Installed VS code and MinGW on windows. Added IntelliSense extension for C/C++. I had added the path of gcc.exe from the bin. Typed a simple C program. It's showing an error. Please help!
Error is:
PS F:\C\Chapter 1> gcc 001_first.c
C:/Program Files (x86)/mingw-w64/i686-8.1.0-posix-dwarf-rt_v6-rev0/mingw32/bin/../lib/gcc/i686-w64-mingw32/8.1.0/../../../../i686-w64-mingw32/lib/../lib/libmingw32.a(lib32_libmingw32_a-crt0_c.o):crt0_c.c:(.text.startup+0x39): undefined reference to `WinMain#16'
collect2.exe: error: ld returned 1 exit status
Also the problem is:
#include errors detected. Please update your include path. Squiggles are disabled for this translation unit( Path is mentioned )
Cannot open source file "stdio.h"
The IntelliSense error only means that VS Code can't find the header files. You need to set your include path. You can learn about it (and many others) here.
The second one is a bit more interesting. I also encountered it several times. The answer is: append -mconsole to the gcc command. It's because by default, it tries to build a desktop application, which requires WinMain.
Related
I am developing executables from source code of llvm. So I downloaded the llvm source code from github.
I am trying everything from command line on Windows OS
I am following the link for libtooling in clang
http://clang.llvm.org/docs/LibASTMatchersTutorial.html
I tried with 2 options
Option First: I ran the below command
cmake -G Ninja "C:\Users\amith.ks\Desktop\Clang-llvm\llvm-project\llvm" -DLLVM_ENABLE_PROJECTS="clang;clang-tools-extra" -DLLVM_BUILD_TESTS=ON -DCMAKE_BUILD_TYPE=Release
Everything Worked..
Second option:
I wan to set cmake_c_compiler and cmake_cxx_compiler from command line.
I dont want to use cmake-gui so I run the below command
cmake -G Ninja "C:\Users\amith.ks\Desktop\Clang-llvm\llvm-project\llvm" -DLLVM_ENABLE_PROJECTS="clang;clang-tools-extra" -DLLVM_BUILD_TESTS=ON -DCMAKE_BUILD_TYPE=Release -DCMAKE_C_COMPILER=clang -DCMAKE_CXX_COMPILER=clang++
From out of no where error was thrown
CMake Error at cmake/modules/CheckAtomic.cmake:53 (message):
Host compiler appears to require libatomic, but cannot find it.
Call Stack (most recent call first):
cmake/config-ix.cmake:343 (include)
CMakeLists.txt:617 (include)
When I saw the cmake error log It was saying this
LINK : fatal error LNK1104: cannot open file 'atomic.lib'
clang: error: linker command failed with exit code 1104 (use -v to see invocation)
ninja: build stopped: subcommand failed.
I searched my whole pc atomic.lib no where to be found.
How to solve this issue on windows?
Please help me with answers.
I know why it fails. It's a bug in the LLVM's cmake files. In order to check if atomic.lib is required CheckAtomic.cmake tries compiling and linking a piece of code with atomic.lib. When it fails (because atomic.lib is not required and thus is not present) it falsely concludes that atomic.lib is needed. LLVM's cmake files is a hot mess with rarely fixed bugs. There are bugs staying there for years. I guess this is because nobody understands anymore how LLVM build system works.
Long story short, I couldn't find why CheckAtomic.cmake inserts atomic.lib while checking that it is not required. As a workaround I just unconditionally set HAVE_CXX_ATOMICS64_WITHOUT_LIB and HAVE_CXX_ATOMICS_WITHOUT_LIB to True in CheckAtomic.cmake:
set(HAVE_CXX_ATOMICS64_WITHOUT_LIB True)
set(HAVE_CXX_ATOMICS_WITHOUT_LIB True)
Use clang-cl (or MSVC cl), they dont require lib atomic.
I'm trying to use Clang 11 (with MSVC ABI) to compile Clang 11. However, MSVC's headers won't compile in C++11 mode, which CheckAtomic.cmake uses:
set(CMAKE_REQUIRED_FLAGS "${CMAKE_REQUIRED_FLAGS} -std=c++11")
results in:
C:\Users\nyanpasu\code\llvm-project>clang -Werror=unguarded-availability-new -std=c++11 uwu.cpp
In file included from uwu.cpp:1:
In file included from C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.27.29110\include\atomic:19:
In file included from C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.27.29110\include\xatomic.h:13:
In file included from C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.27.29110\include\type_traits:11:
C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.27.29110\include\xstddef:283:22: error: 'auto' return
without trailing return type; deduced return types are a C++14 extension
_NODISCARD constexpr auto _Unfancy(_Ptrty _Ptr) noexcept { // converts from a fancy pointer to a plain pointer
^
1 error generated.
I don't know if this is a MSVC bug where it ships an header incompatible with C++11, or a Clang bug where it's using MSVC headers instead of its own.
In any case, changing c++11 to c++14 fixes this error.
As lulle mentioned, you need to use the right tools to compile. If you have installed Visual Studio, use the Developer Command prompt for Visual Studio instead of a regular cmd and you'll have the environment variables you need. It is located in your Visual Studio installation directory, under the Tools subdirectory.
For example:
C:\Program Files (x86)\Microsoft Visual Studio\2019\Enterprise\Common7\Tools\VsDevCmd.bat
I encountered the same error in macOS. Adding the CPP Libraries to the path solved it for me. You can try executing the following commands.
export SDKROOT="$(xcrun --sdk macosx --show-sdk-path)"
export CPLUS_INCLUDE_PATH=/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/../include/c++/v1
Okay I'm having a particular issue during compilation using the command gcc -shared -o attack_test.so -fPIC attack_test.c. Specifically, when I run the command, I get the error fatal error: ec.h: No such file or directory. Any idea how to fix this issue? Thanks.
EDIT: To give a little more issue background, I'm running this command from the desktop, where my C source code is located. Perhaps there's a specific place I need to put the source and run the command?
EDIT 2: So I was able to get past the above error by running the command sudo apt-get install libssl-dev, and then changing the line #include <ec.h> to #include <openssl/ec.h>, which seemed to negate the issue. Adding the <openssl/...> to the other includes did not work the same way. So now, instead of fatal error: ec.h: No such file or directory, it's fatal error: ec_plugins.h: No such file or directory. Any ideas for possible fixes? Thanks so much!
For reference, I'm building upon the the dos_attack.c source code from this link.
TO BE CLEAR: This is for a project and tested upon in a CLOSED ENVIRONMENT. This is COMPLETELY* for EDUCATIONAL use and for the study of Denial-of-Service and Man-In-the-Middle mitigation techniques.
<ec.h> is intended to refer to a file in the Ettercap include directory, not an OpenSSL header. The same goes for the other ec_*.h headers.
Adding -I/path/to/ettercap/include to your compiler flags should fix this.
I'm using MinGW to compile C code. but when I give command "make", appear error : gcc.exe : fatal error: -fuse-linker-plugin, but liblto_plugin-0.dll not found compilation terminated. make: * [all] Error 1. Does anybody know how to solve it?
though a long time since asked, I faced the same issue and I found a fine workaround when compiling.
gcc -fno-use-linker-plugin test.cpp -o test
Check your PATH environment variable and look for a path to another MinGW installation. I have two MinGW installed (first one in Falcon C++ IDE directory), and second one I installed by myself. There was two paths to MinGW in system PATH environment variable and that induced the error.
I'm getting some kind of linking error with a program utilizing MS-s waveform API. The code I'm using can be found here: http://www.planet-source-code.com/vb/scripts/ShowCode.asp?txtCodeId=4422&lngWId=3
It's the first bit of code on the page under the heading "Opening the Sound Device". I'm not getting any errors with the code, however my linker complains quite persistently. The output being the following:
d:\codeblocks\mingw\bin\..\lib\gcc\mingw32\4.4.1\..\..\..\crt2.o:crt1.c|| undefined reference to `SetUnhandledExceptionFilter#4'|
d:\codeblocks\mingw\bin\..\lib\gcc\mingw32\4.4.1\..\..\..\crt2.o:crt1.c|| undefined reference to `ExitProcess#4'|
d:\codeblocks\mingw\bin\..\lib\gcc\mingw32\4.4.1\crtbegin.o:cygming-crtbegin.c|| undefined reference to `GetModuleHandleA#4'|
d:\codeblocks\mingw\bin\..\lib\gcc\mingw32\4.4.1\crtbegin.o:cygming-crtbegin.c|| undefined reference to `GetProcAddress#8'|
d:\codeblocks\mingw\bin\..\lib\gcc\mingw32\4.4.1\..\..\..\libmingw32.a(pseudo-reloc.o):pseudo-reloc.c|| undefined reference to `VirtualQuery#12'|
d:\codeblocks\mingw\bin\..\lib\gcc\mingw32\4.4.1\..\..\..\libmingw32.a(pseudo-reloc.o):pseudo-reloc.c|| undefined reference to `VirtualProtect#16'|
d:\codeblocks\mingw\bin\..\lib\gcc\mingw32\4.4.1\..\..\..\libmingw32.a(pseudo-reloc.o):pseudo-reloc.c|| undefined reference to `VirtualProtect#16'|
d:\codeblocks\mingw\bin\..\lib\gcc\mingw32\4.4.1\libgcc.a(cygming-shared-data.o)||In function `cmshared_get_ptr_from_atom':|
c:\crossdev\b4.4.1-tdm-1\build-sjlj\mingw32\libgcc\..\..\..\..\gcc-4.4.1\libgcc\..\gcc\config\i386\cygming-shared-data.c|91|undefined reference to `GetAtomNameA#12'|
d:\codeblocks\mingw\bin\..\lib\gcc\mingw32\4.4.1\libgcc.a(cygming-shared-data.o)||In function `_cmshared_create_or_grab':|
c:\crossdev\b4.4.1-tdm-1\build-sjlj\mingw32\libgcc\..\..\..\..\gcc-4.4.1\libgcc\..\gcc\config\i386\cygming-shared-data.c|140|undefined reference to `FindAtomA#4'|
d:\codeblocks\mingw\bin\..\lib\gcc\mingw32\4.4.1\libgcc.a(cygming-shared-data.o)||In function `cmshared_add_ptr_as_atom':|
c:\crossdev\b4.4.1-tdm-1\build-sjlj\mingw32\libgcc\..\..\..\..\gcc-4.4.1\libgcc\..\gcc\config\i386\cygming-shared-data.c|118|undefined reference to `AddAtomA#4'|
||=== Build finished: 10 errors, 0 warnings ===|
I can paste the build log here as well if need be. Could anybody let me know what I need to do to fix this? Or even better, how to troubleshoot it myself so I can fix future problems like this myself?
[EDIT]: Okay so I managed to solve this. While playing around in Project->Build Options->Linker Settings->Link Libraries I noticed that if I add something like kernel32.lib I get an error from ld saying "cannot find -lkernel32.lib". So I searched around till I found the libraries I required and I added absolute paths to them, and now my project compiles flawlessly! :D However this was kind of strange since I did add my library search path and it didn't seem to help. So I went over it again to see why it didn't seem to work and it seems like one has to add a trailing backslash to the path name, like if the libraries are located in "C:\SDK\Lib\" one has to add the trailing slash for Code::BLocks to search it correctly. Once that was fixed I changed the option in Linker Settings->Link Libraries to kernel32 as was recommended and now it works without a hitch.
Thanks to everybody who helped me out with this!!! ^_^
I got the same errors when I place a library into system32 and try to link to it with LIBS += L$$quote(C:/Windows/System32).
When I placed it into a normal directory and set the relative path to it, everything works alright.
Looks like you are not linking with Kernel32.lib Check your project properties.
I'd the same problem and after some time I figured out how to do it, the problem relies on the linker, it's trying to link some windows functions into the exe file, and the compiler cannot find those, this is how I solved this:
Install the Windows 7 SDK, you can download it from here: http://www.microsoft.com/download/en/details.aspx?displaylang=en&id=3138
Add the library path to the search libraries on CodeBlocks
Project / Build Options (select The project if you want to link it to all the configurations you have) / Search Directories / Linker / Add and place the SDK Library path, which should be at: C:\Program Files\Microsoft SDKs\Windows\v7.1\Lib
Add the references to the windows libraries on the linker settings:
Project / Build Options / Linker Settings / Add (at Link Libraries) and add kernel32, user32 and gdi32 (this last one should be added if you're using windows gui).
That's it... now rebuild your project and it should work flawlessly.
Note: You need to add this settings to the exe project, if you're creating libraries the error will be shown when try to link the exe file.
I have installed OpenCV C libraries from package management into my Kubuntu 9.10. I wrote a small image modification program that uses those libraries, but I can't compile it, because every time I try to compile I get the error message that the functions from OpenCV library cannot be found.
Those header files are in folder /usr/include/opencv/ but still the compiler says that functions cannot be found.
I have tried to include them by basic
#include <highgui.h>
#include <cv.h>
commands in netbeans (and I have set netbeans to search header files from that specific directory) and also tried to include them with full path.
#include </usr/include/opencv/highgui.h>
#include </usr/include/opencv/cv.h>
Also I tried to compile my source file in console (with full path includes), but again I got the same error message that the functions from those libraries cannot be found.
Edit:
The error message that I get is
undefined reference to 'function_name'
I get that error for every function that I try to use from that opencv library.
Any idea how to fix that problem?
Edit II:
in case someone else is using netbeans 6.7.1 and has the same problem, here's the solution copy pasted from another website:
Here I presume that you have succesfully installed the opencv library either manually or via package management.
Open Netbeans then do the following:
Select Tools -> Options -> C/C++ -> Code Assistance -> add Include Directories
(For me, /usr/local/include/opencv)
Create new project, then:
Right click on Project Name -> Properties ->Build -> C/C++ Compiler -> Include Directories
Include Directories : /usr/local/include/opencv
Right click on Project Name -> Properties -> Build -> linker ->
Include Directories : /usr/local/include/opencv
Additional Options : -I/usr/local/include/opencv -L/usr/local/lib -lcxcore -lcv -lhighgui -lcvaux -lml
Then writing program and compile!!
"undefined reference to" is a linker error. You forgot to link your application against the OpenCV libraries. Make sure you link against cv and highgui (-lcv -lhighgui) or use the pkg-config call that Tobu provided. I'd also second the request for more detailed error messages.
OpenCV uses pkg-config, the standard way to locate libraries and headers on unix. You can run (untested):
make CFLAGS="$(pkg-config --cflags --libs opencv)" your-program