Messages in this thread | | | Date | Wed, 20 Apr 2022 13:46:41 -0300 | From | Arnaldo Carvalho de Melo <> | Subject | Re: [PATCHSET v4 next 0/3] perf stat: add user_time and system_time tool events |
| |
Em Wed, Apr 20, 2022 at 12:23:51PM +0200, Florian Fischer escreveu: > This patch series adds new internal events to perf stat exposing the times spend > in user and kernel mode in nanoseconds reported by rusage. > > During some benchmarking using perf it bothered me that I could not easily > retrieve those times from perf stat when using the machine readable output. > > But perf definitely knows about those values because in the human readable output > they are present. > > Therefore I exposed the times reported by rusage via the new tool events: > user_time and system_time. > > This allows to retrieved them in machine-readable output: > > $ ./perf stat -x, -e duration_time,user_time,system_time,cache-misses -- grep -q -r duration_time tools/perf > 72134524,ns,duration_time:u,72134524,100.00,, > 65225000,ns,user_time:u,65225000,100.00,, > 6865000,ns,ssystem_time:u,6865000,100.00,, > 38705,,cache-misses:u,71189328,100.00,, > > The changes are mostly inspired by the code for the only other available > tool event: 'duration_time'. > > For now the new user_time and system_time events are not usable with metrics > (as far as I understand). > > The patch applies cleanly on linux-next/next-20220420.
I tested and applied the first two patches, please take a look at that "OR" problem with patch 3.
- Arnaldo
> [PATCH v4 1/3] perf stat: introduce stats for the user and system > [PATCH v4 2/3] perf stat: add user_time and system_time events > [PATCH v4 3/3] perf list: print all available tool events > > [v2]: Split up the changes into separate commits. > > [v3]: * Use nanoseconds as suggested by Xing Zhengjun. > * Squash [PATCH v2 2/4] and [PATCH v2 4/4], because it only adds the unit string. > > [v4]: * Simplify event names to only user_time and system_time as suggested by Ian Rogers. > > Florian Fischer
| |