[MLIR][Linalg] Remove elemwise_unary and elemwise_binary #147082
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
rengolin
requested review from
dcaballe,
ftynse,
makslevental,
nicolasvasilache and
stellaraccident
as code owners
Member
|
@llvm/pr-subscribers-mlir @llvm/pr-subscribers-mlir-linalg Author: Renato Golin (rengolin) ChangesRFC: https://discourse.llvm.org/t/rfc-deprecate-linalg-elemwise-unary-and-elemwise-binary/87144 Remove the two operations and fix the tests by:
Nothing else changed. Patch is 54.07 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/147082.diff 15 Files Affected:
diff --git a/mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml b/mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml
index 6344861c53ac5..3637147c5a90d 100644
--- a/mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml
+++ b/mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml
@@ -44,56 +44,6 @@ structured_op: !LinalgStructuredOpConfig
- !ScalarExpression
scalar_arg: I
--- !LinalgOpConfig
-metadata: !LinalgOpMetadata
- name: elemwise_unary
- cpp_class_name: ElemwiseUnaryOp
- doc: |-
- Applies the unary function fun elementwise.
-
- Numeric casting is performed on the input operand, promoting it to the same
- data type as the accumulator/output.
-structured_op: !LinalgStructuredOpConfig
- args:
- - !LinalgOperandDefConfig
- name: I
- kind: input_tensor
- type_var: T1
- shape_map: affine_map<() -> ()>
- - !LinalgOperandDefConfig
- name: O
- kind: output_tensor
- type_var: U
- shape_map: affine_map<() -> ()>
- - !LinalgOperandDefConfig
- name: fun
- kind: unary_fn_attr
- default_fn: exp
- - !LinalgOperandDefConfig
- name: cast
- kind: type_fn_attr
- default_fn: cast_signed
- indexing_maps: !LinalgIndexingMapsConfig
- static_indexing_maps:
- - affine_map<() -> ()>
- - affine_map<() -> ()>
- iterator_types: []
- assignments:
- - !ScalarAssign
- arg: O
- value: !ScalarExpression
- scalar_fn:
- kind: unary
- attr_name: fun
- operands:
- - !ScalarExpression
- scalar_fn:
- kind: type
- attr_name: cast
- type_var: U
- operands:
- - !ScalarExpression
- scalar_arg: I
---- !LinalgOpConfig
metadata: !LinalgOpMetadata
name: exp
cpp_class_name: ExpOp
@@ -549,70 +499,6 @@ structured_op: !LinalgStructuredOpConfig
- !ScalarExpression
scalar_arg: I
--- !LinalgOpConfig
-metadata: !LinalgOpMetadata
- name: elemwise_binary
- cpp_class_name: ElemwiseBinaryOp
- doc: |-
- Applies the binary function fun elementwise.
-
- Numeric casting is performed on the input operand, promoting it to the same
- data type as the accumulator/output.
-structured_op: !LinalgStructuredOpConfig
- args:
- - !LinalgOperandDefConfig
- name: lhs
- kind: input_tensor
- type_var: T1
- shape_map: affine_map<() -> ()>
- - !LinalgOperandDefConfig
- name: rhs
- kind: input_tensor
- type_var: T2
- shape_map: affine_map<() -> ()>
- - !LinalgOperandDefConfig
- name: O
- kind: output_tensor
- type_var: U
- shape_map: affine_map<() -> ()>
- - !LinalgOperandDefConfig
- name: fun
- kind: binary_fn_attr
- default_fn: add
- - !LinalgOperandDefConfig
- name: cast
- kind: type_fn_attr
- default_fn: cast_signed
- indexing_maps: !LinalgIndexingMapsConfig
- static_indexing_maps:
- - affine_map<() -> ()>
- - affine_map<() -> ()>
- - affine_map<() -> ()>
- iterator_types: []
- assignments:
- - !ScalarAssign
- arg: O
- value: !ScalarExpression
- scalar_fn:
- kind: binary
- attr_name: fun
- operands:
- - !ScalarExpression
- scalar_fn:
- kind: type
- attr_name: cast
- type_var: U
- operands:
- - !ScalarExpression
- scalar_arg: lhs
- - !ScalarExpression
- scalar_fn:
- kind: type
- attr_name: cast
- type_var: U
- operands:
- - !ScalarExpression
- scalar_arg: rhs
---- !LinalgOpConfig
metadata: !LinalgOpMetadata
name: add
cpp_class_name: AddOp
diff --git a/mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py b/mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py
index 48e724d80c926..1b359da40a291 100644
--- a/mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py
+++ b/mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py
@@ -21,21 +21,6 @@ def copy(
O[None] = cast(U, I[None])
-@linalg_structured_op
-def elemwise_unary(
- I=TensorDef(T1),
- O=TensorDef(U, output=True),
- fun=UnaryFnAttrDef(default=UnaryFn.exp),
- cast=TypeFnAttrDef(default=TypeFn.cast_signed),
-):
- """Applies the unary function fun elementwise.
-
- Numeric casting is performed on the input operand, promoting it to the same
- data type as the accumulator/output.
- """
- O[None] = fun(cast(U, I[None]))
-
-
@linalg_structured_op
def exp(
I=TensorDef(T1),
@@ -192,22 +177,6 @@ def erf(
O[None] = UnaryFn.erf(I[None])
-@linalg_structured_op
-def elemwise_binary(
- lhs=TensorDef(T1),
- rhs=TensorDef(T2),
- O=TensorDef(U, output=True),
- fun=BinaryFnAttrDef(default=BinaryFn.add),
- cast=TypeFnAttrDef(default=TypeFn.cast_signed),
-):
- """Applies the binary function fun elementwise.
-
- Numeric casting is performed on the input operand, promoting it to the same
- data type as the accumulator/output.
- """
- O[None] = fun(cast(U, lhs[None]), cast(U, rhs[None]))
-
-
@linalg_structured_op
def add(
lhs=TensorDef(T1),
diff --git a/mlir/test/Dialect/Linalg/generalize-named-polymorphic-ops.mlir b/mlir/test/Dialect/Linalg/generalize-named-polymorphic-ops.mlir
index bbd6e0fc8e2cc..290c6c7c36f76 100644
--- a/mlir/test/Dialect/Linalg/generalize-named-polymorphic-ops.mlir
+++ b/mlir/test/Dialect/Linalg/generalize-named-polymorphic-ops.mlir
@@ -434,117 +434,6 @@ func.func @generalize_const(%min: f64, %max: f64, %seed: i32, %O: tensor<16x32xf
// -----
-// Verifies the default value of the fun attribute is an exp op.
-func.func @generalize_elemwise_exp(%lhs : tensor<4x8xf32>, %output : tensor<4x8xf32>) -> tensor<4x8xf32> {
- %0 = linalg.elemwise_unary ins(%lhs: tensor<4x8xf32>) outs(%output: tensor<4x8xf32>) -> tensor<4x8xf32>
- return %0: tensor<4x8xf32>
-}
-
-// CHECK-LABEL: @generalize_elemwise_exp
-// CHECK: = math.exp
-
-// -----
-
-// Verifies the fun attribute controls the unary function used.
-func.func @generalize_elemwise_log(%lhs : tensor<4x8xf32>, %output : tensor<4x8xf32>) -> tensor<4x8xf32> {
- %0 = linalg.elemwise_unary {fun = #linalg.unary_fn<log>}
- ins(%lhs: tensor<4x8xf32>) outs(%output: tensor<4x8xf32>) -> tensor<4x8xf32>
- return %0: tensor<4x8xf32>
-}
-
-// CHECK-LABEL: @generalize_elemwise_log
-// CHECK: = math.log
-
-// -----
-
-// Verifies the fun attribute controls the unary function used.
-func.func @generalize_elemwise_abs(%lhs : tensor<4x8xf32>, %output : tensor<4x8xf32>) -> tensor<4x8xf32> {
- %0 = linalg.elemwise_unary {fun = #linalg.unary_fn<abs>}
- ins(%lhs: tensor<4x8xf32>) outs(%output: tensor<4x8xf32>) -> tensor<4x8xf32>
- return %0: tensor<4x8xf32>
-}
-
-// CHECK-LABEL: @generalize_elemwise_abs
-// CHECK: = math.absf
-
-// -----
-
-// Verifies the fun attribute controls the unary function used.
-func.func @generalize_elemwise_ceil(%lhs : tensor<4x8xf32>, %output : tensor<4x8xf32>) -> tensor<4x8xf32> {
- %0 = linalg.elemwise_unary {fun = #linalg.unary_fn<ceil>}
- ins(%lhs: tensor<4x8xf32>) outs(%output: tensor<4x8xf32>) -> tensor<4x8xf32>
- return %0: tensor<4x8xf32>
-}
-
-// CHECK-LABEL: @generalize_elemwise_ceil
-// CHECK: = math.ceil
-
-// -----
-
-// Verifies the fun attribute controls the unary function used.
-func.func @generalize_elemwise_floor(%lhs : tensor<4x8xf32>, %output : tensor<4x8xf32>) -> tensor<4x8xf32> {
- %0 = linalg.elemwise_unary {fun = #linalg.unary_fn<floor>}
- ins(%lhs: tensor<4x8xf32>) outs(%output: tensor<4x8xf32>) -> tensor<4x8xf32>
- return %0: tensor<4x8xf32>
-}
-
-// CHECK-LABEL: @generalize_elemwise_floor
-// CHECK: = math.floor
-
-// -----
-
-// Verifies the fun attribute controls the unary function used.
-func.func @generalize_elemwise_negf(%lhs : tensor<4x8xf32>, %output : tensor<4x8xf32>) -> tensor<4x8xf32> {
- %0 = linalg.elemwise_unary {fun = #linalg.unary_fn<negf>}
- ins(%lhs: tensor<4x8xf32>) outs(%output: tensor<4x8xf32>) -> tensor<4x8xf32>
- return %0: tensor<4x8xf32>
-}
-
-// CHECK-LABEL: @generalize_elemwise_negf
-// CHECK: = arith.negf
-
-// -----
-
-// Verifies the default value of the fun attribute is an add op.
-func.func @generalize_elemwise_add(%lhs : tensor<4x8xf32>, %rhs : tensor<4x8xf32>, %output : tensor<4x8xf32>) -> tensor<4x8xf32> {
- %0 = linalg.elemwise_binary ins(%lhs, %rhs: tensor<4x8xf32>, tensor<4x8xf32>)
- outs(%output: tensor<4x8xf32>) -> tensor<4x8xf32>
- return %0: tensor<4x8xf32>
-}
-
-// CHECK-LABEL: @generalize_elemwise_add
-// CHECK: = arith.addf
-
-// -----
-
-// Verifies the fun attribute controls the binary function used.
-func.func @generalize_elemwise_mul(%lhs : tensor<4x8xf32>, %rhs : tensor<4x8xf32>, %output : tensor<4x8xf32>) -> tensor<4x8xf32> {
- %0 = linalg.elemwise_binary {fun = #linalg.binary_fn<mul>}
- ins(%lhs, %rhs: tensor<4x8xf32>, tensor<4x8xf32>)
- outs(%output: tensor<4x8xf32>) -> tensor<4x8xf32>
- return %0: tensor<4x8xf32>
-}
-
-// CHECK-LABEL: @generalize_elemwise_mul
-// CHECK: = arith.mulf
-
-// -----
-
-// Verifies pointwise ops support rank zero input tensors
-func.func @generalize_elemwise_rank_zero(%lhs : tensor<f32>, %rhs : tensor<f32>, %output : tensor<4x8xf32>) -> tensor<4x8xf32> {
- %0 = linalg.elemwise_binary {fun = #linalg.binary_fn<sub>}
- ins(%lhs, %rhs: tensor<f32>, tensor<f32>)
- outs(%output: tensor<4x8xf32>) -> tensor<4x8xf32>
- return %0: tensor<4x8xf32>
-}
-
-// CHECK-LABEL: @generalize_elemwise_rank_zero
-// CHECK: linalg.generic
-// CHECK-SAME: iterator_types = ["parallel", "parallel"]
-// CHECK: = arith.subf
-
-// -----
-
// Verifies the fun attribute controls the binary function used.
func.func @generalize_copy(%lhs : tensor<4x8xf32>, %output : tensor<4x8xf32>) -> tensor<4x8xf32> {
%0 = linalg.copy ins(%lhs: tensor<4x8xf32>) outs(%output: tensor<4x8xf32>) -> tensor<4x8xf32>
diff --git a/mlir/test/Dialect/Linalg/invalid.mlir b/mlir/test/Dialect/Linalg/invalid.mlir
index 964681d7dcd92..da1dfc7b6a624 100644
--- a/mlir/test/Dialect/Linalg/invalid.mlir
+++ b/mlir/test/Dialect/Linalg/invalid.mlir
@@ -1909,14 +1909,6 @@ module {
// -----
-func.func @elemwise_unary_invalid_mixed_types(%arg0 : tensor<?xi32>) -> tensor<?xi32> {
- // expected-error @below {{unsupported non numeric type}}
- %0 = linalg.elemwise_unary ins(%arg0 : tensor<?xi32>) outs(%arg0 : tensor<?xi32>) -> tensor<?xi32>
- return %0 : tensor<?xi32>
-}
-
-// -----
-
func.func @matmul_invalid_mixed_types(%t: tensor<?xf16>, %f: vector<4xf16>)
-> (tensor<?xf16>, vector<4xf16>)
{
diff --git a/mlir/test/Dialect/Linalg/library-calls.mlir b/mlir/test/Dialect/Linalg/library-calls.mlir
index 1fa675d8b4b68..77c9d4a911447 100644
--- a/mlir/test/Dialect/Linalg/library-calls.mlir
+++ b/mlir/test/Dialect/Linalg/library-calls.mlir
@@ -59,43 +59,3 @@ module {
return
}
}
-
-
-// -----
-
-// CHECK: func.func private @linalg_elemwise_unary_negf_view16x8xf32_view16x8xf32(memref<16x8xf32, strided<[?, ?], offset: ?>>, memref<16x8xf32, strided<[?, ?], offset: ?>>) attributes {llvm.emit_c_interface}
-// CHECK: func.func private @linalg_elemwise_unary_negf_view16xf32_view16xf32(memref<16xf32, strided<[?], offset: ?>>, memref<16xf32, strided<[?], offset: ?>>) attributes {llvm.emit_c_interface}
-
-func.func @test_neg(%A : memref<16x8xf32>, %B: memref<16x8xf32>, %C: memref<16xf32>, %D: memref<16xf32>) {
- linalg.elemwise_unary {fun = #linalg.unary_fn<negf>}
- ins(%A: memref<16x8xf32>) outs(%B: memref<16x8xf32>)
- linalg.elemwise_unary {fun = #linalg.unary_fn<negf>}
- ins(%C: memref<16xf32>) outs(%D: memref<16xf32>)
- return
-}
-
-// -----
-
-// CHECK: func.func private @linalg_elemwise_unary_exp_view16x8xf32_view16x8xf32(memref<16x8xf32, strided<[?, ?], offset: ?>>, memref<16x8xf32, strided<[?, ?], offset: ?>>) attributes {llvm.emit_c_interface}
-// CHECK: func.func private @linalg_elemwise_unary_exp_view16xf32_view16xf32(memref<16xf32, strided<[?], offset: ?>>, memref<16xf32, strided<[?], offset: ?>>) attributes {llvm.emit_c_interface}
-
-func.func @test_exp(%A : memref<16x8xf32>, %B: memref<16x8xf32>, %C: memref<16xf32>, %D: memref<16xf32>) {
- linalg.elemwise_unary {fun = #linalg.unary_fn<exp>}
- ins(%A: memref<16x8xf32>) outs(%B: memref<16x8xf32>)
- linalg.elemwise_unary {fun = #linalg.unary_fn<exp>}
- ins(%C: memref<16xf32>) outs(%D: memref<16xf32>)
- return
-}
-
-// -----
-
-// CHECK: func.func private @linalg_elemwise_binary_add_view16x8xf32_view16x8xf32_view16x8xf32(memref<16x8xf32, strided<[?, ?], offset: ?>>, memref<16x8xf32, strided<[?, ?], offset: ?>>, memref<16x8xf32, strided<[?, ?], offset: ?>>) attributes {llvm.emit_c_interface}
-// CHECK: func.func private @linalg_elemwise_binary_add_view16xf32_view16xf32_view16xf32(memref<16xf32, strided<[?], offset: ?>>, memref<16xf32, strided<[?], offset: ?>>, memref<16xf32, strided<[?], offset: ?>>) attributes {llvm.emit_c_interface}
-
-func.func @test_add(%A : memref<16x8xf32>, %B: memref<16x8xf32>, %C: memref<16x8xf32>, %D: memref<16xf32>, %E: memref<16xf32>, %F: memref<16xf32>) {
- linalg.elemwise_binary {fun = #linalg.binary_fn<add>}
- ins(%A, %B: memref<16x8xf32>, memref<16x8xf32>) outs(%C: memref<16x8xf32>)
- linalg.elemwise_binary {fun = #linalg.binary_fn<add>}
- ins(%D, %E: memref<16xf32>, memref<16xf32>) outs(%F: memref<16xf32>)
- return
-}
diff --git a/mlir/test/Dialect/Linalg/match-ops-interpreter.mlir b/mlir/test/Dialect/Linalg/match-ops-interpreter.mlir
index bfe7a07cb38a5..618ba3402ff52 100644
--- a/mlir/test/Dialect/Linalg/match-ops-interpreter.mlir
+++ b/mlir/test/Dialect/Linalg/match-ops-interpreter.mlir
@@ -842,15 +842,15 @@ module attributes { transform.with_named_sequence } {
// expected-remark @below {{op result}}
// expected-note @below {{value handle points to an op result #0}}
// expected-remark @below {{single user}}
- linalg.elemwise_unary {fun = #linalg.unary_fn<negf>} ins(%2 : tensor<42x42xf32>) outs(%0 : tensor<42x42xf32>) -> tensor<42x42xf32>
+ linalg.negf ins(%2 : tensor<42x42xf32>) outs(%0 : tensor<42x42xf32>) -> tensor<42x42xf32>
// expected-remark @below {{matched result value}}
// expected-remark @below {{op result}}
// expected-note @below {{value handle points to an op result #0}}
- linalg.elemwise_unary {fun = #linalg.unary_fn<exp>} ins(%3 : tensor<42x42xf32>) outs(%0 : tensor<42x42xf32>) -> tensor<42x42xf32>
+ linalg.exp ins(%3 : tensor<42x42xf32>) outs(%0 : tensor<42x42xf32>) -> tensor<42x42xf32>
// expected-remark @below {{matched result value}}
// expected-remark @below {{op result}}
// expected-note @below {{value handle points to an op result #0}}
- linalg.elemwise_unary {fun = #linalg.unary_fn<exp>} ins(%3 : tensor<42x42xf32>) outs(%0 : tensor<42x42xf32>) -> tensor<42x42xf32>
+ linalg.exp ins(%3 : tensor<42x42xf32>) outs(%0 : tensor<42x42xf32>) -> tensor<42x42xf32>
return
}
}
diff --git a/mlir/test/Dialect/Linalg/one-shot-bufferize-analysis.mlir b/mlir/test/Dialect/Linalg/one-shot-bufferize-analysis.mlir
index 5b7c2baf9d84f..a0922bdfcfbe4 100644
--- a/mlir/test/Dialect/Linalg/one-shot-bufferize-analysis.mlir
+++ b/mlir/test/Dialect/Linalg/one-shot-bufferize-analysis.mlir
@@ -1,43 +1,5 @@
// RUN: mlir-opt %s -one-shot-bufferize="bufferize-function-boundaries test-analysis-only" -split-input-file | FileCheck %s
-// CHECK-LABEL: @elementwise_no_conflict
-func.func @elementwise_no_conflict(%a: tensor<5xf32>,
- %b: tensor<5xf32>) -> tensor<5xf32> {
- // CHECK: linalg.elemwise_binary
- // CHECK-SAME: {__inplace_operands_attr__ = ["true", "true", "true"], fun = #linalg.binary_fn<add>}
- %0 = linalg.elemwise_binary {fun = #linalg.binary_fn<add>}
- ins(%a, %b : tensor<5xf32>, tensor<5xf32>)
- outs(%a : tensor<5xf32>) -> tensor<5xf32>
- return %0 : tensor<5xf32>
-}
-
-// -----
-
-// CHECK-LABEL: @elementwise_no_conflict_2
-func.func @elementwise_no_conflict_2(%a: tensor<5xf32>) -> tensor<5xf32> {
- // CHECK: linalg.elemwise_binary
- // CHECK-SAME: {__inplace_operands_attr__ = ["true", "true", "true"], fun = #linalg.binary_fn<add>}
- %0 = linalg.elemwise_binary {fun = #linalg.binary_fn<add>}
- ins(%a, %a : tensor<5xf32>, tensor<5xf32>)
- outs(%a : tensor<5xf32>) -> tensor<5xf32>
- return %0 : tensor<5xf32>
-}
-
-// -----
-
-// CHECK-LABEL: @elementwise_no_conflict_3
-func.func @elementwise_no_conflict_3(%a: tensor<5xf32>) -> tensor<5xf32> {
- %c0f = arith.constant 1.0 : f32
- // CHECK: linalg.elemwise_binary
- // CHECK-SAME: {__inplace_operands_attr__ = ["true", "none", "true"], fun = #linalg.binary_fn<add>}
- %0 = linalg.elemwise_binary {fun = #linalg.binary_fn<add>}
- ins(%a, %c0f : tensor<5xf32>, f32)
- outs(%a : tensor<5xf32>) -> tensor<5xf32>
- return %0 : tensor<5xf32>
-}
-
-// -----
-
func.func @not_elementwise(%a: tensor<5x6xf32>) -> tensor<5x6xf32> {
%cst = arith.constant 5.0 : f32
// CHECK: tensor.extract_slice
diff --git a/mlir/test/Dialect/Linalg/transform-op-fuse-into-containing.mlir b/mlir/test/Dialect/Linalg/transform-op-fuse-into-containing.mlir
index 5bdb5073ee865..312468970ae6d 100644
--- a/mlir/test/Dialect/Linalg/transform-op-fuse-into-containing.mlir
+++ b/mlir/test/Dialect/Linalg/transform-op-fuse-into-containing.mlir
@@ -26,8 +26,8 @@ module {
// CHECK: %[[T1:.*]] = linalg.fill {{.*}} outs(%[[T0]]
%6 = tensor.extract_slice %0[%3] [%4] [1] : tensor<?xf32> to tensor<?xf32>
- // CHECK: %[[T2:.*]] = linalg.elemwise_unary ins(%[[T1]]
- %7 = linalg.elemwise_unary ins(%6 : tensor<?xf32>) outs(%5 : tensor<?xf32>) -> tensor<?xf32>
+ // CHECK: %[[T2:.*]] = linalg.exp ins(%[[T1]]
+ %7 = linalg.exp ins(%6 : tensor<?xf32>) outs(%5 : tensor<?xf32>) -> tensor<?xf32>
scf.forall.in_parallel {
tensor.parallel_insert_slice %7 into %o[%3] [%4] [1] : tensor<?xf32> into tensor<?xf32>
}
@@ -76,8 +76,8 @@ module {
%4 = affine.min #map2(%arg3)[%arg0]
%5 = tensor.extract_slice %o[%3] [%4] [1] : tensor<64xf32> to tensor<?xf32>
- // CHECK: %[[T2:.*]] = linalg.elemwise_unary ins(%[[INIT_TENSOR]]
- %7 = linalg.elemwise_unary ins(%0 : tensor<?xf32>) outs(%5 : tensor<?xf32>) -> tensor<?xf32>
+ // CHECK: %[[T2:.*]] = linalg.exp ins(%[[INIT_TENSOR]]
+ %7 = linalg.exp ins(%0 : tensor<?xf32>) outs(%5 : tensor<?xf32>) -> tensor<?xf32>
scf.forall.in_parallel {
tensor.parallel_insert_slice %7 into %o[%3] [%4] [1] : tensor<?xf32> into tensor<64xf32>
}
@@ -177,8 +177,8 @@ module {
// CHECK: %[[T1:.*]] = linalg.fill {{.*}} outs(%[[T0]]
%6 = tensor.extract_slice %arg1[%3] [%4] [1] : tensor<?xf32> to tensor<?xf32>
- // CHECK: %[[T2:.*]] = linalg.elemwise_unary {{.*}} outs(%[[T1]]
- %7 = linalg.elemwise_unary ins(%6 : tensor<?xf32>) outs(%5 : tensor<?xf32>) -> tensor<?xf32>
+ // CHECK: %[[T2:.*]] = linalg.exp {{.*}} outs(%[[T1]]
+ %7 = linalg.exp ins(%6 : tensor<?xf32>) outs(%5 : tensor<?xf32>) -> tensor<?xf32>
scf.forall.in_parallel {
tensor.parallel_insert_slice %7 into %o[%3] [%4] [1] : tensor<?xf32> into tensor<?xf32>
}
@@ -228,8 +228,8 @@ module {
// CHECK: %[[T2:.*]] = linalg.fill {{.*}} outs(%[[T1]]
%6 = tensor.extract_slice %0[%3] [%4] [1] : tensor<?xf32> to tensor<?xf32>
- // CHECK: %[[T3:.*]] = linalg.elemwise_unary ins(%[[T2]] : tensor<?xf32>) outs(%[[T0]] : tensor<?xf32>)
- %7 = linalg.elemwise_unary ins(%6 : tensor<?xf32>) outs(%5 : tensor<?xf32>) -> tensor<?xf32>
+ // CHECK: %[[T3:.*]] = linalg.exp ins(%[[T2]] : tensor<?xf32>) outs(%[[T0]] : tensor<?xf32>)
+ %7 = linalg.exp ins(%6 : tensor<?xf32>) outs(%5 : tensor<?xf32>) -> tensor<?xf32>
scf.forall.in_parallel {
tensor.parallel_insert_slice %7 into %o[%3] [%4] [1] : tensor<?xf32> into tensor<?xf32>
}
@@ -261,7 +261,7 @@ module {
%c2 = arith.constant...
[truncated]
|
rengolin
requested review from
MaheshRavishankar,
adam-smnk,
banach-space,
javedabsar1,
rolfmorel and
shahidact
rolfmorel
reviewed
Jul 4, 2025
rolfmorel
reviewed
Jul 4, 2025
mlir/test/Dialect/Linalg/transform-op-fuse-into-containing.mlir
Outdated
Show resolved
Hide resolved
rolfmorel
reviewed
Jul 4, 2025
mlir/test/Dialect/Linalg/transform-op-fuse-into-containing.mlir
Outdated
Show resolved
Hide resolved
rolfmorel
approved these changes
Jul 4, 2025
Contributor
rolfmorel
left a comment
rolfmorel
left a comment
There was a problem hiding this comment.
Besides the above comments, this looks about as expected.
Better wait for others to speak up though, especially any users (in so far as they exist).
ftynse
reviewed
Jul 4, 2025
Member
ftynse
left a comment
ftynse
left a comment
There was a problem hiding this comment.
I remember using elemwise as matching target in the transform dialect tutorial. Could you please double-check that it still runs correctly? It is under examples and the tests may not be run by default unless you specifically build with examples enabled.
mlir/test/Dialect/Linalg/transform-op-fuse-into-containing.mlir
Outdated
Show resolved
Hide resolved
Member
Author
That's what's failing the CI on both Linux and Windows. I'm happy that it caught the issue. :) |
Member
Author
Examples moved to |
Member
Author
|
@makslevental @rolfmorel how do I change the Python interfaces to not try to create these instructions anymore? |
Contributor
Just delete them from https://github.com/llvm/llvm-project/blob/main/mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py |
Member
Author
|
Ok, that's all done. Hopefully no one else is using them, but I'll wait a few days just in case. |
Contributor
Looks like just me and that's fine (I'll remove too). |
adam-smnk
approved these changes
Jul 7, 2025
Contributor
adam-smnk
left a comment
adam-smnk
left a comment
There was a problem hiding this comment.
Great to see this cleanup finally happen 👍
2 tasks
rniczh
added a commit
to PennyLaneAI/catalyst
that referenced
this pull request
Nov 5, 2025
**Context:** Update llvm, stablehlo and enzyme, 2025 Q4 The latest pair of good versions, indicated by stablehlo, is openxla/stablehlo@0a4440a ``` stablehlo=0a4440a5c8de45c4f9649bf3eb4913bf3f97da0d llvm=113f01aa82d055410f22a9d03b3468fa68600589 ``` For Enzyme, we go to the latest release https://github.com/EnzymeAD/Enzyme/releases/tag/v0.0.203 ``` enzyme=v0.0.203 ``` with commit `476c8e3193a8577ba24ff845ae2294109225f83a` **Description of the Change:** - `llvm-project/mlir/include/mlir/InitAllPasses.h` header no longer includes individual pass headers. To incorporate specific passes, you now need to include either `mlir/Conversion/Passes.h` or the header for each dialect pass. This change stems from the upstream cleanup of `mlir/InitAllDialects.h`, which previously handled these includes. Following the decision from the owner (as seen in this pull request comment: https://github.com/ftynse/water/pull/16/files#r2259526343), we will specifically include each required header, faster compilation (no unnecessary headers). For further context, refer to this PR: llvm/llvm-project#151150 - To bufferize custom tensors into custom buffers. The upstream PR llvm/llvm-project#144867 changed the the return type of `bufferization::detail::defaultGetBufferType` to be `BufferLikeType` instead of `BaseMemRefType`. We aligned the behaviour with the upstream PR, return the BufferLikeType from our `getBufferType()` implementation as well. - `elemwise_unary` and `elemwise_binary` of `linalg` have been deprecated from the upstream, replaced with `elementwise`. llvm/llvm-project#147082. Follow this PR, using `linalg.add` directly. register_all_dialects/passes/extension` should now be configured as a static library in CMake, refer to this PR: llvm/llvm-project#151150 - An opt-in feature (`{op}::create` method) be added to support the same behaviour of `rewriter.create` method. But it provides a meaningful message when using this new opt-in feature compared to `rewrite.create`. Since the new generated `create` function will call the `build` function, it requires us to define the instance of the builder (`GraidentOps.td`). Refer to the PR: llvm/llvm-project#147168. **Take `linalg.abs` as an example:** ```cpp void AbsOp::build(::mlir::OpBuilder &odsBuilder, ::mlir::OperationState &odsState, ValueRange inputs, ValueRange outputs, ArrayRef<NamedAttribute> attributes) { buildStructuredOp(odsBuilder, odsState, std::nullopt, inputs, outputs, attributes, AbsOp::getRegionBuilder()); } AbsOp AbsOp::create(::mlir::OpBuilder &builder, ::mlir::Location location, ValueRange inputs, ValueRange outputs, ArrayRef<NamedAttribute> attributes) { ::mlir::OperationState __state__(location, getOperationName()); build(builder, __state__, inputs, outputs, attributes); auto __res__ = ::llvm::dyn_cast<AbsOp>(builder.create(__state__)); assert(__res__ && "builder didn't return the right type"); return __res__; } AbsOp AbsOp::create(::mlir::ImplicitLocOpBuilder &builder, ValueRange inputs, ValueRange outputs, ArrayRef<NamedAttribute> attributes) { return create(builder, builder.getLoc(), inputs, outputs, attributes); } ``` - Changed `EnzymeStatic-21` to `EnzymeStatic-22` - Mock `_ods_cext.globals.register_traceback_file_exclusion` due to API conflicts between Catalyst's MLIR version and the MLIR version used by JAX. The current JAX version we used has not yet updated to the latest MLIR, causing compatibility issues. This workaround will be removed once JAX updates to a compatible MLIR version. llvm/llvm-project#151246 TODO: - [x] Update .dep_version - [ ] Change all `rewrite.create` to `{op}::create` Even it’s compatible to the current `rewrite.create` method, we don’t need to change all `rewrite.create` to `{op}::create` right away, but this new feature provides more friendly development experience. Just for an example: llvm/llvm-project#147311 **Benefits:** **Possible Drawbacks:** **Related GitHub Issues:** [sc-100911] [sc-100912] --------- Co-authored-by: David Ittah <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
7 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
RFC: https://discourse.llvm.org/t/rfc-deprecate-linalg-elemwise-unary-and-elemwise-binary/87144
Remove the two operations and fix the tests by:
linalg.elemwise_{u|bi}narywithlinalg.{exp|add}on transform testslinalg.elementwiseinstead, to broaden test coverageelemwise_*part in the Python testslinalg.elementwiseinsteadNothing else changed.