[Mesa-dev] [PATCH 3/3] st/mesa: use PK2H/UP2H when gs5 is supported

[Ilia Mirkin]

On Thu, Oct 29, 2015 at 2:52 AM, Ilia Mirkin <imirkin at alum.mit.edu> wrote:
> ARB_gpu_shader5 should be a decent proxy for whether those instructions
> are supported.
>
> Signed-off-by: Ilia Mirkin <imirkin at alum.mit.edu>
> ---
>
> Not actually ready to be committed, since r600/radeonsi need support
> for these ops as well. Should be easy to implement, but want to get
> the OK before I do so.

Oh, and to pre-empt the question of "why not the other pack/unpack
primitives", it's because nouveau codegen is able to optimize them
into reasonable forms (it does better with unpack than with pack), esp
after I adjusted the lowering passes to use BFI/BFE. Not ideal, but
the manual f32 <-> f16 stuff was a bit much. I wouldn't be opposed to
the other opcodes getting added too, just didn't seem too important.

>
>  src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 16 +++++++++++-----
>  1 file changed, 11 insertions(+), 5 deletions(-)
>
> diff --git a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp b/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
> index f481e89..fff18a4 100644
> --- a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
> +++ b/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
> @@ -2195,15 +2195,20 @@ glsl_to_tgsi_visitor::visit(ir_expression *ir)
>        }
>        break;
>
> +   case ir_unop_pack_half_2x16:
> +      emit_asm(ir, TGSI_OPCODE_PK2H, result_dst, op[0]);
> +      break;
> +   case ir_unop_unpack_half_2x16:
> +      emit_asm(ir, TGSI_OPCODE_UP2H, result_dst, op[0]);
> +      break;
> +
>     case ir_unop_pack_snorm_2x16:
>     case ir_unop_pack_unorm_2x16:
> -   case ir_unop_pack_half_2x16:
>     case ir_unop_pack_snorm_4x8:
>     case ir_unop_pack_unorm_4x8:
>
>     case ir_unop_unpack_snorm_2x16:
>     case ir_unop_unpack_unorm_2x16:
> -   case ir_unop_unpack_half_2x16:
>     case ir_unop_unpack_half_2x16_split_x:
>     case ir_unop_unpack_half_2x16_split_y:
>     case ir_unop_unpack_snorm_4x8:
> @@ -5794,13 +5799,14 @@ st_link_shader(struct gl_context *ctx, struct gl_shader_program *prog)
>                                 LOWER_PACK_SNORM_4x8 |
>                                 LOWER_UNPACK_SNORM_4x8 |
>                                 LOWER_UNPACK_UNORM_4x8 |
> -                               LOWER_PACK_UNORM_4x8 |
> -                               LOWER_PACK_HALF_2x16 |
> -                               LOWER_UNPACK_HALF_2x16;
> +                               LOWER_PACK_UNORM_4x8;
>
>           if (ctx->Extensions.ARB_gpu_shader5)
>              lower_inst |= LOWER_PACK_USE_BFI |
>                            LOWER_PACK_USE_BFE;
> +         else
> +            lower_inst |= LOWER_PACK_HALF_2x16 |
> +                          LOWER_UNPACK_HALF_2x16;
>
>           lower_packing_builtins(ir, lower_inst);
>        }
> --
> 2.4.10
>