| 1 |
## This test shows that llvm-objcopy does not baulk at overlapping sections. |
| 2 |
## These don't really make sense, but the tool should still handle invalid |
| 3 |
## inputs somehow. |
| 4 |
# RUN: yaml2obj %s -o %t.o |
| 5 |
|
| 6 |
## Check that the contents are as expected before the copy. |
| 7 |
# RUN: llvm-readobj -x .first -x .second %t.o | FileCheck %s --check-prefix=CONTENTS |
| 8 |
|
| 9 |
## Now check that the section contents are still correct after the copy. |
| 10 |
## Also characterize the behavior of llvm-objcopy in that it "unoverlaps" the |
| 11 |
## two sections. |
| 12 |
# RUN: llvm-objcopy %t.o %t2.o |
| 13 |
# RUN: llvm-readobj --section-headers %t2.o | FileCheck %s |
| 14 |
# RUN: llvm-readobj -x .first -x .second %t2.o | FileCheck %s --check-prefix=CONTENTS |
| 15 |
|
| 16 |
# CHECK: Name: .first |
| 17 |
# CHECK: Offset: 0x1000 |
| 18 |
# CHECK: Name: .second |
| 19 |
# CHECK: Offset: 0x1004 |
| 20 |
|
| 21 |
# CONTENTS: Hex dump of section '.first': |
| 22 |
# CONTENTS-NEXT: 0x00000000 01234567 |
| 23 |
# CONTENTS-EMPTY: |
| 24 |
# CONTENTS-NEXT: Hex dump of section '.second': |
| 25 |
# CONTENTS-NEXT: 0x00000000 23456789 |
| 26 |
|
| 27 |
--- !ELF |
| 28 |
FileHeader: |
| 29 |
Class: ELFCLASS64 |
| 30 |
Data: ELFDATA2LSB |
| 31 |
Type: ET_EXEC |
| 32 |
Machine: EM_X86_64 |
| 33 |
Sections: |
| 34 |
- Name: .first |
| 35 |
Type: SHT_PROGBITS |
| 36 |
Content: '01234567' |
| 37 |
AddressAlign: 0x1000 |
| 38 |
- Name: .second |
| 39 |
Type: SHT_PROGBITS |
| 40 |
Content: '89abcdef' |
| 41 |
ShOffset: 0x1001 |