nak/nir: Use correct rounding for fp64 -> fp16 conversions

For up, down, and round towards zero, the rounding accumulates properly
as long as you use the same rounding mode for both.  For RTNE, however,
we need to insert a two-instruction fixup in order to guarantee correct
rounding.

Reviewed-by: Mel Henning <mhenning@darkrefraction.com>
Reviewed-by: Benjamin Lee <benjamin.lee@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34126>
1 file changed