Re: [patch] del_timer_sync scalability patch

[Oleg Nesterov]

I suspect that del_timer_sync() in its current form is racy.

CPU 0						CPU 1

__run_timers() sets timer->base = NULL

						del_timer_sync() starts, calls del_timer(), it returns
						because timer->base == NULL.

						waits until the run is complete:
							while (base->running_timer == timer)
								preempt_check_resched();
						
						calls schedule(), or long interrupt happens.

timer reschedules itself, __run_timers()
exits.

						base->running_timer == NULL, end of loop.

next timer interrupt, __run_timers() picks
this timer again, sets timer->base = NULL

						if (timer_pending(timer))	// no, timer->base == NULL
							goto del_again;		// not taken

						del_timer_sync() returns

timer runs.

No?

Oleg.
-
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/