On 4/13/07, Patrick McNamara <[EMAIL PROTECTED]> wrote:
Timothy Normand Miller wrote:
[snip]
>
> Perhaps it would just be simpler to always set the interrupt in the
> right instruction and have a mask bit in the global config registers.
>
I would certain argue for the second option.  It may take slightly more
hardware to do, but it is certainly the simpler and more portable
solution.  I don't know if we have discussed how video controller
microcode is uploaded or changed, but enabling/disabling the interrupt
should a PIO operation I think.  The reason for this is the expectation
the system may have surrounding the toggling of interrupts.  If it is
done as a DMA command to the card, it could be some time before the DMA
completes, especially if there is a large transfer of some sort
currently in progress.  If interrupts need to be disabled, they
generally need to be disabled immediately, not some "random" time in the
future.

I completely agree.  In fact, it's being handled entirely via PIO at
the moment anyhow, since we don't have a DMA engine.  The DMA engine
may (when the right privileges are set) have access to every register,
but at least by convention, we won't use DMA for certain things.

--
Timothy Normand Miller
http://www.cse.ohio-state.edu/~millerti
Open Graphics Project
_______________________________________________
Open-graphics mailing list
[EMAIL PROTECTED]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)

Reply via email to