i965: Don't use nir_assign_var_locations for VS/TES/GS outputs.

Fixes spec/arb_enhanced_layouts/execution/component-layout/vs-fs-array-dvec3.

v2: Remove nir_outputs field from fs_visitor (caught by Tim and Iago).

Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>

diff --git a/src/mesa/drivers/dri/i965/brw_fs.cpp b/src/mesa/drivers/dri/i965/brw_fs.cpp
index 7738a47550798d81d282ae6cf73023a32aed44af..921cc00a03eedeb552d0726c3a476b56954f139c 100644 (file)
--- a/src/mesa/drivers/dri/i965/brw_fs.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs.cpp
@@ -492,19 +492,6 @@ type_size_scalar(const struct glsl_type *type)
    return 0;
 }
 
-/**
- * Returns the number of scalar components needed to store type, assuming
- * that vectors are padded out to vec4.
- *
- * This has the packing rules of type_size_vec4(), but counts components
- * similar to type_size_scalar().
- */
-extern "C" int
-type_size_vec4_times_4(const struct glsl_type *type)
-{
-   return 4 * type_size_vec4(type);
-}
-
 /* Attribute arrays are loaded as one vec4 per element (or matrix column),
  * except for double-precision types, which are loaded as one dvec4.
  */

diff --git a/src/mesa/drivers/dri/i965/brw_fs.h b/src/mesa/drivers/dri/i965/brw_fs.h
index 663b3eac5deccceadfa991751597dc79556e1320..aad21602a4c7193b06fa35386fcb17f0d24407f2 100644 (file)
--- a/src/mesa/drivers/dri/i965/brw_fs.h
+++ b/src/mesa/drivers/dri/i965/brw_fs.h
@@ -325,7 +325,6 @@ public:
    fs_reg *nir_locals;
    fs_reg *nir_ssa_values;
    fs_reg nir_inputs;
-   fs_reg nir_outputs;
    fs_reg *nir_system_values;
 
    bool failed;

diff --git a/src/mesa/drivers/dri/i965/brw_fs_nir.cpp b/src/mesa/drivers/dri/i965/brw_fs_nir.cpp
index 9cad1a6d29573482aeb3aeefab84daa6533e614e..4baadc9587a2041ba8c2bcd917f58701646133c7 100644 (file)
--- a/src/mesa/drivers/dri/i965/brw_fs_nir.cpp
+++ b/src/mesa/drivers/dri/i965/brw_fs_nir.cpp
@@ -48,48 +48,19 @@ fs_visitor::emit_nir_code()
    }
 }
 
-void
-fs_visitor::nir_setup_single_output_varying(fs_reg *reg,
-                                            const glsl_type *type,
-                                            unsigned *location)
-{
-   if (type->is_array() || type->is_matrix()) {
-      const struct glsl_type *elem_type = glsl_get_array_element(type);
-      const unsigned length = glsl_get_length(type);
-
-      for (unsigned i = 0; i < length; i++) {
-         nir_setup_single_output_varying(reg, elem_type, location);
-      }
-   } else if (type->is_record()) {
-      for (unsigned i = 0; i < type->length; i++) {
-         const struct glsl_type *field_type = type->fields.structure[i].type;
-         nir_setup_single_output_varying(reg, field_type, location);
-      }
-   } else {
-      assert(type->is_scalar() || type->is_vector());
-      unsigned num_iter = 1;
-      if (type->is_dual_slot())
-         num_iter = 2;
-      for (unsigned count = 0; count < num_iter; count++) {
-         this->outputs[*location] = *reg;
-         *reg = offset(*reg, bld, 4);
-         (*location)++;
-      }
-   }
-}
-
 void
 fs_visitor::nir_setup_outputs()
 {
    if (stage == MESA_SHADER_TESS_CTRL || stage == MESA_SHADER_FRAGMENT)
       return;
 
-   nir_outputs = bld.vgrf(BRW_REGISTER_TYPE_F, nir->num_outputs);
-
    nir_foreach_variable(var, &nir->outputs) {
-      fs_reg reg = offset(nir_outputs, bld, var->data.driver_location);
-      unsigned location = var->data.location;
-      nir_setup_single_output_varying(&reg, var->type, &location);
+      const unsigned vec4s = type_size_vec4(var->type);
+      fs_reg reg = bld.vgrf(BRW_REGISTER_TYPE_F, 4 * vec4s);
+      for (unsigned i = 0; i < vec4s; i++) {
+         if (outputs[var->data.driver_location + i].file == BAD_FILE)
+            outputs[var->data.driver_location + i] = offset(reg, bld, 4 * i);
+      }
    }
 }
 
@@ -4242,12 +4213,11 @@ fs_visitor::nir_emit_intrinsic(const fs_builder &bld, nir_intrinsic_instr *instr
 
    case nir_intrinsic_store_output: {
       fs_reg src = get_nir_src(instr->src[0]);
-      fs_reg new_dest = offset(retype(nir_outputs, src.type), bld,
-                               instr->const_index[0]);
 
       nir_const_value *const_offset = nir_src_as_const_value(instr->src[1]);
       assert(const_offset && "Indirect output stores not allowed");
-      new_dest = offset(new_dest, bld, const_offset->u32[0]);
+      fs_reg new_dest = retype(offset(outputs[instr->const_index[0]], bld,
+                                      4 * const_offset->u32[0]), src.type);
 
       unsigned num_components = instr->num_components;
       unsigned first_component = nir_intrinsic_component(instr);

diff --git a/src/mesa/drivers/dri/i965/brw_nir.c b/src/mesa/drivers/dri/i965/brw_nir.c
index a935f42d8e27a28fbaed0f6862dd3d828e6c9ede..de0e2353418c7c4ff8048da165defff5e3d91f68 100644 (file)
--- a/src/mesa/drivers/dri/i965/brw_nir.c
+++ b/src/mesa/drivers/dri/i965/brw_nir.c
@@ -327,16 +327,11 @@ void
 brw_nir_lower_vue_outputs(nir_shader *nir,
                           bool is_scalar)
 {
-   if (is_scalar) {
-      nir_assign_var_locations(&nir->outputs, &nir->num_outputs,
-                               VARYING_SLOT_VAR0,
-                               type_size_vec4_times_4);
-      nir_lower_io(nir, nir_var_shader_out, type_size_vec4_times_4, 0);
-   } else {
-      nir_foreach_variable(var, &nir->outputs)
-         var->data.driver_location = var->data.location;
-      nir_lower_io(nir, nir_var_shader_out, type_size_vec4, 0);
+   nir_foreach_variable(var, &nir->outputs) {
+      var->data.driver_location = var->data.location;
    }
+
+   nir_lower_io(nir, nir_var_shader_out, type_size_vec4, 0);
 }
 
 void

diff --git a/src/mesa/drivers/dri/i965/brw_nir.h b/src/mesa/drivers/dri/i965/brw_nir.h
index aef5c532336a86f458ac3eebae912ceccc87986c..645a818c4e500de570644dffba991607ad0080d0 100644 (file)
--- a/src/mesa/drivers/dri/i965/brw_nir.h
+++ b/src/mesa/drivers/dri/i965/brw_nir.h
@@ -34,7 +34,6 @@ extern "C" {
 int type_size_scalar(const struct glsl_type *type);
 int type_size_vec4(const struct glsl_type *type);
 int type_size_dvec4(const struct glsl_type *type);
-int type_size_vec4_times_4(const struct glsl_type *type);
 int type_size_vs_input(const struct glsl_type *type);
 
 static inline int