From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Garzik Subject: Re: [PATCH] libata: write cache and read ahead Date: Wed, 24 Aug 2005 06:06:29 -0400 Message-ID: <430C46A5.1050702@pobox.com> References: <43085209.50405@torque.net> <430914C2.8030301@pobox.com> <43095EB3.2020202@torque.net> <430BAFD3.9000702@adaptec.com> <20050824091641.GD26447@infradead.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mail.dvmed.net ([216.237.124.58]:31890 "EHLO mail.dvmed.net") by vger.kernel.org with ESMTP id S1750700AbVHXKGd (ORCPT ); Wed, 24 Aug 2005 06:06:33 -0400 In-Reply-To: <20050824091641.GD26447@infradead.org> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Christoph Hellwig Cc: Luben Tuikov , dougg@torque.net, linux-scsi@vger.kernel.org, Tejun Heo Christoph Hellwig wrote: > On Tue, Aug 23, 2005 at 07:22:59PM -0400, Luben Tuikov wrote: > >>On 08/22/05 01:12, Douglas Gilbert wrote: >> >>>I was surprised how much code needed changing. >>>With MODE SELECT's issues with libata addressed >>>various other SAT "extras" should be much easier >>>to implement. That should make libata more attractive >>>as a SAT layer for SAS LLDDs (that don't do it already >>>in firmware). >> >>Doug, how about never needing a SAT layer for SAS LLDDs >>for ATA/ATAPI devices. >> >>Would you object if I "give" you a "domain device" to which >>you can send FISes (+ the packet if ATAPI) through >>a SAM-4 intefrace: Execute Task + TMFs? > > > Personally I think that would be nice, but when I mentioned the basic idea > (sending FISes through BLOCK_PC block later commands and let the scsi layer > handle then as vendor-specific commands) Jeff didn't like that because > then we'd need a non-sd upper driver although using the scsi midlayer. That's something different than when Luben was talking about. Addressing what you [Christoph] are talking about, the block layer needs to grow a generic method for submitting device commands, be they ATA, SCSI, SCSI w/ variable-length CDB, I2O, Promise SX8 message, whatever. Stuff that into the request_queue, and the transport driver handles it from there. Jeff