I am trying to use cmake to link intel mkl for my code. From mkl link advisor, I should link the library by:
-L${MKLROOT}/lib/intel64 -lmkl_intel_lp64 -lmkl_core -lmkl_sequential -lpthread -lm -ldl
I then write a module to find mkl_intel_lp64, mkl_core, and mkl_sequential in Findmkl.cmake:
find_library(lp64_libraries NAMES mkl_intel_lp64 PATHS "${mkl_path}/lib/intel64")
find_library(core_libraries NAMES mkl_core PATHS "${mkl_path}/lib/intel64")
find_library(sequential_libraries NAMES mkl_sequential PATHS "${mkl_path}/lib/intel64")
Then I can link these libraries. While I do not know how to link -lpthread -lm -ldl, I am not sure what this link mean, I can not find any libpthread.a in $MKLROOT/lib/intel64.
Those libraries should exist on each Unix system.
So you can just add:
target_link_libraries(target_name m dl pthread)
Related
I can compile my project by running
g++ main.cpp -l:libpj-x86_64-unknown-linux-gnu.a -lpthread -lm -luuid
or
g++ main.cpp /usr/local/lib/libpj-x86_64-unknown-linux-gnu.a -lpthread -lm -luuid
But when I try adding library with either one of:
meson.get_compiler('cpp').find_library('libpj-x86_64-unknown-linux-gnu.a')
meson.get_compiler('cpp').find_library('/usr/local/lib/libpj-x86_64-unknown-linux-gnu.a')
I'm getting error:
ERROR: C++ library 'libpj-x86_64-unknown-linux-gnu' not found
Solution was to add 'dirs' variable even tho file is in standard /usr/local/lib path and to remove .a extension.
cc.find_library('libpj-x86_64-unknown-linux-gnu', dirs: '/usr/local/lib/')
Later of course lib was available inside meson and was added to executable and tested.
I'm building an SDL2/C++ program that needs to be portable to Windows, Mac, and Linux machines which may not have SDL installed.
I've read that static linking is the solution, but I'm not very good with compiling and don't know how to static link.
My program relies only on SDL2, GLU, and OpenGL. I'm compiling C++ with either MinGW (on Windows 8.1) or gcc (on Ubuntu 14.04) -- both of these OS's have SDL installed natively.
Here is my current makefile, derived from a sample makefile given to me by a professor of mine:
# Executable/file name
EXE=experiment
# MinGW
ifeq "$(OS)" "Windows_NT"
CFLG=-O3 -Wall -DUSEGLEW
LIBS= -lSDL2 -lglu32 -lopengl32
CLEAN=del *.exe *.o *.a
else
# OSX
ifeq "$(shell uname)" "Darwin"
CFLG=-O3 -Wall -Wno-deprecated-declarations
LIBS=-framework SDL2 -framework OpenGL
# Linux\Unix\Solaris
else
CFLG=-O3 -Wall
LIBS= `sdl2-config --cflags --libs` -lGLU -lGL -lm
endif
# OSX\Linux\Unix\Solaris
CLEAN=rm -f $(EXE) *.o *.a
endif
# Dependencies
$(EXE).o: $(EXE).cpp FORCE
.c.o:
gcc -c -o $# $(CFLG) $<
.cpp.o:
g++ -std=c++11 -c -o $# $(CFLG) $<
# Link
$(EXE):$(EXE).o
g++ -std=c++11 -O3 -o $# $^ $(LIBS)
# Clean
clean:
$(CLEAN)
# Force
FORCE:
To link with static library you either specify path to library file
gcc -o out_bin your_object_files.o path/to/lib.a -lfoo
or ask linker to use static version with -Bstatic linker flag. Usually you'll want to reset linking back to dynamic for the rest of the libraries, e.g. for static SDL2 and GLU but dynamic GL:
gcc -o out_bin your_object_files -Wl,-Bstatic -lSDL2 -lGLU -Wl,-Bdynamic -lGL
That of course implies that static versions of libraries are present in library search path list (.a libs for gcc on all specified platforms, although MSVC uses .lib for static libraries).
However you usually don't really want to do that at all. It is common practice for software to either depend on some libs (widespread on linux, with packages and dependendices lists) or bring required libraries with it. You can just distribute SDL dynamic library with your program and load it with LD_LIBRARY_PATH or relative rpath.
Please also note that newer SDL2 implements dynamic loading of functions which provides a way to override SDL with user-specified dynamic library, even if linked statically.
It wasn't related directly to static linking. When static linking, I had to include all of SDL's dependency libraries. Turns out, having -mwindows causes console communication to fail.
The correct link command is
g++ file1.o file2.o xxx.0 -o target -I./ -I/usr/local/libmylibone/
-L./ -L/usr/local/testlib/ ../lib/special_lib/static_lib.a
-lasn1c++ -lmysqlclient -lnsl -lm -lz -lc -ldl -lpthread -lrt -ljson
Please focus on ../lib/special_lib/static_lib.a, this is a static library and not named with libxxx.a. And I don't know how to write a CMake command to get this correct link command.
I've tried TARGET_LINK_LIBRARIES(../lib/special_lib/static_lib.a) and it will be translated to -l../lib/special_lib/static_lib.a. I've also tried TARGET_LINK_LIBRARIES(static_lib) but got -lstatic_lib and failed.
If you put the absolute path to your library it should work:
TARGET_LINK_LIBRARIES(your_binary /usr/local/lib/static_lib.a)
Second option:
ADD_LIBRARY(staticlib STATIC IMPORTED)
SET_TARGET_PROPERTIES(staticlib PROPERTIES IMPORTED_LOCATION /usr/local/lib/static_lib.a)
TARGET_LINK_LIBRARIES(your_binary staticlib)
The official CMake documentation for importing/exporting targets is here.
Is there a way to print a list the filenames for the libraries linked into a target via the target_link_libraries command
or even better, have all a target's dependencies copied to a specific folder?
get_target_property(OUT Target LINK_LIBRARIES)
message(STATUS ${OUT})
I realise this doesn't fully answer the question with regards doing it within cmake, but I faced a similar problem and thought I should share my solution.
First, in your source directory ("project"):
$ mkdir build && cd build
$ cmake ..
Then, use graphviz to create a dot file, as in this answer:
$ cmake --graphviz=graph.dot .
Then, strip out the dependencies from the graph for your target (let's call it "foo"):
$ sed -n 's/.*label="\(.*\)"\s.*/\1/p' graph.dot.foo > foo_dependencies.txt
Now, remove the clutter:
$ rm graph.dot*
Actually not(*).
However, you can use a cmake variable to collect the name of the libraries that you want to link (using the set( ... or the list(APPEND ... command), and then use this variable in your target_link_libraries command:
target_link_libraries(<targetname> ${YOUR_CMAKE_VARIABLE})
The same variable can also be used to create your copy commands (for example using this custom target)
(*) A similar question was asked here, and it got no definitive answer.
Well, all linked libraries to a given target TARGET are in the build.../CMakeFiles/TARGET.dir/link.txt.
For instance, TARGET=dirac.x,
.../build_intel17_mkl_i8/.less CMakeFiles/dirac.x.dir/link.txt
There will be a large number of linked libraries:
/cvmfs/it.gsi.de/compiler/intel/17.0/compilers_and_libraries_2017.4.196/linux/bin/intel64/ifort -Wl,-E -w -assume byterecl -g -traceback -DVAR_IFORT -i8 -w -assume byterecl -g -traceback -DVAR_IFORT -i8 -O3 -ip CMakeFiles/dirac.x.dir/src/main/main.F90.o -o dirac.x -L/tmp/milias-work/software/qch/dirac/devel_trunk/build_intel17_mkl_i8/external/lib -L/tmp/milias-work/software/qch/dirac/devel_trunk/build_intel17_mkl_i8/external/gen1int-build/external/lib -L/tmp/milias-work/software/qch/dirac/devel_trunk/build_intel17_mkl_i8/external/pelib-build/external/lib -Wl,-rpath,/tmp/milias-work/software/qch/dirac/devel_trunk/build_intel17_mkl_i8/external/lib:/tmp/milias-work/software/qch/dirac/devel_trunk/build_intel17_mkl_i8/external/gen1int-build/external/lib:/tmp/milias-work/software/qch/dirac/devel_trunk/build_intel17_mkl_i8/external/pelib-build/external/lib: libobjlib.dirac.x.a src/pelib/libpelib_interface.a external/lib/libstieltjes.a -limf -lsvml -lirng -lstdc++ -lm -lipgo -ldecimal -lstdc++ -lgcc -lgcc_s -lirc -lsvml -lc -lgcc -lgcc_s -lirc_s -ldl -lc external/pcmsolver/install/lib/libpcm.a /usr/lib/x86_64-linux-gnu/libz.so -limf -lsvml -lirng -lstdc++ -lm -lipgo -ldecimal -lstdc++ -lgcc -lgcc_s -lirc -lsvml -lc -lgcc -lgcc_s -lirc_s -ldl -lc /usr/lib/x86_64-linux-gnu/libz.so src/libxcfun_fortran_bindings.a external/xcfun-build/src/libxcfun.a external/lib/libpelib.a libgen1int_interface.a external/lib/libgen1int.a -Wl,--start-group /cvmfs/it.gsi.de/compiler/intel/17.0/compilers_and_libraries_2017.4.196/linux/mkl/lib/intel64/libmkl_lapack95_ilp64.a -lmkl_intel_ilp64 -qopenmp -Wl,--end-group -Wl,--start-group -lmkl_intel_ilp64 -lmkl_intel_thread -lmkl_core -lpthread /usr/lib/x86_64-linux-gnu/libm.so -qopenmp -Wl,--end-group external/pcmsolver/install/lib/libpcm.a external/xcfun-build/src/libxcfun.a external/lib/libpelib.a external/lib/libgen1int.a /cvmfs/it.gsi.de/compiler/intel/17.0/compilers_and_libraries_2017.4.196/linux/mkl/lib/intel64/libmkl_lapack95_ilp64.a -lmkl_intel_ilp64 -lmkl_intel_thread -lmkl_core -lpthread /usr/lib/x86_64-linux-gnu/libm.so -lirng -ldecimal -lstdc++
I have a GNU build system with autoconf-2.69, automake-1.14.1, libtool-2.4.2. I've configured with --host=i686-linux on a x86_64 RHEL6 host OS to build a 32-bit program. The libtool command seems to be:
/bin/sh ../libtool --tag=CXX --mode=link g++ -I/home/STools/RLX/boost/include/boost-1_44 -m32 -g3 -Wall -static -o engine engine-main.o ../components/librlxvm.la /home/STools/RLX/boost/include/boost-1_44/../../lib/libboost_program_options-gcc42-mt-1_44.a -lz -lpthread -ldl -lrt -ldl -lz -lm
But the real command is to search the 64-bit libraries not the 32-bit libraries as shown below:
libtool: link: g++ -I/home/STools/RLX/boost/include/boost-1_44 -m32 -g3 -Wall -o engine engine-main.o -L/home/robert_bu/src/gcc/gcc-4.2.2/build-x86_64/x86_64-unknown-linux-gnu/libstdc++-v3/src -L/home/robert_bu/src/gcc/gcc-4.2.2/build-x86_64/x86_64-unknown-linux-gnu/libstdc++-v3/src/.libs -L/home/robert_bu/src/gcc/gcc-4.2.2/build-x86_64/./gcc ../components/.libs/librlxvm.a /home/STools/RLX/boost/include/boost-1_44/../../lib/libboost_program_options-gcc42-mt-1_44.a /home/STools/RLX/gcc-4.2.2-x86_64/lib/../lib64/libstdc++.so -L/lib/../lib64 -L/usr/lib/../lib64 -lc -lgcc_s -lrt -ldl -lz -lm -pthread -Wl,-rpath -Wl,/home/STools/RLX/gcc-4.2.2-x86_64/lib/../lib64 -Wl,-rpath -Wl,/home/STools/RLX/gcc-4.2.2-x86_64/lib/../lib64
The --host config seems to have no effect. Is there anyway to tell libtool that 32-bit libraries are what we want?
It seems that libtool uses "CC", "CXX" to check the library search path. After I set CC to "gcc -m32", and CXX to "g++ -m32", it works. So libtool does not add "-m32" automatically even if I try to build a 32-bit program on a 64-bit system.
You're being hit by the problem of libtool .la files expansion. In particular libstdc++.la is being expanded for you to a full path rather than a simple -lstdc++.
My suggestion is to remove .la file from the SDK you're using (/home/STools). This way libtool can't assume things for you. Usually the ones you have in the system are fine, because the libraries are already in the search path, so it does not need to use -rpath or the full path to the .so file.
Depending on how well the SDK was crafted, this might or might not work correctly, so take it with a grain of salt.