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>
commit 672917dcc7 ("cpuidle: menu governor: reduce latency on exit")
added an optimization, where the analysis on the past idle period moved
from the end of idle, to the beginning of the new idle.
Unfortunately, this optimization had a bug where it zeroed one key
variable for new use, that is needed for the analysis. The fix is
simple, zero the variable after doing the work from the previous idle.
During the audit of the code that found this issue, another issue was
also found; the ->measured_us data structure member is never set, a
local variable is always used instead.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Corrado Zoccolo <czoccolo@gmail.com>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'for-linus' of git://neil.brown.name/md:
md: restore ability of spare drives to spin down.
md/raid6: Fix raid-6 read-error correction in degraded state
* 'urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/brodo/pcmcia-2.6:
pcmcia: fix compilation after 16bit state locking changes
pcmcia: order userspace suspend and resume requests
pcmcia: avoid pccard_validate_cis failure in resume callpath
* 'for-linus' of git://git.kernel.dk/linux-2.6-block:
blk-cgroup: Fix an RCU warning in blkiocg_create()
blk-cgroup: Fix RCU correctness warning in cfq_init_queue()
drbd: don't expose failed local READ to upper layers
* 'bugfixes' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6:
NFS: Fix RCU issues in the NFSv4 delegation code
NFSv4: Fix the locking in nfs_inode_reclaim_delegation()
* 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-acpi-2.6:
ACPI: sleep: init_set_sci_en_on_resume for Dell Studio 155x
ACPI: fix acpi_hest_firmware_first_pci() caused oops
sbshc: acpi_device_class "smbus_host_controller" too long
power_meter: acpi_device_class "power_meter_resource" too long
acpi_pad: "processor_aggregator" name too long
PNP: don't check for conflicts with bridge windows
ACPI: DMI init_set_sci_en_on_resume for multiple Lenovo ThinkPads
PNPACPI: compute Address Space length rather than using _LEN
ACPI: silence kmemcheck false positive
* 'v4l_for_2.6.34' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6:
V4L/DVB: pxa_camera: move fifo reset direct before dma start
V4L/DVB: video: testing unsigned for less than 0
V4L/DVB: mx1-camera: compile fix
V4L/DVB: budget: Oops: "BUG: unable to handle kernel NULL pointer dereference"
V4L/DVB: ngene: Workaround for stuck DiSEqC pin
V4L/DVB: saa7146: fix regression of the av7110/budget-av driver
V4L/DVB: v4l: fix config dependencies: mxb and saa7191 are V4L2 drivers, not V4L1
V4L/DVB: feature-removal: announce videotext.h removal
V4L/DVB: V4L - vpfe capture - fix for kernel crash
V4L/DVB: gspca: make usb id 0461:0815 get handled by the right driver
V4L/DVB: gspca - stv06xx: Remove the 046d:08da from the stv06xx driver
V4L/DVB: gspca - sn9c20x: Correct onstack wait_queue_head declaration
V4L/DVB: saa7146: fix up bytesperline if it is an impossible value
V4L/DVB: V4L: vpfe_capture - free ccdc_lock when memory allocation fails
V4L/DVB: V4L - Makfile:Removed duplicate entry of davinci
V4L/DVB: omap24xxcam: potential buffer overflow
* 'core-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
rcu: create rcu_my_thread_group_empty() wrapper
memcg: css_id() must be called under rcu_read_lock()
cgroup: Check task_lock in task_subsys_state()
sched: Fix an RCU warning in print_task()
cgroup: Fix an RCU warning in alloc_css_id()
cgroup: Fix an RCU warning in cgroup_path()
KEYS: Fix an RCU warning in the reading of user keys
KEYS: Fix an RCU warning
Some time ago we stopped the clean/active metadata updates
from being written to a 'spare' device in most cases so that
it could spin down and say spun down. Device failure/removal
etc are still recorded on spares.
However commit 51d5668cb2 broke this 50% of the time,
depending on whether the event count is even or odd.
The change log entry said:
This means that the alignment between 'odd/even' and
'clean/dirty' might take a little longer to attain,
how ever the code makes no attempt to create that alignment, so it
could take arbitrarily long.
So when we find that clean/dirty is not aligned with odd/even,
force a second metadata-update immediately. There are already cases
where a second metadata-update is needed immediately (e.g. when a
device fails during the metadata update). We just piggy-back on that.
Reported-by: Joe Bryant <tenminjoe@yahoo.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Cc: stable@kernel.org
with CONFIG_PROVE_RCU=y, a warning can be triggered:
# mount -t cgroup -o blkio xxx /mnt
# mkdir /mnt/subgroup
...
kernel/cgroup.c:4442 invoked rcu_dereference_check() without protection!
...
To fix this, we avoid caling css_depth() here, which is a bit simpler
than the original code.
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
It's unused and buggy in its current form, since it can place a bo
in the reserved state without removing it from lru lists.
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Bring radeon up to speed with the async event synchronization for
drmWaitVblank. See c9a9c5e02a for
more information. Without this patch event never get delivered
to userspace client.
Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Move the fifo reset from pxa_camera_start_capture to pxa_camera_irq direct
before the dma start after an end of frame interrupt to prevent images from
shifting because of old data at the begin of the frame.
Signed-off-by: Stefan Herbrechtsmeier <hbmeier@hni.uni-paderborn.de>
Acked-by: Robert Jarzmik <robert.jarzmik@free.fr>
Tested-by: Antonio Ospite <ospite@studenti.unina.it>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
soc_mbus_bytes_per_line() returns -EINVAL on error but we store it in an
unsigned int so the test for less than zero doesn't work. I think it
always returns "small" positive values so we can just cast it to int
here.
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Never call dvb_frontend_detach if we failed to attach a frontend. This fixes
the following oops, which will be triggered by a missing stv090x module:
[ 8.172997] DVB: registering new adapter (TT-Budget S2-1600 PCI)
[ 8.209018] adapter has MAC addr = 00:d0:5c:cc:a7:29
[ 8.328665] Intel ICH 0000:00:1f.5: PCI INT B -> GSI 17 (level, low) -> IRQ 17
[ 8.328753] Intel ICH 0000:00:1f.5: setting latency timer to 64
[ 8.562047] DVB: Unable to find symbol stv090x_attach()
[ 8.562117] BUG: unable to handle kernel NULL pointer dereference at 000000ac
[ 8.562239] IP: [<e08b04a3>] dvb_frontend_detach+0x4/0x67 [dvb_core]
Ref http://bugs.debian.org/575207
Signed-off-by: Bjørn Mork <bjorn@mork.no>
Cc: stable@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Don Zickus
Frederic Weisbecker
Linus Torvalds
Arjan van de Ven
Stijn Tintel
NeilBrown
Gabriele A. Trombetti
Li Zefan
Len Brown
Thomas Hellstrom
Jerome Glisse
Stefan Herbrechtsmeier
Dan Carpenter
Uwe Kleine-König
Bjørn Mork