cache.objsize = ALIGN(cache_cache.objsize, cache_line_size());
+
+ cache_estimate(0, cache_cache.objsize, cache_line_size(), 0,
+ &left_over, &cache_cache.num);
if (!cache_cache.num)
BUG();
cache_cache.colour = left_over/cache_cache.colour_off;
cache_cache.colour_next = 0;
-
+ cache_cache.slab_size = ALIGN(cache_cache.num*sizeof(kmem_bufctl_t) +
+ sizeof(struct slab), cache_line_size());
/* 2+3) create the kmalloc caches */
sizes = malloc_sizes;
@@ -711,7 +726,7 @@ void __init kmem_cache_init(void)
* allow tighter packing of the smaller caches. */
sizes->cs_cachep = kmem_cache_create(
names->name, sizes->cs_size,
- 0, SLAB_HWCACHE_ALIGN, NULL, NULL);
+ ARCH_KMALLOC_MINALIGN, 0, NULL, NULL);
if (!sizes->cs_cachep)
BUG();
@@ -723,7 +738,7 @@ void __init kmem_cache_init(void)
sizes->cs_dmacachep = kmem_cache_create(
names->name_dma, sizes->cs_size,
- 0, SLAB_CACHE_DMA|SLAB_HWCACHE_ALIGN, NULL, NULL);
+ ARCH_KMALLOC_MINALIGN, SLAB_CACHE_DMA, NULL, NULL);
if (!sizes->cs_dmacachep)
BUG();
@@ -914,7 +929,7 @@ static void print_objinfo(kmem_cache_t *
printk("\n");
}
realobj = (char*)objp+obj_dbghead(cachep);
- size = cachep->objsize;
+ size = obj_reallen(cachep);
for (i=0; i<size && lines;i+=16, lines--) {
int limit;
limit = 16;
@@ -1039,7 +1054,7 @@ static void slab_destroy (kmem_cache_t *
* kmem_cache_create - Create a cache.
* @name: A string which is used in /proc/slabinfo to identify this cache.
* @size: The size of objects to be created in this cache.
- * @offset: The offset to use within the page.
+ * @align: The required alignment for the objects.
* @flags: SLAB flags
* @ctor: A constructor for the objects.
* @dtor: A destructor for the objects.
@@ -1064,16 +1079,15 @@ static void slab_destroy (kmem_cache_t *
* %SLAB_NO_REAP - Don't automatically reap this cache when we're under
* memory pressure.
*
- * %SLAB_HWCACHE_ALIGN - Align the objects in this cache to a hardware
- * cacheline. This can be beneficial if you're counting cycles as closely
- * as davem.
+ * %SLAB_HWCACHE_ALIGN - This flag has no effect and will be removed soon.
+ *
*/
kmem_cache_t *
-kmem_cache_create (const char *name, size_t size, size_t offset,
+kmem_cache_create (const char *name, size_t size, size_t align,
unsigned long flags, void (*ctor)(void*, kmem_cache_t *, unsigned long),
void (*dtor)(void*, kmem_cache_t *, unsigned long))
{
- size_t left_over, align, slab_size;
+ size_t left_over, slab_size;
kmem_cache_t *cachep = NULL;
/*
@@ -1084,7 +1098,7 @@ kmem_cache_create (const char *name, siz
(size < BYTES_PER_WORD) ||
(size > (1<<MAX_OBJ_ORDER)*PAGE_SIZE) ||
(dtor && !ctor) ||
- (offset < 0 || offset > size)) {
+ (align < 0)) {
printk(KERN_ERR "%s: Early error in slab %s\n",
__FUNCTION__, name);
BUG();
@@ -1101,22 +1115,16 @@ kmem_cache_create (const char *name, siz
#if FORCED_DEBUG
/*
- * Enable redzoning and last user accounting, except
- * - for caches with forced alignment: redzoning would violate the
- * alignment
- * - for caches with large objects, if the increased size would
- * increase the object size above the next power of two: caches
- * with object sizes just above a power of two have a significant
- * amount of internal fragmentation
+ * Enable redzoning and last user accounting, except for caches with
+ * large objects, if the increased size would increase the object size
+ * above the next power of two: caches with object sizes just above a
+ * power of two have a significant amount of internal fragmentation.
*/
- if ((size < 4096 || fls(size-1) == fls(size-1+3*BYTES_PER_WORD))
- && !(flags & SLAB_MUST_HWCACHE_ALIGN)) {
+ if ((size < 4096 || fls(size-1) == fls(size-1+3*BYTES_PER_WORD)))
flags |= SLAB_RED_ZONE|SLAB_STORE_USER;
- }
flags |= SLAB_POISON;
#endif
#endif
-
/*
* Always checks flags, a caller might be expecting debug
* support which isn't available.
@@ -1124,15 +1132,23 @@ kmem_cache_create (const char *name, siz
if (flags & ~CREATE_MASK)
BUG();
+ if (align) {
+ /* minimum supported alignment: */
+ if (align < BYTES_PER_WORD)
+ align = BYTES_PER_WORD;
+
+ /* combinations of forced alignment and advanced debugging is
+ * not yet implemented.
+ */
+ flags &= ~(SLAB_RED_ZONE|SLAB_STORE_USER);
+ }
+
/* Get cache's description obj. */
cachep = (kmem_cache_t *) kmem_cache_alloc(&cache_cache, SLAB_KERNEL);
if (!cachep)
goto opps;
memset(cachep, 0, sizeof(kmem_cache_t));
-#if DEBUG
- cachep->reallen = size;
-#endif
/* Check that size is in terms of words. This is needed to avoid
* unaligned accesses for some archs when redzoning is used, and makes
* sure any on-slab bufctl's are also correctly aligned.
@@ -1143,30 +1159,31 @@ kmem_cache_create (const char *name, siz
}
#if DEBUG
+ cachep->reallen = size;
+
if (flags & SLAB_RED_ZONE) {
- /*
- * There is no point trying to honour cache alignment
- * when redzoning.
- */
- flags &= ~SLAB_HWCACHE_ALIGN;
+ /* redzoning only works with word aligned caches */
+ align = BYTES_PER_WORD;
+
/* add space for red zone words */
cachep->dbghead += BYTES_PER_WORD;
size += 2*BYTES_PER_WORD;
}
if (flags & SLAB_STORE_USER) {
- flags &= ~SLAB_HWCACHE_ALIGN;
- size += BYTES_PER_WORD; /* add space */
+ /* user store requires word alignment and
+ * one word storage behind the end of the real
+ * object.
+ */
+ align = BYTES_PER_WORD;
+ size += BYTES_PER_WORD;
}
#if FORCED_DEBUG && defined(CONFIG_DEBUG_PAGEALLOC)
- if (size > 128 && cachep->reallen > L1_CACHE_BYTES && size < PAGE_SIZE) {
+ if (size > 128 && cachep->reallen > cache_line_size() && size < PAGE_SIZE) {
cachep->dbghead += PAGE_SIZE - size;
size = PAGE_SIZE;
}
#endif
#endif
- align = BYTES_PER_WORD;
- if (flags & SLAB_HWCACHE_ALIGN)
- align = L1_CACHE_BYTES;
/* Determine if the slab management is 'on' or 'off' slab. */
if (size >= (PAGE_SIZE>>3))
@@ -1176,13 +1193,16 @@ kmem_cache_create (const char *name, siz
*/
flags |= CFLGS_OFF_SLAB;
- if (flags & SLAB_HWCACHE_ALIGN) {
- /* Need to adjust size so that objs are cache aligned. */
- /* Small obj size, can get at least two per cache line. */
+ if (!align) {
+ /* Default alignment: compile time specified l1 cache size.
+ * Except if an object is really small, then squeeze multiple
+ * into one cacheline.
+ */
+ align = cache_line_size();
while (size <= align/2)
align /= 2;
- size = (size+align-1)&(~(align-1));
}
+ size = ALIGN(size, align);
/* Cal size (in pages) of slabs, and the num of objs per slab.
* This could be made much more intelligent. For now, try to avoid
@@ -1192,7 +1212,7 @@ kmem_cache_create (const char *name, siz
do {
unsigned int break_flag = 0;
cal_wastage:
- cache_estimate(cachep->gfporder, size, flags,
+ cache_estimate(cachep->gfporder, size, align, flags,
&left_over, &cachep->num);
if (break_flag)
break;
@@ -1226,7 +1246,8 @@ next:
cachep = NULL;
goto opps;
}
- slab_size = L1_CACHE_ALIGN(cachep->num*sizeof(kmem_bufctl_t)+sizeof(struct slab));
+ slab_size = ALIGN(cachep->num*sizeof(kmem_bufctl_t)
+ + sizeof(struct slab), align);
/*
* If the slab has been placed off-slab, and we have enough space then
@@ -1237,14 +1258,17 @@ next:
left_over -= slab_size;
}
- /* Offset must be a multiple of the alignment. */
- offset += (align-1);
- offset &= ~(align-1);
- if (!offset)
- offset = L1_CACHE_BYTES;
- cachep->colour_off = offset;
- cachep->colour = left_over/offset;
+ if (flags & CFLGS_OFF_SLAB) {
+ /* really off slab. No need for manual alignment */
+ slab_size = cachep->num*sizeof(kmem_bufctl_t)+sizeof(struct slab);
+ }
+ cachep->colour_off = cache_line_size();
+ /* Offset must be a multiple of the alignment. */
+ if (cachep->colour_off < align)
+ cachep->colour_off = align;
+ cachep->colour = left_over/cachep->colour_off;
+ cachep->slab_size = slab_size;
cachep->flags = flags;
cachep->gfpflags = 0;
if (flags & SLAB_CACHE_DMA)
@@ -1270,10 +1294,13 @@ next:
* the cache that's used by kmalloc(24), otherwise
* the creation of further caches will BUG().
*/
- cachep->array[smp_processor_id()] = &initarray_generic.cache;
+ cachep->array[smp_processor_id()] =
+ &initarray_generic.cache;
g_cpucache_up = PARTIAL;
} else {
- cachep->array[smp_processor_id()] = kmalloc(sizeof(struct arraycache_init),GFP_KERNEL);
+ cachep->array[smp_processor_id()] =
+ kmalloc(sizeof(struct arraycache_init),
+ GFP_KERNEL);
}
BUG_ON(!ac_data(cachep));
ac_data(cachep)->avail = 0;
@@ -1287,7 +1314,7 @@ next:
}
cachep->lists.next_reap = jiffies + REAPTIMEOUT_LIST3 +
- ((unsigned long)cachep)%REAPTIMEOUT_LIST3;
+ ((unsigned long)cachep)%REAPTIMEOUT_LIST3;
/* Need the semaphore to access the chain. */
down(&cache_chain_sem);
@@ -1300,16 +1327,24 @@ next:
list_for_each(p, &cache_chain) {
kmem_cache_t *pc = list_entry(p, kmem_cache_t, next);
char tmp;
- /* This happens when the module gets unloaded and doesn't
- destroy its slab cache and noone else reuses the vmalloc
- area of the module. Print a warning. */
- if (__get_user(tmp,pc->name)) {
- printk("SLAB: cache with size %d has lost its name\n",
- pc->objsize);
+
+ /*
+ * This happens when the module gets unloaded and
+ * doesn't destroy its slab cache and noone else reuses
+ * the vmalloc area of the module. Print a warning.
+ */
+#ifdef CONFIG_X86_UACCESS_INDIRECT
+ if (__direct_get_user(tmp,pc->name)) {
+#else
+ if (__get_user(tmp,pc->name)) {
+#endif
+ printk("SLAB: cache with size %d has lost its "
+ "name\n", pc->objsize);
continue;
}
if (!strcmp(pc->name,name)) {
- printk("kmem_cache_create: duplicate cache %s\n",name);
+ printk("kmem_cache_create: duplicate "
+ "cache %s\n",name);
up(&cache_chain_sem);
BUG();
}
@@ -1512,8 +1547,7 @@ static inline struct slab* alloc_slabmgm
return NULL;
} else {
slabp = objp+colour_off;
- colour_off += L1_CACHE_ALIGN(cachep->num *
- sizeof(kmem_bufctl_t) + sizeof(struct slab));
+ colour_off += cachep->slab_size;
}
slabp->inuse = 0;
slabp->colouroff = colour_off;
@@ -1934,6 +1968,15 @@ cache_alloc_debugcheck_after(kmem_cache_
*dbg_redzone1(cachep, objp) = RED_ACTIVE;
*dbg_redzone2(cachep, objp) = RED_ACTIVE;
}
+ {
+ int objnr;
+ struct slab *slabp;
+
+ slabp = GET_PAGE_SLAB(virt_to_page(objp));
+
+ objnr = (objp - slabp->s_mem) / cachep->objsize;
+ slab_bufctl(slabp)[objnr] = (unsigned long)caller;
+ }
objp += obj_dbghead(cachep);
if (cachep->ctor && cachep->flags & SLAB_POISON) {
unsigned long ctor_flags = SLAB_CTOR_CONSTRUCTOR;
@@ -1995,12 +2038,14 @@ static void free_block(kmem_cache_t *cac
objnr = (objp - slabp->s_mem) / cachep->objsize;
check_slabp(cachep, slabp);
#if DEBUG
+#if 0
if (slab_bufctl(slabp)[objnr] != BUFCTL_FREE) {
printk(KERN_ERR "slab: double free detected in cache '%s', objp %p.\n",
cachep->name, objp);
BUG();
}
#endif
+#endif
slab_bufctl(slabp)[objnr] = slabp->free;
slabp->free = objnr;
STATS_DEC_ACTIVE(cachep);
@@ -2781,6 +2826,29 @@ struct seq_operations slabinfo_op = {
.show = s_show,
};
+static void do_dump_slabp(kmem_cache_t *cachep)
+{
+#if DEBUG
+ struct list_head *q;
+
+ check_irq_on();
+ spin_lock_irq(&cachep->spinlock);
+ list_for_each(q,&cachep->lists.slabs_full) {
+ struct slab *slabp;
+ int i;
+ slabp = list_entry(q, struct slab, list);
+ for (i = 0; i < cachep->num; i++) {
+ unsigned long sym = slab_bufctl(slabp)[i];
+
+ printk("obj %p/%d: %p", slabp, i, (void *)sym);
+ print_symbol(" <%s>", sym);
+ printk("\n");
+ }
+ }
+ spin_unlock_irq(&cachep->spinlock);
+#endif
+}
+
#define MAX_SLABINFO_WRITE 128
/**
* slabinfo_write - Tuning for the slab allocator
@@ -2821,9 +2889,11 @@ ssize_t slabinfo_write(struct file *file
batchcount < 1 ||
batchcount > limit ||
shared < 0) {
- res = -EINVAL;
+ do_dump_slabp(cachep);
+ res = 0;
} else {
- res = do_tune_cpucache(cachep, limit, batchcount, shared);
+ res = do_tune_cpucache(cachep, limit,
+ batchcount, shared);
}
break;
}
--- diff/mm/swap.c 2004-03-11 10:20:29.000000000 +0000
+++ source/mm/swap.c 2004-03-16 11:27:48.692785232 +0000
@@ -67,7 +67,7 @@ int rotate_reclaimable_page(struct page
list_add_tail(&page->lru, &zone->inactive_list);
inc_page_state(pgrotated);
}
- if (!TestClearPageWriteback(page))
+ if (!test_clear_page_writeback(page))
BUG();
spin_unlock_irqrestore(&zone->lru_lock, flags);
return 0;
@@ -350,13 +350,21 @@ void pagevec_strip(struct pagevec *pvec)
*
* pagevec_lookup() returns the number of pages which were found.
*/
-unsigned int pagevec_lookup(struct pagevec *pvec, struct address_space *mapping,
- pgoff_t start, unsigned int nr_pages)
+unsigned pagevec_lookup(struct pagevec *pvec, struct address_space *mapping,
+ pgoff_t start, unsigned nr_pages)
{
pvec->nr = find_get_pages(mapping, start, nr_pages, pvec->pages);
return pagevec_count(pvec);
}
+unsigned pagevec_lookup_tag(struct pagevec *pvec, struct address_space *mapping,
+ pgoff_t *index, int tag, unsigned nr_pages)
+{
+ pvec->nr = find_get_pages_tag(mapping, index, tag,
+ nr_pages, pvec->pages);
+ return pagevec_count(pvec);
+}
+
#ifdef CONFIG_SMP
/*
--- diff/mm/swap_state.c 2003-08-20 13:16:34.000000000 +0000
+++ source/mm/swap_state.c 2004-03-16 11:27:48.695784776 +0000
@@ -19,17 +19,14 @@
static struct backing_dev_info swap_backing_dev_info = {
.ra_pages = 0, /* No readahead */
.memory_backed = 1, /* Does not contribute to dirty memory */
+ .unplug_io_fn = swap_unplug_io_fn,
};
extern struct address_space_operations swap_aops;
struct address_space swapper_space = {
.page_tree = RADIX_TREE_INIT(GFP_ATOMIC),
- .page_lock = SPIN_LOCK_UNLOCKED,
- .clean_pages = LIST_HEAD_INIT(swapper_space.clean_pages),
- .dirty_pages = LIST_HEAD_INIT(swapper_space.dirty_pages),
- .io_pages = LIST_HEAD_INIT(swapper_space.io_pages),
- .locked_pages = LIST_HEAD_INIT(swapper_space.locked_pages),
+ .tree_lock = SPIN_LOCK_UNLOCKED,
.a_ops = &swap_aops,
.backing_dev_info = &swap_backing_dev_info,
.i_mmap = LIST_HEAD_INIT(swapper_space.i_mmap),
@@ -149,7 +146,7 @@ int add_to_swap(struct page * page)
switch (err) {
case 0: /* Success */
SetPageUptodate(page);
- ClearPageDirty(page);
+ __clear_page_dirty(page);
set_page_dirty(page);
INC_CACHE_INFO(add_total);
return 1;
@@ -182,9 +179,9 @@ void delete_from_swap_cache(struct page
entry.val = page->index;
- spin_lock(&swapper_space.page_lock);
+ spin_lock_irq(&swapper_space.tree_lock);
__delete_from_swap_cache(page);
- spin_unlock(&swapper_space.page_lock);
+ spin_unlock_irq(&swapper_space.tree_lock);
swap_free(entry);
page_cache_release(page);
@@ -195,8 +192,8 @@ int move_to_swap_cache(struct page *page
struct address_space *mapping = page->mapping;
int err;
- spin_lock(&swapper_space.page_lock);
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&swapper_space.tree_lock);
+ spin_lock(&mapping->tree_lock);
err = radix_tree_insert(&swapper_space.page_tree, entry.val, page);
if (!err) {
@@ -204,13 +201,12 @@ int move_to_swap_cache(struct page *page
___add_to_page_cache(page, &swapper_space, entry.val);
}
- spin_unlock(&mapping->page_lock);
- spin_unlock(&swapper_space.page_lock);
+ spin_unlock(&mapping->tree_lock);
+ spin_unlock_irq(&swapper_space.tree_lock);
if (!err) {
if (!swap_duplicate(entry))
BUG();
- /* shift page from clean_pages to dirty_pages list */
BUG_ON(PageDirty(page));
set_page_dirty(page);
INC_CACHE_INFO(add_total);
@@ -231,8 +227,8 @@ int move_from_swap_cache(struct page *pa
entry.val = page->index;
- spin_lock(&swapper_space.page_lock);
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&swapper_space.tree_lock);
+ spin_lock(&mapping->tree_lock);
err = radix_tree_insert(&mapping->page_tree, index, page);
if (!err) {
@@ -240,13 +236,12 @@ int move_from_swap_cache(struct page *pa
___add_to_page_cache(page, mapping, index);
}
- spin_unlock(&mapping->page_lock);
- spin_unlock(&swapper_space.page_lock);
+ spin_unlock(&mapping->tree_lock);
+ spin_unlock_irq(&swapper_space.tree_lock);
if (!err) {
swap_free(entry);
- /* shift page from clean_pages to dirty_pages list */
- ClearPageDirty(page);
+ __clear_page_dirty(page);
set_page_dirty(page);
}
return err;
--- diff/mm/swapfile.c 2004-03-16 11:24:02.000000000 +0000
+++ source/mm/swapfile.c 2004-03-16 11:27:48.694784928 +0000
@@ -23,6 +23,7 @@
#include <linux/module.h>
#include <linux/rmap-locking.h>
#include <linux/security.h>
+#include <linux/backing-dev.h>
#include <asm/pgtable.h>
#include <asm/tlbflush.h>
@@ -44,8 +45,64 @@ struct swap_list_t swap_list = {-1, -1};
struct swap_info_struct swap_info[MAX_SWAPFILES];
+/*
+ * Array of backing blockdevs, for swap_unplug_fn. We need this because the
+ * bdev->unplug_fn can sleep and we cannot hold swap_list_lock while calling
+ * the unplug_fn. And swap_list_lock cannot be turned into a semaphore.
+ */
+static DECLARE_MUTEX(swap_bdevs_sem);
+static struct block_device *swap_bdevs[MAX_SWAPFILES];
+
#define SWAPFILE_CLUSTER 256
+/*
+ * Caller holds swap_bdevs_sem
+ */
+static void install_swap_bdev(struct block_device *bdev)
+{
+ int i;
+
+ for (i = 0; i < MAX_SWAPFILES; i++) {
+ if (swap_bdevs[i] == NULL) {
+ swap_bdevs[i] = bdev;
+ return;
+ }
+ }
+ BUG();
+}
+
+static void remove_swap_bdev(struct block_device *bdev)
+{
+ int i;
+
+ for (i = 0; i < MAX_SWAPFILES; i++) {
+ if (swap_bdevs[i] == bdev) {
+ memcpy(&swap_bdevs[i], &swap_bdevs[i + 1],
+ (MAX_SWAPFILES - i - 1) * sizeof(*swap_bdevs));
+ swap_bdevs[MAX_SWAPFILES - 1] = NULL;
+ return;
+ }
+ }
+ BUG();
+}
+
+void swap_unplug_io_fn(struct backing_dev_info *unused_bdi)
+{
+ int i;
+
+ down(&swap_bdevs_sem);
+ for (i = 0; i < MAX_SWAPFILES; i++) {
+ struct block_device *bdev = swap_bdevs[i];
+ struct backing_dev_info *bdi;
+
+ if (bdev == NULL)
+ break;
+ bdi = bdev->bd_inode->i_mapping->backing_dev_info;
+ (*bdi->unplug_io_fn)(bdi);
+ }
+ up(&swap_bdevs_sem);
+}
+
static inline int scan_swap_map(struct swap_info_struct *si)
{
unsigned long offset;
@@ -253,10 +310,10 @@ static int exclusive_swap_page(struct pa
/* Is the only swap cache user the cache itself? */
if (p->swap_map[swp_offset(entry)] == 1) {
/* Recheck the page count with the pagecache lock held.. */
- spin_lock(&swapper_space.page_lock);
+ spin_lock_irq(&swapper_space.tree_lock);
if (page_count(page) - !!PagePrivate(page) == 2)
retval = 1;
- spin_unlock(&swapper_space.page_lock);
+ spin_unlock_irq(&swapper_space.tree_lock);
}
swap_info_put(p);
}
@@ -324,13 +381,13 @@ int remove_exclusive_swap_page(struct pa
retval = 0;
if (p->swap_map[swp_offset(entry)] == 1) {
/* Recheck the page count with the pagecache lock held.. */
- spin_lock(&swapper_space.page_lock);
+ spin_lock_irq(&swapper_space.tree_lock);
if ((page_count(page) == 2) && !PageWriteback(page)) {
__delete_from_swap_cache(page);
SetPageDirty(page);
retval = 1;
}
- spin_unlock(&swapper_space.page_lock);
+ spin_unlock_irq(&swapper_space.tree_lock);
}
swap_info_put(p);
@@ -387,7 +444,7 @@ static void
unuse_pte(struct vm_area_struct *vma, unsigned long address, pte_t *dir,
swp_entry_t entry, struct page *page, struct pte_chain **pte_chainp)
{
- vma->vm_mm->rss++;
+ inc_rss(vma->vm_mm, page);
get_page(page);
set_pte(dir, pte_mkold(mk_pte(page, vma->vm_page_prot)));
*pte_chainp = page_add_rmap(page, dir, *pte_chainp);
@@ -1088,6 +1145,7 @@ asmlinkage long sys_swapoff(const char _
swap_list_unlock();
goto out_dput;
}
+ down(&swap_bdevs_sem);
swap_list_lock();
swap_device_lock(p);
swap_file = p->swap_file;
@@ -1099,6 +1157,8 @@ asmlinkage long sys_swapoff(const char _
destroy_swap_extents(p);
swap_device_unlock(p);
swap_list_unlock();
+ remove_swap_bdev(p->bdev);
+ up(&swap_bdevs_sem);
vfree(swap_map);
if (S_ISBLK(mapping->host->i_mode)) {
struct block_device *bdev = I_BDEV(mapping->host);
@@ -1414,6 +1474,7 @@ asmlinkage long sys_swapon(const char __
if (error)
goto bad_swap;
+ down(&swap_bdevs_sem);
swap_list_lock();
swap_device_lock(p);
p->flags = SWP_ACTIVE;
@@ -1439,6 +1500,8 @@ asmlinkage long sys_swapon(const char __
}
swap_device_unlock(p);
swap_list_unlock();
+ install_swap_bdev(p->bdev);
+ up(&swap_bdevs_sem);
error = 0;
goto out;
bad_swap:
@@ -1458,7 +1521,7 @@ bad_swap_2:
destroy_swap_extents(p);
if (swap_map)
vfree(swap_map);
- if (swap_file && !IS_ERR(swap_file))
+ if (swap_file)
filp_close(swap_file, NULL);
out:
if (page && !IS_ERR(page)) {
--- diff/mm/truncate.c 2004-03-16 11:24:02.000000000 +0000
+++ source/mm/truncate.c 2004-03-16 11:27:48.695784776 +0000
@@ -62,7 +62,7 @@ truncate_complete_page(struct address_sp
* This is for invalidate_inode_pages(). That function can be called at
* any time, and is not supposed to throw away dirty pages. But pages can
* be marked dirty at any time too. So we re-check the dirtiness inside
- * ->page_lock. That provides exclusion against the __set_page_dirty
+ * ->tree_lock. That provides exclusion against the __set_page_dirty
* functions.
*/
static int
@@ -74,13 +74,13 @@ invalidate_complete_page(struct address_
if (PagePrivate(page) && !try_to_release_page(page, 0))
return 0;
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&mapping->tree_lock);
if (PageDirty(page)) {
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
return 0;
}
__remove_from_page_cache(page);
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
ClearPageUptodate(page);
page_cache_release(page); /* pagecache ref */
return 1;
@@ -174,6 +174,9 @@ void truncate_inode_pages(struct address
}
pagevec_release(&pvec);
}
+
+ if (lstart == 0)
+ WARN_ON(mapping->nrpages);
}
EXPORT_SYMBOL(truncate_inode_pages);
--- diff/mm/vmscan.c 2004-03-16 11:24:02.000000000 +0000
+++ source/mm/vmscan.c 2004-03-16 11:27:48.698784320 +0000
@@ -352,7 +352,6 @@ shrink_list(struct list_head *page_list,
goto keep_locked;
if (!may_write_to_queue(mapping->backing_dev_info))
goto keep_locked;
- spin_lock(&mapping->page_lock);
if (test_clear_page_dirty(page)) {
int res;
struct writeback_control wbc = {
@@ -362,9 +361,6 @@ shrink_list(struct list_head *page_list,
.for_reclaim = 1,
};
- list_move(&page->list, &mapping->locked_pages);
- spin_unlock(&mapping->page_lock);
-
SetPageReclaim(page);
res = mapping->a_ops->writepage(page, &wbc);
if (res < 0)
@@ -379,7 +375,6 @@ shrink_list(struct list_head *page_list,
}
goto keep;
}
- spin_unlock(&mapping->page_lock);
}
/*
@@ -413,7 +408,7 @@ shrink_list(struct list_head *page_list,
if (!mapping)
goto keep_locked; /* truncate got there first */
- spin_lock(&mapping->page_lock);
+ spin_lock_irq(&mapping->tree_lock);
/*
* The non-racy check for busy page. It is critical to check
@@ -421,7 +416,7 @@ shrink_list(struct list_head *page_list,
* not in use by anybody. (pagecache + us == 2)
*/
if (page_count(page) != 2 || PageDirty(page)) {
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
goto keep_locked;
}
@@ -429,7 +424,7 @@ shrink_list(struct list_head *page_list,
if (PageSwapCache(page)) {
swp_entry_t swap = { .val = page->index };
__delete_from_swap_cache(page);
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
swap_free(swap);
__put_page(page); /* The pagecache ref */
goto free_it;
@@ -437,7 +432,7 @@ shrink_list(struct list_head *page_list,
#endif /* CONFIG_SWAP */
__remove_from_page_cache(page);
- spin_unlock(&mapping->page_lock);
+ spin_unlock_irq(&mapping->tree_lock);
__put_page(page);
free_it:
--- diff/net/Kconfig 2004-03-16 11:24:02.000000000 +0000
+++ source/net/Kconfig 2004-03-16 11:27:48.704783408 +0000
@@ -71,6 +71,17 @@ config UNIX
Say Y unless you know what you are doing.
+config IPMI_SOCKET
+ tristate "IPMI sockets"
+ depends on IPMI_HANDLER
+ ---help---
+ If you say Y here, you will include support for IPMI sockets;
+ This way you don't have to use devices to access IPMI. You
+ must also enable the IPMI message handler and a low-level
+ driver in the Character Drivers if you enable this.
+
+ If unsure, say N.
+
config NET_KEY
tristate "PF_KEY sockets"
select XFRM
@@ -658,18 +669,17 @@ source "net/irda/Kconfig"
source "net/bluetooth/Kconfig"
+config KGDBOE
+ def_bool X86 && KGDB
+
config NETPOLL
- def_bool NETCONSOLE
+ def_bool NETCONSOLE || KGDBOE
config NETPOLL_RX
- bool "Netpoll support for trapping incoming packets"
- default n
- depends on NETPOLL
+ def_bool KGDBOE
config NETPOLL_TRAP
- bool "Netpoll traffic trapping"
- default n
- depends on NETPOLL
+ def_bool KGDBOE
config NET_POLL_CONTROLLER
def_bool NETPOLL
--- diff/net/Makefile 2003-10-09 08:47:34.000000000 +0000
+++ source/net/Makefile 2004-03-16 11:27:48.704783408 +0000
@@ -38,6 +38,7 @@ obj-$(CONFIG_DECNET) += decnet/
obj-$(CONFIG_ECONET) += econet/
obj-$(CONFIG_VLAN_8021Q) += 8021q/
obj-$(CONFIG_IP_SCTP) += sctp/
+obj-$(CONFIG_IPMI_SOCKET) += ipmi/
ifeq ($(CONFIG_NET),y)
obj-$(CONFIG_SYSCTL) += sysctl_net.o
--- diff/net/core/dev.c 2004-03-16 11:24:02.000000000 +0000
+++ source/net/core/dev.c 2004-03-16 11:27:48.699784168 +0000
@@ -1514,7 +1514,6 @@ static void sample_queue(unsigned long d
}
#endif
-
/**
* netif_rx - post buffer to the network code
* @skb: buffer to post
@@ -1834,7 +1833,6 @@ static void net_rx_action(struct softirq
unsigned long start_time = jiffies;
int budget = netdev_max_backlog;
-
preempt_disable();
local_irq_disable();
@@ -1861,6 +1859,10 @@ static void net_rx_action(struct softirq
dev_put(dev);
local_irq_disable();
}
+
+#ifdef CONFIG_KGDBOE
+ kgdb_process_breakpoint();
+#endif
}
out:
local_irq_enable();
--- diff/net/ipv4/devinet.c 2004-03-11 10:20:29.000000000 +0000
+++ source/net/ipv4/devinet.c 2004-03-16 11:27:48.702783712 +0000
@@ -1210,11 +1210,11 @@ int ipv4_doint_and_flush_strategy(ctl_ta
static struct devinet_sysctl_table {
struct ctl_table_header *sysctl_header;
- ctl_table devinet_vars[20];
- ctl_table devinet_dev[2];
- ctl_table devinet_conf_dir[2];
- ctl_table devinet_proto_dir[2];
- ctl_table devinet_root_dir[2];
+ ctl_table devinet_vars[21];
+ ctl_table devinet_dev[3];
+ ctl_table devinet_conf_dir[3];
+ ctl_table devinet_proto_dir[3];
+ ctl_table devinet_root_dir[3];
} devinet_sysctl = {
.devinet_vars = {
{
@@ -1372,6 +1372,7 @@ static struct devinet_sysctl_table {
.proc_handler = &ipv4_doint_and_flush,
.strategy = &ipv4_doint_and_flush_strategy,
},
+ { .ctl_name = 0 }
},
.devinet_dev = {
{
@@ -1380,6 +1381,7 @@ static struct devinet_sysctl_table {
.mode = 0555,
.child = devinet_sysctl.devinet_vars,
},
+ { .ctl_name = 0 }
},
.devinet_conf_dir = {
{
@@ -1388,6 +1390,7 @@ static struct devinet_sysctl_table {
.mode = 0555,
.child = devinet_sysctl.devinet_dev,
},
+ { .ctl_name = 0 }
},
.devinet_proto_dir = {
{
@@ -1396,6 +1399,7 @@ static struct devinet_sysctl_table {
.mode = 0555,
.child = devinet_sysctl.devinet_conf_dir,
},
+ { .ctl_name = 0 }
},
.devinet_root_dir = {
{
@@ -1404,6 +1408,7 @@ static struct devinet_sysctl_table {
.mode = 0555,
.child = devinet_sysctl.devinet_proto_dir,
},
+ { .ctl_name = 0 }
},
};
--- diff/net/ipv4/netfilter/ipt_owner.c 2004-02-18 08:54:13.000000000 +0000
+++ source/net/ipv4/netfilter/ipt_owner.c 2004-03-16 11:27:48.703783560 +0000
@@ -95,7 +95,7 @@ match_sid(const struct sk_buff *skb, pid
read_lock(&tasklist_lock);
do_each_thread(g, p) {
struct files_struct *files;
- if (p->session != sid)
+ if (p->signal->session != sid)
continue;
task_lock(p);
--- diff/net/ipv6/netfilter/ip6t_owner.c 2004-02-18 08:54:13.000000000 +0000
+++ source/net/ipv6/netfilter/ip6t_owner.c 2004-03-16 11:27:48.703783560 +0000
@@ -61,7 +61,7 @@ match_sid(const struct sk_buff *skb, pid
read_lock(&tasklist_lock);
do_each_thread(g, p) {
struct files_struct *files;
- if (p->session != sid)
+ if (p->signal->session != sid)
continue;
task_lock(p);
--- diff/net/sunrpc/auth_gss/auth_gss.c 2004-03-11 10:20:29.000000000 +0000
+++ source/net/sunrpc/auth_gss/auth_gss.c 2004-03-16 11:27:48.705783256 +0000
@@ -365,7 +365,7 @@ retry:
gss_msg = gss_new;
memset(gss_new, 0, sizeof(*gss_new));
INIT_LIST_HEAD(&gss_new->list);
- INIT_RPC_WAITQ(&gss_new->waitq, "RPCSEC_GSS upcall waitq");
+ rpc_init_wait_queue(&gss_new->waitq, "RPCSEC_GSS upcall waitq");
atomic_set(&gss_new->count, 2);
msg = &gss_new->msg;
msg->data = &gss_new->uid;
--- diff/net/sunrpc/auth_gss/svcauth_gss.c 2004-03-11 10:20:29.000000000 +0000
+++ source/net/sunrpc/auth_gss/svcauth_gss.c 2004-03-16 11:27:48.706783104 +0000
@@ -687,7 +687,7 @@ svcauth_gss_accept(struct svc_rqst *rqst
u32 *reject_stat = resv->iov_base;
int ret;
- dprintk("RPC: svcauth_gss: argv->iov_len = %zd\n",argv->iov_len);
+ dprintk("RPC: svcauth_gss: argv->iov_len = %Zd\n",argv->iov_len);
*authp = rpc_autherr_badcred;
if (!svcdata)
--- diff/net/sunrpc/clnt.c 2004-03-16 11:24:02.000000000 +0000
+++ source/net/sunrpc/clnt.c 2004-03-16 11:27:48.707782952 +0000
@@ -144,7 +144,7 @@ rpc_create_client(struct rpc_xprt *xprt,
clnt->cl_vers = version->number;
clnt->cl_prot = xprt->prot;
clnt->cl_stats = program->stats;
- INIT_RPC_WAITQ(&clnt->cl_pmap_default.pm_bindwait, "bindwait");
+ rpc_init_wait_queue(&clnt->cl_pmap_default.pm_bindwait, "bindwait");
if (!clnt->cl_port)
clnt->cl_autobind = 1;
--- diff/net/sunrpc/sched.c 2004-03-16 11:24:02.000000000 +0000
+++ source/net/sunrpc/sched.c 2004-03-16 11:27:48.708782800 +0000
@@ -162,6 +162,26 @@ rpc_delete_timer(struct rpc_task *task)
}
/*
+ * Add new request to a priority queue.
+ */
+static void __rpc_add_wait_queue_priority(struct rpc_wait_queue *queue, struct rpc_task *task)
+{
+ struct list_head *q;
+ struct rpc_task *t;
+
+ q = &queue->tasks[task->tk_priority];
+ if (unlikely(task->tk_priority > queue->maxpriority))
+ q = &queue->tasks[queue->maxpriority];
+ list_for_each_entry(t, q, tk_list) {
+ if (t->tk_cookie == task->tk_cookie) {
+ list_add_tail(&task->tk_list, &t->tk_links);
+ return;
+ }
+ }
+ list_add_tail(&task->tk_list, q);
+}
+
+/*
* Add new request to wait queue.
*
* Swapper tasks always get inserted at the head of the queue.
@@ -169,8 +189,7 @@ rpc_delete_timer(struct rpc_task *task)
* improve overall performance.
* Everyone else gets appended to the queue to ensure proper FIFO behavior.
*/
-static inline int
-__rpc_add_wait_queue(struct rpc_wait_queue *queue, struct rpc_task *task)
+static int __rpc_add_wait_queue(struct rpc_wait_queue *queue, struct rpc_task *task)
{
if (task->tk_rpcwait == queue)
return 0;
@@ -179,10 +198,12 @@ __rpc_add_wait_queue(struct rpc_wait_que
printk(KERN_WARNING "RPC: doubly enqueued task!\n");
return -EWOULDBLOCK;
}
- if (RPC_IS_SWAPPER(task))
- list_add(&task->tk_list, &queue->tasks);
+ if (RPC_IS_PRIORITY(queue))
+ __rpc_add_wait_queue_priority(queue, task);
+ else if (RPC_IS_SWAPPER(task))
+ list_add(&task->tk_list, &queue->tasks[0]);
else
- list_add_tail(&task->tk_list, &queue->tasks);
+ list_add_tail(&task->tk_list, &queue->tasks[0]);
task->tk_rpcwait = queue;
dprintk("RPC: %4d added to queue %p \"%s\"\n",
@@ -191,8 +212,7 @@ __rpc_add_wait_queue(struct rpc_wait_que
return 0;
}
-int
-rpc_add_wait_queue(struct rpc_wait_queue *q, struct rpc_task *task)
+int rpc_add_wait_queue(struct rpc_wait_queue *q, struct rpc_task *task)
{
int result;
@@ -203,18 +223,35 @@ rpc_add_wait_queue(struct rpc_wait_queue
}
/*
+ * Remove request from a priority queue.
+ */
+static void __rpc_remove_wait_queue_priority(struct rpc_task *task)
+{
+ struct rpc_task *t;
+
+ if (!list_empty(&task->tk_links)) {
+ t = list_entry(task->tk_links.next, struct rpc_task, tk_list);
+ list_move(&t->tk_list, &task->tk_list);
+ list_splice_init(&task->tk_links, &t->tk_links);
+ }
+ list_del(&task->tk_list);
+}
+
+/*
* Remove request from queue.
* Note: must be called with spin lock held.
*/
-static inline void
-__rpc_remove_wait_queue(struct rpc_task *task)
+static void __rpc_remove_wait_queue(struct rpc_task *task)
{
struct rpc_wait_queue *queue = task->tk_rpcwait;
if (!queue)
return;
- list_del(&task->tk_list);
+ if (RPC_IS_PRIORITY(queue))
+ __rpc_remove_wait_queue_priority(task);
+ else
+ list_del(&task->tk_list);
task->tk_rpcwait = NULL;
dprintk("RPC: %4d removed from queue %p \"%s\"\n",
@@ -231,6 +268,48 @@ rpc_remove_wait_queue(struct rpc_task *t
spin_unlock_bh(&rpc_queue_lock);
}
+static inline void rpc_set_waitqueue_priority(struct rpc_wait_queue *queue, int priority)
+{
+ queue->priority = priority;
+ queue->count = 1 << (priority * 2);
+}
+
+static inline void rpc_set_waitqueue_cookie(struct rpc_wait_queue *queue, unsigned long cookie)
+{
+ queue->cookie = cookie;
+ queue->nr = RPC_BATCH_COUNT;
+}
+
+static inline void rpc_reset_waitqueue_priority(struct rpc_wait_queue *queue)
+{
+ rpc_set_waitqueue_priority(queue, queue->maxpriority);
+ rpc_set_waitqueue_cookie(queue, 0);
+}
+
+static void __rpc_init_priority_wait_queue(struct rpc_wait_queue *queue, const char *qname, int maxprio)
+{
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(queue->tasks); i++)
+ INIT_LIST_HEAD(&queue->tasks[i]);
+ queue->maxpriority = maxprio;
+ rpc_reset_waitqueue_priority(queue);
+#ifdef RPC_DEBUG
+ queue->name = qname;
+#endif
+}
+
+void rpc_init_priority_wait_queue(struct rpc_wait_queue *queue, const char *qname)
+{
+ __rpc_init_priority_wait_queue(queue, qname, RPC_PRIORITY_HIGH);
+}
+
+void rpc_init_wait_queue(struct rpc_wait_queue *queue, const char *qname)
+{
+ __rpc_init_priority_wait_queue(queue, qname, 0);
+}
+EXPORT_SYMBOL(rpc_init_wait_queue);
+
/*
* Make an RPC task runnable.
*
@@ -255,13 +334,11 @@ rpc_make_runnable(struct rpc_task *task)
return;
}
rpc_clear_sleeping(task);
- if (waitqueue_active(&rpciod_idle))
- wake_up(&rpciod_idle);
+ wake_up(&rpciod_idle);
}
} else {
rpc_clear_sleeping(task);
- if (waitqueue_active(&task->tk_wait))
- wake_up(&task->tk_wait);
+ wake_up(&task->tk_wait);
}
}
@@ -287,8 +364,7 @@ void rpciod_wake_up(void)
{
if(rpciod_pid==0)
printk(KERN_ERR "rpciod: wot no daemon?\n");
- if (waitqueue_active(&rpciod_idle))
- wake_up(&rpciod_idle);
+ wake_up(&rpciod_idle);
}
/*
@@ -406,17 +482,72 @@ rpc_wake_up_task(struct rpc_task *task)
}
/*
+ * Wake up the next task on a priority queue.
+ */
+static struct rpc_task * __rpc_wake_up_next_priority(struct rpc_wait_queue *queue)
+{
+ struct list_head *q;
+ struct rpc_task *task;
+
+ /*
+ * Service a batch of tasks from a single cookie.
+ */
+ q = &queue->tasks[queue->priority];
+ if (!list_empty(q)) {
+ task = list_entry(q->next, struct rpc_task, tk_list);
+ if (queue->cookie == task->tk_cookie) {
+ if (--queue->nr)
+ goto out;
+ list_move_tail(&task->tk_list, q);
+ }
+ /*
+ * Check if we need to switch queues.
+ */
+ if (--queue->count)
+ goto new_cookie;
+ }
+
+ /*
+ * Service the next queue.
+ */
+ do {
+ if (q == &queue->tasks[0])
+ q = &queue->tasks[queue->maxpriority];
+ else
+ q = q - 1;
+ if (!list_empty(q)) {
+ task = list_entry(q->next, struct rpc_task, tk_list);
+ goto new_queue;
+ }
+ } while (q != &queue->tasks[queue->priority]);
+
+ rpc_reset_waitqueue_priority(queue);
+ return NULL;
+
+new_queue:
+ rpc_set_waitqueue_priority(queue, (unsigned int)(q - &queue->tasks[0]));
+new_cookie:
+ rpc_set_waitqueue_cookie(queue, task->tk_cookie);
+out:
+ __rpc_wake_up_task(task);
+ return task;
+}
+
+/*
* Wake up the next task on the wait queue.
*/
-struct rpc_task *
-rpc_wake_up_next(struct rpc_wait_queue *queue)
+struct rpc_task * rpc_wake_up_next(struct rpc_wait_queue *queue)
{
struct rpc_task *task = NULL;
dprintk("RPC: wake_up_next(%p \"%s\")\n", queue, rpc_qname(queue));
spin_lock_bh(&rpc_queue_lock);
- task_for_first(task, &queue->tasks)
- __rpc_wake_up_task(task);
+ if (RPC_IS_PRIORITY(queue))
+ task = __rpc_wake_up_next_priority(queue);
+ else {
+ task_for_first(task, &queue->tasks[0])
+ __rpc_wake_up_task(task);
+ }
spin_unlock_bh(&rpc_queue_lock);
return task;
@@ -428,15 +559,22 @@ rpc_wake_up_next(struct rpc_wait_queue *
*
* Grabs rpc_queue_lock
*/
-void
-rpc_wake_up(struct rpc_wait_queue *queue)
+void rpc_wake_up(struct rpc_wait_queue *queue)
{
struct rpc_task *task;
+ struct list_head *head;
spin_lock_bh(&rpc_queue_lock);
- while (!list_empty(&queue->tasks))
- task_for_first(task, &queue->tasks)
+ head = &queue->tasks[queue->maxpriority];
+ for (;;) {
+ while (!list_empty(head)) {
+ task = list_entry(head->next, struct rpc_task, tk_list);
__rpc_wake_up_task(task);
+ }
+ if (head == &queue->tasks[0])
+ break;
+ head--;
+ }
spin_unlock_bh(&rpc_queue_lock);
}
@@ -447,17 +585,22 @@ rpc_wake_up(struct rpc_wait_queue *queue
*
* Grabs rpc_queue_lock
*/
-void
-rpc_wake_up_status(struct rpc_wait_queue *queue, int status)
+void rpc_wake_up_status(struct rpc_wait_queue *queue, int status)
{
+ struct list_head *head;
struct rpc_task *task;
spin_lock_bh(&rpc_queue_lock);
- while (!list_empty(&queue->tasks)) {
- task_for_first(task, &queue->tasks) {
+ head = &queue->tasks[queue->maxpriority];
+ for (;;) {
+ while (!list_empty(head)) {
+ task = list_entry(head->next, struct rpc_task, tk_list);
task->tk_status = status;
__rpc_wake_up_task(task);
}
+ if (head == &queue->tasks[0])
+ break;
+ head--;
}
spin_unlock_bh(&rpc_queue_lock);
}
@@ -643,21 +786,22 @@ __rpc_schedule(void)
dprintk("RPC: rpc_schedule enter\n");
while (1) {
- spin_lock_bh(&rpc_queue_lock);
- task_for_first(task, &schedq.tasks) {
+ task_for_first(task, &schedq.tasks[0]) {
__rpc_remove_wait_queue(task);
spin_unlock_bh(&rpc_queue_lock);
__rpc_execute(task);
+ spin_lock_bh(&rpc_queue_lock);
} else {
- spin_unlock_bh(&rpc_queue_lock);
break;
}
if (++count >= 200 || need_resched()) {
count = 0;
+ spin_unlock_bh(&rpc_queue_lock);
schedule();
+ spin_lock_bh(&rpc_queue_lock);
}
}
dprintk("RPC: rpc_schedule leave\n");
@@ -709,9 +853,7 @@ rpc_free(struct rpc_task *task)
/*
* Creation and deletion of RPC task structures
*/
-inline void
-rpc_init_task(struct rpc_task *task, struct rpc_clnt *clnt,
- rpc_action callback, int flags)
+void rpc_init_task(struct rpc_task *task, struct rpc_clnt *clnt, rpc_action callback, int flags)
{
memset(task, 0, sizeof(*task));
init_timer(&task->tk_timer);
@@ -729,6 +871,10 @@ rpc_init_task(struct rpc_task *task, str
task->tk_cred_retry = 2;
task->tk_suid_retry = 1;
+ task->tk_priority = RPC_PRIORITY_NORMAL;
+ task->tk_cookie = (unsigned long)current;
+ INIT_LIST_HEAD(&task->tk_links);
+
/* Add to global list of all tasks */
spin_lock(&rpc_sched_lock);
list_add(&task->tk_task, &all_tasks);
@@ -866,7 +1012,7 @@ rpc_find_parent(struct rpc_task *child)
struct list_head *le;
parent = (struct rpc_task *) child->tk_calldata;
- task_for_each(task, le, &childq.tasks)
+ task_for_each(task, le, &childq.tasks[0])
if (task == parent)
return parent;
@@ -946,7 +1092,7 @@ static DECLARE_MUTEX_LOCKED(rpciod_runni
static inline int
rpciod_task_pending(void)
{
- return !list_empty(&schedq.tasks);
+ return !list_empty(&schedq.tasks[0]);
}
@@ -969,27 +1115,41 @@ rpciod(void *ptr)
allow_signal(SIGKILL);
dprintk("RPC: rpciod starting (pid %d)\n", rpciod_pid);
+ spin_lock_bh(&rpc_queue_lock);
while (rpciod_users) {
+ DEFINE_WAIT(wait);
if (signalled()) {
+ spin_unlock_bh(&rpc_queue_lock);
rpciod_killall();
flush_signals(current);
+ spin_lock_bh(&rpc_queue_lock);
}
__rpc_schedule();
- if (current->flags & PF_FREEZE)
+ if (current->flags & PF_FREEZE) {
+ spin_unlock_bh(&rpc_queue_lock);
refrigerator(PF_IOTHREAD);
+ spin_lock_bh(&rpc_queue_lock);
+ }
if (++rounds >= 64) { /* safeguard */
+ spin_unlock_bh(&rpc_queue_lock);
schedule();
rounds = 0;
+ spin_lock_bh(&rpc_queue_lock);
}
- if (!rpciod_task_pending()) {
- dprintk("RPC: rpciod back to sleep\n");
- wait_event_interruptible(rpciod_idle, rpciod_task_pending());
- dprintk("RPC: switch to rpciod\n");
+ dprintk("RPC: rpciod back to sleep\n");
+ prepare_to_wait(&rpciod_idle, &wait, TASK_INTERRUPTIBLE);
+ if (!rpciod_task_pending() && !signalled()) {
+ spin_unlock_bh(&rpc_queue_lock);
+ schedule();
rounds = 0;
+ spin_lock_bh(&rpc_queue_lock);
}
+ finish_wait(&rpciod_idle, &wait);
+ dprintk("RPC: switch to rpciod\n");
}
+ spin_unlock_bh(&rpc_queue_lock);
dprintk("RPC: rpciod shutdown commences\n");
if (!list_empty(&all_tasks)) {
@@ -1013,7 +1173,9 @@ rpciod_killall(void)
while (!list_empty(&all_tasks)) {
clear_thread_flag(TIF_SIGPENDING);
rpc_killall_tasks(NULL);
+ spin_lock_bh(&rpc_queue_lock);
__rpc_schedule();
+ spin_unlock_bh(&rpc_queue_lock);
if (!list_empty(&all_tasks)) {
dprintk("rpciod_killall: waiting for tasks to exit\n");
yield();
--- diff/net/sunrpc/xprt.c 2004-03-16 11:24:02.000000000 +0000
+++ source/net/sunrpc/xprt.c 2004-03-16 11:27:48.710782496 +0000
@@ -782,8 +782,6 @@ udp_data_ready(struct sock *sk, int len)
dropit:
skb_free_datagram(sk, skb);
out:
- if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
- wake_up_interruptible(sk->sk_sleep);
read_unlock(&sk->sk_callback_lock);
}
@@ -1043,8 +1041,6 @@ tcp_state_change(struct sock *sk)
break;
}
out:
- if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
- wake_up_interruptible_all(sk->sk_sleep);
read_unlock(&sk->sk_callback_lock);
}
@@ -1084,8 +1080,6 @@ xprt_write_space(struct sock *sk)
if (xprt->snd_task && xprt->snd_task->tk_rpcwait == &xprt->pending)
rpc_wake_up_task(xprt->snd_task);
spin_unlock_bh(&xprt->sock_lock);
- if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
- wake_up_interruptible(sk->sk_sleep);
out:
read_unlock(&sk->sk_callback_lock);
}
@@ -1464,10 +1458,10 @@ xprt_setup(int proto, struct sockaddr_in
} else
xprt_default_timeout(&xprt->timeout, xprt->prot);
- INIT_RPC_WAITQ(&xprt->pending, "xprt_pending");
- INIT_RPC_WAITQ(&xprt->sending, "xprt_sending");
- INIT_RPC_WAITQ(&xprt->resend, "xprt_resend");
- INIT_RPC_WAITQ(&xprt->backlog, "xprt_backlog");
+ rpc_init_wait_queue(&xprt->pending, "xprt_pending");
+ rpc_init_wait_queue(&xprt->sending, "xprt_sending");
+ rpc_init_wait_queue(&xprt->resend, "xprt_resend");
+ rpc_init_priority_wait_queue(&xprt->backlog, "xprt_backlog");
/* initialize free list */
for (req = &xprt->slot[entries-1]; req >= &xprt->slot[0]; req--)
@@ -1626,8 +1620,7 @@ xprt_shutdown(struct rpc_xprt *xprt)
rpc_wake_up(&xprt->resend);
rpc_wake_up(&xprt->pending);
rpc_wake_up(&xprt->backlog);
- if (waitqueue_active(&xprt->cong_wait))
- wake_up(&xprt->cong_wait);
+ wake_up(&xprt->cong_wait);
del_timer_sync(&xprt->timer);
}
@@ -1637,8 +1630,7 @@ xprt_shutdown(struct rpc_xprt *xprt)
int
xprt_clear_backlog(struct rpc_xprt *xprt) {
rpc_wake_up_next(&xprt->backlog);
- if (waitqueue_active(&xprt->cong_wait))
- wake_up(&xprt->cong_wait);
+ wake_up(&xprt->cong_wait);
return 1;
}
--- diff/scripts/Makefile 2004-03-11 10:20:29.000000000 +0000
+++ source/scripts/Makefile 2004-03-16 11:27:48.742777632 +0000
@@ -2,14 +2,10 @@
# scripts contains sources for various helper programs used throughout
# the kernel for the build process.
# ---------------------------------------------------------------------------
-# fix-dep: Used to generate dependency information during build process
-# split-include: Divide all config symbols up in a number of files in
-# include/config/...
# docproc: Preprocess .tmpl file in order to generate .sgml docs
# conmakehash: Create arrays for initializing the kernel console tables
-host-progs := fixdep split-include conmakehash docproc kallsyms modpost \
- mk_elfconfig pnmtologo bin2c
+host-progs := conmakehash kallsyms modpost mk_elfconfig pnmtologo bin2c
always := $(host-progs) empty.o
modpost-objs := modpost.o file2alias.o sumversion.o
@@ -17,10 +13,7 @@ modpost-objs := modpost.o file2alias.o s
subdir-$(CONFIG_MODVERSIONS) += genksyms
# Let clean descend into subdirs
-subdir- += lxdialog kconfig
-
-# fixdep is needed to compile other host programs
-$(addprefix $(obj)/,$(filter-out fixdep,$(always)) $(subdir-y)): $(obj)/fixdep
+subdir- += basic lxdialog kconfig
# dependencies on generated files need to be listed explicitly
--- diff/scripts/Makefile.build 2004-03-11 10:20:29.000000000 +0000
+++ source/scripts/Makefile.build 2004-03-16 11:27:48.742777632 +0000
@@ -162,7 +162,7 @@ define rule_cc_o_c
$(if $($(quiet)cmd_cc_o_c),echo ' $($(quiet)cmd_cc_o_c)';) \
$(cmd_cc_o_c); \
$(cmd_modversions) \
- scripts/fixdep $(depfile) $@ '$(cmd_cc_o_c)' > $(@D)/.$(@F).tmp; \
+ scripts/basic/fixdep $(depfile) $@ '$(cmd_cc_o_c)' > $(@D)/.$(@F).tmp; \
rm -f $(depfile); \
mv -f $(@D)/.$(@F).tmp $(@D)/.$(@F).cmd
endef
--- diff/scripts/Makefile.lib 2004-02-18 08:54:13.000000000 +0000
+++ source/scripts/Makefile.lib 2004-03-16 11:27:48.743777480 +0000
@@ -249,7 +249,7 @@ if_changed_dep = $(if $(strip $? $(filte
@set -e; \
$(if $($(quiet)cmd_$(1)),echo ' $(subst ','\'',$($(quiet)cmd_$(1)))';) \
$(cmd_$(1)); \
- scripts/fixdep $(depfile) $@ '$(subst $$,$$$$,$(subst ','\'',$(cmd_$(1))))' > $(@D)/.$(@F).tmp; \
+ scripts/basic/fixdep $(depfile) $@ '$(subst $$,$$$$,$(subst ','\'',$(cmd_$(1))))' > $(@D)/.$(@F).tmp; \
rm -f $(depfile); \
mv -f $(@D)/.$(@F).tmp $(@D)/.$(@F).cmd)
--- diff/scripts/modpost.c 2004-03-16 11:24:02.000000000 +0000
+++ source/scripts/modpost.c 2004-03-16 11:27:48.744777328 +0000
@@ -11,6 +11,7 @@
* Usage: modpost vmlinux module1.o module2.o ...
*/
+#include <ctype.h>
#include "modpost.h"
/* Are we using CONFIG_MODVERSIONS? */
@@ -44,8 +45,6 @@ warn(const char *fmt, ...)
va_end(arglist);
}
-#define NOFAIL(ptr) do_nofail((ptr), __FILE__, __LINE__, #ptr)
-
void *do_nofail(void *ptr, const char *file, int line, const char *expr)
{
if (!ptr) {
@@ -63,21 +62,19 @@ struct module *
new_module(char *modname)
{
struct module *mod;
- char *p;
- size_t len;
+ char *p, *s;
mod = NOFAIL(malloc(sizeof(*mod)));
memset(mod, 0, sizeof(*mod));
p = NOFAIL(strdup(modname));
- len = strlen(p);
-
/* strip trailing .o */
- if (len > 2 && p[len-2] == '.' && p[len-1] == 'o')
- p[len -2] = '\0';
+ if ((s = strrchr(p, '.')) != NULL)
+ if (strcmp(s, ".o") == 0)
+ *s = '\0';
/* add to list */
- mod->name = NOFAIL(strdup(p));
+ mod->name = p;
mod->next = modules;
modules = mod;
@@ -209,6 +206,42 @@ grab_file(const char *filename, unsigned
return map;
}
+/*
+ Return a copy of the next line in a mmap'ed file.
+ spaces in the beginning of the line is trimmed away.
+ Return a pointer to a static buffer.
+*/
+char*
+get_next_line(unsigned long *pos, void *file, unsigned long size)
+{
+ static char line[4096];
+ int skip = 1;
+ size_t len = 0;
+ char *p = (char *)file + *pos;
+ char *s = line;
+
+ for (; *pos < size ; (*pos)++)
+ {
+ if (skip && isspace(*p)) {
+ p++;
+ continue;
+ }
+ skip = 0;
+ if (*p != '\n' && (*pos < size)) {
+ len++;
+ *s++ = *p++;
+ if (len > 4095)
+ break; /* Too long, stop */
+ } else {
+ /* End of string */
+ *s = '\0';
+ return line;
+ }
+ }
+ /* End of buffer */
+ return NULL;
+}
+
void
release_file(void *file, unsigned long size)
{
--- diff/scripts/modpost.h 2004-03-11 10:20:29.000000000 +0000
+++ source/scripts/modpost.h 2004-03-16 11:27:48.744777328 +0000
@@ -53,6 +53,9 @@ static inline void __endian(const void *
#endif
+#define NOFAIL(ptr) do_nofail((ptr), __FILE__, __LINE__, #ptr)
+void *do_nofail(void *ptr, const char *file, int line, const char *expr);
+
struct buffer {
char *p;
int pos;
@@ -95,4 +98,5 @@ void maybe_frob_version(const char *modf
unsigned long modinfo_offset);
void *grab_file(const char *filename, unsigned long *size);
+char* get_next_line(unsigned long *pos, void *file, unsigned long size);
void release_file(void *file, unsigned long size);
--- diff/scripts/sumversion.c 2004-03-11 10:20:29.000000000 +0000
+++ source/scripts/sumversion.c 2004-03-16 11:27:48.745777176 +0000
@@ -323,12 +323,12 @@ static int parse_file(const char *fname,
* figure out source file. */
static int parse_source_files(const char *objfile, struct md4_ctx *md)
{
- char *cmd, *file, *p, *end;
+ char *cmd, *file, *line, *dir;
const char *base;
- unsigned long flen;
- int dirlen, ret = 0;
+ unsigned long flen, pos = 0;
+ int dirlen, ret = 0, check_files = 0;
- cmd = malloc(strlen(objfile) + sizeof("..cmd"));
+ cmd = NOFAIL(malloc(strlen(objfile) + sizeof("..cmd")));
base = strrchr(objfile, '/');
if (base) {
@@ -339,6 +339,9 @@ static int parse_source_files(const char
dirlen = 0;
sprintf(cmd, ".%s.cmd", objfile);
}
+ dir = NOFAIL(malloc(dirlen + 1));
+ strncpy(dir, objfile, dirlen);
+ dir[dirlen] = '\0';
file = grab_file(cmd, &flen);
if (!file) {
@@ -357,48 +360,38 @@ static int parse_source_files(const char
Sum all files in the same dir or subdirs.
*/
- /* Strictly illegal: file is not nul terminated. */
- p = strstr(file, "\ndeps_");
- if (!p) {
- fprintf(stderr, "Warning: could not find deps_ line in %s\n",
- cmd);
- goto out_file;
- }
- p = strstr(p, ":=");
- if (!p) {
- fprintf(stderr, "Warning: could not find := line in %s\n",
- cmd);
- goto out_file;
- }
- p += strlen(":=");
- p += strspn(p, " \\\n");
+ while ((line = get_next_line(&pos, file, flen)) != NULL) {
+ char* p = line;
+ if (strncmp(line, "deps_", sizeof("deps_")-1) == 0) {
+ check_files = 1;
+ continue;
+ }
+ if (!check_files)
+ continue;
- end = strstr(p, "\n\n");
- if (!end) {
- fprintf(stderr, "Warning: could not find end line in %s\n",
- cmd);
- goto out_file;
- }
-
- while (p < end) {
- unsigned int len;
-
- len = strcspn(p, " \\\n");
- if (memcmp(objfile, p, dirlen) == 0) {
- char source[len + 1];
-
- memcpy(source, p, len);
- source[len] = '\0';
- printf("parsing %s\n", source);
- if (!parse_file(source, md)) {
+ /* Continue until line does not end with '\' */
+ if ( *(p + strlen(p)-1) != '\\')
+ break;
+ /* Terminate line at first space, to get rid of final ' \' */
+ while (*p) {
+ if isspace(*p) {
+ *p = '\0';
+ break;
+ }
+ p++;
+ }
+
+ /* Check if this file is in same dir as objfile */
+ if ((strstr(line, dir)+strlen(dir)-1) == strrchr(line, '/')) {
+ if (!parse_file(line, md)) {
fprintf(stderr,
"Warning: could not open %s: %s\n",
- source, strerror(errno));
+ line, strerror(errno));
goto out_file;
}
+
}
- p += len;
- p += strspn(p, " \\\n");
+
}
/* Everyone parsed OK */
@@ -406,6 +399,7 @@ static int parse_source_files(const char
out_file:
release_file(file, flen);
out:
+ free(dir);
free(cmd);
return ret;
}
--- diff/scripts/ver_linux 2003-10-09 08:47:34.000000000 +0000
+++ source/scripts/ver_linux 2004-03-16 11:27:48.746777024 +0000
@@ -21,8 +21,9 @@ gcc --version 2>&1| grep gcc | awk \
make --version 2>&1 | awk -F, '{print $1}' | awk \
'/GNU Make/{print "Gnu make ",$NF}'
-ld -v 2>&1 | awk -F\) '{print $1}' | awk \
- '/BFD/{print "binutils ",$NF}'
+ld -v | awk -F\) '{print $1}' | awk \
+'/BFD/{print "binutils ",$NF} \
+/^GNU/{print "binutils ",$4}'
fdformat --version | awk -F\- '{print "util-linux ", $NF}'
--- diff/security/selinux/avc.c 2004-02-18 08:54:13.000000000 +0000
+++ source/security/selinux/avc.c 2004-03-16 11:27:48.747776872 +0000
@@ -22,6 +22,7 @@
#include <linux/un.h>
#include <net/af_unix.h>
#include <linux/ip.h>
+#include <linux/audit.h>
#include "avc.h"
#include "avc_ss.h"
#include "class_to_string.h"
@@ -66,14 +67,10 @@ struct avc_callback_node {
};
static spinlock_t avc_lock = SPIN_LOCK_UNLOCKED;
-static spinlock_t avc_log_lock = SPIN_LOCK_UNLOCKED;
static struct avc_node *avc_node_freelist = NULL;
static struct avc_cache avc_cache;
-static char *avc_audit_buffer = NULL;
static unsigned avc_cache_stats[AVC_NSTATS];
static struct avc_callback_node *avc_callbacks = NULL;
-static unsigned int avc_log_level = 4; /* default: KERN_WARNING */
-static char avc_level_string[4] = "< >";
static inline int avc_hash(u32 ssid, u32 tsid, u16 tclass)
{
@@ -85,14 +82,14 @@ static inline int avc_hash(u32 ssid, u32
* @tclass: target security class
* @av: access vector
*/
-void avc_dump_av(u16 tclass, u32 av)
+void avc_dump_av(struct audit_buffer *ab, u16 tclass, u32 av)
{
char **common_pts = 0;
u32 common_base = 0;
int i, i2, perm;
if (av == 0) {
- printk(" null");
+ audit_log_format(ab, " null");
return;
}
@@ -104,12 +101,12 @@ void avc_dump_av(u16 tclass, u32 av)
}
}
- printk(" {");
+ audit_log_format(ab, " {");
i = 0;
perm = 1;
while (perm < common_base) {
if (perm & av)
- printk(" %s", common_pts[i]);
+ audit_log_format(ab, " %s", common_pts[i]);
i++;
perm <<= 1;
}
@@ -122,13 +119,14 @@ void avc_dump_av(u16 tclass, u32 av)
break;
}
if (i2 < ARRAY_SIZE(av_perm_to_string))
- printk(" %s", av_perm_to_string[i2].name);
+ audit_log_format(ab, " %s",
+ av_perm_to_string[i2].name);
}
i++;
perm <<= 1;
}
- printk(" }");
+ audit_log_format(ab, " }");
}
/**
@@ -137,7 +135,7 @@ void avc_dump_av(u16 tclass, u32 av)
* @tsid: target security identifier
* @tclass: target security class
*/
-void avc_dump_query(u32 ssid, u32 tsid, u16 tclass)
+void avc_dump_query(struct audit_buffer *ab, u32 ssid, u32 tsid, u16 tclass)
{
int rc;
char *scontext;
@@ -145,20 +143,20 @@ void avc_dump_query(u32 ssid, u32 tsid,
rc = security_sid_to_context(ssid, &scontext, &scontext_len);
if (rc)
- printk("ssid=%d", ssid);
+ audit_log_format(ab, "ssid=%d", ssid);
else {
- printk("scontext=%s", scontext);
+ audit_log_format(ab, "scontext=%s", scontext);
kfree(scontext);
}
rc = security_sid_to_context(tsid, &scontext, &scontext_len);
if (rc)
- printk(" tsid=%d", tsid);
+ audit_log_format(ab, " tsid=%d", tsid);
else {
- printk(" tcontext=%s", scontext);
+ audit_log_format(ab, " tcontext=%s", scontext);
kfree(scontext);
}
- printk(" tclass=%s", class_to_string[tclass]);
+ audit_log_format(ab, " tclass=%s", class_to_string[tclass]);
}
/**
@@ -192,11 +190,7 @@ void __init avc_init(void)
avc_node_freelist = new;
}
- avc_audit_buffer = (char *)__get_free_page(GFP_ATOMIC);
- if (!avc_audit_buffer)
- panic("AVC: unable to allocate audit buffer\n");
-
- avc_level_string[1] = '0' + avc_log_level;
+ audit_log(current->audit_context, "AVC INITIALIZED\n");
}
#if 0
@@ -418,12 +412,13 @@ out:
return rc;
}
-static inline void avc_print_ipv4_addr(u32 addr, u16 port, char *name1, char *name2)
+static inline void avc_print_ipv4_addr(struct audit_buffer *ab, u32 addr,
+ u16 port, char *name1, char *name2)
{
if (addr)
- printk(" %s=%d.%d.%d.%d", name1, NIPQUAD(addr));
+ audit_log_format(ab, " %s=%d.%d.%d.%d", name1, NIPQUAD(addr));
if (port)
- printk(" %s=%d", name2, ntohs(port));
+ audit_log_format(ab, " %s=%d", name2, ntohs(port));
}
/*
@@ -503,9 +498,8 @@ void avc_audit(u32 ssid, u32 tsid,
{
struct task_struct *tsk = current;
struct inode *inode = NULL;
- char *p;
u32 denied, audited;
- unsigned long flags;
+ struct audit_buffer *ab;
denied = requested & ~avd->allowed;
if (denied) {
@@ -523,19 +517,18 @@ void avc_audit(u32 ssid, u32 tsid,
if (!check_avc_ratelimit())
return;
- /* prevent overlapping printks */
- spin_lock_irqsave(&avc_log_lock,flags);
-
- printk("%s\n", avc_level_string);
- printk("%savc: %s ", avc_level_string, denied ? "denied" : "granted");
- avc_dump_av(tclass,audited);
- printk(" for ");
+ ab = audit_log_start(current->audit_context);
+ if (!ab)
+ return; /* audit_panic has been called */
+ audit_log_format(ab, "avc: %s ", denied ? "denied" : "granted");
+ avc_dump_av(ab, tclass,audited);
+ audit_log_format(ab, " for ");
if (a && a->tsk)
tsk = a->tsk;
if (tsk && tsk->pid) {
struct mm_struct *mm;
struct vm_area_struct *vma;
- printk(" pid=%d", tsk->pid);
+ audit_log_format(ab, " pid=%d", tsk->pid);
if (tsk == current)
mm = current->mm;
else
@@ -546,11 +539,9 @@ void avc_audit(u32 ssid, u32 tsid,
while (vma) {
if ((vma->vm_flags & VM_EXECUTABLE) &&
vma->vm_file) {
- p = d_path(vma->vm_file->f_dentry,
- vma->vm_file->f_vfsmnt,
- avc_audit_buffer,
- PAGE_SIZE);
- printk(" exe=%s", p);
+ audit_log_d_path(ab, "exe=",
+ vma->vm_file->f_dentry,
+ vma->vm_file->f_vfsmnt);
break;
}
vma = vma->vm_next;
@@ -560,29 +551,25 @@ void avc_audit(u32 ssid, u32 tsid,
if (tsk != current)
mmput(mm);
} else {
- printk(" comm=%s", tsk->comm);
+ audit_log_format(ab, " comm=%s", tsk->comm);
}
}
if (a) {
switch (a->type) {
case AVC_AUDIT_DATA_IPC:
- printk(" key=%d", a->u.ipc_id);
+ audit_log_format(ab, " key=%d", a->u.ipc_id);
break;
case AVC_AUDIT_DATA_CAP:
- printk(" capability=%d", a->u.cap);
+ audit_log_format(ab, " capability=%d", a->u.cap);
break;
case AVC_AUDIT_DATA_FS:
if (a->u.fs.dentry) {
struct dentry *dentry = a->u.fs.dentry;
if (a->u.fs.mnt) {
- p = d_path(dentry,
- a->u.fs.mnt,
- avc_audit_buffer,
- PAGE_SIZE);
- if (p)
- printk(" path=%s", p);
+ audit_log_d_path(ab, "path=", dentry, a->u.fs.mnt);
} else {
- printk(" name=%s", dentry->d_name.name);
+ audit_log_format(ab, " name=%s",
+ dentry->d_name.name);
}
inode = dentry->d_inode;
} else if (a->u.fs.inode) {
@@ -590,13 +577,15 @@ void avc_audit(u32 ssid, u32 tsid,
inode = a->u.fs.inode;
dentry = d_find_alias(inode);
if (dentry) {
- printk(" name=%s", dentry->d_name.name);
+ audit_log_format(ab, " name=%s",
+ dentry->d_name.name);
dput(dentry);
}
}
if (inode)
- printk(" dev=%s ino=%ld",
- inode->i_sb->s_id, inode->i_ino);
+ audit_log_format(ab, " dev=%s ino=%ld",
+ inode->i_sb->s_id,
+ inode->i_ino);
break;
case AVC_AUDIT_DATA_NET:
if (a->u.net.sk) {
@@ -607,53 +596,45 @@ void avc_audit(u32 ssid, u32 tsid,
switch (sk->sk_family) {
case AF_INET:
inet = inet_sk(sk);
- avc_print_ipv4_addr(inet->rcv_saddr,
- inet->sport,
- "laddr", "lport");
- avc_print_ipv4_addr(inet->daddr,
- inet->dport,
- "faddr", "fport");
+ avc_print_ipv4_addr(ab, inet->rcv_saddr,
+ inet->sport,
+ "laddr", "lport");
+ avc_print_ipv4_addr(ab, inet->daddr,
+ inet->dport,
+ "faddr", "fport");
break;
case AF_UNIX:
u = unix_sk(sk);
if (u->dentry) {
- p = d_path(u->dentry,
- u->mnt,
- avc_audit_buffer,
- PAGE_SIZE);
- printk(" path=%s", p);
+ audit_log_d_path(ab, "path=",
+ u->dentry, u->mnt);
} else if (u->addr) {
- p = avc_audit_buffer;
- memcpy(p,
- u->addr->name->sun_path,
- u->addr->len-sizeof(short));
- if (*p == 0) {
- *p = '@';
- p += u->addr->len-sizeof(short);
- *p = 0;
- }
- printk(" path=%s",
- avc_audit_buffer);
+ int len = u->addr->len-sizeof(short);
+ char *p = &u->addr->name->sun_path[0];
+ if (*p)
+ audit_log_format(ab, "path=%*.*s",
+ len, len, p);
+ else
+ audit_log_format(ab, "path=@%*.*s",
+ len-1, len-1, p+1);
}
break;
}
}
- avc_print_ipv4_addr(a->u.net.saddr, a->u.net.sport,
- "saddr", "src");
- avc_print_ipv4_addr(a->u.net.daddr, a->u.net.dport,
- "daddr", "dest");
+ avc_print_ipv4_addr(ab, a->u.net.saddr, a->u.net.sport,
+ "saddr", "src");
+ avc_print_ipv4_addr(ab, a->u.net.daddr, a->u.net.dport,
+ "daddr", "dest");
if (a->u.net.netif)
- printk(" netif=%s", a->u.net.netif);
+ audit_log_format(ab, " netif=%s", a->u.net.netif);
break;
}
}
- printk(" ");
- avc_dump_query(ssid, tsid, tclass);
- printk("\n");
-
- spin_unlock_irqrestore(&avc_log_lock,flags);
+ audit_log_format(ab, " ");
+ avc_dump_query(ab, ssid, tsid, tclass);
+ audit_log_end(ab);
}
/**
@@ -1083,13 +1064,4 @@ int avc_has_perm(u32 ssid, u32 tsid, u16
return rc;
}
-static int __init avc_log_level_setup(char *str)
-{
- avc_log_level = simple_strtol(str, NULL, 0);
- if (avc_log_level > 7)
- avc_log_level = 7;
- return 1;
-}
-
-__setup("avc_log_level=", avc_log_level_setup);
--- diff/security/selinux/hooks.c 2004-03-16 11:24:02.000000000 +0000
+++ source/security/selinux/hooks.c 2004-03-16 11:27:48.750776416 +0000
@@ -750,6 +750,7 @@ static int inode_doinit_with_dentry(stru
inode->i_ino);
goto out;
}
+ BUG_ON(inode != dentry->d_inode);
len = INITCONTEXTLEN;
context = kmalloc(len, GFP_KERNEL);
--- diff/security/selinux/include/av_perm_to_string.h 2004-02-09 10:36:12.000000000 +0000
+++ source/security/selinux/include/av_perm_to_string.h 2004-03-16 11:27:48.752776112 +0000
@@ -84,6 +84,7 @@ static struct av_perm_to_string av_perm_
{ SECCLASS_SECURITY, SECURITY__COMPUTE_RELABEL, "compute_relabel" },
{ SECCLASS_SECURITY, SECURITY__COMPUTE_USER, "compute_user" },
{ SECCLASS_SECURITY, SECURITY__SETENFORCE, "setenforce" },
+ { SECCLASS_SECURITY, SECURITY__SETBOOL, "setbool" },
{ SECCLASS_SYSTEM, SYSTEM__IPC_INFO, "ipc_info" },
{ SECCLASS_SYSTEM, SYSTEM__SYSLOG_READ, "syslog_read" },
{ SECCLASS_SYSTEM, SYSTEM__SYSLOG_MOD, "syslog_mod" },
--- diff/security/selinux/include/av_permissions.h 2004-02-09 10:36:12.000000000 +0000
+++ source/security/selinux/include/av_permissions.h 2004-03-16 11:27:48.751776264 +0000
@@ -512,6 +512,7 @@
#define SECURITY__COMPUTE_RELABEL 0x00000020UL
#define SECURITY__COMPUTE_USER 0x00000040UL
#define SECURITY__SETENFORCE 0x00000080UL
+#define SECURITY__SETBOOL 0x00000100UL
#define SYSTEM__IPC_INFO 0x00000001UL
#define SYSTEM__SYSLOG_READ 0x00000002UL
--- diff/security/selinux/include/avc.h 2004-02-18 08:54:13.000000000 +0000
+++ source/security/selinux/include/avc.h 2004-03-16 11:27:48.750776416 +0000
@@ -114,9 +114,10 @@ static inline void avc_cache_stats_add(i
/*
* AVC display support
*/
-void avc_dump_av(u16 tclass, u32 av);
-void avc_dump_query(u32 ssid, u32 tsid, u16 tclass);
-void avc_dump_cache(char *tag);
+struct audit_buffer;
+void avc_dump_av(struct audit_buffer *ab, u16 tclass, u32 av);
+void avc_dump_query(struct audit_buffer *ab, u32 ssid, u32 tsid, u16 tclass);
+void avc_dump_cache(struct audit_buffer *ab, char *tag);
/*
* AVC operations
--- diff/security/selinux/include/security.h 2004-02-18 08:54:13.000000000 +0000
+++ source/security/selinux/include/security.h 2004-03-16 11:27:48.753775960 +0000
@@ -2,7 +2,9 @@
* Security server interface.
*
* Author : Stephen Smalley, <sds@epoch.ncsc.mil>
+ *
*/
+
#ifndef _SELINUX_SECURITY_H_
#define _SELINUX_SECURITY_H_
@@ -13,7 +15,8 @@
#define SECCLASS_NULL 0x0000 /* no class */
#define SELINUX_MAGIC 0xf97cff8c
-#define POLICYDB_VERSION 15
+#define POLICYDB_VERSION 16
+#define POLICYDB_VERSION_COMPAT 15
#ifdef CONFIG_SECURITY_SELINUX_BOOTPARAM
extern int selinux_enabled;
@@ -21,6 +24,12 @@ extern int selinux_enabled;
#define selinux_enabled 1
#endif
+#ifdef CONFIG_SECURITY_SELINUX_MLS
+#define selinux_mls_enabled 1
+#else
+#define selinux_mls_enabled 0
+#endif
+
int security_load_policy(void * data, size_t len);
struct av_decision {
--- diff/security/selinux/selinuxfs.c 2004-03-11 10:20:29.000000000 +0000
+++ source/security/selinux/selinuxfs.c 2004-03-16 11:27:48.755775656 +0000
@@ -1,5 +1,16 @@
+/* Updated: Karl MacMillan <kmacmillan@tresys.com>
+ *
+ * Added conditional policy language extensions
+ *
+ * Copyright (C) 2003 - 2004 Tresys Technology, LLC
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, version 2.
+ */
+
#include <linux/config.h>
#include <linux/kernel.h>
+#include <linux/pagemap.h>
#include <linux/slab.h>
#include <linux/vmalloc.h>
#include <linux/fs.h>
@@ -7,6 +18,7 @@
#include <linux/string.h>
#include <linux/security.h>
#include <asm/uaccess.h>
+#include <asm/semaphore.h>
/* selinuxfs pseudo filesystem for exporting the security policy API.
Based on the proc code and the fs/nfsd/nfsctl.c code. */
@@ -16,6 +28,14 @@
#include "avc_ss.h"
#include "security.h"
#include "objsec.h"
+#include "conditional.h"
+
+static DECLARE_MUTEX(sel_sem);
+
+/* global data for booleans */
+static struct dentry *bool_dir = NULL;
+static int bool_num = 0;
+static int *bool_pending_values = NULL;
extern void selnl_notify_setenforce(int val);
@@ -40,7 +60,9 @@ enum sel_inos {
SEL_CREATE, /* compute create labeling decision */
SEL_RELABEL, /* compute relabeling decision */
SEL_USER, /* compute reachable user contexts */
- SEL_POLICYVERS /* return policy version for this kernel */
+ SEL_POLICYVERS, /* return policy version for this kernel */
+ SEL_COMMIT_BOOLS,
+ SEL_MLS /* return if MLS policy is enabled */
};
static ssize_t sel_read_enforce(struct file *filp, char *buf,
@@ -169,24 +191,74 @@ static struct file_operations sel_policy
.read = sel_read_policyvers,
};
+/* declaration for sel_write_load */
+static int sel_make_bools(void);
+
+static ssize_t sel_read_mls(struct file *filp, char *buf,
+ size_t count, loff_t *ppos)
+{
+ char *page;
+ ssize_t length;
+ ssize_t end;
+
+ if (count < 0 || count > PAGE_SIZE)
+ return -EINVAL;
+ if (!(page = (char*)__get_free_page(GFP_KERNEL)))
+ return -ENOMEM;
+ memset(page, 0, PAGE_SIZE);
+
+ length = scnprintf(page, PAGE_SIZE, "%d", selinux_mls_enabled);
+ if (length < 0) {
+ free_page((unsigned long)page);
+ return length;
+ }
+
+ if (*ppos >= length) {
+ free_page((unsigned long)page);
+ return 0;
+ }
+ if (count + *ppos > length)
+ count = length - *ppos;
+ end = count + *ppos;
+ if (copy_to_user(buf, (char *) page + *ppos, count)) {
+ count = -EFAULT;
+ goto out;
+ }
+ *ppos = end;
+out:
+ free_page((unsigned long)page);
+ return count;
+}
+
+static struct file_operations sel_mls_ops = {
+ .read = sel_read_mls,
+};
+
static ssize_t sel_write_load(struct file * file, const char * buf,
size_t count, loff_t *ppos)
{
+ int ret;
ssize_t length;
- void *data;
+ void *data = NULL;
+
+ down(&sel_sem);
length = task_has_security(current, SECURITY__LOAD_POLICY);
if (length)
- return length;
+ goto out;
if (*ppos != 0) {
/* No partial writes. */
- return -EINVAL;
+ length = -EINVAL;
+ goto out;
}
- if ((count < 0) || (count > 64 * 1024 * 1024) || (data = vmalloc(count)) == NULL)
- return -ENOMEM;
+ if ((count < 0) || (count > 64 * 1024 * 1024)
+ || (data = vmalloc(count)) == NULL) {
+ length = -ENOMEM;
+ goto out;
+ }
length = -EFAULT;
if (copy_from_user(data, buf, count) != 0)
@@ -196,8 +268,13 @@ static ssize_t sel_write_load(struct fil
if (length)
goto out;
- length = count;
+ ret = sel_make_bools();
+ if (ret)
+ length = ret;
+ else
+ length = count;
out:
+ up(&sel_sem);
vfree(data);
return length;
}
@@ -601,9 +678,322 @@ out:
return length;
}
+static struct inode *sel_make_inode(struct super_block *sb, int mode)
+{
+ struct inode *ret = new_inode(sb);
+
+ if (ret) {
+ ret->i_mode = mode;
+ ret->i_uid = ret->i_gid = 0;
+ ret->i_blksize = PAGE_CACHE_SIZE;
+ ret->i_blocks = 0;
+ ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME;
+ }
+ return ret;
+}
+
+#define BOOL_INO_OFFSET 30
+
+static ssize_t sel_read_bool(struct file *filep, char *buf,
+ size_t count, loff_t *ppos)
+{
+ char *page = NULL;
+ ssize_t length;
+ ssize_t end;
+ ssize_t ret;
+ int cur_enforcing;
+ struct inode *inode;
+
+ down(&sel_sem);
+
+ ret = -EFAULT;
+
+ /* check to see if this file has been deleted */
+ if (!filep->f_op)
+ goto out;
+
+ if (count < 0 || count > PAGE_SIZE) {
+ ret = -EINVAL;
+ goto out;
+ }
+ if (!(page = (char*)__get_free_page(GFP_KERNEL))) {
+ ret = -ENOMEM;
+ goto out;
+ }
+ memset(page, 0, PAGE_SIZE);
+
+ inode = filep->f_dentry->d_inode;
+ cur_enforcing = security_get_bool_value(inode->i_ino - BOOL_INO_OFFSET);
+ if (cur_enforcing < 0) {
+ ret = cur_enforcing;
+ goto out;
+ }
+
+ length = scnprintf(page, PAGE_SIZE, "%d %d", cur_enforcing,
+ bool_pending_values[inode->i_ino - BOOL_INO_OFFSET]);
+ if (length < 0) {
+ ret = length;
+ goto out;
+ }
+
+ if (*ppos >= length) {
+ ret = 0;
+ goto out;
+ }
+ if (count + *ppos > length)
+ count = length - *ppos;
+ end = count + *ppos;
+ if (copy_to_user(buf, (char *) page + *ppos, count)) {
+ ret = -EFAULT;
+ goto out;
+ }
+ *ppos = end;
+ ret = count;
+out:
+ up(&sel_sem);
+ if (page)
+ free_page((unsigned long)page);
+ return ret;
+}
+
+static ssize_t sel_write_bool(struct file *filep, const char *buf,
+ size_t count, loff_t *ppos)
+{
+ char *page = NULL;
+ ssize_t length = -EFAULT;
+ int new_value;
+ struct inode *inode;
+
+ down(&sel_sem);
+
+ length = task_has_security(current, SECURITY__SETBOOL);
+ if (length)
+ goto out;
+
+ /* check to see if this file has been deleted */
+ if (!filep->f_op)
+ goto out;
+
+ if (count < 0 || count >= PAGE_SIZE) {
+ length = -ENOMEM;
+ goto out;
+ }
+ if (*ppos != 0) {
+ /* No partial writes. */
+ goto out;
+ }
+ page = (char*)__get_free_page(GFP_KERNEL);
+ if (!page)
+ return -ENOMEM;
+ memset(page, 0, PAGE_SIZE);
+
+ if (copy_from_user(page, buf, count))
+ goto out;
+
+ length = -EINVAL;
+ if (sscanf(page, "%d", &new_value) != 1)
+ goto out;
+
+ if (new_value)
+ new_value = 1;
+
+ inode = filep->f_dentry->d_inode;
+ bool_pending_values[inode->i_ino - BOOL_INO_OFFSET] = new_value;
+ length = count;
+
+out:
+ up(&sel_sem);
+ if (page)
+ free_page((unsigned long) page);
+ return length;
+}
+
+static struct file_operations sel_bool_ops = {
+ .read = sel_read_bool,
+ .write = sel_write_bool,
+};
+
+static ssize_t sel_commit_bools_write(struct file *filep, const char *buf,
+ size_t count, loff_t *ppos)
+{
+ char *page = NULL;
+ ssize_t length = -EFAULT;
+ int new_value;
+
+ down(&sel_sem);
+
+ length = task_has_security(current, SECURITY__SETBOOL);
+ if (length)
+ goto out;
+
+ /* check to see if this file has been deleted */
+ if (!filep->f_op)
+ goto out;
+
+ if (count < 0 || count >= PAGE_SIZE) {
+ length = -ENOMEM;
+ goto out;
+ }
+ if (*ppos != 0) {
+ /* No partial writes. */
+ goto out;
+ }
+ page = (char*)__get_free_page(GFP_KERNEL);
+ if (!page)
+ return -ENOMEM;
+
+ memset(page, 0, PAGE_SIZE);
+
+ if (copy_from_user(page, buf, count))
+ goto out;
+
+ length = -EINVAL;
+ if (sscanf(page, "%d", &new_value) != 1)
+ goto out;
+
+ if (new_value) {
+ security_set_bools(bool_num, bool_pending_values);
+ }
+
+ length = count;
+
+out:
+ up(&sel_sem);
+ if (page)
+ free_page((unsigned long) page);
+ return length;
+}
+
+static struct file_operations sel_commit_bools_ops = {
+ .write = sel_commit_bools_write,
+};
+
+/* delete booleans - partial revoke() from
+ * fs/proc/generic.c proc_kill_inodes */
+static void sel_remove_bools(struct dentry *de)
+{
+ struct list_head *p, *node;
+ struct super_block *sb = de->d_sb;
+
+ spin_lock(&dcache_lock);
+ node = de->d_subdirs.next;
+ while (node != &de->d_subdirs) {
+ struct dentry *d = list_entry(node, struct dentry, d_child);
+ list_del_init(node);
+
+ if (d->d_inode) {
+ d = dget_locked(d);
+ spin_unlock(&dcache_lock);
+ d_delete(d);
+ simple_unlink(de->d_inode, d);
+ dput(d);
+ spin_lock(&dcache_lock);
+ }
+ node = de->d_subdirs.next;
+ }
+
+ spin_unlock(&dcache_lock);
+
+ file_list_lock();
+ list_for_each(p, &sb->s_files) {
+ struct file * filp = list_entry(p, struct file, f_list);
+ struct dentry * dentry = filp->f_dentry;
+
+ if (dentry->d_parent != de) {
+ continue;
+ }
+ filp->f_op = NULL;
+ }
+ file_list_unlock();
+}
+
+#define BOOL_DIR_NAME "booleans"
+
+static int sel_make_bools(void)
+{
+ int i, ret = 0;
+ ssize_t len;
+ struct dentry *dentry = NULL;
+ struct dentry *dir = bool_dir;
+ struct inode *inode = NULL;
+ struct inode_security_struct *isec;
+ struct qstr qname;
+ char **names = NULL, *page;
+ int num;
+ int *values = NULL;
+ u32 sid;
+
+ /* remove any existing files */
+ if (bool_pending_values)
+ kfree(bool_pending_values);
+
+ sel_remove_bools(dir);
+
+ if (!(page = (char*)__get_free_page(GFP_KERNEL)))
+ return -ENOMEM;
+ memset(page, 0, PAGE_SIZE);
+
+ ret = security_get_bools(&num, &names, &values);
+ if (ret != 0)
+ goto out;
+
+ for (i = 0; i < num; i++) {
+ qname.name = names[i];
+ qname.len = strlen(qname.name);
+ qname.hash = full_name_hash(qname.name, qname.len);
+ dentry = d_alloc(dir, &qname);
+ if (!dentry) {
+ ret = -ENOMEM;
+ goto err;
+ }
+ inode = sel_make_inode(dir->d_sb, S_IFREG | S_IRUGO | S_IWUSR);
+ if (!inode) {
+ ret = -ENOMEM;
+ goto err;
+ }
+
+ len = snprintf(page, PAGE_SIZE, "/%s/%s", BOOL_DIR_NAME, names[i]);
+ if (len < 0) {
+ ret = -EINVAL;
+ goto err;
+ } else if (len >= PAGE_SIZE) {
+ ret = -ENAMETOOLONG;
+ goto err;
+ }
+ isec = (struct inode_security_struct*)inode->i_security;
+ if ((ret = security_genfs_sid("selinuxfs", page, SECCLASS_FILE, &sid)))
+ goto err;
+ isec->sid = sid;
+ isec->initialized = 1;
+ inode->i_fop = &sel_bool_ops;
+ inode->i_ino = i + BOOL_INO_OFFSET;
+ d_add(dentry, inode);
+ }
+ bool_num = num;
+ bool_pending_values = values;
+out:
+ free_page((unsigned long)page);
+ if (names) {
+ for (i = 0; i < num; i++) {
+ if (names[i])
+ kfree(names[i]);
+ }
+ kfree(names);
+ }
+ return ret;
+err:
+ d_genocide(dir);
+ ret = -ENOMEM;
+ goto out;
+}
static int sel_fill_super(struct super_block * sb, void * data, int silent)
{
+ int ret;
+ struct dentry *dentry;
+ struct inode *inode;
+ struct qstr qname;
+
static struct tree_descr selinux_files[] = {
[SEL_LOAD] = {"load", &sel_load_ops, S_IRUSR|S_IWUSR},
[SEL_ENFORCE] = {"enforce", &sel_enforce_ops, S_IRUGO|S_IWUSR},
@@ -613,9 +1003,37 @@ static int sel_fill_super(struct super_b
[SEL_RELABEL] = {"relabel", &transaction_ops, S_IRUGO|S_IWUGO},
[SEL_USER] = {"user", &transaction_ops, S_IRUGO|S_IWUGO},
[SEL_POLICYVERS] = {"policyvers", &sel_policyvers_ops, S_IRUGO},
+ [SEL_COMMIT_BOOLS] = {"commit_pending_bools", &sel_commit_bools_ops, S_IWUSR},
+ [SEL_MLS] = {"mls", &sel_mls_ops, S_IRUGO},
/* last one */ {""}
};
- return simple_fill_super(sb, SELINUX_MAGIC, selinux_files);
+ ret = simple_fill_super(sb, SELINUX_MAGIC, selinux_files);
+ if (ret)
+ return ret;
+
+ qname.name = BOOL_DIR_NAME;
+ qname.len = strlen(qname.name);
+ qname.hash = full_name_hash(qname.name, qname.len);
+ dentry = d_alloc(sb->s_root, &qname);
+ if (!dentry)
+ return -ENOMEM;
+
+ inode = sel_make_inode(sb, S_IFDIR | S_IRUGO | S_IXUGO);
+ if (!inode)
+ goto out;
+ inode->i_op = &simple_dir_inode_operations;
+ inode->i_fop = &simple_dir_operations;
+ d_add(dentry, inode);
+ bool_dir = dentry;
+ ret = sel_make_bools();
+ if (ret)
+ goto out;
+
+ return 0;
+out:
+ dput(dentry);
+ printk(KERN_ERR "security: error creating conditional out_dput\n");
+ return -ENOMEM;
}
static struct super_block *sel_get_sb(struct file_system_type *fs_type,
--- diff/security/selinux/ss/Makefile 2004-02-09 10:36:12.000000000 +0000
+++ source/security/selinux/ss/Makefile 2004-03-16 11:27:48.760774896 +0000
@@ -5,7 +5,7 @@
EXTRA_CFLAGS += -Isecurity/selinux/include
obj-y := ss.o
-ss-y := ebitmap.o hashtab.o symtab.o sidtab.o avtab.o policydb.o services.o
+ss-y := ebitmap.o hashtab.o symtab.o sidtab.o avtab.o policydb.o services.o conditional.o
ss-$(CONFIG_SECURITY_SELINUX_MLS) += mls.o
--- diff/security/selinux/ss/avtab.c 2004-02-18 08:54:13.000000000 +0000
+++ source/security/selinux/ss/avtab.c 2004-03-16 11:27:48.757775352 +0000
@@ -3,10 +3,22 @@
*
* Author : Stephen Smalley, <sds@epoch.ncsc.mil>
*/
+
+/* Updated: Frank Mayer <mayerf@tresys.com> and Karl MacMillan <kmacmillan@tresys.com>
+ *
+ * Added conditional policy language extensions
+ *
+ * Copyright (C) 2003 Tresys Technology, LLC
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, version 2.
+ */
+
#include <linux/kernel.h>
#include <linux/slab.h>
#include <linux/vmalloc.h>
#include <linux/errno.h>
+
#include "avtab.h"
#include "policydb.h"
@@ -16,6 +28,29 @@
(keyp->source_type << 9)) & \
AVTAB_HASH_MASK)
+static struct avtab_node*
+avtab_insert_node(struct avtab *h, int hvalue, struct avtab_node * prev, struct avtab_node * cur,
+ struct avtab_key *key, struct avtab_datum *datum)
+{
+ struct avtab_node * newnode;
+ newnode = (struct avtab_node *) kmalloc(sizeof(struct avtab_node),GFP_KERNEL);
+ if (newnode == NULL)
+ return NULL;
+ memset(newnode, 0, sizeof(struct avtab_node));
+ newnode->key = *key;
+ newnode->datum = *datum;
+ if (prev) {
+ newnode->next = prev->next;
+ prev->next = newnode;
+ } else {
+ newnode->next = h->htable[hvalue];
+ h->htable[hvalue] = newnode;
+ }
+
+ h->nel++;
+ return newnode;
+}
+
int avtab_insert(struct avtab *h, struct avtab_key *key, struct avtab_datum *datum)
{
int hvalue;
@@ -44,24 +79,48 @@ int avtab_insert(struct avtab *h, struct
break;
}
- newnode = kmalloc(sizeof(*newnode), GFP_KERNEL);
- if (newnode == NULL)
+ newnode = avtab_insert_node(h, hvalue, prev, cur, key, datum);
+ if(!newnode)
return -ENOMEM;
- memset(newnode, 0, sizeof(*newnode));
- newnode->key = *key;
- newnode->datum = *datum;
- if (prev) {
- newnode->next = prev->next;
- prev->next = newnode;
- } else {
- newnode->next = h->htable[hvalue];
- h->htable[hvalue] = newnode;
- }
- h->nel++;
return 0;
}
+/* Unlike avtab_insert(), this function allow multiple insertions of the same
+ * key/specified mask into the table, as needed by the conditional avtab.
+ * It also returns a pointer to the node inserted.
+ */
+struct avtab_node *
+avtab_insert_nonunique(struct avtab * h, struct avtab_key * key, struct avtab_datum * datum)
+{
+ int hvalue;
+ struct avtab_node *prev, *cur, *newnode;
+
+ if (!h)
+ return NULL;
+ hvalue = AVTAB_HASH(key);
+ for (prev = NULL, cur = h->htable[hvalue];
+ cur;
+ prev = cur, cur = cur->next) {
+ if (key->source_type == cur->key.source_type &&
+ key->target_type == cur->key.target_type &&
+ key->target_class == cur->key.target_class &&
+ (datum->specified & cur->datum.specified))
+ break;
+ if (key->source_type < cur->key.source_type)
+ break;
+ if (key->source_type == cur->key.source_type &&
+ key->target_type < cur->key.target_type)
+ break;
+ if (key->source_type == cur->key.source_type &&
+ key->target_type == cur->key.target_type &&
+ key->target_class < cur->key.target_class)
+ break;
+ }
+ newnode = avtab_insert_node(h, hvalue, prev, cur, key, datum);
+
+ return newnode;
+}
struct avtab_datum *avtab_search(struct avtab *h, struct avtab_key *key, int specified)
{
@@ -93,6 +152,67 @@ struct avtab_datum *avtab_search(struct
return NULL;
}
+/* This search function returns a node pointer, and can be used in
+ * conjunction with avtab_search_next_node()
+ */
+struct avtab_node*
+avtab_search_node(struct avtab *h, struct avtab_key *key, int specified)
+{
+ int hvalue;
+ struct avtab_node *cur;
+
+ if (!h)
+ return NULL;
+
+ hvalue = AVTAB_HASH(key);
+ for (cur = h->htable[hvalue]; cur; cur = cur->next) {
+ if (key->source_type == cur->key.source_type &&
+ key->target_type == cur->key.target_type &&
+ key->target_class == cur->key.target_class &&
+ (specified & cur->datum.specified))
+ return cur;
+
+ if (key->source_type < cur->key.source_type)
+ break;
+ if (key->source_type == cur->key.source_type &&
+ key->target_type < cur->key.target_type)
+ break;
+ if (key->source_type == cur->key.source_type &&
+ key->target_type == cur->key.target_type &&
+ key->target_class < cur->key.target_class)
+ break;
+ }
+ return NULL;
+}
+
+struct avtab_node*
+avtab_search_node_next(struct avtab_node *node, int specified)
+{
+ struct avtab_node *cur;
+
+ if (!node)
+ return NULL;
+
+ for (cur = node->next; cur; cur = cur->next) {
+ if (node->key.source_type == cur->key.source_type &&
+ node->key.target_type == cur->key.target_type &&
+ node->key.target_class == cur->key.target_class &&
+ (specified & cur->datum.specified))
+ return cur;
+
+ if (node->key.source_type < cur->key.source_type)
+ break;
+ if (node->key.source_type == cur->key.source_type &&
+ node->key.target_type < cur->key.target_type)
+ break;
+ if (node->key.source_type == cur->key.source_type &&
+ node->key.target_type == cur->key.target_type &&
+ node->key.target_class < cur->key.target_class)
+ break;
+ }
+ return NULL;
+}
+
void avtab_destroy(struct avtab *h)
{
int i;
@@ -179,13 +299,72 @@ void avtab_hash_eval(struct avtab *h, ch
max_chain_len);
}
+int avtab_read_item(void *fp, struct avtab_datum *avdatum, struct avtab_key *avkey)
+{
+ __u32 *buf;
+ __u32 items, items2;
+
+ memset(avkey, 0, sizeof(struct avtab_key));
+ memset(avdatum, 0, sizeof(struct avtab_datum));
+
+ buf = next_entry(fp, sizeof(__u32));
+ if (!buf) {
+ printk(KERN_ERR "security: avtab: truncated entry\n");
+ goto bad;
+ }
+ items2 = le32_to_cpu(buf[0]);
+ buf = next_entry(fp, sizeof(__u32)*items2);
+ if (!buf) {
+ printk(KERN_ERR "security: avtab: truncated entry\n");
+ goto bad;
+ }
+ items = 0;
+ avkey->source_type = le32_to_cpu(buf[items++]);
+ avkey->target_type = le32_to_cpu(buf[items++]);
+ avkey->target_class = le32_to_cpu(buf[items++]);
+ avdatum->specified = le32_to_cpu(buf[items++]);
+ if (!(avdatum->specified & (AVTAB_AV | AVTAB_TYPE))) {
+ printk(KERN_ERR "security: avtab: null entry\n");
+ goto bad;
+ }
+ if ((avdatum->specified & AVTAB_AV) &&
+ (avdatum->specified & AVTAB_TYPE)) {
+ printk(KERN_ERR "security: avtab: entry has both access vectors and types\n");
+ goto bad;
+ }
+ if (avdatum->specified & AVTAB_AV) {
+ if (avdatum->specified & AVTAB_ALLOWED)
+ avtab_allowed(avdatum) = le32_to_cpu(buf[items++]);
+ if (avdatum->specified & AVTAB_AUDITDENY)
+ avtab_auditdeny(avdatum) = le32_to_cpu(buf[items++]);
+ if (avdatum->specified & AVTAB_AUDITALLOW)
+ avtab_auditallow(avdatum) = le32_to_cpu(buf[items++]);
+ } else {
+ if (avdatum->specified & AVTAB_TRANSITION)
+ avtab_transition(avdatum) = le32_to_cpu(buf[items++]);
+ if (avdatum->specified & AVTAB_CHANGE)
+ avtab_change(avdatum) = le32_to_cpu(buf[items++]);
+ if (avdatum->specified & AVTAB_MEMBER)
+ avtab_member(avdatum) = le32_to_cpu(buf[items++]);
+ }
+ if (items != items2) {
+ printk(KERN_ERR "security: avtab: entry only had %d items, expected %d\n",
+ items2, items);
+ goto bad;
+ }
+
+ return 0;
+bad:
+ return -1;
+}
+
int avtab_read(struct avtab *a, void *fp, u32 config)
{
int i, rc = -EINVAL;
struct avtab_key avkey;
struct avtab_datum avdatum;
u32 *buf;
- u32 nel, items, items2;
+ u32 nel;
buf = next_entry(fp, sizeof(u32));
@@ -199,55 +378,8 @@ int avtab_read(struct avtab *a, void *fp
goto bad;
}
for (i = 0; i < nel; i++) {
- memset(&avkey, 0, sizeof(avkey));
- memset(&avdatum, 0, sizeof(avdatum));
-
- buf = next_entry(fp, sizeof(u32));
- if (!buf) {
- printk(KERN_ERR "security: avtab: truncated entry\n");
- goto bad;
- }
- items2 = le32_to_cpu(buf[0]);
- buf = next_entry(fp, sizeof(u32)*items2);
- if (!buf) {
- printk(KERN_ERR "security: avtab: truncated entry\n");
+ if (avtab_read_item(fp, &avdatum, &avkey))
goto bad;
- }
- items = 0;
- avkey.source_type = le32_to_cpu(buf[items++]);
- avkey.target_type = le32_to_cpu(buf[items++]);
- avkey.target_class = le32_to_cpu(buf[items++]);
- avdatum.specified = le32_to_cpu(buf[items++]);
- if (!(avdatum.specified & (AVTAB_AV | AVTAB_TYPE))) {
- printk(KERN_ERR "security: avtab: null entry\n");
- goto bad;
- }
- if ((avdatum.specified & AVTAB_AV) &&
- (avdatum.specified & AVTAB_TYPE)) {
- printk(KERN_ERR "security: avtab: entry has both "
- "access vectors and types\n");
- goto bad;
- }
- if (avdatum.specified & AVTAB_AV) {
- if (avdatum.specified & AVTAB_ALLOWED)
- avtab_allowed(&avdatum) = le32_to_cpu(buf[items++]);
- if (avdatum.specified & AVTAB_AUDITDENY)
- avtab_auditdeny(&avdatum) = le32_to_cpu(buf[items++]);
- if (avdatum.specified & AVTAB_AUDITALLOW)
- avtab_auditallow(&avdatum) = le32_to_cpu(buf[items++]);
- } else {
- if (avdatum.specified & AVTAB_TRANSITION)
- avtab_transition(&avdatum) = le32_to_cpu(buf[items++]);
- if (avdatum.specified & AVTAB_CHANGE)
- avtab_change(&avdatum) = le32_to_cpu(buf[items++]);
- if (avdatum.specified & AVTAB_MEMBER)
- avtab_member(&avdatum) = le32_to_cpu(buf[items++]);
- }
- if (items != items2) {
- printk(KERN_ERR "security: avtab: entry only had %d "
- "items, expected %d\n", items2, items);
- goto bad;
- }
rc = avtab_insert(a, &avkey, &avdatum);
if (rc) {
if (rc == -ENOMEM)
--- diff/security/selinux/ss/avtab.h 2003-08-20 13:16:36.000000000 +0000
+++ source/security/selinux/ss/avtab.h 2004-03-16 11:27:48.757775352 +0000
@@ -7,6 +7,16 @@
*
* Author : Stephen Smalley, <sds@epoch.ncsc.mil>
*/
+
+/* Updated: Frank Mayer <mayerf@tresys.com> and Karl MacMillan <kmacmillan@tresys.com>
+ *
+ * Added conditional policy language extensions
+ *
+ * Copyright (C) 2003 Tresys Technology, LLC
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, version 2.
+ */
#ifndef _SS_AVTAB_H_
#define _SS_AVTAB_H_
@@ -25,6 +35,7 @@ struct avtab_datum {
#define AVTAB_MEMBER 32
#define AVTAB_CHANGE 64
#define AVTAB_TYPE (AVTAB_TRANSITION | AVTAB_MEMBER | AVTAB_CHANGE)
+#define AVTAB_ENABLED 0x80000000 /* reserved for used in cond_avtab */
u32 specified; /* what fields are specified */
u32 data[3]; /* access vectors or types */
#define avtab_allowed(x) (x)->data[0]
@@ -56,8 +67,17 @@ int avtab_map(struct avtab *h,
void *args),
void *args);
void avtab_hash_eval(struct avtab *h, char *tag);
+
+int avtab_read_item(void *fp, struct avtab_datum *avdatum, struct avtab_key *avkey);
int avtab_read(struct avtab *a, void *fp, u32 config);
+struct avtab_node *avtab_insert_nonunique(struct avtab *h, struct avtab_key *key,
+ struct avtab_datum *datum);
+
+struct avtab_node *avtab_search_node(struct avtab *h, struct avtab_key *key, int specified);
+
+struct avtab_node *avtab_search_node_next(struct avtab_node *node, int specified);
+
#define AVTAB_HASH_BITS 15
#define AVTAB_HASH_BUCKETS (1 << AVTAB_HASH_BITS)
#define AVTAB_HASH_MASK (AVTAB_HASH_BUCKETS-1)
--- diff/security/selinux/ss/mls.h 2003-08-20 13:16:36.000000000 +0000
+++ source/security/selinux/ss/mls.h 2004-03-16 11:27:48.760774896 +0000
@@ -48,12 +48,12 @@ usercon.range = __ranges->range;
#define mls_end_user_ranges } }
-#define mls_symtab_names , "levels", "categories"
-#define mls_symtab_sizes , 16, 16
-#define mls_index_f ,sens_index, cat_index
-#define mls_destroy_f ,sens_destroy, cat_destroy
-#define mls_read_f ,sens_read, cat_read
-#define mls_write_f ,sens_write, cat_write
+#define mls_symtab_names "levels", "categories",
+#define mls_symtab_sizes 16, 16,
+#define mls_index_f sens_index, cat_index,
+#define mls_destroy_f sens_destroy, cat_destroy,
+#define mls_read_f sens_read, cat_read,
+#define mls_write_f sens_write, cat_write,
#define mls_policydb_index_others(p) printk(", %d levels", p->nlevels);
#define mls_set_config(config) config |= POLICYDB_CONFIG_MLS
--- diff/security/selinux/ss/policydb.c 2004-02-18 08:54:13.000000000 +0000
+++ source/security/selinux/ss/policydb.c 2004-03-16 11:27:48.761774744 +0000
@@ -3,12 +3,25 @@
*
* Author : Stephen Smalley, <sds@epoch.ncsc.mil>
*/
+
+/* Updated: Frank Mayer <mayerf@tresys.com> and Karl MacMillan <kmacmillan@tresys.com>
+ *
+ * Added conditional policy language extensions
+ *
+ * Copyright (C) 2003 - 2004 Tresys Technology, LLC
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, version 2.
+ */
+
#include <linux/kernel.h>
#include <linux/slab.h>
#include <linux/string.h>
#include <linux/errno.h>
#include "security.h"
+
#include "policydb.h"
+#include "conditional.h"
#include "mls.h"
#define _DEBUG_HASHES
@@ -19,8 +32,9 @@ static char *symtab_name[SYM_NUM] = {
"classes",
"roles",
"types",
- "users"
+ "users",
mls_symtab_names
+ "bools"
};
#endif
@@ -29,8 +43,9 @@ static unsigned int symtab_sizes[SYM_NUM
32,
16,
512,
- 128
+ 128,
mls_symtab_sizes
+ 16
};
/*
@@ -95,6 +110,10 @@ int policydb_init(struct policydb *p)
if (rc)
goto out_free_avtab;
+ rc = cond_policydb_init(p);
+ if (rc)
+ goto out_free_avtab;
+
out:
return rc;
@@ -195,8 +214,9 @@ static int (*index_f[SYM_NUM]) (void *ke
class_index,
role_index,
type_index,
- user_index
+ user_index,
mls_index_f
+ cond_index_bool
};
/*
@@ -267,8 +287,8 @@ int policydb_index_others(struct policyd
{
int i, rc = 0;
- printk(KERN_INFO "security: %d users, %d roles, %d types",
- p->p_users.nprim, p->p_roles.nprim, p->p_types.nprim);
+ printk(KERN_INFO "security: %d users, %d roles, %d types, %d bools",
+ p->p_users.nprim, p->p_roles.nprim, p->p_types.nprim, p->p_bools.nprim);
mls_policydb_index_others(p);
printk("\n");
@@ -296,6 +316,11 @@ int policydb_index_others(struct policyd
goto out;
}
+ if (cond_init_bool_indexes(p)) {
+ rc = -ENOMEM;
+ goto out;
+ }
+
for (i = SYM_ROLES; i < SYM_NUM; i++) {
p->sym_val_to_name[i] =
kmalloc(p->symtab[i].nprim * sizeof(char *), GFP_KERNEL);
@@ -402,8 +427,9 @@ static int (*destroy_f[SYM_NUM]) (void *
class_destroy,
role_destroy,
type_destroy,
- user_destroy
+ user_destroy,
mls_destroy_f
+ cond_destroy_bool
};
void ocontext_destroy(struct ocontext *c, int i)
@@ -467,6 +493,8 @@ void policydb_destroy(struct policydb *p
kfree(gtmp);
}
+ cond_policydb_destroy(p);
+
return;
}
@@ -1040,8 +1068,9 @@ static int (*read_f[SYM_NUM]) (struct po
class_read,
role_read,
type_read,
- user_read
+ user_read,
mls_read_f
+ cond_read_bool
};
#define mls_config(x) \
@@ -1057,7 +1086,7 @@ int policydb_read(struct policydb *p, vo
struct role_trans *tr, *ltr;
struct ocontext *l, *c, *newc;
struct genfs *genfs_p, *genfs, *newgenfs;
- int i, j, rc;
+ int i, j, rc, policy_ver, num_syms;
u32 *buf, len, len2, config, nprim, nel, nel2;
char *policydb_str;
@@ -1122,7 +1151,8 @@ int policydb_read(struct policydb *p, vo
for (i = 0; i < 4; i++)
buf[i] = le32_to_cpu(buf[i]);
- if (buf[0] != POLICYDB_VERSION) {
+ policy_ver = buf[0];
+ if (policy_ver != POLICYDB_VERSION && policy_ver != POLICYDB_VERSION_COMPAT) {
printk(KERN_ERR "security: policydb version %d does not match "
"my version %d\n", buf[0], POLICYDB_VERSION);
goto bad;
@@ -1134,18 +1164,30 @@ int policydb_read(struct policydb *p, vo
mls_config(config));
goto bad;
}
- if (buf[2] != SYM_NUM || buf[3] != OCON_NUM) {
- printk(KERN_ERR "security: policydb table sizes (%d,%d) do "
- "not match mine (%d,%d)\n",
- buf[2], buf[3], SYM_NUM, OCON_NUM);
- goto bad;
+
+ if (policy_ver == POLICYDB_VERSION_COMPAT) {
+ if (buf[2] != (SYM_NUM - 1) || buf[3] != OCON_NUM) {
+ printk(KERN_ERR "security: policydb table sizes (%d,%d) do "
+ "not match mine (%d,%d)\n",
+ buf[2], buf[3], SYM_NUM, OCON_NUM);
+ goto bad;
+ }
+ num_syms = SYM_NUM - 1;
+ } else {
+ if (buf[2] != SYM_NUM || buf[3] != OCON_NUM) {
+ printk(KERN_ERR "security: policydb table sizes (%d,%d) do "
+ "not match mine (%d,%d)\n",
+ buf[2], buf[3], SYM_NUM, OCON_NUM);
+ goto bad;
+ }
+ num_syms = SYM_NUM;
}
rc = mls_read_nlevels(p, fp);
if (rc)
goto bad;
- for (i = 0; i < SYM_NUM; i++) {
+ for (i = 0; i < num_syms; i++) {
buf = next_entry(fp, sizeof(u32)*2);
if (!buf) {
rc = -EINVAL;
@@ -1166,6 +1208,12 @@ int policydb_read(struct policydb *p, vo
if (rc)
goto bad;
+ if (policy_ver == POLICYDB_VERSION) {
+ rc = cond_read_list(p, fp);
+ if (rc)
+ goto bad;
+ }
+
buf = next_entry(fp, sizeof(u32));
if (!buf) {
rc = -EINVAL;
--- diff/security/selinux/ss/policydb.h 2003-10-27 09:20:39.000000000 +0000
+++ source/security/selinux/ss/policydb.h 2004-03-16 11:27:48.762774592 +0000
@@ -4,6 +4,17 @@
*
* Author : Stephen Smalley, <sds@epoch.ncsc.mil>
*/
+
+/* Updated: Frank Mayer <mayerf@tresys.com> and Karl MacMillan <kmacmillan@tresys.com>
+ *
+ * Added conditional policy language extensions
+ *
+ * Copyright (C) 2003 - 2004 Tresys Technology, LLC
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, version 2.
+ */
+
#ifndef _SS_POLICYDB_H_
#define _SS_POLICYDB_H_
@@ -100,6 +111,13 @@ struct cat_datum {
};
#endif
+/* Boolean data type */
+struct cond_bool_datum {
+ __u32 value; /* internal type value */
+ int state;
+};
+
+struct cond_node;
/*
* The configuration data includes security contexts for
@@ -145,9 +163,11 @@ struct genfs {
#ifdef CONFIG_SECURITY_SELINUX_MLS
#define SYM_LEVELS 5
#define SYM_CATS 6
-#define SYM_NUM 7
+#define SYM_BOOLS 7
+#define SYM_NUM 8
#else
-#define SYM_NUM 5
+#define SYM_BOOLS 5
+#define SYM_NUM 6
#endif
/* object context array indices */
@@ -170,6 +190,7 @@ struct policydb {
#define p_users symtab[SYM_USERS]
#define p_levels symtab[SYM_LEVELS]
#define p_cats symtab[SYM_CATS]
+#define p_bools symtab[SYM_BOOLS]
/* symbol names indexed by (value - 1) */
char **sym_val_to_name[SYM_NUM];
@@ -180,6 +201,7 @@ struct policydb {
#define p_user_val_to_name sym_val_to_name[SYM_USERS]
#define p_sens_val_to_name sym_val_to_name[SYM_LEVELS]
#define p_cat_val_to_name sym_val_to_name[SYM_CATS]
+#define p_bool_val_to_name sym_val_to_name[SYM_BOOLS]
/* class, role, and user attributes indexed by (value - 1) */
struct class_datum **class_val_to_struct;
@@ -192,6 +214,13 @@ struct policydb {
/* role transitions */
struct role_trans *role_tr;
+ /* bools indexed by (value - 1) */
+ struct cond_bool_datum **bool_val_to_struct;
+ /* type enforcement conditional access vectors and transitions */
+ struct avtab te_cond_avtab;
+ /* linked list indexing te_cond_avtab by conditional */
+ struct cond_node* cond_list;
+
/* role allows */
struct role_allow *role_allow;
--- diff/security/selinux/ss/services.c 2004-03-11 10:20:29.000000000 +0000
+++ source/security/selinux/ss/services.c 2004-03-16 11:27:48.763774440 +0000
@@ -9,6 +9,15 @@
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2,
* as published by the Free Software Foundation.
+ *
+ * Updated: Frank Mayer <mayerf@tresys.com> and Karl MacMillan <kmacmillan@tresys.com>
+ *
+ * Added conditional policy language extensions
+ *
+ * Copyright (C) 2003 - 2004 Tresys Technology, LLC
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, version 2.
*/
#include <linux/kernel.h>
#include <linux/slab.h>
@@ -26,6 +35,7 @@
#include "policydb.h"
#include "sidtab.h"
#include "services.h"
+#include "conditional.h"
#include "mls.h"
extern void selnl_notify_policyload(u32 seqno);
@@ -225,6 +235,9 @@ static int context_struct_compute_av(str
avd->auditallow = avtab_auditallow(avdatum);
}
+ /* Check conditional av table for additional permissions */
+ cond_compute_av(&policydb.te_cond_avtab, &avkey, avd);
+
/*
* Remove any permissions prohibited by the MLS policy.
*/
@@ -249,7 +262,7 @@ static int context_struct_compute_av(str
* pair.
*/
if (tclass == SECCLASS_PROCESS &&
- avd->allowed && PROCESS__TRANSITION &&
+ (avd->allowed & PROCESS__TRANSITION) &&
scontext->role != tcontext->role) {
for (ra = policydb.role_allow; ra; ra = ra->next) {
if (scontext->role == ra->role &&
@@ -386,7 +399,7 @@ int security_sid_to_context(u32 sid, cha
char *scontextp;
*scontext_len = strlen(initial_sid_to_string[sid]) + 1;
- scontextp = kmalloc(*scontext_len,GFP_KERNEL);
+ scontextp = kmalloc(*scontext_len,GFP_ATOMIC);
strcpy(scontextp, initial_sid_to_string[sid]);
*scontext = scontextp;
goto out;
@@ -573,6 +586,7 @@ static int security_compute_sid(u32 ssid
struct role_trans *roletr = 0;
struct avtab_key avkey;
struct avtab_datum *avdatum;
+ struct avtab_node *node;
unsigned int type_change = 0;
int rc = 0;
@@ -639,6 +653,18 @@ static int security_compute_sid(u32 ssid
avkey.target_type = tcontext->type;
avkey.target_class = tclass;
avdatum = avtab_search(&policydb.te_avtab, &avkey, AVTAB_TYPE);
+
+ /* If no permanent rule, also check for enabled conditional rules */
+ if(!avdatum) {
+ node = avtab_search_node(&policydb.te_cond_avtab, &avkey, specified);
+ for (; node != NULL; node = avtab_search_node_next(node, specified)) {
+ if (node->datum.specified & AVTAB_ENABLED) {
+ avdatum = &node->datum;
+ break;
+ }
+ }
+ }
+
type_change = (avdatum && (avdatum->specified & specified));
if (type_change) {
/* Use the type from the type transition/member/change rule. */
@@ -1000,6 +1026,7 @@ int security_load_policy(void *data, siz
return -EINVAL;
}
ss_initialized = 1;
+
LOAD_UNLOCK;
selinux_complete_init();
return 0;
@@ -1046,6 +1073,7 @@ int security_load_policy(void *data, siz
memcpy(&policydb, &newpolicydb, sizeof policydb);
sidtab_set(&sidtab, &newsidtab);
seqno = ++latest_granting;
+
POLICY_WRUNLOCK;
LOAD_UNLOCK;
@@ -1428,3 +1456,116 @@ out:
POLICY_RDUNLOCK;
return rc;
}
+
+int security_get_bools(int *len, char ***names, int **values)
+{
+ int i, rc = -ENOMEM;
+
+ POLICY_RDLOCK;
+ *names = NULL;
+ *values = NULL;
+
+ *len = policydb.p_bools.nprim;
+ if (!*len) {
+ rc = 0;
+ goto out;
+ }
+
+ *names = (char**)kmalloc(sizeof(char*) * *len, GFP_ATOMIC);
+ if (!*names)
+ goto err;
+ memset(*names, 0, sizeof(char*) * *len);
+
+ *values = (int*)kmalloc(sizeof(int) * *len, GFP_ATOMIC);
+ if (!*values)
+ goto err;
+
+ for (i = 0; i < *len; i++) {
+ size_t name_len;
+ (*values)[i] = policydb.bool_val_to_struct[i]->state;
+ name_len = strlen(policydb.p_bool_val_to_name[i]) + 1;
+ (*names)[i] = (char*)kmalloc(sizeof(char) * name_len, GFP_ATOMIC);
+ if (!(*names)[i])
+ goto err;
+ strncpy((*names)[i], policydb.p_bool_val_to_name[i], name_len);
+ (*names)[i][name_len - 1] = 0;
+ }
+ rc = 0;
+out:
+ POLICY_RDUNLOCK;
+ return rc;
+err:
+ if (*names) {
+ for (i = 0; i < *len; i++)
+ if ((*names)[i])
+ kfree((*names)[i]);
+ }
+ if (*values)
+ kfree(*values);
+ goto out;
+}
+
+
+int security_set_bools(int len, int *values)
+{
+ int i, rc = 0;
+ int lenp, seqno = 0;
+ struct cond_node *cur;
+
+ POLICY_WRLOCK;
+
+ lenp = policydb.p_bools.nprim;
+ if (len != lenp) {
+ rc = -EFAULT;
+ goto out;
+ }
+
+ printk(KERN_INFO "security: committed booleans { ");
+ for (i = 0; i < len; i++) {
+ if (values[i]) {
+ policydb.bool_val_to_struct[i]->state = 1;
+ } else {
+ policydb.bool_val_to_struct[i]->state = 0;
+ }
+ if (i != 0)
+ printk(", ");
+ printk("%s:%d", policydb.p_bool_val_to_name[i],
+ policydb.bool_val_to_struct[i]->state);
+ }
+ printk(" }\n");
+
+ for (cur = policydb.cond_list; cur != NULL; cur = cur->next) {
+ rc = evaluate_cond_node(&policydb, cur);
+ if (rc)
+ goto out;
+ }
+
+ seqno = ++latest_granting;
+
+out:
+ POLICY_WRUNLOCK;
+ if (!rc) {
+ avc_ss_reset(seqno);
+ selnl_notify_policyload(seqno);
+ }
+ return rc;
+}
+
+int security_get_bool_value(int bool)
+{
+ int rc = 0;
+ int len;
+
+ POLICY_RDLOCK;
+
+ len = policydb.p_bools.nprim;
+ if (bool >= len) {
+ rc = -EFAULT;
+ goto out;
+ }
+
+ rc = policydb.bool_val_to_struct[bool]->state;
+out:
+ POLICY_RDUNLOCK;
+ return rc;
+}
--- diff/sound/core/init.c 2004-03-16 11:24:02.000000000 +0000
+++ source/sound/core/init.c 2004-03-16 11:27:48.764774288 +0000
@@ -281,7 +281,8 @@ int snd_card_free(snd_card_t * card)
}
if (card->private_free)
card->private_free(card);
- snd_info_unregister(card->proc_id);
+ if (card->proc_id)
+ snd_info_unregister(card->proc_id);
if (snd_info_card_free(card) < 0) {
snd_printk(KERN_WARNING "unable to free card info\n");
/* Not fatal error */
--- diff/sound/drivers/serial-u16550.c 2004-02-18 08:54:13.000000000 +0000
+++ source/sound/drivers/serial-u16550.c 2004-03-16 11:27:48.765774136 +0000
@@ -342,7 +342,7 @@ static int __init snd_uart16550_detect(s
unsigned char c;
/* Do some vague tests for the presence of the uart */
- if (io_base == 0) {
+ if (io_base == 0 || io_base == SNDRV_AUTO_PORT) {
return -ENODEV; /* Not configured */
}
@@ -804,10 +804,9 @@ static int __init snd_uart16550_create(s
return err;
}
- if (irq >= 0) {
+ if (irq >= 0 && irq != SNDRV_AUTO_IRQ) {
if (request_irq(irq, snd_uart16550_interrupt,
SA_INTERRUPT, "Serial MIDI", (void *) uart)) {
- uart->irq = -1;
snd_printk("irq %d busy. Using Polling.\n", irq);
} else {
uart->irq = irq;
--- diff/sound/i2c/cs8427.c 2004-02-18 08:54:13.000000000 +0000
+++ source/sound/i2c/cs8427.c 2004-03-16 11:27:48.766773984 +0000
@@ -553,6 +553,7 @@ int snd_cs8427_iec958_pcm(snd_i2c_device
SNDRV_CTL_EVENT_MASK_VALUE,
&chip->playback.pcm_ctl->id);
reset = chip->rate != rate;
+ chip->rate = rate;
snd_i2c_unlock(cs8427->bus);
if (reset)
snd_cs8427_reset(cs8427);
--- diff/sound/isa/dt019x.c 2004-03-16 11:24:02.000000000 +0000
+++ source/sound/isa/dt019x.c 2004-03-16 11:27:48.766773984 +0000
@@ -296,8 +296,8 @@ static int __devinit snd_dt019x_pnp_prob
return res;
dev++;
return 0;
- }
- return -ENODEV;
+ }
+ return -ENODEV;
}
static void __devexit snd_dt019x_pnp_remove(struct pnp_card_link * pcard)
@@ -322,8 +322,10 @@ static int __init alsa_card_dt019x_init(
cards += pnp_register_card_driver(&dt019x_pnpc_driver);
#ifdef MODULE
- if (!cards)
+ if (!cards) {
+ pnp_unregister_card_driver(&dt019x_pnpc_driver);
snd_printk(KERN_ERR "no DT-019X / ALS-007 based soundcards found\n");
+ }
#endif
return cards ? 0 : -ENODEV;
}
--- diff/sound/isa/es1688/es1688.c 2004-02-18 08:54:13.000000000 +0000
+++ source/sound/isa/es1688/es1688.c 2004-03-16 11:27:48.767773832 +0000
@@ -182,7 +182,7 @@ static int __init snd_audiodrive_legacy_
static int __init alsa_card_es1688_init(void)
{
static unsigned long possible_ports[] = {0x220, 0x240, 0x260, -1};
- int dev, cards = 0;
+ int dev, cards = 0, i;
for (dev = cards = 0; dev < SNDRV_CARDS && enable[dev]; dev++) {
if (port[dev] == SNDRV_AUTO_PORT)
@@ -190,7 +190,10 @@ static int __init alsa_card_es1688_init(
if (snd_audiodrive_probe(dev) >= 0)
cards++;
}
- cards += snd_legacy_auto_probe(possible_ports, snd_audiodrive_legacy_auto_probe);
+ i = snd_legacy_auto_probe(possible_ports, snd_audiodrive_legacy_auto_probe);
+ if (i > 0)
+ cards += i;
+
if (!cards) {
#ifdef MODULE
printk(KERN_ERR "ESS AudioDrive ES1688 soundcard not found or device busy\n");
--- diff/sound/isa/es18xx.c 2004-03-16 11:24:02.000000000 +0000
+++ source/sound/isa/es18xx.c 2004-03-16 11:27:48.768773680 +0000
@@ -2232,7 +2232,7 @@ static struct pnp_card_driver es18xx_pnp
static int __init alsa_card_es18xx_init(void)
{
static unsigned long possible_ports[] = {0x220, 0x240, 0x260, 0x280, -1};
- int dev, cards = 0;
+ int dev, cards = 0, i;
/* legacy non-auto cards at first */
for (dev = 0; dev < SNDRV_CARDS; dev++) {
@@ -2246,10 +2246,16 @@ static int __init alsa_card_es18xx_init(
cards++;
}
/* legacy auto configured cards */
- cards += snd_legacy_auto_probe(possible_ports, snd_audiodrive_probe_legacy_port);
+ i = snd_legacy_auto_probe(possible_ports, snd_audiodrive_probe_legacy_port);
+ if (i > 0)
+ cards += i;
+
#ifdef CONFIG_PNP
/* ISA PnP cards at last */
- cards += pnp_register_card_driver(&es18xx_pnpc_driver);
+ i = pnp_register_card_driver(&es18xx_pnpc_driver);
+ if (i > 0)
+ cards += i;
+
#endif
if(!cards) {
#ifdef CONFIG_PNP
--- diff/sound/isa/gus/gusclassic.c 2004-02-18 08:54:13.000000000 +0000
+++ source/sound/isa/gus/gusclassic.c 2004-03-16 11:27:48.769773528 +0000
@@ -238,7 +238,7 @@ static int __init snd_gusclassic_legacy_
static int __init alsa_card_gusclassic_init(void)
{
static unsigned long possible_ports[] = {0x220, 0x230, 0x240, 0x250, 0x260, -1};
- int dev, cards;
+ int dev, cards, i;
for (dev = cards = 0; dev < SNDRV_CARDS && enable[dev]; dev++) {
if (port[dev] == SNDRV_AUTO_PORT)
@@ -246,7 +246,10 @@ static int __init alsa_card_gusclassic_i
if (snd_gusclassic_probe(dev) >= 0)
cards++;
}
- cards += snd_legacy_auto_probe(possible_ports, snd_gusclassic_legacy_auto_probe);
+ i = snd_legacy_auto_probe(possible_ports, snd_gusclassic_legacy_auto_probe);
+ if (i > 0)
+ cards += i;
+
if (!cards) {
#ifdef MODULE
printk(KERN_ERR "GUS Classic soundcard not found or device busy\n");
--- diff/sound/isa/gus/gusextreme.c 2004-02-18 08:54:13.000000000 +0000
+++ source/sound/isa/gus/gusextreme.c 2004-03-16 11:27:48.770773376 +0000
@@ -220,17 +220,17 @@ static int __init snd_gusextreme_probe(i
xgf1_irq = gf1_irq[dev];
if (xgf1_irq == SNDRV_AUTO_IRQ) {
if ((xgf1_irq = snd_legacy_find_free_irq(possible_gf1_irqs)) < 0) {
- snd_card_free(card);
snd_printk("unable to find a free IRQ for GF1\n");
- return -EBUSY;
+ err = -EBUSY;
+ goto out;
}
}
xess_irq = irq[dev];
if (xess_irq == SNDRV_AUTO_IRQ) {
if ((xess_irq = snd_legacy_find_free_irq(possible_ess_irqs)) < 0) {
- snd_card_free(card);
snd_printk("unable to find a free IRQ for ES1688\n");
- return -EBUSY;
+ err = -EBUSY;
+ goto out;
}
}
if (mpu_port[dev] == SNDRV_AUTO_PORT)
@@ -241,26 +241,24 @@ static int __init snd_gusextreme_probe(i
xgf1_dma = dma1[dev];
if (xgf1_dma == SNDRV_AUTO_DMA) {
if ((xgf1_dma = snd_legacy_find_free_dma(possible_gf1_dmas)) < 0) {
- snd_card_free(card);
snd_printk("unable to find a free DMA for GF1\n");
- return -EBUSY;
+ err = -EBUSY;
+ goto out;
}
}
xess_dma = dma8[dev];
if (xess_dma == SNDRV_AUTO_DMA) {
if ((xess_dma = snd_legacy_find_free_dma(possible_ess_dmas)) < 0) {
- snd_card_free(card);
snd_printk("unable to find a free DMA for ES1688\n");
- return -EBUSY;
+ err = -EBUSY;
+ goto out;
}
}
if ((err = snd_es1688_create(card, port[dev], mpu_port[dev],
xess_irq, xmpu_irq, xess_dma,
- ES1688_HW_1688, &es1688)) < 0) {
- snd_card_free(card);
- return err;
- }
+ ES1688_HW_1688, &es1688)) < 0)
+ goto out;
if (gf1_port[dev] < 0)
gf1_port[dev] = port[dev] + 0x20;
if ((err = snd_gus_create(card,
@@ -270,77 +268,65 @@ static int __init snd_gusextreme_probe(i
-1,
0, channels[dev],
pcm_channels[dev], 0,
- &gus)) < 0) {
- snd_card_free(card);
- return err;
- }
- if ((err = snd_gusextreme_detect(dev, card, gus, es1688)) < 0) {
- snd_card_free(card);
- return err;
- }
+ &gus)) < 0)
+ goto out;
+
+ if ((err = snd_gusextreme_detect(dev, card, gus, es1688)) < 0)
+ goto out;
+
snd_gusextreme_init(dev, gus);
- if ((err = snd_gus_initialize(gus)) < 0) {
- snd_card_free(card);
- return err;
- }
+ if ((err = snd_gus_initialize(gus)) < 0)
+ goto out;
+
if (!gus->ess_flag) {
snd_printdd("GUS Extreme soundcard was not detected at 0x%lx\n", gus->gf1.port);
- snd_card_free(card);
- return -ENODEV;
- }
- if ((err = snd_es1688_pcm(es1688, 0, NULL)) < 0) {
- snd_card_free(card);
- return err;
- }
- if ((err = snd_es1688_mixer(es1688)) < 0) {
- snd_card_free(card);
- return err;
+ err = -ENODEV;
+ goto out;
}
+ if ((err = snd_es1688_pcm(es1688, 0, NULL)) < 0)
+ goto out;
+
+ if ((err = snd_es1688_mixer(es1688)) < 0)
+ goto out;
+
snd_component_add(card, "ES1688");
if (pcm_channels[dev] > 0) {
- if ((err = snd_gf1_pcm_new(gus, 1, 1, NULL)) < 0) {
- snd_card_free(card);
- return err;
- }
- }
- if ((err = snd_gf1_new_mixer(gus)) < 0) {
- snd_card_free(card);
- return err;
- }
- if ((err = snd_gusextreme_mixer(es1688)) < 0) {
- snd_card_free(card);
- return err;
+ if ((err = snd_gf1_pcm_new(gus, 1, 1, NULL)) < 0)
+ goto out;
}
+ if ((err = snd_gf1_new_mixer(gus)) < 0)
+ goto out;
+
+ if ((err = snd_gusextreme_mixer(es1688)) < 0)
+ goto out;
if (snd_opl3_create(card, es1688->port, es1688->port + 2,
OPL3_HW_OPL3, 0, &opl3) < 0) {
printk(KERN_ERR "gusextreme: opl3 not detected at 0x%lx\n", es1688->port);
} else {
- if ((err = snd_opl3_hwdep_new(opl3, 0, 2, NULL)) < 0) {
- snd_card_free(card);
- return err;
- }
+ if ((err = snd_opl3_hwdep_new(opl3, 0, 2, NULL)) < 0)
+ goto out;
}
- if (es1688->mpu_port >= 0x300) {
- if ((err = snd_mpu401_uart_new(card, 0, MPU401_HW_ES1688,
+ if (es1688->mpu_port >= 0x300 &&
+ (err = snd_mpu401_uart_new(card, 0, MPU401_HW_ES1688,
es1688->mpu_port, 0,
xmpu_irq,
SA_INTERRUPT,
- NULL)) < 0) {
- snd_card_free(card);
- return err;
- }
- }
+ NULL)) < 0)
+ goto out;
sprintf(card->longname, "Gravis UltraSound Extreme at 0x%lx, irq %i&%i, dma %i&%i",
es1688->port, xgf1_irq, xess_irq, xgf1_dma, xess_dma);
- if ((err = snd_card_register(card)) < 0) {
- snd_card_free(card);
- return err;
- }
+ if ((err = snd_card_register(card)) < 0)
+ goto out;
+
snd_gusextreme_cards[dev] = card;
return 0;
+
+ out:
+ snd_card_free(card);
+ return err;
}
static int __init snd_gusextreme_legacy_auto_probe(unsigned long xport)
@@ -363,7 +349,7 @@ static int __init snd_gusextreme_legacy_
static int __init alsa_card_gusextreme_init(void)
{
static unsigned long possible_ports[] = {0x220, 0x240, 0x260, -1};
- int dev, cards;
+ int dev, cards, i;
for (dev = cards = 0; dev < SNDRV_CARDS && enable[dev] > 0; dev++) {
if (port[dev] == SNDRV_AUTO_PORT)
@@ -371,7 +357,10 @@ static int __init alsa_card_gusextreme_i
if (snd_gusextreme_probe(dev) >= 0)
cards++;
}
- cards += snd_legacy_auto_probe(possible_ports, snd_gusextreme_legacy_auto_probe);
+ i = snd_legacy_auto_probe(possible_ports, snd_gusextreme_legacy_auto_probe);
+ if (i > 0)
+ cards += i;
+
if (!cards) {
#ifdef MODULE
printk(KERN_ERR "GUS Extreme soundcard not found or device busy\n");
--- diff/sound/isa/gus/gusmax.c 2004-02-18 08:54:13.000000000 +0000
+++ source/sound/isa/gus/gusmax.c 2004-03-16 11:27:48.770773376 +0000
@@ -378,7 +378,7 @@ static int __init snd_gusmax_legacy_auto
static int __init alsa_card_gusmax_init(void)
{
static unsigned long possible_ports[] = {0x220, 0x230, 0x240, 0x250, 0x260, -1};
- int dev, cards;
+ int dev, cards, i;
for (dev = cards = 0; dev < SNDRV_CARDS && enable[dev] > 0; dev++) {
if (port[dev] == SNDRV_AUTO_PORT)
@@ -386,7 +386,10 @@ static int __init alsa_card_gusmax_init(
if (snd_gusmax_probe(dev) >= 0)
cards++;
}
- cards += snd_legacy_auto_probe(possible_ports, snd_gusmax_legacy_auto_probe);
+ i = snd_legacy_auto_probe(possible_ports, snd_gusmax_legacy_auto_probe);
+ if (i > 0)
+ cards += i;
+
if (!cards) {
#ifdef MODULE
printk(KERN_ERR "GUS MAX soundcard not found or device busy\n");
--- diff/sound/isa/gus/interwave.c 2004-02-18 08:54:13.000000000 +0000
+++ source/sound/isa/gus/interwave.c 2004-03-16 11:27:48.771773224 +0000
@@ -929,7 +929,7 @@ static struct pnp_card_driver interwave_
static int __init alsa_card_interwave_init(void)
{
- int cards = 0;
+ int cards = 0, i;
static long possible_ports[] = {0x210, 0x220, 0x230, 0x240, 0x250, 0x260, -1};
int dev;
@@ -949,10 +949,14 @@ static int __init alsa_card_interwave_in
#endif
}
/* legacy auto configured cards */
- cards += snd_legacy_auto_probe(possible_ports, snd_interwave_probe_legacy_port);
+ i = snd_legacy_auto_probe(possible_ports, snd_interwave_probe_legacy_port);
+ if (i > 0)
+ cards += i;
#ifdef CONFIG_PNP
- /* ISA PnP cards */
- cards += pnp_register_card_driver(&interwave_pnpc_driver);
+ /* ISA PnP cards */
+ i = pnp_register_card_driver(&interwave_pnpc_driver);
+ if (i > 0)
+ cards += i;
#endif
if (!cards) {
--- diff/sound/isa/sb/es968.c 2004-02-18 08:54:13.000000000 +0000
+++ source/sound/isa/sb/es968.c 2004-03-16 11:27:48.772773072 +0000
@@ -203,8 +203,8 @@ static int __devinit snd_es968_pnp_detec
return res;
dev++;
return 0;
- }
- return -ENODEV;
+ }
+ return -ENODEV;
}
static void __devexit snd_es968_pnp_remove(struct pnp_card_link * pcard)
@@ -225,15 +225,14 @@ static struct pnp_card_driver es968_pnpc
static int __init alsa_card_es968_init(void)
{
- int res = pnp_register_card_driver(&es968_pnpc_driver);
- if (res == 0)
- {
- pnp_unregister_card_driver(&es968_pnpc_driver);
+ int cards = pnp_register_card_driver(&es968_pnpc_driver);
#ifdef MODULE
+ if (cards == 0) {
+ pnp_unregister_card_driver(&es968_pnpc_driver);
snd_printk(KERN_ERR "no ES968 based soundcards found\n");
-#endif
}
- return res < 0 ? res : 0;
+#endif
+ return cards ? 0 : -ENODEV;
}
static void __exit alsa_card_es968_exit(void)
--- diff/sound/isa/sb/sb16.c 2004-02-18 08:54:13.000000000 +0000
+++ source/sound/isa/sb/sb16.c 2004-03-16 11:27:48.772773072 +0000
@@ -629,7 +629,7 @@ static struct pnp_card_driver sb16_pnpc_
static int __init alsa_card_sb16_init(void)
{
- int dev, cards = 0;
+ int dev, cards = 0, i;
static unsigned long possible_ports[] = {0x220, 0x240, 0x260, 0x280, -1};
/* legacy non-auto cards at first */
@@ -649,10 +649,15 @@ static int __init alsa_card_sb16_init(vo
#endif
}
/* legacy auto configured cards */
- cards += snd_legacy_auto_probe(possible_ports, snd_sb16_probe_legacy_port);
+ i = snd_legacy_auto_probe(possible_ports, snd_sb16_probe_legacy_port);
+ if (i > 0)
+ cards += i;
+
#ifdef CONFIG_PNP
/* PnP cards at last */
- cards += pnp_register_card_driver(&sb16_pnpc_driver);
+ i = pnp_register_card_driver(&sb16_pnpc_driver);
+ if (i >0)
+ cards += i;
#endif
if (!cards) {
--- diff/sound/isa/sb/sb8.c 2004-02-18 08:54:13.000000000 +0000
+++ source/sound/isa/sb/sb8.c 2004-03-16 11:27:48.773772920 +0000
@@ -199,7 +199,7 @@ static int __init snd_card_sb8_legacy_au
static int __init alsa_card_sb8_init(void)
{
static unsigned long possible_ports[] = {0x220, 0x240, 0x260, -1};
- int dev, cards;
+ int dev, cards, i;
for (dev = cards = 0; dev < SNDRV_CARDS && enable[dev]; dev++) {
if (port[dev] == SNDRV_AUTO_PORT)
@@ -207,7 +207,10 @@ static int __init alsa_card_sb8_init(voi
if (snd_sb8_probe(dev) >= 0)
cards++;
}
- cards += snd_legacy_auto_probe(possible_ports, snd_card_sb8_legacy_auto_probe);
+ i = snd_legacy_auto_probe(possible_ports, snd_card_sb8_legacy_auto_probe);
+ if (i > 0)
+ cards += i;
+
if (!cards) {
#ifdef MODULE
snd_printk(KERN_ERR "Sound Blaster soundcard not found or device busy\n");
--- diff/sound/pci/emu10k1/emu10k1_callback.c 2004-03-16 11:24:02.000000000 +0000
+++ source/sound/pci/emu10k1/emu10k1_callback.c 2004-03-16 11:27:48.774772768 +0000
@@ -426,7 +426,7 @@ start_voice(snd_emux_voice_t *vp)
snd_emu10k1_ptr_write(hw, CDF, ch, sample);
/* invalidate maps */
- temp = ((unsigned int)hw->silent_page_dmaaddr << 1) | MAP_PTI_MASK;
+ temp = ((unsigned int)hw->silent_page.addr << 1) | MAP_PTI_MASK;
snd_emu10k1_ptr_write(hw, MAPA, ch, temp);
snd_emu10k1_ptr_write(hw, MAPB, ch, temp);
--- diff/sound/pci/emu10k1/memory.c 2004-03-16 11:24:02.000000000 +0000
+++ source/sound/pci/emu10k1/memory.c 2004-03-16 11:27:48.775772616 +0000
@@ -62,7 +62,7 @@ static inline void set_silent_ptb(emu10k
page *= UNIT_PAGES;
for (i = 0; i < UNIT_PAGES; i++, page++)
/* do not increment ptr */
- __set_ptb_entry(emu, page, emu->silent_page_dmaaddr);
+ __set_ptb_entry(emu, page, emu->silent_page.addr);
}
#endif /* PAGE_SIZE */
--- diff/sound/pci/mixart/mixart.h 2004-03-16 11:24:02.000000000 +0000
+++ source/sound/pci/mixart/mixart.h 2004-03-16 11:27:48.775772616 +0000
@@ -23,6 +23,7 @@
#ifndef __SOUND_MIXART_H
#define __SOUND_MIXART_H
+#include <linux/interrupt.h>
#include <sound/pcm.h>
#define MIXART_DRIVER_VERSION 0x000100 /* 0.1.0 */
--- diff/sound/pci/trident/trident_memory.c 2004-03-16 11:24:02.000000000 +0000
+++ source/sound/pci/trident/trident_memory.c 2004-03-16 11:27:48.776772464 +0000
@@ -76,8 +76,8 @@ static inline void set_tlb_bus(trident_t
static inline void set_silent_tlb(trident_t *trident, int page)
{
page <<= 1;
- __set_tlb_bus(trident, page, (unsigned long)trident->tlb.silent_page, trident->tlb.silent_page_dmaaddr);
- __set_tlb_bus(trident, page+1, (unsigned long)trident->tlb.silent_page, trident->tlb.silent_page_dmaaddr);
+ __set_tlb_bus(trident, page, (unsigned long)trident->tlb.silent_page.area, trident->tlb.silent_page.addr);
+ __set_tlb_bus(trident, page+1, (unsigned long)trident->tlb.silent_page.area, trident->tlb.silent_page.addr);
}
#else
@@ -111,7 +111,7 @@ static inline void set_silent_tlb(triden
int i;
page *= UNIT_PAGES;
for (i = 0; i < UNIT_PAGES; i++, page++)
- __set_tlb_bus(trident, page, (unsigned long)trident->tlb.silent_page, trident->tlb.silent_page_dmaaddr);
+ __set_tlb_bus(trident, page, (unsigned long)trident->tlb.silent_page.area, trident->tlb.silent_page.addr);
}
#endif /* PAGE_SIZE */
--- diff/Documentation/i386/kgdb/andthen 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/i386/kgdb/andthen 2004-03-16 11:27:47.983893000 +0000
@@ -0,0 +1,100 @@
+
+define set_andthen
+ set var $thp=0
+ set var $thp=(struct kgdb_and_then_struct *)&kgdb_data[0]
+ set var $at_size = (sizeof kgdb_data)/(sizeof *$thp)
+ set var $at_oc=kgdb_and_then_count
+ set var $at_cc=$at_oc
+end
+
+define andthen_next
+ set var $at_cc=$arg0
+end
+
+define andthen
+ andthen_set_edge
+ if ($at_cc >= $at_oc)
+ printf "Outside window. Window size is %d\n",($at_oc-$at_low)
+ else
+ printf "%d: ",$at_cc
+ output *($thp+($at_cc++ % $at_size ))
+ printf "\n"
+ end
+end
+define andthen_set_edge
+ set var $at_oc=kgdb_and_then_count
+ set var $at_low = $at_oc - $at_size
+ if ($at_low < 0 )
+ set var $at_low = 0
+ end
+ if (( $at_cc > $at_oc) || ($at_cc < $at_low))
+ printf "Count outside of window, setting count to "
+ if ($at_cc >= $at_oc)
+ set var $at_cc = $at_oc
+ else
+ set var $at_cc = $at_low
+ end
+ printf "%d\n",$at_cc
+ end
+end
+
+define beforethat
+ andthen_set_edge
+ if ($at_cc <= $at_low)
+ printf "Outside window. Window size is %d\n",($at_oc-$at_low)
+ else
+ printf "%d: ",$at_cc-1
+ output *($thp+(--$at_cc % $at_size ))
+ printf "\n"
+ end
+end
+
+document andthen_next
+ andthen_next <count>
+ . sets the number of the event to display next. If this event
+ . is not in the event pool, either andthen or beforethat will
+ . correct it to the nearest event pool edge. The event pool
+ . ends at the last event recorded and begins <number of events>
+ . prior to that. If beforethat is used next, it will display
+ . event <count> -1.
+.
+ andthen commands are: set_andthen, andthen_next, andthen and beforethat
+end
+
+
+document andthen
+ andthen
+. displays the next event in the list. <set_andthen> sets up to display
+. the oldest saved event first.
+. <count> (optional) count of the event to display.
+. note the number of events saved is specified at configure time.
+. if events are saved between calls to andthen the index will change
+. but the displayed event will be the next one (unless the event buffer
+. is overrun).
+.
+. andthen commands are: set_andthen, andthen_next, andthen and beforethat
+end
+
+document set_andthen
+ set_andthen
+. sets up to use the <andthen> and <beforethat> commands.
+. if you have defined your own struct, use the above and
+. then enter the following:
+. p $thp=(struct kgdb_and_then_structX *)&kgdb_data[0]
+. where <kgdb_and_then_structX> is the name of your structure.
+.
+. andthen commands are: set_andthen, andthen_next, andthen and beforethat
+end
+
+document beforethat
+ beforethat
+. displays the next prior event in the list. <set_andthen> sets up to
+. display the last occuring event first.
+.
+. note the number of events saved is specified at configure time.
+. if events are saved between calls to beforethat the index will change
+. but the displayed event will be the next one (unless the event buffer
+. is overrun).
+.
+. andthen commands are: set_andthen, andthen_next, andthen and beforethat
+end
--- diff/Documentation/i386/kgdb/debug-nmi.txt 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/i386/kgdb/debug-nmi.txt 2004-03-16 11:27:47.983893000 +0000
@@ -0,0 +1,37 @@
+Subject: Debugging with NMI
+Date: Mon, 12 Jul 1999 11:28:31 -0500
+From: David Grothe <dave@gcom.com>
+Organization: Gcom, Inc
+To: David Grothe <dave@gcom.com>
+
+Kernel hackers:
+
+Maybe this is old hat, but it is new to me --
+
+On an ISA bus machine, if you short out the A1 and B1 pins of an ISA
+slot you will generate an NMI to the CPU. This interrupts even a
+machine that is hung in a loop with interrupts disabled. Used in
+conjunction with kgdb <
+ftp://ftp.gcom.com/pub/linux/src/kgdb-2.3.35/kgdb-2.3.35.tgz > you can
+gain debugger control of a machine that is hung in the kernel! Even
+without kgdb the kernel will print a stack trace so you can find out
+where it was hung.
+
+The A1/B1 pins are directly opposite one another and the farthest pins
+towards the bracket end of the ISA bus socket. You can stick a paper
+clip or multi-meter probe between them to short them out.
+
+I had a spare ISA bus to PC104 bus adapter around. The PC104 end of the
+board consists of two rows of wire wrap pins. So I wired a push button
+between the A1/B1 pins and now have an ISA board that I can stick into
+any ISA bus slot for debugger entry.
+
+Microsoft has a circuit diagram of a PCI card at
+http://www.microsoft.com/hwdev/DEBUGGING/DMPSW.HTM. If you want to
+build one you will have to mail them and ask for the PAL equations.
+Nobody makes one comercially.
+
+[THIS TIP COMES WITH NO WARRANTY WHATSOEVER. It works for me, but if
+your machine catches fire, it is your problem, not mine.]
+
+-- Dave (the kgdb guy)
--- diff/Documentation/i386/kgdb/gdb-globals.txt 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/i386/kgdb/gdb-globals.txt 2004-03-16 11:27:47.984892848 +0000
@@ -0,0 +1,71 @@
+Sender: akale@veritas.com
+Date: Fri, 23 Jun 2000 19:26:35 +0530
+From: "Amit S. Kale" <akale@veritas.com>
+Organization: Veritas Software (India)
+To: Dave Grothe <dave@gcom.com>, linux-kernel@vger.rutgers.edu
+CC: David Milburn <dmilburn@wirespeed.com>,
+ "Edouard G. Parmelan" <Edouard.Parmelan@quadratec.fr>,
+ ezannoni@cygnus.com, Keith Owens <kaos@ocs.com.au>
+Subject: Re: Module debugging using kgdb
+
+Dave Grothe wrote:
+>
+> Amit:
+>
+> There is a 2.4.0 version of kgdb on our ftp site:
+> ftp://ftp.gcom.com/pub/linux/src/kgdb. I mirrored your version of gdb
+> and loadmodule.sh there.
+>
+> Have a look at the README file and see if I go it right. If not, send
+> me some corrections and I will update it.
+>
+> Does your version of gdb solve the global variable problem?
+
+Yes.
+Thanks to Elena Zanoni, gdb (developement version) can now calculate
+correctly addresses of dynamically loaded object files. I have not been
+following gdb developement for sometime and am not sure when symbol
+address calculation fix is going to appear in a gdb stable version.
+
+Elena, any idea when the fix will make it to a prebuilt gdb from a
+redhat release?
+
+For the time being I have built a gdb developement version. It can be
+used for module debugging with loadmodule.sh script.
+
+The problem with calculating of module addresses with previous versions
+of gdb was as follows:
+gdb did not use base address of a section while calculating address of
+a symbol in the section in an object file loaded via 'add-symbol-file'.
+It used address of .text segment instead. Due to this addresses of
+symbols in .data, .bss etc. (e.g. global variables) were calculated incorrectly.
+
+Above mentioned fix allow gdb to use base address of a segment while
+calculating address of a symbol in it. It adds a parameter '-s' to
+'add-symbol-file' command for specifying base address of a segment.
+
+loadmodule.sh script works as follows.
+
+1. Copy a module file to target machine.
+2. Load the module on the target machine using insmod with -m parameter.
+insmod produces a module load map which contains base addresses of all
+sections in the module and addresses of symbols in the module file.
+3. Find all sections and their base addresses in the module from
+the module map.
+4. Generate a script that loads the module file. The script uses
+'add-symbol-file' and specifies address of text segment followed by
+addresses of all segments in the module.
+
+Here is an example gdb script produced by loadmodule.sh script.
+
+add-symbol-file foo 0xd082c060 -s .text.lock 0xd08cbfb5
+-s .fixup 0xd08cfbdf -s .rodata 0xd08cfde0 -s __ex_table 0xd08e3b38
+-s .data 0xd08e3d00 -s .bss 0xd08ec8c0 -s __ksymtab 0xd08ee838
+
+With this command gdb can calculate addresses of symbols in ANY segment
+in a module file.
+
+Regards.
+--
+Amit Kale
+Veritas Software ( http://www.veritas.com )
--- diff/Documentation/i386/kgdb/gdbinit 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/i386/kgdb/gdbinit 2004-03-16 11:27:47.984892848 +0000
@@ -0,0 +1,14 @@
+shell echo -e "\003" >/dev/ttyS0
+set remotebaud 38400
+target remote /dev/ttyS0
+define si
+stepi
+printf "EAX=%08x EBX=%08x ECX=%08x EDX=%08x\n", $eax, $ebx, $ecx, $edx
+printf "ESI=%08x EDI=%08x EBP=%08x ESP=%08x\n", $esi, $edi, $ebp, $esp
+x/i $eip
+end
+define ni
+nexti
+printf "EAX=%08x EBX=%08x ECX=%08x EDX=%08x\n", $eax, $ebx, $ecx, $edx
+printf "ESI=%08x EDI=%08x EBP=%08x ESP=%08x\n", $esi, $edi, $ebp, $esp
+x/i $eip
--- diff/Documentation/i386/kgdb/gdbinit-modules 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/i386/kgdb/gdbinit-modules 2004-03-16 11:27:47.985892696 +0000
@@ -0,0 +1,146 @@
+#
+# Usefull GDB user-command to debug Linux Kernel Modules with gdbstub.
+#
+# This don't work for Linux-2.0 or older.
+#
+# Author Edouard G. Parmelan <Edouard.Parmelan@quadratec.fr>
+#
+#
+# Fri Apr 30 20:33:29 CEST 1999
+# First public release.
+#
+# Major cleanup after experiment Linux-2.0 kernel without success.
+# Symbols of a module are not in the correct order, I can't explain
+# why :(
+#
+# Fri Mar 19 15:41:40 CET 1999
+# Initial version.
+#
+# Thu Jan 6 16:29:03 CST 2000
+# A little fixing by Dave Grothe <dave@gcom.com>
+#
+# Mon Jun 19 09:33:13 CDT 2000
+# Alignment changes from Edouard Parmelan
+#
+# The basic idea is to find where insmod load the module and inform
+# GDB to load the symbol table of the module with the GDB command
+# ``add-symbol-file <object> <address>''.
+#
+# The Linux kernel holds the list of all loaded modules in module_list,
+# this list end with &kernel_module (exactly with module->next == NULL,
+# but the last module is not a real module).
+#
+# Insmod allocates the struct module before the object file. Since
+# Linux-2.1, this structure contain his size. The real address of
+# the object file is then (char*)module + module->size_of_struct.
+#
+# You can use three user functions ``mod-list'', ``mod-print-symbols''
+# and ``add-module-symbols''.
+#
+# mod-list list all loaded modules with the format:
+# <module-address> <module-name>
+#
+# As soon as you have found the address of your module, you can
+# print its exported symbols (mod-print-symbols) or inform GDB to add
+# symbols from your module file (mod-add-symbols).
+#
+# The argument that you give to mod-print-symbols or mod-add-symbols
+# is the <module-address> from the mod-list command.
+#
+# When using the mod-add-symbols command you must also give the full
+# pathname of the modules object code file.
+#
+# The command mod-add-lis is an example of how to make this easier.
+# You can edit this macro to contain the path name of your own
+# favorite module and then use it as a shorthand to load it. You
+# still need the module-address, however.
+#
+# The internal function ``mod-validate'' set the GDB variable $mod
+# as a ``struct module*'' if the kernel known the module otherwise
+# $mod is set to NULL. This ensure to not add symbols for a wrong
+# address.
+#
+# Have a nice hacking day !
+#
+#
+define mod-list
+ set $mod = (struct module*)module_list
+ # the last module is the kernel, ignore it
+ while $mod != &kernel_module
+ printf "%p\t%s\n", (long)$mod, ($mod)->name
+ set $mod = $mod->next
+ end
+end
+document mod-list
+List all modules in the form: <module-address> <module-name>
+Use the <module-address> as the argument for the other
+mod-commands: mod-print-symbols, mod-add-symbols.
+end
+
+define mod-validate
+ set $mod = (struct module*)module_list
+ while ($mod != $arg0) && ($mod != &kernel_module)
+ set $mod = $mod->next
+ end
+ if $mod == &kernel_module
+ set $mod = 0
+ printf "%p is not a module\n", $arg0
+ end
+end
+document mod-validate
+mod-validate <module-address>
+Internal user-command used to validate the module parameter.
+If <module> is a real loaded module, set $mod to it otherwise set $mod to 0.
+end
+
+
+define mod-print-symbols
+ mod-validate $arg0
+ if $mod != 0
+ set $i = 0
+ while $i < $mod->nsyms
+ set $sym = $mod->syms[$i]
+ printf "%p\t%s\n", $sym->value, $sym->name
+ set $i = $i + 1
+ end
+ end
+end
+document mod-print-symbols
+mod-print-symbols <module-address>
+Print all exported symbols of the module. see mod-list
+end
+
+
+define mod-add-symbols-align
+ mod-validate $arg0
+ if $mod != 0
+ set $mod_base = ($mod->size_of_struct + (long)$mod)
+ if ($arg2 != 0) && (($mod_base & ($arg2 - 1)) != 0)
+ set $mod_base = ($mod_base | ($arg2 - 1)) + 1
+ end
+ add-symbol-file $arg1 $mod_base
+ end
+end
+document mod-add-symbols-align
+mod-add-symbols-align <module-address> <object file path name> <align>
+Load the symbols table of the module from the object file where
+first section aligment is <align>.
+To retreive alignment, use `objdump -h <object file path name>'.
+end
+
+define mod-add-symbols
+ mod-add-symbols-align $arg0 $arg1 sizeof(long)
+end
+document mod-add-symbols
+mod-add-symbols <module-address> <object file path name>
+Load the symbols table of the module from the object file.
+Default alignment is 4. See mod-add-symbols-align.
+end
+
+define mod-add-lis
+ mod-add-symbols-align $arg0 /usr/src/LiS/streams.o 16
+end
+document mod-add-lis
+mod-add-lis <module-address>
+Does mod-add-symbols <module-address> /usr/src/LiS/streams.o
+end
--- diff/Documentation/i386/kgdb/gdbinit.hw 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/i386/kgdb/gdbinit.hw 2004-03-16 11:27:47.984892848 +0000
@@ -0,0 +1,117 @@
+
+#Using ia-32 hardware breakpoints.
+#
+#4 hardware breakpoints are available in ia-32 processors. These breakpoints
+#do not need code modification. They are set using debug registers.
+#
+#Each hardware breakpoint can be of one of the
+#three types: execution, write, access.
+#1. An Execution breakpoint is triggered when code at the breakpoint address is
+#executed.
+#2. A write breakpoint ( aka watchpoints ) is triggered when memory location
+#at the breakpoint address is written.
+#3. An access breakpoint is triggered when memory location at the breakpoint
+#address is either read or written.
+#
+#As hardware breakpoints are available in limited number, use software
+#breakpoints ( br command in gdb ) instead of execution hardware breakpoints.
+#
+#Length of an access or a write breakpoint defines length of the datatype to
+#be watched. Length is 1 for char, 2 short , 3 int.
+#
+#For placing execution, write and access breakpoints, use commands
+#hwebrk, hwwbrk, hwabrk
+#To remove a breakpoint use hwrmbrk command.
+#
+#These commands take following types of arguments. For arguments associated
+#with each command, use help command.
+#1. breakpointno: 0 to 3
+#2. length: 1 to 3
+#3. address: Memory location in hex ( without 0x ) e.g c015e9bc
+#
+#Use the command exinfo to find which hardware breakpoint occured.
+
+#hwebrk breakpointno address
+define hwebrk
+ maintenance packet Y$arg0,0,0,$arg1
+end
+document hwebrk
+ hwebrk <breakpointno> <address>
+ Places a hardware execution breakpoint
+ <breakpointno> = 0 - 3
+ <address> = Hex digits without leading "0x".
+end
+
+#hwwbrk breakpointno length address
+define hwwbrk
+ maintenance packet Y$arg0,1,$arg1,$arg2
+end
+document hwwbrk
+ hwwbrk <breakpointno> <length> <address>
+ Places a hardware write breakpoint
+ <breakpointno> = 0 - 3
+ <length> = 1 (1 byte), 2 (2 byte), 3 (4 byte)
+ <address> = Hex digits without leading "0x".
+end
+
+#hwabrk breakpointno length address
+define hwabrk
+ maintenance packet Y$arg0,1,$arg1,$arg2
+end
+document hwabrk
+ hwabrk <breakpointno> <length> <address>
+ Places a hardware access breakpoint
+ <breakpointno> = 0 - 3
+ <length> = 1 (1 byte), 2 (2 byte), 3 (4 byte)
+ <address> = Hex digits without leading "0x".
+end
+
+#hwrmbrk breakpointno
+define hwrmbrk
+ maintenance packet y$arg0
+end
+document hwrmbrk
+ hwrmbrk <breakpointno>
+ <breakpointno> = 0 - 3
+ Removes a hardware breakpoint
+end
+
+define reboot
+ maintenance packet r
+end
+#exinfo
+define exinfo
+ maintenance packet qE
+end
+document exinfo
+ exinfo
+ Gives information about a breakpoint.
+end
+define get_th
+ p $th=(struct thread_info *)((int)$esp & ~8191)
+end
+document get_th
+ get_tu
+ Gets and prints the current thread_info pointer, Defines th to be it.
+end
+define get_cu
+ p $cu=((struct thread_info *)((int)$esp & ~8191))->task
+end
+document get_cu
+ get_cu
+ Gets and print the "current" value. Defines $cu to be it.
+end
+define int_off
+ set var $flags=$eflags
+ set $eflags=$eflags&~0x200
+ end
+define int_on
+ set var $eflags|=$flags&0x200
+ end
+document int_off
+ saves the current interrupt state and clears the processor interrupt
+ flag. Use int_on to restore the saved flag.
+end
+document int_on
+ Restores the interrupt flag saved by int_off.
+end
--- diff/Documentation/i386/kgdb/kgdb.txt 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/i386/kgdb/kgdb.txt 2004-03-16 11:27:47.988892240 +0000
@@ -0,0 +1,775 @@
+Last edit: <20030806.1637.12>
+This file has information specific to the i386 kgdb option. Other
+platforms with the kgdb option may behave in a similar fashion.
+
+New features:
+============
+20030806.1557.37
+This version was made against the 2.6.0-test2 kernel. We have made the
+following changes:
+
+- The getthread() code in the stub calls find_task_by_pid(). It fails
+ if we are early in the bring up such that the pid arrays have yet to
+ be allocated. We have added a line to kernel/pid.c to make
+ "kgdb_pid_init_done" true once the arrays are allocated. This way the
+ getthread() code knows not to call. This is only used by the thread
+ debugging stuff and threads will not yet exist at this point in the
+ boot.
+
+- For some reason, gdb was not asking for a new thread list when the
+ "info thread" command was given. We changed to the newer version of
+ the thread info command and gdb now seems to ask when needed. Result,
+ we now get all threads in the thread list.
+
+- We now respond to the ThreadExtraInfo request from gdb with the thread
+ name from task_struct .comm. This then appears in the thread list.
+ Thoughts on additional options for this are welcome. Things such as
+ "has BKL" and "Preempted" come to mind. I think we could have a flag
+ word that could enable different bits of info here.
+
+- We now honor, sort of, the C and S commands. These are continue and
+ single set after delivering a signal. We ignore the signal and do the
+ requested action. This only happens when we told gdb that a signal
+ was the reason for entry, which is only done on memory faults. The
+ result is that you can now continue into the Oops.
+
+- We changed the -g to -gdwarf-2. This seems to be the same as -ggdb,
+ but it is more exact on what language to use.
+
+- We added two dwarf2 include files and a bit of code at the end of
+ entry.S. This does not yet work, so it is disabled. Still we want to
+ keep track of the code and "maybe" someone out there can fix it.
+
+- Randy Dunlap sent some fix ups for this file which are now merged.
+
+- Hugh Dickins sent a fix to a bit of code in traps.c that prevents a
+ compiler warning if CONFIG_KGDB is off (now who would do that :).
+
+- Andrew Morton sent a fix for the serial driver which is now merged.
+
+- Andrew also sent a change to the stub around the cpu managment code
+ which is also merged.
+
+- Andrew also sent a patch to make "f" as well as "g" work as SysRq
+ commands to enter kgdb, merged.
+
+- If CONFIG_KGDB and CONFIG_DEBUG_SPINLOCKS are both set we added a
+ "who" field to the spinlock data struct. This is filled with
+ "current" when ever the spinlock suceeds. Useful if you want to know
+ who has the lock.
+
+_ And last, but not least, we fixed the "get_cu" macro to properly get
+ the current value of "current".
+
+New features:
+============
+20030505.1827.27
+We are starting to align with the sourceforge version, at least in
+commands. To this end, the boot command string to start kgdb at
+boot time has been changed from "kgdb" to "gdb".
+
+Andrew Morton sent a couple of patches which are now included as follows:
+1.) We now return a flag to the interrupt handler.
+2.) We no longer use smp_num_cpus (a conflict with the lock meter).
+3.) And from William Lee Irwin III <wli@holomorphy.com> code to make
+ sure high-mem is set up before we attempt to register our interrupt
+ handler.
+We now include asm/kgdb.h from config.h so you will most likely never
+have to include it. It also 'NULLS' the kgdb macros you might have in
+your code when CONFIG_KGDB is not defined. This allows you to just
+turn off CONFIG_KGDB to turn off all the kgdb_ts() calls and such.
+This include is conditioned on the machine being an x86 so as to not
+mess with other archs.
+
+20020801.1129.03
+This is currently the version for the 2.4.18 (and beyond?) kernel.
+
+We have several new "features" beginning with this version:
+
+1.) Kgdb now syncs the "other" CPUs with a cross-CPU NMI. No more
+ waiting and it will pull that guy out of an IRQ off spin lock :)
+
+2.) We doctored up the code that tells where a task is waiting and
+ included it so that the "info thread" command will show a bit more
+ than "schedule()". Try it...
+
+3.) Added the ability to call a function from gdb. All the standard gdb
+ issues apply, i.e. if you hit a breakpoint in the function, you are
+ not allowed to call another (gdb limitation, not kgdb). To help
+ this capability we added a memory allocation function. Gdb does not
+ return this memory (it is used for strings that you pass to that function
+ you are calling from gdb) so we fixed up a way to allow you to
+ manually return the memory (see below).
+
+4.) Kgdb time stamps (kgdb_ts()) are enhanced to expand what was the
+ interrupt flag to now also include the preemption count and the
+ "in_interrupt" info. The flag is now called "with_pif" to indicate
+ the order, preempt_count, in_interrupt, flag. The preempt_count is
+ shifted left by 4 bits so you can read the count in hex by dropping
+ the low order digit. In_interrupt is in bit 1, and the flag is in
+ bit 0.
+
+5.) The command: "p kgdb_info" is now expanded and prints something
+ like:
+(gdb) p kgdb_info
+$2 = {used_malloc = 0, called_from = 0xc0107506, entry_tsc = 67468627259,
+ errcode = 0, vector = 3, print_debug_info = 0, hold_on_sstep = 1,
+ cpus_waiting = {{task = 0xc027a000, pid = 32768, hold = 0,
+ regs = 0xc027bf84}, {task = 0x0, pid = 0, hold = 0, regs = 0x0}}}
+
+ Things to note here: a.) used_malloc is the amount of memory that
+ has been malloc'ed to do calls from gdb. You can reclaim this
+ memory like this: "p kgdb_info.used_malloc=0" Cool, huh? b.)
+ cpus_waiting is now "sized" by the number of CPUs you enter at
+ configure time in the kgdb configure section. This is NOT used
+ anywhere else in the system, but it is "nice" here. c.) The task's
+ "pid" is now in the structure. This is the pid you will need to use
+ to decode to the thread id to get gdb to look at that thread.
+ Remember that the "info thread" command prints a list of threads
+ wherein it numbers each thread with its reference number followed
+ by the thread's pid. Note that the per-CPU idle threads actually
+ have pids of 0 (yes, there is more than one pid 0 in an SMP system).
+ To avoid confusion, kgdb numbers these threads with numbers beyond
+ the MAX_PID. That is why you see 32768 and above.
+
+6.) A subtle change, we now provide the complete register set for tasks
+ that are active on the other CPUs. This allows better trace back on
+ those tasks.
+
+ And, let's mention what we could not fix. Back-trace from all but the
+ thread that we trapped will, most likely, have a bogus entry in it.
+ The problem is that gdb does not recognize the entry code for
+ functions that use "current" near (at all?) the entry. The compiler
+ is putting the "current" decode as the first two instructions of the
+ function where gdb expects to find %ebp changing code. Back trace
+ also has trouble with interrupt frames. I am talking with Daniel
+ Jacobowitz about some way to fix this, but don't hold your breath.
+
+20011220.0050.35
+Major enhancement with this version is the ability to hold one or more
+CPUs in an SMP system while allowing the others to continue. Also, by
+default only the current CPU is enabled on single-step commands (please
+note that gdb issues single-step commands at times other than when you
+use the si command).
+
+Another change is to collect some useful information in
+a global structure called "kgdb_info". You should be able to just:
+
+p kgdb_info
+
+although I have seen cases where the first time this is done gdb just
+prints the first member but prints the whole structure if you then enter
+CR (carriage return or enter). This also works:
+
+p *&kgdb_info
+
+Here is a sample:
+(gdb) p kgdb_info
+$4 = {called_from = 0xc010732c, entry_tsc = 32804123790856, errcode = 0,
+ vector = 3, print_debug_info = 0}
+
+"Called_from" is the return address from the current entry into kgdb.
+Sometimes it is useful to know why you are in kgdb, for example, was
+it an NMI or a real breakpoint? The simple way to interrogate this
+return address is:
+
+l *0xc010732c
+
+which will print the surrounding few lines of source code.
+
+"Entry_tsc" is the CPU TSC on entry to kgdb (useful to compare to the
+kgdb_ts entries).
+
+"errcode" and "vector" are other entry parameters which may be helpful on
+some traps.
+
+"print_debug_info" is the internal debugging kgdb print enable flag. Yes,
+you can modify it.
+
+In SMP systems kgdb_info also includes the "cpus_waiting" structure and
+"hold_on_step":
+
+(gdb) p kgdb_info
+$7 = {called_from = 0xc0112739, entry_tsc = 1034936624074, errcode = 0,
+ vector = 2, print_debug_info = 0, hold_on_sstep = 1, cpus_waiting = {{
+ task = 0x0, hold = 0, regs = 0x0}, {task = 0xc71b8000, hold = 0,
+ regs = 0xc71b9f70}, {task = 0x0, hold = 0, regs = 0x0}, {task = 0x0,
+ hold = 0, regs = 0x0}, {task = 0x0, hold = 0, regs = 0x0}, {task = 0x0,
+ hold = 0, regs = 0x0}, {task = 0x0, hold = 0, regs = 0x0}, {task = 0x0,
+ hold = 0, regs = 0x0}}}
+
+"Cpus_waiting" has an entry for each CPU other than the current one that
+has been stopped. Each entry contains the task_struct address for that
+CPU, the address of the regs for that task and a hold flag. All these
+have the proper typing so that, for example:
+
+p *kgdb_info.cpus_waiting[1].regs
+
+will print the registers for CPU 1.
+
+"Hold_on_sstep" is a new feature with this version and comes up set or
+true. What this means is that whenever kgdb is asked to single-step all
+other CPUs are held (i.e. not allowed to execute). The flag applies to
+all but the current CPU and, again, can be changed:
+
+p kgdb_info.hold_on_sstep=0
+
+restores the old behavior of letting all CPUs run during single-stepping.
+
+Likewise, each CPU has a "hold" flag, which if set, locks that CPU out
+of execution. Note that this has some risk in cases where the CPUs need
+to communicate with each other. If kgdb finds no CPU available on exit,
+it will push a message thru gdb and stay in kgdb. Note that it is legal
+to hold the current CPU as long as at least one CPU can execute.
+
+20010621.1117.09
+This version implements an event queue. Events are signaled by calling
+a function in the kgdb stub and may be examined from gdb. See EVENTS
+below for details. This version also tightens up the interrupt and SMP
+handling to not allow interrupts on the way to kgdb from a breakpoint
+trap. It is fine to allow these interrupts for user code, but not
+system debugging.
+
+Version
+=======
+
+This version of the kgdb package was developed and tested on
+kernel version 2.4.16. It will not install on any earlier kernels.
+It is possible that it will continue to work on later versions
+of 2.4 and then versions of 2.5 (I hope).
+
+
+Debugging Setup
+===============
+
+Designate one machine as the "development" machine. This is the
+machine on which you run your compiles and which has your source
+code for the kernel. Designate a second machine as the "target"
+machine. This is the machine that will run your experimental
+kernel.
+
+The two machines will be connected together via a serial line out
+one or the other of the COM ports of the PC. You will need the
+appropriate modem eliminator (null modem) cable(s) for this.
+
+Decide on which tty port you want the machines to communicate, then
+connect them up back-to-back using the null modem cable. COM1 is
+/dev/ttyS0 and COM2 is /dev/ttyS1. You should test this connection
+with the two machines prior to trying to debug a kernel. Once you
+have it working, on the TARGET machine, enter:
+
+setserial /dev/ttyS0 (or what ever tty you are using)
+
+and record the port address and the IRQ number.
+
+On the DEVELOPMENT machine you need to apply the patch for the kgdb
+hooks. You have probably already done that if you are reading this
+file.
+
+On your DEVELOPMENT machine, go to your kernel source directory and do
+"make Xconfig" where X is one of "x", "menu", or "". If you are
+configuring in the standard serial driver, it must not be a module.
+Either yes or no is ok, but making the serial driver a module means it
+will initialize after kgdb has set up the UART interrupt code and may
+cause a failure of the control-C option discussed below. The configure
+question for the serial driver is under the "Character devices" heading
+and is:
+
+"Standard/generic (8250/16550 and compatible UARTs) serial support"
+
+Go down to the kernel debugging menu item and open it up. Enable the
+kernel kgdb stub code by selecting that item. You can also choose to
+turn on the "-ggdb -O1" compile options. The -ggdb causes the compiler
+to put more debug info (like local symbols) in the object file. On the
+i386 -g and -ggdb are the same so this option just reduces to "O1". The
+-O1 reduces the optimization level. This may be helpful in some cases,
+be aware, however, that this may also mask the problem you are looking
+for.
+
+The baud rate. Default is 115200. What ever you choose be sure that
+the host machine is set to the same speed. I recommend the default.
+
+The port. This is the I/O address of the serial UART that you should
+have gotten using setserial as described above. The standard COM1 port
+(3f8) using IRQ 4 is default. COM2 is 2f8 which by convention uses IRQ
+3.
+
+The port IRQ (see above).
+
+Stack overflow test. This option makes a minor change in the trap,
+system call and interrupt code to detect stack overflow and transfer
+control to kgdb if it happens. (Some platforms have this in the
+baseline code, but the i386 does not.)
+
+You can also configure the system to recognize the boot option
+"console=kgdb" which if given will cause all console output during
+booting to be put thru gdb as well as other consoles. This option
+requires that gdb and kgdb be connected prior to sending console output
+so, if they are not, a breakpoint is executed to force the connection.
+This will happen before any kernel output (it is going thru gdb, right),
+and will stall the boot until the connection is made.
+
+You can also configure in a patch to SysRq to enable the kGdb SysRq.
+This request generates a breakpoint. Since the serial port IRQ line is
+set up after any serial drivers, it is possible that this command will
+work when the control-C will not.
+
+Save and exit the Xconfig program. Then do "make clean" , "make dep"
+and "make bzImage" (or whatever target you want to make). This gets the
+kernel compiled with the "-g" option set -- necessary for debugging.
+
+You have just built the kernel on your DEVELOPMENT machine that you
+intend to run on your TARGET machine.
+
+To install this new kernel, use the following installation procedure.
+Remember, you are on the DEVELOPMENT machine patching the kernel source
+for the kernel that you intend to run on the TARGET machine.
+
+Copy this kernel to your target machine using your usual procedures. I
+usually arrange to copy development:
+/usr/src/linux/arch/i386/boot/bzImage to /vmlinuz on the TARGET machine
+via a LAN based NFS access. That is, I run the cp command on the target
+and copy from the development machine via the LAN. Run Lilo (see "man
+lilo" for details on how to set this up) on the new kernel on the target
+machine so that it will boot! Then boot the kernel on the target
+machine.
+
+On the DEVELOPMENT machine, create a file called .gdbinit in the
+directory /usr/src/linux. An example .gdbinit file looks like this:
+
+shell echo -e "\003" >/dev/ttyS0
+set remotebaud 38400 (or what ever speed you have chosen)
+target remote /dev/ttyS0
+
+
+Change the "echo" and "target" definition so that it specifies the tty
+port that you intend to use. Change the "remotebaud" definition to
+match the data rate that you are going to use for the com line.
+
+You are now ready to try it out.
+
+Boot your target machine with "kgdb" in the boot command i.e. something
+like:
+
+lilo> test kgdb
+
+or if you also want console output thru gdb:
+
+lilo> test kgdb console=kgdb
+
+You should see the lilo message saying it has loaded the kernel and then
+all output stops. The kgdb stub is trying to connect with gdb. Start
+gdb something like this:
+
+
+On your DEVELOPMENT machine, cd /usr/src/linux and enter "gdb vmlinux".
+When gdb gets the symbols loaded it will read your .gdbinit file and, if
+everything is working correctly, you should see gdb print out a few
+lines indicating that a breakpoint has been taken. It will actually
+show a line of code in the target kernel inside the kgdb activation
+code.
+
+The gdb interaction should look something like this:
+
+ linux-dev:/usr/src/linux# gdb vmlinux
+ GDB is free software and you are welcome to distribute copies of it
+ under certain conditions; type "show copying" to see the conditions.
+ There is absolutely no warranty for GDB; type "show warranty" for details.
+ GDB 4.15.1 (i486-slackware-linux),
+ Copyright 1995 Free Software Foundation, Inc...
+ breakpoint () at i386-stub.c:750
+ 750 }
+ (gdb)
+
+You can now use whatever gdb commands you like to set breakpoints.
+Enter "continue" to start your target machine executing again. At this
+point the target system will run at full speed until it encounters
+your breakpoint or gets a segment violation in the kernel, or whatever.
+
+If you have the kgdb console enabled when you continue, gdb will print
+out all the console messages.
+
+The above example caused a breakpoint relatively early in the boot
+process. For the i386 kgdb it is possible to code a break instruction
+as the first C-language point in init/main.c, i.e. as the first instruction
+in start_kernel(). This could be done as follows:
+
+#include <asm/kgdb.h>
+ breakpoint();
+
+This breakpoint() is really a function that sets up the breakpoint and
+single-step hardware trap cells and then executes a breakpoint. Any
+early hard coded breakpoint will need to use this function. Once the
+trap cells are set up they need not be set again, but doing it again
+does not hurt anything, so you don't need to be concerned about which
+breakpoint is hit first. Once the trap cells are set up (and the kernel
+sets them up in due course even if breakpoint() is never called) the
+macro:
+
+BREAKPOINT;
+
+will generate an inline breakpoint. This may be more useful as it stops
+the processor at the instruction instead of in a function a step removed
+from the location of interest. In either case <asm/kgdb.h> must be
+included to define both breakpoint() and BREAKPOINT.
+
+Triggering kgdbstub at other times
+==================================
+
+Often you don't need to enter the debugger until much later in the boot
+or even after the machine has been running for some time. Once the
+kernel is booted and interrupts are on, you can force the system to
+enter the debugger by sending a control-C to the debug port. This is
+what the first line of the recommended .gdbinit file does. This allows
+you to start gdb any time after the system is up as well as when the
+system is already at a breakpoint. (In the case where the system is
+already at a breakpoint the control-C is not needed, however, it will
+be ignored by the target so no harm is done. Also note the the echo
+command assumes that the port speed is already set. This will be true
+once gdb has connected, but it is best to set the port speed before you
+run gdb.)
+
+Another simple way to do this is to put the following file in you ~/bin
+directory:
+
+#!/bin/bash
+echo -e "\003" > /dev/ttyS0
+
+Here, the ttyS0 should be replaced with what ever port you are using.
+The "\003" is control-C. Once you are connected with gdb, you can enter
+control-C at the command prompt.
+
+An alternative way to get control to the debugger is to enable the kGdb
+SysRq command. Then you would enter Alt-SysRq-g (all three keys at the
+same time, but push them down in the order given). To refresh your
+memory of the available SysRq commands try Alt-SysRq-=. Actually any
+undefined command could replace the "=", but I like to KNOW that what I
+am pushing will never be defined.
+
+Debugging hints
+===============
+
+You can break into the target machine at any time from the development
+machine by typing ^C (see above paragraph). If the target machine has
+interrupts enabled this will stop it in the kernel and enter the
+debugger.
+
+There is unfortunately no way of breaking into the kernel if it is
+in a loop with interrupts disabled, so if this happens to you then
+you need to place exploratory breakpoints or printk's into the kernel
+to find out where it is looping. The exploratory breakpoints can be
+entered either thru gdb or hard coded into the source. This is very
+handy if you do something like:
+
+if (<it hurts>) BREAKPOINT;
+
+
+There is a copy of an e-mail in the Documentation/i386/kgdb/ directory
+(debug-nmi.txt) which describes how to create an NMI on an ISA bus
+machine using a paper clip. I have a sophisticated version of this made
+by wiring a push button switch into a PC104/ISA bus adapter card. The
+adapter card nicely furnishes wire wrap pins for all the ISA bus
+signals.
+
+When you are done debugging the kernel on the target machine it is a
+good idea to leave it in a running state. This makes reboots faster,
+bypassing the fsck. So do a gdb "continue" as the last gdb command if
+this is possible. To terminate gdb itself on the development machine
+and leave the target machine running, first clear all breakpoints and
+continue, then type ^Z to suspend gdb and then kill it with "kill %1" or
+something similar.
+
+If gdbstub Does Not Work
+========================
+
+If it doesn't work, you will have to troubleshoot it. Do the easy
+things first like double checking your cabling and data rates. You
+might try some non-kernel based programs to see if the back-to-back
+connection works properly. Just something simple like cat /etc/hosts
+>/dev/ttyS0 on one machine and cat /dev/ttyS0 on the other will tell you
+if you can send data from one machine to the other. Make sure it works
+in both directions. There is no point in tearing out your hair in the
+kernel if the line doesn't work.
+
+All of the real action takes place in the file
+/usr/src/linux/arch/i386/kernel/kgdb_stub.c. That is the code on the target
+machine that interacts with gdb on the development machine. In gdb you can
+turn on a debug switch with the following command:
+
+ set remotedebug
+
+This will print out the protocol messages that gdb is exchanging with
+the target machine.
+
+Another place to look is /usr/src/arch/i386/lib/kgdb_serial.c. This is
+the code that talks to the serial port on the target side. There might
+be a problem there. In particular there is a section of this code that
+tests the UART which will tell you what UART you have if you define
+"PRNT" (just remove "_off" from the #define PRNT_off). To view this
+report you will need to boot the system without any beakpoints. This
+allows the kernel to run to the point where it calls kgdb to set up
+interrupts. At this time kgdb will test the UART and print out the type
+it finds. (You need to wait so that the printks are actually being
+printed. Early in the boot they are cached, waiting for the console to
+be enabled. Also, if kgdb is entered thru a breakpoint it is possible
+to cause a dead lock by calling printk when the console is locked. The
+stub thus avoids doing printks from breakpoints, especially in the
+serial code.) At this time, if the UART fails to do the expected thing,
+kgdb will print out (using printk) information on what failed. (These
+messages will be buried in all the other boot up messages. Look for
+lines that start with "gdb_hook_interrupt:". You may want to use dmesg
+once the system is up to view the log. If this fails or if you still
+don't connect, review your answers for the port address. Use:
+
+setserial /dev/ttyS0
+
+to get the current port and IRQ information. This command will also
+tell you what the system found for the UART type. The stub recognizes
+the following UART types:
+
+16450, 16550, and 16550A
+
+If you are really desperate you can use printk debugging in the
+kgdbstub code in the target kernel until you get it working. In particular,
+there is a global variable in /usr/src/linux/arch/i386/kernel/kgdb_stub.c
+named "remote_debug". Compile your kernel with this set to 1, rather
+than 0 and the debug stub will print out lots of stuff as it does
+what it does. Likewise there are debug printks in the kgdb_serial.c
+code that can be turned on with simple changes in the macro defines.
+
+
+Debugging Loadable Modules
+==========================
+
+This technique comes courtesy of Edouard Parmelan
+<Edouard.Parmelan@quadratec.fr>
+
+When you run gdb, enter the command
+
+source gdbinit-modules
+
+This will read in a file of gdb macros that was installed in your
+kernel source directory when kgdb was installed. This file implements
+the following commands:
+
+mod-list
+ Lists the loaded modules in the form <module-address> <module-name>
+
+mod-print-symbols <module-address>
+ Prints all the symbols in the indicated module.
+
+mod-add-symbols <module-address> <object-file-path-name>
+ Loads the symbols from the object file and associates them
+ with the indicated module.
+
+After you have loaded the module that you want to debug, use the command
+mod-list to find the <module-address> of your module. Then use that
+address in the mod-add-symbols command to load your module's symbols.
+From that point onward you can debug your module as if it were a part
+of the kernel.
+
+The file gdbinit-modules also contains a command named mod-add-lis as
+an example of how to construct a command of your own to load your
+favorite module. The idea is to "can" the pathname of the module
+in the command so you don't have to type so much.
+
+Threads
+=======
+
+Each process in a target machine is seen as a gdb thread. gdb thread
+related commands (info threads, thread n) can be used.
+
+ia-32 hardware breakpoints
+==========================
+
+kgdb stub contains support for hardware breakpoints using debugging features
+of ia-32(x86) processors. These breakpoints do not need code modification.
+They use debugging registers. 4 hardware breakpoints are available in ia-32
+processors.
+
+Each hardware breakpoint can be of one of the following three types.
+
+1. Execution breakpoint - An Execution breakpoint is triggered when code
+ at the breakpoint address is executed.
+
+ As limited number of hardware breakpoints are available, it is
+ advisable to use software breakpoints ( break command ) instead
+ of execution hardware breakpoints, unless modification of code
+ is to be avoided.
+
+2. Write breakpoint - A write breakpoint is triggered when memory
+ location at the breakpoint address is written.
+
+ A write or can be placed for data of variable length. Length of
+ a write breakpoint indicates length of the datatype to be
+ watched. Length is 1 for 1 byte data , 2 for 2 byte data, 3 for
+ 4 byte data.
+
+3. Access breakpoint - An access breakpoint is triggered when memory
+ location at the breakpoint address is either read or written.
+
+ Access breakpoints also have lengths similar to write breakpoints.
+
+IO breakpoints in ia-32 are not supported.
+
+Since gdb stub at present does not use the protocol used by gdb for hardware
+breakpoints, hardware breakpoints are accessed through gdb macros. gdb macros
+for hardware breakpoints are described below.
+
+hwebrk - Places an execution breakpoint
+ hwebrk breakpointno address
+hwwbrk - Places a write breakpoint
+ hwwbrk breakpointno length address
+hwabrk - Places an access breakpoint
+ hwabrk breakpointno length address
+hwrmbrk - Removes a breakpoint
+ hwrmbrk breakpointno
+exinfo - Tells whether a software or hardware breakpoint has occurred.
+ Prints number of the hardware breakpoint if a hardware breakpoint has
+ occurred.
+
+Arguments required by these commands are as follows
+breakpointno - 0 to 3
+length - 1 to 3
+address - Memory location in hex digits ( without 0x ) e.g c015e9bc
+
+SMP support
+==========
+
+When a breakpoint occurs or user issues a break ( Ctrl + C ) to gdb
+client, all the processors are forced to enter the debugger. Current
+thread corresponds to the thread running on the processor where
+breakpoint occurred. Threads running on other processor(s) appear
+similar to other non-running threads in the 'info threads' output.
+Within the kgdb stub there is a structure "waiting_cpus" in which kgdb
+records the values of "current" and "regs" for each CPU other than the
+one that hit the breakpoint. "current" is a pointer to the task
+structure for the task that CPU is running, while "regs" points to the
+saved registers for the task. This structure can be examined with the
+gdb "p" command.
+
+ia-32 hardware debugging registers on all processors are set to same
+values. Hence any hardware breakpoints may occur on any processor.
+
+gdb troubleshooting
+===================
+
+1. gdb hangs
+Kill it. restart gdb. Connect to target machine.
+
+2. gdb cannot connect to target machine (after killing a gdb and
+restarting another) If the target machine was not inside debugger when
+you killed gdb, gdb cannot connect because the target machine won't
+respond. In this case echo "Ctrl+C"(ASCII 3) to the serial line.
+e.g. echo -e "\003" > /dev/ttyS1
+This forces that target machine into the debugger, after which you
+can connect.
+
+3. gdb cannot connect even after echoing Ctrl+C into serial line
+Try changing serial line settings min to 1 and time to 0
+e.g. stty min 1 time 0 < /dev/ttyS1
+Try echoing again
+
+Check serial line speed and set it to correct value if required
+e.g. stty ispeed 115200 ospeed 115200 < /dev/ttyS1
+
+EVENTS
+======
+
+Ever want to know the order of things happening? Which CPU did what and
+when? How did the spinlock get the way it is? Then events are for
+you. Events are defined by calls to an event collection interface and
+saved for later examination. In this case, kgdb events are saved by a
+very fast bit of code in kgdb which is fully SMP and interrupt protected
+and they are examined by using gdb to display them. Kgdb keeps only
+the last N events, where N must be a power of two and is defined at
+configure time.
+
+
+Events are signaled to kgdb by calling:
+
+kgdb_ts(data0,data1)
+
+For each call kgdb records each call in an array along with other info.
+Here is the array definition:
+
+struct kgdb_and_then_struct {
+#ifdef CONFIG_SMP
+ int on_cpu;
+#endif
+ long long at_time;
+ int from_ln;
+ char * in_src;
+ void *from;
+ int with_if;
+ int data0;
+ int data1;
+};
+
+For SMP machines the CPU is recorded, for all machines the TSC is
+recorded (gets a time stamp) as well as the line number and source file
+the call was made from. The address of the (from), the "if" (interrupt
+flag) and the two data items are also recorded. The macro kgdb_ts casts
+the types to int, so you can put any 32-bit values here. There is a
+configure option to select the number of events you want to keep. A
+nice number might be 128, but you can keep up to 1024 if you want. The
+number must be a power of two. An "andthen" macro library is provided
+for gdb to help you look at these events. It is also possible to define
+a different structure for the event storage and cast the data to this
+structure. For example the following structure is defined in kgdb:
+
+struct kgdb_and_then_struct2 {
+#ifdef CONFIG_SMP
+ int on_cpu;
+#endif
+ long long at_time;
+ int from_ln;
+ char * in_src;
+ void *from;
+ int with_if;
+ struct task_struct *t1;
+ struct task_struct *t2;
+};
+
+If you use this for display, the data elements will be displayed as
+pointers to task_struct entries. You may want to define your own
+structure to use in casting. You should only change the last two items
+and you must keep the structure size the same. Kgdb will handle these
+as 32-bit ints, but within that constraint you can define a structure to
+cast to any 32-bit quantity. This need only be available to gdb and is
+only used for casting in the display code.
+
+Final Items
+===========
+
+I picked up this code from Amit S. Kale and enhanced it.
+
+If you make some really cool modification to this stuff, or if you
+fix a bug, please let me know.
+
+George Anzinger
+<george@mvista.com>
+
+Amit S. Kale
+<akale@veritas.com>
+
+(First kgdb by David Grothe <dave@gcom.com>)
+
+(modified by Tigran Aivazian <tigran@sco.com>)
+ Putting gdbstub into the kernel config menu.
+
+(modified by Scott Foehner <sfoehner@engr.sgi.com>)
+ Hooks for entering gdbstub at boot time.
+
+(modified by Amit S. Kale <akale@veritas.com>)
+ Threads, ia-32 hw debugging, mp support, console support,
+ nmi watchdog handling.
+
+(modified by George Anzinger <george@mvista.com>)
+ Extended threads to include the idle threads.
+ Enhancements to allow breakpoint() at first C code.
+ Use of module_init() and __setup() to automate the configure.
+ Enhanced the cpu "collection" code to work in early bring-up.
+ Added ability to call functions from gdb
+ Print info thread stuff without going back to schedule()
+ Now collect the "other" cpus with an IPI/ NMI.
--- diff/Documentation/i386/kgdb/kgdbeth.txt 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/i386/kgdb/kgdbeth.txt 2004-03-16 11:27:47.985892696 +0000
@@ -0,0 +1,92 @@
+KGDB over ethernet
+==================
+
+Authors
+-------
+
+Robert Walsh <rjwalsh@durables.org> (2.6 port)
+wangdi <wangdi@clusterfs.com> (2.6 port)
+Matt Mackall <mpm@selenic.com> (netpoll api)
+San Mehat (original 2.4 code)
+
+
+Introduction
+------------
+
+KGDB supports debugging over ethernet (kgdboe) via polling of a given
+network interface. Most cards should be supported automatically.
+Debugging facilities are available as soon as the network driver and
+kgdboe have initialized. Unfortunately, this is too late in the boot
+process for debugging some issues, but works quite well for many
+others. This should not interfere with normal network usage and
+doesn't require a dedicated NIC.
+
+Terminology
+-----------
+
+This document uses the following terms:
+
+ TARGET: the machine being debugged.
+ HOST: the machine running gdb.
+
+
+Usage
+-----
+
+You need to use the following command-line option on the TARGET kernel:
+
+ kgdboe=[tgt-port]@<tgt-ip>/[dev],[host-port]@<host-ip>/[host-macaddr]
+
+ where
+ tgt-port source for UDP packets (defaults to 6443)
+ tgt-ip source IP to use (interface address)
+ dev network interface (eth0)
+ host-port HOST UDP port (6442) (not really used)
+ host-ip IP address for HOST machine
+ host-macaddr ethernet MAC address for HOST (ff:ff:ff:ff:ff:ff)
+
+ examples:
+
+ kgdboe=7000@192.168.0.1/eth1,7001@192.168.0.2/00:05:3C:04:47:5D
+ this machine is 192.168.0.1 on eth1
+ remote machine is 192.168.0.2 with MAC address 00:05:3C:04:47:5D
+ listen for gdb packets on port 7000
+ send unsolicited gdb packets to port 7001
+
+ kgdboe=@192.168.0.1/,@192.168.0.2/
+ this machine is 192.168.0.1 on default interface eth0
+ remote machine is 192.168.0.2, use default broadcast MAC address
+ listen for gdb packets on default port 6443
+ send unsolicited gdb packets to port 6442
+
+Only packets originating from the configured HOST IP address will be
+accepted by the debugger.
+
+On the HOST side, run gdb as normal and use a remote UDP host as the
+target:
+
+ % gdb ./vmlinux
+ GNU gdb Red Hat Linux (5.3post-0.20021129.18rh)
+ Copyright 2003 Free Software Foundation, Inc.
+ GDB is free software, covered by the GNU General Public License, and you are
+ welcome to change it and/or distribute copies of it under certain conditions.
+ Type "show copying" to see the conditions.
+ There is absolutely no warranty for GDB. Type "show warranty" for details.
+ This GDB was configured as "i386-redhat-linux-gnu"...
+ (gdb) target remote udp:HOSTNAME:6443
+
+You can now continue as if you were debugging over a serial line.
+
+Limitations
+-----------
+
+The current release of this code is exclusive of using kgdb on a
+serial interface, so you must boot without the kgdboe option to use
+serial debugging. Trying to debug the network driver while using it
+will prove interesting.
+
+Bug reports
+-----------
+
+Send bug reports to Robert Walsh <rjwalsh@durables.org> and Matt
+Mackall <mpm@selenic.com>.
--- diff/Documentation/i386/kgdb/loadmodule.sh 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/i386/kgdb/loadmodule.sh 2004-03-16 11:27:47.988892240 +0000
@@ -0,0 +1,78 @@
+#/bin/sh
+# This script loads a module on a target machine and generates a gdb script.
+# source generated gdb script to load the module file at appropriate addresses
+# in gdb.
+#
+# Usage:
+# Loading the module on target machine and generating gdb script)
+# [foo]$ loadmodule.sh <modulename>
+#
+# Loading the module file into gdb
+# (gdb) source <gdbscriptpath>
+#
+# Modify following variables according to your setup.
+# TESTMACHINE - Name of the target machine
+# GDBSCRIPTS - The directory where a gdb script will be generated
+#
+# Author: Amit S. Kale (akale@veritas.com).
+#
+# If you run into problems, please check files pointed to by following
+# variables.
+# ERRFILE - /tmp/<modulename>.errs contains stderr output of insmod
+# MAPFILE - /tmp/<modulename>.map contains stdout output of insmod
+# GDBSCRIPT - $GDBSCRIPTS/load<modulename> gdb script.
+
+TESTMACHINE=foo
+GDBSCRIPTS=/home/bar
+
+if [ $# -lt 1 ] ; then {
+ echo Usage: $0 modulefile
+ exit
+} ; fi
+
+MODULEFILE=$1
+MODULEFILEBASENAME=`basename $1`
+
+if [ $MODULEFILE = $MODULEFILEBASENAME ] ; then {
+ MODULEFILE=`pwd`/$MODULEFILE
+} fi
+
+ERRFILE=/tmp/$MODULEFILEBASENAME.errs
+MAPFILE=/tmp/$MODULEFILEBASENAME.map
+GDBSCRIPT=$GDBSCRIPTS/load$MODULEFILEBASENAME
+
+function findaddr() {
+ local ADDR=0x$(echo "$SEGMENTS" | \
+ grep "$1" | sed 's/^[^ ]*[ ]*[^ ]*[ ]*//' | \
+ sed 's/[ ]*[^ ]*$//')
+ echo $ADDR
+}
+
+function checkerrs() {
+ if [ "`cat $ERRFILE`" != "" ] ; then {
+ cat $ERRFILE
+ exit
+ } fi
+}
+
+#load the module
+echo Copying $MODULEFILE to $TESTMACHINE
+rcp $MODULEFILE root@${TESTMACHINE}:
+
+echo Loading module $MODULEFILE
+rsh -l root $TESTMACHINE /sbin/insmod -m ./`basename $MODULEFILE` \
+ > $MAPFILE 2> $ERRFILE
+checkerrs
+
+SEGMENTS=`head -n 11 $MAPFILE | tail -n 10`
+TEXTADDR=$(findaddr "\\.text[^.]")
+LOADSTRING="add-symbol-file $MODULEFILE $TEXTADDR"
+SEGADDRS=`echo "$SEGMENTS" | awk '//{
+ if ($1 != ".text" && $1 != ".this" &&
+ $1 != ".kstrtab" && $1 != ".kmodtab") {
+ print " -s " $1 " 0x" $3 " "
+ }
+}'`
+LOADSTRING="$LOADSTRING $SEGADDRS"
+echo Generating script $GDBSCRIPT
+echo $LOADSTRING > $GDBSCRIPT
--- diff/Documentation/laptop-mode.txt 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/laptop-mode.txt 2004-03-16 11:27:47.996891024 +0000
@@ -0,0 +1,480 @@
+How to conserve battery power using laptop-mode
+-----------------------------------------------
+
+Document Author: Bart Samwel (bart@samwel.tk)
+Date created: January 2, 2004
+
+Introduction
+------------
+
+Laptopmode is used to minimize the time that the hard disk needs to be spun up,
+to conserve battery power on laptops. It has been reported to cause significant
+power savings.
+
+Contents
+--------
+
+* Introduction
+* The short story
+* Caveats
+* The details
+* Tips & Tricks
+* Control script
+* ACPI integration
+* Monitoring tool
+
+
+The short story
+---------------
+
+If you just want to use it, run the laptop_mode control script (which is included
+at the end of this document) as follows:
+
+# laptop_mode start
+
+Then set your harddisk spindown time to a relatively low value with hdparm:
+
+hdparm -S 4 /dev/hda
+
+The value -S 4 means 20 seconds idle time before spindown. Your harddisk will
+now only spin up when a disk cache miss occurs, or at least once every 10
+minutes to write back any pending changes.
+
+To stop laptop_mode, remount your filesystems with regular commit intervals
+(e.g., 5 seconds), and run "laptop_mode stop".
+
+
+Caveats
+-------
+
+* The downside of laptop mode is that you have a chance of losing up
+ to 10 minutes of work. If you cannot afford this, don't use it!
+
+* Most desktop hard drives have a very limited lifetime measured in spindown
+ cycles, typically about 50.000 times (it's usually listed on the spec sheet).
+ Check your drive's rating, and don't wear down your drive's lifetime if you
+ don't need to.
+
+* If you mount some of your ext3/reiserfs filesystems with the -n option, then
+ the control script will not be able to remount them correctly. You must set
+ DO_REMOUNTS=0 in the control script, otherwise it will remount them with the
+ wrong options -- or it will fail because it cannot write to /etc/mtab.
+
+* If you have your filesystems listed as type "auto" in fstab, like I did, then
+ the control script will not recognize them as filesystems that need remounting.
+
+The details
+-----------
+
+Laptop-mode is controlled by the flag /proc/sys/vm/laptop_mode. When this
+flag is set, any physical disk read operation (that might have caused the
+hard disk to spin up) causes Linux to flush all dirty blocks. The result
+of this is that after a disk has spun down, it will not be spun up anymore
+to write dirty blocks, because those blocks had already been written
+immediately after the most recent read operation
+
+To increase the effectiveness of the laptop_mode strategy, the laptop_mode
+control script increases dirty_expire_centisecs and dirty_writeback_centisecs in
+/proc/sys/vm to about 10 minutes (by default), which means that pages that are
+dirtied are not forced to be written to disk as often. The control script also
+changes the dirty background ratio, so that background writeback of dirty pages
+is not done anymore. Combined with a higher commit value (also 10 minutes) for
+ext3 or ReiserFS filesystems (also done automatically by the control script),
+this results in concentration of disk activity in a small time interval which
+occurs only once every 10 minutes, or whenever the disk is forced to spin up by
+a cache miss. The disk can then be spun down in the periods of inactivity.
+
+If you want to find out which process caused the disk to spin up, you can
+gather information by setting the flag /proc/sys/vm/block_dump. When this flag
+is set, Linux reports all disk read and write operations that take place, and
+all block dirtyings done to files. This makes it possible to debug why a disk
+needs to spin up, and to increase battery life even more.
+
+If 10 minutes is too much or too little downtime for you, you can configure
+this downtime as follows. In the control script, set the MAX_AGE value to the
+maximum number of seconds of disk downtime that you would like. You should
+then set your filesystem's commit interval to the same value. The dirty ratio
+is also configurable from the control script.
+
+If you don't like the idea of the control script remounting your filesystems
+for you, you can change DO_REMOUNTS to 0 in the script.
+
+Thanks to Kiko Piris, the control script can be used to enable laptop mode on
+both the Linux 2.4 and 2.6 series.
+
+
+Tips & Tricks
+-------------
+
+* Bartek Kania reports getting up to 50 minutes of extra battery life (on top
+ of his regular 3 to 3.5 hours) using very aggressive power management (hdparm
+ -B1) and a spindown time of 5 seconds (hdparm -S1).
+
+* You can spin down the disk while playing MP3, by setting the disk readahead
+ to 8MB (hdparm -a 16384). Effectively, the disk will read a complete MP3 at
+ once, and will then spin down while the MP3 is playing. (Thanks to Bartek
+ Kania.)
+
+* Drew Scott Daniels observed: "I don't know why, but when I decrease the number
+ of colours that my display uses it consumes less battery power. I've seen
+ this on powerbooks too. I hope that this is a piece of information that
+ might be useful to the Laptop Mode patch or it's users."
+
+
+Control script
+--------------
+
+Please note that this control script works for the Linux 2.4 and 2.6 series.
+
+--------------------CONTROL SCRIPT BEGIN------------------------------------------
+#!/bin/sh
+
+# start or stop laptop_mode, best run by a power management daemon when
+# ac gets connected/disconnected from a laptop
+#
+# install as /sbin/laptop_mode
+#
+# Contributors to this script: Kiko Piris
+# Bart Samwel
+# Dax Kelson
+# Original Linux 2.4 version by: Jens Axboe
+
+parse_mount_opts () {
+ echo "$*" | \
+ sed 's/commit=[0-9]*//g' | \
+ sed 's/,,*/,/g' | \
+ sed 's/^,//' | \
+ sed 's/,$//' | \
+ cat -
+}
+
+KLEVEL="$(uname -r | cut -c1-3)"
+case "$KLEVEL" in
+ "2.4")
+ true
+ ;;
+ "2.6")
+ true
+ ;;
+ *)
+ echo "Unhandled kernel level: $KLEVEL ('uname -r' = '$(uname -r)')"
+ exit 1
+ ;;
+esac
+
+# Shall we remount journaled fs. with appropiate commit interval? (1=yes)
+DO_REMOUNTS=1
+
+# age time, in seconds. should be put into a sysconfig file
+MAX_AGE=600
+
+# Allowed dirty ratio, in pct. should be put into a sysconfig file as well.
+DIRTY_RATIO=40
+
+# kernel default dirty buffer age
+DEF_AGE=30
+DEF_UPDATE=5
+DEF_DIRTY_BACKGROUND_RATIO=10
+DEF_DIRTY_RATIO=40
+
+
+if [ ! -e /proc/sys/vm/laptop_mode ]; then
+ echo "Kernel is not patched with laptop_mode patch."
+ exit 1
+fi
+
+if [ ! -w /proc/sys/vm/laptop_mode ]; then
+ echo "You do not have enough privileges to enable laptop_mode."
+ exit 1
+fi
+
+case "$1" in
+ start)
+ AGE=$((100*$MAX_AGE))
+ echo -n "Starting laptop_mode"
+ case "$KLEVEL" in
+ "2.4")
+ echo "1" > /proc/sys/vm/laptop_mode
+ echo "30 500 0 0 $AGE $AGE 60 20 0" > /proc/sys/vm/bdflush
+ ;;
+ "2.6")
+ echo "1" > /proc/sys/vm/laptop_mode
+ echo "$AGE" > /proc/sys/vm/dirty_writeback_centisecs
+ echo "$AGE" > /proc/sys/vm/dirty_expire_centisecs
+ echo "$DIRTY_RATIO" > /proc/sys/vm/dirty_ratio
+ echo "$DIRTY_RATIO" > /proc/sys/vm/dirty_background_ratio
+ ;;
+ esac
+ if [ $DO_REMOUNTS -eq 1 ]; then
+ cat /etc/mtab | while read DEV MP FST OPTS DUMP PASS ; do
+ PARSEDOPTS="$(parse_mount_opts "$OPTS")"
+ case "$FST" in
+ "ext3") mount $DEV -t $FST $MP -o remount,$PARSEDOPTS,commit=$MAX_AGE ;;
+ "reiserfs") mount $DEV -t $FST $MP -o remount,$PARSEDOPTS,commit=$MAX_AGE ;;
+ "xfs") mount $DEV -t $FST $MP -o remount,$PARSEDOPTS,commit=$MAX_AGE ;;
+ esac
+ done
+ fi
+ echo "."
+ ;;
+ stop)
+ U_AGE=$((100*$DEF_UPDATE))
+ B_AGE=$((100*$DEF_AGE))
+ echo -n "Stopping laptop_mode"
+ case "$KLEVEL" in
+ "2.4")
+ echo "0" > /proc/sys/vm/laptop_mode
+ echo "30 500 0 0 $U_AGE $B_AGE 60 20 0" > /proc/sys/vm/bdflush
+ ;;
+ "2.6")
+ echo "0" > /proc/sys/vm/laptop_mode
+ echo "$U_AGE" > /proc/sys/vm/dirty_writeback_centisecs
+ echo "$B_AGE" > /proc/sys/vm/dirty_expire_centisecs
+ echo "$DEF_DIRTY_RATIO" > /proc/sys/vm/dirty_ratio
+ echo "$DEF_DIRTY_BACKGROUND_RATIO" > /proc/sys/vm/dirty_background_ratio
+ ;;
+ esac
+ if [ $DO_REMOUNTS -eq 1 ]; then
+ cat /etc/mtab | while read DEV MP FST OPTS DUMP PASS ; do
+ PARSEDOPTS="$(parse_mount_opts "$OPTS")"
+ case "$FST" in
+ "ext3") mount $DEV -t $FST $MP -o remount,$PARSEDOPTS ;;
+ "reiserfs") mount $DEV -t $FST $MP -o remount,$PARSEDOPTS ;;
+ "xfs") mount $DEV -t $FST $MP -o remount,$PARSEDOPTS ;;
+ esac
+ done
+ fi
+ echo "."
+ ;;
+ *)
+ echo "$0 {start|stop}"
+ ;;
+
+esac
+
+exit 0
+
+--------------------CONTROL SCRIPT END--------------------------------------------
+
+
+ACPI integration
+----------------
+
+Dax Kelson submitted this so that the ACPI acpid daemon will
+kick off the laptop_mode script and run hdparm.
+
+---------------------------/etc/acpi/events/ac_adapter BEGIN-------------------------------------------
+event=ac_adapter
+action=/etc/acpi/actions/battery.sh
+---------------------------/etc/acpi/events/ac_adapter END-------------------------------------------
+
+---------------------------/etc/acpi/actions/battery.sh BEGIN-------------------------------------------
+#!/bin/sh
+
+# cpu throttling
+# cat /proc/acpi/processor/CPU0/throttling for more info
+ACAD_THR=0
+BATT_THR=2
+
+# spindown time for HD (man hdparm for valid values)
+# I prefer 2 hours for acad and 20 seconds for batt
+ACAD_HD=244
+BATT_HD=4
+
+# ac/battery event handler
+
+status=`awk '/^state: / { print $2 }' /proc/acpi/ac_adapter/AC/state`
+
+case $status in
+ "on-line")
+ echo "Setting HD spindown to 2 hours"
+ /sbin/laptop-mode stop
+ /sbin/hdparm -S $ACAD_HD /dev/hda > /dev/null 2>&1
+ /sbin/hdparm -B 255 /dev/hda > /dev/null 2>&1
+ #echo -n $ACAD_CPU:$ACAD_THR > /proc/acpi/processor/CPU0/limit
+ exit 0
+ ;;
+ "off-line")
+ echo "Setting HD spindown to 20 seconds"
+ /sbin/laptop-mode start
+ /sbin/hdparm -S $BATT_HD /dev/hda > /dev/null 2>&1
+ /sbin/hdparm -B 1 /dev/hda > /dev/null 2>&1
+ #echo -n $BATT_CPU:$BATT_THR > /proc/acpi/processor/CPU0/limit
+ exit 0
+ ;;
+esac
+---------------------------/etc/acpi/actions/battery.sh END-------------------------------------------
+
+Monitoring tool
+---------------
+
+Bartek Kania submitted this, it can be used to measure how much time your disk
+spends spun up/down.
+
+---------------------------dslm.c BEGIN-------------------------------------------
+/*
+ * Simple Disk SLeep Monitor
+ * by Bartek Kania
+ * Licenced under the GPL
+ */
+#include <unistd.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <fcntl.h>
+#include <errno.h>
+#include <time.h>
+#include <string.h>
+#include <signal.h>
+#include <sys/ioctl.h>
+#include <linux/hdreg.h>
+
+#ifdef DEBUG
+#define D(x) x
+#else
+#define D(x)
+#endif
+
+int endit = 0;
+
+/* Check if the disk is in powersave-mode
+ * Most of the code is stolen from hdparm.
+ * 1 = active, 0 = standby/sleep, -1 = unknown */
+int check_powermode(int fd)
+{
+ unsigned char args[4] = {WIN_CHECKPOWERMODE1,0,0,0};
+ int state;
+
+ if (ioctl(fd, HDIO_DRIVE_CMD, &args)
+ && (args[0] = WIN_CHECKPOWERMODE2) /* try again with 0x98 */
+ && ioctl(fd, HDIO_DRIVE_CMD, &args)) {
+ if (errno != EIO || args[0] != 0 || args[1] != 0) {
+ state = -1; /* "unknown"; */
+ } else
+ state = 0; /* "sleeping"; */
+ } else {
+ state = (args[2] == 255) ? 1 : 0;
+ }
+ D(printf(" drive state is: %s\n", state));
+
+ return state;
+}
+
+char *state_name(int i)
+{
+ if (i == -1) return "unknown";
+ if (i == 0) return "sleeping";
+ if (i == 1) return "active";
+
+ return "internal error";
+}
+
+char *myctime(time_t time)
+{
+ char *ts = ctime(&time);
+ ts[strlen(ts) - 1] = 0;
+
+ return ts;
+}
+
+void measure(int fd)
+{
+ time_t start_time;
+ int last_state;
+ time_t last_time;
+ int curr_state;
+ time_t curr_time = 0;
+ time_t time_diff;
+ time_t active_time = 0;
+ time_t sleep_time = 0;
+ time_t unknown_time = 0;
+ time_t total_time = 0;
+ int changes = 0;
+ float tmp;
+
+ printf("Starting measurements\n");
+
+ last_state = check_powermode(fd);
+ start_time = last_time = time(0);
+ printf(" System is in state %s\n\n", state_name(last_state));
+
+ while(!endit) {
+ sleep(1);
+ curr_state = check_powermode(fd);
+
+ if (curr_state != last_state || endit) {
+ changes++;
+ curr_time = time(0);
+ time_diff = curr_time - last_time;
+
+ if (last_state == 1) active_time += time_diff;
+ else if (last_state == 0) sleep_time += time_diff;
+ else unknown_time += time_diff;
+
+ last_state = curr_state;
+ last_time = curr_time;
+
+ printf("%s: State-change to %s\n", myctime(curr_time),
+ state_name(curr_state));
+ }
+ }
+ changes--; /* Compensate for SIGINT */
+
+ total_time = time(0) - start_time;
+ printf("\nTotal running time: %lus\n", curr_time - start_time);
+ printf(" State changed %d times\n", changes);
+
+ tmp = (float)sleep_time / (float)total_time * 100;
+ printf(" Time in sleep state: %lus (%.2f%%)\n", sleep_time, tmp);
+ tmp = (float)active_time / (float)total_time * 100;
+ printf(" Time in active state: %lus (%.2f%%)\n", active_time, tmp);
+ tmp = (float)unknown_time / (float)total_time * 100;
+ printf(" Time in unknown state: %lus (%.2f%%)\n", unknown_time, tmp);
+}
+
+void ender(int s)
+{
+ endit = 1;
+}
+
+void usage()
+{
+ puts("usage: dslm [-w <time>] <disk>");
+ exit(0);
+}
+
+int main(int ac, char **av)
+{
+ int fd;
+ char *disk = 0;
+ int settle_time = 60;
+
+ /* Parse the simple command-line */
+ if (ac == 2)
+ disk = av[1];
+ else if (ac == 4) {
+ settle_time = atoi(av[2]);
+ disk = av[3];
+ } else
+ usage();
+
+ if (!(fd = open(disk, O_RDONLY|O_NONBLOCK))) {
+ printf("Can't open %s, because: %s\n", disk, strerror(errno));
+ exit(-1);
+ }
+
+ if (settle_time) {
+ printf("Waiting %d seconds for the system to settle down to "
+ "'normal'\n", settle_time);
+ sleep(settle_time);
+ } else
+ puts("Not waiting for system to settle down");
+
+ signal(SIGINT, ender);
+
+ measure(fd);
+
+ close(fd);
+
+ return 0;
+}
+---------------------------dslm.c END---------------------------------------------
--- diff/Documentation/must-fix.txt 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/must-fix.txt 2004-03-16 11:27:47.997890872 +0000
@@ -0,0 +1,288 @@
+
+Must-fix bugs
+=============
+
+drivers/char/
+~~~~~~~~~~~~~
+
+o TTY locking is broken.
+
+ o see FIXME in do_tty_hangup(). This causes ppp BUGs in local_bh_enable()
+
+ o Other problems: aviro, dipankar, Alan have details.
+
+ o somebody will have to document the tty driver and ldisc API
+
+drivers/tty
+~~~~~~~~~~~
+
+o viro: tty_driver refcounting, tty/misc/upper levels of sound still not
+ completely fixed.
+
+drivers/block/
+~~~~~~~~~~~~~~
+
+o loop.c: Concurrent write access on block devices might cause a deadlock
+ of the complete system. See:
+ http://marc.theaimsgroup.com/?l=linux-kernel&m=106275365925769&w==
+ http://bugzilla.kernel.org/show_bug.cgi?id=1198
+ Thread of possible fix:
+ http://www.kerneli.org/pipermail/cryptoapi-devel/2003-October/000676.html
+
+ (Fruhwirth Clemens)
+
+o ideraid hasn't been ported to 2.5 at all yet.
+
+ We need to understand whether the proposed BIO split code will suffice
+ for this.
+
+drivers/input/
+~~~~~~~~~~~~~~
+
+o rmk: unconverted keyboard/mouse drivers (there's a deadline of 2.6.0
+ currently on these remaining in my/Linus' tree.)
+
+o viro: large absence of locking.
+
+o viro: parport is nearly as bad as that and there the code is more hairy.
+ IMO parport is more of "figure out what API changes are needed for its
+ users, get them done ASAP, then fix generic layer at leisure"
+
+o (Albert Cahalan) Lots of people (check Google) get this message from the
+ kernel:
+
+ psmouse.c: Lost synchronization, throwing 2 bytes away.
+
+ (the number of bytes will be 1, 2, or 3)
+
+ At work, I get it when there is heavy NFS traffic. The mouse goes crazy,
+ jumping around and doing random cut-and-paste all over everything. This
+ is with a decently fast and modern PC.
+
+o There seem to be too many reports of keyboards and mice failing or acting
+ strangely.
+
+
+drivers/misc/
+~~~~~~~~~~~~~
+
+o rmk: UCB1[23]00 drivers, currently sitting in drivers/misc in the ARM
+ tree. (touchscreen, audio, gpio, type device.)
+
+ These need to be moved out of drivers/misc/ and into real places
+
+o viro: actually, misc.c has a good chance to die. With cdev-cidr that's
+ trivial.
+
+drivers/net/
+~~~~~~~~~~~~
+
+drivers/net/irda/
+~~~~~~~~~~~~~~~~~
+
+ (Jean Tourrilhes)
+
+o irport need to be converted to sir-kthread
+
+o dongle drivers need to be converted to sir-dev (in progress)
+
+o new drivers (irtty-sir/smsc-ircc2/donauboe) need more testing (in progress)
+
+
+drivers/pci/
+~~~~~~~~~~~~
+
+o alan: Some cardbus crashes the system
+
+ (bugzilla, please?)
+
+drivers/pcmcia/
+~~~~~~~~~~~~~~~
+
+o alan: This is a locking disaster.
+
+ (rmk, brodo: in progress)
+
+drivers/pld/
+~~~~~~~~~~~~
+
+o rmk: EPXA (ARM platform) PLD hotswap drivers (drivers/pld)
+
+ (rmk: will work out what to do here. maybe drivers/arm/)
+
+drivers/video/
+~~~~~~~~~~~~~~
+
+o Lots of drivers don't compile, others do but don't work.
+
+drivers/scsi/
+~~~~~~~~~~~~~
+
+o Convert am53c974, dpt_i2o, initio and pci2220i to DMA-mapping
+
+o Make inia100, cpqfc, pci2000 and dc390t compile
+
+o Convert
+
+ wd33c99 based: a2091 a3000 gpv11 mvme174 sgiwd93
+
+ 53c7xx based: amiga7xxx bvme6000 mvme16x initio am53c974 pci2000
+ pci2220i dc390t
+
+ To new error handling
+
+ It also might be possible to shift the 53c7xx based drivers over to
+ 53c700 which does the new EH stuff, but I don't have the hardware to check
+ such a shift.
+
+ For the non-compiling stuff, I've probably missed a few that just aren't
+ compilable on my platforms, so any updates would be welcome. Also, are
+ some of our non-compiling or unconverted drivers obsolete?
+
+fs/
+~~~
+
+o AIO/direct-IO writes can race with truncate and wreck filesystems.
+ (Badari has a patch)
+
+o viro: fs/char_dev.c needs removal of aeb stuff and merge of cdev-cidr.
+ In progress.
+
+o forward-port sct's O_DIRECT fixes (Badari has a patch)
+
+o viro: there is some generic stuff for namei/namespace/super, but that's a
+ slow-merge and can go in 2.6 just fine
+
+o trond: NFS has a mmap-versus-truncate problem (fixed? needs testing)
+
+o trond: NFSv4 client, bugs in lockd, RPSEC_GSS for NFSv[23], some atomic open
+ bits. more info: http://www.fys.uio.no/~trondmy/src/Linux-2.6.x/2.6.0-test11/
+
+kernel/sched.c
+~~~~~~~~~~~~~~
+
+o Starvation, general interactivity need close monitoring.
+
+o SMT aware scheduler (Ingo, Rusty, Nick have implementations)
+
+kernel/
+~~~~~~~
+
+o Alan: 32bit uid support is *still* broken for process accounting.
+
+ Create a 32bit uid, turn accounting on. Shock horror it doesn't work
+ because the field is 16bit. We need an acct structure flag day for 2.6
+ IMHO
+
+ (alan has patch)
+
+o viro: core sysctl code is racy. And its interaction wiuth sysfs
+
+o (ingo) rwsems (on x86) are limited to 32766 waiting processes. This
+ means that setting pid_max to above 32K is unsafe :-(
+
+ An option is to use CONFIG_RWSEM_GENERIC_SPINLOCK variant all the time,
+ for all archs, and not inline any part of the ops.
+
+lib/kobject.c
+~~~~~~~~~~~~~
+
+o kobject refcounting (comments from Al Viro):
+
+ _anything_ can grab a temporary reference to kobject. IOW, if kobject is
+ embedded into something that could be freed - it _MUST_ have a destructor
+ and that destructor _MUST_ be the destructor for containing object.
+
+ Any violation of the above (and we already have a bunch of those) is a
+ user-triggerable memory corruption.
+
+ We can tolerate it for a while in 2.5 (e.g. during work on susbsystem we
+ can decide to switch to that way of handling objects and have subsystem
+ vulnerable for a while), but all such windows must be closed before 2.6
+ and during 2.6 we can't open them at all.
+
+o All block drivers which control multiple gendisks with a single
+ request_queue are broken, due to one-to-one assumptions in the request
+ queue sysfs hookup.
+
+mm/
+~~~
+
+o GFP_DMA32 (or something like that). Lots of ideas. jejb, zaitcev,
+ willy, arjan, wli.
+
+ Specifically, 64-bit systems need to be able to enforce 32-bit addressing
+ limits for device metadata like network cards' ring buffers and SCSI
+ command descriptors.
+
+o access_process_vm() doesn't flush right. We probably need new flushing
+ primitives to do this (davem?)
+
+
+modules
+~~~~~~~
+
+ (Rusty)
+
+net/
+~~~~
+
+ (davem)
+
+o UDP apps can in theory deadlock, because the ip_append_data path can end
+ up sleeping while the socket lock is held.
+
+ It is OK to sleep with the socket held held, normally. But in this case
+ the sleep happens while waiting for socket memory/space to become
+ available, if another context needs to take the socket lock to free up the
+ space we could hang.
+
+ I sent a rough patch on how to fix this to Alexey, and he is analyzing
+ the situation. I expect a final fix from him next week or so.
+
+o Semantics for IPSEC during operations such as TCP connect suck currently.
+
+ When we first try to connect to a destination, we may need to ask the
+ IPSEC key management daemon to resolve the IPSEC routes for us. For the
+ purposes of what the kernel needs to do, you can think of it like ARP. We
+ can't send the packet out properly until we resolve the path.
+
+ What happens now for IPSEC is basically this:
+
+ O_NONBLOCK: returns -EAGAIN over and over until route is resolved
+
+ !O_NONBLOCK: Sleeps until route is resolved
+
+ These semantics are total crap. The solution, which Alexey is working
+ on, is to allow incomplete routes to exist. These "incomplete" routes
+ merely put the packet onto a "resolution queue", and once the key manager
+ does it's thing we finish the output of the packet. This is precisely how
+ ARP works.
+
+ I don't know when Alexey will be done with this.
+
+net/*/netfilter/
+~~~~~~~~~~~~~~~~
+
+ (Rusty)
+
+sound/
+~~~~~~
+
+global
+~~~~~~
+
+o viro: 64-bit dev_t (not a mustfix for 2.6.0). 32-bit dev_t is done, 64-bit
+ means extra work on nfsd/raid/etc.
+
+o alan: Forward port 2.4 fixes
+ - Chris Wright: Security fixes including execve holes, execve vs proc races
+
+o There are about 60 or 70 security related checks that need doing
+ (copy_user etc) from Stanford tools. (badari is looking into this, and
+ hollisb)
+
+o A couple of hundred real looking bugzilla bugs
+
+o viro: cdev rework. Mostly done.
+
--- diff/Documentation/power/tricks.txt 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/power/tricks.txt 2004-03-16 11:27:47.997890872 +0000
@@ -0,0 +1,25 @@
+ swsusp/S3 tricks
+ ~~~~~~~~~~~~~~~~
+Pavel Machek <pavel@suse.cz>
+
+If you want to trick swsusp/S3 into working, you might want to try:
+
+* go with minimal config, turn off drivers like USB, AGP you don't
+ really need
+
+* use ext2. At least it has working fsck. [If something seemes to go
+ wrong, force fsck when you have a chance]
+
+* turn off modules
+
+* use vga text console, shut down X. [If you really want X, you might
+ want to try vesafb later]
+
+* try running as few processes as possible, preferably go to single
+ user mode.
+
+* due to video issues, swsusp should be easier to get working than
+ S3. Try that first.
+
+When you make it work, try to find out what exactly was it that broke
+suspend, and preferably fix that.
--- diff/Documentation/sched-domains.txt 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/sched-domains.txt 2004-03-16 11:27:47.998890720 +0000
@@ -0,0 +1,55 @@
+Each CPU has a "base" scheduling domain (struct sched_domain). These are
+accessed via cpu_sched_domain(i) and this_sched_domain() macros. The domain
+hierarchy is built from these base domains via the ->parent pointer. ->parent
+MUST be NULL terminated, and domain structures should be per-CPU as they
+are locklessly updated.
+
+Each scheduling domain spans a number of CPUs (stored in the ->span field).
+A domain's span MUST be a superset of it child's span, and a base domain
+for CPU i MUST span at least i. The top domain for each CPU will generally
+span all CPUs in the system although strictly it doesn't have to, but this
+could lead to a case where some CPUs will never be given tasks to run unless
+the CPUs allowed mask is explicitly set. A sched domain's span means "balance
+process load among these CPUs".
+
+Each scheduling domain must have one or more CPU groups (struct sched_group)
+which are organised as a circular one way linked list from the ->groups
+pointer. The union of cpumasks of these groups MUST be the same as the
+domain's span. The intersection of cpumasks from any two of these groups
+MUST be the empty set. The group pointed to by the ->groups pointer MUST
+contain the CPU to which the domain belongs. Groups may be shared among
+CPUs as they contain read only data after they have been set up.
+
+Balancing within a sched domain occurs between groups. That is, each group
+is treated as one entity. The load of a group is defined as the sum of the
+load of each of its member CPUs, and only when the load of a group becomes
+out of balance are tasks moved between groups.
+
+In kernel/sched.c, rebalance_tick is run periodically on each CPU. This
+function takes its CPU's base sched domain and checks to see if has reached
+its rebalance interval. If so, then it will run load_balance on that domain.
+rebalance_tick then checks the parent sched_domain (if it exists), and the
+parent of the parent and so forth.
+
+*** Implementing sched domains ***
+The "base" domain will "span" the first level of the hierarchy. In the case
+of SMT, you'll span all siblings of the physical CPU, with each group being
+a single virtual CPU.
+
+In SMP, the parent of the base domain will span all physical CPUs in the
+node. Each group being a single physical CPU. Then with NUMA, the parent
+of the SMP domain will span the entire machine, with each group having the
+cpumask of a node. Or, you could do multi-level NUMA or Opteron, for example,
+might have just one domain covering its one NUMA level.
+
+The implementor should read comments in include/linux/sched.h:
+struct sched_domain fields, SD_FLAG_*, SD_*_INIT to get an idea of
+the specifics and what to tune.
+
+Implementors should change the line
+#undef SCHED_DOMAIN_DEBUG
+to
+#define SCHED_DOMAIN_DEBUG
+in kernel/sched.c as this enables an error checking parse of the sched domains
+which should catch most possible errors (described above). It also prints out
+the domain structure in a visual format.
--- diff/Documentation/should-fix.txt 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/should-fix.txt 2004-03-16 11:27:47.999890568 +0000
@@ -0,0 +1,545 @@
+Not-ready features and speedups
+===============================
+
+Legend:
+
+PRI1: We're totally lame if this doesn't get in
+PRI2: Would be nice
+PRI3: Not very important
+
+drivers/block/
+~~~~~~~~~~~~~~
+
+o viro: paride drivers need a big cleanup. Partially done, but ATAPI drivers
+ need serious work and bug fixing.
+
+ PRI2
+
+drivers/char/rtc/
+~~~~~~~~~~~~~~~~~
+
+o rmk, trini: add support for alarms to the existing generic rtc driver.
+
+ PRI2
+
+console drivers
+~~~~~~~~~~~~~~~
+ (Pavel Machek <pavel@ucw.cz>)
+
+o There are few must-fix bugs in cursor handling.
+
+o Play with gpm selection for a while and your cursor gets corrupted with
+ random dots. Ouch.
+
+device mapper
+~~~~~~~~~~~~~
+
+o ioctl interface cleanup patch is ready (redo the structure layouts)
+
+ PRI1
+
+o A port of the 2.4 snapshot and mirror targets is in progress
+
+ PRI1
+
+o the fs interface to dm needs to be redone. gregkh was going to work on
+ this. viro is interested in seeing work thus-far.
+
+ PRI2
+
+drivers/net/wireless/
+~~~~~~~~~~~~~~~~~~~~~
+
+ (Jean Tourrilhes <jt@bougret.hpl.hp.com>)
+
+o get HostAP driver in the kernel. No consolidation of the 802.11
+ management across driver can happen until this one is in (which is probably
+ 2.7.X material). I think Jouni is mostly ready but didn't find time for
+ it.
+
+ PRI2
+
+o get more wireless drivers into the kernel. The most "integrable" drivers
+ at this point seem the NWN driver, Pavel's Spectrum driver.
+
+ PRI1
+
+drivers/usb/gadget/
+~~~~~~~~~~~~~~~~~~~
+
+o rmk: SA11xx USB client/gadget code (David B has been doing some work on
+ this, and keeps trying to prod me, but unfortunately I haven't had the time
+ to look at his work, sorry David.)
+
+ PRI3
+
+fs/
+~~~
+
+o ext3 and ext2 block allocators have serious failure modes - interleaved
+ allocations.
+
+ PRI3
+
+o Integrate Chris Mason's 2.4 reiserfs ordered data and data journaling
+ patches. They make reiserfs a lot safer.
+
+ Ordered: PRI2
+ data journalled: PRI3
+
+o viro: convert more filesystems to use lib/parser.c for options.
+
+ PRI2
+
+o aio: fs IO isn't async at present. suparna has restart patches, they're
+ in -mm. Need to get Ben to review/comment.
+
+ PRI1.
+
+o drepper: various filesystems use ->pid wrongly
+
+ PRI1
+
+o hch: devfs: there's a fundamental lookup vs devfsd race that's only
+ fixable by introducing a lookup vs devfs deadlock. I can't see how this is
+ fixable without getting rid of the current devfsd design. Mandrake seems
+ to have a workaround for this so this is at least not triggered so easily,
+ but that's not what I'd consider a fix..
+
+ PRI2
+
+kernel/
+~~~~~~~
+
+o rusty: Zippel's Reference count simplification. Tricky code, but cuts
+ about 120 lines from module.c. Patch exists, needs stressing.
+
+ PRI3
+
+o rusty: Fix module-failed-init races by starting module "disabled". Patch
+ exists, requires some subsystems (ie. add_partition) to explicitly say
+ "make module live now". Without patch we are no worse off than 2.4 etc.
+
+ PRI1
+
+o Integrate userspace irq balancing daemon.
+
+ PRI2
+
+o kexec. Seems to work, was in -mm.
+
+ PRI3
+
+o rmk: lib/inflate.c must not use static variables (causes these to be
+ referenced via GOTOFF relocations in PIC decompressor. We have a PIC
+ decompressor to avoid having to hard code a per platform zImage link
+ address into the makefiles.)
+
+ PRI2
+
+o klibc merge?
+
+ PRI2
+
+mm/
+~~~
+
+o dropbehind for large files
+
+ PRI2
+
+net/
+~~~~
+
+ (davem)
+
+o Real serious use of IPSEC is hampered by lack of MPLS support. MPLS is a
+ switching technology that works by switching based upon fixed length labels
+ prepended to packets. Many people use this and IPSEC to implement VPNs
+ over public networks, it is also used for things like traffic engineering.
+
+ A good reference site is:
+
+ http://www.mplsrc.com/
+
+ Anyways, an existing (crappy) implementation exists. I've almost
+ completed a rewrite, I should have something in the tree next week.
+
+ PRI1
+
+o Sometimes we generate IP fragments when it truly isn't necessary.
+
+ The way IP fragmentation is specified, each fragment must be modulo 8
+ bytes in length. So suppose the device has an MTU that is not 0 modulo 8,
+ ethernet even classifies in this way. 1500 == (8 * 187) + 4
+
+ Our IP fragmenting engine can fragment on packets that are sized within
+ the last modulo 8 bytes of the MTU. This happens in obscure cases, but it
+ does happen.
+
+ I've proposed a fix to Alexey, whereby very late in the output path we
+ check the packet, if we fragmented but the data length would fit into the
+ MTU we unfragment the packet.
+
+ This is low priority, because technically it creates suboptimal behavior
+ rather than mis-operation.
+
+ PRI1
+
+net/*/netfilter/
+~~~~~~~~~~~~~~~~
+
+o Lots of misc. cleanups, which are happening slowly.
+
+ PRI2
+
+power management
+~~~~~~~~~~~~~~~~
+
+o Pat and Pavel disagree over swsusp. Need to sort that out.
+
+ PRI2
+
+o Frame buffer restore codepaths (that requires some deep PCI magic)
+
+ PRI2
+
+o XFree86 hooks
+
+ PRI2
+
+o AGP restoration
+
+ PRI2
+
+o DRI restoration
+
+ (davej/Alan: not super-critical, can crash laptop on restore. davej
+ looking into it.)
+
+ PRI2
+
+o IDE suspend/resume without races (Ben is looking at this a little)
+
+ PRI2
+
+o Pat: There are already CPU device structures; MTRRs should be a
+ dynamically registered interface of CPUs, which implies there needs
+ to be some other glue to know that there are MTRRs that need to be
+ saved/restored.
+
+ PRI1
+
+global
+~~~~~~
+
+o We need a kernel side API for reporting error events to userspace (could
+ be async to 2.6 itself)
+
+ (Prototype core based on netlink exists)
+
+ PRI2
+
+o Kai: Introduce a sane, easy and standard way to build external modules
+ - make clean and make modules_install are both broken
+
+ PRI2
+
+drivers
+~~~~~~~
+
+o Alan: Cardbus/PCMCIA requires all Russell's stuff is merged to do
+ multiheader right and so on
+
+ PRI1
+
+drivers/acpi/
+~~~~~~~~~~~~~
+
+o Fix acpi for all newer IBM Thinkpads see
+ http://bugme.osdl.org/show_bug.cgi?id=1038 for more information
+
+o alan: VIA APIC stuff is one bit of this, there are also some other
+ reports that were caused by ACPI not setting level v edge trigger some
+ times
+
+ PRI1
+
+o mochel: it seems the acpi irq routing code could use a serious rewrite.
+
+ grover: The problem is the ACPI irq routing code is trying to piggyback
+ on the existing MPS-specific data structures, and it's generally a hack.
+ So yes mochel is right, but it is also purging MPS-ities from common code
+ as well. I've done some preliminary work in this area and it doesn't seem
+ to break anything (yet) but a rewrite in this area imho should not be
+ rushed out the door. And, I think the above bugs can be fixed w/o the
+ rewrite.
+
+ PRI2
+
+o mochel: ACPI suspend doesn't work. Important, not cricital. Pat is
+ working it.
+
+ PRI2
+
+drivers/block/
+~~~~~~~~~~~~~~
+
+o More testing of floppy
+
+ PRI3
+
+drivers/char/
+~~~~~~~~~~~~~
+
+
+drivers/ide/
+~~~~~~~~~~~~
+
+ (Alan)
+
+o IDE PIO has occasional unexplained PIO disk eating reports
+
+ PRI1
+
+o IDE has multiple zillions of races/hangs in 2.5 still
+
+ PRI1
+
+o IDE scsi needs rewriting
+
+ PRI2
+
+o IDE needs significant reworking to handle Simplex right
+
+ PRI2
+
+o IDE hotplug handling for 2.5 is completely broken still
+
+ PRI2
+
+o There are lots of other IDE bugs that wont go away until the taskfile
+ stuff is included, the locking bugs that allow any user to hang the IDE
+ layer in 2.5, and some other updates are forward ported. (esp. HPT372N).
+
+ PRI1
+
+drivers/isdn/
+~~~~~~~~~~~~~
+
+ (Kai, rmk)
+
+o isdn_tty locking is completely broken (cli() and friends)
+
+ PRI2
+
+o fix other drivers
+
+ PRI2
+
+o lots more cleanups, adaption to recent APIs etc
+
+ PRI3
+
+o fixup tty-based ISDN drivers which provide TIOCM* ioctls (see my recent
+ 3-set patch for serial stuff)
+
+ Alternatively, we could re-introduce the fallback to driver ioctl parsing
+ for these if not enough drivers get updated.
+
+ PRI3
+
+drivers/net/
+~~~~~~~~~~~~
+
+o davej: Either Wireless network drivers or PCMCIA broke somewhen. A
+ configuration that worked fine under 2.4 doesn't receive any packets. Need
+ to look into this more to make sure I don't have any misconfiguration that
+ just 'happened to work' under 2.4
+
+ PRI1
+
+drivers/scsi/
+~~~~~~~~~~~~~
+
+o jejb: qlogic -
+
+ o Merge the feral driver. It covers all qlogic chips: 1020 all the way
+ up to 23xxx. http://linux-scsi.bkbits.net/scsi-isp-2.5
+
+ o qla2xxx: only for FC chips. Has significant build issues. hch
+ promises to send me a "must fix" list for this.
+ http://linux-scsi.bkbits.net/scsi-qla2xxx-2.5
+
+ PRI2
+
+o hch, Mike Anderson, Badari Pulavarty: scsi locking issues
+
+ o there are lots of members of struct Scsi_Host/scsi_device/scsi_cmnd
+ with very unclear locking, many of them probably want to become
+ atomic_t's or bitmaps (for the 1bit bitfields).
+
+ o there's lots of volatile abuse in the scsi code that needs to be
+ thought about.
+
+ o there's some global variables incremented without any locks
+
+ PRI2
+
+sound/
+~~~~~~
+
+o rmk: several OSS drivers for SA11xx-based hardware in need of
+ ALSA-ification and L3 bus support code for these.
+
+o rmk: need to complete ALSA-ification of the WaveArtist driver for both
+ NetWinder and other stuff (there's some fairly fundamental differences in
+ the way the mixer needs to be handled for the NetWinder.)
+
+ (Issues with forward-porting 2.4 bugfixes.)
+ (Killing off OSS is 2.7 material)
+
+PRI2
+
+arch/i386/
+~~~~~~~~~~
+
+o Also PC9800 merge needs finishing to the point we want for 2.6 (not all).
+
+ PRI3
+
+o davej: PAT support (for mtrr exhaustion w/ AGP)
+
+ PRI2
+
+o 2.5.x won't boot on some 440GX
+
+ alan: Problem understood now, feasible fix in 2.4/2.4-ac. (440GX has two
+ IRQ routers, we use the $PIR table with the PIIX, but the 440GX doesnt use
+ the PIIX for its IRQ routing). Fall back to BIOS for 440GX works and Intel
+ concurs.
+
+ PRI1
+
+o 2.5.x doesn't handle VIA APIC right yet.
+
+ 1. We must write the PCI_INTERRUPT_LINE
+
+ 2. We have quirk handlers that seem to trash it.
+
+ PRI1
+
+o ECC driver questions are not yet sorted (DaveJ is working on this) (Dan
+ Hollis)
+
+ alan: ECC - I have some test bits from Dan's stuff - they need no kernel
+ core changes for most platforms. That means we can treat it as a random
+ driver merge.
+
+ PRI3
+
+o alan: 2.4 has some fixes for tsc handling bugs. One where some bioses in
+ SMM mode mess up our toggle on the time high/low or mangle the counter and
+ one where a few chips need religious use of _p for timer access and we
+ don't do that. This is forward porting little bits of fixup.
+
+ ACPI HZ stuff we can't trap - a lot of ACPI is implemented as outb's
+ triggering SMM traps
+
+ PRI1
+
+arch/x86_64/
+~~~~~~~~~~~~
+
+ (Andi)
+
+o time handling is broken. Need to move up 2.4 time.c code.
+
+ PRI1
+
+o NMI watchdog seems to tick too fast
+
+ PRI2
+
+o need to coredump 64bit vsyscall code with dwarf2
+
+ PRI2
+
+o move 64bit signal trampolines into vsyscall code and add dwarf2 for it.
+ (in progress)
+
+ PRI1
+
+o describe kernel assembly with dwarf2 annotations for kgdb
+
+ PRI3
+
+arch/alpha/
+~~~~~~~~~~~
+
+o rth: Ptrace writes are broken. This means we can't (reliably) set
+ breakpoints or modify variables from gdb.
+
+ PRI1
+
+arch/arm/
+~~~~~~~~~
+
+o rmk: missing raw keyboard translation tables for all ARM machines.
+ Haven't even looked into this at all. This could be messy since there
+ isn't an ARM architecture standard. I'm presently hoping that it won't be
+ an issue. If it does, I guess we'll see drivers/char/keyboard.c explode.
+
+ PRI2
+
+arch/others/
+~~~~~~~~~~~~
+
+o SH needs resyncing, as do some other ports. SH64 needs merging.
+ No impact on mainstream platforms hopefully.
+
+ PRI2
+
+arch/s390/
+~~~~~~~~~
+
+o A nastly memory management problem causes random crashes. These appear
+ to be fixed/hidden by the objrmap patch, more investigation is needed.
+
+ PRI1
+
+drivers/s390/
+~~~~~~~~~~~~~
+
+o Early userspace and 64 bit dev_t will allow the removal of most of
+ dasd_devmap.c and dasd_genhd.c.
+
+ PRI2
+
+o The 3270 console driver needs to be replaced with a working one
+ (prototype is there, needs to be finished).
+
+ PRI2
+
+o Minor interface changes are pending in cio/ when the z990 machines are
+ out.
+
+ PRI2
+
+o Jan Glauber is working on a fix for the timer issues related to running
+ on virtualized CPUs (wall-clock vs. cpu time).
+
+ PRI1
+
+o a block device driver for ramdisks shared among virtual machines
+
+ PRI3
+
+o driver for crypto hardware
+
+ PRI3
+
+o 'claw' network device driver
+
+ PRI3
+
--- diff/Documentation/usb/mtouchusb.txt 1970-01-01 00:00:00.000000000 +0000
+++ source/Documentation/usb/mtouchusb.txt 2004-03-16 11:27:48.001890264 +0000
@@ -0,0 +1,85 @@
+CHANGES
+
+- Created based off of scanner & INSTALL from the original touchscreen
+ driver on freshmeat (http://freshmeat.net/projects/3mtouchscreendriver)
+- Amended for linux-2.4.18, then 2.4.19
+
+- Complete rewrite using Linux Input in 2.6.3
+ Unfortunately no calibration support at this time
+
+
+DRIVER NOTES:
+
+Installation is simple, you only need to add Linux Input, Linux USB, and the
+driver to the kernel. The driver can also be optionally built as a module.
+
+If you have another MicroTouch device that you wish to experiment with
+or try using this driver with, but the Vendor and Product ID's are not
+coded in, don't despair. If the driver was compiled as a module, you can
+pass options to the driver. Simply try:
+
+ /sbin/modprobe mtouchusb vendor=0x#### product=0x****
+
+If it works, send me the iVendor & iProduct (or a patch) and I will add...
+
+This driver appears to be one of possible 2 Linux USB Input Touchscreen
+drivers. Although 3M produces a binary only driver available for
+download, I persist in updating this driver since I would like to use the
+touchscreen for embedded apps using QTEmbedded, DirectFB, etc. So I feel the
+logical choice is to use Linux Imput.
+
+A little info about the MicroTouch USB controller (14-206):
+
+Y is inverted, and the device has a total possible resolution of 0 - 65535.
+
+Y is inverted by the driver by:
+
+ input.absmin[ABS_Y] = MTOUCHUSB_MAX_YC;
+ input.absmax[ABS_Y] = MTOUCHUSB_MIN_YC;
+
+absmin & absmax are also used to scale the data, sine it is rather high
+resolution.
+
+ ---------------touch screen area-----------------
+ I MicroTouch (xmax,ymax) @I
+ I X I
+ I ########visible monitor area############## I
+ I #@ (xmin,ymin) # I
+ I # # I
+ I # # I
+ I # # I
+ I # # I
+ I # # I
+ I Y # # I
+ I # # I
+ I # # I
+ I # # I
+ I # # I
+ I # # I
+ I # (xmax,ymax) @# I
+ I ########################################## I
+ I I
+ I@ MicroTouch (xmin,ymin) I
+ -------------------------------------------------
+
+Currently there is no way to calibrate the device via this driver. Perhaps
+at some point an abstract function will be placed into evdev so generic
+functions like calibrations, resets, and vendor information can be requested
+(And the drivers would handle the vendor specific tasks).
+
+ADDITIONAL INFORMATION/UPDATES:
+
+http://groomlakelabs.com/grandamp/code/microtouch/
+
+TODO:
+
+Implement a control urb again to handle requests to and from the device
+such as calibration, etc.
+
+DISCLAMER:
+
+I am not a MicroTouch/3M employee, nor have I ever been. 3M does not support
+this driver! If you want touch drivers only supported within X, please go to:
+
+http://www.3m.com/3MTouchSystems/downloads/
+
--- diff/arch/i386/kernel/entry_trampoline.c 1970-01-01 00:00:00.000000000 +0000
+++ source/arch/i386/kernel/entry_trampoline.c 2004-03-16 11:27:47.870910176 +0000
@@ -0,0 +1,75 @@
+/*
+ * linux/arch/i386/kernel/entry_trampoline.c
+ *
+ * (C) Copyright 2003 Ingo Molnar
+ *
+ * This file contains the needed support code for 4GB userspace
+ */
+
+#include <linux/init.h>
+#include <linux/smp.h>
+#include <linux/mm.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/highmem.h>
+#include <asm/desc.h>
+#include <asm/atomic_kmap.h>
+
+extern char __entry_tramp_start, __entry_tramp_end, __start___entry_text;
+
+void __init init_entry_mappings(void)
+{
+#ifdef CONFIG_X86_HIGH_ENTRY
+
+ void *tramp;
+ int p;
+
+ /*
+ * We need a high IDT and GDT for the 4G/4G split:
+ */
+ trap_init_virtual_IDT();
+
+ __set_fixmap(FIX_ENTRY_TRAMPOLINE_0, __pa((unsigned long)&__entry_tramp_start), PAGE_KERNEL);
+ __set_fixmap(FIX_ENTRY_TRAMPOLINE_1, __pa((unsigned long)&__entry_tramp_start) + PAGE_SIZE, PAGE_KERNEL);
+ tramp = (void *)fix_to_virt(FIX_ENTRY_TRAMPOLINE_0);
+
+ printk("mapped 4G/4G trampoline to %p.\n", tramp);
+ BUG_ON((void *)&__start___entry_text != tramp);
+ /*
+ * Virtual kernel stack:
+ */
+ BUG_ON(__kmap_atomic_vaddr(KM_VSTACK_TOP) & (THREAD_SIZE-1));
+ BUG_ON(sizeof(struct desc_struct)*NR_CPUS*GDT_ENTRIES > 2*PAGE_SIZE);
+ BUG_ON((unsigned int)&__entry_tramp_end - (unsigned int)&__entry_tramp_start > 2*PAGE_SIZE);
+
+ /*
+ * set up the initial thread's virtual stack related
+ * fields:
+ */
+ for (p = 0; p < ARRAY_SIZE(current->thread.stack_page); p++)
+ current->thread.stack_page[p] = virt_to_page((char *)current->thread_info + (p*PAGE_SIZE));
+
+ current->thread_info->virtual_stack = (void *)__kmap_atomic_vaddr(KM_VSTACK_TOP);
+
+ for (p = 0; p < ARRAY_SIZE(current->thread.stack_page); p++) {
+ __kunmap_atomic_type(KM_VSTACK_TOP-p);
+ __kmap_atomic(current->thread.stack_page[p], KM_VSTACK_TOP-p);
+ }
+#endif
+ current->thread_info->real_stack = (void *)current->thread_info;
+ current->thread_info->user_pgd = NULL;
+ current->thread.esp0 = (unsigned long)current->thread_info->real_stack + THREAD_SIZE;
+}
+
+
+
+void __init entry_trampoline_setup(void)
+{
+ /*
+ * old IRQ entries set up by the boot code will still hang
+ * around - they are a sign of hw trouble anyway, now they'll
+ * produce a double fault message.
+ */
+ trap_init_virtual_GDT();
+}
--- diff/arch/i386/kernel/kgdb_stub.c 1970-01-01 00:00:00.000000000 +0000
+++ source/arch/i386/kernel/kgdb_stub.c 2004-03-16 11:27:47.881908504 +0000
@@ -0,0 +1,2458 @@
+/*
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2, or (at your option) any
+ * later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ */
+
+/*
+ * Copyright (c) 2000 VERITAS Software Corporation.
+ *
+ */
+/****************************************************************************
+ * Header: remcom.c,v 1.34 91/03/09 12:29:49 glenne Exp $
+ *
+ * Module name: remcom.c $
+ * Revision: 1.34 $
+ * Date: 91/03/09 12:29:49 $
+ * Contributor: Lake Stevens Instrument Division$
+ *
+ * Description: low level support for gdb debugger. $
+ *
+ * Considerations: only works on target hardware $
+ *
+ * Written by: Glenn Engel $
+ * Updated by: David Grothe <dave@gcom.com>
+ * Updated by: Robert Walsh <rjwalsh@durables.org>
+ * Updated by: wangdi <wangdi@clusterfs.com>
+ * ModuleState: Experimental $
+ *
+ * NOTES: See Below $
+ *
+ * Modified for 386 by Jim Kingdon, Cygnus Support.
+ * Compatibility with 2.1.xx kernel by David Grothe <dave@gcom.com>
+ *
+ * Changes to allow auto initilization. All that is needed is that it
+ * be linked with the kernel and a break point (int 3) be executed.
+ * The header file <asm/kgdb.h> defines BREAKPOINT to allow one to do
+ * this. It should also be possible, once the interrupt system is up, to
+ * call putDebugChar("+"). Once this is done, the remote debugger should
+ * get our attention by sending a ^C in a packet. George Anzinger
+ * <george@mvista.com>
+ * Integrated into 2.2.5 kernel by Tigran Aivazian <tigran@sco.com>
+ * Added thread support, support for multiple processors,
+ * support for ia-32(x86) hardware debugging.
+ * Amit S. Kale ( akale@veritas.com )
+ *
+ * Modified to support debugging over ethernet by Robert Walsh
+ * <rjwalsh@durables.org> and wangdi <wangdi@clusterfs.com>, based on
+ * code by San Mehat.
+ *
+ *
+ * To enable debugger support, two things need to happen. One, a
+ * call to set_debug_traps() is necessary in order to allow any breakpoints
+ * or error conditions to be properly intercepted and reported to gdb.
+ * Two, a breakpoint needs to be generated to begin communication. This
+ * is most easily accomplished by a call to breakpoint(). Breakpoint()
+ * simulates a breakpoint by executing an int 3.
+ *
+ *************
+ *
+ * The following gdb commands are supported:
+ *
+ * command function Return value
+ *
+ * g return the value of the CPU registers hex data or ENN
+ * G set the value of the CPU registers OK or ENN
+ *
+ * mAA..AA,LLLL Read LLLL bytes at address AA..AA hex data or ENN
+ * MAA..AA,LLLL: Write LLLL bytes at address AA.AA OK or ENN
+ *
+ * c Resume at current address SNN ( signal NN)
+ * cAA..AA Continue at address AA..AA SNN
+ *
+ * s Step one instruction SNN
+ * sAA..AA Step one instruction from AA..AA SNN
+ *
+ * k kill
+ *
+ * ? What was the last sigval ? SNN (signal NN)
+ *
+ * All commands and responses are sent with a packet which includes a
+ * checksum. A packet consists of
+ *
+ * $<packet info>#<checksum>.
+ *
+ * where
+ * <packet info> :: <characters representing the command or response>
+ * <checksum> :: < two hex digits computed as modulo 256 sum of <packetinfo>>
+ *
+ * When a packet is received, it is first acknowledged with either '+' or '-'.
+ * '+' indicates a successful transfer. '-' indicates a failed transfer.
+ *
+ * Example:
+ *
+ * Host: Reply:
+ * $m0,10#2a +$00010203040506070809101112131415#42
+ *
+ ****************************************************************************/
+#define KGDB_VERSION "<20030915.1651.33>"
+#include <linux/config.h>
+#include <linux/types.h>
+#include <asm/string.h> /* for strcpy */
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <asm/vm86.h>
+#include <asm/system.h>
+#include <asm/ptrace.h> /* for linux pt_regs struct */
+#include <asm/kgdb_local.h>
+#include <linux/list.h>
+#include <asm/atomic.h>
+#include <asm/processor.h>
+#include <linux/irq.h>
+#include <asm/desc.h>
+#include <linux/inet.h>
+#include <linux/netpoll.h>
+
+/************************************************************************
+ *
+ * external low-level support routines
+ */
+typedef void (*Function) (void); /* pointer to a function */
+
+/* Thread reference */
+typedef unsigned char threadref[8];
+
+extern int tty_putDebugChar(int); /* write a single character */
+extern int tty_getDebugChar(void); /* read and return a single char */
+extern void tty_flushDebugChar(void); /* flush pending characters */
+extern int eth_putDebugChar(int); /* write a single character */
+extern int eth_getDebugChar(void); /* read and return a single char */
+extern void eth_flushDebugChar(void); /* flush pending characters */
+
+/************************************************************************/
+/* BUFMAX defines the maximum number of characters in inbound/outbound buffers*/
+/* at least NUMREGBYTES*2 are needed for register packets */
+/* Longer buffer is needed to list all threads */
+#define BUFMAX 400
+
+char *kgdb_version = KGDB_VERSION;
+
+/* debug > 0 prints ill-formed commands in valid packets & checksum errors */
+int debug_regs = 0; /* set to non-zero to print registers */
+
+/* filled in by an external module */
+char *gdb_module_offsets;
+
+static const char hexchars[] = "0123456789abcdef";
+
+/* Number of bytes of registers. */
+#define NUMREGBYTES 64
+/*
+ * Note that this register image is in a different order than
+ * the register image that Linux produces at interrupt time.
+ *
+ * Linux's register image is defined by struct pt_regs in ptrace.h.
+ * Just why GDB uses a different order is a historical mystery.
+ */
+enum regnames { _EAX, /* 0 */
+ _ECX, /* 1 */
+ _EDX, /* 2 */
+ _EBX, /* 3 */
+ _ESP, /* 4 */
+ _EBP, /* 5 */
+ _ESI, /* 6 */
+ _EDI, /* 7 */
+ _PC /* 8 also known as eip */ ,
+ _PS /* 9 also known as eflags */ ,
+ _CS, /* 10 */
+ _SS, /* 11 */
+ _DS, /* 12 */
+ _ES, /* 13 */
+ _FS, /* 14 */
+ _GS /* 15 */
+};
+
+/*************************** ASSEMBLY CODE MACROS *************************/
+/*
+ * Put the error code here just in case the user cares.
+ * Likewise, the vector number here (since GDB only gets the signal
+ * number through the usual means, and that's not very specific).
+ * The called_from is the return address so he can tell how we entered kgdb.
+ * This will allow him to seperate out the various possible entries.
+ */
+#define REMOTE_DEBUG 0 /* set != to turn on printing (also available in info) */
+
+#define PID_MAX PID_MAX_DEFAULT
+
+#ifdef CONFIG_SMP
+void smp_send_nmi_allbutself(void);
+#define IF_SMP(x) x
+#undef MAX_NO_CPUS
+#ifndef CONFIG_NO_KGDB_CPUS
+#define CONFIG_NO_KGDB_CPUS 2
+#endif
+#if CONFIG_NO_KGDB_CPUS > NR_CPUS
+#define MAX_NO_CPUS NR_CPUS
+#else
+#define MAX_NO_CPUS CONFIG_NO_KGDB_CPUS
+#endif
+#define hold_init hold_on_sstep: 1,
+#define MAX_CPU_MASK (unsigned long)((1LL << MAX_NO_CPUS) - 1LL)
+#define NUM_CPUS num_online_cpus()
+#else
+#define IF_SMP(x)
+#define hold_init
+#undef MAX_NO_CPUS
+#define MAX_NO_CPUS 1
+#define NUM_CPUS 1
+#endif
+#define NOCPU (struct task_struct *)0xbad1fbad
+/* *INDENT-OFF* */
+struct kgdb_info {
+ int used_malloc;
+ void *called_from;
+ long long entry_tsc;
+ int errcode;
+ int vector;
+ int print_debug_info;
+#ifdef CONFIG_SMP
+ int hold_on_sstep;
+ struct {
+ volatile struct task_struct *task;
+ int pid;
+ int hold;
+ struct pt_regs *regs;
+ } cpus_waiting[MAX_NO_CPUS];
+#endif
+} kgdb_info = {hold_init print_debug_info:REMOTE_DEBUG, vector:-1};
+
+/* *INDENT-ON* */
+
+#define used_m kgdb_info.used_malloc
+/*
+ * This is little area we set aside to contain the stack we
+ * need to build to allow gdb to call functions. We use one
+ * per cpu to avoid locking issues. We will do all this work
+ * with interrupts off so that should take care of the protection
+ * issues.
+ */
+#define LOOKASIDE_SIZE 200 /* should be more than enough */
+#define MALLOC_MAX 200 /* Max malloc size */
+struct {
+ unsigned int esp;
+ int array[LOOKASIDE_SIZE];
+} fn_call_lookaside[MAX_NO_CPUS];
+
+static int trap_cpu;
+static unsigned int OLD_esp;
+
+#define END_OF_LOOKASIDE &fn_call_lookaside[trap_cpu].array[LOOKASIDE_SIZE]
+#define IF_BIT 0x200
+#define TF_BIT 0x100
+
+#define MALLOC_ROUND 8-1
+
+static char malloc_array[MALLOC_MAX];
+IF_SMP(static void to_gdb(const char *mess));
+void *
+malloc(int size)
+{
+
+ if (size <= (MALLOC_MAX - used_m)) {
+ int old_used = used_m;
+ used_m += ((size + MALLOC_ROUND) & (~MALLOC_ROUND));
+ return &malloc_array[old_used];
+ } else {
+ return NULL;
+ }
+}
+
+/*
+ * I/O dispatch functions...
+ * Based upon kgdboe, either call the ethernet
+ * handler or the serial one..
+ */
+void
+putDebugChar(int c)
+{
+ if (!kgdboe) {
+ tty_putDebugChar(c);
+ } else {
+ eth_putDebugChar(c);
+ }
+}
+
+int
+getDebugChar(void)
+{
+ if (!kgdboe) {
+ return tty_getDebugChar();
+ } else {
+ return eth_getDebugChar();
+ }
+}
+
+void
+flushDebugChar(void)
+{
+ if (!kgdboe) {
+ tty_flushDebugChar();
+ } else {
+ eth_flushDebugChar();
+ }
+}
+
+/*
+ * Gdb calls functions by pushing agruments, including a return address
+ * on the stack and the adjusting EIP to point to the function. The
+ * whole assumption in GDB is that we are on a different stack than the
+ * one the "user" i.e. code that hit the break point, is on. This, of
+ * course is not true in the kernel. Thus various dodges are needed to
+ * do the call without directly messing with EIP (which we can not change
+ * as it is just a location and not a register. To adjust it would then
+ * require that we move every thing below EIP up or down as needed. This
+ * will not work as we may well have stack relative pointer on the stack
+ * (such as the pointer to regs, for example).
+
+ * So here is what we do:
+ * We detect gdb attempting to store into the stack area and instead, store
+ * into the fn_call_lookaside.array at the same relative location as if it
+ * were the area ESP pointed at. We also trap ESP modifications
+ * and uses these to adjust fn_call_lookaside.esp. On entry
+ * fn_call_lookaside.esp will be set to point at the last entry in
+ * fn_call_lookaside.array. This allows us to check if it has changed, and
+ * if so, on exit, we add the registers we will use to do the move and a
+ * trap/ interrupt return exit sequence. We then adjust the eflags in the
+ * regs array (remember we now have a copy in the fn_call_lookaside.array) to
+ * kill the interrupt bit, AND we change EIP to point at our set up stub.
+ * As part of the register set up we preset the registers to point at the
+ * begining and end of the fn_call_lookaside.array, so all the stub needs to
+ * do is move words from the array to the stack until ESP= the desired value
+ * then do the rti. This will then transfer to the desired function with
+ * all the correct registers. Nifty huh?
+ */
+extern asmlinkage void fn_call_stub(void);
+extern asmlinkage void fn_rtn_stub(void);
+/* *INDENT-OFF* */
+__asm__("fn_rtn_stub:\n\t"
+ "movl %eax,%esp\n\t"
+ "fn_call_stub:\n\t"
+ "1:\n\t"
+ "addl $-4,%ebx\n\t"
+ "movl (%ebx), %eax\n\t"
+ "pushl %eax\n\t"
+ "cmpl %esp,%ecx\n\t"
+ "jne 1b\n\t"
+ "popl %eax\n\t"
+ "popl %ebx\n\t"
+ "popl %ecx\n\t"
+ "iret \n\t");
+/* *INDENT-ON* */
+#define gdb_i386vector kgdb_info.vector
+#define gdb_i386errcode kgdb_info.errcode
+#define waiting_cpus kgdb_info.cpus_waiting
+#define remote_debug kgdb_info.print_debug_info
+#define hold_cpu(cpu) kgdb_info.cpus_waiting[cpu].hold
+/* gdb locks */
+
+#ifdef CONFIG_SMP
+static int in_kgdb_called;
+static spinlock_t waitlocks[MAX_NO_CPUS] =
+ {[0 ... MAX_NO_CPUS - 1] = SPIN_LOCK_UNLOCKED };
+/*
+ * The following array has the thread pointer of each of the "other"
+ * cpus. We make it global so it can be seen by gdb.
+ */
+volatile int in_kgdb_entry_log[MAX_NO_CPUS];
+volatile struct pt_regs *in_kgdb_here_log[MAX_NO_CPUS];
+/*
+static spinlock_t continuelocks[MAX_NO_CPUS];
+*/
+spinlock_t kgdb_spinlock = SPIN_LOCK_UNLOCKED;
+/* waiters on our spinlock plus us */
+static atomic_t spinlock_waiters = ATOMIC_INIT(1);
+static int spinlock_count = 0;
+static int spinlock_cpu = 0;
+/*
+ * Note we use nested spin locks to account for the case where a break
+ * point is encountered when calling a function by user direction from
+ * kgdb. Also there is the memory exception recursion to account for.
+ * Well, yes, but this lets other cpus thru too. Lets add a
+ * cpu id to the lock.
+ */
+#define KGDB_SPIN_LOCK(x) if( spinlock_count == 0 || \
+ spinlock_cpu != smp_processor_id()){\
+ atomic_inc(&spinlock_waiters); \
+ while (! spin_trylock(x)) {\
+ in_kgdb(®s);\
+ }\
+ atomic_dec(&spinlock_waiters); \
+ spinlock_count = 1; \
+ spinlock_cpu = smp_processor_id(); \
+ }else{ \
+ spinlock_count++; \
+ }
+#define KGDB_SPIN_UNLOCK(x) if( --spinlock_count == 0) spin_unlock(x)
+#else
+unsigned kgdb_spinlock = 0;
+#define KGDB_SPIN_LOCK(x) --*x
+#define KGDB_SPIN_UNLOCK(x) ++*x
+#endif
+
+int
+hex(char ch)
+{
+ if ((ch >= 'a') && (ch <= 'f'))
+ return (ch - 'a' + 10);
+ if ((ch >= '0') && (ch <= '9'))
+ return (ch - '0');
+ if ((ch >= 'A') && (ch <= 'F'))
+ return (ch - 'A' + 10);
+ return (-1);
+}
+
+/* scan for the sequence $<data>#<checksum> */
+void
+getpacket(char *buffer)
+{
+ unsigned char checksum;
+ unsigned char xmitcsum;
+ int i;
+ int count;
+ char ch;
+
+ do {
+ /* wait around for the start character, ignore all other characters */
+ while ((ch = (getDebugChar() & 0x7f)) != '$') ;
+ checksum = 0;
+ xmitcsum = -1;
+
+ count = 0;
+
+ /* now, read until a # or end of buffer is found */
+ while (count < BUFMAX) {
+ ch = getDebugChar() & 0x7f;
+ if (ch == '#')
+ break;
+ checksum = checksum + ch;
+ buffer[count] = ch;
+ count = count + 1;
+ }
+ buffer[count] = 0;
+
+ if (ch == '#') {
+ xmitcsum = hex(getDebugChar() & 0x7f) << 4;
+ xmitcsum += hex(getDebugChar() & 0x7f);
+ if ((remote_debug) && (checksum != xmitcsum)) {
+ printk
+ ("bad checksum. My count = 0x%x, sent=0x%x. buf=%s\n",
+ checksum, xmitcsum, buffer);
+ }
+
+ if (checksum != xmitcsum)
+ putDebugChar('-'); /* failed checksum */
+ else {
+ putDebugChar('+'); /* successful transfer */
+ /* if a sequence char is present, reply the sequence ID */
+ if (buffer[2] == ':') {
+ putDebugChar(buffer[0]);
+ putDebugChar(buffer[1]);
+ /* remove sequence chars from buffer */
+ count = strlen(buffer);
+ for (i = 3; i <= count; i++)
+ buffer[i - 3] = buffer[i];
+ }
+ }
+ }
+ } while (checksum != xmitcsum);
+
+ if (remote_debug)
+ printk("R:%s\n", buffer);
+ flushDebugChar();
+}
+
+/* send the packet in buffer. */
+
+void
+putpacket(char *buffer)
+{
+ unsigned char checksum;
+ int count;
+ char ch;
+
+ /* $<packet info>#<checksum>. */
+
+ if (!kgdboe) {
+ do {
+ if (remote_debug)
+ printk("T:%s\n", buffer);
+ putDebugChar('$');
+ checksum = 0;
+ count = 0;
+
+ while ((ch = buffer[count])) {
+ putDebugChar(ch);
+ checksum += ch;
+ count += 1;
+ }
+
+ putDebugChar('#');
+ putDebugChar(hexchars[checksum >> 4]);
+ putDebugChar(hexchars[checksum % 16]);
+ flushDebugChar();
+
+ } while ((getDebugChar() & 0x7f) != '+');
+ } else {
+ /*
+ * For udp, we can not transfer too much bytes once.
+ * We only transfer MAX_SEND_COUNT size bytes each time
+ */
+
+#define MAX_SEND_COUNT 30
+
+ int send_count = 0, i = 0;
+ char send_buf[MAX_SEND_COUNT];
+
+ do {
+ if (remote_debug)
+ printk("T:%s\n", buffer);
+ putDebugChar('$');
+ checksum = 0;
+ count = 0;
+ send_count = 0;
+ while ((ch = buffer[count])) {
+ if (send_count >= MAX_SEND_COUNT) {
+ for(i = 0; i < MAX_SEND_COUNT; i++) {
+ putDebugChar(send_buf[i]);
+ }
+ flushDebugChar();
+ send_count = 0;
+ } else {
+ send_buf[send_count] = ch;
+ checksum += ch;
+ count ++;
+ send_count++;
+ }
+ }
+ for(i = 0; i < send_count; i++)
+ putDebugChar(send_buf[i]);
+ putDebugChar('#');
+ putDebugChar(hexchars[checksum >> 4]);
+ putDebugChar(hexchars[checksum % 16]);
+ flushDebugChar();
+ } while ((getDebugChar() & 0x7f) != '+');
+ }
+}
+
+static char remcomInBuffer[BUFMAX];
+static char remcomOutBuffer[BUFMAX];
+static short error;
+
+void
+debug_error(char *format, char *parm)
+{
+ if (remote_debug)
+ printk(format, parm);
+}
+
+static void
+print_regs(struct pt_regs *regs)
+{
+ printk("EAX=%08lx ", regs->eax);
+ printk("EBX=%08lx ", regs->ebx);
+ printk("ECX=%08lx ", regs->ecx);
+ printk("EDX=%08lx ", regs->edx);
+ printk("\n");
+ printk("ESI=%08lx ", regs->esi);
+ printk("EDI=%08lx ", regs->edi);
+ printk("EBP=%08lx ", regs->ebp);
+ printk("ESP=%08lx ", (long) ®s->esp);
+ printk("\n");
+ printk(" DS=%08x ", regs->xds);
+ printk(" ES=%08x ", regs->xes);
+ printk(" SS=%08x ", __KERNEL_DS);
+ printk(" FL=%08lx ", regs->eflags);
+ printk("\n");
+ printk(" CS=%08x ", regs->xcs);
+ printk(" IP=%08lx ", regs->eip);
+#if 0
+ printk(" FS=%08x ", regs->fs);
+ printk(" GS=%08x ", regs->gs);
+#endif
+ printk("\n");
+
+} /* print_regs */
+
+#define NEW_esp fn_call_lookaside[trap_cpu].esp
+
+static void
+regs_to_gdb_regs(int *gdb_regs, struct pt_regs *regs)
+{
+ gdb_regs[_EAX] = regs->eax;
+ gdb_regs[_EBX] = regs->ebx;
+ gdb_regs[_ECX] = regs->ecx;
+ gdb_regs[_EDX] = regs->edx;
+ gdb_regs[_ESI] = regs->esi;
+ gdb_regs[_EDI] = regs->edi;
+ gdb_regs[_EBP] = regs->ebp;
+ gdb_regs[_DS] = regs->xds;
+ gdb_regs[_ES] = regs->xes;
+ gdb_regs[_PS] = regs->eflags;
+ gdb_regs[_CS] = regs->xcs;
+ gdb_regs[_PC] = regs->eip;
+ /* Note, as we are a debugging the kernel, we will always
+ * trap in kernel code, this means no priviledge change,
+ * and so the pt_regs structure is not completely valid. In a non
+ * privilege change trap, only EFLAGS, CS and EIP are put on the stack,
+ * SS and ESP are not stacked, this means that the last 2 elements of
+ * pt_regs is not valid (they would normally refer to the user stack)
+ * also, using regs+1 is no good because you end up will a value that is
+ * 2 longs (8) too high. This used to cause stepping over functions
+ * to fail, so my fix is to use the address of regs->esp, which
+ * should point at the end of the stack frame. Note I have ignored
+ * completely exceptions that cause an error code to be stacked, such
+ * as double fault. Stuart Hughes, Zentropix.
+ * original code: gdb_regs[_ESP] = (int) (regs + 1) ;
+
+ * this is now done on entry and moved to OLD_esp (as well as NEW_esp).
+ */
+ gdb_regs[_ESP] = NEW_esp;
+ gdb_regs[_SS] = __KERNEL_DS;
+ gdb_regs[_FS] = 0xFFFF;
+ gdb_regs[_GS] = 0xFFFF;
+} /* regs_to_gdb_regs */
+
+static void
+gdb_regs_to_regs(int *gdb_regs, struct pt_regs *regs)
+{
+ regs->eax = gdb_regs[_EAX];
+ regs->ebx = gdb_regs[_EBX];
+ regs->ecx = gdb_regs[_ECX];
+ regs->edx = gdb_regs[_EDX];
+ regs->esi = gdb_regs[_ESI];
+ regs->edi = gdb_regs[_EDI];
+ regs->ebp = gdb_regs[_EBP];
+ regs->xds = gdb_regs[_DS];
+ regs->xes = gdb_regs[_ES];
+ regs->eflags = gdb_regs[_PS];
+ regs->xcs = gdb_regs[_CS];
+ regs->eip = gdb_regs[_PC];
+ NEW_esp = gdb_regs[_ESP]; /* keep the value */
+#if 0 /* can't change these */
+ regs->esp = gdb_regs[_ESP];
+ regs->xss = gdb_regs[_SS];
+ regs->fs = gdb_regs[_FS];
+ regs->gs = gdb_regs[_GS];
+#endif
+
+} /* gdb_regs_to_regs */
+extern void scheduling_functions_start_here(void);
+extern void scheduling_functions_end_here(void);
+#define first_sched ((unsigned long) scheduling_functions_start_here)
+#define last_sched ((unsigned long) scheduling_functions_end_here)
+
+int thread_list = 0;
+
+void
+get_gdb_regs(struct task_struct *p, struct pt_regs *regs, int *gdb_regs)
+{
+ unsigned long stack_page;
+ int count = 0;
+ IF_SMP(int i);
+ if (!p || p == current) {
+ regs_to_gdb_regs(gdb_regs, regs);
+ return;
+ }
+#ifdef CONFIG_SMP
+ for (i = 0; i < MAX_NO_CPUS; i++) {
+ if (p == kgdb_info.cpus_waiting[i].task) {
+ regs_to_gdb_regs(gdb_regs,
+ kgdb_info.cpus_waiting[i].regs);
+ gdb_regs[_ESP] =
+ (int) &kgdb_info.cpus_waiting[i].regs->esp;
+
+ return;
+ }
+ }
+#endif
+ memset(gdb_regs, 0, NUMREGBYTES);
+ gdb_regs[_ESP] = p->thread.esp;
+ gdb_regs[_PC] = p->thread.eip;
+ gdb_regs[_EBP] = *(int *) gdb_regs[_ESP];
+ gdb_regs[_EDI] = *(int *) (gdb_regs[_ESP] + 4);
+ gdb_regs[_ESI] = *(int *) (gdb_regs[_ESP] + 8);
+
+/*
+ * This code is to give a more informative notion of where a process
+ * is waiting. It is used only when the user asks for a thread info
+ * list. If he then switches to the thread, s/he will find the task
+ * is in schedule, but a back trace should show the same info we come
+ * up with. This code was shamelessly purloined from process.c. It was
+ * then enhanced to provide more registers than simply the program
+ * counter.
+ */
+
+ if (!thread_list) {
+ return;
+ }
+
+ if (p->state == TASK_RUNNING)
+ return;
+ stack_page = (unsigned long) p->thread_info;
+ if (gdb_regs[_ESP] < stack_page || gdb_regs[_ESP] >
+ THREAD_SIZE - sizeof(long) + stack_page)
+ return;
+ /* include/asm-i386/system.h:switch_to() pushes ebp last. */
+ do {
+ if (gdb_regs[_EBP] < stack_page ||
+ gdb_regs[_EBP] > THREAD_SIZE - 2*sizeof(long) + stack_page)
+ return;
+ gdb_regs[_PC] = *(unsigned long *) (gdb_regs[_EBP] + 4);
+ gdb_regs[_ESP] = gdb_regs[_EBP] + 8;
+ gdb_regs[_EBP] = *(unsigned long *) gdb_regs[_EBP];
+ if (gdb_regs[_PC] < first_sched || gdb_regs[_PC] >= last_sched)
+ return;
+ } while (count++ < 16);
+ return;
+}
+
+/* Indicate to caller of mem2hex or hex2mem that there has been an
+ error. */
+static volatile int mem_err = 0;
+static volatile int mem_err_expected = 0;
+static volatile int mem_err_cnt = 0;
+static int garbage_loc = -1;
+
+int
+get_char(char *addr)
+{
+ return *addr;
+}
+
+void
+set_char(char *addr, int val, int may_fault)
+{
+ /*
+ * This code traps references to the area mapped to the kernel
+ * stack as given by the regs and, instead, stores to the
+ * fn_call_lookaside[cpu].array
+ */
+ if (may_fault &&
+ (unsigned int) addr < OLD_esp &&
+ ((unsigned int) addr > (OLD_esp - (unsigned int) LOOKASIDE_SIZE))) {
+ addr = (char *) END_OF_LOOKASIDE - ((char *) OLD_esp - addr);
+ }
+ *addr = val;
+}
+
+/* convert the memory pointed to by mem into hex, placing result in buf */
+/* return a pointer to the last char put in buf (null) */
+/* If MAY_FAULT is non-zero, then we should set mem_err in response to
+ a fault; if zero treat a fault like any other fault in the stub. */
+char *
+mem2hex(char *mem, char *buf, int count, int may_fault)
+{
+ int i;
+ unsigned char ch;
+
+ if (may_fault) {
+ mem_err_expected = 1;
+ mem_err = 0;
+ }
+ for (i = 0; i < count; i++) {
+ /* printk("%lx = ", mem) ; */
+
+ ch = get_char(mem++);
+
+ /* printk("%02x\n", ch & 0xFF) ; */
+ if (may_fault && mem_err) {
+ if (remote_debug)
+ printk("Mem fault fetching from addr %lx\n",
+ (long) (mem - 1));
+ *buf = 0; /* truncate buffer */
+ return (buf);
+ }
+ *buf++ = hexchars[ch >> 4];
+ *buf++ = hexchars[ch % 16];
+ }
+ *buf = 0;
+ if (may_fault)
+ mem_err_expected = 0;
+ return (buf);
+}
+
+/* convert the hex array pointed to by buf into binary to be placed in mem */
+/* return a pointer to the character AFTER the last byte written */
+/* NOTE: We use the may fault flag to also indicate if the write is to
+ * the registers (0) or "other" memory (!=0)
+ */
+char *
+hex2mem(char *buf, char *mem, int count, int may_fault)
+{
+ int i;
+ unsigned char ch;
+
+ if (may_fault) {
+ mem_err_expected = 1;
+ mem_err = 0;
+ }
+ for (i = 0; i < count; i++) {
+ ch = hex(*buf++) << 4;
+ ch = ch + hex(*buf++);
+ set_char(mem++, ch, may_fault);
+
+ if (may_fault && mem_err) {
+ if (remote_debug)
+ printk("Mem fault storing to addr %lx\n",
+ (long) (mem - 1));
+ return (mem);
+ }
+ }
+ if (may_fault)
+ mem_err_expected = 0;
+ return (mem);
+}
+
+/**********************************************/
+/* WHILE WE FIND NICE HEX CHARS, BUILD AN INT */
+/* RETURN NUMBER OF CHARS PROCESSED */
+/**********************************************/
+int
+hexToInt(char **ptr, int *intValue)
+{
+ int numChars = 0;
+ int hexValue;
+
+ *intValue = 0;
+
+ while (**ptr) {
+ hexValue = hex(**ptr);
+ if (hexValue >= 0) {
+ *intValue = (*intValue << 4) | hexValue;
+ numChars++;
+ } else
+ break;
+
+ (*ptr)++;
+ }
+
+ return (numChars);
+}
+
+#define stubhex(h) hex(h)
+#ifdef old_thread_list
+
+static int
+stub_unpack_int(char *buff, int fieldlength)
+{
+ int nibble;
+ int retval = 0;
+
+ while (fieldlength) {
+ nibble = stubhex(*buff++);
+ retval |= nibble;
+ fieldlength--;
+ if (fieldlength)
+ retval = retval << 4;
+ }
+ return retval;
+}
+#endif
+static char *
+pack_hex_byte(char *pkt, int byte)
+{
+ *pkt++ = hexchars[(byte >> 4) & 0xf];
+ *pkt++ = hexchars[(byte & 0xf)];
+ return pkt;
+}
+
+#define BUF_THREAD_ID_SIZE 16
+
+static char *
+pack_threadid(char *pkt, threadref * id)
+{
+ char *limit;
+ unsigned char *altid;
+
+ altid = (unsigned char *) id;
+ limit = pkt + BUF_THREAD_ID_SIZE;
+ while (pkt < limit)
+ pkt = pack_hex_byte(pkt, *altid++);
+ return pkt;
+}
+
+#ifdef old_thread_list
+static char *
+unpack_byte(char *buf, int *value)
+{
+ *value = stub_unpack_int(buf, 2);
+ return buf + 2;
+}
+
+static char *
+unpack_threadid(char *inbuf, threadref * id)
+{
+ char *altref;
+ char *limit = inbuf + BUF_THREAD_ID_SIZE;
+ int x, y;
+
+ altref = (char *) id;
+
+ while (inbuf < limit) {
+ x = stubhex(*inbuf++);
+ y = stubhex(*inbuf++);
+ *altref++ = (x << 4) | y;
+ }
+ return inbuf;
+}
+#endif
+void
+int_to_threadref(threadref * id, int value)
+{
+ unsigned char *scan;
+
+ scan = (unsigned char *) id;
+ {
+ int i = 4;
+ while (i--)
+ *scan++ = 0;
+ }
+ *scan++ = (value >> 24) & 0xff;
+ *scan++ = (value >> 16) & 0xff;
+ *scan++ = (value >> 8) & 0xff;
+ *scan++ = (value & 0xff);
+}
+int
+int_to_hex_v(unsigned char * id, int value)
+{
+ unsigned char *start = id;
+ int shift;
+ int ch;
+
+ for (shift = 28; shift >= 0; shift -= 4) {
+ if ((ch = (value >> shift) & 0xf) || (id != start)) {
+ *id = hexchars[ch];
+ id++;
+ }
+ }
+ if (id == start)
+ *id++ = '0';
+ return id - start;
+}
+#ifdef old_thread_list
+
+static int
+threadref_to_int(threadref * ref)
+{
+ int i, value = 0;
+ unsigned char *scan;
+
+ scan = (char *) ref;
+ scan += 4;
+ i = 4;
+ while (i-- > 0)
+ value = (value << 8) | ((*scan++) & 0xff);
+ return value;
+}
+#endif
+static int
+cmp_str(char *s1, char *s2, int count)
+{
+ while (count--) {
+ if (*s1++ != *s2++)
+ return 0;
+ }
+ return 1;
+}
+
+#if 1 /* this is a hold over from 2.4 where O(1) was "sometimes" */
+extern struct task_struct *kgdb_get_idle(int cpu);
+#define idle_task(cpu) kgdb_get_idle(cpu)
+#else
+#define idle_task(cpu) init_tasks[cpu]
+#endif
+
+extern int kgdb_pid_init_done;
+
+struct task_struct *
+getthread(int pid)
+{
+ struct task_struct *thread;
+ if (pid >= PID_MAX && pid <= (PID_MAX + MAX_NO_CPUS)) {
+
+ return idle_task(pid - PID_MAX);
+ } else {
+ /*
+ * find_task_by_pid is relatively safe all the time
+ * Other pid functions require lock downs which imply
+ * that we may be interrupting them (as we get here
+ * in the middle of most any lock down).
+ * Still we don't want to call until the table exists!
+ */
+ if (kgdb_pid_init_done){
+ thread = find_task_by_pid(pid);
+ if (thread) {
+ return thread;
+ }
+ }
+ }
+ return NULL;
+}
+/* *INDENT-OFF* */
+struct hw_breakpoint {
+ unsigned enabled;
+ unsigned type;
+ unsigned len;
+ unsigned addr;
+} breakinfo[4] = { {enabled:0},
+ {enabled:0},
+ {enabled:0},
+ {enabled:0}};
+/* *INDENT-ON* */
+unsigned hw_breakpoint_status;
+void
+correct_hw_break(void)
+{
+ int breakno;
+ int correctit;
+ int breakbit;
+ unsigned dr7;
+
+ asm volatile ("movl %%db7, %0\n":"=r" (dr7)
+ :);
+ /* *INDENT-OFF* */
+ do {
+ unsigned addr0, addr1, addr2, addr3;
+ asm volatile ("movl %%db0, %0\n"
+ "movl %%db1, %1\n"
+ "movl %%db2, %2\n"
+ "movl %%db3, %3\n"
+ :"=r" (addr0), "=r"(addr1),
+ "=r"(addr2), "=r"(addr3)
+ :);
+ } while (0);
+ /* *INDENT-ON* */
+ correctit = 0;
+ for (breakno = 0; breakno < 3; breakno++) {
+ breakbit = 2 << (breakno << 1);
+ if (!(dr7 & breakbit) && breakinfo[breakno].enabled) {
+ correctit = 1;
+ dr7 |= breakbit;
+ dr7 &= ~(0xf0000 << (breakno << 2));
+ dr7 |= (((breakinfo[breakno].len << 2) |
+ breakinfo[breakno].type) << 16) <<
+ (breakno << 2);
+ switch (breakno) {
+ case 0:
+ asm volatile ("movl %0, %%dr0\n"::"r"
+ (breakinfo[breakno].addr));
+ break;
+
+ case 1:
+ asm volatile ("movl %0, %%dr1\n"::"r"
+ (breakinfo[breakno].addr));
+ break;
+
+ case 2:
+ asm volatile ("movl %0, %%dr2\n"::"r"
+ (breakinfo[breakno].addr));
+ break;
+
+ case 3:
+ asm volatile ("movl %0, %%dr3\n"::"r"
+ (breakinfo[breakno].addr));
+ break;
+ }
+ } else if ((dr7 & breakbit) && !breakinfo[breakno].enabled) {
+ correctit = 1;
+ dr7 &= ~breakbit;
+ dr7 &= ~(0xf0000 << (breakno << 2));
+ }
+ }
+ if (correctit) {
+ asm volatile ("movl %0, %%db7\n"::"r" (dr7));
+ }
+}
+
+int
+remove_hw_break(unsigned breakno)
+{
+ if (!breakinfo[breakno].enabled) {
+ return -1;
+ }
+ breakinfo[breakno].enabled = 0;
+ return 0;
+}
+
+int
+set_hw_break(unsigned breakno, unsigned type, unsigned len, unsigned addr)
+{
+ if (breakinfo[breakno].enabled) {
+ return -1;
+ }
+ breakinfo[breakno].enabled = 1;
+ breakinfo[breakno].type = type;
+ breakinfo[breakno].len = len;
+ breakinfo[breakno].addr = addr;
+ return 0;
+}
+
+#ifdef CONFIG_SMP
+static int in_kgdb_console = 0;
+
+int
+in_kgdb(struct pt_regs *regs)
+{
+ unsigned flags;
+ int cpu = smp_processor_id();
+ in_kgdb_called = 1;
+ if (!spin_is_locked(&kgdb_spinlock)) {
+ if (in_kgdb_here_log[cpu] || /* we are holding this cpu */
+ in_kgdb_console) { /* or we are doing slow i/o */
+ return 1;
+ }
+ return 0;
+ }
+
+ /* As I see it the only reason not to let all cpus spin on
+ * the same spin_lock is to allow selected ones to proceed.
+ * This would be a good thing, so we leave it this way.
+ * Maybe someday.... Done !
+
+ * in_kgdb() is called from an NMI so we don't pretend
+ * to have any resources, like printk() for example.
+ */
+
+ kgdb_local_irq_save(flags); /* only local here, to avoid hanging */
+ /*
+ * log arival of this cpu
+ * The NMI keeps on ticking. Protect against recurring more
+ * than once, and ignor the cpu that has the kgdb lock
+ */
+ in_kgdb_entry_log[cpu]++;
+ in_kgdb_here_log[cpu] = regs;
+ if (cpu == spinlock_cpu || waiting_cpus[cpu].task)
+ goto exit_in_kgdb;
+
+ /*
+ * For protection of the initilization of the spin locks by kgdb
+ * it locks the kgdb spinlock before it gets the wait locks set
+ * up. We wait here for the wait lock to be taken. If the
+ * kgdb lock goes away first?? Well, it could be a slow exit
+ * sequence where the wait lock is removed prior to the kgdb lock
+ * so if kgdb gets unlocked, we just exit.
+ */
+
+ while (spin_is_locked(&kgdb_spinlock) &&
+ !spin_is_locked(waitlocks + cpu)) ;
+ if (!spin_is_locked(&kgdb_spinlock))
+ goto exit_in_kgdb;
+
+ waiting_cpus[cpu].task = current;
+ waiting_cpus[cpu].pid = (current->pid) ? : (PID_MAX + cpu);
+ waiting_cpus[cpu].regs = regs;
+
+ spin_unlock_wait(waitlocks + cpu);
+
+ /*
+ * log departure of this cpu
+ */
+ waiting_cpus[cpu].task = 0;
+ waiting_cpus[cpu].pid = 0;
+ waiting_cpus[cpu].regs = 0;
+ correct_hw_break();
+ exit_in_kgdb:
+ in_kgdb_here_log[cpu] = 0;
+ kgdb_local_irq_restore(flags);
+ return 1;
+ /*
+ spin_unlock(continuelocks + smp_processor_id());
+ */
+}
+
+void
+smp__in_kgdb(struct pt_regs regs)
+{
+ ack_APIC_irq();
+ in_kgdb(®s);
+}
+#else
+int
+in_kgdb(struct pt_regs *regs)
+{
+ return (kgdb_spinlock);
+}
+#endif
+
+void
+printexceptioninfo(int exceptionNo, int errorcode, char *buffer)
+{
+ unsigned dr6;
+ int i;
+ switch (exceptionNo) {
+ case 1: /* debug exception */
+ break;
+ case 3: /* breakpoint */
+ sprintf(buffer, "Software breakpoint");
+ return;
+ default:
+ sprintf(buffer, "Details not available");
+ return;
+ }
+ asm volatile ("movl %%db6, %0\n":"=r" (dr6)
+ :);
+ if (dr6 & 0x4000) {
+ sprintf(buffer, "Single step");
+ return;
+ }
+ for (i = 0; i < 4; ++i) {
+ if (dr6 & (1 << i)) {
+ sprintf(buffer, "Hardware breakpoint %d", i);
+ return;
+ }
+ }
+ sprintf(buffer, "Unknown trap");
+ return;
+}
+
+/*
+ * This function does all command procesing for interfacing to gdb.
+ *
+ * NOTE: The INT nn instruction leaves the state of the interrupt
+ * enable flag UNCHANGED. That means that when this routine
+ * is entered via a breakpoint (INT 3) instruction from code
+ * that has interrupts enabled, then interrupts will STILL BE
+ * enabled when this routine is entered. The first thing that
+ * we do here is disable interrupts so as to prevent recursive
+ * entries and bothersome serial interrupts while we are
+ * trying to run the serial port in polled mode.
+ *
+ * For kernel version 2.1.xx the kgdb_cli() actually gets a spin lock so
+ * it is always necessary to do a restore_flags before returning
+ * so as to let go of that lock.
+ */
+int
+kgdb_handle_exception(int exceptionVector,
+ int signo, int err_code, struct pt_regs *linux_regs)
+{
+ struct task_struct *usethread = NULL;
+ struct task_struct *thread_list_start = 0, *thread = NULL;
+ int addr, length;
+ int breakno, breaktype;
+ char *ptr;
+ int newPC;
+ threadref thref;
+ int threadid;
+ int thread_min = PID_MAX + MAX_NO_CPUS;
+#ifdef old_thread_list
+ int maxthreads;
+#endif
+ int nothreads;
+ unsigned long flags;
+ int gdb_regs[NUMREGBYTES / 4];
+ int dr6;
+ IF_SMP(int entry_state = 0); /* 0, ok, 1, no nmi, 2 sync failed */
+#define NO_NMI 1
+#define NO_SYNC 2
+#define regs (*linux_regs)
+#define NUMREGS NUMREGBYTES/4
+ /*
+ * If the entry is not from the kernel then return to the Linux
+ * trap handler and let it process the interrupt normally.
+ */
+ if ((linux_regs->eflags & VM_MASK) || (3 & linux_regs->xcs)) {
+ printk("ignoring non-kernel exception\n");
+ print_regs(®s);
+ return (0);
+ }
+ /*
+ * If we're using eth mode, set the 'mode' in the netdevice.
+ */
+
+ if (kgdboe)
+ netpoll_set_trap(1);
+
+ kgdb_local_irq_save(flags);
+
+ /* Get kgdb spinlock */
+
+ KGDB_SPIN_LOCK(&kgdb_spinlock);
+ rdtscll(kgdb_info.entry_tsc);
+ /*
+ * We depend on this spinlock and the NMI watch dog to control the
+ * other cpus. They will arrive at "in_kgdb()" as a result of the
+ * NMI and will wait there for the following spin locks to be
+ * released.
+ */
+#ifdef CONFIG_SMP
+
+#if 0
+ if (cpu_callout_map & ~MAX_CPU_MASK) {
+ printk("kgdb : too many cpus, possibly not mapped"
+ " in contiguous space, change MAX_NO_CPUS"
+ " in kgdb_stub and make new kernel.\n"
+ " cpu_callout_map is %lx\n", cpu_callout_map);
+ goto exit_just_unlock;
+ }
+#endif
+ if (spinlock_count == 1) {
+ int time = 0, end_time, dum = 0;
+ int i;
+ int cpu_logged_in[MAX_NO_CPUS] = {[0 ... MAX_NO_CPUS - 1] = (0)
+ };
+ if (remote_debug) {
+ printk("kgdb : cpu %d entry, syncing others\n",
+ smp_processor_id());
+ }
+ for (i = 0; i < MAX_NO_CPUS; i++) {
+ /*
+ * Use trylock as we may already hold the lock if
+ * we are holding the cpu. Net result is all
+ * locked.
+ */
+ spin_trylock(&waitlocks[i]);
+ }
+ for (i = 0; i < MAX_NO_CPUS; i++)
+ cpu_logged_in[i] = 0;
+ /*
+ * Wait for their arrival. We know the watch dog is active if
+ * in_kgdb() has ever been called, as it is always called on a
+ * watchdog tick.
+ */
+ rdtsc(dum, time);
+ end_time = time + 2; /* Note: we use the High order bits! */
+ i = 1;
+ if (num_online_cpus() > 1) {
+ int me_in_kgdb = in_kgdb_entry_log[smp_processor_id()];
+ smp_send_nmi_allbutself();
+
+ while (i < num_online_cpus() && time != end_time) {
+ int j;
+ for (j = 0; j < MAX_NO_CPUS; j++) {
+ if (waiting_cpus[j].task &&
+ waiting_cpus[j].task != NOCPU &&
+ !cpu_logged_in[j]) {
+ i++;
+ cpu_logged_in[j] = 1;
+ if (remote_debug) {
+ printk
+ ("kgdb : cpu %d arrived at kgdb\n",
+ j);
+ }
+ break;
+ } else if (!waiting_cpus[j].task &&
+ !cpu_online(j)) {
+ waiting_cpus[j].task = NOCPU;
+ cpu_logged_in[j] = 1;
+ waiting_cpus[j].hold = 1;
+ break;
+ }
+ if (!waiting_cpus[j].task &&
+ in_kgdb_here_log[j]) {
+
+ int wait = 100000;
+ while (wait--) ;
+ if (!waiting_cpus[j].task &&
+ in_kgdb_here_log[j]) {
+ printk
+ ("kgdb : cpu %d stall"
+ " in in_kgdb\n",
+ j);
+ i++;
+ cpu_logged_in[j] = 1;
+ waiting_cpus[j].task =
+ (struct task_struct
+ *) 1;
+ }
+ }
+ }
+
+ if (in_kgdb_entry_log[smp_processor_id()] >
+ (me_in_kgdb + 10)) {
+ break;
+ }
+
+ rdtsc(dum, time);
+ }
+ if (i < num_online_cpus()) {
+ printk
+ ("kgdb : time out, proceeding without sync\n");
+#if 0
+ printk("kgdb : Waiting_cpus: 0 = %d, 1 = %d\n",
+ waiting_cpus[0].task != 0,
+ waiting_cpus[1].task != 0);
+ printk("kgdb : Cpu_logged in: 0 = %d, 1 = %d\n",
+ cpu_logged_in[0], cpu_logged_in[1]);
+ printk
+ ("kgdb : in_kgdb_here_log in: 0 = %d, 1 = %d\n",
+ in_kgdb_here_log[0] != 0,
+ in_kgdb_here_log[1] != 0);
+#endif
+ entry_state = NO_SYNC;
+ } else {
+#if 0
+ int ent =
+ in_kgdb_entry_log[smp_processor_id()] -
+ me_in_kgdb;
+ printk("kgdb : sync after %d entries\n", ent);
+#endif
+ }
+ } else {
+ if (remote_debug) {
+ printk
+ ("kgdb : %d cpus, but watchdog not active\n"
+ "proceeding without locking down other cpus\n",
+ num_online_cpus());
+ entry_state = NO_NMI;
+ }
+ }
+ }
+#endif
+
+ if (remote_debug) {
+ unsigned long *lp = (unsigned long *) &linux_regs;
+
+ printk("handle_exception(exceptionVector=%d, "
+ "signo=%d, err_code=%d, linux_regs=%p)\n",
+ exceptionVector, signo, err_code, linux_regs);
+ if (debug_regs) {
+ print_regs(®s);
+ printk("Stk: %8lx %8lx %8lx %8lx"
+ " %8lx %8lx %8lx %8lx\n",
+ lp[0], lp[1], lp[2], lp[3],
+ lp[4], lp[5], lp[6], lp[7]);
+ printk(" %8lx %8lx %8lx %8lx"
+ " %8lx %8lx %8lx %8lx\n",
+ lp[8], lp[9], lp[10], lp[11],
+ lp[12], lp[13], lp[14], lp[15]);
+ printk(" %8lx %8lx %8lx %8lx "
+ "%8lx %8lx %8lx %8lx\n",
+ lp[16], lp[17], lp[18], lp[19],
+ lp[20], lp[21], lp[22], lp[23]);
+ printk(" %8lx %8lx %8lx %8lx "
+ "%8lx %8lx %8lx %8lx\n",
+ lp[24], lp[25], lp[26], lp[27],
+ lp[28], lp[29], lp[30], lp[31]);
+ }
+ }
+
+ /* Disable hardware debugging while we are in kgdb */
+ /* Get the debug register status register */
+/* *INDENT-OFF* */
+ __asm__("movl %0,%%db7"
+ : /* no output */
+ :"r"(0));
+
+ asm volatile ("movl %%db6, %0\n"
+ :"=r" (hw_breakpoint_status)
+ :);
+
+/* *INDENT-ON* */
+ switch (exceptionVector) {
+ case 0: /* divide error */
+ case 1: /* debug exception */
+ case 2: /* NMI */
+ case 3: /* breakpoint */
+ case 4: /* overflow */
+ case 5: /* bounds check */
+ case 6: /* invalid opcode */
+ case 7: /* device not available */
+ case 8: /* double fault (errcode) */
+ case 10: /* invalid TSS (errcode) */
+ case 12: /* stack fault (errcode) */
+ case 16: /* floating point error */
+ case 17: /* alignment check (errcode) */
+ default: /* any undocumented */
+ break;
+ case 11: /* segment not present (errcode) */
+ case 13: /* general protection (errcode) */
+ case 14: /* page fault (special errcode) */
+ case 19: /* cache flush denied */
+ if (mem_err_expected) {
+ /*
+ * This fault occured because of the
+ * get_char or set_char routines. These
+ * two routines use either eax of edx to
+ * indirectly reference the location in
+ * memory that they are working with.
+ * For a page fault, when we return the
+ * instruction will be retried, so we
+ * have to make sure that these
+ * registers point to valid memory.
+ */
+ mem_err = 1; /* set mem error flag */
+ mem_err_expected = 0;
+ mem_err_cnt++; /* helps in debugging */
+ /* make valid address */
+ regs.eax = (long) &garbage_loc;
+ /* make valid address */
+ regs.edx = (long) &garbage_loc;
+ if (remote_debug)
+ printk("Return after memory error: "
+ "mem_err_cnt=%d\n", mem_err_cnt);
+ if (debug_regs)
+ print_regs(®s);
+ goto exit_kgdb;
+ }
+ break;
+ }
+ if (remote_debug)
+ printk("kgdb : entered kgdb on cpu %d\n", smp_processor_id());
+
+ gdb_i386vector = exceptionVector;
+ gdb_i386errcode = err_code;
+ kgdb_info.called_from = __builtin_return_address(0);
+#ifdef CONFIG_SMP
+ /*
+ * OK, we can now communicate, lets tell gdb about the sync.
+ * but only if we had a problem.
+ */
+ switch (entry_state) {
+ case NO_NMI:
+ to_gdb("NMI not active, other cpus not stopped\n");
+ break;
+ case NO_SYNC:
+ to_gdb("Some cpus not stopped, see 'kgdb_info' for details\n");
+ default:;
+ }
+
+#endif
+/*
+ * Set up the gdb function call area.
+ */
+ trap_cpu = smp_processor_id();
+ OLD_esp = NEW_esp = (int) (&linux_regs->esp);
+
+ IF_SMP(once_again:)
+ /* reply to host that an exception has occurred */
+ remcomOutBuffer[0] = 'S';
+ remcomOutBuffer[1] = hexchars[signo >> 4];
+ remcomOutBuffer[2] = hexchars[signo % 16];
+ remcomOutBuffer[3] = 0;
+
+ putpacket(remcomOutBuffer);
+
+ while (1 == 1) {
+ error = 0;
+ remcomOutBuffer[0] = 0;
+ getpacket(remcomInBuffer);
+ switch (remcomInBuffer[0]) {
+ case '?':
+ remcomOutBuffer[0] = 'S';
+ remcomOutBuffer[1] = hexchars[signo >> 4];
+ remcomOutBuffer[2] = hexchars[signo % 16];
+ remcomOutBuffer[3] = 0;
+ break;
+ case 'd':
+ remote_debug = !(remote_debug); /* toggle debug flag */
+ printk("Remote debug %s\n",
+ remote_debug ? "on" : "off");
+ break;
+ case 'g': /* return the value of the CPU registers */
+ get_gdb_regs(usethread, ®s, gdb_regs);
+ mem2hex((char *) gdb_regs,
+ remcomOutBuffer, NUMREGBYTES, 0);
+ break;
+ case 'G': /* set the value of the CPU registers - return OK */
+ hex2mem(&remcomInBuffer[1],
+ (char *) gdb_regs, NUMREGBYTES, 0);
+ if (!usethread || usethread == current) {
+ gdb_regs_to_regs(gdb_regs, ®s);
+ strcpy(remcomOutBuffer, "OK");
+ } else {
+ strcpy(remcomOutBuffer, "E00");
+ }
+ break;
+
+ case 'P':{ /* set the value of a single CPU register -
+ return OK */
+ /*
+ * For some reason, gdb wants to talk about psudo
+ * registers (greater than 15). These may have
+ * meaning for ptrace, but for us it is safe to
+ * ignor them. We do this by dumping them into
+ * _GS which we also ignor, but do have memory for.
+ */
+ int regno;
+
+ ptr = &remcomInBuffer[1];
+ regs_to_gdb_regs(gdb_regs, ®s);
+ if ((!usethread || usethread == current) &&
+ hexToInt(&ptr, ®no) &&
+ *ptr++ == '=' && (regno >= 0)) {
+ regno =
+ (regno >= NUMREGS ? _GS : regno);
+ hex2mem(ptr, (char *) &gdb_regs[regno],
+ 4, 0);
+ gdb_regs_to_regs(gdb_regs, ®s);
+ strcpy(remcomOutBuffer, "OK");
+ break;
+ }
+ strcpy(remcomOutBuffer, "E01");
+ break;
+ }
+
+ /* mAA..AA,LLLL Read LLLL bytes at address AA..AA */
+ case 'm':
+ /* TRY TO READ %x,%x. IF SUCCEED, SET PTR = 0 */
+ ptr = &remcomInBuffer[1];
+ if (hexToInt(&ptr, &addr) &&
+ (*(ptr++) == ',') && (hexToInt(&ptr, &length))) {
+ ptr = 0;
+ /*
+ * hex doubles the byte count
+ */
+ if (length > (BUFMAX / 2))
+ length = BUFMAX / 2;
+ mem2hex((char *) addr,
+ remcomOutBuffer, length, 1);
+ if (mem_err) {
+ strcpy(remcomOutBuffer, "E03");
+ debug_error("memory fault\n", NULL);
+ }
+ }
+
+ if (ptr) {
+ strcpy(remcomOutBuffer, "E01");
+ debug_error
+ ("malformed read memory command: %s\n",
+ remcomInBuffer);
+ }
+ break;
+
+ /* MAA..AA,LLLL:
+ Write LLLL bytes at address AA.AA return OK */
+ case 'M':
+ /* TRY TO READ '%x,%x:'. IF SUCCEED, SET PTR = 0 */
+ ptr = &remcomInBuffer[1];
+ if (hexToInt(&ptr, &addr) &&
+ (*(ptr++) == ',') &&
+ (hexToInt(&ptr, &length)) && (*(ptr++) == ':')) {
+ hex2mem(ptr, (char *) addr, length, 1);
+
+ if (mem_err) {
+ strcpy(remcomOutBuffer, "E03");
+ debug_error("memory fault\n", NULL);
+ } else {
+ strcpy(remcomOutBuffer, "OK");
+ }
+
+ ptr = 0;
+ }
+ if (ptr) {
+ strcpy(remcomOutBuffer, "E02");
+ debug_error
+ ("malformed write memory command: %s\n",
+ remcomInBuffer);
+ }
+ break;
+ case 'S':
+ remcomInBuffer[0] = 's';
+ case 'C':
+ /* Csig;AA..AA where ;AA..AA is optional
+ * continue with signal
+ * Since signals are meaning less to us, delete that
+ * part and then fall into the 'c' code.
+ */
+ ptr = &remcomInBuffer[1];
+ length = 2;
+ while (*ptr && *ptr != ';') {
+ length++;
+ ptr++;
+ }
+ if (*ptr) {
+ do {
+ ptr++;
+ *(ptr - length++) = *ptr;
+ } while (*ptr);
+ } else {
+ remcomInBuffer[1] = 0;
+ }
+
+ /* cAA..AA Continue at address AA..AA(optional) */
+ /* sAA..AA Step one instruction from AA..AA(optional) */
+ /* D detach, reply OK and then continue */
+ case 'c':
+ case 's':
+ case 'D':
+
+ /* try to read optional parameter,
+ pc unchanged if no parm */
+ ptr = &remcomInBuffer[1];
+ if (hexToInt(&ptr, &addr)) {
+ if (remote_debug)
+ printk("Changing EIP to 0x%x\n", addr);
+
+ regs.eip = addr;
+ }
+
+ newPC = regs.eip;
+
+ /* clear the trace bit */
+ regs.eflags &= 0xfffffeff;
+
+ /* set the trace bit if we're stepping */
+ if (remcomInBuffer[0] == 's')
+ regs.eflags |= 0x100;
+
+ /* detach is a friendly version of continue. Note that
+ debugging is still enabled (e.g hit control C)
+ */
+ if (remcomInBuffer[0] == 'D') {
+ strcpy(remcomOutBuffer, "OK");
+ putpacket(remcomOutBuffer);
+ }
+
+ if (remote_debug) {
+ printk("Resuming execution\n");
+ print_regs(®s);
+ }
+ asm volatile ("movl %%db6, %0\n":"=r" (dr6)
+ :);
+ if (!(dr6 & 0x4000)) {
+ for (breakno = 0; breakno < 4; ++breakno) {
+ if (dr6 & (1 << breakno) &&
+ (breakinfo[breakno].type == 0)) {
+ /* Set restore flag */
+ regs.eflags |= 0x10000;
+ break;
+ }
+ }
+ }
+
+ if (kgdboe)
+ netpoll_set_trap(0);
+
+ correct_hw_break();
+ asm volatile ("movl %0, %%db6\n"::"r" (0));
+ goto exit_kgdb;
+
+ /* kill the program */
+ case 'k': /* do nothing */
+ break;
+
+ /* query */
+ case 'q':
+ nothreads = 0;
+ switch (remcomInBuffer[1]) {
+ case 'f':
+ threadid = 1;
+ thread_list = 2;
+ thread_list_start = (usethread ? : current);
+ case 's':
+ if (!cmp_str(&remcomInBuffer[2],
+ "ThreadInfo", 10))
+ break;
+
+ remcomOutBuffer[nothreads++] = 'm';
+ for (; threadid < PID_MAX + MAX_NO_CPUS;
+ threadid++) {
+ thread = getthread(threadid);
+ if (thread) {
+ nothreads += int_to_hex_v(
+ &remcomOutBuffer[
+ nothreads],
+ threadid);
+ if (thread_min > threadid)
+ thread_min = threadid;
+ remcomOutBuffer[
+ nothreads] = ',';
+ nothreads++;
+ if (nothreads > BUFMAX - 10)
+ break;
+ }
+ }
+ if (remcomOutBuffer[nothreads - 1] == 'm') {
+ remcomOutBuffer[nothreads - 1] = 'l';
+ } else {
+ nothreads--;
+ }
+ remcomOutBuffer[nothreads] = 0;
+ break;
+
+#ifdef old_thread_list /* Old thread info request */
+ case 'L':
+ /* List threads */
+ thread_list = 2;
+ thread_list_start = (usethread ? : current);
+ unpack_byte(remcomInBuffer + 3, &maxthreads);
+ unpack_threadid(remcomInBuffer + 5, &thref);
+ do {
+ int buf_thread_limit =
+ (BUFMAX - 22) / BUF_THREAD_ID_SIZE;
+ if (maxthreads > buf_thread_limit) {
+ maxthreads = buf_thread_limit;
+ }
+ } while (0);
+ remcomOutBuffer[0] = 'q';
+ remcomOutBuffer[1] = 'M';
+ remcomOutBuffer[4] = '0';
+ pack_threadid(remcomOutBuffer + 5, &thref);
+
+ threadid = threadref_to_int(&thref);
+ for (nothreads = 0;
+ nothreads < maxthreads &&
+ threadid < PID_MAX + MAX_NO_CPUS;
+ threadid++) {
+ thread = getthread(threadid);
+ if (thread) {
+ int_to_threadref(&thref,
+ threadid);
+ pack_threadid(remcomOutBuffer +
+ 21 +
+ nothreads * 16,
+ &thref);
+ nothreads++;
+ if (thread_min > threadid)
+ thread_min = threadid;
+ }
+ }
+
+ if (threadid == PID_MAX + MAX_NO_CPUS) {
+ remcomOutBuffer[4] = '1';
+ }
+ pack_hex_byte(remcomOutBuffer + 2, nothreads);
+ remcomOutBuffer[21 + nothreads * 16] = '\0';
+ break;
+#endif
+ case 'C':
+ /* Current thread id */
+ remcomOutBuffer[0] = 'Q';
+ remcomOutBuffer[1] = 'C';
+ threadid = current->pid;
+ if (!threadid) {
+ /*
+ * idle thread
+ */
+ for (threadid = PID_MAX;
+ threadid < PID_MAX + MAX_NO_CPUS;
+ threadid++) {
+ if (current ==
+ idle_task(threadid -
+ PID_MAX))
+ break;
+ }
+ }
+ int_to_threadref(&thref, threadid);
+ pack_threadid(remcomOutBuffer + 2, &thref);
+ remcomOutBuffer[18] = '\0';
+ break;
+
+ case 'E':
+ /* Print exception info */
+ printexceptioninfo(exceptionVector,
+ err_code, remcomOutBuffer);
+ break;
+ case 'T':{
+ char * nptr;
+ /* Thread extra info */
+ if (!cmp_str(&remcomInBuffer[2],
+ "hreadExtraInfo,", 15)) {
+ break;
+ }
+ ptr = &remcomInBuffer[17];
+ hexToInt(&ptr, &threadid);
+ thread = getthread(threadid);
+ nptr = &thread->comm[0];
+ length = 0;
+ ptr = &remcomOutBuffer[0];
+ do {
+ length++;
+ ptr = pack_hex_byte(ptr, *nptr++);
+ } while (*nptr && length < 16);
+ /*
+ * would like that 16 to be the size of
+ * task_struct.comm but don't know the
+ * syntax..
+ */
+ *ptr = 0;
+ }
+ }
+ break;
+
+ /* task related */
+ case 'H':
+ switch (remcomInBuffer[1]) {
+ case 'g':
+ ptr = &remcomInBuffer[2];
+ hexToInt(&ptr, &threadid);
+ thread = getthread(threadid);
+ if (!thread) {
+ remcomOutBuffer[0] = 'E';
+ remcomOutBuffer[1] = '\0';
+ break;
+ }
+ /*
+ * Just in case I forget what this is all about,
+ * the "thread info" command to gdb causes it
+ * to ask for a thread list. It then switches
+ * to each thread and asks for the registers.
+ * For this (and only this) usage, we want to
+ * fudge the registers of tasks not on the run
+ * list (i.e. waiting) to show the routine that
+ * called schedule. Also, gdb, is a minimalist
+ * in that if the current thread is the last
+ * it will not re-read the info when done.
+ * This means that in this case we must show
+ * the real registers. So here is how we do it:
+ * Each entry we keep track of the min
+ * thread in the list (the last that gdb will)
+ * get info for. We also keep track of the
+ * starting thread.
+ * "thread_list" is cleared when switching back
+ * to the min thread if it is was current, or
+ * if it was not current, thread_list is set
+ * to 1. When the switch to current comes,
+ * if thread_list is 1, clear it, else do
+ * nothing.
+ */
+ usethread = thread;
+ if ((thread_list == 1) &&
+ (thread == thread_list_start)) {
+ thread_list = 0;
+ }
+ if (thread_list && (threadid == thread_min)) {
+ if (thread == thread_list_start) {
+ thread_list = 0;
+ } else {
+ thread_list = 1;
+ }
+ }
+ /* follow through */
+ case 'c':
+ remcomOutBuffer[0] = 'O';
+ remcomOutBuffer[1] = 'K';
+ remcomOutBuffer[2] = '\0';
+ break;
+ }
+ break;
+
+ /* Query thread status */
+ case 'T':
+ ptr = &remcomInBuffer[1];
+ hexToInt(&ptr, &threadid);
+ thread = getthread(threadid);
+ if (thread) {
+ remcomOutBuffer[0] = 'O';
+ remcomOutBuffer[1] = 'K';
+ remcomOutBuffer[2] = '\0';
+ if (thread_min > threadid)
+ thread_min = threadid;
+ } else {
+ remcomOutBuffer[0] = 'E';
+ remcomOutBuffer[1] = '\0';
+ }
+ break;
+
+ case 'Y': /* set up a hardware breakpoint */
+ ptr = &remcomInBuffer[1];
+ hexToInt(&ptr, &breakno);
+ ptr++;
+ hexToInt(&ptr, &breaktype);
+ ptr++;
+ hexToInt(&ptr, &length);
+ ptr++;
+ hexToInt(&ptr, &addr);
+ if (set_hw_break(breakno & 0x3,
+ breaktype & 0x3,
+ length & 0x3, addr) == 0) {
+ strcpy(remcomOutBuffer, "OK");
+ } else {
+ strcpy(remcomOutBuffer, "ERROR");
+ }
+ break;
+
+ /* Remove hardware breakpoint */
+ case 'y':
+ ptr = &remcomInBuffer[1];
+ hexToInt(&ptr, &breakno);
+ if (remove_hw_break(breakno & 0x3) == 0) {
+ strcpy(remcomOutBuffer, "OK");
+ } else {
+ strcpy(remcomOutBuffer, "ERROR");
+ }
+ break;
+
+ case 'r': /* reboot */
+ strcpy(remcomOutBuffer, "OK");
+ putpacket(remcomOutBuffer);
+ /*to_gdb("Rebooting\n"); */
+ /* triplefault no return from here */
+ {
+ static long no_idt[2];
+ __asm__ __volatile__("lidt %0"::"m"(no_idt[0]));
+ BREAKPOINT;
+ }
+
+ } /* switch */
+
+ /* reply to the request */
+ putpacket(remcomOutBuffer);
+ } /* while(1==1) */
+ /*
+ * reached by goto only.
+ */
+ exit_kgdb:
+ /*
+ * Here is where we set up to trap a gdb function call. NEW_esp
+ * will be changed if we are trying to do this. We handle both
+ * adding and subtracting, thus allowing gdb to put grung on
+ * the stack which it removes later.
+ */
+ if (NEW_esp != OLD_esp) {
+ int *ptr = END_OF_LOOKASIDE;
+ if (NEW_esp < OLD_esp)
+ ptr -= (OLD_esp - NEW_esp) / sizeof (int);
+ *--ptr = linux_regs->eflags;
+ *--ptr = linux_regs->xcs;
+ *--ptr = linux_regs->eip;
+ *--ptr = linux_regs->ecx;
+ *--ptr = linux_regs->ebx;
+ *--ptr = linux_regs->eax;
+ linux_regs->ecx = NEW_esp - (sizeof (int) * 6);
+ linux_regs->ebx = (unsigned int) END_OF_LOOKASIDE;
+ if (NEW_esp < OLD_esp) {
+ linux_regs->eip = (unsigned int) fn_call_stub;
+ } else {
+ linux_regs->eip = (unsigned int) fn_rtn_stub;
+ linux_regs->eax = NEW_esp;
+ }
+ linux_regs->eflags &= ~(IF_BIT | TF_BIT);
+ }
+#ifdef CONFIG_SMP
+ /*
+ * Release gdb wait locks
+ * Sanity check time. Must have at least one cpu to run. Also single
+ * step must not be done if the current cpu is on hold.
+ */
+ if (spinlock_count == 1) {
+ int ss_hold = (regs.eflags & 0x100) && kgdb_info.hold_on_sstep;
+ int cpu_avail = 0;
+ int i;
+
+ for (i = 0; i < MAX_NO_CPUS; i++) {
+ if (!cpu_online(i))
+ break;
+ if (!hold_cpu(i)) {
+ cpu_avail = 1;
+ }
+ }
+ /*
+ * Early in the bring up there will be NO cpus on line...
+ */
+ if (!cpu_avail && !cpus_empty(cpu_online_map)) {
+ to_gdb("No cpus unblocked, see 'kgdb_info.hold_cpu'\n");
+ goto once_again;
+ }
+ if (hold_cpu(smp_processor_id()) && (regs.eflags & 0x100)) {
+ to_gdb
+ ("Current cpu must be unblocked to single step\n");
+ goto once_again;
+ }
+ if (!(ss_hold)) {
+ int i;
+ for (i = 0; i < MAX_NO_CPUS; i++) {
+ if (!hold_cpu(i)) {
+ spin_unlock(&waitlocks[i]);
+ }
+ }
+ } else {
+ spin_unlock(&waitlocks[smp_processor_id()]);
+ }
+ /* Release kgdb spinlock */
+ KGDB_SPIN_UNLOCK(&kgdb_spinlock);
+ /*
+ * If this cpu is on hold, this is where we
+ * do it. Note, the NMI will pull us out of here,
+ * but will return as the above lock is not held.
+ * We will stay here till another cpu releases the lock for us.
+ */
+ spin_unlock_wait(waitlocks + smp_processor_id());
+ kgdb_local_irq_restore(flags);
+ return (0);
+ }
+#if 0
+exit_just_unlock:
+#endif
+#endif
+ /* Release kgdb spinlock */
+ KGDB_SPIN_UNLOCK(&kgdb_spinlock);
+ kgdb_local_irq_restore(flags);
+ return (0);
+}
+
+/* this function is used to set up exception handlers for tracing and
+ * breakpoints.
+ * This function is not needed as the above line does all that is needed.
+ * We leave it for backward compatitability...
+ */
+void
+set_debug_traps(void)
+{
+ /*
+ * linux_debug_hook is defined in traps.c. We store a pointer
+ * to our own exception handler into it.
+
+ * But really folks, every hear of labeled common, an old Fortran
+ * concept. Lots of folks can reference it and it is define if
+ * anyone does. Only one can initialize it at link time. We do
+ * this with the hook. See the statement above. No need for any
+ * executable code and it is ready as soon as the kernel is
+ * loaded. Very desirable in kernel debugging.
+
+ linux_debug_hook = handle_exception ;
+ */
+
+ /* In case GDB is started before us, ack any packets (presumably
+ "$?#xx") sitting there.
+ putDebugChar ('+');
+
+ initialized = 1;
+ */
+}
+
+/* This function will generate a breakpoint exception. It is used at the
+ beginning of a program to sync up with a debugger and can be used
+ otherwise as a quick means to stop program execution and "break" into
+ the debugger. */
+/* But really, just use the BREAKPOINT macro. We will handle the int stuff
+ */
+
+#ifdef later
+/*
+ * possibly we should not go thru the traps.c code at all? Someday.
+ */
+void
+do_kgdb_int3(struct pt_regs *regs, long error_code)
+{
+ kgdb_handle_exception(3, 5, error_code, regs);
+ return;
+}
+#endif
+#undef regs
+#ifdef CONFIG_TRAP_BAD_SYSCALL_EXITS
+asmlinkage void
+bad_sys_call_exit(int stuff)
+{
+ struct pt_regs *regs = (struct pt_regs *) &stuff;
+ printk("Sys call %d return with %x preempt_count\n",
+ (int) regs->orig_eax, preempt_count());
+}
+#endif
+#ifdef CONFIG_STACK_OVERFLOW_TEST
+#include <asm/kgdb.h>
+asmlinkage void
+stack_overflow(void)
+{
+#ifdef BREAKPOINT
+ BREAKPOINT;
+#else
+ printk("Kernel stack overflow, looping forever\n");
+#endif
+ while (1) {
+ }
+}
+#endif
+
+#if defined(CONFIG_SMP) || defined(CONFIG_KGDB_CONSOLE)
+char gdbconbuf[BUFMAX];
+
+static void
+kgdb_gdb_message(const char *s, unsigned count)
+{
+ int i;
+ int wcount;
+ char *bufptr;
+ /*
+ * This takes care of NMI while spining out chars to gdb
+ */
+ IF_SMP(in_kgdb_console = 1);
+ gdbconbuf[0] = 'O';
+ bufptr = gdbconbuf + 1;
+ while (count > 0) {
+ if ((count << 1) > (BUFMAX - 2)) {
+ wcount = (BUFMAX - 2) >> 1;
+ } else {
+ wcount = count;
+ }
+ count -= wcount;
+ for (i = 0; i < wcount; i++) {
+ bufptr = pack_hex_byte(bufptr, s[i]);
+ }
+ *bufptr = '\0';
+ s += wcount;
+
+ putpacket(gdbconbuf);
+
+ }
+ IF_SMP(in_kgdb_console = 0);
+}
+#endif
+#ifdef CONFIG_SMP
+static void
+to_gdb(const char *s)
+{
+ int count = 0;
+ while (s[count] && (count++ < BUFMAX)) ;
+ kgdb_gdb_message(s, count);
+}
+#endif
+#ifdef CONFIG_KGDB_CONSOLE
+#include <linux/console.h>
+#include <linux/init.h>
+#include <linux/fs.h>
+#include <asm/uaccess.h>
+#include <asm/semaphore.h>
+
+void
+kgdb_console_write(struct console *co, const char *s, unsigned count)
+{
+
+ if (gdb_i386vector == -1) {
+ /*
+ * We have not yet talked to gdb. What to do...
+ * lets break, on continue we can do the write.
+ * But first tell him whats up. Uh, well no can do,
+ * as this IS the console. Oh well...
+ * We do need to wait or the messages will be lost.
+ * Other option would be to tell the above code to
+ * ignore this breakpoint and do an auto return,
+ * but that might confuse gdb. Also this happens
+ * early enough in boot up that we don't have the traps
+ * set up yet, so...
+ */
+ breakpoint();
+ }
+ kgdb_gdb_message(s, count);
+}
+
+/*
+ * ------------------------------------------------------------
+ * Serial KGDB driver
+ * ------------------------------------------------------------
+ */
+
+static struct console kgdbcons = {
+ name:"kgdb",
+ write:kgdb_console_write,
+#ifdef CONFIG_KGDB_USER_CONSOLE
+ device:kgdb_console_device,
+#endif
+ flags:CON_PRINTBUFFER | CON_ENABLED,
+ index:-1,
+};
+
+/*
+ * The trick here is that this file gets linked before printk.o
+ * That means we get to peer at the console info in the command
+ * line before it does. If we are up, we register, otherwise,
+ * do nothing. By returning 0, we allow printk to look also.
+ */
+static int kgdb_console_enabled;
+
+int __init
+kgdb_console_init(char *str)
+{
+ if ((strncmp(str, "kgdb", 4) == 0) || (strncmp(str, "gdb", 3) == 0)) {
+ register_console(&kgdbcons);
+ kgdb_console_enabled = 1;
+ }
+ return 0; /* let others look at the string */
+}
+
+__setup("console=", kgdb_console_init);
+
+#ifdef CONFIG_KGDB_USER_CONSOLE
+static kdev_t kgdb_console_device(struct console *c);
+/* This stuff sort of works, but it knocks out telnet devices
+ * we are leaving it here in case we (or you) find time to figure it out
+ * better..
+ */
+
+/*
+ * We need a real char device as well for when the console is opened for user
+ * space activities.
+ */
+
+static int
+kgdb_consdev_open(struct inode *inode, struct file *file)
+{
+ return 0;
+}
+
+static ssize_t
+kgdb_consdev_write(struct file *file, const char *buf,
+ size_t count, loff_t * ppos)
+{
+ int size, ret = 0;
+ static char kbuf[128];
+ static DECLARE_MUTEX(sem);
+
+ /* We are not reentrant... */
+ if (down_interruptible(&sem))
+ return -ERESTARTSYS;
+
+ while (count > 0) {
+ /* need to copy the data from user space */
+ size = count;
+ if (size > sizeof (kbuf))
+ size = sizeof (kbuf);
+ if (copy_from_user(kbuf, buf, size)) {
+ ret = -EFAULT;
+ break;;
+ }
+ kgdb_console_write(&kgdbcons, kbuf, size);
+ count -= size;
+ ret += size;
+ buf += size;
+ }
+
+ up(&sem);
+
+ return ret;
+}
+
+struct file_operations kgdb_consdev_fops = {
+ open:kgdb_consdev_open,
+ write:kgdb_consdev_write
+};
+static kdev_t
+kgdb_console_device(struct console *c)
+{
+ return MKDEV(TTYAUX_MAJOR, 1);
+}
+
+/*
+ * This routine gets called from the serial stub in the i386/lib
+ * This is so it is done late in bring up (just before the console open).
+ */
+void
+kgdb_console_finit(void)
+{
+ if (kgdb_console_enabled) {
+ char *cptr = cdevname(MKDEV(TTYAUX_MAJOR, 1));
+ char *cp = cptr;
+ while (*cptr && *cptr != '(')
+ cptr++;
+ *cptr = 0;
+ unregister_chrdev(TTYAUX_MAJOR, cp);
+ register_chrdev(TTYAUX_MAJOR, "kgdb", &kgdb_consdev_fops);
+ }
+}
+#endif
+#endif
+#ifdef CONFIG_KGDB_TS
+#include <asm/msr.h> /* time stamp code */
+#include <asm/hardirq.h> /* in_interrupt */
+#ifdef CONFIG_KGDB_TS_64
+#define DATA_POINTS 64
+#endif
+#ifdef CONFIG_KGDB_TS_128
+#define DATA_POINTS 128
+#endif
+#ifdef CONFIG_KGDB_TS_256
+#define DATA_POINTS 256
+#endif
+#ifdef CONFIG_KGDB_TS_512
+#define DATA_POINTS 512
+#endif
+#ifdef CONFIG_KGDB_TS_1024
+#define DATA_POINTS 1024
+#endif
+#ifndef DATA_POINTS
+#define DATA_POINTS 128 /* must be a power of two */
+#endif
+#define INDEX_MASK (DATA_POINTS - 1)
+#if (INDEX_MASK & DATA_POINTS)
+#error "CONFIG_KGDB_TS_COUNT must be a power of 2"
+#endif
+struct kgdb_and_then_struct {
+#ifdef CONFIG_SMP
+ int on_cpu;
+#endif
+ struct task_struct *task;
+ long long at_time;
+ int from_ln;
+ char *in_src;
+ void *from;
+ int *with_shpf;
+ int data0;
+ int data1;
+};
+struct kgdb_and_then_struct2 {
+#ifdef CONFIG_SMP
+ int on_cpu;
+#endif
+ struct task_struct *task;
+ long long at_time;
+ int from_ln;
+ char *in_src;
+ void *from;
+ int *with_shpf;
+ struct task_struct *t1;
+ struct task_struct *t2;
+};
+struct kgdb_and_then_struct kgdb_data[DATA_POINTS];
+
+struct kgdb_and_then_struct *kgdb_and_then = &kgdb_data[0];
+int kgdb_and_then_count;
+
+void
+kgdb_tstamp(int line, char *source, int data0, int data1)
+{
+ static spinlock_t ts_spin = SPIN_LOCK_UNLOCKED;
+ int flags;
+ kgdb_local_irq_save(flags);
+ spin_lock(&ts_spin);
+ rdtscll(kgdb_and_then->at_time);
+#ifdef CONFIG_SMP
+ kgdb_and_then->on_cpu = smp_processor_id();
+#endif
+ kgdb_and_then->task = current;
+ kgdb_and_then->from_ln = line;
+ kgdb_and_then->in_src = source;
+ kgdb_and_then->from = __builtin_return_address(0);
+ kgdb_and_then->with_shpf = (int *) (((flags & IF_BIT) >> 9) |
+ (preempt_count() << 8));
+ kgdb_and_then->data0 = data0;
+ kgdb_and_then->data1 = data1;
+ kgdb_and_then = &kgdb_data[++kgdb_and_then_count & INDEX_MASK];
+ spin_unlock(&ts_spin);
+ kgdb_local_irq_restore(flags);
+#ifdef CONFIG_PREEMPT
+
+#endif
+ return;
+}
+#endif
+typedef int gdb_debug_hook(int exceptionVector,
+ int signo, int err_code, struct pt_regs *linux_regs);
+gdb_debug_hook *linux_debug_hook = &kgdb_handle_exception; /* histerical reasons... */
+
+static int kgdb_need_breakpoint[NR_CPUS];
+
+void kgdb_schedule_breakpoint(void)
+{
+ kgdb_need_breakpoint[smp_processor_id()] = 1;
+}
+
+void kgdb_process_breakpoint(void)
+{
+ /*
+ * Handle a breakpoint queued from inside network driver code
+ * to avoid reentrancy issues
+ */
+ if (kgdb_need_breakpoint[smp_processor_id()]) {
+ kgdb_need_breakpoint[smp_processor_id()] = 0;
+ BREAKPOINT;
+ }
+}
+
--- diff/arch/i386/lib/kgdb_serial.c 1970-01-01 00:00:00.000000000 +0000
+++ source/arch/i386/lib/kgdb_serial.c 2004-03-16 11:27:47.899905768 +0000
@@ -0,0 +1,499 @@
+/*
+ * Serial interface GDB stub
+ *
+ * Written (hacked together) by David Grothe (dave@gcom.com)
+ * Modified to allow invokation early in boot see also
+ * kgdb.h for instructions by George Anzinger(george@mvista.com)
+ * Modified to handle debugging over ethernet by Robert Walsh
+ * <rjwalsh@durables.org> and wangdi <wangdi@clusterfs.com>, based on
+ * code by San Mehat.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/timer.h>
+#include <linux/interrupt.h>
+#include <linux/tty.h>
+#include <linux/tty_flip.h>
+#include <linux/serial.h>
+#include <linux/serial_reg.h>
+#include <linux/config.h>
+#include <linux/major.h>
+#include <linux/string.h>
+#include <linux/fcntl.h>
+#include <linux/ptrace.h>
+#include <linux/ioport.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/highmem.h>
+#include <asm/system.h>
+#include <asm/io.h>
+#include <asm/segment.h>
+#include <asm/bitops.h>
+#include <asm/system.h>
+#include <asm/kgdb_local.h>
+#ifdef CONFIG_KGDB_USER_CONSOLE
+extern void kgdb_console_finit(void);
+#endif
+#define PRNT_off
+#define TEST_EXISTANCE
+#ifdef PRNT
+#define dbprintk(s) printk s
+#else
+#define dbprintk(s)
+#endif
+#define TEST_INTERRUPT_off
+#ifdef TEST_INTERRUPT
+#define intprintk(s) printk s
+#else
+#define intprintk(s)
+#endif
+
+#define IRQ_T(info) ((info->flags & ASYNC_SHARE_IRQ) ? SA_SHIRQ : SA_INTERRUPT)
+
+#define GDB_BUF_SIZE 512 /* power of 2, please */
+
+static char gdb_buf[GDB_BUF_SIZE];
+static int gdb_buf_in_inx;
+static atomic_t gdb_buf_in_cnt;
+static int gdb_buf_out_inx;
+
+struct async_struct *gdb_async_info;
+static int gdb_async_irq;
+
+#define outb_px(a,b) outb_p(b,a)
+
+static void program_uart(struct async_struct *info);
+static void write_char(struct async_struct *info, int chr);
+/*
+ * Get a byte from the hardware data buffer and return it
+ */
+static int
+read_data_bfr(struct async_struct *info)
+{
+ char it = inb_p(info->port + UART_LSR);
+
+ if (it & UART_LSR_DR)
+ return (inb_p(info->port + UART_RX));
+ /*
+ * If we have a framing error assume somebody messed with
+ * our uart. Reprogram it and send '-' both ways...
+ */
+ if (it & 0xc) {
+ program_uart(info);
+ write_char(info, '-');
+ return ('-');
+ }
+ return (-1);
+
+} /* read_data_bfr */
+
+/*
+ * Get a char if available, return -1 if nothing available.
+ * Empty the receive buffer first, then look at the interface hardware.
+
+ * Locking here is a bit of a problem. We MUST not lock out communication
+ * if we are trying to talk to gdb about a kgdb entry. ON the other hand
+ * we can loose chars in the console pass thru if we don't lock. It is also
+ * possible that we could hold the lock or be waiting for it when kgdb
+ * NEEDS to talk. Since kgdb locks down the world, it does not need locks.
+ * We do, of course have possible issues with interrupting a uart operation,
+ * but we will just depend on the uart status to help keep that straight.
+
+ */
+static spinlock_t uart_interrupt_lock = SPIN_LOCK_UNLOCKED;
+#ifdef CONFIG_SMP
+extern spinlock_t kgdb_spinlock;
+#endif
+
+static int
+read_char(struct async_struct *info)
+{
+ int chr;
+ unsigned long flags;
+ local_irq_save(flags);
+#ifdef CONFIG_SMP
+ if (!spin_is_locked(&kgdb_spinlock)) {
+ spin_lock(&uart_interrupt_lock);
+ }
+#endif
+ if (atomic_read(&gdb_buf_in_cnt) != 0) { /* intr routine has q'd chars */
+ chr = gdb_buf[gdb_buf_out_inx++];
+ gdb_buf_out_inx &= (GDB_BUF_SIZE - 1);
+ atomic_dec(&gdb_buf_in_cnt);
+ } else {
+ chr = read_data_bfr(info);
+ }
+#ifdef CONFIG_SMP
+ if (!spin_is_locked(&kgdb_spinlock)) {
+ spin_unlock(&uart_interrupt_lock);
+ }
+#endif
+ local_irq_restore(flags);
+ return (chr);
+}
+
+/*
+ * Wait until the interface can accept a char, then write it.
+ */
+static void
+write_char(struct async_struct *info, int chr)
+{
+ while (!(inb_p(info->port + UART_LSR) & UART_LSR_THRE)) ;
+
+ outb_p(chr, info->port + UART_TX);
+
+} /* write_char */
+
+/*
+ * Mostly we don't need a spinlock, but since the console goes
+ * thru here with interrutps on, well, we need to catch those
+ * chars.
+ */
+/*
+ * This is the receiver interrupt routine for the GDB stub.
+ * It will receive a limited number of characters of input
+ * from the gdb host machine and save them up in a buffer.
+ *
+ * When the gdb stub routine tty_getDebugChar() is called it
+ * draws characters out of the buffer until it is empty and
+ * then reads directly from the serial port.
+ *
+ * We do not attempt to write chars from the interrupt routine
+ * since the stubs do all of that via tty_putDebugChar() which
+ * writes one byte after waiting for the interface to become
+ * ready.
+ *
+ * The debug stubs like to run with interrupts disabled since,
+ * after all, they run as a consequence of a breakpoint in
+ * the kernel.
+ *
+ * Perhaps someone who knows more about the tty driver than I
+ * care to learn can make this work for any low level serial
+ * driver.
+ */
+static irqreturn_t
+gdb_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+{
+ struct async_struct *info;
+ unsigned long flags;
+
+ info = gdb_async_info;
+ if (!info || !info->tty || irq != gdb_async_irq)
+ return IRQ_NONE;
+
+ local_irq_save(flags);
+ spin_lock(&uart_interrupt_lock);
+ do {
+ int chr = read_data_bfr(info);
+ intprintk(("Debug char on int: %x hex\n", chr));
+ if (chr < 0)
+ continue;
+
+ if (chr == 3) { /* Ctrl-C means remote interrupt */
+ BREAKPOINT;
+ continue;
+ }
+
+ if (atomic_read(&gdb_buf_in_cnt) >= GDB_BUF_SIZE) {
+ /* buffer overflow tosses early char */
+ read_char(info);
+ }
+ gdb_buf[gdb_buf_in_inx++] = chr;
+ gdb_buf_in_inx &= (GDB_BUF_SIZE - 1);
+ } while (inb_p(info->port + UART_IIR) & UART_IIR_RDI);
+ spin_unlock(&uart_interrupt_lock);
+ local_irq_restore(flags);
+ return IRQ_HANDLED;
+} /* gdb_interrupt */
+
+/*
+ * Just a NULL routine for testing.
+ */
+void
+gdb_null(void)
+{
+} /* gdb_null */
+
+/* These structure are filled in with values defined in asm/kgdb_local.h
+ */
+static struct serial_state state = SB_STATE;
+static struct async_struct local_info = SB_INFO;
+static int ok_to_enable_ints = 0;
+static void kgdb_enable_ints_now(void);
+
+extern char *kgdb_version;
+/*
+ * Hook an IRQ for KGDB.
+ *
+ * This routine is called from tty_putDebugChar, below.
+ */
+static int ints_disabled = 1;
+int
+gdb_hook_interrupt(struct async_struct *info, int verb)
+{
+ struct serial_state *state = info->state;
+ unsigned long flags;
+ int port;
+#ifdef TEST_EXISTANCE
+ int scratch, scratch2;
+#endif
+
+ /* The above fails if memory managment is not set up yet.
+ * Rather than fail the set up, just keep track of the fact
+ * and pick up the interrupt thing later.
+ */
+ gdb_async_info = info;
+ port = gdb_async_info->port;
+ gdb_async_irq = state->irq;
+ if (verb) {
+ printk("kgdb %s : port =%x, IRQ=%d, divisor =%d\n",
+ kgdb_version,
+ port,
+ gdb_async_irq, gdb_async_info->state->custom_divisor);
+ }
+ local_irq_save(flags);
+#ifdef TEST_EXISTANCE
+ /* Existance test */
+ /* Should not need all this, but just in case.... */
+
+ scratch = inb_p(port + UART_IER);
+ outb_px(port + UART_IER, 0);
+ outb_px(0xff, 0x080);
+ scratch2 = inb_p(port + UART_IER);
+ outb_px(port + UART_IER, scratch);
+ if (scratch2) {
+ printk
+ ("gdb_hook_interrupt: Could not clear IER, not a UART!\n");
+ local_irq_restore(flags);
+ return 1; /* We failed; there's nothing here */
+ }
+ scratch2 = inb_p(port + UART_LCR);
+ outb_px(port + UART_LCR, 0xBF); /* set up for StarTech test */
+ outb_px(port + UART_EFR, 0); /* EFR is the same as FCR */
+ outb_px(port + UART_LCR, 0);
+ outb_px(port + UART_FCR, UART_FCR_ENABLE_FIFO);
+ scratch = inb_p(port + UART_IIR) >> 6;
+ if (scratch == 1) {
+ printk("gdb_hook_interrupt: Undefined UART type!"
+ " Not a UART! \n");
+ local_irq_restore(flags);
+ return 1;
+ } else {
+ dbprintk(("gdb_hook_interrupt: UART type "
+ "is %d where 0=16450, 2=16550 3=16550A\n", scratch));
+ }
+ scratch = inb_p(port + UART_MCR);
+ outb_px(port + UART_MCR, UART_MCR_LOOP | scratch);
+ outb_px(port + UART_MCR, UART_MCR_LOOP | 0x0A);
+ scratch2 = inb_p(port + UART_MSR) & 0xF0;
+ outb_px(port + UART_MCR, scratch);
+ if (scratch2 != 0x90) {
+ printk("gdb_hook_interrupt: "
+ "Loop back test failed! Not a UART!\n");
+ local_irq_restore(flags);
+ return scratch2 + 1000; /* force 0 to fail */
+ }
+#endif /* test existance */
+ program_uart(info);
+ local_irq_restore(flags);
+
+ return (0);
+
+} /* gdb_hook_interrupt */
+
+static void
+program_uart(struct async_struct *info)
+{
+ int port = info->port;
+
+ (void) inb_p(port + UART_RX);
+ outb_px(port + UART_IER, 0);
+
+ (void) inb_p(port + UART_RX); /* serial driver comments say */
+ (void) inb_p(port + UART_IIR); /* this clears the interrupt regs */
+ (void) inb_p(port + UART_MSR);
+ outb_px(port + UART_LCR, UART_LCR_WLEN8 | UART_LCR_DLAB);
+ outb_px(port + UART_DLL, info->state->custom_divisor & 0xff); /* LS */
+ outb_px(port + UART_DLM, info->state->custom_divisor >> 8); /* MS */
+ outb_px(port + UART_MCR, info->MCR);
+
+ outb_px(port + UART_FCR, UART_FCR_ENABLE_FIFO | UART_FCR_TRIGGER_1 | UART_FCR_CLEAR_XMIT | UART_FCR_CLEAR_RCVR); /* set fcr */
+ outb_px(port + UART_LCR, UART_LCR_WLEN8); /* reset DLAB */
+ outb_px(port + UART_FCR, UART_FCR_ENABLE_FIFO | UART_FCR_TRIGGER_1); /* set fcr */
+ if (!ints_disabled) {
+ intprintk(("KGDB: Sending %d to port %x offset %d\n",
+ gdb_async_info->IER,
+ (int) gdb_async_info->port, UART_IER));
+ outb_px(gdb_async_info->port + UART_IER, gdb_async_info->IER);
+ }
+ return;
+}
+
+/*
+ * tty_getDebugChar
+ *
+ * This is a GDB stub routine. It waits for a character from the
+ * serial interface and then returns it. If there is no serial
+ * interface connection then it returns a bogus value which will
+ * almost certainly cause the system to hang. In the
+ */
+int kgdb_in_isr = 0;
+int kgdb_in_lsr = 0;
+extern spinlock_t kgdb_spinlock;
+
+/* Caller takes needed protections */
+
+int
+tty_getDebugChar(void)
+{
+ volatile int chr, dum, time, end_time;
+
+ dbprintk(("tty_getDebugChar(port %x): ", gdb_async_info->port));
+
+ if (gdb_async_info == NULL) {
+ gdb_hook_interrupt(&local_info, 0);
+ }
+ /*
+ * This trick says if we wait a very long time and get
+ * no char, return the -1 and let the upper level deal
+ * with it.
+ */
+ rdtsc(dum, time);
+ end_time = time + 2;
+ while (((chr = read_char(gdb_async_info)) == -1) &&
+ (end_time - time) > 0) {
+ rdtsc(dum, time);
+ };
+ /*
+ * This covers our butts if some other code messes with
+ * our uart, hay, it happens :o)
+ */
+ if (chr == -1)
+ program_uart(gdb_async_info);
+
+ dbprintk(("%c\n", chr > ' ' && chr < 0x7F ? chr : ' '));
+ return (chr);
+
+} /* tty_getDebugChar */
+
+static int count = 3;
+static spinlock_t one_at_atime = SPIN_LOCK_UNLOCKED;
+
+static int __init
+kgdb_enable_ints(void)
+{
+ if (kgdboe) {
+ return 0;
+ }
+ if (gdb_async_info == NULL) {
+ gdb_hook_interrupt(&local_info, 1);
+ }
+ ok_to_enable_ints = 1;
+ kgdb_enable_ints_now();
+#ifdef CONFIG_KGDB_USER_CONSOLE
+ kgdb_console_finit();
+#endif
+ return 0;
+}
+
+#ifdef CONFIG_SERIAL_8250
+void shutdown_for_kgdb(struct async_struct *gdb_async_info);
+#endif
+
+#ifdef CONFIG_DISCONTIGMEM
+static inline int kgdb_mem_init_done(void)
+{
+ return highmem_start_page != NULL;
+}
+#else
+static inline int kgdb_mem_init_done(void)
+{
+ return max_mapnr != 0;
+}
+#endif
+
+static void
+kgdb_enable_ints_now(void)
+{
+ if (!spin_trylock(&one_at_atime))
+ return;
+ if (!ints_disabled)
+ goto exit;
+ if (kgdb_mem_init_done() &&
+ ints_disabled) { /* don't try till mem init */
+#ifdef CONFIG_SERIAL_8250
+ /*
+ * The ifdef here allows the system to be configured
+ * without the serial driver.
+ * Don't make it a module, however, it will steal the port
+ */
+ shutdown_for_kgdb(gdb_async_info);
+#endif
+ ints_disabled = request_irq(gdb_async_info->state->irq,
+ gdb_interrupt,
+ IRQ_T(gdb_async_info),
+ "KGDB-stub", NULL);
+ intprintk(("KGDB: request_irq returned %d\n", ints_disabled));
+ }
+ if (!ints_disabled) {
+ intprintk(("KGDB: Sending %d to port %x offset %d\n",
+ gdb_async_info->IER,
+ (int) gdb_async_info->port, UART_IER));
+ outb_px(gdb_async_info->port + UART_IER, gdb_async_info->IER);
+ }
+ exit:
+ spin_unlock(&one_at_atime);
+}
+
+/*
+ * tty_putDebugChar
+ *
+ * This is a GDB stub routine. It waits until the interface is ready
+ * to transmit a char and then sends it. If there is no serial
+ * interface connection then it simply returns to its caller, having
+ * pretended to send the char. Caller takes needed protections.
+ */
+void
+tty_putDebugChar(int chr)
+{
+ dbprintk(("tty_putDebugChar(port %x): chr=%02x '%c', ints_on=%d\n",
+ gdb_async_info->port,
+ chr,
+ chr > ' ' && chr < 0x7F ? chr : ' ', ints_disabled ? 0 : 1));
+
+ if (gdb_async_info == NULL) {
+ gdb_hook_interrupt(&local_info, 0);
+ }
+
+ write_char(gdb_async_info, chr); /* this routine will wait */
+ count = (chr == '#') ? 0 : count + 1;
+ if ((count == 2)) { /* try to enable after */
+ if (ints_disabled & ok_to_enable_ints)
+ kgdb_enable_ints_now(); /* try to enable after */
+
+ /* We do this a lot because, well we really want to get these
+ * interrupts. The serial driver will clear these bits when it
+ * initializes the chip. Every thing else it does is ok,
+ * but this.
+ */
+ if (!ints_disabled) {
+ outb_px(gdb_async_info->port + UART_IER,
+ gdb_async_info->IER);
+ }
+ }
+
+} /* tty_putDebugChar */
+
+/*
+ * This does nothing for the serial port, since it doesn't buffer.
+ */
+
+void tty_flushDebugChar(void)
+{
+}
+
+module_init(kgdb_enable_ints);
--- diff/arch/ppc64/kernel/dma.c 1970-01-01 00:00:00.000000000 +0000
+++ source/arch/ppc64/kernel/dma.c 2004-03-16 11:27:47.915903336 +0000
@@ -0,0 +1,184 @@
+/*
+ * Copyright (C) 2004 IBM Corporation
+ *
+ * Implements the generic device dma API for ppc64. Handles
+ * the pci and vio busses
+ */
+
+#include <linux/config.h>
+#include <linux/device.h>
+#include <linux/dma-mapping.h>
+/* Include the busses we support */
+#include <linux/pci.h>
+#ifdef CONFIG_PPC_PSERIES
+#include <asm/vio.h>
+#endif
+#include <asm/scatterlist.h>
+#include <asm/bug.h>
+
+int dma_supported(struct device *dev, u64 mask)
+{
+ if (dev->bus == &pci_bus_type)
+ return pci_dma_supported(to_pci_dev(dev), mask);
+#ifdef CONFIG_PPC_PSERIES
+ if (dev->bus == &vio_bus_type)
+ return vio_dma_supported(to_vio_dev(dev), mask);
+#endif
+ BUG();
+ return 0;
+}
+EXPORT_SYMBOL(dma_supported);
+
+int dma_set_mask(struct device *dev, u64 dma_mask)
+{
+ if (dev->bus == &pci_bus_type)
+ return pci_set_dma_mask(to_pci_dev(dev), dma_mask);
+#ifdef CONFIG_PPC_PSERIES
+ if (dev->bus == &vio_bus_type)
+ return vio_set_dma_mask(to_vio_dev(dev), dma_mask);
+#endif
+ BUG();
+ return 0;
+}
+EXPORT_SYMBOL(dma_set_mask);
+
+void *dma_alloc_coherent(struct device *dev, size_t size,
+ dma_addr_t *dma_handle, int flag)
+{
+ if (dev->bus == &pci_bus_type)
+ return pci_alloc_consistent(to_pci_dev(dev), size, dma_handle);
+#ifdef CONFIG_PPC_PSERIES
+ if (dev->bus == &vio_bus_type)
+ return vio_alloc_consistent(to_vio_dev(dev), size, dma_handle);
+#endif
+ BUG();
+ return 0;
+}
+EXPORT_SYMBOL(dma_alloc_coherent);
+
+void dma_free_coherent(struct device *dev, size_t size, void *cpu_addr,
+ dma_addr_t dma_handle)
+{
+ if (dev->bus == &pci_bus_type)
+ pci_free_consistent(to_pci_dev(dev), size, cpu_addr, dma_handle);
+#ifdef CONFIG_PPC_PSERIES
+ else if (dev->bus == &vio_bus_type)
+ vio_free_consistent(to_vio_dev(dev), size, cpu_addr, dma_handle);
+#endif
+ else
+ BUG();
+}
+EXPORT_SYMBOL(dma_free_coherent);
+
+dma_addr_t dma_map_single(struct device *dev, void *cpu_addr, size_t size,
+ enum dma_data_direction direction)
+{
+ if (dev->bus == &pci_bus_type)
+ return pci_map_single(to_pci_dev(dev), cpu_addr, size, (int)direction);
+#ifdef CONFIG_PPC_PSERIES
+ if (dev->bus == &vio_bus_type)
+ return vio_map_single(to_vio_dev(dev), cpu_addr, size, (int)direction);
+#endif
+ BUG();
+ return (dma_addr_t)0;
+}
+EXPORT_SYMBOL(dma_map_single);
+
+void dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
+ enum dma_data_direction direction)
+{
+ if (dev->bus == &pci_bus_type)
+ pci_unmap_single(to_pci_dev(dev), dma_addr, size, (int)direction);
+#ifdef CONFIG_PPC_PSERIES
+ else if (dev->bus == &vio_bus_type)
+ vio_unmap_single(to_vio_dev(dev), dma_addr, size, (int)direction);
+#endif
+ else
+ BUG();
+}
+EXPORT_SYMBOL(dma_unmap_single);
+
+dma_addr_t dma_map_page(struct device *dev, struct page *page,
+ unsigned long offset, size_t size,
+ enum dma_data_direction direction)
+{
+ if (dev->bus == &pci_bus_type)
+ return pci_map_page(to_pci_dev(dev), page, offset, size, (int)direction);
+#ifdef CONFIG_PPC_PSERIES
+ if (dev->bus == &vio_bus_type)
+ return vio_map_page(to_vio_dev(dev), page, offset, size, (int)direction);
+#endif
+ BUG();
+ return (dma_addr_t)0;
+}
+EXPORT_SYMBOL(dma_map_page);
+
+void dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
+ enum dma_data_direction direction)
+{
+ if (dev->bus == &pci_bus_type)
+ pci_unmap_page(to_pci_dev(dev), dma_address, size, (int)direction);
+#ifdef CONFIG_PPC_PSERIES
+ else if (dev->bus == &vio_bus_type)
+ vio_unmap_page(to_vio_dev(dev), dma_address, size, (int)direction);
+#endif
+ else
+ BUG();
+}
+EXPORT_SYMBOL(dma_unmap_page);
+
+int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
+ enum dma_data_direction direction)
+{
+ if (dev->bus == &pci_bus_type)
+ return pci_map_sg(to_pci_dev(dev), sg, nents, (int)direction);
+#ifdef CONFIG_PPC_PSERIES
+ if (dev->bus == &vio_bus_type)
+ return vio_map_sg(to_vio_dev(dev), sg, nents, (int)direction);
+#endif
+ BUG();
+ return 0;
+}
+EXPORT_SYMBOL(dma_map_sg);
+
+void dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nhwentries,
+ enum dma_data_direction direction)
+{
+ if (dev->bus == &pci_bus_type)
+ pci_unmap_sg(to_pci_dev(dev), sg, nhwentries, (int)direction);
+#ifdef CONFIG_PPC_PSERIES
+ else if (dev->bus == &vio_bus_type)
+ vio_unmap_sg(to_vio_dev(dev), sg, nhwentries, (int)direction);
+#endif
+ else
+ BUG();
+}
+EXPORT_SYMBOL(dma_unmap_sg);
+
+void dma_sync_single(struct device *dev, dma_addr_t dma_handle, size_t size,
+ enum dma_data_direction direction)
+{
+ if (dev->bus == &pci_bus_type)
+ pci_dma_sync_single(to_pci_dev(dev), dma_handle, size, (int)direction);
+#ifdef CONFIG_PPC_PSERIES
+ else if (dev->bus == &vio_bus_type)
+ vio_dma_sync_single(to_vio_dev(dev), dma_handle, size, (int)direction);
+#endif
+ else
+ BUG();
+}
+EXPORT_SYMBOL(dma_sync_single);
+
+void dma_sync_sg(struct device *dev, struct scatterlist *sg, int nelems,
+ enum dma_data_direction direction)
+{
+ if (dev->bus == &pci_bus_type)
+ pci_dma_sync_sg(to_pci_dev(dev), sg, nelems, (int)direction);
+#ifdef CONFIG_PPC_PSERIES
+ else if (dev->bus == &vio_bus_type)
+ vio_dma_sync_sg(to_vio_dev(dev), sg, nelems, (int)direction);
+#endif
+ else
+ BUG();
+}
+EXPORT_SYMBOL(dma_sync_sg);
--- diff/arch/x86_64/Kconfig.kgdb 1970-01-01 00:00:00.000000000 +0000
+++ source/arch/x86_64/Kconfig.kgdb 2004-03-16 11:27:47.962896192 +0000
@@ -0,0 +1,176 @@
+config KGDB
+ bool "Include kgdb kernel debugger"
+ depends on DEBUG_KERNEL
+ select DEBUG_INFO
+ help
+ If you say Y here, the system will be compiled with the debug
+ option (-g) and a debugging stub will be included in the
+ kernel. This stub communicates with gdb on another (host)
+ computer via a serial port. The host computer should have
+ access to the kernel binary file (vmlinux) and a serial port
+ that is connected to the target machine. Gdb can be made to
+ configure the serial port or you can use stty and setserial to
+ do this. See the 'target' command in gdb. This option also
+ configures in the ability to request a breakpoint early in the
+ boot process. To request the breakpoint just include 'kgdb'
+ as a boot option when booting the target machine. The system
+ will then break as soon as it looks at the boot options. This
+ option also installs a breakpoint in panic and sends any
+ kernel faults to the debugger. For more information see the
+ Documentation/i386/kgdb.txt file.
+
+choice
+ depends on KGDB
+ prompt "Debug serial port BAUD"
+ default KGDB_115200BAUD
+ help
+ Gdb and the kernel stub need to agree on the baud rate to be
+ used. Some systems (x86 family at this writing) allow this to
+ be configured.
+
+config KGDB_9600BAUD
+ bool "9600"
+
+config KGDB_19200BAUD
+ bool "19200"
+
+config KGDB_38400BAUD
+ bool "38400"
+
+config KGDB_57600BAUD
+ bool "57600"
+
+config KGDB_115200BAUD
+ bool "115200"
+endchoice
+
+config KGDB_PORT
+ hex "hex I/O port address of the debug serial port"
+ depends on KGDB
+ default 3f8
+ help
+ Some systems (x86 family at this writing) allow the port
+ address to be configured. The number entered is assumed to be
+ hex, don't put 0x in front of it. The standard address are:
+ COM1 3f8 , irq 4 and COM2 2f8 irq 3. Setserial /dev/ttySx
+ will tell you what you have. It is good to test the serial
+ connection with a live system before trying to debug.
+
+config KGDB_IRQ
+ int "IRQ of the debug serial port"
+ depends on KGDB
+ default 4
+ help
+ This is the irq for the debug port. If everything is working
+ correctly and the kernel has interrupts on a control C to the
+ port should cause a break into the kernel debug stub.
+
+config DEBUG_INFO
+ bool
+ depends on KGDB
+ default y
+
+config KGDB_MORE
+ bool "Add any additional compile options"
+ depends on KGDB
+ default n
+ help
+ Saying yes here turns on the ability to enter additional
+ compile options.
+
+
+config KGDB_OPTIONS
+ depends on KGDB_MORE
+ string "Additional compile arguments"
+ default "-O1"
+ help
+ This option allows you enter additional compile options for
+ the whole kernel compile. Each platform will have a default
+ that seems right for it. For example on PPC "-ggdb -O1", and
+ for i386 "-O1". Note that by configuring KGDB "-g" is already
+ turned on. In addition, on i386 platforms
+ "-fomit-frame-pointer" is deleted from the standard compile
+ options.
+
+config NO_KGDB_CPUS
+ int "Number of CPUs"
+ depends on KGDB && SMP
+ default NR_CPUS
+ help
+
+ This option sets the number of cpus for kgdb ONLY. It is used
+ to prune some internal structures so they look "nice" when
+ displayed with gdb. This is to overcome possibly larger
+ numbers that may have been entered above. Enter the real
+ number to get nice clean kgdb_info displays.
+
+config KGDB_TS
+ bool "Enable kgdb time stamp macros?"
+ depends on KGDB
+ default n
+ help
+ Kgdb event macros allow you to instrument your code with calls
+ to the kgdb event recording function. The event log may be
+ examined with gdb at a break point. Turning on this
+ capability also allows you to choose how many events to
+ keep. Kgdb always keeps the lastest events.
+
+choice
+ depends on KGDB_TS
+ prompt "Max number of time stamps to save?"
+ default KGDB_TS_128
+
+config KGDB_TS_64
+ bool "64"
+
+config KGDB_TS_128
+ bool "128"
+
+config KGDB_TS_256
+ bool "256"
+
+config KGDB_TS_512
+ bool "512"
+
+config KGDB_TS_1024
+ bool "1024"
+
+endchoice
+
+config STACK_OVERFLOW_TEST
+ bool "Turn on kernel stack overflow testing?"
+ depends on KGDB
+ default n
+ help
+ This option enables code in the front line interrupt handlers
+ to check for kernel stack overflow on interrupts and system
+ calls. This is part of the kgdb code on x86 systems.
+
+config KGDB_CONSOLE
+ bool "Enable serial console thru kgdb port"
+ depends on KGDB
+ default n
+ help
+ This option enables the command line "console=kgdb" option.
+ When the system is booted with this option in the command line
+ all kernel printk output is sent to gdb (as well as to other
+ consoles). For this to work gdb must be connected. For this
+ reason, this command line option will generate a breakpoint if
+ gdb has not yet connected. After the gdb continue command is
+ given all pent up console output will be printed by gdb on the
+ host machine. Neither this option, nor KGDB require the
+ serial driver to be configured.
+
+config KGDB_SYSRQ
+ bool "Turn on SysRq 'G' command to do a break?"
+ depends on KGDB
+ default y
+ help
+ This option includes an option in the SysRq code that allows
+ you to enter SysRq G which generates a breakpoint to the KGDB
+ stub. This will work if the keyboard is alive and can
+ interrupt the system. Because of constraints on when the
+ serial port interrupt can be enabled, this code may allow you
+ to interrupt the system before the serial port control C is
+ available. Just say yes here.
+
--- diff/arch/x86_64/kernel/kgdb_stub.c 1970-01-01 00:00:00.000000000 +0000
+++ source/arch/x86_64/kernel/kgdb_stub.c 2004-03-16 11:27:47.969895128 +0000
@@ -0,0 +1,2595 @@
+/*
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2, or (at your option) any
+ * later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ */
+
+/*
+ * Copyright (c) 2000 VERITAS Software Corporation.
+ *
+ */
+/****************************************************************************
+ * Header: remcom.c,v 1.34 91/03/09 12:29:49 glenne Exp $
+ *
+ * Module name: remcom.c $
+ * Revision: 1.34 $
+ * Date: 91/03/09 12:29:49 $
+ * Contributor: Lake Stevens Instrument Division$
+ *
+ * Description: low level support for gdb debugger. $
+ *
+ * Considerations: only works on target hardware $
+ *
+ * Written by: Glenn Engel $
+ * Updated by: David Grothe <dave@gcom.com>
+ * Updated by: Robert Walsh <rjwalsh@durables.org>
+ * Updated by: wangdi <wangdi@clusterfs.com>
+ * ModuleState: Experimental $
+ *
+ * NOTES: See Below $
+ *
+ * Modified for 386 by Jim Kingdon, Cygnus Support.
+ * Compatibility with 2.1.xx kernel by David Grothe <dave@gcom.com>
+ *
+ * Changes to allow auto initilization. All that is needed is that it
+ * be linked with the kernel and a break point (int 3) be executed.
+ * The header file <asm/kgdb.h> defines BREAKPOINT to allow one to do
+ * this. It should also be possible, once the interrupt system is up, to
+ * call putDebugChar("+"). Once this is done, the remote debugger should
+ * get our attention by sending a ^C in a packet. George Anzinger
+ * <george@mvista.com>
+ * Integrated into 2.2.5 kernel by Tigran Aivazian <tigran@sco.com>
+ * Added thread support, support for multiple processors,
+ * support for ia-32(x86) hardware debugging.
+ * Amit S. Kale ( akale@veritas.com )
+ *
+ * Modified to support debugging over ethernet by Robert Walsh
+ * <rjwalsh@durables.org> and wangdi <wangdi@clusterfs.com>, based on
+ * code by San Mehat.
+ *
+ * X86_64 changes from Andi Kleen's patch merged by Jim Houston
+ * (jim.houston@ccur.com). If it works thank Andi if its broken
+ * blame me.
+ *
+ * To enable debugger support, two things need to happen. One, a
+ * call to set_debug_traps() is necessary in order to allow any breakpoints
+ * or error conditions to be properly intercepted and reported to gdb.
+ * Two, a breakpoint needs to be generated to begin communication. This
+ * is most easily accomplished by a call to breakpoint(). Breakpoint()
+ * simulates a breakpoint by executing an int 3.
+ *
+ *************
+ *
+ * The following gdb commands are supported:
+ *
+ * command function Return value
+ *
+ * g return the value of the CPU registers hex data or ENN
+ * G set the value of the CPU registers OK or ENN
+ *
+ * mAA..AA,LLLL Read LLLL bytes at address AA..AA hex data or ENN
+ * MAA..AA,LLLL: Write LLLL bytes at address AA.AA OK or ENN
+ *
+ * c Resume at current address SNN ( signal NN)
+ * cAA..AA Continue at address AA..AA SNN
+ *
+ * s Step one instruction SNN
+ * sAA..AA Step one instruction from AA..AA SNN
+ *
+ * k kill
+ *
+ * ? What was the last sigval ? SNN (signal NN)
+ *
+ * All commands and responses are sent with a packet which includes a
+ * checksum. A packet consists of
+ *
+ * $<packet info>#<checksum>.
+ *
+ * where
+ * <packet info> :: <characters representing the command or response>
+ * <checksum> :: < two hex digits computed as modulo 256 sum of <packetinfo>>
+ *
+ * When a packet is received, it is first acknowledged with either '+' or '-'.
+ * '+' indicates a successful transfer. '-' indicates a failed transfer.
+ *
+ * Example:
+ *
+ * Host: Reply:
+ * $m0,10#2a +$00010203040506070809101112131415#42
+ *
+ ****************************************************************************/
+#define KGDB_VERSION "<20030915.1651.33>"
+#include <linux/config.h>
+#include <linux/types.h>
+#include <asm/string.h> /* for strcpy */
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <asm/system.h>
+#include <asm/ptrace.h> /* for linux pt_regs struct */
+#include <asm/kgdb_local.h>
+#include <linux/list.h>
+#include <asm/atomic.h>
+#include <asm/processor.h>
+#include <linux/irq.h>
+#include <asm/desc.h>
+#include <linux/inet.h>
+#include <linux/netpoll.h>
+#include <linux/cpumask.h>
+#include <linux/bitops.h>
+#include <linux/notifier.h>
+#include <asm/kdebug.h>
+#include <asm/uaccess.h>
+#include <linux/ptrace.h>
+
+#define Dearly_printk(x...)
+int kgdb_enabled = 0;
+
+/************************************************************************
+ *
+ * external low-level support routines
+ */
+typedef void (*Function) (void); /* pointer to a function */
+
+/* Thread reference */
+typedef unsigned char threadref[8];
+
+extern int tty_putDebugChar(int); /* write a single character */
+extern int tty_getDebugChar(void); /* read and return a single char */
+extern void tty_flushDebugChar(void); /* flush pending characters */
+extern int eth_putDebugChar(int); /* write a single character */
+extern int eth_getDebugChar(void); /* read and return a single char */
+extern void eth_flushDebugChar(void); /* flush pending characters */
+
+/************************************************************************/
+/* BUFMAX defines the maximum number of characters in inbound/outbound buffers*/
+/* at least NUMREGBYTES*2 are needed for register packets */
+/* Longer buffer is needed to list all threads */
+#define BUFMAX 400
+
+char *kgdb_version = KGDB_VERSION;
+
+/* debug > 0 prints ill-formed commands in valid packets & checksum errors */
+int debug_regs = 0; /* set to non-zero to print registers */
+
+/* filled in by an external module */
+char *gdb_module_offsets;
+
+static const char hexchars[] = "0123456789abcdef";
+
+/* Number of bytes of registers. */
+#define NUMREGBYTES (NUMREGS * sizeof(unsigned long))
+/*
+ * Note that this register image is in a different order than
+ * the register image that Linux produces at interrupt time.
+ *
+ * Linux's register image is defined by struct pt_regs in ptrace.h.
+ * Just why GDB uses a different order is a historical mystery.
+ *
+ * Could add XMM and segment registers here.
+ */
+enum regnames {_RAX,
+ _RBX,
+ _RCX,
+ _RDX,
+ _RSI,
+ _RDI,
+ _RBP,
+ _RSP,
+ _R8,
+ _R9,
+ _R10,
+ _R11,
+ _R12,
+ _R13,
+ _R14,
+ _R15,
+ _PC,
+ _PS,
+ NUMREGS };
+
+
+/*************************** ASSEMBLY CODE MACROS *************************/
+/*
+ * Put the error code here just in case the user cares.
+ * Likewise, the vector number here (since GDB only gets the signal
+ * number through the usual means, and that's not very specific).
+ * The called_from is the return address so he can tell how we entered kgdb.
+ * This will allow him to seperate out the various possible entries.
+ */
+#define REMOTE_DEBUG 0 /* set != to turn on printing (also available in info) */
+
+#define PID_MAX PID_MAX_DEFAULT
+
+#ifdef CONFIG_SMP
+void smp_send_nmi_allbutself(void);
+#define IF_SMP(x) x
+#undef MAX_NO_CPUS
+#ifndef CONFIG_NO_KGDB_CPUS
+#define CONFIG_NO_KGDB_CPUS 2
+#endif
+#if CONFIG_NO_KGDB_CPUS > NR_CPUS
+#define MAX_NO_CPUS NR_CPUS
+#else
+#define MAX_NO_CPUS CONFIG_NO_KGDB_CPUS
+#endif
+#define hold_init hold_on_sstep: 1,
+#define MAX_CPU_MASK (unsigned long)((1LL << MAX_NO_CPUS) - 1LL)
+#define NUM_CPUS num_online_cpus()
+#else
+#define IF_SMP(x)
+#define hold_init
+#undef MAX_NO_CPUS
+#define MAX_NO_CPUS 1
+#define NUM_CPUS 1
+#endif
+#define NOCPU (struct task_struct *)0xbad1fbad
+/* *INDENT-OFF* */
+struct kgdb_info {
+ int used_malloc;
+ void *called_from;
+ long long entry_tsc;
+ int errcode;
+ int vector;
+ int print_debug_info;
+#ifdef CONFIG_SMP
+ int hold_on_sstep;
+ struct {
+ volatile struct task_struct *task;
+ int pid;
+ int hold;
+ struct pt_regs *regs;
+ } cpus_waiting[MAX_NO_CPUS];
+#endif
+} kgdb_info = {hold_init print_debug_info:REMOTE_DEBUG, vector:-1};
+
+/* *INDENT-ON* */
+
+#define used_m kgdb_info.used_malloc
+/*
+ * This is little area we set aside to contain the stack we
+ * need to build to allow gdb to call functions. We use one
+ * per cpu to avoid locking issues. We will do all this work
+ * with interrupts off so that should take care of the protection
+ * issues.
+ */
+#define LOOKASIDE_SIZE 200 /* should be more than enough */
+#define MALLOC_MAX 200 /* Max malloc size */
+struct {
+ unsigned long rsp;
+ unsigned long array[LOOKASIDE_SIZE];
+} fn_call_lookaside[MAX_NO_CPUS];
+
+static int trap_cpu;
+static unsigned long OLD_esp;
+
+#define END_OF_LOOKASIDE &fn_call_lookaside[trap_cpu].array[LOOKASIDE_SIZE]
+#define IF_BIT 0x200
+#define TF_BIT 0x100
+
+#define MALLOC_ROUND 8-1
+
+static char malloc_array[MALLOC_MAX];
+IF_SMP(static void to_gdb(const char *mess));
+void *
+malloc(int size)
+{
+
+ if (size <= (MALLOC_MAX - used_m)) {
+ int old_used = used_m;
+ used_m += ((size + MALLOC_ROUND) & (~MALLOC_ROUND));
+ return &malloc_array[old_used];
+ } else {
+ return NULL;
+ }
+}
+
+/*
+ * I/O dispatch functions...
+ * Based upon kgdboe, either call the ethernet
+ * handler or the serial one..
+ */
+void
+putDebugChar(int c)
+{
+ if (!kgdboe) {
+ tty_putDebugChar(c);
+ } else {
+ eth_putDebugChar(c);
+ }
+}
+
+int
+getDebugChar(void)
+{
+ if (!kgdboe) {
+ return tty_getDebugChar();
+ } else {
+ return eth_getDebugChar();
+ }
+}
+
+void
+flushDebugChar(void)
+{
+ if (!kgdboe) {
+ tty_flushDebugChar();
+ } else {
+ eth_flushDebugChar();
+ }
+}
+
+/*
+ * Gdb calls functions by pushing agruments, including a return address
+ * on the stack and the adjusting EIP to point to the function. The
+ * whole assumption in GDB is that we are on a different stack than the
+ * one the "user" i.e. code that hit the break point, is on. This, of
+ * course is not true in the kernel. Thus various dodges are needed to
+ * do the call without directly messing with EIP (which we can not change
+ * as it is just a location and not a register. To adjust it would then
+ * require that we move every thing below EIP up or down as needed. This
+ * will not work as we may well have stack relative pointer on the stack
+ * (such as the pointer to regs, for example).
+
+ * So here is what we do:
+ * We detect gdb attempting to store into the stack area and instead, store
+ * into the fn_call_lookaside.array at the same relative location as if it
+ * were the area ESP pointed at. We also trap ESP modifications
+ * and uses these to adjust fn_call_lookaside.esp. On entry
+ * fn_call_lookaside.esp will be set to point at the last entry in
+ * fn_call_lookaside.array. This allows us to check if it has changed, and
+ * if so, on exit, we add the registers we will use to do the move and a
+ * trap/ interrupt return exit sequence. We then adjust the eflags in the
+ * regs array (remember we now have a copy in the fn_call_lookaside.array) to
+ * kill the interrupt bit, AND we change EIP to point at our set up stub.
+ * As part of the register set up we preset the registers to point at the
+ * begining and end of the fn_call_lookaside.array, so all the stub needs to
+ * do is move words from the array to the stack until ESP= the desired value
+ * then do the rti. This will then transfer to the desired function with
+ * all the correct registers. Nifty huh?
+ */
+extern asmlinkage void fn_call_stub(void);
+extern asmlinkage void fn_rtn_stub(void);
+/* *INDENT-OFF* */
+__asm__("fn_rtn_stub:\n\t"
+ "movq %rax,%rsp\n\t"
+ "fn_call_stub:\n\t"
+ "1:\n\t"
+ "addq $-8,%rbx\n\t"
+ "movq (%rbx), %rax\n\t"
+ "pushq %rax\n\t"
+ "cmpq %rsp,%rcx\n\t"
+ "jne 1b\n\t"
+ "popq %rax\n\t"
+ "popq %rbx\n\t"
+ "popq %rcx\n\t"
+ "iret \n\t");
+/* *INDENT-ON* */
+#define gdb_i386vector kgdb_info.vector
+#define gdb_i386errcode kgdb_info.errcode
+#define waiting_cpus kgdb_info.cpus_waiting
+#define remote_debug kgdb_info.print_debug_info
+#define hold_cpu(cpu) kgdb_info.cpus_waiting[cpu].hold
+/* gdb locks */
+
+#ifdef CONFIG_SMP
+static int in_kgdb_called;
+static spinlock_t waitlocks[MAX_NO_CPUS] =
+ {[0 ... MAX_NO_CPUS - 1] = SPIN_LOCK_UNLOCKED };
+/*
+ * The following array has the thread pointer of each of the "other"
+ * cpus. We make it global so it can be seen by gdb.
+ */
+volatile int in_kgdb_entry_log[MAX_NO_CPUS];
+volatile struct pt_regs *in_kgdb_here_log[MAX_NO_CPUS];
+/*
+static spinlock_t continuelocks[MAX_NO_CPUS];
+*/
+spinlock_t kgdb_spinlock = SPIN_LOCK_UNLOCKED;
+/* waiters on our spinlock plus us */
+static atomic_t spinlock_waiters = ATOMIC_INIT(1);
+static int spinlock_count = 0;
+static int spinlock_cpu = 0;
+/*
+ * Note we use nested spin locks to account for the case where a break
+ * point is encountered when calling a function by user direction from
+ * kgdb. Also there is the memory exception recursion to account for.
+ * Well, yes, but this lets other cpus thru too. Lets add a
+ * cpu id to the lock.
+ */
+#define KGDB_SPIN_LOCK(x) if( spinlock_count == 0 || \
+ spinlock_cpu != smp_processor_id()){\
+ atomic_inc(&spinlock_waiters); \
+ while (! spin_trylock(x)) {\
+ in_kgdb(®s);\
+ }\
+ atomic_dec(&spinlock_waiters); \
+ spinlock_count = 1; \
+ spinlock_cpu = smp_processor_id(); \
+ }else{ \
+ spinlock_count++; \
+ }
+#define KGDB_SPIN_UNLOCK(x) if( --spinlock_count == 0) spin_unlock(x)
+#else
+unsigned kgdb_spinlock = 0;
+#define KGDB_SPIN_LOCK(x) --*x
+#define KGDB_SPIN_UNLOCK(x) ++*x
+#endif
+
+int
+hex(char ch)
+{
+ if ((ch >= 'a') && (ch <= 'f'))
+ return (ch - 'a' + 10);
+ if ((ch >= '0') && (ch <= '9'))
+ return (ch - '0');
+ if ((ch >= 'A') && (ch <= 'F'))
+ return (ch - 'A' + 10);
+ return (-1);
+}
+
+/* scan for the sequence $<data>#<checksum> */
+void
+getpacket(char *buffer)
+{
+ unsigned char checksum;
+ unsigned char xmitcsum;
+ int i;
+ int count;
+ char ch;
+
+ do {
+ /* wait around for the start character, ignore all other characters */
+ while ((ch = (getDebugChar() & 0x7f)) != '$') ;
+ checksum = 0;
+ xmitcsum = -1;
+
+ count = 0;
+
+ /* now, read until a # or end of buffer is found */
+ while (count < BUFMAX) {
+ ch = getDebugChar() & 0x7f;
+ if (ch == '#')
+ break;
+ checksum = checksum + ch;
+ buffer[count] = ch;
+ count = count + 1;
+ }
+ buffer[count] = 0;
+
+ if (ch == '#') {
+ xmitcsum = hex(getDebugChar() & 0x7f) << 4;
+ xmitcsum += hex(getDebugChar() & 0x7f);
+ if ((remote_debug) && (checksum != xmitcsum)) {
+ printk
+ ("bad checksum. My count = 0x%x, sent=0x%x. buf=%s\n",
+ checksum, xmitcsum, buffer);
+ }
+
+ if (checksum != xmitcsum)
+ putDebugChar('-'); /* failed checksum */
+ else {
+ putDebugChar('+'); /* successful transfer */
+ /* if a sequence char is present, reply the sequence ID */
+ if (buffer[2] == ':') {
+ putDebugChar(buffer[0]);
+ putDebugChar(buffer[1]);
+ /* remove sequence chars from buffer */
+ count = strlen(buffer);
+ for (i = 3; i <= count; i++)
+ buffer[i - 3] = buffer[i];
+ }
+ }
+ }
+ } while (checksum != xmitcsum);
+
+ if (remote_debug)
+ printk("R:%s\n", buffer);
+ flushDebugChar();
+}
+
+/* send the packet in buffer. */
+
+void
+putpacket(char *buffer)
+{
+ unsigned char checksum;
+ int count;
+ char ch;
+
+ /* $<packet info>#<checksum>. */
+
+ if (!kgdboe) {
+ do {
+ if (remote_debug)
+ printk("T:%s\n", buffer);
+ putDebugChar('$');
+ checksum = 0;
+ count = 0;
+
+ while ((ch = buffer[count])) {
+ putDebugChar(ch);
+ checksum += ch;
+ count += 1;
+ }
+
+ putDebugChar('#');
+ putDebugChar(hexchars[checksum >> 4]);
+ putDebugChar(hexchars[checksum % 16]);
+ flushDebugChar();
+
+ } while ((getDebugChar() & 0x7f) != '+');
+ } else {
+ /*
+ * For udp, we can not transfer too much bytes once.
+ * We only transfer MAX_SEND_COUNT size bytes each time
+ */
+
+#define MAX_SEND_COUNT 30
+
+ int send_count = 0, i = 0;
+ char send_buf[MAX_SEND_COUNT];
+
+ do {
+ if (remote_debug)
+ printk("T:%s\n", buffer);
+ putDebugChar('$');
+ checksum = 0;
+ count = 0;
+ send_count = 0;
+ while ((ch = buffer[count])) {
+ if (send_count >= MAX_SEND_COUNT) {
+ for(i = 0; i < MAX_SEND_COUNT; i++) {
+ putDebugChar(send_buf[i]);
+ }
+ flushDebugChar();
+ send_count = 0;
+ } else {
+ send_buf[send_count] = ch;
+ checksum += ch;
+ count ++;
+ send_count++;
+ }
+ }
+ for(i = 0; i < send_count; i++)
+ putDebugChar(send_buf[i]);
+ putDebugChar('#');
+ putDebugChar(hexchars[checksum >> 4]);
+ putDebugChar(hexchars[checksum % 16]);
+ flushDebugChar();
+ } while ((getDebugChar() & 0x7f) != '+');
+ }
+}
+
+static char remcomInBuffer[BUFMAX];
+static char remcomOutBuffer[BUFMAX];
+static char lbuf[BUFMAX];
+static short error;
+
+void
+debug_error(char *format, char *parm)
+{
+ if (remote_debug)
+ printk(format, parm);
+}
+
+static void
+print_regs(struct pt_regs *regs)
+{
+ printk("RAX=%016lx RBX=%016lx RCX=%016lx\n",
+ regs->rax, regs->rbx, regs->rcx);
+ printk("RDX=%016lx RSI=%016lx RDI=%016lx\n",
+ regs->rdx, regs->rsi, regs->rdi);
+ printk("RBP=%016lx PS=%016lx PC=%016lx\n",
+ regs->rbp, regs->eflags, regs->rip);
+ printk("R8=%016lx R9=%016lx R10=%016lx\n",
+ regs->r8, regs->r9, regs->r10);
+ printk("R11=%016lx R12=%016lx R13=%016lx\n",
+ regs->r11, regs->r12, regs->r13);
+ printk("R14=%016lx R15=%016lx RSP=%016lx\n",
+ regs->r14, regs->r15, regs->rsp);
+}
+
+#define NEW_esp fn_call_lookaside[trap_cpu].rsp
+
+static void
+regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs)
+{
+ gdb_regs[_RAX] = regs->rax;
+ gdb_regs[_RBX] = regs->rbx;
+ gdb_regs[_RCX] = regs->rcx;
+ gdb_regs[_RDX] = regs->rdx;
+ gdb_regs[_RSI] = regs->rsi;
+ gdb_regs[_RDI] = regs->rdi;
+ gdb_regs[_RBP] = regs->rbp;
+ gdb_regs[ _PS] = regs->eflags;
+ gdb_regs[ _PC] = regs->rip;
+ gdb_regs[ _R8] = regs->r8;
+ gdb_regs[ _R9] = regs->r9;
+ gdb_regs[_R10] = regs->r10;
+ gdb_regs[_R11] = regs->r11;
+ gdb_regs[_R12] = regs->r12;
+ gdb_regs[_R13] = regs->r13;
+ gdb_regs[_R14] = regs->r14;
+ gdb_regs[_R15] = regs->r15;
+ gdb_regs[_RSP] = regs->rsp;
+
+ /* Note, as we are a debugging the kernel, we will always
+ * trap in kernel code, this means no priviledge change,
+ * and so the pt_regs structure is not completely valid. In a non
+ * privilege change trap, only EFLAGS, CS and EIP are put on the stack,
+ * SS and ESP are not stacked, this means that the last 2 elements of
+ * pt_regs is not valid (they would normally refer to the user stack)
+ * also, using regs+1 is no good because you end up will a value that is
+ * 2 longs (8) too high. This used to cause stepping over functions
+ * to fail, so my fix is to use the address of regs->esp, which
+ * should point at the end of the stack frame. Note I have ignored
+ * completely exceptions that cause an error code to be stacked, such
+ * as double fault. Stuart Hughes, Zentropix.
+ * original code: gdb_regs[_ESP] = (int) (regs + 1) ;
+
+ * this is now done on entry and moved to OLD_esp (as well as NEW_esp).
+ */
+}
+
+static void
+gdb_regs_to_regs(unsigned long *gdb_regs, struct pt_regs *regs)
+{
+ regs->rax = gdb_regs[_RAX] ;
+ regs->rbx = gdb_regs[_RBX] ;
+ regs->rcx = gdb_regs[_RCX] ;
+ regs->rdx = gdb_regs[_RDX] ;
+ regs->rsi = gdb_regs[_RSI] ;
+ regs->rdi = gdb_regs[_RDI] ;
+ regs->rbp = gdb_regs[_RBP] ;
+ regs->eflags = gdb_regs[ _PS] ;
+ regs->rip = gdb_regs[ _PC] ;
+ regs->r8 = gdb_regs[ _R8] ;
+ regs->r9 = gdb_regs[ _R9] ;
+ regs->r10 = gdb_regs[ _R10] ;
+ regs->r11 = gdb_regs[ _R11] ;
+ regs->r12 = gdb_regs[ _R12] ;
+ regs->r13 = gdb_regs[ _R13] ;
+ regs->r14 = gdb_regs[ _R14] ;
+ regs->r15 = gdb_regs[ _R15] ;
+ #if 0 /* can't change these */
+ regs->rsp = gdb_regs[_RSP] ;
+ regs->ss = gdb_regs[ _SS] ;
+ regs->fs = gdb_regs[_FS];
+ regs->gs = gdb_regs[_GS];
+#endif
+} /* gdb_regs_to_regs */
+
+extern void scheduling_functions_start_here(void);
+extern void scheduling_functions_end_here(void);
+#define first_sched ((unsigned long) scheduling_functions_start_here)
+#define last_sched ((unsigned long) scheduling_functions_end_here)
+
+int thread_list = 0;
+extern void thread_return(void);
+
+void
+get_gdb_regs(struct task_struct *p, struct pt_regs *regs, unsigned long *gdb_regs)
+{
+ unsigned long **rbp, *rsp, *rsp0, pc;
+ int count = 0;
+ IF_SMP(int i);
+ if (!p || p == current) {
+ regs_to_gdb_regs(gdb_regs, regs);
+ return;
+ }
+#ifdef CONFIG_SMP
+ for (i = 0; i < MAX_NO_CPUS; i++) {
+ if (p == kgdb_info.cpus_waiting[i].task) {
+ regs_to_gdb_regs(gdb_regs,
+ kgdb_info.cpus_waiting[i].regs);
+ gdb_regs[_RSP] =
+ (unsigned long)&kgdb_info.cpus_waiting[i].regs->rsp;
+
+ return;
+ }
+ }
+#endif
+ memset(gdb_regs, 0, NUMREGBYTES);
+ rsp = (unsigned long *)p->thread.rsp;
+ rbp = (unsigned long **)rsp[0];
+ rsp += 2;
+ gdb_regs[_PC] = (unsigned long)thread_return;
+ gdb_regs[_RBP] = (unsigned long)rbp;
+ gdb_regs[_RSP] = (unsigned long)rsp;
+
+/*
+ * This code is to give a more informative notion of where a process
+ * is waiting. It is used only when the user asks for a thread info
+ * list. If he then switches to the thread, s/he will find the task
+ * is in schedule, but a back trace should show the same info we come
+ * up with. This code was shamelessly purloined from process.c. It was
+ * then enhanced to provide more registers than simply the program
+ * counter.
+ */
+
+ if (!thread_list) {
+ return;
+ }
+
+ if (p->state == TASK_RUNNING)
+ return;
+ rsp0 = (unsigned long *)p->thread.rsp0;
+ if (rsp < (unsigned long *) p->thread_info || rsp > rsp0)
+ return;
+ /* include/asm-i386/system.h:switch_to() pushes ebp last. */
+ do {
+ if (*rbp < rsp || *rbp > rsp0)
+ break;
+ rbp = (unsigned long **)*rbp;
+ rsp = (unsigned long *)rbp;
+ pc = rsp[1];
+
+ if (pc < first_sched || pc >= last_sched)
+ break;
+ gdb_regs[_PC] = (unsigned long)pc;
+ gdb_regs[_RSP] = (unsigned long)rsp;
+ gdb_regs[_RBP] = (unsigned long)rbp;
+ } while (count++ < 16);
+ return;
+}
+
+/* convert the memory pointed to by mem into hex, placing result in buf */
+/* returns nonzero if any memory access fails. */
+int mem2hex( char* mem, char* buf, int count)
+{
+ int i;
+ unsigned char ch;
+ int ret = 0;
+
+ for (i=0;i<count;i++) {
+ ch = 0;
+ ret |= __get_user(ch, mem);
+ mem++;
+ *buf++ = hexchars[ch >> 4];
+ *buf++ = hexchars[ch % 16];
+ }
+ *buf = 0;
+ if (ret) {
+ Dearly_printk("mem2hex: fault at accessing %p\n", mem);
+ }
+ return(ret);
+}
+
+/* convert the hex array pointed to by buf into binary to be placed in mem */
+/* return nonzero if any memory access fails. */
+int hex2mem( char* buf, char* mem, int count)
+{
+ int i;
+ unsigned char ch;
+ int ret = 0;
+
+ for (i=0;i<count;i++) {
+ ch = hex(*buf++) << 4;
+ ch = ch + hex(*buf++);
+ ret |= __put_user(ch, mem);
+ mem++;
+ }
+ if (ret) {
+ Dearly_printk("hex2mem: fault at %p\n", mem);
+ }
+ return(ret);
+}
+
+#if 0
+/* Indicate to caller of mem2hex or hex2mem that there has been an
+ error. */
+static volatile int mem_err = 0;
+static volatile int mem_err_expected = 0;
+static volatile int mem_err_cnt = 0;
+static int garbage_loc = -1;
+
+int
+get_char(char *addr)
+{
+ return *addr;
+}
+
+void
+set_char(char *addr, int val, int may_fault)
+{
+ /*
+ * This code traps references to the area mapped to the kernel
+ * stack as given by the regs and, instead, stores to the
+ * fn_call_lookaside[cpu].array
+ */
+ if (may_fault &&
+ (unsigned int) addr < OLD_esp &&
+ ((unsigned int) addr > (OLD_esp - (unsigned int) LOOKASIDE_SIZE))) {
+ addr = (char *) END_OF_LOOKASIDE - ((char *) OLD_esp - addr);
+ }
+ *addr = val;
+}
+
+/* convert the memory pointed to by mem into hex, placing result in buf */
+/* return a pointer to the last char put in buf (null) */
+/* If MAY_FAULT is non-zero, then we should set mem_err in response to
+ a fault; if zero treat a fault like any other fault in the stub. */
+char *
+mem2hex(char *mem, char *buf, int count, int may_fault)
+{
+ int i;
+ unsigned char ch;
+
+ if (may_fault) {
+ mem_err_expected = 1;
+ mem_err = 0;
+ }
+ for (i = 0; i < count; i++) {
+ /* printk("%lx = ", mem) ; */
+
+ ch = get_char(mem++);
+
+ /* printk("%02x\n", ch & 0xFF) ; */
+ if (may_fault && mem_err) {
+ if (remote_debug)
+ printk("Mem fault fetching from addr %lx\n",
+ (long) (mem - 1));
+ *buf = 0; /* truncate buffer */
+ return (buf);
+ }
+ *buf++ = hexchars[ch >> 4];
+ *buf++ = hexchars[ch % 16];
+ }
+ *buf = 0;
+ if (may_fault)
+ mem_err_expected = 0;
+ return (buf);
+}
+
+/* convert the hex array pointed to by buf into binary to be placed in mem */
+/* return a pointer to the character AFTER the last byte written */
+/* NOTE: We use the may fault flag to also indicate if the write is to
+ * the registers (0) or "other" memory (!=0)
+ */
+char *
+hex2mem(char *buf, char *mem, int count, int may_fault)
+{
+ int i;
+ unsigned char ch;
+
+ if (may_fault) {
+ mem_err_expected = 1;
+ mem_err = 0;
+ }
+ for (i = 0; i < count; i++) {
+ ch = hex(*buf++) << 4;
+ ch = ch + hex(*buf++);
+ set_char(mem++, ch, may_fault);
+
+ if (may_fault && mem_err) {
+ if (remote_debug)
+ printk("Mem fault storing to addr %lx\n",
+ (long) (mem - 1));
+ return (mem);
+ }
+ }
+ if (may_fault)
+ mem_err_expected = 0;
+ return (mem);
+}
+#endif
+
+/**********************************************/
+/* WHILE WE FIND NICE HEX CHARS, BUILD AN INT */
+/* RETURN NUMBER OF CHARS PROCESSED */
+/**********************************************/
+int
+hexToLong(char **ptr, unsigned long *value)
+{
+ int numChars = 0;
+ int hexValue;
+
+ *value = 0;
+
+ while (**ptr) {
+ hexValue = hex(**ptr);
+ if (hexValue >= 0) {
+ *value = (*value << 4) | hexValue;
+ numChars++;
+ } else
+ break;
+
+ (*ptr)++;
+ }
+
+ return (numChars);
+}
+
+#define stubhex(h) hex(h)
+#ifdef old_thread_list
+
+static int
+stub_unpack_int(char *buff, int fieldlength)
+{
+ int nibble;
+ int retval = 0;
+
+ while (fieldlength) {
+ nibble = stubhex(*buff++);
+ retval |= nibble;
+ fieldlength--;
+ if (fieldlength)
+ retval = retval << 4;
+ }
+ return retval;
+}
+#endif
+static char *
+pack_hex_byte(char *pkt, int byte)
+{
+ *pkt++ = hexchars[(byte >> 4) & 0xf];
+ *pkt++ = hexchars[(byte & 0xf)];
+ return pkt;
+}
+
+#define BUF_THREAD_ID_SIZE 16
+
+static char *
+pack_threadid(char *pkt, threadref * id)
+{
+ char *limit;
+ unsigned char *altid;
+
+ altid = (unsigned char *) id;
+ limit = pkt + BUF_THREAD_ID_SIZE;
+ while (pkt < limit)
+ pkt = pack_hex_byte(pkt, *altid++);
+ return pkt;
+}
+
+#ifdef old_thread_list
+static char *
+unpack_byte(char *buf, int *value)
+{
+ *value = stub_unpack_int(buf, 2);
+ return buf + 2;
+}
+
+static char *
+unpack_threadid(char *inbuf, threadref * id)
+{
+ char *altref;
+ char *limit = inbuf + BUF_THREAD_ID_SIZE;
+ int x, y;
+
+ altref = (char *) id;
+
+ while (inbuf < limit) {
+ x = stubhex(*inbuf++);
+ y = stubhex(*inbuf++);
+ *altref++ = (x << 4) | y;
+ }
+ return inbuf;
+}
+#endif
+void
+int_to_threadref(threadref * id, int value)
+{
+ unsigned char *scan;
+
+ scan = (unsigned char *) id;
+ {
+ int i = 4;
+ while (i--)
+ *scan++ = 0;
+ }
+ *scan++ = (value >> 24) & 0xff;
+ *scan++ = (value >> 16) & 0xff;
+ *scan++ = (value >> 8) & 0xff;
+ *scan++ = (value & 0xff);
+}
+int
+int_to_hex_v(unsigned char * id, int value)
+{
+ unsigned char *start = id;
+ int shift;
+ int ch;
+
+ for (shift = 28; shift >= 0; shift -= 4) {
+ if ((ch = (value >> shift) & 0xf) || (id != start)) {
+ *id = hexchars[ch];
+ id++;
+ }
+ }
+ if (id == start)
+ *id++ = '0';
+ return id - start;
+}
+#ifdef old_thread_list
+
+static int
+threadref_to_int(threadref * ref)
+{
+ int i, value = 0;
+ unsigned char *scan;
+
+ scan = (char *) ref;
+ scan += 4;
+ i = 4;
+ while (i-- > 0)
+ value = (value << 8) | ((*scan++) & 0xff);
+ return value;
+}
+#endif
+static int
+cmp_str(char *s1, char *s2, int count)
+{
+ while (count--) {
+ if (*s1++ != *s2++)
+ return 0;
+ }
+ return 1;
+}
+
+#if 1 /* this is a hold over from 2.4 where O(1) was "sometimes" */
+extern struct task_struct *kgdb_get_idle(int cpu);
+#define idle_task(cpu) kgdb_get_idle(cpu)
+#else
+#define idle_task(cpu) init_tasks[cpu]
+#endif
+
+extern int kgdb_pid_init_done;
+
+struct task_struct *
+getthread(int pid)
+{
+ struct task_struct *thread;
+ if (pid >= PID_MAX && pid <= (PID_MAX + MAX_NO_CPUS)) {
+ if (!cpu_online(pid - PID_MAX))
+ return NULL;
+
+ return idle_task(pid - PID_MAX);
+ } else {
+ /*
+ * find_task_by_pid is relatively safe all the time
+ * Other pid functions require lock downs which imply
+ * that we may be interrupting them (as we get here
+ * in the middle of most any lock down).
+ * Still we don't want to call until the table exists!
+ */
+ if (kgdb_pid_init_done){
+ thread = find_task_by_pid(pid);
+ if (thread) {
+ return thread;
+ }
+ }
+ }
+ return NULL;
+}
+/* *INDENT-OFF* */
+struct hw_breakpoint {
+ unsigned enabled;
+ unsigned type;
+ unsigned len;
+ unsigned long addr;
+} breakinfo[4] = { {enabled:0},
+ {enabled:0},
+ {enabled:0},
+ {enabled:0}};
+/* *INDENT-ON* */
+unsigned long hw_breakpoint_status;
+void
+correct_hw_break(void)
+{
+ int breakno;
+ int correctit;
+ int breakbit;
+ unsigned long dr7;
+
+ asm volatile ("movq %%db7, %0\n":"=r" (dr7)
+ :);
+ /* *INDENT-OFF* */
+ do {
+ unsigned long addr0, addr1, addr2, addr3;
+ asm volatile ("movq %%db0, %0\n"
+ "movq %%db1, %1\n"
+ "movq %%db2, %2\n"
+ "movq %%db3, %3\n"
+ :"=r" (addr0), "=r"(addr1),
+ "=r"(addr2), "=r"(addr3)
+ :);
+ } while (0);
+ /* *INDENT-ON* */
+ correctit = 0;
+ for (breakno = 0; breakno < 3; breakno++) {
+ breakbit = 2 << (breakno << 1);
+ if (!(dr7 & breakbit) && breakinfo[breakno].enabled) {
+ correctit = 1;
+ dr7 |= breakbit;
+ dr7 &= ~(0xf0000 << (breakno << 2));
+ dr7 |= (((breakinfo[breakno].len << 2) |
+ breakinfo[breakno].type) << 16) <<
+ (breakno << 2);
+ switch (breakno) {
+ case 0:
+ asm volatile ("movq %0, %%dr0\n"::"r"
+ (breakinfo[breakno].addr));
+ break;
+
+ case 1:
+ asm volatile ("movq %0, %%dr1\n"::"r"
+ (breakinfo[breakno].addr));
+ break;
+
+ case 2:
+ asm volatile ("movq %0, %%dr2\n"::"r"
+ (breakinfo[breakno].addr));
+ break;
+
+ case 3:
+ asm volatile ("movq %0, %%dr3\n"::"r"
+ (breakinfo[breakno].addr));
+ break;
+ }
+ } else if ((dr7 & breakbit) && !breakinfo[breakno].enabled) {
+ correctit = 1;
+ dr7 &= ~breakbit;
+ dr7 &= ~(0xf0000 << (breakno << 2));
+ }
+ }
+ if (correctit) {
+ asm volatile ("movq %0, %%db7\n"::"r" (dr7));
+ }
+}
+
+int
+remove_hw_break(unsigned breakno)
+{
+ if (!breakinfo[breakno].enabled) {
+ return -1;
+ }
+ breakinfo[breakno].enabled = 0;
+ return 0;
+}
+
+int
+set_hw_break(unsigned breakno, unsigned type, unsigned len, unsigned addr)
+{
+ if (breakinfo[breakno].enabled) {
+ return -1;
+ }
+ breakinfo[breakno].enabled = 1;
+ breakinfo[breakno].type = type;
+ breakinfo[breakno].len = len;
+ breakinfo[breakno].addr = addr;
+ return 0;
+}
+
+#ifdef CONFIG_SMP
+static int in_kgdb_console = 0;
+
+int
+in_kgdb(struct pt_regs *regs)
+{
+ unsigned long flags;
+ int cpu;
+ if (!kgdb_enabled)
+ return 0;
+ cpu = smp_processor_id();
+ in_kgdb_called = 1;
+ if (!spin_is_locked(&kgdb_spinlock)) {
+ if (in_kgdb_here_log[cpu] || /* we are holding this cpu */
+ in_kgdb_console) { /* or we are doing slow i/o */
+ return 1;
+ }
+ return 0;
+ }
+
+ /* As I see it the only reason not to let all cpus spin on
+ * the same spin_lock is to allow selected ones to proceed.
+ * This would be a good thing, so we leave it this way.
+ * Maybe someday.... Done !
+
+ * in_kgdb() is called from an NMI so we don't pretend
+ * to have any resources, like printk() for example.
+ */
+
+ local_irq_save(flags); /* only local here, to avoid hanging */
+ /*
+ * log arival of this cpu
+ * The NMI keeps on ticking. Protect against recurring more
+ * than once, and ignor the cpu that has the kgdb lock
+ */
+ in_kgdb_entry_log[cpu]++;
+ in_kgdb_here_log[cpu] = regs;
+ if (cpu == spinlock_cpu || waiting_cpus[cpu].task)
+ goto exit_in_kgdb;
+
+ /*
+ * For protection of the initilization of the spin locks by kgdb
+ * it locks the kgdb spinlock before it gets the wait locks set
+ * up. We wait here for the wait lock to be taken. If the
+ * kgdb lock goes away first?? Well, it could be a slow exit
+ * sequence where the wait lock is removed prior to the kgdb lock
+ * so if kgdb gets unlocked, we just exit.
+ */
+
+ while (spin_is_locked(&kgdb_spinlock) &&
+ !spin_is_locked(waitlocks + cpu)) ;
+ if (!spin_is_locked(&kgdb_spinlock))
+ goto exit_in_kgdb;
+
+ waiting_cpus[cpu].task = current;
+ waiting_cpus[cpu].pid = (current->pid) ? : (PID_MAX + cpu);
+ waiting_cpus[cpu].regs = regs;
+
+ spin_unlock_wait(waitlocks + cpu);
+
+ /*
+ * log departure of this cpu
+ */
+ waiting_cpus[cpu].task = 0;
+ waiting_cpus[cpu].pid = 0;
+ waiting_cpus[cpu].regs = 0;
+ correct_hw_break();
+ exit_in_kgdb:
+ in_kgdb_here_log[cpu] = 0;
+ local_irq_restore(flags);
+ return 1;
+ /*
+ spin_unlock(continuelocks + smp_processor_id());
+ */
+}
+
+void
+smp__in_kgdb(struct pt_regs regs)
+{
+ ack_APIC_irq();
+ in_kgdb(®s);
+}
+#else
+int
+in_kgdb(struct pt_regs *regs)
+{
+ return (kgdb_spinlock);
+}
+#endif
+
+void
+printexceptioninfo(int exceptionNo, int errorcode, char *buffer)
+{
+ unsigned long dr6;
+ int i;
+ switch (exceptionNo) {
+ case 1: /* debug exception */
+ break;
+ case 3: /* breakpoint */
+ sprintf(buffer, "Software breakpoint");
+ return;
+ default:
+ sprintf(buffer, "Details not available");
+ return;
+ }
+ asm volatile ("movq %%db6, %0\n":"=r" (dr6)
+ :);
+ if (dr6 & 0x4000) {
+ sprintf(buffer, "Single step");
+ return;
+ }
+ for (i = 0; i < 4; ++i) {
+ if (dr6 & (1 << i)) {
+ sprintf(buffer, "Hardware breakpoint %d", i);
+ return;
+ }
+ }
+ sprintf(buffer, "Unknown trap");
+ return;
+}
+
+/*
+ * The ThreadExtraInfo query allows us to pass an arbitrary string
+ * for display with the "info threads" command.
+ */
+
+void
+print_extra_info(task_t *p, char *buf)
+{
+ if (!p) {
+ sprintf(buf, "Invalid thread");
+ return;
+ }
+ sprintf(buf, "0x%p %8d %4d %c %s",
+ (void *)p, p->parent->pid,
+ task_cpu(p),
+ (p->state == 0) ? (task_curr(p)?'R':'r') :
+ (p->state < 0) ? 'U' :
+ (p->state & TASK_UNINTERRUPTIBLE) ? 'D' :
+ (p->state & TASK_STOPPED || p->ptrace & PT_PTRACED) ? 'T' :
+ (p->state & (TASK_ZOMBIE | TASK_DEAD)) ? 'Z' :
+ (p->state & TASK_INTERRUPTIBLE) ? 'S' : '?',
+ p->comm);
+}
+
+/*
+ * This function does all command procesing for interfacing to gdb.
+ *
+ * NOTE: The INT nn instruction leaves the state of the interrupt
+ * enable flag UNCHANGED. That means that when this routine
+ * is entered via a breakpoint (INT 3) instruction from code
+ * that has interrupts enabled, then interrupts will STILL BE
+ * enabled when this routine is entered. The first thing that
+ * we do here is disable interrupts so as to prevent recursive
+ * entries and bothersome serial interrupts while we are
+ * trying to run the serial port in polled mode.
+ *
+ * For kernel version 2.1.xx the kgdb_cli() actually gets a spin lock so
+ * it is always necessary to do a restore_flags before returning
+ * so as to let go of that lock.
+ */
+int
+kgdb_handle_exception(int exceptionVector,
+ int signo, int err_code, struct pt_regs *linux_regs)
+{
+ struct task_struct *usethread = NULL;
+ struct task_struct *thread_list_start = 0, *thread = NULL;
+ struct task_struct *p;
+ unsigned long addr, length;
+ unsigned long breakno, breaktype;
+ char *ptr;
+ unsigned long newPC;
+ threadref thref;
+ unsigned long threadid, tmpid;
+ int thread_min = PID_MAX + MAX_NO_CPUS;
+#ifdef old_thread_list
+ int maxthreads;
+#endif
+ int nothreads;
+ unsigned long flags;
+ unsigned long gdb_regs[NUMREGS];
+ unsigned long dr6;
+ IF_SMP(int entry_state = 0); /* 0, ok, 1, no nmi, 2 sync failed */
+#define NO_NMI 1
+#define NO_SYNC 2
+#define regs (*linux_regs)
+ /*
+ * If the entry is not from the kernel then return to the Linux
+ * trap handler and let it process the interrupt normally.
+ */
+ if ((linux_regs->eflags & VM_MASK) || (3 & linux_regs->cs)) {
+ printk("ignoring non-kernel exception\n");
+ print_regs(®s);
+ return (0);
+ }
+ /*
+ * If we're using eth mode, set the 'mode' in the netdevice.
+ */
+
+ if (kgdboe)
+ netpoll_set_trap(1);
+
+ local_irq_save(flags);
+
+ /* Get kgdb spinlock */
+
+ KGDB_SPIN_LOCK(&kgdb_spinlock);
+ rdtscll(kgdb_info.entry_tsc);
+ /*
+ * We depend on this spinlock and the NMI watch dog to control the
+ * other cpus. They will arrive at "in_kgdb()" as a result of the
+ * NMI and will wait there for the following spin locks to be
+ * released.
+ */
+#ifdef CONFIG_SMP
+
+#if 0
+ if (cpu_callout_map & ~MAX_CPU_MASK) {
+ printk("kgdb : too many cpus, possibly not mapped"
+ " in contiguous space, change MAX_NO_CPUS"
+ " in kgdb_stub and make new kernel.\n"
+ " cpu_callout_map is %lx\n", cpu_callout_map);
+ goto exit_just_unlock;
+ }
+#endif
+ if (spinlock_count == 1) {
+ int time, end_time, dum;
+ int i;
+ int cpu_logged_in[MAX_NO_CPUS] = {[0 ... MAX_NO_CPUS - 1] = (0)
+ };
+ if (remote_debug) {
+ printk("kgdb : cpu %d entry, syncing others\n",
+ smp_processor_id());
+ }
+ for (i = 0; i < MAX_NO_CPUS; i++) {
+ /*
+ * Use trylock as we may already hold the lock if
+ * we are holding the cpu. Net result is all
+ * locked.
+ */
+ spin_trylock(&waitlocks[i]);
+ }
+ for (i = 0; i < MAX_NO_CPUS; i++)
+ cpu_logged_in[i] = 0;
+ /*
+ * Wait for their arrival. We know the watch dog is active if
+ * in_kgdb() has ever been called, as it is always called on a
+ * watchdog tick.
+ */
+ rdtsc(dum, time);
+ end_time = time + 2; /* Note: we use the High order bits! */
+ i = 1;
+ if (num_online_cpus() > 1) {
+ int me_in_kgdb = in_kgdb_entry_log[smp_processor_id()];
+ smp_send_nmi_allbutself();
+
+ while (i < num_online_cpus() && time != end_time) {
+ int j;
+ for (j = 0; j < MAX_NO_CPUS; j++) {
+ if (waiting_cpus[j].task &&
+ waiting_cpus[j].task != NOCPU &&
+ !cpu_logged_in[j]) {
+ i++;
+ cpu_logged_in[j] = 1;
+ if (remote_debug) {
+ printk
+ ("kgdb : cpu %d arrived at kgdb\n",
+ j);
+ }
+ break;
+ } else if (!waiting_cpus[j].task &&
+ !cpu_online(j)) {
+ waiting_cpus[j].task = NOCPU;
+ cpu_logged_in[j] = 1;
+ waiting_cpus[j].hold = 1;
+ break;
+ }
+ if (!waiting_cpus[j].task &&
+ in_kgdb_here_log[j]) {
+
+ int wait = 100000;
+ while (wait--) ;
+ if (!waiting_cpus[j].task &&
+ in_kgdb_here_log[j]) {
+ printk
+ ("kgdb : cpu %d stall"
+ " in in_kgdb\n",
+ j);
+ i++;
+ cpu_logged_in[j] = 1;
+ waiting_cpus[j].task =
+ (struct task_struct
+ *) 1;
+ }
+ }
+ }
+
+ if (in_kgdb_entry_log[smp_processor_id()] >
+ (me_in_kgdb + 10)) {
+ break;
+ }
+
+ rdtsc(dum, time);
+ }
+ if (i < num_online_cpus()) {
+ printk
+ ("kgdb : time out, proceeding without sync\n");
+#if 0
+ printk("kgdb : Waiting_cpus: 0 = %d, 1 = %d\n",
+ waiting_cpus[0].task != 0,
+ waiting_cpus[1].task != 0);
+ printk("kgdb : Cpu_logged in: 0 = %d, 1 = %d\n",
+ cpu_logged_in[0], cpu_logged_in[1]);
+ printk
+ ("kgdb : in_kgdb_here_log in: 0 = %d, 1 = %d\n",
+ in_kgdb_here_log[0] != 0,
+ in_kgdb_here_log[1] != 0);
+#endif
+ entry_state = NO_SYNC;
+ } else {
+#if 0
+ int ent =
+ in_kgdb_entry_log[smp_processor_id()] -
+ me_in_kgdb;
+ printk("kgdb : sync after %d entries\n", ent);
+#endif
+ }
+ } else {
+ if (remote_debug) {
+ printk
+ ("kgdb : %d cpus, but watchdog not active\n"
+ "proceeding without locking down other cpus\n",
+ num_online_cpus());
+ entry_state = NO_NMI;
+ }
+ }
+ }
+#endif
+
+ if (remote_debug) {
+ unsigned long *lp = (unsigned long *) &linux_regs;
+
+ printk("handle_exception(exceptionVector=%d, "
+ "signo=%d, err_code=%d, linux_regs=%p)\n",
+ exceptionVector, signo, err_code, linux_regs);
+ if (debug_regs) {
+ print_regs(®s);
+ printk("Stk: %8lx %8lx %8lx %8lx"
+ " %8lx %8lx %8lx %8lx\n",
+ lp[0], lp[1], lp[2], lp[3],
+ lp[4], lp[5], lp[6], lp[7]);
+ printk(" %8lx %8lx %8lx %8lx"
+ " %8lx %8lx %8lx %8lx\n",
+ lp[8], lp[9], lp[10], lp[11],
+ lp[12], lp[13], lp[14], lp[15]);
+ printk(" %8lx %8lx %8lx %8lx "
+ "%8lx %8lx %8lx %8lx\n",
+ lp[16], lp[17], lp[18], lp[19],
+ lp[20], lp[21], lp[22], lp[23]);
+ printk(" %8lx %8lx %8lx %8lx "
+ "%8lx %8lx %8lx %8lx\n",
+ lp[24], lp[25], lp[26], lp[27],
+ lp[28], lp[29], lp[30], lp[31]);
+ }
+ }
+
+ /* Disable hardware debugging while we are in kgdb */
+ /* Get the debug register status register */
+/* *INDENT-OFF* */
+ __asm__("movq %0,%%db7"
+ : /* no output */
+ :"r"(0UL));
+
+ asm volatile ("movq %%db6, %0\n"
+ :"=r" (hw_breakpoint_status)
+ :);
+
+#if 0
+/* *INDENT-ON* */
+ switch (exceptionVector) {
+ case 0: /* divide error */
+ case 1: /* debug exception */
+ case 2: /* NMI */
+ case 3: /* breakpoint */
+ case 4: /* overflow */
+ case 5: /* bounds check */
+ case 6: /* invalid opcode */
+ case 7: /* device not available */
+ case 8: /* double fault (errcode) */
+ case 10: /* invalid TSS (errcode) */
+ case 12: /* stack fault (errcode) */
+ case 16: /* floating point error */
+ case 17: /* alignment check (errcode) */
+ default: /* any undocumented */
+ break;
+ case 11: /* segment not present (errcode) */
+ case 13: /* general protection (errcode) */
+ case 14: /* page fault (special errcode) */
+ case 19: /* cache flush denied */
+ if (mem_err_expected) {
+ /*
+ * This fault occured because of the
+ * get_char or set_char routines. These
+ * two routines use either eax of edx to
+ * indirectly reference the location in
+ * memory that they are working with.
+ * For a page fault, when we return the
+ * instruction will be retried, so we
+ * have to make sure that these
+ * registers point to valid memory.
+ */
+ mem_err = 1; /* set mem error flag */
+ mem_err_expected = 0;
+ mem_err_cnt++; /* helps in debugging */
+ /* make valid address */
+ regs.eax = (long) &garbage_loc;
+ /* make valid address */
+ regs.edx = (long) &garbage_loc;
+ if (remote_debug)
+ printk("Return after memory error: "
+ "mem_err_cnt=%d\n", mem_err_cnt);
+ if (debug_regs)
+ print_regs(®s);
+ goto exit_kgdb;
+ }
+ break;
+ }
+#endif
+ if (remote_debug)
+ printk("kgdb : entered kgdb on cpu %d\n", smp_processor_id());
+
+ gdb_i386vector = exceptionVector;
+ gdb_i386errcode = err_code;
+ kgdb_info.called_from = __builtin_return_address(0);
+#ifdef CONFIG_SMP
+ /*
+ * OK, we can now communicate, lets tell gdb about the sync.
+ * but only if we had a problem.
+ */
+ switch (entry_state) {
+ case NO_NMI:
+ to_gdb("NMI not active, other cpus not stopped\n");
+ break;
+ case NO_SYNC:
+ to_gdb("Some cpus not stopped, see 'kgdb_info' for details\n");
+ default:;
+ }
+
+#endif
+/*
+ * Set up the gdb function call area.
+ */
+ trap_cpu = smp_processor_id();
+ OLD_esp = NEW_esp = (unsigned long) (&linux_regs->rsp);
+
+ IF_SMP(once_again:)
+ /* reply to host that an exception has occurred */
+ remcomOutBuffer[0] = 'S';
+ remcomOutBuffer[1] = hexchars[signo >> 4];
+ remcomOutBuffer[2] = hexchars[signo % 16];
+ remcomOutBuffer[3] = 0;
+
+ putpacket(remcomOutBuffer);
+
+ while (1 == 1) {
+ error = 0;
+ remcomOutBuffer[0] = 0;
+ getpacket(remcomInBuffer);
+ switch (remcomInBuffer[0]) {
+ case '?':
+ remcomOutBuffer[0] = 'S';
+ remcomOutBuffer[1] = hexchars[signo >> 4];
+ remcomOutBuffer[2] = hexchars[signo % 16];
+ remcomOutBuffer[3] = 0;
+ break;
+ case 'd':
+ remote_debug = !(remote_debug); /* toggle debug flag */
+ printk("Remote debug %s\n",
+ remote_debug ? "on" : "off");
+ break;
+ case 'g': /* return the value of the CPU registers */
+ get_gdb_regs(usethread, ®s, gdb_regs);
+ mem2hex((char *) gdb_regs,
+ remcomOutBuffer, NUMREGBYTES);
+ break;
+ case 'G': /* set the value of the CPU registers - return OK */
+ hex2mem(&remcomInBuffer[1],
+ (char *) gdb_regs, NUMREGBYTES);
+ if (!usethread || usethread == current) {
+ gdb_regs_to_regs(gdb_regs, ®s);
+ strcpy(remcomOutBuffer, "OK");
+ } else {
+ strcpy(remcomOutBuffer, "E00");
+ }
+ break;
+
+ case 'P':{ /* set the value of a single CPU register -
+ return OK */
+ /*
+ * For some reason, gdb wants to talk about psudo
+ * registers (greater than 15).
+ */
+ unsigned long regno;
+
+ ptr = &remcomInBuffer[1];
+ regs_to_gdb_regs(gdb_regs, ®s);
+ if ((!usethread || usethread == current) &&
+ hexToLong(&ptr, ®no) &&
+ *ptr++ == '=' && (regno >= 0)) {
+ if (regno >= NUMREGS)
+ break;
+ hex2mem(ptr, (char *) &gdb_regs[regno],
+ 8);
+ gdb_regs_to_regs(gdb_regs, ®s);
+ strcpy(remcomOutBuffer, "OK");
+ break;
+ }
+ strcpy(remcomOutBuffer, "E01");
+ break;
+ }
+
+ /* mAA..AA,LLLL Read LLLL bytes at address AA..AA */
+ case 'm':
+ /* TRY TO READ %x,%x. IF SUCCEED, SET PTR = 0 */
+ ptr = &remcomInBuffer[1];
+ if (hexToLong(&ptr, &addr) &&
+ (*(ptr++) == ',') && (hexToLong(&ptr, &length))) {
+ ptr = 0;
+ /*
+ * hex doubles the byte count
+ */
+ if (length > (BUFMAX / 2))
+ length = BUFMAX / 2;
+ if (mem2hex((char *) addr,
+ remcomOutBuffer, length)) {
+ strcpy(remcomOutBuffer, "E03");
+ debug_error("memory fault\n", NULL);
+ }
+ }
+
+ if (ptr) {
+ strcpy(remcomOutBuffer, "E01");
+ debug_error
+ ("malformed read memory command: %s\n",
+ remcomInBuffer);
+ }
+ break;
+
+ /* MAA..AA,LLLL:
+ Write LLLL bytes at address AA.AA return OK */
+ case 'M':
+ /* TRY TO READ '%x,%x:'. IF SUCCEED, SET PTR = 0 */
+ ptr = &remcomInBuffer[1];
+ if (hexToLong(&ptr, &addr) &&
+ (*(ptr++) == ',') &&
+ (hexToLong(&ptr, &length)) && (*(ptr++) == ':')) {
+ if (hex2mem(ptr, (char *) addr, length)) {
+ strcpy(remcomOutBuffer, "E03");
+ debug_error("memory fault\n", NULL);
+ } else {
+ strcpy(remcomOutBuffer, "OK");
+ }
+
+ ptr = 0;
+ }
+ if (ptr) {
+ strcpy(remcomOutBuffer, "E02");
+ debug_error
+ ("malformed write memory command: %s\n",
+ remcomInBuffer);
+ }
+ break;
+ case 'S':
+ remcomInBuffer[0] = 's';
+ case 'C':
+ /* Csig;AA..AA where ;AA..AA is optional
+ * continue with signal
+ * Since signals are meaning less to us, delete that
+ * part and then fall into the 'c' code.
+ */
+ ptr = &remcomInBuffer[1];
+ length = 2;
+ while (*ptr && *ptr != ';') {
+ length++;
+ ptr++;
+ }
+ if (*ptr) {
+ do {
+ ptr++;
+ *(ptr - length++) = *ptr;
+ } while (*ptr);
+ } else {
+ remcomInBuffer[1] = 0;
+ }
+
+ /* cAA..AA Continue at address AA..AA(optional) */
+ /* sAA..AA Step one instruction from AA..AA(optional) */
+ /* D detach, reply OK and then continue */
+ case 'c':
+ case 's':
+ case 'D':
+
+ /* try to read optional parameter,
+ pc unchanged if no parm */
+ ptr = &remcomInBuffer[1];
+ if (hexToLong(&ptr, &addr)) {
+ if (remote_debug)
+ printk("Changing EIP to 0x%lx\n", addr);
+
+ regs.rip = addr;
+ }
+
+ newPC = regs.rip;
+
+ /* clear the trace bit */
+ regs.eflags &= 0xfffffeff;
+
+ /* set the trace bit if we're stepping */
+ if (remcomInBuffer[0] == 's')
+ regs.eflags |= 0x100;
+
+ /* detach is a friendly version of continue. Note that
+ debugging is still enabled (e.g hit control C)
+ */
+ if (remcomInBuffer[0] == 'D') {
+ strcpy(remcomOutBuffer, "OK");
+ putpacket(remcomOutBuffer);
+ }
+
+ if (remote_debug) {
+ printk("Resuming execution\n");
+ print_regs(®s);
+ }
+ asm volatile ("movq %%db6, %0\n":"=r" (dr6)
+ :);
+ if (!(dr6 & 0x4000)) {
+ for (breakno = 0; breakno < 4; ++breakno) {
+ if (dr6 & (1 << breakno) &&
+ (breakinfo[breakno].type == 0)) {
+ /* Set restore flag */
+ regs.eflags |= 0x10000;
+ break;
+ }
+ }
+ }
+
+ if (kgdboe)
+ netpoll_set_trap(0);
+
+ correct_hw_break();
+ asm volatile ("movq %0, %%db6\n"::"r" (0UL));
+ goto exit_kgdb;
+
+ /* kill the program */
+ case 'k': /* do nothing */
+ break;
+
+ /* query */
+ case 'q':
+ nothreads = 0;
+ switch (remcomInBuffer[1]) {
+ case 'f':
+ threadid = 1;
+ thread_list = 2;
+ thread_list_start = (usethread ? : current);
+ case 's':
+ if (!cmp_str(&remcomInBuffer[2],
+ "ThreadInfo", 10))
+ break;
+
+ remcomOutBuffer[nothreads++] = 'm';
+ for (; threadid < PID_MAX + MAX_NO_CPUS;
+ threadid++) {
+ thread = getthread(threadid);
+ if (thread) {
+ nothreads += int_to_hex_v(
+ &remcomOutBuffer[
+ nothreads],
+ threadid);
+ if (thread_min > threadid)
+ thread_min = threadid;
+ remcomOutBuffer[
+ nothreads] = ',';
+ nothreads++;
+ if (nothreads > BUFMAX - 10)
+ break;
+ }
+ }
+ if (remcomOutBuffer[nothreads - 1] == 'm') {
+ remcomOutBuffer[nothreads - 1] = 'l';
+ } else {
+ nothreads--;
+ }
+ remcomOutBuffer[nothreads] = 0;
+ break;
+
+#ifdef old_thread_list /* Old thread info request */
+ case 'L':
+ /* List threads */
+ thread_list = 2;
+ thread_list_start = (usethread ? : current);
+ unpack_byte(remcomInBuffer + 3, &maxthreads);
+ unpack_threadid(remcomInBuffer + 5, &thref);
+ do {
+ int buf_thread_limit =
+ (BUFMAX - 22) / BUF_THREAD_ID_SIZE;
+ if (maxthreads > buf_thread_limit) {
+ maxthreads = buf_thread_limit;
+ }
+ } while (0);
+ remcomOutBuffer[0] = 'q';
+ remcomOutBuffer[1] = 'M';
+ remcomOutBuffer[4] = '0';
+ pack_threadid(remcomOutBuffer + 5, &thref);
+
+ /* If start flag set start at 0. */
+ if (remcomInBuffer[2] == '1')
+ threadid = 0;
+ else
+ threadid = threadref_to_int(&thref);
+ for (nothreads = 0;
+ nothreads < maxthreads &&
+ threadid < PID_MAX + MAX_NO_CPUS;
+ threadid++) {
+ thread = getthread(threadid);
+ if (thread) {
+ int_to_threadref(&thref,
+ threadid);
+ pack_threadid(remcomOutBuffer +
+ 21 +
+ nothreads * 16,
+ &thref);
+ nothreads++;
+ if (thread_min > threadid)
+ thread_min = threadid;
+ }
+ }
+
+ if (threadid == PID_MAX + MAX_NO_CPUS) {
+ remcomOutBuffer[4] = '1';
+ }
+ pack_hex_byte(remcomOutBuffer + 2, nothreads);
+ remcomOutBuffer[21 + nothreads * 16] = '\0';
+ break;
+#endif
+ case 'C':
+ /* Current thread id */
+ remcomOutBuffer[0] = 'Q';
+ remcomOutBuffer[1] = 'C';
+ threadid = current->pid;
+ if (!threadid) {
+ /*
+ * idle thread
+ */
+ for (threadid = PID_MAX;
+ threadid < PID_MAX + MAX_NO_CPUS;
+ threadid++) {
+ if (current ==
+ idle_task(threadid -
+ PID_MAX))
+ break;
+ }
+ }
+ int_to_threadref(&thref, threadid);
+ pack_threadid(remcomOutBuffer + 2, &thref);
+ remcomOutBuffer[18] = '\0';
+ break;
+
+ case 'E':
+ /* Print exception info */
+ printexceptioninfo(exceptionVector,
+ err_code, remcomOutBuffer);
+ break;
+ case 'T':
+ ptr = &remcomInBuffer[0];
+ if (strncmp(ptr, "qThreadExtraInfo,",
+ strlen("qThreadExtraInfo,")) == 0) {
+ ptr += strlen("qThreadExtraInfo,");
+ hexToLong(&ptr, &tmpid);
+ p = getthread(tmpid);
+ print_extra_info(p, lbuf);
+ mem2hex(lbuf, remcomOutBuffer,
+ strlen(lbuf));
+ }
+ break;
+#if 0
+ case 'T':{
+ char * nptr;
+ /* Thread extra info */
+ if (!cmp_str(&remcomInBuffer[2],
+ "hreadExtraInfo,", 15)) {
+ break;
+ }
+ ptr = &remcomInBuffer[17];
+ hexToLong(&ptr, &threadid);
+ thread = getthread(threadid);
+ nptr = &thread->comm[0];
+ length = 0;
+ ptr = &remcomOutBuffer[0];
+ do {
+ length++;
+ ptr = pack_hex_byte(ptr, *nptr++);
+ } while (*nptr && length < 16);
+ /*
+ * would like that 16 to be the size of
+ * task_struct.comm but don't know the
+ * syntax..
+ */
+ *ptr = 0;
+ }
+#endif
+ }
+ break;
+
+ /* task related */
+ case 'H':
+ switch (remcomInBuffer[1]) {
+ case 'g':
+ ptr = &remcomInBuffer[2];
+ hexToLong(&ptr, &threadid);
+ thread = getthread(threadid);
+ if (!thread) {
+ remcomOutBuffer[0] = 'E';
+ remcomOutBuffer[1] = '\0';
+ break;
+ }
+ /*
+ * Just in case I forget what this is all about,
+ * the "thread info" command to gdb causes it
+ * to ask for a thread list. It then switches
+ * to each thread and asks for the registers.
+ * For this (and only this) usage, we want to
+ * fudge the registers of tasks not on the run
+ * list (i.e. waiting) to show the routine that
+ * called schedule. Also, gdb, is a minimalist
+ * in that if the current thread is the last
+ * it will not re-read the info when done.
+ * This means that in this case we must show
+ * the real registers. So here is how we do it:
+ * Each entry we keep track of the min
+ * thread in the list (the last that gdb will)
+ * get info for. We also keep track of the
+ * starting thread.
+ * "thread_list" is cleared when switching back
+ * to the min thread if it is was current, or
+ * if it was not current, thread_list is set
+ * to 1. When the switch to current comes,
+ * if thread_list is 1, clear it, else do
+ * nothing.
+ */
+ usethread = thread;
+ if ((thread_list == 1) &&
+ (thread == thread_list_start)) {
+ thread_list = 0;
+ }
+ if (thread_list && (threadid == thread_min)) {
+ if (thread == thread_list_start) {
+ thread_list = 0;
+ } else {
+ thread_list = 1;
+ }
+ }
+ /* follow through */
+ case 'c':
+ remcomOutBuffer[0] = 'O';
+ remcomOutBuffer[1] = 'K';
+ remcomOutBuffer[2] = '\0';
+ break;
+ }
+ break;
+
+ /* Query thread status */
+ case 'T':
+ ptr = &remcomInBuffer[1];
+ hexToLong(&ptr, &threadid);
+ thread = getthread(threadid);
+ if (thread) {
+ remcomOutBuffer[0] = 'O';
+ remcomOutBuffer[1] = 'K';
+ remcomOutBuffer[2] = '\0';
+ if (thread_min > threadid)
+ thread_min = threadid;
+ } else {
+ remcomOutBuffer[0] = 'E';
+ remcomOutBuffer[1] = '\0';
+ }
+ break;
+
+ case 'Y': /* set up a hardware breakpoint */
+ ptr = &remcomInBuffer[1];
+ hexToLong(&ptr, &breakno);
+ ptr++;
+ hexToLong(&ptr, &breaktype);
+ ptr++;
+ hexToLong(&ptr, &length);
+ ptr++;
+ hexToLong(&ptr, &addr);
+ if (set_hw_break(breakno & 0x3,
+ breaktype & 0x3,
+ length & 0x3, addr) == 0) {
+ strcpy(remcomOutBuffer, "OK");
+ } else {
+ strcpy(remcomOutBuffer, "ERROR");
+ }
+ break;
+
+ /* Remove hardware breakpoint */
+ case 'y':
+ ptr = &remcomInBuffer[1];
+ hexToLong(&ptr, &breakno);
+ if (remove_hw_break(breakno & 0x3) == 0) {
+ strcpy(remcomOutBuffer, "OK");
+ } else {
+ strcpy(remcomOutBuffer, "ERROR");
+ }
+ break;
+
+ case 'r': /* reboot */
+ strcpy(remcomOutBuffer, "OK");
+ putpacket(remcomOutBuffer);
+ /*to_gdb("Rebooting\n"); */
+ /* triplefault no return from here */
+ {
+ static long no_idt[2];
+ __asm__ __volatile__("lidt %0"::"m"(no_idt[0]));
+ BREAKPOINT;
+ }
+
+ } /* switch */
+
+ /* reply to the request */
+ putpacket(remcomOutBuffer);
+ } /* while(1==1) */
+ /*
+ * reached by goto only.
+ */
+ exit_kgdb:
+ /*
+ * Here is where we set up to trap a gdb function call. NEW_esp
+ * will be changed if we are trying to do this. We handle both
+ * adding and subtracting, thus allowing gdb to put grung on
+ * the stack which it removes later.
+ */
+ if (NEW_esp != OLD_esp) {
+ unsigned long *ptr = END_OF_LOOKASIDE;
+ if (NEW_esp < OLD_esp)
+ ptr -= (OLD_esp - NEW_esp) / sizeof (unsigned long);
+ *--ptr = linux_regs->eflags;
+ *--ptr = linux_regs->cs;
+ *--ptr = linux_regs->rip;
+ *--ptr = linux_regs->rcx;
+ *--ptr = linux_regs->rbx;
+ *--ptr = linux_regs->rax;
+ linux_regs->rcx = NEW_esp - (sizeof (unsigned long) * 6);
+ linux_regs->rbx = (unsigned long) END_OF_LOOKASIDE;
+ if (NEW_esp < OLD_esp) {
+ linux_regs->rip = (unsigned long) fn_call_stub;
+ } else {
+ linux_regs->rip = (unsigned long) fn_rtn_stub;
+ linux_regs->rax = NEW_esp;
+ }
+ linux_regs->eflags &= ~(IF_BIT | TF_BIT);
+ }
+#ifdef CONFIG_SMP
+ /*
+ * Release gdb wait locks
+ * Sanity check time. Must have at least one cpu to run. Also single
+ * step must not be done if the current cpu is on hold.
+ */
+ if (spinlock_count == 1) {
+ int ss_hold = (regs.eflags & 0x100) && kgdb_info.hold_on_sstep;
+ int cpu_avail = 0;
+ int i;
+
+ for (i = 0; i < MAX_NO_CPUS; i++) {
+ if (!cpu_online(i))
+ break;
+ if (!hold_cpu(i)) {
+ cpu_avail = 1;
+ }
+ }
+ /*
+ * Early in the bring up there will be NO cpus on line...
+ */
+ if (!cpu_avail && !cpus_empty(cpu_online_map)) {
+ to_gdb("No cpus unblocked, see 'kgdb_info.hold_cpu'\n");
+ goto once_again;
+ }
+ if (hold_cpu(smp_processor_id()) && (regs.eflags & 0x100)) {
+ to_gdb
+ ("Current cpu must be unblocked to single step\n");
+ goto once_again;
+ }
+ if (!(ss_hold)) {
+ int i;
+ for (i = 0; i < MAX_NO_CPUS; i++) {
+ if (!hold_cpu(i)) {
+ spin_unlock(&waitlocks[i]);
+ }
+ }
+ } else {
+ spin_unlock(&waitlocks[smp_processor_id()]);
+ }
+ /* Release kgdb spinlock */
+ KGDB_SPIN_UNLOCK(&kgdb_spinlock);
+ /*
+ * If this cpu is on hold, this is where we
+ * do it. Note, the NMI will pull us out of here,
+ * but will return as the above lock is not held.
+ * We will stay here till another cpu releases the lock for us.
+ */
+ spin_unlock_wait(waitlocks + smp_processor_id());
+ local_irq_restore(flags);
+ return (1);
+ }
+#if 0
+exit_just_unlock:
+#endif
+#endif
+ /* Release kgdb spinlock */
+ KGDB_SPIN_UNLOCK(&kgdb_spinlock);
+ local_irq_restore(flags);
+ return (1);
+}
+
+#undef regs
+static int kgdb_notify(struct notifier_block *self, unsigned long cmd, void *ptr)
+{
+ struct die_args *d = ptr;
+
+ if (!kgdb_enabled || (cmd == DIE_DEBUG && user_mode(d->regs)))
+ return NOTIFY_DONE;
+ if (cmd == DIE_NMI_IPI) {
+ if (in_kgdb(d->regs))
+ return NOTIFY_BAD;
+ } else if (kgdb_handle_exception(d->trapnr, d->signr, d->err, d->regs))
+ return NOTIFY_BAD; /* skip */
+
+ return NOTIFY_DONE;
+}
+
+static struct notifier_block kgdb_notifier = {
+ .notifier_call = kgdb_notify,
+ .priority = 0,
+};
+
+void set_debug_traps(void)
+{
+ static int initialized = 0;
+
+ if (!initialized) {
+ initialized = 1;
+ notifier_chain_register(&die_chain, &kgdb_notifier);
+ }
+}
+
+/*
+ * Provide the command line "gdb" initial break
+ */
+int __init kgdb_initial_break(char * str)
+{
+ if (*str == '\0'){
+ breakpoint();
+ return 1;
+ }
+ return 0;
+}
+__setup("gdb",kgdb_initial_break);
+
+/* This function will generate a breakpoint exception. It is used at the
+ beginning of a program to sync up with a debugger and can be used
+ otherwise as a quick means to stop program execution and "break" into
+ the debugger. */
+/* But really, just use the BREAKPOINT macro. We will handle the int stuff
+ */
+
+void breakpoint(void)
+{
+
+ set_debug_traps();
+ kgdb_enabled = 1;
+#if 0
+ /*
+ * These calls were not enough to allow breakpoint to be
+ * called before trap_init(). I moved the argument parsing
+ * after trap_init() and it seems to work.
+ */
+ set_intr_usr_gate(3,&int3); /* disable ints on trap */
+ set_intr_gate(1,&debug);
+ set_intr_gate(14,&page_fault);
+#endif
+
+ BREAKPOINT;
+}
+
+#ifdef later
+/*
+ * possibly we should not go thru the traps.c code at all? Someday.
+ */
+void
+do_kgdb_int3(struct pt_regs *regs, long error_code)
+{
+ kgdb_handle_exception(3, 5, error_code, regs);
+ return;
+}
+#endif
+#undef regs
+#ifdef CONFIG_TRAP_BAD_SYSCALL_EXITS
+asmlinkage void
+bad_sys_call_exit(int stuff)
+{
+ struct pt_regs *regs = (struct pt_regs *) &stuff;
+ printk("Sys call %d return with %x preempt_count\n",
+ (int) regs->orig_eax, preempt_count());
+}
+#endif
+#ifdef CONFIG_STACK_OVERFLOW_TEST
+#include <asm/kgdb.h>
+asmlinkage void
+stack_overflow(void)
+{
+#ifdef BREAKPOINT
+ BREAKPOINT;
+#else
+ printk("Kernel stack overflow, looping forever\n");
+#endif
+ while (1) {
+ }
+}
+#endif
+
+#if defined(CONFIG_SMP) || defined(CONFIG_KGDB_CONSOLE)
+char gdbconbuf[BUFMAX];
+
+static void
+kgdb_gdb_message(const char *s, unsigned count)
+{
+ int i;
+ int wcount;
+ char *bufptr;
+ /*
+ * This takes care of NMI while spining out chars to gdb
+ */
+ IF_SMP(in_kgdb_console = 1);
+ gdbconbuf[0] = 'O';
+ bufptr = gdbconbuf + 1;
+ while (count > 0) {
+ if ((count << 1) > (BUFMAX - 2)) {
+ wcount = (BUFMAX - 2) >> 1;
+ } else {
+ wcount = count;
+ }
+ count -= wcount;
+ for (i = 0; i < wcount; i++) {
+ bufptr = pack_hex_byte(bufptr, s[i]);
+ }
+ *bufptr = '\0';
+ s += wcount;
+
+ putpacket(gdbconbuf);
+
+ }
+ IF_SMP(in_kgdb_console = 0);
+}
+#endif
+#ifdef CONFIG_SMP
+static void
+to_gdb(const char *s)
+{
+ int count = 0;
+ while (s[count] && (count++ < BUFMAX)) ;
+ kgdb_gdb_message(s, count);
+}
+#endif
+#ifdef CONFIG_KGDB_CONSOLE
+#include <linux/console.h>
+#include <linux/init.h>
+#include <linux/fs.h>
+#include <asm/semaphore.h>
+
+void
+kgdb_console_write(struct console *co, const char *s, unsigned count)
+{
+
+ if (gdb_i386vector == -1) {
+ /*
+ * We have not yet talked to gdb. What to do...
+ * lets break, on continue we can do the write.
+ * But first tell him whats up. Uh, well no can do,
+ * as this IS the console. Oh well...
+ * We do need to wait or the messages will be lost.
+ * Other option would be to tell the above code to
+ * ignore this breakpoint and do an auto return,
+ * but that might confuse gdb. Also this happens
+ * early enough in boot up that we don't have the traps
+ * set up yet, so...
+ */
+ breakpoint();
+ }
+ kgdb_gdb_message(s, count);
+}
+
+/*
+ * ------------------------------------------------------------
+ * Serial KGDB driver
+ * ------------------------------------------------------------
+ */
+
+static struct console kgdbcons = {
+ name:"kgdb",
+ write:kgdb_console_write,
+#ifdef CONFIG_KGDB_USER_CONSOLE
+ device:kgdb_console_device,
+#endif
+ flags:CON_PRINTBUFFER | CON_ENABLED,
+ index:-1,
+};
+
+/*
+ * The trick here is that this file gets linked before printk.o
+ * That means we get to peer at the console info in the command
+ * line before it does. If we are up, we register, otherwise,
+ * do nothing. By returning 0, we allow printk to look also.
+ */
+static int kgdb_console_enabled;
+
+int __init
+kgdb_console_init(char *str)
+{
+ if ((strncmp(str, "kgdb", 4) == 0) || (strncmp(str, "gdb", 3) == 0)) {
+ register_console(&kgdbcons);
+ kgdb_console_enabled = 1;
+ }
+ return 0; /* let others look at the string */
+}
+
+__setup("console=", kgdb_console_init);
+
+#ifdef CONFIG_KGDB_USER_CONSOLE
+static kdev_t kgdb_console_device(struct console *c);
+/* This stuff sort of works, but it knocks out telnet devices
+ * we are leaving it here in case we (or you) find time to figure it out
+ * better..
+ */
+
+/*
+ * We need a real char device as well for when the console is opened for user
+ * space activities.
+ */
+
+static int
+kgdb_consdev_open(struct inode *inode, struct file *file)
+{
+ return 0;
+}
+
+static ssize_t
+kgdb_consdev_write(struct file *file, const char *buf,
+ size_t count, loff_t * ppos)
+{
+ int size, ret = 0;
+ static char kbuf[128];
+ static DECLARE_MUTEX(sem);
+
+ /* We are not reentrant... */
+ if (down_interruptible(&sem))
+ return -ERESTARTSYS;
+
+ while (count > 0) {
+ /* need to copy the data from user space */
+ size = count;
+ if (size > sizeof (kbuf))
+ size = sizeof (kbuf);
+ if (copy_from_user(kbuf, buf, size)) {
+ ret = -EFAULT;
+ break;;
+ }
+ kgdb_console_write(&kgdbcons, kbuf, size);
+ count -= size;
+ ret += size;
+ buf += size;
+ }
+
+ up(&sem);
+
+ return ret;
+}
+
+struct file_operations kgdb_consdev_fops = {
+ open:kgdb_consdev_open,
+ write:kgdb_consdev_write
+};
+static kdev_t
+kgdb_console_device(struct console *c)
+{
+ return MKDEV(TTYAUX_MAJOR, 1);
+}
+
+/*
+ * This routine gets called from the serial stub in the i386/lib
+ * This is so it is done late in bring up (just before the console open).
+ */
+void
+kgdb_console_finit(void)
+{
+ if (kgdb_console_enabled) {
+ char *cptr = cdevname(MKDEV(TTYAUX_MAJOR, 1));
+ char *cp = cptr;
+ while (*cptr && *cptr != '(')
+ cptr++;
+ *cptr = 0;
+ unregister_chrdev(TTYAUX_MAJOR, cp);
+ register_chrdev(TTYAUX_MAJOR, "kgdb", &kgdb_consdev_fops);
+ }
+}
+#endif
+#endif
+#ifdef CONFIG_KGDB_TS
+#include <asm/msr.h> /* time stamp code */
+#include <asm/hardirq.h> /* in_interrupt */
+#ifdef CONFIG_KGDB_TS_64
+#define DATA_POINTS 64
+#endif
+#ifdef CONFIG_KGDB_TS_128
+#define DATA_POINTS 128
+#endif
+#ifdef CONFIG_KGDB_TS_256
+#define DATA_POINTS 256
+#endif
+#ifdef CONFIG_KGDB_TS_512
+#define DATA_POINTS 512
+#endif
+#ifdef CONFIG_KGDB_TS_1024
+#define DATA_POINTS 1024
+#endif
+#ifndef DATA_POINTS
+#define DATA_POINTS 128 /* must be a power of two */
+#endif
+#define INDEX_MASK (DATA_POINTS - 1)
+#if (INDEX_MASK & DATA_POINTS)
+#error "CONFIG_KGDB_TS_COUNT must be a power of 2"
+#endif
+struct kgdb_and_then_struct {
+#ifdef CONFIG_SMP
+ int on_cpu;
+#endif
+ struct task_struct *task;
+ long long at_time;
+ int from_ln;
+ char *in_src;
+ void *from;
+ int *with_shpf;
+ int data0;
+ int data1;
+};
+struct kgdb_and_then_struct2 {
+#ifdef CONFIG_SMP
+ int on_cpu;
+#endif
+ struct task_struct *task;
+ long long at_time;
+ int from_ln;
+ char *in_src;
+ void *from;
+ int *with_shpf;
+ struct task_struct *t1;
+ struct task_struct *t2;
+};
+struct kgdb_and_then_struct kgdb_data[DATA_POINTS];
+
+struct kgdb_and_then_struct *kgdb_and_then = &kgdb_data[0];
+int kgdb_and_then_count;
+
+void
+kgdb_tstamp(int line, char *source, int data0, int data1)
+{
+ static spinlock_t ts_spin = SPIN_LOCK_UNLOCKED;
+ int flags;
+ local_irq_save(flags);
+ spin_lock(&ts_spin);
+ rdtscll(kgdb_and_then->at_time);
+#ifdef CONFIG_SMP
+ kgdb_and_then->on_cpu = smp_processor_id();
+#endif
+ kgdb_and_then->task = current;
+ kgdb_and_then->from_ln = line;
+ kgdb_and_then->in_src = source;
+ kgdb_and_then->from = __builtin_return_address(0);
+ kgdb_and_then->with_shpf = (int *) (((flags & IF_BIT) >> 9) |
+ (preempt_count() << 8));
+ kgdb_and_then->data0 = data0;
+ kgdb_and_then->data1 = data1;
+ kgdb_and_then = &kgdb_data[++kgdb_and_then_count & INDEX_MASK];
+ spin_unlock(&ts_spin);
+ local_irq_restore(flags);
+#ifdef CONFIG_PREEMPT
+
+#endif
+ return;
+}
+#endif
+typedef int gdb_debug_hook(int exceptionVector,
+ int signo, int err_code, struct pt_regs *linux_regs);
+gdb_debug_hook *linux_debug_hook = &kgdb_handle_exception; /* histerical reasons... */
+
+static int kgdb_need_breakpoint[NR_CPUS];
+
+void kgdb_schedule_breakpoint(void)
+{
+ kgdb_need_breakpoint[smp_processor_id()] = 1;
+}
+
+void kgdb_process_breakpoint(void)
+{
+ /*
+ * Handle a breakpoint queued from inside network driver code
+ * to avoid reentrancy issues
+ */
+ if (kgdb_need_breakpoint[smp_processor_id()]) {
+ kgdb_need_breakpoint[smp_processor_id()] = 0;
+ kgdb_enabled = 1;
+ BREAKPOINT;
+ }
+}
+
--- diff/arch/x86_64/lib/kgdb_serial.c 1970-01-01 00:00:00.000000000 +0000
+++ source/arch/x86_64/lib/kgdb_serial.c 2004-03-16 11:27:47.973894520 +0000
@@ -0,0 +1,490 @@
+/*
+ * Serial interface GDB stub
+ *
+ * Written (hacked together) by David Grothe (dave@gcom.com)
+ * Modified to allow invokation early in boot see also
+ * kgdb.h for instructions by George Anzinger(george@mvista.com)
+ * Modified to handle debugging over ethernet by Robert Walsh
+ * <rjwalsh@durables.org> and wangdi <wangdi@clusterfs.com>, based on
+ * code by San Mehat.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/timer.h>
+#include <linux/interrupt.h>
+#include <linux/tty.h>
+#include <linux/tty_flip.h>
+#include <linux/serial.h>
+#include <linux/serial_reg.h>
+#include <linux/config.h>
+#include <linux/major.h>
+#include <linux/string.h>
+#include <linux/fcntl.h>
+#include <linux/ptrace.h>
+#include <linux/ioport.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/highmem.h>
+#include <asm/system.h>
+#include <asm/io.h>
+#include <asm/segment.h>
+#include <asm/bitops.h>
+#include <asm/system.h>
+#include <asm/kgdb_local.h>
+#ifdef CONFIG_KGDB_USER_CONSOLE
+extern void kgdb_console_finit(void);
+#endif
+#define PRNT_off
+#define TEST_EXISTANCE
+#ifdef PRNT
+#define dbprintk(s) printk s
+#else
+#define dbprintk(s)
+#endif
+#define TEST_INTERRUPT_off
+#ifdef TEST_INTERRUPT
+#define intprintk(s) printk s
+#else
+#define intprintk(s)
+#endif
+
+#define IRQ_T(info) ((info->flags & ASYNC_SHARE_IRQ) ? SA_SHIRQ : SA_INTERRUPT)
+
+#define GDB_BUF_SIZE 512 /* power of 2, please */
+
+static char gdb_buf[GDB_BUF_SIZE];
+static int gdb_buf_in_inx;
+static atomic_t gdb_buf_in_cnt;
+static int gdb_buf_out_inx;
+
+struct async_struct *gdb_async_info;
+static int gdb_async_irq;
+
+#define outb_px(a,b) outb_p(b,a)
+
+static void program_uart(struct async_struct *info);
+static void write_char(struct async_struct *info, int chr);
+/*
+ * Get a byte from the hardware data buffer and return it
+ */
+static int
+read_data_bfr(struct async_struct *info)
+{
+ char it = inb_p(info->port + UART_LSR);
+
+ if (it & UART_LSR_DR)
+ return (inb_p(info->port + UART_RX));
+ /*
+ * If we have a framing error assume somebody messed with
+ * our uart. Reprogram it and send '-' both ways...
+ */
+ if (it & 0xc) {
+ program_uart(info);
+ write_char(info, '-');
+ return ('-');
+ }
+ return (-1);
+
+} /* read_data_bfr */
+
+/*
+ * Get a char if available, return -1 if nothing available.
+ * Empty the receive buffer first, then look at the interface hardware.
+
+ * Locking here is a bit of a problem. We MUST not lock out communication
+ * if we are trying to talk to gdb about a kgdb entry. ON the other hand
+ * we can loose chars in the console pass thru if we don't lock. It is also
+ * possible that we could hold the lock or be waiting for it when kgdb
+ * NEEDS to talk. Since kgdb locks down the world, it does not need locks.
+ * We do, of course have possible issues with interrupting a uart operation,
+ * but we will just depend on the uart status to help keep that straight.
+
+ */
+static spinlock_t uart_interrupt_lock = SPIN_LOCK_UNLOCKED;
+#ifdef CONFIG_SMP
+extern spinlock_t kgdb_spinlock;
+#endif
+
+static int
+read_char(struct async_struct *info)
+{
+ int chr;
+ unsigned long flags;
+ local_irq_save(flags);
+#ifdef CONFIG_SMP
+ if (!spin_is_locked(&kgdb_spinlock)) {
+ spin_lock(&uart_interrupt_lock);
+ }
+#endif
+ if (atomic_read(&gdb_buf_in_cnt) != 0) { /* intr routine has q'd chars */
+ chr = gdb_buf[gdb_buf_out_inx++];
+ gdb_buf_out_inx &= (GDB_BUF_SIZE - 1);
+ atomic_dec(&gdb_buf_in_cnt);
+ } else {
+ chr = read_data_bfr(info);
+ }
+#ifdef CONFIG_SMP
+ if (!spin_is_locked(&kgdb_spinlock)) {
+ spin_unlock(&uart_interrupt_lock);
+ }
+#endif
+ local_irq_restore(flags);
+ return (chr);
+}
+
+/*
+ * Wait until the interface can accept a char, then write it.
+ */
+static void
+write_char(struct async_struct *info, int chr)
+{
+ while (!(inb_p(info->port + UART_LSR) & UART_LSR_THRE)) ;
+
+ outb_p(chr, info->port + UART_TX);
+
+} /* write_char */
+
+/*
+ * Mostly we don't need a spinlock, but since the console goes
+ * thru here with interrutps on, well, we need to catch those
+ * chars.
+ */
+/*
+ * This is the receiver interrupt routine for the GDB stub.
+ * It will receive a limited number of characters of input
+ * from the gdb host machine and save them up in a buffer.
+ *
+ * When the gdb stub routine tty_getDebugChar() is called it
+ * draws characters out of the buffer until it is empty and
+ * then reads directly from the serial port.
+ *
+ * We do not attempt to write chars from the interrupt routine
+ * since the stubs do all of that via tty_putDebugChar() which
+ * writes one byte after waiting for the interface to become
+ * ready.
+ *
+ * The debug stubs like to run with interrupts disabled since,
+ * after all, they run as a consequence of a breakpoint in
+ * the kernel.
+ *
+ * Perhaps someone who knows more about the tty driver than I
+ * care to learn can make this work for any low level serial
+ * driver.
+ */
+static irqreturn_t
+gdb_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+{
+ struct async_struct *info;
+ unsigned long flags;
+
+ info = gdb_async_info;
+ if (!info || !info->tty || irq != gdb_async_irq)
+ return IRQ_NONE;
+
+ local_irq_save(flags);
+ spin_lock(&uart_interrupt_lock);
+ do {
+ int chr = read_data_bfr(info);
+ intprintk(("Debug char on int: %x hex\n", chr));
+ if (chr < 0)
+ continue;
+
+ if (chr == 3) { /* Ctrl-C means remote interrupt */
+ BREAKPOINT;
+ continue;
+ }
+
+ if (atomic_read(&gdb_buf_in_cnt) >= GDB_BUF_SIZE) {
+ /* buffer overflow tosses early char */
+ read_char(info);
+ }
+ gdb_buf[gdb_buf_in_inx++] = chr;
+ gdb_buf_in_inx &= (GDB_BUF_SIZE - 1);
+ } while (inb_p(info->port + UART_IIR) & UART_IIR_RDI);
+ spin_unlock(&uart_interrupt_lock);
+ local_irq_restore(flags);
+ return IRQ_HANDLED;
+} /* gdb_interrupt */
+
+/*
+ * Just a NULL routine for testing.
+ */
+void
+gdb_null(void)
+{
+} /* gdb_null */
+
+/* These structure are filled in with values defined in asm/kgdb_local.h
+ */
+static struct serial_state state = SB_STATE;
+static struct async_struct local_info = SB_INFO;
+static int ok_to_enable_ints = 0;
+static void kgdb_enable_ints_now(void);
+
+extern char *kgdb_version;
+/*
+ * Hook an IRQ for KGDB.
+ *
+ * This routine is called from tty_putDebugChar, below.
+ */
+static int ints_disabled = 1;
+int
+gdb_hook_interrupt(struct async_struct *info, int verb)
+{
+ struct serial_state *state = info->state;
+ unsigned long flags;
+ int port;
+#ifdef TEST_EXISTANCE
+ int scratch, scratch2;
+#endif
+
+ /* The above fails if memory managment is not set up yet.
+ * Rather than fail the set up, just keep track of the fact
+ * and pick up the interrupt thing later.
+ */
+ gdb_async_info = info;
+ port = gdb_async_info->port;
+ gdb_async_irq = state->irq;
+ if (verb) {
+ printk("kgdb %s : port =%x, IRQ=%d, divisor =%d\n",
+ kgdb_version,
+ port,
+ gdb_async_irq, gdb_async_info->state->custom_divisor);
+ }
+ local_irq_save(flags);
+#ifdef TEST_EXISTANCE
+ /* Existance test */
+ /* Should not need all this, but just in case.... */
+
+ scratch = inb_p(port + UART_IER);
+ outb_px(port + UART_IER, 0);
+ outb_px(0xff, 0x080);
+ scratch2 = inb_p(port + UART_IER);
+ outb_px(port + UART_IER, scratch);
+ if (scratch2) {
+ printk
+ ("gdb_hook_interrupt: Could not clear IER, not a UART!\n");
+ local_irq_restore(flags);
+ return 1; /* We failed; there's nothing here */
+ }
+ scratch2 = inb_p(port + UART_LCR);
+ outb_px(port + UART_LCR, 0xBF); /* set up for StarTech test */
+ outb_px(port + UART_EFR, 0); /* EFR is the same as FCR */
+ outb_px(port + UART_LCR, 0);
+ outb_px(port + UART_FCR, UART_FCR_ENABLE_FIFO);
+ scratch = inb_p(port + UART_IIR) >> 6;
+ if (scratch == 1) {
+ printk("gdb_hook_interrupt: Undefined UART type!"
+ " Not a UART! \n");
+ local_irq_restore(flags);
+ return 1;
+ } else {
+ dbprintk(("gdb_hook_interrupt: UART type "
+ "is %d where 0=16450, 2=16550 3=16550A\n", scratch));
+ }
+ scratch = inb_p(port + UART_MCR);
+ outb_px(port + UART_MCR, UART_MCR_LOOP | scratch);
+ outb_px(port + UART_MCR, UART_MCR_LOOP | 0x0A);
+ scratch2 = inb_p(port + UART_MSR) & 0xF0;
+ outb_px(port + UART_MCR, scratch);
+ if (scratch2 != 0x90) {
+ printk("gdb_hook_interrupt: "
+ "Loop back test failed! Not a UART!\n");
+ local_irq_restore(flags);
+ return scratch2 + 1000; /* force 0 to fail */
+ }
+#endif /* test existance */
+ program_uart(info);
+ local_irq_restore(flags);
+
+ return (0);
+
+} /* gdb_hook_interrupt */
+
+static void
+program_uart(struct async_struct *info)
+{
+ int port = info->port;
+
+ (void) inb_p(port + UART_RX);
+ outb_px(port + UART_IER, 0);
+
+ (void) inb_p(port + UART_RX); /* serial driver comments say */
+ (void) inb_p(port + UART_IIR); /* this clears the interrupt regs */
+ (void) inb_p(port + UART_MSR);
+ outb_px(port + UART_LCR, UART_LCR_WLEN8 | UART_LCR_DLAB);
+ outb_px(port + UART_DLL, info->state->custom_divisor & 0xff); /* LS */
+ outb_px(port + UART_DLM, info->state->custom_divisor >> 8); /* MS */
+ outb_px(port + UART_MCR, info->MCR);
+
+ outb_px(port + UART_FCR, UART_FCR_ENABLE_FIFO | UART_FCR_TRIGGER_1 | UART_FCR_CLEAR_XMIT | UART_FCR_CLEAR_RCVR); /* set fcr */
+ outb_px(port + UART_LCR, UART_LCR_WLEN8); /* reset DLAB */
+ outb_px(port + UART_FCR, UART_FCR_ENABLE_FIFO | UART_FCR_TRIGGER_1); /* set fcr */
+ if (!ints_disabled) {
+ intprintk(("KGDB: Sending %d to port %x offset %d\n",
+ gdb_async_info->IER,
+ (int) gdb_async_info->port, UART_IER));
+ outb_px(gdb_async_info->port + UART_IER, gdb_async_info->IER);
+ }
+ return;
+}
+
+/*
+ * tty_getDebugChar
+ *
+ * This is a GDB stub routine. It waits for a character from the
+ * serial interface and then returns it. If there is no serial
+ * interface connection then it returns a bogus value which will
+ * almost certainly cause the system to hang. In the
+ */
+int kgdb_in_isr = 0;
+int kgdb_in_lsr = 0;
+extern spinlock_t kgdb_spinlock;
+
+/* Caller takes needed protections */
+
+int
+tty_getDebugChar(void)
+{
+ volatile int chr, dum, time, end_time;
+
+ dbprintk(("tty_getDebugChar(port %x): ", gdb_async_info->port));
+
+ if (gdb_async_info == NULL) {
+ gdb_hook_interrupt(&local_info, 0);
+ }
+ /*
+ * This trick says if we wait a very long time and get
+ * no char, return the -1 and let the upper level deal
+ * with it.
+ */
+ rdtsc(dum, time);
+ end_time = time + 2;
+ while (((chr = read_char(gdb_async_info)) == -1) &&
+ (end_time - time) > 0) {
+ rdtsc(dum, time);
+ };
+ /*
+ * This covers our butts if some other code messes with
+ * our uart, hay, it happens :o)
+ */
+ if (chr == -1)
+ program_uart(gdb_async_info);
+
+ dbprintk(("%c\n", chr > ' ' && chr < 0x7F ? chr : ' '));
+ return (chr);
+
+} /* tty_getDebugChar */
+
+static int count = 3;
+static spinlock_t one_at_atime = SPIN_LOCK_UNLOCKED;
+
+static int __init
+kgdb_enable_ints(void)
+{
+ set_debug_traps();
+ if (kgdboe) {
+ return 0;
+ }
+ if (gdb_async_info == NULL) {
+ gdb_hook_interrupt(&local_info, 1);
+ }
+ ok_to_enable_ints = 1;
+ kgdb_enable_ints_now();
+#ifdef CONFIG_KGDB_USER_CONSOLE
+ kgdb_console_finit();
+#endif
+ return 0;
+}
+
+#ifdef CONFIG_SERIAL_8250
+void shutdown_for_kgdb(struct async_struct *gdb_async_info);
+#endif
+
+#define kgdb_mem_init_done() (1)
+
+static void
+kgdb_enable_ints_now(void)
+{
+ if (!spin_trylock(&one_at_atime))
+ return;
+ if (!ints_disabled)
+ goto exit;
+ if (kgdb_mem_init_done() &&
+ ints_disabled) { /* don't try till mem init */
+#ifdef CONFIG_SERIAL_8250
+ /*
+ * The ifdef here allows the system to be configured
+ * without the serial driver.
+ * Don't make it a module, however, it will steal the port
+ */
+ shutdown_for_kgdb(gdb_async_info);
+#endif
+ ints_disabled = request_irq(gdb_async_info->state->irq,
+ gdb_interrupt,
+ IRQ_T(gdb_async_info),
+ "KGDB-stub", NULL);
+ intprintk(("KGDB: request_irq returned %d\n", ints_disabled));
+ }
+ if (!ints_disabled) {
+ intprintk(("KGDB: Sending %d to port %x offset %d\n",
+ gdb_async_info->IER,
+ (int) gdb_async_info->port, UART_IER));
+ outb_px(gdb_async_info->port + UART_IER, gdb_async_info->IER);
+ }
+ exit:
+ spin_unlock(&one_at_atime);
+}
+
+/*
+ * tty_putDebugChar
+ *
+ * This is a GDB stub routine. It waits until the interface is ready
+ * to transmit a char and then sends it. If there is no serial
+ * interface connection then it simply returns to its caller, having
+ * pretended to send the char. Caller takes needed protections.
+ */
+void
+tty_putDebugChar(int chr)
+{
+ dbprintk(("tty_putDebugChar(port %x): chr=%02x '%c', ints_on=%d\n",
+ gdb_async_info->port,
+ chr,
+ chr > ' ' && chr < 0x7F ? chr : ' ', ints_disabled ? 0 : 1));
+
+ if (gdb_async_info == NULL) {
+ gdb_hook_interrupt(&local_info, 0);
+ }
+
+ write_char(gdb_async_info, chr); /* this routine will wait */
+ count = (chr == '#') ? 0 : count + 1;
+ if ((count == 2)) { /* try to enable after */
+ if (ints_disabled & ok_to_enable_ints)
+ kgdb_enable_ints_now(); /* try to enable after */
+
+ /* We do this a lot because, well we really want to get these
+ * interrupts. The serial driver will clear these bits when it
+ * initializes the chip. Every thing else it does is ok,
+ * but this.
+ */
+ if (!ints_disabled) {
+ outb_px(gdb_async_info->port + UART_IER,
+ gdb_async_info->IER);
+ }
+ }
+
+} /* tty_putDebugChar */
+
+/*
+ * This does nothing for the serial port, since it doesn't buffer.
+ */
+
+void tty_flushDebugChar(void)
+{
+}
+
+module_init(kgdb_enable_ints);
--- diff/drivers/block/cfq-iosched.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/block/cfq-iosched.c 2004-03-16 11:27:48.004889808 +0000
@@ -0,0 +1,707 @@
+/*
+ * linux/drivers/block/cfq-iosched.c
+ *
+ * CFQ, or complete fairness queueing, disk scheduler.
+ *
+ * Based on ideas from a previously unfinished io
+ * scheduler (round robin per-process disk scheduling) and Andrea Arcangeli.
+ *
+ * Copyright (C) 2003 Jens Axboe <axboe@suse.de>
+ */
+#include <linux/kernel.h>
+#include <linux/fs.h>
+#include <linux/blkdev.h>
+#include <linux/elevator.h>
+#include <linux/bio.h>
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/init.h>
+#include <linux/compiler.h>
+#include <linux/hash.h>
+#include <linux/rbtree.h>
+#include <linux/mempool.h>
+
+/*
+ * tunables
+ */
+static int cfq_quantum = 4;
+static int cfq_queued = 8;
+
+#define CFQ_QHASH_SHIFT 6
+#define CFQ_QHASH_ENTRIES (1 << CFQ_QHASH_SHIFT)
+#define list_entry_qhash(entry) list_entry((entry), struct cfq_queue, cfq_hash)
+
+#define CFQ_MHASH_SHIFT 8
+#define CFQ_MHASH_BLOCK(sec) ((sec) >> 3)
+#define CFQ_MHASH_ENTRIES (1 << CFQ_MHASH_SHIFT)
+#define CFQ_MHASH_FN(sec) (hash_long(CFQ_MHASH_BLOCK((sec)),CFQ_MHASH_SHIFT))
+#define ON_MHASH(crq) !list_empty(&(crq)->hash)
+#define rq_hash_key(rq) ((rq)->sector + (rq)->nr_sectors)
+#define list_entry_hash(ptr) list_entry((ptr), struct cfq_rq, hash)
+
+#define list_entry_cfqq(ptr) list_entry((ptr), struct cfq_queue, cfq_list)
+
+#define RQ_DATA(rq) ((struct cfq_rq *) (rq)->elevator_private)
+
+static kmem_cache_t *crq_pool;
+static kmem_cache_t *cfq_pool;
+static mempool_t *cfq_mpool;
+
+struct cfq_data {
+ struct list_head rr_list;
+ struct list_head *dispatch;
+ struct list_head *cfq_hash;
+
+ struct list_head *crq_hash;
+
+ unsigned int busy_queues;
+ unsigned int max_queued;
+
+ mempool_t *crq_pool;
+};
+
+struct cfq_queue {
+ struct list_head cfq_hash;
+ struct list_head cfq_list;
+ struct rb_root sort_list;
+ int pid;
+ int queued[2];
+#if 0
+ /*
+ * with a simple addition like this, we can do io priorities. almost.
+ * does need a split request free list, too.
+ */
+ int io_prio
+#endif
+};
+
+struct cfq_rq {
+ struct rb_node rb_node;
+ sector_t rb_key;
+
+ struct request *request;
+
+ struct cfq_queue *cfq_queue;
+
+ struct list_head hash;
+};
+
+static void cfq_put_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq);
+static struct cfq_queue *cfq_find_cfq_hash(struct cfq_data *cfqd, int pid);
+static void cfq_dispatch_sort(struct list_head *head, struct cfq_rq *crq);
+
+/*
+ * lots of deadline iosched dupes, can be abstracted later...
+ */
+static inline void __cfq_del_crq_hash(struct cfq_rq *crq)
+{
+ list_del_init(&crq->hash);
+}
+
+static inline void cfq_del_crq_hash(struct cfq_rq *crq)
+{
+ if (ON_MHASH(crq))
+ __cfq_del_crq_hash(crq);
+}
+
+static void cfq_remove_merge_hints(request_queue_t *q, struct cfq_rq *crq)
+{
+ cfq_del_crq_hash(crq);
+
+ if (q->last_merge == crq->request)
+ q->last_merge = NULL;
+}
+
+static inline void cfq_add_crq_hash(struct cfq_data *cfqd, struct cfq_rq *crq)
+{
+ struct request *rq = crq->request;
+
+ BUG_ON(ON_MHASH(crq));
+
+ list_add(&crq->hash, &cfqd->crq_hash[CFQ_MHASH_FN(rq_hash_key(rq))]);
+}
+
+static struct request *cfq_find_rq_hash(struct cfq_data *cfqd, sector_t offset)
+{
+ struct list_head *hash_list = &cfqd->crq_hash[CFQ_MHASH_FN(offset)];
+ struct list_head *entry, *next = hash_list->next;
+
+ while ((entry = next) != hash_list) {
+ struct cfq_rq *crq = list_entry_hash(entry);
+ struct request *__rq = crq->request;
+
+ next = entry->next;
+
+ BUG_ON(!ON_MHASH(crq));
+
+ if (!rq_mergeable(__rq)) {
+ __cfq_del_crq_hash(crq);
+ continue;
+ }
+
+ if (rq_hash_key(__rq) == offset)
+ return __rq;
+ }
+
+ return NULL;
+}
+
+/*
+ * rb tree support functions
+ */
+#define RB_NONE (2)
+#define RB_EMPTY(node) ((node)->rb_node == NULL)
+#define RB_CLEAR(node) ((node)->rb_color = RB_NONE)
+#define RB_CLEAR_ROOT(root) ((root)->rb_node = NULL)
+#define ON_RB(node) ((node)->rb_color != RB_NONE)
+#define rb_entry_crq(node) rb_entry((node), struct cfq_rq, rb_node)
+#define rq_rb_key(rq) (rq)->sector
+
+static inline void cfq_del_crq_rb(struct cfq_queue *cfqq, struct cfq_rq *crq)
+{
+ if (ON_RB(&crq->rb_node)) {
+ cfqq->queued[rq_data_dir(crq->request)]--;
+ rb_erase(&crq->rb_node, &cfqq->sort_list);
+ crq->cfq_queue = NULL;
+ }
+}
+
+static struct cfq_rq *
+__cfq_add_crq_rb(struct cfq_queue *cfqq, struct cfq_rq *crq)
+{
+ struct rb_node **p = &cfqq->sort_list.rb_node;
+ struct rb_node *parent = NULL;
+ struct cfq_rq *__crq;
+
+ while (*p) {
+ parent = *p;
+ __crq = rb_entry_crq(parent);
+
+ if (crq->rb_key < __crq->rb_key)
+ p = &(*p)->rb_left;
+ else if (crq->rb_key > __crq->rb_key)
+ p = &(*p)->rb_right;
+ else
+ return __crq;
+ }
+
+ rb_link_node(&crq->rb_node, parent, p);
+ return 0;
+}
+
+static void
+cfq_add_crq_rb(struct cfq_data *cfqd, struct cfq_queue *cfqq,struct cfq_rq *crq)
+{
+ struct request *rq = crq->request;
+ struct cfq_rq *__alias;
+
+ crq->rb_key = rq_rb_key(rq);
+ cfqq->queued[rq_data_dir(rq)]++;
+retry:
+ __alias = __cfq_add_crq_rb(cfqq, crq);
+ if (!__alias) {
+ rb_insert_color(&crq->rb_node, &cfqq->sort_list);
+ crq->cfq_queue = cfqq;
+ return;
+ }
+
+ cfq_del_crq_rb(cfqq, __alias);
+ cfq_dispatch_sort(cfqd->dispatch, __alias);
+ goto retry;
+}
+
+static struct request *
+cfq_find_rq_rb(struct cfq_data *cfqd, sector_t sector)
+{
+ struct cfq_queue *cfqq = cfq_find_cfq_hash(cfqd, current->tgid);
+ struct rb_node *n;
+
+ if (!cfqq)
+ goto out;
+
+ n = cfqq->sort_list.rb_node;
+ while (n) {
+ struct cfq_rq *crq = rb_entry_crq(n);
+
+ if (sector < crq->rb_key)
+ n = n->rb_left;
+ else if (sector > crq->rb_key)
+ n = n->rb_right;
+ else
+ return crq->request;
+ }
+
+out:
+ return NULL;
+}
+
+static void cfq_remove_request(request_queue_t *q, struct request *rq)
+{
+ struct cfq_data *cfqd = q->elevator.elevator_data;
+ struct cfq_rq *crq = RQ_DATA(rq);
+
+ if (crq) {
+ struct cfq_queue *cfqq = crq->cfq_queue;
+
+ cfq_remove_merge_hints(q, crq);
+ list_del_init(&rq->queuelist);
+
+ if (cfqq) {
+ cfq_del_crq_rb(cfqq, crq);
+
+ if (RB_EMPTY(&cfqq->sort_list))
+ cfq_put_queue(cfqd, cfqq);
+ }
+ }
+}
+
+static int
+cfq_merge(request_queue_t *q, struct request **req, struct bio *bio)
+{
+ struct cfq_data *cfqd = q->elevator.elevator_data;
+ struct request *__rq;
+ int ret;
+
+ ret = elv_try_last_merge(q, bio);
+ if (ret != ELEVATOR_NO_MERGE) {
+ __rq = q->last_merge;
+ goto out_insert;
+ }
+
+ __rq = cfq_find_rq_hash(cfqd, bio->bi_sector);
+ if (__rq) {
+ BUG_ON(__rq->sector + __rq->nr_sectors != bio->bi_sector);
+
+ if (elv_rq_merge_ok(__rq, bio)) {
+ ret = ELEVATOR_BACK_MERGE;
+ goto out;
+ }
+ }
+
+ __rq = cfq_find_rq_rb(cfqd, bio->bi_sector + bio_sectors(bio));
+ if (__rq) {
+ if (elv_rq_merge_ok(__rq, bio)) {
+ ret = ELEVATOR_FRONT_MERGE;
+ goto out;
+ }
+ }
+
+ return ELEVATOR_NO_MERGE;
+out:
+ q->last_merge = __rq;
+out_insert:
+ *req = __rq;
+ return ret;
+}
+
+static void cfq_merged_request(request_queue_t *q, struct request *req)
+{
+ struct cfq_data *cfqd = q->elevator.elevator_data;
+ struct cfq_rq *crq = RQ_DATA(req);
+
+ cfq_del_crq_hash(crq);
+ cfq_add_crq_hash(cfqd, crq);
+
+ if (ON_RB(&crq->rb_node) && (rq_rb_key(req) != crq->rb_key)) {
+ struct cfq_queue *cfqq = crq->cfq_queue;
+
+ cfq_del_crq_rb(cfqq, crq);
+ cfq_add_crq_rb(cfqd, cfqq, crq);
+ }
+
+ q->last_merge = req;
+}
+
+static void
+cfq_merged_requests(request_queue_t *q, struct request *req,
+ struct request *next)
+{
+ cfq_merged_request(q, req);
+ cfq_remove_request(q, next);
+}
+
+static void cfq_dispatch_sort(struct list_head *head, struct cfq_rq *crq)
+{
+ struct list_head *entry = head;
+ struct request *__rq;
+
+ if (!list_empty(head)) {
+ __rq = list_entry_rq(head->next);
+
+ if (crq->request->sector < __rq->sector) {
+ entry = head->prev;
+ goto link;
+ }
+ }
+
+ while ((entry = entry->prev) != head) {
+ __rq = list_entry_rq(entry);
+
+ if (crq->request->sector <= __rq->sector)
+ break;
+ }
+
+link:
+ list_add_tail(&crq->request->queuelist, entry);
+}
+
+static inline void
+__cfq_dispatch_requests(request_queue_t *q, struct cfq_data *cfqd,
+ struct cfq_queue *cfqq)
+{
+ struct cfq_rq *crq = rb_entry_crq(rb_first(&cfqq->sort_list));
+
+ cfq_del_crq_rb(cfqq, crq);
+ cfq_remove_merge_hints(q, crq);
+ cfq_dispatch_sort(cfqd->dispatch, crq);
+}
+
+static int cfq_dispatch_requests(request_queue_t *q, struct cfq_data *cfqd)
+{
+ struct cfq_queue *cfqq;
+ struct list_head *entry, *tmp;
+ int ret, queued, good_queues;
+
+ if (list_empty(&cfqd->rr_list))
+ return 0;
+
+ queued = ret = 0;
+restart:
+ good_queues = 0;
+ list_for_each_safe(entry, tmp, &cfqd->rr_list) {
+ cfqq = list_entry_cfqq(cfqd->rr_list.next);
+
+ BUG_ON(RB_EMPTY(&cfqq->sort_list));
+
+ __cfq_dispatch_requests(q, cfqd, cfqq);
+
+ if (RB_EMPTY(&cfqq->sort_list))
+ cfq_put_queue(cfqd, cfqq);
+ else
+ good_queues++;
+
+ queued++;
+ ret = 1;
+ }
+
+ if ((queued < cfq_quantum) && good_queues)
+ goto restart;
+
+ return ret;
+}
+
+static struct request *cfq_next_request(request_queue_t *q)
+{
+ struct cfq_data *cfqd = q->elevator.elevator_data;
+ struct request *rq;
+
+ if (!list_empty(cfqd->dispatch)) {
+ struct cfq_rq *crq;
+dispatch:
+ rq = list_entry_rq(cfqd->dispatch->next);
+
+ BUG_ON(q->last_merge == rq);
+ crq = RQ_DATA(rq);
+ if (crq)
+ BUG_ON(ON_MHASH(crq));
+
+ return rq;
+ }
+
+ if (cfq_dispatch_requests(q, cfqd))
+ goto dispatch;
+
+ return NULL;
+}
+
+static inline struct cfq_queue *
+__cfq_find_cfq_hash(struct cfq_data *cfqd, int pid, const int hashval)
+{
+ struct list_head *hash_list = &cfqd->cfq_hash[hashval];
+ struct list_head *entry;
+
+ list_for_each(entry, hash_list) {
+ struct cfq_queue *__cfqq = list_entry_qhash(entry);
+
+ if (__cfqq->pid == pid)
+ return __cfqq;
+ }
+
+ return NULL;
+}
+
+static struct cfq_queue *cfq_find_cfq_hash(struct cfq_data *cfqd, int pid)
+{
+ const int hashval = hash_long(current->tgid, CFQ_QHASH_SHIFT);
+
+ return __cfq_find_cfq_hash(cfqd, pid, hashval);
+}
+
+static void cfq_put_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq)
+{
+ cfqd->busy_queues--;
+ list_del(&cfqq->cfq_list);
+ list_del(&cfqq->cfq_hash);
+ mempool_free(cfqq, cfq_mpool);
+}
+
+static struct cfq_queue *cfq_get_queue(struct cfq_data *cfqd, int pid)
+{
+ const int hashval = hash_long(current->tgid, CFQ_QHASH_SHIFT);
+ struct cfq_queue *cfqq = __cfq_find_cfq_hash(cfqd, pid, hashval);
+
+ if (!cfqq) {
+ cfqq = mempool_alloc(cfq_mpool, GFP_NOIO);
+
+ INIT_LIST_HEAD(&cfqq->cfq_hash);
+ INIT_LIST_HEAD(&cfqq->cfq_list);
+ RB_CLEAR_ROOT(&cfqq->sort_list);
+
+ cfqq->pid = pid;
+ cfqq->queued[0] = cfqq->queued[1] = 0;
+ list_add(&cfqq->cfq_hash, &cfqd->cfq_hash[hashval]);
+ }
+
+ return cfqq;
+}
+
+static void cfq_enqueue(struct cfq_data *cfqd, struct cfq_rq *crq)
+{
+ struct cfq_queue *cfqq;
+
+ cfqq = cfq_get_queue(cfqd, current->tgid);
+
+ cfq_add_crq_rb(cfqd, cfqq, crq);
+
+ if (list_empty(&cfqq->cfq_list)) {
+ list_add(&cfqq->cfq_list, &cfqd->rr_list);
+ cfqd->busy_queues++;
+ }
+}
+
+static void
+cfq_insert_request(request_queue_t *q, struct request *rq, int where)
+{
+ struct cfq_data *cfqd = q->elevator.elevator_data;
+ struct cfq_rq *crq = RQ_DATA(rq);
+
+ switch (where) {
+ case ELEVATOR_INSERT_BACK:
+ while (cfq_dispatch_requests(q, cfqd))
+ ;
+ list_add_tail(&rq->queuelist, cfqd->dispatch);
+ break;
+ case ELEVATOR_INSERT_FRONT:
+ list_add(&rq->queuelist, cfqd->dispatch);
+ break;
+ case ELEVATOR_INSERT_SORT:
+ BUG_ON(!blk_fs_request(rq));
+ cfq_enqueue(cfqd, crq);
+ break;
+ default:
+ printk("%s: bad insert point %d\n", __FUNCTION__,where);
+ return;
+ }
+
+ if (rq_mergeable(rq)) {
+ cfq_add_crq_hash(cfqd, crq);
+
+ if (!q->last_merge)
+ q->last_merge = rq;
+ }
+}
+
+static int cfq_queue_empty(request_queue_t *q)
+{
+ struct cfq_data *cfqd = q->elevator.elevator_data;
+
+ if (list_empty(cfqd->dispatch) && list_empty(&cfqd->rr_list))
+ return 1;
+
+ return 0;
+}
+
+static struct request *
+cfq_former_request(request_queue_t *q, struct request *rq)
+{
+ struct cfq_rq *crq = RQ_DATA(rq);
+ struct rb_node *rbprev = rb_prev(&crq->rb_node);
+
+ if (rbprev)
+ return rb_entry_crq(rbprev)->request;
+
+ return NULL;
+}
+
+static struct request *
+cfq_latter_request(request_queue_t *q, struct request *rq)
+{
+ struct cfq_rq *crq = RQ_DATA(rq);
+ struct rb_node *rbnext = rb_next(&crq->rb_node);
+
+ if (rbnext)
+ return rb_entry_crq(rbnext)->request;
+
+ return NULL;
+}
+
+static int cfq_may_queue(request_queue_t *q, int rw)
+{
+ struct cfq_data *cfqd = q->elevator.elevator_data;
+ struct cfq_queue *cfqq;
+ int ret = 1;
+
+ if (!cfqd->busy_queues)
+ goto out;
+
+ cfqq = cfq_find_cfq_hash(cfqd, current->tgid);
+ if (cfqq) {
+ int limit = (q->nr_requests - cfq_queued) / cfqd->busy_queues;
+
+ if (limit < 3)
+ limit = 3;
+ else if (limit > cfqd->max_queued)
+ limit = cfqd->max_queued;
+
+ if (cfqq->queued[rw] > limit)
+ ret = 0;
+ }
+out:
+ return ret;
+}
+
+static void cfq_put_request(request_queue_t *q, struct request *rq)
+{
+ struct cfq_data *cfqd = q->elevator.elevator_data;
+ struct cfq_rq *crq = RQ_DATA(rq);
+
+ if (crq) {
+ BUG_ON(q->last_merge == rq);
+ BUG_ON(ON_MHASH(crq));
+
+ mempool_free(crq, cfqd->crq_pool);
+ rq->elevator_private = NULL;
+ }
+}
+
+static int cfq_set_request(request_queue_t *q, struct request *rq, int gfp_mask)
+{
+ struct cfq_data *cfqd = q->elevator.elevator_data;
+ struct cfq_rq *crq = mempool_alloc(cfqd->crq_pool, gfp_mask);
+
+ if (crq) {
+ RB_CLEAR(&crq->rb_node);
+ crq->request = rq;
+ crq->cfq_queue = NULL;
+ INIT_LIST_HEAD(&crq->hash);
+ rq->elevator_private = crq;
+ return 0;
+ }
+
+ return 1;
+}
+
+static void cfq_exit(request_queue_t *q, elevator_t *e)
+{
+ struct cfq_data *cfqd = e->elevator_data;
+
+ e->elevator_data = NULL;
+ mempool_destroy(cfqd->crq_pool);
+ kfree(cfqd->crq_hash);
+ kfree(cfqd->cfq_hash);
+ kfree(cfqd);
+}
+
+static int cfq_init(request_queue_t *q, elevator_t *e)
+{
+ struct cfq_data *cfqd;
+ int i;
+
+ cfqd = kmalloc(sizeof(*cfqd), GFP_KERNEL);
+ if (!cfqd)
+ return -ENOMEM;
+
+ memset(cfqd, 0, sizeof(*cfqd));
+ INIT_LIST_HEAD(&cfqd->rr_list);
+
+ cfqd->crq_hash = kmalloc(sizeof(struct list_head) * CFQ_MHASH_ENTRIES, GFP_KERNEL);
+ if (!cfqd->crq_hash)
+ goto out_crqhash;
+
+ cfqd->cfq_hash = kmalloc(sizeof(struct list_head) * CFQ_QHASH_ENTRIES, GFP_KERNEL);
+ if (!cfqd->cfq_hash)
+ goto out_cfqhash;
+
+ cfqd->crq_pool = mempool_create(BLKDEV_MIN_RQ, mempool_alloc_slab, mempool_free_slab, crq_pool);
+ if (!cfqd->crq_pool)
+ goto out_crqpool;
+
+ for (i = 0; i < CFQ_MHASH_ENTRIES; i++)
+ INIT_LIST_HEAD(&cfqd->crq_hash[i]);
+ for (i = 0; i < CFQ_QHASH_ENTRIES; i++)
+ INIT_LIST_HEAD(&cfqd->cfq_hash[i]);
+
+ cfqd->dispatch = &q->queue_head;
+ e->elevator_data = cfqd;
+
+ /*
+ * just set it to some high value, we want anyone to be able to queue
+ * some requests. fairness is handled differently
+ */
+ cfqd->max_queued = q->nr_requests;
+ q->nr_requests = 8192;
+
+ return 0;
+out_crqpool:
+ kfree(cfqd->cfq_hash);
+out_cfqhash:
+ kfree(cfqd->crq_hash);
+out_crqhash:
+ kfree(cfqd);
+ return -ENOMEM;
+}
+
+static int __init cfq_slab_setup(void)
+{
+ crq_pool = kmem_cache_create("crq_pool", sizeof(struct cfq_rq), 0, 0,
+ NULL, NULL);
+
+ if (!crq_pool)
+ panic("cfq_iosched: can't init crq pool\n");
+
+ cfq_pool = kmem_cache_create("cfq_pool", sizeof(struct cfq_queue), 0, 0,
+ NULL, NULL);
+
+ if (!cfq_pool)
+ panic("cfq_iosched: can't init cfq pool\n");
+
+ cfq_mpool = mempool_create(64, mempool_alloc_slab, mempool_free_slab, cfq_pool);
+
+ if (!cfq_mpool)
+ panic("cfq_iosched: can't init cfq mpool\n");
+
+ return 0;
+}
+
+subsys_initcall(cfq_slab_setup);
+
+elevator_t iosched_cfq = {
+ .elevator_name = "cfq",
+ .elevator_merge_fn = cfq_merge,
+ .elevator_merged_fn = cfq_merged_request,
+ .elevator_merge_req_fn = cfq_merged_requests,
+ .elevator_next_req_fn = cfq_next_request,
+ .elevator_add_req_fn = cfq_insert_request,
+ .elevator_remove_req_fn = cfq_remove_request,
+ .elevator_queue_empty_fn = cfq_queue_empty,
+ .elevator_former_req_fn = cfq_former_request,
+ .elevator_latter_req_fn = cfq_latter_request,
+ .elevator_set_req_fn = cfq_set_request,
+ .elevator_put_req_fn = cfq_put_request,
+ .elevator_may_queue_fn = cfq_may_queue,
+ .elevator_init_fn = cfq_init,
+ .elevator_exit_fn = cfq_exit,
+};
+
+EXPORT_SYMBOL(iosched_cfq);
--- diff/drivers/char/drm/drm_irq.h 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/char/drm/drm_irq.h 2004-03-16 11:27:48.023886920 +0000
@@ -0,0 +1,377 @@
+/**
+ * \file drm_irq.h
+ * IRQ support
+ *
+ * \author Rickard E. (Rik) Faith <faith@valinux.com>
+ * \author Gareth Hughes <gareth@valinux.com>
+ */
+
+/*
+ * Created: Fri Mar 19 14:30:16 1999 by faith@valinux.com
+ *
+ * Copyright 1999, 2000 Precision Insight, Inc., Cedar Park, Texas.
+ * Copyright 2000 VA Linux Systems, Inc., Sunnyvale, California.
+ * All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * VA LINUX SYSTEMS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include "drmP.h"
+
+#include <linux/interrupt.h> /* For task queue support */
+
+#ifndef __HAVE_SHARED_IRQ
+#define __HAVE_SHARED_IRQ 0
+#endif
+
+#if __HAVE_SHARED_IRQ
+#define DRM_IRQ_TYPE SA_SHIRQ
+#else
+#define DRM_IRQ_TYPE 0
+#endif
+
+/**
+ * Get interrupt from bus id.
+ *
+ * \param inode device inode.
+ * \param filp file pointer.
+ * \param cmd command.
+ * \param arg user argument, pointing to a drm_irq_busid structure.
+ * \return zero on success or a negative number on failure.
+ *
+ * Finds the PCI device with the specified bus id and gets its IRQ number.
+ * This IOCTL is deprecated, and will now return EINVAL for any busid not equal
+ * to that of the device that this DRM instance attached to.
+ */
+int DRM(irq_by_busid)(struct inode *inode, struct file *filp,
+ unsigned int cmd, unsigned long arg)
+{
+ drm_file_t *priv = filp->private_data;
+ drm_device_t *dev = priv->dev;
+ drm_irq_busid_t p;
+
+ if (copy_from_user(&p, (drm_irq_busid_t *)arg, sizeof(p)))
+ return -EFAULT;
+
+ if ((p.busnum >> 8) != dev->pci_domain ||
+ (p.busnum & 0xff) != dev->pci_bus ||
+ p.devnum != dev->pci_slot ||
+ p.funcnum != dev->pci_func)
+ return -EINVAL;
+
+ p.irq = dev->irq;
+
+ DRM_DEBUG("%d:%d:%d => IRQ %d\n",
+ p.busnum, p.devnum, p.funcnum, p.irq);
+ if (copy_to_user((drm_irq_busid_t *)arg, &p, sizeof(p)))
+ return -EFAULT;
+ return 0;
+}
+
+#if __HAVE_IRQ
+
+/**
+ * Install IRQ handler.
+ *
+ * \param dev DRM device.
+ * \param irq IRQ number.
+ *
+ * Initializes the IRQ related data, and setups drm_device::vbl_queue. Installs the handler, calling the driver
+ * \c DRM(driver_irq_preinstall)() and \c DRM(driver_irq_postinstall)() functions
+ * before and after the installation.
+ */
+int DRM(irq_install)( drm_device_t *dev )
+{
+ int ret;
+
+ if ( dev->irq == 0 )
+ return -EINVAL;
+
+ down( &dev->struct_sem );
+
+ /* Driver must have been initialized */
+ if ( !dev->dev_private ) {
+ up( &dev->struct_sem );
+ return -EINVAL;
+ }
+
+ if ( dev->irq_enabled ) {
+ up( &dev->struct_sem );
+ return -EBUSY;
+ }
+ dev->irq_enabled = 1;
+ up( &dev->struct_sem );
+
+ DRM_DEBUG( "%s: irq=%d\n", __FUNCTION__, dev->irq );
+
+#if 0 /* this is already done in DRM(setup) - why do it here ?? */
+ dev->context_flag = 0;
+ dev->interrupt_flag = 0;
+ dev->dma_flag = 0;
+#endif
+
+#if __HAVE_DMA
+ dev->dma->next_buffer = NULL;
+ dev->dma->next_queue = NULL;
+ dev->dma->this_buffer = NULL;
+#endif
+
+#if __HAVE_IRQ_BH
+ INIT_WORK(&dev->work, DRM(irq_immediate_bh), dev);
+#endif
+
+#if __HAVE_VBL_IRQ
+ init_waitqueue_head(&dev->vbl_queue);
+
+ spin_lock_init( &dev->vbl_lock );
+
+ INIT_LIST_HEAD( &dev->vbl_sigs.head );
+
+ dev->vbl_pending = 0;
+#endif
+
+ /* Before installing handler */
+ DRM(driver_irq_preinstall)(dev);
+
+ /* Install handler */
+ ret = request_irq( dev->irq, DRM(irq_handler),
+ DRM_IRQ_TYPE, dev->devname, dev );
+ if ( ret < 0 ) {
+ down( &dev->struct_sem );
+ dev->irq_enabled = 0;
+ up( &dev->struct_sem );
+ return ret;
+ }
+
+ /* After installing handler */
+ DRM(driver_irq_postinstall)(dev);
+
+ return 0;
+}
+
+/**
+ * Uninstall the IRQ handler.
+ *
+ * \param dev DRM device.
+ *
+ * Calls the driver's \c DRM(driver_irq_uninstall)() function, and stops the irq.
+ */
+int DRM(irq_uninstall)( drm_device_t *dev )
+{
+ int irq_enabled;
+
+ down( &dev->struct_sem );
+ irq_enabled = dev->irq_enabled;
+ dev->irq_enabled = 0;
+ up( &dev->struct_sem );
+
+ if ( !irq_enabled )
+ return -EINVAL;
+
+ DRM_DEBUG( "%s: irq=%d\n", __FUNCTION__, dev->irq );
+
+ DRM(driver_irq_uninstall)( dev );
+
+ free_irq( dev->irq, dev );
+
+ return 0;
+}
+
+/**
+ * IRQ control ioctl.
+ *
+ * \param inode device inode.
+ * \param filp file pointer.
+ * \param cmd command.
+ * \param arg user argument, pointing to a drm_control structure.
+ * \return zero on success or a negative number on failure.
+ *
+ * Calls irq_install() or irq_uninstall() according to \p arg.
+ */
+int DRM(control)( struct inode *inode, struct file *filp,
+ unsigned int cmd, unsigned long arg )
+{
+ drm_file_t *priv = filp->private_data;
+ drm_device_t *dev = priv->dev;
+ drm_control_t ctl;
+
+ if ( copy_from_user( &ctl, (drm_control_t *)arg, sizeof(ctl) ) )
+ return -EFAULT;
+
+ switch ( ctl.func ) {
+ case DRM_INST_HANDLER:
+ if (dev->if_version < DRM_IF_VERSION(1, 2) &&
+ ctl.irq != dev->irq)
+ return -EINVAL;
+ return DRM(irq_install)( dev );
+ case DRM_UNINST_HANDLER:
+ return DRM(irq_uninstall)( dev );
+ default:
+ return -EINVAL;
+ }
+}
+
+#if __HAVE_VBL_IRQ
+
+/**
+ * Wait for VBLANK.
+ *
+ * \param inode device inode.
+ * \param filp file pointer.
+ * \param cmd command.
+ * \param data user argument, pointing to a drm_wait_vblank structure.
+ * \return zero on success or a negative number on failure.
+ *
+ * Verifies the IRQ is installed.
+ *
+ * If a signal is requested checks if this task has already scheduled the same signal
+ * for the same vblank sequence number - nothing to be done in
+ * that case. If the number of tasks waiting for the interrupt exceeds 100 the
+ * function fails. Otherwise adds a new entry to drm_device::vbl_sigs for this
+ * task.
+ *
+ * If a signal is not requested, then calls vblank_wait().
+ */
+int DRM(wait_vblank)( DRM_IOCTL_ARGS )
+{
+ drm_file_t *priv = filp->private_data;
+ drm_device_t *dev = priv->dev;
+ drm_wait_vblank_t vblwait;
+ struct timeval now;
+ int ret = 0;
+ unsigned int flags;
+
+ if (!dev->irq)
+ return -EINVAL;
+
+ DRM_COPY_FROM_USER_IOCTL( vblwait, (drm_wait_vblank_t *)data,
+ sizeof(vblwait) );
+
+ switch ( vblwait.request.type & ~_DRM_VBLANK_FLAGS_MASK ) {
+ case _DRM_VBLANK_RELATIVE:
+ vblwait.request.sequence += atomic_read( &dev->vbl_received );
+ vblwait.request.type &= ~_DRM_VBLANK_RELATIVE;
+ case _DRM_VBLANK_ABSOLUTE:
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ flags = vblwait.request.type & _DRM_VBLANK_FLAGS_MASK;
+
+ if ( flags & _DRM_VBLANK_SIGNAL ) {
+ unsigned long irqflags;
+ drm_vbl_sig_t *vbl_sig;
+
+ vblwait.reply.sequence = atomic_read( &dev->vbl_received );
+
+ spin_lock_irqsave( &dev->vbl_lock, irqflags );
+
+ /* Check if this task has already scheduled the same signal
+ * for the same vblank sequence number; nothing to be done in
+ * that case
+ */
+ list_for_each_entry( vbl_sig, &dev->vbl_sigs.head, head ) {
+ if (vbl_sig->sequence == vblwait.request.sequence
+ && vbl_sig->info.si_signo == vblwait.request.signal
+ && vbl_sig->task == current)
+ {
+ spin_unlock_irqrestore( &dev->vbl_lock, irqflags );
+ goto done;
+ }
+ }
+
+ if ( dev->vbl_pending >= 100 ) {
+ spin_unlock_irqrestore( &dev->vbl_lock, irqflags );
+ return -EBUSY;
+ }
+
+ dev->vbl_pending++;
+
+ spin_unlock_irqrestore( &dev->vbl_lock, irqflags );
+
+ if ( !( vbl_sig = DRM_MALLOC( sizeof( drm_vbl_sig_t ) ) ) ) {
+ return -ENOMEM;
+ }
+
+ memset( (void *)vbl_sig, 0, sizeof(*vbl_sig) );
+
+ vbl_sig->sequence = vblwait.request.sequence;
+ vbl_sig->info.si_signo = vblwait.request.signal;
+ vbl_sig->task = current;
+
+ spin_lock_irqsave( &dev->vbl_lock, irqflags );
+
+ list_add_tail( (struct list_head *) vbl_sig, &dev->vbl_sigs.head );
+
+ spin_unlock_irqrestore( &dev->vbl_lock, irqflags );
+ } else {
+ ret = DRM(vblank_wait)( dev, &vblwait.request.sequence );
+
+ do_gettimeofday( &now );
+ vblwait.reply.tval_sec = now.tv_sec;
+ vblwait.reply.tval_usec = now.tv_usec;
+ }
+
+done:
+ DRM_COPY_TO_USER_IOCTL( (drm_wait_vblank_t *)data, vblwait,
+ sizeof(vblwait) );
+
+ return ret;
+}
+
+/**
+ * Send the VBLANK signals.
+ *
+ * \param dev DRM device.
+ *
+ * Sends a signal for each task in drm_device::vbl_sigs and empties the list.
+ *
+ * If a signal is not requested, then calls vblank_wait().
+ */
+void DRM(vbl_send_signals)( drm_device_t *dev )
+{
+ struct list_head *list, *tmp;
+ drm_vbl_sig_t *vbl_sig;
+ unsigned int vbl_seq = atomic_read( &dev->vbl_received );
+ unsigned long flags;
+
+ spin_lock_irqsave( &dev->vbl_lock, flags );
+
+ list_for_each_safe( list, tmp, &dev->vbl_sigs.head ) {
+ vbl_sig = list_entry( list, drm_vbl_sig_t, head );
+ if ( ( vbl_seq - vbl_sig->sequence ) <= (1<<23) ) {
+ vbl_sig->info.si_code = vbl_seq;
+ send_sig_info( vbl_sig->info.si_signo, &vbl_sig->info, vbl_sig->task );
+
+ list_del( list );
+
+ DRM_FREE( vbl_sig, sizeof(*vbl_sig) );
+
+ dev->vbl_pending--;
+ }
+ }
+
+ spin_unlock_irqrestore( &dev->vbl_lock, flags );
+}
+
+#endif /* __HAVE_VBL_IRQ */
+
+#endif /* __HAVE_IRQ */
--- diff/drivers/char/ipmi/ipmi_bt_sm.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/char/ipmi/ipmi_bt_sm.c 2004-03-16 11:27:48.070879776 +0000
@@ -0,0 +1,515 @@
+/*
+ * ipmi_bt_sm.c
+ *
+ * The state machine for an Open IPMI BT sub-driver under ipmi_si.c, part
+ * of the driver architecture at http://sourceforge.net/project/openipmi
+ *
+ * Author: Rocky Craig <first.last@hp.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ *
+ * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
+ * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
+ * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
+ * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
+ * OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR
+ * TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
+ * USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 675 Mass Ave, Cambridge, MA 02139, USA. */
+
+#include <linux/types.h>
+
+#include <linux/kernel.h> /* For printk. */
+#include <linux/string.h>
+#include <linux/ipmi_msgdefs.h> /* for completion codes */
+#include "ipmi_si_sm.h"
+
+#define IPMI_BT_VERSION "v31"
+
+static int bt_debug = 0x00; /* Production value 0, see following flags */
+
+#define BT_DEBUG_ENABLE 1
+#define BT_DEBUG_MSG 2
+#define BT_DEBUG_STATES 4
+
+/* Typical "Get BT Capabilities" values are 2-3 retries, 5-10 seconds,
+ and 64 byte buffers. However, one HP implementation wants 255 bytes of
+ buffer (with a documented message of 160 bytes) so go for the max.
+ Since the Open IPMI architecture is single-message oriented at this
+ stage, the queue depth of BT is of no concern. */
+
+#define BT_NORMAL_TIMEOUT 2000000 /* seconds in microseconds */
+#define BT_RETRY_LIMIT 2
+#define BT_RESET_DELAY 6000000 /* 6 seconds after warm reset */
+
+enum bt_states {
+ BT_STATE_IDLE,
+ BT_STATE_XACTION_START,
+ BT_STATE_WRITE_BYTES,
+ BT_STATE_WRITE_END,
+ BT_STATE_WRITE_CONSUME,
+ BT_STATE_B2H_WAIT,
+ BT_STATE_READ_END,
+ BT_STATE_RESET1, /* These must come last */
+ BT_STATE_RESET2,
+ BT_STATE_RESET3,
+ BT_STATE_RESTART,
+ BT_STATE_HOSED
+};
+
+struct si_sm_data {
+ enum bt_states state;
+ enum bt_states last_state; /* assist printing and resets */
+ unsigned char seq; /* BT sequence number */
+ struct si_sm_io *io;
+ unsigned char write_data[IPMI_MAX_MSG_LENGTH];
+ int write_count;
+ unsigned char read_data[IPMI_MAX_MSG_LENGTH];
+ int read_count;
+ int truncated;
+ long timeout;
+ unsigned int error_retries; /* end of "common" fields */
+ int nonzero_status; /* hung BMCs stay all 0 */
+};
+
+#define BT_CLR_WR_PTR 0x01 /* See IPMI 1.5 table 11.6.4 */
+#define BT_CLR_RD_PTR 0x02
+#define BT_H2B_ATN 0x04
+#define BT_B2H_ATN 0x08
+#define BT_SMS_ATN 0x10
+#define BT_OEM0 0x20
+#define BT_H_BUSY 0x40
+#define BT_B_BUSY 0x80
+
+/* Some bits are toggled on each write: write once to set it, once
+ more to clear it; writing a zero does nothing. To absolutely
+ clear it, check its state and write if set. This avoids the "get
+ current then use as mask" scheme to modify one bit. Note that the
+ variable "bt" is hardcoded into these macros. */
+
+#define BT_STATUS bt->io->inputb(bt->io, 0)
+#define BT_CONTROL(x) bt->io->outputb(bt->io, 0, x)
+
+#define BMC2HOST bt->io->inputb(bt->io, 1)
+#define HOST2BMC(x) bt->io->outputb(bt->io, 1, x)
+
+#define BT_INTMASK_R bt->io->inputb(bt->io, 2)
+#define BT_INTMASK_W(x) bt->io->outputb(bt->io, 2, x)
+
+/* Convenience routines for debugging. These are not multi-open safe!
+ Note the macros have hardcoded variables in them. */
+
+static char *state2txt(unsigned char state)
+{
+ switch (state) {
+ case BT_STATE_IDLE: return("IDLE");
+ case BT_STATE_XACTION_START: return("XACTION");
+ case BT_STATE_WRITE_BYTES: return("WR_BYTES");
+ case BT_STATE_WRITE_END: return("WR_END");
+ case BT_STATE_WRITE_CONSUME: return("WR_CONSUME");
+ case BT_STATE_B2H_WAIT: return("B2H_WAIT");
+ case BT_STATE_READ_END: return("RD_END");
+ case BT_STATE_RESET1: return("RESET1");
+ case BT_STATE_RESET2: return("RESET2");
+ case BT_STATE_RESET3: return("RESET3");
+ case BT_STATE_RESTART: return("RESTART");
+ case BT_STATE_HOSED: return("HOSED");
+ }
+ return("BAD STATE");
+}
+#define STATE2TXT state2txt(bt->state)
+
+static char *status2txt(unsigned char status, char *buf)
+{
+ strcpy(buf, "[ ");
+ if (status & BT_B_BUSY) strcat(buf, "B_BUSY ");
+ if (status & BT_H_BUSY) strcat(buf, "H_BUSY ");
+ if (status & BT_OEM0) strcat(buf, "OEM0 ");
+ if (status & BT_SMS_ATN) strcat(buf, "SMS ");
+ if (status & BT_B2H_ATN) strcat(buf, "B2H ");
+ if (status & BT_H2B_ATN) strcat(buf, "H2B ");
+ strcat(buf, "]");
+ return buf;
+}
+#define STATUS2TXT(buf) status2txt(status, buf)
+
+/* This will be called from within this module on a hosed condition */
+#define FIRST_SEQ 0
+static unsigned int bt_init_data(struct si_sm_data *bt, struct si_sm_io *io)
+{
+ bt->state = BT_STATE_IDLE;
+ bt->last_state = BT_STATE_IDLE;
+ bt->seq = FIRST_SEQ;
+ bt->io = io;
+ bt->write_count = 0;
+ bt->read_count = 0;
+ bt->error_retries = 0;
+ bt->nonzero_status = 0;
+ bt->truncated = 0;
+ bt->timeout = BT_NORMAL_TIMEOUT;
+ return 3; /* We claim 3 bytes of space; ought to check SPMI table */
+}
+
+static int bt_start_transaction(struct si_sm_data *bt,
+ unsigned char *data,
+ unsigned int size)
+{
+ unsigned int i;
+
+ if ((size < 2) || (size > IPMI_MAX_MSG_LENGTH)) return -1;
+
+ if ((bt->state != BT_STATE_IDLE) && (bt->state != BT_STATE_HOSED))
+ return -2;
+
+ if (bt_debug & BT_DEBUG_MSG) {
+ printk(KERN_WARNING "+++++++++++++++++++++++++++++++++++++\n");
+ printk(KERN_WARNING "BT: write seq=0x%02X:", bt->seq);
+ for (i = 0; i < size; i ++) printk (" %02x", data[i]);
+ printk("\n");
+ }
+ bt->write_data[0] = size + 1; /* all data plus seq byte */
+ bt->write_data[1] = *data; /* NetFn/LUN */
+ bt->write_data[2] = bt->seq;
+ memcpy(bt->write_data + 3, data + 1, size - 1);
+ bt->write_count = size + 2;
+
+ bt->error_retries = 0;
+ bt->nonzero_status = 0;
+ bt->read_count = 0;
+ bt->truncated = 0;
+ bt->state = BT_STATE_XACTION_START;
+ bt->last_state = BT_STATE_IDLE;
+ bt->timeout = BT_NORMAL_TIMEOUT;
+ return 0;
+}
+
+/* After the upper state machine has been told SI_SM_TRANSACTION_COMPLETE
+ it calls this. Strip out the length and seq bytes. */
+
+static int bt_get_result(struct si_sm_data *bt,
+ unsigned char *data,
+ unsigned int length)
+{
+ int i, msg_len;
+
+ msg_len = bt->read_count - 2; /* account for length & seq */
+ /* Always NetFn, Cmd, cCode */
+ if (msg_len < 3 || msg_len > IPMI_MAX_MSG_LENGTH) {
+ printk(KERN_WARNING "BT results: bad msg_len = %d\n", msg_len);
+ data[0] = bt->write_data[1] | 0x4; /* Kludge a response */
+ data[1] = bt->write_data[3];
+ data[2] = IPMI_ERR_UNSPECIFIED;
+ msg_len = 3;
+ } else {
+ data[0] = bt->read_data[1];
+ data[1] = bt->read_data[3];
+ if (length < msg_len) bt->truncated = 1;
+ if (bt->truncated) { /* can be set in read_all_bytes() */
+ data[2] = IPMI_ERR_MSG_TRUNCATED;
+ msg_len = 3;
+ } else memcpy(data + 2, bt->read_data + 4, msg_len - 2);
+
+ if (bt_debug & BT_DEBUG_MSG) {
+ printk (KERN_WARNING "BT: res (raw)");
+ for (i = 0; i < msg_len; i++) printk(" %02x", data[i]);
+ printk ("\n");
+ }
+ }
+ bt->read_count = 0; /* paranoia */
+ return msg_len;
+}
+
+/* This bit's functionality is optional */
+#define BT_BMC_HWRST 0x80
+
+static void reset_flags(struct si_sm_data *bt)
+{
+ if (BT_STATUS & BT_H_BUSY) BT_CONTROL(BT_H_BUSY);
+ if (BT_STATUS & BT_B_BUSY) BT_CONTROL(BT_B_BUSY);
+ BT_CONTROL(BT_CLR_WR_PTR);
+ BT_CONTROL(BT_SMS_ATN);
+ BT_INTMASK_W(BT_BMC_HWRST);
+#ifdef DEVELOPMENT_ONLY_NOT_FOR_PRODUCTION
+ if (BT_STATUS & BT_B2H_ATN) {
+ int i;
+ BT_CONTROL(BT_H_BUSY);
+ BT_CONTROL(BT_B2H_ATN);
+ BT_CONTROL(BT_CLR_RD_PTR);
+ for (i = 0; i < IPMI_MAX_MSG_LENGTH + 2; i++) BMC2HOST;
+ BT_CONTROL(BT_H_BUSY);
+ }
+#endif
+}
+
+static inline void write_all_bytes(struct si_sm_data *bt)
+{
+ int i;
+
+ if (bt_debug & BT_DEBUG_MSG) {
+ printk(KERN_WARNING "BT: write %d bytes seq=0x%02X",
+ bt->write_count, bt->seq);
+ for (i = 0; i < bt->write_count; i++)
+ printk (" %02x", bt->write_data[i]);
+ printk ("\n");
+ }
+ for (i = 0; i < bt->write_count; i++) HOST2BMC(bt->write_data[i]);
+}
+
+static inline int read_all_bytes(struct si_sm_data *bt)
+{
+ unsigned char i;
+
+ bt->read_data[0] = BMC2HOST;
+ bt->read_count = bt->read_data[0];
+ if (bt_debug & BT_DEBUG_MSG)
+ printk(KERN_WARNING "BT: read %d bytes:", bt->read_count);
+
+ /* minimum: length, NetFn, Seq, Cmd, cCode == 5 total, or 4 more
+ following the length byte. */
+ if (bt->read_count < 4 || bt->read_count >= IPMI_MAX_MSG_LENGTH) {
+ if (bt_debug & BT_DEBUG_MSG)
+ printk("bad length %d\n", bt->read_count);
+ bt->truncated = 1;
+ return 1; /* let next XACTION START clean it up */
+ }
+ for (i = 1; i <= bt->read_count; i++) bt->read_data[i] = BMC2HOST;
+ bt->read_count++; /* account for the length byte */
+
+ if (bt_debug & BT_DEBUG_MSG) {
+ for (i = 0; i < bt->read_count; i++)
+ printk (" %02x", bt->read_data[i]);
+ printk ("\n");
+ }
+ if (bt->seq != bt->write_data[2]) /* idiot check */
+ printk(KERN_WARNING "BT: internal error: sequence mismatch\n");
+
+ /* per the spec, the (NetFn, Seq, Cmd) tuples should match */
+ if ((bt->read_data[3] == bt->write_data[3]) && /* Cmd */
+ (bt->read_data[2] == bt->write_data[2]) && /* Sequence */
+ ((bt->read_data[1] & 0xF8) == (bt->write_data[1] & 0xF8)))
+ return 1;
+
+ if (bt_debug & BT_DEBUG_MSG) printk(KERN_WARNING "BT: bad packet: "
+ "want 0x(%02X, %02X, %02X) got (%02X, %02X, %02X)\n",
+ bt->write_data[1], bt->write_data[2], bt->write_data[3],
+ bt->read_data[1], bt->read_data[2], bt->read_data[3]);
+ return 0;
+}
+
+/* Modifies bt->state appropriately, need to get into the bt_event() switch */
+
+static void error_recovery(struct si_sm_data *bt, char *reason)
+{
+ unsigned char status;
+ char buf[40]; /* For getting status */
+
+ bt->timeout = BT_NORMAL_TIMEOUT; /* various places want to retry */
+
+ status = BT_STATUS;
+ printk(KERN_WARNING "BT: %s in %s %s ", reason, STATE2TXT,
+ STATUS2TXT(buf));
+
+ (bt->error_retries)++;
+ if (bt->error_retries > BT_RETRY_LIMIT) {
+ printk("retry limit (%d) exceeded\n", BT_RETRY_LIMIT);
+ bt->state = BT_STATE_HOSED;
+ if (!bt->nonzero_status)
+ printk(KERN_ERR "IPMI: BT stuck, try power cycle\n");
+ else if (bt->seq == FIRST_SEQ + BT_RETRY_LIMIT) {
+ /* most likely during insmod */
+ printk(KERN_WARNING "IPMI: BT reset (takes 5 secs)\n");
+ bt->state = BT_STATE_RESET1;
+ }
+ return;
+ }
+
+ /* Sometimes the BMC queues get in an "off-by-one" state...*/
+ if ((bt->state == BT_STATE_B2H_WAIT) && (status & BT_B2H_ATN)) {
+ printk("retry B2H_WAIT\n");
+ return;
+ }
+
+ printk("restart command\n");
+ bt->state = BT_STATE_RESTART;
+}
+
+/* Check the status and (possibly) advance the BT state machine. The
+ default return is SI_SM_CALL_WITH_DELAY. */
+
+static enum si_sm_result bt_event(struct si_sm_data *bt, long time)
+{
+ unsigned char status;
+ char buf[40]; /* For getting status */
+ int i;
+
+ status = BT_STATUS;
+ bt->nonzero_status |= status;
+
+ if ((bt_debug & BT_DEBUG_STATES) && (bt->state != bt->last_state))
+ printk(KERN_WARNING "BT: %s %s TO=%ld - %ld \n",
+ STATE2TXT,
+ STATUS2TXT(buf),
+ bt->timeout,
+ time);
+ bt->last_state = bt->state;
+
+ if (bt->state == BT_STATE_HOSED) return SI_SM_HOSED;
+
+ if (bt->state != BT_STATE_IDLE) { /* do timeout test */
+
+ /* Certain states, on error conditions, can lock up a CPU
+ because they are effectively in an infinite loop with
+ CALL_WITHOUT_DELAY (right back here with time == 0).
+ Prevent infinite lockup by ALWAYS decrementing timeout. */
+
+ /* FIXME: bt_event is sometimes called with time > BT_NORMAL_TIMEOUT
+ (noticed in ipmi_smic_sm.c January 2004) */
+
+ if ((time <= 0) || (time >= BT_NORMAL_TIMEOUT)) time = 100;
+ bt->timeout -= time;
+ if ((bt->timeout < 0) && (bt->state < BT_STATE_RESET1)) {
+ error_recovery(bt, "timed out");
+ return SI_SM_CALL_WITHOUT_DELAY;
+ }
+ }
+
+ switch (bt->state) {
+
+ case BT_STATE_IDLE: /* check for asynchronous messages */
+ if (status & BT_SMS_ATN) {
+ BT_CONTROL(BT_SMS_ATN); /* clear it */
+ return SI_SM_ATTN;
+ }
+ return SI_SM_IDLE;
+
+ case BT_STATE_XACTION_START:
+ if (status & BT_H_BUSY) {
+ BT_CONTROL(BT_H_BUSY);
+ break;
+ }
+ if (status & BT_B2H_ATN) break;
+ bt->state = BT_STATE_WRITE_BYTES;
+ return SI_SM_CALL_WITHOUT_DELAY; /* for logging */
+
+ case BT_STATE_WRITE_BYTES:
+ if (status & (BT_B_BUSY | BT_H2B_ATN)) break;
+ BT_CONTROL(BT_CLR_WR_PTR);
+ write_all_bytes(bt);
+ BT_CONTROL(BT_H2B_ATN); /* clears too fast to catch? */
+ bt->state = BT_STATE_WRITE_CONSUME;
+ return SI_SM_CALL_WITHOUT_DELAY; /* it MIGHT sail through */
+
+ case BT_STATE_WRITE_CONSUME: /* BMCs usually blow right thru here */
+ if (status & (BT_H2B_ATN | BT_B_BUSY)) break;
+ bt->state = BT_STATE_B2H_WAIT;
+ /* fall through with status */
+
+ /* Stay in BT_STATE_B2H_WAIT until a packet matches. However, spinning
+ hard here, constantly reading status, seems to hold off the
+ generation of B2H_ATN so ALWAYS return CALL_WITH_DELAY. */
+
+ case BT_STATE_B2H_WAIT:
+ if (!(status & BT_B2H_ATN)) break;
+
+ /* Assume ordered, uncached writes: no need to wait */
+ if (!(status & BT_H_BUSY)) BT_CONTROL(BT_H_BUSY); /* set */
+ BT_CONTROL(BT_B2H_ATN); /* clear it, ACK to the BMC */
+ BT_CONTROL(BT_CLR_RD_PTR); /* reset the queue */
+ i = read_all_bytes(bt);
+ BT_CONTROL(BT_H_BUSY); /* clear */
+ if (!i) break; /* Try this state again */
+ bt->state = BT_STATE_READ_END;
+ return SI_SM_CALL_WITHOUT_DELAY; /* for logging */
+
+ case BT_STATE_READ_END:
+
+ /* I could wait on BT_H_BUSY to go clear for a truly clean
+ exit. However, this is already done in XACTION_START
+ and the (possible) extra loop/status/possible wait affects
+ performance. So, as long as it works, just ignore H_BUSY */
+
+#ifdef MAKE_THIS_TRUE_IF_NECESSARY
+
+ if (status & BT_H_BUSY) break;
+#endif
+ bt->seq++;
+ bt->state = BT_STATE_IDLE;
+ return SI_SM_TRANSACTION_COMPLETE;
+
+ case BT_STATE_RESET1:
+ reset_flags(bt);
+ bt->timeout = BT_RESET_DELAY;;
+ bt->state = BT_STATE_RESET2;
+ break;
+
+ case BT_STATE_RESET2: /* Send a soft reset */
+ BT_CONTROL(BT_CLR_WR_PTR);
+ HOST2BMC(3); /* number of bytes following */
+ HOST2BMC(0x18); /* NetFn/LUN == Application, LUN 0 */
+ HOST2BMC(42); /* Sequence number */
+ HOST2BMC(3); /* Cmd == Soft reset */
+ BT_CONTROL(BT_H2B_ATN);
+ bt->state = BT_STATE_RESET3;
+ break;
+
+ case BT_STATE_RESET3:
+ if (bt->timeout > 0) return SI_SM_CALL_WITH_DELAY;
+ bt->state = BT_STATE_RESTART; /* printk in debug modes */
+ break;
+
+ case BT_STATE_RESTART: /* don't reset retries! */
+ bt->write_data[2] = ++bt->seq;
+ bt->read_count = 0;
+ bt->nonzero_status = 0;
+ bt->timeout = BT_NORMAL_TIMEOUT;
+ bt->state = BT_STATE_XACTION_START;
+ break;
+
+ default: /* HOSED is supposed to be caught much earlier */
+ error_recovery(bt, "internal logic error");
+ break;
+ }
+ return SI_SM_CALL_WITH_DELAY;
+}
+
+static int bt_detect(struct si_sm_data *bt)
+{
+ /* It's impossible for the BT status and interrupt registers to be
+ all 1's, (assuming a properly functioning, self-initialized BMC)
+ but that's what you get from reading a bogus address, so we
+ test that first. The calling routine uses negative logic. */
+
+ if ((BT_STATUS == 0xFF) && (BT_INTMASK_R == 0xFF)) return 1;
+ reset_flags(bt);
+ return 0;
+}
+
+static void bt_cleanup(struct si_sm_data *bt)
+{
+}
+
+static int bt_size(void)
+{
+ return sizeof(struct si_sm_data);
+}
+
+struct si_sm_handlers bt_smi_handlers =
+{
+ .version = IPMI_BT_VERSION,
+ .init_data = bt_init_data,
+ .start_transaction = bt_start_transaction,
+ .get_result = bt_get_result,
+ .event = bt_event,
+ .detect = bt_detect,
+ .cleanup = bt_cleanup,
+ .size = bt_size,
+};
--- diff/drivers/char/ipmi/ipmi_si_intf.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/char/ipmi/ipmi_si_intf.c 2004-03-16 11:27:48.095875976 +0000
@@ -0,0 +1,2051 @@
+/*
+ * ipmi_si.c
+ *
+ * The interface to the IPMI driver for the system interfaces (KCS, SMIC,
+ * BT).
+ *
+ * Author: MontaVista Software, Inc.
+ * Corey Minyard <minyard@mvista.com>
+ * source@mvista.com
+ *
+ * Copyright 2002 MontaVista Software Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ *
+ *
+ * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
+ * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
+ * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
+ * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
+ * OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR
+ * TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
+ * USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+/*
+ * This file holds the "policy" for the interface to the SMI state
+ * machine. It does the configuration, handles timers and interrupts,
+ * and drives the real SMI state machine.
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <asm/system.h>
+#include <linux/sched.h>
+#include <linux/timer.h>
+#include <linux/errno.h>
+#include <linux/spinlock.h>
+#include <linux/slab.h>
+#include <linux/delay.h>
+#include <linux/list.h>
+#include <linux/pci.h>
+#include <linux/ioport.h>
+#ifdef CONFIG_HIGH_RES_TIMERS
+#include <linux/hrtime.h>
+# if defined(schedule_next_int)
+/* Old high-res timer code, do translations. */
+# define get_arch_cycles(a) quick_update_jiffies_sub(a)
+# define arch_cycles_per_jiffy cycles_per_jiffies
+# endif
+static inline void add_usec_to_timer(struct timer_list *t, long v)
+{
+ t->sub_expires += nsec_to_arch_cycle(v * 1000);
+ while (t->sub_expires >= arch_cycles_per_jiffy)
+ {
+ t->expires++;
+ t->sub_expires -= arch_cycles_per_jiffy;
+ }
+}
+#endif
+#include <linux/interrupt.h>
+#include <linux/rcupdate.h>
+#include <linux/ipmi_smi.h>
+#include <asm/io.h>
+#include "ipmi_si_sm.h"
+#include <linux/init.h>
+
+#define IPMI_SI_VERSION "v31"
+
+/* Measure times between events in the driver. */
+#undef DEBUG_TIMING
+
+/* Call every 10 ms. */
+#define SI_TIMEOUT_TIME_USEC 10000
+#define SI_USEC_PER_JIFFY (1000000/HZ)
+#define SI_TIMEOUT_JIFFIES (SI_TIMEOUT_TIME_USEC/SI_USEC_PER_JIFFY)
+#define SI_SHORT_TIMEOUT_USEC 250 /* .25ms when the SM request a
+ short timeout */
+
+enum si_intf_state {
+ SI_NORMAL,
+ SI_GETTING_FLAGS,
+ SI_GETTING_EVENTS,
+ SI_CLEARING_FLAGS,
+ SI_CLEARING_FLAGS_THEN_SET_IRQ,
+ SI_GETTING_MESSAGES,
+ SI_ENABLE_INTERRUPTS1,
+ SI_ENABLE_INTERRUPTS2
+ /* FIXME - add watchdog stuff. */
+};
+
+enum si_type {
+ SI_KCS, SI_SMIC, SI_BT
+};
+
+struct smi_info
+{
+ ipmi_smi_t intf;
+ struct si_sm_data *si_sm;
+ struct si_sm_handlers *handlers;
+ enum si_type si_type;
+ spinlock_t si_lock;
+ spinlock_t msg_lock;
+ struct list_head xmit_msgs;
+ struct list_head hp_xmit_msgs;
+ struct ipmi_smi_msg *curr_msg;
+ enum si_intf_state si_state;
+
+ /* Used to handle the various types of I/O that can occur with
+ IPMI */
+ struct si_sm_io io;
+ int (*io_setup)(struct smi_info *info);
+ void (*io_cleanup)(struct smi_info *info);
+ int (*irq_setup)(struct smi_info *info);
+ void (*irq_cleanup)(struct smi_info *info);
+ unsigned int io_size;
+
+ /* Flags from the last GET_MSG_FLAGS command, used when an ATTN
+ is set to hold the flags until we are done handling everything
+ from the flags. */
+#define RECEIVE_MSG_AVAIL 0x01
+#define EVENT_MSG_BUFFER_FULL 0x02
+#define WDT_PRE_TIMEOUT_INT 0x08
+ unsigned char msg_flags;
+
+ /* If set to true, this will request events the next time the
+ state machine is idle. */
+ atomic_t req_events;
+
+ /* If true, run the state machine to completion on every send
+ call. Generally used after a panic to make sure stuff goes
+ out. */
+ int run_to_completion;
+
+ /* The I/O port of an SI interface. */
+ int port;
+
+ /* zero if no irq; */
+ int irq;
+
+ /* The timer for this si. */
+ struct timer_list si_timer;
+
+ /* The time (in jiffies) the last timeout occurred at. */
+ unsigned long last_timeout_jiffies;
+
+ /* Used to gracefully stop the timer without race conditions. */
+ volatile int stop_operation;
+ volatile int timer_stopped;
+
+ /* The driver will disable interrupts when it gets into a
+ situation where it cannot handle messages due to lack of
+ memory. Once that situation clears up, it will re-enable
+ interrupts. */
+ int interrupt_disabled;
+
+ unsigned char ipmi_si_dev_rev;
+ unsigned char ipmi_si_fw_rev_major;
+ unsigned char ipmi_si_fw_rev_minor;
+ unsigned char ipmi_version_major;
+ unsigned char ipmi_version_minor;
+
+ /* Counters and things for the proc filesystem. */
+ spinlock_t count_lock;
+ unsigned long short_timeouts;
+ unsigned long long_timeouts;
+ unsigned long timeout_restarts;
+ unsigned long idles;
+ unsigned long interrupts;
+ unsigned long attentions;
+ unsigned long flag_fetches;
+ unsigned long hosed_count;
+ unsigned long complete_transactions;
+ unsigned long events;
+ unsigned long watchdog_pretimeouts;
+ unsigned long incoming_messages;
+};
+
+static void si_restart_short_timer(struct smi_info *smi_info);
+
+static void deliver_recv_msg(struct smi_info *smi_info,
+ struct ipmi_smi_msg *msg)
+{
+ /* Deliver the message to the upper layer with the lock
+ released. */
+ spin_unlock(&(smi_info->si_lock));
+ ipmi_smi_msg_received(smi_info->intf, msg);
+ spin_lock(&(smi_info->si_lock));
+}
+
+static void return_hosed_msg(struct smi_info *smi_info)
+{
+ struct ipmi_smi_msg *msg = smi_info->curr_msg;
+
+ /* Make it a reponse */
+ msg->rsp[0] = msg->data[0] | 4;
+ msg->rsp[1] = msg->data[1];
+ msg->rsp[2] = 0xFF; /* Unknown error. */
+ msg->rsp_size = 3;
+
+ smi_info->curr_msg = NULL;
+ deliver_recv_msg(smi_info, msg);
+}
+
+static enum si_sm_result start_next_msg(struct smi_info *smi_info)
+{
+ int rv;
+ struct list_head *entry = NULL;
+#ifdef DEBUG_TIMING
+ struct timeval t;
+#endif
+
+ /* No need to save flags, we aleady have interrupts off and we
+ already hold the SMI lock. */
+ spin_lock(&(smi_info->msg_lock));
+
+ /* Pick the high priority queue first. */
+ if (! list_empty(&(smi_info->hp_xmit_msgs))) {
+ entry = smi_info->hp_xmit_msgs.next;
+ } else if (! list_empty(&(smi_info->xmit_msgs))) {
+ entry = smi_info->xmit_msgs.next;
+ }
+
+ if (!entry) {
+ smi_info->curr_msg = NULL;
+ rv = SI_SM_IDLE;
+ } else {
+ int err;
+
+ list_del(entry);
+ smi_info->curr_msg = list_entry(entry,
+ struct ipmi_smi_msg,
+ link);
+#ifdef DEBUG_TIMING
+ do_gettimeofday(&t);
+ printk("**Start2: %d.%9.9d\n", t.tv_sec, t.tv_usec);
+#endif
+ err = smi_info->handlers->start_transaction(
+ smi_info->si_sm,
+ smi_info->curr_msg->data,
+ smi_info->curr_msg->data_size);
+ if (err) {
+ return_hosed_msg(smi_info);
+ }
+
+ rv = SI_SM_CALL_WITHOUT_DELAY;
+ }
+ spin_unlock(&(smi_info->msg_lock));
+
+ return rv;
+}
+
+static void start_enable_irq(struct smi_info *smi_info)
+{
+ unsigned char msg[2];
+
+ /* If we are enabling interrupts, we have to tell the
+ BMC to use them. */
+ msg[0] = (IPMI_NETFN_APP_REQUEST << 2);
+ msg[1] = IPMI_GET_BMC_GLOBAL_ENABLES_CMD;
+
+ smi_info->handlers->start_transaction(smi_info->si_sm, msg, 2);
+ smi_info->si_state = SI_ENABLE_INTERRUPTS1;
+}
+
+static void start_clear_flags(struct smi_info *smi_info)
+{
+ unsigned char msg[3];
+
+ /* Make sure the watchdog pre-timeout flag is not set at startup. */
+ msg[0] = (IPMI_NETFN_APP_REQUEST << 2);
+ msg[1] = IPMI_CLEAR_MSG_FLAGS_CMD;
+ msg[2] = WDT_PRE_TIMEOUT_INT;
+
+ smi_info->handlers->start_transaction(smi_info->si_sm, msg, 3);
+ smi_info->si_state = SI_CLEARING_FLAGS;
+}
+
+/* When we have a situtaion where we run out of memory and cannot
+ allocate messages, we just leave them in the BMC and run the system
+ polled until we can allocate some memory. Once we have some
+ memory, we will re-enable the interrupt. */
+static inline void disable_si_irq(struct smi_info *smi_info)
+{
+ if ((smi_info->irq) && (!smi_info->interrupt_disabled)) {
+ disable_irq_nosync(smi_info->irq);
+ smi_info->interrupt_disabled = 1;
+ }
+}
+
+static inline void enable_si_irq(struct smi_info *smi_info)
+{
+ if ((smi_info->irq) && (smi_info->interrupt_disabled)) {
+ enable_irq(smi_info->irq);
+ smi_info->interrupt_disabled = 0;
+ }
+}
+
+static void handle_flags(struct smi_info *smi_info)
+{
+ if (smi_info->msg_flags & WDT_PRE_TIMEOUT_INT) {
+ /* Watchdog pre-timeout */
+ spin_lock(&smi_info->count_lock);
+ smi_info->watchdog_pretimeouts++;
+ spin_unlock(&smi_info->count_lock);
+
+ start_clear_flags(smi_info);
+ smi_info->msg_flags &= ~WDT_PRE_TIMEOUT_INT;
+ spin_unlock(&(smi_info->si_lock));
+ ipmi_smi_watchdog_pretimeout(smi_info->intf);
+ spin_lock(&(smi_info->si_lock));
+ } else if (smi_info->msg_flags & RECEIVE_MSG_AVAIL) {
+ /* Messages available. */
+ smi_info->curr_msg = ipmi_alloc_smi_msg();
+ if (!smi_info->curr_msg) {
+ disable_si_irq(smi_info);
+ smi_info->si_state = SI_NORMAL;
+ return;
+ }
+ enable_si_irq(smi_info);
+
+ smi_info->curr_msg->data[0] = (IPMI_NETFN_APP_REQUEST << 2);
+ smi_info->curr_msg->data[1] = IPMI_GET_MSG_CMD;
+ smi_info->curr_msg->data_size = 2;
+
+ smi_info->handlers->start_transaction(
+ smi_info->si_sm,
+ smi_info->curr_msg->data,
+ smi_info->curr_msg->data_size);
+ smi_info->si_state = SI_GETTING_MESSAGES;
+ } else if (smi_info->msg_flags & EVENT_MSG_BUFFER_FULL) {
+ /* Events available. */
+ smi_info->curr_msg = ipmi_alloc_smi_msg();
+ if (!smi_info->curr_msg) {
+ disable_si_irq(smi_info);
+ smi_info->si_state = SI_NORMAL;
+ return;
+ }
+ enable_si_irq(smi_info);
+
+ smi_info->curr_msg->data[0] = (IPMI_NETFN_APP_REQUEST << 2);
+ smi_info->curr_msg->data[1] = IPMI_READ_EVENT_MSG_BUFFER_CMD;
+ smi_info->curr_msg->data_size = 2;
+
+ smi_info->handlers->start_transaction(
+ smi_info->si_sm,
+ smi_info->curr_msg->data,
+ smi_info->curr_msg->data_size);
+ smi_info->si_state = SI_GETTING_EVENTS;
+ } else {
+ smi_info->si_state = SI_NORMAL;
+ }
+}
+
+static void handle_transaction_done(struct smi_info *smi_info)
+{
+ struct ipmi_smi_msg *msg;
+#ifdef DEBUG_TIMING
+ struct timeval t;
+
+ do_gettimeofday(&t);
+ printk("**Done: %d.%9.9d\n", t.tv_sec, t.tv_usec);
+#endif
+ switch (smi_info->si_state) {
+ case SI_NORMAL:
+ if (!smi_info->curr_msg)
+ break;
+
+ smi_info->curr_msg->rsp_size
+ = smi_info->handlers->get_result(
+ smi_info->si_sm,
+ smi_info->curr_msg->rsp,
+ IPMI_MAX_MSG_LENGTH);
+
+ /* Do this here becase deliver_recv_msg() releases the
+ lock, and a new message can be put in during the
+ time the lock is released. */
+ msg = smi_info->curr_msg;
+ smi_info->curr_msg = NULL;
+ deliver_recv_msg(smi_info, msg);
+ break;
+
+ case SI_GETTING_FLAGS:
+ {
+ unsigned char msg[4];
+ unsigned int len;
+
+ /* We got the flags from the SMI, now handle them. */
+ len = smi_info->handlers->get_result(smi_info->si_sm, msg, 4);
+ if (msg[2] != 0) {
+ /* Error fetching flags, just give up for
+ now. */
+ smi_info->si_state = SI_NORMAL;
+ } else if (len < 3) {
+ /* Hmm, no flags. That's technically illegal, but
+ don't use uninitialized data. */
+ smi_info->si_state = SI_NORMAL;
+ } else {
+ smi_info->msg_flags = msg[3];
+ handle_flags(smi_info);
+ }
+ break;
+ }
+
+ case SI_CLEARING_FLAGS:
+ case SI_CLEARING_FLAGS_THEN_SET_IRQ:
+ {
+ unsigned char msg[3];
+
+ /* We cleared the flags. */
+ smi_info->handlers->get_result(smi_info->si_sm, msg, 3);
+ if (msg[2] != 0) {
+ /* Error clearing flags */
+ printk(KERN_WARNING
+ "ipmi_si: Error clearing flags: %2.2x\n",
+ msg[2]);
+ }
+ if (smi_info->si_state == SI_CLEARING_FLAGS_THEN_SET_IRQ)
+ start_enable_irq(smi_info);
+ else
+ smi_info->si_state = SI_NORMAL;
+ break;
+ }
+
+ case SI_GETTING_EVENTS:
+ {
+ smi_info->curr_msg->rsp_size
+ = smi_info->handlers->get_result(
+ smi_info->si_sm,
+ smi_info->curr_msg->rsp,
+ IPMI_MAX_MSG_LENGTH);
+
+ /* Do this here becase deliver_recv_msg() releases the
+ lock, and a new message can be put in during the
+ time the lock is released. */
+ msg = smi_info->curr_msg;
+ smi_info->curr_msg = NULL;
+ if (msg->rsp[2] != 0) {
+ /* Error getting event, probably done. */
+ msg->done(msg);
+
+ /* Take off the event flag. */
+ smi_info->msg_flags &= ~EVENT_MSG_BUFFER_FULL;
+ } else {
+ spin_lock(&smi_info->count_lock);
+ smi_info->events++;
+ spin_unlock(&smi_info->count_lock);
+
+ deliver_recv_msg(smi_info, msg);
+ }
+ handle_flags(smi_info);
+ break;
+ }
+
+ case SI_GETTING_MESSAGES:
+ {
+ smi_info->curr_msg->rsp_size
+ = smi_info->handlers->get_result(
+ smi_info->si_sm,
+ smi_info->curr_msg->rsp,
+ IPMI_MAX_MSG_LENGTH);
+
+ /* Do this here becase deliver_recv_msg() releases the
+ lock, and a new message can be put in during the
+ time the lock is released. */
+ msg = smi_info->curr_msg;
+ smi_info->curr_msg = NULL;
+ if (msg->rsp[2] != 0) {
+ /* Error getting event, probably done. */
+ msg->done(msg);
+
+ /* Take off the msg flag. */
+ smi_info->msg_flags &= ~RECEIVE_MSG_AVAIL;
+ } else {
+ spin_lock(&smi_info->count_lock);
+ smi_info->incoming_messages++;
+ spin_unlock(&smi_info->count_lock);
+
+ deliver_recv_msg(smi_info, msg);
+ }
+ handle_flags(smi_info);
+ break;
+ }
+
+ case SI_ENABLE_INTERRUPTS1:
+ {
+ unsigned char msg[4];
+
+ /* We got the flags from the SMI, now handle them. */
+ smi_info->handlers->get_result(smi_info->si_sm, msg, 4);
+ if (msg[2] != 0) {
+ printk(KERN_WARNING
+ "ipmi_si: Could not enable interrupts"
+ ", failed get, using polled mode.\n");
+ smi_info->si_state = SI_NORMAL;
+ } else {
+ msg[0] = (IPMI_NETFN_APP_REQUEST << 2);
+ msg[1] = IPMI_SET_BMC_GLOBAL_ENABLES_CMD;
+ msg[2] = msg[3] | 1; /* enable msg queue int */
+ smi_info->handlers->start_transaction(
+ smi_info->si_sm, msg, 3);
+ smi_info->si_state = SI_ENABLE_INTERRUPTS2;
+ }
+ break;
+ }
+
+ case SI_ENABLE_INTERRUPTS2:
+ {
+ unsigned char msg[4];
+
+ /* We got the flags from the SMI, now handle them. */
+ smi_info->handlers->get_result(smi_info->si_sm, msg, 4);
+ if (msg[2] != 0) {
+ printk(KERN_WARNING
+ "ipmi_si: Could not enable interrupts"
+ ", failed set, using polled mode.\n");
+ }
+ smi_info->si_state = SI_NORMAL;
+ break;
+ }
+ }
+}
+
+/* Called on timeouts and events. Timeouts should pass the elapsed
+ time, interrupts should pass in zero. */
+static enum si_sm_result smi_event_handler(struct smi_info *smi_info,
+ int time)
+{
+ enum si_sm_result si_sm_result;
+
+ restart:
+ /* There used to be a loop here that waited a little while
+ (around 25us) before giving up. That turned out to be
+ pointless, the minimum delays I was seeing were in the 300us
+ range, which is far too long to wait in an interrupt. So
+ we just run until the state machine tells us something
+ happened or it needs a delay. */
+ si_sm_result = smi_info->handlers->event(smi_info->si_sm, time);
+ time = 0;
+ while (si_sm_result == SI_SM_CALL_WITHOUT_DELAY)
+ {
+ si_sm_result = smi_info->handlers->event(smi_info->si_sm, 0);
+ }
+
+ if (si_sm_result == SI_SM_TRANSACTION_COMPLETE)
+ {
+ spin_lock(&smi_info->count_lock);
+ smi_info->complete_transactions++;
+ spin_unlock(&smi_info->count_lock);
+
+ handle_transaction_done(smi_info);
+ si_sm_result = smi_info->handlers->event(smi_info->si_sm, 0);
+ }
+ else if (si_sm_result == SI_SM_HOSED)
+ {
+ spin_lock(&smi_info->count_lock);
+ smi_info->hosed_count++;
+ spin_unlock(&smi_info->count_lock);
+
+ if (smi_info->curr_msg != NULL) {
+ /* If we were handling a user message, format
+ a response to send to the upper layer to
+ tell it about the error. */
+ return_hosed_msg(smi_info);
+ }
+ si_sm_result = smi_info->handlers->event(smi_info->si_sm, 0);
+ smi_info->si_state = SI_NORMAL;
+ }
+
+ /* We prefer handling attn over new messages. */
+ if (si_sm_result == SI_SM_ATTN)
+ {
+ unsigned char msg[2];
+
+ spin_lock(&smi_info->count_lock);
+ smi_info->attentions++;
+ spin_unlock(&smi_info->count_lock);
+
+ /* Got a attn, send down a get message flags to see
+ what's causing it. It would be better to handle
+ this in the upper layer, but due to the way
+ interrupts work with the SMI, that's not really
+ possible. */
+ msg[0] = (IPMI_NETFN_APP_REQUEST << 2);
+ msg[1] = IPMI_GET_MSG_FLAGS_CMD;
+
+ smi_info->handlers->start_transaction(
+ smi_info->si_sm, msg, 2);
+ smi_info->si_state = SI_GETTING_FLAGS;
+ goto restart;
+ }
+
+ /* If we are currently idle, try to start the next message. */
+ if (si_sm_result == SI_SM_IDLE) {
+ spin_lock(&smi_info->count_lock);
+ smi_info->idles++;
+ spin_unlock(&smi_info->count_lock);
+
+ si_sm_result = start_next_msg(smi_info);
+ if (si_sm_result != SI_SM_IDLE)
+ goto restart;
+ }
+
+ if ((si_sm_result == SI_SM_IDLE)
+ && (atomic_read(&smi_info->req_events)))
+ {
+ /* We are idle and the upper layer requested that I fetch
+ events, so do so. */
+ unsigned char msg[2];
+
+ spin_lock(&smi_info->count_lock);
+ smi_info->flag_fetches++;
+ spin_unlock(&smi_info->count_lock);
+
+ atomic_set(&smi_info->req_events, 0);
+ msg[0] = (IPMI_NETFN_APP_REQUEST << 2);
+ msg[1] = IPMI_GET_MSG_FLAGS_CMD;
+
+ smi_info->handlers->start_transaction(
+ smi_info->si_sm, msg, 2);
+ smi_info->si_state = SI_GETTING_FLAGS;
+ goto restart;
+ }
+
+ return si_sm_result;
+}
+
+static void sender(void *send_info,
+ struct ipmi_smi_msg *msg,
+ int priority)
+{
+ struct smi_info *smi_info = send_info;
+ enum si_sm_result result;
+ unsigned long flags;
+#ifdef DEBUG_TIMING
+ struct timeval t;
+#endif
+
+ spin_lock_irqsave(&(smi_info->msg_lock), flags);
+#ifdef DEBUG_TIMING
+ do_gettimeofday(&t);
+ printk("**Enqueue: %d.%9.9d\n", t.tv_sec, t.tv_usec);
+#endif
+
+ if (smi_info->run_to_completion) {
+ /* If we are running to completion, then throw it in
+ the list and run transactions until everything is
+ clear. Priority doesn't matter here. */
+ list_add_tail(&(msg->link), &(smi_info->xmit_msgs));
+
+ /* We have to release the msg lock and claim the smi
+ lock in this case, because of race conditions. */
+ spin_unlock_irqrestore(&(smi_info->msg_lock), flags);
+
+ spin_lock_irqsave(&(smi_info->si_lock), flags);
+ result = smi_event_handler(smi_info, 0);
+ while (result != SI_SM_IDLE) {
+ udelay(SI_SHORT_TIMEOUT_USEC);
+ result = smi_event_handler(smi_info,
+ SI_SHORT_TIMEOUT_USEC);
+ }
+ spin_unlock_irqrestore(&(smi_info->si_lock), flags);
+ return;
+ } else {
+ if (priority > 0) {
+ list_add_tail(&(msg->link), &(smi_info->hp_xmit_msgs));
+ } else {
+ list_add_tail(&(msg->link), &(smi_info->xmit_msgs));
+ }
+ }
+ spin_unlock_irqrestore(&(smi_info->msg_lock), flags);
+
+ spin_lock_irqsave(&(smi_info->si_lock), flags);
+ if ((smi_info->si_state == SI_NORMAL)
+ && (smi_info->curr_msg == NULL))
+ {
+ start_next_msg(smi_info);
+ si_restart_short_timer(smi_info);
+ }
+ spin_unlock_irqrestore(&(smi_info->si_lock), flags);
+}
+
+static void set_run_to_completion(void *send_info, int i_run_to_completion)
+{
+ struct smi_info *smi_info = send_info;
+ enum si_sm_result result;
+ unsigned long flags;
+
+ spin_lock_irqsave(&(smi_info->si_lock), flags);
+
+ smi_info->run_to_completion = i_run_to_completion;
+ if (i_run_to_completion) {
+ result = smi_event_handler(smi_info, 0);
+ while (result != SI_SM_IDLE) {
+ udelay(SI_SHORT_TIMEOUT_USEC);
+ result = smi_event_handler(smi_info,
+ SI_SHORT_TIMEOUT_USEC);
+ }
+ }
+
+ spin_unlock_irqrestore(&(smi_info->si_lock), flags);
+}
+
+static void request_events(void *send_info)
+{
+ struct smi_info *smi_info = send_info;
+
+ atomic_set(&smi_info->req_events, 1);
+}
+
+static int initialized = 0;
+
+/* Must be called with interrupts off and with the si_lock held. */
+static void si_restart_short_timer(struct smi_info *smi_info)
+{
+#if defined(CONFIG_HIGH_RES_TIMERS)
+ unsigned long flags;
+ unsigned long jiffies_now;
+
+ if (del_timer(&(smi_info->si_timer))) {
+ /* If we don't delete the timer, then it will go off
+ immediately, anyway. So we only process if we
+ actually delete the timer. */
+
+ /* We already have irqsave on, so no need for it
+ here. */
+ read_lock(&xtime_lock);
+ jiffies_now = jiffies;
+ smi_info->si_timer.expires = jiffies_now;
+ smi_info->si_timer.sub_expires = get_arch_cycles(jiffies_now);
+
+ add_usec_to_timer(&smi_info->si_timer, SI_SHORT_TIMEOUT_USEC);
+
+ add_timer(&(smi_info->si_timer));
+ spin_lock_irqsave(&smi_info->count_lock, flags);
+ smi_info->timeout_restarts++;
+ spin_unlock_irqrestore(&smi_info->count_lock, flags);
+ }
+#endif
+}
+
+static void smi_timeout(unsigned long data)
+{
+ struct smi_info *smi_info = (struct smi_info *) data;
+ enum si_sm_result smi_result;
+ unsigned long flags;
+ unsigned long jiffies_now;
+ unsigned long time_diff;
+#ifdef DEBUG_TIMING
+ struct timeval t;
+#endif
+
+ if (smi_info->stop_operation) {
+ smi_info->timer_stopped = 1;
+ return;
+ }
+
+ spin_lock_irqsave(&(smi_info->si_lock), flags);
+#ifdef DEBUG_TIMING
+ do_gettimeofday(&t);
+ printk("**Timer: %d.%9.9d\n", t.tv_sec, t.tv_usec);
+#endif
+ jiffies_now = jiffies;
+ time_diff = ((jiffies_now - smi_info->last_timeout_jiffies)
+ * SI_USEC_PER_JIFFY);
+ smi_result = smi_event_handler(smi_info, time_diff);
+
+ spin_unlock_irqrestore(&(smi_info->si_lock), flags);
+
+ smi_info->last_timeout_jiffies = jiffies_now;
+
+ if ((smi_info->irq) && (! smi_info->interrupt_disabled)) {
+ /* Running with interrupts, only do long timeouts. */
+ smi_info->si_timer.expires = jiffies + SI_TIMEOUT_JIFFIES;
+ spin_lock_irqsave(&smi_info->count_lock, flags);
+ smi_info->long_timeouts++;
+ spin_unlock_irqrestore(&smi_info->count_lock, flags);
+ goto do_add_timer;
+ }
+
+ /* If the state machine asks for a short delay, then shorten
+ the timer timeout. */
+ if (smi_result == SI_SM_CALL_WITH_DELAY) {
+ spin_lock_irqsave(&smi_info->count_lock, flags);
+ smi_info->short_timeouts++;
+ spin_unlock_irqrestore(&smi_info->count_lock, flags);
+#if defined(CONFIG_HIGH_RES_TIMERS)
+ read_lock(&xtime_lock);
+ smi_info->si_timer.expires = jiffies;
+ smi_info->si_timer.sub_expires
+ = get_arch_cycles(smi_info->si_timer.expires);
+ read_unlock(&xtime_lock);
+ add_usec_to_timer(&smi_info->si_timer, SI_SHORT_TIMEOUT_USEC);
+#else
+ smi_info->si_timer.expires = jiffies + 1;
+#endif
+ } else {
+ spin_lock_irqsave(&smi_info->count_lock, flags);
+ smi_info->long_timeouts++;
+ spin_unlock_irqrestore(&smi_info->count_lock, flags);
+ smi_info->si_timer.expires = jiffies + SI_TIMEOUT_JIFFIES;
+#if defined(CONFIG_HIGH_RES_TIMERS)
+ smi_info->si_timer.sub_expires = 0;
+#endif
+ }
+
+ do_add_timer:
+ add_timer(&(smi_info->si_timer));
+}
+
+static irqreturn_t si_irq_handler(int irq, void *data, struct pt_regs *regs)
+{
+ struct smi_info *smi_info = data;
+ unsigned long flags;
+#ifdef DEBUG_TIMING
+ struct timeval t;
+#endif
+
+ spin_lock_irqsave(&(smi_info->si_lock), flags);
+
+ spin_lock(&smi_info->count_lock);
+ smi_info->interrupts++;
+ spin_unlock(&smi_info->count_lock);
+
+ if (smi_info->stop_operation)
+ goto out;
+
+#ifdef DEBUG_TIMING
+ do_gettimeofday(&t);
+ printk("**Interrupt: %d.%9.9d\n", t.tv_sec, t.tv_usec);
+#endif
+ smi_event_handler(smi_info, 0);
+ out:
+ spin_unlock_irqrestore(&(smi_info->si_lock), flags);
+ return IRQ_HANDLED;
+}
+
+static struct ipmi_smi_handlers handlers =
+{
+ .owner = THIS_MODULE,
+ .sender = sender,
+ .request_events = request_events,
+ .set_run_to_completion = set_run_to_completion
+};
+
+/* There can be 4 IO ports passed in (with or without IRQs), 4 addresses,
+ a default IO port, and 1 ACPI/SPMI address. That sets SI_MAX_DRIVERS */
+
+#define SI_MAX_PARMS 4
+#define SI_MAX_DRIVERS ((SI_MAX_PARMS * 2) + 2)
+static struct smi_info *smi_infos[SI_MAX_DRIVERS] =
+{ NULL, NULL, NULL, NULL };
+
+#define DEVICE_NAME "ipmi_si"
+
+#define DEFAULT_KCS_IO_PORT 0xca2
+#define DEFAULT_SMIC_IO_PORT 0xca9
+#define DEFAULT_BT_IO_PORT 0xe4
+
+static int si_trydefaults = 1;
+static char *si_type[SI_MAX_PARMS] = { NULL, NULL, NULL, NULL };
+#define MAX_SI_TYPE_STR 30
+static char si_type_str[MAX_SI_TYPE_STR];
+static unsigned long addrs[SI_MAX_PARMS] = { 0, 0, 0, 0 };
+static int num_addrs = 0;
+static unsigned int ports[SI_MAX_PARMS] = { 0, 0, 0, 0 };
+static int num_ports = 0;
+static int irqs[SI_MAX_PARMS] = { 0, 0, 0, 0 };
+static int num_irqs = 0;
+
+
+module_param_named(trydefaults, si_trydefaults, bool, 0);
+MODULE_PARM_DESC(trydefaults, "Setting this to 'false' will disable the"
+ " default scan of the KCS and SMIC interface at the standard"
+ " address");
+module_param_string(type, si_type_str, MAX_SI_TYPE_STR, 0);
+MODULE_PARM_DESC(type, "Defines the type of each interface, each"
+ " interface separated by commas. The types are 'kcs',"
+ " 'smic', and 'bt'. For example si_type=kcs,bt will set"
+ " the first interface to kcs and the second to bt");
+module_param_array(addrs, long, num_addrs, 0);
+MODULE_PARM_DESC(addrs, "Sets the memory address of each interface, the"
+ " addresses separated by commas. Only use if an interface"
+ " is in memory. Otherwise, set it to zero or leave"
+ " it blank.");
+module_param_array(ports, int, num_ports, 0);
+MODULE_PARM_DESC(ports, "Sets the port address of each interface, the"
+ " addresses separated by commas. Only use if an interface"
+ " is a port. Otherwise, set it to zero or leave"
+ " it blank.");
+module_param_array(irqs, int, num_irqs, 0);
+MODULE_PARM_DESC(irqs, "Sets the interrupt of each interface, the"
+ " addresses separated by commas. Only use if an interface"
+ " has an interrupt. Otherwise, set it to zero or leave"
+ " it blank.");
+
+
+#if defined(CONFIG_ACPI_INTERPETER) || defined(CONFIG_X86) || defined(CONFIG_PCI)
+#define IPMI_MEM_ADDR_SPACE 1
+#define IPMI_IO_ADDR_SPACE 2
+static int is_new_interface(int intf, u8 addr_space, unsigned long base_addr)
+{
+ int i;
+
+ for (i = 0; i < SI_MAX_PARMS; ++i) {
+ /* Don't check our address. */
+ if (i == intf)
+ continue;
+ if (si_type[i] != NULL) {
+ if ((addr_space == IPMI_MEM_ADDR_SPACE &&
+ base_addr == addrs[i]) ||
+ (addr_space == IPMI_IO_ADDR_SPACE &&
+ base_addr == ports[i]))
+ return 0;
+ }
+ else
+ break;
+ }
+
+ return 1;
+}
+#endif
+
+static int std_irq_setup(struct smi_info *info)
+{
+ int rv;
+
+ if (!info->irq)
+ return 0;
+
+ rv = request_irq(info->irq,
+ si_irq_handler,
+ SA_INTERRUPT,
+ DEVICE_NAME,
+ info);
+ if (rv) {
+ printk(KERN_WARNING
+ "ipmi_si: %s unable to claim interrupt %d,"
+ " running polled\n",
+ DEVICE_NAME, info->irq);
+ info->irq = 0;
+ } else {
+ printk(" Using irq %d\n", info->irq);
+ }
+
+ return rv;
+}
+
+static void std_irq_cleanup(struct smi_info *info)
+{
+ if (!info->irq)
+ return;
+
+ free_irq(info->irq, info);
+}
+
+static unsigned char port_inb(struct si_sm_io *io, unsigned int offset)
+{
+ unsigned int *addr = io->info;
+
+ return inb((*addr)+offset);
+}
+
+static void port_outb(struct si_sm_io *io, unsigned int offset,
+ unsigned char b)
+{
+ unsigned int *addr = io->info;
+
+ outb(b, (*addr)+offset);
+}
+
+static int port_setup(struct smi_info *info)
+{
+ unsigned int *addr = info->io.info;
+
+ if (!addr || (!*addr))
+ return -ENODEV;
+
+ if (request_region(*addr, info->io_size, DEVICE_NAME) == NULL)
+ return -EIO;
+ return 0;
+}
+
+static void port_cleanup(struct smi_info *info)
+{
+ unsigned int *addr = info->io.info;
+
+ if (addr && (*addr))
+ release_region (*addr, info->io_size);
+ kfree(info);
+}
+
+static int try_init_port(int intf_num, struct smi_info **new_info)
+{
+ struct smi_info *info;
+
+ if (!ports[intf_num])
+ return -ENODEV;
+
+ if (!is_new_interface(intf_num, IPMI_IO_ADDR_SPACE,
+ ports[intf_num]))
+ return -ENODEV;
+
+ info = kmalloc(sizeof(*info), GFP_KERNEL);
+ if (!info) {
+ printk(KERN_ERR "ipmi_si: Could not allocate SI data (1)\n");
+ return -ENOMEM;
+ }
+ memset(info, 0, sizeof(*info));
+
+ info->io_setup = port_setup;
+ info->io_cleanup = port_cleanup;
+ info->io.inputb = port_inb;
+ info->io.outputb = port_outb;
+ info->io.info = &(ports[intf_num]);
+ info->io.addr = NULL;
+ info->irq = 0;
+ info->irq_setup = NULL;
+ *new_info = info;
+
+ if (si_type[intf_num] == NULL)
+ si_type[intf_num] = "kcs";
+
+ printk("ipmi_si: Trying \"%s\" at I/O port 0x%x\n",
+ si_type[intf_num], ports[intf_num]);
+ return 0;
+}
+
+static unsigned char mem_inb(struct si_sm_io *io, unsigned int offset)
+{
+ return readb((io->addr)+offset);
+}
+
+static void mem_outb(struct si_sm_io *io, unsigned int offset,
+ unsigned char b)
+{
+ writeb(b, (io->addr)+offset);
+}
+
+static int mem_setup(struct smi_info *info)
+{
+ unsigned long *addr = info->io.info;
+
+ if (!addr || (!*addr))
+ return -ENODEV;
+
+ if (request_mem_region(*addr, info->io_size, DEVICE_NAME) == NULL)
+ return -EIO;
+
+ info->io.addr = ioremap(*addr, info->io_size);
+ if (info->io.addr == NULL) {
+ release_mem_region(*addr, info->io_size);
+ return -EIO;
+ }
+ return 0;
+}
+
+static void mem_cleanup(struct smi_info *info)
+{
+ unsigned long *addr = info->io.info;
+
+ if (info->io.addr) {
+ iounmap(info->io.addr);
+ release_mem_region(*addr, info->io_size);
+ }
+ kfree(info);
+}
+
+static int try_init_mem(int intf_num, struct smi_info **new_info)
+{
+ struct smi_info *info;
+
+ if (!addrs[intf_num])
+ return -ENODEV;
+
+ if (!is_new_interface(intf_num, IPMI_MEM_ADDR_SPACE,
+ addrs[intf_num]))
+ return -ENODEV;
+
+ info = kmalloc(sizeof(*info), GFP_KERNEL);
+ if (!info) {
+ printk(KERN_ERR "ipmi_si: Could not allocate SI data (2)\n");
+ return -ENOMEM;
+ }
+ memset(info, 0, sizeof(*info));
+
+ info->io_setup = mem_setup;
+ info->io_cleanup = mem_cleanup;
+ info->io.inputb = mem_inb;
+ info->io.outputb = mem_outb;
+ info->io.info = (void *) addrs[intf_num];
+ info->io.addr = NULL;
+ info->irq = 0;
+ info->irq_setup = NULL;
+ *new_info = info;
+
+ if (si_type[intf_num] == NULL)
+ si_type[intf_num] = "kcs";
+
+ printk("ipmi_si: Trying \"%s\" at memory address 0x%lx\n",
+ si_type[intf_num], addrs[intf_num]);
+ return 0;
+}
+
+
+#ifdef CONFIG_ACPI_INTERPRETER
+
+#include <linux/acpi.h>
+
+/* Once we get an ACPI failure, we don't try any more, because we go
+ through the tables sequentially. Once we don't find a table, there
+ are no more. */
+static int acpi_failure = 0;
+
+/* For GPE-type interrupts. */
+void ipmi_acpi_gpe(void *context)
+{
+ struct smi_info *smi_info = context;
+ unsigned long flags;
+#ifdef DEBUG_TIMING
+ struct timeval t;
+#endif
+
+ spin_lock_irqsave(&(smi_info->si_lock), flags);
+
+ spin_lock(&smi_info->count_lock);
+ smi_info->interrupts++;
+ spin_unlock(&smi_info->count_lock);
+
+ if (smi_info->stop_operation)
+ goto out;
+
+#ifdef DEBUG_TIMING
+ do_gettimeofday(&t);
+ printk("**ACPI_GPE: %d.%9.9d\n", t.tv_sec, t.tv_usec);
+#endif
+ smi_event_handler(smi_info, 0);
+ out:
+ spin_unlock_irqrestore(&(smi_info->si_lock), flags);
+}
+
+static int acpi_gpe_irq_setup(struct smi_info *info)
+{
+ acpi_status status;
+
+ if (!info->irq)
+ return 0;
+
+ /* FIXME - is level triggered right? */
+ status = acpi_install_gpe_handler(NULL,
+ info->irq,
+ ACPI_EVENT_LEVEL_TRIGGERED,
+ ipmi_acpi_gpe,
+ info);
+ if (status != AE_OK) {
+ printk(KERN_WARNING
+ "ipmi_si: %s unable to claim ACPI GPE %d,"
+ " running polled\n",
+ DEVICE_NAME, info->irq);
+ info->irq = 0;
+ return -EINVAL;
+ } else {
+ printk(" Using ACPI GPE %d\n", info->irq);
+ return 0;
+ }
+
+}
+
+static void acpi_gpe_irq_cleanup(struct smi_info *info)
+{
+ if (!info->irq)
+ return;
+
+ acpi_remove_gpe_handler(NULL, info->irq, ipmi_acpi_gpe);
+}
+
+/*
+ * Defined at
+ * http://h21007.www2.hp.com/dspp/files/unprotected/devresource/Docs/TechPapers/IA64/hpspmi.pdf
+ */
+struct SPMITable {
+ s8 Signature[4];
+ u32 Length;
+ u8 Revision;
+ u8 Checksum;
+ s8 OEMID[6];
+ s8 OEMTableID[8];
+ s8 OEMRevision[4];
+ s8 CreatorID[4];
+ s8 CreatorRevision[4];
+ u8 InterfaceType;
+ u8 IPMIlegacy;
+ s16 SpecificationRevision;
+
+ /*
+ * Bit 0 - SCI interrupt supported
+ * Bit 1 - I/O APIC/SAPIC
+ */
+ u8 InterruptType;
+
+ /* If bit 0 of InterruptType is set, then this is the SCI
+ interrupt in the GPEx_STS register. */
+ u8 GPE;
+
+ s16 Reserved;
+
+ /* If bit 1 of InterruptType is set, then this is the I/O
+ APIC/SAPIC interrupt. */
+ u32 GlobalSystemInterrupt;
+
+ /* The actual register address. */
+ struct acpi_generic_address addr;
+
+ u8 UID[4];
+
+ s8 spmi_id[1]; /* A '\0' terminated array starts here. */
+};
+
+static int try_init_acpi(int intf_num, struct smi_info **new_info)
+{
+ struct smi_info *info;
+ acpi_status status;
+ struct SPMITable *spmi;
+ char *io_type;
+ u8 addr_space;
+
+ if (acpi_failure)
+ return -ENODEV;
+
+ status = acpi_get_firmware_table("SPMI", intf_num+1,
+ ACPI_LOGICAL_ADDRESSING,
+ (struct acpi_table_header **) &spmi);
+ if (status != AE_OK) {
+ acpi_failure = 1;
+ return -ENODEV;
+ }
+
+ if (spmi->IPMIlegacy != 1) {
+ printk(KERN_INFO "IPMI: Bad SPMI legacy %d\n", spmi->IPMIlegacy);
+ return -ENODEV;
+ }
+
+ if (spmi->addr.address_space_id == ACPI_ADR_SPACE_SYSTEM_MEMORY)
+ addr_space = IPMI_MEM_ADDR_SPACE;
+ else
+ addr_space = IPMI_IO_ADDR_SPACE;
+ if (!is_new_interface(-1, addr_space, spmi->addr.address))
+ return -ENODEV;
+
+ /* Figure out the interface type. */
+ switch (spmi->InterfaceType)
+ {
+ case 1: /* KCS */
+ si_type[intf_num] = "kcs";
+ break;
+
+ case 2: /* SMIC */
+ si_type[intf_num] = "smic";
+ break;
+
+ case 3: /* BT */
+ si_type[intf_num] = "bt";
+ break;
+
+ default:
+ printk(KERN_INFO "ipmi_si: Unknown ACPI/SPMI SI type %d\n",
+ spmi->InterfaceType);
+ return -EIO;
+ }
+
+ info = kmalloc(sizeof(*info), GFP_KERNEL);
+ if (!info) {
+ printk(KERN_ERR "ipmi_si: Could not allocate SI data (3)\n");
+ return -ENOMEM;
+ }
+ memset(info, 0, sizeof(*info));
+
+ if (spmi->InterruptType & 1) {
+ /* We've got a GPE interrupt. */
+ info->irq = spmi->GPE;
+ info->irq_setup = acpi_gpe_irq_setup;
+ info->irq_cleanup = acpi_gpe_irq_cleanup;
+ } else if (spmi->InterruptType & 2) {
+ /* We've got an APIC/SAPIC interrupt. */
+ info->irq = spmi->GlobalSystemInterrupt;
+ info->irq_setup = std_irq_setup;
+ info->irq_cleanup = std_irq_cleanup;
+ } else {
+ /* Use the default interrupt setting. */
+ info->irq = 0;
+ info->irq_setup = NULL;
+ }
+
+ if (spmi->addr.address_space_id == ACPI_ADR_SPACE_SYSTEM_MEMORY) {
+ io_type = "memory";
+ info->io_setup = mem_setup;
+ info->io_cleanup = mem_cleanup;
+ addrs[intf_num] = spmi->addr.address;
+ info->io.inputb = mem_inb;
+ info->io.outputb = mem_outb;
+ info->io.info = &(addrs[intf_num]);
+ } else if (spmi->addr.address_space_id == ACPI_ADR_SPACE_SYSTEM_IO) {
+ io_type = "I/O";
+ info->io_setup = port_setup;
+ info->io_cleanup = port_cleanup;
+ ports[intf_num] = spmi->addr.address;
+ info->io.inputb = port_inb;
+ info->io.outputb = port_outb;
+ info->io.info = &(ports[intf_num]);
+ } else {
+ kfree(info);
+ printk("ipmi_si: Unknown ACPI I/O Address type\n");
+ return -EIO;
+ }
+
+ *new_info = info;
+
+ printk("ipmi_si: ACPI/SPMI specifies \"%s\" %s SI @ 0x%lx\n",
+ si_type[intf_num], io_type, (unsigned long) spmi->addr.address);
+ return 0;
+}
+#endif
+
+#ifdef CONFIG_X86
+
+typedef struct dmi_ipmi_data
+{
+ u8 type;
+ u8 addr_space;
+ unsigned long base_addr;
+ u8 irq;
+}dmi_ipmi_data_t;
+
+typedef struct dmi_header
+{
+ u8 type;
+ u8 length;
+ u16 handle;
+}dmi_header_t;
+
+static int decode_dmi(dmi_header_t *dm, dmi_ipmi_data_t *ipmi_data)
+{
+ u8 *data = (u8 *)dm;
+ unsigned long base_addr;
+
+ ipmi_data->type = data[0x04];
+
+ memcpy(&base_addr,&data[0x08],sizeof(unsigned long));
+ if (base_addr & 1) {
+ /* I/O */
+ base_addr &= 0xFFFE;
+ ipmi_data->addr_space = IPMI_IO_ADDR_SPACE;
+ }
+ else {
+ /* Memory */
+ ipmi_data->addr_space = IPMI_MEM_ADDR_SPACE;
+ }
+
+ ipmi_data->base_addr = base_addr;
+ ipmi_data->irq = data[0x11];
+
+ if (is_new_interface(-1, ipmi_data->addr_space,ipmi_data->base_addr))
+ return 0;
+
+ memset(ipmi_data,0,sizeof(dmi_ipmi_data_t));
+
+ return -1;
+}
+
+static int dmi_table(u32 base, int len, int num,
+ dmi_ipmi_data_t *ipmi_data)
+{
+ u8 *buf;
+ struct dmi_header *dm;
+ u8 *data;
+ int i=1;
+ int status=-1;
+
+ buf = ioremap(base, len);
+ if(buf==NULL)
+ return -1;
+
+ data = buf;
+
+ while(i<num && (data - buf) < len)
+ {
+ dm=(dmi_header_t *)data;
+
+ if((data-buf+dm->length) >= len)
+ break;
+
+ if (dm->type == 38) {
+ if (decode_dmi(dm, ipmi_data) == 0) {
+ status = 0;
+ break;
+ }
+ }
+
+ data+=dm->length;
+ while((data-buf) < len && (*data || data[1]))
+ data++;
+ data+=2;
+ i++;
+ }
+ iounmap(buf);
+
+ return status;
+}
+
+inline static int dmi_checksum(u8 *buf)
+{
+ u8 sum=0;
+ int a;
+
+ for(a=0; a<15; a++)
+ sum+=buf[a];
+ return (sum==0);
+}
+
+static int dmi_iterator(dmi_ipmi_data_t *ipmi_data)
+{
+ u8 buf[15];
+ u32 fp=0xF0000;
+
+#ifdef CONFIG_SIMNOW
+ return -1;
+#endif
+
+ while(fp < 0xFFFFF)
+ {
+ isa_memcpy_fromio(buf, fp, 15);
+ if(memcmp(buf, "_DMI_", 5)==0 && dmi_checksum(buf))
+ {
+ u16 num=buf[13]<<8|buf[12];
+ u16 len=buf[7]<<8|buf[6];
+ u32 base=buf[11]<<24|buf[10]<<16|buf[9]<<8|buf[8];
+
+ if(dmi_table(base, len, num, ipmi_data) == 0)
+ return 0;
+ }
+ fp+=16;
+ }
+
+ return -1;
+}
+
+static int try_init_smbios(int intf_num, struct smi_info **new_info)
+{
+ struct smi_info *info;
+ dmi_ipmi_data_t ipmi_data;
+ char *io_type;
+ int status;
+
+ status = dmi_iterator(&ipmi_data);
+
+ if (status < 0)
+ return -ENODEV;
+
+ switch(ipmi_data.type) {
+ case 0x01: /* KCS */
+ si_type[intf_num] = "kcs";
+ break;
+ case 0x02: /* SMIC */
+ si_type[intf_num] = "smic";
+ break;
+ case 0x03: /* BT */
+ si_type[intf_num] = "bt";
+ break;
+ default:
+ printk("ipmi_si: Unknown SMBIOS SI type.\n");
+ return -EIO;
+ }
+
+ info = kmalloc(sizeof(*info), GFP_KERNEL);
+ if (!info) {
+ printk(KERN_ERR "ipmi_si: Could not allocate SI data (4)\n");
+ return -ENOMEM;
+ }
+ memset(info, 0, sizeof(*info));
+
+ if (ipmi_data.addr_space == 1) {
+ io_type = "memory";
+ info->io_setup = mem_setup;
+ info->io_cleanup = mem_cleanup;
+ addrs[intf_num] = ipmi_data.base_addr;
+ info->io.inputb = mem_inb;
+ info->io.outputb = mem_outb;
+ info->io.info = &(addrs[intf_num]);
+ } else if (ipmi_data.addr_space == 2) {
+ io_type = "I/O";
+ info->io_setup = port_setup;
+ info->io_cleanup = port_cleanup;
+ ports[intf_num] = ipmi_data.base_addr;
+ info->io.inputb = port_inb;
+ info->io.outputb = port_outb;
+ info->io.info = &(ports[intf_num]);
+ } else {
+ kfree(info);
+ printk("ipmi_si: Unknown SMBIOS I/O Address type.\n");
+ return -EIO;
+ }
+
+ irqs[intf_num] = ipmi_data.irq;
+
+ *new_info = info;
+
+ printk("ipmi_si: Found SMBIOS-specified state machine at %s"
+ " address 0x%lx\n",
+ io_type, (unsigned long)ipmi_data.base_addr);
+ return 0;
+}
+#endif /* CONFIG_X86 */
+
+#ifdef CONFIG_PCI
+
+#define PCI_ERMC_CLASSCODE 0x0C0700
+#define PCI_HP_VENDOR_ID 0x103C
+#define PCI_MMC_DEVICE_ID 0x121A
+#define PCI_MMC_ADDR_CW 0x10
+
+/* Avoid more than one attempt to probe pci smic. */
+static int pci_smic_checked = 0;
+
+static int find_pci_smic(int intf_num, struct smi_info **new_info)
+{
+ struct smi_info *info;
+ int error;
+ struct pci_dev *pci_dev = NULL;
+ u16 base_addr;
+ int fe_rmc = 0;
+
+ if (pci_smic_checked)
+ return -ENODEV;
+
+ pci_smic_checked = 1;
+
+ if ((pci_dev = pci_find_device(PCI_HP_VENDOR_ID, PCI_MMC_DEVICE_ID,
+ NULL)))
+ ;
+ else if ((pci_dev = pci_find_class(PCI_ERMC_CLASSCODE, NULL)) &&
+ pci_dev->subsystem_vendor == PCI_HP_VENDOR_ID)
+ fe_rmc = 1;
+ else
+ return -ENODEV;
+
+ error = pci_read_config_word(pci_dev, PCI_MMC_ADDR_CW, &base_addr);
+ if (error)
+ {
+ printk(KERN_ERR
+ "ipmi_si: pci_read_config_word() failed (%d).\n",
+ error);
+ return -ENODEV;
+ }
+
+ /* Bit 0: 1 specifies programmed I/O, 0 specifies memory mapped I/O */
+ if (!(base_addr & 0x0001))
+ {
+ printk(KERN_ERR
+ "ipmi_si: memory mapped I/O not supported for PCI"
+ " smic.\n");
+ return -ENODEV;
+ }
+
+ base_addr &= 0xFFFE;
+ if (!fe_rmc)
+ /* Data register starts at base address + 1 in eRMC */
+ ++base_addr;
+
+ if (!is_new_interface(-1, IPMI_IO_ADDR_SPACE, base_addr))
+ return -ENODEV;
+
+ info = kmalloc(sizeof(*info), GFP_KERNEL);
+ if (!info) {
+ printk(KERN_ERR "ipmi_si: Could not allocate SI data (5)\n");
+ return -ENOMEM;
+ }
+ memset(info, 0, sizeof(*info));
+
+ info->io_setup = port_setup;
+ info->io_cleanup = port_cleanup;
+ ports[intf_num] = base_addr;
+ info->io.inputb = port_inb;
+ info->io.outputb = port_outb;
+ info->io.info = &(ports[intf_num]);
+
+ *new_info = info;
+
+ irqs[intf_num] = pci_dev->irq;
+ si_type[intf_num] = "smic";
+
+ printk("ipmi_si: Found PCI SMIC at I/O address 0x%lx\n",
+ (long unsigned int) base_addr);
+
+ return 0;
+}
+#endif /* CONFIG_PCI */
+
+static int try_init_plug_and_play(int intf_num, struct smi_info **new_info)
+{
+#ifdef CONFIG_PCI
+ if (find_pci_smic(intf_num, new_info)==0)
+ return 0;
+#endif
+ /* Include other methods here. */
+
+ return -ENODEV;
+}
+
+
+static int try_get_dev_id(struct smi_info *smi_info)
+{
+ unsigned char msg[2];
+ unsigned char *resp;
+ unsigned long resp_len;
+ enum si_sm_result smi_result;
+ int rv = 0;
+
+ resp = kmalloc(IPMI_MAX_MSG_LENGTH, GFP_KERNEL);
+ if (!resp)
+ return -ENOMEM;
+
+ /* Do a Get Device ID command, since it comes back with some
+ useful info. */
+ msg[0] = IPMI_NETFN_APP_REQUEST << 2;
+ msg[1] = IPMI_GET_DEVICE_ID_CMD;
+ smi_info->handlers->start_transaction(smi_info->si_sm, msg, 2);
+
+ smi_result = smi_info->handlers->event(smi_info->si_sm, 0);
+ for (;;)
+ {
+ if (smi_result == SI_SM_CALL_WITH_DELAY) {
+ set_current_state(TASK_UNINTERRUPTIBLE);
+ schedule_timeout(1);
+ smi_result = smi_info->handlers->event(
+ smi_info->si_sm, 100);
+ }
+ else if (smi_result == SI_SM_CALL_WITHOUT_DELAY)
+ {
+ smi_result = smi_info->handlers->event(
+ smi_info->si_sm, 0);
+ }
+ else
+ break;
+ }
+ if (smi_result == SI_SM_HOSED) {
+ /* We couldn't get the state machine to run, so whatever's at
+ the port is probably not an IPMI SMI interface. */
+ rv = -ENODEV;
+ goto out;
+ }
+
+ /* Otherwise, we got some data. */
+ resp_len = smi_info->handlers->get_result(smi_info->si_sm,
+ resp, IPMI_MAX_MSG_LENGTH);
+ if (resp_len < 6) {
+ /* That's odd, it should be longer. */
+ rv = -EINVAL;
+ goto out;
+ }
+
+ if ((resp[1] != IPMI_GET_DEVICE_ID_CMD) || (resp[2] != 0)) {
+ /* That's odd, it shouldn't be able to fail. */
+ rv = -EINVAL;
+ goto out;
+ }
+
+ /* Record info from the get device id, in case we need it. */
+ smi_info->ipmi_si_dev_rev = resp[4] & 0xf;
+ smi_info->ipmi_si_fw_rev_major = resp[5] & 0x7f;
+ smi_info->ipmi_si_fw_rev_minor = resp[6];
+ smi_info->ipmi_version_major = resp[7] & 0xf;
+ smi_info->ipmi_version_minor = resp[7] >> 4;
+
+ out:
+ kfree(resp);
+ return rv;
+}
+
+static int type_file_read_proc(char *page, char **start, off_t off,
+ int count, int *eof, void *data)
+{
+ char *out = (char *) page;
+ struct smi_info *smi = data;
+
+ switch (smi->si_type) {
+ case SI_KCS:
+ return sprintf(out, "kcs\n");
+ case SI_SMIC:
+ return sprintf(out, "smic\n");
+ case SI_BT:
+ return sprintf(out, "bt\n");
+ default:
+ return 0;
+ }
+}
+
+static int stat_file_read_proc(char *page, char **start, off_t off,
+ int count, int *eof, void *data)
+{
+ char *out = (char *) page;
+ struct smi_info *smi = data;
+
+ out += sprintf(out, "interrupts_enabled: %d\n",
+ smi->irq && !smi->interrupt_disabled);
+ out += sprintf(out, "short_timeouts: %ld\n",
+ smi->short_timeouts);
+ out += sprintf(out, "long_timeouts: %ld\n",
+ smi->long_timeouts);
+ out += sprintf(out, "timeout_restarts: %ld\n",
+ smi->timeout_restarts);
+ out += sprintf(out, "idles: %ld\n",
+ smi->idles);
+ out += sprintf(out, "interrupts: %ld\n",
+ smi->interrupts);
+ out += sprintf(out, "attentions: %ld\n",
+ smi->attentions);
+ out += sprintf(out, "flag_fetches: %ld\n",
+ smi->flag_fetches);
+ out += sprintf(out, "hosed_count: %ld\n",
+ smi->hosed_count);
+ out += sprintf(out, "complete_transactions: %ld\n",
+ smi->complete_transactions);
+ out += sprintf(out, "events: %ld\n",
+ smi->events);
+ out += sprintf(out, "watchdog_pretimeouts: %ld\n",
+ smi->watchdog_pretimeouts);
+ out += sprintf(out, "incoming_messages: %ld\n",
+ smi->incoming_messages);
+
+ return (out - ((char *) page));
+}
+
+/* Returns 0 if initialized, or negative on an error. */
+static int init_one_smi(int intf_num, struct smi_info **smi)
+{
+ int rv;
+ struct smi_info *new_smi;
+
+ rv = try_init_mem(intf_num, &new_smi);
+ if (rv)
+ rv = try_init_port(intf_num, &new_smi);
+#ifdef CONFIG_ACPI_INTERPRETER
+ if ((rv) && (si_trydefaults)) {
+ rv = try_init_acpi(intf_num, &new_smi);
+ }
+#endif
+#ifdef CONFIG_X86
+ if ((rv) && (si_trydefaults)) {
+ rv = try_init_smbios(intf_num, &new_smi);
+ }
+#endif
+ if ((rv) && (si_trydefaults)) {
+ rv = try_init_plug_and_play(intf_num, &new_smi);
+ }
+
+
+ if (rv)
+ return rv;
+
+ /* So we know not to free it unless we have allocated one. */
+ new_smi->intf = NULL;
+ new_smi->si_sm = NULL;
+ new_smi->handlers = 0;
+
+ if (!new_smi->irq_setup) {
+ new_smi->irq = irqs[intf_num];
+ new_smi->irq_setup = std_irq_setup;
+ new_smi->irq_cleanup = std_irq_cleanup;
+ }
+
+ /* Default to KCS if no type is specified. */
+ if (si_type[intf_num] == NULL) {
+ if (si_trydefaults)
+ si_type[intf_num] = "kcs";
+ else {
+ rv = -EINVAL;
+ goto out_err;
+ }
+ }
+
+ /* Set up the state machine to use. */
+ if (strcmp(si_type[intf_num], "kcs") == 0) {
+ new_smi->handlers = &kcs_smi_handlers;
+ new_smi->si_type = SI_KCS;
+ } else if (strcmp(si_type[intf_num], "smic") == 0) {
+ new_smi->handlers = &smic_smi_handlers;
+ new_smi->si_type = SI_SMIC;
+ } else if (strcmp(si_type[intf_num], "bt") == 0) {
+ new_smi->handlers = &bt_smi_handlers;
+ new_smi->si_type = SI_BT;
+ } else {
+ /* No support for anything else yet. */
+ rv = -EIO;
+ goto out_err;
+ }
+
+ /* Allocate the state machine's data and initialize it. */
+ new_smi->si_sm = kmalloc(new_smi->handlers->size(), GFP_KERNEL);
+ if (!new_smi->si_sm) {
+ printk(" Could not allocate state machine memory\n");
+ rv = -ENOMEM;
+ goto out_err;
+ }
+ new_smi->io_size = new_smi->handlers->init_data(new_smi->si_sm,
+ &new_smi->io);
+
+ /* Now that we know the I/O size, we can set up the I/O. */
+ rv = new_smi->io_setup(new_smi);
+ if (rv) {
+ printk(" Could not set up I/O space\n");
+ goto out_err;
+ }
+
+ spin_lock_init(&(new_smi->si_lock));
+ spin_lock_init(&(new_smi->msg_lock));
+ spin_lock_init(&(new_smi->count_lock));
+
+ /* Do low-level detection first. */
+ if (new_smi->handlers->detect(new_smi->si_sm)) {
+ rv = -ENODEV;
+ goto out_err;
+ }
+
+ /* Attempt a get device id command. If it fails, we probably
+ don't have a SMI here. */
+ rv = try_get_dev_id(new_smi);
+ if (rv)
+ goto out_err;
+
+ /* Try to claim any interrupts. */
+ new_smi->irq_setup(new_smi);
+
+ INIT_LIST_HEAD(&(new_smi->xmit_msgs));
+ INIT_LIST_HEAD(&(new_smi->hp_xmit_msgs));
+ new_smi->curr_msg = NULL;
+ atomic_set(&new_smi->req_events, 0);
+ new_smi->run_to_completion = 0;
+
+ rv = ipmi_register_smi(&handlers,
+ new_smi,
+ new_smi->ipmi_version_major,
+ new_smi->ipmi_version_minor,
+ &(new_smi->intf));
+ if (rv) {
+ printk(KERN_ERR
+ "ipmi_si: Unable to register device: error %d\n",
+ rv);
+ goto out_err;
+ }
+
+ rv = ipmi_smi_add_proc_entry(new_smi->intf, "type",
+ type_file_read_proc, NULL,
+ new_smi, THIS_MODULE);
+ if (rv) {
+ printk(KERN_ERR
+ "ipmi_si: Unable to create proc entry: %d\n",
+ rv);
+ goto out_err;
+ }
+
+ rv = ipmi_smi_add_proc_entry(new_smi->intf, "si_stats",
+ stat_file_read_proc, NULL,
+ new_smi, THIS_MODULE);
+ if (rv) {
+ printk(KERN_ERR
+ "ipmi_si: Unable to create proc entry: %d\n",
+ rv);
+ goto out_err;
+ }
+
+ start_clear_flags(new_smi);
+
+ /* IRQ is defined to be set when non-zero. */
+ if (new_smi->irq)
+ new_smi->si_state = SI_CLEARING_FLAGS_THEN_SET_IRQ;
+
+ new_smi->interrupt_disabled = 0;
+ new_smi->timer_stopped = 0;
+ new_smi->stop_operation = 0;
+
+ init_timer(&(new_smi->si_timer));
+ new_smi->si_timer.data = (long) new_smi;
+ new_smi->si_timer.function = smi_timeout;
+ new_smi->last_timeout_jiffies = jiffies;
+ new_smi->si_timer.expires = jiffies + SI_TIMEOUT_JIFFIES;
+ add_timer(&(new_smi->si_timer));
+
+ *smi = new_smi;
+
+ printk(" IPMI %s interface initialized\n", si_type[intf_num]);
+
+ return 0;
+
+ out_err:
+ if (new_smi->intf)
+ ipmi_unregister_smi(new_smi->intf);
+
+ new_smi->irq_cleanup(new_smi);
+ if (new_smi->si_sm) {
+ if (new_smi->handlers)
+ new_smi->handlers->cleanup(new_smi->si_sm);
+ kfree(new_smi->si_sm);
+ }
+ new_smi->io_cleanup(new_smi);
+ return rv;
+}
+
+static __init int init_ipmi_si(void)
+{
+ int rv = 0;
+ int pos = 0;
+ int i;
+ char *str;
+
+ if (initialized)
+ return 0;
+ initialized = 1;
+
+ /* Parse out the si_type string into its components. */
+ str = si_type_str;
+ if (*str != '\0') {
+ for (i=0; (i<SI_MAX_PARMS) && (*str != '\0'); i++) {
+ si_type[i] = str;
+ str = strchr(str, ',');
+ if (str) {
+ *str = '\0';
+ str++;
+ } else {
+ break;
+ }
+ }
+ }
+
+ printk(KERN_INFO "IPMI System Interface driver version "
+ IPMI_SI_VERSION);
+ if (kcs_smi_handlers.version)
+ printk(", KCS version %s", kcs_smi_handlers.version);
+ if (smic_smi_handlers.version)
+ printk(", SMIC version %s", smic_smi_handlers.version);
+ if (bt_smi_handlers.version)
+ printk(", BT version %s", bt_smi_handlers.version);
+ printk("\n");
+
+ rv = init_one_smi(0, &(smi_infos[pos]));
+ if (rv && !ports[0] && si_trydefaults) {
+ /* If we are trying defaults and the initial port is
+ not set, then set it. */
+ si_type[0] = "kcs";
+ ports[0] = DEFAULT_KCS_IO_PORT;
+ rv = init_one_smi(0, &(smi_infos[pos]));
+ if (rv) {
+ /* No KCS - try SMIC */
+ si_type[0] = "smic";
+ ports[0] = DEFAULT_SMIC_IO_PORT;
+ rv = init_one_smi(0, &(smi_infos[pos]));
+ }
+ if (rv) {
+ /* No SMIC - try BT */
+ si_type[0] = "bt";
+ ports[0] = DEFAULT_BT_IO_PORT;
+ rv = init_one_smi(0, &(smi_infos[pos]));
+ }
+ }
+ if (rv == 0)
+ pos++;
+
+ for (i=1; i < SI_MAX_PARMS; i++) {
+ rv = init_one_smi(i, &(smi_infos[pos]));
+ if (rv == 0)
+ pos++;
+ }
+
+ if (smi_infos[0] == NULL) {
+ printk("ipmi_si: Unable to find any System Interface(s)\n");
+ return -ENODEV;
+ }
+
+ return 0;
+}
+module_init(init_ipmi_si);
+
+void __exit cleanup_one_si(struct smi_info *to_clean)
+{
+ int rv;
+ unsigned long flags;
+
+ if (! to_clean)
+ return;
+
+ /* Tell the timer and interrupt handlers that we are shutting
+ down. */
+ spin_lock_irqsave(&(to_clean->si_lock), flags);
+ spin_lock(&(to_clean->msg_lock));
+
+ to_clean->stop_operation = 1;
+
+ to_clean->irq_cleanup(to_clean);
+
+ spin_unlock(&(to_clean->msg_lock));
+ spin_unlock_irqrestore(&(to_clean->si_lock), flags);
+
+ /* Wait until we know that we are out of any interrupt
+ handlers might have been running before we freed the
+ interrupt. */
+ synchronize_kernel();
+
+ /* Wait for the timer to stop. This avoids problems with race
+ conditions removing the timer here. */
+ while (!to_clean->timer_stopped) {
+ set_current_state(TASK_UNINTERRUPTIBLE);
+ schedule_timeout(1);
+ }
+
+ rv = ipmi_unregister_smi(to_clean->intf);
+ if (rv) {
+ printk(KERN_ERR
+ "ipmi_si: Unable to unregister device: errno=%d\n",
+ rv);
+ }
+
+ to_clean->handlers->cleanup(to_clean->si_sm);
+
+ kfree(to_clean->si_sm);
+
+ to_clean->io_cleanup(to_clean);
+}
+
+static __exit void cleanup_ipmi_si(void)
+{
+ int i;
+
+ if (!initialized)
+ return;
+
+ for (i=0; i<SI_MAX_DRIVERS; i++) {
+ cleanup_one_si(smi_infos[i]);
+ }
+}
+module_exit(cleanup_ipmi_si);
+
+MODULE_LICENSE("GPL");
--- diff/drivers/char/ipmi/ipmi_si_sm.h 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/char/ipmi/ipmi_si_sm.h 2004-03-16 11:27:48.095875976 +0000
@@ -0,0 +1,117 @@
+/*
+ * ipmi_si_sm.h
+ *
+ * State machine interface for low-level IPMI system management
+ * interface state machines. This code is the interface between
+ * the ipmi_smi code (that handles the policy of a KCS, SMIC, or
+ * BT interface) and the actual low-level state machine.
+ *
+ * Author: MontaVista Software, Inc.
+ * Corey Minyard <minyard@mvista.com>
+ * source@mvista.com
+ *
+ * Copyright 2002 MontaVista Software Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ *
+ *
+ * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
+ * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
+ * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
+ * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
+ * OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR
+ * TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
+ * USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+/* This is defined by the state machines themselves, it is an opaque
+ data type for them to use. */
+struct si_sm_data;
+
+/* The structure for doing I/O in the state machine. The state
+ machine doesn't have the actual I/O routines, they are done through
+ this interface. */
+struct si_sm_io
+{
+ unsigned char (*inputb)(struct si_sm_io *io, unsigned int offset);
+ void (*outputb)(struct si_sm_io *io,
+ unsigned int offset,
+ unsigned char b);
+
+ /* Generic info used by the actual handling routines, the
+ state machine shouldn't touch these. */
+ void *info;
+ void *addr;
+};
+
+/* Results of SMI events. */
+enum si_sm_result
+{
+ SI_SM_CALL_WITHOUT_DELAY, /* Call the driver again immediately */
+ SI_SM_CALL_WITH_DELAY, /* Delay some before calling again. */
+ SI_SM_TRANSACTION_COMPLETE, /* A transaction is finished. */
+ SI_SM_IDLE, /* The SM is in idle state. */
+ SI_SM_HOSED, /* The hardware violated the state machine. */
+ SI_SM_ATTN /* The hardware is asserting attn and the
+ state machine is idle. */
+};
+
+/* Handlers for the SMI state machine. */
+struct si_sm_handlers
+{
+ /* Put the version number of the state machine here so the
+ upper layer can print it. */
+ char *version;
+
+ /* Initialize the data and return the amount of I/O space to
+ reserve for the space. */
+ unsigned int (*init_data)(struct si_sm_data *smi,
+ struct si_sm_io *io);
+
+ /* Start a new transaction in the state machine. This will
+ return -2 if the state machine is not idle, -1 if the size
+ is invalid (to large or too small), or 0 if the transaction
+ is successfully completed. */
+ int (*start_transaction)(struct si_sm_data *smi,
+ unsigned char *data, unsigned int size);
+
+ /* Return the results after the transaction. This will return
+ -1 if the buffer is too small, zero if no transaction is
+ present, or the actual length of the result data. */
+ int (*get_result)(struct si_sm_data *smi,
+ unsigned char *data, unsigned int length);
+
+ /* Call this periodically (for a polled interface) or upon
+ receiving an interrupt (for a interrupt-driven interface).
+ If interrupt driven, you should probably poll this
+ periodically when not in idle state. This should be called
+ with the time that passed since the last call, if it is
+ significant. Time is in microseconds. */
+ enum si_sm_result (*event)(struct si_sm_data *smi, long time);
+
+ /* Attempt to detect an SMI. Returns 0 on success or nonzero
+ on failure. */
+ int (*detect)(struct si_sm_data *smi);
+
+ /* The interface is shutting down, so clean it up. */
+ void (*cleanup)(struct si_sm_data *smi);
+
+ /* Return the size of the SMI structure in bytes. */
+ int (*size)(void);
+};
+
+/* Current state machines that we can use. */
+extern struct si_sm_handlers kcs_smi_handlers;
+extern struct si_sm_handlers smic_smi_handlers;
+extern struct si_sm_handlers bt_smi_handlers;
+
--- diff/drivers/char/ipmi/ipmi_smb.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/char/ipmi/ipmi_smb.c 2004-03-16 11:27:48.098875520 +0000
@@ -0,0 +1,1270 @@
+/*
+ * ipmi_smb_intf.c
+ *
+ * The interface to the IPMI driver for SMBus access to a SMBus compliant device.
+ *
+ * Author: Intel Corporation
+ * Todd Davis <todd.c.davis@intel.com>
+ *
+ * Copyright 2003 Intel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ *
+ *
+ * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
+ * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
+ * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
+ * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
+ * OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR
+ * TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
+ * USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+/*
+ * This file holds the "policy" for the interface to the SMB state
+ * machine. It does the configuration, handles timers and interrupts,
+ * and drives the real SMB state machine.
+ */
+
+#include <linux/config.h>
+#include <linux/version.h>
+#if defined(MODVERSIONS)
+#include <linux/modversions.h>
+#endif
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <asm/system.h>
+#include <linux/sched.h>
+#include <linux/timer.h>
+#include <linux/errno.h>
+#include <linux/spinlock.h>
+#include <linux/slab.h>
+#include <linux/delay.h>
+#include <linux/list.h>
+#include <linux/i2c.h>
+#include <linux/ipmi_smi.h>
+#include <linux/init.h>
+
+
+#define IPMI_SMB_VERSION "v31"
+
+/* module feature switches */
+#define REGISTER_SMI 1
+#define SMB_KTHREAD 1
+
+#ifdef SMB_KTHREAD
+#include <linux/kernel.h>
+#include <linux/smp_lock.h>
+
+/* a structure to store all information we need
+ for our thread */
+typedef struct kthread_struct
+{
+ /* Linux task structure of thread */
+ struct task_struct *thread;
+
+ /* semaphore needed on start and creation of thread. */
+ struct semaphore startstop_sem;
+
+ /* queue thread is waiting on. Gets initialized by
+ init_kthread, can be used by thread itself.
+ */
+ wait_queue_head_t queue;
+ /* flag to tell thread whether to die or not.
+ When the thread receives a signal, it must check
+ the value of terminate and call exit_kthread and terminate
+ if set.
+ */
+ int terminate;
+} kthread_t; /* the variable that contains the thread data */
+#endif
+
+#define SMB_IPMI_REQUEST 2
+#define SMB_IPMI_RESPONSE 3
+
+/* smb_debug is a bit-field
+ * SMB_DEBUG_MSG - commands and their responses
+ * SMB_DEBUG_STATES - message states
+ * SMB_DEBUG_TIMING - Measure times between events in the driver
+ */
+#define SMB_DEBUG_TIMING 4
+#define SMB_DEBUG_STATE 2
+#define SMB_DEBUG_MSG 1
+#define SMB_NODEBUG 0
+#define SMB_DEFAULT_DEBUG (SMB_NODEBUG)
+
+#ifdef CONFIG_IPMI_SMB
+/* This forces a dependency to the config file for this option. */
+#endif
+
+enum smb_intf_state {
+ SMB_NORMAL,
+ SMB_GETTING_FLAGS,
+ SMB_GETTING_EVENTS,
+ SMB_CLEARING_FLAGS,
+ SMB_CLEARING_FLAGS_THEN_SET_IRQ,
+ SMB_GETTING_MESSAGES,
+ SMB_ENABLE_INTERRUPTS1,
+ SMB_ENABLE_INTERRUPTS2
+ /* FIXME - add watchdog stuff. */
+};
+
+#define SMB_IDLE(smb) ((smb)->smb_state == SMB_NORMAL \
+ && (smb)->curr_msg == NULL \
+ && ! atomic_read(&(smb)->req_events))
+
+#define SMB_MSG_RETRIES 10
+
+struct smb_info
+{
+ int pos;
+ ipmi_smi_t intf;
+ spinlock_t msg_lock;
+ struct list_head xmit_msgs;
+ struct list_head hp_xmit_msgs;
+ struct ipmi_smi_msg *curr_msg;
+ enum smb_intf_state smb_state;
+ unsigned long smb_debug;
+
+ /* Flags from the last GET_MSG_FLAGS command, used when an ATTN
+ is set to hold the flags until we are done handling everything
+ from the flags. */
+#define RECEIVE_MSG_AVAIL 0x01
+#define EVENT_MSG_BUFFER_FULL 0x02
+#define WDT_PRE_TIMEOUT_INT 0x08
+ unsigned char msg_flags;
+
+ /* If set to true, this will request events the next time the
+ state machine is idle. */
+ atomic_t req_events;
+
+#ifdef CONFIG_IPMI_PANIC_EVENT
+ /* If true, run the state machine to completion on every send
+ call. Generally used after a panic or shutdown to make
+ sure stuff goes out. */
+ int run_to_completion;
+#endif
+
+#ifdef SMB_KTHREAD
+ kthread_t smb_thread;
+#endif
+ struct i2c_client client;
+ unsigned char ipmi_smb_dev_rev;
+ unsigned char ipmi_smb_fw_rev_major;
+ unsigned char ipmi_smb_fw_rev_minor;
+ unsigned char ipmi_version_major;
+ unsigned char ipmi_version_minor;
+};
+
+static int initialized = 0;
+static void return_hosed_msg(struct smb_info *smb_info);
+
+static void deliver_recv_msg(struct smb_info *smb_info,
+ struct ipmi_smi_msg *msg)
+{
+ if (msg->rsp_size < 0) {
+ if (smb_info->curr_msg == NULL) {
+ smb_info->curr_msg = msg;
+ return_hosed_msg(smb_info);
+ } else {
+ printk(KERN_ERR
+ "malformed message in deliver_recv_msg:"
+ " rsp_size = %d\n", msg->rsp_size);
+ ipmi_free_smi_msg(msg);
+ }
+ } else {
+ ipmi_smi_msg_received(smb_info->intf, msg);
+ }
+}
+
+static void return_hosed_msg(struct smb_info *smb_info)
+{
+ struct ipmi_smi_msg *msg = smb_info->curr_msg;
+
+ /* Make it a reponse */
+ msg->rsp[0] = msg->data[0] | 4;
+ msg->rsp[1] = msg->data[1];
+ msg->rsp[2] = 0xFF; /* Unknown error. */
+ msg->rsp_size = 3;
+
+ smb_info->curr_msg = NULL;
+ deliver_recv_msg(smb_info, msg);
+}
+
+static s32 smbus_client_read_block_data(struct i2c_client *client, int debug,
+ u8 *values)
+{
+ s32 resp_len;
+ int retries = 0;
+
+ /* FIXME - is there a way to limit the read size? This culd
+ * be an overrun situation otherwise. */
+ while ((resp_len = i2c_smbus_read_block_data (client,
+ SMB_IPMI_RESPONSE,
+ values)) < 0)
+ {
+ if ((retries += 1) >= SMB_MSG_RETRIES) {
+ printk(KERN_ERR
+ "smb_smbus_read_block_data failed: %d\n",
+ resp_len);
+ break;
+#ifdef CONFIG_IPMI_PANIC_EVENT
+ } else if (retries >= SMB_MSG_RETRIES/2 && signal_pending(current)) {
+ i2c_set_spin_delay(1); /* shutting down */
+#endif
+ } else if (debug & (SMB_DEBUG_MSG|SMB_DEBUG_TIMING)) {
+ printk(KERN_WARNING
+ "smb_smbus_read_block_data: retry %d\n",
+ retries);
+ }
+ }
+
+ if (debug & SMB_DEBUG_TIMING) {
+ struct timeval t;
+ do_gettimeofday(&t);
+ printk(KERN_INFO "**Response %02x %02x %02x: %ld.%6.6ld\n",
+ values[0], values[1], values[2], t.tv_sec, t.tv_usec);
+ }
+
+ if (debug & SMB_DEBUG_MSG) {
+ int i;
+ printk(KERN_INFO "ipmi response:");
+ for (i = 0; i < resp_len; i ++) {
+ printk (" %02x", (unsigned char) (values[i]));
+ }
+ printk ("\n");
+ }
+
+ return resp_len;
+}
+
+static s32 smb_smbus_read_block_data(struct smb_info *smb_info, u8 *values)
+{
+ return smbus_client_read_block_data(&smb_info->client,
+ smb_info->smb_debug, values);
+}
+
+static s32 smbus_client_write_block_data(struct i2c_client *client, int debug,
+ u8 length, u8 *values)
+{
+ s32 ret;
+ int retries = 0;
+
+ if (debug & SMB_DEBUG_MSG) {
+ int i ;
+
+ printk(KERN_INFO "ipmi request:");
+ for (i = 0; i < length; i ++) {
+ printk (" %02x", (unsigned char) (values [i]));
+ }
+ printk ("\n");
+ }
+
+ if (debug & SMB_DEBUG_TIMING) {
+ struct timeval t;
+ do_gettimeofday(&t);
+ printk(KERN_INFO "**Request %02x %02x: %ld.%6.6ld\n",
+ values [0],values [1], t.tv_sec, t.tv_usec);
+ }
+
+ while ((ret = i2c_smbus_write_block_data (client,
+ SMB_IPMI_REQUEST,
+ length, values)) < 0)
+ {
+ if ((retries += 1) >= SMB_MSG_RETRIES) {
+ printk(KERN_ERR
+ "smb_smbus_write_block_data failed: %d\n", ret);
+ break;
+#ifdef CONFIG_IPMI_PANIC_EVENT
+ } else if (retries >= SMB_MSG_RETRIES/2 && signal_pending(current)) {
+ i2c_set_spin_delay(1); /* shutting down */
+#endif
+ } else if (debug & (SMB_DEBUG_MSG|SMB_DEBUG_TIMING)) {
+ printk(KERN_WARNING
+ "smb_smbus_write_block_data: retry %d\n",
+ retries);
+ }
+ }
+ return ret;
+}
+
+static s32 smb_smbus_write_block_data(struct smb_info *smb_info,
+ u8 length, u8 *values)
+{
+ return smbus_client_write_block_data(&smb_info->client,
+ smb_info->smb_debug,
+ length,
+ values);
+}
+
+static int send_next_msg(struct smb_info *smb_info)
+{
+ s32 rv;
+ struct list_head *entry = NULL;
+
+ spin_lock(&(smb_info->msg_lock));
+
+ /* Pick the high priority queue first. */
+ if (! list_empty(&(smb_info->hp_xmit_msgs))) {
+ entry = smb_info->hp_xmit_msgs.next;
+ } else if (! list_empty(&(smb_info->xmit_msgs))) {
+ entry = smb_info->xmit_msgs.next;
+ }
+
+ if (!entry) {
+ spin_unlock(&(smb_info->msg_lock));
+ smb_info->curr_msg = NULL;
+ return 0;
+ }
+ list_del(entry);
+ smb_info->curr_msg = list_entry(entry,
+ struct ipmi_smi_msg,
+ link);
+ spin_unlock(&(smb_info->msg_lock));
+ rv = smb_smbus_write_block_data(
+ smb_info,
+ smb_info->curr_msg->data_size,
+ smb_info->curr_msg->data);
+ if (rv) {
+ return_hosed_msg(smb_info);
+ }
+ return 1;
+}
+
+static void start_enable_irq(struct smb_info *smb_info)
+{
+ unsigned char msg[2];
+
+ /* If we are enabling interrupts, we have to tell the
+ IPMI device to use them. */
+ msg[0] = (IPMI_NETFN_APP_REQUEST << 2);
+ msg[1] = IPMI_GET_BMC_GLOBAL_ENABLES_CMD;
+
+ if (smb_smbus_write_block_data(smb_info, 2, msg) == 0) {
+ smb_info->smb_state = SMB_ENABLE_INTERRUPTS1;
+ } else {
+ smb_info->smb_state = SMB_NORMAL;
+ }
+}
+
+static void start_clear_flags(struct smb_info *smb_info)
+{
+ unsigned char msg[3];
+
+ /* Make sure the watchdog pre-timeout flag is not set at startup. */
+ msg[0] = (IPMI_NETFN_APP_REQUEST << 2);
+ msg[1] = IPMI_CLEAR_MSG_FLAGS_CMD;
+ msg[2] = WDT_PRE_TIMEOUT_INT;
+
+ if (smb_smbus_write_block_data(smb_info, 3, msg) == 0) {
+ smb_info->smb_state = SMB_CLEARING_FLAGS;
+ } else {
+ smb_info->smb_state = SMB_NORMAL;
+ }
+}
+
+static void handle_flags(struct smb_info *smb_info)
+{
+ if (smb_info->msg_flags & WDT_PRE_TIMEOUT_INT) {
+ /* Watchdog pre-timeout */
+ start_clear_flags(smb_info);
+ smb_info->msg_flags &= ~WDT_PRE_TIMEOUT_INT;
+ ipmi_smi_watchdog_pretimeout(smb_info->intf);
+ } else if (smb_info->msg_flags & RECEIVE_MSG_AVAIL) {
+ /* Messages available. */
+ smb_info->curr_msg = ipmi_alloc_smi_msg();
+ if (!smb_info->curr_msg) {
+ smb_info->smb_state = SMB_NORMAL;
+ return;
+ }
+
+ smb_info->curr_msg->data[0] = (IPMI_NETFN_APP_REQUEST << 2);
+ smb_info->curr_msg->data[1] = IPMI_GET_MSG_CMD;
+ smb_info->curr_msg->data_size = 2;
+
+ if (smb_smbus_write_block_data(
+ smb_info,
+ smb_info->curr_msg->data_size,
+ smb_info->curr_msg->data) == 0) {
+ smb_info->smb_state = SMB_GETTING_MESSAGES;
+ } else {
+ ipmi_free_smi_msg(smb_info->curr_msg);
+ smb_info->curr_msg = NULL;
+ smb_info->smb_state = SMB_NORMAL;
+ return;
+ }
+ } else if (smb_info->msg_flags & EVENT_MSG_BUFFER_FULL) {
+ /* Events available. */
+ smb_info->curr_msg = ipmi_alloc_smi_msg();
+ if (!smb_info->curr_msg) {
+ smb_info->smb_state = SMB_NORMAL;
+ return;
+ }
+
+ smb_info->curr_msg->data[0] = (IPMI_NETFN_APP_REQUEST << 2);
+ smb_info->curr_msg->data[1] = IPMI_READ_EVENT_MSG_BUFFER_CMD;
+ smb_info->curr_msg->data_size = 2;
+
+ if (smb_smbus_write_block_data(
+ smb_info,
+ smb_info->curr_msg->data_size,
+ smb_info->curr_msg->data) == 0) {
+ smb_info->smb_state = SMB_GETTING_EVENTS;
+ } else {
+ ipmi_free_smi_msg(smb_info->curr_msg);
+ smb_info->curr_msg = NULL;
+ smb_info->smb_state = SMB_NORMAL;
+ return;
+ }
+ } else {
+ smb_info->smb_state = SMB_NORMAL;
+ }
+}
+
+static void handle_transaction_done(struct smb_info *smb_info)
+{
+ struct ipmi_smi_msg *msg;
+ unsigned int len;
+
+ if (smb_info->smb_debug & SMB_DEBUG_STATE) {
+ printk(KERN_INFO "DONE 1: state = %d.\n", smb_info->smb_state);
+ }
+ switch (smb_info->smb_state) {
+ case SMB_NORMAL:
+ if (!smb_info->curr_msg)
+ break;
+
+ smb_info->curr_msg->rsp_size
+ = smb_smbus_read_block_data(
+ smb_info,
+ smb_info->curr_msg->rsp);
+
+ msg = smb_info->curr_msg;
+ smb_info->curr_msg = NULL;
+ deliver_recv_msg(smb_info, msg);
+ break;
+
+ case SMB_GETTING_FLAGS:
+ {
+ unsigned char msg[4];
+
+ /* We got the flags from the SMB, now handle them. */
+ len = smb_smbus_read_block_data(smb_info, msg);
+ if (len < 0) {
+ /* Error fetching flags, just give up for now. */
+ smb_info->smb_state = SMB_NORMAL;
+ } else if (msg[2] != 0) {
+ /* Error fetching flags, just give up for now. */
+ smb_info->smb_state = SMB_NORMAL;
+ } else if (len < 3) {
+ /* Hmm, no flags. That's technically illegal, but
+ don't use uninitialized data. */
+ smb_info->smb_state = SMB_NORMAL;
+ } else {
+ smb_info->msg_flags = msg[3];
+ handle_flags(smb_info);
+ }
+ break;
+ }
+
+ case SMB_CLEARING_FLAGS:
+ case SMB_CLEARING_FLAGS_THEN_SET_IRQ:
+ {
+ unsigned char msg[3];
+
+ /* We cleared the flags. */
+ len = smb_smbus_read_block_data(smb_info, msg);
+ if (len < 0 || msg[2] != 0) {
+ /* Error clearing flags */
+ printk(KERN_WARNING
+ "ipmi_smb: Error clearing flags: %2.2x\n",
+ msg[2]);
+ }
+ if (smb_info->smb_state == SMB_CLEARING_FLAGS_THEN_SET_IRQ)
+ start_enable_irq(smb_info);
+ else
+ smb_info->smb_state = SMB_NORMAL;
+ break;
+ }
+
+ case SMB_GETTING_EVENTS:
+ {
+ smb_info->curr_msg->rsp_size
+ = smb_smbus_read_block_data(smb_info,
+ smb_info->curr_msg->rsp);
+
+ msg = smb_info->curr_msg;
+ smb_info->curr_msg = NULL;
+ if (msg->rsp_size < 0 || msg->rsp[2] != 0) {
+ /* Error getting event, probably done. */
+ msg->done(msg);
+
+ /* Take off the event flag. */
+ smb_info->msg_flags &= ~EVENT_MSG_BUFFER_FULL;
+ } else {
+ deliver_recv_msg(smb_info, msg);
+ }
+ handle_flags(smb_info);
+ break;
+ }
+
+ case SMB_GETTING_MESSAGES:
+ {
+ smb_info->curr_msg->rsp_size
+ = smb_smbus_read_block_data(smb_info,
+ smb_info->curr_msg->rsp);
+
+ msg = smb_info->curr_msg;
+ smb_info->curr_msg = NULL;
+ if (msg->rsp_size < 0 || msg->rsp[2] != 0) {
+ /* Error getting event, probably done. */
+ msg->done(msg);
+
+ /* Take off the msg flag. */
+ smb_info->msg_flags &= ~RECEIVE_MSG_AVAIL;
+ } else {
+ deliver_recv_msg(smb_info, msg);
+ }
+ handle_flags(smb_info);
+ break;
+ }
+
+ case SMB_ENABLE_INTERRUPTS1:
+ {
+ unsigned char msg[4];
+
+ /* We got the flags from the SMB, now handle them. */
+ len = smb_smbus_read_block_data(smb_info, msg);
+ if (len < 0 || msg[2] != 0) {
+ printk(KERN_WARNING
+ "ipmi_smb: Could not enable interrupts"
+ ", failed get, using polled mode.\n");
+ smb_info->smb_state = SMB_NORMAL;
+ } else {
+ msg[0] = (IPMI_NETFN_APP_REQUEST << 2);
+ msg[1] = IPMI_SET_BMC_GLOBAL_ENABLES_CMD;
+ msg[2] = msg[3] | 1; /* enable msg queue int */
+ if (smb_smbus_write_block_data(
+ smb_info,3, msg) == 0) {
+ smb_info->smb_state = SMB_ENABLE_INTERRUPTS2;
+ } else
+ smb_info->smb_state = SMB_NORMAL;
+ }
+ break;
+ }
+
+ case SMB_ENABLE_INTERRUPTS2:
+ {
+ unsigned char msg[4];
+
+ /* We got the flags from the SMB, now handle them. */
+ len = smb_smbus_read_block_data(smb_info, msg);
+ if (len < 0 || msg[2] != 0) {
+ printk(KERN_WARNING
+ "ipmi_smb: Could not enable interrupts"
+ ", failed set, using polled mode.\n");
+ }
+ smb_info->smb_state = SMB_NORMAL;
+ break;
+ }
+ }
+ if (smb_info->smb_debug & SMB_DEBUG_STATE) {
+ printk(KERN_INFO "DONE 2: state = %d.\n", smb_info->smb_state);
+ }
+}
+
+/*
+ * smb_event_handler must have a user context for calls to lm_sensors'
+ * SMBus interface
+ */
+static void smb_event_handler(struct smb_info *smb_info)
+{
+ s32 rv;
+ if (smb_info->smb_debug & SMB_DEBUG_STATE) {
+ printk(KERN_INFO "smb_event_handler: state = %d.\n",
+ smb_info->smb_state);
+ }
+ if (smb_info->smb_state == SMB_NORMAL) {
+ if (smb_info->curr_msg != NULL)
+ {
+ handle_transaction_done(smb_info);
+ } else {
+ /* If we are currently idle, try to start the
+ * next message. */
+ send_next_msg(smb_info);
+ }
+ } else {
+ handle_transaction_done(smb_info);
+ }
+
+ if (smb_info->smb_state == SMB_NORMAL &&
+ smb_info->curr_msg == NULL &&
+ atomic_read(&smb_info->req_events))
+ {
+ /* We are idle and the upper layer requested that I fetch
+ events, so do so. */
+ unsigned char msg[2];
+
+ atomic_set(&smb_info->req_events, 0);
+ msg[0] = (IPMI_NETFN_APP_REQUEST << 2);
+ msg[1] = IPMI_GET_MSG_FLAGS_CMD;
+
+ rv = smb_smbus_write_block_data(smb_info, 2, msg);
+ if(rv == 0) {
+ smb_info->smb_state = SMB_GETTING_FLAGS;
+ }
+ }
+}
+
+#ifdef REGISTER_SMI
+static void sender(void *send_info,
+ struct ipmi_smi_msg *msg,
+ int priority)
+{
+ struct smb_info *smb_info = (struct smb_info *) send_info;
+
+#ifdef CONFIG_IPMI_PANIC_EVENT
+ if (smb_info->run_to_completion) {
+ /* If we are running to completion, then throw it in
+ the list and run transactions until everything is
+ clear. Priority doesn't matter here. */
+ list_add_tail(&(msg->link), &(smb_info->xmit_msgs));
+
+ smb_event_handler(smb_info);
+ while (! SMB_IDLE(smb_info)) {
+ smb_event_handler(smb_info);
+ }
+ return;
+ }
+#endif
+ if (smb_info->smb_debug & SMB_DEBUG_TIMING) {
+ struct timeval t;
+ do_gettimeofday(&t);
+ printk(KERN_INFO
+ "**Enqueue %02x %02x: %ld.%6.6ld\n",
+ msg->data[0], msg->data[1], t.tv_sec, t.tv_usec);
+ }
+
+ spin_lock(&(smb_info->msg_lock));
+ if (priority > 0) {
+ list_add_tail(&(msg->link), &(smb_info->hp_xmit_msgs));
+ } else {
+ list_add_tail(&(msg->link), &(smb_info->xmit_msgs));
+ }
+ spin_unlock(&(smb_info->msg_lock));
+
+#ifdef SMB_KTHREAD
+ wake_up_interruptible(&smb_info->smb_thread.queue);
+#endif
+}
+
+static void request_events(void *send_info)
+{
+ struct smb_info *smb_info = (struct smb_info *) send_info;
+
+ atomic_set(&smb_info->req_events, 1);
+#ifdef SMB_KTHREAD
+ wake_up_interruptible(&smb_info->smb_thread.queue);
+#endif
+}
+
+static void set_run_to_completion(void *send_info, int i_run_to_completion)
+{
+#ifdef CONFIG_IPMI_PANIC_EVENT
+ struct smb_info *smb_info = (struct smb_info *) send_info;
+
+ smb_info->run_to_completion = i_run_to_completion;
+ /* Note that if this does not compile, there are some I2C
+ changes that you need to handle this properly. */
+ i2c_set_spin_delay(i_run_to_completion);
+ if (i_run_to_completion) {
+ smb_event_handler(smb_info);
+ while (! SMB_IDLE(smb_info)) {
+ smb_event_handler(smb_info);
+ }
+ }
+#endif
+}
+
+static struct ipmi_smi_handlers handlers =
+{
+ .owner = THIS_MODULE,
+ .sender = sender,
+ .request_events = request_events,
+ .set_run_to_completion = set_run_to_completion
+};
+#endif /* REGISTER_SMI */
+
+#ifdef SMB_KTHREAD
+/* initialize new created thread. Called by the new thread. */
+static void init_kthread(kthread_t *kthread, char *name)
+{
+ /* lock the kernel. A new kernel thread starts without
+ the big kernel lock, regardless of the lock state
+ of the creator (the lock level is *not* inheritated)
+ */
+ lock_kernel();
+
+ /* fill in thread structure */
+ kthread->thread = current;
+
+ /* set signal mask to what we want to respond */
+ siginitsetinv(¤t->blocked,
+ sigmask(SIGKILL)|sigmask(SIGINT)|sigmask(SIGTERM));
+
+ /* initialise wait queue */
+ init_waitqueue_head(&kthread->queue);
+
+ /* initialise termination flag */
+ kthread->terminate = 0;
+
+ /* set name of this process (max 15 chars + 0 !) */
+ sprintf(current->comm, name);
+
+ /* let others run */
+ unlock_kernel();
+
+ /* tell the creator that we are ready and let him continue */
+ up(&kthread->startstop_sem);
+
+}
+
+/* cleanup of thread. Called by the exiting thread. */
+static void exit_kthread(kthread_t *kthread)
+{
+ /* we are terminating */
+
+ /* lock the kernel, the exit will unlock it */
+ lock_kernel();
+ kthread->thread = NULL;
+ mb();
+
+ /* notify the stop_kthread() routine that we are terminating. */
+ up(&kthread->startstop_sem);
+ /* the kernel_thread that called clone() does a do_exit here. */
+
+ /* there is no race here between execution of the "killer" and
+ real termination of the thread (race window between up and
+ do_exit), since both the thread and the "killer" function
+ are running with the kernel lock held. The kernel lock
+ will be freed after the thread exited, so the code is
+ really not executed anymore as soon as the unload functions
+ gets the kernel lock back. The init process may not have
+ made the cleanup of the process here, but the cleanup can
+ be done safely with the module unloaded.
+ */
+
+}
+
+/* this is the thread function for making SMBus function calls into
+ * the lm_sensors' i2c-core module */
+static int smb_thread(void *data)
+{
+ struct smb_info *smb_info = data;
+ kthread_t *kthread = &smb_info->smb_thread;
+ /* setup the thread environment */
+ init_kthread(kthread, "IPMI SMBus thread");
+
+ /* an endless loop in which we are doing our work */
+ for(;;)
+ {
+ smb_event_handler(smb_info);
+ while (! SMB_IDLE(smb_info)) {
+ smb_event_handler(smb_info);
+ }
+
+ interruptible_sleep_on(&kthread->queue);
+ if (! send_next_msg(smb_info)) {
+ interruptible_sleep_on(&kthread->queue);
+ }
+
+ /* We need to do a memory barrier here to be sure that
+ the flags are visible on all CPUs.
+ */
+ mb();
+
+ /* here we are back from sleep, either due to the timeout
+ (one second), or because we caught a signal.
+ */
+ if (kthread->terminate)
+ {
+ /* we received a request to terminate ourself */
+ break;
+ }
+ }
+
+ /* cleanup the thread, leave */
+ exit_kthread(kthread);
+
+ /* returning from the thread here calls the exit functions */
+ return 0;
+}
+
+/* create a new kernel thread. Called by the creator. */
+static void start_kthread(struct smb_info *smb_info)
+{
+ kthread_t *kthread = &smb_info->smb_thread;
+ /* initialize the semaphore:
+ we start with the semaphore locked. The new kernel
+ thread will setup its stuff and unlock it. This
+ control flow (the one that creates the thread) blocks
+ in the down operation below until the thread has reached
+ the up() operation.
+ */
+ init_MUTEX_LOCKED(&kthread->startstop_sem);
+
+ kernel_thread(smb_thread, smb_info, 0);
+
+ /* wait till it has reached the setup_thread routine */
+ down(&kthread->startstop_sem);
+
+}
+
+/* stop a kernel thread. Called by the removing instance */
+static void stop_kthread(kthread_t *kthread)
+{
+ if (kthread->thread == NULL)
+ {
+ printk(KERN_WARNING
+ "stop_kthread: killing non existing thread!\n");
+ return;
+ }
+
+ /* this function needs to be protected with the big
+ kernel lock (lock_kernel()). The lock must be
+ grabbed before changing the terminate
+ flag and released after the down() call. */
+ lock_kernel();
+
+ /* initialize the semaphore. We lock it here, the
+ leave_thread call of the thread to be terminated
+ will unlock it. As soon as we see the semaphore
+ unlocked, we know that the thread has exited.
+ */
+ init_MUTEX_LOCKED(&kthread->startstop_sem);
+
+ /* We need to do a memory barrier here to be sure that
+ the flags are visible on all CPUs.
+ */
+ mb();
+
+ /* set flag to request thread termination */
+ kthread->terminate = 1;
+ wake_up_interruptible(&kthread->queue);
+
+ /* We need to do a memory barrier here to be sure that
+ the flags are visible on all CPUs.
+ */
+ mb();
+ kill_proc(kthread->thread->pid, SIGKILL, 1);
+
+ /* block till thread terminated */
+ down(&kthread->startstop_sem);
+
+ /* release the big kernel lock */
+ unlock_kernel();
+
+ /* now we are sure the thread is in zombie state. We
+ notify keventd to clean the process up.
+ */
+ kill_proc(2, SIGCHLD, 1);
+
+}
+#endif
+
+static int ipmi_smb_detect_hardware(struct i2c_client *client, int debug,
+ struct smb_info **smb_info)
+{
+ unsigned char msg[2];
+ unsigned char *resp;
+ unsigned long resp_len;
+ s32 ret;
+ struct smb_info *info;
+ int rv = 0;
+
+ resp = kmalloc(IPMI_MAX_MSG_LENGTH, GFP_KERNEL);
+ if (!resp)
+ return -ENOMEM;
+
+ /* Do a Get Device ID command, since it comes back with some
+ useful info. */
+ msg[0] = IPMI_NETFN_APP_REQUEST << 2;
+ msg[1] = IPMI_GET_DEVICE_ID_CMD;
+
+ ret = smbus_client_write_block_data(client, debug, 2, msg);
+ if (ret) {
+ rv = -ENODEV;
+ goto out;
+ }
+
+ /* Otherwise, we got some data. */
+ resp_len = smbus_client_read_block_data(client, debug, resp);
+ if (resp_len < 6) {
+ /* That's odd, it should be longer. */
+ rv = -EINVAL;
+ goto out;
+ }
+
+ if ((resp[1] != IPMI_GET_DEVICE_ID_CMD) || (resp[2] != 0)) {
+ /* That's odd, it shouldn't be able to fail. */
+ rv = -EINVAL;
+ goto out;
+ }
+
+ info = kmalloc(sizeof(*info), GFP_KERNEL);
+ if (!info) {
+ rv = -ENOMEM;
+ goto out;
+ }
+ memset(info, 0, sizeof(*info));
+
+ info->ipmi_smb_dev_rev = resp[4] & 0xf;
+ info->ipmi_smb_fw_rev_major = resp[5] & 0x7f;
+ info->ipmi_smb_fw_rev_minor = resp[6];
+ info->ipmi_version_major = resp[7] & 0xf;
+ info->ipmi_version_minor = resp[7] >> 4;
+ info->client = *client;
+ i2c_set_clientdata(&info->client, info);
+ info->smb_debug = debug;
+
+ *smb_info = info;
+
+ out:
+ kfree(resp);
+ return rv;
+}
+
+#define MAX_SMB_BMCS 4
+/* no expressions allowed in __MODULE_STRING */
+#define MAX_SMB_ADDR_PAIRS 8
+
+/* An array of SMB interfaces. */
+static struct smb_info *smb_infos[MAX_SMB_BMCS];
+
+/*
+ * An array of pairs of numbers to specify where specific BMCs exist.
+ * Each pair specifies the adapter number and address on the adapter.
+ * A BMC will be forced at that position.
+ *
+ * Always provide the i2c address if it is known.
+ */
+/* force list has 3 entries - 0:bus/adapter no 1: i2c addr 2: unknown/unused */
+#define FORCE_LIST_ENTRIES 3
+static unsigned short __initdata addr[MAX_SMB_BMCS*2];
+static int num_addrs = 0;
+static unsigned short smb_force_list[MAX_SMB_BMCS*FORCE_LIST_ENTRIES + FORCE_LIST_ENTRIES];
+
+module_param_array(addr, ushort, num_addrs, 0);
+
+/*
+ * If this is sent, don't probe for adapters anywhere but where the
+ * addr array gives.
+ */
+static int smb_defaultprobe = 1;
+module_param_named(defaultprobe, smb_defaultprobe, int, 0);
+
+/*
+ * Turn debugging on for specific BMCs. This array is indexed by
+ * BMC number.
+ *
+ * Debug bit flags: IPMI messages: 1, driver state: 2, msg timing: 4
+ */
+static int dbg[MAX_SMB_BMCS];
+static int num_dbg = 0;
+module_param_array(dbg, int, num_dbg, 0);
+
+/*
+ * Debug the probing of adapters.
+ */
+static int smb_dbg_probe = 0;
+module_param_named(dbg_probe, smb_dbg_probe, int, 0);
+
+#define SMB_I2C_START_ADDR 0x20
+#define SMB_I2C_END_ADDR 0x4f
+static unsigned short normal_i2c[] = { I2C_CLIENT_END, I2C_CLIENT_END };
+static unsigned short normal_i2c_range[] = { SMB_I2C_START_ADDR,
+ SMB_I2C_END_ADDR,
+ I2C_CLIENT_END };
+/*
+static unsigned int normal_isa[] = { SENSORS_ISA_END };
+static unsigned int normal_isa_range[] = { SENSORS_ISA_END };
+*/
+static unsigned short reserved[] =
+{
+/* As defined by SMBus Spec. Appendix C */
+ 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x28,
+ 0x37,
+/* As defined by SMBus Spec. Sect. 5.2 */
+ 0x01, 0x02, 0x03, 0x04, 0x05,
+ 0x06, 0x07, 0x78, 0x79, 0x7a, 0x7b,
+ 0x7c, 0x7d, 0x7e, 0x7f,
+/* Common PC addresses (bad idea) */
+ 0x2d, 0x48, 0x49, /* sensors */
+ 0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, /* eeproms */
+ 0x69, /* clock chips */
+
+ I2C_CLIENT_END
+};
+
+static unsigned short smb_empty_list[] = { I2C_CLIENT_END, I2C_CLIENT_END };
+
+static struct i2c_client_address_data smb_address_data = {
+ .normal_i2c = normal_i2c,
+ .normal_i2c_range = normal_i2c_range,
+ .probe = smb_empty_list,
+ .probe_range = smb_empty_list,
+ .ignore = reserved,
+ .ignore_range = smb_empty_list,
+ .force = smb_force_list,
+};
+
+static unsigned int pos_reserved_as(int pos)
+{
+ if (addr[pos*2+1] != 0)
+ return (addr[pos*2] << 16) | addr[pos*2+1];
+
+ return 0;
+}
+
+static int smb_found_addr_proc(struct i2c_adapter *adapter, int addr, int kind)
+{
+ int id = i2c_adapter_id(adapter);
+ int debug = dbg[id];
+ int rv;
+ int i;
+ int next_pos;
+ struct i2c_client client;
+ struct smb_info *smb_info;
+
+ if( id >= MAX_SMB_BMCS )
+ return 0;
+ memset(&client, 0, sizeof(&client));
+ strcpy(client.name, "IPMI");
+ client.addr = addr;
+ client.adapter = adapter;
+
+ rv = ipmi_smb_detect_hardware(&client, debug, &smb_info);
+ if (rv) {
+ if (smb_dbg_probe) {
+ printk(KERN_INFO
+ "smb_found_addr_proc:No IPMI client 0x%x: %d\n",
+ addr, rv);
+ }
+ return 0;
+ }
+
+ if (smb_dbg_probe) {
+ printk(KERN_INFO
+ "smb_found_addr_proc: i2c_probe found device at"
+ " i2c address %x\n", addr);
+ }
+
+ spin_lock_init(&(smb_info->msg_lock));
+ INIT_LIST_HEAD(&(smb_info->xmit_msgs));
+ INIT_LIST_HEAD(&(smb_info->hp_xmit_msgs));
+ smb_info->curr_msg = NULL;
+ atomic_set(&smb_info->req_events, 0);
+#ifdef CONFIG_IPMI_PANIC_EVENT
+ smb_info->run_to_completion = 0;
+#endif
+ smb_info->smb_state = SMB_NORMAL;
+
+ next_pos = -1;
+ for (i=0; i < MAX_SMB_BMCS; i++) {
+ unsigned int res = pos_reserved_as(i);
+
+ if (res == ((id << 16) | addr)) {
+ /* We have a reserved position, use it. */
+ next_pos = i;
+ break;
+ }
+
+ /* Claim the first unused position */
+ if (!res && (next_pos == -1) && (smb_infos[next_pos] == NULL))
+ next_pos = i;
+ }
+ if (next_pos == -1) {
+ rv = -EBUSY;
+ goto out_err;
+ }
+
+ rv = i2c_attach_client(&smb_info->client);
+ if (rv) {
+ printk(KERN_ERR
+ "smb_found_one_addr_proc:"
+ " Unable to attach i2c client: error %d\n",
+ rv);
+ goto out_err;
+ }
+
+ smb_info->pos = next_pos;
+ smb_infos[next_pos] = smb_info;
+
+#ifdef SMB_KTHREAD
+ start_kthread(smb_info);
+#endif
+
+#ifdef REGISTER_SMI
+ rv = ipmi_register_smi(&handlers,
+ smb_info,
+ smb_info->ipmi_version_major,
+ smb_info->ipmi_version_minor,
+ &(smb_info->intf));
+ if (rv) {
+ i2c_detach_client(&smb_info->client);
+ smb_infos[next_pos] = NULL;
+ printk(KERN_ERR
+ "ipmi_smb: Unable to register device: error %d\n",
+ rv);
+ goto out_err;
+ }
+#endif
+
+ start_clear_flags(smb_info);
+ smb_event_handler(smb_info);
+
+ return addr;
+
+ out_err:
+ kfree(smb_info);
+ return 0;
+}
+
+static int attach_adapter(struct i2c_adapter *adapter)
+{
+ int id = i2c_adapter_id(adapter);
+
+ if (smb_dbg_probe) {
+ printk(KERN_INFO "init_one_smb: Checking SMBus adapter %d:"
+ " %s\n", id, adapter->name);
+ }
+ if ((i2c_get_functionality(adapter) & (I2C_FUNC_SMBUS_BLOCK_DATA))
+ == (I2C_FUNC_SMBUS_BLOCK_DATA))
+ {
+ if (smb_dbg_probe) {
+ printk(KERN_INFO "init_one_smb: found SMBus adapter:"
+ " %s\n", adapter->name);
+ }
+ i2c_probe(adapter, &smb_address_data, smb_found_addr_proc);
+ }
+
+ return 0;
+}
+
+void cleanup_one_smb(struct smb_info *to_clean)
+{
+ int rv;
+
+ if (! to_clean)
+ return;
+
+#ifdef SMB_KTHREAD
+ stop_kthread(&to_clean->smb_thread);
+#endif
+
+#ifdef REGISTER_SMI
+ rv = ipmi_unregister_smi(to_clean->intf);
+ if (rv) {
+ printk(KERN_ERR
+ "ipmi_smb: Unable to unregister device: errno=%d\n",
+ rv);
+ }
+#endif
+
+ rv = i2c_detach_client(&to_clean->client);
+ if (rv) {
+ printk(KERN_ERR
+ "ipmi_smb: Unable to detach SMBUS client: errno=%d\n",
+ rv);
+ }
+
+ smb_infos[to_clean->pos] = NULL;
+ kfree(to_clean);
+}
+
+static int detach_client(struct i2c_client *client)
+{
+ struct smb_info *smb_info = i2c_get_clientdata(client);
+
+ cleanup_one_smb(smb_info);
+
+ return(0);
+}
+
+static struct i2c_driver smb_i2c_driver =
+{
+ .name = "IPMI",
+ .flags = I2C_DF_NOTIFY,
+ .attach_adapter = attach_adapter,
+ .detach_client = detach_client,
+ .command = NULL,
+};
+
+static __init int init_ipmi_smb(void)
+{
+ int i;
+ int rv;
+
+ if (initialized)
+ return 0;
+
+ printk(KERN_INFO "IPMI SMB Interface driver version "
+ IPMI_SMB_VERSION "\n");
+
+ /* build force list from addr list */
+ for (i=0; i<MAX_SMB_BMCS; i++) {
+ if (addr[i*2+1] == 0)
+ break;
+ smb_force_list[i*FORCE_LIST_ENTRIES] = addr[i*2];
+ smb_force_list[i*FORCE_LIST_ENTRIES+1] = addr[i*2+1];
+ }
+ smb_force_list[i*FORCE_LIST_ENTRIES] = I2C_CLIENT_END;
+ smb_force_list[i*FORCE_LIST_ENTRIES+1] = I2C_CLIENT_END;
+
+ /* If the default probing is turned off, then disable the
+ * range scanning. */
+ if (!smb_defaultprobe)
+ normal_i2c_range[0] = I2C_CLIENT_END;
+
+ rv = i2c_add_driver(&smb_i2c_driver);
+ if (!rv)
+ initialized = 1;
+
+ return rv;
+}
+module_init(init_ipmi_smb);
+
+static __exit void cleanup_ipmi_smb(void)
+{
+ int i;
+ int rv;
+
+ if (!initialized)
+ return;
+
+ for (i=0; i<MAX_SMB_BMCS; i++) {
+ cleanup_one_smb(smb_infos[i]);
+ }
+
+ initialized = 0;
+
+ rv = i2c_del_driver(&smb_i2c_driver);
+ if (!rv)
+ initialized = 0;
+}
+module_exit(cleanup_ipmi_smb);
+
+MODULE_AUTHOR("Todd C Davis <todd.c.davis@intel.com>");
+MODULE_DESCRIPTION("IPMI system interface driver for management controllers on a SMBus");
+MODULE_LICENSE("GPL");
--- diff/drivers/char/ipmi/ipmi_smic_sm.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/char/ipmi/ipmi_smic_sm.c 2004-03-16 11:27:48.100875216 +0000
@@ -0,0 +1,601 @@
+/*
+ * ipmi_smic_sm.c
+ *
+ * The state-machine driver for an IPMI SMIC driver
+ *
+ * It started as a copy of Corey Minyard's driver for the KSC interface
+ * and the kernel patch "mmcdev-patch-245" by HP
+ *
+ * modified by: Hannes Schulz <schulz@schwaar.com>
+ * ipmi@schwaar.com
+ *
+ *
+ * Corey Minyard's driver for the KSC interface has the following
+ * copyright notice:
+ * Copyright 2002 MontaVista Software Inc.
+ *
+ * the kernel patch "mmcdev-patch-245" by HP has the following
+ * copyright notice:
+ * (c) Copyright 2001 Grant Grundler (c) Copyright
+ * 2001 Hewlett-Packard Company
+ *
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ *
+ *
+ * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
+ * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
+ * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
+ * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
+ * OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR
+ * TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
+ * USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 675 Mass Ave, Cambridge, MA 02139, USA. */
+
+#include <linux/types.h>
+
+#include <linux/kernel.h> /* For printk. */
+#include <linux/string.h>
+#include <linux/ipmi_msgdefs.h> /* for completion codes */
+#include "ipmi_si_sm.h"
+
+#define IPMI_SMIC_VERSION "v31"
+
+/* smic_debug is a bit-field
+ * SMIC_DEBUG_ENABLE - turned on for now
+ * SMIC_DEBUG_MSG - commands and their responses
+ * SMIC_DEBUG_STATES - state machine
+*/
+#define SMIC_DEBUG_STATES 4
+#define SMIC_DEBUG_MSG 2
+#define SMIC_DEBUG_ENABLE 1
+
+static int smic_debug = 1;
+
+enum smic_states {
+ SMIC_IDLE,
+ SMIC_START_OP,
+ SMIC_OP_OK,
+ SMIC_WRITE_START,
+ SMIC_WRITE_NEXT,
+ SMIC_WRITE_END,
+ SMIC_WRITE2READ,
+ SMIC_READ_START,
+ SMIC_READ_NEXT,
+ SMIC_READ_END,
+ SMIC_HOSED
+};
+
+#define MAX_SMIC_READ_SIZE 80
+#define MAX_SMIC_WRITE_SIZE 80
+#define SMIC_MAX_ERROR_RETRIES 3
+
+/* Timeouts in microseconds. */
+#define SMIC_RETRY_TIMEOUT 100000
+
+/* SMIC Flags Register Bits */
+#define SMIC_RX_DATA_READY 0x80
+#define SMIC_TX_DATA_READY 0x40
+#define SMIC_SMI 0x10
+#define SMIC_EVM_DATA_AVAIL 0x08
+#define SMIC_SMS_DATA_AVAIL 0x04
+#define SMIC_FLAG_BSY 0x01
+
+/* SMIC Error Codes */
+#define EC_NO_ERROR 0x00
+#define EC_ABORTED 0x01
+#define EC_ILLEGAL_CONTROL 0x02
+#define EC_NO_RESPONSE 0x03
+#define EC_ILLEGAL_COMMAND 0x04
+#define EC_BUFFER_FULL 0x05
+
+struct si_sm_data
+{
+ enum smic_states state;
+ struct si_sm_io *io;
+ unsigned char write_data[MAX_SMIC_WRITE_SIZE];
+ int write_pos;
+ int write_count;
+ int orig_write_count;
+ unsigned char read_data[MAX_SMIC_READ_SIZE];
+ int read_pos;
+ int truncated;
+ unsigned int error_retries;
+ long smic_timeout;
+};
+
+static unsigned int init_smic_data (struct si_sm_data *smic,
+ struct si_sm_io *io)
+{
+ smic->state = SMIC_IDLE;
+ smic->io = io;
+ smic->write_pos = 0;
+ smic->write_count = 0;
+ smic->orig_write_count = 0;
+ smic->read_pos = 0;
+ smic->error_retries = 0;
+ smic->truncated = 0;
+ smic->smic_timeout = SMIC_RETRY_TIMEOUT;
+
+ /* We use 3 bytes of I/O. */
+ return 3;
+}
+
+static int start_smic_transaction(struct si_sm_data *smic,
+ unsigned char *data, unsigned int size)
+{
+ unsigned int i;
+
+ if ((size < 2) || (size > MAX_SMIC_WRITE_SIZE)) {
+ return -1;
+ }
+ if ((smic->state != SMIC_IDLE) && (smic->state != SMIC_HOSED)) {
+ return -2;
+ }
+ if (smic_debug & SMIC_DEBUG_MSG) {
+ printk(KERN_INFO "start_smic_transaction -");
+ for (i = 0; i < size; i ++) {
+ printk (" %02x", (unsigned char) (data [i]));
+ }
+ printk ("\n");
+ }
+ smic->error_retries = 0;
+ memcpy(smic->write_data, data, size);
+ smic->write_count = size;
+ smic->orig_write_count = size;
+ smic->write_pos = 0;
+ smic->read_pos = 0;
+ smic->state = SMIC_START_OP;
+ smic->smic_timeout = SMIC_RETRY_TIMEOUT;
+ return 0;
+}
+
+static int smic_get_result(struct si_sm_data *smic,
+ unsigned char *data, unsigned int length)
+{
+ int i;
+
+ if (smic_debug & SMIC_DEBUG_MSG) {
+ printk (KERN_INFO "smic_get result -");
+ for (i = 0; i < smic->read_pos; i ++) {
+ printk (" %02x", (smic->read_data [i]));
+ }
+ printk ("\n");
+ }
+ if (length < smic->read_pos) {
+ smic->read_pos = length;
+ smic->truncated = 1;
+ }
+ memcpy(data, smic->read_data, smic->read_pos);
+
+ if ((length >= 3) && (smic->read_pos < 3)) {
+ data[2] = IPMI_ERR_UNSPECIFIED;
+ smic->read_pos = 3;
+ }
+ if (smic->truncated) {
+ data[2] = IPMI_ERR_MSG_TRUNCATED;
+ smic->truncated = 0;
+ }
+ return smic->read_pos;
+}
+
+static inline unsigned char read_smic_flags(struct si_sm_data *smic)
+{
+ return smic->io->inputb(smic->io, 2);
+}
+
+static inline unsigned char read_smic_status(struct si_sm_data *smic)
+{
+ return smic->io->inputb(smic->io, 1);
+}
+
+static inline unsigned char read_smic_data(struct si_sm_data *smic)
+{
+ return smic->io->inputb(smic->io, 0);
+}
+
+static inline void write_smic_flags(struct si_sm_data *smic,
+ unsigned char flags)
+{
+ smic->io->outputb(smic->io, 2, flags);
+}
+
+static inline void write_smic_control(struct si_sm_data *smic,
+ unsigned char control)
+{
+ smic->io->outputb(smic->io, 1, control);
+}
+
+static inline void write_si_sm_data (struct si_sm_data *smic,
+ unsigned char data)
+{
+ smic->io->outputb(smic->io, 0, data);
+}
+
+static inline void start_error_recovery(struct si_sm_data *smic, char *reason)
+{
+ (smic->error_retries)++;
+ if (smic->error_retries > SMIC_MAX_ERROR_RETRIES) {
+ if (smic_debug & SMIC_DEBUG_ENABLE) {
+ printk(KERN_WARNING
+ "ipmi_smic_drv: smic hosed: %s\n", reason);
+ }
+ smic->state = SMIC_HOSED;
+ } else {
+ smic->write_count = smic->orig_write_count;
+ smic->write_pos = 0;
+ smic->read_pos = 0;
+ smic->state = SMIC_START_OP;
+ smic->smic_timeout = SMIC_RETRY_TIMEOUT;
+ }
+}
+
+static inline void write_next_byte(struct si_sm_data *smic)
+{
+ write_si_sm_data(smic, smic->write_data[smic->write_pos]);
+ (smic->write_pos)++;
+ (smic->write_count)--;
+}
+
+static inline void read_next_byte (struct si_sm_data *smic)
+{
+ if (smic->read_pos >= MAX_SMIC_READ_SIZE) {
+ read_smic_data (smic);
+ smic->truncated = 1;
+ } else {
+ smic->read_data[smic->read_pos] = read_smic_data(smic);
+ (smic->read_pos)++;
+ }
+}
+
+/* SMIC Control/Status Code Components */
+#define SMIC_GET_STATUS 0x00 /* Control form's name */
+#define SMIC_READY 0x00 /* Status form's name */
+#define SMIC_WR_START 0x01 /* Unified Control/Status names... */
+#define SMIC_WR_NEXT 0x02
+#define SMIC_WR_END 0x03
+#define SMIC_RD_START 0x04
+#define SMIC_RD_NEXT 0x05
+#define SMIC_RD_END 0x06
+#define SMIC_CODE_MASK 0x0f
+
+#define SMIC_CONTROL 0x00
+#define SMIC_STATUS 0x80
+#define SMIC_CS_MASK 0x80
+
+#define SMIC_SMS 0x40
+#define SMIC_SMM 0x60
+#define SMIC_STREAM_MASK 0x60
+
+/* SMIC Control Codes */
+#define SMIC_CC_SMS_GET_STATUS (SMIC_CONTROL|SMIC_SMS|SMIC_GET_STATUS)
+#define SMIC_CC_SMS_WR_START (SMIC_CONTROL|SMIC_SMS|SMIC_WR_START)
+#define SMIC_CC_SMS_WR_NEXT (SMIC_CONTROL|SMIC_SMS|SMIC_WR_NEXT)
+#define SMIC_CC_SMS_WR_END (SMIC_CONTROL|SMIC_SMS|SMIC_WR_END)
+#define SMIC_CC_SMS_RD_START (SMIC_CONTROL|SMIC_SMS|SMIC_RD_START)
+#define SMIC_CC_SMS_RD_NEXT (SMIC_CONTROL|SMIC_SMS|SMIC_RD_NEXT)
+#define SMIC_CC_SMS_RD_END (SMIC_CONTROL|SMIC_SMS|SMIC_RD_END)
+
+#define SMIC_CC_SMM_GET_STATUS (SMIC_CONTROL|SMIC_SMM|SMIC_GET_STATUS)
+#define SMIC_CC_SMM_WR_START (SMIC_CONTROL|SMIC_SMM|SMIC_WR_START)
+#define SMIC_CC_SMM_WR_NEXT (SMIC_CONTROL|SMIC_SMM|SMIC_WR_NEXT)
+#define SMIC_CC_SMM_WR_END (SMIC_CONTROL|SMIC_SMM|SMIC_WR_END)
+#define SMIC_CC_SMM_RD_START (SMIC_CONTROL|SMIC_SMM|SMIC_RD_START)
+#define SMIC_CC_SMM_RD_NEXT (SMIC_CONTROL|SMIC_SMM|SMIC_RD_NEXT)
+#define SMIC_CC_SMM_RD_END (SMIC_CONTROL|SMIC_SMM|SMIC_RD_END)
+
+/* SMIC Status Codes */
+#define SMIC_SC_SMS_READY (SMIC_STATUS|SMIC_SMS|SMIC_READY)
+#define SMIC_SC_SMS_WR_START (SMIC_STATUS|SMIC_SMS|SMIC_WR_START)
+#define SMIC_SC_SMS_WR_NEXT (SMIC_STATUS|SMIC_SMS|SMIC_WR_NEXT)
+#define SMIC_SC_SMS_WR_END (SMIC_STATUS|SMIC_SMS|SMIC_WR_END)
+#define SMIC_SC_SMS_RD_START (SMIC_STATUS|SMIC_SMS|SMIC_RD_START)
+#define SMIC_SC_SMS_RD_NEXT (SMIC_STATUS|SMIC_SMS|SMIC_RD_NEXT)
+#define SMIC_SC_SMS_RD_END (SMIC_STATUS|SMIC_SMS|SMIC_RD_END)
+
+#define SMIC_SC_SMM_READY (SMIC_STATUS|SMIC_SMM|SMIC_READY)
+#define SMIC_SC_SMM_WR_START (SMIC_STATUS|SMIC_SMM|SMIC_WR_START)
+#define SMIC_SC_SMM_WR_NEXT (SMIC_STATUS|SMIC_SMM|SMIC_WR_NEXT)
+#define SMIC_SC_SMM_WR_END (SMIC_STATUS|SMIC_SMM|SMIC_WR_END)
+#define SMIC_SC_SMM_RD_START (SMIC_STATUS|SMIC_SMM|SMIC_RD_START)
+#define SMIC_SC_SMM_RD_NEXT (SMIC_STATUS|SMIC_SMM|SMIC_RD_NEXT)
+#define SMIC_SC_SMM_RD_END (SMIC_STATUS|SMIC_SMM|SMIC_RD_END)
+
+/* these are the control/status codes we actually use
+ SMIC_CC_SMS_GET_STATUS 0x40
+ SMIC_CC_SMS_WR_START 0x41
+ SMIC_CC_SMS_WR_NEXT 0x42
+ SMIC_CC_SMS_WR_END 0x43
+ SMIC_CC_SMS_RD_START 0x44
+ SMIC_CC_SMS_RD_NEXT 0x45
+ SMIC_CC_SMS_RD_END 0x46
+
+ SMIC_SC_SMS_READY 0xC0
+ SMIC_SC_SMS_WR_START 0xC1
+ SMIC_SC_SMS_WR_NEXT 0xC2
+ SMIC_SC_SMS_WR_END 0xC3
+ SMIC_SC_SMS_RD_START 0xC4
+ SMIC_SC_SMS_RD_NEXT 0xC5
+ SMIC_SC_SMS_RD_END 0xC6
+*/
+
+static enum si_sm_result smic_event (struct si_sm_data *smic, long time)
+{
+ unsigned char status;
+ unsigned char flags;
+ unsigned char data;
+
+ if (smic->state == SMIC_HOSED) {
+ init_smic_data(smic, smic->io);
+ return SI_SM_HOSED;
+ }
+ if (smic->state != SMIC_IDLE) {
+ if (smic_debug & SMIC_DEBUG_STATES) {
+ printk(KERN_INFO
+ "smic_event - smic->smic_timeout = %ld,"
+ " time = %ld\n",
+ smic->smic_timeout, time);
+ }
+/* FIXME: smic_event is sometimes called with time > SMIC_RETRY_TIMEOUT */
+ if (time < SMIC_RETRY_TIMEOUT) {
+ smic->smic_timeout -= time;
+ if (smic->smic_timeout < 0) {
+ start_error_recovery(smic, "smic timed out.");
+ return SI_SM_CALL_WITH_DELAY;
+ }
+ }
+ }
+ flags = read_smic_flags(smic);
+ if (flags & SMIC_FLAG_BSY)
+ return SI_SM_CALL_WITH_DELAY;
+
+ status = read_smic_status (smic);
+ if (smic_debug & SMIC_DEBUG_STATES)
+ printk(KERN_INFO
+ "smic_event - state = %d, flags = 0x%02x,"
+ " status = 0x%02x\n",
+ smic->state, flags, status);
+
+ switch (smic->state) {
+ case SMIC_IDLE:
+ /* in IDLE we check for available messages */
+ if (flags & (SMIC_SMI |
+ SMIC_EVM_DATA_AVAIL | SMIC_SMS_DATA_AVAIL))
+ {
+ return SI_SM_ATTN;
+ }
+ return SI_SM_IDLE;
+
+ case SMIC_START_OP:
+ /* sanity check whether smic is really idle */
+ write_smic_control(smic, SMIC_CC_SMS_GET_STATUS);
+ write_smic_flags(smic, flags | SMIC_FLAG_BSY);
+ smic->state = SMIC_OP_OK;
+ break;
+
+ case SMIC_OP_OK:
+ if (status != SMIC_SC_SMS_READY) {
+ /* this should not happen */
+ start_error_recovery(smic,
+ "state = SMIC_OP_OK,"
+ " status != SMIC_SC_SMS_READY");
+ return SI_SM_CALL_WITH_DELAY;
+ }
+ /* OK so far; smic is idle let us start ... */
+ write_smic_control(smic, SMIC_CC_SMS_WR_START);
+ write_next_byte(smic);
+ write_smic_flags(smic, flags | SMIC_FLAG_BSY);
+ smic->state = SMIC_WRITE_START;
+ break;
+
+ case SMIC_WRITE_START:
+ if (status != SMIC_SC_SMS_WR_START) {
+ start_error_recovery(smic,
+ "state = SMIC_WRITE_START, "
+ "status != SMIC_SC_SMS_WR_START");
+ return SI_SM_CALL_WITH_DELAY;
+ }
+ /* we must not issue WR_(NEXT|END) unless
+ TX_DATA_READY is set */
+ if (flags & SMIC_TX_DATA_READY) {
+ if (smic->write_count == 1) {
+ /* last byte */
+ write_smic_control(smic, SMIC_CC_SMS_WR_END);
+ smic->state = SMIC_WRITE_END;
+ } else {
+ write_smic_control(smic, SMIC_CC_SMS_WR_NEXT);
+ smic->state = SMIC_WRITE_NEXT;
+ }
+ write_next_byte(smic);
+ write_smic_flags(smic, flags | SMIC_FLAG_BSY);
+ }
+ else {
+ return SI_SM_CALL_WITH_DELAY;
+ }
+ break;
+
+ case SMIC_WRITE_NEXT:
+ if (status != SMIC_SC_SMS_WR_NEXT) {
+ start_error_recovery(smic,
+ "state = SMIC_WRITE_NEXT, "
+ "status != SMIC_SC_SMS_WR_NEXT");
+ return SI_SM_CALL_WITH_DELAY;
+ }
+ /* this is the same code as in SMIC_WRITE_START */
+ if (flags & SMIC_TX_DATA_READY) {
+ if (smic->write_count == 1) {
+ write_smic_control(smic, SMIC_CC_SMS_WR_END);
+ smic->state = SMIC_WRITE_END;
+ }
+ else {
+ write_smic_control(smic, SMIC_CC_SMS_WR_NEXT);
+ smic->state = SMIC_WRITE_NEXT;
+ }
+ write_next_byte(smic);
+ write_smic_flags(smic, flags | SMIC_FLAG_BSY);
+ }
+ else {
+ return SI_SM_CALL_WITH_DELAY;
+ }
+ break;
+
+ case SMIC_WRITE_END:
+ if (status != SMIC_SC_SMS_WR_END) {
+ start_error_recovery (smic,
+ "state = SMIC_WRITE_END, "
+ "status != SMIC_SC_SMS_WR_END");
+ return SI_SM_CALL_WITH_DELAY;
+ }
+ /* data register holds an error code */
+ data = read_smic_data(smic);
+ if (data != 0) {
+ if (smic_debug & SMIC_DEBUG_ENABLE) {
+ printk(KERN_INFO
+ "SMIC_WRITE_END: data = %02x\n", data);
+ }
+ start_error_recovery(smic,
+ "state = SMIC_WRITE_END, "
+ "data != SUCCESS");
+ return SI_SM_CALL_WITH_DELAY;
+ } else {
+ smic->state = SMIC_WRITE2READ;
+ }
+ break;
+
+ case SMIC_WRITE2READ:
+ /* we must wait for RX_DATA_READY to be set before we
+ can continue */
+ if (flags & SMIC_RX_DATA_READY) {
+ write_smic_control(smic, SMIC_CC_SMS_RD_START);
+ write_smic_flags(smic, flags | SMIC_FLAG_BSY);
+ smic->state = SMIC_READ_START;
+ } else {
+ return SI_SM_CALL_WITH_DELAY;
+ }
+ break;
+
+ case SMIC_READ_START:
+ if (status != SMIC_SC_SMS_RD_START) {
+ start_error_recovery(smic,
+ "state = SMIC_READ_START, "
+ "status != SMIC_SC_SMS_RD_START");
+ return SI_SM_CALL_WITH_DELAY;
+ }
+ if (flags & SMIC_RX_DATA_READY) {
+ read_next_byte(smic);
+ write_smic_control(smic, SMIC_CC_SMS_RD_NEXT);
+ write_smic_flags(smic, flags | SMIC_FLAG_BSY);
+ smic->state = SMIC_READ_NEXT;
+ } else {
+ return SI_SM_CALL_WITH_DELAY;
+ }
+ break;
+
+ case SMIC_READ_NEXT:
+ switch (status) {
+ /* smic tells us that this is the last byte to be read
+ --> clean up */
+ case SMIC_SC_SMS_RD_END:
+ read_next_byte(smic);
+ write_smic_control(smic, SMIC_CC_SMS_RD_END);
+ write_smic_flags(smic, flags | SMIC_FLAG_BSY);
+ smic->state = SMIC_READ_END;
+ break;
+ case SMIC_SC_SMS_RD_NEXT:
+ if (flags & SMIC_RX_DATA_READY) {
+ read_next_byte(smic);
+ write_smic_control(smic, SMIC_CC_SMS_RD_NEXT);
+ write_smic_flags(smic, flags | SMIC_FLAG_BSY);
+ smic->state = SMIC_READ_NEXT;
+ } else {
+ return SI_SM_CALL_WITH_DELAY;
+ }
+ break;
+ default:
+ start_error_recovery(
+ smic,
+ "state = SMIC_READ_NEXT, "
+ "status != SMIC_SC_SMS_RD_(NEXT|END)");
+ return SI_SM_CALL_WITH_DELAY;
+ }
+ break;
+
+ case SMIC_READ_END:
+ if (status != SMIC_SC_SMS_READY) {
+ start_error_recovery(smic,
+ "state = SMIC_READ_END, "
+ "status != SMIC_SC_SMS_READY");
+ return SI_SM_CALL_WITH_DELAY;
+ }
+ data = read_smic_data(smic);
+ /* data register holds an error code */
+ if (data != 0) {
+ if (smic_debug & SMIC_DEBUG_ENABLE) {
+ printk(KERN_INFO
+ "SMIC_READ_END: data = %02x\n", data);
+ }
+ start_error_recovery(smic,
+ "state = SMIC_READ_END, "
+ "data != SUCCESS");
+ return SI_SM_CALL_WITH_DELAY;
+ } else {
+ smic->state = SMIC_IDLE;
+ return SI_SM_TRANSACTION_COMPLETE;
+ }
+
+ case SMIC_HOSED:
+ init_smic_data(smic, smic->io);
+ return SI_SM_HOSED;
+
+ default:
+ if (smic_debug & SMIC_DEBUG_ENABLE) {
+ printk(KERN_WARNING "smic->state = %d\n", smic->state);
+ start_error_recovery(smic, "state = UNKNOWN");
+ return SI_SM_CALL_WITH_DELAY;
+ }
+ }
+ smic->smic_timeout = SMIC_RETRY_TIMEOUT;
+ return SI_SM_CALL_WITHOUT_DELAY;
+}
+
+static int smic_detect(struct si_sm_data *smic)
+{
+ /* It's impossible for the SMIC fnags register to be all 1's,
+ (assuming a properly functioning, self-initialized BMC)
+ but that's what you get from reading a bogus address, so we
+ test that first. */
+ if (read_smic_flags(smic) == 0xff)
+ return 1;
+
+ return 0;
+}
+
+static void smic_cleanup(struct si_sm_data *kcs)
+{
+}
+
+static int smic_size(void)
+{
+ return sizeof(struct si_sm_data);
+}
+
+struct si_sm_handlers smic_smi_handlers =
+{
+ .version = IPMI_SMIC_VERSION,
+ .init_data = init_smic_data,
+ .start_transaction = start_smic_transaction,
+ .get_result = smic_get_result,
+ .event = smic_event,
+ .detect = smic_detect,
+ .cleanup = smic_cleanup,
+ .size = smic_size,
+};
--- diff/drivers/ide/pci/atiixp.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/ide/pci/atiixp.c 2004-03-16 11:27:48.126871264 +0000
@@ -0,0 +1,512 @@
+/*
+ * linux/drivers/ide/pci/atiixp.c Version 0.01-bart2 Feb. 26, 2004
+ *
+ * Copyright (C) 2003 ATI Inc. <hyu@ati.com>
+ * Copyright (C) 2004 Bartlomiej Zolnierkiewicz
+ *
+ */
+
+#include <linux/config.h>
+#include <linux/types.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/ioport.h>
+#include <linux/pci.h>
+#include <linux/hdreg.h>
+#include <linux/ide.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+
+#include <asm/io.h>
+
+#define ATIIXP_IDE_PIO_TIMING 0x40
+#define ATIIXP_IDE_MDMA_TIMING 0x44
+#define ATIIXP_IDE_PIO_CONTROL 0x48
+#define ATIIXP_IDE_PIO_MODE 0x4a
+#define ATIIXP_IDE_UDMA_CONTROL 0x54
+#define ATIIXP_IDE_UDMA_MODE 0x56
+
+typedef struct {
+ u8 command_width;
+ u8 recover_width;
+} atiixp_ide_timing;
+
+static atiixp_ide_timing pio_timing[] = {
+ { 0x05, 0x0d },
+ { 0x04, 0x07 },
+ { 0x03, 0x04 },
+ { 0x02, 0x02 },
+ { 0x02, 0x00 },
+};
+
+static atiixp_ide_timing mdma_timing[] = {
+ { 0x07, 0x07 },
+ { 0x02, 0x01 },
+ { 0x02, 0x00 },
+};
+
+static int save_mdma_mode[4];
+
+#define DISPLAY_ATIIXP_TIMINGS
+
+#if defined(DISPLAY_ATIIXP_TIMINGS) && defined(CONFIG_PROC_FS)
+
+#include <linux/stat.h>
+#include <linux/proc_fs.h>
+
+static u8 atiixp_proc;
+static struct pci_dev *bmide_dev;
+
+/**
+ * atiixp_get_info - fill in /proc for ATIIXP IDE
+ * @buffer: buffer to fill
+ * @addr: address of user start in buffer
+ * @offset: offset into 'file'
+ * @count: buffer count
+ *
+ * Output summary data on the tuning.
+ */
+
+static int atiixp_get_info(char *buffer, char **addr, off_t offset, int count)
+{
+ char *p = buffer;
+ struct pci_dev *dev = bmide_dev;
+ unsigned long bibma = pci_resource_start(dev, 4);
+ u32 mdma_timing = 0;
+ u16 udma_mode = 0, pio_mode = 0;
+ u8 c0, c1, udma_control = 0;
+
+ p += sprintf(p, "\n ATI ");
+ p += sprintf(p, "ATIIXP Ultra100 IDE Chipset.\n");
+
+ pci_read_config_byte(dev, ATIIXP_IDE_UDMA_CONTROL, &udma_control);
+ pci_read_config_word(dev, ATIIXP_IDE_UDMA_MODE, &udma_mode);
+ pci_read_config_word(dev, ATIIXP_IDE_PIO_MODE, &pio_mode);
+ pci_read_config_dword(dev, ATIIXP_IDE_MDMA_TIMING, &mdma_timing);
+
+ /*
+ * at that point bibma+0x2 et bibma+0xa are byte registers
+ * to investigate:
+ */
+ c0 = inb(bibma + 0x02);
+ c1 = inb(bibma + 0x0a);
+
+ p += sprintf(p, "--------------- Primary Channel "
+ "---------------- Secondary Channel "
+ "-------------\n");
+ p += sprintf(p, " %sabled "
+ " %sabled\n",
+ (c0 & 0x80) ? "dis" : " en",
+ (c1 & 0x80) ? "dis" : " en");
+ p += sprintf(p, "--------------- drive0 --------- drive1 "
+ "-------- drive0 ---------- drive1 ------\n");
+ p += sprintf(p, "DMA enabled: %s %s "
+ " %s %s\n",
+ (c0 & 0x20) ? "yes" : "no ",
+ (c0 & 0x40) ? "yes" : "no ",
+ (c1 & 0x20) ? "yes" : "no ",
+ (c1 & 0x40) ? "yes" : "no " );
+ p += sprintf(p, "UDMA enabled: %s %s "
+ " %s %s\n",
+ (udma_control & 0x01) ? "yes" : "no ",
+ (udma_control & 0x02) ? "yes" : "no ",
+ (udma_control & 0x04) ? "yes" : "no ",
+ (udma_control & 0x08) ? "yes" : "no " );
+ p += sprintf(p, "UDMA mode: %c %c "
+ " %c %c\n",
+ (udma_control & 0x01) ?
+ ((udma_mode & 0x07) + 48) : 'X',
+ (udma_control & 0x02) ?
+ (((udma_mode >> 4) & 0x07) + 48) : 'X',
+ (udma_control & 0x04) ?
+ (((udma_mode >> 8) & 0x07) + 48) : 'X',
+ (udma_control & 0x08) ?
+ (((udma_mode >> 12) & 0x07) + 48) : 'X');
+ p += sprintf(p, "MDMA mode: %c %c "
+ " %c %c\n",
+ (save_mdma_mode[0] && (c0 & 0x20)) ?
+ ((save_mdma_mode[0] & 0xf) + 48) : 'X',
+ (save_mdma_mode[1] && (c0 & 0x40)) ?
+ ((save_mdma_mode[1] & 0xf) + 48) : 'X',
+ (save_mdma_mode[2] && (c1 & 0x20)) ?
+ ((save_mdma_mode[2] & 0xf) + 48) : 'X',
+ (save_mdma_mode[3] && (c1 & 0x40)) ?
+ ((save_mdma_mode[3] & 0xf) + 48) : 'X');
+ p += sprintf(p, "PIO mode: %c %c "
+ " %c %c\n",
+ (c0 & 0x20) ? 'X' : ((pio_mode & 0x07) + 48),
+ (c0 & 0x40) ? 'X' : (((pio_mode >> 4) & 0x07) + 48),
+ (c1 & 0x20) ? 'X' : (((pio_mode >> 8) & 0x07) + 48),
+ (c1 & 0x40) ? 'X' : (((pio_mode >> 12) & 0x07) + 48));
+
+ return p - buffer; /* => must be less than 4k! */
+}
+#endif /* defined(DISPLAY_ATIIXP_TIMINGS) && defined(CONFIG_PROC_FS) */
+
+/**
+ * atiixp_ratemask - compute rate mask for ATIIXP IDE
+ * @drive: IDE drive to compute for
+ *
+ * Returns the available modes for the ATIIXP IDE controller.
+ */
+
+static u8 atiixp_ratemask(ide_drive_t *drive)
+{
+ u8 mode = 3;
+
+ if (!eighty_ninty_three(drive))
+ mode = min(mode, (u8)1);
+ return mode;
+}
+
+/**
+ * atiixp_dma_2_pio - return the PIO mode matching DMA
+ * @xfer_rate: transfer speed
+ *
+ * Returns the nearest equivalent PIO timing for the PIO or DMA
+ * mode requested by the controller.
+ */
+
+static u8 atiixp_dma_2_pio(u8 xfer_rate) {
+ switch(xfer_rate) {
+ case XFER_UDMA_6:
+ case XFER_UDMA_5:
+ case XFER_UDMA_4:
+ case XFER_UDMA_3:
+ case XFER_UDMA_2:
+ case XFER_UDMA_1:
+ case XFER_UDMA_0:
+ case XFER_MW_DMA_2:
+ case XFER_PIO_4:
+ return 4;
+ case XFER_MW_DMA_1:
+ case XFER_PIO_3:
+ return 3;
+ case XFER_SW_DMA_2:
+ case XFER_PIO_2:
+ return 2;
+ case XFER_MW_DMA_0:
+ case XFER_SW_DMA_1:
+ case XFER_SW_DMA_0:
+ case XFER_PIO_1:
+ case XFER_PIO_0:
+ case XFER_PIO_SLOW:
+ default:
+ return 0;
+ }
+}
+
+static int atiixp_ide_dma_host_on(ide_drive_t *drive)
+{
+ struct pci_dev *dev = drive->hwif->pci_dev;
+ unsigned long flags;
+ u16 tmp16;
+
+ spin_lock_irqsave(&ide_lock, flags);
+
+ pci_read_config_word(dev, ATIIXP_IDE_UDMA_CONTROL, &tmp16);
+ if (save_mdma_mode[drive->dn])
+ tmp16 &= ~(1 << drive->dn);
+ else
+ tmp16 |= (1 << drive->dn);
+ pci_write_config_word(dev, ATIIXP_IDE_UDMA_CONTROL, tmp16);
+
+ spin_unlock_irqrestore(&ide_lock, flags);
+
+ return __ide_dma_host_on(drive);
+}
+
+static int atiixp_ide_dma_host_off(ide_drive_t *drive)
+{
+ struct pci_dev *dev = drive->hwif->pci_dev;
+ unsigned long flags;
+ u16 tmp16;
+
+ spin_lock_irqsave(&ide_lock, flags);
+
+ pci_read_config_word(dev, ATIIXP_IDE_UDMA_CONTROL, &tmp16);
+ tmp16 &= ~(1 << drive->dn);
+ pci_write_config_word(dev, ATIIXP_IDE_UDMA_CONTROL, tmp16);
+
+ spin_unlock_irqrestore(&ide_lock, flags);
+
+ return __ide_dma_host_off(drive);
+}
+
+/**
+ * atiixp_tune_drive - tune a drive attached to a ATIIXP
+ * @drive: drive to tune
+ * @pio: desired PIO mode
+ *
+ * Set the interface PIO mode.
+ */
+
+static void atiixp_tuneproc(ide_drive_t *drive, u8 pio)
+{
+ struct pci_dev *dev = drive->hwif->pci_dev;
+ unsigned long flags;
+ int timing_shift = (drive->dn & 2) ? 16 : 0 + (drive->dn & 1) ? 0 : 8;
+ u32 pio_timing_data;
+ u16 pio_mode_data;
+
+ spin_lock_irqsave(&ide_lock, flags);
+
+ pci_read_config_word(dev, ATIIXP_IDE_PIO_MODE, &pio_mode_data);
+ pio_mode_data &= ~(0x07 << (drive->dn * 4));
+ pio_mode_data |= (pio << (drive->dn * 4));
+ pci_write_config_word(dev, ATIIXP_IDE_PIO_MODE, pio_mode_data);
+
+ pci_read_config_dword(dev, ATIIXP_IDE_PIO_TIMING, &pio_timing_data);
+ pio_timing_data &= ~(0xff << timing_shift);
+ pio_timing_data |= (pio_timing[pio].recover_width << timing_shift) |
+ (pio_timing[pio].command_width << (timing_shift + 4));
+ pci_write_config_dword(dev, ATIIXP_IDE_PIO_TIMING, pio_timing_data);
+
+ spin_unlock_irqrestore(&ide_lock, flags);
+}
+
+/**
+ * atiixp_tune_chipset - tune a ATIIXP interface
+ * @drive: IDE drive to tune
+ * @xferspeed: speed to configure
+ *
+ * Set a ATIIXP interface channel to the desired speeds. This involves
+ * requires the right timing data into the ATIIXP configuration space
+ * then setting the drive parameters appropriately
+ */
+
+static int atiixp_speedproc(ide_drive_t *drive, u8 xferspeed)
+{
+ struct pci_dev *dev = drive->hwif->pci_dev;
+ unsigned long flags;
+ int timing_shift = (drive->dn & 2) ? 16 : 0 + (drive->dn & 1) ? 0 : 8;
+ u32 tmp32;
+ u16 tmp16;
+ u8 speed, pio;
+
+ speed = ide_rate_filter(atiixp_ratemask(drive), xferspeed);
+
+ spin_lock_irqsave(&ide_lock, flags);
+
+ save_mdma_mode[drive->dn] = 0;
+ if (speed >= XFER_UDMA_0) {
+ pci_read_config_word(dev, ATIIXP_IDE_UDMA_MODE, &tmp16);
+ tmp16 &= ~(0x07 << (drive->dn * 4));
+ tmp16 |= ((speed & 0x07) << (drive->dn * 4));
+ pci_write_config_word(dev, ATIIXP_IDE_UDMA_MODE, tmp16);
+ } else {
+ if ((speed >= XFER_MW_DMA_0) && (speed <= XFER_MW_DMA_2)) {
+ save_mdma_mode[drive->dn] = speed;
+ pci_read_config_dword(dev, ATIIXP_IDE_MDMA_TIMING, &tmp32);
+ tmp32 &= ~(0xff << timing_shift);
+ tmp32 |= (mdma_timing[speed & 0x03].recover_width << timing_shift) |
+ (mdma_timing[speed & 0x03].command_width << (timing_shift + 4));
+ pci_write_config_dword(dev, ATIIXP_IDE_MDMA_TIMING, tmp32);
+ }
+ }
+
+ spin_unlock_irqrestore(&ide_lock, flags);
+
+ if (speed >= XFER_SW_DMA_0)
+ pio = atiixp_dma_2_pio(speed);
+ else
+ pio = speed - XFER_PIO_0;
+
+ atiixp_tuneproc(drive, pio);
+
+ return ide_config_drive_speed(drive, speed);
+}
+
+/**
+ * atiixp_config_drive_for_dma - configure drive for DMA
+ * @drive: IDE drive to configure
+ *
+ * Set up a ATIIXP interface channel for the best available speed.
+ * We prefer UDMA if it is available and then MWDMA. If DMA is
+ * not available we switch to PIO and return 0.
+ */
+
+static int atiixp_config_drive_for_dma(ide_drive_t *drive)
+{
+ u8 speed = ide_dma_speed(drive, atiixp_ratemask(drive));
+
+ /* If no DMA speed was available then disable DMA and use PIO. */
+ if (!speed) {
+ u8 tspeed = ide_get_best_pio_mode(drive, 255, 5, NULL);
+ speed = atiixp_dma_2_pio(XFER_PIO_0 + tspeed) + XFER_PIO_0;
+ }
+
+ (void) atiixp_speedproc(drive, speed);
+ return ide_dma_enable(drive);
+}
+
+/**
+ * atiixp_dma_check - set up an IDE device
+ * @drive: IDE drive to configure
+ *
+ * Set up the ATIIXP interface for the best available speed on this
+ * interface, preferring DMA to PIO.
+ */
+
+static int atiixp_dma_check(ide_drive_t *drive)
+{
+ ide_hwif_t *hwif = HWIF(drive);
+ struct hd_driveid *id = drive->id;
+ u8 tspeed, speed;
+
+ drive->init_speed = 0;
+
+ if ((id->capability & 1) && drive->autodma) {
+ /* Consult the list of known "bad" drives */
+ if (__ide_dma_bad_drive(drive))
+ goto fast_ata_pio;
+ if (id->field_valid & 4) {
+ if (id->dma_ultra & hwif->ultra_mask) {
+ /* Force if Capable UltraDMA */
+ if ((id->field_valid & 2) &&
+ (!atiixp_config_drive_for_dma(drive)))
+ goto try_dma_modes;
+ }
+ } else if (id->field_valid & 2) {
+try_dma_modes:
+ if ((id->dma_mword & hwif->mwdma_mask) ||
+ (id->dma_1word & hwif->swdma_mask)) {
+ /* Force if Capable regular DMA modes */
+ if (!atiixp_config_drive_for_dma(drive))
+ goto no_dma_set;
+ }
+ } else if (__ide_dma_good_drive(drive) &&
+ (id->eide_dma_time < 150)) {
+ /* Consult the list of known "good" drives */
+ if (!atiixp_config_drive_for_dma(drive))
+ goto no_dma_set;
+ } else {
+ goto fast_ata_pio;
+ }
+ return hwif->ide_dma_on(drive);
+ } else if ((id->capability & 8) || (id->field_valid & 2)) {
+fast_ata_pio:
+no_dma_set:
+ tspeed = ide_get_best_pio_mode(drive, 255, 5, NULL);
+ speed = atiixp_dma_2_pio(XFER_PIO_0 + tspeed) + XFER_PIO_0;
+ hwif->speedproc(drive, speed);
+ return hwif->ide_dma_off_quietly(drive);
+ }
+ /* IORDY not supported */
+ return 0;
+}
+
+/**
+ * init_chipset_atiixp - set up the ATIIXP chipset
+ * @dev: PCI device to set up
+ * @name: Name of the device
+ *
+ * Initialize the PCI device as required. For the ATIIXP this turns
+ * out to be nice and simple
+ */
+
+static unsigned int __devinit init_chipset_atiixp(struct pci_dev *dev, const char *name)
+{
+#if defined(DISPLAY_ATIIXP_TIMINGS) && defined(CONFIG_PROC_FS)
+ if (!atiixp_proc) {
+ atiixp_proc = 1;
+ bmide_dev = dev;
+ ide_pci_create_host_proc("atiixp", atiixp_get_info);
+ }
+#endif /* DISPLAY_ATIIXP_TIMINGS && CONFIG_PROC_FS */
+ return 0;
+}
+
+/**
+ * init_hwif_atiixp - fill in the hwif for the ATIIXP
+ * @hwif: IDE interface
+ *
+ * Set up the ide_hwif_t for the ATIIXP interface according to the
+ * capabilities of the hardware.
+ */
+
+static void __devinit init_hwif_atiixp(ide_hwif_t *hwif)
+{
+ if (!hwif->irq)
+ hwif->irq = hwif->channel ? 15 : 14;
+
+ hwif->autodma = 0;
+ hwif->tuneproc = &atiixp_tuneproc;
+ hwif->speedproc = &atiixp_speedproc;
+ hwif->drives[0].autotune = 1;
+ hwif->drives[1].autotune = 1;
+
+ if (!hwif->dma_base)
+ return;
+
+ hwif->atapi_dma = 1;
+ hwif->ultra_mask = 0x3f;
+ hwif->mwdma_mask = 0x06;
+ hwif->swdma_mask = 0x04;
+
+ /* FIXME: proper cable detection needed */
+ hwif->udma_four = 1;
+ hwif->ide_dma_host_on = &atiixp_ide_dma_host_on;
+ hwif->ide_dma_host_off = &atiixp_ide_dma_host_off;
+ hwif->ide_dma_check = &atiixp_dma_check;
+ if (!noautodma)
+ hwif->autodma = 1;
+
+ hwif->drives[1].autodma = hwif->autodma;
+ hwif->drives[0].autodma = hwif->autodma;
+}
+
+static ide_pci_device_t atiixp_pci_info[] __devinitdata = {
+ { /* 0 */
+ .vendor = PCI_VENDOR_ID_ATI,
+ .device = PCI_DEVICE_ID_ATI_IXP_IDE,
+ .name = "ATIIXP",
+ .init_chipset = init_chipset_atiixp,
+ .init_hwif = init_hwif_atiixp,
+ .channels = 2,
+ .autodma = AUTODMA,
+ .enablebits = {{0x48,0x01,0x00}, {0x48,0x08,0x00}},
+ .bootable = ON_BOARD,
+ }
+};
+
+/**
+ * atiixp_init_one - called when a ATIIXP is found
+ * @dev: the atiixp device
+ * @id: the matching pci id
+ *
+ * Called when the PCI registration layer (or the IDE initialization)
+ * finds a device matching our IDE device tables.
+ */
+
+static int __devinit atiixp_init_one(struct pci_dev *dev, const struct pci_device_id *id)
+{
+ ide_pci_device_t *d = &atiixp_pci_info[id->driver_data];
+
+ if (dev->device != d->device)
+ BUG();
+ ide_setup_pci_device(dev, d);
+ return 0;
+}
+
+static struct pci_device_id atiixp_pci_tbl[] = {
+ { PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
+ { 0, },
+};
+
+static struct pci_driver driver = {
+ .name = "ATIIXP IDE",
+ .id_table = atiixp_pci_tbl,
+ .probe = atiixp_init_one,
+};
+
+static int atiixp_ide_init(void)
+{
+ return ide_pci_register_driver(&driver);
+}
+
+module_init(atiixp_ide_init);
+
+MODULE_AUTHOR("HUI YU");
+MODULE_DESCRIPTION("PCI driver module for ATI IXP IDE");
+MODULE_LICENSE("GPL");
--- diff/drivers/input/keyboard/lkkbd.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/input/keyboard/lkkbd.c 2004-03-16 11:27:48.143868680 +0000
@@ -0,0 +1,625 @@
+/*
+ * Copyright (C) 2004 by Jan-Benedict Glaw <jbglaw@lug-owl.de>
+ */
+
+/*
+ * LK keyboard driver for Linux, based on sunkbd.c (C) by Vojtech Pavlik
+ */
+
+/*
+ * DEC LK201 and LK401 keyboard driver for Linux (primary for DECstations
+ * and VAXstations, but can also be used on any standard RS232 with an
+ * adaptor).
+ *
+ * DISCLAUNER: This works for _me_. If you break anything by using the
+ * information given below, I will _not_ be lieable!
+ *
+ * RJ11 pinout: To DB9: Or DB25:
+ * 1 - RxD <----> Pin 3 (TxD) <-> Pin 2 (TxD)
+ * 2 - GND <----> Pin 5 (GND) <-> Pin 7 (GND)
+ * 4 - TxD <----> Pin 2 (RxD) <-> Pin 3 (RxD)
+ * 3 - +12V (from HDD drive connector), DON'T connect to DB9 or DB25!!!
+ *
+ * Pin numbers for DB9 and DB25 are noted on the plug (quite small:). For
+ * RJ11, it's like this:
+ *
+ * __=__ Hold the plug in front of you, cable downwards,
+ * /___/| nose is hidden behind the plug. Now, pin 1 is at
+ * |1234|| the left side, pin 4 at the right and 2 and 3 are
+ * |IIII|| in between, of course:)
+ * | ||
+ * |____|/
+ * || So the adaptor consists of three connected cables
+ * || for data transmission (RxD and TxD) and signal ground.
+ * Additionally, you have to get +12V from somewhere.
+ * Most easily, you'll get that from a floppy or HDD power connector.
+ * It's the yellow cable there (black is ground and red is +5V).
+ */
+
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Should you need to contact me, the author, you can do so either by
+ * email or by paper mail:
+ * Jan-Benedict Glaw, Lilienstra�e 16, 33790 H�rste (near Halle/Westf.),
+ * Germany.
+ */
+
+#include <linux/delay.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/interrupt.h>
+#include <linux/init.h>
+#include <linux/input.h>
+#include <linux/serio.h>
+#include <linux/workqueue.h>
+
+
+MODULE_AUTHOR ("Jan-Benedict Glaw <jblaw@lug-owl.de>");
+MODULE_DESCRIPTION ("LK keyboard driver");
+MODULE_LICENSE ("GPL");
+
+/*
+ * Known parameters:
+ * bell_volume
+ * keyclick_volume
+ * ctrlclick_volume
+ *
+ * Please notice that there's not yet an API to set these at runtime.
+ */
+static int bell_volume = 100; /* % */
+module_param (bell_volume, int, 0);
+MODULE_PARM_DESC (bell_volume, "Bell volume (in %). default is 100%");
+
+static int keyclick_volume = 100; /* % */
+module_param (keyclick_volume, int, 0);
+MODULE_PARM_DESC (keyclick_volume, "Keyclick volume (in %), default is 100%");
+
+static int ctrlclick_volume = 100; /* % */
+module_param (ctrlclick_volume, int, 0);
+MODULE_PARM_DESC (ctrlclick_volume, "Ctrlclick volume (in %), default is 100%");
+
+
+
+#undef LKKBD_DEBUG
+#ifdef LKKBD_DEBUG
+#define DBG(x...) printk (x)
+#else
+#define DBG(x...) do {} while (0)
+#endif
+
+/* LED control */
+#define LK_LED_WAIT 0x81
+#define LK_LED_COMPOSE 0x82
+#define LK_LED_SHIFTLOCK 0x84
+#define LK_LED_SCROLLLOCK 0x88
+#define LK_CMD_LED_ON 0x13
+#define LK_CMD_LED_OFF 0x11
+
+/* Mode control */
+#define LK_MODE_DOWN 0x80
+#define LK_MODE_AUTODOWN 0x82
+#define LK_MODE_UPDOWN 0x86
+#define LK_CMD_SET_MODE(mode,div) ((mode) | ((div) << 3))
+
+/* Misc commands */
+#define LK_CMD_ENABLE_KEYCLICK 0x1b
+#define LK_CMD_DISABLE_KEYCLICK 0x99
+#define LK_CMD_DISABLE_BELL 0xa1
+#define LK_CMD_SOUND_BELL 0xa7
+#define LK_CMD_ENABLE_BELL 0x23
+#define LK_CMD_DISABLE_CTRCLICK 0xb9
+#define LK_CMD_ENABLE_CTRCLICK 0xbb
+#define LK_CMD_SET_DEFAULTS 0xd3
+#define LK_CMD_POWERCYCLE_RESET 0xfd
+#define LK_CMD_ENABLE_LK401 0xe9
+
+/* Misc responses from keyboard */
+#define LK_ALL_KEYS_UP 0xb3
+#define LK_METRONOME 0xb4
+#define LK_OUTPUT_ERROR 0xb5
+#define LK_INPUT_ERROR 0xb6
+#define LK_KBD_LOCKED 0xb7
+#define LK_KBD_TEST_MODE_ACK 0xb8
+#define LK_PREFIX_KEY_DOWN 0xb9
+#define LK_MODE_CHANGE_ACK 0xba
+#define LK_RESPONSE_RESERVED 0xbb
+
+#define LK_NUM_KEYCODES 256
+typedef u_int16_t lk_keycode_t;
+
+
+
+static lk_keycode_t lkkbd_keycode[LK_NUM_KEYCODES] = {
+ [0x56] = KEY_F1,
+ [0x57] = KEY_F2,
+ [0x58] = KEY_F3,
+ [0x59] = KEY_F4,
+ [0x5a] = KEY_F5,
+ [0x64] = KEY_F6,
+ [0x65] = KEY_F7,
+ [0x66] = KEY_F8,
+ [0x67] = KEY_F9,
+ [0x68] = KEY_F10,
+ [0x71] = KEY_F11,
+ [0x72] = KEY_F12,
+ [0x73] = KEY_F13,
+ [0x74] = KEY_F14,
+ [0x7c] = KEY_F15,
+ [0x7d] = KEY_F16,
+ [0x80] = KEY_F17,
+ [0x81] = KEY_F18,
+ [0x82] = KEY_F19,
+ [0x83] = KEY_F20,
+ [0x8a] = KEY_FIND,
+ [0x8b] = KEY_INSERT,
+ [0x8c] = KEY_DELETE,
+ [0x8d] = KEY_SELECT,
+ [0x8e] = KEY_PAGEUP,
+ [0x8f] = KEY_PAGEDOWN,
+ [0x92] = KEY_KP0,
+ [0x94] = KEY_KPDOT,
+ [0x95] = KEY_KPENTER,
+ [0x96] = KEY_KP1,
+ [0x97] = KEY_KP2,
+ [0x98] = KEY_KP3,
+ [0x99] = KEY_KP4,
+ [0x9a] = KEY_KP5,
+ [0x9b] = KEY_KP6,
+ [0x9c] = KEY_KPCOMMA,
+ [0x9d] = KEY_KP7,
+ [0x9e] = KEY_KP8,
+ [0x9f] = KEY_KP9,
+ [0xa0] = KEY_KPMINUS,
+ [0xa1] = KEY_PROG1,
+ [0xa2] = KEY_PROG2,
+ [0xa3] = KEY_PROG3,
+ [0xa4] = KEY_PROG4,
+ [0xa7] = KEY_LEFT,
+ [0xa8] = KEY_RIGHT,
+ [0xa9] = KEY_DOWN,
+ [0xaa] = KEY_UP,
+ [0xab] = KEY_RIGHTSHIFT,
+ [0xac] = KEY_LEFTALT,
+ [0xad] = KEY_COMPOSE, /* Right Compose, that is. */
+ [0xae] = KEY_LEFTSHIFT, /* Same as KEY_RIGHTSHIFT on LK201 */
+ [0xaf] = KEY_LEFTCTRL,
+ [0xb0] = KEY_CAPSLOCK,
+ [0xb1] = KEY_COMPOSE, /* Left Compose, that is. */
+ [0xb2] = KEY_RIGHTALT,
+ [0xbc] = KEY_BACKSPACE,
+ [0xbd] = KEY_ENTER,
+ [0xbe] = KEY_TAB,
+ [0xbf] = KEY_ESC,
+ [0xc0] = KEY_1,
+ [0xc1] = KEY_Q,
+ [0xc2] = KEY_A,
+ [0xc3] = KEY_Z,
+ [0xc5] = KEY_2,
+ [0xc6] = KEY_W,
+ [0xc7] = KEY_S,
+ [0xc8] = KEY_X,
+ [0xc9] = KEY_102ND,
+ [0xcb] = KEY_3,
+ [0xcc] = KEY_E,
+ [0xcd] = KEY_D,
+ [0xce] = KEY_C,
+ [0xd0] = KEY_4,
+ [0xd1] = KEY_R,
+ [0xd2] = KEY_F,
+ [0xd3] = KEY_V,
+ [0xd4] = KEY_SPACE,
+ [0xd6] = KEY_5,
+ [0xd7] = KEY_T,
+ [0xd8] = KEY_G,
+ [0xd9] = KEY_B,
+ [0xdb] = KEY_6,
+ [0xdc] = KEY_Y,
+ [0xdd] = KEY_H,
+ [0xde] = KEY_N,
+ [0xe0] = KEY_7,
+ [0xe1] = KEY_U,
+ [0xe2] = KEY_J,
+ [0xe3] = KEY_M,
+ [0xe5] = KEY_8,
+ [0xe6] = KEY_I,
+ [0xe7] = KEY_K,
+ [0xe8] = KEY_COMMA,
+ [0xea] = KEY_9,
+ [0xeb] = KEY_O,
+ [0xec] = KEY_L,
+ [0xed] = KEY_DOT,
+ [0xef] = KEY_0,
+ [0xf0] = KEY_P,
+ [0xf2] = KEY_SEMICOLON,
+ [0xf3] = KEY_SLASH,
+ [0xf5] = KEY_EQUAL,
+ [0xf6] = KEY_RIGHTBRACE,
+ [0xf7] = KEY_BACKSLASH,
+ [0xf9] = KEY_MINUS,
+ [0xfa] = KEY_LEFTBRACE,
+ [0xfb] = KEY_APOSTROPHE,
+};
+
+#define CHECK_LED(LED, BITS) do { \
+ if (test_bit (LED, lk->dev.led)) \
+ leds_on |= BITS; \
+ else \
+ leds_off |= BITS; \
+ } while (0)
+
+/*
+ * Per-keyboard data
+ */
+struct lkkbd {
+ lk_keycode_t keycode[LK_NUM_KEYCODES];
+ int ignore_bytes;
+ struct input_dev dev;
+ struct serio *serio;
+ struct work_struct tq;
+ char name[64];
+ char phys[32];
+ char type;
+ int bell_volume;
+ int keyclick_volume;
+ int ctrlclick_volume;
+};
+
+/*
+ * Calculate volume parameter byte for a given volume.
+ */
+static unsigned char
+volume_to_hw (int volume_percent)
+{
+ unsigned char ret = 0;
+
+ if (volume_percent < 0)
+ volume_percent = 0;
+ if (volume_percent > 100)
+ volume_percent = 100;
+
+ if (volume_percent >= 0)
+ ret = 7;
+ if (volume_percent >= 13) /* 12.5 */
+ ret = 6;
+ if (volume_percent >= 25)
+ ret = 5;
+ if (volume_percent >= 38) /* 37.5 */
+ ret = 4;
+ if (volume_percent >= 50)
+ ret = 3;
+ if (volume_percent >= 63) /* 62.5 */
+ ret = 2; /* This is the default volume */
+ if (volume_percent >= 75)
+ ret = 1;
+ if (volume_percent >= 88) /* 87.5 */
+ ret = 0;
+
+ ret |= 0x80;
+
+ return ret;
+}
+
+/*
+ * lkkbd_interrupt() is called by the low level driver when a character
+ * is received.
+ */
+static irqreturn_t
+lkkbd_interrupt (struct serio *serio, unsigned char data, unsigned int flags,
+ struct pt_regs *regs)
+{
+ struct lkkbd *lk = serio->private;
+ int i;
+
+ DBG (KERN_INFO "Got byte 0x%02x\n", data);
+
+ if (lk->ignore_bytes > 0) {
+ DBG (KERN_INFO "Ignoring a byte on %s\n",
+ lk->name);
+ lk->ignore_bytes--;
+ return IRQ_HANDLED;
+ }
+
+ switch (data) {
+ case LK_ALL_KEYS_UP:
+ input_regs (&lk->dev, regs);
+ for (i = 0; i < ARRAY_SIZE (lkkbd_keycode); i++)
+ if (lk->keycode[i] != KEY_RESERVED)
+ input_report_key (&lk->dev, lk->keycode[i], 0);
+ input_sync (&lk->dev);
+ break;
+ case LK_METRONOME:
+ DBG (KERN_INFO "Got LK_METRONOME and don't "
+ "know how to handle...\n");
+ break;
+ case LK_OUTPUT_ERROR:
+ DBG (KERN_INFO "Got LK_OUTPUT_ERROR and don't "
+ "know how to handle...\n");
+ break;
+ case LK_INPUT_ERROR:
+ DBG (KERN_INFO "Got LK_INPUT_ERROR and don't "
+ "know how to handle...\n");
+ break;
+ case LK_KBD_LOCKED:
+ DBG (KERN_INFO "Got LK_KBD_LOCKED and don't "
+ "know how to handle...\n");
+ break;
+ case LK_KBD_TEST_MODE_ACK:
+ DBG (KERN_INFO "Got LK_KBD_TEST_MODE_ACK and don't "
+ "know how to handle...\n");
+ break;
+ case LK_PREFIX_KEY_DOWN:
+ DBG (KERN_INFO "Got LK_PREFIX_KEY_DOWN and don't "
+ "know how to handle...\n");
+ break;
+ case LK_MODE_CHANGE_ACK:
+ DBG (KERN_INFO "Got LK_MODE_CHANGE_ACK and ignored "
+ "it properly...\n");
+ break;
+ case LK_RESPONSE_RESERVED:
+ DBG (KERN_INFO "Got LK_RESPONSE_RESERVED and don't "
+ "know how to handle...\n");
+ break;
+ case 0x01:
+ DBG (KERN_INFO "Got 0x01, scheduling re-initialization\n");
+ lk->ignore_bytes = 3;
+ schedule_work (&lk->tq);
+ break;
+
+ default:
+ if (lk->keycode[data] != KEY_RESERVED) {
+ input_regs (&lk->dev, regs);
+ if (!test_bit (lk->keycode[data], lk->dev.key))
+ input_report_key (&lk->dev, lk->keycode[data], 1);
+ else
+ input_report_key (&lk->dev, lk->keycode[data], 0);
+ input_sync (&lk->dev);
+ } else
+ printk (KERN_WARNING "%s: Unknown key with "
+ "scancode %02x on %s.\n",
+ __FILE__, data, lk->name);
+ }
+
+ return IRQ_HANDLED;
+}
+
+/*
+ * lkkbd_event() handles events from the input module.
+ */
+static int
+lkkbd_event (struct input_dev *dev, unsigned int type, unsigned int code,
+ int value)
+{
+ struct lkkbd *lk = dev->private;
+ unsigned char leds_on = 0;
+ unsigned char leds_off = 0;
+
+ switch (type) {
+ case EV_LED:
+ CHECK_LED (LED_CAPSL, LK_LED_SHIFTLOCK);
+ CHECK_LED (LED_COMPOSE, LK_LED_COMPOSE);
+ CHECK_LED (LED_SCROLLL, LK_LED_SCROLLLOCK);
+ CHECK_LED (LED_SLEEP, LK_LED_WAIT);
+ if (leds_on != 0) {
+ lk->serio->write (lk->serio, LK_CMD_LED_ON);
+ lk->serio->write (lk->serio, leds_on);
+ }
+ if (leds_off != 0) {
+ lk->serio->write (lk->serio, LK_CMD_LED_OFF);
+ lk->serio->write (lk->serio, leds_off);
+ }
+ return 0;
+
+ case EV_SND:
+ switch (code) {
+ case SND_CLICK:
+ if (value == 0) {
+ DBG ("%s: Deactivating key clicks\n", __FUNCTION__);
+ lk->serio->write (lk->serio, LK_CMD_DISABLE_KEYCLICK);
+ lk->serio->write (lk->serio, LK_CMD_DISABLE_CTRCLICK);
+ } else {
+ DBG ("%s: Activating key clicks\n", __FUNCTION__);
+ lk->serio->write (lk->serio, LK_CMD_ENABLE_KEYCLICK);
+ lk->serio->write (lk->serio, volume_to_hw (lk->keyclick_volume));
+ lk->serio->write (lk->serio, LK_CMD_ENABLE_CTRCLICK);
+ lk->serio->write (lk->serio, volume_to_hw (lk->ctrlclick_volume));
+ }
+ return 0;
+
+ case SND_BELL:
+ if (value != 0)
+ lk->serio->write (lk->serio, LK_CMD_SOUND_BELL);
+
+ return 0;
+ }
+ break;
+
+ default:
+ printk (KERN_ERR "%s (): Got unknown type %d, code %d, value %d\n",
+ __FUNCTION__, type, code, value);
+ }
+
+ return -1;
+}
+
+/*
+ * lkkbd_reinit() sets leds and beeps to a state the computer remembers they
+ * were in.
+ */
+static void
+lkkbd_reinit (void *data)
+{
+ struct lkkbd *lk = data;
+ int division;
+ unsigned char leds_on = 0;
+ unsigned char leds_off = 0;
+
+ /* Reset parameters */
+ lk->serio->write (lk->serio, LK_CMD_SET_DEFAULTS);
+
+ /* Set LEDs */
+ CHECK_LED (LED_CAPSL, LK_LED_SHIFTLOCK);
+ CHECK_LED (LED_COMPOSE, LK_LED_COMPOSE);
+ CHECK_LED (LED_SCROLLL, LK_LED_SCROLLLOCK);
+ CHECK_LED (LED_SLEEP, LK_LED_WAIT);
+ if (leds_on != 0) {
+ lk->serio->write (lk->serio, LK_CMD_LED_ON);
+ lk->serio->write (lk->serio, leds_on);
+ }
+ if (leds_off != 0) {
+ lk->serio->write (lk->serio, LK_CMD_LED_OFF);
+ lk->serio->write (lk->serio, leds_off);
+ }
+
+ /*
+ * Try to activate extended LK401 mode. This command will
+ * only work with a LK401 keyboard and grants access to
+ * LAlt, RAlt, RCompose and RShift.
+ */
+ lk->serio->write (lk->serio, LK_CMD_ENABLE_LK401);
+
+ /* Set all keys to UPDOWN mode */
+ for (division = 1; division <= 14; division++)
+ lk->serio->write (lk->serio, LK_CMD_SET_MODE (LK_MODE_UPDOWN,
+ division));
+
+ /* Enable bell and set volume */
+ lk->serio->write (lk->serio, LK_CMD_ENABLE_BELL);
+ lk->serio->write (lk->serio, volume_to_hw (lk->bell_volume));
+
+ /* Enable/disable keyclick (and possibly set volume) */
+ if (test_bit (SND_CLICK, lk->dev.snd)) {
+ lk->serio->write (lk->serio, LK_CMD_ENABLE_KEYCLICK);
+ lk->serio->write (lk->serio, volume_to_hw (lk->keyclick_volume));
+ lk->serio->write (lk->serio, LK_CMD_ENABLE_CTRCLICK);
+ lk->serio->write (lk->serio, volume_to_hw (lk->ctrlclick_volume));
+ } else {
+ lk->serio->write (lk->serio, LK_CMD_DISABLE_KEYCLICK);
+ lk->serio->write (lk->serio, LK_CMD_DISABLE_CTRCLICK);
+ }
+
+ /* Sound the bell if needed */
+ if (test_bit (SND_BELL, lk->dev.snd))
+ lk->serio->write (lk->serio, LK_CMD_SOUND_BELL);
+}
+
+/*
+ * lkkbd_connect() probes for a LK keyboard and fills the necessary structures.
+ */
+static void
+lkkbd_connect (struct serio *serio, struct serio_dev *dev)
+{
+ struct lkkbd *lk;
+ int i;
+
+ if ((serio->type & SERIO_TYPE) != SERIO_RS232)
+ return;
+ if (!(serio->type & SERIO_PROTO))
+ return;
+ if ((serio->type & SERIO_PROTO) && (serio->type & SERIO_PROTO) != SERIO_LKKBD)
+ return;
+
+ if (!(lk = kmalloc (sizeof (struct lkkbd), GFP_KERNEL)))
+ return;
+ memset (lk, 0, sizeof (struct lkkbd));
+
+ init_input_dev (&lk->dev);
+
+ lk->dev.evbit[0] = BIT (EV_KEY) | BIT (EV_LED) | BIT (EV_SND) | BIT (EV_REP);
+ lk->dev.ledbit[0] = BIT (LED_CAPSL) | BIT (LED_COMPOSE) | BIT (LED_SCROLLL) | BIT (LED_SLEEP);
+ lk->dev.sndbit[0] = BIT (SND_CLICK) | BIT (SND_BELL);
+
+ lk->serio = serio;
+
+ INIT_WORK (&lk->tq, lkkbd_reinit, lk);
+
+ lk->bell_volume = bell_volume;
+ lk->keyclick_volume = keyclick_volume;
+ lk->ctrlclick_volume = ctrlclick_volume;
+
+ lk->dev.keycode = lk->keycode;
+ lk->dev.keycodesize = sizeof (lk_keycode_t);
+ lk->dev.keycodemax = LK_NUM_KEYCODES;
+
+ lk->dev.event = lkkbd_event;
+ lk->dev.private = lk;
+
+ serio->private = lk;
+
+ if (serio_open (serio, dev)) {
+ kfree (lk);
+ return;
+ }
+
+ sprintf (lk->name, "LK keyboard");
+
+ memcpy (lk->keycode, lkkbd_keycode, sizeof (lk_keycode_t) * LK_NUM_KEYCODES);
+ for (i = 0; i < LK_NUM_KEYCODES; i++)
+ set_bit (lk->keycode[i], lk->dev.keybit);
+
+ sprintf (lk->name, "%s/input0", serio->phys);
+
+ lk->dev.name = lk->name;
+ lk->dev.phys = lk->phys;
+ lk->dev.id.bustype = BUS_RS232;
+ lk->dev.id.vendor = SERIO_LKKBD;
+ lk->dev.id.product = 0;
+ lk->dev.id.version = 0x0100;
+
+ input_register_device (&lk->dev);
+
+ printk (KERN_INFO "input: %s on %s, initiating reset\n", lk->name, serio->phys);
+ lk->serio->write (lk->serio, LK_CMD_POWERCYCLE_RESET);
+}
+
+/*
+ * lkkbd_disconnect() unregisters and closes behind us.
+ */
+static void
+lkkbd_disconnect (struct serio *serio)
+{
+ struct lkkbd *lk = serio->private;
+
+ input_unregister_device (&lk->dev);
+ serio_close (serio);
+ kfree (lk);
+}
+
+static struct serio_dev lkkbd_dev = {
+ .interrupt = lkkbd_interrupt,
+ .connect = lkkbd_connect,
+ .disconnect = lkkbd_disconnect,
+};
+
+/*
+ * The functions for insering/removing us as a module.
+ */
+int __init
+lkkbd_init (void)
+{
+ serio_register_device (&lkkbd_dev);
+ return 0;
+}
+
+void __exit
+lkkbd_exit (void)
+{
+ serio_unregister_device (&lkkbd_dev);
+}
+
+module_init (lkkbd_init);
+module_exit (lkkbd_exit);
+
--- diff/drivers/input/mouse/vsxxxaa.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/input/mouse/vsxxxaa.c 2004-03-16 11:27:48.155866856 +0000
@@ -0,0 +1,550 @@
+/*
+ * DEC VSXXX-AA and VSXXX-GA mouse driver.
+ *
+ * Copyright (C) 2003-2004 by Jan-Benedict Glaw <jbglaw@lug-owl.de>
+ *
+ * The packet format was taken from a patch to GPM which is (C) 2001
+ * by Karsten Merker <merker@linuxtag.org>
+ * and Maciej W. Rozycki <macro@ds2.pg.gda.pl>
+ */
+
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+/*
+ * Building an adaptor to DB9 / DB25 RS232
+ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+ *
+ * DISCLAIMER: Use this description AT YOUR OWN RISK! I'll not pay for
+ * anything if you break your mouse, your computer or whatever!
+ *
+ * In theory, this mouse is a simple RS232 device. In practice, it has got
+ * a quite uncommon plug and the requirement to additionally get a power
+ * supply at +5V and -12V.
+ *
+ * If you look at the socket/jack (_not_ at the plug), we use this pin
+ * numbering:
+ * _______
+ * / 7 6 5 \
+ * | 4 --- 3 |
+ * \ 2 1 /
+ * -------
+ *
+ * DEC socket DB9 DB25 Note
+ * 1 (GND) 5 7 -
+ * 2 (RxD) 3 3 -
+ * 3 (TxD) 2 2 -
+ * 4 (-12V) - - Somewhere from the PSU. At ATX, it's
+ * the blue wire at pin 12 of the ATX
+ * power connector. Please note that the
+ * docs say this should be +12V! However,
+ * I measured -12V...
+ * 5 (+5V) - - PSU (red wire of ATX power connector
+ * on pin 4, 6, 19 or 20) or HDD power
+ * connector (also red wire)
+ * 6 (not conn.) - - -
+ * 7 (dev. avail.) - - The mouse shorts this one to pin 1.
+ * This way, the host computer can detect
+ * the mouse. To use it with the adaptor,
+ * simply don't connect this pin.
+ *
+ * So to get a working adaptor, you need to connect the mouse with three
+ * wires to a RS232 port and two additional wires for +5V and -12V to the
+ * PSU.
+ *
+ * Flow specification for the link is 4800, 8o1.
+ */
+
+/*
+ * TODO list:
+ * - Automatically attach to a given serial port (no need for inputattach).
+ */
+
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/interrupt.h>
+#include <linux/input.h>
+#include <linux/config.h>
+#include <linux/serio.h>
+#include <linux/init.h>
+
+MODULE_AUTHOR ("Jan-Benedict Glaw <jbglaw@lug-owl.de>");
+MODULE_DESCRIPTION ("Serial DEC VSXXX-AA/GA mouse / DEC tablet driver");
+MODULE_LICENSE ("GPL");
+
+#undef VSXXXAA_DEBUG
+#ifdef VSXXXAA_DEBUG
+#define DBG(x...) printk (x)
+#else
+#define DBG(x...) do {} while (0)
+#endif
+
+#define VSXXXAA_INTRO_MASK 0x80
+#define VSXXXAA_INTRO_HEAD 0x80
+#define IS_HDR_BYTE(x) (((x) & VSXXXAA_INTRO_MASK) \
+ == VSXXXAA_INTRO_HEAD)
+
+#define VSXXXAA_PACKET_MASK 0xe0
+#define VSXXXAA_PACKET_REL 0x80
+#define VSXXXAA_PACKET_ABS 0xc0
+#define VSXXXAA_PACKET_POR 0xa0
+#define MATCH_PACKET_TYPE(data, type) (((data) & VSXXXAA_PACKET_MASK) == type)
+
+
+
+struct vsxxxaa {
+ struct input_dev dev;
+ struct serio *serio;
+#define BUFLEN 15 /* At least 5 is needed for a full tablet packet */
+ unsigned char buf[BUFLEN];
+ unsigned char count;
+ unsigned char version;
+ unsigned char country;
+ unsigned char type;
+ char phys[32];
+};
+
+static void
+vsxxxaa_drop_bytes (struct vsxxxaa *mouse, int num)
+{
+ if (num >= mouse->count)
+ mouse->count = 0;
+ else {
+ memmove (mouse->buf, mouse->buf + num - 1, BUFLEN - num);
+ mouse->count -= num;
+ }
+}
+
+static void
+vsxxxaa_queue_byte (struct vsxxxaa *mouse, unsigned char byte)
+{
+ if (mouse->count == BUFLEN) {
+ printk (KERN_ERR "%s on %s: Dropping a byte of full buffer.\n",
+ mouse->dev.name, mouse->dev.phys);
+ vsxxxaa_drop_bytes (mouse, 1);
+ }
+
+ mouse->buf[mouse->count++] = byte;
+}
+
+static void
+vsxxxaa_report_mouse (struct vsxxxaa *mouse)
+{
+ char *devtype;
+
+ switch (mouse->type) {
+ case 0x02: devtype = "DEC mouse"; break;
+ case 0x04: devtype = "DEC tablet"; break;
+ default: devtype = "unknown DEC device"; break;
+ }
+
+ printk (KERN_INFO "Found %s version 0x%x from country 0x%x "
+ "on port %s\n", devtype, mouse->version,
+ mouse->country, mouse->dev.phys);
+}
+
+/*
+ * Returns number of bytes to be dropped, 0 if packet is okay.
+ */
+static int
+vsxxxaa_check_packet (struct vsxxxaa *mouse, int packet_len)
+{
+ int i;
+
+ /* First byte must be a header byte */
+ if (!IS_HDR_BYTE (mouse->buf[0])) {
+ DBG ("vsck: len=%d, 1st=0x%02x\n", packet_len, mouse->buf[0]);
+ return 1;
+ }
+
+ /* Check all following bytes */
+ if (packet_len > 1) {
+ for (i = 1; i < packet_len; i++) {
+ if (IS_HDR_BYTE (mouse->buf[i])) {
+ printk (KERN_ERR "Need to drop %d bytes "
+ "of a broken packet.\n",
+ i - 1);
+ DBG (KERN_INFO "check: len=%d, b[%d]=0x%02x\n",
+ packet_len, i, mouse->buf[i]);
+ return i - 1;
+ }
+ }
+ }
+
+ return 0;
+}
+
+static __inline__ int
+vsxxxaa_smells_like_packet (struct vsxxxaa *mouse, unsigned char type, size_t len)
+{
+ return (mouse->count >= len) && MATCH_PACKET_TYPE (mouse->buf[0], type);
+}
+
+static void
+vsxxxaa_handle_REL_packet (struct vsxxxaa *mouse, struct pt_regs *regs)
+{
+ struct input_dev *dev = &mouse->dev;
+ unsigned char *buf = mouse->buf;
+ int left, middle, right;
+ int dx, dy;
+
+ /*
+ * Check for normal stream packets. This is three bytes,
+ * with the first byte's 3 MSB set to 100.
+ *
+ * [0]: 1 0 0 SignX SignY Left Middle Right
+ * [1]: 0 dx dx dx dx dx dx dx
+ * [2]: 0 dy dy dy dy dy dy dy
+ */
+
+ /*
+ * Low 7 bit of byte 1 are abs(dx), bit 7 is
+ * 0, bit 4 of byte 0 is direction.
+ */
+ dx = buf[1] & 0x7f;
+ dx *= ((buf[0] >> 4) & 0x01)? -1: 1;
+
+ /*
+ * Low 7 bit of byte 2 are abs(dy), bit 7 is
+ * 0, bit 3 of byte 0 is direction.
+ */
+ dy = buf[2] & 0x7f;
+ dy *= ((buf[0] >> 3) & 0x01)? -1: 1;
+
+ /*
+ * Get button state. It's the low three bits
+ * (for three buttons) of byte 0.
+ */
+ left = (buf[0] & 0x04)? 1: 0;
+ middle = (buf[0] & 0x02)? 1: 0;
+ right = (buf[0] & 0x01)? 1: 0;
+
+ vsxxxaa_drop_bytes (mouse, 3);
+
+ DBG (KERN_INFO "%s on %s: dx=%d, dy=%d, buttons=%s%s%s\n",
+ mouse->dev.name, mouse->dev.phys, dx, dy,
+ left? "L": "l", middle? "M": "m", right? "R": "r");
+
+ /*
+ * Report what we've found so far...
+ */
+ input_regs (dev, regs);
+ input_report_key (dev, BTN_LEFT, left);
+ input_report_key (dev, BTN_MIDDLE, middle);
+ input_report_key (dev, BTN_RIGHT, right);
+ input_report_rel (dev, REL_X, dx);
+ input_report_rel (dev, REL_Y, dy);
+ input_sync (dev);
+}
+
+static void
+vsxxxaa_handle_ABS_packet (struct vsxxxaa *mouse, struct pt_regs *regs)
+{
+ struct input_dev *dev = &mouse->dev;
+ unsigned char *buf = mouse->buf;
+ int left, middle, right, extra;
+ int x, y;
+
+ /*
+ * Tablet position / button packet
+ *
+ * [0]: 1 1 0 B4 B3 B2 B1 Pr
+ * [1]: 0 0 X5 X4 X3 X2 X1 X0
+ * [2]: 0 0 X11 X10 X9 X8 X7 X6
+ * [3]: 0 0 Y5 Y4 Y3 Y2 Y1 Y0
+ * [4]: 0 0 Y11 Y10 Y9 Y8 Y7 Y6
+ */
+
+ /*
+ * Get X/Y position
+ */
+ x = ((buf[2] & 0x3f) << 6) | (buf[1] & 0x3f);
+ y = ((buf[4] & 0x3f) << 6) | (buf[3] & 0x3f);
+
+ /*
+ * Get button state. It's bits <4..1> of byte 0.
+ */
+ left = (buf[0] & 0x02)? 1: 0;
+ middle = (buf[0] & 0x04)? 1: 0;
+ right = (buf[0] & 0x08)? 1: 0;
+ extra = (buf[0] & 0x10)? 1: 0;
+
+ vsxxxaa_drop_bytes (mouse, 5);
+
+ DBG (KERN_INFO "%s on %s: x=%d, y=%d, buttons=%s%s%s%s\n",
+ mouse->dev.name, mouse->dev.phys, x, y,
+ left? "L": "l", middle? "M": "m",
+ right? "R": "r", extra? "E": "e");
+
+ /*
+ * Report what we've found so far...
+ */
+ input_regs (dev, regs);
+ input_report_key (dev, BTN_LEFT, left);
+ input_report_key (dev, BTN_MIDDLE, middle);
+ input_report_key (dev, BTN_RIGHT, right);
+ input_report_key (dev, BTN_EXTRA, extra);
+ input_report_abs (dev, ABS_X, x);
+ input_report_abs (dev, ABS_Y, y);
+ input_sync (dev);
+}
+
+static void
+vsxxxaa_handle_POR_packet (struct vsxxxaa *mouse, struct pt_regs *regs)
+{
+ struct input_dev *dev = &mouse->dev;
+ unsigned char *buf = mouse->buf;
+ int left, middle, right;
+ unsigned char error;
+
+ /*
+ * Check for Power-On-Reset packets. These are sent out
+ * after plugging the mouse in, or when explicitely
+ * requested by sending 'T'.
+ *
+ * [0]: 1 0 1 0 R3 R2 R1 R0
+ * [1]: 0 M2 M1 M0 D3 D2 D1 D0
+ * [2]: 0 E6 E5 E4 E3 E2 E1 E0
+ * [3]: 0 0 0 0 0 Left Middle Right
+ *
+ * M: manufacturer location code
+ * R: revision code
+ * E: Error code. I'm not sure about these, but gpm's sources,
+ * which support this mouse, too, tell about them:
+ * E = [0x00 .. 0x1f]: no error, byte #3 is button state
+ * E = 0x3d: button error, byte #3 tells which one.
+ * E = <else>: other error
+ * D: <0010> == mouse, <0100> == tablet
+ *
+ */
+
+ mouse->version = buf[0] & 0x0f;
+ mouse->country = (buf[1] >> 4) & 0x07;
+ mouse->type = buf[1] & 0x07;
+ error = buf[2] & 0x7f;
+
+ /*
+ * Get button state. It's the low three bits
+ * (for three buttons) of byte 0. Maybe even the bit <3>
+ * has some meaning if a tablet is attached.
+ */
+ left = (buf[0] & 0x04)? 1: 0;
+ middle = (buf[0] & 0x02)? 1: 0;
+ right = (buf[0] & 0x01)? 1: 0;
+
+ vsxxxaa_drop_bytes (mouse, 4);
+ vsxxxaa_report_mouse (mouse);
+
+ if (error <= 0x1f) {
+ /* No error. Report buttons */
+ input_regs (dev, regs);
+ input_report_key (dev, BTN_LEFT, left);
+ input_report_key (dev, BTN_MIDDLE, middle);
+ input_report_key (dev, BTN_RIGHT, right);
+ input_sync (dev);
+ } else {
+ printk (KERN_ERR "Your %s on %s reports an undefined error, "
+ "please check it...\n", mouse->dev.name,
+ mouse->dev.phys);
+ }
+
+ /*
+ * If the mouse was hot-plugged, we need to
+ * force differential mode now...
+ */
+ printk (KERN_NOTICE "%s on %s: Forceing standard packet format and "
+ "streaming mode\n", mouse->dev.name, mouse->dev.phys);
+ mouse->serio->write (mouse->serio, 'S');
+ mouse->serio->write (mouse->serio, 'R');
+}
+
+static void
+vsxxxaa_parse_buffer (struct vsxxxaa *mouse, struct pt_regs *regs)
+{
+ unsigned char *buf = mouse->buf;
+ int stray_bytes;
+
+ /*
+ * Parse buffer to death...
+ */
+ do {
+ /*
+ * Out of sync? Throw away what we don't understand. Each
+ * packet starts with a byte whose bit 7 is set. Unhandled
+ * packets (ie. which we don't know about or simply b0rk3d
+ * data...) will get shifted out of the buffer after some
+ * activity on the mouse.
+ */
+ while (mouse->count > 0 && !IS_HDR_BYTE(buf[0])) {
+ printk (KERN_ERR "%s on %s: Dropping a byte to regain "
+ "sync with mouse data stream...\n",
+ mouse->dev.name, mouse->dev.phys);
+ vsxxxaa_drop_bytes (mouse, 1);
+ }
+
+ /*
+ * Check for packets we know about.
+ */
+
+ if (vsxxxaa_smells_like_packet (mouse, VSXXXAA_PACKET_REL, 3)) {
+ /* Check for broken packet */
+ stray_bytes = vsxxxaa_check_packet (mouse, 3);
+ if (stray_bytes > 0) {
+ printk (KERN_ERR "Dropping %d bytes now...\n",
+ stray_bytes);
+ vsxxxaa_drop_bytes (mouse, stray_bytes);
+ continue;
+ }
+
+ vsxxxaa_handle_REL_packet (mouse, regs);
+ continue; /* More to parse? */
+ }
+
+ if (vsxxxaa_smells_like_packet (mouse, VSXXXAA_PACKET_ABS, 5)) {
+ /* Check for broken packet */
+ stray_bytes = vsxxxaa_check_packet (mouse, 5);
+ if (stray_bytes > 0) {
+ printk (KERN_ERR "Dropping %d bytes now...\n",
+ stray_bytes);
+ vsxxxaa_drop_bytes (mouse, stray_bytes);
+ continue;
+ }
+
+ vsxxxaa_handle_ABS_packet (mouse, regs);
+ continue; /* More to parse? */
+ }
+
+ if (vsxxxaa_smells_like_packet (mouse, VSXXXAA_PACKET_POR, 4)) {
+ /* Check for broken packet */
+ stray_bytes = vsxxxaa_check_packet (mouse, 4);
+ if (stray_bytes > 0) {
+ printk (KERN_ERR "Dropping %d bytes now...\n",
+ stray_bytes);
+ vsxxxaa_drop_bytes (mouse, stray_bytes);
+ continue;
+ }
+
+ vsxxxaa_handle_POR_packet (mouse, regs);
+ continue; /* More to parse? */
+ }
+
+ break; /* No REL, ABS or POR packet found */
+ } while (1);
+}
+
+static irqreturn_t
+vsxxxaa_interrupt (struct serio *serio, unsigned char data, unsigned int flags,
+ struct pt_regs *regs)
+{
+ struct vsxxxaa *mouse = serio->private;
+
+ vsxxxaa_queue_byte (mouse, data);
+ vsxxxaa_parse_buffer (mouse, regs);
+
+ return IRQ_HANDLED;
+}
+
+static void
+vsxxxaa_disconnect (struct serio *serio)
+{
+ struct vsxxxaa *mouse = serio->private;
+
+ input_unregister_device (&mouse->dev);
+ serio_close (serio);
+ kfree (mouse);
+}
+
+static void
+vsxxxaa_connect (struct serio *serio, struct serio_dev *dev)
+{
+ struct vsxxxaa *mouse;
+
+ if ((serio->type & SERIO_TYPE) != SERIO_RS232)
+ return;
+
+ if ((serio->type & SERIO_PROTO) != SERIO_VSXXXAA)
+ return;
+
+ if (!(mouse = kmalloc (sizeof (struct vsxxxaa), GFP_KERNEL)))
+ return;
+
+ memset (mouse, 0, sizeof (struct vsxxxaa));
+
+ init_input_dev (&mouse->dev);
+ set_bit (EV_KEY, mouse->dev.evbit); /* We have buttons */
+ set_bit (EV_REL, mouse->dev.evbit); /* We can move */
+ set_bit (BTN_LEFT, mouse->dev.keybit); /* We have 3 buttons */
+ set_bit (BTN_MIDDLE, mouse->dev.keybit);
+ set_bit (BTN_RIGHT, mouse->dev.keybit);
+ set_bit (BTN_EXTRA, mouse->dev.keybit); /* ...and Tablet */
+ set_bit (REL_X, mouse->dev.relbit); /* We can move in */
+ set_bit (REL_Y, mouse->dev.relbit); /* two dimensions */
+ set_bit (ABS_X, mouse->dev.absbit); /* DEC tablet support */
+ set_bit (ABS_Y, mouse->dev.absbit);
+
+ mouse->dev.absmin[ABS_X] = 0;
+ mouse->dev.absmax[ABS_X] = 1023;
+ mouse->dev.absmin[ABS_Y] = 0;
+ mouse->dev.absmax[ABS_Y] = 1023;
+
+ mouse->dev.private = mouse;
+ serio->private = mouse;
+
+ sprintf (mouse->phys, "%s/input0", serio->phys);
+ mouse->dev.phys = mouse->phys;
+ mouse->dev.name = "DEC VSXXX-AA/GA mouse or DEC tablet";
+ mouse->dev.id.bustype = BUS_RS232;
+ mouse->serio = serio;
+
+ if (serio_open (serio, dev)) {
+ kfree (mouse);
+ return;
+ }
+
+ /*
+ * Request selftest and differential stream mode.
+ */
+ mouse->serio->write (mouse->serio, 'T'); /* Test */
+ mouse->serio->write (mouse->serio, 'R'); /* Differential stream */
+
+ input_register_device (&mouse->dev);
+
+ printk (KERN_INFO "input: %s on %s\n", mouse->dev.name, serio->phys);
+}
+
+static struct serio_dev vsxxxaa_dev = {
+ .interrupt = vsxxxaa_interrupt,
+ .connect = vsxxxaa_connect,
+ .disconnect = vsxxxaa_disconnect
+};
+
+int __init
+vsxxxaa_init (void)
+{
+ serio_register_device (&vsxxxaa_dev);
+ return 0;
+}
+
+void __exit
+vsxxxaa_exit (void)
+{
+ serio_unregister_device (&vsxxxaa_dev);
+}
+
+module_init (vsxxxaa_init);
+module_exit (vsxxxaa_exit);
+
--- diff/drivers/input/serio/gscps2.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/input/serio/gscps2.c 2004-03-16 11:27:48.156866704 +0000
@@ -0,0 +1,470 @@
+/*
+ * drivers/input/serio/gscps2.c
+ *
+ * Copyright (c) 2004 Helge Deller <deller@gmx.de>
+ * Copyright (c) 2002 Laurent Canet <canetl@esiee.fr>
+ * Copyright (c) 2002 Thibaut Varene <varenet@esiee.fr>
+ *
+ * Pieces of code based on linux-2.4's hp_mouse.c & hp_keyb.c
+ * Copyright (c) 1999 Alex deVries <adevries@thepuffingroup.com>
+ * Copyright (c) 1999-2000 Philipp Rumpf <prumpf@tux.org>
+ * Copyright (c) 2000 Xavier Debacker <debackex@esiee.fr>
+ * Copyright (c) 2000-2001 Thomas Marteau <marteaut@esiee.fr>
+ *
+ * HP GSC PS/2 port driver, found in PA/RISC Workstations
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * TODO:
+ * - Dino testing (did HP ever shipped a machine on which this port
+ * was usable/enabled ?)
+ */
+
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/serio.h>
+#include <linux/input.h>
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/delay.h>
+#include <linux/ioport.h>
+#include <linux/pci_ids.h>
+
+#include <asm/irq.h>
+#include <asm/io.h>
+#include <asm/parisc-device.h>
+
+MODULE_AUTHOR("Laurent Canet <canetl@esiee.fr>, Thibaut Varene <varenet@esiee.fr>, Helge Deller <deller@gmx.de>");
+MODULE_DESCRIPTION("HP GSC PS/2 port driver");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(parisc, gscps2_device_tbl);
+
+#define PFX "gscps2.c: "
+
+/*
+ * Driver constants
+ */
+
+/* various constants */
+#define ENABLE 1
+#define DISABLE 0
+
+#define GSC_DINO_OFFSET 0x0800 /* offset for DINO controller versus LASI one */
+
+/* PS/2 IO port offsets */
+#define GSC_ID 0x00 /* device ID offset (see: GSC_ID_XXX) */
+#define GSC_RESET 0x00 /* reset port offset */
+#define GSC_RCVDATA 0x04 /* receive port offset */
+#define GSC_XMTDATA 0x04 /* transmit port offset */
+#define GSC_CONTROL 0x08 /* see: Control register bits */
+#define GSC_STATUS 0x0C /* see: Status register bits */
+
+/* Control register bits */
+#define GSC_CTRL_ENBL 0x01 /* enable interface */
+#define GSC_CTRL_LPBXR 0x02 /* loopback operation */
+#define GSC_CTRL_DIAG 0x20 /* directly control clock/data line */
+#define GSC_CTRL_DATDIR 0x40 /* data line direct control */
+#define GSC_CTRL_CLKDIR 0x80 /* clock line direct control */
+
+/* Status register bits */
+#define GSC_STAT_RBNE 0x01 /* Receive Buffer Not Empty */
+#define GSC_STAT_TBNE 0x02 /* Transmit Buffer Not Empty */
+#define GSC_STAT_TERR 0x04 /* Timeout Error */
+#define GSC_STAT_PERR 0x08 /* Parity Error */
+#define GSC_STAT_CMPINTR 0x10 /* Composite Interrupt = irq on any port */
+#define GSC_STAT_DATSHD 0x40 /* Data Line Shadow */
+#define GSC_STAT_CLKSHD 0x80 /* Clock Line Shadow */
+
+/* IDs returned by GSC_ID port register */
+#define GSC_ID_KEYBOARD 0 /* device ID values */
+#define GSC_ID_MOUSE 1
+
+
+static irqreturn_t gscps2_interrupt(int irq, void *dev, struct pt_regs *regs);
+
+#define BUFFER_SIZE 0x0f
+
+/* GSC PS/2 port device struct */
+struct gscps2port {
+ struct list_head node;
+ struct parisc_device *padev;
+ struct serio port;
+ spinlock_t lock;
+ char *addr;
+ u8 act, append; /* position in buffer[] */
+ struct {
+ u8 data;
+ u8 str;
+ } buffer[BUFFER_SIZE+1];
+ int id;
+ char name[32];
+};
+
+/*
+ * Various HW level routines
+ */
+
+#define gscps2_readb_input(x) readb((x)+GSC_RCVDATA)
+#define gscps2_readb_control(x) readb((x)+GSC_CONTROL)
+#define gscps2_readb_status(x) readb((x)+GSC_STATUS)
+#define gscps2_writeb_control(x, y) writeb((x), (y)+GSC_CONTROL)
+
+
+/*
+ * wait_TBE() - wait for Transmit Buffer Empty
+ */
+
+static int wait_TBE(char *addr)
+{
+ int timeout = 25000; /* device is expected to react within 250 msec */
+ while (gscps2_readb_status(addr) & GSC_STAT_TBNE) {
+ if (!--timeout)
+ return 0; /* This should not happen */
+ udelay(10);
+ }
+ return 1;
+}
+
+
+/*
+ * gscps2_flush() - flush the receive buffer
+ */
+
+static void gscps2_flush(struct gscps2port *ps2port)
+{
+ while (gscps2_readb_status(ps2port->addr) & GSC_STAT_RBNE)
+ gscps2_readb_input(ps2port->addr);
+ ps2port->act = ps2port->append = 0;
+}
+
+/*
+ * gscps2_writeb_output() - write a byte to the port
+ *
+ * returns 1 on sucess, 0 on error
+ */
+
+static inline int gscps2_writeb_output(struct gscps2port *ps2port, u8 data)
+{
+ unsigned long flags;
+ char *addr = ps2port->addr;
+
+ if (!wait_TBE(addr)) {
+ printk(KERN_DEBUG PFX "timeout - could not write byte %#x\n", data);
+ return 0;
+ }
+
+ while (gscps2_readb_status(ps2port->addr) & GSC_STAT_RBNE)
+ /* wait */;
+
+ spin_lock_irqsave(&ps2port->lock, flags);
+ writeb(data, addr+GSC_XMTDATA);
+ spin_unlock_irqrestore(&ps2port->lock, flags);
+
+ /* this is ugly, but due to timing of the port it seems to be necessary. */
+ mdelay(6);
+
+ /* make sure any received data is returned as fast as possible */
+ /* this is important e.g. when we set the LEDs on the keyboard */
+ gscps2_interrupt(0, NULL, NULL);
+
+ return 1;
+}
+
+
+/*
+ * gscps2_enable() - enables or disables the port
+ */
+
+static void gscps2_enable(struct gscps2port *ps2port, int enable)
+{
+ unsigned long flags;
+ u8 data;
+
+ /* now enable/disable the port */
+ spin_lock_irqsave(&ps2port->lock, flags);
+ gscps2_flush(ps2port);
+ data = gscps2_readb_control(ps2port->addr);
+ if (enable)
+ data |= GSC_CTRL_ENBL;
+ else
+ data &= ~GSC_CTRL_ENBL;
+ gscps2_writeb_control(data, ps2port->addr);
+ spin_unlock_irqrestore(&ps2port->lock, flags);
+ wait_TBE(ps2port->addr);
+ gscps2_flush(ps2port);
+}
+
+/*
+ * gscps2_reset() - resets the PS/2 port
+ */
+
+static void gscps2_reset(struct gscps2port *ps2port)
+{
+ char *addr = ps2port->addr;
+ unsigned long flags;
+
+ /* reset the interface */
+ spin_lock_irqsave(&ps2port->lock, flags);
+ gscps2_flush(ps2port);
+ writeb(0xff, addr+GSC_RESET);
+ gscps2_flush(ps2port);
+ spin_unlock_irqrestore(&ps2port->lock, flags);
+
+ /* enable it */
+ gscps2_enable(ps2port, ENABLE);
+}
+
+static LIST_HEAD(ps2port_list);
+
+/**
+ * gscps2_interrupt() - Interruption service routine
+ *
+ * This function reads received PS/2 bytes and processes them on
+ * all interfaces.
+ * The problematic part here is, that the keyboard and mouse PS/2 port
+ * share the same interrupt and it's not possible to send data if any
+ * one of them holds input data. To solve this problem we try to receive
+ * the data as fast as possible and handle the reporting to the upper layer
+ * later.
+ */
+
+static irqreturn_t gscps2_interrupt(int irq, void *dev, struct pt_regs *regs)
+{
+ struct gscps2port *ps2port;
+
+ list_for_each_entry(ps2port, &ps2port_list, node) {
+
+ unsigned long flags;
+ spin_lock_irqsave(&ps2port->lock, flags);
+
+ while ( (ps2port->buffer[ps2port->append].str =
+ gscps2_readb_status(ps2port->addr)) & GSC_STAT_RBNE ) {
+ ps2port->buffer[ps2port->append].data =
+ gscps2_readb_input(ps2port->addr);
+ ps2port->append = ((ps2port->append+1) & BUFFER_SIZE);
+ }
+
+ spin_unlock_irqrestore(&ps2port->lock, flags);
+
+ } /* list_for_each_entry */
+
+ /* all data was read from the ports - now report the data to upper layer */
+
+ list_for_each_entry(ps2port, &ps2port_list, node) {
+
+ while (ps2port->act != ps2port->append) {
+
+ unsigned int rxflags;
+ u8 data, status;
+
+ /* Did new data arrived while we read existing data ?
+ If yes, exit now and let the new irq handler start over again */
+ if (gscps2_readb_status(ps2port->addr) & GSC_STAT_CMPINTR)
+ return IRQ_HANDLED;
+
+ status = ps2port->buffer[ps2port->act].str;
+ data = ps2port->buffer[ps2port->act].data;
+
+ ps2port->act = ((ps2port->act+1) & BUFFER_SIZE);
+ rxflags = ((status & GSC_STAT_TERR) ? SERIO_TIMEOUT : 0 ) |
+ ((status & GSC_STAT_PERR) ? SERIO_PARITY : 0 );
+
+ serio_interrupt(&ps2port->port, data, rxflags, regs);
+
+ } /* while() */
+
+ } /* list_for_each_entry */
+
+ return IRQ_HANDLED;
+}
+
+
+/*
+ * gscps2_write() - send a byte out through the aux interface.
+ */
+
+static int gscps2_write(struct serio *port, unsigned char data)
+{
+ struct gscps2port *ps2port = port->driver;
+
+ if (!gscps2_writeb_output(ps2port, data)) {
+ printk(KERN_DEBUG PFX "sending byte %#x failed.\n", data);
+ return -1;
+ }
+ return 0;
+}
+
+/*
+ * gscps2_open() is called when a port is opened by the higher layer.
+ * It resets and enables the port.
+ */
+
+static int gscps2_open(struct serio *port)
+{
+ struct gscps2port *ps2port = port->driver;
+
+ gscps2_reset(ps2port);
+
+ gscps2_interrupt(0, NULL, NULL);
+
+ return 0;
+}
+
+/*
+ * gscps2_close() disables the port
+ */
+
+static void gscps2_close(struct serio *port)
+{
+ struct gscps2port *ps2port = port->driver;
+ gscps2_enable(ps2port, DISABLE);
+}
+
+static struct serio gscps2_serio_port =
+{
+ .name = "GSC PS/2",
+ .idbus = BUS_GSC,
+ .idvendor = PCI_VENDOR_ID_HP,
+ .idproduct = 0x0001,
+ .idversion = 0x0010,
+ .type = SERIO_8042,
+ .write = gscps2_write,
+ .open = gscps2_open,
+ .close = gscps2_close,
+};
+
+/**
+ * gscps2_probe() - Probes PS2 devices
+ * @return: success/error report
+ */
+
+static int __init gscps2_probe(struct parisc_device *dev)
+{
+ struct gscps2port *ps2port;
+ unsigned long hpa = dev->hpa;
+ int ret;
+
+ if (!dev->irq)
+ return -ENODEV;
+
+ /* Offset for DINO PS/2. Works with LASI even */
+ if (dev->id.sversion == 0x96)
+ hpa += GSC_DINO_OFFSET;
+
+ ps2port = kmalloc(sizeof(struct gscps2port), GFP_KERNEL);
+ if (!ps2port)
+ return -ENOMEM;
+
+ dev_set_drvdata(&dev->dev, ps2port);
+
+ memset(ps2port, 0, sizeof(struct gscps2port));
+ ps2port->padev = dev;
+ ps2port->addr = ioremap(hpa, GSC_STATUS + 4);
+ spin_lock_init(&ps2port->lock);
+
+ gscps2_reset(ps2port);
+ ps2port->id = readb(ps2port->addr+GSC_ID) & 0x0f;
+ snprintf(ps2port->name, sizeof(ps2port->name)-1, "%s %s",
+ gscps2_serio_port.name,
+ (ps2port->id == GSC_ID_KEYBOARD) ? "keyboard" : "mouse" );
+
+ memcpy(&ps2port->port, &gscps2_serio_port, sizeof(gscps2_serio_port));
+ ps2port->port.driver = ps2port;
+ ps2port->port.name = ps2port->name;
+ ps2port->port.phys = dev->dev.bus_id;
+
+ list_add_tail(&ps2port->node, &ps2port_list);
+
+ ret = -EBUSY;
+ if (request_irq(dev->irq, gscps2_interrupt, SA_SHIRQ, ps2port->name, ps2port))
+ goto fail_miserably;
+
+ if ( (ps2port->id != GSC_ID_KEYBOARD) && (ps2port->id != GSC_ID_MOUSE) ) {
+ printk(KERN_WARNING PFX "Unsupported PS/2 port at 0x%08lx (id=%d) ignored\n",
+ hpa, ps2port->id);
+ ret = -ENODEV;
+ goto fail;
+ }
+
+#if 0
+ if (!request_mem_region(hpa, GSC_STATUS + 4, ps2port->port.name))
+ goto fail;
+#endif
+
+ printk(KERN_INFO "serio: %s port at 0x%p irq %d @ %s\n",
+ ps2port->name,
+ ps2port->addr,
+ ps2port->padev->irq,
+ ps2port->port.phys);
+
+ serio_register_port(&ps2port->port);
+
+ return 0;
+
+fail:
+ free_irq(dev->irq, ps2port);
+
+fail_miserably:
+ list_del(&ps2port->node);
+ iounmap(ps2port->addr);
+ release_mem_region(dev->hpa, GSC_STATUS + 4);
+ kfree(ps2port);
+ return ret;
+}
+
+/**
+ * gscps2_remove() - Removes PS2 devices
+ * @return: success/error report
+ */
+
+static int __devexit gscps2_remove(struct parisc_device *dev)
+{
+ struct gscps2port *ps2port = dev_get_drvdata(&dev->dev);
+
+ serio_unregister_port(&ps2port->port);
+ free_irq(dev->irq, ps2port);
+ gscps2_flush(ps2port);
+ list_del(&ps2port->node);
+ iounmap(ps2port->addr);
+#if 0
+ release_mem_region(dev->hpa, GSC_STATUS + 4);
+#endif
+ dev_set_drvdata(&dev->dev, NULL);
+ kfree(ps2port);
+ return 0;
+}
+
+
+static struct parisc_device_id gscps2_device_tbl[] = {
+ { HPHW_FIO, HVERSION_REV_ANY_ID, HVERSION_ANY_ID, 0x00084 }, /* LASI PS/2 */
+#ifdef DINO_TESTED
+ { HPHW_FIO, HVERSION_REV_ANY_ID, HVERSION_ANY_ID, 0x00096 }, /* DINO PS/2 */
+#endif
+ { 0, } /* 0 terminated list */
+};
+
+static struct parisc_driver parisc_ps2_driver = {
+ .name = "GSC PS/2",
+ .id_table = gscps2_device_tbl,
+ .probe = gscps2_probe,
+ .remove = gscps2_remove,
+};
+
+static int __init gscps2_init(void)
+{
+ register_parisc_driver(&parisc_ps2_driver);
+ return 0;
+}
+
+static void __exit gscps2_exit(void)
+{
+ unregister_parisc_driver(&parisc_ps2_driver);
+}
+
+
+module_init(gscps2_init);
+module_exit(gscps2_exit);
+
--- diff/drivers/net/kgdb_eth.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/net/kgdb_eth.c 2004-03-16 11:27:48.189861688 +0000
@@ -0,0 +1,131 @@
+/*
+ * Network interface GDB stub
+ *
+ * Written by San Mehat (nettwerk@biodome.org)
+ * Based upon 'gdbserial' by David Grothe (dave@gcom.com)
+ * and Scott Foehner (sfoehner@engr.sgi.com)
+ *
+ * Twiddled for 2.6 by Robert Walsh <rjwalsh@durables.org>
+ * and wangdi <wangdi@clusterfs.com>.
+ *
+ * Refactored for netpoll API by Matt Mackall <mpm@selenic.com>
+ *
+ */
+
+#include <linux/sched.h>
+#include <linux/interrupt.h>
+#include <linux/config.h>
+#include <linux/string.h>
+#include <linux/netpoll.h>
+
+#include <asm/system.h>
+#include <asm/kgdb.h>
+#include <asm/io.h>
+#include <asm/bitops.h>
+#include <asm/system.h>
+#include <asm/irq.h>
+#include <asm/atomic.h>
+
+#define IN_BUF_SIZE 512 /* power of 2, please */
+#define OUT_BUF_SIZE 256
+
+static char in_buf[IN_BUF_SIZE], out_buf[OUT_BUF_SIZE];
+static int in_head, in_tail, out_count;
+static atomic_t in_count;
+int kgdboe = 0; /* Default to tty mode */
+
+extern void set_debug_traps(void);
+extern void breakpoint(void);
+static void rx_hook(struct netpoll *np, int port, char *msg, int len);
+
+static struct netpoll np = {
+ .name = "kgdboe",
+ .dev_name = "eth0",
+ .rx_hook = rx_hook,
+ .local_port = 6443,
+ .remote_port = 6442,
+ .remote_mac = {0xff, 0xff, 0xff, 0xff, 0xff, 0xff},
+};
+
+int eth_getDebugChar(void)
+{
+ int chr;
+
+ while (atomic_read(&in_count) == 0)
+ netpoll_poll(&np);
+
+ chr = in_buf[in_tail++];
+ in_tail &= (IN_BUF_SIZE - 1);
+ atomic_dec(&in_count);
+ return chr;
+}
+
+void eth_flushDebugChar(void)
+{
+ if(out_count && np.dev) {
+ netpoll_send_udp(&np, out_buf, out_count);
+ out_count = 0;
+ }
+}
+
+void eth_putDebugChar(int chr)
+{
+ out_buf[out_count++] = chr;
+ if(out_count == OUT_BUF_SIZE)
+ eth_flushDebugChar();
+}
+
+static void rx_hook(struct netpoll *np, int port, char *msg, int len)
+{
+ int i;
+
+ np->remote_port = port;
+
+ /* Is this gdb trying to attach? */
+ if (!netpoll_trap() && len == 8 && !strncmp(msg, "$Hc-1#09", 8))
+ kgdb_schedule_breakpoint();
+
+ for (i = 0; i < len; i++) {
+ if (msg[i] == 3)
+ kgdb_schedule_breakpoint();
+
+ if (atomic_read(&in_count) >= IN_BUF_SIZE) {
+ /* buffer overflow, clear it */
+ in_head = in_tail = 0;
+ atomic_set(&in_count, 0);
+ break;
+ }
+ in_buf[in_head++] = msg[i];
+ in_head &= (IN_BUF_SIZE - 1);
+ atomic_inc(&in_count);
+ }
+}
+
+static int option_setup(char *opt)
+{
+ return netpoll_parse_options(&np, opt);
+}
+
+__setup("kgdboe=", option_setup);
+
+static int init_kgdboe(void)
+{
+#ifdef CONFIG_SMP
+ if (num_online_cpus() > CONFIG_NO_KGDB_CPUS) {
+ printk("kgdb: too manu cpus. Cannot enable debugger with more than %d cpus\n", CONFIG_NO_KGDB_CPUS);
+ return -1;
+ }
+#endif
+
+ set_debug_traps();
+
+ if(!np.remote_ip || netpoll_setup(&np))
+ return 1;
+
+ kgdboe = 1;
+ printk(KERN_INFO "kgdb: debugging over ethernet enabled\n");
+
+ return 0;
+}
+
+module_init(init_kgdboe);
--- diff/drivers/pci/hotplug/rpaphp_slot.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/pci/hotplug/rpaphp_slot.c 2004-03-16 11:27:48.223856520 +0000
@@ -0,0 +1,188 @@
+/*
+ * RPA Virtual I/O device functions
+ * Copyright (C) 2004 Linda Xie <lxie@us.ibm.com>
+ *
+ * All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or (at
+ * your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
+ * NON INFRINGEMENT. See the GNU General Public License for more
+ * details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ * Send feedback to <lxie@us.ibm.com>
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/kobject.h>
+#include <linux/sysfs.h>
+#include <linux/pci.h>
+#include "rpaphp.h"
+
+/* free up the memory user by a slot */
+
+static void rpaphp_release_slot(struct hotplug_slot *hotplug_slot)
+{
+ struct slot *slot = hotplug_slot? (struct slot *) hotplug_slot->private:NULL;
+
+ if (slot == NULL)
+ return;
+
+ dealloc_slot_struct(slot);
+}
+
+void dealloc_slot_struct(struct slot *slot)
+{
+ kfree(slot->hotplug_slot->info);
+ kfree(slot->hotplug_slot->name);
+ kfree(slot->hotplug_slot);
+ kfree(slot);
+ return;
+}
+
+struct slot *alloc_slot_struct(struct device_node *dn, int drc_index, char *drc_name,
+ int power_domain)
+{
+ struct slot *slot;
+
+ dbg("Enter alloc_slot_struct(): dn->full_name=%s drc_index=0x%x drc_name=%s\n",
+ dn->full_name, drc_index, drc_name);
+
+ slot = kmalloc(sizeof (struct slot), GFP_KERNEL);
+ if (!slot)
+ return (NULL);
+ memset(slot, 0, sizeof (struct slot));
+ slot->hotplug_slot = kmalloc(sizeof (struct hotplug_slot), GFP_KERNEL);
+ if (!slot->hotplug_slot) {
+ kfree(slot);
+ return (NULL);
+ }
+ memset(slot->hotplug_slot, 0, sizeof (struct hotplug_slot));
+ slot->hotplug_slot->info = kmalloc(sizeof (struct hotplug_slot_info),
+ GFP_KERNEL);
+ if (!slot->hotplug_slot->info) {
+ kfree(slot->hotplug_slot);
+ kfree(slot);
+ return (NULL);
+ }
+ memset(slot->hotplug_slot->info, 0, sizeof (struct hotplug_slot_info));
+ slot->hotplug_slot->name = kmalloc(strlen(drc_name) + 1, GFP_KERNEL);
+ if (!slot->hotplug_slot->name) {
+ kfree(slot->hotplug_slot->info);
+ kfree(slot->hotplug_slot);
+ kfree(slot);
+ return (NULL);
+ }
+ slot->name = slot->hotplug_slot->name;
+ slot->dn = dn;
+ slot->index = drc_index;
+ strcpy(slot->name, drc_name);
+ slot->power_domain = power_domain;
+ slot->magic = SLOT_MAGIC;
+ slot->hotplug_slot->private = slot;
+ slot->hotplug_slot->ops = &rpaphp_hotplug_slot_ops;
+ slot->hotplug_slot->release = &rpaphp_release_slot;
+ dbg("Exit alloc_slot_struct(): slot->dn->full_name=%s drc_index=0x%x drc_name=%s\n",
+ slot->dn->full_name, slot->index, slot->name);
+ return (slot);
+}
+
+int register_slot(struct slot *slot)
+{
+ int retval;
+ char *vio_uni_addr = NULL;
+
+ dbg("%s registering slot:path[%s] index[%x], name[%s] pdomain[%x] type[%d]\n", __FUNCTION__, slot->dn->full_name, slot->index, slot->name, slot->power_domain, slot->type);
+
+ retval = pci_hp_register(slot->hotplug_slot);
+ if (retval) {
+ err("pci_hp_register failed with error %d\n", retval);
+ rpaphp_release_slot(slot->hotplug_slot);
+ return (retval);
+ }
+ switch (slot->dev_type) {
+ case PCI_DEV:
+ /* create symlink between slot->name and it's bus_id */
+
+ dbg("%s: sysfs_create_link: %s --> %s\n", __FUNCTION__,
+ pci_name(slot->bridge), slot->name);
+
+ retval = sysfs_create_link(slot->hotplug_slot->kobj.parent,
+ &slot->hotplug_slot->kobj,
+ pci_name(slot->bridge));
+ if (retval) {
+ err("sysfs_create_link failed with error %d\n", retval);
+ rpaphp_release_slot(slot->hotplug_slot);
+ return (retval);
+ }
+ break;
+ case VIO_DEV:
+ /* create symlink between slot->name and it's uni-address */
+ vio_uni_addr = strchr(slot->dn->full_name, '@');
+ if (!vio_uni_addr)
+ return (1);
+ dbg("%s: sysfs_create_link: %s --> %s\n", __FUNCTION__,
+ vio_uni_addr, slot->name);
+ retval = sysfs_create_link(slot->hotplug_slot->kobj.parent,
+ &slot->hotplug_slot->kobj,
+ vio_uni_addr);
+ if (retval) {
+ err("sysfs_create_link failed with error %d\n", retval);
+ rpaphp_release_slot(slot->hotplug_slot);
+ return (retval);
+ }
+ break;
+ default:
+ return (1);
+ }
+
+ /* add slot to our internal list */
+ dbg("%s adding slot[%s] to rpaphp_slot_list\n",
+ __FUNCTION__, slot->name);
+
+ list_add(&slot->rpaphp_slot_list, &rpaphp_slot_head);
+
+ if (vio_uni_addr)
+ info("Slot [%s](vio_uni_addr=%s) registered\n",
+ slot->name, vio_uni_addr);
+ else
+ info("Slot [%s](bus_id=%s) registered\n",
+ slot->name, pci_name(slot->bridge));
+ num_slots++;
+ return (0);
+}
+
+int rpaphp_get_power_status(struct slot *slot, u8 * value)
+{
+ int rc;
+
+ rc = rtas_get_power_level(slot->power_domain, (int *) value);
+ if (rc)
+ err("failed to get power-level for slot(%s), rc=0x%x\n",
+ slot->name, rc);
+
+ return rc;
+}
+
+int rpaphp_set_attention_status(struct slot *slot, u8 status)
+{
+ int rc;
+
+ /* status: LED_OFF or LED_ON */
+ rc = rtas_set_indicator(DR_INDICATOR, slot->index, status);
+ if (rc)
+ err("slot(%s) set attention-status(%d) failed! rc=0x%x\n",
+ slot->name, status, rc);
+
+ return rc;
+}
--- diff/drivers/pci/hotplug/rpaphp_vio.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/pci/hotplug/rpaphp_vio.c 2004-03-16 11:27:48.224856368 +0000
@@ -0,0 +1,121 @@
+/*
+ * RPA Hot Plug Virtual I/O device functions
+ * Copyright (C) 2004 Linda Xie <lxie@us.ibm.com>
+ *
+ * All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or (at
+ * your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
+ * NON INFRINGEMENT. See the GNU General Public License for more
+ * details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ * Send feedback to <lxie@us.ibm.com>
+ *
+ */
+#include <asm/vio.h>
+#include "rpaphp.h"
+
+/*
+ * get_vio_adapter_status - get the status of a slot
+ *
+ * status:
+ *
+ * 1-- adapter is configured
+ * 2-- adapter is not configured
+ * 3-- not valid
+ */
+inline int rpaphp_get_vio_adapter_status(struct slot *slot, int is_init, u8 *value)
+{
+ *value = slot->state;
+ return 0;
+}
+
+int rpaphp_unconfig_vio_adapter(struct slot *slot)
+{
+ int retval = 0;
+
+ dbg("Entry %s: slot[%s]\n", __FUNCTION__, slot->name);
+ if (!slot->dev.vio_dev) {
+ info("%s: no VIOA in slot[%s]\n", __FUNCTION__, slot->name);
+ retval = -EINVAL;
+ goto exit;
+ }
+ /* remove the device from the vio core */
+ vio_unregister_device(slot->dev.vio_dev);
+ slot->state = NOT_CONFIGURED;
+ info("%s: adapter in slot[%s] unconfigured.\n", __FUNCTION__, slot->name);
+exit:
+ dbg("Exit %s, rc=0x%x\n", __FUNCTION__, retval);
+ return retval;
+}
+
+static int setup_vio_hotplug_slot_info(struct slot *slot)
+{
+ slot->hotplug_slot->info->power_status = 1;
+ rpaphp_get_vio_adapter_status(slot, 1,
+ &slot->hotplug_slot->info->adapter_status);
+ return 0;
+}
+
+int register_vio_slot(struct device_node *dn)
+{
+ u32 *index;
+ char *name;
+ int rc = 1;
+ struct slot *slot = NULL;
+
+ index = (u32 *) get_property(dn, "ibm,my-drc-index", NULL);
+ if (!index)
+ goto exit_rc;
+ name = get_property(dn, "ibm,loc-code", NULL);
+ if (!name)
+ goto exit_rc;
+ if (!(slot = alloc_slot_struct(dn, *index, name, 0))) {
+ rc = -ENOMEM;
+ goto exit_rc;
+ }
+ slot->dev_type = VIO_DEV;
+ slot->dev.vio_dev = vio_find_node(dn);
+ if (!slot->dev.vio_dev)
+ slot->dev.vio_dev = vio_register_device(dn);
+ if (slot->dev.vio_dev)
+ slot->state = CONFIGURED;
+ else
+ slot->state = NOT_CONFIGURED;
+ if (setup_vio_hotplug_slot_info(slot))
+ goto exit_rc;
+ info("%s: registered VIO device[name=%s vio_dev=%p]\n",
+ __FUNCTION__, slot->name, slot->dev.vio_dev);
+ rc = register_slot(slot);
+exit_rc:
+ if (rc && slot)
+ dealloc_slot_struct(slot);
+ return (rc);
+}
+
+int rpaphp_enable_vio_slot(struct slot *slot)
+{
+ int retval = 0;
+
+ if ((slot->dev.vio_dev = vio_register_device(slot->dn))) {
+ info("%s: VIO adapter %s in slot[%s] has been configured\n",
+ __FUNCTION__, slot->dn->name, slot->name);
+ slot->state = CONFIGURED;
+ } else {
+ info("%s: no vio_dev struct for adapter in slot[%s]\n",
+ __FUNCTION__, slot->name);
+ slot->state = NOT_CONFIGURED;
+ }
+
+ return retval;
+}
--- diff/drivers/usb/gadget/config.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/usb/gadget/config.c 2004-03-16 11:27:48.269849528 +0000
@@ -0,0 +1,116 @@
+/*
+ * usb/gadget/config.c -- simplify building config descriptors
+ *
+ * Copyright (C) 2003 David Brownell
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/string.h>
+#include <linux/device.h>
+
+#include <linux/usb_ch9.h>
+
+
+/**
+ * usb_descriptor_fillbuf - fill buffer with descriptors
+ * @buf: Buffer to be filled
+ * @buflen: Size of buf
+ * @src: Array of descriptor pointers, terminated by null pointer.
+ *
+ * Copies descriptors into the buffer, returning the length or a
+ * negative error code if they can't all be copied. Useful when
+ * assembling descriptors for an associated set of interfaces used
+ * as part of configuring a composite device; or in other cases where
+ * sets of descriptors need to be marshaled.
+ */
+int
+usb_descriptor_fillbuf(void *buf, unsigned buflen,
+ const struct usb_descriptor_header **src)
+{
+ u8 *dest = buf;
+
+ if (!src)
+ return -EINVAL;
+
+ /* fill buffer from src[] until null descriptor ptr */
+ for (; 0 != *src; src++) {
+ unsigned len = (*src)->bLength;
+
+ if (len > buflen)
+ return -EINVAL;
+ memcpy(dest, *src, len);
+ buflen -= len;
+ dest += len;
+ }
+ return dest - (u8 *)buf;
+}
+
+
+/**
+ * usb_gadget_config_buf - builts a complete configuration descriptor
+ * @config: Header for the descriptor, including characteristics such
+ * as power requirements and number of interfaces.
+ * @desc: Null-terminated vector of pointers to the descriptors (interface,
+ * endpoint, etc) defining all functions in this device configuration.
+ * @buf: Buffer for the resulting configuration descriptor.
+ * @length: Length of buffer. If this is not big enough to hold the
+ * entire configuration descriptor, an error code will be returned.
+ *
+ * This copies descriptors into the response buffer, building a descriptor
+ * for that configuration. It returns the buffer length or a negative
+ * status code. The config.wTotalLength field is set to match the length
+ * of the result, but other descriptor fields (including power usage and
+ * interface count) must be set by the caller.
+ *
+ * Gadget drivers could use this when constructing a config descriptor
+ * in response to USB_REQ_GET_DESCRIPTOR. They will need to patch the
+ * resulting bDescriptorType value if USB_DT_OTHER_SPEED_CONFIG is needed.
+ */
+int usb_gadget_config_buf(
+ const struct usb_config_descriptor *config,
+ void *buf,
+ unsigned length,
+ const struct usb_descriptor_header **desc
+)
+{
+ struct usb_config_descriptor *cp = buf;
+ int len;
+
+ /* config descriptor first */
+ if (length < USB_DT_CONFIG_SIZE || !desc)
+ return -EINVAL;
+ *cp = *config;
+
+ /* then interface/endpoint/class/vendor/... */
+ len = usb_descriptor_fillbuf(USB_DT_CONFIG_SIZE + (u8*)buf,
+ length - USB_DT_CONFIG_SIZE, desc);
+ if (len < 0)
+ return len;
+ len += USB_DT_CONFIG_SIZE;
+ if (len > 0xffff)
+ return -EINVAL;
+
+ /* patch up the config descriptor */
+ cp->bLength = USB_DT_CONFIG_SIZE;
+ cp->bDescriptorType = USB_DT_CONFIG;
+ cp->wTotalLength = cpu_to_le16(len);
+ cp->bmAttributes |= USB_CONFIG_ATT_ONE;
+ return len;
+}
+
--- diff/drivers/usb/gadget/gadget_chips.h 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/usb/gadget/gadget_chips.h 2004-03-16 11:27:48.271849224 +0000
@@ -0,0 +1,57 @@
+/*
+ * USB device controllers have lots of quirks. Use these macros in
+ * gadget drivers or other code that needs to deal with them, and which
+ * autoconfigures instead of using early binding to the hardware.
+ *
+ * This could eventually work like the ARM mach_is_*() stuff, driven by
+ * some config file that gets updated as new hardware is supported.
+ *
+ * NOTE: some of these controller drivers may not be available yet.
+ */
+#ifdef CONFIG_USB_GADGET_NET2280
+#define gadget_is_net2280(g) !strcmp("net2280", (g)->name)
+#else
+#define gadget_is_net2280(g) 0
+#endif
+
+#ifdef CONFIG_USB_GADGET_PXA
+#define gadget_is_pxa(g) !strcmp("pxa2xx_udc", (g)->name)
+#else
+#define gadget_is_pxa(g) 0
+#endif
+
+#ifdef CONFIG_USB_GADGET_GOKU
+#define gadget_is_goku(g) !strcmp("goku_udc", (g)->name)
+#else
+#define gadget_is_goku(g) 0
+#endif
+
+#ifdef CONFIG_USB_GADGET_SUPERH
+#define gadget_is_sh(g) !strcmp("sh_udc", (g)->name)
+#else
+#define gadget_is_sh(g) 0
+#endif
+
+#ifdef CONFIG_USB_GADGET_SA1100
+#define gadget_is_sa1100(g) !strcmp("sa1100_udc", (g)->name)
+#else
+#define gadget_is_sa1100(g) 0
+#endif
+
+#ifdef CONFIG_USB_GADGET_MQ11XX
+#define gadget_is_mq11xx(g) !strcmp("mq11xx_udc", (g)->name)
+#else
+#define gadget_is_mq11xx(g) 0
+#endif
+
+#ifdef CONFIG_USB_GADGET_OMAP
+#define gadget_is_omap(g) !strcmp("omap_udc", (g)->name)
+#else
+#define gadget_is_omap(g) 0
+#endif
+
+// CONFIG_USB_GADGET_AT91RM9200
+// CONFIG_USB_GADGET_SX2
+// CONFIG_USB_GADGET_AU1X00
+// ...
+
--- diff/drivers/usb/input/ati_remote.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/usb/input/ati_remote.c 2004-03-16 11:27:48.292846032 +0000
@@ -0,0 +1,851 @@
+/*
+ * USB ATI Remote support
+ *
+ * Version 2.2.0 Copyright (c) 2004 Torrey Hoffman <thoffman@arnor.net>
+ * Version 2.1.1 Copyright (c) 2002 Vladimir Dergachev
+ *
+ * This 2.2.0 version is a rewrite / cleanup of the 2.1.1 driver, including
+ * porting to the 2.6 kernel interfaces, along with other modification
+ * to better match the style of the existing usb/input drivers. However, the
+ * protocol and hardware handling is essentially unchanged from 2.1.1.
+ *
+ * The 2.1.1 driver was derived from the usbati_remote and usbkbd drivers by
+ * Vojtech Pavlik.
+ *
+ * Changes:
+ *
+ * Feb 2004: Torrey Hoffman <thoffman@arnor.net>
+ * Version 2.2.0
+ *
+ * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
+ *
+ * Hardware & software notes
+ *
+ * These remote controls are distributed by ATI as part of their
+ * "All-In-Wonder" video card packages. The receiver self-identifies as a
+ * "USB Receiver" with manufacturer "X10 Wireless Technology Inc".
+ *
+ * It is possible to use multiple receivers and remotes on multiple computers
+ * simultaneously by configuring them to use specific channels.
+ *
+ * The RF protocol used by the remote supports 16 distinct channels, 1 to 16.
+ * Actually, it may even support more, at least in some revisions of the
+ * hardware.
+ *
+ * Each remote can be configured to transmit on one channel as follows:
+ * - Press and hold the "hand icon" button.
+ * - When the red LED starts to blink, let go of the "hand icon" button.
+ * - When it stops blinking, input the channel code as two digits, from 01
+ * to 16, and press the hand icon again.
+ *
+ * The timing can be a little tricky. Try loading the module with debug=1
+ * to have the kernel print out messages about the remote control number
+ * and mask. Note: debugging prints remote numbers as zero-based hexadecimal.
+ *
+ * The driver has a "channel_mask" parameter. This bitmask specifies which
+ * channels will be ignored by the module. To mask out channels, just add
+ * all the 2^channel_number values together.
+ *
+ * For instance, set channel_mask = 2^4 = 16 (binary 10000) to make ati_remote
+ * ignore signals coming from remote controls transmitting on channel 4, but
+ * accept all other channels.
+ *
+ * Or, set channel_mask = 65533, (0xFFFD), and all channels except 1 will be
+ * ignored.
+ *
+ * The default is 0 (respond to all channels). Bit 0 and bits 17-32 of this
+ * parameter are unused.
+ *
+ */
+
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/input.h>
+#include <linux/usb.h>
+
+/*
+ * Module and Version Information, Module Parameters
+ */
+
+#define ATI_REMOTE_VENDOR_ID 0x0bc7
+#define ATI_REMOTE_PRODUCT_ID 0x004
+
+#define DRIVER_VERSION "2.2.0"
+#define DRIVER_AUTHOR "Torrey Hoffman <thoffman@arnor.net>"
+#define DRIVER_DESC "ATI/X10 RF USB Remote Control"
+
+#define NAME_BUFSIZE 80 /* size of product name, path buffers */
+#define DATA_BUFSIZE 63 /* size of URB data buffers */
+#define ATI_INPUTNUM 1 /* Which input device to register as */
+
+unsigned long channel_mask = 0;
+module_param(channel_mask, ulong, 444);
+MODULE_PARM_DESC(channel_mask, "Bitmask of remote control channels to ignore");
+
+static int debug = 0;
+module_param(debug, int, 444);
+MODULE_PARM_DESC(debug, "Enable extra debug messages and information");
+
+#define dbginfo(dev, format, arg...) do { if (debug) dev_info(dev , format , ## arg); } while (0)
+#undef err
+#define err(format, arg...) printk(KERN_ERR format , ## arg)
+
+static struct usb_device_id ati_remote_table[] = {
+ { USB_DEVICE(ATI_REMOTE_VENDOR_ID, ATI_REMOTE_PRODUCT_ID) },
+ {} /* Terminating entry */
+};
+
+MODULE_DEVICE_TABLE(usb, ati_remote_table);
+
+/* Get hi and low bytes of a 16-bits int */
+#define HI(a) ((unsigned char)((a) >> 8))
+#define LO(a) ((unsigned char)((a) & 0xff))
+
+#define SEND_FLAG_IN_PROGRESS 1
+#define SEND_FLAG_COMPLETE 2
+
+/* Device initialization strings */
+static char init1[] = { 0x01, 0x00, 0x20, 0x14 };
+static char init2[] = { 0x01, 0x00, 0x20, 0x14, 0x20, 0x20, 0x20 };
+
+/* Acceleration curve for directional control pad */
+static char accel[] = { 1, 2, 4, 6, 9, 13, 20 };
+
+/* Duplicate event filtering time.
+ * Sequential, identical KIND_FILTERED inputs with less than
+ * FILTER_TIME jiffies between them are dropped.
+ * (HZ >> 4) == 1/16th of a second and works well for me.
+ */
+#define FILTER_TIME (HZ >> 4)
+
+struct ati_remote {
+ struct input_dev idev;
+ struct usb_device *udev;
+ struct usb_interface *interface;
+
+ struct urb *irq_urb;
+ struct urb *out_urb;
+ struct usb_endpoint_descriptor *endpoint_in;
+ struct usb_endpoint_descriptor *endpoint_out;
+ unsigned char *inbuf;
+ unsigned char *outbuf;
+ dma_addr_t inbuf_dma;
+ dma_addr_t outbuf_dma;
+
+ int open; /* open counter */
+ int present; /* device plugged in? */
+
+ unsigned char old_data[2]; /* Detect duplicate events */
+ unsigned long old_jiffies;
+ unsigned long acc_jiffies; /* handle acceleration */
+
+ char name[NAME_BUFSIZE];
+ char phys[NAME_BUFSIZE];
+
+ wait_queue_head_t wait;
+ int send_flags;
+};
+
+/* "Kinds" of messages sent from the hardware to the driver. */
+#define KIND_END 0
+#define KIND_LITERAL 1 /* Simply pass to input system */
+#define KIND_FILTERED 2 /* Add artificial key-up events, drop keyrepeats */
+#define KIND_LU 3 /* Directional keypad diagonals - left up, */
+#define KIND_RU 4 /* right up, */
+#define KIND_LD 5 /* left down, */
+#define KIND_RD 6 /* right down */
+#define KIND_ACCEL 7 /* Directional keypad - left, right, up, down.*/
+
+/* Translation table from hardware messages to input events. */
+static struct
+{
+ short kind;
+ unsigned char data1, data2;
+ int type;
+ unsigned int code;
+ int value;
+} ati_remote_tbl[] =
+{
+ /* Directional control pad axes */
+ {KIND_ACCEL, 0x35, 0x70, EV_REL, REL_X, -1}, /* left */
+ {KIND_ACCEL, 0x36, 0x71, EV_REL, REL_X, 1}, /* right */
+ {KIND_ACCEL, 0x37, 0x72, EV_REL, REL_Y, -1}, /* up */
+ {KIND_ACCEL, 0x38, 0x73, EV_REL, REL_Y, 1}, /* down */
+ /* Directional control pad diagonals */
+ {KIND_LU, 0x39, 0x74, EV_REL, 0, 0}, /* left up */
+ {KIND_RU, 0x3a, 0x75, EV_REL, 0, 0}, /* right up */
+ {KIND_LD, 0x3c, 0x77, EV_REL, 0, 0}, /* left down */
+ {KIND_RD, 0x3b, 0x76, EV_REL, 0, 0}, /* right down */
+
+ /* "Mouse button" buttons */
+ {KIND_LITERAL, 0x3d, 0x78, EV_KEY, BTN_LEFT, 1}, /* left btn down */
+ {KIND_LITERAL, 0x3e, 0x79, EV_KEY, BTN_LEFT, 0}, /* left btn up */
+ {KIND_LITERAL, 0x41, 0x7c, EV_KEY, BTN_RIGHT, 1},/* right btn down */
+ {KIND_LITERAL, 0x42, 0x7d, EV_KEY, BTN_RIGHT, 0},/* right btn up */
+
+ /* Artificial "doubleclick" events are generated by the hardware.
+ * They are mapped to the "side" and "extra" mouse buttons here. */
+ {KIND_FILTERED, 0x3f, 0x7a, EV_KEY, BTN_SIDE, 1}, /* left dblclick */
+ {KIND_FILTERED, 0x43, 0x7e, EV_KEY, BTN_EXTRA, 1},/* right dblclick */
+
+ /* keyboard. */
+ {KIND_FILTERED, 0xd2, 0x0d, EV_KEY, KEY_1, 1},
+ {KIND_FILTERED, 0xd3, 0x0e, EV_KEY, KEY_2, 1},
+ {KIND_FILTERED, 0xd4, 0x0f, EV_KEY, KEY_3, 1},
+ {KIND_FILTERED, 0xd5, 0x10, EV_KEY, KEY_4, 1},
+ {KIND_FILTERED, 0xd6, 0x11, EV_KEY, KEY_5, 1},
+ {KIND_FILTERED, 0xd7, 0x12, EV_KEY, KEY_6, 1},
+ {KIND_FILTERED, 0xd8, 0x13, EV_KEY, KEY_7, 1},
+ {KIND_FILTERED, 0xd9, 0x14, EV_KEY, KEY_8, 1},
+ {KIND_FILTERED, 0xda, 0x15, EV_KEY, KEY_9, 1},
+ {KIND_FILTERED, 0xdc, 0x17, EV_KEY, KEY_0, 1},
+ {KIND_FILTERED, 0xc5, 0x00, EV_KEY, KEY_A, 1},
+ {KIND_FILTERED, 0xc6, 0x01, EV_KEY, KEY_B, 1},
+ {KIND_FILTERED, 0xde, 0x19, EV_KEY, KEY_C, 1},
+ {KIND_FILTERED, 0xe0, 0x1b, EV_KEY, KEY_D, 1},
+ {KIND_FILTERED, 0xe6, 0x21, EV_KEY, KEY_E, 1},
+ {KIND_FILTERED, 0xe8, 0x23, EV_KEY, KEY_F, 1},
+
+ /* "special" keys */
+ {KIND_FILTERED, 0xdd, 0x18, EV_KEY, KEY_KPENTER, 1}, /* "check" */
+ {KIND_FILTERED, 0xdb, 0x16, EV_KEY, KEY_MENU, 1}, /* "menu" */
+ {KIND_FILTERED, 0xc7, 0x02, EV_KEY, KEY_POWER, 1}, /* Power */
+ {KIND_FILTERED, 0xc8, 0x03, EV_KEY, KEY_PROG1, 1}, /* TV */
+ {KIND_FILTERED, 0xc9, 0x04, EV_KEY, KEY_PROG2, 1}, /* DVD */
+ {KIND_FILTERED, 0xca, 0x05, EV_KEY, KEY_WWW, 1}, /* WEB */
+ {KIND_FILTERED, 0xcb, 0x06, EV_KEY, KEY_BOOKMARKS, 1}, /* "book" */
+ {KIND_FILTERED, 0xcc, 0x07, EV_KEY, KEY_EDIT, 1}, /* "hand" */
+ {KIND_FILTERED, 0xe1, 0x1c, EV_KEY, KEY_COFFEE, 1}, /* "timer" */
+ {KIND_FILTERED, 0xe5, 0x20, EV_KEY, KEY_FRONT, 1}, /* "max" */
+ {KIND_FILTERED, 0xe2, 0x1d, EV_KEY, KEY_LEFT, 1}, /* left */
+ {KIND_FILTERED, 0xe4, 0x1f, EV_KEY, KEY_RIGHT, 1}, /* right */
+ {KIND_FILTERED, 0xe7, 0x22, EV_KEY, KEY_DOWN, 1}, /* down */
+ {KIND_FILTERED, 0xdf, 0x1a, EV_KEY, KEY_UP, 1}, /* up */
+ {KIND_FILTERED, 0xe3, 0x1e, EV_KEY, KEY_ENTER, 1}, /* "OK" */
+ {KIND_FILTERED, 0xce, 0x09, EV_KEY, KEY_VOLUMEDOWN, 1}, /* VOL + */
+ {KIND_FILTERED, 0xcd, 0x08, EV_KEY, KEY_VOLUMEUP, 1}, /* VOL - */
+ {KIND_FILTERED, 0xcf, 0x0a, EV_KEY, KEY_MUTE, 1}, /* MUTE */
+ {KIND_FILTERED, 0xd1, 0x0c, EV_KEY, KEY_CHANNELUP, 1}, /* CH + */
+ {KIND_FILTERED, 0xd0, 0x0b, EV_KEY, KEY_CHANNELDOWN, 1},/* CH - */
+ {KIND_FILTERED, 0xec, 0x27, EV_KEY, KEY_RECORD, 1}, /* ( o) red */
+ {KIND_FILTERED, 0xea, 0x25, EV_KEY, KEY_PLAYCD, 1}, /* ( >) */
+ {KIND_FILTERED, 0xe9, 0x24, EV_KEY, KEY_REWIND, 1}, /* (<<) */
+ {KIND_FILTERED, 0xeb, 0x26, EV_KEY, KEY_FORWARD, 1}, /* (>>) */
+ {KIND_FILTERED, 0xed, 0x28, EV_KEY, KEY_STOP, 1}, /* ([]) */
+ {KIND_FILTERED, 0xee, 0x29, EV_KEY, KEY_PAUSE, 1}, /* ('') */
+
+ {KIND_END, 0x00, 0x00, EV_MAX + 1, 0, 0}
+};
+
+/* Local function prototypes */
+static void ati_remote_dump (unsigned char *data, unsigned int actual_length);
+static void ati_remote_delete (struct ati_remote *dev);
+static int ati_remote_open (struct input_dev *inputdev);
+static void ati_remote_close (struct input_dev *inputdev);
+static int ati_remote_sendpacket (struct ati_remote *ati_remote, u16 cmd, unsigned char *data);
+static void ati_remote_irq_out (struct urb *urb, struct pt_regs *regs);
+static void ati_remote_irq_in (struct urb *urb, struct pt_regs *regs);
+static void ati_remote_input_report (struct urb *urb, struct pt_regs *regs);
+static int ati_remote_initialize (struct ati_remote *ati_remote);
+static int ati_remote_probe (struct usb_interface *interface, const struct usb_device_id *id);
+static void ati_remote_disconnect (struct usb_interface *interface);
+
+/* usb specific object to register with the usb subsystem */
+static struct usb_driver ati_remote_driver = {
+ .owner = THIS_MODULE,
+ .name = "ati_remote",
+ .probe = ati_remote_probe,
+ .disconnect = ati_remote_disconnect,
+ .id_table = ati_remote_table,
+};
+
+/*
+ * ati_remote_dump_input
+ */
+static void ati_remote_dump(unsigned char *data, unsigned int len)
+{
+ if ((len == 1) && (data[0] != (unsigned char)0xff) && (data[0] != 0x00))
+ warn("Weird byte 0x%02x\n", data[0]);
+ else if (len == 4)
+ warn("Weird key %02x %02x %02x %02x\n",
+ data[0], data[1], data[2], data[3]);
+ else
+ warn("Weird data, len=%d %02x %02x %02x %02x %02x %02x ...\n",
+ len, data[0], data[1], data[2], data[3], data[4], data[5]);
+}
+
+/*
+ * ati_remote_open
+ */
+static int ati_remote_open(struct input_dev *inputdev)
+{
+ struct ati_remote *ati_remote = inputdev->private;
+
+ if (ati_remote->open++)
+ return 0;
+
+ /* On first open, submit the read urb which was set up previously. */
+ ati_remote->irq_urb->dev = ati_remote->udev;
+ if (usb_submit_urb(ati_remote->irq_urb, GFP_KERNEL)) {
+ dev_err(&ati_remote->interface->dev,
+ "%s: usb_submit_urb failed!\n", __FUNCTION__);
+ ati_remote->open--;
+ return -EIO;
+ }
+
+ return 0;
+}
+
+/*
+ * ati_remote_close
+ */
+static void ati_remote_close(struct input_dev *inputdev)
+{
+ struct ati_remote *ati_remote = inputdev->private;
+
+ if (ati_remote == NULL) {
+ err("ati_remote: %s: object is NULL!\n", __FUNCTION__);
+ return;
+ }
+
+ if (ati_remote->open <= 0)
+ dev_dbg(&ati_remote->interface->dev, "%s: Not open.\n", __FUNCTION__);
+ else
+ --ati_remote->open;
+
+ /* If still present, disconnect will call delete. */
+ if (!ati_remote->present && !ati_remote->open)
+ ati_remote_delete(ati_remote);
+}
+
+/*
+ * ati_remote_irq_out
+ */
+static void ati_remote_irq_out(struct urb *urb, struct pt_regs *regs)
+{
+ struct ati_remote *ati_remote = urb->context;
+
+ if (urb->status) {
+ dev_dbg(&ati_remote->interface->dev, "%s: status %d\n",
+ __FUNCTION__, urb->status);
+ return;
+ }
+
+ ati_remote->send_flags |= SEND_FLAG_COMPLETE;
+ wmb();
+ if (waitqueue_active(&ati_remote->wait))
+ wake_up(&ati_remote->wait);
+}
+
+/*
+ * ati_remote_sendpacket
+ *
+ * Used to send device initialization strings
+ */
+static int ati_remote_sendpacket(struct ati_remote *ati_remote, u16 cmd, unsigned char *data)
+{
+ DECLARE_WAITQUEUE(wait, current);
+ int timeout = HZ; /* 1 second */
+ int retval = 0;
+
+ /* Set up out_urb */
+ memcpy(ati_remote->out_urb->transfer_buffer + 1, data, LO(cmd));
+ ((char *) ati_remote->out_urb->transfer_buffer)[0] = HI(cmd);
+
+ ati_remote->out_urb->transfer_buffer_length = LO(cmd) + 1;
+ ati_remote->out_urb->dev = ati_remote->udev;
+ ati_remote->send_flags = SEND_FLAG_IN_PROGRESS;
+
+ set_current_state(TASK_INTERRUPTIBLE);
+ add_wait_queue(&ati_remote->wait, &wait);
+
+ retval = usb_submit_urb(ati_remote->out_urb, GFP_KERNEL);
+ if (retval) {
+ set_current_state(TASK_RUNNING);
+ remove_wait_queue(&ati_remote->wait, &wait);
+ dev_dbg(&ati_remote->interface->dev,
+ "sendpacket: usb_submit_urb failed: %d\n", retval);
+ return retval;
+ }
+
+ while (timeout && (ati_remote->out_urb->status == -EINPROGRESS)
+ && !(ati_remote->send_flags & SEND_FLAG_COMPLETE)) {
+ timeout = schedule_timeout(timeout);
+ rmb();
+ }
+
+ set_current_state(TASK_RUNNING);
+ remove_wait_queue(&ati_remote->wait, &wait);
+ usb_unlink_urb(ati_remote->out_urb);
+
+ return retval;
+}
+
+/*
+ * ati_remote_event_lookup
+ */
+static int ati_remote_event_lookup(int rem, unsigned char d1, unsigned char d2)
+{
+ int i;
+
+ for (i = 0; ati_remote_tbl[i].kind != KIND_END; i++) {
+ /*
+ * Decide if the table entry matches the remote input.
+ */
+ if ((((ati_remote_tbl[i].data1 & 0x0f) == (d1 & 0x0f))) &&
+ ((((ati_remote_tbl[i].data1 >> 4) -
+ (d1 >> 4) + rem) & 0x0f) == 0x0f) &&
+ (ati_remote_tbl[i].data2 == d2))
+ return i;
+
+ }
+ return -1;
+}
+
+/*
+ * ati_remote_report_input
+ */
+static void ati_remote_input_report(struct urb *urb, struct pt_regs *regs)
+{
+ struct ati_remote *ati_remote = urb->context;
+ unsigned char *data= ati_remote->inbuf;
+ struct input_dev *dev = &ati_remote->idev;
+ int index, acc;
+ int remote_num;
+
+ /* Deal with strange looking inputs */
+ if ( (urb->actual_length != 4) || (data[0] != 0x14) ||
+ ((data[3] & 0x0f) != 0x00) ) {
+ ati_remote_dump(data, urb->actual_length);
+ return;
+ }
+
+ /* Mask unwanted remote channels. */
+ /* note: remote_num is 0-based, channel 1 on remote == 0 here */
+ remote_num = (data[3] >> 4) & 0x0f;
+ if (channel_mask & (1 << (remote_num + 1))) {
+ dbginfo(&ati_remote->interface->dev,
+ "Masked input from channel 0x%02x: data %02x,%02x, mask= 0x%02lx\n",
+ remote_num, data[1], data[2], channel_mask);
+ return;
+ }
+
+ /* Look up event code index in translation table */
+ index = ati_remote_event_lookup(remote_num, data[1], data[2]);
+ if (index < 0) {
+ dev_warn(&ati_remote->interface->dev,
+ "Unknown input from channel 0x%02x: data %02x,%02x\n",
+ remote_num, data[1], data[2]);
+ return;
+ }
+ dbginfo(&ati_remote->interface->dev,
+ "channel 0x%02x; data %02x,%02x; index %d; keycode %d\n",
+ remote_num, data[1], data[2], index, ati_remote_tbl[index].code);
+
+ if (ati_remote_tbl[index].kind == KIND_LITERAL) {
+ input_regs(dev, regs);
+ input_event(dev, ati_remote_tbl[index].type,
+ ati_remote_tbl[index].code,
+ ati_remote_tbl[index].value);
+ input_sync(dev);
+
+ ati_remote->old_jiffies = jiffies;
+ return;
+ }
+
+ if (ati_remote_tbl[index].kind == KIND_FILTERED) {
+ /* Filter duplicate events which happen "too close" together. */
+ if ((ati_remote->old_data[0] == data[1]) &&
+ (ati_remote->old_data[1] == data[2]) &&
+ ((ati_remote->old_jiffies + FILTER_TIME) > jiffies)) {
+ ati_remote->old_jiffies = jiffies;
+ return;
+ }
+
+ input_regs(dev, regs);
+ input_event(dev, ati_remote_tbl[index].type,
+ ati_remote_tbl[index].code, 1);
+ input_event(dev, ati_remote_tbl[index].type,
+ ati_remote_tbl[index].code, 0);
+ input_sync(dev);
+
+ ati_remote->old_data[0] = data[1];
+ ati_remote->old_data[1] = data[2];
+ ati_remote->old_jiffies = jiffies;
+ return;
+ }
+
+ /*
+ * Other event kinds are from the directional control pad, and have an
+ * acceleration factor applied to them. Without this acceleration, the
+ * control pad is mostly unusable.
+ *
+ * If elapsed time since last event is > 1/4 second, user "stopped",
+ * so reset acceleration. Otherwise, user is probably holding the control
+ * pad down, so we increase acceleration, ramping up over two seconds to
+ * a maximum speed. The acceleration curve is #defined above.
+ */
+ if ((jiffies - ati_remote->old_jiffies) > (HZ >> 2)) {
+ acc = 1;
+ ati_remote->acc_jiffies = jiffies;
+ }
+ else if ((jiffies - ati_remote->acc_jiffies) < (HZ >> 3)) acc = accel[0];
+ else if ((jiffies - ati_remote->acc_jiffies) < (HZ >> 2)) acc = accel[1];
+ else if ((jiffies - ati_remote->acc_jiffies) < (HZ >> 1)) acc = accel[2];
+ else if ((jiffies - ati_remote->acc_jiffies) < HZ ) acc = accel[3];
+ else if ((jiffies - ati_remote->acc_jiffies) < HZ+(HZ>>1)) acc = accel[4];
+ else if ((jiffies - ati_remote->acc_jiffies) < (HZ << 1)) acc = accel[5];
+ else acc = accel[6];
+
+ input_regs(dev, regs);
+ switch (ati_remote_tbl[index].kind) {
+ case KIND_ACCEL:
+ input_event(dev, ati_remote_tbl[index].type,
+ ati_remote_tbl[index].code,
+ ati_remote_tbl[index].value * acc);
+ break;
+ case KIND_LU:
+ input_report_rel(dev, REL_X, -acc);
+ input_report_rel(dev, REL_Y, -acc);
+ break;
+ case KIND_RU:
+ input_report_rel(dev, REL_X, acc);
+ input_report_rel(dev, REL_Y, -acc);
+ break;
+ case KIND_LD:
+ input_report_rel(dev, REL_X, -acc);
+ input_report_rel(dev, REL_Y, acc);
+ break;
+ case KIND_RD:
+ input_report_rel(dev, REL_X, acc);
+ input_report_rel(dev, REL_Y, acc);
+ break;
+ default:
+ dev_dbg(&ati_remote->interface->dev, "ati_remote kind=%d\n",
+ ati_remote_tbl[index].kind);
+ }
+ input_sync(dev);
+
+ ati_remote->old_jiffies = jiffies;
+ ati_remote->old_data[0] = data[1];
+ ati_remote->old_data[1] = data[2];
+}
+
+/*
+ * ati_remote_irq_in
+ */
+static void ati_remote_irq_in(struct urb *urb, struct pt_regs *regs)
+{
+ struct ati_remote *ati_remote = urb->context;
+ int retval;
+
+ switch (urb->status) {
+ case 0: /* success */
+ ati_remote_input_report(urb, regs);
+ break;
+ case -ECONNRESET: /* unlink */
+ case -ENOENT:
+ case -ESHUTDOWN:
+ dev_dbg(&ati_remote->interface->dev, "%s: urb error status, unlink? \n",
+ __FUNCTION__);
+ return;
+ default: /* error */
+ dev_dbg(&ati_remote->interface->dev, "%s: Nonzero urb status %d\n",
+ __FUNCTION__, urb->status);
+ }
+
+ retval = usb_submit_urb(urb, SLAB_ATOMIC);
+ if (retval)
+ dev_err(&ati_remote->interface->dev, "%s: usb_submit_urb()=%d\n",
+ __FUNCTION__, retval);
+}
+
+/*
+ * ati_remote_delete
+ */
+static void ati_remote_delete(struct ati_remote *ati_remote)
+{
+ if (!ati_remote) return;
+
+ if (ati_remote->irq_urb)
+ usb_unlink_urb(ati_remote->irq_urb);
+
+ if (ati_remote->out_urb)
+ usb_unlink_urb(ati_remote->out_urb);
+
+ if (ati_remote->irq_urb)
+ usb_free_urb(ati_remote->irq_urb);
+
+ if (ati_remote->out_urb)
+ usb_free_urb(ati_remote->out_urb);
+
+ if (ati_remote->inbuf)
+ usb_buffer_free(ati_remote->udev, DATA_BUFSIZE,
+ ati_remote->inbuf, ati_remote->inbuf_dma);
+
+ if (ati_remote->outbuf)
+ usb_buffer_free(ati_remote->udev, DATA_BUFSIZE,
+ ati_remote->inbuf, ati_remote->outbuf_dma);
+
+ kfree(ati_remote);
+}
+
+static void ati_remote_input_init(struct ati_remote *ati_remote)
+{
+ struct input_dev *idev = &(ati_remote->idev);
+ int i;
+
+ idev->evbit[0] = BIT(EV_KEY) | BIT(EV_REL);
+ idev->keybit[LONG(BTN_MOUSE)] = ( BIT(BTN_LEFT) | BIT(BTN_RIGHT) |
+ BIT(BTN_SIDE) | BIT(BTN_EXTRA) );
+ idev->relbit[0] = BIT(REL_X) | BIT(REL_Y);
+ for (i = 0; ati_remote_tbl[i].kind != KIND_END; i++)
+ if (ati_remote_tbl[i].type == EV_KEY)
+ set_bit(ati_remote_tbl[i].code, idev->keybit);
+
+ idev->private = ati_remote;
+ idev->open = ati_remote_open;
+ idev->close = ati_remote_close;
+
+ idev->name = ati_remote->name;
+ idev->phys = ati_remote->phys;
+
+ idev->id.bustype = BUS_USB;
+ idev->id.vendor = ati_remote->udev->descriptor.idVendor;
+ idev->id.product = ati_remote->udev->descriptor.idProduct;
+ idev->id.version = ati_remote->udev->descriptor.bcdDevice;
+}
+
+static int ati_remote_initialize(struct ati_remote *ati_remote)
+{
+ struct usb_device *udev = ati_remote->udev;
+ int pipe, maxp;
+
+ init_waitqueue_head(&ati_remote->wait);
+
+ /* Set up irq_urb */
+ pipe = usb_rcvintpipe(udev, ati_remote->endpoint_in->bEndpointAddress);
+ maxp = usb_maxpacket(udev, pipe, usb_pipeout(pipe));
+ maxp = (maxp > DATA_BUFSIZE) ? DATA_BUFSIZE : maxp;
+
+ usb_fill_int_urb(ati_remote->irq_urb, udev, pipe, ati_remote->inbuf,
+ maxp, ati_remote_irq_in, ati_remote,
+ ati_remote->endpoint_in->bInterval);
+ ati_remote->irq_urb->transfer_dma = ati_remote->inbuf_dma;
+ ati_remote->irq_urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
+
+ /* Set up out_urb */
+ pipe = usb_sndintpipe(udev, ati_remote->endpoint_out->bEndpointAddress);
+ maxp = usb_maxpacket(udev, pipe, usb_pipeout(pipe));
+ maxp = (maxp > DATA_BUFSIZE) ? DATA_BUFSIZE : maxp;
+
+ usb_fill_int_urb(ati_remote->out_urb, udev, pipe, ati_remote->outbuf,
+ maxp, ati_remote_irq_out, ati_remote,
+ ati_remote->endpoint_out->bInterval);
+ ati_remote->out_urb->transfer_dma = ati_remote->outbuf_dma;
+ ati_remote->out_urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
+
+ /* send initialization strings */
+ if ((ati_remote_sendpacket(ati_remote, 0x8004, init1)) ||
+ (ati_remote_sendpacket(ati_remote, 0x8007, init2))) {
+ dev_err(&ati_remote->interface->dev,
+ "Initializing ati_remote hardware failed.\n");
+ return 1;
+ }
+
+ return 0;
+}
+
+/*
+ * ati_remote_probe
+ */
+static int ati_remote_probe(struct usb_interface *interface, const struct usb_device_id *id)
+{
+ struct usb_device *udev = interface_to_usbdev(interface);
+ struct ati_remote *ati_remote = NULL;
+ struct usb_host_interface *iface_host;
+ int retval = -ENOMEM;
+ char path[64];
+ char *buf = NULL;
+
+ /* See if the offered device matches what we can accept */
+ if ((udev->descriptor.idVendor != ATI_REMOTE_VENDOR_ID) ||
+ (udev->descriptor.idProduct != ATI_REMOTE_PRODUCT_ID)) {
+ return -ENODEV;
+ }
+
+ /* Allocate and clear an ati_remote struct */
+ if (!(ati_remote = kmalloc(sizeof (struct ati_remote), GFP_KERNEL)))
+ return -ENOMEM;
+ memset(ati_remote, 0x00, sizeof (struct ati_remote));
+
+ iface_host = interface->cur_altsetting;
+ if (iface_host->desc.bNumEndpoints != 2) {
+ err("%s: Unexpected desc.bNumEndpoints\n", __FUNCTION__);
+ retval = -ENODEV;
+ goto error;
+ }
+
+ ati_remote->endpoint_in = &(iface_host->endpoint[0].desc);
+ ati_remote->endpoint_out = &(iface_host->endpoint[1].desc);
+ ati_remote->udev = udev;
+ ati_remote->interface = interface;
+
+ if (!(ati_remote->endpoint_in->bEndpointAddress & 0x80)) {
+ err("%s: Unexpected endpoint_in->bEndpointAddress\n", __FUNCTION__);
+ retval = -ENODEV;
+ goto error;
+ }
+ if ((ati_remote->endpoint_in->bmAttributes & 3) != 3) {
+ err("%s: Unexpected endpoint_in->bmAttributes\n", __FUNCTION__);
+ retval = -ENODEV;
+ goto error;
+ }
+ if (ati_remote->endpoint_in->wMaxPacketSize == 0) {
+ err("%s: endpoint_in message size==0? \n", __FUNCTION__);
+ retval = -ENODEV;
+ goto error;
+ }
+ if (!(buf = kmalloc(NAME_BUFSIZE, GFP_KERNEL)))
+ goto error;
+
+ /* Allocate URB buffers, URBs */
+ ati_remote->inbuf = usb_buffer_alloc(udev, DATA_BUFSIZE, SLAB_ATOMIC,
+ &ati_remote->inbuf_dma);
+ if (!ati_remote->inbuf)
+ goto error;
+
+ ati_remote->outbuf = usb_buffer_alloc(udev, DATA_BUFSIZE, SLAB_ATOMIC,
+ &ati_remote->outbuf_dma);
+ if (!ati_remote->outbuf)
+ goto error;
+
+ ati_remote->irq_urb = usb_alloc_urb(0, GFP_KERNEL);
+ if (!ati_remote->irq_urb)
+ goto error;
+
+ ati_remote->out_urb = usb_alloc_urb(0, GFP_KERNEL);
+ if (!ati_remote->out_urb)
+ goto error;
+
+ usb_make_path(udev, path, NAME_BUFSIZE);
+ sprintf(ati_remote->phys, "%s/input%d", path, ATI_INPUTNUM);
+ if (udev->descriptor.iManufacturer &&
+ (usb_string(udev, udev->descriptor.iManufacturer, buf,
+ NAME_BUFSIZE) > 0))
+ strcat(ati_remote->name, buf);
+
+ if (udev->descriptor.iProduct &&
+ (usb_string(udev, udev->descriptor.iProduct, buf, NAME_BUFSIZE) > 0))
+ sprintf(ati_remote->name, "%s %s", ati_remote->name, buf);
+
+ if (!strlen(ati_remote->name))
+ sprintf(ati_remote->name, DRIVER_DESC "(%04x,%04x)",
+ ati_remote->udev->descriptor.idVendor,
+ ati_remote->udev->descriptor.idProduct);
+
+ /* Device Hardware Initialization - fills in ati_remote->idev from udev. */
+ retval = ati_remote_initialize(ati_remote);
+ if (retval)
+ goto error;
+
+ /* Set up and register input device */
+ ati_remote_input_init(ati_remote);
+ input_register_device(&ati_remote->idev);
+
+ dev_info(&ati_remote->interface->dev, "Input registered: %s on %s\n",
+ ati_remote->name, path);
+
+ usb_set_intfdata(interface, ati_remote);
+ ati_remote->present = 1;
+ kfree(buf);
+ return 0;
+
+error:
+ if (buf)
+ kfree(buf);
+
+ ati_remote_delete(ati_remote);
+ return retval;
+}
+
+/*
+ * ati_remote_disconnect
+ */
+static void ati_remote_disconnect(struct usb_interface *interface)
+{
+ struct ati_remote *ati_remote;
+
+ ati_remote = usb_get_intfdata(interface);
+ usb_set_intfdata(interface, NULL);
+ if (!ati_remote) {
+ warn("%s - null device?\n", __FUNCTION__);
+ return;
+ }
+
+ input_unregister_device(&ati_remote->idev);
+
+ /* Mark device as unplugged */
+ ati_remote->present = 0;
+
+ /* If device is still open, ati_remote_close will call delete. */
+ if (!ati_remote->open)
+ ati_remote_delete(ati_remote);
+}
+
+/*
+ * ati_remote_init
+ */
+static int __init ati_remote_init(void)
+{
+ int result;
+
+ result = usb_register(&ati_remote_driver);
+ if (result)
+ err("usb_register error #%d\n", result);
+ else
+ info("Registered USB driver " DRIVER_DESC " v. " DRIVER_VERSION);
+
+ return result;
+}
+
+/*
+ * ati_remote_exit
+ */
+static void __exit ati_remote_exit(void)
+{
+ usb_deregister(&ati_remote_driver);
+}
+
+/*
+ * module specification
+ */
+
+module_init(ati_remote_init);
+module_exit(ati_remote_exit);
+
+MODULE_AUTHOR(DRIVER_AUTHOR);
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("GPL");
+
--- diff/drivers/usb/input/mtouchusb.c 1970-01-01 00:00:00.000000000 +0000
+++ source/drivers/usb/input/mtouchusb.c 2004-03-16 11:27:48.351837064 +0000
@@ -0,0 +1,391 @@
+/******************************************************************************
+ * mtouchusb.c -- Driver for Microtouch (Now 3M) USB Touchscreens
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ * Based upon original work by Radoslaw Garbacz (usb-support@ite.pl)
+ * (http://freshmeat.net/projects/3mtouchscreendriver)
+ *
+ * History
+ *
+ * 0.3 & 0.4 2002 (TEJ) tejohnson@yahoo.com
+ * Updated to 2.4.18, then 2.4.19
+ * Old version still relied on stealing a minor
+ *
+ * 0.5 02/26/2004 (TEJ) tejohnson@yahoo.com
+ * Complete rewrite using Linux Input in 2.6.3
+ * Unfortunately no calibration support at this time
+ *
+ *****************************************************************************/
+
+#include <linux/config.h>
+
+#ifdef CONFIG_USB_DEBUG
+ #define DEBUG
+#else
+ #undef DEBUG
+#endif
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/input.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/usb.h>
+
+#define MTOUCHUSB_MIN_XC 0xc8
+#define MTOUCHUSB_MAX_XC 0xff78
+#define MTOUCHUSB_XC_FUZZ 0x0
+#define MTOUCHUSB_XC_FLAT 0x0
+#define MTOUCHUSB_MIN_YC 0x0
+#define MTOUCHUSB_MAX_YC 0xff78
+#define MTOUCHUSB_YC_FUZZ 0x0
+#define MTOUCHUSB_YC_FLAT 0x0
+#define MTOUCHUSB_ASYC_REPORT 1
+#define MTOUCHUSB_REPORT_SIZE_DATA 11
+#define MTOUCHUSB_REQ_CTRLLR_ID 10
+
+#define MTOUCHUSB_GET_XC(data) (data[4]<<8 | data[3])
+#define MTOUCHUSB_GET_YC(data) (data[6]<<8 | data[5])
+#define MTOUCHUSB_GET_TOUCHED(data) ((data[2] & 0x40) ? 1:0)
+
+#define DRIVER_VERSION "v0.1"
+#define DRIVER_AUTHOR "Todd E. Johnson, tejohnson@yahoo.com"
+#define DRIVER_DESC "Microtouch USB HID Touchscreen Driver"
+
+struct mtouch_usb {
+ unsigned char *data;
+ dma_addr_t data_dma;
+ struct urb *irq;
+ struct usb_device *udev;
+ struct input_dev input;
+ int open;
+ char name[128];
+ char phys[64];
+};
+
+static __s32 vendor=-1, product=-1;
+
+static struct usb_device_id mtouchusb_devices [] = {
+ { USB_DEVICE(0x0596, 0x0001) }, /* 3M (Formerly MicroTouch) 14-206 */
+ { } /* Terminating entry */
+};
+
+static void mtouchusb_irq(struct urb *urb, struct pt_regs *regs)
+{
+ struct mtouch_usb *mtouch = urb->context;
+ int retval;
+
+ switch (urb->status) {
+ case 0:
+ /* success */
+ break;
+ case -ETIMEDOUT:
+ /* this urb is timing out */
+ dbg("%s - urb timed out - was the device unplugged?",
+ __FUNCTION__);
+ return;
+ case -ECONNRESET:
+ case -ENOENT:
+ case -ESHUTDOWN:
+ /* this urb is terminated, clean up */
+ dbg("%s - urb shutting down with status: %d",
+ __FUNCTION__, urb->status);
+ return;
+ default:
+ dbg("%s - nonzero urb status received: %d",
+ __FUNCTION__, urb->status);
+ goto exit;
+ }
+
+ input_regs(&mtouch->input, regs);
+ input_report_key(&mtouch->input, BTN_TOUCH,
+ MTOUCHUSB_GET_TOUCHED(mtouch->data));
+ input_report_abs(&mtouch->input, ABS_X,
+ MTOUCHUSB_GET_XC(mtouch->data));
+ input_report_abs(&mtouch->input, ABS_Y,
+ MTOUCHUSB_GET_YC(mtouch->data));
+ input_sync(&mtouch->input);
+
+exit:
+ retval = usb_submit_urb (urb, GFP_ATOMIC);
+ if (retval)
+ err ("%s - usb_submit_urb failed with result: %d",
+ __FUNCTION__, retval);
+}
+
+static int mtouchusb_open (struct input_dev *input)
+{
+ struct mtouch_usb *mtouch = input->private;
+
+ if (mtouch->open++)
+ return 0;
+
+ mtouch->irq->dev = mtouch->udev;
+
+ if (usb_submit_urb (mtouch->irq, GFP_ATOMIC))
+ return -EIO;
+
+ return 0;
+}
+
+static void mtouchusb_close (struct input_dev *input)
+{
+ struct mtouch_usb *mtouch = input->private;
+
+ if (!--mtouch->open)
+ usb_unlink_urb (mtouch->irq);
+}
+
+static int mtouchusb_alloc_buffers(struct usb_device *udev, struct mtouch_usb *mtouch)
+{
+ dbg("%s - called", __FUNCTION__);
+
+ mtouch->data = usb_buffer_alloc(udev, MTOUCHUSB_REPORT_SIZE_DATA,
+ SLAB_ATOMIC, &mtouch->data_dma);
+
+ if (!mtouch->data)
+ return -1;
+
+ return 0;
+}
+
+static void mtouchusb_free_buffers(struct usb_device *udev, struct mtouch_usb *mtouch)
+{
+ dbg("%s - called", __FUNCTION__);
+
+ if (mtouch->data)
+ usb_buffer_free(udev, MTOUCHUSB_REPORT_SIZE_DATA,
+ mtouch->data, mtouch->data_dma);
+}
+
+static int mtouchusb_probe(struct usb_interface *intf, const struct usb_device_id *id)
+{
+ struct mtouch_usb *mtouch;
+ struct usb_host_interface *interface;
+ struct usb_endpoint_descriptor *endpoint;
+ struct usb_device *udev = interface_to_usbdev (intf);
+ char path[64];
+ char *buf;
+ int nRet;
+ int ix;
+ char valid_device = 0;
+
+ dbg("%s - called", __FUNCTION__);
+ if (vendor != -1 && product != -1) {
+ info("%s - User specified USB Touch -- Vend:Prod - %x:%x",
+ __FUNCTION__, vendor, product);
+ }
+
+ for (ix = 0; ix < sizeof (mtouchusb_devices) /
+ sizeof (struct usb_device_id); ix++) {
+ if ((udev->descriptor.idVendor ==
+ mtouchusb_devices [ix].idVendor) &&
+ (udev->descriptor.idProduct ==
+ mtouchusb_devices [ix].idProduct)) {
+ valid_device = 1;
+ break;
+ }
+ }
+
+ if (udev->descriptor.idVendor == vendor &&
+ udev->descriptor.idProduct == product) { /* User specified */
+ valid_device = 1;
+ }
+
+ if (!valid_device) {
+ err("%s - No valid device!", __FUNCTION__);
+ return -EIO;
+ }
+
+ if (udev->descriptor.bNumConfigurations != 1) {
+ err("%s - Only one device configuration is supported.",
+ __FUNCTION__);
+ return -EIO;
+ }
+
+ dbg("%s - setting interface", __FUNCTION__);
+ interface = intf->cur_altsetting;
+
+ dbg("%s - setting endpoint", __FUNCTION__);
+ endpoint = &interface->endpoint[0].desc;
+
+ if (interface->desc.bNumEndpoints != 1) {
+ err("%s - Only one endpoint is supported.", __FUNCTION__);
+ return -EIO;
+ }
+
+ if (!(mtouch = kmalloc (sizeof (struct mtouch_usb), GFP_KERNEL))) {
+ err("%s - Out of memory.", __FUNCTION__);
+ return -ENOMEM;
+ }
+
+ memset(mtouch, 0, sizeof(struct mtouch_usb));
+ mtouch->udev = udev;
+
+ dbg("%s - allocating buffers", __FUNCTION__);
+ if (mtouchusb_alloc_buffers(udev, mtouch)) {
+ mtouchusb_free_buffers(udev, mtouch);
+ kfree(mtouch);
+ return -ENOMEM;
+ }
+
+ mtouch->input.private = mtouch;
+ mtouch->input.open = mtouchusb_open;
+ mtouch->input.close = mtouchusb_close;
+
+ usb_make_path(udev, path, 64);
+ sprintf(mtouch->phys, "%s/input0", path);
+
+ mtouch->input.name = mtouch->name;
+ mtouch->input.phys = mtouch->phys;
+ mtouch->input.id.bustype = BUS_USB;
+ mtouch->input.id.vendor = udev->descriptor.idVendor;
+ mtouch->input.id.product = udev->descriptor.idProduct;
+ mtouch->input.id.version = udev->descriptor.bcdDevice;
+ mtouch->input.dev = &intf->dev;
+
+ mtouch->input.evbit[0] = BIT(EV_KEY) | BIT(EV_ABS);
+ mtouch->input.absbit[0] = BIT(ABS_X) | BIT(ABS_Y);
+ mtouch->input.keybit[LONG(BTN_TOUCH)] = BIT(BTN_TOUCH);
+
+ /* Used to Scale Compensated Data and Flip Y */
+ mtouch->input.absmin[ABS_X] = MTOUCHUSB_MIN_XC;
+ mtouch->input.absmax[ABS_X] = MTOUCHUSB_MAX_XC;
+ mtouch->input.absfuzz[ABS_X] = MTOUCHUSB_XC_FUZZ;
+ mtouch->input.absflat[ABS_X] = MTOUCHUSB_XC_FLAT;
+ mtouch->input.absmin[ABS_Y] = MTOUCHUSB_MAX_YC;
+ mtouch->input.absmax[ABS_Y] = MTOUCHUSB_MIN_YC;
+ mtouch->input.absfuzz[ABS_Y] = MTOUCHUSB_YC_FUZZ;
+ mtouch->input.absflat[ABS_Y] = MTOUCHUSB_YC_FLAT;
+
+ if (!(buf = kmalloc(63, GFP_KERNEL))) {
+ kfree(mtouch);
+ return -ENOMEM;
+ }
+
+ if (udev->descriptor.iManufacturer &&
+ usb_string(udev, udev->descriptor.iManufacturer, buf, 63) > 0)
+ strcat(mtouch->name, buf);
+ if (udev->descriptor.iProduct &&
+ usb_string(udev, udev->descriptor.iProduct, buf, 63) > 0)
+ sprintf(mtouch->name, "%s %s", mtouch->name, buf);
+
+ if (!strlen(mtouch->name))
+ sprintf(mtouch->name, "USB Touchscreen %04x:%04x",
+ mtouch->input.id.vendor, mtouch->input.id.product);
+
+ kfree(buf);
+
+ nRet = usb_control_msg(mtouch->udev,
+ usb_rcvctrlpipe(udev, 0x80),
+ USB_REQ_GET_CONFIGURATION,
+ USB_DIR_IN | USB_TYPE_STANDARD | USB_RECIP_DEVICE,
+ 0,
+ 0x81,
+ NULL,
+ 0,
+ HZ * USB_CTRL_SET_TIMEOUT);
+ dbg("%s - usb_control_msg - USB_REQ_GET_CONFIGURATION - bytes|err: %d",
+ __FUNCTION__, nRet);
+
+ dbg("%s - usb_alloc_urb: mtouch->irq", __FUNCTION__);
+ mtouch->irq = usb_alloc_urb(0, GFP_KERNEL);
+ if (!mtouch->irq) {
+ dbg("%s - usb_alloc_urb failed: mtouch->irq", __FUNCTION__);
+ mtouchusb_free_buffers(udev, mtouch);
+ kfree(mtouch);
+ return -ENOMEM;
+ }
+
+ dbg("%s - usb_fill_int_urb", __FUNCTION__);
+ usb_fill_int_urb(mtouch->irq,
+ mtouch->udev,
+ usb_rcvintpipe(mtouch->udev, 0x81),
+ mtouch->data,
+ MTOUCHUSB_REPORT_SIZE_DATA,
+ mtouchusb_irq,
+ mtouch,
+ endpoint->bInterval);
+
+ dbg("%s - input_register_device", __FUNCTION__);
+ input_register_device(&mtouch->input);
+
+ nRet = usb_control_msg(mtouch->udev,
+ usb_rcvctrlpipe(udev, 0x80),
+ MTOUCHUSB_ASYC_REPORT,
+ USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+ MTOUCHUSB_ASYC_REPORT,
+ MTOUCHUSB_ASYC_REPORT,
+ NULL,
+ 0,
+ HZ * USB_CTRL_SET_TIMEOUT);
+ dbg("%s - usb_control_msg - MTOUCHUSB_ASYC_REPORT - bytes|err: %d",
+ __FUNCTION__, nRet);
+
+ printk(KERN_INFO "input: %s on %s\n", mtouch->name, path);
+ usb_set_intfdata(intf, mtouch);
+
+ return 0;
+}
+
+static void mtouchusb_disconnect(struct usb_interface *intf)
+{
+ struct mtouch_usb *mtouch = usb_get_intfdata (intf);
+
+ dbg("%s - called", __FUNCTION__);
+ usb_set_intfdata(intf, NULL);
+ if (mtouch) {
+ dbg("%s - mtouch is initialized, cleaning up", __FUNCTION__);
+ usb_unlink_urb(mtouch->irq);
+ input_unregister_device(&mtouch->input);
+ usb_free_urb(mtouch->irq);
+ mtouchusb_free_buffers(interface_to_usbdev(intf), mtouch);
+ kfree(mtouch);
+ }
+}
+
+MODULE_DEVICE_TABLE (usb, mtouchusb_devices);
+
+static struct usb_driver mtouchusb_driver = {
+ .owner = THIS_MODULE,
+ .name = "mtouchusb",
+ .probe = mtouchusb_probe,
+ .disconnect = mtouchusb_disconnect,
+ .id_table = mtouchusb_devices,
+};
+
+static int __init mtouchusb_init(void) {
+ dbg("%s - called", __FUNCTION__);
+ return usb_register(&mtouchusb_driver);
+}
+
+static void __exit mtouchusb_cleanup(void) {
+ dbg("%s - called", __FUNCTION__);
+ usb_deregister(&mtouchusb_driver);
+}
+
+module_init(mtouchusb_init);
+module_exit(mtouchusb_cleanup);
+
+MODULE_AUTHOR( DRIVER_AUTHOR );
+MODULE_DESCRIPTION( DRIVER_DESC );
+MODULE_LICENSE("GPL");
+MODULE_PARM(vendor, "i");
+MODULE_PARM_DESC(vendor, "User specified USB idVendor");
+MODULE_PARM(product, "i");
+MODULE_PARM_DESC(product, "User specified USB idProduct");
+
+
--- diff/include/asm-alpha/lockmeter.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/asm-alpha/lockmeter.h 2004-03-16 11:27:48.442823232 +0000
@@ -0,0 +1,84 @@
+/*
+ * Written by John Hawkes (hawkes@sgi.com)
+ * Based on klstat.h by Jack Steiner (steiner@sgi.com)
+ *
+ * Modified by Peter Rival (frival@zk3.dec.com)
+ */
+
+#ifndef _ALPHA_LOCKMETER_H
+#define _ALPHA_LOCKMETER_H
+
+#include <asm/hwrpb.h>
+#define CPU_CYCLE_FREQUENCY hwrpb->cycle_freq
+
+#define get_cycles64() get_cycles()
+
+#define THIS_CPU_NUMBER smp_processor_id()
+
+#include <linux/version.h>
+
+#define SPINLOCK_MAGIC_INIT /**/
+
+/*
+ * Macros to cache and retrieve an index value inside of a lock
+ * these macros assume that there are less than 65536 simultaneous
+ * (read mode) holders of a rwlock.
+ * We also assume that the hash table has less than 32767 entries.
+ * the high order bit is used for write locking a rw_lock
+ * Note: although these defines and macros are the same as what is being used
+ * in include/asm-i386/lockmeter.h, they are present here to easily
+ * allow an alternate Alpha implementation.
+ */
+/*
+ * instrumented spinlock structure -- never used to allocate storage
+ * only used in macros below to overlay a spinlock_t
+ */
+typedef struct inst_spinlock_s {
+ /* remember, Alpha is little endian */
+ unsigned short lock;
+ unsigned short index;
+} inst_spinlock_t;
+#define PUT_INDEX(lock_ptr,indexv) ((inst_spinlock_t *)(lock_ptr))->index = indexv
+#define GET_INDEX(lock_ptr) ((inst_spinlock_t *)(lock_ptr))->index
+
+/*
+ * macros to cache and retrieve an index value in a read/write lock
+ * as well as the cpu where a reader busy period started
+ * we use the 2nd word (the debug word) for this, so require the
+ * debug word to be present
+ */
+/*
+ * instrumented rwlock structure -- never used to allocate storage
+ * only used in macros below to overlay a rwlock_t
+ */
+typedef struct inst_rwlock_s {
+ volatile int lock;
+ unsigned short index;
+ unsigned short cpu;
+} inst_rwlock_t;
+#define PUT_RWINDEX(rwlock_ptr,indexv) ((inst_rwlock_t *)(rwlock_ptr))->index = indexv
+#define GET_RWINDEX(rwlock_ptr) ((inst_rwlock_t *)(rwlock_ptr))->index
+#define PUT_RW_CPU(rwlock_ptr,cpuv) ((inst_rwlock_t *)(rwlock_ptr))->cpu = cpuv
+#define GET_RW_CPU(rwlock_ptr) ((inst_rwlock_t *)(rwlock_ptr))->cpu
+
+/*
+ * return true if rwlock is write locked
+ * (note that other lock attempts can cause the lock value to be negative)
+ */
+#define RWLOCK_IS_WRITE_LOCKED(rwlock_ptr) (((inst_rwlock_t *)rwlock_ptr)->lock & 1)
+#define IABS(x) ((x) > 0 ? (x) : -(x))
+
+#define RWLOCK_READERS(rwlock_ptr) rwlock_readers(rwlock_ptr)
+extern inline int rwlock_readers(rwlock_t *rwlock_ptr)
+{
+ int tmp = (int) ((inst_rwlock_t *)rwlock_ptr)->lock;
+ /* readers subtract 2, so we have to: */
+ /* - andnot off a possible writer (bit 0) */
+ /* - get the absolute value */
+ /* - divide by 2 (right shift by one) */
+ /* to find the number of readers */
+ if (tmp == 0) return(0);
+ else return(IABS(tmp & ~1)>>1);
+}
+
+#endif /* _ALPHA_LOCKMETER_H */
--- diff/include/asm-generic/compat_signal.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/asm-generic/compat_signal.h 2004-03-16 11:27:48.446822624 +0000
@@ -0,0 +1,25 @@
+#ifndef _ASM_GENERIC_COMPAT_SIGNAL_H
+#define _ASM_GENERIC_COMPAT_SIGNAL_H
+
+#ifndef __ASSEMBLY__
+#include <linux/compat.h>
+
+typedef compat_uptr_t compat_sighandler_t;
+
+typedef struct compat_sigaltstack {
+ compat_uptr_t ss_sp;
+ compat_int_t ss_flags;
+ compat_size_t ss_size;
+} compat_stack_t;
+
+/* Most things should be clean enough to redefine this at will, if care
+ is taken to make libc match. */
+
+struct compat_sigaction {
+ compat_sighandler_t sa_handler;
+ compat_uint_t sa_flags;
+ compat_sigset_t sa_mask; /* mask last for extensibility */
+};
+
+#endif /* !__ASSEMBLY__ */
+#endif /* !_ASM_GENERIC_COMPAT_SIGNAL_H */
--- diff/include/asm-i386/1 1970-01-01 00:00:00.000000000 +0000
+++ source/include/asm-i386/1 2004-03-16 11:27:48.447822472 +0000
@@ -0,0 +1,81 @@
+--- pgtable.h.orig 2004-03-10 11:47:23.000000000 +0100
++++ pgtable.h 2004-03-10 12:31:10.000000000 +0100
+@@ -21,15 +21,25 @@
+ #include <asm/bitops.h>
+ #endif
+
+-extern pgd_t swapper_pg_dir[1024];
+-extern void paging_init(void);
++#include <linux/slab.h>
++#include <linux/list.h>
++#include <linux/spinlock.h>
+
+ /*
+ * ZERO_PAGE is a global shared page that is always zero: used
+ * for zero-mapped memory areas etc..
+ */
+-extern unsigned long empty_zero_page[1024];
+ #define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
++extern unsigned long empty_zero_page[1024];
++extern pgd_t swapper_pg_dir[1024];
++extern kmem_cache_t *pgd_cache, *pmd_cache, *kpmd_cache;
++
++void pmd_ctor(void *, kmem_cache_t *, unsigned long);
++void kpmd_ctor(void *, kmem_cache_t *, unsigned long);
++void pgd_ctor(void *, kmem_cache_t *, unsigned long);
++void pgtable_cache_init(void);
++void paging_init(void);
++void setup_identity_mappings(pgd_t *pgd_base, unsigned long start, unsigned long end);
+
+ #endif /* !__ASSEMBLY__ */
+
+@@ -39,22 +49,15 @@ extern unsigned long empty_zero_page[102
+ * newer 3-level PAE-mode page tables.
+ */
+ #ifndef __ASSEMBLY__
+-#ifdef CONFIG_X86_PAE
+-# include <asm/pgtable-3level.h>
+
+-/*
+- * Need to initialise the X86 PAE caches
+- */
+-extern void pgtable_cache_init(void);
++extern void set_system_gate(unsigned int n, void *addr);
++extern void init_entry_mappings(void);
++extern void entry_trampoline_setup(void);
+
++#ifdef CONFIG_X86_PAE
++# include <asm/pgtable-3level.h>
+ #else
+ # include <asm/pgtable-2level.h>
+-
+-/*
+- * No page table caches to initialise
+- */
+-#define pgtable_cache_init() do { } while (0)
+-
+ #endif
+ #endif
+
+@@ -63,7 +66,12 @@ extern void pgtable_cache_init(void);
+ #define PGDIR_SIZE (1UL << PGDIR_SHIFT)
+ #define PGDIR_MASK (~(PGDIR_SIZE-1))
+
+-#define USER_PTRS_PER_PGD (TASK_SIZE/PGDIR_SIZE)
++#if defined(CONFIG_X86_PAE) && defined(CONFIG_X86_4G_VM_LAYOUT)
++# define USER_PTRS_PER_PGD 4
++#else
++# define USER_PTRS_PER_PGD ((TASK_SIZE/PGDIR_SIZE) + ((TASK_SIZE % PGDIR_SIZE) + PGDIR_SIZE-1)/PGDIR_SIZE)
++#endif
++
+ #define FIRST_USER_PGD_NR 0
+
+ #define USER_PGD_PTRS (PAGE_OFFSET >> PGDIR_SHIFT)
+@@ -233,6 +241,7 @@ static inline void ptep_mkdirty(pte_t *p
+
+ #define mk_pte(page, pgprot) pfn_pte(page_to_pfn(page), (pgprot))
+ #define mk_pte_huge(entry) ((entry).pte_low |= _PAGE_PRESENT | _PAGE_PSE)
++#define mk_pte_phys(physpage, pgprot) pfn_pte((physpage) >> PAGE_SHIFT, pgprot)
+
+ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
+ {
--- diff/include/asm-i386/atomic_kmap.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/asm-i386/atomic_kmap.h 2004-03-16 11:27:48.448822320 +0000
@@ -0,0 +1,95 @@
+/*
+ * atomic_kmap.h: temporary virtual kernel memory mappings
+ *
+ * Copyright (C) 2003 Ingo Molnar <mingo@redhat.com>
+ */
+
+#ifndef _ASM_ATOMIC_KMAP_H
+#define _ASM_ATOMIC_KMAP_H
+
+#ifdef __KERNEL__
+
+#include <linux/config.h>
+#include <asm/tlbflush.h>
+
+#ifdef CONFIG_DEBUG_HIGHMEM
+#define HIGHMEM_DEBUG 1
+#else
+#define HIGHMEM_DEBUG 0
+#endif
+
+extern pte_t *kmap_pte;
+#define kmap_prot PAGE_KERNEL
+
+#define PKMAP_BASE (0xff000000UL)
+#define NR_SHARED_PMDS ((0xffffffff-PKMAP_BASE+1)/PMD_SIZE)
+
+static inline unsigned long __kmap_atomic_vaddr(enum km_type type)
+{
+ enum fixed_addresses idx;
+
+ idx = type + KM_TYPE_NR*smp_processor_id();
+ return __fix_to_virt(FIX_KMAP_BEGIN + idx);
+}
+
+static inline void *__kmap_atomic_noflush(struct page *page, enum km_type type)
+{
+ enum fixed_addresses idx;
+ unsigned long vaddr;
+
+ idx = type + KM_TYPE_NR*smp_processor_id();
+ vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
+ /*
+ * NOTE: entries that rely on some secondary TLB-flush
+ * effect must not be global:
+ */
+ set_pte(kmap_pte-idx, mk_pte(page, PAGE_KERNEL));
+
+ return (void*) vaddr;
+}
+
+static inline void *__kmap_atomic(struct page *page, enum km_type type)
+{
+ enum fixed_addresses idx;
+ unsigned long vaddr;
+
+ idx = type + KM_TYPE_NR*smp_processor_id();
+ vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
+#if HIGHMEM_DEBUG
+ BUG_ON(!pte_none(*(kmap_pte-idx)));
+#else
+ /*
+ * Performance optimization - do not flush if the new
+ * pte is the same as the old one:
+ */
+ if (pte_val(*(kmap_pte-idx)) == pte_val(mk_pte(page, kmap_prot)))
+ return (void *) vaddr;
+#endif
+ set_pte(kmap_pte-idx, mk_pte(page, kmap_prot));
+ __flush_tlb_one(vaddr);
+
+ return (void*) vaddr;
+}
+
+static inline void __kunmap_atomic(void *kvaddr, enum km_type type)
+{
+#if HIGHMEM_DEBUG
+ unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK;
+ enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id();
+
+ BUG_ON(vaddr != __fix_to_virt(FIX_KMAP_BEGIN+idx));
+ /*
+ * force other mappings to Oops if they'll try to access
+ * this pte without first remap it
+ */
+ pte_clear(kmap_pte-idx);
+ __flush_tlb_one(vaddr);
+#endif
+}
+
+#define __kunmap_atomic_type(type) \
+ __kunmap_atomic((void *)__kmap_atomic_vaddr(type), (type))
+
+#endif /* __KERNEL__ */
+
+#endif /* _ASM_ATOMIC_KMAP_H */
--- diff/include/asm-i386/kgdb.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/asm-i386/kgdb.h 2004-03-16 11:27:48.451821864 +0000
@@ -0,0 +1,69 @@
+#ifndef __KGDB
+#define __KGDB
+
+/*
+ * This file should not include ANY others. This makes it usable
+ * most anywhere without the fear of include order or inclusion.
+ * Make it so!
+ *
+ * This file may be included all the time. It is only active if
+ * CONFIG_KGDB is defined, otherwise it stubs out all the macros
+ * and entry points.
+ */
+#if defined(CONFIG_KGDB) && !defined(__ASSEMBLY__)
+
+extern void breakpoint(void);
+#define INIT_KGDB_INTS kgdb_enable_ints()
+
+#ifndef BREAKPOINT
+#define BREAKPOINT asm(" int $3")
+#endif
+
+extern void kgdb_schedule_breakpoint(void);
+extern void kgdb_process_breakpoint(void);
+
+extern int kgdb_tty_hook(void);
+extern int kgdb_eth_hook(void);
+extern int kgdboe;
+
+/*
+ * GDB debug stub (or any debug stub) can point the 'linux_debug_hook'
+ * pointer to its routine and it will be entered as the first thing
+ * when a trap occurs.
+ *
+ * Return values are, at present, undefined.
+ *
+ * The debug hook routine does not necessarily return to its caller.
+ * It has the register image and thus may choose to resume execution
+ * anywhere it pleases.
+ */
+struct pt_regs;
+
+extern int kgdb_handle_exception(int trapno,
+ int signo, int err_code, struct pt_regs *regs);
+extern int in_kgdb(struct pt_regs *regs);
+
+#ifdef CONFIG_KGDB_TS
+void kgdb_tstamp(int line, char *source, int data0, int data1);
+/*
+ * This is the time stamp function. The macro adds the source info and
+ * does a cast on the data to allow most any 32-bit value.
+ */
+
+#define kgdb_ts(data0,data1) kgdb_tstamp(__LINE__,__FILE__,(int)data0,(int)data1)
+#else
+#define kgdb_ts(data0,data1)
+#endif
+#else /* CONFIG_KGDB && ! __ASSEMBLY__ ,stubs follow... */
+#ifndef BREAKPOINT
+#define BREAKPOINT
+#endif
+#define kgdb_ts(data0,data1)
+#define in_kgdb
+#define kgdb_handle_exception
+#define breakpoint
+#define INIT_KGDB_INTS
+#define kgdb_process_breakpoint() do {} while(0)
+
+#endif
+#endif /* __KGDB */
--- diff/include/asm-i386/kgdb_local.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/asm-i386/kgdb_local.h 2004-03-16 11:27:48.484816848 +0000
@@ -0,0 +1,102 @@
+#ifndef __KGDB_LOCAL
+#define ___KGDB_LOCAL
+#include <linux/config.h>
+#include <linux/types.h>
+#include <linux/serial.h>
+#include <linux/serialP.h>
+#include <linux/spinlock.h>
+#include <asm/processor.h>
+#include <asm/msr.h>
+#include <asm/kgdb.h>
+
+#define PORT 0x3f8
+#ifdef CONFIG_KGDB_PORT
+#undef PORT
+#define PORT CONFIG_KGDB_PORT
+#endif
+#define IRQ 4
+#ifdef CONFIG_KGDB_IRQ
+#undef IRQ
+#define IRQ CONFIG_KGDB_IRQ
+#endif
+#define SB_CLOCK 1843200
+#define SB_BASE (SB_CLOCK/16)
+#define SB_BAUD9600 SB_BASE/9600
+#define SB_BAUD192 SB_BASE/19200
+#define SB_BAUD384 SB_BASE/38400
+#define SB_BAUD576 SB_BASE/57600
+#define SB_BAUD1152 SB_BASE/115200
+#ifdef CONFIG_KGDB_9600BAUD
+#define SB_BAUD SB_BAUD9600
+#endif
+#ifdef CONFIG_KGDB_19200BAUD
+#define SB_BAUD SB_BAUD192
+#endif
+#ifdef CONFIG_KGDB_38400BAUD
+#define SB_BAUD SB_BAUD384
+#endif
+#ifdef CONFIG_KGDB_57600BAUD
+#define SB_BAUD SB_BAUD576
+#endif
+#ifdef CONFIG_KGDB_115200BAUD
+#define SB_BAUD SB_BAUD1152
+#endif
+#ifndef SB_BAUD
+#define SB_BAUD SB_BAUD1152 /* Start with this if not given */
+#endif
+
+#ifndef CONFIG_X86_TSC
+#undef rdtsc
+#define rdtsc(a,b) if (a++ > 10000){a = 0; b++;}
+#undef rdtscll
+#define rdtscll(s) s++
+#endif
+
+#ifdef _raw_read_unlock /* must use a name that is "define"ed, not an inline */
+#undef spin_lock
+#undef spin_trylock
+#undef spin_unlock
+#define spin_lock _raw_spin_lock
+#define spin_trylock _raw_spin_trylock
+#define spin_unlock _raw_spin_unlock
+#else
+#endif
+#undef spin_unlock_wait
+#define spin_unlock_wait(x) do { cpu_relax(); barrier();} \
+ while(spin_is_locked(x))
+
+#define SB_IER 1
+#define SB_MCR UART_MCR_OUT2 | UART_MCR_DTR | UART_MCR_RTS
+
+#define FLAGS 0
+#define SB_STATE { \
+ magic: SSTATE_MAGIC, \
+ baud_base: SB_BASE, \
+ port: PORT, \
+ irq: IRQ, \
+ flags: FLAGS, \
+ custom_divisor:SB_BAUD}
+#define SB_INFO { \
+ magic: SERIAL_MAGIC, \
+ port: PORT,0,FLAGS, \
+ state: &state, \
+ tty: (struct tty_struct *)&state, \
+ IER: SB_IER, \
+ MCR: SB_MCR}
+extern void putDebugChar(int);
+/* RTAI support needs us to really stop/start interrupts */
+
+#define kgdb_sti() __asm__ __volatile__("sti": : :"memory")
+#define kgdb_cli() __asm__ __volatile__("cli": : :"memory")
+#define kgdb_local_save_flags(x) __asm__ __volatile__(\
+ "pushfl ; popl %0":"=g" (x): /* no input */)
+#define kgdb_local_irq_restore(x) __asm__ __volatile__(\
+ "pushl %0 ; popfl": \
+ /* no output */ :"g" (x):"memory", "cc")
+#define kgdb_local_irq_save(x) kgdb_local_save_flags(x); kgdb_cli()
+
+#ifdef CONFIG_SERIAL
+extern void shutdown_for_kgdb(struct async_struct *info);
+#endif
+#define INIT_KDEBUG putDebugChar("+");
+#endif /* __KGDB_LOCAL */
--- diff/include/asm-i386/lockmeter.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/asm-i386/lockmeter.h 2004-03-16 11:27:48.485816696 +0000
@@ -0,0 +1,115 @@
+/*
+ * Copyright (C) 1999,2000 Silicon Graphics, Inc.
+ *
+ * Written by John Hawkes (hawkes@sgi.com)
+ * Based on klstat.h by Jack Steiner (steiner@sgi.com)
+ *
+ * Modified by Ray Bryant (raybry@us.ibm.com)
+ * Changes Copyright (C) 2000 IBM, Inc.
+ * Added save of index in spinlock_t to improve efficiency
+ * of "hold" time reporting for spinlocks.
+ * Added support for hold time statistics for read and write
+ * locks.
+ * Moved machine dependent code here from include/lockmeter.h.
+ *
+ */
+
+#ifndef _I386_LOCKMETER_H
+#define _I386_LOCKMETER_H
+
+#include <asm/spinlock.h>
+#include <asm/rwlock.h>
+
+#include <linux/version.h>
+
+#ifdef __KERNEL__
+extern unsigned long cpu_khz;
+#define CPU_CYCLE_FREQUENCY (cpu_khz * 1000)
+#else
+#define CPU_CYCLE_FREQUENCY 450000000
+#endif
+
+#define THIS_CPU_NUMBER smp_processor_id()
+
+/*
+ * macros to cache and retrieve an index value inside of a spin lock
+ * these macros assume that there are less than 65536 simultaneous
+ * (read mode) holders of a rwlock. Not normally a problem!!
+ * we also assume that the hash table has less than 65535 entries.
+ */
+/*
+ * instrumented spinlock structure -- never used to allocate storage
+ * only used in macros below to overlay a spinlock_t
+ */
+typedef struct inst_spinlock_s {
+ /* remember, Intel is little endian */
+ unsigned short lock;
+ unsigned short index;
+} inst_spinlock_t;
+#define PUT_INDEX(lock_ptr,indexv) ((inst_spinlock_t *)(lock_ptr))->index = indexv
+#define GET_INDEX(lock_ptr) ((inst_spinlock_t *)(lock_ptr))->index
+
+/*
+ * macros to cache and retrieve an index value in a read/write lock
+ * as well as the cpu where a reader busy period started
+ * we use the 2nd word (the debug word) for this, so require the
+ * debug word to be present
+ */
+/*
+ * instrumented rwlock structure -- never used to allocate storage
+ * only used in macros below to overlay a rwlock_t
+ */
+typedef struct inst_rwlock_s {
+ volatile int lock;
+ unsigned short index;
+ unsigned short cpu;
+} inst_rwlock_t;
+#define PUT_RWINDEX(rwlock_ptr,indexv) ((inst_rwlock_t *)(rwlock_ptr))->index = indexv
+#define GET_RWINDEX(rwlock_ptr) ((inst_rwlock_t *)(rwlock_ptr))->index
+#define PUT_RW_CPU(rwlock_ptr,cpuv) ((inst_rwlock_t *)(rwlock_ptr))->cpu = cpuv
+#define GET_RW_CPU(rwlock_ptr) ((inst_rwlock_t *)(rwlock_ptr))->cpu
+
+/*
+ * return the number of readers for a rwlock_t
+ */
+#define RWLOCK_READERS(rwlock_ptr) rwlock_readers(rwlock_ptr)
+
+extern inline int rwlock_readers(rwlock_t *rwlock_ptr)
+{
+ int tmp = (int) rwlock_ptr->lock;
+ /* read and write lock attempts may cause the lock value to temporarily */
+ /* be negative. Until it is >= 0 we know nothing (i. e. can't tell if */
+ /* is -1 because it was write locked and somebody tried to read lock it */
+ /* or if it is -1 because it was read locked and somebody tried to write*/
+ /* lock it. ........................................................... */
+ do {
+ tmp = (int) rwlock_ptr->lock;
+ } while (tmp < 0);
+ if (tmp == 0) return(0);
+ else return(RW_LOCK_BIAS-tmp);
+}
+
+/*
+ * return true if rwlock is write locked
+ * (note that other lock attempts can cause the lock value to be negative)
+ */
+#define RWLOCK_IS_WRITE_LOCKED(rwlock_ptr) ((rwlock_ptr)->lock <= 0)
+#define IABS(x) ((x) > 0 ? (x) : -(x))
+#define RWLOCK_IS_READ_LOCKED(rwlock_ptr) ((IABS((rwlock_ptr)->lock) % RW_LOCK_BIAS) != 0)
+
+/* this is a lot of typing just to get gcc to emit "rdtsc" */
+static inline long long get_cycles64 (void)
+{
+ union longlong_u {
+ long long intlong;
+ struct intint_s {
+ uint32_t eax;
+ uint32_t edx;
+ } intint;
+ } longlong;
+
+ rdtsc(longlong.intint.eax,longlong.intint.edx);
+ return longlong.intlong;
+}
+
+#endif /* _I386_LOCKMETER_H */
--- diff/include/asm-ia64/lockmeter.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/asm-ia64/lockmeter.h 2004-03-16 11:27:48.498814720 +0000
@@ -0,0 +1,72 @@
+/*
+ * Copyright (C) 1999,2000 Silicon Graphics, Inc.
+ *
+ * Written by John Hawkes (hawkes@sgi.com)
+ * Based on klstat.h by Jack Steiner (steiner@sgi.com)
+ */
+
+#ifndef _IA64_LOCKMETER_H
+#define _IA64_LOCKMETER_H
+
+#ifdef local_cpu_data
+#define CPU_CYCLE_FREQUENCY local_cpu_data->itc_freq
+#else
+#define CPU_CYCLE_FREQUENCY my_cpu_data.itc_freq
+#endif
+#define get_cycles64() get_cycles()
+
+#define THIS_CPU_NUMBER smp_processor_id()
+
+/*
+ * macros to cache and retrieve an index value inside of a lock
+ * these macros assume that there are less than 65536 simultaneous
+ * (read mode) holders of a rwlock.
+ * we also assume that the hash table has less than 32767 entries.
+ */
+/*
+ * instrumented spinlock structure -- never used to allocate storage
+ * only used in macros below to overlay a spinlock_t
+ */
+typedef struct inst_spinlock_s {
+ /* remember, Intel is little endian */
+ volatile unsigned short lock;
+ volatile unsigned short index;
+} inst_spinlock_t;
+#define PUT_INDEX(lock_ptr,indexv) ((inst_spinlock_t *)(lock_ptr))->index = indexv
+#define GET_INDEX(lock_ptr) ((inst_spinlock_t *)(lock_ptr))->index
+
+/*
+ * macros to cache and retrieve an index value in a read/write lock
+ * as well as the cpu where a reader busy period started
+ * we use the 2nd word (the debug word) for this, so require the
+ * debug word to be present
+ */
+/*
+ * instrumented rwlock structure -- never used to allocate storage
+ * only used in macros below to overlay a rwlock_t
+ */
+typedef struct inst_rwlock_s {
+ volatile int read_counter:31;
+ volatile int write_lock:1;
+ volatile unsigned short index;
+ volatile unsigned short cpu;
+} inst_rwlock_t;
+#define PUT_RWINDEX(rwlock_ptr,indexv) ((inst_rwlock_t *)(rwlock_ptr))->index = indexv
+#define GET_RWINDEX(rwlock_ptr) ((inst_rwlock_t *)(rwlock_ptr))->index
+#define PUT_RW_CPU(rwlock_ptr,cpuv) ((inst_rwlock_t *)(rwlock_ptr))->cpu = cpuv
+#define GET_RW_CPU(rwlock_ptr) ((inst_rwlock_t *)(rwlock_ptr))->cpu
+
+/*
+ * return the number of readers for a rwlock_t
+ */
+#define RWLOCK_READERS(rwlock_ptr) ((rwlock_ptr)->read_counter)
+
+/*
+ * return true if rwlock is write locked
+ * (note that other lock attempts can cause the lock value to be negative)
+ */
+#define RWLOCK_IS_WRITE_LOCKED(rwlock_ptr) ((rwlock_ptr)->write_lock)
+#define RWLOCK_IS_READ_LOCKED(rwlock_ptr) ((rwlock_ptr)->read_counter)
+
+#endif /* _IA64_LOCKMETER_H */
+
--- diff/include/asm-mips/lockmeter.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/asm-mips/lockmeter.h 2004-03-16 11:27:48.502814112 +0000
@@ -0,0 +1,126 @@
+/*
+ * Copyright (C) 1999,2000 Silicon Graphics, Inc.
+ *
+ * Written by John Hawkes (hawkes@sgi.com)
+ * Based on klstat.h by Jack Steiner (steiner@sgi.com)
+ * Ported to mips32 for Asita Technologies
+ * by D.J. Barrow ( dj.barrow@asitatechnologies.com )
+ */
+#ifndef _ASM_LOCKMETER_H
+#define _ASM_LOCKMETER_H
+
+/* do_gettimeoffset is a function pointer on mips */
+/* & it is not included by <linux/time.h> */
+#include <asm/time.h>
+#include <linux/time.h>
+#include <asm/div64.h>
+
+#define SPINLOCK_MAGIC_INIT /* */
+
+#define CPU_CYCLE_FREQUENCY get_cpu_cycle_frequency()
+
+#define THIS_CPU_NUMBER smp_processor_id()
+
+static uint32_t cpu_cycle_frequency = 0;
+
+static uint32_t get_cpu_cycle_frequency(void)
+{
+ /* a total hack, slow and invasive, but ... it works */
+ int sec;
+ uint32_t start_cycles;
+ struct timeval tv;
+
+ if (cpu_cycle_frequency == 0) { /* uninitialized */
+ do_gettimeofday(&tv);
+ sec = tv.tv_sec; /* set up to catch the tv_sec rollover */
+ while (sec == tv.tv_sec) { do_gettimeofday(&tv); }
+ sec = tv.tv_sec; /* rolled over to a new sec value */
+ start_cycles = get_cycles();
+ while (sec == tv.tv_sec) { do_gettimeofday(&tv); }
+ cpu_cycle_frequency = get_cycles() - start_cycles;
+ }
+
+ return cpu_cycle_frequency;
+}
+
+extern struct timeval xtime;
+
+static uint64_t get_cycles64(void)
+{
+ static uint64_t last_get_cycles64 = 0;
+ uint64_t ret;
+ unsigned long sec;
+ unsigned long usec, usec_offset;
+
+again:
+ sec = xtime.tv_sec;
+ usec = xtime.tv_usec;
+ usec_offset = do_gettimeoffset();
+ if ((xtime.tv_sec != sec) ||
+ (xtime.tv_usec != usec)||
+ (usec_offset >= 20000))
+ goto again;
+
+ ret = ((uint64_t)(usec + usec_offset) * cpu_cycle_frequency);
+ /* We can't do a normal 64 bit division on mips without libgcc.a */
+ do_div(ret,1000000);
+ ret += ((uint64_t)sec * cpu_cycle_frequency);
+
+ /* XXX why does time go backwards? do_gettimeoffset? general time adj? */
+ if (ret <= last_get_cycles64)
+ ret = last_get_cycles64+1;
+ last_get_cycles64 = ret;
+
+ return ret;
+}
+
+/*
+ * macros to cache and retrieve an index value inside of a lock
+ * these macros assume that there are less than 65536 simultaneous
+ * (read mode) holders of a rwlock.
+ * we also assume that the hash table has less than 32767 entries.
+ * the high order bit is used for write locking a rw_lock
+ */
+#define INDEX_MASK 0x7FFF0000
+#define READERS_MASK 0x0000FFFF
+#define INDEX_SHIFT 16
+#define PUT_INDEX(lockp,index) \
+ lockp->lock = (((lockp->lock) & ~INDEX_MASK) | (index) << INDEX_SHIFT)
+#define GET_INDEX(lockp) \
+ (((lockp->lock) & INDEX_MASK) >> INDEX_SHIFT)
+
+/*
+ * macros to cache and retrieve an index value in a read/write lock
+ * as well as the cpu where a reader busy period started
+ * we use the 2nd word (the debug word) for this, so require the
+ * debug word to be present
+ */
+/*
+ * instrumented rwlock structure -- never used to allocate storage
+ * only used in macros below to overlay a rwlock_t
+ */
+typedef struct inst_rwlock_s {
+ volatile int lock;
+ unsigned short index;
+ unsigned short cpu;
+} inst_rwlock_t;
+#define PUT_RWINDEX(rwlock_ptr,indexv) ((inst_rwlock_t *)(rwlock_ptr))->index = indexv
+#define GET_RWINDEX(rwlock_ptr) ((inst_rwlock_t *)(rwlock_ptr))->index
+#define PUT_RW_CPU(rwlock_ptr,cpuv) ((inst_rwlock_t *)(rwlock_ptr))->cpu = cpuv
+#define GET_RW_CPU(rwlock_ptr) ((inst_rwlock_t *)(rwlock_ptr))->cpu
+
+/*
+ * return the number of readers for a rwlock_t
+ */
+#define RWLOCK_READERS(rwlock_ptr) rwlock_readers(rwlock_ptr)
+
+extern inline int rwlock_readers(rwlock_t *rwlock_ptr)
+{
+ int tmp = (int) rwlock_ptr->lock;
+ return (tmp >= 0) ? tmp : 0;
+}
+
+#define RWLOCK_IS_WRITE_LOCKED(rwlock_ptr) ((rwlock_ptr)->lock < 0)
+#define RWLOCK_IS_READ_LOCKED(rwlock_ptr) ((rwlock_ptr)->lock > 0)
+
+#endif /* _ASM_LOCKMETER_H */
--- diff/include/asm-sparc64/lockmeter.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/asm-sparc64/lockmeter.h 2004-03-16 11:27:48.517811832 +0000
@@ -0,0 +1,45 @@
+/*
+ * Copyright (C) 2000 Anton Blanchard (anton@linuxcare.com)
+ * Copyright (C) 2003 David S. Miller (davem@redhat.com)
+ */
+
+#ifndef _SPARC64_LOCKMETER_H
+#define _SPARC64_LOCKMETER_H
+
+#include <linux/smp.h>
+#include <asm/spinlock.h>
+#include <asm/timer.h>
+#include <asm/timex.h>
+
+/* Actually, this is not the CPU frequency by the system tick
+ * frequency which is good enough for lock metering.
+ */
+#define CPU_CYCLE_FREQUENCY (timer_tick_offset * HZ)
+#define THIS_CPU_NUMBER smp_processor_id()
+
+#define PUT_INDEX(lock_ptr,indexv) (lock_ptr)->index = (indexv)
+#define GET_INDEX(lock_ptr) (lock_ptr)->index
+
+#define PUT_RWINDEX(rwlock_ptr,indexv) (rwlock_ptr)->index = (indexv)
+#define GET_RWINDEX(rwlock_ptr) (rwlock_ptr)->index
+#define PUT_RW_CPU(rwlock_ptr,cpuv) (rwlock_ptr)->cpu = (cpuv)
+#define GET_RW_CPU(rwlock_ptr) (rwlock_ptr)->cpu
+
+#define RWLOCK_READERS(rwlock_ptr) rwlock_readers(rwlock_ptr)
+
+extern inline int rwlock_readers(rwlock_t *rwlock_ptr)
+{
+ signed int tmp = rwlock_ptr->lock;
+
+ if (tmp > 0)
+ return tmp;
+ else
+ return 0;
+}
+
+#define RWLOCK_IS_WRITE_LOCKED(rwlock_ptr) ((signed int)((rwlock_ptr)->lock) < 0)
+#define RWLOCK_IS_READ_LOCKED(rwlock_ptr) ((signed int)((rwlock_ptr)->lock) > 0)
+
+#define get_cycles64() get_cycles()
+
+#endif /* _SPARC64_LOCKMETER_H */
--- diff/include/asm-x86_64/kgdb.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/asm-x86_64/kgdb.h 2004-03-16 11:27:48.523810920 +0000
@@ -0,0 +1,71 @@
+#ifndef __KGDB
+#define __KGDB
+
+/*
+ * This file should not include ANY others. This makes it usable
+ * most anywhere without the fear of include order or inclusion.
+ * Make it so!
+ *
+ * This file may be included all the time. It is only active if
+ * CONFIG_KGDB is defined, otherwise it stubs out all the macros
+ * and entry points.
+ */
+#if defined(CONFIG_KGDB) && !defined(__ASSEMBLY__)
+
+extern void breakpoint(void);
+#define INIT_KGDB_INTS kgdb_enable_ints()
+
+#ifndef BREAKPOINT
+#define BREAKPOINT asm(" int $3")
+#endif
+
+extern void kgdb_schedule_breakpoint(void);
+extern void kgdb_process_breakpoint(void);
+
+extern int kgdb_tty_hook(void);
+extern int kgdb_eth_hook(void);
+extern int kgdboe;
+
+/*
+ * GDB debug stub (or any debug stub) can point the 'linux_debug_hook'
+ * pointer to its routine and it will be entered as the first thing
+ * when a trap occurs.
+ *
+ * Return values are, at present, undefined.
+ *
+ * The debug hook routine does not necessarily return to its caller.
+ * It has the register image and thus may choose to resume execution
+ * anywhere it pleases.
+ */
+struct pt_regs;
+
+extern int kgdb_handle_exception(int trapno,
+ int signo, int err_code, struct pt_regs *regs);
+extern int in_kgdb(struct pt_regs *regs);
+
+extern void set_debug_traps(void);
+
+#ifdef CONFIG_KGDB_TS
+void kgdb_tstamp(int line, char *source, int data0, int data1);
+/*
+ * This is the time stamp function. The macro adds the source info and
+ * does a cast on the data to allow most any 32-bit value.
+ */
+
+#define kgdb_ts(data0,data1) kgdb_tstamp(__LINE__,__FILE__,(int)data0,(int)data1)
+#else
+#define kgdb_ts(data0,data1)
+#endif
+#else /* CONFIG_KGDB && ! __ASSEMBLY__ ,stubs follow... */
+#ifndef BREAKPOINT
+#define BREAKPOINT
+#endif
+#define kgdb_ts(data0,data1)
+#define in_kgdb (0)
+#define kgdb_handle_exception
+#define breakpoint
+#define INIT_KGDB_INTS
+#define kgdb_process_breakpoint() do {} while(0)
+
+#endif
+#endif /* __KGDB */
--- diff/include/asm-x86_64/kgdb_local.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/asm-x86_64/kgdb_local.h 2004-03-16 11:27:48.524810768 +0000
@@ -0,0 +1,102 @@
+#ifndef __KGDB_LOCAL
+#define ___KGDB_LOCAL
+#include <linux/config.h>
+#include <linux/types.h>
+#include <linux/serial.h>
+#include <linux/serialP.h>
+#include <linux/spinlock.h>
+#include <asm/processor.h>
+#include <asm/msr.h>
+#include <asm/kgdb.h>
+
+#define PORT 0x3f8
+#ifdef CONFIG_KGDB_PORT
+#undef PORT
+#define PORT CONFIG_KGDB_PORT
+#endif
+#define IRQ 4
+#ifdef CONFIG_KGDB_IRQ
+#undef IRQ
+#define IRQ CONFIG_KGDB_IRQ
+#endif
+#define SB_CLOCK 1843200
+#define SB_BASE (SB_CLOCK/16)
+#define SB_BAUD9600 SB_BASE/9600
+#define SB_BAUD192 SB_BASE/19200
+#define SB_BAUD384 SB_BASE/38400
+#define SB_BAUD576 SB_BASE/57600
+#define SB_BAUD1152 SB_BASE/115200
+#ifdef CONFIG_KGDB_9600BAUD
+#define SB_BAUD SB_BAUD9600
+#endif
+#ifdef CONFIG_KGDB_19200BAUD
+#define SB_BAUD SB_BAUD192
+#endif
+#ifdef CONFIG_KGDB_38400BAUD
+#define SB_BAUD SB_BAUD384
+#endif
+#ifdef CONFIG_KGDB_57600BAUD
+#define SB_BAUD SB_BAUD576
+#endif
+#ifdef CONFIG_KGDB_115200BAUD
+#define SB_BAUD SB_BAUD1152
+#endif
+#ifndef SB_BAUD
+#define SB_BAUD SB_BAUD1152 /* Start with this if not given */
+#endif
+
+#ifndef CONFIG_X86_TSC
+#undef rdtsc
+#define rdtsc(a,b) if (a++ > 10000){a = 0; b++;}
+#undef rdtscll
+#define rdtscll(s) s++
+#endif
+
+#ifdef _raw_read_unlock /* must use a name that is "define"ed, not an inline */
+#undef spin_lock
+#undef spin_trylock
+#undef spin_unlock
+#define spin_lock _raw_spin_lock
+#define spin_trylock _raw_spin_trylock
+#define spin_unlock _raw_spin_unlock
+#else
+#endif
+#undef spin_unlock_wait
+#define spin_unlock_wait(x) do { cpu_relax(); barrier();} \
+ while(spin_is_locked(x))
+
+#define SB_IER 1
+#define SB_MCR UART_MCR_OUT2 | UART_MCR_DTR | UART_MCR_RTS
+
+#define FLAGS 0
+#define SB_STATE { \
+ magic: SSTATE_MAGIC, \
+ baud_base: SB_BASE, \
+ port: PORT, \
+ irq: IRQ, \
+ flags: FLAGS, \
+ custom_divisor:SB_BAUD}
+#define SB_INFO { \
+ magic: SERIAL_MAGIC, \
+ port: PORT,0,FLAGS, \
+ state: &state, \
+ tty: (struct tty_struct *)&state, \
+ IER: SB_IER, \
+ MCR: SB_MCR}
+extern void putDebugChar(int);
+/* RTAI support needs us to really stop/start interrupts */
+
+#define kgdb_sti() __asm__ __volatile__("sti": : :"memory")
+#define kgdb_cli() __asm__ __volatile__("cli": : :"memory")
+#define kgdb_local_save_flags(x) __asm__ __volatile__(\
+ "pushfl ; popl %0":"=g" (x): /* no input */)
+#define kgdb_local_irq_restore(x) __asm__ __volatile__(\
+ "pushl %0 ; popfl": \
+ /* no output */ :"g" (x):"memory", "cc")
+#define kgdb_local_irq_save(x) kgdb_local_save_flags(x); kgdb_cli()
+
+#ifdef CONFIG_SERIAL
+extern void shutdown_for_kgdb(struct async_struct *info);
+#endif
+#define INIT_KDEBUG putDebugChar("+");
+#endif /* __KGDB_LOCAL */
--- diff/include/linux/audit.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/linux/audit.h 2004-03-16 11:27:48.528810160 +0000
@@ -0,0 +1,202 @@
+/* audit.h -- Auditing support -*- linux-c -*-
+ *
+ * Copyright 2003-2004 Red Hat Inc., Durham, North Carolina.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Written by Rickard E. (Rik) Faith <faith@redhat.com>
+ *
+ */
+
+#ifndef _LINUX_AUDIT_H_
+#define _LINUX_AUDIT_H_
+
+/* Request and reply types */
+#define AUDIT_GET 1000 /* Get status */
+#define AUDIT_SET 1001 /* Set status (enable/disable/auditd) */
+#define AUDIT_LIST 1002 /* List filtering rules */
+#define AUDIT_ADD 1003 /* Add filtering rule */
+#define AUDIT_DEL 1004 /* Delete filtering rule */
+#define AUDIT_USER 1005 /* Send a message from user-space */
+#define AUDIT_LOGIN 1006 /* Define the login id and informaiton */
+#define AUDIT_KERNEL 2000 /* Asynchronous audit record. NOT A REQUEST. */
+
+/* Rule flags */
+#define AUDIT_PER_TASK 0x01 /* Apply rule at task creation (not syscall) */
+#define AUDIT_AT_ENTRY 0x02 /* Apply rule at syscall entry */
+#define AUDIT_AT_EXIT 0x04 /* Apply rule at syscall exit */
+#define AUDIT_PREPEND 0x10 /* Prepend to front of list */
+
+/* Rule actions */
+#define AUDIT_NEVER 0 /* Do not build context if rule matches */
+#define AUDIT_POSSIBLE 1 /* Build context if rule matches */
+#define AUDIT_ALWAYS 2 /* Generate audit record if rule matches */
+
+/* Rule structure sizes -- if these change, different AUDIT_ADD and
+ * AUDIT_LIST commands must be implemented. */
+#define AUDIT_MAX_FIELDS 64
+#define AUDIT_BITMASK_SIZE 64
+#define AUDIT_WORD(nr) ((__u32)((nr)/32))
+#define AUDIT_BIT(nr) (1 << ((nr) - AUDIT_WORD(nr)*32))
+
+/* Rule fields */
+ /* These are useful when checking the
+ * task structure at task creation time
+ * (AUDIT_PER_TASK). */
+#define AUDIT_PID 0
+#define AUDIT_UID 1
+#define AUDIT_EUID 2
+#define AUDIT_SUID 3
+#define AUDIT_FSUID 4
+#define AUDIT_GID 5
+#define AUDIT_EGID 6
+#define AUDIT_SGID 7
+#define AUDIT_FSGID 8
+#define AUDIT_LOGINUID 9
+
+ /* These are ONLY useful when checking
+ * at syscall exit time (AUDIT_AT_EXIT). */
+#define AUDIT_DEVMAJOR 100
+#define AUDIT_DEVMINOR 101
+#define AUDIT_INODE 102
+#define AUDIT_EXIT 103
+#define AUDIT_NOTEXIT 104
+
+
+/* Status symbols */
+ /* Mask values */
+#define AUDIT_STATUS_ENABLED 0x0001
+#define AUDIT_STATUS_FAILURE 0x0002
+#define AUDIT_STATUS_PID 0x0004
+#define AUDIT_STATUS_RATE_LIMIT 0x0008
+#define AUDIT_STATUS_BACKLOG_LIMIT 0x0010
+ /* Failure-to-log actions */
+#define AUDIT_FAIL_SILENT 0
+#define AUDIT_FAIL_PRINTK 1
+#define AUDIT_FAIL_PANIC 2
+
+#ifndef __KERNEL__
+struct audit_message {
+ struct nlmsghdr nlh;
+ char data[1200];
+};
+#endif
+
+struct audit_status {
+ __u32 mask; /* Bit mask for valid entries */
+ __u32 enabled; /* 1 = enabled, 0 = disbaled */
+ __u32 failure; /* Failure-to-log action */
+ __u32 pid; /* pid of auditd process */
+ __u32 rate_limit; /* messages rate limit (per second) */
+ __u32 backlog_limit; /* waiting messages limit */
+ __u32 lost; /* messages lost */
+ __u32 backlog; /* messages waiting in queue */
+};
+
+struct audit_login {
+ __u32 loginuid;
+ int msglen;
+ char msg[1024];
+};
+
+struct audit_rule { /* for AUDIT_LIST, AUDIT_ADD, and AUDIT_DEL */
+ __u32 flags; /* AUDIT_PER_{TASK,CALL}, AUDIT_PREPEND */
+ __u32 action; /* AUDIT_NEVER, AUDIT_POSSIBLE, AUDIT_ALWAYS */
+ __u32 field_count;
+ __u32 mask[AUDIT_BITMASK_SIZE];
+ __u32 fields[AUDIT_MAX_FIELDS];
+ __u32 values[AUDIT_MAX_FIELDS];
+};
+
+#ifdef __KERNEL__
+
+#ifdef CONFIG_AUDIT
+struct audit_buffer;
+struct audit_context;
+#endif
+
+#ifdef CONFIG_AUDITSYSCALL
+/* These are defined in auditsc.c */
+ /* Public API */
+extern int audit_alloc(struct task_struct *task);
+extern void audit_free(struct task_struct *task);
+extern void audit_syscall_entry(struct task_struct *task,
+ int major, int minor);
+extern void audit_syscall_exit(struct task_struct *task);
+extern void audit_getname(const char *name);
+extern void audit_putname(const char *name);
+extern void audit_inode(const char *name, unsigned long ino, dev_t rdev);
+
+ /* Private API (for audit.c only) */
+extern int audit_receive_filter(int type, int pid, int uid, int seq,
+ void *data);
+extern void audit_get_stamp(struct audit_context *ctx,
+ struct timespec *t, int *serial);
+extern int audit_set_loginuid(struct audit_context *ctx, uid_t loginuid);
+#else
+#define audit_alloc(t) ({ 0; })
+#define audit_free(t) do { ; } while (0)
+#define audit_syscall_entry(t,a,b) do { ; } while (0)
+#define audit_syscall_exit(t) do { ; } while (0)
+#define audit_getname(n) do { ; } while (0)
+#define audit_putname(n) do { ; } while (0)
+#define audit_inode(n,i,d) do { ; } while (0)
+#endif
+
+#ifdef CONFIG_AUDIT
+/* These are defined in audit.c */
+ /* Public API */
+extern void audit_log(struct audit_context *ctx,
+ const char *fmt, ...)
+ __attribute__((format(printf,2,3)));
+
+extern struct audit_buffer *audit_log_start(struct audit_context *ctx);
+extern void audit_log_format(struct audit_buffer *ab,
+ const char *fmt, ...)
+ __attribute__((format(printf,2,3)));
+extern void audit_log_end(struct audit_buffer *ab);
+extern void audit_log_end_fast(struct audit_buffer *ab);
+extern void audit_log_end_irq(struct audit_buffer *ab);
+extern void audit_log_d_path(struct audit_buffer *ab,
+ const char *prefix,
+ struct dentry *dentry,
+ struct vfsmount *vfsmnt);
+extern int audit_set_rate_limit(int limit);
+extern int audit_set_backlog_limit(int limit);
+extern int audit_set_enabled(int state);
+extern int audit_set_failure(int state);
+
+ /* Private API (for auditsc.c only) */
+extern void audit_send_reply(int pid, int seq, int type,
+ int done, int multi,
+ void *payload, int size);
+extern void audit_log_lost(const char *message);
+#else
+#define audit_log(t,f,...) do { ; } while (0)
+#define audit_log_start(t) ({ NULL; })
+#define audit_log_vformat(b,f,a) do { ; } while (0)
+#define audit_log_format(b,f,...) do { ; } while (0)
+#define audit_log_end(b) do { ; } while (0)
+#define audit_log_end_fast(b) do { ; } while (0)
+#define audit_log_end_irq(b) do { ; } while (0)
+#define audit_log_d_path(b,p,d,v) do { ; } while (0)
+#define audit_set_rate_limit(l) do { ; } while (0)
+#define audit_set_backlog_limit(l) do { ; } while (0)
+#define audit_set_enabled(s) do { ; } while (0)
+#define audit_set_failure(s) do { ; } while (0)
+#endif
+#endif
+#endif
--- diff/include/linux/compat_siginfo.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/linux/compat_siginfo.h 2004-03-16 11:27:48.537808792 +0000
@@ -0,0 +1,170 @@
+#ifndef _ASM_GENERIC_COMPAT_SIGINFO_H
+#define _ASM_GENERIC_COMPAT_SIGINFO_H
+
+#include <linux/config.h>
+#include <linux/compat.h>
+
+#ifndef CONFIG_COMPAT
+
+/* No compatibility layer required, add empty definitions for the compiler */
+
+typedef struct compat_siginfo{
+} compat_siginfo_t;
+
+static inline int compat_copy_siginfo_to_user(compat_siginfo_t __user *to,
+ struct siginfo *from)
+{
+ return -1;
+}
+
+#else
+
+#include <linux/compiler.h>
+#include <asm/siginfo.h>
+
+/* compat view of sigval_t */
+typedef union compat_sigval {
+ compat_int_t sival_int;
+ compat_uptr_t sival_ptr;
+} compat_sigval_t;
+
+/*
+ * This is the size (including padding) of the part of the
+ * struct siginfo that is before the union.
+ */
+#ifndef __ARCH_SI_COMPAT_PREAMBLE_SIZE
+#define __ARCH_SI_COMPAT_PREAMBLE_SIZE (3 * sizeof(int))
+#endif
+
+#define SI_COMPAT_MAX_SIZE 128
+#ifndef SI_COMPAT_PAD_SIZE
+#define SI_COMPAT_PAD_SIZE ((SI_COMPAT_MAX_SIZE - __ARCH_SI_COMPAT_PREAMBLE_SIZE) / sizeof(int))
+#endif
+
+/* 32-bit view of si.uid_t */
+#ifndef __ARCH_SI_COMPAT_UID_T
+#define __ARCH_SI_COMPAT_UID_T compat_uid_t
+#endif
+
+/* 32-bit view of si.band_t */
+#ifndef __ARCH_SI_COMPAT_BAND_T
+#define __ARCH_SI_COMPAT_BAND_T compat_int_t
+#endif
+
+#ifndef HAVE_ARCH_COMPAT_SIGINFO_T
+
+/* Compat view of siginfo_t */
+typedef struct compat_siginfo {
+ compat_int_t si_signo;
+ compat_int_t si_errno;
+ compat_int_t si_code;
+
+ union {
+ compat_int_t _pad[SI_COMPAT_PAD_SIZE];
+
+ /* kill() */
+ struct {
+ compat_pid_t _pid; /* sender's pid */
+ __ARCH_SI_COMPAT_UID_T _uid; /* sender's uid */
+ } _kill;
+
+ /* POSIX.1b timers */
+ struct {
+ compat_timer_t _tid; /* timer id */
+ compat_int_t _overrun; /* overrun count */
+ char _pad[sizeof( __ARCH_SI_UID_T) - sizeof(int)];
+ compat_sigval_t _sigval; /* same as below */
+ compat_int_t _sys_private; /* not to be passed to user */
+ } _timer;
+
+ /* POSIX.1b signals */
+ struct {
+ compat_pid_t _pid; /* sender's pid */
+ __ARCH_SI_COMPAT_UID_T _uid; /* sender's uid */
+ compat_sigval_t _sigval;
+ } _rt;
+
+ /* SIGCHLD */
+ struct {
+ compat_pid_t _pid; /* which child */
+ __ARCH_SI_COMPAT_UID_T _uid; /* sender's uid */
+ compat_int_t _status; /* exit code */
+ compat_clock_t _utime;
+ compat_clock_t _stime;
+ } _sigchld;
+
+ /* SIGILL, SIGFPE, SIGSEGV, SIGBUS */
+ struct {
+ compat_uptr_t _addr; /* faulting insn/memory ref. */
+#ifdef __ARCH_SI_COMPAT_TRAPNO
+ compat_int_t _trapno; /* TRAP # which caused the signal */
+#endif
+ } _sigfault;
+
+ /* SIGPOLL */
+ struct {
+ __ARCH_SI_COMPAT_BAND_T _band; /* POLL_IN, POLL_OUT, POLL_MSG */
+ compat_int_t _fd;
+ } _sigpoll;
+ } _sifields;
+} compat_siginfo_t;
+#endif /* !HAVE_ARCH_COMPAT_SIGINFO_T */
+
+#ifdef __ARCH_SI_COMPAT_TRAPNO
+#define si_trapno _sifields._sigfault._trapno
+#endif
+
+/*
+ * sigevent definitions
+ *
+ * It seems likely that SIGEV_THREAD will have to be handled from
+ * userspace, libpthread transmuting it to SIGEV_SIGNAL, which the
+ * thread manager then catches and does the appropriate nonsense.
+ * However, everything is written out here so as to not get lost.
+ */
+
+#define SIGEV_COMPAT_MAX_SIZE 64
+#ifndef SIGEV_COMPAT_PAD_SIZE
+#define SIGEV_COMPAT_PAD_SIZE ((SIGEV_COMPAT_MAX_SIZE/sizeof(int)) - 3)
+#endif
+
+#ifndef HAVE_ARCH_COMPAT_SIGEVENT_T
+
+/* 32-bit view of sigevent_t */
+typedef struct compat_sigevent {
+ compat_sigval_t sigev_value;
+ compat_int_t sigev_signo;
+ compat_int_t sigev_notify;
+ union {
+ compat_int_t _pad[SIGEV_COMPAT_PAD_SIZE];
+ compat_int_t _tid;
+
+ struct {
+ compat_uptr_t _function;
+ compat_uptr_t _attribute; /* really pthread_attr_t */
+ } _sigev_thread;
+ } _sigev_un;
+} compat_sigevent_t;
+
+#endif /* HAVE_ARCH_COMPAT_SIGEVENT_T */
+
+#ifndef HAVE_ARCH_COMPAT_COPY_SIGINFO
+
+#include <linux/string.h>
+
+static inline void compat_copy_siginfo(struct compat_siginfo *to, struct compat_siginfo *from)
+{
+ if (from->si_code < 0)
+ memcpy(to, from, sizeof(*to));
+ else
+ /* _sigchld is currently the largest know union member */
+ memcpy(to, from, __ARCH_SI_COMPAT_PREAMBLE_SIZE + sizeof(from->_sifields._sigchld));
+}
+
+#endif /* !HAVE_ARCH_COMPAT_COPY_SIGINFO */
+
+extern int compat_copy_siginfo_to_user(compat_siginfo_t __user *to, struct siginfo *from);
+
+#endif /* CONFIG_COMPAT */
+#endif /* _ASM_GENERIC_COMPAT_SIGINFO_H */
+
--- diff/include/linux/dwarf2-lang.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/linux/dwarf2-lang.h 2004-03-16 11:27:48.540808336 +0000
@@ -0,0 +1,132 @@
+#ifndef DWARF2_LANG
+#define DWARF2_LANG
+#include <linux/dwarf2.h>
+
+/*
+ * This is free software; you can redistribute it and/or modify it under
+ * the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2, or (at your option) any later
+ * version.
+ */
+/*
+ * This file defines macros that allow generation of DWARF debug records
+ * for asm files. This file is platform independent. Register numbers
+ * (which are about the only thing that is platform dependent) are to be
+ * supplied by a platform defined file.
+ */
+#define DWARF_preamble() .section .debug_frame,"",@progbits
+/*
+ * This macro starts a debug frame section. The debug_frame describes
+ * where to find the registers that the enclosing function saved on
+ * entry.
+ *
+ * ORD is use by the label generator and should be the same as what is
+ * passed to CFI_postamble.
+ *
+ * pc, pc register gdb ordinal.
+ *
+ * code_align this is the factor used to define locations or regions
+ * where the given definitions apply. If you use labels to define these
+ * this should be 1.
+ *
+ * data_align this is the factor used to define register offsets. If
+ * you use struct offset, this should be the size of the register in
+ * bytes or the negative of that. This is how it is used: you will
+ * define a register as the reference register, say the stack pointer,
+ * then you will say where a register is located relative to this
+ * reference registers value, say 40 for register 3 (the gdb register
+ * number). The <40> will be multiplied by <data_align> to define the
+ * byte offset of the given register (3, in this example). So if your
+ * <40> is the byte offset and the reference register points at the
+ * begining, you would want 1 for the data_offset. If <40> was the 40th
+ * 4-byte element in that structure you would want 4. And if your
+ * reference register points at the end of the structure you would want
+ * a negative data_align value(and you would have to do other math as
+ * well).
+ */
+
+#define CFI_preamble(ORD, pc, code_align, data_align) \
+.section .debug_frame,"",@progbits ; \
+frame/**/_/**/ORD: \
+ .long end/**/_/**/ORD-start/**/_/**/ORD; \
+start/**/_/**/ORD: \
+ .long DW_CIE_ID; \
+ .byte DW_CIE_VERSION; \
+ .byte 0 ; \
+ .uleb128 code_align; \
+ .sleb128 data_align; \
+ .byte pc;
+
+/*
+ * After the above macro and prior to the CFI_postamble, you need to
+ * define the initial state. This starts with defining the reference
+ * register and, usually the pc. Here are some helper macros:
+ */
+
+#define CFA_define_reference(reg, offset) \
+ .byte DW_CFA_def_cfa; \
+ .uleb128 reg; \
+ .uleb128 (offset);
+
+#define CFA_define_offset(reg, offset) \
+ .byte (DW_CFA_offset + reg); \
+ .uleb128 (offset);
+
+#define CFI_postamble(ORD) \
+ .align 4; \
+end/**/_/**/ORD:
+/*
+ * So now your code pushs stuff on the stack, you need a new location
+ * and the rules for what to do. This starts a running description of
+ * the call frame. You need to describe what changes with respect to
+ * the call registers as the location of the pc moves through the code.
+ * The following builds an FDE (fram descriptor entry?). Like the
+ * above, it has a preamble and a postamble. It also is tied to the CFI
+ * above.
+ * The first entry after the preamble must be the location in the code
+ * that the call frame is being described for.
+ */
+#define FDE_preamble(ORD, fde_no, initial_address, length) \
+ .long FDE_end/**/_/**/fde_no-FDE_start/**/_/**/fde_no; \
+FDE_start/**/_/**/fde_no: \
+ .long frame/**/_/**/ORD; \
+ .long initial_address; \
+ .long length;
+
+#define FDE_postamble(fde_no) \
+ .align 4; \
+FDE_end/**/_/**/fde_no:
+/*
+ * That done, you can now add registers, subtract registers, move the
+ * reference and even change the reference. You can also define a new
+ * area of code the info applies to. For discontinuous bits you should
+ * start a new FDE. You may have as many as you like.
+ */
+
+/*
+ * To advance the address by <bytes>
+ */
+
+#define FDE_advance(bytes) \
+ .byte DW_CFA_advance_loc4 \
+ .long bytes
+
+
+
+/*
+ * With the above you can define all the register locations. But
+ * suppose the reference register moves... Takes the new offset NOT an
+ * increment. This is how esp is tracked if it is not saved.
+ */
+
+#define CFA_define_cfa_offset(offset) \
+ .byte $DW_CFA_def_cfa_offset; \
+ .uleb128 (offset);
+/*
+ * Or suppose you want to use a different reference register...
+ */
+#define CFA_define_cfa_register(reg) \
+ .byte DW_CFA_def_cfa_register; \
+ .uleb128 reg;
+
+#endif
--- diff/include/linux/dwarf2.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/linux/dwarf2.h 2004-03-16 11:27:48.540808336 +0000
@@ -0,0 +1,738 @@
+/* Declarations and definitions of codes relating to the DWARF2 symbolic
+ debugging information format.
+ Copyright (C) 1992, 1993, 1995, 1996, 1997, 1999, 2000, 2001, 2002
+ Free Software Foundation, Inc.
+
+ Written by Gary Funck (gary@intrepid.com) The Ada Joint Program
+ Office (AJPO), Florida State Unviversity and Silicon Graphics Inc.
+ provided support for this effort -- June 21, 1995.
+
+ Derived from the DWARF 1 implementation written by Ron Guilmette
+ (rfg@netcom.com), November 1990.
+
+ This file is part of GCC.
+
+ GCC is free software; you can redistribute it and/or modify it under
+ the terms of the GNU General Public License as published by the Free
+ Software Foundation; either version 2, or (at your option) any later
+ version.
+
+ GCC is distributed in the hope that it will be useful, but WITHOUT
+ ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
+ License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with GCC; see the file COPYING. If not, write to the Free
+ Software Foundation, 59 Temple Place - Suite 330, Boston, MA
+ 02111-1307, USA. */
+
+/* This file is derived from the DWARF specification (a public document)
+ Revision 2.0.0 (July 27, 1993) developed by the UNIX International
+ Programming Languages Special Interest Group (UI/PLSIG) and distributed
+ by UNIX International. Copies of this specification are available from
+ UNIX International, 20 Waterview Boulevard, Parsippany, NJ, 07054.
+
+ This file also now contains definitions from the DWARF 3 specification. */
+
+/* This file is shared between GCC and GDB, and should not contain
+ prototypes. */
+
+#ifndef _ELF_DWARF2_H
+#define _ELF_DWARF2_H
+
+/* Structure found in the .debug_line section. */
+#ifndef __ASSEMBLY__
+typedef struct
+{
+ unsigned char li_length [4];
+ unsigned char li_version [2];
+ unsigned char li_prologue_length [4];
+ unsigned char li_min_insn_length [1];
+ unsigned char li_default_is_stmt [1];
+ unsigned char li_line_base [1];
+ unsigned char li_line_range [1];
+ unsigned char li_opcode_base [1];
+}
+DWARF2_External_LineInfo;
+
+typedef struct
+{
+ unsigned long li_length;
+ unsigned short li_version;
+ unsigned int li_prologue_length;
+ unsigned char li_min_insn_length;
+ unsigned char li_default_is_stmt;
+ int li_line_base;
+ unsigned char li_line_range;
+ unsigned char li_opcode_base;
+}
+DWARF2_Internal_LineInfo;
+
+/* Structure found in .debug_pubnames section. */
+typedef struct
+{
+ unsigned char pn_length [4];
+ unsigned char pn_version [2];
+ unsigned char pn_offset [4];
+ unsigned char pn_size [4];
+}
+DWARF2_External_PubNames;
+
+typedef struct
+{
+ unsigned long pn_length;
+ unsigned short pn_version;
+ unsigned long pn_offset;
+ unsigned long pn_size;
+}
+DWARF2_Internal_PubNames;
+
+/* Structure found in .debug_info section. */
+typedef struct
+{
+ unsigned char cu_length [4];
+ unsigned char cu_version [2];
+ unsigned char cu_abbrev_offset [4];
+ unsigned char cu_pointer_size [1];
+}
+DWARF2_External_CompUnit;
+
+typedef struct
+{
+ unsigned long cu_length;
+ unsigned short cu_version;
+ unsigned long cu_abbrev_offset;
+ unsigned char cu_pointer_size;
+}
+DWARF2_Internal_CompUnit;
+
+typedef struct
+{
+ unsigned char ar_length [4];
+ unsigned char ar_version [2];
+ unsigned char ar_info_offset [4];
+ unsigned char ar_pointer_size [1];
+ unsigned char ar_segment_size [1];
+}
+DWARF2_External_ARange;
+
+typedef struct
+{
+ unsigned long ar_length;
+ unsigned short ar_version;
+ unsigned long ar_info_offset;
+ unsigned char ar_pointer_size;
+ unsigned char ar_segment_size;
+}
+DWARF2_Internal_ARange;
+
+#define ENUM(name) enum name {
+#define IF_NOT_ASM(a) a
+#define COMMA ,
+#else
+#define ENUM(name)
+#define IF_NOT_ASM(a)
+#define COMMA
+
+#endif
+
+/* Tag names and codes. */
+ENUM(dwarf_tag)
+
+ DW_TAG_padding = 0x00 COMMA
+ DW_TAG_array_type = 0x01 COMMA
+ DW_TAG_class_type = 0x02 COMMA
+ DW_TAG_entry_point = 0x03 COMMA
+ DW_TAG_enumeration_type = 0x04 COMMA
+ DW_TAG_formal_parameter = 0x05 COMMA
+ DW_TAG_imported_declaration = 0x08 COMMA
+ DW_TAG_label = 0x0a COMMA
+ DW_TAG_lexical_block = 0x0b COMMA
+ DW_TAG_member = 0x0d COMMA
+ DW_TAG_pointer_type = 0x0f COMMA
+ DW_TAG_reference_type = 0x10 COMMA
+ DW_TAG_compile_unit = 0x11 COMMA
+ DW_TAG_string_type = 0x12 COMMA
+ DW_TAG_structure_type = 0x13 COMMA
+ DW_TAG_subroutine_type = 0x15 COMMA
+ DW_TAG_typedef = 0x16 COMMA
+ DW_TAG_union_type = 0x17 COMMA
+ DW_TAG_unspecified_parameters = 0x18 COMMA
+ DW_TAG_variant = 0x19 COMMA
+ DW_TAG_common_block = 0x1a COMMA
+ DW_TAG_common_inclusion = 0x1b COMMA
+ DW_TAG_inheritance = 0x1c COMMA
+ DW_TAG_inlined_subroutine = 0x1d COMMA
+ DW_TAG_module = 0x1e COMMA
+ DW_TAG_ptr_to_member_type = 0x1f COMMA
+ DW_TAG_set_type = 0x20 COMMA
+ DW_TAG_subrange_type = 0x21 COMMA
+ DW_TAG_with_stmt = 0x22 COMMA
+ DW_TAG_access_declaration = 0x23 COMMA
+ DW_TAG_base_type = 0x24 COMMA
+ DW_TAG_catch_block = 0x25 COMMA
+ DW_TAG_const_type = 0x26 COMMA
+ DW_TAG_constant = 0x27 COMMA
+ DW_TAG_enumerator = 0x28 COMMA
+ DW_TAG_file_type = 0x29 COMMA
+ DW_TAG_friend = 0x2a COMMA
+ DW_TAG_namelist = 0x2b COMMA
+ DW_TAG_namelist_item = 0x2c COMMA
+ DW_TAG_packed_type = 0x2d COMMA
+ DW_TAG_subprogram = 0x2e COMMA
+ DW_TAG_template_type_param = 0x2f COMMA
+ DW_TAG_template_value_param = 0x30 COMMA
+ DW_TAG_thrown_type = 0x31 COMMA
+ DW_TAG_try_block = 0x32 COMMA
+ DW_TAG_variant_part = 0x33 COMMA
+ DW_TAG_variable = 0x34 COMMA
+ DW_TAG_volatile_type = 0x35 COMMA
+ /* DWARF 3. */
+ DW_TAG_dwarf_procedure = 0x36 COMMA
+ DW_TAG_restrict_type = 0x37 COMMA
+ DW_TAG_interface_type = 0x38 COMMA
+ DW_TAG_namespace = 0x39 COMMA
+ DW_TAG_imported_module = 0x3a COMMA
+ DW_TAG_unspecified_type = 0x3b COMMA
+ DW_TAG_partial_unit = 0x3c COMMA
+ DW_TAG_imported_unit = 0x3d COMMA
+ /* SGI/MIPS Extensions. */
+ DW_TAG_MIPS_loop = 0x4081 COMMA
+ /* GNU extensions. */
+ DW_TAG_format_label = 0x4101 COMMA /* For FORTRAN 77 and Fortran 90. */
+ DW_TAG_function_template = 0x4102 COMMA /* For C++. */
+ DW_TAG_class_template = 0x4103 COMMA /* For C++. */
+ DW_TAG_GNU_BINCL = 0x4104 COMMA
+ DW_TAG_GNU_EINCL = 0x4105 COMMA
+ /* Extensions for UPC. See: http://upc.gwu.edu/~upc. */
+ DW_TAG_upc_shared_type = 0x8765 COMMA
+ DW_TAG_upc_strict_type = 0x8766 COMMA
+ DW_TAG_upc_relaxed_type = 0x8767
+IF_NOT_ASM(};)
+
+#define DW_TAG_lo_user 0x4080
+#define DW_TAG_hi_user 0xffff
+
+/* Flag that tells whether entry has a child or not. */
+#define DW_children_no 0
+#define DW_children_yes 1
+
+/* Form names and codes. */
+ENUM(dwarf_form)
+
+ DW_FORM_addr = 0x01 COMMA
+ DW_FORM_block2 = 0x03 COMMA
+ DW_FORM_block4 = 0x04 COMMA
+ DW_FORM_data2 = 0x05 COMMA
+ DW_FORM_data4 = 0x06 COMMA
+ DW_FORM_data8 = 0x07 COMMA
+ DW_FORM_string = 0x08 COMMA
+ DW_FORM_block = 0x09 COMMA
+ DW_FORM_block1 = 0x0a COMMA
+ DW_FORM_data1 = 0x0b COMMA
+ DW_FORM_flag = 0x0c COMMA
+ DW_FORM_sdata = 0x0d COMMA
+ DW_FORM_strp = 0x0e COMMA
+ DW_FORM_udata = 0x0f COMMA
+ DW_FORM_ref_addr = 0x10 COMMA
+ DW_FORM_ref1 = 0x11 COMMA
+ DW_FORM_ref2 = 0x12 COMMA
+ DW_FORM_ref4 = 0x13 COMMA
+ DW_FORM_ref8 = 0x14 COMMA
+ DW_FORM_ref_udata = 0x15 COMMA
+ DW_FORM_indirect = 0x16
+IF_NOT_ASM(};)
+
+/* Attribute names and codes. */
+
+ENUM(dwarf_attribute)
+
+ DW_AT_sibling = 0x01 COMMA
+ DW_AT_location = 0x02 COMMA
+ DW_AT_name = 0x03 COMMA
+ DW_AT_ordering = 0x09 COMMA
+ DW_AT_subscr_data = 0x0a COMMA
+ DW_AT_byte_size = 0x0b COMMA
+ DW_AT_bit_offset = 0x0c COMMA
+ DW_AT_bit_size = 0x0d COMMA
+ DW_AT_element_list = 0x0f COMMA
+ DW_AT_stmt_list = 0x10 COMMA
+ DW_AT_low_pc = 0x11 COMMA
+ DW_AT_high_pc = 0x12 COMMA
+ DW_AT_language = 0x13 COMMA
+ DW_AT_member = 0x14 COMMA
+ DW_AT_discr = 0x15 COMMA
+ DW_AT_discr_value = 0x16 COMMA
+ DW_AT_visibility = 0x17 COMMA
+ DW_AT_import = 0x18 COMMA
+ DW_AT_string_length = 0x19 COMMA
+ DW_AT_common_reference = 0x1a COMMA
+ DW_AT_comp_dir = 0x1b COMMA
+ DW_AT_const_value = 0x1c COMMA
+ DW_AT_containing_type = 0x1d COMMA
+ DW_AT_default_value = 0x1e COMMA
+ DW_AT_inline = 0x20 COMMA
+ DW_AT_is_optional = 0x21 COMMA
+ DW_AT_lower_bound = 0x22 COMMA
+ DW_AT_producer = 0x25 COMMA
+ DW_AT_prototyped = 0x27 COMMA
+ DW_AT_return_addr = 0x2a COMMA
+ DW_AT_start_scope = 0x2c COMMA
+ DW_AT_stride_size = 0x2e COMMA
+ DW_AT_upper_bound = 0x2f COMMA
+ DW_AT_abstract_origin = 0x31 COMMA
+ DW_AT_accessibility = 0x32 COMMA
+ DW_AT_address_class = 0x33 COMMA
+ DW_AT_artificial = 0x34 COMMA
+ DW_AT_base_types = 0x35 COMMA
+ DW_AT_calling_convention = 0x36 COMMA
+ DW_AT_count = 0x37 COMMA
+ DW_AT_data_member_location = 0x38 COMMA
+ DW_AT_decl_column = 0x39 COMMA
+ DW_AT_decl_file = 0x3a COMMA
+ DW_AT_decl_line = 0x3b COMMA
+ DW_AT_declaration = 0x3c COMMA
+ DW_AT_discr_list = 0x3d COMMA
+ DW_AT_encoding = 0x3e COMMA
+ DW_AT_external = 0x3f COMMA
+ DW_AT_frame_base = 0x40 COMMA
+ DW_AT_friend = 0x41 COMMA
+ DW_AT_identifier_case = 0x42 COMMA
+ DW_AT_macro_info = 0x43 COMMA
+ DW_AT_namelist_items = 0x44 COMMA
+ DW_AT_priority = 0x45 COMMA
+ DW_AT_segment = 0x46 COMMA
+ DW_AT_specification = 0x47 COMMA
+ DW_AT_static_link = 0x48 COMMA
+ DW_AT_type = 0x49 COMMA
+ DW_AT_use_location = 0x4a COMMA
+ DW_AT_variable_parameter = 0x4b COMMA
+ DW_AT_virtuality = 0x4c COMMA
+ DW_AT_vtable_elem_location = 0x4d COMMA
+ /* DWARF 3 values. */
+ DW_AT_allocated = 0x4e COMMA
+ DW_AT_associated = 0x4f COMMA
+ DW_AT_data_location = 0x50 COMMA
+ DW_AT_stride = 0x51 COMMA
+ DW_AT_entry_pc = 0x52 COMMA
+ DW_AT_use_UTF8 = 0x53 COMMA
+ DW_AT_extension = 0x54 COMMA
+ DW_AT_ranges = 0x55 COMMA
+ DW_AT_trampoline = 0x56 COMMA
+ DW_AT_call_column = 0x57 COMMA
+ DW_AT_call_file = 0x58 COMMA
+ DW_AT_call_line = 0x59 COMMA
+ /* SGI/MIPS extensions. */
+ DW_AT_MIPS_fde = 0x2001 COMMA
+ DW_AT_MIPS_loop_begin = 0x2002 COMMA
+ DW_AT_MIPS_tail_loop_begin = 0x2003 COMMA
+ DW_AT_MIPS_epilog_begin = 0x2004 COMMA
+ DW_AT_MIPS_loop_unroll_factor = 0x2005 COMMA
+ DW_AT_MIPS_software_pipeline_depth = 0x2006 COMMA
+ DW_AT_MIPS_linkage_name = 0x2007 COMMA
+ DW_AT_MIPS_stride = 0x2008 COMMA
+ DW_AT_MIPS_abstract_name = 0x2009 COMMA
+ DW_AT_MIPS_clone_origin = 0x200a COMMA
+ DW_AT_MIPS_has_inlines = 0x200b COMMA
+ /* GNU extensions. */
+ DW_AT_sf_names = 0x2101 COMMA
+ DW_AT_src_info = 0x2102 COMMA
+ DW_AT_mac_info = 0x2103 COMMA
+ DW_AT_src_coords = 0x2104 COMMA
+ DW_AT_body_begin = 0x2105 COMMA
+ DW_AT_body_end = 0x2106 COMMA
+ DW_AT_GNU_vector = 0x2107 COMMA
+ /* VMS extensions. */
+ DW_AT_VMS_rtnbeg_pd_address = 0x2201 COMMA
+ /* UPC extension. */
+ DW_AT_upc_threads_scaled = 0x3210
+IF_NOT_ASM(};)
+
+#define DW_AT_lo_user 0x2000 /* Implementation-defined range start. */
+#define DW_AT_hi_user 0x3ff0 /* Implementation-defined range end. */
+
+/* Location atom names and codes. */
+ENUM(dwarf_location_atom)
+
+ DW_OP_addr = 0x03 COMMA
+ DW_OP_deref = 0x06 COMMA
+ DW_OP_const1u = 0x08 COMMA
+ DW_OP_const1s = 0x09 COMMA
+ DW_OP_const2u = 0x0a COMMA
+ DW_OP_const2s = 0x0b COMMA
+ DW_OP_const4u = 0x0c COMMA
+ DW_OP_const4s = 0x0d COMMA
+ DW_OP_const8u = 0x0e COMMA
+ DW_OP_const8s = 0x0f COMMA
+ DW_OP_constu = 0x10 COMMA
+ DW_OP_consts = 0x11 COMMA
+ DW_OP_dup = 0x12 COMMA
+ DW_OP_drop = 0x13 COMMA
+ DW_OP_over = 0x14 COMMA
+ DW_OP_pick = 0x15 COMMA
+ DW_OP_swap = 0x16 COMMA
+ DW_OP_rot = 0x17 COMMA
+ DW_OP_xderef = 0x18 COMMA
+ DW_OP_abs = 0x19 COMMA
+ DW_OP_and = 0x1a COMMA
+ DW_OP_div = 0x1b COMMA
+ DW_OP_minus = 0x1c COMMA
+ DW_OP_mod = 0x1d COMMA
+ DW_OP_mul = 0x1e COMMA
+ DW_OP_neg = 0x1f COMMA
+ DW_OP_not = 0x20 COMMA
+ DW_OP_or = 0x21 COMMA
+ DW_OP_plus = 0x22 COMMA
+ DW_OP_plus_uconst = 0x23 COMMA
+ DW_OP_shl = 0x24 COMMA
+ DW_OP_shr = 0x25 COMMA
+ DW_OP_shra = 0x26 COMMA
+ DW_OP_xor = 0x27 COMMA
+ DW_OP_bra = 0x28 COMMA
+ DW_OP_eq = 0x29 COMMA
+ DW_OP_ge = 0x2a COMMA
+ DW_OP_gt = 0x2b COMMA
+ DW_OP_le = 0x2c COMMA
+ DW_OP_lt = 0x2d COMMA
+ DW_OP_ne = 0x2e COMMA
+ DW_OP_skip = 0x2f COMMA
+ DW_OP_lit0 = 0x30 COMMA
+ DW_OP_lit1 = 0x31 COMMA
+ DW_OP_lit2 = 0x32 COMMA
+ DW_OP_lit3 = 0x33 COMMA
+ DW_OP_lit4 = 0x34 COMMA
+ DW_OP_lit5 = 0x35 COMMA
+ DW_OP_lit6 = 0x36 COMMA
+ DW_OP_lit7 = 0x37 COMMA
+ DW_OP_lit8 = 0x38 COMMA
+ DW_OP_lit9 = 0x39 COMMA
+ DW_OP_lit10 = 0x3a COMMA
+ DW_OP_lit11 = 0x3b COMMA
+ DW_OP_lit12 = 0x3c COMMA
+ DW_OP_lit13 = 0x3d COMMA
+ DW_OP_lit14 = 0x3e COMMA
+ DW_OP_lit15 = 0x3f COMMA
+ DW_OP_lit16 = 0x40 COMMA
+ DW_OP_lit17 = 0x41 COMMA
+ DW_OP_lit18 = 0x42 COMMA
+ DW_OP_lit19 = 0x43 COMMA
+ DW_OP_lit20 = 0x44 COMMA
+ DW_OP_lit21 = 0x45 COMMA
+ DW_OP_lit22 = 0x46 COMMA
+ DW_OP_lit23 = 0x47 COMMA
+ DW_OP_lit24 = 0x48 COMMA
+ DW_OP_lit25 = 0x49 COMMA
+ DW_OP_lit26 = 0x4a COMMA
+ DW_OP_lit27 = 0x4b COMMA
+ DW_OP_lit28 = 0x4c COMMA
+ DW_OP_lit29 = 0x4d COMMA
+ DW_OP_lit30 = 0x4e COMMA
+ DW_OP_lit31 = 0x4f COMMA
+ DW_OP_reg0 = 0x50 COMMA
+ DW_OP_reg1 = 0x51 COMMA
+ DW_OP_reg2 = 0x52 COMMA
+ DW_OP_reg3 = 0x53 COMMA
+ DW_OP_reg4 = 0x54 COMMA
+ DW_OP_reg5 = 0x55 COMMA
+ DW_OP_reg6 = 0x56 COMMA
+ DW_OP_reg7 = 0x57 COMMA
+ DW_OP_reg8 = 0x58 COMMA
+ DW_OP_reg9 = 0x59 COMMA
+ DW_OP_reg10 = 0x5a COMMA
+ DW_OP_reg11 = 0x5b COMMA
+ DW_OP_reg12 = 0x5c COMMA
+ DW_OP_reg13 = 0x5d COMMA
+ DW_OP_reg14 = 0x5e COMMA
+ DW_OP_reg15 = 0x5f COMMA
+ DW_OP_reg16 = 0x60 COMMA
+ DW_OP_reg17 = 0x61 COMMA
+ DW_OP_reg18 = 0x62 COMMA
+ DW_OP_reg19 = 0x63 COMMA
+ DW_OP_reg20 = 0x64 COMMA
+ DW_OP_reg21 = 0x65 COMMA
+ DW_OP_reg22 = 0x66 COMMA
+ DW_OP_reg23 = 0x67 COMMA
+ DW_OP_reg24 = 0x68 COMMA
+ DW_OP_reg25 = 0x69 COMMA
+ DW_OP_reg26 = 0x6a COMMA
+ DW_OP_reg27 = 0x6b COMMA
+ DW_OP_reg28 = 0x6c COMMA
+ DW_OP_reg29 = 0x6d COMMA
+ DW_OP_reg30 = 0x6e COMMA
+ DW_OP_reg31 = 0x6f COMMA
+ DW_OP_breg0 = 0x70 COMMA
+ DW_OP_breg1 = 0x71 COMMA
+ DW_OP_breg2 = 0x72 COMMA
+ DW_OP_breg3 = 0x73 COMMA
+ DW_OP_breg4 = 0x74 COMMA
+ DW_OP_breg5 = 0x75 COMMA
+ DW_OP_breg6 = 0x76 COMMA
+ DW_OP_breg7 = 0x77 COMMA
+ DW_OP_breg8 = 0x78 COMMA
+ DW_OP_breg9 = 0x79 COMMA
+ DW_OP_breg10 = 0x7a COMMA
+ DW_OP_breg11 = 0x7b COMMA
+ DW_OP_breg12 = 0x7c COMMA
+ DW_OP_breg13 = 0x7d COMMA
+ DW_OP_breg14 = 0x7e COMMA
+ DW_OP_breg15 = 0x7f COMMA
+ DW_OP_breg16 = 0x80 COMMA
+ DW_OP_breg17 = 0x81 COMMA
+ DW_OP_breg18 = 0x82 COMMA
+ DW_OP_breg19 = 0x83 COMMA
+ DW_OP_breg20 = 0x84 COMMA
+ DW_OP_breg21 = 0x85 COMMA
+ DW_OP_breg22 = 0x86 COMMA
+ DW_OP_breg23 = 0x87 COMMA
+ DW_OP_breg24 = 0x88 COMMA
+ DW_OP_breg25 = 0x89 COMMA
+ DW_OP_breg26 = 0x8a COMMA
+ DW_OP_breg27 = 0x8b COMMA
+ DW_OP_breg28 = 0x8c COMMA
+ DW_OP_breg29 = 0x8d COMMA
+ DW_OP_breg30 = 0x8e COMMA
+ DW_OP_breg31 = 0x8f COMMA
+ DW_OP_regx = 0x90 COMMA
+ DW_OP_fbreg = 0x91 COMMA
+ DW_OP_bregx = 0x92 COMMA
+ DW_OP_piece = 0x93 COMMA
+ DW_OP_deref_size = 0x94 COMMA
+ DW_OP_xderef_size = 0x95 COMMA
+ DW_OP_nop = 0x96 COMMA
+ /* DWARF 3 extensions. */
+ DW_OP_push_object_address = 0x97 COMMA
+ DW_OP_call2 = 0x98 COMMA
+ DW_OP_call4 = 0x99 COMMA
+ DW_OP_call_ref = 0x9a COMMA
+ /* GNU extensions. */
+ DW_OP_GNU_push_tls_address = 0xe0
+IF_NOT_ASM(};)
+
+#define DW_OP_lo_user 0xe0 /* Implementation-defined range start. */
+#define DW_OP_hi_user 0xff /* Implementation-defined range end. */
+
+/* Type encodings. */
+ENUM(dwarf_type)
+
+ DW_ATE_void = 0x0 COMMA
+ DW_ATE_address = 0x1 COMMA
+ DW_ATE_boolean = 0x2 COMMA
+ DW_ATE_complex_float = 0x3 COMMA
+ DW_ATE_float = 0x4 COMMA
+ DW_ATE_signed = 0x5 COMMA
+ DW_ATE_signed_char = 0x6 COMMA
+ DW_ATE_unsigned = 0x7 COMMA
+ DW_ATE_unsigned_char = 0x8 COMMA
+ /* DWARF 3. */
+ DW_ATE_imaginary_float = 0x9
+IF_NOT_ASM(};)
+
+#define DW_ATE_lo_user 0x80
+#define DW_ATE_hi_user 0xff
+
+/* Array ordering names and codes. */
+ENUM(dwarf_array_dim_ordering)
+
+ DW_ORD_row_major = 0 COMMA
+ DW_ORD_col_major = 1
+IF_NOT_ASM(};)
+
+/* Access attribute. */
+ENUM(dwarf_access_attribute)
+
+ DW_ACCESS_public = 1 COMMA
+ DW_ACCESS_protected = 2 COMMA
+ DW_ACCESS_private = 3
+IF_NOT_ASM(};)
+
+/* Visibility. */
+ENUM(dwarf_visibility_attribute)
+
+ DW_VIS_local = 1 COMMA
+ DW_VIS_exported = 2 COMMA
+ DW_VIS_qualified = 3
+IF_NOT_ASM(};)
+
+/* Virtuality. */
+ENUM(dwarf_virtuality_attribute)
+
+ DW_VIRTUALITY_none = 0 COMMA
+ DW_VIRTUALITY_virtual = 1 COMMA
+ DW_VIRTUALITY_pure_virtual = 2
+IF_NOT_ASM(};)
+
+/* Case sensitivity. */
+ENUM(dwarf_id_case)
+
+ DW_ID_case_sensitive = 0 COMMA
+ DW_ID_up_case = 1 COMMA
+ DW_ID_down_case = 2 COMMA
+ DW_ID_case_insensitive = 3
+IF_NOT_ASM(};)
+
+/* Calling convention. */
+ENUM(dwarf_calling_convention)
+
+ DW_CC_normal = 0x1 COMMA
+ DW_CC_program = 0x2 COMMA
+ DW_CC_nocall = 0x3
+IF_NOT_ASM(};)
+
+#define DW_CC_lo_user 0x40
+#define DW_CC_hi_user 0xff
+
+/* Inline attribute. */
+ENUM(dwarf_inline_attribute)
+
+ DW_INL_not_inlined = 0 COMMA
+ DW_INL_inlined = 1 COMMA
+ DW_INL_declared_not_inlined = 2 COMMA
+ DW_INL_declared_inlined = 3
+IF_NOT_ASM(};)
+
+/* Discriminant lists. */
+ENUM(dwarf_discrim_list)
+
+ DW_DSC_label = 0 COMMA
+ DW_DSC_range = 1
+IF_NOT_ASM(};)
+
+/* Line number opcodes. */
+ENUM(dwarf_line_number_ops)
+
+ DW_LNS_extended_op = 0 COMMA
+ DW_LNS_copy = 1 COMMA
+ DW_LNS_advance_pc = 2 COMMA
+ DW_LNS_advance_line = 3 COMMA
+ DW_LNS_set_file = 4 COMMA
+ DW_LNS_set_column = 5 COMMA
+ DW_LNS_negate_stmt = 6 COMMA
+ DW_LNS_set_basic_block = 7 COMMA
+ DW_LNS_const_add_pc = 8 COMMA
+ DW_LNS_fixed_advance_pc = 9 COMMA
+ /* DWARF 3. */
+ DW_LNS_set_prologue_end = 10 COMMA
+ DW_LNS_set_epilogue_begin = 11 COMMA
+ DW_LNS_set_isa = 12
+IF_NOT_ASM(};)
+
+/* Line number extended opcodes. */
+ENUM(dwarf_line_number_x_ops)
+
+ DW_LNE_end_sequence = 1 COMMA
+ DW_LNE_set_address = 2 COMMA
+ DW_LNE_define_file = 3
+IF_NOT_ASM(};)
+
+/* Call frame information. */
+ENUM(dwarf_call_frame_info)
+
+ DW_CFA_advance_loc = 0x40 COMMA
+ DW_CFA_offset = 0x80 COMMA
+ DW_CFA_restore = 0xc0 COMMA
+ DW_CFA_nop = 0x00 COMMA
+ DW_CFA_set_loc = 0x01 COMMA
+ DW_CFA_advance_loc1 = 0x02 COMMA
+ DW_CFA_advance_loc2 = 0x03 COMMA
+ DW_CFA_advance_loc4 = 0x04 COMMA
+ DW_CFA_offset_extended = 0x05 COMMA
+ DW_CFA_restore_extended = 0x06 COMMA
+ DW_CFA_undefined = 0x07 COMMA
+ DW_CFA_same_value = 0x08 COMMA
+ DW_CFA_register = 0x09 COMMA
+ DW_CFA_remember_state = 0x0a COMMA
+ DW_CFA_restore_state = 0x0b COMMA
+ DW_CFA_def_cfa = 0x0c COMMA
+ DW_CFA_def_cfa_register = 0x0d COMMA
+ DW_CFA_def_cfa_offset = 0x0e COMMA
+
+ /* DWARF 3. */
+ DW_CFA_def_cfa_expression = 0x0f COMMA
+ DW_CFA_expression = 0x10 COMMA
+ DW_CFA_offset_extended_sf = 0x11 COMMA
+ DW_CFA_def_cfa_sf = 0x12 COMMA
+ DW_CFA_def_cfa_offset_sf = 0x13 COMMA
+
+ /* SGI/MIPS specific. */
+ DW_CFA_MIPS_advance_loc8 = 0x1d COMMA
+
+ /* GNU extensions. */
+ DW_CFA_GNU_window_save = 0x2d COMMA
+ DW_CFA_GNU_args_size = 0x2e COMMA
+ DW_CFA_GNU_negative_offset_extended = 0x2f
+IF_NOT_ASM(};)
+
+#define DW_CIE_ID 0xffffffff
+#define DW_CIE_VERSION 1
+
+#define DW_CFA_extended 0
+#define DW_CFA_lo_user 0x1c
+#define DW_CFA_hi_user 0x3f
+
+#define DW_CHILDREN_no 0x00
+#define DW_CHILDREN_yes 0x01
+
+#define DW_ADDR_none 0
+
+/* Source language names and codes. */
+ENUM(dwarf_source_language)
+
+ DW_LANG_C89 = 0x0001 COMMA
+ DW_LANG_C = 0x0002 COMMA
+ DW_LANG_Ada83 = 0x0003 COMMA
+ DW_LANG_C_plus_plus = 0x0004 COMMA
+ DW_LANG_Cobol74 = 0x0005 COMMA
+ DW_LANG_Cobol85 = 0x0006 COMMA
+ DW_LANG_Fortran77 = 0x0007 COMMA
+ DW_LANG_Fortran90 = 0x0008 COMMA
+ DW_LANG_Pascal83 = 0x0009 COMMA
+ DW_LANG_Modula2 = 0x000a COMMA
+ DW_LANG_Java = 0x000b COMMA
+ /* DWARF 3. */
+ DW_LANG_C99 = 0x000c COMMA
+ DW_LANG_Ada95 = 0x000d COMMA
+ DW_LANG_Fortran95 = 0x000e COMMA
+ /* MIPS. */
+ DW_LANG_Mips_Assembler = 0x8001 COMMA
+ /* UPC. */
+ DW_LANG_Upc = 0x8765
+IF_NOT_ASM(};)
+
+#define DW_LANG_lo_user 0x8000 /* Implementation-defined range start. */
+#define DW_LANG_hi_user 0xffff /* Implementation-defined range start. */
+
+/* Names and codes for macro information. */
+ENUM(dwarf_macinfo_record_type)
+
+ DW_MACINFO_define = 1 COMMA
+ DW_MACINFO_undef = 2 COMMA
+ DW_MACINFO_start_file = 3 COMMA
+ DW_MACINFO_end_file = 4 COMMA
+ DW_MACINFO_vendor_ext = 255
+IF_NOT_ASM(};)
+
+/* @@@ For use with GNU frame unwind information. */
+
+#define DW_EH_PE_absptr 0x00
+#define DW_EH_PE_omit 0xff
+
+#define DW_EH_PE_uleb128 0x01
+#define DW_EH_PE_udata2 0x02
+#define DW_EH_PE_udata4 0x03
+#define DW_EH_PE_udata8 0x04
+#define DW_EH_PE_sleb128 0x09
+#define DW_EH_PE_sdata2 0x0A
+#define DW_EH_PE_sdata4 0x0B
+#define DW_EH_PE_sdata8 0x0C
+#define DW_EH_PE_signed 0x08
+
+#define DW_EH_PE_pcrel 0x10
+#define DW_EH_PE_textrel 0x20
+#define DW_EH_PE_datarel 0x30
+#define DW_EH_PE_funcrel 0x40
+#define DW_EH_PE_aligned 0x50
+
+#define DW_EH_PE_indirect 0x80
+
+#endif /* _ELF_DWARF2_H */
--- diff/include/linux/lockmeter.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/linux/lockmeter.h 2004-03-16 11:27:48.598799520 +0000
@@ -0,0 +1,320 @@
+/*
+ * Copyright (C) 1999-2002 Silicon Graphics, Inc.
+ *
+ * Written by John Hawkes (hawkes@sgi.com)
+ * Based on klstat.h by Jack Steiner (steiner@sgi.com)
+ *
+ * Modified by Ray Bryant (raybry@us.ibm.com) Feb-Apr 2000
+ * Changes Copyright (C) 2000 IBM, Inc.
+ * Added save of index in spinlock_t to improve efficiency
+ * of "hold" time reporting for spinlocks
+ * Added support for hold time statistics for read and write
+ * locks.
+ * Moved machine dependent code to include/asm/lockmeter.h.
+ *
+ */
+
+#ifndef _LINUX_LOCKMETER_H
+#define _LINUX_LOCKMETER_H
+
+
+/*---------------------------------------------------
+ * architecture-independent lockmeter.h
+ *-------------------------------------------------*/
+
+/*
+ * raybry -- version 2: added efficient hold time statistics
+ * requires lstat recompile, so flagged as new version
+ * raybry -- version 3: added global reader lock data
+ * hawkes -- version 4: removed some unnecessary fields to simplify mips64 port
+ */
+#define LSTAT_VERSION 5
+
+int lstat_update(void*, void*, int);
+int lstat_update_time(void*, void*, int, uint32_t);
+
+/*
+ * Currently, the mips64 and sparc64 kernels talk to a 32-bit lockstat, so we
+ * need to force compatibility in the inter-communication data structure.
+ */
+
+#if defined(CONFIG_MIPS32_COMPAT)
+#define TIME_T uint32_t
+#elif defined(CONFIG_SPARC) || defined(CONFIG_SPARC64)
+#define TIME_T uint64_t
+#else
+#define TIME_T time_t
+#endif
+
+#if defined(__KERNEL__) || (!defined(CONFIG_MIPS32_COMPAT) && !defined(CONFIG_SPARC) && !defined(CONFIG_SPARC64)) || (_MIPS_SZLONG==32)
+#define POINTER void *
+#else
+#define POINTER int64_t
+#endif
+
+/*
+ * Values for the "action" parameter passed to lstat_update.
+ * ZZZ - do we want a try-success status here???
+ */
+#define LSTAT_ACT_NO_WAIT 0
+#define LSTAT_ACT_SPIN 1
+#define LSTAT_ACT_REJECT 2
+#define LSTAT_ACT_WW_SPIN 3
+#define LSTAT_ACT_SLEPT 4 /* UNUSED */
+
+#define LSTAT_ACT_MAX_VALUES 4 /* NOTE: Increase to 5 if use ACT_SLEPT */
+
+/*
+ * Special values for the low 2 bits of an RA passed to
+ * lstat_update.
+ */
+/* we use these values to figure out what kind of lock data */
+/* is stored in the statistics table entry at index ....... */
+#define LSTAT_RA_SPIN 0 /* spin lock data */
+#define LSTAT_RA_READ 1 /* read lock statistics */
+#define LSTAT_RA_SEMA 2 /* RESERVED */
+#define LSTAT_RA_WRITE 3 /* write lock statistics*/
+
+#define LSTAT_RA(n) \
+ ((void*)( ((unsigned long)__builtin_return_address(0) & ~3) | n) )
+
+/*
+ * Constants used for lock addresses in the lstat_directory
+ * to indicate special values of the lock address.
+ */
+#define LSTAT_MULTI_LOCK_ADDRESS NULL
+
+/*
+ * Maximum size of the lockstats tables. Increase this value
+ * if its not big enough. (Nothing bad happens if its not
+ * big enough although some locks will not be monitored.)
+ * We record overflows of this quantity in lstat_control.dir_overflows
+ *
+ * Note: The max value here must fit into the field set
+ * and obtained by the macro's PUT_INDEX() and GET_INDEX().
+ * This value depends on how many bits are available in the
+ * lock word in the particular machine implementation we are on.
+ */
+#define LSTAT_MAX_STAT_INDEX 2000
+
+/*
+ * Size and mask for the hash table into the directory.
+ */
+#define LSTAT_HASH_TABLE_SIZE 4096 /* must be 2**N */
+#define LSTAT_HASH_TABLE_MASK (LSTAT_HASH_TABLE_SIZE-1)
+
+#define DIRHASH(ra) ((unsigned long)(ra)>>2 & LSTAT_HASH_TABLE_MASK)
+
+/*
+ * This defines an entry in the lockstat directory. It contains
+ * information about a lock being monitored.
+ * A directory entry only contains the lock identification -
+ * counts on usage of the lock are kept elsewhere in a per-cpu
+ * data structure to minimize cache line pinging.
+ */
+typedef struct {
+ POINTER caller_ra; /* RA of code that set lock */
+ POINTER lock_ptr; /* lock address */
+ ushort next_stat_index; /* Used to link multiple locks that have the same hash table value */
+} lstat_directory_entry_t;
+
+/*
+ * A multi-dimensioned array used to contain counts for lock accesses.
+ * The array is 3-dimensional:
+ * - CPU number. Keep from thrashing cache lines between CPUs
+ * - Directory entry index. Identifies the lock
+ * - Action. Indicates what kind of contention occurred on an
+ * access to the lock.
+ *
+ * The index of an entry in the directory is the same as the 2nd index
+ * of the entry in the counts array.
+ */
+/*
+ * This table contains data for spin_locks, write locks, and read locks
+ * Not all data is used for all cases. In particular, the hold time
+ * information is not stored here for read locks since that is a global
+ * (e. g. cannot be separated out by return address) quantity.
+ * See the lstat_read_lock_counts_t structure for the global read lock
+ * hold time.
+ */
+typedef struct {
+ uint64_t cum_wait_ticks; /* sum of wait times */
+ /* for write locks, sum of time a */
+ /* writer is waiting for a reader */
+ int64_t cum_hold_ticks; /* cumulative sum of holds */
+ /* not used for read mode locks */
+ /* must be signed. ............... */
+ uint32_t max_wait_ticks; /* max waiting time */
+ uint32_t max_hold_ticks; /* max holding time */
+ uint64_t cum_wait_ww_ticks; /* sum times writer waits on writer*/
+ uint32_t max_wait_ww_ticks; /* max wait time writer vs writer */
+ /* prev 2 only used for write locks*/
+ uint32_t acquire_time; /* time lock acquired this CPU */
+ uint32_t count[LSTAT_ACT_MAX_VALUES];
+} lstat_lock_counts_t;
+
+typedef lstat_lock_counts_t lstat_cpu_counts_t[LSTAT_MAX_STAT_INDEX];
+
+/*
+ * User request to:
+ * - turn statistic collection on/off, or to reset
+ */
+#define LSTAT_OFF 0
+#define LSTAT_ON 1
+#define LSTAT_RESET 2
+#define LSTAT_RELEASE 3
+
+#define LSTAT_MAX_READ_LOCK_INDEX 1000
+typedef struct {
+ POINTER lock_ptr; /* address of lock for output stats */
+ uint32_t read_lock_count;
+ int64_t cum_hold_ticks; /* sum of read lock hold times over */
+ /* all callers. ....................*/
+ uint32_t write_index; /* last write lock hash table index */
+ uint32_t busy_periods; /* count of busy periods ended this */
+ uint64_t start_busy; /* time this busy period started. ..*/
+ uint64_t busy_ticks; /* sum of busy periods this lock. ..*/
+ uint64_t max_busy; /* longest busy period for this lock*/
+ uint32_t max_readers; /* maximum number of readers ...... */
+#ifdef USER_MODE_TESTING
+ rwlock_t entry_lock; /* lock for this read lock entry... */
+ /* avoid having more than one rdr at*/
+ /* needed for user space testing... */
+ /* not needed for kernel 'cause it */
+ /* is non-preemptive. ............. */
+#endif
+} lstat_read_lock_counts_t;
+typedef lstat_read_lock_counts_t lstat_read_lock_cpu_counts_t[LSTAT_MAX_READ_LOCK_INDEX];
+
+#if defined(__KERNEL__) || defined(USER_MODE_TESTING)
+
+#ifndef USER_MODE_TESTING
+#include <asm/lockmeter.h>
+#else
+#include "asm_newlockmeter.h"
+#endif
+
+/*
+ * Size and mask for the hash table into the directory.
+ */
+#define LSTAT_HASH_TABLE_SIZE 4096 /* must be 2**N */
+#define LSTAT_HASH_TABLE_MASK (LSTAT_HASH_TABLE_SIZE-1)
+
+#define DIRHASH(ra) ((unsigned long)(ra)>>2 & LSTAT_HASH_TABLE_MASK)
+
+/*
+ * This version eliminates the per processor lock stack. What we do is to
+ * store the index of the lock hash structure in unused bits in the lock
+ * itself. Then on unlock we can find the statistics record without doing
+ * any additional hash or lock stack lookup. This works for spin_locks.
+ * Hold time reporting is now basically as cheap as wait time reporting
+ * so we ignore the difference between LSTAT_ON_HOLD and LSTAT_ON_WAIT
+ * as in version 1.1.* of lockmeter.
+ *
+ * For rw_locks, we store the index of a global reader stats structure in
+ * the lock and the writer index is stored in the latter structure.
+ * For read mode locks we hash at the time of the lock to find an entry
+ * in the directory for reader wait time and the like.
+ * At unlock time for read mode locks, we update just the global structure
+ * so we don't need to know the reader directory index value at unlock time.
+ *
+ */
+
+/*
+ * Protocol to change lstat_control.state
+ * This is complicated because we don't want the cum_hold_time for
+ * a rw_lock to be decremented in _read_lock_ without making sure it
+ * is incremented in _read_lock_ and vice versa. So here is the
+ * way we change the state of lstat_control.state:
+ * I. To Turn Statistics On
+ * After allocating storage, set lstat_control.state non-zero.
+ * This works because we don't start updating statistics for in use
+ * locks until the reader lock count goes to zero.
+ * II. To Turn Statistics Off:
+ * (0) Disable interrupts on this CPU
+ * (1) Seize the lstat_control.directory_lock
+ * (2) Obtain the current value of lstat_control.next_free_read_lock_index
+ * (3) Store a zero in lstat_control.state.
+ * (4) Release the lstat_control.directory_lock
+ * (5) For each lock in the read lock list up to the saved value
+ * (well, -1) of the next_free_read_lock_index, do the following:
+ * (a) Check validity of the stored lock address
+ * by making sure that the word at the saved addr
+ * has an index that matches this entry. If not
+ * valid, then skip this entry.
+ * (b) If there is a write lock already set on this lock,
+ * skip to (d) below.
+ * (c) Set a non-metered write lock on the lock
+ * (d) set the cached INDEX in the lock to zero
+ * (e) Release the non-metered write lock.
+ * (6) Re-enable interrupts
+ *
+ * These rules ensure that a read lock will not have its statistics
+ * partially updated even though the global lock recording state has
+ * changed. See put_lockmeter_info() for implementation.
+ *
+ * The reason for (b) is that there may be write locks set on the
+ * syscall path to put_lockmeter_info() from user space. If we do
+ * not do this check, then we can deadlock. A similar problem would
+ * occur if the lock was read locked by the current CPU. At the
+ * moment this does not appear to happen.
+ */
+
+/*
+ * Main control structure for lockstat. Used to turn statistics on/off
+ * and to maintain directory info.
+ */
+typedef struct {
+ int state;
+ spinlock_t control_lock; /* used to serialize turning statistics on/off */
+ spinlock_t directory_lock; /* for serialize adding entries to directory */
+ volatile int next_free_dir_index;/* next free entry in the directory */
+ /* FIXME not all of these fields are used / needed .............. */
+ /* the following fields represent data since */
+ /* first "lstat on" or most recent "lstat reset" */
+ TIME_T first_started_time; /* time when measurement first enabled */
+ TIME_T started_time; /* time when measurement last started */
+ TIME_T ending_time; /* time when measurement last disabled */
+ uint64_t started_cycles64; /* cycles when measurement last started */
+ uint64_t ending_cycles64; /* cycles when measurement last disabled */
+ uint64_t enabled_cycles64; /* total cycles with measurement enabled */
+ int intervals; /* number of measurement intervals recorded */
+ /* i. e. number of times did lstat on;lstat off */
+ lstat_directory_entry_t *dir; /* directory */
+ int dir_overflow; /* count of times ran out of space in directory */
+ int rwlock_overflow; /* count of times we couldn't allocate a rw block*/
+ ushort *hashtab; /* hash table for quick dir scans */
+ lstat_cpu_counts_t *counts[NR_CPUS]; /* Array of pointers to per-cpu stats */
+ int next_free_read_lock_index; /* next rwlock reader (global) stats block */
+ lstat_read_lock_cpu_counts_t *read_lock_counts[NR_CPUS]; /* per cpu read lock stats */
+} lstat_control_t;
+
+#endif /* defined(__KERNEL__) || defined(USER_MODE_TESTING) */
+
+typedef struct {
+ short lstat_version; /* version of the data */
+ short state; /* the current state is returned */
+ int maxcpus; /* Number of cpus present */
+ int next_free_dir_index; /* index of the next free directory entry */
+ TIME_T first_started_time; /* when measurement enabled for first time */
+ TIME_T started_time; /* time in secs since 1969 when stats last turned on */
+ TIME_T ending_time; /* time in secs since 1969 when stats last turned off */
+ uint32_t cycleval; /* cycles per second */
+#ifdef notyet
+ void *kernel_magic_addr; /* address of kernel_magic */
+ void *kernel_end_addr; /* contents of kernel magic (points to "end") */
+#endif
+ int next_free_read_lock_index; /* index of next (global) read lock stats struct */
+ uint64_t started_cycles64; /* cycles when measurement last started */
+ uint64_t ending_cycles64; /* cycles when stats last turned off */
+ uint64_t enabled_cycles64; /* total cycles with measurement enabled */
+ int intervals; /* number of measurement intervals recorded */
+ /* i.e. number of times we did lstat on;lstat off*/
+ int dir_overflow; /* number of times we wanted more space in directory */
+ int rwlock_overflow; /* # of times we wanted more space in read_locks_count */
+ struct new_utsname uts; /* info about machine where stats are measured */
+ /* -T option of lockstat allows data to be */
+ /* moved to another machine. ................. */
+} lstat_user_request_t;
+
+#endif /* _LINUX_LOCKMETER_H */
--- diff/include/linux/mqueue.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/linux/mqueue.h 2004-03-16 11:27:48.602798912 +0000
@@ -0,0 +1,37 @@
+/* Copyright (C) 2003 Krzysztof Benedyczak & Michal Wronski
+
+ This program is free software; you can redistribute it and/or
+ modify it under the terms of the GNU Lesser General Public
+ License as published by the Free Software Foundation; either
+ version 2.1 of the License, or (at your option) any later version.
+
+ It is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public
+ License along with this software; if not, write to the Free
+ Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+ 02111-1307 USA. */
+
+#ifndef _LINUX_MQUEUE_H
+#define _LINUX_MQUEUE_H
+
+#define MQ_PRIO_MAX 32768
+
+typedef int mqd_t;
+
+struct mq_attr {
+ long mq_flags; /* message queue flags */
+ long mq_maxmsg; /* maximum number of messages */
+ long mq_msgsize; /* maximum message size */
+ long mq_curmsgs; /* number of messages currently queued */
+ long __reserved[4]; /* ignored for input, zeroed for output */
+};
+
+#define NOTIFY_NONE 0
+#define NOTIFY_WOKENUP 1
+#define NOTIFY_REMOVED 2
+
+#endif
--- diff/include/net/af_ipmi.h 1970-01-01 00:00:00.000000000 +0000
+++ source/include/net/af_ipmi.h 2004-03-16 11:27:48.627795112 +0000
@@ -0,0 +1,59 @@
+/*
+ * IPMI Socket Glue
+ *
+ * Author: Louis Zhuang <louis.zhuang@linux.intel.com>
+ * Copyright by Intel Corp., 2003
+ */
+#ifndef _NET_IPMI_H
+#define _NET_IPMI_H
+
+#include <linux/ipmi.h>
+
+/*
+ * This is ipmi address for socket
+ */
+struct sockaddr_ipmi {
+ sa_family_t sipmi_family; /* AF_IPMI */
+ int if_num; /* IPMI interface number */
+ struct ipmi_addr ipmi_addr;
+};
+#define SOCKADDR_IPMI_OVERHEAD (sizeof(struct sockaddr_ipmi) \
+ - sizeof(struct ipmi_addr))
+
+/* A msg_control item, this takes a 'struct ipmi_timing_parms' */
+#define IPMI_CMSG_TIMING_PARMS 0x01
+
+/*
+ * This is ipmi message for socket
+ */
+struct ipmi_sock_msg {
+ int recv_type;
+ long msgid;
+
+ unsigned char netfn;
+ unsigned char cmd;
+ int data_len;
+ unsigned char data[0];
+};
+
+#define IPMI_MAX_SOCK_MSG_LENGTH (sizeof(struct ipmi_sock_msg)+IPMI_MAX_MSG_LENGTH)
+
+/* Register/unregister to receive specific commands. Uses struct
+ ipmi_cmdspec from linux/ipmi.h */
+#define SIOCIPMIREGCMD (SIOCPROTOPRIVATE + 0)
+#define SIOCIPMIUNREGCMD (SIOCPROTOPRIVATE + 1)
+
+/* Register to receive events. Takes an integer */
+#define SIOCIPMIGETEVENT (SIOCPROTOPRIVATE + 2)
+
+/* Set the default timing parameters for the socket. Takes a struct
+ ipmi_timing_parms from linux/ipmi.h */
+#define SIOCIPMISETTIMING (SIOCPROTOPRIVATE + 3)
+#define SIOCIPMIGETTIMING (SIOCPROTOPRIVATE + 4)
+
+/* Set/Get the IPMB address of the MC we are connected to, takes an
+ unsigned int. */
+#define SIOCIPMISETADDR (SIOCPROTOPRIVATE + 5)
+#define SIOCIPMIGETADDR (SIOCPROTOPRIVATE + 6)
+
+#endif/*_NET_IPMI_H*/
--- diff/ipc/mqueue.c 1970-01-01 00:00:00.000000000 +0000
+++ source/ipc/mqueue.c 2004-03-16 11:27:48.633794200 +0000
@@ -0,0 +1,1256 @@
+/*
+ * POSIX message queues filesystem for Linux.
+ *
+ * Copyright (C) 2003,2004 Krzysztof Benedyczak (golbi@mat.uni.torun.pl)
+ * Michal Wronski (wrona@mat.uni.torun.pl)
+ *
+ * Spinlocks: Mohamed Abbas (abbas.mohamed@intel.com)
+ * Lockless receive & send, fd based notify:
+ * Manfred Spraul (manfred@colorfullife.com)
+ *
+ * This file is released under the GPL.
+ */
+
+#include <linux/init.h>
+#include <linux/pagemap.h>
+#include <linux/file.h>
+#include <linux/mount.h>
+#include <linux/namei.h>
+#include <linux/sysctl.h>
+#include <linux/poll.h>
+#include <linux/mqueue.h>
+#include <linux/msg.h>
+#include "util.h"
+
+#define MQUEUE_MAGIC 0x19800202
+#define DIRENT_SIZE 20
+#define FILENT_SIZE 80
+
+#define SEND 0
+#define RECV 1
+
+#define STATE_NONE 0
+#define STATE_PENDING 1
+#define STATE_READY 2
+
+#define NP_NONE ((void*)NOTIFY_NONE)
+#define NP_WOKENUP ((void*)NOTIFY_WOKENUP)
+#define NP_REMOVED ((void*)NOTIFY_REMOVED)
+/* used by sysctl */
+#define FS_MQUEUE 1
+#define CTL_QUEUESMAX 2
+#define CTL_MSGMAX 3
+#define CTL_MSGSIZEMAX 4
+
+/* default values */
+#define DFLT_QUEUESMAX 64 /* max number of message queues */
+#define DFLT_MSGMAX 40 /* max number of messages in each queue */
+#define HARD_MSGMAX (131072/sizeof(void*))
+#define DFLT_MSGSIZEMAX 16384 /* max message size */
+
+struct ext_wait_queue { /* queue of sleeping tasks */
+ struct task_struct *task;
+ struct list_head list;
+ struct msg_msg *msg; /* ptr of loaded message */
+ int state; /* one of STATE_* values */
+};
+
+struct mqueue_inode_info {
+ struct mq_attr attr;
+ struct msg_msg **messages;
+
+ pid_t notify_owner; /* != 0 means notification registered */
+ struct sigevent notify;
+ struct file *notify_filp;
+
+ /* for tasks waiting for free space and messages, respectively */
+ struct ext_wait_queue e_wait_q[2];
+ wait_queue_head_t wait_q;
+
+ unsigned long qsize; /* size of queue in memory (sum of all msgs) */
+ spinlock_t lock;
+ struct inode vfs_inode;
+};
+
+static struct inode_operations mqueue_dir_inode_operations;
+static struct file_operations mqueue_file_operations;
+static struct file_operations mqueue_notify_fops;
+static struct super_operations mqueue_super_ops;
+static void remove_notification(struct mqueue_inode_info *info);
+
+static spinlock_t mq_lock;
+static kmem_cache_t *mqueue_inode_cachep;
+static struct vfsmount *mqueue_mnt;
+
+static unsigned int queues_count;
+static unsigned int queues_max = DFLT_QUEUESMAX;
+static unsigned int msg_max = DFLT_MSGMAX;
+static unsigned int msgsize_max = DFLT_MSGSIZEMAX;
+
+static struct ctl_table_header * mq_sysctl_table;
+
+static inline struct mqueue_inode_info *MQUEUE_I(struct inode *inode)
+{
+ return container_of(inode, struct mqueue_inode_info, vfs_inode);
+}
+
+static struct inode *mqueue_get_inode(struct super_block *sb, int mode)
+{
+ struct inode *inode;
+
+ inode = new_inode(sb);
+ if (inode) {
+ inode->i_mode = mode;
+ inode->i_uid = current->fsuid;
+ inode->i_gid = current->fsgid;
+ inode->i_blksize = PAGE_CACHE_SIZE;
+ inode->i_blocks = 0;
+ inode->i_mtime = inode->i_ctime = inode->i_atime =
+ CURRENT_TIME;
+
+ if (S_ISREG(mode)) {
+ struct mqueue_inode_info *info;
+
+ inode->i_fop = &mqueue_file_operations;
+ inode->i_size = FILENT_SIZE;
+ /* mqueue specific info */
+ info = MQUEUE_I(inode);
+ spin_lock_init(&info->lock);
+ init_waitqueue_head(&info->wait_q);
+ INIT_LIST_HEAD(&info->e_wait_q[0].list);
+ INIT_LIST_HEAD(&info->e_wait_q[1].list);
+ info->notify_owner = 0;
+ info->qsize = 0;
+ memset(&info->attr, 0, sizeof(info->attr));
+ info->attr.mq_maxmsg = DFLT_MSGMAX;
+ info->attr.mq_msgsize = DFLT_MSGSIZEMAX;
+ info->messages = kmalloc(DFLT_MSGMAX * sizeof(struct msg_msg *), GFP_KERNEL);
+ if (!info->messages) {
+ make_bad_inode(inode);
+ iput(inode);
+ inode = NULL;
+ }
+ } else if (S_ISDIR(mode)) {
+ inode->i_nlink++;
+ /* Some things misbehave if size == 0 on a directory */
+ inode->i_size = 2 * DIRENT_SIZE;
+ inode->i_op = &mqueue_dir_inode_operations;
+ inode->i_fop = &simple_dir_operations;
+ }
+ }
+ return inode;
+}
+
+static int mqueue_fill_super(struct super_block *sb, void *data, int silent)
+{
+ struct inode *inode;
+
+ sb->s_blocksize = PAGE_CACHE_SIZE;
+ sb->s_blocksize_bits = PAGE_CACHE_SHIFT;
+ sb->s_magic = MQUEUE_MAGIC;
+ sb->s_op = &mqueue_super_ops;
+
+ inode = mqueue_get_inode(sb, S_IFDIR | S_IRWXUGO);
+ if (!inode)
+ return -ENOMEM;
+
+ sb->s_root = d_alloc_root(inode);
+ if (!sb->s_root) {
+ iput(inode);
+ return -ENOMEM;
+ }
+
+ return 0;
+}
+
+static struct super_block *mqueue_get_sb(struct file_system_type *fs_type,
+ int flags, const char *dev_name,
+ void *data)
+{
+ return get_sb_single(fs_type, flags, data, mqueue_fill_super);
+}
+
+static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags)
+{
+ struct mqueue_inode_info *p = (struct mqueue_inode_info *) foo;
+
+ if ((flags & (SLAB_CTOR_VERIFY | SLAB_CTOR_CONSTRUCTOR)) ==
+ SLAB_CTOR_CONSTRUCTOR)
+ inode_init_once(&p->vfs_inode);
+}
+
+static struct inode *mqueue_alloc_inode(struct super_block *sb)
+{
+ struct mqueue_inode_info *ei;
+
+ ei = kmem_cache_alloc(mqueue_inode_cachep, SLAB_KERNEL);
+ if (!ei)
+ return NULL;
+ return &ei->vfs_inode;
+}
+
+static void mqueue_destroy_inode(struct inode *inode)
+{
+ kmem_cache_free(mqueue_inode_cachep, MQUEUE_I(inode));
+}
+
+static void mqueue_delete_inode(struct inode *inode)
+{
+ struct mqueue_inode_info *info;
+ int i;
+
+ if (S_ISDIR(inode->i_mode)) {
+ clear_inode(inode);
+ return;
+ }
+ info = MQUEUE_I(inode);
+ spin_lock(&info->lock);
+ for (i = 0; i < info->attr.mq_curmsgs; i++)
+ free_msg(info->messages[i]);
+ kfree(info->messages);
+ spin_unlock(&info->lock);
+
+ clear_inode(inode);
+
+ spin_lock(&mq_lock);
+ queues_count--;
+ spin_unlock(&mq_lock);
+}
+
+static int mqueue_create(struct inode *dir, struct dentry *dentry,
+ int mode, struct nameidata *nd)
+{
+ struct inode *inode;
+ int error;
+
+ spin_lock(&mq_lock);
+ if (queues_count >= queues_max && !capable(CAP_SYS_RESOURCE)) {
+ error = -ENOSPC;
+ goto out_lock;
+ }
+ queues_count++;
+ spin_unlock(&mq_lock);
+
+ inode = mqueue_get_inode(dir->i_sb, mode);
+ if (!inode) {
+ error = -ENOMEM;
+ spin_lock(&mq_lock);
+ queues_count--;
+ goto out_lock;
+ }
+
+ dir->i_size += DIRENT_SIZE;
+ dir->i_ctime = dir->i_mtime = dir->i_atime = CURRENT_TIME;
+
+ d_instantiate(dentry, inode);
+ dget(dentry);
+ return 0;
+out_lock:
+ spin_unlock(&mq_lock);
+ return error;
+}
+
+static int mqueue_unlink(struct inode *dir, struct dentry *dentry)
+{
+ struct inode *inode = dentry->d_inode;
+
+ dir->i_ctime = dir->i_mtime = dir->i_atime = CURRENT_TIME;
+ dir->i_size -= DIRENT_SIZE;
+ inode->i_nlink--;
+ dput(dentry);
+ return 0;
+}
+
+/*
+* This is routine for system read from queue file.
+* To avoid mess with doing here some sort of mq_receive we allow
+* to read only queue size & notification info (the only values
+* that are interesting from user point of view and aren't accessible
+* through std routines)
+*/
+static ssize_t mqueue_read_file(struct file *filp, char __user *u_data,
+ size_t count, loff_t * off)
+{
+ struct mqueue_inode_info *info = MQUEUE_I(filp->f_dentry->d_inode);
+ char buffer[FILENT_SIZE];
+ size_t slen;
+ loff_t o;
+
+ if (!count)
+ return 0;
+
+ spin_lock(&info->lock);
+ snprintf(buffer, sizeof(buffer),
+ "QSIZE:%-10lu NOTIFY:%-5d SIGNO:%-5d NOTIFY_PID:%-6d\n",
+ info->qsize,
+ info->notify_owner ? info->notify.sigev_notify : SIGEV_NONE,
+ (info->notify_owner && info->notify.sigev_notify == SIGEV_SIGNAL ) ?
+ info->notify.sigev_signo : 0,
+ info->notify_owner);
+ spin_unlock(&info->lock);
+ buffer[sizeof(buffer)-1] = '\0';
+ slen = strlen(buffer)+1;
+
+ o = *off;
+ if (o > slen)
+ return 0;
+
+ if (o + count > slen)
+ count = slen - o;
+
+ if (copy_to_user(u_data, buffer + o, count))
+ return -EFAULT;
+
+ *off = o + count;
+ filp->f_dentry->d_inode->i_atime = filp->f_dentry->d_inode->i_ctime = CURRENT_TIME;
+ return count;
+}
+
+static int mqueue_flush_file(struct file *filp)
+{
+ struct mqueue_inode_info *info = MQUEUE_I(filp->f_dentry->d_inode);
+
+ spin_lock(&info->lock);
+ if (current->tgid == info->notify_owner)
+ remove_notification(info);
+
+ spin_unlock(&info->lock);
+ return 0;
+}
+
+static unsigned int mqueue_poll_file(struct file *filp, struct poll_table_struct *poll_tab)
+{
+ struct mqueue_inode_info *info = MQUEUE_I(filp->f_dentry->d_inode);
+ int retval = 0;
+
+ poll_wait(filp, &info->wait_q, poll_tab);
+
+ spin_lock(&info->lock);
+ if (info->attr.mq_curmsgs)
+ retval = POLLIN | POLLRDNORM;
+
+ if (info->attr.mq_curmsgs < info->attr.mq_maxmsg)
+ retval |= POLLOUT | POLLWRNORM;
+ spin_unlock(&info->lock);
+
+ return retval;
+}
+
+/* Adds current to info->e_wait_q[sr] before element with smaller prio */
+static void wq_add(struct mqueue_inode_info *info, int sr,
+ struct ext_wait_queue *ewp)
+{
+ struct ext_wait_queue *walk;
+
+ ewp->task = current;
+
+ list_for_each_entry(walk, &info->e_wait_q[sr].list, list) {
+ if (walk->task->static_prio <= current->static_prio) {
+ list_add_tail(&ewp->list, &walk->list);
+ return;
+ }
+ }
+ list_add_tail(&ewp->list, &info->e_wait_q[sr].list);
+}
+
+/*
+ * Puts current task to sleep. Caller must hold queue lock. After return
+ * lock isn't held.
+ * sr: SEND or RECV
+ */
+static int wq_sleep(struct mqueue_inode_info *info, int sr,
+ long timeout, struct ext_wait_queue *ewp)
+{
+ int retval;
+ signed long time;
+
+ wq_add(info, sr, ewp);
+
+ for (;;) {
+ set_current_state(TASK_INTERRUPTIBLE);
+
+ spin_unlock(&info->lock);
+ time = schedule_timeout(timeout);
+
+ while (ewp->state == STATE_PENDING)
+ cpu_relax();
+
+ if (ewp->state == STATE_READY) {
+ retval = 0;
+ goto out;
+ }
+ spin_lock(&info->lock);
+ if (ewp->state == STATE_READY) {
+ retval = 0;
+ goto out_unlock;
+ }
+ if (signal_pending(current)) {
+ retval = -ERESTARTSYS;
+ break;
+ }
+ if (time == 0) {
+ retval = -ETIMEDOUT;
+ break;
+ }
+ }
+ list_del(&ewp->list);
+out_unlock:
+ spin_unlock(&info->lock);
+out:
+ return retval;
+}
+
+/*
+ * Returns waiting task that should be serviced first or NULL if none exists
+ */
+static struct ext_wait_queue *wq_get_first_waiter(
+ struct mqueue_inode_info *info, int sr)
+{
+ struct list_head *ptr;
+
+ ptr = info->e_wait_q[sr].list.prev;
+ if (ptr == &info->e_wait_q[sr].list)
+ return NULL;
+ return list_entry(ptr, struct ext_wait_queue, list);
+}
+
+/* Auxiliary functions to manipulate messages' list */
+static void msg_insert(struct msg_msg *ptr, struct mqueue_inode_info *info)
+{
+ int k;
+
+ k = info->attr.mq_curmsgs - 1;
+ while (k >= 0 && info->messages[k]->m_type >= ptr->m_type) {
+ info->messages[k + 1] = info->messages[k];
+ k--;
+ }
+ info->attr.mq_curmsgs++;
+ info->qsize += ptr->m_ts;
+ info->messages[k + 1] = ptr;
+}
+
+static inline struct msg_msg *msg_get(struct mqueue_inode_info *info)
+{
+ info->qsize -= info->messages[--info->attr.mq_curmsgs]->m_ts;
+ return info->messages[info->attr.mq_curmsgs];
+}
+
+/*
+ * The next function is only to split too long sys_mq_timedsend
+ */
+static void __do_notify(struct mqueue_inode_info *info)
+{
+ /* notification
+ * invoked when there is registered process and there isn't process
+ * waiting synchronously for message AND state of queue changed from
+ * empty to not empty. Here we are sure that no one is waiting
+ * synchronously. */
+ if (info->notify_owner && info->attr.mq_curmsgs == 1) {
+ /* sends signal */
+ if (info->notify.sigev_notify == SIGEV_SIGNAL) {
+ struct siginfo sig_i;
+
+ sig_i.si_signo = info->notify.sigev_signo;
+ sig_i.si_errno = 0;
+ sig_i.si_code = SI_MESGQ;
+ sig_i.si_value = info->notify.sigev_value;
+ sig_i.si_pid = current->tgid;
+ sig_i.si_uid = current->uid;
+
+ kill_proc_info(info->notify.sigev_signo,
+ &sig_i, info->notify_owner);
+ } else if (info->notify.sigev_notify == SIGEV_THREAD) {
+ info->notify_filp->private_data = (void*)NP_WOKENUP;
+ }
+ /* after notification unregisters process */
+ info->notify_owner = 0;
+ }
+ wake_up(&info->wait_q);
+}
+
+static long prepare_timeout(const struct timespec __user *u_arg)
+{
+ struct timespec ts, nowts;
+ long timeout;
+
+ if (u_arg) {
+ if (unlikely(copy_from_user(&ts, u_arg,
+ sizeof(struct timespec))))
+ return -EFAULT;
+
+ if (unlikely(ts.tv_nsec < 0 || ts.tv_sec < 0
+ || ts.tv_nsec >= NSEC_PER_SEC))
+ return -EINVAL;
+ nowts = CURRENT_TIME;
+ /* first subtract as jiffies can't be too big */
+ ts.tv_sec -= nowts.tv_sec;
+ if (ts.tv_nsec < nowts.tv_nsec) {
+ ts.tv_nsec += NSEC_PER_SEC;
+ ts.tv_sec--;
+ }
+ ts.tv_nsec -= nowts.tv_nsec;
+ if (ts.tv_sec < 0)
+ return 0;
+
+ timeout = timespec_to_jiffies(&ts) + 1;
+ } else
+ return MAX_SCHEDULE_TIMEOUT;
+
+ return timeout;
+}
+
+/*
+ * File descriptor based notification, intended to be used to implement
+ * SIGEV_THREAD:
+ * SIGEV_THREAD means that a notification function should be called in the
+ * context of a new thread. The kernel can't do that. Therefore mq_notify
+ * calls with SIGEV_THREAD return a new file descriptor. A user space helper
+ * must create a new thread and then read from the given file descriptor.
+ * The read always returns one byte. If it's NOTIFY_WOKENUP, then it must
+ * call the notification function. If it's NOTIFY_REMOVED, then the
+ * notification was removed. The file descriptor supports poll, thus one
+ * supervisor thread can manage multiple message queue notifications.
+ *
+ * The implementation must support multiple outstanding notifications:
+ * It's possible that a new notification is added and signaled before user
+ * space calls mqueue_notify_read for the previous notification.
+ * Therefore the notification state is stored in the private_data field of
+ * the file descriptor.
+ */
+static unsigned int mqueue_notify_poll(struct file *filp,
+ struct poll_table_struct *poll_tab)
+{
+ struct mqueue_inode_info *info = MQUEUE_I(filp->f_dentry->d_inode);
+ int retval;
+
+ poll_wait(filp, &info->wait_q, poll_tab);
+
+ if (filp->private_data == NP_NONE)
+ retval = 0;
+ else
+ retval = POLLIN | POLLRDNORM;
+ return retval;
+}
+
+static ssize_t mqueue_notify_read(struct file *filp, char __user *buf,
+ size_t count, loff_t *ppos)
+{
+ struct mqueue_inode_info *info = MQUEUE_I(filp->f_dentry->d_inode);
+ char result;
+
+ if (!count)
+ return 0;
+ if (*ppos != 0)
+ return 0;
+ spin_lock(&info->lock);
+ while (filp->private_data == NP_NONE) {
+ DEFINE_WAIT(wait);
+ if (filp->f_flags & O_NONBLOCK) {
+ spin_unlock(&info->lock);
+ return -EAGAIN;
+ }
+ prepare_to_wait(&info->wait_q, &wait, TASK_INTERRUPTIBLE);
+ spin_unlock(&info->lock);
+ schedule();
+ finish_wait(&info->wait_q, &wait);
+ spin_lock(&info->lock);
+ }
+ spin_unlock(&info->lock);
+ result = (char)(unsigned long)filp->private_data;
+ if (put_user(result, buf))
+ return -EFAULT;
+ *ppos = 1;
+ return 1;
+}
+
+static int mqueue_notify_release(struct inode *inode, struct file *filp)
+{
+ struct mqueue_inode_info *info = MQUEUE_I(filp->f_dentry->d_inode);
+
+ spin_lock(&info->lock);
+ if (info->notify_owner && info->notify_filp == filp)
+ info->notify_owner = 0;
+ filp->private_data = NP_REMOVED;
+ spin_unlock(&info->lock);
+
+ return 0;
+}
+
+static void remove_notification(struct mqueue_inode_info *info)
+{
+ if (info->notify.sigev_notify == SIGEV_THREAD) {
+ info->notify_filp->private_data = NP_REMOVED;
+ wake_up(&info->wait_q);
+ }
+ info->notify_owner = 0;
+}
+
+/*
+ * Invoked when creating a new queue via sys_mq_open
+ */
+static struct file *do_create(struct dentry *dir, struct dentry *dentry,
+ int oflag, mode_t mode, struct mq_attr __user *u_attr)
+{
+ struct file *filp;
+ struct inode *inode;
+ struct mqueue_inode_info *info;
+ struct msg_msg **msgs = NULL;
+ struct mq_attr attr;
+ int ret;
+
+ if (u_attr != NULL) {
+ if (copy_from_user(&attr, u_attr, sizeof(attr)))
+ return ERR_PTR(-EFAULT);
+
+ if (attr.mq_maxmsg <= 0 || attr.mq_msgsize <= 0)
+ return ERR_PTR(-EINVAL);
+ if (capable(CAP_SYS_RESOURCE)) {
+ if (attr.mq_maxmsg > HARD_MSGMAX)
+ return ERR_PTR(-EINVAL);
+ } else {
+ if (attr.mq_maxmsg > msg_max ||
+ attr.mq_msgsize > msgsize_max)
+ return ERR_PTR(-EINVAL);
+ }
+ msgs = kmalloc(attr.mq_maxmsg * sizeof(*msgs), GFP_KERNEL);
+ if (!msgs)
+ return ERR_PTR(-ENOMEM);
+ } else {
+ msgs = NULL;
+ }
+
+ ret = vfs_create(dir->d_inode, dentry, mode, NULL);
+ if (ret) {
+ kfree(msgs);
+ return ERR_PTR(ret);
+ }
+
+ inode = dentry->d_inode;
+ info = MQUEUE_I(inode);
+
+ if (msgs) {
+ info->attr.mq_maxmsg = attr.mq_maxmsg;
+ info->attr.mq_msgsize = attr.mq_msgsize;
+ kfree(info->messages);
+ info->messages = msgs;
+ }
+
+ filp = dentry_open(dentry, mqueue_mnt, oflag);
+ if (!IS_ERR(filp))
+ dget(dentry);
+
+ return filp;
+}
+
+/* Opens existing queue */
+static struct file *do_open(struct dentry *dentry, int oflag)
+{
+static int oflag2acc[O_ACCMODE] = { MAY_READ, MAY_WRITE,
+ MAY_READ | MAY_WRITE };
+ struct file *filp;
+
+ if ((oflag & O_ACCMODE) == (O_RDWR | O_WRONLY))
+ return ERR_PTR(-EINVAL);
+
+ if (permission(dentry->d_inode, oflag2acc[oflag & O_ACCMODE], NULL))
+ return ERR_PTR(-EACCES);
+
+ filp = dentry_open(dentry, mqueue_mnt, oflag);
+
+ if (!IS_ERR(filp))
+ dget(dentry);
+
+ return filp;
+}
+
+asmlinkage long sys_mq_open(const char __user *u_name, int oflag, mode_t mode,
+ struct mq_attr __user *u_attr)
+{
+ struct dentry *dentry;
+ struct file *filp;
+ char *name;
+ int fd, error;
+
+ if (IS_ERR(name = getname(u_name)))
+ return PTR_ERR(name);
+
+ fd = get_unused_fd();
+ if (fd < 0)
+ goto out_putname;
+
+ down(&mqueue_mnt->mnt_root->d_inode->i_sem);
+ dentry = lookup_one_len(name, mqueue_mnt->mnt_root, strlen(name));
+ if (IS_ERR(dentry)) {
+ error = PTR_ERR(dentry);
+ goto out_err;
+ }
+ mntget(mqueue_mnt);
+
+ if (oflag & O_CREAT) {
+ if (dentry->d_inode) { /* entry already exists */
+ filp = (oflag & O_EXCL) ? ERR_PTR(-EEXIST) :
+ do_open(dentry, oflag);
+ } else {
+ filp = do_create(mqueue_mnt->mnt_root, dentry,
+ oflag, mode, u_attr);
+ }
+ } else
+ filp = (dentry->d_inode) ? do_open(dentry, oflag) :
+ ERR_PTR(-ENOENT);
+
+ dput(dentry);
+
+ if (IS_ERR(filp)) {
+ error = PTR_ERR(filp);
+ goto out_putfd;
+ }
+
+ fd_install(fd, filp);
+ goto out_upsem;
+
+out_putfd:
+ mntput(mqueue_mnt);
+ put_unused_fd(fd);
+out_err:
+ fd = error;
+out_upsem:
+ up(&mqueue_mnt->mnt_root->d_inode->i_sem);
+out_putname:
+ putname(name);
+ return fd;
+}
+
+asmlinkage long sys_mq_unlink(const char __user *u_name)
+{
+ int err;
+ char *name;
+ struct dentry *dentry;
+ struct inode *inode = NULL;
+
+ name = getname(u_name);
+ if (IS_ERR(name))
+ return PTR_ERR(name);
+
+ down(&mqueue_mnt->mnt_root->d_inode->i_sem);
+ dentry = lookup_one_len(name, mqueue_mnt->mnt_root, strlen(name));
+ if (IS_ERR(dentry)) {
+ err = PTR_ERR(dentry);
+ goto out_unlock;
+ }
+
+ if (!dentry->d_inode) {
+ err = -ENOENT;
+ goto out_err;
+ }
+
+ if (permission(dentry->d_inode, MAY_WRITE, NULL)) {
+ err = -EACCES;
+ goto out_err;
+ }
+ inode = dentry->d_inode;
+ if (inode)
+ atomic_inc(&inode->i_count);
+
+ err = vfs_unlink(dentry->d_parent->d_inode, dentry);
+out_err:
+ dput(dentry);
+
+out_unlock:
+ up(&mqueue_mnt->mnt_root->d_inode->i_sem);
+ putname(name);
+ if (inode)
+ iput(inode);
+
+ return err;
+}
+
+/* Pipelined send and receive functions.
+ *
+ * If a receiver finds no waiting message, then it registers itself in the
+ * list of waiting receivers. A sender checks that list before adding the new
+ * message into the message array. If there is a waiting receiver, then it
+ * bypasses the message array and directly hands the message over to the
+ * receiver.
+ * The receiver accepts the message and returns without grabbing the queue
+ * spinlock. Therefore an intermediate STATE_PENDING state and memory barriers
+ * are necessary. The same algorithm is used for sysv semaphores, see
+ * ipc/sem.c fore more details.
+ *
+ * The same algorithm is used for senders.
+ */
+
+/* pipelined_send() - send a message directly to the task waiting in
+ * sys_mq_timedreceive() (without inserting message into a queue). */
+static inline void pipelined_send(struct mqueue_inode_info *info,
+ struct msg_msg *message,
+ struct ext_wait_queue *receiver)
+{
+ receiver->msg = message;
+ list_del(&receiver->list);
+ receiver->state = STATE_PENDING;
+ wake_up_process(receiver->task);
+ wmb();
+ receiver->state = STATE_READY;
+}
+
+/* pipelined_receive() - if there is task waiting in sys_mq_timedsend()
+ * gets its message and put to the queue (we have one free place for sure). */
+static inline void pipelined_receive(struct mqueue_inode_info *info)
+{
+ struct ext_wait_queue *sender = wq_get_first_waiter(info, SEND);
+
+ if (!sender) {
+ /* for poll */
+ wake_up_interruptible(&info->wait_q);
+ return;
+ }
+ msg_insert(sender->msg, info);
+ list_del(&sender->list);
+ sender->state = STATE_PENDING;
+ wake_up_process(sender->task);
+ wmb();
+ sender->state = STATE_READY;
+}
+
+asmlinkage long sys_mq_timedsend(mqd_t mqdes, const char __user *u_msg_ptr,
+ size_t msg_len, unsigned int msg_prio,
+ const struct timespec __user *u_abs_timeout)
+{
+ struct file *filp;
+ struct inode *inode;
+ struct ext_wait_queue wait;
+ struct ext_wait_queue *receiver;
+ struct msg_msg *msg_ptr;
+ struct mqueue_inode_info *info;
+ long timeout;
+ int ret;
+
+ if (unlikely(msg_prio >= (unsigned long) MQ_PRIO_MAX))
+ return -EINVAL;
+
+ if (unlikely((timeout = prepare_timeout(u_abs_timeout)) < 0))
+ return timeout;
+
+ ret = -EBADF;
+ filp = fget(mqdes);
+ if (unlikely(!filp))
+ goto out;
+
+ inode = filp->f_dentry->d_inode;
+ if (unlikely(filp->f_op != &mqueue_file_operations))
+ goto out_fput;
+ info = MQUEUE_I(inode);
+
+ if (unlikely(!(filp->f_mode & FMODE_WRITE)))
+ goto out_fput;
+
+ if (unlikely(msg_len > info->attr.mq_msgsize)) {
+ ret = -EMSGSIZE;
+ goto out_fput;
+ }
+
+ /* First try to allocate memory, before doing anything with
+ * existing queues. */
+ msg_ptr = load_msg((void *)u_msg_ptr, msg_len);
+ if (unlikely(IS_ERR(msg_ptr))) {
+ ret = PTR_ERR(msg_ptr);
+ goto out_fput;
+ }
+ msg_ptr->m_ts = msg_len;
+ msg_ptr->m_type = msg_prio;
+
+ spin_lock(&info->lock);
+
+ if (info->attr.mq_curmsgs == info->attr.mq_maxmsg) {
+ if (filp->f_flags & O_NONBLOCK) {
+ spin_unlock(&info->lock);
+ ret = -EAGAIN;
+ } else {
+ wait.task = current;
+ wait.msg = (void *) msg_ptr;
+ wait.state = STATE_NONE;
+ ret = wq_sleep(info, SEND, timeout, &wait);
+ if (ret < 0)
+ free_msg(msg_ptr);
+ }
+ } else {
+ receiver = wq_get_first_waiter(info, RECV);
+ if (receiver) {
+ pipelined_send(info, msg_ptr, receiver);
+ } else {
+ /* adds message to the queue */
+ msg_insert(msg_ptr, info);
+ __do_notify(info);
+ }
+ inode->i_atime = inode->i_mtime = inode->i_ctime =
+ CURRENT_TIME;
+ spin_unlock(&info->lock);
+ ret = 0;
+ }
+out_fput:
+ fput(filp);
+out:
+ return ret;
+}
+
+asmlinkage ssize_t sys_mq_timedreceive(mqd_t mqdes, char __user *u_msg_ptr,
+ size_t msg_len, unsigned int __user *u_msg_prio,
+ const struct timespec __user *u_abs_timeout)
+{
+ long timeout;
+ ssize_t ret;
+ struct msg_msg *msg_ptr;
+ struct file *filp;
+ struct inode *inode;
+ struct mqueue_inode_info *info;
+ struct ext_wait_queue wait;
+
+
+ if (unlikely((timeout = prepare_timeout(u_abs_timeout)) < 0))
+ return timeout;
+
+ ret = -EBADF;
+ filp = fget(mqdes);
+ if (unlikely(!filp))
+ goto out;
+
+ inode = filp->f_dentry->d_inode;
+ if (unlikely(filp->f_op != &mqueue_file_operations))
+ goto out_fput;
+ info = MQUEUE_I(inode);
+
+ if (unlikely(!(filp->f_mode & FMODE_READ)))
+ goto out_fput;
+
+ /* checks if buffer is big enough */
+ if (unlikely(msg_len < info->attr.mq_msgsize)) {
+ ret = -EMSGSIZE;
+ goto out_fput;
+ }
+
+ spin_lock(&info->lock);
+ if (info->attr.mq_curmsgs == 0) {
+ if (filp->f_flags & O_NONBLOCK) {
+ spin_unlock(&info->lock);
+ ret = -EAGAIN;
+ msg_ptr = NULL;
+ } else {
+ wait.task = current;
+ wait.state = STATE_NONE;
+ ret = wq_sleep(info, RECV, timeout, &wait);
+ msg_ptr = wait.msg;
+ }
+ } else {
+ msg_ptr = msg_get(info);
+
+ inode->i_atime = inode->i_mtime = inode->i_ctime =
+ CURRENT_TIME;
+
+ /* There is now free space in queue. */
+ pipelined_receive(info);
+ spin_unlock(&info->lock);
+ ret = 0;
+ }
+ if (ret == 0) {
+ ret = msg_ptr->m_ts;
+
+ if ((u_msg_prio && put_user(msg_ptr->m_type, u_msg_prio)) ||
+ store_msg(u_msg_ptr, msg_ptr, msg_ptr->m_ts)) {
+ ret = -EFAULT;
+ }
+ free_msg(msg_ptr);
+ }
+out_fput:
+ fput(filp);
+out:
+ return ret;
+}
+
+/*
+ * Notes: the case when user wants us to deregister (with NULL as pointer
+ * or SIGEV_NONE) and he isn't currently owner of notification will be
+ * silently discarded. It isn't explicitly defined in the POSIX.
+ */
+asmlinkage long sys_mq_notify(mqd_t mqdes,
+ const struct sigevent __user *u_notification)
+{
+ int ret, fd;
+ struct file *filp, *nfilp;
+ struct inode *inode;
+ struct sigevent notification;
+ struct mqueue_inode_info *info;
+
+ if (u_notification == NULL) {
+ notification.sigev_notify = SIGEV_NONE;
+ } else {
+ if (copy_from_user(¬ification, u_notification,
+ sizeof(struct sigevent)))
+ return -EFAULT;
+
+ if (unlikely(notification.sigev_notify != SIGEV_NONE &&
+ notification.sigev_notify != SIGEV_SIGNAL &&
+ notification.sigev_notify != SIGEV_THREAD))
+ return -EINVAL;
+ if (notification.sigev_notify == SIGEV_SIGNAL &&
+ (notification.sigev_signo < 0 ||
+ notification.sigev_signo > _NSIG)) {
+ return -EINVAL;
+ }
+ }
+
+ ret = -EBADF;
+ filp = fget(mqdes);
+ if (!filp)
+ goto out;
+
+ inode = filp->f_dentry->d_inode;
+ if (unlikely(filp->f_op != &mqueue_file_operations))
+ goto out_fput;
+ info = MQUEUE_I(inode);
+
+ ret = 0;
+ if (notification.sigev_notify == SIGEV_THREAD) {
+ ret = get_unused_fd();
+ if (ret < 0)
+ goto out_fput;
+ fd = ret;
+ nfilp = get_empty_filp();
+ if (!nfilp) {
+ ret = -ENFILE;
+ goto out_dropfd;
+ }
+ nfilp->private_data = NP_NONE;
+ nfilp->f_op = &mqueue_notify_fops;
+ nfilp->f_vfsmnt = mntget(mqueue_mnt);
+ nfilp->f_dentry = dget(filp->f_dentry);
+ nfilp->f_mapping = filp->f_dentry->d_inode->i_mapping;
+ nfilp->f_flags = O_RDONLY;
+ nfilp->f_mode = FMODE_READ;
+ } else {
+ nfilp = NULL;
+ fd = -1;
+ }
+
+ spin_lock(&info->lock);
+
+ if (notification.sigev_notify == SIGEV_NONE) {
+ if (info->notify_owner == current->tgid) {
+ remove_notification(info);
+ inode->i_atime = inode->i_ctime = CURRENT_TIME;
+ }
+ } else if (info->notify_owner) {
+ ret = -EBUSY;
+ } else if (notification.sigev_notify == SIGEV_THREAD) {
+ info->notify_filp = nfilp;
+ fd_install(fd, nfilp);
+ ret = fd;
+ fd = -1;
+ nfilp = NULL;
+ info->notify.sigev_notify = SIGEV_THREAD;
+ info->notify_owner = current->tgid;
+ inode->i_atime = inode->i_ctime = CURRENT_TIME;
+ } else {
+ info->notify.sigev_signo = notification.sigev_signo;
+ info->notify.sigev_value = notification.sigev_value;
+ info->notify.sigev_notify = SIGEV_SIGNAL;
+ info->notify_owner = current->tgid;
+ inode->i_atime = inode->i_ctime = CURRENT_TIME;
+ }
+ spin_unlock(&info->lock);
+out_dropfd:
+ if (fd != -1)
+ put_unused_fd(fd);
+out_fput:
+ fput(filp);
+out:
+ return ret;
+}
+
+asmlinkage long sys_mq_getsetattr(mqd_t mqdes,
+ const struct mq_attr __user *u_mqstat,
+ struct mq_attr __user *u_omqstat)
+{
+ int ret;
+ struct mq_attr mqstat, omqstat;
+ struct file *filp;
+ struct inode *inode;
+ struct mqueue_inode_info *info;
+
+ if (u_mqstat != NULL) {
+ if (copy_from_user(&mqstat, u_mqstat, sizeof(struct mq_attr)))
+ return -EFAULT;
+ if (mqstat.mq_flags & (~O_NONBLOCK))
+ return -EINVAL;
+ }
+
+ ret = -EBADF;
+ filp = fget(mqdes);
+ if (!filp)
+ goto out;
+
+ inode = filp->f_dentry->d_inode;
+ if (unlikely(filp->f_op != &mqueue_file_operations))
+ goto out_fput;
+ info = MQUEUE_I(inode);
+
+ spin_lock(&info->lock);
+
+ omqstat = info->attr;
+ omqstat.mq_flags = filp->f_flags & O_NONBLOCK;
+ if (u_mqstat) {
+ if (mqstat.mq_flags & O_NONBLOCK)
+ filp->f_flags |= O_NONBLOCK;
+ else
+ filp->f_flags &= ~O_NONBLOCK;
+
+ inode->i_atime = inode->i_ctime = CURRENT_TIME;
+ }
+
+ spin_unlock(&info->lock);
+
+ ret = 0;
+ if (u_omqstat != NULL && copy_to_user(u_omqstat, &omqstat,
+ sizeof(struct mq_attr)))
+ ret = -EFAULT;
+
+out_fput:
+ fput(filp);
+out:
+ return ret;
+}
+
+static struct inode_operations mqueue_dir_inode_operations = {
+ .lookup = simple_lookup,
+ .create = mqueue_create,
+ .unlink = mqueue_unlink,
+};
+
+static struct file_operations mqueue_file_operations = {
+ .flush = mqueue_flush_file,
+ .poll = mqueue_poll_file,
+ .read = mqueue_read_file,
+};
+
+static struct file_operations mqueue_notify_fops = {
+ .poll = mqueue_notify_poll,
+ .read = mqueue_notify_read,
+ .release = mqueue_notify_release,
+};
+
+
+static struct super_operations mqueue_super_ops = {
+ .alloc_inode = mqueue_alloc_inode,
+ .destroy_inode = mqueue_destroy_inode,
+ .statfs = simple_statfs,
+ .delete_inode = mqueue_delete_inode,
+ .drop_inode = generic_delete_inode,
+};
+
+static struct file_system_type mqueue_fs_type = {
+ .name = "mqueue",
+ .get_sb = mqueue_get_sb,
+ .kill_sb = kill_litter_super,
+};
+
+static int msg_max_limit_min = DFLT_MSGMAX;
+static int msg_max_limit_max = HARD_MSGMAX;
+
+static int msg_maxsize_limit_min = DFLT_MSGSIZEMAX;
+static int msg_maxsize_limit_max = INT_MAX;
+
+static ctl_table mq_sysctls[] = {
+ {
+ .ctl_name = CTL_QUEUESMAX,
+ .procname = "queues_max",
+ .data = &queues_max,
+ .maxlen = sizeof(int),
+ .mode = 0644,
+ .proc_handler = &proc_dointvec,
+ },
+ {
+ .ctl_name = CTL_MSGMAX,
+ .procname = "msg_max",
+ .data = &msg_max,
+ .maxlen = sizeof(int),
+ .mode = 0644,
+ .proc_handler = &proc_dointvec_minmax,
+ .extra1 = &msg_max_limit_min,
+ .extra2 = &msg_max_limit_max,
+ },
+ {
+ .ctl_name = CTL_MSGSIZEMAX,
+ .procname = "msgsize_max",
+ .data = &msgsize_max,
+ .maxlen = sizeof(int),
+ .mode = 0644,
+ .proc_handler = &proc_dointvec_minmax,
+ .extra1 = &msg_maxsize_limit_min,
+ .extra2 = &msg_maxsize_limit_max,
+ },
+ { .ctl_name = 0 }
+};
+
+static ctl_table mq_sysctl_dir[] = {
+ {
+ .ctl_name = FS_MQUEUE,
+ .procname = "mqueue",
+ .mode = 0555,
+ .child = mq_sysctls,
+ },
+ { .ctl_name = 0 }
+};
+
+static ctl_table mq_sysctl_root[] = {
+ {
+ .ctl_name = CTL_FS,
+ .procname = "fs",
+ .mode = 0555,
+ .child = mq_sysctl_dir,
+ },
+ { .ctl_name = 0 }
+};
+
+static int __init init_mqueue_fs(void)
+{
+ int error;
+
+ mqueue_inode_cachep = kmem_cache_create("mqueue_inode_cache",
+ sizeof(struct mqueue_inode_info), 0,
+ SLAB_HWCACHE_ALIGN, init_once, NULL);
+ if (mqueue_inode_cachep == NULL)
+ return -ENOMEM;
+
+ mq_sysctl_table = register_sysctl_table(mq_sysctl_root, 0);
+ if (!mq_sysctl_table) {
+ error = -ENOMEM;
+ goto out_cache;
+ }
+
+ error = register_filesystem(&mqueue_fs_type);
+ if (error)
+ goto out_sysctl;
+
+ if (IS_ERR(mqueue_mnt = kern_mount(&mqueue_fs_type))) {
+ error = PTR_ERR(mqueue_mnt);
+ goto out_filesystem;
+ }
+
+ /* internal initialization - not common for vfs */
+ queues_count = 0;
+ spin_lock_init(&mq_lock);
+
+ return 0;
+
+out_filesystem:
+ unregister_filesystem(&mqueue_fs_type);
+out_sysctl:
+ unregister_sysctl_table(mq_sysctl_table);
+out_cache:
+ if (kmem_cache_destroy(mqueue_inode_cachep)) {
+ printk(KERN_INFO
+ "mqueue_inode_cache: not all structures were freed\n");
+ }
+ return error;
+}
+
+__initcall(init_mqueue_fs);
--- diff/ipc/msgutil.c 1970-01-01 00:00:00.000000000 +0000
+++ source/ipc/msgutil.c 2004-03-16 11:27:48.635793896 +0000
@@ -0,0 +1,127 @@
+/*
+ * linux/ipc/util.c
+ * Copyright (C) 1999, 2004 Manfred Spraul
+ *
+ * This file is released under GNU General Public Licence version 2 or
+ * (at your option) any later version.
+ *
+ * See the file COPYING for more details.
+ */
+
+#include <linux/spinlock.h>
+#include <linux/init.h>
+#include <linux/security.h>
+#include <linux/slab.h>
+#include <linux/ipc.h>
+#include <asm/uaccess.h>
+
+#include "util.h"
+
+struct msg_msgseg {
+ struct msg_msgseg* next;
+ /* the next part of the message follows immediately */
+};
+
+#define DATALEN_MSG (PAGE_SIZE-sizeof(struct msg_msg))
+#define DATALEN_SEG (PAGE_SIZE-sizeof(struct msg_msgseg))
+
+struct msg_msg *load_msg(void __user *src, int len)
+{
+ struct msg_msg *msg;
+ struct msg_msgseg **pseg;
+ int err;
+ int alen;
+
+ alen = len;
+ if (alen > DATALEN_MSG)
+ alen = DATALEN_MSG;
+
+ msg = (struct msg_msg *)kmalloc(sizeof(*msg) + alen, GFP_KERNEL);
+ if (msg == NULL)
+ return ERR_PTR(-ENOMEM);
+
+ msg->next = NULL;
+ msg->security = NULL;
+
+ if (copy_from_user(msg + 1, src, alen)) {
+ err = -EFAULT;
+ goto out_err;
+ }
+
+ len -= alen;
+ src = ((char *)src) + alen;
+ pseg = &msg->next;
+ while (len > 0) {
+ struct msg_msgseg *seg;
+ alen = len;
+ if (alen > DATALEN_SEG)
+ alen = DATALEN_SEG;
+ seg = (struct msg_msgseg *)kmalloc(sizeof(*seg) + alen,
+ GFP_KERNEL);
+ if (seg == NULL) {
+ err = -ENOMEM;
+ goto out_err;
+ }
+ *pseg = seg;
+ seg->next = NULL;
+ if (copy_from_user(seg + 1, src, alen)) {
+ err = -EFAULT;
+ goto out_err;
+ }
+ pseg = &seg->next;
+ len -= alen;
+ src = ((char *)src) + alen;
+ }
+
+ err = security_msg_msg_alloc(msg);
+ if (err)
+ goto out_err;
+
+ return msg;
+
+out_err:
+ free_msg(msg);
+ return ERR_PTR(err);
+}
+
+int store_msg(void __user *dest, struct msg_msg *msg, int len)
+{
+ int alen;
+ struct msg_msgseg *seg;
+
+ alen = len;
+ if (alen > DATALEN_MSG)
+ alen = DATALEN_MSG;
+ if (copy_to_user(dest, msg + 1, alen))
+ return -1;
+
+ len -= alen;
+ dest = ((char *)dest) + alen;
+ seg = msg->next;
+ while (len > 0) {
+ alen = len;
+ if (alen > DATALEN_SEG)
+ alen = DATALEN_SEG;
+ if (copy_to_user(dest, seg + 1, alen))
+ return -1;
+ len -= alen;
+ dest = ((char *)dest) + alen;
+ seg = seg->next;
+ }
+ return 0;
+}
+
+void free_msg(struct msg_msg *msg)
+{
+ struct msg_msgseg *seg;
+
+ security_msg_msg_free(msg);
+
+ seg = msg->next;
+ kfree(msg);
+ while (seg != NULL) {
+ struct msg_msgseg *tmp = seg->next;
+ kfree(seg);
+ seg = tmp;
+ }
+}
--- diff/kernel/audit.c 1970-01-01 00:00:00.000000000 +0000
+++ source/kernel/audit.c 2004-03-16 11:27:48.640793136 +0000
@@ -0,0 +1,816 @@
+/* audit.c -- Auditing support -*- linux-c -*-
+ * Gateway between the kernel (e.g., selinux) and the user-space audit daemon.
+ * System-call specific features have moved to auditsc.c
+ *
+ * Copyright 2003-2004 Red Hat Inc., Durham, North Carolina.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Written by Rickard E. (Rik) Faith <faith@redhat.com>
+ *
+ * Goals: 1) Integrate fully with SELinux.
+ * 2) Minimal run-time overhead:
+ * a) Minimal when syscall auditing is disabled (audit_enable=0).
+ * b) Small when syscall auditing is enabled and no audit record
+ * is generated (defer as much work as possible to record
+ * generation time):
+ * i) context is allocated,
+ * ii) names from getname are stored without a copy, and
+ * iii) inode information stored from path_lookup.
+ * 3) Ability to disable syscall auditing at boot time (audit=0).
+ * 4) Usable by other parts of the kernel (if audit_log* is called,
+ * then a syscall record will be generated automatically for the
+ * current syscall).
+ * 5) Netlink interface to user-space.
+ * 6) Support low-overhead kernel-based filtering to minimize the
+ * information that must be passed to user-space.
+ *
+ * Example user-space utilities: http://people.redhat.com/faith/audit/
+ */
+
+#include <linux/init.h>
+#include <asm/atomic.h>
+#include <asm/types.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+
+#include <linux/audit.h>
+
+#include <net/sock.h>
+#include <linux/skbuff.h>
+#include <linux/netlink.h>
+
+/* No auditing will take place until audit_initialized != 0.
+ * (Initialization happens after skb_init is called.) */
+static int audit_initialized;
+
+/* No syscall auditing will take place unless audit_enabled != 0. */
+int audit_enabled;
+
+/* Default state when kernel boots without any parameters. */
+static int audit_default;
+
+/* If auditing cannot proceed, audit_failure selects what happens. */
+static int audit_failure = AUDIT_FAIL_PRINTK;
+
+/* If audit records are to be written to the netlink socket, audit_pid
+ * contains the (non-zero) pid. */
+static int audit_pid;
+
+/* If audit_limit is non-zero, limit the rate of sending audit records
+ * to that number per second. This prevents DoS attacks, but results in
+ * audit records being dropped. */
+static int audit_rate_limit;
+
+/* Number of outstanding audit_buffers allowed. */
+static int audit_backlog_limit = 64;
+static atomic_t audit_backlog = ATOMIC_INIT(0);
+
+/* Records can be lost in several ways:
+ 0) [suppressed in audit_alloc]
+ 1) out of memory in audit_log_start [kmalloc of struct audit_buffer]
+ 2) out of memory in audit_log_move [alloc_skb]
+ 3) suppressed due to audit_rate_limit
+ 4) suppressed due to audit_backlog_limit
+*/
+static atomic_t audit_lost = ATOMIC_INIT(0);
+
+/* The netlink socket. */
+static struct sock *audit_sock;
+
+/* There are two lists of audit buffers. The txlist contains audit
+ * buffers that cannot be sent immediately to the netlink device because
+ * we are in an irq context (these are sent later in a tasklet).
+ *
+ * The second list is a list of pre-allocated audit buffers (if more
+ * than AUDIT_MAXFREE are in use, the audit buffer is freed instead of
+ * being placed on the freelist). */
+static spinlock_t audit_txlist_lock = SPIN_LOCK_UNLOCKED;
+static spinlock_t audit_freelist_lock = SPIN_LOCK_UNLOCKED;
+static int audit_freelist_count = 0;
+static LIST_HEAD(audit_txlist);
+static LIST_HEAD(audit_freelist);
+
+/* There are three lists of rules -- one to search at task creation
+ * time, one to search at syscall entry time, and another to search at
+ * syscall exit time. */
+static LIST_HEAD(audit_tsklist);
+static LIST_HEAD(audit_entlist);
+static LIST_HEAD(audit_extlist);
+
+/* AUDIT_BUFSIZ is the size of the temporary buffer used for formatting
+ * audit records. Since printk uses a 1024 byte buffer, this buffer
+ * should be at least that large. */
+#define AUDIT_BUFSIZ 1024
+
+/* AUDIT_MAXFREE is the number of empty audit_buffers we keep on the
+ * audit_freelist. Doing so eliminates many kmalloc/kfree calls. */
+#define AUDIT_MAXFREE (2*NR_CPUS)
+
+/* The audit_buffer is used when formatting an audit record. The caller
+ * locks briefly to get the record off the freelist or to allocate the
+ * buffer, and locks briefly to send the buffer to the netlink layer or
+ * to place it on a transmit queue. Multiple audit_buffers can be in
+ * use simultaneously. */
+struct audit_buffer {
+ struct list_head list;
+ struct sk_buff_head sklist; /* formatted skbs ready to send */
+ struct audit_context *ctx; /* NULL or associated context */
+ int len; /* used area of tmp */
+ char tmp[AUDIT_BUFSIZ];
+
+ /* Pointer to header and contents */
+ struct nlmsghdr *nlh;
+ int total;
+ int type;
+ int pid;
+ int count; /* Times requeued */
+};
+
+struct audit_entry {
+ struct list_head list;
+ struct audit_rule rule;
+};
+
+static void audit_panic(const char *message)
+{
+ switch (audit_failure)
+ {
+ case AUDIT_FAIL_SILENT:
+ break;
+ case AUDIT_FAIL_PRINTK:
+ printk(KERN_ERR "audit: %s\n", message);
+ break;
+ case AUDIT_FAIL_PANIC:
+ panic(message);
+ break;
+ }
+}
+
+static inline int audit_rate_check(void)
+{
+ static unsigned long last_check = 0;
+ static int messages = 0;
+ static spinlock_t lock = SPIN_LOCK_UNLOCKED;
+ unsigned long flags;
+ unsigned long now;
+ unsigned long elapsed;
+ int retval = 0;
+
+ if (!audit_rate_limit) return 1;
+
+ spin_lock_irqsave(&lock, flags);
+ if (++messages < audit_rate_limit) {
+ retval = 1;
+ } else {
+ now = jiffies;
+ elapsed = now - last_check;
+ if (elapsed > HZ) {
+ last_check = now;
+ messages = 0;
+ retval = 1;
+ }
+ }
+ spin_unlock_irqrestore(&lock, flags);
+
+ return retval;
+}
+
+/* Emit at least 1 message per second, even if audit_rate_check is
+ * throttling. */
+void audit_log_lost(const char *message)
+{
+ static unsigned long last_msg = 0;
+ static spinlock_t lock = SPIN_LOCK_UNLOCKED;
+ unsigned long flags;
+ unsigned long now;
+ int print;
+
+ atomic_inc(&audit_lost);
+
+ print = (audit_failure == AUDIT_FAIL_PANIC || !audit_rate_limit);
+
+ if (!print) {
+ spin_lock_irqsave(&lock, flags);
+ now = jiffies;
+ if (now - last_msg > HZ) {
+ print = 1;
+ last_msg = now;
+ }
+ spin_unlock_irqrestore(&lock, flags);
+ }
+
+ if (print) {
+ printk(KERN_WARNING
+ "audit: audit_lost=%d audit_backlog=%d"
+ " audit_rate_limit=%d audit_backlog_limit=%d\n",
+ atomic_read(&audit_lost),
+ atomic_read(&audit_backlog),
+ audit_rate_limit,
+ audit_backlog_limit);
+ audit_panic(message);
+ }
+
+}
+
+int audit_set_rate_limit(int limit)
+{
+ int old = audit_rate_limit;
+ audit_rate_limit = limit;
+ audit_log(current->audit_context, "audit_rate_limit=%d old=%d",
+ audit_rate_limit, old);
+ return old;
+}
+
+int audit_set_backlog_limit(int limit)
+{
+ int old = audit_backlog_limit;
+ audit_backlog_limit = limit;
+ audit_log(current->audit_context, "audit_backlog_limit=%d old=%d",
+ audit_backlog_limit, old);
+ return old;
+}
+
+int audit_set_enabled(int state)
+{
+ int old = audit_enabled;
+ if (state != 0 && state != 1)
+ return -EINVAL;
+ audit_enabled = state;
+ audit_log(current->audit_context, "audit_enabled=%d old=%d",
+ audit_enabled, old);
+ return old;
+}
+
+int audit_set_failure(int state)
+{
+ int old = audit_failure;
+ if (state != AUDIT_FAIL_SILENT
+ && state != AUDIT_FAIL_PRINTK
+ && state != AUDIT_FAIL_PANIC)
+ return -EINVAL;
+ audit_failure = state;
+ audit_log(current->audit_context, "audit_failure=%d old=%d",
+ audit_failure, old);
+ return old;
+}
+
+#ifdef CONFIG_NET
+void audit_send_reply(int pid, int seq, int type, int done, int multi,
+ void *payload, int size)
+{
+ struct sk_buff *skb;
+ struct nlmsghdr *nlh;
+ int len = NLMSG_SPACE(size);
+ void *data;
+ int flags = multi ? NLM_F_MULTI : 0;
+ int t = done ? NLMSG_DONE : type;
+
+ skb = alloc_skb(len, GFP_KERNEL);
+ if (!skb)
+ goto nlmsg_failure;
+
+ nlh = NLMSG_PUT(skb, pid, seq, t, len - sizeof(*nlh));
+ nlh->nlmsg_flags = flags;
+ data = NLMSG_DATA(nlh);
+ memcpy(data, payload, size);
+ netlink_unicast(audit_sock, skb, pid, MSG_DONTWAIT);
+ return;
+
+nlmsg_failure: /* Used by NLMSG_PUT */
+ if (skb)
+ kfree_skb(skb);
+}
+
+static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
+{
+ u32 uid, pid, seq;
+ void *data;
+ struct audit_status *status_get, status_set;
+ struct audit_login *login;
+ int err = 0;
+ struct audit_buffer *ab;
+
+ pid = NETLINK_CREDS(skb)->pid;
+ uid = NETLINK_CREDS(skb)->uid;
+ seq = nlh->nlmsg_seq;
+ data = NLMSG_DATA(nlh);
+
+ switch (nlh->nlmsg_type) {
+ case AUDIT_GET:
+ status_set.enabled = audit_enabled;
+ status_set.failure = audit_failure;
+ status_set.pid = audit_pid;
+ status_set.rate_limit = audit_rate_limit;
+ status_set.backlog_limit = audit_backlog_limit;
+ status_set.lost = atomic_read(&audit_lost);
+ status_set.backlog = atomic_read(&audit_backlog);
+ audit_send_reply(pid, seq, AUDIT_GET, 0, 0,
+ &status_set, sizeof(status_set));
+ break;
+ case AUDIT_SET:
+ if (!capable(CAP_SYS_ADMIN))
+ return -EPERM;
+ status_get = (struct audit_status *)data;
+ if (status_get->mask & AUDIT_STATUS_ENABLED) {
+ err = audit_set_enabled(status_get->enabled);
+ if (err < 0) return err;
+ }
+ if (status_get->mask & AUDIT_STATUS_FAILURE) {
+ err = audit_set_failure(status_get->failure);
+ if (err < 0) return err;
+ }
+ if (status_get->mask & AUDIT_STATUS_PID) {
+ int old = audit_pid;
+ audit_pid = status_get->pid;
+ audit_log(current->audit_context,
+ "audit_pid=%d old=%d", audit_pid, old);
+ }
+ if (status_get->mask & AUDIT_STATUS_RATE_LIMIT)
+ audit_set_rate_limit(status_get->rate_limit);
+ if (status_get->mask & AUDIT_STATUS_BACKLOG_LIMIT)
+ audit_set_backlog_limit(status_get->backlog_limit);
+ break;
+ case AUDIT_USER:
+ ab = audit_log_start(NULL);
+ if (!ab)
+ break; /* audit_panic has been called */
+ audit_log_format(ab,
+ "user pid=%d uid=%d length=%d msg='%.1024s'",
+ pid, uid,
+ (int)(nlh->nlmsg_len
+ - ((char *)data - (char *)nlh)),
+ (char *)data);
+ ab->type = AUDIT_USER;
+ ab->pid = pid;
+ audit_log_end(ab);
+ break;
+ case AUDIT_LOGIN:
+ if (!capable(CAP_SYS_ADMIN))
+ return -EPERM;
+ login = (struct audit_login *)data;
+ ab = audit_log_start(NULL);
+ if (ab) {
+ audit_log_format(ab, "login pid=%d uid=%d loginuid=%d"
+ " length=%d msg='%.1024s'",
+ pid, uid,
+ login->loginuid,
+ login->msglen,
+ login->msg);
+ ab->type = AUDIT_LOGIN;
+ ab->pid = pid;
+ audit_log_end(ab);
+ }
+#ifdef CONFIG_AUDITSYSCALL
+ err = audit_set_loginuid(current->audit_context,
+ login->loginuid);
+#endif
+ break;
+ case AUDIT_LIST:
+ case AUDIT_ADD:
+ case AUDIT_DEL:
+#ifdef CONFIG_AUDITSYSCALL
+ err = audit_receive_filter(nlh->nlmsg_type, pid, uid, seq,
+ data);
+#else
+ err = -EOPNOTSUPP;
+#endif
+ break;
+ default:
+ err = -EINVAL;
+ break;
+ }
+
+ return err < 0 ? err : 0;
+}
+
+/* Get message from skb (based on rtnetlink_rcv_skb). Each message is
+ * processed by audit_receive_msg. Malformed skbs with wrong length are
+ * discarded silently. */
+static int audit_receive_skb(struct sk_buff *skb)
+{
+ int err;
+ struct nlmsghdr *nlh;
+ u32 rlen;
+
+ while (skb->len >= NLMSG_SPACE(0)) {
+ nlh = (struct nlmsghdr *)skb->data;
+ if (nlh->nlmsg_len < sizeof(*nlh) || skb->len < nlh->nlmsg_len)
+ return 0;
+ rlen = NLMSG_ALIGN(nlh->nlmsg_len);
+ if (rlen > skb->len)
+ rlen = skb->len;
+ if ((err = audit_receive_msg(skb, nlh))) {
+ netlink_ack(skb, nlh, -err);
+ } else if (nlh->nlmsg_flags & NLM_F_ACK)
+ netlink_ack(skb, nlh, 0);
+ skb_pull(skb, rlen);
+ }
+ return 0;
+}
+
+/* Receive messages from netlink socket. */
+static void audit_receive(struct sock *sk, int length)
+{
+ struct sk_buff *skb;
+
+ /* FIXME: this must not cause starvation */
+ while ((skb = skb_dequeue(&sk->sk_receive_queue))) {
+ if (audit_receive_skb(skb) && skb->len)
+ skb_queue_head(&sk->sk_receive_queue, skb);
+ else
+ kfree_skb(skb);
+ }
+}
+
+/* Move data from tmp buffer into an skb. This is an extra copy, and
+ * that is unfortunate. However, the copy will only occur when a record
+ * is being written to user space, which is already a high-overhead
+ * operation. (Elimination of the copy is possible, for example, by
+ * writing directly into a pre-allocated skb, at the cost of wasting
+ * memory. */
+static void audit_log_move(struct audit_buffer *ab)
+{
+ struct sk_buff *skb;
+ char *start;
+ int extra = ab->nlh ? 0 : NLMSG_SPACE(0);
+
+ skb = skb_peek(&ab->sklist);
+ if (!skb || skb_tailroom(skb) <= ab->len + extra) {
+ skb = alloc_skb(2 * ab->len + extra, GFP_ATOMIC);
+ if (!skb) {
+ ab->len = 0; /* Lose information in ab->tmp */
+ audit_log_lost("out of memory in audit_log_move");
+ return;
+ }
+ __skb_queue_tail(&ab->sklist, skb);
+ if (!ab->nlh)
+ ab->nlh = (struct nlmsghdr *)skb_put(skb,
+ NLMSG_SPACE(0));
+ }
+ start = skb_put(skb, ab->len);
+ memcpy(start, ab->tmp, ab->len);
+ ab->len = 0;
+}
+
+/* Iterate over the skbuff in the audit_buffer, sending their contents
+ * to user space. */
+static inline int audit_log_drain(struct audit_buffer *ab)
+{
+ struct sk_buff *skb;
+
+ while ((skb = skb_dequeue(&ab->sklist))) {
+ int retval = 0;
+
+ if (audit_pid) {
+ if (ab->nlh) {
+ ab->nlh->nlmsg_len = ab->total;
+ ab->nlh->nlmsg_type = ab->type;
+ ab->nlh->nlmsg_flags = 0;
+ ab->nlh->nlmsg_seq = 0;
+ ab->nlh->nlmsg_pid = ab->pid;
+ }
+ skb_get(skb); /* because netlink_* frees */
+ retval = netlink_unicast(audit_sock, skb, audit_pid,
+ MSG_DONTWAIT);
+ }
+ if (retval == -EAGAIN && ab->count < 5) {
+ ++ab->count;
+ audit_log_end_irq(ab);
+ return 1;
+ }
+ if (retval < 0) {
+ if (retval == -ECONNREFUSED) {
+ printk(KERN_ERR
+ "audit: *NO* daemon at audit_pid=%d\n",
+ audit_pid);
+ audit_pid = 0;
+ } else
+ audit_log_lost("netlink socket too busy");
+ }
+ if (!audit_pid) { /* No daemon */
+ int offset = ab->nlh ? NLMSG_SPACE(0) : 0;
+ int len = skb->len - offset;
+ printk(KERN_ERR "%*.*s\n",
+ len, len, skb->data + offset);
+ }
+ kfree_skb(skb);
+ ab->nlh = NULL;
+ }
+ return 0;
+}
+
+/* Initialize audit support at boot time. */
+int __init audit_init(void)
+{
+ printk(KERN_INFO "audit: initializing netlink socket (%s)\n",
+ audit_default ? "enabled" : "disabled");
+ audit_sock = netlink_kernel_create(NETLINK_AUDIT, audit_receive);
+ if (!audit_sock)
+ audit_panic("cannot initialize netlink socket");
+
+ audit_initialized = 1;
+ audit_enabled = audit_default;
+ audit_log(NULL, "initialized");
+ return 0;
+}
+
+#else
+/* Without CONFIG_NET, we have no skbuffs. For now, print what we have
+ * in the buffer. */
+static void audit_log_move(struct audit_buffer *ab)
+{
+ printk(KERN_ERR "%*.*s\n", ab->len, ab->len, ab->tmp);
+ ab->len = 0;
+}
+
+static inline int audit_log_drain(struct audit_buffer *ab)
+{
+ return 0;
+}
+
+/* Initialize audit support at boot time. */
+int __init audit_init(void)
+{
+ printk(KERN_INFO "audit: initializing WITHOUT netlink support\n");
+ audit_sock = NULL;
+ audit_pid = 0;
+
+ audit_initialized = 1;
+ audit_enabled = audit_default;
+ audit_log(NULL, "initialized");
+ return 0;
+}
+#endif
+
+__initcall(audit_init);
+
+/* Process kernel command-line parameter at boot time. audit=0 or audit=1. */
+static int __init audit_enable(char *str)
+{
+ audit_default = !!simple_strtol(str, NULL, 0);
+ printk(KERN_INFO "audit: %s%s\n",
+ audit_default ? "enabled" : "disabled",
+ audit_initialized ? "" : " (after initialization)");
+ if (audit_initialized)
+ audit_enabled = audit_default;
+ return 0;
+}
+
+__setup("audit=", audit_enable);
+
+
+/* Obtain an audit buffer. This routine does locking to obtain the
+ * audit buffer, but then no locking is required for calls to
+ * audit_log_*format. If the tsk is a task that is currently in a
+ * syscall, then the syscall is marked as auditable and an audit record
+ * will be written at syscall exit. If there is no associated task, tsk
+ * should be NULL. */
+struct audit_buffer *audit_log_start(struct audit_context *ctx)
+{
+ struct audit_buffer *ab = NULL;
+ unsigned long flags;
+ struct timespec t;
+ int serial = 0;
+
+ if (!audit_initialized)
+ return NULL;
+
+ if (audit_backlog_limit
+ && atomic_read(&audit_backlog) > audit_backlog_limit) {
+ if (audit_rate_check())
+ printk(KERN_WARNING
+ "audit: audit_backlog=%d > "
+ "audit_backlog_limit=%d\n",
+ atomic_read(&audit_backlog),
+ audit_backlog_limit);
+ audit_log_lost("backlog limit exceeded");
+ return NULL;
+ }
+
+ spin_lock_irqsave(&audit_freelist_lock, flags);
+ if (!list_empty(&audit_freelist)) {
+ ab = list_entry(audit_freelist.next,
+ struct audit_buffer, list);
+ list_del(&ab->list);
+ --audit_freelist_count;
+ }
+ spin_unlock_irqrestore(&audit_freelist_lock, flags);
+
+ if (!ab)
+ ab = kmalloc(sizeof(*ab), GFP_ATOMIC);
+ if (!ab)
+ audit_log_lost("audit: out of memory in audit_log_start");
+ if (!ab)
+ return NULL;
+
+ atomic_inc(&audit_backlog);
+ skb_queue_head_init(&ab->sklist);
+
+ ab->ctx = ctx;
+ ab->len = 0;
+ ab->nlh = NULL;
+ ab->total = 0;
+ ab->type = AUDIT_KERNEL;
+ ab->pid = 0;
+ ab->count = 0;
+
+#ifdef CONFIG_AUDITSYSCALL
+ if (ab->ctx)
+ audit_get_stamp(ab->ctx, &t, &serial);
+ else
+#endif
+ t = CURRENT_TIME;
+
+ audit_log_format(ab, "audit(%lu.%03lu:%u): ",
+ t.tv_sec, t.tv_nsec/1000000, serial);
+ return ab;
+}
+
+
+/* Format an audit message into the audit buffer. If there isn't enough
+ * room in the audit buffer, more room will be allocated and vsnprint
+ * will be called a second time. Currently, we assume that a printk
+ * can't format message larger than 1024 bytes, so we don't either. */
+static void audit_log_vformat(struct audit_buffer *ab, const char *fmt,
+ va_list args)
+{
+ int len, avail;
+
+ if (!ab)
+ return;
+
+ avail = sizeof(ab->tmp) - ab->len;
+ if (avail <= 0) {
+ audit_log_move(ab);
+ avail = sizeof(ab->tmp) - ab->len;
+ }
+ len = vsnprintf(ab->tmp + ab->len, avail, fmt, args);
+ if (len >= avail) {
+ /* The printk buffer is 1024 bytes long, so if we get
+ * here and AUDIT_BUFSIZ is at least 1024, then we can
+ * log everything that printk could have logged. */
+ audit_log_move(ab);
+ avail = sizeof(ab->tmp) - ab->len;
+ len = vsnprintf(ab->tmp + ab->len, avail, fmt, args);
+ }
+ ab->len += (len < avail) ? len : avail;
+ ab->total += (len < avail) ? len : avail;
+}
+
+/* Format a message into the audit buffer. All the work is done in
+ * audit_log_vformat. */
+void audit_log_format(struct audit_buffer *ab, const char *fmt, ...)
+{
+ va_list args;
+
+ if (!ab)
+ return;
+ va_start(args, fmt);
+ audit_log_vformat(ab, fmt, args);
+ va_end(args);
+}
+
+/* This is a helper-function to print the d_path without using a static
+ * buffer or allocating another buffer in addition to the one in
+ * audit_buffer. */
+void audit_log_d_path(struct audit_buffer *ab, const char *prefix,
+ struct dentry *dentry, struct vfsmount *vfsmnt)
+{
+ char *p;
+ int len, avail;
+
+ if (prefix) audit_log_format(ab, " %s", prefix);
+
+ if (ab->len > 128)
+ audit_log_move(ab);
+ avail = sizeof(ab->tmp) - ab->len;
+ p = d_path(dentry, vfsmnt, ab->tmp + ab->len, avail);
+ if (p == ERR_PTR(-ENAMETOOLONG)) {
+ /* FIXME: can we save some information here? */
+ audit_log_format(ab, "<toolong>");
+ } else {
+ /* path isn't at start of buffer */
+ len = (ab->tmp + sizeof(ab->tmp) - 1) - p;
+ memmove(ab->tmp + ab->len, p, len);
+ ab->len += len;
+ ab->total += len;
+ }
+}
+
+/* Remove queued messages from the audit_txlist and send them to userspace. */
+static void audit_tasklet_handler(unsigned long arg)
+{
+ LIST_HEAD(list);
+ struct audit_buffer *ab;
+ unsigned long flags;
+
+ spin_lock_irqsave(&audit_txlist_lock, flags);
+ list_splice_init(&audit_txlist, &list);
+ spin_unlock_irqrestore(&audit_txlist_lock, flags);
+
+ while (!list_empty(&list)) {
+ ab = list_entry(list.next, struct audit_buffer, list);
+ list_del(&ab->list);
+ audit_log_end_fast(ab);
+ }
+}
+
+static DECLARE_TASKLET(audit_tasklet, audit_tasklet_handler, 0);
+
+/* The netlink_* functions cannot be called inside an irq context, so
+ * the audit buffer is places on a queue and a tasklet is scheduled to
+ * remove them from the queue outside the irq context. May be called in
+ * any context. */
+void audit_log_end_irq(struct audit_buffer *ab)
+{
+ unsigned long flags;
+
+ if (!ab)
+ return;
+ spin_lock_irqsave(&audit_txlist_lock, flags);
+ list_add_tail(&ab->list, &audit_txlist);
+ spin_unlock_irqrestore(&audit_txlist_lock, flags);
+
+ tasklet_schedule(&audit_tasklet);
+}
+
+/* Send the message in the audit buffer directly to user space. May not
+ * be called in an irq context. */
+void audit_log_end_fast(struct audit_buffer *ab)
+{
+ unsigned long flags;
+
+ BUG_ON(in_irq());
+ if (!ab)
+ return;
+ if (!audit_rate_check()) {
+ audit_log_lost("rate limit exceeded");
+ } else {
+ audit_log_move(ab);
+ if (audit_log_drain(ab))
+ return;
+ }
+
+ atomic_dec(&audit_backlog);
+ spin_lock_irqsave(&audit_freelist_lock, flags);
+ if (++audit_freelist_count > AUDIT_MAXFREE)
+ kfree(ab);
+ else
+ list_add(&ab->list, &audit_freelist);
+ spin_unlock_irqrestore(&audit_freelist_lock, flags);
+}
+
+/* Send or queue the message in the audit buffer, depending on the
+ * current context. (A convenience function that may be called in any
+ * context.) */
+void audit_log_end(struct audit_buffer *ab)
+{
+ if (in_irq())
+ audit_log_end_irq(ab);
+ else
+ audit_log_end_fast(ab);
+}
+
+/* Log an audit record. This is a convenience function that calls
+ * audit_log_start, audit_log_vformat, and audit_log_end. It may be
+ * called in any context. */
+void audit_log(struct audit_context *ctx, const char *fmt, ...)
+{
+ struct audit_buffer *ab;
+ va_list args;
+
+ ab = audit_log_start(ctx);
+ if (ab) {
+ va_start(args, fmt);
+ audit_log_vformat(ab, fmt, args);
+ va_end(args);
+ audit_log_end(ab);
+ }
+}
+
+EXPORT_SYMBOL_GPL(audit_set_rate_limit);
+EXPORT_SYMBOL_GPL(audit_set_backlog_limit);
+EXPORT_SYMBOL_GPL(audit_set_enabled);
+EXPORT_SYMBOL_GPL(audit_set_failure);
+
+EXPORT_SYMBOL_GPL(audit_log_start);
+EXPORT_SYMBOL_GPL(audit_log_format);
+EXPORT_SYMBOL_GPL(audit_log_end_irq);
+EXPORT_SYMBOL_GPL(audit_log_end_fast);
+EXPORT_SYMBOL_GPL(audit_log_end);
+EXPORT_SYMBOL_GPL(audit_log);
+EXPORT_SYMBOL_GPL(audit_log_d_path);
--- diff/kernel/auditsc.c 1970-01-01 00:00:00.000000000 +0000
+++ source/kernel/auditsc.c 2004-03-16 11:27:48.642792832 +0000
@@ -0,0 +1,869 @@
+/* auditsc.c -- System-call auditing support -*- linux-c -*-
+ * Handles all system-call specific auditing features.
+ *
+ * Copyright 2003-2004 Red Hat Inc., Durham, North Carolina.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Written by Rickard E. (Rik) Faith <faith@redhat.com>
+ *
+ * Many of the ideas implemented here are from Stephen C. Tweedie,
+ * especially the idea of avoiding a copy by using getname.
+ *
+ * The method for actual interception of syscall entry and exit (not in
+ * this file -- see entry.S) is based on a GPL'd patch written by
+ * okir@suse.de and Copyright 2003 SuSE Linux AG.
+ *
+ */
+
+#include <linux/init.h>
+#include <asm/atomic.h>
+#include <asm/types.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+
+#include <linux/audit.h>
+#include <linux/time.h>
+#include <asm/unistd.h>
+
+/* 0 = no checking
+ 1 = put_count checking
+ 2 = verbose put_count checking
+*/
+#define AUDIT_DEBUG 0
+
+/* No syscall auditing will take place unless audit_enabled != 0. */
+extern int audit_enabled;
+
+/* AUDIT_NAMES is the number of slots we reserve in the audit_context
+ * for saving names from getname(). */
+#define AUDIT_NAMES 20
+
+/* AUDIT_NAMES_RESERVED is the number of slots we reserve in the
+ * audit_context from being used for nameless inodes from
+ * path_lookup. */
+#define AUDIT_NAMES_RESERVED 7
+
+/* At task start time, the audit_state is set in the audit_context using
+ a per-task filter. At syscall entry, the audit_state is augmented by
+ the syscall filter. */
+enum audit_state {
+ AUDIT_DISABLED, /* Do not create per-task audit_context.
+ * No syscall-specific audit records can
+ * be generated. */
+ AUDIT_SETUP_CONTEXT, /* Create the per-task audit_context,
+ * but don't necessarily fill it in at
+ * syscall entry time (i.e., filter
+ * instead). */
+ AUDIT_BUILD_CONTEXT, /* Create the per-task audit_context,
+ * and always fill it in at syscall
+ * entry time. This makes a full
+ * syscall record available if some
+ * other part of the kernel decides it
+ * should be recorded. */
+ AUDIT_RECORD_CONTEXT /* Create the per-task audit_context,
+ * always fill it in at syscall entry
+ * time, and always write out the audit
+ * record at syscall exit time. */
+};
+
+/* When fs/namei.c:getname() is called, we store the pointer in name and
+ * we don't let putname() free it (instead we free all of the saved
+ * pointers at syscall exit time).
+ *
+ * Further, in fs/namei.c:path_lookup() we store the inode and device. */
+struct audit_names {
+ const char *name;
+ unsigned long ino;
+ dev_t rdev;
+};
+
+/* The per-task audit context. */
+struct audit_context {
+ int in_syscall; /* 1 if task is in a syscall */
+ enum audit_state state;
+ unsigned int serial; /* serial number for record */
+ struct timespec ctime; /* time of syscall entry */
+ uid_t loginuid; /* login uid (identity) */
+ int major; /* syscall number */
+ int minor; /* function for multiplex syscalls */
+ int auditable; /* 1 if record should be written */
+ int name_count;
+ struct audit_names names[AUDIT_NAMES];
+ struct audit_context *previous; /* For nested syscalls */
+
+ /* Save things to print about task_struct */
+ pid_t pid;
+ uid_t uid, euid, suid, fsuid;
+ gid_t gid, egid, sgid, fsgid;
+
+#if AUDIT_DEBUG
+ int put_count;
+ int ino_count;
+#endif
+};
+
+ /* Public API */
+/* There are three lists of rules -- one to search at task creation
+ * time, one to search at syscall entry time, and another to search at
+ * syscall exit time. */
+static LIST_HEAD(audit_tsklist);
+static LIST_HEAD(audit_entlist);
+static LIST_HEAD(audit_extlist);
+
+struct audit_entry {
+ struct list_head list;
+ struct rcu_head rcu;
+ struct audit_rule rule;
+};
+
+/* Check to see if two rules are identical. It is called from
+ * audit_del_rule during AUDIT_DEL. */
+static int audit_compare_rule(struct audit_rule *a, struct audit_rule *b)
+{
+ int i;
+
+ if (a->flags != b->flags)
+ return 1;
+
+ if (a->action != b->action)
+ return 1;
+
+ if (a->field_count != b->field_count)
+ return 1;
+
+ for (i = 0; i < a->field_count; i++) {
+ if (a->fields[i] != b->fields[i]
+ || a->values[i] != b->values[i])
+ return 1;
+ }
+
+ for (i = 0; i < AUDIT_BITMASK_SIZE; i++)
+ if (a->mask[i] != b->mask[i])
+ return 1;
+
+ return 0;
+}
+
+static inline int audit_add_rule(struct audit_entry *entry,
+ struct list_head *list)
+{
+ if (entry->rule.flags & AUDIT_PREPEND) {
+ entry->rule.flags &= ~AUDIT_PREPEND;
+ list_add_rcu(&entry->list, list);
+ } else {
+ list_add_tail_rcu(&entry->list, list);
+ }
+ return 0;
+}
+
+static void audit_free_rule(void *arg)
+{
+ kfree(arg);
+}
+
+static inline int audit_del_rule(struct audit_rule *rule,
+ struct list_head *list)
+{
+ struct audit_entry *e;
+
+ list_for_each_entry_rcu(e, list, list) {
+ if (!audit_compare_rule(rule, &e->rule)) {
+ list_del_rcu(&e->list);
+ call_rcu(&e->rcu, audit_free_rule, e);
+ return 0;
+ }
+ }
+ return -EFAULT; /* No matching rule */
+}
+
+#ifdef CONFIG_NET
+/* Copy rule from user-space to kernel-space. Called during
+ * AUDIT_ADD. */
+static int audit_copy_rule(struct audit_rule *d, struct audit_rule *s)
+{
+ int i;
+
+ if (s->action != AUDIT_NEVER
+ && s->action != AUDIT_POSSIBLE
+ && s->action != AUDIT_ALWAYS)
+ return -1;
+ if (s->field_count < 0 || s->field_count > AUDIT_MAX_FIELDS)
+ return -1;
+
+ d->flags = s->flags;
+ d->action = s->action;
+ d->field_count = s->field_count;
+ for (i = 0; i < d->field_count; i++) {
+ d->fields[i] = s->fields[i];
+ d->values[i] = s->values[i];
+ }
+ for (i = 0; i < AUDIT_BITMASK_SIZE; i++) d->mask[i] = s->mask[i];
+ return 0;
+}
+
+int audit_receive_filter(int type, int pid, int uid, int seq, void *data)
+{
+ u32 flags;
+ struct audit_entry *entry;
+ int err = 0;
+
+ switch (type) {
+ case AUDIT_LIST:
+ list_for_each_entry_rcu(entry, &audit_tsklist, list)
+ audit_send_reply(pid, seq, AUDIT_LIST, 0, 1,
+ &entry->rule, sizeof(entry->rule));
+ list_for_each_entry_rcu(entry, &audit_entlist, list)
+ audit_send_reply(pid, seq, AUDIT_LIST, 0, 1,
+ &entry->rule, sizeof(entry->rule));
+ list_for_each_entry_rcu(entry, &audit_extlist, list)
+ audit_send_reply(pid, seq, AUDIT_LIST, 0, 1,
+ &entry->rule, sizeof(entry->rule));
+ audit_send_reply(pid, seq, AUDIT_LIST, 1, 1, NULL, 0);
+ break;
+ case AUDIT_ADD:
+ if (!capable(CAP_SYS_ADMIN))
+ return -EPERM;
+ if (!(entry = kmalloc(sizeof(*entry), GFP_KERNEL)))
+ return -ENOMEM;
+ if (audit_copy_rule(&entry->rule, data)) {
+ kfree(entry);
+ return -EINVAL;
+ }
+ flags = entry->rule.flags;
+ if (!err && (flags & AUDIT_PER_TASK))
+ err = audit_add_rule(entry, &audit_tsklist);
+ if (!err && (flags & AUDIT_AT_ENTRY))
+ err = audit_add_rule(entry, &audit_entlist);
+ if (!err && (flags & AUDIT_AT_EXIT))
+ err = audit_add_rule(entry, &audit_extlist);
+ break;
+ case AUDIT_DEL:
+ flags =((struct audit_rule *)data)->flags;
+ if (!err && (flags & AUDIT_PER_TASK))
+ err = audit_del_rule(data, &audit_tsklist);
+ if (!err && (flags & AUDIT_AT_ENTRY))
+ err = audit_del_rule(data, &audit_entlist);
+ if (!err && (flags & AUDIT_AT_EXIT))
+ err = audit_del_rule(data, &audit_extlist);
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return err;
+}
+#endif
+
+/* Compare a task_struct with an audit_rule. Return 1 on match, 0
+ * otherwise. */
+static int audit_filter_rules(struct task_struct *tsk,
+ struct audit_rule *rule,
+ struct audit_context *ctx,
+ enum audit_state *state)
+{
+ int i, j;
+ int found;
+
+ for (i = 0; i < rule->field_count; i++) {
+ u32 value = rule->values[i];
+
+ switch (rule->fields[i]) {
+ case AUDIT_PID:
+ if (tsk->pid != value)
+ return 0;
+ break;
+ case AUDIT_UID:
+ if (tsk->uid != value)
+ return 0;
+ break;
+ case AUDIT_EUID:
+ if (tsk->euid != value)
+ return 0;
+ break;
+ case AUDIT_SUID:
+ if (tsk->suid != value)
+ return 0;
+ break;
+ case AUDIT_FSUID:
+ if (tsk->fsuid != value)
+ return 0;
+ break;
+ case AUDIT_GID:
+ if (tsk->gid != value)
+ return 0;
+ break;
+ case AUDIT_EGID:
+ if (tsk->egid != value)
+ return 0;
+ break;
+ case AUDIT_SGID:
+ if (tsk->sgid != value)
+ return 0;
+ break;
+ case AUDIT_FSGID:
+ if (tsk->fsgid != value)
+ return 0;
+ break;
+ case AUDIT_EXIT:
+ if (tsk->exit_code!=value)
+ return 0;
+ break;
+ case AUDIT_NOTEXIT:
+ if (tsk->exit_code==value)
+ return 0;
+ break;
+ case AUDIT_DEVMAJOR:
+ if (!ctx)
+ return 0;
+ found = 0;
+ for (j = 0; !found && j < ctx->name_count; j++) {
+ if (MAJOR(ctx->names[j].rdev) == value)
+ ++found;
+ }
+ if (!found)
+ return 0;
+ break;
+ case AUDIT_DEVMINOR:
+ if (!ctx)
+ return 0;
+ found = 0;
+ for (j = 0; !found && j < ctx->name_count; j++) {
+ if (MINOR(ctx->names[j].rdev) == value)
+ ++found;
+ }
+ if (!found)
+ return 0;
+ break;
+ case AUDIT_INODE:
+ if (!ctx)
+ return 0;
+ found = 0;
+ for (j = 0; !found && j < ctx->name_count; j++) {
+ if (ctx->names[j].ino == value)
+ ++found;
+ }
+ if (!found)
+ return 0;
+ break;
+ case AUDIT_LOGINUID:
+ if (!ctx || ctx->loginuid != value)
+ return 0;
+ break;
+ }
+ }
+ switch (rule->action) {
+ case AUDIT_NEVER: *state = AUDIT_DISABLED; break;
+ case AUDIT_POSSIBLE: *state = AUDIT_BUILD_CONTEXT; break;
+ case AUDIT_ALWAYS: *state = AUDIT_RECORD_CONTEXT; break;
+ }
+ return 1;
+}
+
+/* At process creation time, we can determine if system-call auditing is
+ * completely disabled for this task. Since we only have the task
+ * structure at this point, we can only check uid and gid.
+ */
+static enum audit_state audit_filter_task(struct task_struct *tsk)
+{
+ struct audit_entry *e;
+ enum audit_state state;
+
+ list_for_each_entry_rcu(e, &audit_tsklist, list) {
+ if (audit_filter_rules(tsk, &e->rule, NULL, &state))
+ return state;
+ }
+ return AUDIT_BUILD_CONTEXT;
+}
+
+/* At syscall entry and exit time, this filter is called if the
+ * audit_state is not low enough that auditing cannot take place, but is
+ * also not high enough that we already know we have to write and audit
+ * record (i.e., the state is AUDIT_SETUP_CONTEXT or AUDIT_BUILD_CONTEXT).
+ */
+static enum audit_state audit_filter_syscall(struct task_struct *tsk,
+ struct audit_context *ctx,
+ struct list_head *list)
+{
+ struct audit_entry *e;
+ enum audit_state state;
+ int word = AUDIT_WORD(ctx->major);
+ int bit = AUDIT_BIT(ctx->major);
+
+ list_for_each_entry_rcu(e, list, list) {
+ if ((e->rule.mask[word] & bit) == bit
+ && audit_filter_rules(tsk, &e->rule, ctx, &state))
+ return state;
+ }
+ return AUDIT_BUILD_CONTEXT;
+}
+
+/* This should be called with task_lock() held. */
+static inline struct audit_context *audit_get_context(struct task_struct *tsk)
+{
+ struct audit_context *context = tsk->audit_context;
+
+ if (likely(!context))
+ return NULL;
+ if (context->in_syscall && !context->auditable) {
+ enum audit_state state;
+ state = audit_filter_syscall(tsk, context, &audit_extlist);
+ if (state == AUDIT_RECORD_CONTEXT)
+ context->auditable = 1;
+ }
+
+ context->pid = tsk->pid;
+ context->uid = tsk->uid;
+ context->gid = tsk->gid;
+ context->euid = tsk->euid;
+ context->suid = tsk->suid;
+ context->fsuid = tsk->fsuid;
+ context->egid = tsk->egid;
+ context->sgid = tsk->sgid;
+ context->fsgid = tsk->fsgid;
+ tsk->audit_context = NULL;
+ return context;
+}
+
+static inline void audit_free_names(struct audit_context *context)
+{
+ int i;
+
+#if AUDIT_DEBUG == 2
+ if (context->auditable
+ ||context->put_count + context->ino_count != context->name_count) {
+ printk(KERN_ERR "audit.c:%d(:%d): major=%d in_syscall=%d"
+ " name_count=%d put_count=%d"
+ " ino_count=%d [NOT freeing]\n",
+ __LINE__,
+ context->serial, context->major, context->in_syscall,
+ context->name_count, context->put_count,
+ context->ino_count);
+ for (i = 0; i < context->name_count; i++)
+ printk(KERN_ERR "names[%d] = %p = %s\n", i,
+ context->names[i].name,
+ context->names[i].name);
+ dump_stack();
+ return;
+ }
+#endif
+#if AUDIT_DEBUG
+ context->put_count = 0;
+ context->ino_count = 0;
+#endif
+
+ for (i = 0; i < context->name_count; i++)
+ if (context->names[i].name)
+ __putname(context->names[i].name);
+ context->name_count = 0;
+}
+
+static inline void audit_zero_context(struct audit_context *context,
+ enum audit_state state)
+{
+ uid_t loginuid = context->loginuid;
+
+ memset(context, 0, sizeof(*context));
+ context->state = state;
+ context->loginuid = loginuid;
+}
+
+static inline struct audit_context *audit_alloc_context(enum audit_state state)
+{
+ struct audit_context *context;
+
+ if (!(context = kmalloc(sizeof(*context), GFP_KERNEL)))
+ return NULL;
+ audit_zero_context(context, state);
+ return context;
+}
+
+/* Filter on the task information and allocate a per-task audit context
+ * if necessary. Doing so turns on system call auditing for the
+ * specified task. This is called from copy_process, so no lock is
+ * needed. */
+int audit_alloc(struct task_struct *tsk)
+{
+ struct audit_context *context;
+ enum audit_state state;
+
+ if (likely(!audit_enabled))
+ return 0; /* Return if not auditing. */
+
+ state = audit_filter_task(tsk);
+ if (likely(state == AUDIT_DISABLED))
+ return 0;
+
+ if (!(context = audit_alloc_context(state))) {
+ audit_log_lost("out of memory in audit_alloc");
+ return -ENOMEM;
+ }
+
+ /* Preserve login uid */
+ context->loginuid = -1;
+ if (tsk->audit_context)
+ context->loginuid = tsk->audit_context->loginuid;
+
+ tsk->audit_context = context;
+ set_tsk_thread_flag(tsk, TIF_SYSCALL_AUDIT);
+ return 0;
+}
+
+static inline void audit_free_context(struct audit_context *context)
+{
+ struct audit_context *previous;
+ int count = 0;
+
+ do {
+ previous = context->previous;
+ if (previous || (count && count < 10)) {
+ ++count;
+ printk(KERN_ERR "audit(:%d): major=%d name_count=%d:"
+ " freeing multiple contexts (%d)\n",
+ context->serial, context->major,
+ context->name_count, count);
+ }
+ audit_free_names(context);
+ kfree(context);
+ context = previous;
+ } while (context);
+ if (count >= 10)
+ printk(KERN_ERR "audit: freed %d contexts\n", count);
+}
+
+static void audit_log_exit(struct audit_context *context)
+{
+ int i;
+
+ audit_log(context,
+ "syscall=%d,0x%x items=%d"
+ " pid=%d loginuid=%d uid=%d gid=%d"
+ " euid=%d suid=%d fsuid=%d"
+ " egid=%d sgid=%d fsgid=%d",
+ context->major,
+ context->minor,
+ context->name_count,
+ context->pid,
+ context->loginuid,
+ context->uid,
+ context->gid,
+ context->euid, context->suid, context->fsuid,
+ context->egid, context->sgid, context->fsgid);
+ for (i = 0; i < context->name_count; i++) {
+ struct audit_buffer *ab = audit_log_start(context);
+ if (!ab)
+ continue; /* audit_panic has been called */
+ audit_log_format(ab, "item=%d", i);
+ if (context->names[i].name)
+ audit_log_format(ab, " name=%s",
+ context->names[i].name);
+ if (context->names[i].ino != (unsigned long)-1)
+ audit_log_format(ab, " inode=%lu",
+ context->names[i].ino);
+ /* FIXME: should use format_dev_t, but ab structure is
+ * opaque. */
+ if (context->names[i].rdev != -1)
+ audit_log_format(ab, " dev=%02x:%02x",
+ MAJOR(context->names[i].rdev),
+ MINOR(context->names[i].rdev));
+ audit_log_end(ab);
+ }
+}
+
+/* Free a per-task audit context. Called from copy_process and
+ * __put_task_struct. */
+void audit_free(struct task_struct *tsk)
+{
+ struct audit_context *context;
+
+ task_lock(tsk);
+ context = audit_get_context(tsk);
+ task_unlock(tsk);
+
+ if (likely(!context))
+ return;
+
+ /* Check for system calls that do not go through the exit
+ * function (e.g., exit_group), then free context block. */
+ if (context->in_syscall && context->auditable)
+ audit_log_exit(context);
+
+ audit_free_context(context);
+}
+
+/* Compute a serial number for the audit record. Audit records are
+ * written to user-space as soon as they are generated, so a complete
+ * audit record may be written in several pieces. The timestamp of the
+ * record and this serial number are used by the user-space daemon to
+ * determine which pieces belong to the same audit record. The
+ * (timestamp,serial) tuple is unique for each syscall and is live from
+ * syscall entry to syscall exit.
+ *
+ * Atomic values are only guaranteed to be 24-bit, so we count down.
+ *
+ * NOTE: Another possibility is to store the formatted records off the
+ * audit context (for those records that have a context), and emit them
+ * all at syscall exit. However, this could delay the reporting of
+ * significant errors until syscall exit (or never, if the system
+ * halts). */
+static inline unsigned int audit_serial(void)
+{
+ static atomic_t serial = ATOMIC_INIT(0xffffff);
+ unsigned int a, b;
+
+ do {
+ a = atomic_read(&serial);
+ if (atomic_dec_and_test(&serial))
+ atomic_set(&serial, 0xffffff);
+ b = atomic_read(&serial);
+ } while (b != a - 1);
+
+ return 0xffffff - b;
+}
+
+/* Fill in audit context at syscall entry. This only happens if the
+ * audit context was created when the task was created and the state or
+ * filters demand the audit context be built. If the state from the
+ * per-task filter or from the per-syscall filter is AUDIT_RECORD_CONTEXT,
+ * then the record will be written at syscall exit time (otherwise, it
+ * will only be written if another part of the kernel requests that it
+ * be written). */
+void audit_syscall_entry(struct task_struct *tsk, int major, int minor)
+{
+ struct audit_context *context = tsk->audit_context;
+ enum audit_state state;
+
+ BUG_ON(!context);
+
+ /* This happens only on certain architectures that make system
+ * calls in kernel_thread via the entry.S interface, instead of
+ * with direct calls. (If you are porting to a new
+ * architecture, hitting this condition can indicate that you
+ * got the _exit/_leave calls backward in entry.S.)
+ *
+ * i386 no
+ * x86_64 no
+ * ppc64 yes (see arch/ppc64/kernel/misc.S)
+ *
+ * This also happens with vm86 emulation in a non-nested manner
+ * (entries without exits), so this case must be caught.
+ */
+ if (context->in_syscall) {
+ struct audit_context *newctx;
+
+#if defined(__NR_vm86) && defined(__NR_vm86old)
+ /* vm86 mode should only be entered once */
+ if (major == __NR_vm86 || major == __NR_vm86old)
+ return;
+#endif
+#if AUDIT_DEBUG
+ printk(KERN_ERR
+ "audit(:%d) pid=%d in syscall=%d;"
+ " entering syscall=%d\n",
+ context->serial, tsk->pid, context->major, major);
+#endif
+ newctx = audit_alloc_context(context->state);
+ if (newctx) {
+ newctx->previous = context;
+ context = newctx;
+ tsk->audit_context = newctx;
+ } else {
+ /* If we can't alloc a new context, the best we
+ * can do is to leak memory (any pending putname
+ * will be lost). The only other alternative is
+ * to abandon auditing. */
+ audit_zero_context(context, context->state);
+ }
+ }
+ BUG_ON(context->in_syscall || context->name_count);
+
+ if (!audit_enabled)
+ return;
+
+ context->major = major;
+ context->minor = minor; /* Only valid for some calls */
+
+ state = context->state;
+ if (state == AUDIT_SETUP_CONTEXT || state == AUDIT_BUILD_CONTEXT)
+ state = audit_filter_syscall(tsk, context, &audit_entlist);
+ if (likely(state == AUDIT_DISABLED))
+ return;
+
+ context->serial = audit_serial();
+ context->ctime = CURRENT_TIME;
+ context->in_syscall = 1;
+ context->auditable = !!(state == AUDIT_RECORD_CONTEXT);
+}
+
+/* Tear down after system call. If the audit context has been marked as
+ * auditable (either because of the AUDIT_RECORD_CONTEXT state from
+ * filtering, or because some other part of the kernel write an audit
+ * message), then write out the syscall information. In call cases,
+ * free the names stored from getname(). */
+void audit_syscall_exit(struct task_struct *tsk)
+{
+ struct audit_context *context;
+
+ get_task_struct(tsk);
+ task_lock(tsk);
+ context = audit_get_context(tsk);
+ task_unlock(tsk);
+
+ /* Not having a context here is ok, since the parent may have
+ * called __put_task_struct. */
+ if (likely(!context))
+ return;
+
+ if (context->in_syscall && context->auditable)
+ audit_log_exit(context);
+
+ context->in_syscall = 0;
+ context->auditable = 0;
+ if (context->previous) {
+ struct audit_context *new_context = context->previous;
+ context->previous = NULL;
+ audit_free_context(context);
+ tsk->audit_context = new_context;
+ } else {
+ audit_free_names(context);
+ audit_zero_context(context, context->state);
+ tsk->audit_context = context;
+ }
+ put_task_struct(tsk);
+}
+
+/* Add a name to the list. Called from fs/namei.c:getname(). */
+void audit_getname(const char *name)
+{
+ struct audit_context *context = current->audit_context;
+
+ BUG_ON(!context);
+ if (!context->in_syscall) {
+#if AUDIT_DEBUG == 2
+ printk(KERN_ERR "audit.c:%d(:%d): ignoring getname(%p)\n",
+ __LINE__, context->serial, name);
+ dump_stack();
+#endif
+ return;
+ }
+ BUG_ON(context->name_count >= AUDIT_NAMES);
+ context->names[context->name_count].name = name;
+ context->names[context->name_count].ino = (unsigned long)-1;
+ context->names[context->name_count].rdev = -1;
+ ++context->name_count;
+}
+
+/* Intercept a putname request. Called from
+ * include/linux/fs.h:putname(). If we have stored the name from
+ * getname in the audit context, then we delay the putname until syscall
+ * exit. */
+void audit_putname(const char *name)
+{
+ struct audit_context *context = current->audit_context;
+
+ BUG_ON(!context);
+ if (!context->in_syscall) {
+#if AUDIT_DEBUG == 2
+ printk(KERN_ERR "audit.c:%d(:%d): __putname(%p)\n",
+ __LINE__, context->serial, name);
+ if (context->name_count) {
+ int i;
+ for (i = 0; i < context->name_count; i++)
+ printk(KERN_ERR "name[%d] = %p = %s\n", i,
+ context->names[i].name,
+ context->names[i].name);
+ }
+#endif
+ __putname(name);
+ }
+#if AUDIT_DEBUG
+ else {
+ ++context->put_count;
+ if (context->put_count > context->name_count) {
+ printk(KERN_ERR "audit.c:%d(:%d): major=%d"
+ " in_syscall=%d putname(%p) name_count=%d"
+ " put_count=%d\n",
+ __LINE__,
+ context->serial, context->major,
+ context->in_syscall, name, context->name_count,
+ context->put_count);
+ dump_stack();
+ }
+ }
+#endif
+}
+
+/* Store the inode and device from a lookup. Called from
+ * fs/namei.c:path_lookup(). */
+void audit_inode(const char *name, unsigned long ino, dev_t rdev)
+{
+ int idx;
+ struct audit_context *context = current->audit_context;
+
+ if (context->name_count
+ && context->names[context->name_count-1].name
+ && context->names[context->name_count-1].name == name)
+ idx = context->name_count - 1;
+ else if (context->name_count > 1
+ && context->names[context->name_count-2].name
+ && context->names[context->name_count-2].name == name)
+ idx = context->name_count - 2;
+ else {
+ /* FIXME: how much do we care about inodes that have no
+ * associated name? */
+ if (context->name_count >= AUDIT_NAMES - AUDIT_NAMES_RESERVED)
+ return;
+ idx = context->name_count++;
+ context->names[idx].name = NULL;
+#if AUDIT_DEBUG
+ ++context->ino_count;
+#endif
+ }
+ context->names[idx].ino = ino;
+ context->names[idx].rdev = rdev;
+}
+
+void audit_get_stamp(struct audit_context *ctx,
+ struct timespec *t, int *serial)
+{
+ if (ctx) {
+ t->tv_sec = ctx->ctime.tv_sec;
+ t->tv_nsec = ctx->ctime.tv_nsec;
+ *serial = ctx->serial;
+ ctx->auditable = 1;
+ } else {
+ *t = CURRENT_TIME;
+ *serial = 0;
+ }
+}
+
+int audit_set_loginuid(struct audit_context *ctx, uid_t loginuid)
+{
+ if (ctx) {
+ if (loginuid < 0)
+ return -EINVAL;
+ ctx->loginuid = loginuid;
+ }
+ return 0;
+}
+
+EXPORT_SYMBOL_GPL(audit_alloc);
+EXPORT_SYMBOL_GPL(audit_free);
+EXPORT_SYMBOL_GPL(audit_syscall_entry);
+EXPORT_SYMBOL_GPL(audit_syscall_exit);
+EXPORT_SYMBOL_GPL(audit_getname);
+EXPORT_SYMBOL_GPL(audit_putname);
+EXPORT_SYMBOL_GPL(audit_inode);
--- diff/kernel/compat_signal.c 1970-01-01 00:00:00.000000000 +0000
+++ source/kernel/compat_signal.c 2004-03-16 11:27:48.643792680 +0000
@@ -0,0 +1,124 @@
+/*
+ * Copyright (C) 2003 Carlos O'Donell
+ *
+ * 2003-12-20 Carlos O'Donell
+ * Copied linux/kernel/compat_signal.c (copy_siginfo_to_user)
+ * and modified to use compat_siginfo_t for thunking down to
+ * 32-bit userspace from a 64-bit kernel.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or (at
+ * your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or
+ * NON INFRINGEMENT. See the GNU General Public License for more
+ * details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#include <linux/compat_siginfo.h>
+#include <asm/errno.h>
+#include <asm/uaccess.h>
+#include <asm/siginfo.h>
+
+#ifndef HAVE_ARCH_COMPAT_COPY_SIGINFO_TO_USER
+
+int compat_copy_siginfo_to_user(compat_siginfo_t __user *to, siginfo_t *from)
+{
+ int err;
+ compat_siginfo_t compat_from;
+
+ if (!access_ok (VERIFY_WRITE, to, sizeof(compat_siginfo_t)))
+ return -EFAULT;
+
+ /*
+ * If you change compat_siginfo_t structure *or* siginfo_t,
+ * please be sure this code is fixed accordingly.
+ * It should never copy any pad contained in the structure
+ * to avoid security leaks, but must copy the generic
+ * 3 ints plus the relevant union member.
+ */
+
+ /* Convert structure, don't leak anything in the copy */
+ memset(&compat_from,'\0',sizeof(compat_siginfo_t));
+ compat_from.si_signo = (compat_int_t)(from->si_signo);
+ compat_from.si_errno = (compat_int_t)(from->si_errno);
+ compat_from.si_code = (compat_int_t)(from->si_code);
+
+ if (from->si_code < 0)
+ return __copy_to_user(to, &compat_from, sizeof(compat_siginfo_t))
+ ? -EFAULT : 0;
+
+ err = __put_user(compat_from.si_signo, &to->si_signo);
+ err |= __put_user(compat_from.si_errno, &to->si_errno);
+ err |= __put_user(compat_from.si_code, &to->si_code);
+
+ switch (from->si_code & __SI_MASK) {
+ case __SI_KILL:
+ compat_from.si_pid = (compat_pid_t)(from->si_pid);
+ compat_from.si_uid = (__ARCH_SI_COMPAT_UID_T)(from->si_uid);
+ err |= __put_user(compat_from.si_pid, &to->si_pid);
+ err |= __put_user(compat_from.si_uid, &to->si_uid);
+ break;
+ case __SI_TIMER:
+ compat_from.si_pid = (compat_timer_t)(from->si_tid);
+ compat_from.si_overrun = (compat_int_t)(from->si_overrun);
+ compat_from.si_ptr = (compat_uptr_t)((u64)(from->si_ptr) & 0xffffffffUL);
+ err |= __put_user(compat_from.si_tid, &to->si_tid);
+ err |= __put_user(compat_from.si_overrun, &to->si_overrun);
+ err |= __put_user(compat_from.si_ptr, &to->si_ptr);
+ break;
+ case __SI_POLL:
+ compat_from.si_band = (__ARCH_SI_COMPAT_BAND_T)(from->si_band);
+ compat_from.si_fd = (compat_int_t)(from->si_fd);
+ err |= __put_user(compat_from.si_band, &to->si_band);
+ err |= __put_user(compat_from.si_fd, &to->si_fd);
+ break;
+ case __SI_FAULT:
+ compat_from.si_addr = (compat_uptr_t)((u64)(from->si_addr) & 0xffffffffUL);
+ err |= __put_user(compat_from.si_addr, &to->si_addr);
+#ifdef __ARCH_SI_COMPAT_TRAPNO
+ compat_from.si_trapno = (compat_int_t)(from->si_addr);
+ err |= __put_user(compat_from.si_trapno, &to->si_trapno);
+#endif
+ break;
+ case __SI_CHLD:
+ compat_from.si_pid = (compat_pid_t)(from->si_pid);
+ compat_from.si_uid = (__ARCH_SI_COMPAT_UID_T)(from->si_uid);
+ compat_from.si_status = (compat_int_t)(from->si_status);
+ compat_from.si_utime = (compat_clock_t)(from->si_utime);
+ compat_from.si_stime = (compat_clock_t)(from->si_stime);
+ err |= __put_user(compat_from.si_pid, &to->si_pid);
+ err |= __put_user(compat_from.si_uid, &to->si_uid);
+ err |= __put_user(compat_from.si_status, &to->si_status);
+ err |= __put_user(compat_from.si_utime, &to->si_utime);
+ err |= __put_user(compat_from.si_stime, &to->si_stime);
+ break;
+ case __SI_RT: /* This is not generated by the kernel as of now. */
+ compat_from.si_pid = (compat_pid_t)(from->si_pid);
+ compat_from.si_uid = (__ARCH_SI_COMPAT_UID_T)(from->si_uid);
+ compat_from.si_int = (compat_int_t)(from->si_int);
+ compat_from.si_ptr = (compat_uptr_t)((u64)(from->si_ptr) & 0xffffffffUL);
+ err |= __put_user(compat_from.si_pid, &to->si_pid);
+ err |= __put_user(compat_from.si_uid, &to->si_uid);
+ err |= __put_user(compat_from.si_int, &to->si_int);
+ err |= __put_user(compat_from.si_ptr, &to->si_ptr);
+ break;
+ default: /* this is just in case for now ... */
+ compat_from.si_pid = (compat_pid_t)(from->si_pid);
+ compat_from.si_uid = (__ARCH_SI_COMPAT_UID_T)(from->si_uid);
+ err |= __put_user(compat_from.si_pid, &to->si_pid);
+ err |= __put_user(compat_from.si_uid, &to->si_uid);
+ break;
+ }
+ return err;
+}
+
+#endif
--- diff/kernel/lockmeter.c 1970-01-01 00:00:00.000000000 +0000
+++ source/kernel/lockmeter.c 2004-03-16 11:27:48.649791768 +0000
@@ -0,0 +1,1178 @@
+/*
+ * Copyright (C) 1999,2000 Silicon Graphics, Inc.
+ *
+ * Written by John Hawkes (hawkes@sgi.com)
+ * Based on klstat.c by Jack Steiner (steiner@sgi.com)
+ *
+ * Modified by Ray Bryant (raybry@us.ibm.com)
+ * Changes Copyright (C) 2000 IBM, Inc.
+ * Added save of index in spinlock_t to improve efficiency
+ * of "hold" time reporting for spinlocks
+ * Added support for hold time statistics for read and write
+ * locks.
+ */
+
+#include <linux/config.h>
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/sched.h>
+#include <linux/smp.h>
+#include <linux/threads.h>
+#include <linux/version.h>
+#include <linux/vmalloc.h>
+#include <linux/spinlock.h>
+#include <linux/utsname.h>
+#include <linux/module.h>
+#include <asm/system.h>
+#include <asm/uaccess.h>
+
+#include <linux/lockmeter.h>
+
+#define ASSERT(cond)
+#define bzero(loc,size) memset(loc,0,size)
+
+/*<---------------------------------------------------*/
+/* lockmeter.c */
+/*>---------------------------------------------------*/
+
+static lstat_control_t lstat_control __cacheline_aligned =
+ { LSTAT_OFF, SPIN_LOCK_UNLOCKED, SPIN_LOCK_UNLOCKED,
+ 19 * 0, NR_CPUS * 0, 0, NR_CPUS * 0 };
+
+static ushort lstat_make_dir_entry(void *, void *);
+
+/*
+ * lstat_lookup
+ *
+ * Given a RA, locate the directory entry for the lock.
+ */
+static ushort
+lstat_lookup(void *lock_ptr, void *caller_ra)
+{
+ ushort index;
+ lstat_directory_entry_t *dirp;
+
+ dirp = lstat_control.dir;
+
+ index = lstat_control.hashtab[DIRHASH(caller_ra)];
+ while (dirp[index].caller_ra != caller_ra) {
+ if (index == 0) {
+ return lstat_make_dir_entry(lock_ptr, caller_ra);
+ }
+ index = dirp[index].next_stat_index;
+ }
+
+ if (dirp[index].lock_ptr != NULL && dirp[index].lock_ptr != lock_ptr) {
+ dirp[index].lock_ptr = NULL;
+ }
+
+ return index;
+}
+
+/*
+ * lstat_make_dir_entry
+ * Called to add a new lock to the lock directory.
+ */
+static ushort
+lstat_make_dir_entry(void *lock_ptr, void *caller_ra)
+{
+ lstat_directory_entry_t *dirp;
+ ushort index, hindex;
+ unsigned long flags;
+
+ /* lock the table without recursively reentering this metering code */
+ local_irq_save(flags);
+ _raw_spin_lock(&lstat_control.directory_lock);
+
+ hindex = DIRHASH(caller_ra);
+ index = lstat_control.hashtab[hindex];
+ dirp = lstat_control.dir;
+ while (index && dirp[index].caller_ra != caller_ra)
+ index = dirp[index].next_stat_index;
+
+ if (index == 0) {
+ if (lstat_control.next_free_dir_index < LSTAT_MAX_STAT_INDEX) {
+ index = lstat_control.next_free_dir_index++;
+ lstat_control.dir[index].caller_ra = caller_ra;
+ lstat_control.dir[index].lock_ptr = lock_ptr;
+ lstat_control.dir[index].next_stat_index =
+ lstat_control.hashtab[hindex];
+ lstat_control.hashtab[hindex] = index;
+ } else {
+ lstat_control.dir_overflow++;
+ }
+ }
+ _raw_spin_unlock(&lstat_control.directory_lock);
+ local_irq_restore(flags);
+ return index;
+}
+
+int
+lstat_update(void *lock_ptr, void *caller_ra, int action)
+{
+ int index;
+ int cpu;
+
+ ASSERT(action < LSTAT_ACT_MAX_VALUES);
+
+ if (lstat_control.state == LSTAT_OFF)
+ return 0;
+
+ index = lstat_lookup(lock_ptr, caller_ra);
+ cpu = THIS_CPU_NUMBER;
+ (*lstat_control.counts[cpu])[index].count[action]++;
+ (*lstat_control.counts[cpu])[index].acquire_time = get_cycles();
+
+ return index;
+}
+
+int
+lstat_update_time(void *lock_ptr, void *caller_ra, int action, uint32_t ticks)
+{
+ ushort index;
+ int cpu;
+
+ ASSERT(action < LSTAT_ACT_MAX_VALUES);
+
+ if (lstat_control.state == LSTAT_OFF)
+ return 0;
+
+ index = lstat_lookup(lock_ptr, caller_ra);
+ cpu = THIS_CPU_NUMBER;
+ (*lstat_control.counts[cpu])[index].count[action]++;
+ (*lstat_control.counts[cpu])[index].cum_wait_ticks += (uint64_t) ticks;
+ if ((*lstat_control.counts[cpu])[index].max_wait_ticks < ticks)
+ (*lstat_control.counts[cpu])[index].max_wait_ticks = ticks;
+
+ (*lstat_control.counts[cpu])[index].acquire_time = get_cycles();
+
+ return index;
+}
+
+void
+_metered_spin_lock(spinlock_t * lock_ptr)
+{
+ if (lstat_control.state == LSTAT_OFF) {
+ _raw_spin_lock(lock_ptr); /* do the real lock */
+ PUT_INDEX(lock_ptr, 0); /* clean index in case lockmetering */
+ /* gets turned on before unlock */
+ } else {
+ void *this_pc = LSTAT_RA(LSTAT_RA_SPIN);
+ int index;
+
+ if (_raw_spin_trylock(lock_ptr)) {
+ index = lstat_update(lock_ptr, this_pc,
+ LSTAT_ACT_NO_WAIT);
+ } else {
+ uint32_t start_cycles = get_cycles();
+ _raw_spin_lock(lock_ptr); /* do the real lock */
+ index = lstat_update_time(lock_ptr, this_pc,
+ LSTAT_ACT_SPIN, get_cycles() - start_cycles);
+ }
+ /* save the index in the lock itself for use in spin unlock */
+ PUT_INDEX(lock_ptr, index);
+ }
+}
+
+int
+_metered_spin_trylock(spinlock_t * lock_ptr)
+{
+ if (lstat_control.state == LSTAT_OFF) {
+ return _raw_spin_trylock(lock_ptr);
+ } else {
+ int retval;
+ void *this_pc = LSTAT_RA(LSTAT_RA_SPIN);
+
+ if ((retval = _raw_spin_trylock(lock_ptr))) {
+ int index = lstat_update(lock_ptr, this_pc,
+ LSTAT_ACT_NO_WAIT);
+ /*
+ * save the index in the lock itself for use in spin
+ * unlock
+ */
+ PUT_INDEX(lock_ptr, index);
+ } else {
+ lstat_update(lock_ptr, this_pc, LSTAT_ACT_REJECT);
+ }
+
+ return retval;
+ }
+}
+
+void
+_metered_spin_unlock(spinlock_t * lock_ptr)
+{
+ int index = -1;
+
+ if (lstat_control.state != LSTAT_OFF) {
+ index = GET_INDEX(lock_ptr);
+ /*
+ * If statistics were turned off when we set the lock,
+ * then the index can be zero. If that is the case,
+ * then collect no stats on this call.
+ */
+ if (index > 0) {
+ uint32_t hold_time;
+ int cpu = THIS_CPU_NUMBER;
+ hold_time = get_cycles() -
+ (*lstat_control.counts[cpu])[index].acquire_time;
+ (*lstat_control.counts[cpu])[index].cum_hold_ticks +=
+ (uint64_t) hold_time;
+ if ((*lstat_control.counts[cpu])[index].max_hold_ticks <
+ hold_time)
+ (*lstat_control.counts[cpu])[index].
+ max_hold_ticks = hold_time;
+ }
+ }
+
+ /* make sure we don't have a stale index value saved */
+ PUT_INDEX(lock_ptr, 0);
+ _raw_spin_unlock(lock_ptr); /* do the real unlock */
+}
+
+/*
+ * allocate the next global read lock structure and store its index
+ * in the rwlock at "lock_ptr".
+ */
+uint32_t
+alloc_rwlock_struct(rwlock_t * rwlock_ptr)
+{
+ int index;
+ unsigned long flags;
+ int cpu = THIS_CPU_NUMBER;
+
+ /* If we've already overflowed, then do a quick exit */
+ if (lstat_control.next_free_read_lock_index >
+ LSTAT_MAX_READ_LOCK_INDEX) {
+ lstat_control.rwlock_overflow++;
+ return 0;
+ }
+
+ local_irq_save(flags);
+ _raw_spin_lock(&lstat_control.directory_lock);
+
+ /* It is possible this changed while we were waiting for the directory_lock */
+ if (lstat_control.state == LSTAT_OFF) {
+ index = 0;
+ goto unlock;
+ }
+
+ /* It is possible someone else got here first and set the index */
+ if ((index = GET_RWINDEX(rwlock_ptr)) == 0) {
+ /*
+ * we can't turn on read stats for this lock while there are
+ * readers (this would mess up the running hold time sum at
+ * unlock time)
+ */
+ if (RWLOCK_READERS(rwlock_ptr) != 0) {
+ index = 0;
+ goto unlock;
+ }
+
+ /*
+ * if stats are turned on after being off, we may need to
+ * return an old index from when the statistics were on last
+ * time.
+ */
+ for (index = 1; index < lstat_control.next_free_read_lock_index;
+ index++)
+ if ((*lstat_control.read_lock_counts[cpu])[index].
+ lock_ptr == rwlock_ptr)
+ goto put_index_and_unlock;
+
+ /* allocate the next global read lock structure */
+ if (lstat_control.next_free_read_lock_index >=
+ LSTAT_MAX_READ_LOCK_INDEX) {
+ lstat_control.rwlock_overflow++;
+ index = 0;
+ goto unlock;
+ }
+ index = lstat_control.next_free_read_lock_index++;
+
+ /*
+ * initialize the global read stats data structure for each
+ * cpu
+ */
+ for (cpu = 0; cpu < num_online_cpus(); cpu++) {
+ (*lstat_control.read_lock_counts[cpu])[index].lock_ptr =
+ rwlock_ptr;
+ }
+put_index_and_unlock:
+ /* store the index for the read lock structure into the lock */
+ PUT_RWINDEX(rwlock_ptr, index);
+ }
+
+unlock:
+ _raw_spin_unlock(&lstat_control.directory_lock);
+ local_irq_restore(flags);
+ return index;
+}
+
+void
+_metered_read_lock(rwlock_t * rwlock_ptr)
+{
+ void *this_pc;
+ uint32_t start_cycles;
+ int index;
+ int cpu;
+ unsigned long flags;
+ int readers_before, readers_after;
+ uint64_t cycles64;
+
+ if (lstat_control.state == LSTAT_OFF) {
+ _raw_read_lock(rwlock_ptr);
+ /* clean index in case lockmetering turns on before an unlock */
+ PUT_RWINDEX(rwlock_ptr, 0);
+ return;
+ }
+
+ this_pc = LSTAT_RA(LSTAT_RA_READ);
+ cpu = THIS_CPU_NUMBER;
+ index = GET_RWINDEX(rwlock_ptr);
+
+ /* allocate the global stats entry for this lock, if needed */
+ if (index == 0)
+ index = alloc_rwlock_struct(rwlock_ptr);
+
+ readers_before = RWLOCK_READERS(rwlock_ptr);
+ if (_raw_read_trylock(rwlock_ptr)) {
+ /*
+ * We have decremented the lock to count a new reader,
+ * and have confirmed that no writer has it locked.
+ */
+ /* update statistics if enabled */
+ if (index > 0) {
+ local_irq_save(flags);
+ lstat_update((void *) rwlock_ptr, this_pc,
+ LSTAT_ACT_NO_WAIT);
+ /* preserve value of TSC so cum_hold_ticks and start_busy use same value */
+ cycles64 = get_cycles64();
+ (*lstat_control.read_lock_counts[cpu])[index].
+ cum_hold_ticks -= cycles64;
+
+ /* record time and cpu of start of busy period */
+ /* this is not perfect (some race conditions are possible) */
+ if (readers_before == 0) {
+ (*lstat_control.read_lock_counts[cpu])[index].
+ start_busy = cycles64;
+ PUT_RW_CPU(rwlock_ptr, cpu);
+ }
+ readers_after = RWLOCK_READERS(rwlock_ptr);
+ if (readers_after >
+ (*lstat_control.read_lock_counts[cpu])[index].
+ max_readers)
+ (*lstat_control.read_lock_counts[cpu])[index].
+ max_readers = readers_after;
+ local_irq_restore(flags);
+ }
+
+ return;
+ }
+ /* If we get here, then we could not quickly grab the read lock */
+
+ start_cycles = get_cycles(); /* start counting the wait time */
+
+ /* Now spin until read_lock is successful */
+ _raw_read_lock(rwlock_ptr);
+
+ lstat_update_time((void *) rwlock_ptr, this_pc, LSTAT_ACT_SPIN,
+ get_cycles() - start_cycles);
+
+ /* update statistics if they are enabled for this lock */
+ if (index > 0) {
+ local_irq_save(flags);
+ cycles64 = get_cycles64();
+ (*lstat_control.read_lock_counts[cpu])[index].cum_hold_ticks -=
+ cycles64;
+
+ /* this is not perfect (some race conditions are possible) */
+ if (readers_before == 0) {
+ (*lstat_control.read_lock_counts[cpu])[index].
+ start_busy = cycles64;
+ PUT_RW_CPU(rwlock_ptr, cpu);
+ }
+ readers_after = RWLOCK_READERS(rwlock_ptr);
+ if (readers_after >
+ (*lstat_control.read_lock_counts[cpu])[index].max_readers)
+ (*lstat_control.read_lock_counts[cpu])[index].
+ max_readers = readers_after;
+ local_irq_restore(flags);
+ }
+}
+
+void
+_metered_read_unlock(rwlock_t * rwlock_ptr)
+{
+ int index;
+ int cpu;
+ unsigned long flags;
+ uint64_t busy_length;
+ uint64_t cycles64;
+
+ if (lstat_control.state == LSTAT_OFF) {
+ _raw_read_unlock(rwlock_ptr);
+ return;
+ }
+
+ index = GET_RWINDEX(rwlock_ptr);
+ cpu = THIS_CPU_NUMBER;
+
+ if (index > 0) {
+ local_irq_save(flags);
+ /*
+ * preserve value of TSC so cum_hold_ticks and busy_ticks are
+ * consistent.
+ */
+ cycles64 = get_cycles64();
+ (*lstat_control.read_lock_counts[cpu])[index].cum_hold_ticks +=
+ cycles64;
+ (*lstat_control.read_lock_counts[cpu])[index].read_lock_count++;
+
+ /*
+ * once again, this is not perfect (some race conditions are
+ * possible)
+ */
+ if (RWLOCK_READERS(rwlock_ptr) == 1) {
+ int cpu1 = GET_RW_CPU(rwlock_ptr);
+ uint64_t last_start_busy =
+ (*lstat_control.read_lock_counts[cpu1])[index].
+ start_busy;
+ (*lstat_control.read_lock_counts[cpu])[index].
+ busy_periods++;
+ if (cycles64 > last_start_busy) {
+ busy_length = cycles64 - last_start_busy;
+ (*lstat_control.read_lock_counts[cpu])[index].
+ busy_ticks += busy_length;
+ if (busy_length >
+ (*lstat_control.
+ read_lock_counts[cpu])[index].
+ max_busy)
+ (*lstat_control.
+ read_lock_counts[cpu])[index].
+ max_busy = busy_length;
+ }
+ }
+ local_irq_restore(flags);
+ }
+ _raw_read_unlock(rwlock_ptr);
+}
+
+void
+_metered_write_lock(rwlock_t * rwlock_ptr)
+{
+ uint32_t start_cycles;
+ void *this_pc;
+ uint32_t spin_ticks = 0; /* in anticipation of a potential wait */
+ int index;
+ int write_index = 0;
+ int cpu;
+ enum {
+ writer_writer_conflict,
+ writer_reader_conflict
+ } why_wait = writer_writer_conflict;
+
+ if (lstat_control.state == LSTAT_OFF) {
+ _raw_write_lock(rwlock_ptr);
+ /* clean index in case lockmetering turns on before an unlock */
+ PUT_RWINDEX(rwlock_ptr, 0);
+ return;
+ }
+
+ this_pc = LSTAT_RA(LSTAT_RA_WRITE);
+ cpu = THIS_CPU_NUMBER;
+ index = GET_RWINDEX(rwlock_ptr);
+
+ /* allocate the global stats entry for this lock, if needed */
+ if (index == 0) {
+ index = alloc_rwlock_struct(rwlock_ptr);
+ }
+
+ if (_raw_write_trylock(rwlock_ptr)) {
+ /* We acquired the lock on the first try */
+ write_index = lstat_update((void *) rwlock_ptr, this_pc,
+ LSTAT_ACT_NO_WAIT);
+ /* save the write_index for use in unlock if stats enabled */
+ if (index > 0)
+ (*lstat_control.read_lock_counts[cpu])[index].
+ write_index = write_index;
+ return;
+ }
+
+ /* If we get here, then we could not quickly grab the write lock */
+ start_cycles = get_cycles(); /* start counting the wait time */
+
+ why_wait = RWLOCK_READERS(rwlock_ptr) ?
+ writer_reader_conflict : writer_writer_conflict;
+
+ /* Now set the lock and wait for conflicts to disappear */
+ _raw_write_lock(rwlock_ptr);
+
+ spin_ticks = get_cycles() - start_cycles;
+
+ /* update stats -- if enabled */
+ if (index > 0 && spin_ticks) {
+ if (why_wait == writer_reader_conflict) {
+ /* waited due to a reader holding the lock */
+ write_index = lstat_update_time((void *)rwlock_ptr,
+ this_pc, LSTAT_ACT_SPIN, spin_ticks);
+ } else {
+ /*
+ * waited due to another writer holding the lock
+ */
+ write_index = lstat_update_time((void *)rwlock_ptr,
+ this_pc, LSTAT_ACT_WW_SPIN, spin_ticks);
+ (*lstat_control.counts[cpu])[write_index].
+ cum_wait_ww_ticks += spin_ticks;
+ if (spin_ticks >
+ (*lstat_control.counts[cpu])[write_index].
+ max_wait_ww_ticks) {
+ (*lstat_control.counts[cpu])[write_index].
+ max_wait_ww_ticks = spin_ticks;
+ }
+ }
+
+ /* save the directory index for use on write_unlock */
+ (*lstat_control.read_lock_counts[cpu])[index].
+ write_index = write_index;
+ }
+}
+
+void
+_metered_write_unlock(rwlock_t * rwlock_ptr)
+{
+ int index;
+ int cpu;
+ int write_index;
+ uint32_t hold_time;
+
+ if (lstat_control.state == LSTAT_OFF) {
+ _raw_write_unlock(rwlock_ptr);
+ return;
+ }
+
+ cpu = THIS_CPU_NUMBER;
+ index = GET_RWINDEX(rwlock_ptr);
+
+ /* update statistics if stats enabled for this lock */
+ if (index > 0) {
+ write_index =
+ (*lstat_control.read_lock_counts[cpu])[index].write_index;
+
+ hold_time = get_cycles() -
+ (*lstat_control.counts[cpu])[write_index].acquire_time;
+ (*lstat_control.counts[cpu])[write_index].cum_hold_ticks +=
+ (uint64_t) hold_time;
+ if ((*lstat_control.counts[cpu])[write_index].max_hold_ticks <
+ hold_time)
+ (*lstat_control.counts[cpu])[write_index].
+ max_hold_ticks = hold_time;
+ }
+ _raw_write_unlock(rwlock_ptr);
+}
+
+int
+_metered_write_trylock(rwlock_t * rwlock_ptr)
+{
+ int retval;
+ void *this_pc = LSTAT_RA(LSTAT_RA_WRITE);
+
+ if ((retval = _raw_write_trylock(rwlock_ptr))) {
+ lstat_update(rwlock_ptr, this_pc, LSTAT_ACT_NO_WAIT);
+ } else {
+ lstat_update(rwlock_ptr, this_pc, LSTAT_ACT_REJECT);
+ }
+
+ return retval;
+}
+
+static void
+init_control_space(void)
+{
+ /* Set all control space pointers to null and indices to "empty" */
+ int cpu;
+
+ /*
+ * Access CPU_CYCLE_FREQUENCY at the outset, which in some
+ * architectures may trigger a runtime calculation that uses a
+ * spinlock. Let's do this before lockmetering is turned on.
+ */
+ if (CPU_CYCLE_FREQUENCY == 0)
+ BUG();
+
+ lstat_control.hashtab = NULL;
+ lstat_control.dir = NULL;
+ for (cpu = 0; cpu < NR_CPUS; cpu++) {
+ lstat_control.counts[cpu] = NULL;
+ lstat_control.read_lock_counts[cpu] = NULL;
+ }
+}
+
+static int
+reset_lstat_data(void)
+{
+ int cpu, flags;
+
+ flags = 0;
+ lstat_control.next_free_dir_index = 1; /* 0 is for overflows */
+ lstat_control.next_free_read_lock_index = 1;
+ lstat_control.dir_overflow = 0;
+ lstat_control.rwlock_overflow = 0;
+
+ lstat_control.started_cycles64 = 0;
+ lstat_control.ending_cycles64 = 0;
+ lstat_control.enabled_cycles64 = 0;
+ lstat_control.first_started_time = 0;
+ lstat_control.started_time = 0;
+ lstat_control.ending_time = 0;
+ lstat_control.intervals = 0;
+
+ /*
+ * paranoia -- in case someone does a "lockstat reset" before
+ * "lockstat on"
+ */
+ if (lstat_control.hashtab) {
+ bzero(lstat_control.hashtab,
+ LSTAT_HASH_TABLE_SIZE * sizeof (short));
+ bzero(lstat_control.dir, LSTAT_MAX_STAT_INDEX *
+ sizeof (lstat_directory_entry_t));
+
+ for (cpu = 0; cpu < num_online_cpus(); cpu++) {
+ bzero(lstat_control.counts[cpu],
+ sizeof (lstat_cpu_counts_t));
+ bzero(lstat_control.read_lock_counts[cpu],
+ sizeof (lstat_read_lock_cpu_counts_t));
+ }
+ }
+#ifdef NOTDEF
+ _raw_spin_unlock(&lstat_control.directory_lock);
+ local_irq_restore(flags);
+#endif
+ return 1;
+}
+
+static void
+release_control_space(void)
+{
+ /*
+ * Called when either (1) allocation of kmem
+ * or (2) when user writes LSTAT_RELEASE to /pro/lockmeter.
+ * Assume that all pointers have been initialized to zero,
+ * i.e., nonzero pointers are valid addresses.
+ */
+ int cpu;
+
+ if (lstat_control.hashtab) {
+ kfree(lstat_control.hashtab);
+ lstat_control.hashtab = NULL;
+ }
+
+ if (lstat_control.dir) {
+ vfree(lstat_control.dir);
+ lstat_control.dir = NULL;
+ }
+
+ for (cpu = 0; cpu < NR_CPUS; cpu++) {
+ if (lstat_control.counts[cpu]) {
+ vfree(lstat_control.counts[cpu]);
+ lstat_control.counts[cpu] = NULL;
+ }
+ if (lstat_control.read_lock_counts[cpu]) {
+ kfree(lstat_control.read_lock_counts[cpu]);
+ lstat_control.read_lock_counts[cpu] = NULL;
+ }
+ }
+}
+
+int
+get_lockmeter_info_size(void)
+{
+ return sizeof (lstat_user_request_t)
+ + num_online_cpus() * sizeof (lstat_cpu_counts_t)
+ + num_online_cpus() * sizeof (lstat_read_lock_cpu_counts_t)
+ + (LSTAT_MAX_STAT_INDEX * sizeof (lstat_directory_entry_t));
+}
+
+ssize_t
+get_lockmeter_info(char *buffer, size_t max_len, loff_t * last_index)
+{
+ lstat_user_request_t req;
+ struct timeval tv;
+ ssize_t next_ret_bcount;
+ ssize_t actual_ret_bcount = 0;
+ int cpu;
+
+ *last_index = 0; /* a one-shot read */
+
+ req.lstat_version = LSTAT_VERSION;
+ req.state = lstat_control.state;
+ req.maxcpus = num_online_cpus();
+ req.cycleval = CPU_CYCLE_FREQUENCY;
+#ifdef notyet
+ req.kernel_magic_addr = (void *) &_etext;
+ req.kernel_end_addr = (void *) &_etext;
+#endif
+ req.uts = system_utsname;
+ req.intervals = lstat_control.intervals;
+
+ req.first_started_time = lstat_control.first_started_time;
+ req.started_time = lstat_control.started_time;
+ req.started_cycles64 = lstat_control.started_cycles64;
+
+ req.next_free_dir_index = lstat_control.next_free_dir_index;
+ req.next_free_read_lock_index = lstat_control.next_free_read_lock_index;
+ req.dir_overflow = lstat_control.dir_overflow;
+ req.rwlock_overflow = lstat_control.rwlock_overflow;
+
+ if (lstat_control.state == LSTAT_OFF) {
+ if (req.intervals == 0) {
+ /* mesasurement is off and no valid data present */
+ next_ret_bcount = sizeof (lstat_user_request_t);
+ req.enabled_cycles64 = 0;
+
+ if ((actual_ret_bcount + next_ret_bcount) > max_len)
+ return actual_ret_bcount;
+
+ copy_to_user(buffer, (void *) &req, next_ret_bcount);
+ actual_ret_bcount += next_ret_bcount;
+ return actual_ret_bcount;
+ } else {
+ /*
+ * measurement is off but valid data present
+ * fetch time info from lstat_control
+ */
+ req.ending_time = lstat_control.ending_time;
+ req.ending_cycles64 = lstat_control.ending_cycles64;
+ req.enabled_cycles64 = lstat_control.enabled_cycles64;
+ }
+ } else {
+ /*
+ * this must be a read while data active--use current time,
+ * etc
+ */
+ do_gettimeofday(&tv);
+ req.ending_time = tv.tv_sec;
+ req.ending_cycles64 = get_cycles64();
+ req.enabled_cycles64 = req.ending_cycles64 -
+ req.started_cycles64 + lstat_control.enabled_cycles64;
+ }
+
+ next_ret_bcount = sizeof (lstat_user_request_t);
+ if ((actual_ret_bcount + next_ret_bcount) > max_len)
+ return actual_ret_bcount;
+
+ copy_to_user(buffer, (void *) &req, next_ret_bcount);
+ actual_ret_bcount += next_ret_bcount;
+
+ if (!lstat_control.counts[0]) /* not initialized? */
+ return actual_ret_bcount;
+
+ next_ret_bcount = sizeof (lstat_cpu_counts_t);
+ for (cpu = 0; cpu < num_online_cpus(); cpu++) {
+ if ((actual_ret_bcount + next_ret_bcount) > max_len)
+ return actual_ret_bcount; /* leave early */
+ copy_to_user(buffer + actual_ret_bcount,
+ lstat_control.counts[cpu], next_ret_bcount);
+ actual_ret_bcount += next_ret_bcount;
+ }
+
+ next_ret_bcount = LSTAT_MAX_STAT_INDEX *
+ sizeof (lstat_directory_entry_t);
+ if (((actual_ret_bcount + next_ret_bcount) > max_len)
+ || !lstat_control.dir)
+ return actual_ret_bcount; /* leave early */
+
+ copy_to_user(buffer + actual_ret_bcount, lstat_control.dir,
+ next_ret_bcount);
+ actual_ret_bcount += next_ret_bcount;
+
+ next_ret_bcount = sizeof (lstat_read_lock_cpu_counts_t);
+ for (cpu = 0; cpu < num_online_cpus(); cpu++) {
+ if (actual_ret_bcount + next_ret_bcount > max_len)
+ return actual_ret_bcount;
+ copy_to_user(buffer + actual_ret_bcount,
+ lstat_control.read_lock_counts[cpu],
+ next_ret_bcount);
+ actual_ret_bcount += next_ret_bcount;
+ }
+
+ return actual_ret_bcount;
+}
+
+/*
+ * Writing to the /proc lockmeter node enables or disables metering.
+ * based upon the first byte of the "written" data.
+ * The following values are defined:
+ * LSTAT_ON: 1st call: allocates storage, intializes and turns on measurement
+ * subsequent calls just turn on measurement
+ * LSTAT_OFF: turns off measurement
+ * LSTAT_RESET: resets statistics
+ * LSTAT_RELEASE: releases statistics storage
+ *
+ * This allows one to accumulate statistics over several lockstat runs:
+ *
+ * lockstat on
+ * lockstat off
+ * ...repeat above as desired...
+ * lockstat get
+ * ...now start a new set of measurements...
+ * lockstat reset
+ * lockstat on
+ * ...
+ *
+ */
+ssize_t
+put_lockmeter_info(const char *buffer, size_t len)
+{
+ int error = 0;
+ int dirsize, countsize, read_lock_countsize, hashsize;
+ int cpu;
+ char put_char;
+ int i, read_lock_blocks;
+ unsigned long flags;
+ rwlock_t *lock_ptr;
+ struct timeval tv;
+
+ if (len <= 0)
+ return -EINVAL;
+
+ _raw_spin_lock(&lstat_control.control_lock);
+
+ get_user(put_char, buffer);
+ switch (put_char) {
+
+ case LSTAT_OFF:
+ if (lstat_control.state != LSTAT_OFF) {
+ /*
+ * To avoid seeing read lock hold times in an
+ * inconsisent state, we have to follow this protocol
+ * to turn off statistics
+ */
+ local_irq_save(flags);
+ /*
+ * getting this lock will stop any read lock block
+ * allocations
+ */
+ _raw_spin_lock(&lstat_control.directory_lock);
+ /*
+ * keep any more read lock blocks from being
+ * allocated
+ */
+ lstat_control.state = LSTAT_OFF;
+ /* record how may read lock blocks there are */
+ read_lock_blocks =
+ lstat_control.next_free_read_lock_index;
+ _raw_spin_unlock(&lstat_control.directory_lock);
+ /* now go through the list of read locks */
+ cpu = THIS_CPU_NUMBER;
+ for (i = 1; i < read_lock_blocks; i++) {
+ lock_ptr =
+ (*lstat_control.read_lock_counts[cpu])[i].
+ lock_ptr;
+ /* is this saved lock address still valid? */
+ if (GET_RWINDEX(lock_ptr) == i) {
+ /*
+ * lock address appears to still be
+ * valid because we only hold one lock
+ * at a time, this can't cause a
+ * deadlock unless this is a lock held
+ * as part of the current system call
+ * path. At the moment there
+ * are no READ mode locks held to get
+ * here from user space, so we solve
+ * this by skipping locks held in
+ * write mode.
+ */
+ if (RWLOCK_IS_WRITE_LOCKED(lock_ptr)) {
+ PUT_RWINDEX(lock_ptr, 0);
+ continue;
+ }
+ /*
+ * now we know there are no read
+ * holders of this lock! stop
+ * statistics collection for this
+ * lock
+ */
+ _raw_write_lock(lock_ptr);
+ PUT_RWINDEX(lock_ptr, 0);
+ _raw_write_unlock(lock_ptr);
+ }
+ /*
+ * it may still be possible for the hold time
+ * sum to be negative e.g. if a lock is
+ * reallocated while "busy" we will have to fix
+ * this up in the data reduction program.
+ */
+ }
+ local_irq_restore(flags);
+ lstat_control.intervals++;
+ lstat_control.ending_cycles64 = get_cycles64();
+ lstat_control.enabled_cycles64 +=
+ lstat_control.ending_cycles64 -
+ lstat_control.started_cycles64;
+ do_gettimeofday(&tv);
+ lstat_control.ending_time = tv.tv_sec;
+ /*
+ * don't deallocate the structures -- we may do a
+ * lockstat on to add to the data that is already
+ * there. Use LSTAT_RELEASE to release storage
+ */
+ } else {
+ error = -EBUSY; /* already OFF */
+ }
+ break;
+
+ case LSTAT_ON:
+ if (lstat_control.state == LSTAT_OFF) {
+#ifdef DEBUG_LOCKMETER
+ printk("put_lockmeter_info(cpu=%d): LSTAT_ON\n",
+ THIS_CPU_NUMBER);
+#endif
+ lstat_control.next_free_dir_index = 1; /* 0 is for overflows */
+
+ dirsize = LSTAT_MAX_STAT_INDEX *
+ sizeof (lstat_directory_entry_t);
+ hashsize =
+ (1 + LSTAT_HASH_TABLE_SIZE) * sizeof (ushort);
+ countsize = sizeof (lstat_cpu_counts_t);
+ read_lock_countsize =
+ sizeof (lstat_read_lock_cpu_counts_t);
+#ifdef DEBUG_LOCKMETER
+ printk(" dirsize:%d", dirsize);
+ printk(" hashsize:%d", hashsize);
+ printk(" countsize:%d", countsize);
+ printk(" read_lock_countsize:%d\n",
+ read_lock_countsize);
+#endif
+#ifdef DEBUG_LOCKMETER
+ {
+ int secs;
+ unsigned long cycles;
+ uint64_t cycles64;
+
+ do_gettimeofday(&tv);
+ secs = tv.tv_sec;
+ do {
+ do_gettimeofday(&tv);
+ } while (secs == tv.tv_sec);
+ cycles = get_cycles();
+ cycles64 = get_cycles64();
+ secs = tv.tv_sec;
+ do {
+ do_gettimeofday(&tv);
+ } while (secs == tv.tv_sec);
+ cycles = get_cycles() - cycles;
+ cycles64 = get_cycles64() - cycles;
+ printk("lockmeter: cycleFrequency:%d "
+ "cycles:%d cycles64:%d\n",
+ CPU_CYCLE_FREQUENCY, cycles, cycles64);
+ }
+#endif
+
+ /*
+ * if this is the first call, allocate storage and
+ * initialize
+ */
+ if (!lstat_control.hashtab) {
+
+ spin_lock_init(&lstat_control.directory_lock);
+
+ /* guarantee all pointers at zero */
+ init_control_space();
+
+ lstat_control.hashtab =
+ kmalloc(hashsize, GFP_KERNEL);
+ if (!lstat_control.hashtab) {
+ error = -ENOSPC;
+#ifdef DEBUG_LOCKMETER
+ printk("!!error kmalloc of hashtab\n");
+#endif
+ }
+ lstat_control.dir = vmalloc(dirsize);
+ if (!lstat_control.dir) {
+ error = -ENOSPC;
+#ifdef DEBUG_LOCKMETER
+ printk("!!error kmalloc of dir\n");
+#endif
+ }
+
+ for (cpu = 0; cpu < num_online_cpus(); cpu++) {
+ lstat_control.counts[cpu] =
+ vmalloc(countsize);
+ if (!lstat_control.counts[cpu]) {
+ error = -ENOSPC;
+#ifdef DEBUG_LOCKMETER
+ printk("!!error vmalloc of "
+ "counts[%d]\n", cpu);
+#endif
+ }
+ lstat_control.read_lock_counts[cpu] =
+ (lstat_read_lock_cpu_counts_t *)
+ kmalloc(read_lock_countsize,
+ GFP_KERNEL);
+ if (!lstat_control.
+ read_lock_counts[cpu]) {
+ error = -ENOSPC;
+#ifdef DEBUG_LOCKMETER
+ printk("!!error kmalloc of "
+ "read_lock_counts[%d]\n",
+ cpu);
+#endif
+ }
+ }
+ }
+
+ if (error) {
+ /*
+ * One or more kmalloc failures -- free
+ * everything
+ */
+ release_control_space();
+ } else {
+
+ if (!reset_lstat_data()) {
+ error = -EINVAL;
+ break;
+ };
+
+ /*
+ * record starting and ending times and the
+ * like
+ */
+ if (lstat_control.intervals == 0) {
+ do_gettimeofday(&tv);
+ lstat_control.first_started_time =
+ tv.tv_sec;
+ }
+ lstat_control.started_cycles64 = get_cycles64();
+ do_gettimeofday(&tv);
+ lstat_control.started_time = tv.tv_sec;
+
+ lstat_control.state = LSTAT_ON;
+ }
+ } else {
+ error = -EBUSY; /* already ON */
+ }
+ break;
+
+ case LSTAT_RESET:
+ if (lstat_control.state == LSTAT_OFF) {
+ if (!reset_lstat_data())
+ error = -EINVAL;
+ } else {
+ error = -EBUSY; /* still on; can't reset */
+ }
+ break;
+
+ case LSTAT_RELEASE:
+ if (lstat_control.state == LSTAT_OFF) {
+ release_control_space();
+ lstat_control.intervals = 0;
+ lstat_control.enabled_cycles64 = 0;
+ } else {
+ error = -EBUSY;
+ }
+ break;
+
+ default:
+ error = -EINVAL;
+ } /* switch */
+
+ _raw_spin_unlock(&lstat_control.control_lock);
+ return error ? error : len;
+}
+
+#ifdef USER_MODE_TESTING
+/* following used for user mode testing */
+void
+lockmeter_init()
+{
+ int dirsize, hashsize, countsize, read_lock_countsize, cpu;
+
+ printf("lstat_control is at %x size=%d\n", &lstat_control,
+ sizeof (lstat_control));
+ printf("sizeof(spinlock_t)=%d\n", sizeof (spinlock_t));
+ lstat_control.state = LSTAT_ON;
+
+ lstat_control.directory_lock = SPIN_LOCK_UNLOCKED;
+ lstat_control.next_free_dir_index = 1; /* 0 is for overflows */
+ lstat_control.next_free_read_lock_index = 1;
+
+ dirsize = LSTAT_MAX_STAT_INDEX * sizeof (lstat_directory_entry_t);
+ hashsize = (1 + LSTAT_HASH_TABLE_SIZE) * sizeof (ushort);
+ countsize = sizeof (lstat_cpu_counts_t);
+ read_lock_countsize = sizeof (lstat_read_lock_cpu_counts_t);
+
+ lstat_control.hashtab = (ushort *) malloc(hashsize);
+
+ if (lstat_control.hashtab == 0) {
+ printf("malloc failure for at line %d in lockmeter.c\n",
+ __LINE__);
+ exit(0);
+ }
+
+ lstat_control.dir = (lstat_directory_entry_t *) malloc(dirsize);
+
+ if (lstat_control.dir == 0) {
+ printf("malloc failure for at line %d in lockmeter.c\n", cpu,
+ __LINE__);
+ exit(0);
+ }
+
+ for (cpu = 0; cpu < num_online_cpus(); cpu++) {
+ int j, k;
+ j = (int) (lstat_control.counts[cpu] =
+ (lstat_cpu_counts_t *) malloc(countsize));
+ k = (int) (lstat_control.read_lock_counts[cpu] =
+ (lstat_read_lock_cpu_counts_t *)
+ malloc(read_lock_countsize));
+ if (j * k == 0) {
+ printf("malloc failure for cpu=%d at line %d in "
+ "lockmeter.c\n", cpu, __LINE__);
+ exit(0);
+ }
+ }
+
+ memset(lstat_control.hashtab, 0, hashsize);
+ memset(lstat_control.dir, 0, dirsize);
+
+ for (cpu = 0; cpu < num_online_cpus(); cpu++) {
+ memset(lstat_control.counts[cpu], 0, countsize);
+ memset(lstat_control.read_lock_counts[cpu], 0,
+ read_lock_countsize);
+ }
+}
+
+asm(" \
+.align 4 \
+.globl __write_lock_failed \
+__write_lock_failed: \
+ " LOCK "addl $" RW_LOCK_BIAS_STR ",(%eax) \
+1: cmpl $" RW_LOCK_BIAS_STR ",(%eax) \
+ jne 1b \
+\
+ " LOCK "subl $" RW_LOCK_BIAS_STR ",(%eax) \
+ jnz __write_lock_failed \
+ ret \
+\
+\
+.align 4 \
+.globl __read_lock_failed \
+__read_lock_failed: \
+ lock ; incl (%eax) \
+1: cmpl $1,(%eax) \
+ js 1b \
+\
+ lock ; decl (%eax) \
+ js __read_lock_failed \
+ ret \
+");
+#endif
+
+EXPORT_SYMBOL(_metered_spin_lock);
+EXPORT_SYMBOL(_metered_spin_unlock);
+EXPORT_SYMBOL(_metered_spin_trylock);
+EXPORT_SYMBOL(_metered_read_lock);
+EXPORT_SYMBOL(_metered_read_unlock);
+EXPORT_SYMBOL(_metered_write_lock);
+EXPORT_SYMBOL(_metered_write_unlock);
--- diff/mm/usercopy.c 1970-01-01 00:00:00.000000000 +0000
+++ source/mm/usercopy.c 2004-03-16 11:27:48.696784624 +0000
@@ -0,0 +1,291 @@
+/*
+ * linux/mm/usercopy.c
+ *
+ * (C) Copyright 2003 Ingo Molnar
+ *
+ * Generic implementation of all the user-VM access functions, without
+ * relying on being able to access the VM directly.
+ */
+
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/errno.h>
+#include <linux/mm.h>
+#include <linux/highmem.h>
+#include <linux/pagemap.h>
+#include <linux/smp_lock.h>
+#include <linux/ptrace.h>
+#include <linux/interrupt.h>
+
+#include <asm/pgtable.h>
+#include <asm/uaccess.h>
+#include <asm/atomic_kmap.h>
+
+/*
+ * Get kernel address of the user page and pin it.
+ */
+static inline struct page *pin_page(unsigned long addr, int write)
+{
+ struct mm_struct *mm = current->mm ? : &init_mm;
+ struct page *page = NULL;
+ int ret;
+
+ /*
+ * Do a quick atomic lookup first - this is the fastpath.
+ */
+retry:
+ page = follow_page(mm, addr, write);
+ if (likely(page != NULL)) {
+ if (!PageReserved(page))
+ get_page(page);
+ return page;
+ }
+
+ /*
+ * No luck - bad address or need to fault in the page:
+ */
+
+ /* Release the lock so get_user_pages can sleep */
+ spin_unlock(&mm->page_table_lock);
+
+ /*
+ * In the context of filemap_copy_from_user(), we are not allowed
+ * to sleep. We must fail this usercopy attempt and allow
+ * filemap_copy_from_user() to recover: drop its atomic kmap and use
+ * a sleeping kmap instead.
+ */
+ if (in_atomic()) {
+ spin_lock(&mm->page_table_lock);
+ return NULL;
+ }
+
+ down_read(&mm->mmap_sem);
+ ret = get_user_pages(current, mm, addr, 1, write, 0, NULL, NULL);
+ up_read(&mm->mmap_sem);
+ spin_lock(&mm->page_table_lock);
+
+ if (ret <= 0)
+ return NULL;
+
+ /*
+ * Go try the follow_page again.
+ */
+ goto retry;
+}
+
+static inline void unpin_page(struct page *page)
+{
+ put_page(page);
+}
+
+/*
+ * Access another process' address space.
+ * Source/target buffer must be kernel space,
+ * Do not walk the page table directly, use get_user_pages
+ */
+static int rw_vm(unsigned long addr, void *buf, int len, int write)
+{
+ struct mm_struct *mm = current->mm ? : &init_mm;
+
+ if (!len)
+ return 0;
+
+ spin_lock(&mm->page_table_lock);
+
+ /* ignore errors, just check how much was sucessfully transfered */
+ while (len) {
+ struct page *page = NULL;
+ int bytes, offset;
+ void *maddr;
+
+ page = pin_page(addr, write);
+ if (!page)
+ break;
+
+ bytes = len;
+ offset = addr & (PAGE_SIZE-1);
+ if (bytes > PAGE_SIZE-offset)
+ bytes = PAGE_SIZE-offset;
+
+ maddr = kmap_atomic(page, KM_USER_COPY);
+
+#define HANDLE_TYPE(type) \
+ case sizeof(type): *(type *)(maddr+offset) = *(type *)(buf); break;
+
+ if (write) {
+ switch (bytes) {
+ HANDLE_TYPE(char);
+ HANDLE_TYPE(int);
+ HANDLE_TYPE(long long);
+ default:
+ memcpy(maddr + offset, buf, bytes);
+ }
+ } else {
+#undef HANDLE_TYPE
+#define HANDLE_TYPE(type) \
+ case sizeof(type): *(type *)(buf) = *(type *)(maddr+offset); break;
+ switch (bytes) {
+ HANDLE_TYPE(char);
+ HANDLE_TYPE(int);
+ HANDLE_TYPE(long long);
+ default:
+ memcpy(buf, maddr + offset, bytes);
+ }
+#undef HANDLE_TYPE
+ }
+ kunmap_atomic(maddr, KM_USER_COPY);
+ unpin_page(page);
+ len -= bytes;
+ buf += bytes;
+ addr += bytes;
+ }
+ spin_unlock(&mm->page_table_lock);
+
+ return len;
+}
+
+static int str_vm(unsigned long addr, void *buf0, int len, int copy)
+{
+ struct mm_struct *mm = current->mm ? : &init_mm;
+ struct page *page;
+ void *buf = buf0;
+
+ if (!len)
+ return len;
+
+ spin_lock(&mm->page_table_lock);
+
+ /* ignore errors, just check how much was sucessfully transfered */
+ while (len) {
+ int bytes, offset, left, copied;
+ char *maddr;
+
+ page = pin_page(addr, copy == 2);
+ if (!page) {
+ spin_unlock(&mm->page_table_lock);
+ return -EFAULT;
+ }
+ bytes = len;
+ offset = addr & (PAGE_SIZE-1);
+ if (bytes > PAGE_SIZE-offset)
+ bytes = PAGE_SIZE-offset;
+
+ maddr = kmap_atomic(page, KM_USER_COPY);
+ if (copy == 2) {
+ memset(maddr + offset, 0, bytes);
+ copied = bytes;
+ left = 0;
+ } else if (copy == 1) {
+ left = strncpy_count(buf, maddr + offset, bytes);
+ copied = bytes - left;
+ } else {
+ copied = strnlen(maddr + offset, bytes);
+ left = bytes - copied;
+ }
+ BUG_ON(bytes < 0 || copied < 0);
+ kunmap_atomic(maddr, KM_USER_COPY);
+ unpin_page(page);
+ len -= copied;
+ buf += copied;
+ addr += copied;
+ if (left)
+ break;
+ }
+ spin_unlock(&mm->page_table_lock);
+
+ return len;
+}
+
+/*
+ * Copies memory from userspace (ptr) into kernelspace (val).
+ *
+ * returns # of bytes not copied.
+ */
+int get_user_size(unsigned int size, void *val, const void *ptr)
+{
+ int ret;
+
+ if (unlikely(segment_eq(get_fs(), KERNEL_DS)))
+ ret = __direct_copy_from_user(val, ptr, size);
+ else
+ ret = rw_vm((unsigned long)ptr, val, size, 0);
+ if (ret)
+ /*
+ * Zero the rest:
+ */
+ memset(val + size - ret, 0, ret);
+ return ret;
+}
+
+/*
+ * Copies memory from kernelspace (val) into userspace (ptr).
+ *
+ * returns # of bytes not copied.
+ */
+int put_user_size(unsigned int size, const void *val, void *ptr)
+{
+ if (unlikely(segment_eq(get_fs(), KERNEL_DS)))
+ return __direct_copy_to_user(ptr, val, size);
+ else
+ return rw_vm((unsigned long)ptr, (void *)val, size, 1);
+}
+
+int copy_str_fromuser_size(unsigned int size, void *val, const void *ptr)
+{
+ int copied, left;
+
+ if (unlikely(segment_eq(get_fs(), KERNEL_DS))) {
+ left = strncpy_count(val, ptr, size);
+ copied = size - left;
+ BUG_ON(copied < 0);
+
+ return copied;
+ }
+ left = str_vm((unsigned long)ptr, val, size, 1);
+ if (left < 0)
+ return left;
+ copied = size - left;
+ BUG_ON(copied < 0);
+
+ return copied;
+}
+
+int strlen_fromuser_size(unsigned int size, const void *ptr)
+{
+ int copied, left;
+
+ if (unlikely(segment_eq(get_fs(), KERNEL_DS))) {
+ copied = strnlen(ptr, size) + 1;
+ BUG_ON(copied < 0);
+
+ return copied;
+ }
+ left = str_vm((unsigned long)ptr, NULL, size, 0);
+ if (left < 0)
+ return 0;
+ copied = size - left + 1;
+ BUG_ON(copied < 0);
+
+ return copied;
+}
+
+int zero_user_size(unsigned int size, void *ptr)
+{
+ int left;
+
+ if (unlikely(segment_eq(get_fs(), KERNEL_DS))) {
+ memset(ptr, 0, size);
+ return 0;
+ }
+ left = str_vm((unsigned long)ptr, NULL, size, 2);
+ if (left < 0)
+ return size;
+ return left;
+}
+
+EXPORT_SYMBOL(get_user_size);
+EXPORT_SYMBOL(put_user_size);
+EXPORT_SYMBOL(zero_user_size);
+EXPORT_SYMBOL(copy_str_fromuser_size);
+EXPORT_SYMBOL(strlen_fromuser_size);
+
--- diff/net/ipmi/Makefile 1970-01-01 00:00:00.000000000 +0000
+++ source/net/ipmi/Makefile 2004-03-16 11:27:48.701783864 +0000
@@ -0,0 +1 @@
+obj-$(CONFIG_IPMI_SOCKET) = af_ipmi.o
--- diff/net/ipmi/af_ipmi.c 1970-01-01 00:00:00.000000000 +0000
+++ source/net/ipmi/af_ipmi.c 2004-03-16 11:27:48.701783864 +0000
@@ -0,0 +1,612 @@
+/*
+ * IPMI Socket Glue
+ *
+ * Author: Louis Zhuang <louis.zhuang@linux.intel.com>
+ * Copyright by Intel Corp., 2003
+ */
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/major.h>
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/stat.h>
+#include <linux/socket.h>
+#include <linux/fcntl.h>
+#include <linux/sockios.h>
+#include <linux/net.h>
+#include <linux/in.h>
+#include <linux/fs.h>
+#include <linux/slab.h>
+#include <asm/uaccess.h>
+#include <linux/skbuff.h>
+#include <linux/tcp.h>
+#include <net/sock.h>
+#include <linux/proc_fs.h>
+#include <linux/init.h>
+#include <linux/poll.h>
+#include <linux/smp_lock.h>
+#include <linux/mount.h>
+#include <linux/security.h>
+#include <linux/ipmi.h>
+#include <net/af_ipmi.h>
+
+#define IPMI_SOCKINTF_VERSION "v31"
+
+#ifdef CONFIG_DEBUG_KERNEL
+static int debug = 0;
+#define dbg(format, arg...) \
+ do { \
+ if(debug) \
+ printk (KERN_DEBUG "%s: " format "\n", \
+ __FUNCTION__, ## arg); \
+ } while(0)
+#else
+#define dbg(format, arg...)
+#endif /* CONFIG_DEBUG_KERNEL */
+
+#define err(format, arg...) \
+ printk(KERN_ERR "%s: " format "\n", \
+ __FUNCTION__ , ## arg)
+#define info(format, arg...) \
+ printk(KERN_INFO "%s: " format "\n", \
+ __FUNCTION__ , ## arg)
+#define warn(format, arg...) \
+ printk(KERN_WARNING "%s: " format "\n", \
+ __FUNCTION__ , ## arg)
+#define trace(format, arg...) \
+ printk(KERN_INFO "%s(" format ")\n", \
+ __FUNCTION__ , ## arg)
+
+struct ipmi_sock {
+ /* WARNING: sk has to be the first member */
+ struct sock sk;
+
+ ipmi_user_t user;
+ struct sockaddr_ipmi addr;
+ struct list_head msg_list;
+
+ wait_queue_head_t wait;
+ spinlock_t lock;
+
+ int default_retries;
+ unsigned int default_retry_time_ms;
+};
+
+static kmem_cache_t *ipmi_sk_cachep = NULL;
+
+static atomic_t ipmi_nr_socks = ATOMIC_INIT(0);
+
+
+
+/*
+ * utility functions
+ */
+static inline struct ipmi_sock *to_ipmi_sock(struct sock *sk)
+{
+ return container_of(sk, struct ipmi_sock, sk);
+}
+
+static inline void ipmi_release_sock(struct sock *sk, int embrion)
+{
+ struct ipmi_sock *i = to_ipmi_sock(sk);
+ struct sk_buff *skb;
+
+ if (i->user) {
+ ipmi_destroy_user(i->user);
+ i->user = NULL;
+ }
+
+ sock_orphan(&i->sk);
+ sk->sk_shutdown = SHUTDOWN_MASK;
+ sk->sk_state = TCP_CLOSE;
+
+ while((skb=skb_dequeue(&sk->sk_receive_queue))!=NULL)
+ kfree_skb(skb);
+
+ sock_put(sk);
+}
+
+static inline long ipmi_wait_for_queue(struct ipmi_sock *i, long timeo)
+{
+
+ DECLARE_WAITQUEUE(wait, current);
+
+ set_current_state(TASK_INTERRUPTIBLE);
+ add_wait_queue_exclusive(&i->wait, &wait);
+ timeo = schedule_timeout(timeo);
+ remove_wait_queue(&i->wait, &wait);
+ return timeo;
+}
+
+/*
+ * IPMI operation functions
+ */
+static void sock_receive_handler(struct ipmi_recv_msg *msg,
+ void *handler_data)
+{
+ struct ipmi_sock *i = (struct ipmi_sock *)handler_data;
+ unsigned long flags;
+
+ spin_lock_irqsave(&i->lock, flags);
+ list_add_tail(&msg->link, &i->msg_list);
+ spin_unlock_irqrestore(&i->lock, flags);
+
+ wake_up_interruptible(&i->wait);
+}
+
+/*
+ * protocol operation functions
+ */
+static int ipmi_release(struct socket *sock)
+{
+ struct sock *sk = sock->sk;
+
+ if (!sk)
+ return 0;
+
+ sock->sk=NULL;
+ ipmi_release_sock(sk, 0);
+ return 0;
+}
+
+static struct ipmi_user_hndl ipmi_hnd = {
+ .ipmi_recv_hndl = sock_receive_handler
+};
+
+static int ipmi_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
+{
+ struct ipmi_sock *i = to_ipmi_sock(sock->sk);
+ struct sockaddr_ipmi *addr = (struct sockaddr_ipmi *)uaddr;
+ int err = -EINVAL;
+
+ if (i->user != NULL) {
+ dbg("Cannot bind twice: %p", i->user);
+ return -EINVAL;
+ }
+
+ err = ipmi_create_user(addr->if_num, &ipmi_hnd, i, &i->user);
+ if (err) {
+ dbg("Cannot create user for the socket: %p", i->user);
+ return err;
+ }
+
+ memcpy(&i->addr, addr, sizeof(i->addr));
+ return 0;
+}
+
+static int ipmi_getname(struct socket *sock, struct sockaddr *uaddr, int *uaddr_len, int peer)
+{
+ struct ipmi_sock *i = to_ipmi_sock(sock->sk);
+ memcpy(uaddr, &i->addr, sizeof(i->addr));
+ return 0;
+}
+
+static unsigned int ipmi_poll(struct file * file, struct socket *sock, poll_table *wait)
+{
+ unsigned int has_msg = 0;
+ struct ipmi_sock *i = to_ipmi_sock(sock->sk);
+ unsigned long flags;
+
+ poll_wait(file, &i->wait, wait);
+ spin_lock_irqsave(&i->lock, flags);
+ if (!list_empty(&i->msg_list))
+ has_msg = 1;
+ spin_unlock_irqrestore(&i->lock, flags);
+
+ if (has_msg)
+ return POLLIN | POLLRDNORM;
+ return 0;
+}
+
+static int ipmi_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
+{
+ struct ipmi_sock *i = to_ipmi_sock(sock->sk);
+ struct ipmi_cmdspec val;
+ int ival;
+ unsigned int uival;
+ int err;
+
+ dbg("cmd=%#x, arg=%#lx", cmd, arg);
+ switch(cmd) {
+ case SIOCIPMIREGCMD:
+ err = copy_from_user((void *)&val, (void *)arg,
+ sizeof(cmd));
+ if (err) {
+ err = -EFAULT;
+ break;
+ }
+
+ err = ipmi_register_for_cmd(i->user, val.netfn,
+ val.cmd);
+ break;
+
+ case SIOCIPMIUNREGCMD:
+ err = copy_from_user((void *)&val, (void *)arg,
+ sizeof(cmd));
+ if (err) {
+ err = -EFAULT;
+ break;
+ }
+
+ err = ipmi_unregister_for_cmd(i->user, val.netfn,
+ val.cmd);
+ break;
+
+ case SIOCIPMIGETEVENT:
+ err = copy_from_user((void *)&ival, (void *)arg,
+ sizeof(ival));
+ if (err) {
+ err = -EFAULT;
+ break;
+ }
+
+ err = ipmi_set_gets_events(i->user, ival);
+ break;
+
+ case SIOCIPMISETADDR:
+ err = copy_from_user((void *)&uival, (void *)arg,
+ sizeof(uival));
+ if (err) {
+ err = -EFAULT;
+ break;
+ }
+
+ ipmi_set_my_address(i->user, uival);
+ break;
+
+ case SIOCIPMIGETADDR:
+ uival = ipmi_get_my_address(i->user);
+
+ if (copy_to_user((void *) arg, &uival, sizeof(uival))) {
+ err = -EFAULT;
+ break;
+ }
+ err = 0;
+ break;
+
+ case SIOCIPMISETTIMING:
+ {
+ struct ipmi_timing_parms parms;
+
+ if (copy_from_user(&parms, (void *) arg, sizeof(parms))) {
+ err = -EFAULT;
+ break;
+ }
+
+ i->default_retries = parms.retries;
+ i->default_retry_time_ms = parms.retry_time_ms;
+ err = 0;
+ break;
+ }
+
+ case SIOCIPMIGETTIMING:
+ {
+ struct ipmi_timing_parms parms;
+
+ parms.retries = i->default_retries;
+ parms.retry_time_ms = i->default_retry_time_ms;
+
+ if (copy_to_user((void *) arg, &parms, sizeof(parms))) {
+ err = -EFAULT;
+ break;
+ }
+
+ err = 0;
+ break;
+ }
+
+ default:
+ err = dev_ioctl(cmd, (void *)arg);
+ break;
+ }
+
+ return err;
+}
+
+static int ipmi_recvmsg(struct kiocb *iocb, struct socket *sock,
+ struct msghdr *msg, size_t size,
+ int rflags)
+{
+ struct ipmi_sock *i = to_ipmi_sock(sock->sk);
+ long timeo;
+ struct ipmi_recv_msg *rcvmsg;
+ struct sockaddr_ipmi addr;
+ char buf[IPMI_MAX_SOCK_MSG_LENGTH];
+ struct ipmi_sock_msg *smsg = (struct ipmi_sock_msg *)buf;
+ int err;
+ unsigned long flags;
+
+
+ timeo = sock_rcvtimeo(&i->sk, rflags & MSG_DONTWAIT);
+
+ while (1) {
+ spin_lock_irqsave(&i->lock, flags);
+ if (!list_empty(&i->msg_list))
+ break;
+ spin_unlock_irqrestore(&i->lock, flags);
+ if (!timeo) {
+ return -EAGAIN;
+ } else if (signal_pending (current)) {
+ dbg("Signal pending: %d", 1);
+ return -EINTR;
+ }
+
+ timeo = ipmi_wait_for_queue(i, timeo);
+ }
+
+ rcvmsg = list_entry(i->msg_list.next, struct ipmi_recv_msg, link);
+ list_del(&rcvmsg->link);
+ spin_unlock_irqrestore(&i->lock, flags);
+
+ memcpy(&addr.ipmi_addr, &rcvmsg->addr, sizeof(addr.ipmi_addr));
+ addr.if_num = i->addr.if_num;
+ addr.sipmi_family = i->addr.sipmi_family;
+ memcpy(msg->msg_name, &addr, sizeof(addr));
+ msg->msg_namelen = (SOCKADDR_IPMI_OVERHEAD
+ + ipmi_addr_length(rcvmsg->addr.addr_type));
+
+ smsg->recv_type = rcvmsg->recv_type;
+ smsg->msgid = rcvmsg->msgid;
+ smsg->netfn = rcvmsg->msg.netfn;
+ smsg->cmd = rcvmsg->msg.cmd;
+ smsg->data_len = rcvmsg->msg.data_len;
+ memcpy(smsg->data, rcvmsg->msg.data, smsg->data_len);
+
+ ipmi_free_recv_msg(rcvmsg);
+
+ err = memcpy_toiovec(msg->msg_iov, (void *)smsg,
+ sizeof(struct ipmi_sock_msg) + smsg->data_len);
+ if (err) {
+ dbg("Cannot copy data to user: %p", i->user);
+ return err;
+ }
+
+ dbg("user=%p", i->user);
+ dbg("addr_type=%x, channel=%x",
+ addr.ipmi_addr.addr_type, addr.ipmi_addr.channel);
+ dbg("netfn=%#02x, cmd=%#02x, data=%p, data_len=%x",
+ smsg->netfn, smsg->cmd, smsg->data, smsg->data_len);
+
+ return (sizeof(struct ipmi_sock_msg) + smsg->data_len);
+}
+
+static int ipmi_sendmsg(struct kiocb *iocb, struct socket *sock,
+ struct msghdr *msg, size_t len)
+{
+ struct ipmi_sock *i = to_ipmi_sock(sock->sk);
+ struct sockaddr_ipmi *addr = (struct sockaddr_ipmi *)msg->msg_name;
+ struct ipmi_msg imsg;
+ unsigned char buf[IPMI_MAX_SOCK_MSG_LENGTH];
+ struct ipmi_sock_msg *smsg = (struct ipmi_sock_msg *) buf;
+ int err;
+ struct ipmi_timing_parms tparms;
+ struct cmsghdr *cmsg;
+
+ err = ipmi_validate_addr(&addr->ipmi_addr,
+ msg->msg_namelen - SOCKADDR_IPMI_OVERHEAD);
+ if (err) {
+ dbg("Invalid IPMI address: %p", i->user);
+ goto err;
+ }
+
+ if (len > IPMI_MAX_SOCK_MSG_LENGTH) {
+ err = -EINVAL;
+ dbg("Message too long: %p", i->user);
+ goto err;
+ }
+
+ if (len < sizeof(struct ipmi_sock_msg)) {
+ err = -EINVAL;
+ dbg("Msg data too small for header: %p", i->user);
+ goto err;
+ }
+
+ err = memcpy_fromiovec((void *)smsg, msg->msg_iov, len);
+ if (err) {
+ dbg("Cannot copy data to kernel: %p", i->user);
+ goto err;
+ }
+
+ if (len < smsg->data_len+sizeof(struct ipmi_sock_msg)) {
+ err = -EINVAL;
+ dbg("Msg data is out of bound: %p", i->user);
+ goto err;
+ }
+
+ /* Set defaults. */
+ tparms.retries = i->default_retries;
+ tparms.retry_time_ms = i->default_retry_time_ms;
+
+ for (cmsg=CMSG_FIRSTHDR(msg);
+ cmsg;
+ cmsg = CMSG_NXTHDR(msg, cmsg))
+ {
+ if (cmsg->cmsg_len < sizeof(struct cmsghdr)) {
+ err = -EINVAL;
+ dbg("cmsg length too short: %p", i->user);
+ goto err;
+ }
+
+ if (cmsg->cmsg_level != SOL_SOCKET)
+ continue;
+
+ if (cmsg->cmsg_type == IPMI_CMSG_TIMING_PARMS) {
+ struct ipmi_timing_parms *pparms;
+
+ if (cmsg->cmsg_len != CMSG_LEN(sizeof(*pparms))) {
+ err = -EINVAL;
+ dbg("timing parms cmsg not right size: %p",
+ i->user);
+ goto err;
+ }
+ pparms = (struct ipmi_timing_parms *) CMSG_DATA(cmsg);
+ tparms.retries = pparms->retries;
+ tparms.retry_time_ms = pparms->retry_time_ms;
+ }
+ }
+
+ imsg.netfn = smsg->netfn;
+ imsg.cmd = smsg->cmd;
+ imsg.data = smsg->data;
+ imsg.data_len = smsg->data_len;
+
+ dbg("user=%p", i->user);
+ dbg("addr_type=%x, channel=%x",
+ addr->ipmi_addr.addr_type, addr->ipmi_addr.channel);
+ dbg("netfn=%#02x, cmd=%#02x, data=%p, data_len=%x",
+ imsg.netfn, imsg.cmd, imsg.data, imsg.data_len);
+ err = ipmi_request_settime(i->user, &addr->ipmi_addr,
+ smsg->msgid, &imsg, NULL, 0,
+ tparms.retries, tparms.retry_time_ms);
+ if (err) {
+ dbg("Cannot send message: %p", i->user);
+ goto err;
+ }
+
+err:
+ return err;
+}
+
+static struct proto_ops ipmi_ops = {
+ .family = PF_IPMI,
+ .owner = THIS_MODULE,
+ .release = ipmi_release,
+ .bind = ipmi_bind,
+ .connect = sock_no_connect,
+ .socketpair = sock_no_socketpair,
+ .accept = sock_no_accept,
+ .getname = ipmi_getname,
+ .poll = ipmi_poll,
+ .ioctl = ipmi_ioctl,
+ .listen = sock_no_listen,
+ .shutdown = sock_no_shutdown,
+ .setsockopt = sock_no_setsockopt,
+ .getsockopt = sock_no_getsockopt,
+ .sendmsg = ipmi_sendmsg,
+ .recvmsg = ipmi_recvmsg,
+ .mmap = sock_no_mmap,
+ .sendpage = sock_no_sendpage
+};
+
+
+static void ipmi_sock_destructor(struct sock *sk)
+{
+ skb_queue_purge(&sk->sk_receive_queue);
+
+ BUG_TRAP(!atomic_read(&sk->sk_wmem_alloc));
+ BUG_TRAP(sk_unhashed(sk));
+ BUG_TRAP(!sk->sk_socket);
+ if (!sock_flag(sk, SOCK_DEAD)) {
+ printk("Attempt to release alive ipmi socket: %p\n", sk);
+ return;
+ }
+
+ atomic_dec(&ipmi_nr_socks);
+}
+
+/*
+ * net protocol functions
+ */
+static struct ipmi_sock *ipmi_socket_create1(struct socket *sock)
+{
+ struct ipmi_sock *i;
+
+ if (atomic_read(&ipmi_nr_socks) >= 2*files_stat.max_files)
+ return NULL;
+
+ i = (struct ipmi_sock *)sk_alloc(PF_IPMI, GFP_KERNEL,
+ sizeof(struct ipmi_sock), ipmi_sk_cachep);
+ if (!i) {
+ return NULL;
+ }
+
+
+ sock_init_data(sock, &i->sk);
+ sk_set_owner(&i->sk, THIS_MODULE);
+ i->sk.sk_destruct = ipmi_sock_destructor;
+ i->sk.sk_rcvtimeo = 5*HZ;
+ spin_lock_init(&i->lock);
+ INIT_LIST_HEAD(&i->msg_list);
+ init_waitqueue_head(&i->wait);
+
+ /* Set to use default values. */
+ i->default_retries = -1;
+ i->default_retry_time_ms = 0;
+
+ atomic_inc(&ipmi_nr_socks);
+ return i;
+}
+
+static int ipmi_socket_create(struct socket *sock, int protocol)
+{
+ if (!capable(CAP_NET_RAW))
+ return -EPERM;
+ if (protocol && protocol != PF_IPMI)
+ return -EPROTONOSUPPORT;
+
+ sock->state = SS_UNCONNECTED;
+
+ switch (sock->type) {
+ case SOCK_RAW:
+ sock->type=SOCK_DGRAM;
+ case SOCK_DGRAM:
+ sock->ops = &ipmi_ops;
+ break;
+ default:
+ return -EPROTONOSUPPORT;
+ }
+
+ return ipmi_socket_create1(sock)? 0 : -ENOMEM;
+}
+
+static struct net_proto_family ipmi_family_ops = {
+ .family = PF_IPMI,
+ .create = ipmi_socket_create,
+ .owner = THIS_MODULE,
+};
+
+
+/*
+ * init/exit functions
+ */
+static int __init ipmi_socket_init(void)
+{
+
+ int err=0;
+
+ printk(KERN_INFO "ipmi socket interface version "
+ IPMI_SOCKINTF_VERSION "\n");
+
+ ipmi_sk_cachep = kmem_cache_create("ipmi_sock",
+ sizeof(struct ipmi_sock), 0,
+ SLAB_HWCACHE_ALIGN, 0, 0);
+ if (!ipmi_sk_cachep) {
+ printk(KERN_CRIT "%s: Unable to create ipmi_sock SLAB cache!\n", __func__);
+ err = -ENOMEM;
+ goto out;
+ }
+
+ err = sock_register(&ipmi_family_ops);
+ if (err)
+ kmem_cache_destroy(ipmi_sk_cachep);
+out:
+ return err;
+}
+
+static void __exit ipmi_socket_exit(void)
+{
+ sock_unregister(PF_IPMI);
+ kmem_cache_destroy(ipmi_sk_cachep);
+}
+
+#ifdef CONFIG_DEBUG_KERNEL
+module_param(debug, int, 0);
+#endif
+module_init(ipmi_socket_init);
+module_exit(ipmi_socket_exit);
+
+MODULE_LICENSE("GPL");
--- diff/scripts/basic/Makefile 1970-01-01 00:00:00.000000000 +0000
+++ source/scripts/basic/Makefile 2004-03-16 11:27:48.712782192 +0000
@@ -0,0 +1,18 @@
+###
+# Makefile.basic list the most basic programs used during the build process.
+# The programs listed herein is what is needed to do the basic stuff,
+# such as splitting .config and fix dependency file.
+# This initial step is needed to avoid files to be recompiled
+# when kernel configuration changes (which is what happens when
+# .config is included by main Makefile.
+# ---------------------------------------------------------------------------
+# fixdep: Used to generate dependency information during build process
+# split-include: Divide all config symbols up in a number of files in
+# include/config/...
+# docproc: Used in Documentation/docbook
+
+host-progs := fixdep split-include docproc
+always := $(host-progs)
+
+# fixdep is needed to compile other host programs
+$(addprefix $(obj)/,$(filter-out fixdep,$(always))): $(obj)/fixdep
--- diff/scripts/basic/docproc.c 1970-01-01 00:00:00.000000000 +0000
+++ source/scripts/basic/docproc.c 2004-03-16 11:27:48.711782344 +0000
@@ -0,0 +1,387 @@
+/*
+ * docproc is a simple preprocessor for the template files
+ * used as placeholders for the kernel internal documentation.
+ * docproc is used for documentation-frontend and
+ * dependency-generator.
+ * The two usages have in common that they require
+ * some knowledge of the .tmpl syntax, therefore they
+ * are kept together.
+ *
+ * documentation-frontend
+ * Scans the template file and call kernel-doc for
+ * all occurrences of ![EIF]file
+ * Beforehand each referenced file are scanned for
+ * any exported sympols "EXPORT_SYMBOL()" statements.
+ * This is used to create proper -function and
+ * -nofunction arguments in calls to kernel-doc.
+ * Usage: docproc doc file.tmpl
+ *
+ * dependency-generator:
+ * Scans the template file and list all files
+ * referenced in a format recognized by make.
+ * Usage: docproc depend file.tmpl
+ * Writes dependency information to stdout
+ * in the following format:
+ * file.tmpl src.c src2.c
+ * The filenames are obtained from the following constructs:
+ * !Efilename
+ * !Ifilename
+ * !Dfilename
+ * !Ffilename
+ *
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <unistd.h>
+#include <limits.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+
+/* exitstatus is used to keep track of any failing calls to kernel-doc,
+ * but execution continues. */
+int exitstatus = 0;
+
+typedef void DFL(char *);
+DFL *defaultline;
+
+typedef void FILEONLY(char * file);
+FILEONLY *internalfunctions;
+FILEONLY *externalfunctions;
+FILEONLY *symbolsonly;
+
+typedef void FILELINE(char * file, char * line);
+FILELINE * singlefunctions;
+FILELINE * entity_system;
+
+#define MAXLINESZ 2048
+#define MAXFILES 250
+#define KERNELDOCPATH "scripts/"
+#define KERNELDOC "kernel-doc"
+#define DOCBOOK "-docbook"
+#define FUNCTION "-function"
+#define NOFUNCTION "-nofunction"
+
+void usage (void)
+{
+ fprintf(stderr, "Usage: docproc {doc|depend} file\n");
+ fprintf(stderr, "Input is read from file.tmpl. Output is sent to stdout\n");
+ fprintf(stderr, "doc: frontend when generating kernel documentation\n");
+ fprintf(stderr, "depend: generate list of files referenced within file\n");
+}
+
+/*
+ * Execute kernel-doc with parameters givin in svec
+ */
+void exec_kernel_doc(char **svec)
+{
+ pid_t pid;
+ int ret;
+ /* Make sure output generated so far are flushed */
+ fflush(stdout);
+ switch(pid=fork()) {
+ case -1:
+ perror("fork");
+ exit(1);
+ case 0:
+ execvp(KERNELDOCPATH KERNELDOC, svec);
+ perror("exec " KERNELDOCPATH KERNELDOC);
+ exit(1);
+ default:
+ waitpid(pid, &ret ,0);
+ }
+ if (WIFEXITED(ret))
+ exitstatus |= WEXITSTATUS(ret);
+ else
+ exitstatus = 0xff;
+}
+
+/* Types used to create list of all exported symbols in a number of files */
+struct symbols
+{
+ char *name;
+};
+
+struct symfile
+{
+ char *filename;
+ struct symbols *symbollist;
+ int symbolcnt;
+};
+
+struct symfile symfilelist[MAXFILES];
+int symfilecnt = 0;
+
+void add_new_symbol(struct symfile *sym, char * symname)
+{
+ sym->symbollist =
+ realloc(sym->symbollist, (sym->symbolcnt + 1) * sizeof(char *));
+ sym->symbollist[sym->symbolcnt++].name = strdup(symname);
+}
+
+/* Add a filename to the list */
+struct symfile * add_new_file(char * filename)
+{
+ symfilelist[symfilecnt++].filename = strdup(filename);
+ return &symfilelist[symfilecnt - 1];
+}
+/* Check if file already are present in the list */
+struct symfile * filename_exist(char * filename)
+{
+ int i;
+ for (i=0; i < symfilecnt; i++)
+ if (strcmp(symfilelist[i].filename, filename) == 0)
+ return &symfilelist[i];
+ return NULL;
+}
+
+/*
+ * List all files referenced within the template file.
+ * Files are separated by tabs.
+ */
+void adddep(char * file) { printf("\t%s", file); }
+void adddep2(char * file, char * line) { line = line; adddep(file); }
+void noaction(char * line) { line = line; }
+void noaction2(char * file, char * line) { file = file; line = line; }
+
+/* Echo the line without further action */
+void printline(char * line) { printf("%s", line); }
+
+/*
+ * Find all symbols exported with EXPORT_SYMBOL and EXPORT_SYMBOL_GPL
+ * in filename.
+ * All symbols located are stored in symfilelist.
+ */
+void find_export_symbols(char * filename)
+{
+ FILE * fp;
+ struct symfile *sym;
+ char line[MAXLINESZ];
+ if (filename_exist(filename) == NULL) {
+ sym = add_new_file(filename);
+ fp = fopen(filename, "r");
+ if (fp == NULL)
+ {
+ fprintf(stderr, "docproc: ");
+ perror(filename);
+ }
+ while(fgets(line, MAXLINESZ, fp)) {
+ char *p;
+ char *e;
+ if (((p = strstr(line, "EXPORT_SYMBOL_GPL")) != 0) ||
+ ((p = strstr(line, "EXPORT_SYMBOL")) != 0)) {
+ /* Skip EXPORT_SYMBOL{_GPL} */
+ while (isalnum(*p) || *p == '_')
+ p++;
+ /* Remove paranteses and additional ws */
+ while (isspace(*p))
+ p++;
+ if (*p != '(')
+ continue; /* Syntax error? */
+ else
+ p++;
+ while (isspace(*p))
+ p++;
+ e = p;
+ while (isalnum(*e) || *e == '_')
+ e++;
+ *e = '\0';
+ add_new_symbol(sym, p);
+ }
+ }
+ fclose(fp);
+ }
+}
+
+/*
+ * Document all external or internal functions in a file.
+ * Call kernel-doc with following parameters:
+ * kernel-doc -docbook -nofunction function_name1 filename
+ * function names are obtained from all the the src files
+ * by find_export_symbols.
+ * intfunc uses -nofunction
+ * extfunc uses -function
+ */
+void docfunctions(char * filename, char * type)
+{
+ int i,j;
+ int symcnt = 0;
+ int idx = 0;
+ char **vec;
+
+ for (i=0; i <= symfilecnt; i++)
+ symcnt += symfilelist[i].symbolcnt;
+ vec = malloc((2 + 2 * symcnt + 2) * sizeof(char*));
+ if (vec == NULL) {
+ perror("docproc: ");
+ exit(1);
+ }
+ vec[idx++] = KERNELDOC;
+ vec[idx++] = DOCBOOK;
+ for (i=0; i < symfilecnt; i++) {
+ struct symfile * sym = &symfilelist[i];
+ for (j=0; j < sym->symbolcnt; j++) {
+ vec[idx++] = type;
+ vec[idx++] = sym->symbollist[j].name;
+ }
+ }
+ vec[idx++] = filename;
+ vec[idx] = NULL;
+ printf("<!-- %s -->\n", filename);
+ exec_kernel_doc(vec);
+ fflush(stdout);
+ free(vec);
+}
+void intfunc(char * filename) { docfunctions(filename, NOFUNCTION); }
+void extfunc(char * filename) { docfunctions(filename, FUNCTION); }
+
+/*
+ * Document sp�ecific function(s) in a file.
+ * Call kernel-doc with the following parameters:
+ * kernel-doc -docbook -function function1 [-function function2]
+ */
+void singfunc(char * filename, char * line)
+{
+ char *vec[200]; /* Enough for specific functions */
+ int i, idx = 0;
+ int startofsym = 1;
+ vec[idx++] = KERNELDOC;
+ vec[idx++] = DOCBOOK;
+
+ /* Split line up in individual parameters preceeded by FUNCTION */
+ for (i=0; line[i]; i++) {
+ if (isspace(line[i])) {
+ line[i] = '\0';
+ startofsym = 1;
+ continue;
+ }
+ if (startofsym) {
+ startofsym = 0;
+ vec[idx++] = FUNCTION;
+ vec[idx++] = &line[i];
+ }
+ }
+ vec[idx++] = filename;
+ vec[idx] = NULL;
+ exec_kernel_doc(vec);
+}
+
+/*
+ * Parse file, calling action specific functions for:
+ * 1) Lines containing !E
+ * 2) Lines containing !I
+ * 3) Lines containing !D
+ * 4) Lines containing !F
+ * 5) Default lines - lines not matching the above
+ */
+void parse_file(FILE *infile)
+{
+ char line[MAXLINESZ];
+ char * s;
+ while(fgets(line, MAXLINESZ, infile)) {
+ if (line[0] == '!') {
+ s = line + 2;
+ switch (line[1]) {
+ case 'E':
+ while (*s && !isspace(*s)) s++;
+ *s = '\0';
+ externalfunctions(line+2);
+ break;
+ case 'I':
+ while (*s && !isspace(*s)) s++;
+ *s = '\0';
+ internalfunctions(line+2);
+ break;
+ case 'D':
+ while (*s && !isspace(*s)) s++;
+ *s = '\0';
+ symbolsonly(line+2);
+ break;
+ case 'F':
+ /* filename */
+ while (*s && !isspace(*s)) s++;
+ *s++ = '\0';
+ /* function names */
+ while (isspace(*s))
+ s++;
+ singlefunctions(line +2, s);
+ break;
+ default:
+ defaultline(line);
+ }
+ }
+ else {
+ defaultline(line);
+ }
+ }
+ fflush(stdout);
+}
+
+
+int main(int argc, char *argv[])
+{
+ FILE * infile;
+ if (argc != 3) {
+ usage();
+ exit(1);
+ }
+ /* Open file, exit on error */
+ infile = fopen(argv[2], "r");
+ if (infile == NULL) {
+ fprintf(stderr, "docproc: ");
+ perror(argv[2]);
+ exit(2);
+ }
+
+ if (strcmp("doc", argv[1]) == 0)
+ {
+ /* Need to do this in two passes.
+ * First pass is used to collect all symbols exported
+ * in the various files.
+ * Second pass generate the documentation.
+ * This is required because function are declared
+ * and exported in different files :-((
+ */
+ /* Collect symbols */
+ defaultline = noaction;
+ internalfunctions = find_export_symbols;
+ externalfunctions = find_export_symbols;
+ symbolsonly = find_export_symbols;
+ singlefunctions = noaction2;
+ parse_file(infile);
+
+ /* Rewind to start from beginning of file again */
+ fseek(infile, 0, SEEK_SET);
+ defaultline = printline;
+ internalfunctions = intfunc;
+ externalfunctions = extfunc;
+ symbolsonly = printline;
+ singlefunctions = singfunc;
+
+ parse_file(infile);
+ }
+ else if (strcmp("depend", argv[1]) == 0)
+ {
+ /* Create first part of dependency chain
+ * file.tmpl */
+ printf("%s\t", argv[2]);
+ defaultline = noaction;
+ internalfunctions = adddep;
+ externalfunctions = adddep;
+ symbolsonly = adddep;
+ singlefunctions = adddep2;
+ parse_file(infile);
+ printf("\n");
+ }
+ else
+ {
+ fprintf(stderr, "Unknown option: %s\n", argv[1]);
+ exit(1);
+ }
+ fclose(infile);
+ fflush(stdout);
+ return exitstatus;
+}
+
--- diff/scripts/basic/fixdep.c 1970-01-01 00:00:00.000000000 +0000
+++ source/scripts/basic/fixdep.c 2004-03-16 11:27:48.712782192 +0000
@@ -0,0 +1,381 @@
+/*
+ * "Optimize" a list of dependencies as spit out by gcc -MD
+ * for the kernel build
+ * ===========================================================================
+ *
+ * Author Kai Germaschewski
+ * Copyright 2002 by Kai Germaschewski <kai.germaschewski@gmx.de>
+ *
+ * This software may be used and distributed according to the terms
+ * of the GNU General Public License, incorporated herein by reference.
+ *
+ *
+ * Introduction:
+ *
+ * gcc produces a very nice and correct list of dependencies which
+ * tells make when to remake a file.
+ *
+ * To use this list as-is however has the drawback that virtually
+ * every file in the kernel includes <linux/config.h> which then again
+ * includes <linux/autoconf.h>
+ *
+ * If the user re-runs make *config, linux/autoconf.h will be
+ * regenerated. make notices that and will rebuild every file which
+ * includes autoconf.h, i.e. basically all files. This is extremely
+ * annoying if the user just changed CONFIG_HIS_DRIVER from n to m.
+ *
+ * So we play the same trick that "mkdep" played before. We replace
+ * the dependency on linux/autoconf.h by a dependency on every config
+ * option which is mentioned in any of the listed prequisites.
+ *
+ * To be exact, split-include populates a tree in include/config/,
+ * e.g. include/config/his/driver.h, which contains the #define/#undef
+ * for the CONFIG_HIS_DRIVER option.
+ *
+ * So if the user changes his CONFIG_HIS_DRIVER option, only the objects
+ * which depend on "include/linux/config/his/driver.h" will be rebuilt,
+ * so most likely only his driver ;-)
+ *
+ * The idea above dates, by the way, back to Michael E Chastain, AFAIK.
+ *
+ * So to get dependencies right, there two issues:
+ * o if any of the files the compiler read changed, we need to rebuild
+ * o if the command line given to the compile the file changed, we
+ * better rebuild as well.
+ *
+ * The former is handled by using the -MD output, the later by saving
+ * the command line used to compile the old object and comparing it
+ * to the one we would now use.
+ *
+ * Again, also this idea is pretty old and has been discussed on
+ * kbuild-devel a long time ago. I don't have a sensibly working
+ * internet connection right now, so I rather don't mention names
+ * without double checking.
+ *
+ * This code here has been based partially based on mkdep.c, which
+ * says the following about its history:
+ *
+ * Copyright abandoned, Michael Chastain, <mailto:mec@shout.net>.
+ * This is a C version of syncdep.pl by Werner Almesberger.
+ *
+ *
+ * It is invoked as
+ *
+ * fixdep <depfile> <target> <cmdline>
+ *
+ * and will read the dependency file <depfile>
+ *
+ * The transformed dependency snipped is written to stdout.
+ *
+ * It first generates a line
+ *
+ * cmd_<target> = <cmdline>
+ *
+ * and then basically copies the .<target>.d file to stdout, in the
+ * process filtering out the dependency on linux/autoconf.h and adding
+ * dependencies on include/config/my/option.h for every
+ * CONFIG_MY_OPTION encountered in any of the prequisites.
+ *
+ * It will also filter out all the dependencies on *.ver. We need
+ * to make sure that the generated version checksum are globally up
+ * to date before even starting the recursive build, so it's too late
+ * at this point anyway.
+ *
+ * The algorithm to grep for "CONFIG_..." is bit unusual, but should
+ * be fast ;-) We don't even try to really parse the header files, but
+ * merely grep, i.e. if CONFIG_FOO is mentioned in a comment, it will
+ * be picked up as well. It's not a problem with respect to
+ * correctness, since that can only give too many dependencies, thus
+ * we cannot miss a rebuild. Since people tend to not mention totally
+ * unrelated CONFIG_ options all over the place, it's not an
+ * efficiency problem either.
+ *
+ * (Note: it'd be easy to port over the complete mkdep state machine,
+ * but I don't think the added complexity is worth it)
+ */
+
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/mman.h>
+#include <unistd.h>
+#include <fcntl.h>
+#include <string.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <limits.h>
+#include <ctype.h>
+#include <netinet/in.h>
+
+#define INT_CONF ntohl(0x434f4e46)
+#define INT_ONFI ntohl(0x4f4e4649)
+#define INT_NFIG ntohl(0x4e464947)
+#define INT_FIG_ ntohl(0x4649475f)
+
+char *target;
+char *depfile;
+char *cmdline;
+
+void usage(void)
+
+{
+ fprintf(stderr, "Usage: fixdep <depfile> <target> <cmdline>\n");
+ exit(1);
+}
+
+void print_cmdline(void)
+{
+ printf("cmd_%s := %s\n\n", target, cmdline);
+}
+
+char * str_config = NULL;
+int size_config = 0;
+int len_config = 0;
+
+/*
+ * Grow the configuration string to a desired length.
+ * Usually the first growth is plenty.
+ */
+void grow_config(int len)
+{
+ while (len_config + len > size_config) {
+ if (size_config == 0)
+ size_config = 2048;
+ str_config = realloc(str_config, size_config *= 2);
+ if (str_config == NULL)
+ { perror("fixdep:malloc"); exit(1); }
+ }
+}
+
+
+
+/*
+ * Lookup a value in the configuration string.
+ */
+int is_defined_config(const char * name, int len)
+{
+ const char * pconfig;
+ const char * plast = str_config + len_config - len;
+ for ( pconfig = str_config + 1; pconfig < plast; pconfig++ ) {
+ if (pconfig[ -1] == '\n'
+ && pconfig[len] == '\n'
+ && !memcmp(pconfig, name, len))
+ return 1;
+ }
+ return 0;
+}
+
+/*
+ * Add a new value to the configuration string.
+ */
+void define_config(const char * name, int len)
+{
+ grow_config(len + 1);
+
+ memcpy(str_config+len_config, name, len);
+ len_config += len;
+ str_config[len_config++] = '\n';
+}
+
+/*
+ * Clear the set of configuration strings.
+ */
+void clear_config(void)
+{
+ len_config = 0;
+ define_config("", 0);
+}
+
+/*
+ * Record the use of a CONFIG_* word.
+ */
+void use_config(char *m, int slen)
+{
+ char s[PATH_MAX];
+ char *p;
+
+ if (is_defined_config(m, slen))
+ return;
+
+ define_config(m, slen);
+
+ memcpy(s, m, slen); s[slen] = 0;
+
+ for (p = s; p < s + slen; p++) {
+ if (*p == '_')
+ *p = '/';
+ else
+ *p = tolower((unsigned char)*p);
+ }
+ printf(" $(wildcard include/config/%s.h) \\\n", s);
+}
+
+void parse_config_file(char *map, size_t len)
+{
+ int *end = (int *) (map + len);
+ /* start at +1, so that p can never be < map */
+ int *m = (int *) map + 1;
+ char *p, *q;
+
+ for (; m < end; m++) {
+ if (*m == INT_CONF) { p = (char *) m ; goto conf; }
+ if (*m == INT_ONFI) { p = (char *) m-1; goto conf; }
+ if (*m == INT_NFIG) { p = (char *) m-2; goto conf; }
+ if (*m == INT_FIG_) { p = (char *) m-3; goto conf; }
+ continue;
+ conf:
+ if (p > map + len - 7)
+ continue;
+ if (memcmp(p, "CONFIG_", 7))
+ continue;
+ for (q = p + 7; q < map + len; q++) {
+ if (!(isalnum(*q) || *q == '_'))
+ goto found;
+ }
+ continue;
+
+ found:
+ use_config(p+7, q-p-7);
+ }
+}
+
+/* test is s ends in sub */
+int strrcmp(char *s, char *sub)
+{
+ int slen = strlen(s);
+ int sublen = strlen(sub);
+
+ if (sublen > slen)
+ return 1;
+
+ return memcmp(s + slen - sublen, sub, sublen);
+}
+
+void do_config_file(char *filename)
+{
+ struct stat st;
+ int fd;
+ void *map;
+
+ fd = open(filename, O_RDONLY);
+ if (fd < 0) {
+ fprintf(stderr, "fixdep: ");
+ perror(filename);
+ exit(2);
+ }
+ fstat(fd, &st);
+ if (st.st_size == 0) {
+ close(fd);
+ return;
+ }
+ map = mmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fd, 0);
+ if ((long) map == -1) {
+ perror("fixdep: mmap");
+ close(fd);
+ return;
+ }
+
+ parse_config_file(map, st.st_size);
+
+ munmap(map, st.st_size);
+
+ close(fd);
+}
+
+void parse_dep_file(void *map, size_t len)
+{
+ char *m = map;
+ char *end = m + len;
+ char *p;
+ char s[PATH_MAX];
+
+ p = strchr(m, ':');
+ if (!p) {
+ fprintf(stderr, "fixdep: parse error\n");
+ exit(1);
+ }
+ memcpy(s, m, p-m); s[p-m] = 0;
+ printf("deps_%s := \\\n", target);
+ m = p+1;
+
+ clear_config();
+
+ while (m < end) {
+ while (m < end && (*m == ' ' || *m == '\\' || *m == '\n'))
+ m++;
+ p = m;
+ while (p < end && *p != ' ') p++;
+ if (p == end) {
+ do p--; while (!isalnum(*p));
+ p++;
+ }
+ memcpy(s, m, p-m); s[p-m] = 0;
+ if (strrcmp(s, "include/linux/autoconf.h") &&
+ strrcmp(s, ".ver")) {
+ printf(" %s \\\n", s);
+ do_config_file(s);
+ }
+ m = p + 1;
+ }
+ printf("\n%s: $(deps_%s)\n\n", target, target);
+ printf("$(deps_%s):\n", target);
+}
+
+void print_deps(void)
+{
+ struct stat st;
+ int fd;
+ void *map;
+
+ fd = open(depfile, O_RDONLY);
+ if (fd < 0) {
+ fprintf(stderr, "fixdep: ");
+ perror(depfile);
+ exit(2);
+ }
+ fstat(fd, &st);
+ if (st.st_size == 0) {
+ fprintf(stderr,"fixdep: %s is empty\n",depfile);
+ close(fd);
+ return;
+ }
+ map = mmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fd, 0);
+ if ((long) map == -1) {
+ perror("fixdep: mmap");
+ close(fd);
+ return;
+ }
+
+ parse_dep_file(map, st.st_size);
+
+ munmap(map, st.st_size);
+
+ close(fd);
+}
+
+void traps(void)
+{
+ static char test[] __attribute__((aligned(sizeof(int)))) = "CONF";
+
+ if (*(int *)test != INT_CONF) {
+ fprintf(stderr, "fixdep: sizeof(int) != 4 or wrong endianess? %#x\n",
+ *(int *)test);
+ exit(2);
+ }
+}
+
+int main(int argc, char *argv[])
+{
+ traps();
+
+ if (argc != 4)
+ usage();
+
+ depfile = argv[1];
+ target = argv[2];
+ cmdline = argv[3];
+
+ print_cmdline();
+ print_deps();
+
+ return 0;
+}
--- diff/scripts/basic/split-include.c 1970-01-01 00:00:00.000000000 +0000
+++ source/scripts/basic/split-include.c 2004-03-16 11:27:48.739778088 +0000
@@ -0,0 +1,226 @@
+/*
+ * split-include.c
+ *
+ * Copyright abandoned, Michael Chastain, <mailto:mec@shout.net>.
+ * This is a C version of syncdep.pl by Werner Almesberger.
+ *
+ * This program takes autoconf.h as input and outputs a directory full
+ * of one-line include files, merging onto the old values.
+ *
+ * Think of the configuration options as key-value pairs. Then there
+ * are five cases:
+ *
+ * key old value new value action
+ *
+ * KEY-1 VALUE-1 VALUE-1 leave file alone
+ * KEY-2 VALUE-2A VALUE-2B write VALUE-2B into file
+ * KEY-3 - VALUE-3 write VALUE-3 into file
+ * KEY-4 VALUE-4 - write an empty file
+ * KEY-5 (empty) - leave old empty file alone
+ */
+
+#include <sys/stat.h>
+#include <sys/types.h>
+
+#include <ctype.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+
+#define ERROR_EXIT(strExit) \
+ { \
+ const int errnoSave = errno; \
+ fprintf(stderr, "%s: ", str_my_name); \
+ errno = errnoSave; \
+ perror((strExit)); \
+ exit(1); \
+ }
+
+
+
+int main(int argc, const char * argv [])
+{
+ const char * str_my_name;
+ const char * str_file_autoconf;
+ const char * str_dir_config;
+
+ FILE * fp_config;
+ FILE * fp_target;
+ FILE * fp_find;
+
+ int buffer_size;
+
+ char * line;
+ char * old_line;
+ char * list_target;
+ char * ptarget;
+
+ struct stat stat_buf;
+
+ /* Check arg count. */
+ if (argc != 3)
+ {
+ fprintf(stderr, "%s: wrong number of arguments.\n", argv[0]);
+ exit(1);
+ }
+
+ str_my_name = argv[0];
+ str_file_autoconf = argv[1];
+ str_dir_config = argv[2];
+
+ /* Find a buffer size. */
+ if (stat(str_file_autoconf, &stat_buf) != 0)
+ ERROR_EXIT(str_file_autoconf);
+ buffer_size = 2 * stat_buf.st_size + 4096;
+
+ /* Allocate buffers. */
+ if ( (line = malloc(buffer_size)) == NULL
+ || (old_line = malloc(buffer_size)) == NULL
+ || (list_target = malloc(buffer_size)) == NULL )
+ ERROR_EXIT(str_file_autoconf);
+
+ /* Open autoconfig file. */
+ if ((fp_config = fopen(str_file_autoconf, "r")) == NULL)
+ ERROR_EXIT(str_file_autoconf);
+
+ /* Make output directory if needed. */
+ if (stat(str_dir_config, &stat_buf) != 0)
+ {
+ if (mkdir(str_dir_config, 0755) != 0)
+ ERROR_EXIT(str_dir_config);
+ }
+
+ /* Change to output directory. */
+ if (chdir(str_dir_config) != 0)
+ ERROR_EXIT(str_dir_config);
+
+ /* Put initial separator into target list. */
+ ptarget = list_target;
+ *ptarget++ = '\n';
+
+ /* Read config lines. */
+ while (fgets(line, buffer_size, fp_config))
+ {
+ const char * str_config;
+ int is_same;
+ int itarget;
+
+ if (line[0] != '#')
+ continue;
+ if ((str_config = strstr(line, "CONFIG_")) == NULL)
+ continue;
+
+ /* Make the output file name. */
+ str_config += sizeof("CONFIG_") - 1;
+ for (itarget = 0; !isspace(str_config[itarget]); itarget++)
+ {
+ int c = (unsigned char) str_config[itarget];
+ if (isupper(c)) c = tolower(c);
+ if (c == '_') c = '/';
+ ptarget[itarget] = c;
+ }
+ ptarget[itarget++] = '.';
+ ptarget[itarget++] = 'h';
+ ptarget[itarget++] = '\0';
+
+ /* Check for existing file. */
+ is_same = 0;
+ if ((fp_target = fopen(ptarget, "r")) != NULL)
+ {
+ fgets(old_line, buffer_size, fp_target);
+ if (fclose(fp_target) != 0)
+ ERROR_EXIT(ptarget);
+ if (!strcmp(line, old_line))
+ is_same = 1;
+ }
+
+ if (!is_same)
+ {
+ /* Auto-create directories. */
+ int islash;
+ for (islash = 0; islash < itarget; islash++)
+ {
+ if (ptarget[islash] == '/')
+ {
+ ptarget[islash] = '\0';
+ if (stat(ptarget, &stat_buf) != 0
+ && mkdir(ptarget, 0755) != 0)
+ ERROR_EXIT( ptarget );
+ ptarget[islash] = '/';
+ }
+ }
+
+ /* Write the file. */
+ if ((fp_target = fopen(ptarget, "w" )) == NULL)
+ ERROR_EXIT(ptarget);
+ fputs(line, fp_target);
+ if (ferror(fp_target) || fclose(fp_target) != 0)
+ ERROR_EXIT(ptarget);
+ }
+
+ /* Update target list */
+ ptarget += itarget;
+ *(ptarget-1) = '\n';
+ }
+
+ /*
+ * Close autoconfig file.
+ * Terminate the target list.
+ */
+ if (fclose(fp_config) != 0)
+ ERROR_EXIT(str_file_autoconf);
+ *ptarget = '\0';
+
+ /*
+ * Fix up existing files which have no new value.
+ * This is Case 4 and Case 5.
+ *
+ * I re-read the tree and filter it against list_target.
+ * This is crude. But it avoids data copies. Also, list_target
+ * is compact and contiguous, so it easily fits into cache.
+ *
+ * Notice that list_target contains strings separated by \n,
+ * with a \n before the first string and after the last.
+ * fgets gives the incoming names a terminating \n.
+ * So by having an initial \n, strstr will find exact matches.
+ */
+
+ fp_find = popen("find * -type f -name \"*.h\" -print", "r");
+ if (fp_find == 0)
+ ERROR_EXIT( "find" );
+
+ line[0] = '\n';
+ while (fgets(line+1, buffer_size, fp_find))
+ {
+ if (strstr(list_target, line) == NULL)
+ {
+ /*
+ * This is an old file with no CONFIG_* flag in autoconf.h.
+ */
+
+ /* First strip the \n. */
+ line[strlen(line)-1] = '\0';
+
+ /* Grab size. */
+ if (stat(line+1, &stat_buf) != 0)
+ ERROR_EXIT(line);
+
+ /* If file is not empty, make it empty and give it a fresh date. */
+ if (stat_buf.st_size != 0)
+ {
+ if ((fp_target = fopen(line+1, "w")) == NULL)
+ ERROR_EXIT(line);
+ if (fclose(fp_target) != 0)
+ ERROR_EXIT(line);
+ }
+ }
+ }
+
+ if (pclose(fp_find) != 0)
+ ERROR_EXIT("find");
+
+ return 0;
+}
--- diff/security/selinux/include/conditional.h 1970-01-01 00:00:00.000000000 +0000
+++ source/security/selinux/include/conditional.h 2004-03-16 11:27:48.752776112 +0000
@@ -0,0 +1,22 @@
+/*
+ * Interface to booleans in the security server. This is exported
+ * for the selinuxfs.
+ *
+ * Author: Karl MacMillan <kmacmillan@tresys.com>
+ *
+ * Copyright (C) 2003 - 2004 Tresys Technology, LLC
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, version 2.
+ */
+
+#ifndef _SELINUX_CONDITIONAL_H_
+#define _SELINUX_CONDITIONAL_H_
+
+int security_get_bools(int *len, char ***names, int **values);
+
+int security_set_bools(int len, int *values);
+
+int security_get_bool_value(int bool);
+
+#endif
--- diff/security/selinux/ss/conditional.c 1970-01-01 00:00:00.000000000 +0000
+++ source/security/selinux/ss/conditional.c 2004-03-16 11:27:48.759775048 +0000
@@ -0,0 +1,487 @@
+/* Authors: Karl MacMillan <kmacmillan@tresys.com>
+ * Frank Mayer <mayerf@tresys.com>
+ *
+ * Copyright (C) 2003 - 2004 Tresys Technology, LLC
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, version 2.
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/spinlock.h>
+#include <asm/semaphore.h>
+#include <linux/slab.h>
+
+#include "security.h"
+#include "conditional.h"
+
+/*
+ * cond_evaluate_expr evaluates a conditional expr
+ * in reverse polish notation. It returns true (1), false (0),
+ * or undefined (-1). Undefined occurs when the expression
+ * exceeds the stack depth of COND_EXPR_MAXDEPTH.
+ */
+static int cond_evaluate_expr(struct policydb *p, struct cond_expr *expr)
+{
+
+ struct cond_expr *cur;
+ int s[COND_EXPR_MAXDEPTH];
+ int sp = -1;
+
+ for (cur = expr; cur != NULL; cur = cur->next) {
+ switch (cur->expr_type) {
+ case COND_BOOL:
+ if (sp == (COND_EXPR_MAXDEPTH - 1))
+ return -1;
+ sp++;
+ s[sp] = p->bool_val_to_struct[cur->bool - 1]->state;
+ break;
+ case COND_NOT:
+ if (sp < 0)
+ return -1;
+ s[sp] = !s[sp];
+ break;
+ case COND_OR:
+ if (sp < 1)
+ return -1;
+ sp--;
+ s[sp] |= s[sp + 1];
+ break;
+ case COND_AND:
+ if (sp < 1)
+ return -1;
+ sp--;
+ s[sp] &= s[sp + 1];
+ break;
+ case COND_XOR:
+ if (sp < 1)
+ return -1;
+ sp--;
+ s[sp] ^= s[sp + 1];
+ break;
+ case COND_EQ:
+ if (sp < 1)
+ return -1;
+ sp--;
+ s[sp] = (s[sp] == s[sp + 1]);
+ break;
+ case COND_NEQ:
+ if (sp < 1)
+ return -1;
+ sp--;
+ s[sp] = (s[sp] != s[sp + 1]);
+ break;
+ default:
+ return -1;
+ }
+ }
+ return s[0];
+}
+
+/*
+ * evaluate_cond_node evaluates the conditional stored in
+ * a struct cond_node and if the result is different than the
+ * current state of the node it sets the rules in the true/false
+ * list appropriately. If the result of the expression is undefined
+ * all of the rules are disabled for safety.
+ */
+int evaluate_cond_node(struct policydb *p, struct cond_node *node)
+{
+ int new_state;
+ struct cond_av_list* cur;
+
+ new_state = cond_evaluate_expr(p, node->expr);
+ if (new_state != node->cur_state) {
+ node->cur_state = new_state;
+ if (new_state == -1)
+ printk(KERN_ERR "security: expression result was undefined - disabling all rules.\n");
+ /* turn the rules on or off */
+ for (cur = node->true_list; cur != NULL; cur = cur->next) {
+ if (new_state <= 0) {
+ cur->node->datum.specified &= ~AVTAB_ENABLED;
+ } else {
+ cur->node->datum.specified |= AVTAB_ENABLED;
+ }
+ }
+
+ for (cur = node->false_list; cur != NULL; cur = cur->next) {
+ /* -1 or 1 */
+ if (new_state) {
+ cur->node->datum.specified &= ~AVTAB_ENABLED;
+ } else {
+ cur->node->datum.specified |= AVTAB_ENABLED;
+ }
+ }
+ }
+ return 0;
+}
+
+int cond_policydb_init(struct policydb *p)
+{
+ p->bool_val_to_struct = NULL;
+ p->cond_list = NULL;
+ if (avtab_init(&p->te_cond_avtab))
+ return -1;
+
+ return 0;
+}
+
+static void cond_av_list_destroy(struct cond_av_list *list)
+{
+ struct cond_av_list *cur, *next;
+ for (cur = list; cur != NULL; cur = next) {
+ next = cur->next;
+ /* the avtab_ptr_t node is destroy by the avtab */
+ kfree(cur);
+ }
+}
+
+static void cond_node_destroy(struct cond_node *node)
+{
+ struct cond_expr *cur_expr, *next_expr;
+
+ for (cur_expr = node->expr; cur_expr != NULL; cur_expr = next_expr) {
+ next_expr = cur_expr->next;
+ kfree(cur_expr);
+ }
+ cond_av_list_destroy(node->true_list);
+ cond_av_list_destroy(node->false_list);
+ kfree(node);
+}
+
+static void cond_list_destroy(struct cond_node *list)
+{
+ struct cond_node *next, *cur;
+
+ if (list == NULL)
+ return;
+
+ for (cur = list; cur != NULL; cur = next) {
+ next = cur->next;
+ cond_node_destroy(cur);
+ }
+}
+
+void cond_policydb_destroy(struct policydb *p)
+{
+ if (p->bool_val_to_struct != NULL)
+ kfree(p->bool_val_to_struct);
+ avtab_destroy(&p->te_cond_avtab);
+ cond_list_destroy(p->cond_list);
+}
+
+int cond_init_bool_indexes(struct policydb *p)
+{
+ if (p->bool_val_to_struct)
+ kfree(p->bool_val_to_struct);
+ p->bool_val_to_struct = (struct cond_bool_datum**)
+ kmalloc(p->p_bools.nprim * sizeof(struct cond_bool_datum*), GFP_KERNEL);
+ if (!p->bool_val_to_struct)
+ return -1;
+ return 0;
+}
+
+int cond_destroy_bool(void *key, void *datum, void *p)
+{
+ if (key)
+ kfree(key);
+ kfree(datum);
+ return 0;
+}
+
+int cond_index_bool(void *key, void *datum, void *datap)
+{
+ struct policydb *p;
+ struct cond_bool_datum *booldatum;
+
+ booldatum = datum;
+ p = datap;
+
+ if (!booldatum->value || booldatum->value > p->p_bools.nprim)
+ return -EINVAL;
+
+ p->p_bool_val_to_name[booldatum->value - 1] = key;
+ p->bool_val_to_struct[booldatum->value -1] = booldatum;
+
+ return 0;
+}
+
+int bool_isvalid(struct cond_bool_datum *b)
+{
+ if (!(b->state == 0 || b->state == 1))
+ return 0;
+ return 1;
+}
+
+int cond_read_bool(struct policydb *p, struct hashtab *h, void *fp)
+{
+ char *key = 0;
+ struct cond_bool_datum *booldatum;
+ __u32 *buf, len;
+
+ booldatum = kmalloc(sizeof(struct cond_bool_datum), GFP_KERNEL);
+ if (!booldatum)
+ return -1;
+ memset(booldatum, 0, sizeof(struct cond_bool_datum));
+
+ buf = next_entry(fp, sizeof(__u32) * 3);
+ if (!buf)
+ goto err;
+
+ booldatum->value = le32_to_cpu(buf[0]);
+ booldatum->state = le32_to_cpu(buf[1]);
+
+ if (!bool_isvalid(booldatum))
+ goto err;
+
+ len = le32_to_cpu(buf[2]);
+
+ buf = next_entry(fp, len);
+ if (!buf)
+ goto err;
+ key = kmalloc(len + 1, GFP_KERNEL);
+ if (!key)
+ goto err;
+ memcpy(key, buf, len);
+ key[len] = 0;
+ if (hashtab_insert(h, key, booldatum))
+ goto err;
+
+ return 0;
+err:
+ cond_destroy_bool(key, booldatum, 0);
+ return -1;
+}
+
+static int cond_read_av_list(struct policydb *p, void *fp, struct cond_av_list **ret_list,
+ struct cond_av_list *other)
+{
+ struct cond_av_list *list, *last = NULL, *cur;
+ struct avtab_key key;
+ struct avtab_datum datum;
+ struct avtab_node *node_ptr;
+ int len, i;
+ __u32 *buf;
+ __u8 found;
+
+ *ret_list = NULL;
+
+ len = 0;
+ buf = next_entry(fp, sizeof(__u32));
+ if (!buf)
+ return -1;
+
+ len = le32_to_cpu(buf[0]);
+ if (len == 0) {
+ return 0;
+ }
+
+ for (i = 0; i < len; i++) {
+ if (avtab_read_item(fp, &datum, &key))
+ goto err;
+
+ /*
+ * For type rules we have to make certain there aren't any
+ * conflicting rules by searching the te_avtab and the
+ * cond_te_avtab.
+ */
+ if (datum.specified & AVTAB_TYPE) {
+ if (avtab_search(&p->te_avtab, &key, AVTAB_TYPE)) {
+ printk("security: type rule already exists outside of a conditional.");
+ goto err;
+ }
+ /*
+ * If we are reading the false list other will be a pointer to
+ * the true list. We can have duplicate entries if there is only
+ * 1 other entry and it is in our true list.
+ *
+ * If we are reading the true list (other == NULL) there shouldn't
+ * be any other entries.
+ */
+ if (other) {
+ node_ptr = avtab_search_node(&p->te_cond_avtab, &key, AVTAB_TYPE);
+ if (node_ptr) {
+ if (avtab_search_node_next(node_ptr, AVTAB_TYPE)) {
+ printk("security: too many conflicting type rules.");
+ goto err;
+ }
+ found = 0;
+ for (cur = other; cur != NULL; cur = cur->next) {
+ if (cur->node == node_ptr) {
+ found = 1;
+ break;
+ }
+ }
+ if (!found) {
+ printk("security: conflicting type rules.");
+ goto err;
+ }
+ }
+ } else {
+ if (avtab_search(&p->te_cond_avtab, &key, AVTAB_TYPE)) {
+ printk("security: conflicting type rules when adding type rule for true.");
+ goto err;
+ }
+ }
+ }
+ node_ptr = avtab_insert_nonunique(&p->te_cond_avtab, &key, &datum);
+ if (!node_ptr) {
+ printk("security: could not insert rule.");
+ goto err;
+ }
+
+ list = kmalloc(sizeof(struct cond_av_list), GFP_KERNEL);
+ if (!list)
+ goto err;
+ memset(list, 0, sizeof(struct cond_av_list));
+
+ list->node = node_ptr;
+ if (i == 0)
+ *ret_list = list;
+ else
+ last->next = list;
+ last = list;
+
+ }
+
+ return 0;
+err:
+ cond_av_list_destroy(*ret_list);
+ *ret_list = NULL;
+ return -1;
+}
+
+static int expr_isvalid(struct policydb *p, struct cond_expr *expr)
+{
+ if (expr->expr_type <= 0 || expr->expr_type > COND_LAST) {
+ printk("security: conditional expressions uses unknown operator.\n");
+ return 0;
+ }
+
+ if (expr->bool > p->p_bools.nprim) {
+ printk("security: conditional expressions uses unknown bool.\n");
+ return 0;
+ }
+ return 1;
+}
+
+static int cond_read_node(struct policydb *p, struct cond_node *node, void *fp)
+{
+ __u32 *buf;
+ int len, i;
+ struct cond_expr *expr = NULL, *last = NULL;
+
+ buf = next_entry(fp, sizeof(__u32));
+ if (!buf)
+ return -1;
+
+ node->cur_state = le32_to_cpu(buf[0]);
+
+ len = 0;
+ buf = next_entry(fp, sizeof(__u32));
+ if (!buf)
+ return -1;
+
+ /* expr */
+ len = le32_to_cpu(buf[0]);
+
+ for (i = 0; i < len; i++ ) {
+ buf = next_entry(fp, sizeof(__u32) * 2);
+ if (!buf)
+ goto err;
+
+ expr = kmalloc(sizeof(struct cond_expr), GFP_KERNEL);
+ if (!expr) {
+ goto err;
+ }
+ memset(expr, 0, sizeof(struct cond_expr));
+
+ expr->expr_type = le32_to_cpu(buf[0]);
+ expr->bool = le32_to_cpu(buf[1]);
+
+ if (!expr_isvalid(p, expr))
+ goto err;
+
+ if (i == 0) {
+ node->expr = expr;
+ } else {
+ last->next = expr;
+ }
+ last = expr;
+ }
+
+ if (cond_read_av_list(p, fp, &node->true_list, NULL) != 0)
+ goto err;
+ if (cond_read_av_list(p, fp, &node->false_list, node->true_list) != 0)
+ goto err;
+ return 0;
+err:
+ cond_node_destroy(node);
+ return -1;
+}
+
+int cond_read_list(struct policydb *p, void *fp)
+{
+ struct cond_node *node, *last = NULL;
+ __u32 *buf;
+ int i, len;
+
+ buf = next_entry(fp, sizeof(__u32));
+ if (!buf)
+ return -1;
+
+ len = le32_to_cpu(buf[0]);
+
+ for (i = 0; i < len; i++) {
+ node = kmalloc(sizeof(struct cond_node), GFP_KERNEL);
+ if (!node)
+ goto err;
+ memset(node, 0, sizeof(struct cond_node));
+
+ if (cond_read_node(p, node, fp) != 0)
+ goto err;
+
+ if (i == 0) {
+ p->cond_list = node;
+ } else {
+ last->next = node;
+ }
+ last = node;
+ }
+ return 0;
+err:
+ cond_list_destroy(p->cond_list);
+ return -1;
+}
+
+/* Determine whether additional permissions are granted by the conditional
+ * av table, and if so, add them to the result
+ */
+void cond_compute_av(struct avtab *ctab, struct avtab_key *key, struct av_decision *avd)
+{
+ struct avtab_node *node;
+
+ if(!ctab || !key || !avd)
+ return;
+
+ for(node = avtab_search_node(ctab, key, AVTAB_AV); node != NULL;
+ node = avtab_search_node_next(node, AVTAB_AV)) {
+ if ( (__u32) (AVTAB_ALLOWED|AVTAB_ENABLED) ==
+ (node->datum.specified & (AVTAB_ALLOWED|AVTAB_ENABLED)))
+ avd->allowed |= avtab_allowed(&node->datum);
+ if ( (__u32) (AVTAB_AUDITDENY|AVTAB_ENABLED) ==
+ (node->datum.specified & (AVTAB_AUDITDENY|AVTAB_ENABLED)))
+ /* Since a '0' in an auditdeny mask represents a
+ * permission we do NOT want to audit (dontaudit), we use
+ * the '&' operand to ensure that all '0's in the mask
+ * are retained (much unlike the allow and auditallow cases).
+ */
+ avd->auditdeny &= avtab_auditdeny(&node->datum);
+ if ( (__u32) (AVTAB_AUDITALLOW|AVTAB_ENABLED) ==
+ (node->datum.specified & (AVTAB_AUDITALLOW|AVTAB_ENABLED)))
+ avd->auditallow |= avtab_auditallow(&node->datum);
+ }
+ return;
+}
--- diff/security/selinux/ss/conditional.h 1970-01-01 00:00:00.000000000 +0000
+++ source/security/selinux/ss/conditional.h 2004-03-16 11:27:48.759775048 +0000
@@ -0,0 +1,77 @@
+/* Authors: Karl MacMillan <kmacmillan@tresys.com>
+ * Frank Mayer <mayerf@tresys.com>
+ *
+ * Copyright (C) 2003 - 2004 Tresys Technology, LLC
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, version 2.
+ */
+
+#ifndef _CONDITIONAL_H_
+#define _CONDITIONAL_H_
+
+#include "avtab.h"
+#include "symtab.h"
+#include "policydb.h"
+
+#define COND_EXPR_MAXDEPTH 10
+
+/*
+ * A conditional expression is a list of operators and operands
+ * in reverse polish notation.
+ */
+struct cond_expr {
+#define COND_BOOL 1 /* plain bool */
+#define COND_NOT 2 /* !bool */
+#define COND_OR 3 /* bool || bool */
+#define COND_AND 4 /* bool && bool */
+#define COND_XOR 5 /* bool ^ bool */
+#define COND_EQ 6 /* bool == bool */
+#define COND_NEQ 7 /* bool != bool */
+#define COND_LAST 8
+ __u32 expr_type;
+ __u32 bool;
+ struct cond_expr *next;
+};
+
+/*
+ * Each cond_node contains a list of rules to be enabled/disabled
+ * depending on the current value of the conditional expression. This
+ * struct is for that list.
+ */
+struct cond_av_list {
+ struct avtab_node *node;
+ struct cond_av_list *next;
+};
+
+/*
+ * A cond node represents a conditional block in a policy. It
+ * contains a conditional expression, the current state of the expression,
+ * two lists of rules to enable/disable depending on the value of the
+ * expression (the true list corresponds to if and the false list corresponds
+ * to else)..
+ */
+struct cond_node {
+ int cur_state;
+ struct cond_expr *expr;
+ struct cond_av_list *true_list;
+ struct cond_av_list *false_list;
+ struct cond_node *next;
+};
+
+int cond_policydb_init(struct policydb* p);
+void cond_policydb_destroy(struct policydb* p);
+
+int cond_init_bool_indexes(struct policydb* p);
+int cond_destroy_bool(void *key, void *datum, void *p);
+
+int cond_index_bool(void *key, void *datum, void *datap);
+
+int cond_read_bool(struct policydb *p, struct hashtab *h, void *fp);
+int cond_read_list(struct policydb *p, void *fp);
+
+void cond_compute_av(struct avtab *ctab, struct avtab_key *key, struct av_decision *avd);
+
+int evaluate_cond_node(struct policydb *p, struct cond_node *node);
+
+#endif /* _CONDITIONAL_H_ */
--- diff/Documentation/usb/brlvger.txt 2002-10-16 03:27:49.000000000 +0000
+++ source/Documentation/usb/brlvger.txt 1970-01-01 00:00:00.000000000 +0000
@@ -1,36 +0,0 @@
-Kernel Driver for the Tieman Voyager Braille Display (USB)
-
-Authors:
-St�phane Dalton <sdalton@videotron.ca>
-St�phane Doyon <s.doyon@videotron.ca>
-
-Version 0.8, April 17, 2002
-
-The brlvger driver supports a Braille display (aka Braille terminal)
-model Voyager from Tieman.
-
-The driver has been in heavy use for about six months now (as of April
-17th 2002) by a very few users (about 3-4), who say it has worked very
-well for them.
-
-We have tested it with a Voyager 44, but it should also support
-the Voyager 70.
-
-This driver implements a character device which allows userspace programs
-access to the braille displays raw functions. You still need a userspace
-program to perform the screen-review functions and control the
-display. Get BRLTTY from http://mielke.cc/brltty/ (version 2.99.8 or
-later). It has a Voyager driver which interfaces with this kernel driver.
-
-The interface is through a character device, major 180, minor 128, called
-"brlvger" under devfs.
-
-Many thanks to the Tieman people: Corand van Strien, Ivar Illing, Daphne
-Vogelaar and Ingrid Vogel. They provided us with a Braille display (as
-well as programming information) so that we could write this driver. They
-replaced the display when it broke and they answered our technical
-questions. It is very motivating when companies take an interest in such
-projects and are so supportive.
-
-Thanks to Andor Demarteau <ademarte@students.cs.uu.nl> who got this whole
-project started and beta-tested all our early buggy attempts.
--- diff/Documentation/usb/scanner.txt 2004-03-11 10:20:19.000000000 +0000
+++ source/Documentation/usb/scanner.txt 1970-01-01 00:00:00.000000000 +0000
@@ -1,336 +0,0 @@
-Copyright (C) 1999, 2000 David E. Nelson <dnelson@jump.net>
-Updated 2003 by Henning Meier-Geinitz <henning@meier-geinitz.de>
-
-
-OVERVIEW
-
-This README addresses issues regarding how to configure the kernel to access a
-USB scanner. Although the driver was originally conceived for USB HP
-scanners, it's general enough so that it can be used with most other USB
-scanners. Also, one can pass the USB Vendor and Product IDs using module
-parameters for unknown scanners.
-
-There are two drivers for SCSI-over-USB scanners:
-* The "hpusbscsi" module for Hewlett-Packard 53xx series, Hewlett-Packard 7400,
- Minolta Scan Dual II, Minolta Elite II
-* The "microtek" module for the Microtek Scanmaker X6
-
-In addition to the kernel driver, user-space tools like SANE are necessary to
-actually use the scanner. SANE ("Scanner Access Now Easy") provides drivers
-for a variety of USB scanners. See the appropriate SANE man page for details,
-e.g. man sane-usb and man sane-hp (for HP scanners).
-
-NOTE: Just because a product is detected by this driver does not mean that
-applications exist that support the product. It's in the hopes that this will
-allow developers a means to produce applications that will support the listed
-USB products.
-
-
-ADDITIONAL INFORMATION
-
-http://www.linux-usb.org/ (General information, mailing lists, links)
-http://www.mostang.com/sane/ (SANE user-space tools)
-http://www.meier-geinitz.de/kernel/ (USB scanner driver information and patches)
-
-
-REQUIREMENTS
-
-A host with a USB port. Ideally, either a UHCI (Intel), OHCI (Compaq and
-others) or EHCI hardware should work.
-
-Using "make menuconfig" or your preferred method for configuring the kernel,
-select "Support for USB", "OHCI HCD/UHCI HCD/EHCI HCD" depending on your
-hardware, "USB Scanner support", and "USB device filesystem". Compile and
-install the modules (you may need to execute "depmod -a" to update the module
-dependencies). If any of the USB sections were compiled into the kernel, a
-reboot is necessary. NOTE: Updating the boot disk with "lilo" may also be
-required. Testing was performed only as modules, YMMV.
-
-Up to 16 scanners can be connected/used simultaneously. If devfs support is
-enabled, see next section. Otherwise, the device files must be created
-manually if they don't exist yet, either by MAKEDEV or mknod.
-
-MAKEDEV method:
- cd /dev
- MAKEDEV usb
- Check that the device files "/dev/usb/scanner0" - "/dev/usb/scanner15" have
- been created.
-
-mknod method:
- mknod /dev/usb/scanner0 c 180 48
- mknod /dev/usb/scanner1 c 180 49
- .
- .
- mknod /dev/usb/scanner15 c 180 63
-
-Set appropriate permissions for /dev/usb/scanner[0-15] (don't forget
-about group and world permissions). Both read and write permissions
-are required for proper operation. For example:
- chmod 666 /dev/usb/scanner0
-
-Load the appropriate modules (if compiled as modules):
-
- modprobe ohci-hcd (or uhci-hcd, ehci-hcd)
- modprobe scanner
-
-
-DEVFS
-
-The later versions of the Linux kernel (2.4.8'ish) included a dynamic
-device filesystem call "devfs". With devfs, there is no need to
-create the device files as explained above; instead, they are
-dynamically created for you. For USB Scanner, the device is created
-in /dev/usb/scannerX where X can range from 0 to 15 depending on the
-number of scanners connected to the system.
-
-To see if you have devfs, issue the command "cat /proc/filesytems".
-If devfs is listed you should be ready to go. You should also have a
-process running called "devfsd". In order to make sure, issue the
-command "ps aux | grep '[d]evfsd'".
-
-
-CONCLUSION
-
-That's it. SANE should now be able to access the device. To make sure the
-device was detected, use "cat /proc/bus/usb/devices". Your scanner should be
-listed and the line starting with "I:" should look similar to this example:
-
- I: If#= 0 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=usbscanner
-
-The important part is "Driver=usbscanner". If it reads "Driver=(none)", the
-USB scanner driver didn't recognize the scanner. Have a look at the MODULE
-PARAMETERS section for what to do in this case.
-
-For more details on the format of "/proc/bus/usb/devices" see
-Documentation/usb/proc_usb_info.txt.
-
-
-MESSAGES
-
-usb_control/bulk_msg: timeout -- On occasions this message will appear
-in "/var/adm/messages", on the console, or both depending on how
-your system is configured. This is a side effect that scanners are
-sometimes very slow at warming up and/or initializing. In most cases,
-however, only several of these messages should appear and is generally
-considered to be normal.
-
-excessive NAK's received -- This message should be considered abnormal
-and generally indicates that the USB system is unable to communicate
-with the scanner for some particular reason.
-
-probe_scanner: Undetected endpoint -- The USB Scanner driver is fairly
-general when it comes to communicating to scanners. Unfortunately,
-some vendors have designed their scanners in one way or another that
-this driver doesn't account for.
-
-probe_scanner: Endpoint determination failed -- This means that the
-driver is unable to detect a supported configuration for means to
-communicate with the scanner. See also "probe_scanner: Undetected
-endpoint".
-
-funky result -- Most of the time the data flow between the computer
-and the scanner goes smoothly. However, due to whatever reason,
-whether it be solar flares or stray neutrons, sometimes the
-communications don't work as expected. The driver tries to handle
-most types of errors but not all. When this message is seen,
-something weird happened. Please contact the mailing list (see
-CONTACT section for details).
-
-
-MODULE PARAMETERS
-
-If you have a device that you wish to experiment with or try using
-this driver with, but the Vendor and Product IDs are not coded in,
-don't despair. If the driver was compiled as a module, you can pass
-options to the driver. Simply add
-
- options scanner vendor=0x#### product=0x****
-
-to the /etc/modprobe.conf file replacing the #'s and the *'s with the
-correct IDs. The IDs can be retrieved from the messages file or
-using "cat /proc/bus/usb/devices".
-
-If the default timeout is too low, i.e. there are frequent "timeout" messages,
-you may want to increase the timeout manually by using the parameter
-"read_timeout". The time is given in seconds. This is an example for
-modprobe.conf with a timeout of 60 seconds:
-
- options scanner read_timeout=60
-
-If the "scanner" module is already loaded into memory, it must be reloaded for
-the module parameters to take effect. In essence, "rmmod scanner; modprobe
-scanner" must be performed.
-
-
-BUGS
-
-Just look at the list of fixes in the source files.
-
-
-CONTACT
-
-For asking about problems and fixes, use the linux-usb-users mailing list. For
-patches, linux-usb-devel should be used. Information on both lists can be
-found on http://www.linux-usb.org/.
-
-
-CHANGES
-
-- Amended for linux-2.5.54
-- Added information about read_timeout
-- Added more details about /proc/bus/usb/devices
-- Added/updated links
-- Added pointers two "special" scanner drivers
-- Reordering, spell-checking, formatting
-- Used /dev/usb/scanner[0-15] instead of /dev/usbscanner[0-15]
-- Removed some basic USB configuration stuff
-- Added EHCI
-- Removed some more references to HP
-- Amended for linux-2.4.12
-- Updated devfs support
-- Amended for linux-2.3.99-pre6-3
-- Appended hp_scan.c to end of this README
-- Removed most references to HP
-- Updated uhci/ohci host controller info
-- Updated support for multiple scanner support
-- Updated supported scanners list
-- Updated usbdevfs info
-- Spellcheck
-
-
-HP TEST PROGRAM
-
-There is a small test program (hp_scan.c -- appended below) that can
-be used to test the scanner device if it's an HP scanner that supports
-SCL (Scanner Control Language). Known HP scanner that support SCL are
-the 4100, 5200, 6200, the 6300 -- note that the 4200 is *not*
-supported since it does not understand SCL; it's also strongly
-suspected that the 3300 and the PhotoSmart S20 are not SCL compliant.
-Hp_scan.c's purpose is to test the driver without having to
-retrieve/configure SANE. Hp_scan.c will scan the entire bed and put
-the output into a file called "out.dat" in the current directory. The
-data in the file is raw data so it's not very useful for imaging.
-
---------------- snip -- hp_scan.c -- snip ---------------
-/*
-
-This is a really crude attempt at writing a short test program. It's
-mostly only to be used to test connectivity with USB HP scanners that
-understand SCL. Currently, the supported models are 4100C, 5200C,
-6200C, and the 6300C. Note that the 4200C is *NOT* acceptable.
-
-Copyright (C) David E. Nelson <dnelson@jump.net>, 1999
-
-This program is free software; you can redistribute it and/or modify
-it under the terms of the GNU General Public License as published by
-the Free Software Foundation; either version 2 of the License, or (at
-your option) any later version.
-
-*/
-
-#include <stdio.h>
-#include <stdlib.h>
-#include <error.h>
-#include <unistd.h>
-#include <fcntl.h>
-
-/*
- Gray Output produces about a 8945400 byte file.
- Color Output produces a 26836200 byte file.
-
- To compile: gcc -o hp_scan hp_scan.c
-*/
-
-// #define COLOR /* Undef to scan GrayScale */
-
-int send_cmd(int, const char *, int);
-int read_cmd(int, char *, int);
-
-int
-main(void) {
-
- ssize_t cnt = 0, total_cnt = 0;
-
- FILE *fpout;
-
- int fp;
- int data_size = 32768;
-
- char *data;
-
- static char reset_cmd[] = {'\x1b','E'};
-
-#ifdef COLOR
- static char data_type_cmd[] = {'\x1b','*','a','5','T'}; /* Color */
- static char data_width_cmd[] = {'\x1b','*','a','2','4','G'}; /* 24 Bit Color */
-#else
- static char data_type_cmd[] = {'\x1b','*','a','4','T'}; /* Gray */
- static char data_width_cmd[] = {'\x1b','*','a','8','G'}; /* 8 Bit Gray */
-#endif
-
- static char query_cmd[] = {'\x1b', '*', 's', '2', '5', '7', 'E'};
- static char start_scan_cmd[] = {'\x1b','*','f','0','S'};
-
- if(!(data=malloc(data_size))) {
- perror("malloc failed");
- exit (1);
- }
-
- if((fp=open("/dev/usb/scanner0", O_RDWR)) < 0) {
- perror("Unable to open scanner device");
- exit (1);
- }
-
- if((fpout=fopen("out.dat", "w+")) == NULL) {
- perror("Unable to open output file");
- exit(1);
- }
-
- send_cmd(fp, reset_cmd, sizeof(reset_cmd));
- send_cmd(fp, data_type_cmd, sizeof(data_type_cmd));
- send_cmd(fp, data_width_cmd, sizeof(data_width_cmd));
- send_cmd(fp, start_scan_cmd, sizeof(start_scan_cmd));
-
- while ((cnt = read(fp, data, data_size)) > 0) {
- printf("Read: %u\n", cnt);
- if(fwrite(data, sizeof(char), cnt, fpout) < 0) {
- perror("Write to output file failed");
- exit (1);
- }
- total_cnt += cnt;
- }
- if (cnt < 0) {
- perror("Read from scanner failed");
- exit (1);
- }
-
- printf("\nRead %lu bytes.\n", total_cnt);
-
- send_cmd(fp, reset_cmd, sizeof(reset_cmd));
-
- close(fp);
- fclose(fpout);
- return (0);
-}
-
-int
-send_cmd(int fp, const char * cmd, int length) {
-
- int result;
- int x;
-
- if((result = write(fp, cmd, length)) != length) {
- printf ("Write warning: %d bytes requested, %d written\n");
- } else if (result < 0) {
- perror ("send_cmd failure");
- exit (1);
- }
- return (result);
-}
-
-int
-read_cmd(int fp, char * response, int length) {
-
- return read(fp, response, length);
-
-}
--- diff/drivers/char/ipmi/ipmi_kcs_intf.c 2004-03-11 10:20:23.000000000 +0000
+++ source/drivers/char/ipmi/ipmi_kcs_intf.c 1970-01-01 00:00:00.000000000 +0000
@@ -1,1305 +0,0 @@
-/*
- * ipmi_kcs_intf.c
- *
- * The interface to the IPMI driver for the KCS.
- *
- * Author: MontaVista Software, Inc.
- * Corey Minyard <minyard@mvista.com>
- * source@mvista.com
- *
- * Copyright 2002 MontaVista Software Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the
- * Free Software Foundation; either version 2 of the License, or (at your
- * option) any later version.
- *
- *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
- * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
- * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
- * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
- * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
- * OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR
- * TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
- * USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program; if not, write to the Free Software Foundation, Inc.,
- * 675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-/*
- * This file holds the "policy" for the interface to the KCS state
- * machine. It does the configuration, handles timers and interrupts,
- * and drives the real KCS state machine.
- */
-
-#include <linux/config.h>
-#include <linux/module.h>
-#include <asm/system.h>
-#include <linux/sched.h>
-#include <linux/timer.h>
-#include <linux/errno.h>
-#include <linux/spinlock.h>
-#include <linux/slab.h>
-#include <linux/delay.h>
-#include <linux/list.h>
-#include <linux/ioport.h>
-#ifdef CONFIG_HIGH_RES_TIMERS
-#include <linux/hrtime.h>
-#endif
-#include <linux/interrupt.h>
-#include <linux/rcupdate.h>
-#include <linux/ipmi_smi.h>
-#include <asm/io.h>
-#include <asm/irq.h>
-#include "ipmi_kcs_sm.h"
-#include <linux/init.h>
-
-/* Measure times between events in the driver. */
-#undef DEBUG_TIMING
-
-/* Timing parameters. Call every 10 ms when not doing anything,
- otherwise call every KCS_SHORT_TIMEOUT_USEC microseconds. */
-#define KCS_TIMEOUT_TIME_USEC 10000
-#define KCS_USEC_PER_JIFFY (1000000/HZ)
-#define KCS_TIMEOUT_JIFFIES (KCS_TIMEOUT_TIME_USEC/KCS_USEC_PER_JIFFY)
-#define KCS_SHORT_TIMEOUT_USEC 250 /* .25ms when the SM request a
- short timeout */
-
-#ifdef CONFIG_IPMI_KCS
-/* This forces a dependency to the config file for this option. */
-#endif
-
-enum kcs_intf_state {
- KCS_NORMAL,
- KCS_GETTING_FLAGS,
- KCS_GETTING_EVENTS,
- KCS_CLEARING_FLAGS,
- KCS_CLEARING_FLAGS_THEN_SET_IRQ,
- KCS_GETTING_MESSAGES,
- KCS_ENABLE_INTERRUPTS1,
- KCS_ENABLE_INTERRUPTS2
- /* FIXME - add watchdog stuff. */
-};
-
-struct kcs_info
-{
- ipmi_smi_t intf;
- struct kcs_data *kcs_sm;
- spinlock_t kcs_lock;
- spinlock_t msg_lock;
- struct list_head xmit_msgs;
- struct list_head hp_xmit_msgs;
- struct ipmi_smi_msg *curr_msg;
- enum kcs_intf_state kcs_state;
-
- /* Flags from the last GET_MSG_FLAGS command, used when an ATTN
- is set to hold the flags until we are done handling everything
- from the flags. */
-#define RECEIVE_MSG_AVAIL 0x01
-#define EVENT_MSG_BUFFER_FULL 0x02
-#define WDT_PRE_TIMEOUT_INT 0x08
- unsigned char msg_flags;
-
- /* If set to true, this will request events the next time the
- state machine is idle. */
- atomic_t req_events;
-
- /* If true, run the state machine to completion on every send
- call. Generally used after a panic to make sure stuff goes
- out. */
- int run_to_completion;
-
- /* The I/O port of a KCS interface. */
- int port;
-
- /* zero if no irq; */
- int irq;
-
- /* The physical and remapped memory addresses of a KCS interface. */
- unsigned long physaddr;
- unsigned char *addr;
-
- /* The timer for this kcs. */
- struct timer_list kcs_timer;
-
- /* The time (in jiffies) the last timeout occurred at. */
- unsigned long last_timeout_jiffies;
-
- /* Used to gracefully stop the timer without race conditions. */
- volatile int stop_operation;
- volatile int timer_stopped;
-
- /* The driver will disable interrupts when it gets into a
- situation where it cannot handle messages due to lack of
- memory. Once that situation clears up, it will re-enable
- interrupts. */
- int interrupt_disabled;
-};
-
-static void kcs_restart_short_timer(struct kcs_info *kcs_info);
-
-static void deliver_recv_msg(struct kcs_info *kcs_info, struct ipmi_smi_msg *msg)
-{
- /* Deliver the message to the upper layer with the lock
- released. */
- spin_unlock(&(kcs_info->kcs_lock));
- ipmi_smi_msg_received(kcs_info->intf, msg);
- spin_lock(&(kcs_info->kcs_lock));
-}
-
-static void return_hosed_msg(struct kcs_info *kcs_info)
-{
- struct ipmi_smi_msg *msg = kcs_info->curr_msg;
-
- /* Make it a reponse */
- msg->rsp[0] = msg->data[0] | 4;
- msg->rsp[1] = msg->data[1];
- msg->rsp[2] = 0xFF; /* Unknown error. */
- msg->rsp_size = 3;
-
- kcs_info->curr_msg = NULL;
- deliver_recv_msg(kcs_info, msg);
-}
-
-static enum kcs_result start_next_msg(struct kcs_info *kcs_info)
-{
- int rv;
- struct list_head *entry = NULL;
-#ifdef DEBUG_TIMING
- struct timeval t;
-#endif
-
- /* No need to save flags, we aleady have interrupts off and we
- already hold the KCS lock. */
- spin_lock(&(kcs_info->msg_lock));
-
- /* Pick the high priority queue first. */
- if (! list_empty(&(kcs_info->hp_xmit_msgs))) {
- entry = kcs_info->hp_xmit_msgs.next;
- } else if (! list_empty(&(kcs_info->xmit_msgs))) {
- entry = kcs_info->xmit_msgs.next;
- }
-
- if (!entry) {
- kcs_info->curr_msg = NULL;
- rv = KCS_SM_IDLE;
- } else {
- int err;
-
- list_del(entry);
- kcs_info->curr_msg = list_entry(entry,
- struct ipmi_smi_msg,
- link);
-#ifdef DEBUG_TIMING
- do_gettimeofday(&t);
- printk("**Start2: %d.%9.9d\n", t.tv_sec, t.tv_usec);
-#endif
- err = start_kcs_transaction(kcs_info->kcs_sm,
- kcs_info->curr_msg->data,
- kcs_info->curr_msg->data_size);
- if (err) {
- return_hosed_msg(kcs_info);
- }
-
- rv = KCS_CALL_WITHOUT_DELAY;
- }
- spin_unlock(&(kcs_info->msg_lock));
-
- return rv;
-}
-
-static void start_enable_irq(struct kcs_info *kcs_info)
-{
- unsigned char msg[2];
-
- /* If we are enabling interrupts, we have to tell the
- BMC to use them. */
- msg[0] = (IPMI_NETFN_APP_REQUEST << 2);
- msg[1] = IPMI_GET_BMC_GLOBAL_ENABLES_CMD;
-
- start_kcs_transaction(kcs_info->kcs_sm, msg, 2);
- kcs_info->kcs_state = KCS_ENABLE_INTERRUPTS1;
-}
-
-static void start_clear_flags(struct kcs_info *kcs_info)
-{
- unsigned char msg[3];
-
- /* Make sure the watchdog pre-timeout flag is not set at startup. */
- msg[0] = (IPMI_NETFN_APP_REQUEST << 2);
- msg[1] = IPMI_CLEAR_MSG_FLAGS_CMD;
- msg[2] = WDT_PRE_TIMEOUT_INT;
-
- start_kcs_transaction(kcs_info->kcs_sm, msg, 3);
- kcs_info->kcs_state = KCS_CLEARING_FLAGS;
-}
-
-/* When we have a situtaion where we run out of memory and cannot
- allocate messages, we just leave them in the BMC and run the system
- polled until we can allocate some memory. Once we have some
- memory, we will re-enable the interrupt. */
-static inline void disable_kcs_irq(struct kcs_info *kcs_info)
-{
- if ((kcs_info->irq) && (!kcs_info->interrupt_disabled)) {
- disable_irq_nosync(kcs_info->irq);
- kcs_info->interrupt_disabled = 1;
- }
-}
-
-static inline void enable_kcs_irq(struct kcs_info *kcs_info)
-{
- if ((kcs_info->irq) && (kcs_info->interrupt_disabled)) {
- enable_irq(kcs_info->irq);
- kcs_info->interrupt_disabled = 0;
- }
-}
-
-static void handle_flags(struct kcs_info *kcs_info)
-{
- if (kcs_info->msg_flags & WDT_PRE_TIMEOUT_INT) {
- /* Watchdog pre-timeout */
- start_clear_flags(kcs_info);
- kcs_info->msg_flags &= ~WDT_PRE_TIMEOUT_INT;
- spin_unlock(&(kcs_info->kcs_lock));
- ipmi_smi_watchdog_pretimeout(kcs_info->intf);
- spin_lock(&(kcs_info->kcs_lock));
- } else if (kcs_info->msg_flags & RECEIVE_MSG_AVAIL) {
- /* Messages available. */
- kcs_info->curr_msg = ipmi_alloc_smi_msg();
- if (!kcs_info->curr_msg) {
- disable_kcs_irq(kcs_info);
- kcs_info->kcs_state = KCS_NORMAL;
- return;
- }
- enable_kcs_irq(kcs_info);
-
- kcs_info->curr_msg->data[0] = (IPMI_NETFN_APP_REQUEST << 2);
- kcs_info->curr_msg->data[1] = IPMI_GET_MSG_CMD;
- kcs_info->curr_msg->data_size = 2;
-
- start_kcs_transaction(kcs_info->kcs_sm,
- kcs_info->curr_msg->data,
- kcs_info->curr_msg->data_size);
- kcs_info->kcs_state = KCS_GETTING_MESSAGES;
- } else if (kcs_info->msg_flags & EVENT_MSG_BUFFER_FULL) {
- /* Events available. */
- kcs_info->curr_msg = ipmi_alloc_smi_msg();
- if (!kcs_info->curr_msg) {
- disable_kcs_irq(kcs_info);
- kcs_info->kcs_state = KCS_NORMAL;
- return;
- }
- enable_kcs_irq(kcs_info);
-
- kcs_info->curr_msg->data[0] = (IPMI_NETFN_APP_REQUEST << 2);
- kcs_info->curr_msg->data[1] = IPMI_READ_EVENT_MSG_BUFFER_CMD;
- kcs_info->curr_msg->data_size = 2;
-
- start_kcs_transaction(kcs_info->kcs_sm,
- kcs_info->curr_msg->data,
- kcs_info->curr_msg->data_size);
- kcs_info->kcs_state = KCS_GETTING_EVENTS;
- } else {
- kcs_info->kcs_state = KCS_NORMAL;
- }
-}
-
-static void handle_transaction_done(struct kcs_info *kcs_info)
-{
- struct ipmi_smi_msg *msg;
-#ifdef DEBUG_TIMING
- struct timeval t;
-
- do_gettimeofday(&t);
- printk("**Done: %d.%9.9d\n", t.tv_sec, t.tv_usec);
-#endif
- switch (kcs_info->kcs_state) {
- case KCS_NORMAL:
- if (!kcs_info->curr_msg)
- break;
-
- kcs_info->curr_msg->rsp_size
- = kcs_get_result(kcs_info->kcs_sm,
- kcs_info->curr_msg->rsp,
- IPMI_MAX_MSG_LENGTH);
-
- /* Do this here becase deliver_recv_msg() releases the
- lock, and a new message can be put in during the
- time the lock is released. */
- msg = kcs_info->curr_msg;
- kcs_info->curr_msg = NULL;
- deliver_recv_msg(kcs_info, msg);
- break;
-
- case KCS_GETTING_FLAGS:
- {
- unsigned char msg[4];
- unsigned int len;
-
- /* We got the flags from the KCS, now handle them. */
- len = kcs_get_result(kcs_info->kcs_sm, msg, 4);
- if (msg[2] != 0) {
- /* Error fetching flags, just give up for
- now. */
- kcs_info->kcs_state = KCS_NORMAL;
- } else if (len < 3) {
- /* Hmm, no flags. That's technically illegal, but
- don't use uninitialized data. */
- kcs_info->kcs_state = KCS_NORMAL;
- } else {
- kcs_info->msg_flags = msg[3];
- handle_flags(kcs_info);
- }
- break;
- }
-
- case KCS_CLEARING_FLAGS:
- case KCS_CLEARING_FLAGS_THEN_SET_IRQ:
- {
- unsigned char msg[3];
-
- /* We cleared the flags. */
- kcs_get_result(kcs_info->kcs_sm, msg, 3);
- if (msg[2] != 0) {
- /* Error clearing flags */
- printk(KERN_WARNING
- "ipmi_kcs: Error clearing flags: %2.2x\n",
- msg[2]);
- }
- if (kcs_info->kcs_state == KCS_CLEARING_FLAGS_THEN_SET_IRQ)
- start_enable_irq(kcs_info);
- else
- kcs_info->kcs_state = KCS_NORMAL;
- break;
- }
-
- case KCS_GETTING_EVENTS:
- {
- kcs_info->curr_msg->rsp_size
- = kcs_get_result(kcs_info->kcs_sm,
- kcs_info->curr_msg->rsp,
- IPMI_MAX_MSG_LENGTH);
-
- /* Do this here becase deliver_recv_msg() releases the
- lock, and a new message can be put in during the
- time the lock is released. */
- msg = kcs_info->curr_msg;
- kcs_info->curr_msg = NULL;
- if (msg->rsp[2] != 0) {
- /* Error getting event, probably done. */
- msg->done(msg);
-
- /* Take off the event flag. */
- kcs_info->msg_flags &= ~EVENT_MSG_BUFFER_FULL;
- } else {
- deliver_recv_msg(kcs_info, msg);
- }
- handle_flags(kcs_info);
- break;
- }
-
- case KCS_GETTING_MESSAGES:
- {
- kcs_info->curr_msg->rsp_size
- = kcs_get_result(kcs_info->kcs_sm,
- kcs_info->curr_msg->rsp,
- IPMI_MAX_MSG_LENGTH);
-
- /* Do this here becase deliver_recv_msg() releases the
- lock, and a new message can be put in during the
- time the lock is released. */
- msg = kcs_info->curr_msg;
- kcs_info->curr_msg = NULL;
- if (msg->rsp[2] != 0) {
- /* Error getting event, probably done. */
- msg->done(msg);
-
- /* Take off the msg flag. */
- kcs_info->msg_flags &= ~RECEIVE_MSG_AVAIL;
- } else {
- deliver_recv_msg(kcs_info, msg);
- }
- handle_flags(kcs_info);
- break;
- }
-
- case KCS_ENABLE_INTERRUPTS1:
- {
- unsigned char msg[4];
-
- /* We got the flags from the KCS, now handle them. */
- kcs_get_result(kcs_info->kcs_sm, msg, 4);
- if (msg[2] != 0) {
- printk(KERN_WARNING
- "ipmi_kcs: Could not enable interrupts"
- ", failed get, using polled mode.\n");
- kcs_info->kcs_state = KCS_NORMAL;
- } else {
- msg[0] = (IPMI_NETFN_APP_REQUEST << 2);
- msg[1] = IPMI_SET_BMC_GLOBAL_ENABLES_CMD;
- msg[2] = msg[3] | 1; /* enable msg queue int */
- start_kcs_transaction(kcs_info->kcs_sm, msg,3);
- kcs_info->kcs_state = KCS_ENABLE_INTERRUPTS2;
- }
- break;
- }
-
- case KCS_ENABLE_INTERRUPTS2:
- {
- unsigned char msg[4];
-
- /* We got the flags from the KCS, now handle them. */
- kcs_get_result(kcs_info->kcs_sm, msg, 4);
- if (msg[2] != 0) {
- printk(KERN_WARNING
- "ipmi_kcs: Could not enable interrupts"
- ", failed set, using polled mode.\n");
- }
- kcs_info->kcs_state = KCS_NORMAL;
- break;
- }
- }
-}
-
-/* Called on timeouts and events. Timeouts should pass the elapsed
- time, interrupts should pass in zero. */
-static enum kcs_result kcs_event_handler(struct kcs_info *kcs_info, int time)
-{
- enum kcs_result kcs_result;
-
- restart:
- /* There used to be a loop here that waited a little while
- (around 25us) before giving up. That turned out to be
- pointless, the minimum delays I was seeing were in the 300us
- range, which is far too long to wait in an interrupt. So
- we just run until the state machine tells us something
- happened or it needs a delay. */
- kcs_result = kcs_event(kcs_info->kcs_sm, time);
- time = 0;
- while (kcs_result == KCS_CALL_WITHOUT_DELAY)
- {
- kcs_result = kcs_event(kcs_info->kcs_sm, 0);
- }
-
- if (kcs_result == KCS_TRANSACTION_COMPLETE)
- {
- handle_transaction_done(kcs_info);
- kcs_result = kcs_event(kcs_info->kcs_sm, 0);
- }
- else if (kcs_result == KCS_SM_HOSED)
- {
- if (kcs_info->curr_msg != NULL) {
- /* If we were handling a user message, format
- a response to send to the upper layer to
- tell it about the error. */
- return_hosed_msg(kcs_info);
- }
- kcs_result = kcs_event(kcs_info->kcs_sm, 0);
- kcs_info->kcs_state = KCS_NORMAL;
- }
-
- /* We prefer handling attn over new messages. */
- if (kcs_result == KCS_ATTN)
- {
- unsigned char msg[2];
-
- /* Got a attn, send down a get message flags to see
- what's causing it. It would be better to handle
- this in the upper layer, but due to the way
- interrupts work with the KCS, that's not really
- possible. */
- msg[0] = (IPMI_NETFN_APP_REQUEST << 2);
- msg[1] = IPMI_GET_MSG_FLAGS_CMD;
-
- start_kcs_transaction(kcs_info->kcs_sm, msg, 2);
- kcs_info->kcs_state = KCS_GETTING_FLAGS;
- goto restart;
- }
-
- /* If we are currently idle, try to start the next message. */
- if (kcs_result == KCS_SM_IDLE) {
- kcs_result = start_next_msg(kcs_info);
- if (kcs_result != KCS_SM_IDLE)
- goto restart;
- }
-
- if ((kcs_result == KCS_SM_IDLE)
- && (atomic_read(&kcs_info->req_events)))
- {
- /* We are idle and the upper layer requested that I fetch
- events, so do so. */
- unsigned char msg[2];
-
- atomic_set(&kcs_info->req_events, 0);
- msg[0] = (IPMI_NETFN_APP_REQUEST << 2);
- msg[1] = IPMI_GET_MSG_FLAGS_CMD;
-
- start_kcs_transaction(kcs_info->kcs_sm, msg, 2);
- kcs_info->kcs_state = KCS_GETTING_FLAGS;
- goto restart;
- }
-
- return kcs_result;
-}
-
-static void sender(void *send_info,
- struct ipmi_smi_msg *msg,
- int priority)
-{
- struct kcs_info *kcs_info = (struct kcs_info *) send_info;
- enum kcs_result result;
- unsigned long flags;
-#ifdef DEBUG_TIMING
- struct timeval t;
-#endif
-
- spin_lock_irqsave(&(kcs_info->msg_lock), flags);
-#ifdef DEBUG_TIMING
- do_gettimeofday(&t);
- printk("**Enqueue: %d.%9.9d\n", t.tv_sec, t.tv_usec);
-#endif
-
- if (kcs_info->run_to_completion) {
- /* If we are running to completion, then throw it in
- the list and run transactions until everything is
- clear. Priority doesn't matter here. */
- list_add_tail(&(msg->link), &(kcs_info->xmit_msgs));
-
- /* We have to release the msg lock and claim the kcs
- lock in this case, because of race conditions. */
- spin_unlock_irqrestore(&(kcs_info->msg_lock), flags);
-
- spin_lock_irqsave(&(kcs_info->kcs_lock), flags);
- result = kcs_event_handler(kcs_info, 0);
- while (result != KCS_SM_IDLE) {
- udelay(KCS_SHORT_TIMEOUT_USEC);
- result = kcs_event_handler(kcs_info,
- KCS_SHORT_TIMEOUT_USEC);
- }
- spin_unlock_irqrestore(&(kcs_info->kcs_lock), flags);
- return;
- } else {
- if (priority > 0) {
- list_add_tail(&(msg->link), &(kcs_info->hp_xmit_msgs));
- } else {
- list_add_tail(&(msg->link), &(kcs_info->xmit_msgs));
- }
- }
- spin_unlock_irqrestore(&(kcs_info->msg_lock), flags);
-
- spin_lock_irqsave(&(kcs_info->kcs_lock), flags);
- if ((kcs_info->kcs_state == KCS_NORMAL)
- && (kcs_info->curr_msg == NULL))
- {
- start_next_msg(kcs_info);
- kcs_restart_short_timer(kcs_info);
- }
- spin_unlock_irqrestore(&(kcs_info->kcs_lock), flags);
-}
-
-static void set_run_to_completion(void *send_info, int i_run_to_completion)
-{
- struct kcs_info *kcs_info = (struct kcs_info *) send_info;
- enum kcs_result result;
- unsigned long flags;
-
- spin_lock_irqsave(&(kcs_info->kcs_lock), flags);
-
- kcs_info->run_to_completion = i_run_to_completion;
- if (i_run_to_completion) {
- result = kcs_event_handler(kcs_info, 0);
- while (result != KCS_SM_IDLE) {
- udelay(KCS_SHORT_TIMEOUT_USEC);
- result = kcs_event_handler(kcs_info,
- KCS_SHORT_TIMEOUT_USEC);
- }
- }
-
- spin_unlock_irqrestore(&(kcs_info->kcs_lock), flags);
-}
-
-static void request_events(void *send_info)
-{
- struct kcs_info *kcs_info = (struct kcs_info *) send_info;
-
- atomic_set(&kcs_info->req_events, 1);
-}
-
-static int initialized = 0;
-
-/* Must be called with interrupts off and with the kcs_lock held. */
-static void kcs_restart_short_timer(struct kcs_info *kcs_info)
-{
- if (del_timer(&(kcs_info->kcs_timer))) {
-#ifdef CONFIG_HIGH_RES_TIMERS
- unsigned long jiffies_now;
-
- /* If we don't delete the timer, then it will go off
- immediately, anyway. So we only process if we
- actually delete the timer. */
-
- /* We already have irqsave on, so no need for it
- here. */
- read_lock(&xtime_lock);
- jiffies_now = jiffies;
- kcs_info->kcs_timer.expires = jiffies_now;
-
- kcs_info->kcs_timer.sub_expires
- = quick_update_jiffies_sub(jiffies_now);
- read_unlock(&xtime_lock);
-
- kcs_info->kcs_timer.sub_expires
- += usec_to_arch_cycles(KCS_SHORT_TIMEOUT_USEC);
- while (kcs_info->kcs_timer.sub_expires >= cycles_per_jiffies) {
- kcs_info->kcs_timer.expires++;
- kcs_info->kcs_timer.sub_expires -= cycles_per_jiffies;
- }
-#else
- kcs_info->kcs_timer.expires = jiffies + 1;
-#endif
- add_timer(&(kcs_info->kcs_timer));
- }
-}
-
-static void kcs_timeout(unsigned long data)
-{
- struct kcs_info *kcs_info = (struct kcs_info *) data;
- enum kcs_result kcs_result;
- unsigned long flags;
- unsigned long jiffies_now;
- unsigned long time_diff;
-#ifdef DEBUG_TIMING
- struct timeval t;
-#endif
-
- if (kcs_info->stop_operation) {
- kcs_info->timer_stopped = 1;
- return;
- }
-
- spin_lock_irqsave(&(kcs_info->kcs_lock), flags);
-#ifdef DEBUG_TIMING
- do_gettimeofday(&t);
- printk("**Timer: %d.%9.9d\n", t.tv_sec, t.tv_usec);
-#endif
- jiffies_now = jiffies;
-
- time_diff = ((jiffies_now - kcs_info->last_timeout_jiffies)
- * KCS_USEC_PER_JIFFY);
- kcs_result = kcs_event_handler(kcs_info, time_diff);
-
- kcs_info->last_timeout_jiffies = jiffies_now;
-
- if ((kcs_info->irq) && (! kcs_info->interrupt_disabled)) {
- /* Running with interrupts, only do long timeouts. */
- kcs_info->kcs_timer.expires = jiffies + KCS_TIMEOUT_JIFFIES;
- goto do_add_timer;
- }
-
- /* If the state machine asks for a short delay, then shorten
- the timer timeout. */
-#ifdef CONFIG_HIGH_RES_TIMERS
- if (kcs_result == KCS_CALL_WITH_DELAY) {
- kcs_info->kcs_timer.sub_expires
- += usec_to_arch_cycles(KCS_SHORT_TIMEOUT_USEC);
- while (kcs_info->kcs_timer.sub_expires >= cycles_per_jiffies) {
- kcs_info->kcs_timer.expires++;
- kcs_info->kcs_timer.sub_expires -= cycles_per_jiffies;
- }
- } else {
- kcs_info->kcs_timer.expires = jiffies + KCS_TIMEOUT_JIFFIES;
- kcs_info->kcs_timer.sub_expires = 0;
- }
-#else
- /* If requested, take the shortest delay possible */
- if (kcs_result == KCS_CALL_WITH_DELAY) {
- kcs_info->kcs_timer.expires = jiffies + 1;
- } else {
- kcs_info->kcs_timer.expires = jiffies + KCS_TIMEOUT_JIFFIES;
- }
-#endif
-
- do_add_timer:
- add_timer(&(kcs_info->kcs_timer));
- spin_unlock_irqrestore(&(kcs_info->kcs_lock), flags);
-}
-
-static irqreturn_t kcs_irq_handler(int irq, void *data, struct pt_regs *regs)
-{
- struct kcs_info *kcs_info = (struct kcs_info *) data;
- unsigned long flags;
-#ifdef DEBUG_TIMING
- struct timeval t;
-#endif
-
- spin_lock_irqsave(&(kcs_info->kcs_lock), flags);
- if (kcs_info->stop_operation)
- goto out;
-
-#ifdef DEBUG_TIMING
- do_gettimeofday(&t);
- printk("**Interrupt: %d.%9.9d\n", t.tv_sec, t.tv_usec);
-#endif
- kcs_event_handler(kcs_info, 0);
- out:
- spin_unlock_irqrestore(&(kcs_info->kcs_lock), flags);
- return IRQ_HANDLED;
-}
-
-static struct ipmi_smi_handlers handlers =
-{
- .owner = THIS_MODULE,
- .sender = sender,
- .request_events = request_events,
- .set_run_to_completion = set_run_to_completion,
-};
-
-static unsigned char ipmi_kcs_dev_rev;
-static unsigned char ipmi_kcs_fw_rev_major;
-static unsigned char ipmi_kcs_fw_rev_minor;
-static unsigned char ipmi_version_major;
-static unsigned char ipmi_version_minor;
-
-extern int kcs_dbg;
-static int ipmi_kcs_detect_hardware(unsigned int port,
- unsigned char *addr,
- struct kcs_data *data)
-{
- unsigned char msg[2];
- unsigned char resp[IPMI_MAX_MSG_LENGTH];
- unsigned long resp_len;
- enum kcs_result kcs_result;
-
- /* It's impossible for the KCS status register to be all 1's,
- (assuming a properly functioning, self-initialized BMC)
- but that's what you get from reading a bogus address, so we
- test that first. */
-
- if (port) {
- if (inb(port+1) == 0xff) return -ENODEV;
- } else {
- if (readb(addr+1) == 0xff) return -ENODEV;
- }
-
- /* Do a Get Device ID command, since it comes back with some
- useful info. */
- msg[0] = IPMI_NETFN_APP_REQUEST << 2;
- msg[1] = IPMI_GET_DEVICE_ID_CMD;
- start_kcs_transaction(data, msg, 2);
-
- kcs_result = kcs_event(data, 0);
- for (;;)
- {
- if (kcs_result == KCS_CALL_WITH_DELAY) {
- udelay(100);
- kcs_result = kcs_event(data, 100);
- }
- else if (kcs_result == KCS_CALL_WITHOUT_DELAY)
- {
- kcs_result = kcs_event(data, 0);
- }
- else
- break;
- }
- if (kcs_result == KCS_SM_HOSED) {
- /* We couldn't get the state machine to run, so whatever's at
- the port is probably not an IPMI KCS interface. */
- return -ENODEV;
- }
- /* Otherwise, we got some data. */
- resp_len = kcs_get_result(data, resp, IPMI_MAX_MSG_LENGTH);
- if (resp_len < 6)
- /* That's odd, it should be longer. */
- return -EINVAL;
-
- if ((resp[1] != IPMI_GET_DEVICE_ID_CMD) || (resp[2] != 0))
- /* That's odd, it shouldn't be able to fail. */
- return -EINVAL;
-
- ipmi_kcs_dev_rev = resp[4] & 0xf;
- ipmi_kcs_fw_rev_major = resp[5] & 0x7f;
- ipmi_kcs_fw_rev_minor = resp[6];
- ipmi_version_major = resp[7] & 0xf;
- ipmi_version_minor = resp[7] >> 4;
-
- return 0;
-}
-
-/* There can be 4 IO ports passed in (with or without IRQs), 4 addresses,
- a default IO port, and 1 ACPI/SPMI address. That sets KCS_MAX_DRIVERS */
-
-#define KCS_MAX_PARMS 4
-#define KCS_MAX_DRIVERS ((KCS_MAX_PARMS * 2) + 2)
-static struct kcs_info *kcs_infos[KCS_MAX_DRIVERS] =
-{ NULL, NULL, NULL, NULL };
-
-#define DEVICE_NAME "ipmi_kcs"
-
-#define DEFAULT_IO_PORT 0xca2
-
-static int kcs_trydefaults = 1;
-static unsigned long kcs_addrs[KCS_MAX_PARMS] = { 0, 0, 0, 0 };
-static int kcs_ports[KCS_MAX_PARMS] = { 0, 0, 0, 0 };
-static int kcs_irqs[KCS_MAX_PARMS] = { 0, 0, 0, 0 };
-
-MODULE_PARM(kcs_trydefaults, "i");
-MODULE_PARM(kcs_addrs, "1-4l");
-MODULE_PARM(kcs_irqs, "1-4i");
-MODULE_PARM(kcs_ports, "1-4i");
-
-/* Returns 0 if initialized, or negative on an error. */
-static int init_one_kcs(int kcs_port,
- int irq,
- unsigned long kcs_physaddr,
- struct kcs_info **kcs)
-{
- int rv;
- struct kcs_info *new_kcs;
-
- /* Did anything get passed in at all? Both == zero disables the
- driver. */
-
- if (!(kcs_port || kcs_physaddr))
- return -ENODEV;
-
- /* Only initialize a port OR a physical address on this call.
- Also, IRQs can go with either ports or addresses. */
-
- if (kcs_port && kcs_physaddr)
- return -EINVAL;
-
- new_kcs = kmalloc(sizeof(*new_kcs), GFP_KERNEL);
- if (!new_kcs) {
- printk(KERN_ERR "ipmi_kcs: out of memory\n");
- return -ENOMEM;
- }
-
- /* So we know not to free it unless we have allocated one. */
- new_kcs->kcs_sm = NULL;
-
- new_kcs->addr = NULL;
- new_kcs->physaddr = kcs_physaddr;
- new_kcs->port = kcs_port;
-
- if (kcs_port) {
- if (request_region(kcs_port, 2, DEVICE_NAME) == NULL) {
- kfree(new_kcs);
- printk(KERN_ERR
- "ipmi_kcs: can't reserve port @ 0x%4.4x\n",
- kcs_port);
- return -EIO;
- }
- } else {
- if (request_mem_region(kcs_physaddr, 2, DEVICE_NAME) == NULL) {
- kfree(new_kcs);
- printk(KERN_ERR
- "ipmi_kcs: can't reserve memory @ 0x%lx\n",
- kcs_physaddr);
- return -EIO;
- }
- if ((new_kcs->addr = ioremap(kcs_physaddr, 2)) == NULL) {
- kfree(new_kcs);
- printk(KERN_ERR
- "ipmi_kcs: can't remap memory at 0x%lx\n",
- kcs_physaddr);
- return -EIO;
- }
- }
-
- new_kcs->kcs_sm = kmalloc(kcs_size(), GFP_KERNEL);
- if (!new_kcs->kcs_sm) {
- printk(KERN_ERR "ipmi_kcs: out of memory\n");
- rv = -ENOMEM;
- goto out_err;
- }
- init_kcs_data(new_kcs->kcs_sm, kcs_port, new_kcs->addr);
- spin_lock_init(&(new_kcs->kcs_lock));
- spin_lock_init(&(new_kcs->msg_lock));
-
- rv = ipmi_kcs_detect_hardware(kcs_port, new_kcs->addr, new_kcs->kcs_sm);
- if (rv) {
- if (kcs_port)
- printk(KERN_ERR
- "ipmi_kcs: No KCS @ port 0x%4.4x\n",
- kcs_port);
- else
- printk(KERN_ERR
- "ipmi_kcs: No KCS @ addr 0x%lx\n",
- kcs_physaddr);
- goto out_err;
- }
-
- if (irq != 0) {
- rv = request_irq(irq,
- kcs_irq_handler,
- SA_INTERRUPT,
- DEVICE_NAME,
- new_kcs);
- if (rv) {
- printk(KERN_WARNING
- "ipmi_kcs: %s unable to claim interrupt %d,"
- " running polled\n",
- DEVICE_NAME, irq);
- irq = 0;
- }
- }
- new_kcs->irq = irq;
-
- INIT_LIST_HEAD(&(new_kcs->xmit_msgs));
- INIT_LIST_HEAD(&(new_kcs->hp_xmit_msgs));
- new_kcs->curr_msg = NULL;
- atomic_set(&new_kcs->req_events, 0);
- new_kcs->run_to_completion = 0;
-
- start_clear_flags(new_kcs);
-
- if (irq) {
- new_kcs->kcs_state = KCS_CLEARING_FLAGS_THEN_SET_IRQ;
-
- printk(KERN_INFO
- "ipmi_kcs: Acquiring BMC @ port=0x%x irq=%d\n",
- kcs_port, irq);
-
- } else {
- if (kcs_port)
- printk(KERN_INFO
- "ipmi_kcs: Acquiring BMC @ port=0x%x\n",
- kcs_port);
- else
- printk(KERN_INFO
- "ipmi_kcs: Acquiring BMC @ addr=0x%lx\n",
- kcs_physaddr);
- }
-
- rv = ipmi_register_smi(&handlers,
- new_kcs,
- ipmi_version_major,
- ipmi_version_minor,
- &(new_kcs->intf));
- if (rv) {
- free_irq(irq, new_kcs);
- printk(KERN_ERR
- "ipmi_kcs: Unable to register device: error %d\n",
- rv);
- goto out_err;
- }
-
- new_kcs->interrupt_disabled = 0;
- new_kcs->timer_stopped = 0;
- new_kcs->stop_operation = 0;
-
- init_timer(&(new_kcs->kcs_timer));
- new_kcs->kcs_timer.data = (long) new_kcs;
- new_kcs->kcs_timer.function = kcs_timeout;
- new_kcs->last_timeout_jiffies = jiffies;
- new_kcs->kcs_timer.expires = jiffies + KCS_TIMEOUT_JIFFIES;
- add_timer(&(new_kcs->kcs_timer));
-
- *kcs = new_kcs;
-
- return 0;
-
- out_err:
- if (kcs_port)
- release_region (kcs_port, 2);
- if (new_kcs->addr)
- iounmap(new_kcs->addr);
- if (kcs_physaddr)
- release_mem_region(kcs_physaddr, 2);
- if (new_kcs->kcs_sm)
- kfree(new_kcs->kcs_sm);
- kfree(new_kcs);
- return rv;
-}
-
-#ifdef CONFIG_ACPI_INTERPRETER
-
-#include <linux/acpi.h>
-
-struct SPMITable {
- s8 Signature[4];
- u32 Length;
- u8 Revision;
- u8 Checksum;
- s8 OEMID[6];
- s8 OEMTableID[8];
- s8 OEMRevision[4];
- s8 CreatorID[4];
- s8 CreatorRevision[4];
- u8 InterfaceType[2];
- s16 SpecificationRevision;
-
- /*
- * Bit 0 - SCI interrupt supported
- * Bit 1 - I/O APIC/SAPIC
- */
- u8 InterruptType;
-
- /* If bit 0 of InterruptType is set, then this is the SCI
- interrupt in the GPEx_STS register. */
- u8 GPE;
-
- s16 Reserved;
-
- /* If bit 1 of InterruptType is set, then this is the I/O
- APIC/SAPIC interrupt. */
- u32 GlobalSystemInterrupt;
-
- /* The actual register address. */
- struct acpi_generic_address addr;
-
- u8 UID[4];
-
- s8 spmi_id[1]; /* A '\0' terminated array starts here. */
-};
-
-static int acpi_find_bmc(unsigned long *physaddr, int *port)
-{
- acpi_status status;
- struct SPMITable *spmi;
-
- status = acpi_get_firmware_table("SPMI", 1,
- ACPI_LOGICAL_ADDRESSING,
- (struct acpi_table_header **) &spmi);
- if (status != AE_OK)
- goto not_found;
-
- if (spmi->InterfaceType[0] != 1)
- /* Not IPMI. */
- goto not_found;
-
- if (spmi->InterfaceType[1] != 1)
- /* Not KCS. */
- goto not_found;
-
- if (spmi->addr.address_space_id == ACPI_ADR_SPACE_SYSTEM_MEMORY) {
- *physaddr = spmi->addr.address;
- printk("ipmi_kcs_intf: Found ACPI-specified state machine"
- " at memory address 0x%lx\n",
- (unsigned long) spmi->addr.address);
- } else if (spmi->addr.address_space_id == ACPI_ADR_SPACE_SYSTEM_IO) {
- *port = spmi->addr.address;
- printk("ipmi_kcs_intf: Found ACPI-specified state machine"
- " at I/O address 0x%lx\n",
- (unsigned long) spmi->addr.address);
- } else
- goto not_found; /* Not an address type we recognise. */
-
- return 0;
-
- not_found:
- return -ENODEV;
-}
-#endif
-
-static __init int init_ipmi_kcs(void)
-{
- int rv = 0;
- int pos = 0;
- int i = 0;
-#ifdef CONFIG_ACPI_INTERPRETER
- unsigned long physaddr = 0;
- int port = 0;
-#endif
-
- if (initialized)
- return 0;
- initialized = 1;
-
- /* First do the "command-line" parameters */
-
- for (i=0; i < KCS_MAX_PARMS; i++) {
- rv = init_one_kcs(kcs_ports[i],
- kcs_irqs[i],
- 0,
- &(kcs_infos[pos]));
- if (rv == 0)
- pos++;
-
- rv = init_one_kcs(0,
- kcs_irqs[i],
- kcs_addrs[i],
- &(kcs_infos[pos]));
- if (rv == 0)
- pos++;
- }
-
- /* Only try the defaults if enabled and resources are available
- (because they weren't already specified above). */
-
- if (kcs_trydefaults && (pos == 0)) {
- rv = -EINVAL;
-#ifdef CONFIG_ACPI_INTERPRETER
- if (rv && (physaddr = acpi_find_bmc(&physaddr, &port) == 0)) {
- rv = init_one_kcs(port,
- 0,
- physaddr,
- &(kcs_infos[pos]));
- if (rv == 0)
- pos++;
- }
-#endif
- if (rv) {
- rv = init_one_kcs(DEFAULT_IO_PORT,
- 0,
- 0,
- &(kcs_infos[pos]));
- if (rv == 0)
- pos++;
- }
- }
-
- if (kcs_infos[0] == NULL) {
- printk("ipmi_kcs: Unable to find any KCS interfaces\n");
- return -ENODEV;
- }
-
- return 0;
-}
-module_init(init_ipmi_kcs);
-
-#ifdef MODULE
-void __exit cleanup_one_kcs(struct kcs_info *to_clean)
-{
- int rv;
- unsigned long flags;
-
- if (! to_clean)
- return;
-
- /* Tell the timer and interrupt handlers that we are shutting
- down. */
- spin_lock_irqsave(&(to_clean->kcs_lock), flags);
- spin_lock(&(to_clean->msg_lock));
-
- to_clean->stop_operation = 1;
-
- if (to_clean->irq != 0)
- free_irq(to_clean->irq, to_clean);
- if (to_clean->port) {
- printk(KERN_INFO
- "ipmi_kcs: Releasing BMC @ port=0x%x\n",
- to_clean->port);
- release_region (to_clean->port, 2);
- }
- if (to_clean->addr) {
- printk(KERN_INFO
- "ipmi_kcs: Releasing BMC @ addr=0x%lx\n",
- to_clean->physaddr);
- iounmap(to_clean->addr);
- release_mem_region(to_clean->physaddr, 2);
- }
-
- spin_unlock(&(to_clean->msg_lock));
- spin_unlock_irqrestore(&(to_clean->kcs_lock), flags);
-
- /* Wait until we know that we are out of any interrupt
- handlers might have been running before we freed the
- interrupt. */
- synchronize_kernel();
-
- /* Wait for the timer to stop. This avoids problems with race
- conditions removing the timer here. */
- while (!to_clean->timer_stopped) {
- schedule_timeout(1);
- }
-
- rv = ipmi_unregister_smi(to_clean->intf);
- if (rv) {
- printk(KERN_ERR
- "ipmi_kcs: Unable to unregister device: errno=%d\n",
- rv);
- }
-
- initialized = 0;
-
- kfree(to_clean->kcs_sm);
- kfree(to_clean);
-}
-
-static __exit void cleanup_ipmi_kcs(void)
-{
- int i;
-
- if (!initialized)
- return;
-
- for (i=0; i<KCS_MAX_DRIVERS; i++) {
- cleanup_one_kcs(kcs_infos[i]);
- }
-}
-module_exit(cleanup_ipmi_kcs);
-#else
-
-/* Unfortunately, cmdline::get_options() only returns integers, not
- longs. Since we need ulongs (64-bit physical addresses) parse the
- comma-separated list manually. Arguments can be one of these forms:
- m0xaabbccddeeff A physical memory address without an IRQ
- m0xaabbccddeeff:cc A physical memory address with an IRQ
- p0xaabb An IO port without an IRQ
- p0xaabb:cc An IO port with an IRQ
- nodefaults Suppress trying the default IO port or ACPI address
-
- For example, to pass one IO port with an IRQ, one address, and
- suppress the use of the default IO port and ACPI address,
- use this option string: ipmi_kcs=p0xCA2:5,m0xFF5B0022,nodefaults
-
- Remember, ipmi_kcs_setup() is passed the string after the equal sign. */
-
-static int __init ipmi_kcs_setup(char *str)
-{
- unsigned long val;
- char *cur, *colon;
- int pos;
-
- pos = 0;
-
- cur = strsep(&str, ",");
- while ((cur) && (*cur) && (pos < KCS_MAX_PARMS)) {
- switch (*cur) {
- case 'n':
- if (strcmp(cur, "nodefaults") == 0)
- kcs_trydefaults = 0;
- else
- printk(KERN_INFO
- "ipmi_kcs: bad parameter value %s\n",
- cur);
- break;
-
- case 'm':
- case 'p':
- val = simple_strtoul(cur + 1,
- &colon,
- 0);
- if (*cur == 'p')
- kcs_ports[pos] = val;
- else
- kcs_addrs[pos] = val;
- if (*colon == ':') {
- val = simple_strtoul(colon + 1,
- &colon,
- 0);
- kcs_irqs[pos] = val;
- }
- pos++;
- break;
-
- default:
- printk(KERN_INFO
- "ipmi_kcs: bad parameter value %s\n",
- cur);
- }
- cur = strsep(&str, ",");
- }
-
- return 1;
-}
-__setup("ipmi_kcs=", ipmi_kcs_setup);
-#endif
-
-MODULE_LICENSE("GPL");
--- diff/drivers/char/ipmi/ipmi_kcs_sm.h 2003-01-16 11:30:40.000000000 +0000
+++ source/drivers/char/ipmi/ipmi_kcs_sm.h 1970-01-01 00:00:00.000000000 +0000
@@ -1,70 +0,0 @@
-/*
- * ipmi_kcs_sm.h
- *
- * State machine for handling IPMI KCS interfaces.
- *
- * Author: MontaVista Software, Inc.
- * Corey Minyard <minyard@mvista.com>
- * source@mvista.com
- *
- * Copyright 2002 MontaVista Software Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the
- * Free Software Foundation; either version 2 of the License, or (at your
- * option) any later version.
- *
- *
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
- * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
- * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
- * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
- * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
- * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
- * OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
- * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR
- * TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
- * USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program; if not, write to the Free Software Foundation, Inc.,
- * 675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-struct kcs_data;
-
-void init_kcs_data(struct kcs_data *kcs,
- unsigned int port,
- unsigned char *addr);
-
-/* Start a new transaction in the state machine. This will return -2
- if the state machine is not idle, -1 if the size is invalid (to
- large or too small), or 0 if the transaction is successfully
- completed. */
-int start_kcs_transaction(struct kcs_data *kcs, char *data, unsigned int size);
-
-/* Return the results after the transaction. This will return -1 if
- the buffer is too small, zero if no transaction is present, or the
- actual length of the result data. */
-int kcs_get_result(struct kcs_data *kcs, unsigned char *data, int length);
-
-enum kcs_result
-{
- KCS_CALL_WITHOUT_DELAY, /* Call the driver again immediately */
- KCS_CALL_WITH_DELAY, /* Delay some before calling again. */
- KCS_TRANSACTION_COMPLETE, /* A transaction is finished. */
- KCS_SM_IDLE, /* The SM is in idle state. */
- KCS_SM_HOSED, /* The hardware violated the state machine. */
- KCS_ATTN /* The hardware is asserting attn and the
- state machine is idle. */
-};
-
-/* Call this periodically (for a polled interface) or upon receiving
- an interrupt (for a interrupt-driven interface). If interrupt
- driven, you should probably poll this periodically when not in idle
- state. This should be called with the time that passed since the
- last call, if it is significant. Time is in microseconds. */
-enum kcs_result kcs_event(struct kcs_data *kcs, long time);
-
-/* Return the size of the KCS structure in bytes. */
-int kcs_size(void);
--- diff/drivers/input/misc/gsc_ps2.c 2003-07-22 17:54:27.000000000 +0000
+++ source/drivers/input/misc/gsc_ps2.c 1970-01-01 00:00:00.000000000 +0000
@@ -1,712 +0,0 @@
-/*
- * drivers/input/misc/gsc_ps2.c
- *
- * Copyright (c) 2002 Laurent Canet <canetl@esiee.fr>
- * Copyright (c) 2002 Thibaut Varene <varenet@esiee.fr>
- *
- * Pieces of code based on linux-2.4's hp_mouse.c & hp_keyb.c
- * Copyright (c) 1999 Alex deVries <adevries@thepuffingroup.com>
- * Copyright (c) 1999-2000 Philipp Rumpf <prumpf@tux.org>
- * Copyright (c) 2000 Xavier Debacker <debackex@esiee.fr>
- * Copyright (c) 2000-2001 Thomas Marteau <marteaut@esiee.fr>
- *
- * HP PS/2 Keyboard, found in PA/RISC Workstations
- * very similar to AT keyboards, but without i8042
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- *
- * STATUS:
- * 11/09: lc: Only basic keyboard is supported, mouse still needs to be done.
- * 11/12: tv: switching iomapping; cleaning code; improving module stuff.
- * 11/13: lc & tv: leds aren't working. auto_repeat/meta are. Generaly good behavior.
- * 11/15: tv: 2AM: leds ARE working !
- * 11/16: tv: 3AM: escaped keycodes emulation *handled*, some keycodes are
- * still deliberately ignored (18), what are they used for ?
- * 11/21: lc: mouse is now working
- * 11/29: tv: first try for error handling in init sequence
- *
- * TODO:
- * Error handling in init sequence
- * SysRq handling
- * Pause key handling
- * Intellimouse & other rodents handling (at least send an error when
- * such a mouse is plugged : it will totally fault)
- * Mouse: set scaling / Dino testing
- * Bug chasing...
- *
- */
-
-#include <linux/input.h>
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/ptrace.h> /* interrupt.h wants struct pt_regs defined */
-#include <linux/interrupt.h>
-#include <linux/sched.h> /* for request_irq/free_irq */
-#include <linux/spinlock.h>
-#include <linux/delay.h>
-#include <linux/ioport.h>
-#include <linux/kd.h>
-#include <linux/pci_ids.h>
-
-#include <asm/irq.h>
-#include <asm/io.h>
-#include <asm/parisc-device.h>
-
-/* Debugging stuff */
-#undef KBD_DEBUG
-#ifdef KBD_DEBUG
- #define DPRINTK(fmt,args...) printk(KERN_DEBUG __FILE__ ":" fmt, ##args)
-#else
- #define DPRINTK(x,...)
-#endif
-
-
-/*
- * Driver constants
- */
-
-/* PS/2 keyboard and mouse constants */
-#define AUX_RECONNECT 0xAA /* PS/2 Mouse end of test successful */
-#define AUX_REPLY_ACK 0xFA
-#define AUX_ENABLE_DEV 0xF4 /* Enables aux device */
-
-/* Order of the mouse bytes coming to the host */
-#define PACKET_X 1
-#define PACKET_Y 2
-#define PACKET_CTRL 0
-
-#define GSC_MOUSE_OFFSET 0x0100 /* offset from keyboard to mouse port */
-#define GSC_DINO_OFFSET 0x800 /* offset for DINO controller versus LASI one */
-
-#define GSC_ID 0x00 /* ID and reset port offsets */
-#define GSC_RESET 0x00
-#define GSC_RCVDATA 0x04 /* receive and transmit port offsets */
-#define GSC_XMTDATA 0x04
-#define GSC_CONTROL 0x08 /* see: control register bits */
-#define GSC_STATUS 0x0C /* see: status register bits */
-
-/* Control register bits */
-#define GSC_CTRL_ENBL 0x01 /* enable interface */
-#define GSC_CTRL_LPBXR 0x02 /* loopback operation */
-#define GSC_CTRL_DIAG 0x20 /* directly control clock/data line */
-#define GSC_CTRL_DATDIR 0x40 /* data line direct control */
-#define GSC_CTRL_CLKDIR 0x80 /* clock line direct control */
-
-/* Status register bits */
-#define GSC_STAT_RBNE 0x01 /* Receive Buffer Not Empty */
-#define GSC_STAT_TBNE 0x02 /* Transmit Buffer Not Empty */
-#define GSC_STAT_TERR 0x04 /* Timeout Error */
-#define GSC_STAT_PERR 0x08 /* Parity Error */
-#define GSC_STAT_CMPINTR 0x10 /* Composite Interrupt */
-#define GSC_STAT_DATSHD 0x40 /* Data Line Shadow */
-#define GSC_STAT_CLKSHD 0x80 /* Clock Line Shadow */
-
-/* Keycode map */
-#define KBD_ESCAPE0 0xe0
-#define KBD_ESCAPE1 0xe1
-#define KBD_RELEASE 0xf0
-#define KBD_ACK 0xfa
-#define KBD_RESEND 0xfe
-#define KBD_UNKNOWN 0
-
-#define KBD_TBLSIZE 512
-
-/* Mouse */
-#define MOUSE_LEFTBTN 0x1
-#define MOUSE_MIDBTN 0x4
-#define MOUSE_RIGHTBTN 0x2
-#define MOUSE_ALWAYS1 0x8
-#define MOUSE_XSIGN 0x10
-#define MOUSE_YSIGN 0x20
-#define MOUSE_XOVFLOW 0x40
-#define MOUSE_YOVFLOW 0x80
-
-/* Remnant of pc_keyb.h */
-#define KBD_CMD_SET_LEDS 0xED /* Sets keyboard leds */
-#define KBD_CMD_SET_RATE 0xF3 /* Sets typematic rate */
-#define KBD_CMD_ENABLE 0xF4 /* Enables scanning */
-#define KBD_CMD_DISABLE 0xF5
-#define KBD_CMD_RESET 0xFF
-
-static unsigned char hpkeyb_keycode[KBD_TBLSIZE] =
-{
- /* 00 */ KBD_UNKNOWN, KEY_F9, KBD_UNKNOWN, KEY_F5, KEY_F3, KEY_F1, KEY_F2, KEY_F12,
- /* 08 */ KBD_UNKNOWN, KEY_F10, KEY_F8, KEY_F6, KEY_F4, KEY_TAB, KEY_GRAVE, KBD_UNKNOWN,
- /* 10 */ KBD_UNKNOWN, KEY_LEFTALT, KEY_LEFTSHIFT, KBD_UNKNOWN, KEY_LEFTCTRL, KEY_Q, KEY_1, KBD_UNKNOWN,
- /* 18 */ KBD_UNKNOWN, KBD_UNKNOWN, KEY_Z, KEY_S, KEY_A, KEY_W, KEY_2, KBD_UNKNOWN,
- /* 20 */ KBD_UNKNOWN, KEY_C, KEY_X, KEY_D, KEY_E, KEY_4, KEY_3, KBD_UNKNOWN,
- /* 28 */ KBD_UNKNOWN, KEY_SPACE, KEY_V, KEY_F, KEY_T, KEY_R, KEY_5, KBD_UNKNOWN,
- /* 30 */ KBD_UNKNOWN, KEY_N, KEY_B, KEY_H, KEY_G, KEY_Y, KEY_6, KBD_UNKNOWN,
- /* 38 */ KBD_UNKNOWN, KBD_UNKNOWN, KEY_M, KEY_J, KEY_U, KEY_7, KEY_8, KBD_UNKNOWN,
- /* 40 */ KBD_UNKNOWN, KEY_COMMA, KEY_K, KEY_I, KEY_O, KEY_0, KEY_9, KBD_UNKNOWN,
- /* 48 */ KBD_UNKNOWN, KEY_DOT, KEY_SLASH, KEY_L, KEY_SEMICOLON, KEY_P, KEY_MINUS, KBD_UNKNOWN,
- /* 50 */ KBD_UNKNOWN, KBD_UNKNOWN, KEY_APOSTROPHE,KBD_UNKNOWN, KEY_LEFTBRACE, KEY_EQUAL, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 58 */ KEY_CAPSLOCK, KEY_RIGHTSHIFT,KEY_ENTER, KEY_RIGHTBRACE,KBD_UNKNOWN, KEY_BACKSLASH,KBD_UNKNOWN, KBD_UNKNOWN,
- /* 60 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KEY_BACKSPACE, KBD_UNKNOWN,
- /* 68 */ KBD_UNKNOWN, KEY_KP1, KBD_UNKNOWN, KEY_KP4, KEY_KP7, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 70 */ KEY_KP0, KEY_KPDOT, KEY_KP2, KEY_KP5, KEY_KP6, KEY_KP8, KEY_ESC, KEY_NUMLOCK,
- /* 78 */ KEY_F11, KEY_KPPLUS, KEY_KP3, KEY_KPMINUS, KEY_KPASTERISK,KEY_KP9, KEY_SCROLLLOCK,KEY_103RD,
- /* 80 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KEY_F7, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 88 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 90 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 98 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* a0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* a8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* b0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* b8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* c0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* c8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* d0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* d8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* e0 */ KBD_ESCAPE0, KBD_ESCAPE1, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* e8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* f0 */ KBD_RELEASE, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* f8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_ACK, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_RESEND, KBD_UNKNOWN,
-/* These are offset for escaped keycodes */
- /* 00 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 08 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 10 */ KBD_UNKNOWN, KEY_RIGHTALT, KBD_UNKNOWN, KBD_UNKNOWN, KEY_RIGHTCTRL, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 18 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 20 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 28 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 30 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 38 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 40 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 48 */ KBD_UNKNOWN, KBD_UNKNOWN, KEY_KPSLASH, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 50 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 58 */ KBD_UNKNOWN, KBD_UNKNOWN, KEY_KPENTER, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 60 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 68 */ KBD_UNKNOWN, KEY_END, KBD_UNKNOWN, KEY_LEFT, KEY_HOME, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 70 */ KEY_INSERT, KEY_DELETE, KEY_DOWN, KBD_UNKNOWN, KEY_RIGHT, KEY_UP, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 78 */ KBD_UNKNOWN, KBD_UNKNOWN, KEY_PAGEDOWN, KBD_UNKNOWN, KEY_SYSRQ, KEY_PAGEUP, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 80 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 88 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 90 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* 98 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* a0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* a8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* b0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* b8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* c0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* c8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* d0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* d8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* e0 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* e8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* f0 */ KBD_RELEASE, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN,
- /* f8 */ KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN, KBD_UNKNOWN
-};
-
-
-/* Keyboard struct */
-static struct {
- struct input_dev dev;
- char * addr;
- unsigned int irq;
- unsigned int scancode;
- unsigned int escaped;
- unsigned int released;
- unsigned int initialized;
-}
-hpkeyb = {
- .escaped = 0,
- .released = 0,
- .initialized = 0
-};
-
-/* Mouse struct */
-static struct {
- struct input_dev dev;
- char * addr;
- unsigned long irq;
- unsigned long initialized;
- int nbread;
- unsigned char bytes[3];
- unsigned long last;
-}
-hpmouse = {
- .initialized = 0,
- .nbread = 0
-};
-
-static spinlock_t gscps2_lock = SPIN_LOCK_UNLOCKED;
-
-
-/*
- * Various HW level routines
- */
-
-#define gscps2_readb_input(x) readb(x+GSC_RCVDATA)
-#define gscps2_readb_control(x) readb(x+GSC_CONTROL)
-#define gscps2_readb_status(x) readb(x+GSC_STATUS)
-#define gscps2_writeb_control(x, y) writeb(x, y+GSC_CONTROL)
-
-static inline void gscps2_writeb_output(u8 val, char * addr)
-{
- int wait = 250; /* Keyboard is expected to react within 250ms */
-
- while (gscps2_readb_status(addr) & GSC_STAT_TBNE) {
- if (!--wait)
- return; /* This should not happen */
- mdelay(1);
- }
- writeb(val, addr+GSC_XMTDATA);
-}
-
-static inline unsigned char gscps2_wait_input(char * addr)
-{
- int wait = 250; /* Keyboard is expected to react within 250ms */
-
- while (!(gscps2_readb_status(addr) & GSC_STAT_RBNE)) {
- if (!--wait)
- return 0; /* This should not happen */
- mdelay(1);
- }
- return gscps2_readb_input(addr);
-}
-
-static int gscps2_writeb_safe_output(u8 val)
-{
- /* This function waits for keyboard's ACK */
- u8 scanread = KBD_UNKNOWN;
- int loop = 5;
-
- while (hpkeyb_keycode[scanread]!=KBD_ACK && --loop > 0) {
- gscps2_writeb_output(val, hpkeyb.addr);
- mdelay(5);
- scanread = gscps2_wait_input(hpkeyb.addr);
- }
-
- if (loop <= 0)
- return -1;
-
- return 0;
-}
-
-/* Reset the PS2 port */
-static void __init gscps2_reset(char * addr)
-{
- /* reset the interface */
- writeb(0xff, addr+GSC_RESET);
- writeb(0x0 , addr+GSC_RESET);
-
- /* enable it */
- gscps2_writeb_control(gscps2_readb_control(addr) | GSC_CTRL_ENBL, addr);
-}
-
-
-/**
- * gscps2_kbd_docode() - PS2 Keyboard basic handler
- *
- * Receives a keyboard scancode, analyses it and sends it to the input layer.
- */
-
-static void gscps2_kbd_docode(struct pt_regs *regs)
-{
- int scancode = gscps2_readb_input(hpkeyb.addr);
- DPRINTK("rel=%d scancode=%d, esc=%d ", hpkeyb.released, scancode, hpkeyb.escaped);
-
- /* Handle previously escaped scancodes */
- if (hpkeyb.escaped == KBD_ESCAPE0)
- scancode |= 0x100; /* jump to the next 256 chars of the table */
-
- switch (hpkeyb_keycode[scancode]) {
- case KBD_RELEASE:
- DPRINTK("release\n");
- hpkeyb.released = 1;
- break;
- case KBD_RESEND:
- DPRINTK("resend request\n");
- break;
- case KBD_ACK:
- DPRINTK("ACK\n");
- break;
- case KBD_ESCAPE0:
- case KBD_ESCAPE1:
- DPRINTK("escape code %d\n", hpkeyb_keycode[scancode]);
- hpkeyb.escaped = hpkeyb_keycode[scancode];
- break;
- case KBD_UNKNOWN:
- DPRINTK("received unknown scancode %d, escape %d.\n",
- scancode, hpkeyb.escaped); /* This is a DPRINTK atm since we do not handle escaped scancodes cleanly */
- if (hpkeyb.escaped)
- hpkeyb.escaped = 0;
- if (hpkeyb.released)
- hpkeyb.released = 0;
- return;
- default:
- hpkeyb.scancode = scancode;
- DPRINTK("sent=%d, rel=%d\n",hpkeyb.scancode, hpkeyb.released);
- /*input_regs(regs);*/
- input_report_key(&hpkeyb.dev, hpkeyb_keycode[hpkeyb.scancode], !hpkeyb.released);
- input_sync(&hpkeyb.dev);
- if (hpkeyb.escaped)
- hpkeyb.escaped = 0;
- if (hpkeyb.released)
- hpkeyb.released = 0;
- break;
- }
-}
-
-
-/**
- * gscps2_mouse_docode() - PS2 Mouse basic handler
- *
- * Receives mouse codes, processes them by packets of three, and sends
- * correct events to the input layer.
- */
-
-static void gscps2_mouse_docode(struct pt_regs *regs)
-{
- int xrel, yrel;
-
- /* process BAT (end of basic tests) command */
- if ((hpmouse.nbread == 1) && (hpmouse.bytes[0] == AUX_RECONNECT))
- hpmouse.nbread--;
-
- /* stolen from psmouse.c */
- if (hpmouse.nbread && time_after(jiffies, hpmouse.last + HZ/2)) {
- printk(KERN_DEBUG "%s:%d : Lost mouse synchronization, throwing %d bytes away.\n", __FILE__, __LINE__,
- hpmouse.nbread);
- hpmouse.nbread = 0;
- }
-
- hpmouse.last = jiffies;
- hpmouse.bytes[hpmouse.nbread++] = gscps2_readb_input(hpmouse.addr);
-
- /* process packet */
- if (hpmouse.nbread == 3) {
-
- if (!(hpmouse.bytes[PACKET_CTRL] & MOUSE_ALWAYS1))
- DPRINTK("Mouse: error on packet always1 bit checking\n");
- /* XXX should exit now, bad data on the line! */
-
- if ((hpmouse.bytes[PACKET_CTRL] & (MOUSE_XOVFLOW | MOUSE_YOVFLOW)))
- DPRINTK("Mouse: position overflow\n");
-
- /*input_regs(regs);*/
-
- input_report_key(&hpmouse.dev, BTN_LEFT, hpmouse.bytes[PACKET_CTRL] & MOUSE_LEFTBTN);
- input_report_key(&hpmouse.dev, BTN_MIDDLE, hpmouse.bytes[PACKET_CTRL] & MOUSE_MIDBTN);
- input_report_key(&hpmouse.dev, BTN_RIGHT, hpmouse.bytes[PACKET_CTRL] & MOUSE_RIGHTBTN);
-
- xrel = hpmouse.bytes[PACKET_X];
- yrel = hpmouse.bytes[PACKET_Y];
-
- /* Data sent by mouse are 9-bit signed, the sign bit is in the control packet */
- if (xrel && (hpmouse.bytes[PACKET_CTRL] & MOUSE_XSIGN))
- xrel = xrel - 0x100;
- if (yrel && (hpmouse.bytes[PACKET_CTRL] & MOUSE_YSIGN))
- yrel = yrel - 0x100;
-
- input_report_rel(&hpmouse.dev, REL_X, xrel);
- input_report_rel(&hpmouse.dev, REL_Y, -yrel); /* Y axis is received upside-down */
-
- input_sync(&hpmouse.dev);
-
- hpmouse.nbread = 0;
- }
-}
-
-
-/**
- * gscps2_interrupt() - Interruption service routine
- *
- * This processes the list of scancodes queued and sends appropriate
- * key value to the system.
- */
-
-static irqreturn_t gscps2_interrupt(int irq, void *dev, struct pt_regs *reg)
-{
- /* process mouse actions */
- while (gscps2_readb_status(hpmouse.addr) & GSC_STAT_RBNE)
- gscps2_mouse_docode(reg);
-
- /* process keyboard scancode */
- while (gscps2_readb_status(hpkeyb.addr) & GSC_STAT_RBNE)
- gscps2_kbd_docode(reg);
-
- return IRQ_HANDLED;
-}
-
-
-/**
- * gscps2_hpkeyb_event() - Event handler
- * @return: success/error report
- *
- * Currently only updates leds on keyboard
- */
-
-int gscps2_hpkeyb_event(struct input_dev *dev, unsigned int type, unsigned int code, int value)
-{
- DPRINTK("Calling %s, type=%d, code=%d, value=%d\n",
- __FUNCTION__, type, code, value);
-
- if (!hpkeyb.initialized)
- return -1;
-
- if (type == EV_LED) {
- u8 leds[2];
-
- if (gscps2_writeb_safe_output(KBD_CMD_SET_LEDS)) {
- printk(KERN_ERR "gsckbd_leds: timeout\n");
- return -1;
- }
- DPRINTK("KBD_CMD_SET_LEDS\n");
-
- *leds = (test_bit(LED_SCROLLL, dev->led) ? LED_SCR : 0)
- | (test_bit(LED_NUML, dev->led) ? LED_NUM : 0)
- | (test_bit(LED_CAPSL, dev->led) ? LED_CAP : 0);
- DPRINTK("Sending leds=%x\n", *leds);
-
- if (gscps2_writeb_safe_output(*leds)) {
- printk(KERN_ERR "gsckbd_leds: timeout\n");
- return -1;
- }
- DPRINTK("leds sent\n");
-
- if (gscps2_writeb_safe_output(KBD_CMD_ENABLE)) {
- printk(KERN_ERR "gsckbd_leds: timeout\n");
- return -1;
- }
- DPRINTK("End\n");
-
- return 0;
-
- }
- return -1;
-}
-
-
-/**
- * gscps2_kbd_probe() - Probes keyboard device and init input_dev structure
- * @return: number of device initialized (1, 0 on error)
- */
-
-static int __init gscps2_kbd_probe(void)
-{
- int i, res = 0;
- unsigned long flags;
-
- if (hpkeyb.initialized) {
- printk(KERN_ERR "GSC PS/2 keyboard driver already registered\n");
- return 0;
- }
-
- spin_lock_irqsave(&gscps2_lock, flags);
-
- if (!gscps2_writeb_safe_output(KBD_CMD_SET_LEDS) &&
- !gscps2_writeb_safe_output(0) &&
- !gscps2_writeb_safe_output(KBD_CMD_ENABLE))
- res = 1;
-
- spin_unlock_irqrestore(&gscps2_lock, flags);
-
- if (!res)
- printk(KERN_ERR "Keyboard initialization sequence failled\n");
-
- init_input_dev(&hpkeyb.dev);
-
- for (i = 0; i < KBD_TBLSIZE; i++)
- if (hpkeyb_keycode[i] != KBD_UNKNOWN)
- set_bit(hpkeyb_keycode[i], hpkeyb.dev.keybit);
-
- hpkeyb.dev.evbit[0] = BIT(EV_KEY) | BIT(EV_LED) | BIT(EV_REP);
- hpkeyb.dev.ledbit[0] = BIT(LED_NUML) | BIT(LED_CAPSL) | BIT(LED_SCROLLL);
- hpkeyb.dev.keycode = hpkeyb_keycode;
- hpkeyb.dev.keycodesize = sizeof(unsigned char);
- hpkeyb.dev.keycodemax = KBD_TBLSIZE;
- hpkeyb.dev.name = "GSC Keyboard";
- hpkeyb.dev.phys = "hpkbd/input0";
-
- hpkeyb.dev.event = gscps2_hpkeyb_event;
-
- /* TODO These need some adjustement, are they really useful ? */
- hpkeyb.dev.id.bustype = BUS_GSC;
- hpkeyb.dev.id.vendor = PCI_VENDOR_ID_HP;
- hpkeyb.dev.id.product = 0x0001;
- hpkeyb.dev.id.version = 0x0010;
- hpkeyb.initialized = 1;
-
- return 1;
-}
-
-
-/**
- * gscps2_mouse_probe() - Probes mouse device and init input_dev structure
- * @return: number of device initialized (1, 0 on error)
- *
- * Currently no check on initialization is performed
- */
-
-static int __init gscps2_mouse_probe(void)
-{
- if (hpmouse.initialized) {
- printk(KERN_ERR "GSC PS/2 Mouse driver already registered\n");
- return 0;
- }
-
- init_input_dev(&hpmouse.dev);
-
- hpmouse.dev.name = "GSC Mouse";
- hpmouse.dev.phys = "hpmouse/input0";
- hpmouse.dev.evbit[0] = BIT(EV_KEY) | BIT(EV_REL);
- hpmouse.dev.keybit[LONG(BTN_MOUSE)] = BIT(BTN_LEFT) | BIT(BTN_MIDDLE) | BIT(BTN_RIGHT);
- hpmouse.dev.relbit[0] = BIT(REL_X) | BIT(REL_Y);
- hpmouse.last = 0;
-
- gscps2_writeb_output(AUX_ENABLE_DEV, hpmouse.addr);
- /* Try it a second time, this will give status if the device is available */
- gscps2_writeb_output(AUX_ENABLE_DEV, hpmouse.addr);
-
- /* TODO These need some adjustement, are they really useful ? */
- hpmouse.dev.id.bustype = BUS_GSC;
- hpmouse.dev.id.vendor = 0x0001;
- hpmouse.dev.id.product = 0x0001;
- hpmouse.dev.id.version = 0x0010;
- hpmouse.initialized = 1;
- return 1; /* XXX: we don't check if initialization failed */
-}
-
-
-/**
- * gscps2_probe() - Probes PS2 devices
- * @return: success/error report
- */
-
-static int __init gscps2_probe(struct parisc_device *dev)
-{
- u8 id;
- char *addr, *name;
- int ret = 0, device_found = 0;
- unsigned long hpa = dev->hpa;
-
- if (!dev->irq)
- goto fail_pitifully;
-
- /* Offset for DINO PS/2. Works with LASI even */
- if (dev->id.sversion == 0x96)
- hpa += GSC_DINO_OFFSET;
-
- addr = ioremap(hpa, 256);
-
- if (!hpmouse.initialized || !hpkeyb.initialized)
- gscps2_reset(addr);
-
- ret = -EINVAL;
- id = readb(addr+GSC_ID) & 0x0f;
- switch (id) {
- case 0: /* keyboard */
- hpkeyb.addr = addr;
- name = "keyboard";
- device_found = gscps2_kbd_probe();
- break;
- case 1: /* mouse */
- hpmouse.addr = addr;
- name = "mouse";
- device_found = gscps2_mouse_probe();
- break;
- default:
- printk(KERN_WARNING "%s: Unsupported PS/2 port (id=%d) ignored\n",
- __FUNCTION__, id);
- goto fail_miserably;
- }
-
- /* No valid device found */
- ret = -ENODEV;
- if (!device_found)
- goto fail_miserably;
-
- /* Here we claim only if we have a device attached */
- /* Allocate the irq and memory region for that device */
- ret = -EBUSY;
- if (request_irq(dev->irq, gscps2_interrupt, 0, name, NULL))
- goto fail_miserably;
-
- if (!request_mem_region(hpa, GSC_STATUS + 4, name))
- goto fail_request_mem;
-
- /* Finalize input struct and register it */
- switch (id) {
- case 0: /* keyboard */
- hpkeyb.irq = dev->irq;
- input_register_device(&hpkeyb.dev);
- break;
- case 1: /* mouse */
- hpmouse.irq = dev->irq;
- input_register_device(&hpmouse.dev);
- break;
- default:
- break;
- }
-
- printk(KERN_INFO "input: PS/2 %s port at 0x%08lx (irq %d) found and attached\n",
- name, hpa, dev->irq);
-
- return 0;
-
-fail_request_mem: free_irq(dev->irq, NULL);
-fail_miserably: iounmap(addr);
-fail_pitifully: return ret;
-}
-
-
-
-static struct parisc_device_id gscps2_device_tbl[] = {
- { HPHW_FIO, HVERSION_REV_ANY_ID, HVERSION_ANY_ID, 0x00084 }, /* LASI PS/2 */
-/* { HPHW_FIO, HVERSION_REV_ANY_ID, HVERSION_ANY_ID, 0x00096 }, DINO PS/2 (XXX Not yet tested) */
- { 0, } /* 0 terminated list */
-};
-
-static struct parisc_driver gscps2_driver = {
- .name = "GSC PS2",
- .id_table = gscps2_device_tbl,
- .probe = gscps2_probe,
-};
-
-static int __init gscps2_init(void)
-{
- if (register_parisc_driver(&gscps2_driver))
- return -EBUSY;
- return 0;
-}
-
-static void __exit gscps2_exit(void)
-{
- /* TODO this is probably not very good and needs to be checked */
- if (hpkeyb.initialized) {
- free_irq(hpkeyb.irq, gscps2_interrupt);
- iounmap(hpkeyb.addr);
- hpkeyb.initialized = 0;
- input_unregister_device(&hpkeyb.dev);
- }
- if (hpmouse.initialized) {
- free_irq(hpmouse.irq, gscps2_interrupt);
- iounmap(hpmouse.addr);
- hpmouse.initialized = 0;
- input_unregister_device(&hpmouse.dev);
- }
- unregister_parisc_driver(&gscps2_driver);
-}
-
-
-MODULE_AUTHOR("Laurent Canet <canetl@esiee.fr>, Thibaut Varene <varenet@esiee.fr>");
-MODULE_DESCRIPTION("GSC PS/2 keyboard/mouse driver");
-MODULE_LICENSE("GPL");
-MODULE_DEVICE_TABLE(parisc, gscps2_device_tbl);
-
-
-module_init(gscps2_init);
-module_exit(gscps2_exit);
--- diff/drivers/net/wan/lmc/lmc_ver.h 2003-09-30 14:46:16.000000000 +0000
+++ source/drivers/net/wan/lmc/lmc_ver.h 1970-01-01 00:00:00.000000000 +0000
@@ -1,123 +0,0 @@
-#include <linux/version.h>
-
-#ifndef _IF_LMC_LINUXVER_
-#define _IF_LMC_LINUXVER_
-
- /*
- * Copyright (c) 1997-2000 LAN Media Corporation (LMC)
- * All rights reserved. www.lanmedia.com
- *
- * This code is written by:
- * Andrew Stanley-Jones (asj@cban.com)
- * Rob Braun (bbraun@vix.com),
- * Michael Graff (explorer@vix.com) and
- * Matt Thomas (matt@3am-software.com).
- *
- * This software may be used and distributed according to the terms
- * of the GNU General Public License version 2, incorporated herein by reference.
- */
-
- /*
- * This file defines and controls all linux version
- * differences.
- *
- * This is being done to keep 1 central location where all linux
- * version differences can be kept and maintained. as this code was
- * found version issues where pepered throughout the source code and
- * made the souce code not only hard to read but version problems hard
- * to track down. If I'm overiding a function/etc with something in
- * this file it will be prefixed by "LMC_" which will mean look
- * here for the version dependent change that's been done.
- *
- */
-
-#if LINUX_VERSION_CODE < 0x20363
-#define net_device device
-#endif
-
-#if LINUX_VERSION_CODE < 0x20363
-#define LMC_XMITTER_BUSY(x) (x)->tbusy = 1
-#define LMC_XMITTER_FREE(x) (x)->tbusy = 0
-#define LMC_XMITTER_INIT(x) (x)->tbusy = 0
-#else
-#define LMC_XMITTER_BUSY(x) netif_stop_queue(x)
-#define LMC_XMITTER_FREE(x) netif_wake_queue(x)
-#define LMC_XMITTER_INIT(x) netif_start_queue(x)
-
-#endif
-
-
-#if LINUX_VERSION_CODE < 0x20100
-//typedef unsigned int u_int32_t;
-
-#define LMC_SETUP_20_DEV {\
- int indx; \
- for (indx = 0; indx < DEV_NUMBUFFS; indx++) \
- skb_queue_head_init (&dev->buffs[indx]); \
- } \
- dev->family = AF_INET; \
- dev->pa_addr = 0; \
- dev->pa_brdaddr = 0; \
- dev->pa_mask = 0xFCFFFFFF; \
- dev->pa_alen = 4; /* IP addr. sizeof(u32) */
-
-#else
-
-#define LMC_SETUP_20_DEV
-
-#endif
-
-
-#if LINUX_VERSION_CODE < 0x20155 /* basically 2.2 plus */
-
-#define LMC_DEV_KFREE_SKB(skb) dev_kfree_skb((skb), FREE_WRITE)
-
-#else /* Mostly 2.0 kernels */
-
-#define LMC_DEV_KFREE_SKB(skb) dev_kfree_skb(skb)
-
-#endif
-
-#if LINUX_VERSION_CODE < 0x20200
-#else
-
-#endif
-
-#if LINUX_VERSION_CODE < 0x20100
-#define LMC_SKB_FREE(skb, val) (skb->free = val)
-#else
-#define LMC_SKB_FREE(skb, val)
-#endif
-
-
-#if (LINUX_VERSION_CODE >= 0x20200)
-
-#define LMC_SPIN_FLAGS unsigned long flags;
-#define LMC_SPIN_LOCK_INIT(x) spin_lock_init(&(x)->lmc_lock);
-#define LMC_SPIN_UNLOCK(x) ((x)->lmc_lock = SPIN_LOCK_UNLOCKED)
-#define LMC_SPIN_LOCK_IRQSAVE(x) spin_lock_irqsave (&(x)->lmc_lock, flags);
-#define LMC_SPIN_UNLOCK_IRQRESTORE(x) spin_unlock_irqrestore (&(x)->lmc_lock, flags);
-#else
-#define LMC_SPIN_FLAGS
-#define LMC_SPIN_LOCK_INIT(x)
-#define LMC_SPIN_UNLOCK(x)
-#define LMC_SPIN_LOCK_IRQSAVE(x)
-#define LMC_SPIN_UNLOCK_IRQRESTORE(x)
-#endif
-
-
-#if LINUX_VERSION_CODE >= 0x20100
-#define LMC_COPY_FROM_USER(x, y, z) if(copy_from_user ((x), (y), (z))) return -EFAULT
-#define LMC_COPY_TO_USER(x, y, z) if(copy_to_user ((x), (y), (z))) return -EFAULT
-#else
-#define LMC_COPY_FROM_USER(x, y, z) if(verify_area(VERIFY_READ, (y), (z))) \
- return -EFAULT; \
- memcpy_fromfs ((x), (y), (z))
-
-#define LMC_COPY_TO_USER(x, y, z) if(verify_area(VERIFY_WRITE, (x), (z))) \
- return -EFAULT; \
- memcpy_tofs ((x), (y), (z))
-#endif
-
-
-#endif
--- diff/drivers/usb/misc/brlvger.c 2003-10-27 09:20:44.000000000 +0000
+++ source/drivers/usb/misc/brlvger.c 1970-01-01 00:00:00.000000000 +0000
@@ -1,1016 +0,0 @@
-/*
- * Tieman Voyager braille display USB driver.
- *
- * Copyright 2001-2002 Stephane Dalton <sdalton@videotron.ca>
- * and St�phane Doyon <s.doyon@videotron.ca>
- * Maintained by St�phane Doyon <s.doyon@videotron.ca>.
- */
-/*
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- */
-/* History:
- * 0.8 April 2002: Integration into the kernel tree.
- * 0.7 October 2001: First public release as a module, distributed with
- * the BRLTTY package (beta versions around 2.99y).
- */
-
-#define DRIVER_VERSION "v0.8"
-#define DATE "April 2002"
-#define DRIVER_AUTHOR \
- "Stephane Dalton <sdalton@videotron.ca> " \
- "and St�phane Doyon <s.doyon@videotron.ca>"
-#define DRIVER_DESC "Tieman Voyager braille display USB driver for Linux 2.4"
-#define DRIVER_SHORTDESC "Voyager"
-
-#define BANNER \
- KERN_INFO DRIVER_SHORTDESC " " DRIVER_VERSION " (" DATE ")\n" \
- KERN_INFO " by " DRIVER_AUTHOR "\n"
-
-static const char longbanner[] = {
- DRIVER_DESC ", " DRIVER_VERSION " (" DATE "), by " DRIVER_AUTHOR
-};
-
-#include <linux/module.h>
-#include <linux/usb.h>
-#include <linux/init.h>
-#include <linux/slab.h>
-#include <linux/sched.h>
-#include <asm/uaccess.h>
-#include <asm/atomic.h>
-#include <linux/poll.h>
-#include <linux/brlvger.h>
-
-MODULE_AUTHOR( DRIVER_AUTHOR );
-MODULE_DESCRIPTION( DRIVER_DESC );
-MODULE_LICENSE("GPL");
-
-/* Module parameters */
-
-static int debug = 1;
-MODULE_PARM(debug, "i");
-MODULE_PARM_DESC(debug, "Debug level, 0-3");
-
-static int write_repeats = 2;
-MODULE_PARM(write_repeats, "i");
-MODULE_PARM_DESC(write_repeats, "Hack: repetitions for command to "
- "display braille pattern");
- /* to get rid of weird extra dots (perhaps only on
- early hardware versions?) */
-
-static int stall_tries = 3;
-MODULE_PARM(stall_tries, "i");
-MODULE_PARM_DESC(stall_tries, "Hack: retransmits of stalled USB "
- "control messages");
- /* broken early hardware versions? */
-
-#define BRLVGER_RAW_VOLTAGE 89
-/* from 0->300V to 255->200V, we are told 265V is normal operating voltage,
- but we don't know the scale. Assuming it is linear. */
-static int raw_voltage = BRLVGER_RAW_VOLTAGE;
-MODULE_PARM(raw_voltage, "i");
-MODULE_PARM_DESC(raw_voltage, "Parameter for the call to SET_DISPLAY_VOLTAGE");
-
-
-/* protocol and display type defines */
-#define MAX_BRLVGER_CELLS 72
-#define MAX_INTERRUPT_DATA 8
-/* control message request types */
-#define BRLVGER_READ_REQ 0xC2
-#define BRLVGER_WRITE_REQ 0x42
-/* control message request codes */
-#define BRLVGER_SET_DISPLAY_ON 0
-#define BRLVGER_SET_DISPLAY_VOLTAGE 1
-#define BRLVGER_GET_SERIAL 3
-#define BRLVGER_GET_HWVERSION 4
-#define BRLVGER_GET_FWVERSION 5
-#define BRLVGER_GET_LENGTH 6
-#define BRLVGER_SEND_BRAILLE 7
-#define BRLVGER_BEEP 9
-#if 0 /* not used and not sure they're working */
-#define BRLVGER_GET_DISPLAY_VOLTAGE 2
-#define BRLVGER_GET_CURRENT 8
-#endif
-
-/* Prototypes */
-static int brlvger_probe (struct usb_interface *intf,
- const struct usb_device_id *id);
-static void brlvger_disconnect(struct usb_interface *intf);
-static int brlvger_open(struct inode *inode, struct file *file);
-static int brlvger_release(struct inode *inode, struct file *file);
-static ssize_t brlvger_write(struct file *file, const char __user *buffer,
- size_t count, loff_t *pos);
-static ssize_t brlvger_read(struct file *file, char __user *buffer,
- size_t count, loff_t *unused_pos);
-static int brlvger_ioctl(struct inode *inode, struct file *file,
- unsigned cmd, unsigned long arg);
-static unsigned brlvger_poll(struct file *file, poll_table *wait);
-static loff_t brlvger_llseek(struct file * file, loff_t offset, int orig);
-static void intr_callback(struct urb *urb, struct pt_regs *regs);
-struct brlvger_priv;
-static int brlvger_get_hw_version(struct brlvger_priv *priv,
- unsigned char *verbuf);
-static int brlvger_get_fw_version(struct brlvger_priv *priv,
- unsigned char *buf);
-static int brlvger_get_serial(struct brlvger_priv *priv,
- unsigned char *buf);
-static int brlvger_get_display_length(struct brlvger_priv *priv);
-static int brlvger_set_display_on_off(struct brlvger_priv *priv, __u16 on);
-static int brlvger_beep(struct brlvger_priv *priv, __u16 duration);
-static int brlvger_set_display_voltage(struct brlvger_priv *priv,
- __u16 voltage);
-static int mycontrolmsg(const char *funcname,
- struct brlvger_priv *priv, unsigned pipe_dir,
- __u8 request, __u8 requesttype, __u16 value,
- __u16 index, void *data, __u16 size);
-
-#define controlmsg(priv,pipe_dir,a,b,c,d,e,f) \
- mycontrolmsg(__FUNCTION__, priv, pipe_dir, \
- a,b,c,d,e,f)
-#define sndcontrolmsg(priv,a,b,c,d,e,f) \
- controlmsg(priv, 0, a,b,c,d,e,f)
-#define rcvcontrolmsg(priv,a,b,c,d,e,f) \
- controlmsg(priv, USB_DIR_IN, a,b,c,d,e,f)
-
-/* ----------------------------------------------------------------------- */
-
-/* Data */
-
-/* key event queue size */
-#define MAX_INTERRUPT_BUFFER 10
-
-/* private state */
-struct brlvger_priv {
- struct usb_device *dev; /* USB device handle */
- struct usb_endpoint_descriptor *in_interrupt;
- struct urb *intr_urb;
-
- int subminor; /* which minor dev #? */
-
- unsigned char hwver[BRLVGER_HWVER_SIZE]; /* hardware version */
- unsigned char fwver[BRLVGER_FWVER_SIZE]; /* firmware version */
- unsigned char serialnum[BRLVGER_SERIAL_SIZE];
-
- int llength; /* logical length */
- int plength; /* physical length */
-
- __u8 obuf[MAX_BRLVGER_CELLS];
- __u8 intr_buff[MAX_INTERRUPT_DATA];
- __u8 event_queue[MAX_INTERRUPT_BUFFER][MAX_INTERRUPT_DATA];
- atomic_t intr_idx, read_idx;
- spinlock_t intr_idx_lock; /* protects intr_idx */
- wait_queue_head_t read_wait;
-
- int opened;
- struct semaphore open_sem; /* protects ->opened */
- struct semaphore dev_sem; /* protects ->dev */
-};
-
-/* Globals */
-
-/* For blocking open */
-static DECLARE_WAIT_QUEUE_HEAD(open_wait);
-
-/* Some print macros */
-#ifdef dbg
-#undef dbg
-#endif
-#ifdef info
-#undef info
-#endif
-#ifdef err
-#undef err
-#endif
-#define info(args...) \
- ({ printk(KERN_INFO "Voyager: " args); \
- printk("\n"); })
-#define err(args...) \
- ({ printk(KERN_ERR "Voyager: " args); \
- printk("\n"); })
-#define dbgprint(fmt, args...) \
- ({ printk(KERN_DEBUG "Voyager: %s: " fmt, __FUNCTION__ , ##args); \
- printk("\n"); })
-#define dbg(args...) \
- ({ if(debug >= 1) dbgprint(args); })
-#define dbg2(args...) \
- ({ if(debug >= 2) dbgprint(args); })
-#define dbg3(args...) \
- ({ if(debug >= 3) dbgprint(args); })
-
-/* ----------------------------------------------------------------------- */
-
-/* Driver registration */
-
-static struct usb_device_id brlvger_ids [] = {
- { USB_DEVICE(0x0798, 0x0001) },
- { } /* Terminating entry */
-};
-MODULE_DEVICE_TABLE (usb, brlvger_ids);
-
-static struct file_operations brlvger_fops =
-{
- .owner = THIS_MODULE,
- .llseek = brlvger_llseek,
- .read = brlvger_read,
- .write = brlvger_write,
- .ioctl = brlvger_ioctl,
- .open = brlvger_open,
- .release = brlvger_release,
- .poll = brlvger_poll,
-};
-
-static struct usb_class_driver brlvger_class = {
- .name = "usb/brlvger%d",
- .fops = &brlvger_fops,
- .mode = S_IFCHR | S_IRUSR |S_IWUSR | S_IRGRP | S_IWGRP,
- .minor_base = BRLVGER_MINOR,
-};
-
-static struct usb_driver brlvger_driver =
-{
- .owner = THIS_MODULE,
- .name = "brlvger",
- .probe = brlvger_probe,
- .disconnect = brlvger_disconnect,
- .id_table = brlvger_ids,
-};
-
-static int
-__init brlvger_init (void)
-{
- int retval;
- printk(BANNER);
-
- if(stall_tries < 1 || write_repeats < 1)
- return -EINVAL;
-
- retval = usb_register(&brlvger_driver);
- if (retval) {
- err("USB registration failed");
- goto out;
- }
-
-out:
- return retval;
-}
-
-static void
-__exit brlvger_cleanup (void)
-{
- usb_deregister (&brlvger_driver);
- dbg("Driver unregistered");
-}
-
-module_init (brlvger_init);
-module_exit (brlvger_cleanup);
-
-/* ----------------------------------------------------------------------- */
-
-/* Probe and disconnect functions */
-
-static int
-brlvger_probe (struct usb_interface *intf,
- const struct usb_device_id *id)
-{
- struct usb_device *dev = interface_to_usbdev(intf);
- struct brlvger_priv *priv = NULL;
- int retval;
- struct usb_endpoint_descriptor *endpoint;
- struct usb_host_interface *actifsettings;
- /* protects against reentrance: once we've found a free slot
- we reserve it.*/
- static DECLARE_MUTEX(reserve_sem);
-
- actifsettings = dev->actconfig->interface[0]->altsetting;
-
- if( dev->descriptor.bNumConfigurations != 1
- || dev->config->desc.bNumInterfaces != 1
- || actifsettings->desc.bNumEndpoints != 1 ) {
- err ("Bogus braille display config info");
- return -ENODEV;
- }
-
- endpoint = &actifsettings->endpoint [0].desc;
- if (!(endpoint->bEndpointAddress & 0x80) ||
- ((endpoint->bmAttributes & 3) != 0x03)) {
- err ("Bogus braille display config info, wrong endpoints");
- return -ENODEV;
- }
-
- down(&reserve_sem);
-
- retval = usb_register_dev(intf, &brlvger_class);
- if (retval) {
- err("Not able to get a minor for this device.");
- goto error;
- }
-
- if( !(priv = kmalloc (sizeof *priv, GFP_KERNEL)) ){
- err("No more memory");
- goto error;
- }
-
- memset(priv, 0, sizeof(*priv));
- atomic_set(&priv->intr_idx, 0);
- atomic_set(&priv->read_idx, MAX_INTERRUPT_BUFFER-1);
- spin_lock_init(&priv->intr_idx_lock);
- init_waitqueue_head(&priv->read_wait);
- /* opened is memset'ed to 0 */
- init_MUTEX(&priv->open_sem);
- init_MUTEX(&priv->dev_sem);
-
- priv->subminor = intf->minor;
-
- /* we found a interrupt in endpoint */
- priv->in_interrupt = endpoint;
-
- priv->dev = dev;
-
- if(brlvger_get_hw_version(priv, priv->hwver) <0) {
- err("Unable to get hardware version");
- goto error;
- }
- dbg("Hw ver %d.%d", priv->hwver[0], priv->hwver[1]);
- if(brlvger_get_fw_version(priv, priv->fwver) <0) {
- err("Unable to get firmware version");
- goto error;
- }
- dbg("Fw ver: %s", priv->fwver);
-
- if(brlvger_get_serial(priv, priv->serialnum) <0) {
- err("Unable to get serial number");
- goto error;
- }
- dbg("Serial number: %s", priv->serialnum);
-
- if( (priv->llength = brlvger_get_display_length(priv)) <0 ){
- err("Unable to get display length");
- goto error;
- }
- switch(priv->llength) {
- case 48:
- priv->plength = 44;
- break;
- case 72:
- priv->plength = 70;
- break;
- default:
- err("Unsupported display length: %d", priv->llength);
- goto error;
- };
- dbg("Display length: %d", priv->plength);
-
- usb_set_intfdata (intf, priv);
- info( "Braille display %d is device major %d minor %d",
- intf->minor, USB_MAJOR, BRLVGER_MINOR + intf->minor);
-
- /* Tell anyone waiting on a blocking open */
- wake_up_interruptible(&open_wait);
-
- goto out;
-
- error:
- if(priv) {
- kfree( priv );
- priv = NULL;
- }
-
- out:
- up(&reserve_sem);
- if (priv) {
- usb_set_intfdata (intf, priv);
- return 0;
- }
- return -EIO;
-}
-
-static void
-brlvger_disconnect(struct usb_interface *intf)
-{
- struct brlvger_priv *priv = usb_get_intfdata (intf);
- int r;
-
- usb_set_intfdata (intf, NULL);
- if(priv){
- info("Display %d disconnecting", priv->subminor);
-
- usb_deregister_dev(intf, &brlvger_class);
-
- down(&priv->open_sem);
- down(&priv->dev_sem);
- if(priv->opened) {
- /* Disable interrupts */
- if((r = usb_unlink_urb(priv->intr_urb)) <0)
- err("usb_unlink_urb returns %d", r);
- usb_free_urb(priv->intr_urb);
- /* mark device as dead and prevent control
- messages to it */
- priv->dev = NULL;
- /* Tell anyone hung up on a read that it
- won't be coming */
- wake_up_interruptible(&priv->read_wait);
- up(&priv->dev_sem);
- up(&priv->open_sem);
- }else
- /* no corresponding up()s */
- kfree(priv);
- }
-}
-
-/* ----------------------------------------------------------------------- */
-
-/* fops implementation */
-
-static int
-brlvger_open(struct inode *inode, struct file *file)
-{
- int devnum = iminor(inode);
- struct usb_interface *intf = NULL;
- struct brlvger_priv *priv = NULL;
- int n, ret;
-
- if (devnum < 0)
- return -ENXIO;
-
- n = devnum - BRLVGER_MINOR;
-
- do {
- intf = usb_find_interface(&brlvger_driver, devnum);
- if (!intf) {
- if (file->f_flags & O_NONBLOCK) {
- dbg3("Failing non-blocking open: "
- "device %d not connected", n);
- return -EAGAIN;
- }
- /* Blocking open. One global wait queue will
- suffice. We wait until a device for the selected
- minor is connected. */
- dbg2("Waiting for device %d to be connected", n);
- ret = wait_event_interruptible(open_wait,
- (intf = usb_find_interface(&brlvger_driver, devnum)));
- if (ret) {
- dbg2("Interrupted wait for device %d", n);
- return ret;
- }
- }
- } while(!intf);
- priv = usb_get_intfdata(intf);
-
- /* We grabbed an existing device. */
- if(down_interruptible(&priv->open_sem))
- return -ERESTARTSYS;
-
- /* Only one process can open each device, no sharing. */
- ret = -EBUSY;
- if(priv->opened)
- goto out;
-
- dbg("Opening display %d", priv->subminor);
-
- /* Setup interrupt handler for receiving key input */
- priv->intr_urb = usb_alloc_urb(0, GFP_KERNEL);
- if(!priv->intr_urb) {
- err("Unable to allocate URB");
- goto out;
- }
- usb_fill_int_urb( priv->intr_urb, priv->dev,
- usb_rcvintpipe(priv->dev,
- priv->in_interrupt->bEndpointAddress),
- priv->intr_buff, sizeof(priv->intr_buff),
- intr_callback, priv, priv->in_interrupt->bInterval);
- if((ret = usb_submit_urb(priv->intr_urb, GFP_KERNEL)) <0){
- err("Error %d while submitting URB", ret);
- goto out;
- }
-
- /* Set voltage */
- if(brlvger_set_display_voltage(priv, raw_voltage) <0) {
- err("Unable to set voltage");
- goto out;
- }
-
- /* Turn display on */
- if((ret = brlvger_set_display_on_off(priv, 1)) <0) {
- err("Error %d while turning display on", ret);
- goto out;
- }
-
- /* Mark as opened, so disconnect cannot free priv. */
- priv->opened = 1;
-
- file->private_data = priv;
-
- ret = 0;
- goto out;
-
- out:
- up(&priv->open_sem);
- return ret;
-}
-
-static int
-brlvger_release(struct inode *inode, struct file *file)
-{
- struct brlvger_priv *priv = file->private_data;
- int r;
-
- /* Turn display off. Safe even if disconnected. */
- brlvger_set_display_on_off(priv, 0);
-
- /* mutex with disconnect and with open */
- down(&priv->open_sem);
-
- if(!priv->dev) {
- dbg("Releasing disconnected device %d", priv->subminor);
- /* no up(&priv->open_sem) */
- kfree(priv);
- }else{
- dbg("Closing display %d", priv->subminor);
- /* Disable interrupts */
- if((r = usb_unlink_urb(priv->intr_urb)) <0)
- err("usb_unlink_urb returns %d", r);
- usb_free_urb(priv->intr_urb);
- priv->opened = 0;
- up(&priv->open_sem);
- }
-
- return 0;
-}
-
-static ssize_t
-brlvger_write(struct file *file, const char __user *buffer,
- size_t count, loff_t *pos)
-{
- struct brlvger_priv *priv = file->private_data;
- char buf[MAX_BRLVGER_CELLS];
- int ret;
- size_t rs;
- loff_t off;
- __u16 written;
-
- if(!priv->dev)
- return -ENOLINK;
-
- off = *pos;
-
- if(off > priv->plength)
- return -ESPIPE;;
-
- rs = priv->plength - off;
-
- if(count > rs)
- count = rs;
- written = count;
-
- if (copy_from_user (buf, buffer, count ) )
- return -EFAULT;
-
- memset(priv->obuf, 0xaa, sizeof(priv->obuf));
-
- /* Firmware supports multiples of 8cells, so some cells are absent
- and for some reason there actually are holes! euurkkk! */
-
- if( priv->plength == 44 ) {
- /* Two ghost cells at the beginning of the display, plus
- two more after the sixth physical cell. */
- if(off > 5) {
- off +=4;
- memcpy(priv->obuf, buf, count);
- }else{
- int firstpart = 6 - off;
-
-#ifdef WRITE_DEBUG
- dbg3("off: %lld, rs: %d, count: %d, firstpart: %d",
- off, rs, count, firstpart);
-#endif
-
- firstpart = (firstpart < count) ? firstpart : count;
-
-#ifdef WRITE_DEBUG
- dbg3("off: %lld", off);
- dbg3("firstpart: %d", firstpart);
-#endif
-
- memcpy(priv->obuf, buf, firstpart);
-
- if(firstpart != count) {
- int secondpart = count - firstpart;
-#ifdef WRITE_DEBUG
- dbg3("secondpart: %d", secondpart);
-#endif
-
- memcpy(priv->obuf+(firstpart+2),
- buf+firstpart, secondpart);
- written +=2;
- }
-
- off +=2;
-
-#ifdef WRITE_DEBUG
- dbg3("off: %lld, rs: %d, count: %d, firstpart: %d, "
- "written: %d", off, rs, count, firstpart, written);
-#endif
- }
- }else{
- /* Two ghost cells at the beginningg of the display. */
- memcpy(priv->obuf, buf, count);
- off += 2;
- }
-
- {
- int repeat = write_repeats;
- /* Dirty hack: sometimes some of the dots are wrong and somehow
- right themselves if the command is repeated. */
- while(repeat--) {
- ret = sndcontrolmsg(priv,
- BRLVGER_SEND_BRAILLE, BRLVGER_WRITE_REQ, 0,
- off, priv->obuf, written);
- if(ret <0)
- return ret;
- }
- }
-
- return count;
-}
-
-static int
-read_index(struct brlvger_priv *priv)
-{
- int intr_idx, read_idx;
-
- read_idx = atomic_read(&priv->read_idx);
- read_idx = ++read_idx == MAX_INTERRUPT_BUFFER ? 0 : read_idx;
-
- intr_idx = atomic_read(&priv->intr_idx);
-
- return(read_idx == intr_idx ? -1 : read_idx);
-}
-
-static ssize_t
-brlvger_read(struct file *file, char __user *buffer,
- size_t count, loff_t *unused_pos)
-{
- struct brlvger_priv *priv = file->private_data;
- int read_idx;
-
- if(count != MAX_INTERRUPT_DATA)
- return -EINVAL;
-
- if(!priv->dev)
- return -ENOLINK;
-
- if((read_idx = read_index(priv)) == -1) {
- /* queue empty */
- if (file->f_flags & O_NONBLOCK)
- return -EAGAIN;
- else{
- int r = wait_event_interruptible(priv->read_wait,
- (!priv->dev || (read_idx = read_index(priv)) != -1));
- if(!priv->dev)
- return -ENOLINK;
- if(r)
- return r;
- if(read_idx == -1)
- /* should not happen */
- return 0;
- }
- }
-
- if (copy_to_user (buffer, priv->event_queue[read_idx], count) )
- return( -EFAULT);
-
- atomic_set(&priv->read_idx, read_idx);
- /* Multiple opens are not allowed. Yet on SMP, two processes could
- read at the same time (on a shared file descriptor); then it is not
- deterministic whether or not they will get duplicates of a key
- event. */
- return MAX_INTERRUPT_DATA;
-}
-
-static int
-brlvger_ioctl(struct inode *inode, struct file *file,
- unsigned cmd, unsigned long arg)
-{
- struct brlvger_priv *priv = file->private_data;
-
- if(!priv->dev)
- return -ENOLINK;
-
- switch(cmd) {
- case BRLVGER_GET_INFO: {
- struct brlvger_info vi;
-
- memset(&vi, 0, sizeof(vi));
- strlcpy(vi.driver_version, DRIVER_VERSION,
- sizeof(vi.driver_version));
- strlcpy(vi.driver_banner, longbanner,
- sizeof(vi.driver_banner));
-
- vi.display_length = priv->plength;
-
- memcpy(&vi.hwver, priv->hwver, BRLVGER_HWVER_SIZE);
- memcpy(&vi.fwver, priv->fwver, BRLVGER_FWVER_SIZE);
- memcpy(&vi.serialnum, priv->serialnum, BRLVGER_SERIAL_SIZE);
-
- if(copy_to_user((void __user *)arg, &vi, sizeof(vi)))
- return -EFAULT;
- return 0;
- }
- case BRLVGER_DISPLAY_ON:
- return brlvger_set_display_on_off(priv, 1);
- case BRLVGER_DISPLAY_OFF:
- return brlvger_set_display_on_off(priv, 0);
- case BRLVGER_BUZZ: {
- __u16 duration;
- if(get_user(duration, (__u16 *)arg))
- return -EFAULT;
- return brlvger_beep(priv, duration);
- }
-
-#if 0 /* Underlying commands don't seem to work for some reason; not clear if
- we'd want to export these anyway. */
- case BRLVGER_SET_VOLTAGE: {
- __u16 voltage;
- if(get_user(voltage, (__u16 *)arg))
- return -EFAULT;
- return brlvger_set_display_voltage(priv, voltage);
- }
- case BRLVGER_GET_VOLTAGE: {
- __u8 voltage;
- int r = brlvger_get_display_voltage(priv);
- if(r <0)
- return r;
- voltage = r;
- if(put_user(voltage, (__u8 *)arg))
- return -EFAULT;
- return 0;
- }
-#endif
- default:
- return -EINVAL;
- };
-}
-
-static loff_t
-brlvger_llseek(struct file *file, loff_t offset, int orig)
-{
- struct brlvger_priv *priv = file->private_data;
-
- if(!priv->dev)
- return -ENOLINK;
-
- switch (orig) {
- case 0:
- /* nothing to do */
- break;
- case 1:
- offset +=file->f_pos;
- break;
- case 2:
- offset += priv->plength;
- default:
- return -EINVAL;
- }
-
- if((offset >= priv->plength) || (offset < 0))
- return -EINVAL;
-
- return (file->f_pos = offset);
-}
-
-static unsigned
-brlvger_poll(struct file *file, poll_table *wait)
-{
- struct brlvger_priv *priv = file->private_data;
-
- if(!priv->dev)
- return POLLERR | POLLHUP;
-
- poll_wait(file, &priv->read_wait, wait);
-
- if(!priv->dev)
- return POLLERR | POLLHUP;
- if(read_index(priv) != -1)
- return POLLIN | POLLRDNORM;
-
- return 0;
-}
-
-static void
-intr_callback(struct urb *urb, struct pt_regs *regs)
-{
- struct brlvger_priv *priv = urb->context;
- int intr_idx, read_idx;
- int status;
-
- switch (urb->status) {
- case 0:
- /* success */
- break;
- case -ECONNRESET:
- case -ENOENT:
- case -ESHUTDOWN:
- /* this urb is terminated, clean up */
- dbg("%s - urb shutting down with status: %d", __FUNCTION__, urb->status);
- return;
- default:
- dbg("%s - nonzero urb status received: %d", __FUNCTION__, urb->status);
- goto exit;
- }
-
- read_idx = atomic_read(&priv->read_idx);
- spin_lock(&priv->intr_idx_lock);
- intr_idx = atomic_read(&priv->intr_idx);
- if(read_idx == intr_idx) {
- dbg2("Queue full, dropping braille display input");
- spin_unlock(&priv->intr_idx_lock);
- goto exit; /* queue full */
- }
-
- memcpy(priv->event_queue[intr_idx], urb->transfer_buffer,
- MAX_INTERRUPT_DATA);
-
- intr_idx = (++intr_idx == MAX_INTERRUPT_BUFFER)? 0 : intr_idx;
- atomic_set(&priv->intr_idx, intr_idx);
- spin_unlock(&priv->intr_idx_lock);
-
- wake_up_interruptible(&priv->read_wait);
-
-exit:
- status = usb_submit_urb (urb, GFP_ATOMIC);
- if (status)
- err ("%s - usb_submit_urb failed with result %d",
- __FUNCTION__, status);
-}
-
-/* ----------------------------------------------------------------------- */
-
-/* Hardware access functions */
-
-static int
-mycontrolmsg(const char *funcname,
- struct brlvger_priv *priv, unsigned pipe_dir,
- __u8 request, __u8 requesttype, __u16 value,
- __u16 index, void *data, __u16 size)
-{
- int ret=0, tries = stall_tries;
-
- /* Make sure the device was not disconnected */
- if(down_interruptible(&priv->dev_sem))
- return -ERESTARTSYS;
- if(!priv->dev) {
- up(&priv->dev_sem);
- return -ENOLINK;
- }
-
- /* Dirty hack for retransmission: stalls and fails all the time
- without this on the hardware we tested. */
- while(tries--) {
- ret = usb_control_msg(priv->dev,
- usb_sndctrlpipe(priv->dev,0) |pipe_dir,
- request, requesttype, value,
- index, data, size,
- HZ);
- if(ret != -EPIPE)
- break;
- dbg2("Stalled, remaining %d tries", tries);
- }
- up(&priv->dev_sem);
- if(ret <0) {
- err("%s: usb_control_msg returns %d",
- funcname, ret);
- return -EIO;
- }
- return 0;
-}
-
-static int
-brlvger_get_hw_version(struct brlvger_priv *priv, unsigned char *verbuf)
-{
- return rcvcontrolmsg(priv,
- BRLVGER_GET_HWVERSION, BRLVGER_READ_REQ, 0,
- 0, verbuf, BRLVGER_HWVER_SIZE);
- /* verbuf should be 2 bytes */
-}
-
-static int
-brlvger_get_fw_version(struct brlvger_priv *priv, unsigned char *buf)
-{
- unsigned char rawbuf[(BRLVGER_FWVER_SIZE-1)*2+2];
- int i, len;
- int r = rcvcontrolmsg(priv,
- BRLVGER_GET_FWVERSION, BRLVGER_READ_REQ, 0,
- 0, rawbuf, sizeof(rawbuf));
- if(r<0)
- return r;
-
- /* If I guess correctly: succession of 16bit words, the string is
- formed of the first byte of each of these words. First byte in
- buffer indicates total length of data; not sure what second byte is
- for. */
- len = rawbuf[0]-2;
- if(len<0)
- len = 0;
- else if(len+1 > BRLVGER_FWVER_SIZE)
- len = BRLVGER_FWVER_SIZE-1;
- for(i=0; i<len; i++)
- buf[i] = rawbuf[2+2*i];
- buf[i] = 0;
- return 0;
-}
-
-static int
-brlvger_get_serial(struct brlvger_priv *priv, unsigned char *buf)
-{
- unsigned char rawserial[BRLVGER_SERIAL_BIN_SIZE];
- int i;
- int r = rcvcontrolmsg(priv,
- BRLVGER_GET_SERIAL, BRLVGER_READ_REQ, 0,
- 0, rawserial, sizeof(rawserial));
- if(r<0)
- return r;
-
- for(i=0; i<BRLVGER_SERIAL_BIN_SIZE; i++) {
-#define NUM_TO_HEX(n) (((n)>9) ? (n)+'A' : (n)+'0')
- buf[2*i] = NUM_TO_HEX(rawserial[i] >>4);
- buf[2*i+1] = NUM_TO_HEX(rawserial[i] &0xf);
- }
- buf[2*i] = 0;
- return 0;
-}
-
-static int
-brlvger_get_display_length(struct brlvger_priv *priv)
-{
- unsigned char data[2];
- int ret = rcvcontrolmsg(priv,
- BRLVGER_GET_LENGTH, BRLVGER_READ_REQ, 0,
- 0, data, 2);
- if(ret<0)
- return ret;
- return data[1];
-}
-
-static int
-brlvger_beep(struct brlvger_priv *priv, __u16 duration)
-{
- return sndcontrolmsg(priv,
- BRLVGER_BEEP, BRLVGER_WRITE_REQ, duration,
- 0, NULL, 0);
-}
-
-static int
-brlvger_set_display_on_off(struct brlvger_priv *priv, __u16 on)
-{
- dbg2("Turning display %s", ((on) ? "on" : "off"));
- return sndcontrolmsg(priv,
- BRLVGER_SET_DISPLAY_ON, BRLVGER_WRITE_REQ, on,
- 0, NULL, 0);
-}
-
-static int
-brlvger_set_display_voltage(struct brlvger_priv *priv, __u16 voltage)
-{
- dbg("SET_DISPLAY_VOLTAGE to %u", voltage);
- return sndcontrolmsg(priv,
- BRLVGER_SET_DISPLAY_VOLTAGE, BRLVGER_WRITE_REQ, voltage,
- 0, NULL, 0);
-}
-
-#if 0 /* Had problems testing these commands. Not particularly useful anyway.*/
-
-static int
-brlvger_get_display_voltage(struct brlvger_priv *priv)
-{
- __u8 voltage = 0;
- int ret = rcvcontrolmsg(priv,
- BRLVGER_GET_DISPLAY_VOLTAGE, BRLVGER_READ_REQ, 0,
- 0, &voltage, 1);
- if(ret<0)
- return ret;
- return voltage;
-}
-
-static int
-brlvger_get_current(struct brlvger_priv *priv)
-{
- unsigned char data;
- int ret = rcvcontrolmsg(priv,
- BRLVGER_GET_CURRENT, BRLVGER_READ_REQ, 0,
- 0, &data, 1);
- if(ret<0)
- return ret;
- return data;
-}
-#endif
--- diff/include/linux/brlvger.h 2003-06-09 13:18:20.000000000 +0000
+++ source/include/linux/brlvger.h 1970-01-01 00:00:00.000000000 +0000
@@ -1,54 +0,0 @@
-/*
- * Tieman Voyager braille display USB driver.
- *
- * Copyright 2001-2002 Stephane Dalton <sdalton@videotron.ca>
- * and St�phane Doyon <s.doyon@videotron.ca>
- * Maintained by St�phane Doyon <s.doyon@videotron.ca>.
- */
-/*
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
- */
-
-#ifndef _LINUX_BRLVGER_H
-#define _LINUX_BRLVGER_H
-
-/* Ioctl request codes */
-#define BRLVGER_GET_INFO 0
-#define BRLVGER_DISPLAY_ON 2
-#define BRLVGER_DISPLAY_OFF 3
-#define BRLVGER_BUZZ 4
-
-/* Base minor for the char devices */
-#define BRLVGER_MINOR 128
-
-/* Size of some fields */
-#define BRLVGER_HWVER_SIZE 2
-#define BRLVGER_FWVER_SIZE 200 /* arbitrary, a long string */
-#define BRLVGER_SERIAL_BIN_SIZE 8
-#define BRLVGER_SERIAL_SIZE ((2*BRLVGER_SERIAL_BIN_SIZE)+1)
-
-struct brlvger_info {
- __u8 driver_version[12];
- __u8 driver_banner[200];
-
- __u32 display_length;
- /* All other char[] fields are strings except this one.
- Hardware version: first byte is major, second byte is minor. */
- __u8 hwver[BRLVGER_HWVER_SIZE];
- __u8 fwver[BRLVGER_FWVER_SIZE];
- __u8 serialnum[BRLVGER_SERIAL_SIZE];
-};
-
-#endif
--- diff/scripts/docproc.c 2003-06-30 09:07:24.000000000 +0000
+++ source/scripts/docproc.c 1970-01-01 00:00:00.000000000 +0000
@@ -1,387 +0,0 @@
-/*
- * docproc is a simple preprocessor for the template files
- * used as placeholders for the kernel internal documentation.
- * docproc is used for documentation-frontend and
- * dependency-generator.
- * The two usages have in common that they require
- * some knowledge of the .tmpl syntax, therefore they
- * are kept together.
- *
- * documentation-frontend
- * Scans the template file and call kernel-doc for
- * all occurrences of ![EIF]file
- * Beforehand each referenced file are scanned for
- * any exported sympols "EXPORT_SYMBOL()" statements.
- * This is used to create proper -function and
- * -nofunction arguments in calls to kernel-doc.
- * Usage: docproc doc file.tmpl
- *
- * dependency-generator:
- * Scans the template file and list all files
- * referenced in a format recognized by make.
- * Usage: docproc depend file.tmpl
- * Writes dependency information to stdout
- * in the following format:
- * file.tmpl src.c src2.c
- * The filenames are obtained from the following constructs:
- * !Efilename
- * !Ifilename
- * !Dfilename
- * !Ffilename
- *
- */
-
-#include <stdio.h>
-#include <stdlib.h>
-#include <string.h>
-#include <ctype.h>
-#include <unistd.h>
-#include <limits.h>
-#include <sys/types.h>
-#include <sys/wait.h>
-
-/* exitstatus is used to keep track of any failing calls to kernel-doc,
- * but execution continues. */
-int exitstatus = 0;
-
-typedef void DFL(char *);
-DFL *defaultline;
-
-typedef void FILEONLY(char * file);
-FILEONLY *internalfunctions;
-FILEONLY *externalfunctions;
-FILEONLY *symbolsonly;
-
-typedef void FILELINE(char * file, char * line);
-FILELINE * singlefunctions;
-FILELINE * entity_system;
-
-#define MAXLINESZ 2048
-#define MAXFILES 250
-#define KERNELDOCPATH "scripts/"
-#define KERNELDOC "kernel-doc"
-#define DOCBOOK "-docbook"
-#define FUNCTION "-function"
-#define NOFUNCTION "-nofunction"
-
-void usage (void)
-{
- fprintf(stderr, "Usage: docproc {doc|depend} file\n");
- fprintf(stderr, "Input is read from file.tmpl. Output is sent to stdout\n");
- fprintf(stderr, "doc: frontend when generating kernel documentation\n");
- fprintf(stderr, "depend: generate list of files referenced within file\n");
-}
-
-/*
- * Execute kernel-doc with parameters givin in svec
- */
-void exec_kernel_doc(char **svec)
-{
- pid_t pid;
- int ret;
- /* Make sure output generated so far are flushed */
- fflush(stdout);
- switch(pid=fork()) {
- case -1:
- perror("fork");
- exit(1);
- case 0:
- execvp(KERNELDOCPATH KERNELDOC, svec);
- perror("exec " KERNELDOCPATH KERNELDOC);
- exit(1);
- default:
- waitpid(pid, &ret ,0);
- }
- if (WIFEXITED(ret))
- exitstatus |= WEXITSTATUS(ret);
- else
- exitstatus = 0xff;
-}
-
-/* Types used to create list of all exported symbols in a number of files */
-struct symbols
-{
- char *name;
-};
-
-struct symfile
-{
- char *filename;
- struct symbols *symbollist;
- int symbolcnt;
-};
-
-struct symfile symfilelist[MAXFILES];
-int symfilecnt = 0;
-
-void add_new_symbol(struct symfile *sym, char * symname)
-{
- sym->symbollist =
- realloc(sym->symbollist, (sym->symbolcnt + 1) * sizeof(char *));
- sym->symbollist[sym->symbolcnt++].name = strdup(symname);
-}
-
-/* Add a filename to the list */
-struct symfile * add_new_file(char * filename)
-{
- symfilelist[symfilecnt++].filename = strdup(filename);
- return &symfilelist[symfilecnt - 1];
-}
-/* Check if file already are present in the list */
-struct symfile * filename_exist(char * filename)
-{
- int i;
- for (i=0; i < symfilecnt; i++)
- if (strcmp(symfilelist[i].filename, filename) == 0)
- return &symfilelist[i];
- return NULL;
-}
-
-/*
- * List all files referenced within the template file.
- * Files are separated by tabs.
- */
-void adddep(char * file) { printf("\t%s", file); }
-void adddep2(char * file, char * line) { line = line; adddep(file); }
-void noaction(char * line) { line = line; }
-void noaction2(char * file, char * line) { file = file; line = line; }
-
-/* Echo the line without further action */
-void printline(char * line) { printf("%s", line); }
-
-/*
- * Find all symbols exported with EXPORT_SYMBOL and EXPORT_SYMBOL_GPL
- * in filename.
- * All symbols located are stored in symfilelist.
- */
-void find_export_symbols(char * filename)
-{
- FILE * fp;
- struct symfile *sym;
- char line[MAXLINESZ];
- if (filename_exist(filename) == NULL) {
- sym = add_new_file(filename);
- fp = fopen(filename, "r");
- if (fp == NULL)
- {
- fprintf(stderr, "docproc: ");
- perror(filename);
- }
- while(fgets(line, MAXLINESZ, fp)) {
- char *p;
- char *e;
- if (((p = strstr(line, "EXPORT_SYMBOL_GPL")) != 0) ||
- ((p = strstr(line, "EXPORT_SYMBOL")) != 0)) {
- /* Skip EXPORT_SYMBOL{_GPL} */
- while (isalnum(*p) || *p == '_')
- p++;
- /* Remove paranteses and additional ws */
- while (isspace(*p))
- p++;
- if (*p != '(')
- continue; /* Syntax error? */
- else
- p++;
- while (isspace(*p))
- p++;
- e = p;
- while (isalnum(*e) || *e == '_')
- e++;
- *e = '\0';
- add_new_symbol(sym, p);
- }
- }
- fclose(fp);
- }
-}
-
-/*
- * Document all external or internal functions in a file.
- * Call kernel-doc with following parameters:
- * kernel-doc -docbook -nofunction function_name1 filename
- * function names are obtained from all the the src files
- * by find_export_symbols.
- * intfunc uses -nofunction
- * extfunc uses -function
- */
-void docfunctions(char * filename, char * type)
-{
- int i,j;
- int symcnt = 0;
- int idx = 0;
- char **vec;
-
- for (i=0; i <= symfilecnt; i++)
- symcnt += symfilelist[i].symbolcnt;
- vec = malloc((2 + 2 * symcnt + 2) * sizeof(char*));
- if (vec == NULL) {
- perror("docproc: ");
- exit(1);
- }
- vec[idx++] = KERNELDOC;
- vec[idx++] = DOCBOOK;
- for (i=0; i < symfilecnt; i++) {
- struct symfile * sym = &symfilelist[i];
- for (j=0; j < sym->symbolcnt; j++) {
- vec[idx++] = type;
- vec[idx++] = sym->symbollist[j].name;
- }
- }
- vec[idx++] = filename;
- vec[idx] = NULL;
- printf("<!-- %s -->\n", filename);
- exec_kernel_doc(vec);
- fflush(stdout);
- free(vec);
-}
-void intfunc(char * filename) { docfunctions(filename, NOFUNCTION); }
-void extfunc(char * filename) { docfunctions(filename, FUNCTION); }
-
-/*
- * Document sp�ecific function(s) in a file.
- * Call kernel-doc with the following parameters:
- * kernel-doc -docbook -function function1 [-function function2]
- */
-void singfunc(char * filename, char * line)
-{
- char *vec[200]; /* Enough for specific functions */
- int i, idx = 0;
- int startofsym = 1;
- vec[idx++] = KERNELDOC;
- vec[idx++] = DOCBOOK;
-
- /* Split line up in individual parameters preceeded by FUNCTION */
- for (i=0; line[i]; i++) {
- if (isspace(line[i])) {
- line[i] = '\0';
- startofsym = 1;
- continue;
- }
- if (startofsym) {
- startofsym = 0;
- vec[idx++] = FUNCTION;
- vec[idx++] = &line[i];
- }
- }
- vec[idx++] = filename;
- vec[idx] = NULL;
- exec_kernel_doc(vec);
-}
-
-/*
- * Parse file, calling action specific functions for:
- * 1) Lines containing !E
- * 2) Lines containing !I
- * 3) Lines containing !D
- * 4) Lines containing !F
- * 5) Default lines - lines not matching the above
- */
-void parse_file(FILE *infile)
-{
- char line[MAXLINESZ];
- char * s;
- while(fgets(line, MAXLINESZ, infile)) {
- if (line[0] == '!') {
- s = line + 2;
- switch (line[1]) {
- case 'E':
- while (*s && !isspace(*s)) s++;
- *s = '\0';
- externalfunctions(line+2);
- break;
- case 'I':
- while (*s && !isspace(*s)) s++;
- *s = '\0';
- internalfunctions(line+2);
- break;
- case 'D':
- while (*s && !isspace(*s)) s++;
- *s = '\0';
- symbolsonly(line+2);
- break;
- case 'F':
- /* filename */
- while (*s && !isspace(*s)) s++;
- *s++ = '\0';
- /* function names */
- while (isspace(*s))
- s++;
- singlefunctions(line +2, s);
- break;
- default:
- defaultline(line);
- }
- }
- else {
- defaultline(line);
- }
- }
- fflush(stdout);
-}
-
-
-int main(int argc, char *argv[])
-{
- FILE * infile;
- if (argc != 3) {
- usage();
- exit(1);
- }
- /* Open file, exit on error */
- infile = fopen(argv[2], "r");
- if (infile == NULL) {
- fprintf(stderr, "docproc: ");
- perror(argv[2]);
- exit(2);
- }
-
- if (strcmp("doc", argv[1]) == 0)
- {
- /* Need to do this in two passes.
- * First pass is used to collect all symbols exported
- * in the various files.
- * Second pass generate the documentation.
- * This is required because function are declared
- * and exported in different files :-((
- */
- /* Collect symbols */
- defaultline = noaction;
- internalfunctions = find_export_symbols;
- externalfunctions = find_export_symbols;
- symbolsonly = find_export_symbols;
- singlefunctions = noaction2;
- parse_file(infile);
-
- /* Rewind to start from beginning of file again */
- fseek(infile, 0, SEEK_SET);
- defaultline = printline;
- internalfunctions = intfunc;
- externalfunctions = extfunc;
- symbolsonly = printline;
- singlefunctions = singfunc;
-
- parse_file(infile);
- }
- else if (strcmp("depend", argv[1]) == 0)
- {
- /* Create first part of dependency chain
- * file.tmpl */
- printf("%s\t", argv[2]);
- defaultline = noaction;
- internalfunctions = adddep;
- externalfunctions = adddep;
- symbolsonly = adddep;
- singlefunctions = adddep2;
- parse_file(infile);
- printf("\n");
- }
- else
- {
- fprintf(stderr, "Unknown option: %s\n", argv[1]);
- exit(1);
- }
- fclose(infile);
- fflush(stdout);
- return exitstatus;
-}
-
--- diff/scripts/fixdep.c 2003-06-09 13:18:21.000000000 +0000
+++ source/scripts/fixdep.c 1970-01-01 00:00:00.000000000 +0000
@@ -1,381 +0,0 @@
-/*
- * "Optimize" a list of dependencies as spit out by gcc -MD
- * for the kernel build
- * ===========================================================================
- *
- * Author Kai Germaschewski
- * Copyright 2002 by Kai Germaschewski <kai.germaschewski@gmx.de>
- *
- * This software may be used and distributed according to the terms
- * of the GNU General Public License, incorporated herein by reference.
- *
- *
- * Introduction:
- *
- * gcc produces a very nice and correct list of dependencies which
- * tells make when to remake a file.
- *
- * To use this list as-is however has the drawback that virtually
- * every file in the kernel includes <linux/config.h> which then again
- * includes <linux/autoconf.h>
- *
- * If the user re-runs make *config, linux/autoconf.h will be
- * regenerated. make notices that and will rebuild every file which
- * includes autoconf.h, i.e. basically all files. This is extremely
- * annoying if the user just changed CONFIG_HIS_DRIVER from n to m.
- *
- * So we play the same trick that "mkdep" played before. We replace
- * the dependency on linux/autoconf.h by a dependency on every config
- * option which is mentioned in any of the listed prequisites.
- *
- * To be exact, split-include populates a tree in include/config/,
- * e.g. include/config/his/driver.h, which contains the #define/#undef
- * for the CONFIG_HIS_DRIVER option.
- *
- * So if the user changes his CONFIG_HIS_DRIVER option, only the objects
- * which depend on "include/linux/config/his/driver.h" will be rebuilt,
- * so most likely only his driver ;-)
- *
- * The idea above dates, by the way, back to Michael E Chastain, AFAIK.
- *
- * So to get dependencies right, there two issues:
- * o if any of the files the compiler read changed, we need to rebuild
- * o if the command line given to the compile the file changed, we
- * better rebuild as well.
- *
- * The former is handled by using the -MD output, the later by saving
- * the command line used to compile the old object and comparing it
- * to the one we would now use.
- *
- * Again, also this idea is pretty old and has been discussed on
- * kbuild-devel a long time ago. I don't have a sensibly working
- * internet connection right now, so I rather don't mention names
- * without double checking.
- *
- * This code here has been based partially based on mkdep.c, which
- * says the following about its history:
- *
- * Copyright abandoned, Michael Chastain, <mailto:mec@shout.net>.
- * This is a C version of syncdep.pl by Werner Almesberger.
- *
- *
- * It is invoked as
- *
- * fixdep <depfile> <target> <cmdline>
- *
- * and will read the dependency file <depfile>
- *
- * The transformed dependency snipped is written to stdout.
- *
- * It first generates a line
- *
- * cmd_<target> = <cmdline>
- *
- * and then basically copies the .<target>.d file to stdout, in the
- * process filtering out the dependency on linux/autoconf.h and adding
- * dependencies on include/config/my/option.h for every
- * CONFIG_MY_OPTION encountered in any of the prequisites.
- *
- * It will also filter out all the dependencies on *.ver. We need
- * to make sure that the generated version checksum are globally up
- * to date before even starting the recursive build, so it's too late
- * at this point anyway.
- *
- * The algorithm to grep for "CONFIG_..." is bit unusual, but should
- * be fast ;-) We don't even try to really parse the header files, but
- * merely grep, i.e. if CONFIG_FOO is mentioned in a comment, it will
- * be picked up as well. It's not a problem with respect to
- * correctness, since that can only give too many dependencies, thus
- * we cannot miss a rebuild. Since people tend to not mention totally
- * unrelated CONFIG_ options all over the place, it's not an
- * efficiency problem either.
- *
- * (Note: it'd be easy to port over the complete mkdep state machine,
- * but I don't think the added complexity is worth it)
- */
-
-#include <sys/types.h>
-#include <sys/stat.h>
-#include <sys/mman.h>
-#include <unistd.h>
-#include <fcntl.h>
-#include <string.h>
-#include <stdlib.h>
-#include <stdio.h>
-#include <limits.h>
-#include <ctype.h>
-#include <netinet/in.h>
-
-#define INT_CONF ntohl(0x434f4e46)
-#define INT_ONFI ntohl(0x4f4e4649)
-#define INT_NFIG ntohl(0x4e464947)
-#define INT_FIG_ ntohl(0x4649475f)
-
-char *target;
-char *depfile;
-char *cmdline;
-
-void usage(void)
-
-{
- fprintf(stderr, "Usage: fixdep <depfile> <target> <cmdline>\n");
- exit(1);
-}
-
-void print_cmdline(void)
-{
- printf("cmd_%s := %s\n\n", target, cmdline);
-}
-
-char * str_config = NULL;
-int size_config = 0;
-int len_config = 0;
-
-/*
- * Grow the configuration string to a desired length.
- * Usually the first growth is plenty.
- */
-void grow_config(int len)
-{
- while (len_config + len > size_config) {
- if (size_config == 0)
- size_config = 2048;
- str_config = realloc(str_config, size_config *= 2);
- if (str_config == NULL)
- { perror("fixdep:malloc"); exit(1); }
- }
-}
-
-
-
-/*
- * Lookup a value in the configuration string.
- */
-int is_defined_config(const char * name, int len)
-{
- const char * pconfig;
- const char * plast = str_config + len_config - len;
- for ( pconfig = str_config + 1; pconfig < plast; pconfig++ ) {
- if (pconfig[ -1] == '\n'
- && pconfig[len] == '\n'
- && !memcmp(pconfig, name, len))
- return 1;
- }
- return 0;
-}
-
-/*
- * Add a new value to the configuration string.
- */
-void define_config(const char * name, int len)
-{
- grow_config(len + 1);
-
- memcpy(str_config+len_config, name, len);
- len_config += len;
- str_config[len_config++] = '\n';
-}
-
-/*
- * Clear the set of configuration strings.
- */
-void clear_config(void)
-{
- len_config = 0;
- define_config("", 0);
-}
-
-/*
- * Record the use of a CONFIG_* word.
- */
-void use_config(char *m, int slen)
-{
- char s[PATH_MAX];
- char *p;
-
- if (is_defined_config(m, slen))
- return;
-
- define_config(m, slen);
-
- memcpy(s, m, slen); s[slen] = 0;
-
- for (p = s; p < s + slen; p++) {
- if (*p == '_')
- *p = '/';
- else
- *p = tolower((unsigned char)*p);
- }
- printf(" $(wildcard include/config/%s.h) \\\n", s);
-}
-
-void parse_config_file(char *map, size_t len)
-{
- int *end = (int *) (map + len);
- /* start at +1, so that p can never be < map */
- int *m = (int *) map + 1;
- char *p, *q;
-
- for (; m < end; m++) {
- if (*m == INT_CONF) { p = (char *) m ; goto conf; }
- if (*m == INT_ONFI) { p = (char *) m-1; goto conf; }
- if (*m == INT_NFIG) { p = (char *) m-2; goto conf; }
- if (*m == INT_FIG_) { p = (char *) m-3; goto conf; }
- continue;
- conf:
- if (p > map + len - 7)
- continue;
- if (memcmp(p, "CONFIG_", 7))
- continue;
- for (q = p + 7; q < map + len; q++) {
- if (!(isalnum(*q) || *q == '_'))
- goto found;
- }
- continue;
-
- found:
- use_config(p+7, q-p-7);
- }
-}
-
-/* test is s ends in sub */
-int strrcmp(char *s, char *sub)
-{
- int slen = strlen(s);
- int sublen = strlen(sub);
-
- if (sublen > slen)
- return 1;
-
- return memcmp(s + slen - sublen, sub, sublen);
-}
-
-void do_config_file(char *filename)
-{
- struct stat st;
- int fd;
- void *map;
-
- fd = open(filename, O_RDONLY);
- if (fd < 0) {
- fprintf(stderr, "fixdep: ");
- perror(filename);
- exit(2);
- }
- fstat(fd, &st);
- if (st.st_size == 0) {
- close(fd);
- return;
- }
- map = mmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fd, 0);
- if ((long) map == -1) {
- perror("fixdep: mmap");
- close(fd);
- return;
- }
-
- parse_config_file(map, st.st_size);
-
- munmap(map, st.st_size);
-
- close(fd);
-}
-
-void parse_dep_file(void *map, size_t len)
-{
- char *m = map;
- char *end = m + len;
- char *p;
- char s[PATH_MAX];
-
- p = strchr(m, ':');
- if (!p) {
- fprintf(stderr, "fixdep: parse error\n");
- exit(1);
- }
- memcpy(s, m, p-m); s[p-m] = 0;
- printf("deps_%s := \\\n", target);
- m = p+1;
-
- clear_config();
-
- while (m < end) {
- while (m < end && (*m == ' ' || *m == '\\' || *m == '\n'))
- m++;
- p = m;
- while (p < end && *p != ' ') p++;
- if (p == end) {
- do p--; while (!isalnum(*p));
- p++;
- }
- memcpy(s, m, p-m); s[p-m] = 0;
- if (strrcmp(s, "include/linux/autoconf.h") &&
- strrcmp(s, ".ver")) {
- printf(" %s \\\n", s);
- do_config_file(s);
- }
- m = p + 1;
- }
- printf("\n%s: $(deps_%s)\n\n", target, target);
- printf("$(deps_%s):\n", target);
-}
-
-void print_deps(void)
-{
- struct stat st;
- int fd;
- void *map;
-
- fd = open(depfile, O_RDONLY);
- if (fd < 0) {
- fprintf(stderr, "fixdep: ");
- perror(depfile);
- exit(2);
- }
- fstat(fd, &st);
- if (st.st_size == 0) {
- fprintf(stderr,"fixdep: %s is empty\n",depfile);
- close(fd);
- return;
- }
- map = mmap(NULL, st.st_size, PROT_READ, MAP_PRIVATE, fd, 0);
- if ((long) map == -1) {
- perror("fixdep: mmap");
- close(fd);
- return;
- }
-
- parse_dep_file(map, st.st_size);
-
- munmap(map, st.st_size);
-
- close(fd);
-}
-
-void traps(void)
-{
- static char test[] __attribute__((aligned(sizeof(int)))) = "CONF";
-
- if (*(int *)test != INT_CONF) {
- fprintf(stderr, "fixdep: sizeof(int) != 4 or wrong endianess? %#x\n",
- *(int *)test);
- exit(2);
- }
-}
-
-int main(int argc, char *argv[])
-{
- traps();
-
- if (argc != 4)
- usage();
-
- depfile = argv[1];
- target = argv[2];
- cmdline = argv[3];
-
- print_cmdline();
- print_deps();
-
- return 0;
-}
--- diff/scripts/split-include.c 2002-10-16 03:27:52.000000000 +0000
+++ source/scripts/split-include.c 1970-01-01 00:00:00.000000000 +0000
@@ -1,226 +0,0 @@
-/*
- * split-include.c
- *
- * Copyright abandoned, Michael Chastain, <mailto:mec@shout.net>.
- * This is a C version of syncdep.pl by Werner Almesberger.
- *
- * This program takes autoconf.h as input and outputs a directory full
- * of one-line include files, merging onto the old values.
- *
- * Think of the configuration options as key-value pairs. Then there
- * are five cases:
- *
- * key old value new value action
- *
- * KEY-1 VALUE-1 VALUE-1 leave file alone
- * KEY-2 VALUE-2A VALUE-2B write VALUE-2B into file
- * KEY-3 - VALUE-3 write VALUE-3 into file
- * KEY-4 VALUE-4 - write an empty file
- * KEY-5 (empty) - leave old empty file alone
- */
-
-#include <sys/stat.h>
-#include <sys/types.h>
-
-#include <ctype.h>
-#include <errno.h>
-#include <fcntl.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include <string.h>
-#include <unistd.h>
-
-#define ERROR_EXIT(strExit) \
- { \
- const int errnoSave = errno; \
- fprintf(stderr, "%s: ", str_my_name); \
- errno = errnoSave; \
- perror((strExit)); \
- exit(1); \
- }
-
-
-
-int main(int argc, const char * argv [])
-{
- const char * str_my_name;
- const char * str_file_autoconf;
- const char * str_dir_config;
-
- FILE * fp_config;
- FILE * fp_target;
- FILE * fp_find;
-
- int buffer_size;
-
- char * line;
- char * old_line;
- char * list_target;
- char * ptarget;
-
- struct stat stat_buf;
-
- /* Check arg count. */
- if (argc != 3)
- {
- fprintf(stderr, "%s: wrong number of arguments.\n", argv[0]);
- exit(1);
- }
-
- str_my_name = argv[0];
- str_file_autoconf = argv[1];
- str_dir_config = argv[2];
-
- /* Find a buffer size. */
- if (stat(str_file_autoconf, &stat_buf) != 0)
- ERROR_EXIT(str_file_autoconf);
- buffer_size = 2 * stat_buf.st_size + 4096;
-
- /* Allocate buffers. */
- if ( (line = malloc(buffer_size)) == NULL
- || (old_line = malloc(buffer_size)) == NULL
- || (list_target = malloc(buffer_size)) == NULL )
- ERROR_EXIT(str_file_autoconf);
-
- /* Open autoconfig file. */
- if ((fp_config = fopen(str_file_autoconf, "r")) == NULL)
- ERROR_EXIT(str_file_autoconf);
-
- /* Make output directory if needed. */
- if (stat(str_dir_config, &stat_buf) != 0)
- {
- if (mkdir(str_dir_config, 0755) != 0)
- ERROR_EXIT(str_dir_config);
- }
-
- /* Change to output directory. */
- if (chdir(str_dir_config) != 0)
- ERROR_EXIT(str_dir_config);
-
- /* Put initial separator into target list. */
- ptarget = list_target;
- *ptarget++ = '\n';
-
- /* Read config lines. */
- while (fgets(line, buffer_size, fp_config))
- {
- const char * str_config;
- int is_same;
- int itarget;
-
- if (line[0] != '#')
- continue;
- if ((str_config = strstr(line, "CONFIG_")) == NULL)
- continue;
-
- /* Make the output file name. */
- str_config += sizeof("CONFIG_") - 1;
- for (itarget = 0; !isspace(str_config[itarget]); itarget++)
- {
- int c = (unsigned char) str_config[itarget];
- if (isupper(c)) c = tolower(c);
- if (c == '_') c = '/';
- ptarget[itarget] = c;
- }
- ptarget[itarget++] = '.';
- ptarget[itarget++] = 'h';
- ptarget[itarget++] = '\0';
-
- /* Check for existing file. */
- is_same = 0;
- if ((fp_target = fopen(ptarget, "r")) != NULL)
- {
- fgets(old_line, buffer_size, fp_target);
- if (fclose(fp_target) != 0)
- ERROR_EXIT(ptarget);
- if (!strcmp(line, old_line))
- is_same = 1;
- }
-
- if (!is_same)
- {
- /* Auto-create directories. */
- int islash;
- for (islash = 0; islash < itarget; islash++)
- {
- if (ptarget[islash] == '/')
- {
- ptarget[islash] = '\0';
- if (stat(ptarget, &stat_buf) != 0
- && mkdir(ptarget, 0755) != 0)
- ERROR_EXIT( ptarget );
- ptarget[islash] = '/';
- }
- }
-
- /* Write the file. */
- if ((fp_target = fopen(ptarget, "w" )) == NULL)
- ERROR_EXIT(ptarget);
- fputs(line, fp_target);
- if (ferror(fp_target) || fclose(fp_target) != 0)
- ERROR_EXIT(ptarget);
- }
-
- /* Update target list */
- ptarget += itarget;
- *(ptarget-1) = '\n';
- }
-
- /*
- * Close autoconfig file.
- * Terminate the target list.
- */
- if (fclose(fp_config) != 0)
- ERROR_EXIT(str_file_autoconf);
- *ptarget = '\0';
-
- /*
- * Fix up existing files which have no new value.
- * This is Case 4 and Case 5.
- *
- * I re-read the tree and filter it against list_target.
- * This is crude. But it avoids data copies. Also, list_target
- * is compact and contiguous, so it easily fits into cache.
- *
- * Notice that list_target contains strings separated by \n,
- * with a \n before the first string and after the last.
- * fgets gives the incoming names a terminating \n.
- * So by having an initial \n, strstr will find exact matches.
- */
-
- fp_find = popen("find * -type f -name \"*.h\" -print", "r");
- if (fp_find == 0)
- ERROR_EXIT( "find" );
-
- line[0] = '\n';
- while (fgets(line+1, buffer_size, fp_find))
- {
- if (strstr(list_target, line) == NULL)
- {
- /*
- * This is an old file with no CONFIG_* flag in autoconf.h.
- */
-
- /* First strip the \n. */
- line[strlen(line)-1] = '\0';
-
- /* Grab size. */
- if (stat(line+1, &stat_buf) != 0)
- ERROR_EXIT(line);
-
- /* If file is not empty, make it empty and give it a fresh date. */
- if (stat_buf.st_size != 0)
- {
- if ((fp_target = fopen(line+1, "w")) == NULL)
- ERROR_EXIT(line);
- if (fclose(fp_target) != 0)
- ERROR_EXIT(line);
- }
- }
- }
-
- if (pclose(fp_find) != 0)
- ERROR_EXIT("find");
-
- return 0;
-}