The path to the 32 bit libraries in the Intel windows/redist folder use
ia32. I don't remember if this has changed at some point, but ia32 has
been used at least since Intel Fortran XE 2018.
The version is determined in two steps. First, the "spec date" is
detected and cached. Second, the date is converted to a version.
Move the second step out of the spec date cache guard condition
so that it runs every time even if the spec date is already cached.
Fixes: #19150
Since commit dd378258f1 (FindJava: Do not accept OS X stub 'java' as
Java, 2014-10-24, v3.1.0-rc3~29^2) we try to avoid using the macOS
`/usr/bin/java` stub if no underlying implementation of Java is actually
installed. However, the message that `/usr/bin/java` prints when there
is no Java available has changed since then. Update our check to also
look for the new message.
While at it, revise the way we suppress `Java_JAVA_EXECUTABLE`.
Previously we set its cache entry to `Java_JAVA_EXECUTABLE-NOTFOUND`,
but that would cause the same find-and-reject sequence to be followed
every time CMake runs in a build tree. Instead, use the approach from
commit 2c0db404d1 (FindSubversion: Do not accept macOS stub without
Xcode implementation, 2020-05-28, v3.18.0-rc1~67^2). Leave the cache
entry alone and just set a normal variable of the same name to hide it.
The Google test framework allows to write the results into an XML file
since commit e9ab39eb1d (GoogleTest: Add XML_OUTPUT_DIR parameter,
2020-03-06, v3.18.0-rc1~538^2~2). This file is passed on the command
line: `--gtest_output=xml:FILE_NAME`.
The module allows to specify a directory to save those files with
**TEST_XML_OUTPUT_PARAM**.
If the option is set, the filename will be set to
`${prefix}${pretty_suite}.${pretty_test}${suffix}.xml`.
The pretty names contain parameters for the tests, if
value-parameterized tests are used. These parameters may not be safe to
use in file names.
There are two possible options:
1. sanitize the file name
2. omit the values and use the internal numbering of gtest
This commit chose option 2. The testname needs to be a valid C++
identifier and should therefore be reasonable for a filename.
Note that the generated names contain slashes. This will lead to
subdirectories, but works on both Linux and Windows.
Fixes: #20877
When using a Visual Studio generator, we do not know the path to
`CMAKE_CUDA_COMPILER` until after the compiler id detection project
extracts it. However, based on the VS toolchain configuration we know
that the compiler id will be NVIDIA, so use that to drive the CUDA
architecture flag logic. After the main compiler id step we know
`CMAKE_CUDA_COMPILER`, so `CMAKE_CUDA_COMPILER_TOOLKIT_ROOT` can then be
computed.
The result variable was `CUDAToolkit_NVCC_EXECUTABLE` instead of
`_CUDA_NVCC_EXECUTABLE`, which the rest of the code used. Also, we can
simply `set()` the `_CUDA_NVCC_EXECUTABLE` value without searching.
Fixes: #20847
947cfc732d Help: Explicitly say transaction, don't abbreviate to trans
90c73479bc Help: Correct and condense examples for VS_SOLUTION_DEPLOY
0420de10b5 Help: Add missing cross-referencing for generator expressions
013f7a2647 Help: Formatting, crossref for FRAMEWORK_MULTI_CONFIG_POSTFIX_<CONFIG>
a12d53acf3 Help: Add missing PCH_WARN_INVALID docs and related cleanups
95a16f7805 Help: Fix formatting error and consistency for gtest_discover_tests()
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !4915
The ability to disable adding architectures completely for packaging purposes
and cases requiring passing the architectures flags explicitly has been
requested.
Support a false value for CUDA_ARCHITECTURES and CMAKE_CUDA_ARCHITECTURES
for this purpose.
Implements #20821.
Clang isn't very good at finding the installed CUDA toolkit.
The upstream recommendation is that we should pass the toolkit explicitly.
Additionally:
* Avoids Clang having to search for the toolkit on every invocation.
* Allows the user to use a toolkit from a non-standard location by simply
setting CUDAToolkit_ROOT. The same way as with FindCUDAToolkit.
Clang wants the directory containing the device library and version.txt as the
toolkit path.
We thus pass the newly introduced CUDAToolkit_LIBRARY_ROOT as the toolkit path.
We save CUDAToolkit_ROOT_DIR and CUDAToolkit_LIBRARY_ROOT on Clang to have them
available in try_compile() and avoid unnecessary re-searching or a possibly
different installation being found in FindCUDAToolkit.
This however means that the selected toolkit can't be changed after the initial
language enablement.
We now determine CUDA compiler ID before doing actual detection, as we don't
want to spend time finding the CUDA toolkit for NVIDIA.
Implements #20754.
Brad King