[Libre-soc-isa] [Bug 933] prefix-code (like huffman code) decode/encode instructions

[bugzilla-daemon at libre-soc.org]

https://bugs.libre-soc.org/show_bug.cgi?id=933

--- Comment #3 from Jacob Lifshay <programmerjake at gmail.com> ---
I filled out the prefix-code decode instruction description:
https://libre-soc.org/openpower/prefix_codes/

(In reply to Luke Kenneth Casson Leighton from comment #2)
> ooo, i like it, and it's exactly the kind of fascinating thing
> that "emerges" like this.

:)

> 1. practical detail: 4-operand is barely enough room to fit, takes up
>    26 bits on its own, is a 5/6-bit XO. if the imm is 1 bit that's 2of
>    5-bit XOs. costly! has to be *really* worth it which...

imm is 1 bit. that's *1* 5-bit XO, since imm fits where Rc would be. If Rc=1 is
useful (probably, but I didn't work that out yet), then imho we just have Rc
always be set, so it doesn't need to be encoded.
> 
> 2. ... are there any other uses? other common compression algorithms?

added some more:
https://libre-soc.org/openpower/prefix_codes/

-- 
You are receiving this mail because:
You are on the CC list for the bug.