perf, tools: Refactor and support interval and CSV metrics v3
[v3: Addressed all review feedback. Update manpage for CSV. Various changes
(see individual patches). Remove some more redundant code in printout callers.]
[v2: Addressed (near) all review feedback. No manpage updates so far.
Add support for --per-core metrics. Various cleanups.]
[v3: Everything compiles again. Some more cleanups.]
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-2.6 perf/stat-metrics-5
PATCH 1/9 perf, tools, stat: Move sw clock metrics printout to stat-shadow
From: Andi Kleen <[email protected]>
The sw clock metrics printing was missed in the earlier move to stat-shadow of all the other metric printouts. Move it too.
--- tools/perf/builtin-stat.c | 9 --------- tools/perf/util/stat-shadow.c | 3 +++ 2 files changed, 3 insertions(+), 9 deletions(-)
tools/perf/builtin-stat.c +0 -9
@@ -510,15 +510,6 @@ 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 || stat_config.interval) -return ; - -if (perf_evsel__match (evsel,SOFTWARE ,SW_TASK_CLOCK )) -fprintf (output," # %8.3f CPUs utilized " , - avg /avg_stats (&walltime_nsecs_stats)); -else -fprintf (output," " ); }static void abs_printout (int id,int nr,struct perf_evsel *evsel,double avg)
tools/perf/util/stat-shadow.c +3 -0
@@ -413,6 +413,9 @@ void perf_stat__print_shadow_stats(FILE *out, struct perf_evsel *evsel, ratio = total / avg;fprintf (out," # %8.0f cycles / elision " , ratio); + }else if (perf_evsel__match (evsel,SOFTWARE ,SW_TASK_CLOCK ) && + (ratio =avg_stats (&walltime_nsecs_stats)) !=0 ) { +fprintf (out," # %8.3f CPUs utilized " , avg / ratio); }else if (runtime_nsecs_stats[cpu].n !=0 ) {char unit ='M' ;
Signature · 1 line
2.4.3
Re: [PATCH 1/9] perf, tools, stat: Move sw clock metrics printout to stat-shadow
Quoted 36 lines · Andi Kleen
From: Andi Kleen <[email protected]>
The sw clock metrics printing was missed in the earlier move to stat-shadow of all the other metric printouts. Move it too.
Signed-off-by Andi Kleen <[email protected]>--- tools/perf/builtin-stat.c | 9 --------- tools/perf/util/stat-shadow.c | 3 +++ 2 files changed, 3 insertions(+), 9 deletions(-)tools/perf/builtin-stat.c +0
-9@@ -510,15 +510,6 @@ 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 || stat_config.interval) -return ; - -if (perf_evsel__match (evsel,SOFTWARE ,SW_TASK_CLOCK )) -fprintf (output," # %8.3f CPUs utilized " , - avg /avg_stats (&walltime_nsecs_stats)); -else -fprintf (output," " ); }static void abs_printout (int id,int nr,struct perf_evsel *evsel,double avg)tools/perf/util/stat-shadow.c +0
-0
tools/perf/util/stat-shadow.c void perf_stat__print_shadow_stats(FILE *out, struct perf_evsel *evsel, · line 418
Quoted 6 lines
tools/perf/util/stat-shadow.c +3
-0@@ -413,6 +413,9 @@ void perf_stat__print_shadow_stats(FILE *out, struct perf_evsel *evsel, ratio = total / avg;fprintf (out," # %8.0f cycles / elision " , ratio); + }else if (perf_evsel__match (evsel,SOFTWARE ,SW_TASK_CLOCK ) && + (ratio =avg_stats (&walltime_nsecs_stats)) !=0 ) { +fprintf (out," # %8.3f CPUs utilized " , avg / ratio);
perf_stat__print_shadow_stats is at this point called only from abs_printout, while this is nsec_printout.. but it's all factored later, so I believe this is ok for now
thanks,
jirka
PATCH 2/9 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.
v2: Refactor functions, printout now does more. Move shadow printing. Improve fallback callbacks. Don't use void * callback data.
--- tools/perf/builtin-stat.c | 102 +++++++++++++------- tools/perf/util/stat-shadow.c | 212 +++++++++++++++++++++++------------------- tools/perf/util/stat.h | 14 ++- 3 files changed, 196 insertions(+), 132 deletions(-)
tools/perf/builtin-stat.c +69 -33
@@ -484,6 +484,47 @@ static void aggr_printout(struct perf_evsel *evsel, int id, int nr) } } +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) {FILE *output = stat_config.output;
@@ -517,7 +558,6 @@ static void abs_printout(int id, int nr, struct perf_evsel *evsel, double avg)FILE *output = stat_config.output;double sc = evsel->scale;const char *fmt; -int cpu =cpu_map__id_to_cpu (id);if (csv_output) { fmt = sc !=1.0 ?"%.2f%s" :"%.0f%s" ;
@@ -530,9 +570,6 @@ static void abs_printout(int id, int nr, struct perf_evsel *evsel, double avg)aggr_printout (evsel, id, nr); -if (stat_config.aggr_mode ==AGGR_GLOBAL ) - cpu =0 ; -fprintf (output, fmt, avg, csv_sep);if (evsel->unit)
@@ -544,12 +581,32 @@ 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 || stat_config.interval) -return ; +static void printout (int id,int nr,struct perf_evsel *counter,double uval) +{ +struct outstate os = { .fh = stat_config.output }; +struct perf_stat_output_ctx out; +print_metric_t pm = print_metric_std; +void (*nl)(void *); -perf_stat__print_shadow_stats (output, evsel, avg, cpu, - stat_config.aggr_mode); +if (stat_config.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); +else +abs_printout (id, nr, counter, uval); + + out.print_metric = pm; + out.new_line = nl; + out.ctx = &os; +perf_stat__print_shadow_stats (counter, uval, + stat_config.aggr_mode ==AGGR_GLOBAL ?0 : +cpu_map__id_to_cpu (id), + &out); }static void print_aggr (char *prefix)
@@ -605,12 +662,7 @@ 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 );
@@ -641,11 +693,7 @@ 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 );
@@ -695,11 +743,7 @@ 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);
@@ -752,15 +796,7 @@ 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); - -if (!csv_output) -print_noise (counter,1.0 ); -print_running (run, ena); +printout (cpu,0 , counter, uval);fputc ('\n' , output); }
tools/perf/util/stat-shadow.c +115 -97
366 lines not on this page · raw message
tools/perf/util/stat.h +12 -2
@@ -67,11 +67,21 @@ 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); +struct perf_stat_output_ctx { +void *ctx; +print_metric_t print_metric; +void (*new_line)(void *ctx); +}; + +void perf_stat__print_shadow_stats (struct perf_evsel *evsel, +double avg,int cpu, +struct perf_stat_output_ctx *out);void perf_evsel__reset_stat_priv (struct perf_evsel *evsel);int perf_evsel__alloc_stat_priv (struct perf_evsel *evsel);
Signature · 1 line
2.4.3
Re: [PATCH 2/9] perf, tools, stat: Abstract stat metrics printing
On Wed, Sep 02, 2015 at 04:35:47PM -0700, Andi Kleen wrote:
SNIP
Quoted 4 lines
tools/perf/builtin-stat.c +0
-0@@ -752,15 +796,7 @@ static void print_counter(struct perf_evsel *counter, char *prefix) }
tools/perf/builtin-stat.c static void print_counter(struct perf_evsel *counter, char *prefix) · line 763
Quoted 10 lines
tools/perf/builtin-stat.c +0
-9uval = val * counter->scale; - -if (nsec_counter (counter)) -nsec_printout (cpu,0 , counter, uval); -else -abs_printout (cpu,0 , counter, uval); - -if (!csv_output) -print_noise (counter,1.0 ); -print_running (run, ena);
why removing print_running/print_noise now?
jirka
Re: [PATCH 2/9] perf, tools, stat: Abstract stat metrics printing
In reply to Andi KleenOn Wed, Sep 02, 2015 at 04:35:47PM -0700, Andi Kleen wrote:
SNIP
Quoted 10 lines
tools/perf/util/stat.h +2
-0@@ -67,11 +67,21 @@ 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);
tools/perf/util/stat.h void perf_stat_evsel_id_init(struct perf_evsel *evsel); · line 79
Quoted 10 lines
tools/perf/util/stat.h +5
-2+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); +struct perf_stat_output_ctx { +void *ctx; +print_metric_t print_metric; +void (*new_line)(void *ctx);
please add typedef for new_line as well
jirka
PATCH 3/9 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 | 21 +++++++++++++-------- 1 file changed, 13 insertions(+), 8 deletions(-)
tools/perf/builtin-stat.c +13 -8
@@ -486,6 +486,7 @@ static void aggr_printout(struct perf_evsel *evsel, int id, int nr)struct outstate {FILE *fh; +const char *prefix; };#define BASE_INDENT 41
@@ -496,13 +497,13 @@ 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,
@@ -583,10 +584,14 @@ static void abs_printout(int id, int nr, struct perf_evsel *evsel, double avg)fprintf (output,"%s%s" , csv_sep, evsel->cgrp->name); } -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 = stat_config.output };struct perf_stat_output_ctx out; +struct outstate os = { + .fh = stat_config.output, + .prefix = prefix ? prefix :"" + };print_metric_t pm = print_metric_std;void (*nl)(void *);
@@ -662,7 +667,7 @@ 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 );
@@ -693,7 +698,7 @@ 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 );
@@ -743,7 +748,7 @@ 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);
@@ -796,7 +801,7 @@ 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);fputc ('\n' , output); }
Signature · 1 line
2.4.3
PATCH 4/9 perf, tools, stat: Move noise/running printing into printout
In reply to Andi KleenFrom: Andi Kleen <[email protected]>
Move the running/noise printing into printout to avoid duplicated code in the callers.
--- tools/perf/builtin-stat.c | 29 +++++++++++------------------ 1 file changed, 11 insertions(+), 18 deletions(-)
tools/perf/builtin-stat.c +11 -18
@@ -585,7 +585,7 @@ 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 perf_stat_output_ctx out;struct outstate os = {
@@ -612,6 +612,12 @@ static void printout(int id, int nr, struct perf_evsel *counter, double uval, stat_config.aggr_mode ==AGGR_GLOBAL ?0 :cpu_map__id_to_cpu (id), &out); + +if (!csv_output) { +print_noise (counter, noise); +if (run != ena) +fprintf (stat_config.output," (%.2f%%)" ,100.0 * run / ena); + } }static void print_aggr (char *prefix)
@@ -667,11 +673,7 @@ 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); } }
@@ -698,12 +700,7 @@ 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); } }
@@ -748,11 +745,7 @@ 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" ); }
@@ -801,7 +794,7 @@ static void print_counter(struct perf_evsel *counter, char *prefix) } uval = val * counter->scale; -printout (cpu,0 , counter, uval, prefix); +printout (cpu,0 , counter, uval, prefix, run, ena,1.0 );fputc ('\n' , output); }
Signature · 1 line
2.4.3
PATCH 5/9 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
v2: Split out function argument changes
--- tools/perf/builtin-stat.c | 64 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 64 insertions(+)
tools/perf/builtin-stat.c +64 -0
@@ -487,6 +487,8 @@ 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
@@ -526,6 +528,49 @@ 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) {FILE *output = stat_config.output;
@@ -600,6 +645,25 @@ 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[stat_config.aggr_mode]; +if (counter->cgrp) + os.nfields++; + os.run = run; + os.ena = ena; + } +if (nsec_counter (counter))nsec_printout (id, nr, counter, uval);else
Signature · 1 line
2.4.3
PATCH 6/9 perf, tools, stat: Move non counting counter printing to printout
In reply to Andi KleenFrom: Andi Kleen <[email protected]>
Move the special case printing for non-running counters to printout, so it can be shared by all the output options.
--- tools/perf/builtin-stat.c | 73 ++++++++++++++++------------------------------- 1 file changed, 24 insertions(+), 49 deletions(-)
tools/perf/builtin-stat.c +24 -49
@@ -664,6 +664,30 @@ static void printout(int id, int nr, struct perf_evsel *counter, double uval, os.ena = ena; } +if (run ==0 || ena ==0 ) { +aggr_printout (counter, id, nr); + +fprintf (stat_config.output,"%*s%s" , + csv_output ?0 :18 , + counter->supported ?CNTR_NOT_COUNTED :CNTR_NOT_SUPPORTED , + csv_sep); + +fprintf (stat_config.output,"%-*s%s" , + csv_output ?0 : unit_width, + counter->unit, csv_sep); + +fprintf (stat_config.output,"%*s" , + csv_output ?0 :-25 , +perf_evsel__name (counter)); + +if (counter->cgrp) +fprintf (stat_config.output,"%s%s" , + csv_sep, counter->cgrp->name); + +print_running (run, ena); +return ; + } +if (nsec_counter (counter))nsec_printout (id, nr, counter, uval);else
@@ -712,30 +736,6 @@ static void print_aggr(char *prefix)if (prefix)fprintf (output,"%s" , prefix); -if (run ==0 || ena ==0 ) { -aggr_printout (counter, id, nr); - -fprintf (output,"%*s%s" , - csv_output ?0 :18 , - counter->supported ?CNTR_NOT_COUNTED :CNTR_NOT_SUPPORTED , - csv_sep); - -fprintf (output,"%-*s%s" , - csv_output ?0 : unit_width, - counter->unit, csv_sep); - -fprintf (output,"%*s" , - csv_output ?0 :-25 , -perf_evsel__name (counter)); - -if (counter->cgrp) -fprintf (output,"%s%s" , - csv_sep, counter->cgrp->name); - -print_running (run, ena); -fputc ('\n' , output); -continue ; - } uval = val * counter->scale;printout (id, nr, counter, uval, prefix, run, ena,1.0 );fputc ('\n' , output);
@@ -832,31 +832,6 @@ static void print_counter(struct perf_evsel *counter, char *prefix)if (prefix)fprintf (output,"%s" , prefix); -if (run ==0 || ena ==0 ) { -fprintf (output,"CPU%*d%s%*s%s" , - csv_output ?0 :-4 , -perf_evsel__cpus (counter)->map[cpu], csv_sep, - csv_output ?0 :18 , - counter->supported ?CNTR_NOT_COUNTED :CNTR_NOT_SUPPORTED , - csv_sep); - -fprintf (output,"%-*s%s" , - csv_output ?0 : unit_width, - counter->unit, csv_sep); - -fprintf (output,"%*s" , - csv_output ?0 :-25 , -perf_evsel__name (counter)); - -if (counter->cgrp) -fprintf (output,"%s%s" , - csv_sep, counter->cgrp->name); - -print_running (run, ena); -fputc ('\n' , output); -continue ; - } - uval = val * counter->scale;printout (cpu,0 , counter, uval, prefix, run, ena,1.0 );
Signature · 1 line
2.4.3
PATCH 7/9 perf, tools: Support metrics in --per-core/socket mode
In reply to Andi KleenFrom: Andi Kleen <[email protected]>
Enable metrics printing in --per-core / --per-socket mode. We need to save the shadow metrics in a unique place. Always use the first CPU in the aggregation. Then use the same CPU to retrieve the shadow value later.
Example output:
% perf stat --per-core -a ./BC1s
Performance counter stats for 'system wide':
S0-C0 2 2966.020381 task-clock (msec) # 2.004 CPUs utilized (100.00%) S0-C0 2 49 context-switches # 0.017 K/sec (100.00%) S0-C0 2 4 cpu-migrations # 0.001 K/sec (100.00%) S0-C0 2 467 page-faults # 0.157 K/sec S0-C0 2 4,599,061,773 cycles # 1.551 GHz (100.00%) S0-C0 2 9,755,886,883 instructions # 2.12 insn per cycle (100.00%) S0-C0 2 1,906,272,125 branches # 642.704 M/sec (100.00%) S0-C0 2 81,180,867 branch-misses # 4.26% of all branches S0-C1 2 2965.995373 task-clock (msec) # 2.003 CPUs utilized (100.00%) S0-C1 2 62 context-switches # 0.021 K/sec (100.00%) S0-C1 2 8 cpu-migrations # 0.003 K/sec (100.00%) S0-C1 2 281 page-faults # 0.095 K/sec S0-C1 2 6,347,290 cycles # 0.002 GHz (100.00%) S0-C1 2 4,654,156 instructions # 0.73 insn per cycle (100.00%) S0-C1 2 947,121 branches # 0.319 M/sec (100.00%) S0-C1 2 37,322 branch-misses # 3.94% of all branches
1.480409747 seconds time elapsed
--- tools/perf/builtin-stat.c | 63 +++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 58 insertions(+), 5 deletions(-)
tools/perf/builtin-stat.c +58 -5
@@ -489,6 +489,8 @@ struct outstate {const char *prefix;int nfields;u64 run, ena; +int id, nr; +struct perf_evsel *evsel; };#define BASE_INDENT 41
@@ -499,13 +501,19 @@ struct outstate {static void new_line_no_aggr_std (void *ctx) {struct outstate *os = ctx; -fprintf (os->fh,"\n%s%-*s" , os->prefix,BASE_INDENT +NA_INDENT ,"" ); + +fprintf (os->fh,"\n%s" , os->prefix); +aggr_printout (os->evsel, os->id, os->nr); +fprintf (os->fh,"%-*s" ,BASE_INDENT +NA_INDENT ,"" ); }static void new_line_std (void *ctx) {struct outstate *os = ctx; -fprintf (os->fh,"\n%s%-*s" , os->prefix,BASE_INDENT +AGGR_INDENT ,"" ); + +fprintf (os->fh,"\n%s" , os->prefix); +aggr_printout (os->evsel, os->id, os->nr); +fprintf (os->fh,"%-*s" ,BASE_INDENT +AGGR_INDENT ,"" ); } static void print_metric_std(void *ctx, const char *color, const char *fmt,
@@ -536,6 +544,7 @@ static void new_line_csv(void *ctx)fputc ('\n' , os->fh);if (os->prefix)fprintf (os->fh,"%s%s" , os->prefix, csv_sep); +aggr_printout (os->evsel, os->id, os->nr);for (i =0 ; i < os->nfields; i++)fputs (csv_sep, os->fh); }
@@ -599,6 +608,22 @@ static void nsec_printout(int id, int nr, struct perf_evsel *evsel, double avg)fprintf (output,"%s%s" , csv_sep, evsel->cgrp->name); } +static int first_shadow_cpu (struct perf_evsel *evsel,int id) +{ +int i; + +if (aggr_get_id ==NULL ) +return 0 ; + +for (i =0 ; i <perf_evsel__nr_cpus (evsel); i++) { +int cpu2 =perf_evsel__cpus (evsel)->map[i]; + +if (aggr_get_id (evsel_list->cpus, cpu2) == id) +return cpu2; + } +return 0 ; +} +static void abs_printout (int id,int nr,struct perf_evsel *evsel,double avg) {FILE *output = stat_config.output;
@@ -635,7 +660,10 @@ static void printout(int id, int nr, struct perf_evsel *counter, double uval,struct perf_stat_output_ctx out;struct outstate os = { .fh = stat_config.output, - .prefix = prefix ? prefix :"" + .prefix = prefix ? prefix :"" , + .id = id, + .nr = nr, + .evsel = counter, };print_metric_t pm = print_metric_std;void (*nl)(void *);
@@ -698,9 +726,8 @@ static void printout(int id, int nr, struct perf_evsel *counter, double uval, out.ctx = &os;perf_stat__print_shadow_stats (counter, uval, stat_config.aggr_mode ==AGGR_GLOBAL ?0 : -cpu_map__id_to_cpu (id), +first_shadow_cpu (counter, id), &out); -if (!csv_output) {print_noise (counter, noise);if (run != ena)
@@ -708,6 +735,30 @@ static void printout(int id, int nr, struct perf_evsel *counter, double uval, } } +static void aggr_update_shadow (void ) +{ +int cpu, cpu2, s2, id, s; +u64 val; +struct perf_evsel *counter; + +for (s =0 ; s < aggr_map->nr; s++) { + id = aggr_map->map[s]; +evlist__for_each (evsel_list, counter) { + val =0 ; +for (cpu =0 ; cpu <perf_evsel__nr_cpus (counter); cpu++) { + cpu2 =perf_evsel__cpus (counter)->map[cpu]; + s2 =aggr_get_id (evsel_list->cpus, cpu2); +if (s2 != id) +continue ; + val +=perf_counts (counter->counts, cpu,0 )->val; + } + val = val * counter->scale; +perf_stat__update_shadow_stats (counter, &val, +first_shadow_cpu (counter, id)); + } + } +} +static void print_aggr (char *prefix) {FILE *output = stat_config.output;
@@ -719,6 +770,8 @@ static void print_aggr(char *prefix)if (!(aggr_map || aggr_get_id))return ; +aggr_update_shadow (); +for (s =0 ; s < aggr_map->nr; s++) { id = aggr_map->map[s];evlist__for_each (evsel_list, counter) {
Signature · 1 line
2.4.3
PATCH 8/9 perf, tools, stat: Remove extra unsupported/not counted handling
In reply to Andi KleenFrom: Andi Kleen <[email protected]>
printout now handles unsupported/not counted printing, so we can remove one extra instance in the aggregated printing.
--- tools/perf/builtin-stat.c | 22 ++-------------------- 1 file changed, 2 insertions(+), 20 deletions(-)
tools/perf/builtin-stat.c +2 -20
@@ -841,26 +841,8 @@ static void print_counter_aggr(struct perf_evsel *counter, char *prefix)if (prefix)fprintf (output,"%s" , prefix); -if (scaled ==-1 || !counter->supported) { -fprintf (output,"%*s%s" , - csv_output ?0 :18 , - counter->supported ?CNTR_NOT_COUNTED :CNTR_NOT_SUPPORTED , - csv_sep); -fprintf (output,"%-*s%s" , - csv_output ?0 : unit_width, - counter->unit, csv_sep); -fprintf (output,"%*s" , - csv_output ?0 :-25 , -perf_evsel__name (counter)); - -if (counter->cgrp) -fprintf (output,"%s%s" , csv_sep, counter->cgrp->name); - -print_running (avg_running, avg_enabled); -fputc ('\n' , output); -return ; - } - +if (scaled ==-1 ) + avg_running = avg_enabled =0 ; uval = avg * counter->scale;printout (-1 ,0 , counter, uval, prefix, avg_running, avg_enabled, avg);fprintf (output,"\n" );
Signature · 1 line
2.4.3
Re: [PATCH 8/9] perf, tools, stat: Remove extra unsupported/not counted handling
Quoted 30 lines · Andi Kleen
From: Andi Kleen <[email protected]>
printout now handles unsupported/not counted printing, so we can remove one extra instance in the aggregated printing.
Signed-off-by Andi Kleen <[email protected]>--- tools/perf/builtin-stat.c | 22 ++-------------------- 1 file changed, 2 insertions(+), 20 deletions(-)tools/perf/builtin-stat.c +0
-12@@ -841,26 +841,8 @@ static void print_counter_aggr(struct perf_evsel *counter, char *prefix)if (prefix)fprintf (output,"%s" , prefix); -if (scaled ==-1 || !counter->supported) { -fprintf (output,"%*s%s" , - csv_output ?0 :18 , - counter->supported ?CNTR_NOT_COUNTED :CNTR_NOT_SUPPORTED , - csv_sep); -fprintf (output,"%-*s%s" , - csv_output ?0 : unit_width, - counter->unit, csv_sep); -fprintf (output,"%*s" , - csv_output ?0 :-25 , -perf_evsel__name (counter)); -
tools/perf/builtin-stat.c static void print_counter_aggr(struct perf_evsel *counter, char *prefix) · line 845
Quoted 10 lines
tools/perf/builtin-stat.c +2
-8-if (counter->cgrp) -fprintf (output,"%s%s" , csv_sep, counter->cgrp->name); - -print_running (avg_running, avg_enabled); -fputc ('\n' , output); -return ; - } - +if (scaled ==-1 ) + avg_running = avg_enabled =0 ;
why the extra initialization?
jirka
Quoted 6 lines
tools/perf/builtin-stat.c +0
-0uval = avg * counter->scale;printout (-1 ,0 , counter, uval, prefix, avg_running, avg_enabled, avg);fprintf (output,"\n" );Signature · 1 line
2.4.3
Re: [PATCH 8/9] perf, tools, stat: Remove extra unsupported/not counted handling
In reply to Andi KleenQuoted 27 lines · Andi Kleen
From: Andi Kleen <[email protected]>
printout now handles unsupported/not counted printing, so we can remove one extra instance in the aggregated printing.
Signed-off-by Andi Kleen <[email protected]>--- tools/perf/builtin-stat.c | 22 ++-------------------- 1 file changed, 2 insertions(+), 20 deletions(-)tools/perf/builtin-stat.c +0
-9@@ -841,26 +841,8 @@ static void print_counter_aggr(struct perf_evsel *counter, char *prefix)if (prefix)fprintf (output,"%s" , prefix); -if (scaled ==-1 || !counter->supported) { -fprintf (output,"%*s%s" , - csv_output ?0 :18 , - counter->supported ?CNTR_NOT_COUNTED :CNTR_NOT_SUPPORTED , - csv_sep); -fprintf (output,"%-*s%s" , - csv_output ?0 : unit_width, - counter->unit, csv_sep); - fprintf(output, "%*s",
tools/perf/builtin-stat.c static void print_counter_aggr(struct perf_evsel *counter, char *prefix) · line 862
Quoted 10 lines
tools/perf/builtin-stat.c +0
-10- csv_output ?0 :-25 , - perf_evsel__name(counter)); - -if (counter->cgrp) -fprintf (output,"%s%s" , csv_sep, counter->cgrp->name); - -print_running (avg_running, avg_enabled); -fputc ('\n' , output); -return ; - }
why is this not part of the:
perf, tools, stat: Move non counting counter printing to printout
jirka
PATCH 9/9 perf, tools, stat: Document CSV format in manpage
In reply to Andi KleenFrom: Andi Kleen <[email protected]>
With all the recently added fields in the perf stat CSV output we should finally document them in the man page. Do this here.
--- tools/perf/Documentation/perf-stat.txt | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+)
tools/perf/Documentation/perf-stat.txt +19 -0
@@ -176,6 +176,25 @@ $ perf stat -- make -j Wall-clock time elapsed: 719.554352 msecs +CSV FORMAT +---------- + +With -x, perf stat is able to output a not-quite-CSV format output +(commas in the output are not put into "") + +The fields are in this order: + + - optional usec time stamp in fractions of second (with -I xxx) + - unit of the counter value or empty + - event name + - counter value + - run time of counter + - percentage of measurement time the counter was running + - optional metric value + - optional unit of metric + +Additional metrics may be printed with all earlier fields being empty. + SEE ALSO -------- linkperf:perf-top[1], linkperf:perf-list[1]
Signature · 1 line
2.4.3
Quoted 49 lines · Andi Kleen
[v3: Addressed all review feedback. Update manpage for CSV. Various changes
(see individual patches). Remove some more redundant code in printout callers.]
[v2: Addressed (near) all review feedback. No manpage updates so far.
Add support for --per-core metrics. Various cleanups.]
[v3: Everything compiles again. Some more cleanups.]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 branchesCSV 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 branches
Quoted 2 lines
Available in
git://git.kernel.org/pub/scm/linux/kernel/git/ak/linux-misc-2.6perf/stat-metrics-5
can't see perf/stat-metrics-5 branch
[jolsa@krava perf]$ git remote update ak
Fetching ak
[jolsa@krava perf]$ git branch -r | grep perf/stat-metrics-
ak/perf/stat-metrics-3
ak/perf/stat-metrics-4 ak/perf/stat-metrics-refactor-2
[jolsa@krava perf]$
jirka
Quoted 9 lines
can't see perf/stat-metrics-5 branch
[jolsa@krava perf]$ git remote update ak
Fetching ak
[jolsa@krava perf]$ git branch -r | grep perf/stat-metrics-
ak/perf/stat-metrics-3
ak/perf/stat-metrics-4 ak/perf/stat-metrics-refactor-2
[jolsa@krava perf]$
Try again. I pushed now.
-Andi