[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20190831175337.GB5447@tassilo.jf.intel.com>
Date: Sat, 31 Aug 2019 10:53:37 -0700
From: Andi Kleen <ak@...ux.intel.com>
To: Stephane Eranian <eranian@...gle.com>
Cc: "Liang, Kan" <kan.liang@...ux.intel.com>,
Peter Zijlstra <peterz@...radead.org>,
Arnaldo Carvalho de Melo <acme@...nel.org>,
Ingo Molnar <mingo@...hat.com>,
LKML <linux-kernel@...r.kernel.org>,
Thomas Gleixner <tglx@...utronix.de>,
Jiri Olsa <jolsa@...nel.org>,
Alexander Shishkin <alexander.shishkin@...ux.intel.com>
Subject: Re: [RESEND PATCH V3 3/8] perf/x86/intel: Support hardware TopDown
metrics
On Sat, Aug 31, 2019 at 02:13:05AM -0700, Stephane Eranian wrote:
> Andi,
>
> On Fri, Aug 30, 2019 at 5:31 PM Andi Kleen <ak@...ux.intel.com> wrote:
> >
> > > the same manner. It would greatly simplify the kernel implementation.
> >
> > I tried that originally. It was actually more complicated.
> >
> > You can't really do deltas on raw metrics, and a lot of the perf
> > infrastructure is built around deltas.
> >
> How is RAPL handled? No deltas there either. It uses the snapshot model.
RAPL doesn't support any context switch or CPU context.
Also it has no concept of "accumulate with clear on read"
-Andi
Powered by blists - more mailing lists