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
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=riscv32 -disable-block-placement -verify-machineinstrs < %s \
; RUN: | FileCheck -check-prefixes=RV32I %s
; RUN: llc -mtriple=riscv64 -disable-block-placement -verify-machineinstrs < %s \
; RUN: | FileCheck -check-prefixes=RV64I %s
define signext i32 @foo(i32 signext %a, ptr %b) nounwind {
; RV32I-LABEL: foo:
; RV32I: # %bb.0:
; RV32I-NEXT: lw a2, 0(a1)
; RV32I-NEXT: beq a0, a2, .LBB0_2
; RV32I-NEXT: # %bb.1:
; RV32I-NEXT: mv a0, a2
; RV32I-NEXT: .LBB0_2:
; RV32I-NEXT: lw a2, 0(a1)
; RV32I-NEXT: bne a0, a2, .LBB0_4
; RV32I-NEXT: # %bb.3:
; RV32I-NEXT: mv a0, a2
; RV32I-NEXT: .LBB0_4:
; RV32I-NEXT: lw a2, 0(a1)
; RV32I-NEXT: bltu a2, a0, .LBB0_6
; RV32I-NEXT: # %bb.5:
; RV32I-NEXT: mv a0, a2
; RV32I-NEXT: .LBB0_6:
; RV32I-NEXT: lw a2, 0(a1)
; RV32I-NEXT: bgeu a0, a2, .LBB0_8
; RV32I-NEXT: # %bb.7:
; RV32I-NEXT: mv a0, a2
; RV32I-NEXT: .LBB0_8:
; RV32I-NEXT: lw a2, 0(a1)
; RV32I-NEXT: bltu a0, a2, .LBB0_10
; RV32I-NEXT: # %bb.9:
; RV32I-NEXT: mv a0, a2
; RV32I-NEXT: .LBB0_10:
; RV32I-NEXT: lw a2, 0(a1)
; RV32I-NEXT: bgeu a2, a0, .LBB0_12
; RV32I-NEXT: # %bb.11:
; RV32I-NEXT: mv a0, a2
; RV32I-NEXT: .LBB0_12:
; RV32I-NEXT: lw a2, 0(a1)
; RV32I-NEXT: blt a2, a0, .LBB0_14
; RV32I-NEXT: # %bb.13:
; RV32I-NEXT: mv a0, a2
; RV32I-NEXT: .LBB0_14:
; RV32I-NEXT: lw a2, 0(a1)
; RV32I-NEXT: bge a0, a2, .LBB0_16
; RV32I-NEXT: # %bb.15:
; RV32I-NEXT: mv a0, a2
; RV32I-NEXT: .LBB0_16:
; RV32I-NEXT: lw a2, 0(a1)
; RV32I-NEXT: blt a0, a2, .LBB0_18
; RV32I-NEXT: # %bb.17:
; RV32I-NEXT: mv a0, a2
; RV32I-NEXT: .LBB0_18:
; RV32I-NEXT: lw a2, 0(a1)
; RV32I-NEXT: bge a2, a0, .LBB0_20
; RV32I-NEXT: # %bb.19:
; RV32I-NEXT: mv a0, a2
; RV32I-NEXT: .LBB0_20:
; RV32I-NEXT: lw a2, 0(a1)
; RV32I-NEXT: blez a2, .LBB0_22
; RV32I-NEXT: # %bb.21:
; RV32I-NEXT: mv a0, a2
; RV32I-NEXT: .LBB0_22:
; RV32I-NEXT: lw a3, 0(a1)
; RV32I-NEXT: bgez a2, .LBB0_24
; RV32I-NEXT: # %bb.23:
; RV32I-NEXT: mv a0, a3
; RV32I-NEXT: .LBB0_24:
; RV32I-NEXT: lw a3, 0(a1)
; RV32I-NEXT: li a4, 1024
; RV32I-NEXT: blt a4, a3, .LBB0_26
; RV32I-NEXT: # %bb.25:
; RV32I-NEXT: mv a0, a3
; RV32I-NEXT: .LBB0_26:
; RV32I-NEXT: lw a1, 0(a1)
; RV32I-NEXT: li a3, 2046
; RV32I-NEXT: bltu a3, a2, .LBB0_28
; RV32I-NEXT: # %bb.27:
; RV32I-NEXT: mv a0, a1
; RV32I-NEXT: .LBB0_28:
; RV32I-NEXT: ret
;
; RV64I-LABEL: foo:
; RV64I: # %bb.0:
; RV64I-NEXT: lw a2, 0(a1)
; RV64I-NEXT: beq a0, a2, .LBB0_2
; RV64I-NEXT: # %bb.1:
; RV64I-NEXT: mv a0, a2
; RV64I-NEXT: .LBB0_2:
; RV64I-NEXT: lw a2, 0(a1)
; RV64I-NEXT: bne a0, a2, .LBB0_4
; RV64I-NEXT: # %bb.3:
; RV64I-NEXT: mv a0, a2
; RV64I-NEXT: .LBB0_4:
; RV64I-NEXT: lw a2, 0(a1)
; RV64I-NEXT: bltu a2, a0, .LBB0_6
; RV64I-NEXT: # %bb.5:
; RV64I-NEXT: mv a0, a2
; RV64I-NEXT: .LBB0_6:
; RV64I-NEXT: lw a2, 0(a1)
; RV64I-NEXT: bgeu a0, a2, .LBB0_8
; RV64I-NEXT: # %bb.7:
; RV64I-NEXT: mv a0, a2
; RV64I-NEXT: .LBB0_8:
; RV64I-NEXT: lw a2, 0(a1)
; RV64I-NEXT: bltu a0, a2, .LBB0_10
; RV64I-NEXT: # %bb.9:
; RV64I-NEXT: mv a0, a2
; RV64I-NEXT: .LBB0_10:
; RV64I-NEXT: lw a2, 0(a1)
; RV64I-NEXT: bgeu a2, a0, .LBB0_12
; RV64I-NEXT: # %bb.11:
; RV64I-NEXT: mv a0, a2
; RV64I-NEXT: .LBB0_12:
; RV64I-NEXT: lw a2, 0(a1)
; RV64I-NEXT: blt a2, a0, .LBB0_14
; RV64I-NEXT: # %bb.13:
; RV64I-NEXT: mv a0, a2
; RV64I-NEXT: .LBB0_14:
; RV64I-NEXT: lw a2, 0(a1)
; RV64I-NEXT: bge a0, a2, .LBB0_16
; RV64I-NEXT: # %bb.15:
; RV64I-NEXT: mv a0, a2
; RV64I-NEXT: .LBB0_16:
; RV64I-NEXT: lw a2, 0(a1)
; RV64I-NEXT: blt a0, a2, .LBB0_18
; RV64I-NEXT: # %bb.17:
; RV64I-NEXT: mv a0, a2
; RV64I-NEXT: .LBB0_18:
; RV64I-NEXT: lw a2, 0(a1)
; RV64I-NEXT: bge a2, a0, .LBB0_20
; RV64I-NEXT: # %bb.19:
; RV64I-NEXT: mv a0, a2
; RV64I-NEXT: .LBB0_20:
; RV64I-NEXT: lw a2, 0(a1)
; RV64I-NEXT: blez a2, .LBB0_22
; RV64I-NEXT: # %bb.21:
; RV64I-NEXT: mv a0, a2
; RV64I-NEXT: .LBB0_22:
; RV64I-NEXT: lw a3, 0(a1)
; RV64I-NEXT: bgez a2, .LBB0_24
; RV64I-NEXT: # %bb.23:
; RV64I-NEXT: mv a0, a3
; RV64I-NEXT: .LBB0_24:
; RV64I-NEXT: lw a3, 0(a1)
; RV64I-NEXT: li a4, 1024
; RV64I-NEXT: blt a4, a3, .LBB0_26
; RV64I-NEXT: # %bb.25:
; RV64I-NEXT: mv a0, a3
; RV64I-NEXT: .LBB0_26:
; RV64I-NEXT: lw a1, 0(a1)
; RV64I-NEXT: li a3, 2046
; RV64I-NEXT: bltu a3, a2, .LBB0_28
; RV64I-NEXT: # %bb.27:
; RV64I-NEXT: mv a0, a1
; RV64I-NEXT: .LBB0_28:
; RV64I-NEXT: ret
%val1 = load volatile i32, ptr %b
%tst1 = icmp eq i32 %a, %val1
%val2 = select i1 %tst1, i32 %a, i32 %val1
%val3 = load volatile i32, ptr %b
%tst2 = icmp ne i32 %val2, %val3
%val4 = select i1 %tst2, i32 %val2, i32 %val3
%val5 = load volatile i32, ptr %b
%tst3 = icmp ugt i32 %val4, %val5
%val6 = select i1 %tst3, i32 %val4, i32 %val5
%val7 = load volatile i32, ptr %b
%tst4 = icmp uge i32 %val6, %val7
%val8 = select i1 %tst4, i32 %val6, i32 %val7
%val9 = load volatile i32, ptr %b
%tst5 = icmp ult i32 %val8, %val9
%val10 = select i1 %tst5, i32 %val8, i32 %val9
%val11 = load volatile i32, ptr %b
%tst6 = icmp ule i32 %val10, %val11
%val12 = select i1 %tst6, i32 %val10, i32 %val11
%val13 = load volatile i32, ptr %b
%tst7 = icmp sgt i32 %val12, %val13
%val14 = select i1 %tst7, i32 %val12, i32 %val13
%val15 = load volatile i32, ptr %b
%tst8 = icmp sge i32 %val14, %val15
%val16 = select i1 %tst8, i32 %val14, i32 %val15
%val17 = load volatile i32, ptr %b
%tst9 = icmp slt i32 %val16, %val17
%val18 = select i1 %tst9, i32 %val16, i32 %val17
%val19 = load volatile i32, ptr %b
%tst10 = icmp sle i32 %val18, %val19
%val20 = select i1 %tst10, i32 %val18, i32 %val19
%val21 = load volatile i32, ptr %b
%tst11 = icmp slt i32 %val21, 1
%val22 = select i1 %tst11, i32 %val20, i32 %val21
%val23 = load volatile i32, ptr %b
%tst12 = icmp sgt i32 %val21, -1
%val24 = select i1 %tst12, i32 %val22, i32 %val23
%val25 = load volatile i32, ptr %b
%tst13 = icmp sgt i32 %val25, 1024
%val26 = select i1 %tst13, i32 %val24, i32 %val25
%val27 = load volatile i32, ptr %b
%tst14 = icmp ugt i32 %val21, 2046
%val28 = select i1 %tst14, i32 %val26, i32 %val27
ret i32 %val28
}
; Test that we can ComputeNumSignBits across basic blocks when the live out is
; RISCVISD::SELECT_CC. There should be no slli+srai or sext.h in the output.
define signext i16 @numsignbits(i16 signext %0, i16 signext %1, i16 signext %2, i16 signext %3) nounwind {
; RV32I-LABEL: numsignbits:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: sw ra, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s0, 8(sp) # 4-byte Folded Spill
; RV32I-NEXT: mv s0, a3
; RV32I-NEXT: beqz a0, .LBB1_2
; RV32I-NEXT: # %bb.1:
; RV32I-NEXT: mv s0, a2
; RV32I-NEXT: .LBB1_2:
; RV32I-NEXT: beqz a1, .LBB1_4
; RV32I-NEXT: # %bb.3:
; RV32I-NEXT: mv a0, s0
; RV32I-NEXT: call bar@plt
; RV32I-NEXT: .LBB1_4:
; RV32I-NEXT: mv a0, s0
; RV32I-NEXT: lw ra, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s0, 8(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: ret
;
; RV64I-LABEL: numsignbits:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: sd ra, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s0, 0(sp) # 8-byte Folded Spill
; RV64I-NEXT: mv s0, a3
; RV64I-NEXT: beqz a0, .LBB1_2
; RV64I-NEXT: # %bb.1:
; RV64I-NEXT: mv s0, a2
; RV64I-NEXT: .LBB1_2:
; RV64I-NEXT: beqz a1, .LBB1_4
; RV64I-NEXT: # %bb.3:
; RV64I-NEXT: mv a0, s0
; RV64I-NEXT: call bar@plt
; RV64I-NEXT: .LBB1_4:
; RV64I-NEXT: mv a0, s0
; RV64I-NEXT: ld ra, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s0, 0(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: ret
%5 = icmp eq i16 %0, 0
%6 = select i1 %5, i16 %3, i16 %2
%7 = icmp eq i16 %1, 0
br i1 %7, label %9, label %8
8: ; preds = %4
tail call void @bar(i16 signext %6)
br label %9
9: ; preds = %8, %4
ret i16 %6
}
declare void @bar(i16 signext)
define i32 @select_sge_int16min(i32 signext %x, i32 signext %y, i32 signext %z) {
; RV32I-LABEL: select_sge_int16min:
; RV32I: # %bb.0:
; RV32I-NEXT: lui a3, 1048560
; RV32I-NEXT: addi a3, a3, -1
; RV32I-NEXT: blt a3, a0, .LBB2_2
; RV32I-NEXT: # %bb.1:
; RV32I-NEXT: mv a1, a2
; RV32I-NEXT: .LBB2_2:
; RV32I-NEXT: mv a0, a1
; RV32I-NEXT: ret
;
; RV64I-LABEL: select_sge_int16min:
; RV64I: # %bb.0:
; RV64I-NEXT: lui a3, 1048560
; RV64I-NEXT: addiw a3, a3, -1
; RV64I-NEXT: blt a3, a0, .LBB2_2
; RV64I-NEXT: # %bb.1:
; RV64I-NEXT: mv a1, a2
; RV64I-NEXT: .LBB2_2:
; RV64I-NEXT: mv a0, a1
; RV64I-NEXT: ret
%a = icmp sge i32 %x, -65536
%b = select i1 %a, i32 %y, i32 %z
ret i32 %b
}
define i64 @select_sge_int32min(i64 %x, i64 %y, i64 %z) {
; RV32I-LABEL: select_sge_int32min:
; RV32I: # %bb.0:
; RV32I-NEXT: li a6, -1
; RV32I-NEXT: bne a1, a6, .LBB3_2
; RV32I-NEXT: # %bb.1:
; RV32I-NEXT: slti a0, a0, 0
; RV32I-NEXT: j .LBB3_3
; RV32I-NEXT: .LBB3_2:
; RV32I-NEXT: slti a0, a1, 0
; RV32I-NEXT: xori a0, a0, 1
; RV32I-NEXT: .LBB3_3:
; RV32I-NEXT: bnez a0, .LBB3_5
; RV32I-NEXT: # %bb.4:
; RV32I-NEXT: mv a2, a4
; RV32I-NEXT: mv a3, a5
; RV32I-NEXT: .LBB3_5:
; RV32I-NEXT: mv a0, a2
; RV32I-NEXT: mv a1, a3
; RV32I-NEXT: ret
;
; RV64I-LABEL: select_sge_int32min:
; RV64I: # %bb.0:
; RV64I-NEXT: lui a3, 524288
; RV64I-NEXT: addi a3, a3, -1
; RV64I-NEXT: blt a3, a0, .LBB3_2
; RV64I-NEXT: # %bb.1:
; RV64I-NEXT: mv a1, a2
; RV64I-NEXT: .LBB3_2:
; RV64I-NEXT: mv a0, a1
; RV64I-NEXT: ret
%a = icmp sge i64 %x, -2147483648
%b = select i1 %a, i64 %y, i64 %z
ret i64 %b
}
|
