Techwizz/kernel-fxtec-pro1x
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

tracing/ftrace: fix a race condition in sched_switch tracer

Browse source 
Impact: fix race condition in sched_switch tracer

This patch fixes a race condition in the sched_switch tracer. If
several tasks (IE: concurrent initcalls) are playing with
tracing_start_cmdline_record() and tracing_stop_cmdline_record(), the
following situation could happen:

_ Task A and B are using the same tracepoint probe. Task A holds it.
  Task B is sleeping and doesn't hold it.

_ Task A frees the sched tracer, then sched_ref is decremented to 0.

_ Task A is preempted and hadn't yet unregistered its tracepoint
  probe, then B runs.

_ B increments sched_ref, sees it's 1 and then guess it has to
  register its probe. But it has not been freed by task A.

_ A lot of bad things can happen after that...

Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This commit is contained in:
Frederic Weisbecker 2008-10-31 13:08:28 +01:00 committed by Ingo Molnar
parent 71566a0d16
commit 07695fa04e
1 changed files with 5 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

5
kernel/trace/trace_sched_switch.c
View file

@ -17,6 +17,7 @@
static struct trace_array *ctx_trace;
static int __read_mostly tracer_enabled;
static atomic_t sched_ref;
static DEFINE_MUTEX(tracepoint_mutex);
static void
probe_sched_switch(struct rq *__rq, struct task_struct *prev,
@ -125,18 +126,22 @@ static void tracing_start_sched_switch(void)
{
long ref;
mutex_lock(&tracepoint_mutex);
ref = atomic_inc_return(&sched_ref);
if (ref == 1)
tracing_sched_register();
mutex_unlock(&tracepoint_mutex);
}
static void tracing_stop_sched_switch(void)
{
long ref;
mutex_lock(&tracepoint_mutex);
ref = atomic_dec_and_test(&sched_ref);
if (ref)
tracing_sched_unregister();
mutex_unlock(&tracepoint_mutex);
}
void tracing_start_cmdline_record(void)