On 22 Dec 2010 at 18:09, jim s wrote:
Developers didn't know about it. A friend of mine
found and
documented it. It was not even in the Red covers for the processor.
Actually, I find that entirely believable. We were working with the
early steppings of the 186. We also had early steppings of the 286
in-house but hadn't done much other than test real-mode operation;
Intel was working on the kernel for Xenix; In retrospect, I wonder if
the Intel Xenix team even knew about the LOADALL instruction.
Anyway, back to the 186.
I recall took us about 2 weeks to track down one such nasty bug, all
while our Intel apps contact sort of stood by with his hands in his
pockets. When we informed him of the bug, the response from the chip
development group was "oh yeah, we knew about that a couple of weeks
ago".
Communication within Intel was terrible by the time it got to the
customer. And Bill Davidow was on our board of directors, so it
wasn't as if we didn't have any influence.
--Chuck