NetNews Usenet Archive 1992 #18/NN_1992_18.iso/spool/comp/os/msdos/programm/8634

Internet Message Format | 1992-08-19 | 1.6 KB

Path: sparky!uunet!charon.amdahl.com!pacbell.com!mips!decwrl!csus.edu!netcom.com!netcomsv!cruzio!aki
From: [e-mail]
Newsgroups: comp.os.msdos.programmer
Subject: Really weird things with 8250's, PC/XT's, IRQ's and I/O
Keywords: What's wrong now?
Message-ID: <[e-mail]>
Date: Wed, 19 Aug 92 01:50:25 PDT
Sender: [e-mail]
Reply-To: [e-mail]
Lines: 28


I have been working on a PC diagnostics program for some time
now.  Today I became aware of a problem that had never surfaced
earlier.  It seems that serial port testing code that works
perfectly on every computer I've tried it on;  suddenly starts
malfunctioning on a PC/XT with an 8 MHz V20.

The code consists of a _huge_ amount of inp's and outp's with
IRQ's blazing at full speed (well... c. 900 times per sec anyway).

The funny thing is,  that after disabling interrupts for the
duration of all I/O functions,  I ended up with less problems.
Even more elusive is the exact nature of the "bug".  I run the
same function two times perfectly with _no_ problems;  on the
third try it bombs.

Is this an obvious "see figure 1",  or am I tracking something
that my time is going to be wasted on?  I know there won't be
too many PC/XT users with NEC V20's out there rushing to use
my program;  thus I could label this as a "feature".

Anyway,  I'd greatly appreciate any input on this phenomena.

Aki.

--
/  Phone: [phone]   Fax: [phone]  |  "Aki" pronounced: Ah-Key. I know  \
\  125 Searidge Ct #D, Aptos, CA 95003  |  what I'm doing most of the time.  /