diff options
author | Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> | 2011-10-04 19:44:38 +0900 |
---|---|---|
committer | Steven Rostedt <rostedt@goodmis.org> | 2011-10-10 15:13:03 -0400 |
commit | 02ca1521ad404cf566e0075848f80d064c0a0503 (patch) | |
tree | 11989ae5bfc1e32f355cc0bb0a64cb7ae801ed78 /tools/perf/util/probe-event.c | |
parent | 9d3ec7a0c41d010153c9d20577cb1bcf9c4f65df (diff) | |
download | linux-02ca1521ad404cf566e0075848f80d064c0a0503.tar.gz linux-02ca1521ad404cf566e0075848f80d064c0a0503.tar.bz2 linux-02ca1521ad404cf566e0075848f80d064c0a0503.zip |
ftrace/kprobes: Fix not to delete probes if in use
Fix kprobe-tracer not to delete a probe if the probe is in use.
In that case, delete operation will return -EBUSY.
This bug can cause a kernel panic if enabled probes are deleted
during perf record.
(Add some probes on functions)
sh-4.2# perf probe --del probe:\*
sh-4.2# exit
(kernel panic)
This is originally reported on the fedora bugzilla:
https://bugzilla.redhat.com/show_bug.cgi?id=742383
I've also checked that this problem doesn't happen on
tracepoints when module removing because perf event
locks target module.
$ sudo ./perf record -e xfs:\* -aR sh
sh-4.2# rmmod xfs
ERROR: Module xfs is in use
sh-4.2# exit
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.203 MB perf.data (~8862 samples) ]
Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Frank Ch. Eigler <fche@redhat.com>
Cc: stable@kernel.org
Link: http://lkml.kernel.org/r/20111004104438.14591.6553.stgit@fedora15
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Diffstat (limited to 'tools/perf/util/probe-event.c')
0 files changed, 0 insertions, 0 deletions