Commit Graph

3436 Commits

Author SHA1 Message Date
Chris Robinson 290badea95 Ensure PulseAudio's mixing loop is signaled when stopping 2015-01-21 22:02:02 -08:00
Chris Robinson 5cbafaa014 Remove a couple unused variables 2015-01-20 16:44:47 -08:00
Chris Robinson 0b6bb76259 Fix variable check 2015-01-20 16:42:23 -08:00
Chris Robinson 199646ac2b Fix X7DOT1_NARROW name to X7DOT1_WIDE 2015-01-15 09:27:37 -08:00
Chris Robinson c1607300bb Accept a "narrow" layout for 7.1 with mmdevapi 2015-01-15 01:56:04 -08:00
Chris Robinson 205e258d1a Fix high-pass coefficients 2015-01-13 09:38:13 -08:00
Chris Robinson f9a9235333 Add missing alignas to CubicLUT declaration 2015-01-13 09:29:03 -08:00
Chris Robinson ed05c83ee7 Fix the flags type 2015-01-11 15:02:26 -08:00
Chris Robinson 4dc9956a12 Remove some IN_IDE_PARSER uses 2014-12-24 17:15:50 -08:00
Chris Robinson bc17d6fc45 Remove some unnecessary restrict uses 2014-12-24 17:14:45 -08:00
Chris Robinson ca28d81b36 Update some function comments (bytes -> elements) 2014-12-24 16:42:57 -08:00
Chris Robinson c71d8c0fc5 Update a comment 2014-12-24 01:17:55 -08:00
Chris Robinson ad11a90720 Don't update the channel format if all JACK ports were allocated 2014-12-23 22:02:56 -08:00
Chris Robinson 93b69583d3 Try the device-specified channel configuration first
This does basically no checking that the channel config is correct, but should
be good enough for when non-stereo modes are requested.
2014-12-23 21:06:10 -08:00
Chris Robinson 2f162f0fc3 Use the custom ringbuffer instead of JACK's 2014-12-23 20:24:10 -08:00
Chris Robinson a9cce5de3e Add a lockless ringbuffer
Largely copied from JACK, it's extended to work with user-specified element
sizes instead of bytes. This is necessary to be able to work with 6- and 7-
channel output modes.
2014-12-23 20:17:34 -08:00
Chris Robinson 6ccf10bbde Don't set unused pointers to NULL, don't try to write 0 samples 2014-12-23 11:56:55 -08:00
Chris Robinson e84b3f4707 At least measure the ringbuffer size for JACK's latency
We should also add the port latency, but there's currently no way to ensure
it's synchronized with the ringbuffer (the ringbuffer will update before the
port latency gets updated).
2014-12-23 11:42:57 -08:00
Chris Robinson beb670a939 Add an option to prevent spawning a JACK server 2014-12-22 19:29:55 -08:00
Chris Robinson 5e189624b7 Add JACK to the config utility 2014-12-22 18:56:15 -08:00
Chris Robinson b8d98dd6d0 Add an option for an increased ringbuffer size with JACK 2014-12-22 18:48:54 -08:00
Chris Robinson 7e6d0d6b4f Fill out the JACK playback backend
A few notes about it:

The OpenAL device's requested buffer metrics are ignored, and instead the
device will keep one JACK-sized buffer's worth of audio prepared for JACK's
next process request.

Output is restricted to 32-bit float stereo. Part of this is because JACK
requires a buffer size that's a power of 2 (measured in samples), and the
ringbuffer requires a buffer size that's a power of 2 (measured in bytes). A
channel count of 6 (5.1) or 7 (6.1) will not work without causing a sample to
split over the edge of the ringbuffer. Additioanlly, JACK doesn't provide
information about what channel configuration a device or set of ports has, so
there's no way to know what ports 3 and up map to (even the first two ports are
unknown, but assuming stereo seems to work well enough).

There is no device latency measurement (for AL_SOFT_source_latency) due to the
non-atomicity between the ringbuffer's read size and port latency. A method is
needed to ensure the ringbuffer's read size and port latency can both be
measured between the end of a JACK update cycle (when the port latency has been
updated with the newly-retrieved audio) and the following ringbuffer read
update.
2014-12-22 13:30:35 -08:00
Chris Robinson 23197ddbc3 Add a skeleton backend for JACK 2014-12-21 15:51:16 -08:00
Chris Robinson f2d59671b2 Add a cmake module to find JACK 2014-12-21 15:48:40 -08:00
Chris Robinson b56d50912e Set error if capture device fails to start 2014-12-21 12:46:04 -08:00
Chris Robinson c2975b5f44 Trace lparam and wparam message values 2014-12-21 12:37:02 -08:00
Chris Robinson a91bca64f8 Check the PROPVARIANT type before use 2014-12-21 12:20:31 -08:00
Chris Robinson 13092d94b7 Use VECTOR_FOR_EACH instead of a manual loop 2014-12-21 10:45:10 -08:00
Chris Robinson 28a9f0826c Use a macro for the record thread name 2014-12-21 10:38:40 -08:00
Chris Robinson 6574fa4863 Cast to the pointer-to-type to increment the buffer 2014-12-21 09:48:36 -08:00
Chris Robinson aa4cf99e63 Fix logging on Windows 2014-12-20 06:24:55 -08:00
Chris Robinson abf0bd13ca Support capture with mmdevapi 2014-12-19 15:49:37 -08:00
Chris Robinson f300be00ae Handle logging Unicode strings on Windows 2014-12-19 10:14:02 -08:00
Chris Robinson 7f696edf13 Set the right variable when assuming CPU extensions 2014-12-19 08:50:30 -08:00
Chris Robinson c3a36d9b19 Do up to 256 samples at a time with multi-step loops 2014-12-18 09:23:55 -08:00
Chris Robinson 9897fca794 Inline a couple functions 2014-12-18 08:52:04 -08:00
Chris Robinson 13c2dd7115 Offset to the buffer's channel start first 2014-12-18 08:42:03 -08:00
Chris Robinson 94ad5b289f Avoid duplicate calculations 2014-12-18 07:42:14 -08:00
Chris Robinson 6086c344dc Assert that there's a buffer for mixing
For Clang's static analysis.
2014-12-17 10:57:35 -08:00
Chris Robinson 4b91d34d58 Set a couple pulse callbacks to NULL during reset, and minor cleanups 2014-12-17 10:10:28 -08:00
Chris Robinson af193e9fb3 Don't kill pulseaudio's mixer thread if it's already killed 2014-12-17 09:47:58 -08:00
Chris Robinson c37275efdc Use aluVector and aluMatrix in a couple more places 2014-12-16 10:36:44 -08:00
Chris Robinson c48a6196ad Constify some variables 2014-12-16 09:20:35 -08:00
Chris Robinson d2adefe063 Pass a vectory to aluMatrixVector 2014-12-16 07:42:17 -08:00
Chris Robinson 3b8f54d572 Use aluVector in some more places 2014-12-16 07:20:27 -08:00
Chris Robinson 97f6d302fe Add explicit matrix and vector types to operate with 2014-12-16 06:29:31 -08:00
Chris Robinson 1b7c554068 Multiply samples with the cubic coeffs before transposing
This avoids having to transpose the cubic coefficients.
2014-12-15 17:13:31 -08:00
Chris Robinson a0e7ad493b Load samples with _mm_loadu_ps in the cubic SSE resamplers 2014-12-15 15:27:31 -08:00
Chris Robinson 177dd95686 Add SSE2 and SSE4.1 cubic resamplers 2014-12-15 13:58:41 -08:00
Chris Robinson a606bbc7f1 Use a lookup table to do cubic resampling 2014-12-15 12:23:28 -08:00