From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tao Ma Subject: Re: [PATCH v2] ocfs2: Let ocfs2_setattr use new truncate sequence. Date: Thu, 10 Jun 2010 13:06:27 +0800 Message-ID: <4C1072D3.4020109@oracle.com> References: <1276141986-9200-1-git-send-email-tao.ma@oracle.com> <20100610044234.GD26335@laptop> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, Joel Becker , Christoph Hellwig To: Nick Piggin Return-path: Received: from rcsinet10.oracle.com ([148.87.113.121]:19119 "EHLO rcsinet10.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751797Ab0FJFHL (ORCPT ); Thu, 10 Jun 2010 01:07:11 -0400 In-Reply-To: <20100610044234.GD26335@laptop> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: Hi Nick, On 06/10/2010 12:42 PM, Nick Piggin wrote: > On Thu, Jun 10, 2010 at 11:53:06AM +0800, Tao Ma wrote: >> Let ocfs2 use the new truncate sequence. The changes include: >> 1. Remove the extra check for inode_newsize_ok since Christoph >> has moved it into inode_change_ok. So we will check it at the >> beginning of ocfs2_setattr. > > So this deals with our questions regarding check of i_size outside > the inode cluster lock? (see fsdevel discussion) oh, I forget about this. yes, we should have cluster lock and shouldn't remove this check. > > >> 2. Use truncate_setsize directly since we don't implement our >> own ->truncate and what we need is "update i_size and >> truncate_pagecache" which truncate_setsize now does. >> 3. For direct write, ocfs2 actually don't allow write to pass >> i_size(see ocfs2_prepare_inode_for_write), so we don't have >> a chance to increase i_size. So remove the bogus check. >> >> Cc: Joel Becker >> Cc: Christoph Hellwig >> Cc: Nick Piggin >> Signed-off-by: Tao Ma >> --- >> fs/ocfs2/file.c | 34 +++++----------------------------- >> 1 files changed, 5 insertions(+), 29 deletions(-) >> >> diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c >> index 1fb0985..764fffb 100644 >> --- a/fs/ocfs2/file.c >> +++ b/fs/ocfs2/file.c >> @@ -983,10 +983,6 @@ int ocfs2_setattr(struct dentry *dentry, struct iattr *attr) >> } >> >> if (size_change&& attr->ia_size != i_size_read(inode)) { >> - status = inode_newsize_ok(inode, attr->ia_size); >> - if (status) >> - goto bail_unlock; >> - >> if (i_size_read(inode)> attr->ia_size) { > > While you're here, you should be able to use inode->i_size if you're > under i_mutex, no? ok, will change it and the correpsonding part in truncate_setsize. Regards, Tao