kernel/tools/perf/util/stat-display.c, branch linux-5.14.y

perf stat: Merge uncore events by default for hybrid platform

2021-07-14T13:05:35Z

On a hybrid platform, by default 'perf stat' aggregates and reports the event counts per PMU. For example, # perf stat -e cycles -a true Performance counter stats for 'system wide': 1,400,445 cpu_core/cycles/ 680,881 cpu_atom/cycles/ 0.001770773 seconds time elapsed But for uncore events that's not a suitable method. Uncore has nothing to do with hybrid. So for uncore events, we aggregate event counts from all PMUs and report the counts without PMUs. Before: # perf stat -e arb/event=0x81,umask=0x1/,arb/event=0x84,umask=0x1/ -a true Performance counter stats for 'system wide': 2,058 uncore_arb_0/event=0x81,umask=0x1/ 2,028 uncore_arb_1/event=0x81,umask=0x1/ 0 uncore_arb_0/event=0x84,umask=0x1/ 0 uncore_arb_1/event=0x84,umask=0x1/ 0.000614498 seconds time elapsed After: # perf stat -e arb/event=0x81,umask=0x1/,arb/event=0x84,umask=0x1/ -a true Performance counter stats for 'system wide': 3,996 arb/event=0x81,umask=0x1/ 0 arb/event=0x84,umask=0x1/ 0.000630046 seconds time elapsed Of course, we also keep the '--no-merge' working for uncore events. # perf stat -e arb/event=0x81,umask=0x1/,arb/event=0x84,umask=0x1/ --no-merge true Performance counter stats for 'system wide': 1,952 uncore_arb_0/event=0x81,umask=0x1/ 1,921 uncore_arb_1/event=0x81,umask=0x1/ 0 uncore_arb_0/event=0x84,umask=0x1/ 0 uncore_arb_1/event=0x84,umask=0x1/ 0.000575536 seconds time elapsed Signed-off-by: Jin Yao Acked-by: Jiri Olsa Cc: Alexander Shishkin Cc: Andi Kleen Cc: Ingo Molnar Cc: Kan Liang Cc: Peter Zijlstra Link: https://lore.kernel.org/r/20210707055652.962-1-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo

perf stat: Disable the NMI watchdog message on hybrid

2021-07-07T14:37:23Z

If we run a single workload that only runs on big core, there is always a ugly message about disabling the NMI watchdog because the atom is not counted. Before: # ./perf stat true Performance counter stats for 'true': 0.43 msec task-clock # 0.396 CPUs utilized 0 context-switches # 0.000 /sec 0 cpu-migrations # 0.000 /sec 45 page-faults # 103.918 K/sec 639,634 cpu_core/cycles/ # 1.477 G/sec cpu_atom/cycles/ (0.00%) 643,498 cpu_core/instructions/ # 1.486 G/sec cpu_atom/instructions/ (0.00%) 123,715 cpu_core/branches/ # 285.694 M/sec cpu_atom/branches/ (0.00%) 4,094 cpu_core/branch-misses/ # 9.454 M/sec cpu_atom/branch-misses/ (0.00%) 0.001092407 seconds time elapsed 0.001144000 seconds user 0.000000000 seconds sys Some events weren't counted. Try disabling the NMI watchdog: echo 0 > /proc/sys/kernel/nmi_watchdog perf stat ... echo 1 > /proc/sys/kernel/nmi_watchdog # ./perf stat -e '{cpu_atom/cycles/,msr/tsc/}' true Performance counter stats for 'true': cpu_atom/cycles/ (0.00%) msr/tsc/ (0.00%) 0.001904106 seconds time elapsed 0.001947000 seconds user 0.000000000 seconds sys Some events weren't counted. Try disabling the NMI watchdog: echo 0 > /proc/sys/kernel/nmi_watchdog perf stat ... echo 1 > /proc/sys/kernel/nmi_watchdog The events in group usually have to be from the same PMU. Try reorganizing the group. Now we disable the NMI watchdog message on hybrid, otherwise there are too many false positives. After: # ./perf stat true Performance counter stats for 'true': 0.79 msec task-clock # 0.419 CPUs utilized 0 context-switches # 0.000 /sec 0 cpu-migrations # 0.000 /sec 48 page-faults # 60.889 K/sec 777,692 cpu_core/cycles/ # 986.519 M/sec cpu_atom/cycles/ (0.00%) 669,147 cpu_core/instructions/ # 848.828 M/sec cpu_atom/instructions/ (0.00%) 128,635 cpu_core/branches/ # 163.176 M/sec cpu_atom/branches/ (0.00%) 4,089 cpu_core/branch-misses/ # 5.187 M/sec cpu_atom/branch-misses/ (0.00%) 0.001880649 seconds time elapsed 0.001935000 seconds user 0.000000000 seconds sys # ./perf stat -e '{cpu_atom/cycles/,msr/tsc/}' true Performance counter stats for 'true': cpu_atom/cycles/ (0.00%) msr/tsc/ (0.00%) 0.000963319 seconds time elapsed 0.000999000 seconds user 0.000000000 seconds sys Signed-off-by: Jin Yao Acked-by: Jiri Olsa Cc: Alexander Shishkin Cc: Andi Kleen Cc: Jin Yao Cc: Kan Liang Cc: Peter Zijlstra Link: http://lore.kernel.org/lkml/20210610034557.29766-1-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo

Merge remote-tracking branch 'torvalds/master' into perf/core

2021-06-22T16:56:50Z

To pick up fixes, since perf/urgent is already upstream. Signed-off-by: Arnaldo Carvalho de Melo

perf stat: Honor event config name on --no-merge

2021-06-04T13:05:23Z

If user gave an event name explicitly, it should be displayed in the output as is. But with --no-merge option it adds a pmu name at the end so might confuse users. Actually this is true for hybrid pmus, I think we should do the same for others. Signed-off-by: Namhyung Kim Acked-by: Ian Rogers Cc: Alexander Shishkin Cc: Andi Kleen Cc: Jin Yao Cc: Jiri Olsa Cc: Mark Rutland Cc: Peter Zijlstra Link: http://lore.kernel.org/lkml/20210602212241.2175005-3-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo

perf stat: Use aggregated counts directly

2021-05-12T15:43:11Z

The ps->res_stats is for repeated runs, so the interval code should not touch it. Actually the aggregated counts are available in the counter->counts->aggr, so we can (and should) use it directly IMHO. No functional change intended. Signed-off-by: Namhyung Kim Acked-by: Jiri Olsa Cc: Alexander Shishkin Cc: Andi Kleen Cc: Ian Rogers Cc: Jin Yao Cc: Mark Rutland Cc: Peter Zijlstra Link: http://lore.kernel.org/lkml/20210423023833.1430520-1-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo

perf stat: Filter out unmatched aggregation for hybrid event

2021-04-29T13:30:59Z

perf-stat has supported some aggregation modes, such as --per-core, --per-socket and etc. While for hybrid event, it may only available on part of cpus. So for --per-core, we need to filter out the unavailable cores, for --per-socket, filter out the unavailable sockets, and so on. Before: # perf stat --per-core -e cpu_core/cycles/ -a -- sleep 1 Performance counter stats for 'system wide': S0-D0-C0 2 479,530 cpu_core/cycles/ S0-D0-C4 2 175,007 cpu_core/cycles/ S0-D0-C8 2 166,240 cpu_core/cycles/ S0-D0-C12 2 704,673 cpu_core/cycles/ S0-D0-C16 2 865,835 cpu_core/cycles/ S0-D0-C20 2 2,958,461 cpu_core/cycles/ S0-D0-C24 2 163,988 cpu_core/cycles/ S0-D0-C28 2 164,729 cpu_core/cycles/ S0-D0-C32 0 cpu_core/cycles/ S0-D0-C33 0 cpu_core/cycles/ S0-D0-C34 0 cpu_core/cycles/ S0-D0-C35 0 cpu_core/cycles/ S0-D0-C36 0 cpu_core/cycles/ S0-D0-C37 0 cpu_core/cycles/ S0-D0-C38 0 cpu_core/cycles/ S0-D0-C39 0 cpu_core/cycles/ 1.003597211 seconds time elapsed After: # perf stat --per-core -e cpu_core/cycles/ -a -- sleep 1 Performance counter stats for 'system wide': S0-D0-C0 2 210,428 cpu_core/cycles/ S0-D0-C4 2 444,830 cpu_core/cycles/ S0-D0-C8 2 435,241 cpu_core/cycles/ S0-D0-C12 2 423,976 cpu_core/cycles/ S0-D0-C16 2 859,350 cpu_core/cycles/ S0-D0-C20 2 1,559,589 cpu_core/cycles/ S0-D0-C24 2 163,924 cpu_core/cycles/ S0-D0-C28 2 376,610 cpu_core/cycles/ 1.003621290 seconds time elapsed Signed-off-by: Jin Yao Co-developed-by: Jiri Olsa Reviewed-by: Jiri Olsa Cc: Alexander Shishkin Cc: Andi Kleen Cc: Ingo Molnar Cc: Kan Liang Cc: Peter Zijlstra Link: https://lore.kernel.org/r/20210427070139.25256-16-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo

perf stat: Uniquify hybrid event name

2021-04-29T13:30:59Z

It would be useful to let user know the pmu which the event belongs to. perf-stat has supported '--no-merge' option and it can print the pmu name after the event name, such as: "cycles [cpu_core]" Now this option is enabled by default for hybrid platform but change the format to: "cpu_core/cycles/" If user configs the name, we still use the user specified name. Signed-off-by: Jin Yao Reviewed-by: Jiri Olsa Cc: Alexander Shishkin Cc: Andi Kleen Cc: Ingo Molnar Cc: Kan Liang Cc: Peter Zijlstra ink: https://lore.kernel.org/r/20210427070139.25256-8-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo

perf stat: Basic support for iostat in perf

2021-04-20T11:40:20Z

Add basic flow for a new iostat mode in perf. Mode is intended to provide four I/O performance metrics per each PCIe root port: Inbound Read, Inbound Write, Outbound Read, Outbound Write. The actual code to compute the metrics and attribute it to root port is in follow-on patches. Signed-off-by: Alexander Antonov Acked-by: Namhyung Kim Cc: Alexander Shishkin Cc: Alexey V Bayduraev Cc: Andi Kleen Cc: Ian Rogers Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mark Rutland Cc: Peter Zijlstra Link: https://lore.kernel.org/r/20210419094147.15909-2-alexander.antonov@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo

perf stat: Align CSV output for summary mode

2021-03-24T13:21:49Z

The 'perf stat' subcommand supports the request for a summary of the interval counter readings. But the summary lines break the CSV output so it's hard for scripts to parse the result. Before: # perf stat -x, -I1000 --interval-count 1 --summary 1.001323097,8013.48,msec,cpu-clock,8013483384,100.00,8.013,CPUs utilized 1.001323097,270,,context-switches,8013513297,100.00,0.034,K/sec 1.001323097,13,,cpu-migrations,8013530032,100.00,0.002,K/sec 1.001323097,184,,page-faults,8013546992,100.00,0.023,K/sec 1.001323097,20574191,,cycles,8013551506,100.00,0.003,GHz 1.001323097,10562267,,instructions,8013564958,100.00,0.51,insn per cycle 1.001323097,2019244,,branches,8013575673,100.00,0.252,M/sec 1.001323097,106152,,branch-misses,8013585776,100.00,5.26,of all branches 8013.48,msec,cpu-clock,8013483384,100.00,7.984,CPUs utilized 270,,context-switches,8013513297,100.00,0.034,K/sec 13,,cpu-migrations,8013530032,100.00,0.002,K/sec 184,,page-faults,8013546992,100.00,0.023,K/sec 20574191,,cycles,8013551506,100.00,0.003,GHz 10562267,,instructions,8013564958,100.00,0.51,insn per cycle 2019244,,branches,8013575673,100.00,0.252,M/sec 106152,,branch-misses,8013585776,100.00,5.26,of all branches The summary line loses the timestamp column, which breaks the CSV output. We add a column at the original 'timestamp' position and it just says 'summary' for the summary line. After: # perf stat -x, -I1000 --interval-count 1 --summary 1.001196053,8012.72,msec,cpu-clock,8012722903,100.00,8.013,CPUs utilized 1.001196053,218,,context-switches,8012753271,100.00,0.027,K/sec 1.001196053,9,,cpu-migrations,8012769767,100.00,0.001,K/sec 1.001196053,0,,page-faults,8012786257,100.00,0.000,K/sec 1.001196053,15004518,,cycles,8012790637,100.00,0.002,GHz 1.001196053,7954691,,instructions,8012804027,100.00,0.53,insn per cycle 1.001196053,1590259,,branches,8012814766,100.00,0.198,M/sec 1.001196053,82601,,branch-misses,8012824365,100.00,5.19,of all branches summary,8012.72,msec,cpu-clock,8012722903,100.00,7.986,CPUs utilized summary,218,,context-switches,8012753271,100.00,0.027,K/sec summary,9,,cpu-migrations,8012769767,100.00,0.001,K/sec summary,0,,page-faults,8012786257,100.00,0.000,K/sec summary,15004518,,cycles,8012790637,100.00,0.002,GHz summary,7954691,,instructions,8012804027,100.00,0.53,insn per cycle summary,1590259,,branches,8012814766,100.00,0.198,M/sec summary,82601,,branch-misses,8012824365,100.00,5.19,of all branches Now it's easy for script to analyse the summary lines. Of course, we also consider not to break possible existing scripts which can continue to use the broken CSV format by using a new '--no-csv-summary.' option. # perf stat -x, -I1000 --interval-count 1 --summary --no-csv-summary 1.001213261,8012.67,msec,cpu-clock,8012672327,100.00,8.013,CPUs utilized 1.001213261,197,,context-switches,8012703742,100.00,24.586,/sec 1.001213261,9,,cpu-migrations,8012720902,100.00,1.123,/sec 1.001213261,644,,page-faults,8012738266,100.00,80.373,/sec 1.001213261,18350698,,cycles,8012744109,100.00,0.002,GHz 1.001213261,12745021,,instructions,8012759001,100.00,0.69,insn per cycle 1.001213261,2458033,,branches,8012770864,100.00,306.768,K/sec 1.001213261,102107,,branch-misses,8012781751,100.00,4.15,of all branches 8012.67,msec,cpu-clock,8012672327,100.00,7.985,CPUs utilized 197,,context-switches,8012703742,100.00,24.586,/sec 9,,cpu-migrations,8012720902,100.00,1.123,/sec 644,,page-faults,8012738266,100.00,80.373,/sec 18350698,,cycles,8012744109,100.00,0.002,GHz 12745021,,instructions,8012759001,100.00,0.69,insn per cycle 2458033,,branches,8012770864,100.00,306.768,K/sec 102107,,branch-misses,8012781751,100.00,4.15,of all branches This option can be enabled in perf config by setting the variable 'stat.no-csv-summary'. # perf config stat.no-csv-summary=true # perf config -l stat.no-csv-summary=true # perf stat -x, -I1000 --interval-count 1 --summary 1.001330198,8013.28,msec,cpu-clock,8013279201,100.00,8.013,CPUs utilized 1.001330198,205,,context-switches,8013308394,100.00,25.583,/sec 1.001330198,10,,cpu-migrations,8013324681,100.00,1.248,/sec 1.001330198,0,,page-faults,8013340926,100.00,0.000,/sec 1.001330198,8027742,,cycles,8013344503,100.00,0.001,GHz 1.001330198,2871717,,instructions,8013356501,100.00,0.36,insn per cycle 1.001330198,553564,,branches,8013366204,100.00,69.081,K/sec 1.001330198,54021,,branch-misses,8013375952,100.00,9.76,of all branches 8013.28,msec,cpu-clock,8013279201,100.00,7.985,CPUs utilized 205,,context-switches,8013308394,100.00,25.583,/sec 10,,cpu-migrations,8013324681,100.00,1.248,/sec 0,,page-faults,8013340926,100.00,0.000,/sec 8027742,,cycles,8013344503,100.00,0.001,GHz 2871717,,instructions,8013356501,100.00,0.36,insn per cycle 553564,,branches,8013366204,100.00,69.081,K/sec 54021,,branch-misses,8013375952,100.00,9.76,of all branches Signed-off-by: Jin Yao Acked-by: Andi Kleen Acked-by: Jiri Olsa Tested-by: Arnaldo Carvalho de Melo Cc: Alexander Shishkin Cc: Jin Yao Cc: Kan Liang Cc: Peter Zijlstra Link: http://lore.kernel.org/lkml/20210319070156.20394-1-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Cast (struct timeval).tv_sec when printing

2021-03-06T19:54:24Z

The musl-libc [1] defines (struct timeval).tv_sec as a 'long long' for arm and other architectures. The default build having a '-Wformat' flag, not casting the field when printing prevents from building perf. This patch casts the (struct timeval).tv_sec fields to the expected format. [1] git://git.musl-libc.org/musl Signed-off-by: Pierre Gondois Acked-by: Jiri Olsa Cc: Alexander Shishkin Cc: Douglas.raillard@arm.com Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Link: http://lore.kernel.org/lkml/20210224182410.5366-1-Pierre.Gondois@arm.com Signed-off-by: Arnaldo Carvalho de Melo