From: Jan Kara Subject: Re: [PATCH v2] fs/ext4: increase parallelism in updating ext4 orphan list Date: Tue, 15 Apr 2014 19:25:22 +0200 Message-ID: <20140415172522.GA13276@quack.suse.cz> References: <1380728283-61038-1-git-send-email-tmac@hp.com> <1396456148-20455-1-git-send-email-tmac@hp.com> <20140402174109.GD8657@quack.suse.cz> <534C135A.9010803@hp.com> <20140414174024.GC13860@quack.suse.cz> <534D5E02.6030500@hp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Jan Kara , T Makphaibulchoke , tytso@mit.edu, adilger.kernel@dilger.ca, linux-ext4@vger.kernel.org, linux-kernel@vger.kernel.org, aswin@hp.com To: Thavatchai Makphaibulchoke Return-path: Received: from cantor2.suse.de ([195.135.220.15]:60033 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750866AbaDORZZ (ORCPT ); Tue, 15 Apr 2014 13:25:25 -0400 Content-Disposition: inline In-Reply-To: <534D5E02.6030500@hp.com> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Tue 15-04-14 10:27:46, Thavatchai Makphaibulchoke wrote: > On 04/14/2014 11:40 AM, Jan Kara wrote: > > Thanks for trying that out! Can you please send me a patch you have been > > testing? Because it doesn't quite make sense to me why using i_mutex should > > be worse than using hashed locks... > > > > Thanks again for the comments. > > Since i_mutex is also used for serialization in other operations on an > inode, in the case that the i_mutex is not held using it for > serialization could cause contention with other operations on the inode. > As the number shows substantial instances of orphan add or delete calls > without holding the i_mutex, I presume the performance degradation is due > to the contention. I have checked the source and I didn't find many places where i_mutex was not held. But maybe I'm wrong. That's why I wanted to see the patch where you are using i_mutex instead of hashed mutexes and which didn't perform good enough. > As for the patch, could you please let me know if you need the patch > using i_mutex or the patch I'm planning to submit. If it's the latter, > I'm thinking of go ahead and resubmit it. Honza -- Jan Kara SUSE Labs, CR