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
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sve < %s | FileCheck %s
; EXTRACT VECTOR ELT
define i32 @promote_extract_2i32_idx(<vscale x 2 x i32> %a, i32 %idx) {
; CHECK-LABEL: promote_extract_2i32_idx:
; CHECK: // %bb.0:
; CHECK-NEXT: // kill: def $w0 killed $w0 def $x0
; CHECK-NEXT: sxtw x8, w0
; CHECK-NEXT: whilels p0.d, xzr, x8
; CHECK-NEXT: lastb x0, p0, z0.d
; CHECK-NEXT: // kill: def $w0 killed $w0 killed $x0
; CHECK-NEXT: ret
%ext = extractelement <vscale x 2 x i32> %a, i32 %idx
ret i32 %ext
}
define i8 @split_extract_32i8_idx(<vscale x 32 x i8> %a, i32 %idx) {
; CHECK-LABEL: split_extract_32i8_idx:
; CHECK: // %bb.0:
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT: addvl sp, sp, #-2
; CHECK-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x10, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 16 * VG
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: mov x8, #-1
; CHECK-NEXT: // kill: def $w0 killed $w0 def $x0
; CHECK-NEXT: sxtw x9, w0
; CHECK-NEXT: ptrue p0.b
; CHECK-NEXT: st1b { z1.b }, p0, [sp, #1, mul vl]
; CHECK-NEXT: st1b { z0.b }, p0, [sp]
; CHECK-NEXT: addvl x8, x8, #2
; CHECK-NEXT: cmp x9, x8
; CHECK-NEXT: csel x8, x9, x8, lo
; CHECK-NEXT: mov x9, sp
; CHECK-NEXT: ldrb w0, [x9, x8]
; CHECK-NEXT: addvl sp, sp, #2
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = extractelement <vscale x 32 x i8> %a, i32 %idx
ret i8 %ext
}
define i16 @split_extract_16i16_idx(<vscale x 16 x i16> %a, i32 %idx) {
; CHECK-LABEL: split_extract_16i16_idx:
; CHECK: // %bb.0:
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT: addvl sp, sp, #-2
; CHECK-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x10, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 16 * VG
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: mov x8, #-1
; CHECK-NEXT: // kill: def $w0 killed $w0 def $x0
; CHECK-NEXT: sxtw x9, w0
; CHECK-NEXT: ptrue p0.h
; CHECK-NEXT: st1h { z1.h }, p0, [sp, #1, mul vl]
; CHECK-NEXT: st1h { z0.h }, p0, [sp]
; CHECK-NEXT: addvl x8, x8, #1
; CHECK-NEXT: cmp x9, x8
; CHECK-NEXT: csel x8, x9, x8, lo
; CHECK-NEXT: mov x9, sp
; CHECK-NEXT: ldrh w0, [x9, x8, lsl #1]
; CHECK-NEXT: addvl sp, sp, #2
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = extractelement <vscale x 16 x i16> %a, i32 %idx
ret i16 %ext
}
define i32 @split_extract_8i32_idx(<vscale x 8 x i32> %a, i32 %idx) {
; CHECK-LABEL: split_extract_8i32_idx:
; CHECK: // %bb.0:
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT: addvl sp, sp, #-2
; CHECK-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x10, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 16 * VG
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: cnth x8
; CHECK-NEXT: // kill: def $w0 killed $w0 def $x0
; CHECK-NEXT: sxtw x9, w0
; CHECK-NEXT: sub x8, x8, #1
; CHECK-NEXT: ptrue p0.s
; CHECK-NEXT: cmp x9, x8
; CHECK-NEXT: st1w { z1.s }, p0, [sp, #1, mul vl]
; CHECK-NEXT: csel x8, x9, x8, lo
; CHECK-NEXT: mov x9, sp
; CHECK-NEXT: st1w { z0.s }, p0, [sp]
; CHECK-NEXT: ldr w0, [x9, x8, lsl #2]
; CHECK-NEXT: addvl sp, sp, #2
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = extractelement <vscale x 8 x i32> %a, i32 %idx
ret i32 %ext
}
define i64 @split_extract_8i64_idx(<vscale x 8 x i64> %a, i32 %idx) {
; CHECK-LABEL: split_extract_8i64_idx:
; CHECK: // %bb.0:
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT: addvl sp, sp, #-4
; CHECK-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x20, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 32 * VG
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: cnth x8
; CHECK-NEXT: // kill: def $w0 killed $w0 def $x0
; CHECK-NEXT: sxtw x9, w0
; CHECK-NEXT: sub x8, x8, #1
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: cmp x9, x8
; CHECK-NEXT: st1d { z3.d }, p0, [sp, #3, mul vl]
; CHECK-NEXT: csel x8, x9, x8, lo
; CHECK-NEXT: mov x9, sp
; CHECK-NEXT: st1d { z2.d }, p0, [sp, #2, mul vl]
; CHECK-NEXT: st1d { z1.d }, p0, [sp, #1, mul vl]
; CHECK-NEXT: st1d { z0.d }, p0, [sp]
; CHECK-NEXT: ldr x0, [x9, x8, lsl #3]
; CHECK-NEXT: addvl sp, sp, #4
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = extractelement <vscale x 8 x i64> %a, i32 %idx
ret i64 %ext
}
; EXTRACT VECTOR ELT, CONSTANT IDX
define i16 @promote_extract_4i16(<vscale x 4 x i16> %a) {
; CHECK-LABEL: promote_extract_4i16:
; CHECK: // %bb.0:
; CHECK-NEXT: mov w0, v0.s[1]
; CHECK-NEXT: ret
%ext = extractelement <vscale x 4 x i16> %a, i32 1
ret i16 %ext
}
define i8 @split_extract_32i8(<vscale x 32 x i8> %a) {
; CHECK-LABEL: split_extract_32i8:
; CHECK: // %bb.0:
; CHECK-NEXT: umov w0, v0.b[3]
; CHECK-NEXT: ret
%ext = extractelement <vscale x 32 x i8> %a, i32 3
ret i8 %ext
}
define i16 @split_extract_16i16(<vscale x 16 x i16> %a) {
; CHECK-LABEL: split_extract_16i16:
; CHECK: // %bb.0:
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT: addvl sp, sp, #-2
; CHECK-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x10, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 16 * VG
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: mov x8, #-1
; CHECK-NEXT: mov w9, #128
; CHECK-NEXT: ptrue p0.h
; CHECK-NEXT: st1h { z1.h }, p0, [sp, #1, mul vl]
; CHECK-NEXT: st1h { z0.h }, p0, [sp]
; CHECK-NEXT: addvl x8, x8, #1
; CHECK-NEXT: cmp x8, #128
; CHECK-NEXT: csel x8, x8, x9, lo
; CHECK-NEXT: mov x9, sp
; CHECK-NEXT: ldrh w0, [x9, x8, lsl #1]
; CHECK-NEXT: addvl sp, sp, #2
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = extractelement <vscale x 16 x i16> %a, i32 128
ret i16 %ext
}
define i32 @split_extract_16i32(<vscale x 16 x i32> %a) {
; CHECK-LABEL: split_extract_16i32:
; CHECK: // %bb.0:
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT: addvl sp, sp, #-4
; CHECK-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x20, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 32 * VG
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: mov x8, #-1
; CHECK-NEXT: mov w9, #34464
; CHECK-NEXT: movk w9, #1, lsl #16
; CHECK-NEXT: ptrue p0.s
; CHECK-NEXT: st1w { z3.s }, p0, [sp, #3, mul vl]
; CHECK-NEXT: st1w { z2.s }, p0, [sp, #2, mul vl]
; CHECK-NEXT: addvl x8, x8, #1
; CHECK-NEXT: st1w { z1.s }, p0, [sp, #1, mul vl]
; CHECK-NEXT: cmp x8, x9
; CHECK-NEXT: st1w { z0.s }, p0, [sp]
; CHECK-NEXT: csel x8, x8, x9, lo
; CHECK-NEXT: mov x9, sp
; CHECK-NEXT: ldr w0, [x9, x8, lsl #2]
; CHECK-NEXT: addvl sp, sp, #4
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = extractelement <vscale x 16 x i32> %a, i32 100000
ret i32 %ext
}
define i64 @split_extract_4i64(<vscale x 4 x i64> %a) {
; CHECK-LABEL: split_extract_4i64:
; CHECK: // %bb.0:
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT: addvl sp, sp, #-2
; CHECK-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x10, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 16 * VG
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: cntw x8
; CHECK-NEXT: mov w9, #10
; CHECK-NEXT: sub x8, x8, #1
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: cmp x8, #10
; CHECK-NEXT: st1d { z1.d }, p0, [sp, #1, mul vl]
; CHECK-NEXT: csel x8, x8, x9, lo
; CHECK-NEXT: mov x9, sp
; CHECK-NEXT: st1d { z0.d }, p0, [sp]
; CHECK-NEXT: ldr x0, [x9, x8, lsl #3]
; CHECK-NEXT: addvl sp, sp, #2
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = extractelement <vscale x 4 x i64> %a, i32 10
ret i64 %ext
}
|
