Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Don't link with libwinmm unless WinMM is enabled | Chris Robinson | 2014-04-19 | 1 | -11/+8 |
| | | | | | We don't use timeGetTime for timing anymore, so winmm only needs to be linked when the winmm backend is enabled. | ||||
* | Use a find module to look for DirectSound | Chris Robinson | 2014-04-19 | 2 | -28/+60 |
| | |||||
* | Use find cmake modules for more backends | Chris Robinson | 2014-04-19 | 7 | -79/+259 |
| | |||||
* | Add a GCC-specific STATIC_UPCAST macro that checks the object type | Chris Robinson | 2014-04-19 | 4 | -4/+11 |
| | | | | | The check is compile time, and is functionally identical to the old/alternate version. | ||||
* | Use C11 alignas when available | Chris Robinson | 2014-04-19 | 7 | -17/+55 |
| | |||||
* | Dont bother trying _alloca | Chris Robinson | 2014-04-19 | 1 | -6/+1 |
| | | | | | We prefer C99 variable-length arrays, and other systems (including MSVC9) don't have an issue with alloca. | ||||
* | Set -D_GNU_SOURCE=1 -D_POSIX_C_SOURCE=200809L -D_XOPEN_SOURCE=700 earlier | Chris Robinson | 2014-04-19 | 1 | -4/+5 |
| | |||||
* | Use FindALSA.cmake to look for ALSA | Chris Robinson | 2014-04-19 | 1 | -15/+14 |
| | | | | Also, only provide the ALSA_BACKEND_ALSA option if it's available | ||||
* | Only define SSE, Neon, and FluidSynth options if they're available | Chris Robinson | 2014-04-18 | 1 | -17/+15 |
| | |||||
* | Simplify some error checking | Chris Robinson | 2014-04-18 | 2 | -21/+5 |
| | |||||
* | Check pthread_mutex_unlock for errors | Chris Robinson | 2014-04-18 | 1 | -3/+4 |
| | |||||
* | Remove unnecessary includes | Chris Robinson | 2014-04-18 | 1 | -4/+0 |
| | |||||
* | Alter the al_fopen macro to not directly take parameters | Chris Robinson | 2014-04-18 | 1 | -1/+1 |
| | |||||
* | Rename AL_ONCE_INIT to AL_ONCE_FLAG_INIT | Chris Robinson | 2014-04-18 | 2 | -3/+3 |
| | |||||
* | Check TlsSetValue for error | Chris Robinson | 2014-04-17 | 1 | -1/+2 |
| | |||||
* | Rename althread_once to be more C11-like | Chris Robinson | 2014-04-17 | 5 | -24/+30 |
| | |||||
* | Handle the lib name as UTF-8 | Chris Robinson | 2014-04-17 | 1 | -14/+28 |
| | |||||
* | Use the thread ID for althrd_t on Windows | Chris Robinson | 2014-04-17 | 2 | -18/+34 |
| | |||||
* | Rename SetThreadName to althrd_setname | Chris Robinson | 2014-04-17 | 13 | -27/+25 |
| | |||||
* | Check explicitly for pthread_mutexattr_setkind_np before use | Chris Robinson | 2014-04-17 | 3 | -1/+7 |
| | |||||
* | Fix SetThreadName for 64-bit MSVC builds | Chris Robinson | 2014-04-17 | 1 | -1/+3 |
| | |||||
* | Don't inline al_nssleep | Chris Robinson | 2014-04-17 | 2 | -13/+13 |
| | |||||
* | Make and use a C11-like altimespec_get wrapper function | Chris Robinson | 2014-04-17 | 6 | -98/+110 |
| | |||||
* | Fix a signed-unsigned comparison warning and clarify a comment | Chris Robinson | 2014-04-17 | 1 | -5/+5 |
| | |||||
* | Keep TlsDestructors within threads.c | Chris Robinson | 2014-04-17 | 4 | -5/+9 |
| | | | | | This basically makes the threads implementation self-contained in threads.c and threads.h, except for the UIntMap/RWLock implementations. | ||||
* | Don't use DllMain to call altss destructors | Chris Robinson | 2014-04-17 | 3 | -20/+43 |
| | |||||
* | Rename althread_key_ wrappers to altss_ and move it to threads.h/c | Chris Robinson | 2014-04-17 | 5 | -51/+82 |
| | |||||
* | Fix some almtx_ return values | Chris Robinson | 2014-04-16 | 2 | -13/+36 |
| | |||||
* | Remove almtx_normal and almtx_errorcheck | Chris Robinson | 2014-04-16 | 2 | -17/+8 |
| | |||||
* | Make sure the duration given to althrd_sleep is valid | Chris Robinson | 2014-04-16 | 1 | -2/+8 |
| | |||||
* | Bump _WIN32_WINNT to 0x0502 (XP SP2) for GetThreadId | Chris Robinson | 2014-04-16 | 1 | -4/+4 |
| | |||||
* | Fix althrd_sleep return value | Chris Robinson | 2014-04-16 | 1 | -3/+9 |
| | |||||
* | Avoid using a Sleep() wrapper | Chris Robinson | 2014-04-16 | 9 | -21/+24 |
| | |||||
* | Use althrd_yield instead of alsched_yield | Chris Robinson | 2014-04-16 | 2 | -9/+2 |
| | |||||
* | Remove the old thread wrappers for the new ones | Chris Robinson | 2014-04-16 | 10 | -221/+97 |
| | |||||
* | Move the threads and mutex wrappers to threads.c | Chris Robinson | 2014-04-16 | 2 | -256/+255 |
| | |||||
* | Fix Windows' almtx_timedlock | Chris Robinson | 2014-04-16 | 1 | -6/+6 |
| | |||||
* | Implement a C11-like thread wrapper and use it in mmdevapi and pulseaudio | Chris Robinson | 2014-04-16 | 4 | -34/+250 |
| | |||||
* | Use a C11-like mutex wrapper instead of CRITICAL_SECTIONs | Chris Robinson | 2014-04-16 | 7 | -55/+215 |
| | |||||
* | Avoid memory leak if preset ID lookup fails | Chris Robinson | 2014-04-14 | 1 | -0/+1 |
| | |||||
* | Ensure VECTOR_ITER_BEGIN gives a pointer-to-type instead of an array-of-type | Chris Robinson | 2014-04-14 | 1 | -1/+1 |
| | |||||
* | Use a helper to detect a path slash on Windows | Chris Robinson | 2014-04-13 | 1 | -2/+5 |
| | |||||
* | Use VECTOR_INSERT to copy and append strings | Chris Robinson | 2014-04-13 | 1 | -12/+8 |
| | |||||
* | Add a GCC-specific VECTOR_INSERT | Chris Robinson | 2014-04-13 | 1 | -0/+12 |
| | | | | | | This version is capable of doing a compile-time type checks, to ensure the objects being inserted are compatible with the vector data type. It should otherwise be functionally identical. | ||||
* | Pass in the vector insertion point as a pointer | Chris Robinson | 2014-04-13 | 2 | -3/+5 |
| | |||||
* | Avoid referencing the start iterator twice for vector insertion | Chris Robinson | 2014-04-13 | 2 | -4/+5 |
| | |||||
* | Insert all new effect slots into the active effect slots at once | Chris Robinson | 2014-04-10 | 1 | -26/+27 |
| | |||||
* | Add a VECTOR_INSERT method to insert a range of elements at once | Chris Robinson | 2014-04-09 | 3 | -10/+29 |
| | |||||
* | Fix a NULL deref when cloning a 0-sized vector | Chris Robinson | 2014-04-09 | 1 | -6/+11 |
| | |||||
* | Silence an MSVC warning about -1u | Chris Robinson | 2014-04-07 | 1 | -4/+4 |
| |