Producing small Windows binaries - windows

When developing and deploying native Windows applications, I often need to install a runtime before being able to run my binary, or statically link the library with my binary. For instance, after building a "Win32 Console" project with Visual Studio 2008, attempting to run the program on a fresh Windows 7 image results in:
The application has failed to start because its side-by-side configuration is incorrect. Please see the application event log or use the command-line sxstrace.exe tool for more detail.
Issues like this one have been brought up in other posts on StackOverflow.
How does one develop applications that don't require runtimes that aren't already on the target OS (i.e. don't require installing redistributable packages or private/shared side-by-side assemblies)? How does one avoid using msvc[mpr]90.dll and just use the Windows API in \windows\system32*.{dll,sys}?
I'm thinking along the lines of the code that comes out of the demoscene, but that's frequently not made available.

Others have already responded with respect to linking CRT statically. If you also want a small binary at the same time, then your best bet is forego CRT entirely, and use only Win32 API functions as much as possible. You'll still get some CRT code, most notably related to startup (i.e. that which calls main) and shutdown (atexit handling etc), but otherwise the linker won't link CRT functions that you do not use.
You can avoid linking CRT altogether by using /Zl compiler switch. This means that main will no longer work, however - you'll need to define WinMain (name doesn't matter, but signature must match, and it must be __stdcall), and you will have to specify the name of your WinMain-like function as an entry point via linker /entry: switch. This will save you ~30Kb of CRT code (tested on a .cpp with an empty main).
If you go the latter route, you might also have to deal with issue of compiler intrinsics. There are some functions that are nominally defined by the CRT (and declared in its headers), but which are treated specially by the compiler, so that it inserts optimized assembly instructions at the point of the call where possible - examples are memset, strlen, and a good chunk of functions in <math.h>; a complete list can be found here. Since you don't have CRT, if you need these functions, or could avoid it but prefer the intrinsic because of improved performance (hard to do better than memset, for example), then you have to declare them yourself, and use #pragma intrinsic. E.g.:
// Contains macros and typedef only, so safe to include without CRT.
// We need it here for size_t.
#include <stddef.h>
extern "C"
{
int abs(int);
void* memset(void*, int, size_t);
}
#pragma intrinsic(abs, memset)
int __stdcall main(void*, void*, char*, int)
{
char tmp[10];
memset(tmp, abs(-123), 10);
return 0;
}
The above can be compiled with:
cl /c /Zl foo.cpp
link /entry:main foo.obj

Link the CRT statically via the /MT switch (and likewise MFC, if you're using it).
Static linking limits what you can do with DLLs somewhat, but for simple executables it works like a charm. (And if you're shipping DLLs, you can always ship private assemblies anyway.)

Use the static CRT. This doesn't create a dependency on msvc*.dll. The CRT is linked directly into your program. This doesn't create dependencies, but does increase the size of your executable.
More info on different CRT options here.

Statically link the runtime. MS Visual C++ has option /MT for that (default is /MD)

I think one way to do this is to just not use Visual Studio and instead rely on the command line SDK tools. (You can alternatively figure out how to config VS to do what you want, but that seems harder.) E.g.:
cl /c app.cpp
link app.obj ws2_32.lib

Related

Hiding the console from D while linked with C-version SDL2

Basically, I get a console window I don't want to see while writing SDL-windowed applications.
In C, the answer is simple: change your linker SubSystem to Windows; then SDL does the rest. It has a macro that redefines your main to SDL_main, which it calls from inside its own hidden WinMain function. Just make sure you take arguments in your main function and all goes smoothly.
In D, I'm having a little more trouble with it. Upon simply changing the subsystem by passing -L/SUBSYSTEM:WINDOWS to the linker, it informs me that I have declared main, SDL has declared WinMain, and it doesn't know how to deal. I thought changing the signature from void main() to extern(C) int SDL_main(int argc, char*[] argv) would solve the problem, but then the linker says it can't locate any of D's runtime symbols. Might have something to do with the entry point being written in C?
On the other side, my problem might be with how I have set up Derelict. I am working from this repository, which is a live-updated amalgam of all the different Derelict repositories on GitHub. In my VisualD solution, I have three relevant projects: a home-brewed Derelict project containing the source to SDL2 and the Utilities library, and the C-API SDL2 and SDL2main projects as found here. I compile and link to all three resulting libraries -- Derelict and SDL2main are static, SDL2 is dynamic. I am less inclined to say this setup is to blame, because it works just fine save the undying console window.
In the command line passed to DMD, add the linker option -L/SUBSYSTEM:WINDOWS
In case you use dub (which i recommend) and gdc, insert the following into your dub.json file:
"dflags": ["-Wl,--subsystem,windows"]

MS C/C++ Linkeroption /MT but still unresolved _allmul() when linking with different IDE

I am writing lib files in VS that have to be imported into CVI.
Recently a linker problem occured.
It says that _allmul() is an undefined symbol.
_allmul() and freinds are implemented as calls to the CRT library functions to handle various 64bit operations.
The lib file that i write is static because i want all the code to live inside it.
Linking with VS is no problem and all unittests pass.
Linking with the CVI-IDE leaves the CVI Linkter complaining about an unresolved _allmul().
I thougth that setting the /MT switch in VS is enough to make the CRT link statically, this seems to be wrong.
Why is my assumtion wrong?
How can i link the CRT calls statically?
Edit:
Here is a short demo of a project that i can compile in VS2010 but that i can not link to in CVI
The following is compiled as a LIB project:
HEADER
void print( unsigned A, unsigned B );
CODE
#include "MyprintInterface.h"
#include <stdio.h>
void print( unsigned A, unsigned B ){
long long copyA = ( long long ) A;
long long copyB = ( long long ) B;
printf_s( " %lli * %lli = %lli ", copyA, copyB, copyA * copyB );
// copyA * copyB -> this invokes allmul
// printf_s this is a ms specific function
}
This is a little late, but maybe it will help someone else.
I had very similar experience when using the gSoap toolkit with LabWindows/CVI. The toolkit generates ANSI C bindings used in creating web service clients or servers. However it sometimes generated versions of MS functions (some deprecated) that were prototyped slightly different than those in CVI. Solving this was kind of tedious, but generally followed these steps:
1) install appropriate Microsoft SDK onto your dev box. Here is one option for Microsoft SDKs. (There are others, Google it)
2) search MSDN website for the unresolved symbol to determine what .lib/.h support it. (MSDN list of variations of sprintf, for example)
3) include any required .lib/.h files.
4) make sure msvcr80.dll or msvcr100.dll (or other as appropriate) are visible to your executable. (i.e. place either in the executable run-time directory, or in the system directory.)
In addition to this I had to bring some of the #defines from the header files supporting file write/read functions, just
for example:
#define _O_TEXT 0x4000 /* file mode is text (translated) */
#define _O_BINARY 0x8000 /* file mode is binary (untranslated) */
And: (Requires Microsoft Platform SDK be installed
c:\program files\microsoft platform sdk\include\crt (copied to project dir and included in project))
fcntl.h (definition for file mode etc., _O_BINARY)
float.h (definition for isnan etc.)
io.h (definition for setmode etc.)
Also, there is an example CVI project with source code illustrating what I have described above, (including using crt) here.
[EDIT]
Regarding COFF compatible .libs when using third party libraries with respect to LabWindows/CVI - Read the posts here.
[EDIT 2]
Once you install the Windows SDK, the headers for everything in msvcrxxx.dll will be on your system. But for finding .h dependancies of a specific API call I typically have good luck just searching for it on Google: i.e. for sprintf_s et. al., the first link returned was this. It includes notes on how to use sprintf_s including source code.
As for allmul issues, I have not run into this before, but this guy did.

Microsoft visual C++ backwards compatibility

Consider a C++ API defined as a series of __options(declexport/import) classes.
Further, assume that the caller is never permitted to call the ordinary operator new(size_t) on these classes. Either a static factory method does the new-ing or there is a class-specific operator new. And ditto marks on the delete size as needed (frequently just a virtual destructor).
Now, if you compile and link a DLL and an IMPLIB of with the tools from VS2010, can you hand that implib and DLL to a user of VS2005 and expect it to work?
MFC is not involved here at all.
I'd be particularly grateful to any reference to any relatively formal Microsoft statement on the subject.
So long as the name mangling on the C++ API is identical (they are), and does not use STL-type specific parameters, such as basic_string or std::map, whose implementation may have changed between releases of the compiler (and they have), then it should just work.
Of course, you'll want to make sure you either compiled your DLL using /MT mode (static linked runtimes), or include the redistributables for VS2010 runtimes with your supplied libraries and link targets.
EDIT: Expanding on "don't pass in types that have version-specific implementations". A partial list is most easily found by looking at the output of the exports of MSVC100P.DLL.
cd %VS100COMNTOOLS%\..\VC\redist\x86\Microsoft.VC100.CRT
DUMPBIN /exports MSVCP100.DLL
The next issue will be header-only implementations of things like map or set which have changed under the hood between versions of the compiler.
This is why it's highly recommended that only scalar types be passed across boundaries between memory arenas. And thus, simple tests will pass, and be reliable.
You have not mentioned if you have used MFC to create the DLL's .If you have, regular DLL's should work , but I dont think extension shall work as the latter links to the MFC dlls .I am including links for your reference.
http://www.codeguru.com/cpp/cpp/cpp_mfc/tutorials/article.php/c4017
http://www.experts-exchange.com/Programming/System/Windows__Programming/MFC/Q_20385543.html
http://msdn.microsoft.com/en-us/library/26h8x9sy%28v=VS.100%29.aspx
EDIT
If its a normal DLL, there should not be any problem.Also depends on the linkage type.

DLL linker issue

I am compiling a DLL twice (once for x86, once for x64) and I have set /ENTRY to "DllMain". I am using the /MT runtime library option to statically link against the runtime library. This all work fine when doing the x86 build, but the x64 build fails with this:
error LNK2019: unresolved external symbol main referenced in function __tmainCRTStartup
{project directory}\LIBCMT.lib(crt0.obj)
Why does this work for the x86 build and not the x64 build? Is there something I am missing here?
Not a direct answer but it may be strictly related: as said in the comment, you should avoid changing the entrypoint in that way: normally the real entrypoint is taken by a "fake" DllMain provided by the CRT to initialize its internal data structures (as explained here), so you're bypassing it. Probably the size reduction is due to CRT init code being removed.
Your dll is working with a non-initialized CRT, which is very bad. You should leave the default entrypoint, which, incidentally, should solve your problem.
By the way, notice that actually you could make a dll without the CRT (and it would become really small), but you shouldn't use the CRT at all, without even linking against it (/NODEFAULTLIB switch). This means that you could just use libraries you explicitly link against (e.g. the Windows API), but I suspect you would lose several C++ features (I think at least exceptions and RTTI).
This could be a silly question, but are you sure you're linking as a DLL in the x64 case (ie. specifying the /DLL switch) - since the complaint is about main, I wonder whether it's trying to link as a an executable?

Which headers should I not use if I don't want my program to be linked with any of msvc*.dll's?

Which headers should I not use if I don't want my program to be linked with any of msvc*.dll ?
At the moment my application uses:
kernel32
user32
shell32
msvcp90
msvcr90
I want to get rid of the bottom two files. I don't mind if I will have to rewrite certain aspects of the program.
Because I know if you code in C and then link it won't link any msvc's
I believe you have to change the way the CRT is linked into your program. I think for that you have to change the C++->Code Generation->Runtime-Library to the static version. This is for Visual Studio 2005, don't know about newer versions.
Those libraries contain the C++ runtime - heap management and other stuff hard to get rid from.
You could link the C++ statically instead - use "C++ -> Code Generation -> Runtime Library" setting. Then you will not need those .dll files. However this is not the recommended way - if a vulnerability is found in the C++ runtime you'll have to recompile and reship your program.
Static link is the right answer. A related bit of advice is to use depends.exe to see what functions your exe is actually hitting in the dependent dlls. Those dependencies might be due to explicit use on your part or due to CRT implementation that you don't explicitly invoke.

Resources