From owner-cvs-all Sat Oct 10 03:51:16 1998 Return-Path: Received: (from daemon@localhost) by hub.freebsd.org (8.8.8/8.8.8) id DAA21737 for cvs-all-outgoing; Sat, 10 Oct 1998 03:51:16 -0700 (PDT) (envelope-from owner-cvs-all) Received: from ceia.nordier.com (slip139-92-122-75.joh.za.ibm.net [139.92.122.75]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id DAA21710; Sat, 10 Oct 1998 03:51:05 -0700 (PDT) (envelope-from rnordier@nordier.com) Received: (from rnordier@localhost) by ceia.nordier.com (8.8.7/8.6.12) id MAA11838; Sat, 10 Oct 1998 12:47:44 +0200 (SAT) From: Robert Nordier Message-Id: <199810101047.MAA11838@ceia.nordier.com> Subject: Re: cvs commit: src/sys/boot/i386 Makefile In-Reply-To: <15674.908014483@time.cdrom.com> from "Jordan K. Hubbard" at "Oct 10, 98 03:14:43 am" To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Sat, 10 Oct 1998 12:47:41 +0200 (SAT) Cc: rnordier@nordier.com, peter@netplex.com.au, skynyrd@opus.cts.cwu.edu, rnordier@FreeBSD.ORG, committers@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-cvs-all@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Jordan K. Hubbard wrote: > > > Hmmm. Is that really the right makefile for it? > > > > Well, it works. But (assuming your question was rhetorical), what do > > you prefer? > > Actually it wasn't - I was confused about the behavior of .MAKEFLAGS > there. So, just to make sure we're all on the same page here, you'd > propose: > > RCS file: /home/ncvs/src/sys/boot/i386/Makefile,v > retrieving revision 1.4 > diff -u -r1.4 Makefile > --- Makefile 1998/10/09 23:30:16 1.4 > +++ Makefile 1998/10/10 10:10:53 > @@ -1,7 +1,7 @@ > SUBDIR= boot0 > > -.if ${OBJFORMAT} == elf > +.MAKEFLAGS: BTXLDFMT=${OBJFORMAT} OBJFORMAT=elf > + > SUBDIR+= btx libi386 loader > -.endif > > .include > > ? That was the suggestion. I think the essential problem is that this isn't "legacy boot" code. When a make world forces OBJFORMAT=aout, I assume it is, in effect, saying "Give 'em an a.out" binary, not "Use the a.out toolchain", because only the former makes sense. It therefore seems sensible to convert OBJFORMAT to BTXLDFMT (since OBJFORMAT then means *something*). We also need to force OBJFORMAT back to elf, thereafter. Though I suppose just ignoring it in each separate /sys/boot/i386/* Makefile is also a possibility. Anyway, I mainly want the problem to go away: I don't know enough about the FreeBSD build environment to argue the technical merits of any particular solution. :) -- Robert Nordier