[llvm-libgcc][CMake] Refactor llvm-libgcc (#65455)
There are some issues in `llvm-libgcc` before this patch:
Commit c5a20b518203613497fa864867fc232648006068 ([llvm-libgcc] initial
commit)
uses `$<TARGET_OBJECTS:unwind_static>` to get libunwind objects, which
is empty.
The built library is actually a shared version of libclang_rt.builtins.
When configuring with `llvm/CMakeLists.txt`, target `llvm-libgcc`
requires a
corresponding target in `llvm-libgcc/CMakeLists.txt`.
Per target installation is not handled by `llvm-libgcc`, which is not
consistent
with `libunwind`.
This patch fixes those issues by:
Reusing target `unwind_shared` in `libunwind`, linking
`compiler-rt.builtins`
objects into it, and applying version script.
Adding target `llvm-libgcc`, creating symlinks, and utilizing cmake's
dependency
and component mechanism to ensure link targets will be built and
installed along
with symlinks.
Mimicking `libunwind` to handle per target installation.
It is quite hard to set necessary options without further modifying the
order of
runtime projects in `runtimes/CMakeLists.txt`. So though this patch
reveals the
possibility of co-existence of `llvm-libgcc` and
`compiler-rt`/`libunwind` in
`LLVM_ENABLE_RUNTIMES`, we still inhibit it to minimize influence on
other
projects, considering that `llvm-libgcc` is only intended for toolchain
vendors,
and not for casual use.
2023-09-19 13:56:03 +08:00
|
|
|
#===============================================================================
|
|
|
|
# Setup Project
|
|
|
|
#===============================================================================
|
[llvm-libgcc] initial commit
Note: the term "libgcc" refers to the all of `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`.
Enabling libunwind as a replacement for libgcc on Linux has proven to be
challenging since libgcc_s.so is a required dependency in the [Linux
standard base][5]. Some software is transitively dependent on libgcc
because glibc makes hardcoded calls to functions in libgcc_s. For example,
the function `__GI___backtrace` eventually makes its way to a [hardcoded
dlopen to libgcc_s' _Unwind_Backtrace][1]. Since libgcc_{eh.a,s.so} and
libunwind have the same ABI, but different implementations, the two
libraries end up [cross-talking, which ultimately results in a
segfault][2].
To solve this problem, libunwind needs to build a “libgcc”. That is, link
the necessary functions from compiler-rt and libunwind into an archive
and shared object that advertise themselves as `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`, so that glibc’s baked calls are diverted to the
correct objects in memory. Fortunately for us, compiler-rt and libunwind
use the same ABI as the libgcc family, so the problem is solvable at the
llvm-project configuration level: no program source needs to be edited.
Thus, the end result is for a user to configure their LLVM build with a
flag that indicates they want to archive compiler-rt/unwind as libgcc.
We achieve this by compiling libunwind with all the symbols necessary
for compiler-rt to emulate the libgcc family, and then generate symlinks
named for our "libgcc" that point to their corresponding libunwind
counterparts.
We alternatively considered patching glibc so that the source doesn't
directly refer to libgcc, but rather _defaults_ to libgcc, so that a
system preferring compiler-rt/libunwind can point to these libraries
at the config stage instead. Even if we modified the Linux standard
base, this alternative won't work because binaries that are built using
libgcc will still end up having crosstalk between the differing
implementations.
This problem has been solved in this manner for [FreeBSD][3], and this
CL has been tested against [Chrome OS][4].
[1]: https://github.com/bminor/glibc/blob/master/sysdeps/arm/backtrace.c#L68
[2]: https://bugs.chromium.org/p/chromium/issues/detail?id=1162190#c16
[3]: https://github.com/freebsd/freebsd-src/tree/main/lib/libgcc_s
[4]: https://chromium-review.googlesource.com/c/chromiumos/overlays/chromiumos-overlay/+/2945947
[5]: https://refspecs.linuxbase.org/LSB_5.0.0/LSB-Core-generic/LSB-Core-generic/libgcc-s.html
Differential Revision: https://reviews.llvm.org/D108416
2021-08-17 20:23:22 +00:00
|
|
|
|
[llvm-libgcc][CMake] Refactor llvm-libgcc (#65455)
There are some issues in `llvm-libgcc` before this patch:
Commit c5a20b518203613497fa864867fc232648006068 ([llvm-libgcc] initial
commit)
uses `$<TARGET_OBJECTS:unwind_static>` to get libunwind objects, which
is empty.
The built library is actually a shared version of libclang_rt.builtins.
When configuring with `llvm/CMakeLists.txt`, target `llvm-libgcc`
requires a
corresponding target in `llvm-libgcc/CMakeLists.txt`.
Per target installation is not handled by `llvm-libgcc`, which is not
consistent
with `libunwind`.
This patch fixes those issues by:
Reusing target `unwind_shared` in `libunwind`, linking
`compiler-rt.builtins`
objects into it, and applying version script.
Adding target `llvm-libgcc`, creating symlinks, and utilizing cmake's
dependency
and component mechanism to ensure link targets will be built and
installed along
with symlinks.
Mimicking `libunwind` to handle per target installation.
It is quite hard to set necessary options without further modifying the
order of
runtime projects in `runtimes/CMakeLists.txt`. So though this patch
reveals the
possibility of co-existence of `llvm-libgcc` and
`compiler-rt`/`libunwind` in
`LLVM_ENABLE_RUNTIMES`, we still inhibit it to minimize influence on
other
projects, considering that `llvm-libgcc` is only intended for toolchain
vendors,
and not for casual use.
2023-09-19 13:56:03 +08:00
|
|
|
cmake_minimum_required(VERSION 3.20.0)
|
2024-05-25 17:41:21 +02:00
|
|
|
set(LLVM_SUBPROJECT_TITLE "LLVM libgcc")
|
[llvm-libgcc] initial commit
Note: the term "libgcc" refers to the all of `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`.
Enabling libunwind as a replacement for libgcc on Linux has proven to be
challenging since libgcc_s.so is a required dependency in the [Linux
standard base][5]. Some software is transitively dependent on libgcc
because glibc makes hardcoded calls to functions in libgcc_s. For example,
the function `__GI___backtrace` eventually makes its way to a [hardcoded
dlopen to libgcc_s' _Unwind_Backtrace][1]. Since libgcc_{eh.a,s.so} and
libunwind have the same ABI, but different implementations, the two
libraries end up [cross-talking, which ultimately results in a
segfault][2].
To solve this problem, libunwind needs to build a “libgcc”. That is, link
the necessary functions from compiler-rt and libunwind into an archive
and shared object that advertise themselves as `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`, so that glibc’s baked calls are diverted to the
correct objects in memory. Fortunately for us, compiler-rt and libunwind
use the same ABI as the libgcc family, so the problem is solvable at the
llvm-project configuration level: no program source needs to be edited.
Thus, the end result is for a user to configure their LLVM build with a
flag that indicates they want to archive compiler-rt/unwind as libgcc.
We achieve this by compiling libunwind with all the symbols necessary
for compiler-rt to emulate the libgcc family, and then generate symlinks
named for our "libgcc" that point to their corresponding libunwind
counterparts.
We alternatively considered patching glibc so that the source doesn't
directly refer to libgcc, but rather _defaults_ to libgcc, so that a
system preferring compiler-rt/libunwind can point to these libraries
at the config stage instead. Even if we modified the Linux standard
base, this alternative won't work because binaries that are built using
libgcc will still end up having crosstalk between the differing
implementations.
This problem has been solved in this manner for [FreeBSD][3], and this
CL has been tested against [Chrome OS][4].
[1]: https://github.com/bminor/glibc/blob/master/sysdeps/arm/backtrace.c#L68
[2]: https://bugs.chromium.org/p/chromium/issues/detail?id=1162190#c16
[3]: https://github.com/freebsd/freebsd-src/tree/main/lib/libgcc_s
[4]: https://chromium-review.googlesource.com/c/chromiumos/overlays/chromiumos-overlay/+/2945947
[5]: https://refspecs.linuxbase.org/LSB_5.0.0/LSB-Core-generic/LSB-Core-generic/libgcc-s.html
Differential Revision: https://reviews.llvm.org/D108416
2021-08-17 20:23:22 +00:00
|
|
|
|
|
|
|
set(LLVM_COMMON_CMAKE_UTILS "${CMAKE_CURRENT_SOURCE_DIR}/../cmake")
|
|
|
|
|
[llvm-libgcc][CMake] Refactor llvm-libgcc (#65455)
There are some issues in `llvm-libgcc` before this patch:
Commit c5a20b518203613497fa864867fc232648006068 ([llvm-libgcc] initial
commit)
uses `$<TARGET_OBJECTS:unwind_static>` to get libunwind objects, which
is empty.
The built library is actually a shared version of libclang_rt.builtins.
When configuring with `llvm/CMakeLists.txt`, target `llvm-libgcc`
requires a
corresponding target in `llvm-libgcc/CMakeLists.txt`.
Per target installation is not handled by `llvm-libgcc`, which is not
consistent
with `libunwind`.
This patch fixes those issues by:
Reusing target `unwind_shared` in `libunwind`, linking
`compiler-rt.builtins`
objects into it, and applying version script.
Adding target `llvm-libgcc`, creating symlinks, and utilizing cmake's
dependency
and component mechanism to ensure link targets will be built and
installed along
with symlinks.
Mimicking `libunwind` to handle per target installation.
It is quite hard to set necessary options without further modifying the
order of
runtime projects in `runtimes/CMakeLists.txt`. So though this patch
reveals the
possibility of co-existence of `llvm-libgcc` and
`compiler-rt`/`libunwind` in
`LLVM_ENABLE_RUNTIMES`, we still inhibit it to minimize influence on
other
projects, considering that `llvm-libgcc` is only intended for toolchain
vendors,
and not for casual use.
2023-09-19 13:56:03 +08:00
|
|
|
# Check if llvm-libgcc is built as a standalone project
|
|
|
|
if(CMAKE_SOURCE_DIR STREQUAL CMAKE_CURRENT_SOURCE_DIR OR LLVM_LIBGCC_STANDALONE_BUILD)
|
|
|
|
project(llvm-libgcc LANGUAGES C CXX ASM)
|
|
|
|
set(COMPILER_RT_STANDALONE_BUILD ON)
|
|
|
|
set_property(GLOBAL PROPERTY USE_FOLDERS ON)
|
|
|
|
set(LLVM_LIBGCC_COMPILER_RT_BINARY_DIR "compiler-rt")
|
|
|
|
set(LLVM_LIBGCC_LIBUNWIND_BINARY_DIR "libunwind")
|
|
|
|
else()
|
|
|
|
set(LLVM_LIBGCC_COMPILER_RT_BINARY_DIR "../compiler-rt")
|
|
|
|
set(LLVM_LIBGCC_LIBUNWIND_BINARY_DIR "../libunwind")
|
|
|
|
endif()
|
|
|
|
|
|
|
|
# Add path for custom modules
|
|
|
|
list(INSERT CMAKE_MODULE_PATH 0
|
[llvm-libgcc] initial commit
Note: the term "libgcc" refers to the all of `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`.
Enabling libunwind as a replacement for libgcc on Linux has proven to be
challenging since libgcc_s.so is a required dependency in the [Linux
standard base][5]. Some software is transitively dependent on libgcc
because glibc makes hardcoded calls to functions in libgcc_s. For example,
the function `__GI___backtrace` eventually makes its way to a [hardcoded
dlopen to libgcc_s' _Unwind_Backtrace][1]. Since libgcc_{eh.a,s.so} and
libunwind have the same ABI, but different implementations, the two
libraries end up [cross-talking, which ultimately results in a
segfault][2].
To solve this problem, libunwind needs to build a “libgcc”. That is, link
the necessary functions from compiler-rt and libunwind into an archive
and shared object that advertise themselves as `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`, so that glibc’s baked calls are diverted to the
correct objects in memory. Fortunately for us, compiler-rt and libunwind
use the same ABI as the libgcc family, so the problem is solvable at the
llvm-project configuration level: no program source needs to be edited.
Thus, the end result is for a user to configure their LLVM build with a
flag that indicates they want to archive compiler-rt/unwind as libgcc.
We achieve this by compiling libunwind with all the symbols necessary
for compiler-rt to emulate the libgcc family, and then generate symlinks
named for our "libgcc" that point to their corresponding libunwind
counterparts.
We alternatively considered patching glibc so that the source doesn't
directly refer to libgcc, but rather _defaults_ to libgcc, so that a
system preferring compiler-rt/libunwind can point to these libraries
at the config stage instead. Even if we modified the Linux standard
base, this alternative won't work because binaries that are built using
libgcc will still end up having crosstalk between the differing
implementations.
This problem has been solved in this manner for [FreeBSD][3], and this
CL has been tested against [Chrome OS][4].
[1]: https://github.com/bminor/glibc/blob/master/sysdeps/arm/backtrace.c#L68
[2]: https://bugs.chromium.org/p/chromium/issues/detail?id=1162190#c16
[3]: https://github.com/freebsd/freebsd-src/tree/main/lib/libgcc_s
[4]: https://chromium-review.googlesource.com/c/chromiumos/overlays/chromiumos-overlay/+/2945947
[5]: https://refspecs.linuxbase.org/LSB_5.0.0/LSB-Core-generic/LSB-Core-generic/libgcc-s.html
Differential Revision: https://reviews.llvm.org/D108416
2021-08-17 20:23:22 +00:00
|
|
|
"${CMAKE_CURRENT_SOURCE_DIR}/cmake"
|
|
|
|
"${CMAKE_CURRENT_SOURCE_DIR}/cmake/Modules"
|
[llvm-libgcc][CMake] Refactor llvm-libgcc (#65455)
There are some issues in `llvm-libgcc` before this patch:
Commit c5a20b518203613497fa864867fc232648006068 ([llvm-libgcc] initial
commit)
uses `$<TARGET_OBJECTS:unwind_static>` to get libunwind objects, which
is empty.
The built library is actually a shared version of libclang_rt.builtins.
When configuring with `llvm/CMakeLists.txt`, target `llvm-libgcc`
requires a
corresponding target in `llvm-libgcc/CMakeLists.txt`.
Per target installation is not handled by `llvm-libgcc`, which is not
consistent
with `libunwind`.
This patch fixes those issues by:
Reusing target `unwind_shared` in `libunwind`, linking
`compiler-rt.builtins`
objects into it, and applying version script.
Adding target `llvm-libgcc`, creating symlinks, and utilizing cmake's
dependency
and component mechanism to ensure link targets will be built and
installed along
with symlinks.
Mimicking `libunwind` to handle per target installation.
It is quite hard to set necessary options without further modifying the
order of
runtime projects in `runtimes/CMakeLists.txt`. So though this patch
reveals the
possibility of co-existence of `llvm-libgcc` and
`compiler-rt`/`libunwind` in
`LLVM_ENABLE_RUNTIMES`, we still inhibit it to minimize influence on
other
projects, considering that `llvm-libgcc` is only intended for toolchain
vendors,
and not for casual use.
2023-09-19 13:56:03 +08:00
|
|
|
"${CMAKE_CURRENT_SOURCE_DIR}/../runtimes/cmake/Modules"
|
[llvm-libgcc] initial commit
Note: the term "libgcc" refers to the all of `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`.
Enabling libunwind as a replacement for libgcc on Linux has proven to be
challenging since libgcc_s.so is a required dependency in the [Linux
standard base][5]. Some software is transitively dependent on libgcc
because glibc makes hardcoded calls to functions in libgcc_s. For example,
the function `__GI___backtrace` eventually makes its way to a [hardcoded
dlopen to libgcc_s' _Unwind_Backtrace][1]. Since libgcc_{eh.a,s.so} and
libunwind have the same ABI, but different implementations, the two
libraries end up [cross-talking, which ultimately results in a
segfault][2].
To solve this problem, libunwind needs to build a “libgcc”. That is, link
the necessary functions from compiler-rt and libunwind into an archive
and shared object that advertise themselves as `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`, so that glibc’s baked calls are diverted to the
correct objects in memory. Fortunately for us, compiler-rt and libunwind
use the same ABI as the libgcc family, so the problem is solvable at the
llvm-project configuration level: no program source needs to be edited.
Thus, the end result is for a user to configure their LLVM build with a
flag that indicates they want to archive compiler-rt/unwind as libgcc.
We achieve this by compiling libunwind with all the symbols necessary
for compiler-rt to emulate the libgcc family, and then generate symlinks
named for our "libgcc" that point to their corresponding libunwind
counterparts.
We alternatively considered patching glibc so that the source doesn't
directly refer to libgcc, but rather _defaults_ to libgcc, so that a
system preferring compiler-rt/libunwind can point to these libraries
at the config stage instead. Even if we modified the Linux standard
base, this alternative won't work because binaries that are built using
libgcc will still end up having crosstalk between the differing
implementations.
This problem has been solved in this manner for [FreeBSD][3], and this
CL has been tested against [Chrome OS][4].
[1]: https://github.com/bminor/glibc/blob/master/sysdeps/arm/backtrace.c#L68
[2]: https://bugs.chromium.org/p/chromium/issues/detail?id=1162190#c16
[3]: https://github.com/freebsd/freebsd-src/tree/main/lib/libgcc_s
[4]: https://chromium-review.googlesource.com/c/chromiumos/overlays/chromiumos-overlay/+/2945947
[5]: https://refspecs.linuxbase.org/LSB_5.0.0/LSB-Core-generic/LSB-Core-generic/libgcc-s.html
Differential Revision: https://reviews.llvm.org/D108416
2021-08-17 20:23:22 +00:00
|
|
|
"${LLVM_COMMON_CMAKE_UTILS}"
|
|
|
|
"${LLVM_COMMON_CMAKE_UTILS}/Modules"
|
[llvm-libgcc][CMake] Refactor llvm-libgcc (#65455)
There are some issues in `llvm-libgcc` before this patch:
Commit c5a20b518203613497fa864867fc232648006068 ([llvm-libgcc] initial
commit)
uses `$<TARGET_OBJECTS:unwind_static>` to get libunwind objects, which
is empty.
The built library is actually a shared version of libclang_rt.builtins.
When configuring with `llvm/CMakeLists.txt`, target `llvm-libgcc`
requires a
corresponding target in `llvm-libgcc/CMakeLists.txt`.
Per target installation is not handled by `llvm-libgcc`, which is not
consistent
with `libunwind`.
This patch fixes those issues by:
Reusing target `unwind_shared` in `libunwind`, linking
`compiler-rt.builtins`
objects into it, and applying version script.
Adding target `llvm-libgcc`, creating symlinks, and utilizing cmake's
dependency
and component mechanism to ensure link targets will be built and
installed along
with symlinks.
Mimicking `libunwind` to handle per target installation.
It is quite hard to set necessary options without further modifying the
order of
runtime projects in `runtimes/CMakeLists.txt`. So though this patch
reveals the
possibility of co-existence of `llvm-libgcc` and
`compiler-rt`/`libunwind` in
`LLVM_ENABLE_RUNTIMES`, we still inhibit it to minimize influence on
other
projects, considering that `llvm-libgcc` is only intended for toolchain
vendors,
and not for casual use.
2023-09-19 13:56:03 +08:00
|
|
|
)
|
|
|
|
|
|
|
|
set(LLVM_LIBGCC_LIBUNWIND_PATH "${CMAKE_CURRENT_LIST_DIR}/../libunwind"
|
|
|
|
CACHE PATH "Specify path to libunwind source.")
|
|
|
|
set(LLVM_LIBGCC_COMPILER_RT_PATH "${CMAKE_CURRENT_LIST_DIR}/../compiler-rt"
|
|
|
|
CACHE PATH "Specify path to compiler-rt source.")
|
[llvm-libgcc] initial commit
Note: the term "libgcc" refers to the all of `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`.
Enabling libunwind as a replacement for libgcc on Linux has proven to be
challenging since libgcc_s.so is a required dependency in the [Linux
standard base][5]. Some software is transitively dependent on libgcc
because glibc makes hardcoded calls to functions in libgcc_s. For example,
the function `__GI___backtrace` eventually makes its way to a [hardcoded
dlopen to libgcc_s' _Unwind_Backtrace][1]. Since libgcc_{eh.a,s.so} and
libunwind have the same ABI, but different implementations, the two
libraries end up [cross-talking, which ultimately results in a
segfault][2].
To solve this problem, libunwind needs to build a “libgcc”. That is, link
the necessary functions from compiler-rt and libunwind into an archive
and shared object that advertise themselves as `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`, so that glibc’s baked calls are diverted to the
correct objects in memory. Fortunately for us, compiler-rt and libunwind
use the same ABI as the libgcc family, so the problem is solvable at the
llvm-project configuration level: no program source needs to be edited.
Thus, the end result is for a user to configure their LLVM build with a
flag that indicates they want to archive compiler-rt/unwind as libgcc.
We achieve this by compiling libunwind with all the symbols necessary
for compiler-rt to emulate the libgcc family, and then generate symlinks
named for our "libgcc" that point to their corresponding libunwind
counterparts.
We alternatively considered patching glibc so that the source doesn't
directly refer to libgcc, but rather _defaults_ to libgcc, so that a
system preferring compiler-rt/libunwind can point to these libraries
at the config stage instead. Even if we modified the Linux standard
base, this alternative won't work because binaries that are built using
libgcc will still end up having crosstalk between the differing
implementations.
This problem has been solved in this manner for [FreeBSD][3], and this
CL has been tested against [Chrome OS][4].
[1]: https://github.com/bminor/glibc/blob/master/sysdeps/arm/backtrace.c#L68
[2]: https://bugs.chromium.org/p/chromium/issues/detail?id=1162190#c16
[3]: https://github.com/freebsd/freebsd-src/tree/main/lib/libgcc_s
[4]: https://chromium-review.googlesource.com/c/chromiumos/overlays/chromiumos-overlay/+/2945947
[5]: https://refspecs.linuxbase.org/LSB_5.0.0/LSB-Core-generic/LSB-Core-generic/libgcc-s.html
Differential Revision: https://reviews.llvm.org/D108416
2021-08-17 20:23:22 +00:00
|
|
|
|
[llvm-libgcc][CMake] Refactor llvm-libgcc (#65455)
There are some issues in `llvm-libgcc` before this patch:
Commit c5a20b518203613497fa864867fc232648006068 ([llvm-libgcc] initial
commit)
uses `$<TARGET_OBJECTS:unwind_static>` to get libunwind objects, which
is empty.
The built library is actually a shared version of libclang_rt.builtins.
When configuring with `llvm/CMakeLists.txt`, target `llvm-libgcc`
requires a
corresponding target in `llvm-libgcc/CMakeLists.txt`.
Per target installation is not handled by `llvm-libgcc`, which is not
consistent
with `libunwind`.
This patch fixes those issues by:
Reusing target `unwind_shared` in `libunwind`, linking
`compiler-rt.builtins`
objects into it, and applying version script.
Adding target `llvm-libgcc`, creating symlinks, and utilizing cmake's
dependency
and component mechanism to ensure link targets will be built and
installed along
with symlinks.
Mimicking `libunwind` to handle per target installation.
It is quite hard to set necessary options without further modifying the
order of
runtime projects in `runtimes/CMakeLists.txt`. So though this patch
reveals the
possibility of co-existence of `llvm-libgcc` and
`compiler-rt`/`libunwind` in
`LLVM_ENABLE_RUNTIMES`, we still inhibit it to minimize influence on
other
projects, considering that `llvm-libgcc` is only intended for toolchain
vendors,
and not for casual use.
2023-09-19 13:56:03 +08:00
|
|
|
include(GNUInstallDirs)
|
[llvm-libgcc] initial commit
Note: the term "libgcc" refers to the all of `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`.
Enabling libunwind as a replacement for libgcc on Linux has proven to be
challenging since libgcc_s.so is a required dependency in the [Linux
standard base][5]. Some software is transitively dependent on libgcc
because glibc makes hardcoded calls to functions in libgcc_s. For example,
the function `__GI___backtrace` eventually makes its way to a [hardcoded
dlopen to libgcc_s' _Unwind_Backtrace][1]. Since libgcc_{eh.a,s.so} and
libunwind have the same ABI, but different implementations, the two
libraries end up [cross-talking, which ultimately results in a
segfault][2].
To solve this problem, libunwind needs to build a “libgcc”. That is, link
the necessary functions from compiler-rt and libunwind into an archive
and shared object that advertise themselves as `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`, so that glibc’s baked calls are diverted to the
correct objects in memory. Fortunately for us, compiler-rt and libunwind
use the same ABI as the libgcc family, so the problem is solvable at the
llvm-project configuration level: no program source needs to be edited.
Thus, the end result is for a user to configure their LLVM build with a
flag that indicates they want to archive compiler-rt/unwind as libgcc.
We achieve this by compiling libunwind with all the symbols necessary
for compiler-rt to emulate the libgcc family, and then generate symlinks
named for our "libgcc" that point to their corresponding libunwind
counterparts.
We alternatively considered patching glibc so that the source doesn't
directly refer to libgcc, but rather _defaults_ to libgcc, so that a
system preferring compiler-rt/libunwind can point to these libraries
at the config stage instead. Even if we modified the Linux standard
base, this alternative won't work because binaries that are built using
libgcc will still end up having crosstalk between the differing
implementations.
This problem has been solved in this manner for [FreeBSD][3], and this
CL has been tested against [Chrome OS][4].
[1]: https://github.com/bminor/glibc/blob/master/sysdeps/arm/backtrace.c#L68
[2]: https://bugs.chromium.org/p/chromium/issues/detail?id=1162190#c16
[3]: https://github.com/freebsd/freebsd-src/tree/main/lib/libgcc_s
[4]: https://chromium-review.googlesource.com/c/chromiumos/overlays/chromiumos-overlay/+/2945947
[5]: https://refspecs.linuxbase.org/LSB_5.0.0/LSB-Core-generic/LSB-Core-generic/libgcc-s.html
Differential Revision: https://reviews.llvm.org/D108416
2021-08-17 20:23:22 +00:00
|
|
|
|
|
|
|
if(NOT LLVM_LIBGCC_EXPLICIT_OPT_IN)
|
|
|
|
message(FATAL_ERROR
|
|
|
|
"llvm-libgcc is not for the casual LLVM user. It is intended to be used by distro "
|
|
|
|
"managers who want to replace libgcc with compiler-rt and libunwind, but cannot "
|
|
|
|
"fully abandon the libgcc family (e.g. because they are dependent on glibc). Such "
|
|
|
|
"managers must have worked out their compatibility requirements ahead of using "
|
|
|
|
"llvm-libgcc. If you want to build llvm-libgcc, please add -DLLVM_LIBGCC_EXPLICIT_OPT_IN=Yes "
|
|
|
|
"to your CMake invocation and try again.")
|
|
|
|
endif()
|
|
|
|
|
[llvm-libgcc][CMake] Refactor llvm-libgcc (#65455)
There are some issues in `llvm-libgcc` before this patch:
Commit c5a20b518203613497fa864867fc232648006068 ([llvm-libgcc] initial
commit)
uses `$<TARGET_OBJECTS:unwind_static>` to get libunwind objects, which
is empty.
The built library is actually a shared version of libclang_rt.builtins.
When configuring with `llvm/CMakeLists.txt`, target `llvm-libgcc`
requires a
corresponding target in `llvm-libgcc/CMakeLists.txt`.
Per target installation is not handled by `llvm-libgcc`, which is not
consistent
with `libunwind`.
This patch fixes those issues by:
Reusing target `unwind_shared` in `libunwind`, linking
`compiler-rt.builtins`
objects into it, and applying version script.
Adding target `llvm-libgcc`, creating symlinks, and utilizing cmake's
dependency
and component mechanism to ensure link targets will be built and
installed along
with symlinks.
Mimicking `libunwind` to handle per target installation.
It is quite hard to set necessary options without further modifying the
order of
runtime projects in `runtimes/CMakeLists.txt`. So though this patch
reveals the
possibility of co-existence of `llvm-libgcc` and
`compiler-rt`/`libunwind` in
`LLVM_ENABLE_RUNTIMES`, we still inhibit it to minimize influence on
other
projects, considering that `llvm-libgcc` is only intended for toolchain
vendors,
and not for casual use.
2023-09-19 13:56:03 +08:00
|
|
|
if(HAVE_COMPILER_RT)
|
|
|
|
message(FATAL_ERROR
|
|
|
|
"Attempting to build both compiler-rt and llvm-libgcc will cause irreconcilable "
|
|
|
|
"target clashes. Please choose one or the other, but not both.")
|
|
|
|
endif()
|
|
|
|
|
|
|
|
if(HAVE_LIBUNWIND)
|
|
|
|
message(FATAL_ERROR
|
|
|
|
"Attempting to build both libunwind and llvm-libgcc will cause irreconcilable "
|
|
|
|
"target clashes. Please choose one or the other, but not both.")
|
|
|
|
endif()
|
|
|
|
|
|
|
|
#===============================================================================
|
|
|
|
# Configure System
|
|
|
|
#===============================================================================
|
|
|
|
|
|
|
|
if(LLVM_ENABLE_PER_TARGET_RUNTIME_DIR AND NOT APPLE)
|
2024-05-31 11:48:45 -07:00
|
|
|
set(TARGET_SUBDIR ${LLVM_DEFAULT_TARGET_TRIPLE})
|
|
|
|
if(LLVM_LIBGCC_LIBDIR_SUBDIR)
|
|
|
|
string(APPEND TARGET_SUBDIR /${LLVM_LIBGCC_LIBDIR_SUBDIR})
|
[llvm-libgcc][CMake] Refactor llvm-libgcc (#65455)
There are some issues in `llvm-libgcc` before this patch:
Commit c5a20b518203613497fa864867fc232648006068 ([llvm-libgcc] initial
commit)
uses `$<TARGET_OBJECTS:unwind_static>` to get libunwind objects, which
is empty.
The built library is actually a shared version of libclang_rt.builtins.
When configuring with `llvm/CMakeLists.txt`, target `llvm-libgcc`
requires a
corresponding target in `llvm-libgcc/CMakeLists.txt`.
Per target installation is not handled by `llvm-libgcc`, which is not
consistent
with `libunwind`.
This patch fixes those issues by:
Reusing target `unwind_shared` in `libunwind`, linking
`compiler-rt.builtins`
objects into it, and applying version script.
Adding target `llvm-libgcc`, creating symlinks, and utilizing cmake's
dependency
and component mechanism to ensure link targets will be built and
installed along
with symlinks.
Mimicking `libunwind` to handle per target installation.
It is quite hard to set necessary options without further modifying the
order of
runtime projects in `runtimes/CMakeLists.txt`. So though this patch
reveals the
possibility of co-existence of `llvm-libgcc` and
`compiler-rt`/`libunwind` in
`LLVM_ENABLE_RUNTIMES`, we still inhibit it to minimize influence on
other
projects, considering that `llvm-libgcc` is only intended for toolchain
vendors,
and not for casual use.
2023-09-19 13:56:03 +08:00
|
|
|
endif()
|
2024-05-31 11:48:45 -07:00
|
|
|
set(LLVM_LIBGCC_LIBRARY_DIR ${LLVM_LIBRARY_OUTPUT_INTDIR}/${TARGET_SUBDIR})
|
|
|
|
set(LLVM_LIBGCC_INSTALL_LIBRARY_DIR lib${LLVM_LIBDIR_SUFFIX}/${TARGET_SUBDIR} CACHE PATH
|
|
|
|
"Path where built llvm-libgcc libraries should be installed.")
|
|
|
|
unset(TARGET_SUBDIR)
|
[llvm-libgcc][CMake] Refactor llvm-libgcc (#65455)
There are some issues in `llvm-libgcc` before this patch:
Commit c5a20b518203613497fa864867fc232648006068 ([llvm-libgcc] initial
commit)
uses `$<TARGET_OBJECTS:unwind_static>` to get libunwind objects, which
is empty.
The built library is actually a shared version of libclang_rt.builtins.
When configuring with `llvm/CMakeLists.txt`, target `llvm-libgcc`
requires a
corresponding target in `llvm-libgcc/CMakeLists.txt`.
Per target installation is not handled by `llvm-libgcc`, which is not
consistent
with `libunwind`.
This patch fixes those issues by:
Reusing target `unwind_shared` in `libunwind`, linking
`compiler-rt.builtins`
objects into it, and applying version script.
Adding target `llvm-libgcc`, creating symlinks, and utilizing cmake's
dependency
and component mechanism to ensure link targets will be built and
installed along
with symlinks.
Mimicking `libunwind` to handle per target installation.
It is quite hard to set necessary options without further modifying the
order of
runtime projects in `runtimes/CMakeLists.txt`. So though this patch
reveals the
possibility of co-existence of `llvm-libgcc` and
`compiler-rt`/`libunwind` in
`LLVM_ENABLE_RUNTIMES`, we still inhibit it to minimize influence on
other
projects, considering that `llvm-libgcc` is only intended for toolchain
vendors,
and not for casual use.
2023-09-19 13:56:03 +08:00
|
|
|
else()
|
|
|
|
if(LLVM_LIBRARY_OUTPUT_INTDIR)
|
|
|
|
set(LLVM_LIBGCC_LIBRARY_DIR ${LLVM_LIBRARY_OUTPUT_INTDIR})
|
|
|
|
else()
|
|
|
|
set(LLVM_LIBGCC_LIBRARY_DIR ${CMAKE_BINARY_DIR}/lib${LLVM_LIBGCC_LIBDIR_SUFFIX})
|
|
|
|
endif()
|
|
|
|
set(LLVM_LIBGCC_INSTALL_LIBRARY_DIR lib${LLVM_LIBGCC_LIBDIR_SUFFIX} CACHE PATH
|
|
|
|
"Path where built llvm-libgcc libraries should be installed.")
|
|
|
|
endif()
|
|
|
|
|
|
|
|
set(CMAKE_ARCHIVE_OUTPUT_DIRECTORY ${LLVM_LIBGCC_LIBRARY_DIR})
|
|
|
|
set(CMAKE_LIBRARY_OUTPUT_DIRECTORY ${LLVM_LIBGCC_LIBRARY_DIR})
|
|
|
|
set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${LLVM_LIBGCC_LIBRARY_DIR})
|
|
|
|
|
|
|
|
#===============================================================================
|
|
|
|
# Build libraries
|
|
|
|
#===============================================================================
|
|
|
|
|
|
|
|
set(COMPILER_RT_BUILD_BUILTINS ON)
|
|
|
|
set(COMPILER_RT_BUILTINS_HIDE_SYMBOLS OFF)
|
|
|
|
add_subdirectory(${LLVM_LIBGCC_COMPILER_RT_PATH} ${LLVM_LIBGCC_COMPILER_RT_BINARY_DIR})
|
|
|
|
|
|
|
|
set(LIBUNWIND_ENABLE_STATIC ON)
|
|
|
|
set(LIBUNWIND_ENABLE_SHARED ON)
|
|
|
|
set(LIBUNWIND_USE_COMPILER_RT OFF)
|
|
|
|
set(LIBUNWIND_HAS_GCC_LIB OFF)
|
|
|
|
set(LIBUNWIND_HAS_GCC_S_LIB OFF)
|
|
|
|
add_subdirectory(${LLVM_LIBGCC_LIBUNWIND_PATH} ${LLVM_LIBGCC_LIBUNWIND_BINARY_DIR})
|
|
|
|
|
|
|
|
add_custom_target(gcc_s.ver
|
|
|
|
SOURCES ${CMAKE_CURRENT_SOURCE_DIR}/gcc_s.ver.in
|
|
|
|
COMMAND ${CMAKE_C_COMPILER} -E
|
|
|
|
-xc ${CMAKE_CURRENT_SOURCE_DIR}/gcc_s.ver.in
|
|
|
|
-o ${CMAKE_CURRENT_BINARY_DIR}/gcc_s.ver
|
|
|
|
)
|
|
|
|
|
|
|
|
add_dependencies(unwind_shared gcc_s.ver)
|
|
|
|
|
|
|
|
construct_compiler_rt_default_triple()
|
|
|
|
|
|
|
|
target_link_options(unwind_shared PUBLIC
|
|
|
|
-Wl,--version-script,${CMAKE_CURRENT_BINARY_DIR}/gcc_s.ver
|
|
|
|
)
|
|
|
|
|
|
|
|
target_link_libraries(unwind_shared PUBLIC
|
|
|
|
$<TARGET_OBJECTS:clang_rt.builtins-${COMPILER_RT_DEFAULT_TARGET_ARCH}>
|
|
|
|
m
|
|
|
|
)
|
|
|
|
|
|
|
|
#===============================================================================
|
|
|
|
# Install Symlinks
|
|
|
|
#===============================================================================
|
|
|
|
|
|
|
|
get_compiler_rt_install_dir(${COMPILER_RT_DEFAULT_TARGET_ARCH} install_dir_builtins)
|
|
|
|
string(REGEX REPLACE "^lib/" "" install_dir_builtins "${install_dir_builtins}")
|
|
|
|
string(FIND "${install_dir_builtins}" "clang" install_path_contains_triple)
|
|
|
|
if(install_path_contains_triple EQUAL -1)
|
|
|
|
set(builtins_suffix "-${COMPILER_RT_DEFAULT_TARGET_ARCH}")
|
|
|
|
else()
|
|
|
|
string(PREPEND install_dir_builtins "../")
|
|
|
|
endif()
|
|
|
|
set(LLVM_LIBGCC_COMPILER_RT ${install_dir_builtins}/libclang_rt.builtins${builtins_suffix}.a)
|
|
|
|
|
|
|
|
add_custom_target(llvm-libgcc ALL
|
|
|
|
DEPENDS unwind_shared unwind_static clang_rt.builtins-${COMPILER_RT_DEFAULT_TARGET_ARCH}
|
|
|
|
COMMAND ${CMAKE_COMMAND} -E create_symlink ${LLVM_LIBGCC_COMPILER_RT} libgcc.a
|
|
|
|
COMMAND ${CMAKE_COMMAND} -E create_symlink libunwind.a libgcc_eh.a
|
|
|
|
COMMAND ${CMAKE_COMMAND} -E create_symlink libunwind.so libgcc_s.so.1.0
|
|
|
|
COMMAND ${CMAKE_COMMAND} -E create_symlink libgcc_s.so.1.0 libgcc_s.so.1
|
|
|
|
COMMAND ${CMAKE_COMMAND} -E create_symlink libgcc_s.so.1 libgcc_s.so
|
|
|
|
)
|
[llvm-libgcc] initial commit
Note: the term "libgcc" refers to the all of `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`.
Enabling libunwind as a replacement for libgcc on Linux has proven to be
challenging since libgcc_s.so is a required dependency in the [Linux
standard base][5]. Some software is transitively dependent on libgcc
because glibc makes hardcoded calls to functions in libgcc_s. For example,
the function `__GI___backtrace` eventually makes its way to a [hardcoded
dlopen to libgcc_s' _Unwind_Backtrace][1]. Since libgcc_{eh.a,s.so} and
libunwind have the same ABI, but different implementations, the two
libraries end up [cross-talking, which ultimately results in a
segfault][2].
To solve this problem, libunwind needs to build a “libgcc”. That is, link
the necessary functions from compiler-rt and libunwind into an archive
and shared object that advertise themselves as `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`, so that glibc’s baked calls are diverted to the
correct objects in memory. Fortunately for us, compiler-rt and libunwind
use the same ABI as the libgcc family, so the problem is solvable at the
llvm-project configuration level: no program source needs to be edited.
Thus, the end result is for a user to configure their LLVM build with a
flag that indicates they want to archive compiler-rt/unwind as libgcc.
We achieve this by compiling libunwind with all the symbols necessary
for compiler-rt to emulate the libgcc family, and then generate symlinks
named for our "libgcc" that point to their corresponding libunwind
counterparts.
We alternatively considered patching glibc so that the source doesn't
directly refer to libgcc, but rather _defaults_ to libgcc, so that a
system preferring compiler-rt/libunwind can point to these libraries
at the config stage instead. Even if we modified the Linux standard
base, this alternative won't work because binaries that are built using
libgcc will still end up having crosstalk between the differing
implementations.
This problem has been solved in this manner for [FreeBSD][3], and this
CL has been tested against [Chrome OS][4].
[1]: https://github.com/bminor/glibc/blob/master/sysdeps/arm/backtrace.c#L68
[2]: https://bugs.chromium.org/p/chromium/issues/detail?id=1162190#c16
[3]: https://github.com/freebsd/freebsd-src/tree/main/lib/libgcc_s
[4]: https://chromium-review.googlesource.com/c/chromiumos/overlays/chromiumos-overlay/+/2945947
[5]: https://refspecs.linuxbase.org/LSB_5.0.0/LSB-Core-generic/LSB-Core-generic/libgcc-s.html
Differential Revision: https://reviews.llvm.org/D108416
2021-08-17 20:23:22 +00:00
|
|
|
|
[llvm-libgcc][CMake] Refactor llvm-libgcc (#65455)
There are some issues in `llvm-libgcc` before this patch:
Commit c5a20b518203613497fa864867fc232648006068 ([llvm-libgcc] initial
commit)
uses `$<TARGET_OBJECTS:unwind_static>` to get libunwind objects, which
is empty.
The built library is actually a shared version of libclang_rt.builtins.
When configuring with `llvm/CMakeLists.txt`, target `llvm-libgcc`
requires a
corresponding target in `llvm-libgcc/CMakeLists.txt`.
Per target installation is not handled by `llvm-libgcc`, which is not
consistent
with `libunwind`.
This patch fixes those issues by:
Reusing target `unwind_shared` in `libunwind`, linking
`compiler-rt.builtins`
objects into it, and applying version script.
Adding target `llvm-libgcc`, creating symlinks, and utilizing cmake's
dependency
and component mechanism to ensure link targets will be built and
installed along
with symlinks.
Mimicking `libunwind` to handle per target installation.
It is quite hard to set necessary options without further modifying the
order of
runtime projects in `runtimes/CMakeLists.txt`. So though this patch
reveals the
possibility of co-existence of `llvm-libgcc` and
`compiler-rt`/`libunwind` in
`LLVM_ENABLE_RUNTIMES`, we still inhibit it to minimize influence on
other
projects, considering that `llvm-libgcc` is only intended for toolchain
vendors,
and not for casual use.
2023-09-19 13:56:03 +08:00
|
|
|
install(TARGETS unwind_shared unwind_static
|
|
|
|
LIBRARY DESTINATION ${LLVM_LIBGCC_INSTALL_LIBRARY_DIR} COMPONENT llvm-libgcc
|
|
|
|
ARCHIVE DESTINATION ${LLVM_LIBGCC_INSTALL_LIBRARY_DIR} COMPONENT llvm-libgcc
|
|
|
|
RUNTIME DESTINATION ${LLVM_LIBGCC_INSTALL_RUNTIME_DIR} COMPONENT llvm-libgcc)
|
[llvm-libgcc] initial commit
Note: the term "libgcc" refers to the all of `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`.
Enabling libunwind as a replacement for libgcc on Linux has proven to be
challenging since libgcc_s.so is a required dependency in the [Linux
standard base][5]. Some software is transitively dependent on libgcc
because glibc makes hardcoded calls to functions in libgcc_s. For example,
the function `__GI___backtrace` eventually makes its way to a [hardcoded
dlopen to libgcc_s' _Unwind_Backtrace][1]. Since libgcc_{eh.a,s.so} and
libunwind have the same ABI, but different implementations, the two
libraries end up [cross-talking, which ultimately results in a
segfault][2].
To solve this problem, libunwind needs to build a “libgcc”. That is, link
the necessary functions from compiler-rt and libunwind into an archive
and shared object that advertise themselves as `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`, so that glibc’s baked calls are diverted to the
correct objects in memory. Fortunately for us, compiler-rt and libunwind
use the same ABI as the libgcc family, so the problem is solvable at the
llvm-project configuration level: no program source needs to be edited.
Thus, the end result is for a user to configure their LLVM build with a
flag that indicates they want to archive compiler-rt/unwind as libgcc.
We achieve this by compiling libunwind with all the symbols necessary
for compiler-rt to emulate the libgcc family, and then generate symlinks
named for our "libgcc" that point to their corresponding libunwind
counterparts.
We alternatively considered patching glibc so that the source doesn't
directly refer to libgcc, but rather _defaults_ to libgcc, so that a
system preferring compiler-rt/libunwind can point to these libraries
at the config stage instead. Even if we modified the Linux standard
base, this alternative won't work because binaries that are built using
libgcc will still end up having crosstalk between the differing
implementations.
This problem has been solved in this manner for [FreeBSD][3], and this
CL has been tested against [Chrome OS][4].
[1]: https://github.com/bminor/glibc/blob/master/sysdeps/arm/backtrace.c#L68
[2]: https://bugs.chromium.org/p/chromium/issues/detail?id=1162190#c16
[3]: https://github.com/freebsd/freebsd-src/tree/main/lib/libgcc_s
[4]: https://chromium-review.googlesource.com/c/chromiumos/overlays/chromiumos-overlay/+/2945947
[5]: https://refspecs.linuxbase.org/LSB_5.0.0/LSB-Core-generic/LSB-Core-generic/libgcc-s.html
Differential Revision: https://reviews.llvm.org/D108416
2021-08-17 20:23:22 +00:00
|
|
|
|
[llvm-libgcc][CMake] Refactor llvm-libgcc (#65455)
There are some issues in `llvm-libgcc` before this patch:
Commit c5a20b518203613497fa864867fc232648006068 ([llvm-libgcc] initial
commit)
uses `$<TARGET_OBJECTS:unwind_static>` to get libunwind objects, which
is empty.
The built library is actually a shared version of libclang_rt.builtins.
When configuring with `llvm/CMakeLists.txt`, target `llvm-libgcc`
requires a
corresponding target in `llvm-libgcc/CMakeLists.txt`.
Per target installation is not handled by `llvm-libgcc`, which is not
consistent
with `libunwind`.
This patch fixes those issues by:
Reusing target `unwind_shared` in `libunwind`, linking
`compiler-rt.builtins`
objects into it, and applying version script.
Adding target `llvm-libgcc`, creating symlinks, and utilizing cmake's
dependency
and component mechanism to ensure link targets will be built and
installed along
with symlinks.
Mimicking `libunwind` to handle per target installation.
It is quite hard to set necessary options without further modifying the
order of
runtime projects in `runtimes/CMakeLists.txt`. So though this patch
reveals the
possibility of co-existence of `llvm-libgcc` and
`compiler-rt`/`libunwind` in
`LLVM_ENABLE_RUNTIMES`, we still inhibit it to minimize influence on
other
projects, considering that `llvm-libgcc` is only intended for toolchain
vendors,
and not for casual use.
2023-09-19 13:56:03 +08:00
|
|
|
install(TARGETS clang_rt.builtins-${COMPILER_RT_DEFAULT_TARGET_ARCH}
|
|
|
|
LIBRARY DESTINATION ${LLVM_LIBGCC_INSTALL_LIBRARY_DIR}/${install_dir_builtins} COMPONENT llvm-libgcc
|
|
|
|
ARCHIVE DESTINATION ${LLVM_LIBGCC_INSTALL_LIBRARY_DIR}/${install_dir_builtins} COMPONENT llvm-libgcc
|
|
|
|
RUNTIME DESTINATION ${LLVM_LIBGCC_INSTALL_RUNTIME_DIR}/${install_dir_builtins} COMPONENT llvm-libgcc)
|
[llvm-libgcc] initial commit
Note: the term "libgcc" refers to the all of `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`.
Enabling libunwind as a replacement for libgcc on Linux has proven to be
challenging since libgcc_s.so is a required dependency in the [Linux
standard base][5]. Some software is transitively dependent on libgcc
because glibc makes hardcoded calls to functions in libgcc_s. For example,
the function `__GI___backtrace` eventually makes its way to a [hardcoded
dlopen to libgcc_s' _Unwind_Backtrace][1]. Since libgcc_{eh.a,s.so} and
libunwind have the same ABI, but different implementations, the two
libraries end up [cross-talking, which ultimately results in a
segfault][2].
To solve this problem, libunwind needs to build a “libgcc”. That is, link
the necessary functions from compiler-rt and libunwind into an archive
and shared object that advertise themselves as `libgcc.a`, `libgcc_eh.a`,
and `libgcc_s.so`, so that glibc’s baked calls are diverted to the
correct objects in memory. Fortunately for us, compiler-rt and libunwind
use the same ABI as the libgcc family, so the problem is solvable at the
llvm-project configuration level: no program source needs to be edited.
Thus, the end result is for a user to configure their LLVM build with a
flag that indicates they want to archive compiler-rt/unwind as libgcc.
We achieve this by compiling libunwind with all the symbols necessary
for compiler-rt to emulate the libgcc family, and then generate symlinks
named for our "libgcc" that point to their corresponding libunwind
counterparts.
We alternatively considered patching glibc so that the source doesn't
directly refer to libgcc, but rather _defaults_ to libgcc, so that a
system preferring compiler-rt/libunwind can point to these libraries
at the config stage instead. Even if we modified the Linux standard
base, this alternative won't work because binaries that are built using
libgcc will still end up having crosstalk between the differing
implementations.
This problem has been solved in this manner for [FreeBSD][3], and this
CL has been tested against [Chrome OS][4].
[1]: https://github.com/bminor/glibc/blob/master/sysdeps/arm/backtrace.c#L68
[2]: https://bugs.chromium.org/p/chromium/issues/detail?id=1162190#c16
[3]: https://github.com/freebsd/freebsd-src/tree/main/lib/libgcc_s
[4]: https://chromium-review.googlesource.com/c/chromiumos/overlays/chromiumos-overlay/+/2945947
[5]: https://refspecs.linuxbase.org/LSB_5.0.0/LSB-Core-generic/LSB-Core-generic/libgcc-s.html
Differential Revision: https://reviews.llvm.org/D108416
2021-08-17 20:23:22 +00:00
|
|
|
|
[llvm-libgcc][CMake] Refactor llvm-libgcc (#65455)
There are some issues in `llvm-libgcc` before this patch:
Commit c5a20b518203613497fa864867fc232648006068 ([llvm-libgcc] initial
commit)
uses `$<TARGET_OBJECTS:unwind_static>` to get libunwind objects, which
is empty.
The built library is actually a shared version of libclang_rt.builtins.
When configuring with `llvm/CMakeLists.txt`, target `llvm-libgcc`
requires a
corresponding target in `llvm-libgcc/CMakeLists.txt`.
Per target installation is not handled by `llvm-libgcc`, which is not
consistent
with `libunwind`.
This patch fixes those issues by:
Reusing target `unwind_shared` in `libunwind`, linking
`compiler-rt.builtins`
objects into it, and applying version script.
Adding target `llvm-libgcc`, creating symlinks, and utilizing cmake's
dependency
and component mechanism to ensure link targets will be built and
installed along
with symlinks.
Mimicking `libunwind` to handle per target installation.
It is quite hard to set necessary options without further modifying the
order of
runtime projects in `runtimes/CMakeLists.txt`. So though this patch
reveals the
possibility of co-existence of `llvm-libgcc` and
`compiler-rt`/`libunwind` in
`LLVM_ENABLE_RUNTIMES`, we still inhibit it to minimize influence on
other
projects, considering that `llvm-libgcc` is only intended for toolchain
vendors,
and not for casual use.
2023-09-19 13:56:03 +08:00
|
|
|
foreach(VAR libgcc.a libgcc_eh.a libgcc_s.so.1.0 libgcc_s.so.1 libgcc_s.so)
|
|
|
|
install(FILES ${CMAKE_CURRENT_BINARY_DIR}/${VAR}
|
|
|
|
DESTINATION ${LLVM_LIBGCC_INSTALL_LIBRARY_DIR}
|
|
|
|
COMPONENT llvm-libgcc)
|
|
|
|
endforeach()
|