From owner-freebsd-stable Wed Mar 7 10:10:31 2001 Delivered-To: freebsd-stable@freebsd.org Received: from linux.intcon.net (linux.intcon.net [206.230.48.2]) by hub.freebsd.org (Postfix) with ESMTP id 95B8E37B719 for ; Wed, 7 Mar 2001 10:10:27 -0800 (PST) (envelope-from steve@portal.megahack.com) Received: from portal.megahack.com (portal.megahack.com [206.230.54.106]) by linux.intcon.net (8.11.0/8.11.0) with ESMTP id f27IAJu17621; Wed, 7 Mar 2001 12:10:19 -0600 Received: (from steve@localhost) by portal.megahack.com (8.11.2/8.11.2) id f27IAI973914; Wed, 7 Mar 2001 12:10:18 -0600 (CST) (envelope-from steve) From: Steven Farmer MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <15014.31114.28633.215085@catbert.megahack.com> Date: Wed, 7 Mar 2001 12:10:18 -0600 (CST) To: freebsd-stable@FreeBSD.ORG Subject: Re: ARCH flag in new make.conf In-Reply-To: <20010307025918.A37732@mollari.cthul.hu> References: <20010307013439I.jkh@osd.bsdi.com> <20010307025918.A37732@mollari.cthul.hu> X-Mailer: VM 6.75 under 21.1 (patch 14) "Cuyahoga Valley" XEmacs Lucid Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG >>>>> "Kris" == Kris Kennaway writes: ... Kris> * Jordan's warning added to make.conf is under debate, and I Kris> think it's excessive. I'm happy to see CPUTYPE, but I've had recent firsthand experience with the use of "-O -march=pentiumpro" breaking /usr/bin/make. All was well until the next buildlworld, when make would get SIGSEGVs at various points. To recover, I had to go back to plain "-O" in CFLAGS, cd to /usr/src/usr.bin/make, rebuild and install it, and then start the buildworld again. I was fortunate that make didn't die while rebuilding make - it could have been worse. This was a couple of weeks ago. I didn't post anything about it at the time for a couple of reasons: first, there have been various "don't come crying to us" pronoucements in the past from developers regarding the use of anything but "-O" for builds. Second, I didn't want to cause a bunch of "it's your hardware" postings. It's NOT my hardware! Well, ok maybe -march=pentiumpro tickles a microcode bug in cpus of a particular stepping, so maybe that is "hardware". But the fact is that everything, so far as I know, works fine with plain "-O". BTW - I have a _genuine_ Pentium Pro, not one of those Johnny-come-lately Pentium IIs or IIIs. :-) It won't be the end of the world for me if I can't use CPUTYPE, but it would be wonderful if there was a way to override the global CPUTYPE setting on a case-by-case basis to fix the occasional breakage. I'd really like to have my cake and eat it too. Just a thought. Steve To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message