On Wed, 2007-01-17 at 15:29 +0100, Peter Zijlstra wrote: > I was thinking that since the server needs to actually sync the page a > commit might be quite expensive (timewise), hence I didn't want to flush > too much, and interleave them with writing out some real pages to > utilise bandwidth. Most servers just call fsync()/fdatasync() whenever we send a COMMIT, in which case the extra round trips are just adding unnecessary latency. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
- References:
- [RFC 0/8] Cpuset aware writeback
- From: Christoph Lameter <[email protected]>
- Re: [RFC 0/8] Cpuset aware writeback
- From: Andrew Morton <[email protected]>
- [PATCH] nfs: fix congestion control
- From: Peter Zijlstra <[email protected]>
- Re: [PATCH] nfs: fix congestion control
- From: Trond Myklebust <[email protected]>
- Re: [PATCH] nfs: fix congestion control
- From: Peter Zijlstra <[email protected]>
- Re: [PATCH] nfs: fix congestion control
- From: Trond Myklebust <[email protected]>
- Re: [PATCH] nfs: fix congestion control
- From: Peter Zijlstra <[email protected]>
- Re: [PATCH] nfs: fix congestion control
- From: Trond Myklebust <[email protected]>
- Re: [PATCH] nfs: fix congestion control
- From: Peter Zijlstra <[email protected]>
- [RFC 0/8] Cpuset aware writeback
- Prev by Date: Re: [take33 10/10] kevent: Kevent based AIO (aio_sendfile()/aio_sendfile_path()).
- Next by Date: Re: data corruption with nvidia chipsets and IDE/SATA drives (k8 cpu errata needed?)
- Previous by thread: Re: [PATCH] nfs: fix congestion control
- Next by thread: Re: [PATCH] nfs: fix congestion control
- Index(es):
 |