avx2-pmovxrm.ll
8.07 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -disable-peephole -mtriple=i686-apple-darwin -mattr=+avx2 | FileCheck %s --check-prefix=X32
; RUN: llc < %s -disable-peephole -mtriple=x86_64-apple-darwin -mattr=+avx2 | FileCheck %s --check-prefix=X64
; RUN: llc < %s -disable-peephole -mtriple=i686-apple-darwin -mattr=+avx512vl,avx512bw | FileCheck %s --check-prefix=X32
; RUN: llc < %s -disable-peephole -mtriple=x86_64-apple-darwin -mattr=+avx512vl,avx512bw | FileCheck %s --check-prefix=X64
define <16 x i16> @test_llvm_x86_avx2_pmovsxbw(<16 x i8>* %a) {
; X32-LABEL: test_llvm_x86_avx2_pmovsxbw:
; X32: ## %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: vpmovsxbw (%eax), %ymm0
; X32-NEXT: retl
;
; X64-LABEL: test_llvm_x86_avx2_pmovsxbw:
; X64: ## %bb.0:
; X64-NEXT: vpmovsxbw (%rdi), %ymm0
; X64-NEXT: retq
%1 = load <16 x i8>, <16 x i8>* %a, align 1
%2 = sext <16 x i8> %1 to <16 x i16>
ret <16 x i16> %2
}
define <8 x i32> @test_llvm_x86_avx2_pmovsxbd(<16 x i8>* %a) {
; X32-LABEL: test_llvm_x86_avx2_pmovsxbd:
; X32: ## %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: vpmovsxbd (%eax), %ymm0
; X32-NEXT: retl
;
; X64-LABEL: test_llvm_x86_avx2_pmovsxbd:
; X64: ## %bb.0:
; X64-NEXT: vpmovsxbd (%rdi), %ymm0
; X64-NEXT: retq
%1 = load <16 x i8>, <16 x i8>* %a, align 1
%2 = shufflevector <16 x i8> %1, <16 x i8> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
%3 = sext <8 x i8> %2 to <8 x i32>
ret <8 x i32> %3
}
define <4 x i64> @test_llvm_x86_avx2_pmovsxbq(<16 x i8>* %a) {
; X32-LABEL: test_llvm_x86_avx2_pmovsxbq:
; X32: ## %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: vpmovsxbq (%eax), %ymm0
; X32-NEXT: retl
;
; X64-LABEL: test_llvm_x86_avx2_pmovsxbq:
; X64: ## %bb.0:
; X64-NEXT: vpmovsxbq (%rdi), %ymm0
; X64-NEXT: retq
%1 = load <16 x i8>, <16 x i8>* %a, align 1
%2 = shufflevector <16 x i8> %1, <16 x i8> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
%3 = sext <4 x i8> %2 to <4 x i64>
ret <4 x i64> %3
}
define <8 x i32> @test_llvm_x86_avx2_pmovsxwd(<8 x i16>* %a) {
; X32-LABEL: test_llvm_x86_avx2_pmovsxwd:
; X32: ## %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: vpmovsxwd (%eax), %ymm0
; X32-NEXT: retl
;
; X64-LABEL: test_llvm_x86_avx2_pmovsxwd:
; X64: ## %bb.0:
; X64-NEXT: vpmovsxwd (%rdi), %ymm0
; X64-NEXT: retq
%1 = load <8 x i16>, <8 x i16>* %a, align 1
%2 = sext <8 x i16> %1 to <8 x i32>
ret <8 x i32> %2
}
define <4 x i64> @test_llvm_x86_avx2_pmovsxwq(<8 x i16>* %a) {
; X32-LABEL: test_llvm_x86_avx2_pmovsxwq:
; X32: ## %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: vpmovsxwq (%eax), %ymm0
; X32-NEXT: retl
;
; X64-LABEL: test_llvm_x86_avx2_pmovsxwq:
; X64: ## %bb.0:
; X64-NEXT: vpmovsxwq (%rdi), %ymm0
; X64-NEXT: retq
%1 = load <8 x i16>, <8 x i16>* %a, align 1
%2 = shufflevector <8 x i16> %1, <8 x i16> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
%3 = sext <4 x i16> %2 to <4 x i64>
ret <4 x i64> %3
}
define <4 x i64> @test_llvm_x86_avx2_pmovsxdq(<4 x i32>* %a) {
; X32-LABEL: test_llvm_x86_avx2_pmovsxdq:
; X32: ## %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: vpmovsxdq (%eax), %ymm0
; X32-NEXT: retl
;
; X64-LABEL: test_llvm_x86_avx2_pmovsxdq:
; X64: ## %bb.0:
; X64-NEXT: vpmovsxdq (%rdi), %ymm0
; X64-NEXT: retq
%1 = load <4 x i32>, <4 x i32>* %a, align 1
%2 = sext <4 x i32> %1 to <4 x i64>
ret <4 x i64> %2
}
define <16 x i16> @test_llvm_x86_avx2_pmovzxbw(<16 x i8>* %a) {
; X32-LABEL: test_llvm_x86_avx2_pmovzxbw:
; X32: ## %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: vpmovzxbw {{.*#+}} ymm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero,mem[8],zero,mem[9],zero,mem[10],zero,mem[11],zero,mem[12],zero,mem[13],zero,mem[14],zero,mem[15],zero
; X32-NEXT: retl
;
; X64-LABEL: test_llvm_x86_avx2_pmovzxbw:
; X64: ## %bb.0:
; X64-NEXT: vpmovzxbw {{.*#+}} ymm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero,mem[8],zero,mem[9],zero,mem[10],zero,mem[11],zero,mem[12],zero,mem[13],zero,mem[14],zero,mem[15],zero
; X64-NEXT: retq
%1 = load <16 x i8>, <16 x i8>* %a, align 1
%2 = zext <16 x i8> %1 to <16 x i16>
ret <16 x i16> %2
}
define <8 x i32> @test_llvm_x86_avx2_pmovzxbd(<16 x i8>* %a) {
; X32-LABEL: test_llvm_x86_avx2_pmovzxbd:
; X32: ## %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: vpmovzxbd {{.*#+}} ymm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero
; X32-NEXT: retl
;
; X64-LABEL: test_llvm_x86_avx2_pmovzxbd:
; X64: ## %bb.0:
; X64-NEXT: vpmovzxbd {{.*#+}} ymm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero
; X64-NEXT: retq
%1 = load <16 x i8>, <16 x i8>* %a, align 1
%2 = shufflevector <16 x i8> %1, <16 x i8> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
%3 = zext <8 x i8> %2 to <8 x i32>
ret <8 x i32> %3
}
define <4 x i64> @test_llvm_x86_avx2_pmovzxbq(<16 x i8>* %a) {
; X32-LABEL: test_llvm_x86_avx2_pmovzxbq:
; X32: ## %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: vpmovzxbq {{.*#+}} ymm0 = mem[0],zero,zero,zero,zero,zero,zero,zero,mem[1],zero,zero,zero,zero,zero,zero,zero,mem[2],zero,zero,zero,zero,zero,zero,zero,mem[3],zero,zero,zero,zero,zero,zero,zero
; X32-NEXT: retl
;
; X64-LABEL: test_llvm_x86_avx2_pmovzxbq:
; X64: ## %bb.0:
; X64-NEXT: vpmovzxbq {{.*#+}} ymm0 = mem[0],zero,zero,zero,zero,zero,zero,zero,mem[1],zero,zero,zero,zero,zero,zero,zero,mem[2],zero,zero,zero,zero,zero,zero,zero,mem[3],zero,zero,zero,zero,zero,zero,zero
; X64-NEXT: retq
%1 = load <16 x i8>, <16 x i8>* %a, align 1
%2 = shufflevector <16 x i8> %1, <16 x i8> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
%3 = zext <4 x i8> %2 to <4 x i64>
ret <4 x i64> %3
}
define <8 x i32> @test_llvm_x86_avx2_pmovzxwd(<8 x i16>* %a) {
; X32-LABEL: test_llvm_x86_avx2_pmovzxwd:
; X32: ## %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: vpmovzxwd {{.*#+}} ymm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero
; X32-NEXT: retl
;
; X64-LABEL: test_llvm_x86_avx2_pmovzxwd:
; X64: ## %bb.0:
; X64-NEXT: vpmovzxwd {{.*#+}} ymm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero,mem[4],zero,mem[5],zero,mem[6],zero,mem[7],zero
; X64-NEXT: retq
%1 = load <8 x i16>, <8 x i16>* %a, align 1
%2 = zext <8 x i16> %1 to <8 x i32>
ret <8 x i32> %2
}
define <4 x i64> @test_llvm_x86_avx2_pmovzxwq(<8 x i16>* %a) {
; X32-LABEL: test_llvm_x86_avx2_pmovzxwq:
; X32: ## %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: vpmovzxwq {{.*#+}} ymm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero
; X32-NEXT: retl
;
; X64-LABEL: test_llvm_x86_avx2_pmovzxwq:
; X64: ## %bb.0:
; X64-NEXT: vpmovzxwq {{.*#+}} ymm0 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero
; X64-NEXT: retq
%1 = load <8 x i16>, <8 x i16>* %a, align 1
%2 = shufflevector <8 x i16> %1, <8 x i16> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
%3 = zext <4 x i16> %2 to <4 x i64>
ret <4 x i64> %3
}
define <4 x i64> @test_llvm_x86_avx2_pmovzxdq(<4 x i32>* %a) {
; X32-LABEL: test_llvm_x86_avx2_pmovzxdq:
; X32: ## %bb.0:
; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
; X32-NEXT: vpmovzxdq {{.*#+}} ymm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero
; X32-NEXT: retl
;
; X64-LABEL: test_llvm_x86_avx2_pmovzxdq:
; X64: ## %bb.0:
; X64-NEXT: vpmovzxdq {{.*#+}} ymm0 = mem[0],zero,mem[1],zero,mem[2],zero,mem[3],zero
; X64-NEXT: retq
%1 = load <4 x i32>, <4 x i32>* %a, align 1
%2 = zext <4 x i32> %1 to <4 x i64>
ret <4 x i64> %2
}