From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Garzik Subject: Re: [PATCH 14/17] sym53c8xx: Get rid of IRQ_FMT and IRQ_PRM Date: Tue, 09 Oct 2007 11:47:03 -0400 Message-ID: <470BA277.2000308@garzik.org> References: <11916141143747-git-send-email-matthew@wil.cx> <11916141153666-git-send-email-matthew@wil.cx> <4706A5C3.6000005@garzik.org> <20071009152846.GF13842@parisc-linux.org> <470B9FB2.9010300@garzik.org> <20071009154413.GH13842@parisc-linux.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from srv5.dvmed.net ([207.36.208.214]:52171 "EHLO mail.dvmed.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752137AbXJIPrJ (ORCPT ); Tue, 9 Oct 2007 11:47:09 -0400 In-Reply-To: <20071009154413.GH13842@parisc-linux.org> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Matthew Wilcox Cc: linux-scsi@vger.kernel.org, Matthew Wilcox Matthew Wilcox wrote: > On Tue, Oct 09, 2007 at 11:35:14AM -0400, Jeff Garzik wrote: >> Not true. The original code used 'int irq', and the printf format was >> correct as a result. >> >> You changed the code to use 'unsigned int irq' as found in struct >> pci_dev, without changing IRQ_FMT as needed. >> >> If you wanted to avoid the regression another way, I suppose patch #1 >> should update IRQ_FMT to use "%u", and patch #14 should move the "%u" >> use to its caller. > > What do you think the regression is, exactly? > > The *only* difference in how vsnprintf handles %d vs %u is the sign bit. Because when you change the type, you should change the printf format to match? It used to match, before your patchset. Now it doesn't. Jeff