From mboxrd@z Thu Jan 1 00:00:00 1970 From: rubisher Subject: Re: cgcc and REAL_CC help Date: Mon, 21 Dec 2009 16:42:01 +0000 Message-ID: <4B2FA559.4040102@scarlet.be> References: <4B29F6C2.7040800@scarlet.be> <36ca99e90912170142p26b9b0aal1a08509d65c53047@mail.gmail.com> <4B2BB887.6060508@scarlet.be> <70318cbf0912201445m37b31c71r6209a26b40892461@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from hel.is.scarlet.be ([193.74.71.26]:39396 "EHLO hel.is.scarlet.be" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753074AbZLUQmI (ORCPT ); Mon, 21 Dec 2009 11:42:08 -0500 In-Reply-To: <70318cbf0912201445m37b31c71r6209a26b40892461@mail.gmail.com> Sender: linux-sparse-owner@vger.kernel.org List-Id: linux-sparse@vger.kernel.org To: Chris Li Cc: Bert Wesarg , linux-sparse@vger.kernel.org Hello Chris, Chris Li wrote: > On Fri, Dec 18, 2009 at 9:14 AM, rubisher wrote: >> That said, my next issue is that "-E" which is disable by cgcc: >> [snip] >> # If someone adds "-E", don't pre-process twice. >> $do_compile = 0 if $_ eq '-E'; >> [snip] >> >> If ever some of you knows why, I would be interested in? > > My guess is that, both gcc and sparse pre-process print to stdout. > So allow them both pre-process will mess up the result. > That was also my guess but when I comment out this line and run the same script one time with 'REAL_CC=gcc-4.4 cgcc' and the second time with 'gcc-4.4' the results are the same, thought? > The current cgcc chose to let sparse do the pre-process and don't > run the real cc. You can make the gcc does the real pro-process and > sparse just check though. > Yes I think that I would have to do more effort to sparse this code. Tx, J. > Chris >