perf, tools: Refactor and support interval and CSV metrics
Currently perf stat does not support printing computed metrics for interval (-I xxx) or CSV (-x,) mode. For example IPC or TSX metrics over time are quite useful to know.
This patch implements them. The main obstacle was that the
metrics printing was all open coded all over the metrics computation code. The second patch refactors the metrics printing to work through call backs that can be more easily changed. This also cleans up the metrics printing significantly. The indentation is now handled through printf, no more need to manually count spaces.
Then based on that it implements metrics printing for CSV and interval mode.
Example output:
% perf stat -I1000 -a sleep 1
# time counts unit events metric multiplex
1.001301370 12020.049593 task-clock (msec) (100.00%)
1.001301370 3,952 context-switches # 0.329 K/sec (100.00%)
1.001301370 69 cpu-migrations # 0.006 K/sec (100.00%)
1.001301370 76 page-faults # 0.006 K/sec
1.001301370 386,582,789 cycles # 0.032 GHz (100.00%)
1.001301370 716,441,544 stalled-cycles-frontend # 185.33% frontend cycles idle (100.00%)
1.001301370 <not supported> stalled-cycles-backend
1.001301370 101,751,678 instructions # 0.26 insn per cycle
1.001301370 # 7.04 stalled cycles per insn (100.00%)
1.001301370 20,914,692 branches # 1.740 M/sec (100.00%)
1.001301370 1,943,630 branch-misses # 9.29% of all branches CSV mode
% perf stat -x, -I1000 -a sleep 1
1.000852081,12016.143006,,task-clock
1.000852081,4457,,context-switches,12015168277,100.00,0.371,K/sec
1.000852081,50,,cpu-migrations,12014024424,100.00,0.004,K/sec
1.000852081,76,,page-faults,12013076716,100.00,0.006,K/sec
1.000852081,515854373,,cycles,12011235336,100.00,0.043,GHz
1.000852081,1030742150,,stalled-cycles-frontend,12010984057,100.00,199.81,frontend cycles idle
1.000852081,<not supported>,,stalled-cycles-backend,0,100.00
1.000852081,116782495,,instructions,12011130729,100.00,0.23,insn per cycle
1.000852081,,,,12011130729,100.00,8.83,stalled cycles per insn
1.000852081,23748237,,branches,12010745125,100.00,1.976,M/sec
1.000852081,1976560,,branch-misses,12010501884,100.00,8.32,of all branchesAvailable ingit://git.kernel.org/pub/scm/linux/kernel/git/ak/linux-misc perf/stat-metrics
Note: for some of the --per-*/-A modes metrics are not printed correctly. That was already the case before, so I didn't change it. I think some of it may be related to Jiri's earlier stat changes.
PATCH 1/4 perf, tools: Do not include escape sequences in color_vfprintf return
From: Andi Kleen <[email protected]>
color_vprintf was including the length of the invisible escape sequences in its return argument. Don't include them to make the return value usable for indentation calculations.
--- tools/perf/util/color.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
tools/perf/util/color.c +3 -3
@@ -83,12 +83,12 @@ static int __color_vfprintf(FILE *fp, const char *color, const char *fmt, }if (perf_use_color_default && *color) - r +=fprintf (fp,"%s" , color); +fprintf (fp,"%s" , color); r +=vfprintf (fp, fmt, args);if (perf_use_color_default && *color) - r +=fprintf (fp,"%s" ,PERF_COLOR_RESET ); +fprintf (fp,"%s" ,PERF_COLOR_RESET );if (trail) - r +=fprintf (fp,"%s" , trail); +fprintf (fp,"%s" , trail);return r; }
Signature · 1 line
2.4.3
PATCH 1/1 perf tools: Remove color_fprintf_ln function
Quoted 16 lines · Andi Kleen
From: Andi Kleen <[email protected]>
color_vprintfwas including the length of the invisible escape sequences in its return argument. Don't include them to make the return value usable for indentation calculations.Signed-off-by Andi Kleen <[email protected]>--- tools/perf/util/color.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)tools/perf/util/color.c +0
-0
tools/perf/util/color.c static int __color_vfprintf(FILE *fp, const char *color, const char *fmt, · line 86
Quoted 5 lines
tools/perf/util/color.c +1
-1@@ -83,12 +83,12 @@ static int __color_vfprintf(FILE *fp, const char *color, const char *fmt, }if (perf_use_color_default && *color) - r +=fprintf (fp,"%s" , color); +fprintf (fp,"%s" , color);
could you please put in a comment saying we dont include colors in the return length..
tools/perf/util/color.c static int __color_vfprintf(FILE *fp, const char *color, const char *fmt, · line 91
Quoted 7 lines
tools/perf/util/color.c +2
-2r +=vfprintf (fp, fmt, args);if (perf_use_color_default && *color) - r +=fprintf (fp,"%s" ,PERF_COLOR_RESET ); +fprintf (fp,"%s" ,PERF_COLOR_RESET );if (trail) - r +=fprintf (fp,"%s" , trail); +fprintf (fp,"%s" , trail);
trail's not color, right?
anyway seems like it's always '\n' through color_fprintf_ln, which is not used at all, removing.. ;-)
jirka
---
It's not used anymore.
--- tools/perf/util/color.c | 16 ++-------------- tools/perf/util/color.h | 1 - 2 files changed, 2 insertions(+), 15 deletions(-)
tools/perf/util/color.c +2 -14
@@ -68,7 +68,7 @@ static int __color_vsnprintf(char *bf, size_t size, const char *color, }static int __color_vfprintf (FILE *fp,const char *color,const char *fmt, -va_list args,const char *trail) +va_list args) {int r =0 ;
@@ -87,8 +87,6 @@ static int __color_vfprintf(FILE *fp, const char *color, const char *fmt, r +=vfprintf (fp, fmt, args);if (perf_use_color_default && *color) r +=fprintf (fp,"%s" ,PERF_COLOR_RESET ); -if (trail) - r +=fprintf (fp,"%s" , trail);return r; }
@@ -100,7 +98,7 @@ int color_vsnprintf(char *bf, size_t size, const char *color,int color_vfprintf (FILE *fp,const char *color,const char *fmt,va_list args) { -return __color_vfprintf (fp, color, fmt, args,NULL ); +return __color_vfprintf (fp, color, fmt, args); } int color_snprintf(char *bf, size_t size, const char *color,
@@ -126,16 +124,6 @@ int color_fprintf(FILE *fp, const char *color, const char *fmt, ...)return r; } -int color_fprintf_ln (FILE *fp,const char *color,const char *fmt, ...) -{ -va_list args; -int r; -va_start (args, fmt); - r =__color_vfprintf (fp, color, fmt, args,"\n" ); -va_end (args); -return r; -} -/* * This function splits the buffer by newlines and colors the lines individually. *
tools/perf/util/color.h +0 -1
@@ -35,7 +35,6 @@ int color_vsnprintf(char *bf, size_t size, const char *color,int color_vfprintf (FILE *fp,const char *color,const char *fmt,va_list args);int color_fprintf (FILE *fp,const char *color,const char *fmt, ...);int color_snprintf (char *bf,size_t size,const char *color,const char *fmt, ...); -int color_fprintf_ln (FILE *fp,const char *color,const char *fmt, ...);int color_fwrite_lines (FILE *fp,const char *color,size_t count,const char *buf);int value_color_snprintf (char *bf,size_t size,const char *fmt,double value);int percent_color_snprintf (char *bf,size_t size,const char *fmt, ...);
Signature · 1 line
2.4.3
PATCH 2/4 perf, tools, stat: Abstract stat metrics printing
In reply to Andi KleenFrom: Andi Kleen <[email protected]>
Abstract the printing of shadow metrics. Instead of every metric calling fprintf directly and taking care of indentation, use two call backs: one to print metrics and another to start a new line.
This will allow adding metrics to CSV mode and also using them for other purposes.
The computation of padding is now done in the central callback, instead of every metric doing it manually. This makes it easier to add new metrics.
Right now there is no (intentional) behavior change, just refactoring.
--- tools/perf/builtin-stat.c | 98 +++++++++++++++++++-------- tools/perf/util/stat-shadow.c | 154 ++++++++++++++++++++++-------------------- tools/perf/util/stat.h | 10 ++- 3 files changed, 160 insertions(+), 102 deletions(-)
tools/perf/builtin-stat.c +71 -27
@@ -617,7 +617,49 @@ static void aggr_printout(struct perf_evsel *evsel, int id, int nr) } } -static void nsec_printout (int id,int nr,struct perf_evsel *evsel,double avg) +struct outstate { +FILE *fh; +}; + +#define BASE_INDENT 41 +#define AGGR_INDENT 8 +#define METRIC_LEN 35 +#define NA_INDENT 16 + +static void new_line_no_aggr_std (void *ctx) +{ +struct outstate *os = ctx; +fprintf (os->fh,"\n%*s" ,BASE_INDENT +NA_INDENT ,"" ); +} + +static void new_line_std (void *ctx) +{ +struct outstate *os = ctx; +fprintf (os->fh,"\n%-*s" ,BASE_INDENT +AGGR_INDENT ,"" ); +} + +static void print_metric_std (void *ctx,const char *color,const char *fmt, +const char *unit,double val) +{ +struct outstate *os = ctx; +FILE *out = os->fh; +int n; + +if (unit ==NULL ) { +fprintf (out,"%-*s" ,METRIC_LEN ,"" ); +return ; + } + + n =fprintf (out," # " ); +if (color) + n +=color_fprintf (out, color, fmt, val); +else + n +=fprintf (out, fmt, val); +fprintf (out," %-*s" ,METRIC_LEN - n -1 , unit); +} + +static void nsec_printout (int id,int nr,struct perf_evsel *evsel,double avg, +void *ctx,print_metric_t print_metric) {double msecs = avg /1e6 ;const char *fmt_v, *fmt_n;
@@ -647,13 +689,16 @@ static void nsec_printout(int id, int nr, struct perf_evsel *evsel, double avg)return ;if (perf_evsel__match (evsel,SOFTWARE ,SW_TASK_CLOCK )) -fprintf (output," # %8.3f CPUs utilized " , - avg /avg_stats (&walltime_nsecs_stats)); +print_metric (ctx,NULL ,"%8.3f" ,"CPUs utilized" , + avg /avg_stats (&walltime_nsecs_stats));else -fprintf (output," " ); +print_metric (ctx,NULL ,NULL ,NULL ,0 ); } -static void abs_printout (int id,int nr,struct perf_evsel *evsel,double avg) +static void abs_printout (int id,int nr,struct perf_evsel *evsel,double avg, +void *ctx, +print_metric_t print_metric, +void (*new_line)(void *)) {double sc = evsel->scale;const char *fmt;
@@ -688,7 +733,23 @@ static void abs_printout(int id, int nr, struct perf_evsel *evsel, double avg)if (csv_output || interval)return ; -perf_stat__print_shadow_stats (output, evsel, avg, cpu, aggr_mode); +perf_stat__print_shadow_stats (evsel, avg, cpu, + print_metric, + new_line, + ctx); +} + +static void printout (int id,int nr,struct perf_evsel *counter,double uval) +{ +struct outstate os = { .fh = output }; + +if (nsec_counter (counter)) +nsec_printout (id, nr, counter, uval, &os, print_metric_std); +else +abs_printout (id, nr, counter, uval, &os, + print_metric_std, + aggr_mode ==AGGR_NONE ? + new_line_no_aggr_std : new_line_std); }static void print_aggr (char *prefix)
@@ -744,12 +805,7 @@ static void print_aggr(char *prefix)continue ; } uval = val * counter->scale; - -if (nsec_counter (counter)) -nsec_printout (id, nr, counter, uval); -else -abs_printout (id, nr, counter, uval); - +printout (id, nr, counter, uval);if (!csv_output)print_noise (counter,1.0 );
@@ -779,11 +835,7 @@ static void print_aggr_thread(struct perf_evsel *counter, char *prefix)fprintf (output,"%s" , prefix); uval = val * counter->scale; - -if (nsec_counter (counter)) -nsec_printout (thread,0 , counter, uval); -else -abs_printout (thread,0 , counter, uval); +printout (thread,0 , counter, uval);if (!csv_output)print_noise (counter,1.0 );
@@ -832,11 +884,7 @@ static void print_counter_aggr(struct perf_evsel *counter, char *prefix) } uval = avg * counter->scale; - -if (nsec_counter (counter)) -nsec_printout (-1 ,0 , counter, uval); -else -abs_printout (-1 ,0 , counter, uval); +printout (-1 ,0 , counter, uval);print_noise (counter, avg);
@@ -888,11 +936,7 @@ static void print_counter(struct perf_evsel *counter, char *prefix) } uval = val * counter->scale; - -if (nsec_counter (counter)) -nsec_printout (cpu,0 , counter, uval); -else -abs_printout (cpu,0 , counter, uval); +printout (cpu,0 , counter, uval);if (!csv_output)print_noise (counter,1.0 );
tools/perf/util/stat-shadow.c +81 -73
324 lines not on this page · raw message
tools/perf/util/stat.h +8 -2
@@ -83,11 +83,17 @@ void perf_stat_evsel_id_init(struct perf_evsel *evsel);extern struct stats walltime_nsecs_stats; +typedef void (*print_metric_t )(void *ctx,const char *color,const char *unit, +const char *fmt,double val); +void perf_stat__reset_shadow_stats (void );void perf_stat__update_shadow_stats (struct perf_evsel *counter,u64 *count,int cpu); -void perf_stat__print_shadow_stats (FILE *out,struct perf_evsel *evsel, -double avg,int cpu,enum aggr_mode aggr); +void perf_stat__print_shadow_stats (struct perf_evsel *evsel, +double avg,int cpu, +print_metric_t print_metric, +void (*new_line)(void *ctx), +void *ctx);struct perf_counts *perf_counts__new (int ncpus,int nthreads);void perf_counts__delete (struct perf_counts *counts);
Signature · 1 line
2.4.3
Re: [PATCH 2/4] perf, tools, stat: Abstract stat metrics printing
On Wed, Jul 29, 2015 at 05:21:38PM -0700, Andi Kleen wrote:
SNIP
Quoted 7 lines
tools/perf/builtin-stat.c +2
-0+static void nsec_printout (int id,int nr,struct perf_evsel *evsel,double avg, +void *ctx,print_metric_t print_metric) {double msecs = avg /1e6 ;const char *fmt_v, *fmt_n;@@ -647,13 +689,16 @@ static void nsec_printout(int id, int nr, struct perf_evsel *evsel, double avg)return ;
tools/perf/builtin-stat.c static void nsec_printout(int id, int nr, struct perf_evsel *evsel, double avg) · line 696
Quoted 10 lines
tools/perf/builtin-stat.c +3
-3if (perf_evsel__match (evsel,SOFTWARE ,SW_TASK_CLOCK )) -fprintf (output," # %8.3f CPUs utilized " , - avg /avg_stats (&walltime_nsecs_stats)); +print_metric (ctx,NULL ,"%8.3f" ,"CPUs utilized" , + avg /avg_stats (&walltime_nsecs_stats));else -fprintf (output," " ); +print_metric (ctx,NULL ,NULL ,NULL ,0 ); }
hum, this should go to perf_stat__print_shadow_stats, could you please move it there before this patch?
thanks,
jirka
Re: [PATCH 2/4] perf, tools, stat: Abstract stat metrics printing
In reply to Andi KleenQuoted 28 lines · Andi Kleen
From: Andi Kleen <[email protected]>
Abstract the printing of shadow metrics. Instead of every metric calling fprintf directly and taking care of indentation, use two call backs: one to print metrics and another to start a new line.
This will allow adding metrics to CSV mode and also using them for other purposes.
The computation of padding is now done in the central callback, instead of every metric doing it manually. This makes it easier to add new metrics.
Right now there is no (intentional) behavior change, just refactoring.
Signed-off-by Andi Kleen <[email protected]>--- tools/perf/builtin-stat.c | 98 +++++++++++++++++++-------- tools/perf/util/stat-shadow.c | 154 ++++++++++++++++++++++-------------------- tools/perf/util/stat.h | 10 ++- 3 files changed, 160 insertions(+), 102 deletions(-)tools/perf/builtin-stat.c +0
-0
tools/perf/builtin-stat.c static void aggr_printout(struct perf_evsel *evsel, int id, int nr) · line 622
Quoted 7 lines
tools/perf/builtin-stat.c +3
-1@@ -617,7 +617,49 @@ static void aggr_printout(struct perf_evsel *evsel, int id, int nr) } } -static void nsec_printout (int id,int nr,struct perf_evsel *evsel,double avg) +struct outstate { +FILE *fh; +};
because we already need to make the print_metric callback global, would it be better to make this struct global, having all the needed callbacks defined within? something like:
typedef void (*perf_stat_output_metric_t)(void *ctx, const char *color, const char *unit,
const char *fmt, double val);
typedef void (*perf_stat_output_newln_t)(void *ctx);
struct perf_stat_output_ctx {
FILE *output
perf_stat_output_metric_t metric;
perf_stat_output_newln_t nl;
};not sure about the naming, but IMO should have some perf_sta_.. global form
it'd also ease the new arguments count from 3 to 1
jirka
Re: [PATCH 2/4] perf, tools, stat: Abstract stat metrics printing
Quoted 4 lines
because we already need to make the
print_metriccallback global, would it be better to make this struct global, having all the needed callbacks defined within? something like:
It's actually not global, but static.
I skipped this change. After some other changes there is only a single function call with these arguments left, so it's not an issue to pass it around.
-Andi
Re: [PATCH 2/4] perf, tools, stat: Abstract stat metrics printing
Quoted 10 lines · Andi Kleen
Quoted 4 lines
because we already need to make the
print_metriccallback global, would it be better to make this struct global, having all the needed callbacks defined within? something like:It's actually not global, but static.
I skipped this change. After some other changes there is only a single function call with these arguments left, so it's not an issue to pass it around.
hum, I dont understand.. I can se the same code in the new version:
+typedef void (*print_metric_t)(void *ctx, const char *color, const char *unit,
+ const char *fmt, double val);
+
void perf_stat__reset_shadow_stats(void);
void perf_stat__update_shadow_stats(struct perf_evsel *counter, u64 *count,
int cpu); -void perf_stat__print_shadow_stats(FILE *out, struct perf_evsel *evsel, - double avg, int cpu, enum aggr_mode aggr); +void perf_stat__print_shadow_stats(struct perf_evsel *evsel, + double avg, int cpu, + print_metric_t print_metric, + void (*new_line)(void *ctx), + void *ctx);
both:
print_metric_t print_metric
void (*new_line)(void *ctx)
are still there.. what single function call did you mean?
jirka
PATCH 3/4 perf, tools, stat: Add support for metrics in interval mode
In reply to Andi KleenFrom: Andi Kleen <[email protected]>
Now that we can modify the metrics printout functions easily, it's straight forward to support metric printing for interval mode. All that is needed is to print the time stamp on every new line. Pass the prefix into the context and print it out.
--- tools/perf/builtin-stat.c | 62 ++++++++++++++++++++++++++++++----------------- 1 file changed, 40 insertions(+), 22 deletions(-)
tools/perf/builtin-stat.c +40 -22
@@ -619,6 +619,9 @@ static void aggr_printout(struct perf_evsel *evsel, int id, int nr)struct outstate {FILE *fh; +const char *prefix; +int nfields; +u64 run, ena; };#define BASE_INDENT 41
@@ -629,13 +632,13 @@ struct outstate {static void new_line_no_aggr_std (void *ctx) {struct outstate *os = ctx; -fprintf (os->fh,"\n%*s" ,BASE_INDENT +NA_INDENT ,"" ); +fprintf (os->fh,"\n%s%-*s" , os->prefix,BASE_INDENT +NA_INDENT ,"" ); }static void new_line_std (void *ctx) {struct outstate *os = ctx; -fprintf (os->fh,"\n%-*s" ,BASE_INDENT +AGGR_INDENT ,"" ); +fprintf (os->fh,"\n%s%-*s" , os->prefix,BASE_INDENT +AGGR_INDENT ,"" ); } static void print_metric_std(void *ctx, const char *color, const char *fmt,
@@ -662,6 +665,7 @@ static void nsec_printout(int id, int nr, struct perf_evsel *evsel, double avg, void *ctx, print_metric_t print_metric) { double msecs = avg / 1e6; +double ratio;const char *fmt_v, *fmt_n;char name[25 ];
@@ -685,12 +689,13 @@ static void nsec_printout(int id, int nr, struct perf_evsel *evsel, double avg,if (evsel->cgrp)fprintf (output,"%s%s" , csv_sep, evsel->cgrp->name); -if (csv_output || interval) +if (csv_output)return ; -if (perf_evsel__match (evsel,SOFTWARE ,SW_TASK_CLOCK )) + ratio =avg_stats (&walltime_nsecs_stats); +if (perf_evsel__match (evsel,SOFTWARE ,SW_TASK_CLOCK ) && ratio)print_metric (ctx,NULL ,"%8.3f" ,"CPUs utilized" , - avg /avg_stats (&walltime_nsecs_stats)); + avg / ratio);else print_metric (ctx,NULL ,NULL ,NULL ,0 ); }
@@ -730,7 +735,7 @@ static void abs_printout(int id, int nr, struct perf_evsel *evsel, double avg,if (evsel->cgrp)fprintf (output,"%s%s" , csv_sep, evsel->cgrp->name); -if (csv_output || interval) +if (csv_output)return ; perf_stat__print_shadow_stats(evsel, avg, cpu,
@@ -739,17 +744,25 @@ static void abs_printout(int id, int nr, struct perf_evsel *evsel, double avg, ctx); } -static void printout (int id,int nr,struct perf_evsel *counter,double uval) +static void printout (int id,int nr,struct perf_evsel *counter,double uval, +char *prefix) { -struct outstate os = { .fh = output }; +struct outstate os = { + .fh = output, + .prefix = prefix ? prefix :"" + }; +print_metric_t pm = print_metric_std; +void (*nl)(void *); + +if (aggr_mode ==AGGR_NONE ) + nl = new_line_no_aggr_std; +else + nl = new_line_std;if (nsec_counter (counter)) -nsec_printout (id, nr, counter, uval, &os, print_metric_std); +nsec_printout (id, nr, counter, uval, &os, pm);else -abs_printout (id, nr, counter, uval, &os, - print_metric_std, - aggr_mode ==AGGR_NONE ? - new_line_no_aggr_std : new_line_std); +abs_printout (id, nr, counter, uval, &os, pm, nl); }static void print_aggr (char *prefix)
@@ -805,7 +818,7 @@ static void print_aggr(char *prefix)continue ; } uval = val * counter->scale; -printout (id, nr, counter, uval); +printout (id, nr, counter, uval, prefix);if (!csv_output)print_noise (counter,1.0 );
@@ -835,7 +848,7 @@ static void print_aggr_thread(struct perf_evsel *counter, char *prefix)fprintf (output,"%s" , prefix); uval = val * counter->scale; -printout (thread,0 , counter, uval); +printout (thread,0 , counter, uval, prefix);if (!csv_output)print_noise (counter,1.0 );
@@ -884,7 +897,7 @@ static void print_counter_aggr(struct perf_evsel *counter, char *prefix) } uval = avg * counter->scale; -printout (-1 ,0 , counter, uval); +printout (-1 ,0 , counter, uval, prefix);print_noise (counter, avg);
@@ -936,7 +949,7 @@ static void print_counter(struct perf_evsel *counter, char *prefix) } uval = val * counter->scale; -printout (cpu,0 , counter, uval); +printout (cpu,0 , counter, uval, prefix);if (!csv_output)print_noise (counter,1.0 );
@@ -955,20 +968,25 @@ static void print_interval(char *prefix, struct timespec *ts)if (num_print_interval ==0 && !csv_output) {switch (aggr_mode) {case AGGR_SOCKET : -fprintf (output,"# time socket cpus counts %*s events\n" , unit_width,"unit" ); +fprintf (output,"# time socket cpus counts %*s events metric multiplex\n" , + unit_width,"unit" );break ;case AGGR_CORE : -fprintf (output,"# time core cpus counts %*s events\n" , unit_width,"unit" ); +fprintf (output,"# time core cpus counts %*s events metric multiplex\n" , + unit_width,"unit" );break ;case AGGR_NONE : -fprintf (output,"# time CPU counts %*s events\n" , unit_width,"unit" ); +fprintf (output,"# time CPU counts %*s events metric multiplex\n" , + unit_width,"unit" );break ;case AGGR_THREAD : -fprintf (output,"# time comm-pid counts %*s events\n" , unit_width,"unit" ); +fprintf (output,"# time comm-pid counts %*s events metric\n" , + unit_width,"unit" );break ;case AGGR_GLOBAL :default : -fprintf (output,"# time counts %*s events\n" , unit_width,"unit" ); +fprintf (output,"# time counts %*s events metric multiplex\n" , + unit_width,"unit" ); } }
Signature · 1 line
2.4.3
Re: [PATCH 3/4] perf, tools, stat: Add support for metrics in interval mode
On Wed, Jul 29, 2015 at 05:21:39PM -0700, Andi Kleen wrote:
SNIP
Quoted 15 lines
tools/perf/builtin-stat.c +6
-3case AGGR_SOCKET : -fprintf (output,"# time socket cpus counts %*s events\n" , unit_width,"unit" ); +fprintf (output,"# time socket cpus counts %*s events metric multiplex\n" , + unit_width,"unit" );break ;case AGGR_CORE : -fprintf (output,"# time core cpus counts %*s events\n" , unit_width,"unit" ); +fprintf (output,"# time core cpus counts %*s events metric multiplex\n" , + unit_width,"unit" );break ;case AGGR_NONE : -fprintf (output,"# time CPU counts %*s events\n" , unit_width,"unit" ); +fprintf (output,"# time CPU counts %*s events metric multiplex\n" , + unit_width,"unit" );break ;
tools/perf/builtin-stat.c static void print_interval(char *prefix, struct timespec *ts) · line 989
Quoted 10 lines
tools/perf/builtin-stat.c +4
-2case AGGR_THREAD : -fprintf (output,"# time comm-pid counts %*s events\n" , unit_width,"unit" ); +fprintf (output,"# time comm-pid counts %*s events metric\n" , + unit_width,"unit" );break ;case AGGR_GLOBAL :default : -fprintf (output,"# time counts %*s events\n" , unit_width,"unit" ); +fprintf (output,"# time counts %*s events metric multiplex\n" , + unit_width,"unit" );
can't see why those header display is adjusted
jirka
PATCH 4/4 perf, tools, stat: Implement CSV metrics output
In reply to Andi KleenFrom: Andi Kleen <[email protected]>
Now support CSV output for metrics. With the new output callbacks this is relatively straight forward by creating new callbacks.
The new line callback needs to know the number of fields to skip them correctly
To avoid reordering the existing CSV fields, I had to move the noise printing into the lower level print metrics call back, so that noise can be printed before metrics.
This actually cleans up the callers because it avoids a lot of duplicated code.
--- tools/perf/builtin-stat.c | 98 +++++++++++++++++++++++++++++++++++------------ 1 file changed, 73 insertions(+), 25 deletions(-)
tools/perf/builtin-stat.c +73 -25
@@ -661,6 +661,49 @@ static void print_metric_std(void *ctx, const char *color, const char *fmt,fprintf (out," %-*s" ,METRIC_LEN - n -1 , unit); } +static void new_line_csv (void *ctx) +{ +struct outstate *os = ctx; +int i; + +fputc ('\n' , os->fh); +if (os->prefix) +fprintf (os->fh,"%s%s" , os->prefix, csv_sep); +for (i =0 ; i < os->nfields; i++) +fputs (csv_sep, os->fh); +} + +static void print_metric_csv (void *ctx, +const char *color__maybe_unused , +const char *fmt,const char *unit,double val) +{ +struct outstate *os = ctx; +FILE *out = os->fh; +char buf[64 ], *vals, *ends; + +if (unit ==NULL ) { +fprintf (out,"%s%s%s%s" , csv_sep, csv_sep, csv_sep, csv_sep); +return ; + } +fprintf (out,"%s%" PRIu64"%s%.2f%s" , + csv_sep, + os->run, + csv_sep, + os->ena ?100.0 * os->run / os->ena :100.0 , + csv_sep); +snprintf (buf,sizeof (buf), fmt, val); + vals = buf; +while (isspace (*vals)) + vals++; + ends = vals; +while (isdigit (*ends) || *ends =='.' ) + ends++; + *ends =0 ; +while (isspace (*unit)) + unit++; +fprintf (out,"%s%s%s" , vals, csv_sep, unit); +} +static void nsec_printout (int id,int nr,struct perf_evsel *evsel,double avg,void *ctx,print_metric_t print_metric) {
@@ -735,9 +778,6 @@ static void abs_printout(int id, int nr, struct perf_evsel *evsel, double avg,if (evsel->cgrp)fprintf (output,"%s%s" , csv_sep, evsel->cgrp->name); -if (csv_output) -return ; - perf_stat__print_shadow_stats(evsel, avg, cpu, print_metric, new_line,
@@ -745,7 +785,7 @@ static void abs_printout(int id, int nr, struct perf_evsel *evsel, double avg, }static void printout (int id,int nr,struct perf_evsel *counter,double uval, -char *prefix) +char *prefix,u64 run,u64 ena,double noise) {struct outstate os = { .fh = output,
@@ -759,10 +799,35 @@ static void printout(int id, int nr, struct perf_evsel *counter, double uval,else nl = new_line_std; +if (csv_output) { +static int aggr_fields[] = { + [AGGR_GLOBAL ] =0 , + [AGGR_THREAD ] =1 , + [AGGR_NONE ] =1 , + [AGGR_SOCKET ] =2 , + [AGGR_CORE ] =2 , + }; + + pm = print_metric_csv; + nl = new_line_csv; + os.nfields =1 ; + os.nfields += aggr_fields[aggr_mode]; +if (counter->cgrp) + os.nfields++; + os.run = run; + os.ena = ena; + } +if (nsec_counter (counter))nsec_printout (id, nr, counter, uval, &os, pm);else abs_printout (id, nr, counter, uval, &os, pm, nl); + +if (!csv_output) { +print_noise (counter, noise); +if (run != ena) +fprintf (output," (%.2f%%)" ,100.0 * run / ena); + } }static void print_aggr (char *prefix)
@@ -818,11 +883,7 @@ static void print_aggr(char *prefix)continue ; } uval = val * counter->scale; -printout (id, nr, counter, uval, prefix); -if (!csv_output) -print_noise (counter,1.0 ); - -print_running (run, ena); +printout (id, nr, counter, uval, prefix, run, ena,1.0 );fputc ('\n' , output); } }
@@ -848,12 +909,7 @@ static void print_aggr_thread(struct perf_evsel *counter, char *prefix)fprintf (output,"%s" , prefix); uval = val * counter->scale; -printout (thread,0 , counter, uval, prefix); - -if (!csv_output) -print_noise (counter,1.0 ); - -print_running (run, ena); +printout (thread,0 , counter, uval, prefix, run, ena,1.0 );fputc ('\n' , output); } }
@@ -897,11 +953,7 @@ static void print_counter_aggr(struct perf_evsel *counter, char *prefix) } uval = avg * counter->scale; -printout (-1 ,0 , counter, uval, prefix); - -print_noise (counter, avg); - -print_running (avg_running, avg_enabled); +printout (-1 ,0 , counter, uval, prefix, avg_running, avg_enabled, avg);fprintf (output,"\n" ); }
@@ -949,11 +1001,7 @@ static void print_counter(struct perf_evsel *counter, char *prefix) } uval = val * counter->scale; -printout (cpu,0 , counter, uval, prefix); - -if (!csv_output) -print_noise (counter,1.0 ); -print_running (run, ena); +printout (cpu,0 , counter, uval, prefix, run, ena,1.0 );fputc ('\n' , output); }
Signature · 1 line
2.4.3
Re: [PATCH 4/4] perf, tools, stat: Implement CSV metrics output
On Wed, Jul 29, 2015 at 05:21:40PM -0700, Andi Kleen wrote:
SNIP
Quoted 45 lines
tools/perf/builtin-stat.c +4
-16+ } }static void print_aggr (char *prefix)@@ -818,11 +883,7 @@ static void print_aggr(char *prefix)continue ; } uval = val * counter->scale; -printout (id, nr, counter, uval, prefix); -if (!csv_output) -print_noise (counter,1.0 ); - -print_running (run, ena); +printout (id, nr, counter, uval, prefix, run, ena,1.0 );fputc ('\n' , output); } }@@ -848,12 +909,7 @@ static void print_aggr_thread(struct perf_evsel *counter, char *prefix)fprintf (output,"%s" , prefix); uval = val * counter->scale; -printout (thread,0 , counter, uval, prefix); - -if (!csv_output) -print_noise (counter,1.0 ); - -print_running (run, ena); +printout (thread,0 , counter, uval, prefix, run, ena,1.0 );fputc ('\n' , output); } }@@ -897,11 +953,7 @@ static void print_counter_aggr(struct perf_evsel *counter, char *prefix) } uval = avg * counter->scale; -printout (-1 ,0 , counter, uval, prefix); - -print_noise (counter, avg); - -print_running (avg_running, avg_enabled); +printout (-1 ,0 , counter, uval, prefix, avg_running, avg_enabled, avg);fprintf (output,"\n" ); }
tools/perf/builtin-stat.c static void print_counter(struct perf_evsel *counter, char *prefix) · line 1004
Quoted 9 lines
tools/perf/builtin-stat.c +1
-5@@ -949,11 +1001,7 @@ static void print_counter(struct perf_evsel *counter, char *prefix) } uval = val * counter->scale; -printout (cpu,0 , counter, uval, prefix); - -if (!csv_output) -print_noise (counter,1.0 ); -print_running (run, ena); +printout (cpu,0 , counter, uval, prefix, run, ena,1.0 );
print_running still stayed in print_aggr
how about moving the (run == 0 || ena == 0) leg from print_aggr into printout
jirka
Re: [PATCH 4/4] perf, tools, stat: Implement CSV metrics output
In reply to Andi KleenQuoted 8 lines · Andi Kleen
From: Andi Kleen <[email protected]>
Now support CSV output for metrics. With the new output callbacks this is relatively straight forward by creating new callbacks.
The new line callback needs to know the number of fields to skip them correctly
Quoted 6 lines
To avoid reordering the existing CSV fields, I had to move the noise printing into the lower level print metrics call back, so that noise can be printed before metrics.
This actually cleans up the callers because it avoids a lot of duplicated code.
nice, please separate the cleanup from the actual addition of the metrics CSV fields
could you please also start some documentation in perf stat man? I dont think we have anything yet.. and it's getting bigger ;-)
thanks,
jirka