[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAP-5=fXQGfEiMn7DVR4UX6=7rcQ02ycF2A9AA+GAGjmpD0+Vfg@mail.gmail.com>
Date: Fri, 7 Feb 2025 21:15:17 -0800
From: Ian Rogers <irogers@...gle.com>
To: "Liang, Kan" <kan.liang@...ux.intel.com>
Cc: Peter Zijlstra <peterz@...radead.org>, Ingo Molnar <mingo@...hat.com>,
Arnaldo Carvalho de Melo <acme@...nel.org>, Namhyung Kim <namhyung@...nel.org>,
Mark Rutland <mark.rutland@....com>,
Alexander Shishkin <alexander.shishkin@...ux.intel.com>, Jiri Olsa <jolsa@...nel.org>,
Adrian Hunter <adrian.hunter@...el.com>, Andreas Färber <afaerber@...e.de>,
Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>, Weilin Wang <weilin.wang@...el.com>,
linux-kernel@...r.kernel.org, linux-perf-users@...r.kernel.org,
Perry Taylor <perry.taylor@...el.com>, Samantha Alt <samantha.alt@...el.com>,
Caleb Biggers <caleb.biggers@...el.com>, Edward Baker <edward.baker@...el.com>,
Michael Petlan <mpetlan@...hat.com>, Thomas Falcon <thomas.falcon@...el.com>
Subject: Re: [PATCH v7 03/24] perf vendor events: Add Arrowlake events/metrics
On Fri, Feb 7, 2025 at 5:19 PM Liang, Kan <kan.liang@...ux.intel.com> wrote:
>
> On 2025-02-07 6:00 p.m., Ian Rogers wrote:
> > On Fri, Feb 7, 2025 at 12:44 PM Liang, Kan <kan.liang@...ux.intel.com> wrote:
> >>
> >>
> >> On 2025-02-07 1:51 p.m., Ian Rogers wrote:
> >>> + {
> >>> + "BriefDescription": "Percentage of time that the core is stalled due to a TPAUSE or UMWAIT instruction",
> >>> + "MetricExpr": "100 * cpu_atom@...IALIZATION.C01_MS_SCB@ / (8 * cpu_atom@..._CLK_UNHALTED.CORE@)",
> >>> + "MetricName": "tma_info_serialization _%_tpause_cycles",
> >> It doesn't seem a correct MetricName. A convert script bug?
> > They come from the E-core_TMA_Metrics.csv line 91:
> > https://github.com/captain5050/perfmon/blob/main/E-core_TMA_Metrics.csv
> > Info.Serialization %_Tpause_Cycles 100 * SERIALIZATION.C01_MS_SCB /
> > #SLOTS 100 * SERIALIZATION.C01_MS_SCB / #SLOTS 100 *
> > SERIALIZATION.C01_MS_SCB / #SLOTS 100 * SERIALIZATION.C01_MS_SCB /
> > #SLOTS Percentage of time that the core is stalled due to a TPAUSE or
> > UMWAIT instruction
> >
> > The % works okay in the metric name but we could swap it if you wan
>
> Yes, the % works, but there is a space between tma_info_serialization
> and _%_tpause_cycles.
>
> I'm not sure if it's an issue of E-core_TMA_Metrics.csv or the convertor
> script.
The space was in the CSV file but the converter shouldn't have passed
it through. I've sent a fix:
https://github.com/intel/perfmon/pull/286
I'll send v8 when Thomas fixes the E-core units.
Thanks,
Ian
Powered by blists - more mailing lists