GitLab

The implicit include directory extraction added by commit 5990ecb7
(Compute implicit include directories from compiler output, 2018-12-07,
v3.14.0-rc1~108^2) leaves paths like `/usr/lib/../include` unchanged.
Fix the logic to canonicalize such paths (e.g. to `/usr/include`)
as we do for implicit link directories already.  This is important
to ensure the set of implicit directories is represented in the same
form as the include directories that will be compared to them.

Issue: #19095

Both commit 88ecfd8b (CPackIFW: Add some options, 2016-11-11,
v3.8.0-rc1~248^2) and commit e5089c56 (CPackIFW: Add some options,
2017-01-24, v3.8.0-rc1~53^2) added some options to this macro's
documentation and implementation, but the actual processing of the
parsed options was never added.  Add it now.
Reviewed-by: Konstantin Podsvirov <konstantin@podsvirov.pro>

Previously compilers that had full support for a language
standard level was forced to verify this every time a new build
directory was created. Now we record this information and insert
the correct granular compile features instead of doing a try_compile.

Our documented standard for find module variable names is to match the
case of the find module package name.  This was overlooked when the
module was added by commit 84e7920b (FindFontconfig: Add module to
find Fontconfig, 2018-09-27, v3.14.0-rc1~523^2).

The module was released with the upper case names in CMake 3.14.0, so
fix it to have camel case names in 3.14.1.  This is incompatible but
anyone using a given release series should be using the latest patch on
it and we've made breaking fixups on newly released features like this
before.
Reported-by: Christophe Giboudeaux <christophe@krop.fr>
Fixes: #19094

The variable is unused since commit ab358d6a (Improve find_dependency
argument handling, 2017-05-22, v3.9.0-rc1~57^2) where handling of EXACT
keyword in find_dependency was reworked.

The `PROMPT` environment variable affects nvcc's output.  Fix
our parsing of that output to tolerate square brackets.

Fixes: #19089

Any `include_directories()` calls in toolchain files are used during our
ABI detection step even though it does not include any system headers.
Since commit 5990ecb7 (Compute implicit include directories from
compiler output, 2018-12-07, v3.14.0-rc1~108^2), that check is also used
to detect implicit include directories.  Any `include_directories()` in
a toolchain file are detected as implicit and later excluded from
explicit specification on compiler command lines, thus breaking the
purpose of the calls in the first place.

Fix the implicit include directory detection step to avoid using paths
from `include_directories()` calls in the toolchain file.

Fixes: #19079

Fixes: #15805

Previously Compilers always had to run a try_compile to determine what
language level each feature mapped to. Now we can skip the try_compile
when a language level has no features.

According to ARMCC 5.06 documentation:

    http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0472m/chr1359125007083.html

the compiler may define `__GNUC__` in addition to `__ARMCC_VERSION`.
Re-order our preprocessor checks to consider the ARM-specific macro
first so that the ARM compiler is not mistaken for a GNU compiler.

Fixes: #19065

These are added through CMAKE_SYSTEM_PREFIX_PATH on the platforms automatically.

Resolve issue building with missing variable declarations, error:

```
/src/cmake_clang/CMakeFiles/CheckTypeSize/CMAKE_SIZEOF_UNSIGNED_LONG.c:24:6: error: no previous extern declaration for non-static variable 'info_size' [-Werror,-Wmissing-variable-declarations]
char info_size[] =  {'I', 'N', 'F', 'O', ':', 's','i','z','e','[',
     ^
1 error generated.
```

Fixes: #19052

Replicate behaviour of upstream GLEW.

* define the imported target GLEW::glew for the shared library GLEW

* if GLEW_USE_STATIC_LIBS is defined and set to TRUE, this module instead
  defines the imported target GLEW::glew_s for the static library GLEW

* an imported target GLEW::GLEW is created as a copy of either GLEW::glew
  or GLEW::glew_s

* by setting GLEW_VERBOSE you print out the log of the module.

Fixes: #17638, #17864, #18989

When GIT_SHALLOW is used, the '--depth 1 --no-single-branch'
arguments are add. It means that only branch names and tags
is downloaded to repository. Most Commit Hash is not working.
With this commit the documentation was updated, to describe
the limitation of GIT_SHALLOW.

Ensure interpreter and libraries architecture matches CMake build
configuration.
Update documentation about interpreter constraints.

Fixes: #19024

Since commit 79410eeb (ExternalProject: Initialize Git submodules
recursively and on update (#16083), 2016-04-26, v3.6.0-rc1~105^2) our
`git submodule update` step uses the `--init` flag.  This makes the
prior `git submodule init` unnecessary.

Revise script generation to hard-code selected options instead of
building them with logic inside the script.

The purpose of the `Boost::dynamic_linking` interface library is to
cause `BOOST_ALL_DYN_LINK` to be defined.  Do this on all platforms
instead of just Windows.

In particular, using Boost::log trivial_logger requires to set
BOOST_ALL_NO_LIB when Boost::Log is compiled as dynamic library.

Fixes: #17813

Our check for just `pthread_create` is not sufficient because there are
cases where that symbol may be found in libc but not all of pthreads.
We first tried to address this in commit e9a1ddc5 (FindThreads:
Replace the pthread symbol checking in libc., 2018-11-18,
v3.14.0-rc1~292^2) by switching to `pthread_kill`, but that had to be
reverted by commit 18320230 (FindThreads: Revert libc symbol check to
pthread_create, 2019-02-27, v3.14.0-rc3~6^2) because there are other
cases where `pthread_kill` is in libc but the rest of pthreads is not.
Update our check to try a complete program using pthreads as an actual
application might.

Fixes: #18994

Since commit 11da882a (Apple: Introduce separate system name for iOS,
tvOS, and watchOS, 2018-01-15, v3.14.0-rc1~14^2~1) we support setting
`CMAKE_SYSTEM_NAME` to `iOS`.  Existing iOS toolchain files already
set `IOS` as a short-hand variable, so do the same here.

Update the component added by commit 513e7755 (FindPython: Introduce
NumPy component, 2018-12-12, v3.14.0-rc1~95^2).  The `numpy/`
sub-directory should not be part of the include directory.  It should be
part of the `#include` line.

Java 9 restructured the standard location of the AWT libraries due to
the removal of the JRE/JDK separation. We should check all possible
combinations of subdirectories to the Java root directories to ensure
that the libraries will be found after an upgrade.

Furthermore, a root directory would contain both, include and library
paths, so the search should be unified to ease maintenance on the
module.

llvm-rc can't handle definitions given with /D and without a space.

In commit b6f6cac3 (ExternalProject: add LOG_DIR option that allows
overriding of log location, 2018-10-12, v3.14.0-rc1~515^2~1) the log
directory got its own option.  The intention was to fall back to the
stamp directory by default.  However, the implementation actually only
falls back to the same default as the stamp directory and does not
consider a custom stamp dir.

Update the default log dir computation to fall back to whatever is the
final selection for the stamp dir.

Fixes: #19000

Code removed for MIPSpro by commit 214fcefa (Remove now-unused code
once used for MIPSpro on IRIX, 2019-02-21) actually changed a
public-facing API by dropping the `<prefix>_COMPILER_IS_MIPSpro`
definition from the generated compiler detection header.  Restore the
definition hard-coded to `0` since the compiler will never be MIPSpro.
Reported-by: Hans Johnson <hans-johnson@uiowa.edu>

Prior to LLVM 8.0, `llvm-rc` does not recognize `/fo` without a space
after it.  Add the space unconditionally because MS `rc` accepts it too.

Issue: #18957

Since commit e9a1ddc5 (FindThreads: Replace the pthread symbol
checking in libc., 2018-11-18, v3.14.0-rc1~292^2) we check libc for
`pthread_kill` instead of `pthread_create`.  However, on FreeBSD
`pthread_kill` is in libc but not `pthread_create`.  Discussion in the
original merge request for the above commit also considered
`pthread_key_create`, `pthread_self`, and `pthread_attr_init`.  Every
symbol seems to have some reason it is not an appropriate choice.

Revert to the pre-3.14 behavior of using `pthread_create` pending
further investigation.