From owner-freebsd-hackers Thu Nov 13 11:57:08 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id LAA13955 for hackers-outgoing; Thu, 13 Nov 1997 11:57:08 -0800 (PST) (envelope-from owner-freebsd-hackers) Received: from server.local.sunyit.edu (A-T34.rh.sunyit.edu [150.156.210.241]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id LAA13940 for ; Thu, 13 Nov 1997 11:56:57 -0800 (PST) (envelope-from perlsta@cs.sunyit.edu) Received: from localhost (perlsta@localhost) by server.local.sunyit.edu (8.8.7/8.8.5) with SMTP id QAA15677 for ; Thu, 13 Nov 1997 16:00:54 -0500 (EST) X-Authentication-Warning: server.local.sunyit.edu: perlsta owned process doing -bs Date: Thu, 13 Nov 1997 16:00:54 -0500 (EST) From: Alfred Perlstein X-Sender: perlsta@server.local.sunyit.edu To: hackers@FreeBSD.org Subject: Re: BSDI F0 bug workaround implementation (help with understanding) In-Reply-To: <199711131907.MAA28899@usr08.primenet.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk I've been following the thread on the Pentium CPU bug, i was wondering how a workaround is possible? it just seems very interesting. -Alfred On Thu, 13 Nov 1997, Terry Lambert wrote: > > I'm not sure I understand the full implications of the impact of this > > hack, although it is worrisome. Judging by what my pentium book says about > > the layout of the IDT, it seems like it will increase interrupt latency > > for page faults and many maskable interrupts. Can anyone more > > knowledgeable than I comment on this? Page fault overhead on freebsd is > > pretty high: would a short-cut make sense that does not go through the > > full vm system for this? Otherwise page fault overhead may come close to > > doubling ... > > Only the first 7 IDT entries are affected (at least in the Linux > workaround), not the whole table. > > On the minus side, the impact *is* non-zero. > > Like the FPU bug, since a workaround exists, it's likely to be swept > under the software as well. 8-(. > > > Terry Lambert > terry@lambert.org > --- > Any opinions in this posting are my own and not those of my present > or previous employers. >