Messages in this thread | | | Subject | Re: [PATCH 2/2] i386: sched_clock() follows percpu frequency changes | From | Daniel Walker <> | Date | Sat, 12 May 2007 12:25:40 -0700 |
| |
On Sat, 2007-05-12 at 20:58 +0200, Andi Kleen wrote: > On Saturday 12 May 2007 17:43:46 Daniel Walker wrote: > > Considering that Andi has a similar patch, I'm going to assume there > > is already justification for this.. > > I just have a patch for sched_clock, which is somewhat different. > > > What it ends up doing is allows the TSC to be used in cases which > > we didn't use it before, and hope it's still mostly accurate. > > Yes the current code is badly broken in this regard. e.g. i always > hated that it always falls back on powernow systems. > > Jiri will hopefully repost his true per CPU TSC patch soon, that should > make it all obsolete.
All I could find on LKML from Jiri was an x86_64 specific gettimeofday implementation .. Do you have a reference to the first post of the per cpu tsc patch?
Daniel
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |