Ingo Molnar dd86e72abd perf stat: Count branches first ... Count branches first, cache-misses second. The reason is that on x86 branches are not counted by all counters on all CPUs. Before: Performance counter stats for 'ls': 0.756653 task-clock-msecs # 0.802 CPUs 0 context-switches # 0.000 M/sec 0 CPU-migrations # 0.000 M/sec 250 page-faults # 0.330 M/sec 2375725 cycles # 3139.781 M/sec 1628129 instructions # 0.685 IPC 19643 cache-references # 25.960 M/sec 4608 cache-misses # 6.090 M/sec 342532 branches # 452.694 M/sec <not counted> branch-misses 0.000943356 seconds time elapsed After: Performance counter stats for 'ls': 1.056734 task-clock-msecs # 0.859 CPUs 0 context-switches # 0.000 M/sec 0 CPU-migrations # 0.000 M/sec 259 page-faults # 0.245 M/sec 3345932 cycles # 3166.295 M/sec 3074090 instructions # 0.919 IPC 616928 branches # 583.806 M/sec 39279 branch-misses # 6.367 % 21312 cache-references # 20.168 M/sec 3661 cache-misses # 3.464 M/sec 0.001230551 seconds time elapsed (also prettify the printout of branch misses, in case it's getting scaled.) Cc: Tim Blechmann <tim@klingt.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> LKML-Reference: <4ADC3975.8050109@klingt.org> Signed-off-by: Ingo Molnar <mingo@elte.hu> --- tools/perf/builtin-stat.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index c373683..95a55ea 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -59,6 +59,8 @@ static struct perf_event_attr default_attrs[] = { { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS }, { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_REFERENCES}, { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_MISSES }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_INSTRUCTIONS}, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_MISSES }, }; --- tools/perf/builtin-stat.c | 20 ++++++++++---------- 1 files changed, 10 insertions(+), 10 deletions(-) diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 95a55ea..90e0a26 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -50,17 +50,17 @@ static struct perf_event_attr default_attrs[] = { - { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK }, - { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CONTEXT_SWITCHES}, - { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CPU_MIGRATIONS }, - { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_PAGE_FAULTS }, - - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CPU_CYCLES }, - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS }, - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_REFERENCES}, - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_MISSES }, - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_INSTRUCTIONS}, - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_MISSES }, + { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK }, + { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CONTEXT_SWITCHES }, + { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CPU_MIGRATIONS }, + { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_PAGE_FAULTS }, + + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CPU_CYCLES }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_REFERENCES }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_MISSES }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_INSTRUCTIONS }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_MISSES }, };		2009-10-19 13:36:32 +02:00
..
Documentation	perf timechart: Add a power-only mode	2009-10-01 09:26:40 +02:00
util	perf tools: Use DECLARE_BITMAP instead of an open-coded array	2009-10-19 09:26:35 +02:00
.gitignore	perf tools: .gitignore += perf*.html	2009-09-24 14:01:22 +02:00
builtin-annotate.c	perf tools: Move dereference after NULL test	2009-10-17 09:29:10 +02:00
builtin-help.c	perf: Enable more compiler warnings	2009-08-16 10:47:47 +02:00
builtin-list.c	perf list: Fix large list output by using the pager	2009-08-13 09:05:48 +02:00
builtin-record.c	perf tools: Introduce bitmask'ed additional headers	2009-10-19 09:26:35 +02:00
builtin-report.c	perf tools: Move dereference after NULL test	2009-10-17 09:29:10 +02:00
builtin-sched.c	perf tools: Use kernel bitmap library	2009-10-19 09:26:34 +02:00
builtin-stat.c	perf stat: Count branches first	2009-10-19 13:36:32 +02:00
builtin-timechart.c	perf timechart: Add a power-only mode	2009-10-01 09:26:40 +02:00
builtin-top.c	perf tools: Remove expensive old debug code from perf top	2009-10-13 15:30:15 +02:00
builtin-trace.c	perf tools: Move dereference after NULL test	2009-10-17 09:29:10 +02:00
builtin.h	perf: Add the timechart tool	2009-09-19 11:42:13 +02:00
command-list.txt	perf: Add timechart help text and add timechart to "perf help"	2009-09-19 18:57:53 +02:00
CREDITS	perf_counter tools: Add CREDITS file for Git contributors	2009-06-24 19:54:29 +02:00
design.txt	perf_event: Provide vmalloc() based mmap() backing	2009-10-06 14:21:50 +02:00
Makefile	perf tools: Display better error messages on missing packages	2009-10-19 10:06:37 +02:00
perf.c	perf tools: Do not manually count string lengths	2009-10-13 11:55:31 +02:00
perf.h	perf: Do the big rename: Performance Counters -> Performance Events	2009-09-21 14:28:04 +02:00