| blob | 50a2e41f45756d60c1471804c38532abf0bc7c74 |
1 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
2 ; RUN: llc < %s -mtriple=aarch64 -mattr=+sve2 -aarch64-stack-hazard-size=0 | FileCheck %s --check-prefixes=CHECK,CHECK0
3 ; RUN: llc < %s -mtriple=aarch64 -mattr=+sve2 -aarch64-stack-hazard-size=64 | FileCheck %s --check-prefixes=CHECK,CHECK64
4 ; RUN: llc < %s -mtriple=aarch64 -mattr=+sve2 -aarch64-stack-hazard-size=1024 | FileCheck %s --check-prefixes=CHECK,CHECK1024
6 define i32 @basic(i32 noundef %num) {
7 ; CHECK-LABEL: basic:
8 ; CHECK: // %bb.0: // %entry
9 ; CHECK-NEXT: mov w0, wzr
10 ; CHECK-NEXT: ret
11 entry:
12 ret i32 0
13 }
15 ; Non-streaming functions don't need hazards
16 define i32 @csr_d8_notsc(i32 noundef %num) {
17 ; CHECK-LABEL: csr_d8_notsc:
18 ; CHECK: // %bb.0: // %entry
19 ; CHECK-NEXT: str d8, [sp, #-16]! // 8-byte Folded Spill
20 ; CHECK-NEXT: .cfi_def_cfa_offset 16
21 ; CHECK-NEXT: .cfi_offset b8, -16
22 ; CHECK-NEXT: mov w0, wzr
23 ; CHECK-NEXT: //APP
24 ; CHECK-NEXT: //NO_APP
25 ; CHECK-NEXT: ldr d8, [sp], #16 // 8-byte Folded Reload
26 ; CHECK-NEXT: ret
27 entry:
28 tail call void asm sideeffect "", "~{d8}"() #1
29 ret i32 0
30 }
32 ; Very simple - doesn't require hazards
33 define i32 @basic_sc(i32 noundef %num) "aarch64_pstate_sm_compatible" {
34 ; CHECK-LABEL: basic_sc:
35 ; CHECK: // %bb.0: // %entry
36 ; CHECK-NEXT: mov w0, wzr
37 ; CHECK-NEXT: ret
38 entry:
39 ret i32 0
40 }
42 ; No fpr accesses/csrs - doesn't require hazards
43 define i32 @nocsr_alloci64(i64 %d) "aarch64_pstate_sm_compatible" {
44 ; CHECK-LABEL: nocsr_alloci64:
45 ; CHECK: // %bb.0: // %entry
46 ; CHECK-NEXT: sub sp, sp, #16
47 ; CHECK-NEXT: .cfi_def_cfa_offset 16
48 ; CHECK-NEXT: mov x8, x0
49 ; CHECK-NEXT: mov w0, wzr
50 ; CHECK-NEXT: str x8, [sp, #8]
51 ; CHECK-NEXT: add sp, sp, #16
52 ; CHECK-NEXT: ret
53 entry:
54 %a = alloca i64
55 store i64 %d, ptr %a
56 ret i32 0
57 }
59 ; No fpr accesses/csrs - doesn't require hazards
60 define i32 @csr_x20(i32 noundef %num) "aarch64_pstate_sm_compatible" {
61 ; CHECK-LABEL: csr_x20:
62 ; CHECK: // %bb.0: // %entry
63 ; CHECK-NEXT: str x20, [sp, #-16]! // 8-byte Folded Spill
64 ; CHECK-NEXT: .cfi_def_cfa_offset 16
65 ; CHECK-NEXT: .cfi_offset w20, -16
66 ; CHECK-NEXT: mov w0, wzr
67 ; CHECK-NEXT: //APP
68 ; CHECK-NEXT: //NO_APP
69 ; CHECK-NEXT: ldr x20, [sp], #16 // 8-byte Folded Reload
70 ; CHECK-NEXT: ret
71 entry:
72 tail call void asm sideeffect "", "~{x20}"() #1
73 ret i32 0
74 }
76 ; CSR of d8. Make sure there is a gap between FPR and GPR
77 define i32 @csr_d8(i32 noundef %num) "aarch64_pstate_sm_compatible" {
78 ; CHECK0-LABEL: csr_d8:
79 ; CHECK0: // %bb.0: // %entry
80 ; CHECK0-NEXT: str d8, [sp, #-16]! // 8-byte Folded Spill
81 ; CHECK0-NEXT: .cfi_def_cfa_offset 16
82 ; CHECK0-NEXT: .cfi_offset b8, -16
83 ; CHECK0-NEXT: mov w0, wzr
84 ; CHECK0-NEXT: //APP
85 ; CHECK0-NEXT: //NO_APP
86 ; CHECK0-NEXT: ldr d8, [sp], #16 // 8-byte Folded Reload
87 ; CHECK0-NEXT: ret
88 ;
89 ; CHECK64-LABEL: csr_d8:
90 ; CHECK64: // %bb.0: // %entry
91 ; CHECK64-NEXT: sub sp, sp, #144
92 ; CHECK64-NEXT: str d8, [sp, #64] // 8-byte Folded Spill
93 ; CHECK64-NEXT: .cfi_def_cfa_offset 144
94 ; CHECK64-NEXT: .cfi_offset b8, -80
95 ; CHECK64-NEXT: //APP
96 ; CHECK64-NEXT: //NO_APP
97 ; CHECK64-NEXT: mov w0, wzr
98 ; CHECK64-NEXT: ldr d8, [sp, #64] // 8-byte Folded Reload
99 ; CHECK64-NEXT: add sp, sp, #144
100 ; CHECK64-NEXT: ret
101 ;
102 ; CHECK1024-LABEL: csr_d8:
103 ; CHECK1024: // %bb.0: // %entry
104 ; CHECK1024-NEXT: sub sp, sp, #1040
105 ; CHECK1024-NEXT: str d8, [sp] // 8-byte Folded Spill
106 ; CHECK1024-NEXT: str x29, [sp, #1032] // 8-byte Folded Spill
107 ; CHECK1024-NEXT: sub sp, sp, #1024
108 ; CHECK1024-NEXT: .cfi_def_cfa_offset 2064
109 ; CHECK1024-NEXT: .cfi_offset w29, -8
110 ; CHECK1024-NEXT: .cfi_offset b8, -1040
111 ; CHECK1024-NEXT: mov w0, wzr
112 ; CHECK1024-NEXT: //APP
113 ; CHECK1024-NEXT: //NO_APP
114 ; CHECK1024-NEXT: add sp, sp, #1024
115 ; CHECK1024-NEXT: ldr x29, [sp, #1032] // 8-byte Folded Reload
116 ; CHECK1024-NEXT: ldr d8, [sp] // 8-byte Folded Reload
117 ; CHECK1024-NEXT: add sp, sp, #1040
118 ; CHECK1024-NEXT: ret
119 entry:
120 tail call void asm sideeffect "", "~{d8}"() #1
121 ret i32 0
122 }
124 ; Stack fpr objects.
125 define i32 @nocsr_allocd(double %d) "aarch64_pstate_sm_compatible" {
126 ; CHECK0-LABEL: nocsr_allocd:
127 ; CHECK0: // %bb.0: // %entry
128 ; CHECK0-NEXT: sub sp, sp, #16
129 ; CHECK0-NEXT: .cfi_def_cfa_offset 16
130 ; CHECK0-NEXT: mov w0, wzr
131 ; CHECK0-NEXT: str d0, [sp, #8]
132 ; CHECK0-NEXT: add sp, sp, #16
133 ; CHECK0-NEXT: ret
134 ;
135 ; CHECK64-LABEL: nocsr_allocd:
136 ; CHECK64: // %bb.0: // %entry
137 ; CHECK64-NEXT: sub sp, sp, #80
138 ; CHECK64-NEXT: .cfi_def_cfa_offset 80
139 ; CHECK64-NEXT: mov w0, wzr
140 ; CHECK64-NEXT: str d0, [sp, #72]
141 ; CHECK64-NEXT: add sp, sp, #80
142 ; CHECK64-NEXT: ret
143 ;
144 ; CHECK1024-LABEL: nocsr_allocd:
145 ; CHECK1024: // %bb.0: // %entry
146 ; CHECK1024-NEXT: sub sp, sp, #1040
147 ; CHECK1024-NEXT: str x29, [sp, #1024] // 8-byte Folded Spill
148 ; CHECK1024-NEXT: sub sp, sp, #1040
149 ; CHECK1024-NEXT: .cfi_def_cfa_offset 2080
150 ; CHECK1024-NEXT: .cfi_offset w29, -16
151 ; CHECK1024-NEXT: mov w0, wzr
152 ; CHECK1024-NEXT: str d0, [sp, #1032]
153 ; CHECK1024-NEXT: add sp, sp, #1040
154 ; CHECK1024-NEXT: ldr x29, [sp, #1024] // 8-byte Folded Reload
155 ; CHECK1024-NEXT: add sp, sp, #1040
156 ; CHECK1024-NEXT: ret
157 entry:
158 %a = alloca double
159 store double %d, ptr %a
160 ret i32 0
161 }
163 define i32 @csr_d8d9(i32 noundef %num) "aarch64_pstate_sm_compatible" {
164 ; CHECK0-LABEL: csr_d8d9:
165 ; CHECK0: // %bb.0: // %entry
166 ; CHECK0-NEXT: stp d9, d8, [sp, #-16]! // 16-byte Folded Spill
167 ; CHECK0-NEXT: .cfi_def_cfa_offset 16
168 ; CHECK0-NEXT: .cfi_offset b8, -8
169 ; CHECK0-NEXT: .cfi_offset b9, -16
170 ; CHECK0-NEXT: mov w0, wzr
171 ; CHECK0-NEXT: //APP
172 ; CHECK0-NEXT: //NO_APP
173 ; CHECK0-NEXT: ldp d9, d8, [sp], #16 // 16-byte Folded Reload
174 ; CHECK0-NEXT: ret
175 ;
176 ; CHECK64-LABEL: csr_d8d9:
177 ; CHECK64: // %bb.0: // %entry
178 ; CHECK64-NEXT: sub sp, sp, #144
179 ; CHECK64-NEXT: stp d9, d8, [sp, #64] // 16-byte Folded Spill
180 ; CHECK64-NEXT: .cfi_def_cfa_offset 144
181 ; CHECK64-NEXT: .cfi_offset b8, -72
182 ; CHECK64-NEXT: .cfi_offset b9, -80
183 ; CHECK64-NEXT: //APP
184 ; CHECK64-NEXT: //NO_APP
185 ; CHECK64-NEXT: mov w0, wzr
186 ; CHECK64-NEXT: ldp d9, d8, [sp, #64] // 16-byte Folded Reload
187 ; CHECK64-NEXT: add sp, sp, #144
188 ; CHECK64-NEXT: ret
189 ;
190 ; CHECK1024-LABEL: csr_d8d9:
191 ; CHECK1024: // %bb.0: // %entry
192 ; CHECK1024-NEXT: sub sp, sp, #1056
193 ; CHECK1024-NEXT: stp d9, d8, [sp] // 16-byte Folded Spill
194 ; CHECK1024-NEXT: str x29, [sp, #1040] // 8-byte Folded Spill
195 ; CHECK1024-NEXT: sub sp, sp, #1024
196 ; CHECK1024-NEXT: .cfi_def_cfa_offset 2080
197 ; CHECK1024-NEXT: .cfi_offset w29, -16
198 ; CHECK1024-NEXT: .cfi_offset b8, -1048
199 ; CHECK1024-NEXT: .cfi_offset b9, -1056
200 ; CHECK1024-NEXT: mov w0, wzr
201 ; CHECK1024-NEXT: //APP
202 ; CHECK1024-NEXT: //NO_APP
203 ; CHECK1024-NEXT: add sp, sp, #1024
204 ; CHECK1024-NEXT: ldp d9, d8, [sp] // 16-byte Folded Reload
205 ; CHECK1024-NEXT: ldr x29, [sp, #1040] // 8-byte Folded Reload
206 ; CHECK1024-NEXT: add sp, sp, #1056
207 ; CHECK1024-NEXT: ret
208 entry:
209 tail call void asm sideeffect "", "~{d8},~{d9}"() #1
210 ret i32 0
211 }
213 define i32 @csr_d8_allocd(double %d) "aarch64_pstate_sm_compatible" {
214 ; CHECK0-LABEL: csr_d8_allocd:
215 ; CHECK0: // %bb.0: // %entry
216 ; CHECK0-NEXT: stp d8, d0, [sp, #-16]! // 8-byte Folded Spill
217 ; CHECK0-NEXT: .cfi_def_cfa_offset 16
218 ; CHECK0-NEXT: .cfi_offset b8, -16
219 ; CHECK0-NEXT: mov w0, wzr
220 ; CHECK0-NEXT: //APP
221 ; CHECK0-NEXT: //NO_APP
222 ; CHECK0-NEXT: ldr d8, [sp], #16 // 8-byte Folded Reload
223 ; CHECK0-NEXT: ret
224 ;
225 ; CHECK64-LABEL: csr_d8_allocd:
226 ; CHECK64: // %bb.0: // %entry
227 ; CHECK64-NEXT: sub sp, sp, #160
228 ; CHECK64-NEXT: stp d0, d8, [sp, #72] // 8-byte Folded Spill
229 ; CHECK64-NEXT: .cfi_def_cfa_offset 160
230 ; CHECK64-NEXT: .cfi_offset b8, -80
231 ; CHECK64-NEXT: //APP
232 ; CHECK64-NEXT: //NO_APP
233 ; CHECK64-NEXT: mov w0, wzr
234 ; CHECK64-NEXT: ldr d8, [sp, #80] // 8-byte Folded Reload
235 ; CHECK64-NEXT: add sp, sp, #160
236 ; CHECK64-NEXT: ret
237 ;
238 ; CHECK1024-LABEL: csr_d8_allocd:
239 ; CHECK1024: // %bb.0: // %entry
240 ; CHECK1024-NEXT: sub sp, sp, #1040
241 ; CHECK1024-NEXT: str d8, [sp] // 8-byte Folded Spill
242 ; CHECK1024-NEXT: str x29, [sp, #1032] // 8-byte Folded Spill
243 ; CHECK1024-NEXT: sub sp, sp, #1040
244 ; CHECK1024-NEXT: .cfi_def_cfa_offset 2080
245 ; CHECK1024-NEXT: .cfi_offset w29, -8
246 ; CHECK1024-NEXT: .cfi_offset b8, -1040
247 ; CHECK1024-NEXT: mov w0, wzr
248 ; CHECK1024-NEXT: //APP
249 ; CHECK1024-NEXT: //NO_APP
250 ; CHECK1024-NEXT: str d0, [sp, #1032]
251 ; CHECK1024-NEXT: add sp, sp, #1040
252 ; CHECK1024-NEXT: ldr x29, [sp, #1032] // 8-byte Folded Reload
253 ; CHECK1024-NEXT: ldr d8, [sp] // 8-byte Folded Reload
254 ; CHECK1024-NEXT: add sp, sp, #1040
255 ; CHECK1024-NEXT: ret
256 entry:
257 %a = alloca double
258 tail call void asm sideeffect "", "~{d8}"() #1
259 store double %d, ptr %a
260 ret i32 0
261 }
263 define i32 @csr_d8_alloci64(i64 %d) "aarch64_pstate_sm_compatible" {
264 ; CHECK0-LABEL: csr_d8_alloci64:
265 ; CHECK0: // %bb.0: // %entry
266 ; CHECK0-NEXT: str d8, [sp, #-16]! // 8-byte Folded Spill
267 ; CHECK0-NEXT: .cfi_def_cfa_offset 16
268 ; CHECK0-NEXT: .cfi_offset b8, -16
269 ; CHECK0-NEXT: mov x8, x0
270 ; CHECK0-NEXT: mov w0, wzr
271 ; CHECK0-NEXT: //APP
272 ; CHECK0-NEXT: //NO_APP
273 ; CHECK0-NEXT: str x8, [sp, #8]
274 ; CHECK0-NEXT: ldr d8, [sp], #16 // 8-byte Folded Reload
275 ; CHECK0-NEXT: ret
276 ;
277 ; CHECK64-LABEL: csr_d8_alloci64:
278 ; CHECK64: // %bb.0: // %entry
279 ; CHECK64-NEXT: sub sp, sp, #160
280 ; CHECK64-NEXT: str d8, [sp, #80] // 8-byte Folded Spill
281 ; CHECK64-NEXT: .cfi_def_cfa_offset 160
282 ; CHECK64-NEXT: .cfi_offset b8, -80
283 ; CHECK64-NEXT: //APP
284 ; CHECK64-NEXT: //NO_APP
285 ; CHECK64-NEXT: mov x8, x0
286 ; CHECK64-NEXT: mov w0, wzr
287 ; CHECK64-NEXT: ldr d8, [sp, #80] // 8-byte Folded Reload
288 ; CHECK64-NEXT: str x8, [sp, #8]
289 ; CHECK64-NEXT: add sp, sp, #160
290 ; CHECK64-NEXT: ret
291 ;
292 ; CHECK1024-LABEL: csr_d8_alloci64:
293 ; CHECK1024: // %bb.0: // %entry
294 ; CHECK1024-NEXT: sub sp, sp, #1040
295 ; CHECK1024-NEXT: str d8, [sp] // 8-byte Folded Spill
296 ; CHECK1024-NEXT: str x29, [sp, #1032] // 8-byte Folded Spill
297 ; CHECK1024-NEXT: sub sp, sp, #1040
298 ; CHECK1024-NEXT: .cfi_def_cfa_offset 2080
299 ; CHECK1024-NEXT: .cfi_offset w29, -8
300 ; CHECK1024-NEXT: .cfi_offset b8, -1040
301 ; CHECK1024-NEXT: mov x8, x0
302 ; CHECK1024-NEXT: mov w0, wzr
303 ; CHECK1024-NEXT: //APP
304 ; CHECK1024-NEXT: //NO_APP
305 ; CHECK1024-NEXT: str x8, [sp, #8]
306 ; CHECK1024-NEXT: add sp, sp, #1040
307 ; CHECK1024-NEXT: ldr x29, [sp, #1032] // 8-byte Folded Reload
308 ; CHECK1024-NEXT: ldr d8, [sp] // 8-byte Folded Reload
309 ; CHECK1024-NEXT: add sp, sp, #1040
310 ; CHECK1024-NEXT: ret
311 entry:
312 %a = alloca i64
313 tail call void asm sideeffect "", "~{d8}"() #1
314 store i64 %d, ptr %a
315 ret i32 0
316 }
318 ; Check the frame pointer is in the right place
319 define i32 @csr_d8_allocd_framepointer(double %d) "aarch64_pstate_sm_compatible" "frame-pointer"="all" {
320 ; CHECK0-LABEL: csr_d8_allocd_framepointer:
321 ; CHECK0: // %bb.0: // %entry
322 ; CHECK0-NEXT: str d8, [sp, #-32]! // 8-byte Folded Spill
323 ; CHECK0-NEXT: stp x29, x30, [sp, #16] // 16-byte Folded Spill
324 ; CHECK0-NEXT: add x29, sp, #16
325 ; CHECK0-NEXT: .cfi_def_cfa w29, 16
326 ; CHECK0-NEXT: .cfi_offset w30, -8
327 ; CHECK0-NEXT: .cfi_offset w29, -16
328 ; CHECK0-NEXT: .cfi_offset b8, -32
329 ; CHECK0-NEXT: //APP
330 ; CHECK0-NEXT: //NO_APP
331 ; CHECK0-NEXT: ldp x29, x30, [sp, #16] // 16-byte Folded Reload
332 ; CHECK0-NEXT: mov w0, wzr
333 ; CHECK0-NEXT: str d0, [sp, #8]
334 ; CHECK0-NEXT: ldr d8, [sp], #32 // 8-byte Folded Reload
335 ; CHECK0-NEXT: ret
336 ;
337 ; CHECK64-LABEL: csr_d8_allocd_framepointer:
338 ; CHECK64: // %bb.0: // %entry
339 ; CHECK64-NEXT: sub sp, sp, #176
340 ; CHECK64-NEXT: str d8, [sp, #80] // 8-byte Folded Spill
341 ; CHECK64-NEXT: stp x29, x30, [sp, #152] // 16-byte Folded Spill
342 ; CHECK64-NEXT: add x29, sp, #80
343 ; CHECK64-NEXT: .cfi_def_cfa w29, 96
344 ; CHECK64-NEXT: .cfi_offset w30, -16
345 ; CHECK64-NEXT: .cfi_offset w29, -24
346 ; CHECK64-NEXT: .cfi_offset b8, -96
347 ; CHECK64-NEXT: //APP
348 ; CHECK64-NEXT: //NO_APP
349 ; CHECK64-NEXT: stur d0, [x29, #-8]
350 ; CHECK64-NEXT: ldr x29, [sp, #152] // 8-byte Folded Reload
351 ; CHECK64-NEXT: ldr d8, [sp, #80] // 8-byte Folded Reload
352 ; CHECK64-NEXT: mov w0, wzr
353 ; CHECK64-NEXT: add sp, sp, #176
354 ; CHECK64-NEXT: ret
355 ;
356 ; CHECK1024-LABEL: csr_d8_allocd_framepointer:
357 ; CHECK1024: // %bb.0: // %entry
358 ; CHECK1024-NEXT: sub sp, sp, #1056
359 ; CHECK1024-NEXT: str d8, [sp] // 8-byte Folded Spill
360 ; CHECK1024-NEXT: str x29, [sp, #1032] // 8-byte Folded Spill
361 ; CHECK1024-NEXT: mov x29, sp
362 ; CHECK1024-NEXT: str x30, [sp, #1040] // 8-byte Folded Spill
363 ; CHECK1024-NEXT: sub sp, sp, #1040
364 ; CHECK1024-NEXT: .cfi_def_cfa w29, 1056
365 ; CHECK1024-NEXT: .cfi_offset w30, -16
366 ; CHECK1024-NEXT: .cfi_offset w29, -24
367 ; CHECK1024-NEXT: .cfi_offset b8, -1056
368 ; CHECK1024-NEXT: mov w0, wzr
369 ; CHECK1024-NEXT: //APP
370 ; CHECK1024-NEXT: //NO_APP
371 ; CHECK1024-NEXT: stur d0, [x29, #-8]
372 ; CHECK1024-NEXT: add sp, sp, #1040
373 ; CHECK1024-NEXT: ldr x30, [sp, #1040] // 8-byte Folded Reload
374 ; CHECK1024-NEXT: ldr x29, [sp, #1032] // 8-byte Folded Reload
375 ; CHECK1024-NEXT: ldr d8, [sp] // 8-byte Folded Reload
376 ; CHECK1024-NEXT: add sp, sp, #1056
377 ; CHECK1024-NEXT: ret
378 entry:
379 %a = alloca double
380 tail call void asm sideeffect "", "~{d8}"() #1
381 store double %d, ptr %a
382 ret i32 0
383 }
385 ; sve stack objects should live with other fpr registers
386 define i32 @csr_d8_allocnxv4i32(i64 %d) "aarch64_pstate_sm_compatible" {
387 ; CHECK0-LABEL: csr_d8_allocnxv4i32:
388 ; CHECK0: // %bb.0: // %entry
389 ; CHECK0-NEXT: str d8, [sp, #-16]! // 8-byte Folded Spill
390 ; CHECK0-NEXT: str x29, [sp, #8] // 8-byte Folded Spill
391 ; CHECK0-NEXT: addvl sp, sp, #-1
392 ; CHECK0-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 8 * VG
393 ; CHECK0-NEXT: .cfi_offset w29, -8
394 ; CHECK0-NEXT: .cfi_offset b8, -16
395 ; CHECK0-NEXT: mov z0.s, #0 // =0x0
396 ; CHECK0-NEXT: ptrue p0.s
397 ; CHECK0-NEXT: mov w0, wzr
398 ; CHECK0-NEXT: //APP
399 ; CHECK0-NEXT: //NO_APP
400 ; CHECK0-NEXT: st1w { z0.s }, p0, [sp]
401 ; CHECK0-NEXT: addvl sp, sp, #1
402 ; CHECK0-NEXT: ldr x29, [sp, #8] // 8-byte Folded Reload
403 ; CHECK0-NEXT: ldr d8, [sp], #16 // 8-byte Folded Reload
404 ; CHECK0-NEXT: ret
405 ;
406 ; CHECK64-LABEL: csr_d8_allocnxv4i32:
407 ; CHECK64: // %bb.0: // %entry
408 ; CHECK64-NEXT: str d8, [sp, #-80]! // 8-byte Folded Spill
409 ; CHECK64-NEXT: str x29, [sp, #72] // 8-byte Folded Spill
410 ; CHECK64-NEXT: sub sp, sp, #64
411 ; CHECK64-NEXT: addvl sp, sp, #-1
412 ; CHECK64-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0x90, 0x01, 0x22, 0x11, 0x08, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 144 + 8 * VG
413 ; CHECK64-NEXT: .cfi_offset w29, -8
414 ; CHECK64-NEXT: .cfi_offset b8, -80
415 ; CHECK64-NEXT: mov z0.s, #0 // =0x0
416 ; CHECK64-NEXT: ptrue p0.s
417 ; CHECK64-NEXT: add x8, sp, #64
418 ; CHECK64-NEXT: mov w0, wzr
419 ; CHECK64-NEXT: //APP
420 ; CHECK64-NEXT: //NO_APP
421 ; CHECK64-NEXT: st1w { z0.s }, p0, [x8]
422 ; CHECK64-NEXT: addvl sp, sp, #1
423 ; CHECK64-NEXT: add sp, sp, #64
424 ; CHECK64-NEXT: ldr x29, [sp, #72] // 8-byte Folded Reload
425 ; CHECK64-NEXT: ldr d8, [sp], #80 // 8-byte Folded Reload
426 ; CHECK64-NEXT: ret
427 ;
428 ; CHECK1024-LABEL: csr_d8_allocnxv4i32:
429 ; CHECK1024: // %bb.0: // %entry
430 ; CHECK1024-NEXT: sub sp, sp, #1040
431 ; CHECK1024-NEXT: str d8, [sp] // 8-byte Folded Spill
432 ; CHECK1024-NEXT: str x29, [sp, #1032] // 8-byte Folded Spill
433 ; CHECK1024-NEXT: sub sp, sp, #1024
434 ; CHECK1024-NEXT: addvl sp, sp, #-1
435 ; CHECK1024-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0x90, 0x10, 0x22, 0x11, 0x08, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 2064 + 8 * VG
436 ; CHECK1024-NEXT: .cfi_offset w29, -8
437 ; CHECK1024-NEXT: .cfi_offset b8, -1040
438 ; CHECK1024-NEXT: mov z0.s, #0 // =0x0
439 ; CHECK1024-NEXT: ptrue p0.s
440 ; CHECK1024-NEXT: add x8, sp, #1024
441 ; CHECK1024-NEXT: mov w0, wzr
442 ; CHECK1024-NEXT: //APP
443 ; CHECK1024-NEXT: //NO_APP
444 ; CHECK1024-NEXT: st1w { z0.s }, p0, [x8]
445 ; CHECK1024-NEXT: addvl sp, sp, #1
446 ; CHECK1024-NEXT: add sp, sp, #1024
447 ; CHECK1024-NEXT: ldr x29, [sp, #1032] // 8-byte Folded Reload
448 ; CHECK1024-NEXT: ldr d8, [sp] // 8-byte Folded Reload
449 ; CHECK1024-NEXT: add sp, sp, #1040
450 ; CHECK1024-NEXT: ret
451 entry:
452 %a = alloca <vscale x 4 x i32>
453 tail call void asm sideeffect "", "~{d8}"() #1
454 store <vscale x 4 x i32> zeroinitializer, ptr %a
455 ret i32 0
456 }
458 define i32 @csr_x18_25_d8_15_allocdi64(i64 %d, double %e) "aarch64_pstate_sm_compatible" {
459 ; CHECK0-LABEL: csr_x18_25_d8_15_allocdi64:
460 ; CHECK0: // %bb.0: // %entry
461 ; CHECK0-NEXT: sub sp, sp, #144
462 ; CHECK0-NEXT: stp d15, d14, [sp, #16] // 16-byte Folded Spill
463 ; CHECK0-NEXT: stp d13, d12, [sp, #32] // 16-byte Folded Spill
464 ; CHECK0-NEXT: stp d11, d10, [sp, #48] // 16-byte Folded Spill
465 ; CHECK0-NEXT: stp d9, d8, [sp, #64] // 16-byte Folded Spill
466 ; CHECK0-NEXT: str x25, [sp, #80] // 8-byte Folded Spill
467 ; CHECK0-NEXT: stp x24, x23, [sp, #96] // 16-byte Folded Spill
468 ; CHECK0-NEXT: stp x22, x21, [sp, #112] // 16-byte Folded Spill
469 ; CHECK0-NEXT: stp x20, x19, [sp, #128] // 16-byte Folded Spill
470 ; CHECK0-NEXT: .cfi_def_cfa_offset 144
471 ; CHECK0-NEXT: .cfi_offset w19, -8
472 ; CHECK0-NEXT: .cfi_offset w20, -16
473 ; CHECK0-NEXT: .cfi_offset w21, -24
474 ; CHECK0-NEXT: .cfi_offset w22, -32
475 ; CHECK0-NEXT: .cfi_offset w23, -40
476 ; CHECK0-NEXT: .cfi_offset w24, -48
477 ; CHECK0-NEXT: .cfi_offset w25, -64
478 ; CHECK0-NEXT: .cfi_offset b8, -72
479 ; CHECK0-NEXT: .cfi_offset b9, -80
480 ; CHECK0-NEXT: .cfi_offset b10, -88
481 ; CHECK0-NEXT: .cfi_offset b11, -96
482 ; CHECK0-NEXT: .cfi_offset b12, -104
483 ; CHECK0-NEXT: .cfi_offset b13, -112
484 ; CHECK0-NEXT: .cfi_offset b14, -120
485 ; CHECK0-NEXT: .cfi_offset b15, -128
486 ; CHECK0-NEXT: //APP
487 ; CHECK0-NEXT: //NO_APP
488 ; CHECK0-NEXT: //APP
489 ; CHECK0-NEXT: //NO_APP
490 ; CHECK0-NEXT: mov x8, x0
491 ; CHECK0-NEXT: ldp x20, x19, [sp, #128] // 16-byte Folded Reload
492 ; CHECK0-NEXT: ldr x25, [sp, #80] // 8-byte Folded Reload
493 ; CHECK0-NEXT: ldp x22, x21, [sp, #112] // 16-byte Folded Reload
494 ; CHECK0-NEXT: mov w0, wzr
495 ; CHECK0-NEXT: ldp x24, x23, [sp, #96] // 16-byte Folded Reload
496 ; CHECK0-NEXT: str x8, [sp, #88]
497 ; CHECK0-NEXT: ldp d9, d8, [sp, #64] // 16-byte Folded Reload
498 ; CHECK0-NEXT: str d0, [sp, #8]
499 ; CHECK0-NEXT: ldp d11, d10, [sp, #48] // 16-byte Folded Reload
500 ; CHECK0-NEXT: ldp d13, d12, [sp, #32] // 16-byte Folded Reload
501 ; CHECK0-NEXT: ldp d15, d14, [sp, #16] // 16-byte Folded Reload
502 ; CHECK0-NEXT: add sp, sp, #144
503 ; CHECK0-NEXT: ret
504 ;
505 ; CHECK64-LABEL: csr_x18_25_d8_15_allocdi64:
506 ; CHECK64: // %bb.0: // %entry
507 ; CHECK64-NEXT: sub sp, sp, #288
508 ; CHECK64-NEXT: stp d15, d14, [sp, #96] // 16-byte Folded Spill
509 ; CHECK64-NEXT: stp d13, d12, [sp, #112] // 16-byte Folded Spill
510 ; CHECK64-NEXT: stp d11, d10, [sp, #128] // 16-byte Folded Spill
511 ; CHECK64-NEXT: stp d9, d8, [sp, #144] // 16-byte Folded Spill
512 ; CHECK64-NEXT: stp x29, x25, [sp, #224] // 16-byte Folded Spill
513 ; CHECK64-NEXT: stp x24, x23, [sp, #240] // 16-byte Folded Spill
514 ; CHECK64-NEXT: stp x22, x21, [sp, #256] // 16-byte Folded Spill
515 ; CHECK64-NEXT: stp x20, x19, [sp, #272] // 16-byte Folded Spill
516 ; CHECK64-NEXT: .cfi_def_cfa_offset 288
517 ; CHECK64-NEXT: .cfi_offset w19, -8
518 ; CHECK64-NEXT: .cfi_offset w20, -16
519 ; CHECK64-NEXT: .cfi_offset w21, -24
520 ; CHECK64-NEXT: .cfi_offset w22, -32
521 ; CHECK64-NEXT: .cfi_offset w23, -40
522 ; CHECK64-NEXT: .cfi_offset w24, -48
523 ; CHECK64-NEXT: .cfi_offset w25, -56
524 ; CHECK64-NEXT: .cfi_offset w29, -64
525 ; CHECK64-NEXT: .cfi_offset b8, -136
526 ; CHECK64-NEXT: .cfi_offset b9, -144
527 ; CHECK64-NEXT: .cfi_offset b10, -152
528 ; CHECK64-NEXT: .cfi_offset b11, -160
529 ; CHECK64-NEXT: .cfi_offset b12, -168
530 ; CHECK64-NEXT: .cfi_offset b13, -176
531 ; CHECK64-NEXT: .cfi_offset b14, -184
532 ; CHECK64-NEXT: .cfi_offset b15, -192
533 ; CHECK64-NEXT: //APP
534 ; CHECK64-NEXT: //NO_APP
535 ; CHECK64-NEXT: //APP
536 ; CHECK64-NEXT: //NO_APP
537 ; CHECK64-NEXT: mov x8, x0
538 ; CHECK64-NEXT: ldp x20, x19, [sp, #272] // 16-byte Folded Reload
539 ; CHECK64-NEXT: mov w0, wzr
540 ; CHECK64-NEXT: ldp x22, x21, [sp, #256] // 16-byte Folded Reload
541 ; CHECK64-NEXT: str x8, [sp, #8]
542 ; CHECK64-NEXT: ldp x24, x23, [sp, #240] // 16-byte Folded Reload
543 ; CHECK64-NEXT: str d0, [sp, #88]
544 ; CHECK64-NEXT: ldp x29, x25, [sp, #224] // 16-byte Folded Reload
545 ; CHECK64-NEXT: ldp d9, d8, [sp, #144] // 16-byte Folded Reload
546 ; CHECK64-NEXT: ldp d11, d10, [sp, #128] // 16-byte Folded Reload
547 ; CHECK64-NEXT: ldp d13, d12, [sp, #112] // 16-byte Folded Reload
548 ; CHECK64-NEXT: ldp d15, d14, [sp, #96] // 16-byte Folded Reload
549 ; CHECK64-NEXT: add sp, sp, #288
550 ; CHECK64-NEXT: ret
551 ;
552 ; CHECK1024-LABEL: csr_x18_25_d8_15_allocdi64:
553 ; CHECK1024: // %bb.0: // %entry
554 ; CHECK1024-NEXT: sub sp, sp, #1152
555 ; CHECK1024-NEXT: stp d15, d14, [sp] // 16-byte Folded Spill
556 ; CHECK1024-NEXT: stp d13, d12, [sp, #16] // 16-byte Folded Spill
557 ; CHECK1024-NEXT: stp d11, d10, [sp, #32] // 16-byte Folded Spill
558 ; CHECK1024-NEXT: stp d9, d8, [sp, #48] // 16-byte Folded Spill
559 ; CHECK1024-NEXT: str x29, [sp, #1088] // 8-byte Folded Spill
560 ; CHECK1024-NEXT: str x25, [sp, #1096] // 8-byte Folded Spill
561 ; CHECK1024-NEXT: str x24, [sp, #1104] // 8-byte Folded Spill
562 ; CHECK1024-NEXT: str x23, [sp, #1112] // 8-byte Folded Spill
563 ; CHECK1024-NEXT: str x22, [sp, #1120] // 8-byte Folded Spill
564 ; CHECK1024-NEXT: str x21, [sp, #1128] // 8-byte Folded Spill
565 ; CHECK1024-NEXT: str x20, [sp, #1136] // 8-byte Folded Spill
566 ; CHECK1024-NEXT: str x19, [sp, #1144] // 8-byte Folded Spill
567 ; CHECK1024-NEXT: sub sp, sp, #1056
568 ; CHECK1024-NEXT: .cfi_def_cfa_offset 2208
569 ; CHECK1024-NEXT: .cfi_offset w19, -8
570 ; CHECK1024-NEXT: .cfi_offset w20, -16
571 ; CHECK1024-NEXT: .cfi_offset w21, -24
572 ; CHECK1024-NEXT: .cfi_offset w22, -32
573 ; CHECK1024-NEXT: .cfi_offset w23, -40
574 ; CHECK1024-NEXT: .cfi_offset w24, -48
575 ; CHECK1024-NEXT: .cfi_offset w25, -56
576 ; CHECK1024-NEXT: .cfi_offset w29, -64
577 ; CHECK1024-NEXT: .cfi_offset b8, -1096
578 ; CHECK1024-NEXT: .cfi_offset b9, -1104
579 ; CHECK1024-NEXT: .cfi_offset b10, -1112
580 ; CHECK1024-NEXT: .cfi_offset b11, -1120
581 ; CHECK1024-NEXT: .cfi_offset b12, -1128
582 ; CHECK1024-NEXT: .cfi_offset b13, -1136
583 ; CHECK1024-NEXT: .cfi_offset b14, -1144
584 ; CHECK1024-NEXT: .cfi_offset b15, -1152
585 ; CHECK1024-NEXT: mov x8, x0
586 ; CHECK1024-NEXT: mov w0, wzr
587 ; CHECK1024-NEXT: //APP
588 ; CHECK1024-NEXT: //NO_APP
589 ; CHECK1024-NEXT: //APP
590 ; CHECK1024-NEXT: //NO_APP
591 ; CHECK1024-NEXT: str x8, [sp, #8]
592 ; CHECK1024-NEXT: str d0, [sp, #1048]
593 ; CHECK1024-NEXT: add sp, sp, #1056
594 ; CHECK1024-NEXT: ldp d9, d8, [sp, #48] // 16-byte Folded Reload
595 ; CHECK1024-NEXT: ldr x19, [sp, #1144] // 8-byte Folded Reload
596 ; CHECK1024-NEXT: ldp d11, d10, [sp, #32] // 16-byte Folded Reload
597 ; CHECK1024-NEXT: ldr x20, [sp, #1136] // 8-byte Folded Reload
598 ; CHECK1024-NEXT: ldr x21, [sp, #1128] // 8-byte Folded Reload
599 ; CHECK1024-NEXT: ldr x22, [sp, #1120] // 8-byte Folded Reload
600 ; CHECK1024-NEXT: ldr x23, [sp, #1112] // 8-byte Folded Reload
601 ; CHECK1024-NEXT: ldr x24, [sp, #1104] // 8-byte Folded Reload
602 ; CHECK1024-NEXT: ldr x25, [sp, #1096] // 8-byte Folded Reload
603 ; CHECK1024-NEXT: ldr x29, [sp, #1088] // 8-byte Folded Reload
604 ; CHECK1024-NEXT: ldp d13, d12, [sp, #16] // 16-byte Folded Reload
605 ; CHECK1024-NEXT: ldp d15, d14, [sp] // 16-byte Folded Reload
606 ; CHECK1024-NEXT: add sp, sp, #1152
607 ; CHECK1024-NEXT: ret
608 entry:
609 %a = alloca i64
610 %b = alloca double
611 tail call void asm sideeffect "", "~{x18},~{x19},~{x20},~{x21},~{x22},~{x23},~{x24},~{x25}"()
612 tail call void asm sideeffect "", "~{d8},~{d9},~{d10},~{d11},~{d12},~{d13},~{d14},~{d15}"()
613 store i64 %d, ptr %a
614 store double %e, ptr %b
615 ret i32 0
616 }
618 define i32 @csr_x18_25_d8_15_allocdi64_locallystreaming(i64 %d, double %e) "aarch64_pstate_sm_body" "target-features"="+sme" {
619 ; CHECK0-LABEL: csr_x18_25_d8_15_allocdi64_locallystreaming:
620 ; CHECK0: // %bb.0: // %entry
621 ; CHECK0-NEXT: sub sp, sp, #176
622 ; CHECK0-NEXT: .cfi_def_cfa_offset 176
623 ; CHECK0-NEXT: rdsvl x9, #1
624 ; CHECK0-NEXT: stp d15, d14, [sp, #48] // 16-byte Folded Spill
625 ; CHECK0-NEXT: lsr x9, x9, #3
626 ; CHECK0-NEXT: stp d13, d12, [sp, #64] // 16-byte Folded Spill
627 ; CHECK0-NEXT: stp d11, d10, [sp, #80] // 16-byte Folded Spill
628 ; CHECK0-NEXT: str x9, [sp, #32] // 8-byte Folded Spill
629 ; CHECK0-NEXT: cntd x9
630 ; CHECK0-NEXT: str x9, [sp, #40] // 8-byte Folded Spill
631 ; CHECK0-NEXT: stp d9, d8, [sp, #96] // 16-byte Folded Spill
632 ; CHECK0-NEXT: str x25, [sp, #112] // 8-byte Folded Spill
633 ; CHECK0-NEXT: stp x24, x23, [sp, #128] // 16-byte Folded Spill
634 ; CHECK0-NEXT: stp x22, x21, [sp, #144] // 16-byte Folded Spill
635 ; CHECK0-NEXT: stp x20, x19, [sp, #160] // 16-byte Folded Spill
636 ; CHECK0-NEXT: .cfi_offset w19, -8
637 ; CHECK0-NEXT: .cfi_offset w20, -16
638 ; CHECK0-NEXT: .cfi_offset w21, -24
639 ; CHECK0-NEXT: .cfi_offset w22, -32
640 ; CHECK0-NEXT: .cfi_offset w23, -40
641 ; CHECK0-NEXT: .cfi_offset w24, -48
642 ; CHECK0-NEXT: .cfi_offset w25, -64
643 ; CHECK0-NEXT: .cfi_offset b8, -72
644 ; CHECK0-NEXT: .cfi_offset b9, -80
645 ; CHECK0-NEXT: .cfi_offset b10, -88
646 ; CHECK0-NEXT: .cfi_offset b11, -96
647 ; CHECK0-NEXT: .cfi_offset b12, -104
648 ; CHECK0-NEXT: .cfi_offset b13, -112
649 ; CHECK0-NEXT: .cfi_offset b14, -120
650 ; CHECK0-NEXT: .cfi_offset b15, -128
651 ; CHECK0-NEXT: .cfi_offset vg, -136
652 ; CHECK0-NEXT: str d0, [sp, #8] // 8-byte Folded Spill
653 ; CHECK0-NEXT: smstart sm
654 ; CHECK0-NEXT: //APP
655 ; CHECK0-NEXT: //NO_APP
656 ; CHECK0-NEXT: //APP
657 ; CHECK0-NEXT: //NO_APP
658 ; CHECK0-NEXT: ldr d0, [sp, #8] // 8-byte Folded Reload
659 ; CHECK0-NEXT: str x0, [sp, #24]
660 ; CHECK0-NEXT: str d0, [sp, #16]
661 ; CHECK0-NEXT: smstop sm
662 ; CHECK0-NEXT: ldp x20, x19, [sp, #160] // 16-byte Folded Reload
663 ; CHECK0-NEXT: mov w0, wzr
664 ; CHECK0-NEXT: ldp x22, x21, [sp, #144] // 16-byte Folded Reload
665 ; CHECK0-NEXT: ldr x25, [sp, #112] // 8-byte Folded Reload
666 ; CHECK0-NEXT: ldp x24, x23, [sp, #128] // 16-byte Folded Reload
667 ; CHECK0-NEXT: ldp d9, d8, [sp, #96] // 16-byte Folded Reload
668 ; CHECK0-NEXT: ldp d11, d10, [sp, #80] // 16-byte Folded Reload
669 ; CHECK0-NEXT: ldp d13, d12, [sp, #64] // 16-byte Folded Reload
670 ; CHECK0-NEXT: ldp d15, d14, [sp, #48] // 16-byte Folded Reload
671 ; CHECK0-NEXT: add sp, sp, #176
672 ; CHECK0-NEXT: .cfi_def_cfa_offset 0
673 ; CHECK0-NEXT: .cfi_restore w19
674 ; CHECK0-NEXT: .cfi_restore w20
675 ; CHECK0-NEXT: .cfi_restore w21
676 ; CHECK0-NEXT: .cfi_restore w22
677 ; CHECK0-NEXT: .cfi_restore w23
678 ; CHECK0-NEXT: .cfi_restore w24
679 ; CHECK0-NEXT: .cfi_restore w25
680 ; CHECK0-NEXT: .cfi_restore b8
681 ; CHECK0-NEXT: .cfi_restore b9
682 ; CHECK0-NEXT: .cfi_restore b10
683 ; CHECK0-NEXT: .cfi_restore b11
684 ; CHECK0-NEXT: .cfi_restore b12
685 ; CHECK0-NEXT: .cfi_restore b13
686 ; CHECK0-NEXT: .cfi_restore b14
687 ; CHECK0-NEXT: .cfi_restore b15
688 ; CHECK0-NEXT: ret
689 ;
690 ; CHECK64-LABEL: csr_x18_25_d8_15_allocdi64_locallystreaming:
691 ; CHECK64: // %bb.0: // %entry
692 ; CHECK64-NEXT: sub sp, sp, #304
693 ; CHECK64-NEXT: .cfi_def_cfa_offset 304
694 ; CHECK64-NEXT: rdsvl x9, #1
695 ; CHECK64-NEXT: stp d15, d14, [sp, #112] // 16-byte Folded Spill
696 ; CHECK64-NEXT: lsr x9, x9, #3
697 ; CHECK64-NEXT: stp d13, d12, [sp, #128] // 16-byte Folded Spill
698 ; CHECK64-NEXT: stp d11, d10, [sp, #144] // 16-byte Folded Spill
699 ; CHECK64-NEXT: str x9, [sp, #96] // 8-byte Folded Spill
700 ; CHECK64-NEXT: cntd x9
701 ; CHECK64-NEXT: str x9, [sp, #104] // 8-byte Folded Spill
702 ; CHECK64-NEXT: stp d9, d8, [sp, #160] // 16-byte Folded Spill
703 ; CHECK64-NEXT: stp x29, x25, [sp, #240] // 16-byte Folded Spill
704 ; CHECK64-NEXT: stp x24, x23, [sp, #256] // 16-byte Folded Spill
705 ; CHECK64-NEXT: stp x22, x21, [sp, #272] // 16-byte Folded Spill
706 ; CHECK64-NEXT: stp x20, x19, [sp, #288] // 16-byte Folded Spill
707 ; CHECK64-NEXT: .cfi_offset w19, -8
708 ; CHECK64-NEXT: .cfi_offset w20, -16
709 ; CHECK64-NEXT: .cfi_offset w21, -24
710 ; CHECK64-NEXT: .cfi_offset w22, -32
711 ; CHECK64-NEXT: .cfi_offset w23, -40
712 ; CHECK64-NEXT: .cfi_offset w24, -48
713 ; CHECK64-NEXT: .cfi_offset w25, -56
714 ; CHECK64-NEXT: .cfi_offset w29, -64
715 ; CHECK64-NEXT: .cfi_offset b8, -136
716 ; CHECK64-NEXT: .cfi_offset b9, -144
717 ; CHECK64-NEXT: .cfi_offset b10, -152
718 ; CHECK64-NEXT: .cfi_offset b11, -160
719 ; CHECK64-NEXT: .cfi_offset b12, -168
720 ; CHECK64-NEXT: .cfi_offset b13, -176
721 ; CHECK64-NEXT: .cfi_offset b14, -184
722 ; CHECK64-NEXT: .cfi_offset b15, -192
723 ; CHECK64-NEXT: .cfi_offset vg, -200
724 ; CHECK64-NEXT: str d0, [sp, #80] // 8-byte Folded Spill
725 ; CHECK64-NEXT: smstart sm
726 ; CHECK64-NEXT: //APP
727 ; CHECK64-NEXT: //NO_APP
728 ; CHECK64-NEXT: //APP
729 ; CHECK64-NEXT: //NO_APP
730 ; CHECK64-NEXT: ldr d0, [sp, #80] // 8-byte Folded Reload
731 ; CHECK64-NEXT: str x0, [sp, #8]
732 ; CHECK64-NEXT: str d0, [sp, #88]
733 ; CHECK64-NEXT: smstop sm
734 ; CHECK64-NEXT: ldp x20, x19, [sp, #288] // 16-byte Folded Reload
735 ; CHECK64-NEXT: mov w0, wzr
736 ; CHECK64-NEXT: ldp x22, x21, [sp, #272] // 16-byte Folded Reload
737 ; CHECK64-NEXT: ldp x24, x23, [sp, #256] // 16-byte Folded Reload
738 ; CHECK64-NEXT: ldp x29, x25, [sp, #240] // 16-byte Folded Reload
739 ; CHECK64-NEXT: ldp d9, d8, [sp, #160] // 16-byte Folded Reload
740 ; CHECK64-NEXT: ldp d11, d10, [sp, #144] // 16-byte Folded Reload
741 ; CHECK64-NEXT: ldp d13, d12, [sp, #128] // 16-byte Folded Reload
742 ; CHECK64-NEXT: ldp d15, d14, [sp, #112] // 16-byte Folded Reload
743 ; CHECK64-NEXT: add sp, sp, #304
744 ; CHECK64-NEXT: .cfi_def_cfa_offset 0
745 ; CHECK64-NEXT: .cfi_restore w19
746 ; CHECK64-NEXT: .cfi_restore w20
747 ; CHECK64-NEXT: .cfi_restore w21
748 ; CHECK64-NEXT: .cfi_restore w22
749 ; CHECK64-NEXT: .cfi_restore w23
750 ; CHECK64-NEXT: .cfi_restore w24
751 ; CHECK64-NEXT: .cfi_restore w25
752 ; CHECK64-NEXT: .cfi_restore w29
753 ; CHECK64-NEXT: .cfi_restore b8
754 ; CHECK64-NEXT: .cfi_restore b9
755 ; CHECK64-NEXT: .cfi_restore b10
756 ; CHECK64-NEXT: .cfi_restore b11
757 ; CHECK64-NEXT: .cfi_restore b12
758 ; CHECK64-NEXT: .cfi_restore b13
759 ; CHECK64-NEXT: .cfi_restore b14
760 ; CHECK64-NEXT: .cfi_restore b15
761 ; CHECK64-NEXT: ret
762 ;
763 ; CHECK1024-LABEL: csr_x18_25_d8_15_allocdi64_locallystreaming:
764 ; CHECK1024: // %bb.0: // %entry
765 ; CHECK1024-NEXT: rdsvl x9, #1
766 ; CHECK1024-NEXT: lsr x9, x9, #3
767 ; CHECK1024-NEXT: sub sp, sp, #1168
768 ; CHECK1024-NEXT: .cfi_def_cfa_offset 1168
769 ; CHECK1024-NEXT: str x9, [sp] // 8-byte Folded Spill
770 ; CHECK1024-NEXT: cntd x9
771 ; CHECK1024-NEXT: str x9, [sp, #8] // 8-byte Folded Spill
772 ; CHECK1024-NEXT: stp d15, d14, [sp, #16] // 16-byte Folded Spill
773 ; CHECK1024-NEXT: stp d13, d12, [sp, #32] // 16-byte Folded Spill
774 ; CHECK1024-NEXT: stp d11, d10, [sp, #48] // 16-byte Folded Spill
775 ; CHECK1024-NEXT: stp d9, d8, [sp, #64] // 16-byte Folded Spill
776 ; CHECK1024-NEXT: str x29, [sp, #1104] // 8-byte Folded Spill
777 ; CHECK1024-NEXT: str x25, [sp, #1112] // 8-byte Folded Spill
778 ; CHECK1024-NEXT: str x24, [sp, #1120] // 8-byte Folded Spill
779 ; CHECK1024-NEXT: str x23, [sp, #1128] // 8-byte Folded Spill
780 ; CHECK1024-NEXT: str x22, [sp, #1136] // 8-byte Folded Spill
781 ; CHECK1024-NEXT: str x21, [sp, #1144] // 8-byte Folded Spill
782 ; CHECK1024-NEXT: str x20, [sp, #1152] // 8-byte Folded Spill
783 ; CHECK1024-NEXT: str x19, [sp, #1160] // 8-byte Folded Spill
784 ; CHECK1024-NEXT: .cfi_offset w19, -8
785 ; CHECK1024-NEXT: .cfi_offset w20, -16
786 ; CHECK1024-NEXT: .cfi_offset w21, -24
787 ; CHECK1024-NEXT: .cfi_offset w22, -32
788 ; CHECK1024-NEXT: .cfi_offset w23, -40
789 ; CHECK1024-NEXT: .cfi_offset w24, -48
790 ; CHECK1024-NEXT: .cfi_offset w25, -56
791 ; CHECK1024-NEXT: .cfi_offset w29, -64
792 ; CHECK1024-NEXT: .cfi_offset b8, -1096
793 ; CHECK1024-NEXT: .cfi_offset b9, -1104
794 ; CHECK1024-NEXT: .cfi_offset b10, -1112
795 ; CHECK1024-NEXT: .cfi_offset b11, -1120
796 ; CHECK1024-NEXT: .cfi_offset b12, -1128
797 ; CHECK1024-NEXT: .cfi_offset b13, -1136
798 ; CHECK1024-NEXT: .cfi_offset b14, -1144
799 ; CHECK1024-NEXT: .cfi_offset b15, -1152
800 ; CHECK1024-NEXT: .cfi_offset vg, -1160
801 ; CHECK1024-NEXT: sub sp, sp, #1056
802 ; CHECK1024-NEXT: .cfi_def_cfa_offset 2224
803 ; CHECK1024-NEXT: str d0, [sp, #1040] // 8-byte Folded Spill
804 ; CHECK1024-NEXT: smstart sm
805 ; CHECK1024-NEXT: //APP
806 ; CHECK1024-NEXT: //NO_APP
807 ; CHECK1024-NEXT: //APP
808 ; CHECK1024-NEXT: //NO_APP
809 ; CHECK1024-NEXT: ldr d0, [sp, #1040] // 8-byte Folded Reload
810 ; CHECK1024-NEXT: str x0, [sp, #8]
811 ; CHECK1024-NEXT: str d0, [sp, #1048]
812 ; CHECK1024-NEXT: smstop sm
813 ; CHECK1024-NEXT: mov w0, wzr
814 ; CHECK1024-NEXT: add sp, sp, #1056
815 ; CHECK1024-NEXT: .cfi_def_cfa_offset 1168
816 ; CHECK1024-NEXT: ldp d9, d8, [sp, #64] // 16-byte Folded Reload
817 ; CHECK1024-NEXT: ldr x19, [sp, #1160] // 8-byte Folded Reload
818 ; CHECK1024-NEXT: ldp d11, d10, [sp, #48] // 16-byte Folded Reload
819 ; CHECK1024-NEXT: ldr x20, [sp, #1152] // 8-byte Folded Reload
820 ; CHECK1024-NEXT: ldr x21, [sp, #1144] // 8-byte Folded Reload
821 ; CHECK1024-NEXT: ldr x22, [sp, #1136] // 8-byte Folded Reload
822 ; CHECK1024-NEXT: ldr x23, [sp, #1128] // 8-byte Folded Reload
823 ; CHECK1024-NEXT: ldr x24, [sp, #1120] // 8-byte Folded Reload
824 ; CHECK1024-NEXT: ldr x25, [sp, #1112] // 8-byte Folded Reload
825 ; CHECK1024-NEXT: ldr x29, [sp, #1104] // 8-byte Folded Reload
826 ; CHECK1024-NEXT: ldp d13, d12, [sp, #32] // 16-byte Folded Reload
827 ; CHECK1024-NEXT: ldp d15, d14, [sp, #16] // 16-byte Folded Reload
828 ; CHECK1024-NEXT: add sp, sp, #1168
829 ; CHECK1024-NEXT: .cfi_def_cfa_offset 0
830 ; CHECK1024-NEXT: .cfi_restore w19
831 ; CHECK1024-NEXT: .cfi_restore w20
832 ; CHECK1024-NEXT: .cfi_restore w21
833 ; CHECK1024-NEXT: .cfi_restore w22
834 ; CHECK1024-NEXT: .cfi_restore w23
835 ; CHECK1024-NEXT: .cfi_restore w24
836 ; CHECK1024-NEXT: .cfi_restore w25
837 ; CHECK1024-NEXT: .cfi_restore w29
838 ; CHECK1024-NEXT: .cfi_restore b8
839 ; CHECK1024-NEXT: .cfi_restore b9
840 ; CHECK1024-NEXT: .cfi_restore b10
841 ; CHECK1024-NEXT: .cfi_restore b11
842 ; CHECK1024-NEXT: .cfi_restore b12
843 ; CHECK1024-NEXT: .cfi_restore b13
844 ; CHECK1024-NEXT: .cfi_restore b14
845 ; CHECK1024-NEXT: .cfi_restore b15
846 ; CHECK1024-NEXT: ret
847 entry:
848 %a = alloca i64
849 %b = alloca double
850 tail call void asm sideeffect "", "~{x18},~{x19},~{x20},~{x21},~{x22},~{x23},~{x24},~{x25}"()
851 tail call void asm sideeffect "", "~{d8},~{d9},~{d10},~{d11},~{d12},~{d13},~{d14},~{d15}"()
852 store i64 %d, ptr %a
853 store double %e, ptr %b
854 ret i32 0
855 }
857 ; We don't currently handle fpr stack arguments very well (they are hopefully relatively rare).
858 define float @nocsr_stackargs(float %a, float %b, float %c, float %d, float %e, float %f, float %g, float %h, float %i) "aarch64_pstate_sm_compatible" {
859 ; CHECK-LABEL: nocsr_stackargs:
860 ; CHECK: // %bb.0: // %entry
861 ; CHECK-NEXT: ldr s0, [sp]
862 ; CHECK-NEXT: ret
863 entry:
864 ret float %i
865 }
867 define float @csr_x20_stackargs(float %a, float %b, float %c, float %d, float %e, float %f, float %g, float %h, float %i) "aarch64_pstate_sm_compatible" {
868 ; CHECK-LABEL: csr_x20_stackargs:
869 ; CHECK: // %bb.0: // %entry
870 ; CHECK-NEXT: str x20, [sp, #-16]! // 8-byte Folded Spill
871 ; CHECK-NEXT: .cfi_def_cfa_offset 16
872 ; CHECK-NEXT: .cfi_offset w20, -16
873 ; CHECK-NEXT: ldr s0, [sp, #16]
874 ; CHECK-NEXT: //APP
875 ; CHECK-NEXT: //NO_APP
876 ; CHECK-NEXT: ldr x20, [sp], #16 // 8-byte Folded Reload
877 ; CHECK-NEXT: ret
878 entry:
879 tail call void asm sideeffect "", "~{x20}"() #1
880 ret float %i
881 }
883 define float @csr_d8_stackargs(float %a, float %b, float %c, float %d, float %e, float %f, float %g, float %h, float %i) "aarch64_pstate_sm_compatible" {
884 ; CHECK0-LABEL: csr_d8_stackargs:
885 ; CHECK0: // %bb.0: // %entry
886 ; CHECK0-NEXT: str d8, [sp, #-16]! // 8-byte Folded Spill
887 ; CHECK0-NEXT: .cfi_def_cfa_offset 16
888 ; CHECK0-NEXT: .cfi_offset b8, -16
889 ; CHECK0-NEXT: ldr s0, [sp, #16]
890 ; CHECK0-NEXT: //APP
891 ; CHECK0-NEXT: //NO_APP
892 ; CHECK0-NEXT: ldr d8, [sp], #16 // 8-byte Folded Reload
893 ; CHECK0-NEXT: ret
894 ;
895 ; CHECK64-LABEL: csr_d8_stackargs:
896 ; CHECK64: // %bb.0: // %entry
897 ; CHECK64-NEXT: sub sp, sp, #144
898 ; CHECK64-NEXT: str d8, [sp, #64] // 8-byte Folded Spill
899 ; CHECK64-NEXT: .cfi_def_cfa_offset 144
900 ; CHECK64-NEXT: .cfi_offset b8, -80
901 ; CHECK64-NEXT: //APP
902 ; CHECK64-NEXT: //NO_APP
903 ; CHECK64-NEXT: ldr s0, [sp, #144]
904 ; CHECK64-NEXT: ldr d8, [sp, #64] // 8-byte Folded Reload
905 ; CHECK64-NEXT: add sp, sp, #144
906 ; CHECK64-NEXT: ret
907 ;
908 ; CHECK1024-LABEL: csr_d8_stackargs:
909 ; CHECK1024: // %bb.0: // %entry
910 ; CHECK1024-NEXT: sub sp, sp, #1040
911 ; CHECK1024-NEXT: str d8, [sp] // 8-byte Folded Spill
912 ; CHECK1024-NEXT: str x29, [sp, #1032] // 8-byte Folded Spill
913 ; CHECK1024-NEXT: sub sp, sp, #1024
914 ; CHECK1024-NEXT: .cfi_def_cfa_offset 2064
915 ; CHECK1024-NEXT: .cfi_offset w29, -8
916 ; CHECK1024-NEXT: .cfi_offset b8, -1040
917 ; CHECK1024-NEXT: ldr s0, [sp, #2064]
918 ; CHECK1024-NEXT: //APP
919 ; CHECK1024-NEXT: //NO_APP
920 ; CHECK1024-NEXT: add sp, sp, #1024
921 ; CHECK1024-NEXT: ldr x29, [sp, #1032] // 8-byte Folded Reload
922 ; CHECK1024-NEXT: ldr d8, [sp] // 8-byte Folded Reload
923 ; CHECK1024-NEXT: add sp, sp, #1040
924 ; CHECK1024-NEXT: ret
925 entry:
926 tail call void asm sideeffect "", "~{d8}"() #1
927 ret float %i
928 }
930 ; SVE calling conventions
931 define i32 @svecc_basic(i32 noundef %num, <vscale x 4 x i32> %vs) "aarch64_pstate_sm_compatible" {
932 ; CHECK-LABEL: svecc_basic:
933 ; CHECK: // %bb.0: // %entry
934 ; CHECK-NEXT: mov w0, wzr
935 ; CHECK-NEXT: ret
936 entry:
937 ret i32 0
938 }
940 define i32 @svecc_csr_x20(i32 noundef %num, <vscale x 4 x i32> %vs) "aarch64_pstate_sm_compatible" {
941 ; CHECK-LABEL: svecc_csr_x20:
942 ; CHECK: // %bb.0: // %entry
943 ; CHECK-NEXT: str x20, [sp, #-16]! // 8-byte Folded Spill
944 ; CHECK-NEXT: .cfi_def_cfa_offset 16
945 ; CHECK-NEXT: .cfi_offset w20, -16
946 ; CHECK-NEXT: mov w0, wzr
947 ; CHECK-NEXT: //APP
948 ; CHECK-NEXT: //NO_APP
949 ; CHECK-NEXT: ldr x20, [sp], #16 // 8-byte Folded Reload
950 ; CHECK-NEXT: ret
951 entry:
952 tail call void asm sideeffect "", "~{x20}"() #1
953 ret i32 0
954 }
956 define i32 @svecc_csr_d8(i32 noundef %num, <vscale x 4 x i32> %vs) "aarch64_pstate_sm_compatible" {
957 ; CHECK0-LABEL: svecc_csr_d8:
958 ; CHECK0: // %bb.0: // %entry
959 ; CHECK0-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
960 ; CHECK0-NEXT: addvl sp, sp, #-1
961 ; CHECK0-NEXT: str z8, [sp] // 16-byte Folded Spill
962 ; CHECK0-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 8 * VG
963 ; CHECK0-NEXT: .cfi_offset w29, -16
964 ; CHECK0-NEXT: .cfi_escape 0x10, 0x48, 0x0a, 0x11, 0x70, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 16 - 8 * VG
965 ; CHECK0-NEXT: //APP
966 ; CHECK0-NEXT: //NO_APP
967 ; CHECK0-NEXT: mov w0, wzr
968 ; CHECK0-NEXT: ldr z8, [sp] // 16-byte Folded Reload
969 ; CHECK0-NEXT: addvl sp, sp, #1
970 ; CHECK0-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
971 ; CHECK0-NEXT: ret
972 ;
973 ; CHECK64-LABEL: svecc_csr_d8:
974 ; CHECK64: // %bb.0: // %entry
975 ; CHECK64-NEXT: sub sp, sp, #80
976 ; CHECK64-NEXT: str x29, [sp, #64] // 8-byte Folded Spill
977 ; CHECK64-NEXT: addvl sp, sp, #-1
978 ; CHECK64-NEXT: str z8, [sp] // 16-byte Folded Spill
979 ; CHECK64-NEXT: sub sp, sp, #64
980 ; CHECK64-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0x90, 0x01, 0x22, 0x11, 0x08, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 144 + 8 * VG
981 ; CHECK64-NEXT: .cfi_offset w29, -16
982 ; CHECK64-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0xb0, 0x7f, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 80 - 8 * VG
983 ; CHECK64-NEXT: mov w0, wzr
984 ; CHECK64-NEXT: //APP
985 ; CHECK64-NEXT: //NO_APP
986 ; CHECK64-NEXT: add sp, sp, #64
987 ; CHECK64-NEXT: ldr z8, [sp] // 16-byte Folded Reload
988 ; CHECK64-NEXT: addvl sp, sp, #1
989 ; CHECK64-NEXT: ldr x29, [sp, #64] // 8-byte Folded Reload
990 ; CHECK64-NEXT: add sp, sp, #80
991 ; CHECK64-NEXT: ret
992 ;
993 ; CHECK1024-LABEL: svecc_csr_d8:
994 ; CHECK1024: // %bb.0: // %entry
995 ; CHECK1024-NEXT: sub sp, sp, #1040
996 ; CHECK1024-NEXT: str x29, [sp, #1024] // 8-byte Folded Spill
997 ; CHECK1024-NEXT: addvl sp, sp, #-1
998 ; CHECK1024-NEXT: str z8, [sp] // 16-byte Folded Spill
999 ; CHECK1024-NEXT: sub sp, sp, #1024
1000 ; CHECK1024-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0x90, 0x10, 0x22, 0x11, 0x08, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 2064 + 8 * VG
1001 ; CHECK1024-NEXT: .cfi_offset w29, -16
1002 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0xf0, 0x77, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 1040 - 8 * VG
1003 ; CHECK1024-NEXT: mov w0, wzr
1004 ; CHECK1024-NEXT: //APP
1005 ; CHECK1024-NEXT: //NO_APP
1006 ; CHECK1024-NEXT: add sp, sp, #1024
1007 ; CHECK1024-NEXT: ldr z8, [sp] // 16-byte Folded Reload
1008 ; CHECK1024-NEXT: addvl sp, sp, #1
1009 ; CHECK1024-NEXT: ldr x29, [sp, #1024] // 8-byte Folded Reload
1010 ; CHECK1024-NEXT: add sp, sp, #1040
1011 ; CHECK1024-NEXT: ret
1012 entry:
1013 tail call void asm sideeffect "", "~{d8}"() #1
1014 ret i32 0
1015 }
1017 define i32 @svecc_csr_d8d9(i32 noundef %num, <vscale x 4 x i32> %vs) "aarch64_pstate_sm_compatible" {
1018 ; CHECK0-LABEL: svecc_csr_d8d9:
1019 ; CHECK0: // %bb.0: // %entry
1020 ; CHECK0-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
1021 ; CHECK0-NEXT: addvl sp, sp, #-2
1022 ; CHECK0-NEXT: str z9, [sp] // 16-byte Folded Spill
1023 ; CHECK0-NEXT: str z8, [sp, #1, mul vl] // 16-byte Folded Spill
1024 ; CHECK0-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x10, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 16 * VG
1025 ; CHECK0-NEXT: .cfi_offset w29, -16
1026 ; CHECK0-NEXT: .cfi_escape 0x10, 0x48, 0x0a, 0x11, 0x70, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 16 - 8 * VG
1027 ; CHECK0-NEXT: .cfi_escape 0x10, 0x49, 0x0a, 0x11, 0x70, 0x22, 0x11, 0x70, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d9 @ cfa - 16 - 16 * VG
1028 ; CHECK0-NEXT: //APP
1029 ; CHECK0-NEXT: //NO_APP
1030 ; CHECK0-NEXT: mov w0, wzr
1031 ; CHECK0-NEXT: ldr z9, [sp] // 16-byte Folded Reload
1032 ; CHECK0-NEXT: ldr z8, [sp, #1, mul vl] // 16-byte Folded Reload
1033 ; CHECK0-NEXT: addvl sp, sp, #2
1034 ; CHECK0-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
1035 ; CHECK0-NEXT: ret
1036 ;
1037 ; CHECK64-LABEL: svecc_csr_d8d9:
1038 ; CHECK64: // %bb.0: // %entry
1039 ; CHECK64-NEXT: sub sp, sp, #80
1040 ; CHECK64-NEXT: str x29, [sp, #64] // 8-byte Folded Spill
1041 ; CHECK64-NEXT: addvl sp, sp, #-2
1042 ; CHECK64-NEXT: str z9, [sp] // 16-byte Folded Spill
1043 ; CHECK64-NEXT: str z8, [sp, #1, mul vl] // 16-byte Folded Spill
1044 ; CHECK64-NEXT: sub sp, sp, #64
1045 ; CHECK64-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0x90, 0x01, 0x22, 0x11, 0x10, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 144 + 16 * VG
1046 ; CHECK64-NEXT: .cfi_offset w29, -16
1047 ; CHECK64-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0xb0, 0x7f, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 80 - 8 * VG
1048 ; CHECK64-NEXT: .cfi_escape 0x10, 0x49, 0x0b, 0x11, 0xb0, 0x7f, 0x22, 0x11, 0x70, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d9 @ cfa - 80 - 16 * VG
1049 ; CHECK64-NEXT: mov w0, wzr
1050 ; CHECK64-NEXT: //APP
1051 ; CHECK64-NEXT: //NO_APP
1052 ; CHECK64-NEXT: add sp, sp, #64
1053 ; CHECK64-NEXT: ldr z9, [sp] // 16-byte Folded Reload
1054 ; CHECK64-NEXT: ldr z8, [sp, #1, mul vl] // 16-byte Folded Reload
1055 ; CHECK64-NEXT: addvl sp, sp, #2
1056 ; CHECK64-NEXT: ldr x29, [sp, #64] // 8-byte Folded Reload
1057 ; CHECK64-NEXT: add sp, sp, #80
1058 ; CHECK64-NEXT: ret
1059 ;
1060 ; CHECK1024-LABEL: svecc_csr_d8d9:
1061 ; CHECK1024: // %bb.0: // %entry
1062 ; CHECK1024-NEXT: sub sp, sp, #1040
1063 ; CHECK1024-NEXT: str x29, [sp, #1024] // 8-byte Folded Spill
1064 ; CHECK1024-NEXT: addvl sp, sp, #-2
1065 ; CHECK1024-NEXT: str z9, [sp] // 16-byte Folded Spill
1066 ; CHECK1024-NEXT: str z8, [sp, #1, mul vl] // 16-byte Folded Spill
1067 ; CHECK1024-NEXT: sub sp, sp, #1024
1068 ; CHECK1024-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0x90, 0x10, 0x22, 0x11, 0x10, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 2064 + 16 * VG
1069 ; CHECK1024-NEXT: .cfi_offset w29, -16
1070 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0xf0, 0x77, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 1040 - 8 * VG
1071 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x49, 0x0b, 0x11, 0xf0, 0x77, 0x22, 0x11, 0x70, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d9 @ cfa - 1040 - 16 * VG
1072 ; CHECK1024-NEXT: mov w0, wzr
1073 ; CHECK1024-NEXT: //APP
1074 ; CHECK1024-NEXT: //NO_APP
1075 ; CHECK1024-NEXT: add sp, sp, #1024
1076 ; CHECK1024-NEXT: ldr z9, [sp] // 16-byte Folded Reload
1077 ; CHECK1024-NEXT: ldr z8, [sp, #1, mul vl] // 16-byte Folded Reload
1078 ; CHECK1024-NEXT: addvl sp, sp, #2
1079 ; CHECK1024-NEXT: ldr x29, [sp, #1024] // 8-byte Folded Reload
1080 ; CHECK1024-NEXT: add sp, sp, #1040
1081 ; CHECK1024-NEXT: ret
1082 entry:
1083 tail call void asm sideeffect "", "~{d8},~{d9}"() #1
1084 ret i32 0
1085 }
1087 define i32 @svecc_csr_d8_allocd(double %d, <vscale x 4 x i32> %vs) "aarch64_pstate_sm_compatible" {
1088 ; CHECK0-LABEL: svecc_csr_d8_allocd:
1089 ; CHECK0: // %bb.0: // %entry
1090 ; CHECK0-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
1091 ; CHECK0-NEXT: addvl sp, sp, #-1
1092 ; CHECK0-NEXT: str z8, [sp] // 16-byte Folded Spill
1093 ; CHECK0-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 8 * VG
1094 ; CHECK0-NEXT: .cfi_offset w29, -16
1095 ; CHECK0-NEXT: .cfi_escape 0x10, 0x48, 0x0a, 0x11, 0x70, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 16 - 8 * VG
1096 ; CHECK0-NEXT: //APP
1097 ; CHECK0-NEXT: //NO_APP
1098 ; CHECK0-NEXT: addvl x8, sp, #1
1099 ; CHECK0-NEXT: mov w0, wzr
1100 ; CHECK0-NEXT: ldr z8, [sp] // 16-byte Folded Reload
1101 ; CHECK0-NEXT: str d0, [x8, #8]
1102 ; CHECK0-NEXT: addvl sp, sp, #1
1103 ; CHECK0-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
1104 ; CHECK0-NEXT: ret
1105 ;
1106 ; CHECK64-LABEL: svecc_csr_d8_allocd:
1107 ; CHECK64: // %bb.0: // %entry
1108 ; CHECK64-NEXT: sub sp, sp, #80
1109 ; CHECK64-NEXT: str x29, [sp, #64] // 8-byte Folded Spill
1110 ; CHECK64-NEXT: addvl sp, sp, #-1
1111 ; CHECK64-NEXT: str z8, [sp] // 16-byte Folded Spill
1112 ; CHECK64-NEXT: sub sp, sp, #80
1113 ; CHECK64-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0xa0, 0x01, 0x22, 0x11, 0x08, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 160 + 8 * VG
1114 ; CHECK64-NEXT: .cfi_offset w29, -16
1115 ; CHECK64-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0xb0, 0x7f, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 80 - 8 * VG
1116 ; CHECK64-NEXT: mov w0, wzr
1117 ; CHECK64-NEXT: //APP
1118 ; CHECK64-NEXT: //NO_APP
1119 ; CHECK64-NEXT: str d0, [sp, #72]
1120 ; CHECK64-NEXT: add sp, sp, #80
1121 ; CHECK64-NEXT: ldr z8, [sp] // 16-byte Folded Reload
1122 ; CHECK64-NEXT: addvl sp, sp, #1
1123 ; CHECK64-NEXT: ldr x29, [sp, #64] // 8-byte Folded Reload
1124 ; CHECK64-NEXT: add sp, sp, #80
1125 ; CHECK64-NEXT: ret
1126 ;
1127 ; CHECK1024-LABEL: svecc_csr_d8_allocd:
1128 ; CHECK1024: // %bb.0: // %entry
1129 ; CHECK1024-NEXT: sub sp, sp, #1040
1130 ; CHECK1024-NEXT: str x29, [sp, #1024] // 8-byte Folded Spill
1131 ; CHECK1024-NEXT: addvl sp, sp, #-1
1132 ; CHECK1024-NEXT: str z8, [sp] // 16-byte Folded Spill
1133 ; CHECK1024-NEXT: sub sp, sp, #1040
1134 ; CHECK1024-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0xa0, 0x10, 0x22, 0x11, 0x08, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 2080 + 8 * VG
1135 ; CHECK1024-NEXT: .cfi_offset w29, -16
1136 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0xf0, 0x77, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 1040 - 8 * VG
1137 ; CHECK1024-NEXT: mov w0, wzr
1138 ; CHECK1024-NEXT: //APP
1139 ; CHECK1024-NEXT: //NO_APP
1140 ; CHECK1024-NEXT: str d0, [sp, #1032]
1141 ; CHECK1024-NEXT: add sp, sp, #1040
1142 ; CHECK1024-NEXT: ldr z8, [sp] // 16-byte Folded Reload
1143 ; CHECK1024-NEXT: addvl sp, sp, #1
1144 ; CHECK1024-NEXT: ldr x29, [sp, #1024] // 8-byte Folded Reload
1145 ; CHECK1024-NEXT: add sp, sp, #1040
1146 ; CHECK1024-NEXT: ret
1147 entry:
1148 %a = alloca double
1149 tail call void asm sideeffect "", "~{d8}"() #1
1150 store double %d, ptr %a
1151 ret i32 0
1152 }
1154 define i32 @svecc_csr_d8_alloci64(i64 %d, <vscale x 4 x i32> %vs) "aarch64_pstate_sm_compatible" {
1155 ; CHECK0-LABEL: svecc_csr_d8_alloci64:
1156 ; CHECK0: // %bb.0: // %entry
1157 ; CHECK0-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
1158 ; CHECK0-NEXT: addvl sp, sp, #-1
1159 ; CHECK0-NEXT: str z8, [sp] // 16-byte Folded Spill
1160 ; CHECK0-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 8 * VG
1161 ; CHECK0-NEXT: .cfi_offset w29, -16
1162 ; CHECK0-NEXT: .cfi_escape 0x10, 0x48, 0x0a, 0x11, 0x70, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 16 - 8 * VG
1163 ; CHECK0-NEXT: //APP
1164 ; CHECK0-NEXT: //NO_APP
1165 ; CHECK0-NEXT: mov x8, x0
1166 ; CHECK0-NEXT: addvl x9, sp, #1
1167 ; CHECK0-NEXT: ldr z8, [sp] // 16-byte Folded Reload
1168 ; CHECK0-NEXT: mov w0, wzr
1169 ; CHECK0-NEXT: str x8, [x9, #8]
1170 ; CHECK0-NEXT: addvl sp, sp, #1
1171 ; CHECK0-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
1172 ; CHECK0-NEXT: ret
1173 ;
1174 ; CHECK64-LABEL: svecc_csr_d8_alloci64:
1175 ; CHECK64: // %bb.0: // %entry
1176 ; CHECK64-NEXT: sub sp, sp, #80
1177 ; CHECK64-NEXT: str x29, [sp, #64] // 8-byte Folded Spill
1178 ; CHECK64-NEXT: addvl sp, sp, #-1
1179 ; CHECK64-NEXT: str z8, [sp] // 16-byte Folded Spill
1180 ; CHECK64-NEXT: sub sp, sp, #80
1181 ; CHECK64-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0xa0, 0x01, 0x22, 0x11, 0x08, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 160 + 8 * VG
1182 ; CHECK64-NEXT: .cfi_offset w29, -16
1183 ; CHECK64-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0xb0, 0x7f, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 80 - 8 * VG
1184 ; CHECK64-NEXT: mov x8, x0
1185 ; CHECK64-NEXT: mov w0, wzr
1186 ; CHECK64-NEXT: //APP
1187 ; CHECK64-NEXT: //NO_APP
1188 ; CHECK64-NEXT: str x8, [sp, #8]
1189 ; CHECK64-NEXT: add sp, sp, #80
1190 ; CHECK64-NEXT: ldr z8, [sp] // 16-byte Folded Reload
1191 ; CHECK64-NEXT: addvl sp, sp, #1
1192 ; CHECK64-NEXT: ldr x29, [sp, #64] // 8-byte Folded Reload
1193 ; CHECK64-NEXT: add sp, sp, #80
1194 ; CHECK64-NEXT: ret
1195 ;
1196 ; CHECK1024-LABEL: svecc_csr_d8_alloci64:
1197 ; CHECK1024: // %bb.0: // %entry
1198 ; CHECK1024-NEXT: sub sp, sp, #1040
1199 ; CHECK1024-NEXT: str x29, [sp, #1024] // 8-byte Folded Spill
1200 ; CHECK1024-NEXT: addvl sp, sp, #-1
1201 ; CHECK1024-NEXT: str z8, [sp] // 16-byte Folded Spill
1202 ; CHECK1024-NEXT: sub sp, sp, #1040
1203 ; CHECK1024-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0xa0, 0x10, 0x22, 0x11, 0x08, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 2080 + 8 * VG
1204 ; CHECK1024-NEXT: .cfi_offset w29, -16
1205 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0xf0, 0x77, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 1040 - 8 * VG
1206 ; CHECK1024-NEXT: mov x8, x0
1207 ; CHECK1024-NEXT: mov w0, wzr
1208 ; CHECK1024-NEXT: //APP
1209 ; CHECK1024-NEXT: //NO_APP
1210 ; CHECK1024-NEXT: str x8, [sp, #8]
1211 ; CHECK1024-NEXT: add sp, sp, #1040
1212 ; CHECK1024-NEXT: ldr z8, [sp] // 16-byte Folded Reload
1213 ; CHECK1024-NEXT: addvl sp, sp, #1
1214 ; CHECK1024-NEXT: ldr x29, [sp, #1024] // 8-byte Folded Reload
1215 ; CHECK1024-NEXT: add sp, sp, #1040
1216 ; CHECK1024-NEXT: ret
1217 entry:
1218 %a = alloca i64
1219 tail call void asm sideeffect "", "~{d8}"() #1
1220 store i64 %d, ptr %a
1221 ret i32 0
1222 }
1224 define i32 @svecc_csr_d8_allocnxv4i32(i64 %d, <vscale x 4 x i32> %vs) "aarch64_pstate_sm_compatible" {
1225 ; CHECK0-LABEL: svecc_csr_d8_allocnxv4i32:
1226 ; CHECK0: // %bb.0: // %entry
1227 ; CHECK0-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
1228 ; CHECK0-NEXT: addvl sp, sp, #-1
1229 ; CHECK0-NEXT: str z8, [sp] // 16-byte Folded Spill
1230 ; CHECK0-NEXT: addvl sp, sp, #-1
1231 ; CHECK0-NEXT: .cfi_escape 0x0f, 0x0c, 0x8f, 0x00, 0x11, 0x10, 0x22, 0x11, 0x10, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 16 + 16 * VG
1232 ; CHECK0-NEXT: .cfi_offset w29, -16
1233 ; CHECK0-NEXT: .cfi_escape 0x10, 0x48, 0x0a, 0x11, 0x70, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 16 - 8 * VG
1234 ; CHECK0-NEXT: mov z0.s, #0 // =0x0
1235 ; CHECK0-NEXT: ptrue p0.s
1236 ; CHECK0-NEXT: mov w0, wzr
1237 ; CHECK0-NEXT: //APP
1238 ; CHECK0-NEXT: //NO_APP
1239 ; CHECK0-NEXT: st1w { z0.s }, p0, [sp]
1240 ; CHECK0-NEXT: addvl sp, sp, #1
1241 ; CHECK0-NEXT: ldr z8, [sp] // 16-byte Folded Reload
1242 ; CHECK0-NEXT: addvl sp, sp, #1
1243 ; CHECK0-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
1244 ; CHECK0-NEXT: ret
1245 ;
1246 ; CHECK64-LABEL: svecc_csr_d8_allocnxv4i32:
1247 ; CHECK64: // %bb.0: // %entry
1248 ; CHECK64-NEXT: sub sp, sp, #80
1249 ; CHECK64-NEXT: str x29, [sp, #64] // 8-byte Folded Spill
1250 ; CHECK64-NEXT: addvl sp, sp, #-1
1251 ; CHECK64-NEXT: str z8, [sp] // 16-byte Folded Spill
1252 ; CHECK64-NEXT: sub sp, sp, #64
1253 ; CHECK64-NEXT: addvl sp, sp, #-1
1254 ; CHECK64-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0x90, 0x01, 0x22, 0x11, 0x10, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 144 + 16 * VG
1255 ; CHECK64-NEXT: .cfi_offset w29, -16
1256 ; CHECK64-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0xb0, 0x7f, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 80 - 8 * VG
1257 ; CHECK64-NEXT: mov z0.s, #0 // =0x0
1258 ; CHECK64-NEXT: ptrue p0.s
1259 ; CHECK64-NEXT: add x8, sp, #64
1260 ; CHECK64-NEXT: mov w0, wzr
1261 ; CHECK64-NEXT: //APP
1262 ; CHECK64-NEXT: //NO_APP
1263 ; CHECK64-NEXT: st1w { z0.s }, p0, [x8]
1264 ; CHECK64-NEXT: add sp, sp, #64
1265 ; CHECK64-NEXT: addvl sp, sp, #1
1266 ; CHECK64-NEXT: ldr z8, [sp] // 16-byte Folded Reload
1267 ; CHECK64-NEXT: addvl sp, sp, #1
1268 ; CHECK64-NEXT: ldr x29, [sp, #64] // 8-byte Folded Reload
1269 ; CHECK64-NEXT: add sp, sp, #80
1270 ; CHECK64-NEXT: ret
1271 ;
1272 ; CHECK1024-LABEL: svecc_csr_d8_allocnxv4i32:
1273 ; CHECK1024: // %bb.0: // %entry
1274 ; CHECK1024-NEXT: sub sp, sp, #1040
1275 ; CHECK1024-NEXT: str x29, [sp, #1024] // 8-byte Folded Spill
1276 ; CHECK1024-NEXT: addvl sp, sp, #-1
1277 ; CHECK1024-NEXT: str z8, [sp] // 16-byte Folded Spill
1278 ; CHECK1024-NEXT: sub sp, sp, #1024
1279 ; CHECK1024-NEXT: addvl sp, sp, #-1
1280 ; CHECK1024-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0x90, 0x10, 0x22, 0x11, 0x10, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 2064 + 16 * VG
1281 ; CHECK1024-NEXT: .cfi_offset w29, -16
1282 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0xf0, 0x77, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 1040 - 8 * VG
1283 ; CHECK1024-NEXT: mov z0.s, #0 // =0x0
1284 ; CHECK1024-NEXT: ptrue p0.s
1285 ; CHECK1024-NEXT: add x8, sp, #1024
1286 ; CHECK1024-NEXT: mov w0, wzr
1287 ; CHECK1024-NEXT: //APP
1288 ; CHECK1024-NEXT: //NO_APP
1289 ; CHECK1024-NEXT: st1w { z0.s }, p0, [x8]
1290 ; CHECK1024-NEXT: add sp, sp, #1024
1291 ; CHECK1024-NEXT: addvl sp, sp, #1
1292 ; CHECK1024-NEXT: ldr z8, [sp] // 16-byte Folded Reload
1293 ; CHECK1024-NEXT: addvl sp, sp, #1
1294 ; CHECK1024-NEXT: ldr x29, [sp, #1024] // 8-byte Folded Reload
1295 ; CHECK1024-NEXT: add sp, sp, #1040
1296 ; CHECK1024-NEXT: ret
1297 entry:
1298 %a = alloca <vscale x 4 x i32>
1299 tail call void asm sideeffect "", "~{d8}"() #1
1300 store <vscale x 4 x i32> zeroinitializer, ptr %a
1301 ret i32 0
1302 }
1304 define i32 @svecc_csr_x18_25_d8_15_allocdi64(i64 %d, double %e, <vscale x 4 x i32> %vs) "aarch64_pstate_sm_compatible" {
1305 ; CHECK0-LABEL: svecc_csr_x18_25_d8_15_allocdi64:
1306 ; CHECK0: // %bb.0: // %entry
1307 ; CHECK0-NEXT: stp x29, x25, [sp, #-64]! // 16-byte Folded Spill
1308 ; CHECK0-NEXT: stp x24, x23, [sp, #16] // 16-byte Folded Spill
1309 ; CHECK0-NEXT: stp x22, x21, [sp, #32] // 16-byte Folded Spill
1310 ; CHECK0-NEXT: stp x20, x19, [sp, #48] // 16-byte Folded Spill
1311 ; CHECK0-NEXT: addvl sp, sp, #-8
1312 ; CHECK0-NEXT: str z15, [sp] // 16-byte Folded Spill
1313 ; CHECK0-NEXT: str z14, [sp, #1, mul vl] // 16-byte Folded Spill
1314 ; CHECK0-NEXT: str z13, [sp, #2, mul vl] // 16-byte Folded Spill
1315 ; CHECK0-NEXT: str z12, [sp, #3, mul vl] // 16-byte Folded Spill
1316 ; CHECK0-NEXT: str z11, [sp, #4, mul vl] // 16-byte Folded Spill
1317 ; CHECK0-NEXT: str z10, [sp, #5, mul vl] // 16-byte Folded Spill
1318 ; CHECK0-NEXT: str z9, [sp, #6, mul vl] // 16-byte Folded Spill
1319 ; CHECK0-NEXT: str z8, [sp, #7, mul vl] // 16-byte Folded Spill
1320 ; CHECK0-NEXT: sub sp, sp, #16
1321 ; CHECK0-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xd0, 0x00, 0x22, 0x11, 0xc0, 0x00, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 80 + 64 * VG
1322 ; CHECK0-NEXT: .cfi_offset w19, -8
1323 ; CHECK0-NEXT: .cfi_offset w20, -16
1324 ; CHECK0-NEXT: .cfi_offset w21, -24
1325 ; CHECK0-NEXT: .cfi_offset w22, -32
1326 ; CHECK0-NEXT: .cfi_offset w23, -40
1327 ; CHECK0-NEXT: .cfi_offset w24, -48
1328 ; CHECK0-NEXT: .cfi_offset w25, -56
1329 ; CHECK0-NEXT: .cfi_offset w29, -64
1330 ; CHECK0-NEXT: .cfi_escape 0x10, 0x48, 0x0a, 0x11, 0x40, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 64 - 8 * VG
1331 ; CHECK0-NEXT: .cfi_escape 0x10, 0x49, 0x0a, 0x11, 0x40, 0x22, 0x11, 0x70, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d9 @ cfa - 64 - 16 * VG
1332 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4a, 0x0a, 0x11, 0x40, 0x22, 0x11, 0x68, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d10 @ cfa - 64 - 24 * VG
1333 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4b, 0x0a, 0x11, 0x40, 0x22, 0x11, 0x60, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d11 @ cfa - 64 - 32 * VG
1334 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4c, 0x0a, 0x11, 0x40, 0x22, 0x11, 0x58, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d12 @ cfa - 64 - 40 * VG
1335 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4d, 0x0a, 0x11, 0x40, 0x22, 0x11, 0x50, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d13 @ cfa - 64 - 48 * VG
1336 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4e, 0x0a, 0x11, 0x40, 0x22, 0x11, 0x48, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d14 @ cfa - 64 - 56 * VG
1337 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4f, 0x0a, 0x11, 0x40, 0x22, 0x11, 0x40, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d15 @ cfa - 64 - 64 * VG
1338 ; CHECK0-NEXT: mov x8, x0
1339 ; CHECK0-NEXT: mov w0, wzr
1340 ; CHECK0-NEXT: //APP
1341 ; CHECK0-NEXT: //NO_APP
1342 ; CHECK0-NEXT: //APP
1343 ; CHECK0-NEXT: //NO_APP
1344 ; CHECK0-NEXT: str x8, [sp, #8]
1345 ; CHECK0-NEXT: str d0, [sp], #16
1346 ; CHECK0-NEXT: ldr z15, [sp] // 16-byte Folded Reload
1347 ; CHECK0-NEXT: ldr z14, [sp, #1, mul vl] // 16-byte Folded Reload
1348 ; CHECK0-NEXT: ldr z13, [sp, #2, mul vl] // 16-byte Folded Reload
1349 ; CHECK0-NEXT: ldr z12, [sp, #3, mul vl] // 16-byte Folded Reload
1350 ; CHECK0-NEXT: ldr z11, [sp, #4, mul vl] // 16-byte Folded Reload
1351 ; CHECK0-NEXT: ldr z10, [sp, #5, mul vl] // 16-byte Folded Reload
1352 ; CHECK0-NEXT: ldr z9, [sp, #6, mul vl] // 16-byte Folded Reload
1353 ; CHECK0-NEXT: ldr z8, [sp, #7, mul vl] // 16-byte Folded Reload
1354 ; CHECK0-NEXT: addvl sp, sp, #8
1355 ; CHECK0-NEXT: ldp x20, x19, [sp, #48] // 16-byte Folded Reload
1356 ; CHECK0-NEXT: ldp x22, x21, [sp, #32] // 16-byte Folded Reload
1357 ; CHECK0-NEXT: ldp x24, x23, [sp, #16] // 16-byte Folded Reload
1358 ; CHECK0-NEXT: ldp x29, x25, [sp], #64 // 16-byte Folded Reload
1359 ; CHECK0-NEXT: ret
1360 ;
1361 ; CHECK64-LABEL: svecc_csr_x18_25_d8_15_allocdi64:
1362 ; CHECK64: // %bb.0: // %entry
1363 ; CHECK64-NEXT: sub sp, sp, #128
1364 ; CHECK64-NEXT: stp x29, x25, [sp, #64] // 16-byte Folded Spill
1365 ; CHECK64-NEXT: stp x24, x23, [sp, #80] // 16-byte Folded Spill
1366 ; CHECK64-NEXT: stp x22, x21, [sp, #96] // 16-byte Folded Spill
1367 ; CHECK64-NEXT: stp x20, x19, [sp, #112] // 16-byte Folded Spill
1368 ; CHECK64-NEXT: addvl sp, sp, #-8
1369 ; CHECK64-NEXT: str z15, [sp] // 16-byte Folded Spill
1370 ; CHECK64-NEXT: str z14, [sp, #1, mul vl] // 16-byte Folded Spill
1371 ; CHECK64-NEXT: str z13, [sp, #2, mul vl] // 16-byte Folded Spill
1372 ; CHECK64-NEXT: str z12, [sp, #3, mul vl] // 16-byte Folded Spill
1373 ; CHECK64-NEXT: str z11, [sp, #4, mul vl] // 16-byte Folded Spill
1374 ; CHECK64-NEXT: str z10, [sp, #5, mul vl] // 16-byte Folded Spill
1375 ; CHECK64-NEXT: str z9, [sp, #6, mul vl] // 16-byte Folded Spill
1376 ; CHECK64-NEXT: str z8, [sp, #7, mul vl] // 16-byte Folded Spill
1377 ; CHECK64-NEXT: sub sp, sp, #96
1378 ; CHECK64-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xe0, 0x01, 0x22, 0x11, 0xc0, 0x00, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 224 + 64 * VG
1379 ; CHECK64-NEXT: .cfi_offset w19, -8
1380 ; CHECK64-NEXT: .cfi_offset w20, -16
1381 ; CHECK64-NEXT: .cfi_offset w21, -24
1382 ; CHECK64-NEXT: .cfi_offset w22, -32
1383 ; CHECK64-NEXT: .cfi_offset w23, -40
1384 ; CHECK64-NEXT: .cfi_offset w24, -48
1385 ; CHECK64-NEXT: .cfi_offset w25, -56
1386 ; CHECK64-NEXT: .cfi_offset w29, -64
1387 ; CHECK64-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0x80, 0x7f, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 128 - 8 * VG
1388 ; CHECK64-NEXT: .cfi_escape 0x10, 0x49, 0x0b, 0x11, 0x80, 0x7f, 0x22, 0x11, 0x70, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d9 @ cfa - 128 - 16 * VG
1389 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4a, 0x0b, 0x11, 0x80, 0x7f, 0x22, 0x11, 0x68, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d10 @ cfa - 128 - 24 * VG
1390 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4b, 0x0b, 0x11, 0x80, 0x7f, 0x22, 0x11, 0x60, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d11 @ cfa - 128 - 32 * VG
1391 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4c, 0x0b, 0x11, 0x80, 0x7f, 0x22, 0x11, 0x58, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d12 @ cfa - 128 - 40 * VG
1392 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4d, 0x0b, 0x11, 0x80, 0x7f, 0x22, 0x11, 0x50, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d13 @ cfa - 128 - 48 * VG
1393 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4e, 0x0b, 0x11, 0x80, 0x7f, 0x22, 0x11, 0x48, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d14 @ cfa - 128 - 56 * VG
1394 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4f, 0x0b, 0x11, 0x80, 0x7f, 0x22, 0x11, 0x40, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d15 @ cfa - 128 - 64 * VG
1395 ; CHECK64-NEXT: mov x8, x0
1396 ; CHECK64-NEXT: mov w0, wzr
1397 ; CHECK64-NEXT: //APP
1398 ; CHECK64-NEXT: //NO_APP
1399 ; CHECK64-NEXT: //APP
1400 ; CHECK64-NEXT: //NO_APP
1401 ; CHECK64-NEXT: str x8, [sp, #8]
1402 ; CHECK64-NEXT: str d0, [sp, #88]
1403 ; CHECK64-NEXT: add sp, sp, #96
1404 ; CHECK64-NEXT: ldr z15, [sp] // 16-byte Folded Reload
1405 ; CHECK64-NEXT: ldr z14, [sp, #1, mul vl] // 16-byte Folded Reload
1406 ; CHECK64-NEXT: ldr z13, [sp, #2, mul vl] // 16-byte Folded Reload
1407 ; CHECK64-NEXT: ldr z12, [sp, #3, mul vl] // 16-byte Folded Reload
1408 ; CHECK64-NEXT: ldr z11, [sp, #4, mul vl] // 16-byte Folded Reload
1409 ; CHECK64-NEXT: ldr z10, [sp, #5, mul vl] // 16-byte Folded Reload
1410 ; CHECK64-NEXT: ldr z9, [sp, #6, mul vl] // 16-byte Folded Reload
1411 ; CHECK64-NEXT: ldr z8, [sp, #7, mul vl] // 16-byte Folded Reload
1412 ; CHECK64-NEXT: addvl sp, sp, #8
1413 ; CHECK64-NEXT: ldp x20, x19, [sp, #112] // 16-byte Folded Reload
1414 ; CHECK64-NEXT: ldp x22, x21, [sp, #96] // 16-byte Folded Reload
1415 ; CHECK64-NEXT: ldp x24, x23, [sp, #80] // 16-byte Folded Reload
1416 ; CHECK64-NEXT: ldp x29, x25, [sp, #64] // 16-byte Folded Reload
1417 ; CHECK64-NEXT: add sp, sp, #128
1418 ; CHECK64-NEXT: ret
1419 ;
1420 ; CHECK1024-LABEL: svecc_csr_x18_25_d8_15_allocdi64:
1421 ; CHECK1024: // %bb.0: // %entry
1422 ; CHECK1024-NEXT: sub sp, sp, #1088
1423 ; CHECK1024-NEXT: str x29, [sp, #1024] // 8-byte Folded Spill
1424 ; CHECK1024-NEXT: str x25, [sp, #1032] // 8-byte Folded Spill
1425 ; CHECK1024-NEXT: str x24, [sp, #1040] // 8-byte Folded Spill
1426 ; CHECK1024-NEXT: str x23, [sp, #1048] // 8-byte Folded Spill
1427 ; CHECK1024-NEXT: str x22, [sp, #1056] // 8-byte Folded Spill
1428 ; CHECK1024-NEXT: str x21, [sp, #1064] // 8-byte Folded Spill
1429 ; CHECK1024-NEXT: str x20, [sp, #1072] // 8-byte Folded Spill
1430 ; CHECK1024-NEXT: str x19, [sp, #1080] // 8-byte Folded Spill
1431 ; CHECK1024-NEXT: addvl sp, sp, #-8
1432 ; CHECK1024-NEXT: str z15, [sp] // 16-byte Folded Spill
1433 ; CHECK1024-NEXT: str z14, [sp, #1, mul vl] // 16-byte Folded Spill
1434 ; CHECK1024-NEXT: str z13, [sp, #2, mul vl] // 16-byte Folded Spill
1435 ; CHECK1024-NEXT: str z12, [sp, #3, mul vl] // 16-byte Folded Spill
1436 ; CHECK1024-NEXT: str z11, [sp, #4, mul vl] // 16-byte Folded Spill
1437 ; CHECK1024-NEXT: str z10, [sp, #5, mul vl] // 16-byte Folded Spill
1438 ; CHECK1024-NEXT: str z9, [sp, #6, mul vl] // 16-byte Folded Spill
1439 ; CHECK1024-NEXT: str z8, [sp, #7, mul vl] // 16-byte Folded Spill
1440 ; CHECK1024-NEXT: sub sp, sp, #1056
1441 ; CHECK1024-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xe0, 0x10, 0x22, 0x11, 0xc0, 0x00, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 2144 + 64 * VG
1442 ; CHECK1024-NEXT: .cfi_offset w19, -8
1443 ; CHECK1024-NEXT: .cfi_offset w20, -16
1444 ; CHECK1024-NEXT: .cfi_offset w21, -24
1445 ; CHECK1024-NEXT: .cfi_offset w22, -32
1446 ; CHECK1024-NEXT: .cfi_offset w23, -40
1447 ; CHECK1024-NEXT: .cfi_offset w24, -48
1448 ; CHECK1024-NEXT: .cfi_offset w25, -56
1449 ; CHECK1024-NEXT: .cfi_offset w29, -64
1450 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0xc0, 0x77, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 1088 - 8 * VG
1451 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x49, 0x0b, 0x11, 0xc0, 0x77, 0x22, 0x11, 0x70, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d9 @ cfa - 1088 - 16 * VG
1452 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4a, 0x0b, 0x11, 0xc0, 0x77, 0x22, 0x11, 0x68, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d10 @ cfa - 1088 - 24 * VG
1453 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4b, 0x0b, 0x11, 0xc0, 0x77, 0x22, 0x11, 0x60, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d11 @ cfa - 1088 - 32 * VG
1454 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4c, 0x0b, 0x11, 0xc0, 0x77, 0x22, 0x11, 0x58, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d12 @ cfa - 1088 - 40 * VG
1455 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4d, 0x0b, 0x11, 0xc0, 0x77, 0x22, 0x11, 0x50, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d13 @ cfa - 1088 - 48 * VG
1456 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4e, 0x0b, 0x11, 0xc0, 0x77, 0x22, 0x11, 0x48, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d14 @ cfa - 1088 - 56 * VG
1457 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4f, 0x0b, 0x11, 0xc0, 0x77, 0x22, 0x11, 0x40, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d15 @ cfa - 1088 - 64 * VG
1458 ; CHECK1024-NEXT: mov x8, x0
1459 ; CHECK1024-NEXT: mov w0, wzr
1460 ; CHECK1024-NEXT: //APP
1461 ; CHECK1024-NEXT: //NO_APP
1462 ; CHECK1024-NEXT: //APP
1463 ; CHECK1024-NEXT: //NO_APP
1464 ; CHECK1024-NEXT: str x8, [sp, #8]
1465 ; CHECK1024-NEXT: str d0, [sp, #1048]
1466 ; CHECK1024-NEXT: add sp, sp, #1056
1467 ; CHECK1024-NEXT: ldr z15, [sp] // 16-byte Folded Reload
1468 ; CHECK1024-NEXT: ldr z14, [sp, #1, mul vl] // 16-byte Folded Reload
1469 ; CHECK1024-NEXT: ldr z13, [sp, #2, mul vl] // 16-byte Folded Reload
1470 ; CHECK1024-NEXT: ldr z12, [sp, #3, mul vl] // 16-byte Folded Reload
1471 ; CHECK1024-NEXT: ldr z11, [sp, #4, mul vl] // 16-byte Folded Reload
1472 ; CHECK1024-NEXT: ldr z10, [sp, #5, mul vl] // 16-byte Folded Reload
1473 ; CHECK1024-NEXT: ldr z9, [sp, #6, mul vl] // 16-byte Folded Reload
1474 ; CHECK1024-NEXT: ldr z8, [sp, #7, mul vl] // 16-byte Folded Reload
1475 ; CHECK1024-NEXT: addvl sp, sp, #8
1476 ; CHECK1024-NEXT: ldr x19, [sp, #1080] // 8-byte Folded Reload
1477 ; CHECK1024-NEXT: ldr x20, [sp, #1072] // 8-byte Folded Reload
1478 ; CHECK1024-NEXT: ldr x21, [sp, #1064] // 8-byte Folded Reload
1479 ; CHECK1024-NEXT: ldr x22, [sp, #1056] // 8-byte Folded Reload
1480 ; CHECK1024-NEXT: ldr x23, [sp, #1048] // 8-byte Folded Reload
1481 ; CHECK1024-NEXT: ldr x24, [sp, #1040] // 8-byte Folded Reload
1482 ; CHECK1024-NEXT: ldr x25, [sp, #1032] // 8-byte Folded Reload
1483 ; CHECK1024-NEXT: ldr x29, [sp, #1024] // 8-byte Folded Reload
1484 ; CHECK1024-NEXT: add sp, sp, #1088
1485 ; CHECK1024-NEXT: ret
1486 entry:
1487 %a = alloca i64
1488 %b = alloca double
1489 tail call void asm sideeffect "", "~{x18},~{x19},~{x20},~{x21},~{x22},~{x23},~{x24},~{x25}"()
1490 tail call void asm sideeffect "", "~{d8},~{d9},~{d10},~{d11},~{d12},~{d13},~{d14},~{d15}"()
1491 store i64 %d, ptr %a
1492 store double %e, ptr %b
1493 ret i32 0
1494 }
1497 define [2 x <vscale x 4 x i1>] @sve_signature_pred_2xv4i1([2 x <vscale x 4 x i1>] %arg1, [2 x <vscale x 4 x i1>] %arg2) nounwind "aarch64_pstate_sm_compatible" {
1498 ; CHECK-LABEL: sve_signature_pred_2xv4i1:
1499 ; CHECK: // %bb.0:
1500 ; CHECK-NEXT: mov p1.b, p3.b
1501 ; CHECK-NEXT: mov p0.b, p2.b
1502 ; CHECK-NEXT: ret
1503 ret [2 x <vscale x 4 x i1>] %arg2
1504 }
1506 define [2 x <vscale x 4 x i1>] @sve_signature_pred_2xv4i1_caller([2 x <vscale x 4 x i1>] %arg1, [2 x <vscale x 4 x i1>] %arg2) nounwind "aarch64_pstate_sm_compatible" {
1507 ; CHECK0-LABEL: sve_signature_pred_2xv4i1_caller:
1508 ; CHECK0: // %bb.0:
1509 ; CHECK0-NEXT: stp x29, x30, [sp, #-16]! // 16-byte Folded Spill
1510 ; CHECK0-NEXT: addvl sp, sp, #-1
1511 ; CHECK0-NEXT: str p5, [sp, #6, mul vl] // 2-byte Folded Spill
1512 ; CHECK0-NEXT: mov p5.b, p0.b
1513 ; CHECK0-NEXT: str p4, [sp, #7, mul vl] // 2-byte Folded Spill
1514 ; CHECK0-NEXT: mov p4.b, p1.b
1515 ; CHECK0-NEXT: mov p0.b, p2.b
1516 ; CHECK0-NEXT: mov p1.b, p3.b
1517 ; CHECK0-NEXT: mov p2.b, p5.b
1518 ; CHECK0-NEXT: mov p3.b, p4.b
1519 ; CHECK0-NEXT: bl sve_signature_pred_2xv4i1
1520 ; CHECK0-NEXT: ldr p5, [sp, #6, mul vl] // 2-byte Folded Reload
1521 ; CHECK0-NEXT: ldr p4, [sp, #7, mul vl] // 2-byte Folded Reload
1522 ; CHECK0-NEXT: addvl sp, sp, #1
1523 ; CHECK0-NEXT: ldp x29, x30, [sp], #16 // 16-byte Folded Reload
1524 ; CHECK0-NEXT: ret
1525 ;
1526 ; CHECK64-LABEL: sve_signature_pred_2xv4i1_caller:
1527 ; CHECK64: // %bb.0:
1528 ; CHECK64-NEXT: sub sp, sp, #80
1529 ; CHECK64-NEXT: stp x29, x30, [sp, #64] // 16-byte Folded Spill
1530 ; CHECK64-NEXT: addvl sp, sp, #-1
1531 ; CHECK64-NEXT: str p5, [sp, #6, mul vl] // 2-byte Folded Spill
1532 ; CHECK64-NEXT: str p4, [sp, #7, mul vl] // 2-byte Folded Spill
1533 ; CHECK64-NEXT: sub sp, sp, #64
1534 ; CHECK64-NEXT: mov p4.b, p1.b
1535 ; CHECK64-NEXT: mov p5.b, p0.b
1536 ; CHECK64-NEXT: mov p0.b, p2.b
1537 ; CHECK64-NEXT: mov p1.b, p3.b
1538 ; CHECK64-NEXT: mov p2.b, p5.b
1539 ; CHECK64-NEXT: mov p3.b, p4.b
1540 ; CHECK64-NEXT: bl sve_signature_pred_2xv4i1
1541 ; CHECK64-NEXT: add sp, sp, #64
1542 ; CHECK64-NEXT: ldr p5, [sp, #6, mul vl] // 2-byte Folded Reload
1543 ; CHECK64-NEXT: ldr p4, [sp, #7, mul vl] // 2-byte Folded Reload
1544 ; CHECK64-NEXT: addvl sp, sp, #1
1545 ; CHECK64-NEXT: ldp x29, x30, [sp, #64] // 16-byte Folded Reload
1546 ; CHECK64-NEXT: add sp, sp, #80
1547 ; CHECK64-NEXT: ret
1548 ;
1549 ; CHECK1024-LABEL: sve_signature_pred_2xv4i1_caller:
1550 ; CHECK1024: // %bb.0:
1551 ; CHECK1024-NEXT: sub sp, sp, #1040
1552 ; CHECK1024-NEXT: str x29, [sp, #1024] // 8-byte Folded Spill
1553 ; CHECK1024-NEXT: str x30, [sp, #1032] // 8-byte Folded Spill
1554 ; CHECK1024-NEXT: addvl sp, sp, #-1
1555 ; CHECK1024-NEXT: str p5, [sp, #6, mul vl] // 2-byte Folded Spill
1556 ; CHECK1024-NEXT: str p4, [sp, #7, mul vl] // 2-byte Folded Spill
1557 ; CHECK1024-NEXT: sub sp, sp, #1024
1558 ; CHECK1024-NEXT: mov p4.b, p1.b
1559 ; CHECK1024-NEXT: mov p5.b, p0.b
1560 ; CHECK1024-NEXT: mov p0.b, p2.b
1561 ; CHECK1024-NEXT: mov p1.b, p3.b
1562 ; CHECK1024-NEXT: mov p2.b, p5.b
1563 ; CHECK1024-NEXT: mov p3.b, p4.b
1564 ; CHECK1024-NEXT: bl sve_signature_pred_2xv4i1
1565 ; CHECK1024-NEXT: add sp, sp, #1024
1566 ; CHECK1024-NEXT: ldr p5, [sp, #6, mul vl] // 2-byte Folded Reload
1567 ; CHECK1024-NEXT: ldr p4, [sp, #7, mul vl] // 2-byte Folded Reload
1568 ; CHECK1024-NEXT: addvl sp, sp, #1
1569 ; CHECK1024-NEXT: ldr x30, [sp, #1032] // 8-byte Folded Reload
1570 ; CHECK1024-NEXT: ldr x29, [sp, #1024] // 8-byte Folded Reload
1571 ; CHECK1024-NEXT: add sp, sp, #1040
1572 ; CHECK1024-NEXT: ret
1573 %res = call [2 x <vscale x 4 x i1>] @sve_signature_pred_2xv4i1([2 x <vscale x 4 x i1>] %arg2, [2 x <vscale x 4 x i1>] %arg1)
1574 ret [2 x <vscale x 4 x i1>] %res
1575 }
1577 define i32 @f128_libcall(fp128 %v0, fp128 %v1, fp128 %v2, fp128 %v3, i32 %a, i32 %b) "aarch64_pstate_sm_compatible" {
1578 ; CHECK0-LABEL: f128_libcall:
1579 ; CHECK0: // %bb.0:
1580 ; CHECK0-NEXT: sub sp, sp, #176
1581 ; CHECK0-NEXT: .cfi_def_cfa_offset 176
1582 ; CHECK0-NEXT: cntd x9
1583 ; CHECK0-NEXT: stp d15, d14, [sp, #64] // 16-byte Folded Spill
1584 ; CHECK0-NEXT: stp d13, d12, [sp, #80] // 16-byte Folded Spill
1585 ; CHECK0-NEXT: stp d11, d10, [sp, #96] // 16-byte Folded Spill
1586 ; CHECK0-NEXT: stp d9, d8, [sp, #112] // 16-byte Folded Spill
1587 ; CHECK0-NEXT: stp x30, x9, [sp, #128] // 16-byte Folded Spill
1588 ; CHECK0-NEXT: stp x22, x21, [sp, #144] // 16-byte Folded Spill
1589 ; CHECK0-NEXT: stp x20, x19, [sp, #160] // 16-byte Folded Spill
1590 ; CHECK0-NEXT: .cfi_offset w19, -8
1591 ; CHECK0-NEXT: .cfi_offset w20, -16
1592 ; CHECK0-NEXT: .cfi_offset w21, -24
1593 ; CHECK0-NEXT: .cfi_offset w22, -32
1594 ; CHECK0-NEXT: .cfi_offset w30, -48
1595 ; CHECK0-NEXT: .cfi_offset b8, -56
1596 ; CHECK0-NEXT: .cfi_offset b9, -64
1597 ; CHECK0-NEXT: .cfi_offset b10, -72
1598 ; CHECK0-NEXT: .cfi_offset b11, -80
1599 ; CHECK0-NEXT: .cfi_offset b12, -88
1600 ; CHECK0-NEXT: .cfi_offset b13, -96
1601 ; CHECK0-NEXT: .cfi_offset b14, -104
1602 ; CHECK0-NEXT: .cfi_offset b15, -112
1603 ; CHECK0-NEXT: mov w19, w1
1604 ; CHECK0-NEXT: mov w20, w0
1605 ; CHECK0-NEXT: stp q0, q1, [sp] // 32-byte Folded Spill
1606 ; CHECK0-NEXT: stp q2, q3, [sp, #32] // 32-byte Folded Spill
1607 ; CHECK0-NEXT: bl __arm_sme_state
1608 ; CHECK0-NEXT: and x21, x0, #0x1
1609 ; CHECK0-NEXT: .cfi_offset vg, -40
1610 ; CHECK0-NEXT: tbz w21, #0, .LBB27_2
1611 ; CHECK0-NEXT: // %bb.1:
1612 ; CHECK0-NEXT: smstop sm
1613 ; CHECK0-NEXT: .LBB27_2:
1614 ; CHECK0-NEXT: ldp q0, q1, [sp] // 32-byte Folded Reload
1615 ; CHECK0-NEXT: bl __lttf2
1616 ; CHECK0-NEXT: tbz w21, #0, .LBB27_4
1617 ; CHECK0-NEXT: // %bb.3:
1618 ; CHECK0-NEXT: smstart sm
1619 ; CHECK0-NEXT: .LBB27_4:
1620 ; CHECK0-NEXT: cmp w0, #0
1621 ; CHECK0-NEXT: .cfi_restore vg
1622 ; CHECK0-NEXT: cset w21, lt
1623 ; CHECK0-NEXT: bl __arm_sme_state
1624 ; CHECK0-NEXT: and x22, x0, #0x1
1625 ; CHECK0-NEXT: .cfi_offset vg, -40
1626 ; CHECK0-NEXT: tbz w22, #0, .LBB27_6
1627 ; CHECK0-NEXT: // %bb.5:
1628 ; CHECK0-NEXT: smstop sm
1629 ; CHECK0-NEXT: .LBB27_6:
1630 ; CHECK0-NEXT: ldp q0, q1, [sp, #32] // 32-byte Folded Reload
1631 ; CHECK0-NEXT: bl __getf2
1632 ; CHECK0-NEXT: tbz w22, #0, .LBB27_8
1633 ; CHECK0-NEXT: // %bb.7:
1634 ; CHECK0-NEXT: smstart sm
1635 ; CHECK0-NEXT: .LBB27_8:
1636 ; CHECK0-NEXT: cmp w0, #0
1637 ; CHECK0-NEXT: cset w8, ge
1638 ; CHECK0-NEXT: tst w8, w21
1639 ; CHECK0-NEXT: csel w0, w20, w19, ne
1640 ; CHECK0-NEXT: .cfi_restore vg
1641 ; CHECK0-NEXT: ldp x20, x19, [sp, #160] // 16-byte Folded Reload
1642 ; CHECK0-NEXT: ldr x30, [sp, #128] // 8-byte Folded Reload
1643 ; CHECK0-NEXT: ldp x22, x21, [sp, #144] // 16-byte Folded Reload
1644 ; CHECK0-NEXT: ldp d9, d8, [sp, #112] // 16-byte Folded Reload
1645 ; CHECK0-NEXT: ldp d11, d10, [sp, #96] // 16-byte Folded Reload
1646 ; CHECK0-NEXT: ldp d13, d12, [sp, #80] // 16-byte Folded Reload
1647 ; CHECK0-NEXT: ldp d15, d14, [sp, #64] // 16-byte Folded Reload
1648 ; CHECK0-NEXT: add sp, sp, #176
1649 ; CHECK0-NEXT: .cfi_def_cfa_offset 0
1650 ; CHECK0-NEXT: .cfi_restore w19
1651 ; CHECK0-NEXT: .cfi_restore w20
1652 ; CHECK0-NEXT: .cfi_restore w21
1653 ; CHECK0-NEXT: .cfi_restore w22
1654 ; CHECK0-NEXT: .cfi_restore w30
1655 ; CHECK0-NEXT: .cfi_restore b8
1656 ; CHECK0-NEXT: .cfi_restore b9
1657 ; CHECK0-NEXT: .cfi_restore b10
1658 ; CHECK0-NEXT: .cfi_restore b11
1659 ; CHECK0-NEXT: .cfi_restore b12
1660 ; CHECK0-NEXT: .cfi_restore b13
1661 ; CHECK0-NEXT: .cfi_restore b14
1662 ; CHECK0-NEXT: .cfi_restore b15
1663 ; CHECK0-NEXT: ret
1664 ;
1665 ; CHECK64-LABEL: f128_libcall:
1666 ; CHECK64: // %bb.0:
1667 ; CHECK64-NEXT: sub sp, sp, #320
1668 ; CHECK64-NEXT: .cfi_def_cfa_offset 320
1669 ; CHECK64-NEXT: cntd x9
1670 ; CHECK64-NEXT: stp d15, d14, [sp, #128] // 16-byte Folded Spill
1671 ; CHECK64-NEXT: stp d13, d12, [sp, #144] // 16-byte Folded Spill
1672 ; CHECK64-NEXT: stp d11, d10, [sp, #160] // 16-byte Folded Spill
1673 ; CHECK64-NEXT: stp d9, d8, [sp, #176] // 16-byte Folded Spill
1674 ; CHECK64-NEXT: stp x29, x30, [sp, #256] // 16-byte Folded Spill
1675 ; CHECK64-NEXT: stp x9, x22, [sp, #272] // 16-byte Folded Spill
1676 ; CHECK64-NEXT: stp x21, x20, [sp, #288] // 16-byte Folded Spill
1677 ; CHECK64-NEXT: str x19, [sp, #304] // 8-byte Folded Spill
1678 ; CHECK64-NEXT: .cfi_offset w19, -16
1679 ; CHECK64-NEXT: .cfi_offset w20, -24
1680 ; CHECK64-NEXT: .cfi_offset w21, -32
1681 ; CHECK64-NEXT: .cfi_offset w22, -40
1682 ; CHECK64-NEXT: .cfi_offset w30, -56
1683 ; CHECK64-NEXT: .cfi_offset w29, -64
1684 ; CHECK64-NEXT: .cfi_offset b8, -136
1685 ; CHECK64-NEXT: .cfi_offset b9, -144
1686 ; CHECK64-NEXT: .cfi_offset b10, -152
1687 ; CHECK64-NEXT: .cfi_offset b11, -160
1688 ; CHECK64-NEXT: .cfi_offset b12, -168
1689 ; CHECK64-NEXT: .cfi_offset b13, -176
1690 ; CHECK64-NEXT: .cfi_offset b14, -184
1691 ; CHECK64-NEXT: .cfi_offset b15, -192
1692 ; CHECK64-NEXT: mov w19, w1
1693 ; CHECK64-NEXT: mov w20, w0
1694 ; CHECK64-NEXT: stp q0, q1, [sp, #64] // 32-byte Folded Spill
1695 ; CHECK64-NEXT: stp q2, q3, [sp, #96] // 32-byte Folded Spill
1696 ; CHECK64-NEXT: bl __arm_sme_state
1697 ; CHECK64-NEXT: and x21, x0, #0x1
1698 ; CHECK64-NEXT: .cfi_offset vg, -48
1699 ; CHECK64-NEXT: tbz w21, #0, .LBB27_2
1700 ; CHECK64-NEXT: // %bb.1:
1701 ; CHECK64-NEXT: smstop sm
1702 ; CHECK64-NEXT: .LBB27_2:
1703 ; CHECK64-NEXT: ldp q0, q1, [sp, #64] // 32-byte Folded Reload
1704 ; CHECK64-NEXT: bl __lttf2
1705 ; CHECK64-NEXT: tbz w21, #0, .LBB27_4
1706 ; CHECK64-NEXT: // %bb.3:
1707 ; CHECK64-NEXT: smstart sm
1708 ; CHECK64-NEXT: .LBB27_4:
1709 ; CHECK64-NEXT: cmp w0, #0
1710 ; CHECK64-NEXT: .cfi_restore vg
1711 ; CHECK64-NEXT: cset w21, lt
1712 ; CHECK64-NEXT: bl __arm_sme_state
1713 ; CHECK64-NEXT: and x22, x0, #0x1
1714 ; CHECK64-NEXT: .cfi_offset vg, -48
1715 ; CHECK64-NEXT: tbz w22, #0, .LBB27_6
1716 ; CHECK64-NEXT: // %bb.5:
1717 ; CHECK64-NEXT: smstop sm
1718 ; CHECK64-NEXT: .LBB27_6:
1719 ; CHECK64-NEXT: ldp q0, q1, [sp, #96] // 32-byte Folded Reload
1720 ; CHECK64-NEXT: bl __getf2
1721 ; CHECK64-NEXT: tbz w22, #0, .LBB27_8
1722 ; CHECK64-NEXT: // %bb.7:
1723 ; CHECK64-NEXT: smstart sm
1724 ; CHECK64-NEXT: .LBB27_8:
1725 ; CHECK64-NEXT: cmp w0, #0
1726 ; CHECK64-NEXT: cset w8, ge
1727 ; CHECK64-NEXT: tst w8, w21
1728 ; CHECK64-NEXT: csel w0, w20, w19, ne
1729 ; CHECK64-NEXT: .cfi_restore vg
1730 ; CHECK64-NEXT: ldp x20, x19, [sp, #296] // 16-byte Folded Reload
1731 ; CHECK64-NEXT: ldp x22, x21, [sp, #280] // 16-byte Folded Reload
1732 ; CHECK64-NEXT: ldp x29, x30, [sp, #256] // 16-byte Folded Reload
1733 ; CHECK64-NEXT: ldp d9, d8, [sp, #176] // 16-byte Folded Reload
1734 ; CHECK64-NEXT: ldp d11, d10, [sp, #160] // 16-byte Folded Reload
1735 ; CHECK64-NEXT: ldp d13, d12, [sp, #144] // 16-byte Folded Reload
1736 ; CHECK64-NEXT: ldp d15, d14, [sp, #128] // 16-byte Folded Reload
1737 ; CHECK64-NEXT: add sp, sp, #320
1738 ; CHECK64-NEXT: .cfi_def_cfa_offset 0
1739 ; CHECK64-NEXT: .cfi_restore w19
1740 ; CHECK64-NEXT: .cfi_restore w20
1741 ; CHECK64-NEXT: .cfi_restore w21
1742 ; CHECK64-NEXT: .cfi_restore w22
1743 ; CHECK64-NEXT: .cfi_restore w30
1744 ; CHECK64-NEXT: .cfi_restore w29
1745 ; CHECK64-NEXT: .cfi_restore b8
1746 ; CHECK64-NEXT: .cfi_restore b9
1747 ; CHECK64-NEXT: .cfi_restore b10
1748 ; CHECK64-NEXT: .cfi_restore b11
1749 ; CHECK64-NEXT: .cfi_restore b12
1750 ; CHECK64-NEXT: .cfi_restore b13
1751 ; CHECK64-NEXT: .cfi_restore b14
1752 ; CHECK64-NEXT: .cfi_restore b15
1753 ; CHECK64-NEXT: ret
1754 ;
1755 ; CHECK1024-LABEL: f128_libcall:
1756 ; CHECK1024: // %bb.0:
1757 ; CHECK1024-NEXT: sub sp, sp, #1152
1758 ; CHECK1024-NEXT: .cfi_def_cfa_offset 1152
1759 ; CHECK1024-NEXT: cntd x9
1760 ; CHECK1024-NEXT: stp d15, d14, [sp] // 16-byte Folded Spill
1761 ; CHECK1024-NEXT: stp d13, d12, [sp, #16] // 16-byte Folded Spill
1762 ; CHECK1024-NEXT: stp d11, d10, [sp, #32] // 16-byte Folded Spill
1763 ; CHECK1024-NEXT: stp d9, d8, [sp, #48] // 16-byte Folded Spill
1764 ; CHECK1024-NEXT: str x29, [sp, #1088] // 8-byte Folded Spill
1765 ; CHECK1024-NEXT: str x30, [sp, #1096] // 8-byte Folded Spill
1766 ; CHECK1024-NEXT: str x9, [sp, #1104] // 8-byte Folded Spill
1767 ; CHECK1024-NEXT: str x22, [sp, #1112] // 8-byte Folded Spill
1768 ; CHECK1024-NEXT: str x21, [sp, #1120] // 8-byte Folded Spill
1769 ; CHECK1024-NEXT: str x20, [sp, #1128] // 8-byte Folded Spill
1770 ; CHECK1024-NEXT: str x19, [sp, #1136] // 8-byte Folded Spill
1771 ; CHECK1024-NEXT: .cfi_offset w19, -16
1772 ; CHECK1024-NEXT: .cfi_offset w20, -24
1773 ; CHECK1024-NEXT: .cfi_offset w21, -32
1774 ; CHECK1024-NEXT: .cfi_offset w22, -40
1775 ; CHECK1024-NEXT: .cfi_offset w30, -56
1776 ; CHECK1024-NEXT: .cfi_offset w29, -64
1777 ; CHECK1024-NEXT: .cfi_offset b8, -1096
1778 ; CHECK1024-NEXT: .cfi_offset b9, -1104
1779 ; CHECK1024-NEXT: .cfi_offset b10, -1112
1780 ; CHECK1024-NEXT: .cfi_offset b11, -1120
1781 ; CHECK1024-NEXT: .cfi_offset b12, -1128
1782 ; CHECK1024-NEXT: .cfi_offset b13, -1136
1783 ; CHECK1024-NEXT: .cfi_offset b14, -1144
1784 ; CHECK1024-NEXT: .cfi_offset b15, -1152
1785 ; CHECK1024-NEXT: sub sp, sp, #1088
1786 ; CHECK1024-NEXT: .cfi_def_cfa_offset 2240
1787 ; CHECK1024-NEXT: mov w19, w1
1788 ; CHECK1024-NEXT: mov w20, w0
1789 ; CHECK1024-NEXT: str q3, [sp, #1072] // 16-byte Folded Spill
1790 ; CHECK1024-NEXT: str q2, [sp, #1056] // 16-byte Folded Spill
1791 ; CHECK1024-NEXT: str q1, [sp, #1040] // 16-byte Folded Spill
1792 ; CHECK1024-NEXT: str q0, [sp, #1024] // 16-byte Folded Spill
1793 ; CHECK1024-NEXT: bl __arm_sme_state
1794 ; CHECK1024-NEXT: and x21, x0, #0x1
1795 ; CHECK1024-NEXT: .cfi_offset vg, -48
1796 ; CHECK1024-NEXT: tbz w21, #0, .LBB27_2
1797 ; CHECK1024-NEXT: // %bb.1:
1798 ; CHECK1024-NEXT: smstop sm
1799 ; CHECK1024-NEXT: .LBB27_2:
1800 ; CHECK1024-NEXT: ldr q0, [sp, #1024] // 16-byte Folded Reload
1801 ; CHECK1024-NEXT: ldr q1, [sp, #1040] // 16-byte Folded Reload
1802 ; CHECK1024-NEXT: bl __lttf2
1803 ; CHECK1024-NEXT: tbz w21, #0, .LBB27_4
1804 ; CHECK1024-NEXT: // %bb.3:
1805 ; CHECK1024-NEXT: smstart sm
1806 ; CHECK1024-NEXT: .LBB27_4:
1807 ; CHECK1024-NEXT: cmp w0, #0
1808 ; CHECK1024-NEXT: .cfi_restore vg
1809 ; CHECK1024-NEXT: cset w21, lt
1810 ; CHECK1024-NEXT: bl __arm_sme_state
1811 ; CHECK1024-NEXT: and x22, x0, #0x1
1812 ; CHECK1024-NEXT: .cfi_offset vg, -48
1813 ; CHECK1024-NEXT: tbz w22, #0, .LBB27_6
1814 ; CHECK1024-NEXT: // %bb.5:
1815 ; CHECK1024-NEXT: smstop sm
1816 ; CHECK1024-NEXT: .LBB27_6:
1817 ; CHECK1024-NEXT: ldr q0, [sp, #1056] // 16-byte Folded Reload
1818 ; CHECK1024-NEXT: ldr q1, [sp, #1072] // 16-byte Folded Reload
1819 ; CHECK1024-NEXT: bl __getf2
1820 ; CHECK1024-NEXT: tbz w22, #0, .LBB27_8
1821 ; CHECK1024-NEXT: // %bb.7:
1822 ; CHECK1024-NEXT: smstart sm
1823 ; CHECK1024-NEXT: .LBB27_8:
1824 ; CHECK1024-NEXT: cmp w0, #0
1825 ; CHECK1024-NEXT: cset w8, ge
1826 ; CHECK1024-NEXT: tst w8, w21
1827 ; CHECK1024-NEXT: csel w0, w20, w19, ne
1828 ; CHECK1024-NEXT: .cfi_restore vg
1829 ; CHECK1024-NEXT: add sp, sp, #1088
1830 ; CHECK1024-NEXT: .cfi_def_cfa_offset 1152
1831 ; CHECK1024-NEXT: ldp d9, d8, [sp, #48] // 16-byte Folded Reload
1832 ; CHECK1024-NEXT: ldr x19, [sp, #1136] // 8-byte Folded Reload
1833 ; CHECK1024-NEXT: ldp d11, d10, [sp, #32] // 16-byte Folded Reload
1834 ; CHECK1024-NEXT: ldr x20, [sp, #1128] // 8-byte Folded Reload
1835 ; CHECK1024-NEXT: ldr x21, [sp, #1120] // 8-byte Folded Reload
1836 ; CHECK1024-NEXT: ldr x22, [sp, #1112] // 8-byte Folded Reload
1837 ; CHECK1024-NEXT: ldr x30, [sp, #1096] // 8-byte Folded Reload
1838 ; CHECK1024-NEXT: ldr x29, [sp, #1088] // 8-byte Folded Reload
1839 ; CHECK1024-NEXT: ldp d13, d12, [sp, #16] // 16-byte Folded Reload
1840 ; CHECK1024-NEXT: ldp d15, d14, [sp] // 16-byte Folded Reload
1841 ; CHECK1024-NEXT: add sp, sp, #1152
1842 ; CHECK1024-NEXT: .cfi_def_cfa_offset 0
1843 ; CHECK1024-NEXT: .cfi_restore w19
1844 ; CHECK1024-NEXT: .cfi_restore w20
1845 ; CHECK1024-NEXT: .cfi_restore w21
1846 ; CHECK1024-NEXT: .cfi_restore w22
1847 ; CHECK1024-NEXT: .cfi_restore w30
1848 ; CHECK1024-NEXT: .cfi_restore w29
1849 ; CHECK1024-NEXT: .cfi_restore b8
1850 ; CHECK1024-NEXT: .cfi_restore b9
1851 ; CHECK1024-NEXT: .cfi_restore b10
1852 ; CHECK1024-NEXT: .cfi_restore b11
1853 ; CHECK1024-NEXT: .cfi_restore b12
1854 ; CHECK1024-NEXT: .cfi_restore b13
1855 ; CHECK1024-NEXT: .cfi_restore b14
1856 ; CHECK1024-NEXT: .cfi_restore b15
1857 ; CHECK1024-NEXT: ret
1858 %c0 = fcmp olt fp128 %v0, %v1
1859 %c1 = fcmp oge fp128 %v2, %v3
1860 %cr = and i1 %c1, %c0
1861 %sel = select i1 %cr, i32 %a, i32 %b
1862 ret i32 %sel
1863 }
1865 define i32 @svecc_call(<4 x i16> %P0, ptr %P1, i32 %P2, <vscale x 16 x i8> %P3, i16 %P4) "aarch64_pstate_sm_compatible" {
1866 ; CHECK0-LABEL: svecc_call:
1867 ; CHECK0: // %bb.0: // %entry
1868 ; CHECK0-NEXT: stp x29, x30, [sp, #-48]! // 16-byte Folded Spill
1869 ; CHECK0-NEXT: .cfi_def_cfa_offset 48
1870 ; CHECK0-NEXT: cntd x9
1871 ; CHECK0-NEXT: stp x9, x28, [sp, #16] // 16-byte Folded Spill
1872 ; CHECK0-NEXT: stp x27, x19, [sp, #32] // 16-byte Folded Spill
1873 ; CHECK0-NEXT: .cfi_offset w19, -8
1874 ; CHECK0-NEXT: .cfi_offset w27, -16
1875 ; CHECK0-NEXT: .cfi_offset w28, -24
1876 ; CHECK0-NEXT: .cfi_offset w30, -40
1877 ; CHECK0-NEXT: .cfi_offset w29, -48
1878 ; CHECK0-NEXT: addvl sp, sp, #-18
1879 ; CHECK0-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0x30, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 48 + 144 * VG
1880 ; CHECK0-NEXT: str p15, [sp, #4, mul vl] // 2-byte Folded Spill
1881 ; CHECK0-NEXT: str p14, [sp, #5, mul vl] // 2-byte Folded Spill
1882 ; CHECK0-NEXT: str p13, [sp, #6, mul vl] // 2-byte Folded Spill
1883 ; CHECK0-NEXT: str p12, [sp, #7, mul vl] // 2-byte Folded Spill
1884 ; CHECK0-NEXT: str p11, [sp, #8, mul vl] // 2-byte Folded Spill
1885 ; CHECK0-NEXT: str p10, [sp, #9, mul vl] // 2-byte Folded Spill
1886 ; CHECK0-NEXT: str p9, [sp, #10, mul vl] // 2-byte Folded Spill
1887 ; CHECK0-NEXT: str p8, [sp, #11, mul vl] // 2-byte Folded Spill
1888 ; CHECK0-NEXT: str p7, [sp, #12, mul vl] // 2-byte Folded Spill
1889 ; CHECK0-NEXT: str p6, [sp, #13, mul vl] // 2-byte Folded Spill
1890 ; CHECK0-NEXT: str p5, [sp, #14, mul vl] // 2-byte Folded Spill
1891 ; CHECK0-NEXT: str p4, [sp, #15, mul vl] // 2-byte Folded Spill
1892 ; CHECK0-NEXT: str z23, [sp, #2, mul vl] // 16-byte Folded Spill
1893 ; CHECK0-NEXT: str z22, [sp, #3, mul vl] // 16-byte Folded Spill
1894 ; CHECK0-NEXT: str z21, [sp, #4, mul vl] // 16-byte Folded Spill
1895 ; CHECK0-NEXT: str z20, [sp, #5, mul vl] // 16-byte Folded Spill
1896 ; CHECK0-NEXT: str z19, [sp, #6, mul vl] // 16-byte Folded Spill
1897 ; CHECK0-NEXT: str z18, [sp, #7, mul vl] // 16-byte Folded Spill
1898 ; CHECK0-NEXT: str z17, [sp, #8, mul vl] // 16-byte Folded Spill
1899 ; CHECK0-NEXT: str z16, [sp, #9, mul vl] // 16-byte Folded Spill
1900 ; CHECK0-NEXT: str z15, [sp, #10, mul vl] // 16-byte Folded Spill
1901 ; CHECK0-NEXT: str z14, [sp, #11, mul vl] // 16-byte Folded Spill
1902 ; CHECK0-NEXT: str z13, [sp, #12, mul vl] // 16-byte Folded Spill
1903 ; CHECK0-NEXT: str z12, [sp, #13, mul vl] // 16-byte Folded Spill
1904 ; CHECK0-NEXT: str z11, [sp, #14, mul vl] // 16-byte Folded Spill
1905 ; CHECK0-NEXT: str z10, [sp, #15, mul vl] // 16-byte Folded Spill
1906 ; CHECK0-NEXT: str z9, [sp, #16, mul vl] // 16-byte Folded Spill
1907 ; CHECK0-NEXT: str z8, [sp, #17, mul vl] // 16-byte Folded Spill
1908 ; CHECK0-NEXT: .cfi_escape 0x10, 0x48, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 48 - 8 * VG
1909 ; CHECK0-NEXT: .cfi_escape 0x10, 0x49, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x70, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d9 @ cfa - 48 - 16 * VG
1910 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4a, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x68, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d10 @ cfa - 48 - 24 * VG
1911 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4b, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x60, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d11 @ cfa - 48 - 32 * VG
1912 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4c, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x58, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d12 @ cfa - 48 - 40 * VG
1913 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4d, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x50, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d13 @ cfa - 48 - 48 * VG
1914 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4e, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x48, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d14 @ cfa - 48 - 56 * VG
1915 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4f, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x40, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d15 @ cfa - 48 - 64 * VG
1916 ; CHECK0-NEXT: mov x8, x0
1917 ; CHECK0-NEXT: //APP
1918 ; CHECK0-NEXT: //NO_APP
1919 ; CHECK0-NEXT: bl __arm_sme_state
1920 ; CHECK0-NEXT: and x19, x0, #0x1
1921 ; CHECK0-NEXT: .cfi_offset vg, -32
1922 ; CHECK0-NEXT: tbz w19, #0, .LBB28_2
1923 ; CHECK0-NEXT: // %bb.1: // %entry
1924 ; CHECK0-NEXT: smstop sm
1925 ; CHECK0-NEXT: .LBB28_2: // %entry
1926 ; CHECK0-NEXT: mov x0, x8
1927 ; CHECK0-NEXT: mov w1, #45 // =0x2d
1928 ; CHECK0-NEXT: mov w2, #37 // =0x25
1929 ; CHECK0-NEXT: bl memset
1930 ; CHECK0-NEXT: tbz w19, #0, .LBB28_4
1931 ; CHECK0-NEXT: // %bb.3: // %entry
1932 ; CHECK0-NEXT: smstart sm
1933 ; CHECK0-NEXT: .LBB28_4: // %entry
1934 ; CHECK0-NEXT: mov w0, #22647 // =0x5877
1935 ; CHECK0-NEXT: movk w0, #59491, lsl #16
1936 ; CHECK0-NEXT: .cfi_restore vg
1937 ; CHECK0-NEXT: ldr z23, [sp, #2, mul vl] // 16-byte Folded Reload
1938 ; CHECK0-NEXT: ldr z22, [sp, #3, mul vl] // 16-byte Folded Reload
1939 ; CHECK0-NEXT: ldr z21, [sp, #4, mul vl] // 16-byte Folded Reload
1940 ; CHECK0-NEXT: ldr z20, [sp, #5, mul vl] // 16-byte Folded Reload
1941 ; CHECK0-NEXT: ldr z19, [sp, #6, mul vl] // 16-byte Folded Reload
1942 ; CHECK0-NEXT: ldr z18, [sp, #7, mul vl] // 16-byte Folded Reload
1943 ; CHECK0-NEXT: ldr z17, [sp, #8, mul vl] // 16-byte Folded Reload
1944 ; CHECK0-NEXT: ldr z16, [sp, #9, mul vl] // 16-byte Folded Reload
1945 ; CHECK0-NEXT: ldr z15, [sp, #10, mul vl] // 16-byte Folded Reload
1946 ; CHECK0-NEXT: ldr z14, [sp, #11, mul vl] // 16-byte Folded Reload
1947 ; CHECK0-NEXT: ldr z13, [sp, #12, mul vl] // 16-byte Folded Reload
1948 ; CHECK0-NEXT: ldr z12, [sp, #13, mul vl] // 16-byte Folded Reload
1949 ; CHECK0-NEXT: ldr z11, [sp, #14, mul vl] // 16-byte Folded Reload
1950 ; CHECK0-NEXT: ldr z10, [sp, #15, mul vl] // 16-byte Folded Reload
1951 ; CHECK0-NEXT: ldr z9, [sp, #16, mul vl] // 16-byte Folded Reload
1952 ; CHECK0-NEXT: ldr z8, [sp, #17, mul vl] // 16-byte Folded Reload
1953 ; CHECK0-NEXT: ldr p15, [sp, #4, mul vl] // 2-byte Folded Reload
1954 ; CHECK0-NEXT: ldr p14, [sp, #5, mul vl] // 2-byte Folded Reload
1955 ; CHECK0-NEXT: ldr p13, [sp, #6, mul vl] // 2-byte Folded Reload
1956 ; CHECK0-NEXT: ldr p12, [sp, #7, mul vl] // 2-byte Folded Reload
1957 ; CHECK0-NEXT: ldr p11, [sp, #8, mul vl] // 2-byte Folded Reload
1958 ; CHECK0-NEXT: ldr p10, [sp, #9, mul vl] // 2-byte Folded Reload
1959 ; CHECK0-NEXT: ldr p9, [sp, #10, mul vl] // 2-byte Folded Reload
1960 ; CHECK0-NEXT: ldr p8, [sp, #11, mul vl] // 2-byte Folded Reload
1961 ; CHECK0-NEXT: ldr p7, [sp, #12, mul vl] // 2-byte Folded Reload
1962 ; CHECK0-NEXT: ldr p6, [sp, #13, mul vl] // 2-byte Folded Reload
1963 ; CHECK0-NEXT: ldr p5, [sp, #14, mul vl] // 2-byte Folded Reload
1964 ; CHECK0-NEXT: ldr p4, [sp, #15, mul vl] // 2-byte Folded Reload
1965 ; CHECK0-NEXT: addvl sp, sp, #18
1966 ; CHECK0-NEXT: .cfi_def_cfa wsp, 48
1967 ; CHECK0-NEXT: .cfi_restore z8
1968 ; CHECK0-NEXT: .cfi_restore z9
1969 ; CHECK0-NEXT: .cfi_restore z10
1970 ; CHECK0-NEXT: .cfi_restore z11
1971 ; CHECK0-NEXT: .cfi_restore z12
1972 ; CHECK0-NEXT: .cfi_restore z13
1973 ; CHECK0-NEXT: .cfi_restore z14
1974 ; CHECK0-NEXT: .cfi_restore z15
1975 ; CHECK0-NEXT: ldp x27, x19, [sp, #32] // 16-byte Folded Reload
1976 ; CHECK0-NEXT: ldr x28, [sp, #24] // 8-byte Folded Reload
1977 ; CHECK0-NEXT: ldp x29, x30, [sp], #48 // 16-byte Folded Reload
1978 ; CHECK0-NEXT: .cfi_def_cfa_offset 0
1979 ; CHECK0-NEXT: .cfi_restore w19
1980 ; CHECK0-NEXT: .cfi_restore w27
1981 ; CHECK0-NEXT: .cfi_restore w28
1982 ; CHECK0-NEXT: .cfi_restore w30
1983 ; CHECK0-NEXT: .cfi_restore w29
1984 ; CHECK0-NEXT: ret
1985 ;
1986 ; CHECK64-LABEL: svecc_call:
1987 ; CHECK64: // %bb.0: // %entry
1988 ; CHECK64-NEXT: sub sp, sp, #112
1989 ; CHECK64-NEXT: .cfi_def_cfa_offset 112
1990 ; CHECK64-NEXT: cntd x9
1991 ; CHECK64-NEXT: stp x29, x30, [sp, #64] // 16-byte Folded Spill
1992 ; CHECK64-NEXT: stp x9, x28, [sp, #80] // 16-byte Folded Spill
1993 ; CHECK64-NEXT: stp x27, x19, [sp, #96] // 16-byte Folded Spill
1994 ; CHECK64-NEXT: .cfi_offset w19, -8
1995 ; CHECK64-NEXT: .cfi_offset w27, -16
1996 ; CHECK64-NEXT: .cfi_offset w28, -24
1997 ; CHECK64-NEXT: .cfi_offset w30, -40
1998 ; CHECK64-NEXT: .cfi_offset w29, -48
1999 ; CHECK64-NEXT: addvl sp, sp, #-18
2000 ; CHECK64-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xf0, 0x00, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 112 + 144 * VG
2001 ; CHECK64-NEXT: str p15, [sp, #4, mul vl] // 2-byte Folded Spill
2002 ; CHECK64-NEXT: str p14, [sp, #5, mul vl] // 2-byte Folded Spill
2003 ; CHECK64-NEXT: str p13, [sp, #6, mul vl] // 2-byte Folded Spill
2004 ; CHECK64-NEXT: str p12, [sp, #7, mul vl] // 2-byte Folded Spill
2005 ; CHECK64-NEXT: str p11, [sp, #8, mul vl] // 2-byte Folded Spill
2006 ; CHECK64-NEXT: str p10, [sp, #9, mul vl] // 2-byte Folded Spill
2007 ; CHECK64-NEXT: str p9, [sp, #10, mul vl] // 2-byte Folded Spill
2008 ; CHECK64-NEXT: str p8, [sp, #11, mul vl] // 2-byte Folded Spill
2009 ; CHECK64-NEXT: str p7, [sp, #12, mul vl] // 2-byte Folded Spill
2010 ; CHECK64-NEXT: str p6, [sp, #13, mul vl] // 2-byte Folded Spill
2011 ; CHECK64-NEXT: str p5, [sp, #14, mul vl] // 2-byte Folded Spill
2012 ; CHECK64-NEXT: str p4, [sp, #15, mul vl] // 2-byte Folded Spill
2013 ; CHECK64-NEXT: str z23, [sp, #2, mul vl] // 16-byte Folded Spill
2014 ; CHECK64-NEXT: str z22, [sp, #3, mul vl] // 16-byte Folded Spill
2015 ; CHECK64-NEXT: str z21, [sp, #4, mul vl] // 16-byte Folded Spill
2016 ; CHECK64-NEXT: str z20, [sp, #5, mul vl] // 16-byte Folded Spill
2017 ; CHECK64-NEXT: str z19, [sp, #6, mul vl] // 16-byte Folded Spill
2018 ; CHECK64-NEXT: str z18, [sp, #7, mul vl] // 16-byte Folded Spill
2019 ; CHECK64-NEXT: str z17, [sp, #8, mul vl] // 16-byte Folded Spill
2020 ; CHECK64-NEXT: str z16, [sp, #9, mul vl] // 16-byte Folded Spill
2021 ; CHECK64-NEXT: str z15, [sp, #10, mul vl] // 16-byte Folded Spill
2022 ; CHECK64-NEXT: str z14, [sp, #11, mul vl] // 16-byte Folded Spill
2023 ; CHECK64-NEXT: str z13, [sp, #12, mul vl] // 16-byte Folded Spill
2024 ; CHECK64-NEXT: str z12, [sp, #13, mul vl] // 16-byte Folded Spill
2025 ; CHECK64-NEXT: str z11, [sp, #14, mul vl] // 16-byte Folded Spill
2026 ; CHECK64-NEXT: str z10, [sp, #15, mul vl] // 16-byte Folded Spill
2027 ; CHECK64-NEXT: str z9, [sp, #16, mul vl] // 16-byte Folded Spill
2028 ; CHECK64-NEXT: str z8, [sp, #17, mul vl] // 16-byte Folded Spill
2029 ; CHECK64-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 112 - 8 * VG
2030 ; CHECK64-NEXT: .cfi_escape 0x10, 0x49, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x70, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d9 @ cfa - 112 - 16 * VG
2031 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4a, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x68, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d10 @ cfa - 112 - 24 * VG
2032 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4b, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x60, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d11 @ cfa - 112 - 32 * VG
2033 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4c, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x58, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d12 @ cfa - 112 - 40 * VG
2034 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4d, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x50, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d13 @ cfa - 112 - 48 * VG
2035 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4e, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x48, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d14 @ cfa - 112 - 56 * VG
2036 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4f, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x40, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d15 @ cfa - 112 - 64 * VG
2037 ; CHECK64-NEXT: sub sp, sp, #64
2038 ; CHECK64-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xb0, 0x01, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 176 + 144 * VG
2039 ; CHECK64-NEXT: mov x8, x0
2040 ; CHECK64-NEXT: //APP
2041 ; CHECK64-NEXT: //NO_APP
2042 ; CHECK64-NEXT: bl __arm_sme_state
2043 ; CHECK64-NEXT: and x19, x0, #0x1
2044 ; CHECK64-NEXT: .cfi_offset vg, -32
2045 ; CHECK64-NEXT: tbz w19, #0, .LBB28_2
2046 ; CHECK64-NEXT: // %bb.1: // %entry
2047 ; CHECK64-NEXT: smstop sm
2048 ; CHECK64-NEXT: .LBB28_2: // %entry
2049 ; CHECK64-NEXT: mov x0, x8
2050 ; CHECK64-NEXT: mov w1, #45 // =0x2d
2051 ; CHECK64-NEXT: mov w2, #37 // =0x25
2052 ; CHECK64-NEXT: bl memset
2053 ; CHECK64-NEXT: tbz w19, #0, .LBB28_4
2054 ; CHECK64-NEXT: // %bb.3: // %entry
2055 ; CHECK64-NEXT: smstart sm
2056 ; CHECK64-NEXT: .LBB28_4: // %entry
2057 ; CHECK64-NEXT: mov w0, #22647 // =0x5877
2058 ; CHECK64-NEXT: movk w0, #59491, lsl #16
2059 ; CHECK64-NEXT: .cfi_restore vg
2060 ; CHECK64-NEXT: add sp, sp, #64
2061 ; CHECK64-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xf0, 0x00, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 112 + 144 * VG
2062 ; CHECK64-NEXT: ldr z23, [sp, #2, mul vl] // 16-byte Folded Reload
2063 ; CHECK64-NEXT: ldr z22, [sp, #3, mul vl] // 16-byte Folded Reload
2064 ; CHECK64-NEXT: ldr z21, [sp, #4, mul vl] // 16-byte Folded Reload
2065 ; CHECK64-NEXT: ldr z20, [sp, #5, mul vl] // 16-byte Folded Reload
2066 ; CHECK64-NEXT: ldr z19, [sp, #6, mul vl] // 16-byte Folded Reload
2067 ; CHECK64-NEXT: ldr z18, [sp, #7, mul vl] // 16-byte Folded Reload
2068 ; CHECK64-NEXT: ldr z17, [sp, #8, mul vl] // 16-byte Folded Reload
2069 ; CHECK64-NEXT: ldr z16, [sp, #9, mul vl] // 16-byte Folded Reload
2070 ; CHECK64-NEXT: ldr z15, [sp, #10, mul vl] // 16-byte Folded Reload
2071 ; CHECK64-NEXT: ldr z14, [sp, #11, mul vl] // 16-byte Folded Reload
2072 ; CHECK64-NEXT: ldr z13, [sp, #12, mul vl] // 16-byte Folded Reload
2073 ; CHECK64-NEXT: ldr z12, [sp, #13, mul vl] // 16-byte Folded Reload
2074 ; CHECK64-NEXT: ldr z11, [sp, #14, mul vl] // 16-byte Folded Reload
2075 ; CHECK64-NEXT: ldr z10, [sp, #15, mul vl] // 16-byte Folded Reload
2076 ; CHECK64-NEXT: ldr z9, [sp, #16, mul vl] // 16-byte Folded Reload
2077 ; CHECK64-NEXT: ldr z8, [sp, #17, mul vl] // 16-byte Folded Reload
2078 ; CHECK64-NEXT: ldr p15, [sp, #4, mul vl] // 2-byte Folded Reload
2079 ; CHECK64-NEXT: ldr p14, [sp, #5, mul vl] // 2-byte Folded Reload
2080 ; CHECK64-NEXT: ldr p13, [sp, #6, mul vl] // 2-byte Folded Reload
2081 ; CHECK64-NEXT: ldr p12, [sp, #7, mul vl] // 2-byte Folded Reload
2082 ; CHECK64-NEXT: ldr p11, [sp, #8, mul vl] // 2-byte Folded Reload
2083 ; CHECK64-NEXT: ldr p10, [sp, #9, mul vl] // 2-byte Folded Reload
2084 ; CHECK64-NEXT: ldr p9, [sp, #10, mul vl] // 2-byte Folded Reload
2085 ; CHECK64-NEXT: ldr p8, [sp, #11, mul vl] // 2-byte Folded Reload
2086 ; CHECK64-NEXT: ldr p7, [sp, #12, mul vl] // 2-byte Folded Reload
2087 ; CHECK64-NEXT: ldr p6, [sp, #13, mul vl] // 2-byte Folded Reload
2088 ; CHECK64-NEXT: ldr p5, [sp, #14, mul vl] // 2-byte Folded Reload
2089 ; CHECK64-NEXT: ldr p4, [sp, #15, mul vl] // 2-byte Folded Reload
2090 ; CHECK64-NEXT: addvl sp, sp, #18
2091 ; CHECK64-NEXT: .cfi_def_cfa wsp, 112
2092 ; CHECK64-NEXT: .cfi_restore z8
2093 ; CHECK64-NEXT: .cfi_restore z9
2094 ; CHECK64-NEXT: .cfi_restore z10
2095 ; CHECK64-NEXT: .cfi_restore z11
2096 ; CHECK64-NEXT: .cfi_restore z12
2097 ; CHECK64-NEXT: .cfi_restore z13
2098 ; CHECK64-NEXT: .cfi_restore z14
2099 ; CHECK64-NEXT: .cfi_restore z15
2100 ; CHECK64-NEXT: ldp x27, x19, [sp, #96] // 16-byte Folded Reload
2101 ; CHECK64-NEXT: ldr x28, [sp, #88] // 8-byte Folded Reload
2102 ; CHECK64-NEXT: ldp x29, x30, [sp, #64] // 16-byte Folded Reload
2103 ; CHECK64-NEXT: add sp, sp, #112
2104 ; CHECK64-NEXT: .cfi_def_cfa_offset 0
2105 ; CHECK64-NEXT: .cfi_restore w19
2106 ; CHECK64-NEXT: .cfi_restore w27
2107 ; CHECK64-NEXT: .cfi_restore w28
2108 ; CHECK64-NEXT: .cfi_restore w30
2109 ; CHECK64-NEXT: .cfi_restore w29
2110 ; CHECK64-NEXT: ret
2111 ;
2112 ; CHECK1024-LABEL: svecc_call:
2113 ; CHECK1024: // %bb.0: // %entry
2114 ; CHECK1024-NEXT: sub sp, sp, #1072
2115 ; CHECK1024-NEXT: .cfi_def_cfa_offset 1072
2116 ; CHECK1024-NEXT: cntd x9
2117 ; CHECK1024-NEXT: str x29, [sp, #1024] // 8-byte Folded Spill
2118 ; CHECK1024-NEXT: str x30, [sp, #1032] // 8-byte Folded Spill
2119 ; CHECK1024-NEXT: str x9, [sp, #1040] // 8-byte Folded Spill
2120 ; CHECK1024-NEXT: str x28, [sp, #1048] // 8-byte Folded Spill
2121 ; CHECK1024-NEXT: str x27, [sp, #1056] // 8-byte Folded Spill
2122 ; CHECK1024-NEXT: str x19, [sp, #1064] // 8-byte Folded Spill
2123 ; CHECK1024-NEXT: .cfi_offset w19, -8
2124 ; CHECK1024-NEXT: .cfi_offset w27, -16
2125 ; CHECK1024-NEXT: .cfi_offset w28, -24
2126 ; CHECK1024-NEXT: .cfi_offset w30, -40
2127 ; CHECK1024-NEXT: .cfi_offset w29, -48
2128 ; CHECK1024-NEXT: addvl sp, sp, #-18
2129 ; CHECK1024-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xb0, 0x08, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 1072 + 144 * VG
2130 ; CHECK1024-NEXT: str p15, [sp, #4, mul vl] // 2-byte Folded Spill
2131 ; CHECK1024-NEXT: str p14, [sp, #5, mul vl] // 2-byte Folded Spill
2132 ; CHECK1024-NEXT: str p13, [sp, #6, mul vl] // 2-byte Folded Spill
2133 ; CHECK1024-NEXT: str p12, [sp, #7, mul vl] // 2-byte Folded Spill
2134 ; CHECK1024-NEXT: str p11, [sp, #8, mul vl] // 2-byte Folded Spill
2135 ; CHECK1024-NEXT: str p10, [sp, #9, mul vl] // 2-byte Folded Spill
2136 ; CHECK1024-NEXT: str p9, [sp, #10, mul vl] // 2-byte Folded Spill
2137 ; CHECK1024-NEXT: str p8, [sp, #11, mul vl] // 2-byte Folded Spill
2138 ; CHECK1024-NEXT: str p7, [sp, #12, mul vl] // 2-byte Folded Spill
2139 ; CHECK1024-NEXT: str p6, [sp, #13, mul vl] // 2-byte Folded Spill
2140 ; CHECK1024-NEXT: str p5, [sp, #14, mul vl] // 2-byte Folded Spill
2141 ; CHECK1024-NEXT: str p4, [sp, #15, mul vl] // 2-byte Folded Spill
2142 ; CHECK1024-NEXT: str z23, [sp, #2, mul vl] // 16-byte Folded Spill
2143 ; CHECK1024-NEXT: str z22, [sp, #3, mul vl] // 16-byte Folded Spill
2144 ; CHECK1024-NEXT: str z21, [sp, #4, mul vl] // 16-byte Folded Spill
2145 ; CHECK1024-NEXT: str z20, [sp, #5, mul vl] // 16-byte Folded Spill
2146 ; CHECK1024-NEXT: str z19, [sp, #6, mul vl] // 16-byte Folded Spill
2147 ; CHECK1024-NEXT: str z18, [sp, #7, mul vl] // 16-byte Folded Spill
2148 ; CHECK1024-NEXT: str z17, [sp, #8, mul vl] // 16-byte Folded Spill
2149 ; CHECK1024-NEXT: str z16, [sp, #9, mul vl] // 16-byte Folded Spill
2150 ; CHECK1024-NEXT: str z15, [sp, #10, mul vl] // 16-byte Folded Spill
2151 ; CHECK1024-NEXT: str z14, [sp, #11, mul vl] // 16-byte Folded Spill
2152 ; CHECK1024-NEXT: str z13, [sp, #12, mul vl] // 16-byte Folded Spill
2153 ; CHECK1024-NEXT: str z12, [sp, #13, mul vl] // 16-byte Folded Spill
2154 ; CHECK1024-NEXT: str z11, [sp, #14, mul vl] // 16-byte Folded Spill
2155 ; CHECK1024-NEXT: str z10, [sp, #15, mul vl] // 16-byte Folded Spill
2156 ; CHECK1024-NEXT: str z9, [sp, #16, mul vl] // 16-byte Folded Spill
2157 ; CHECK1024-NEXT: str z8, [sp, #17, mul vl] // 16-byte Folded Spill
2158 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 1072 - 8 * VG
2159 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x49, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x70, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d9 @ cfa - 1072 - 16 * VG
2160 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4a, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x68, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d10 @ cfa - 1072 - 24 * VG
2161 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4b, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x60, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d11 @ cfa - 1072 - 32 * VG
2162 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4c, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x58, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d12 @ cfa - 1072 - 40 * VG
2163 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4d, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x50, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d13 @ cfa - 1072 - 48 * VG
2164 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4e, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x48, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d14 @ cfa - 1072 - 56 * VG
2165 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4f, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x40, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d15 @ cfa - 1072 - 64 * VG
2166 ; CHECK1024-NEXT: sub sp, sp, #1024
2167 ; CHECK1024-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xb0, 0x10, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 2096 + 144 * VG
2168 ; CHECK1024-NEXT: mov x8, x0
2169 ; CHECK1024-NEXT: //APP
2170 ; CHECK1024-NEXT: //NO_APP
2171 ; CHECK1024-NEXT: bl __arm_sme_state
2172 ; CHECK1024-NEXT: and x19, x0, #0x1
2173 ; CHECK1024-NEXT: .cfi_offset vg, -32
2174 ; CHECK1024-NEXT: tbz w19, #0, .LBB28_2
2175 ; CHECK1024-NEXT: // %bb.1: // %entry
2176 ; CHECK1024-NEXT: smstop sm
2177 ; CHECK1024-NEXT: .LBB28_2: // %entry
2178 ; CHECK1024-NEXT: mov x0, x8
2179 ; CHECK1024-NEXT: mov w1, #45 // =0x2d
2180 ; CHECK1024-NEXT: mov w2, #37 // =0x25
2181 ; CHECK1024-NEXT: bl memset
2182 ; CHECK1024-NEXT: tbz w19, #0, .LBB28_4
2183 ; CHECK1024-NEXT: // %bb.3: // %entry
2184 ; CHECK1024-NEXT: smstart sm
2185 ; CHECK1024-NEXT: .LBB28_4: // %entry
2186 ; CHECK1024-NEXT: mov w0, #22647 // =0x5877
2187 ; CHECK1024-NEXT: movk w0, #59491, lsl #16
2188 ; CHECK1024-NEXT: .cfi_restore vg
2189 ; CHECK1024-NEXT: add sp, sp, #1024
2190 ; CHECK1024-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xb0, 0x08, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 1072 + 144 * VG
2191 ; CHECK1024-NEXT: ldr z23, [sp, #2, mul vl] // 16-byte Folded Reload
2192 ; CHECK1024-NEXT: ldr z22, [sp, #3, mul vl] // 16-byte Folded Reload
2193 ; CHECK1024-NEXT: ldr z21, [sp, #4, mul vl] // 16-byte Folded Reload
2194 ; CHECK1024-NEXT: ldr z20, [sp, #5, mul vl] // 16-byte Folded Reload
2195 ; CHECK1024-NEXT: ldr z19, [sp, #6, mul vl] // 16-byte Folded Reload
2196 ; CHECK1024-NEXT: ldr z18, [sp, #7, mul vl] // 16-byte Folded Reload
2197 ; CHECK1024-NEXT: ldr z17, [sp, #8, mul vl] // 16-byte Folded Reload
2198 ; CHECK1024-NEXT: ldr z16, [sp, #9, mul vl] // 16-byte Folded Reload
2199 ; CHECK1024-NEXT: ldr z15, [sp, #10, mul vl] // 16-byte Folded Reload
2200 ; CHECK1024-NEXT: ldr z14, [sp, #11, mul vl] // 16-byte Folded Reload
2201 ; CHECK1024-NEXT: ldr z13, [sp, #12, mul vl] // 16-byte Folded Reload
2202 ; CHECK1024-NEXT: ldr z12, [sp, #13, mul vl] // 16-byte Folded Reload
2203 ; CHECK1024-NEXT: ldr z11, [sp, #14, mul vl] // 16-byte Folded Reload
2204 ; CHECK1024-NEXT: ldr z10, [sp, #15, mul vl] // 16-byte Folded Reload
2205 ; CHECK1024-NEXT: ldr z9, [sp, #16, mul vl] // 16-byte Folded Reload
2206 ; CHECK1024-NEXT: ldr z8, [sp, #17, mul vl] // 16-byte Folded Reload
2207 ; CHECK1024-NEXT: ldr p15, [sp, #4, mul vl] // 2-byte Folded Reload
2208 ; CHECK1024-NEXT: ldr p14, [sp, #5, mul vl] // 2-byte Folded Reload
2209 ; CHECK1024-NEXT: ldr p13, [sp, #6, mul vl] // 2-byte Folded Reload
2210 ; CHECK1024-NEXT: ldr p12, [sp, #7, mul vl] // 2-byte Folded Reload
2211 ; CHECK1024-NEXT: ldr p11, [sp, #8, mul vl] // 2-byte Folded Reload
2212 ; CHECK1024-NEXT: ldr p10, [sp, #9, mul vl] // 2-byte Folded Reload
2213 ; CHECK1024-NEXT: ldr p9, [sp, #10, mul vl] // 2-byte Folded Reload
2214 ; CHECK1024-NEXT: ldr p8, [sp, #11, mul vl] // 2-byte Folded Reload
2215 ; CHECK1024-NEXT: ldr p7, [sp, #12, mul vl] // 2-byte Folded Reload
2216 ; CHECK1024-NEXT: ldr p6, [sp, #13, mul vl] // 2-byte Folded Reload
2217 ; CHECK1024-NEXT: ldr p5, [sp, #14, mul vl] // 2-byte Folded Reload
2218 ; CHECK1024-NEXT: ldr p4, [sp, #15, mul vl] // 2-byte Folded Reload
2219 ; CHECK1024-NEXT: addvl sp, sp, #18
2220 ; CHECK1024-NEXT: .cfi_def_cfa wsp, 1072
2221 ; CHECK1024-NEXT: .cfi_restore z8
2222 ; CHECK1024-NEXT: .cfi_restore z9
2223 ; CHECK1024-NEXT: .cfi_restore z10
2224 ; CHECK1024-NEXT: .cfi_restore z11
2225 ; CHECK1024-NEXT: .cfi_restore z12
2226 ; CHECK1024-NEXT: .cfi_restore z13
2227 ; CHECK1024-NEXT: .cfi_restore z14
2228 ; CHECK1024-NEXT: .cfi_restore z15
2229 ; CHECK1024-NEXT: ldr x19, [sp, #1064] // 8-byte Folded Reload
2230 ; CHECK1024-NEXT: ldr x27, [sp, #1056] // 8-byte Folded Reload
2231 ; CHECK1024-NEXT: ldr x28, [sp, #1048] // 8-byte Folded Reload
2232 ; CHECK1024-NEXT: ldr x30, [sp, #1032] // 8-byte Folded Reload
2233 ; CHECK1024-NEXT: ldr x29, [sp, #1024] // 8-byte Folded Reload
2234 ; CHECK1024-NEXT: add sp, sp, #1072
2235 ; CHECK1024-NEXT: .cfi_def_cfa_offset 0
2236 ; CHECK1024-NEXT: .cfi_restore w19
2237 ; CHECK1024-NEXT: .cfi_restore w27
2238 ; CHECK1024-NEXT: .cfi_restore w28
2239 ; CHECK1024-NEXT: .cfi_restore w30
2240 ; CHECK1024-NEXT: .cfi_restore w29
2241 ; CHECK1024-NEXT: ret
2242 entry:
2243 tail call void asm sideeffect "", "~{x0},~{x28},~{x27},~{x3}"() #2
2244 %call = call ptr @memset(ptr noundef nonnull %P1, i32 noundef 45, i32 noundef 37)
2245 ret i32 -396142473
2246 }
2248 define i32 @svecc_alloca_call(<4 x i16> %P0, ptr %P1, i32 %P2, <vscale x 16 x i8> %P3, i16 %P4) "aarch64_pstate_sm_compatible" {
2249 ; CHECK0-LABEL: svecc_alloca_call:
2250 ; CHECK0: // %bb.0: // %entry
2251 ; CHECK0-NEXT: stp x29, x30, [sp, #-48]! // 16-byte Folded Spill
2252 ; CHECK0-NEXT: .cfi_def_cfa_offset 48
2253 ; CHECK0-NEXT: cntd x9
2254 ; CHECK0-NEXT: stp x9, x28, [sp, #16] // 16-byte Folded Spill
2255 ; CHECK0-NEXT: stp x27, x19, [sp, #32] // 16-byte Folded Spill
2256 ; CHECK0-NEXT: .cfi_offset w19, -8
2257 ; CHECK0-NEXT: .cfi_offset w27, -16
2258 ; CHECK0-NEXT: .cfi_offset w28, -24
2259 ; CHECK0-NEXT: .cfi_offset w30, -40
2260 ; CHECK0-NEXT: .cfi_offset w29, -48
2261 ; CHECK0-NEXT: addvl sp, sp, #-18
2262 ; CHECK0-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0x30, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 48 + 144 * VG
2263 ; CHECK0-NEXT: str p15, [sp, #4, mul vl] // 2-byte Folded Spill
2264 ; CHECK0-NEXT: str p14, [sp, #5, mul vl] // 2-byte Folded Spill
2265 ; CHECK0-NEXT: str p13, [sp, #6, mul vl] // 2-byte Folded Spill
2266 ; CHECK0-NEXT: str p12, [sp, #7, mul vl] // 2-byte Folded Spill
2267 ; CHECK0-NEXT: str p11, [sp, #8, mul vl] // 2-byte Folded Spill
2268 ; CHECK0-NEXT: str p10, [sp, #9, mul vl] // 2-byte Folded Spill
2269 ; CHECK0-NEXT: str p9, [sp, #10, mul vl] // 2-byte Folded Spill
2270 ; CHECK0-NEXT: str p8, [sp, #11, mul vl] // 2-byte Folded Spill
2271 ; CHECK0-NEXT: str p7, [sp, #12, mul vl] // 2-byte Folded Spill
2272 ; CHECK0-NEXT: str p6, [sp, #13, mul vl] // 2-byte Folded Spill
2273 ; CHECK0-NEXT: str p5, [sp, #14, mul vl] // 2-byte Folded Spill
2274 ; CHECK0-NEXT: str p4, [sp, #15, mul vl] // 2-byte Folded Spill
2275 ; CHECK0-NEXT: str z23, [sp, #2, mul vl] // 16-byte Folded Spill
2276 ; CHECK0-NEXT: str z22, [sp, #3, mul vl] // 16-byte Folded Spill
2277 ; CHECK0-NEXT: str z21, [sp, #4, mul vl] // 16-byte Folded Spill
2278 ; CHECK0-NEXT: str z20, [sp, #5, mul vl] // 16-byte Folded Spill
2279 ; CHECK0-NEXT: str z19, [sp, #6, mul vl] // 16-byte Folded Spill
2280 ; CHECK0-NEXT: str z18, [sp, #7, mul vl] // 16-byte Folded Spill
2281 ; CHECK0-NEXT: str z17, [sp, #8, mul vl] // 16-byte Folded Spill
2282 ; CHECK0-NEXT: str z16, [sp, #9, mul vl] // 16-byte Folded Spill
2283 ; CHECK0-NEXT: str z15, [sp, #10, mul vl] // 16-byte Folded Spill
2284 ; CHECK0-NEXT: str z14, [sp, #11, mul vl] // 16-byte Folded Spill
2285 ; CHECK0-NEXT: str z13, [sp, #12, mul vl] // 16-byte Folded Spill
2286 ; CHECK0-NEXT: str z12, [sp, #13, mul vl] // 16-byte Folded Spill
2287 ; CHECK0-NEXT: str z11, [sp, #14, mul vl] // 16-byte Folded Spill
2288 ; CHECK0-NEXT: str z10, [sp, #15, mul vl] // 16-byte Folded Spill
2289 ; CHECK0-NEXT: str z9, [sp, #16, mul vl] // 16-byte Folded Spill
2290 ; CHECK0-NEXT: str z8, [sp, #17, mul vl] // 16-byte Folded Spill
2291 ; CHECK0-NEXT: .cfi_escape 0x10, 0x48, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 48 - 8 * VG
2292 ; CHECK0-NEXT: .cfi_escape 0x10, 0x49, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x70, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d9 @ cfa - 48 - 16 * VG
2293 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4a, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x68, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d10 @ cfa - 48 - 24 * VG
2294 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4b, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x60, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d11 @ cfa - 48 - 32 * VG
2295 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4c, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x58, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d12 @ cfa - 48 - 40 * VG
2296 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4d, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x50, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d13 @ cfa - 48 - 48 * VG
2297 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4e, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x48, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d14 @ cfa - 48 - 56 * VG
2298 ; CHECK0-NEXT: .cfi_escape 0x10, 0x4f, 0x0a, 0x11, 0x50, 0x22, 0x11, 0x40, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d15 @ cfa - 48 - 64 * VG
2299 ; CHECK0-NEXT: sub sp, sp, #48
2300 ; CHECK0-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xe0, 0x00, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 96 + 144 * VG
2301 ; CHECK0-NEXT: //APP
2302 ; CHECK0-NEXT: //NO_APP
2303 ; CHECK0-NEXT: bl __arm_sme_state
2304 ; CHECK0-NEXT: and x19, x0, #0x1
2305 ; CHECK0-NEXT: .cfi_offset vg, -32
2306 ; CHECK0-NEXT: tbz w19, #0, .LBB29_2
2307 ; CHECK0-NEXT: // %bb.1: // %entry
2308 ; CHECK0-NEXT: smstop sm
2309 ; CHECK0-NEXT: .LBB29_2: // %entry
2310 ; CHECK0-NEXT: mov x0, sp
2311 ; CHECK0-NEXT: mov w1, #45 // =0x2d
2312 ; CHECK0-NEXT: mov w2, #37 // =0x25
2313 ; CHECK0-NEXT: bl memset
2314 ; CHECK0-NEXT: tbz w19, #0, .LBB29_4
2315 ; CHECK0-NEXT: // %bb.3: // %entry
2316 ; CHECK0-NEXT: smstart sm
2317 ; CHECK0-NEXT: .LBB29_4: // %entry
2318 ; CHECK0-NEXT: mov w0, #22647 // =0x5877
2319 ; CHECK0-NEXT: movk w0, #59491, lsl #16
2320 ; CHECK0-NEXT: .cfi_restore vg
2321 ; CHECK0-NEXT: add sp, sp, #48
2322 ; CHECK0-NEXT: .cfi_escape 0x0f, 0x0d, 0x8f, 0x00, 0x11, 0x30, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 48 + 144 * VG
2323 ; CHECK0-NEXT: ldr z23, [sp, #2, mul vl] // 16-byte Folded Reload
2324 ; CHECK0-NEXT: ldr z22, [sp, #3, mul vl] // 16-byte Folded Reload
2325 ; CHECK0-NEXT: ldr z21, [sp, #4, mul vl] // 16-byte Folded Reload
2326 ; CHECK0-NEXT: ldr z20, [sp, #5, mul vl] // 16-byte Folded Reload
2327 ; CHECK0-NEXT: ldr z19, [sp, #6, mul vl] // 16-byte Folded Reload
2328 ; CHECK0-NEXT: ldr z18, [sp, #7, mul vl] // 16-byte Folded Reload
2329 ; CHECK0-NEXT: ldr z17, [sp, #8, mul vl] // 16-byte Folded Reload
2330 ; CHECK0-NEXT: ldr z16, [sp, #9, mul vl] // 16-byte Folded Reload
2331 ; CHECK0-NEXT: ldr z15, [sp, #10, mul vl] // 16-byte Folded Reload
2332 ; CHECK0-NEXT: ldr z14, [sp, #11, mul vl] // 16-byte Folded Reload
2333 ; CHECK0-NEXT: ldr z13, [sp, #12, mul vl] // 16-byte Folded Reload
2334 ; CHECK0-NEXT: ldr z12, [sp, #13, mul vl] // 16-byte Folded Reload
2335 ; CHECK0-NEXT: ldr z11, [sp, #14, mul vl] // 16-byte Folded Reload
2336 ; CHECK0-NEXT: ldr z10, [sp, #15, mul vl] // 16-byte Folded Reload
2337 ; CHECK0-NEXT: ldr z9, [sp, #16, mul vl] // 16-byte Folded Reload
2338 ; CHECK0-NEXT: ldr z8, [sp, #17, mul vl] // 16-byte Folded Reload
2339 ; CHECK0-NEXT: ldr p15, [sp, #4, mul vl] // 2-byte Folded Reload
2340 ; CHECK0-NEXT: ldr p14, [sp, #5, mul vl] // 2-byte Folded Reload
2341 ; CHECK0-NEXT: ldr p13, [sp, #6, mul vl] // 2-byte Folded Reload
2342 ; CHECK0-NEXT: ldr p12, [sp, #7, mul vl] // 2-byte Folded Reload
2343 ; CHECK0-NEXT: ldr p11, [sp, #8, mul vl] // 2-byte Folded Reload
2344 ; CHECK0-NEXT: ldr p10, [sp, #9, mul vl] // 2-byte Folded Reload
2345 ; CHECK0-NEXT: ldr p9, [sp, #10, mul vl] // 2-byte Folded Reload
2346 ; CHECK0-NEXT: ldr p8, [sp, #11, mul vl] // 2-byte Folded Reload
2347 ; CHECK0-NEXT: ldr p7, [sp, #12, mul vl] // 2-byte Folded Reload
2348 ; CHECK0-NEXT: ldr p6, [sp, #13, mul vl] // 2-byte Folded Reload
2349 ; CHECK0-NEXT: ldr p5, [sp, #14, mul vl] // 2-byte Folded Reload
2350 ; CHECK0-NEXT: ldr p4, [sp, #15, mul vl] // 2-byte Folded Reload
2351 ; CHECK0-NEXT: addvl sp, sp, #18
2352 ; CHECK0-NEXT: .cfi_def_cfa wsp, 48
2353 ; CHECK0-NEXT: .cfi_restore z8
2354 ; CHECK0-NEXT: .cfi_restore z9
2355 ; CHECK0-NEXT: .cfi_restore z10
2356 ; CHECK0-NEXT: .cfi_restore z11
2357 ; CHECK0-NEXT: .cfi_restore z12
2358 ; CHECK0-NEXT: .cfi_restore z13
2359 ; CHECK0-NEXT: .cfi_restore z14
2360 ; CHECK0-NEXT: .cfi_restore z15
2361 ; CHECK0-NEXT: ldp x27, x19, [sp, #32] // 16-byte Folded Reload
2362 ; CHECK0-NEXT: ldr x28, [sp, #24] // 8-byte Folded Reload
2363 ; CHECK0-NEXT: ldp x29, x30, [sp], #48 // 16-byte Folded Reload
2364 ; CHECK0-NEXT: .cfi_def_cfa_offset 0
2365 ; CHECK0-NEXT: .cfi_restore w19
2366 ; CHECK0-NEXT: .cfi_restore w27
2367 ; CHECK0-NEXT: .cfi_restore w28
2368 ; CHECK0-NEXT: .cfi_restore w30
2369 ; CHECK0-NEXT: .cfi_restore w29
2370 ; CHECK0-NEXT: ret
2371 ;
2372 ; CHECK64-LABEL: svecc_alloca_call:
2373 ; CHECK64: // %bb.0: // %entry
2374 ; CHECK64-NEXT: sub sp, sp, #112
2375 ; CHECK64-NEXT: .cfi_def_cfa_offset 112
2376 ; CHECK64-NEXT: cntd x9
2377 ; CHECK64-NEXT: stp x29, x30, [sp, #64] // 16-byte Folded Spill
2378 ; CHECK64-NEXT: stp x9, x28, [sp, #80] // 16-byte Folded Spill
2379 ; CHECK64-NEXT: stp x27, x19, [sp, #96] // 16-byte Folded Spill
2380 ; CHECK64-NEXT: .cfi_offset w19, -8
2381 ; CHECK64-NEXT: .cfi_offset w27, -16
2382 ; CHECK64-NEXT: .cfi_offset w28, -24
2383 ; CHECK64-NEXT: .cfi_offset w30, -40
2384 ; CHECK64-NEXT: .cfi_offset w29, -48
2385 ; CHECK64-NEXT: addvl sp, sp, #-18
2386 ; CHECK64-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xf0, 0x00, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 112 + 144 * VG
2387 ; CHECK64-NEXT: str p15, [sp, #4, mul vl] // 2-byte Folded Spill
2388 ; CHECK64-NEXT: str p14, [sp, #5, mul vl] // 2-byte Folded Spill
2389 ; CHECK64-NEXT: str p13, [sp, #6, mul vl] // 2-byte Folded Spill
2390 ; CHECK64-NEXT: str p12, [sp, #7, mul vl] // 2-byte Folded Spill
2391 ; CHECK64-NEXT: str p11, [sp, #8, mul vl] // 2-byte Folded Spill
2392 ; CHECK64-NEXT: str p10, [sp, #9, mul vl] // 2-byte Folded Spill
2393 ; CHECK64-NEXT: str p9, [sp, #10, mul vl] // 2-byte Folded Spill
2394 ; CHECK64-NEXT: str p8, [sp, #11, mul vl] // 2-byte Folded Spill
2395 ; CHECK64-NEXT: str p7, [sp, #12, mul vl] // 2-byte Folded Spill
2396 ; CHECK64-NEXT: str p6, [sp, #13, mul vl] // 2-byte Folded Spill
2397 ; CHECK64-NEXT: str p5, [sp, #14, mul vl] // 2-byte Folded Spill
2398 ; CHECK64-NEXT: str p4, [sp, #15, mul vl] // 2-byte Folded Spill
2399 ; CHECK64-NEXT: str z23, [sp, #2, mul vl] // 16-byte Folded Spill
2400 ; CHECK64-NEXT: str z22, [sp, #3, mul vl] // 16-byte Folded Spill
2401 ; CHECK64-NEXT: str z21, [sp, #4, mul vl] // 16-byte Folded Spill
2402 ; CHECK64-NEXT: str z20, [sp, #5, mul vl] // 16-byte Folded Spill
2403 ; CHECK64-NEXT: str z19, [sp, #6, mul vl] // 16-byte Folded Spill
2404 ; CHECK64-NEXT: str z18, [sp, #7, mul vl] // 16-byte Folded Spill
2405 ; CHECK64-NEXT: str z17, [sp, #8, mul vl] // 16-byte Folded Spill
2406 ; CHECK64-NEXT: str z16, [sp, #9, mul vl] // 16-byte Folded Spill
2407 ; CHECK64-NEXT: str z15, [sp, #10, mul vl] // 16-byte Folded Spill
2408 ; CHECK64-NEXT: str z14, [sp, #11, mul vl] // 16-byte Folded Spill
2409 ; CHECK64-NEXT: str z13, [sp, #12, mul vl] // 16-byte Folded Spill
2410 ; CHECK64-NEXT: str z12, [sp, #13, mul vl] // 16-byte Folded Spill
2411 ; CHECK64-NEXT: str z11, [sp, #14, mul vl] // 16-byte Folded Spill
2412 ; CHECK64-NEXT: str z10, [sp, #15, mul vl] // 16-byte Folded Spill
2413 ; CHECK64-NEXT: str z9, [sp, #16, mul vl] // 16-byte Folded Spill
2414 ; CHECK64-NEXT: str z8, [sp, #17, mul vl] // 16-byte Folded Spill
2415 ; CHECK64-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 112 - 8 * VG
2416 ; CHECK64-NEXT: .cfi_escape 0x10, 0x49, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x70, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d9 @ cfa - 112 - 16 * VG
2417 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4a, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x68, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d10 @ cfa - 112 - 24 * VG
2418 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4b, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x60, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d11 @ cfa - 112 - 32 * VG
2419 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4c, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x58, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d12 @ cfa - 112 - 40 * VG
2420 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4d, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x50, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d13 @ cfa - 112 - 48 * VG
2421 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4e, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x48, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d14 @ cfa - 112 - 56 * VG
2422 ; CHECK64-NEXT: .cfi_escape 0x10, 0x4f, 0x0b, 0x11, 0x90, 0x7f, 0x22, 0x11, 0x40, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d15 @ cfa - 112 - 64 * VG
2423 ; CHECK64-NEXT: sub sp, sp, #112
2424 ; CHECK64-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xe0, 0x01, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 224 + 144 * VG
2425 ; CHECK64-NEXT: //APP
2426 ; CHECK64-NEXT: //NO_APP
2427 ; CHECK64-NEXT: bl __arm_sme_state
2428 ; CHECK64-NEXT: and x19, x0, #0x1
2429 ; CHECK64-NEXT: .cfi_offset vg, -32
2430 ; CHECK64-NEXT: tbz w19, #0, .LBB29_2
2431 ; CHECK64-NEXT: // %bb.1: // %entry
2432 ; CHECK64-NEXT: smstop sm
2433 ; CHECK64-NEXT: .LBB29_2: // %entry
2434 ; CHECK64-NEXT: mov x0, sp
2435 ; CHECK64-NEXT: mov w1, #45 // =0x2d
2436 ; CHECK64-NEXT: mov w2, #37 // =0x25
2437 ; CHECK64-NEXT: bl memset
2438 ; CHECK64-NEXT: tbz w19, #0, .LBB29_4
2439 ; CHECK64-NEXT: // %bb.3: // %entry
2440 ; CHECK64-NEXT: smstart sm
2441 ; CHECK64-NEXT: .LBB29_4: // %entry
2442 ; CHECK64-NEXT: mov w0, #22647 // =0x5877
2443 ; CHECK64-NEXT: movk w0, #59491, lsl #16
2444 ; CHECK64-NEXT: .cfi_restore vg
2445 ; CHECK64-NEXT: add sp, sp, #112
2446 ; CHECK64-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xf0, 0x00, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 112 + 144 * VG
2447 ; CHECK64-NEXT: ldr z23, [sp, #2, mul vl] // 16-byte Folded Reload
2448 ; CHECK64-NEXT: ldr z22, [sp, #3, mul vl] // 16-byte Folded Reload
2449 ; CHECK64-NEXT: ldr z21, [sp, #4, mul vl] // 16-byte Folded Reload
2450 ; CHECK64-NEXT: ldr z20, [sp, #5, mul vl] // 16-byte Folded Reload
2451 ; CHECK64-NEXT: ldr z19, [sp, #6, mul vl] // 16-byte Folded Reload
2452 ; CHECK64-NEXT: ldr z18, [sp, #7, mul vl] // 16-byte Folded Reload
2453 ; CHECK64-NEXT: ldr z17, [sp, #8, mul vl] // 16-byte Folded Reload
2454 ; CHECK64-NEXT: ldr z16, [sp, #9, mul vl] // 16-byte Folded Reload
2455 ; CHECK64-NEXT: ldr z15, [sp, #10, mul vl] // 16-byte Folded Reload
2456 ; CHECK64-NEXT: ldr z14, [sp, #11, mul vl] // 16-byte Folded Reload
2457 ; CHECK64-NEXT: ldr z13, [sp, #12, mul vl] // 16-byte Folded Reload
2458 ; CHECK64-NEXT: ldr z12, [sp, #13, mul vl] // 16-byte Folded Reload
2459 ; CHECK64-NEXT: ldr z11, [sp, #14, mul vl] // 16-byte Folded Reload
2460 ; CHECK64-NEXT: ldr z10, [sp, #15, mul vl] // 16-byte Folded Reload
2461 ; CHECK64-NEXT: ldr z9, [sp, #16, mul vl] // 16-byte Folded Reload
2462 ; CHECK64-NEXT: ldr z8, [sp, #17, mul vl] // 16-byte Folded Reload
2463 ; CHECK64-NEXT: ldr p15, [sp, #4, mul vl] // 2-byte Folded Reload
2464 ; CHECK64-NEXT: ldr p14, [sp, #5, mul vl] // 2-byte Folded Reload
2465 ; CHECK64-NEXT: ldr p13, [sp, #6, mul vl] // 2-byte Folded Reload
2466 ; CHECK64-NEXT: ldr p12, [sp, #7, mul vl] // 2-byte Folded Reload
2467 ; CHECK64-NEXT: ldr p11, [sp, #8, mul vl] // 2-byte Folded Reload
2468 ; CHECK64-NEXT: ldr p10, [sp, #9, mul vl] // 2-byte Folded Reload
2469 ; CHECK64-NEXT: ldr p9, [sp, #10, mul vl] // 2-byte Folded Reload
2470 ; CHECK64-NEXT: ldr p8, [sp, #11, mul vl] // 2-byte Folded Reload
2471 ; CHECK64-NEXT: ldr p7, [sp, #12, mul vl] // 2-byte Folded Reload
2472 ; CHECK64-NEXT: ldr p6, [sp, #13, mul vl] // 2-byte Folded Reload
2473 ; CHECK64-NEXT: ldr p5, [sp, #14, mul vl] // 2-byte Folded Reload
2474 ; CHECK64-NEXT: ldr p4, [sp, #15, mul vl] // 2-byte Folded Reload
2475 ; CHECK64-NEXT: addvl sp, sp, #18
2476 ; CHECK64-NEXT: .cfi_def_cfa wsp, 112
2477 ; CHECK64-NEXT: .cfi_restore z8
2478 ; CHECK64-NEXT: .cfi_restore z9
2479 ; CHECK64-NEXT: .cfi_restore z10
2480 ; CHECK64-NEXT: .cfi_restore z11
2481 ; CHECK64-NEXT: .cfi_restore z12
2482 ; CHECK64-NEXT: .cfi_restore z13
2483 ; CHECK64-NEXT: .cfi_restore z14
2484 ; CHECK64-NEXT: .cfi_restore z15
2485 ; CHECK64-NEXT: ldp x27, x19, [sp, #96] // 16-byte Folded Reload
2486 ; CHECK64-NEXT: ldr x28, [sp, #88] // 8-byte Folded Reload
2487 ; CHECK64-NEXT: ldp x29, x30, [sp, #64] // 16-byte Folded Reload
2488 ; CHECK64-NEXT: add sp, sp, #112
2489 ; CHECK64-NEXT: .cfi_def_cfa_offset 0
2490 ; CHECK64-NEXT: .cfi_restore w19
2491 ; CHECK64-NEXT: .cfi_restore w27
2492 ; CHECK64-NEXT: .cfi_restore w28
2493 ; CHECK64-NEXT: .cfi_restore w30
2494 ; CHECK64-NEXT: .cfi_restore w29
2495 ; CHECK64-NEXT: ret
2496 ;
2497 ; CHECK1024-LABEL: svecc_alloca_call:
2498 ; CHECK1024: // %bb.0: // %entry
2499 ; CHECK1024-NEXT: sub sp, sp, #1072
2500 ; CHECK1024-NEXT: .cfi_def_cfa_offset 1072
2501 ; CHECK1024-NEXT: cntd x9
2502 ; CHECK1024-NEXT: str x29, [sp, #1024] // 8-byte Folded Spill
2503 ; CHECK1024-NEXT: str x30, [sp, #1032] // 8-byte Folded Spill
2504 ; CHECK1024-NEXT: str x9, [sp, #1040] // 8-byte Folded Spill
2505 ; CHECK1024-NEXT: str x28, [sp, #1048] // 8-byte Folded Spill
2506 ; CHECK1024-NEXT: str x27, [sp, #1056] // 8-byte Folded Spill
2507 ; CHECK1024-NEXT: str x19, [sp, #1064] // 8-byte Folded Spill
2508 ; CHECK1024-NEXT: .cfi_offset w19, -8
2509 ; CHECK1024-NEXT: .cfi_offset w27, -16
2510 ; CHECK1024-NEXT: .cfi_offset w28, -24
2511 ; CHECK1024-NEXT: .cfi_offset w30, -40
2512 ; CHECK1024-NEXT: .cfi_offset w29, -48
2513 ; CHECK1024-NEXT: addvl sp, sp, #-18
2514 ; CHECK1024-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xb0, 0x08, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 1072 + 144 * VG
2515 ; CHECK1024-NEXT: str p15, [sp, #4, mul vl] // 2-byte Folded Spill
2516 ; CHECK1024-NEXT: str p14, [sp, #5, mul vl] // 2-byte Folded Spill
2517 ; CHECK1024-NEXT: str p13, [sp, #6, mul vl] // 2-byte Folded Spill
2518 ; CHECK1024-NEXT: str p12, [sp, #7, mul vl] // 2-byte Folded Spill
2519 ; CHECK1024-NEXT: str p11, [sp, #8, mul vl] // 2-byte Folded Spill
2520 ; CHECK1024-NEXT: str p10, [sp, #9, mul vl] // 2-byte Folded Spill
2521 ; CHECK1024-NEXT: str p9, [sp, #10, mul vl] // 2-byte Folded Spill
2522 ; CHECK1024-NEXT: str p8, [sp, #11, mul vl] // 2-byte Folded Spill
2523 ; CHECK1024-NEXT: str p7, [sp, #12, mul vl] // 2-byte Folded Spill
2524 ; CHECK1024-NEXT: str p6, [sp, #13, mul vl] // 2-byte Folded Spill
2525 ; CHECK1024-NEXT: str p5, [sp, #14, mul vl] // 2-byte Folded Spill
2526 ; CHECK1024-NEXT: str p4, [sp, #15, mul vl] // 2-byte Folded Spill
2527 ; CHECK1024-NEXT: str z23, [sp, #2, mul vl] // 16-byte Folded Spill
2528 ; CHECK1024-NEXT: str z22, [sp, #3, mul vl] // 16-byte Folded Spill
2529 ; CHECK1024-NEXT: str z21, [sp, #4, mul vl] // 16-byte Folded Spill
2530 ; CHECK1024-NEXT: str z20, [sp, #5, mul vl] // 16-byte Folded Spill
2531 ; CHECK1024-NEXT: str z19, [sp, #6, mul vl] // 16-byte Folded Spill
2532 ; CHECK1024-NEXT: str z18, [sp, #7, mul vl] // 16-byte Folded Spill
2533 ; CHECK1024-NEXT: str z17, [sp, #8, mul vl] // 16-byte Folded Spill
2534 ; CHECK1024-NEXT: str z16, [sp, #9, mul vl] // 16-byte Folded Spill
2535 ; CHECK1024-NEXT: str z15, [sp, #10, mul vl] // 16-byte Folded Spill
2536 ; CHECK1024-NEXT: str z14, [sp, #11, mul vl] // 16-byte Folded Spill
2537 ; CHECK1024-NEXT: str z13, [sp, #12, mul vl] // 16-byte Folded Spill
2538 ; CHECK1024-NEXT: str z12, [sp, #13, mul vl] // 16-byte Folded Spill
2539 ; CHECK1024-NEXT: str z11, [sp, #14, mul vl] // 16-byte Folded Spill
2540 ; CHECK1024-NEXT: str z10, [sp, #15, mul vl] // 16-byte Folded Spill
2541 ; CHECK1024-NEXT: str z9, [sp, #16, mul vl] // 16-byte Folded Spill
2542 ; CHECK1024-NEXT: str z8, [sp, #17, mul vl] // 16-byte Folded Spill
2543 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x48, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x78, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d8 @ cfa - 1072 - 8 * VG
2544 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x49, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x70, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d9 @ cfa - 1072 - 16 * VG
2545 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4a, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x68, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d10 @ cfa - 1072 - 24 * VG
2546 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4b, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x60, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d11 @ cfa - 1072 - 32 * VG
2547 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4c, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x58, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d12 @ cfa - 1072 - 40 * VG
2548 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4d, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x50, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d13 @ cfa - 1072 - 48 * VG
2549 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4e, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x48, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d14 @ cfa - 1072 - 56 * VG
2550 ; CHECK1024-NEXT: .cfi_escape 0x10, 0x4f, 0x0b, 0x11, 0xd0, 0x77, 0x22, 0x11, 0x40, 0x92, 0x2e, 0x00, 0x1e, 0x22 // $d15 @ cfa - 1072 - 64 * VG
2551 ; CHECK1024-NEXT: sub sp, sp, #1072
2552 ; CHECK1024-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xe0, 0x10, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 2144 + 144 * VG
2553 ; CHECK1024-NEXT: //APP
2554 ; CHECK1024-NEXT: //NO_APP
2555 ; CHECK1024-NEXT: bl __arm_sme_state
2556 ; CHECK1024-NEXT: and x19, x0, #0x1
2557 ; CHECK1024-NEXT: .cfi_offset vg, -32
2558 ; CHECK1024-NEXT: tbz w19, #0, .LBB29_2
2559 ; CHECK1024-NEXT: // %bb.1: // %entry
2560 ; CHECK1024-NEXT: smstop sm
2561 ; CHECK1024-NEXT: .LBB29_2: // %entry
2562 ; CHECK1024-NEXT: mov x0, sp
2563 ; CHECK1024-NEXT: mov w1, #45 // =0x2d
2564 ; CHECK1024-NEXT: mov w2, #37 // =0x25
2565 ; CHECK1024-NEXT: bl memset
2566 ; CHECK1024-NEXT: tbz w19, #0, .LBB29_4
2567 ; CHECK1024-NEXT: // %bb.3: // %entry
2568 ; CHECK1024-NEXT: smstart sm
2569 ; CHECK1024-NEXT: .LBB29_4: // %entry
2570 ; CHECK1024-NEXT: mov w0, #22647 // =0x5877
2571 ; CHECK1024-NEXT: movk w0, #59491, lsl #16
2572 ; CHECK1024-NEXT: .cfi_restore vg
2573 ; CHECK1024-NEXT: add sp, sp, #1072
2574 ; CHECK1024-NEXT: .cfi_escape 0x0f, 0x0e, 0x8f, 0x00, 0x11, 0xb0, 0x08, 0x22, 0x11, 0x90, 0x01, 0x92, 0x2e, 0x00, 0x1e, 0x22 // sp + 1072 + 144 * VG
2575 ; CHECK1024-NEXT: ldr z23, [sp, #2, mul vl] // 16-byte Folded Reload
2576 ; CHECK1024-NEXT: ldr z22, [sp, #3, mul vl] // 16-byte Folded Reload
2577 ; CHECK1024-NEXT: ldr z21, [sp, #4, mul vl] // 16-byte Folded Reload
2578 ; CHECK1024-NEXT: ldr z20, [sp, #5, mul vl] // 16-byte Folded Reload
2579 ; CHECK1024-NEXT: ldr z19, [sp, #6, mul vl] // 16-byte Folded Reload
2580 ; CHECK1024-NEXT: ldr z18, [sp, #7, mul vl] // 16-byte Folded Reload
2581 ; CHECK1024-NEXT: ldr z17, [sp, #8, mul vl] // 16-byte Folded Reload
2582 ; CHECK1024-NEXT: ldr z16, [sp, #9, mul vl] // 16-byte Folded Reload
2583 ; CHECK1024-NEXT: ldr z15, [sp, #10, mul vl] // 16-byte Folded Reload
2584 ; CHECK1024-NEXT: ldr z14, [sp, #11, mul vl] // 16-byte Folded Reload
2585 ; CHECK1024-NEXT: ldr z13, [sp, #12, mul vl] // 16-byte Folded Reload
2586 ; CHECK1024-NEXT: ldr z12, [sp, #13, mul vl] // 16-byte Folded Reload
2587 ; CHECK1024-NEXT: ldr z11, [sp, #14, mul vl] // 16-byte Folded Reload
2588 ; CHECK1024-NEXT: ldr z10, [sp, #15, mul vl] // 16-byte Folded Reload
2589 ; CHECK1024-NEXT: ldr z9, [sp, #16, mul vl] // 16-byte Folded Reload
2590 ; CHECK1024-NEXT: ldr z8, [sp, #17, mul vl] // 16-byte Folded Reload
2591 ; CHECK1024-NEXT: ldr p15, [sp, #4, mul vl] // 2-byte Folded Reload
2592 ; CHECK1024-NEXT: ldr p14, [sp, #5, mul vl] // 2-byte Folded Reload
2593 ; CHECK1024-NEXT: ldr p13, [sp, #6, mul vl] // 2-byte Folded Reload
2594 ; CHECK1024-NEXT: ldr p12, [sp, #7, mul vl] // 2-byte Folded Reload
2595 ; CHECK1024-NEXT: ldr p11, [sp, #8, mul vl] // 2-byte Folded Reload
2596 ; CHECK1024-NEXT: ldr p10, [sp, #9, mul vl] // 2-byte Folded Reload
2597 ; CHECK1024-NEXT: ldr p9, [sp, #10, mul vl] // 2-byte Folded Reload
2598 ; CHECK1024-NEXT: ldr p8, [sp, #11, mul vl] // 2-byte Folded Reload
2599 ; CHECK1024-NEXT: ldr p7, [sp, #12, mul vl] // 2-byte Folded Reload
2600 ; CHECK1024-NEXT: ldr p6, [sp, #13, mul vl] // 2-byte Folded Reload
2601 ; CHECK1024-NEXT: ldr p5, [sp, #14, mul vl] // 2-byte Folded Reload
2602 ; CHECK1024-NEXT: ldr p4, [sp, #15, mul vl] // 2-byte Folded Reload
2603 ; CHECK1024-NEXT: addvl sp, sp, #18
2604 ; CHECK1024-NEXT: .cfi_def_cfa wsp, 1072
2605 ; CHECK1024-NEXT: .cfi_restore z8
2606 ; CHECK1024-NEXT: .cfi_restore z9
2607 ; CHECK1024-NEXT: .cfi_restore z10
2608 ; CHECK1024-NEXT: .cfi_restore z11
2609 ; CHECK1024-NEXT: .cfi_restore z12
2610 ; CHECK1024-NEXT: .cfi_restore z13
2611 ; CHECK1024-NEXT: .cfi_restore z14
2612 ; CHECK1024-NEXT: .cfi_restore z15
2613 ; CHECK1024-NEXT: ldr x19, [sp, #1064] // 8-byte Folded Reload
2614 ; CHECK1024-NEXT: ldr x27, [sp, #1056] // 8-byte Folded Reload
2615 ; CHECK1024-NEXT: ldr x28, [sp, #1048] // 8-byte Folded Reload
2616 ; CHECK1024-NEXT: ldr x30, [sp, #1032] // 8-byte Folded Reload
2617 ; CHECK1024-NEXT: ldr x29, [sp, #1024] // 8-byte Folded Reload
2618 ; CHECK1024-NEXT: add sp, sp, #1072
2619 ; CHECK1024-NEXT: .cfi_def_cfa_offset 0
2620 ; CHECK1024-NEXT: .cfi_restore w19
2621 ; CHECK1024-NEXT: .cfi_restore w27
2622 ; CHECK1024-NEXT: .cfi_restore w28
2623 ; CHECK1024-NEXT: .cfi_restore w30
2624 ; CHECK1024-NEXT: .cfi_restore w29
2625 ; CHECK1024-NEXT: ret
2626 entry:
2627 tail call void asm sideeffect "", "~{x0},~{x28},~{x27},~{x3}"() #2
2628 %0 = alloca [37 x i8], align 16
2629 %call = call ptr @memset(ptr noundef nonnull %0, i32 noundef 45, i32 noundef 37)
2630 ret i32 -396142473
2631 }
2632 declare ptr @memset(ptr, i32, i32)
2634 define void @call_with_doubles() "aarch64_pstate_sm_compatible" {
2635 ; CHECK0-LABEL: call_with_doubles:
2636 ; CHECK0: // %bb.0: // %entry
2637 ; CHECK0-NEXT: str d8, [sp, #-16]! // 8-byte Folded Spill
2638 ; CHECK0-NEXT: str x30, [sp, #8] // 8-byte Folded Spill
2639 ; CHECK0-NEXT: .cfi_def_cfa_offset 16
2640 ; CHECK0-NEXT: .cfi_offset w30, -8
2641 ; CHECK0-NEXT: .cfi_offset b8, -16
2642 ; CHECK0-NEXT: mov x8, #9221120237041090560 // =0x7ff8000000000000
2643 ; CHECK0-NEXT: fmov d8, x8
2644 ; CHECK0-NEXT: fmov d0, d8
2645 ; CHECK0-NEXT: bl calld
2646 ; CHECK0-NEXT: ldr x30, [sp, #8] // 8-byte Folded Reload
2647 ; CHECK0-NEXT: fmov d0, d8
2648 ; CHECK0-NEXT: ldr d8, [sp], #16 // 8-byte Folded Reload
2649 ; CHECK0-NEXT: b calld
2650 ;
2651 ; CHECK64-LABEL: call_with_doubles:
2652 ; CHECK64: // %bb.0: // %entry
2653 ; CHECK64-NEXT: sub sp, sp, #144
2654 ; CHECK64-NEXT: str d8, [sp, #64] // 8-byte Folded Spill
2655 ; CHECK64-NEXT: str x30, [sp, #136] // 8-byte Folded Spill
2656 ; CHECK64-NEXT: .cfi_def_cfa_offset 144
2657 ; CHECK64-NEXT: .cfi_offset w30, -8
2658 ; CHECK64-NEXT: .cfi_offset b8, -80
2659 ; CHECK64-NEXT: mov x8, #9221120237041090560 // =0x7ff8000000000000
2660 ; CHECK64-NEXT: fmov d8, x8
2661 ; CHECK64-NEXT: fmov d0, d8
2662 ; CHECK64-NEXT: bl calld
2663 ; CHECK64-NEXT: fmov d0, d8
2664 ; CHECK64-NEXT: ldr x30, [sp, #136] // 8-byte Folded Reload
2665 ; CHECK64-NEXT: ldr d8, [sp, #64] // 8-byte Folded Reload
2666 ; CHECK64-NEXT: add sp, sp, #144
2667 ; CHECK64-NEXT: b calld
2668 ;
2669 ; CHECK1024-LABEL: call_with_doubles:
2670 ; CHECK1024: // %bb.0: // %entry
2671 ; CHECK1024-NEXT: sub sp, sp, #1056
2672 ; CHECK1024-NEXT: str d8, [sp] // 8-byte Folded Spill
2673 ; CHECK1024-NEXT: str x29, [sp, #1032] // 8-byte Folded Spill
2674 ; CHECK1024-NEXT: str x30, [sp, #1040] // 8-byte Folded Spill
2675 ; CHECK1024-NEXT: sub sp, sp, #1024
2676 ; CHECK1024-NEXT: .cfi_def_cfa_offset 2080
2677 ; CHECK1024-NEXT: .cfi_offset w30, -16
2678 ; CHECK1024-NEXT: .cfi_offset w29, -24
2679 ; CHECK1024-NEXT: .cfi_offset b8, -1056
2680 ; CHECK1024-NEXT: mov x8, #9221120237041090560 // =0x7ff8000000000000
2681 ; CHECK1024-NEXT: fmov d8, x8
2682 ; CHECK1024-NEXT: fmov d0, d8
2683 ; CHECK1024-NEXT: bl calld
2684 ; CHECK1024-NEXT: fmov d0, d8
2685 ; CHECK1024-NEXT: add sp, sp, #1024
2686 ; CHECK1024-NEXT: ldr x30, [sp, #1040] // 8-byte Folded Reload
2687 ; CHECK1024-NEXT: ldr x29, [sp, #1032] // 8-byte Folded Reload
2688 ; CHECK1024-NEXT: ldr d8, [sp] // 8-byte Folded Reload
2689 ; CHECK1024-NEXT: add sp, sp, #1056
2690 ; CHECK1024-NEXT: b calld
2691 entry:
2692 %call = tail call i32 @calld(double 0x7FF8000000000000)
2693 %call.1 = tail call i32 @calld(double 0x7FF8000000000000)
2694 ret void
2695 }
2696 declare i32 @calld(double) "aarch64_pstate_sm_compatible"
2698 ; Check that stack objects are ordererd fpr > hazard > gpr
2699 define void @ordering_test(double %d, half %h, <4 x i32> %v) "aarch64_pstate_sm_compatible" {
2700 ; CHECK0-LABEL: ordering_test:
2701 ; CHECK0: // %bb.0: // %entry
2702 ; CHECK0-NEXT: sub sp, sp, #48
2703 ; CHECK0-NEXT: .cfi_def_cfa_offset 48
2704 ; CHECK0-NEXT: str wzr, [sp, #32]
2705 ; CHECK0-NEXT: str d0, [sp, #24]
2706 ; CHECK0-NEXT: str wzr, [sp, #44]
2707 ; CHECK0-NEXT: str h1, [sp, #22]
2708 ; CHECK0-NEXT: str wzr, [sp, #16]
2709 ; CHECK0-NEXT: str q2, [sp], #48
2710 ; CHECK0-NEXT: ret
2711 ;
2712 ; CHECK64-LABEL: ordering_test:
2713 ; CHECK64: // %bb.0: // %entry
2714 ; CHECK64-NEXT: sub sp, sp, #128
2715 ; CHECK64-NEXT: .cfi_def_cfa_offset 128
2716 ; CHECK64-NEXT: stp wzr, wzr, [sp, #12]
2717 ; CHECK64-NEXT: str d0, [sp, #120]
2718 ; CHECK64-NEXT: str wzr, [sp, #28]
2719 ; CHECK64-NEXT: str h1, [sp, #118]
2720 ; CHECK64-NEXT: str q2, [sp, #96]
2721 ; CHECK64-NEXT: add sp, sp, #128
2722 ; CHECK64-NEXT: ret
2723 ;
2724 ; CHECK1024-LABEL: ordering_test:
2725 ; CHECK1024: // %bb.0: // %entry
2726 ; CHECK1024-NEXT: sub sp, sp, #1040
2727 ; CHECK1024-NEXT: str x29, [sp, #1024] // 8-byte Folded Spill
2728 ; CHECK1024-NEXT: sub sp, sp, #1088
2729 ; CHECK1024-NEXT: .cfi_def_cfa_offset 2128
2730 ; CHECK1024-NEXT: .cfi_offset w29, -16
2731 ; CHECK1024-NEXT: stp wzr, wzr, [sp, #12]
2732 ; CHECK1024-NEXT: str d0, [sp, #1080]
2733 ; CHECK1024-NEXT: str wzr, [sp, #28]
2734 ; CHECK1024-NEXT: str h1, [sp, #1078]
2735 ; CHECK1024-NEXT: str q2, [sp, #1056]
2736 ; CHECK1024-NEXT: add sp, sp, #1088
2737 ; CHECK1024-NEXT: ldr x29, [sp, #1024] // 8-byte Folded Reload
2738 ; CHECK1024-NEXT: add sp, sp, #1040
2739 ; CHECK1024-NEXT: ret
2740 entry:
2741 %i32 = alloca i32
2742 %i64 = alloca i64
2743 %f64 = alloca double
2744 %f16 = alloca half
2745 %i32b = alloca i32
2746 %v4i32 = alloca <4 x i32>
2747 store i32 0, ptr %i64
2748 store double %d, ptr %f64
2749 store i32 0, ptr %i32
2750 store half %h, ptr %f16
2751 store i32 0, ptr %i32b
2752 store <4 x i32> %v, ptr %v4i32
2753 ret void
2754 }
2757 define void @ordering_test_array(i64 %o, i64 %p, float %f, i32 %x) "aarch64_pstate_sm_compatible" {
2758 ; CHECK0-LABEL: ordering_test_array:
2759 ; CHECK0: // %bb.0: // %entry
2760 ; CHECK0-NEXT: sub sp, sp, #272
2761 ; CHECK0-NEXT: str x29, [sp, #256] // 8-byte Folded Spill
2762 ; CHECK0-NEXT: .cfi_def_cfa_offset 272
2763 ; CHECK0-NEXT: .cfi_offset w29, -16
2764 ; CHECK0-NEXT: add x8, sp, #128
2765 ; CHECK0-NEXT: str w2, [x8, x0, lsl #2]
2766 ; CHECK0-NEXT: mov x8, sp
2767 ; CHECK0-NEXT: str s0, [x8, x1, lsl #2]
2768 ; CHECK0-NEXT: add sp, sp, #272
2769 ; CHECK0-NEXT: ret
2770 ;
2771 ; CHECK64-LABEL: ordering_test_array:
2772 ; CHECK64: // %bb.0: // %entry
2773 ; CHECK64-NEXT: sub sp, sp, #400
2774 ; CHECK64-NEXT: str x29, [sp, #384] // 8-byte Folded Spill
2775 ; CHECK64-NEXT: .cfi_def_cfa_offset 400
2776 ; CHECK64-NEXT: .cfi_offset w29, -16
2777 ; CHECK64-NEXT: mov x8, sp
2778 ; CHECK64-NEXT: str w2, [x8, x0, lsl #2]
2779 ; CHECK64-NEXT: add x8, sp, #192
2780 ; CHECK64-NEXT: str s0, [x8, x1, lsl #2]
2781 ; CHECK64-NEXT: add sp, sp, #400
2782 ; CHECK64-NEXT: ret
2783 ;
2784 ; CHECK1024-LABEL: ordering_test_array:
2785 ; CHECK1024: // %bb.0: // %entry
2786 ; CHECK1024-NEXT: sub sp, sp, #1040
2787 ; CHECK1024-NEXT: str x29, [sp, #1024] // 8-byte Folded Spill
2788 ; CHECK1024-NEXT: sub sp, sp, #1280
2789 ; CHECK1024-NEXT: .cfi_def_cfa_offset 2320
2790 ; CHECK1024-NEXT: .cfi_offset w29, -16
2791 ; CHECK1024-NEXT: mov x8, sp
2792 ; CHECK1024-NEXT: str w2, [x8, x0, lsl #2]
2793 ; CHECK1024-NEXT: add x8, sp, #1152
2794 ; CHECK1024-NEXT: str s0, [x8, x1, lsl #2]
2795 ; CHECK1024-NEXT: add sp, sp, #1280
2796 ; CHECK1024-NEXT: ldr x29, [sp, #1024] // 8-byte Folded Reload
2797 ; CHECK1024-NEXT: add sp, sp, #1040
2798 ; CHECK1024-NEXT: ret
2799 entry:
2800 %i32 = alloca [32 x i32]
2801 %f32 = alloca [32 x float]
2802 %g = getelementptr i32, ptr %i32, i64 %o
2803 store i32 %x, ptr %g
2804 %h = getelementptr float, ptr %f32, i64 %p
2805 store float %f, ptr %h
2806 ret void
2807 }
2809 ; The VA register currently ends up in VLA space. Lets hope that doesn't come up very often.
2810 define i32 @vastate(i32 %x) "aarch64_inout_za" "aarch64_pstate_sm_enabled" "target-features"="+sme" {
2811 ; CHECK0-LABEL: vastate:
2812 ; CHECK0: // %bb.0: // %entry
2813 ; CHECK0-NEXT: stp d15, d14, [sp, #-112]! // 16-byte Folded Spill
2814 ; CHECK0-NEXT: .cfi_def_cfa_offset 112
2815 ; CHECK0-NEXT: cntd x9
2816 ; CHECK0-NEXT: stp d13, d12, [sp, #16] // 16-byte Folded Spill
2817 ; CHECK0-NEXT: stp d11, d10, [sp, #32] // 16-byte Folded Spill
2818 ; CHECK0-NEXT: stp d9, d8, [sp, #48] // 16-byte Folded Spill
2819 ; CHECK0-NEXT: stp x29, x30, [sp, #64] // 16-byte Folded Spill
2820 ; CHECK0-NEXT: str x9, [sp, #80] // 8-byte Folded Spill
2821 ; CHECK0-NEXT: stp x20, x19, [sp, #96] // 16-byte Folded Spill
2822 ; CHECK0-NEXT: add x29, sp, #64
2823 ; CHECK0-NEXT: .cfi_def_cfa w29, 48
2824 ; CHECK0-NEXT: .cfi_offset w19, -8
2825 ; CHECK0-NEXT: .cfi_offset w20, -16
2826 ; CHECK0-NEXT: .cfi_offset w30, -40
2827 ; CHECK0-NEXT: .cfi_offset w29, -48
2828 ; CHECK0-NEXT: .cfi_offset b8, -56
2829 ; CHECK0-NEXT: .cfi_offset b9, -64
2830 ; CHECK0-NEXT: .cfi_offset b10, -72
2831 ; CHECK0-NEXT: .cfi_offset b11, -80
2832 ; CHECK0-NEXT: .cfi_offset b12, -88
2833 ; CHECK0-NEXT: .cfi_offset b13, -96
2834 ; CHECK0-NEXT: .cfi_offset b14, -104
2835 ; CHECK0-NEXT: .cfi_offset b15, -112
2836 ; CHECK0-NEXT: sub sp, sp, #16
2837 ; CHECK0-NEXT: rdsvl x8, #1
2838 ; CHECK0-NEXT: mov x9, sp
2839 ; CHECK0-NEXT: mov w20, w0
2840 ; CHECK0-NEXT: msub x9, x8, x8, x9
2841 ; CHECK0-NEXT: mov sp, x9
2842 ; CHECK0-NEXT: stur x9, [x29, #-80]
2843 ; CHECK0-NEXT: sub x9, x29, #80
2844 ; CHECK0-NEXT: sturh wzr, [x29, #-70]
2845 ; CHECK0-NEXT: stur wzr, [x29, #-68]
2846 ; CHECK0-NEXT: sturh w8, [x29, #-72]
2847 ; CHECK0-NEXT: msr TPIDR2_EL0, x9
2848 ; CHECK0-NEXT: .cfi_offset vg, -32
2849 ; CHECK0-NEXT: smstop sm
2850 ; CHECK0-NEXT: bl other
2851 ; CHECK0-NEXT: smstart sm
2852 ; CHECK0-NEXT: .cfi_restore vg
2853 ; CHECK0-NEXT: smstart za
2854 ; CHECK0-NEXT: mrs x8, TPIDR2_EL0
2855 ; CHECK0-NEXT: sub x0, x29, #80
2856 ; CHECK0-NEXT: cbnz x8, .LBB33_2
2857 ; CHECK0-NEXT: // %bb.1: // %entry
2858 ; CHECK0-NEXT: bl __arm_tpidr2_restore
2859 ; CHECK0-NEXT: .LBB33_2: // %entry
2860 ; CHECK0-NEXT: mov w0, w20
2861 ; CHECK0-NEXT: msr TPIDR2_EL0, xzr
2862 ; CHECK0-NEXT: sub sp, x29, #64
2863 ; CHECK0-NEXT: .cfi_def_cfa wsp, 112
2864 ; CHECK0-NEXT: ldp x20, x19, [sp, #96] // 16-byte Folded Reload
2865 ; CHECK0-NEXT: ldp x29, x30, [sp, #64] // 16-byte Folded Reload
2866 ; CHECK0-NEXT: ldp d9, d8, [sp, #48] // 16-byte Folded Reload
2867 ; CHECK0-NEXT: ldp d11, d10, [sp, #32] // 16-byte Folded Reload
2868 ; CHECK0-NEXT: ldp d13, d12, [sp, #16] // 16-byte Folded Reload
2869 ; CHECK0-NEXT: ldp d15, d14, [sp], #112 // 16-byte Folded Reload
2870 ; CHECK0-NEXT: .cfi_def_cfa_offset 0
2871 ; CHECK0-NEXT: .cfi_restore w19
2872 ; CHECK0-NEXT: .cfi_restore w20
2873 ; CHECK0-NEXT: .cfi_restore w30
2874 ; CHECK0-NEXT: .cfi_restore w29
2875 ; CHECK0-NEXT: .cfi_restore b8
2876 ; CHECK0-NEXT: .cfi_restore b9
2877 ; CHECK0-NEXT: .cfi_restore b10
2878 ; CHECK0-NEXT: .cfi_restore b11
2879 ; CHECK0-NEXT: .cfi_restore b12
2880 ; CHECK0-NEXT: .cfi_restore b13
2881 ; CHECK0-NEXT: .cfi_restore b14
2882 ; CHECK0-NEXT: .cfi_restore b15
2883 ; CHECK0-NEXT: ret
2884 ;
2885 ; CHECK64-LABEL: vastate:
2886 ; CHECK64: // %bb.0: // %entry
2887 ; CHECK64-NEXT: stp d15, d14, [sp, #-176]! // 16-byte Folded Spill
2888 ; CHECK64-NEXT: .cfi_def_cfa_offset 176
2889 ; CHECK64-NEXT: cntd x9
2890 ; CHECK64-NEXT: stp d13, d12, [sp, #16] // 16-byte Folded Spill
2891 ; CHECK64-NEXT: stp d11, d10, [sp, #32] // 16-byte Folded Spill
2892 ; CHECK64-NEXT: stp d9, d8, [sp, #48] // 16-byte Folded Spill
2893 ; CHECK64-NEXT: stp x29, x30, [sp, #128] // 16-byte Folded Spill
2894 ; CHECK64-NEXT: stp x9, x20, [sp, #144] // 16-byte Folded Spill
2895 ; CHECK64-NEXT: str x19, [sp, #160] // 8-byte Folded Spill
2896 ; CHECK64-NEXT: mov x29, sp
2897 ; CHECK64-NEXT: .cfi_def_cfa w29, 176
2898 ; CHECK64-NEXT: .cfi_offset w19, -16
2899 ; CHECK64-NEXT: .cfi_offset w20, -24
2900 ; CHECK64-NEXT: .cfi_offset w30, -40
2901 ; CHECK64-NEXT: .cfi_offset w29, -48
2902 ; CHECK64-NEXT: .cfi_offset b8, -120
2903 ; CHECK64-NEXT: .cfi_offset b9, -128
2904 ; CHECK64-NEXT: .cfi_offset b10, -136
2905 ; CHECK64-NEXT: .cfi_offset b11, -144
2906 ; CHECK64-NEXT: .cfi_offset b12, -152
2907 ; CHECK64-NEXT: .cfi_offset b13, -160
2908 ; CHECK64-NEXT: .cfi_offset b14, -168
2909 ; CHECK64-NEXT: .cfi_offset b15, -176
2910 ; CHECK64-NEXT: sub sp, sp, #80
2911 ; CHECK64-NEXT: rdsvl x8, #1
2912 ; CHECK64-NEXT: mov x9, sp
2913 ; CHECK64-NEXT: mov w20, w0
2914 ; CHECK64-NEXT: msub x9, x8, x8, x9
2915 ; CHECK64-NEXT: mov sp, x9
2916 ; CHECK64-NEXT: stur x9, [x29, #-80]
2917 ; CHECK64-NEXT: sub x9, x29, #80
2918 ; CHECK64-NEXT: sturh wzr, [x29, #-70]
2919 ; CHECK64-NEXT: stur wzr, [x29, #-68]
2920 ; CHECK64-NEXT: sturh w8, [x29, #-72]
2921 ; CHECK64-NEXT: msr TPIDR2_EL0, x9
2922 ; CHECK64-NEXT: .cfi_offset vg, -32
2923 ; CHECK64-NEXT: smstop sm
2924 ; CHECK64-NEXT: bl other
2925 ; CHECK64-NEXT: smstart sm
2926 ; CHECK64-NEXT: .cfi_restore vg
2927 ; CHECK64-NEXT: smstart za
2928 ; CHECK64-NEXT: mrs x8, TPIDR2_EL0
2929 ; CHECK64-NEXT: sub x0, x29, #80
2930 ; CHECK64-NEXT: cbnz x8, .LBB33_2
2931 ; CHECK64-NEXT: // %bb.1: // %entry
2932 ; CHECK64-NEXT: bl __arm_tpidr2_restore
2933 ; CHECK64-NEXT: .LBB33_2: // %entry
2934 ; CHECK64-NEXT: mov w0, w20
2935 ; CHECK64-NEXT: msr TPIDR2_EL0, xzr
2936 ; CHECK64-NEXT: mov sp, x29
2937 ; CHECK64-NEXT: .cfi_def_cfa wsp, 176
2938 ; CHECK64-NEXT: ldp x20, x19, [sp, #152] // 16-byte Folded Reload
2939 ; CHECK64-NEXT: ldr d14, [sp, #8] // 8-byte Folded Reload
2940 ; CHECK64-NEXT: ldp x29, x30, [sp, #128] // 16-byte Folded Reload
2941 ; CHECK64-NEXT: ldp d9, d8, [sp, #48] // 16-byte Folded Reload
2942 ; CHECK64-NEXT: ldp d11, d10, [sp, #32] // 16-byte Folded Reload
2943 ; CHECK64-NEXT: ldp d13, d12, [sp, #16] // 16-byte Folded Reload
2944 ; CHECK64-NEXT: ldr d15, [sp], #176 // 8-byte Folded Reload
2945 ; CHECK64-NEXT: .cfi_def_cfa_offset 0
2946 ; CHECK64-NEXT: .cfi_restore w19
2947 ; CHECK64-NEXT: .cfi_restore w20
2948 ; CHECK64-NEXT: .cfi_restore w30
2949 ; CHECK64-NEXT: .cfi_restore w29
2950 ; CHECK64-NEXT: .cfi_restore b8
2951 ; CHECK64-NEXT: .cfi_restore b9
2952 ; CHECK64-NEXT: .cfi_restore b10
2953 ; CHECK64-NEXT: .cfi_restore b11
2954 ; CHECK64-NEXT: .cfi_restore b12
2955 ; CHECK64-NEXT: .cfi_restore b13
2956 ; CHECK64-NEXT: .cfi_restore b14
2957 ; CHECK64-NEXT: .cfi_restore b15
2958 ; CHECK64-NEXT: ret
2959 ;
2960 ; CHECK1024-LABEL: vastate:
2961 ; CHECK1024: // %bb.0: // %entry
2962 ; CHECK1024-NEXT: sub sp, sp, #1136
2963 ; CHECK1024-NEXT: .cfi_def_cfa_offset 1136
2964 ; CHECK1024-NEXT: cntd x9
2965 ; CHECK1024-NEXT: stp d15, d14, [sp] // 16-byte Folded Spill
2966 ; CHECK1024-NEXT: stp d13, d12, [sp, #16] // 16-byte Folded Spill
2967 ; CHECK1024-NEXT: stp d11, d10, [sp, #32] // 16-byte Folded Spill
2968 ; CHECK1024-NEXT: stp d9, d8, [sp, #48] // 16-byte Folded Spill
2969 ; CHECK1024-NEXT: str x29, [sp, #1088] // 8-byte Folded Spill
2970 ; CHECK1024-NEXT: str x30, [sp, #1096] // 8-byte Folded Spill
2971 ; CHECK1024-NEXT: str x9, [sp, #1104] // 8-byte Folded Spill
2972 ; CHECK1024-NEXT: str x28, [sp, #1112] // 8-byte Folded Spill
2973 ; CHECK1024-NEXT: str x20, [sp, #1120] // 8-byte Folded Spill
2974 ; CHECK1024-NEXT: str x19, [sp, #1128] // 8-byte Folded Spill
2975 ; CHECK1024-NEXT: mov x29, sp
2976 ; CHECK1024-NEXT: .cfi_def_cfa w29, 1136
2977 ; CHECK1024-NEXT: .cfi_offset w19, -8
2978 ; CHECK1024-NEXT: .cfi_offset w20, -16
2979 ; CHECK1024-NEXT: .cfi_offset w28, -24
2980 ; CHECK1024-NEXT: .cfi_offset w30, -40
2981 ; CHECK1024-NEXT: .cfi_offset w29, -48
2982 ; CHECK1024-NEXT: .cfi_offset b8, -1080
2983 ; CHECK1024-NEXT: .cfi_offset b9, -1088
2984 ; CHECK1024-NEXT: .cfi_offset b10, -1096
2985 ; CHECK1024-NEXT: .cfi_offset b11, -1104
2986 ; CHECK1024-NEXT: .cfi_offset b12, -1112
2987 ; CHECK1024-NEXT: .cfi_offset b13, -1120
2988 ; CHECK1024-NEXT: .cfi_offset b14, -1128
2989 ; CHECK1024-NEXT: .cfi_offset b15, -1136
2990 ; CHECK1024-NEXT: sub sp, sp, #1040
2991 ; CHECK1024-NEXT: rdsvl x8, #1
2992 ; CHECK1024-NEXT: mov x9, sp
2993 ; CHECK1024-NEXT: mov w20, w0
2994 ; CHECK1024-NEXT: msub x9, x8, x8, x9
2995 ; CHECK1024-NEXT: mov sp, x9
2996 ; CHECK1024-NEXT: sub x10, x29, #784
2997 ; CHECK1024-NEXT: stur x9, [x10, #-256]
2998 ; CHECK1024-NEXT: sub x9, x29, #774
2999 ; CHECK1024-NEXT: sub x10, x29, #772
3000 ; CHECK1024-NEXT: sturh wzr, [x9, #-256]
3001 ; CHECK1024-NEXT: sub x9, x29, #1040
3002 ; CHECK1024-NEXT: stur wzr, [x10, #-256]
3003 ; CHECK1024-NEXT: sub x10, x29, #776
3004 ; CHECK1024-NEXT: sturh w8, [x10, #-256]
3005 ; CHECK1024-NEXT: msr TPIDR2_EL0, x9
3006 ; CHECK1024-NEXT: .cfi_offset vg, -32
3007 ; CHECK1024-NEXT: smstop sm
3008 ; CHECK1024-NEXT: bl other
3009 ; CHECK1024-NEXT: smstart sm
3010 ; CHECK1024-NEXT: .cfi_restore vg
3011 ; CHECK1024-NEXT: smstart za
3012 ; CHECK1024-NEXT: mrs x8, TPIDR2_EL0
3013 ; CHECK1024-NEXT: sub x0, x29, #1040
3014 ; CHECK1024-NEXT: cbnz x8, .LBB33_2
3015 ; CHECK1024-NEXT: // %bb.1: // %entry
3016 ; CHECK1024-NEXT: bl __arm_tpidr2_restore
3017 ; CHECK1024-NEXT: .LBB33_2: // %entry
3018 ; CHECK1024-NEXT: mov w0, w20
3019 ; CHECK1024-NEXT: msr TPIDR2_EL0, xzr
3020 ; CHECK1024-NEXT: mov sp, x29
3021 ; CHECK1024-NEXT: .cfi_def_cfa wsp, 1136
3022 ; CHECK1024-NEXT: ldp d9, d8, [sp, #48] // 16-byte Folded Reload
3023 ; CHECK1024-NEXT: ldr x19, [sp, #1128] // 8-byte Folded Reload
3024 ; CHECK1024-NEXT: ldp d11, d10, [sp, #32] // 16-byte Folded Reload
3025 ; CHECK1024-NEXT: ldr x20, [sp, #1120] // 8-byte Folded Reload
3026 ; CHECK1024-NEXT: ldr x28, [sp, #1112] // 8-byte Folded Reload
3027 ; CHECK1024-NEXT: ldr x30, [sp, #1096] // 8-byte Folded Reload
3028 ; CHECK1024-NEXT: ldr x29, [sp, #1088] // 8-byte Folded Reload
3029 ; CHECK1024-NEXT: ldp d13, d12, [sp, #16] // 16-byte Folded Reload
3030 ; CHECK1024-NEXT: ldp d15, d14, [sp] // 16-byte Folded Reload
3031 ; CHECK1024-NEXT: add sp, sp, #1136
3032 ; CHECK1024-NEXT: .cfi_def_cfa_offset 0
3033 ; CHECK1024-NEXT: .cfi_restore w19
3034 ; CHECK1024-NEXT: .cfi_restore w20
3035 ; CHECK1024-NEXT: .cfi_restore w28
3036 ; CHECK1024-NEXT: .cfi_restore w30
3037 ; CHECK1024-NEXT: .cfi_restore w29
3038 ; CHECK1024-NEXT: .cfi_restore b8
3039 ; CHECK1024-NEXT: .cfi_restore b9
3040 ; CHECK1024-NEXT: .cfi_restore b10
3041 ; CHECK1024-NEXT: .cfi_restore b11
3042 ; CHECK1024-NEXT: .cfi_restore b12
3043 ; CHECK1024-NEXT: .cfi_restore b13
3044 ; CHECK1024-NEXT: .cfi_restore b14
3045 ; CHECK1024-NEXT: .cfi_restore b15
3046 ; CHECK1024-NEXT: ret
3047 entry:
3048 tail call void @other()
3049 ret i32 %x
3050 }
3051 declare void @other()