| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <CAP-5=fWGrzsO15kszYvTGBus+ojxJg3bQMntJrcsfhpAAjiBmA@mail.gmail.com> Date: Wed, 8 Jan 2025 08:18:45 -0800 From: Ian Rogers <irogers@...gle.com> To: Arnaldo Carvalho de Melo <acme@...nel.org>, Caleb Biggers <caleb.biggers@...el.com> Cc: Peter Zijlstra <peterz@...radead.org>, Ingo Molnar <mingo@...hat.com>, 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>, Kan Liang <kan.liang@...ux.intel.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>, Edward Baker <edward.baker@...el.com>, Michael Petlan <mpetlan@...hat.com> Subject: Re: [PATCH v1 01/22] perf vendor events: Update Alderlake events/metrics On Wed, Jan 8, 2025 at 6:51 AM Arnaldo Carvalho de Melo <acme@...nel.org> wrote: > > On Mon, Dec 09, 2024 at 02:27:38PM -0800, Ian Rogers wrote: > > Update events from v1.27 to v1.28. > > Update TMA metrics from 4.8 to 5.01. > > > > Bring in the event updates v1.28: > > https://github.com/intel/perfmon/commit/801f43f22ec6bd23fbb5d18860f395d61e7f4081 > > > > The TMA 5.01 update is from (with subsequent fixes): > > https://github.com/intel/perfmon/commit/1d72913b2d938781fb28f3cc3507aaec5c22d782 > > On my workstation this patch is causing: > > ⬢ [acme@...lbox perf-tools-next]$ perf stat sleep 1 > event syntax error: '..ED.L3_HIT/,cpu/DTLB_LOAD_MISSES.STLB_HIT,cmask=0x1,metric-id=cpu!3DTLB_LOAD_MISSES.STLB_HIT!0cmask!20x1!3/,topdown-mem-bound/metric-id=topdown!1mem!1bound/,cpu/UOPS_..' > \___ Bad event or PMU > > Unable to find PMU or event on a PMU of 'cpu' > > Performance counter stats for 'sleep 1': > > 0.79 msec task-clock:u # 0.001 CPUs utilized > 0 context-switches:u # 0.000 /sec > 0 cpu-migrations:u # 0.000 /sec > 68 page-faults:u # 86.026 K/sec > 381,023 cpu_atom/instructions/u # 0.36 insn per cycle (60.24%) > 92,367 cpu_core/instructions/u # 0.44 insn per cycle (39.76%) > 1,064,063 cpu_atom/cycles/u # 1.346 GHz (60.24%) > 209,521 cpu_core/cycles/u # 0.265 GHz (39.76%) > 82,585 cpu_atom/branches/u # 104.477 M/sec (60.24%) > 19,921 cpu_core/branches/u # 25.202 M/sec (39.76%) > 6,049 cpu_atom/branch-misses/u # 7.32% of all branches (60.24%) > 1,685 cpu_core/branch-misses/u # 8.46% of all branches (39.76%) > > 1.001365193 seconds time elapsed > > 0.000000000 seconds user > 0.001284000 seconds sys > > > ⬢ [acme@...lbox perf-tools-next]$ grep -m1 'model name' /proc/cpuinfo > model name : Intel(R) Core(TM) i7-14700K > ⬢ [acme@...lbox perf-tools-next]$ > > And on the notebook, that is also hybrid: > > acme@x1:~$ grep -m1 'model name' /proc/cpuinfo > model name : 13th Gen Intel(R) Core(TM) i7-1365U > > ⬢ [acme@...lbox perf-tools-next]$ perf stat sleep 0.1 > event syntax error: '..ED.L3_HIT/,cpu/DTLB_LOAD_MISSES.STLB_HIT,cmask=0x1,metric-id=cpu!3DTLB_LOAD_MISSES.STLB_HIT!0cmask!20x1!3/,topdown-mem-bound/metric-id=topdown!1mem!1bound/,cpu/UOPS_..' > \___ Bad event or PMU > > Unable to find PMU or event on a PMU of 'cpu' > > Performance counter stats for 'sleep 0.1': > > 0.94 msec task-clock:u # 0.009 CPUs utilized > 0 context-switches:u # 0.000 /sec > 0 cpu-migrations:u # 0.000 /sec > 66 page-faults:u # 70.334 K/sec > 308,869 cpu_atom/instructions/u # 0.50 insn per cycle (51.07%) > 209,685 cpu_core/instructions/u # 0.75 insn per cycle (48.93%) > 616,738 cpu_atom/cycles/u # 0.657 GHz (51.07%) > 281,210 cpu_core/cycles/u # 0.300 GHz (48.93%) > 63,094 cpu_atom/branches/u # 67.237 M/sec (51.07%) > 48,598 cpu_core/branches/u # 51.789 M/sec (48.93%) > 5,086 cpu_atom/branch-misses/u # 8.06% of all branches (51.07%) > 2,653 cpu_core/branch-misses/u # 5.46% of all branches (48.93%) > > 0.102117022 seconds time elapsed > > 0.001682000 seconds user > 0.000000000 seconds sys > > ⬢ [acme@...lbox perf-tools-next]$ Thanks Arnaldo, it looks like this problem exists on all the hybrid metrics which are sourced from the following (look for cpu@ that should be cpu_core@): https://github.com/intel/perfmon/blob/main/ADL/metrics/perf/alderlake_metrics_goldencove_core_perf.json https://github.com/intel/perfmon/blob/main/ARL/metrics/perf/arrowlake_metrics_lioncove_core_perf.json https://github.com/intel/perfmon/blob/main/LNL/metrics/perf/lunarlake_metrics_lioncove_core_perf.json https://github.com/intel/perfmon/blob/main/MTL/metrics/perf/meteorlake_metrics_redwoodcove_core_perf.json I'll need Caleb to do a new version of these metrics before I can send out the updated version. Thanks, Ian
Powered by blists - more mailing lists