From owner-freebsd-current@FreeBSD.ORG Thu Jul 7 09:30:02 2005 Return-Path: X-Original-To: current@FreeBSD.ORG Delivered-To: freebsd-current@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id AB69516A41C for ; Thu, 7 Jul 2005 09:30:02 +0000 (GMT) (envelope-from sos@DeepCore.dk) Received: from spider.deepcore.dk (cpe.atm2-0-53484.0x50a6c9a6.abnxx9.customer.tele.dk [80.166.201.166]) by mx1.FreeBSD.org (Postfix) with ESMTP id EEDD743D49 for ; Thu, 7 Jul 2005 09:30:01 +0000 (GMT) (envelope-from sos@DeepCore.dk) Received: from [172.18.2.1] (axiell-gw1.novi.dk [130.225.63.24]) by spider.deepcore.dk (8.13.3/8.13.3) with ESMTP id j679Mdp8090107; Thu, 7 Jul 2005 11:22:40 +0200 (CEST) (envelope-from sos@DeepCore.dk) Message-ID: <42CCF630.5070308@DeepCore.dk> Date: Thu, 07 Jul 2005 11:30:24 +0200 From: =?ISO-8859-1?Q?S=F8ren_Schmidt?= User-Agent: Mozilla Thunderbird 1.0 (X11/20050116) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Luigi Rizzo References: <20050706143050.A96472@xorpc.icir.org> In-Reply-To: <20050706143050.A96472@xorpc.icir.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: quoted-printable X-mail-scanned: by DeepCore Virus & Spam killer v1.12 Cc: current@FreeBSD.ORG Subject: Re: disksort disappeared from ATA devices in HEAD ? X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 07 Jul 2005 09:30:02 -0000 Luigi Rizzo wrote: > I notice that the ata drivers in HEAD do not use anymore > bioq_disksort, but rather they have their own routine, > (ad_queue_request() in turn calling ata_sort_queue()) > which probably does something similar. >=20 > I wonder: > 1) how different is ata_sort_queue() from bioq_disksort() > in terms of design ? The difference is that ata_sort_queue() has a way to make certain that=20 new requests are inserted after a specific point in the ATA queue=20 (freezepoint). It also knows how to deal with non-r/w operations. This is needed to make "composite" ATA operations work properly.=20 Composite ATA operations are several ATA operations chained together=20 over different ATA channels, which are *real* handy for RAID operations e= tc. > 2) is there any strong reason for this change ? Since this is special usage for ATA ops alone I didn't want to pollute=20 the system disksort with this functionality. If/when we grow a way to=20 register specialized sort routines with the system that could perhaps be = used then. > 3) any pointers to mail threads where this changes have been > discussed ? Nope. This was my decision during the development of ATA mkIII. -- -S=F8ren