lkml.org 
[lkml]   [2011]   [May]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH 2/2] x86: Make the x86-64 stacktrace code safely callable from scheduler
From
Date
On Fri, 2011-05-13 at 14:48 +0200, Frederic Weisbecker wrote:
> I haven't observed any deadlock. trace events disable preemption and
> other tracers do too (my changelog was buggy).
>
> I just worried about potential other users, like a WARN_ON in the
> scheduler or so.
>
> My worry is the following scenario:
>
> schedule() {
> acquire(rq)
> set_tsk_need_resched
> WARN_ON() {
> stack_trace() {
> preempt_enable() {
> preempt_schedule() {

Would never happen, because rq->lock is a spinlock which holds another
preempt count so preempt_enable() would never schedule.

> acquire(rq)
> }
> }
> }
> }
> }



> I don't know if it happens that one set TIF_NEED_RESCHED remotely,

Yes

> or if TIF_NEED_RESCHED can be set when we hold the rq,

Yes

> and then we
> can be followed by a WARN_ON, ...

Not quite sure, but possible.

> So I preferred to be careful.

Still not quite seeing how all things could go bang.


\
 
 \ /
  Last update: 2011-05-13 15:23    [from the cache]
©2003-2011 Jasper Spaans