Re: 2.6.6-rc{1,2} bad VM/NFS interaction in case of dirty page writeback

[Trond Myklebust <trond.myklebust@fys.uio.no>]

On Tue, 2004-04-27 at 21:02, Andrew Morton wrote:
> Shantanu Goel <sgoel01@yahoo.com> wrote:
> >
> > Andrew/Trond,
> > 
> > Any consensus as to what the right approach is here?
> 
> For now I suggest you do
> 
> -	err = WRITEPAGE_ACTIVATE;
> +	err = 0;
> 
> in nfs_writepage().

That will just cause the page to be put back onto the inactive list
without starting writeback. Won't that cause precisely those kswapd
loops that Shantanu was worried about?
AFAICS if you want to do this, you probably need to flush the page
immediately to disk on the server using a STABLE write as per the
appeanded patch. The problem is that screws the server over pretty hard
as it will get flooded with what are in effect a load of 4k O_SYNC
writes.

Cheers,
  Trond


--- linux-2.6.6-rc2/fs/nfs/write.c.orig	2004-04-27 16:01:01.000000000 -0400
+++ linux-2.6.6-rc2/fs/nfs/write.c	2004-04-27 21:18:58.000000000 -0400
@@ -313,7 +313,7 @@
 		if (err >= 0) {
 			err = 0;
 			if (wbc->for_reclaim)
-				err = WRITEPAGE_ACTIVATE;
+				nfs_flush_inode(inode, 0, 0, FLUSH_STABLE);
 		}
 	} else {
 		err = nfs_writepage_sync(NULL, inode, page, 0,