Home » Mailing list archives » linux-kernel » 2010 » August » 4

excessive wakeups in 2.6.35

Previous thread: [PATCH] drivers/staging/iio: call mutex_unlock in error handling code by Julia Lawall on Wednesday, August 4, 2010 - 12:54 pm. (2 messages)

Next thread: [PATCH v2] isofs: Fix lseek() to position beyond 4 GB by Jan Andres on Wednesday, August 4, 2010 - 1:52 pm. (3 messages)
[view original message]
From: Dave Jones
Subject: excessive wakeups in 2.6.35
Date: Wednesday, August 4, 2010 - 1:38 pm

My laptop felt like it was about to burst into flames after I ran
2.6.35 on it for a while.  It actually got too hot to place on my legs.
Powertop showed it running over a 1000 wakeups a second while completely idle.

Chuck Ebbert pointed me at this diff..
http://git.kernel.org/?p=linux/kernel/git/x86/linux-2.6-tip.git;a=commitdiff;h=396e894...

which brings it back down to a much more respectable 80 wakeups a second.

Is this lined up for 2.6.35.1 ?

	Dave

--


[ message continues ]
[view original message]
From: Mike Galbraith
Subject: Re: excessive wakeups in 2.6.35
Date: Wednesday, August 4, 2010 - 8:48 pm

Yeah, it's tagged.  More trouble than it's worth, optimization dies.

	-Mike

--


[ message continues ]
Previous thread: [PATCH] drivers/staging/iio: call mutex_unlock in error handling code by Julia Lawall on Wednesday, August 4, 2010 - 12:54 pm. (2 messages)

Next thread: [PATCH v2] isofs: Fix lseek() to position beyond 4 GB by Jan Andres on Wednesday, August 4, 2010 - 1:52 pm. (3 messages)