hello,

On Sat, 23 Jun 2001, Anders Magnusson wrote:
> What happens is that you get an interrupt that changes things.

ok, thats what i thought, that forbbing the status register causes it
to generate an interrupt. have i done something wrong that an interrupt
handler didnt pick it up and then restore the old psl on the REI? 
did it, and just not restore the right status? 
or should interrupts be blocked when im doing this? 

> Note that single-step on the 4000 machines does a "software" single-step
> and do not follow interrupts like the 11/750. (It actually sets the
> T bit and executes one instruction, just like a debugger).

ack! well, this explains why it works when i halt around it.. 

isildur