perf, tools: Support spark lines in perf stat v5
PATCH 1/1 perf, tools: Support spark lines in perf stat v5
From: Andi Kleen <[email protected]>
perf stat -rX prints the stddev for multiple measurements. Just looking at the stddev for judging the quality of the data is a bit dangerous The simplest sanity check is to just look at a simple plot. This patchs add a sparkline to the end of the measurements to make it simple to judge the data.
The sparkline only uses UTF-8, so should be readable in all modern tools and terminals.
The sparkline is between the minimum and maximum of the data, so it's mainly a indicator of variance. To keep the code simple and make the output not too wide only the first 8 values are printed. If more values are there it adds '..'
The code is inspired by Zach Holman's spark shell script.
Example output (view in non-proportial font):
Performance counter stats for 'true' (10 runs):
0.175672 task-clock (msec) # 0.555 CPUs utilized ( +- 1.77% ) █▄▁▁▁▁▁▁..
0 context-switches # 0.000 K/sec
0 cpu-migrations # 0.000 K/sec
114 page-faults # 0.647 M/sec ( +- 0.14% ) ▁█▁▁████..
520,798 cycles # 2.965 GHz ( +- 1.75% ) █▄▁▁▁▁▁▁..
433,525 instructions # 0.83 insns per cycle ( +- 0.28% ) ▅▇▅▄▇█▁▆..
83,012 branches # 472.537 M/sec ( +- 0.31% ) ▅▇▆▄▇█▁▆..
3,157 branch-misses # 3.80% of all branches ( +- 2.55% ) ▇█▃▅▁▃▁▂..0.000316660 seconds time elapsed ( +- 1.78% ) █▅▁▁▁▁▁▁..
As you can see even in the most simple run there are quite interesting patterns. The time sparkline suggests it would be also useful to have an option to throw the first measurement away.
Known issues:
- Makes the perf stat output wider. Could be adjust by shrinking some white space. Not done so far.
- No output for -A/--per-socket/--per-core with -rX. This code is missing the basic noise detection code. Once it's added there sparklines could be shown too.
v2: Avoid printing spark lines for normal CSV case (Jiri)
v3: LONG->ULONG, random changes
v4: Add some missing changes from the forked v2: checks value is not zero instead of all the same. Update documentation. Remove n variable. Remove #pragma once
v5: Add overflow fixes from Jiri.
--- tools/perf/Documentation/perf-stat.txt | 4 ++++ tools/perf/Makefile.perf | 1 + tools/perf/builtin-stat.c | 12 ++++++++++++ tools/perf/util/spark.c | 31 +++++++++++++++++++++++++++++++ tools/perf/util/spark.h | 4 ++++ tools/perf/util/stat.c | 34 ++++++++++++++++++++++++++++++++++ tools/perf/util/stat.h | 10 ++++++++++ 7 files changed, 96 insertions(+) create mode 100644 tools/perf/util/spark.c create mode 100644 tools/perf/util/spark.h
tools/perf/Documentation/perf-stat.txt +4 -0
@@ -53,6 +53,10 @@ OPTIONS -r:: --repeat=<n>:: repeat command and print average + stddev (max: 100). 0 means forever. + In addition it prints a spark line (when not in CSV mode), which visualizes the + variance between minimum and maximum of the measurements. This allows a simple sanity + check of the measurements. Only 8 values are printed, when more are available + it adds .. -B:: --big-num::
tools/perf/Makefile.perf +1 -0
@@ -359,6 +359,7 @@ LIB_OBJS += $(OUTPUT)util/trace-event-scripting.o LIB_OBJS += $(OUTPUT)util/trace-event.o LIB_OBJS += $(OUTPUT)util/svghelper.o LIB_OBJS += $(OUTPUT)util/sort.o +LIB_OBJS += $(OUTPUT)util/spark.o LIB_OBJS += $(OUTPUT)util/hist.o LIB_OBJS += $(OUTPUT)util/probe-event.o LIB_OBJS += $(OUTPUT)util/util.o
tools/perf/builtin-stat.c +12 -0
@@ -1176,6 +1176,9 @@ static void print_aggr(char *prefix)if (run != ena)fprintf (output," (%.2f%%)" ,100.0 * run / ena); + +fputc (' ' , output); +print_stat_spark (output, counter->priv); }fputc ('\n' , output); }
@@ -1229,6 +1232,9 @@ static void print_counter_aggr(struct perf_evsel *counter, char *prefix)return ; } +fputc (' ' , output); +print_stat_spark (output, counter->priv); +if (scaled) {double avg_enabled, avg_running;
@@ -1295,6 +1301,9 @@ static void print_counter(struct perf_evsel *counter, char *prefix)if (run != ena)fprintf (output," (%.2f%%)" ,100.0 * run / ena); + +fputc (' ' , output); +print_stat_spark (output, counter->priv); }fputc ('\n' , output); }
@@ -1355,6 +1364,9 @@ static void print_stat(int argc, const char **argv)fprintf (output," " );print_noise_pct (stddev_stats (&walltime_nsecs_stats),avg_stats (&walltime_nsecs_stats)); + +fputc (' ' , output); +print_stat_spark (output, &walltime_nsecs_stats); }fprintf (output,"\n\n" ); }
tools/perf/util/spark.c new file +31 -0
@@ -0,0 +1,31 @@ +#include <stdio.h> +#include <limits.h> +#include "spark.h" + +#define NUM_SPARKS 8 +#define SPARK_SHIFT 8 + +/* Print spark lines on outf for numval values in val. */ +void print_spark (FILE *outf,unsigned long *val,int numval) +{ +static const char *ticks[NUM_SPARKS ] = { +"▁" ,"▂" ,"▃" ,"▄" ,"▅" ,"▆" ,"▇" ,"█" + }; +int i; +unsigned long long min =ULLONG_MAX , max =0 , f; + +for (i =0 ; i < numval; i++) { +if (val[i] < min) + min = val[i]; +if (val[i] > max) + max = val[i]; + } + f = ((max - min) <<SPARK_SHIFT ) / (NUM_SPARKS -1 ); +if (f <1 ) + f =1 ; +for (i =0 ; i < numval; i++) { +int index = ((val[i] - min) <<SPARK_SHIFT ) / f; +if (index >=0 && index <NUM_SPARKS ) +fputs (ticks[index], outf); + } +}
tools/perf/util/spark.h new file +4 -0
@@ -0,0 +1,4 @@ +#ifndef SPARK_H +#define SPARK_H 1 +void print_spark (FILE *outf,unsigned long *val,int numval); +#endif
tools/perf/util/stat.c +34 -0
@@ -1,10 +1,16 @@#include <math.h> +#include <stdio.h> #include "stat.h" +#include "spark.h" void update_stats (struct stats *stats,u64 val) {double delta; +int n = stats->n; + +if (n <NUM_SPARK_VALS ) + stats->svals[n] = val; stats->n++; delta = val - stats->mean;
@@ -61,3 +67,31 @@ double rel_stddev_stats(double stddev, double avg)return pct; } + +static int all_zero (unsigned long *vals,int len) +{ +int i; + +for (i =0 ; i < len; i++) +if (vals[i] !=0 ) +return 0 ; +return 1 ; +} + +void print_stat_spark (FILE *f,struct stats *stat) +{ +int len; + +if (stat->n <=1 ) +return ; + + len = stat->n; +if (len >NUM_SPARK_VALS ) + len =NUM_SPARK_VALS ; +if (all_zero (stat->svals, len)) +return ; + +print_spark (f, stat->svals, len); +if (stat->n >NUM_SPARK_VALS ) +fputs (".." , f); +}
tools/perf/util/stat.h +10 -0
@@ -1,12 +1,16 @@#ifndef __PERF_STATS_H#define __PERF_STATS_H +#include <stdio.h> #include "types.h" +#define NUM_SPARK_VALS 8/* support spark line on first N items */ +struct stats {double n, mean, M2;u64 max, min; +u64 svals[NUM_SPARK_VALS ]; };void update_stats (struct stats *stats,u64 val);
@@ -14,12 +18,18 @@ double avg_stats(struct stats *stats);double stddev_stats (struct stats *stats);double rel_stddev_stats (double stddev,double avg); +void print_stat_spark (FILE *f,struct stats *stat); +static inline void init_stats (struct stats *stats) { +int i; + stats->n =0.0 ; stats->mean =0.0 ; stats->M2 =0.0 ; stats->min = (u64) -1 ; stats->max =0 ; +for (i =0 ; i <NUM_SPARK_VALS ; i++) + stats->svals[i] =0 ; }#endif
Signature · 1 line
1.9.0
Quoted 44 lines · Andi Kleen
From: Andi Kleen <[email protected]>
perf stat -rX prints the stddev for multiple measurements. Just looking at the stddev for judging the quality of the data is a bit dangerous The simplest sanity check is to just look at a simple plot. This patchs add a sparkline to the end of the measurements to make it simple to judge the data.
The sparkline only uses UTF-8, so should be readable in all modern tools and terminals.
The sparkline is between the minimum and maximum of the data, so it's mainly a indicator of variance. To keep the code simple and make the output not too wide only the first 8 values are printed. If more values are there it adds '..'
The code is inspired by Zach Holman's spark shell script.
Example output (view in non-proportial font):
Performance counter stats for 'true' (10 runs):
0.175672 task-clock (msec) # 0.555 CPUs utilized ( +- 1.77% ) █▄▁▁▁▁▁▁.. 0 context-switches # 0.000 K/sec 0 cpu-migrations # 0.000 K/sec 114 page-faults # 0.647 M/sec ( +- 0.14% ) ▁█▁▁████.. 520,798 cycles # 2.965 GHz ( +- 1.75% ) █▄▁▁▁▁▁▁.. 433,525 instructions # 0.83 insns per cycle ( +- 0.28% ) ▅▇▅▄▇█▁▆.. 83,012 branches # 472.537 M/sec ( +- 0.31% ) ▅▇▆▄▇█▁▆.. 3,157 branch-misses # 3.80% of all branches ( +- 2.55% ) ▇█▃▅▁▃▁▂..0.000316660 seconds time elapsed ( +- 1.78% ) █▅▁▁▁▁▁▁..
As you can see even in the most simple run there are quite interesting patterns. The time sparkline suggests it would be also useful to have an option to throw the first measurement away.
Known issues:
- Makes the perf stat output wider. Could be adjust by shrinking some white space. Not done so far.
- No output for -A/--per-socket/--per-core with -rX. This code is missing the basic noise detection code. Once it's added there sparklines could be shown too.
Quoted 6 lines
v2: Avoid printing spark lines for normal CSV case (Jiri)
v3:LONG->ULONG, random changes
v4: Add some missing changes from the forked v2: checks value is not zero instead of all the same. Update documentation. Remove n variable. Remove #pragma once
v5: Add overflow fixes from Jiri.
got compilation error on Fedora 15 i386:
CC util/stat.outil/stat.c: In function ‘print_stat_spark’:util/stat.c:91:2: error: passing argument 1 of ‘all_zero’ from incompatible pointer type [-Werror] util/stat.c:71:12: note: expected ‘long unsigned int *’ but argument is of type ‘u64 *’ util/stat.c:94:2: error: passing argument 2 of ‘print_spark’ from incompatible pointer type [-Werror] util/spark.h:3:6: note: expected ‘long unsigned int *’ but argument is of type ‘u64 *’
cc1: all warnings being treated as errors
make[1]: *** [util/stat.o] Error 1
make[1]: *** Waiting for unfinished jobs....
make: *** [all] Error 2
jirka