1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
; RUN: llc < %s -mtriple=aarch64 -mattr=+mte -aarch64-order-frame-objects=0 | FileCheck %s
declare void @use(ptr %p)
declare void @llvm.aarch64.settag(ptr %p, i64 %a)
declare void @llvm.aarch64.settag.zero(ptr %p, i64 %a)
define void @stg16_16() {
; CHECK-LABEL: stg16_16:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: sub sp, sp, #32
; CHECK-NEXT: .cfi_def_cfa_offset 32
; CHECK-NEXT: st2g sp, [sp], #32
; CHECK-NEXT: ret
entry:
%a = alloca i8, i32 16, align 16
%b = alloca i8, i32 16, align 16
call void @llvm.aarch64.settag(ptr %a, i64 16)
call void @llvm.aarch64.settag(ptr %b, i64 16)
ret void
}
define i32 @stg16_16_16_16_ret() {
; CHECK-LABEL: stg16_16_16_16_ret:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: sub sp, sp, #64
; CHECK-NEXT: .cfi_def_cfa_offset 64
; CHECK-NEXT: mov w0, wzr
; CHECK-NEXT: st2g sp, [sp, #32]
; CHECK-NEXT: st2g sp, [sp], #64
; CHECK-NEXT: ret
entry:
%a = alloca i8, i32 16, align 16
%b = alloca i8, i32 16, align 16
%c = alloca i8, i32 16, align 16
%d = alloca i8, i32 16, align 16
call void @llvm.aarch64.settag(ptr %a, i64 16)
call void @llvm.aarch64.settag(ptr %b, i64 16)
call void @llvm.aarch64.settag(ptr %c, i64 16)
call void @llvm.aarch64.settag(ptr %d, i64 16)
ret i32 0
}
define void @stg16_16_16_16() {
; CHECK-LABEL: stg16_16_16_16:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: sub sp, sp, #64
; CHECK-NEXT: .cfi_def_cfa_offset 64
; CHECK-NEXT: st2g sp, [sp, #32]
; CHECK-NEXT: st2g sp, [sp], #64
; CHECK-NEXT: ret
entry:
%a = alloca i8, i32 16, align 16
%b = alloca i8, i32 16, align 16
%c = alloca i8, i32 16, align 16
%d = alloca i8, i32 16, align 16
call void @llvm.aarch64.settag(ptr %a, i64 16)
call void @llvm.aarch64.settag(ptr %b, i64 16)
call void @llvm.aarch64.settag(ptr %c, i64 16)
call void @llvm.aarch64.settag(ptr %d, i64 16)
ret void
}
define void @stg128_128_128_128() {
; CHECK-LABEL: stg128_128_128_128:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT: sub sp, sp, #512
; CHECK-NEXT: .cfi_def_cfa_offset 528
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: mov x8, #512 // =0x200
; CHECK-NEXT: .LBB3_1: // %entry
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: st2g sp, [sp], #32
; CHECK-NEXT: subs x8, x8, #32
; CHECK-NEXT: b.ne .LBB3_1
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
entry:
%a = alloca i8, i32 128, align 16
%b = alloca i8, i32 128, align 16
%c = alloca i8, i32 128, align 16
%d = alloca i8, i32 128, align 16
call void @llvm.aarch64.settag(ptr %a, i64 128)
call void @llvm.aarch64.settag(ptr %b, i64 128)
call void @llvm.aarch64.settag(ptr %c, i64 128)
call void @llvm.aarch64.settag(ptr %d, i64 128)
ret void
}
define void @stg16_512_16() {
; CHECK-LABEL: stg16_512_16:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT: sub sp, sp, #544
; CHECK-NEXT: .cfi_def_cfa_offset 560
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: mov x8, #544 // =0x220
; CHECK-NEXT: .LBB4_1: // %entry
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: st2g sp, [sp], #32
; CHECK-NEXT: subs x8, x8, #32
; CHECK-NEXT: b.ne .LBB4_1
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
entry:
%a = alloca i8, i32 16, align 16
%b = alloca i8, i32 512, align 16
%c = alloca i8, i32 16, align 16
call void @llvm.aarch64.settag(ptr %a, i64 16)
call void @llvm.aarch64.settag(ptr %b, i64 512)
call void @llvm.aarch64.settag(ptr %c, i64 16)
ret void
}
define void @stg512_512_512() {
; CHECK-LABEL: stg512_512_512:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT: sub sp, sp, #1536
; CHECK-NEXT: .cfi_def_cfa_offset 1552
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: mov x8, #1536 // =0x600
; CHECK-NEXT: .LBB5_1: // %entry
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: st2g sp, [sp], #32
; CHECK-NEXT: subs x8, x8, #32
; CHECK-NEXT: b.ne .LBB5_1
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
entry:
%a = alloca i8, i32 512, align 16
%b = alloca i8, i32 512, align 16
%c = alloca i8, i32 512, align 16
call void @llvm.aarch64.settag(ptr %a, i64 512)
call void @llvm.aarch64.settag(ptr %b, i64 512)
call void @llvm.aarch64.settag(ptr %c, i64 512)
ret void
}
define void @early(i1 %flag) {
; CHECK-LABEL: early:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: sub sp, sp, #144
; CHECK-NEXT: .cfi_def_cfa_offset 144
; CHECK-NEXT: tbz w0, #0, .LBB6_2
; CHECK-NEXT: // %bb.1: // %if.then
; CHECK-NEXT: st2g sp, [sp, #48]
; CHECK-NEXT: st2g sp, [sp, #80]
; CHECK-NEXT: st2g sp, [sp, #112]
; CHECK-NEXT: .LBB6_2: // %if.end
; CHECK-NEXT: stg sp, [sp, #32]
; CHECK-NEXT: st2g sp, [sp], #144
; CHECK-NEXT: ret
entry:
%a = alloca i8, i32 48, align 16
%b = alloca i8, i32 48, align 16
%c = alloca i8, i32 48, align 16
br i1 %flag, label %if.then, label %if.end
if.then:
call void @llvm.aarch64.settag(ptr %a, i64 48)
call void @llvm.aarch64.settag(ptr %b, i64 48)
br label %if.end
if.end:
call void @llvm.aarch64.settag(ptr %c, i64 48)
ret void
}
define void @early_128_128(i1 %flag) {
; CHECK-LABEL: early_128_128:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: sub sp, sp, #320
; CHECK-NEXT: str x29, [sp, #304] // 8-byte Folded Spill
; CHECK-NEXT: .cfi_def_cfa_offset 320
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: tbz w0, #0, .LBB7_4
; CHECK-NEXT: // %bb.1: // %if.then
; CHECK-NEXT: add x9, sp, #48
; CHECK-NEXT: mov x8, #256 // =0x100
; CHECK-NEXT: .LBB7_2: // %if.then
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: st2g x9, [x9], #32
; CHECK-NEXT: subs x8, x8, #32
; CHECK-NEXT: b.ne .LBB7_2
; CHECK-NEXT: // %bb.3: // %if.then
; CHECK-NEXT: .LBB7_4: // %if.end
; CHECK-NEXT: stg sp, [sp, #32]
; CHECK-NEXT: st2g sp, [sp], #304
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
entry:
%a = alloca i8, i32 128, align 16
%b = alloca i8, i32 128, align 16
%c = alloca i8, i32 48, align 16
br i1 %flag, label %if.then, label %if.end
if.then:
call void @llvm.aarch64.settag(ptr %a, i64 128)
call void @llvm.aarch64.settag(ptr %b, i64 128)
br label %if.end
if.end:
call void @llvm.aarch64.settag(ptr %c, i64 48)
ret void
}
define void @early_512_512(i1 %flag) {
; CHECK-LABEL: early_512_512:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: str x29, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT: sub sp, sp, #1072
; CHECK-NEXT: .cfi_def_cfa_offset 1088
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: tbz w0, #0, .LBB8_4
; CHECK-NEXT: // %bb.1: // %if.then
; CHECK-NEXT: add x9, sp, #48
; CHECK-NEXT: mov x8, #1024 // =0x400
; CHECK-NEXT: .LBB8_2: // %if.then
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: st2g x9, [x9], #32
; CHECK-NEXT: subs x8, x8, #32
; CHECK-NEXT: b.ne .LBB8_2
; CHECK-NEXT: // %bb.3: // %if.then
; CHECK-NEXT: .LBB8_4: // %if.end
; CHECK-NEXT: stg sp, [sp, #32]
; CHECK-NEXT: st2g sp, [sp], #1072
; CHECK-NEXT: ldr x29, [sp], #16 // 8-byte Folded Reload
; CHECK-NEXT: ret
entry:
%a = alloca i8, i32 512, align 16
%b = alloca i8, i32 512, align 16
%c = alloca i8, i32 48, align 16
br i1 %flag, label %if.then, label %if.end
if.then:
call void @llvm.aarch64.settag(ptr %a, i64 512)
call void @llvm.aarch64.settag(ptr %b, i64 512)
br label %if.end
if.end:
call void @llvm.aarch64.settag(ptr %c, i64 48)
ret void
}
; Two loops of size 256; the second loop updates SP.
define void @stg128_128_gap_128_128() {
; CHECK-LABEL: stg128_128_gap_128_128:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: stp x29, x30, [sp, #-16]! // 16-byte Folded Spill
; CHECK-NEXT: sub sp, sp, #544
; CHECK-NEXT: .cfi_def_cfa_offset 560
; CHECK-NEXT: .cfi_offset w30, -8
; CHECK-NEXT: .cfi_offset w29, -16
; CHECK-NEXT: add x0, sp, #256
; CHECK-NEXT: bl use
; CHECK-NEXT: mov x9, sp
; CHECK-NEXT: mov x8, #256 // =0x100
; CHECK-NEXT: .LBB9_1: // %entry
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: st2g x9, [x9], #32
; CHECK-NEXT: subs x8, x8, #32
; CHECK-NEXT: b.ne .LBB9_1
; CHECK-NEXT: // %bb.2: // %entry
; CHECK-NEXT: add sp, sp, #288
; CHECK-NEXT: mov x8, #256 // =0x100
; CHECK-NEXT: .LBB9_3: // %entry
; CHECK-NEXT: // =>This Inner Loop Header: Depth=1
; CHECK-NEXT: st2g sp, [sp], #32
; CHECK-NEXT: subs x8, x8, #32
; CHECK-NEXT: b.ne .LBB9_3
; CHECK-NEXT: // %bb.4: // %entry
; CHECK-NEXT: ldp x29, x30, [sp], #16 // 16-byte Folded Reload
; CHECK-NEXT: ret
entry:
%a = alloca i8, i32 128, align 16
%a2 = alloca i8, i32 128, align 16
%b = alloca i8, i32 32, align 16
%c = alloca i8, i32 128, align 16
%c2 = alloca i8, i32 128, align 16
call void @use(ptr %b)
call void @llvm.aarch64.settag(ptr %a, i64 128)
call void @llvm.aarch64.settag(ptr %a2, i64 128)
call void @llvm.aarch64.settag(ptr %c, i64 128)
call void @llvm.aarch64.settag(ptr %c2, i64 128)
ret void
}
; Function Attrs: nounwind
declare i32 @printf(ptr, ...) #0
@.str = private unnamed_addr constant [4 x i8] c"%d\0A\00", align 1
; Case 1
; Insert point of stg merge is followed by nzcv read
; Don't merge in this case
define i32 @nzcv_clobber(i32 %in) {
entry:
; CHECK-LABEL: nzcv_clobber:
; CHECK: stg sp, [sp, #528]
; CHECK-NEXT: .LBB10_1:
; CHECK: st2g x9, [x9], #32
; CHECK-NEXT: subs x8, x8, #32
; CHECK-NEXT: b.ne .LBB10_1
; CHECK-NEXT: // %bb.2:
; CHECK-NEXT: cmp w0, #10
; CHECK-NEXT: stg sp, [sp]
; CHECK-NEXT: b.ge .LBB10_4
%a = alloca i8, i32 16, align 16
%b = alloca i8, i32 512, align 16
%c = alloca i8, i32 16, align 16
call void @llvm.aarch64.settag(ptr %a, i64 16)
call void @llvm.aarch64.settag(ptr %b, i64 512)
%cmp = icmp slt i32 %in, 10
call void @llvm.aarch64.settag(ptr %c, i64 16)
br i1 %cmp, label %return0, label %return1
return0: ; preds = %entry
%call = call i32 (ptr, ...) @printf(ptr @.str, i32 10) #1
ret i32 0
return1:
ret i32 1
}
; Case 2
; Insert point of stg merge is not followed by nzcv read
; Merge in this case
define i32 @nzcv_no_clobber(i32 %in) {
entry:
; CHECK-LABEL: nzcv_no_clobber:
; CHECK: mov x8, #544
; CHECK-NEXT: .LBB11_1:
; CHECK: st2g sp, [sp], #32
; CHECK-NEXT: subs x8, x8, #32
; CHECK-NEXT: b.ne .LBB11_1
%a = alloca i8, i32 16, align 16
%b = alloca i8, i32 512, align 16
%c = alloca i8, i32 16, align 16
call void @llvm.aarch64.settag(ptr %a, i64 16)
call void @llvm.aarch64.settag(ptr %b, i64 512)
call void @llvm.aarch64.settag(ptr %c, i64 16)
br label %return1
return0: ; preds = %entry
%call = call i32 (ptr, ...) @printf(ptr @.str, i32 10) #1
ret i32 0
return1:
ret i32 1
}
|
