🚀 go-pugleaf

In article <1992Aug29.044807.15808@access.digex.com> terryfry@access.digex.com (Terry Fry) writes:
>
>DosPortAccess is part of the API, however it was never implemented.
>
>Run a program that writes to a set of ports in an IOPL segment,
>then call DosPortAcess restricting access to these ports, and write
>again.  Nothing is going to happen, no protection fault.

Are you sure?  I thought it works.  I'll have to find some way to test
it.  Note that you can't restrict yourself from a port with
DosPortAccess.  You can only restrict other apps from it.  Your app
will have to lock the port (via DosPortAccess) and go into a loop.
Some other app must then try and access the port.

If I read the API specs correctly, DosPortAccess MUST be called before
performing any PORT I/O, in order to secure permission from the OS
before using the ports.  Of course, the ports must be accessed from an
IOPL segment.

It may, however, be voluntary locking.  But this runs counter to
OS/2's design philosophy.

>I can't remember why IBM never finished its implementation, and
>I havn't remmebered where I read that it doesn't do anything.
>
>Any IBM'ers want to shed some light?

Yes, please do.  I'm confused about this now...


-- 
   |)  David Charlap           "I don't even represent myself
  /|_  dic5340@hertz.njit.edu   sometimes so NJIT is right out!.
 ((|,)
  ~|~  Hi! I am a .signature virus, copy me into your .signature file.