From owner-cvs-src@FreeBSD.ORG Tue Nov 30 18:45:13 2004 Return-Path: Delivered-To: cvs-src@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C80BF16A4D6; Tue, 30 Nov 2004 18:45:13 +0000 (GMT) Received: from ns1.xcllnt.net (209-128-86-226.BAYAREA.NET [209.128.86.226]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3E0C643D2D; Tue, 30 Nov 2004 18:45:13 +0000 (GMT) (envelope-from marcel@xcllnt.net) Received: from [192.168.4.250] (dhcp50.pn.xcllnt.net [192.168.4.250]) by ns1.xcllnt.net (8.13.1/8.13.1) with ESMTP id iAUIj5rZ092466; Tue, 30 Nov 2004 10:45:05 -0800 (PST) (envelope-from marcel@xcllnt.net) In-Reply-To: <20041130.013428.25281327.imp@bsdimp.com> References: <200411300256.iAU2urva052399@repoman.freebsd.org> <20041130073515.GA92857@ip.net.ua> <41690958-42A5-11D9-A383-000D93C47836@xcllnt.net> <20041130.013428.25281327.imp@bsdimp.com> Mime-Version: 1.0 (Apple Message framework v619) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: Content-Transfer-Encoding: 7bit From: Marcel Moolenaar Date: Tue, 30 Nov 2004 10:45:04 -0800 To: "M. Warner Losh" X-Mailer: Apple Mail (2.619) cc: cvs-src@FreeBSD.ORG cc: src-committers@FreeBSD.ORG cc: ru@FreeBSD.ORG cc: cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/gnu/usr.bin/gdb/kgdb Makefile X-BeenThere: cvs-src@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: CVS commit messages for the src tree List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 30 Nov 2004 18:45:14 -0000 On Nov 30, 2004, at 12:34 AM, M. Warner Losh wrote: > : I disagree. The fact that gdb is not a cross-tool in the buildworld > : sense of the meaning is because we don't need gdb to cross-build > world. > : This does not mean that gdb cannot be a cross-tool from a developers > : point of view. In fact, it's really handy to be able to debug a > kernel > : remotely when the debugger doesn't run on the same architecture as > : the kernel. I for one do not want to have 2 machines per architecture > : for the only reason that I want to debug a kernel remotely. So, I > made > : some changes to work towards that goal. I see no problems or harm, > : especially since I'm not misusing TARGET_ARCH. The ability to build a > : cross-debugger is simply not utilized as part of a buildworld. > > But MACHINE_ARCH is set correctly during the build of gdb when you've > set TARGET_ARCH to build the rest of the world you need to build gdb > with. Or am I missing something? I'm having problems figuring out what exactly you mean and thus what your point is. I think the following addresses your presumed point: As soon as we do the _everything part of a buildworld both MACHINE_ARCH and TARGET_ARCH represent the architecture we're doing the buildworld *for* (if TARGET_ARCH is undefined, we define it as MACHINE_ARCH). Hence, we cannot build a cross debugger as part of buildworld if we don't do it as part of the cross-tools phase. This is also not my immediate goal. Rather, I'm working towards an easy recipe involving binutils, gdb and libkvm that allows building and installing a cross debugger. By hand for now. This may or may not be given a target in src/Makefile and may or may not evolve into something that allows doing it during a buildworld. -- Marcel Moolenaar USPA: A-39004 marcel@xcllnt.net