[Mesa-dev] [PATCH v2 2/3] nir: handle double-precision in fsign, fsat, fnot and frcp

[Matt Turner]

On Fri, May 20, 2016 at 12:55 AM, Iago Toral Quiroga <itoral at igalia.com> wrote:
> I think these are not strictly necessary since the floats in them
> should be automatically promoted to doubles when operated with
> double sources, but it makes things more explicit at least.
> ---

I would be somewhat worried that without this, some non-zero
double-precision values would be rounded to 0.0f when converted to a
float.

Reviewed-by: Matt Turner <mattst88 at gmail.com>