[Libclc-dev] [PATCH 1/1] amdgcn/fmin: Fix typos that reduced precision

[Aaron Watry via Libclc-dev]

Yeah, that was obviously wrong...

Reviewed-by: Aaron Watry <awatry at gmail.com>

On Thu, 2018-04-12 at 22:12 -0400, Jan Vesely via Libclc-dev wrote:
> Not sure how these sneaked in.
> Fixes fminD and few other tests(fractD, cosD) on carrizo
> Signed-off-by: Jan Vesely <jan.vesely at rutgers.edu>
> ---
>  amdgcn/lib/math/fmin.cl | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/amdgcn/lib/math/fmin.cl b/amdgcn/lib/math/fmin.cl
> index 910f68c..35dea8b 100644
> --- a/amdgcn/lib/math/fmin.cl
> +++ b/amdgcn/lib/math/fmin.cl
> @@ -19,9 +19,9 @@ _CLC_BINARY_VECTORIZE(_CLC_OVERLOAD _CLC_DEF,
> float, fmin, float, float)
>  
>  _CLC_DEF _CLC_OVERLOAD double fmin(double x, double y)
>  {
> -   x = __builtin_canonicalizef(x);
> -   y = __builtin_canonicalizef(y);
> -   return __builtin_fminf(x, y);
> +   x = __builtin_canonicalize(x);
> +   y = __builtin_canonicalize(y);
> +   return __builtin_fmin(x, y);
>  }
>  _CLC_BINARY_VECTORIZE(_CLC_OVERLOAD _CLC_DEF, double, fmin, double,
> double)
>