Re: PERF_EVENT_IOC_SET_OUTPUT

From: Frederic Weisbecker
Date: Wed Oct 02 2013 - 07:43:41 EST


On Wed, Oct 02, 2013 at 01:27:30PM +0200, Peter Zijlstra wrote:
> On Wed, Oct 02, 2013 at 12:29:56PM +0200, Frederic Weisbecker wrote:
> > On Wed, Oct 02, 2013 at 12:03:50PM +0200, Peter Zijlstra wrote:
> > > On Tue, Oct 01, 2013 at 10:11:56PM +0300, Adrian Hunter wrote:
> > > > Hi
> > > >
> > > > It does not seem possible to use set-output between
> > > > task contexts of different types (e.g. a software event
> > > > to a hardware event)
> > > >
> > > > If you look at perf_event_set_output():
> > > >
> > > > /*
> > > > * If its not a per-cpu rb, it must be the same task.
> > > > */
> > > > if (output_event->cpu == -1 && output_event->ctx != event->ctx)
> > > > goto out;
> > > >
> > > > ctx (perf_event_context) won't be the same for events
> > > > of different types. Is this restriction necessary?
> > >
> > > Hmm.. so last night I wrote me a big reply saying we couldn't do it;
> > > then this morning I reconsidered and thing that something like:
> > >
> > > output_event->ctx->task != event->ctx->task
> > >
> > > should actually work.
> > >
> > > The reason it should be OK I think is because perf_mmap() will refuse to
> > > create a buffer for inherited events that have ->cpu == -1.
> > >
> > > My initial response was going to say that it wouldn't be possible
> > > because __perf_event_task_sched_out() could 'break' one ctx while still
> > > swapping the other, at which point the buffer would have to service two
> > > different tasks, potentially from different CPUs and with the buffers
> > > not actually being SMP safe that's a problem.
> >
> > I don't get what you mean with breaking or swapping a ctx.
> > But I can confirm that perf_mmap() won't allow a buffer to be remotely
> > accessed from another CPU. Now there may be other issues than locality which
> > I'm missing :)
>
> The way we 'optimize' context switches between tasks with identical
> contexts is to simply swap the context and leave the hardware alone.
>
> So counters belonging to prev will then belong to next and vice versa.
> This avoids having to read hardware counters, update stats, removes
> counters from hardware, and re-program hardware with possible the exact
> same set.
>
> When a child context changes its context (eg, inserts or removes a
> counter) we break this swapping because now the contexts don't match
> anymore and we have to take the slow and painful way of prodding
> hardware.

Ah right, I remember that now. This caused me quite some headaches
a few years ago :)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/