From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gdb-patches-return-64223-listarch-gdb-patches=sources.redhat.com@sourceware.org>
Received: (qmail 11628 invoked by alias); 5 Jun 2009 18:49:03 -0000
Received: (qmail 11620 invoked by uid 22791); 5 Jun 2009 18:49:02 -0000
X-SWARE-Spam-Status: No, hits=-2.4 required=5.0 	tests=AWL,BAYES_00,SPF_PASS
X-Spam-Check-By: sourceware.org
Received: from mail.codesourcery.com (HELO mail.codesourcery.com) (65.74.133.4)     by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Fri, 05 Jun 2009 18:48:57 +0000
Received: (qmail 25980 invoked from network); 5 Jun 2009 18:48:55 -0000
Received: from unknown (HELO orlando.local) (pedro@127.0.0.2)   by mail.codesourcery.com with ESMTPA; 5 Jun 2009 18:48:55 -0000
From: Pedro Alves <pedro@codesourcery.com>
To: tromey@redhat.com
Subject: Re: [RFA] Fix too many "no debugging symbols found" warnings.
Date: Fri, 05 Jun 2009 18:49:00 -0000
User-Agent: KMail/1.9.10
Cc: gdb-patches@sourceware.org,  Doug Evans <dje@google.com>,  Eli Zaretskii <eliz@gnu.org>
References: <20090523232414.BED2B846C2@localhost> <200906042221.22713.pedro@codesourcery.com> <m3y6s6tuja.fsf@fleche.redhat.com>
In-Reply-To: <m3y6s6tuja.fsf@fleche.redhat.com>
MIME-Version: 1.0
Content-Type: text/plain;   charset="us-ascii"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
Message-Id: <200906051949.45647.pedro@codesourcery.com>
X-IsSubscribed: yes
Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm
Precedence: bulk
List-Id: <gdb-patches.sourceware.org>
List-Subscribe: <mailto:gdb-patches-subscribe@sourceware.org>
List-Archive: <http://sourceware.org/ml/gdb-patches/>
List-Post: <mailto:gdb-patches@sourceware.org>
List-Help: <mailto:gdb-patches-help@sourceware.org>, <http://sourceware.org/ml/#faqs>
Sender: gdb-patches-owner@sourceware.org
X-SW-Source: 2009-06/txt/msg00110.txt.bz2

On Friday 05 June 2009 19:17:13, Tom Tromey wrote:

> IMO, the ideal would be to share objfiles across inferiors (perhaps a
> big task in itself) and then issue this warning at most once per
> objfile.  That way the user won't see 1000 warnings for /lib/libc.so.

100% agreed, both in the ideal bit, and on the task size bit.  :-)  For
true/always/full sharing, I'm thinking that we'd have to rework how gdb
addresses symbol values (we relocate everything early; we'd have to apply
relocation offsets on demand), and, we do some sorting of symbols, since
sections may end up loaded at different relative offsets within the
same objfile for different inferiors, in some targets.  E.g.,

objfiles.c:objfile_relocate:
...

 /* Relocating different sections by different amounts may cause the symbols
     to be out of order.  */
  msymbols_sort (objfile);

-- 
Pedro Alves