From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 11120 invoked by alias); 6 Aug 2004 22:58:12 -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 11113 invoked from network); 6 Aug 2004 22:58:11 -0000 Received: from unknown (HELO mx1.redhat.com) (66.187.233.31) by sourceware.org with SMTP; 6 Aug 2004 22:58:11 -0000 Received: from int-mx1.corp.redhat.com (int-mx1.corp.redhat.com [172.16.52.254]) by mx1.redhat.com (8.12.10/8.12.10) with ESMTP id i76MwBe1013124 for ; Fri, 6 Aug 2004 18:58:11 -0400 Received: from zenia.home.redhat.com (porkchop.devel.redhat.com [172.16.58.2]) by int-mx1.corp.redhat.com (8.11.6/8.11.6) with ESMTP id i76Mw9a26494; Fri, 6 Aug 2004 18:58:10 -0400 To: Daniel Jacobowitz Cc: gdb-patches@sources.redhat.com, ezannoni@redhat.com Subject: Re: [rfa/dwarf/doc] Inter-compilation-unit reference support for partial DIEs References: <20040420170855.GA31548@nevyn.them.org> <20040617034201.GD23443@nevyn.them.org> <20040715184431.GA25807@nevyn.them.org> <20040804230536.GA30848@nevyn.them.org> <20040805180341.GB9011@nevyn.them.org> From: Jim Blandy Date: Fri, 06 Aug 2004 22:58:00 -0000 In-Reply-To: <20040805180341.GB9011@nevyn.them.org> Message-ID: User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.3 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-SW-Source: 2004-08/txt/msg00188.txt.bz2 When looking up references to dies in other compilation units, the patch asserts a few times that the offset is valid. I have in mind the gdb_assert in find_partial_die, and the second gdb_assert in dwarf2_find_containing_comp_unit. But if the offset is wrong, that doesn't indicate an internal inconsistency in GDB; it indicates an error in the debugging info. So these should be complaints or perhaps errors, but not internal errors. (The other gdb_asserts in dwarf2_build_psymtabs_hard and dwarf2_find_containing_comp_unit are cross-checking offsets in the CU tree against other offsets created by reading the same data, so those look right.) I have to admit, there isn't much support for robust error handling in the Dwarf reader. I don't want to hold up this patch pending the creation of that support, but we should at least do better than an internal error.