From mboxrd@z Thu Jan 1 00:00:00 1970 From: Boaz Harrosh Subject: Re: [PATCH 3/3] scsi_lib: Collapse scsi_end_request into only user Date: Mon, 04 Jan 2010 15:56:32 +0200 Message-ID: <4B41F390.9060108@panasas.com> References: <4B41DADD.4000102@panasas.com> <4B41DC42.1090706@panasas.com> <20100104132338.GA9882@parisc-linux.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Return-path: Received: from daytona.panasas.com ([67.152.220.89]:42417 "EHLO daytona.int.panasas.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1750738Ab0ADN4g (ORCPT ); Mon, 4 Jan 2010 08:56:36 -0500 In-Reply-To: <20100104132338.GA9882@parisc-linux.org> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Matthew Wilcox Cc: James Bottomley , Alan Stern , Mike Christie , Hannes Reinecke , linux-scsi On 01/04/2010 03:23 PM, Matthew Wilcox wrote: > On Mon, Jan 04, 2010 at 02:17:06PM +0200, Boaz Harrosh wrote: >> Embedding scsi_end_request() into scsi_io_completion actually simplifies the >> code and makes it clearer what's going on. > > I'm not entirely convinced about that -- scsi_io_completion is currently > over 200 lines long and needs to be made shorter, not longer. That said, > I see no reason that the current factoring of scsi_io_completion() makes > sense; pushing the decoding of the sense key into a separate function > looks like a more profitable idea. It would also let you do without > the nasty gotos you add in this patch. > I agree, but I want that Alan's retries handling, Hannes's sense for FS_COMMANDS and mainly the question about good_bytes!=0 and retries are answered first. Then we can see what the code is really like and make a clean cut. I'll attempt a second version without the goto. >> There is absolutely no functional and/or side effects changes after this patch. >> >> Patch was inspired by Alan Stern. >> >> CC: Alan Stern >> Signed-off-by: Boaz Harrosh Boaz