[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: vm02: calling all IIcs - Take 3



David Schmenk wrote:
David Schmenk wrote:
Whew. I really hope this works. Thanks again to all who tested and submitted feedback on this project. While spending so much time working out the IIc VBL issues, I was negligent with the II/II+. I went to re-test the classes that use the 80 column card and found a mess. My videx doesn't do well being interrupted. So, I finally worked out a solution by disabling interrupts during initialization and setting MSLOT myself before calling the output routine (with interrupts enabled). I'm not terribly pleased with the character output code which uses the CSW function pointer, so I may write a real output routine for the videx and IIe/IIc. I understand that the un-enhanced IIes disable interrupts during screen access which disrupts high speed serial I/O. Can't have that.

Here is the latest image:

http://schmenk.is-a-geek.com/tarfiles/VM02ALPHA4.DSK.zip

Dave...

Darn it - while testing this, my IIc stopped getting interrupts. Of course I forgot to add the back door Ctrl-C check. This image has the Ctrl-C test to break into the monitor - even if interrupts get lost.

http://schmenk.is-a-geek.com/tarfiles/VM02ALPHA4A.DSK.zip

Sorry,

Dave...

I seem to be replying to myself a lot on this version. I have actually captured a failure on my IIc (and remembered to assemble in the Ctrl-C break code). Unfortunately it look like something is deeply confused about the thread state. The good news is mouse interrupts are probably working better than I thought. Time to review my critical sections.

Dave...