From owner-freebsd-hackers Mon Nov 17 12:28:13 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id MAA11199 for hackers-outgoing; Mon, 17 Nov 1997 12:28:13 -0800 (PST) (envelope-from owner-freebsd-hackers) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id MAA11146 for ; Mon, 17 Nov 1997 12:27:56 -0800 (PST) (envelope-from toor@dyson.iquest.net) Received: (from root@localhost) by dyson.iquest.net (8.8.7/8.8.5) id PAA05034; Mon, 17 Nov 1997 15:27:44 -0500 (EST) From: "John S. Dyson" Message-Id: <199711172027.PAA05034@dyson.iquest.net> Subject: Re: Status of patch? In-Reply-To: from Alfred Perlstein at "Nov 17, 97 08:32:32 am" To: perlsta@cs.sunyit.edu (Alfred Perlstein) Date: Mon, 17 Nov 1997 15:27:44 -0500 (EST) Cc: hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Alfred Perlstein said: > i'm about to do a 'make world' and remake my kernel too, i was wondering > if and what has been done to the kernel to fix the 'f00f' bug? > There are people working fairly aggressively on the problem. > > i don't want to rebuild everything if there has been no fix for the 'f00f' > bug. > The fix isn't in the tree yet, however, if you are running an ISP or shell machine and NEED the fix now, the SEF fix is likely the right way to go for now. I wouldn't apply any fixes unless you really need them yet. > > what kind of performance impact are we talking about with the fix? > Probably barely measurable. > > i was also wondering if any of the alternative fixes people have brought > up seem to work? > So far, it appears that the Intel-suggested approach is the safest and most reliable today. That can change, but I think that everyone agrees that the standard fix is ugly, and people are still looking at alternatives for inclusion into the official sources. There are some other, innovative approaches being tried. The Intel-suggested approach has some negative impact on some of the esoteric features of FreeBSD, and will require mods in various parts of the kernel. -- John dyson@freebsd.org jdyson@nc.com