[Mesa-dev] [PATCH v2 45/52] nir: Make ballot intrinsics variable-size

[Lionel Landwerlin]

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin at intel.com>

On 13/10/17 06:48, Jason Ekstrand wrote:
> This way they can return either a uvec4 or a uint64_t.  At the moment,
> this is a no-op since we still always return a uint64_t.
> ---
>   src/compiler/glsl/glsl_to_nir.cpp          |  1 +
>   src/compiler/nir/nir_intrinsics.h          | 12 ++++++------
>   src/compiler/nir/nir_lower_system_values.c |  1 +
>   3 files changed, 8 insertions(+), 6 deletions(-)
>
> diff --git a/src/compiler/glsl/glsl_to_nir.cpp b/src/compiler/glsl/glsl_to_nir.cpp
> index 5e9544f..6110aa9 100644
> --- a/src/compiler/glsl/glsl_to_nir.cpp
> +++ b/src/compiler/glsl/glsl_to_nir.cpp
> @@ -1165,6 +1165,7 @@ nir_visitor::visit(ir_call *ir)
>         case nir_intrinsic_ballot: {
>            nir_ssa_dest_init(&instr->instr, &instr->dest,
>                              ir->return_deref->type->vector_elements, 64, NULL);
> +         instr->num_components = ir->return_deref->type->vector_elements;
>   
>            ir_rvalue *value = (ir_rvalue *) ir->actual_parameters.get_head();
>            instr->src[0] = nir_src_for_ssa(evaluate_rvalue(value));
> diff --git a/src/compiler/nir/nir_intrinsics.h b/src/compiler/nir/nir_intrinsics.h
> index 54a51f8..c346c0e 100644
> --- a/src/compiler/nir/nir_intrinsics.h
> +++ b/src/compiler/nir/nir_intrinsics.h
> @@ -102,7 +102,7 @@ INTRINSIC(shader_clock, 0, ARR(0), true, 2, 0, 0, xx, xx, xx, NIR_INTRINSIC_CAN_
>    *
>    * GLSL functions from ARB_shader_ballot.
>    */
> -INTRINSIC(ballot, 1, ARR(1), true, 1, 0, 0, xx, xx, xx, NIR_INTRINSIC_CAN_ELIMINATE)
> +INTRINSIC(ballot, 1, ARR(1), true, 0, 0, 0, xx, xx, xx, NIR_INTRINSIC_CAN_ELIMINATE)
>   INTRINSIC(read_invocation, 2, ARR(0, 1), true, 0, 0, 0, xx, xx, xx, NIR_INTRINSIC_CAN_ELIMINATE)
>   INTRINSIC(read_first_invocation, 1, ARR(0), true, 0, 0, 0, xx, xx, xx, NIR_INTRINSIC_CAN_ELIMINATE)
>   
> @@ -350,11 +350,11 @@ SYSTEM_VALUE(layer_id, 1, 0, xx, xx, xx)
>   SYSTEM_VALUE(view_index, 1, 0, xx, xx, xx)
>   SYSTEM_VALUE(subgroup_size, 1, 0, xx, xx, xx)
>   SYSTEM_VALUE(subgroup_invocation, 1, 0, xx, xx, xx)
> -SYSTEM_VALUE(subgroup_eq_mask, 1, 0, xx, xx, xx)
> -SYSTEM_VALUE(subgroup_ge_mask, 1, 0, xx, xx, xx)
> -SYSTEM_VALUE(subgroup_gt_mask, 1, 0, xx, xx, xx)
> -SYSTEM_VALUE(subgroup_le_mask, 1, 0, xx, xx, xx)
> -SYSTEM_VALUE(subgroup_lt_mask, 1, 0, xx, xx, xx)
> +SYSTEM_VALUE(subgroup_eq_mask, 0, 0, xx, xx, xx)
> +SYSTEM_VALUE(subgroup_ge_mask, 0, 0, xx, xx, xx)
> +SYSTEM_VALUE(subgroup_gt_mask, 0, 0, xx, xx, xx)
> +SYSTEM_VALUE(subgroup_le_mask, 0, 0, xx, xx, xx)
> +SYSTEM_VALUE(subgroup_lt_mask, 0, 0, xx, xx, xx)
>   SYSTEM_VALUE(subgroup_id, 1, 0, xx, xx, xx)
>   
>   /* Blend constant color values.  Float values are clamped. */
> diff --git a/src/compiler/nir/nir_lower_system_values.c b/src/compiler/nir/nir_lower_system_values.c
> index ba20d30..c21a468 100644
> --- a/src/compiler/nir/nir_lower_system_values.c
> +++ b/src/compiler/nir/nir_lower_system_values.c
> @@ -125,6 +125,7 @@ convert_block(nir_block *block, nir_builder *b)
>               nir_intrinsic_from_system_value(var->data.location);
>            nir_intrinsic_instr *load = nir_intrinsic_instr_create(b->shader, op);
>            nir_ssa_dest_init(&load->instr, &load->dest, 1, 64, NULL);
> +         load->num_components = 1;
>            nir_builder_instr_insert(b, &load->instr);
>            sysval = &load->dest.ssa;
>            break;