If FindMPI can't interrogate any of the available compilers, it attempts to compile simple MPI
programs with CMAKE_${lang}_COMPILER. If this works, it uses that as MPI_${lang}_COMPILER.
This allows MPI to be discovered on Cray XT/XE systems, where modules are used and cc, CC, and ftn
compilers *are* MPI compilers.
Without this, regular swig (not 2.0) executable is not found
automatically.
Commit 55b7c87e (Add support for new swig 2.0 application, 2011-03-28)
added support for finding swig 2.0 at the (accidental) expense of not
finding 1.x. Fix this regression.
Patch from Aaron C. Meadows
Adds support for detecting VS 2010 SP1
Also improves performance using CMAKE_CXX_COMPILER if it is available
and for robustness falls back using try_compile() and try_run()
1ed19bc multiarch: Set CMAKE_LIBRARY_ARCHITECTURE_REGEX for Linux|Hurd|kFreeBSD
52a6ed2 Test find_package multiarch support (#12037)
b41ad3b Teach find_(library|package) about Linux multiarch (#12037)
Fix typo introduced by commit e28c16b4 (Split GNU compiler information
files, 2009-12-02).
Reported-by: Campbell Barton <ideasman42@gmail.com>
Suggested-by: Michael Hertling <mhertling@online.de>
* Fix linux CMAKE_LIBRARY_ARCHITECTURE_REGEX to support armel-linux-gnueabi.
* Add CMAKE_LIBRARY_ARCHITECTURE_REGEX on kFreeBSD.
* Add CMAKE_LIBRARY_ARCHITECTURE_REGEX on GNU (Hurd).
Also regex is improved to support quadlets. Even if I have not seen this
in the wild yet, reportedly they are possible.
Workaround an issue where users wanted to use Boost_LIBRARIES with the
install() command and debug/optimized keywords were interfering. Now
debug/optimized keywords are removed if the release & debug library
are the same.
Implement support for multiarch as specified here:
http://wiki.debian.org/Multiarchhttps://wiki.ubuntu.com/MultiarchSpec
Detect the <arch> part of <prefix>/lib/<arch> from the implicit library
search path from each compiler to set CMAKE_<lang>_LIBRARY_ARCHITECTURE.
Define CMAKE_LIBRARY_ARCHITECTURE using one of these values (they should
all be the same). Teach the find_library and find_package commands to
search <prefix>/lib/<arch> whenever they would search <prefix>/lib.
51e16c0 BundleUtilities: Avoid test on Watcom dashboards (#12034)
41f962a Revert "BundleUtilities: Run test on Windows if either MSVC or dumpbin was found."
e17135e BundleUtilities: Add rpath to loadable modules in test.
8064044 BundleUtilities: Print reason for not loading module.so
f3de459 BundleUtilities: Run test on Windows if either MSVC or dumpbin was found.
900bf98 BundleUtilities: Disable running test on Windows unless using MSVC.
fa4dc08 BundleUtilities: Fix issues with custom target DEPENDS in test (#12034)
e40b79e BundleUtilities: Fix test when using xcode (#12034)
b68d3dc BundleUtilities: Fix regex to extract dependents from ldd (#12034)
7ac7b43 BundleUtilities: Work w/ non .app exes on Mac (#12034)
When the path to "resolved_embedded_item" was shorter than
the path to the bundle being fixed up, fixup_bundle would
fail with a cmake error like:
"string end index: 110 is out of range 0 - 85"
Detect when the path of resolved_embedded_item is too short
to be embedded in the bundle, and report the proper error
message, so the poor developer reading it has a snowball's
chance of actually fixing the issue.
517837f Fix , to - in Copyright message so it passes CMake.ModuleNotices test
162f3fb Merge branch 'master' of git://public.kitware.com/cmake into 12128_FindProtobuf_module_behavior_under_Windows_is_annoying
ca000a0 FindProtobuf: Better MSVC support, Searching for protobuf lite
e38216c Remove unnecessary mark_as_advanced from FindHDF5.
93ba19e FindHDF5 ensures good link lines when libraries are duplicated.
0584701 Fix for bug 11752, mixed debug and release libraries.
4e12284 Use HDF5_FOUND to control autoconf and CMake built FindHDF5.
2ba826e Use CMAKE_CURRENT_LIST_DIR to locate FindPackageHandleStandardArgs.
3978f32 Add logic for CMake built HDF5 install.
063ac87 HDF5 high level library is a find COMPONENT now.
34ea179 Modified the FindHDF5.cmake file to locate the Fortran bindings.
Also add a test of BundleUtilities including an exe,
some shared libs, a plugin, and a framework-style lib.
This test presently runs (and this functionality works)
on Linux, Mac and Windows.
For now, the framework-style lib is built as a plain old
shared lib because there is another yet-unresolved issue
with local frameworks without rpaths on the Mac.
Add support for finding debug libraries
Add support for searching Google provided MSVC project dir structure for libs
Add support for finding Protobuf "Lite" libraries
cdc2b41 Fix CompileCommandOutput test build on Windows
7039d1f Fix CompileCommandOutput test for Make tools not supporting spaces
4268e3d run_compile_commands: Cast istream::get() result to char
c45c60b run_compile_commands: Avoid extra stl vector conversion
7c5be51 run_compile_commands: Avoid shadow in std::map<>::at workaround
169bb05 Provide std::map<>::at for use in run_compile_commands
4e2185c Make std::map usage more portable in language=>flags/defines maps
a7e7a04 Fix run_compile_commands build on Apple GCC 3.3
c9174c0 Fix signed/unsigned comparison in EscapeJSON
8346a28 Only offer the compile command output feature on unix systems
0e6b05f Adds a test for the compile command line output.
5674844 make compile command output optional
fe07b05 implement cxx command output
65c0c24 cache flags and defines
3f064ef refactor flags and defines
8bd3e51 Absoft: Enable FortranCInterface check in Fortran test
d7b376b Absoft: Detect implicit link libraries on Linux and Mac
ac5b999 Add Absoft Fortran compiler id and basic flags
Use the "-X -v" flag to the Absoft front-end to pass "-v" to the gcc it
invokes under the hood. Teach CMakeParseImplicitLinkInfo to exclude
linker version lines from consideration as link lines. Fix parsing of
Sun's linker search path option "-Y..." to avoid conflict with the Mac
linker option "-Y<num>".
The find module calls find_package( HDF5 QUIET NO_MODULE ). If that succeeds,
the find module variables are set based on the imported target locations.
The HDF5 high level library was originally hard coded as a dependency when the C
bindings are desired. The high level API is now requested as a COMPONENT.
a55da06 Added backward compatibility for input as well as output vars.
a32a633 FindMPI: Fix documentation formatting
706b73e FindMPI: Handle multiple languages
Users can now supply MPI_COMPILER, MPI_INCLUDE_PATH, MPI_LIBRARY,
and others as with the old FindMPI. These are mapped to their
respective equivalents for C and CXX. Fortran is not touched, as
there was no Fortran support in the old FindMPI.
The custom error message that mentions PREFIX/lib/InsightToolkit is not
accurate for ITKv4. Just use the more generic message that find_package
generates by default. This module is now almost a no-op but exists to
tell find_package to look for the InsightToolkit name as well as ITK.
The previous commit added --non-interactive as a "side effect"
of turning on SVN_TRUST_CERT. While reviewing that commit, we
decided all ExternalProject svn usage should be non-interactive.
That way, if there's any sort of problem, svn will return an
error right away rather than hang forever waiting for input...
Which adds --non-interactive and --trust-server-cert to the svn
checkout and update command lines. This allows ExternalProject
clients to pull from an https:// based svn server even though
the server may not have a valid or trusted certificate.
Caveat emptor: I would NOT recommend using this except as a
short-term work-around. Rather, the server should have a valid,
trusted certificate, or the client should be using "http" instead
of "https".
This commit fixes BUG: 0011782. UseSWIG would be using the same variable
to declare module information. The problem would only be noticed in parallel builds
Fix this variable declaration by properly resetting it.
This commit fixes BUG: 0011215 by properly expanding $(OutDir)
Instead of creating the output directory using file(MAKE_DIRECTORY)
we use cmake -E to create the directory at execution time
This commit fixes BUG: 0004147 it directly uses swig executable
to compute a list of dependencies directly from the .i files
to make sure to rebuild the swig module any of its direct dep.
is touched
The XL toolchain supports shared object files stored in archives. Since
CMake lists libraries on link lines by full path it is common for a
shared library link line to contain the path to an archive file.
When linking a shared library the compiler front-end by default runs
CreateExportList to construct the list of symbols to be exported.
Unfortunately it passes all files found on the command line to the tool
so archive and library files get processed along with the object files.
The tool returns a list of all symbols in all objects, archives, and
libraries on the command line. This causes the linker to copy every
object file out of every archive into the shared library whether they
are dependencies of the original object files or not.
Work around this problem by running CreateExportList ourselves with just
the original object files intended for inclusion in the shared library.
Then pass the list it produces on the link line to prevent the compiler
front-end from constructing its own. This tells the linker to export
only the symbols provided by the original source files of the shared
library.
Factor duplicate information out of Compiler/XL-<lang>.cmake modules
into a macro in a new Compiler/XL.cmake module. Invoke it from the
per-language files to produce the original settings.
Since commit e1729238 (Add initial XL C compiler flags for safer builds,
2009-09-30) CMake sets the initial XL C flags to include "-qthreaded"
and "-qhalt=e". Do the same for C++ and Fortran with this toolchain.
112f1dd FindOpenSSL: Use find_package_handle_standard_args for version check.
a091ba6 FindOpenSSL: Fixed crypto und ssl variable names.
a164649 FindOpenSSL: We should only use hints to find OpenSSL.
0fb5142 FindOpenSSL: Added support for pkg-config.
The command generating the jar file depends on java_class_filelist which
is generated by another command.
Signed-off-by: Andreas Schneider <asn@cryptomilk.org>
Compile java sources within a custom_command using automatically checked
file dependencies and create java_class_filelist after java compile.
Signed-off-by: Andreas Schneider <asn@cryptomilk.org>
The regex was slightly wrong, it excluded the last line, so e.g.
/usr/include/ didn't end up in the .cproject file.
Thanks to Shash Chatterjee for the patch.
Alex
The AMD64 ABI document http://www.x86-64.org/documentation/abi.pdf
does specify that 64bits binary libraries should end up in <prefix>/lib64
and 32bits ones in <prefix>/lib. All but debian based distros do so,
and some like OpenSUSE even enforce the rule when packaging with RPM
and refuse to build the RPM if this is not the case.
After some discussion (see the bug notes) we cannot do that behind
the scene and the current fix supposes that the user shall use
the CMAKE_INSTALL_LIBDIR variables content in its INSTALL rules if
he wants to put the lib in the right place. CMAKE_INSTALL_LIBDIR
shall have the appropriate value depending on the Linux distribution
found and 32/64bitness of the host.
The cross-compiling case (even 32bits compile on a 64bits host)
is not handled.
The Cray Fortran compiler started using module init symbols in version 7.3.2.
Starting in commit 71287734 (Teach FortranC interface for Intel, PGI, and gcc
4.2, 2009-08-05) we provide C versions of the module init symbols so that the
detection executable can link when the C versions of the module-mangled symbols
are picked up.
If no C module-mangled symbol matches then we cannot let the C module init
symbol appear because it will be duplicated by the Fortran copy that provides
the module-mangled symbol. This was first handled for the PathScale compiler
in commit 21faaa5d (FortranCInterface: Fix PathScale detection, 2010-01-22) and
commit 46858720 (FortranCInterface: Fix PathScale detection again, 2010-02-16).
Handle it now for the Cray compiler too.
86cb17b Pass include directories with response files to GNU on Windows
9a0b9bc Optionally pass include directories with response files
6e8a67f Generate target-wide flags before individual build rules
d099546 Factor old-style -D flags out from -I flag generation
Also, comment out all "debugging" calls to message() that helped
us interpret the output on other platforms when running on the
dashboard clients.
Using ERROR_QUIET avoids unnecessary stderr output while calling
external tools to determine the processor count. If there's an
error parsing the output, we set the count to 0 anyhow.
Also, the test will fail on a CMake dashboard run if the count
comes back equal to 0.
Now that the code is "done"-ish, remove the debugging output.
Expect no output on stdout or stderr when calling the
ProcessorCount function from now on.
This check if the function exists and the prototype we want to use is
correct. There are still functions which have different prototypes on
different UNIX systems.
The GNU 4.x toolchain on MinGW (and therefore MSYS) allows compiler
options to be passed via response files. Use this to pass include
directory -I options. This allows the include file search path to be
very long despite shell and mingw32-make command line length limits.
Adjust whitespace to make the output of "--help-module FindMPI" look
good. Also separate the comment containing the copyright and license
notice so it does not appear in the documentation.
Adds support for:
- MPI_<lang>_COMPILER and other useful variables for C, CXX, Fortran
- Better compiler interrogation (handles mvapich)
- Supports specifying an MPI compiler name directly on the command line
without and absolute path, e.g.: cmake -D MPI_CXX_COMPILER=mpixlC
- Better compiler name searching tries to match MPI compiler to regular
CMAKE_<lang>_COMPILER_ID, if it's available.
Gets rid of:
- MPI_LIBRARY, MPI_EXTRA_LIBRARY cache variables. These and other old
vars are still exported for backward compatibility, but they're not
cached.
Also, add message output (temporarily) for gathering data
on all the dashboard machines. After the test runs on the
overnight dashboards tonight, I'll comment out the message
output and commit/push again.
Credit goes to David Cole ( http://www.kitware.com/blog/home/post/63 ).
Also add a script-based test of the new module.
Signed-off-by: Michael Wild <themiwi@users.sourceforge.net>
ade04de Remove debbuging typo
a201028 CPack try to please SUSE 64 bits and install lib in lib64 and not lib.
7ebbcf1 CPackRPM non matching ENDIF
0e07b42 CPackRPM even more trace in debug mode or in case of failure
564b731 CPackRPM add more trace output in order to help failing diagnostics
b3ce420 Do not bother enabling C++ in Assembler test
80f6a34 Teach Assembler test to generate main.s at build time
1dafa74 Fix Assembler test to parse C flags string before using
4139a73 -only enable the asm test for the Intel compiler if we are under UNIX
1f6c6b1 -use CMAKE_C_FLAGS when generating the assembler file
d0f71e2 Add ASM platform information for HP compiler on HP
bd580be Merge branch 'hp-compiler-info' into ReworkedAsmSupport
a0bab7a Add ASM platform information for XL compiler on AIX
c03b610 Merge branch 'aix-xl-platform-info' into ReworkedAsmSupport
c623008 Initialize ASM rpath flags for executables with those for shared libs
17c658f Add support for the Intel compiler used for ASM under Windows
4258b24 Add more regex for gcc, always print the ASM compiler ID
9071b8b Add temporary debug output for compiler ID detection for ASM
48f7199 It's ELSEIF(), not ELSIF()
d103c75 Fix bad comparison in the detect assembler-code
20fe0be Only try assembler support for Makefile-based generators
8614470 Use a regexp instead a lot of ORs for checking the compiler ID
cf88092 The Assembler test now tests ASM for GNU, Intel, HP, XL and SunPro
66614a8 Add assemble- and preprocess commands for HP
78f7c59 Fix the default CMAKE_ASM_COMPILE_OBJECT, make XL-ASM use it
7456461 Change the default rules so they fit better to the new ASM handling
5542d58 Set the HP asm file suffix
f745220 Add support for ASm for the HP compiler.
00735d4 Add suport for ASM for the IBM XL compiler
d3e9e8a Add support for ASM for the SunPro compiler
64e66eb Actually use CMAKE_ASM_COMPILER for asm, instead of CMAKE_C_COMPILER
38f92bf Add ASM support for the Intel compiler
4b40d42 Rework the way assembler is handled, use the C/CXX compiler by default
Move HP flags out of Platform/HP-UX.cmake into platform-specific
compiler information files "Platform/HP-UX-HP-<lang>.cmake". Factor
common values into "Platform/HP-UX-HP.cmake" and load it from the
per-language files.
Since the flags for executables are almost always the same as those for
shared libraries each language information file just uses the latter for
the former by default. This reduces duplication in the compiler and
platform flag information files.
Move RPATH flags out of Platform/AIX.cmake into platform-specific
compiler information files Platform/AIX-XL and Platform/AIX-GNU.
The flags need to be set for each compiler of each language.
CMAKE_ASM_COMPILE_OBJECT is now so that it fits for GNU, Intel, HP and Sun
ASM_ATT had to be adjusted. Also adjusted the default
CMAKE_ASM_LINK_EXECUTABLE
Alex
Commit 6a61a8a5 (Honor module .def files with MinGW tools, 2011-02-21)
set CMAKE_LINK_DEF_FILE_FLAG to "-Wl," to enable passing .def files to
the linker on MinGW. However, older GNU tools in the MSYS shell do not
know how to translate "-Wl,/c/..." to "c:/..." and complain that the
file does not exist. Instead set the flag to just "" which tells CMake
it can pass the file through the front-end with no special flag.
7a1027a When checking find_package() components, special case qtmain.
4c1c358 FindQt4: Include builtin FindPackageHandleStandardArgs directly
43cb9b8 Change to use fphsa to check required variables and version.
Since commit 024d05ad (Fix use of module .def files for MS tools,
2009-09-29) module .def files work for any platform that sets
CMAKE_LINK_DEF_FILE_FLAG correctly. Set it in the Windows-GNU platform
information file to enable support with MinGW tools. Also enable the
test added by commit 0db2c850 (Test use of module .def files for MS
tools, 2009-09-29) for MinGW and MSYS generators.
When initializing CMAKE_(EXE|SHARED|MODULE)_LINKER_FLAGS from LDFLAGS
and CMAKE_(EXE|SHARED|MODULE)_LINKER_FLAGS_INIT quote the whole string
in case both are set.
Reported-by: Daniel R. Gomez <gomez@teragram.com>
Clients may define CTEST_TEST_TARGET_ALIAS in a project's
initial cache to get a target of another name that executes
the same underlying command as "make test"
When running
$ mpif77 -c example.f -o example.f.o
mpif77 recognizes -o and produces example.f.o, but when running
$ mpif77 -o example.f.o -c example.f
the -o option is ignored and the object file is example.o. Performing
the same experiment on the underlying compiler tool or with the mpicc
and mpiCC wrappers does not exhibit this behavior, so the issue appears
to be specific to mpif77.
Reported-by: Zhen Wang <zwang26@emory.edu>
This commit changes the way how the assembler support works in cmake.
The language "ASM" now always uses the C/Cxx compiler instead
of the assembler directly. This fixes#8392, assembler files are
not preprocessed.
If one wants to use the assembler directly, the specific
assembler "dialect" has to be enabled. I.e. to get as/gas,
you have to use now ASM-ATT, the same way for ASM_MASM and ASM_NASM.
Implemented this now for gcc.
SunStudio, IBM, HP and Intel still todo.
Alex
dd2f814 Merge branch 'dev/add_test-working-directory' into dev/strict-mode
949d32c Unwatch manual variables upon removal in cmake-gui
3939032 Unwatch manual variables upon removal in ccmake
8354413 Add method to unwatch a manual variable
8ed3c85 Give a better message for unused variables
729db48 Fix ArgumentExpansion test expected results
89c2544 Checking for a definition is a usage
5625dee Don't output to stderr in the GUI
ad25a96 Merge branch 'ImprovedDotSupport2' into dev/strict-mode
c128abe Merge branch 'AddCMAKE_CURRENT_LIST_DIR' into dev/strict-mode
9bcaff0 Merge branch 'cmake-guiRememberAdvancedCheckbox' into dev/strict-mode
544d0c3 Fix expected output for WarnUninitialized test
4e3bea4 Update expected messages to new format
8e8c9e4 Don't check at destruction for usage
668e005 Use cmake::IssueMessage for warnings
88cd4c1 Use 'CMake Warning' versus 'warning' for CDash
3c3b98d Initialize the class before setting warn flags
cf8b15a Ignore files under the CMakeFiles directory
fd50f06 Don't check for unused vars at configure time
447a04c Don't warn during configure when doing everything
b97ee21 Check for unused variables at the end of generate
c18c977 When checking for variables, specify a reason
3f1121f Use a long int since Line is a long as well
2507f93 Change the failure case string to 'Unexpected'
fe390a2 Add 'ArgumentExpansion' test
8dbb209 Wrong boolean value for CLI warnings
d4ee998 Hard-code the --no-warn-unused-cli flag
a267b99 Fix line lengths
82ed104 Flag that the directories have been set
5aa535b Add argument to arg parsing to not set directories
367e5c3 Revert "Revert "When calling CMake, set the args and the cache""
ab5d4e4 Revert "When calling CMake, set the args and the cache"
9b90040 When calling CMake, set the args and the cache
fe56002 Fix long lines for KWStyle
5d30cfc Set a watch on variables added through the gui
33c63b1 Add a method to put a watch for variables
535253f Initialize the warning variables earlier
cbb286c Fix the path detection to work for top-level
62be1f7 Initialize the usage stack earlier
c6e7fab Factor out the checks for unused variables
5e41ba8 When using the API, check for Add vs. Remove
dee1976 Fix typo in VariableUnusedViaUnset test
f231ce5 Remove old false positive avoidance code
a117e02 Revert "Add test for unused warnings at the end of scope"
2c82f2b Exempt CMAKE(CURRENT|PARENT)_LIST_FILE from usage
6d7d449 Ignore CLI warnings for ABI determination
7740a73 Only return local keys that are defined
bef3aee Use the API so that warnings can be tracked
05cb0f4 Check for unused variables in the dtor
91c4c99 Add test for unused warnings at the end of scope
ca90f67 Fix detection of unused variables when setting
f7438ca Add test for unused detection via setting it
995cfb0 Don't warn if the variable wasn't defined
aefc91d Add test for usage checks via unset
a8e97f8 Remove VarRemoved code since it's been superceded
59463ef Rework CheckVariableForUnused usage
f117423 Fix line lengths to be no more than 78
e49a935 Improve unused warning logic
e01e40c Mark ARGC, ARGV*, and ARGN as used
a17aff7 Ignore CMAKE_MATCH_* variables for usage
02a114d Add method to allow variables to be marked as used
a0b0d23 CMAKE_DO_TRY_COMPILE is no longer used
ae3eff3 Fix the path used for ignoring system warnings
056b441 Fix missing case for usage of a variable
980e048 Factor out checks for unused variables
83acb0a Remove now unused variables
3801463 Use built-ins for readability and maintainability
8b52015 Push the initialize and unused states when copying
439877f Be consistent with single and double quotes
4cf1706 Add documentation for check-system-vars
b74777f Fix the spelling of the flag for warn-unused-vars
b948120 Change logic of flag to turn off cli unused checks
f047a17 Add test for uninitialized variables
75bda38 Add tests for unused command line variables
300fc15 Fix detection of system files
d784e6a Run the unused variables check on the final pass
9efc057 VariableWatch is not available when bootstrapping
2e78224 Add a missing comma to the warning message
7499700 Add a flag to warn about system files
fff9f6d Rename flags again and use variablewatch for cli
786e269 Add warn-unused to the Qt interface
636e6c4 Default to marking things as used
4ff0340 Rename find-unused to warn-unused
d7999e9 Rename strict-mode to warn-uninitialized
e141bc9 Detect unused variables
d3e8eb5 Add flags to detect unused variables
f332e14 Complete strict-mode checks for uninitialized vars
52f9637 Add method to get the local scope variables
f794d58 Make --strict-mode option, and integrate with cmake-gui
48b5b85 Add a warning when variables are used uninitialized.
cd626ea For macros make sure the FilePath points to a valid pointer in the args.
The FindPackageHandleStandardArgs module was originally created outside
of CMake. It was added for CMake 2.6.0 by commit e118a627 (add a macro
FIND_PACKAGE_HANDLE_STANDARD_ARGS..., 2007-07-18). However, it also
proliferated into a number of other projects that at the time required
only CMake 2.4 and thus could not depend on CMake to provide the module.
CMake's own find modules started using the module in commit b5f656e0
(use the new FIND_PACKAGE_HANDLE_STANDARD_ARGS in some of the FindXXX
modules..., 2007-07-18).
Then commit d358cf5c (add 2nd, more powerful mode to
find_package_handle_standard_args, 2010-07-29) added a new feature to
the interface of the module that was fully optional and backward
compatible with all existing users of the module. Later commit 5f183caa
(FindZLIB: use the FPHSA version mode, 2010-08-04) and others shortly
thereafter started using the new interface in CMake's own find modules.
This change was also backward compatible because it was only an
implementation detail within each module.
Unforutnately these changes introduced a problem for projects that still
have an old copy of FindPackageHandleStandardArgs in CMAKE_MODULE_PATH.
When any such project uses one of CMake's builtin find modules the line
include(FindPackageHandleStandardArgs)
loads the copy from the project which does not have the new interface!
Then the including find module tries to use the new interface with the
old module and fails.
Whether this breakage can be considered a backward incompatible change
in CMake is debatable. The situation is analagous to copying a standard
library header from one version of a compiler into a project and then
observing problems when the next version of the compiler reports errors
in its other headers that depend on its new version of the original
header. Nevertheless it is a change to CMake that causes problems for
projects that worked with previous versions.
This problem was discovered during the 2.8.3 release candidate cycle.
It is an instance of a more general problem with projects that provide
their own versions of CMake modules when other CMake modules depend on
them. At the time we resolved this instance of the problem with commit
b0118402 (Use absolute path to FindPackageHandleStandardArgs.cmake
everywhere, 2010-09-28) for the 2.8.3 release.
In order to address the more general problem we introduced policy
CMP0017 in commit db44848f (Prefer files from CMAKE_ROOT when including
from CMAKE_ROOT, 2010-11-17). That change was followed by commit
ce28737c (Remove usage of CMAKE_CURRENT_LIST_DIR now that we have
CMP0017, 2010-12-20) which reverted the original workaround in favor of
using the policy. However, existing project releases do not set the
policy behavior to NEW and therefore still exhibit the problem.
We introduced in commit a364daf1 (Allow users to specify defaults for
unset policies, 2011-01-03) an option for users to build existing
projects by adding -DCMAKE_POLICY_DEFAULT_CMP0017=NEW to the command
line. Unfortunately this solution still does not allow such projects to
build out of the box, and there is no good way to suggest the use of the
new option.
The only remaining solution to keep existing projects that exhibit this
problem building is to restore the change originally made in commit
b0118402 (Use absolute path to FindPackageHandleStandardArgs.cmake
everywhere, 2010-09-28). This also avoids policy CMP0017 warnings for
this particular instance of the problem the policy addresses.
In commit 295b5b60 (Honor CMAKE_USER_MAKE_RULES_OVERRIDE in try_compile,
2010-06-29) we started passing the value of this variable when building
a try_compile project. If the variable contains a relative path it must
be treated with respect to the file where it is first used. Ensure that
the value is converted to a full path so that it is correctly referenced
in the try_compile projects.
The module header was being placed in the source tree before. Thanks to
Marcel Loose for the patch, this ensures the file is written to the
build tree.
The parent commit added a warning message whenever a required file
does not exist.
As it turns out, the "required" files never exist when built with
Visual Studio Express editions. Add a variable to suppress these
warning messages because only packagers or naive includers of
this file will care to see such warning messages.
We want to warn about this condition by default so that people who
are using InstallRequiredSystemLibraries without understanding it
fully will have a chance of understanding why it's not working in
the event of missing required files.
But we also want to give projects the ability to suppress this warning
(by "project's choice default") so that they can encourage users who
are restricted to using an Express edition to build their project.
Packagers should explicitly use...
-DCMAKE_INSTALL_SYSTEM_RUNTIME_LIBS_NO_WARNINGS=OFF
...when building releases. That way, their release build process will warn
them about any missing files, but only if their project CMakeLists files
use a construct similar to CMake's:
IF(NOT DEFINED CMAKE_INSTALL_SYSTEM_RUNTIME_LIBS_NO_WARNINGS)
SET(CMAKE_INSTALL_SYSTEM_RUNTIME_LIBS_NO_WARNINGS ON)
ENDIF()
Thanks to "J Decker" on the CMake mailing list for pointing out
that one of the MSVC10_CRT_DIR settings was using "VC90" instead
of "VC100".
After fixing that, I added the code to generate a CMake warning
if one of the files we think is "required" does not exist.
Then, with VS10, there were several other problems that the
warning revealed:
- MSVC10_REDIST_DIR needed more PATHS to be found correctly
- the 64-bit directory is named "x64" now, not "amd64" as in
previous VS versions
- manifest files no longer exist as separate files in the
redist subdirectories (they must be built-in as resources
to the dlls...?)
The check works for macros, functions, and variables, but not for types
or enumeration values. Clearly describe the behavior of the check with
respect to each symbol type.
ce28737 Remove usage of CMAKE_CURRENT_LIST_DIR now that we have CMP0017
7db8db5 Improve documentation and messages for the new CMP0017
db44848 Prefer files from CMAKE_ROOT when including from CMAKE_ROOT
1e69c6f Merge branch 'user-policy-defaults' into policy-CMP0017
65a0a2a Merge branch 'include-command-whitespace' into policy-CMP0017
Control the root directory of the default directory presented to
the end user of an NSIS installer by a CPack variable.
Previously, the value used in the NSIS script was $PROGRAMFILES,
which is equivalent to the "ProgramFiles" environment variable.
That default value is still the same, but now a project may
override the value by setting this new variable.
Introduced an additional variable, Python_ADDITIONAL_VERSIONS, to both
FindPythonLibs and FindPythonInterp. Changed FindPythonInterp to loop
over versions rather than hardcoding all versions (more like libs).
Commit aaa88d33 (Build large archives incrementally, 2008-08-04)
hard-coded rules to create static archives incrementally in each
language information file. Set each rule conditionally to allow
compiler and platform information files to override the language
default rules.
Inspired-by: Harald Pohl <pohl.h@eppendorf.de>
Unset VS_UNICODE_OUTPUT when executing a command whose output
is being logged to a file. Previously, running Microsoft tools
in sub-processes of Visual Studio would send their output to
the Visual Studio output pipe named by this environment variable.
Unsetting it forces the output back to the normal stdout and stderr
channels where cmake can intercept it and direct it to the
appropriate log files.
MUI_FINISHPAGE_RUN is frequently used with NSIS and provides a checkbox
on the finish page of an installer which specifies whether the specified
executable should be run when the installer exits. This commit adds support
for this setting in CPack.
NSIS installers default to assuming the executables exist in a
directory named "bin" under the installation directory. As this
isn't usual for Windows programs, the addition of this variable
allows the customization of this directory and links still to be
created correctly.
Add support to InstallRequiredSystemLibraries to only install
debug libraries when both debug and release versions are available.
This is as if you are building a debug package then only the debug
versions are needed but not the release.
InstallRequiredSystemLibraries currently defaults to installing to
bin on WIN32 and lib otherwise. This patch allows you to configure
this by using the variable CMAKE_INSTALL_SYSTEM_RUNTIME_DESTINATION.
It also switches the logic to use a single INSTALL(PROGRAMS) command
rather than two deprecated uses of the INSTALL_PROGRAMS command.
In InstallRequiredSystemLibraries the documentation details the
variable CMAKE_SKIP_INSTALL_RULES to skip installation. This
actually doesn't do anything, the variable required is named
CMAKE_INSTALL_SYSTEM_RUNTIME_LIBS_SKIP. This commit amends the
documentation to point to the correct variable.
971692c Build enable_language command during bootstrap
960ace1 Add testing for windows resources for mingw/msys/cygwin and remove for watcom.
060d6e8 Add support for windres to cygwin.
b2f308c Add support for windows resources with mingw/msys.
This puts the new search behaviour for included files in action, i.e.
now when a file from Modules/ include()s another file, it also gets the
one from Modules/ included, i.e. the one it expects.
Alex
Since commit 70c2dc8a (Make compiler id detection more robust,
2008-03-10) we store compiler identification strings in test binaries
using the form
char* info = "info";
Use the const-correct
char const* info = "info";
form instead. This allows the C++ compiler identification to work with
"-Werror -Wall" or equivalent flags if the compiler would warn about
const-to-non-const conversion.
The TI DSP compiler predefines "__TI_COMPILER_VERSION__". Use this to
identify the C and C++ compilers. For assembler language the C compiler
executable is used:
$ cl6x -h
TMS320C6x C/C++ Compiler v6.1.11
Tools Copyright (c) 1996-2009 Texas Instruments Incorporated
Use this command-line option and output to recognize the assembler.
The Mac linker defines flag -headerpad_max_install_names but not all
front-ends recognize the flag and pass it through (many did in the past,
such as the Apple port of GCC). Use the -Wl, option prefix to tell
front-ends to pass it through without trying to interpret it.
One of Cygwin's goals is to build projects using the POSIX API with no
Windows awareness. Many CMake-built projects have been written to test
for UNIX and WIN32 but not CYGWIN. The preferred behavior under Cygwin
in such projects is to take the UNIX path but not the WIN32 path.
Unfortunately this change is BACKWARDS INCOMPATIBLE for Cygwin-aware
CMake projects! Some projects that previously built under Cygwin and
are Cygwin-aware when they test for WIN32 may now behave differently.
Eventually these projects will need to be updated, but to help users
build them in the meantime we print a warning about the change in
behavior. Furthermore, one may set CMAKE_LEGACY_CYGWIN_WIN32 to request
old behavior during the transition.
Normally we avoid backwards incompatible changes, but we make an
exception in this case for a few reasons:
(1) This behavior is preferred by Cygwin's design goals.
(2) A warning provides a clear path forward for everyone who may see
incompatible behavior, and CMAKE_LEGACY_CYGWIN_WIN32 provides a
compatibility option. The warning and compatibility option both
disappear when the minimum required version of CMake in a project is
sufficiently new, so this issue will simply go away over time as
projects are updated to account for the change.
(3) The fixes required to update projects are fairly insignificant.
Furthermore, the Cygwin distribution has no releases itself so project
versions that predate said fixes tend to be difficult to build anyway.
(4) This change enables many CMake-built projects that did not
previously build under Cygwin to work out-of-the-box. From bug #10122:
"I have built over 120 different source packages with (my patched)
CMake, including most of KDE4, and have found that NOT defining
WIN32 on Cygwin is much more accurate." -- Yaakov Selkowitz
A fully compatible change would require patches on top of these project
releases for Cygwin even though they otherwise need not be aware of it.
(5) Yaakov has been maintaining a fork of CMake with this change for the
Cygwin Ports distribution. It works well in practice. By accepting the
change in upstream CMake we avoid confusion between the versions.
CMake itself builds without WIN32 defined on Cygwin. Simply disable
CMAKE_LEGACY_CYGWIN_WIN32 explicitly in our own CMakeLists.txt file.
09d1c10 FortranCInterface: Recognize NAG Fortran module symbols
af2ad90 Add NAG Fortran compiler information files
24cc3d4 Recognize the NAG Fortran compiler
83892c4 Allow Fortran platform files to set empty values
fe3f878 Detect object files in implicit link information
Use "int main(void)" instead of just "int main()" so that compiling with
"gcc -Werror=strict-prototypes" works. Test this check using the flags
"-Werror -Wstrict-prototypes" to work with old GCC versions.
Since commit aff31479 (Modernize GNU compiler info on Windows,
2009-12-02) the file Modules/Platform/Windows-g++.cmake has been unused.
It just includes the non-existent Modules/Platform/Windows-gcc.cmake so
remove it outright.
This moves Intel compiler info on Windows into new-style modules
Platform/Windows-Intel-<lang>.cmake
using language-independent helper module
Platform/Windows-Intel.cmake
to define macros consolidating the information.
Commit 4430bccc (Change the way 32/64 bit compiles are detected with
MSVC and intel, 2009-11-19) added detection of the target processor to C
and CXX language builds with MS and Intel tools. Do the same for Intel
Fortran for Windows (ifort). Use /machine:<arch> to link executables.
3bc828d Fixed bug where last entry would be lost.
b316087 Escape file write expansion, and build up lists.
68cd3fe Added CMAKE_CACHE_ARGS to ExternalProject.
27ee50a Add ECLIPSE_CDT4_GENERATE_SOURCE_PROJECT as a ADVANCED cache variable (#9631)
50d21d4 Add cache var CMAKE_ECLIPSE_MAKE_ARGUMENTS when using the Eclipse generator
This variable can be set to command line arguments which will be passed
to make when eclipse invokes make, e.g. you can enter "-j8" to get
8 parallel builds (#9930)
Alex
Escaped the @var@ in the file writes - this was being expanded at file
write and so not causing a reconfigure at the right time. I also took
care of build up lists of lists in the variables, especially important
for things like MPI_EXTRA_LIBRARY. Added some error checking, and use
the tmp_dir for initial cache file.
On Windows the limit for command line arguments is 8192 characters, and
this was limiting longer paths with some of our more nested projects
such as Library. Placing the -D arguments into CMAKE_CACHE_ARGS will
write out an initial cache file, that will be passed to CMake with a -C
argument as the initial cache.
By forcing the cache variables we preserve the existing behavior with
-D, to change the values of cache variables in our inner projects.
On Linux the NAG Fortran compiler uses gcc under the hood to link. Use
"-Wl,-v" to pass "-v" to the underlying gcc compiler to get verbose link
output. Detect the NAG Fortran directory (using -dryrun) and then honor
object files in the directory referenced in the implicit link line.
Pass real linker options with "-Wl,-Xlinker,". The -Wl, gets through
the NAG front-end and the -Xlinker gets through the gcc front-end.
The Numerical Algorithms Group (NAG) Fortran compiler does not document
a preprocessor macro to identify it. Check for identifying output using
the -V option.
Teach CMakeFortranInformation to use default flags only for variables
that have not been set at all, rather then not set or empty. This will
allow platform or compiler-specific information files to set empty
values without getting the defaults.
The NAG Fortran compiler implicitly passes object files by full path to
the linker. Teach CMakeParseImplicitLinkInfo to parse object files that
match some tool-specific regular expression.
FindTCL.cmake switched variables in the FIND_LIBRARY invocation. The
FIND_LIBRARY() statement for TCL used the TK variables and vice versa.
This patch reverses that into the right usage.
Closes debian issue 600245.
Also, improve the documentation of the fixup_bundle and fixup_bundle_item
functions to clarify that plugin type "libs" need to be copied into
the bundle *before* calling fixup_bundle.
Commit e93a4b4d34 changed the way that
the libs parameter to fixup_bundle is interpreted. Before the commit,
the libs were copied into the bundle first and then fixed up. After
the commit, the copy was skipped, assuming the libs were in the bundle
in the first place, and then the fixups occurred as before.
However, before the commit, it was possible to name a lib from outside
the bundle, and have it copied in and then fixed up. Its resolved
embedded name was always inside the bundle before. After, its resolved
embedded name was just the same as its resolved name, which is in its
original location, and not necessarily inside the bundle.
This manifested itself as a problem with the ParaView call to
fixup_bundle and its many plugins. Previously, ParaView had simply
passed in the list of plugin file names as they existed in the build
tree, and left the copying into the bundle up to the fixup_bundle
function. When built with CMake 2.8.3 (the first version to contain
the above named commit) the fixup_bundle call would inadventently
fixup libraries in the build tree, not libraries that were in the
bundle. Furthermore, the plugins would not be in the final bundle.
This points out the fact that the fix for the bugs made by the above
commit was a backwards-incompatible change in behavior.
This commit makes it an error to try to fixup an item that is not
already inside the bundle to make the change in behavior apparent
to folks who were depending on the prior copy-in behavior: now,
they should get an error, and hopefully, reading the new and
improved documentation, should be able to resolve it in their
projects by adding code to install or copy in such libraries prior
to calling fixup_bundle.
Whew.
e0b6016 Some more fixes for nasm support, from Etienne (#10069)
d25c2eb Use CMAKE_ASM_NASM_FLAGS for nasm instead of FLAGS
e614e9b Add support for yasm, a nasm compatible assembler
79dd9be We already have 2010, fix copyright year.
ffeca06 Add missing copyright headers
7b337ac Improve misleading comments.
e1fc9b9 Add support for nasm assembler, patch by Peter Collingbourne (see #10069)
Use the Config mode of find_package to search for ITKConfig. This makes
FindITK a thin-wrapper around a standard find_package, bringing benefits
like searching lib64 paths when appropriate. This does for FindITK what
commit 2c1a01dc (Modernize FindVTK module, 2009-10-07) did for FindVTK.
The Mac linker defines -headerpad_max_install_names and the GCC
front-end passes this flag through. The PGI compiler does not know
about this flag, so we must use -Wl,-headerpad_max_install_names to pass
it to the linker instead.
CMakeDetermineASMCompiler.cmake relied on that somebody else (usually
during enabling C or CXX) already included that file, and broke if that
was not the case.
Thanks to Louis for the patch
Alex
The Cray Fortran compiler needs "-em" to enable module output and also
"-J." to place the .mod files in the current working directory (instead
of next to the .o file).
Clinton Stimpson
Todd Gamblin
Bill Hoffman
Brad King
Andreas Schneider
Matej Hribernik
Modestas Vainius
Julien Malik
Christoph Höger
Philip Lowman
Eric NOULARD
David Cole
Will Dicharry
Tim Gallagher
Manuel Klimek
Mathieu Malaterre
Rolf Eike Beer
Oliver Buchtala
Alex Neundorf
Clement Creusot
Martin Konrad
Alexey Ozeritsky
Michael Wild
Markus Rathgeb
Nikita Krupen'ko
Marcus D. Hanwell
Mike McQuaid
Wojciech Migda
Yaakov Selkowitz
Kai Wasserbäch
Ben Boeckel
Andrius Štikonas