vkd3d-shader/dxil: Implement the DXIL EXTRACTVAL instruction.

Approved-by: Giovanni Mascellani (@giomasce)
Approved-by: Henri Verbeet (@hverbeet)
Approved-by: Alexandre Julliard (@julliard)
Merge-Request: https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/408

libs/vkd3d-shader/dxil.c

@@ -1483,6 +1483,11 @@ static inline bool sm6_type_is_floating_point(const struct sm6_type *type)
     return type->class == TYPE_CLASS_FLOAT;
 }
 
+static bool sm6_type_is_scalar(const struct sm6_type *type)
+{
+    return type->class == TYPE_CLASS_INTEGER || type->class == TYPE_CLASS_FLOAT || type->class == TYPE_CLASS_POINTER;
+}
+
 static inline bool sm6_type_is_numeric(const struct sm6_type *type)
 {
     return type->class == TYPE_CLASS_INTEGER || type->class == TYPE_CLASS_FLOAT;
@@ -1493,6 +1498,11 @@ static inline bool sm6_type_is_pointer(const struct sm6_type *type)
     return type->class == TYPE_CLASS_POINTER;
 }
 
+static bool sm6_type_is_aggregate(const struct sm6_type *type)
+{
+    return type->class == TYPE_CLASS_STRUCT || type->class == TYPE_CLASS_VECTOR || type->class == TYPE_CLASS_ARRAY;
+}
+
 static bool sm6_type_is_numeric_aggregate(const struct sm6_type *type)
 {
     unsigned int i;
@@ -1563,6 +1573,27 @@ static const struct sm6_type *sm6_type_get_pointer_to_type(const struct sm6_type
     return NULL;
 }
 
+/* Call for aggregate types only. */
+static const struct sm6_type *sm6_type_get_element_type_at_index(const struct sm6_type *type, uint64_t elem_idx)
+{
+    switch (type->class)
+    {
+        case TYPE_CLASS_ARRAY:
+        case TYPE_CLASS_VECTOR:
+            if (elem_idx >= type->u.array.count)
+                return NULL;
+            return type->u.array.elem_type;
+
+        case TYPE_CLASS_STRUCT:
+            if (elem_idx >= type->u.struc->elem_count)
+                return NULL;
+            return type->u.struc->elem_types[elem_idx];
+
+        default:
+            vkd3d_unreachable();
+    }
+}
+
 /* Never returns null for elem_idx 0. */
 static const struct sm6_type *sm6_type_get_scalar_type(const struct sm6_type *type, unsigned int elem_idx)
 {
@@ -2994,6 +3025,64 @@ static void sm6_parser_emit_call(struct sm6_parser *sm6, const struct dxil_recor
             fn_value->u.function.name, &operands[1], operand_count - 1, ins, dst);
 }
 
+static void sm6_parser_emit_extractval(struct sm6_parser *sm6, const struct dxil_record *record,
+        struct vkd3d_shader_instruction *ins, struct sm6_value *dst)
+{
+    struct vkd3d_shader_src_param *src_param;
+    const struct sm6_type *type;
+    const struct sm6_value *src;
+    unsigned int i = 0;
+    uint64_t elem_idx;
+
+    if (!(src = sm6_parser_get_value_by_ref(sm6, record, NULL, &i)))
+        return;
+
+    if (!dxil_record_validate_operand_min_count(record, i + 1, sm6))
+        return;
+
+    if (record->operand_count > i + 1)
+    {
+        FIXME("Unhandled multiple indices.\n");
+        vkd3d_shader_parser_error(&sm6->p, VKD3D_SHADER_ERROR_DXIL_INVALID_OPERAND,
+                "Multiple extractval indices are not supported.");
+        return;
+    }
+
+    type = src->type;
+    if (!sm6_type_is_aggregate(type))
+    {
+        WARN("Invalid extraction from non-aggregate.\n");
+        vkd3d_shader_parser_error(&sm6->p, VKD3D_SHADER_ERROR_DXIL_INVALID_OPERAND,
+                "Source type of an extractval instruction is not an aggregate.");
+        return;
+    }
+
+    elem_idx = record->operands[i];
+    if (!(type = sm6_type_get_element_type_at_index(type, elem_idx)))
+    {
+        WARN("Invalid element index %"PRIu64".\n", elem_idx);
+        vkd3d_shader_parser_error(&sm6->p, VKD3D_SHADER_ERROR_DXIL_INVALID_OPERAND,
+                "Element index %"PRIu64" for an extractval instruction is out of bounds.", elem_idx);
+        return;
+    }
+    if (!sm6_type_is_scalar(type))
+    {
+        FIXME("Nested extraction is not supported.\n");
+        vkd3d_shader_parser_error(&sm6->p, VKD3D_SHADER_ERROR_DXIL_INVALID_OPERAND,
+                "Extraction from nested aggregates is not supported.");
+        return;
+    }
+    dst->type = type;
+
+    ins->handler_idx = VKD3DSIH_MOV;
+
+    src_param = instruction_src_params_alloc(ins, 1, sm6);
+    src_param_init_from_value(src_param, src);
+    src_param->swizzle = vkd3d_shader_create_swizzle(elem_idx, elem_idx, elem_idx, elem_idx);
+
+    instruction_dst_param_init_ssa_scalar(ins, sm6);
+}
+
 static void sm6_parser_emit_ret(struct sm6_parser *sm6, const struct dxil_record *record,
         struct sm6_block *code_block, struct vkd3d_shader_instruction *ins)
 {
@@ -3149,6 +3238,9 @@ static enum vkd3d_result sm6_parser_function_init(struct sm6_parser *sm6, const
             case FUNC_CODE_INST_CALL:
                 sm6_parser_emit_call(sm6, record, code_block, ins, dst);
                 break;
+            case FUNC_CODE_INST_EXTRACTVAL:
+                sm6_parser_emit_extractval(sm6, record, ins, dst);
+                break;
             case FUNC_CODE_INST_RET:
                 sm6_parser_emit_ret(sm6, record, code_block, ins);
                 is_terminator = true;

tests/hlsl/asfloat.shader_test

@@ -36,7 +36,7 @@ float4 main() : sv_target
 uniform 0 float4 11 12 0 0
 uniform 4 float4 13 14 0 0
 uniform 8 float4 20 21 22 23
-todo(sm>=6) draw quad
+draw quad
 probe (320,240) rgba (13.0, 21.0, 0.0, 0.0)
 
 [pixel shader fail]

tests/hlsl/cbuffer.shader_test

@@ -13,7 +13,7 @@ float4 main() : sv_target
 
 [test]
 uniform 0 float4 1.0 2.0 3.0 4.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (1.0, 2.0, 3.0, 4.0)
 
 [pixel shader]
@@ -31,7 +31,7 @@ float4 main() : sv_target
 
 [test]
 uniform 0 float4 1.0 2.0 3.0 4.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (1.0, 2.0, 3.0, 4.0)
 
 % SM1 buffer offset allocation follows different rules than SM4.
@@ -72,7 +72,7 @@ uniform 0 float4 0.0 1.0 2.0 3.0
 uniform 4 float4 4.0 5.0 6.0 7.0
 uniform 8 float4 8.0 9.0 10.0 11.0
 uniform 12 float4 12.0 13.0 14.0 15.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.0, 2.0, 4.0, 8.0)
 
 
@@ -93,7 +93,7 @@ float4 main() : sv_target
 uniform 0 float4 0.0 1.0 2.0 3.0
 uniform 4 float4 4.0 5.0 6.0 7.0
 uniform 8 float4 8.0 9.0 10.0 11.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.0, 4.0, 8.0, 9.0)
 
 
@@ -119,7 +119,7 @@ uniform 0 float4 0.0 1.0 2.0 3.0
 uniform 4 float4 4.0 5.0 6.0 7.0
 uniform 8 float4 8.0 9.0 10.0 11.0
 uniform 12 float4 12.0 13.0 14.0 15.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.0, 4.0, 5.0, 6.0)
 
 
@@ -142,7 +142,7 @@ uniform 0 float4 1.0 0.0 0.0 0.0
 uniform 4 float4 2.0 0.0 0.0 0.0
 uniform 8 float4 3.0 0.0 0.0 0.0
 uniform 12 float4 4.0 0.0 0.0 0.0
-todo draw quad
+todo(sm<6) draw quad
 probe all rgba (1.0, 3.0, 3.0, 4.0)
 
 
@@ -221,7 +221,7 @@ float4 main() : sv_target
 
 [test]
 uniform 0 float4 1.0 2.0 3.0 4.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (2.0, 3.0, 2.0, 3.0)
 
 
@@ -357,7 +357,7 @@ float4 main() : sv_target
 
 [test]
 uniform 0 float4 1.0 2.0 3.0 4.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (3.0, 4.0, 3.0, 4.0)
 
 

tests/hlsl/expr-indexing.shader_test

@@ -40,7 +40,7 @@ float4 main() : sv_target
 
 [test]
 uniform 0 float4 1.0 2.0 3.0 4.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (3.0, 3.0, 3.0, 3.0)
 
 
@@ -78,7 +78,7 @@ float4 main() : sv_target
 
 [test]
 uniform 0 float4 1.0 2.0 3.0 4.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (4.0, 4.0, 4.0, 4.0)
 
 

tests/hlsl/half.shader_test

@@ -19,5 +19,5 @@ float4 main() : sv_target
 
 [test]
 uniform 0 float 10.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (10.0, 10.0, 10.0, 10.0)

tests/hlsl/majority-pragma.shader_test

@@ -40,7 +40,7 @@ float4 main() : sv_target
 [test]
 uniform  0 float4 0.1 0.2 0.0 0.0
 uniform  4 float4 0.3 0.4 0.0 0.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.1, 0.2, 0.3, 0.4)
 
 
@@ -61,7 +61,7 @@ uniform  0 float4 0.0 0.0 0.0 0.0
 uniform  4 float4 0.0 0.0 0.0 0.0
 uniform  8 float4 0.5 0.6 0.0 0.0
 uniform 12 float4 0.7 0.8 0.0 0.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.5, 0.6, 0.7, 0.8)
 
 
@@ -111,7 +111,7 @@ float4 main() : sv_target
 [test]
 uniform  0 float4 0.1 0.2 0.0 0.0
 uniform  4 float4 0.3 0.4 0.0 0.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.1, 0.2, 0.3, 0.4)
 
 
@@ -149,7 +149,7 @@ uniform  0 float4 0.3 0.4 0.0 0.0
 uniform  4 float4 0.0 0.0 0.0 0.0
 uniform  8 float4 0.0 0.0 0.0 0.0
 uniform 12 float4 0.5 0.6 0.0 0.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.3, 0.4, 0.5, 0.6)
 
 
@@ -173,7 +173,7 @@ float4 main() : sv_target
 [test]
 uniform  0 float4 0.1 0.2 0.0 0.0
 uniform  4 float4 0.3 0.4 0.0 0.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.1, 0.2, 0.3, 0.4)
 
 
@@ -201,7 +201,7 @@ float4 main() : sv_target
 [test]
 uniform  0 float4 0.2 0.4 0.0 0.0
 uniform  4 float4 0.3 0.5 0.0 0.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.2, 0.3, 0.4, 0.5)
 
 
@@ -221,7 +221,7 @@ uniform  0 float4 0.3 0.0 0.0 0.0
 uniform  4 float4 0.4 0.0 0.0 0.0
 uniform  8 float4 0.0 0.5 0.0 0.0
 uniform 12 float4 0.0 0.6 0.0 0.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.3, 0.4, 0.5, 0.6)
 
 % Compiler options
@@ -245,7 +245,7 @@ uniform  0 float4 0.1 0.5 0.9 1.3
 uniform  4 float4 0.2 0.6 1.0 1.4
 uniform  8 float4 0.3 0.7 1.1 1.5
 uniform 12 float4 0.4 0.8 1.2 1.6
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.2, 0.3, 0.6, 0.7) 1
 
 [require]
@@ -267,7 +267,7 @@ uniform  0 float4 0.1 0.5 0.9 1.3
 uniform  4 float4 0.2 0.6 1.0 1.4
 uniform  8 float4 0.3 0.7 1.1 1.5
 uniform 12 float4 0.4 0.8 1.2 1.6
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.2, 0.3, 0.6, 0.7) 1
 
 [require]
@@ -289,7 +289,7 @@ uniform  0 float4 0.1 0.5 0.9 1.3
 uniform  4 float4 0.2 0.6 1.0 1.4
 uniform  8 float4 0.3 0.7 1.1 1.5
 uniform 12 float4 0.4 0.8 1.2 1.6
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.5, 0.9, 0.6, 1.0) 1
 
 [require]
@@ -317,7 +317,7 @@ uniform 16 float4 1.7 2.1 2.5 2.9
 uniform 20 float4 1.8 2.2 2.6 3.0
 uniform 24 float4 1.9 2.3 2.7 3.1
 uniform 28 float4 2.0 2.4 2.8 3.2
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.3, 0.4, 2.5, 2.9) 1
 
 [require]
@@ -345,7 +345,7 @@ uniform 16 float4 1.7 2.1 2.5 2.9
 uniform 20 float4 1.8 2.2 2.6 3.0
 uniform 24 float4 1.9 2.3 2.7 3.1
 uniform 28 float4 2.0 2.4 2.8 3.2
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (1.2, 1.6, 3.1, 3.2) 1
 
 [require]
@@ -365,5 +365,5 @@ float4 main() : sv_target
 [test]
 uniform  0 float4 0.1 0.2 0.0 0.0
 uniform  4 float4 0.3 0.4 0.0 0.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.1, 0.3, 0.2, 0.4) 1

tests/hlsl/majority-syntax.shader_test

@@ -11,7 +11,7 @@ float4 main() : sv_target
 [test]
 uniform 0 float4 0.1 0.3 0.0 0.0
 uniform 4 float4 0.2 0.4 0.0 0.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.1, 0.3, 0.2, 0.4)
 
 [pixel shader fail(sm<6)]

tests/hlsl/matrix-indexing.shader_test

@@ -11,7 +11,7 @@ uniform 0 float4 1.0 2.0 3.0 4.0
 uniform 4 float4 5.0 6.0 7.0 8.0
 uniform 8 float4 9.0 10.0 11.0 12.0
 uniform 12 float4 13.0 14.0 15.0 16.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (1.0, 2.0, 10.0, 15.0)
 
 [pixel shader]
@@ -27,7 +27,7 @@ uniform 0 float4 1.0 2.0 3.0 4.0
 uniform 4 float4 5.0 6.0 7.0 8.0
 uniform 8 float4 9.0 10.0 11.0 12.0
 uniform 12 float4 13.0 14.0 15.0 16.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (1.0, 2.0, 10.0, 15.0)
 
 [pixel shader]
@@ -43,7 +43,7 @@ uniform 0 float4 1.0 2.0 3.0 4.0
 uniform 4 float4 5.0 6.0 7.0 8.0
 uniform 8 float4 9.0 10.0 11.0 12.0
 uniform 12 float4 13.0 14.0 15.0 16.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (1.0, 5.0, 7.0, 12.0)
 
 [pixel shader]
@@ -58,7 +58,7 @@ float4 main() : SV_TARGET
 [test]
 uniform 0 float4 1.0 2.0 3.0 0.0
 uniform 4 float4 5.0 6.0 7.0 0.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (1.0, 3.0, 6.0, 7.0)
 
 [pixel shader]

tests/hlsl/nested-arrays.shader_test

@@ -21,5 +21,5 @@ uniform  8 float4 0.3 0.0 0.0 0.0
 uniform 12 float4 0.4 0.0 0.0 0.0
 uniform 16 float4 0.5 0.0 0.0 0.0
 uniform 20 float4 0.6 0.0 0.0 0.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.4, 0.1, 0.6, 0.3)

tests/hlsl/object-field-offsets.shader_test

@@ -19,7 +19,7 @@ float4 main() : sv_target
 [test]
 uniform 0 float4 1.0 2.0 3.0 4.0
 uniform 4 float4 5.0 6.0 7.0 8.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (1.0, 2.0, 3.0, 0.0)
 
 
@@ -66,5 +66,5 @@ float4 main() : sv_target
 [test]
 uniform 0 float4 1.0 2.0 3.0 4.0
 uniform 4 float4 5.0 6.0 7.0 8.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (1.0, 2.0, 5.0, 0.0)

tests/hlsl/storage-qualifiers.shader_test

@@ -42,5 +42,5 @@ void main(out float4 o : sv_target)
 [test]
 uniform 0 float4 0.1 0.0 0.0 0.0
 uniform 4 float4 0.2 0.0 0.0 0.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.1, 0.2, 0.3, 0.4)

tests/hlsl/struct-array.shader_test

@@ -18,5 +18,5 @@ float4 main() : sv_target
 uniform 0 float4 0.1 0.2 0.0 0.0
 uniform 4 float4 0.3 0.4 0.0 0.0
 uniform 8 float4 0.5 0.6 0.0 0.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.2, 0.3, 0.6, 0.5)

tests/hlsl/swizzle-matrix.shader_test

@@ -9,7 +9,7 @@ float4 main() : sv_target
 [test]
 uniform 0 float4 11 21 31 -1
 uniform 4 float4 12 22 32 -1
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (21.0, 31.0, 11.0, 12.0)
 
 
@@ -24,7 +24,7 @@ float4 main() : sv_target
 [test]
 uniform 0 float4 11 21 31 -1
 uniform 4 float4 12 22 32 -1
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (11.0, 31.0, 12.0, 32.0)
 
 
@@ -40,7 +40,7 @@ float4 main() : sv_target
 uniform 0 float4 11 12 -1 -1
 uniform 4 float4 21 22 -1 -1
 uniform 8 float4 31 32 -1 -1
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (11.0, 31.0, 12.0, 32.0)
 
 
@@ -169,8 +169,8 @@ float4 main() : sv_target
 
 [test]
 uniform 0 float4 20 30 40 -1
-todo draw quad
+todo(sm<6) draw quad
-todo probe all rgba (10.0, 20.0, 30.0, 40.0)
+todo(sm<6) probe all rgba (10.0, 20.0, 30.0, 40.0)
 
 
 [pixel shader todo]
@@ -187,8 +187,8 @@ float4 main() : sv_target
 
 [test]
 uniform 0 float4 20 30 80 -1
-todo draw quad
+todo(sm<6) draw quad
-todo probe all rgba (80.0, 30.0, 20.0, 10.0)
+todo(sm<6) probe all rgba (80.0, 30.0, 20.0, 10.0)
 
 
 % Cannot repeat components when assigning to a swizzle.

tests/hlsl/swizzles.shader_test

@@ -11,7 +11,7 @@ float4 main() : sv_target
 
 [test]
 uniform 0 float4 0.0303 0.08 0.07 0.0202
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (0.0101, 0.0303, 0.0202, 0.0404)
 
 
@@ -149,7 +149,7 @@ float4 main() : sv_target
 
 [test]
 uniform 0 float4 1.0 2.0 3.0 4.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (1.0, 2.0, 3.0, 4.0)
 
 
@@ -166,5 +166,5 @@ float4 main() : sv_target
 
 [test]
 uniform 0 float4 1.0 2.0 3.0 4.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (1.0, 4.0, 2.0, 3.0)

tests/hlsl/uniform-semantics.shader_test

@@ -10,7 +10,7 @@ float4 main() : sv_target
 
 [test]
 uniform 0 float 3.5
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (3.5, 3.5, 3.5, 3.5)
 
 
@@ -24,5 +24,5 @@ float4 main() : sv_target
 
 [test]
 uniform 0 float4 4.0 5.0 6.0 7.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (4.0, 5.0, 4.0, 5.0)

tests/hlsl/vector-indexing.shader_test

@@ -23,7 +23,7 @@ float4 main() : SV_TARGET
 
 [test]
 uniform 0 float4 1.0 2.0 3.0 4.0
-todo(sm>=6) draw quad
+draw quad
 probe all rgba (1.0, 2.0, 2.0, 3.0)