lkml.org 
[lkml]   [2007]   [Dec]   [8]   [last100]   RSS Feed
Views: [more markup]  [less markup]  [headers]  [forward] 
 
Messages in this thread
/
DateSat, 8 Dec 2007 12:18:11 -0800
FromArjan van de Ven <>
SubjectRe: [patch] x86: scale cyc_2_nsec according to CPU frequency
On Sat, 8 Dec 2007 20:16:29 +0100
Ingo Molnar <mingo@elte.hu> wrote:

> 
> * Arjan van de Ven <arjan@infradead.org> wrote:
> 
> > > > > Firstly, we dont need the 'offset' anymore because
> > > > > cpu_clock() maintains offsets itself.
> > > > 
> > > > Yes, but a lower quality one. __update_rq_clock tries to 
> > > > compensate large jumping clocks with a jiffy resolution, while
> > > > my offset arranges for a very smooth frequency transition.
> > > 
> > > yes, but that would be easy to fix up via calling 
> > > sched_clock_idle_wakeup_event(0) when doing a frequency
> > > transition, without burdening the normal sched_clock() codepath
> > > with the offset. See the attached latest version.
> > 
> > can this deal with dual/quad core where the frequency of one core 
> > changes if the sofware changes the frequency of the other core?
> 
> doesnt the notifier still get run on the target CPU?
> 

.... if and only if the BIOS actually gives correct information to the
OS.
In reality... that's not a very common thing on this field sadly

-- 
If you want to reach me at my work email, use arjan@linux.intel.com
For development, discussion and tips for power savings, 
visit http://www.lesswatts.org


\
 
 \ /
  Last update: 2007-12-08 21:21    [from the cache]
©2003-2008