search

mewmew / float

Binary floating-point formats in Go (IEEE 754 half and quadruple precision, x86 extended precision and PowerPC quadruple precision with double-double arithmetic).
The Unlicense
24 stars 3 forks source link

Implement `ppc_fp128`: 128-bit floating point type (two 64-bits, PowerPC) #11

Open mewmew opened 6 years ago

mewmew commented 6 years ago

Close this issue when float128ppc is feature complete and tested.

pwaller commented 4 years ago

Fixed in #18? @dannypsnl @mewmew

mewmew commented 4 years ago

Fixed in #18? @dannypsnl @mewmew

Not yet, we'd first need test cases :)

dannypsnl commented 2 years ago

@mewmew https://gcc.gnu.org/wiki/Ieee128PowerPC I'm just thinking maybe we generate test data to check our implementation. And a reference is https://juliamath.github.io/DoubleFloats.jl/stable/characteristics/ I think we can look into implementation

mewmew commented 2 years ago

@mewmew https://gcc.gnu.org/wiki/Ieee128PowerPC I'm just thinking maybe we generate test data to check our implementation. And a reference is https://juliamath.github.io/DoubleFloats.jl/stable/characteristics/ I think we can look into implementation

Sounds good.