[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20190429165634.GD2182@lakrids.cambridge.arm.com>
Date: Mon, 29 Apr 2019 17:56:34 +0100
From: Mark Rutland <mark.rutland@....com>
To: "Liang, Kan" <kan.liang@...ux.intel.com>
Cc: peterz@...radead.org, tglx@...utronix.de, mingo@...hat.com,
linux-kernel@...r.kernel.org, eranian@...gle.com, tj@...nel.org,
ak@...ux.intel.com
Subject: Re: [PATCH 2/4] perf: Add filter_match() as a parameter for
pinned/flexible_sched_in()
On Mon, Apr 29, 2019 at 11:31:26AM -0400, Liang, Kan wrote:
>
>
> On 4/29/2019 11:12 AM, Mark Rutland wrote:
> > On Mon, Apr 29, 2019 at 07:44:03AM -0700, kan.liang@...ux.intel.com wrote:
> > > From: Kan Liang <kan.liang@...ux.intel.com>
> > >
> > > A fast path will be introduced in the following patches to speed up the
> > > cgroup events sched in, which only needs a simpler filter_match().
> > >
> > > Add filter_match() as a parameter for pinned/flexible_sched_in().
> > >
> > > No functional change.
> >
> > I suspect that the cost you're trying to avoid is pmu_filter_match()
> > iterating over the entire group, which arm systems rely upon for correct
> > behaviour on big.LITTLE systems.
> >
> > Is that the case?
>
> No. In X86, we don't use pmu_filter_match(). The fast path still keeps this
> filter.
> perf_cgroup_match() is the one I want to avoid.
Understood; sorry for the silly question, and thanks for confirming! :)
Thanks,
Mark.
Powered by blists - more mailing lists