linux/tools/perf/Documentation
Jin Yao 1af62ce61c perf stat: Show percore counts in per CPU output
We have supported the event modifier "percore" which sums up the event
counts for all hardware threads in a core and show the counts per core.

For example,

 # perf stat -e cpu/event=cpu-cycles,percore/ -a -A -- sleep 1

  Performance counter stats for 'system wide':

 S0-D0-C0                395,072      cpu/event=cpu-cycles,percore/
 S0-D0-C1                851,248      cpu/event=cpu-cycles,percore/
 S0-D0-C2                954,226      cpu/event=cpu-cycles,percore/
 S0-D0-C3              1,233,659      cpu/event=cpu-cycles,percore/

This patch provides a new option "--percore-show-thread". It is used
with event modifier "percore" together to sum up the event counts for
all hardware threads in a core but show the counts per hardware thread.

This is essentially a replacement for the any bit (which is gone in
Icelake). Per core counts are useful for some formulas, e.g. CoreIPC.
The original percore version was inconvenient to post process. This
variant matches the output of the any bit.

With this patch, for example,

 # perf stat -e cpu/event=cpu-cycles,percore/ -a -A --percore-show-thread  -- sleep 1

  Performance counter stats for 'system wide':

 CPU0               2,453,061      cpu/event=cpu-cycles,percore/
 CPU1               1,823,921      cpu/event=cpu-cycles,percore/
 CPU2               1,383,166      cpu/event=cpu-cycles,percore/
 CPU3               1,102,652      cpu/event=cpu-cycles,percore/
 CPU4               2,453,061      cpu/event=cpu-cycles,percore/
 CPU5               1,823,921      cpu/event=cpu-cycles,percore/
 CPU6               1,383,166      cpu/event=cpu-cycles,percore/
 CPU7               1,102,652      cpu/event=cpu-cycles,percore/

We can see counts are duplicated in CPU pairs (CPU0/CPU4, CPU1/CPU5,
CPU2/CPU6, CPU3/CPU7).

The interval mode also works. For example,

 # perf stat -e cpu/event=cpu-cycles,percore/ -a -A --percore-show-thread  -I 1000
 #           time CPU                    counts unit events
      1.000425421 CPU0                 925,032      cpu/event=cpu-cycles,percore/
      1.000425421 CPU1                 430,202      cpu/event=cpu-cycles,percore/
      1.000425421 CPU2                 436,843      cpu/event=cpu-cycles,percore/
      1.000425421 CPU3               1,192,504      cpu/event=cpu-cycles,percore/
      1.000425421 CPU4                 925,032      cpu/event=cpu-cycles,percore/
      1.000425421 CPU5                 430,202      cpu/event=cpu-cycles,percore/
      1.000425421 CPU6                 436,843      cpu/event=cpu-cycles,percore/
      1.000425421 CPU7               1,192,504      cpu/event=cpu-cycles,percore/

If we offline CPU5, the result is:

 # perf stat -e cpu/event=cpu-cycles,percore/ -a -A --percore-show-thread -- sleep 1

  Performance counter stats for 'system wide':

 CPU0               2,752,148      cpu/event=cpu-cycles,percore/
 CPU1               1,009,312      cpu/event=cpu-cycles,percore/
 CPU2               2,784,072      cpu/event=cpu-cycles,percore/
 CPU3               2,427,922      cpu/event=cpu-cycles,percore/
 CPU4               2,752,148      cpu/event=cpu-cycles,percore/
 CPU6               2,784,072      cpu/event=cpu-cycles,percore/
 CPU7               2,427,922      cpu/event=cpu-cycles,percore/

        1.001416041 seconds time elapsed

 v4:
 ---
 Ravi Bangoria reports an issue in v3. Once we offline a CPU,
 the output is not correct. The issue is we should use the cpu
 idx in print_percore_thread rather than using the cpu value.

 v3:
 ---
 1. Fix the interval mode output error
 2. Use cpu value (not cpu index) in config->aggr_get_id().
 3. Refine the code according to Jiri's comments.

 v2:
 ---
 Add the explanation in change log. This is essentially a replacement
 for the any bit. No code change.

Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Tested-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20200214080452.26402-1-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2020-03-04 10:34:09 -03:00
..
android.txt perf tools: Update android build documentation 2016-07-04 20:27:27 -03:00
asciidoc.conf perf docs: Allow man page date to be specified 2019-09-27 09:26:14 -03:00
asciidoctor-extensions.rb perf Documentation: Support for asciidoctor 2018-04-26 13:47:10 -03:00
build-xed.txt perf script: Add --insn-trace for instruction decoding 2018-10-24 15:29:50 -03:00
Build.txt perf tools: Add doc about how to build perf with Asan and UBSan 2019-03-19 16:52:04 -03:00
callchain-overhead-calculation.txt perf tools: Document --children option in more detail 2015-04-29 10:38:06 -03:00
db-export.txt perf db-export: Add brief documentation 2019-06-05 09:47:57 -03:00
examples.txt perf record: Remove -f/--force option 2013-07-08 17:37:25 -03:00
intel-bts.txt perf tools: Add Intel BTS support 2015-08-21 11:34:10 -03:00
intel-pt.txt perf intel-pt: Add support for recording AUX area samples 2019-11-22 10:48:13 -03:00
itrace.txt perf tools: Add itrace option 'o' to synthesize aux-output events 2019-08-14 10:59:59 -03:00
jit-interface.txt perf symbols: Add description of JIT interface 2012-08-13 14:55:02 -03:00
jitdump-specification.txt perf docs: Correct and clarify jitdump spec 2019-09-30 17:29:51 -03:00
Makefile perf tools: Fix a typo in a variable name in the Documentation Makefile 2019-08-08 15:41:10 -03:00
manpage-1.72.xsl
manpage-base.xsl
manpage-bold-literal.xsl
manpage-normal.xsl
manpage-suppress-sp.xsl
perf-annotate.txt perf tools: Support --prefix/--prefix-strip 2020-01-14 12:02:19 -03:00
perf-archive.txt perf archive: Remove duplicated 'runs' in man page 2013-12-09 15:21:45 -03:00
perf-bench.txt perf bench: Add epoll_ctl(2) benchmark 2018-11-21 22:39:55 -03:00
perf-buildid-cache.txt perf buildid-cache: Support --purge-all option 2018-04-26 09:30:26 -03:00
perf-buildid-list.txt perf report: Accept fifos as input file 2011-12-23 17:01:03 -02:00
perf-c2c.txt perf mem/c2c: Fix perf_mem_events to support powerpc 2019-02-04 11:32:14 -03:00
perf-config.txt perf config: Document missing config options 2020-02-27 10:45:19 -03:00
perf-data.txt perf tools: Correct title markers for asciidoctor 2018-03-07 10:26:32 -03:00
perf-diff.txt perf diff: Report noisy for cycles diff 2019-10-11 10:57:00 -03:00
perf-evlist.txt perf evlist: Document missing --force option 2017-11-16 14:50:07 -03:00
perf-ftrace.txt perf tools: Correct title markers for asciidoctor 2018-03-07 10:26:32 -03:00
perf-help.txt
perf-inject.txt perf inject: Document missing options 2017-11-16 14:50:05 -03:00
perf-kallsyms.txt perf tools: Correct title markers for asciidoctor 2018-03-07 10:26:32 -03:00
perf-kmem.txt perf kmem: Document a missing option & an argument 2018-02-16 14:55:42 -03:00
perf-kvm.txt perf kvm: Clarify the 'perf kvm' -i and -o command line options 2019-12-02 15:38:59 -03:00
perf-list.txt perf list: Hide deprecated events by default 2019-10-19 15:35:01 -03:00
perf-lock.txt perf lock: Document missing options 2017-11-16 14:50:04 -03:00
perf-mem.txt perf mem/c2c: Fix perf_mem_events to support powerpc 2019-02-04 11:32:14 -03:00
perf-probe.txt perf-probe: Add user memory access attribute support 2019-05-25 23:04:42 -04:00
perf-record.txt perf record: Add aux-sample-size config term 2019-11-22 10:48:13 -03:00
perf-report.txt perf tools: Support --prefix/--prefix-strip 2020-01-14 12:02:19 -03:00
perf-sched.txt perf sched timehist: Add support for filtering on CPU 2020-01-06 11:46:09 -03:00
perf-script-perl.txt perf tools: Correct title markers for asciidoctor 2018-03-07 10:26:32 -03:00
perf-script-python.txt perf script python: Add dict fields introduction to Documentation 2018-06-06 15:40:10 -03:00
perf-script.txt perf script: Allow specifying event to switch off processing of other events 2019-08-15 12:24:16 -03:00
perf-stat.txt perf stat: Show percore counts in per CPU output 2020-03-04 10:34:09 -03:00
perf-test.txt perf test: Add -F/--dont-fork option 2016-06-30 18:27:45 -03:00
perf-timechart.txt perf timechart: Document missing --force option 2017-11-16 14:50:06 -03:00
perf-top.txt perf tools: Support --prefix/--prefix-strip 2020-01-14 12:02:19 -03:00
perf-trace.txt perf trace: Introduce --errno-summary 2019-10-15 13:03:49 -03:00
perf-version.txt perf version: Add man page 2018-04-02 13:52:23 -03:00
perf.data-directory-format.txt perf record: Put a copy of kcore into the perf.data directory 2019-11-06 15:43:05 -03:00
perf.data-file-format.txt perf.data documentation: Clarify HEADER_SAMPLE_TOPOLOGY format 2019-08-14 10:59:59 -03:00
perf.txt perf tool: Provide an option to print perf_event_open args and return value 2019-11-12 08:32:27 -03:00
perfconfig.example perf config: Show default report configuration in example and docs 2016-09-01 09:44:13 -03:00
tips.txt perf tools: Fix typos / broken sentences 2019-07-02 16:08:16 -03:00