From: Eric Sandeen Subject: Re: [PATCH] ext4: sync the directory inode in ext4_sync_parent() Date: Tue, 05 Apr 2011 12:46:07 -0700 Message-ID: <4D9B717F.3010300@redhat.com> References: <1302031663-13547-1-git-send-email-curtw@google.com> <4D9B6F0A.5080705@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: tytso@mit.edu, linux-ext4@vger.kernel.org To: Curt Wohlgemuth Return-path: Received: from mx1.redhat.com ([209.132.183.28]:60033 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752954Ab1DETqM (ORCPT ); Tue, 5 Apr 2011 15:46:12 -0400 In-Reply-To: Sender: linux-ext4-owner@vger.kernel.org List-ID: On 4/5/11 12:43 PM, Curt Wohlgemuth wrote: > Well, you could argue that a failure to sync the parent shouldn't > cause the inode's fsync() to fail, but I probably wouldn't :-) . I'll > resend a patch that implements this. > > Thanks, > Curt It might be a valid argument, since we're doing it "just to be nice" - but if you're getting EIOs you'd probably rather know sooner than later... -Eric