![]() | |||||||||||||
Messages in this thread |
Bill, would you please check the boot messages for the NUMA scheduler before doing the run. Martin sent me an example where he has: CPU pools : 1 pool 0 :0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 node level 0 : 10 pool_delay matrix: 129 which is clearly wrong. In that case we need to fix the cpu-pools setup first. Regards, Erich On Sunday 22 September 2002 10:09, William Lee Irwin III wrote: > On Sat, Sep 21, 2002 at 09:46:05AM -0700, Martin J. Bligh wrote: > >> An old compile off 2.5.31-mm1 + extras (I don't have 37, but similar) > > On Sat, Sep 21, 2002 at 04:18:10PM -0700, William Lee Irwin III wrote: > > Some 8-quad numbers for 2.5.37 (virgin) follow. > > Okay, 2.5.37 virgin with overcommit_memory set to 1 this time. > (compiles with -j256 seem to do better than -j32 or -j48, here is -j256): > > ... will follow up with 2.5.38-mm1 with and without NUMA sched, at > least if the arrival rate of releases doesn't exceed the benchtime. - 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/ | ||||||||||||
| Last update: 2005-03-22 12:29 [from the cache] ©2003-2008 | |||||||||||||