From owner-freebsd-hackers@freebsd.org Sat Jun 6 14:11:36 2020 Return-Path: Delivered-To: freebsd-hackers@mailman.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.nyi.freebsd.org (Postfix) with ESMTP id C41BB334CD6 for ; Sat, 6 Jun 2020 14:11:36 +0000 (UTC) (envelope-from freebsd-rwg@gndrsh.dnsmgr.net) Received: from gndrsh.dnsmgr.net (br1.CN84in.dnsmgr.net [69.59.192.140]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 49fLzX43P4z410P; Sat, 6 Jun 2020 14:11:36 +0000 (UTC) (envelope-from freebsd-rwg@gndrsh.dnsmgr.net) Received: from gndrsh.dnsmgr.net (localhost [127.0.0.1]) by gndrsh.dnsmgr.net (8.13.3/8.13.3) with ESMTP id 056EBYPB036236; Sat, 6 Jun 2020 07:11:34 -0700 (PDT) (envelope-from freebsd-rwg@gndrsh.dnsmgr.net) Received: (from freebsd-rwg@localhost) by gndrsh.dnsmgr.net (8.13.3/8.13.3/Submit) id 056EBYje036235; Sat, 6 Jun 2020 07:11:34 -0700 (PDT) (envelope-from freebsd-rwg) From: "Rodney W. Grimes" Message-Id: <202006061411.056EBYje036235@gndrsh.dnsmgr.net> Subject: Re: Cross compile FreeBSD on amd64 for arm64 failes via compile determination error In-Reply-To: To: Mark Murray Date: Sat, 6 Jun 2020 07:11:34 -0700 (PDT) CC: Ian Lepore , freebsd-hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL121h (25)] MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII X-Rspamd-Queue-Id: 49fLzX43P4z410P X-Spamd-Bar: ---- Authentication-Results: mx1.freebsd.org; none X-Spamd-Result: default: False [-4.00 / 15.00]; REPLY(-4.00)[]; ASN(0.00)[asn:13868, ipnet:69.59.192.0/19, country:US] X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.33 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 06 Jun 2020 14:11:36 -0000 > > On 6 Jun 2020, at 00:25, Ian Lepore wrote: > > > > You cannot do that. The installworld process uses a path that points > > into the obj tree that got built (so that new binaries are used to do > > the install work), and those binaries are compiled for the arch that > > did the build. > > That is my experience. > > I used to work around a closely related problem when I built build* on > a fast machine, and did the install* on the slow ones. As I was using > automounts, I ended up with incompatible paths baked into the build, > so I worked around those with symlinks. > > I never tried this with anything other than i386/i386 or amd64/amd64, > so the above is interesting. Could a viable cross-build* be "fixed" by > symlinks, and somehow and easily forcing a cross-build of the bootstrap > tools? Thats an interesting idea, or a form of a special "populatebuildtools" target that says to populate the obj/ tools area from the binaries on the running system, as often those are adaquate for an installworld to complete. Or possibly a simple "usehosttools" to say don't even try to use the tools from obj/ just use the tools on the running system. > My RPis wish to know :-) My diskless boot and build system wants to know too. :-) > M > -- > -- Rod Grimes rgrimes@freebsd.org