[MiNT] Current beta problems - solved

[Joerg_Westheide@su.maus.de (Jörg Westheide)]

Hi Thomas!

TB>Well, it's just as reentrant as your special workaround code, i.e. it
TB>uses additional fields in the context struct.
As these fields in my workarounds are function specific these workarounds
work unless one of these functions is called from within code that is called
from these function, which means it works unless there's a recursion.
If you use one field for multiple functions that will work at the moment, but
also will cause problems later when some new functionality provided by the
new handler is used (and that's the reason why I use one field for each
(function specific) workaround)

TB>Sure, but reentrancy will only be possible at kernel level,
TB>anyway (i.e. no GEMDOS call of the same process may be made within a
TB>GEMDOS call), because the context concept doesn't allow that, unless
TB>I'm missing something.
What you're missing is that the context isn't saved directly at the begining
of the handler

TB>But for background DMA, that wouldn't be a problem (again, unless I'm
TB>missing something ...)
Wasn't that a result of a discussion here? I thought so

Bye

                Joerg