From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: [PATCH] change scsi_cmd_ioctl to take a gendisk instead of a queue Date: Sun, 1 Feb 2004 22:13:55 +0000 Sender: linux-scsi-owner@vger.kernel.org Message-ID: <20040201221355.A18561@infradead.org> References: <20040201124803.A18050@beaverton.ibm.com> <20040201213334.A17988@infradead.org> <20040201134916.A18509@beaverton.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from phoenix.infradead.org ([213.86.99.234]:3853 "EHLO phoenix.infradead.org") by vger.kernel.org with ESMTP id S265479AbUBAWOA (ORCPT ); Sun, 1 Feb 2004 17:14:00 -0500 Content-Disposition: inline In-Reply-To: <20040201134916.A18509@beaverton.ibm.com>; from patmans@us.ibm.com on Sun, Feb 01, 2004 at 01:49:16PM -0800 List-Id: linux-scsi@vger.kernel.org To: Patrick Mansfield Cc: Christoph Hellwig , Jens Axboe , linux-scsi@vger.kernel.org On Sun, Feb 01, 2004 at 01:49:16PM -0800, Patrick Mansfield wrote: > Yes. That would also clean up the upper level driver code (for example, > sd.c's sd_init_command), but I have not tried this. Heh, I just took a short look at sd_rw_intr vs rw_intr (sr) and moving those into common code defintily looks like a good thing. There's a bunch of small difference, but I think we'd better off having those two unified instead of two sets of bugs. It's probably worth investigating. > Then in the future we could use sg_io instead of the scsi_do_req, > scsi_wait_req, etc., and get rid of REQ_CMD in favor of REQ_BLOCK_PC. Hmmm.. Sound like an interesting path to investigate. But unlike the REQ_PC unification it sounds more like a 2.7 thing.