[PATCH vkd3d 08/11] vkd3d-shader/hlsl: Fold constant division.

[Giovanni Mascellani]

This commit includes work by Francisco Casas.

Signed-off-by: Giovanni Mascellani <gmascellani at codeweavers.com>
---
 libs/vkd3d-shader/hlsl_constant_ops.c    | 67 ++++++++++++++++++++++++
 libs/vkd3d-shader/vkd3d_shader_private.h |  2 +
 tests/arithmetic-float.shader_test       | 16 ++++++
 tests/arithmetic-int.shader_test         | 11 +++-
 tests/arithmetic-uint.shader_test        |  9 ++++
 5 files changed, 104 insertions(+), 1 deletion(-)

diff --git a/libs/vkd3d-shader/hlsl_constant_ops.c b/libs/vkd3d-shader/hlsl_constant_ops.c
index 5cac4bde..9a8bc6b0 100644
--- a/libs/vkd3d-shader/hlsl_constant_ops.c
+++ b/libs/vkd3d-shader/hlsl_constant_ops.c
@@ -255,6 +255,69 @@ static bool fold_nequal(struct hlsl_ctx *ctx, struct hlsl_ir_constant *dst,
     return true;
 }
 
+static bool fold_div(struct hlsl_ctx *ctx, struct hlsl_ir_constant *dst,
+        struct hlsl_ir_constant *src1, struct hlsl_ir_constant *src2)
+{
+    enum hlsl_base_type type = dst->node.data_type->base_type;
+    unsigned int k;
+
+    assert(type == src1->node.data_type->base_type);
+    assert(type == src2->node.data_type->base_type);
+
+    for (k = 0; k < dst->node.data_type->dimx; ++k)
+    {
+        switch (type)
+        {
+            case HLSL_TYPE_FLOAT:
+            case HLSL_TYPE_HALF:
+                if (src2->value[k].f == 0)
+                {
+                    hlsl_warning(ctx, &dst->node.loc, VKD3D_SHADER_WARNING_HLSL_DIVISION_BY_ZERO,
+                            "Floating point division by zero");
+                }
+                dst->value[k].f = src1->value[k].f / src2->value[k].f;
+                break;
+
+            case HLSL_TYPE_DOUBLE:
+                if (src2->value[k].d == 0)
+                {
+                    hlsl_warning(ctx, &dst->node.loc, VKD3D_SHADER_WARNING_HLSL_DIVISION_BY_ZERO,
+                            "Floating point division by zero");
+                }
+                dst->value[k].d = src1->value[k].d / src2->value[k].d;
+                break;
+
+            case HLSL_TYPE_INT:
+                if (src2->value[k].i == 0)
+                {
+                    hlsl_error(ctx, &dst->node.loc, VKD3D_SHADER_ERROR_HLSL_DIVISION_BY_ZERO,
+                            "Division by zero.");
+                    return false;
+                }
+                if (src1->value[k].i == INT_MIN && src2->value[k].i == -1)
+                    dst->value[k].i = INT_MIN;
+                else
+                    dst->value[k].i = src1->value[k].i / src2->value[k].i;
+                break;
+
+            case HLSL_TYPE_UINT:
+                if (src2->value[k].u == 0)
+                {
+                    hlsl_error(ctx, &dst->node.loc, VKD3D_SHADER_ERROR_HLSL_DIVISION_BY_ZERO,
+                            "Division by zero.");
+                    return false;
+                }
+                dst->value[k].u = src1->value[k].u / src2->value[k].u;
+                break;
+
+            default:
+                FIXME("Fold division for type %s.\n", debug_hlsl_type(ctx, dst->node.data_type));
+                return false;
+        }
+    }
+    return true;
+}
+
 bool hlsl_fold_constants(struct hlsl_ctx *ctx, struct hlsl_ir_node *instr, void *context)
 {
     struct hlsl_ir_constant *arg1, *arg2 = NULL, *res;
@@ -308,6 +371,10 @@ bool hlsl_fold_constants(struct hlsl_ctx *ctx, struct hlsl_ir_node *instr, void
             success = fold_nequal(ctx, res, arg1, arg2);
             break;
 
+        case HLSL_OP2_DIV:
+            success = fold_div(ctx, res, arg1, arg2);
+            break;
+
         default:
             FIXME("Fold \"%s\" expression.\n", debug_hlsl_expr_op(expr->op));
             success = false;
diff --git a/libs/vkd3d-shader/vkd3d_shader_private.h b/libs/vkd3d-shader/vkd3d_shader_private.h
index 21f23709..29e178cc 100644
--- a/libs/vkd3d-shader/vkd3d_shader_private.h
+++ b/libs/vkd3d-shader/vkd3d_shader_private.h
@@ -117,8 +117,10 @@ enum vkd3d_shader_error
     VKD3D_SHADER_ERROR_HLSL_INVALID_TEXEL_OFFSET        = 5018,
     VKD3D_SHADER_ERROR_HLSL_OFFSET_OUT_OF_BOUNDS        = 5019,
     VKD3D_SHADER_ERROR_HLSL_INCOMPATIBLE_PROFILE        = 5020,
+    VKD3D_SHADER_ERROR_HLSL_DIVISION_BY_ZERO            = 5021,
 
     VKD3D_SHADER_WARNING_HLSL_IMPLICIT_TRUNCATION       = 5300,
+    VKD3D_SHADER_WARNING_HLSL_DIVISION_BY_ZERO          = 5301,
 
     VKD3D_SHADER_ERROR_GLSL_INTERNAL                    = 6000,
 
diff --git a/tests/arithmetic-float.shader_test b/tests/arithmetic-float.shader_test
index d55dc9d4..f99b9728 100644
--- a/tests/arithmetic-float.shader_test
+++ b/tests/arithmetic-float.shader_test
@@ -23,3 +23,19 @@ float4 main() : SV_TARGET
 [test]
 todo draw quad
 probe all rgba (5.0, 5.0, -5.0, 3.0)
+
+[require]
+shader model >= 4.0
+
+[pixel shader]
+float4 main() : SV_TARGET
+{
+    float x = 1;
+    float y = 0;
+
+    return x / y;
+}
+
+[test]
+draw quad
+probe all rgba (1e99, 1e99, 1e99, 1e99)
diff --git a/tests/arithmetic-int.shader_test b/tests/arithmetic-int.shader_test
index 5d9cc029..c85b9a3a 100644
--- a/tests/arithmetic-int.shader_test
+++ b/tests/arithmetic-int.shader_test
@@ -8,7 +8,7 @@ float4 main() : SV_TARGET
 }
 
 [test]
-todo draw quad
+draw quad
 probe all rgba (20.0, -10.0, 75.0, 0.0)
 
 [pixel shader]
@@ -23,3 +23,12 @@ float4 main() : SV_TARGET
 [test]
 todo draw quad
 probe all rgba (5.0, 5.0, -5.0, 3.0)
+
+[pixel shader fail]
+float4 main() : SV_TARGET
+{
+    int x = 1;
+    int y = 0;
+
+    return x / y;
+}
diff --git a/tests/arithmetic-uint.shader_test b/tests/arithmetic-uint.shader_test
index b016dbb6..f37c881e 100644
--- a/tests/arithmetic-uint.shader_test
+++ b/tests/arithmetic-uint.shader_test
@@ -26,3 +26,12 @@ float4 main() : SV_TARGET
 [test]
 draw quad
 probe all rgba (5.0, 5.0, 4294967296.0, 3.0)
+
+[pixel shader fail]
+float4 main() : SV_TARGET
+{
+    uint x = 1;
+    uint y = 0;
+
+    return x / y;
+}
-- 
2.35.2