14 years ago[PATCH] CRIS update: IRQ
Mikael Starvik [Wed, 27 Jul 2005 18:44:36 +0000]
[PATCH] CRIS update: IRQ

Use the generic IRQ framework

Signed-off-by: Mikael Starvik <starvik@axis.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] CRIS update: I/O and DMA allocator
Mikael Starvik [Wed, 27 Jul 2005 18:44:35 +0000]
[PATCH] CRIS update: I/O and DMA allocator

Added I/O and DMA allocators to be used by drivers.

Signed-off-by: Mikael Starvik <starvik@axis.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] CRIS update: drivers
Mikael Starvik [Wed, 27 Jul 2005 18:44:34 +0000]
[PATCH] CRIS update: drivers

Updates to device drivers.

* Use I/O and DMA allocators.
* Use wait_event_interruptible instead of interrutiple_sleep_on.
* Added spinlocks SMP.
* Changed restore_flags to local_irq_restore etc.
* Updated IDE driver include to fit 2.6.12.

Signed-off-by: Mikael Starvik <starvik@axis.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] CRIS update: debug
Mikael Starvik [Wed, 27 Jul 2005 18:44:32 +0000]
[PATCH] CRIS update: debug

Improvements to crash debug code.

Signed-off-by: Mikael Starvik <starvik@axis.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] CRIS update: console
Mikael Starvik [Wed, 27 Jul 2005 18:44:31 +0000]
[PATCH] CRIS update: console

Changes to console.

* Added LF->CRLF translation
* Make use of real console framework.

Signed-off-by: Mikael Starvik <starvik@axis.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] CRIS update: configuration and build
Mikael Starvik [Wed, 27 Jul 2005 18:44:30 +0000]
[PATCH] CRIS update: configuration and build

Changes to configuration and build system.

* Added v32 sub architecture.
* Use generic hard IRQ.
* Added SMP options.
* Added options to OOPS at NMI and reboot at OOM.
* Made it possible to set objtree.
* Added option to select Kernel GDB serial port.
* Corrected Kconfig usage.
* Added system profiler.

Signed-off-by: Mikael Starvik <starvik@axis.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] CRIS update: arch split
Mikael Starvik [Wed, 27 Jul 2005 18:44:27 +0000]
[PATCH] CRIS update: arch split

Changes necessary to make the sub-arch split complete.

Signed-off-by: Mikael Starvik <starvik@axis.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] m32r: add missing Kconfig help text
Adrian Bunk [Wed, 27 Jul 2005 18:44:24 +0000]
[PATCH] m32r: add missing Kconfig help text

There's no help text for CONFIG_DEBUG_STACKOVERFLOW - add one.

Signed-off-by: Jesper Juhl <juhl-lkml@dif.dk>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] i386: add missing Kconfig help text
Adrian Bunk [Wed, 27 Jul 2005 18:44:23 +0000]
[PATCH] i386: add missing Kconfig help text

There's no help text for CONFIG_DEBUG_STACKOVERFLOW - add one.

Signed-off-by: Jesper Juhl <juhl-lkml@dif.dk>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] mips: remove obsolete GIU driver for vr41xx
Yoichi Yuasa [Wed, 27 Jul 2005 18:44:22 +0000]
[PATCH] mips: remove obsolete GIU driver for vr41xx

This patch has removed obsolete GIU driver for vr41xx.  This patch already
has been applied to mips tree.

Signed-off-by: Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc64: remove another fixed address constraint
David Gibson [Wed, 27 Jul 2005 18:44:21 +0000]
[PATCH] ppc64: remove another fixed address constraint

Presently the LparMap, one of the structures the kernel shares with the
legacy iSeries hypervisor has a fixed offset address in head.S.  This patch
changes this so the LparMap is a normally initialized structure, without
fixed address.  This allows us to use macros to compute some of the values
in the structure, which wasn't previously possible because the assembler
always uses signed-% which gets the wrong answers for the computations in
question.

Unfortunately, a gcc bug means that doing this requires another structure
(hvReleaseData) to be initialized in asm instead of C, but on the whole the
result is cleaner than before.

Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc64: dynamically allocate segment tables
David Gibson [Wed, 27 Jul 2005 18:44:19 +0000]
[PATCH] ppc64: dynamically allocate segment tables

PPC64 machines before Power4 need a segment table page allocated for each
CPU.  Currently these are allocated statically in a big array in head.S for
all CPUs.  The segment tables need to be in the first segment (so
do_stab_bolted doesn't take a recursive fault on the stab itself), but
other than that there are no constraints which require the stabs for the
secondary CPUs to be statically allocated.

This patch allocates segment tables dynamically during boot, using
lmb_alloc() to ensure they are within the first 256M segment.  This reduces
the kernel image size by 192k...

Tested on RS64 iSeries, POWER3 pSeries, and POWER5.

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] make a few functions static in pmac_setup.c
Olaf Hering [Wed, 27 Jul 2005 18:44:18 +0000]
[PATCH] make a few functions static in pmac_setup.c

Making a few functions static saves a few bytes, but only on ppc32.

     text    data     bss     dec     hex filename
  3752421 1605208  345608 5703237  570645 ../O-ppc64_iSeries_defconfig/vmlinux
  3709411 2042552  339992 6091955  5cf4b3 ../O-ppc64_maple_defconfig/vmlinux
  5397329 3054824  679856 9132009  8b57e9 ../O-ppc64_pSeries_defconfig/vmlinux
 -3882695  417552  197760 4498007  44a257 ../O-ppc_common_defconfig/vmlinux
 -3414510  574500  241440 4230450  408d32 ../O-ppc_pmac_defconfig/vmlinux
 +3882627  417168  197760 4497555  44a093 ../O-ppc_common_defconfig/vmlinux
 +3414442  575428  241440 4231310  40908e ../O-ppc_pmac_defconfig/vmlinux

Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc64: genrtc build fix
Andrew Morton [Wed, 27 Jul 2005 18:44:17 +0000]
[PATCH] ppc64: genrtc build fix

genrtc.c won't compile on ppc64.  Seems that ppc32 does support it though?

We do this wrong btw - we should be selecting GEN_RTC in each
arch/xxx/Kconfig.

Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc64: hide CONFIG_ADB
Olaf Hering [Wed, 27 Jul 2005 18:44:16 +0000]
[PATCH] ppc64: hide CONFIG_ADB

This bites me all day when I use our default config for ppc64.  We use a
patch to fix the compile errors and provide the CONFIG_MAC_EMUMOUSEBTN
functionality (which is behind CONFIG_INPUT_ADBHID).  But Benh doesnt like
it.

http://ozlabs.org/pipermail/linuxppc64-dev/2005-March/003423.html

Just hide all the ADB parts from via-pmu on ppc64 instead.

drivers/macintosh/adbhid.c: In function `adbhid_init':
drivers/macintosh/adbhid.c:1199: error: `_MACH_chrp' undeclared (first use in this function)
drivers/macintosh/adbhid.c:1199: error: (Each undeclared identifier is reported only once
drivers/macintosh/adbhid.c:1199: error: for each function it appears in.)

Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc64: update defconfigs
Olaf Hering [Wed, 27 Jul 2005 18:44:15 +0000]
[PATCH] ppc64: update defconfigs

update defconfig, use new CONFIG_HZ and set it to 100 just for the kicks.

Signed-off-by: Olaf Hering <olh@suse.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc32: Fix building of TQM8260 board
Kumar Gala [Wed, 27 Jul 2005 18:44:14 +0000]
[PATCH] ppc32: Fix building of TQM8260 board

Added missing include of cpm2.h in correct order to allow TQM8260 to build

Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc32: fix 440SP MAL channels count
Eugene Surovegin [Wed, 27 Jul 2005 18:44:13 +0000]
[PATCH] ppc32: fix 440SP MAL channels count

Fix the MAL channels count in PPC 440SP OCP definition.  PPC 440SP has only
1 EMAC attached to MAL.

Signed-off-by: Eugene Surovegin <ebs@ebshome.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc32: fix dma_map_page() to use page_to_bus()
Matt Porter [Wed, 27 Jul 2005 18:44:12 +0000]
[PATCH] ppc32: fix dma_map_page() to use page_to_bus()

The following trivial patch changes dma_map_page() to use page_to_bus()
instead of open-coding it (incorrectly in some cases).

Signed-off-by: Eugene Surovegin <ebs@ebshome.net>
Signed-off-by: Matt Porter <mporter@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc32: Fix building of radstone_ppc7d
Kumar Gala [Wed, 27 Jul 2005 18:44:11 +0000]
[PATCH] ppc32: Fix building of radstone_ppc7d

Updated radstone_ppc7d_defconfig to include the ds1337 driver which is used
by the platform code.  This fixes the link error when building.

Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc32: Fix building of prpmc750
Kumar Gala [Wed, 27 Jul 2005 18:44:10 +0000]
[PATCH] ppc32: Fix building of prpmc750

Updated prpmc750 platform code to include serial_reg.h to fix building.

Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc32: Fix typo in setup of 2nd PCI bus on 85xx
Kumar Gala [Wed, 27 Jul 2005 18:44:10 +0000]
[PATCH] ppc32: Fix typo in setup of 2nd PCI bus on 85xx

Typo bug that was using PCI1 defines instead of PCI2 when setting up the
second PCI bus controller on 85xx based systems.  This hasn't been a real
issue since currently the PCI2 sizes are the same as the PCI1 sizes for
currently supported boards.

Thanks to Andrew Klossner @ Xerox for point this out.

Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc32: fix compilation error with CONFIG_PQ2FADS
Downing, Thomas [Wed, 27 Jul 2005 18:44:09 +0000]
[PATCH] ppc32: fix compilation error with CONFIG_PQ2FADS

The 2.6.12.3 kernel compilation fails for ARCH=ppc when CONFIG_PQ2FADS=y.
This patch has been tested on Freescale PQ2FADS-ZU and -VR boards.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc32: 8xx: update DataTLBMiss exception comment
Marcelo Tosatti [Wed, 27 Jul 2005 18:44:08 +0000]
[PATCH] ppc32: 8xx: update DataTLBMiss exception comment

On PPC 8xx, the DataTLBMiss handler does not jump directly to the page
fault handler, as was the case in v2.4.

It instead loads an invalid TLB which causes a subsequent DataTLBError
exception.

The comment on top of it haven't been update to reflect the change, though.

Signed-off-by: Marcelo Tosatti <marcelo.tosatti@cyclades.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc32: Make the UARTs on MPC824x individual platform devices
Kumar Gala [Wed, 27 Jul 2005 18:44:07 +0000]
[PATCH] ppc32: Make the UARTs on MPC824x individual platform devices

The UARTs on the MPC824x are unique devices and really shouldn't be thought
of as a DUART.  In addition, if both UARTs are in use we need to configure
the part to enable the 2nd UART since the pins for the UARTs are
multiplexed.  Adds support to run the 824x Sandpoint with both UARTs if
desired.

Signed-off-by: Matt McClintock <msm@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc32: Add proper prototype for cpm2_reset()
Kumar Gala [Wed, 27 Jul 2005 18:44:06 +0000]
[PATCH] ppc32: Add proper prototype for cpm2_reset()

Added a proper prototype for cpm2_reset() which gets rid of a build
warning.

Signed-off-by: Jon Loeliger <jdl@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc32: update defconfigs
Olaf Hering [Wed, 27 Jul 2005 18:44:04 +0000]
[PATCH] ppc32: update defconfigs

update pmac_defconfig

enable all relevant options in common_defconfig,
so it can serve as a compiletest for PPC_MULTIPLATFORM configuration

Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ppc/ppc64: use Kconfig.hz
Olaf Hering [Wed, 27 Jul 2005 18:44:03 +0000]
[PATCH] ppc/ppc64: use Kconfig.hz

use new Kconfig.hz on ppc/ppc64, use also Kconfig.preempt for ppc

Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove bogus warning in page_alloc.c
Andy Whitcroft [Wed, 27 Jul 2005 18:44:02 +0000]
[PATCH] Remove bogus warning in page_alloc.c

Originally __free_pages_bulk used the relative page number within a zone to
define its buddies.  This meant that to maintain the "maximally aligned"
requirements (that an allocation of size N will be aligned at least to N
physically) zones had to also be aligned to 1<<MAX_ORDER pages.  When
__free_pages_bulk was updated to use the relative page frame numbers of the
free'd pages to pair buddies this released the alignment constraint on the
'left' edge of the zone.  This allows _either_ edge of the zone to contain
partial MAX_ORDER sized buddies.  These simply never will have matching
buddies and thus will never make it to the 'top' of the pyramid.

The patch below removes a now redundant check ensuring that the mem_map was
aligned to MAX_ORDER.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Cc: Christoph Lameter <christoph@lameter.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] madvise() does not always return -EBADF on non-file mapped area
suzuki [Wed, 27 Jul 2005 18:43:59 +0000]
[PATCH] madvise() does not always return -EBADF on non-file mapped area

The madvise() system call returns -EBADF for areas which does not map to
files, only for *behaviour* request MADV_WILLNEED.

According to man pages, madvise returns :

EBADF - the map exists, but the area maps something that isn't a file.

Fixes bug 2995.

Signed-off-by: Suzuki K P <suzuki@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] consolidate CONFIG_WATCHDOG_NOWAYOUT handling
Andrey Panin [Wed, 27 Jul 2005 18:43:58 +0000]
[PATCH] consolidate CONFIG_WATCHDOG_NOWAYOUT handling

Attached patch removes #ifdef CONFIG_WATCHDOG_NOWAYOUT mess duplicated in
almost every watchdog driver and replaces it with common define in
linux/watchdog.h.

Signed-off-by: Andrey Panin <pazke@donpac.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Fix warning in powernow-k8.c
Brian Gerst [Wed, 27 Jul 2005 18:43:57 +0000]
[PATCH] Fix warning in powernow-k8.c

powernow-k8.c: In function `query_current_values_with_pending_wait':
powernow-k8.c:110: warning: `hi' may be used uninitialized in this function

Signed-off-by: Brian Gerst <bgerst@didntduck.org>
Cc: Dave Jones <davej@codemonkey.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] drm: via: fix sparse warnings
Alexey Dobriyan [Wed, 27 Jul 2005 18:43:56 +0000]
[PATCH] drm: via: fix sparse warnings

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Dave Airlie <airlied@linux.ie>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] serial: MRi MRI-PCIDS1 dual port serial card
David Ranson [Wed, 27 Jul 2005 18:43:55 +0000]
[PATCH] serial: MRi MRI-PCIDS1 dual port serial card

Add support for the MRi PCIDS1 dual port serial card.  This card is a
little controversial since it is the subject of a PCI vendor/device ID
clash.  (See
http://www.ussg.iu.edu/hypermail/linux/kernel/0303.1/0516.html).  I have
for now just used the hex ID 0x950a.  The divisor was part calculated part
iterated, so may not be exactly correct (but works for me at all settings
between 300 - 115300 bps).

Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] statically link halfmd4
Andrew Morton [Wed, 27 Jul 2005 18:43:55 +0000]
[PATCH] statically link halfmd4

For some reason halfmd4 isn't being linked into the kernel any more and
modular ext3 wants it.

So statically link the halfmd4 code into the kernel.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] check_user_page_readable() deadlock fix
Andrew Morton [Wed, 27 Jul 2005 18:43:54 +0000]
[PATCH] check_user_page_readable() deadlock fix

Fix bug identifued by Richard Purdie <rpurdie@rpsys.net>.

oprofile calls check_user_page_readable() from interrupt context, so we
deadlock over various VFS locks.

But check_user_page_readable() doesn't imply either a read or a write of the
page's contents.  Change __follow_page() so that check_user_page_readable()
can tell __follow_page() that we're not accessing the page's contents, and use
that info to avoid the troublesome lock-takings.

Also, make follow_page() inline for the single callsite in memory.c to save a
bit of stack space.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fix xip sparse file handling in ext2
Carsten Otte [Wed, 27 Jul 2005 18:43:52 +0000]
[PATCH] fix xip sparse file handling in ext2

Oliver Paukstadt from our test department is testing the xip patches in
Linus' git-tree.  He found a problem that shows when reading a file that
contains sparse blocks (holes) on a -o xip mounted ext2 filesystem: the
BUG_ON() in fs/ext2/xip.c:40 triggers where it should not.  The problem was
introduced by a cleanup in my previous patch, this patch fixes it.

Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ub: fix for blank CDs
Pete Zaitcev [Wed, 27 Jul 2005 18:43:51 +0000]
[PATCH] ub: fix for blank CDs

This patch fixes a microcode lockup in my CD-ROM adapters when a blank CD
is inserted.  However, do not try to burn CDs yet!  I'm pretty sure that
trying it will end in coasters.

 - Fix a few cases where we were unable to resynchronize with replies
   for previous commands. The main thing is to keep reading replies
   in case of a stall. This is done with the new state CLRRS.
 - Since I am forgetting the basic state machine already, document it.
 - Move counter increments in the looping path in its own function.
 - Fix a harmless buglet in case CSW read fails to submit: do not
   override state.
 - Implement the Alan Stern's idea for adaptive signature checking.

Signed-off-by: Pete Zaitcev <zaitcev@yahoo.com>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Undo mempolicy shared policy rbtree microoptimization
Andi Kleen [Wed, 27 Jul 2005 18:43:50 +0000]
[PATCH] Undo mempolicy shared policy rbtree microoptimization

All mempolicy changes must be inside the spinlock and readding the rb_erase
prevents a crash while doing:

> echo "1" > /tmp/numatest
> numactl --length=0x4000 --shm /tmp/numatest --localalloc
> numactl --length=0x2000 --offset=0 --shm /tmp/numatest --membind=0
> numactl --length=0x2000 --offset=0x2000 --shm /tmp/numatest --membind=1
> ipcs
> ipcrm -M "the_key_value_of_this_shm_area"

Based on a patch by John Blackwood

Cc: <john.blackwood@ccur.com>
Cc: <andrea@suse.de>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] jsm: warning fixes
V. ANANDA KRISHNAN [Wed, 27 Jul 2005 18:43:49 +0000]
[PATCH] jsm: warning fixes

- updates the version

- fix mixing of declarations and code.  The mixing of declarations and
  code displays warnings when used against RedHat RHEL4.0 distro (compiler
  version is 3.4.3-22.1) and hence I separated them out.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] jsm: use dynamic major number allocation
V. ANANDA KRISHNAN [Wed, 27 Jul 2005 18:43:48 +0000]
[PATCH] jsm: use dynamic major number allocation

The jsm driver uses a static number of 253.  The major number 253 is a
reserved for "LOCAL/EXPERIMENTAL USE" by both char and block devices.  So
take advantage of the dynamic allocation of major number by the kernel.

Signed-off-by: V. Ananda Krishnan <mansarov@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] i4l: add Olitec ISDN PCI card in hisax gazel driver
Olivier Blin [Wed, 27 Jul 2005 18:43:47 +0000]
[PATCH] i4l: add Olitec ISDN PCI card in hisax gazel driver

This patch adds support for the Olitec ISDN PCI card in the hisax gazel
driver.  The gazel driver supports this card, but wasn't aware of its PCI
ids.  Users used to modify the PCI ids of a supported card in
include/linux/pci_ids.h and recompile their kernel to get this card
running, as said in most Howtos.  This patch makes the hisax gazel driver
recognize the PCI ids of the Olitec ISDN PCI card.

Signed-off-by: Olivier Blin <oblin@mandriva.com>
Signed-off-by: Karsten Keil <kkeil@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] scsi_scan: check return code from scsi_sysfs_add_sdev
Alan Stern [Wed, 27 Jul 2005 18:43:46 +0000]
[PATCH] scsi_scan: check return code from scsi_sysfs_add_sdev

Adds a missing check for an error return code from scsi_sysfs_add_sdev.
This resolves entry #4863 in the OSDL bugzilla.  Although in that bug
report the failure occurred because of a confusion over scanning vs.
rescanning, in general add_sdev can fail for a number of reasons (the
simplest being insufficient memory) and the caller should cope properly.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Cc: James Bottomley <James.Bottomley@SteelEye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] autofs4: fix infamous "Busy inodes after umount ..." message
Ian Kent [Wed, 27 Jul 2005 18:43:45 +0000]
[PATCH] autofs4: fix infamous "Busy inodes after umount ..." message

If the automount daemon receives a signal which causes it to sumarily
terminate the autofs4 module leaks dentries.  The same problem exists with
detached mount requests without the warning.

This patch cleans these dentries at umount.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Add pcibios_bus_to_resource for parisc
Dominik Brodowski [Wed, 27 Jul 2005 18:43:44 +0000]
[PATCH] Add pcibios_bus_to_resource for parisc

Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Acked-by: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] itimer fixes
George Anzinger [Wed, 27 Jul 2005 18:43:44 +0000]
[PATCH] itimer fixes

Fix the recent off-by-one fix in the itimer code:

1. The repeating timer is figured using the requested time
(not +1 as we know where we are in the jiffie).

2. The tests for interval too large are left to the time_val to jiffie code.

Signed-off-by: George Anzinger <george@mvista.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] watchdog: add missing 0x in alim1535_wdt.c
Rolf Eike Beer [Wed, 27 Jul 2005 18:43:42 +0000]
[PATCH] watchdog: add missing 0x in alim1535_wdt.c

Usually the device IDs are given in hex.  This one is a bit strange: it is
without 0x in the first place and used with it some lines later.  I suspect
the first one to be the wrong.

Signed-off-by: Rolf Eike Beer <eike-kernel@sf-tec.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Address BUG: using smp_processor_id() in preemptible [00000001] code
Nigel Cunningham [Wed, 27 Jul 2005 18:43:41 +0000]
[PATCH] Address BUG: using smp_processor_id() in preemptible [00000001] code

This patch fixes a warning in the disable_nonboot_cpus call in
kernel/power/smp.c.

Signed-off by: Nigel Cunningham <nigel@suspend2.net>

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] drivers/pnp/pnpbios/rsparser.c: fix compile error with PCI=n
Adrian Bunk [Wed, 27 Jul 2005 18:43:41 +0000]
[PATCH] drivers/pnp/pnpbios/rsparser.c: fix compile error with PCI=n

drivers/pnp/pnpbios/rsparser.c: In function 'pnpbios_parse_allocated_irqresource':
drivers/pnp/pnpbios/rsparser.c:67: error: too many arguments to function 'pcibios_penalize_isa_irq'

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ext3: drop quota references before releasing inode
Jan Kara [Wed, 27 Jul 2005 18:43:40 +0000]
[PATCH] ext3: drop quota references before releasing inode

We must drop references to quota structures before releasing the inode.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] ext2: drop quota reference before releasing inode
Jan Kara [Wed, 27 Jul 2005 18:43:39 +0000]
[PATCH] ext2: drop quota reference before releasing inode

We must drop references to quota structures before releasing the inode.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] reiserfs: fix deadlock in inode creation failure path w/ default ACL
Jeff Mahoney [Wed, 27 Jul 2005 18:43:38 +0000]
[PATCH] reiserfs: fix deadlock in inode creation failure path w/ default ACL

reiserfs_new_inode() can call iput() with the xattr lock held.  This will
cause a deadlock to occur when reiserfs_delete_xattrs() is called to clean
up.

The following patch releases the lock and reacquires it after the iput.
This is safe because interaction with xattrs is complete, and the relock is
just to balance out the release in the caller.

The locking needs some reworking to be more sane, but that's more intrusive
and I was just looking to fix this bug.

Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] as-iosched tunable encoding fix
Jens Axboe [Wed, 27 Jul 2005 18:43:37 +0000]
[PATCH] as-iosched tunable encoding fix

AS is doing internal msec<->jiffies conversions twice, so the sysfs tunables
which represent time are coming out wrong.  The switch from HZ=1000 exposed
this.

Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Fix missing refrigerator invocation in jffs2
Nigel Cunningham [Wed, 27 Jul 2005 18:43:36 +0000]
[PATCH] Fix missing refrigerator invocation in jffs2

Here's a patch to fix a missing refrigerator call in jffs2.

Signed-off-by: Nigel Cunningham <nigel@suspend2.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Add missing tvaudio try_to_freeze()
Nigel Cunningham [Wed, 27 Jul 2005 18:43:35 +0000]
[PATCH] Add missing tvaudio try_to_freeze()

Tvaudio lacks a refrigerator call. This patch fixes that.

Signed-off-by: Nigel Cunningham <ncunningham@suspend2.net>
Cc: <video4linux-list@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] try_to_freeze() call fixes
Nigel Cunningham [Wed, 27 Jul 2005 18:43:34 +0000]
[PATCH] try_to_freeze() call fixes

Here are fixes for four try_to_freeze calls that are still (incorrectly)
using a parameter after the recent try_to_freeze() changes.

Signed-off-by: Nigel Cunningham <nigel@suspend2.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: fix SMP boot lockup on some machines
Andi Kleen [Wed, 27 Jul 2005 18:43:34 +0000]
[PATCH] x86_64: fix SMP boot lockup on some machines

Fixes boot up lockups on some machines where CPU apic ids don't start with
0

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] uml: fix misdeclared function
Jeff Dike [Wed, 27 Jul 2005 18:43:33 +0000]
[PATCH] uml: fix misdeclared function

This fixes an interface which differed from its declaration, and includes
the relevant header so that this doesn't happen again.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] uml: update module interface
Dominik Hackl [Wed, 27 Jul 2005 18:43:32 +0000]
[PATCH] uml: update module interface

This patch replaces the deprecated MODULE_PARM function by the new
module_param function.

Signed-off-by: Dominik Hackl <dominik@hackl.dhs.org>
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] uml: add skas0 command-line option
Paolo 'Blaisorblade' Giarrusso [Wed, 27 Jul 2005 18:43:31 +0000]
[PATCH] uml: add skas0 command-line option

This adds the "skas0" parameter to force skas0 operation on SKAS3 host and
shows which operating mode has been selected.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] uml: add dependency to arch/um/Makefile for parallel builds
Olaf Hering [Wed, 27 Jul 2005 18:43:30 +0000]
[PATCH] uml: add dependency to arch/um/Makefile for parallel builds

the header file must be build before mk_user_constants.  Adding it as a
direct dep doesnt work for some reason.

arch/um/os-Linux/util/mk_user_constants.c:2:26: error: user-offsets.h: No such file or directory
arch/um/os-Linux/util/mk_user_constants.c: In function 'main':
arch/um/os-Linux/util/mk_user_constants.c:17: error: '__UM_FRAME_SIZE' undeclared (first use in this function)
arch/um/os-Linux/util/mk_user_constants.c:17: error: (Each undeclared identifier is reported only once
arch/um/os-Linux/util/mk_user_constants.c:17: error: for each function it appears in.)
make[1]: *** [arch/um/os-Linux/util/mk_user_constants] Error 1

Signed-off-by: Olaf Hering <olh@suse.de>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] uml: readd missing define to arch/um/Makefile-i386
Olaf Hering [Wed, 27 Jul 2005 18:43:29 +0000]
[PATCH] uml: readd missing define to arch/um/Makefile-i386

scripts/Makefile.build:13: /Makefile: No such file or directory
scripts/Makefile.build:64: kbuild: Makefile.build is included improperly

the define was removed, but its still required to build some targets.

Signed-off-by: Olaf Hering <olh@suse.de>
Cc: Paolo Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] md: when resizing an array, we need to update resync_max_sectors as well...
NeilBrown [Wed, 27 Jul 2005 18:43:28 +0000]
[PATCH] md: when resizing an array, we need to update resync_max_sectors as well as size

Without this, and attempt to 'grow' an array will claim to have synced the
extra part without actually having done anything.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] mips: fbdev Kcofnig fix
Yoichi Yuasa [Wed, 27 Jul 2005 18:43:28 +0000]
[PATCH] mips: fbdev Kcofnig fix

arch/mips/Kconfig is defining CONFIG_FB as bool and drivers/video/Kconfig
was changed a while ago to define it as tristate.  Remove the MIPS
definition.

Signed-off-by: Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Really __nocast-annotate kmalloc_node()
Alexey Dobriyan [Wed, 27 Jul 2005 18:43:27 +0000]
[PATCH] Really __nocast-annotate kmalloc_node()

One chunk was lost somewhere between my and Andrew's machine.

Noticed by Victor Fusco.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] I2C-MPC: Restore code removed
Kumar Gala [Wed, 27 Jul 2005 18:43:26 +0000]
[PATCH] I2C-MPC: Restore code removed

A previous patch to remove support for the OCP device model was way to
generious and moved some of the platform device model code, oops.

Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] user_mode_vm() build fix
Andrew Morton [Wed, 27 Jul 2005 18:43:25 +0000]
[PATCH] user_mode_vm() build fix

include/asm/ptrace.h: In function `user_mode_vm':
include/asm/ptrace.h:67: `VM_MASK' undeclared (first use in this function)

Cc: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64 fsnotify build fix
Andrew Morton [Wed, 27 Jul 2005 04:41:38 +0000]
[PATCH] x86_64 fsnotify build fix

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] softdog build fix
Andrew Morton [Wed, 27 Jul 2005 04:41:38 +0000]
[PATCH] softdog build fix

drivers/char/watchdog/softdog.c:94: too many arguments to function `emergency_restart'

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] eurotechwdt build fix
Andrew Morton [Wed, 27 Jul 2005 04:41:37 +0000]
[PATCH] eurotechwdt build fix

drivers/char/watchdog/eurotechwdt.c:165: too many arguments to function `emergency_restart'

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] i386: clean up user_mode macros
Chuck Ebbert [Wed, 27 Jul 2005 01:57:24 +0000]
[PATCH] i386: clean up user_mode macros

 - make the new user_mode() return 0 or 1 (same as x86_64)

 - remove conditional jump from user_mode_vm() it's called every timer
   tick on each CPU on SMP)

Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Tue, 26 Jul 2005 23:43:39 +0000]
Merge /pub/scm/linux/kernel/git/davem/net-2.6

14 years ago[PATCH] Change PowerPC MPC8xx maintainer
Tom Rini [Tue, 26 Jul 2005 19:49:53 +0000]
[PATCH] Change PowerPC MPC8xx maintainer

As Marcelo has been spending a great deal of time working on MPC8xx
systems of late (thanks!) and has more time than I do now for it, I'm
handing this over to him.

Signed-off-by: Tom Rini <trini@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[XFRM]: Fix possible overflow of sock->sk_policy
Herbert Xu [Tue, 26 Jul 2005 22:43:17 +0000]
[XFRM]: Fix possible overflow of sock->sk_policy

Spotted by, and original patch by, Balazs Scheidler.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>

14 years ago[PATCH] fix MAX_USER_RT_PRIO and MAX_RT_PRIO
Steven Rostedt [Mon, 25 Jul 2005 20:28:39 +0000]
[PATCH] fix MAX_USER_RT_PRIO and MAX_RT_PRIO

Here's the patch again to fix the code to handle if the values between
MAX_USER_RT_PRIO and MAX_RT_PRIO are different.

Without this patch, an SMP system will crash if the values are
different.

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Dean Nelson <dcn@sgi.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[NETFILTER]: Fix ip_conntrack_put() prototype.
Adrian Bunk [Tue, 26 Jul 2005 22:39:28 +0000]
[NETFILTER]: Fix ip_conntrack_put() prototype.

The function is not inline.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

14 years ago[PATCH] Fix RLIMIT_RTPRIO breakage
Andreas Steinmetz [Sat, 23 Jul 2005 11:42:04 +0000]
[PATCH] Fix RLIMIT_RTPRIO breakage

RLIMIT_RTPRIO is supposed to grant non privileged users the right to use
SCHED_FIFO/SCHED_RR scheduling policies with priorites bounded by the
RLIMIT_RTPRIO value via sched_setscheduler(). This is usually used by
audio users.

Unfortunately this is broken in 2.6.13rc3 as you can see in the excerpt
from sched_setscheduler below:

        /*
         * Allow unprivileged RT tasks to decrease priority:
         */
        if (!capable(CAP_SYS_NICE)) {
                /* can't change policy */
                if (policy != p->policy)
                        return -EPERM;

After the above unconditional test which causes sched_setscheduler to
fail with no regard to the RLIMIT_RTPRIO value the following check is made:

               /* can't increase priority */
                if (policy != SCHED_NORMAL &&
                    param->sched_priority > p->rt_priority &&
                    param->sched_priority >
                                p->signal->rlim[RLIMIT_RTPRIO].rlim_cur)
                        return -EPERM;

Thus I do believe that the RLIMIT_RTPRIO value must be taken into
account for the policy check, especially as the RLIMIT_RTPRIO limit is
of no use without this change.

The attached patch fixes this problem.

Signed-off-by: Andreas Steinmetz <ast@domdv.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years agoMerge master.kernel.org:/home/rmk/linux-2.6-arm
Linus Torvalds [Tue, 26 Jul 2005 22:13:26 +0000]
Merge master.kernel.org:/home/rmk/linux-2.6-arm

14 years agoMerge master.kernel.org:/home/rmk/linux-2.6-arm-smp
Linus Torvalds [Tue, 26 Jul 2005 22:12:54 +0000]
Merge kernel.org:/home/rmk/linux-2.6-arm-smp

14 years agoFix compiler warning in qla_iocb.c
Linus Torvalds [Tue, 26 Jul 2005 21:50:02 +0000]
Fix compiler warning in qla_iocb.c

Remove bogus initialization that was re-done (correctly) later.

14 years ago[PATCH] ARM: 2831/1: S3C2440 - split s3c2440 clocks from central clock code
Ben Dooks [Tue, 26 Jul 2005 21:39:14 +0000]
[PATCH] ARM: 2831/1: S3C2440 - split s3c2440 clocks from central clock code

Patch from Ben Dooks

Split the s3c2440 specific clocks from the arch clock support, to
make the code clearer.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

14 years ago[PATCH] acpi: Don't call acpi_sleep_prepare from acpi_power_off
Eric W. Biederman [Tue, 26 Jul 2005 18:17:52 +0000]
[PATCH] acpi: Don't call acpi_sleep_prepare from acpi_power_off

Now that all of the code paths that call acpi_power_off
have been modified to call either call kernel_power_off
(which calls apci_sleep_prepare by way of acpi_shutdown)
or to call acpi_sleep_prepare directly it is redundant to call
acpi_sleep_prepare from acpi_power_off.

So simplify the code and simply don't call acpi_sleep_prepare.

In addition there is a little error handling done so if we
can't register the acpi class we don't hook pm_power_off.

I think I have done the right thing with the CONFIG_PM define
but I'm not certain.  Can this code even be compiled if
CONFIG_PM is false?

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] acpi_power_off: Don't switch to the boot cpu
Eric W. Biederman [Tue, 26 Jul 2005 18:16:00 +0000]
[PATCH] acpi_power_off: Don't switch to the boot cpu

machine_power_off on i386 and x86_64 now switch to the
boot cpu out of paranoia and because the MP Specification indicates it
is a good idea on reboot, so for those architectures it is a noop.
I can't see anything in the acpi spec that requires you to be on
the boot cpu to power off the system, so this should not be an issue
for ia64.  In addition ia64 has the altix a massive multi-node
system where switching to the boot cpu sounds insane as we may
hot removed the boot cpu.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64 sync machine_power_off with i386
Eric W. Biederman [Tue, 26 Jul 2005 18:14:16 +0000]
[PATCH] x86_64 sync machine_power_off with i386

i386 machine_power_off was disabling the local apic
and all of it's users wanted to be on the boot cpu.
So call machine_shutdown which places us on the boot
cpu and disables the apics.  This keeps us in sync
and reduces the number of cases we need to worry about in
the power management code.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] APM: Remove redundant call to set_cpus_allowed
Eric W. Biederman [Tue, 26 Jul 2005 18:10:02 +0000]
[PATCH] APM: Remove redundant call to set_cpus_allowed

machine_power_off now always switches to the boot cpu so there
is no reason for APM to also do that.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] i386 machine_power_off cleanup
Eric W. Biederman [Tue, 26 Jul 2005 18:08:53 +0000]
[PATCH] i386 machine_power_off cleanup

Call machine_shutdown() to move to the boot cpu
and disable apics.  Both acpi_power_off and
apm_power_off want to move to the boot cpu.
and we are already disabling the local apics
so calling machine_shutdown simply reuses
code.

ia64 doesn't have a special path in power_off
for efi so there is no reason i386 should.  If
we really need to call the efi power off path
the efi driver can set pm_power_off like everyone
else.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] machine_shutdown: Typo fix to actually allow specifying which cpu to reboot on
Eric W. Biederman [Tue, 26 Jul 2005 18:07:01 +0000]
[PATCH] machine_shutdown: Typo fix to actually allow specifying which cpu to reboot on

This appears to be a typo I introduced when cleaning
this code up earlier. Ooops.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] pcwd.c: Call kernel_power_off not machine_power_off
Eric W. Biederman [Tue, 26 Jul 2005 18:03:08 +0000]
[PATCH] pcwd.c: Call kernel_power_off not machine_power_off

The call appears to come from process context so kernel_power_off
should be safe.  And acpi_power_off won't necessarily work if you just
call machine_power_off.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] swpsuspend: Have suspend to disk use factors of sys_reboot
Eric W. Biederman [Tue, 26 Jul 2005 18:01:17 +0000]
[PATCH] swpsuspend: Have suspend to disk use factors of sys_reboot

The suspend to disk code was a poor copy of the code in
sys_reboot now that we have kernel_power_off, kernel_restart
and kernel_halt use them instead of poorly duplicating them inline.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] 68328serial: sysrq should use emergency_reboot
Eric W. Biederman [Tue, 26 Jul 2005 17:59:54 +0000]
[PATCH] 68328serial: sysrq should use emergency_reboot

The 68328serial.c driver has a weird local reimplementation of
magic sysrq.  The code is architecture specific enough that calling
machine_restart() is probably ok.  But there is no reason not to call
emergency_restart() so do so.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] In hangcheck-timer.c call emergency_restart()
Eric W. Biederman [Tue, 26 Jul 2005 17:55:59 +0000]
[PATCH] In hangcheck-timer.c call emergency_restart()

If we've hung a clean reboot does not sound like a real
option.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Fix watchdog drivers to call emergency_reboot()
Eric W. Biederman [Tue, 26 Jul 2005 17:53:19 +0000]
[PATCH] Fix watchdog drivers to call emergency_reboot()

If a watchdog driver has decided it is time to reboot the system
we know something is wrong and we are in interrupt context
so emergency_reboot() is what we want.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Update sysrq-B to use emergency_restart()
Eric W. Biederman [Tue, 26 Jul 2005 17:51:06 +0000]
[PATCH] Update sysrq-B to use emergency_restart()

sysrq calls into the reboot path from an interrupt handler
we can either push the code do into process context and
call kernel_restart and get a clean reboot or we can simply
reboot the machine, and increase our chances of actually
rebooting.  emergency_reboot() seems like the closest match
to what we have previously done, and what we want.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Call emergency_reboot from panic
Eric W. Biederman [Tue, 26 Jul 2005 17:49:23 +0000]
[PATCH] Call emergency_reboot from panic

We know the system is in trouble so there is no question if this
is an emergecy :)

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Use kernel_power_off in sysrq-o
Eric W. Biederman [Tue, 26 Jul 2005 17:47:32 +0000]
[PATCH] Use kernel_power_off in sysrq-o

We already do all of the gymnastics to run from process context
to call the power off code so call into the power off code cleanly.

This especially helps acpi as part of it's shutdown logic should
run acpi_shutdown called from device_shutdown which was not
being called from here.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: Implemenent machine_emergency_restart
Eric W. Biederman [Tue, 26 Jul 2005 17:45:31 +0000]
[PATCH] x86_64: Implemenent machine_emergency_restart

It is not safe to call set_cpus_allowed() in interrupt
context and disabling the apics is complicated code.
So unconditionally skip machine_shutdown in machine_emergency_reboot
on x86_64.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: Fix reboot_force
Eric W. Biederman [Tue, 26 Jul 2005 17:44:21 +0000]
[PATCH] x86_64: Fix reboot_force

We only want to shutdown the apics if reboot_force
is not specified.  Be we are doing this both
in machine_shutdown which is called unconditionally
and if (!reboot_force).  So simply call machine_shutdown
if (!reboot_force).  It looks like something
went weird with merging some of the kexec patches for
x86_64, and caused this.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] i386: Implement machine_emergency_reboot
Eric W. Biederman [Tue, 26 Jul 2005 17:41:26 +0000]
[PATCH] i386: Implement machine_emergency_reboot

set_cpus_allowed is not safe in interrupt context
and disabling apics is complicated code so don't
call machine_shutdown on i386 from emergency_restart().

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Don't export machine_restart, machine_halt, or machine_power_off.
Eric W. Biederman [Tue, 26 Jul 2005 17:36:01 +0000]
[PATCH] Don't export machine_restart, machine_halt, or machine_power_off.

machine_restart, machine_halt and machine_power_off are machine
specific hooks deep into the reboot logic, that modules
have no business messing with.  Usually code should be calling
kernel_restart, kernel_halt, kernel_power_off, or
emergency_restart. So don't export machine_restart,
machine_halt, and machine_power_off so we can catch buggy users.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Fix the arguments to machine_restart on cris
Eric W. Biederman [Tue, 26 Jul 2005 17:32:34 +0000]
[PATCH] Fix the arguments to machine_restart on cris

It appears machine_restart has been working cris just
by luck.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>