riddle@woton.UUCP (Prentiss Riddle) (07/06/88)
According to discussion I've seen on the net, the current release of
the 80386 chip contains a bug which can cause a machine to crash when it
is paging and doing floating point operations on an 80387 math
co-processor. Some vendors of '386 boxes have come up with hardware
fixes to get around this problem.
Does anyone know what the status of the AT&T 6386 is in relation to
this matter? Our local tech people haven't been able to find any
reference to the problem, which may mean either that AT&T found and
fixed it early on, or that AT&T still hasn't dealt with it. Since
we're in the process of buying a 6386 with a '387, we'd like to
know...
Please reply by mail and I will summarize.
-- Prentiss Riddle ("Aprendiz de todo, maestro de nada.")
-- Opinions expressed are not necessarily those of my employer.
-- riddle%woton.uucp@cs.utexas.edu uunet!ut-sally!cs.utexas.edu!woton!riddletam@mtunf.ATT.COM (TED MYSLINSKI) (07/07/88)
In article <1072@woton.UUCP> riddle@woton.UUCP (Prentiss Riddle) writes: >According to discussion I've seen on the net, the current release of >the 80386 chip contains a bug which can cause a machine to crash when it >is paging and doing floating point operations on an 80387 math >co-processor. Some vendors of '386 boxes have come up with hardware >fixes to get around this problem. > >Does anyone know what the status of the AT&T 6386 is in relation to >this matter? Our local tech people haven't been able to find any >reference to the problem, which may mean either that AT&T found and >fixed it early on, or that AT&T still hasn't dealt with it. Since >we're in the process of buying a 6386 with a '387, we'd like to >know... > >Please reply by mail and I will summarize. > >-- Prentiss Riddle ("Aprendiz de todo, maestro de nada.") >-- Opinions expressed are not necessarily those of my employer. >-- riddle%woton.uucp@cs.utexas.edu uunet!ut-sally!cs.utexas.edu!woton!riddle (I tried to mail this but it was returned. Anyway.....) Their exixts a daughterboard made by Ironwood Electronics, of Minneapolis, Minnesota, that fixes the paging problem on the 6386 machines. The board, called the Fix-386, is about the same size as an 80386 and is placed into the 80386 socket. The 80386 is then plugged into the Ironwood daughterboard. Fix-386 is available directly from the manufacturer as part number FX-PGA 132-01. The list price of the board is $145, but AT&T referrals will receive a 25% discount. To order the Fix-386, contact: Ironwood Electronics Attention: Paul Jasmin P. O. Box 21-151 St. Paul, MN 55121 Ted Myslinski