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
| ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -basicaa -slp-vectorizer -S | FileCheck %s
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"
@b = common global [4 x i32] zeroinitializer, align 16
@c = common global [4 x i32] zeroinitializer, align 16
@d = common global [4 x i32] zeroinitializer, align 16
@e = common global [4 x i32] zeroinitializer, align 16
@a = common global [4 x i32] zeroinitializer, align 16
@fb = common global [4 x float] zeroinitializer, align 16
@fc = common global [4 x float] zeroinitializer, align 16
@fa = common global [4 x float] zeroinitializer, align 16
@fd = common global [4 x float] zeroinitializer, align 16
; Function Attrs: nounwind uwtable
define void @addsub() #0 {
; CHECK-LABEL: @addsub(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP0:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([4 x i32]* @b to <4 x i32>*), align 4
; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([4 x i32]* @c to <4 x i32>*), align 4
; CHECK-NEXT: [[TMP2:%.*]] = add nsw <4 x i32> [[TMP0]], [[TMP1]]
; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([4 x i32]* @d to <4 x i32>*), align 4
; CHECK-NEXT: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([4 x i32]* @e to <4 x i32>*), align 4
; CHECK-NEXT: [[TMP5:%.*]] = add nsw <4 x i32> [[TMP3]], [[TMP4]]
; CHECK-NEXT: [[TMP6:%.*]] = add nsw <4 x i32> [[TMP2]], [[TMP5]]
; CHECK-NEXT: [[TMP7:%.*]] = sub nsw <4 x i32> [[TMP2]], [[TMP5]]
; CHECK-NEXT: [[TMP8:%.*]] = shufflevector <4 x i32> [[TMP6]], <4 x i32> [[TMP7]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT: store <4 x i32> [[TMP8]], <4 x i32>* bitcast ([4 x i32]* @a to <4 x i32>*), align 4
; CHECK-NEXT: ret void
;
entry:
%0 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 0), align 4
%1 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 0), align 4
%add = add nsw i32 %0, %1
%2 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 0), align 4
%3 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 0), align 4
%add1 = add nsw i32 %2, %3
%add2 = add nsw i32 %add, %add1
store i32 %add2, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 0), align 4
%4 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 1), align 4
%5 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 1), align 4
%add3 = add nsw i32 %4, %5
%6 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 1), align 4
%7 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 1), align 4
%add4 = add nsw i32 %6, %7
%sub = sub nsw i32 %add3, %add4
store i32 %sub, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 1), align 4
%8 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 2), align 4
%9 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 2), align 4
%add5 = add nsw i32 %8, %9
%10 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 2), align 4
%11 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 2), align 4
%add6 = add nsw i32 %10, %11
%add7 = add nsw i32 %add5, %add6
store i32 %add7, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 2), align 4
%12 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 3), align 4
%13 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 3), align 4
%add8 = add nsw i32 %12, %13
%14 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 3), align 4
%15 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 3), align 4
%add9 = add nsw i32 %14, %15
%sub10 = sub nsw i32 %add8, %add9
store i32 %sub10, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 3), align 4
ret void
}
; Function Attrs: nounwind uwtable
define void @subadd() #0 {
; CHECK-LABEL: @subadd(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP0:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([4 x i32]* @b to <4 x i32>*), align 4
; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([4 x i32]* @c to <4 x i32>*), align 4
; CHECK-NEXT: [[TMP2:%.*]] = add nsw <4 x i32> [[TMP0]], [[TMP1]]
; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([4 x i32]* @d to <4 x i32>*), align 4
; CHECK-NEXT: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([4 x i32]* @e to <4 x i32>*), align 4
; CHECK-NEXT: [[TMP5:%.*]] = add nsw <4 x i32> [[TMP3]], [[TMP4]]
; CHECK-NEXT: [[TMP6:%.*]] = sub nsw <4 x i32> [[TMP2]], [[TMP5]]
; CHECK-NEXT: [[TMP7:%.*]] = add nsw <4 x i32> [[TMP2]], [[TMP5]]
; CHECK-NEXT: [[TMP8:%.*]] = shufflevector <4 x i32> [[TMP6]], <4 x i32> [[TMP7]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT: store <4 x i32> [[TMP8]], <4 x i32>* bitcast ([4 x i32]* @a to <4 x i32>*), align 4
; CHECK-NEXT: ret void
;
entry:
%0 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 0), align 4
%1 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 0), align 4
%add = add nsw i32 %0, %1
%2 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 0), align 4
%3 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 0), align 4
%add1 = add nsw i32 %2, %3
%sub = sub nsw i32 %add, %add1
store i32 %sub, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 0), align 4
%4 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 1), align 4
%5 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 1), align 4
%add2 = add nsw i32 %4, %5
%6 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 1), align 4
%7 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 1), align 4
%add3 = add nsw i32 %6, %7
%add4 = add nsw i32 %add2, %add3
store i32 %add4, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 1), align 4
%8 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 2), align 4
%9 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 2), align 4
%add5 = add nsw i32 %8, %9
%10 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 2), align 4
%11 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 2), align 4
%add6 = add nsw i32 %10, %11
%sub7 = sub nsw i32 %add5, %add6
store i32 %sub7, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 2), align 4
%12 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 3), align 4
%13 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 3), align 4
%add8 = add nsw i32 %12, %13
%14 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 3), align 4
%15 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 3), align 4
%add9 = add nsw i32 %14, %15
%add10 = add nsw i32 %add8, %add9
store i32 %add10, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 3), align 4
ret void
}
; Function Attrs: nounwind uwtable
define void @faddfsub() #0 {
; CHECK-LABEL: @faddfsub(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP0:%.*]] = load <4 x float>, <4 x float>* bitcast ([4 x float]* @fb to <4 x float>*), align 4
; CHECK-NEXT: [[TMP1:%.*]] = load <4 x float>, <4 x float>* bitcast ([4 x float]* @fc to <4 x float>*), align 4
; CHECK-NEXT: [[TMP2:%.*]] = fadd <4 x float> [[TMP0]], [[TMP1]]
; CHECK-NEXT: [[TMP3:%.*]] = fsub <4 x float> [[TMP0]], [[TMP1]]
; CHECK-NEXT: [[TMP4:%.*]] = shufflevector <4 x float> [[TMP2]], <4 x float> [[TMP3]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT: store <4 x float> [[TMP4]], <4 x float>* bitcast ([4 x float]* @fa to <4 x float>*), align 4
; CHECK-NEXT: ret void
;
entry:
%0 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 0), align 4
%1 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 0), align 4
%add = fadd float %0, %1
store float %add, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 0), align 4
%2 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 1), align 4
%3 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 1), align 4
%sub = fsub float %2, %3
store float %sub, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 1), align 4
%4 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 2), align 4
%5 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 2), align 4
%add1 = fadd float %4, %5
store float %add1, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 2), align 4
%6 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 3), align 4
%7 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 3), align 4
%sub2 = fsub float %6, %7
store float %sub2, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 3), align 4
ret void
}
; Function Attrs: nounwind uwtable
define void @fsubfadd() #0 {
; CHECK-LABEL: @fsubfadd(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP0:%.*]] = load <4 x float>, <4 x float>* bitcast ([4 x float]* @fb to <4 x float>*), align 4
; CHECK-NEXT: [[TMP1:%.*]] = load <4 x float>, <4 x float>* bitcast ([4 x float]* @fc to <4 x float>*), align 4
; CHECK-NEXT: [[TMP2:%.*]] = fsub <4 x float> [[TMP0]], [[TMP1]]
; CHECK-NEXT: [[TMP3:%.*]] = fadd <4 x float> [[TMP0]], [[TMP1]]
; CHECK-NEXT: [[TMP4:%.*]] = shufflevector <4 x float> [[TMP2]], <4 x float> [[TMP3]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT: store <4 x float> [[TMP4]], <4 x float>* bitcast ([4 x float]* @fa to <4 x float>*), align 4
; CHECK-NEXT: ret void
;
entry:
%0 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 0), align 4
%1 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 0), align 4
%sub = fsub float %0, %1
store float %sub, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 0), align 4
%2 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 1), align 4
%3 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 1), align 4
%add = fadd float %2, %3
store float %add, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 1), align 4
%4 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 2), align 4
%5 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 2), align 4
%sub1 = fsub float %4, %5
store float %sub1, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 2), align 4
%6 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 3), align 4
%7 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 3), align 4
%add2 = fadd float %6, %7
store float %add2, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 3), align 4
ret void
}
; Function Attrs: nounwind uwtable
define void @faddfsub_select() #0 {
; CHECK-LABEL: @faddfsub_select(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP0:%.*]] = load <4 x float>, <4 x float>* bitcast ([4 x float]* @fb to <4 x float>*), align 4
; CHECK-NEXT: [[TMP1:%.*]] = load <4 x float>, <4 x float>* bitcast ([4 x float]* @fc to <4 x float>*), align 4
; CHECK-NEXT: [[TMP2:%.*]] = fadd <4 x float> [[TMP0]], [[TMP1]]
; CHECK-NEXT: [[TMP3:%.*]] = fsub <4 x float> [[TMP0]], [[TMP1]]
; CHECK-NEXT: [[TMP4:%.*]] = shufflevector <4 x float> [[TMP2]], <4 x float> [[TMP3]], <4 x i32> <i32 0, i32 1, i32 2, i32 7>
; CHECK-NEXT: store <4 x float> [[TMP4]], <4 x float>* bitcast ([4 x float]* @fa to <4 x float>*), align 4
; CHECK-NEXT: ret void
;
entry:
%0 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 0), align 4
%1 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 0), align 4
%add = fadd float %0, %1
store float %add, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 0), align 4
%2 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 1), align 4
%3 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 1), align 4
%add1 = fadd float %2, %3
store float %add1, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 1), align 4
%4 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 2), align 4
%5 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 2), align 4
%add2 = fadd float %4, %5
store float %add2, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 2), align 4
%6 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 3), align 4
%7 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 3), align 4
%sub = fsub float %6, %7
store float %sub, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 3), align 4
ret void
}
; Check vectorization of following code for float data type-
; fc[0] = fb[0]+fa[0]; //swapped fb and fa
; fc[1] = fa[1]-fb[1];
; fc[2] = fa[2]+fb[2];
; fc[3] = fa[3]-fb[3];
define void @reorder_alt() #0 {
; CHECK-LABEL: @reorder_alt(
; CHECK-NEXT: [[TMP1:%.*]] = load <4 x float>, <4 x float>* bitcast ([4 x float]* @fa to <4 x float>*), align 4
; CHECK-NEXT: [[TMP2:%.*]] = load <4 x float>, <4 x float>* bitcast ([4 x float]* @fb to <4 x float>*), align 4
; CHECK-NEXT: [[TMP3:%.*]] = fadd <4 x float> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = fsub <4 x float> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP5:%.*]] = shufflevector <4 x float> [[TMP3]], <4 x float> [[TMP4]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT: store <4 x float> [[TMP5]], <4 x float>* bitcast ([4 x float]* @fc to <4 x float>*), align 4
; CHECK-NEXT: ret void
;
%1 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 0), align 4
%2 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 0), align 4
%3 = fadd float %1, %2
store float %3, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 0), align 4
%4 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 1), align 4
%5 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 1), align 4
%6 = fsub float %4, %5
store float %6, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 1), align 4
%7 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 2), align 4
%8 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 2), align 4
%9 = fadd float %7, %8
store float %9, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 2), align 4
%10 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 3), align 4
%11 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 3), align 4
%12 = fsub float %10, %11
store float %12, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 3), align 4
ret void
}
; Check vectorization of following code for float data type-
; fc[0] = fa[0]+(fb[0]-fd[0]);
; fc[1] = fa[1]-(fb[1]+fd[1]);
; fc[2] = fa[2]+(fb[2]-fd[2]);
; fc[3] = fa[3]-(fd[3]+fb[3]); //swapped fd and fb
define void @reorder_alt_subTree() #0 {
; CHECK-LABEL: @reorder_alt_subTree(
; CHECK-NEXT: [[TMP1:%.*]] = load <4 x float>, <4 x float>* bitcast ([4 x float]* @fa to <4 x float>*), align 4
; CHECK-NEXT: [[TMP2:%.*]] = load <4 x float>, <4 x float>* bitcast ([4 x float]* @fd to <4 x float>*), align 4
; CHECK-NEXT: [[TMP3:%.*]] = load <4 x float>, <4 x float>* bitcast ([4 x float]* @fb to <4 x float>*), align 4
; CHECK-NEXT: [[TMP4:%.*]] = fsub <4 x float> [[TMP3]], [[TMP2]]
; CHECK-NEXT: [[TMP5:%.*]] = fadd <4 x float> [[TMP3]], [[TMP2]]
; CHECK-NEXT: [[TMP6:%.*]] = shufflevector <4 x float> [[TMP4]], <4 x float> [[TMP5]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT: [[TMP7:%.*]] = fadd <4 x float> [[TMP1]], [[TMP6]]
; CHECK-NEXT: [[TMP8:%.*]] = fsub <4 x float> [[TMP1]], [[TMP6]]
; CHECK-NEXT: [[TMP9:%.*]] = shufflevector <4 x float> [[TMP7]], <4 x float> [[TMP8]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT: store <4 x float> [[TMP9]], <4 x float>* bitcast ([4 x float]* @fc to <4 x float>*), align 4
; CHECK-NEXT: ret void
;
%1 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 0), align 4
%2 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 0), align 4
%3 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fd, i32 0, i64 0), align 4
%4 = fsub float %2, %3
%5 = fadd float %1, %4
store float %5, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 0), align 4
%6 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 1), align 4
%7 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 1), align 4
%8 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fd, i32 0, i64 1), align 4
%9 = fadd float %7, %8
%10 = fsub float %6, %9
store float %10, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 1), align 4
%11 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 2), align 4
%12 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 2), align 4
%13 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fd, i32 0, i64 2), align 4
%14 = fsub float %12, %13
%15 = fadd float %11, %14
store float %15, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 2), align 4
%16 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 3), align 4
%17 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fd, i32 0, i64 3), align 4
%18 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 3), align 4
%19 = fadd float %17, %18
%20 = fsub float %16, %19
store float %20, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 3), align 4
ret void
}
; Check vectorization of following code for double data type-
; c[0] = (a[0]+b[0])-d[0];
; c[1] = d[1]+(a[1]+b[1]); //swapped d[1] and (a[1]+b[1])
define void @reorder_alt_rightsubTree(double* nocapture %c, double* noalias nocapture readonly %a, double* noalias nocapture readonly %b, double* noalias nocapture readonly %d) {
; CHECK-LABEL: @reorder_alt_rightsubTree(
; CHECK-NEXT: [[TMP1:%.*]] = getelementptr inbounds double, double* [[D:%.*]], i64 1
; CHECK-NEXT: [[TMP2:%.*]] = bitcast double* [[D]] to <2 x double>*
; CHECK-NEXT: [[TMP3:%.*]] = load <2 x double>, <2 x double>* [[TMP2]], align 8
; CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds double, double* [[A:%.*]], i64 1
; CHECK-NEXT: [[TMP5:%.*]] = bitcast double* [[A]] to <2 x double>*
; CHECK-NEXT: [[TMP6:%.*]] = load <2 x double>, <2 x double>* [[TMP5]], align 8
; CHECK-NEXT: [[TMP7:%.*]] = getelementptr inbounds double, double* [[B:%.*]], i64 1
; CHECK-NEXT: [[TMP8:%.*]] = bitcast double* [[B]] to <2 x double>*
; CHECK-NEXT: [[TMP9:%.*]] = load <2 x double>, <2 x double>* [[TMP8]], align 8
; CHECK-NEXT: [[TMP10:%.*]] = fadd <2 x double> [[TMP6]], [[TMP9]]
; CHECK-NEXT: [[TMP11:%.*]] = fsub <2 x double> [[TMP10]], [[TMP3]]
; CHECK-NEXT: [[TMP12:%.*]] = fadd <2 x double> [[TMP10]], [[TMP3]]
; CHECK-NEXT: [[TMP13:%.*]] = shufflevector <2 x double> [[TMP11]], <2 x double> [[TMP12]], <2 x i32> <i32 0, i32 3>
; CHECK-NEXT: [[TMP14:%.*]] = getelementptr inbounds double, double* [[C:%.*]], i64 1
; CHECK-NEXT: [[TMP15:%.*]] = bitcast double* [[C]] to <2 x double>*
; CHECK-NEXT: store <2 x double> [[TMP13]], <2 x double>* [[TMP15]], align 8
; CHECK-NEXT: ret void
;
%1 = load double, double* %a
%2 = load double, double* %b
%3 = fadd double %1, %2
%4 = load double, double* %d
%5 = fsub double %3, %4
store double %5, double* %c
%6 = getelementptr inbounds double, double* %d, i64 1
%7 = load double, double* %6
%8 = getelementptr inbounds double, double* %a, i64 1
%9 = load double, double* %8
%10 = getelementptr inbounds double, double* %b, i64 1
%11 = load double, double* %10
%12 = fadd double %9, %11
%13 = fadd double %7, %12
%14 = getelementptr inbounds double, double* %c, i64 1
store double %13, double* %14
ret void
}
; Dont vectorization of following code for float data type as sub is not commutative-
; fc[0] = fb[0]+fa[0];
; fc[1] = fa[1]-fb[1];
; fc[2] = fa[2]+fb[2];
; fc[3] = fb[3]-fa[3];
; In the above code we can swap the 1st and 2nd operation as fadd is commutative
; but not 2nd or 4th as fsub is not commutative.
define void @no_vec_shuff_reorder() #0 {
; CHECK-LABEL: @no_vec_shuff_reorder(
; CHECK-NEXT: [[TMP1:%.*]] = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 0), align 4
; CHECK-NEXT: [[TMP2:%.*]] = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 0), align 4
; CHECK-NEXT: [[TMP3:%.*]] = fadd float [[TMP1]], [[TMP2]]
; CHECK-NEXT: store float [[TMP3]], float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 0), align 4
; CHECK-NEXT: [[TMP4:%.*]] = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 1), align 4
; CHECK-NEXT: [[TMP5:%.*]] = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 1), align 4
; CHECK-NEXT: [[TMP6:%.*]] = fsub float [[TMP4]], [[TMP5]]
; CHECK-NEXT: store float [[TMP6]], float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 1), align 4
; CHECK-NEXT: [[TMP7:%.*]] = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 2), align 4
; CHECK-NEXT: [[TMP8:%.*]] = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 2), align 4
; CHECK-NEXT: [[TMP9:%.*]] = fadd float [[TMP7]], [[TMP8]]
; CHECK-NEXT: store float [[TMP9]], float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 2), align 4
; CHECK-NEXT: [[TMP10:%.*]] = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 3), align 4
; CHECK-NEXT: [[TMP11:%.*]] = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 3), align 4
; CHECK-NEXT: [[TMP12:%.*]] = fsub float [[TMP10]], [[TMP11]]
; CHECK-NEXT: store float [[TMP12]], float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 3), align 4
; CHECK-NEXT: ret void
;
%1 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 0), align 4
%2 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 0), align 4
%3 = fadd float %1, %2
store float %3, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 0), align 4
%4 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 1), align 4
%5 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 1), align 4
%6 = fsub float %4, %5
store float %6, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 1), align 4
%7 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 2), align 4
%8 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 2), align 4
%9 = fadd float %7, %8
store float %9, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 2), align 4
%10 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 3), align 4
%11 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 3), align 4
%12 = fsub float %10, %11
store float %12, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 3), align 4
ret void
}
attributes #0 = { nounwind }
|