Re: [RFC] 2.5 TASK_INTERRUPTIBLE preemption race

From: Robert Love (rml@tech9.net)
Date: Mon Apr 14 2003 - 17:30:57 EST


On Mon, 2003-04-14 at 18:27, Joe Korty wrote:

> I see. It is because the 'goto pick_next_task' skips the
> 'deactivate_task' call. Therefore the previous task remains on the
> run queue in spite of its TASK_UNINTERRUPTIBLE state. Clever!

Yep :)

We actually did away with it for a bit in mid 2.5... it turned out to
not be worth it. Its a little odd to use the flag like that, but its
quite simply and the results are good.

Means we can safely schedule anytime, no matter the state. There are
tons of other places where we would not want to sleep if it were not for
PREEMPT_ACTIVE.

        Robert Love

-
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/



This archive was generated by hypermail 2b29 : Tue Apr 15 2003 - 22:00:34 EST