From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 1302 invoked by alias); 18 Feb 2004 00:32:28 -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 1293 invoked from network); 18 Feb 2004 00:32:27 -0000 Received: from unknown (HELO nevyn.them.org) (66.93.172.17) by sources.redhat.com with SMTP; 18 Feb 2004 00:32:27 -0000 Received: from drow by nevyn.them.org with local (Exim 4.30 #1 (Debian)) id 1AtFdb-0002yb-59 for ; Tue, 17 Feb 2004 19:32:27 -0500 Date: Wed, 18 Feb 2004 00:32:00 -0000 From: Daniel Jacobowitz To: gdb-patches@sources.redhat.com Subject: Re: [rfa] Add SYMBOL_SET_LINKAGE_NAME Message-ID: <20040218003227.GA11352@nevyn.them.org> Mail-Followup-To: gdb-patches@sources.redhat.com References: <20040216212406.GC17141@nevyn.them.org> <20040218002329.GA11115@nevyn.them.org> <16434.45185.855584.104606@localhost.redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <16434.45185.855584.104606@localhost.redhat.com> User-Agent: Mutt/1.5.1i X-SW-Source: 2004-02/txt/msg00492.txt.bz2 On Tue, Feb 17, 2004 at 07:23:29PM -0500, Elena Zannoni wrote: > Daniel Jacobowitz writes: > > On Tue, Feb 17, 2004 at 09:23:47AM -0800, David Carlton wrote: > > > On Mon, 16 Feb 2004 16:24:06 -0500, Daniel Jacobowitz said: > > > > > > > This patch adds a macro, SYMBOL_SET_LINKAGE_NAME, which is used to > > > > set a symbol's name when the name should not be demangled. Used for > > > > things like typedefs whose name comes from debug info. > > > > > > The idea is okay, but I don't like the name all that much. I once had > > > a goal, which I've admittedly been lax about pursuing recently, that > > > we would have a very clear distinction between linkage names (which > > > really did mean names used by the linker) and natural names (i.e. the > > > names in the source code), to the extent that, if we were to represent > > > these by different types, then our code would almost compile. > > > > > > When we're talking about types, however, linkage names don't make much > > > sense, only natural names. So, while it's true that your macro does > > > set the field that, in the case of a symbol with both linkage and > > > natural names, corresponds to the linkage name, that's really an > > > implementation detail that should be shielded behind this macro. > > > > > > Having said that, I don't have any great suggestions for a better > > > name. SYMBOL_SET_NATURAL_NAME? SYMBOL_SET_NATURAL_ONLY_NAME? Hmm. > > > > I don't want to call it SYMBOL_SET_NATURAL_NAME. It's not necessarily > > the natural name. Other than that, I don't know. > > > > I'm just going to sit on this. The HP patches need to be revised > > anyway, and people want me to draft a complete interface before doing > > any cleanups. > > come on, that's not what I asked. I know. You haven't asked anything about the patch, which you haven't reviewed yet, which is fine. It's only a day old. The discussion about interfaces is on a tangent thread about my future intentions. But there's no point in just continuing the cleanups now that I've stirred up this much annoyed discussion asking for interface descriptions. So I'll write the interface description and come back later. -- Daniel Jacobowitz MontaVista Software Debian GNU/Linux Developer