Errors in Spin Byte Code wiki page?

johnfos

I think there are a couple of minor errors in the wiki page for Spin byte code: http://propeller.wikispaces.com/Spin+Byte+Code

I'm willing to edit the page if any expert out there can reassure me that I won't be perpetrating an error instead of fixing one.

1. In the big table, opcodes 8c to 8f are listed as PUSH_INDEXED_LOCALMEM_BYTE, POP_(ditto), EFFECT_(ditto) and REFERENCE_(ditto). I think these four should have the _INDEXED deleted.

2. Earlier, under the heading "Structure of UPPER opcodes" and under the sub-heading "Structure of Memory ops", the description of S-field contains the sentence "Used only if I-field is 1." I think that sentence should be deleted.

Anyone care to confirm or dispute?

Dave Hein

Item #1 is an error, and it should be corrected. Item #2 is trying to say that the index is adjusted by the access size. The index is multiplied times 4 for a long, and times 2 for a word. This sentence should remain, but it should be clarified a bit.

johnfos

Thanks Dave for that.

On #2, I'm not quite convinced. As I see it, the SS field has two effects. It specifies the size of the memory location to be accessed, and if the I field is set it also determines the amount by which the index is shifted. The first effect applies to all the memop instructions, which is why I thought the sentence should be deleted.

I'd be happy to replace it with a clarification, maybe:

S-field: Size of address operand. 00: Byte; 01: Word; 10: Long; 11: illegal (would be a math op). If the I-field is set to 1, the index value is shifted left by S bits.

Dave Hein

Your clarification looks good.

johnfos

Dave, thanks for the reassurance and the suggestion. I've now edited the page.

Cluso99

If you follow the links in my signature and look at the thread 'faster spin interpreter' (or something like that), I have a detailed description of the interpreter bytes. This is taken from all the known data at the time including Chip's listing, and includes my own interpretation from traces, etc. This may help.

johnfos

Thanks Cluso - checking that now.