[Intel-gfx] high number of wakeups/second shown with 2.5 branch

Tino Keitel tino.keitel+xorg at tikei.de
Tue Oct 21 07:25:59 CEST 2008


On Mon, Oct 20, 2008 at 15:22:58 -0700, Eric Anholt wrote:
> On Mon, 2008-10-20 at 22:51 +0200, Lukas Hejtmanek wrote:
> > On Mon, Oct 20, 2008 at 01:31:20PM -0700, Eric Anholt wrote:
> > > > The workload is a plain 2D desktop (no compositing), with clicking
> > > > around a bit, scrolling, changing window focus etc. A completely idle
> > > > desktop with no moving parts doesn't produce any wakeups in the i915
> > > > driver.
> > > 
> > > Sounds like exactly expected behavior then.
> > 
> > I guess, people around powertop will not like this behavior..
> 
> powertop helps you decrease interrupts to decrease power consumption on
> an idle system.  In this case, we're saving power on an active system by
> letting the CPU sleep at all instead of spinning waiting for results.

I already thought the same, so no need to worry. Thanks for the
clarification. Maybe this should be documented somewhere, to have a
link that other people can read if they notice the same.

Regards,
Tino



More information about the Intel-gfx mailing list