From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756808Ab1LWM04 (ORCPT ); Fri, 23 Dec 2011 07:26:56 -0500 Received: from mail1-relais-roc.national.inria.fr ([192.134.164.82]:35382 "EHLO mail1-relais-roc.national.inria.fr" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753915Ab1LWM0x (ORCPT ); Fri, 23 Dec 2011 07:26:53 -0500 X-IronPort-AV: E=Sophos;i="4.71,398,1320620400"; d="scan'208";a="136607699" Date: Fri, 23 Dec 2011 13:26:49 +0100 (CET) From: Julia Lawall X-X-Sender: jll@localhost6.localdomain6 To: Lars-Peter Clausen cc: Gilles Muller , Nicolas Palix , linux-kernel@vger.kernel.org, cocci@diku.dk, mmarek@suse.cz Subject: Re: [Cocci] [PATCH 2/2] coccinelle: Add patch for replacing open-coded IS_ERR_OR_NULL In-Reply-To: <4EF471ED.1030902@metafoo.de> Message-ID: References: <1324301063-16117-1-git-send-email-lars@metafoo.de> <1324301063-16117-2-git-send-email-lars@metafoo.de> <4EF1C7D3.7080301@metafoo.de> <4EF471ED.1030902@metafoo.de> User-Agent: Alpine 2.02 (DEB 1266 2009-07-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 23 Dec 2011, Lars-Peter Clausen wrote: > On 12/21/2011 01:10 PM, Julia Lawall wrote: >> On Wed, 21 Dec 2011, Lars-Peter Clausen wrote: >> >>> On 12/20/2011 11:24 PM, Julia Lawall wrote: >>>> For this one, I made quite a number of changes, some of which required >>>> fixing a bug in Coccinelle... I will try to release the bug fix >>>> shortly. In general, though, I have added ... between the two tests. I >>>> also got rid of the focus on ifs in the org and report rules. >>>> >>> >>> Ok, thanks. Was the removing of the ifs in related to the Coccinelle >>> bug? I >>> always get an error if I start a match with a conditional. >> >> Not at all. The bug was that a patterm like: >> >> @depends on patch@ >> expression x; >> @@ >> - IS_ERR(x) >> + IS_ERR_OR_NULL(x) >> || ... >> - || x == NULL >> >> thought that it should be applied in two ways, giving an error about a >> double transformation of the IS_ERR call. >> >> Could you reproduce the example that caused problems? > > Hm, I can reproduce it anymore, even with my original patch, must have been > caused by some other error I had in the semantic patch at that time. OK, should do you want to resubmit the semantic patch that I sent you? The version of Coccinelle you have does not perfectly support it, but it should just cause failures on a few files. The problem is only with the patch case, when there is the possibility to mak two calls to IS_ERR_OR_NULL in a single test expression, which doesn't happen very often. julia