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
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=aarch64-unknown-linux-gnu < %s | FileCheck %s
define <4 x i16> @fold_srem_vec_1(<4 x i16> %x) {
; CHECK-LABEL: fold_srem_vec_1:
; CHECK: // %bb.0:
; CHECK-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-NEXT: smov w8, v0.h[1]
; CHECK-NEXT: smov w9, v0.h[0]
; CHECK-NEXT: mov w10, #63421
; CHECK-NEXT: mov w11, #37253
; CHECK-NEXT: movk w10, #31710, lsl #16
; CHECK-NEXT: movk w11, #44150, lsl #16
; CHECK-NEXT: smov w13, v0.h[2]
; CHECK-NEXT: mov w12, #33437
; CHECK-NEXT: smull x10, w8, w10
; CHECK-NEXT: movk w12, #21399, lsl #16
; CHECK-NEXT: smull x11, w9, w11
; CHECK-NEXT: lsr x10, x10, #32
; CHECK-NEXT: lsr x11, x11, #32
; CHECK-NEXT: sub w10, w10, w8
; CHECK-NEXT: add w11, w11, w9
; CHECK-NEXT: asr w14, w10, #6
; CHECK-NEXT: asr w15, w11, #6
; CHECK-NEXT: add w10, w14, w10, lsr #31
; CHECK-NEXT: add w11, w15, w11, lsr #31
; CHECK-NEXT: mov w14, #95
; CHECK-NEXT: mov w15, #-124
; CHECK-NEXT: smull x12, w13, w12
; CHECK-NEXT: msub w9, w11, w14, w9
; CHECK-NEXT: msub w8, w10, w15, w8
; CHECK-NEXT: lsr x10, x12, #63
; CHECK-NEXT: asr x11, x12, #37
; CHECK-NEXT: smov w12, v0.h[3]
; CHECK-NEXT: add w10, w11, w10
; CHECK-NEXT: mov w11, #98
; CHECK-NEXT: fmov s0, w9
; CHECK-NEXT: mov w9, #63249
; CHECK-NEXT: movk w9, #48808, lsl #16
; CHECK-NEXT: msub w10, w10, w11, w13
; CHECK-NEXT: smull x9, w12, w9
; CHECK-NEXT: mov v0.h[1], w8
; CHECK-NEXT: lsr x8, x9, #63
; CHECK-NEXT: asr x9, x9, #40
; CHECK-NEXT: add w8, w9, w8
; CHECK-NEXT: mov w9, #-1003
; CHECK-NEXT: mov v0.h[2], w10
; CHECK-NEXT: msub w8, w8, w9, w12
; CHECK-NEXT: mov v0.h[3], w8
; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-NEXT: ret
%1 = srem <4 x i16> %x, <i16 95, i16 -124, i16 98, i16 -1003>
ret <4 x i16> %1
}
define <4 x i16> @fold_srem_vec_2(<4 x i16> %x) {
; CHECK-LABEL: fold_srem_vec_2:
; CHECK: // %bb.0:
; CHECK-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-NEXT: smov w9, v0.h[0]
; CHECK-NEXT: mov w8, #37253
; CHECK-NEXT: movk w8, #44150, lsl #16
; CHECK-NEXT: smov w10, v0.h[1]
; CHECK-NEXT: smov w14, v0.h[2]
; CHECK-NEXT: mov w12, #95
; CHECK-NEXT: smull x11, w9, w8
; CHECK-NEXT: smull x13, w10, w8
; CHECK-NEXT: lsr x11, x11, #32
; CHECK-NEXT: add w11, w11, w9
; CHECK-NEXT: lsr x13, x13, #32
; CHECK-NEXT: asr w15, w11, #6
; CHECK-NEXT: add w13, w13, w10
; CHECK-NEXT: add w11, w15, w11, lsr #31
; CHECK-NEXT: smov w15, v0.h[3]
; CHECK-NEXT: asr w16, w13, #6
; CHECK-NEXT: msub w9, w11, w12, w9
; CHECK-NEXT: add w13, w16, w13, lsr #31
; CHECK-NEXT: smull x11, w14, w8
; CHECK-NEXT: msub w10, w13, w12, w10
; CHECK-NEXT: lsr x11, x11, #32
; CHECK-NEXT: smull x8, w15, w8
; CHECK-NEXT: add w11, w11, w14
; CHECK-NEXT: fmov s0, w9
; CHECK-NEXT: asr w9, w11, #6
; CHECK-NEXT: lsr x8, x8, #32
; CHECK-NEXT: add w9, w9, w11, lsr #31
; CHECK-NEXT: add w8, w8, w15
; CHECK-NEXT: mov v0.h[1], w10
; CHECK-NEXT: asr w10, w8, #6
; CHECK-NEXT: msub w9, w9, w12, w14
; CHECK-NEXT: add w8, w10, w8, lsr #31
; CHECK-NEXT: msub w8, w8, w12, w15
; CHECK-NEXT: mov v0.h[2], w9
; CHECK-NEXT: mov v0.h[3], w8
; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-NEXT: ret
%1 = srem <4 x i16> %x, <i16 95, i16 95, i16 95, i16 95>
ret <4 x i16> %1
}
; Don't fold if we can combine srem with sdiv.
define <4 x i16> @combine_srem_sdiv(<4 x i16> %x) {
; CHECK-LABEL: combine_srem_sdiv:
; CHECK: // %bb.0:
; CHECK-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-NEXT: smov w9, v0.h[0]
; CHECK-NEXT: mov w8, #37253
; CHECK-NEXT: movk w8, #44150, lsl #16
; CHECK-NEXT: smov w10, v0.h[1]
; CHECK-NEXT: smov w11, v0.h[2]
; CHECK-NEXT: smov w12, v0.h[3]
; CHECK-NEXT: mov w14, #95
; CHECK-NEXT: smull x13, w9, w8
; CHECK-NEXT: smull x15, w10, w8
; CHECK-NEXT: lsr x13, x13, #32
; CHECK-NEXT: smull x16, w11, w8
; CHECK-NEXT: add w13, w13, w9
; CHECK-NEXT: lsr x15, x15, #32
; CHECK-NEXT: asr w17, w13, #6
; CHECK-NEXT: add w15, w15, w10
; CHECK-NEXT: add w13, w17, w13, lsr #31
; CHECK-NEXT: asr w17, w15, #6
; CHECK-NEXT: add w15, w17, w15, lsr #31
; CHECK-NEXT: smull x8, w12, w8
; CHECK-NEXT: msub w9, w13, w14, w9
; CHECK-NEXT: lsr x16, x16, #32
; CHECK-NEXT: add w16, w16, w11
; CHECK-NEXT: msub w10, w15, w14, w10
; CHECK-NEXT: asr w17, w16, #6
; CHECK-NEXT: lsr x8, x8, #32
; CHECK-NEXT: fmov s1, w13
; CHECK-NEXT: add w16, w17, w16, lsr #31
; CHECK-NEXT: fmov s0, w9
; CHECK-NEXT: add w8, w8, w12
; CHECK-NEXT: asr w9, w8, #6
; CHECK-NEXT: add w8, w9, w8, lsr #31
; CHECK-NEXT: msub w9, w16, w14, w11
; CHECK-NEXT: mov v0.h[1], w10
; CHECK-NEXT: mov v1.h[1], w15
; CHECK-NEXT: msub w10, w8, w14, w12
; CHECK-NEXT: mov v0.h[2], w9
; CHECK-NEXT: mov v1.h[2], w16
; CHECK-NEXT: mov v0.h[3], w10
; CHECK-NEXT: mov v1.h[3], w8
; CHECK-NEXT: add v0.4h, v0.4h, v1.4h
; CHECK-NEXT: ret
%1 = srem <4 x i16> %x, <i16 95, i16 95, i16 95, i16 95>
%2 = sdiv <4 x i16> %x, <i16 95, i16 95, i16 95, i16 95>
%3 = add <4 x i16> %1, %2
ret <4 x i16> %3
}
; Don't fold for divisors that are a power of two.
define <4 x i16> @dont_fold_srem_power_of_two(<4 x i16> %x) {
; CHECK-LABEL: dont_fold_srem_power_of_two:
; CHECK: // %bb.0:
; CHECK-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-NEXT: smov w9, v0.h[1]
; CHECK-NEXT: smov w10, v0.h[0]
; CHECK-NEXT: mov w8, #37253
; CHECK-NEXT: movk w8, #44150, lsl #16
; CHECK-NEXT: add w11, w9, #31
; CHECK-NEXT: cmp w9, #0
; CHECK-NEXT: add w12, w10, #63
; CHECK-NEXT: csel w11, w11, w9, lt
; CHECK-NEXT: cmp w10, #0
; CHECK-NEXT: and w11, w11, #0xffffffe0
; CHECK-NEXT: csel w12, w12, w10, lt
; CHECK-NEXT: sub w9, w9, w11
; CHECK-NEXT: and w12, w12, #0xffffffc0
; CHECK-NEXT: sub w10, w10, w12
; CHECK-NEXT: smov w12, v0.h[3]
; CHECK-NEXT: fmov s1, w10
; CHECK-NEXT: smov w10, v0.h[2]
; CHECK-NEXT: smull x8, w12, w8
; CHECK-NEXT: mov v1.h[1], w9
; CHECK-NEXT: lsr x8, x8, #32
; CHECK-NEXT: add w9, w10, #7
; CHECK-NEXT: cmp w10, #0
; CHECK-NEXT: csel w9, w9, w10, lt
; CHECK-NEXT: add w8, w8, w12
; CHECK-NEXT: and w9, w9, #0xfffffff8
; CHECK-NEXT: sub w9, w10, w9
; CHECK-NEXT: asr w10, w8, #6
; CHECK-NEXT: add w8, w10, w8, lsr #31
; CHECK-NEXT: mov w10, #95
; CHECK-NEXT: mov v1.h[2], w9
; CHECK-NEXT: msub w8, w8, w10, w12
; CHECK-NEXT: mov v1.h[3], w8
; CHECK-NEXT: fmov d0, d1
; CHECK-NEXT: ret
%1 = srem <4 x i16> %x, <i16 64, i16 32, i16 8, i16 95>
ret <4 x i16> %1
}
; Don't fold if the divisor is one.
define <4 x i16> @dont_fold_srem_one(<4 x i16> %x) {
; CHECK-LABEL: dont_fold_srem_one:
; CHECK: // %bb.0:
; CHECK-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-NEXT: smov w8, v0.h[1]
; CHECK-NEXT: smov w9, v0.h[2]
; CHECK-NEXT: mov w10, #30865
; CHECK-NEXT: mov w11, #17097
; CHECK-NEXT: movk w10, #51306, lsl #16
; CHECK-NEXT: movk w11, #45590, lsl #16
; CHECK-NEXT: mov w12, #654
; CHECK-NEXT: smull x10, w8, w10
; CHECK-NEXT: smull x11, w9, w11
; CHECK-NEXT: lsr x10, x10, #32
; CHECK-NEXT: lsr x11, x11, #32
; CHECK-NEXT: add w10, w10, w8
; CHECK-NEXT: add w11, w11, w9
; CHECK-NEXT: asr w13, w10, #9
; CHECK-NEXT: add w10, w13, w10, lsr #31
; CHECK-NEXT: asr w13, w11, #4
; CHECK-NEXT: add w11, w13, w11, lsr #31
; CHECK-NEXT: smov w13, v0.h[3]
; CHECK-NEXT: msub w8, w10, w12, w8
; CHECK-NEXT: movi d0, #0000000000000000
; CHECK-NEXT: mov w12, #47143
; CHECK-NEXT: mov w10, #23
; CHECK-NEXT: movk w12, #24749, lsl #16
; CHECK-NEXT: msub w9, w11, w10, w9
; CHECK-NEXT: smull x10, w13, w12
; CHECK-NEXT: mov v0.h[1], w8
; CHECK-NEXT: lsr x8, x10, #63
; CHECK-NEXT: asr x10, x10, #43
; CHECK-NEXT: add w8, w10, w8
; CHECK-NEXT: mov w10, #5423
; CHECK-NEXT: mov v0.h[2], w9
; CHECK-NEXT: msub w8, w8, w10, w13
; CHECK-NEXT: mov v0.h[3], w8
; CHECK-NEXT: // kill: def $d0 killed $d0 killed $q0
; CHECK-NEXT: ret
%1 = srem <4 x i16> %x, <i16 1, i16 654, i16 23, i16 5423>
ret <4 x i16> %1
}
; Don't fold if the divisor is 2^15.
define <4 x i16> @dont_fold_srem_i16_smax(<4 x i16> %x) {
; CHECK-LABEL: dont_fold_srem_i16_smax:
; CHECK: // %bb.0:
; CHECK-NEXT: // kill: def $d0 killed $d0 def $q0
; CHECK-NEXT: smov w8, v0.h[2]
; CHECK-NEXT: mov w9, #17097
; CHECK-NEXT: smov w10, v0.h[1]
; CHECK-NEXT: movk w9, #45590, lsl #16
; CHECK-NEXT: mov w11, #32767
; CHECK-NEXT: smov w12, v0.h[3]
; CHECK-NEXT: movi d1, #0000000000000000
; CHECK-NEXT: smull x9, w8, w9
; CHECK-NEXT: add w11, w10, w11
; CHECK-NEXT: cmp w10, #0
; CHECK-NEXT: lsr x9, x9, #32
; CHECK-NEXT: csel w11, w11, w10, lt
; CHECK-NEXT: add w9, w9, w8
; CHECK-NEXT: and w11, w11, #0xffff8000
; CHECK-NEXT: asr w13, w9, #4
; CHECK-NEXT: sub w10, w10, w11
; CHECK-NEXT: mov w11, #47143
; CHECK-NEXT: add w9, w13, w9, lsr #31
; CHECK-NEXT: mov w13, #23
; CHECK-NEXT: movk w11, #24749, lsl #16
; CHECK-NEXT: mov v1.h[1], w10
; CHECK-NEXT: msub w8, w9, w13, w8
; CHECK-NEXT: smull x9, w12, w11
; CHECK-NEXT: lsr x10, x9, #63
; CHECK-NEXT: asr x9, x9, #43
; CHECK-NEXT: add w9, w9, w10
; CHECK-NEXT: mov w10, #5423
; CHECK-NEXT: mov v1.h[2], w8
; CHECK-NEXT: msub w8, w9, w10, w12
; CHECK-NEXT: mov v1.h[3], w8
; CHECK-NEXT: fmov d0, d1
; CHECK-NEXT: ret
%1 = srem <4 x i16> %x, <i16 1, i16 32768, i16 23, i16 5423>
ret <4 x i16> %1
}
; Don't fold i64 srem.
define <4 x i64> @dont_fold_srem_i64(<4 x i64> %x) {
; CHECK-LABEL: dont_fold_srem_i64:
; CHECK: // %bb.0:
; CHECK-NEXT: mov x8, #8549
; CHECK-NEXT: fmov x9, d1
; CHECK-NEXT: movk x8, #22795, lsl #16
; CHECK-NEXT: mov x12, #6055
; CHECK-NEXT: movk x8, #17096, lsl #32
; CHECK-NEXT: movk x12, #58853, lsl #16
; CHECK-NEXT: movk x8, #45590, lsl #48
; CHECK-NEXT: mov x14, #21445
; CHECK-NEXT: mov x10, v1.d[1]
; CHECK-NEXT: movk x12, #47142, lsl #32
; CHECK-NEXT: smulh x8, x9, x8
; CHECK-NEXT: movk x14, #1603, lsl #16
; CHECK-NEXT: mov x11, v0.d[1]
; CHECK-NEXT: movk x12, #24749, lsl #48
; CHECK-NEXT: add x8, x8, x9
; CHECK-NEXT: movk x14, #15432, lsl #32
; CHECK-NEXT: asr x13, x8, #4
; CHECK-NEXT: movk x14, #25653, lsl #48
; CHECK-NEXT: add x8, x13, x8, lsr #63
; CHECK-NEXT: mov w13, #23
; CHECK-NEXT: smulh x12, x10, x12
; CHECK-NEXT: smulh x14, x11, x14
; CHECK-NEXT: msub x8, x8, x13, x9
; CHECK-NEXT: asr x13, x12, #11
; CHECK-NEXT: add x12, x13, x12, lsr #63
; CHECK-NEXT: asr x13, x14, #8
; CHECK-NEXT: mov w9, #5423
; CHECK-NEXT: add x13, x13, x14, lsr #63
; CHECK-NEXT: mov w14, #654
; CHECK-NEXT: msub x9, x12, x9, x10
; CHECK-NEXT: fmov d1, x8
; CHECK-NEXT: msub x10, x13, x14, x11
; CHECK-NEXT: movi v0.2d, #0000000000000000
; CHECK-NEXT: mov v1.d[1], x9
; CHECK-NEXT: mov v0.d[1], x10
; CHECK-NEXT: ret
%1 = srem <4 x i64> %x, <i64 1, i64 654, i64 23, i64 5423>
ret <4 x i64> %1
}
|
