Re: [MiNT] Mintlib: Bindings for EsetShift() and VsetMask()

[Paul Wratt <paul.wratt@gmail.com>]

On Sat, Dec 12, 2009 at 12:49 PM, Peter Persson <pep.fishmoose@gmail.com> wrote:
> Howdy,
>
> The current bindings for VsetMask() (falcon.h) and EgetShift() (osbind.h) looks as follows:
>
> #define VsetMask(andmask,ormask)                                        \
>        (short)trap_14_www((short)150,(short)(andmask),(short)(ormask))
>
> #define         EsetShift(mode)                                         \
>        (void)trap_14_ww((short)(80),(short)mode)
>
>
> According to the Atari Compendium and Tos.hyp, it should be as follows:
>
> #define VsetMask(andmask,ormask)                                        \
>        (short)trap_14_www((short)150,(long)(andmask),(long)(ormask),(short)(overlay))
>
> #define         EsetShift(mode)                                         \
>        (short)trap_14_ww((short)(80),(short)mode)
>
> I don't know which procedures are needed to get this into the CVS, but I guess it could be a good idea to fix it..
>
> -- PeP
>
With these things, TOSHYP is very up to date, but it still has known
"wrong things", simply because they follow the official documentation,
which was wrong and never fixed (before Atari went bust), and no one
has taken the time to clarify or document the differences. I could
probably find the exact reference I have not quoted here, it might
take me a while, but it was related to getting video "something"

If you want to clear this up (as I do) then do some tests, like I want
to do a VDI & AES versions and compaibility profiler. If you dont have
all the hardware, make sure your tests are simple to run on others
hardware. And find every piece of documentation you can that relates
to what needs fixing.

If worst comes to worst, the community as a whole can decide what
should be done, in light of how it might affect ACP/Coldfire.. I dont
think it will come to this, but I know there are still divisions among
those who believe one way, and those who dont

As far as I know most of TOSHYP is up to scratch, but you can try and
search for a blog made by someone in november that discuses some of
the things you are going to turn up, the comment to his blog turn up
some interesting things. There is also some online message systems
from the 90's that are available and mirrored around the net, sorry I
can find a reference, you will know it when you find it, it had an
Atari Section, and an ST subsection, was text based and in the states

Anyway, just know that is "delacate", but you are right, it needs
fixing, or at least documenting its difference until a fix can be
confirmed. I will try find that blog for you.. it was not the only
article I read concerning this, but I would have no idea where to find
the others (except what I said above), although they may be on some
ftp/cdroms..

Hey, BTW, when you find anything like this, can you keep them in a
list in a text file, so I can use them in a scripted search of source
files.. I have a few things to get finished before I can get around to
doing practical stuff (like the profiler), but I will get there, it
will be handy to know someone has recently covered the same ground, so
keep me/us posted


Paul
PS I think tests will be the quickest way to find a solution, but lots
of different hardware, as much as possible