> > 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.
> Disabling it temporarily seems like a good solution, as the program can run on
> until it actually hits the fb. However this can make for a bigger performance
> impact in some situations (where one process does much accel and another
> much fb) that just disallowing the scheduling for fb-mapping processes
> while the accel is running.
> However for the general case I think the unmap method is best.
Well the unmapp solution isn't that bad on a UMP machine but on SMP this
is really bad. On SMP machines IMO proper scheduling will be the best
solution. This solution will not be that bad on a UMP machine either. With
cards framebuffers growing in memory size the unmapping solution will grow
in penalty. The proper scheduling will cost pretty much the same either
way.
