[Mesa-dev] [PATCH] glsl_to_tgsi: try harder to lower unsupported ir_binop_vector_extract

[Nicolai Hähnle]

On 16.04.2018 23:02, Marek Olšák wrote:
> From: Marek Olšák <marek.olsak at amd.com>
> 
> Cc: 18.0 <mesa-stable at lists.freedesktop.org>

Would be nice to have an indication of what this fixes. Apart from that 
it looks reasonable.

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


> ---
>   src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 5 +++++
>   1 file changed, 5 insertions(+)
> 
> diff --git a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp b/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
> index 01a58d7d2c6..b321112cf88 100644
> --- a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
> +++ b/src/mesa/state_tracker/st_glsl_to_tgsi.cpp
> @@ -7117,20 +7117,25 @@ st_link_shader(struct gl_context *ctx, struct gl_shader_program *prog)
>            /* Repeat it until it stops making changes. */
>            bool progress;
>            do {
>               progress = do_common_optimization(ir, true, true, options,
>                                                 ctx->Const.NativeIntegers);
>               progress |= lower_if_to_cond_assign((gl_shader_stage)i, ir,
>                                                   options->MaxIfDepth, if_threshold);
>            } while (progress);
>         }
>   
> +      /* Do this again to lower ir_binop_vector_extract introduced
> +       * by optimization passes.
> +       */
> +      do_vec_index_to_cond_assign(ir);
> +
>         validate_ir_tree(ir);
>      }
>   
>      build_program_resource_list(ctx, prog);
>   
>      if (use_nir)
>         return st_link_nir(ctx, prog);
>   
>      for (unsigned i = 0; i < MESA_SHADER_STAGES; i++) {
>         struct gl_linked_shader *shader = prog->_LinkedShaders[i];
> 


-- 
Lerne, wie die Welt wirklich ist,
Aber vergiss niemals, wie sie sein sollte.