From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 17570 invoked by alias); 24 Apr 2013 23:16:07 -0000 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 Received: (qmail 17560 invoked by uid 89); 24 Apr 2013 23:16:06 -0000 X-Spam-SWARE-Status: No, score=-3.0 required=5.0 tests=AWL,BAYES_00,FREEMAIL_FROM,KHOP_THREADED,RCVD_IN_DNSWL_LOW,RCVD_IN_HOSTKARMA_YE,SPF_PASS autolearn=ham version=3.3.1 Received: from mail-ob0-f181.google.com (HELO mail-ob0-f181.google.com) (209.85.214.181) by sourceware.org (qpsmtpd/0.84/v0.84-167-ge50287c) with ESMTP; Wed, 24 Apr 2013 23:16:04 +0000 Received: by mail-ob0-f181.google.com with SMTP id ta17so2061508obb.12 for ; Wed, 24 Apr 2013 16:16:03 -0700 (PDT) X-Received: by 10.60.136.129 with SMTP id qa1mr14451636oeb.47.1366845363454; Wed, 24 Apr 2013 16:16:03 -0700 (PDT) MIME-Version: 1.0 Received: by 10.60.13.133 with HTTP; Wed, 24 Apr 2013 16:15:23 -0700 (PDT) In-Reply-To: <877gjsndpv.fsf@fleche.redhat.com> References: <87haiwngpr.fsf@fleche.redhat.com> <877gjsndpv.fsf@fleche.redhat.com> From: Hui Zhu Date: Thu, 25 Apr 2013 12:49:00 -0000 Message-ID: Subject: Re: [PATCH/7.6] Fix wrong release (maybe crash GDB) in build_target_command_list To: Tom Tromey Cc: Yao Qi , gdb-patches ml , Joel Brobecker Content-Type: text/plain; charset=ISO-8859-1 X-SW-Source: 2013-04/txt/msg00767.txt.bz2 On Wed, Apr 24, 2013 at 11:35 PM, Tom Tromey wrote: >>>>>> "Hui" == Hui Zhu writes: > > Hui> I tried but looks free conditions cannot crash GDB. :( > > I was wondering if there could be a non-crashing reproducer. > It seems a little tricky. > > I think this particular patch is ok without a test though. > Please check it in. Checked in http://www.sourceware.org/ml/gdb-cvs/2013-04/msg00230.html Can I check in it to 7.6 branch? > >>> The "continue" seems to mean that null_command_or_parse_error will not >>> be set in the "parse error" case. > > Hui> I think it is right, because even if one of breakpoint loc doesn't > Hui> have commands. GDB still need send other commands to target. > > I think that would be an argument for removing the "break" -- but not > necessarily for keeping the "continue". > > The "return" in the null_command_or_parse_error case also seems weird to > me, but I didn't think hard about it. Because current function is hard to handle two or more breakpoints in same address with different commands and conditions. So maybe I will rewrite this function. Please keep help me with that when I post patch for them. Thanks, Hui > > Tom