Date: Sun, 16 Nov 1997 01:44:55 -0800 From: David Greenman <dg@root.com> To: "David E. Cross" <dec@phoenix.its.rpi.edu> Cc: freebsd-hackers@FreeBSD.ORG Subject: Re: 'Official' Intel Fix Message-ID: <199711160944.BAA13671@implode.root.com> In-Reply-To: Your message of "Sun, 16 Nov 1997 03:46:00 EST." <Pine.BSF.3.96.971116034050.26480E-100000@phoenix.its.rpi.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
>Well, I have seen the 'official' intel fix... it is the one to move the >IDT over a page boundry. As has been said before, it is a nasty hairy >kludge. > >Q: Does our fix 'work'.. in the SIGBUS code can we acurately determine if >it was really SIGILL and return it? (what I am asking is why Intel picked >that evil workaround) Yes, I've found a way to distiguish the illegal instruction and generate the proper SIGILL in all the cases that matter. I'm having trouble with the fix working reliably, however, so we may end up doing the Intel fix afterall. It is too early to tell. >Q2: Considering that Intel's 'official' fix can result in a fair >performance hit, any word on if they will be doing a recall? I haven't heard anything from Intel other than they intend to have the bug fixed in a future stepping of the part. If I were calling the shots at AMD, I'd issue rebate coupons for people upgrading their broken Pentiums. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199711160944.BAA13671>