aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Update documentation in dlfcn.h, specially for RTLD_LAZYPali Rohár2019-08-292-11/+26
|
* Merge pull request #63 from pali/masterSilvio Traversaro2019-08-231-18/+20
|\ | | | | Makefile: Fix building both static and shared library at the same time
| * Makefile: Fix building both static and shared library at the same timePali Rohár2019-08-221-18/+20
|/ | | | | | | | Static library libdl.a must be build without -DSHARED flag for obvious reason. So -DSHARED cannot be in global CFLAGS. Define also two separate test targets: test.exe for shared library and test-static.exe for static library.
* Merge pull request #62 from pali/masterSilvio Traversaro2019-08-221-15/+9
|\ | | | | Reduce memory usage by 64K
| * Reduce memory usage by 64KPali Rohár2019-08-221-15/+9
|/ | | | One buffer for error message is enough.
* Merge pull request #59 from pali/masterSilvio Traversaro2019-08-213-47/+47
|\ | | | | Fix compile warnings
| * Fix more gcc warningsPali Rohár2019-07-261-6/+10
| | | | | | | | | | ISO C90 forbids mixed declarations and code ISO C++ forbids converting a string constant to 'char*'
| * Fix gcc warning: ISO C forbids return between function pointer and void *Pali Rohár2019-07-253-37/+33
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Instead of using compiler specific pragma to disable particular warning, rewrite code which cast from function pointer to data pointer according to POSIX dlopen() documentation. This also fix compile warning under MSVC. According to the ISO C standard, casting between function pointers and 'void *', as done above, produces undefined results. POSIX.1-2003 and POSIX.1-2008 accepted this state of affairs and proposed the following workaround: *(void **) (&cosine) = dlsym(handle, "cos"); This (clumsy) cast conforms with the ISO C standard and will avoid any compiler warnings.
| * Fix gcc warning: comparison between signed and unsigned integer expressionsPali Rohár2019-07-251-3/+4
| | | | | | | | | | GetTempPathA() returns DWORD (32/64bit unsigned integer) and not int (32 signed integer).
| * Do not include intrin.h filePali Rohár2019-07-251-1/+0
|/ | | | | | | | | | | | Include file intrin.h is not needed for compiling dlfcn.c as _ReturnAddress intrinsic is defined in dlfcn.c manually. Moreover _ReturnAddress is available only as an intrinsic, there is no function implementation. So even including intrin.h file does not provide function variant for _ReturnAddress. More important, include file intrin.h is not available in older Windows Driver Kit 7.1.0. So removing usage of intrin.h file makes compilation of dlfcn-win32 also under these older WDK versions.
* Merge pull request #58 from pali/masterSilvio Traversaro2019-07-251-0/+5
|\ | | | | Correctly process malloc() error in dlsym()
| * Correctly process malloc() error in dlsym()Pali Rohár2019-07-241-0/+5
|/ | | | malloc() may fail, so propagate this error to caller.
* Merge pull request #55 from pali/masterSilvio Traversaro2019-07-231-0/+7
|\ | | | | Turn off GUI error messages around LoadLibraryA also in test
| * Turn off GUI error messages around LoadLibraryA also in testPali Rohár2019-06-111-0/+7
|/ | | | Otherwise running test show GUI error messages which needs to be closed.
* Merge pull request #54 from pali/masterSilvio Traversaro2019-06-112-67/+228
|\ | | | | Correctly process and indicate errors
| * Add test for non-library filePali Rohár2019-05-231-0/+61
| |
| * Add tests for non-existent file and file with too long namePali Rohár2019-05-231-0/+50
| |
| * Test that dlerror() returns non-NULL error for failed callsPali Rohár2019-05-231-20/+37
| |
| * Correctly process and indicate error when file name is too longPali Rohár2019-05-231-47/+55
| |
| * Correctly process malloc() error in local_add()Pali Rohár2019-05-231-8/+15
| | | | | | | | malloc() may fail, so propagate this error to caller of dlopen().
| * Correctly process and indicate error in dlsym() functionPali Rohár2019-05-231-1/+10
| | | | | | | | | | Function save_err_str() checks for error by GetLastError() call. So ensure that last error is always set when error occurs.
| * Correctly process and indicate error from LoadLibraryExA() functionPali Rohár2019-05-231-16/+25
|/ | | | | | Function save_err_str() checks for error by GetLastError() call. Calling EnumProcessModules() may change or reset it. So call save_err_str() immediately after LoadLibraryExA().
* Merge pull request #52 from pali/masterSilvio Traversaro2019-05-211-1/+1
|\ | | | | Call FormatMessage() with FORMAT_MESSAGE_IGNORE_INSERTS
| * Call FormatMessage() with FORMAT_MESSAGE_IGNORE_INSERTSPali Rohár2019-05-211-1/+1
|/ | | | | | Documentation says: In particular, it is unsafe to take an arbitrary system error code returned from an API and use FORMAT_MESSAGE_FROM_SYSTEM without FORMAT_MESSAGE_IGNORE_INSERTS.
* Merge pull request #51 from pali/masterv1.2.0Silvio Traversaro2019-05-206-30/+26
|\ | | | | Load Psapi.dll at runtime, this avoids linking caveat
| * Load Psapi.dll at runtime, this avoids linking caveatPali Rohár2019-05-066-30/+26
|/
* Merge pull request #50 from pali/masterSilvio Traversaro2019-05-062-95/+134
|\ | | | | Simplify UNICODE build
| * Test also UNICODE builds on AppVeyor CIPali Rohár2019-04-251-0/+102
| |
| * Simplify implementation of save_err_str()Pali Rohár2019-04-251-26/+22
| | | | | | | | | | Check return value of FormatMessageA() function and remove copy_string() function as it is not needed.
| * Remove ifdef hack for snprintf()Pali Rohár2019-04-251-5/+5
| | | | | | | | | | | | Old version of MSVC does not support snprintf() function and sprintf_s() is not replacement for C99 snprintf(). As the only usage of snprintf() is to format void* pointer we can use sprintf() with enough long buffer.
| * Simplify code around #ifdef UNICODEPali Rohár2019-04-251-69/+10
|/ | | | | | | | | | The whole dlfcn.h API works with char* (ANSI) strings. For WINAPI UNICODE builds it is still possible to call WINAPI ANSI functions with -A suffix. E.g. LoadLibraryExA() instead of LoadLibraryEx() or FormatMessageA() instead of FormatMessage(). This simplify whole implementation when compiling with UNICODE support as there is no need to do conversion from wchar_t to char and vice-versa.
* Merge pull request #44 from pali/masterSilvio Traversaro2019-02-149-120/+335
|\ | | | | Fix resolving global symbols and implement RTLD_DEFAULT and RTLD_NEXT
| * Implement support for dlsym() with RTLD_DEFAULT and RTLD_NEXTPali Rohár2019-02-148-10/+193
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | dlsym() with RTLD_DEFAULT handle behaves in same way like with global handle returned by dlopen() with NULL file name. dlsym() with RTLD_NEXT handle search for next loaded module which provides specified symbol. "Next" means module which in EnumProcessModules() result after the module which called dlsym(). To get caller function of dlsym() use _ReturnAddress() intrinsic. To get module where is caller function use the fact that HMODULE is the same value as the module's base address. When compiling under gcc, defines _ReturnAddress() macro via gcc's builtin as it does not provide MSC's specific _ReturnAddress() intrinsic. Added tests demonstrate that both RTLD_DEFAULT and RTLD_NEXT are working as expected.
| * Fix resolving global symbols when LoadLibrary() is called after dlopen()Pali Rohár2019-02-146-112/+144
|/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Usage of first_automatic_object cache is wrong. This cache is filled by all loaded DLL files (either implicitly or explicitly with LoadLibrary() call) by EnumProcessModules() call at first usage of dlopen(). So dlsym() can resolve global symbols only if they were loaded prior to dlopen() call. Any future usage of LoadLibrary() does not include newly loaded DLLs into first_automatic_object cache. To fix this problem, first_automatic_object cache is fully removed and EnumProcessModules() call is issued directly in dlsym() call. As EnumProcessModules() returns all DLLs, included those which were loaded by dlopen() with RTLD_LOCAL, it may break RTLD_LOCAL support. To address this problem switch linked-list of all loaded DLLs with RTLD_GLOBAL to linked-list of all loaded DLLs with RTLD_LOCAL flag. And then skip modules from EnumProcessModules() which are in linked-list. Also in WinAPI all DLLs loaded by LoadLibrary() behaves like RTLD_GLOBAL. So above change is compatible with this behavior. There may be another problem. Before retrieving HMODULE for DLL filename (which is done by LoadLibrary()), it is not possible to detect if DLL was already loaded by RTLD_LOCAL or not. And after calling LoadLibrary() it is not possible to know if DLL was loaded either by dlsym() with RTLD_LOCAL or by LoadLibrary() (which is equivalent to RTLD_GLOBAL). To address this problem, compare number of loaded modules (counted by EnumProcessModules()) before and after LoadLibrary() called from dlsym(). If number does not change it means that DLL was already loaded. So based on this result either add or remove HMODULE from linked-list of RTLD_LOCAL modules. Added test demonstrate usage of: global = dlopen(NULL, RTLD_GLOBAL); /* global handle */ LoadLibrary("library.dll"); /* this provides function */ function = dlsym(global, "function"); /* resolve function from library.dll */
* Merge pull request #48 from dlfcn-win32/appveyor-mingw-take-twoSilvio Traversaro2019-02-141-5/+84
|\ | | | | Add MinGW and MinGW-w64 builds to AppVeyor (take 2)
| * Add MinGW and MinGW-w64 tests to AppVeyorSilvio Traversaro2019-02-121-5/+84
|/
* Merge pull request #41 from jddurand/masterv1.1.2Silvio Traversaro2018-03-081-0/+1
|\ | | | | #include <stdlib.h>
| * #include <stdlib.h>Jean-Damien Durand2018-01-171-0/+1
|/
* Merge pull request #40 from dlfcn-win32/document-cmake-usageSilvio Traversaro2017-09-271-0/+24
|\ | | | | Document how to use the library when using CMake
| * Mention the possibility of defining CMAKE_DL_LIBSSilvio Traversaro2017-08-241-0/+13
| |
| * Document how to use the library when using CMakeSilvio Traversaro2017-08-241-0/+11
|/
* Merge pull request #39 from dlfcn-win32/traversaro-patch-2Silvio Traversaro2017-05-041-1/+1
|\ | | | | [README] Fix AppVeyor badge
| * [README] Fix AppVeyor badgeSilvio Traversaro2017-05-041-1/+1
|/
* Merge pull request #38 from dlfcn-win32/appveyor-test-vs2017v1.1.1Silvio Traversaro2017-05-041-6/+16
|\ | | | | [appveyor] Test the library using Visual Studio 15 2017
| * [appveyor] Test the library using Visual Studio 15Silvio Traversaro2017-05-011-6/+16
| | | | | | See https://www.appveyor.com/docs/build-environment/ for the logic behind the option used.
* | Merge pull request #36 from dlfcn-win32/fix-dlerrorSilvio Traversaro2017-05-042-0/+21
|\ \ | |/ |/| Fix bug in dlerror second consecutive call
| * Fix bug in dlerror second consecutive callSilvio2017-05-012-0/+21
|/ | | | | | | | | | | | | According to the specs, a second consecutive call to dlerror should always return NULL . This was the case in dlfcn-win32 before https://github.com/dlfcn-win32/dlfcn-win32/pull/20 introduce a regression that caused dlerror to crash on the second consecutive call. In this commit the issue is fixed as suggested in https://github.com/dlfcn-win32/dlfcn-win32/issues/34 and a regression test has been added.
* Merge pull request #30 from Kamekameha/appveyor-fixSilvio Traversaro2017-04-091-16/+16
|\ | | | | Minor AppVeyor configuration cleanup
| * Minor AppVeyor configuration cleanupKamekameha2017-03-191-16/+16
|/
* [readme] Update AppVeyor badgeSilvio Traversaro2017-03-191-1/+1
|
generated by cgit v1.2.3-55-g6feb (git 2.39.1) at 2025-04-09 09:41:52 +0000