[Mesa-dev] [PATCH 06/95] i965/vec4/nir: fix nir_intrinsic_load_uniform for doubles

[Timothy Arceri]

On Tue, 2016-07-19 at 12:40 +0200, Iago Toral Quiroga wrote:
> From: Connor Abbott <connor.w.abbott at intel.com>
> 
> ---
>  src/mesa/drivers/dri/i965/brw_vec4_nir.cpp | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/src/mesa/drivers/dri/i965/brw_vec4_nir.cpp
> b/src/mesa/drivers/dri/i965/brw_vec4_nir.cpp
> index 5a7ee0b..df927e7 100644
> --- a/src/mesa/drivers/dri/i965/brw_vec4_nir.cpp
> +++ b/src/mesa/drivers/dri/i965/brw_vec4_nir.cpp
> @@ -712,6 +712,7 @@
> vec4_visitor::nir_emit_intrinsic(nir_intrinsic_instr *instr)
>           src.reg_offset = offset / 16;
>           shift = (offset % 16) / 4;
>           src.swizzle += BRW_SWIZZLE4(shift, shift, shift, shift);
> +         src.type = dest.type;

This seems to already be set above:

      src = src_reg(dst_reg(UNIFORM, nir_intrinsic_base(instr) / 16));
      src.type = dest.type;

I'm I missing something?

>  
>           emit(MOV(dest, src));
>        } else {