dialects: (arith) split generic binary op definition into specific on…

…es (#3274)

This makes the next step for Pyright updating easier, as we can't
actually properly represent generics in the IRDL constraint system.
Currently, the arith base class helpers work because the generics are
all specified by the time the ops are "defined" with the annotation. But
my proposed solution of replacing the Annotated with VarConstraints
means that we need to define the constraint on a specific type already.
My understanding is that this is not a functional change, only requiring
some clients of the base classes to update the names that they refer to.

Part of #3264

Note stacked PR.

---------

Co-authored-by: Alex Rice <alexrice999@hotmail.co.uk>

tests/dialects/test_arith.py

@@ -7,7 +7,6 @@
     Addi,
     AddUIExtended,
     AndI,
-    BinaryOperation,
     CeilDivSI,
     CeilDivUI,
     Cmpf,
@@ -20,7 +19,7 @@
     ExtSIOp,
     ExtUIOp,
     FastMathFlagsAttr,
-    FloatingPointLikeBinaryOp,
+    FloatingPointLikeBinaryOperation,
     FloorDivSI,
     FPToSIOp,
     IndexCastOp,
@@ -41,6 +40,7 @@
     ShLI,
     ShRSI,
     ShRUI,
+    SignlessIntegerBinaryOperation,
     SIToFPOp,
     Subf,
     Subi,
@@ -103,7 +103,7 @@ class Test_integer_arith_construction:
     @pytest.mark.parametrize("return_type", [None, operand_type])
     def test_arith_ops_init(
         self,
-        OpClass: type[BinaryOperation[_BinOpArgT]],
+        OpClass: type[SignlessIntegerBinaryOperation],
         return_type: Attribute,
     ):
         op = OpClass(self.a, self.b)
@@ -210,7 +210,9 @@ class Test_float_arith_construction:
         "flags", [FastMathFlagsAttr("none"), FastMathFlagsAttr("fast"), None]
     )
     def test_arith_ops(
-        self, func: type[FloatingPointLikeBinaryOp], flags: FastMathFlagsAttr | None
+        self,
+        func: type[FloatingPointLikeBinaryOperation],
+        flags: FastMathFlagsAttr | None,
     ):
         op = func(self.a, self.b, flags)
         assert op.operands[0].owner is self.a

tests/dialects/test_math.py

@@ -2,7 +2,7 @@
 
 import pytest
 
-from xdsl.dialects.arith import BinaryOperation, Constant
+from xdsl.dialects.arith import Constant, FloatingPointLikeBinaryOperation
 from xdsl.dialects.builtin import (
     DenseIntOrFPElementsAttr,
     FloatAttr,
@@ -84,7 +84,7 @@ class Test_float_math_binary_construction:
     @pytest.mark.parametrize("return_type", [None, operand_type])
     def test_float_binary_ops_constant_math_init(
         self,
-        OpClass: type[BinaryOperation[_BinOpArgT]],
+        OpClass: type[FloatingPointLikeBinaryOperation],
         return_type: Attribute,
     ):
         op = OpClass(self.a, self.b)
@@ -104,7 +104,7 @@ def test_float_binary_ops_constant_math_init(
     )
     @pytest.mark.parametrize("return_type", [None, f32_vector_type])
     def test_flaot_binary_vector_ops_init(
-        self, OpClass: type[BinaryOperation[_BinOpArgT]], return_type: Attribute
+        self, OpClass: type[FloatingPointLikeBinaryOperation], return_type: Attribute
     ):
         op = OpClass(self.lhs_vector, self.rhs_vector)
         assert isinstance(op, OpClass)
@@ -123,7 +123,7 @@ def test_flaot_binary_vector_ops_init(
     )
     @pytest.mark.parametrize("return_type", [None, f32_tensor_type])
     def test_float_binary_ops_tensor_math_init(
-        self, OpClass: type[BinaryOperation[_BinOpArgT]], return_type: Attribute
+        self, OpClass: type[FloatingPointLikeBinaryOperation], return_type: Attribute
     ):
         op = OpClass(self.lhs_tensor, self.rhs_tensor)
         assert isinstance(op, OpClass)

xdsl/backend/riscv/lowering/convert_arith_to_riscv.py

@@ -152,7 +152,7 @@ def match_and_rewrite(
 
 @dataclass
 class LowerBinaryIntegerOp(RewritePattern):
-    arith_op_cls: type[arith.SignlessIntegerBinaryOp]
+    arith_op_cls: type[arith.SignlessIntegerBinaryOperation]
     riscv_op_cls: type[RdRsRsIntegerOperation]
 
     def match_and_rewrite(self, op: Operation, rewriter: PatternRewriter) -> None:
@@ -169,7 +169,7 @@ def match_and_rewrite(self, op: Operation, rewriter: PatternRewriter) -> None:
 
 @dataclass
 class LowerBinaryFloatOp(RewritePattern):
-    arith_op_cls: type[arith.FloatingPointLikeBinaryOp]
+    arith_op_cls: type[arith.FloatingPointLikeBinaryOperation]
     riscv_f_op_cls: type[riscv.RdRsRsFloatOperationWithFastMath]
     riscv_d_op_cls: type[riscv.RdRsRsFloatOperationWithFastMath]
 

xdsl/backend/riscv/lowering/convert_arith_to_riscv_snitch.py

@@ -26,7 +26,7 @@
 
 @dataclass
 class LowerBinaryFloatVectorOp(RewritePattern):
-    arith_op_cls: type[arith.FloatingPointLikeBinaryOp]
+    arith_op_cls: type[arith.FloatingPointLikeBinaryOperation]
     riscv_d_op_cls: type[riscv.RdRsRsFloatOperationWithFastMath]
     riscv_snitch_v_f32_op_cls: type[riscv.RdRsRsFloatOperationWithFastMath]
     riscv_snitch_v_f16_op_cls: type[riscv.RdRsRsFloatOperationWithFastMath]