* Rob Hussey <robjhussey@gmail.com> wrote:btw., it's likely that if you turn off CONFIG_PREEMPT for .21 and for .22-ck1 they'll improve a bit too - so it's not fair to put the .23 !PREEMPT numbers on the graph as the PREEMPT numbers of the other kernels. (it shows the .23 scheduler being faster than it really is) the pipe-test behavior looks like an outlier. !PREEMPT only removes code (which makes the code faster), so this could be a cache layout artifact. (or perhaps we preempt at a different point which is disadvantageous to caching?) Pipe-test is equivalent to "lat_ctx -s 0 2" so if there was a genuine slowdown it would show up in the lat_ctx graph - but the graph shows a speedup. Ingo -
| Greg Kroah-Hartman | [PATCH 005/196] Chinese: add translation of SubmittingDrivers |
| Christian Kujau | 2.6.20.4: NETDEV WATCHDOG and lockups |
| Tarkan Erimer | Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3 |
| Jack Steiner | Re: [patch] my mmu notifiers |
git: | |
| Gerrit Renker | [PATCH 27/37] dccp: Integration of dynamic feature activation - part 2 (server side) |
| Arjan van de Ven | Re: [GIT]: Networking |
| Jarek Poplawski | [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock(). |
| Jens Axboe | Re: [BUG] New Kernel Bugs |
| YAMAMOTO Takashi | removing VOPs |
| Lennart Augustsson | Re: FreeBSD 5/6/7 kernel emulator for NetBSD 2.x |
| Daniel Carosone | Re: direct I/O |
| Brian Buhrow | Re: /sbin/reboot and secmodel |
