Some libraries require the -all_load linker flag when linking to an Xcode project. However, this leads to a linker error if there are symbol conflicts among libraries. The solution is to use -force_load, which effectively lets you use -all_load on some libraries, but not on others.
However, this in turn leads to a new problem, at least for me. Whenever I use -force_load with a relative path to a library, the linker always finds symbol conflicts between the library and itself. It appears that the linker thinks that the library with its absolute path and the library with its relative path are different libraries, and therefore finds conflicts between the library and itself.
I can avoid this by using an absolute path with the flag. But this is not a wonderful solution, as it is convenient to keep source code for libraries within my documents directory. But the path to the documents directory will be different on other machines.
Question: Can anyone get force_load to work with a relative path to the library?
EDIT: for background information, see this question
With Xcode 4, if you include the library project into your app project, then you can add this to the Other Linker Flags:
-force_load $(BUILT_PRODUCTS_DIR)/<library_name.a>
You still need the dependency, and you need to add the library in the Link Phase list of frameworks and libraries too.
EDIT: Apple now says as of some Xcode 4 release that you can simply use this linker flag: "-ObjC" to get libraries with categories to properly load. That flag is working just fine for me in Xcode 5. People are still up voting this answer, but I suspect that the -ObjC flag is the best solution now.
This worked for me. Like the above answers you still need to include the library in the project.
-force_load $(SRCROOT)/pathToLibraryFromProject/libname.a
For the path it's just the folders in your project that lead to where you put your library, for example BaseFoler/Subfolder/libName.a.
Related
Long story short, I rewrote the Godot build system to cmake (only windows part), mostly because I wanted to learn it, but I have trouble compiling Godot with mingw. When I'm trying to compile it, at first everything goes fine, up until the point of linking final exe, where I get a lot of "undefined reference to" errors. It looks like main libraries (core/scene/editor/..) can't see functions from each other. MSVC build is working fine, and scons version is also compiling under mingw, so I clearly just missed something in my cmake version.. I tried to remove some compile/linking options throughout my cmake scripts as a test, but nothing changed. I don't really know how even debug this problem, so if someone could kick me in the right direction I would be really glad for it.
Ok, I finally got time to came back to this.
The problem
So basically the problem was in circular dependency on godot libs. I didn't thought this was the problem because I'm spoiled by MSVC (which doesn't depend on link order). Also, I tried to replicate circular dependency in mingw on a test project with much smaller scale. The project was a 30 libs with two functions in each, first function printing string, and another calling all first functions from all 30 libs, so there is 30 libs of circular dependency. Strangely enough, the project linked no problems, and printed 30^2 strings..
The solution
The solution is to use -Wl,--start-group/-Wl,--end-group linking flags around all libraries. There is two ways you can do it.
First way, is to add all your libraries to the some list of sorts. This could be global property, or property on some target (not just a simple variable), so it could be accessed from other subdirectories. After you formed your list of libraries you simply link it to the executable as follows
# getting all your libs from the global property..
get_property(__LIBS_LIST GLOBAL PROPERTY EXE_LIBS_LIST)
# linking all libraries to the exe..
target_link_libraries(my-exe PRIVATE -Wl,--start-group ${__LIBS_LIST} -Wl,--end-group)
This is the easiest solution, but be cautious about dependencies on libraries which you link to your exe, because it seems that when CMake creates link line for your exe, it first lists all libraries which are linked directly to your exe, and only after it places libraries which came from dependencies of libraries linked directly. Basically if your target dependency tree looks something like this:
exe // your main exe file
- lib_A // lib A linked directly to the main exe
- lib_AA // lib AA linked to the lib_A
- lib_AAA // lib AAA linked to the lib_AA
- lib_B // lib B linked directly to the main exe
- lib_BB // lib BB linked to the lib_B
- lib_BBB // lib BBB linked to the lib_BB
your link order for the exe will look something like this:
// first libs linked directly to the exe
lib_A
lib_B
// only after recursively initial libs dependencies
lib_AA
lib_AAA
lib_BB
lib_BBB
That's meen, that if you will link your libs like target_link_libraries(my-exe PRIVATE -Wl,--start-group ${__LIBS_LIST} -Wl,--end-group), --start-group and --end-group will guard only the libraries linked directly to your exe. I didn't found this described in documentation, but I found SO question which talks pretty much about same behaviour (CMake library linking order). Also, as I tested this on mingw, it didn't mattered how exactly libs lib_AA/lib_AAA/lib_BB/lib_BBB were linked, via PRIVATE or via INTERFACE, the results were the same.
Second way, is to exploit recursivity of link expanding for dependencies of libraries linked directly to the exe. From my example you can see, that dependencies of lib_A (lib_AA/lib_AAA) weren't mixed with dependencies of lib_B (lib_BB/lib_BBB). So basiacaly what we can do, is to create INTERFACE library and connect to it -Wl,--start-group immediately after that. Then add any number of libraries to it's interface and link global-libs to your exe (order does not matter). And in very end, close the group in your global-libs library
add_library(global-libs INTERFACE)
target_link_libraries(global-libs INTERFACE -Wl,--start-group)
# ...
# linking another libs, and linking global-libs to exe
# ...
target_link_libraries(global-libs INTERFACE -Wl,--end-group)
This will ensure, that all libraries connected to global-libs will be surrounded by -Wl,--start-group/-Wl,--end-group.
Now, theoretically, CMake should handle circular dependency by itself, by placing libraries in link line multiple times (how many times controlled by LINK_INTERFACE_MULTIPLICITY). But this method didn't worked for me (mb I just missed something..). Plus, you need declare dependencies between cmake targets, and with -Wl,--start-group/-Wl,--end-group you can just set one specific interface library as a holder for all libs with circular dependencies..
I have a project that uses several different open source libraries (FFmpeg, OpenSSL, for example).
No matter what I try, CMake is linking against the SYSTEM installed versions of these libraries, rather than the custom built ones I need for my projects.
Here is an example of what I've tried to add the FFmpeg library 'libswresample':
set(FFMPEG_PATH "/shared/dev/libs/ffmpeg/3.4.2")
find_library(LIBSWRESAMPLE_LIB NAMES swresample
PATHS ${FFMPEG_PATH}/lib/darwin-amd64
NO_DEFAULT_PATH
NO_CMAKE_ENVIRONMENT PATH
NO_CMAKE_PATH
NO_SYSTEM_ENVIRONMENT_PATH
NO_CMAKE_SYSTEM_PATH
NO_CMAKE_FIND_ROOT_PATH)
list(APPEND includes "${FFMPEG_PATH}/include")
link_directories("${FFMPEG_PATH}/lib/${ARCH}")
list(APPEND libs ${LIBSWRESAMPLE_LIB})
MESSAGE(STATUS "libs: ${libs}")
I've tried setting CMAKE_PREFIX_PATH to the location of my shared libraries - no luck.
I've tried this with various combinations of the "NO_SOMETHING_PATH" options, which doesn't seem to help. I also tried just giving ${FFMPEG_PATH} to the PATHS parameter of find_library() and providing PATH_SUFFIXES lib ${ARCH} (which would be ideal, since I build this project for multiple platforms), but that didn't work either.
No matter what I've tried, MESSAGE outputs libs: /usr/local/lib/libswresample.dylib, rather than libs: /shared/dev/libs/ffmpeg/3.4.2/lib/darwin-amd64/libswresample.dylib
I've found several FindFFMpeg modules, but they all are pretty much doing what I'm trying here and wind up finding the system installed FFmpeg libraries rather than the one I actually want to link with.
If I explicitly provide the absolute path to the library I can get it to work, but this is obviously not optimal since some platforms use static libraries, some use shared libs, and so on. If I were to go that route, I'd have to additional work to figure out which platform I'm building for, and that doesn't seem like the right way to go about it anyways.
I know I must be missing something simple. Can anyone point me in the right direction?
The code works, but you should clear CMake cache (remove CMakeCache.txt file from build directory) for re-search the library.
Option NO_DEFAULT_PATH implies all other NO_* options, so you may omit them.
I am running Linux Redhat, I have Anaconda installed and I am trying to install a program (libspimage) using CMAKE amd I get the following warning/error:
CMake Warning at src/CMakeLists.txt:74 (ADD_LIBRARY):
Cannot generate a safe runtime search path for target _spimage_pybackend
because files in some directories may conflict with libraries in implicit
directories:
runtime library [libtiff.so.5] in /usr/lib64 may be hidden by files in:
/home/michantia/anaconda2/lib
Some of these libraries may not be found correctly.
When I do:
echo $PATH
I get:
/home/mi_a/anaconda2/bin:/usr/lib64/qt-3.3/bin:/usr/local/bin:/usr/local/sbin:/usr/bin:/usr/sbin:/bin:/sbin:/home/michantia/.local/bin:/home/michantia/bin
I tried:
export PATH=/usr/lib64:$PATH
hoping cmake would find the libraries in this directory before finding them in anancoda's, but that did not work. I also tried two other similar suggestions for a similar problem that I saw in stackoverflow, but that did not work.
Any other ideas are highly welcomed.
Warning message
Cannot generate a safe runtime search path for target
is related neither with CMake ability to find a library (libtiff.so.5 in your case) nor with a linker ability to link the library.
The warning message means that when a target (_spimage_pybackend) will be loaded, the loader will be unable to choose the correct library: according to the loader's algorithm and the target's setting, file /home/michantia/anaconda2/lib/libtiff.so.5 will be choosen instead of proper one /usr/lib64/libtiff.so.5.
The error is usually resulted in linking into the single target two libraries from different directories, when the directory with a second library also contains a file with the name of the first library:
Directory /usr/lib64 contains a library libtiff.so.5, which is linked into the target.
Directory /home/michantia/anaconda2/lib contains a library <A> which is also linked into the target; but this directory also contains a file libtiff.so.5.
According to CMake algorithm, runpath for the binary file of such target will include both directories, so both libraries could be found. But such runpath confuses the loader to find the first library properly.
Except from avoiding such situation (when a library is contained in two directories), one hardly is able to handle this warning.
I'm learning to build a llvm project, this is the reference: http://llvm.org/docs/Projects.html. I use the "llvm/projects/sample" directory as the primary project skeleton, and it works. Then I want to build tools from "llvm/examples" to my project, such as Fibonacci, it can't work. I do it this way: first copy the "llvm/examples/Fabonacci" directory to "MyProj/tools" ("MyProj" is top level of my project) and change Makefile to contain Fabonacci target, then configure & make. but the Fabonacci tool seems can't be built. It depends on some libs when linking. So what can I do if I want to build the source code from "llvm/projects/example" in my project?
You need to provide LLVM libraries to linker when building your own project. This means adding some flags, library directories and libraries themselves to link command. Build script probably needs some editing.
llvm-config tool can be used for providing necessary options to compiler/linker. Check documentation and examples.
I've build a Framework for an Objective-C App. I've tested it on minimalistic Programs where it worked. I was trying to use it in a real App now. Unfortunately the linker can't find the definition of my classes. =(
When I try to run it, I get the following Error Message:
Undefined symbols for architecture x86_64:
"_OBJC_CLASS_$_ClassInMyFramework", referenced from:
objc-class-ref in libMyLib.a(MyLib.o)
ld: symbol(s) not found for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see invocation)
The Dependency:
App -> libMyLib.a -> MyFramework.framework
The usual suggestion for this error message is to add the framework into the "Link Binary with Library" Build Phase... I can assure you that this has happened ;)
My first thought is that something might be wrong with the Build Settings which results in this linking error.
On second thought It may have to do with the Project-Setup. Is it possible to statically link a framework into an '.a' library file?
Update:
I've linked the Framework into the App and now it's working. But I don't consider this a clean solution. Help still appriciated. =)
The short answer is no, you can't link the framework statically into your .a file. See this discussion.
The reason is, the static library doesn't include the object code (the definition) of classes from the dynamic framework. The static library links to object code in the framework the same way the app links to the framework code: at run-time.
From Apple's Framework Programming Guide: "Dynamic shared libraries have characteristics that set them apart from static linked shared libraries. For static linked shared libraries, the symbols in the library are checked at link time to make sure they exist. If they don’t exist, link errors occur. With dynamic shared libraries, the binding of undefined symbols is delayed until the execution of the program."
It depends on what you want to be able to do with your code. You could add a static "target" for your framework project, so your framework project outputs both a framework and a static library. You could include this static library into apps.
But, one benefit of frameworks is that you can include nibs, images, headers, etc. So, linking your framework into your apps directly is not a bad way to go. Otherwise, you need to include these assets directly into your project. If you want this framework to be distributed with your app, you'll need to package it inside the app wrapper.
It looks like some people create a "static framework" for inclusion into an iOS project, but this looks a bit hacky to me.
As an interesting exercise, you can explore the symbols in your object code. Let's say you are using a ClassInMyFramework (from your framework) somewhere in your static library, like:
ClassInMyFramework *myFramework = [[ClassInMyFramework alloc] init];
The static library will then include the _OBJC_CLASS_$_ClassInMyFramework symbol. You can see the list of symbols in your static library file at the command line:
$ nm /path/to/libMyLib.a
This will output a list of symbols, which will show that _OBJC_CLASS_$_ClassInMyFramework is undefined (note, the "U" designates that the class is undefined):
U _OBJC_CLASS_$_ClassInMyFramework
Whereas, if you were do to the nm command on your framework:
$ nm /path/to/MyFramework.framework/Versions/A/MyFramework
Your output would show that the symbol is defined in your framework (though the definition will still only linked at run-time), which would look something like this, showing an address of the definition:
0000000000001100 S _OBJC_CLASS_$_ClassInMyFramework