* [PATCH] 2.6.3 Posix scheduling violation for !SCHED_OTHER
@ 2004-03-23 23:35 Joe Korty
2004-03-24 0:01 ` Ingo Molnar
0 siblings, 1 reply; 2+ messages in thread
From: Joe Korty @ 2004-03-23 23:35 UTC (permalink / raw)
To: akpm; +Cc: linux-kernel, mingo
Andrew,
The following fixes a problem where a SCHED_FIFO task would on occasion
be moved to the end of its runqueue when returned to from a preemption.
Cause was do to some SCHED_OTHER code in schedule() which was being
run for tasks of every policy.
Regards,
Joe
--- 2.6.3/kernel/sched.c.orig 2004-02-17 22:59:10.000000000 -0500
+++ 2.6.3/kernel/sched.c 2004-03-23 18:34:19.000000000 -0500
@@ -1677,7 +1677,7 @@
queue = array->queue + idx;
next = list_entry(queue->next, task_t, run_list);
- if (next->activated > 0) {
+ if (!rt_task(next) && next->activated > 0) {
unsigned long long delta = now - next->timestamp;
if (next->activated == 1)
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [PATCH] 2.6.3 Posix scheduling violation for !SCHED_OTHER
2004-03-23 23:35 [PATCH] 2.6.3 Posix scheduling violation for !SCHED_OTHER Joe Korty
@ 2004-03-24 0:01 ` Ingo Molnar
0 siblings, 0 replies; 2+ messages in thread
From: Ingo Molnar @ 2004-03-24 0:01 UTC (permalink / raw)
To: Joe Korty; +Cc: akpm, linux-kernel
* Joe Korty <joe.korty@ccur.com> wrote:
> Andrew,
> The following fixes a problem where a SCHED_FIFO task would on
> occasion be moved to the end of its runqueue when returned to from a
> preemption. Cause was do to some SCHED_OTHER code in schedule() which
> was being run for tasks of every policy.
yes, indeed this is a bug.
> - if (next->activated > 0) {
> + if (!rt_task(next) && next->activated > 0) {
the patch looks good. Andrew, please apply.
Ingo
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2004-03-24 0:00 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-03-23 23:35 [PATCH] 2.6.3 Posix scheduling violation for !SCHED_OTHER Joe Korty
2004-03-24 0:01 ` Ingo Molnar
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox