Messages in this thread Patch in this message |  | | Date | Sat, 8 Oct 2016 14:54:06 +0200 (CEST) | From | Thomas Gleixner <> | Subject | [GIT pull] sched fix for 4.9 |
| |
Linus,
please pull the latest sched-urgent-for-linus git tree from:
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git sched-urgent-for-linus
A revert of a commit which pointelessly widened a preempt disabled section which in turn caused might_sleep() to trigger. The patch intended to prevent usage of smp_processor_id() in preemptible context, but the usage in that case is fine because the thread is pinned on a single cpu and therefore cannot be migrated off.
Thanks,
tglx
------------------> Ingo Molnar (1): Revert "sched/core: Do not use smp_processor_id() with preempt enabled in smpboot_thread_fn()"
kernel/smpboot.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/smpboot.c b/kernel/smpboot.c index fc0d8270f69e..13bc43d1fb22 100644 --- a/kernel/smpboot.c +++ b/kernel/smpboot.c @@ -122,12 +122,12 @@ static int smpboot_thread_fn(void *data) if (kthread_should_park()) { __set_current_state(TASK_RUNNING); + preempt_enable(); if (ht->park && td->status == HP_THREAD_ACTIVE) { BUG_ON(td->cpu != smp_processor_id()); ht->park(td->cpu); td->status = HP_THREAD_PARKED; } - preempt_enable(); kthread_parkme(); /* We might have been woken for stop */ continue;
|  |