memory-attrbution.mlir
11 KB
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
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
// RUN: mlir-opt -allow-unregistered-dialect --convert-gpu-to-nvvm --split-input-file %s | FileCheck --check-prefix=NVVM %s
// RUN: mlir-opt -allow-unregistered-dialect --convert-gpu-to-rocdl --split-input-file %s | FileCheck --check-prefix=ROCDL %s
gpu.module @kernel {
// NVVM-LABEL: llvm.func @private
gpu.func @private(%arg0: f32) private(%arg1: memref<4xf32, 5>) {
// Allocate private memory inside the function.
// NVVM: %[[size:.*]] = llvm.mlir.constant(4 : i64) : !llvm.i64
// NVVM: %[[raw:.*]] = llvm.alloca %[[size]] x !llvm.float : (!llvm.i64) -> !llvm.ptr<float>
// ROCDL: %[[size:.*]] = llvm.mlir.constant(4 : i64) : !llvm.i64
// ROCDL: %[[raw:.*]] = llvm.alloca %[[size]] x !llvm.float : (!llvm.i64) -> !llvm.ptr<float, 5>
// Populate the memref descriptor.
// NVVM: %[[descr1:.*]] = llvm.mlir.undef : !llvm.struct<(ptr<float>, ptr<float>, i64, array<1 x i64>, array<1 x i64>)>
// NVVM: %[[descr2:.*]] = llvm.insertvalue %[[raw]], %[[descr1]][0]
// NVVM: %[[descr3:.*]] = llvm.insertvalue %[[raw]], %[[descr2]][1]
// NVVM: %[[c0:.*]] = llvm.mlir.constant(0 : index) : !llvm.i64
// NVVM: %[[descr4:.*]] = llvm.insertvalue %[[c0]], %[[descr3]][2]
// NVVM: %[[c4:.*]] = llvm.mlir.constant(4 : index) : !llvm.i64
// NVVM: %[[descr5:.*]] = llvm.insertvalue %[[c4]], %[[descr4]][3, 0]
// NVVM: %[[c1:.*]] = llvm.mlir.constant(1 : index) : !llvm.i64
// NVVM: %[[descr6:.*]] = llvm.insertvalue %[[c1]], %[[descr5]][4, 0]
// ROCDL: %[[descr1:.*]] = llvm.mlir.undef : !llvm.struct<(ptr<float, 5>, ptr<float, 5>, i64, array<1 x i64>, array<1 x i64>)>
// ROCDL: %[[descr2:.*]] = llvm.insertvalue %[[raw]], %[[descr1]][0]
// ROCDL: %[[descr3:.*]] = llvm.insertvalue %[[raw]], %[[descr2]][1]
// ROCDL: %[[c0:.*]] = llvm.mlir.constant(0 : index) : !llvm.i64
// ROCDL: %[[descr4:.*]] = llvm.insertvalue %[[c0]], %[[descr3]][2]
// ROCDL: %[[c4:.*]] = llvm.mlir.constant(4 : index) : !llvm.i64
// ROCDL: %[[descr5:.*]] = llvm.insertvalue %[[c4]], %[[descr4]][3, 0]
// ROCDL: %[[c1:.*]] = llvm.mlir.constant(1 : index) : !llvm.i64
// ROCDL: %[[descr6:.*]] = llvm.insertvalue %[[c1]], %[[descr5]][4, 0]
// "Store" lowering should work just as any other memref, only check that
// we emit some core instructions.
// NVVM: llvm.extractvalue %[[descr6:.*]]
// NVVM: llvm.getelementptr
// NVVM: llvm.store
// ROCDL: llvm.extractvalue %[[descr6:.*]]
// ROCDL: llvm.getelementptr
// ROCDL: llvm.store
%c0 = constant 0 : index
store %arg0, %arg1[%c0] : memref<4xf32, 5>
"terminator"() : () -> ()
}
}
// -----
gpu.module @kernel {
// Workgroup buffers are allocated as globals.
// NVVM: llvm.mlir.global internal @[[$buffer:.*]]()
// NVVM-SAME: addr_space = 3
// NVVM-SAME: !llvm.array<4 x float>
// ROCDL: llvm.mlir.global internal @[[$buffer:.*]]()
// ROCDL-SAME: addr_space = 3
// ROCDL-SAME: !llvm.array<4 x float>
// NVVM-LABEL: llvm.func @workgroup
// NVVM-SAME: {
// ROCDL-LABEL: llvm.func @workgroup
// ROCDL-SAME: {
gpu.func @workgroup(%arg0: f32) workgroup(%arg1: memref<4xf32, 3>) {
// Get the address of the first element in the global array.
// NVVM: %[[c0:.*]] = llvm.mlir.constant(0 : i32) : !llvm.i32
// NVVM: %[[addr:.*]] = llvm.mlir.addressof @[[$buffer]] : !llvm.ptr<array<4 x float>, 3>
// NVVM: %[[raw:.*]] = llvm.getelementptr %[[addr]][%[[c0]], %[[c0]]]
// NVVM-SAME: !llvm.ptr<float, 3>
// ROCDL: %[[c0:.*]] = llvm.mlir.constant(0 : i32) : !llvm.i32
// ROCDL: %[[addr:.*]] = llvm.mlir.addressof @[[$buffer]] : !llvm.ptr<array<4 x float>, 3>
// ROCDL: %[[raw:.*]] = llvm.getelementptr %[[addr]][%[[c0]], %[[c0]]]
// ROCDL-SAME: !llvm.ptr<float, 3>
// Populate the memref descriptor.
// NVVM: %[[descr1:.*]] = llvm.mlir.undef : !llvm.struct<(ptr<float, 3>, ptr<float, 3>, i64, array<1 x i64>, array<1 x i64>)>
// NVVM: %[[descr2:.*]] = llvm.insertvalue %[[raw]], %[[descr1]][0]
// NVVM: %[[descr3:.*]] = llvm.insertvalue %[[raw]], %[[descr2]][1]
// NVVM: %[[c0:.*]] = llvm.mlir.constant(0 : index) : !llvm.i64
// NVVM: %[[descr4:.*]] = llvm.insertvalue %[[c0]], %[[descr3]][2]
// NVVM: %[[c4:.*]] = llvm.mlir.constant(4 : index) : !llvm.i64
// NVVM: %[[descr5:.*]] = llvm.insertvalue %[[c4]], %[[descr4]][3, 0]
// NVVM: %[[c1:.*]] = llvm.mlir.constant(1 : index) : !llvm.i64
// NVVM: %[[descr6:.*]] = llvm.insertvalue %[[c1]], %[[descr5]][4, 0]
// ROCDL: %[[descr1:.*]] = llvm.mlir.undef : !llvm.struct<(ptr<float, 3>, ptr<float, 3>, i64, array<1 x i64>, array<1 x i64>)>
// ROCDL: %[[descr2:.*]] = llvm.insertvalue %[[raw]], %[[descr1]][0]
// ROCDL: %[[descr3:.*]] = llvm.insertvalue %[[raw]], %[[descr2]][1]
// ROCDL: %[[c0:.*]] = llvm.mlir.constant(0 : index) : !llvm.i64
// ROCDL: %[[descr4:.*]] = llvm.insertvalue %[[c0]], %[[descr3]][2]
// ROCDL: %[[c4:.*]] = llvm.mlir.constant(4 : index) : !llvm.i64
// ROCDL: %[[descr5:.*]] = llvm.insertvalue %[[c4]], %[[descr4]][3, 0]
// ROCDL: %[[c1:.*]] = llvm.mlir.constant(1 : index) : !llvm.i64
// ROCDL: %[[descr6:.*]] = llvm.insertvalue %[[c1]], %[[descr5]][4, 0]
// "Store" lowering should work just as any other memref, only check that
// we emit some core instructions.
// NVVM: llvm.extractvalue %[[descr6:.*]]
// NVVM: llvm.getelementptr
// NVVM: llvm.store
// ROCDL: llvm.extractvalue %[[descr6:.*]]
// ROCDL: llvm.getelementptr
// ROCDL: llvm.store
%c0 = constant 0 : index
store %arg0, %arg1[%c0] : memref<4xf32, 3>
"terminator"() : () -> ()
}
}
// -----
gpu.module @kernel {
// Check that the total size was computed correctly.
// NVVM: llvm.mlir.global internal @[[$buffer:.*]]()
// NVVM-SAME: addr_space = 3
// NVVM-SAME: !llvm.array<48 x float>
// ROCDL: llvm.mlir.global internal @[[$buffer:.*]]()
// ROCDL-SAME: addr_space = 3
// ROCDL-SAME: !llvm.array<48 x float>
// NVVM-LABEL: llvm.func @workgroup3d
// ROCDL-LABEL: llvm.func @workgroup3d
gpu.func @workgroup3d(%arg0: f32) workgroup(%arg1: memref<4x2x6xf32, 3>) {
// Get the address of the first element in the global array.
// NVVM: %[[c0:.*]] = llvm.mlir.constant(0 : i32) : !llvm.i32
// NVVM: %[[addr:.*]] = llvm.mlir.addressof @[[$buffer]] : !llvm.ptr<array<48 x float>, 3>
// NVVM: %[[raw:.*]] = llvm.getelementptr %[[addr]][%[[c0]], %[[c0]]]
// NVVM-SAME: !llvm.ptr<float, 3>
// ROCDL: %[[c0:.*]] = llvm.mlir.constant(0 : i32) : !llvm.i32
// ROCDL: %[[addr:.*]] = llvm.mlir.addressof @[[$buffer]] : !llvm.ptr<array<48 x float>, 3>
// ROCDL: %[[raw:.*]] = llvm.getelementptr %[[addr]][%[[c0]], %[[c0]]]
// ROCDL-SAME: !llvm.ptr<float, 3>
// Populate the memref descriptor.
// NVVM: %[[descr1:.*]] = llvm.mlir.undef : !llvm.struct<(ptr<float, 3>, ptr<float, 3>, i64, array<3 x i64>, array<3 x i64>)>
// NVVM: %[[descr2:.*]] = llvm.insertvalue %[[raw]], %[[descr1]][0]
// NVVM: %[[descr3:.*]] = llvm.insertvalue %[[raw]], %[[descr2]][1]
// NVVM: %[[c0:.*]] = llvm.mlir.constant(0 : index) : !llvm.i64
// NVVM: %[[descr4:.*]] = llvm.insertvalue %[[c0]], %[[descr3]][2]
// NVVM: %[[c4:.*]] = llvm.mlir.constant(4 : index) : !llvm.i64
// NVVM: %[[descr5:.*]] = llvm.insertvalue %[[c4]], %[[descr4]][3, 0]
// NVVM: %[[c12:.*]] = llvm.mlir.constant(12 : index) : !llvm.i64
// NVVM: %[[descr6:.*]] = llvm.insertvalue %[[c12]], %[[descr5]][4, 0]
// NVVM: %[[c2:.*]] = llvm.mlir.constant(2 : index) : !llvm.i64
// NVVM: %[[descr7:.*]] = llvm.insertvalue %[[c2]], %[[descr6]][3, 1]
// NVVM: %[[c6:.*]] = llvm.mlir.constant(6 : index) : !llvm.i64
// NVVM: %[[descr8:.*]] = llvm.insertvalue %[[c6]], %[[descr7]][4, 1]
// NVVM: %[[c6:.*]] = llvm.mlir.constant(6 : index) : !llvm.i64
// NVVM: %[[descr9:.*]] = llvm.insertvalue %[[c6]], %[[descr8]][3, 2]
// NVVM: %[[c1:.*]] = llvm.mlir.constant(1 : index) : !llvm.i64
// NVVM: %[[descr10:.*]] = llvm.insertvalue %[[c1]], %[[descr9]][4, 2]
// ROCDL: %[[descr1:.*]] = llvm.mlir.undef : !llvm.struct<(ptr<float, 3>, ptr<float, 3>, i64, array<3 x i64>, array<3 x i64>)>
// ROCDL: %[[descr2:.*]] = llvm.insertvalue %[[raw]], %[[descr1]][0]
// ROCDL: %[[descr3:.*]] = llvm.insertvalue %[[raw]], %[[descr2]][1]
// ROCDL: %[[c0:.*]] = llvm.mlir.constant(0 : index) : !llvm.i64
// ROCDL: %[[descr4:.*]] = llvm.insertvalue %[[c0]], %[[descr3]][2]
// ROCDL: %[[c4:.*]] = llvm.mlir.constant(4 : index) : !llvm.i64
// ROCDL: %[[descr5:.*]] = llvm.insertvalue %[[c4]], %[[descr4]][3, 0]
// ROCDL: %[[c12:.*]] = llvm.mlir.constant(12 : index) : !llvm.i64
// ROCDL: %[[descr6:.*]] = llvm.insertvalue %[[c12]], %[[descr5]][4, 0]
// ROCDL: %[[c2:.*]] = llvm.mlir.constant(2 : index) : !llvm.i64
// ROCDL: %[[descr7:.*]] = llvm.insertvalue %[[c2]], %[[descr6]][3, 1]
// ROCDL: %[[c6:.*]] = llvm.mlir.constant(6 : index) : !llvm.i64
// ROCDL: %[[descr8:.*]] = llvm.insertvalue %[[c6]], %[[descr7]][4, 1]
// ROCDL: %[[c6:.*]] = llvm.mlir.constant(6 : index) : !llvm.i64
// ROCDL: %[[descr9:.*]] = llvm.insertvalue %[[c6]], %[[descr8]][3, 2]
// ROCDL: %[[c1:.*]] = llvm.mlir.constant(1 : index) : !llvm.i64
// ROCDL: %[[descr10:.*]] = llvm.insertvalue %[[c1]], %[[descr9]][4, 2]
%c0 = constant 0 : index
store %arg0, %arg1[%c0,%c0,%c0] : memref<4x2x6xf32, 3>
"terminator"() : () -> ()
}
}
// -----
gpu.module @kernel {
// Check that several buffers are defined.
// NVVM: llvm.mlir.global internal @[[$buffer1:.*]]()
// NVVM-SAME: !llvm.array<1 x float>
// NVVM: llvm.mlir.global internal @[[$buffer2:.*]]()
// NVVM-SAME: !llvm.array<2 x float>
// ROCDL: llvm.mlir.global internal @[[$buffer1:.*]]()
// ROCDL-SAME: !llvm.array<1 x float>
// ROCDL: llvm.mlir.global internal @[[$buffer2:.*]]()
// ROCDL-SAME: !llvm.array<2 x float>
// NVVM-LABEL: llvm.func @multiple
// ROCDL-LABEL: llvm.func @multiple
gpu.func @multiple(%arg0: f32)
workgroup(%arg1: memref<1xf32, 3>, %arg2: memref<2xf32, 3>)
private(%arg3: memref<3xf32, 5>, %arg4: memref<4xf32, 5>) {
// Workgroup buffers.
// NVVM: llvm.mlir.addressof @[[$buffer1]]
// NVVM: llvm.mlir.addressof @[[$buffer2]]
// ROCDL: llvm.mlir.addressof @[[$buffer1]]
// ROCDL: llvm.mlir.addressof @[[$buffer2]]
// Private buffers.
// NVVM: %[[c3:.*]] = llvm.mlir.constant(3 : i64)
// NVVM: llvm.alloca %[[c3]] x !llvm.float : (!llvm.i64) -> !llvm.ptr<float>
// NVVM: %[[c4:.*]] = llvm.mlir.constant(4 : i64)
// NVVM: llvm.alloca %[[c4]] x !llvm.float : (!llvm.i64) -> !llvm.ptr<float>
// ROCDL: %[[c3:.*]] = llvm.mlir.constant(3 : i64)
// ROCDL: llvm.alloca %[[c3]] x !llvm.float : (!llvm.i64) -> !llvm.ptr<float, 5>
// ROCDL: %[[c4:.*]] = llvm.mlir.constant(4 : i64)
// ROCDL: llvm.alloca %[[c4]] x !llvm.float : (!llvm.i64) -> !llvm.ptr<float, 5>
%c0 = constant 0 : index
store %arg0, %arg1[%c0] : memref<1xf32, 3>
store %arg0, %arg2[%c0] : memref<2xf32, 3>
store %arg0, %arg3[%c0] : memref<3xf32, 5>
store %arg0, %arg4[%c0] : memref<4xf32, 5>
"terminator"() : () -> ()
}
}