[Libre-soc-bugs] [Bug 230] Video opcode development and discussion

[bugzilla-daemon at libre-soc.org]

https://bugs.libre-soc.org/show_bug.cgi?id=230

--- Comment #67 from Luke Kenneth Casson Leighton <lkcl at lkcl.net> ---
lauri i found the comment about IEEE754 saturate as an option.  yes it is,
however there does actually already exist an FP SPR flag dedicated to "less
accuracy".


Float estimates

vec_expte - float 2^x
vec_loge - float log2(x)
vec_re - float 1/x
vec_rsqrte - float 1/sqrt(x)
The spec says the max relative inaccuracy is 1/4096.

In conjunction with the FPSPR "accuracy" bit These could be done by assigning
meaning to the "sat mode" SVP64 bits in a FP context. 0b00 is IEEE754 FP, 0b01
is 212 accuracy for FP32. These can be applied to standard scalar FP ops

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