[Libre-soc-isa] [Bug 1057] split out all int/fp min/max ops into their own RFC ls013

bugzilla-daemon at libre-soc.org bugzilla-daemon at libre-soc.org
Wed Apr 26 02:52:52 BST 2023


https://bugs.libre-soc.org/show_bug.cgi?id=1057

--- Comment #28 from Jacob Lifshay <programmerjake at gmail.com> ---
I realized that fminmaxs[.] is redundant since it always produces the exact
same results as fminmax[.] (for scalar at least), should we just remove it?

It produces the exact same results because min/max is independent of
precision/range and has no round off error, it behaves identically for
f32/f64/f32-in-f64/etc. the only part that varies is which bits it reads from
and sets for quieting sNaNs, and the changes caused by SINGLE/DOUBLE (only
relevant for not-really-f32 in f32-in-f64), which is only relevant for f32, not
f64/f32-in-f64.

Luke, what do you think?

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


More information about the Libre-SOC-ISA mailing list