From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263793AbTLXS4W (ORCPT ); Wed, 24 Dec 2003 13:56:22 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263795AbTLXS4W (ORCPT ); Wed, 24 Dec 2003 13:56:22 -0500 Received: from mta7.pltn13.pbi.net ([64.164.98.8]:7602 "EHLO mta7.pltn13.pbi.net") by vger.kernel.org with ESMTP id S263793AbTLXS4V (ORCPT ); Wed, 24 Dec 2003 13:56:21 -0500 Date: Wed, 24 Dec 2003 10:56:06 -0800 From: Mike Fedyk To: Neil Brown Cc: Jim Lawson , linux-kernel@vger.kernel.org Subject: Re: 2.6.0, SiI3112, md raid1 problem: bio too big device (128 > 15) Message-ID: <20031224185606.GZ6438@matchmail.com> Mail-Followup-To: Neil Brown , Jim Lawson , linux-kernel@vger.kernel.org References: <16361.28564.275984.580859@notabene.cse.unsw.edu.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <16361.28564.275984.580859@notabene.cse.unsw.edu.au> User-Agent: Mutt/1.5.4i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Dec 24, 2003 at 09:51:00PM +1100, Neil Brown wrote: > This is a raid1 bug. > Some block devices have limits on the sizes of io requests that they > can handle, and advertise these limits with ->max_sectors and > ->merge_bvec_fn (and a few others). Any device MUST be able to accept > a single page IO at any offset. So, how can I determine the max io request support by my underlying devices from userspace? > When raid1 does a 'resync' it does all IO in 64K requests, ignoring > the restrictions. > Getting raid1 resync to handle the restrictions is non-trivial (I have > code, but it is still buggy). Do you have patches posted anywhere? > A simple interim fix is to replace > > #define RESYNC_BLOCK_SIZE (64*1024) > > with > > #define RESYNC_BLOCK_SIZE PAGE_SIZE Then by all means, let's get this into -mm. Then once the new resync code is ready it can be merged.