Mail Archives: djgpp/1994/04/09/13:31:27
> > I expect that the future will require some form of public DPMI 1.0
> > host to satisfy the masses. Charles Sandmann has mutated go32 into
> > enough of a DPMI provider to run V2.0 apps, but it's not a good
> > solution (one app per go32 still).
>
> Note that Windows 3.1 is a DPMI 0.9 provider only. Who knows what
> Windows 4.0 does, but I'd be willing to bet that it still has DPMI 0.9.
Well, I should find out fairly soon :-)
BTW, I still can't run V2.0 apps, just simple DPMI programs. I still have
a few buggy (and missing) DPMI calls. But I don't see any show stoppers,
just debugging... And it's been a month or so since I looked at it.
> Does this mean that future DJGPP programs will not work in a
> Windows 3.1 DOS session, because DPMI 1.0 is required?
Working with DPMI 0.9 will remain a requirement as far as the eye can see.
DPMI 1.0 would provide some additional features in V2.0 which we have
taken for granted with GO32, things like null pointer detection, stack
expansion, and direct access to low memory mapped in our address space.
My goal is to provide the minimum API set to get V2.0 apps running, add
a few DPMI 1.0 extensions, and then make it available. Eventually it should
be completely re-written in djasm and load things more things into
protected memory (so the extender uses extended memory itself) but this
is a long term project.
- Raw text -