Linus Torvalds [Fri, 25 May 2007 00:38:18 +0000 (17:38 -0700)]
Merge branch 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6
* 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6:
[IA64] acpi_get_sysname() should be __init
[IA64] Cleanup acpi header to reuse the generic _PDC defines
[IA64] Fix using uninitialized data in _PDC setup
[IA64] start_secondary() and smp_callin() should be __cpuinit
Ayaz Abdulla [Tue, 22 May 2007 00:23:11 +0000 (20:23 -0400)]
forcedeth: fix cpu irq mask
This patch fixes the cpu irq mask define to include the timer irq.
Another flag check was setting up the timer bit in all cases so we
didn't notice the issue.
Signed-off-by: Ayaz Abdulla <aabdulla@nvidia.com> Signed-off-by: Jeff Garzik <jeff@garzik.org>
Florin Malita [Tue, 22 May 2007 23:09:42 +0000 (18:09 -0500)]
spidernet: skb used after netif_receive_skb
The stats update code in spider_net_pass_skb_up() is touching the skb
after it's been passed up to the stack. To avoid that, just update the
stats first.
Ralf Baechle [Thu, 24 May 2007 11:54:04 +0000 (12:54 +0100)]
meth driver renovation
The meth ethernet driver for the SGI IP32 aka O2 is so far still an old
style driver which does not use the device driver model. This is now
causing issues with some udev based gadgetry in debian-stable. Fixed by
converting the meth driver to a platform device.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
--
Fixes since previous patch:
o Fixed typo in meth_exit_module() Signed-off-by: Jeff Garzik <jeff@garzik.org>
Tony Luck [Thu, 24 May 2007 17:59:44 +0000 (10:59 -0700)]
[IA64] acpi_get_sysname() should be __init
Section mismatch: reference to .init.text:acpi_find_rsdp
(between 'acpi_get_sysname' and 'acpi_request_vector')
acpi_get_sysname() needs to call the __init function acpi_find_rsdp, but it
doesn't have the __init attribute itself, hence the warning. Luckily it is
only called from machvec_init() which has __init attribute, so the fix
is to define acpi_get_sysname() as __init too.
Linus Torvalds [Thu, 24 May 2007 16:17:12 +0000 (09:17 -0700)]
Merge branch 'fixes' of git://git.linux-nfs.org/pub/linux/nfs-2.6
* 'fixes' of git://git.linux-nfs.org/pub/linux/nfs-2.6:
NFS: Fix nfs_direct_dirty_pages()
NFS: Fix handful of compiler warnings in direct.c
NFS: Avoid a deadlock situation on write
Linus Torvalds [Thu, 24 May 2007 15:37:52 +0000 (08:37 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/drzeus/mmc
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/drzeus/mmc:
mmc: add maintainer for ARM Primecell controller
mmc: add maintainer for iMX MMC interface
mmc: Add maintainers for TI OMAP MMC interface
mmc: mark unmaintained drivers
mmc: clean up unused parts of block driver
Linus Torvalds [Thu, 24 May 2007 15:37:14 +0000 (08:37 -0700)]
Fix crash with irqpoll due to the IRQF_IRQPOLL flag testing
With irqpoll enabled, trying to test the IRQF_IRQPOLL flag in the
actions would cause a NULL pointer dereference if no action was
installed (for example, the driver might have been unloaded with
interrupts still pending).
So be a bit more careful about testing the flag by making sure to test
for that case.
(The actual _change_ is trivial, the patch is more than a one-liner
because I rewrote the testing to also be much more readable.
Trond Myklebust [Tue, 22 May 2007 14:22:27 +0000 (10:22 -0400)]
NFS: Fix nfs_direct_dirty_pages()
We only need to dirty the pages that were actually read in.
Also convert nfs_direct_dirty_pages() to call set_page_dirty() instead of
set_page_dirty_lock(). A call to lock_page() is unacceptable in an rpciod
callback function.
Chuck Lever [Sat, 19 May 2007 21:22:46 +0000 (17:22 -0400)]
NFS: Fix handful of compiler warnings in direct.c
This patch fixes a couple of signage issues that were causing an Oops
when running the LTP diotest4 test. get_user_pages() returns a signed
error, hence we need to be careful when comparing with the unsigned
number of pages from data->npages.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Trond Myklebust [Sun, 20 May 2007 14:18:27 +0000 (10:18 -0400)]
NFS: Avoid a deadlock situation on write
When processes are allowed to attempt to lock a non-contiguous range of nfs
write requests, it is possible for generic_writepages to 'wrap round' the
address space, and call writepage() on a request that is already locked by
the same process.
We avoid the deadlock by checking if the page index is contiguous with the
list of nfs write requests that is already held in our
nfs_pageio_descriptor prior to attempting to lock a new request.
Jordan Crouse [Thu, 24 May 2007 11:23:24 +0000 (21:23 +1000)]
[CRYPTO] geode: Fix in-place operations and set key
Allow in-place crypto operations. Also remove the coherent user flag
(we use it automagically now), and by default use the user written
key rather then the HW hidden key - this makes crypto just work without
any special considerations, and thats OK, since its our only usage
model.
Signed-off-by: Jordan Crouse <jordan.crouse@amd.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Syed Khasim [Tue, 15 May 2007 23:07:22 +0000 (01:07 +0200)]
mmc: Add maintainers for TI OMAP MMC interface
Carlos Aguiar and Anderson Briglia are interested in making sure
the driver works for existing boards as they have access to them,
and Syed Khasim can make it work for new omaps (2430, 3430).
Signed-off-by: Syed Khasim <x0khasim@ti.com> Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Pierre Ossman [Mon, 14 May 2007 19:25:26 +0000 (21:25 +0200)]
mmc: mark unmaintained drivers
Most of the host controller drivers in the MMC layer lacks an
official maintainer. Make sure this is mentioned in MAINTAINERS
in case someone wants to pick up the ball.
drivers/isdn/hardware/eicon/message.c: In function 'api_put':
drivers/isdn/hardware/eicon/message.c:536: warning: cast from pointer to integer of different size
drivers/isdn/hardware/eicon/message.c: In function 'plci_free_msg_in_queue':
drivers/isdn/hardware/eicon/message.c:1035: warning: cast to pointer from integer of different size
drivers/isdn/hardware/eicon/message.c: In function 'data_b3_req':
drivers/isdn/hardware/eicon/message.c:3121: warning: cast to pointer from integer of different size
drivers/isdn/hardware/eicon/message.c:3154: warning: cast to pointer from integer of different size
drivers/isdn/hardware/eicon/message.c: In function 'callback':
drivers/isdn/hardware/eicon/message.c:4060: warning: cast to pointer from integer of different size
drivers/isdn/hardware/eicon/message.c: In function 'nl_ind':
drivers/isdn/hardware/eicon/message.c:7137: warning: cast from pointer to integer of different size
Andrew Morton [Wed, 23 May 2007 20:58:22 +0000 (13:58 -0700)]
capifunc warning fixes
squish these:
drivers/isdn/hardware/eicon/capifunc.c: In function 'TransmitBufferSet':
drivers/isdn/hardware/eicon/capifunc.c:192: warning: cast to pointer from integer of different size
drivers/isdn/hardware/eicon/capifunc.c: In function 'TransmitBufferGet':
drivers/isdn/hardware/eicon/capifunc.c:197: warning: cast from pointer to integer of different size
drivers/isdn/hardware/eicon/capifunc.c:198: warning: cast from pointer to integer of different size
drivers/isdn/hardware/eicon/capifunc.c:200: warning: cast from pointer to integer of different size
drivers/isdn/hardware/eicon/capifunc.c: In function 'TransmitBufferFree':
drivers/isdn/hardware/eicon/capifunc.c:205: warning: cast from pointer to integer of different size
drivers/isdn/hardware/eicon/capifunc.c:206: warning: cast from pointer to integer of different size
drivers/isdn/hardware/eicon/capifunc.c: In function 'sendf':
drivers/isdn/hardware/eicon/capifunc.c:304: warning: cast to pointer from integer of different size
drivers/isdn/hardware/eicon/capifunc.c:304: warning: cast to pointer from integer of different size
drivers/isdn/hardware/eicon/capifunc.c:321: warning: cast to pointer from integer of different size
Thomas Gleixner [Wed, 23 May 2007 20:58:19 +0000 (13:58 -0700)]
i386: fix early usage of atomic_add_return and local_add_return on real i386
The check (boot_cpu_data.x86 == 3) in atomic_add_return() and
local_add_return() fails, when those operations are used before
boot_cpu_data is filled in.
Change the check to (boot_cpu_data.x86 <= 3) to fix this.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Thomas Gleixner [Wed, 23 May 2007 20:58:18 +0000 (13:58 -0700)]
Prevent going idle with softirq pending
The NOHZ patch contains a check for softirqs pending when a CPU goes idle.
The BUG is unrelated to NOHZ, it just was made visible by the NOHZ patch.
The BUG showed up mainly on P4 / hyperthreading enabled machines which lead
the investigations into the wrong direction in the first place. The real
cause is in cond_resched_softirq():
cond_resched_softirq() is enabling softirqs without invoking the softirq
daemon when softirqs are pending. This leads to the warning message in the
NOHZ idle code:
t1 runs softirq disabled code on CPU#0
interrupt happens, softirq is raised, but deferred (softirqs disabled)
t1 calls cond_resched_softirq()
enables softirqs via _local_bh_enable()
calls schedule()
t2 runs
t1 is migrated to CPU#1
t2 is done and invokes idle()
NOHZ detects the pending softirq
Fix: change _local_bh_enable() to local_bh_enable() so the softirq
daemon is invoked.
Thanks to Anant Nitya for debugging this with great patience !
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Cc: <stable@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Li Yang [Wed, 23 May 2007 20:58:17 +0000 (13:58 -0700)]
ehci-fsl: fix cache coherency problem on system with large memory
The patch fixes bug http://bugzilla.kernel.org/show_bug.cgi?id=7482.
It sets USB snooping on 4G space for PowerPC platforms without
CONFIG_NOT_COHERENT_CACHE defined.
Reported-by: Stefan Meyer <reyems@telkomsa.net> Signed-off-by: Li Yang <leoli@freescale.com> Cc: Greg KH <greg@kroah.com> Cc: David Brownell <david-b@pacbell.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Alan Stern [Wed, 23 May 2007 20:58:12 +0000 (13:58 -0700)]
HiSax: fix error checking for hisax_register()]
This patch (as875) adds error-checking to the callers of hisax_register().
It also changes an error pathway in that routine, making it return an error
code rather than 0.
This fixes Bugzilla #7960.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Acked-by: Karsten Keil <kkeil@suse.de> Cc: Martin Bachem <info@colognechip.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
NeilBrown [Wed, 23 May 2007 20:58:10 +0000 (13:58 -0700)]
md: fix bug with linear hot-add and elsewhere
Adding a drive to a linear array seems to have stopped working, due to changes
elsewhere in md, and insufficient ongoing testing...
So the patch to make linear hot-add work in the first place introduced a
subtle bug elsewhere that interracts poorly with older version of mdadm.
This fixes it all up.
Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
NeilBrown [Wed, 23 May 2007 20:58:10 +0000 (13:58 -0700)]
md: don't write more than is required of the last page of a bitmap
It is possible that real data or metadata follows the bitmap without full page
alignment.
So limit the last write to be only the required number of bytes, rounded up to
the hard sector size of the device.
Signed-off-by: Neil Brown <neilb@suse.de> Cc: <stable@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Stefan Richter [Wed, 23 May 2007 20:58:08 +0000 (13:58 -0700)]
i386, x86-64: show that CONFIG_HOTPLUG_CPU is required for suspend on SMP
It's not sufficiently documented that CONFIG_HOTPLUG_CPU is required for
suspend/hibernation on SMP.
Point out the non-obvious.
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Akinobu Mita [Wed, 23 May 2007 20:58:05 +0000 (13:58 -0700)]
i2o: destroy event queue only when drv->event is set
i2o_driver_register() initalizes event queue for driver only when
drv->event is set. So similarly the event queue should be destroyed only
when drv->event is set in the error path. Otherwise destroy_workqueue()
will called with NULL.
Cc: Markus Lidel <Markus.Lidel@shadowconnect.com> Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Davi Arnaut [Wed, 23 May 2007 20:58:04 +0000 (13:58 -0700)]
signalfd: retrieve multiple signals with one read() call
Gathering signals in bulk enables server applications to drain a signal
queue (almost full of realtime signals) more efficiently by reducing the
syscall and file look-up overhead.
Very similar to the sigtimedwait4() call described by Niels Provos, Chuck
Lever, and Stephen Tweedie in a paper entitled "Analyzing the Overload
Behavior of a Simple Web Server". The paper lists more details and
advantages.
Signed-off-by: Davi E. M. Arnaut <davi@haxent.com.br> Acked-by: Davide Libenzi <davidel@xmailserver.org> Cc: Oleg Nesterov <oleg@tv-sign.ru> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
SLUB Debug: fix check for super sized slabs (>512k 64bit, >256k 32bit)
The check for super sized slabs where we can no longer move the free
pointer behind the object for debugging purposes etc is accessing a
field that is not setup yet. We must use objsize here since the size of
the slab has not been determined yet.
The effect of this is that a global slab shrink via "slabinfo -s" will
show errors about offsets being wrong if booted with slub_debug.
Potentially there are other troubles with huge slabs under slub_debug
because the calculated free pointer offset is truncated.
Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Sascha Hauer [Wed, 23 May 2007 20:57:51 +0000 (13:57 -0700)]
imxfb: remove ifdefs
Here are some small fixes for the imxfb framebuffer driver:
- remove ifndef for MX1FS2 platform which is not present in the kernel.
- REV, CLS, PS and SPL_SPR are only needed with some sharp displays.
Only setup the corresponding io pins when such a display is connected.
Krzysztof Helt [Wed, 23 May 2007 20:57:48 +0000 (13:57 -0700)]
pm3fb: various fixes
- fix 16-bit RGBA 4444 and 5551 modes (use fbset -rgba to set them)
- kill wmb() after each register write by PM3_SLOW_WRITE_REG
- kill inaccurate comments (leftovers from skeletonfb.c)
- kill PCI class restriction, now it is more general and shorter
- white spaces and other minor fixes and source reducing
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Signed-off-by: Antonino Daplas <adaplas@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Roland McGrath [Wed, 23 May 2007 20:57:44 +0000 (13:57 -0700)]
recalc_sigpending_tsk fixes
Steve Hawkes discovered a problem where recalc_sigpending_tsk was called in
do_sigaction but no signal_wake_up call was made, preventing later signals
from waking up blocked threads with TIF_SIGPENDING already set.
In fact, the few other calls to recalc_sigpending_tsk outside the signals
code are also subject to this problem in other race conditions.
This change makes recalc_sigpending_tsk private to the signals code. It
changes the outside calls, as well as do_sigaction, to use the new
recalc_sigpending_and_wake instead.
Signed-off-by: Roland McGrath <roland@redhat.com> Cc: <Steve.Hawkes@motorola.com> Cc: Oleg Nesterov <oleg@tv-sign.ru> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ben Collins [Wed, 23 May 2007 20:57:43 +0000 (13:57 -0700)]
Avoid zero size allocation in cache_k8_northbridges()
kmalloc for flush_words resulted in zero size allocation when no
k8_northbridges existed. Short circuit the code path for this case.
Also remove uneeded zeroing of num_k8_northbridges just after checking if
it is zero.
Signed-off-by: Ben Collins <bcollins@ubuntu.com> Cc: Andi Kleen <ak@suse.de> Cc: Dave Jones <davej@codemonkey.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Robert P. J. Day [Wed, 23 May 2007 20:57:42 +0000 (13:57 -0700)]
Documentation: fix the explanation of Kconfig files
Fix the obvious errors in the explanation of Kconfig files.
Signed-off-by: Robert P. J. Day <rpjday@mindspring.com> Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Paul E. McKenney [Wed, 23 May 2007 20:57:41 +0000 (13:57 -0700)]
prohibit rcutorture from being compiled into the kernel
There have been a number of instances where people have accidentally compiled
rcutorture into the kernel (CONFIG_RCU_TORTURE_TEST=y), which has never been
useful, and has often resulted in great frustration.
The attached patch prohibits rcutorture from being compiled into the
kernel. It may be excluded altogether or compiled as a module. People
wishing to have rcutorture hammer their machine immediately upon boot
are free to hand-edit lib/Kconfig.debug to remove the "depends on m"
line.
Thanks to Randy Dunlap for the trick that makes this work.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by: Josh Triplett <josh@kernel.org> Cc: "Randy.Dunlap" <rdunlap@xenotime.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Jeff Dike [Wed, 23 May 2007 20:57:40 +0000 (13:57 -0700)]
uml: improve PTRACE_SYSEMU checking
Make the PTRACE_SYSEMU checking more robust. It will make sure that system
call numbers are reported correctly. If there is a problem, it will disable
PTRACE_SYSEMU use and use PTRACE_SYSCALL instead.
Thanks to Balaji G for helping reproduce this problem.
Signed-off-by: Jeff Dike <jdike@linux.intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Andrew Morton [Wed, 23 May 2007 20:57:39 +0000 (13:57 -0700)]
capability.h warning fix
include/linux/capability.h:397: warning: "struct task_struct" declared inside parameter list
include/linux/capability.h:397: warning: its scope is only this definition or declaration, which is probably not what you want
Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Allow user space to determine if a disk supports Asynchronous Notification of
media changes. This is done by adding a new sysfs file "capability_flags",
which is documented in (insert file name). This sysfs file will export all
disk capabilities flags to user space. We also define a new flag to define
the media change notification capability.
Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Thomas Gleixner [Wed, 23 May 2007 20:57:37 +0000 (13:57 -0700)]
NOHZ: Rate limit the local softirq pending warning output
The warning in the NOHZ code, which triggers when a CPU goes idle with
softirqs pending can fill up the logs quite quickly. Rate limit the output
until we found the root cause of that problem.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: <stable@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Domen Puncer [Wed, 23 May 2007 20:57:32 +0000 (13:57 -0700)]
mpc52xx_psc_spi: fix it for CONFIG_PPC_MERGE
Fix mpc5200 PSC SPI driver to actually work for CONFIG_PPC_MERGE
- s/mpc52xx/mpc5200/, as this was changed in device tree some time ago
- fix spi id detection
The move from 'ppc' to 'powerpc' didn't quite match the latest
kernel.org code.
Signed-off-by: Domen Puncer <domen.puncer@telargo.com> Acked-by: Dragos Carp <dragos.carp@toptica.com> Acked-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The object size calculation is wrong if !CONFIG_SLUB_DEBUG because the
#ifdef CONFIG_SLUB_DEBUG is now switching off the size adjustments for
DESTROY_BY_RCU and ctor.
Signed-off-by: Christoph Lameter <clameter@sgi.com> Acked-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Thomas Gleixner [Wed, 23 May 2007 20:57:30 +0000 (13:57 -0700)]
Ignore bogus ACPI info for offline CPUs
Booting a SMP kernel with maxcpus=1 on a SMP system leads to a hard hang,
because ACPI ignores the maxcpus setting and sends timer broadcast info for
the offline CPUs. This results in a stuck for ever call to
smp_call_function_single() on an offline CPU.
Ignore the bogus information and print a kernel error to remind ACPI
folks to fix it.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: <stable@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Gautham R Shenoy [Wed, 23 May 2007 20:57:28 +0000 (13:57 -0700)]
freezer: fix PF_NOFREEZE vs freezeable race
This patch fixes the race pointed out by Oleg Nesterov.
* Freezer marks a thread as freezeable.
* The thread now marks itself PF_NOFREEZE, but it will be frozen on
on calling try_to_freeze(). Thus the task is frozen, even though it doesn't
want to.
* Subsequent thaw_processes() will also fail to thaw the task since it is
marked PF_NOFREEZE.
Avoid this problem by checking the task's PF_NOFREEZE status in
frozen_processes() before marking the task as frozen.
Signed-off-by: Gautham R Shenoy <ego@in.ibm.com> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Cc: Oleg Nesterov <oleg@tv-sign.ru> Cc: Pavel Machek <pavel@ucw.cz> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Oleg Nesterov [Wed, 23 May 2007 20:57:27 +0000 (13:57 -0700)]
freezer: fix kthread_create vs freezer theoretical race
kthread() sleeps in TASK_INTERRUPTIBLE state waiting for the first wakeup. In
theory, this wakeup may come from freeze_process()->signal_wake_up(), so the
task can disappear even before kthread_create() sets its ->comm.
Change kthread() to use TASK_UNINTERRUPTIBLE.
[akpm@linux-foundation.org: s/BUG_ON/WARN_ON+recover] Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru> Acked-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Cc: Gautham R Shenoy <ego@in.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Kernel threads can become userland processes by calling kernel_execve().
In particular, this may happen right after the try_to_freeze_tasks()
called with FREEZER_USER_SPACE has returned, so try_to_freeze_tasks()
needs to take userspace processes into consideration even if it is
called with FREEZER_KERNEL_THREADS.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Acked-by: Pavel Machek <pavel@ucw.cz> Cc: Gautham R Shenoy <ego@in.ibm.com> Cc: Oleg Nesterov <oleg@tv-sign.ru> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently try_to_freeze_tasks() has to wait until all of the vforked processes
exit and for this reason every user can make it fail. To fix this problem we
can introduce the additional process flag PF_FREEZER_SKIP to be used by tasks
that do not want to be counted as freezable by the freezer and want to have
TIF_FREEZE set nevertheless. Then, this flag can be set by tasks using
sys_vfork() before they call wait_for_completion(&vfork) and cleared after
they have woken up. After clearing it, the tasks should call try_to_freeze()
as soon as possible.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Cc: Gautham R Shenoy <ego@in.ibm.com> Cc: Oleg Nesterov <oleg@tv-sign.ru> Cc: Pavel Machek <pavel@ucw.cz> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
freezer: close potential race between refrigerator and thaw_tasks
If the freezing of tasks fails and a task is preempted in refrigerator()
before calling frozen_process(), then thaw_tasks() may run before this task is
frozen. In that case the task will freeze and no one will thaw it.
To fix this race we can call freezing(current) in refrigerator() along with
frozen_process(current) under the task_lock() which also should be taken in
the error path of try_to_freeze_tasks() as well as in thaw_process().
Moreover, if thaw_process() additionally clears TIF_FREEZE for tasks that are
not frozen, we can be sure that all tasks are thawed and there are no pending
"freeze" requests after thaw_tasks() has run.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Acked-by: Pavel Machek <pavel@ucw.cz> Cc: Gautham R Shenoy <ego@in.ibm.com> Cc: Oleg Nesterov <oleg@tv-sign.ru> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Junio C Hamano [Thu, 24 May 2007 00:42:38 +0000 (02:42 +0200)]
Match DMA blacklist entries between ide-dma.c and libata-core.c
There are a few entries in ata_device_blacklist[] in libata-core.c
marked with HORKAGE_NODMA but are missing from drive_blacklist[]
in ide-dma.c. This patch makes the lists in sync.
Also remove a duplicated entry for "SanDisk SDP3B-64".
Signed-off-by: Junio C Hamano <junkio@cox.net> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Cc: Dave Jones <davej@redhat.com> Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Mika Kukkonen [Thu, 24 May 2007 00:42:38 +0000 (02:42 +0200)]
Add two missing chipsets to drivers/ide/ide-proc.c
Compiling with '-Wswitch-enum' I noticed following:
CC drivers/ide/ide-proc.o
drivers/ide/ide-proc.c: In function â\80\98proc_ide_read_imodelâ\80\99:
drivers/ide/ide-proc.c:54: warning: enumeration value â\80\98ide_etrax100â\80\99 not handled in switch
drivers/ide/ide-proc.c:54: warning: enumeration value â\80\98ide_acornâ\80\99 not handled in switch
How about the following patch?
Signed-off-by: Mika Kukkonen <mikukkon@iki.fi> Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Alan Cox [Thu, 24 May 2007 00:42:38 +0000 (02:42 +0200)]
ide/pci/serverworks.c: Fix corruption/timeouts with MegaIDE
It turns out from customer reports to Red Hat and some PCI dumps that the
MegaIDE in RAID mode doesn't provide the drive tuning data that the
serverworks driver expects but sometimes does provide something that
fools the code.
For the RAID class case skip the oem setup and don't trust the BIOS data.
We then tune from scratch and this sorts it out. (This has been confirmed
on an afflicted IBM blade)
[libata serverworks.c never trusts the BIOS in the first place so is
accidentally immune]
Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
projects / linux-2.6-omap-h63xx.git / log