aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Put per-channel filter properties togetherChris Robinson2014-05-193-39/+46
|
* Return a sample pointer from DoFiltersChris Robinson2014-05-191-14/+18
| | | | | This allows us to avoid copying samples when no filtering is needed. Doing the same for resampling would require another temporary buffer in the device.
* Move an HRTF mixer parameter and shorten a couple variable namesChris Robinson2014-05-184-33/+28
|
* Don't pass the SendParams to the wet-path mixerChris Robinson2014-05-186-41/+44
|
* Don't pass the DirectParams to the dry-path mixerChris Robinson2014-05-187-96/+119
|
* Use different parameters for HRTF mixersChris Robinson2014-05-186-29/+51
|
* Pass some DirectParams as function parametersChris Robinson2014-05-187-29/+40
|
* Use _mm_setr_ps instead of _mm_set_psChris Robinson2014-05-181-4/+4
| | | | | Apparently _mm_set_ps loads in reverse order compared to _mm_load_ps, so _mm_setr_ps should give what we really want.
* Add a ALfilterState method to process multiple samples at onceChris Robinson2014-05-184-9/+26
|
* Reset filter properties when setting a null filter typeChris Robinson2014-05-171-0/+6
|
* Support the band-pass filterChris Robinson2014-05-172-3/+88
|
* Apply high-pass source filters as neededChris Robinson2014-05-175-10/+88
|
* Implementing handling high-pass filter propertiesChris Robinson2014-05-173-2/+82
|
* Add a flag to specify when the low-pass filter needs to applyChris Robinson2014-05-173-29/+54
|
* Use logarithmic adjustment for the gain in the autowah effectChris Robinson2014-05-171-7/+7
|
* Document the different filter types, and combine some split linesChris Robinson2014-05-172-36/+20
|
* Rename freq_scale parameter to freq_multChris Robinson2014-05-172-3/+3
|
* Ensure that LoopEnd > LoopStartChris Robinson2014-05-151-0/+1
| | | | | | For the benefit of Clang's static analysis, which doesn't know that a static source will always have a valid buffer and a buffer must have a non-0 loop range.
* Initialize a pointer to NULL instead of a stringChris Robinson2014-05-151-2/+1
|
* Don't bother stripping whitespace from commentsChris Robinson2014-05-151-5/+1
|
* Early-out when outN is 0Chris Robinson2014-05-151-0/+3
|
* Remove some unnecessary castsChris Robinson2014-05-151-12/+13
|
* Initialize some variables when declaring themChris Robinson2014-05-151-4/+2
|
* Initialize newformat to AL_NONE instead of a valid formatChris Robinson2014-05-151-6/+2
|
* Use an enum for the chorus and flanger waveformsChris Robinson2014-05-142-16/+54
|
* Don't try to read the internal capture buffer into itselfChris Robinson2014-05-141-6/+9
|
* Remove a dead assignmentChris Robinson2014-05-141-1/+0
|
* Add a couple asserts to ensure a proper bufferChris Robinson2014-05-141-0/+3
| | | | | | It should not be possible for a playing or paused source to not have a valid buffer, but Clang's static analyzer doesn't know that. Hopefully an assert will convince it.
* Avoid freeing an in-use capture bufferChris Robinson2014-05-141-5/+3
| | | | | When stopping, ALSA may capture into its own storage buffer. Do not free the storage buffer if it first reads from it.
* Make RefCount a non-integer typeChris Robinson2014-05-1414-70/+88
| | | | | It should only be accessed through the appropriate functions to ensure proper atomicity.
* Add HF Reference as a filter propertyChris Robinson2014-05-143-0/+6
|
* Make LOWPASSFREQREF a float valueChris Robinson2014-05-114-5/+5
|
* Store the filter reference frequency in the sourceChris Robinson2014-05-113-12/+24
|
* Update the source send target gains properlyChris Robinson2014-05-111-4/+4
|
* Avoid locking the mixer when unqueueing buffersChris Robinson2014-05-111-4/+16
|
* Avoid accessing the source's buffer queue head multiple timesChris Robinson2014-05-111-17/+16
|
* Use a struct to store the source's direct gain/gainhf propertiesChris Robinson2014-05-113-15/+16
|
* Don't mix old- and new-style FIND_PACKAGE_HANDLE_STANDARD_ARGS callsChris Robinson2014-05-101-6/+2
| | | | | | | Some (all?) versions of cmake seem to get a bit confused when an old-style FIND_PACKAGE_HANDLE_STANDARD_ARGS call fails, and the scripts attempt a new- style call later on. It continues to think it's using old-style, acting as if REQUIRED_VARS is the message and looks for a VERSION_VAR variable.
* Avoid locking the mixer while queueing buffers onto a source.Chris Robinson2014-05-101-24/+34
|
* Mark the buffer queue pointers volatileChris Robinson2014-05-101-2/+2
|
* Add a missing pointer initChris Robinson2014-05-101-3/+3
|
* Update the output buffer pointer in the Write_* methodsChris Robinson2014-05-101-15/+11
|
* Add a couple constsChris Robinson2014-05-101-2/+2
|
* Explicit init the source queue and current_buffer pointersChris Robinson2014-05-101-0/+3
|
* Use a RWLock to help protect the source's buffer queueChris Robinson2014-05-103-35/+61
| | | | | In some instances this allows to to remove the device/mixer lock, or reduce how long it's held.
* Don't explicitly store the BuffersInQueueChris Robinson2014-05-102-10/+14
|
* Store the current buffer queue item, rather than played buffer countChris Robinson2014-05-104-101/+76
|
* Avoid checking the previous buffer multiple timesChris Robinson2014-05-101-9/+9
|
* Append the number of bits, not bytes, to typesChris Robinson2014-05-081-37/+27
| | | | Also use ALuint64SOFT from alext.h for uint64
* Always use unsigned int for RefCountChris Robinson2014-05-071-18/+22
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-02 17:10:36 +0000