Messages in this thread | | | From | Namhyung Kim <> | Subject | Re: [GIT PULL 0/8] Annotation weekly ponies delivery | Date | Wed, 16 May 2012 14:57:49 +0900 |
| |
Hi,
On Tue, 15 May 2012 19:07:13 +0200, Peter Zijlstra wrote: > This does indeed make it work again: > > pre patch: > > [ perf record: Woken up 1 times to write data ] > [ perf record: Captured and wrote 0.005 MB perf.data (~212 samples) ] > > post patch: > > [ perf record: Woken up 209 times to write data ] > [ perf record: Captured and wrote 54.931 MB perf.data (~2399974 samples) ] > > So please do send as a proper patch, and you can add: > > Acked-and-tested-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Hmm.. but it seems this breaks my 'perf stat -g' - it doesn't respond any more (even for ^C):
namhyung@sejong:perf$ ./perf stat -g sleep 1 ^C^C^C^C^\Quit (core dumped)
namhyung@sejong:perf$ gdb ./perf core.11964 [SNIP] Core was generated by `./perf stat -g sleep 1'. Program terminated with signal 3, Quit. #0 0x0000003fd800e090 in __read_nocancel () from /lib64/libpthread.so.0 [SNIP] (gdb) bt #0 0x0000003fd800e090 in __read_nocancel () from /lib64/libpthread.so.0 #1 0x000000000047fa00 in read (__nbytes=24, __buf=0x7fffc5e1d0b0, __fd=0) at /usr/include/bits/unistd.h:45 #2 readn (fd=0, buf=<optimized out>, buf@entry=0x7fffc5e1d0b0, n=n@entry=24) at util/util.c:140 #3 0x000000000043e4bb in __perf_evsel__read (evsel=evsel@entry=0xc7e700, ncpus=<optimized out>, nthreads=1, scale=true) at util/evsel.c:267 #4 0x0000000000422777 in read_counter_aggr (counter=0xc7e700) at builtin-stat.c:360 #5 run_perf_stat (argv=0x7fffc5e1d510, argc=2) at builtin-stat.c:528 #6 cmd_stat (argc=2, argv=0x7fffc5e1d510, prefix=<optimized out>) at builtin-stat.c:1251 #7 0x0000000000414723 in run_builtin (p=p@entry=0x77e1d8, argc=argc@entry=4, argv=argv@entry=0x7fffc5e1d510) at perf.c:273 #8 0x0000000000413f3b in handle_internal_command (argv=0x7fffc5e1d510, argc=4) at perf.c:345 #9 run_argv (argv=0x7fffc5e1d300, argcp=0x7fffc5e1d30c) at perf.c:389 #10 main (argc=4, argv=0x7fffc5e1d510) at perf.c:487
In fact, my perf stat -g was already broken - it cannot count at all on my SNB box. I'll investigate it more.
Any thoughts?
Namhyung
| |