Date: Sun, 13 Jul 2014 20:00:02 -0400 From: Julio Merino <jmmv@freebsd.org> To: =?UTF-8?Q?Dag=2DErling_Sm=C3=B8rgrav?= <des@freebsd.org> Cc: src-committers@freebsd.org, svn-src-user@freebsd.org Subject: Re: svn commit: r266617 - user/des/tinderbox Message-ID: <CAFY7cWDaMMn=mU2C2s8bDCBaH_PWE=khZZ93cGatUO06Nk1=OA@mail.gmail.com> In-Reply-To: <201405241334.s4ODYf4M000124@svn.freebsd.org> References: <201405241334.s4ODYf4M000124@svn.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, May 24, 2014 at 9:34 AM, Dag-Erling Sm=C3=B8rgrav <des@freebsd.org>= wrote: > Author: des > Date: Sat May 24 13:34:41 2014 > New Revision: 266617 > URL: http://svnweb.freebsd.org/changeset/base/266617 > > Log: > Set MAKESYSPATH before building to ensure we use the correct set of .mk > files right from the start. Is this supposed to work though? The host make may not match current's make, and thus current's share/mk/ files may be using features not yet supported by the host make. I'd expect the top-level Makefile to deal with any such issues by ensuring everything leading to the build of a current bmake works. This seem to be the case with a "make buildworld" *without* overriding MAKESYSPATH. This uncovered an oddity in the build (r268598), but I'm not sure tinderbox is doing the right thing.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAFY7cWDaMMn=mU2C2s8bDCBaH_PWE=khZZ93cGatUO06Nk1=OA>