[Mesa-dev] [PATCH 7/8] st/mesa: explicitly handle/list 64bit ops in visit_expression()

[Nicolai Hähnle]

On 26.01.2017 14:18, Emil Velikov wrote:
> From: Emil Velikov <emil.velikov at collabora.com>
>
> Follow the approach set in the file and handle all the ops, as otherwise
> the compiler throws a bunch of lovely warnings.
>
> Note that some versions of GCC have -Wswitch implied by -Wall, latter of
> which set in out autoconf and scons builds.
>
> Signed-off-by: Emil Velikov <emil.velikov at collabora.com>

This patch is likely to clash with the 64-bit enablement patch that I'm 
about to land.

Patches 4-6:

Reviewed-by: Nicolai Hähnle <nicolai.haehnle at amd.com>

> ---
>  src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 27 +++++++++++++++++++++++++++
>  1 file changed, 27 insertions(+)
>
> diff --git a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp b/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
> index a437645d9e..b154fad185 100644
> --- a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
> +++ b/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
> @@ -2319,6 +2319,33 @@ glsl_to_tgsi_visitor::visit_expression(ir_expression* ir, st_src_reg *op)
>     case ir_binop_carry:
>     case ir_binop_borrow:
>     case ir_unop_ssbo_unsized_array_length:
> +   case ir_unop_bitcast_i642d:
> +   case ir_unop_bitcast_d2u64:
> +   case ir_unop_bitcast_d2i64:
> +   case ir_unop_i642i:
> +   case ir_unop_u642i:
> +   case ir_unop_i642u:
> +   case ir_unop_u642u:
> +   case ir_unop_i642b:
> +   case ir_unop_i642f:
> +   case ir_unop_u642f:
> +   case ir_unop_i642d:
> +   case ir_unop_u642d:
> +   case ir_unop_i2i64:
> +   case ir_unop_u2i64:
> +   case ir_unop_b2i64:
> +   case ir_unop_f2i64:
> +   case ir_unop_d2i64:
> +   case ir_unop_i2u64:
> +   case ir_unop_u2u64:
> +   case ir_unop_f2u64:
> +   case ir_unop_d2u64:
> +   case ir_unop_u642i64:
> +   case ir_unop_i642u64:
> +   case ir_unop_pack_int_2x32:
> +   case ir_unop_pack_uint_2x32:
> +   case ir_unop_unpack_int_2x32:
> +   case ir_unop_unpack_uint_2x32:
>        /* This operation is not supported, or should have already been handled.
>         */
>        assert(!"Invalid ir opcode in glsl_to_tgsi_visitor::visit()");
>