BUG? possible race between tasklet_action and tasklet_kill

[Chris Friesen]

I think I may have discovered a race between tasklet_action and
tasklet_kill.  Given how long this code has been around, I assume that
I'm simply not understanding the purposes of tasklet_kill correctly, but
it seems like there's a mechanism whereby we could hit the BUG path in
tasklet_action().

cpu A                             cpu B
starts tasklet_kill()
does while loop until
  TASKLET_STATE_SCHED not set
                                  runs tasklet_schedule()
                                  starts tasklet_action()
tasklet_unlock_wait()
                                  if (tasklet_trylock(t)) {
clear_bit(TASKLET_STATE_SCHED)
                                  test_and_clear_bit(TASKLET_STATE_SCHED
                                  BUG


Is this a valid issue, or is this an improper usage pattern somehow?

Thanks,

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