From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Kara Subject: Re: [PATCH 0/8 v3] Checkpointing fixes and cleanups Date: Wed, 29 Feb 2012 12:03:42 +0100 Message-ID: <20120229110342.GC18846@quack.suse.cz> References: <1329330854-14237-1-git-send-email-jack@suse.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-ext4@vger.kernel.org To: Ted Tso Return-path: Received: from cantor2.suse.de ([195.135.220.15]:40457 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757734Ab2B2LDp (ORCPT ); Wed, 29 Feb 2012 06:03:45 -0500 Content-Disposition: inline In-Reply-To: <1329330854-14237-1-git-send-email-jack@suse.cz> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Wed 15-02-12 19:34:06, Jan Kara wrote: > Hello, > > this is a third round of my checkpointing fixes. The patch set is primarily > aimed at fixing filesystem corruption happening occasionally (rather rarely) > after power failure. First three patches should fix the issue. Patches 4-7 are > assorted checkpointing cleanups I've gathered when inspecting checkpointing > code. Finally patch 8 is a possible speedup of checkpoining - we can use cache > flushes happening during transaction commits for pushing the journal tail > safely. The observable speedup is disputable since jbd2_cleanup_journal_tail() > is called rather rarely (for metadata heavy load I saw about one > jbd2_cleanup_journal_tail() for about 200 commits) so the cost of additional > cache flush will be likely in the noise. But the patch is simple enough so I > send it for others to judge whether it makes sense or not. > > Changes since v2: > Rewrote the code pushing log tail to fix race described by Ted. > Improved some comments Ted, do you plan to have a look at this? It would be nice to get these fixes in for the next merge window... Honza -- Jan Kara SUSE Labs, CR