A 'softlockup' is defined as a bug that causes the kernel to loop in
kernel mode for more than a predefined period to time, without giving
other tasks a chance to run.
Currently, upon detection of this condition by the per-cpu watchdog
task, debug information (including a stack trace) is sent to the system
log.
On some occasions, we have observed that the "victim" rather than the
actual "culprit" (i.e. the owner/holder of the contended resource) is
reported to the user. Often this information has proven to be
insufficient to assist debugging efforts.
To avoid loss of useful debug information, for architectures which
support NMI, this patch makes it possible to improve soft lockup
reporting. This is accomplished by issuing an NMI to each cpu to obtain
a stack trace.
If NMI is not supported we just revert back to the old method. A sysctl
and boot-time parameter is available to toggle this feature.
[dzickus@redhat.com: add CONFIG_SMP in certain areas]
[akpm@linux-foundation.org: additional CONFIG_SMP=n optimisations]
[mq@suse.cz: fix warning]
Signed-off-by: Aaron Tomlin <atomlin@redhat.com>
Signed-off-by: Don Zickus <dzickus@redhat.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Mateusz Guzik <mguzik@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Jan Moskyto Matejka <mq@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Sometimes it is preferred not to use the trigger_all_cpu_backtrace()
routine when one wants to avoid capturing a back trace for current. For
instance if one was previously captured recently.
This patch provides a new routine namely
trigger_allbutself_cpu_backtrace() which offers the flexibility to issue
an NMI to every cpu but current and capture a back trace accordingly.
Patch x86 and sparc to support new routine.
[dzickus@redhat.com: add stub in #else clause]
[dzickus@redhat.com: don't print message in single processor case, wrap with get/put_cpu based on Oleg's suggestion]
[sfr@canb.auug.org.au: undo C99ism]
Signed-off-by: Aaron Tomlin <atomlin@redhat.com>
Signed-off-by: Don Zickus <dzickus@redhat.com>
Acked-by: David S. Miller <davem@davemloft.net>
Cc: Mateusz Guzik <mguzik@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We have two very conflicting state variable names in the
watchdog:
* watchdog_enabled: This one reflects the user interface. It's
set to 1 by default and can be overriden with boot options
or sysctl/procfs interface.
* watchdog_disabled: This is the internal toggle state that
tells if watchdog threads, timers and NMI events are currently
running or not. This state mostly depends on the user settings.
It's a convenient state latch.
Now we really need to find clearer names because those
are just too confusing to encourage deep review.
watchdog_enabled now becomes watchdog_user_enabled to reflect
its purpose as an interface.
watchdog_disabled becomes watchdog_running to suggest its
role as a pure internal state.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Cc: Anish Singh <anish198519851985@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Don Zickus <dzickus@redhat.com>
ARCH_HAS_NMI_WATCHDOG is a macro defined by arch, but config
HARDLOCKUP_DETECTOR depends on it. This is wrong, ARCH_HAS_NMI_WATCHDOG
has to be a Kconfig config, and arch's need it should select it
explicitly.
Signed-off-by: WANG Cong <xiyou.wangcong@gmail.com>
Acked-by: Don Zickus <dzickus@redhat.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Cc: David Howells <dhowells@redhat.com>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Before the conversion of the NMI watchdog to perf event, the
watchdog timeout was 5 seconds. Now it is 60 seconds. For my
particular application, netbooks, 5 seconds was a better
timeout. With a short timeout, we catch faults earlier and are
able to send back a panic. With a 60 second timeout, the user is
unlikely to wait and will instead hit the power button, causing
us to lose the panic info.
This change configures the NMI period to watchdog_thresh and
sets the softlockup_thresh to watchdog_thresh * 2. In addition,
watchdog_thresh was reduced to 10 seconds as suggested by Ingo
Molnar.
Signed-off-by: Mandeep Singh Baines <msb@chromium.org>
Cc: Marcin Slusarz <marcin.slusarz@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Link: http://lkml.kernel.org/r/1306127423-3347-4-git-send-email-msb@chromium.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
LKML-Reference: <20110517071642.GF22305@elte.hu>
This restores the previous behavior of softlock_thresh.
Currently, setting watchdog_thresh to zero causes the watchdog
kthreads to consume a lot of CPU.
In addition, the logic of proc_dowatchdog_thresh and
proc_dowatchdog_enabled has been factored into proc_dowatchdog.
Signed-off-by: Mandeep Singh Baines <msb@chromium.org>
Cc: Marcin Slusarz <marcin.slusarz@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Link: http://lkml.kernel.org/r/1306127423-3347-3-git-send-email-msb@chromium.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
LKML-Reference: <20110517071018.GE22305@elte.hu>
The x86 arch has shifted its use of the nmi_watchdog from a
local implementation to the global one provide by
kernel/watchdog.c. This shift has caused a whole bunch of
compile problems under different config options. I attempt to
simplify things with the patch below.
In order to simplify things, I had to come to terms with the
meaning of two terms ARCH_HAS_NMI_WATCHDOG and
CONFIG_HARDLOCKUP_DETECTOR. Basically they mean the same thing,
the former on a local level and the latter on a global level.
With the old x86 nmi watchdog gone, there is no need to rely on
defining the ARCH_HAS_NMI_WATCHDOG variable because it doesn't
make sense any more. x86 will now use the global
implementation.
The changes below do a few things. First it changes the few
places that relied on ARCH_HAS_NMI_WATCHDOG to use
CONFIG_X86_LOCAL_APIC (the former was an alias for the latter
anyway, so nothing unusual here). Those pieces of code were
relying more on local apic functionality the nmi watchdog
functionality, so the change should make sense.
Second, I removed the x86 implementation of
touch_nmi_watchdog(). It isn't need now, instead x86 will rely
on kernel/watchdog.c's implementation.
Third, I removed the #define ARCH_HAS_NMI_WATCHDOG itself from
x86. And tweaked the include/linux/nmi.h file to tell users to
look for an externally defined touch_nmi_watchdog in the case of
ARCH_HAS_NMI_WATCHDOG _or_ CONFIG_HARDLOCKUP_DETECTOR. This
changes removes some of the ugliness in that file.
Finally, I added a Kconfig dependency for
CONFIG_HARDLOCKUP_DETECTOR that said you can't have
ARCH_HAS_NMI_WATCHDOG _and_ CONFIG_HARDLOCKUP_DETECTOR. You can
only have one nmi_watchdog.
Tested with
ARCH=i386: allnoconfig, defconfig, allyesconfig, (various broken
configs) ARCH=x86_64: allnoconfig, defconfig, allyesconfig,
(various broken configs)
Hopefully, after this patch I won't get any more compile broken
emails. :-)
v3:
changed a couple of 'linux/nmi.h' -> 'asm/nmi.h' to pick-up correct function
prototypes when CONFIG_HARDLOCKUP_DETECTOR is not set.
Signed-off-by: Don Zickus <dzickus@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: fweisbec@gmail.com
LKML-Reference: <1293044403-14117-1-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
My patch that removed the old x86 nmi watchdog broke other
arches. This change reverts a piece of that patch and puts the
change in the correct spot.
Signed-off-by: Don Zickus <dzickus@redhat.com>
Reviewed-by: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: fweisbec@gmail.com
Cc: yinghai@kernel.org
LKML-Reference: <1291068437-5331-2-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Now that the bulk of the old nmi_watchdog is gone, remove all
the stub variables and hooks associated with it.
This touches lots of files mainly because of how the io_apic
nmi_watchdog was implemented. Now that the io_apic nmi_watchdog
is forever gone, remove all its fingers.
Most of this code was not being exercised by virtue of
nmi_watchdog != NMI_IO_APIC, so there shouldn't be anything to
risky here.
Signed-off-by: Don Zickus <dzickus@redhat.com>
Cc: fweisbec@gmail.com
Cc: gorcunov@openvz.org
LKML-Reference: <1289578944-28564-3-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Now that we have a new nmi_watchdog that is more generic and
sits on top of the perf subsystem, we really do not need the old
nmi_watchdog any more.
In addition, the old nmi_watchdog doesn't really work if you are
using the default clocksource, hpet. The old nmi_watchdog code
relied on local apic interrupts to determine if the cpu is still
alive. With hpet as the clocksource, these interrupts don't
increment any more and the old nmi_watchdog triggers false
postives.
This piece removes the old nmi_watchdog code and stubs out any
variables and functions calls. The stubs are the same ones used
by the new nmi_watchdog code, so it should be well tested.
Signed-off-by: Don Zickus <dzickus@redhat.com>
Cc: fweisbec@gmail.com
Cc: gorcunov@openvz.org
LKML-Reference: <1289578944-28564-2-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Combining the softlockup and hardlockup code causes watchdog.c
to build even without the hardlockup detection support.
So if an arch, that has the previous and the new nmi watchdog
implementations cohabiting, wants to know if the generic one
is in use, CONFIG_LOCKUP_DETECTOR is not a reliable check.
We need to use CONFIG_HARDLOCKUP_DETECTOR instead.
Fixes:
kernel/built-in.o: In function `touch_nmi_watchdog':
(.text+0x449bc): multiple definition of `touch_nmi_watchdog'
arch/sparc/kernel/built-in.o:(.text+0x11b28): first defined here
Signed-off-by: Don Zickus <dzickus@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
LKML-Reference: <20100514151121.GR15159@redhat.com>
[ use CONFIG_HARDLOCKUP_DETECTOR instead of CONFIG_PERF_EVENTS_NMI]
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
The new nmi_watchdog (which uses the perf event subsystem) is very
similar in structure to the softlockup detector. Using Ingo's
suggestion, I combined the two functionalities into one file:
kernel/watchdog.c.
Now both the nmi_watchdog (or hardlockup detector) and softlockup
detector sit on top of the perf event subsystem, which is run every
60 seconds or so to see if there are any lockups.
To detect hardlockups, cpus not responding to interrupts, I
implemented an hrtimer that runs 5 times for every perf event
overflow event. If that stops counting on a cpu, then the cpu is
most likely in trouble.
To detect softlockups, tasks not yielding to the scheduler, I used the
previous kthread idea that now gets kicked every time the hrtimer fires.
If the kthread isn't being scheduled neither is anyone else and the
warning is printed to the console.
I tested this on x86_64 and both the softlockup and hardlockup paths
work.
V2:
- cleaned up the Kconfig and softlockup combination
- surrounded hardlockup cases with #ifdef CONFIG_PERF_EVENTS_NMI
- seperated out the softlockup case from perf event subsystem
- re-arranged the enabling/disabling nmi watchdog from proc space
- added cpumasks for hardlockup failure cases
- removed fallback to soft events if no PMU exists for hard events
V3:
- comment cleanups
- drop support for older softlockup code
- per_cpu cleanups
- completely remove software clock base hardlockup detector
- use per_cpu masking on hard/soft lockup detection
- #ifdef cleanups
- rename config option NMI_WATCHDOG to LOCKUP_DETECTOR
- documentation additions
V4:
- documentation fixes
- convert per_cpu to __get_cpu_var
- powerpc compile fixes
V5:
- split apart warn flags for hard and soft lockups
TODO:
- figure out how to make an arch-agnostic clock2cycles call
(if possible) to feed into perf events as a sample period
[fweisbec: merged conflict patch]
Signed-off-by: Don Zickus <dzickus@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Cyrill Gorcunov <gorcunov@gmail.com>
Cc: Eric Paris <eparis@redhat.com>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
LKML-Reference: <1273266711-18706-2-git-send-email-dzickus@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
The original patch was x86_64 centric. Changed the code to make
it less so.
ested by building and running on a powerpc.
Signed-off-by: Don Zickus <dzickus@redhat.com>
Cc: peterz@infradead.org
Cc: gorcunov@gmail.com
Cc: aris@redhat.com
LKML-Reference: <1266013161-31197-2-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
These are the bits that enable the new nmi_watchdog and safely
isolate the old nmi_watchdog. Only one or the other can run,
not both at the same time.
Signed-off-by: Don Zickus <dzickus@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: gorcunov@gmail.com
Cc: aris@redhat.com
Cc: peterz@infradead.org
LKML-Reference: <1265424425-31562-4-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
As Andrew noted, my previous patch ("debug lockups: Improve lockup
detection") broke/removed SysRq-L support from architecture that do
not provide a __trigger_all_cpu_backtrace implementation.
Restore a fallback path and clean up the SysRq-L machinery a bit:
- Rename the arch method to arch_trigger_all_cpu_backtrace()
- Simplify the define
- Document the method a bit - in the hope of more architectures
adding support for it.
[ The patch touches Sparc code for the rename. ]
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: "David S. Miller" <davem@davemloft.net>
LKML-Reference: <20090802140809.7ec4bb6b.akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
During kernel bootup, a new T60 laptop (CoreDuo, 32-bit) hangs about
10%-20% of the time in acpi_init():
Calling initcall 0xc055ce1a: topology_init+0x0/0x2f()
Calling initcall 0xc055d75e: mtrr_init_finialize+0x0/0x2c()
Calling initcall 0xc05664f3: param_sysfs_init+0x0/0x175()
Calling initcall 0xc014cb65: pm_sysrq_init+0x0/0x17()
Calling initcall 0xc0569f99: init_bio+0x0/0xf4()
Calling initcall 0xc056b865: genhd_device_init+0x0/0x50()
Calling initcall 0xc056c4bd: fbmem_init+0x0/0x87()
Calling initcall 0xc056dd74: acpi_init+0x0/0x1ee()
It's a hard hang that not even an NMI could punch through! Frustratingly,
adding printks or function tracing to the ACPI code made the hangs go away
...
After some time an additional detail emerged: disabling the NMI watchdog
made these occasional hangs go away.
So i spent the better part of today trying to debug this and trying out
various theories when i finally found the likely reason for the hang: if
acpi_ns_initialize_devices() executes an _INI AML method and an NMI
happens to hit that AML execution in the wrong moment, the machine would
hang. (my theory is that this must be some sort of chipset setup method
doing stores to chipset mmio registers?)
Unfortunately given the characteristics of the hang it was sheer
impossible to figure out which of the numerous AML methods is impacted
by this problem.
As a workaround i wrote an interface to disable chipset-based NMIs while
executing _INI sections - and indeed this fixed the hang. I did a
boot-loop of 100 separate reboots and none hung - while without the patch
it would hang every 5-10 attempts. Out of caution i did not touch the
nmi_watchdog=2 case (it's not related to the chipset anyway and didnt
hang).
I implemented this for both x86_64 and i686, tested the i686 laptop both
with nmi_watchdog=1 [which triggered the hangs] and nmi_watchdog=2, and
tested an Athlon64 box with the 64-bit kernel as well. Everything builds
and works with the patch applied.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Cc: Len Brown <lenb@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
When a spinlock lockup occurs, arrange for the NMI code to emit an all-cpu
backtrace, so we get to see which CPU is holding the lock, and where.
Cc: Andi Kleen <ak@muc.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
touch_nmi_watchdog() calls touch_softlockup_watchdog() on both
architectures that implement it (i386 and x86_64). On other architectures
it does nothing at all. touch_nmi_watchdog() should imply
touch_softlockup_watchdog() on all architectures. Suggested by Andi Kleen.
[heiko.carstens@de.ibm.com: s390 fix]
Signed-off-by: Michal Schmidt <xschmi00@stud.feec.vutbr.cz>
Cc: Andi Kleen <ak@muc.de>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Michal Schmidt <xschmi00@stud.feec.vutbr.cz>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.
Let it rip!
Aaron Tomlin