From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 7499 invoked by alias); 16 Feb 2010 22:57:18 -0000 Received: (qmail 7490 invoked by uid 22791); 16 Feb 2010 22:57:18 -0000 X-SWARE-Spam-Status: No, hits=-2.5 required=5.0 tests=AWL,BAYES_00,SPF_PASS X-Spam-Check-By: sourceware.org Received: from mail.codesourcery.com (HELO mail.codesourcery.com) (38.113.113.100) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Tue, 16 Feb 2010 22:57:15 +0000 Received: (qmail 32109 invoked from network); 16 Feb 2010 22:57:13 -0000 Received: from unknown (HELO caradoc.them.org) (dan@127.0.0.2) by mail.codesourcery.com with ESMTPA; 16 Feb 2010 22:57:13 -0000 Date: Tue, 16 Feb 2010 22:57:00 -0000 From: Daniel Jacobowitz To: Tom Tromey Cc: Chris Moller , gdb-patches@sourceware.org Subject: Re: beta patch for PR 9065 Message-ID: <20100216225702.GA9914@caradoc.them.org> Mail-Followup-To: Tom Tromey , Chris Moller , gdb-patches@sourceware.org References: <4B7AE477.9060606@redhat.com> <20100216215313.GA4858@caradoc.them.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.20 (2009-06-14) 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: 2010-02/txt/msg00408.txt.bz2 On Tue, Feb 16, 2010 at 03:27:47PM -0700, Tom Tromey wrote: > >>>>> "Daniel" == Daniel Jacobowitz writes: > > Daniel> It's been a while since I looked at this, but the typeinfo for a class > Daniel> with RTTI should be available in the symbol table (_ZTI > Daniel> symbols). > > Oh, right. Maybe I was confusing this w/ a vtable slot, whoops. It's both, but there are also _ZTI symbols for things without vtables; e.g. standard types ("int") and classes without a vtable but where typeid was used. The vtable's probably preferable where available. > Daniel> We could synthesize a type_info object; the Itanium C++ ABI defines > Daniel> the layout (section 2.9.5). But we might also have to synthesize its > Daniel> name() method, as it is unlikely to be emitted out of line. > > I think it would be acceptable, though not ideal, to give an error in > this situation. Yes - it's a more general problem - IMO it'd be nice to be able to have scripts implement not-outlined methods. std::string.length() is probably similar... -- Daniel Jacobowitz CodeSourcery