Hi !
> > > Once we have accepted to
> > > memory map the fb into the userspace process adress space the only
> > > way to prevent it (even temporarily) to access this area is to
> > > unmap everything.
> > Well - to unmap the part that might conflict. This may vary between cards.
> That was my question the other day. So some cards do conflict on a region
> access bases.
Yeah, but probably not quite as you think. It's not directly conflicting
Vidmem access, but any FB-access on Virge IIRC. Access to accelspace still
works.
> Well the unmapp solution isn't that bad on a UMP machine but on SMP this
> is really bad.
With a high "M", of course this gets more and more ugly.
> On SMP machines IMO proper scheduling will be the best solution.
Yeah. I'd implement both and make it a runtime/compiletime switch.
> This solution will not be that bad on a UMP machine either.
Well - it has a bad penalty, if both apps rarely make use of the mmap,
but rather use accel almost always.
> With cards framebuffers growing in memory size
Who says you need to map it all ?
CU, ANdy
--
= Andreas Beck | Email : <[EMAIL PROTECTED]> =
