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
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=aarch64--linux-gnu -mattr=sve < %s | FileCheck %s
; ld2b
define { <vscale x 16 x i8>, <vscale x 16 x i8> } @ld2.nxv32i8(<vscale x 16 x i1> %Pg, i8 *%addr, i64 %a) {
; CHECK-LABEL: ld2.nxv32i8:
; CHECK: // %bb.0:
; CHECK-NEXT: ld2b { z0.b, z1.b }, p0/z, [x0, x1]
; CHECK-NEXT: ret
%addr2 = getelementptr i8, i8 * %addr, i64 %a
%res = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ld2.sret.nxv32i8.nxv16i1.p0i8(<vscale x 16 x i1> %Pg, i8 *%addr2)
ret { <vscale x 16 x i8>, <vscale x 16 x i8> } %res
}
; ld2h
define { <vscale x 8 x i16>, <vscale x 8 x i16> } @ld2.nxv16i16(<vscale x 8 x i1> %Pg, i16 *%addr, i64 %a) {
; CHECK-LABEL: ld2.nxv16i16:
; CHECK: // %bb.0:
; CHECK-NEXT: ld2h { z0.h, z1.h }, p0/z, [x0, x1, lsl #1]
; CHECK-NEXT: ret
%addr2 = getelementptr i16, i16 * %addr, i64 %a
%res = call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ld2.sret.nxv16i16.nxv8i1.p0i16(<vscale x 8 x i1> %Pg, i16 *%addr2)
ret { <vscale x 8 x i16>, <vscale x 8 x i16> } %res
}
define { <vscale x 8 x half>, <vscale x 8 x half> } @ld2.nxv16f16(<vscale x 8 x i1> %Pg, half *%addr, i64 %a) {
; CHECK-LABEL: ld2.nxv16f16:
; CHECK: // %bb.0:
; CHECK-NEXT: ld2h { z0.h, z1.h }, p0/z, [x0, x1, lsl #1]
; CHECK-NEXT: ret
%addr2 = getelementptr half, half * %addr, i64 %a
%res = call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sve.ld2.sret.nxv16f16.nxv8i1.p0f16(<vscale x 8 x i1> %Pg, half *%addr2)
ret { <vscale x 8 x half>, <vscale x 8 x half> } %res
}
define { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @ld2.nxv16bf16(<vscale x 8 x i1> %Pg, bfloat *%addr, i64 %a) #0 {
; CHECK-LABEL: ld2.nxv16bf16:
; CHECK: // %bb.0:
; CHECK-NEXT: ld2h { z0.h, z1.h }, p0/z, [x0, x1, lsl #1]
; CHECK-NEXT: ret
%addr2 = getelementptr bfloat, bfloat * %addr, i64 %a
%res = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sve.ld2.sret.nxv16bf16.nxv8i1.p0bf16(<vscale x 8 x i1> %Pg, bfloat *%addr2)
ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } %res
}
; ld2w
define { <vscale x 4 x i32>, <vscale x 4 x i32> } @ld2.nxv8i32(<vscale x 4 x i1> %Pg, i32 *%addr, i64 %a) {
; CHECK-LABEL: ld2.nxv8i32:
; CHECK: // %bb.0:
; CHECK-NEXT: ld2w { z0.s, z1.s }, p0/z, [x0, x1, lsl #2]
; CHECK-NEXT: ret
%addr2 = getelementptr i32, i32 * %addr, i64 %a
%res = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ld2.sret.nxv8i32.nxv4i1.p0i32(<vscale x 4 x i1> %Pg, i32 *%addr2)
ret { <vscale x 4 x i32>, <vscale x 4 x i32> } %res
}
define { <vscale x 4 x float>, <vscale x 4 x float> } @ld2.nxv8f32(<vscale x 4 x i1> %Pg, float *%addr, i64 %a) {
; CHECK-LABEL: ld2.nxv8f32:
; CHECK: // %bb.0:
; CHECK-NEXT: ld2w { z0.s, z1.s }, p0/z, [x0, x1, lsl #2]
; CHECK-NEXT: ret
%addr2 = getelementptr float, float * %addr, i64 %a
%res = call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sve.ld2.sret.nxv8f32.nxv4i1.p0f32(<vscale x 4 x i1> %Pg, float *%addr2)
ret { <vscale x 4 x float>, <vscale x 4 x float> } %res
}
; ld2d
define { <vscale x 2 x i64>, <vscale x 2 x i64> } @ld2.nxv4i64(<vscale x 2 x i1> %Pg, i64 *%addr, i64 %a) {
; CHECK-LABEL: ld2.nxv4i64:
; CHECK: // %bb.0:
; CHECK-NEXT: ld2d { z0.d, z1.d }, p0/z, [x0, x1, lsl #3]
; CHECK-NEXT: ret
%addr2 = getelementptr i64, i64 * %addr, i64 %a
%res = call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ld2.sret.nxv4i64.nxv2i1.p0i64(<vscale x 2 x i1> %Pg, i64 *%addr2)
ret { <vscale x 2 x i64>, <vscale x 2 x i64> } %res
}
define { <vscale x 2 x double>, <vscale x 2 x double> } @ld2.nxv4f64(<vscale x 2 x i1> %Pg, double *%addr, i64 %a) {
; CHECK-LABEL: ld2.nxv4f64:
; CHECK: // %bb.0:
; CHECK-NEXT: ld2d { z0.d, z1.d }, p0/z, [x0, x1, lsl #3]
; CHECK-NEXT: ret
%addr2 = getelementptr double, double * %addr, i64 %a
%res = call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sve.ld2.sret.nxv4f64.nxv2i1.p0f64(<vscale x 2 x i1> %Pg, double *%addr2)
ret { <vscale x 2 x double>, <vscale x 2 x double> } %res
}
; ld3b
define { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @ld3.nxv48i8(<vscale x 16 x i1> %Pg, i8 *%addr, i64 %a) {
; CHECK-LABEL: ld3.nxv48i8:
; CHECK: // %bb.0:
; CHECK-NEXT: ld3b { z0.b, z1.b, z2.b }, p0/z, [x0, x1]
; CHECK-NEXT: ret
%addr2 = getelementptr i8, i8 * %addr, i64 %a
%res = call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ld3.sret.nxv48i8.nxv16i1.p0i8(<vscale x 16 x i1> %Pg, i8 *%addr2)
ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res
}
; ld3h
define { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @ld3.nxv24i16(<vscale x 8 x i1> %Pg, i16 *%addr, i64 %a) {
; CHECK-LABEL: ld3.nxv24i16:
; CHECK: // %bb.0:
; CHECK-NEXT: ld3h { z0.h, z1.h, z2.h }, p0/z, [x0, x1, lsl #1]
; CHECK-NEXT: ret
%addr2 = getelementptr i16, i16 * %addr, i64 %a
%res = call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ld3.sret.nxv24i16.nxv8i1.p0i16(<vscale x 8 x i1> %Pg, i16 *%addr2)
ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res
}
define { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @ld3.nxv24f16(<vscale x 8 x i1> %Pg, half *%addr, i64 %a) {
; CHECK-LABEL: ld3.nxv24f16:
; CHECK: // %bb.0:
; CHECK-NEXT: ld3h { z0.h, z1.h, z2.h }, p0/z, [x0, x1, lsl #1]
; CHECK-NEXT: ret
%addr2 = getelementptr half, half * %addr, i64 %a
%res = call { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sve.ld3.sret.nxv24f16.nxv8i1.p0f16(<vscale x 8 x i1> %Pg, half *%addr2)
ret { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } %res
}
define { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @ld3.nxv24bf16(<vscale x 8 x i1> %Pg, bfloat *%addr, i64 %a) #0 {
; CHECK-LABEL: ld3.nxv24bf16:
; CHECK: // %bb.0:
; CHECK-NEXT: ld3h { z0.h, z1.h, z2.h }, p0/z, [x0, x1, lsl #1]
; CHECK-NEXT: ret
%addr2 = getelementptr bfloat, bfloat * %addr, i64 %a
%res = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sve.ld3.sret.nxv24bf16.nxv8i1.p0bf16(<vscale x 8 x i1> %Pg, bfloat *%addr2)
ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } %res
}
; ld3w
define { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @ld3.nxv12i32(<vscale x 4 x i1> %Pg, i32 *%addr, i64 %a) {
; CHECK-LABEL: ld3.nxv12i32:
; CHECK: // %bb.0:
; CHECK-NEXT: ld3w { z0.s, z1.s, z2.s }, p0/z, [x0, x1, lsl #2]
; CHECK-NEXT: ret
%addr2 = getelementptr i32, i32 * %addr, i64 %a
%res = call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ld3.sret.nxv12i32.nxv4i1.p0i32(<vscale x 4 x i1> %Pg, i32 *%addr2)
ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res
}
define { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @ld3.nxv12f32(<vscale x 4 x i1> %Pg, float *%addr, i64 %a) {
; CHECK-LABEL: ld3.nxv12f32:
; CHECK: // %bb.0:
; CHECK-NEXT: ld3w { z0.s, z1.s, z2.s }, p0/z, [x0, x1, lsl #2]
; CHECK-NEXT: ret
%addr2 = getelementptr float, float * %addr, i64 %a
%res = call { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sve.ld3.sret.nxv12f32.nxv4i1.p0f32(<vscale x 4 x i1> %Pg, float *%addr2)
ret { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } %res
}
; ld3d
define { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @ld3.nxv6i64(<vscale x 2 x i1> %Pg, i64 *%addr, i64 %a) {
; CHECK-LABEL: ld3.nxv6i64:
; CHECK: // %bb.0:
; CHECK-NEXT: ld3d { z0.d, z1.d, z2.d }, p0/z, [x0, x1, lsl #3]
; CHECK-NEXT: ret
%addr2 = getelementptr i64, i64 * %addr, i64 %a
%res = call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ld3.sret.nxv6i64.nxv2i1.p0i64(<vscale x 2 x i1> %Pg, i64 *%addr2)
ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res
}
define { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @ld3.nxv6f64(<vscale x 2 x i1> %Pg, double *%addr, i64 %a) {
; CHECK-LABEL: ld3.nxv6f64:
; CHECK: // %bb.0:
; CHECK-NEXT: ld3d { z0.d, z1.d, z2.d }, p0/z, [x0, x1, lsl #3]
; CHECK-NEXT: ret
%addr2 = getelementptr double, double * %addr, i64 %a
%res = call { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sve.ld3.sret.nxv6f64.nxv2i1.p0f64(<vscale x 2 x i1> %Pg, double *%addr2)
ret { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } %res
}
; ld4b
define { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @ld4.nxv64i8(<vscale x 16 x i1> %Pg, i8 *%addr, i64 %a) {
; CHECK-LABEL: ld4.nxv64i8:
; CHECK: // %bb.0:
; CHECK-NEXT: ld4b { z0.b, z1.b, z2.b, z3.b }, p0/z, [x0, x1]
; CHECK-NEXT: ret
%addr2 = getelementptr i8, i8 * %addr, i64 %a
%res = call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ld4.sret.nxv64i8.nxv16i1.p0i8(<vscale x 16 x i1> %Pg, i8 *%addr2)
ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } %res
}
; ld4h
define { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @ld4.nxv32i16(<vscale x 8 x i1> %Pg, i16 *%addr, i64 %a) {
; CHECK-LABEL: ld4.nxv32i16:
; CHECK: // %bb.0:
; CHECK-NEXT: ld4h { z0.h, z1.h, z2.h, z3.h }, p0/z, [x0, x1, lsl #1]
; CHECK-NEXT: ret
%addr2 = getelementptr i16, i16 * %addr, i64 %a
%res = call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ld4.sret.nxv32i16.nxv8i1.p0i16(<vscale x 8 x i1> %Pg, i16 *%addr2)
ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } %res
}
define { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @ld4.nxv32f16(<vscale x 8 x i1> %Pg, half *%addr, i64 %a) {
; CHECK-LABEL: ld4.nxv32f16:
; CHECK: // %bb.0:
; CHECK-NEXT: ld4h { z0.h, z1.h, z2.h, z3.h }, p0/z, [x0, x1, lsl #1]
; CHECK-NEXT: ret
%addr2 = getelementptr half, half * %addr, i64 %a
%res = call { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sve.ld4.sret.nxv32f16.nxv8i1.p0f16(<vscale x 8 x i1> %Pg, half *%addr2)
ret { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } %res
}
define { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @ld4.nxv32bf16(<vscale x 8 x i1> %Pg, bfloat *%addr, i64 %a) #0 {
; CHECK-LABEL: ld4.nxv32bf16:
; CHECK: // %bb.0:
; CHECK-NEXT: ld4h { z0.h, z1.h, z2.h, z3.h }, p0/z, [x0, x1, lsl #1]
; CHECK-NEXT: ret
%addr2 = getelementptr bfloat, bfloat * %addr, i64 %a
%res = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sve.ld4.sret.nxv32bf16.nxv8i1.p0bf16(<vscale x 8 x i1> %Pg, bfloat *%addr2)
ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } %res
}
; ld4w
define { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @ld4.nxv16i32(<vscale x 4 x i1> %Pg, i32 *%addr, i64 %a) {
; CHECK-LABEL: ld4.nxv16i32:
; CHECK: // %bb.0:
; CHECK-NEXT: ld4w { z0.s, z1.s, z2.s, z3.s }, p0/z, [x0, x1, lsl #2]
; CHECK-NEXT: ret
%addr2 = getelementptr i32, i32 * %addr, i64 %a
%res = call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ld4.sret.nxv16i32.nxv4i1.p0i32(<vscale x 4 x i1> %Pg, i32 *%addr2)
ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } %res
}
define { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @ld4.nxv16f32(<vscale x 4 x i1> %Pg, float *%addr, i64 %a) {
; CHECK-LABEL: ld4.nxv16f32:
; CHECK: // %bb.0:
; CHECK-NEXT: ld4w { z0.s, z1.s, z2.s, z3.s }, p0/z, [x0, x1, lsl #2]
; CHECK-NEXT: ret
%addr2 = getelementptr float, float * %addr, i64 %a
%res = call { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sve.ld4.sret.nxv16f32.nxv4i1.p0f32(<vscale x 4 x i1> %Pg, float *%addr2)
ret { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } %res
}
; ld4d
define { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @ld4.nxv8i64(<vscale x 2 x i1> %Pg, i64 *%addr, i64 %a) {
; CHECK-LABEL: ld4.nxv8i64:
; CHECK: // %bb.0:
; CHECK-NEXT: ld4d { z0.d, z1.d, z2.d, z3.d }, p0/z, [x0, x1, lsl #3]
; CHECK-NEXT: ret
%addr2 = getelementptr i64, i64 * %addr, i64 %a
%res = call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ld4.sret.nxv8i64.nxv2i1.p0i64(<vscale x 2 x i1> %Pg, i64 *%addr2)
ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } %res
}
define { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @ld4.nxv8f64(<vscale x 2 x i1> %Pg, double *%addr, i64 %a) {
; CHECK-LABEL: ld4.nxv8f64:
; CHECK: // %bb.0:
; CHECK-NEXT: ld4d { z0.d, z1.d, z2.d, z3.d }, p0/z, [x0, x1, lsl #3]
; CHECK-NEXT: ret
%addr2 = getelementptr double, double * %addr, i64 %a
%res = call { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sve.ld4.sret.nxv8f64.nxv2i1.p0f64(<vscale x 2 x i1> %Pg, double *%addr2)
ret { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } %res
}
declare { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ld2.sret.nxv32i8.nxv16i1.p0i8(<vscale x 16 x i1>, i8*)
declare { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ld2.sret.nxv16i16.nxv8i1.p0i16(<vscale x 8 x i1>, i16*)
declare { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ld2.sret.nxv8i32.nxv4i1.p0i32(<vscale x 4 x i1>, i32*)
declare { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ld2.sret.nxv4i64.nxv2i1.p0i64(<vscale x 2 x i1>, i64*)
declare { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sve.ld2.sret.nxv16f16.nxv8i1.p0f16(<vscale x 8 x i1>, half*)
declare { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sve.ld2.sret.nxv16bf16.nxv8i1.p0bf16(<vscale x 8 x i1>, bfloat*)
declare { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sve.ld2.sret.nxv8f32.nxv4i1.p0f32(<vscale x 4 x i1>, float*)
declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sve.ld2.sret.nxv4f64.nxv2i1.p0f64(<vscale x 2 x i1>, double*)
declare { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ld3.sret.nxv48i8.nxv16i1.p0i8(<vscale x 16 x i1>, i8*)
declare { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ld3.sret.nxv24i16.nxv8i1.p0i16(<vscale x 8 x i1>, i16*)
declare { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ld3.sret.nxv12i32.nxv4i1.p0i32(<vscale x 4 x i1>, i32*)
declare { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ld3.sret.nxv6i64.nxv2i1.p0i64(<vscale x 2 x i1>, i64*)
declare { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sve.ld3.sret.nxv24f16.nxv8i1.p0f16(<vscale x 8 x i1>, half*)
declare { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sve.ld3.sret.nxv24bf16.nxv8i1.p0bf16(<vscale x 8 x i1>, bfloat*)
declare { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sve.ld3.sret.nxv12f32.nxv4i1.p0f32(<vscale x 4 x i1>, float*)
declare { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sve.ld3.sret.nxv6f64.nxv2i1.p0f64(<vscale x 2 x i1>, double*)
declare { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sve.ld4.sret.nxv64i8.nxv16i1.p0i8(<vscale x 16 x i1>, i8*)
declare { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sve.ld4.sret.nxv32i16.nxv8i1.p0i16(<vscale x 8 x i1>, i16*)
declare { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sve.ld4.sret.nxv16i32.nxv4i1.p0i32(<vscale x 4 x i1>, i32*)
declare { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sve.ld4.sret.nxv8i64.nxv2i1.p0i64(<vscale x 2 x i1>, i64*)
declare { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sve.ld4.sret.nxv32f16.nxv8i1.p0f16(<vscale x 8 x i1>, half*)
declare { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sve.ld4.sret.nxv32bf16.nxv8i1.p0bf16(<vscale x 8 x i1>, bfloat*)
declare { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sve.ld4.sret.nxv16f32.nxv4i1.p0f32(<vscale x 4 x i1>, float*)
declare { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sve.ld4.sret.nxv8f64.nxv2i1.p0f64(<vscale x 2 x i1>, double*)
; +bf16 is required for the bfloat version.
attributes #0 = { "target-features"="+sve,+bf16" }
|
