Julian Hall wrote:
>
> Quick followup: up until the 1st, this still happens with the CVS version.
> Haven't tried a more recent version than that.
>
> One thing that did occur to me: the machine I'm running this on is a 386 -
> is anything being done that might not work on it?
For now we're targeting Pentium+, and using feature specific to
that level. We need to save/restore all the CPU state on
switches between the host and monitor contexts. Some
of that context is Pentium specific, like the CR4 as
Ulrich mentioned. Also, we'll need to deal with the
Time Stamp Counter (TSC) and other registers which are not
on the 386. You might be able to omit access to those
registers for the moment, but it will be more difficult
as the code progresses.
-Kevin
