aboutsummaryrefslogtreecommitdiffstats
path: root/Alc
Commit message (Collapse)AuthorAgeFilesLines
* Implement capture for the DirectSound backendChris Robinson2012-01-311-61/+410
| | | | Based on a patch by Jeffrey Knockel <jeff250.at.gmail.com>
* Only initialize the default reverb effect when a slot is availableChris Robinson2012-01-291-1/+2
|
* Use sizeof(ALfloat) when finding the max stepping valueChris Robinson2012-01-261-4/+4
| | | | | The mixer always reads from floats, so the size of the storage type is unimportant.
* Make sure context attributes are specified for loopback devicesChris Robinson2012-01-261-0/+6
|
* Make sure InitializeEffect on the default slot doesn't set a context errorChris Robinson2012-01-261-0/+1
|
* Return the error code from UpdateDeviceParamsChris Robinson2012-01-251-22/+14
|
* Make sure the attributes list specifies a format for loopback devicesChris Robinson2012-01-251-3/+19
|
* Don't process a default device slot if it doesn't existChris Robinson2012-01-251-12/+15
|
* Set the source's slot send parameter to NULL if it's the AL_EFFECT_NULL typeChris Robinson2012-01-232-2/+5
|
* Avoid duplicating some initialization codeChris Robinson2012-01-201-13/+3
|
* Not all devices will have a default effect slotChris Robinson2012-01-201-2/+5
|
* Fix WaveIn shutdown for realChris Robinson2012-01-201-11/+8
| | | | | | The processing thread must be messaged when shutting down, and depending on the state of the device, waveInReset may not do it. Additionally, all buffers must be removed from the device before they can be unprepared and the device closed.
* Remove an unnecessary else statementChris Robinson2012-01-201-41/+38
|
* Always message the WinMM processing thread when a buffer is doneChris Robinson2012-01-201-12/+2
| | | | | So the processing thread can keep track of completed buffers when shutting down, too.
* Fix a possible race condition when shutting down WinMM devicesChris Robinson2012-01-201-23/+21
| | | | | | If the processing threads are interrupted between the shutdown check and re- adding the buffer back to the device, the device can be left with a buffer that will get unprepared and freed while in use.
* Add a default-reverb config option, and rename __ALSOFT_FORCE_REVERB to matchChris Robinson2012-01-201-1/+3
|
* Add the ability to specify a preset for forced reverbChris Robinson2012-01-191-1/+1
|
* Add a global option to apply a reverb effect on source send 0Chris Robinson2012-01-192-2/+53
| | | | | | A special slot on the device is created and processed, so it can be shared across all contexts on the device. Sources that don't have a slot set on send 0 will use this special slot instead.
* Avoid printing messages about device parameter changes in the backendsChris Robinson2012-01-179-187/+38
|
* Report parameter request changes in UpdateDeviceParamsChris Robinson2012-01-171-18/+44
|
* Change the default output format to 32-bit floatChris Robinson2012-01-171-1/+1
|
* Retry opening the PortAudio device with 16-bit short if 32-bit float output ↵Chris Robinson2012-01-171-3/+9
| | | | fails
* Retry opening the DSound device with 16-bit short if 32-bit float output failsChris Robinson2012-01-171-0/+11
|
* Retry opening the WinMM device with 16-bit short if 32-bit float output failsChris Robinson2012-01-171-0/+6
|
* Support 32-bit float in the WinMM backendChris Robinson2012-01-171-4/+11
|
* Safely insert the new context into the listChris Robinson2012-01-121-2/+3
|
* Finalize AL_SOFT_buffer_samplesChris Robinson2012-01-101-1/+1
|
* Make LockDevice, LockContext, etc, inlineChris Robinson2012-01-091-21/+0
|
* Remove the warning when a context being released is globally currentChris Robinson2012-01-011-3/+0
| | | | | | Just about every implementation allows this, and there's no direct harm caused by it. However, leave the warnings for when it's still current on specific threads, as this can cause issues.
* Use the device lock when removing the context from the device's listChris Robinson2012-01-011-9/+2
|
* Use a proper typedef for handling atomic pointer swapsChris Robinson2012-01-011-6/+6
|
* Use a string to specify the resampler config optionChris Robinson2011-12-201-4/+18
|
* Use a struct to associate channel enums and their anglesChris Robinson2011-12-201-50/+51
|
* Restore the old FPU mode if the effects fail to update with new device ↵Chris Robinson2011-12-201-0/+1
| | | | parameters
* Reset the ALSA device if the number of available samples grows too largeChris Robinson2011-11-281-2/+15
|
* Make ALSA XRUN when the number of available samples exceeds the buffer sizeChris Robinson2011-11-281-0/+5
|
* Don't fail to insert a map entry when the key exists and the limit is reachedChris Robinson2011-11-261-9/+9
|
* Initialize listener matrix to identityChris Robinson2011-11-261-0/+7
|
* Define TlsDestructor for Windows static builds, tooChris Robinson2011-11-121-1/+1
|
* Use loops to write reverb outputChris Robinson2011-11-021-37/+7
|
* Use a loop to write the modulator effect outputChris Robinson2011-11-011-10/+5
|
* Use panning for echo instead of specifying discrete channelsChris Robinson2011-11-011-28/+38
|
* Append _SOFT to the ALC loopback format enumsChris Robinson2011-11-011-26/+26
|
* Calculate the listener matrix when a new orientation is specifiedChris Robinson2011-10-301-57/+14
| | | | | This is so the matrix isn't derived each time a source is updated, and it will make supporting user-defined matrices easier.
* Don't translate the listener velocityChris Robinson2011-10-301-2/+5
|
* Don't transform/translate vectors with an invalid listener orientationChris Robinson2011-10-301-22/+26
|
* Initialize/Uninitialize COM as neededChris Robinson2011-10-241-2/+11
|
* Don't close devices that are left open on lib shutdownChris Robinson2011-10-241-24/+12
|
* Don't update the device update parameters when PulseAudio does an async ↵Chris Robinson2011-10-191-24/+18
| | | | | | | buffer update It risks a deadlock, and though the UpdateSize won't change, we'll still fill as much of the buffer as we can up to a multiple of it
* Use the local Device variableChris Robinson2011-10-111-5/+5
|