From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bart Van Assche Subject: Re: [PATCH 5/4] scsi: remove MSG_*_TAG defines Date: Mon, 01 Dec 2014 12:44:23 +0100 Message-ID: <547C5497.5020706@acm.org> References: <1416839781-26988-1-git-send-email-hch@lst.de> <20141124150725.GA23165@infradead.org> <54784762.8020304@acm.org> <20141201090315.GA4042@infradead.org> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from sf2.bxl.stone.is ([5.134.1.239]:50919 "EHLO sf2.bxl.stone.is" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753013AbaLALot (ORCPT ); Mon, 1 Dec 2014 06:44:49 -0500 In-Reply-To: <20141201090315.GA4042@infradead.org> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Christoph Hellwig Cc: linux-scsi@vger.kernel.org, Doug Ledford On 12/01/14 10:03, Christoph Hellwig wrote: > On Fri, Nov 28, 2014 at 10:58:58AM +0100, Bart Van Assche wrote: >>> if (disable_syn_offset_one_fix) { >>> - scsiq->q2.tag_code &= ~MSG_SIMPLE_TAG; >>> + scsiq->q2.tag_code &= ~SIMPLE_QUEUE_TAG; >>> scsiq->q2.tag_code |= (ASC_TAG_FLAG_DISABLE_ASYN_USE_SYN_FIX | >>> ASC_TAG_FLAG_DISABLE_DISCONNECT); >>> } else { >> >> One of the statements in advansys.c (tag_code &= 0xDC) makes assumptions >> about the numeric value of MSG_SIMPLE_TAG. This makes me wonder whether it >> is safe to change MSG_SIMPLE_TAG into another value ? > > MSG_SIMPLE_TAG and SIMPLE_QUEUE_TAG are two different symbolic names > for the 0x20 value assigned to the SCSI-2 protocol "SIMPLE QUEUE TAG" > message. Having both is confusing, so I decided to only keep the one > that is named closer to the spec, and in the same namespace as the > symbolic names we use for other SCSI-2 messages. Thanks for the clarification. Using different namespaces for symbolic constants for initiator and target drivers definitely looks like a good idea to me. The patch itself also looks fine to me. Hence: Reviewed-by: Bart Van Assche