Messages in this thread | | | Date | Wed, 26 Jan 2011 22:19:31 +0100 | From | Oleg Nesterov <> | Subject | Re: Q: perf_install_in_context/perf_event_enable are racy? |
| |
On 01/26, Peter Zijlstra wrote: > > On Wed, 2011-01-26 at 20:05 +0100, Peter Zijlstra wrote: > > On Wed, 2011-01-26 at 19:49 +0100, Oleg Nesterov wrote: > > > On 01/26, Oleg Nesterov wrote: > > > > > > > > Please see the untested patch below. It doesn't change perf_event_enable(), > > > > only perf_install_in_context(). > > > > > > Forgot to mention... Also, it doesn't try to fix the race with do_exit(), > > > this needs another change. > > > > > > And, damn, can't resist. This is mostly cosmetic issue, but I feel > > > discomfort every time I look at task_oncpu_function_call(). It _looks_ > > > obviously wrong, even if the problem doesn't exist in practice. I'll > > > send the pedantic fix to keep the maintainers busy ;) > > > > I've been trying to sit down and work my way through it today, your last > > suggestion very nearly seemed to make sense, but I kept getting > > distracted. > > > > FWIW I think perf_event_enable() has the very same issue...
Yes, yes, note the "doesn't change perf_event_enable()" above.
In fact, I _suspect_ perf_event_enable() has more problems, but I need to recheck.
> +void task_function_trampoline(void *data) > +{ > + struct task_function_call *tfc = data; > + > + if (this_rq()->curr != tfc->p) > + return;
Yes, I was thinking about checking rq->curr too, but this doesn't really help. This closes the race with "prev", but we have the similar race with "next".
__perf_install_in_context() should not set ->task_ctx before next does perf_event_context_sched_in(). Otherwise it will do nothing, it checks cpuctx->task_ctx == ctx.
Oleg.
| |