perf_counter tools: Also display time-normalized stat results
Commit-ID: be1ac0d81d0e3ab655f8c8ade31fb860ef6aa186 Gitweb: http://git.kernel.org/tip/be1ac0d81d0e3ab655f8c8ade31fb860ef6aa186 Author: Ingo Molnar <[email protected]> AuthorDate: Fri, 29 May 2009 09:10:54 +0200 Committer: Ingo Molnar <[email protected]> CommitDate: Fri, 29 May 2009 09:46:45 +0200
perf_counter tools: Also display time-normalized stat results
Add new column that normalizes counter results by
'nanoseconds spent running' unit.
Before:
Performance counter stats for '/home/mingo/hackbench':
10469.403605 task clock ticks (msecs)
75502 context switches (events)
9501 CPU migrations (events)
36158 pagefaults (events)
31975676185 CPU cycles (events)
26257738659 instructions (events)
108740581 cache references (events)
54606088 cache misses (events)Wall-clock time elapsed: 810.514504 msecs
After:
Performance counter stats for '/home/mingo/hackbench':
10469.403605 task clock ticks (msecs)
75502 context switches # 0.007 M/sec
9501 CPU migrations # 0.001 M/sec
36158 pagefaults # 0.003 M/sec
31975676185 CPU cycles # 3054.202 M/sec
26257738659 instructions # 2508.045 M/sec
108740581 cache references # 10.387 M/sec
54606088 cache misses # 5.216 M/secWall-clock time elapsed: 810.514504 msecs
The advantage of that column is that it is characteristic of the execution workflow, regardless of runtime. Hence 'hackbench 10' will look similar to 'hackbench 15' - while the absolute counter values are very different.
Cc: Peter Zijlstra <[email protected]>
Cc: Mike Galbraith <[email protected]>
Cc: Paul Mackerras <[email protected]>
Cc: Corey Ashford <[email protected]>
Cc: Marcelo Tosatti <[email protected]>
Cc: Arnaldo Carvalho de Melo <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: John Kacur <[email protected]>
LKML-Reference: <new-submission>
--- Documentation/perf_counter/builtin-stat.c | 12 +++++++++++- 1 files changed, 11 insertions(+), 1 deletions(-)
Documentation/perf_counter/builtin-stat.c +11 -1
@@ -74,6 +74,8 @@ static const unsigned int default_count[] = {static __u64 event_res[MAX_COUNTERS ][3 ];static __u64 event_scaled[MAX_COUNTERS ]; +static __u64 runtime_nsecs; +static void create_perfstat_counter (int counter) {struct perf_counter_hw_event hw_event;
@@ -165,6 +167,11 @@ static void read_counter(int counter) ((double )count[0 ] * count[1 ] / count[2 ] +0.5 ); } } +/* +* Save the full runtime - to allow normalization during printout: +*/ +if (event_id[counter] ==EID (PERF_TYPE_SOFTWARE ,PERF_COUNT_TASK_CLOCK )) + runtime_nsecs = count[0 ]; }/*
@@ -190,8 +197,11 @@ static void print_counter(int counter)fprintf (stderr," %14.6f %-20s (msecs)" , msecs,event_name (counter)); }else { -fprintf (stderr," %14Ld %-20s (events)" , +fprintf (stderr," %14Ld %-20s" , count[0 ],event_name (counter)); +if (runtime_nsecs) +fprintf (stderr," # %12.3f M/sec" , + (double )count[0 ]/runtime_nsecs*1000.0 ); }if (scaled) fprintf(stderr, " (scaled from %.2f%%)",