From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gdb-patches-return-16734-listarch-gdb-patches=sourceware.cygnus.com@sources.redhat.com>
Received: (qmail 16611 invoked by alias); 9 Jul 2002 03:49:10 -0000
Mailing-List: contact gdb-patches-help@sources.redhat.com; run by ezmlm
Precedence: bulk
List-Subscribe: <mailto:gdb-patches-subscribe@sources.redhat.com>
List-Archive: <http://sources.redhat.com/ml/gdb-patches/>
List-Post: <mailto:gdb-patches@sources.redhat.com>
List-Help: <mailto:gdb-patches-help@sources.redhat.com>, <http://sources.redhat.com/ml/#faqs>
Sender: gdb-patches-owner@sources.redhat.com
Received: (qmail 16582 invoked from network); 9 Jul 2002 03:49:07 -0000
Received: from unknown (HELO localhost.redhat.com) (24.112.240.27)
  by sources.redhat.com with SMTP; 9 Jul 2002 03:49:07 -0000
Received: from ges.redhat.com (localhost [127.0.0.1])
	by localhost.redhat.com (Postfix) with ESMTP
	id A73313CC5; Mon,  8 Jul 2002 23:49:00 -0400 (EDT)
Message-ID: <3D2A5D2C.5040506@ges.redhat.com>
Date: Mon, 08 Jul 2002 21:12:00 -0000
From: Andrew Cagney <ac131313@ges.redhat.com>
User-Agent: Mozilla/5.0 (X11; U; NetBSD macppc; en-US; rv:1.0.0) Gecko/20020613
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: jimb@redhat.com
Cc: Daniel Berlin <dberlin@dberlin.org>,
	gdb-patches@sources.redhat.com
Subject: Re: [RFA]: dwarf2expr.[ch]
References: <Pine.LNX.4.44.0207081941190.31676-100000@dberlin.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
X-SW-Source: 2002-07/txt/msg00137.txt.bz2

Jim,

Is it possible to simplify some of this.  For instance:

> +   void (*error) (const char *fmt, ...);

In GDB, the function error() has well understood semantics - no return, 
long jump, fprintf parameter checking, .....  Why not use error() 
directly?  Failing that, the function probably wants to be marked up 
with NORETURN and a printf attribute.

> + 
> +   /* If it has a location expression for fbreg, record it.  
> +      Since not all symbols use location expressions exclusively yet,
> +      we still need to decode it above. */
> +   if (attr)
> +     {
> +       baton = obstack_alloc (&objfile->symbol_obstack, sizeof (struct locexpr_baton));
> +       baton->sym = new->name;
> +       baton->obj = objfile;
> +       memcpy (&baton->locexpr, DW_BLOCK (attr), sizeof (struct dwarf_block));
> +       SYMBOL_LOCATION_FUNCS (new->name) = &locexpr_funcs;
> +       SYMBOL_LOCATION_BATON (new->name) = baton;
> +     }
> + 

Is there also a way of implementing these objects such that they check, 
at compile time, a match between initialized members and those that 
require assignment?  Perhaphs a memset(0) will be easiest.

I suspect, for instance, you'll need to add a thread_local_storgage() 
method pretty soon :-)

Andrew