performance - Linux Perf Counting Instructions, Cycles, Cache misses more lightweight over a long time -


i want profile programs under different restrictions research. use perf tool measure counters. profiling go long time (hours), , perf tool generates data don't need. instance:

$ perf record -e cycles,instructions,cache-misses,cpu-clock,context-switches,cpu-migrations ./stress-ng --cpu 4 --cpu-ops 10000 stress-ng: info:  [6642] defaulting 86400 second run per stressor stress-ng: info:  [6642] dispatching hogs: 4 cpu stress-ng: info:  [6642] successful run completed in 15.57s [ perf record: woken 117 times write data ] [ perf record: captured , wrote 29.971 mb perf.data (~1309444 samples) ] 

it approrixmately 1.92 mbs per second. in 24 hours, 165 gbs. when used perf report tool, saw kept function calls not need. how can make perf-record work more lightweight , record data need? need analyze following data on time see how change changes make system. considering records lot of data, , use in 24 core machines makes data large. best option far record periodically , analyze perf.data integers want , delete it, believe there should better option since can miss profiling data in window.

 performance counter stats './stress-ng --cpu 4 --cpu-ops 10000':     128.290.294.651      cycles                       201.217.788.817      instructions              #    1,57  insns per cycle                  3.048.096      cache-misses                                                       40340,258757      cpu-clock (msec)                                                            123      context-switches                                                             25      cpu-migrations                                                      10,205760993 seconds time elapsed      

does simple perf stat output you? example

$ perf stat -e cycles,instructions,cache-misses,cpu-clock,context-switches,cpu-migrations date  sun jan 24 18:46:10 est 2016   performance counter stats 'date':           1,269,265      cycles                               660,875      instructions              #    0.52  insns per cycle                      2,265      cache-misses                                                           0.644391      cpu-clock (msec)                                                              2      context-switches                                                              1      cpu-migrations                                                       0.001031207 seconds time elapsed 

Comments