From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754083Ab1HBPU3 (ORCPT ); Tue, 2 Aug 2011 11:20:29 -0400 Received: from mgw2.diku.dk ([130.225.96.92]:51278 "EHLO mgw2.diku.dk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753525Ab1HBPUV (ORCPT ); Tue, 2 Aug 2011 11:20:21 -0400 Date: Tue, 2 Aug 2011 17:20:14 +0200 (CEST) From: Julia Lawall To: Andrea Righi Cc: Andrew Morton , trivial@kernel.org, kernel-janitors@vger.kernel.org, Stefani Seibold , "Ira W. Snyder" , linux-kernel@vger.kernel.org Subject: Re: [PATCH 1/11] kernel/kfifo.c: trivial: use BUG_ON In-Reply-To: <20110802120329.GA9088@thinkpad> Message-ID: References: <1312281304-11847-1-git-send-email-julia@diku.dk> <1312281304-11847-2-git-send-email-julia@diku.dk> <20110802113335.GA4185@thinkpad> <20110802120329.GA9088@thinkpad> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2 Aug 2011, Andrea Righi wrote: > On Tue, Aug 02, 2011 at 01:52:49PM +0200, Julia Lawall wrote: > > On Tue, 2 Aug 2011, Andrea Righi wrote: > > > > > On Tue, Aug 02, 2011 at 12:34:54PM +0200, Julia Lawall wrote: > > > > From: Julia Lawall > > > > > > > > Use BUG_ON(x) rather than if(x) BUG(); > > > > > > > > The semantic patch that fixes this problem is as follows: > > > > (http://coccinelle.lip6.fr/) > > > > > > > > // > > > > @@ identifier x; @@ > > > > -if (x) BUG(); > > > > +BUG_ON(x); > > > > > > > > @@ identifier x; @@ > > > > -if (!x) BUG(); > > > > +BUG_ON(!x); > > > > // > > > > > > > > Signed-off-by: Julia Lawall > > > > > > > > > > The patch looks good. However, I think we should just get rid of the > > > BUG_ON() there. If nents is 0, setup_sgl_buf() would just returns 0 and > > > nothing critical will happen. > > > > I don't know the code at all. Indeed, I'm not sure to see any uses of eg > > __kfifo_dma_in_prepare_r, except in demo code. But that would imply > > returning 0 for two different reasons: asking for nothing and being out of > > space. > > > > julia > > Right. But I'd prefer to return nothing also when asking nothing rather > than crashing the kernel. OK, that seems like a good point too. I will send a new patch. julia