aboutsummaryrefslogtreecommitdiffstats
path: root/Alc
Commit message (Collapse)AuthorAgeFilesLines
* Use floats for the listener transformsChris Robinson2016-05-162-66/+28
|
* Don't store the source's update method with the voiceChris Robinson2016-05-161-38/+42
|
* Avoid separate updates to sources that should apply togetherChris Robinson2016-05-152-12/+19
|
* Allocate context storage before starting/resetting the deviceChris Robinson2016-05-151-19/+23
| | | | | In case allocation fails, we don't have to worry about the playing status of the backend.
* Avoid using a flag to specify if the effect state needs to be updatedChris Robinson2016-05-152-6/+10
| | | | | This fixes a potential missed state change if an update with a new state got replaced with one that doesn't.
* Provide asynchronous property updates for sourcesChris Robinson2016-05-143-181/+153
| | | | | | | | | | | | | | | | | | | | | | | | | This necessitates a change in how source updates are handled. Rather than just being able to update sources when a dependent object state is changed (e.g. a listener gain change), now all source updates must be proactively provided. Consequently, apps that do not utilize any deferring (AL_SOFT_defer_updates or alcSuspendContext/alcProcessContext) may utilize more CPU since it'll be filling out more update containers for the mixer thread to use. The upside is that there's less blocking between the app's calling thread and the mixer thread, particularly for vectors and other multi-value properties (filters and sends). Deferring behavior when used is also improved, since updates that shouldn't be applied yet are simply not provided. And when they are provided, the mixer doesn't have to ignore them, meaning the actual deferring of a context doesn't have to synchrnously force an update -- the process call will send any pending updates, which the mixer will apply even if another deferral occurs before the mixer runs, because it'll still be there waiting on the next mixer invocation. There is one slight bug introduced by this commit. When a listener change is made, or changes to multiple sources while updates are being deferred, it is possible for the mixer to run while the sources are prepping their updates, causing some of the source updates to be seen before the other. This will be fixed in short order.
* Store the remaining context properties with the listener propertiesChris Robinson2016-05-131-2/+6
|
* Get rid of an unnecessary copy of ALeffectPropsChris Robinson2016-05-1312-61/+59
|
* Recognize AUX0...AUX15 for decoder speaker labelsChris Robinson2016-05-121-3/+11
|
* Hold the effect and filter maps while handling effects and filtersChris Robinson2016-05-121-8/+0
|
* Call the effect state update method after "returning" the container object.Chris Robinson2016-05-121-2/+2
|
* Avoid updating the effect state object if it's not changedChris Robinson2016-05-122-20/+19
|
* Provide (mostly) lockless updates for effect slotsChris Robinson2016-05-1213-132/+177
| | | | | | | | | | | | | | | | | Similar to the listener, separate containers are provided atomically for the mixer thread to apply updates without needing to block, and a free-list is used to reuse container objects. A couple things to note. First, the lock is still used when the effect state's deviceUpdate method is called to prevent asynchronous calls to reset the device from interfering. This can be fixed by using the list lock in ALc.c instead. Secondly, old effect states aren't immediately deleted when the effect type changes (the actual type, not just its properties). This is because the mixer thread is intended to be real-time safe, and so can't be freeing anything. They are cleared away when updates reuse the container they were kept in, and they don't incur any extra processing cost, but there may be cases where the memory is kept around until the effect slot is deleted.
* Use a lockless method for updating listener and context propertiesChris Robinson2016-05-112-22/+71
| | | | | | | | | | | This uses a separate container to provide the relevant properties to the internal update method, using atomic pointer swaps. A free-list is used to avoid having too many individual containers. This allows the mixer to update the internal listener properties without requiring the lock to protect against async updates. It also allows concurrent read access to the user-facing property values, even the multi-value ones (e.g. the vectors).
* Use the source's offset type to determine if there's an offsetChris Robinson2016-05-091-1/+1
|
* Find a valid source buffer before updating the voiceChris Robinson2016-05-092-50/+59
|
* Store more "active" listener and context properties separatelyChris Robinson2016-05-092-11/+25
| | | | | This helps ensure async listener/context property changes affect all playing sources at the same time.
* Fix placement of alignas - fixes Mac OS X buildrdb2016-05-033-12/+12
|
* Don't crash when there's no backend to probeChris Robinson2016-04-301-2/+7
|
* Start AL_SOFT_buffer_samples2 as a replacement for AL_SOFT_buffer_samplesChris Robinson2016-04-251-8/+22
|
* Add support for AL_EXT_SOURCE_RADIUSChris Robinson2016-04-251-4/+6
|
* Drop support for AL_SOFT_buffer_samples and AL_SOFT_buffer_sub_dataChris Robinson2016-04-241-12/+3
| | | | | | Unfortunately they conflict with AL_EXT_SOURCE_RADIUS, as AL_SOURCE_RADIUS and AL_BYTE_RW_OFFSETS_SOFT share the same source property value. A replacement for AL_SOFT_buffer_samples will eventually be made.
* Avoid an unnecessary aluVectorChris Robinson2016-04-241-7/+7
|
* Improve radius behavior with scaling of ambisonic coefficientsChris Robinson2016-04-2410-69/+114
|
* Increase max output channels to 16Chris Robinson2016-04-191-11/+14
| | | | This also enables fully periphonic 3rd order HQ decoding.
* Combine two if checks into oneChris Robinson2016-04-191-5/+3
|
* Remove unneeded ChannelMaps for BFormat formatsChris Robinson2016-04-171-14/+2
|
* Trace the requested attribute valuesChris Robinson2016-04-171-0/+24
|
* Avoid storing channel names for the dry bufferChris Robinson2016-04-162-25/+16
|
* Don't look up the dry buffer's FrontCenter in the Dedicated effectChris Robinson2016-04-161-8/+4
| | | | | | The real FrontCenter output is used if it exists, and if it doesn't, it's unlikely the dry buffer will have it (and even if it does, it won't be any better than panning).
* Reorganize a bit of aluInitRenderer codeChris Robinson2016-04-161-46/+30
|
* Silence possible out-of-bounds warning againChris Robinson2016-04-161-1/+1
|
* More directly map coefficients for ambisonic mixing buffersChris Robinson2016-04-1512-123/+126
| | | | | | Instead of looping over all the coefficients for each channel with multiplies, when we know only one will have a non-0 factor for ambisonic mixing buffers, just index the one with a non-0 factor.
* Update some commentsChris Robinson2016-04-151-3/+3
|
* Avoid mixing all coefficients together when only some are usedChris Robinson2016-04-157-59/+75
|
* Fix possible out-of-bounds warningChris Robinson2016-04-151-2/+2
|
* Shorten VECTOR_ITER_ macros to VECTOR_Chris Robinson2016-04-159-65/+62
|
* Avoid unnecessary loops for setting up effect slot b-format buffer mixingChris Robinson2016-04-142-24/+40
|
* Prepare the custom decoder in aluInitRendererChris Robinson2016-04-141-200/+193
|
* Move the InitRenderer method to panning.cChris Robinson2016-04-142-184/+174
|
* Separate some long if-else blocks into a separate functionChris Robinson2016-04-141-164/+179
|
* Split aluInitPanning into separate functions for HRTF or UHJChris Robinson2016-04-143-249/+241
|
* Improve error trace for bad matrix definitionsChris Robinson2016-04-101-2/+2
|
* Use frequency-dependent processing for the ambisonic up-samplerChris Robinson2016-04-091-3/+37
|
* Avoid double slashes when constructing pathsChris Robinson2016-04-071-3/+11
|
* Use al_malloc/al_free for allocationsChris Robinson2016-04-051-9/+7
|
* Also disable ALSA's resampler when not requesting a sample rateChris Robinson2016-04-051-1/+4
|
* Add a hack to workaround erroneous prebuf values from pulseChris Robinson2016-03-311-5/+28
|
* Use the 3D panned output for reverb with HQ decodingChris Robinson2016-03-311-3/+3
| | | | | | This is less than ideal, but matching each reverb line to a speaker with surround sound output is way too loud without the ambient volume scaling offered by the "direct" panning.
* Allow dual-band decoders for basic renderingChris Robinson2016-03-301-8/+2
| | | | | Since the basic renderer does not do frequency-dependent processing, the high- frequency matrix is used for panning.