refactor func to be a class hierarchy

mlir_utils/_configuration/configuration.py

@@ -5,13 +5,16 @@
 from base64 import urlsafe_b64encode
 from importlib.metadata import distribution, packages_distributions
 from importlib.resources import files
+from importlib.resources.readers import MultiplexedPath
 from pathlib import Path
 
 from .module_alias_map import get_meta_path_insertion_index, AliasedModuleFinder
 
 __MLIR_PYTHON_PACKAGE_PREFIX__ = "__MLIR_PYTHON_PACKAGE_PREFIX__"
 PACKAGE = __package__.split(".")[0]
 PACKAGE_ROOT_PATH = files(PACKAGE)
+if isinstance(PACKAGE_ROOT_PATH, MultiplexedPath):
+    PACKAGE_ROOT_PATH = PACKAGE_ROOT_PATH._paths[0]
 DIST = distribution(packages_distributions()[PACKAGE][0])
 MLIR_PYTHON_PACKAGE_PREFIX_TOKEN_PATH = (
     Path(__file__).parent / __MLIR_PYTHON_PACKAGE_PREFIX__

mlir_utils/dialects/ext/func.py

@@ -18,95 +18,114 @@
 )
 
 
-def func_base(
-    FuncOp,
-    ReturnOp,
-    CallOp,
-    sym_visibility=None,
-    arg_attrs=None,
-    res_attrs=None,
-    loc=None,
-    ip=None,
-):
-    ip = ip or InsertionPoint.current
-
-    # if this is set to true then wrapper below won't emit a call op
-    # it is set below by a def emit fn that is attached to the body_builder
-    # wrapper; thus you can call wrapped_fn.emit() (i.e., without an operands)
-    # and the func will be emitted.
-    _emit = False
-
-    def builder_wrapper(body_builder):
-        @wraps(body_builder)
-        def wrapper(*call_args):
-            # TODO(max): implement constexpr ie enable passing constants that skip being
-            # part of the signature
-            sig = inspect.signature(body_builder)
-            implicit_return = sig.return_annotation is inspect._empty
-            input_types = [p.annotation for p in sig.parameters.values()]
-            if not (
-                len(input_types) == len(sig.parameters)
-                and all(isinstance(t, Type) for t in input_types)
-            ):
-                input_types = [a.type for a in call_args]
-            function_type = TypeAttr.get(
-                FunctionType.get(
-                    inputs=input_types,
-                    results=[] if implicit_return else sig.return_annotation,
-                )
+class FuncOpMeta(type):
+    def __call__(cls, *args, **kwargs):
+        cls_obj = cls.__new__(cls)
+        if len(args) == 1 and len(kwargs) == 0 and inspect.isfunction(args[0]):
+            return cls.__init__(cls_obj, args[0])
+        else:
+
+            def init_wrapper(f):
+                cls.__init__(cls_obj, f, *args, **kwargs)
+                return cls_obj
+
+            return lambda f: init_wrapper(f)
+
+
+class FuncBase(metaclass=FuncOpMeta):
+    def __init__(
+        self,
+        body_builder,
+        func_op_ctor,
+        return_op_ctor,
+        call_op_ctor,
+        sym_visibility=None,
+        arg_attrs=None,
+        res_attrs=None,
+        loc=None,
+        ip=None,
+    ):
+        assert inspect.isfunction(body_builder), body_builder
+        assert inspect.isclass(func_op_ctor), func_op_ctor
+        assert inspect.isclass(return_op_ctor), return_op_ctor
+        assert inspect.isclass(call_op_ctor), call_op_ctor
+
+        self.body_builder = body_builder
+        self.func_name = self.body_builder.__name__
+
+        self.func_op_ctor = func_op_ctor
+        self.return_op_ctor = return_op_ctor
+        self.call_op_ctor = call_op_ctor
+        self.sym_visibility = (
+            StringAttr.get(str(sym_visibility)) if sym_visibility is not None else None
+        )
+        self.arg_attrs = arg_attrs
+        self.res_attrs = res_attrs
+        self.loc = loc
+        self.ip = ip or InsertionPoint.current
+        self.emitted = False
+
+    def __str__(self):
+        return str(f"{self.__class__} {self.__dict__}")
+
+    def body_builder_wrapper(self, *call_args):
+        sig = inspect.signature(self.body_builder)
+        implicit_return = sig.return_annotation is inspect._empty
+        input_types = [p.annotation for p in sig.parameters.values()]
+        if not (
+            len(input_types) == len(sig.parameters)
+            and all(isinstance(t, Type) for t in input_types)
+        ):
+            input_types = [a.type for a in call_args]
+        function_type = TypeAttr.get(
+            FunctionType.get(
+                inputs=input_types,
+                results=[] if implicit_return else sig.return_annotation,
             )
-            # FuncOp is extended but we do really want the base
-            func_name = body_builder.__name__
-            func_op = FuncOp(
-                func_name,
-                function_type,
-                sym_visibility=StringAttr.get(str(sym_visibility))
-                if sym_visibility is not None
-                else None,
-                arg_attrs=arg_attrs,
-                res_attrs=res_attrs,
-                loc=loc,
-                ip=ip,
+        )
+        func_op = self.func_op_ctor(
+            self.func_name,
+            function_type,
+            sym_visibility=self.sym_visibility,
+            arg_attrs=self.arg_attrs,
+            res_attrs=self.res_attrs,
+            loc=self.loc,
+            ip=self.ip,
+        )
+        func_op.regions[0].blocks.append(*input_types)
+        with InsertionPoint(func_op.regions[0].blocks[0]):
+            results = get_result_or_results(
+                self.body_builder(*func_op.regions[0].blocks[0].arguments)
             )
-            func_op.regions[0].blocks.append(*input_types)
-            with InsertionPoint(func_op.regions[0].blocks[0]):
-                results = get_result_or_results(
-                    body_builder(*func_op.regions[0].blocks[0].arguments)
-                )
-                if results is not None:
-                    if isinstance(results, (tuple, list)):
-                        results = list(results)
-                    else:
-                        results = [results]
+            if results is not None:
+                if isinstance(results, (tuple, list)):
+                    results = list(results)
                 else:
-                    results = []
-                ReturnOp(results)
-            # Recompute the function type.
-            return_types = [v.type for v in results]
-            function_type = FunctionType.get(inputs=input_types, results=return_types)
-            func_op.attributes["function_type"] = TypeAttr.get(function_type)
-
-            if _emit:
-                return maybe_cast(get_result_or_results(func_op))
+                    results = [results]
             else:
-                call_op = CallOp(
-                    [r.type for r in results],
-                    FlatSymbolRefAttr.get(func_name),
-                    call_args,
-                )
-                return maybe_cast(get_result_or_results(call_op))
+                results = []
+            self.return_op_ctor(results)
 
-        def emit():
-            nonlocal _emit
-            _emit = True
-            wrapper()
+        return results, input_types, func_op
 
-        wrapper.emit = emit
-        return wrapper
+    def emit(self):
+        self.results, input_types, func_op = self.body_builder_wrapper()
+        return_types = [v.type for v in self.results]
+        function_type = FunctionType.get(inputs=input_types, results=return_types)
+        func_op.attributes["function_type"] = TypeAttr.get(function_type)
+        self.emitted = True
+        # this is the func op itself (funcs never have a resulting ssa value)
+        return maybe_cast(get_result_or_results(func_op))
 
-    return builder_wrapper
+    def __call__(self, *call_args):
+        if not self.emitted:
+            self.emit()
+        call_op = CallOp(
+            [r.type for r in self.results],
+            FlatSymbolRefAttr.get(self.func_name),
+            call_args,
+        )
+        return maybe_cast(get_result_or_results(call_op))
 
 
-func = make_maybe_no_args_decorator(
-    partial(func_base, FuncOp=FuncOp.__base__, ReturnOp=ReturnOp, CallOp=CallOp)
-)
+func = FuncBase(FuncOp.__base__, ReturnOp, CallOp.__base__)

mlir_utils/dialects/ext/tensor.py

@@ -2,7 +2,7 @@
 from typing import Union, Tuple, Sequence
 
 import numpy as np
-from mlir.dialects.tensor import EmptyOp
+from mlir.dialects.tensor import EmptyOp, GenerateOp
 from mlir.ir import Type, Value, RankedTensorType, DenseElementsAttr, ShapedType
 
 from mlir_utils.dialects.ext.arith import ArithValue
@@ -62,7 +62,6 @@ def empty(
         shape: Union[list[Union[int, Value]], tuple[Union[int, Value], ...]],
         el_type: Type,
     ) -> "Tensor":
-
         return cls(EmptyOp(shape, el_type).result)
 
 

tests/test_func.py

@@ -2,7 +2,6 @@
 from textwrap import dedent
 
 import pytest
-
 from mlir_utils.dialects.ext.arith import constant
 from mlir_utils.dialects.ext.func import func
 
@@ -20,7 +19,7 @@ def demo_fun1():
         return one
 
     assert hasattr(demo_fun1, "emit")
-    assert inspect.isfunction(demo_fun1.emit)
+    assert inspect.ismethod(demo_fun1.emit)
     demo_fun1.emit()
     correct = dedent(
         """\
@@ -33,3 +32,63 @@ def demo_fun1():
     """
     )
     filecheck(correct, ctx.module)
+
+
+def test_func_base_meta(ctx: MLIRContext):
+    print()
+
+    @func
+    def foo1():
+        one = constant(1)
+        return one
+
+    # print("wrapped foo", foo1)
+    foo1.emit()
+    correct = dedent(
+        """\
+    module {
+      func.func @foo1() -> i64 {
+        %c1_i64 = arith.constant 1 : i64
+        return %c1_i64 : i64
+      }
+    }
+    """
+    )
+    filecheck(correct, ctx.module)
+
+    foo1()
+    correct = dedent(
+        """\
+    module {
+      func.func @foo1() -> i64 {
+        %c1_i64 = arith.constant 1 : i64
+        return %c1_i64 : i64
+      }
+      %0 = func.call @foo1() : () -> i64
+    }
+    """
+    )
+    filecheck(correct, ctx.module)
+
+
+def test_func_base_meta2(ctx: MLIRContext):
+    print()
+
+    @func
+    def foo1():
+        one = constant(1)
+        return one
+
+    foo1()
+    correct = dedent(
+        """\
+    module {
+      func.func @foo1() -> i64 {
+        %c1_i64 = arith.constant 1 : i64
+        return %c1_i64 : i64
+      }
+      %0 = func.call @foo1() : () -> i64
+    }
+    """
+    )
+    filecheck(correct, ctx.module)

tests/test_regions.py

@@ -7,7 +7,7 @@
 
 from mlir_utils.dialects.ext.arith import constant
 from mlir_utils.dialects.ext.func import func
-from mlir_utils.dialects.ext.tensor import Tensor, S, rank
+from mlir_utils.dialects.ext.tensor import S, rank
 
 # noinspection PyUnresolvedReferences
 from mlir_utils.testing import mlir_ctx as ctx, filecheck, MLIRContext

tests/test_value_caster.py

@@ -1,5 +1,4 @@
 import pytest
-from mlir.ir import OpResult
 
 from mlir_utils.dialects.ext.tensor import S, empty
 from mlir_utils.dialects.ext.arith import constant
@@ -9,6 +8,8 @@
 from mlir_utils.testing import mlir_ctx as ctx, filecheck, MLIRContext
 from mlir_utils.types import f64_t, RankedTensorType
 
+from mlir.ir import OpResult
+
 # needed since the fix isn't defined here nor conftest.py
 pytest.mark.usefixtures("ctx")