From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757313Ab3GEMX4 (ORCPT ); Fri, 5 Jul 2013 08:23:56 -0400 Received: from zeniv.linux.org.uk ([195.92.253.2]:49873 "EHLO ZenIV.linux.org.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751162Ab3GEMXz (ORCPT ); Fri, 5 Jul 2013 08:23:55 -0400 Date: Fri, 5 Jul 2013 13:23:52 +0100 From: Al Viro To: Ramkumar Ramachandra Cc: LKML , Richard Weinberger , Jeff Dike Subject: Re: [PATCH 1/3] arch/um/Makefile: remove dead code Message-ID: <20130705122352.GR4165@ZenIV.linux.org.uk> References: <1372971797-23518-1-git-send-email-artagnon@gmail.com> <1372971797-23518-2-git-send-email-artagnon@gmail.com> <20130704235313.GP4165@ZenIV.linux.org.uk> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Jul 05, 2013 at 04:12:30PM +0530, Ramkumar Ramachandra wrote: > Al Viro wrote: > > [...] > > IOW, any make variable (== macro in POSIX quote above) can be explicitly > > set in make(1) command line in form name=value and in that case nothing in > > makefile(s) can change its value. It's true for GNU make, it's true for > > BSD make and it's been true for historical Unix make since its introduction > > in v7. > > Thanks Al, the education is much appreciated. I always thought = > would override it, and ?= was for default value. ?= is a different story - it refuses to override _any_ earlier declarations, including ones in makefile itself. GNU make has added it fairly late, BTW - in 3.77 (1998); BSD make has grown it at some point during the rewrite Adam de Boor had been doing in late 80s. NFI whether he'd invented it or picked it elsewhere - there had been a lot of cross-pollination between the variants of make(1) all along...