On Tue, Jul 30, 2013 at 06:18:57PM +0900, Namhyung Kim wrote:
> This patchset implements a front-end tool for kernel's ftrace. It
> uses function_graph tracer by default and normal function tracer is
> also supported. (Of course you need to enable those tracers in your
> kernel first.)
>
> v4 changes:
> * use pid instead of tid for -p option (David)
> * not to poll() for reading ftrace pipes (Jiri)
> * rename to create_perf_header() (Jiri)
> * canonicalize directory name (Jiri)
> * show more info when -D option was given (Jiri)
> * update documentation (Jiri)
> * and few more bug fixes
>
> I pushed it out to 'perf/ftrace-v4' branch on my tree at:
>
> git://git.kernel.org/pub/scm/linux/kernel/git/namhyung/linux-perf.git
got following for record:
[root@dhcp-26-214 perf]# ./perf ftrace record kill
usage: kill [ -s signal | -p ] [ -a ] pid ...
kill -l [ signal ]
recorder failed for some reason on cpu0
recorder failed for some reason on cpu1
afterwards 'perf ftrace report' gets stuck with 2 processes:
[jolsa@dhcp-26-214 perf]$ sudo pstack 1470
#0 0x00007f23ab61e997 in munmap () from /lib64/libc.so.6
#1 0x000000000047b5f7 in __perf_session__process_events ()
#2 0x000000000043f230 in do_ftrace_report ()
#3 0x0000000000440555 in cmd_ftrace ()
#4 0x000000000041aad3 in run_builtin ()
#5 0x000000000041a379 in main ()
[jolsa@dhcp-26-214 perf]$ sudo pstack 1471
#0 0x00007f23ab61b623 in __select_nocancel () from /lib64/libc.so.6
#1 0x000000000046a337 in pager_preexec ()
#2 0x000000000045f5a3 in start_command ()
#3 0x000000000046a4ab in setup_pager ()
#4 0x0000000000440535 in cmd_ftrace ()
#5 0x000000000041aad3 in run_builtin ()
#6 0x000000000041a379 in main ()
jirka
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
