From owner-freebsd-qa Mon Mar 12 11: 0:36 2001 Delivered-To: freebsd-qa@freebsd.org Received: from lerami.lerctr.org (lerami.lerctr.org [207.158.72.11]) by hub.freebsd.org (Postfix) with ESMTP id E86A037B71F; Mon, 12 Mar 2001 11:00:22 -0800 (PST) (envelope-from ler@lerctr.org) Received: from ler-freebie.iadfw.net (ler-freebie.iadfw.net [206.66.13.221]) by lerami.lerctr.org (8.11.3/8.11.3/20010112/$Revision: 1.13 $) with SMTP id f2CJ0KI25861; Mon, 12 Mar 2001 13:00:21 -0600 (CST) (envelope-from ler@lerctr.org) From: Larry Rosenman Date: Mon, 12 Mar 2001 19:00:19 GMT Message-ID: <20010312.19001900@ler-freebie.iadfw.net> Subject: Re: cputype=486 To: John Baldwin Cc: Larry Rosenman , Kris Kennaway , Mike Pritchard , stable@FreeBSD.org, qa@FreeBSD.org In-Reply-To: References: X-Mailer: Mozilla/3.0 (compatible; StarOffice/5.2; Linux) X-Priority: 3 (Normal) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Sender: owner-freebsd-qa@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG See my answer to Jim. I MEANT to compile with a HOST system that has=20 -march=3Dpentiumpro executables, and compile a new world with CPUTYPE=3D= i486=20 or=20 no CPUTYPE and expect the new code to run on the 486, correct method?=20 LER >>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<< On 3/12/01, 12:49:50 PM, John Baldwin wrote regarding = Re:=20 cputype=3D486: > On 12-Mar-01 Larry Rosenman wrote: > > Not in the log I posted. isatty is NOT that large, so what ever thi= s is, > > is after isatty in the executable. > *sigh* > isatty() is in libc. That's the only place that that code comes from.= > It doesn't live in the executable's object files or source at all, so > it has to be coming from some copy of libc. > > I wonder if something forgot to pick up the /usr/obj libraries, sinc= e the > > HOST system *IS* compiled with -march=3Dpentiumpro. > Now that could be. > > This is valid, isn't it? (to compile on one box, and install on anot= her > > with DIFFERENT /etc/make.conf settings)? > Well, as Jim pointed out, not really. This is why releases are built > from scratch inside of a clean "white-room" chroot. Regardless, the > libc in /usr/obj should be used when compiling static binaries during > world. (Dynamic binaries should still need to make sure they use > /usr/lib/libc.so for their rpath, though if we don't use rpath, I gues= s > they could be linked against the libc.so in usr/obj as well, but a > bintools/linker person (jdp/obrien) should be asked about that > to be sure). > -- > John Baldwin -- http://www.FreeBSD.org/~jhb/ > PGP Key: http://www.baldwin.cx/~john/pgpkey.asc > "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-qa" in the body of the message