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
328
329
330
331
332
333
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sve < %s -o - | FileCheck %s
; Extracting a legal fixed-length vector from an illegal subvector
define <4 x i32> @extract_v4i32_nxv16i32_12(<vscale x 16 x i32> %arg) {
; CHECK-LABEL: extract_v4i32_nxv16i32_12:
; 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, #-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: 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: st1w { z1.s }, p0, [sp, #1, mul vl]
; CHECK-NEXT: st1w { z0.s }, p0, [sp]
; CHECK-NEXT: ldr q0, [sp, #48]
; CHECK-NEXT: addvl sp, sp, #4
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = call <4 x i32> @llvm.vector.extract.v4i32.nxv16i32(<vscale x 16 x i32> %arg, i64 12)
ret <4 x i32> %ext
}
define <8 x i16> @extract_v8i16_nxv32i16_8(<vscale x 32 x i16> %arg) {
; CHECK-LABEL: extract_v8i16_nxv32i16_8:
; 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, #-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: ptrue p0.h
; CHECK-NEXT: st1h { z1.h }, p0, [sp, #1, mul vl]
; CHECK-NEXT: st1h { z0.h }, p0, [sp]
; CHECK-NEXT: ldr q0, [sp, #16]
; CHECK-NEXT: addvl sp, sp, #2
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = call <8 x i16> @llvm.vector.extract.v8i16.nxv32i16(<vscale x 32 x i16> %arg, i64 8)
ret <8 x i16> %ext
}
define <4 x i16> @extract_v4i16_nxv32i16_8(<vscale x 32 x i16> %arg) {
; CHECK-LABEL: extract_v4i16_nxv32i16_8:
; 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, #-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: ptrue p0.h
; CHECK-NEXT: st1h { z3.h }, p0, [sp, #3, mul vl]
; CHECK-NEXT: st1h { z2.h }, p0, [sp, #2, mul vl]
; CHECK-NEXT: st1h { z1.h }, p0, [sp, #1, mul vl]
; CHECK-NEXT: st1h { z0.h }, p0, [sp]
; CHECK-NEXT: ldr d0, [sp, #32]
; CHECK-NEXT: addvl sp, sp, #4
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = call <4 x i16> @llvm.vector.extract.v4i16.nxv32i16(<vscale x 32 x i16> %arg, i64 16)
ret <4 x i16> %ext
}
; The result type gets promoted, leading to us extracting 2 elements from a nxv32i16.
; Hence we don't end up in SplitVecOp_EXTRACT_SUBVECTOR, but in SplitVecOp_EXTRACT_VECTOR_ELT instead.
define <2 x i16> @extract_v2i16_nxv32i16_8(<vscale x 32 x i16> %arg) {
; CHECK-LABEL: extract_v2i16_nxv32i16_8:
; 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, #-8
; CHECK-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0xc0, 0x00, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 64 * VG
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: ptrue p0.h
; CHECK-NEXT: add x8, x8, #32
; CHECK-NEXT: st1h { z3.h }, p0, [sp, #3, mul vl]
; CHECK-NEXT: st1h { z2.h }, p0, [sp, #2, mul vl]
; CHECK-NEXT: st1h { z1.h }, p0, [sp, #1, mul vl]
; CHECK-NEXT: st1h { z0.h }, p0, [sp]
; CHECK-NEXT: st1h { z3.h }, p0, [sp, #7, mul vl]
; CHECK-NEXT: st1h { z2.h }, p0, [sp, #6, mul vl]
; CHECK-NEXT: st1h { z1.h }, p0, [sp, #5, mul vl]
; CHECK-NEXT: st1h { z0.h }, p0, [sp, #4, mul vl]
; CHECK-NEXT: ld1 { v0.h }[0], [x8]
; CHECK-NEXT: addvl x8, sp, #4
; CHECK-NEXT: add x8, x8, #34
; CHECK-NEXT: ld1 { v0.h }[2], [x8]
; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-NEXT: addvl sp, sp, #8
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = call <2 x i16> @llvm.vector.extract.v2i16.nxv32i16(<vscale x 32 x i16> %arg, i64 16)
ret <2 x i16> %ext
}
define <2 x i64> @extract_v2i64_nxv8i64_8(<vscale x 8 x i64> %arg) {
; CHECK-LABEL: extract_v2i64_nxv8i64_8:
; 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, #-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: cnth x8
; CHECK-NEXT: mov w9, #8
; CHECK-NEXT: sub x8, x8, #2
; CHECK-NEXT: ptrue p0.d
; CHECK-NEXT: cmp x8, #8
; CHECK-NEXT: st1d { z3.d }, p0, [sp, #3, mul vl]
; CHECK-NEXT: csel x8, x8, x9, lo
; CHECK-NEXT: mov x9, sp
; CHECK-NEXT: lsl x8, x8, #3
; 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 q0, [x9, x8]
; CHECK-NEXT: addvl sp, sp, #4
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = call <2 x i64> @llvm.vector.extract.v2i64.nxv8i64(<vscale x 8 x i64> %arg, i64 8)
ret <2 x i64> %ext
}
define <4 x float> @extract_v4f32_nxv16f32_12(<vscale x 16 x float> %arg) {
; CHECK-LABEL: extract_v4f32_nxv16f32_12:
; 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, #-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: 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: st1w { z1.s }, p0, [sp, #1, mul vl]
; CHECK-NEXT: st1w { z0.s }, p0, [sp]
; CHECK-NEXT: ldr q0, [sp, #48]
; CHECK-NEXT: addvl sp, sp, #4
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = call <4 x float> @llvm.vector.extract.v4f32.nxv16f32(<vscale x 16 x float> %arg, i64 12)
ret <4 x float> %ext
}
define <2 x float> @extract_v2f32_nxv16f32_2(<vscale x 16 x float> %arg) {
; CHECK-LABEL: extract_v2f32_nxv16f32_2:
; 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, #-1
; CHECK-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 8 * VG
; CHECK-NEXT: ptrue p0.s
; CHECK-NEXT: st1w { z0.s }, p0, [sp]
; CHECK-NEXT: ldr d0, [sp, #8]
; CHECK-NEXT: addvl sp, sp, #1
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = call <2 x float> @llvm.vector.extract.v2f32.nxv16f32(<vscale x 16 x float> %arg, i64 2)
ret <2 x float> %ext
}
define <4 x i1> @extract_v4i1_nxv32i1_0(<vscale x 32 x i1> %arg) {
; CHECK-LABEL: extract_v4i1_nxv32i1_0:
; CHECK: // %bb.0:
; CHECK-NEXT: mov z1.b, p0/z, #1 // =0x1
; CHECK-NEXT: umov w8, v1.b[1]
; CHECK-NEXT: umov w9, v1.b[2]
; CHECK-NEXT: mov v0.16b, v1.16b
; CHECK-NEXT: mov v0.h[1], w8
; CHECK-NEXT: umov w8, v1.b[3]
; CHECK-NEXT: mov v0.h[2], w9
; CHECK-NEXT: mov v0.h[3], w8
; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-NEXT: ret
%ext = call <4 x i1> @llvm.vector.extract.v4i1.nxv32i1(<vscale x 32 x i1> %arg, i64 0)
ret <4 x i1> %ext
}
; The result type gets promoted, leading to us extracting 4 elements from a nxv32i16.
; Hence we don't end up in SplitVecOp_EXTRACT_SUBVECTOR, but in SplitVecOp_EXTRACT_VECTOR_ELT instead.
define <4 x i1> @extract_v4i1_nxv32i1_16(<vscale x 32 x i1> %arg) {
; CHECK-LABEL: extract_v4i1_nxv32i1_16:
; 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, #-8
; CHECK-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0xc0, 0x00, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 64 * VG
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: ptrue p2.b
; CHECK-NEXT: add x8, x8, #16
; CHECK-NEXT: mov z0.b, p1/z, #1 // =0x1
; CHECK-NEXT: mov z1.b, p0/z, #1 // =0x1
; CHECK-NEXT: st1b { z0.b }, p2, [sp, #1, mul vl]
; CHECK-NEXT: st1b { z1.b }, p2, [sp]
; CHECK-NEXT: st1b { z0.b }, p2, [sp, #3, mul vl]
; CHECK-NEXT: st1b { z1.b }, p2, [sp, #2, mul vl]
; CHECK-NEXT: st1b { z0.b }, p2, [sp, #5, mul vl]
; CHECK-NEXT: st1b { z1.b }, p2, [sp, #4, mul vl]
; CHECK-NEXT: st1b { z0.b }, p2, [sp, #7, mul vl]
; CHECK-NEXT: st1b { z1.b }, p2, [sp, #6, mul vl]
; CHECK-NEXT: ld1 { v0.b }[0], [x8]
; CHECK-NEXT: addvl x8, sp, #2
; CHECK-NEXT: add x8, x8, #17
; CHECK-NEXT: ld1 { v0.b }[2], [x8]
; CHECK-NEXT: addvl x8, sp, #4
; CHECK-NEXT: add x8, x8, #18
; CHECK-NEXT: ld1 { v0.b }[4], [x8]
; CHECK-NEXT: addvl x8, sp, #6
; CHECK-NEXT: add x8, x8, #19
; CHECK-NEXT: ld1 { v0.b }[6], [x8]
; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-NEXT: addvl sp, sp, #8
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = call <4 x i1> @llvm.vector.extract.v4i1.nxv32i1(<vscale x 32 x i1> %arg, i64 16)
ret <4 x i1> %ext
}
define <4 x i1> @extract_v4i1_v32i1_16(<32 x i1> %arg) {
; CHECK-LABEL: extract_v4i1_v32i1_16:
; CHECK: // %bb.0:
; CHECK-NEXT: ldr w8, [sp, #64]
; CHECK-NEXT: ldr w9, [sp, #72]
; CHECK-NEXT: fmov s0, w8
; CHECK-NEXT: ldr w8, [sp, #80]
; CHECK-NEXT: mov v0.h[1], w9
; CHECK-NEXT: mov v0.h[2], w8
; CHECK-NEXT: ldr w8, [sp, #88]
; CHECK-NEXT: mov v0.h[3], w8
; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-NEXT: ret
%ext = call <4 x i1> @llvm.vector.extract.v4i1.v32i1(<32 x i1> %arg, i64 16)
ret <4 x i1> %ext
}
; The result type gets promoted, leading to us extracting 4 elements from a nxv32i3.
; Hence we don't end up in SplitVecOp_EXTRACT_SUBVECTOR, but in SplitVecOp_EXTRACT_VECTOR_ELT instead.
define <4 x i3> @extract_v4i3_nxv32i3_16(<vscale x 32 x i3> %arg) {
; CHECK-LABEL: extract_v4i3_nxv32i3_16:
; 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, #-8
; CHECK-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0xc0, 0x00, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 64 * VG
; CHECK-NEXT: mov x8, sp
; CHECK-NEXT: ptrue p0.b
; CHECK-NEXT: add x8, x8, #16
; CHECK-NEXT: st1b { z1.b }, p0, [sp, #1, mul vl]
; CHECK-NEXT: st1b { z0.b }, p0, [sp]
; CHECK-NEXT: st1b { z1.b }, p0, [sp, #3, mul vl]
; CHECK-NEXT: st1b { z0.b }, p0, [sp, #2, mul vl]
; CHECK-NEXT: st1b { z1.b }, p0, [sp, #5, mul vl]
; CHECK-NEXT: st1b { z0.b }, p0, [sp, #4, mul vl]
; CHECK-NEXT: st1b { z1.b }, p0, [sp, #7, mul vl]
; CHECK-NEXT: st1b { z0.b }, p0, [sp, #6, mul vl]
; CHECK-NEXT: ld1 { v0.b }[0], [x8]
; CHECK-NEXT: addvl x8, sp, #2
; CHECK-NEXT: add x8, x8, #17
; CHECK-NEXT: ld1 { v0.b }[2], [x8]
; CHECK-NEXT: addvl x8, sp, #4
; CHECK-NEXT: add x8, x8, #18
; CHECK-NEXT: ld1 { v0.b }[4], [x8]
; CHECK-NEXT: addvl x8, sp, #6
; CHECK-NEXT: add x8, x8, #19
; CHECK-NEXT: ld1 { v0.b }[6], [x8]
; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-NEXT: addvl sp, sp, #8
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = call <4 x i3> @llvm.vector.extract.v4i3.nxv32i3(<vscale x 32 x i3> %arg, i64 16)
ret <4 x i3> %ext
}
; Extracting an illegal fixed-length vector from an illegal subvector
define <2 x i32> @extract_v2i32_nxv16i32_2(<vscale x 16 x i32> %arg) {
; CHECK-LABEL: extract_v2i32_nxv16i32_2:
; 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, #-1
; CHECK-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 8 * VG
; CHECK-NEXT: ptrue p0.s
; CHECK-NEXT: st1w { z0.s }, p0, [sp]
; CHECK-NEXT: ldr d0, [sp, #8]
; CHECK-NEXT: addvl sp, sp, #1
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = call <2 x i32> @llvm.vector.extract.v2i32.nxv16i32(<vscale x 16 x i32> %arg, i64 2)
ret <2 x i32> %ext
}
define <4 x i64> @extract_v4i64_nxv8i64_0(<vscale x 8 x i64> %arg) {
; CHECK-LABEL: extract_v4i64_nxv8i64_0:
; 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, #-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: ptrue p0.d
; CHECK-NEXT: st1d { z1.d }, p0, [sp, #1, mul vl]
; CHECK-NEXT: st1d { z0.d }, p0, [sp]
; CHECK-NEXT: ldr q1, [sp, #16]
; CHECK-NEXT: // kill: def $q0 killed $q0 killed $z0
; CHECK-NEXT: addvl sp, sp, #2
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
%ext = call <4 x i64> @llvm.vector.extract.v4i64.nxv8i64(<vscale x 8 x i64> %arg, i64 0)
ret <4 x i64> %ext
}
declare <2 x i64> @llvm.vector.extract.v2i64.nxv8i64(<vscale x 8 x i64>, i64)
declare <4 x i64> @llvm.vector.extract.v4i64.nxv8i64(<vscale x 8 x i64>, i64)
declare <4 x float> @llvm.vector.extract.v4f32.nxv16f32(<vscale x 16 x float>, i64)
declare <2 x float> @llvm.vector.extract.v2f32.nxv16f32(<vscale x 16 x float>, i64)
declare <4 x i32> @llvm.vector.extract.v4i32.nxv16i32(<vscale x 16 x i32>, i64)
declare <2 x i32> @llvm.vector.extract.v2i32.nxv16i32(<vscale x 16 x i32>, i64)
declare <8 x i16> @llvm.vector.extract.v8i16.nxv32i16(<vscale x 32 x i16>, i64)
declare <4 x i16> @llvm.vector.extract.v4i16.nxv32i16(<vscale x 32 x i16>, i64)
declare <2 x i16> @llvm.vector.extract.v2i16.nxv32i16(<vscale x 32 x i16>, i64)
declare <4 x i1> @llvm.vector.extract.v4i1.nxv32i1(<vscale x 32 x i1>, i64)
declare <4 x i1> @llvm.vector.extract.v4i1.v32i1(<32 x i1>, i64)
declare <4 x i3> @llvm.vector.extract.v4i3.nxv32i3(<vscale x 32 x i3>, i64)
|
