From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 21606 invoked by alias); 18 Jun 2002 15:42:22 -0000 Mailing-List: contact gdb-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-owner@sources.redhat.com Received: (qmail 21562 invoked from network); 18 Jun 2002 15:42:04 -0000 Received: from unknown (HELO localhost.redhat.com) (216.138.202.10) by sources.redhat.com with SMTP; 18 Jun 2002 15:42:04 -0000 Received: from cygnus.com (localhost [127.0.0.1]) by localhost.redhat.com (Postfix) with ESMTP id 363E43D0C; Tue, 18 Jun 2002 11:15:42 -0400 (EDT) Message-ID: <3D0F4E9E.6030407@cygnus.com> Date: Tue, 18 Jun 2002 08:42:00 -0000 From: Andrew Cagney 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: "Frank Ch. Eigler" Cc: cagney@redhat.com, gdb@sources.redhat.com Subject: Re: Obsolete GDB's m32r support References: <200206181406.g5IE6kb05361@tooth.toronto.redhat.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-SW-Source: 2002-06/txt/msg00136.txt.bz2 > Hi - > > >> > Can you summarize the technical reasons for this dichotomy of >> > outcomes? In what way does the presence of old targets pose a >> > technical challenge to more modern multiarch'd ones? In other words, >> > what technical reasons exist to not allow old targets to stick around >> > as they are? > >> >> The rationale for this move has been discussed on a number of occasions >> in the past, can I recommend a quick search through the e-mail archives >> and a review of that previous debate. > > > A quick search resulted only in (valid) descriptions of multi-arch > superiority, and dicta that old targets shall convert or die. It > didn't answer my question above. Perhaps you could spare time for > a more specific pointer? Interesting. I didn't say the search would be easy. Of the top of my head: Each GDB target incures overhead. Old, non-multi-arch targets, incure greater overhead then newer multi-arch ones (hint, how long is a multi-arch rebuild). By either obsoleting or multi-arching a target, the overall maintenance overhead of GDB is reduced. Non multi-arch targets rely on deprecated mechanisms and those mechanims are being be deleted from GDB. Rather than putting targets relying on such mechanisms on life support (applying untested bandaid patches) GDB is eliminating those targets. There is no benefit in retaining a broken target (the old code is always available in CVS). The option of blindly multi-arching non- multi-arch targets was considered (i.e. convert the code but not test the result). It was concluded that, there was little benefit. You would end up with a broken multi-arch target that no one is interested in maintaining. The targets being obsoleted show no evidence of life (your use of the word old was unfortunate). The vax has been multi-arched, the z8k is going to be multi-arched, the !*&^$#*&@!$ PA RISC will need to be multi-arched, I suspect that even the 88k will get multi-arched! enjoy, Andrew