| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|\
| |
| |
| | |
Merge-request: !2631
|
| |
| |
| |
| |
| |
| | |
Refactoring in commit v3.12.0-rc1~92^2 (FindPkgConfig: export the list
of found libraries also as variable, 2018-05-11) dropped use of FPHSA
to set `BLAS_FOUND`. Set it explicitly instead.
|
|\ \
| | |
| | |
| | | |
Merge-request: !2624
|
| |/ |
|
|\ \
| | |
| | |
| | | |
Merge-request: !2637
|
| |/
| |
| |
| |
| |
| |
| | |
Windows SDK version 10.0.17763.0 now places the uCRT libraries in a
versioned directory.
Fixes: #18603
|
|\ \
| | |
| | |
| | | |
Merge-request: !2635
|
| |/
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Represent the input file path internally in canonical form. Otherwise
multiple `configure_file` calls that share the same input file but specify
it relative to different directories (e.g. via `../`) result in multiple
copies of the dependency on the rule to re-run CMake. This causes the
Ninja generator to emit duplicate phony build statements for these
dependencies, which generates an error with `-w dupbuild=err`, which
will be default in Ninja 1.9.
Also canonicalize the output path for consistency.
Add a test case.
Fixes: #18584
|
|\ \
| | |
| | |
| | | |
Merge-request: !2618
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Since commit v3.11.0-rc1~117^2 (CTest: Re-implement test process
handling using libuv, 2017-12-10) we do not give the child test
processes any stdin. Prior to that change we let the child test
processes inherit stdin from ctest itself. Tests that run serially
might be able to use the real stdin meaningfully, so restore that
behavior and add a test case.
Fixes: #18591
|
| |/ |
|
|\ \
| | |
| | |
| | | |
Merge-request: !2601
|
| |/ |
|
|\ \
| | |
| | |
| | | |
Merge-request: !2579
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Add recognition of compiler version and prefix for clang.
Accommodate changes to version numbering scheme for GCC 5+ and clang 4+
- Minor number becomes patch, so excluded it from compiler prefix.
- Improves searching for Boost 1.69+ libraries built with GCC 5+ and
clang 4+, where library names are generated with compiler tag based on
major version only eg. -gcc5 for GC 5.5.0 or -clang6 for clang 6.0.0
- Follows up related changes in upcoming Boost 1.69
https://github.com/boostorg/build/pull/349
Dump detected compiler version in Boost_DEBUG message.
Replace use of CMAKE_COMPILER_IS_GNUCXX with CMAKE_CXX_COMPILER_ID.
|
|\ \ \
| |/ /
|/| |
| | | |
Merge-request: !2584
|
| |/
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Add missing vendor table entries for Clang and AppleClang to identify
them. Previously this worked only if Clang was also enabled as a C or
CXX compiler first because we used to copy the C compiler id. However,
that was removed by commit v3.13.0-rc1~44^2 (ASM: Search for full path
even when using C or C++ compiler, 2018-09-26).
Fixes: #18575
|
|\ \
| |/
|/|
| | |
Merge-request: !2570
|
|/
|
|
|
|
|
|
|
| |
Add the system thread library to Boost_LIBRARIES when the boost thread
component has been found.
The Boost::thread imported target already pulls in Threads::Threads.
This changes does the same for projects using the Boost_LIBRARIES variable
instead.
|
| |
|
|\
| |
| |
| | |
Merge-request: !2574
|
| | |
|
| |
| |
| |
| | |
With -Wstrict-prototype -Werror, the check would fail
|
|\ \
| | |
| | |
| | | |
Merge-request: !2563
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
CPack generator names were not used consistently
throughout the documentation, resulting in ambiguity
about what the correct name was for use with the
`cpack -G` option. With the changes in this commit, the
cpack-generators(7) page of the manual now shows the
correct names and other help pages no longer use
inconsistent or incorrect names.
|
|\ \ \
| | | |
| | | |
| | | | |
Merge-request: !2568
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Boost 1.66 and above built with `--layout=versioned` add an architecture
tag to the library file names. We already try to compute this tag
automatically when `CMAKE_CXX_COMPILER_ARCHITECTURE_ID` is available,
but that is currently not computed everywhere. Add an explicit
`Boost_ARCHITECTURE` option that a user can set to specify the
architecture tag.
Issue: #17701
|
|\ \ \ \
| | |/ /
| |/| |
| | | | |
Merge-request: !2566
|
| | |/
| |/|
| | |
| | |
| | | |
Ext and External were used inconsistently in the code and the
docs. This change converts all uses of Ext to External, including
within variable names used by the generator.
|
|\ \ \
| | | |
| | | |
| | | | |
Merge-request: !2560
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
The symbol "CHEEV", originally used to determine if a library provides
Fortran 95 wrappers for LAPACK, has been replaced by "cheev_f95". "CHEEV"
is provided by libmkl_intel_(i)lp64, which does not provide the generic
Fortran 95 wrappers. Instead, libmkl_lapack95_(i)lp64 does; one of the
specializations of the type-generic interfaces contained in that library
is "lapack_f95".
Also, FindLAPACK used libmkl_intel_(i)lp64 instead of the correct
libmkl_lapack95_(i)lp64 library for LAPACK95 functionality. This has
been fixed, too.
|
| |/ /
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The symbol "SGEMM", originally used to determine if a library provides
Fortran 95 wrappers for BLAS, has been replaced by "sgemm_f95". "SGEMM"
is provided by libmkl_intel_(i)lp64, which does not provide the generic
Fortran 95 wrappers. Instead, libmkl_blas95_(i)lp does; one of the
specializations of the type-generic interfaces contained in that library
is "sgemm_f95".
|
|\ \ \
| | | |
| | | |
| | | | |
Merge-request: !2559
|
| |/ /
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Since commit v3.13.0-rc1~39^2 (add_custom_{command,target}:
WORKING_DIRECTORY generator expressions, 2018-09-22) the
`WORKING_DIRECTORY` option accepts generator expressions.
Fix support for the case of a leading generator expression
by deferring conversion to an absolute path until after
evaluation of the generator expression.
Fixes: #18543
|
|\ \ \
| | | |
| | | |
| | | | |
Merge-request: !2550
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Backport upstream curl commit 2c5ec339ea (Curl_follow: accept
non-supported schemes for "fake" redirects, 2018-11-01) to get
a fix to curl issue 3210, a regression in 7.62.0.
|
| | | |
| | | |
| | | |
| | | |
| | | | |
Drop unused code. Report the error message on failure.
Format the source file using clang-format.
|
| | | | |
|
| |\ \ \
| | | | |
| | | | |
| | | | |
| | | | | |
* upstream-curl:
curl 2018-10-30 (19667715)
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Code extracted from:
https://github.com/curl/curl.git
at commit 196677150f711a96c38ed123e621f1d4e995b2e5 (curl-7_62_0).
|
| | |/ /
| |/| | |
|
|\ \ \ \
| | | | |
| | | | |
| | | | | |
Merge-request: !2562
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Minor rewording around the updated link and extra blank line
added to improve readability and remove a potential
ambiguity around the same area.
Fixes: #18545
|
|\ \ \ \ \
| | | | | |
| | | | | |
| | | | | | |
Merge-request: !2556
|
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Fixes: #18532
|
|\ \ \ \ \ \
| | | | | | |
| | | | | | |
| | | | | | | |
Merge-request: !2551
|
| | |_|/ / /
| |/| | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Protobuf headers have dependencies on threads. On UNIX platforms this
requires linking to a threads library. We've long done this in the
`Protobuf_LIBRARIES` result variable. However, the imported targets
added by commit v3.9.0-rc1~81^2~2 (FindProtobuf: add targets,
2017-05-17) and commit v3.9.0-rc1~68^2 (FindProtobuf: Rename imported
targets to match upstream names, 2017-05-22) were missing a dependency
on threads.
Add the dependency to the imported targets, and to the variables
`Protobuf_LITE_LIBRARIES` and `Protobuf_PROTOC_LIBRARIES`. While this
did not seem to matter in practice for a long time, protobuf 3.6 throws
exceptions in some cases when threads are missing.
Fixes: #18533
|
|\ \ \ \ \ \
| | | | | | |
| | | | | | |
| | | | | | | |
Merge-request: !2549
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
A target created by `add_custom_target` should always be a `.vcxproj`
file even if it has `.cs` sources involved in custom commands and such.
The latter case was broken by refactoring in commit v3.12.0-rc1~160^2~7
(remove TargetIsCSharpOnly() and use methods from cmGeneratorTarget,
2018-03-19). The reason is that the `HasLanguage` method added by
commit v3.12.0-rc1~239^2~6 (cmGeneratorTarget: add HasLanguage() as
wrapper for GetLanguages(), 2018-03-19) does not check the target type
and so is not a suitable check for deciding the project file extension.
The `HasLanguage` method was an attempt at an abstraction that turns
out not to work very well. Replace it with a dedicated `IsCSharpOnly`
method that considers the target type, sources, and non-transitive
`LINKER_LANGUAGE`.
Fixes: #18515
|