kernel/linux.git/tools/perf/util/stat-shadow.c, branch v6.12.80

perf stat: Fix find_stat for mixed legacy/non-legacy events

2025-04-10T12:39:21+00:00

[ Upstream commit 8ce0d2da14d3fb62844dd0e95982c194326b1a5f ] Legacy events typically don't have a PMU when added leading to mismatched legacy/non-legacy cases in find_stat. Use evsel__find_pmu to make sure the evsel PMU is looked up. Update the evsel__find_pmu code to look for the PMU using the extended config type or, for legacy hardware/hw_cache events on non-hybrid systems, just use the core PMU. Before: ``` $ perf stat -e cycles,cpu/instructions/ -a sleep 1 Performance counter stats for 'system wide': 215,309,764 cycles 44,326,491 cpu/instructions/ 1.002555314 seconds time elapsed ``` After: ``` $ perf stat -e cycles,cpu/instructions/ -a sleep 1 Performance counter stats for 'system wide': 990,676,332 cycles 1,235,762,487 cpu/instructions/ # 1.25 insn per cycle 1.002667198 seconds time elapsed ``` Fixes: 3612ca8e2935 ("perf stat: Fix the hard-coded metrics calculation on the hybrid") Signed-off-by: Ian Rogers Tested-by: James Clark Tested-by: Leo Yan Tested-by: Atish Patra Link: https://lore.kernel.org/r/20250109222109.567031-3-irogers@google.com Signed-off-by: Namhyung Kim Signed-off-by: Sasha Levin

perf evsel: Add accessor for tool_event

2024-09-11T14:28:27+00:00

Currently tool events use a dedicated variable within the evsel. Later changes will move this to the unused struct perf_event_attr config for these events. Add an accessor to allow the later change to be well typed and avoid changing all uses. Signed-off-by: Ian Rogers Link: https://lore.kernel.org/r/20240907050830.6752-4-irogers@google.com Cc: Ravi Bangoria Cc: Sandipan Das Cc: Mark Rutland Cc: Yang Jihong Cc: Dominique Martinet Cc: Clément Le Goffic Cc: Colin Ian King Cc: Howard Chu Cc: Ze Gao Cc: Yicong Yang Cc: Changbin Du Cc: Junhao He Cc: Peter Zijlstra Cc: Adrian Hunter Cc: Weilin Wang Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Namhyung Kim Cc: Will Deacon Cc: James Clark Cc: Mike Leach Cc: Jing Zhang Cc: Leo Yan Cc: Oliver Upton Cc: Benjamin Gray Cc: Andi Kleen Cc: Alexander Shishkin Cc: Kan Liang Cc: Athira Jajeev Cc: linux-arm-kernel@lists.infradead.org Cc: Sun Haiyong Cc: Tiezhu Yang Cc: Xu Yang Cc: John Garry Cc: Ingo Molnar Cc: Veronika Molnarova Cc: Dr. David Alan Gilbert Cc: linux-kernel@vger.kernel.org Cc: linux-perf-users@vger.kernel.org Signed-off-by: Arnaldo Carvalho de Melo

perf stat: Fix the hard-coded metrics calculation on the hybrid

2024-06-21T16:43:09+00:00

The hard-coded metrics is wrongly calculated on the hybrid machine. $ perf stat -e cycles,instructions -a sleep 1 Performance counter stats for 'system wide': 18,205,487 cpu_atom/cycles/ 9,733,603 cpu_core/cycles/ 9,423,111 cpu_atom/instructions/ # 0.52 insn per cycle 4,268,965 cpu_core/instructions/ # 0.23 insn per cycle The insn per cycle for cpu_core should be 4,268,965 / 9,733,603 = 0.44. When finding the metric events, the find_stat() doesn't take the PMU type into account. The cpu_atom/cycles/ is wrongly used to calculate the IPC of the cpu_core. In the hard-coded metrics, the events from a different PMU are only SW_CPU_CLOCK and SW_TASK_CLOCK. They both have the stat type, STAT_NSECS. Except the SW CLOCK events, check the PMU type as well. Fixes: 0a57b910807a ("perf stat: Use counts rather than saved_value") Reported-by: Khalil, Amiri Reviewed-by: Ian Rogers Signed-off-by: Kan Liang Acked-by: Namhyung Kim Cc: stable@vger.kernel.org Signed-off-by: Namhyung Kim Link: https://lore.kernel.org/r/20240606180316.4122904-1-kan.liang@linux.intel.com

perf metrics: Compute unmerged uncore metrics individually

2024-02-22T16:57:09+00:00

When merging counts from multiple uncore PMUs the metric is only computed for the metric leader. When merging/aggregation is disabled, prior to this patch just the leader's metric would be computed. Fix this by computing the metric for each PMU. On a SkylakeX: Before: ``` $ perf stat -A -M memory_bandwidth_total -a sleep 1 Performance counter stats for 'system wide': CPU0 82,217 UNC_M_CAS_COUNT.RD [uncore_imc_0] # 9.2 MB/s memory_bandwidth_total CPU18 0 UNC_M_CAS_COUNT.RD [uncore_imc_0] # 0.0 MB/s memory_bandwidth_total CPU0 61,395 UNC_M_CAS_COUNT.WR [uncore_imc_0] CPU18 0 UNC_M_CAS_COUNT.WR [uncore_imc_0] CPU0 0 UNC_M_CAS_COUNT.RD [uncore_imc_1] CPU18 0 UNC_M_CAS_COUNT.RD [uncore_imc_1] CPU0 0 UNC_M_CAS_COUNT.WR [uncore_imc_1] CPU18 0 UNC_M_CAS_COUNT.WR [uncore_imc_1] CPU0 81,570 UNC_M_CAS_COUNT.RD [uncore_imc_2] CPU18 113,886 UNC_M_CAS_COUNT.RD [uncore_imc_2] CPU0 62,330 UNC_M_CAS_COUNT.WR [uncore_imc_2] CPU18 66,942 UNC_M_CAS_COUNT.WR [uncore_imc_2] CPU0 75,489 UNC_M_CAS_COUNT.RD [uncore_imc_3] CPU18 27,958 UNC_M_CAS_COUNT.RD [uncore_imc_3] CPU0 55,864 UNC_M_CAS_COUNT.WR [uncore_imc_3] CPU18 38,727 UNC_M_CAS_COUNT.WR [uncore_imc_3] CPU0 0 UNC_M_CAS_COUNT.RD [uncore_imc_4] CPU18 0 UNC_M_CAS_COUNT.RD [uncore_imc_4] CPU0 0 UNC_M_CAS_COUNT.WR [uncore_imc_4] CPU18 0 UNC_M_CAS_COUNT.WR [uncore_imc_4] CPU0 75,423 UNC_M_CAS_COUNT.RD [uncore_imc_5] CPU18 104,527 UNC_M_CAS_COUNT.RD [uncore_imc_5] CPU0 57,596 UNC_M_CAS_COUNT.WR [uncore_imc_5] CPU18 56,777 UNC_M_CAS_COUNT.WR [uncore_imc_5] CPU0 1,003,440,851 ns duration_time 1.003440851 seconds time elapsed ``` After: ``` $ perf stat -A -M memory_bandwidth_total -a sleep 1 Performance counter stats for 'system wide': CPU0 88,968 UNC_M_CAS_COUNT.RD [uncore_imc_0] # 9.5 MB/s memory_bandwidth_total CPU18 0 UNC_M_CAS_COUNT.RD [uncore_imc_0] # 0.0 MB/s memory_bandwidth_total CPU0 59,498 UNC_M_CAS_COUNT.WR [uncore_imc_0] CPU18 0 UNC_M_CAS_COUNT.WR [uncore_imc_0] CPU0 0 UNC_M_CAS_COUNT.RD [uncore_imc_1] # 0.0 MB/s memory_bandwidth_total CPU18 0 UNC_M_CAS_COUNT.RD [uncore_imc_1] # 0.0 MB/s memory_bandwidth_total CPU0 0 UNC_M_CAS_COUNT.WR [uncore_imc_1] CPU18 0 UNC_M_CAS_COUNT.WR [uncore_imc_1] CPU0 88,635 UNC_M_CAS_COUNT.RD [uncore_imc_2] # 9.5 MB/s memory_bandwidth_total CPU18 117,975 UNC_M_CAS_COUNT.RD [uncore_imc_2] # 11.5 MB/s memory_bandwidth_total CPU0 60,829 UNC_M_CAS_COUNT.WR [uncore_imc_2] CPU18 62,105 UNC_M_CAS_COUNT.WR [uncore_imc_2] CPU0 82,238 UNC_M_CAS_COUNT.RD [uncore_imc_3] # 8.7 MB/s memory_bandwidth_total CPU18 22,906 UNC_M_CAS_COUNT.RD [uncore_imc_3] # 3.6 MB/s memory_bandwidth_total CPU0 53,959 UNC_M_CAS_COUNT.WR [uncore_imc_3] CPU18 32,990 UNC_M_CAS_COUNT.WR [uncore_imc_3] CPU0 0 UNC_M_CAS_COUNT.RD [uncore_imc_4] # 0.0 MB/s memory_bandwidth_total CPU18 0 UNC_M_CAS_COUNT.RD [uncore_imc_4] # 0.0 MB/s memory_bandwidth_total CPU0 0 UNC_M_CAS_COUNT.WR [uncore_imc_4] CPU18 0 UNC_M_CAS_COUNT.WR [uncore_imc_4] CPU0 83,595 UNC_M_CAS_COUNT.RD [uncore_imc_5] # 8.9 MB/s memory_bandwidth_total CPU18 110,151 UNC_M_CAS_COUNT.RD [uncore_imc_5] # 10.5 MB/s memory_bandwidth_total CPU0 56,540 UNC_M_CAS_COUNT.WR [uncore_imc_5] CPU18 53,816 UNC_M_CAS_COUNT.WR [uncore_imc_5] CPU0 1,003,353,416 ns duration_time ``` Signed-off-by: Ian Rogers | Acked-by: Namhyung Kim Cc: K Prateek Nayak Cc: Stephane Eranian Cc: Kaige Ye Cc: Kajol Jain Cc: Kan Liang Cc: John Garry Signed-off-by: Namhyung Kim Link: https://lore.kernel.org/r/20240221070754.4163916-2-irogers@google.com

perf stat: Pass fewer metric arguments

2024-02-22T16:56:45+00:00

Pass metric_expr and evsel rather than specific variables from the struct, thereby reducing the number of arguments. This will enable later fixes. To reduce the size of the diff, local variables are added to match the previous parameter names. This isn't done in the case of "name" as evsel->name is more intention revealing. A whitespace issue is also addressed. Signed-off-by: Ian Rogers Acked-by: Namhyung Kim Cc: K Prateek Nayak Cc: Stephane Eranian Cc: Kaige Ye Cc: Kajol Jain Cc: Kan Liang Cc: John Garry Signed-off-by: Namhyung Kim Link: https://lore.kernel.org/r/20240221070754.4163916-1-irogers@google.com

perf metric: Don't remove scale from counts

2024-02-13T21:48:09+00:00

Counts were switched from the scaled saved value form to the aggregated count to avoid double accounting. When this happened the removing of scaling for a count should have been removed, however, it wasn't and this wasn't observed as it normally doesn't matter because a counter's scale is 1. A problem was observed with RAPL events that are scaled. Fixes: 37cc8ad77cf8 ("perf metric: Directly use counts rather than saved_value") Signed-off-by: Ian Rogers Reviewed-by: Kan Liang Cc: K Prateek Nayak Cc: James Clark Cc: Kaige Ye Cc: John Garry Signed-off-by: Namhyung Kim Link: https://lore.kernel.org/r/20240209204947.3873294-5-irogers@google.com

perf stat: Fix hard coded LL miss units

2024-01-03T20:55:01+00:00

Copy-paste error where LL cache misses are reported as l1i. Fixes: 0a57b910807ad163 ("perf stat: Use counts rather than saved_value") Suggested-by: Guillaume Endignoux Reviewed-by: Kan Liang Signed-off-by: Ian Rogers Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Ian Rogers Cc: Ingo Molnar Cc: Jiri Olsa Cc: John Garry Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Link: https://lore.kernel.org/r/20231211181242.1721059-1-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo

perf stat: New metricgroup output for the default mode

2023-06-16T12:57:19+00:00

In the default mode, the current output of the metricgroup include both events and metrics, which is not necessary and just makes the output hard to read. Since different ARCHs (even different generations in the same ARCH) may use different events. The output also vary on different platforms. For a metricgroup, only outputting the value of each metric is good enough. Add a new field default_metricgroup in evsel to indicate an event of the default metricgroup. For those events, printout() should print the metricgroup name rather than each event. Add perf_stat__skip_metric_event() to skip the evsel in the Default metricgroup, if it's not running or not the metric event. Add print_metricgroup_header_t to pass the functions which print the display name of each metricgroup in the Default metricgroup. Support all three output methods. Factor out perf_stat__print_shadow_stats_metricgroup() to print out each metrics. On SPR: Before: ./perf_old stat sleep 1 Performance counter stats for 'sleep 1': 0.54 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 # 125.445 K/sec 540,970 cycles:u # 0.998 GHz 556,325 instructions:u # 1.03 insn per cycle 123,602 branches:u # 228.018 M/sec 6,889 branch-misses:u # 5.57% of all branches 3,245,820 TOPDOWN.SLOTS:u # 18.4 % tma_backend_bound # 17.2 % tma_retiring # 23.1 % tma_bad_speculation # 41.4 % tma_frontend_bound 564,859 topdown-retiring:u 1,370,999 topdown-fe-bound:u 603,271 topdown-be-bound:u 744,874 topdown-bad-spec:u 12,661 INT_MISC.UOP_DROPPING:u # 23.357 M/sec 1.001798215 seconds time elapsed 0.000193000 seconds user 0.001700000 seconds sys After: $ ./perf stat sleep 1 Performance counter stats for 'sleep 1': 0.51 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 # 132.683 K/sec 545,228 cycles:u # 1.064 GHz 555,509 instructions:u # 1.02 insn per cycle 123,574 branches:u # 241.120 M/sec 6,957 branch-misses:u # 5.63% of all branches TopdownL1 # 17.5 % tma_backend_bound # 22.6 % tma_bad_speculation # 42.7 % tma_frontend_bound # 17.1 % tma_retiring TopdownL2 # 21.8 % tma_branch_mispredicts # 11.5 % tma_core_bound # 13.4 % tma_fetch_bandwidth # 29.3 % tma_fetch_latency # 2.7 % tma_heavy_operations # 14.5 % tma_light_operations # 0.8 % tma_machine_clears # 6.1 % tma_memory_bound 1.001712086 seconds time elapsed 0.000151000 seconds user 0.001618000 seconds sys Reviewed-by: Ian Rogers Signed-off-by: Kan Liang Cc: Adrian Hunter Cc: Ahmad Yasin Cc: Andi Kleen Cc: Ingo Molnar Cc: Jiri Olsa Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Stephane Eranian Link: https://lore.kernel.org/r/20230616031420.3751973-3-kan.liang@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo

perf metric: Change divide by zero and !support events behavior

2023-05-10T15:35:02+00:00

Division by zero causes expression parsing to fail and no metric to be generated. This can mean for short running benchmarks metrics are not shown. Change the behavior to make the value nan, which gets shown like: ''' $ perf stat -M TopdownL2 true Performance counter stats for 'true': 1,031,492 INST_RETIRED.ANY # nan % tma_fetch_bandwidth # nan % tma_heavy_operations # nan % tma_light_operations 29,304 CPU_CLK_UNHALTED.REF_XCLK # nan % tma_fetch_latency # nan % tma_branch_mispredicts # nan % tma_machine_clears # nan % tma_core_bound # nan % tma_memory_bound 2,658,319 IDQ_UOPS_NOT_DELIVERED.CORE 11,167 EXE_ACTIVITY.BOUND_ON_STORES 262,058 EXE_ACTIVITY.1_PORTS_UTIL BR_MISP_RETIRED.ALL_BRANCHES (0.00%) INT_MISC.RECOVERY_CYCLES_ANY (0.00%) CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE (0.00%) CPU_CLK_UNHALTED.THREAD (0.00%) UOPS_RETIRED.RETIRE_SLOTS (0.00%) CYCLE_ACTIVITY.STALLS_MEM_ANY (0.00%) UOPS_RETIRED.MACRO_FUSED (0.00%) IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE (0.00%) EXE_ACTIVITY.2_PORTS_UTIL (0.00%) CYCLE_ACTIVITY.STALLS_TOTAL (0.00%) MACHINE_CLEARS.COUNT (0.00%) UOPS_ISSUED.ANY (0.00%) 0.002864879 seconds time elapsed 0.003012000 seconds user 0.000000000 seconds sys ''' When events aren't supported a count of 0 can be confusing and make metrics look meaningful. Change these to be nan also which, with the next change, gets shown like: ''' $ perf stat true Performance counter stats for 'true': 1.25 msec task-clock:u # 0.387 CPUs utilized 0 context-switches:u # 0.000 /sec 0 cpu-migrations:u # 0.000 /sec 46 page-faults:u # 36.702 K/sec 255,942 cycles:u # 0.204 GHz (88.66%) 123,046 instructions:u # 0.48 insn per cycle 28,301 branches:u # 22.580 M/sec 2,489 branch-misses:u # 8.79% of all branches 4,719 CPU_CLK_UNHALTED.REF_XCLK:u # 3.765 M/sec # nan % tma_frontend_bound # nan % tma_retiring # nan % tma_backend_bound # nan % tma_bad_speculation 344,855 IDQ_UOPS_NOT_DELIVERED.CORE:u # 275.147 M/sec INT_MISC.RECOVERY_CYCLES_ANY:u CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE:u (0.00%) CPU_CLK_UNHALTED.THREAD:u (0.00%) UOPS_RETIRED.RETIRE_SLOTS:u (0.00%) UOPS_ISSUED.ANY:u (0.00%) 0.003238142 seconds time elapsed 0.000000000 seconds user 0.003434000 seconds sys ''' Ensure that nan metric values are quoted as nan isn't a valid number in JSON. Signed-off-by: Ian Rogers Tested-by: Kan Liang Cc: Adrian Hunter Cc: Ahmad Yasin Cc: Alexander Shishkin Cc: Andi Kleen Cc: Athira Rajeev Cc: Caleb Biggers Cc: Edward Baker Cc: Florian Fischer Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: John Garry Cc: Kajol Jain Cc: Kang Minchul Cc: Leo Yan Cc: Mark Rutland Cc: Namhyung Kim Cc: Perry Taylor Cc: Peter Zijlstra Cc: Ravi Bangoria Cc: Rob Herring Cc: Samantha Alt Cc: Stephane Eranian Cc: Sumanth Korikkar Cc: Suzuki Poulouse Cc: Thomas Richter Cc: Tiezhu Yang Cc: Weilin Wang Cc: Xing Zhengjun Cc: Yang Jihong Link: https://lore.kernel.org/r/20230502223851.2234828-2-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo

perf stat: Modify the group test

2023-03-13T18:12:31+00:00

Currently nr_members is 0 for an event with no group, however, they are always a leader of their own group. A later change will make that count 1 because the event is its own leader. Make the find_stat logic consistent with this, an improvement suggested by Namhyung Kim. Suggested-by: Namhyung Kim Signed-off-by: Ian Rogers Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Florian Fischer Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: John Garry Cc: Kajol Jain Cc: Kan Liang Cc: Kim Phillips Cc: Leo Yan Cc: Mark Rutland Cc: Peter Zijlstra Cc: Ravi Bangoria Cc: Sean Christopherson Cc: Steinar H. Gunderson Cc: Stephane Eranian Cc: Suzuki Poulouse Cc: Xing Zhengjun Link: https://lore.kernel.org/r/20230312021543.3060328-2-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo