Subject: Probe/Match routine causing interrupt
To: None <tech-kern@NetBSD.ORG>
From: Martin Husemann <>
List: tech-kern
Date: 08/05/1997 06:54:27
[Yet another stupid autoconfig question, thanks to everyone answering my
previous ones, especially Jonathan. I finaly got enlighted and found the
solution to the puzzle - well, most of...]

In the daemon book I read the probe routine should try to cause an interrupt
by the device so the system can figure the interrupt and configure it
accordingly. Only if thats not possible, the probe routine should hardwire
the interrupt.

I didn't find this in our current code - is it still valid? If yes, could
anyone point me to an example where this is done (an i386 driver prefered)?
How can my probe routine verify the interrupt happened?