From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jens Axboe Subject: Re: Alternative TRIM proposal Date: Tue, 7 Oct 2008 13:39:31 +0200 Message-ID: <20081007113931.GU19428@kernel.dk> References: <200810021732.m92HW7qL015836@coles02.co.lsil.com> <20081003154237.GT13822@parisc-linux.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from pasmtpb.tele.dk ([80.160.77.98]:36757 "EHLO pasmtpB.tele.dk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753007AbYJGLkB (ORCPT ); Tue, 7 Oct 2008 07:40:01 -0400 Content-Disposition: inline In-Reply-To: <20081003154237.GT13822@parisc-linux.org> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Matthew Wilcox Cc: Gerry.Houlder@seagate.com, dougg@torque.net, "Knight, Frederick" , linux-scsi@vger.kernel.org, t10@t10.org, David Woodhouse On Fri, Oct 03 2008, Matthew Wilcox wrote: > It is today the case in Linux that each extent will be sent from the > filesystem to the block layer individually. I can't speak to other > operating systems, maybe some of them allow filesystems to send a list > of extents to the SCSI layer, maybe their SCSI layer can take extents > out of the queue and bundle them together into a single command. We do that for lots of regular IO as well, but still support merging them into a single command - the same could be done for trim, basically making any of them mergable into a single command. -- Jens Axboe