maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/lld/test/ELF/gdb-index.s

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

135 lines
3.6 KiB
ArmAsm
Raw Normal View History

Add REQUIRES: zlib to gdb-index.s test llvm-svn: 324978
2018-02-13 09:19:56 +08:00
# REQUIRES: x86, zlib
[ELF] - Remove dependency on precompiled inputs from -gdb-index testcases. Previously we used precompiled objects in gdb-index.s and debug-gnu-pubnames.s testcases. We can avoid that. Differential revision: https://reviews.llvm.org/D35360 llvm-svn: 308005
2017-07-14 16:26:44 +08:00
# RUN: llvm-mc -filetype=obj -triple=x86_64-pc-linux %s -o %t1.o
# RUN: llvm-mc -filetype=obj -triple=x86_64-pc-linux %p/Inputs/gdb-index.s -o %t2.o
Fix off-by-one error. llvm-svn: 314093
2017-09-25 11:40:45 +08:00
# RUN: ld.lld --gdb-index %t1.o %t2.o -o %t
Remove 'z' in .zdebug when decompressing a section. When decompressing a compressed debug section, we drop SHF_COMPRESSED flag but we didn't drop "z" in ".zdebug" section name. This patch does that for consistency. This change also fixes the issue that .zdebug_gnu_pubnames are not dropped when we are creating a .gdb_index section. llvm-svn: 324949
2018-02-13 06:25:45 +08:00
# RUN: llvm-objdump -d %t | FileCheck %s --check-prefix=DISASM
# RUN: llvm-dwarfdump -gdb-index %t | FileCheck %s --check-prefix=DWARF
[ELF] Drop .rel[a].debug_gnu_pub{names,types} for --gdb-index --emit-relocs Fixes PR48693: --emit-relocs keeps relocation sections. --gdb-index drops .debug_gnu_pubnames and .debug_gnu_pubtypes but not their relocation sections. This can cause a null pointer dereference in `getOutputSectionName`. Also delete debug-gnu-pubnames.s which is covered by gdb-index.s Reviewed By: grimar Differential Revision: https://reviews.llvm.org/D94354
2021-01-12 16:07:28 +08:00
## Drop .debug_gnu_pubnames and .debug_gnu_pubtypes.
## Also drop their relocation sections if --emit-relocs is specified.
# RUN: ld.lld --gdb-index --emit-relocs %t1.o %t2.o -o %t1
# RUN: llvm-readelf --sections %t1 | FileCheck %s --check-prefix=SECTION
# SECTION-NOT: .debug_gnu_pubnames
# SECTION-NOT: .debug_gnu_pubtypes
Remove 'z' in .zdebug when decompressing a section. When decompressing a compressed debug section, we drop SHF_COMPRESSED flag but we didn't drop "z" in ".zdebug" section name. This patch does that for consistency. This change also fixes the issue that .zdebug_gnu_pubnames are not dropped when we are creating a .gdb_index section. llvm-svn: 324949
2018-02-13 06:25:45 +08:00
# RUN: llvm-mc -compress-debug-sections=zlib-gnu -filetype=obj -triple=x86_64-pc-linux \
# RUN: %p/Inputs/gdb-index.s -o %t2.o
# RUN: ld.lld --gdb-index %t1.o %t2.o -o %t
[ELF] - Partial support of --gdb-index command line option (Part 1). In this patch partial gdb_index section is created. For costructing the .gdb_index section 6 steps should be performed (details are in SplitDebugInfo.cpp file header), this patch do first 3: Creates proper section header. Fills list of compilation units. Types CU list area is not supposed to be supported, so it is ignored and therefore can be treated as implemented either. Differential revision: https://reviews.llvm.org/D24706 llvm-svn: 284708
2016-10-20 17:19:48 +08:00
# RUN: llvm-objdump -d %t | FileCheck %s --check-prefix=DISASM
Remove 'z' in .zdebug when decompressing a section. When decompressing a compressed debug section, we drop SHF_COMPRESSED flag but we didn't drop "z" in ".zdebug" section name. This patch does that for consistency. This change also fixes the issue that .zdebug_gnu_pubnames are not dropped when we are creating a .gdb_index section. llvm-svn: 324949
2018-02-13 06:25:45 +08:00
# RUN: llvm-dwarfdump -gdb-index %t | FileCheck %s --check-prefix=DWARF
[ELF][test] Avoid llvm-readelf/llvm-readobj one-dash long options
2021-07-17 01:02:47 +08:00
# RUN: llvm-readelf -S %t | FileCheck %s --check-prefix=SECTION
[ELF] - Partial support of --gdb-index command line option (Part 1). In this patch partial gdb_index section is created. For costructing the .gdb_index section 6 steps should be performed (details are in SplitDebugInfo.cpp file header), this patch do first 3: Creates proper section header. Fills list of compilation units. Types CU list area is not supposed to be supported, so it is ignored and therefore can be treated as implemented either. Differential revision: https://reviews.llvm.org/D24706 llvm-svn: 284708
2016-10-20 17:19:48 +08:00
# DISASM: Disassembly of section .text:
[llvm-objdump] Print newlines before and after "Disassembly of section ...:" This improves readability and the behavior is consistent with GNU objdump. The new test test/tools/llvm-objdump/X86/disassemble-section-name.s checks we print newlines before and after "Disassembly of section ...:" Differential Revision: https://reviews.llvm.org/D61127 llvm-svn: 359668
2019-05-01 18:40:48 +08:00
# DISASM-EMPTY:
[llvm-objdump] -d: print `00000000 <foo>:` instead of `00000000 foo:` The new behavior matches GNU objdump. A pair of angle brackets makes tests slightly easier. `.foo:` is not unique and thus cannot be used in a `CHECK-LABEL:` directive. Without `-LABEL`, the CHECK line can match the `Disassembly of section` line and causes the next `CHECK-NEXT:` to fail. ``` Disassembly of section .foo: 0000000000001634 .foo: ``` Bdragon: <> has metalinguistic connotation. it just "feels right" Reviewed By: rupprecht Differential Revision: https://reviews.llvm.org/D75713
2020-03-06 06:18:38 +08:00
# DISASM: <entrypoint>:
[ELF] - Remove dependency on precompiled inputs from -gdb-index testcases. Previously we used precompiled objects in gdb-index.s and debug-gnu-pubnames.s testcases. We can avoid that. Differential revision: https://reviews.llvm.org/D35360 llvm-svn: 308005
2017-07-14 16:26:44 +08:00
# DISASM-CHECK: 201000: 90 nop
# DISASM-CHECK: 201001: cc int3
# DISASM-CHECK: 201002: cc int3
# DISASM-CHECK: 201003: cc int3
[llvm-objdump] -d: print `00000000 <foo>:` instead of `00000000 foo:` The new behavior matches GNU objdump. A pair of angle brackets makes tests slightly easier. `.foo:` is not unique and thus cannot be used in a `CHECK-LABEL:` directive. Without `-LABEL`, the CHECK line can match the `Disassembly of section` line and causes the next `CHECK-NEXT:` to fail. ``` Disassembly of section .foo: 0000000000001634 .foo: ``` Bdragon: <> has metalinguistic connotation. it just "feels right" Reviewed By: rupprecht Differential Revision: https://reviews.llvm.org/D75713
2020-03-06 06:18:38 +08:00
# DISASM: <aaaaaaaaaaaaaaaa>:
[ELF] - Remove dependency on precompiled inputs from -gdb-index testcases. Previously we used precompiled objects in gdb-index.s and debug-gnu-pubnames.s testcases. We can avoid that. Differential revision: https://reviews.llvm.org/D35360 llvm-svn: 308005
2017-07-14 16:26:44 +08:00
# DISASM-CHECK: 201004: 90 nop
# DISASM-CHECK: 201005: 90 nop
[ELF] - Partial support of --gdb-index command line option (Part 1). In this patch partial gdb_index section is created. For costructing the .gdb_index section 6 steps should be performed (details are in SplitDebugInfo.cpp file header), this patch do first 3: Creates proper section header. Fills list of compilation units. Types CU list area is not supposed to be supported, so it is ignored and therefore can be treated as implemented either. Differential revision: https://reviews.llvm.org/D24706 llvm-svn: 284708
2016-10-20 17:19:48 +08:00
[DWARF] Fix typo, .gnu_index -> .gdb_index llvm-svn: 346040
2018-11-03 04:34:42 +08:00
# DWARF: .gdb_index contents:
Remove 'z' in .zdebug when decompressing a section. When decompressing a compressed debug section, we drop SHF_COMPRESSED flag but we didn't drop "z" in ".zdebug" section name. This patch does that for consistency. This change also fixes the issue that .zdebug_gnu_pubnames are not dropped when we are creating a .gdb_index section. llvm-svn: 324949
2018-02-13 06:25:45 +08:00
# DWARF-NEXT: Version = 7
# DWARF: CU list offset = 0x18, has 2 entries:
# DWARF-NEXT: 0: Offset = 0x0, Length = 0x34
# DWARF-NEXT: 1: Offset = 0x34, Length = 0x34
# DWARF: Address area offset = 0x38, has 2 entries:
[ELF][X86] Allow PT_LOAD to have overlapping p_offset ranges on EM_X86_64 Port the D64906 technique to EM_X86_64. Differential Revision: https://reviews.llvm.org/D67482 llvm-svn: 371958
2019-09-16 15:05:34 +08:00
# DWARF-NEXT: Low/High address = [0x201120, 0x201121) (Size: 0x1), CU id = 0
# DWARF-NEXT: Low/High address = [0x201124, 0x201126) (Size: 0x2), CU id = 1
Remove 'z' in .zdebug when decompressing a section. When decompressing a compressed debug section, we drop SHF_COMPRESSED flag but we didn't drop "z" in ".zdebug" section name. This patch does that for consistency. This change also fixes the issue that .zdebug_gnu_pubnames are not dropped when we are creating a .gdb_index section. llvm-svn: 324949
2018-02-13 06:25:45 +08:00
# DWARF: Symbol table offset = 0x60, size = 1024, filled slots:
Parallelize GdbIndexSection's symbol table creation. Since .gdb_index sections contain all known symbols, they can be very large. One of my executables has a .gdb_index section of 1350 GiB. Uniquifying symbols by name takes 3.77 seconds on my machine. This patch parallelize it. Time to call createSymbols() with 8.4 million unique symbols: Without this patch: 3773 ms Parallelism = 1: 4374 ms Parallelism = 2: 2628 ms Parallelism = 16: 837 ms As you can see above, this algorithm is a bit more inefficient than the non-parallelized version, but even with dual-core, it is faster than that, so I think it is overall a win. Differential Revision: https://reviews.llvm.org/D49164 llvm-svn: 336790
2018-07-11 19:37:10 +08:00
# DWARF-NEXT: 512: Name offset = 0x1c, CU vector offset = 0x0
# DWARF-NEXT: String name: aaaaaaaaaaaaaaaa, CU vector index: 0
# DWARF-NEXT: 754: Name offset = 0x38, CU vector offset = 0x10
# DWARF-NEXT: String name: int, CU vector index: 2
# DWARF-NEXT: 822: Name offset = 0x2d, CU vector offset = 0x8
# DWARF-NEXT: String name: entrypoint, CU vector index: 1
Remove 'z' in .zdebug when decompressing a section. When decompressing a compressed debug section, we drop SHF_COMPRESSED flag but we didn't drop "z" in ".zdebug" section name. This patch does that for consistency. This change also fixes the issue that .zdebug_gnu_pubnames are not dropped when we are creating a .gdb_index section. llvm-svn: 324949
2018-02-13 06:25:45 +08:00
# DWARF: Constant pool offset = 0x2060, has 3 CU vectors:
Parallelize GdbIndexSection's symbol table creation. Since .gdb_index sections contain all known symbols, they can be very large. One of my executables has a .gdb_index section of 1350 GiB. Uniquifying symbols by name takes 3.77 seconds on my machine. This patch parallelize it. Time to call createSymbols() with 8.4 million unique symbols: Without this patch: 3773 ms Parallelism = 1: 4374 ms Parallelism = 2: 2628 ms Parallelism = 16: 837 ms As you can see above, this algorithm is a bit more inefficient than the non-parallelized version, but even with dual-core, it is faster than that, so I think it is overall a win. Differential Revision: https://reviews.llvm.org/D49164 llvm-svn: 336790
2018-07-11 19:37:10 +08:00
# DWARF-NEXT: 0(0x0): 0x30000001
# DWARF-NEXT: 1(0x8): 0x30000000
# DWARF-NEXT: 2(0x10): 0x90000000 0x90000001
Remove 'z' in .zdebug when decompressing a section. When decompressing a compressed debug section, we drop SHF_COMPRESSED flag but we didn't drop "z" in ".zdebug" section name. This patch does that for consistency. This change also fixes the issue that .zdebug_gnu_pubnames are not dropped when we are creating a .gdb_index section. llvm-svn: 324949
2018-02-13 06:25:45 +08:00
Fix off-by-one error. llvm-svn: 314093
2017-09-25 11:40:45 +08:00
# RUN: ld.lld --gdb-index --no-gdb-index %t1.o %t2.o -o %t2
[test] Change llvm-readobj -long-option to --long-option or well-known short options. NFC Also change some options that have different semantics (cause confusion) in llvm-readelf mode: -s => -S -t => --symbols -sd => --section-data llvm-svn: 359651
2019-05-01 13:49:01 +08:00
# RUN: llvm-readobj --sections %t2 | FileCheck -check-prefix=NOGDB %s
Add -no-gdb-index which negates -gdb-index option. llvm-svn: 312753
2017-09-08 06:40:54 +08:00
# NOGDB-NOT: Name: .gdb_index
[ELF] - Remove dependency on precompiled inputs from -gdb-index testcases. Previously we used precompiled objects in gdb-index.s and debug-gnu-pubnames.s testcases. We can avoid that. Differential revision: https://reviews.llvm.org/D35360 llvm-svn: 308005
2017-07-14 16:26:44 +08:00
## The following section contents are created by this using gcc 7.1.0:
Fix off-by-one error. llvm-svn: 314093
2017-09-25 11:40:45 +08:00
## echo 'int entrypoint() { return 0; }' | gcc -gsplit-dwarf -xc++ -S -o- -
[ELF] - Remove dependency on precompiled inputs from -gdb-index testcases. Previously we used precompiled objects in gdb-index.s and debug-gnu-pubnames.s testcases. We can avoid that. Differential revision: https://reviews.llvm.org/D35360 llvm-svn: 308005
2017-07-14 16:26:44 +08:00
.text
.Ltext0:
Fix off-by-one error. llvm-svn: 314093
2017-09-25 11:40:45 +08:00
.globl entrypoint
.type entrypoint, @function
entrypoint:
[ELF] - Remove dependency on precompiled inputs from -gdb-index testcases. Previously we used precompiled objects in gdb-index.s and debug-gnu-pubnames.s testcases. We can avoid that. Differential revision: https://reviews.llvm.org/D35360 llvm-svn: 308005
2017-07-14 16:26:44 +08:00
nop
.Letext0:
.section .debug_info,"",@progbits
.long 0x30
.value 0x4
.long 0
.byte 0x8
.uleb128 0x1
.quad .Ltext0
.quad .Letext0-.Ltext0
.long 0
.long 0
.long 0
.long 0
.byte 0x63
.byte 0x88
.byte 0xb4
.byte 0x61
.byte 0xaa
.byte 0xb6
.byte 0xb0
.byte 0x67
.section .debug_abbrev,"",@progbits
.uleb128 0x1
.uleb128 0x11
.byte 0
.uleb128 0x11
.uleb128 0x1
.uleb128 0x12
.uleb128 0x7
.uleb128 0x10
.uleb128 0x17
.uleb128 0x2130
.uleb128 0xe
.uleb128 0x1b
.uleb128 0xe
.uleb128 0x2134
.uleb128 0x19
.uleb128 0x2133
.uleb128 0x17
.uleb128 0x2131
.uleb128 0x7
.byte 0
.byte 0
.byte 0
.section .debug_gnu_pubnames,"",@progbits
[DebugInfo] Add more checks to parsing .debug_pub* sections. The patch adds checking for various potential issues in parsing name lookup tables and reporting them as recoverable errors, similarly as we do for other tables. Differential Revision: https://reviews.llvm.org/D83050
2020-07-09 20:15:31 +08:00
.long 0x1e
[ELF] - Remove dependency on precompiled inputs from -gdb-index testcases. Previously we used precompiled objects in gdb-index.s and debug-gnu-pubnames.s testcases. We can avoid that. Differential revision: https://reviews.llvm.org/D35360 llvm-svn: 308005
2017-07-14 16:26:44 +08:00
.value 0x2
[ELF] Drop .rel[a].debug_gnu_pub{names,types} for --gdb-index --emit-relocs Fixes PR48693: --emit-relocs keeps relocation sections. --gdb-index drops .debug_gnu_pubnames and .debug_gnu_pubtypes but not their relocation sections. This can cause a null pointer dereference in `getOutputSectionName`. Also delete debug-gnu-pubnames.s which is covered by gdb-index.s Reviewed By: grimar Differential Revision: https://reviews.llvm.org/D94354
2021-01-12 16:07:28 +08:00
.long .debug_info
[ELF] - Remove dependency on precompiled inputs from -gdb-index testcases. Previously we used precompiled objects in gdb-index.s and debug-gnu-pubnames.s testcases. We can avoid that. Differential revision: https://reviews.llvm.org/D35360 llvm-svn: 308005
2017-07-14 16:26:44 +08:00
.long 0x33
.long 0x18
.byte 0x30
Fix off-by-one error. llvm-svn: 314093
2017-09-25 11:40:45 +08:00
.string "entrypoint"
[ELF] - Remove dependency on precompiled inputs from -gdb-index testcases. Previously we used precompiled objects in gdb-index.s and debug-gnu-pubnames.s testcases. We can avoid that. Differential revision: https://reviews.llvm.org/D35360 llvm-svn: 308005
2017-07-14 16:26:44 +08:00
.long 0
.section .debug_gnu_pubtypes,"",@progbits
.long 0x17
.value 0x2
[ELF] Drop .rel[a].debug_gnu_pub{names,types} for --gdb-index --emit-relocs Fixes PR48693: --emit-relocs keeps relocation sections. --gdb-index drops .debug_gnu_pubnames and .debug_gnu_pubtypes but not their relocation sections. This can cause a null pointer dereference in `getOutputSectionName`. Also delete debug-gnu-pubnames.s which is covered by gdb-index.s Reviewed By: grimar Differential Revision: https://reviews.llvm.org/D94354
2021-01-12 16:07:28 +08:00
.long .debug_info
[ELF] - Remove dependency on precompiled inputs from -gdb-index testcases. Previously we used precompiled objects in gdb-index.s and debug-gnu-pubnames.s testcases. We can avoid that. Differential revision: https://reviews.llvm.org/D35360 llvm-svn: 308005
2017-07-14 16:26:44 +08:00
.long 0x33
.long 0x2b
.byte 0x90
.string "int"
.long 0