summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Use gettimeofday if clock_gettime isn't availableChris Robinson2014-05-061-5/+11
|
* Test _Alignas with 16 instead of 4Chris Robinson2014-05-051-1/+1
|
* C11's alignas is also sufficient for SSE aligningChris Robinson2014-05-051-7/+9
|
* Add a cast to silence an MSVC warningChris Robinson2014-05-041-2/+3
|
* Trace skipped info sub-chunksChris Robinson2014-05-041-0/+2
|
* Use ALuint for the amount to skipChris Robinson2014-05-041-1/+1
|
* Don't use the next power of two when duplicating config valuesChris Robinson2014-05-041-1/+1
| | | | Speed isn't terribly important here, and reallocs should be extremely rare.
* Remove unnecessary ifdefsChris Robinson2014-05-042-4/+0
| | | | | mixer_sse.c and mixer_neon.c are only compiled when the relavent headers are found anyway.
* Always use the current gains when mixingChris Robinson2014-05-043-17/+22
| | | | | | The current gain gets explicitly set to the target when the stepping is finished to ensure the target is still used. This way, however, will allow for asynchronously 'canceling' a fade by setting the counter to 0.
* Better pack HRTF mixing propertiesChris Robinson2014-05-034-70/+65
|
* Clamp the current and target gain lower bound to epsilonChris Robinson2014-05-032-11/+11
| | | | | Should give a bit more wiggle room for the gain stepping to get lower than the silence threshold.
* Make sure all gain steps are applied with the SSE and Neon mixersChris Robinson2014-05-032-44/+22
|
* Revert e9aee578a797c694655538cd0a00ad6c0bf6d5dcChris Robinson2014-05-021-2/+2
| | | | | The vector iterator needs to be the start of the Data array even if the vector itself is NULL, or else insertion can have a negative insert position.
* Iniitialize some mmdevapi backend fields in the constructorChris Robinson2014-05-021-0/+23
|
* Use a backup in case pthread_mutex_timedlock isn't availableChris Robinson2014-05-013-1/+28
|
* Use ALboolean for al_string_empty's return typeChris Robinson2014-04-302-2/+2
|
* Add a helper VECTOR_FOR_EACH macroChris Robinson2014-04-306-55/+49
|
* Use _mm_set_ps() to set an __m128 instead of {}Chris Robinson2014-04-261-2/+2
|
* Convert the mmdevapi backend to the new backend APIChris Robinson2014-04-254-397/+511
|
* Use a vector for mmdevapi device listsChris Robinson2014-04-241-93/+77
|
* More device list cleanup cleanupChris Robinson2014-04-242-40/+25
|
* Use a helper function to clear PulseAudio device listsChris Robinson2014-04-241-36/+20
|
* Use a helper method to clear winmm device listsChris Robinson2014-04-241-20/+15
|
* Avoid using superfluous LP* typesChris Robinson2014-04-231-10/+10
|
* Use a RefCount for counting committed winmm buffersChris Robinson2014-04-231-7/+7
|
* Avoid forward-declaring backend vtablesChris Robinson2014-04-234-42/+44
|
* Avoid using assert() in createBackendChris Robinson2014-04-232-14/+20
|
* Include FindALSA.cmakeChris Robinson2014-04-231-0/+73
| | | | Older versions of CMake don't seem to have it.
* Properly handle HAVE_DYNLOAD in the dsound backendChris Robinson2014-04-221-1/+7
|
* Avoid some superfluous typesChris Robinson2014-04-221-8/+8
|
* Simplify the cmake backend handling a bitChris Robinson2014-04-221-31/+24
|
* Fix adding backend include directoriesChris Robinson2014-04-221-16/+48
|
* Use a helper to return mmdevapi message thread responsesChris Robinson2014-04-221-16/+15
|
* Avoid dereferencing a NULL pointerChris Robinson2014-04-211-2/+2
|
* Use a vector to store the ALSA device listsChris Robinson2014-04-211-96/+82
|
* Use al_string for ALSA's device mapChris Robinson2014-04-211-21/+26
|
* Trace the device name being opened in ALSA and PulseAudioChris Robinson2014-04-212-0/+4
|
* Use the althrd methods in winmmChris Robinson2014-04-211-78/+15
|
* Only declare backend options when the backend is availableChris Robinson2014-04-201-22/+20
|
* Use the stored format block align instead of calculating itChris Robinson2014-04-201-7/+4
|
* Use a vector for the PulseAudio DevMapsChris Robinson2014-04-201-76/+82
|
* Use an al_string in the PulseAudio backend instead of char*Chris Robinson2014-04-201-44/+63
|
* Convert the DSound backend to the new APIChris Robinson2014-04-194-224/+293
|
* Add a SZFMT macro for a size_t string formatterChris Robinson2014-04-192-2/+11
|
* Don't link with libwinmm unless WinMM is enabledChris Robinson2014-04-191-11/+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 DirectSoundChris Robinson2014-04-192-28/+60
|
* Use find cmake modules for more backendsChris Robinson2014-04-197-79/+259
|
* Add a GCC-specific STATIC_UPCAST macro that checks the object typeChris Robinson2014-04-194-4/+11
| | | | | The check is compile time, and is functionally identical to the old/alternate version.
* Use C11 alignas when availableChris Robinson2014-04-197-17/+55
|
* Dont bother trying _allocaChris Robinson2014-04-191-6/+1
| | | | | We prefer C99 variable-length arrays, and other systems (including MSVC9) don't have an issue with alloca.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-15 03:35:57 +0000