From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 13963 invoked by alias); 15 Aug 2008 15:02:53 -0000 Received: (qmail 13951 invoked by uid 22791); 15 Aug 2008 15:02:51 -0000 X-Spam-Check-By: sourceware.org Received: from mail.codesourcery.com (HELO mail.codesourcery.com) (65.74.133.4) by sourceware.org (qpsmtpd/0.31) with ESMTP; Fri, 15 Aug 2008 15:02:16 +0000 Received: (qmail 5692 invoked from network); 15 Aug 2008 15:02:15 -0000 Received: from unknown (HELO localhost) (vladimir@127.0.0.2) by mail.codesourcery.com with ESMTPA; 15 Aug 2008 15:02:15 -0000 From: Vladimir Prus To: Pedro Alves Subject: Re: [RFA] Remove unnecessary target defaults. Date: Fri, 15 Aug 2008 15:02:00 -0000 User-Agent: KMail/1.9.9 Cc: gdb-patches@sourceware.org References: <200808151645.31042.vladimir@codesourcery.com> <200808151840.53192.vladimir@codesourcery.com> <200808151552.19945.pedro@codesourcery.com> In-Reply-To: <200808151552.19945.pedro@codesourcery.com> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200808151902.07713.vladimir@codesourcery.com> 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: 2008-08/txt/msg00411.txt.bz2 On Friday 15 August 2008 18:52:19 Pedro Alves wrote: > On Friday 15 August 2008 15:40:53, Vladimir Prus wrote: > > > > It sounds sound to me. This would be like a generalised version of > > > inf-child.c, from which all targets that can have execution > > > inherit. > > > > Well, actually, I meant implementing target_can_async_p in such a way that > > it: > > > > - iterates over all targets > > - breaks after it examines a target at process stratum > > > > Should work too... > > > Of course, we can also have a common target, but it would require manually > > adjusting all target on process stratum. > > ... yeah. I keep meaning to try adding a base target from which all > targets inherit, so we could perhaps get rid of this INHERIT + > de_fault business, and I somehow thought you could read minds. Well, > you do, but you set it to block dumb ideas. :-) In fact, I cannot shake the feeling that current target stack is several designs lumped together, and it would be beneficial to switch to a more coherent design, where, in particular: - there's base target every other target derives from and most methods of the base target forward to the target beneath - dummy target catches all methods - all INHERIT and de_fault code is removed - all target_xxx methods (iterating over target stack) are removed, or make just call current_target->xxx But this is fairly large thing to do. - Volodya