lkml.org 
[lkml]   [2006]   [Nov]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
SubjectPATCH? hrtimer_wakeup: fix a theoretical race wrt rt_mutex_slowlock()
When task->array != NULL, try_to_wake_up() just goes to "out_running" and sets
task->state = TASK_RUNNING.

In that case hrtimer_wakeup() does:

timeout->task = NULL; <----- [1]

spin_lock(runqueues->lock);

task->state = TASK_RUNNING; <----- [2]

from Documentation/memory-barriers.txt

Memory operations that occur before a LOCK operation may appear to
happen after it completes.

This means that [2] may be completed before [1], and

CPU_0 CPU_1
rt_mutex_slowlock:

for (;;) {
...
if (timeout && !timeout->task)
return -ETIMEDOUT;
...

schedule();
hrtimer_wakeup() sets
... task->state = TASK_RUNNING,
but "timeout->task = NULL"
is not completed
set_current_state(TASK_INTERRUPTIBLE);
}

we can miss a timeout.

Of course, this all is scholasticism, this can't happen in practice, but
may be this patch makes sense as a documentation update.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>

--- STATS/kernel/hrtimer.c~1_hrtw 2006-10-22 18:24:03.000000000 +0400
+++ STATS/kernel/hrtimer.c 2006-11-05 22:32:36.000000000 +0300
@@ -662,9 +662,12 @@ static int hrtimer_wakeup(struct hrtimer
container_of(timer, struct hrtimer_sleeper, timer);
struct task_struct *task = t->task;

- t->task = NULL;
- if (task)
+ if (task) {
+ t->task = NULL;
+ /* must be visible before task->state = TASK_RUNNING */
+ smp_wmb();
wake_up_process(task);
+ }

return HRTIMER_NORESTART;
}
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2006-11-05 20:39    [W:0.065 / U:0.084 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site