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
| ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=riscv32 -verify-machineinstrs < %s \
; RUN: | FileCheck -check-prefixes=CHECK,RV32I %s
; RUN: llc -mtriple=riscv32 -mattr=+m -verify-machineinstrs < %s \
; RUN: | FileCheck -check-prefixes=CHECK,RV32IM %s
; RUN: llc -mtriple=riscv64 -verify-machineinstrs < %s \
; RUN: | FileCheck -check-prefixes=CHECK,RV64I %s
; RUN: llc -mtriple=riscv64 -mattr=+m -verify-machineinstrs < %s \
; RUN: | FileCheck -check-prefixes=CHECK,RV64IM %s
define i32 @fold_urem_positive_odd(i32 %x) {
; RV32I-LABEL: fold_urem_positive_odd:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: .cfi_def_cfa_offset 16
; RV32I-NEXT: sw ra, 12(sp)
; RV32I-NEXT: .cfi_offset ra, -4
; RV32I-NEXT: addi a1, zero, 95
; RV32I-NEXT: call __umodsi3
; RV32I-NEXT: lw ra, 12(sp)
; RV32I-NEXT: .cfi_restore ra
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: .cfi_def_cfa_offset 0
; RV32I-NEXT: ret
;
; RV32IM-LABEL: fold_urem_positive_odd:
; RV32IM: # %bb.0:
; RV32IM-NEXT: lui a1, 364242
; RV32IM-NEXT: addi a1, a1, 777
; RV32IM-NEXT: mulhu a1, a0, a1
; RV32IM-NEXT: sub a2, a0, a1
; RV32IM-NEXT: srli a2, a2, 1
; RV32IM-NEXT: add a1, a2, a1
; RV32IM-NEXT: srli a1, a1, 6
; RV32IM-NEXT: addi a2, zero, 95
; RV32IM-NEXT: mul a1, a1, a2
; RV32IM-NEXT: sub a0, a0, a1
; RV32IM-NEXT: .cfi_def_cfa_offset 0
; RV32IM-NEXT: ret
;
; RV64I-LABEL: fold_urem_positive_odd:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: .cfi_def_cfa_offset 16
; RV64I-NEXT: sd ra, 8(sp)
; RV64I-NEXT: .cfi_offset ra, -8
; RV64I-NEXT: slli a0, a0, 32
; RV64I-NEXT: srli a0, a0, 32
; RV64I-NEXT: addi a1, zero, 95
; RV64I-NEXT: call __umoddi3
; RV64I-NEXT: ld ra, 8(sp)
; RV64I-NEXT: .cfi_restore ra
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: .cfi_def_cfa_offset 0
; RV64I-NEXT: ret
;
; RV64IM-LABEL: fold_urem_positive_odd:
; RV64IM: # %bb.0:
; RV64IM-NEXT: slli a0, a0, 32
; RV64IM-NEXT: srli a0, a0, 32
; RV64IM-NEXT: lui a1, 1423
; RV64IM-NEXT: addiw a1, a1, -733
; RV64IM-NEXT: slli a1, a1, 15
; RV64IM-NEXT: addi a1, a1, 1035
; RV64IM-NEXT: slli a1, a1, 13
; RV64IM-NEXT: addi a1, a1, -1811
; RV64IM-NEXT: slli a1, a1, 12
; RV64IM-NEXT: addi a1, a1, 561
; RV64IM-NEXT: mulhu a1, a0, a1
; RV64IM-NEXT: sub a2, a0, a1
; RV64IM-NEXT: srli a2, a2, 1
; RV64IM-NEXT: add a1, a2, a1
; RV64IM-NEXT: srli a1, a1, 6
; RV64IM-NEXT: addi a2, zero, 95
; RV64IM-NEXT: mul a1, a1, a2
; RV64IM-NEXT: sub a0, a0, a1
; RV64IM-NEXT: .cfi_def_cfa_offset 0
; RV64IM-NEXT: ret
%1 = urem i32 %x, 95
ret i32 %1
}
define i32 @fold_urem_positive_even(i32 %x) {
; RV32I-LABEL: fold_urem_positive_even:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: .cfi_def_cfa_offset 16
; RV32I-NEXT: sw ra, 12(sp)
; RV32I-NEXT: .cfi_offset ra, -4
; RV32I-NEXT: addi a1, zero, 1060
; RV32I-NEXT: call __umodsi3
; RV32I-NEXT: lw ra, 12(sp)
; RV32I-NEXT: .cfi_restore ra
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: .cfi_def_cfa_offset 0
; RV32I-NEXT: ret
;
; RV32IM-LABEL: fold_urem_positive_even:
; RV32IM: # %bb.0:
; RV32IM-NEXT: lui a1, 1012964
; RV32IM-NEXT: addi a1, a1, -61
; RV32IM-NEXT: mulhu a1, a0, a1
; RV32IM-NEXT: srli a1, a1, 10
; RV32IM-NEXT: addi a2, zero, 1060
; RV32IM-NEXT: mul a1, a1, a2
; RV32IM-NEXT: sub a0, a0, a1
; RV32IM-NEXT: .cfi_def_cfa_offset 0
; RV32IM-NEXT: ret
;
; RV64I-LABEL: fold_urem_positive_even:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: .cfi_def_cfa_offset 16
; RV64I-NEXT: sd ra, 8(sp)
; RV64I-NEXT: .cfi_offset ra, -8
; RV64I-NEXT: slli a0, a0, 32
; RV64I-NEXT: srli a0, a0, 32
; RV64I-NEXT: addi a1, zero, 1060
; RV64I-NEXT: call __umoddi3
; RV64I-NEXT: ld ra, 8(sp)
; RV64I-NEXT: .cfi_restore ra
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: .cfi_def_cfa_offset 0
; RV64I-NEXT: ret
;
; RV64IM-LABEL: fold_urem_positive_even:
; RV64IM: # %bb.0:
; RV64IM-NEXT: slli a0, a0, 32
; RV64IM-NEXT: srli a0, a0, 32
; RV64IM-NEXT: lui a1, 1048020
; RV64IM-NEXT: addiw a1, a1, -1793
; RV64IM-NEXT: slli a1, a1, 12
; RV64IM-NEXT: addi a1, a1, 139
; RV64IM-NEXT: slli a1, a1, 14
; RV64IM-NEXT: addi a1, a1, 1793
; RV64IM-NEXT: slli a1, a1, 12
; RV64IM-NEXT: addi a1, a1, -139
; RV64IM-NEXT: mulhu a1, a0, a1
; RV64IM-NEXT: srli a1, a1, 10
; RV64IM-NEXT: addi a2, zero, 1060
; RV64IM-NEXT: mul a1, a1, a2
; RV64IM-NEXT: sub a0, a0, a1
; RV64IM-NEXT: .cfi_def_cfa_offset 0
; RV64IM-NEXT: ret
%1 = urem i32 %x, 1060
ret i32 %1
}
; Don't fold if we can combine urem with udiv.
define i32 @combine_urem_udiv(i32 %x) {
; RV32I-LABEL: combine_urem_udiv:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: .cfi_def_cfa_offset 16
; RV32I-NEXT: sw ra, 12(sp)
; RV32I-NEXT: sw s0, 8(sp)
; RV32I-NEXT: sw s1, 4(sp)
; RV32I-NEXT: .cfi_offset ra, -4
; RV32I-NEXT: .cfi_offset s0, -8
; RV32I-NEXT: .cfi_offset s1, -12
; RV32I-NEXT: mv s0, a0
; RV32I-NEXT: addi a1, zero, 95
; RV32I-NEXT: call __umodsi3
; RV32I-NEXT: mv s1, a0
; RV32I-NEXT: addi a1, zero, 95
; RV32I-NEXT: mv a0, s0
; RV32I-NEXT: call __udivsi3
; RV32I-NEXT: add a0, s1, a0
; RV32I-NEXT: lw s1, 4(sp)
; RV32I-NEXT: lw s0, 8(sp)
; RV32I-NEXT: lw ra, 12(sp)
; RV32I-NEXT: .cfi_restore ra
; RV32I-NEXT: .cfi_restore s0
; RV32I-NEXT: .cfi_restore s1
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: .cfi_def_cfa_offset 0
; RV32I-NEXT: ret
;
; RV32IM-LABEL: combine_urem_udiv:
; RV32IM: # %bb.0:
; RV32IM-NEXT: lui a1, 364242
; RV32IM-NEXT: addi a1, a1, 777
; RV32IM-NEXT: mulhu a1, a0, a1
; RV32IM-NEXT: sub a2, a0, a1
; RV32IM-NEXT: srli a2, a2, 1
; RV32IM-NEXT: add a1, a2, a1
; RV32IM-NEXT: srli a1, a1, 6
; RV32IM-NEXT: addi a2, zero, 95
; RV32IM-NEXT: mul a2, a1, a2
; RV32IM-NEXT: sub a0, a0, a2
; RV32IM-NEXT: add a0, a0, a1
; RV32IM-NEXT: .cfi_def_cfa_offset 0
; RV32IM-NEXT: ret
;
; RV64I-LABEL: combine_urem_udiv:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -32
; RV64I-NEXT: .cfi_def_cfa_offset 32
; RV64I-NEXT: sd ra, 24(sp)
; RV64I-NEXT: sd s0, 16(sp)
; RV64I-NEXT: sd s1, 8(sp)
; RV64I-NEXT: .cfi_offset ra, -8
; RV64I-NEXT: .cfi_offset s0, -16
; RV64I-NEXT: .cfi_offset s1, -24
; RV64I-NEXT: slli a0, a0, 32
; RV64I-NEXT: srli s0, a0, 32
; RV64I-NEXT: addi a1, zero, 95
; RV64I-NEXT: mv a0, s0
; RV64I-NEXT: call __umoddi3
; RV64I-NEXT: mv s1, a0
; RV64I-NEXT: addi a1, zero, 95
; RV64I-NEXT: mv a0, s0
; RV64I-NEXT: call __udivdi3
; RV64I-NEXT: add a0, s1, a0
; RV64I-NEXT: ld s1, 8(sp)
; RV64I-NEXT: ld s0, 16(sp)
; RV64I-NEXT: ld ra, 24(sp)
; RV64I-NEXT: .cfi_restore ra
; RV64I-NEXT: .cfi_restore s0
; RV64I-NEXT: .cfi_restore s1
; RV64I-NEXT: addi sp, sp, 32
; RV64I-NEXT: .cfi_def_cfa_offset 0
; RV64I-NEXT: ret
;
; RV64IM-LABEL: combine_urem_udiv:
; RV64IM: # %bb.0:
; RV64IM-NEXT: slli a0, a0, 32
; RV64IM-NEXT: srli a0, a0, 32
; RV64IM-NEXT: lui a1, 1423
; RV64IM-NEXT: addiw a1, a1, -733
; RV64IM-NEXT: slli a1, a1, 15
; RV64IM-NEXT: addi a1, a1, 1035
; RV64IM-NEXT: slli a1, a1, 13
; RV64IM-NEXT: addi a1, a1, -1811
; RV64IM-NEXT: slli a1, a1, 12
; RV64IM-NEXT: addi a1, a1, 561
; RV64IM-NEXT: mulhu a1, a0, a1
; RV64IM-NEXT: sub a2, a0, a1
; RV64IM-NEXT: srli a2, a2, 1
; RV64IM-NEXT: add a1, a2, a1
; RV64IM-NEXT: srli a1, a1, 6
; RV64IM-NEXT: addi a2, zero, 95
; RV64IM-NEXT: mul a2, a1, a2
; RV64IM-NEXT: sub a0, a0, a2
; RV64IM-NEXT: add a0, a0, a1
; RV64IM-NEXT: .cfi_def_cfa_offset 0
; RV64IM-NEXT: ret
%1 = urem i32 %x, 95
%2 = udiv i32 %x, 95
%3 = add i32 %1, %2
ret i32 %3
}
; Don't fold for divisors that are a power of two.
define i32 @dont_fold_urem_power_of_two(i32 %x) {
; CHECK-LABEL: dont_fold_urem_power_of_two:
; CHECK: # %bb.0:
; CHECK-NEXT: andi a0, a0, 63
; CHECK-NEXT: .cfi_def_cfa_offset 0
; CHECK-NEXT: ret
%1 = urem i32 %x, 64
ret i32 %1
}
; Don't fold if the divisor is one.
define i32 @dont_fold_urem_one(i32 %x) {
; CHECK-LABEL: dont_fold_urem_one:
; CHECK: # %bb.0:
; CHECK-NEXT: mv a0, zero
; CHECK-NEXT: .cfi_def_cfa_offset 0
; CHECK-NEXT: ret
%1 = urem i32 %x, 1
ret i32 %1
}
; Don't fold if the divisor is 2^32.
define i32 @dont_fold_urem_i32_umax(i32 %x) {
; CHECK-LABEL: dont_fold_urem_i32_umax:
; CHECK: # %bb.0:
; CHECK-NEXT: .cfi_def_cfa_offset 0
; CHECK-NEXT: ret
%1 = urem i32 %x, 4294967296
ret i32 %1
}
; Don't fold i64 urem
define i64 @dont_fold_urem_i64(i64 %x) {
; RV32I-LABEL: dont_fold_urem_i64:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: .cfi_def_cfa_offset 16
; RV32I-NEXT: sw ra, 12(sp)
; RV32I-NEXT: .cfi_offset ra, -4
; RV32I-NEXT: addi a2, zero, 98
; RV32I-NEXT: mv a3, zero
; RV32I-NEXT: call __umoddi3
; RV32I-NEXT: lw ra, 12(sp)
; RV32I-NEXT: .cfi_restore ra
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: .cfi_def_cfa_offset 0
; RV32I-NEXT: ret
;
; RV32IM-LABEL: dont_fold_urem_i64:
; RV32IM: # %bb.0:
; RV32IM-NEXT: addi sp, sp, -16
; RV32IM-NEXT: .cfi_def_cfa_offset 16
; RV32IM-NEXT: sw ra, 12(sp)
; RV32IM-NEXT: .cfi_offset ra, -4
; RV32IM-NEXT: addi a2, zero, 98
; RV32IM-NEXT: mv a3, zero
; RV32IM-NEXT: call __umoddi3
; RV32IM-NEXT: lw ra, 12(sp)
; RV32IM-NEXT: .cfi_restore ra
; RV32IM-NEXT: addi sp, sp, 16
; RV32IM-NEXT: .cfi_def_cfa_offset 0
; RV32IM-NEXT: ret
;
; RV64I-LABEL: dont_fold_urem_i64:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: .cfi_def_cfa_offset 16
; RV64I-NEXT: sd ra, 8(sp)
; RV64I-NEXT: .cfi_offset ra, -8
; RV64I-NEXT: addi a1, zero, 98
; RV64I-NEXT: call __umoddi3
; RV64I-NEXT: ld ra, 8(sp)
; RV64I-NEXT: .cfi_restore ra
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: .cfi_def_cfa_offset 0
; RV64I-NEXT: ret
;
; RV64IM-LABEL: dont_fold_urem_i64:
; RV64IM: # %bb.0:
; RV64IM-NEXT: srli a1, a0, 1
; RV64IM-NEXT: lui a2, 2675
; RV64IM-NEXT: addiw a2, a2, -251
; RV64IM-NEXT: slli a2, a2, 13
; RV64IM-NEXT: addi a2, a2, 1839
; RV64IM-NEXT: slli a2, a2, 13
; RV64IM-NEXT: addi a2, a2, 167
; RV64IM-NEXT: slli a2, a2, 13
; RV64IM-NEXT: addi a2, a2, 1505
; RV64IM-NEXT: mulhu a1, a1, a2
; RV64IM-NEXT: srli a1, a1, 4
; RV64IM-NEXT: addi a2, zero, 98
; RV64IM-NEXT: mul a1, a1, a2
; RV64IM-NEXT: sub a0, a0, a1
; RV64IM-NEXT: .cfi_def_cfa_offset 0
; RV64IM-NEXT: ret
%1 = urem i64 %x, 98
ret i64 %1
}
|