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
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
|
; NOTE: Assertions have been autogenerated by update_llc_test_checks.py
; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=knl -mattr=+avx512vl --show-mc-encoding| FileCheck %s
declare void @llvm.x86.avx512.mask.store.pd.128(i8*, <2 x double>, i8)
define void@test_int_x86_avx512_mask_store_pd_128(i8* %ptr1, i8* %ptr2, <2 x double> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_store_pd_128:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovapd %xmm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0xfd,0x09,0x29,0x07]
; CHECK-NEXT: vmovapd %xmm0, (%rsi) ## encoding: [0x62,0xf1,0xfd,0x08,0x29,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.store.pd.128(i8* %ptr1, <2 x double> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.store.pd.128(i8* %ptr2, <2 x double> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.store.pd.256(i8*, <4 x double>, i8)
define void@test_int_x86_avx512_mask_store_pd_256(i8* %ptr1, i8* %ptr2, <4 x double> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_store_pd_256:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovapd %ymm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0xfd,0x29,0x29,0x07]
; CHECK-NEXT: vmovapd %ymm0, (%rsi) ## encoding: [0x62,0xf1,0xfd,0x28,0x29,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.store.pd.256(i8* %ptr1, <4 x double> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.store.pd.256(i8* %ptr2, <4 x double> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.storeu.pd.128(i8*, <2 x double>, i8)
define void@test_int_x86_avx512_mask_storeu_pd_128(i8* %ptr1, i8* %ptr2, <2 x double> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_storeu_pd_128:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovupd %xmm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0xfd,0x09,0x11,0x07]
; CHECK-NEXT: vmovupd %xmm0, (%rsi) ## encoding: [0x62,0xf1,0xfd,0x08,0x11,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.storeu.pd.128(i8* %ptr1, <2 x double> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.storeu.pd.128(i8* %ptr2, <2 x double> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.storeu.pd.256(i8*, <4 x double>, i8)
define void@test_int_x86_avx512_mask_storeu_pd_256(i8* %ptr1, i8* %ptr2, <4 x double> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_storeu_pd_256:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovupd %ymm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0xfd,0x29,0x11,0x07]
; CHECK-NEXT: vmovupd %ymm0, (%rsi) ## encoding: [0x62,0xf1,0xfd,0x28,0x11,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.storeu.pd.256(i8* %ptr1, <4 x double> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.storeu.pd.256(i8* %ptr2, <4 x double> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.store.ps.128(i8*, <4 x float>, i8)
define void@test_int_x86_avx512_mask_store_ps_128(i8* %ptr1, i8* %ptr2, <4 x float> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_store_ps_128:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovaps %xmm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0x7c,0x09,0x29,0x07]
; CHECK-NEXT: vmovaps %xmm0, (%rsi) ## encoding: [0x62,0xf1,0x7c,0x08,0x29,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.store.ps.128(i8* %ptr1, <4 x float> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.store.ps.128(i8* %ptr2, <4 x float> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.store.ps.256(i8*, <8 x float>, i8)
define void@test_int_x86_avx512_mask_store_ps_256(i8* %ptr1, i8* %ptr2, <8 x float> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_store_ps_256:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovaps %ymm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0x7c,0x29,0x29,0x07]
; CHECK-NEXT: vmovaps %ymm0, (%rsi) ## encoding: [0x62,0xf1,0x7c,0x28,0x29,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.store.ps.256(i8* %ptr1, <8 x float> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.store.ps.256(i8* %ptr2, <8 x float> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.storeu.ps.128(i8*, <4 x float>, i8)
define void@test_int_x86_avx512_mask_storeu_ps_128(i8* %ptr1, i8* %ptr2, <4 x float> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_storeu_ps_128:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovups %xmm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0x7c,0x09,0x11,0x07]
; CHECK-NEXT: vmovups %xmm0, (%rsi) ## encoding: [0x62,0xf1,0x7c,0x08,0x11,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.storeu.ps.128(i8* %ptr1, <4 x float> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.storeu.ps.128(i8* %ptr2, <4 x float> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.storeu.ps.256(i8*, <8 x float>, i8)
define void@test_int_x86_avx512_mask_storeu_ps_256(i8* %ptr1, i8* %ptr2, <8 x float> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_storeu_ps_256:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovups %ymm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0x7c,0x29,0x11,0x07]
; CHECK-NEXT: vmovups %ymm0, (%rsi) ## encoding: [0x62,0xf1,0x7c,0x28,0x11,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.storeu.ps.256(i8* %ptr1, <8 x float> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.storeu.ps.256(i8* %ptr2, <8 x float> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.storeu.q.128(i8*, <2 x i64>, i8)
define void@test_int_x86_avx512_mask_storeu_q_128(i8* %ptr1, i8* %ptr2, <2 x i64> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_storeu_q_128:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovdqu64 %xmm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0xfe,0x09,0x7f,0x07]
; CHECK-NEXT: vmovdqu64 %xmm0, (%rsi) ## encoding: [0x62,0xf1,0xfe,0x08,0x7f,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.storeu.q.128(i8* %ptr1, <2 x i64> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.storeu.q.128(i8* %ptr2, <2 x i64> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.storeu.q.256(i8*, <4 x i64>, i8)
define void@test_int_x86_avx512_mask_storeu_q_256(i8* %ptr1, i8* %ptr2, <4 x i64> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_storeu_q_256:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovdqu64 %ymm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0xfe,0x29,0x7f,0x07]
; CHECK-NEXT: vmovdqu64 %ymm0, (%rsi) ## encoding: [0x62,0xf1,0xfe,0x28,0x7f,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.storeu.q.256(i8* %ptr1, <4 x i64> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.storeu.q.256(i8* %ptr2, <4 x i64> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.storeu.d.128(i8*, <4 x i32>, i8)
define void@test_int_x86_avx512_mask_storeu_d_128(i8* %ptr1, i8* %ptr2, <4 x i32> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_storeu_d_128:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovdqu32 %xmm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0x7e,0x09,0x7f,0x07]
; CHECK-NEXT: vmovdqu32 %xmm0, (%rsi) ## encoding: [0x62,0xf1,0x7e,0x08,0x7f,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.storeu.d.128(i8* %ptr1, <4 x i32> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.storeu.d.128(i8* %ptr2, <4 x i32> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.storeu.d.256(i8*, <8 x i32>, i8)
define void@test_int_x86_avx512_mask_storeu_d_256(i8* %ptr1, i8* %ptr2, <8 x i32> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_storeu_d_256:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovdqu32 %ymm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0x7e,0x29,0x7f,0x07]
; CHECK-NEXT: vmovdqu32 %ymm0, (%rsi) ## encoding: [0x62,0xf1,0x7e,0x28,0x7f,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.storeu.d.256(i8* %ptr1, <8 x i32> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.storeu.d.256(i8* %ptr2, <8 x i32> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.store.q.128(i8*, <2 x i64>, i8)
define void@test_int_x86_avx512_mask_store_q_128(i8* %ptr1, i8* %ptr2, <2 x i64> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_store_q_128:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovdqa64 %xmm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0xfd,0x09,0x7f,0x07]
; CHECK-NEXT: vmovdqa64 %xmm0, (%rsi) ## encoding: [0x62,0xf1,0xfd,0x08,0x7f,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.store.q.128(i8* %ptr1, <2 x i64> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.store.q.128(i8* %ptr2, <2 x i64> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.store.q.256(i8*, <4 x i64>, i8)
define void@test_int_x86_avx512_mask_store_q_256(i8* %ptr1, i8* %ptr2, <4 x i64> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_store_q_256:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovdqa64 %ymm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0xfd,0x29,0x7f,0x07]
; CHECK-NEXT: vmovdqa64 %ymm0, (%rsi) ## encoding: [0x62,0xf1,0xfd,0x28,0x7f,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.store.q.256(i8* %ptr1, <4 x i64> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.store.q.256(i8* %ptr2, <4 x i64> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.store.d.128(i8*, <4 x i32>, i8)
define void@test_int_x86_avx512_mask_store_d_128(i8* %ptr1, i8* %ptr2, <4 x i32> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_store_d_128:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovdqa32 %xmm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0x7d,0x09,0x7f,0x07]
; CHECK-NEXT: vmovdqa32 %xmm0, (%rsi) ## encoding: [0x62,0xf1,0x7d,0x08,0x7f,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.store.d.128(i8* %ptr1, <4 x i32> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.store.d.128(i8* %ptr2, <4 x i32> %x1, i8 -1)
ret void
}
declare void @llvm.x86.avx512.mask.store.d.256(i8*, <8 x i32>, i8)
define void@test_int_x86_avx512_mask_store_d_256(i8* %ptr1, i8* %ptr2, <8 x i32> %x1, i8 %x2) {
; CHECK-LABEL: test_int_x86_avx512_mask_store_d_256:
; CHECK: ## BB#0:
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovdqa32 %ymm0, (%rdi) {%k1} ## encoding: [0x62,0xf1,0x7d,0x29,0x7f,0x07]
; CHECK-NEXT: vmovdqa32 %ymm0, (%rsi) ## encoding: [0x62,0xf1,0x7d,0x28,0x7f,0x06]
; CHECK-NEXT: retq ## encoding: [0xc3]
call void @llvm.x86.avx512.mask.store.d.256(i8* %ptr1, <8 x i32> %x1, i8 %x2)
call void @llvm.x86.avx512.mask.store.d.256(i8* %ptr2, <8 x i32> %x1, i8 -1)
ret void
}
define <8 x float> @test_mask_load_aligned_ps_256(<8 x float> %data, i8* %ptr, i8 %mask) {
; CHECK-LABEL: test_mask_load_aligned_ps_256:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovaps (%rdi), %ymm0 ## encoding: [0x62,0xf1,0x7c,0x28,0x28,0x07]
; CHECK-NEXT: kmovw %esi, %k1 ## encoding: [0xc5,0xf8,0x92,0xce]
; CHECK-NEXT: vmovaps (%rdi), %ymm0 {%k1} ## encoding: [0x62,0xf1,0x7c,0x29,0x28,0x07]
; CHECK-NEXT: vmovaps (%rdi), %ymm1 {%k1} {z} ## encoding: [0x62,0xf1,0x7c,0xa9,0x28,0x0f]
; CHECK-NEXT: vaddps %ymm0, %ymm1, %ymm0 ## encoding: [0x62,0xf1,0x74,0x28,0x58,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <8 x float> @llvm.x86.avx512.mask.load.ps.256(i8* %ptr, <8 x float> zeroinitializer, i8 -1)
%res1 = call <8 x float> @llvm.x86.avx512.mask.load.ps.256(i8* %ptr, <8 x float> %res, i8 %mask)
%res2 = call <8 x float> @llvm.x86.avx512.mask.load.ps.256(i8* %ptr, <8 x float> zeroinitializer, i8 %mask)
%res4 = fadd <8 x float> %res2, %res1
ret <8 x float> %res4
}
declare <8 x float> @llvm.x86.avx512.mask.load.ps.256(i8*, <8 x float>, i8)
define <8 x float> @test_mask_load_unaligned_ps_256(<8 x float> %data, i8* %ptr, i8 %mask) {
; CHECK-LABEL: test_mask_load_unaligned_ps_256:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovups (%rdi), %ymm0 ## encoding: [0x62,0xf1,0x7c,0x28,0x10,0x07]
; CHECK-NEXT: kmovw %esi, %k1 ## encoding: [0xc5,0xf8,0x92,0xce]
; CHECK-NEXT: vmovups (%rdi), %ymm0 {%k1} ## encoding: [0x62,0xf1,0x7c,0x29,0x10,0x07]
; CHECK-NEXT: vmovups (%rdi), %ymm1 {%k1} {z} ## encoding: [0x62,0xf1,0x7c,0xa9,0x10,0x0f]
; CHECK-NEXT: vaddps %ymm0, %ymm1, %ymm0 ## encoding: [0x62,0xf1,0x74,0x28,0x58,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <8 x float> @llvm.x86.avx512.mask.loadu.ps.256(i8* %ptr, <8 x float> zeroinitializer, i8 -1)
%res1 = call <8 x float> @llvm.x86.avx512.mask.loadu.ps.256(i8* %ptr, <8 x float> %res, i8 %mask)
%res2 = call <8 x float> @llvm.x86.avx512.mask.loadu.ps.256(i8* %ptr, <8 x float> zeroinitializer, i8 %mask)
%res4 = fadd <8 x float> %res2, %res1
ret <8 x float> %res4
}
declare <8 x float> @llvm.x86.avx512.mask.loadu.ps.256(i8*, <8 x float>, i8)
define <4 x double> @test_mask_load_aligned_pd_256(<4 x double> %data, i8* %ptr, i8 %mask) {
; CHECK-LABEL: test_mask_load_aligned_pd_256:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovapd (%rdi), %ymm0 ## encoding: [0x62,0xf1,0xfd,0x28,0x28,0x07]
; CHECK-NEXT: kmovw %esi, %k1 ## encoding: [0xc5,0xf8,0x92,0xce]
; CHECK-NEXT: vmovapd (%rdi), %ymm0 {%k1} ## encoding: [0x62,0xf1,0xfd,0x29,0x28,0x07]
; CHECK-NEXT: vmovapd (%rdi), %ymm1 {%k1} {z} ## encoding: [0x62,0xf1,0xfd,0xa9,0x28,0x0f]
; CHECK-NEXT: vaddpd %ymm0, %ymm1, %ymm0 ## encoding: [0x62,0xf1,0xf5,0x28,0x58,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <4 x double> @llvm.x86.avx512.mask.load.pd.256(i8* %ptr, <4 x double> zeroinitializer, i8 -1)
%res1 = call <4 x double> @llvm.x86.avx512.mask.load.pd.256(i8* %ptr, <4 x double> %res, i8 %mask)
%res2 = call <4 x double> @llvm.x86.avx512.mask.load.pd.256(i8* %ptr, <4 x double> zeroinitializer, i8 %mask)
%res4 = fadd <4 x double> %res2, %res1
ret <4 x double> %res4
}
declare <4 x double> @llvm.x86.avx512.mask.load.pd.256(i8*, <4 x double>, i8)
define <4 x double> @test_mask_load_unaligned_pd_256(<4 x double> %data, i8* %ptr, i8 %mask) {
; CHECK-LABEL: test_mask_load_unaligned_pd_256:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovupd (%rdi), %ymm0 ## encoding: [0x62,0xf1,0xfd,0x28,0x10,0x07]
; CHECK-NEXT: kmovw %esi, %k1 ## encoding: [0xc5,0xf8,0x92,0xce]
; CHECK-NEXT: vmovupd (%rdi), %ymm0 {%k1} ## encoding: [0x62,0xf1,0xfd,0x29,0x10,0x07]
; CHECK-NEXT: vmovupd (%rdi), %ymm1 {%k1} {z} ## encoding: [0x62,0xf1,0xfd,0xa9,0x10,0x0f]
; CHECK-NEXT: vaddpd %ymm0, %ymm1, %ymm0 ## encoding: [0x62,0xf1,0xf5,0x28,0x58,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <4 x double> @llvm.x86.avx512.mask.loadu.pd.256(i8* %ptr, <4 x double> zeroinitializer, i8 -1)
%res1 = call <4 x double> @llvm.x86.avx512.mask.loadu.pd.256(i8* %ptr, <4 x double> %res, i8 %mask)
%res2 = call <4 x double> @llvm.x86.avx512.mask.loadu.pd.256(i8* %ptr, <4 x double> zeroinitializer, i8 %mask)
%res4 = fadd <4 x double> %res2, %res1
ret <4 x double> %res4
}
declare <4 x double> @llvm.x86.avx512.mask.loadu.pd.256(i8*, <4 x double>, i8)
define <4 x float> @test_mask_load_aligned_ps_128(<4 x float> %data, i8* %ptr, i8 %mask) {
; CHECK-LABEL: test_mask_load_aligned_ps_128:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovaps (%rdi), %xmm0 ## encoding: [0x62,0xf1,0x7c,0x08,0x28,0x07]
; CHECK-NEXT: kmovw %esi, %k1 ## encoding: [0xc5,0xf8,0x92,0xce]
; CHECK-NEXT: vmovaps (%rdi), %xmm0 {%k1} ## encoding: [0x62,0xf1,0x7c,0x09,0x28,0x07]
; CHECK-NEXT: vmovaps (%rdi), %xmm1 {%k1} {z} ## encoding: [0x62,0xf1,0x7c,0x89,0x28,0x0f]
; CHECK-NEXT: vaddps %xmm0, %xmm1, %xmm0 ## encoding: [0x62,0xf1,0x74,0x08,0x58,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <4 x float> @llvm.x86.avx512.mask.load.ps.128(i8* %ptr, <4 x float> zeroinitializer, i8 -1)
%res1 = call <4 x float> @llvm.x86.avx512.mask.load.ps.128(i8* %ptr, <4 x float> %res, i8 %mask)
%res2 = call <4 x float> @llvm.x86.avx512.mask.load.ps.128(i8* %ptr, <4 x float> zeroinitializer, i8 %mask)
%res4 = fadd <4 x float> %res2, %res1
ret <4 x float> %res4
}
declare <4 x float> @llvm.x86.avx512.mask.load.ps.128(i8*, <4 x float>, i8)
define <4 x float> @test_mask_load_unaligned_ps_128(<4 x float> %data, i8* %ptr, i8 %mask) {
; CHECK-LABEL: test_mask_load_unaligned_ps_128:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovups (%rdi), %xmm0 ## encoding: [0x62,0xf1,0x7c,0x08,0x10,0x07]
; CHECK-NEXT: kmovw %esi, %k1 ## encoding: [0xc5,0xf8,0x92,0xce]
; CHECK-NEXT: vmovups (%rdi), %xmm0 {%k1} ## encoding: [0x62,0xf1,0x7c,0x09,0x10,0x07]
; CHECK-NEXT: vmovups (%rdi), %xmm1 {%k1} {z} ## encoding: [0x62,0xf1,0x7c,0x89,0x10,0x0f]
; CHECK-NEXT: vaddps %xmm0, %xmm1, %xmm0 ## encoding: [0x62,0xf1,0x74,0x08,0x58,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <4 x float> @llvm.x86.avx512.mask.loadu.ps.128(i8* %ptr, <4 x float> zeroinitializer, i8 -1)
%res1 = call <4 x float> @llvm.x86.avx512.mask.loadu.ps.128(i8* %ptr, <4 x float> %res, i8 %mask)
%res2 = call <4 x float> @llvm.x86.avx512.mask.loadu.ps.128(i8* %ptr, <4 x float> zeroinitializer, i8 %mask)
%res4 = fadd <4 x float> %res2, %res1
ret <4 x float> %res4
}
declare <4 x float> @llvm.x86.avx512.mask.loadu.ps.128(i8*, <4 x float>, i8)
define <2 x double> @test_mask_load_aligned_pd_128(<2 x double> %data, i8* %ptr, i8 %mask) {
; CHECK-LABEL: test_mask_load_aligned_pd_128:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovapd (%rdi), %xmm0 ## encoding: [0x62,0xf1,0xfd,0x08,0x28,0x07]
; CHECK-NEXT: kmovw %esi, %k1 ## encoding: [0xc5,0xf8,0x92,0xce]
; CHECK-NEXT: vmovapd (%rdi), %xmm0 {%k1} ## encoding: [0x62,0xf1,0xfd,0x09,0x28,0x07]
; CHECK-NEXT: vmovapd (%rdi), %xmm1 {%k1} {z} ## encoding: [0x62,0xf1,0xfd,0x89,0x28,0x0f]
; CHECK-NEXT: vaddpd %xmm0, %xmm1, %xmm0 ## encoding: [0x62,0xf1,0xf5,0x08,0x58,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <2 x double> @llvm.x86.avx512.mask.load.pd.128(i8* %ptr, <2 x double> zeroinitializer, i8 -1)
%res1 = call <2 x double> @llvm.x86.avx512.mask.load.pd.128(i8* %ptr, <2 x double> %res, i8 %mask)
%res2 = call <2 x double> @llvm.x86.avx512.mask.load.pd.128(i8* %ptr, <2 x double> zeroinitializer, i8 %mask)
%res4 = fadd <2 x double> %res2, %res1
ret <2 x double> %res4
}
declare <2 x double> @llvm.x86.avx512.mask.load.pd.128(i8*, <2 x double>, i8)
define <2 x double> @test_mask_load_unaligned_pd_128(<2 x double> %data, i8* %ptr, i8 %mask) {
; CHECK-LABEL: test_mask_load_unaligned_pd_128:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovupd (%rdi), %xmm0 ## encoding: [0x62,0xf1,0xfd,0x08,0x10,0x07]
; CHECK-NEXT: kmovw %esi, %k1 ## encoding: [0xc5,0xf8,0x92,0xce]
; CHECK-NEXT: vmovupd (%rdi), %xmm0 {%k1} ## encoding: [0x62,0xf1,0xfd,0x09,0x10,0x07]
; CHECK-NEXT: vmovupd (%rdi), %xmm1 {%k1} {z} ## encoding: [0x62,0xf1,0xfd,0x89,0x10,0x0f]
; CHECK-NEXT: vaddpd %xmm0, %xmm1, %xmm0 ## encoding: [0x62,0xf1,0xf5,0x08,0x58,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <2 x double> @llvm.x86.avx512.mask.loadu.pd.128(i8* %ptr, <2 x double> zeroinitializer, i8 -1)
%res1 = call <2 x double> @llvm.x86.avx512.mask.loadu.pd.128(i8* %ptr, <2 x double> %res, i8 %mask)
%res2 = call <2 x double> @llvm.x86.avx512.mask.loadu.pd.128(i8* %ptr, <2 x double> zeroinitializer, i8 %mask)
%res4 = fadd <2 x double> %res2, %res1
ret <2 x double> %res4
}
declare <2 x double> @llvm.x86.avx512.mask.loadu.pd.128(i8*, <2 x double>, i8)
declare <4 x i32> @llvm.x86.avx512.mask.loadu.d.128(i8*, <4 x i32>, i8)
define <4 x i32> @test_mask_load_unaligned_d_128(i8* %ptr, i8* %ptr2, <4 x i32> %data, i8 %mask) {
; CHECK-LABEL: test_mask_load_unaligned_d_128:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovdqu32 (%rdi), %xmm0 ## encoding: [0x62,0xf1,0x7e,0x08,0x6f,0x07]
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovdqu32 (%rsi), %xmm0 {%k1} ## encoding: [0x62,0xf1,0x7e,0x09,0x6f,0x06]
; CHECK-NEXT: vmovdqu32 (%rdi), %xmm1 {%k1} {z} ## encoding: [0x62,0xf1,0x7e,0x89,0x6f,0x0f]
; CHECK-NEXT: vpaddd %xmm0, %xmm1, %xmm0 ## encoding: [0x62,0xf1,0x75,0x08,0xfe,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <4 x i32> @llvm.x86.avx512.mask.loadu.d.128(i8* %ptr, <4 x i32> zeroinitializer, i8 -1)
%res1 = call <4 x i32> @llvm.x86.avx512.mask.loadu.d.128(i8* %ptr2, <4 x i32> %res, i8 %mask)
%res2 = call <4 x i32> @llvm.x86.avx512.mask.loadu.d.128(i8* %ptr, <4 x i32> zeroinitializer, i8 %mask)
%res4 = add <4 x i32> %res2, %res1
ret <4 x i32> %res4
}
declare <8 x i32> @llvm.x86.avx512.mask.loadu.d.256(i8*, <8 x i32>, i8)
define <8 x i32> @test_mask_load_unaligned_d_256(i8* %ptr, i8* %ptr2, <8 x i32> %data, i8 %mask) {
; CHECK-LABEL: test_mask_load_unaligned_d_256:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovdqu32 (%rdi), %ymm0 ## encoding: [0x62,0xf1,0x7e,0x28,0x6f,0x07]
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovdqu32 (%rsi), %ymm0 {%k1} ## encoding: [0x62,0xf1,0x7e,0x29,0x6f,0x06]
; CHECK-NEXT: vmovdqu32 (%rdi), %ymm1 {%k1} {z} ## encoding: [0x62,0xf1,0x7e,0xa9,0x6f,0x0f]
; CHECK-NEXT: vpaddd %ymm0, %ymm1, %ymm0 ## encoding: [0x62,0xf1,0x75,0x28,0xfe,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <8 x i32> @llvm.x86.avx512.mask.loadu.d.256(i8* %ptr, <8 x i32> zeroinitializer, i8 -1)
%res1 = call <8 x i32> @llvm.x86.avx512.mask.loadu.d.256(i8* %ptr2, <8 x i32> %res, i8 %mask)
%res2 = call <8 x i32> @llvm.x86.avx512.mask.loadu.d.256(i8* %ptr, <8 x i32> zeroinitializer, i8 %mask)
%res4 = add <8 x i32> %res2, %res1
ret <8 x i32> %res4
}
declare <2 x i64> @llvm.x86.avx512.mask.loadu.q.128(i8*, <2 x i64>, i8)
define <2 x i64> @test_mask_load_unaligned_q_128(i8* %ptr, i8* %ptr2, <2 x i64> %data, i8 %mask) {
; CHECK-LABEL: test_mask_load_unaligned_q_128:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovdqu64 (%rdi), %xmm0 ## encoding: [0x62,0xf1,0xfe,0x08,0x6f,0x07]
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovdqu64 (%rsi), %xmm0 {%k1} ## encoding: [0x62,0xf1,0xfe,0x09,0x6f,0x06]
; CHECK-NEXT: vmovdqu64 (%rdi), %xmm1 {%k1} {z} ## encoding: [0x62,0xf1,0xfe,0x89,0x6f,0x0f]
; CHECK-NEXT: vpaddq %xmm0, %xmm1, %xmm0 ## encoding: [0x62,0xf1,0xf5,0x08,0xd4,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <2 x i64> @llvm.x86.avx512.mask.loadu.q.128(i8* %ptr, <2 x i64> zeroinitializer, i8 -1)
%res1 = call <2 x i64> @llvm.x86.avx512.mask.loadu.q.128(i8* %ptr2, <2 x i64> %res, i8 %mask)
%res2 = call <2 x i64> @llvm.x86.avx512.mask.loadu.q.128(i8* %ptr, <2 x i64> zeroinitializer, i8 %mask)
%res4 = add <2 x i64> %res2, %res1
ret <2 x i64> %res4
}
declare <4 x i64> @llvm.x86.avx512.mask.loadu.q.256(i8*, <4 x i64>, i8)
define <4 x i64> @test_mask_load_unaligned_q_256(i8* %ptr, i8* %ptr2, <4 x i64> %data, i8 %mask) {
; CHECK-LABEL: test_mask_load_unaligned_q_256:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovdqu64 (%rdi), %ymm0 ## encoding: [0x62,0xf1,0xfe,0x28,0x6f,0x07]
; CHECK-NEXT: kmovw %edx, %k1 ## encoding: [0xc5,0xf8,0x92,0xca]
; CHECK-NEXT: vmovdqu64 (%rsi), %ymm0 {%k1} ## encoding: [0x62,0xf1,0xfe,0x29,0x6f,0x06]
; CHECK-NEXT: vmovdqu64 (%rdi), %ymm1 {%k1} {z} ## encoding: [0x62,0xf1,0xfe,0xa9,0x6f,0x0f]
; CHECK-NEXT: vpaddq %ymm0, %ymm1, %ymm0 ## encoding: [0x62,0xf1,0xf5,0x28,0xd4,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <4 x i64> @llvm.x86.avx512.mask.loadu.q.256(i8* %ptr, <4 x i64> zeroinitializer, i8 -1)
%res1 = call <4 x i64> @llvm.x86.avx512.mask.loadu.q.256(i8* %ptr2, <4 x i64> %res, i8 %mask)
%res2 = call <4 x i64> @llvm.x86.avx512.mask.loadu.q.256(i8* %ptr, <4 x i64> zeroinitializer, i8 %mask)
%res4 = add <4 x i64> %res2, %res1
ret <4 x i64> %res4
}
declare <4 x i32> @llvm.x86.avx512.mask.load.d.128(i8*, <4 x i32>, i8)
define <4 x i32> @test_mask_load_aligned_d_128(<4 x i32> %data, i8* %ptr, i8 %mask) {
; CHECK-LABEL: test_mask_load_aligned_d_128:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovdqa32 (%rdi), %xmm0 ## encoding: [0x62,0xf1,0x7d,0x08,0x6f,0x07]
; CHECK-NEXT: kmovw %esi, %k1 ## encoding: [0xc5,0xf8,0x92,0xce]
; CHECK-NEXT: vmovdqa32 (%rdi), %xmm0 {%k1} ## encoding: [0x62,0xf1,0x7d,0x09,0x6f,0x07]
; CHECK-NEXT: vmovdqa32 (%rdi), %xmm1 {%k1} {z} ## encoding: [0x62,0xf1,0x7d,0x89,0x6f,0x0f]
; CHECK-NEXT: vpaddd %xmm0, %xmm1, %xmm0 ## encoding: [0x62,0xf1,0x75,0x08,0xfe,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <4 x i32> @llvm.x86.avx512.mask.load.d.128(i8* %ptr, <4 x i32> zeroinitializer, i8 -1)
%res1 = call <4 x i32> @llvm.x86.avx512.mask.load.d.128(i8* %ptr, <4 x i32> %res, i8 %mask)
%res2 = call <4 x i32> @llvm.x86.avx512.mask.load.d.128(i8* %ptr, <4 x i32> zeroinitializer, i8 %mask)
%res4 = add <4 x i32> %res2, %res1
ret <4 x i32> %res4
}
declare <8 x i32> @llvm.x86.avx512.mask.load.d.256(i8*, <8 x i32>, i8)
define <8 x i32> @test_mask_load_aligned_d_256(<8 x i32> %data, i8* %ptr, i8 %mask) {
; CHECK-LABEL: test_mask_load_aligned_d_256:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovdqa32 (%rdi), %ymm0 ## encoding: [0x62,0xf1,0x7d,0x28,0x6f,0x07]
; CHECK-NEXT: kmovw %esi, %k1 ## encoding: [0xc5,0xf8,0x92,0xce]
; CHECK-NEXT: vmovdqa32 (%rdi), %ymm0 {%k1} ## encoding: [0x62,0xf1,0x7d,0x29,0x6f,0x07]
; CHECK-NEXT: vmovdqa32 (%rdi), %ymm1 {%k1} {z} ## encoding: [0x62,0xf1,0x7d,0xa9,0x6f,0x0f]
; CHECK-NEXT: vpaddd %ymm0, %ymm1, %ymm0 ## encoding: [0x62,0xf1,0x75,0x28,0xfe,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <8 x i32> @llvm.x86.avx512.mask.load.d.256(i8* %ptr, <8 x i32> zeroinitializer, i8 -1)
%res1 = call <8 x i32> @llvm.x86.avx512.mask.load.d.256(i8* %ptr, <8 x i32> %res, i8 %mask)
%res2 = call <8 x i32> @llvm.x86.avx512.mask.load.d.256(i8* %ptr, <8 x i32> zeroinitializer, i8 %mask)
%res4 = add <8 x i32> %res2, %res1
ret <8 x i32> %res4
}
declare <2 x i64> @llvm.x86.avx512.mask.load.q.128(i8*, <2 x i64>, i8)
define <2 x i64> @test_mask_load_aligned_q_128(<2 x i64> %data, i8* %ptr, i8 %mask) {
; CHECK-LABEL: test_mask_load_aligned_q_128:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovdqa64 (%rdi), %xmm0 ## encoding: [0x62,0xf1,0xfd,0x08,0x6f,0x07]
; CHECK-NEXT: kmovw %esi, %k1 ## encoding: [0xc5,0xf8,0x92,0xce]
; CHECK-NEXT: vmovdqa64 (%rdi), %xmm0 {%k1} ## encoding: [0x62,0xf1,0xfd,0x09,0x6f,0x07]
; CHECK-NEXT: vmovdqa64 (%rdi), %xmm1 {%k1} {z} ## encoding: [0x62,0xf1,0xfd,0x89,0x6f,0x0f]
; CHECK-NEXT: vpaddq %xmm0, %xmm1, %xmm0 ## encoding: [0x62,0xf1,0xf5,0x08,0xd4,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <2 x i64> @llvm.x86.avx512.mask.load.q.128(i8* %ptr, <2 x i64> zeroinitializer, i8 -1)
%res1 = call <2 x i64> @llvm.x86.avx512.mask.load.q.128(i8* %ptr, <2 x i64> %res, i8 %mask)
%res2 = call <2 x i64> @llvm.x86.avx512.mask.load.q.128(i8* %ptr, <2 x i64> zeroinitializer, i8 %mask)
%res4 = add <2 x i64> %res2, %res1
ret <2 x i64> %res4
}
declare <4 x i64> @llvm.x86.avx512.mask.load.q.256(i8*, <4 x i64>, i8)
define <4 x i64> @test_mask_load_aligned_q_256(<4 x i64> %data, i8* %ptr, i8 %mask) {
; CHECK-LABEL: test_mask_load_aligned_q_256:
; CHECK: ## BB#0:
; CHECK-NEXT: vmovdqa64 (%rdi), %ymm0 ## encoding: [0x62,0xf1,0xfd,0x28,0x6f,0x07]
; CHECK-NEXT: kmovw %esi, %k1 ## encoding: [0xc5,0xf8,0x92,0xce]
; CHECK-NEXT: vmovdqa64 (%rdi), %ymm0 {%k1} ## encoding: [0x62,0xf1,0xfd,0x29,0x6f,0x07]
; CHECK-NEXT: vmovdqa64 (%rdi), %ymm1 {%k1} {z} ## encoding: [0x62,0xf1,0xfd,0xa9,0x6f,0x0f]
; CHECK-NEXT: vpaddq %ymm0, %ymm1, %ymm0 ## encoding: [0x62,0xf1,0xf5,0x28,0xd4,0xc0]
; CHECK-NEXT: retq ## encoding: [0xc3]
%res = call <4 x i64> @llvm.x86.avx512.mask.load.q.256(i8* %ptr, <4 x i64> zeroinitializer, i8 -1)
%res1 = call <4 x i64> @llvm.x86.avx512.mask.load.q.256(i8* %ptr, <4 x i64> %res, i8 %mask)
%res2 = call <4 x i64> @llvm.x86.avx512.mask.load.q.256(i8* %ptr, <4 x i64> zeroinitializer, i8 %mask)
%res4 = add <4 x i64> %res2, %res1
ret <4 x i64> %res4
}
|
