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
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=SSE --check-prefix=X86-SSE
; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=AVX --check-prefix=AVX1 --check-prefix=X86-AVX --check-prefix=X86-AVX1
; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefix=AVX --check-prefix=AVX2 --check-prefix=X86-AVX --check-prefix=X86-AVX2
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=SSE --check-prefix=X64-SSE
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=AVX --check-prefix=AVX1 --check-prefix=X64-AVX --check-prefix=X64-AVX1
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefix=AVX --check-prefix=AVX2 --check-prefix=X64-AVX --check-prefix=X64-AVX2
define <4 x i32> @trunc_ashr_v4i64(<4 x i64> %a) nounwind {
; SSE-LABEL: trunc_ashr_v4i64:
; SSE: # %bb.0:
; SSE-NEXT: psrad $31, %xmm1
; SSE-NEXT: pshufd {{.*#+}} xmm1 = xmm1[1,1,3,3]
; SSE-NEXT: psrad $31, %xmm0
; SSE-NEXT: pshufd {{.*#+}} xmm0 = xmm0[1,1,3,3]
; SSE-NEXT: packssdw %xmm1, %xmm0
; SSE-NEXT: ret{{[l|q]}}
;
; AVX1-LABEL: trunc_ashr_v4i64:
; AVX1: # %bb.0:
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2
; AVX1-NEXT: vpcmpgtq %xmm1, %xmm2, %xmm1
; AVX1-NEXT: vpcmpgtq %xmm0, %xmm2, %xmm0
; AVX1-NEXT: vpackssdw %xmm1, %xmm0, %xmm0
; AVX1-NEXT: vzeroupper
; AVX1-NEXT: ret{{[l|q]}}
;
; AVX2-LABEL: trunc_ashr_v4i64:
; AVX2: # %bb.0:
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX2-NEXT: vpcmpgtq %ymm0, %ymm1, %ymm0
; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX2-NEXT: vpackssdw %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vzeroupper
; AVX2-NEXT: ret{{[l|q]}}
%1 = ashr <4 x i64> %a, <i64 63, i64 63, i64 63, i64 63>
%2 = trunc <4 x i64> %1 to <4 x i32>
ret <4 x i32> %2
}
define <8 x i16> @trunc_ashr_v4i64_bitcast(<4 x i64> %a0) {
; SSE-LABEL: trunc_ashr_v4i64_bitcast:
; SSE: # %bb.0:
; SSE-NEXT: movdqa %xmm1, %xmm2
; SSE-NEXT: psrad $31, %xmm2
; SSE-NEXT: pshufd {{.*#+}} xmm2 = xmm2[1,3,2,3]
; SSE-NEXT: psrad $17, %xmm1
; SSE-NEXT: pshufd {{.*#+}} xmm1 = xmm1[1,3,2,3]
; SSE-NEXT: punpckldq {{.*#+}} xmm1 = xmm1[0],xmm2[0],xmm1[1],xmm2[1]
; SSE-NEXT: movdqa %xmm0, %xmm2
; SSE-NEXT: psrad $31, %xmm2
; SSE-NEXT: pshufd {{.*#+}} xmm2 = xmm2[1,3,2,3]
; SSE-NEXT: psrad $17, %xmm0
; SSE-NEXT: pshufd {{.*#+}} xmm0 = xmm0[1,3,2,3]
; SSE-NEXT: punpckldq {{.*#+}} xmm0 = xmm0[0],xmm2[0],xmm0[1],xmm2[1]
; SSE-NEXT: packssdw %xmm1, %xmm0
; SSE-NEXT: ret{{[l|q]}}
;
; AVX1-LABEL: trunc_ashr_v4i64_bitcast:
; AVX1: # %bb.0:
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX1-NEXT: vpsrad $31, %xmm1, %xmm2
; AVX1-NEXT: vpsrad $17, %xmm1, %xmm1
; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[1,1,3,3]
; AVX1-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1],xmm2[2,3],xmm1[4,5],xmm2[6,7]
; AVX1-NEXT: vpsrad $31, %xmm0, %xmm2
; AVX1-NEXT: vpsrad $17, %xmm0, %xmm0
; AVX1-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[1,1,3,3]
; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm2[2,3],xmm0[4,5],xmm2[6,7]
; AVX1-NEXT: vpackssdw %xmm1, %xmm0, %xmm0
; AVX1-NEXT: vzeroupper
; AVX1-NEXT: ret{{[l|q]}}
;
; AVX2-LABEL: trunc_ashr_v4i64_bitcast:
; AVX2: # %bb.0:
; AVX2-NEXT: vpsrad $31, %ymm0, %ymm1
; AVX2-NEXT: vpsrad $17, %ymm0, %ymm0
; AVX2-NEXT: vpshufd {{.*#+}} ymm0 = ymm0[1,1,3,3,5,5,7,7]
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3],ymm0[4],ymm1[5],ymm0[6],ymm1[7]
; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX2-NEXT: vpackssdw %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vzeroupper
; AVX2-NEXT: ret{{[l|q]}}
%1 = ashr <4 x i64> %a0, <i64 49, i64 49, i64 49, i64 49>
%2 = bitcast <4 x i64> %1 to <8 x i32>
%3 = trunc <8 x i32> %2 to <8 x i16>
ret <8 x i16> %3
}
define <8 x i16> @trunc_ashr_v8i32(<8 x i32> %a) nounwind {
; SSE-LABEL: trunc_ashr_v8i32:
; SSE: # %bb.0:
; SSE-NEXT: psrad $31, %xmm1
; SSE-NEXT: psrad $31, %xmm0
; SSE-NEXT: packssdw %xmm1, %xmm0
; SSE-NEXT: ret{{[l|q]}}
;
; AVX1-LABEL: trunc_ashr_v8i32:
; AVX1: # %bb.0:
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX1-NEXT: vpsrad $31, %xmm1, %xmm1
; AVX1-NEXT: vpsrad $31, %xmm0, %xmm0
; AVX1-NEXT: vpackssdw %xmm1, %xmm0, %xmm0
; AVX1-NEXT: vzeroupper
; AVX1-NEXT: ret{{[l|q]}}
;
; AVX2-LABEL: trunc_ashr_v8i32:
; AVX2: # %bb.0:
; AVX2-NEXT: vpsrad $31, %ymm0, %ymm0
; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm1
; AVX2-NEXT: vpackssdw %xmm1, %xmm0, %xmm0
; AVX2-NEXT: vzeroupper
; AVX2-NEXT: ret{{[l|q]}}
%1 = ashr <8 x i32> %a, <i32 31, i32 31, i32 31, i32 31, i32 31, i32 31, i32 31, i32 31>
%2 = trunc <8 x i32> %1 to <8 x i16>
ret <8 x i16> %2
}
define <8 x i16> @trunc_ashr_v4i32_icmp_v4i32(<4 x i32> %a, <4 x i32> %b) nounwind {
; X86-SSE-LABEL: trunc_ashr_v4i32_icmp_v4i32:
; X86-SSE: # %bb.0:
; X86-SSE-NEXT: psrad $31, %xmm0
; X86-SSE-NEXT: pcmpgtd {{\.LCPI.*}}, %xmm1
; X86-SSE-NEXT: packssdw %xmm1, %xmm0
; X86-SSE-NEXT: retl
;
; X86-AVX-LABEL: trunc_ashr_v4i32_icmp_v4i32:
; X86-AVX: # %bb.0:
; X86-AVX-NEXT: vpsrad $31, %xmm0, %xmm0
; X86-AVX-NEXT: vpcmpgtd {{\.LCPI.*}}, %xmm1, %xmm1
; X86-AVX-NEXT: vpackssdw %xmm1, %xmm0, %xmm0
; X86-AVX-NEXT: retl
;
; X64-SSE-LABEL: trunc_ashr_v4i32_icmp_v4i32:
; X64-SSE: # %bb.0:
; X64-SSE-NEXT: psrad $31, %xmm0
; X64-SSE-NEXT: pcmpgtd {{.*}}(%rip), %xmm1
; X64-SSE-NEXT: packssdw %xmm1, %xmm0
; X64-SSE-NEXT: retq
;
; X64-AVX-LABEL: trunc_ashr_v4i32_icmp_v4i32:
; X64-AVX: # %bb.0:
; X64-AVX-NEXT: vpsrad $31, %xmm0, %xmm0
; X64-AVX-NEXT: vpcmpgtd {{.*}}(%rip), %xmm1, %xmm1
; X64-AVX-NEXT: vpackssdw %xmm1, %xmm0, %xmm0
; X64-AVX-NEXT: retq
%1 = ashr <4 x i32> %a, <i32 31, i32 31, i32 31, i32 31>
%2 = icmp sgt <4 x i32> %b, <i32 1, i32 16, i32 255, i32 65535>
%3 = sext <4 x i1> %2 to <4 x i32>
%4 = shufflevector <4 x i32> %1, <4 x i32> %3, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
%5 = trunc <8 x i32> %4 to <8 x i16>
ret <8 x i16> %5
}
define <8 x i16> @trunc_ashr_v4i64_demandedelts(<4 x i64> %a0) {
; X86-SSE-LABEL: trunc_ashr_v4i64_demandedelts:
; X86-SSE: # %bb.0:
; X86-SSE-NEXT: psllq $63, %xmm1
; X86-SSE-NEXT: psllq $63, %xmm0
; X86-SSE-NEXT: psrlq $63, %xmm0
; X86-SSE-NEXT: movdqa {{.*#+}} xmm2 = [4.9406564584124654E-324,-0.0E+0]
; X86-SSE-NEXT: pxor %xmm2, %xmm0
; X86-SSE-NEXT: psubq %xmm2, %xmm0
; X86-SSE-NEXT: psrlq $63, %xmm1
; X86-SSE-NEXT: pxor %xmm2, %xmm1
; X86-SSE-NEXT: psubq %xmm2, %xmm1
; X86-SSE-NEXT: pshufd {{.*#+}} xmm1 = xmm1[0,0,0,0]
; X86-SSE-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; X86-SSE-NEXT: packssdw %xmm1, %xmm0
; X86-SSE-NEXT: retl
;
; X86-AVX1-LABEL: trunc_ashr_v4i64_demandedelts:
; X86-AVX1: # %bb.0:
; X86-AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
; X86-AVX1-NEXT: vpsllq $63, %xmm1, %xmm2
; X86-AVX1-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0,1,2,3],xmm1[4,5,6,7]
; X86-AVX1-NEXT: vpsllq $63, %xmm0, %xmm2
; X86-AVX1-NEXT: vpsrlq $63, %xmm1, %xmm1
; X86-AVX1-NEXT: vmovdqa {{.*#+}} xmm3 = [1,0,0,2147483648]
; X86-AVX1-NEXT: vpxor %xmm3, %xmm1, %xmm1
; X86-AVX1-NEXT: vpsubq %xmm3, %xmm1, %xmm1
; X86-AVX1-NEXT: vpsrlq $63, %xmm2, %xmm2
; X86-AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm2[0,1,2,3],xmm0[4,5,6,7]
; X86-AVX1-NEXT: vpxor %xmm3, %xmm0, %xmm0
; X86-AVX1-NEXT: vpsubq %xmm3, %xmm0, %xmm0
; X86-AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
; X86-AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,0,0,0,4,4,4,4]
; X86-AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
; X86-AVX1-NEXT: vpackssdw %xmm1, %xmm0, %xmm0
; X86-AVX1-NEXT: vzeroupper
; X86-AVX1-NEXT: retl
;
; X86-AVX2-LABEL: trunc_ashr_v4i64_demandedelts:
; X86-AVX2: # %bb.0:
; X86-AVX2-NEXT: vbroadcasti128 {{.*#+}} ymm1 = [63,0,0,0,63,0,0,0]
; X86-AVX2-NEXT: # ymm1 = mem[0,1,0,1]
; X86-AVX2-NEXT: vpsllvq %ymm1, %ymm0, %ymm0
; X86-AVX2-NEXT: vmovdqa {{.*#+}} ymm2 = [0,2147483648,0,2147483648,0,2147483648,0,2147483648]
; X86-AVX2-NEXT: vpsrlvq %ymm1, %ymm2, %ymm2
; X86-AVX2-NEXT: vpsrlvq %ymm1, %ymm0, %ymm0
; X86-AVX2-NEXT: vpxor %ymm2, %ymm0, %ymm0
; X86-AVX2-NEXT: vpsubq %ymm2, %ymm0, %ymm0
; X86-AVX2-NEXT: vpshufd {{.*#+}} ymm0 = ymm0[0,0,0,0,4,4,4,4]
; X86-AVX2-NEXT: vextracti128 $1, %ymm0, %xmm1
; X86-AVX2-NEXT: vpackssdw %xmm1, %xmm0, %xmm0
; X86-AVX2-NEXT: vzeroupper
; X86-AVX2-NEXT: retl
;
; X64-SSE-LABEL: trunc_ashr_v4i64_demandedelts:
; X64-SSE: # %bb.0:
; X64-SSE-NEXT: psllq $63, %xmm1
; X64-SSE-NEXT: psllq $63, %xmm0
; X64-SSE-NEXT: psrlq $63, %xmm0
; X64-SSE-NEXT: movdqa {{.*#+}} xmm2 = [1,9223372036854775808]
; X64-SSE-NEXT: pxor %xmm2, %xmm0
; X64-SSE-NEXT: psubq %xmm2, %xmm0
; X64-SSE-NEXT: psrlq $63, %xmm1
; X64-SSE-NEXT: pxor %xmm2, %xmm1
; X64-SSE-NEXT: psubq %xmm2, %xmm1
; X64-SSE-NEXT: pshufd {{.*#+}} xmm1 = xmm1[0,0,0,0]
; X64-SSE-NEXT: pshufd {{.*#+}} xmm0 = xmm0[0,0,0,0]
; X64-SSE-NEXT: packssdw %xmm1, %xmm0
; X64-SSE-NEXT: retq
;
; X64-AVX1-LABEL: trunc_ashr_v4i64_demandedelts:
; X64-AVX1: # %bb.0:
; X64-AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
; X64-AVX1-NEXT: vpsllq $63, %xmm1, %xmm2
; X64-AVX1-NEXT: vpblendw {{.*#+}} xmm1 = xmm2[0,1,2,3],xmm1[4,5,6,7]
; X64-AVX1-NEXT: vpsllq $63, %xmm0, %xmm2
; X64-AVX1-NEXT: vpsrlq $63, %xmm1, %xmm1
; X64-AVX1-NEXT: vmovdqa {{.*#+}} xmm3 = [1,9223372036854775808]
; X64-AVX1-NEXT: vpxor %xmm3, %xmm1, %xmm1
; X64-AVX1-NEXT: vpsubq %xmm3, %xmm1, %xmm1
; X64-AVX1-NEXT: vpsrlq $63, %xmm2, %xmm2
; X64-AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm2[0,1,2,3],xmm0[4,5,6,7]
; X64-AVX1-NEXT: vpxor %xmm3, %xmm0, %xmm0
; X64-AVX1-NEXT: vpsubq %xmm3, %xmm0, %xmm0
; X64-AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
; X64-AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,0,0,0,4,4,4,4]
; X64-AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
; X64-AVX1-NEXT: vpackssdw %xmm1, %xmm0, %xmm0
; X64-AVX1-NEXT: vzeroupper
; X64-AVX1-NEXT: retq
;
; X64-AVX2-LABEL: trunc_ashr_v4i64_demandedelts:
; X64-AVX2: # %bb.0:
; X64-AVX2-NEXT: vpbroadcastq {{.*#+}} ymm1 = [1,1,1,1]
; X64-AVX2-NEXT: vpand %ymm1, %ymm0, %ymm0
; X64-AVX2-NEXT: vbroadcasti128 {{.*#+}} ymm1 = [1,9223372036854775808,1,9223372036854775808]
; X64-AVX2-NEXT: # ymm1 = mem[0,1,0,1]
; X64-AVX2-NEXT: vpxor %ymm1, %ymm0, %ymm0
; X64-AVX2-NEXT: vpsubq %ymm1, %ymm0, %ymm0
; X64-AVX2-NEXT: vpshufd {{.*#+}} ymm0 = ymm0[0,0,0,0,4,4,4,4]
; X64-AVX2-NEXT: vextracti128 $1, %ymm0, %xmm1
; X64-AVX2-NEXT: vpackssdw %xmm1, %xmm0, %xmm0
; X64-AVX2-NEXT: vzeroupper
; X64-AVX2-NEXT: retq
%1 = shl <4 x i64> %a0, <i64 63, i64 0, i64 63, i64 0>
%2 = ashr <4 x i64> %1, <i64 63, i64 0, i64 63, i64 0>
%3 = bitcast <4 x i64> %2 to <8 x i32>
%4 = shufflevector <8 x i32> %3, <8 x i32> undef, <8 x i32> <i32 0, i32 0, i32 0, i32 0, i32 4, i32 4, i32 4, i32 4>
%5 = trunc <8 x i32> %4 to <8 x i16>
ret <8 x i16> %5
}
define <16 x i8> @packsswb_icmp_128_zero(<8 x i16> %a0) {
; SSE-LABEL: packsswb_icmp_128_zero:
; SSE: # %bb.0:
; SSE-NEXT: pxor %xmm1, %xmm1
; SSE-NEXT: pcmpeqw %xmm0, %xmm1
; SSE-NEXT: packsswb %xmm0, %xmm1
; SSE-NEXT: movq {{.*#+}} xmm0 = xmm1[0],zero
; SSE-NEXT: ret{{[l|q]}}
;
; AVX-LABEL: packsswb_icmp_128_zero:
; AVX: # %bb.0:
; AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX-NEXT: vpcmpeqw %xmm1, %xmm0, %xmm0
; AVX-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,2,4,6,8,10,12,14],zero,zero,zero,zero,zero,zero,zero,zero
; AVX-NEXT: ret{{[l|q]}}
%1 = icmp eq <8 x i16> %a0, zeroinitializer
%2 = sext <8 x i1> %1 to <8 x i8>
%3 = shufflevector <8 x i8> %2, <8 x i8> zeroinitializer, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
ret <16 x i8> %3
}
define <32 x i8> @packsswb_icmp_zero_256(<16 x i16> %a0) {
; SSE-LABEL: packsswb_icmp_zero_256:
; SSE: # %bb.0:
; SSE-NEXT: pxor %xmm2, %xmm2
; SSE-NEXT: pcmpeqw %xmm2, %xmm1
; SSE-NEXT: pcmpeqw %xmm2, %xmm0
; SSE-NEXT: packsswb %xmm1, %xmm0
; SSE-NEXT: shufps {{.*#+}} xmm2 = xmm2[0,1],xmm0[2,3]
; SSE-NEXT: pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5,6,7]
; SSE-NEXT: movaps %xmm2, %xmm1
; SSE-NEXT: ret{{[l|q]}}
;
; AVX1-LABEL: packsswb_icmp_zero_256:
; AVX1: # %bb.0:
; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2
; AVX1-NEXT: vpcmpeqw %xmm2, %xmm1, %xmm1
; AVX1-NEXT: vpcmpeqw %xmm2, %xmm0, %xmm0
; AVX1-NEXT: vpacksswb %xmm1, %xmm0, %xmm1
; AVX1-NEXT: vpacksswb %xmm0, %xmm0, %xmm0
; AVX1-NEXT: vinsertf128 $1, %xmm1, %ymm0, %ymm0
; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,3],ymm1[4,5],ymm0[6,7]
; AVX1-NEXT: ret{{[l|q]}}
;
; AVX2-LABEL: packsswb_icmp_zero_256:
; AVX2: # %bb.0:
; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX2-NEXT: vpcmpeqw %ymm1, %ymm0, %ymm0
; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm2
; AVX2-NEXT: vpacksswb %xmm2, %xmm0, %xmm0
; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,0,2,1]
; AVX2-NEXT: vpblendd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,3],ymm1[4,5],ymm0[6,7]
; AVX2-NEXT: ret{{[l|q]}}
%1 = icmp eq <16 x i16> %a0, zeroinitializer
%2 = sext <16 x i1> %1 to <16 x i8>
%3 = shufflevector <16 x i8> zeroinitializer, <16 x i8> %2, <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
ret <32 x i8> %3
}
|
