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
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
; RUN: llc -verify-machineinstrs -csky-no-aliases -mattr=+e2 < %s -mtriple=csky | FileCheck %s
; RUN: llc -verify-machineinstrs -csky-no-aliases < %s -mtriple=csky | FileCheck %s --check-prefix=GENERIC
;; This file shows if a multiplication can be simplified to an addition/subtraction
;; of left shifts.
define i32 @mul_i32_4097(i32 %x) {
; CHECK-LABEL: mul_i32_4097:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: lsli16 a1, a0, 12
; CHECK-NEXT: addu16 a0, a1
; CHECK-NEXT: rts16
;
; GENERIC-LABEL: mul_i32_4097:
; GENERIC: # %bb.0: # %entry
; GENERIC-NEXT: .cfi_def_cfa_offset 0
; GENERIC-NEXT: subi16 sp, sp, 4
; GENERIC-NEXT: .cfi_def_cfa_offset 4
; GENERIC-NEXT: lsli16 a1, a0, 12
; GENERIC-NEXT: addu16 a0, a1, a0
; GENERIC-NEXT: addi16 sp, sp, 4
; GENERIC-NEXT: rts16
entry:
%y = mul nsw i32 %x, 4097
ret i32 %y
}
define i32 @mul_i32_4095(i32 %x) {
; CHECK-LABEL: mul_i32_4095:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: lsli16 a1, a0, 12
; CHECK-NEXT: subu16 a0, a1, a0
; CHECK-NEXT: rts16
;
; GENERIC-LABEL: mul_i32_4095:
; GENERIC: # %bb.0: # %entry
; GENERIC-NEXT: .cfi_def_cfa_offset 0
; GENERIC-NEXT: subi16 sp, sp, 4
; GENERIC-NEXT: .cfi_def_cfa_offset 4
; GENERIC-NEXT: lsli16 a1, a0, 12
; GENERIC-NEXT: subu16 a0, a1, a0
; GENERIC-NEXT: addi16 sp, sp, 4
; GENERIC-NEXT: rts16
entry:
%y = mul nsw i32 %x, 4095
ret i32 %y
}
define i32 @mul_i32_minus_4095(i32 %x) {
; CHECK-LABEL: mul_i32_minus_4095:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: lsli16 a1, a0, 12
; CHECK-NEXT: subu16 a0, a1
; CHECK-NEXT: rts16
;
; GENERIC-LABEL: mul_i32_minus_4095:
; GENERIC: # %bb.0: # %entry
; GENERIC-NEXT: .cfi_def_cfa_offset 0
; GENERIC-NEXT: subi16 sp, sp, 4
; GENERIC-NEXT: .cfi_def_cfa_offset 4
; GENERIC-NEXT: lsli16 a1, a0, 12
; GENERIC-NEXT: subu16 a0, a0, a1
; GENERIC-NEXT: addi16 sp, sp, 4
; GENERIC-NEXT: rts16
entry:
%y = mul nsw i32 %x, -4095
ret i32 %y
}
define i16 @mul_i16_4097(i16 %x) {
; CHECK-LABEL: mul_i16_4097:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: lsli16 a1, a0, 12
; CHECK-NEXT: addu16 a0, a1
; CHECK-NEXT: rts16
;
; GENERIC-LABEL: mul_i16_4097:
; GENERIC: # %bb.0: # %entry
; GENERIC-NEXT: .cfi_def_cfa_offset 0
; GENERIC-NEXT: subi16 sp, sp, 4
; GENERIC-NEXT: .cfi_def_cfa_offset 4
; GENERIC-NEXT: lsli16 a1, a0, 12
; GENERIC-NEXT: addu16 a0, a1, a0
; GENERIC-NEXT: addi16 sp, sp, 4
; GENERIC-NEXT: rts16
entry:
%y = mul nsw i16 %x, 4097
ret i16 %y
}
define i16 @mul_i16_4095(i16 %x) {
; CHECK-LABEL: mul_i16_4095:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: lsli16 a1, a0, 12
; CHECK-NEXT: subu16 a0, a1, a0
; CHECK-NEXT: rts16
;
; GENERIC-LABEL: mul_i16_4095:
; GENERIC: # %bb.0: # %entry
; GENERIC-NEXT: .cfi_def_cfa_offset 0
; GENERIC-NEXT: subi16 sp, sp, 4
; GENERIC-NEXT: .cfi_def_cfa_offset 4
; GENERIC-NEXT: lsli16 a1, a0, 12
; GENERIC-NEXT: subu16 a0, a1, a0
; GENERIC-NEXT: addi16 sp, sp, 4
; GENERIC-NEXT: rts16
entry:
%y = mul nsw i16 %x, 4095
ret i16 %y
}
define i16 @mul_i16_minus_4095(i16 %x) {
; CHECK-LABEL: mul_i16_minus_4095:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: lsli16 a1, a0, 12
; CHECK-NEXT: subu16 a0, a1
; CHECK-NEXT: rts16
;
; GENERIC-LABEL: mul_i16_minus_4095:
; GENERIC: # %bb.0: # %entry
; GENERIC-NEXT: .cfi_def_cfa_offset 0
; GENERIC-NEXT: subi16 sp, sp, 4
; GENERIC-NEXT: .cfi_def_cfa_offset 4
; GENERIC-NEXT: lsli16 a1, a0, 12
; GENERIC-NEXT: subu16 a0, a0, a1
; GENERIC-NEXT: addi16 sp, sp, 4
; GENERIC-NEXT: rts16
entry:
%y = mul nsw i16 %x, -4095
ret i16 %y
}
define i8 @mul_i8_65(i8 %x) {
; CHECK-LABEL: mul_i8_65:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: lsli16 a1, a0, 6
; CHECK-NEXT: addu16 a0, a1
; CHECK-NEXT: rts16
;
; GENERIC-LABEL: mul_i8_65:
; GENERIC: # %bb.0: # %entry
; GENERIC-NEXT: .cfi_def_cfa_offset 0
; GENERIC-NEXT: subi16 sp, sp, 4
; GENERIC-NEXT: .cfi_def_cfa_offset 4
; GENERIC-NEXT: lsli16 a1, a0, 6
; GENERIC-NEXT: addu16 a0, a1, a0
; GENERIC-NEXT: addi16 sp, sp, 4
; GENERIC-NEXT: rts16
entry:
%y = mul nsw i8 %x, 65
ret i8 %y
}
define i8 @mul_i8_63(i8 %x) {
; CHECK-LABEL: mul_i8_63:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: lsli16 a1, a0, 6
; CHECK-NEXT: subu16 a0, a1, a0
; CHECK-NEXT: rts16
;
; GENERIC-LABEL: mul_i8_63:
; GENERIC: # %bb.0: # %entry
; GENERIC-NEXT: .cfi_def_cfa_offset 0
; GENERIC-NEXT: subi16 sp, sp, 4
; GENERIC-NEXT: .cfi_def_cfa_offset 4
; GENERIC-NEXT: lsli16 a1, a0, 6
; GENERIC-NEXT: subu16 a0, a1, a0
; GENERIC-NEXT: addi16 sp, sp, 4
; GENERIC-NEXT: rts16
entry:
%y = mul nsw i8 %x, 63
ret i8 %y
}
define i8 @mul_i8_minus_63(i8 %x) {
; CHECK-LABEL: mul_i8_minus_63:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: lsli16 a1, a0, 6
; CHECK-NEXT: subu16 a0, a1
; CHECK-NEXT: rts16
;
; GENERIC-LABEL: mul_i8_minus_63:
; GENERIC: # %bb.0: # %entry
; GENERIC-NEXT: .cfi_def_cfa_offset 0
; GENERIC-NEXT: subi16 sp, sp, 4
; GENERIC-NEXT: .cfi_def_cfa_offset 4
; GENERIC-NEXT: lsli16 a1, a0, 6
; GENERIC-NEXT: subu16 a0, a0, a1
; GENERIC-NEXT: addi16 sp, sp, 4
; GENERIC-NEXT: rts16
entry:
%y = mul nsw i8 %x, -63
ret i8 %y
}
define i32 @mul_i32_minus_4097(i32 %x) {
; CHECK-LABEL: mul_i32_minus_4097:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: movih32 a1, 65535
; CHECK-NEXT: ori32 a1, a1, 61439
; CHECK-NEXT: mult16 a0, a1
; CHECK-NEXT: rts16
;
; GENERIC-LABEL: mul_i32_minus_4097:
; GENERIC: # %bb.0: # %entry
; GENERIC-NEXT: .cfi_def_cfa_offset 0
; GENERIC-NEXT: subi16 sp, sp, 4
; GENERIC-NEXT: .cfi_def_cfa_offset 4
; GENERIC-NEXT: lsli16 a1, a0, 12
; GENERIC-NEXT: addu16 a0, a1, a0
; GENERIC-NEXT: movi16 a1, 0
; GENERIC-NEXT: subu16 a0, a1, a0
; GENERIC-NEXT: addi16 sp, sp, 4
; GENERIC-NEXT: rts16
entry:
%y = mul nsw i32 %x, -4097
ret i32 %y
}
define i16 @mul_i16_minus_4097(i16 %x) {
; CHECK-LABEL: mul_i16_minus_4097:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: movih32 a1, 65535
; CHECK-NEXT: ori32 a1, a1, 61439
; CHECK-NEXT: mult16 a0, a1
; CHECK-NEXT: rts16
;
; GENERIC-LABEL: mul_i16_minus_4097:
; GENERIC: # %bb.0: # %entry
; GENERIC-NEXT: .cfi_def_cfa_offset 0
; GENERIC-NEXT: subi16 sp, sp, 4
; GENERIC-NEXT: .cfi_def_cfa_offset 4
; GENERIC-NEXT: lsli16 a1, a0, 12
; GENERIC-NEXT: addu16 a0, a1, a0
; GENERIC-NEXT: movi16 a1, 0
; GENERIC-NEXT: subu16 a0, a1, a0
; GENERIC-NEXT: addi16 sp, sp, 4
; GENERIC-NEXT: rts16
entry:
%y = mul nsw i16 %x, -4097
ret i16 %y
}
define i8 @mul_i8_minus_65(i8 %x) {
; CHECK-LABEL: mul_i8_minus_65:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: movih32 a1, 65535
; CHECK-NEXT: ori32 a1, a1, 65471
; CHECK-NEXT: mult16 a0, a1
; CHECK-NEXT: rts16
;
; GENERIC-LABEL: mul_i8_minus_65:
; GENERIC: # %bb.0: # %entry
; GENERIC-NEXT: .cfi_def_cfa_offset 0
; GENERIC-NEXT: subi16 sp, sp, 4
; GENERIC-NEXT: .cfi_def_cfa_offset 4
; GENERIC-NEXT: lsli16 a1, a0, 6
; GENERIC-NEXT: addu16 a0, a1, a0
; GENERIC-NEXT: movi16 a1, 0
; GENERIC-NEXT: subu16 a0, a1, a0
; GENERIC-NEXT: addi16 sp, sp, 4
; GENERIC-NEXT: rts16
entry:
%y = mul nsw i8 %x, -65
ret i8 %y
}
;; This case can not be optimized, due to the data type exceeds.
define i64 @mul_i64_4097(i64 %x) {
; CHECK-LABEL: mul_i64_4097:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: subi16 sp, sp, 4
; CHECK-NEXT: .cfi_def_cfa_offset 4
; CHECK-NEXT: st32.w lr, (sp, 0) # 4-byte Folded Spill
; CHECK-NEXT: .cfi_offset lr, -4
; CHECK-NEXT: .cfi_def_cfa_offset 4
; CHECK-NEXT: lrw32 t0, [.LCPI12_0]
; CHECK-NEXT: movi32 a2, 4097
; CHECK-NEXT: movi16 a3, 0
; CHECK-NEXT: jsr16 t0
; CHECK-NEXT: ld32.w lr, (sp, 0) # 4-byte Folded Reload
; CHECK-NEXT: addi16 sp, sp, 4
; CHECK-NEXT: rts16
; CHECK-NEXT: .p2align 1
; CHECK-NEXT: # %bb.1:
; CHECK-NEXT: .p2align 2, 0x0
; CHECK-NEXT: .LCPI12_0:
; CHECK-NEXT: .long __muldi3
;
; GENERIC-LABEL: mul_i64_4097:
; GENERIC: # %bb.0: # %entry
; GENERIC-NEXT: subi16 sp, sp, 8
; GENERIC-NEXT: .cfi_def_cfa_offset 8
; GENERIC-NEXT: st16.w l0, (sp, 4) # 4-byte Folded Spill
; GENERIC-NEXT: st32.w lr, (sp, 0) # 4-byte Folded Spill
; GENERIC-NEXT: .cfi_offset l0, -4
; GENERIC-NEXT: .cfi_offset lr, -8
; GENERIC-NEXT: subi16 sp, sp, 4
; GENERIC-NEXT: .cfi_def_cfa_offset 12
; GENERIC-NEXT: movi16 a2, 0
; GENERIC-NEXT: lsli16 a3, a2, 24
; GENERIC-NEXT: lsli16 a2, a2, 16
; GENERIC-NEXT: or16 a2, a3
; GENERIC-NEXT: movi16 a3, 16
; GENERIC-NEXT: lsli16 a3, a3, 8
; GENERIC-NEXT: or16 a3, a2
; GENERIC-NEXT: movi16 a2, 1
; GENERIC-NEXT: or16 a2, a3
; GENERIC-NEXT: lrw32 l0, [.LCPI12_0]
; GENERIC-NEXT: movi16 a3, 0
; GENERIC-NEXT: jsr16 l0
; GENERIC-NEXT: addi16 sp, sp, 4
; GENERIC-NEXT: ld32.w lr, (sp, 0) # 4-byte Folded Reload
; GENERIC-NEXT: ld16.w l0, (sp, 4) # 4-byte Folded Reload
; GENERIC-NEXT: addi16 sp, sp, 8
; GENERIC-NEXT: rts16
; GENERIC-NEXT: .p2align 1
; GENERIC-NEXT: # %bb.1:
; GENERIC-NEXT: .p2align 2, 0x0
; GENERIC-NEXT: .LCPI12_0:
; GENERIC-NEXT: .long __muldi3
entry:
%y = mul nsw i64 %x, 4097
ret i64 %y
}
|
