Namhyung Kim 3599705789 perf stat: Do not delay the workload with --delay ... [ Upstream commit c587e77e10 ] The -D/--delay option is to delay the measure after the program starts. But the current code goes to sleep before starting the program so the program is delayed too. This is not the intention, let's fix it. Before: $ time sudo ./perf stat -a -e cycles -D 3000 sleep 4 Events disabled Events enabled Performance counter stats for 'system wide': 4,326,949,337 cycles 4.007494118 seconds time elapsed real 0m7.474s user 0m0.356s sys 0m0.120s It ran the workload for 4 seconds and gave the 3 second delay. So it should skip the first 3 second and measure the last 1 second only. But as you can see, it delays 3 seconds and ran the workload after that for 4 seconds. So the total time (real) was 7 seconds. After: $ time sudo ./perf stat -a -e cycles -D 3000 sleep 4 Events disabled Events enabled Performance counter stats for 'system wide': 1,063,551,013 cycles 1.002769510 seconds time elapsed real 0m4.484s user 0m0.385s sys 0m0.086s The bug was introduced when it changed enablement of system-wide events with a command line workload. But it should've considered the initial delay case. The code was reworked since then (in bb8bc52e75) so I'm afraid it won't be applied cleanly. Fixes: d0a0a51149 ("perf stat: Fix forked applications enablement of counters") Reported-by: Kevin Nomura <nomurak@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org> Tested-by: Thomas Richter <tmricht@linux.ibm.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Sumanth Korikkar <sumanthk@linux.ibm.com> Link: https://lore.kernel.org/r/20221212230820.901382-1-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Sasha Levin <sashal@kernel.org>		2022-12-31 13:26:31 +01:00
..
arch	perf intel-pt: Fix system_wide dummy event for hybrid	2022-10-21 12:39:26 +02:00
bench
dlfilters	perf dlfilter dlfilter-show-cycles: Fix types for print format	2022-09-08 12:17:45 -03:00
Documentation
examples/bpf
include
jvmti
pmu-events	perf vendor events arm64: Fix incorrect Hisi hip08 L3 metrics	2022-11-04 00:00:32 +09:00
python
scripts
tests	perf test: Fix skipping branch stack sampling test	2022-11-16 10:03:59 +01:00
trace
ui
util	perf off_cpu: Fix a typo in BTF tracepoint name, it should be 'btf_trace_sched_switch'	2022-12-31 13:26:31 +01:00
.gitignore	perf tools: Add the include/perf/ directory to .gitignore	2022-11-16 10:03:59 +01:00
Build
builtin-annotate.c
builtin-bench.c
builtin-buildid-cache.c
builtin-buildid-list.c
builtin-c2c.c	perf c2c: Prevent potential memory leak in c2c_he_zalloc()	2022-09-06 09:45:23 -03:00
builtin-config.c
builtin-daemon.c
builtin-data.c
builtin-diff.c
builtin-evlist.c
builtin-ftrace.c
builtin-help.c
builtin-inject.c
builtin-kallsyms.c
builtin-kmem.c
builtin-kvm.c
builtin-kwork.c
builtin-list.c
builtin-lock.c	perf lock: Remove redundant word 'contention' in help message	2022-09-08 15:23:42 -03:00
builtin-mem.c
builtin-probe.c
builtin-record.c	perf record: Fix cpu mask bit setting for mixed mmaps	2022-09-21 16:08:00 -03:00
builtin-report.c
builtin-sched.c
builtin-script.c	perf script: Fix Cannot print 'iregs' field for hybrid systems	2022-09-08 15:27:39 -03:00
builtin-stat.c	perf stat: Do not delay the workload with --delay	2022-12-31 13:26:31 +01:00
builtin-timechart.c
builtin-top.c
builtin-trace.c	perf trace: Handle failure when trace point folder is missed	2022-12-31 13:26:30 +01:00
builtin-version.c
builtin.h
check-headers.sh
command-list.txt
CREDITS
design.txt
Makefile
Makefile.config
Makefile.perf	perf tools: Don't install data files with x permissions	2022-09-08 15:55:56 -03:00
MANIFEST
perf-archive.sh
perf-completion.sh
perf-iostat.sh
perf-read-vdso.c
perf-sys.h
perf.c
perf.h