On Thu, 2 Jun 2005 07:37, Con Kolivas wrote: > On Thu, 2 Jun 2005 04:41, Steve Rotolo wrote: > > I guess the bottom-line is: given N logical cpus, 1/N of all > > SCHED_NORMAL tasks may get stuck on a sibling cpu with no chance to > > run. All it takes is one spinning SCHED_FIFO task. Sounds like a bug. > > You're right, and excuse me for missing it. Oh and thanks for picking it up! Cheers, Con
Attachment:
pgpkVIECYnHiZ.pgp
Description: PGP signature
- References:
- SD_SHARE_CPUPOWER breaks scheduler fairness
- From: Steve Rotolo <[email protected]>
- Re: SD_SHARE_CPUPOWER breaks scheduler fairness
- From: Steve Rotolo <[email protected]>
- Re: SD_SHARE_CPUPOWER breaks scheduler fairness
- From: Con Kolivas <[email protected]>
- SD_SHARE_CPUPOWER breaks scheduler fairness
- Prev by Date: Re: RT patch acceptance
- Next by Date: Re: 2.6.12-rc5-mm2 JFS problems ?
- Previous by thread: Re: SD_SHARE_CPUPOWER breaks scheduler fairness
- Next by thread: Re: SD_SHARE_CPUPOWER breaks scheduler fairness
- Index(es):