Re: [MiNT] Shutdown() discussion

[Petr Stehlik <joy@sophics.cz>]

On Thu, 2003-12-11 at 11:21, Frank Naumann wrote:
> > The NatFeats were designed to slowly replace all currently emulated
> > hardware. It already happened for graphics, sound and mass storage I/O
> > while networking was always our implementation (not based on any
> > previous HW). So basically we are almost ready for getting rid of all HW
> > emulation in ARAnyM and so we are definitely not going to implement a
> > special hardware I/O again.
> 
> To come back to the poweroff feature, I don't think this is very
> timecritical and need to be optimized very much.

I am against defining two different ways of accessing the NatFeats. We
discussed this at LARGE in ARAnyM list and while I was suggesting the HW
I/O originally I gave up on that after the dicussion and was convinced
to follow the illegal instructions approach (since the cookie jar was
just a helper for TOS applications and a wrapper for NatFeats that was
there to allow different platforms to implement NatFeats differently -
ARAnyM itself uses two illegal instructions for actual invocation of the
functions).

So let me repeat again, the Cookie was a wrapper for TOS. Feel free to
define a different wrapper for MiNT. Both would call the currently
illegal instructions of MC680x0 if running on ARAnyM. But I don't agree
with adding HW I/O to ARAnyM itself (feel free to add it to CT60, for
example, and follow the NatFeats proposals - the cookie/syscall wrapper
ensures that applications will continue to work with it).

> And on work I use vmware a lot. It's a nice thing and it emulate all
> hardware except the processor. It's very fast. For graphics it install
> drivers for the guest os. Why not go such way?

We do go the very same way, if you haven't noticed! NatFeats are the
bridge between the guest and host OS.

BTW, I can still feel signs of misunderstanding of NatFeats. I am glad
we are discussing it but if you wanted to know more just download ARAnyM
source tar ball and read aranym/src/natfeat/doc/native.txt. That could
help a LOT in this discussion. Thanks.

Petr