RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-01 16:33:37 +01:00
Code Issues Projects Releases Wiki Activity
076597a026
llvm-mirror/test/MC/MachO/darwin-x86_64-reloc.s

406 lines
12 KiB
ArmAsm
Raw Normal View History

MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// RUN: llvm-mc -n -triple x86_64-apple-darwin9 %s -filetype=obj -o - | macho-dump --dump-section-data | FileCheck %s
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// These examples are taken from <mach-o/x86_64/reloc.h>.
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
.data
.long 0
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
.text
_foo:
ret
_baz:
call _foo
call _foo+4
movq _foo@GOTPCREL(%rip), %rax
pushq _foo@GOTPCREL(%rip)
movl _foo(%rip), %eax
movl _foo+4(%rip), %eax
movb $0x12, _foo(%rip)
movl $0x12345678, _foo(%rip)
.quad _foo
_bar:
.quad _foo+4
.quad _foo - _bar
.quad _foo - _bar + 4
.long _foo - _bar
leaq L1(%rip), %rax
leaq L0(%rip), %rax
addl $6,L0(%rip)
addw $500,L0(%rip)
addl $500,L0(%rip)
_prev:
.space 12,0x90
.quad L1
L0:
.quad L0
L_pc:
.quad _foo - L_pc
.quad _foo - L1
L1:
.quad L1 - _prev
MC/Mach-O/x86_64: Support @GOTPCREL on symbols, even for non-PCrel relocations! llvm-svn: 99853
2010-03-30 01:56:40 +02:00
.data
.long _foobar@GOTPCREL+4
.long _foo@GOTPCREL+4
MC/Mach-O/x86_64: Relocations in debug sections should use local relocations when possible. - <rdar://problem/7934873> llvm-svn: 103092
2010-05-05 19:22:39 +02:00
.section __DWARF,__debug_frame,regular,debug
.quad L1
.quad _ext_foo
MC/Mach-O x86_64: Switch to using fragment atom symbol. - This eliminates getAtomForAddress() (which was a linear search) and simplifies getAtom(). - This also fixes some correctness problems where local labels at the same address as non-local labels could be assigned to the wrong atom. llvm-svn: 103480
2010-05-11 19:22:50 +02:00
// Make sure local label which overlaps with non-local one is assigned to the
// right atom.
.text
_f2:
L2_0:
addl $0, %eax
L2_1:
_f3:
addl L2_1 - L2_0, %eax
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
.data
L4:
.long 0
.text
movl L4(%rip), %eax
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
.section __TEXT,__literal8,8byte_literals
.quad 0
L5:
.quad 0
f6:
.quad 0
L6:
.quad 0
.text
movl L5(%rip), %eax
movl f6(%rip), %eax
movl L6(%rip), %eax
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
.data
.quad L5
.quad f6
.quad L6
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
.text
cmpq $0, _foo@GOTPCREL(%rip)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('cputype', 16777223)
// CHECK: ('cpusubtype', 3)
// CHECK: ('filetype', 1)
// CHECK: ('num_load_commands', 1)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('load_commands_size', 496)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('flag', 0)
// CHECK: ('reserved', 0)
// CHECK: ('load_commands', [
// CHECK: # Load Command 0
// CHECK: (('command', 25)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('size', 392)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('segment_name', '\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00')
// CHECK: ('vm_addr', 0)
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('vm_size', 311)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('file_offset', 528)
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('file_size', 311)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('maxprot', 7)
// CHECK: ('initprot', 7)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('num_sections', 4)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('flags', 0)
// CHECK: ('sections', [
// CHECK: # Section 0
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: (('section_name', '__data\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00')
// CHECK: ('segment_name', '__DATA\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00')
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('address', 0)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('size', 40)
// CHECK: ('offset', 528)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('alignment', 0)
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('reloc_offset', 840)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('num_reloc', 5)
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: ('flags', 0x0)
// CHECK: ('reserved1', 0)
// CHECK: ('reserved2', 0)
// CHECK: ('reserved3', 0)
// CHECK: ),
// CHECK: ('_relocations', [
// CHECK: # Relocation 0
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: (('word-0', 0x20),
// CHECK: ('word-1', 0x6000004)),
// CHECK: # Relocation 1
// CHECK: (('word-0', 0x18),
// CHECK: ('word-1', 0xe000006)),
// CHECK: # Relocation 2
// CHECK: (('word-0', 0x10),
// CHECK: ('word-1', 0x6000004)),
// CHECK: # Relocation 3
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: (('word-0', 0x8),
// CHECK: ('word-1', 0x4d000000)),
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: # Relocation 4
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: (('word-0', 0x4),
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('word-1', 0x4d000008)),
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: ])
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('_section_data', '\x00\x00\x00\x00\x04\x00\x00\x00\x04\x00\x00\x00\x00\x00\x00\x00\x1f\x01\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00/\x01\x00\x00\x00\x00\x00\x00')
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: # Section 1
// CHECK: (('section_name', '__text\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00')
// CHECK: ('segment_name', '__TEXT\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00')
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('address', 40)
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('size', 223)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('offset', 568)
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: ('alignment', 0)
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('reloc_offset', 880)
// CHECK: ('num_reloc', 32)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('flags', 0x80000400)
// CHECK: ('reserved1', 0)
// CHECK: ('reserved2', 0)
// CHECK: ('reserved3', 0)
// CHECK: ),
// CHECK: ('_relocations', [
// CHECK: # Relocation 0
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: (('word-0', 0xda),
// CHECK: ('word-1', 0x4d000000)),
// CHECK: # Relocation 1
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: (('word-0', 0xd3),
// CHECK: ('word-1', 0x15000004)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 2
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: (('word-0', 0xcd),
// CHECK: ('word-1', 0x1d000006)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 3
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: (('word-0', 0xc7),
// CHECK: ('word-1', 0x15000004)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 4
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: (('word-0', 0xc1),
// CHECK: ('word-1', 0x15000001)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 5
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0xa5),
// CHECK: ('word-1', 0x5e000003)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 6
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0xa5),
// CHECK: ('word-1', 0xe000000)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 7
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x9d),
// CHECK: ('word-1', 0x5e000003)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 8
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x9d),
// CHECK: ('word-1', 0xe000000)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 9
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x95),
// CHECK: ('word-1', 0xe000003)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 10
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x8d),
// CHECK: ('word-1', 0xe000003)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 11
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x79),
// CHECK: ('word-1', 0x8d000003)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 12
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x71),
// CHECK: ('word-1', 0x7d000003)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 13
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x69),
// CHECK: ('word-1', 0x6d000003)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 14
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x63),
// CHECK: ('word-1', 0x1d000003)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 15
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x5c),
// CHECK: ('word-1', 0x1d000003)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 16
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x55),
// CHECK: ('word-1', 0x5c000002)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 17
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x55),
// CHECK: ('word-1', 0xc000000)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 18
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x4d),
// CHECK: ('word-1', 0x5e000002)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 19
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x4d),
// CHECK: ('word-1', 0xe000000)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 20
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x45),
// CHECK: ('word-1', 0x5e000002)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 21
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x45),
// CHECK: ('word-1', 0xe000000)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 22
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x3d),
// CHECK: ('word-1', 0xe000000)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 23
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x35),
// CHECK: ('word-1', 0xe000000)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 24
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x2d),
// CHECK: ('word-1', 0x8d000000)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 25
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x26),
// CHECK: ('word-1', 0x6d000000)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 26
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x20),
// CHECK: ('word-1', 0x1d000000)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 27
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x1a),
// CHECK: ('word-1', 0x1d000000)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 28
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x14),
// CHECK: ('word-1', 0x4d000000)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 29
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0xe),
// CHECK: ('word-1', 0x3d000000)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 30
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x7),
// CHECK: ('word-1', 0x2d000000)),
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: # Relocation 31
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: (('word-0', 0x2),
// CHECK: ('word-1', 0x2d000000)),
// CHECK: ])
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('_section_data', '\xc3\xe8\x00\x00\x00\x00\xe8\x04\x00\x00\x00H\x8b\x05\x00\x00\x00\x00\xff5\x00\x00\x00\x00\x8b\x05\x00\x00\x00\x00\x8b\x05\x04\x00\x00\x00\xc6\x05\xff\xff\xff\xff\x12\xc7\x05\xfc\xff\xff\xffxV4\x12\x00\x00\x00\x00\x00\x00\x00\x00\x04\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x04\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00H\x8d\x05,\x00\x00\x00H\x8d\x05\x14\x00\x00\x00\x83\x05\x13\x00\x00\x00\x06f\x81\x05\x12\x00\x00\x00\xf4\x01\x81\x05\x10\x00\x00\x00\xf4\x01\x00\x00\x90\x90\x90\x90\x90\x90\x90\x90\x90\x90\x90\x90,\x00\x00\x00\x00\x00\x00\x00\x14\x00\x00\x00\x00\x00\x00\x00\xe4\xff\xff\xff\xff\xff\xff\xff\xd4\xff\xff\xff\xff\xff\xff\xff,\x00\x00\x00\x00\x00\x00\x00\x83\xc0\x00\x03\x04%\x03\x00\x00\x00\x8b\x05\x1f\xff\xff\xff\x8b\x05,\x00\x00\x00\x8b\x05\x00\x00\x00\x00\x8b\x050\x00\x00\x00H\x83=\xff\xff\xff\xff\x00')
MC/Mach-O/x86_64: Relocations in debug sections should use local relocations when possible. - <rdar://problem/7934873> llvm-svn: 103092
2010-05-05 19:22:39 +02:00
// CHECK: # Section 2
// CHECK: (('section_name', '__debug_frame\x00\x00\x00')
// CHECK: ('segment_name', '__DWARF\x00\x00\x00\x00\x00\x00\x00\x00\x00')
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('address', 263)
MC/Mach-O/x86_64: Relocations in debug sections should use local relocations when possible. - <rdar://problem/7934873> llvm-svn: 103092
2010-05-05 19:22:39 +02:00
// CHECK: ('size', 16)
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('offset', 791)
MC/Mach-O/x86_64: Relocations in debug sections should use local relocations when possible. - <rdar://problem/7934873> llvm-svn: 103092
2010-05-05 19:22:39 +02:00
// CHECK: ('alignment', 0)
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('reloc_offset', 1136)
MC/Mach-O/x86_64: Relocations in debug sections should use local relocations when possible. - <rdar://problem/7934873> llvm-svn: 103092
2010-05-05 19:22:39 +02:00
// CHECK: ('num_reloc', 2)
// CHECK: ('flags', 0x2000000)
// CHECK: ('reserved1', 0)
// CHECK: ('reserved2', 0)
// CHECK: ('reserved3', 0)
// CHECK: ),
// CHECK: ('_relocations', [
// CHECK: # Relocation 0
// CHECK: (('word-0', 0x8),
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('word-1', 0xe000007)),
MC/Mach-O/x86_64: Relocations in debug sections should use local relocations when possible. - <rdar://problem/7934873> llvm-svn: 103092
2010-05-05 19:22:39 +02:00
// CHECK: # Relocation 1
// CHECK: (('word-0', 0x0),
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: ('word-1', 0x6000002)),
MC/Mach-O/x86_64: Relocations in debug sections should use local relocations when possible. - <rdar://problem/7934873> llvm-svn: 103092
2010-05-05 19:22:39 +02:00
// CHECK: ])
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('_section_data', '\xd5\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00')
// CHECK: # Section 3
// CHECK: (('section_name', '__literal8\x00\x00\x00\x00\x00\x00')
// CHECK: ('segment_name', '__TEXT\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00')
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('address', 279)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('size', 32)
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('offset', 807)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('alignment', 0)
// CHECK: ('reloc_offset', 0)
// CHECK: ('num_reloc', 0)
// CHECK: ('flags', 0x4)
// CHECK: ('reserved1', 0)
// CHECK: ('reserved2', 0)
// CHECK: ('reserved3', 0)
// CHECK: ),
// CHECK: ('_relocations', [
// CHECK: ])
// CHECK: ('_section_data', '\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00')
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ])
// CHECK: ),
// CHECK: # Load Command 1
// CHECK: (('command', 2)
// CHECK: ('size', 24)
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('symoff', 1152)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('nsyms', 9)
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('stroff', 1296)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('strsize', 52)
// CHECK: ('_string_data', '\x00_foobar\x00_ext_foo\x00_foo\x00_baz\x00_bar\x00_prev\x00_f2\x00_f3\x00f6\x00\x00\x00')
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('_symbols', [
// CHECK: # Symbol 0
MC/Mach-O/x86_64: Relocations in debug sections should use local relocations when possible. - <rdar://problem/7934873> llvm-svn: 103092
2010-05-05 19:22:39 +02:00
// CHECK: (('n_strx', 18)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('n_type', 0xe)
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: ('n_sect', 2)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('n_desc', 0)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('n_value', 40)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('_string', '_foo')
// CHECK: ),
// CHECK: # Symbol 1
MC/Mach-O/x86_64: Relocations in debug sections should use local relocations when possible. - <rdar://problem/7934873> llvm-svn: 103092
2010-05-05 19:22:39 +02:00
// CHECK: (('n_strx', 23)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('n_type', 0xe)
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: ('n_sect', 2)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('n_desc', 0)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('n_value', 41)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('_string', '_baz')
// CHECK: ),
// CHECK: # Symbol 2
MC/Mach-O/x86_64: Relocations in debug sections should use local relocations when possible. - <rdar://problem/7934873> llvm-svn: 103092
2010-05-05 19:22:39 +02:00
// CHECK: (('n_strx', 28)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('n_type', 0xe)
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: ('n_sect', 2)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('n_desc', 0)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('n_value', 101)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('_string', '_bar')
// CHECK: ),
// CHECK: # Symbol 3
MC/Mach-O/x86_64: Relocations in debug sections should use local relocations when possible. - <rdar://problem/7934873> llvm-svn: 103092
2010-05-05 19:22:39 +02:00
// CHECK: (('n_strx', 33)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('n_type', 0xe)
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: ('n_sect', 2)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('n_desc', 0)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('n_value', 169)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('_string', '_prev')
// CHECK: ),
MC/Mach-O/x86_64: Support @GOTPCREL on symbols, even for non-PCrel relocations! llvm-svn: 99853
2010-03-30 01:56:40 +02:00
// CHECK: # Symbol 4
MC/Mach-O x86_64: Switch to using fragment atom symbol. - This eliminates getAtomForAddress() (which was a linear search) and simplifies getAtom(). - This also fixes some correctness problems where local labels at the same address as non-local labels could be assigned to the wrong atom. llvm-svn: 103480
2010-05-11 19:22:50 +02:00
// CHECK: (('n_strx', 39)
// CHECK: ('n_type', 0xe)
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: ('n_sect', 2)
MC/Mach-O x86_64: Switch to using fragment atom symbol. - This eliminates getAtomForAddress() (which was a linear search) and simplifies getAtom(). - This also fixes some correctness problems where local labels at the same address as non-local labels could be assigned to the wrong atom. llvm-svn: 103480
2010-05-11 19:22:50 +02:00
// CHECK: ('n_desc', 0)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('n_value', 221)
MC/Mach-O x86_64: Switch to using fragment atom symbol. - This eliminates getAtomForAddress() (which was a linear search) and simplifies getAtom(). - This also fixes some correctness problems where local labels at the same address as non-local labels could be assigned to the wrong atom. llvm-svn: 103480
2010-05-11 19:22:50 +02:00
// CHECK: ('_string', '_f2')
// CHECK: ),
// CHECK: # Symbol 5
// CHECK: (('n_strx', 43)
// CHECK: ('n_type', 0xe)
MC/Mach-O/x86_64: Fix PCrel adjustment for x86_64, which was using the fixup offset instead of the fixup address as intended. llvm-svn: 103527
2010-05-12 01:53:11 +02:00
// CHECK: ('n_sect', 2)
MC/Mach-O x86_64: Switch to using fragment atom symbol. - This eliminates getAtomForAddress() (which was a linear search) and simplifies getAtom(). - This also fixes some correctness problems where local labels at the same address as non-local labels could be assigned to the wrong atom. llvm-svn: 103480
2010-05-11 19:22:50 +02:00
// CHECK: ('n_desc', 0)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('n_value', 224)
MC/Mach-O x86_64: Switch to using fragment atom symbol. - This eliminates getAtomForAddress() (which was a linear search) and simplifies getAtom(). - This also fixes some correctness problems where local labels at the same address as non-local labels could be assigned to the wrong atom. llvm-svn: 103480
2010-05-11 19:22:50 +02:00
// CHECK: ('_string', '_f3')
// CHECK: ),
// CHECK: # Symbol 6
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: (('n_strx', 47)
// CHECK: ('n_type', 0xe)
// CHECK: ('n_sect', 4)
// CHECK: ('n_desc', 0)
MC/Mach-O/x86_64: Darwin's special "signed_N" relocation types should only be used to replace a normal relocation, not a reference to a GOT entry. llvm-svn: 103789
2010-05-14 20:53:40 +02:00
// CHECK: ('n_value', 295)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('_string', 'f6')
// CHECK: ),
// CHECK: # Symbol 7
MC/Mach-O/x86_64: Relocations in debug sections should use local relocations when possible. - <rdar://problem/7934873> llvm-svn: 103092
2010-05-05 19:22:39 +02:00
// CHECK: (('n_strx', 9)
// CHECK: ('n_type', 0x1)
// CHECK: ('n_sect', 0)
// CHECK: ('n_desc', 0)
// CHECK: ('n_value', 0)
// CHECK: ('_string', '_ext_foo')
// CHECK: ),
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: # Symbol 8
MC/Mach-O/x86_64: Support @GOTPCREL on symbols, even for non-PCrel relocations! llvm-svn: 99853
2010-03-30 01:56:40 +02:00
// CHECK: (('n_strx', 1)
// CHECK: ('n_type', 0x1)
// CHECK: ('n_sect', 0)
// CHECK: ('n_desc', 0)
// CHECK: ('n_value', 0)
// CHECK: ('_string', '_foobar')
// CHECK: ),
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ])
// CHECK: ),
// CHECK: # Load Command 2
// CHECK: (('command', 11)
// CHECK: ('size', 80)
// CHECK: ('ilocalsym', 0)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('nlocalsym', 7)
// CHECK: ('iextdefsym', 7)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('nextdefsym', 0)
MC/Mach-O/x86_64: Add a new hook for checking whether a particular section can be diced into atoms, and adjust getAtom() to take this into account. - This fixes relocations to symbols in fixed size literal sections, for example. llvm-svn: 103532
2010-05-12 02:38:17 +02:00
// CHECK: ('iundefsym', 7)
MC/Mach-O/x86_64: Relocations in debug sections should use local relocations when possible. - <rdar://problem/7934873> llvm-svn: 103092
2010-05-05 19:22:39 +02:00
// CHECK: ('nundefsym', 2)
MC/Mach-O/x86_64: Add relocation support. - This is "extraordinarily" Darwin 'as' compatible. See the litany of FIXMEs littered about for more information. - There are a few cases which seem to clearly be 'as' bugs which I have left unsupported, and there is one cases where we diverge but should fix if it blocks diffing .o files (Darwin 'as' ends up widening a jump unnecessarily). - 403.gcc build, runs, and diffs equivalently to the 'as' built version now (using llvm-mc). However, it builds so slowly that I wouldn't recommend trying it quite yet. :) llvm-svn: 98974
2010-03-19 19:07:55 +01:00
// CHECK: ('tocoff', 0)
// CHECK: ('ntoc', 0)
// CHECK: ('modtaboff', 0)
// CHECK: ('nmodtab', 0)
// CHECK: ('extrefsymoff', 0)
// CHECK: ('nextrefsyms', 0)
// CHECK: ('indirectsymoff', 0)
// CHECK: ('nindirectsyms', 0)
// CHECK: ('extreloff', 0)
// CHECK: ('nextrel', 0)
// CHECK: ('locreloff', 0)
// CHECK: ('nlocrel', 0)
// CHECK: ('_indirect_symbols', [
// CHECK: ])
// CHECK: ),
// CHECK: ])
Reference in New Issue Copy Permalink