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
|
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -passes=memcpyopt -S %s -verify-memoryssa | FileCheck %s
; memset -> memcpy forwarding, if memcpy is larger than memset, but trailing
; bytes are known to be undef.
%T = type { i64, i32, i32 }
define void @test_alloca(ptr %result) {
; CHECK-LABEL: @test_alloca(
; CHECK-NEXT: [[A:%.*]] = alloca [[T:%.*]], align 8
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 8 [[A]], i8 0, i64 12, i1 false)
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr [[RESULT:%.*]], i8 0, i64 12, i1 false)
; CHECK-NEXT: ret void
;
%a = alloca %T, align 8
call void @llvm.memset.p0.i64(ptr align 8 %a, i8 0, i64 12, i1 false)
call void @llvm.memcpy.p0.p0.i64(ptr %result, ptr align 8 %a, i64 16, i1 false)
ret void
}
define void @test_alloca_with_lifetimes(ptr %result) {
; CHECK-LABEL: @test_alloca_with_lifetimes(
; CHECK-NEXT: [[A:%.*]] = alloca [[T:%.*]], align 8
; CHECK-NEXT: call void @llvm.lifetime.start.p0(i64 16, ptr [[A]])
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 8 [[A]], i8 0, i64 12, i1 false)
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr [[RESULT:%.*]], i8 0, i64 12, i1 false)
; CHECK-NEXT: call void @llvm.lifetime.end.p0(i64 16, ptr [[A]])
; CHECK-NEXT: ret void
;
%a = alloca %T, align 8
call void @llvm.lifetime.start.p0(i64 16, ptr %a)
call void @llvm.memset.p0.i64(ptr align 8 %a, i8 0, i64 12, i1 false)
call void @llvm.memcpy.p0.p0.i64(ptr %result, ptr align 8 %a, i64 16, i1 false)
call void @llvm.lifetime.end.p0(i64 16, ptr %a)
ret void
}
define void @test_malloc_with_lifetimes(ptr %result) {
; CHECK-LABEL: @test_malloc_with_lifetimes(
; CHECK-NEXT: [[A:%.*]] = call ptr @malloc(i64 16)
; CHECK-NEXT: call void @llvm.lifetime.start.p0(i64 16, ptr [[A]])
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 8 [[A]], i8 0, i64 12, i1 false)
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr [[RESULT:%.*]], i8 0, i64 12, i1 false)
; CHECK-NEXT: call void @llvm.lifetime.end.p0(i64 16, ptr [[A]])
; CHECK-NEXT: call void @free(ptr [[A]])
; CHECK-NEXT: ret void
;
%a = call ptr @malloc(i64 16)
call void @llvm.lifetime.start.p0(i64 16, ptr %a)
call void @llvm.memset.p0.i64(ptr align 8 %a, i8 0, i64 12, i1 false)
call void @llvm.memcpy.p0.p0.i64(ptr %result, ptr align 8 %a, i64 16, i1 false)
call void @llvm.lifetime.end.p0(i64 16, ptr %a)
call void @free(ptr %a)
ret void
}
; memcpy size is larger than lifetime, don't optimize.
define void @test_copy_larger_than_lifetime_size(ptr %result) {
; CHECK-LABEL: @test_copy_larger_than_lifetime_size(
; CHECK-NEXT: [[A:%.*]] = call ptr @malloc(i64 16)
; CHECK-NEXT: call void @llvm.lifetime.start.p0(i64 12, ptr [[A]])
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 8 [[A]], i8 0, i64 12, i1 false)
; CHECK-NEXT: call void @llvm.memcpy.p0.p0.i64(ptr [[RESULT:%.*]], ptr align 8 [[A]], i64 16, i1 false)
; CHECK-NEXT: call void @llvm.lifetime.end.p0(i64 12, ptr [[A]])
; CHECK-NEXT: call void @free(ptr [[A]])
; CHECK-NEXT: ret void
;
%a = call ptr @malloc(i64 16)
call void @llvm.lifetime.start.p0(i64 12, ptr %a)
call void @llvm.memset.p0.i64(ptr align 8 %a, i8 0, i64 12, i1 false)
call void @llvm.memcpy.p0.p0.i64(ptr %result, ptr align 8 %a, i64 16, i1 false)
call void @llvm.lifetime.end.p0(i64 12, ptr %a)
call void @free(ptr %a)
ret void
}
; The trailing bytes are not known to be undef, we can't ignore them.
define void @test_not_undef_memory(ptr %result, ptr %input) {
; CHECK-LABEL: @test_not_undef_memory(
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 8 [[INPUT:%.*]], i8 0, i64 12, i1 false)
; CHECK-NEXT: call void @llvm.memcpy.p0.p0.i64(ptr [[RESULT:%.*]], ptr align 8 [[INPUT]], i64 16, i1 false)
; CHECK-NEXT: ret void
;
call void @llvm.memset.p0.i64(ptr align 8 %input, i8 0, i64 12, i1 false)
call void @llvm.memcpy.p0.p0.i64(ptr %result, ptr align 8 %input, i64 16, i1 false)
ret void
}
; Memset is volatile, memcpy is not. Can be optimized.
define void @test_volatile_memset(ptr %result) {
; CHECK-LABEL: @test_volatile_memset(
; CHECK-NEXT: [[A:%.*]] = alloca [[T:%.*]], align 8
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 8 [[A]], i8 0, i64 12, i1 true)
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr [[RESULT:%.*]], i8 0, i64 12, i1 false)
; CHECK-NEXT: ret void
;
%a = alloca %T, align 8
call void @llvm.memset.p0.i64(ptr align 8 %a, i8 0, i64 12, i1 true)
call void @llvm.memcpy.p0.p0.i64(ptr %result, ptr align 8 %a, i64 16, i1 false)
ret void
}
; Memcpy is volatile, memset is not. Cannot be optimized.
define void @test_volatile_memcpy(ptr %result) {
; CHECK-LABEL: @test_volatile_memcpy(
; CHECK-NEXT: [[A:%.*]] = alloca [[T:%.*]], align 8
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 8 [[A]], i8 0, i64 12, i1 false)
; CHECK-NEXT: call void @llvm.memcpy.p0.p0.i64(ptr [[RESULT:%.*]], ptr align 8 [[A]], i64 16, i1 true)
; CHECK-NEXT: ret void
;
%a = alloca %T, align 8
call void @llvm.memset.p0.i64(ptr align 8 %a, i8 0, i64 12, i1 false)
call void @llvm.memcpy.p0.p0.i64(ptr %result, ptr align 8 %a, i64 16, i1 true)
ret void
}
; Write between memset and memcpy, can't optimize.
define void @test_write_between(ptr %result) {
; CHECK-LABEL: @test_write_between(
; CHECK-NEXT: [[A:%.*]] = alloca [[T:%.*]], align 8
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 8 [[A]], i8 0, i64 12, i1 false)
; CHECK-NEXT: store i8 -1, ptr [[A]], align 1
; CHECK-NEXT: call void @llvm.memcpy.p0.p0.i64(ptr [[RESULT:%.*]], ptr align 8 [[A]], i64 16, i1 false)
; CHECK-NEXT: ret void
;
%a = alloca %T, align 8
call void @llvm.memset.p0.i64(ptr align 8 %a, i8 0, i64 12, i1 false)
store i8 -1, ptr %a
call void @llvm.memcpy.p0.p0.i64(ptr %result, ptr align 8 %a, i64 16, i1 false)
ret void
}
; A write prior to the memset, which is part of the memset region.
; We could optimize this, but currently don't, because the used memory location is imprecise.
define void @test_write_before_memset_in_memset_region(ptr %result) {
; CHECK-LABEL: @test_write_before_memset_in_memset_region(
; CHECK-NEXT: [[A:%.*]] = alloca [[T:%.*]], align 8
; CHECK-NEXT: store i8 -1, ptr [[A]], align 1
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 8 [[A]], i8 0, i64 8, i1 false)
; CHECK-NEXT: call void @llvm.memcpy.p0.p0.i64(ptr [[RESULT:%.*]], ptr align 8 [[A]], i64 16, i1 false)
; CHECK-NEXT: ret void
;
%a = alloca %T, align 8
store i8 -1, ptr %a
call void @llvm.memset.p0.i64(ptr align 8 %a, i8 0, i64 8, i1 false)
call void @llvm.memcpy.p0.p0.i64(ptr %result, ptr align 8 %a, i64 16, i1 false)
ret void
}
; A write prior to the memset, which is part of the memcpy (but not memset) region.
; This cannot be optimized.
define void @test_write_before_memset_in_memcpy_region(ptr %result) {
; CHECK-LABEL: @test_write_before_memset_in_memcpy_region(
; CHECK-NEXT: [[A:%.*]] = alloca [[T:%.*]], align 8
; CHECK-NEXT: [[C:%.*]] = getelementptr inbounds [[T]], ptr [[A]], i64 0, i32 2
; CHECK-NEXT: store i32 -1, ptr [[C]], align 4
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 8 [[A]], i8 0, i64 8, i1 false)
; CHECK-NEXT: call void @llvm.memcpy.p0.p0.i64(ptr [[RESULT:%.*]], ptr align 8 [[A]], i64 16, i1 false)
; CHECK-NEXT: ret void
;
%a = alloca %T, align 8
%c = getelementptr inbounds %T, ptr %a, i64 0, i32 2
store i32 -1, ptr %c
call void @llvm.memset.p0.i64(ptr align 8 %a, i8 0, i64 8, i1 false)
call void @llvm.memcpy.p0.p0.i64(ptr %result, ptr align 8 %a, i64 16, i1 false)
ret void
}
; A write prior to the memset, which is part of both the memset and memcpy regions.
; This cannot be optimized.
define void @test_write_before_memset_in_both_regions(ptr %result) {
; CHECK-LABEL: @test_write_before_memset_in_both_regions(
; CHECK-NEXT: [[A:%.*]] = alloca [[T:%.*]], align 8
; CHECK-NEXT: [[C:%.*]] = getelementptr inbounds [[T]], ptr [[A]], i64 0, i32 1
; CHECK-NEXT: store i32 -1, ptr [[C]], align 4
; CHECK-NEXT: call void @llvm.memset.p0.i64(ptr align 8 [[A]], i8 0, i64 10, i1 false)
; CHECK-NEXT: call void @llvm.memcpy.p0.p0.i64(ptr [[RESULT:%.*]], ptr align 8 [[A]], i64 16, i1 false)
; CHECK-NEXT: ret void
;
%a = alloca %T, align 8
%c = getelementptr inbounds %T, ptr %a, i64 0, i32 1
store i32 -1, ptr %c
call void @llvm.memset.p0.i64(ptr align 8 %a, i8 0, i64 10, i1 false)
call void @llvm.memcpy.p0.p0.i64(ptr %result, ptr align 8 %a, i64 16, i1 false)
ret void
}
declare ptr @malloc(i64)
declare void @free(ptr)
declare void @llvm.memset.p0.i64(ptr nocapture, i8, i64, i1)
declare void @llvm.memcpy.p0.p0.i64(ptr nocapture, ptr nocapture readonly, i64, i1)
declare void @llvm.lifetime.start.p0(i64, ptr nocapture)
declare void @llvm.lifetime.end.p0(i64, ptr nocapture)
|
