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
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=thumbv6m-none-unknown-eabi -mcpu=cortex-m0 | FileCheck %s --check-prefix=ARM
declare i4 @llvm.umul.fix.sat.i4 (i4, i4, i32)
declare i32 @llvm.umul.fix.sat.i32 (i32, i32, i32)
declare i64 @llvm.umul.fix.sat.i64 (i64, i64, i32)
define i32 @func(i32 %x, i32 %y) nounwind {
; ARM-LABEL: func:
; ARM: @ %bb.0:
; ARM-NEXT: .save {r4, lr}
; ARM-NEXT: push {r4, lr}
; ARM-NEXT: mov r2, r1
; ARM-NEXT: movs r4, #0
; ARM-NEXT: mov r1, r4
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: cmp r1, #3
; ARM-NEXT: bhi .LBB0_2
; ARM-NEXT: @ %bb.1:
; ARM-NEXT: lsrs r0, r0, #2
; ARM-NEXT: lsls r1, r1, #30
; ARM-NEXT: adds r0, r1, r0
; ARM-NEXT: pop {r4, pc}
; ARM-NEXT: .LBB0_2:
; ARM-NEXT: mvns r0, r4
; ARM-NEXT: pop {r4, pc}
%tmp = call i32 @llvm.umul.fix.sat.i32(i32 %x, i32 %y, i32 2)
ret i32 %tmp
}
define i64 @func2(i64 %x, i64 %y) nounwind {
; ARM-LABEL: func2:
; ARM: @ %bb.0:
; ARM-NEXT: .save {r4, r5, r6, r7, lr}
; ARM-NEXT: push {r4, r5, r6, r7, lr}
; ARM-NEXT: .pad #28
; ARM-NEXT: sub sp, #28
; ARM-NEXT: str r3, [sp, #24] @ 4-byte Spill
; ARM-NEXT: mov r5, r1
; ARM-NEXT: str r1, [sp, #8] @ 4-byte Spill
; ARM-NEXT: movs r4, #0
; ARM-NEXT: mov r6, r0
; ARM-NEXT: str r0, [sp, #12] @ 4-byte Spill
; ARM-NEXT: mov r1, r4
; ARM-NEXT: mov r7, r2
; ARM-NEXT: str r2, [sp, #16] @ 4-byte Spill
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: str r0, [sp] @ 4-byte Spill
; ARM-NEXT: str r1, [sp, #20] @ 4-byte Spill
; ARM-NEXT: mov r0, r5
; ARM-NEXT: mov r1, r4
; ARM-NEXT: mov r2, r7
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: mov r5, r1
; ARM-NEXT: ldr r1, [sp, #20] @ 4-byte Reload
; ARM-NEXT: adds r0, r0, r1
; ARM-NEXT: str r0, [sp, #20] @ 4-byte Spill
; ARM-NEXT: adcs r5, r4
; ARM-NEXT: mov r0, r6
; ARM-NEXT: mov r1, r4
; ARM-NEXT: ldr r7, [sp, #24] @ 4-byte Reload
; ARM-NEXT: mov r2, r7
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: ldr r2, [sp, #20] @ 4-byte Reload
; ARM-NEXT: adds r0, r0, r2
; ARM-NEXT: str r0, [sp, #20] @ 4-byte Spill
; ARM-NEXT: adcs r1, r4
; ARM-NEXT: adds r0, r5, r1
; ARM-NEXT: str r0, [sp, #4] @ 4-byte Spill
; ARM-NEXT: mov r6, r4
; ARM-NEXT: adcs r6, r4
; ARM-NEXT: ldr r5, [sp, #8] @ 4-byte Reload
; ARM-NEXT: mov r0, r5
; ARM-NEXT: mov r1, r4
; ARM-NEXT: mov r2, r7
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: mov r7, r1
; ARM-NEXT: ldr r1, [sp, #4] @ 4-byte Reload
; ARM-NEXT: adds r0, r0, r1
; ARM-NEXT: str r0, [sp, #4] @ 4-byte Spill
; ARM-NEXT: adcs r7, r6
; ARM-NEXT: ldr r0, [sp, #16] @ 4-byte Reload
; ARM-NEXT: ldr r1, [sp, #24] @ 4-byte Reload
; ARM-NEXT: mov r2, r4
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: mov r6, r0
; ARM-NEXT: str r1, [sp, #24] @ 4-byte Spill
; ARM-NEXT: ldr r0, [sp, #12] @ 4-byte Reload
; ARM-NEXT: mov r1, r5
; ARM-NEXT: mov r2, r4
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: adds r0, r0, r6
; ARM-NEXT: ldr r2, [sp, #24] @ 4-byte Reload
; ARM-NEXT: adcs r1, r2
; ARM-NEXT: ldr r2, [sp, #4] @ 4-byte Reload
; ARM-NEXT: adds r2, r2, r0
; ARM-NEXT: adcs r1, r7
; ARM-NEXT: lsrs r3, r2, #2
; ARM-NEXT: orrs r3, r1
; ARM-NEXT: mvns r1, r4
; ARM-NEXT: cmp r3, #0
; ARM-NEXT: mov r0, r1
; ARM-NEXT: beq .LBB1_3
; ARM-NEXT: @ %bb.1:
; ARM-NEXT: cmp r3, #0
; ARM-NEXT: beq .LBB1_4
; ARM-NEXT: .LBB1_2:
; ARM-NEXT: add sp, #28
; ARM-NEXT: pop {r4, r5, r6, r7, pc}
; ARM-NEXT: .LBB1_3:
; ARM-NEXT: ldr r0, [sp, #20] @ 4-byte Reload
; ARM-NEXT: lsls r0, r0, #30
; ARM-NEXT: ldr r4, [sp] @ 4-byte Reload
; ARM-NEXT: lsrs r4, r4, #2
; ARM-NEXT: adds r0, r0, r4
; ARM-NEXT: cmp r3, #0
; ARM-NEXT: bne .LBB1_2
; ARM-NEXT: .LBB1_4:
; ARM-NEXT: lsls r1, r2, #30
; ARM-NEXT: ldr r2, [sp, #20] @ 4-byte Reload
; ARM-NEXT: lsrs r2, r2, #2
; ARM-NEXT: adds r1, r1, r2
; ARM-NEXT: add sp, #28
; ARM-NEXT: pop {r4, r5, r6, r7, pc}
%tmp = call i64 @llvm.umul.fix.sat.i64(i64 %x, i64 %y, i32 2)
ret i64 %tmp
}
define i4 @func3(i4 %x, i4 %y) nounwind {
; ARM-LABEL: func3:
; ARM: @ %bb.0:
; ARM-NEXT: .save {r4, lr}
; ARM-NEXT: push {r4, lr}
; ARM-NEXT: movs r2, #15
; ARM-NEXT: ands r2, r1
; ARM-NEXT: lsls r0, r0, #28
; ARM-NEXT: movs r4, #0
; ARM-NEXT: mov r1, r4
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: cmp r1, #3
; ARM-NEXT: bhi .LBB2_2
; ARM-NEXT: @ %bb.1:
; ARM-NEXT: lsrs r0, r0, #2
; ARM-NEXT: lsls r1, r1, #30
; ARM-NEXT: adds r0, r1, r0
; ARM-NEXT: lsrs r0, r0, #28
; ARM-NEXT: pop {r4, pc}
; ARM-NEXT: .LBB2_2:
; ARM-NEXT: mvns r0, r4
; ARM-NEXT: lsrs r0, r0, #28
; ARM-NEXT: pop {r4, pc}
%tmp = call i4 @llvm.umul.fix.sat.i4(i4 %x, i4 %y, i32 2)
ret i4 %tmp
}
;; These result in regular integer multiplication with a saturation check.
define i32 @func4(i32 %x, i32 %y) nounwind {
; ARM-LABEL: func4:
; ARM: @ %bb.0:
; ARM-NEXT: .save {r4, lr}
; ARM-NEXT: push {r4, lr}
; ARM-NEXT: mov r2, r1
; ARM-NEXT: movs r4, #0
; ARM-NEXT: mov r1, r4
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: cmp r1, #0
; ARM-NEXT: bls .LBB3_2
; ARM-NEXT: @ %bb.1:
; ARM-NEXT: mvns r0, r4
; ARM-NEXT: .LBB3_2:
; ARM-NEXT: pop {r4, pc}
%tmp = call i32 @llvm.umul.fix.sat.i32(i32 %x, i32 %y, i32 0)
ret i32 %tmp
}
define i64 @func5(i64 %x, i64 %y) {
; ARM-LABEL: func5:
; ARM: @ %bb.0:
; ARM-NEXT: .save {r4, r5, r6, r7, lr}
; ARM-NEXT: push {r4, r5, r6, r7, lr}
; ARM-NEXT: .pad #12
; ARM-NEXT: sub sp, #12
; ARM-NEXT: mov r6, r3
; ARM-NEXT: str r2, [sp, #8] @ 4-byte Spill
; ARM-NEXT: mov r4, r1
; ARM-NEXT: mov r2, r0
; ARM-NEXT: str r0, [sp, #4] @ 4-byte Spill
; ARM-NEXT: movs r5, #0
; ARM-NEXT: mov r0, r3
; ARM-NEXT: mov r1, r5
; ARM-NEXT: mov r3, r5
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: str r0, [sp] @ 4-byte Spill
; ARM-NEXT: mov r7, r1
; ARM-NEXT: subs r0, r1, #1
; ARM-NEXT: sbcs r7, r0
; ARM-NEXT: mov r0, r4
; ARM-NEXT: mov r1, r5
; ARM-NEXT: ldr r2, [sp, #8] @ 4-byte Reload
; ARM-NEXT: mov r3, r5
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: subs r2, r1, #1
; ARM-NEXT: sbcs r1, r2
; ARM-NEXT: subs r2, r6, #1
; ARM-NEXT: sbcs r6, r2
; ARM-NEXT: subs r2, r4, #1
; ARM-NEXT: sbcs r4, r2
; ARM-NEXT: ands r4, r6
; ARM-NEXT: orrs r4, r1
; ARM-NEXT: orrs r4, r7
; ARM-NEXT: ldr r1, [sp] @ 4-byte Reload
; ARM-NEXT: adds r6, r0, r1
; ARM-NEXT: ldr r0, [sp, #4] @ 4-byte Reload
; ARM-NEXT: mov r1, r5
; ARM-NEXT: ldr r2, [sp, #8] @ 4-byte Reload
; ARM-NEXT: mov r3, r5
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: adds r3, r1, r6
; ARM-NEXT: mov r6, r5
; ARM-NEXT: adcs r6, r5
; ARM-NEXT: orrs r6, r4
; ARM-NEXT: mvns r1, r5
; ARM-NEXT: cmp r6, #0
; ARM-NEXT: mov r2, r1
; ARM-NEXT: bne .LBB4_2
; ARM-NEXT: @ %bb.1:
; ARM-NEXT: mov r2, r0
; ARM-NEXT: .LBB4_2:
; ARM-NEXT: cmp r6, #0
; ARM-NEXT: bne .LBB4_4
; ARM-NEXT: @ %bb.3:
; ARM-NEXT: mov r1, r3
; ARM-NEXT: .LBB4_4:
; ARM-NEXT: mov r0, r2
; ARM-NEXT: add sp, #12
; ARM-NEXT: pop {r4, r5, r6, r7, pc}
%tmp = call i64 @llvm.umul.fix.sat.i64(i64 %x, i64 %y, i32 0)
ret i64 %tmp
}
define i4 @func6(i4 %x, i4 %y) nounwind {
; ARM-LABEL: func6:
; ARM: @ %bb.0:
; ARM-NEXT: .save {r4, lr}
; ARM-NEXT: push {r4, lr}
; ARM-NEXT: movs r2, #15
; ARM-NEXT: ands r2, r1
; ARM-NEXT: lsls r0, r0, #28
; ARM-NEXT: movs r4, #0
; ARM-NEXT: mov r1, r4
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: cmp r1, #0
; ARM-NEXT: bls .LBB5_2
; ARM-NEXT: @ %bb.1:
; ARM-NEXT: mvns r0, r4
; ARM-NEXT: .LBB5_2:
; ARM-NEXT: lsrs r0, r0, #28
; ARM-NEXT: pop {r4, pc}
%tmp = call i4 @llvm.umul.fix.sat.i4(i4 %x, i4 %y, i32 0)
ret i4 %tmp
}
define <4 x i32> @vec2(<4 x i32> %x, <4 x i32> %y) nounwind {
; ARM-LABEL: vec2:
; ARM: @ %bb.0:
; ARM-NEXT: .save {r4, r5, r6, r7, lr}
; ARM-NEXT: push {r4, r5, r6, r7, lr}
; ARM-NEXT: .pad #12
; ARM-NEXT: sub sp, #12
; ARM-NEXT: str r3, [sp, #8] @ 4-byte Spill
; ARM-NEXT: mov r7, r2
; ARM-NEXT: mov r5, r1
; ARM-NEXT: ldr r2, [sp, #32]
; ARM-NEXT: movs r6, #0
; ARM-NEXT: mov r1, r6
; ARM-NEXT: mov r3, r6
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: mvns r4, r6
; ARM-NEXT: cmp r1, #0
; ARM-NEXT: mov r1, r4
; ARM-NEXT: bhi .LBB6_2
; ARM-NEXT: @ %bb.1:
; ARM-NEXT: mov r1, r0
; ARM-NEXT: .LBB6_2:
; ARM-NEXT: str r1, [sp, #4] @ 4-byte Spill
; ARM-NEXT: ldr r2, [sp, #36]
; ARM-NEXT: mov r0, r5
; ARM-NEXT: mov r1, r6
; ARM-NEXT: mov r3, r6
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: cmp r1, #0
; ARM-NEXT: mov r5, r4
; ARM-NEXT: bhi .LBB6_4
; ARM-NEXT: @ %bb.3:
; ARM-NEXT: mov r5, r0
; ARM-NEXT: .LBB6_4:
; ARM-NEXT: ldr r2, [sp, #40]
; ARM-NEXT: mov r0, r7
; ARM-NEXT: mov r1, r6
; ARM-NEXT: mov r3, r6
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: cmp r1, #0
; ARM-NEXT: mov r7, r4
; ARM-NEXT: bhi .LBB6_6
; ARM-NEXT: @ %bb.5:
; ARM-NEXT: mov r7, r0
; ARM-NEXT: .LBB6_6:
; ARM-NEXT: ldr r2, [sp, #44]
; ARM-NEXT: ldr r0, [sp, #8] @ 4-byte Reload
; ARM-NEXT: mov r1, r6
; ARM-NEXT: mov r3, r6
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: cmp r1, #0
; ARM-NEXT: bhi .LBB6_8
; ARM-NEXT: @ %bb.7:
; ARM-NEXT: mov r4, r0
; ARM-NEXT: .LBB6_8:
; ARM-NEXT: ldr r0, [sp, #4] @ 4-byte Reload
; ARM-NEXT: mov r1, r5
; ARM-NEXT: mov r2, r7
; ARM-NEXT: mov r3, r4
; ARM-NEXT: add sp, #12
; ARM-NEXT: pop {r4, r5, r6, r7, pc}
%tmp = call <4 x i32> @llvm.umul.fix.sat.v4i32(<4 x i32> %x, <4 x i32> %y, i32 0)
ret <4 x i32> %tmp
}
define i64 @func7(i64 %x, i64 %y) nounwind {
; ARM-LABEL: func7:
; ARM: @ %bb.0:
; ARM-NEXT: .save {r4, r5, r6, r7, lr}
; ARM-NEXT: push {r4, r5, r6, r7, lr}
; ARM-NEXT: .pad #28
; ARM-NEXT: sub sp, #28
; ARM-NEXT: str r3, [sp, #24] @ 4-byte Spill
; ARM-NEXT: mov r7, r2
; ARM-NEXT: str r2, [sp, #20] @ 4-byte Spill
; ARM-NEXT: mov r5, r1
; ARM-NEXT: str r1, [sp, #12] @ 4-byte Spill
; ARM-NEXT: movs r4, #0
; ARM-NEXT: mov r6, r0
; ARM-NEXT: str r0, [sp, #16] @ 4-byte Spill
; ARM-NEXT: mov r1, r4
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: str r1, [sp, #8] @ 4-byte Spill
; ARM-NEXT: mov r0, r5
; ARM-NEXT: mov r1, r4
; ARM-NEXT: mov r2, r7
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: mov r7, r1
; ARM-NEXT: ldr r1, [sp, #8] @ 4-byte Reload
; ARM-NEXT: adds r5, r0, r1
; ARM-NEXT: adcs r7, r4
; ARM-NEXT: mov r0, r6
; ARM-NEXT: mov r1, r4
; ARM-NEXT: ldr r6, [sp, #24] @ 4-byte Reload
; ARM-NEXT: mov r2, r6
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: adds r0, r0, r5
; ARM-NEXT: str r0, [sp, #4] @ 4-byte Spill
; ARM-NEXT: adcs r1, r4
; ARM-NEXT: adds r0, r7, r1
; ARM-NEXT: str r0, [sp, #8] @ 4-byte Spill
; ARM-NEXT: mov r5, r4
; ARM-NEXT: adcs r5, r4
; ARM-NEXT: ldr r7, [sp, #12] @ 4-byte Reload
; ARM-NEXT: mov r0, r7
; ARM-NEXT: mov r1, r4
; ARM-NEXT: mov r2, r6
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: mov r6, r1
; ARM-NEXT: ldr r1, [sp, #8] @ 4-byte Reload
; ARM-NEXT: adds r0, r0, r1
; ARM-NEXT: str r0, [sp, #8] @ 4-byte Spill
; ARM-NEXT: adcs r6, r5
; ARM-NEXT: ldr r0, [sp, #20] @ 4-byte Reload
; ARM-NEXT: ldr r1, [sp, #24] @ 4-byte Reload
; ARM-NEXT: mov r2, r4
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: mov r5, r0
; ARM-NEXT: str r1, [sp, #24] @ 4-byte Spill
; ARM-NEXT: ldr r0, [sp, #16] @ 4-byte Reload
; ARM-NEXT: mov r1, r7
; ARM-NEXT: mov r2, r4
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: adds r0, r0, r5
; ARM-NEXT: ldr r2, [sp, #24] @ 4-byte Reload
; ARM-NEXT: adcs r1, r2
; ARM-NEXT: ldr r2, [sp, #8] @ 4-byte Reload
; ARM-NEXT: adds r3, r2, r0
; ARM-NEXT: adcs r1, r6
; ARM-NEXT: mvns r2, r4
; ARM-NEXT: cmp r1, #0
; ARM-NEXT: mov r0, r2
; ARM-NEXT: bne .LBB7_2
; ARM-NEXT: @ %bb.1:
; ARM-NEXT: ldr r0, [sp, #4] @ 4-byte Reload
; ARM-NEXT: .LBB7_2:
; ARM-NEXT: cmp r1, #0
; ARM-NEXT: bne .LBB7_4
; ARM-NEXT: @ %bb.3:
; ARM-NEXT: mov r2, r3
; ARM-NEXT: .LBB7_4:
; ARM-NEXT: mov r1, r2
; ARM-NEXT: add sp, #28
; ARM-NEXT: pop {r4, r5, r6, r7, pc}
%tmp = call i64 @llvm.umul.fix.sat.i64(i64 %x, i64 %y, i32 32)
ret i64 %tmp
}
define i64 @func8(i64 %x, i64 %y) nounwind {
; ARM-LABEL: func8:
; ARM: @ %bb.0:
; ARM-NEXT: .save {r4, r5, r6, r7, lr}
; ARM-NEXT: push {r4, r5, r6, r7, lr}
; ARM-NEXT: .pad #28
; ARM-NEXT: sub sp, #28
; ARM-NEXT: str r3, [sp, #24] @ 4-byte Spill
; ARM-NEXT: mov r7, r2
; ARM-NEXT: str r2, [sp, #20] @ 4-byte Spill
; ARM-NEXT: mov r5, r1
; ARM-NEXT: str r1, [sp, #12] @ 4-byte Spill
; ARM-NEXT: movs r4, #0
; ARM-NEXT: mov r6, r0
; ARM-NEXT: str r0, [sp, #16] @ 4-byte Spill
; ARM-NEXT: mov r1, r4
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: str r1, [sp, #8] @ 4-byte Spill
; ARM-NEXT: mov r0, r5
; ARM-NEXT: mov r1, r4
; ARM-NEXT: mov r2, r7
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: mov r7, r1
; ARM-NEXT: ldr r1, [sp, #8] @ 4-byte Reload
; ARM-NEXT: adds r5, r0, r1
; ARM-NEXT: adcs r7, r4
; ARM-NEXT: mov r0, r6
; ARM-NEXT: mov r1, r4
; ARM-NEXT: ldr r6, [sp, #24] @ 4-byte Reload
; ARM-NEXT: mov r2, r6
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: adds r0, r0, r5
; ARM-NEXT: str r0, [sp, #4] @ 4-byte Spill
; ARM-NEXT: adcs r1, r4
; ARM-NEXT: adds r0, r7, r1
; ARM-NEXT: str r0, [sp, #8] @ 4-byte Spill
; ARM-NEXT: mov r5, r4
; ARM-NEXT: adcs r5, r4
; ARM-NEXT: ldr r7, [sp, #12] @ 4-byte Reload
; ARM-NEXT: mov r0, r7
; ARM-NEXT: mov r1, r4
; ARM-NEXT: mov r2, r6
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: mov r6, r1
; ARM-NEXT: ldr r1, [sp, #8] @ 4-byte Reload
; ARM-NEXT: adds r0, r0, r1
; ARM-NEXT: str r0, [sp, #8] @ 4-byte Spill
; ARM-NEXT: adcs r6, r5
; ARM-NEXT: ldr r0, [sp, #20] @ 4-byte Reload
; ARM-NEXT: ldr r1, [sp, #24] @ 4-byte Reload
; ARM-NEXT: mov r2, r4
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: mov r5, r0
; ARM-NEXT: str r1, [sp, #24] @ 4-byte Spill
; ARM-NEXT: ldr r0, [sp, #16] @ 4-byte Reload
; ARM-NEXT: mov r1, r7
; ARM-NEXT: mov r2, r4
; ARM-NEXT: mov r3, r4
; ARM-NEXT: bl __aeabi_lmul
; ARM-NEXT: adds r0, r0, r5
; ARM-NEXT: ldr r2, [sp, #24] @ 4-byte Reload
; ARM-NEXT: adcs r1, r2
; ARM-NEXT: ldr r2, [sp, #8] @ 4-byte Reload
; ARM-NEXT: adds r5, r2, r0
; ARM-NEXT: adcs r1, r6
; ARM-NEXT: lsrs r3, r5, #31
; ARM-NEXT: mvns r2, r4
; ARM-NEXT: cmp r3, #0
; ARM-NEXT: mov r0, r2
; ARM-NEXT: bne .LBB8_2
; ARM-NEXT: @ %bb.1:
; ARM-NEXT: lsls r0, r5, #1
; ARM-NEXT: ldr r4, [sp, #4] @ 4-byte Reload
; ARM-NEXT: lsrs r4, r4, #31
; ARM-NEXT: adds r0, r0, r4
; ARM-NEXT: .LBB8_2:
; ARM-NEXT: cmp r3, #0
; ARM-NEXT: bne .LBB8_4
; ARM-NEXT: @ %bb.3:
; ARM-NEXT: lsls r1, r1, #1
; ARM-NEXT: adds r2, r1, r3
; ARM-NEXT: .LBB8_4:
; ARM-NEXT: mov r1, r2
; ARM-NEXT: add sp, #28
; ARM-NEXT: pop {r4, r5, r6, r7, pc}
%tmp = call i64 @llvm.umul.fix.sat.i64(i64 %x, i64 %y, i32 63)
ret i64 %tmp
}
|
