Nick Piggin writes:
Secondly, with or without swap prefetch, I think we can do a better job of handling these use-once patterns to begin with.
Absolutely! But there will always be workloads that cause appropriate use-once swapping. Qemu, games, big print jobs, big open documents, big graphic edits and so on.
Cheers, Con
Attachment:
pgpRtIvKKW5qE.pgp
Description: PGP signature
- Follow-Ups:
- Re: [ck] swap prefetching merge plans
- From: Con Kolivas <[email protected]>
- Re: [ck] swap prefetching merge plans
- References:
- -mm merge plans
- From: Andrew Morton <[email protected]>
- swap prefetching merge plans
- From: Con Kolivas <[email protected]>
- Re: [ck] swap prefetching merge plans
- From: jos poortvliet <[email protected]>
- Re: [ck] swap prefetching merge plans
- From: Nick Piggin <[email protected]>
- -mm merge plans
- Prev by Date: Re: passing command line args to the kernel
- Next by Date: RE: 2.6.16-rc5: known regressions [TP 600X S3, vanilla DSDT]
- Previous by thread: Re: [ck] swap prefetching merge plans
- Next by thread: Re: [ck] swap prefetching merge plans
- Index(es):
![]() |