e5743d666c
SDL2: Improve Emscripten support by limiting the period size.
...
Setting the period size to something too small results in glitching so
this commit will clamp it to a minimum size on the Emscripten build.
2026-01-13 13:03:15 +10:00
629e751d56
Fix the SDL2 backend for Emscripten.
2026-01-13 12:56:07 +10:00
cdd3d39ac0
Emscripten: Force MA_NO_THREADING if not compiling with pthread support.
2026-01-13 12:53:30 +10:00
b3f32437b8
Update an out of date comment.
2026-01-13 12:46:51 +10:00
2b50979bd3
Minor reorganization.
2026-01-13 12:44:35 +10:00
2ead596562
PulseAudio: Revert a PipeWire compatibility workaround.
...
This is no longer needed since a native PipeWire backend is about to be
integrated.
2026-01-13 11:22:46 +10:00
27f7eeba91
Change default period count from 3 to 2.
2026-01-13 11:12:08 +10:00
51f8235bef
Force a minimum of two periods for duplex mode.
...
Not all backends have the notion of a period, but for those that do this
can make it a bit more reliable.
2026-01-13 10:47:32 +10:00
d1316a58cf
ALSA: Ensure at least two periods are used for duplex mode.
2026-01-13 10:37:25 +10:00
98a23f9551
ALSA: Fix an error with underrun recovery.
...
When an underrun occurs the buffer needs to be filled with data before
restarting or else snd_pcm_start() will return EPIPE.
2026-01-13 10:36:30 +10:00
55cdba084d
ALSA: Fix an error when priming the playback buffer.
...
There was a possibility that this process could result in
snd_pcm_writei() getting stuck because the frame count is not being
clamped to the buffer size properly.
2026-01-13 10:22:39 +10:00
92a58fae11
ALSA: Fix some typos.
2026-01-13 09:49:10 +10:00
1adad94ef8
Try fixing the Emscripten build.
2026-01-12 19:51:40 +10:00
994ae48feb
CMake: Minor output cleanup.
2026-01-12 19:39:15 +10:00
c50cc675c9
CMake: Try fixing the build for Clang and FORCE_C89.
2026-01-12 19:38:33 +10:00
676b84cad1
ALSA: Remove some out of date comments.
2026-01-12 19:25:02 +10:00
cd521f9440
Comment out a printf() debugging statement.
2026-01-12 17:45:48 +10:00
26029b2250
Add a basic engine test.
...
This is currently just for some specific tests, but might be expanded
later to be something a bit more practical.
2026-01-12 17:30:05 +10:00
628f2c1640
Add a debugging VFS.
2026-01-12 17:29:18 +10:00
82ec45e349
CMake: Fix an error with libvorbis.
2026-01-12 17:25:33 +10:00
f028249019
Add a command line option for exclusive mode to deviceio test.
2026-01-12 15:54:27 +10:00
94a79c9cda
Fix a bug in ma_calculate_frame_count_after_resampling().
...
This now uses the same calculation that the linear resampler uses.
Public issue https://github.com/mackron/miniaudio/issues/760
2026-01-12 12:43:09 +10:00
9a1b551658
WASAPI: Fix an error with rerouting.
...
When WASAPI reports `AUDCLNT_E_DEVICE_INVALIDATED` miniaudio will
attempt to reinitialize the device and continue processing. Prior to
this commit the reinit process would use NULL as the device ID which
is incorrect when an explicit device was requested.
With this commit the reinit process will correctly pass in the ID of the
device that it was initialized with. In practice, this will mean the
device will be put into an errored state if it is unplugged.
2026-01-12 10:57:09 +10:00
e9abcccd86
Update fs.
2026-01-12 07:11:43 +10:00
c649733b34
WASAPI: Remove old unused code.
2026-01-12 06:54:35 +10:00
4d583a4508
WASAPI: Fix device enumeration.
2026-01-12 06:37:14 +10:00
6a3d5fde05
Remove some line breaks from log entries.
2026-01-12 06:36:14 +10:00
df88373cd0
Silence a warning.
2026-01-11 20:21:19 +10:00
d4631208f9
WASAPI: Clean up a memory leak.
2026-01-11 20:15:51 +10:00
8709bef653
WASAPI: Improvement to buffer size negotiation for exclusive mode.
2026-01-11 20:02:47 +10:00
de3946c9ec
WASAPI: Properly recover from overruns in duplex mode.
2026-01-11 17:54:37 +10:00
12256314a9
WASAPI: Add a debug log message.
2026-01-11 17:19:58 +10:00
39abeb167e
WASAPI: Prime the playback buffer before starting.
2026-01-11 17:03:55 +10:00
3debb4a20f
ALSA: Add some missing snd_pcm_state_t values.
2026-01-11 07:23:14 +10:00
c4c484794b
ALSA: Add a timeout to poll().
...
I don't trust there won't be edge cases that result in this getting
stuck forever.
2026-01-11 07:22:44 +10:00
adce75cba6
Show device IDs in the deviceio test.
2026-01-11 07:06:23 +10:00
660d5f6f9e
ASLA: Fall back to NAME for the device description when DESC is empty.
2026-01-10 13:39:59 +10:00
96c4b105b0
ALSA: Use more descriptive error messages.
2026-01-10 13:07:10 +10:00
9a4318997b
ALSA: Whitespace.
...
This was off by one space and it was annoying me!
2026-01-10 12:45:30 +10:00
280c118dfd
ALSA: Try improving native channel count detection heuristics.
2026-01-10 12:41:19 +10:00
999c16d834
Merge branch 'dev' into dev-0.12
2026-01-10 12:14:40 +10:00
88776cedb7
Whitespace.
2026-01-10 08:49:12 +10:00
e00cee2af1
Cast tv_sec to 64-bit int before converting
2026-01-10 08:47:03 +10:00
760765ec93
ALSA: Use SND_PCM_NONBLOCK when enumerating devices.
2026-01-10 08:10:21 +10:00
fb132046d4
Add command line switches for periods and period sizes to deviceio test.
2026-01-09 19:21:24 +10:00
4bed03a65e
Fix a bug in ma_prev_power_of_2().
2026-01-09 18:51:02 +10:00
e9eccf49a3
ALSA: Improvements buffer size negotiation.
...
This makes it so period sizes are a power of 2, which in turns makes it
so the value set by `snd_pcm_sw_params_set_avail_min()` is also a power
of 2. This is inline with a suggestion by the ALSA documentation, and
seems to have an actual positive impact in practice with my testing as
well.
2026-01-09 15:37:05 +10:00
46adfae3bb
Update the deviceio test with a few more command line switches.
2026-01-09 13:38:49 +10:00
948967dcbb
ALSA: Change to device enumeration.
...
In order to get detailed information about a device, that is supported
formats, channels and sample rates, the PCM needs to be opened. This
can fail sometimes, in which case enumeration would previously just not
enumerate the device.
This is OK, but it creates issues. The first is that the enumerated
devices will not be consistent with what's reported by `aplay`. The
other is that when a hardware device is opened, iteration will not
include that device because it'll be opened in exclusive mode. This
creates a practical issue when trying to get the name of an already
opened device.
This commit makes it so that these devices will still be enumerated,
only they'll be missing detailed format, channels and rate information.
2026-01-09 13:38:07 +10:00
224a4c9d3a
ALSA: Fix a crash when failing to initialize a device.
2026-01-09 13:24:32 +10:00
David Reid
spevnev