Messages in this thread Patch in this message | | | Subject | Re: bdflush/rpciod high CPU utilization, profile does not make sense | From | Trond Myklebust <> | Date | Thu, 07 Apr 2005 12:17:51 -0400 |
| |
to den 07.04.2005 Klokka 17:38 (+0200) skreiv Jakob Oestergaard:
> I tweaked the VM a bit, put the following in /etc/sysctl.conf: > vm.dirty_writeback_centisecs=100 > vm.dirty_expire_centisecs=200 > > The defaults are 500 and 3000 respectively... > > This improved things a lot; the client is now "almost not very laggy", > and load stays in the saner 1-2 range.
OK. That hints at what is causing the latencies on the server: I'll bet it is the fact that the page reclaim code tries to be clever, and uses NFSv3 STABLE writes in order to be able to free up the dirty pages immediately. Could you try the following patch, and see if that makes a difference too?
Cheers, Trond ---- fs/nfs/write.c | 2 +- 1 files changed, 1 insertion(+), 1 deletion(-)
Index: linux-2.6.12-rc2/fs/nfs/write.c =================================================================== --- linux-2.6.12-rc2.orig/fs/nfs/write.c +++ linux-2.6.12-rc2/fs/nfs/write.c @@ -305,7 +305,7 @@ do_it: if (err >= 0) { err = 0; if (wbc->for_reclaim) - nfs_flush_inode(inode, 0, 0, FLUSH_STABLE); + nfs_flush_inode(inode, 0, 0, 0); } } else { err = nfs_writepage_sync(ctx, inode, page, 0,
-- Trond Myklebust <trond.myklebust@fys.uio.no>
- 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/
| |