* [PATCH 1/3] wait_task_inactive: "improve" the returned value for ->nvcsw == 0
@ 2008-07-30 17:09 Oleg Nesterov
2008-07-31 0:37 ` Roland McGrath
0 siblings, 1 reply; 2+ messages in thread
From: Oleg Nesterov @ 2008-07-30 17:09 UTC (permalink / raw)
To: Andrew Morton; +Cc: Ingo Molnar, Linus Torvalds, Roland McGrath, linux-kernel
(on top of wait_task_inactive-dont-consider-task-nivcsw.patch)
wait_task_inactive() returns 1 when p->nvcsw == 0 || p->nvcsw == 1.
This means that two subsequent calls can return the same number while
the task was scheduled in between.
Change the code to return "nvcsw | LONG_MIN" instead of "nvcsw ?: 1",
now the overlap always needs LONG_MAX schedules.
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
--- 27/kernel/sched.c~1_WTI_RET 2008-07-30 19:32:09.000000000 +0400
+++ 27/kernel/sched.c 2008-07-30 19:47:56.000000000 +0400
@@ -1923,7 +1923,7 @@ unsigned long wait_task_inactive(struct
on_rq = p->se.on_rq;
ncsw = 0;
if (!match_state || p->state == match_state)
- ncsw = p->nvcsw ?: 1;
+ ncsw = p->nvcsw | LONG_MIN; /* sets MSB */
task_rq_unlock(rq, &flags);
/*
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2008-07-31 0:38 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2008-07-30 17:09 [PATCH 1/3] wait_task_inactive: "improve" the returned value for ->nvcsw == 0 Oleg Nesterov
2008-07-31 0:37 ` Roland McGrath
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®