Daniel Thornburgh

Long term, yes. Short term, no. Getting this far in the project has required doing a considerable amount of fiddling with upstream passes; however, we're still not quite to human-level...

> We may work on 8-bit codegen at some point, but that is not where our project is oriented at this time; as such, when/if we do get around to...

> Getting GCC working shouldn't be that difficult This is the first time I've heard someone describe working with GCC this way, and given the project's history, I hope you...

It seems that I never actually implemented the part of emitPrologue and emitEpilogue that are supposed to handle this, so this doesn't work yet for soft stacks either.

This is harder to support than it initially seemed, due to the way we handle frame pointers. We take the frame pointer to be the value after SP was incremented,...

I'd be a bit careful with "officially"; none of these open source projects are officially affiliated with or supported by the owners of the 6502 or 65816 IP. There is...

Regardless, it's a bit presumptive to declare anything we make "the official X for Y", particularly when Y may be a trademark owned by someone else. Everything we do should...

I think in this example the optimization I'd hope to see here would be to use the Y part of the indirect indexed addressing mode, even without unrolling the loop....

This is actually pretty difficult, due to how libcalls work. They're C function calls generated at a very low level in LLVM, below the C calling convention handling in Clang....

Closing as obsolete; this could only be done with great difficulty, and 64-bit integers just aren't that important.