2016-04-13 06:36:05 +08:00
|
|
|
// REQUIRES: x86
|
|
|
|
|
|
|
|
// RUN: llvm-mc -filetype=obj -triple=x86_64-pc-linux %s -o %t.o
|
|
|
|
// RUN: llvm-mc -filetype=obj -triple=x86_64-pc-linux %p/Inputs/relocation-copy.s -o %t2.o
|
2019-09-11 14:20:14 +08:00
|
|
|
// RUN: ld.lld %t2.o -o %t2.so -shared -soname=so
|
2017-10-06 17:37:44 +08:00
|
|
|
// RUN: ld.lld --hash-style=sysv %t.o %t2.so -o %t.exe
|
2019-05-01 13:49:01 +08:00
|
|
|
// RUN: llvm-readobj -S --section-data -r %t.exe | FileCheck %s
|
2016-04-13 06:36:05 +08:00
|
|
|
|
2019-09-06 23:10:31 +08:00
|
|
|
// Copy relocate x in a non-writable position.
|
2016-04-13 06:36:05 +08:00
|
|
|
.global _start
|
|
|
|
_start:
|
2016-06-10 04:22:35 +08:00
|
|
|
.quad x
|
2016-04-13 06:36:05 +08:00
|
|
|
|
2019-09-06 23:10:31 +08:00
|
|
|
// Resolved to 0 in a non-alloc section.
|
2016-04-13 06:36:05 +08:00
|
|
|
.section foo
|
2016-06-10 04:22:35 +08:00
|
|
|
.quad y
|
2016-04-13 06:36:05 +08:00
|
|
|
|
2019-09-06 23:10:31 +08:00
|
|
|
// Produce a dynamic relocation in a writable position.
|
2016-04-14 09:48:11 +08:00
|
|
|
.section bar, "aw"
|
2016-06-10 04:22:35 +08:00
|
|
|
.quad z
|
2016-04-14 09:48:11 +08:00
|
|
|
|
2016-04-13 06:36:05 +08:00
|
|
|
// CHECK: Name: .text
|
2019-09-06 23:10:31 +08:00
|
|
|
// CHECK: SectionData (
|
[ELF] Simplify RelRo, TLS, NOBITS section ranks and make RW PT_LOAD start with RelRo
Old: PT_LOAD(.data | PT_GNU_RELRO(.data.rel.ro .bss.rel.ro) | .bss)
New: PT_LOAD(PT_GNU_RELRO(.data.rel.ro .bss.rel.ro) | .data .bss)
The placement of | indicates page alignment caused by PT_GNU_RELRO. The
new layout has simpler rules and saves space for many cases.
Old size: roundup(.data) + roundup(.data.rel.ro)
New size: roundup(.data.rel.ro + .bss.rel.ro) + .data
Other advantages:
* At runtime the 3 memory mappings decrease to 2.
* start(PT_TLS) = start(PT_GNU_RELRO) = start(RW PT_LOAD). This
simplifies binary manipulation tools.
GNU strip before 2.31 discards PT_GNU_RELRO if its
address is not equal to the start of its associated PT_LOAD.
This has been fixed by https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=f2731e0c374e5323ce4cdae2bcc7b7fe22da1a6f
But with this change, we will be compatible with GNU strip before 2.31
* Before, .got.plt (non-relro by default) was placed before .got (relro
by default), which made it impossible to have _GLOBAL_OFFSET_TABLE_
(start of .got.plt on x86-64) equal to the end of .got (R_GOT*_FROM_END)
(https://bugs.llvm.org/show_bug.cgi?id=36555). With the new ordering, we
can improve on this regard if we'd like to.
Reviewers: ruiu, espindola, pcc
Subscribers: emaste, arichardson, llvm-commits, joerg, jdoerfert
Differential Revision: https://reviews.llvm.org/D56828
llvm-svn: 356117
2019-03-14 11:47:45 +08:00
|
|
|
// CHECK-NEXT: 0000: 10302000
|
2016-04-14 09:48:11 +08:00
|
|
|
// CHECK-NEXT: )
|
|
|
|
|
|
|
|
// CHECK: Name: bar
|
|
|
|
// CHECK-NEXT: Type: SHT_PROGBITS
|
|
|
|
// CHECK-NEXT: Flags [
|
|
|
|
// CHECK-NEXT: SHF_ALLOC
|
|
|
|
// CHECK-NEXT: SHF_WRITE
|
|
|
|
// CHECK-NEXT: ]
|
[ELF] Simplify RelRo, TLS, NOBITS section ranks and make RW PT_LOAD start with RelRo
Old: PT_LOAD(.data | PT_GNU_RELRO(.data.rel.ro .bss.rel.ro) | .bss)
New: PT_LOAD(PT_GNU_RELRO(.data.rel.ro .bss.rel.ro) | .data .bss)
The placement of | indicates page alignment caused by PT_GNU_RELRO. The
new layout has simpler rules and saves space for many cases.
Old size: roundup(.data) + roundup(.data.rel.ro)
New size: roundup(.data.rel.ro + .bss.rel.ro) + .data
Other advantages:
* At runtime the 3 memory mappings decrease to 2.
* start(PT_TLS) = start(PT_GNU_RELRO) = start(RW PT_LOAD). This
simplifies binary manipulation tools.
GNU strip before 2.31 discards PT_GNU_RELRO if its
address is not equal to the start of its associated PT_LOAD.
This has been fixed by https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=f2731e0c374e5323ce4cdae2bcc7b7fe22da1a6f
But with this change, we will be compatible with GNU strip before 2.31
* Before, .got.plt (non-relro by default) was placed before .got (relro
by default), which made it impossible to have _GLOBAL_OFFSET_TABLE_
(start of .got.plt on x86-64) equal to the end of .got (R_GOT*_FROM_END)
(https://bugs.llvm.org/show_bug.cgi?id=36555). With the new ordering, we
can improve on this regard if we'd like to.
Reviewers: ruiu, espindola, pcc
Subscribers: emaste, arichardson, llvm-commits, joerg, jdoerfert
Differential Revision: https://reviews.llvm.org/D56828
llvm-svn: 356117
2019-03-14 11:47:45 +08:00
|
|
|
// CHECK-NEXT: Address: 0x203000
|
|
|
|
// CHECK-NEXT: Offset: 0x3000
|
2016-06-10 04:22:35 +08:00
|
|
|
// CHECK-NEXT: Size: 8
|
2016-04-14 09:48:11 +08:00
|
|
|
// CHECK-NEXT: Link: 0
|
|
|
|
// CHECK-NEXT: Info: 0
|
|
|
|
// CHECK-NEXT: AddressAlignment: 1
|
|
|
|
// CHECK-NEXT: EntrySize: 0
|
|
|
|
// CHECK-NEXT: SectionData (
|
|
|
|
// CHECK-NEXT: 0000: 00000000
|
2016-04-13 06:36:05 +08:00
|
|
|
// CHECK-NEXT: )
|
|
|
|
|
|
|
|
// CHECK: Name: foo
|
|
|
|
// CHECK-NEXT: Type: SHT_PROGBITS
|
|
|
|
// CHECK-NEXT: Flags [
|
|
|
|
// CHECK-NEXT: ]
|
|
|
|
// CHECK-NEXT: Address: 0x0
|
[ELF] Simplify RelRo, TLS, NOBITS section ranks and make RW PT_LOAD start with RelRo
Old: PT_LOAD(.data | PT_GNU_RELRO(.data.rel.ro .bss.rel.ro) | .bss)
New: PT_LOAD(PT_GNU_RELRO(.data.rel.ro .bss.rel.ro) | .data .bss)
The placement of | indicates page alignment caused by PT_GNU_RELRO. The
new layout has simpler rules and saves space for many cases.
Old size: roundup(.data) + roundup(.data.rel.ro)
New size: roundup(.data.rel.ro + .bss.rel.ro) + .data
Other advantages:
* At runtime the 3 memory mappings decrease to 2.
* start(PT_TLS) = start(PT_GNU_RELRO) = start(RW PT_LOAD). This
simplifies binary manipulation tools.
GNU strip before 2.31 discards PT_GNU_RELRO if its
address is not equal to the start of its associated PT_LOAD.
This has been fixed by https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=f2731e0c374e5323ce4cdae2bcc7b7fe22da1a6f
But with this change, we will be compatible with GNU strip before 2.31
* Before, .got.plt (non-relro by default) was placed before .got (relro
by default), which made it impossible to have _GLOBAL_OFFSET_TABLE_
(start of .got.plt on x86-64) equal to the end of .got (R_GOT*_FROM_END)
(https://bugs.llvm.org/show_bug.cgi?id=36555). With the new ordering, we
can improve on this regard if we'd like to.
Reviewers: ruiu, espindola, pcc
Subscribers: emaste, arichardson, llvm-commits, joerg, jdoerfert
Differential Revision: https://reviews.llvm.org/D56828
llvm-svn: 356117
2019-03-14 11:47:45 +08:00
|
|
|
// CHECK-NEXT: Offset: 0x3008
|
2016-06-10 04:22:35 +08:00
|
|
|
// CHECK-NEXT: Size: 8
|
2016-04-13 06:36:05 +08:00
|
|
|
// CHECK-NEXT: Link: 0
|
|
|
|
// CHECK-NEXT: Info: 0
|
|
|
|
// CHECK-NEXT: AddressAlignment: 1
|
|
|
|
// CHECK-NEXT: EntrySize: 0
|
|
|
|
// CHECK-NEXT: SectionData (
|
|
|
|
// CHECK-NEXT: 0000: 00000000
|
|
|
|
// CHECK-NEXT: )
|
|
|
|
|
|
|
|
// CHECK: Relocations [
|
|
|
|
// CHECK-NEXT: Section (4) .rela.dyn {
|
[ELF] Simplify RelRo, TLS, NOBITS section ranks and make RW PT_LOAD start with RelRo
Old: PT_LOAD(.data | PT_GNU_RELRO(.data.rel.ro .bss.rel.ro) | .bss)
New: PT_LOAD(PT_GNU_RELRO(.data.rel.ro .bss.rel.ro) | .data .bss)
The placement of | indicates page alignment caused by PT_GNU_RELRO. The
new layout has simpler rules and saves space for many cases.
Old size: roundup(.data) + roundup(.data.rel.ro)
New size: roundup(.data.rel.ro + .bss.rel.ro) + .data
Other advantages:
* At runtime the 3 memory mappings decrease to 2.
* start(PT_TLS) = start(PT_GNU_RELRO) = start(RW PT_LOAD). This
simplifies binary manipulation tools.
GNU strip before 2.31 discards PT_GNU_RELRO if its
address is not equal to the start of its associated PT_LOAD.
This has been fixed by https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=f2731e0c374e5323ce4cdae2bcc7b7fe22da1a6f
But with this change, we will be compatible with GNU strip before 2.31
* Before, .got.plt (non-relro by default) was placed before .got (relro
by default), which made it impossible to have _GLOBAL_OFFSET_TABLE_
(start of .got.plt on x86-64) equal to the end of .got (R_GOT*_FROM_END)
(https://bugs.llvm.org/show_bug.cgi?id=36555). With the new ordering, we
can improve on this regard if we'd like to.
Reviewers: ruiu, espindola, pcc
Subscribers: emaste, arichardson, llvm-commits, joerg, jdoerfert
Differential Revision: https://reviews.llvm.org/D56828
llvm-svn: 356117
2019-03-14 11:47:45 +08:00
|
|
|
// CHECK-NEXT: 0x203010 R_X86_64_COPY x 0x0
|
|
|
|
// CHECK-NEXT: 0x203000 R_X86_64_64 z 0x0
|
2016-04-13 06:36:05 +08:00
|
|
|
// CHECK-NEXT: }
|
|
|
|
// CHECK-NEXT: ]
|