From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dave Chinner Subject: Re: [xfs] f876e446: +1.2% fileio.requests_per_sec Date: Tue, 4 Mar 2014 10:17:40 +1100 Message-ID: <20140303231740.GA6851@dastard> References: <20140303142151.GA7935@localhost> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Dave Chinner , linux-fsdevel@vger.kernel.org, LKML To: Fengguang Wu Return-path: Received: from ipmail05.adl6.internode.on.net ([150.101.137.143]:27402 "EHLO ipmail05.adl6.internode.on.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755182AbaCCXXU (ORCPT ); Mon, 3 Mar 2014 18:23:20 -0500 Content-Disposition: inline In-Reply-To: <20140303142151.GA7935@localhost> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Mon, Mar 03, 2014 at 10:21:51PM +0800, Fengguang Wu wrote: > Hi Dave, > > We noticed the below changes on commit f876e44603ad091c840a5fae5b0753bbb421c037 > ("xfs: always do log forces via the workqueue"): > > Basically there are +1.2% increased throughput with +29.5% increased > context switches. Yes, the increase in context switches is expected. Every sync IO causes a log force if it changed inode metadata, and the above commit, as it describes, moves them to a work queue. Hence there's more context switches involved in a log force that has to do work. Cheers, Dave. -- Dave Chinner david@fromorbit.com