[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 19 20:05:20 BST 2023


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

--- Comment #10 from Jacob Lifshay <programmerjake at gmail.com> ---
(In reply to Jacob Lifshay from comment #9)
> (In reply to Luke Kenneth Casson Leighton from comment #8)
> > (In reply to Jacob Lifshay from comment #7)
> > > while we're changing stuff, adding min[u]w/max[u]w would be useful, since
> > > lots of programs often use 32-bit types, and PowerISA has *w variants where
> > > they'd be different than the lower 32-bits of 64-bit variants.
> > 
> > hmmm.... what other ISAs have them? i'm getting concerned about the number
> > of *w variants.  are those present in VSX?
> 
> arm has smax on 32/64-bit scalar registers and presumably also
> umax/smin/umin.
> arm also has atomic fetch-and-min/max
> arm and Power also have unsigned/signed simd min/max. VSX has no scalar
> integer min/max.

x86 has fp/int simd min/max but only has fp scalar min/max.

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


More information about the Libre-SOC-ISA mailing list