[lkml]   [2009]   [Jan]   [7]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH] configure HAVE_UNSTABLE_SCHED_CLOCK for SGI_SN systems
On Wed, 2009-01-07 at 07:32 -0600, Dimitri Sivanich wrote:
> Peter,
> On Wed, Jan 07, 2009 at 10:53:57AM +0100, Peter Zijlstra wrote:
> >
> > Clock state is kept per-cpu, and locked with a spinlock. When we request
> Admittedly I have not looked at this possibility too closely, but my
> initial concern upon looking at sched_clock_cpu() for the UNSTABLE
> case was the lock_double_clock() and what sort of contention that
> might cause on larger systems under certain conditions.

Similar contention would already exist on rq->lock.

 \ /
  Last update: 2009-01-07 16:19    [W:0.082 / U:2.216 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site