From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 9590 invoked by alias); 19 Feb 2010 20:28:38 -0000 Received: (qmail 9582 invoked by uid 22791); 19 Feb 2010 20:28:38 -0000 X-SWARE-Spam-Status: No, hits=-2.5 required=5.0 tests=AWL,BAYES_00,SPF_PASS X-Spam-Check-By: sourceware.org Received: from mail.codesourcery.com (HELO mail.codesourcery.com) (38.113.113.100) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Fri, 19 Feb 2010 20:28:34 +0000 Received: (qmail 17741 invoked from network); 19 Feb 2010 20:28:32 -0000 Received: from unknown (HELO orlando.localnet) (pedro@127.0.0.2) by mail.codesourcery.com with ESMTPA; 19 Feb 2010 20:28:32 -0000 From: Pedro Alves To: Tom Tromey Subject: Re: RFA: valgrind and the test suite, take 2 Date: Fri, 19 Feb 2010 20:28:00 -0000 User-Agent: KMail/1.12.2 (Linux/2.6.31-19-generic; KDE/4.3.2; x86_64; ; ) Cc: gdb-patches@sourceware.org References: <201002191745.18600.pedro@codesourcery.com> In-Reply-To: MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <201002192028.30653.pedro@codesourcery.com> X-IsSubscribed: yes Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org X-SW-Source: 2010-02/txt/msg00510.txt.bz2 On Friday 19 February 2010 20:02:40, Tom Tromey wrote: > >>>>> "Pedro" == Pedro Alves writes: > > Pedro> On Friday 19 February 2010 17:23:52, Tom Tromey wrote: > >> This patch adds support for valgrind to the test suite. Unlike my last > >> patch along these lines, this one adds value above just setting GDB. > > Pedro> Couldn't this be done with a board file? > > Maybe, but then it won't be in the tree, and it won't be documented. Those are both fixable. I don't know if you seen my other more recent email. > If that is an objection, could you explain why you'd prefer it to be > done that way? It's not an objection that I'll spend much more energy defending, but FYI, I prefer not to hack specific solutions when a more general solution already exists. Another advantange is that you can tweak it to your hearts content without having to touch generic testsuite code. If it allows things that a board file doesn't, then it's another story. But maybe that could be considered a problem of missing hooks instead. > The point of doing it this way is to encourage more frequent runs using > valgrind. What's wrong with encouraging testing with a board file? We do the same for local gdbserver testing. We could come up with N other board files that did similar things. > I find bugs with valgrind so often that I wish this mode of > running were the default. If it's not going to be made the default, then that's irrelevant. -- Pedro Alves