Xref: utzoo comp.unix.i386:1791 comp.unix.xenix:9049
Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!samsung!caesar.cs.montana.edu!milton!milton.acs.washington.edu!tom
From: tom@yang.cpac.washington.edu (Tom May)
Newsgroups: comp.unix.i386,comp.unix.xenix
Subject: FP problems - buggy 386/7?
Message-ID: <TOM.89Dec16120605@yang.cpac.washington.edu>
Date: 16 Dec 89 20:06:05 GMT
Sender: news@milton.acs.washington.edu
Distribution: comp
Organization: Center for Process Analytical Chemistry, U of Wash, Seattle
Lines: 27


This is probably asked all the time, etc., etc., etc., but I just
started using 386/ix and Xenix 386, so I haven't been following these
newsgroups.  Which 386 and/or 387 processors have bugs?  I have a very
simple floating point program which produces varying incorrect results
under both 386/ix and Xenix, so I suspect my hardware, which is a
Compaq 386/25 with the following identification on the chips:

	80386:	A80386DX-25
		L8281253
		(m)(c)i '85 '86
		and two sigmas in the lower right corner

	80387:	A80387-25
		8X106
		L8300722
		(m)(c) INTEL '86

	82385:	A82385-25 (B)
		SZ121
		L8271301
		(m)(c) INTEL '87

If these chips are hosed, which ones should I get?  Thanks.
--
Tom May
tom@yang.cpac.washington.edu