On Fri, Nov 08, 2024 at 09:49:29AM -0800, Ian Rogers wrote:
> Following the convention of the tool PMU, create a hwmon PMU that
> exposes hwmon data for reading. For example, the following shows
> reading the CPU temperature and 2 fan speeds alongside the uncore
> frequency:
> ```
> $ perf stat -e temp_cpu,fan1,hwmon_thinkpad/fan2/,tool/num_cpus_online/ -M UNCORE_FREQ -I 1000
> 1.001153138 52.00 'C temp_cpu
> 1.001153138 2,588 rpm fan1
> 1.001153138 2,482 rpm hwmon_thinkpad/fan2/
> 1.001153138 8 tool/num_cpus_online/
> 1.001153138 1,077,101,397 UNC_CLOCK.SOCKET # 1.08 UNCORE_FREQ
> 1.001153138 1,012,773,595 duration_time
> ...
> ```
>
> Additional data on the hwmon events is in perf list:
> ```
> $ perf list
> ...
> hwmon:
> ...
> temp_core_0 OR temp2
> [Temperature in unit coretemp named Core 0. crit=100'C,max=100'C crit_alarm=0'C. Unit:
> hwmon_coretemp]
> ...
> ```
>
> v7: Try to address feedback from Namhyung wrt size/comprehensibility
> of the hwmon PMU patch. To better explain the config encoding put
> the parser into a patch ahead of the PMU patch and add a unit
> test. This shows how a hwmon filename can be moved back-and-forth
> with a config value as a single patch (two with the test). The
> hwmon PMU patch is then putting the wrapping around that for the
> pmus', parse_event's and evsel reading's benefit.
Thanks for doing this!
Namhyung
>
> The alternate approach of not caring about the mapping, name to
> config value, and using the evsel->name wasn't followed. In the
> tool PMU it was made so we can go back-and-forth between event
> names and a type+config, this removed the tool enum value from the
> evsel as now tool events were more 1st class and evsel should
> really just be a wrapper on a perf_event_attr. Using the
> evsel->name wasn't doing that and so the approach of mapping
> filenames and configs, as better highlighted in this series was
> pursued.
>
> It is expected future PMUs won't be able to cleanly map fixed
> strings to config values, in those cases tables would be
> necessary. Making a hwmon PMU do this and then removing it to
> follow the parsing/mapping done in patch 2 here, would have been a
> lot of additional work which then would be removed.
>
> v6: Add string.h #include for issue reported by kernel test robot.
> v5: Fix asan issue in parse_hwmon_filename caught by a TMA metric.
> v4: Drop merged patches 1 to 10. Separate adding the hwmon_pmu from
> the update to perf_pmu to use it. Try to make source of literal
> strings clearer via named #defines. Fix a number of GCC warnings.
> v3: Rebase, add Namhyung's acked-by to patches 1 to 10.
> v2: Address Namhyung's review feedback. Rebase dropping 4 patches
> applied by Arnaldo, fix build breakage reported by Arnaldo.
>
> Ian Rogers (7):
> tools api io: Ensure line_len_out is always initialized
> perf hwmon_pmu: Add hwmon filename parser
> perf test: Add hwmon filename parser test
> perf hwmon_pmu: Add a tool PMU exposing events from hwmon in sysfs
> perf pmu: Add calls enabling the hwmon_pmu
> perf test: Add hwmon "PMU" test
> perf docs: Document tool and hwmon events
>
> tools/lib/api/io.h | 1 +
> tools/perf/Documentation/perf-list.txt | 15 +
> tools/perf/tests/Build | 1 +
> tools/perf/tests/builtin-test.c | 1 +
> tools/perf/tests/hwmon_pmu.c | 340 ++++++++++
> tools/perf/tests/tests.h | 1 +
> tools/perf/util/Build | 1 +
> tools/perf/util/evsel.c | 9 +
> tools/perf/util/hwmon_pmu.c | 825 +++++++++++++++++++++++++
> tools/perf/util/hwmon_pmu.h | 156 +++++
> tools/perf/util/pmu.c | 20 +
> tools/perf/util/pmu.h | 2 +
> tools/perf/util/pmus.c | 9 +
> tools/perf/util/pmus.h | 3 +
> 14 files changed, 1384 insertions(+)
> create mode 100644 tools/perf/tests/hwmon_pmu.c
> create mode 100644 tools/perf/util/hwmon_pmu.c
> create mode 100644 tools/perf/util/hwmon_pmu.h
>
> --
> 2.47.0.277.g8800431eea-goog
>