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
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
; RUN: llc < %s -debug-only=legalize-types 2>&1 | FileCheck %s --check-prefix=CHECK-LEGALIZATION
; RUN: llc < %s | FileCheck %s
; REQUIRES: asserts
target triple = "aarch64-unknown-linux-gnu"
attributes #0 = {"target-features"="+sve" uwtable}
declare <vscale x 2 x i64> @llvm.vector.insert.nxv2i64.v8i64(<vscale x 2 x i64>, <8 x i64>, i64)
declare <vscale x 2 x double> @llvm.vector.insert.nxv2f64.v8f64(<vscale x 2 x double>, <8 x double>, i64)
define <vscale x 2 x i64> @test_nxv2i64_v8i64(<vscale x 2 x i64> %a, <8 x i64> %b) #0 {
; CHECK-LEGALIZATION-LABEL: test_nxv2i64_v8i64:
; CHECK-LEGALIZATION: // %bb.0:
; CHECK-LEGALIZATION-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-LEGALIZATION-NEXT: .cfi_def_cfa_offset 16
; CHECK-LEGALIZATION-NEXT: .cfi_offset w29, -16
; CHECK-LEGALIZATION-NEXT: addvl sp, sp, #-3
; CHECK-LEGALIZATION-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x18, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 24 * VG
; CHECK-LEGALIZATION-NEXT: cntd x8
; CHECK-LEGALIZATION-NEXT: ptrue p0.d, vl2
; CHECK-LEGALIZATION-NEXT: mov w9, #2 // =0x2
; CHECK-LEGALIZATION-NEXT: sub x8, x8, #2
; CHECK-LEGALIZATION-NEXT: // kill: def $q1 killed $q1 def $z1
; CHECK-LEGALIZATION-NEXT: mov x10, sp
; CHECK-LEGALIZATION-NEXT: cmp x8, #2
; CHECK-LEGALIZATION-NEXT: mov z0.d, p0/m, z1.d
; CHECK-LEGALIZATION-NEXT: ptrue p0.d
; CHECK-LEGALIZATION-NEXT: csel x9, x8, x9, lo
; CHECK-LEGALIZATION-NEXT: cmp x8, #4
; CHECK-LEGALIZATION-NEXT: lsl x9, x9, #3
; CHECK-LEGALIZATION-NEXT: st1d { z0.d }, p0, [sp]
; CHECK-LEGALIZATION-NEXT: str q2, [x10, x9]
; CHECK-LEGALIZATION-NEXT: mov w9, #4 // =0x4
; CHECK-LEGALIZATION-NEXT: addvl x10, sp, #1
; CHECK-LEGALIZATION-NEXT: ld1d { z0.d }, p0/z, [sp]
; CHECK-LEGALIZATION-NEXT: csel x9, x8, x9, lo
; CHECK-LEGALIZATION-NEXT: cmp x8, #6
; CHECK-LEGALIZATION-NEXT: lsl x9, x9, #3
; CHECK-LEGALIZATION-NEXT: st1d { z0.d }, p0, [sp, #1, mul vl]
; CHECK-LEGALIZATION-NEXT: str q3, [x10, x9]
; CHECK-LEGALIZATION-NEXT: mov w9, #6 // =0x6
; CHECK-LEGALIZATION-NEXT: ld1d { z0.d }, p0/z, [sp, #1, mul vl]
; CHECK-LEGALIZATION-NEXT: csel x8, x8, x9, lo
; CHECK-LEGALIZATION-NEXT: addvl x9, sp, #2
; CHECK-LEGALIZATION-NEXT: lsl x8, x8, #3
; CHECK-LEGALIZATION-NEXT: st1d { z0.d }, p0, [sp, #2, mul vl]
; CHECK-LEGALIZATION-NEXT: str q4, [x9, x8]
; CHECK-LEGALIZATION-NEXT: ld1d { z0.d }, p0/z, [sp, #2, mul vl]
; CHECK-LEGALIZATION-NEXT: addvl sp, sp, #3
; CHECK-LEGALIZATION-NEXT: .cfi_def_cfa wsp, 16
; CHECK-LEGALIZATION-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-LEGALIZATION-NEXT: .cfi_def_cfa_offset 0
; CHECK-LEGALIZATION-NEXT: .cfi_restore w29
; CHECK-LEGALIZATION-NEXT: ret
;
; CHECK-LABEL: test_nxv2i64_v8i64:
; CHECK: // %bb.0:
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT: .cfi_def_cfa_offset 16
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: addvl sp, sp, #-3
; CHECK-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x18, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 24 * VG
; CHECK-NEXT: cntd x8
; CHECK-NEXT: ptrue p0.d, vl2
; CHECK-NEXT: mov w9, #2 // =0x2
; CHECK-NEXT: sub x8, x8, #2
; CHECK-NEXT: // kill: def $q1 killed $q1 def $z1
; CHECK-NEXT: mov x10, sp
; CHECK-NEXT: cmp x8, #2
; CHECK-NEXT: mov z0.d, p0/m, z1.d
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: csel x9, x8, x9, lo
; CHECK-NEXT: cmp x8, #4
; CHECK-NEXT: lsl x9, x9, #3
; CHECK-NEXT: st1d { z0.d }, p0, [sp]
; CHECK-NEXT: str q2, [x10, x9]
; CHECK-NEXT: mov w9, #4 // =0x4
; CHECK-NEXT: addvl x10, sp, #1
; CHECK-NEXT: ld1d { z0.d }, p0/z, [sp]
; CHECK-NEXT: csel x9, x8, x9, lo
; CHECK-NEXT: cmp x8, #6
; CHECK-NEXT: lsl x9, x9, #3
; CHECK-NEXT: st1d { z0.d }, p0, [sp, #1, mul vl]
; CHECK-NEXT: str q3, [x10, x9]
; CHECK-NEXT: mov w9, #6 // =0x6
; CHECK-NEXT: ld1d { z0.d }, p0/z, [sp, #1, mul vl]
; CHECK-NEXT: csel x8, x8, x9, lo
; CHECK-NEXT: addvl x9, sp, #2
; CHECK-NEXT: lsl x8, x8, #3
; CHECK-NEXT: st1d { z0.d }, p0, [sp, #2, mul vl]
; CHECK-NEXT: str q4, [x9, x8]
; CHECK-NEXT: ld1d { z0.d }, p0/z, [sp, #2, mul vl]
; CHECK-NEXT: addvl sp, sp, #3
; CHECK-NEXT: .cfi_def_cfa wsp, 16
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: .cfi_def_cfa_offset 0
; CHECK-NEXT: .cfi_restore w29
; CHECK-NEXT: ret
%r = call <vscale x 2 x i64> @llvm.vector.insert.nxv2i64.v8i64(<vscale x 2 x i64> %a, <8 x i64> %b, i64 0)
ret <vscale x 2 x i64> %r
}
define <vscale x 2 x double> @test_nxv2f64_v8f64(<vscale x 2 x double> %a, <8 x double> %b) #0 {
; CHECK-LEGALIZATION-LABEL: test_nxv2f64_v8f64:
; CHECK-LEGALIZATION: // %bb.0:
; CHECK-LEGALIZATION-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-LEGALIZATION-NEXT: .cfi_def_cfa_offset 16
; CHECK-LEGALIZATION-NEXT: .cfi_offset w29, -16
; CHECK-LEGALIZATION-NEXT: addvl sp, sp, #-3
; CHECK-LEGALIZATION-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x18, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 24 * VG
; CHECK-LEGALIZATION-NEXT: cntd x8
; CHECK-LEGALIZATION-NEXT: ptrue p0.d, vl2
; CHECK-LEGALIZATION-NEXT: mov w9, #2 // =0x2
; CHECK-LEGALIZATION-NEXT: sub x8, x8, #2
; CHECK-LEGALIZATION-NEXT: // kill: def $q1 killed $q1 def $z1
; CHECK-LEGALIZATION-NEXT: mov x10, sp
; CHECK-LEGALIZATION-NEXT: cmp x8, #2
; CHECK-LEGALIZATION-NEXT: mov z0.d, p0/m, z1.d
; CHECK-LEGALIZATION-NEXT: ptrue p0.d
; CHECK-LEGALIZATION-NEXT: csel x9, x8, x9, lo
; CHECK-LEGALIZATION-NEXT: cmp x8, #4
; CHECK-LEGALIZATION-NEXT: lsl x9, x9, #3
; CHECK-LEGALIZATION-NEXT: st1d { z0.d }, p0, [sp]
; CHECK-LEGALIZATION-NEXT: str q2, [x10, x9]
; CHECK-LEGALIZATION-NEXT: mov w9, #4 // =0x4
; CHECK-LEGALIZATION-NEXT: addvl x10, sp, #1
; CHECK-LEGALIZATION-NEXT: ld1d { z0.d }, p0/z, [sp]
; CHECK-LEGALIZATION-NEXT: csel x9, x8, x9, lo
; CHECK-LEGALIZATION-NEXT: cmp x8, #6
; CHECK-LEGALIZATION-NEXT: lsl x9, x9, #3
; CHECK-LEGALIZATION-NEXT: st1d { z0.d }, p0, [sp, #1, mul vl]
; CHECK-LEGALIZATION-NEXT: str q3, [x10, x9]
; CHECK-LEGALIZATION-NEXT: mov w9, #6 // =0x6
; CHECK-LEGALIZATION-NEXT: ld1d { z0.d }, p0/z, [sp, #1, mul vl]
; CHECK-LEGALIZATION-NEXT: csel x8, x8, x9, lo
; CHECK-LEGALIZATION-NEXT: addvl x9, sp, #2
; CHECK-LEGALIZATION-NEXT: lsl x8, x8, #3
; CHECK-LEGALIZATION-NEXT: st1d { z0.d }, p0, [sp, #2, mul vl]
; CHECK-LEGALIZATION-NEXT: str q4, [x9, x8]
; CHECK-LEGALIZATION-NEXT: ld1d { z0.d }, p0/z, [sp, #2, mul vl]
; CHECK-LEGALIZATION-NEXT: addvl sp, sp, #3
; CHECK-LEGALIZATION-NEXT: .cfi_def_cfa wsp, 16
; CHECK-LEGALIZATION-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-LEGALIZATION-NEXT: .cfi_def_cfa_offset 0
; CHECK-LEGALIZATION-NEXT: .cfi_restore w29
; CHECK-LEGALIZATION-NEXT: ret
;
; CHECK-LABEL: test_nxv2f64_v8f64:
; CHECK: // %bb.0:
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT: .cfi_def_cfa_offset 16
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: addvl sp, sp, #-3
; CHECK-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x18, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 24 * VG
; CHECK-NEXT: cntd x8
; CHECK-NEXT: ptrue p0.d, vl2
; CHECK-NEXT: mov w9, #2 // =0x2
; CHECK-NEXT: sub x8, x8, #2
; CHECK-NEXT: // kill: def $q1 killed $q1 def $z1
; CHECK-NEXT: mov x10, sp
; CHECK-NEXT: cmp x8, #2
; CHECK-NEXT: mov z0.d, p0/m, z1.d
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: csel x9, x8, x9, lo
; CHECK-NEXT: cmp x8, #4
; CHECK-NEXT: lsl x9, x9, #3
; CHECK-NEXT: st1d { z0.d }, p0, [sp]
; CHECK-NEXT: str q2, [x10, x9]
; CHECK-NEXT: mov w9, #4 // =0x4
; CHECK-NEXT: addvl x10, sp, #1
; CHECK-NEXT: ld1d { z0.d }, p0/z, [sp]
; CHECK-NEXT: csel x9, x8, x9, lo
; CHECK-NEXT: cmp x8, #6
; CHECK-NEXT: lsl x9, x9, #3
; CHECK-NEXT: st1d { z0.d }, p0, [sp, #1, mul vl]
; CHECK-NEXT: str q3, [x10, x9]
; CHECK-NEXT: mov w9, #6 // =0x6
; CHECK-NEXT: ld1d { z0.d }, p0/z, [sp, #1, mul vl]
; CHECK-NEXT: csel x8, x8, x9, lo
; CHECK-NEXT: addvl x9, sp, #2
; CHECK-NEXT: lsl x8, x8, #3
; CHECK-NEXT: st1d { z0.d }, p0, [sp, #2, mul vl]
; CHECK-NEXT: str q4, [x9, x8]
; CHECK-NEXT: ld1d { z0.d }, p0/z, [sp, #2, mul vl]
; CHECK-NEXT: addvl sp, sp, #3
; CHECK-NEXT: .cfi_def_cfa wsp, 16
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: .cfi_def_cfa_offset 0
; CHECK-NEXT: .cfi_restore w29
; CHECK-NEXT: ret
%r = call <vscale x 2 x double> @llvm.vector.insert.nxv2f64.v8f64(<vscale x 2 x double> %a, <8 x double> %b, i64 0)
ret <vscale x 2 x double> %r
}
|
