linux-toradex.git/kernel/softlockup.c, branch v2.6.27.27

softlockup: minor cleanup, don't check task->state twice

2008-09-02T17:49:51+00:00

The recent commit 16d9679f33caf7e683471647d1472bfe133d858 changed
check_hung_task() to filter out the TASK_KILLABLE tasks. We can
move this check to the caller which has to test t->state anyway.

Signed-off-by: Oleg Nesterov 
Acked-by: Andi Kleen 
Signed-off-by: Linus Torvalds

Don't trigger softlockup detector on network fs blocked tasks

2008-08-29T21:46:29+00:00

Pulling the ethernet cable on a 2.6.27-rc system with NFS mounts
currently leads to an ongoing flood of soft lockup detector backtraces
for all tasks blocked on the NFS mounts when the hickup takes
longer than 120s.

I don't think NFS problems should be all that noisy.

Luckily there's a reasonably easy way to distingush this case.

Don't report task softlockup warnings for tasks in TASK_KILLABLE
state, which is used by the network file systems.

I believe this patch is a 2.6.27 candidate.

Signed-off-by: Andi Kleen 
Signed-off-by: Linus Torvalds

Full conversion to early_initcall() interface, remove old interface

2008-07-26T19:00:04+00:00

A previous patch added the early_initcall(), to allow a cleaner hooking of
pre-SMP initcalls.  Now we remove the older interface, converting all
existing users to the new one.

[akpm@linux-foundation.org: cleanups]
[akpm@linux-foundation.org: build fix]
[kosaki.motohiro@jp.fujitsu.com: warning fix]
[kosaki.motohiro@jp.fujitsu.com: warning fix]
Signed-off-by: Eduard - Gabriel Munteanu 
Cc: Tom Zanussi 
Signed-off-by: KOSAKI Motohiro 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

softlockup: fix watchdog task wakeup frequency

2008-07-01T07:22:49+00:00

The print_timestamp can never be bigger than the touch_timestamp, at
maximum it can be equal.  And if it is, the second check for
touch_timestamp + 1 bigger print_timestamp is always true, too.

The check for equality is sufficient as we proceed in one-second-steps
and are at least one second away from the last print-out if we have
another timestamp.

Signed-off-by: Johannes Weiner 
Cc: Peter Zijlstra 
Signed-off-by: Ingo Molnar

softlockup: fix watchdog task wakeup frequency

2008-06-30T13:03:08+00:00

Updating the timestamp more often is pointless as we print the warnings
only if we exceed the threshold.  And the check for hung tasks relies on
the last timestamp, so it will keep working correctly, too.

Signed-off-by: Johannes Weiner 
Cc: Peter Zijlstra 
Signed-off-by: Ingo Molnar

softlockup: show irqtrace

2008-06-25T15:49:48+00:00

This patch adds some information about when interrupts were last
enabled and disabled to the output of the softlockup detector.

Signed-off-by: Vegard Nossum 
Cc: Peter Zijlstra 
Cc: Johannes Weiner 
Cc: Arjan van de Ven 
Signed-off-by: Ingo Molnar

softlockup: print a module list on being stuck

2008-06-18T13:26:54+00:00

Most places in the kernel that go BUG: print a module list
(which is very useful for doing statistics and finding patterns),
however the softlockup detector does not do this yet.

This patch adds the one line change to fix this gap.

Signed-off-by: Arjan van de Ven 
Signed-off-by: Ingo Molnar

softlockup: fix NMI hangs due to lock race - 2.6.26-rc regression

2008-06-02T10:38:27+00:00

The touch_nmi_watchdog() routine on x86 ultimately calls
touch_softlockup_watchdog().  The problem is that to touch the
softlockup watchdog, the cpu_clock code has to be called which could
involve multiple cpu locks and can lead to a hard hang if one of the
locks is held by a processor that is not going to return anytime soon
(such as could be the case with kgdb or perhaps even with some other
kind of exception).

This patch causes the public version of the
touch_softlockup_watchdog() to defer the cpu clock access to a later
point.

The test case for this problem is to use the following kernel config
options:

CONFIG_KGDB_TESTS=y
CONFIG_KGDB_TESTS_ON_BOOT=y
CONFIG_KGDB_TESTS_BOOT_STRING="V1F100I100000"

It should be noted that kgdb test suite and these options were not
available until 2.6.26-rc2, so it was necessary to patch the kgdb
test suite during the bisection.

I would consider this patch a regression fix because the problem first
appeared in commit 27ec4407790d075c325e1f4da0a19c56953cce23 when some
logic was added to try to periodically sync the clocks.  It was
possible to work around this particular problem by simply not
performing the sync anytime the system was in a critical context.
This was ok until commit 3e51f33fcc7f55e6df25d15b55ed10c8b4da84cd,
which added config option CONFIG_HAVE_UNSTABLE_SCHED_CLOCK and some
multi-cpu locks to sync the clocks.  It became clear that accessing
this code from an nmi was the source of the lockups.  Avoiding the
access to the low level clock code from an code inside the NMI
processing also fixed the problem with the 27ec44... commit.

Signed-off-by: Jason Wessel 
Signed-off-by: Ingo Molnar

softlockup: fix softlockup_thresh unaligned access and disable detection at runtime

2008-05-25T04:35:03+00:00

Fix unaligned access errors when setting softlockup_thresh on
64 bit platforms.

Allow softlockup detection to be disabled by setting
softlockup_thresh <= 0.

Detect that boot time softlockup detection has been disabled
earlier in softlockup_tick.

Signed-off-by: Dimitri Sivanich 
Signed-off-by: Ingo Molnar 
Signed-off-by: Thomas Gleixner

softlockup: allow panic on lockup

2008-05-25T04:34:44+00:00

allow users to configure the softlockup detector to generate a panic
instead of a warning message.

high-availability systems might opt for this strict method (combined
with panic_timeout= boot option/sysctl), instead of generating
softlockup warnings ad infinitum.

also, automated tests work better if the system reboots reliably (into
a safe kernel) in case of a lockup.

The full spectrum of configurability is supported: boot option, sysctl
option and Kconfig option.

it's default-disabled.

Signed-off-by: Ingo Molnar 
Signed-off-by: Thomas Gleixner