Messages in this thread | | | Date | Wed, 30 Mar 2011 21:53:54 +0200 | From | Oleg Nesterov <> | Subject | Re: [PATCH,RFC] perf: panic due to inclied cpu context task_ctx value |
| |
On 03/30, Paul E. McKenney wrote: > > On Wed, Mar 30, 2011 at 06:37:30PM +0200, Oleg Nesterov wrote: > > > > So. synchronize_sched() above should ensure that all CPUs do context > > switch at least once (ignoring idle). And I _thought_ that in practice > > this should work. > > > > But, unles I misread the comment above synchronize_sched(), it seems that > > it only guarantees the end of "everything" which disables preemption, > > explicitly or not. IOW, say, in theory rcu_read_unlock_sched() could > > trigger ->passed_quiesc == T without reschedule. > > For rcu_read_lock() in preemptible RCU, this is true.
Hmm, not sure I understand... Do you mean that with the current implementation rcu_read_unlock() can imply rcu_sched_qs() without rescheduling ?
> But for > rcu_read_unlock_sched(), the only way rcu_note_context_switch() is called > is if the code is preempted, which ends up calling schedule().
Indeed, that is why I thought synchronize_sched() can help in this case. I meant, according to the documentation it could in theory.
But,
> > Oh, and this is not theoretical, afaics. run_ksoftirqd() does > > rcu_note_context_switch(). > > Interesting... Color me confused. > > Suppose the rcu_note_context_switch() in run_ksoftirqd() was replaced > with schedule(). This has to be OK, right? But schedule() itself > invokes rcu_note_context_switch(). So if it is OK to call schedule(), > it should be OK to call rcu_note_context_switch() directly, right? > > So, what am I missing here?
It is me, not you.
Damn. It is even worse than I thought. Somehow I missed the simple fact, schedule() does not necessarily mean context_switch(). So the idea to use synchronize_sched() was simply wrong. Sorry to all for wasting your time ;)
> > So, I think we need something else :/ > > The thing that I would be more concerned about is the idle loop. > If a CPU is in the idle loop, then rcu_sched_qs() will be invoked > (and which is invoked by rcu_note_context_switch()). So is it > illegal to use the above in the idle loop?
Not sure I understand what you mean, but the idle loop is fine. An idle thread can't have the counters attached, we don't care about them.
Thanks Paul,
Oleg.
| |