On Wed, May 19, 2010 at 10:18:43AM +0200, Peter Zijlstra wrote: > On Wed, 2010-05-19 at 09:58 +0200, Frederic Weisbecker wrote: > > On Tue, May 18, 2010 at 05:15:42PM +0000, tip-bot for Peter Zijlstra wrote: > > > Commit-ID: 4f41c013f553957765902fb01475972f0af3e8e7 > > > Gitweb: http://git.kernel.org/tip/4f41c013f553957765902fb01475972f0af3e8e7 > > > Author: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx> > > > AuthorDate: Tue, 18 May 2010 18:08:32 +0200 > > > Committer: Ingo Molnar <mingo@xxxxxxx> > > > CommitDate: Tue, 18 May 2010 18:35:46 +0200 > > > > > > perf/ftrace: Optimize perf/tracepoint interaction for single events > > > > > > When we've got but a single event per tracepoint > > > there is no reason to try and multiplex it so don't. > > > > > > I haven't seen this patch going. > > > > But yeah that's a nice optimization for UP machines. > > Uhm, doesn't have anything to do with UP machines... it works quite well > on SMP as well. > > Its about tracepoints with a single event attached. Yeah, then if you launch perf sched for example, you'll have NR_CPU perf events attached to each lock trace events, since we create one event per cpu. Or may be I'm somewhat confused... -- To unsubscribe from this list: send the line "unsubscribe linux-tip-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
|