1 ; RUN: llc -aarch64-sve-vector-bits-min=128 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=16 -check-prefix=NO_SVE
2 ; RUN: llc -aarch64-sve-vector-bits-min=256 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=32
3 ; RUN: llc -aarch64-sve-vector-bits-min=384 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=32
4 ; RUN: llc -aarch64-sve-vector-bits-min=512 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=64 -check-prefixes=CHECK,VBITS_GE_512
5 ; RUN: llc -aarch64-sve-vector-bits-min=640 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=64 -check-prefixes=CHECK,VBITS_GE_512
6 ; RUN: llc -aarch64-sve-vector-bits-min=768 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=64 -check-prefixes=CHECK,VBITS_GE_512
7 ; RUN: llc -aarch64-sve-vector-bits-min=896 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=64 -check-prefixes=CHECK,VBITS_GE_512
8 ; RUN: llc -aarch64-sve-vector-bits-min=1024 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=128 -check-prefixes=CHECK,VBITS_GE_512,VBITS_GE_1024
9 ; RUN: llc -aarch64-sve-vector-bits-min=1152 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=128 -check-prefixes=CHECK,VBITS_GE_512,VBITS_GE_1024
10 ; RUN: llc -aarch64-sve-vector-bits-min=1280 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=128 -check-prefixes=CHECK,VBITS_GE_512,VBITS_GE_1024
11 ; RUN: llc -aarch64-sve-vector-bits-min=1408 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=128 -check-prefixes=CHECK,VBITS_GE_512,VBITS_GE_1024
12 ; RUN: llc -aarch64-sve-vector-bits-min=1536 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=128 -check-prefixes=CHECK,VBITS_GE_512,VBITS_GE_1024
13 ; RUN: llc -aarch64-sve-vector-bits-min=1664 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=128 -check-prefixes=CHECK,VBITS_GE_512,VBITS_GE_1024
14 ; RUN: llc -aarch64-sve-vector-bits-min=1792 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=128 -check-prefixes=CHECK,VBITS_GE_512,VBITS_GE_1024
15 ; RUN: llc -aarch64-sve-vector-bits-min=1920 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=128 -check-prefixes=CHECK,VBITS_GE_512,VBITS_GE_1024
16 ; RUN: llc -aarch64-sve-vector-bits-min=2048 -asm-verbose=0 < %s | FileCheck %s -D#VBYTES=256 -check-prefixes=CHECK,VBITS_GE_512,VBITS_GE_1024,VBITS_GE_2048
18 target triple = "aarch64-unknown-linux-gnu"
20 ; Don't use SVE when its registers are no bigger than NEON.
23 ; Don't use SVE for 64-bit vectors.
24 define <4 x half> @select_v4f16(<4 x half> %op1, <4 x half> %op2, <4 x i1> %mask) #0 {
25 ; CHECK-LABEL: select_v4f16:
26 ; CHECK: bif v0.8b, v1.8b, v2.8b
28 %sel = select <4 x i1> %mask, <4 x half> %op1, <4 x half> %op2
32 ; Don't use SVE for 128-bit vectors.
33 define <8 x half> @select_v8f16(<8 x half> %op1, <8 x half> %op2, <8 x i1> %mask) #0 {
34 ; CHECK-LABEL: select_v8f16:
35 ; CHECK: bif v0.16b, v1.16b, v2.16b
37 %sel = select <8 x i1> %mask, <8 x half> %op1, <8 x half> %op2
41 define void @select_v16f16(<16 x half>* %a, <16 x half>* %b, <16 x i1>* %c) #0 {
42 ; CHECK-LABEL: select_v16f16:
43 ; CHECK: ptrue [[PG:p[0-9]+]].h, vl[[#min(div(VBYTES,2),16)]]
44 ; CHECK: ptrue [[PG1:p[0-9]+]].h
45 ; CHECK: ld1h { [[MASK:z[0-9]+]].h }, [[PG]]/z, [x9]
46 ; CHECK-NEXT: ld1h { [[OP1:z[0-9]+]].h }, [[PG]]/z, [x0]
47 ; CHECK-NEXT: ld1h { [[OP2:z[0-9]+]].h }, [[PG]]/z, [x1]
48 ; CHECK-NEXT: and [[AND:z[0-9]+]].h, [[MASK]].h, #0x1
49 ; CHECK-NEXT: cmpne [[COND:p[0-9]+]].h, [[PG1]]/z, [[AND]].h, #0
50 ; CHECK-NEXT: sel [[RES:z[0-9]+]].h, [[COND]], [[OP1]].h, [[OP2]].h
51 ; CHECK-NEXT: st1h { [[RES]].h }, [[PG]], [x0]
53 %mask = load <16 x i1>, <16 x i1>* %c
54 %op1 = load <16 x half>, <16 x half>* %a
55 %op2 = load <16 x half>, <16 x half>* %b
56 %sel = select <16 x i1> %mask, <16 x half> %op1, <16 x half> %op2
57 store <16 x half> %sel, <16 x half>* %a
61 define void @select_v32f16(<32 x half>* %a, <32 x half>* %b, <32 x i1>* %c) #0 {
62 ; CHECK-LABEL: select_v32f16:
63 ; CHECK: ptrue [[PG:p[0-9]+]].h, vl[[#min(div(VBYTES,2),32)]]
64 ; CHECK: ptrue [[PG1:p[0-9]+]].h
65 ; VBITS_GE_512: ld1h { [[MASK:z[0-9]+]].h }, [[PG]]/z, [x9]
66 ; VBITS_GE_512-NEXT: ld1h { [[OP1:z[0-9]+]].h }, [[PG]]/z, [x0]
67 ; VBITS_GE_512-NEXT: ld1h { [[OP2:z[0-9]+]].h }, [[PG]]/z, [x1]
68 ; VBITS_GE_512-NEXT: and [[AND:z[0-9]+]].h, [[MASK]].h, #0x1
69 ; VBITS_GE_512-NEXT: cmpne [[COND:p[0-9]+]].h, [[PG1]]/z, [[AND]].h, #0
70 ; VBITS_GE_512-NEXT: sel [[RES:z[0-9]+]].h, [[COND]], [[OP1]].h, [[OP2]].h
71 ; VBITS_GE_512-NEXT: st1h { [[RES]].h }, [[PG]], [x0]
73 %mask = load <32 x i1>, <32 x i1>* %c
74 %op1 = load <32 x half>, <32 x half>* %a
75 %op2 = load <32 x half>, <32 x half>* %b
76 %sel = select <32 x i1> %mask, <32 x half> %op1, <32 x half> %op2
77 store <32 x half> %sel, <32 x half>* %a
81 define void @select_v64f16(<64 x half>* %a, <64 x half>* %b, <64 x i1>* %c) #0 {
82 ; CHECK-LABEL: select_v64f16:
83 ; CHECK: ptrue [[PG:p[0-9]+]].h, vl[[#min(div(VBYTES,2),64)]]
84 ; CHECK: ptrue [[PG1:p[0-9]+]].h
85 ; VBITS_GE_1024: ld1h { [[MASK:z[0-9]+]].h }, [[PG]]/z, [x9]
86 ; VBITS_GE_1024-NEXT: ld1h { [[OP1:z[0-9]+]].h }, [[PG]]/z, [x0]
87 ; VBITS_GE_1024-NEXT: ld1h { [[OP2:z[0-9]+]].h }, [[PG]]/z, [x1]
88 ; VBITS_GE_1024-NEXT: and [[AND:z[0-9]+]].h, [[MASK]].h, #0x1
89 ; VBITS_GE_1024-NEXT: cmpne [[COND:p[0-9]+]].h, [[PG1]]/z, [[AND]].h, #0
90 ; VBITS_GE_1024-NEXT: sel [[RES:z[0-9]+]].h, [[COND]], [[OP1]].h, [[OP2]].h
91 ; VBITS_GE_1024-NEXT: st1h { [[RES]].h }, [[PG]], [x0]
93 %mask = load <64 x i1>, <64 x i1>* %c
94 %op1 = load <64 x half>, <64 x half>* %a
95 %op2 = load <64 x half>, <64 x half>* %b
96 %sel = select <64 x i1> %mask, <64 x half> %op1, <64 x half> %op2
97 store <64 x half> %sel, <64 x half>* %a
101 define void @select_v128f16(<128 x half>* %a, <128 x half>* %b, <128 x i1>* %c) #0 {
102 ; CHECK-LABEL: select_v128f16:
103 ; CHECK: ptrue [[PG:p[0-9]+]].h, vl[[#min(div(VBYTES,2),128)]]
104 ; CHECK: ptrue [[PG1:p[0-9]+]].h
105 ; VBITS_GE_2048: ld1h { [[MASK:z[0-9]+]].h }, [[PG]]/z, [x9]
106 ; VBITS_GE_2048-NEXT: ld1h { [[OP1:z[0-9]+]].h }, [[PG]]/z, [x0]
107 ; VBITS_GE_2048-NEXT: ld1h { [[OP2:z[0-9]+]].h }, [[PG]]/z, [x1]
108 ; VBITS_GE_2048-NEXT: and [[AND:z[0-9]+]].h, [[MASK]].h, #0x1
109 ; VBITS_GE_2048-NEXT: cmpne [[COND:p[0-9]+]].h, [[PG1]]/z, [[AND]].h, #0
110 ; VBITS_GE_2048-NEXT: sel [[RES:z[0-9]+]].h, [[COND]], [[OP1]].h, [[OP2]].h
111 ; VBITS_GE_2048-NEXT: st1h { [[RES]].h }, [[PG]], [x0]
113 %mask = load <128 x i1>, <128 x i1>* %c
114 %op1 = load <128 x half>, <128 x half>* %a
115 %op2 = load <128 x half>, <128 x half>* %b
116 %sel = select <128 x i1> %mask, <128 x half> %op1, <128 x half> %op2
117 store <128 x half> %sel, <128 x half>* %a
121 ; Don't use SVE for 64-bit vectors.
122 define <2 x float> @select_v2f32(<2 x float> %op1, <2 x float> %op2, <2 x i1> %mask) #0 {
123 ; CHECK-LABEL: select_v2f32:
124 ; CHECK: bif v0.8b, v1.8b, v2.8b
126 %sel = select <2 x i1> %mask, <2 x float> %op1, <2 x float> %op2
130 ; Don't use SVE for 128-bit vectors.
131 define <4 x float> @select_v4f32(<4 x float> %op1, <4 x float> %op2, <4 x i1> %mask) #0 {
132 ; CHECK-LABEL: select_v4f32:
133 ; CHECK: bif v0.16b, v1.16b, v2.16b
135 %sel = select <4 x i1> %mask, <4 x float> %op1, <4 x float> %op2
139 define void @select_v8f32(<8 x float>* %a, <8 x float>* %b, <8 x i1>* %c) #0 {
140 ; CHECK-LABEL: select_v8f32:
141 ; CHECK: ptrue [[PG:p[0-9]+]].s, vl[[#min(div(VBYTES,4),8)]]
142 ; CHECK: ptrue [[PG1:p[0-9]+]].s
143 ; CHECK: ld1w { [[MASK:z[0-9]+]].s }, [[PG]]/z, [x9]
144 ; CHECK-NEXT: ld1w { [[OP1:z[0-9]+]].s }, [[PG]]/z, [x0]
145 ; CHECK-NEXT: ld1w { [[OP2:z[0-9]+]].s }, [[PG]]/z, [x1]
146 ; CHECK-NEXT: and [[AND:z[0-9]+]].s, [[MASK]].s, #0x1
147 ; CHECK-NEXT: cmpne [[COND:p[0-9]+]].s, [[PG1]]/z, [[AND]].s, #0
148 ; CHECK-NEXT: sel [[RES:z[0-9]+]].s, [[COND]], [[OP1]].s, [[OP2]].s
149 ; CHECK-NEXT: st1w { [[RES]].s }, [[PG]], [x0]
151 %mask = load <8 x i1>, <8 x i1>* %c
152 %op1 = load <8 x float>, <8 x float>* %a
153 %op2 = load <8 x float>, <8 x float>* %b
154 %sel = select <8 x i1> %mask, <8 x float> %op1, <8 x float> %op2
155 store <8 x float> %sel, <8 x float>* %a
159 define void @select_v16f32(<16 x float>* %a, <16 x float>* %b, <16 x i1>* %c) #0 {
160 ; CHECK-LABEL: select_v16f32:
161 ; CHECK: ptrue [[PG:p[0-9]+]].s, vl[[#min(div(VBYTES,4),16)]]
162 ; CHECK: ptrue [[PG1:p[0-9]+]].s
163 ; VBITS_GE_512: ld1w { [[MASK:z[0-9]+]].s }, [[PG]]/z, [x9]
164 ; VBITS_GE_512-NEXT: ld1w { [[OP1:z[0-9]+]].s }, [[PG]]/z, [x0]
165 ; VBITS_GE_512-NEXT: ld1w { [[OP2:z[0-9]+]].s }, [[PG]]/z, [x1]
166 ; VBITS_GE_512-NEXT: and [[AND:z[0-9]+]].s, [[MASK]].s, #0x1
167 ; VBITS_GE_512-NEXT: cmpne [[COND:p[0-9]+]].s, [[PG1]]/z, [[AND]].s, #0
168 ; VBITS_GE_512-NEXT: sel [[RES:z[0-9]+]].s, [[COND]], [[OP1]].s, [[OP2]].s
169 ; VBITS_GE_512-NEXT: st1w { [[RES]].s }, [[PG]], [x0]
171 %mask = load <16 x i1>, <16 x i1>* %c
172 %op1 = load <16 x float>, <16 x float>* %a
173 %op2 = load <16 x float>, <16 x float>* %b
174 %sel = select <16 x i1> %mask, <16 x float> %op1, <16 x float> %op2
175 store <16 x float> %sel, <16 x float>* %a
179 define void @select_v32f32(<32 x float>* %a, <32 x float>* %b, <32 x i1>* %c) #0 {
180 ; CHECK-LABEL: select_v32f32:
181 ; CHECK: ptrue [[PG:p[0-9]+]].s, vl[[#min(div(VBYTES,4),32)]]
182 ; CHECK: ptrue [[PG1:p[0-9]+]].s
183 ; VBITS_GE_1024: ld1w { [[MASK:z[0-9]+]].s }, [[PG]]/z, [x9]
184 ; VBITS_GE_1024-NEXT: ld1w { [[OP1:z[0-9]+]].s }, [[PG]]/z, [x0]
185 ; VBITS_GE_1024-NEXT: ld1w { [[OP2:z[0-9]+]].s }, [[PG]]/z, [x1]
186 ; VBITS_GE_1024-NEXT: and [[AND:z[0-9]+]].s, [[MASK]].s, #0x1
187 ; VBITS_GE_1024-NEXT: cmpne [[COND:p[0-9]+]].s, [[PG1]]/z, [[AND]].s, #0
188 ; VBITS_GE_1024-NEXT: sel [[RES:z[0-9]+]].s, [[COND]], [[OP1]].s, [[OP2]].s
189 ; VBITS_GE_1024-NEXT: st1w { [[RES]].s }, [[PG]], [x0]
191 %mask = load <32 x i1>, <32 x i1>* %c
192 %op1 = load <32 x float>, <32 x float>* %a
193 %op2 = load <32 x float>, <32 x float>* %b
194 %sel = select <32 x i1> %mask, <32 x float> %op1, <32 x float> %op2
195 store <32 x float> %sel, <32 x float>* %a
199 define void @select_v64f32(<64 x float>* %a, <64 x float>* %b, <64 x i1>* %c) #0 {
200 ; CHECK-LABEL: select_v64f32:
201 ; CHECK: ptrue [[PG:p[0-9]+]].s, vl[[#min(div(VBYTES,4),64)]]
202 ; CHECK: ptrue [[PG1:p[0-9]+]].s
203 ; VBITS_GE_2048: ld1w { [[MASK:z[0-9]+]].s }, [[PG]]/z, [x9]
204 ; VBITS_GE_2048-NEXT: ld1w { [[OP1:z[0-9]+]].s }, [[PG]]/z, [x0]
205 ; VBITS_GE_2048-NEXT: ld1w { [[OP2:z[0-9]+]].s }, [[PG]]/z, [x1]
206 ; VBITS_GE_2048-NEXT: and [[AND:z[0-9]+]].s, [[MASK]].s, #0x1
207 ; VBITS_GE_2048-NEXT: cmpne [[COND:p[0-9]+]].s, [[PG1]]/z, [[AND]].s, #0
208 ; VBITS_GE_2048-NEXT: sel [[RES:z[0-9]+]].s, [[COND]], [[OP1]].s, [[OP2]].s
209 ; VBITS_GE_2048-NEXT: st1w { [[RES]].s }, [[PG]], [x0]
211 %mask = load <64 x i1>, <64 x i1>* %c
212 %op1 = load <64 x float>, <64 x float>* %a
213 %op2 = load <64 x float>, <64 x float>* %b
214 %sel = select <64 x i1> %mask, <64 x float> %op1, <64 x float> %op2
215 store <64 x float> %sel, <64 x float>* %a
219 ; Don't use SVE for 64-bit vectors.
220 define <1 x double> @select_v1f64(<1 x double> %op1, <1 x double> %op2, <1 x i1> %mask) #0 {
221 ; CHECK-LABEL: select_v1f64:
222 ; CHECK: bif v0.8b, v1.8b, v2.8b
224 %sel = select <1 x i1> %mask, <1 x double> %op1, <1 x double> %op2
225 ret <1 x double> %sel
228 ; Don't use SVE for 128-bit vectors.
229 define <2 x double> @select_v2f64(<2 x double> %op1, <2 x double> %op2, <2 x i1> %mask) #0 {
230 ; CHECK-LABEL: select_v2f64:
231 ; CHECK: bif v0.16b, v1.16b, v2.16b
233 %sel = select <2 x i1> %mask, <2 x double> %op1, <2 x double> %op2
234 ret <2 x double> %sel
237 define void @select_v4f64(<4 x double>* %a, <4 x double>* %b, <4 x i1>* %c) #0 {
238 ; CHECK-LABEL: select_v4f64:
239 ; CHECK: ptrue [[PG:p[0-9]+]].d, vl[[#min(div(VBYTES,8),4)]]
240 ; CHECK: ptrue [[PG1:p[0-9]+]].d
241 ; CHECK: ld1d { [[MASK:z[0-9]+]].d }, [[PG]]/z, [x9]
242 ; CHECK-NEXT: ld1d { [[OP1:z[0-9]+]].d }, [[PG]]/z, [x0]
243 ; CHECK-NEXT: ld1d { [[OP2:z[0-9]+]].d }, [[PG]]/z, [x1]
244 ; CHECK-NEXT: and [[AND:z[0-9]+]].d, [[MASK]].d, #0x1
245 ; CHECK-NEXT: cmpne [[COND:p[0-9]+]].d, [[PG1]]/z, [[AND]].d, #0
246 ; CHECK-NEXT: sel [[RES:z[0-9]+]].d, [[COND]], [[OP1]].d, [[OP2]].d
247 ; CHECK-NEXT: st1d { [[RES]].d }, [[PG]], [x0]
249 %mask = load <4 x i1>, <4 x i1>* %c
250 %op1 = load <4 x double>, <4 x double>* %a
251 %op2 = load <4 x double>, <4 x double>* %b
252 %sel = select <4 x i1> %mask, <4 x double> %op1, <4 x double> %op2
253 store <4 x double> %sel, <4 x double>* %a
257 define void @select_v8f64(<8 x double>* %a, <8 x double>* %b, <8 x i1>* %c) #0 {
258 ; CHECK-LABEL: select_v8f64:
259 ; CHECK: ptrue [[PG:p[0-9]+]].d, vl[[#min(div(VBYTES,8),8)]]
260 ; CHECK: ptrue [[PG1:p[0-9]+]].d
261 ; VBITS_GE_512: ld1d { [[MASK:z[0-9]+]].d }, [[PG]]/z, [x9]
262 ; VBITS_GE_512-NEXT: ld1d { [[OP1:z[0-9]+]].d }, [[PG]]/z, [x0]
263 ; VBITS_GE_512-NEXT: ld1d { [[OP2:z[0-9]+]].d }, [[PG]]/z, [x1]
264 ; VBITS_GE_512-NEXT: and [[AND:z[0-9]+]].d, [[MASK]].d, #0x1
265 ; VBITS_GE_512-NEXT: cmpne [[COND:p[0-9]+]].d, [[PG1]]/z, [[AND]].d, #0
266 ; VBITS_GE_512-NEXT: sel [[RES:z[0-9]+]].d, [[COND]], [[OP1]].d, [[OP2]].d
267 ; VBITS_GE_512-NEXT: st1d { [[RES]].d }, [[PG]], [x0]
269 %mask = load <8 x i1>, <8 x i1>* %c
270 %op1 = load <8 x double>, <8 x double>* %a
271 %op2 = load <8 x double>, <8 x double>* %b
272 %sel = select <8 x i1> %mask, <8 x double> %op1, <8 x double> %op2
273 store <8 x double> %sel, <8 x double>* %a
277 define void @select_v16f64(<16 x double>* %a, <16 x double>* %b, <16 x i1>* %c) #0 {
278 ; CHECK-LABEL: select_v16f64:
279 ; CHECK: ptrue [[PG:p[0-9]+]].d, vl[[#min(div(VBYTES,8),16)]]
280 ; CHECK: ptrue [[PG1:p[0-9]+]].d
281 ; VBITS_GE_1024: ld1d { [[MASK:z[0-9]+]].d }, [[PG]]/z, [x9]
282 ; VBITS_GE_1024-NEXT: ld1d { [[OP1:z[0-9]+]].d }, [[PG]]/z, [x0]
283 ; VBITS_GE_1024-NEXT: ld1d { [[OP2:z[0-9]+]].d }, [[PG]]/z, [x1]
284 ; VBITS_GE_1024-NEXT: and [[AND:z[0-9]+]].d, [[MASK]].d, #0x1
285 ; VBITS_GE_1024-NEXT: cmpne [[COND:p[0-9]+]].d, [[PG1]]/z, [[AND]].d, #0
286 ; VBITS_GE_1024-NEXT: sel [[RES:z[0-9]+]].d, [[COND]], [[OP1]].d, [[OP2]].d
287 ; VBITS_GE_1024-NEXT: st1d { [[RES]].d }, [[PG]], [x0]
289 %mask = load <16 x i1>, <16 x i1>* %c
290 %op1 = load <16 x double>, <16 x double>* %a
291 %op2 = load <16 x double>, <16 x double>* %b
292 %sel = select <16 x i1> %mask, <16 x double> %op1, <16 x double> %op2
293 store <16 x double> %sel, <16 x double>* %a
297 define void @select_v32f64(<32 x double>* %a, <32 x double>* %b, <32 x i1>* %c) #0 {
298 ; CHECK-LABEL: select_v32f64:
299 ; CHECK: ptrue [[PG:p[0-9]+]].d, vl[[#min(div(VBYTES,8),32)]]
300 ; CHECK: ptrue [[PG1:p[0-9]+]].d
301 ; VBITS_GE_2048: ld1d { [[MASK:z[0-9]+]].d }, [[PG]]/z, [x9]
302 ; VBITS_GE_2048-NEXT: ld1d { [[OP1:z[0-9]+]].d }, [[PG]]/z, [x0]
303 ; VBITS_GE_2048-NEXT: ld1d { [[OP2:z[0-9]+]].d }, [[PG]]/z, [x1]
304 ; VBITS_GE_2048-NEXT: and [[AND:z[0-9]+]].d, [[MASK]].d, #0x1
305 ; VBITS_GE_2048-NEXT: cmpne [[COND:p[0-9]+]].d, [[PG1]]/z, [[AND]].d, #0
306 ; VBITS_GE_2048-NEXT: sel [[RES:z[0-9]+]].d, [[COND]], [[OP1]].d, [[OP2]].d
307 ; VBITS_GE_2048-NEXT: st1d { [[RES]].d }, [[PG]], [x0]
309 %mask = load <32 x i1>, <32 x i1>* %c
310 %op1 = load <32 x double>, <32 x double>* %a
311 %op2 = load <32 x double>, <32 x double>* %b
312 %sel = select <32 x i1> %mask, <32 x double> %op1, <32 x double> %op2
313 store <32 x double> %sel, <32 x double>* %a
317 attributes #0 = { "target-features"="+sve" }