Update: 2.0 kernels, tulip driver, crashes and reboots (long)

[Al Youngwerth]

At 04:04 PM 1/25/99 -0600, Dave Weis wrote:
>
>
>> > We still don't know exactly what causes the problem or why different
>> > kernels seem to affect the problem. What we do know is that with APM and
>> > USB disabled we have no problems.
>> There are some background hints on how the SMI stuff works with USB devices
>> and USB "legacy mode". Thats the mode where SMM code is used to fake I/O
>> to the keyboard port and talk USB hidbp to the USB devices from DOS. With
>> no keyboard present its reasonable to expect the USB polls to be active.
>
>The box that I was having the problems with had an attached keyboard with
>APM and USB off.

If you have access to an oscilliscope or better yet a digital oscilliscope,
you can try probing the SMIACT# signal when the system is running normally.
You should see this signal high all of the time. If it ever pulses low,
then you're getting SMIs.

If you have access to a DMM (Digital Multi Meter), then the next time it
locks, try probing the SMIACT# pin. It's not that hard to do and a DMM can
be had for less than $50 bucks. If want to try this and you need any help,
e-mail me privately and I can help you set it up.

Thanks,

Al Youngwerth
alberty@apexxtech.com