>From: Tom Ivar Helbekkmo <tih@kpnQwest.no>
>Date: 11 Mar 2002 10:58:02 +0100
>
>Anders Magnusson <ragge@ludd.luth.se> writes:
>
>>> I noticed that the kernel is busy-waiting when idle. Can this not be
>>> improved? I am used to the Motorola 68000 which has a WAIT instruction
>>> that halts the CPU until an interrupt occurs...
>>> 
>> Unfortunately no, the vax do not have any such instruction, busy-wait
>> is the only solution.
>
>Maybe it would be a good idea to pretend that one exists?  Add a
>fictive VAX model that differs from the one that SIMH actually
>emulates only in that it has a special WAIT instruction, and then have
>the kernel use that when it detects that it's running on SIMH.  It
>might be necessary to make SIMH do something that allows the kernel to
>realize that's what it's on, of course...

Or perhaps the busy-wait could be done from an emulated instruction on
real vax hardware.  Then SIMH could implement the instruction in a
more efficent way, and you don't need any special checks.  (Just an
idea, I'm not volunteering to write any code for this :-)

--
entropy -- it's not just a good idea, it's the second law.