| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Metrics are their own unit and these variables held broken metrics
previously and now just hold the value NULL. Remove code that used
these variables.
Reviewed-by: John Garry <[email protected]>
Reviewed-by: Kajol Jain <[email protected]>
Signed-off-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Caleb Biggers <[email protected]>
Cc: Florian Fischer <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jing Zhang <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Kang Minchul <[email protected]>
Cc: Kim Phillips <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mike Leach <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Perry Taylor <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Ravi Bangoria <[email protected]>
Cc: Rob Herring <[email protected]>
Cc: Sandipan Das <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Cc: [email protected]
Cc: [email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
The data type of the verbose variable is integer and can be negative,
replace improperly used cases in a unified manner:
1. if (verbose) => if (verbose > 0)
2. if (!verbose) => if (verbose <= 0)
3. if (XX && verbose) => if (XX && verbose > 0)
4. if (XX && !verbose) => if (XX && verbose <= 0)
Reviewed-by: Adrian Hunter <[email protected]>
Signed-off-by: Yang Jihong <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Carsten Haitzler <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Martin KaFai Lau <[email protected]>
Cc: Masami Hiramatsu <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Ravi Bangoria <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
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 bb8bc52e7578) so I'm
afraid it won't be applied cleanly.
Fixes: d0a0a511493d2695 ("perf stat: Fix forked applications enablement of counters")
Reported-by: Kevin Nomura <[email protected]>
Signed-off-by: Namhyung Kim <[email protected]>
Tested-by: Thomas Richter <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sumanth Korikkar <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
The group option predates grouping events using curly braces added in
commit 89efb029502d7f2d ("perf tools: Add support to parse event group
syntax").
The --group option was retained for legacy support (in August
2012) but keeping it adds complexity.
Signed-off-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Alexey Bayduraev <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Anshuman Khandual <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Eelco Chaudron <[email protected]>
Cc: German Gomez <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Kim Phillips <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Riccardo Mancini <[email protected]>
Cc: Sandipan Das <[email protected]>
Cc: Sean Christopherson <[email protected]>
Cc: Shaomin Deng <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Timothy Hayes <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Cc: [email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Remove unnecessary include of internal threadmap.h and refcount.h in
thread_map.h. Switch to using public APIs when possible or including
the internal header file in the C file. Fix a transitive dependency in
openat-syscall.c broken by the clean up.
Signed-off-by: Ian Rogers <[email protected]>
Acked-by: Namhyung Kim <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andrii Nakryiko <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Masahiro Yamada <[email protected]>
Cc: Nick Desaulniers <[email protected]>
Cc: Nicolas Schier <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
The --interval-clear option makes perf stat to clear the terminal at
each interval. But it doesn't need to clear the screen when it saves
to a file. Make it fail when it's enabled with the output options.
$ perf stat -I 1 --interval-clear -o myfile true
--interval-clear does not work with output
Usage: perf stat [<options>] [<command>]
-o, --output <file> output file name
--log-fd <n> log output to fd, instead of stderr
--interval-clear clear screen in between new interval
Signed-off-by: Namhyung Kim <[email protected]>
Tested-by: Arnaldo Carvalho de Melo <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Use sig_atomic_t for variables written/accessed in signal
handlers. This is undefined behavior as per:
https://wiki.sei.cmu.edu/confluence/display/c/SIG31-C.+Do+not+access+shared+objects+in+signal+handlers
Signed-off-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Alexey Bayduraev <[email protected]>
Cc: German Gomez <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Use the global quiet variable everywhere so that all tools hide warnings
in quiet mode and update the documentation to reflect this.
'perf probe' claimed that errors are not printed in quiet mode but I
don't see this so remove it from the docs.
Signed-off-by: James Clark <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
The recent change in the perf stat broke the percore event display.
Note that the aggr counts are already processed so that the every
sibling thread in the same core will get the per-core counter values.
Check percore evsels and skip the sibling threads in the display.
Signed-off-by: Namhyung Kim <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
This function updates the shadow stats using the aggregated counts
uniformly since it uses the aggr_counts for the every aggr mode.
It'd have duplicate shadow stats for each items for now since the
display routines will update them once again. But that'd be fine
as it shows the average values and it'd be gone eventually.
Signed-off-by: Namhyung Kim <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
The perf_stat_process_percore() is to aggregate counts for an event per-core
even if the aggr_mode is AGGR_NONE. This is enabled when user requested it
on the command line.
To handle that, it keeps the per-cpu counts at first. And then it aggregates
the counts that have the same core id in the aggr->counts and updates the
values for each cpu back.
Later, per-core events will skip one of the CPUs unless percore-show-thread
option is given. In that case, it can simply print all cpu stats with the
updated (per-core) values.
Signed-off-by: Namhyung Kim <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
The perf_stat_merge_counters() is to aggregate the same events in different
PMUs like in case of uncore or hybrid. The same logic is in the stat-display
routines but I think it should be handled when it processes the event counters.
As it works on the aggr_counters, it doesn't change the output yet.
Signed-off-by: Namhyung Kim <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
It'd do more processing with aggregation. Let's split the function so that it
can be shared with by process_stat_round_event() too.
Signed-off-by: Namhyung Kim <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
The evsel->stats->aggr->count should be reset for interval processing
since we want to use the values directly for display.
Signed-off-by: Namhyung Kim <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
In the process_stat_config_event() it sets the aggr_mode that means the
earlier evlist__alloc_stats() cannot allocate the aggr counts due to the
missing aggr_mode.
Do it after setting the aggr_map using evlist__alloc_aggr_stats().
Signed-off-by: Namhyung Kim <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Per-thread aggregation doesn't use the CPU numbers but the logic should
be the same. Initialize cpu_aggr_map separately for AGGR_THREAD and use
thread map idx to aggregate counter values.
Signed-off-by: Namhyung Kim <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Add a logic to aggregate counter values to the new evsel->stats->aggr.
This is not used yet so shadow stats are not updated. But later patch
will convert the existing code to use it.
With that, we don't need to handle AGGR_GLOBAL specially anymore. It
can use the same logic with counts, prev_counts and aggr_counts.
Signed-off-by: Namhyung Kim <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
The perf_stat_config.aggr_map should have a correct size of the
aggregation map. Use it to allocate aggr_counts.
Also AGGR_NONE with per-core events can be tricky because it doesn't
aggreate basically but it needs to do so for per-core events only.
So only per-core evsels will have stats->aggr data.
Note that other caller of evlist__alloc_stat() might not have
stat_config or aggr_map.
Signed-off-by: Namhyung Kim <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
In case of no aggregation, it needs to keep the original (cpu) ordering
in the aggr_map so that it can be in sync with the cpu map. This will
make the code easier to handle AGGR_NONE similar to others.
Signed-off-by: Namhyung Kim <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Likewise, add an aggr_id for cpu for none aggregation mode. This is not
used actually yet but later code will use to unify the aggregation code.
No functional change intended.
Signed-off-by: Namhyung Kim <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
To make the code simpler, I'd like to use the same aggregation code for
the global mode. We can simply add an id function to return cpu 0 and
use print_aggr().
No functional change intended.
Signed-off-by: Namhyung Kim <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Athira Jajeev <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Michael Petlan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Now it's using the global rt_stat, no need to use per-thread stats. Let
get rid of them.
Reviewed-by: James Clark <[email protected]>
Signed-off-by: Namhyung Kim <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Since the first if statment is covered by the following one, clean up
the first if statment.
Signed-off-by: Shang XiaoJing <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
As two cases in process_evlist has same behavior, make the first fall
through to the second.
Commiter notes:
Added __fallthrough, the kernel has "fallthrough", we need to make
tools/ use it.
Signed-off-by: Shang XiaoJing <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Pass state necessary for core_wide into the expression parser. Add
system_wide and user_requested_cpu_list to perf_stat_config to make it
available at display time. evlist isn't used as the
evlist__create_maps, that computes user_requested_cpus, needs the list
of events which is generated by the metric.
Signed-off-by: Ian Rogers <[email protected]>
Cc: Ahmad Yasin <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Caleb Biggers <[email protected]>
Cc: Florian Fischer <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: John Garry <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Kshipra Bopardikar <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Miaoqian Lin <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Perry Taylor <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Thomas Richter <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Having metric parsing as part of argument processing causes issues as
flags like metric-no-group may be specified later. It also denies the
opportunity to optimize the events on SMT systems where fewer events
may be possible if we know the target is system-wide. Move metric
parsing to after command line option parsing. Because of how stat runs
this moves the parsing after record/report which fail to work with
metrics currently anyway.
Signed-off-by: Ian Rogers <[email protected]>
Cc: Ahmad Yasin <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Caleb Biggers <[email protected]>
Cc: Florian Fischer <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: John Garry <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Kshipra Bopardikar <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Miaoqian Lin <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Perry Taylor <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Thomas Richter <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
In perf/Documentation/perf-stat.txt, for "--td-level" the default "0" means
the max level that the current hardware support.
So we need initialize the stat_config.topdown_level to TOPDOWN_MAX_LEVEL
when “--td-level=0” or no “--td-level” option. Otherwise, for the
hardware with a max level is 2, the 2nd level metrics disappear for raw
events in this case.
The issue cannot be observed for the perf stat default or "--topdown"
options. This commit fixes the raw events issue and removes the
duplicated code for the perf stat default.
Before:
# ./perf stat -e "cpu-clock,context-switches,cpu-migrations,page-faults,instructions,cycles,ref-cycles,branches,branch-misses,{slots,topdown-retiring,topdown-bad-spec,topdown-fe-bound,topdown-be-bound,topdown-heavy-ops,topdown-br-mispredict,topdown-fetch-lat,topdown-mem-bound}" sleep 1
Performance counter stats for 'sleep 1':
1.03 msec cpu-clock # 0.001 CPUs utilized
1 context-switches # 966.216 /sec
0 cpu-migrations # 0.000 /sec
60 page-faults # 57.973 K/sec
1,132,112 instructions # 1.41 insn per cycle
803,872 cycles # 0.777 GHz
1,909,120 ref-cycles # 1.845 G/sec
236,634 branches # 228.640 M/sec
6,367 branch-misses # 2.69% of all branches
4,823,232 slots # 4.660 G/sec
1,210,536 topdown-retiring # 25.1% Retiring
699,841 topdown-bad-spec # 14.5% Bad Speculation
1,777,975 topdown-fe-bound # 36.9% Frontend Bound
1,134,878 topdown-be-bound # 23.5% Backend Bound
189,146 topdown-heavy-ops # 182.756 M/sec
662,012 topdown-br-mispredict # 639.647 M/sec
1,097,048 topdown-fetch-lat # 1.060 G/sec
416,121 topdown-mem-bound # 402.063 M/sec
1.002423690 seconds time elapsed
0.002494000 seconds user
0.000000000 seconds sys
After:
# ./perf stat -e "cpu-clock,context-switches,cpu-migrations,page-faults,instructions,cycles,ref-cycles,branches,branch-misses,{slots,topdown-retiring,topdown-bad-spec,topdown-fe-bound,topdown-be-bound,topdown-heavy-ops,topdown-br-mispredict,topdown-fetch-lat,topdown-mem-bound}" sleep 1
Performance counter stats for 'sleep 1':
1.13 msec cpu-clock # 0.001 CPUs utilized
1 context-switches # 882.128 /sec
0 cpu-migrations # 0.000 /sec
61 page-faults # 53.810 K/sec
1,137,612 instructions # 1.29 insn per cycle
881,477 cycles # 0.778 GHz
2,093,496 ref-cycles # 1.847 G/sec
236,356 branches # 208.496 M/sec
7,090 branch-misses # 3.00% of all branches
5,288,862 slots # 4.665 G/sec
1,223,697 topdown-retiring # 23.1% Retiring
767,403 topdown-bad-spec # 14.5% Bad Speculation
2,053,322 topdown-fe-bound # 38.8% Frontend Bound
1,244,438 topdown-be-bound # 23.5% Backend Bound
186,665 topdown-heavy-ops # 3.5% Heavy Operations # 19.6% Light Operations
725,922 topdown-br-mispredict # 13.7% Branch Mispredict # 0.8% Machine Clears
1,327,400 topdown-fetch-lat # 25.1% Fetch Latency # 13.7% Fetch Bandwidth
497,775 topdown-mem-bound # 9.4% Memory Bound # 14.1% Core Bound
1.002701530 seconds time elapsed
0.002744000 seconds user
0.000000000 seconds sys
Fixes: 63e39aa6ae103451 ("perf stat: Support L2 Topdown events")
Reviewed-by: Kan Liang <[email protected]>
Signed-off-by: Xing Zhengjun <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
If a weak group is broken then the reset_group flag remains set for
the next run. Having reset_group set means the counter isn't created
and ultimately a segfault.
A simple reproduction of this is:
# perf stat -r2 -e '{cycles,cycles,cycles,cycles,cycles,cycles,cycles,cycles,cycles,cycles}:W
which will be added as a test in the next patch.
Fixes: 4804e0111662d7d8 ("perf stat: Use affinity for opening events")
Reviewed-by: Andi Kleen <[email protected]>
Signed-off-by: Ian Rogers <[email protected]>
Tested-by: Arnaldo Carvalho de Melo <[email protected]>
Tested-by: Xing Zhengjun <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
util/topdown.h is included twice in builtin-stat.c,
remove one of them.
Reported-by: Abaci Robot <[email protected]>
Signed-off-by: Yang Li <[email protected]>
Tested-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=1818
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
CSV output is tricky to format and column layout changes are susceptible
to breaking parsers. New JSON-formatted output has variable names to
identify fields that are consistent and informative, making the output
parseable.
CSV output example:
1.20,msec,task-clock:u,1204272,100.00,0.697,CPUs utilized
0,,context-switches:u,1204272,100.00,0.000,/sec
0,,cpu-migrations:u,1204272,100.00,0.000,/sec
70,,page-faults:u,1204272,100.00,58.126,K/sec
JSON output example:
{"counter-value" : "3805.723968", "unit" : "msec", "event" :
"cpu-clock", "event-runtime" : 3805731510100.00, "pcnt-running"
: 100.00, "metric-value" : 4.007571, "metric-unit" : "CPUs utilized"}
{"counter-value" : "6166.000000", "unit" : "", "event" :
"context-switches", "event-runtime" : 3805723045100.00, "pcnt-running"
: 100.00, "metric-value" : 1.620191, "metric-unit" : "K/sec"}
{"counter-value" : "466.000000", "unit" : "", "event" :
"cpu-migrations", "event-runtime" : 3805727613100.00, "pcnt-running"
: 100.00, "metric-value" : 122.447136, "metric-unit" : "/sec"}
{"counter-value" : "208.000000", "unit" : "", "event" :
"page-faults", "event-runtime" : 3805726799100.00, "pcnt-running"
: 100.00, "metric-value" : 54.654516, "metric-unit" : "/sec"}
Also added documentation for JSON option.
There is some tidy up of CSV code including a potential memory over run
in the os.nfields set up. To facilitate this an AGGR_MAX value is added.
Committer notes:
Fixed up using PRIu64 to format u64 values, not %lu.
Committer testing:
⬢[acme@toolbox perf]$ perf stat -j sleep 1
{"counter-value" : "0.731750", "unit" : "msec", "event" : "task-clock:u", "event-runtime" : 731750, "pcnt-running" : 100.00, "metric-value" : 0.000731, "metric-unit" : "CPUs utilized"}
{"counter-value" : "0.000000", "unit" : "", "event" : "context-switches:u", "event-runtime" : 731750, "pcnt-running" : 100.00, "metric-value" : 0.000000, "metric-unit" : "/sec"}
{"counter-value" : "0.000000", "unit" : "", "event" : "cpu-migrations:u", "event-runtime" : 731750, "pcnt-running" : 100.00, "metric-value" : 0.000000, "metric-unit" : "/sec"}
{"counter-value" : "75.000000", "unit" : "", "event" : "page-faults:u", "event-runtime" : 731750, "pcnt-running" : 100.00, "metric-value" : 102.494021, "metric-unit" : "K/sec"}
{"counter-value" : "578765.000000", "unit" : "", "event" : "cycles:u", "event-runtime" : 379366, "pcnt-running" : 49.00, "metric-value" : 0.790933, "metric-unit" : "GHz"}
{"counter-value" : "1298.000000", "unit" : "", "event" : "stalled-cycles-frontend:u", "event-runtime" : 768020, "pcnt-running" : 100.00, "metric-value" : 0.224271, "metric-unit" : "frontend cycles idle"}
{"counter-value" : "21984.000000", "unit" : "", "event" : "stalled-cycles-backend:u", "event-runtime" : 768020, "pcnt-running" : 100.00, "metric-value" : 3.798433, "metric-unit" : "backend cycles idle"}
{"counter-value" : "468197.000000", "unit" : "", "event" : "instructions:u", "event-runtime" : 768020, "pcnt-running" : 100.00, "metric-value" : 0.808959, "metric-unit" : "insn per cycle"}
{"metric-value" : 0.046955, "metric-unit" : "stalled cycles per insn"}
{"counter-value" : "103335.000000", "unit" : "", "event" : "branches:u", "event-runtime" : 768020, "pcnt-running" : 100.00, "metric-value" : 141.216262, "metric-unit" : "M/sec"}
{"counter-value" : "2381.000000", "unit" : "", "event" : "branch-misses:u", "event-runtime" : 388654, "pcnt-running" : 50.00, "metric-value" : 2.304156, "metric-unit" : "of all branches"}
⬢[acme@toolbox perf]$
Signed-off-by: Claire Jensen <[email protected]>
Acked-by: Namhyung Kim <[email protected]>
Tested-by: Arnaldo Carvalho de Melo <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Alyssa Ross <[email protected]>
Cc: Claire Jensen <[email protected]>
Cc: Florian Fischer <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Like Xu <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sandipan Das <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Ian Rogers <[email protected]>
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
This extracts common code from the branches of the forks if-then-else.
enable_counters(), which was at the beginning of both branches of the
conditional, is now unconditional; evlist__start_workload() is extracted
to a different if, which enables making the common clocking code
unconditional.
Reviewed-by: Leo Yan <[email protected]>
Signed-off-by: Adrián Herrera Arcila <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Song Liu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Topdown metrics are missed in the default perf stat on the hybrid machine,
add Topdown metrics in default perf stat for hybrid systems.
Currently, we support the perf metrics Topdown for the p-core PMU in the
perf stat default, the perf metrics Topdown support for e-core PMU will be
implemented later separately. Refactor the code adds two x86 specific
functions. Widen the size of the event name column by 7 chars, so that all
metrics after the "#" become aligned again.
The perf metrics topdown feature is supported on the cpu_core of ADL. The
dedicated perf metrics counter and the fixed counter 3 are used for the
topdown events. Adding the topdown metrics doesn't trigger multiplexing.
Before:
# ./perf stat -a true
Performance counter stats for 'system wide':
53.70 msec cpu-clock # 25.736 CPUs utilized
80 context-switches # 1.490 K/sec
24 cpu-migrations # 446.951 /sec
52 page-faults # 968.394 /sec
2,788,555 cpu_core/cycles/ # 51.931 M/sec
851,129 cpu_atom/cycles/ # 15.851 M/sec
2,974,030 cpu_core/instructions/ # 55.385 M/sec
416,919 cpu_atom/instructions/ # 7.764 M/sec
586,136 cpu_core/branches/ # 10.916 M/sec
79,872 cpu_atom/branches/ # 1.487 M/sec
14,220 cpu_core/branch-misses/ # 264.819 K/sec
7,691 cpu_atom/branch-misses/ # 143.229 K/sec
0.002086438 seconds time elapsed
After:
# ./perf stat -a true
Performance counter stats for 'system wide':
61.39 msec cpu-clock # 24.874 CPUs utilized
76 context-switches # 1.238 K/sec
24 cpu-migrations # 390.968 /sec
52 page-faults # 847.097 /sec
2,753,695 cpu_core/cycles/ # 44.859 M/sec
903,899 cpu_atom/cycles/ # 14.725 M/sec
2,927,529 cpu_core/instructions/ # 47.690 M/sec
428,498 cpu_atom/instructions/ # 6.980 M/sec
581,299 cpu_core/branches/ # 9.470 M/sec
83,409 cpu_atom/branches/ # 1.359 M/sec
13,641 cpu_core/branch-misses/ # 222.216 K/sec
8,008 cpu_atom/branch-misses/ # 130.453 K/sec
14,761,308 cpu_core/slots/ # 240.466 M/sec
3,288,625 cpu_core/topdown-retiring/ # 22.3% retiring
1,323,323 cpu_core/topdown-bad-spec/ # 9.0% bad speculation
5,477,470 cpu_core/topdown-fe-bound/ # 37.1% frontend bound
4,679,199 cpu_core/topdown-be-bound/ # 31.7% backend bound
646,194 cpu_core/topdown-heavy-ops/ # 4.4% heavy operations # 17.9% light operations
1,244,999 cpu_core/topdown-br-mispredict/ # 8.4% branch mispredict # 0.5% machine clears
3,891,800 cpu_core/topdown-fetch-lat/ # 26.4% fetch latency # 10.7% fetch bandwidth
1,879,034 cpu_core/topdown-mem-bound/ # 12.7% memory bound # 19.0% Core bound
0.002467839 seconds time elapsed
Reviewed-by: Kan Liang <[email protected]>
Signed-off-by: Xing Zhengjun <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Acked-by: Namhyung Kim <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Current perf stat uses the evlist__add_default_attrs() to add the
generic default attrs, and uses arch_evlist__add_default_attrs() to add
the Arch specific default attrs, e.g., Topdown for x86.
It works well for the non-hybrid platforms. However, for a hybrid
platform, the hard code generic default attrs don't work.
Uses arch_evlist__add_default_attrs() to replace the
evlist__add_default_attrs(). The arch_evlist__add_default_attrs() is
modified to invoke the same __evlist__add_default_attrs() for the
generic default attrs. No functional change.
Add default_null_attrs[] to indicate the arch specific attrs.
No functional change for the arch specific default attrs either.
Signed-off-by: Kan Liang <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Acked-by: Namhyung Kim <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Xing Zhengjun <[email protected]>
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
This reverts commit Fixes: ac2dc29edd21f9ec ("perf stat: Add default
hybrid events")
Between this patch and the reverted patch, the commit 6c1912898ed21bef
("perf parse-events: Rename parse_events_error functions") and the
commit 07eafd4e053a41d7 ("perf parse-event: Add init and exit to
parse_event_error") clean up the parse_events_error_*() codes. The
related change is also reverted.
The reverted patch is hard to be extended to support new default events,
e.g., Topdown events, and the existing "--detailed" option on a hybrid
platform.
A new solution will be proposed in the following patch to enable the
perf stat default on a hybrid platform.
Signed-off-by: Kan Liang <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Acked-by: Namhyung Kim <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Xing Zhengjun <[email protected]>
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
perf already support ignore_missing_thread for -p, but not yet
applied to `perf stat -p <pid>`. This patch enables ignore_missing_thread
for `perf stat -p <pid>`.
Committer notes:
And here is a refresher about the 'ignore_missing_thread' knob, from a
previous patch using it:
ca8000684ec4e66f ("perf evsel: Enable ignore_missing_thread for pid option")
---
While monitoring a multithread process with pid option, perf sometimes
may return sys_perf_event_open failure with 3(No such process) if any of
the process's threads die before we open the event. However, we want
perf continue monitoring the remaining threads and do not exit with
error.
---
Signed-off-by: Gang Li <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Uncore events require a CPU i.e. it cannot be -1.
The evsel system_wide flag is intended for events that should be on every
CPU, which does not make sense for uncore events because uncore events do
not map one-to-one with CPUs.
These 2 requirements are not exactly the same, so introduce a new flag
'requires_cpu' for the uncore case.
Signed-off-by: Adrian Hunter <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Acked-by: Namhyung Kim <[email protected]>
Cc: Alexey Bayduraev <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Leo Yan <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
If any member in a group has a different cpu mask than the other
members, the current perf stat disables group. when the perf metrics
topdown events are part of the group, the below <not supported> error
will be triggered.
$ perf stat -e "{slots,topdown-retiring,uncore_imc_free_running_0/dclk/}" -a sleep 1
WARNING: grouped events cpus do not match, disabling group:
anon group { slots, topdown-retiring, uncore_imc_free_running_0/dclk/ }
Performance counter stats for 'system wide':
141,465,174 slots
<not supported> topdown-retiring
1,605,330,334 uncore_imc_free_running_0/dclk/
The perf metrics topdown events must always be grouped with a slots
event as leader.
Factor out evsel__remove_from_group() to only remove the regular events
from the group.
Remove evsel__must_be_in_group(), since no one use it anymore.
With the patch, the topdown events aren't broken from the group for the
splitting.
$ perf stat -e "{slots,topdown-retiring,uncore_imc_free_running_0/dclk/}" -a sleep 1
WARNING: grouped events cpus do not match, disabling group:
anon group { slots, topdown-retiring, uncore_imc_free_running_0/dclk/ }
Performance counter stats for 'system wide':
346,110,588 slots
124,608,256 topdown-retiring
1,606,869,976 uncore_imc_free_running_0/dclk/
1.003877592 seconds time elapsed
Fixes: a9a1790247bdcf3b ("perf stat: Ensure group is defined on top of the same cpu mask")
Signed-off-by: Kan Liang <[email protected]>
Acked-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Since for cpu_core or cpu_atom, they have different topdown events
groups.
For cpu_core, --topdown equals to:
"{slots,cpu_core/topdown-retiring/,cpu_core/topdown-bad-spec/,
cpu_core/topdown-fe-bound/,cpu_core/topdown-be-bound/,
cpu_core/topdown-heavy-ops/,cpu_core/topdown-br-mispredict/,
cpu_core/topdown-fetch-lat/,cpu_core/topdown-mem-bound/}"
For cpu_atom, --topdown equals to:
"{cpu_atom/topdown-retiring/,cpu_atom/topdown-bad-spec/,
cpu_atom/topdown-fe-bound/,cpu_atom/topdown-be-bound/}"
To simplify the implementation, on hybrid, --topdown is used
together with --cputype. If without --cputype, it uses cpu_core
topdown events by default.
# ./perf stat --topdown -a sleep 1
WARNING: default to use cpu_core topdown events
Performance counter stats for 'system wide':
retiring bad speculation frontend bound backend bound heavy operations light operations branch mispredict machine clears fetch latency fetch bandwidth memory bound Core bound
4.1% 0.0% 5.1% 90.8% 2.3% 1.8% 0.0% 0.0% 4.2% 0.9% 9.9% 81.0%
1.002624229 seconds time elapsed
# ./perf stat --topdown -a --cputype atom sleep 1
Performance counter stats for 'system wide':
retiring bad speculation frontend bound backend bound
13.5% 0.1% 31.2% 55.2%
1.002366987 seconds time elapsed
Reviewed-by: Kan Liang <[email protected]>
Signed-off-by: Xing Zhengjun <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
For hybrid events, by default stat aggregates and reports the event counts
per pmu.
# ./perf stat -e cycles -a sleep 1
Performance counter stats for 'system wide':
14,066,877,268 cpu_core/cycles/
6,814,443,147 cpu_atom/cycles/
1.002760625 seconds time elapsed
Sometimes, it's also useful to aggregate event counts from all PMUs.
Create a new option '--hybrid-merge' to enable that behavior and report
the counts without PMUs.
# ./perf stat -e cycles -a --hybrid-merge sleep 1
Performance counter stats for 'system wide':
20,732,982,512 cycles
1.002776793 seconds time elapsed
Reviewed-by: Kan Liang <[email protected]>
Signed-off-by: Xing Zhengjun <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
It bothered me that during benchmarking using 'perf stat' (to collect
for example CPU cache events) I could not simultaneously retrieve the
times spend in user or kernel mode in a machine readable format.
When running 'perf stat' the output for humans contains the times
reported by rusage and wait4.
$ perf stat -e cache-misses:u -- true
Performance counter stats for 'true':
4,206 cache-misses:u
0.001113619 seconds time elapsed
0.001175000 seconds user
0.000000000 seconds sys
But 'perf stat's machine-readable format does not provide this information.
$ perf stat -x, -e cache-misses:u -- true
4282,,cache-misses:u,492859,100.00,,
I found no way to retrieve this information using the available events
while using machine-readable output.
This patch adds two new tool internal events 'user_time' and
'system_time', similarly to the already present 'duration_time' event.
Both events use the already collected rusage information obtained by
wait4 and tracked in the global ru_stats.
Examples presenting cache-misses and rusage information in both human
and machine-readable form:
$ perf stat -e duration_time,user_time,system_time,cache-misses -- grep -q -r duration_time .
Performance counter stats for 'grep -q -r duration_time .':
67,422,542 ns duration_time:u
50,517,000 ns user_time:u
16,839,000 ns system_time:u
30,937 cache-misses:u
0.067422542 seconds time elapsed
0.050517000 seconds user
0.016839000 seconds sys
$ perf stat -x, -e duration_time,user_time,system_time,cache-misses -- grep -q -r duration_time .
72134524,ns,duration_time:u,72134524,100.00,,
65225000,ns,user_time:u,65225000,100.00,,
6865000,ns,system_time:u,6865000,100.00,,
38705,,cache-misses:u,71189328,100.00,,
Signed-off-by: Florian Fischer <[email protected]>
Tested-by: Arnaldo Carvalho de Melo <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
This is preparation for exporting rusage values as tool events.
Add new global stats tracking the values obtained via rusage.
For now only ru_utime and ru_stime are part of the tracked stats.
Both are stored as nanoseconds to be consistent with 'duration_time',
although the finest resolution the struct timeval data in rusage
provides are microseconds.
Signed-off-by: Florian Fischer <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Xing Zhengjun <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
evlist contains cpus and all_cpus. all_cpus is the union of the cpu maps
of all evsels.
For non-task targets, cpus is set to be cpus requested from the command
line, defaulting to all online cpus if no cpus are specified.
For an uncore event, all_cpus may be just CPU 0 or every online CPU.
This causes all_cpus to have fewer values than the cpus variable which
is confusing given the 'all' in the name.
To try to make the behavior clearer, rename cpus to user_requested_cpus
and add comments on the two struct variables.
Signed-off-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Alexander Antonov <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Alexei Starovoitov <[email protected]>
Cc: Alexey Bayduraev <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Andrii Nakryiko <[email protected]>
Cc: Daniel Borkmann <[email protected]>
Cc: German Gomez <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: John Fastabend <[email protected]>
Cc: John Garry <[email protected]>
Cc: KP Singh <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Martin KaFai Lau <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Mike Leach <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Riccardo Mancini <[email protected]>
Cc: Song Liu <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Suzuki Poulouse <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Yonghong Song <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Passing NULL to perf_cpu_map__max doesn't make sense as there is no
valid max. Avoid this problem by null checking in
perf_stat_init_aggr_mode.
Signed-off-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Alexander Antonov <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Alexei Starovoitov <[email protected]>
Cc: Alexey Bayduraev <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Andrii Nakryiko <[email protected]>
Cc: Daniel Borkmann <[email protected]>
Cc: German Gomez <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: John Fastabend <[email protected]>
Cc: John Garry <[email protected]>
Cc: KP Singh <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Martin KaFai Lau <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Mike Leach <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Riccardo Mancini <[email protected]>
Cc: Song Liu <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Suzuki Poulouse <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Yonghong Song <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
We support short command 'rec*' for 'record' and 'rep*' for 'report' in
lots of sub-commands, but the matching is not quite strict currnetly.
It may be puzzling sometime, like we mis-type a 'recport' to report but
it will perform 'record' in fact without any message.
To fix this, add a check to ensure that the short cmd is valid prefix
of the real command.
Committer testing:
[root@quaco ~]# perf c2c re sleep 1
Usage: perf c2c {record|report}
-v, --verbose be more verbose (show counter open errors, etc)
# perf c2c rec sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.038 MB perf.data (16 samples) ]
# perf c2c recport sleep 1
Usage: perf c2c {record|report}
-v, --verbose be more verbose (show counter open errors, etc)
# perf c2c record sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.038 MB perf.data (15 samples) ]
# perf c2c records sleep 1
Usage: perf c2c {record|report}
-v, --verbose be more verbose (show counter open errors, etc)
#
Signed-off-by: Wei Li <[email protected]>
Tested-by: Arnaldo Carvalho de Melo <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Hanjun Guo <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Rui Xiang <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
I have run into the following issue:
# perf stat -a -e new_pmu/INSTRUCTION_7/ -- mytest -c1 7
Performance counter stats for 'system wide':
0 new_pmu/INSTRUCTION_7/
0.000366428 seconds time elapsed
#
The new PMU for s390 counts the execution of certain CPU instructions.
The root cause is the extremely small run time of the mytest program. It
just executes some assembly instructions and then exits.
In above invocation the instruction is executed exactly one time (-c1
option). The PMU is expected to report this one time execution by a
counter value of one, but fails to do so in some cases, not all.
Debugging reveals the invocation of the child process is done
*before* the counter events are installed and enabled.
Tracing reveals that sometimes the child process starts and exits before
the event is installed on all CPUs. The more CPUs the machine has, the
more often this miscount happens.
Fix this by reversing the start of the work load after the events have
been installed on the specified CPUs. Now the comment also matches the
code.
Output after:
# perf stat -a -e new_pmu/INSTRUCTION_7/ -- mytest -c1 7
Performance counter stats for 'system wide':
1 new_pmu/INSTRUCTION_7/
0.000366428 seconds time elapsed
#
Now the correct result is reported rock solid all the time regardless
how many CPUs are online.
Reviewers notes:
Jiri:
Right, without -a the event has enable_on_exec so the race does not
matter, but it's a problem for system wide with fork.
Namhyung:
Agreed. Also we may move the enable_counters() and the clock code out of
the if block to be shared with the else block.
Fixes: acf2892270dcc428 ("perf stat: Use perf_evlist__prepare/start_workload()")
Signed-off-by: Thomas Richter <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
Acked-by: Namhyung Kim <[email protected]>
Acked-by: Sumanth Korikkar <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Sven Schnelle <[email protected]>
Cc: Vasily Gorbik <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Switch from directly accessing the perf_cpu_map to using the appropriate
libperf API when possible. Using the API simplifies the job of
refactoring use of perf_cpu_map.
Signed-off-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Alexey Bayduraev <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: André Almeida <[email protected]>
Cc: Andy Shevchenko <[email protected]>
Cc: Darren Hart <[email protected]>
Cc: Davidlohr Bueso <[email protected]>
Cc: Dmitriy Vyukov <[email protected]>
Cc: Eric Dumazet <[email protected]>
Cc: German Gomez <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jin Yao <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: John Garry <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Madhavan Srinivasan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Masami Hiramatsu <[email protected]>
Cc: Miaoqian Lin <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Riccardo Mancini <[email protected]>
Cc: Shunsuke Nakamura <[email protected]>
Cc: Song Liu <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Stephen Brennan <[email protected]>
Cc: Steven Rostedt (VMware) <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Thomas Richter <[email protected]>
Cc: Yury Norov <[email protected]>
Link: http://lore.kernel.org/lkml/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
I.e. the simple:
$ perf stat sleep 1
Uses a dummy CPU map and thus there is no need to setup/cleanup
affinities to avoid IPIs, etc.
With this we're down to a sched_getaffinity() call, in the libnuma
initialization, that probably can be removed in a followup patch.
Acked-by: Ian Rogers <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ian Rogers <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Namhyung Kim <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
A common problem is confusing CPU map indices with the CPU, by wrapping
the CPU with a struct then this is avoided. This approach is similar to
atomic_t.
Committer notes:
To make it build with BUILD_BPF_SKEL=1 these files needed the
conversions to 'struct perf_cpu' usage:
tools/perf/util/bpf_counter.c
tools/perf/util/bpf_counter_cgroup.c
tools/perf/util/bpf_ftrace.c
Also perf_env__get_cpu() was removed back in "perf cpumap: Switch
cpu_map__build_map to cpu function".
Additionally these needed to be fixed for the ARM builds to complete:
tools/perf/arch/arm/util/cs-etm.c
tools/perf/arch/arm64/util/pmu.c
Suggested-by: John Garry <[email protected]>
Signed-off-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Mike Leach <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Riccardo Mancini <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Suzuki Poulouse <[email protected]>
Cc: Vineet Singh <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
Switch from cpu to cpu_map_idx to reduce confusion.
Signed-off-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: John Garry <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Mike Leach <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Riccardo Mancini <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Suzuki Poulouse <[email protected]>
Cc: Vineet Singh <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|
|
evsel__write_stat_event() was incorrectly passing a cpu map index rather
than a CPU to perf_event__synthesize_stat().
Signed-off-by: Ian Rogers <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Andi Kleen <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: James Clark <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: John Garry <[email protected]>
Cc: Kajol Jain <[email protected]>
Cc: Kan Liang <[email protected]>
Cc: Leo Yan <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: Mathieu Poirier <[email protected]>
Cc: Mike Leach <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Paul Clarke <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Riccardo Mancini <[email protected]>
Cc: Stephane Eranian <[email protected]>
Cc: Suzuki Poulouse <[email protected]>
Cc: Vineet Singh <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
|