tracing: trace_remove_event_call() should fail if call/file is in use
authorOleg Nesterov <oleg@redhat.com>
Mon, 29 Jul 2013 17:50:33 +0000 (19:50 +0200)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Thu, 29 Aug 2013 16:47:34 +0000 (09:47 -0700)
commit8169887b694aa3d9632fdd0308aaa1db7c422e75
tree521d15d1f21cdcbbf25d490a3a0f850e31635fbe
parent012dc156d6af49e02a30fcba7d688e251608d97c
tracing: trace_remove_event_call() should fail if call/file is in use

commit 2816c551c796ec14620325b2c9ed75b9979d3125 upstream.

Change trace_remove_event_call(call) to return the error if this
call is active. This is what the callers assume but can't verify
outside of the tracing locks. Both trace_kprobe.c/trace_uprobe.c
need the additional changes, unregister_trace_probe() should abort
if trace_remove_event_call() fails.

The caller is going to free this call/file so we must ensure that
nobody can use them after trace_remove_event_call() succeeds.
debugfs should be fine after the previous changes and event_remove()
does TRACE_REG_UNREGISTER, but still there are 2 reasons why we need
the additional checks:

- There could be a perf_event(s) attached to this tp_event, so the
  patch checks ->perf_refcount.

- TRACE_REG_UNREGISTER can be suppressed by FTRACE_EVENT_FL_SOFT_MODE,
  so we simply check FTRACE_EVENT_FL_ENABLED protected by event_mutex.

Link: http://lkml.kernel.org/r/20130729175033.GB26284@redhat.com

Reviewed-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
include/linux/ftrace_event.h
kernel/trace/trace_events.c