[Mesa-dev] [PATCH v2 1/3] nir: sge operation is defined for floating-point types

[Juan A. Suarez Romero]

According to GLSL.std.450 spec, the operand for step() function must be
a floating-point. It does not restrict the value to 32-bit floats.

Reviewed by: Elie Tournier <elie.tournier at collabora.com>
---
 src/compiler/nir/nir_opcodes.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/compiler/nir/nir_opcodes.py b/src/compiler/nir/nir_opcodes.py
index 31b46159a4..39c01a702f 100644
--- a/src/compiler/nir/nir_opcodes.py
+++ b/src/compiler/nir/nir_opcodes.py
@@ -467,7 +467,7 @@ binop_reduce("fany_nequal", 1, tfloat32, tfloat32, "{src0} != {src1}",
 # and false respectively
 
 binop("slt", tfloat32, "", "(src0 < src1) ? 1.0f : 0.0f") # Set on Less Than
-binop("sge", tfloat32, "", "(src0 >= src1) ? 1.0f : 0.0f") # Set on Greater or Equal
+binop("sge", tfloat, "", "(src0 >= src1) ? 1.0f : 0.0f") # Set on Greater or Equal
 binop("seq", tfloat32, commutative, "(src0 == src1) ? 1.0f : 0.0f") # Set on Equal
 binop("sne", tfloat32, commutative, "(src0 != src1) ? 1.0f : 0.0f") # Set on Not Equal
 
-- 
2.11.0