On Sun, 12 Jun 2005 09:27, Martin J. Bligh wrote: > >> not sure what the benefits of the patch are, I should have answered this. Since we moved to one runqueue per cpu with the current scheduler, 'nice' levels basically fall apart on SMP. Balancing tends to group together all the wrong tasks to have any meaningful 'nice' support where often on a 2 cpu machine if we run 4 tasks, 2 nice 0 and 2 nice 19 we end up with: cpu 1: nice 19 + nice 19 cpu 2: nice 0 + nice 0 which means each nice 19 task gets half a cpu and each nice 0 task gets half a cpu which is lousy fairness. The smp nice patches should end up with cpu 1: nice 0 + nice 19 cpu 2: nice 0 + nice 19 so that the nice 0 tasks get 95% of a cpu and nice 19 tasks get 5% of a cpu. The patches should balance things as fairly as possible according to nice levels across cpus. As you can see this is clearly a bug in behaviour and has been a showstopper for many trying to move from 2.4. Cheers, Con
Attachment:
pgpuoXgLl753F.pgp
Description: PGP signature
- Follow-Ups:
- Re: 2.6.12-rc6-mm1
- From: "Martin J. Bligh" <[email protected]>
- Re: 2.6.12-rc6-mm1
- References:
- Re: 2.6.12-rc6-mm1
- From: Andrew Morton <[email protected]>
- Re: 2.6.12-rc6-mm1
- From: Con Kolivas <[email protected]>
- Re: 2.6.12-rc6-mm1
- From: "Martin J. Bligh" <[email protected]>
- Re: 2.6.12-rc6-mm1
- Prev by Date: Re: [PATCH] local_irq_disable removal
- Next by Date: Re: [PATCH] local_irq_disable removal
- Previous by thread: Re: 2.6.12-rc6-mm1
- Next by thread: Re: 2.6.12-rc6-mm1
- Index(es):