aeusesef@csun.UUCP (Sean Eric Fagan) (07/02/87)
[]
After hearing (here and elsewhere) about the 80386 multiplication bug,
I was hoping someone could tell me more about it. For example, I know
it shows up in 32-bit multiplies, but how? What happens? Can I get
some examples of data with the corresponding wrong answers?
Thank you for your support.
-----
Sean Eric Fagan Office of Computing/Communications Resources
(213) 852 5742 Suite 2600
1GTLSEF@CALSTATE.BITNET 5670 Wilshire Boulevard
Los Angeles, CA 90036
{litvax, rdlvax, psivax, hplabs, ihnp4}!csun!{aeusesef,titan!eectrsef}
--------------------------------------------------------------------------------
My employers do not endorse my | "I may be slow, but I'm not stupid.
opinions, and, at least in my | I can count up to five *real* good."
preference of Unix, heartily | The Great Skeeve
disagree. | (Robert Asprin)jim@applix.UUCP (07/14/87)
In article <657@csun.UUCP>, aeusesef@csun.UUCP (Sean Eric Fagan) writes: > [] > After hearing (here and elsewhere) about the 80386 multiplication bug, > I was hoping someone could tell me more about it. For example, I know > it shows up in 32-bit multiplies, but how? What happens? Can I get If you are a COMPAQ Deskpro 386 owner, a copy of the DOS program "MULTEST" is available from your dealer or COMPAQ. Other 386 owners should contact their dealer, the manufacturer, Intel, in that order. The program actually appears to be a part of the Intel test suite for the 386 but I may be wrong. The program is run on a cold 386 and will display whether or not the 32 bit multiply bug is present. -- -- Jim Morton, APPLiX Inc., Westboro, MA UUCP: ...seismo!husc6!necntc!m2c!applix!jim ...seismo!harvard!m2c!applix!jim
farber@udel.EDU (Dave Farber) (07/14/87)
In article <546@applix.UUCP> jim@applix.UUCP (Jim Morton) writes: >In article <657@csun.UUCP>, aeusesef@csun.UUCP (Sean Eric Fagan) writes: >> [] >> After hearing (here and elsewhere) about the 80386 multiplication bug, >> I was hoping someone could tell me more about it. For example, I know >> it shows up in 32-bit multiplies, but how? What happens? Can I get > >If you are a COMPAQ Deskpro 386 owner, a copy of the DOS program >"MULTEST" is available from your dealer or COMPAQ. Other 386 owners >should contact their dealer, the manufacturer, Intel, in that order. >The program actually appears to be a part of the Intel test suite >for the 386 but I may be wrong. The program is run on a cold 386 >and will display whether or not the 32 bit multiply bug is present. > > >-- >-- >Jim Morton, APPLiX Inc., Westboro, MA >UUCP: ...seismo!husc6!necntc!m2c!applix!jim > ...seismo!harvard!m2c!applix!jim If anyone wants a copy of Multest let me know Dave Farber _______________________________________________________________________________ Office: University of Delaware Department of Electrical Engineering Newark, De 19716 Home: RD 1, Box 279 Landenberg, Pa 19350 Arpa:farber@udel.edu Compuserve: 76010,104 Telex: 6501066405 MciMail: dfarber Telephones: Office(UD): 302-451-1163 Home: 215-274-8292 _______________________________________________________________________________
johnw@astroatc.UUCP (John F. Wardale) (07/15/87)
Instead of all this "test program" non-sense, will someone
*PLEASE* *POST* two values that give incorrect results'
(correct and expected-buggy-values) appreciated too.
Does this ONLY occur on these two values, or is it more general?
Can someone at Intel tell us what caused the bug, and/or
more specifically how it got thru simulation without detection?
--> I'm **NOT** trying to smear anyone with this....I'd like to
know cuz we're doing LOTS of simulation here, and are constatnly
looking for pitfalls (to make improvements).
John W
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Name: John F. Wardale
UUCP: ... {seismo | harvard | ihnp4} !uwvax!astroatc!johnw
arpa: astroatc!johnw@rsch.wisc.edu
snail: 5800 Cottage Gr. Rd. ;;; Madison WI 53716
audio: 608-221-9001 eXt 110
To err is human, to really foul up world news requires the net!