Re: [PATCH 0/4] TopDown metrics support for Ice Lake (perf tool)

[Liang, Kan]

On 8/26/2020 11:54 AM, Jiri Olsa wrote:
> On Thu, Aug 20, 2020 at 09:45:28AM -0700, kan.liang@xxxxxxxxxxxxxxx wrote:
> > From: Kan Liang <kan.liang@xxxxxxxxxxxxxxx>
> >
> > The kernel patches have been merged into the tip's perf/core branch.
> > The patch set is on top of commit 2cb5383b30d4 ("perf/x86/intel: Support
> > per-thread RDPMC TopDown metrics") of the tip's perf/core branch.
> >
> > The changes for the perf tool include:
> > - Extend --topdown option to support per thread TopDown metrics
> > - Support sample-read topdown metric group
> > - Add a complete document for the TopDown usage.
> >
> > Ice Lake has support for measuring the level 1 TopDown metrics
> > directly in hardware. This is implemented by an additional METRICS
> > register, and a new Fixed Counter 3 that measures pipeline SLOTS.
> >
> > New in Icelake
> > - Do not require generic counters. This allows to collect TopDown always
> >    in addition to other events.
> > - Measuring TopDown per thread/process instead of only per core
> >
> > For the Ice Lake implementation of performance metrics, the values in
> > PERF_METRICS MSR are derived from fixed counter 3. Software should start
> > both registers, PERF_METRICS and fixed counter 3, from zero.
> > Additionally, software is recommended to periodically clear both
> > registers in order to maintain accurate measurements. The latter is
> > required for certain scenarios that involve sampling metrics at high
> > rates. Software should always write fixed counter 3 before write to
> > PERF_METRICS.
> >
> > IA32_PERF_GLOBAL_STATUS. OVF_PERF_METRICS[48]: If this bit is set,
> > it indicates that some PERF_METRICS-related counter has overflowed and
> > a PMI is triggered. Software has to synchronize, e.g. re-start,
> > PERF_METRICS as well as fixed counter 3. Otherwise, PERF_METRICS may
> > return invalid values.
> >
> > Limitation
> > - To get accurate result and avoid reading the METRICS register multiple
> >    times, the TopDown metrics events and SLOTS event have to be in the
> >    same group.
> > - METRICS and SLOTS registers have to be cleared after each read by SW.
> >    That is to prevent the lose of precision.
> > - Cannot do sampling read SLOTS and TopDown metric events
> >
> > Please refer SDM Vol3, 18.3.9.3 Performance Metrics for the details of
> > TopDown metrics.
> >
> > Andi Kleen (2):
> >    perf stat: Support new per thread TopDown metrics
> >    perf, tools: Add documentation for topdown metrics
> >
> > Kan Liang (2):
> >    perf tools: Rename group to topdown
> >    perf record: Support sample-read topdown metric group
>
> I don't have Ice lake to actualy check, but it looks good
>
> Acked-by: Jiri Olsa <jolsa@xxxxxxxxxx>


Thanks Jirka for the review.

Hi Arnaldo,

Andi has a comment regarding a grammar error in the printf message. I 
once plan to apply a fix in the next version. I'm not sure whether you 
will have more comments for this patch set.
If yes, I will wait for the comments.
If not, should I send a V2 patch set with this fixed?

On 8/20/2020 4:05 PM, Andi Kleen wrote:
>> +				fprintf(stat_config.output,
>> +					"Topdown accuracy may decreases when measuring long period.\n"
> "may decrease" ... "long periods".
>
> (s needs to move to the end)


Thanks,
Kan