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/ifunc.test

57 lines
1.7 KiB
Plaintext
Raw Normal View History

[lld][LayoutPass] This commit fixes a problem in the Layoutpass by ordering atoms if they appear in the override list first and then looking at the way of ordering atoms in the default way. The fix also fixes issues with the sizes of the sections, that appear in the output properly too. The commit also adds a testcase(orderatoms-by-override.test) to test it and fixes all the other relevant testcases. llvm-svn: 181605
2013-05-11 00:44:02 +08:00
# This test checks that IRELATIVE relocations are created for symbols that
# need relocation even for static links.
This is my Driver refactoring patch. The major changes are: 1) LinkerOptions has been merged into TargetInfo 2) LinkerInvocation has been merged into Driver 3) Drivers no longer convert arguments into an intermediate (core) argument list, but instead create a TargetInfo object and call setter methods on it. This is only how in-process linking would work. That is, you can programmatically set up a TargetInfo object which controls the linking. 4) Lots of tweaks to test suite to work with driver changes 5) Add the DarwinDriver 6) I heavily doxygen commented TargetInfo.h Things to do after this patch is committed: a) Consider renaming TargetInfo, given its new roll. b) Consider pulling the list of input files out of TargetInfo. This will enable in-process clients to create one TargetInfo the re-use it with different input file lists. c) Work out a way for Drivers to format the warnings and error done in core linking. llvm-svn: 178776
2013-04-05 02:59:24 +08:00
RUN: lld -flavor gnu -target x86_64-linux -emit-yaml -r \
RUN: %p/Inputs/ifunc.x86-64 | FileCheck %s
[ELF][x86-64] Fix ifunc and add test. In an previous commit I managed to completely disable the IRELATIVE relocation writing code. I also used the wrong addend for the static relocation. Fix both these issues and add a test. This test is quite brittle because there's no way to do arithmetic on variables in FileCheck. llvm-svn: 174161
2013-02-01 17:14:21 +08:00
This is my Driver refactoring patch. The major changes are: 1) LinkerOptions has been merged into TargetInfo 2) LinkerInvocation has been merged into Driver 3) Drivers no longer convert arguments into an intermediate (core) argument list, but instead create a TargetInfo object and call setter methods on it. This is only how in-process linking would work. That is, you can programmatically set up a TargetInfo object which controls the linking. 4) Lots of tweaks to test suite to work with driver changes 5) Add the DarwinDriver 6) I heavily doxygen commented TargetInfo.h Things to do after this patch is committed: a) Consider renaming TargetInfo, given its new roll. b) Consider pulling the list of input files out of TargetInfo. This will enable in-process clients to create one TargetInfo the re-use it with different input file lists. c) Work out a way for Drivers to format the warnings and error done in core linking. llvm-svn: 178776
2013-04-05 02:59:24 +08:00
RUN: lld -flavor gnu -target x86_64-linux -emit-yaml --noinhibit-exec \
RUN: %p/Inputs/ifunc.x86-64 %p/Inputs/ifunc.cpp.x86-64 \
[ELF][x86-64] Handle PLT32 relocations to IFUNC. llvm-svn: 174428
2013-02-06 03:15:03 +08:00
RUN: | FileCheck %s --check-prefix=PLT
[ELF] Read IFUNC symbols correctly as typeResolver. llvm-svn: 173299
2013-01-24 06:08:46 +08:00
This is my Driver refactoring patch. The major changes are: 1) LinkerOptions has been merged into TargetInfo 2) LinkerInvocation has been merged into Driver 3) Drivers no longer convert arguments into an intermediate (core) argument list, but instead create a TargetInfo object and call setter methods on it. This is only how in-process linking would work. That is, you can programmatically set up a TargetInfo object which controls the linking. 4) Lots of tweaks to test suite to work with driver changes 5) Add the DarwinDriver 6) I heavily doxygen commented TargetInfo.h Things to do after this patch is committed: a) Consider renaming TargetInfo, given its new roll. b) Consider pulling the list of input files out of TargetInfo. This will enable in-process clients to create one TargetInfo the re-use it with different input file lists. c) Work out a way for Drivers to format the warnings and error done in core linking. llvm-svn: 178776
2013-04-05 02:59:24 +08:00
RUN: lld -flavor gnu -target x86_64-linux -o %t %p/Inputs/ifunc.x86-64 \
RUN: -e main -static %p/Inputs/ifunc.cpp.x86-64 \
[Core,Driver,ELF] Differentiate static and dynamic executables. This also adds a simple relocation change for dynamic executables to x86-64 ELF. llvm-svn: 175208
2013-02-15 04:32:00 +08:00
RUN: && llvm-objdump -d -s %t| FileCheck %s --check-prefix=BIN
[lld][ELF][All Archs] Addend is used by dynamic relocations only if they are relative. This removes the FIXME when the relocations are being emitted and checks if the relocation is relative and only then populates the addend information. I couldnt add a testcase for this as llvm-readobj lacks functionality of printing dynamic relocations. When the functionality is added, remove the commented lines from elf/ifunc.test to test functionality. llvm-svn: 182077
2013-05-17 13:10:30 +08:00
#REMOVE THE BELOW LINE WHEN llvm-readobj adds functionality to print
#Dynamic relocations
#llvm-readobj -r %t | FileCheck %s --check-prefix=RELATIVEADDEND
This is my Driver refactoring patch. The major changes are: 1) LinkerOptions has been merged into TargetInfo 2) LinkerInvocation has been merged into Driver 3) Drivers no longer convert arguments into an intermediate (core) argument list, but instead create a TargetInfo object and call setter methods on it. This is only how in-process linking would work. That is, you can programmatically set up a TargetInfo object which controls the linking. 4) Lots of tweaks to test suite to work with driver changes 5) Add the DarwinDriver 6) I heavily doxygen commented TargetInfo.h Things to do after this patch is committed: a) Consider renaming TargetInfo, given its new roll. b) Consider pulling the list of input files out of TargetInfo. This will enable in-process clients to create one TargetInfo the re-use it with different input file lists. c) Work out a way for Drivers to format the warnings and error done in core linking. llvm-svn: 178776
2013-04-05 02:59:24 +08:00
PLT: defined-atoms:
PLT: name: plt
[ELF][x86-64] Handle PLT32 relocations to IFUNC. llvm-svn: 174428
2013-02-06 03:15:03 +08:00
PLT: scope: global
This is my Driver refactoring patch. The major changes are: 1) LinkerOptions has been merged into TargetInfo 2) LinkerInvocation has been merged into Driver 3) Drivers no longer convert arguments into an intermediate (core) argument list, but instead create a TargetInfo object and call setter methods on it. This is only how in-process linking would work. That is, you can programmatically set up a TargetInfo object which controls the linking. 4) Lots of tweaks to test suite to work with driver changes 5) Add the DarwinDriver 6) I heavily doxygen commented TargetInfo.h Things to do after this patch is committed: a) Consider renaming TargetInfo, given its new roll. b) Consider pulling the list of input files out of TargetInfo. This will enable in-process clients to create one TargetInfo the re-use it with different input file lists. c) Work out a way for Drivers to format the warnings and error done in core linking. llvm-svn: 178776
2013-04-05 02:59:24 +08:00
PLT: references:
PLT: kind: R_X86_64_PC32
PLT: target: [[PLTNAME:[-a-zA-Z0-9_]+]]
PLT: kind: layout-before
PLT: target: __hey_1
[lld][LayoutPass] This commit fixes a problem in the Layoutpass by ordering atoms if they appear in the override list first and then looking at the way of ordering atoms in the default way. The fix also fixes issues with the sizes of the sections, that appear in the output properly too. The commit also adds a testcase(orderatoms-by-override.test) to test it and fixes all the other relevant testcases. llvm-svn: 181605
2013-05-11 00:44:02 +08:00
[ELF] Add support for IFUNC. This sadly doesn't have a test for the final output because llvm-objdump can't dump relocations that don't belong to a section :( llvm-svn: 173808
2013-01-30 00:38:03 +08:00
PLT: name: main
PLT: scope: global
This is my Driver refactoring patch. The major changes are: 1) LinkerOptions has been merged into TargetInfo 2) LinkerInvocation has been merged into Driver 3) Drivers no longer convert arguments into an intermediate (core) argument list, but instead create a TargetInfo object and call setter methods on it. This is only how in-process linking would work. That is, you can programmatically set up a TargetInfo object which controls the linking. 4) Lots of tweaks to test suite to work with driver changes 5) Add the DarwinDriver 6) I heavily doxygen commented TargetInfo.h Things to do after this patch is committed: a) Consider renaming TargetInfo, given its new roll. b) Consider pulling the list of input files out of TargetInfo. This will enable in-process clients to create one TargetInfo the re-use it with different input file lists. c) Work out a way for Drivers to format the warnings and error done in core linking. llvm-svn: 178776
2013-04-05 02:59:24 +08:00
PLT: references:
PLT: kind: R_X86_64_PC32
PLT: target: [[PLTNAME]]
[ELF] Add support for IFUNC. This sadly doesn't have a test for the final output because llvm-objdump can't dump relocations that don't belong to a section :( llvm-svn: 173808
2013-01-30 00:38:03 +08:00
// Make sure the target of main's relocation is a stub with a PC32 relocation.
// This relocation is to the got atom, but you can't really write that check in
// FileCheck.
This is my Driver refactoring patch. The major changes are: 1) LinkerOptions has been merged into TargetInfo 2) LinkerInvocation has been merged into Driver 3) Drivers no longer convert arguments into an intermediate (core) argument list, but instead create a TargetInfo object and call setter methods on it. This is only how in-process linking would work. That is, you can programmatically set up a TargetInfo object which controls the linking. 4) Lots of tweaks to test suite to work with driver changes 5) Add the DarwinDriver 6) I heavily doxygen commented TargetInfo.h Things to do after this patch is committed: a) Consider renaming TargetInfo, given its new roll. b) Consider pulling the list of input files out of TargetInfo. This will enable in-process clients to create one TargetInfo the re-use it with different input file lists. c) Work out a way for Drivers to format the warnings and error done in core linking. llvm-svn: 178776
2013-04-05 02:59:24 +08:00
PLT: name:
[ELF] Add support for IFUNC. This sadly doesn't have a test for the final output because llvm-objdump can't dump relocations that don't belong to a section :( llvm-svn: 173808
2013-01-30 00:38:03 +08:00
PLT: type: stub
PLT: references
This is my Driver refactoring patch. The major changes are: 1) LinkerOptions has been merged into TargetInfo 2) LinkerInvocation has been merged into Driver 3) Drivers no longer convert arguments into an intermediate (core) argument list, but instead create a TargetInfo object and call setter methods on it. This is only how in-process linking would work. That is, you can programmatically set up a TargetInfo object which controls the linking. 4) Lots of tweaks to test suite to work with driver changes 5) Add the DarwinDriver 6) I heavily doxygen commented TargetInfo.h Things to do after this patch is committed: a) Consider renaming TargetInfo, given its new roll. b) Consider pulling the list of input files out of TargetInfo. This will enable in-process clients to create one TargetInfo the re-use it with different input file lists. c) Work out a way for Drivers to format the warnings and error done in core linking. llvm-svn: 178776
2013-04-05 02:59:24 +08:00
PLT: kind: R_X86_64_PC32
[lld][LayoutPass] An earlier commit moved the functionality so that Atoms would be laid out by their ordinal overrides first, there was a bug that two atoms may get the same override index due to which atoms were not ordered properly. This commit fixes the problem. Now the atoms are ordered by - Section Position hints - Atom override (Using layout-after/layout-before/in-group) - Content Permissions - Content Type - File Ordinal This also fixes the problem of running c++ static executables that was broken by an earlier patch. llvm-svn: 182494
2013-05-23 01:41:04 +08:00
// Make sure there's a got entry with a IRELATIVE relocation.
PLT: type: got
PLT: references:
PLT: kind: R_X86_64_IRELATIVE
PLT: target: hey
This is my Driver refactoring patch. The major changes are: 1) LinkerOptions has been merged into TargetInfo 2) LinkerInvocation has been merged into Driver 3) Drivers no longer convert arguments into an intermediate (core) argument list, but instead create a TargetInfo object and call setter methods on it. This is only how in-process linking would work. That is, you can programmatically set up a TargetInfo object which controls the linking. 4) Lots of tweaks to test suite to work with driver changes 5) Add the DarwinDriver 6) I heavily doxygen commented TargetInfo.h Things to do after this patch is committed: a) Consider renaming TargetInfo, given its new roll. b) Consider pulling the list of input files out of TargetInfo. This will enable in-process clients to create one TargetInfo the re-use it with different input file lists. c) Work out a way for Drivers to format the warnings and error done in core linking. llvm-svn: 178776
2013-04-05 02:59:24 +08:00
CHECK: name: hey
CHECK: scope: global
CHECK: type: resolver
[ELF][x86-64] Fix ifunc and add test. In an previous commit I managed to completely disable the IRELATIVE relocation writing code. I also used the wrong addend for the static relocation. Fix both these issues and add a test. This test is quite brittle because there's no way to do arithmetic on variables in FileCheck. llvm-svn: 174161
2013-02-01 17:14:21 +08:00
This is my Driver refactoring patch. The major changes are: 1) LinkerOptions has been merged into TargetInfo 2) LinkerInvocation has been merged into Driver 3) Drivers no longer convert arguments into an intermediate (core) argument list, but instead create a TargetInfo object and call setter methods on it. This is only how in-process linking would work. That is, you can programmatically set up a TargetInfo object which controls the linking. 4) Lots of tweaks to test suite to work with driver changes 5) Add the DarwinDriver 6) I heavily doxygen commented TargetInfo.h Things to do after this patch is committed: a) Consider renaming TargetInfo, given its new roll. b) Consider pulling the list of input files out of TargetInfo. This will enable in-process clients to create one TargetInfo the re-use it with different input file lists. c) Work out a way for Drivers to format the warnings and error done in core linking. llvm-svn: 178776
2013-04-05 02:59:24 +08:00
// This is a horribly brittle test. We need a way to do arithmetic on captured
// variables.
[ELF][x86-64] Separate static and dynamic GOT/PLT generation. llvm-svn: 176294
2013-03-01 04:04:32 +08:00
BIN: {{[0-9a-f]+}}: ff 25 {{[0-9a-f]+}} {{[0-9a-f]+}} 00 00 jmpq *{{[0-9]+}}(%rip)
[ELF][x86-64] Fix ifunc and add test. In an previous commit I managed to completely disable the IRELATIVE relocation writing code. I also used the wrong addend for the static relocation. Fix both these issues and add a test. This test is quite brittle because there's no way to do arithmetic on variables in FileCheck. llvm-svn: 174161
2013-02-01 17:14:21 +08:00
BIN: .got.plt:
[ELF][x86-64] Separate static and dynamic GOT/PLT generation. llvm-svn: 176294
2013-03-01 04:04:32 +08:00
BIN-NEXT: {{[0-9a-f]+}} 00000000 00000000