and2.ll
6.97 KB
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
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -instcombine -S | FileCheck %s
define i1 @test2(i1 %X, i1 %Y) {
; CHECK-LABEL: @test2(
; CHECK-NEXT: [[A:%.*]] = and i1 [[X:%.*]], [[Y:%.*]]
; CHECK-NEXT: ret i1 [[A]]
;
%a = and i1 %X, %Y
%b = and i1 %a, %X
ret i1 %b
}
define i32 @test3(i32 %X, i32 %Y) {
; CHECK-LABEL: @test3(
; CHECK-NEXT: [[A:%.*]] = and i32 [[X:%.*]], [[Y:%.*]]
; CHECK-NEXT: ret i32 [[A]]
;
%a = and i32 %X, %Y
%b = and i32 %Y, %a
ret i32 %b
}
define i1 @test7(i32 %i, i1 %b) {
; CHECK-LABEL: @test7(
; CHECK-NEXT: [[TMP1:%.*]] = icmp eq i32 [[I:%.*]], 0
; CHECK-NEXT: [[TMP2:%.*]] = and i1 [[TMP1]], [[B:%.*]]
; CHECK-NEXT: ret i1 [[TMP2]]
;
%cmp1 = icmp slt i32 %i, 1
%cmp2 = icmp sgt i32 %i, -1
%and1 = and i1 %cmp1, %b
%and2 = and i1 %and1, %cmp2
ret i1 %and2
}
define i1 @test8(i32 %i) {
; CHECK-LABEL: @test8(
; CHECK-NEXT: [[I_OFF:%.*]] = add i32 [[I:%.*]], -1
; CHECK-NEXT: [[TMP1:%.*]] = icmp ult i32 [[I_OFF]], 13
; CHECK-NEXT: ret i1 [[TMP1]]
;
%cmp1 = icmp ne i32 %i, 0
%cmp2 = icmp ult i32 %i, 14
%cond = and i1 %cmp1, %cmp2
ret i1 %cond
}
; FIXME: Vectors should fold too.
define <2 x i1> @test8vec(<2 x i32> %i) {
; CHECK-LABEL: @test8vec(
; CHECK-NEXT: [[CMP1:%.*]] = icmp ne <2 x i32> [[I:%.*]], zeroinitializer
; CHECK-NEXT: [[CMP2:%.*]] = icmp ult <2 x i32> [[I]], <i32 14, i32 14>
; CHECK-NEXT: [[COND:%.*]] = and <2 x i1> [[CMP1]], [[CMP2]]
; CHECK-NEXT: ret <2 x i1> [[COND]]
;
%cmp1 = icmp ne <2 x i32> %i, zeroinitializer
%cmp2 = icmp ult <2 x i32> %i, <i32 14, i32 14>
%cond = and <2 x i1> %cmp1, %cmp2
ret <2 x i1> %cond
}
; combine -x & 1 into x & 1
define i64 @test9(i64 %x) {
; CHECK-LABEL: @test9(
; CHECK-NEXT: [[AND:%.*]] = and i64 [[X:%.*]], 1
; CHECK-NEXT: ret i64 [[AND]]
;
%sub = sub nsw i64 0, %x
%and = and i64 %sub, 1
ret i64 %and
}
; combine -x & 1 into x & 1
define <2 x i64> @test9vec(<2 x i64> %x) {
; CHECK-LABEL: @test9vec(
; CHECK-NEXT: [[AND:%.*]] = and <2 x i64> [[X:%.*]], <i64 1, i64 1>
; CHECK-NEXT: ret <2 x i64> [[AND]]
;
%sub = sub nsw <2 x i64> <i64 0, i64 0>, %x
%and = and <2 x i64> %sub, <i64 1, i64 1>
ret <2 x i64> %and
}
define i64 @test10(i64 %x) {
; CHECK-LABEL: @test10(
; CHECK-NEXT: [[TMP1:%.*]] = and i64 [[X:%.*]], -2
; CHECK-NEXT: [[ADD:%.*]] = sub i64 0, [[TMP1]]
; CHECK-NEXT: ret i64 [[ADD]]
;
%sub = sub nsw i64 0, %x
%and = and i64 %sub, 1
%add = add i64 %sub, %and
ret i64 %add
}
; (1 << x) & 1 --> zext(x == 0)
define i8 @and1_shl1_is_cmp_eq_0(i8 %x) {
; CHECK-LABEL: @and1_shl1_is_cmp_eq_0(
; CHECK-NEXT: [[TMP1:%.*]] = icmp eq i8 [[X:%.*]], 0
; CHECK-NEXT: [[AND:%.*]] = zext i1 [[TMP1]] to i8
; CHECK-NEXT: ret i8 [[AND]]
;
%sh = shl i8 1, %x
%and = and i8 %sh, 1
ret i8 %and
}
; Don't do it if the shift has another use.
define i8 @and1_shl1_is_cmp_eq_0_multiuse(i8 %x) {
; CHECK-LABEL: @and1_shl1_is_cmp_eq_0_multiuse(
; CHECK-NEXT: [[SH:%.*]] = shl i8 1, [[X:%.*]]
; CHECK-NEXT: [[AND:%.*]] = and i8 [[SH]], 1
; CHECK-NEXT: [[ADD:%.*]] = add i8 [[SH]], [[AND]]
; CHECK-NEXT: ret i8 [[ADD]]
;
%sh = shl i8 1, %x
%and = and i8 %sh, 1
%add = add i8 %sh, %and
ret i8 %add
}
; (1 << x) & 1 --> zext(x == 0)
define <2 x i8> @and1_shl1_is_cmp_eq_0_vec(<2 x i8> %x) {
; CHECK-LABEL: @and1_shl1_is_cmp_eq_0_vec(
; CHECK-NEXT: [[TMP1:%.*]] = icmp eq <2 x i8> [[X:%.*]], zeroinitializer
; CHECK-NEXT: [[AND:%.*]] = zext <2 x i1> [[TMP1]] to <2 x i8>
; CHECK-NEXT: ret <2 x i8> [[AND]]
;
%sh = shl <2 x i8> <i8 1, i8 1>, %x
%and = and <2 x i8> %sh, <i8 1, i8 1>
ret <2 x i8> %and
}
; (1 >> x) & 1 --> zext(x == 0)
define i8 @and1_lshr1_is_cmp_eq_0(i8 %x) {
; CHECK-LABEL: @and1_lshr1_is_cmp_eq_0(
; CHECK-NEXT: [[TMP1:%.*]] = icmp eq i8 [[X:%.*]], 0
; CHECK-NEXT: [[AND:%.*]] = zext i1 [[TMP1]] to i8
; CHECK-NEXT: ret i8 [[AND]]
;
%sh = lshr i8 1, %x
%and = and i8 %sh, 1
ret i8 %and
}
; Don't do it if the shift has another use.
define i8 @and1_lshr1_is_cmp_eq_0_multiuse(i8 %x) {
; CHECK-LABEL: @and1_lshr1_is_cmp_eq_0_multiuse(
; CHECK-NEXT: [[SH:%.*]] = lshr i8 1, [[X:%.*]]
; CHECK-NEXT: [[AND:%.*]] = and i8 [[SH]], 1
; CHECK-NEXT: [[ADD:%.*]] = add nuw nsw i8 [[SH]], [[AND]]
; CHECK-NEXT: ret i8 [[ADD]]
;
%sh = lshr i8 1, %x
%and = and i8 %sh, 1
%add = add i8 %sh, %and
ret i8 %add
}
; (1 >> x) & 1 --> zext(x == 0)
define <2 x i8> @and1_lshr1_is_cmp_eq_0_vec(<2 x i8> %x) {
; CHECK-LABEL: @and1_lshr1_is_cmp_eq_0_vec(
; CHECK-NEXT: [[TMP1:%.*]] = icmp eq <2 x i8> [[X:%.*]], zeroinitializer
; CHECK-NEXT: [[AND:%.*]] = zext <2 x i1> [[TMP1]] to <2 x i8>
; CHECK-NEXT: ret <2 x i8> [[AND]]
;
%sh = lshr <2 x i8> <i8 1, i8 1>, %x
%and = and <2 x i8> %sh, <i8 1, i8 1>
ret <2 x i8> %and
}
; The add in this test is unnecessary because the LSBs of the LHS are 0 and the 'and' only consumes bits from those LSBs. It doesn't matter what happens to the upper bits.
define i32 @test11(i32 %a, i32 %b) {
; CHECK-LABEL: @test11(
; CHECK-NEXT: [[X:%.*]] = shl i32 [[A:%.*]], 8
; CHECK-NEXT: [[Z:%.*]] = and i32 [[B:%.*]], 128
; CHECK-NEXT: [[W:%.*]] = mul i32 [[Z]], [[X]]
; CHECK-NEXT: ret i32 [[W]]
;
%x = shl i32 %a, 8
%y = add i32 %x, %b
%z = and i32 %y, 128
%w = mul i32 %z, %x ; to keep the shift from being removed
ret i32 %w
}
; The add in this test is unnecessary because the LSBs of the RHS are 0 and the 'and' only consumes bits from those LSBs. It doesn't matter what happens to the upper bits.
define i32 @test12(i32 %a, i32 %b) {
; CHECK-LABEL: @test12(
; CHECK-NEXT: [[X:%.*]] = shl i32 [[A:%.*]], 8
; CHECK-NEXT: [[Z:%.*]] = and i32 [[B:%.*]], 128
; CHECK-NEXT: [[W:%.*]] = mul i32 [[Z]], [[X]]
; CHECK-NEXT: ret i32 [[W]]
;
%x = shl i32 %a, 8
%y = add i32 %b, %x
%z = and i32 %y, 128
%w = mul i32 %z, %x ; to keep the shift from being removed
ret i32 %w
}
; The sub in this test is unnecessary because the LSBs of the RHS are 0 and the 'and' only consumes bits from those LSBs. It doesn't matter what happens to the upper bits.
define i32 @test13(i32 %a, i32 %b) {
; CHECK-LABEL: @test13(
; CHECK-NEXT: [[X:%.*]] = shl i32 [[A:%.*]], 8
; CHECK-NEXT: [[Z:%.*]] = and i32 [[B:%.*]], 128
; CHECK-NEXT: [[W:%.*]] = mul i32 [[Z]], [[X]]
; CHECK-NEXT: ret i32 [[W]]
;
%x = shl i32 %a, 8
%y = sub i32 %b, %x
%z = and i32 %y, 128
%w = mul i32 %z, %x ; to keep the shift from being removed
ret i32 %w
}
; The sub in this test cannot be removed because we need to keep the negation of %b. TODO: But we should be able to replace the LHS of it with a 0.
define i32 @test14(i32 %a, i32 %b) {
; CHECK-LABEL: @test14(
; CHECK-NEXT: [[X:%.*]] = shl i32 [[A:%.*]], 8
; CHECK-NEXT: [[Y:%.*]] = sub i32 0, [[B:%.*]]
; CHECK-NEXT: [[Z:%.*]] = and i32 [[Y]], 128
; CHECK-NEXT: [[W:%.*]] = mul i32 [[Z]], [[X]]
; CHECK-NEXT: ret i32 [[W]]
;
%x = shl i32 %a, 8
%y = sub i32 %x, %b
%z = and i32 %y, 128
%w = mul i32 %z, %x ; to keep the shift from being removed
ret i32 %w
}