Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Remove the VirtOut buffer alias | Chris Robinson | 2016-07-05 | 3 | -30/+24 |
| | |||||
* | Use separate arrays for UIntMap keys and values | Chris Robinson | 2016-07-04 | 8 | -58/+87 |
| | |||||
* | Use al_malloc/al_free for the UIntMap array | Chris Robinson | 2016-07-04 | 2 | -9/+17 |
| | |||||
* | Ensure voices has been updated once before mixing them | Chris Robinson | 2016-06-16 | 2 | -2/+10 |
| | | | | | | | | Sometimes the mixer is temporarily prevented from applying updates, when multiple sources need to be updated simultaneously for example, but does not prevent mixing. If the mixer runs during that time and a voice was just started, it would've mixed the voice without any internal properties being set for it. | ||||
* | Rename MaxNoOfSources for consistency | Chris Robinson | 2016-06-08 | 2 | -16/+16 |
| | |||||
* | Remove an IN_IDE_PARSER hack | Chris Robinson | 2016-06-07 | 1 | -6/+0 |
| | | | | | | Not all IDE parsers necessarily choke on restrict, and even those that do can probably have their own configuration to define macros that can hide the parsing errors caused by it. | ||||
* | Look in the executable's dir for another config file | Chris Robinson | 2016-06-04 | 3 | -0/+133 |
| | | | | | On Windows it'll look for alsoft.ini, and elsewhere is alsoft.conf. This applies after the user-local settings and before ALSOFT_CONF. | ||||
* | Only define 8 HRTF filter states and params for the device | Chris Robinson | 2016-06-04 | 1 | -2/+2 |
| | | | | | It will only be used with a cube channel setup, so there's no need to have one for every possible output channel. | ||||
* | Add property queries to get the device latency with the clock | Chris Robinson | 2016-06-03 | 2 | -11/+34 |
| | |||||
* | Improve the filter processing function | Chris Robinson | 2016-06-03 | 1 | -2/+35 |
| | |||||
* | Remove some unnecessary volatile keywords | Chris Robinson | 2016-06-03 | 2 | -33/+33 |
| | |||||
* | Implement a Neon-enhanced MixRow | Chris Robinson | 2016-06-01 | 3 | -0/+31 |
| | |||||
* | Make a function static | Chris Robinson | 2016-06-01 | 2 | -7/+7 |
| | |||||
* | Use a macro to specify the ambisonic periphonic channel mask | Chris Robinson | 2016-06-01 | 3 | -4/+13 |
| | |||||
* | Use SSE for applying the HQ B-Format decoder matrices | Chris Robinson | 2016-05-31 | 5 | -27/+82 |
| | |||||
* | Don't access the band splitter fields in the processing loops | Chris Robinson | 2016-05-31 | 1 | -9/+16 |
| | | | | | | perf shows a 5% drop in relative execution time on the alffplay example with an audio-only file (20% to 15%). Kinda figured the optimizer would handle it better, but I guess not. | ||||
* | Don't try to emulate almtx_timedlock | Chris Robinson | 2016-05-30 | 1 | -43/+12 |
| | |||||
* | Clean up a couple variable names and declarations | Chris Robinson | 2016-05-30 | 2 | -7/+6 |
| | |||||
* | Remove unnecessary VECTOR_INSERT | Chris Robinson | 2016-05-30 | 2 | -54/+30 |
| | |||||
* | Hold the effectslot map lock while handling it | Chris Robinson | 2016-05-29 | 3 | -2/+38 |
| | |||||
* | Use a linked list for active effect slots | Chris Robinson | 2016-05-29 | 5 | -60/+59 |
| | |||||
* | Avoid the mixer lock when getting the plain source offset | Chris Robinson | 2016-05-28 | 1 | -11/+15 |
| | | | | i.e. without the latency | ||||
* | Avoid an explicit mixer lock for getting the source offset and latency | Chris Robinson | 2016-05-28 | 1 | -27/+72 |
| | | | | | The only mixer locking involved is with the backend, as determined by it's ability to get the device clock and latency atomically. | ||||
* | Change the backend getLatency method to return the clock time too | Chris Robinson | 2016-05-28 | 15 | -53/+119 |
| | | | | | | This will also allow backends to better synchronize the tracked clock time with the device output latency, without necessarily needing to lock if the backend API can allow for it. | ||||
* | Use the backend lock for the effectstate's deviceUpdate call | Chris Robinson | 2016-05-27 | 1 | -8/+4 |
| | |||||
* | Use a specific lock for the backend's stop/reset/play calls | Chris Robinson | 2016-05-27 | 2 | -30/+46 |
| | | | | | This helps protect against the device changing unexpectedly from multiple threads, instead of using the global list/library lock. | ||||
* | Remove a couple unneeded functions | Chris Robinson | 2016-05-25 | 3 | -24/+7 |
| | |||||
* | Increment the device's mix count closer to the mixing loops | Chris Robinson | 2016-05-23 | 1 | -3/+2 |
| | |||||
* | Properly pluralize some messages | Chris Robinson | 2016-05-22 | 2 | -8/+14 |
| | |||||
* | Improve locking for device attribute queries | Chris Robinson | 2016-05-22 | 1 | -8/+28 |
| | | | | | Avoids the backend device lock, instead using the list lock to prevent the device from changing while being queried, and adds some missing locks. | ||||
* | Avoid using realloc in a number of places | Chris Robinson | 2016-05-21 | 9 | -32/+53 |
| | |||||
* | Make the source position calues atomic | Chris Robinson | 2016-05-19 | 4 | -30/+31 |
| | |||||
* | Copy the source's Looping property into the voice | Chris Robinson | 2016-05-18 | 1 | -0/+2 |
| | |||||
* | Avoid redundantly storing distance model settings | Chris Robinson | 2016-05-17 | 6 | -33/+11 |
| | |||||
* | Update the right effect state when the device is reset | Chris Robinson | 2016-05-17 | 1 | -2/+2 |
| | |||||
* | Don't assume the "real" output buffer follows the dry buffer | Chris Robinson | 2016-05-17 | 1 | -15/+44 |
| | |||||
* | Improve reverb panning gains for "3D" output. | Chris Robinson | 2016-05-17 | 1 | -8/+8 |
| | |||||
* | Ignore the listening angle for the wet path sound cones | Chris Robinson | 2016-05-16 | 1 | -32/+51 |
| | | | | | | | Since the wet path is essentially the room response to a sound, the direction of the sound to the listener doesn't change the amount of energy the room receives. Instead, the surface area defined by the cones dictate the volume the room gets for the sound. | ||||
* | Use floats for the listener transforms | Chris Robinson | 2016-05-16 | 4 | -92/+29 |
| | |||||
* | Don't store the source's update method with the voice | Chris Robinson | 2016-05-16 | 4 | -48/+42 |
| | |||||
* | Avoid separate updates to sources that should apply together | Chris Robinson | 2016-05-15 | 5 | -15/+37 |
| | |||||
* | Allocate context storage before starting/resetting the device | Chris Robinson | 2016-05-15 | 1 | -19/+23 |
| | | | | | In case allocation fails, we don't have to worry about the playing status of the backend. | ||||
* | Avoid duplicate effect state objects in the freelist | Chris Robinson | 2016-05-15 | 1 | -12/+25 |
| | | | | | | | If an unapplied update was superceded, it would be placed in the freelist with its effect state object intact. This would cause another update with the same effect state object to be placed into the freelist as well, or worse, cause it to get deleted while in use when the container had its effect state cleared. | ||||
* | Avoid using a flag to specify if the effect state needs to be updated | Chris Robinson | 2016-05-15 | 4 | -19/+20 |
| | | | | | 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 sources | Chris Robinson | 2016-05-14 | 10 | -356/+627 |
| | | | | | | | | | | | | | | | | | | | | | | | | | 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. | ||||
* | Merge pull request #41 from ColdPie1/source_limit | kcat | 2016-05-14 | 1 | -1/+1 |
|\ | | | | | alsoft-config: Raise source limit to 4096 | ||||
| * | alsoft-config: Raise source limit to 4096 | Andrew Eikum | 2016-05-13 | 1 | -1/+1 |
| | | |||||
* | | Store the remaining context properties with the listener properties | Chris Robinson | 2016-05-13 | 4 | -8/+20 |
| | | |||||
* | | Get rid of an unnecessary copy of ALeffectProps | Chris Robinson | 2016-05-13 | 13 | -64/+61 |
|/ | |||||
* | Properly load the effect state pointer from the property container | Chris Robinson | 2016-05-13 | 1 | -2/+6 |
| |