#885 R800 (Z80 variant) simulator

[Philipp Klaus Krause]

To be able to add an r800 port to SDCC, we want uCsim to support the R800.
The R800

• Supports the official Z80 instructions
• Supports the iyl/iyh/ixl/ixh instructions (unofficial on Z80, official on eZ80 in Z80 mode)
• Supports a few of the Z280 multiplication instructions (those at opcodes ed c3, ed f3, ed c1, ed c9, ed d1, ed d8).
• Executes instruction in fewer cycles than Z80.

Instruction set documentation (in Japanese, but still useable without knowing Japanese) can be found at https://archive.org/details/MsxTurboR/page/n173/mode/2up
Personally, I think we should just use Z280 menmonics, instead of what that documentation uses, though. Also, Z280 documentation (see Z280 e.g. preliminary technical manual from https://oldcomputers.dyndns.org/public/pub/rechner/zilog/z280/manual/index.html) is more detailed wrt. flags on the multiplication instructions. So we could just use the Japanese docs for the R800 cycle counts, and Z280 documentation for everything else.

P.S.: I'm creating an r800 branch for this work now.

[Philipp Klaus Krause]

There are two more difference between the Z80 and the R800 (though these are unlikely to ever be used by SDCC):

• The undocumented (for both Z80 and R800) sll instructions behave differently: On the R800, they behave just like the official sla instructions.
• Undocumented (for both Z80 and R800) bits 3 and 5 in the flag register behave differently.

[Daniel Drotos]

https://oldcomputers.dyndns.org/public/pub/rechner/zilog/z280/manual/index.html is not available, is there any other documents?

[Philipp Klaus Krause]

The link https://oldcomputers.dyndns.org/public/pub/rechner/zilog/z280/manual/index.html works for me (like it did an hour ago).

An alternative (but not OCR'd) is https://web.archive.org/web/20190911051642/http://www.classiccmp.org/hp/zilog/z280_manual.pdf

P.S.: Maybe this direct link to the pdf works for you:
https://oldcomputers.dyndns.org/public/pub/rechner/zilog/z280/manual/z280_mpu_(ocr_bw_600).pdf

[Philipp Klaus Krause]

In the r800 branch, today I implemented (very basic, but sufficient to pass our regression tests) support for multu and multuw.

[Philipp Klaus Krause]

The r800 branch has been merged to trunk yesterday.