From: Ingo Molnar Date: Thu, 19 May 2011 11:30:56 +0000 (+0200) Subject: perf stat: Add -d -d and -d -d -d options to show more CPU events X-Git-Url: https://git.stricted.de/?a=commitdiff_plain;h=2cba3ffb9a9d;p=GitHub%2FLineageOS%2FG12%2Fandroid_kernel_amlogic_linux-4.9.git perf stat: Add -d -d and -d -d -d options to show more CPU events Print even more detailed statistics if requested via perf stat -d: -d: detailed events, L1 and LLC data cache -d -d: more detailed events, dTLB and iTLB events -d -d -d: very detailed events, adding prefetch events Full output looks like this now: Performance counter stats for '/home/mingo/hackbench 10' (5 runs): 1703.674707 task-clock # 8.709 CPUs utilized ( +- 4.19% ) 49,068 context-switches # 0.029 M/sec ( +- 16.66% ) 8,303 CPU-migrations # 0.005 M/sec ( +- 24.90% ) 17,397 page-faults # 0.010 M/sec ( +- 0.46% ) 2,345,389,239 cycles # 1.377 GHz ( +- 4.61% ) [55.90%] 1,884,503,527 stalled-cycles-frontend # 80.35% frontend cycles idle ( +- 5.67% ) [50.39%] 743,919,737 stalled-cycles-backend # 31.72% backend cycles idle ( +- 8.75% ) [49.91%] 1,314,416,379 instructions # 0.56 insns per cycle # 1.43 stalled cycles per insn ( +- 2.53% ) [60.87%] 272,592,567 branches # 160.003 M/sec ( +- 1.74% ) [56.56%] 3,794,846 branch-misses # 1.39% of all branches ( +- 6.59% ) [58.50%] 449,982,778 L1-dcache-loads # 264.125 M/sec ( +- 2.47% ) [49.88%] 22,404,961 L1-dcache-load-misses # 4.98% of all L1-dcache hits ( +- 6.08% ) [55.05%] 6,204,750 LLC-loads # 3.642 M/sec ( +- 8.91% ) [43.75%] 1,837,411 LLC-load-misses # 1.078 M/sec ( +- 7.27% ) [12.07%] 411,440,421 L1-icache-loads # 241.502 M/sec ( +- 5.60% ) [36.52%] 27,556,832 L1-icache-load-misses # 16.175 M/sec ( +- 7.46% ) [46.72%] 464,067,627 dTLB-loads # 272.392 M/sec ( +- 4.46% ) [54.17%] 10,765,648 dTLB-load-misses # 6.319 M/sec ( +- 3.18% ) [48.68%] 1,273,080,386 iTLB-loads # 747.256 M/sec ( +- 3.38% ) [47.53%] 117,481 iTLB-load-misses # 0.069 M/sec ( +- 14.99% ) [47.01%] 4,590,653 L1-dcache-prefetches # 2.695 M/sec ( +- 4.49% ) [46.19%] 1,712,660 L1-dcache-prefetch-misses # 1.005 M/sec ( +- 3.75% ) [44.82%] 0.195622057 seconds time elapsed ( +- 6.84% ) Also clean up the attribute construction code to be appending, and factor it out into add_default_attributes(). Tweak the coverage percentage printout a bit, so that it's easier to view it alongside the +- sttddev colum. Cc: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Frederic Weisbecker Cc: Mike Galbraith Cc: Steven Rostedt Link: http://lkml.kernel.org/n/tip-to3kgu04449s64062val8b62@git.kernel.org Signed-off-by: Ingo Molnar --- diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 602c3c96fa1e..a89fc0835367 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -6,24 +6,28 @@ * * Sample output: - $ perf stat ~/hackbench 10 - Time: 0.104 + $ perf stat ./hackbench 10 - Performance counter stats for '/home/mingo/hackbench': + Time: 0.118 - 1255.538611 task clock ticks # 10.143 CPU utilization factor - 54011 context switches # 0.043 M/sec - 385 CPU migrations # 0.000 M/sec - 17755 pagefaults # 0.014 M/sec - 3808323185 CPU cycles # 3033.219 M/sec - 1575111190 instructions # 1254.530 M/sec - 17367895 cache references # 13.833 M/sec - 7674421 cache misses # 6.112 M/sec + Performance counter stats for './hackbench 10': - Wall-clock time elapsed: 123.786620 msecs + 1708.761321 task-clock # 11.037 CPUs utilized + 41,190 context-switches # 0.024 M/sec + 6,735 CPU-migrations # 0.004 M/sec + 17,318 page-faults # 0.010 M/sec + 5,205,202,243 cycles # 3.046 GHz + 3,856,436,920 stalled-cycles-frontend # 74.09% frontend cycles idle + 1,600,790,871 stalled-cycles-backend # 30.75% backend cycles idle + 2,603,501,247 instructions # 0.50 insns per cycle + # 1.48 stalled cycles per insn + 484,357,498 branches # 283.455 M/sec + 6,388,934 branch-misses # 1.32% of all branches + + 0.154822978 seconds time elapsed * - * Copyright (C) 2008, Red Hat Inc, Ingo Molnar + * Copyright (C) 2008-2011, Red Hat Inc, Ingo Molnar * * Improvements and fixes by: * @@ -75,22 +79,10 @@ static struct perf_event_attr default_attrs[] = { }; /* - * Detailed stats: + * Detailed stats (-d), covering the L1 and last level data caches: */ static struct perf_event_attr detailed_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_STALLED_CYCLES_FRONTEND }, - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_STALLED_CYCLES_BACKEND }, - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS }, - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_INSTRUCTIONS }, - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_MISSES }, - { .type = PERF_TYPE_HW_CACHE, .config = PERF_COUNT_HW_CACHE_L1D << 0 | @@ -116,6 +108,69 @@ static struct perf_event_attr detailed_attrs[] = { (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) }, }; +/* + * Very detailed stats (-d -d), covering the instruction cache and the TLB caches: + */ +static struct perf_event_attr very_detailed_attrs[] = { + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_L1I << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_L1I << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_DTLB << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_DTLB << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_ITLB << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_ITLB << 0 | + (PERF_COUNT_HW_CACHE_OP_READ << 8) | + (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) }, + +}; + +/* + * Very, very detailed stats (-d -d -d), adding prefetch events: + */ +static struct perf_event_attr very_very_detailed_attrs[] = { + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_L1D << 0 | + (PERF_COUNT_HW_CACHE_OP_PREFETCH << 8) | + (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) }, + + { .type = PERF_TYPE_HW_CACHE, + .config = + PERF_COUNT_HW_CACHE_L1D << 0 | + (PERF_COUNT_HW_CACHE_OP_PREFETCH << 8) | + (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) }, +}; + + + struct perf_evlist *evsel_list; static bool system_wide = false; @@ -129,7 +184,7 @@ static pid_t target_pid = -1; static pid_t target_tid = -1; static pid_t child_pid = -1; static bool null_run = false; -static bool detailed_run = false; +static int detailed_run = 0; static bool sync_run = false; static bool big_num = true; static int big_num_opt = -1; @@ -464,7 +519,7 @@ static void nsec_printout(int cpu, struct perf_evsel *evsel, double avg) { double msecs = avg / 1e6; char cpustr[16] = { '\0', }; - const char *fmt = csv_output ? "%s%.6f%s%s" : "%s%18.6f%s%-24s"; + const char *fmt = csv_output ? "%s%.6f%s%s" : "%s%18.6f%s%-25s"; if (no_aggr) sprintf(cpustr, "CPU%*d%s", @@ -584,9 +639,9 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg) if (csv_output) fmt = "%s%.0f%s%s"; else if (big_num) - fmt = "%s%'18.0f%s%-24s"; + fmt = "%s%'18.0f%s%-25s"; else - fmt = "%s%18.0f%s%-24s"; + fmt = "%s%18.0f%s%-25s"; if (no_aggr) sprintf(cpustr, "CPU%*d%s", @@ -616,7 +671,7 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg) if (total && avg) { ratio = total / avg; - fprintf(stderr, "\n # %5.2f stalled cycles per insn", ratio); + fprintf(stderr, "\n # %5.2f stalled cycles per insn", ratio); } } else if (perf_evsel__match(evsel, HARDWARE, HW_BRANCH_MISSES) && @@ -704,7 +759,7 @@ static void print_counter_aggr(struct perf_evsel *counter) avg_enabled = avg_stats(&ps->res_stats[1]); avg_running = avg_stats(&ps->res_stats[2]); - fprintf(stderr, " (%.2f%%)", 100 * avg_running / avg_enabled); + fprintf(stderr, " [%5.2f%%]", 100 * avg_running / avg_enabled); } fprintf(stderr, "\n"); } @@ -854,7 +909,7 @@ static const struct option options[] = { "repeat command and print average + stddev (max: 100)"), OPT_BOOLEAN('n', "null", &null_run, "null run - dont start any counters"), - OPT_BOOLEAN('d', "detailed", &detailed_run, + OPT_INCR('d', "detailed", &detailed_run, "detailed run - start a lot of events"), OPT_BOOLEAN('S', "sync", &sync_run, "call sync() before starting a run"), @@ -873,6 +928,70 @@ static const struct option options[] = { OPT_END() }; +/* + * Add default attributes, if there were no attributes specified or + * if -d/--detailed, -d -d or -d -d -d is used: + */ +static int add_default_attributes(void) +{ + struct perf_evsel *pos; + size_t attr_nr = 0; + size_t c; + + /* Set attrs if no event is selected and !null_run: */ + if (null_run) + return 0; + + if (!evsel_list->nr_entries) { + for (c = 0; c < ARRAY_SIZE(default_attrs); c++) { + pos = perf_evsel__new(default_attrs + c, c + attr_nr); + if (pos == NULL) + return -1; + perf_evlist__add(evsel_list, pos); + } + attr_nr += c; + } + + /* Detailed events get appended to the event list: */ + + if (detailed_run < 1) + return 0; + + /* Append detailed run extra attributes: */ + for (c = 0; c < ARRAY_SIZE(detailed_attrs); c++) { + pos = perf_evsel__new(detailed_attrs + c, c + attr_nr); + if (pos == NULL) + return -1; + perf_evlist__add(evsel_list, pos); + } + attr_nr += c; + + if (detailed_run < 2) + return 0; + + /* Append very detailed run extra attributes: */ + for (c = 0; c < ARRAY_SIZE(very_detailed_attrs); c++) { + pos = perf_evsel__new(very_detailed_attrs + c, c + attr_nr); + if (pos == NULL) + return -1; + perf_evlist__add(evsel_list, pos); + } + + if (detailed_run < 3) + return 0; + + /* Append very, very detailed run extra attributes: */ + for (c = 0; c < ARRAY_SIZE(very_very_detailed_attrs); c++) { + pos = perf_evsel__new(very_very_detailed_attrs + c, c + attr_nr); + if (pos == NULL) + return -1; + perf_evlist__add(evsel_list, pos); + } + + + return 0; +} + int cmd_stat(int argc, const char **argv, const char *prefix __used) { struct perf_evsel *pos; @@ -918,28 +1037,8 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used) usage_with_options(stat_usage, options); } - /* Set attrs and nr_counters if no event is selected and !null_run */ - if (detailed_run) { - size_t c; - - for (c = 0; c < ARRAY_SIZE(detailed_attrs); ++c) { - pos = perf_evsel__new(&detailed_attrs[c], c); - if (pos == NULL) - goto out; - perf_evlist__add(evsel_list, pos); - } - } - /* Set attrs and nr_counters if no event is selected and !null_run */ - if (!detailed_run && !null_run && !evsel_list->nr_entries) { - size_t c; - - for (c = 0; c < ARRAY_SIZE(default_attrs); ++c) { - pos = perf_evsel__new(&default_attrs[c], c); - if (pos == NULL) - goto out; - perf_evlist__add(evsel_list, pos); - } - } + if (add_default_attributes()) + goto out; if (target_pid != -1) target_tid = target_pid;