Does the gpftools cpu profiler not support a multiprocessing process?

Question

according to the document http://gperftools.googlecode.com/svn/trunk/doc/cpuprofile.html , cpu profiles support the multiprocessing process and generate an independent output file:

If your software plugs, children will also be profiled (since they inherit the same CPUPROFILE setting). Each process is profiled separately; to distinguish between child profiles from the parent profile and from each other, all children will have their own process identifier added to the name CPUPROFILE.

but when I try to do the following:

//main_cmd_argv.cpp

#include <stdio.h> #include <stdlib.h> #include <unistd.h> #include <string.h> #include <sys/types.h> #include <sys/wait.h> #include <gperftools/profiler.h> int loop(int n) { int sum = 0; for (int i = 0; i < n; i++) { for (int j = 0; j < n; j++) { sum = i + j; if (sum %3 == 0) { sum /= 3; } } } return 0; } int main(int argc, char* argv[]) { printf("%s\n%s\n", getenv("CPUPROFILE"), getenv("CPUPROFILESIGNAL")); if (argc > 1 && strcmp(argv[1], "-s")==0) { // single process loop(100000); printf("stoped\n"); } else if (argc > 1 && strcmp(argv[1], "-m")==0) { // multi process pid_t pid = fork(); if (pid < 0) { printf("fork error\n"); return -1; } if (pid == 0) { loop(100000); printf("child stoped\n"); } else if (pid > 0) { loop(10000); printf("father stoped\n"); wait(NULL); } } return 0; } 

// makefile

 GPerfTools=/home/adenzhang/tools/gperftools CCFLAGS=-fno-omit-frame-pointer -g -Wall ALL_BINS=main_cmd_argv all:$(ALL_BINS) main_cmd_argv:main_cmd_argv.o g++ $(CCFLAGS) -o $@ $^ -L./ -L$(GPerfTools)/lib -Wl,-Bdynamic -lprofiler -lunwind .cpp.o: g++ $(CCFLAGS) -c -I./ -I$(GPerfTools)/include -fPIC -o $@ $< clean: rm -f $(ALL_BINS) *.o *.prof 

// shell command

 $ make g++ -fno-omit-frame-pointer -g -Wall -c -I./ -I/home/adenzhang/tools/gperftools/include -fPIC -o main_cmd_argv.o main_cmd_argv.cpp g++ -fno-omit-frame-pointer -g -Wall -o main_cmd_argv main_cmd_argv.o -L./ -L/home/adenzhang/tools/gperftools/lib -Wl,-Bdynamic -lprofiler -lunwind $ env CPUPROFILE=main_cmd_argv.prof ./main_cmd_argv -s 젩n_cmd_argv.prof (null) stoped PROFILE: interrupts/evictions/bytes = 6686/3564/228416 $ /home/adenzhang/tools/gperftools/bin/pprof --text ./main_cmd_argv ./main_cmd_argv.prof Using local file ./main_cmd_argv. Using local file ./main_cmd_argv.prof. Removing killpg from all stack traces. Total: 6686 samples 6686 100.0% 100.0% 6686 100.0% loop 0 0.0% 100.0% 6686 100.0% __libc_start_main 0 0.0% 100.0% 6686 100.0% _start 0 0.0% 100.0% 6686 100.0% main $ rm main_cmd_argv.prof $ env CPUPROFILE=main_cmd_argv.prof ./main_cmd_argv -m 젩n_cmd_argv.prof (null) father stoped child stoped PROFILE: interrupts/evictions/bytes = 0/0/64 PROFILE: interrupts/evictions/bytes = 68/36/2624 $ ls main_cmd_argv main_cmd_argv.cpp main_cmd_argv.o main_cmd_argv.prof Makefile $ /home/adenzhang/tools/gperftools/bin/pprof --text ./main_cmd_argv ./main_cmd_argv.prof Using local file ./main_cmd_argv. Using local file ./main_cmd_argv.prof. $ 

This means that gperf does not support a multiprocessing process, can anyone explain this? thanks!

Answer 1

Pretty old, I don't know if you found an answer or not, but ...

It seems that every thread / fork should register using ProfilerRegisterThread (); You can find more information on these two issues: Here and Here .

An example code is also provided, similar to your test case, where forks can be registered .

Answer 2

I am currently using gperftools to profile the mpi program and run into this problem. After googling, I found that ProfilerStart(_YOUR_PROF_FILE_NAME_) and ProfilerStop() should be called during each subprocess, and _YOUR_PRO_FILE_NAME_ should be different in different processes. Then you can analyze the performance of each process.

link (also given by ZRJ):
https://groups.google.com/forum/#!topic/google-perftools/bmysZILR4ik