From mboxrd@z Thu Jan 1 00:00:00 1970 From: Trond Myklebust Subject: Re: [PATCH] NFS: return -EAGAIN when skipped commit in nfs_commit_unstable_pages() Date: Mon, 13 Jun 2011 15:38:06 -0400 Message-ID: <1307993886.5569.3.camel@lade.trondhjem.org> References: <20110610140459.GA8021@localhost> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 8BIT Cc: Christoph Hellwig , linux-nfs@vger.kernel.org, linux-fsdevel@vger.kernel.org, LKML To: Wu Fengguang Return-path: Received: from mx2.netapp.com ([216.240.18.37]:11318 "EHLO mx2.netapp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752160Ab1FMTiH convert rfc822-to-8bit (ORCPT ); Mon, 13 Jun 2011 15:38:07 -0400 In-Reply-To: <20110610140459.GA8021@localhost> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Fri, 2011-06-10 at 22:04 +0800, Wu Fengguang wrote: > It's confusing to return success while redirtying the inode at the > same time in ->write_inode(). Return -EAGAIN to indicate that we've > not finished with this inode. > > Impact: it's a cleanup, not bug fix. > > CC: Christoph Hellwig > Signed-off-by: Wu Fengguang > --- > fs/nfs/write.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > --- linux-next.orig/fs/nfs/write.c 2011-06-10 21:52:34.000000000 +0800 > +++ linux-next/fs/nfs/write.c 2011-06-10 21:52:37.000000000 +0800 > @@ -1521,7 +1521,7 @@ static int nfs_commit_unstable_pages(str > { > struct nfs_inode *nfsi = NFS_I(inode); > int flags = FLUSH_SYNC; > - int ret = 0; > + int ret = -EAGAIN; > > if (wbc->sync_mode == WB_SYNC_NONE) { > /* Don't commit yet if this is a non-blocking flush and there Looks good, but since this is a cleanup, I'm queueing it for 3.1 rather than pushing it in the regression-fixing window. Cheers Trond -- Trond Myklebust Linux NFS client maintainer NetApp Trond.Myklebust@netapp.com www.netapp.com