* [PATCH] perf: add -o and --append options to perf stat
@ 2011-08-15 20:22 Stephane Eranian
2011-08-15 21:32 ` David Ahern
2011-08-18 20:14 ` [tip:perf/core] perf stat: Add -o and --append options tip-bot for Stephane Eranian
0 siblings, 2 replies; 4+ messages in thread
From: Stephane Eranian @ 2011-08-15 20:22 UTC (permalink / raw)
To: linux-kernel; +Cc: acme, mingo, peterz
This patch adds an option (-o) to save the output of perf stat
into a file. You could do this with perf record but not with
perf stat. Instead, you had to fiddle with stderr to save
the counts into a separate file.
The patch also adds the --append option so that results can
be concatenated into a single file across runs. Each run of
the tool is clearly separated by a comment line starting with
a hash mark. The -A option of perf record is already used by
perf stat, so we only add a long option.
$ perf stat -o res.txt date
$ cat res.txt
# started on Mon Aug 15 19:46:55 2011
Performance counter stats for 'date':
0.791306 task-clock # 0.668 CPUs utilized
2 context-switches # 0.003 M/sec
0 CPU-migrations # 0.000 M/sec
197 page-faults # 0.249 M/sec
1878143 cycles # 2.373 GHz
<not supported> stalled-cycles-frontend
<not supported> stalled-cycles-backend
1083367 instructions # 0.58 insns per cycle
193027 branches # 243.935 M/sec
9014 branch-misses # 4.67% of all branches
0.001184746 seconds time elapsed
The option can be combined with -x to make the output file
much easier to parse.
Signed-off-by: Stephane Eranian <eranian@google.com>
---
diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
index 918cc38..08394c4 100644
--- a/tools/perf/Documentation/perf-stat.txt
+++ b/tools/perf/Documentation/perf-stat.txt
@@ -94,6 +94,13 @@ an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must ha
corresponding events, i.e., they always refer to events defined earlier on the command
line.
+-o file::
+-output file::
+Print the output into the designated file.
+
+--append::
+Append to the output file designated with the -o option. Ignored if -o is not specified.
+
EXAMPLES
--------
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 1ad04ce..a22393d 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -193,6 +193,8 @@ static int big_num_opt = -1;
static const char *cpu_list;
static const char *csv_sep = NULL;
static bool csv_output = false;
+static const char *output_name = NULL;
+static FILE *output = NULL;
static volatile int done = 0;
@@ -351,7 +353,7 @@ static int read_counter_aggr(struct perf_evsel *counter)
update_stats(&ps->res_stats[i], count[i]);
if (verbose) {
- fprintf(stderr, "%s: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n",
+ fprintf(output, "%s: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n",
event_name(counter), count[0], count[1], count[2]);
}
@@ -518,9 +520,9 @@ static void print_noise_pct(double total, double avg)
pct = 100.0*total/avg;
if (csv_output)
- fprintf(stderr, "%s%.2f%%", csv_sep, pct);
+ fprintf(output, "%s%.2f%%", csv_sep, pct);
else
- fprintf(stderr, " ( +-%6.2f%% )", pct);
+ fprintf(output, " ( +-%6.2f%% )", pct);
}
static void print_noise(struct perf_evsel *evsel, double avg)
@@ -545,16 +547,17 @@ static void nsec_printout(int cpu, struct perf_evsel *evsel, double avg)
csv_output ? 0 : -4,
evsel_list->cpus->map[cpu], csv_sep);
- fprintf(stderr, fmt, cpustr, msecs, csv_sep, event_name(evsel));
+ fprintf(output, fmt, cpustr, msecs, csv_sep, event_name(evsel));
if (evsel->cgrp)
- fprintf(stderr, "%s%s", csv_sep, evsel->cgrp->name);
+ fprintf(output, "%s%s", csv_sep, evsel->cgrp->name);
if (csv_output)
return;
if (perf_evsel__match(evsel, SOFTWARE, SW_TASK_CLOCK))
- fprintf(stderr, " # %8.3f CPUs utilized ", avg / avg_stats(&walltime_nsecs_stats));
+ fprintf(output, " # %8.3f CPUs utilized ",
+ avg / avg_stats(&walltime_nsecs_stats));
}
static void print_stalled_cycles_frontend(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -575,9 +578,9 @@ static void print_stalled_cycles_frontend(int cpu, struct perf_evsel *evsel __us
else if (ratio > 10.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " frontend cycles idle ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " frontend cycles idle ");
}
static void print_stalled_cycles_backend(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -598,9 +601,9 @@ static void print_stalled_cycles_backend(int cpu, struct perf_evsel *evsel __use
else if (ratio > 20.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " backend cycles idle ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " backend cycles idle ");
}
static void print_branch_misses(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -621,9 +624,9 @@ static void print_branch_misses(int cpu, struct perf_evsel *evsel __used, double
else if (ratio > 5.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " of all branches ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all branches ");
}
static void print_l1_dcache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -644,9 +647,9 @@ static void print_l1_dcache_misses(int cpu, struct perf_evsel *evsel __used, dou
else if (ratio > 5.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " of all L1-dcache hits ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all L1-dcache hits ");
}
static void print_l1_icache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -667,9 +670,9 @@ static void print_l1_icache_misses(int cpu, struct perf_evsel *evsel __used, dou
else if (ratio > 5.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " of all L1-icache hits ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all L1-icache hits ");
}
static void print_dtlb_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -690,9 +693,9 @@ static void print_dtlb_cache_misses(int cpu, struct perf_evsel *evsel __used, do
else if (ratio > 5.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " of all dTLB cache hits ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all dTLB cache hits ");
}
static void print_itlb_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -713,9 +716,9 @@ static void print_itlb_cache_misses(int cpu, struct perf_evsel *evsel __used, do
else if (ratio > 5.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " of all iTLB cache hits ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all iTLB cache hits ");
}
static void print_ll_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -736,9 +739,9 @@ static void print_ll_cache_misses(int cpu, struct perf_evsel *evsel __used, doub
else if (ratio > 5.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " of all LL-cache hits ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all LL-cache hits ");
}
static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
@@ -761,10 +764,10 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
else
cpu = 0;
- fprintf(stderr, fmt, cpustr, avg, csv_sep, event_name(evsel));
+ fprintf(output, fmt, cpustr, avg, csv_sep, event_name(evsel));
if (evsel->cgrp)
- fprintf(stderr, "%s%s", csv_sep, evsel->cgrp->name);
+ fprintf(output, "%s%s", csv_sep, evsel->cgrp->name);
if (csv_output)
return;
@@ -775,14 +778,14 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
if (total)
ratio = avg / total;
- fprintf(stderr, " # %5.2f insns per cycle ", ratio);
+ fprintf(output, " # %5.2f insns per cycle ", ratio);
total = avg_stats(&runtime_stalled_cycles_front_stats[cpu]);
total = max(total, avg_stats(&runtime_stalled_cycles_back_stats[cpu]));
if (total && avg) {
ratio = total / avg;
- fprintf(stderr, "\n # %5.2f stalled cycles per insn", ratio);
+ fprintf(output, "\n # %5.2f stalled cycles per insn", ratio);
}
} else if (perf_evsel__match(evsel, HARDWARE, HW_BRANCH_MISSES) &&
@@ -830,7 +833,7 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
if (total)
ratio = avg * 100 / total;
- fprintf(stderr, " # %8.3f %% of all cache refs ", ratio);
+ fprintf(output, " # %8.3f %% of all cache refs ", ratio);
} else if (perf_evsel__match(evsel, HARDWARE, HW_STALLED_CYCLES_FRONTEND)) {
print_stalled_cycles_frontend(cpu, evsel, avg);
@@ -842,16 +845,16 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
if (total)
ratio = 1.0 * avg / total;
- fprintf(stderr, " # %8.3f GHz ", ratio);
+ fprintf(output, " # %8.3f GHz ", ratio);
} else if (runtime_nsecs_stats[cpu].n != 0) {
total = avg_stats(&runtime_nsecs_stats[cpu]);
if (total)
ratio = 1000.0 * avg / total;
- fprintf(stderr, " # %8.3f M/sec ", ratio);
+ fprintf(output, " # %8.3f M/sec ", ratio);
} else {
- fprintf(stderr, " ");
+ fprintf(output, " ");
}
}
@@ -866,7 +869,7 @@ static void print_counter_aggr(struct perf_evsel *counter)
int scaled = counter->counts->scaled;
if (scaled == -1) {
- fprintf(stderr, "%*s%s%*s",
+ fprintf(output, "%*s%s%*s",
csv_output ? 0 : 18,
counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED,
csv_sep,
@@ -874,9 +877,9 @@ static void print_counter_aggr(struct perf_evsel *counter)
event_name(counter));
if (counter->cgrp)
- fprintf(stderr, "%s%s", csv_sep, counter->cgrp->name);
+ fprintf(output, "%s%s", csv_sep, counter->cgrp->name);
- fputc('\n', stderr);
+ fputc('\n', output);
return;
}
@@ -888,7 +891,7 @@ static void print_counter_aggr(struct perf_evsel *counter)
print_noise(counter, avg);
if (csv_output) {
- fputc('\n', stderr);
+ fputc('\n', output);
return;
}
@@ -898,9 +901,9 @@ 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, " [%5.2f%%]", 100 * avg_running / avg_enabled);
+ fprintf(output, " [%5.2f%%]", 100 * avg_running / avg_enabled);
}
- fprintf(stderr, "\n");
+ fprintf(output, "\n");
}
/*
@@ -917,7 +920,7 @@ static void print_counter(struct perf_evsel *counter)
ena = counter->counts->cpu[cpu].ena;
run = counter->counts->cpu[cpu].run;
if (run == 0 || ena == 0) {
- fprintf(stderr, "CPU%*d%s%*s%s%*s",
+ fprintf(output, "CPU%*d%s%*s%s%*s",
csv_output ? 0 : -4,
evsel_list->cpus->map[cpu], csv_sep,
csv_output ? 0 : 18,
@@ -927,9 +930,10 @@ static void print_counter(struct perf_evsel *counter)
event_name(counter));
if (counter->cgrp)
- fprintf(stderr, "%s%s", csv_sep, counter->cgrp->name);
+ fprintf(output, "%s%s",
+ csv_sep, counter->cgrp->name);
- fputc('\n', stderr);
+ fputc('\n', output);
continue;
}
@@ -942,9 +946,10 @@ static void print_counter(struct perf_evsel *counter)
print_noise(counter, 1.0);
if (run != ena)
- fprintf(stderr, " (%.2f%%)", 100.0 * run / ena);
+ fprintf(output, " (%.2f%%)",
+ 100.0 * run / ena);
}
- fputc('\n', stderr);
+ fputc('\n', output);
}
}
@@ -956,21 +961,21 @@ static void print_stat(int argc, const char **argv)
fflush(stdout);
if (!csv_output) {
- fprintf(stderr, "\n");
- fprintf(stderr, " Performance counter stats for ");
+ fprintf(output, "\n");
+ fprintf(output, " Performance counter stats for ");
if(target_pid == -1 && target_tid == -1) {
- fprintf(stderr, "\'%s", argv[0]);
+ fprintf(output, "\'%s", argv[0]);
for (i = 1; i < argc; i++)
- fprintf(stderr, " %s", argv[i]);
+ fprintf(output, " %s", argv[i]);
} else if (target_pid != -1)
- fprintf(stderr, "process id \'%d", target_pid);
+ fprintf(output, "process id \'%d", target_pid);
else
- fprintf(stderr, "thread id \'%d", target_tid);
+ fprintf(output, "thread id \'%d", target_tid);
- fprintf(stderr, "\'");
+ fprintf(output, "\'");
if (run_count > 1)
- fprintf(stderr, " (%d runs)", run_count);
- fprintf(stderr, ":\n\n");
+ fprintf(output, " (%d runs)", run_count);
+ fprintf(output, ":\n\n");
}
if (no_aggr) {
@@ -983,15 +988,15 @@ static void print_stat(int argc, const char **argv)
if (!csv_output) {
if (!null_run)
- fprintf(stderr, "\n");
- fprintf(stderr, " %17.9f seconds time elapsed",
+ fprintf(output, "\n");
+ fprintf(output, " %17.9f seconds time elapsed",
avg_stats(&walltime_nsecs_stats)/1e9);
if (run_count > 1) {
- fprintf(stderr, " ");
+ fprintf(output, " ");
print_noise_pct(stddev_stats(&walltime_nsecs_stats),
avg_stats(&walltime_nsecs_stats));
}
- fprintf(stderr, "\n\n");
+ fprintf(output, "\n\n");
}
}
@@ -1029,6 +1034,8 @@ static int stat__set_big_num(const struct option *opt __used,
return 0;
}
+static bool append_file;
+
static const struct option options[] = {
OPT_CALLBACK('e', "event", &evsel_list, "event",
"event selector. use 'perf list' to list available events",
@@ -1067,6 +1074,9 @@ static const struct option options[] = {
OPT_CALLBACK('G', "cgroup", &evsel_list, "name",
"monitor event in cgroup name only",
parse_cgroups),
+ OPT_STRING('o', "output", &output_name, "file",
+ "output file name"),
+ OPT_BOOLEAN(0, "append", &append_file, "append to the output file"),
OPT_END()
};
@@ -1138,6 +1148,7 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
{
struct perf_evsel *pos;
int status = -ENOMEM;
+ const char *mode;
setlocale(LC_ALL, "");
@@ -1148,6 +1159,23 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
argc = parse_options(argc, argv, options, stat_usage,
PARSE_OPT_STOP_AT_NON_OPTION);
+ output = stderr;
+ if (output_name && strcmp(output_name, "-"))
+ output = NULL;
+
+ if (!output) {
+ struct timespec tm;
+ mode = append_file ? "a" : "w";
+
+ output = fopen(output_name, mode);
+ if (!output) {
+ perror("failed to create output file");
+ exit(-1);
+ }
+ clock_gettime(CLOCK_REALTIME, &tm);
+ fprintf(output, "# started on %s\n", ctime(&tm.tv_sec));
+ }
+
if (csv_sep)
csv_output = true;
else
@@ -1223,7 +1251,8 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
status = 0;
for (run_idx = 0; run_idx < run_count; run_idx++) {
if (run_count != 1 && verbose)
- fprintf(stderr, "[ perf stat: executing run #%d ... ]\n", run_idx + 1);
+ fprintf(output, "[ perf stat: executing run #%d ... ]\n",
+ run_idx + 1);
if (sync_run)
sync();
diff --git a/tools/perf/util/color.c b/tools/perf/util/color.c
index e191eb9..521c38a 100644
--- a/tools/perf/util/color.c
+++ b/tools/perf/util/color.c
@@ -200,7 +200,7 @@ static int __color_vfprintf(FILE *fp, const char *color, const char *fmt,
* Auto-detect:
*/
if (perf_use_color_default < 0) {
- if (isatty(1) || pager_in_use())
+ if (isatty(fileno(fp)) || pager_in_use())
perf_use_color_default = 1;
else
perf_use_color_default = 0;
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH] perf: add -o and --append options to perf stat
2011-08-15 20:22 [PATCH] perf: add -o and --append options to perf stat Stephane Eranian
@ 2011-08-15 21:32 ` David Ahern
2011-08-15 21:33 ` Stephane Eranian
2011-08-18 20:14 ` [tip:perf/core] perf stat: Add -o and --append options tip-bot for Stephane Eranian
1 sibling, 1 reply; 4+ messages in thread
From: David Ahern @ 2011-08-15 21:32 UTC (permalink / raw)
To: Stephane Eranian; +Cc: linux-kernel, acme, mingo, peterz
On 08/15/2011 02:22 PM, Stephane Eranian wrote:
>
> This patch adds an option (-o) to save the output of perf stat
> into a file. You could do this with perf record but not with
> perf stat. Instead, you had to fiddle with stderr to save
> the counts into a separate file.
>
> The patch also adds the --append option so that results can
> be concatenated into a single file across runs. Each run of
> the tool is clearly separated by a comment line starting with
> a hash mark. The -A option of perf record is already used by
> perf stat, so we only add a long option.
>
> $ perf stat -o res.txt date
Why not just use redirection to put the output into a file?
perf stat date > res.txt
David
> $ cat res.txt
> # started on Mon Aug 15 19:46:55 2011
>
>
> Performance counter stats for 'date':
>
> 0.791306 task-clock # 0.668 CPUs utilized
> 2 context-switches # 0.003 M/sec
> 0 CPU-migrations # 0.000 M/sec
> 197 page-faults # 0.249 M/sec
> 1878143 cycles # 2.373 GHz
> <not supported> stalled-cycles-frontend
> <not supported> stalled-cycles-backend
> 1083367 instructions # 0.58 insns per cycle
> 193027 branches # 243.935 M/sec
> 9014 branch-misses # 4.67% of all branches
>
> 0.001184746 seconds time elapsed
>
> The option can be combined with -x to make the output file
> much easier to parse.
>
> Signed-off-by: Stephane Eranian <eranian@google.com>
> ---
>
> diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
> index 918cc38..08394c4 100644
> --- a/tools/perf/Documentation/perf-stat.txt
> +++ b/tools/perf/Documentation/perf-stat.txt
> @@ -94,6 +94,13 @@ an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must ha
> corresponding events, i.e., they always refer to events defined earlier on the command
> line.
>
> +-o file::
> +-output file::
> +Print the output into the designated file.
> +
> +--append::
> +Append to the output file designated with the -o option. Ignored if -o is not specified.
> +
> EXAMPLES
> --------
>
> diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
> index 1ad04ce..a22393d 100644
> --- a/tools/perf/builtin-stat.c
> +++ b/tools/perf/builtin-stat.c
> @@ -193,6 +193,8 @@ static int big_num_opt = -1;
> static const char *cpu_list;
> static const char *csv_sep = NULL;
> static bool csv_output = false;
> +static const char *output_name = NULL;
> +static FILE *output = NULL;
>
> static volatile int done = 0;
>
> @@ -351,7 +353,7 @@ static int read_counter_aggr(struct perf_evsel *counter)
> update_stats(&ps->res_stats[i], count[i]);
>
> if (verbose) {
> - fprintf(stderr, "%s: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n",
> + fprintf(output, "%s: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n",
> event_name(counter), count[0], count[1], count[2]);
> }
>
> @@ -518,9 +520,9 @@ static void print_noise_pct(double total, double avg)
> pct = 100.0*total/avg;
>
> if (csv_output)
> - fprintf(stderr, "%s%.2f%%", csv_sep, pct);
> + fprintf(output, "%s%.2f%%", csv_sep, pct);
> else
> - fprintf(stderr, " ( +-%6.2f%% )", pct);
> + fprintf(output, " ( +-%6.2f%% )", pct);
> }
>
> static void print_noise(struct perf_evsel *evsel, double avg)
> @@ -545,16 +547,17 @@ static void nsec_printout(int cpu, struct perf_evsel *evsel, double avg)
> csv_output ? 0 : -4,
> evsel_list->cpus->map[cpu], csv_sep);
>
> - fprintf(stderr, fmt, cpustr, msecs, csv_sep, event_name(evsel));
> + fprintf(output, fmt, cpustr, msecs, csv_sep, event_name(evsel));
>
> if (evsel->cgrp)
> - fprintf(stderr, "%s%s", csv_sep, evsel->cgrp->name);
> + fprintf(output, "%s%s", csv_sep, evsel->cgrp->name);
>
> if (csv_output)
> return;
>
> if (perf_evsel__match(evsel, SOFTWARE, SW_TASK_CLOCK))
> - fprintf(stderr, " # %8.3f CPUs utilized ", avg / avg_stats(&walltime_nsecs_stats));
> + fprintf(output, " # %8.3f CPUs utilized ",
> + avg / avg_stats(&walltime_nsecs_stats));
> }
>
> static void print_stalled_cycles_frontend(int cpu, struct perf_evsel *evsel __used, double avg)
> @@ -575,9 +578,9 @@ static void print_stalled_cycles_frontend(int cpu, struct perf_evsel *evsel __us
> else if (ratio > 10.0)
> color = PERF_COLOR_YELLOW;
>
> - fprintf(stderr, " # ");
> - color_fprintf(stderr, color, "%6.2f%%", ratio);
> - fprintf(stderr, " frontend cycles idle ");
> + fprintf(output, " # ");
> + color_fprintf(output, color, "%6.2f%%", ratio);
> + fprintf(output, " frontend cycles idle ");
> }
>
> static void print_stalled_cycles_backend(int cpu, struct perf_evsel *evsel __used, double avg)
> @@ -598,9 +601,9 @@ static void print_stalled_cycles_backend(int cpu, struct perf_evsel *evsel __use
> else if (ratio > 20.0)
> color = PERF_COLOR_YELLOW;
>
> - fprintf(stderr, " # ");
> - color_fprintf(stderr, color, "%6.2f%%", ratio);
> - fprintf(stderr, " backend cycles idle ");
> + fprintf(output, " # ");
> + color_fprintf(output, color, "%6.2f%%", ratio);
> + fprintf(output, " backend cycles idle ");
> }
>
> static void print_branch_misses(int cpu, struct perf_evsel *evsel __used, double avg)
> @@ -621,9 +624,9 @@ static void print_branch_misses(int cpu, struct perf_evsel *evsel __used, double
> else if (ratio > 5.0)
> color = PERF_COLOR_YELLOW;
>
> - fprintf(stderr, " # ");
> - color_fprintf(stderr, color, "%6.2f%%", ratio);
> - fprintf(stderr, " of all branches ");
> + fprintf(output, " # ");
> + color_fprintf(output, color, "%6.2f%%", ratio);
> + fprintf(output, " of all branches ");
> }
>
> static void print_l1_dcache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
> @@ -644,9 +647,9 @@ static void print_l1_dcache_misses(int cpu, struct perf_evsel *evsel __used, dou
> else if (ratio > 5.0)
> color = PERF_COLOR_YELLOW;
>
> - fprintf(stderr, " # ");
> - color_fprintf(stderr, color, "%6.2f%%", ratio);
> - fprintf(stderr, " of all L1-dcache hits ");
> + fprintf(output, " # ");
> + color_fprintf(output, color, "%6.2f%%", ratio);
> + fprintf(output, " of all L1-dcache hits ");
> }
>
> static void print_l1_icache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
> @@ -667,9 +670,9 @@ static void print_l1_icache_misses(int cpu, struct perf_evsel *evsel __used, dou
> else if (ratio > 5.0)
> color = PERF_COLOR_YELLOW;
>
> - fprintf(stderr, " # ");
> - color_fprintf(stderr, color, "%6.2f%%", ratio);
> - fprintf(stderr, " of all L1-icache hits ");
> + fprintf(output, " # ");
> + color_fprintf(output, color, "%6.2f%%", ratio);
> + fprintf(output, " of all L1-icache hits ");
> }
>
> static void print_dtlb_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
> @@ -690,9 +693,9 @@ static void print_dtlb_cache_misses(int cpu, struct perf_evsel *evsel __used, do
> else if (ratio > 5.0)
> color = PERF_COLOR_YELLOW;
>
> - fprintf(stderr, " # ");
> - color_fprintf(stderr, color, "%6.2f%%", ratio);
> - fprintf(stderr, " of all dTLB cache hits ");
> + fprintf(output, " # ");
> + color_fprintf(output, color, "%6.2f%%", ratio);
> + fprintf(output, " of all dTLB cache hits ");
> }
>
> static void print_itlb_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
> @@ -713,9 +716,9 @@ static void print_itlb_cache_misses(int cpu, struct perf_evsel *evsel __used, do
> else if (ratio > 5.0)
> color = PERF_COLOR_YELLOW;
>
> - fprintf(stderr, " # ");
> - color_fprintf(stderr, color, "%6.2f%%", ratio);
> - fprintf(stderr, " of all iTLB cache hits ");
> + fprintf(output, " # ");
> + color_fprintf(output, color, "%6.2f%%", ratio);
> + fprintf(output, " of all iTLB cache hits ");
> }
>
> static void print_ll_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
> @@ -736,9 +739,9 @@ static void print_ll_cache_misses(int cpu, struct perf_evsel *evsel __used, doub
> else if (ratio > 5.0)
> color = PERF_COLOR_YELLOW;
>
> - fprintf(stderr, " # ");
> - color_fprintf(stderr, color, "%6.2f%%", ratio);
> - fprintf(stderr, " of all LL-cache hits ");
> + fprintf(output, " # ");
> + color_fprintf(output, color, "%6.2f%%", ratio);
> + fprintf(output, " of all LL-cache hits ");
> }
>
> static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
> @@ -761,10 +764,10 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
> else
> cpu = 0;
>
> - fprintf(stderr, fmt, cpustr, avg, csv_sep, event_name(evsel));
> + fprintf(output, fmt, cpustr, avg, csv_sep, event_name(evsel));
>
> if (evsel->cgrp)
> - fprintf(stderr, "%s%s", csv_sep, evsel->cgrp->name);
> + fprintf(output, "%s%s", csv_sep, evsel->cgrp->name);
>
> if (csv_output)
> return;
> @@ -775,14 +778,14 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
> if (total)
> ratio = avg / total;
>
> - fprintf(stderr, " # %5.2f insns per cycle ", ratio);
> + fprintf(output, " # %5.2f insns per cycle ", ratio);
>
> total = avg_stats(&runtime_stalled_cycles_front_stats[cpu]);
> total = max(total, avg_stats(&runtime_stalled_cycles_back_stats[cpu]));
>
> if (total && avg) {
> ratio = total / avg;
> - fprintf(stderr, "\n # %5.2f stalled cycles per insn", ratio);
> + fprintf(output, "\n # %5.2f stalled cycles per insn", ratio);
> }
>
> } else if (perf_evsel__match(evsel, HARDWARE, HW_BRANCH_MISSES) &&
> @@ -830,7 +833,7 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
> if (total)
> ratio = avg * 100 / total;
>
> - fprintf(stderr, " # %8.3f %% of all cache refs ", ratio);
> + fprintf(output, " # %8.3f %% of all cache refs ", ratio);
>
> } else if (perf_evsel__match(evsel, HARDWARE, HW_STALLED_CYCLES_FRONTEND)) {
> print_stalled_cycles_frontend(cpu, evsel, avg);
> @@ -842,16 +845,16 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
> if (total)
> ratio = 1.0 * avg / total;
>
> - fprintf(stderr, " # %8.3f GHz ", ratio);
> + fprintf(output, " # %8.3f GHz ", ratio);
> } else if (runtime_nsecs_stats[cpu].n != 0) {
> total = avg_stats(&runtime_nsecs_stats[cpu]);
>
> if (total)
> ratio = 1000.0 * avg / total;
>
> - fprintf(stderr, " # %8.3f M/sec ", ratio);
> + fprintf(output, " # %8.3f M/sec ", ratio);
> } else {
> - fprintf(stderr, " ");
> + fprintf(output, " ");
> }
> }
>
> @@ -866,7 +869,7 @@ static void print_counter_aggr(struct perf_evsel *counter)
> int scaled = counter->counts->scaled;
>
> if (scaled == -1) {
> - fprintf(stderr, "%*s%s%*s",
> + fprintf(output, "%*s%s%*s",
> csv_output ? 0 : 18,
> counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED,
> csv_sep,
> @@ -874,9 +877,9 @@ static void print_counter_aggr(struct perf_evsel *counter)
> event_name(counter));
>
> if (counter->cgrp)
> - fprintf(stderr, "%s%s", csv_sep, counter->cgrp->name);
> + fprintf(output, "%s%s", csv_sep, counter->cgrp->name);
>
> - fputc('\n', stderr);
> + fputc('\n', output);
> return;
> }
>
> @@ -888,7 +891,7 @@ static void print_counter_aggr(struct perf_evsel *counter)
> print_noise(counter, avg);
>
> if (csv_output) {
> - fputc('\n', stderr);
> + fputc('\n', output);
> return;
> }
>
> @@ -898,9 +901,9 @@ 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, " [%5.2f%%]", 100 * avg_running / avg_enabled);
> + fprintf(output, " [%5.2f%%]", 100 * avg_running / avg_enabled);
> }
> - fprintf(stderr, "\n");
> + fprintf(output, "\n");
> }
>
> /*
> @@ -917,7 +920,7 @@ static void print_counter(struct perf_evsel *counter)
> ena = counter->counts->cpu[cpu].ena;
> run = counter->counts->cpu[cpu].run;
> if (run == 0 || ena == 0) {
> - fprintf(stderr, "CPU%*d%s%*s%s%*s",
> + fprintf(output, "CPU%*d%s%*s%s%*s",
> csv_output ? 0 : -4,
> evsel_list->cpus->map[cpu], csv_sep,
> csv_output ? 0 : 18,
> @@ -927,9 +930,10 @@ static void print_counter(struct perf_evsel *counter)
> event_name(counter));
>
> if (counter->cgrp)
> - fprintf(stderr, "%s%s", csv_sep, counter->cgrp->name);
> + fprintf(output, "%s%s",
> + csv_sep, counter->cgrp->name);
>
> - fputc('\n', stderr);
> + fputc('\n', output);
> continue;
> }
>
> @@ -942,9 +946,10 @@ static void print_counter(struct perf_evsel *counter)
> print_noise(counter, 1.0);
>
> if (run != ena)
> - fprintf(stderr, " (%.2f%%)", 100.0 * run / ena);
> + fprintf(output, " (%.2f%%)",
> + 100.0 * run / ena);
> }
> - fputc('\n', stderr);
> + fputc('\n', output);
> }
> }
>
> @@ -956,21 +961,21 @@ static void print_stat(int argc, const char **argv)
> fflush(stdout);
>
> if (!csv_output) {
> - fprintf(stderr, "\n");
> - fprintf(stderr, " Performance counter stats for ");
> + fprintf(output, "\n");
> + fprintf(output, " Performance counter stats for ");
> if(target_pid == -1 && target_tid == -1) {
> - fprintf(stderr, "\'%s", argv[0]);
> + fprintf(output, "\'%s", argv[0]);
> for (i = 1; i < argc; i++)
> - fprintf(stderr, " %s", argv[i]);
> + fprintf(output, " %s", argv[i]);
> } else if (target_pid != -1)
> - fprintf(stderr, "process id \'%d", target_pid);
> + fprintf(output, "process id \'%d", target_pid);
> else
> - fprintf(stderr, "thread id \'%d", target_tid);
> + fprintf(output, "thread id \'%d", target_tid);
>
> - fprintf(stderr, "\'");
> + fprintf(output, "\'");
> if (run_count > 1)
> - fprintf(stderr, " (%d runs)", run_count);
> - fprintf(stderr, ":\n\n");
> + fprintf(output, " (%d runs)", run_count);
> + fprintf(output, ":\n\n");
> }
>
> if (no_aggr) {
> @@ -983,15 +988,15 @@ static void print_stat(int argc, const char **argv)
>
> if (!csv_output) {
> if (!null_run)
> - fprintf(stderr, "\n");
> - fprintf(stderr, " %17.9f seconds time elapsed",
> + fprintf(output, "\n");
> + fprintf(output, " %17.9f seconds time elapsed",
> avg_stats(&walltime_nsecs_stats)/1e9);
> if (run_count > 1) {
> - fprintf(stderr, " ");
> + fprintf(output, " ");
> print_noise_pct(stddev_stats(&walltime_nsecs_stats),
> avg_stats(&walltime_nsecs_stats));
> }
> - fprintf(stderr, "\n\n");
> + fprintf(output, "\n\n");
> }
> }
>
> @@ -1029,6 +1034,8 @@ static int stat__set_big_num(const struct option *opt __used,
> return 0;
> }
>
> +static bool append_file;
> +
> static const struct option options[] = {
> OPT_CALLBACK('e', "event", &evsel_list, "event",
> "event selector. use 'perf list' to list available events",
> @@ -1067,6 +1074,9 @@ static const struct option options[] = {
> OPT_CALLBACK('G', "cgroup", &evsel_list, "name",
> "monitor event in cgroup name only",
> parse_cgroups),
> + OPT_STRING('o', "output", &output_name, "file",
> + "output file name"),
> + OPT_BOOLEAN(0, "append", &append_file, "append to the output file"),
> OPT_END()
> };
>
> @@ -1138,6 +1148,7 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
> {
> struct perf_evsel *pos;
> int status = -ENOMEM;
> + const char *mode;
>
> setlocale(LC_ALL, "");
>
> @@ -1148,6 +1159,23 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
> argc = parse_options(argc, argv, options, stat_usage,
> PARSE_OPT_STOP_AT_NON_OPTION);
>
> + output = stderr;
> + if (output_name && strcmp(output_name, "-"))
> + output = NULL;
> +
> + if (!output) {
> + struct timespec tm;
> + mode = append_file ? "a" : "w";
> +
> + output = fopen(output_name, mode);
> + if (!output) {
> + perror("failed to create output file");
> + exit(-1);
> + }
> + clock_gettime(CLOCK_REALTIME, &tm);
> + fprintf(output, "# started on %s\n", ctime(&tm.tv_sec));
> + }
> +
> if (csv_sep)
> csv_output = true;
> else
> @@ -1223,7 +1251,8 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
> status = 0;
> for (run_idx = 0; run_idx < run_count; run_idx++) {
> if (run_count != 1 && verbose)
> - fprintf(stderr, "[ perf stat: executing run #%d ... ]\n", run_idx + 1);
> + fprintf(output, "[ perf stat: executing run #%d ... ]\n",
> + run_idx + 1);
>
> if (sync_run)
> sync();
> diff --git a/tools/perf/util/color.c b/tools/perf/util/color.c
> index e191eb9..521c38a 100644
> --- a/tools/perf/util/color.c
> +++ b/tools/perf/util/color.c
> @@ -200,7 +200,7 @@ static int __color_vfprintf(FILE *fp, const char *color, const char *fmt,
> * Auto-detect:
> */
> if (perf_use_color_default < 0) {
> - if (isatty(1) || pager_in_use())
> + if (isatty(fileno(fp)) || pager_in_use())
> perf_use_color_default = 1;
> else
> perf_use_color_default = 0;
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] perf: add -o and --append options to perf stat
2011-08-15 21:32 ` David Ahern
@ 2011-08-15 21:33 ` Stephane Eranian
0 siblings, 0 replies; 4+ messages in thread
From: Stephane Eranian @ 2011-08-15 21:33 UTC (permalink / raw)
To: David Ahern; +Cc: linux-kernel, acme, mingo, peterz
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=UTF-8, Size: 23588 bytes --]
On Mon, Aug 15, 2011 at 11:32 PM, David Ahern <dsahern@gmail.com> wrote:
> On 08/15/2011 02:22 PM, Stephane Eranian wrote:
>>
>> This patch adds an option (-o) to save the output of perf stat
>> into a file. You could do this with perf record but not with
>> perf stat. Instead, you had to fiddle with stderr to save
>> the counts into a separate file.
>>
>> The patch also adds the --append option so that results can
>> be concatenated into a single file across runs. Each run of
>> the tool is clearly separated by a comment line starting with
>> a hash mark. The -A option of perf record is already used by
>> perf stat, so we only add a long option.
>>
>> $ perf stat -o res.txt date
>
> Why not just use redirection to put the output into a file?
> perf stat date > res.txt
>
Because results are not printed on stdout but stderr, so it gets messy
to sort this out, especially if the program you're measuring itself spits
out data on stderr....
> David
>
>> $ cat res.txt
>> # started on Mon Aug 15 19:46:55 2011
>>
>>
>> Â Performance counter stats for 'date':
>>
>>      0.791306 task-clock         #   0.668 CPUs utilized
>>          2 context-switches      #   0.003 M/sec
>>          0 CPU-migrations       #   0.000 M/sec
>>         197 page-faults        #   0.249 M/sec
>>       1878143 cycles           #   2.373 GHz
>> Â Â <not supported> stalled-cycles-frontend
>> Â Â <not supported> stalled-cycles-backend
>>       1083367 instructions        #   0.58  insns per cycle
>>       193027 branches          #  243.935 M/sec
>>        9014 branch-misses       #   4.67% of all branches
>>
>> Â Â Â Â 0.001184746 seconds time elapsed
>>
>> The option can be combined with -x to make the output file
>> much easier to parse.
>>
>> Signed-off-by: Stephane Eranian <eranian@google.com>
>> ---
>>
>> diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
>> index 918cc38..08394c4 100644
>> --- a/tools/perf/Documentation/perf-stat.txt
>> +++ b/tools/perf/Documentation/perf-stat.txt
>> @@ -94,6 +94,13 @@ an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must ha
>> Â corresponding events, i.e., they always refer to events defined earlier on the command
>> Â line.
>>
>> +-o file::
>> +-output file::
>> +Print the output into the designated file.
>> +
>> +--append::
>> +Append to the output file designated with the -o option. Ignored if -o is not specified.
>> +
>> Â EXAMPLES
>> Â --------
>>
>> diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
>> index 1ad04ce..a22393d 100644
>> --- a/tools/perf/builtin-stat.c
>> +++ b/tools/perf/builtin-stat.c
>> @@ -193,6 +193,8 @@ static int             big_num_opt           =  -1;
>>  static const char       *cpu_list;
>>  static const char       *csv_sep             = NULL;
>>  static bool          csv_output            = false;
>> +static const char       *output_name           = NULL;
>> +static FILE          *output             = NULL;
>>
>> Â static volatile int done = 0;
>>
>> @@ -351,7 +353,7 @@ static int read_counter_aggr(struct perf_evsel *counter)
>> Â Â Â Â Â Â Â update_stats(&ps->res_stats[i], count[i]);
>>
>> Â Â Â if (verbose) {
>> - Â Â Â Â Â Â fprintf(stderr, "%s: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n",
>> + Â Â Â Â Â Â fprintf(output, "%s: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n",
>> Â Â Â Â Â Â Â Â Â Â Â event_name(counter), count[0], count[1], count[2]);
>> Â Â Â }
>>
>> @@ -518,9 +520,9 @@ static void print_noise_pct(double total, double avg)
>> Â Â Â Â Â Â Â pct = 100.0*total/avg;
>>
>> Â Â Â if (csv_output)
>> - Â Â Â Â Â Â fprintf(stderr, "%s%.2f%%", csv_sep, pct);
>> + Â Â Â Â Â Â fprintf(output, "%s%.2f%%", csv_sep, pct);
>> Â Â Â else
>> - Â Â Â Â Â Â fprintf(stderr, " Â ( +-%6.2f%% )", pct);
>> + Â Â Â Â Â Â fprintf(output, " Â ( +-%6.2f%% )", pct);
>> Â }
>>
>> Â static void print_noise(struct perf_evsel *evsel, double avg)
>> @@ -545,16 +547,17 @@ static void nsec_printout(int cpu, struct perf_evsel *evsel, double avg)
>> Â Â Â Â Â Â Â Â Â Â Â csv_output ? 0 : -4,
>> Â Â Â Â Â Â Â Â Â Â Â evsel_list->cpus->map[cpu], csv_sep);
>>
>> - Â Â fprintf(stderr, fmt, cpustr, msecs, csv_sep, event_name(evsel));
>> + Â Â fprintf(output, fmt, cpustr, msecs, csv_sep, event_name(evsel));
>>
>> Â Â Â if (evsel->cgrp)
>> - Â Â Â Â Â Â fprintf(stderr, "%s%s", csv_sep, evsel->cgrp->name);
>> + Â Â Â Â Â Â fprintf(output, "%s%s", csv_sep, evsel->cgrp->name);
>>
>> Â Â Â if (csv_output)
>> Â Â Â Â Â Â Â return;
>>
>> Â Â Â if (perf_evsel__match(evsel, SOFTWARE, SW_TASK_CLOCK))
>> -       fprintf(stderr, " # %8.3f CPUs utilized      ", avg / avg_stats(&walltime_nsecs_stats));
>> +       fprintf(output, " # %8.3f CPUs utilized      ",
>> + Â Â Â Â Â Â Â Â Â Â avg / avg_stats(&walltime_nsecs_stats));
>> Â }
>>
>> Â static void print_stalled_cycles_frontend(int cpu, struct perf_evsel *evsel __used, double avg)
>> @@ -575,9 +578,9 @@ static void print_stalled_cycles_frontend(int cpu, struct perf_evsel *evsel __us
>> Â Â Â else if (ratio > 10.0)
>> Â Â Â Â Â Â Â color = PERF_COLOR_YELLOW;
>>
>> - Â Â fprintf(stderr, " # Â ");
>> - Â Â color_fprintf(stderr, color, "%6.2f%%", ratio);
>> -   fprintf(stderr, " frontend cycles idle  ");
>> + Â Â fprintf(output, " # Â ");
>> + Â Â color_fprintf(output, color, "%6.2f%%", ratio);
>> +   fprintf(output, " frontend cycles idle  ");
>> Â }
>>
>> Â static void print_stalled_cycles_backend(int cpu, struct perf_evsel *evsel __used, double avg)
>> @@ -598,9 +601,9 @@ static void print_stalled_cycles_backend(int cpu, struct perf_evsel *evsel __use
>> Â Â Â else if (ratio > 20.0)
>> Â Â Â Â Â Â Â color = PERF_COLOR_YELLOW;
>>
>> - Â Â fprintf(stderr, " # Â ");
>> - Â Â color_fprintf(stderr, color, "%6.2f%%", ratio);
>> -   fprintf(stderr, " backend  cycles idle  ");
>> + Â Â fprintf(output, " # Â ");
>> + Â Â color_fprintf(output, color, "%6.2f%%", ratio);
>> +   fprintf(output, " backend  cycles idle  ");
>> Â }
>>
>> Â static void print_branch_misses(int cpu, struct perf_evsel *evsel __used, double avg)
>> @@ -621,9 +624,9 @@ static void print_branch_misses(int cpu, struct perf_evsel *evsel __used, double
>> Â Â Â else if (ratio > 5.0)
>> Â Â Â Â Â Â Â color = PERF_COLOR_YELLOW;
>>
>> - Â Â fprintf(stderr, " # Â ");
>> - Â Â color_fprintf(stderr, color, "%6.2f%%", ratio);
>> -   fprintf(stderr, " of all branches     ");
>> + Â Â fprintf(output, " # Â ");
>> + Â Â color_fprintf(output, color, "%6.2f%%", ratio);
>> +   fprintf(output, " of all branches     ");
>> Â }
>>
>> Â static void print_l1_dcache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
>> @@ -644,9 +647,9 @@ static void print_l1_dcache_misses(int cpu, struct perf_evsel *evsel __used, dou
>> Â Â Â else if (ratio > 5.0)
>> Â Â Â Â Â Â Â color = PERF_COLOR_YELLOW;
>>
>> - Â Â fprintf(stderr, " # Â ");
>> - Â Â color_fprintf(stderr, color, "%6.2f%%", ratio);
>> -   fprintf(stderr, " of all L1-dcache hits  ");
>> + Â Â fprintf(output, " # Â ");
>> + Â Â color_fprintf(output, color, "%6.2f%%", ratio);
>> +   fprintf(output, " of all L1-dcache hits  ");
>> Â }
>>
>> Â static void print_l1_icache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
>> @@ -667,9 +670,9 @@ static void print_l1_icache_misses(int cpu, struct perf_evsel *evsel __used, dou
>> Â Â Â else if (ratio > 5.0)
>> Â Â Â Â Â Â Â color = PERF_COLOR_YELLOW;
>>
>> - Â Â fprintf(stderr, " # Â ");
>> - Â Â color_fprintf(stderr, color, "%6.2f%%", ratio);
>> -   fprintf(stderr, " of all L1-icache hits  ");
>> + Â Â fprintf(output, " # Â ");
>> + Â Â color_fprintf(output, color, "%6.2f%%", ratio);
>> +   fprintf(output, " of all L1-icache hits  ");
>> Â }
>>
>> Â static void print_dtlb_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
>> @@ -690,9 +693,9 @@ static void print_dtlb_cache_misses(int cpu, struct perf_evsel *evsel __used, do
>> Â Â Â else if (ratio > 5.0)
>> Â Â Â Â Â Â Â color = PERF_COLOR_YELLOW;
>>
>> - Â Â fprintf(stderr, " # Â ");
>> - Â Â color_fprintf(stderr, color, "%6.2f%%", ratio);
>> - Â Â fprintf(stderr, " of all dTLB cache hits ");
>> + Â Â fprintf(output, " # Â ");
>> + Â Â color_fprintf(output, color, "%6.2f%%", ratio);
>> + Â Â fprintf(output, " of all dTLB cache hits ");
>> Â }
>>
>> Â static void print_itlb_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
>> @@ -713,9 +716,9 @@ static void print_itlb_cache_misses(int cpu, struct perf_evsel *evsel __used, do
>> Â Â Â else if (ratio > 5.0)
>> Â Â Â Â Â Â Â color = PERF_COLOR_YELLOW;
>>
>> - Â Â fprintf(stderr, " # Â ");
>> - Â Â color_fprintf(stderr, color, "%6.2f%%", ratio);
>> - Â Â fprintf(stderr, " of all iTLB cache hits ");
>> + Â Â fprintf(output, " # Â ");
>> + Â Â color_fprintf(output, color, "%6.2f%%", ratio);
>> + Â Â fprintf(output, " of all iTLB cache hits ");
>> Â }
>>
>> Â static void print_ll_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
>> @@ -736,9 +739,9 @@ static void print_ll_cache_misses(int cpu, struct perf_evsel *evsel __used, doub
>> Â Â Â else if (ratio > 5.0)
>> Â Â Â Â Â Â Â color = PERF_COLOR_YELLOW;
>>
>> - Â Â fprintf(stderr, " # Â ");
>> - Â Â color_fprintf(stderr, color, "%6.2f%%", ratio);
>> -   fprintf(stderr, " of all LL-cache hits  ");
>> + Â Â fprintf(output, " # Â ");
>> + Â Â color_fprintf(output, color, "%6.2f%%", ratio);
>> +   fprintf(output, " of all LL-cache hits  ");
>> Â }
>>
>> Â static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
>> @@ -761,10 +764,10 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
>> Â Â Â else
>> Â Â Â Â Â Â Â cpu = 0;
>>
>> - Â Â fprintf(stderr, fmt, cpustr, avg, csv_sep, event_name(evsel));
>> + Â Â fprintf(output, fmt, cpustr, avg, csv_sep, event_name(evsel));
>>
>> Â Â Â if (evsel->cgrp)
>> - Â Â Â Â Â Â fprintf(stderr, "%s%s", csv_sep, evsel->cgrp->name);
>> + Â Â Â Â Â Â fprintf(output, "%s%s", csv_sep, evsel->cgrp->name);
>>
>> Â Â Â if (csv_output)
>> Â Â Â Â Â Â Â return;
>> @@ -775,14 +778,14 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
>> Â Â Â Â Â Â Â if (total)
>> Â Â Â Â Â Â Â Â Â Â Â ratio = avg / total;
>>
>> -       fprintf(stderr, " #  %5.2f  insns per cycle     ", ratio);
>> +       fprintf(output, " #  %5.2f  insns per cycle     ", ratio);
>>
>> Â Â Â Â Â Â Â total = avg_stats(&runtime_stalled_cycles_front_stats[cpu]);
>> Â Â Â Â Â Â Â total = max(total, avg_stats(&runtime_stalled_cycles_back_stats[cpu]));
>>
>> Â Â Â Â Â Â Â if (total && avg) {
>> Â Â Â Â Â Â Â Â Â Â Â ratio = total / avg;
>> -           fprintf(stderr, "\n                       #  %5.2f  stalled cycles per insn", ratio);
>> +           fprintf(output, "\n                       #  %5.2f  stalled cycles per insn", ratio);
>> Â Â Â Â Â Â Â }
>>
>> Â Â Â } else if (perf_evsel__match(evsel, HARDWARE, HW_BRANCH_MISSES) &&
>> @@ -830,7 +833,7 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
>> Â Â Â Â Â Â Â if (total)
>> Â Â Â Â Â Â Â Â Â Â Â ratio = avg * 100 / total;
>>
>> -       fprintf(stderr, " # %8.3f %% of all cache refs   ", ratio);
>> +       fprintf(output, " # %8.3f %% of all cache refs   ", ratio);
>>
>> Â Â Â } else if (perf_evsel__match(evsel, HARDWARE, HW_STALLED_CYCLES_FRONTEND)) {
>> Â Â Â Â Â Â Â print_stalled_cycles_frontend(cpu, evsel, avg);
>> @@ -842,16 +845,16 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
>> Â Â Â Â Â Â Â if (total)
>> Â Â Â Â Â Â Â Â Â Â Â ratio = 1.0 * avg / total;
>>
>> -       fprintf(stderr, " # %8.3f GHz           ", ratio);
>> +       fprintf(output, " # %8.3f GHz           ", ratio);
>> Â Â Â } else if (runtime_nsecs_stats[cpu].n != 0) {
>> Â Â Â Â Â Â Â total = avg_stats(&runtime_nsecs_stats[cpu]);
>>
>> Â Â Â Â Â Â Â if (total)
>> Â Â Â Â Â Â Â Â Â Â Â ratio = 1000.0 * avg / total;
>>
>> -       fprintf(stderr, " # %8.3f M/sec          ", ratio);
>> +       fprintf(output, " # %8.3f M/sec          ", ratio);
>> Â Â Â } else {
>> - Â Â Â Â Â Â fprintf(stderr, " Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ");
>> + Â Â Â Â Â Â fprintf(output, " Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ");
>> Â Â Â }
>> Â }
>>
>> @@ -866,7 +869,7 @@ static void print_counter_aggr(struct perf_evsel *counter)
>> Â Â Â int scaled = counter->counts->scaled;
>>
>> Â Â Â if (scaled == -1) {
>> - Â Â Â Â Â Â fprintf(stderr, "%*s%s%*s",
>> + Â Â Â Â Â Â fprintf(output, "%*s%s%*s",
>> Â Â Â Â Â Â Â Â Â Â Â csv_output ? 0 : 18,
>> Â Â Â Â Â Â Â Â Â Â Â counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED,
>> Â Â Â Â Â Â Â Â Â Â Â csv_sep,
>> @@ -874,9 +877,9 @@ static void print_counter_aggr(struct perf_evsel *counter)
>> Â Â Â Â Â Â Â Â Â Â Â event_name(counter));
>>
>> Â Â Â Â Â Â Â if (counter->cgrp)
>> - Â Â Â Â Â Â Â Â Â Â fprintf(stderr, "%s%s", csv_sep, counter->cgrp->name);
>> + Â Â Â Â Â Â Â Â Â Â fprintf(output, "%s%s", csv_sep, counter->cgrp->name);
>>
>> - Â Â Â Â Â Â fputc('\n', stderr);
>> + Â Â Â Â Â Â fputc('\n', output);
>> Â Â Â Â Â Â Â return;
>> Â Â Â }
>>
>> @@ -888,7 +891,7 @@ static void print_counter_aggr(struct perf_evsel *counter)
>> Â Â Â print_noise(counter, avg);
>>
>> Â Â Â if (csv_output) {
>> - Â Â Â Â Â Â fputc('\n', stderr);
>> + Â Â Â Â Â Â fputc('\n', output);
>> Â Â Â Â Â Â Â return;
>> Â Â Â }
>>
>> @@ -898,9 +901,9 @@ static void print_counter_aggr(struct perf_evsel *counter)
>> Â Â Â Â Â Â Â avg_enabled = avg_stats(&ps->res_stats[1]);
>> Â Â Â Â Â Â Â avg_running = avg_stats(&ps->res_stats[2]);
>>
>> - Â Â Â Â Â Â fprintf(stderr, " [%5.2f%%]", 100 * avg_running / avg_enabled);
>> + Â Â Â Â Â Â fprintf(output, " [%5.2f%%]", 100 * avg_running / avg_enabled);
>> Â Â Â }
>> - Â Â fprintf(stderr, "\n");
>> + Â Â fprintf(output, "\n");
>> Â }
>>
>> Â /*
>> @@ -917,7 +920,7 @@ static void print_counter(struct perf_evsel *counter)
>> Â Â Â Â Â Â Â ena = counter->counts->cpu[cpu].ena;
>> Â Â Â Â Â Â Â run = counter->counts->cpu[cpu].run;
>> Â Â Â Â Â Â Â if (run == 0 || ena == 0) {
>> - Â Â Â Â Â Â Â Â Â Â fprintf(stderr, "CPU%*d%s%*s%s%*s",
>> + Â Â Â Â Â Â Â Â Â Â fprintf(output, "CPU%*d%s%*s%s%*s",
>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â csv_output ? 0 : -4,
>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â evsel_list->cpus->map[cpu], csv_sep,
>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â csv_output ? 0 : 18,
>> @@ -927,9 +930,10 @@ static void print_counter(struct perf_evsel *counter)
>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â event_name(counter));
>>
>> Â Â Â Â Â Â Â Â Â Â Â if (counter->cgrp)
>> - Â Â Â Â Â Â Â Â Â Â Â Â Â Â fprintf(stderr, "%s%s", csv_sep, counter->cgrp->name);
>> + Â Â Â Â Â Â Â Â Â Â Â Â Â Â fprintf(output, "%s%s",
>> + Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â csv_sep, counter->cgrp->name);
>>
>> - Â Â Â Â Â Â Â Â Â Â fputc('\n', stderr);
>> + Â Â Â Â Â Â Â Â Â Â fputc('\n', output);
>> Â Â Â Â Â Â Â Â Â Â Â continue;
>> Â Â Â Â Â Â Â }
>>
>> @@ -942,9 +946,10 @@ static void print_counter(struct perf_evsel *counter)
>> Â Â Â Â Â Â Â Â Â Â Â print_noise(counter, 1.0);
>>
>> Â Â Â Â Â Â Â Â Â Â Â if (run != ena)
>> - Â Â Â Â Â Â Â Â Â Â Â Â Â Â fprintf(stderr, " Â (%.2f%%)", 100.0 * run / ena);
>> + Â Â Â Â Â Â Â Â Â Â Â Â Â Â fprintf(output, " Â (%.2f%%)",
>> + Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â 100.0 * run / ena);
>> Â Â Â Â Â Â Â }
>> - Â Â Â Â Â Â fputc('\n', stderr);
>> + Â Â Â Â Â Â fputc('\n', output);
>> Â Â Â }
>> Â }
>>
>> @@ -956,21 +961,21 @@ static void print_stat(int argc, const char **argv)
>> Â Â Â fflush(stdout);
>>
>> Â Â Â if (!csv_output) {
>> - Â Â Â Â Â Â fprintf(stderr, "\n");
>> - Â Â Â Â Â Â fprintf(stderr, " Performance counter stats for ");
>> + Â Â Â Â Â Â fprintf(output, "\n");
>> + Â Â Â Â Â Â fprintf(output, " Performance counter stats for ");
>> Â Â Â Â Â Â Â if(target_pid == -1 && target_tid == -1) {
>> - Â Â Â Â Â Â Â Â Â Â fprintf(stderr, "\'%s", argv[0]);
>> + Â Â Â Â Â Â Â Â Â Â fprintf(output, "\'%s", argv[0]);
>> Â Â Â Â Â Â Â Â Â Â Â for (i = 1; i < argc; i++)
>> - Â Â Â Â Â Â Â Â Â Â Â Â Â Â fprintf(stderr, " %s", argv[i]);
>> + Â Â Â Â Â Â Â Â Â Â Â Â Â Â fprintf(output, " %s", argv[i]);
>> Â Â Â Â Â Â Â } else if (target_pid != -1)
>> - Â Â Â Â Â Â Â Â Â Â fprintf(stderr, "process id \'%d", target_pid);
>> + Â Â Â Â Â Â Â Â Â Â fprintf(output, "process id \'%d", target_pid);
>> Â Â Â Â Â Â Â else
>> - Â Â Â Â Â Â Â Â Â Â fprintf(stderr, "thread id \'%d", target_tid);
>> + Â Â Â Â Â Â Â Â Â Â fprintf(output, "thread id \'%d", target_tid);
>>
>> - Â Â Â Â Â Â fprintf(stderr, "\'");
>> + Â Â Â Â Â Â fprintf(output, "\'");
>> Â Â Â Â Â Â Â if (run_count > 1)
>> - Â Â Â Â Â Â Â Â Â Â fprintf(stderr, " (%d runs)", run_count);
>> - Â Â Â Â Â Â fprintf(stderr, ":\n\n");
>> + Â Â Â Â Â Â Â Â Â Â fprintf(output, " (%d runs)", run_count);
>> + Â Â Â Â Â Â fprintf(output, ":\n\n");
>> Â Â Â }
>>
>> Â Â Â if (no_aggr) {
>> @@ -983,15 +988,15 @@ static void print_stat(int argc, const char **argv)
>>
>> Â Â Â if (!csv_output) {
>> Â Â Â Â Â Â Â if (!null_run)
>> - Â Â Â Â Â Â Â Â Â Â fprintf(stderr, "\n");
>> - Â Â Â Â Â Â fprintf(stderr, " %17.9f seconds time elapsed",
>> + Â Â Â Â Â Â Â Â Â Â fprintf(output, "\n");
>> + Â Â Â Â Â Â fprintf(output, " %17.9f seconds time elapsed",
>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â avg_stats(&walltime_nsecs_stats)/1e9);
>> Â Â Â Â Â Â Â if (run_count > 1) {
>> - Â Â Â Â Â Â Â Â Â Â fprintf(stderr, " Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ");
>> + Â Â Â Â Â Â Â Â Â Â fprintf(output, " Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â ");
>> Â Â Â Â Â Â Â Â Â Â Â print_noise_pct(stddev_stats(&walltime_nsecs_stats),
>> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â avg_stats(&walltime_nsecs_stats));
>> Â Â Â Â Â Â Â }
>> - Â Â Â Â Â Â fprintf(stderr, "\n\n");
>> + Â Â Â Â Â Â fprintf(output, "\n\n");
>> Â Â Â }
>> Â }
>>
>> @@ -1029,6 +1034,8 @@ static int stat__set_big_num(const struct option *opt __used,
>> Â Â Â return 0;
>> Â }
>>
>> +static bool append_file;
>> +
>> Â static const struct option options[] = {
>> Â Â Â OPT_CALLBACK('e', "event", &evsel_list, "event",
>> Â Â Â Â Â Â Â Â Â Â "event selector. use 'perf list' to list available events",
>> @@ -1067,6 +1074,9 @@ static const struct option options[] = {
>> Â Â Â OPT_CALLBACK('G', "cgroup", &evsel_list, "name",
>> Â Â Â Â Â Â Â Â Â Â "monitor event in cgroup name only",
>> Â Â Â Â Â Â Â Â Â Â parse_cgroups),
>> + Â Â OPT_STRING('o', "output", &output_name, "file",
>> + Â Â Â Â Â Â Â Â "output file name"),
>> + Â Â OPT_BOOLEAN(0, "append", &append_file, "append to the output file"),
>> Â Â Â OPT_END()
>> Â };
>>
>> @@ -1138,6 +1148,7 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
>> Â {
>> Â Â Â struct perf_evsel *pos;
>> Â Â Â int status = -ENOMEM;
>> + Â Â const char *mode;
>>
>> Â Â Â setlocale(LC_ALL, "");
>>
>> @@ -1148,6 +1159,23 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
>> Â Â Â argc = parse_options(argc, argv, options, stat_usage,
>> Â Â Â Â Â Â Â PARSE_OPT_STOP_AT_NON_OPTION);
>>
>> + Â Â output = stderr;
>> + Â Â if (output_name && strcmp(output_name, "-"))
>> + Â Â Â Â Â Â output = NULL;
>> +
>> + Â Â if (!output) {
>> + Â Â Â Â Â Â struct timespec tm;
>> + Â Â Â Â Â Â mode = append_file ? "a" : "w";
>> +
>> + Â Â Â Â Â Â output = fopen(output_name, mode);
>> + Â Â Â Â Â Â if (!output) {
>> + Â Â Â Â Â Â Â Â Â Â perror("failed to create output file");
>> + Â Â Â Â Â Â Â Â Â Â exit(-1);
>> + Â Â Â Â Â Â }
>> + Â Â Â Â Â Â clock_gettime(CLOCK_REALTIME, &tm);
>> + Â Â Â Â Â Â fprintf(output, "# started on %s\n", ctime(&tm.tv_sec));
>> + Â Â }
>> +
>> Â Â Â if (csv_sep)
>> Â Â Â Â Â Â Â csv_output = true;
>> Â Â Â else
>> @@ -1223,7 +1251,8 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
>> Â Â Â status = 0;
>> Â Â Â for (run_idx = 0; run_idx < run_count; run_idx++) {
>> Â Â Â Â Â Â Â if (run_count != 1 && verbose)
>> - Â Â Â Â Â Â Â Â Â Â fprintf(stderr, "[ perf stat: executing run #%d ... ]\n", run_idx + 1);
>> + Â Â Â Â Â Â Â Â Â Â fprintf(output, "[ perf stat: executing run #%d ... ]\n",
>> + Â Â Â Â Â Â Â Â Â Â Â Â Â Â run_idx + 1);
>>
>> Â Â Â Â Â Â Â if (sync_run)
>> Â Â Â Â Â Â Â Â Â Â Â sync();
>> diff --git a/tools/perf/util/color.c b/tools/perf/util/color.c
>> index e191eb9..521c38a 100644
>> --- a/tools/perf/util/color.c
>> +++ b/tools/perf/util/color.c
>> @@ -200,7 +200,7 @@ static int __color_vfprintf(FILE *fp, const char *color, const char *fmt,
>> Â Â Â Â * Auto-detect:
>> Â Â Â Â */
>> Â Â Â if (perf_use_color_default < 0) {
>> - Â Â Â Â Â Â if (isatty(1) || pager_in_use())
>> + Â Â Â Â Â Â if (isatty(fileno(fp)) || pager_in_use())
>> Â Â Â Â Â Â Â Â Â Â Â perf_use_color_default = 1;
>> Â Â Â Â Â Â Â else
>> Â Â Â Â Â Â Â Â Â Â Â perf_use_color_default = 0;
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>> Please read the FAQ at  http://www.tux.org/lkml/
>
ÿôèº{.nÇ+‰·Ÿ®‰†+%ŠËÿ±éݶ\x17¥Šwÿº{.nÇ+‰·¥Š{±þG«éÿŠ{ayº\x1dʇڙë,j\a¢f£¢·hšïêÿ‘êçz_è®\x03(階ŠÝ¢j"ú\x1a¶^[m§ÿÿ¾\a«þG«éÿ¢¸?™¨èÚ&£ø§~á¶iO•æ¬z·švØ^\x14\x04\x1a¶^[m§ÿÿÃ\fÿ¶ìÿ¢¸?–I¥
^ permalink raw reply [flat|nested] 4+ messages in thread
* [tip:perf/core] perf stat: Add -o and --append options
2011-08-15 20:22 [PATCH] perf: add -o and --append options to perf stat Stephane Eranian
2011-08-15 21:32 ` David Ahern
@ 2011-08-18 20:14 ` tip-bot for Stephane Eranian
1 sibling, 0 replies; 4+ messages in thread
From: tip-bot for Stephane Eranian @ 2011-08-18 20:14 UTC (permalink / raw)
To: linux-tip-commits
Cc: acme, linux-kernel, eranian, hpa, mingo, peterz, tglx, mingo
Commit-ID: 4aa9015f8bfd2c8d7cc33a360275b71a9d708b37
Gitweb: http://git.kernel.org/tip/4aa9015f8bfd2c8d7cc33a360275b71a9d708b37
Author: Stephane Eranian <eranian@google.com>
AuthorDate: Mon, 15 Aug 2011 22:22:33 +0200
Committer: Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Thu, 18 Aug 2011 07:46:13 -0300
perf stat: Add -o and --append options
This patch adds an option (-o) to save the output of perf stat into a
file. You could do this with perf record but not with perf stat.
Instead, you had to fiddle with stderr to save the counts into a
separate file.
The patch also adds the --append option so that results can be
concatenated into a single file across runs. Each run of the tool is
clearly separated by a comment line starting with a hash mark. The -A
option of perf record is already used by perf stat, so we only add a
long option.
$ perf stat -o res.txt date
$ cat res.txt
Performance counter stats for 'date':
0.791306 task-clock # 0.668 CPUs utilized
2 context-switches # 0.003 M/sec
0 CPU-migrations # 0.000 M/sec
197 page-faults # 0.249 M/sec
1878143 cycles # 2.373 GHz
<not supported> stalled-cycles-frontend
<not supported> stalled-cycles-backend
1083367 instructions # 0.58 insns per cycle
193027 branches # 243.935 M/sec
9014 branch-misses # 4.67% of all branches
0.001184746 seconds time elapsed
The option can be combined with -x to make the output file much easier
to parse.
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20110815202233.GA18535@quad
Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/Documentation/perf-stat.txt | 7 ++
tools/perf/builtin-stat.c | 155 +++++++++++++++++++-------------
tools/perf/util/color.c | 2 +-
3 files changed, 100 insertions(+), 64 deletions(-)
diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
index 918cc38..08394c4 100644
--- a/tools/perf/Documentation/perf-stat.txt
+++ b/tools/perf/Documentation/perf-stat.txt
@@ -94,6 +94,13 @@ an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must ha
corresponding events, i.e., they always refer to events defined earlier on the command
line.
+-o file::
+-output file::
+Print the output into the designated file.
+
+--append::
+Append to the output file designated with the -o option. Ignored if -o is not specified.
+
EXAMPLES
--------
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 1ad04ce..a22393d 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -193,6 +193,8 @@ static int big_num_opt = -1;
static const char *cpu_list;
static const char *csv_sep = NULL;
static bool csv_output = false;
+static const char *output_name = NULL;
+static FILE *output = NULL;
static volatile int done = 0;
@@ -351,7 +353,7 @@ static int read_counter_aggr(struct perf_evsel *counter)
update_stats(&ps->res_stats[i], count[i]);
if (verbose) {
- fprintf(stderr, "%s: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n",
+ fprintf(output, "%s: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n",
event_name(counter), count[0], count[1], count[2]);
}
@@ -518,9 +520,9 @@ static void print_noise_pct(double total, double avg)
pct = 100.0*total/avg;
if (csv_output)
- fprintf(stderr, "%s%.2f%%", csv_sep, pct);
+ fprintf(output, "%s%.2f%%", csv_sep, pct);
else
- fprintf(stderr, " ( +-%6.2f%% )", pct);
+ fprintf(output, " ( +-%6.2f%% )", pct);
}
static void print_noise(struct perf_evsel *evsel, double avg)
@@ -545,16 +547,17 @@ static void nsec_printout(int cpu, struct perf_evsel *evsel, double avg)
csv_output ? 0 : -4,
evsel_list->cpus->map[cpu], csv_sep);
- fprintf(stderr, fmt, cpustr, msecs, csv_sep, event_name(evsel));
+ fprintf(output, fmt, cpustr, msecs, csv_sep, event_name(evsel));
if (evsel->cgrp)
- fprintf(stderr, "%s%s", csv_sep, evsel->cgrp->name);
+ fprintf(output, "%s%s", csv_sep, evsel->cgrp->name);
if (csv_output)
return;
if (perf_evsel__match(evsel, SOFTWARE, SW_TASK_CLOCK))
- fprintf(stderr, " # %8.3f CPUs utilized ", avg / avg_stats(&walltime_nsecs_stats));
+ fprintf(output, " # %8.3f CPUs utilized ",
+ avg / avg_stats(&walltime_nsecs_stats));
}
static void print_stalled_cycles_frontend(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -575,9 +578,9 @@ static void print_stalled_cycles_frontend(int cpu, struct perf_evsel *evsel __us
else if (ratio > 10.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " frontend cycles idle ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " frontend cycles idle ");
}
static void print_stalled_cycles_backend(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -598,9 +601,9 @@ static void print_stalled_cycles_backend(int cpu, struct perf_evsel *evsel __use
else if (ratio > 20.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " backend cycles idle ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " backend cycles idle ");
}
static void print_branch_misses(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -621,9 +624,9 @@ static void print_branch_misses(int cpu, struct perf_evsel *evsel __used, double
else if (ratio > 5.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " of all branches ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all branches ");
}
static void print_l1_dcache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -644,9 +647,9 @@ static void print_l1_dcache_misses(int cpu, struct perf_evsel *evsel __used, dou
else if (ratio > 5.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " of all L1-dcache hits ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all L1-dcache hits ");
}
static void print_l1_icache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -667,9 +670,9 @@ static void print_l1_icache_misses(int cpu, struct perf_evsel *evsel __used, dou
else if (ratio > 5.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " of all L1-icache hits ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all L1-icache hits ");
}
static void print_dtlb_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -690,9 +693,9 @@ static void print_dtlb_cache_misses(int cpu, struct perf_evsel *evsel __used, do
else if (ratio > 5.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " of all dTLB cache hits ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all dTLB cache hits ");
}
static void print_itlb_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -713,9 +716,9 @@ static void print_itlb_cache_misses(int cpu, struct perf_evsel *evsel __used, do
else if (ratio > 5.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " of all iTLB cache hits ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all iTLB cache hits ");
}
static void print_ll_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
@@ -736,9 +739,9 @@ static void print_ll_cache_misses(int cpu, struct perf_evsel *evsel __used, doub
else if (ratio > 5.0)
color = PERF_COLOR_YELLOW;
- fprintf(stderr, " # ");
- color_fprintf(stderr, color, "%6.2f%%", ratio);
- fprintf(stderr, " of all LL-cache hits ");
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all LL-cache hits ");
}
static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
@@ -761,10 +764,10 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
else
cpu = 0;
- fprintf(stderr, fmt, cpustr, avg, csv_sep, event_name(evsel));
+ fprintf(output, fmt, cpustr, avg, csv_sep, event_name(evsel));
if (evsel->cgrp)
- fprintf(stderr, "%s%s", csv_sep, evsel->cgrp->name);
+ fprintf(output, "%s%s", csv_sep, evsel->cgrp->name);
if (csv_output)
return;
@@ -775,14 +778,14 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
if (total)
ratio = avg / total;
- fprintf(stderr, " # %5.2f insns per cycle ", ratio);
+ fprintf(output, " # %5.2f insns per cycle ", ratio);
total = avg_stats(&runtime_stalled_cycles_front_stats[cpu]);
total = max(total, avg_stats(&runtime_stalled_cycles_back_stats[cpu]));
if (total && avg) {
ratio = total / avg;
- fprintf(stderr, "\n # %5.2f stalled cycles per insn", ratio);
+ fprintf(output, "\n # %5.2f stalled cycles per insn", ratio);
}
} else if (perf_evsel__match(evsel, HARDWARE, HW_BRANCH_MISSES) &&
@@ -830,7 +833,7 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
if (total)
ratio = avg * 100 / total;
- fprintf(stderr, " # %8.3f %% of all cache refs ", ratio);
+ fprintf(output, " # %8.3f %% of all cache refs ", ratio);
} else if (perf_evsel__match(evsel, HARDWARE, HW_STALLED_CYCLES_FRONTEND)) {
print_stalled_cycles_frontend(cpu, evsel, avg);
@@ -842,16 +845,16 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
if (total)
ratio = 1.0 * avg / total;
- fprintf(stderr, " # %8.3f GHz ", ratio);
+ fprintf(output, " # %8.3f GHz ", ratio);
} else if (runtime_nsecs_stats[cpu].n != 0) {
total = avg_stats(&runtime_nsecs_stats[cpu]);
if (total)
ratio = 1000.0 * avg / total;
- fprintf(stderr, " # %8.3f M/sec ", ratio);
+ fprintf(output, " # %8.3f M/sec ", ratio);
} else {
- fprintf(stderr, " ");
+ fprintf(output, " ");
}
}
@@ -866,7 +869,7 @@ static void print_counter_aggr(struct perf_evsel *counter)
int scaled = counter->counts->scaled;
if (scaled == -1) {
- fprintf(stderr, "%*s%s%*s",
+ fprintf(output, "%*s%s%*s",
csv_output ? 0 : 18,
counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED,
csv_sep,
@@ -874,9 +877,9 @@ static void print_counter_aggr(struct perf_evsel *counter)
event_name(counter));
if (counter->cgrp)
- fprintf(stderr, "%s%s", csv_sep, counter->cgrp->name);
+ fprintf(output, "%s%s", csv_sep, counter->cgrp->name);
- fputc('\n', stderr);
+ fputc('\n', output);
return;
}
@@ -888,7 +891,7 @@ static void print_counter_aggr(struct perf_evsel *counter)
print_noise(counter, avg);
if (csv_output) {
- fputc('\n', stderr);
+ fputc('\n', output);
return;
}
@@ -898,9 +901,9 @@ 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, " [%5.2f%%]", 100 * avg_running / avg_enabled);
+ fprintf(output, " [%5.2f%%]", 100 * avg_running / avg_enabled);
}
- fprintf(stderr, "\n");
+ fprintf(output, "\n");
}
/*
@@ -917,7 +920,7 @@ static void print_counter(struct perf_evsel *counter)
ena = counter->counts->cpu[cpu].ena;
run = counter->counts->cpu[cpu].run;
if (run == 0 || ena == 0) {
- fprintf(stderr, "CPU%*d%s%*s%s%*s",
+ fprintf(output, "CPU%*d%s%*s%s%*s",
csv_output ? 0 : -4,
evsel_list->cpus->map[cpu], csv_sep,
csv_output ? 0 : 18,
@@ -927,9 +930,10 @@ static void print_counter(struct perf_evsel *counter)
event_name(counter));
if (counter->cgrp)
- fprintf(stderr, "%s%s", csv_sep, counter->cgrp->name);
+ fprintf(output, "%s%s",
+ csv_sep, counter->cgrp->name);
- fputc('\n', stderr);
+ fputc('\n', output);
continue;
}
@@ -942,9 +946,10 @@ static void print_counter(struct perf_evsel *counter)
print_noise(counter, 1.0);
if (run != ena)
- fprintf(stderr, " (%.2f%%)", 100.0 * run / ena);
+ fprintf(output, " (%.2f%%)",
+ 100.0 * run / ena);
}
- fputc('\n', stderr);
+ fputc('\n', output);
}
}
@@ -956,21 +961,21 @@ static void print_stat(int argc, const char **argv)
fflush(stdout);
if (!csv_output) {
- fprintf(stderr, "\n");
- fprintf(stderr, " Performance counter stats for ");
+ fprintf(output, "\n");
+ fprintf(output, " Performance counter stats for ");
if(target_pid == -1 && target_tid == -1) {
- fprintf(stderr, "\'%s", argv[0]);
+ fprintf(output, "\'%s", argv[0]);
for (i = 1; i < argc; i++)
- fprintf(stderr, " %s", argv[i]);
+ fprintf(output, " %s", argv[i]);
} else if (target_pid != -1)
- fprintf(stderr, "process id \'%d", target_pid);
+ fprintf(output, "process id \'%d", target_pid);
else
- fprintf(stderr, "thread id \'%d", target_tid);
+ fprintf(output, "thread id \'%d", target_tid);
- fprintf(stderr, "\'");
+ fprintf(output, "\'");
if (run_count > 1)
- fprintf(stderr, " (%d runs)", run_count);
- fprintf(stderr, ":\n\n");
+ fprintf(output, " (%d runs)", run_count);
+ fprintf(output, ":\n\n");
}
if (no_aggr) {
@@ -983,15 +988,15 @@ static void print_stat(int argc, const char **argv)
if (!csv_output) {
if (!null_run)
- fprintf(stderr, "\n");
- fprintf(stderr, " %17.9f seconds time elapsed",
+ fprintf(output, "\n");
+ fprintf(output, " %17.9f seconds time elapsed",
avg_stats(&walltime_nsecs_stats)/1e9);
if (run_count > 1) {
- fprintf(stderr, " ");
+ fprintf(output, " ");
print_noise_pct(stddev_stats(&walltime_nsecs_stats),
avg_stats(&walltime_nsecs_stats));
}
- fprintf(stderr, "\n\n");
+ fprintf(output, "\n\n");
}
}
@@ -1029,6 +1034,8 @@ static int stat__set_big_num(const struct option *opt __used,
return 0;
}
+static bool append_file;
+
static const struct option options[] = {
OPT_CALLBACK('e', "event", &evsel_list, "event",
"event selector. use 'perf list' to list available events",
@@ -1067,6 +1074,9 @@ static const struct option options[] = {
OPT_CALLBACK('G', "cgroup", &evsel_list, "name",
"monitor event in cgroup name only",
parse_cgroups),
+ OPT_STRING('o', "output", &output_name, "file",
+ "output file name"),
+ OPT_BOOLEAN(0, "append", &append_file, "append to the output file"),
OPT_END()
};
@@ -1138,6 +1148,7 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
{
struct perf_evsel *pos;
int status = -ENOMEM;
+ const char *mode;
setlocale(LC_ALL, "");
@@ -1148,6 +1159,23 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
argc = parse_options(argc, argv, options, stat_usage,
PARSE_OPT_STOP_AT_NON_OPTION);
+ output = stderr;
+ if (output_name && strcmp(output_name, "-"))
+ output = NULL;
+
+ if (!output) {
+ struct timespec tm;
+ mode = append_file ? "a" : "w";
+
+ output = fopen(output_name, mode);
+ if (!output) {
+ perror("failed to create output file");
+ exit(-1);
+ }
+ clock_gettime(CLOCK_REALTIME, &tm);
+ fprintf(output, "# started on %s\n", ctime(&tm.tv_sec));
+ }
+
if (csv_sep)
csv_output = true;
else
@@ -1223,7 +1251,8 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used)
status = 0;
for (run_idx = 0; run_idx < run_count; run_idx++) {
if (run_count != 1 && verbose)
- fprintf(stderr, "[ perf stat: executing run #%d ... ]\n", run_idx + 1);
+ fprintf(output, "[ perf stat: executing run #%d ... ]\n",
+ run_idx + 1);
if (sync_run)
sync();
diff --git a/tools/perf/util/color.c b/tools/perf/util/color.c
index e191eb9..521c38a 100644
--- a/tools/perf/util/color.c
+++ b/tools/perf/util/color.c
@@ -200,7 +200,7 @@ static int __color_vfprintf(FILE *fp, const char *color, const char *fmt,
* Auto-detect:
*/
if (perf_use_color_default < 0) {
- if (isatty(1) || pager_in_use())
+ if (isatty(fileno(fp)) || pager_in_use())
perf_use_color_default = 1;
else
perf_use_color_default = 0;
^ permalink raw reply related [flat|nested] 4+ messages in thread
end of thread, other threads:[~2011-08-18 20:14 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-08-15 20:22 [PATCH] perf: add -o and --append options to perf stat Stephane Eranian
2011-08-15 21:32 ` David Ahern
2011-08-15 21:33 ` Stephane Eranian
2011-08-18 20:14 ` [tip:perf/core] perf stat: Add -o and --append options tip-bot for Stephane Eranian
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox