Re: [PATCH] perf_events: improve Intel event scheduling

[Stephane Eranian]

Hi,

[Repost because of HTML]

On Mon, Dec 21, 2009 at 4:40 PM, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
>
> On Fri, 2009-12-11 at 12:59 +0100, stephane eranian wrote:
>
> > There is a major difference between PPC and X86 here. PPC has a
> > centralized register to control start/stop. This register Âuses
> > bitmask to enable or disable counters. Thus, in hw_perf_enable(), if
> > n_added=0, then you just need to use the pre-computed bitmask.
> > Otherwise, you need to recompute the bitmask to include the new
> > registers. The assignment of events and validation is done in
> > hw_group_sched_in().
> >
> > In X86, assignment and validation is done in hw_group_sched_in().
> > Activation is done individually for each counter. There is no
> > centralized register used here, thus no bitmask to update.
>
> intel core2 has the global control reg, but for all intents and purposes
> the perf_enable/disable calls emulate this global enable/disable.
>
> > Disabling a counter does not trigger a complete reschedule of events.
> > This happens only when hw_group_sched_in() is called.
> >
> > The n_events = 0 in hw_perf_disable() is used to signal that something
> > is changing. It should not be here but here. The problem is that
> > hw_group_sched_in() needs a way to know that it is called for a
> > completely new series of group scheduling so it can discard any
> > previous assignment. This goes back to the issue I raised in my
> > previous email. You could add a parameter to hw_group_sched_in() that
> > would indicate this is the first group. that would cause n_events =0
> > and the function would start accumulating events for the new
> > scheduling period.
>
> I'm not really seeing the problem here...
>
>
> Âperf_disable() <-- shut down the full pmu
>
> Âpmu->disable() <-- hey someone got removed (easy free the reg)
> Âpmu->enable() Â<-- hey someone got added (harder, check constraints)
>
> Âhw_perf_group_sched_in() <-- hey a full group got added
> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â(better than multiple ->enable)
>
> Âperf_enable() <-- re-enable pmu
>
>
> So ->disable() is used to track freeing, ->enable is used to add
> individual counters, check constraints etc..
>
> hw_perf_group_sched_in() is used to optimize the full group enable.
>

Does that mean that after a disable() I can assume that there won't
be an enable() without a group_sched_in()?

I suspect not. In fact, there is a counter-example in perf_ctx_adjust_freq().

> Afaict that is what power does (Paul?) and that should I think be
> sufficient to track x86 as well.
>
> Since sched_in() is balanced with sched_out(), the ->disable() calls
> should provide the required information as to the occupation of the pmu.
> I don't see the need for more hooks.
>
> Paul, could you comment, since you did all this for power?
>
--
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/