[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [MiNT] Future (was Re: MiNT 1.16)

> > Yes, but the fvdi, IIRC, is loaded AFTER MiNT.
> no. It's loaded before MiNT.

Ok, I didn't know.

> > Of course, this also applies to NVDI, currently.
> NVDI is also started *before* MiNT.

Yes, it is. But anyways, MiNT currently doesn't touch trap #2 and trap #15
vectors at all. This is why it doesn't have control on NVDI, even if it is
loaded before MiNT. I just have no idea, how to solve the control -
restrict VDI to root? This doesn't seem plausible. Perhaps it rather
should be restricted to software which is running on the console. The same
for most XBIOS functions, which are currently restricted to root only
(yes, by me). Changing the restriction could definitely make non-root
users able to run machine-specific software on the console, and prevent
remote users from doing the same.
> > > Seems like giving up on virtual memory would be easier than patching all
> > > of our TOS software base :)
> > 
> > I.e. we have to drop the progress in this direction and stay with bad
> > programs forever. Not good.
> Is it real true that virtual memory does not work when someone calls
> Super()?

Frank wold know better, but as I understand it, the problem is two types
of memory. The virtual memory system is easier to do, if user programs
reside in user memory only, and the OS in supervisor memory. User mem is
swapped, super mem is not. Now, if an user program runs supervisor code,
it cannot be swapped... in fact this introduces a third class of memory,
which doesn't have support from the hardware side.

But it is possible that I understand it wrong :-)

Konrad M.Kokoszkiewicz
mail: draco@atari.org

** Ea natura multitudinis est,
** aut servit humiliter, aut superbe dominatur (Liv. XXIV,25)
** Taka to juz natura pospolstwa, ze albo sluzy ono unizenie,
** albo bezczelnie sie panoszy.