From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 4898 invoked by alias); 9 Apr 2002 04:39:50 -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 4809 invoked from network); 9 Apr 2002 04:39:49 -0000 Received: from unknown (HELO nevyn.them.org) (128.2.145.6) by sources.redhat.com with SMTP; 9 Apr 2002 04:39:49 -0000 Received: from drow by nevyn.them.org with local (Exim 3.35 #1 (Debian)) id 16unQA-0004mZ-00 for ; Tue, 09 Apr 2002 00:39:54 -0400 Date: Mon, 08 Apr 2002 21:39:00 -0000 From: Daniel Jacobowitz To: gdb-patches@sources.redhat.com Subject: Re: [patch] gdb.c++/local.exp: make usable with gcc Message-ID: <20020409003954.A18360@nevyn.them.org> Mail-Followup-To: gdb-patches@sources.redhat.com References: <200204090315.g393FFD04125@duracef.shout.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200204090315.g393FFD04125@duracef.shout.net> User-Agent: Mutt/1.3.23i X-SW-Source: 2002-04/txt/msg00347.txt.bz2 On Mon, Apr 08, 2002 at 10:15:15PM -0500, Michael Elizabeth Chastain wrote: > This is about as much fun as a trip to the dentist. And about as > necessary, too. > > gdb.c++/local.exp has four tests for classes defined in nested scopes. > In sound-byte terms, the tests "were never known to work with gcc". > Most tests in most configurations FAILed or XFAILed (look at one of the > weekly attention reports). Those that did PASS were passing because > their error messages were deemed legal output. > > There is some nice testing functionality buried in here so I polished it > up and ran it on my testbed. The big change was to access the nested > classes inside the scope where they are defined rather than outside. > The four tests reveal two genuine gdb bugs, which now result in FAILs. > > The bugs are: > > . For a nested class definition with v3 dwarf-2 (plus 2.96-rh dwarf-2), > gdb prints two copies of the member function declaration. > > . gdb treats the commands "ptype InnerLocal::NestedInnerLocal" as a > field lookup in the InnerLocal class, failing to find the nested > class named NestedInnerLocal. This happens with both v2 and v3, > both dwarf-2 and stabs+. > > I'm going to put this up for comment for 24 hours and then commit. > After commit, I will file PR's for the bugs. Thanks! I'll take a look at those to-be PRs when I have a chance. -- Daniel Jacobowitz Carnegie Mellon University MontaVista Software Debian GNU/Linux Developer