1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
|
// RUN: mlir-opt %s -test-transform-dialect-interpreter -split-input-file | FileCheck %s
func.func @vectorize_dynamic_identity(%arg0: tensor<?xf32>,
%arg1: tensor<?xf32>,
%arg2: tensor<?xf32>) -> tensor<?xf32> {
%0 = linalg.generic { indexing_maps = [affine_map<(d0) -> (d0)>,
affine_map<(d0) -> (d0)>,
affine_map<(d0) -> (d0)>],
iterator_types = ["parallel"] }
ins(%arg0, %arg1 : tensor<?xf32>, tensor<?xf32>)
outs(%arg2 : tensor<?xf32>) {
^bb(%in0: f32, %in1: f32, %out: f32) :
%0 = arith.addf %in0, %in1 : f32
linalg.yield %0 : f32
} -> tensor<?xf32>
return %0 : tensor<?xf32>
}
// CHECK-LABEL: @vectorize_dynamic_identity
// CHECK: %[[VAL_3:.*]] = arith.constant 0 : index
// CHECK: %[[VAL_4:.*]] = tensor.dim %{{.*}}, %[[VAL_3]] : tensor<?xf32>
// CHECK: %[[VAL_7:.*]] = vector.create_mask %[[VAL_4]] : vector<[4]xi1>
// CHECK: %[[VAL_8:.*]] = vector.mask %[[VAL_7]] { vector.transfer_read %{{.*}} {in_bounds = [true]} : tensor<?xf32>, vector<[4]xf32> } : vector<[4]xi1> -> vector<[4]xf32>
// CHECK: %[[VAL_10:.*]] = vector.mask %[[VAL_7]] { vector.transfer_read %{{.*}} {in_bounds = [true]} : tensor<?xf32>, vector<[4]xf32> } : vector<[4]xi1> -> vector<[4]xf32>
// CHECK: %[[VAL_12:.*]] = vector.mask %[[VAL_7]] { vector.transfer_read %{{.*}} {in_bounds = [true]} : tensor<?xf32>, vector<[4]xf32> } : vector<[4]xi1> -> vector<[4]xf32>
// CHECK: %[[VAL_13:.*]] = arith.addf %[[VAL_8]], %[[VAL_10]] : vector<[4]xf32>
// CHECK: %[[VAL_14:.*]] = vector.mask %[[VAL_7]] { vector.transfer_write %{{.*}} {in_bounds = [true]} : vector<[4]xf32>, tensor<?xf32> } : vector<[4]xi1> -> tensor<?xf32>
transform.sequence failures(propagate) {
^bb1(%arg1: !transform.any_op):
%0 = transform.structured.match ops{["linalg.generic"]} in %arg1 : (!transform.any_op) -> !transform.any_op
transform.structured.masked_vectorize %0 vector_sizes [[4]] : !transform.any_op
}
// -----
func.func @vectorize_partial_dynamic_identity(%arg0: tensor<8x?xf32>,
%arg1: tensor<8x?xf32>,
%arg2: tensor<8x?xf32>) -> tensor<8x?xf32> {
%0 = linalg.generic { indexing_maps = [affine_map<(d0, d1) -> (d0, d1)>,
affine_map<(d0, d1) -> (d0, d1)>,
affine_map<(d0, d1) -> (d0, d1)>],
iterator_types = ["parallel", "parallel"] }
ins(%arg0, %arg1 : tensor<8x?xf32>, tensor<8x?xf32>)
outs(%arg2 : tensor<8x?xf32>) {
^bb(%in0: f32, %in1: f32, %out: f32) :
%0 = arith.addf %in0, %in1 : f32
linalg.yield %0 : f32
} -> tensor<8x?xf32>
return %0 : tensor<8x?xf32>
}
// CHECK-LABEL: func.func @vectorize_partial_dynamic_identity(
// CHECK-SAME: %[[VAL_0:.*]]: tensor<8x?xf32>, %[[VAL_1:.*]]: tensor<8x?xf32>, %[[VAL_2:.*]]: tensor<8x?xf32>) -> tensor<8x?xf32> {
// CHECK-DAG: %[[VAL_3:.*]] = arith.constant 1 : index
// CHECK-DAG: %[[VAL_4:.*]] = tensor.dim %[[VAL_0]], %[[VAL_3]] : tensor<8x?xf32>
// CHECK-DAG: %[[VAL_5:.*]] = arith.constant 0 : index
// CHECK-DAG: %[[VAL_6:.*]] = arith.constant 0.000000e+00 : f32
// CHECK-DAG: %[[VAL_7:.*]] = arith.constant 8 : index
// CHECK: %[[VAL_8:.*]] = vector.create_mask %[[VAL_7]], %[[VAL_4]] : vector<8x[32]xi1>
// CHECK: %[[VAL_9:.*]] = vector.mask %[[VAL_8]] { vector.transfer_read %[[VAL_0]][%[[VAL_5]], %[[VAL_5]]], %[[VAL_6]] {in_bounds = [true, true]} : tensor<8x?xf32>, vector<8x[32]xf32> } : vector<8x[32]xi1> -> vector<8x[32]xf32>
// CHECK: %[[VAL_10:.*]] = arith.constant 0.000000e+00 : f32
// CHECK: %[[VAL_11:.*]] = vector.mask %[[VAL_8]] { vector.transfer_read %[[VAL_1]][%[[VAL_5]], %[[VAL_5]]], %[[VAL_10]] {in_bounds = [true, true]} : tensor<8x?xf32>, vector<8x[32]xf32> } : vector<8x[32]xi1> -> vector<8x[32]xf32>
// CHECK: %[[VAL_12:.*]] = arith.constant 0.000000e+00 : f32
// CHECK: %[[VAL_13:.*]] = vector.mask %[[VAL_8]] { vector.transfer_read %[[VAL_2]][%[[VAL_5]], %[[VAL_5]]], %[[VAL_12]] {in_bounds = [true, true]} : tensor<8x?xf32>, vector<8x[32]xf32> } : vector<8x[32]xi1> -> vector<8x[32]xf32>
// CHECK: %[[VAL_14:.*]] = arith.addf %[[VAL_9]], %[[VAL_11]] : vector<8x[32]xf32>
// CHECK: %[[VAL_15:.*]] = arith.constant 0 : index
// CHECK: %[[VAL_16:.*]] = vector.mask %[[VAL_8]] { vector.transfer_write %[[VAL_14]], %[[VAL_2]][%[[VAL_15]], %[[VAL_15]]] {in_bounds = [true, true]} : vector<8x[32]xf32>, tensor<8x?xf32> } : vector<8x[32]xi1> -> tensor<8x?xf32>
transform.sequence failures(propagate) {
^bb1(%arg1: !transform.any_op):
%0 = transform.structured.match ops{["linalg.generic"]} in %arg1 : (!transform.any_op) -> !transform.any_op
transform.structured.masked_vectorize %0 vector_sizes [8, [32]] : !transform.any_op
}
// -----
func.func @vectorize_static_shape_with_mask(%arg0: tensor<8x30xf32>,
%arg1: tensor<8x30xf32>,
%arg2: tensor<8x30xf32>) -> tensor<8x30xf32> {
%0 = linalg.generic { indexing_maps = [affine_map<(d0, d1) -> (d0, d1)>,
affine_map<(d0, d1) -> (d0, d1)>,
affine_map<(d0, d1) -> (d0, d1)>],
iterator_types = ["parallel", "parallel"] }
ins(%arg0, %arg1 : tensor<8x30xf32>, tensor<8x30xf32>)
outs(%arg2 : tensor<8x30xf32>) {
^bb(%in0: f32, %in1: f32, %out: f32) :
%0 = arith.addf %in0, %in1 : f32
linalg.yield %0 : f32
} -> tensor<8x30xf32>
return %0 : tensor<8x30xf32>
}
// CHECK-LABEL: func.func @vectorize_static_shape_with_mask(
// CHECK-SAME: %[[VAL_0:.*]]: tensor<8x30xf32>, %[[VAL_1:.*]]: tensor<8x30xf32>, %[[VAL_2:.*]]: tensor<8x30xf32>) -> tensor<8x30xf32> {
// CHECK-DAG: %[[VAL_3:.*]] = arith.constant 0 : index
// CHECK-DAG: %[[VAL_4:.*]] = arith.constant 0.000000e+00 : f32
// CHECK-DAG: %[[VAL_5:.*]] = arith.constant 8 : index
// CHECK-DAG: %[[VAL_6:.*]] = arith.constant 30 : index
// CHECK: %[[VAL_7:.*]] = vector.create_mask %[[VAL_5]], %[[VAL_6]] : vector<8x[32]xi1>
// CHECK: %[[VAL_8:.*]] = vector.mask %[[VAL_7]] { vector.transfer_read %[[VAL_0]][%[[VAL_3]], %[[VAL_3]]], %[[VAL_4]] {in_bounds = [true, true]} : tensor<8x30xf32>, vector<8x[32]xf32> } : vector<8x[32]xi1> -> vector<8x[32]xf32>
// CHECK: %[[VAL_9:.*]] = arith.constant 0.000000e+00 : f32
// CHECK: %[[VAL_10:.*]] = vector.mask %[[VAL_7]] { vector.transfer_read %[[VAL_1]][%[[VAL_3]], %[[VAL_3]]], %[[VAL_9]] {in_bounds = [true, true]} : tensor<8x30xf32>, vector<8x[32]xf32> } : vector<8x[32]xi1> -> vector<8x[32]xf32>
// CHECK: %[[VAL_11:.*]] = arith.constant 0.000000e+00 : f32
// CHECK: %[[VAL_12:.*]] = vector.mask %[[VAL_7]] { vector.transfer_read %[[VAL_2]][%[[VAL_3]], %[[VAL_3]]], %[[VAL_11]] {in_bounds = [true, true]} : tensor<8x30xf32>, vector<8x[32]xf32> } : vector<8x[32]xi1> -> vector<8x[32]xf32>
// CHECK: %[[VAL_13:.*]] = arith.addf %[[VAL_8]], %[[VAL_10]] : vector<8x[32]xf32>
// CHECK: %[[VAL_14:.*]] = arith.constant 0 : index
// CHECK: %[[VAL_15:.*]] = vector.mask %[[VAL_7]] { vector.transfer_write %[[VAL_13]], %[[VAL_2]][%[[VAL_14]], %[[VAL_14]]] {in_bounds = [true, true]} : vector<8x[32]xf32>, tensor<8x30xf32> } : vector<8x[32]xi1> -> tensor<8x30xf32>
transform.sequence failures(propagate) {
^bb1(%arg1: !transform.any_op):
%0 = transform.structured.match ops{["linalg.generic"]} in %arg1 : (!transform.any_op) -> !transform.any_op
transform.structured.masked_vectorize %0 vector_sizes [8, [32]] : !transform.any_op
}
// -----
func.func @vectorize_dynamic_fill(%A : tensor<?x?xf32>, %arg0 : f32) -> tensor<?x?xf32> {
%0 = linalg.fill ins(%arg0 : f32) outs(%A : tensor<?x?xf32>) -> tensor<?x?xf32>
return %0 : tensor<?x?xf32>
}
// CHECK-LABEL: func.func @vectorize_dynamic_fill
// CHECK: %[[DIM0:.*]] = tensor.dim
// CHECK: %[[DIM1:.*]] = tensor.dim
// CHECK: %[[MASK:.*]] = vector.create_mask %[[DIM0]], %[[DIM1]] : vector<8x[16]xi1>
// CHECK: %[[BCAST:.*]] = vector.broadcast %{{.*}} : f32 to vector<8x[16]xf32>
// CHECK: vector.mask %[[MASK]] { vector.transfer_write %[[BCAST]], {{.*}} {in_bounds = [true, true]} : vector<8x[16]xf32>, tensor<?x?xf32> } : vector<8x[16]xi1>
transform.sequence failures(propagate) {
^bb1(%arg1: !transform.any_op):
%0 = transform.structured.match ops{["linalg.fill"]} in %arg1 : (!transform.any_op) -> !transform.any_op
transform.structured.masked_vectorize %0 vector_sizes [8, [16]] : !transform.any_op
}
|
