On Sat, 09 Jul 2011 15:07:28 -0700, Keith Packard <kei...@keithp.com> wrote: Non-text part: multipart/signed > On Sat, 09 Jul 2011 22:23:28 +0100, Chris Wilson <ch...@chris-wilson.co.uk> > wrote: > > I can stick some counters in there and find out what the impact actually > > is.
On a SNB laptop, so surplus memory and surplus processing power, we don't see that many faults at all whilst messing around with firefox, gitk and a couple of timedemos (padman, openarena): no-prefaulting: pread: shmem fast: 1273764/3395, slow: 0/0 pwrite: sheme fast: 51163148/14554, slow: 23552/9 pwrite: gtt fast: 32744702068/12658489, slow: 29376/10 all-prefaulted: pread: shmem fast: 6081544/5999, slow: 0/0 pwrite: shmem fast: 56867580/15932, slow: 0/0 pwrite: gtt fast: 32976168476/12753355, slow: 0/0 I'll look at a PNV netbook next. -Chris -- Chris Wilson, Intel Open Source Technology Centre _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/intel-gfx
