Travis Geiselbrecht

Hmm, where is the current version of the spec? I can't find it on Notch's site anymore. Is the 1.7 spec you're following 'final'?

Hmm, where did they get that from? The 0x10c site still has the old one.

Looks like it came from http://www.reddit.com/r/dcpu16/comments/sqfre/rfe_dcpu16_11/

Well, this is still a draft I assume. He hasn't officially updated the spec on the 0x10c website. Presumably he'll bump the rev when he posts it there.

Ah I see what it is. 'data' is a keyword so it's getting confused.

Yeah, I think the best solution would be to put the lexer into a 'look for label' mode when at that part of the parser. I'd guess it'd not try...

I was thinking of just running it through cpp as a command line option. I've done it for some other assembler projects, and it works reasonably well.

FWIW, that's precisely how the arm64 ABI works. Passes the first 8 int and and first 8 quad fp args via registers, and the first thing the variadic function does...

x86-64 has an ABI hack to pass a boolean in I believe eax to signify if any fp args are in use, which saves the trouble of having the callee...

Also remember that bug that a particular vendor had where it would stop transmitting in the middle of the character if the receiver dropped CTS in the middle of it?...