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
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
|
// FIXME: We should not be testing with -O2 (ie, a dependency on the entire IR optimizer).
// RUN: %clang_cc1 -ffreestanding %s -O2 -triple=x86_64-apple-darwin -target-cpu skylake-avx512 -emit-llvm -o - -Wall -Werror |opt -instnamer -S |FileCheck %s
#include <immintrin.h>
long long test_mm512_reduce_max_epi64(__m512i __W){
// CHECK: %shuffle1.i = shufflevector <8 x i64> %__W, <8 x i64> undef, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp = icmp slt <8 x i64> %shuffle1.i, %__W
// CHECK: %tmp1 = select <8 x i1> %tmp, <8 x i64> %__W, <8 x i64> %shuffle1.i
// CHECK: %shuffle3.i = shufflevector <8 x i64> %tmp1, <8 x i64> undef, <8 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = icmp sgt <8 x i64> %tmp1, %shuffle3.i
// CHECK: %tmp3 = select <8 x i1> %tmp2, <8 x i64> %tmp1, <8 x i64> %shuffle3.i
// CHECK: %shuffle6.i = shufflevector <8 x i64> %tmp3, <8 x i64> undef, <8 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp4 = icmp sgt <8 x i64> %tmp3, %shuffle6.i
// CHECK: %.elt.i = extractelement <8 x i1> %tmp4, i32 0
// CHECK: %.elt20.i = extractelement <8 x i64> %tmp3, i32 0
// CHECK: %shuffle6.elt.i = extractelement <8 x i64> %tmp3, i32 1
// CHECK: %vecext.i = select i1 %.elt.i, i64 %.elt20.i, i64 %shuffle6.elt.i
// CHECK: ret i64 %vecext.i
return _mm512_reduce_max_epi64(__W);
}
unsigned long long test_mm512_reduce_max_epu64(__m512i __W){
// CHECK: %shuffle1.i = shufflevector <8 x i64> %__W, <8 x i64> undef, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp = icmp ult <8 x i64> %shuffle1.i, %__W
// CHECK: %tmp1 = select <8 x i1> %tmp, <8 x i64> %__W, <8 x i64> %shuffle1.i
// CHECK: %shuffle3.i = shufflevector <8 x i64> %tmp1, <8 x i64> undef, <8 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = icmp ugt <8 x i64> %tmp1, %shuffle3.i
// CHECK: %tmp3 = select <8 x i1> %tmp2, <8 x i64> %tmp1, <8 x i64> %shuffle3.i
// CHECK: %shuffle6.i = shufflevector <8 x i64> %tmp3, <8 x i64> undef, <8 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp4 = icmp ugt <8 x i64> %tmp3, %shuffle6.i
// CHECK: %.elt.i = extractelement <8 x i1> %tmp4, i32 0
// CHECK: %.elt20.i = extractelement <8 x i64> %tmp3, i32 0
// CHECK: %shuffle6.elt.i = extractelement <8 x i64> %tmp3, i32 1
// CHECK: %vecext.i = select i1 %.elt.i, i64 %.elt20.i, i64 %shuffle6.elt.i
// CHECK: ret i64 %vecext.i
return _mm512_reduce_max_epu64(__W);
}
double test_mm512_reduce_max_pd(__m512d __W){
// CHECK: %shuffle1.i = shufflevector <8 x double> %__W, <8 x double> undef, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp = tail call <8 x double> @llvm.x86.avx512.mask.max.pd.512(<8 x double> %__W, <8 x double> %shuffle1.i, <8 x double> zeroinitializer, i8 -1, i32 4) #3
// CHECK: %shuffle3.i = shufflevector <8 x double> %tmp, <8 x double> undef, <8 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp1 = tail call <8 x double> @llvm.x86.avx512.mask.max.pd.512(<8 x double> %tmp, <8 x double> %shuffle3.i, <8 x double> zeroinitializer, i8 -1, i32 4) #3
// CHECK: %shuffle6.i = shufflevector <8 x double> %tmp1, <8 x double> undef, <8 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = tail call <8 x double> @llvm.x86.avx512.mask.max.pd.512(<8 x double> %tmp1, <8 x double> %shuffle6.i, <8 x double> zeroinitializer, i8 -1, i32 4) #3
// CHECK: %vecext.i = extractelement <8 x double> %tmp2, i32 0
// CHECK: ret double %vecext.i
return _mm512_reduce_max_pd(__W);
}
long long test_mm512_reduce_min_epi64(__m512i __W){
// CHECK: %shuffle1.i = shufflevector <8 x i64> %__W, <8 x i64> undef, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp = icmp slt <8 x i64> %shuffle1.i, %__W
// CHECK: %tmp1 = select <8 x i1> %tmp, <8 x i64> %__W, <8 x i64> %shuffle1.i
// CHECK: %shuffle3.i = shufflevector <8 x i64> %tmp1, <8 x i64> undef, <8 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = icmp sgt <8 x i64> %tmp1, %shuffle3.i
// CHECK: %tmp3 = select <8 x i1> %tmp2, <8 x i64> %tmp1, <8 x i64> %shuffle3.i
// CHECK: %shuffle6.i = shufflevector <8 x i64> %tmp3, <8 x i64> undef, <8 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp4 = icmp sgt <8 x i64> %tmp3, %shuffle6.i
// CHECK: %.elt.i = extractelement <8 x i1> %tmp4, i32 0
// CHECK: %.elt20.i = extractelement <8 x i64> %tmp3, i32 0
// CHECK: %shuffle6.elt.i = extractelement <8 x i64> %tmp3, i32 1
// CHECK: %vecext.i = select i1 %.elt.i, i64 %.elt20.i, i64 %shuffle6.elt.i
// CHECK: ret i64 %vecext.i
return _mm512_reduce_max_epi64(__W);
}
unsigned long long test_mm512_reduce_min_epu64(__m512i __W){
// CHECK: %shuffle1.i = shufflevector <8 x i64> %__W, <8 x i64> undef, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp = icmp ult <8 x i64> %shuffle1.i, %__W
// CHECK: %tmp1 = select <8 x i1> %tmp, <8 x i64> %__W, <8 x i64> %shuffle1.i
// CHECK: %shuffle3.i = shufflevector <8 x i64> %tmp1, <8 x i64> undef, <8 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = icmp ugt <8 x i64> %tmp1, %shuffle3.i
// CHECK: %tmp3 = select <8 x i1> %tmp2, <8 x i64> %tmp1, <8 x i64> %shuffle3.i
// CHECK: %shuffle6.i = shufflevector <8 x i64> %tmp3, <8 x i64> undef, <8 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp4 = icmp ugt <8 x i64> %tmp3, %shuffle6.i
// CHECK: %.elt.i = extractelement <8 x i1> %tmp4, i32 0
// CHECK: %.elt20.i = extractelement <8 x i64> %tmp3, i32 0
// CHECK: %shuffle6.elt.i = extractelement <8 x i64> %tmp3, i32 1
// CHECK: %vecext.i = select i1 %.elt.i, i64 %.elt20.i, i64 %shuffle6.elt.i
// CHECK: ret i64 %vecext.i
return _mm512_reduce_max_epu64(__W);
}
double test_mm512_reduce_min_pd(__m512d __W){
// CHECK: %shuffle1.i = shufflevector <8 x double> %__W, <8 x double> undef, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp = tail call <8 x double> @llvm.x86.avx512.mask.min.pd.512(<8 x double> %__W, <8 x double> %shuffle1.i, <8 x double> zeroinitializer, i8 -1, i32 4) #3
// CHECK: %shuffle3.i = shufflevector <8 x double> %tmp, <8 x double> undef, <8 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp1 = tail call <8 x double> @llvm.x86.avx512.mask.min.pd.512(<8 x double> %tmp, <8 x double> %shuffle3.i, <8 x double> zeroinitializer, i8 -1, i32 4) #3
// CHECK: %shuffle6.i = shufflevector <8 x double> %tmp1, <8 x double> undef, <8 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = tail call <8 x double> @llvm.x86.avx512.mask.min.pd.512(<8 x double> %tmp1, <8 x double> %shuffle6.i, <8 x double> zeroinitializer, i8 -1, i32 4) #3
// CHECK: %vecext.i = extractelement <8 x double> %tmp2, i32 0
// CHECK: ret double %vecext.i
return _mm512_reduce_min_pd(__W);
}
long long test_mm512_mask_reduce_max_epi64(__mmask8 __M, __m512i __W){
// CHECK: %tmp = bitcast i8 %__M to <8 x i1>
// CHECK: %tmp1 = select <8 x i1> %tmp, <8 x i64> %__W, <8 x i64> <i64 -9223372036854775808, i64 -9223372036854775808, i64 -9223372036854775808, i64 -9223372036854775808, i64 -9223372036854775808, i64 -9223372036854775808, i64 -9223372036854775808, i64 -9223372036854775808>
// CHECK: %shuffle1.i = shufflevector <8 x i64> %tmp1, <8 x i64> undef, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = icmp sgt <8 x i64> %tmp1, %shuffle1.i
// CHECK: %tmp3 = select <8 x i1> %tmp2, <8 x i64> %tmp1, <8 x i64> %shuffle1.i
// CHECK: %shuffle4.i = shufflevector <8 x i64> %tmp3, <8 x i64> undef, <8 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp4 = icmp sgt <8 x i64> %tmp3, %shuffle4.i
// CHECK: %tmp5 = select <8 x i1> %tmp4, <8 x i64> %tmp3, <8 x i64> %shuffle4.i
// CHECK: %shuffle7.i = shufflevector <8 x i64> %tmp5, <8 x i64> undef, <8 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp6 = icmp sgt <8 x i64> %tmp5, %shuffle7.i
// CHECK: %.elt.i = extractelement <8 x i1> %tmp6, i32 0
// CHECK: %.elt22.i = extractelement <8 x i64> %tmp5, i32 0
// CHECK: %shuffle7.elt.i = extractelement <8 x i64> %tmp5, i32 1
// CHECK: %vecext.i = select i1 %.elt.i, i64 %.elt22.i, i64 %shuffle7.elt.i
// CHECK: ret i64 %vecext.i
return _mm512_mask_reduce_max_epi64(__M, __W);
}
unsigned long test_mm512_mask_reduce_max_epu64(__mmask8 __M, __m512i __W){
// CHECK: %tmp = bitcast i8 %__M to <8 x i1>
// CHECK: %tmp1 = select <8 x i1> %tmp, <8 x i64> %__W, <8 x i64> zeroinitializer
// CHECK: %shuffle1.i = shufflevector <8 x i64> %tmp1, <8 x i64> undef, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = icmp ugt <8 x i64> %tmp1, %shuffle1.i
// CHECK: %tmp3 = select <8 x i1> %tmp2, <8 x i64> %tmp1, <8 x i64> %shuffle1.i
// CHECK: %shuffle4.i = shufflevector <8 x i64> %tmp3, <8 x i64> undef, <8 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp4 = icmp ugt <8 x i64> %tmp3, %shuffle4.i
// CHECK: %tmp5 = select <8 x i1> %tmp4, <8 x i64> %tmp3, <8 x i64> %shuffle4.i
// CHECK: %shuffle7.i = shufflevector <8 x i64> %tmp5, <8 x i64> undef, <8 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp6 = icmp ugt <8 x i64> %tmp5, %shuffle7.i
// CHECK: %.elt.i = extractelement <8 x i1> %tmp6, i32 0
// CHECK: %.elt22.i = extractelement <8 x i64> %tmp5, i32 0
// CHECK: %shuffle7.elt.i = extractelement <8 x i64> %tmp5, i32 1
// CHECK: %vecext.i = select i1 %.elt.i, i64 %.elt22.i, i64 %shuffle7.elt.i
// CHECK: ret i64 %vecext.i
return _mm512_mask_reduce_max_epu64(__M, __W);
}
long long test_mm512_mask_reduce_max_pd(__mmask8 __M, __m512d __W){
// CHECK: %tmp = bitcast i8 %__M to <8 x i1>
// CHECK: %tmp1 = select <8 x i1> %tmp, <8 x double> %__W, <8 x double> <double 0xFFF0000000000000, double 0xFFF0000000000000, double 0xFFF0000000000000, double 0xFFF0000000000000, double 0xFFF0000000000000, double 0xFFF0000000000000, double 0xFFF0000000000000, double 0xFFF0000000000000>
// CHECK: %shuffle1.i = shufflevector <8 x double> %tmp1, <8 x double> undef, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = tail call <8 x double> @llvm.x86.avx512.mask.max.pd.512(<8 x double> %tmp1, <8 x double> %shuffle1.i, <8 x double> zeroinitializer, i8 -1, i32 4) #3
// CHECK: %shuffle4.i = shufflevector <8 x double> %tmp2, <8 x double> undef, <8 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp3 = tail call <8 x double> @llvm.x86.avx512.mask.max.pd.512(<8 x double> %tmp2, <8 x double> %shuffle4.i, <8 x double> zeroinitializer, i8 -1, i32 4) #3
// CHECK: %shuffle7.i = shufflevector <8 x double> %tmp3, <8 x double> undef, <8 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp4 = tail call <8 x double> @llvm.x86.avx512.mask.max.pd.512(<8 x double> %tmp3, <8 x double> %shuffle7.i, <8 x double> zeroinitializer, i8 -1, i32 4) #3
// CHECK: %vecext.i = extractelement <8 x double> %tmp4, i32 0
// CHECK: %conv = fptosi double %vecext.i to i64
// CHECK: ret i64 %conv
return _mm512_mask_reduce_max_pd(__M, __W);
}
long long test_mm512_mask_reduce_min_epi64(__mmask8 __M, __m512i __W){
// CHECK: %tmp = bitcast i8 %__M to <8 x i1>
// CHECK: %tmp1 = select <8 x i1> %tmp, <8 x i64> %__W, <8 x i64> <i64 9223372036854775807, i64 9223372036854775807, i64 9223372036854775807, i64 9223372036854775807, i64 9223372036854775807, i64 9223372036854775807, i64 9223372036854775807, i64 9223372036854775807>
// CHECK: %shuffle1.i = shufflevector <8 x i64> %tmp1, <8 x i64> undef, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = icmp slt <8 x i64> %tmp1, %shuffle1.i
// CHECK: %tmp3 = select <8 x i1> %tmp2, <8 x i64> %tmp1, <8 x i64> %shuffle1.i
// CHECK: %shuffle4.i = shufflevector <8 x i64> %tmp3, <8 x i64> undef, <8 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp4 = icmp slt <8 x i64> %tmp3, %shuffle4.i
// CHECK: %tmp5 = select <8 x i1> %tmp4, <8 x i64> %tmp3, <8 x i64> %shuffle4.i
// CHECK: %shuffle7.i = shufflevector <8 x i64> %tmp5, <8 x i64> undef, <8 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp6 = icmp slt <8 x i64> %tmp5, %shuffle7.i
// CHECK: %.elt.i = extractelement <8 x i1> %tmp6, i32 0
// CHECK: %.elt22.i = extractelement <8 x i64> %tmp5, i32 0
// CHECK: %shuffle7.elt.i = extractelement <8 x i64> %tmp5, i32 1
// CHECK: %vecext.i = select i1 %.elt.i, i64 %.elt22.i, i64 %shuffle7.elt.i
// CHECK: ret i64 %vecext.i
return _mm512_mask_reduce_min_epi64(__M, __W);
}
long long test_mm512_mask_reduce_min_epu64(__mmask8 __M, __m512i __W){
// CHECK: %tmp = bitcast i8 %__M to <8 x i1>
// CHECK: %tmp1 = select <8 x i1> %tmp, <8 x i64> %__W, <8 x i64> zeroinitializer
// CHECK: %shuffle1.i = shufflevector <8 x i64> %tmp1, <8 x i64> undef, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = icmp ugt <8 x i64> %tmp1, %shuffle1.i
// CHECK: %tmp3 = select <8 x i1> %tmp2, <8 x i64> %tmp1, <8 x i64> %shuffle1.i
// CHECK: %shuffle4.i = shufflevector <8 x i64> %tmp3, <8 x i64> undef, <8 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp4 = icmp ugt <8 x i64> %tmp3, %shuffle4.i
// CHECK: %tmp5 = select <8 x i1> %tmp4, <8 x i64> %tmp3, <8 x i64> %shuffle4.i
// CHECK: %shuffle7.i = shufflevector <8 x i64> %tmp5, <8 x i64> undef, <8 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp6 = icmp ugt <8 x i64> %tmp5, %shuffle7.i
// CHECK: %.elt.i = extractelement <8 x i1> %tmp6, i32 0
// CHECK: %.elt22.i = extractelement <8 x i64> %tmp5, i32 0
// CHECK: %shuffle7.elt.i = extractelement <8 x i64> %tmp5, i32 1
// CHECK: %vecext.i = select i1 %.elt.i, i64 %.elt22.i, i64 %shuffle7.elt.i
// CHECK: ret i64 %vecext.i
return _mm512_mask_reduce_max_epu64(__M, __W);
}
double test_mm512_mask_reduce_min_pd(__mmask8 __M, __m512d __W){
// CHECK: %tmp = bitcast i8 %__M to <8 x i1>
// CHECK: %tmp1 = select <8 x i1> %tmp, <8 x double> %__W, <8 x double> <double 0x7FF0000000000000, double 0x7FF0000000000000, double 0x7FF0000000000000, double 0x7FF0000000000000, double 0x7FF0000000000000, double 0x7FF0000000000000, double 0x7FF0000000000000, double 0x7FF0000000000000>
// CHECK: %shuffle1.i = shufflevector <8 x double> %tmp1, <8 x double> undef, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = tail call <8 x double> @llvm.x86.avx512.mask.min.pd.512(<8 x double> %tmp1, <8 x double> %shuffle1.i, <8 x double> zeroinitializer, i8 -1, i32 4) #3
// CHECK: %shuffle4.i = shufflevector <8 x double> %tmp2, <8 x double> undef, <8 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp3 = tail call <8 x double> @llvm.x86.avx512.mask.min.pd.512(<8 x double> %tmp2, <8 x double> %shuffle4.i, <8 x double> zeroinitializer, i8 -1, i32 4) #3
// CHECK: %shuffle7.i = shufflevector <8 x double> %tmp3, <8 x double> undef, <8 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp4 = tail call <8 x double> @llvm.x86.avx512.mask.min.pd.512(<8 x double> %tmp3, <8 x double> %shuffle7.i, <8 x double> zeroinitializer, i8 -1, i32 4) #3
// CHECK: %vecext.i = extractelement <8 x double> %tmp4, i32 0
// CHECK: ret double %vecext.i
return _mm512_mask_reduce_min_pd(__M, __W);
}
int test_mm512_reduce_max_epi32(__m512i __W){
// CHECK: %tmp = bitcast <8 x i64> %__W to <16 x i32>
// CHECK: %shuffle1.i = shufflevector <16 x i32> %tmp, <16 x i32> undef, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp1 = icmp slt <16 x i32> %shuffle1.i, %tmp
// CHECK: %tmp2 = select <16 x i1> %tmp1, <16 x i32> %tmp, <16 x i32> %shuffle1.i
// CHECK: %shuffle3.i = shufflevector <16 x i32> %tmp2, <16 x i32> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp3 = icmp sgt <16 x i32> %tmp2, %shuffle3.i
// CHECK: %tmp4 = select <16 x i1> %tmp3, <16 x i32> %tmp2, <16 x i32> %shuffle3.i
// CHECK: %shuffle6.i = shufflevector <16 x i32> %tmp4, <16 x i32> undef, <16 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp5 = icmp sgt <16 x i32> %tmp4, %shuffle6.i
// CHECK: %tmp6 = select <16 x i1> %tmp5, <16 x i32> %tmp4, <16 x i32> %shuffle6.i
// CHECK: %shuffle9.i = shufflevector <16 x i32> %tmp6, <16 x i32> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp7 = icmp sgt <16 x i32> %tmp6, %shuffle9.i
// CHECK: %tmp8 = select <16 x i1> %tmp7, <16 x i32> %tmp6, <16 x i32> %shuffle9.i
// CHECK: %tmp9 = bitcast <16 x i32> %tmp8 to <8 x i64>
// CHECK: %vecext.i = extractelement <8 x i64> %tmp9, i32 0
// CHECK: %conv.i = trunc i64 %vecext.i to i32
// CHECK: ret i32 %conv.i
return _mm512_reduce_max_epi32(__W);
}
unsigned int test_mm512_reduce_max_epu32(__m512i __W){
// CHECK: %tmp = bitcast <8 x i64> %__W to <16 x i32>
// CHECK: %shuffle1.i = shufflevector <16 x i32> %tmp, <16 x i32> undef, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp1 = icmp ult <16 x i32> %shuffle1.i, %tmp
// CHECK: %tmp2 = select <16 x i1> %tmp1, <16 x i32> %tmp, <16 x i32> %shuffle1.i
// CHECK: %shuffle3.i = shufflevector <16 x i32> %tmp2, <16 x i32> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp3 = icmp ugt <16 x i32> %tmp2, %shuffle3.i
// CHECK: %tmp4 = select <16 x i1> %tmp3, <16 x i32> %tmp2, <16 x i32> %shuffle3.i
// CHECK: %shuffle6.i = shufflevector <16 x i32> %tmp4, <16 x i32> undef, <16 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp5 = icmp ugt <16 x i32> %tmp4, %shuffle6.i
// CHECK: %tmp6 = select <16 x i1> %tmp5, <16 x i32> %tmp4, <16 x i32> %shuffle6.i
// CHECK: %shuffle9.i = shufflevector <16 x i32> %tmp6, <16 x i32> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp7 = icmp ugt <16 x i32> %tmp6, %shuffle9.i
// CHECK: %tmp8 = select <16 x i1> %tmp7, <16 x i32> %tmp6, <16 x i32> %shuffle9.i
// CHECK: %tmp9 = bitcast <16 x i32> %tmp8 to <8 x i64>
// CHECK: %vecext.i = extractelement <8 x i64> %tmp9, i32 0
// CHECK: %conv.i = trunc i64 %vecext.i to i32
// CHECK: ret i32 %conv.i
return _mm512_reduce_max_epu32(__W);
}
float test_mm512_reduce_max_ps(__m512 __W){
// CHECK: %shuffle1.i = shufflevector <16 x float> %__W, <16 x float> undef, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp = tail call <16 x float> @llvm.x86.avx512.mask.max.ps.512(<16 x float> %__W, <16 x float> %shuffle1.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %shuffle3.i = shufflevector <16 x float> %tmp, <16 x float> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp1 = tail call <16 x float> @llvm.x86.avx512.mask.max.ps.512(<16 x float> %tmp, <16 x float> %shuffle3.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %shuffle6.i = shufflevector <16 x float> %tmp1, <16 x float> undef, <16 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = tail call <16 x float> @llvm.x86.avx512.mask.max.ps.512(<16 x float> %tmp1, <16 x float> %shuffle6.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %shuffle9.i = shufflevector <16 x float> %tmp2, <16 x float> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp3 = tail call <16 x float> @llvm.x86.avx512.mask.max.ps.512(<16 x float> %tmp2, <16 x float> %shuffle9.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %vecext.i = extractelement <16 x float> %tmp3, i32 0
// CHECK: ret float %vecext.i
return _mm512_reduce_max_ps(__W);
}
int test_mm512_reduce_min_epi32(__m512i __W){
// CHECK: %tmp = bitcast <8 x i64> %__W to <16 x i32>
// CHECK: %shuffle1.i = shufflevector <16 x i32> %tmp, <16 x i32> undef, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp1 = icmp sgt <16 x i32> %shuffle1.i, %tmp
// CHECK: %tmp2 = select <16 x i1> %tmp1, <16 x i32> %tmp, <16 x i32> %shuffle1.i
// CHECK: %shuffle3.i = shufflevector <16 x i32> %tmp2, <16 x i32> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp3 = icmp slt <16 x i32> %tmp2, %shuffle3.i
// CHECK: %tmp4 = select <16 x i1> %tmp3, <16 x i32> %tmp2, <16 x i32> %shuffle3.i
// CHECK: %shuffle6.i = shufflevector <16 x i32> %tmp4, <16 x i32> undef, <16 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp5 = icmp slt <16 x i32> %tmp4, %shuffle6.i
// CHECK: %tmp6 = select <16 x i1> %tmp5, <16 x i32> %tmp4, <16 x i32> %shuffle6.i
// CHECK: %shuffle9.i = shufflevector <16 x i32> %tmp6, <16 x i32> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp7 = icmp slt <16 x i32> %tmp6, %shuffle9.i
// CHECK: %tmp8 = select <16 x i1> %tmp7, <16 x i32> %tmp6, <16 x i32> %shuffle9.i
// CHECK: %tmp9 = bitcast <16 x i32> %tmp8 to <8 x i64>
// CHECK: %vecext.i = extractelement <8 x i64> %tmp9, i32 0
// CHECK: %conv.i = trunc i64 %vecext.i to i32
// CHECK: ret i32 %conv.i
return _mm512_reduce_min_epi32(__W);
}
unsigned int test_mm512_reduce_min_epu32(__m512i __W){
// CHECK: %tmp = bitcast <8 x i64> %__W to <16 x i32>
// CHECK: %shuffle1.i = shufflevector <16 x i32> %tmp, <16 x i32> undef, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp1 = icmp ugt <16 x i32> %shuffle1.i, %tmp
// CHECK: %tmp2 = select <16 x i1> %tmp1, <16 x i32> %tmp, <16 x i32> %shuffle1.i
// CHECK: %shuffle3.i = shufflevector <16 x i32> %tmp2, <16 x i32> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp3 = icmp ult <16 x i32> %tmp2, %shuffle3.i
// CHECK: %tmp4 = select <16 x i1> %tmp3, <16 x i32> %tmp2, <16 x i32> %shuffle3.i
// CHECK: %shuffle6.i = shufflevector <16 x i32> %tmp4, <16 x i32> undef, <16 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp5 = icmp ult <16 x i32> %tmp4, %shuffle6.i
// CHECK: %tmp6 = select <16 x i1> %tmp5, <16 x i32> %tmp4, <16 x i32> %shuffle6.i
// CHECK: %shuffle9.i = shufflevector <16 x i32> %tmp6, <16 x i32> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp7 = icmp ult <16 x i32> %tmp6, %shuffle9.i
// CHECK: %tmp8 = select <16 x i1> %tmp7, <16 x i32> %tmp6, <16 x i32> %shuffle9.i
// CHECK: %tmp9 = bitcast <16 x i32> %tmp8 to <8 x i64>
// CHECK: %vecext.i = extractelement <8 x i64> %tmp9, i32 0
// CHECK: %conv.i = trunc i64 %vecext.i to i32
// CHECK: ret i32 %conv.i
return _mm512_reduce_min_epu32(__W);
}
float test_mm512_reduce_min_ps(__m512 __W){
// CHECK: %shuffle1.i = shufflevector <16 x float> %__W, <16 x float> undef, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp = tail call <16 x float> @llvm.x86.avx512.mask.min.ps.512(<16 x float> %__W, <16 x float> %shuffle1.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %shuffle3.i = shufflevector <16 x float> %tmp, <16 x float> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp1 = tail call <16 x float> @llvm.x86.avx512.mask.min.ps.512(<16 x float> %tmp, <16 x float> %shuffle3.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %shuffle6.i = shufflevector <16 x float> %tmp1, <16 x float> undef, <16 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = tail call <16 x float> @llvm.x86.avx512.mask.min.ps.512(<16 x float> %tmp1, <16 x float> %shuffle6.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %shuffle9.i = shufflevector <16 x float> %tmp2, <16 x float> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp3 = tail call <16 x float> @llvm.x86.avx512.mask.min.ps.512(<16 x float> %tmp2, <16 x float> %shuffle9.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %vecext.i = extractelement <16 x float> %tmp3, i32 0
// CHECK: ret float %vecext.i
return _mm512_reduce_min_ps(__W);
}
int test_mm512_mask_reduce_max_epi32(__mmask16 __M, __m512i __W){
// CHECK: %tmp = bitcast <8 x i64> %__W to <16 x i32>
// CHECK: %tmp1 = bitcast i16 %__M to <16 x i1>
// CHECK: %tmp2 = select <16 x i1> %tmp1, <16 x i32> %tmp, <16 x i32> <i32 -2147483648, i32 -2147483648, i32 -2147483648, i32 -2147483648, i32 -2147483648, i32 -2147483648, i32 -2147483648, i32 -2147483648, i32 -2147483648, i32 -2147483648, i32 -2147483648, i32 -2147483648, i32 -2147483648, i32 -2147483648, i32 -2147483648, i32 -2147483648>
// CHECK: %shuffle1.i = shufflevector <16 x i32> %tmp2, <16 x i32> undef, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp3 = icmp sgt <16 x i32> %tmp2, %shuffle1.i
// CHECK: %tmp4 = select <16 x i1> %tmp3, <16 x i32> %tmp2, <16 x i32> %shuffle1.i
// CHECK: %shuffle4.i = shufflevector <16 x i32> %tmp4, <16 x i32> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp5 = icmp sgt <16 x i32> %tmp4, %shuffle4.i
// CHECK: %tmp6 = select <16 x i1> %tmp5, <16 x i32> %tmp4, <16 x i32> %shuffle4.i
// CHECK: %shuffle7.i = shufflevector <16 x i32> %tmp6, <16 x i32> undef, <16 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp7 = icmp sgt <16 x i32> %tmp6, %shuffle7.i
// CHECK: %tmp8 = select <16 x i1> %tmp7, <16 x i32> %tmp6, <16 x i32> %shuffle7.i
// CHECK: %shuffle10.i = shufflevector <16 x i32> %tmp8, <16 x i32> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp9 = icmp sgt <16 x i32> %tmp8, %shuffle10.i
// CHECK: %tmp10 = select <16 x i1> %tmp9, <16 x i32> %tmp8, <16 x i32> %shuffle10.i
// CHECK: %tmp11 = bitcast <16 x i32> %tmp10 to <8 x i64>
// CHECK: %vecext.i = extractelement <8 x i64> %tmp11, i32 0
// CHECK: %conv.i = trunc i64 %vecext.i to i32
// CHECK: ret i32 %conv.i
return _mm512_mask_reduce_max_epi32(__M, __W);
}
unsigned int test_mm512_mask_reduce_max_epu32(__mmask16 __M, __m512i __W){
// CHECK: %tmp = bitcast <8 x i64> %__W to <16 x i32>
// CHECK: %tmp1 = bitcast i16 %__M to <16 x i1>
// CHECK: %tmp2 = select <16 x i1> %tmp1, <16 x i32> %tmp, <16 x i32> zeroinitializer
// CHECK: %shuffle1.i = shufflevector <16 x i32> %tmp2, <16 x i32> undef, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp3 = icmp ugt <16 x i32> %tmp2, %shuffle1.i
// CHECK: %tmp4 = select <16 x i1> %tmp3, <16 x i32> %tmp2, <16 x i32> %shuffle1.i
// CHECK: %shuffle4.i = shufflevector <16 x i32> %tmp4, <16 x i32> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp5 = icmp ugt <16 x i32> %tmp4, %shuffle4.i
// CHECK: %tmp6 = select <16 x i1> %tmp5, <16 x i32> %tmp4, <16 x i32> %shuffle4.i
// CHECK: %shuffle7.i = shufflevector <16 x i32> %tmp6, <16 x i32> undef, <16 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp7 = icmp ugt <16 x i32> %tmp6, %shuffle7.i
// CHECK: %tmp8 = select <16 x i1> %tmp7, <16 x i32> %tmp6, <16 x i32> %shuffle7.i
// CHECK: %shuffle10.i = shufflevector <16 x i32> %tmp8, <16 x i32> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp9 = icmp ugt <16 x i32> %tmp8, %shuffle10.i
// CHECK: %tmp10 = select <16 x i1> %tmp9, <16 x i32> %tmp8, <16 x i32> %shuffle10.i
// CHECK: %tmp11 = bitcast <16 x i32> %tmp10 to <8 x i64>
// CHECK: %vecext.i = extractelement <8 x i64> %tmp11, i32 0
// CHECK: %conv.i = trunc i64 %vecext.i to i32
// CHECK: ret i32 %conv.i
return _mm512_mask_reduce_max_epu32(__M, __W);
}
float test_mm512_mask_reduce_max_ps(__mmask16 __M, __m512 __W){
// CHECK: %tmp = bitcast i16 %__M to <16 x i1>
// CHECK: %tmp1 = select <16 x i1> %tmp, <16 x float> %__W, <16 x float> <float 0xFFF0000000000000, float 0xFFF0000000000000, float 0xFFF0000000000000, float 0xFFF0000000000000, float 0xFFF0000000000000, float 0xFFF0000000000000, float 0xFFF0000000000000, float 0xFFF0000000000000, float 0xFFF0000000000000, float 0xFFF0000000000000, float 0xFFF0000000000000, float 0xFFF0000000000000, float 0xFFF0000000000000, float 0xFFF0000000000000, float 0xFFF0000000000000, float 0xFFF0000000000000>
// CHECK: %shuffle1.i = shufflevector <16 x float> %tmp1, <16 x float> undef, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = tail call <16 x float> @llvm.x86.avx512.mask.max.ps.512(<16 x float> %tmp1, <16 x float> %shuffle1.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %shuffle4.i = shufflevector <16 x float> %tmp2, <16 x float> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp3 = tail call <16 x float> @llvm.x86.avx512.mask.max.ps.512(<16 x float> %tmp2, <16 x float> %shuffle4.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %shuffle7.i = shufflevector <16 x float> %tmp3, <16 x float> undef, <16 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp4 = tail call <16 x float> @llvm.x86.avx512.mask.max.ps.512(<16 x float> %tmp3, <16 x float> %shuffle7.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %shuffle10.i = shufflevector <16 x float> %tmp4, <16 x float> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp5 = tail call <16 x float> @llvm.x86.avx512.mask.max.ps.512(<16 x float> %tmp4, <16 x float> %shuffle10.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %vecext.i = extractelement <16 x float> %tmp5, i32 0
// CHECK: ret float %vecext.i
return _mm512_mask_reduce_max_ps(__M, __W);
}
int test_mm512_mask_reduce_min_epi32(__mmask16 __M, __m512i __W){
// CHECK: %tmp = bitcast <8 x i64> %__W to <16 x i32>
// CHECK: %tmp1 = bitcast i16 %__M to <16 x i1>
// CHECK: %tmp2 = select <16 x i1> %tmp1, <16 x i32> %tmp, <16 x i32> <i32 2147483647, i32 2147483647, i32 2147483647, i32 2147483647, i32 2147483647, i32 2147483647, i32 2147483647, i32 2147483647, i32 2147483647, i32 2147483647, i32 2147483647, i32 2147483647, i32 2147483647, i32 2147483647, i32 2147483647, i32 2147483647>
// CHECK: %shuffle1.i = shufflevector <16 x i32> %tmp2, <16 x i32> undef, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp3 = icmp slt <16 x i32> %tmp2, %shuffle1.i
// CHECK: %tmp4 = select <16 x i1> %tmp3, <16 x i32> %tmp2, <16 x i32> %shuffle1.i
// CHECK: %shuffle4.i = shufflevector <16 x i32> %tmp4, <16 x i32> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp5 = icmp slt <16 x i32> %tmp4, %shuffle4.i
// CHECK: %tmp6 = select <16 x i1> %tmp5, <16 x i32> %tmp4, <16 x i32> %shuffle4.i
// CHECK: %shuffle7.i = shufflevector <16 x i32> %tmp6, <16 x i32> undef, <16 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp7 = icmp slt <16 x i32> %tmp6, %shuffle7.i
// CHECK: %tmp8 = select <16 x i1> %tmp7, <16 x i32> %tmp6, <16 x i32> %shuffle7.i
// CHECK: %shuffle10.i = shufflevector <16 x i32> %tmp8, <16 x i32> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp9 = icmp slt <16 x i32> %tmp8, %shuffle10.i
// CHECK: %tmp10 = select <16 x i1> %tmp9, <16 x i32> %tmp8, <16 x i32> %shuffle10.i
// CHECK: %tmp11 = bitcast <16 x i32> %tmp10 to <8 x i64>
// CHECK: %vecext.i = extractelement <8 x i64> %tmp11, i32 0
// CHECK: %conv.i = trunc i64 %vecext.i to i32
// CHECK: ret i32 %conv.i
return _mm512_mask_reduce_min_epi32(__M, __W);
}
unsigned int test_mm512_mask_reduce_min_epu32(__mmask16 __M, __m512i __W){
// CHECK: %tmp = bitcast <8 x i64> %__W to <16 x i32>
// CHECK: %tmp1 = bitcast i16 %__M to <16 x i1>
// CHECK: %tmp2 = select <16 x i1> %tmp1, <16 x i32> %tmp, <16 x i32> <i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1>
// CHECK: %shuffle1.i = shufflevector <16 x i32> %tmp2, <16 x i32> undef, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp3 = icmp ult <16 x i32> %tmp2, %shuffle1.i
// CHECK: %tmp4 = select <16 x i1> %tmp3, <16 x i32> %tmp2, <16 x i32> %shuffle1.i
// CHECK: %shuffle4.i = shufflevector <16 x i32> %tmp4, <16 x i32> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp5 = icmp ult <16 x i32> %tmp4, %shuffle4.i
// CHECK: %tmp6 = select <16 x i1> %tmp5, <16 x i32> %tmp4, <16 x i32> %shuffle4.i
// CHECK: %shuffle7.i = shufflevector <16 x i32> %tmp6, <16 x i32> undef, <16 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp7 = icmp ult <16 x i32> %tmp6, %shuffle7.i
// CHECK: %tmp8 = select <16 x i1> %tmp7, <16 x i32> %tmp6, <16 x i32> %shuffle7.i
// CHECK: %shuffle10.i = shufflevector <16 x i32> %tmp8, <16 x i32> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp9 = icmp ult <16 x i32> %tmp8, %shuffle10.i
// CHECK: %tmp10 = select <16 x i1> %tmp9, <16 x i32> %tmp8, <16 x i32> %shuffle10.i
// CHECK: %tmp11 = bitcast <16 x i32> %tmp10 to <8 x i64>
// CHECK: %vecext.i = extractelement <8 x i64> %tmp11, i32 0
// CHECK: %conv.i = trunc i64 %vecext.i to i32
// CHECK: ret i32 %conv.i
return _mm512_mask_reduce_min_epu32(__M, __W);
}
float test_mm512_mask_reduce_min_ps(__mmask16 __M, __m512 __W){
// CHECK: %tmp = bitcast i16 %__M to <16 x i1>
// CHECK: %tmp1 = select <16 x i1> %tmp, <16 x float> %__W, <16 x float> <float 0x7FF0000000000000, float 0x7FF0000000000000, float 0x7FF0000000000000, float 0x7FF0000000000000, float 0x7FF0000000000000, float 0x7FF0000000000000, float 0x7FF0000000000000, float 0x7FF0000000000000, float 0x7FF0000000000000, float 0x7FF0000000000000, float 0x7FF0000000000000, float 0x7FF0000000000000, float 0x7FF0000000000000, float 0x7FF0000000000000, float 0x7FF0000000000000, float 0x7FF0000000000000>
// CHECK: %shuffle1.i = shufflevector <16 x float> %tmp1, <16 x float> undef, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp2 = tail call <16 x float> @llvm.x86.avx512.mask.min.ps.512(<16 x float> %tmp1, <16 x float> %shuffle1.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %shuffle4.i = shufflevector <16 x float> %tmp2, <16 x float> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp3 = tail call <16 x float> @llvm.x86.avx512.mask.min.ps.512(<16 x float> %tmp2, <16 x float> %shuffle4.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %shuffle7.i = shufflevector <16 x float> %tmp3, <16 x float> undef, <16 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp4 = tail call <16 x float> @llvm.x86.avx512.mask.min.ps.512(<16 x float> %tmp3, <16 x float> %shuffle7.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %shuffle10.i = shufflevector <16 x float> %tmp4, <16 x float> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
// CHECK: %tmp5 = tail call <16 x float> @llvm.x86.avx512.mask.min.ps.512(<16 x float> %tmp4, <16 x float> %shuffle10.i, <16 x float> zeroinitializer, i16 -1, i32 4) #3
// CHECK: %vecext.i = extractelement <16 x float> %tmp5, i32 0
// CHECK: ret float %vecext.i
return _mm512_mask_reduce_min_ps(__M, __W);
}
|
