Currently there are users of queue_delayed_work() who specify
system_long_wq, the per-cpu workqueue. This workqueue should
be used for long per-cpu works, but queue_delayed_work()
queue the work using:
queue_delayed_work_on(WORK_CPU_UNBOUND, ...);
This would end up calling __queue_delayed_work() that does:
if (housekeeping_enabled(HK_TYPE_TIMER)) {
// [....]
} else {
if (likely(cpu == WORK_CPU_UNBOUND))
add_timer_global(timer);
else
add_timer_on(timer, cpu);
}
So when cpu == WORK_CPU_UNBOUND the timer is global and is
not using a specific CPU. Later, when __queue_work() is called:
if (req_cpu == WORK_CPU_UNBOUND) {
if (wq->flags & WQ_UNBOUND)
cpu = wq_select_unbound_cpu(raw_smp_processor_id());
else
cpu = raw_smp_processor_id();
}
Because the wq is not unbound, it takes the CPU where the timer
fired and enqueue the work on that CPU.
The consequence of all of this is that the work can run anywhere,
depending on where the timer fired.
Introduce system_dfl_long_wq in order to change, in a future step,
users that are still calling:
queue_delayed_work(system_long_wq, ...);
with the new system_dfl_long_wq instead, so that the work may
benefit from scheduler task placement.
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
show_cpu_pool_hog() and show_cpu_pools_hogs() no longer only dump CPU
hogs — since commit 8823eaef45 ("workqueue: Show all busy workers in
stall diagnostics"), they dump every in-flight worker in the pool's
busy_hash.
Rename them to show_cpu_pool_busy_workers() and
show_cpu_pools_busy_workers() to accurately describe what they do.
Also fix the pr_info() message to say "stalled worker pools" instead of
"stalled CPU-bound worker pools", since sleeping/blocked workers are now
included.
No functional change.
Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
show_cpu_pool_hog() only prints workers whose task is currently running
on the CPU (task_is_running()). This misses workers that are busy
processing a work item but are sleeping or blocked — for example, a
worker that clears PF_WQ_WORKER and enters wait_event_idle(). Such a
worker still occupies a pool slot and prevents progress, yet produces
an empty backtrace section in the watchdog output.
This is happening on real arm64 systems, where
toggle_allocation_gate() IPIs every single CPU in the machine (which
lacks NMI), causing workqueue stalls that show empty backtraces because
toggle_allocation_gate() is sleeping in wait_event_idle().
Remove the task_is_running() filter so every in-flight worker in the
pool's busy_hash is dumped. The busy_hash is protected by pool->lock,
which is already held.
Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Song Liu <song@kernel.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
When diagnosing workqueue stalls, knowing how long each in-flight work
item has been executing is valuable. Add a current_start timestamp
(jiffies) to struct worker, set it when a work item begins execution in
process_one_work(), and print the elapsed wall-clock time in show_pwq().
Unlike current_at (which tracks CPU runtime and resets on wakeup for
CPU-intensive detection), current_start is never reset because the
diagnostic cares about total wall-clock time including sleeps.
Before: in-flight: 165:stall_work_fn [wq_stall]
After: in-flight: 165:stall_work_fn [wq_stall] for 100s
Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Song Liu <song@kernel.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
The watchdog_ts name doesn't convey what the timestamp actually tracks.
This field tracks the last time a workqueue got progress.
Rename it to last_progress_ts to make it clear that it records when the
pool last made forward progress (started processing new work items).
No functional change.
Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Song Liu <song@kernel.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
pr_cont_worker_id() checks pool->flags against WQ_BH, which is a
workqueue-level flag (defined in workqueue.h). Pool flags use a
separate namespace with POOL_* constants (defined in workqueue.c).
The correct constant is POOL_BH. Both WQ_BH and POOL_BH are defined
as (1 << 0) so this has no behavioral impact, but it is semantically
wrong and inconsistent with every other pool-level BH check in the
file.
Fixes: 4cb1ef6460 ("workqueue: Implement BH workqueues to eventually replace tasklets")
Signed-off-by: Breno Leitao <leitao@debian.org>
Acked-by: Song Liu <song@kernel.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
Ordered workqueues are not exposed via sysfs because the 'max_active'
attribute changes the number actives worker. More than one active worker
can break ordering guarantees.
This can be avoided by forbidding writes the file for ordered
workqueues. Exposing it via sysfs allows to alter other attributes such
as the cpumask on which CPU the worker can run.
The 'max_active' value shouldn't be changed for BH worker because the
core never spawns additional worker and the worker itself can not be
preempted. So this make no sense.
Allow to expose ordered workqueues via sysfs if requested and forbid
changing 'max_active' value for ordered and BH worker.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Acked-by: Tejun Heo <tj@kernel.org>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
This is the exact same thing as the 'alloc_obj()' version, only much
smaller because there are a lot fewer users of the *alloc_flex()
interface.
As with alloc_obj() version, this was done entirely with mindless brute
force, using the same script, except using 'flex' in the pattern rather
than 'objs*'.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This was done entirely with mindless brute force, using
git grep -l '\<k[vmz]*alloc_objs*(.*, GFP_KERNEL)' |
xargs sed -i 's/\(alloc_objs*(.*\), GFP_KERNEL)/\1)/'
to convert the new alloc_obj() users that had a simple GFP_KERNEL
argument to just drop that argument.
Note that due to the extreme simplicity of the scripting, any slightly
more complex cases spread over multiple lines would not be triggered:
they definitely exist, but this covers the vast bulk of the cases, and
the resulting diff is also then easier to check automatically.
For the same reason the 'flex' versions will be done as a separate
conversion.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is the result of running the Coccinelle script from
scripts/coccinelle/api/kmalloc_objs.cocci. The script is designed to
avoid scalar types (which need careful case-by-case checking), and
instead replace kmalloc-family calls that allocate struct or union
object instances:
Single allocations: kmalloc(sizeof(TYPE), ...)
are replaced with: kmalloc_obj(TYPE, ...)
Array allocations: kmalloc_array(COUNT, sizeof(TYPE), ...)
are replaced with: kmalloc_objs(TYPE, COUNT, ...)
Flex array allocations: kmalloc(struct_size(PTR, FAM, COUNT), ...)
are replaced with: kmalloc_flex(*PTR, FAM, COUNT, ...)
(where TYPE may also be *VAR)
The resulting allocations no longer return "void *", instead returning
"TYPE *".
Signed-off-by: Kees Cook <kees@kernel.org>
Pull workqueue updates from Tejun Heo:
- Rework the rescuer to process work items one-by-one instead of
slurping all pending work items in a single pass.
As there is only one rescuer per workqueue, a single long-blocking
work item could cause high latency for all tasks queued behind it,
even after memory pressure is relieved and regular kworkers become
available to service them.
- Add CONFIG_BOOTPARAM_WQ_STALL_PANIC build-time option and
workqueue.panic_on_stall_time parameter for time-based stall panic,
giving systems more control over workqueue stall handling.
- Replace BUG_ON() with panic() in the stall panic path for clearer
intent and more informative output.
* tag 'wq-for-6.20' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
workqueue: replace BUG_ON with panic in panic_on_wq_watchdog
workqueue: add time-based panic for stalls
workqueue: add CONFIG_BOOTPARAM_WQ_STALL_PANIC option
workqueue: Process extra works in rescuer on memory pressure
workqueue: Process rescuer work items one-by-one using a cursor
workqueue: Make send_mayday() take a PWQ argument directly
Replace BUG_ON() with panic() in panic_on_wq_watchdog(). This is not
a bug condition but a deliberate forced panic requested by the user
via module parameters to crash the system for debugging purposes.
Using panic() instead of BUG_ON() makes this intent clearer and provides
more informative output about which threshold was exceeded and the actual
values, making it easier to diagnose the stall condition from crash dumps.
Signed-off-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
Add a new module parameter 'panic_on_stall_time' that triggers a panic
when a workqueue stall persists for longer than the specified duration
in seconds.
Unlike 'panic_on_stall' which counts accumulated stall events, this
parameter triggers based on the duration of a single continuous stall.
This is useful for catching truly stuck workqueues rather than
accumulating transient stalls.
Usage:
workqueue.panic_on_stall_time=120
This would panic if any workqueue pool has been stalled for 120 seconds
or more.
The stall duration is measured from the workqueue last progress
(poll_ts) which accounts for legitimate system stalls.
Signed-off-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
Add a kernel config option to set the default value of
workqueue.panic_on_stall, similar to CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC,
CONFIG_BOOTPARAM_HARDLOCKUP_PANIC and CONFIG_BOOTPARAM_HUNG_TASK_PANIC.
This allows setting the number of workqueue stalls before triggering
a kernel panic at build time, which is useful for high-availability
systems that need consistent panic-on-stall, in other words, those
servers which run with CONFIG_BOOTPARAM_*_PANIC=y already.
The default remains 0 (disabled). Setting it to 1 will panic on the
first stall, and higher values will panic after that many stall
warnings. The value can still be overridden at runtime via the
workqueue.panic_on_stall boot parameter or sysfs.
Signed-off-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
Make the rescuer process more work on the last pwq when there are no
more to rescue for the whole workqueue to help the regular workers in
case it is a temporary memory pressure relief and to reduce relapse.
Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Previously, the rescuer scanned for all matching work items at once and
processed them within a single rescuer thread, which could cause one
blocking work item to stall all others.
Make the rescuer process work items one-by-one instead of slurping all
matches in a single pass.
Break the rescuer loop after finding and processing the first matching
work item, then restart the search to pick up the next. This gives
normal worker threads a chance to process other items which gives them
the opportunity to be processed instead of waiting on the rescuer's
queue and prevents a blocking work item from stalling the rest once
memory pressure is relieved.
Introduce a dummy cursor work item to avoid potentially O(N^2)
rescans of the work list. The marker records the resume position for
the next scan, eliminating redundant traversals.
Also introduce RESCUER_BATCH to control the maximum number of work items
the rescuer processes in each turn, and move on to other PWQs when the
limit is reached.
Cc: ying chen <yc1082463@gmail.com>
Reported-by: ying chen <yc1082463@gmail.com>
Fixes: e22bee782b ("workqueue: implement concurrency managed dynamic worker pool")
Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Make send_mayday() operate on a PWQ directly instead of taking a work
item, so that rescuer_thread() now calls send_mayday(pwq) instead of
open-coding the mayday list manipulation.
Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
The commit1 def98c84b6 ("workqueue: Fix spurious sanity check failures
in destroy_workqueue()") tries to fix spurious sanity check failures by
stopping send_mayday() via setting wq->rescuer to NULL.
But it fails to stop the pwq->mayday_node requeuing in the rescuer, and
the commit2 e66b39af00 ("workqueue: Fix pwq ref leak in
rescuer_thread()") fixes it by checking wq->rescuer which is the result
of commit1.
Both commits together really fix spurious sanity check failures caused
by the rescuer, but they both use a convoluted method by relying on
wq->rescuer state rather than the real count of work items.
Actually __WQ_DESTROYING and drain_workqueue() together already stop
send_mayday() by draining all the work items and ensuring no new work
item requeuing.
And the more proper fix to stop the pwq->mayday_node requeuing in the
rescuer is from commit3 4f3f4cf388 ("workqueue: avoid unneeded
requeuing the pwq in rescuer thread") and renders the checking of
wq->rescuer in commit2 unnecessary.
So __WQ_DESTROYING, drain_workqueue() and commit3 together fix spurious
sanity check failures introduced by the rescuer.
Just remove the convoluted code of using wq->rescuer.
Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
If the pwq does not need rescue (normal workers have been created or
become available), the rescuer can immediately move on to other stalled
pwqs.
Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Move the code to assign work to rescuer and assign_rescuer_work().
Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
The affinity to set to the rescuers should be consistent in all paths
when a rescuer is in detached state. The affinity could be either
wq_unbound_cpumask or unbound_effective_cpumask(wq).
Related paths:
rescuer's worker_detach_from_pool()
update wq_unbound_cpumask
update wq's cpumask
init_rescuer()
Both affinities are Ok as long as they are consistent in all paths.
In the commit 449b31ad29 ("workqueue: Init rescuer's affinities as
the wq's effective cpumask") makes init_rescuer use
unbound_effective_cpumask(wq) which is consistent with then
apply_wqattrs_commit().
But using unbound_effective_cpumask(wq) requres much more code to
maintain the consistency, and it doesn't make much sense since the
affinity is only effective when the rescuer is not processing works.
wq_unbound_cpumask is more favorable.
So apply_wqattrs_commit() and the path of "updating wq's cpumask" had
been changed to not update the rescuer's affinity, and both the paths
of "updating wq_unbound_cpumask" and "rescuer's
worker_detach_from_pool()" had been changed to use wq_unbound_cpumask.
Now, make init_rescuer() use wq_unbound_cpumask for rescuer's affinity
and make all the paths consistent.
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Waiman Long <longman@redhat.com>
Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
When workqueue cpumask changes are committed, the DISASSOCIATED workers
affinity is not touched and this might be a problem down the line for
isolated setups when the DISASSOCIATED pools still have works to run
after the cpu is offline.
Make sure the workers' affinity is updated every time a workqueue cpumask
changes, so these workers can't break isolation.
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Waiman Long <longman@redhat.com>
Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
When a rescuer is attached to a pool, its affinity should be only
managed by the pool.
But updating the detached rescuer's affinity is still meaningful so
that it will not disrupt isolated CPUs when it is to be waken up.
But the commit d64f2fa064 ("kernel/workqueue: Let rescuers follow
unbound wq cpumask changes") updates the affinity unconditionally, and
causes some issues
1) it also changes the affinity when the rescuer is already attached to
a pool, which violates the affinity management.
2) the said commit tries to update the affinity of the rescuers, but it
misses the rescuers of the PERCPU workqueues, and isolated CPUs can
be possibly disrupted by these rescuers when they are summoned.
3) The affinity to set to the rescuers should be consistent in all paths
when a rescuer is in detached state. The affinity could be either
wq_unbound_cpumask or unbound_effective_cpumask(wq). Related paths:
rescuer's worker_detach_from_pool()
update wq_unbound_cpumask
update wq's cpumask
init_rescuer()
Both affinities are Ok as long as they are consistent in all paths.
But using unbound_effective_cpumask(wq) requres much more code to
maintain the consistency, and it doesn't make much sense since the
affinity is only effective when the rescuer is not processing works.
wq_unbound_cpumask is more favorable.
Fix the 1) issue by testing rescuer->pool before updating with
wq_pool_attach_mutex held.
Fix the 2) issue by moving the rescuer's affinity updating code to
the place updating wq_unbound_cpumask and make it also update for
PERCPU workqueues.
Partially cleanup the 3) consistency issue by using wq_unbound_cpumask.
So that the path of "updating wq's cpumask" doesn't need to maintain it.
and both the paths of "updating wq_unbound_cpumask" and "rescuer's
worker_detach_from_pool()" use wq_unbound_cpumask.
Cleanup for init_rescuer()'s consistency for affinity can be done in
future.
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Waiman Long <longman@redhat.com>
Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Marco CrivellariandTejun Heo
Breno LeitaoandTejun Heo
Sebastian Andrzej SiewiorandTejun Heo
Linus Torvalds
Kees Cook
Frederic Weisbecker
Lai JiangshanandTejun Heo
zhang jiaoandTejun Heo