Subject: Re: Possible (but Evil) work-around for Pentium bug?
To: Michael K. Sanders <firstname.lastname@example.org>
From: Rick Byers <email@example.com>
Date: 11/12/1997 16:36:48
I think the problem here is that we are confusing two issues. There are
TWO recently reported "pentium bugs". The first (and probably the concern
of people on this list), is the F00FC7C8 bug in Pentium and Pentium MMX
chips which causes them to lock up. This bug DOESN'T affect Pentium Pro
or Pentium II chips. The second bug is (although I don't know all the
details), a "minor" floating point problem in Pentium II and Pentium Pros
where a flag isn't being set when a floating point number doesn't convert
to an integer properly (atleast, this is what I've read).
According to Intel, only the Pentium Pro and Pentium IIs have software
upgradable microcode. This would explain the SCO fix - it fixes the
floating point problem. Does anyone know which problem BSDI has fixed?
I'm not sure if the F00FC7C8 bug can properly be fixed in software
(without diabling the L1 cache) - but it appears the floating point bug is
being fixed by a microcode update.
Hope this helps clear up some confusion,
On Wed, 12 Nov 1997, Michael K. Sanders wrote:
> In message <firstname.lastname@example.org>, "Erik E. Fair" writes:
> >I got the following today, which hints at a work-around with unacceptable
> >performance loss: turning off the L1 cache.
> Hrm. I just noticed that the referenced SCO document specifically
> mentions Pentium Pro and Pentium II, but it still seems to imply
> that a microcode upgrade should be possible for the P5 for this
> particular problem. Anyone have any other information?
Rick Byers Internet Access Worldwide
email@example.com System Admin
University of Waterloo, Computer Science (905)714-1400