[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230926042938.509234-1-changbin.du@huawei.com>
Date: Tue, 26 Sep 2023 12:29:33 +0800
From: Changbin Du <changbin.du@...wei.com>
To: Peter Zijlstra <peterz@...radead.org>,
Ingo Molnar <mingo@...hat.com>,
Arnaldo Carvalho de Melo <acme@...nel.org>
CC: Mark Rutland <mark.rutland@....com>,
Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
Jiri Olsa <jolsa@...nel.org>,
Namhyung Kim <namhyung@...nel.org>,
Ian Rogers <irogers@...gle.com>,
Adrian Hunter <adrian.hunter@...el.com>,
<linux-perf-users@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
<changbin.du@...il.com>, Changbin Du <changbin.du@...wei.com>
Subject: [PATCH v5 0/5] perf: Add new option '--workload-config' to set workload sched_policy/prio/cpumask
[I still think this is a handy option, so do some improvment and resend it.]
To get consistent benchmarking results, sometimes we need to set the
sched_policy/priority/cpumask of the workload to reduce system noise.
For example, CPU binding is required on big.little system.
$ perf stat -- taskset -c 0 ls
However, the events of 'taskset' itself are also counted here. To get more
accurate result, this should be avoided.
To get away of the middleman, this adds a new option '--workload-config' to
do the same jobs for stat and record commands.
--workload-config <[sched_policy=policy][,sched_prio=priority][,cpu-list=list]>
setup target workload (the <command>) attributes:
sched_policy: other|fifo|rr|batch|idle
sched_prio: scheduling priority for fifo|rr, nice value for other
cpu-list: CPU affinity. e.g. 1-3:5 is processors #1, #2, #3 and #5
For example,
$ sudo perf stat --workload-config sched_policy=fifo,sched_prio=40,cpu-list=0-3:7 -- ls
Above command will make 'ls' run on CPU #0-#3 and #7 with fifo scheduler and
realtime priority is 40.
v5:
o rename '--workload-attr' as '--workload-config' (sounds better) .
o transform to key-value pair style option.
v4:
- add a test case for perf-stat. (suggeted by Ian Rogers)
- fix warning found by 0-DAY.
v3:
- replace taskset with --workload-attr option in documents and tests.
v2:
- Use cpu list spec instead of cpu mask number.
- Update documents.
Changbin Du (5):
perf cpumap: Add __perf_cpu_map__new and perf_cpu_map__2_cpuset
perf: util: support string type option for perf_parse_sublevel_options
perf: add new option '--workload-config' to set workload
sched_policy/prio/cpumask
perf: replace taskset with --workload-config option
perf test: add test case for --workload-config option
tools/lib/perf/cpumap.c | 45 +++++++-
tools/lib/perf/include/perf/cpumap.h | 4 +
tools/lib/perf/libperf.map | 2 +
tools/perf/Documentation/intel-hybrid.txt | 2 +-
tools/perf/Documentation/perf-record.txt | 7 ++
tools/perf/Documentation/perf-stat.txt | 8 +-
tools/perf/builtin-record.c | 27 +++++
tools/perf/builtin-stat.c | 19 +++
tools/perf/tests/cpumap.c | 23 ++++
tools/perf/tests/shell/stat.sh | 19 +++
.../tests/shell/stat_bpf_counters_cgrp.sh | 2 +-
tools/perf/tests/shell/test_arm_coresight.sh | 2 +-
tools/perf/tests/shell/test_data_symbol.sh | 2 +-
tools/perf/tests/shell/test_intel_pt.sh | 2 +-
tools/perf/util/evlist.c | 108 ++++++++++++++++++
tools/perf/util/evlist.h | 3 +
tools/perf/util/parse-sublevel-options.c | 12 +-
tools/perf/util/parse-sublevel-options.h | 7 ++
tools/perf/util/target.h | 9 ++
19 files changed, 291 insertions(+), 12 deletions(-)
--
2.25.1
Powered by blists - more mailing lists