Messages in this thread | | | Date | Tue, 19 Nov 2002 13:53:43 -0800 | From | Andrew Morton <> | Subject | Re: [BENCHMARK] 2.5.48-mm1 with contest |
| |
Robert Love wrote: > > On Tue, 2002-11-19 at 16:28, Con Kolivas wrote: > > > xtar_load: > > Kernel [runs] Time CPU% Loads LCPU% Ratio > > 2.5.48 [5] 184.4 41 2 6 2.52 > > 2.5.48-mm1 [5] 210.7 35 2 6 2.88 > > > > read_load: > > Kernel [runs] Time CPU% Loads LCPU% Ratio > > 2.5.48 [5] 102.9 74 6 4 1.41 > > 2.5.48-mm1 [5] 256.7 29 11 2 3.51* > > What changed, Andrew? > > Wall time is up but CPU is down... spending more time on I/O?
Big changes in the IO scheduler. I have not really sat down and looked at the effects of those changes - Jens keeps on changing it, and I keep on diddling the queue sizes. So it has only really been "stability tested" at this time. - 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/
| |