Ramon van Handel wrote:
> Perhaps we can solve problems like these by
> explicitly having a guest_flags and a
> monitor_flags in the context structure, and updating
> both fields according to different criteria; how they're
> emulated is a different matter (not too hard I think,
> we can make a set of macros which will generate
> a "real" eflags from monitor_eflags+guest_eflags)
> but at least not ambiguity can arise. We'd probably
> need something similar for the control registers.
> What do you think ? Shall I work on this next weekend ?
Yes, I agree. We'll need 'guest' variants of
EFLAGS (IF, IOPL, NT?, ...)
segment registers (RPL)
control registers (various flags, not sure right now)
debug registers (maybe)
Bye,
Ulrich
--
Ulrich Weigand,
IMMD 1, Universitaet Erlangen-Nuernberg,
Martensstr. 3, D-91058 Erlangen, Phone: +49 9131 85-27688
