sched/core: Add missing completion for affine_move_task() waiters
PATCH 1/1 sched/core: Add missing completion for affine_move_task() waiters
Qian reported that some fuzzer issuing sched_setaffinity() ends up stuck on a wait_for_completion(). The problematic pattern seems to be:
affine_move_task()
// task_running() case
stop_one_cpu();
wait_for_completion(&pending->done);Combined with, on the stopper side:
migration_cpu_stop()
// Task moved between unlocks and scheduling the stopper
task_rq(p) != rq &&
// task_running() case
dest_cpu >= 0
=> no complete_all()This can happen with both PREEMPT and !PREEMPT, although !PREEMPT should be more likely to see this given the targeted task has a much bigger window to block and be woken up elsewhere before the stopper runs.
Make migration_cpu_stop() always look at pending affinity requests; signal their completion if the stopper hits a rq mismatch but the task is still within its allowed mask. When Migrate-Disable isn't involved, this matches the previous set_cpus_allowed_ptr() vs migration_cpu_stop() behaviour.
--- kernel/sched/core.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-)
kernel/sched/core.c +12 -1
@@ -1923,7 +1923,7 @@ static int migration_cpu_stop(void *data)else p->wake_cpu = dest_cpu; - }else if (dest_cpu <0 ) { + }else if (dest_cpu <0 || pending) {/* * This happens when we get migrated between migrate_enable()'s * preempt_enable() and scheduling the stopper task. At that
@@ -1933,6 +1933,17 @@ static int migration_cpu_stop(void *data)* more likely. */ +/* +* The task moved before the stopper got to run. We're holding +* ->pi_lock, so the allowed mask is stable - if it got +* somewhere allowed, we're done. +*/ +if (pending &&cpumask_test_cpu (task_cpu (p), p->cpus_ptr)) { + p->migration_pending =NULL ; + complete = true; + goto out; + } +/* * When this was migrate_enable() but we no longer have an * @pending, a concurrent SCA 'fixed' things and we should be
Signature · 1 line
2.27.0
Hi,
Quoted 68 lines · Valentin Schneider
Qian reported that some fuzzer issuing
sched_setaffinity()ends up stuck on await_for_completion(). The problematic pattern seems to be:
affine_move_task()// task_running() case stop_one_cpu(); wait_for_completion(&pending->done);Combined with, on the stopper side:
migration_cpu_stop()// Task moved between unlocks and scheduling the stopper task_rq(p) != rq && // task_running() case dest_cpu >= 0 => no complete_all()This can happen with both PREEMPT and !PREEMPT, although !PREEMPT should be more likely to see this given the targeted task has a much bigger window to block and be woken up elsewhere before the stopper runs.
Make
migration_cpu_stop()always look at pending affinity requests; signal their completion if the stopper hits a rq mismatch but the task is still within its allowed mask. When Migrate-Disable isn't involved, this matches the previousset_cpus_allowed_ptr()vsmigration_cpu_stop()behaviour.Fixes 6d337eab041d ("sched: Fix migrate_disable() vs set_cpus_allowed_ptr()")Reported-by Qian Cai <[email protected]>Signed-off-by Valentin Schneider <[email protected]>--- kernel/sched/core.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-)kernel/sched/core.c +12
-1@@ -1923,7 +1923,7 @@ static int migration_cpu_stop(void *data)else p->wake_cpu = dest_cpu; - }else if (dest_cpu <0 ) { + }else if (dest_cpu <0 || pending) {/* * This happens when we get migrated between migrate_enable()'s * preempt_enable() and scheduling the stopper task. At that @@ -1933,6 +1933,17 @@ static int migration_cpu_stop(void *data)* more likely. */ +/* +* The task moved before the stopper got to run. We're holding +* ->pi_lock, so the allowed mask is stable - if it got +* somewhere allowed, we're done. +*/ +if (pending &&cpumask_test_cpu (task_cpu (p), p->cpus_ptr)) { + p->migration_pending =NULL ; + complete = true; + goto out; + } +/* * When this was migrate_enable() but we no longer have an * @pending, a concurrent SCA 'fixed' things and we should be Signature · 1 line
2.27.0
Oh, I did not receive this patch from '[email protected]' account. Checked that you sent the patch to that mail address from web. If 'ouwen210' is not a good mail account name(I have used this name since 2002), I will change to use this one(Now is smooth enough and can go to lkml).
Thanks,
Tao
Quoted 6 lines · Tao Zhou
Oh, I did not receive this patch from '[email protected]' account. Checked that you sent the patch to that mail address from web. If 'ouwen210' is not a good mail account name(I have used this name since 2002), I will change to use this one(Now is smooth enough and can go to lkml).
Nah, any address is fine. I grabbed whatever emails were cc'd to the patch this is fixing - just make sure you're using the right address (i.e. the one you want LKML stuff to go to) when interacting on the list and you should be good.
Quoted 2 lines
Thanks,
Tao