Messages in this thread | | | Date | Mon, 28 Sep 2009 08:39:31 -0700 (PDT) | From | Linus Torvalds <> | Subject | Re: Linux 2.6.32-rc1 |
| |
On Mon, 28 Sep 2009, Eric Dumazet wrote: > Linus Torvalds a écrit : > > > > Go wild, test it out, and let us know about any regressions you find, > > Something seems wrong with process time accounting. > > Following program should consume 10*8 seconds of cpu on a 8 cpu machine, but > with 2.6.32-rc1 numbers are crazy.
Ok, so the top-level process time looks sane _while_ the thing is running (sum of all threads), but the per-thread times look broken (as if they had randomly had the times of the other threads added into them - looks to me like they on average had half the other threads' time accounted into them).
And then at the end, it looks like the time of the threads (which was over-accounted) get re-accounted back into the main process, so the final time is indeed wildly inflated.
IOW, looks like we're adding thread times multiple times to other threads (and then finally to the parent).
I'm adding the usual suspects to the cc, and leaving your results and test-program quoted here for them.. Thomas, Martin, John - if you're not the people I should blame for this, let me know.
Linus
--- > $ gcc -O2 -o process process.c -lpthread > $ ./process > PID TTY STAT TIME COMMAND > 5532 pts/1 - 0:08 ./process > - - Sl+ 0:00 - > - - Rl+ 0:05 - > - - Rl+ 0:05 - > - - Rl+ 0:09 - > - - Rl+ 0:05 - > - - Rl+ 0:05 - > - - Rl+ 0:05 - > - - Rl+ 0:05 - > - - Rl+ 0:01 - > PID TTY STAT TIME COMMAND > 5532 pts/1 - 0:16 ./process > - - Sl+ 0:00 - > - - Rl+ 0:14 - > - - Rl+ 0:06 - > - - Rl+ 0:14 - > - - Rl+ 0:10 - > - - Rl+ 0:06 - > - - Rl+ 0:06 - > - - Rl+ 0:10 - > - - Rl+ 0:02 - > PID TTY STAT TIME COMMAND > 5532 pts/1 - 0:24 ./process > - - Sl+ 0:00 - > - - Rl+ 0:20 - > - - Rl+ 0:11 - > - - Rl+ 0:24 - > - - Rl+ 0:15 - > - - Rl+ 0:07 - > - - Rl+ 0:07 - > - - Rl+ 0:24 - > - - Rl+ 0:03 - > PID TTY STAT TIME COMMAND > 5532 pts/1 - 0:32 ./process > - - Sl+ 0:00 - > - - Rl+ 0:29 - > - - Rl+ 0:12 - > - - Rl+ 0:29 - > - - Rl+ 0:21 - > - - Rl+ 0:08 - > - - Rl+ 0:08 - > - - Rl+ 0:29 - > - - Rl+ 0:04 - > PID TTY STAT TIME COMMAND > 5532 pts/1 - 0:40 ./process > - - Sl+ 0:00 - > - - Rl+ 0:35 - > - - Rl+ 0:13 - > - - Rl+ 0:35 - > - - Rl+ 0:26 - > - - Rl+ 0:09 - > - - Rl+ 0:09 - > - - Rl+ 0:35 - > - - Rl+ 0:05 - > PID TTY STAT TIME COMMAND > 5532 pts/1 - 0:48 ./process > - - Sl+ 0:00 - > - - Rl+ 0:40 - > - - Rl+ 0:14 - > - - Rl+ 0:40 - > - - Rl+ 0:31 - > - - Rl+ 0:10 - > - - Rl+ 0:10 - > - - Rl+ 0:40 - > - - Rl+ 0:06 - > PID TTY STAT TIME COMMAND > 5532 pts/1 - 0:56 ./process > - - Sl+ 0:00 - > - - Rl+ 0:45 - > - - Rl+ 0:15 - > - - Rl+ 0:45 - > - - Rl+ 0:37 - > - - Rl+ 0:11 - > - - Rl+ 0:11 - > - - Rl+ 0:45 - > - - Rl+ 0:07 - > PID TTY STAT TIME COMMAND > 5532 pts/1 - 1:05 ./process > - - Sl+ 0:00 - > - - Rl+ 0:51 - > - - Rl+ 0:21 - > - - Rl+ 0:55 - > - - Rl+ 0:46 - > - - Rl+ 0:12 - > - - Rl+ 0:12 - > - - Rl+ 0:59 - > - - Rl+ 0:08 - > PID TTY STAT TIME COMMAND > 5532 pts/1 - 1:13 ./process > - - Sl+ 0:00 - > - - Rl+ 0:56 - > - - Rl+ 0:22 - > - - Rl+ 1:00 - > - - Rl+ 0:52 - > - - Rl+ 0:13 - > - - Rl+ 0:13 - > - - Rl+ 1:04 - > - - Rl+ 0:09 - > PID TTY STAT TIME COMMAND > 5532 pts/1 - 5:13 ./process > - - S+ 5:13 - > > > $ cat process.c > #include <pthread.h> > #include <time.h> > #include <stdio.h> > > /* > * Burn cpu cycles for about 10 seconds > */ > static void *do_unit(void *arg) > { > time_t t0,t1; > > t0 = time(NULL); > do { > t1 = time(NULL); > } while (t1 - t0 <= 10); > return NULL; > } > > > > > int main(int argc, char *argv[]) > { > pthread_t tids[8]; > int i; > char cmd[128]; > > sprintf(cmd, "ps m -p %d", getpid()); > for (i = 0; i < 8 ; i++) > pthread_create(&tids[i], NULL, do_unit, NULL); > for (i = 0; i < 9; i++) { > sleep(1); > system(cmd); > } > for (i = 0; i < 8 ; i++) > pthread_join(tids[i], NULL); > system(cmd); > return 0; > } > -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |