From mboxrd@z Thu Jan 1 00:00:00 1970 From: Daniel Walker Subject: Re: [PATCH] qla2xxx: firmware semaphore to mutex Date: Mon, 28 Apr 2008 12:54:58 -0700 Message-ID: <1209412498.12461.78.camel@localhost.localdomain> References: <20080428174742.163523860@mvista.com> <20080428180846.GX14990@parisc-linux.org> <1209409033.12461.64.camel@localhost.localdomain> <20080428190425.GY14990@parisc-linux.org> <1209410518.12461.71.camel@localhost.localdomain> <20080428192416.GZ14990@parisc-linux.org> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from gateway-1237.mvista.com ([63.81.120.158]:54903 "EHLO gateway-1237.mvista.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1765978AbYD1TzA (ORCPT ); Mon, 28 Apr 2008 15:55:00 -0400 In-Reply-To: <20080428192416.GZ14990@parisc-linux.org> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Matthew Wilcox Cc: James.Bottomley@HansenPartnership.com, mingo@elte.hu, linux-driver@qlogic.com, linux-scsi@vger.kernel.org, akpm@linux-foundation.org On Mon, 2008-04-28 at 13:24 -0600, Matthew Wilcox wrote: > On Mon, Apr 28, 2008 at 12:21:58PM -0700, Daniel Walker wrote: > > Don't you need to return the -EINTR so the caller knows the nature of > > the failure? You might also need to re-factor the caller of this > > function so it properly reports the failure to userspace .. In this case > > your just returning NULL .. > > The signal is _fatal_. Userspace doesn't get to check the return value. > It's dead. Kernel space does still observe the failure, right? Otherwise you wouldn't return anything from mutex_lock_killable() .. If that's the case just returning NULL , is like just injecting a failure.. I'm not against adding mutex_lock_killable() , I'm just wondering if your changes are complete in this case.. Daniel