Thread (177 messages) 177 messages, 16 authors, 2012-07-14

Re: [PATCH 21/40] autonuma: avoid CFS select_task_rq_fair to return -1

From: Rik van Riel <hidden>
Date: 2012-06-29 18:59:08
Also in: lkml

On 06/28/2012 08:56 AM, Andrea Arcangeli wrote:
quoted hunk ↗ jump to hunk
Fix to avoid -1 retval.

Includes fixes from Hillf Danton[off-list ref].

Signed-off-by: Andrea Arcangeli<redacted>
---
  kernel/sched/fair.c |    4 ++++
  1 files changed, 4 insertions(+), 0 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index c099cc6..fa96810 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -2789,6 +2789,9 @@ select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags)
  		if (new_cpu == -1 || new_cpu == cpu) {
  			/* Now try balancing at a lower domain level of cpu */
  			sd = sd->child;
+			if (new_cpu<  0)
+				/* Return prev_cpu is find_idlest_cpu failed */
+				new_cpu = prev_cpu;
  			continue;
  		}
@@ -2807,6 +2810,7 @@ select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags)
  unlock:
  	rcu_read_unlock();

+	BUG_ON(new_cpu<  0);
  	return new_cpu;
  }
  #endif /* CONFIG_SMP */
Wait, what?

Either this is a scheduler bugfix, in which case you
are better off submitting it separately and reducing
the size of your autonuma patch queue, or this is a
behaviour change in the scheduler that needs better
arguments than a 1-line changelog.

-- 
All rights reversed

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help