From mboxrd@z Thu Jan 1 00:00:00 1970 From: tytso@mit.edu Subject: Re: [PATCH 0/3] ext4: don't use quota reservation for speculative metadata blocks Date: Sun, 11 Apr 2010 22:20:27 -0400 Message-ID: <20100412022027.GK1849@thunk.org> References: <4BBCFD10.3030504@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: ext4 development , Jan Kara To: Eric Sandeen Return-path: Received: from thunk.org ([69.25.196.29]:54147 "EHLO thunker.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751784Ab0DLCUa (ORCPT ); Sun, 11 Apr 2010 22:20:30 -0400 Content-Disposition: inline In-Reply-To: <4BBCFD10.3030504@redhat.com> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Wed, Apr 07, 2010 at 04:45:52PM -0500, Eric Sandeen wrote: > Because we can badly over-reserve metadata when we > calculate worst-case, it complicates things for quota, since > we must reserve and then claim later, retry on EDQUOT, etc. > Quota is also a generally smaller pool than fs free blocks, > so this over-reservation hurts more, and more often. > > I'm of the opinion that it's not the worst thing to allow > metadata to push a user slightly over quota. This simplifies > the code and avoids the false quota rejections that result > from worst-case speculation. This patch series looks good to me in general; Jan, it requires relatively minor changes to the quota system, so it would be good to get your Acked-by for the first two patches. Since the changes to the ext4 layer are more in-depth, any objections if I carry all three patches in the ext4 tree? - Ted