Re: Linux vs MS Security
From: Moe Trin (ibuprofin_at_painkiller.example.tld)
Date: 09/15/05
- Next message: Rick Moen: "Re: Linux vs MS Security"
- Previous message: George Ruch: "Re: Protection Software for Linux"
- In reply to: Rick Moen: "Re: Linux vs MS Security"
- Next in thread: Rick Moen: "Re: Linux vs MS Security"
- Reply: Rick Moen: "Re: Linux vs MS Security"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Wed, 14 Sep 2005 19:20:11 -0500
In the Usenet newsgroup comp.os.linux, in article
<440e$432800b9$c690c3ba$12436@TSOFT.COM>, Rick Moen wrote:
>It was a grave bug in Pentium / PPro processors, discovered in 1997,
>that Intel managed to talk its way out of fixing by some subtle
>misdirection that somehow convinced people that the CPU defect was OS
>vendors' problem.
----------------------------------------
] Path: excalibur.flash.net!nntp.flash.net!sunqbc.risq.qc.ca
!cpk-news-hub1.bbnplanet.com!news.bbnplanet.com!cs.utexas.edu
!geraldo.cc.utexas.edu!not-for-mail
] From: noname@noname.com
] Newsgroups: comp.os.linux.advocacy
] Subject: This code will lock up any P5 machine, even usermode Linux!
(F0 0F C7 C8)
] Date: Thu, 06 Nov 1997 21:57:33 -0800
] Organization: The University of Texas at Austin, Austin, Texas
] Lines: 7
] Message-ID: <3462ADCD.135B@noname.com>
] NNTP-Posting-Host: dial-102-5.ots.utexas.edu
] Mime-Version: 1.0
] Content-Type: text/plain; charset=us-ascii
] Content-Transfer-Encoding: 7bit
] X-Mailer: Mozilla 3.0 (Win95; I)
]
] Hi,
]
] Check this out. If you execute F0 0F C7 C8 on a P5 it will lock the
] machine up. This is true for any operating system including usermode
] Linux. It's pretty cool. Basically, the opcodes are an invalid form
] of cmpxchg8b eax with a lock prefix. Has anyone seen this before? The
] problem doesn't show itself for the Pentium Pro or Pentium 2.
----------------------------------------
I don't think anyone ever identified the discoverer. There was some talk
that whoever it was worked for Cyrix and was reverse engineering the chip.
Others claimed that it was more likely a computer science student at the
University of Texas at Austin. I imagine if Intel were serious, they could
have filed a complaint and the police would have had a look the the dialin
logs - never heard a word about that.
>Regardless, after the bug was publicised on 1997-11-10,
Are you sure of the date? The original posting (above) was late on the
sixth, and the next few days were like someone stomped on a fire ant nest.
>the BSDi people were first to produce a fix, using information they
>received from Intel under NDA -- in something like 2-3 days.
google groups has a thread in comp.unix.bsd.freebsd.misc dated Tue, 11
Nov 1997 16:38:48 -0700 announcing the BSDi fix. It was withdrawn on the
12th apparently because it was released in violation of the NDA, and there
was a lot of bickering in that group over the Linux fix.
>The Linux kernel coders, working _without_ NDA information, were able to
>do likewise within, if memory serves, about one day more.
-------------------------------------------
]From: torvalds@transmeta.com (Linus Torvalds)
]Newsgroups: comp.os.linux.misc,comp.os.linux.hardware
]Subject: Pentium bug workaround, please test!
]Date: 12 Nov 1997 19:27:02 GMT
]Organization: Transmeta Corporation, Santa Clara, CA
]Lines: 20
]Message-ID: <64cvu6$b3f$1@palladium.transmeta.com>
]NNTP-Posting-Host: penguin.transmeta.com
]
]I just made 2.1.63 available on the normal ftp site (ftp.kernel.org,
]directory pub/linux/kernel/v2.1). The most exciting change is probably
]the preliminary patch by Ingo Molnar that should work around the by now
]well-known Pentium lock-up bug. Many thanks to Ingo who put together
]the patch from various snippets of information floating around.
[...]
]Please give it a good testing, especially the Pentium bug workaround.
]Throw all the tests you have at it, to see that it really works. We'll
]be doing a 2.0.x patch for that too, but it's probably not going to
]appear for a few days, so in the meantime testing this fix on 2.1.x
]would be a GoodThing(tm)...
]
] Linus
-------------------------------------------
and
-------------------------------------------
]From: set-usenet-879492588@reality.samiam.org (Sam Trenholme)
]Newsgroups: comp.os.linux.development.system,comp.os.linux.advocacy,
comp.os.linux.misc
]Subject: F00F bug *fixed* in 2.0.x kernels
]Date: 14 Nov 1997 07:27:58 GMT
]Organization: Linux reality.samiam.org 2.0.30 #2 Mon Sep 15 1997 i686 unknown
]Lines: 41
]Message-ID: <64guhu$p7k@news9.noc.netcom.net>
]NNTP-Posting-Host: reality.samiam.org
]X-NETCOM-Date: Fri Nov 14 1:27:58 AM CST 1997
]
]Linux 2.0.x users:
]
]The Linux developers have, again, done the impossible. Within seven days
]of the serious FOOF bug in the Pentium being discovered, the kernel
]developers have not only figured out a software fix for the problem, but
]have patches for *both* the 2.1.63 and the 2.0.31 kernels which make
]Linux immune to the F00F bug.
]
]The patch for the F00F bug successfully works on the latest pre-2.0.32
]version. As I type these words on my Pentium Linux laptop, the sequence
]F0 0F C7 C8 is being run over and over again in an infinite loop.
-------------------------------------------
>Microsoft? They got around to hotfixing some but not all of their
>then-supported OS releases about six months later.
Did they ever fix it at all? I know Novell didn't. My understanding was
that they claimed there was no need to fix it, as no compiler would
produce that machine code. Obviously wrong, as anyone who has ever coded
ANYTHING knows (how do they think it was discovered in the first place,
magic?), but that was their story at the time.
Old guy
- Next message: Rick Moen: "Re: Linux vs MS Security"
- Previous message: George Ruch: "Re: Protection Software for Linux"
- In reply to: Rick Moen: "Re: Linux vs MS Security"
- Next in thread: Rick Moen: "Re: Linux vs MS Security"
- Reply: Rick Moen: "Re: Linux vs MS Security"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
|
