From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nick Piggin Subject: Re: [PATCH] as i/o hang with aacraid driver 2.6.0-test1 Date: Thu, 31 Jul 2003 17:40:45 +1000 Sender: linux-scsi-owner@vger.kernel.org Message-ID: <3F28C7FD.9030204@cyberone.com.au> References: <20030716132036.GB833@suse.de> <1058364455.1856.28.camel@mulgrave> <20030716 170456.GK833@suse.de> <20030717015756.135a3f5a.akpm@osdl.org> <2003071708595 2.GX833@suse.de> <3F1672D9.7070309@cyberone.com.au> <20030717102926.GE833@su se.de> <3F167F98.60006@cyberone.com.au> <20030717105641.GF833@suse.de> <3F16 83F5.4030107@cyberone.com.au> <20030717111059.GI833@suse.de> <3F168846.90902 @cyberone.com.au> <1058474814.4638.11.camel@markh1.pdx.osdl.net> <1058481553 .19508.5.camel@markh1.pdx.osdl.net> <1058485621.7424.30.camel@dell_ss5.pdx.o sdl.net> <20030717170055.5dbe20c1.akpm@osdl.org> <3F17821A.307@cyberone.com. au> <20030717222534.0295c44e.akpm@osdl.org> <3F1785D9.5000905@cyberone.com.au> <1058537776.1863.6.camel@mulgrave> <20030718093006.5e07b6e2.akpm@osdl.org> <1058546495.1863.57.camel@mulgrave> <1058549104.19558.63.camel@dhcp22.swansea.linux.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from dyn-ctb-210-9-243-68.webone.com.au ([210.9.243.68]:28947 "EHLO chimp.local.net") by vger.kernel.org with ESMTP id S272807AbTGaHlK (ORCPT ); Thu, 31 Jul 2003 03:41:10 -0400 In-Reply-To: <1058549104.19558.63.camel@dhcp22.swansea.linux.org.uk> List-Id: linux-scsi@vger.kernel.org To: Alan Cox Cc: James Bottomley , Andrew Morton , daniel@osdl.org, markh@osdl.org, Jens Axboe , cliffw@osdl.org, SCSI Mailing List Alan Cox wrote: >On Gwe, 2003-07-18 at 17:41, James Bottomley wrote: > >>On Fri, 2003-07-18 at 11:30, Andrew Morton wrote: >> >>>Nick is probably on a plane by now. I sent Mark's original one-liner to >>>Linus last night. >>> >>That's not a fix, it only catches one out of the five requeue cases in >>SCSI. >> >>why don't we just use the deadline elevator until this can be sorted out >>properly? >> > >For aacraid you can just throw commands at the card, its got a 233Mhz >ARM and a mind of its own anyway > > The problem that AS solves isn't that a card/disk doesn't know what to do with the requests it has, its that they don't know what requests they are going to get (likely to get). On the other hand high end SCSI devices running database stuff often take some performance hit when using AS vs deadline.