Re: [PATCH v1 00/25] Lazily load PMU data

[Ian Rogers]

On Wed, Aug 23, 2023 at 10:11 AM Arnaldo Carvalho de Melo
<acme@xxxxxxxxxx> wrote:
>
> Em Wed, Aug 23, 2023 at 09:45:50AM -0700, Ian Rogers escreveu:
> > On Wed, Aug 23, 2023, 8:56 AM Arnaldo Carvalho de Melo <acme@xxxxxxxxxx>
> > wrote:
> >
> > > Em Wed, Aug 23, 2023 at 01:08:03AM -0700, Ian Rogers escreveu:
> > > > Lazily load PMU data both from sysfs and json files. Reorganize
> > > > json data to be more PMU oriented to facilitate this, for
> > > > example, json data is now sorted into arrays for their PMU.
> > > >
> > > > In refactoring the code some changes were made to get rid of maximum
> > > > encoding sizes for events (256 bytes), with input files being directly
> > > > passed to the lex generated code. There is also a small event parse
> > > > error message improvement.
> > > >
> > > > Some results from an Intel tigerlake laptop running Debian:
> > > >
> > > > Binary size reduction of 1.4% or 143,264 bytes because the PMU
> > > > name no longer appears in the string.
> > > >
> > > > stat -e cpu/cycles/ minor faults reduced from 1733 to 1667, open calls
> > > reduced
> > > > from 171 to 94.
> > > >
> > > > stat default minor faults reduced from 1085 to 1727, open calls reduced
> > > > from 654 to 343.
> > > >
> > > > Average PMU scanning reduced from 4720.641usec to 2927.293usec.
> > > > Average core PMU scanning reduced from 1004.658usec to 232.668usec
> > > > (4.3x faster).
> > >
> > > I'm now chasing this one when building it on ubuntu arm64
> > >
> >
> > I'll fix and send a v2.
>
> Its fixed already, I'm pushing it to tmp.perf-tools-next

Thanks!
Ian

> diff --git a/tools/perf/arch/arm/util/cs-etm.c b/tools/perf/arch/arm/util/cs-etm.c
> index 7c51fa182b51dab0..b8d6a953fd7423e1 100644
> --- a/tools/perf/arch/arm/util/cs-etm.c
> +++ b/tools/perf/arch/arm/util/cs-etm.c
> @@ -79,9 +79,9 @@ static int cs_etm_validate_context_id(struct auxtrace_record *itr,
>         int err;
>         u32 val;
>         u64 contextid = evsel->core.attr.config &
> -               (perf_pmu__format_bits(&cs_etm_pmu->format, "contextid") |
> -                perf_pmu__format_bits(&cs_etm_pmu->format, "contextid1") |
> -                perf_pmu__format_bits(&cs_etm_pmu->format, "contextid2"));
> +               (perf_pmu__format_bits(cs_etm_pmu, "contextid") |
> +                perf_pmu__format_bits(cs_etm_pmu, "contextid1") |
> +                perf_pmu__format_bits(cs_etm_pmu, "contextid2"));
>
>         if (!contextid)
>                 return 0;
> @@ -106,7 +106,7 @@ static int cs_etm_validate_context_id(struct auxtrace_record *itr,
>         }
>
>         if (contextid &
> -           perf_pmu__format_bits(&cs_etm_pmu->format, "contextid1")) {
> +           perf_pmu__format_bits(cs_etm_pmu, "contextid1")) {
>                 /*
>                  * TRCIDR2.CIDSIZE, bit [9-5], indicates whether contextID
>                  * tracing is supported:
> @@ -122,7 +122,7 @@ static int cs_etm_validate_context_id(struct auxtrace_record *itr,
>         }
>
>         if (contextid &
> -           perf_pmu__format_bits(&cs_etm_pmu->format, "contextid2")) {
> +           perf_pmu__format_bits(cs_etm_pmu, "contextid2")) {
>                 /*
>                  * TRCIDR2.VMIDOPT[30:29] != 0 and
>                  * TRCIDR2.VMIDSIZE[14:10] == 0b00100 (32bit virtual contextid)
> @@ -151,7 +151,7 @@ static int cs_etm_validate_timestamp(struct auxtrace_record *itr,
>         u32 val;
>
>         if (!(evsel->core.attr.config &
> -             perf_pmu__format_bits(&cs_etm_pmu->format, "timestamp")))
> +             perf_pmu__format_bits(cs_etm_pmu, "timestamp")))
>                 return 0;
>
>         if (!cs_etm_is_etmv4(itr, cpu)) {