Add fmaximum, fminimum functions [committed]

[Joseph Myers]

On Mon, 4 Oct 2021, Palmer Dabbelt via Libc-alpha wrote:

> > Add these functions to glibc.  The implementations use type-generic
> > templates based on those for fmax, fmin, fmaxmag and fminmag, and test
> > inputs are based on those for those functions with appropriate
> > adjustments to the expected results.  The RISC-V maintainers might
> > wish to add optimized versions of fmaximum_num and fminimum_num (for
> > float and double), since RISC-V (F extension version 2.2 and later)
> 
> Thanks for pointing this out.
> 
> I don't actually remember if we had any pre-2.2 hardware that could run glibc:
> pretty sure the HiFive Unleashed is post-2.2, and that's the only thing I can
> think of that would be that old.  I'm never quite sure about the dates on
> specs, but I'm seeing May 2017 for 2.2 and Feb 2018 for the HiFive Unleashed
> so I think it's likely fine.  Andrew would know for sure, though.

Note that this is F (and D) extension version 2.2 rather than architecture 
version 2.2 (architecture version 2.2 has F extension 2.0, where FMAX and 
FMIN still referred to the IEEE 754-2008 operations, i.e. fmax and fmin 
functions rather than fmaximum_num and fminimum_num).

-- 
Joseph S. Myers
joseph@codesourcery.com