aboutsummaryrefslogtreecommitdiffstats
path: root/test/CodeGen/Mips/msa/elm_copy.ll
blob: ed3e52cbffc2942c5f574756c65039b6002d3f26 (plain)
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
; Test the MSA intrinsics that are encoded with the ELM instruction format and
; are element extraction operations.

; RUN: llc -march=mips -mattr=+msa,+fp64 < %s | FileCheck %s
; RUN: llc -march=mipsel -mattr=+msa,+fp64 < %s | FileCheck %s

@llvm_mips_copy_s_b_ARG1 = global <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>, align 16
@llvm_mips_copy_s_b_RES  = global i32 0, align 16

define void @llvm_mips_copy_s_b_test() nounwind {
entry:
  %0 = load <16 x i8>* @llvm_mips_copy_s_b_ARG1
  %1 = tail call i32 @llvm.mips.copy.s.b(<16 x i8> %0, i32 1)
  store i32 %1, i32* @llvm_mips_copy_s_b_RES
  ret void
}

declare i32 @llvm.mips.copy.s.b(<16 x i8>, i32) nounwind

; CHECK: llvm_mips_copy_s_b_test:
; CHECK: ld.b
; CHECK: copy_s.b
; CHECK: sw
; CHECK: .size llvm_mips_copy_s_b_test
;
@llvm_mips_copy_s_h_ARG1 = global <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>, align 16
@llvm_mips_copy_s_h_RES  = global i32 0, align 16

define void @llvm_mips_copy_s_h_test() nounwind {
entry:
  %0 = load <8 x i16>* @llvm_mips_copy_s_h_ARG1
  %1 = tail call i32 @llvm.mips.copy.s.h(<8 x i16> %0, i32 1)
  store i32 %1, i32* @llvm_mips_copy_s_h_RES
  ret void
}

declare i32 @llvm.mips.copy.s.h(<8 x i16>, i32) nounwind

; CHECK: llvm_mips_copy_s_h_test:
; CHECK: ld.h
; CHECK: copy_s.h
; CHECK: sw
; CHECK: .size llvm_mips_copy_s_h_test
;
@llvm_mips_copy_s_w_ARG1 = global <4 x i32> <i32 0, i32 1, i32 2, i32 3>, align 16
@llvm_mips_copy_s_w_RES  = global i32 0, align 16

define void @llvm_mips_copy_s_w_test() nounwind {
entry:
  %0 = load <4 x i32>* @llvm_mips_copy_s_w_ARG1
  %1 = tail call i32 @llvm.mips.copy.s.w(<4 x i32> %0, i32 1)
  store i32 %1, i32* @llvm_mips_copy_s_w_RES
  ret void
}

declare i32 @llvm.mips.copy.s.w(<4 x i32>, i32) nounwind

; CHECK: llvm_mips_copy_s_w_test:
; CHECK: ld.w
; CHECK: copy_s.w
; CHECK: sw
; CHECK: .size llvm_mips_copy_s_w_test
;
@llvm_mips_copy_s_d_ARG1 = global <2 x i64> <i64 0, i64 1>, align 16
@llvm_mips_copy_s_d_RES  = global i64 0, align 16

define void @llvm_mips_copy_s_d_test() nounwind {
entry:
  %0 = load <2 x i64>* @llvm_mips_copy_s_d_ARG1
  %1 = tail call i64 @llvm.mips.copy.s.d(<2 x i64> %0, i32 1)
  store i64 %1, i64* @llvm_mips_copy_s_d_RES
  ret void
}

declare i64 @llvm.mips.copy.s.d(<2 x i64>, i32) nounwind

; CHECK: llvm_mips_copy_s_d_test:
; CHECK: ld.w
; CHECK: copy_s.w
; CHECK: copy_s.w
; CHECK: sw
; CHECK: sw
; CHECK: .size llvm_mips_copy_s_d_test
;
@llvm_mips_copy_u_b_ARG1 = global <16 x i8> <i8 0, i8 1, i8 2, i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10, i8 11, i8 12, i8 13, i8 14, i8 15>, align 16
@llvm_mips_copy_u_b_RES  = global i32 0, align 16

define void @llvm_mips_copy_u_b_test() nounwind {
entry:
  %0 = load <16 x i8>* @llvm_mips_copy_u_b_ARG1
  %1 = tail call i32 @llvm.mips.copy.u.b(<16 x i8> %0, i32 1)
  store i32 %1, i32* @llvm_mips_copy_u_b_RES
  ret void
}

declare i32 @llvm.mips.copy.u.b(<16 x i8>, i32) nounwind

; CHECK: llvm_mips_copy_u_b_test:
; CHECK: ld.b
; CHECK: copy_u.b
; CHECK: sw
; CHECK: .size llvm_mips_copy_u_b_test
;
@llvm_mips_copy_u_h_ARG1 = global <8 x i16> <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>, align 16
@llvm_mips_copy_u_h_RES  = global i32 0, align 16

define void @llvm_mips_copy_u_h_test() nounwind {
entry:
  %0 = load <8 x i16>* @llvm_mips_copy_u_h_ARG1
  %1 = tail call i32 @llvm.mips.copy.u.h(<8 x i16> %0, i32 1)
  store i32 %1, i32* @llvm_mips_copy_u_h_RES
  ret void
}

declare i32 @llvm.mips.copy.u.h(<8 x i16>, i32) nounwind

; CHECK: llvm_mips_copy_u_h_test:
; CHECK: ld.h
; CHECK: copy_u.h
; CHECK: sw
; CHECK: .size llvm_mips_copy_u_h_test
;
@llvm_mips_copy_u_w_ARG1 = global <4 x i32> <i32 0, i32 1, i32 2, i32 3>, align 16
@llvm_mips_copy_u_w_RES  = global i32 0, align 16

define void @llvm_mips_copy_u_w_test() nounwind {
entry:
  %0 = load <4 x i32>* @llvm_mips_copy_u_w_ARG1
  %1 = tail call i32 @llvm.mips.copy.u.w(<4 x i32> %0, i32 1)
  store i32 %1, i32* @llvm_mips_copy_u_w_RES
  ret void
}

declare i32 @llvm.mips.copy.u.w(<4 x i32>, i32) nounwind

; CHECK: llvm_mips_copy_u_w_test:
; CHECK: ld.w
; CHECK: copy_u.w
; CHECK: sw
; CHECK: .size llvm_mips_copy_u_w_test
;
@llvm_mips_copy_u_d_ARG1 = global <2 x i64> <i64 0, i64 1>, align 16
@llvm_mips_copy_u_d_RES  = global i64 0, align 16

define void @llvm_mips_copy_u_d_test() nounwind {
entry:
  %0 = load <2 x i64>* @llvm_mips_copy_u_d_ARG1
  %1 = tail call i64 @llvm.mips.copy.u.d(<2 x i64> %0, i32 1)
  store i64 %1, i64* @llvm_mips_copy_u_d_RES
  ret void
}

declare i64 @llvm.mips.copy.u.d(<2 x i64>, i32) nounwind

; CHECK: llvm_mips_copy_u_d_test:
; CHECK: ld.w
; CHECK: copy_s.w
; CHECK: copy_s.w
; CHECK: sw
; CHECK: sw
; CHECK: .size llvm_mips_copy_u_d_test
;