uport V386 fp

[Bill Hatch]

Subject: uport 386 fp problems
About 6 months ago there were several news articles about 
floating point problems with microport V386 with and without
the 80387 coprocessor.  My specific problems (without 80387)
were solved when microport/john p. shipped me an engineering
update to the dos merge kernel in december.  

I purchased a hardware fix to the 80387 problem (erratum 21) from
Ironwood Electronics.  However, in the process of reinstalling the
80387, I accidently burned out the chip.  Therefore I was unable
to test the Ironwood fix.  (Note: the hardware fix from Bell
Technologies was not compatable with my ALR 80386/220).  

We tested uport 3.0e on a compaq 386/20 portable (with 80387)
and there was no problem as long as only one process was using
floating point operations.  If 2 processes used floating point,
then the os crashed.

My company is in the process of purchasing several 80386 machines
with 80387 coprocessors and sockets for a Wytek.  One of the
machines is a Hewett Packard (Vector ?) and the others are by
Wells American.  I would like
to get some answers to the following questions:

1. Has anyone installed the Ironwood hardware fix; and did it fix
the problem?  On what hardware configuration and what brand/release
of 80386 Unix?  Also the Bell Technologies hardware fix ?
Does anyone have Ironwood's address or phone - i lost these in a
disk crash last month.

2. Is this strictly a Microport V386 problem or is it a hardware
(80383 chip) problem present to disrupt all varieties of 80386
Unix unless hardware remedies are applied ?

3. Has anyone managed to get any version of V386 unix working
with an 80387 so that more than one process can do floating point
operations on a "generic" AT386 machine?  (Our intention is to
use our unix machines as multi-user software development machines.)


Our busines depends heavily on floating point intensive applications.
I have been successful in persuading the management to try V386 Unix on
2 projects as a replacement for individual 8086/8087 DOS clunkers.  
Thus, I will be very grateful for any information in response to the 
above questions and comments on related issues. 


bill hatch
uunet!bts!bill

	Computational Engineering
	14504 Greenview Drive Suite 500
	Laurel, Maryland 20708
	Phone (301)470-3839
	FAX (301)776-5451
	HOME: (301)441-1675