Why is the high speed channel light not updated?

[Bill E]

It's commented out in panel1.c updatelights(). The light is there, and now there is a high speed channel effectively in the drum impl, it does 'dma' like the hardware drum did, it could update the hsc flag.

Any reason to not enable it?

Bill

[Bill E]

And, hmm, it's labeled brk cycle on our panels, but h.s. cycle in the manual. Another random PDP-1 variation?

Bill

[Bill E]

Last data point, I enabled the hs update just to check it out, the brk cycle light doesn't seem to ever be updated in the distribution, it does seem to be the hs light. I have the drum IOT set hs when it starts a transfer, turn it off when done, light works. This mimics what would have been happening in the real hardware with the high speed data channel hardware.

Angelo, opinion? Enable the hs light, aka brk cycle or not?

Bill

[Bill E]

Turns out the physical panel does have the hs light, it's the virtual panel that's wrong. So, I'm going to check enabling the hs light into my fork, can't have enough blinking lights. The drum is the only thing using it now, but I'm going to generalize the dma in it to emulate the High Speed Channel hardware to make dma more easily available to people writing device drivers and such. It will control the hs light. Stay tuned.

Bill

[Angelo Papenhoff/aap]

I'm pretty sure both panels should have it, it's just that my emulator doesn't implement it (yet). we should really figure out how to merge our code before it diverges too much. really sorry for being so absent here X_X

[Bill E]

It was commented out in pdp1.h. I just put it back in, seems to work fine.

Bill