Instead of relying on SDKSettings.plist which does not contain the
certain architectures, deduce the supported architectures by
inspecting libSystem.tbd and libSystem.dylib.
.tbd files are text files, so just parse out the archs string.
.dylib files can be fat or non-fat, so use lipo -info to extract
the architectures and parse lipo output.
Fixes: #20588
Use `<arch>-SDK-NOTFOUND` instead of an empty string as a placeholder in
`CMAKE_APPLE_ARCH_SYSROOTS` for architectures whose SDK is not found.
This ensures the length of `CMAKE_APPLE_ARCH_SYSROOTS` matches the
length of `CMAKE_OSX_ARCHITECTURES`. It also makes the missing SDKs
more visible in the value.
Issue: #20534
Since commit 1c2d031cbd (Add -E cmake_llvm_rc to preprocess files for
llvm-rc, 2020-01-14, v3.17.0-rc1~24^2) with llvm-rc we explicitly
preprocess RC source files and then compile separately without -I flags.
This broke cases where the RC source references data files adjacent to
itself or in the include path.
This change adds the expansion of the include paths when calling the
llvm-rc in order for the resource files to be picked up correctly by
llvm-rc. Since the RC compiled file is first preprocessed, the file
being compiled by llvm-rc resides in the build directory. In order for
llvm-rc to find the resource data specified relative to the .rc file
being compiled, the source file path is preppended in the include list
so that the original source path takes priority over all the other
includes paths specified.
A space was added in the CMAKE_INCLUDE_FLAG_RC to make the include
directive work properly for llvm-rc. Checks on the rc.exe showed that
the syntax change doesn't affect it's proper operation.
Fixes: #20529
Move the depfile flags to `CMAKE_DEPFILE_FLAGS_RC` so that they
are only usedwith generators that use depfiles. Also switch to
using the `<DEPFILE>` placeholder for the location of the depfile
so that it goes where the generator expects.
Fixes: #20493
The llvm-rc preprocessor is selected according to the
CMAKE_C_COMPILER_ID / CMAKE_CXX_COMPILER_ID which are only defined when
the C or CXX language is active.
Since commit 1c2d031cbd (Add -E cmake_llvm_rc to preprocess files for
llvm-rc, 2020-01-14, v3.17.0-rc1~24^2) we pass the full target `<FLAGS>`
to the llvm-rc resource compiler, but we should pass only `<DEFINES>`.
Fixes: #20414
Add an additional include flag to PCH usage command line to fix programs
that rely on `compile_commands.json` file. Pass it to the preprocessor
directly to avoid compiler driver to change it to '-include-pch'.
When preprocessor is requested to preprocess a file, it tries to get
the original filename from '.pch' and uses that file for preprocessing.
CMake generates a '.pch' file from the '.hxx' file by passing an empty
'.cxx' source file to the compiler as a compilation unit and the header
file with the '-include' flag. After that, compiler puts compilation
unit filename in the '.pch' as the original filename.
However, CMake build system uses empty file as the source file and
passes the header file using '-include-pch' flag. As a result, Clang
uses the wrong file for preprocessing and produces the corrupted
preprocessed file.
Fixes: #20355
Signed-off-by: Sergey Larin <cerg2010cerg2010@mail.ru>
Populate `CMAKE_Swift_IMPLICIT_INCLUDE_DIRECTORIES` with the macOS SDK's
include directory so that we filter such implicit directories out of
Swift targets.
Fixes: #19845
llvm-rc requires preprocessed rc files. The CMake command line tool
cmake_llvm_rc enables channing the preprocessor call and the resource
compiler and make this appear as single compilation step.
When llvm-rc is detected as resource compiler, the RC compilation step
is set to use this command.
In commit 97bca2f9fa (Android: Use unified toolchain in NDK r19+,
2019-07-26, v3.16.0-rc1~342^2) we hard-coded use of the unified
toolchain for NDK r19+ and skipped most of the old detection logic.
However, in that fast path we left out setting `_CMAKE_TOOLCHAIN_PREFIX`
for `CMakeFindBinutils` to select the matching binutils. Add it.
Fixes: #20038
This allows building .s/.S assembly code, and makes sure that the
`-fPIC` option isn't passed to the compiler (as it errors out on
Clang and causes a loud warning on GCC).
Since commit 0f150b69d3 (AIX: Explicitly compute shared object exports
for both XL and GNU, 2019-07-11, v3.16.0-rc1~418^2~2) we always export
all symbols from shared libraries by default. Add a new target property
called `AIX_EXPORT_ALL_SYMBOLS` that can be explicitly set to OFF to
suppress this behavior and export no symbols by default.
Fixes: #20290
In commit 0f150b69d3 (AIX: Explicitly compute shared object exports for
both XL and GNU, 2019-07-11, v3.16.0-rc1~418^2~2) we dropped use of the
old `CMAKE_XL_CreateExportList` cache entry for XL exports. However,
some people were setting the value to an empty string as a way to
disable automatic export of symbols. Restore this behavior when the
option is explicitly set to an empty string.
Issue: #20290
Fixes#17559
Replace our hard-coded default of cudart=static with a first-class abstraction to select the runtime library from an enumeration of logical names.
Modern distributions of packages built with MinGW tools provide `.dll.a`
import libraries. Prefer those instead of finding plain `.dll` files.
This avoids accidentally finding unrelated Windows `.dll` files.
Fixes: #20019
Add a `JOB_POOL_PRECOMPILE_HEADER` target property to specify the pool
name, and its associated `CMAKE_JOB_POOL_PRECOMPILE_HEADER` variable.
Fixes: #20217
fcde42751a FindPython: ensure new Xcode framework for Python3 is detected
dd7b741b81 macOS: Add support for new Xcode 11 frameworks directory
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !4198
fcde42751a FindPython: ensure new Xcode framework for Python3 is detected
dd7b741b81 macOS: Add support for new Xcode 11 frameworks directory
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !4198
Extend the logic from commit abe8a623d9 (GNUtoMS: Add search path for VS
2017 environment scripts, 2017-05-19, v3.8.2~1^2) to consider VS 2019
paths too.
Fixes: #20162
This moves the definition of the import library flags for Windows into a
platform specific file (much like the Darwin `-sdk` flag). This is in
preparation for re-use of the flag to make the flags handling for
executables more precise.
a9b41195d2 Handle multi-arch sysroots on Apple platforms
372c89ea38 Refactor -arch and -isysroot code usage
7a44e16d7f Refactor SDK name to path into a function
777d364913 Use xcrun instead of xcodebuild for resolving SDK path
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !4125
Starting with CUDA 10.2 the nvcc compiler has gained support
to automatically forward unknown flags to the host compiler.
This behavior is highly desired as projcts that mix CUDA, C, C++
run into situation where flags such as `-pthread` which aren't
supported by nvcc, are being applied to all source files and
therefore break CUDA compilation.
2d0100fac7 replace remove and remove_directory with rm in tests
5239fc5c75 cmake -E: Add rm with improved semantics over remove and remove_directory
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !3949
7b2dd9dedc Refactor: Use added message types in various modules
949a1e120a message: New message types to mark checks performed by CMake
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !3689
Pass the value to the Swift compiler driver via `-sdk`. We already do
this for C/C++ via `-isysroot`.
This fixes command-line builds on macOS 10.15 with Xcode 11 Swift tools.
Fixes: #19880
The CMake support for Android assumes cross-compilation using the NDK,
so stub out that Android NDK support and use the Linux support that's
already invoked. Set CMAKE_HOST_SYSTEM_NAME to "Android", rather than
"Linux".
Issue: #19840
