From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jens Axboe Subject: Re: [PATCH 00/26] AIO performance improvements/cleanups, v2 Date: Sat, 15 Dec 2012 14:16:48 +0100 Message-ID: <50CC7840.4090403@fusionio.com> References: <1354568322-29029-1-git-send-email-koverstreet@google.com> <20121213211808.GJ25017@kernel.dk> <50CAD6D9.5070703@fusionio.com> <20121215092526.GA10411@moria.home.lan> <50CC46F8.3070600@fusionio.com> <20121215103658.GB10411@moria.home.lan> Mime-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit Cc: Jack Wang , "linux-kernel@vger.kernel.org" , "linux-aio@kvack.org" , "linux-fsdevel@vger.kernel.org" , "zab@redhat.com" , "bcrl@kvack.org" , "jmoyer@redhat.com" , "viro@zeniv.linux.org.uk" To: Kent Overstreet Return-path: Received: from mx1.fusionio.com ([66.114.96.30]:55688 "EHLO mx1.fusionio.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756520Ab2LONQx (ORCPT ); Sat, 15 Dec 2012 08:16:53 -0500 In-Reply-To: <20121215103658.GB10411@moria.home.lan> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On 2012-12-15 11:36, Kent Overstreet wrote: >> Knock yourself out - I already took a quick look at it, and conversion >> should be pretty simple. It's the mtip32xx driver, it's in the kernel. I >> would suggest getting rid of the ->async_callback() (since it's always >> bio_endio()) since that'll make it cleaner. > > Just pushed my conversion - it's untested, but it's pretty > straightforward. You forgot a batch_complete_init(). With that, it works. Single device is ~1050K now, so still slower than jaio without batching (which was ~1220K). But it's an improvement over kaio-dio, which was roughly ~930K IOPS. -- Jens Axboe