From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 14987 invoked by alias); 12 Dec 2002 09:42:36 -0000 Mailing-List: contact gdb-patches-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sources.redhat.com Received: (qmail 14967 invoked from network); 12 Dec 2002 09:42:33 -0000 Received: from unknown (HELO takamaka.act-europe.fr) (212.157.227.202) by sources.redhat.com with SMTP; 12 Dec 2002 09:42:33 -0000 Received: by takamaka.act-europe.fr (Postfix, from userid 507) id ABD1CD2D29; Thu, 12 Dec 2002 10:42:32 +0100 (CET) Date: Thu, 12 Dec 2002 02:04:00 -0000 From: Joel Brobecker To: Michael Snyder Cc: Klee Dienes , gdb-patches@sources.redhat.com Subject: Re: [RFC/RFA] GDB crash when using command lines due to memory corruption Message-ID: <20021212094232.GJ25575@gnat.com> References: <8096FEF2-0D32-11D7-9BDD-00039396EEB8@apple.com> <3DF7C8AB.6822A575@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <3DF7C8AB.6822A575@redhat.com> User-Agent: Mutt/1.4i X-SW-Source: 2002-12/txt/msg00399.txt.bz2 > > A safer change for 5.3 might be the patch I submitted on October 30th. > > > > http://sources.redhat.com/ml/gdb-patches/2002-10/msg00586.html > > > > Joel, would Klee's change satisfy your requirement? Absolutely. Klee's change is actually the first solution I came up with, but for some reason I remembered that there was a reluctance to copy this command lines structure more often than necessary. To be fair, I really like Klee's change over mine, as it is much simpler and much more maintainable. Can Klee's change be approved, and can Klee do the checkin? I can also do the checkin on Klee's behalf if he's busy. He's addition testcase is also a useful addition to our testsuite. One small detail, is about memory management. Did anybody make sure that there is no memory leak introduced in this change? Speaking of memory leaks, there is the small change is bpstat_stop_status which I will resubmit separately. Thanks, -- Joel