From: Nigel Cunningham Subject: Re: [PATCH] Introduce freeze_super and thaw_super for the fsfreeze ioctl Date: Wed, 24 Mar 2010 16:16:56 +1100 Message-ID: <4BA9A048.6050705@crca.org.au> References: <20100323142200.GA2381@localhost.localdomain> <4BA93FE3.9020502@crca.org.au> <20100324011714.GA5658@localhost.localdomain> 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, chris.mason@oracle.com, viro@ZenIV.linux.org.uk, hch@lst.de To: Josef Bacik Return-path: Received: from crca.org.au ([74.207.252.120]:38983 "EHLO crca.org.au" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751561Ab0CXFQI (ORCPT ); Wed, 24 Mar 2010 01:16:08 -0400 In-Reply-To: <20100324011714.GA5658@localhost.localdomain> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: Hi. On 24/03/10 12:17, Josef Bacik wrote: > On Wed, Mar 24, 2010 at 09:25:39AM +1100, Nigel Cunningham wrote: >> Hi. >> >> A cc would have been nice :) >> >> Thankfully the subject caught my eye. >> > > Well it's not really tux-on-ice's fault, freeze wouldn't work _at all_ via the > ioctl or any other method, we just happened to notice it was broken because > people using tux-on-ice were getting corrupt filesystems after suspend :). Btw, > it may be good to have freezer_sync set to 1 by default, otherwise if an fs > doesnt support freezing it will likely end up corrupted too. Thanks, Corrupted because the sys_sync was disabled? (We were recently discussing adding a tuneable to swsusp for disabling sys_sync there too). I've just committed a patch making freezer_sync default to 1. Regards, Nigel