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
|
# REQUIRES: x86
# RUN: rm -rf %t; split-file %s %t
# RUN: llvm-mc -filetype=obj -triple=x86_64-apple-darwin %t/foo1.s -o %t/foo1.o
# RUN: llvm-mc -filetype=obj -triple=x86_64-apple-darwin %t/foo2.s -o %t/foo2.o
# RUN: %lld -dylib --icf=all -framework CoreFoundation %t/foo1.o %t/foo2.o -o %t/foo
# RUN: llvm-objdump --no-print-imm-hex --macho --rebase --bind --syms -d %t/foo | FileCheck %s --check-prefixes=CHECK,LITERALS
# RUN: %lld -dylib -framework CoreFoundation %t/foo1.o %t/foo2.o -o %t/foo
# RUN: llvm-objdump --no-print-imm-hex --macho --rebase --bind --syms -d %t/foo | FileCheck %s --check-prefix=LITERALS
# CHECK: (__TEXT,__text) section
# CHECK-NEXT: _foo1:
# CHECK-NEXT: _foo2:
# CHECK-NEXT: movq _named_cfstring(%rip), %rax
# CHECK-NEXT: _foo1_utf16:
# CHECK-NEXT: movq [[#]](%rip), %rax
# CHECK-NEXT: _named_foo1:
# CHECK-NEXT: _named_foo2:
# CHECK-NEXT: movq _named_cfstring(%rip), %rax
# CHECK-NEXT: _foo2_utf16:
# CHECK-NEXT: movq [[#]](%rip), %rax
# CHECK: SYMBOL TABLE:
# CHECK-DAG: [[#%.16x,FOO:]] g F __TEXT,__text _foo1
# CHECK-DAG: [[#FOO]] g F __TEXT,__text _foo2
## Make sure we don't emit redundant bind / rebase opcodes for folded sections.
# LITERALS: Rebase table:
# LITERALS-NEXT: segment section address type
# LITERALS-NEXT: __DATA_CONST __cfstring {{.*}} pointer
# LITERALS-NEXT: __DATA_CONST __cfstring {{.*}} pointer
# LITERALS-NEXT: __DATA_CONST __cfstring {{.*}} pointer
# LITERALS-EMPTY:
# LITERALS-NEXT: Bind table:
# LITERALS-NEXT: segment section address type addend dylib symbol
# LITERALS-NEXT: __DATA_CONST __cfstring {{.*}} pointer 0 CoreFoundation ___CFConstantStringClassReference
# LITERALS-NEXT: __DATA_CONST __cfstring {{.*}} pointer 0 CoreFoundation ___CFConstantStringClassReference
# LITERALS-NEXT: __DATA_CONST __cfstring {{.*}} pointer 0 CoreFoundation ___CFConstantStringClassReference
# LITERALS-EMPTY:
#--- foo1.s
.cstring
L_.str.0:
.asciz "bar"
## This string is at a different offset than the corresponding "foo" string in
## foo2.s. Make sure that we treat references to either string as equivalent.
L_.str:
.asciz "foo"
.section __DATA,__cfstring
.p2align 3
L__unnamed_cfstring_:
.quad ___CFConstantStringClassReference
.long 1992 ## utf-8
.space 4
.quad L_.str
.quad 3 ## strlen
_named_cfstring:
.quad ___CFConstantStringClassReference
.long 1992 ## utf-8
.space 4
.quad L_.str
.quad 3 ## strlen
.section __TEXT,__ustring
l_.ustr:
.short 102 ## f
.short 111 ## o
.short 0 ## \0
.short 111 ## o
.short 0 ## \0
## FIXME: We should be able to deduplicate UTF-16 CFStrings too.
## Note that this string contains a null byte in the middle -- any dedup code
## we add should take care to handle this correctly.
## Technically, UTF-8 should support encoding null bytes too, but since we
## atomize the __cstring section at every null byte, this isn't supported. ld64
## doesn't support it either, and clang seems to always emit a UTF-16 CFString
## if it needs to contain a null, so I think we're good here.
.section __DATA,__cfstring
.p2align 3
L__unnamed_cfstring_.2:
.quad ___CFConstantStringClassReference
.long 2000 ## utf-16
.space 4
.quad l_.ustr
.quad 4 ## strlen
.text
.globl _foo1, _foo1_utf16, _named_foo1
_foo1:
movq L__unnamed_cfstring_(%rip), %rax
_foo1_utf16:
movq L__unnamed_cfstring_.2(%rip), %rax
_named_foo1:
movq _named_cfstring(%rip), %rax
.subsections_via_symbols
#--- foo2.s
.cstring
L_.str:
.asciz "foo"
.section __DATA,__cfstring
.p2align 3
L__unnamed_cfstring_:
.quad ___CFConstantStringClassReference
.long 1992 ## utf-8
.space 4
.quad L_.str
.quad 3 ## strlen
_named_cfstring:
.quad ___CFConstantStringClassReference
.long 1992 ## utf-8
.space 4
.quad L_.str
.quad 3 ## strlen
.section __TEXT,__ustring
.p2align 1
l_.ustr:
.short 102 ## f
.short 111 ## o
.short 0 ## \0
.short 111 ## o
.short 0 ## \0
.section __DATA,__cfstring
.p2align 3
L__unnamed_cfstring_.2:
.quad ___CFConstantStringClassReference
.long 2000 ## utf-16
.space 4
.quad l_.ustr
.quad 4 ## strlen
.text
.globl _foo2, _foo2_utf16, _named_foo2
_foo2:
movq L__unnamed_cfstring_(%rip), %rax
_foo2_utf16:
movq L__unnamed_cfstring_.2(%rip), %rax
_named_foo2:
movq _named_cfstring(%rip), %rax
.subsections_via_symbols
|
