From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 1890 invoked by alias); 23 Oct 2008 19:16:19 -0000 Received: (qmail 1866 invoked by uid 22791); 23 Oct 2008 19:16:15 -0000 X-Spam-Check-By: sourceware.org Received: from mail.codesourcery.com (HELO mail.codesourcery.com) (65.74.133.4) by sourceware.org (qpsmtpd/0.31) with ESMTP; Thu, 23 Oct 2008 19:15:39 +0000 Received: (qmail 21384 invoked from network); 23 Oct 2008 19:15:37 -0000 Received: from unknown (HELO orlando.local) (pedro@127.0.0.2) by mail.codesourcery.com with ESMTPA; 23 Oct 2008 19:15:37 -0000 From: Pedro Alves To: gdb-patches@sourceware.org Subject: Re: [patch] [python] Fix for double-free. Date: Thu, 23 Oct 2008 19:16:00 -0000 User-Agent: KMail/1.9.9 Cc: Paul Pluzhnikov References: <20081023172158.D13083A6AFD@localhost> In-Reply-To: <20081023172158.D13083A6AFD@localhost> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200810232015.36765.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: 2008-10/txt/msg00582.txt.bz2 I'll take this one. On Thursday 23 October 2008 18:21:58, Paul Pluzhnikov wrote: > Greetings, > > I made the change below on archer-tromey-python. > Thanks for sending it here. (((( When I asked before please to send breakage fixes here, I was refering to breakages that affected everybody's non-python GDB usage, like the build breakage of the other thread. It's certainly nice to have the code maintained though, in case people are already trying simple things out. Thanks again. )))) > Unfortunately, I can't seem to be able to construct a test case > which exposes this bug. > Weird. Maybe the fact that you can't trigger it in FSF has something to do with owned_by_gdb not being anywhere? That's strange, because one would expect to see a double-free then anyway, I guess. Still, with that in mind, the patch looks correct to me, though. > OK to commit without a test case? IMO, yes. I'm sure you'll have testcases that trigger this when more of the python stuff is in. > Thanks, > -- > Paul Pluzhnikov > > 2008-10-23 Paul Pluzhnikov > > * python/python-value (valpy_getitem): Fix heap corruption. > -- Pedro Alves