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
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
|
// RUN: mlir-opt -int-range-optimizations -split-input-file %s | FileCheck %s
// CHECK-LABEL: func @launch_func
func.func @launch_func(%arg0 : index) {
%0 = test.with_bounds {
umin = 3 : index, umax = 5 : index,
smin = 3 : index, smax = 5 : index
} : index
%1 = test.with_bounds {
umin = 7 : index, umax = 11 : index,
smin = 7 : index, smax = 11 : index
} : index
gpu.launch blocks(%block_id_x, %block_id_y, %block_id_z) in (%grid_dim_x = %0, %grid_dim_y = %1, %grid_dim_z = %arg0)
threads(%thread_id_x, %thread_id_y, %thread_id_z) in (%block_dim_x = %arg0, %block_dim_y = %0, %block_dim_z = %1) {
// CHECK: test.reflect_bounds {smax = 5 : index, smin = 3 : index, umax = 5 : index, umin = 3 : index}
// CHECK: test.reflect_bounds {smax = 11 : index, smin = 7 : index, umax = 11 : index, umin = 7 : index}
// CHECK: test.reflect_bounds {smax = 4294967295 : index, smin = 1 : index, umax = 4294967295 : index, umin = 1 : index}
%grid_dim_x0 = test.reflect_bounds %grid_dim_x : index
%grid_dim_y0 = test.reflect_bounds %grid_dim_y : index
%grid_dim_z0 = test.reflect_bounds %grid_dim_z : index
// CHECK: test.reflect_bounds {smax = 4 : index, smin = 0 : index, umax = 4 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 10 : index, smin = 0 : index, umax = 10 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 4294967294 : index, smin = 0 : index, umax = 4294967294 : index, umin = 0 : index}
%block_id_x0 = test.reflect_bounds %block_id_x : index
%block_id_y0 = test.reflect_bounds %block_id_y : index
%block_id_z0 = test.reflect_bounds %block_id_z : index
// CHECK: test.reflect_bounds {smax = 4294967295 : index, smin = 1 : index, umax = 4294967295 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 5 : index, smin = 3 : index, umax = 5 : index, umin = 3 : index}
// CHECK: test.reflect_bounds {smax = 11 : index, smin = 7 : index, umax = 11 : index, umin = 7 : index}
%block_dim_x0 = test.reflect_bounds %block_dim_x : index
%block_dim_y0 = test.reflect_bounds %block_dim_y : index
%block_dim_z0 = test.reflect_bounds %block_dim_z : index
// CHECK: test.reflect_bounds {smax = 4294967294 : index, smin = 0 : index, umax = 4294967294 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 4 : index, smin = 0 : index, umax = 4 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 10 : index, smin = 0 : index, umax = 10 : index, umin = 0 : index}
%thread_id_x0 = test.reflect_bounds %thread_id_x : index
%thread_id_y0 = test.reflect_bounds %thread_id_y : index
%thread_id_z0 = test.reflect_bounds %thread_id_z : index
// The launch bounds are not constant, and so this can't infer anything
// CHECK: test.reflect_bounds {smax = 4294967294 : index, smin = 0 : index, umax = 4294967294 : index, umin = 0 : index}
%thread_id_op = gpu.thread_id y
%thread_id_op0 = test.reflect_bounds %thread_id_op : index
gpu.terminator
}
func.return
}
// -----
// CHECK-LABEL: func @kernel
module attributes {gpu.container_module} {
gpu.module @gpu_module {
llvm.func @kernel() attributes {gpu.kernel} {
%grid_dim_x = gpu.grid_dim x
%grid_dim_y = gpu.grid_dim y
%grid_dim_z = gpu.grid_dim z
// CHECK: test.reflect_bounds {smax = 4294967295 : index, smin = 1 : index, umax = 4294967295 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 4294967295 : index, smin = 1 : index, umax = 4294967295 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 4294967295 : index, smin = 1 : index, umax = 4294967295 : index, umin = 1 : index}
%grid_dim_x0 = test.reflect_bounds %grid_dim_x : index
%grid_dim_y0 = test.reflect_bounds %grid_dim_y : index
%grid_dim_z0 = test.reflect_bounds %grid_dim_z : index
%block_id_x = gpu.block_id x
%block_id_y = gpu.block_id y
%block_id_z = gpu.block_id z
// CHECK: test.reflect_bounds {smax = 4294967294 : index, smin = 0 : index, umax = 4294967294 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 4294967294 : index, smin = 0 : index, umax = 4294967294 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 4294967294 : index, smin = 0 : index, umax = 4294967294 : index, umin = 0 : index}
%block_id_x0 = test.reflect_bounds %block_id_x : index
%block_id_y0 = test.reflect_bounds %block_id_y : index
%block_id_z0 = test.reflect_bounds %block_id_z : index
%block_dim_x = gpu.block_dim x
%block_dim_y = gpu.block_dim y
%block_dim_z = gpu.block_dim z
// CHECK: test.reflect_bounds {smax = 4294967295 : index, smin = 1 : index, umax = 4294967295 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 4294967295 : index, smin = 1 : index, umax = 4294967295 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 4294967295 : index, smin = 1 : index, umax = 4294967295 : index, umin = 1 : index}
%block_dim_x0 = test.reflect_bounds %block_dim_x : index
%block_dim_y0 = test.reflect_bounds %block_dim_y : index
%block_dim_z0 = test.reflect_bounds %block_dim_z : index
%thread_id_x = gpu.thread_id x
%thread_id_y = gpu.thread_id y
%thread_id_z = gpu.thread_id z
// CHECK: test.reflect_bounds {smax = 4294967294 : index, smin = 0 : index, umax = 4294967294 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 4294967294 : index, smin = 0 : index, umax = 4294967294 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 4294967294 : index, smin = 0 : index, umax = 4294967294 : index, umin = 0 : index}
%thread_id_x0 = test.reflect_bounds %thread_id_x : index
%thread_id_y0 = test.reflect_bounds %thread_id_y : index
%thread_id_z0 = test.reflect_bounds %thread_id_z : index
%global_id_x = gpu.global_id x
%global_id_y = gpu.global_id y
%global_id_z = gpu.global_id z
// CHECK: test.reflect_bounds {smax = 9223372036854775807 : index, smin = -9223372036854775808 : index, umax = -8589934592 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 9223372036854775807 : index, smin = -9223372036854775808 : index, umax = -8589934592 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 9223372036854775807 : index, smin = -9223372036854775808 : index, umax = -8589934592 : index, umin = 0 : index}
%global_id_x0 = test.reflect_bounds %global_id_x : index
%global_id_y0 = test.reflect_bounds %global_id_y : index
%global_id_z0 = test.reflect_bounds %global_id_z : index
%subgroup_size = gpu.subgroup_size : index
%lane_id = gpu.lane_id
%num_subgroups = gpu.num_subgroups : index
%subgroup_id = gpu.subgroup_id : index
// CHECK: test.reflect_bounds {smax = 128 : index, smin = 1 : index, umax = 128 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 127 : index, smin = 0 : index, umax = 127 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 4294967295 : index, smin = 1 : index, umax = 4294967295 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 4294967294 : index, smin = 0 : index, umax = 4294967294 : index, umin = 0 : index}
%subgroup_size0 = test.reflect_bounds %subgroup_size : index
%lane_id0 = test.reflect_bounds %lane_id : index
%num_subgroups0 = test.reflect_bounds %num_subgroups : index
%subgroup_id0 = test.reflect_bounds %subgroup_id : index
llvm.return
}
}
}
// -----
// CHECK-LABEL: func @annotated_kernel
module attributes {gpu.container_module} {
gpu.module @gpu_module {
gpu.func @annotated_kernel() kernel
attributes {known_block_size = array<i32: 8, 12, 16>,
known_grid_size = array<i32: 20, 24, 28>} {
%grid_dim_x = gpu.grid_dim x
%grid_dim_y = gpu.grid_dim y
%grid_dim_z = gpu.grid_dim z
// CHECK: test.reflect_bounds {smax = 20 : index, smin = 20 : index, umax = 20 : index, umin = 20 : index}
// CHECK: test.reflect_bounds {smax = 24 : index, smin = 24 : index, umax = 24 : index, umin = 24 : index}
// CHECK: test.reflect_bounds {smax = 28 : index, smin = 28 : index, umax = 28 : index, umin = 28 : index}
%grid_dim_x0 = test.reflect_bounds %grid_dim_x : index
%grid_dim_y0 = test.reflect_bounds %grid_dim_y : index
%grid_dim_z0 = test.reflect_bounds %grid_dim_z : index
%block_id_x = gpu.block_id x
%block_id_y = gpu.block_id y
%block_id_z = gpu.block_id z
// CHECK: test.reflect_bounds {smax = 19 : index, smin = 0 : index, umax = 19 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 23 : index, smin = 0 : index, umax = 23 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 27 : index, smin = 0 : index, umax = 27 : index, umin = 0 : index}
%block_id_x0 = test.reflect_bounds %block_id_x : index
%block_id_y0 = test.reflect_bounds %block_id_y : index
%block_id_z0 = test.reflect_bounds %block_id_z : index
%block_dim_x = gpu.block_dim x
%block_dim_y = gpu.block_dim y
%block_dim_z = gpu.block_dim z
// CHECK: test.reflect_bounds {smax = 8 : index, smin = 8 : index, umax = 8 : index, umin = 8 : index}
// CHECK: test.reflect_bounds {smax = 12 : index, smin = 12 : index, umax = 12 : index, umin = 12 : index}
// CHECK: test.reflect_bounds {smax = 16 : index, smin = 16 : index, umax = 16 : index, umin = 16 : index}
%block_dim_x0 = test.reflect_bounds %block_dim_x : index
%block_dim_y0 = test.reflect_bounds %block_dim_y : index
%block_dim_z0 = test.reflect_bounds %block_dim_z : index
%thread_id_x = gpu.thread_id x
%thread_id_y = gpu.thread_id y
%thread_id_z = gpu.thread_id z
// CHECK: test.reflect_bounds {smax = 7 : index, smin = 0 : index, umax = 7 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 11 : index, smin = 0 : index, umax = 11 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 15 : index, smin = 0 : index, umax = 15 : index, umin = 0 : index}
%thread_id_x0 = test.reflect_bounds %thread_id_x : index
%thread_id_y0 = test.reflect_bounds %thread_id_y : index
%thread_id_z0 = test.reflect_bounds %thread_id_z : index
%global_id_x = gpu.global_id x
%global_id_y = gpu.global_id y
%global_id_z = gpu.global_id z
// CHECK: test.reflect_bounds {smax = 159 : index, smin = 0 : index, umax = 159 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 287 : index, smin = 0 : index, umax = 287 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 447 : index, smin = 0 : index, umax = 447 : index, umin = 0 : index}
%global_id_x0 = test.reflect_bounds %global_id_x : index
%global_id_y0 = test.reflect_bounds %global_id_y : index
%global_id_z0 = test.reflect_bounds %global_id_z : index
%subgroup_size = gpu.subgroup_size : index
%lane_id = gpu.lane_id
%num_subgroups = gpu.num_subgroups : index
%subgroup_id = gpu.subgroup_id : index
// CHECK: test.reflect_bounds {smax = 128 : index, smin = 1 : index, umax = 128 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 127 : index, smin = 0 : index, umax = 127 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 4294967295 : index, smin = 1 : index, umax = 4294967295 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 4294967294 : index, smin = 0 : index, umax = 4294967294 : index, umin = 0 : index}
%subgroup_size0 = test.reflect_bounds %subgroup_size : index
%lane_id0 = test.reflect_bounds %lane_id : index
%num_subgroups0 = test.reflect_bounds %num_subgroups : index
%subgroup_id0 = test.reflect_bounds %subgroup_id : index
gpu.return
}
}
}
// -----
// CHECK-LABEL: func @annotated_kernel
module {
func.func @annotated_kernel()
attributes {gpu.known_block_size = array<i32: 8, 12, 16>,
gpu.known_grid_size = array<i32: 20, 24, 28>} {
%block_id_x = gpu.block_id x
%block_id_y = gpu.block_id y
%block_id_z = gpu.block_id z
// CHECK: test.reflect_bounds {smax = 19 : index, smin = 0 : index, umax = 19 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 23 : index, smin = 0 : index, umax = 23 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 27 : index, smin = 0 : index, umax = 27 : index, umin = 0 : index}
%block_id_x0 = test.reflect_bounds %block_id_x : index
%block_id_y0 = test.reflect_bounds %block_id_y : index
%block_id_z0 = test.reflect_bounds %block_id_z : index
%thread_id_x = gpu.thread_id x
%thread_id_y = gpu.thread_id y
%thread_id_z = gpu.thread_id z
// CHECK: test.reflect_bounds {smax = 7 : index, smin = 0 : index, umax = 7 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 11 : index, smin = 0 : index, umax = 11 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 15 : index, smin = 0 : index, umax = 15 : index, umin = 0 : index}
%thread_id_x0 = test.reflect_bounds %thread_id_x : index
%thread_id_y0 = test.reflect_bounds %thread_id_y : index
%thread_id_z0 = test.reflect_bounds %thread_id_z : index
return
}
}
// -----
// CHECK-LABEL: func @local_bounds_kernel
module attributes {gpu.container_module} {
gpu.module @gpu_module {
gpu.func @local_bounds_kernel() kernel {
%grid_dim_x = gpu.grid_dim x upper_bound 20
%grid_dim_y = gpu.grid_dim y upper_bound 24
%grid_dim_z = gpu.grid_dim z upper_bound 28
// CHECK: test.reflect_bounds {smax = 20 : index, smin = 1 : index, umax = 20 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 24 : index, smin = 1 : index, umax = 24 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 28 : index, smin = 1 : index, umax = 28 : index, umin = 1 : index}
%grid_dim_x0 = test.reflect_bounds %grid_dim_x : index
%grid_dim_y0 = test.reflect_bounds %grid_dim_y : index
%grid_dim_z0 = test.reflect_bounds %grid_dim_z : index
%block_id_x = gpu.block_id x upper_bound 20
%block_id_y = gpu.block_id y upper_bound 24
%block_id_z = gpu.block_id z upper_bound 28
// CHECK: test.reflect_bounds {smax = 19 : index, smin = 0 : index, umax = 19 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 23 : index, smin = 0 : index, umax = 23 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 27 : index, smin = 0 : index, umax = 27 : index, umin = 0 : index}
%block_id_x0 = test.reflect_bounds %block_id_x : index
%block_id_y0 = test.reflect_bounds %block_id_y : index
%block_id_z0 = test.reflect_bounds %block_id_z : index
%block_dim_x = gpu.block_dim x upper_bound 8
%block_dim_y = gpu.block_dim y upper_bound 12
%block_dim_z = gpu.block_dim z upper_bound 16
// CHECK: test.reflect_bounds {smax = 8 : index, smin = 1 : index, umax = 8 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 12 : index, smin = 1 : index, umax = 12 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 16 : index, smin = 1 : index, umax = 16 : index, umin = 1 : index}
%block_dim_x0 = test.reflect_bounds %block_dim_x : index
%block_dim_y0 = test.reflect_bounds %block_dim_y : index
%block_dim_z0 = test.reflect_bounds %block_dim_z : index
%thread_id_x = gpu.thread_id x upper_bound 8
%thread_id_y = gpu.thread_id y upper_bound 12
%thread_id_z = gpu.thread_id z upper_bound 16
// CHECK: test.reflect_bounds {smax = 7 : index, smin = 0 : index, umax = 7 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 11 : index, smin = 0 : index, umax = 11 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 15 : index, smin = 0 : index, umax = 15 : index, umin = 0 : index}
%thread_id_x0 = test.reflect_bounds %thread_id_x : index
%thread_id_y0 = test.reflect_bounds %thread_id_y : index
%thread_id_z0 = test.reflect_bounds %thread_id_z : index
%global_id_x = gpu.global_id x upper_bound 160
%global_id_y = gpu.global_id y upper_bound 288
%global_id_z = gpu.global_id z upper_bound 448
// CHECK: test.reflect_bounds {smax = 159 : index, smin = 0 : index, umax = 159 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 287 : index, smin = 0 : index, umax = 287 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 447 : index, smin = 0 : index, umax = 447 : index, umin = 0 : index}
%global_id_x0 = test.reflect_bounds %global_id_x : index
%global_id_y0 = test.reflect_bounds %global_id_y : index
%global_id_z0 = test.reflect_bounds %global_id_z : index
%subgroup_size = gpu.subgroup_size upper_bound 32 : index
%subgroup_id = gpu.subgroup_id upper_bound 32 : index
%num_subgroups = gpu.num_subgroups upper_bound 8 : index
%lane_id = gpu.lane_id upper_bound 64
// CHECK: test.reflect_bounds {smax = 32 : index, smin = 1 : index, umax = 32 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 31 : index, smin = 0 : index, umax = 31 : index, umin = 0 : index}
// CHECK: test.reflect_bounds {smax = 8 : index, smin = 1 : index, umax = 8 : index, umin = 1 : index}
// CHECK: test.reflect_bounds {smax = 63 : index, smin = 0 : index, umax = 63 : index, umin = 0 : index}
%subgroup_size0 = test.reflect_bounds %subgroup_size : index
%subgroup_id0 = test.reflect_bounds %subgroup_id : index
%num_subgroups0 = test.reflect_bounds %num_subgroups : index
%lane_id0 = test.reflect_bounds %lane_id : index
gpu.return
}
}
}
|
