I am writing a program that is meant to be extended by some function definitions. One of the way of compiling the program is to create a single executable linking your code to the main code. The problem is: one of the function to define is optional and I need to test for that.
On Linux, here is what I am doing:
Compile the program with the "-rdynamic" option to g++ or "--export-dynamic" option to ld. Then, Use ldsym like this:
fct_type myfct = (fct_type)dlsym(RTLD_DEFAULT, "fct");
If the function "fct" exists in the program, this will return its address, otherwise, it returns NULL.
Now, on Windows, I am supposed to be able to do so:
dll_handle = GetModuleHandle(0);
fct_type myfct = (fct_type)GetProcAddress(dll_handle, "fct");
But there is no "-rdynamic" or "--export-dynamic" option to g++ on MinGW32! So this doesn't work. Does anybody knows what to do on windows with MinGW32 ?
Ok, so in the end, I will answer my own question ...
You have to link with the flag -Wl,--export-all-symbols and it works ...
Related
I have used the C code from the following verbatim: https://wiki.tcl-lang.org/page/Hello+World+as+a+C+extension
/*
* hello.c -- A minimal Tcl C extension.
*/
#include <tcl.h>
static int
Hello_Cmd(ClientData cdata, Tcl_Interp *interp, int objc, Tcl_Obj *const objv[])
{
Tcl_SetObjResult(interp, Tcl_NewStringObj("Hello, World!", -1));
return TCL_OK;
}
/*
* Hello_Init -- Called when Tcl loads your extension.
*/
int DLLEXPORT
Hello_Init(Tcl_Interp *interp)
{
if (Tcl_InitStubs(interp, TCL_VERSION, 0) == NULL) {
return TCL_ERROR;
}
/* changed this to check for an error - GPS */
if (Tcl_PkgProvide(interp, "Hello", "1.0") == TCL_ERROR) {
return TCL_ERROR;
}
Tcl_CreateObjCommand(interp, "hello", Hello_Cmd, NULL, NULL);
return TCL_OK;
}
My command for compiling is nearly verbatim except for the last character, indicating Tcl version 8.6 rather than 8.4, and it compiles without error:
gcc -shared -o hello.dll -DUSE_TCL_STUBS -I$TCLINC -L$TCLLIB -ltclstub86
Then I created the following Tcl program:
load hello.dll Hello
puts "got here"
But when running it with tclsh get the following error:
cannot find symbol "Hello_Init"
while executing
"load ./hello.dll Hello"
(file "hello.tcl" line 1)
So I am essentially following a couple of suggestions from Donal Fellows answer here: cannot find symbol "Embeddedrcall_Init" The OP there however commented that, like me, the suggestion(s) hadn't resolved their issue. One thing that I didn't try from that answer was "You should have an exported (extern "C") function symbol in your library" -- could that be the difference maker? Shouldn't it have been in the example all along then?
At the suggestion of somebody on comp.lang.tcl I found "DLL Export Viewer" but when I run it against the DLL it reports 0 functions found :( What am I doing wrong?
Could it be an issue with MinGW/gcc on Windows, and I need to bite the bullet and do this with Visual Studio? That's overkill I'd like to avoid if possible.
The core of the problem is that your function Hello_Init is not ending up in the global symbol table exported by the resulting DLL. (Some linkers would put such things in as _Hello_Init instead of Hello_Init; Tcl adapts to them transparently.) The symbol must be there for Tcl's load command to work: without it, there's simply no consistent way to tell your extension code what the Tcl_Interp context handle is (which allows it to make commands, variables, etc.)
(If you'd been working with C++, one of the possible problem is a missing extern "C" whose actual meaning is to turn off name mangling. That's probably not the problem here.)
Since you are on Windows — going by the symbols in your DLL, such as EnterCriticalSection and GetLastError — the problem is probably linked to exactly how you are linking. I'm guessing that Tcl is defining your function to have __declspec(dllexport) (assuming you've not defined STATIC_BUILD, which absolutely should not be used when building a DLL) and yet that's not getting respected. Assuming you're using a modern-enough version of GCC… which you probably are.
I'm also going through the process of how to build tcl extensions in C and had exactly the same problem when working though this same example using tcl 8.6.
i.e. I was compiling using MinGW GCC (64-bit), and used the following:
gcc -shared -o hello.dll -DUSE_TCL_STUBS "-IC:\\ActiveTcl\\include" "-LC:\\ActiveTcl\\lib" -ltclstub86
And like the OP I got no compile error, but when loading the dll at a tclsh prompt tcl complained :
'cannot find symbol "Hello_Init"'
I can't say that I understand, but I was able to find a solution that works thanks to some trial and error, and some information on the tcl wiki here
https://wiki.tcl-lang.org/page/Building+Tcl+DLL%27s+for+Windows
In my case I had to adjust the compiler statement to the following
gcc -shared -o hello.dll hello.c "-IC:\\ActiveTcl\\include" "-LC:\\ActiveTcl\\bin" -ltcl86t
Obviously those file paths are specific to my system, but basically
I had to add an explicit reference to the .c file
I had to include the tcl86t dll library from the tcl bin directory
I had to remove the -DUSE_TCL_STUBS flag ( meaning that the references -LC:\\ActiveTcl\\lib and -ltclstub86 could also be removed)
(attempting to use the -DUSE_TCL_STUBS flag caused the compiler to complain with C:\ActiveTcl\lib/tclstub86.lib: error adding symbols: File format not recognized )
This successfully compiled a dll that I could load, and then call the hello function to print my 'Hello World' message.
Something else I stumbled over, and which wasn't immediately obvious:
reading https://www.tcl.tk/man/tcl8.6/TclCmd/load.htm, tcl epxects to find an 'init' function based on a certain naming convention.
if the C extension does not define a package name then the name of that init function will be derived from the dll filename.
This caused a few problems for me (when compiling via Eclipse IDE), as the dll name was being automatically determined from the eclipse projet name.
For example, if I recompile the same example, but call the .dll something else, eg.
gcc -shared -o helloWorldExenstion.dll hello.c "-IC:\\ActiveTcl\\include" "-LC:\\ActiveTcl\\bin" -ltcl86t
Then at tclsh prompt:
% load helloWorldExtension
cannot find symbol "Helloworldextension_Init"
I'm handed a variable CC which contains the executable that refers to the compiler on the system. I need to take this executable and eventually call it with some compiler-specific link arguments.
My current approach seems very fragile:
def guess_compiler(cc):
out = subprocess.check_output((cc, '--version'))
for possible in (b'clang', b'gcc'):
if possible in out:
return possible.decode()
else:
# Fall back to gcc
return 'gcc'
From this I'm using a mapping of the specific linker arguments I care about:
return {
'clang': '-Wl,-undefined,dynamic_lookup',
'gcc': '-Wl,--unresolved-symbols=ignore-all',
}[cc]
In short, I'm hoping there's a better (less fragile) way to do what I've accomplished.
For those looking for why I want something like this, it's mostly for portability in a project I'm implementing.
I'd rather call a compiler with some dummy code and these flags passed in. This is also what tools like Autotools and CMake do.
The problem with your current approach is that text string you see in --version output can actually be arbitrary. For instance, when clang wasn't that popular, FreeBSD's cc --version have been giving
GCC 4.2.1 -compatible Clang bla bla
So, just call the compiler with each flag you are interested in and then look at exit code.
An example of how to do this:
for flag in FLAGS:
try:
subprocess.check_call((cc, 'test.c', flag), cwd=tmpdir)
return flag
except subprocess.CalledProcessError:
pass
else:
# wellp, none of them worked, fall back to gcc and they'll get a
# hopefully reasonable error message
return FLAG_GCC
I am trying to write a game similar to code hunt (https://www.codehunt.com/about.aspx)
So How this will work, is the player can modify a .cpp file, which will be compiled and and ran within the code, but I have problems about how to compile it.
I'am using the latest g++ compiler, and this is how I try to achive this:
void Builder::build(const char* file){
std::string s = "g++ ";
s += file;
s += " -o test.o";
system(s.c_str());
}
Where we get the .cpp file's name, and this code piece is supposed to build test.o
This is just a test now, it might get more complex, I just wanted to test if the compiler will work within the code, but I get this error message when I try to run this:
c:/mingw/bin/../lib/gcc/mingw32/4.9.3/../../../libmingw32.a(main.o):(.text.startup+0xa7):undefined reference to 'WinMain#16'
collect2.exe: error: ld returned 1 exit status
PATH for mingw is set correctly, I checked.
I am using Windows 8.1, g++ 4.9.3 and Code::Blocks.
In Windows execution doesn't normally start at the main functions, it starts at the WinMain function, which takes some Windows-specific arguments. You should read e.g. this WinMain reference for more information.
That some programs still seems to start at a main function is because there is an object file linked with the program that contains the WinMain function which calls your main function.
OK I'm an idiot, so the problem was that I was trying to build a file whitout a main function. This was deliberat design choice at first, but...well yeah. Sorry about that.
Thank you Joachim Pileborg for leading me to it.
Is there a way to change the specs file so that it will pass -march=native if nothing is specified in command line?
Related things in the default specs file is:
*cc1:
%(cc1_cpu)
*cc1_cpu:
%{march=native:%>march=native %:local_cpu_detect(arch) %{!mtune=*:%>mtune=native %:local_cpu_detect(tune)}} %{mtune=native:%>mtune=native %:local_cpu_detect(tune)}
I am not sure how specs works. Simply specifying -march=native before or after %(cc1_cpu) doesn't work. However, this line does take effect because GCC will report error if I put -something_wierd instead of -march=native.
Another thing I noticed is if I put %{march=i386:-something_wierd} before %(cc1_cpu), gcc reports error so looks like -march=i386 is always passed in if nothing is specified, so is there a way to distinguish between nothing specified and -march=i386 in specs file?
BTW, what does %> do? Seems like it is not specified in the documentation.
I am using MinGW's gcc-4.6.2.
Referring to your last question: The gcc 4.6.1 sources (gcc/gcc.c) contain the following comment on %>:
%>S Similar to "%<S", but keep it in the GCC command line.
For the sake of completeness following the comment for %< form the same file:
%<S remove all occurrences of -S from the command line.
Note - this command is position dependent. % commands in the
spec string before this one will see -S, % commands in the
spec string after this one will not.
To answer the first question in short: yes, but ....
... the only generic solution I found has the significant drawback that the -march option will be ignored, so every build is done as if -march=native had been specified. Anyhow there is a workaround to that.
1 The solution (without workaround)
Create a specs-file called let's say specs.nativealways containing:
*cc1_cpu:
%<march=* -march=native %>march=native %:local_cpu_detect(arch) %{!mtune=*:%>mtune=native %:local_cpu_detect(tune)} %{mtune=native:%>mtune=native %:local_cpu_detect(tune)}
When using the specs-file (for example by invoking gcc with the option -specs=specs.nativealways) the build will be done as if -march=native was specified (with the mentioned drawback that any occurrence of option -march=<arch> would have simply been ignored).
2 The workaround
To still by able to override the newly configured default behavior one can use a modified version of the specs-file described above, introducing a new option called -myarch using the same syntax as -march (except for -myarch=native, which won't work, which does not metter as native now is the default).
The modfied specs-file looks like this:
*cc1_cpu:
%<march=* %{myarch=*:%<myarch* -march=%* ; :-march=native %>march=native %:local_cpu_detect(arch) %{!mtune=*:%>mtune=native %:local_cpu_detect(tune)}} %{mtune=native:%>mtune=native %:local_cpu_detect(tune)}
PS: This has been tested with with gcc 4.6.2 on Linux, but should work on MinGW.
While not a direct answer to your question, you can reach a very similar effect by defining CFLAGS and CXXFLAGS in your shell's initialization file. 99% of the Makefiles are sufficiently standard to pick up the environment values and pass the flags to gcc.
*cc1_cpu:
+ %{!march*:-march=native}
In this episode of "let's be stupid", we have the following problem: a C++ library has been wrapped with a layer of code that exports its functionality in a way that allows it to be called from C. This results in a separate library that must be linked (along with the original C++ library and some object files specific to the program) into a C program to produce the desired result.
The tricky part is that this is being done in the context of a rigid build system that was built in-house and consists of literally dozens of include makefiles. This system has a separate step for the linking of libraries and object files into the final executable but it insists on using gcc for this step instead of g++ because the program source files all have a .c extension, so the result is a profusion of undefined symbols. If the command line is manually pasted at a prompt and g++ is substituted for gcc, then everything works fine.
There is a well-known (to this build system) make variable that allows flags to be passed to the linking step, and it would be nice if there were some incantation that could be added to this variable that would force gcc to act like g++ (since both are just driver programs).
I have spent quality time with the gcc documentation searching for something that would do this but haven't found anything that looks right, does anybody have suggestions?
Considering such a terrible build system write a wrapper around gcc that exec's gcc or g++ dependent upon the arguments. Replace /usr/bin/gcc with this script, or modify your PATH to use this script in preference to the real binary.
#!/bin/sh
if [ "$1" == "wibble wobble" ]
then
exec /usr/bin/gcc-4.5 $*
else
exec /usr/bin/g++-4.5 $*
fi
The problem is that C linkage produces object files with C name mangling, and that C++ linkage produces object files with C++ name mangling.
Your best bet is to use
extern "C"
before declarations in your C++ builds, and no prefix on your C builds.
You can detect C++ using
#if __cplusplus
Many thanks to bmargulies for his comment on the original question. By comparing the output of running the link line with both gcc and g++ using the -v option and doing a bit of experimenting, I was able to determine that "-lstdc++" was the magic ingredient to add to my linking flags (in the appropriate order relative to other libraries) in order to avoid the problem of undefined symbols.
For those of you who wish to play "let's be stupid" at home, I should note that I have avoided any use of static initialization in the C++ code (as is generally wise), so I wasn't forced to compile the translation unit containing the main() function with g++ as indicated in item 32.1 of FAQ-Lite (http://www.parashift.com/c++-faq-lite/mixing-c-and-cpp.html).