Re: A channel access problem probably due to something else.

[winans@aps.anl.gov]

>These records are all processed synchronously, so it appears that I
>don't need to mess with PACT.  Is this a reasonable assumption?

Yes.  This is true for synchronous records.

>Any pointers/suggestions/wild guesses will be greatly appreciated.

Going from periodic to passive _should_ not make any difference to the
driver.  I would make sure that a passive record that gets processed
due to a poke at the PROC field is working properly before embarking on
other fun things like CA links.

What you probably have is a spin in an IRQ handler... is the VME light
stuck on?  That would indicate that something is spinning on a value it
is getting from an IO board.  Do you have a VME backplane tracer?  I
have found them to be of great value at times like this.

--John