mirrors/gstreamer
mirrors/gstreamer
1
1
Fork 0
mirror of https://gitlab.freedesktop.org/gstreamer/gstreamer.git synced 2024-11-30 21:51:09 +00:00
Code Issues 1 Projects Releases Wiki Activity
13158 commits 44 branches 1509 tags 464 MiB
Commit graph

1344 commits

Author SHA1 Message Date
Josep Torra
fc087f6419 opensles: initial attempt to reduce the src latency 2012-10-18 14:03:09 +02:00
Josep Torra
920354eb0d opensles: sprinkle comments and cosmetic fixes 2012-10-18 14:03:09 +02:00
Josep Torra
78e3b9f428 opensles: check for device outputs in the mixer 2012-10-18 14:03:09 +02:00
Josep Torra
1d9f48a33d opensles: drop 48kHz sample rate 
OpenSL ES implementation in Android is just a 'facade' API on top of
AudioFlinger which will downsample 48kHz into 44.1kHz before
delivering the audio to the underlaying hardware.

We found that it suffer some sort of underrun when the downsample
enters in action so relay on our good resampler to take care of that
and fix the clicks issue. And get an extra bonus of a lower latency.
 2012-10-18 14:03:09 +02:00
Josep Torra
9cc7e6a74d opensles: change the defaults to use 20 ms ringbuffer segments 
In my nexus7 seems that the internal min buffer size is 20 ms so make
our segments match.
 2012-10-18 14:03:09 +02:00
Josep Torra
97a1ccfab1 opensles: ensure that we register the callback only in STOPPED 
Fixes the error registering the callback on the PLAYING -> PAUSE ->
PLAYING state change sequence.
 2012-10-18 14:03:09 +02:00
Josep Torra
e265cec514 opensles: cap queue size 
Just in case we want to tweak the sink behaviour with buffer-time and
latency-time properties cap the queue size to something reasonable.
 2012-10-18 14:03:09 +02:00
Josep Torra
cc6fc15674 opensles: sink to provide the audioclock by default 2012-10-18 14:03:09 +02:00
Josep Torra
08ef2e3eed opensles: only drain half ringbuffer on start 
At start drain half ringbuffer into the OpenSL so the writting/reading
pointers will start at half ringbuffer distance.
 2012-10-18 14:03:09 +02:00
Josep Torra
3698d98921 opensles: monitor some player events 2012-10-18 14:03:09 +02:00
Josep Torra
3ff51bb88b opensles: rework around the _delay function 2012-10-18 14:03:09 +02:00
Josep Torra
27cdb7c2ca opensles: implement _delay function 2012-10-18 14:03:09 +02:00
Josep Torra
9fcfa00680 opensles: refactor to behave more like other sinks 
Reflect the queue in our own data buffer.
Drop the _commit hook that wasn't usefull.
Don't mess with the segsize/segtotal.
 2012-10-18 14:03:09 +02:00
Josep Torra
ed5870b605 opensles: read player position and show it in the log 2012-10-18 14:03:09 +02:00
Josep Torra
1e69918f3e opensles: do not provide a clock in the sink element. 2012-10-18 14:03:08 +02:00
Josep Torra
b00049a9f6 opensles: rework on start/stop operations and callback function handling 2012-10-18 14:03:08 +02:00
Josep Torra
f00d5a5cac opensles: fixes 8 bit format which is unsigned on android. 2012-10-18 14:03:08 +02:00
Josep Torra
eb6715a305 opensles: attempt to reduce playback latency 2012-10-18 14:03:08 +02:00
Josep Torra
1d9e16fb78 opensles: drop _buffer_clear calls and refactor to a shared _enqueue_cb 2012-10-18 14:03:08 +02:00
Josep Torra
1694befac6 opensles: use 0.25s segments in the sink to lower latency 2012-10-18 14:03:08 +02:00
Josep Torra
5612936d1a opensles: finish remaining bits for source element. 2012-10-18 14:03:08 +02:00
Josep Torra
c006973a4e opensles: attempt to query device for capabilities 2012-10-18 14:03:08 +02:00
Josep Torra
85bd75a2b1 opensles: add the mute property and handle volume/mute changes on the fly 2012-10-18 14:03:08 +02:00
Josep Torra
f06688c7b2 opensles: make the volume property actually work 2012-10-18 14:03:08 +02:00
Josep Torra
0291953997 opensles: rework on the ringbufffer to properly clear segments 
Make the segments bigger (1 second) as it seems to be the minimum size
we need to not introduce noise.
Sink works in my nexus 7 with rates from 8000 to 44100 and some noise
can be noticed on higger sample rates.
 2012-10-18 14:03:08 +02:00
Josep Torra
29334f3233 opensles: fixes the license headers 2012-10-18 14:03:08 +02:00
Josep Torra
805a010dc9 opensles: produces expected output until ringbuffer wraps 
Add some log messages.
Fixed a bit the _player_cb function and properly advance reding in the
ringbuffer.
Still produces noise when the ringbuffer wraps.
 2012-10-18 14:03:08 +02:00
Josep Torra
f6aa2f29bc opensles: Add initial draft implementation of OpenSL ES plugin. 
Initial draft implementation for a OpenSL ES based plugin for Android
that provides audio src and sink.
 2012-10-18 14:03:08 +02:00
Sebastian Dröge
77364e2b3c androidmedia: Use correct variable name in Makefile.am 2012-10-18 09:25:52 +02:00
Raimo Järvi
01853745a3 directsoundsrc: Fix compiler warning 
https://bugzilla.gnome.org/show_bug.cgi?id=673414
 2012-10-17 21:01:39 +01:00
Tim-Philipp Müller
32ba17cd0f Use gst_element_class_set_static_metadata() 
where possible. Avoids some string copies. Also re-indent
some stuff. Also some indent fixes here and there.
 2012-10-17 17:46:34 +01:00
Raimo Järvi
a7258842ab directsoundsrc: port to 1.0 
https://bugzilla.gnome.org/show_bug.cgi?id=673414
 2012-10-17 13:25:25 +01:00
Sebastian Dröge
80533fa3a3 androidmedia: Add missing file 2012-10-16 11:33:50 +02:00
Sebastian Dröge
6fa3f058b1 androidmedia: Rename plugin 2012-10-15 16:37:54 +02:00
Sebastian Dröge
5e954a7d0a androidmedia: Make everything buildable 2012-10-15 16:28:43 +02:00
Sebastian Dröge
f3682a0a6b Don't try to use the NVidia DRM codecs 2012-10-15 16:28:43 +02:00
Sebastian Dröge
31f0f163bd Try to handle format changes more gracefully 
And make stop() faster and more robust
 2012-10-15 16:28:43 +02:00
Sebastian Dröge
b0b642d8ab Add workaround for Google MP3 decoder outputting garbage in first output buffer 
And assume one decoded input frame per output buffer to fix timestamp
handling by the base class.
 2012-10-15 16:28:43 +02:00
Sebastian Dröge
2f3b2c586f Also add some more output format checks to the audio decoder 2012-10-15 16:28:42 +02:00
Sebastian Dröge
a870e6a5c3 Check output format metadata some more 
And implement workaround for NVidia Tegra 3 not setting the slice_height.
Thanks to Josep Torra for debugging this issue.
 2012-10-15 16:28:42 +02:00
Sebastian Dröge
7341ed62fa Add some more default channel layouts, these should be good for AAC at least 2012-10-15 16:28:42 +02:00
Sebastian Dröge
6ed3ea7cad Don't set timestamps, the baseclass handles this for us anyway 2012-10-15 16:28:42 +02:00
Sebastian Dröge
86176bd2a2 List profiles in reverse to minimize caps 2012-10-15 16:28:42 +02:00
Sebastian Dröge
fc5a18c091 Iterate levels in reverse order to minimize caps 2012-10-15 16:28:42 +02:00
Sebastian Dröge
0b0255ed12 Merge structures into caps instead of appending them 2012-10-15 16:28:42 +02:00
Sebastian Dröge
9f7ed3f180 Use orc_memcpy() instead of memcpy() as it should be faster for larger memory areas 2012-10-15 16:28:42 +02:00
Sebastian Dröge
7a53743bfd Add support for audio/mpeg-L2 for the Nexus 7 2012-10-15 16:28:42 +02:00
Sebastian Dröge
3159b7b5dd Add support for MPEG1/2 video 2012-10-15 16:28:42 +02:00
Sebastian Dröge
41862dff59 Ignore the *law and Vorbis decoders 
They are broken unfortunately.
 2012-10-15 16:28:42 +02:00
Sebastian Dröge
f098ae6b88 Add FIXME comment for not handling odd widths/height with the TI hardware decoders 
Must likely an issue with the codec itself though.
 2012-10-15 16:28:42 +02:00
Sebastian Dröge
42d11e8c1b Fix handling of I420 buffers with odd width 2012-10-15 16:28:41 +02:00
Sebastian Dröge
cb6f703669 Only mpegversion=4 is supported for AAC 2012-10-15 16:28:41 +02:00
Sebastian Dröge
742816df69 Remove some obsolete FIXMEs 2012-10-15 16:28:41 +02:00
Sebastian Dröge
f839a615a0 Fix flushing logic to unbreak handling of flushing seeks 2012-10-15 16:28:41 +02:00
Sebastian Dröge
bc7d2006e3 Fix chroma stride for I420 stride/crop conversion 2012-10-15 16:28:41 +02:00
Sebastian Dröge
7dbef6415f Fix Vorbis decoding 2012-10-15 16:28:41 +02:00
Sebastian Dröge
562b033872 Wait at most 100ms for dequeueing a buffer, and retry after that 
Some codecs don't fail the dequeueing if the codec is flushed,
causing deadlocks when shutting down the element.
 2012-10-15 16:28:41 +02:00
Sebastian Dröge
9a034f2beb Release output buffers a bit earlier 2012-10-15 16:28:41 +02:00
Sebastian Dröge
4c3617f7f0 Keep another reference of the input buffer in handle_frame() 
It might be unreffed in the meantime by the other thread due
to finish_frame().
 2012-10-15 16:28:41 +02:00
Sebastian Dröge
613a75ad92 Pass -1 as number of frames to the base class 
This will automatically do the right thing if timestamps are good enough
 2012-10-15 16:28:41 +02:00
Sebastian Dröge
0143418e0c Use correct buffer variable 2012-10-15 16:28:41 +02:00
Sebastian Dröge
bb8e53d4b8 Always pass 1 as number of frames to finish_frame() 2012-10-15 16:28:41 +02:00
Sebastian Dröge
99e287b86b Correctly remember number of channels and samplerate 2012-10-15 16:28:40 +02:00
Sebastian Dröge
efac01d096 Don't unref input buffer from handle_frame(), it's owned by the baseclass 2012-10-15 16:28:40 +02:00
Sebastian Dröge
3c40a9ca56 Remove invalid GST_ERROR() call 2012-10-15 16:28:40 +02:00
Sebastian Dröge
b07cd212ac Only try to get the channel-mask if it is set 2012-10-15 16:28:40 +02:00
Sebastian Dröge
afbc23ceed Add support for FLAC 2012-10-15 16:28:40 +02:00
Sebastian Dröge
b64e57ce42 Fix src template caps 2012-10-15 16:28:40 +02:00
Sebastian Dröge
6cc535581f Fix plugin initialization 2012-10-15 16:28:40 +02:00
Sebastian Dröge
77fbaae250 Add support for audio decoders, completely untested so far 2012-10-15 16:28:40 +02:00
Sebastian Dröge
de5375da81 Only set the SYNC_FRAME flag on the first buffer if we need to split the input 2012-10-15 16:28:40 +02:00
Sebastian Dröge
290d6f61c5 Ignore codecs with unknown color formats for stability purposes 
They can be registered nonetheless if an environment variable
or compile-time #define is set.
 2012-10-15 16:28:40 +02:00
Sebastian Dröge
65448a3790 Implement cropping and correctly handle the different color formats 2012-10-15 16:28:40 +02:00
Sebastian Dröge
3acc776dd6 Really implement stride conversion for NV12 too 2012-10-15 16:28:40 +02:00
Sebastian Dröge
266bc5bf69 Implement stride-conversion for I420 and NV12 2012-10-15 16:28:39 +02:00
Sebastian Dröge
024ac3560d Fix typo in previous commit 2012-10-15 16:28:39 +02:00
Sebastian Dröge
f2c5afd031 Don't allow using non-Google codecs if we started the Java VM ourselves 
Hardware codecs, e.g. the TI ones, only work in processes that had their
Java VM started via the non-public AndroidRuntime class.
 2012-10-15 16:28:39 +02:00
Sebastian Dröge
6feb6433d5 Rename to amc from androidmediacodec everywhere 2012-10-15 16:28:39 +02:00
Sebastian Dröge
2de542373b Improve debug output 2012-10-15 16:28:39 +02:00
Sebastian Dröge
0df00d68b5 Print the MediaFormats we get in the debug logs 2012-10-15 16:28:39 +02:00
Sebastian Dröge
11b8d33666 Improve debug output 2012-10-15 16:28:39 +02:00
Sebastian Dröge
d06af02637 Improve level/profile handling 2012-10-15 16:28:39 +02:00
Sebastian Dröge
2eff8bf82d Allow registering as static plugin with the GStreamer SDK API 2012-10-15 16:28:39 +02:00
Sebastian Dröge
b7a923234c If the buffer sizes are the same, just memcpy() to the output buffers 2012-10-15 16:28:39 +02:00
Sebastian Dröge
cb262e2465 Correctly convert timestamps between us and ns 2012-10-15 16:28:39 +02:00
Sebastian Dröge
2fe74efc04 Set SYNC_FRAME flag on buffers if required 2012-10-15 16:28:39 +02:00
Sebastian Dröge
ed98e8cf60 Improve flushing 2012-10-15 16:28:38 +02:00
Sebastian Dröge
d2aff4ecd3 Wait at most 0.5 seconds for getting an input buffer for EOS 2012-10-15 16:28:38 +02:00
Sebastian Dröge
3be4965d00 Correctly check for flushing 2012-10-15 16:28:38 +02:00
Sebastian Dröge
6a48db3d0f Retry for negative return values when dequeueing buffers 2012-10-15 16:28:38 +02:00
Sebastian Dröge
911ca50971 Fix refcount bug 2012-10-15 16:28:38 +02:00
Sebastian Dröge
b63c58e850 The MediaCodec API is apparently not threadsafe between stop() and release() 
So we don't call release() and just hope for the GC to release the
codec in time.
 2012-10-15 16:28:38 +02:00
Sebastian Dröge
525ecbb76a Fix plugin cache creation 2012-10-15 16:28:38 +02:00
Sebastian Dröge
e7b21d5796 Cache the codec informations inside the registry 
Otherwise we would always load all codecs during plugin initialization
which can take quite some time (because of hardware) and also loads
lots of shared libraries (which number is limited by 64 in Android).
 2012-10-15 16:28:38 +02:00
Sebastian Dröge
a98a627370 Properly attach and detach the current threads to the Java VM 
Attaching a thread after it was detached will cause segfaults,
as such we use a pthread_key_t to keep track of the JNIEnv* of
the attached threads and in the destructor (i.e. when the
thread exits) we detach the thread.
 2012-10-15 16:28:38 +02:00
Sebastian Dröge
bafd4a0b19 Don't get the output format from the codec before it is ready 
Apparently things are crashing otherwise.
 2012-10-15 16:28:38 +02:00
Sebastian Dröge
c4258069bb Get the input buffers from the codec 2012-10-15 16:28:38 +02:00
Sebastian Dröge
30b25fea35 Fix JNI method call signatures 2012-10-15 16:28:38 +02:00
Sebastian Dröge
686b9cf2fa Initialize the VM will full debugging for now 2012-10-15 16:28:37 +02:00
Sebastian Dröge
d7148efb8e Use NewStringUTF() instead of our own method for that 2012-10-15 16:28:37 +02:00