Hi;
Todd L. Miller wrote:
> Er, I phrased this poorly because we don't distinguish between
> IRQ's and interrupts. Basically, we don't currently use
> InterrupterDispatcher at all. This induces a hack in the keyboard driver
> that will lose scancodes if >1 is generated per interrupt. This might be
> 'fixable' by re-implementing InterrupterDispatcher. However, I don't know
Hmm... I vaguely and fuzzily recall something about this. Was the "bug"
or misfeature in the native code, the Java, or in the interaction
between the two?
> how much work this might entail, and as long as you're cloning the JavaOS
> APIs for memory, I might as well clone the APIs for IRQs/interrupts.
> (Rather, clone the parts that I need.)
Ah. Cloning the APIs. (Well, I guess I misread subject, verb, and
object of your description. Three in one sentence. Pretty much
everything -- didn't miss a single chance to misintepret something.
This is some sort of personal record. :=)
> Er, these virtual consoles are implemented entirely in Java, and
> besides being "cool" (and wonderful for debugging), it's a good test of
> the driver system. If the VGA driver ends up doing our text-display work
> (and/or includes a text/video mode switch), I can use that method/code to
> (help) handle switching the consoles.
(Well, I'm 2 for 2 here. Thought you were talking about native code.)
> > I was planning to hook up the GC stuff this weekend. By the flurry of
> > email from me today, I am sure you can tell I'm at home hacking away,
> > and not at the Day Job (and I won't be again until Monday).
>
> Hey, great, anything that makes my job easier :)
Let's hope my C++ and Java are better than my English.
Embarrassedly Yours,
-jm
--
==== John Morrison ==== [EMAIL PROTECTED] == http://www.mak.com/welcome.html
==== MaK Technologies Inc., 185 Alewife Brook Pkwy, Cambridge, MA 02138
==== vox:617-876-8085 x115
==== fax:617-876-9208
_______________________________________________
Kernel maillist - [EMAIL PROTECTED]
http://jos.org/mailman/listinfo/kernel
