OSDN Git Service
Eino-Ville Talvala [Wed, 22 Oct 2014 21:21:12 +0000 (14:21 -0700)]
CameraService: Disconnect: Release mutex while waiting for joins.
The threads shutting down may have callpaths that require taking the
binder interface mutex, so waiting to join them with that mutex held
can lead to deadlocks.
A specific instance is StreamingProcessor calling dataCallbackTimestamp,
which can immediately lead to a Camera2Client::releaseRecordingFrame call,
which requires the binder interface mutex. If this call happens right when
shutdown is occurring, and Camera2Client::disconnect is holding the mutex,
deadlock ensues.
Bug:
17997578
Change-Id: I71253cd5542b5920ad205976d315110ca0043d94
Chong Zhang [Wed, 22 Oct 2014 00:56:56 +0000 (00:56 +0000)]
Merge "release camera without holding CameraSource mutex" into lmp-dev
Lajos Molnar [Wed, 22 Oct 2014 00:29:43 +0000 (00:29 +0000)]
Merge "stagefright: try to free codec instance if MediaCodec.release hangs" into lmp-dev
Lajos Molnar [Wed, 22 Oct 2014 00:29:29 +0000 (00:29 +0000)]
Merge "Stagefright: MediaCodec: shutdown allocated codec on error" into lmp-dev
Chong Zhang [Tue, 21 Oct 2014 21:12:25 +0000 (14:12 -0700)]
release camera without holding CameraSource mutex
Bug:
17997578
Change-Id: Iba93848ad10ba84d2d836573da7ce242c761582f
Lajos Molnar [Tue, 21 Oct 2014 23:07:52 +0000 (16:07 -0700)]
stagefright: try to free codec instance if MediaCodec.release hangs
Bug:
18033275
Change-Id: If86cd26566d7b75941976f37829bbec619800778
Lajos Molnar [Tue, 21 Oct 2014 16:36:43 +0000 (09:36 -0700)]
mediaplayer: don't report position past last queued position
also reanchor offloaded audio track correctly
Bug:
17653702
Change-Id: If05394d3b119be8d80b7764847496246d34a2133
Praveen Chavan [Wed, 15 Oct 2014 09:24:34 +0000 (02:24 -0700)]
Stagefright: MediaCodec: shutdown allocated codec on error
If MediaCodec sees a fatal error and transitions to
UNINITIALIZED state, The codec may still be alive (with an
exception of 'mediaserver-died' error).
Handle Shutdown of the codec during release().
Bug:
17784012
Bug:
18033275
Change-Id: I891e036499d9b440a57f77fb735a5ba4da9a6e43
Eric Laurent [Tue, 21 Oct 2014 16:07:11 +0000 (09:07 -0700)]
audio policy: fix effect handle leak
Fix ref counting error in AudioPolicyEffect::addInputEffects()
and addOutputSessionEffects()
Bug:
17675112.
Change-Id: I5267d1024a585919f66ead83662e57b78ba53540
Lajos Molnar [Tue, 21 Oct 2014 00:47:54 +0000 (17:47 -0700)]
mediaplayer: limit scheduling video frames into the future
This addresses when video timestamps jumps before an audio timestamp,
but still works on slideshow video clips (<=1fps).
This, however, will not skip time-changes on video-only live video
streams, as we cannot distinguish live slideshow video clips from
non-slideshow ones.
Bug:
18032127
Change-Id: I959a714edfe1c8cf3b84704c693dcd1b3e5b7855
Ronghua Wu [Mon, 20 Oct 2014 06:12:50 +0000 (23:12 -0700)]
NuPlayer: set anchor time for each audio buffer.
And use anchor time to compute current position
Bug:
17999949
Bug:
18008307
Bug:
18032127
Change-Id: Ie493c9a1d45d7b788aef65d863f710da6326fcc1
Marco Nelissen [Mon, 20 Oct 2014 20:15:38 +0000 (13:15 -0700)]
Fix the fix
We need to clear the buffer a little more often.
Bug:
18023017
Change-Id: If8a522b9133a4269e2435af8958f6ced322e71e9
Marco Nelissen [Mon, 20 Oct 2014 14:44:58 +0000 (14:44 +0000)]
Merge "Clear effect buffer before sending to chain" into lmp-dev
The Android Automerger [Mon, 20 Oct 2014 00:10:22 +0000 (17:10 -0700)]
merge in lmp-release history after reset to
e9234f17dfbb192c96b122b9e0ce8cd9015d1589
Marco Nelissen [Fri, 17 Oct 2014 23:39:39 +0000 (16:39 -0700)]
Clear effect buffer before sending to chain
otherwise the effect might get a buffer with previous data instead
of silence.
Bug:
18023017
Change-Id: Ic6be5ce0e74552aac2a0cce9c660893f9ad27f5c
Andy Hung [Fri, 17 Oct 2014 20:38:10 +0000 (20:38 +0000)]
Merge "Fix AudioTrack retrograde getPosition when restoring tracks" into lmp-dev
Andy Hung [Fri, 17 Oct 2014 00:54:34 +0000 (17:54 -0700)]
Fix AudioTrack retrograde getPosition when restoring tracks
mReleased represents the number of frames written to the
track, but was cleared on start() causing a mismatch if the
client wrote to the track before starting. Moved the clearing
to entering the STATE_STOPPED or STATE_FLUSHED state.
Bug:
18017947
Bug:
18022276
Change-Id: I3788c98c4c3c4d9cc004378432797b3f3138e22e
Chong Zhang [Thu, 16 Oct 2014 00:05:08 +0000 (17:05 -0700)]
move audio sink open/close to NuPlayerRenderer
Bug:
17675112
Change-Id: I7eb3d02380658f848baedafe2aea287586ccf016
Chong Zhang [Fri, 17 Oct 2014 01:10:23 +0000 (01:10 +0000)]
Merge "move audio sink open/close to NuPlayerRenderer" into lmp-dev
Chong Zhang [Thu, 16 Oct 2014 00:05:08 +0000 (17:05 -0700)]
move audio sink open/close to NuPlayerRenderer
Bug:
17675112
Change-Id: I7eb3d02380658f848baedafe2aea287586ccf016
Marco Nelissen [Thu, 16 Oct 2014 17:51:23 +0000 (10:51 -0700)]
Fix freed memory references
MPEG4Source references memory owned by MPEG4Extractor, and therefore an
MPEG4Extractor needs to be kept around as long as the MPEG4Sources obtained
from it exist.
Bug:
17890354
Change-Id: I399e18ec78517559ccc0914ffc7e099687c0ba51
Lajos Molnar [Thu, 16 Oct 2014 18:59:22 +0000 (18:59 +0000)]
Merge "Fix freed memory references" into lmp-dev
Chris Elliott [Thu, 16 Oct 2014 18:23:13 +0000 (18:23 +0000)]
am
b6dcdbf4: (-s ours) Revert "audio: prevent larger than required sleeps - DO NOT MERGE"
* commit '
b6dcdbf436ffacb52a41ac8ef9b07f99fd64779f':
Revert "audio: prevent larger than required sleeps - DO NOT MERGE"
Marco Nelissen [Thu, 16 Oct 2014 17:51:23 +0000 (10:51 -0700)]
Fix freed memory references
MPEG4Source references memory owned by MPEG4Extractor, and therefore an
MPEG4Extractor needs to be kept around as long as the MPEG4Sources obtained
from it exist.
Bug:
17890354
Change-Id: I399e18ec78517559ccc0914ffc7e099687c0ba51
The Android Automerger [Thu, 16 Oct 2014 13:11:49 +0000 (06:11 -0700)]
merge in lmp-release history after reset to
a2d4a6dac432f0c4f543d79b9a63c55ae91f81d6
Chris Elliott [Thu, 16 Oct 2014 02:17:27 +0000 (02:17 +0000)]
Revert "audio: prevent larger than required sleeps - DO NOT MERGE"
Revert change if we require a build beyond P
This reverts commit
620208dc0bbd7a0792702df3ab08800fdad60cec.
Change-Id: I07f65e35dbd9f814725f50f7bb346f7b285f41aa
Haynes Mathew George [Mon, 13 Oct 2014 20:05:22 +0000 (13:05 -0700)]
audio policy: Fix for voice call audio loss
Audio Policy service filters out one of the create patch
commands when it finds two pending commands with the same
patch handles. Due to this routing command is not received
to audio HAL and the voice call set up fails.
Fix this by filtering create patch commands only when
they are issued on the same output.
authored-by: Karthik Reddy Katta <a_katta@codeaurora.org>
Bug:
17787282
Change-Id: If36f0ab71e9b72d6a8eb61d31f762bc5e1683b89
Chris Elliott [Wed, 15 Oct 2014 22:10:42 +0000 (22:10 +0000)]
am
620208dc: (-s ours) audio: prevent larger than required sleeps - DO NOT MERGE
* commit '
620208dc0bbd7a0792702df3ab08800fdad60cec':
audio: prevent larger than required sleeps - DO NOT MERGE
Chris Elliott [Wed, 15 Oct 2014 20:46:15 +0000 (20:46 +0000)]
audio: prevent larger than required sleeps - DO NOT MERGE
This will be cherry picked into P build after O completes for other devices.
This reverts the revert from commit
2feb18ddfdbd20941ebcc3314780a582da5608e4.
b/
17962037
From: Haynes Mathew George <hgeorge@codeaurora.org>
Date: Wed, 26 Mar 2014 16:18:42 -0700
Subject: [PATCH] AudioTrack: prevent larger than required sleeps
AudioTrackThread can end up waiting for larger than necessary
time for free space to be available in the cblk. Fix this by
waiting on the cblk futex instead of the (internal) condition variable.
Change-Id: Iba5b266f9b4082b3833f4abd52ebc6601c8e3034
Chris Elliott [Wed, 15 Oct 2014 19:11:37 +0000 (19:11 +0000)]
am
2feb18dd: (-s ours) Revert "audio: prevent larger than required sleeps - DO NOT MERGE"
* commit '
2feb18ddfdbd20941ebcc3314780a582da5608e4':
Revert "audio: prevent larger than required sleeps - DO NOT MERGE"
The Android Automerger [Wed, 15 Oct 2014 13:11:39 +0000 (06:11 -0700)]
merge in lmp-release history after reset to
9d2189ffcf1142b2948ce3065c32521ed7d966c4
Chris Elliott [Wed, 15 Oct 2014 00:49:17 +0000 (00:49 +0000)]
Revert "audio: prevent larger than required sleeps - DO NOT MERGE"
This reverts commit
8c0417c173d1fd618ae54e2d721b0ec360ea70cc.
Reverting as we only wanted to include this CL the KNX01M build
Change-Id: Ib53292de658078285a38710915168e1746486fea
Zhijun He [Tue, 14 Oct 2014 23:31:50 +0000 (23:31 +0000)]
Merge "Camera2Client: avoid two consecutive sets of configure_streams calls" into lmp-dev
Haynes Mathew George [Tue, 14 Oct 2014 22:03:36 +0000 (22:03 +0000)]
am
8c0417c1: (-s ours) audio: prevent larger than required sleeps - DO NOT MERGE
* commit '
8c0417c173d1fd618ae54e2d721b0ec360ea70cc':
audio: prevent larger than required sleeps - DO NOT MERGE
Zhijun He [Tue, 14 Oct 2014 05:20:37 +0000 (22:20 -0700)]
Camera2Client: avoid two consecutive sets of configure_streams calls
When mediaRecorder starts without an active preview stream, Camera2Client
starts preview then immediately start recording, which could cause the second
configure_streams to call into HAL before any preview request is sent. This
could cause HAL to run into bad state. This change work around this issue
by making sure the first preview request is submitted to the HAL before
start recording.
Bug:
17915062
Change-Id: I94ae64ee76487603695a469240da601ddcb29a66
Haynes Mathew George [Thu, 9 Oct 2014 12:06:00 +0000 (20:06 +0800)]
audio: prevent larger than required sleeps - DO NOT MERGE
b/
17962037
From: Haynes Mathew George <hgeorge@codeaurora.org>
Date: Wed, 26 Mar 2014 16:18:42 -0700
Subject: [PATCH] AudioTrack: prevent larger than required sleeps
AudioTrackThread can end up waiting for larger than necessary
time for free space to be available in the cblk. Fix this by
waiting on the cblk futex instead of the (internal) condition variable.
Change-Id: I75bd14cd46f3b75afe0ea8bd9b48d7a2f95f654a
The Android Automerger [Tue, 14 Oct 2014 13:11:19 +0000 (06:11 -0700)]
merge in lmp-release history after reset to
447898bca790fae2fdae304073162e980d816b79
Lajos Molnar [Tue, 14 Oct 2014 08:31:20 +0000 (08:31 +0000)]
Merge "ACodec: only check crop rect on output port" into lmp-dev
Lajos Molnar [Tue, 14 Oct 2014 08:00:15 +0000 (08:00 +0000)]
Merge "NuPlayer: query current position from NuPlayerRenderer." into lmp-dev
Lajos Molnar [Sat, 11 Oct 2014 00:15:17 +0000 (17:15 -0700)]
stagefright: fix surface input handling of software encoders
- added SoftVideoEncoder for common color conversion and
extension handling logic
- fix YUV420 SemiPlanar handling that should be NV12 not NV21
Bug:
17935149
Change-Id: I9b8d05678b1862dd37bf349ea83d67bdf1bb5560
Lajos Molnar [Sat, 11 Oct 2014 00:14:17 +0000 (17:14 -0700)]
stagefright: support passing GraphicBuffer in metadata buffer
Bug:
17935149
Change-Id: I6bb5dd654e498a7153410afc052c2c8f7f35e44d
Ronghua Wu [Wed, 8 Oct 2014 22:13:29 +0000 (15:13 -0700)]
NuPlayer: query current position from NuPlayerRenderer.
Bug:
17653702
Change-Id: Ie0b1f92420b071a0cfcd389f5e7917a54d332541
Wei Jia [Mon, 13 Oct 2014 20:27:43 +0000 (20:27 +0000)]
Merge "NuPlayerDriver: request seek notification when prepare*() is called in STOPPED state." into lmp-dev
Wei Jia [Sat, 11 Oct 2014 22:33:00 +0000 (15:33 -0700)]
NuPlayerDriver: request seek notification when prepare*() is called in STOPPED state.
Bug:
17596535
Change-Id: Ib43f91f9b0d0a52112aa1fa8352adac15aaf80f8
Chong Zhang [Mon, 13 Oct 2014 19:48:51 +0000 (19:48 +0000)]
Merge "handle case OMX_AUDIO_CodingG711 in ACodec::getPortFormat" into lmp-dev
Wei Jia [Mon, 13 Oct 2014 18:41:03 +0000 (18:41 +0000)]
Merge "ACodec: process deferred messages when entering Loaded state." into lmp-dev
Chong Zhang [Mon, 13 Oct 2014 05:28:11 +0000 (22:28 -0700)]
handle case OMX_AUDIO_CodingG711 in ACodec::getPortFormat
Bug:
17935293
Change-Id: Id09190ae813aa9e00a3c1ef6fb8a1bd3021deb6e
The Android Automerger [Mon, 13 Oct 2014 13:11:02 +0000 (06:11 -0700)]
merge in lmp-release history after reset to
380757ba5c927015d403e3b97cc418413330453e
Guang Zhu [Mon, 13 Oct 2014 00:02:36 +0000 (00:02 +0000)]
Merge "Revert "handle case OMX_AUDIO_CodingG711 in ACodec::getPortFormat"" into lmp-dev
Guang Zhu [Mon, 13 Oct 2014 00:02:20 +0000 (00:02 +0000)]
Revert "handle case OMX_AUDIO_CodingG711 in ACodec::getPortFormat"
Bug:
17935293
fix 64-bit builds
This reverts commit
9c78328f4dfd2c8a1dd5d31010f10ffcd762f04b.
Change-Id: Ifbc499342fb47fd67cae02a79cb900e701caed68
The Android Automerger [Sun, 12 Oct 2014 13:11:26 +0000 (06:11 -0700)]
merge in lmp-release history after reset to
9286354c0a864ce2348cc7dfc6ca81e84a15f8e2
Dave Burke [Sun, 12 Oct 2014 05:30:16 +0000 (05:30 +0000)]
Merge "Revert "audio policy: add support for ro.audio.media_deep_buffer"" into lmp-dev
Vineeta Srivastava [Sat, 11 Oct 2014 21:59:36 +0000 (21:59 +0000)]
Revert "audio policy: add support for ro.audio.media_deep_buffer"
This reverts commit
95511add8ad4d440c67272503559f5042c4e6f1f.
Change-Id: I76a8b86790b06b08d8ca0f8d374e585f964ffba8
Wei Jia [Sat, 11 Oct 2014 01:13:00 +0000 (18:13 -0700)]
ACodec: process deferred messages when entering Loaded state.
Bug:
17792260
Change-Id: Ie8154cf41be5672fc8a9f3a2a82b7ede0a8b828f
Eric Laurent [Fri, 10 Oct 2014 23:39:36 +0000 (23:39 +0000)]
Merge "audio policy: add support for ro.audio.media_deep_buffer" into lmp-dev
Eric Laurent [Fri, 10 Oct 2014 22:31:38 +0000 (15:31 -0700)]
audio policy: add support for ro.audio.media_deep_buffer
Bug:
17931716.
Change-Id: Ife170f20f5d02b71059fbd384bd29759fafee826
Chong Zhang [Fri, 10 Oct 2014 22:19:44 +0000 (22:19 +0000)]
Merge "handle case OMX_AUDIO_CodingG711 in ACodec::getPortFormat" into lmp-dev
Eric Laurent [Fri, 10 Oct 2014 00:49:00 +0000 (17:49 -0700)]
audio policy: move intializations to onFirstRef()
Bug:
17896324.
Change-Id: I3bb74f5397142ac354bd815eb362e5ddacdcb463
Eric Laurent [Fri, 10 Oct 2014 00:49:00 +0000 (17:49 -0700)]
audio policy: move intializations to onFirstRef()
Bug:
17896324.
Change-Id: I3bb74f5397142ac354bd815eb362e5ddacdcb463
Chong Zhang [Fri, 10 Oct 2014 04:33:53 +0000 (21:33 -0700)]
handle case OMX_AUDIO_CodingG711 in ACodec::getPortFormat
Bug:
17935293
Change-Id: I8eab0acf55df6934e05947b0dbdd1fd456130088
Lajos Molnar [Fri, 10 Oct 2014 03:05:40 +0000 (03:05 +0000)]
Merge "MediaCodec: fix onError and onInputBufferAvailable" into lmp-dev
Wei Jia [Fri, 10 Oct 2014 00:30:58 +0000 (00:30 +0000)]
Merge "NuPlayerRenderer: adjust anchor time correctly for video only case." into lmp-dev
Lajos Molnar [Thu, 9 Oct 2014 23:32:33 +0000 (16:32 -0700)]
MediaCodec: fix onError and onInputBufferAvailable
- don't call onInputBufferAvailable if we created an input surface
- keep component name alive if an error occurs
Bug:
17934966
Change-Id: I396c6e6cfd9fa589a1a95c6169492089255e6993
Marco Nelissen [Thu, 9 Oct 2014 16:47:12 +0000 (09:47 -0700)]
Fix VideoFrame class layout for 64/32 bit
This structure is passed between 64 and 32 bit processes via shared
memory, so ensure it's the same size, and that the members we care
about are in the same place.
Bug:
17569156
Change-Id: Id776bc825af1fbf43a6dd3407cca064f6d932902
Marco Nelissen [Thu, 9 Oct 2014 19:34:31 +0000 (19:34 +0000)]
Merge "Fix VideoFrame class layout for 64/32 bit" into lmp-dev
Wei Jia [Thu, 9 Oct 2014 01:44:45 +0000 (18:44 -0700)]
NuPlayerRenderer: adjust anchor time correctly for video only case.
Bug:
17922171
Change-Id: I4c7e9e7f5ab96d3675d012ae30f78ef945394103
Marco Nelissen [Thu, 9 Oct 2014 16:47:12 +0000 (09:47 -0700)]
Fix VideoFrame class layout for 64/32 bit
This structure is passed between 64 and 32 bit processes via shared
memory, so ensure it's the same size, and that the members we care
about are in the same place.
Bug:
17569156
Change-Id: Id776bc825af1fbf43a6dd3407cca064f6d932902
The Android Automerger [Thu, 9 Oct 2014 13:11:10 +0000 (06:11 -0700)]
merge in lmp-release history after reset to
ce26e9d13da7a4a4443b9be775bb12fb3ac97d23
The Android Automerger [Wed, 8 Oct 2014 13:14:12 +0000 (06:14 -0700)]
merge in lmp-release history after reset to
ce26e9d13da7a4a4443b9be775bb12fb3ac97d23
Lajos Molnar [Wed, 8 Oct 2014 01:42:45 +0000 (18:42 -0700)]
ACodec: only check crop rect on output port
Bug:
17789956
Change-Id: I0e17660d95ac0b3b051e7f555820188d9286571a
Eric Laurent [Wed, 8 Oct 2014 00:16:25 +0000 (00:16 +0000)]
Merge "AudioTrack: fix AudioTrackThread crash upon exit" into lmp-dev
Andy Hung [Fri, 3 Oct 2014 16:53:53 +0000 (09:53 -0700)]
NuPlayer: wait for renderer flush before decoder shutdown
Bug:
17679341
Change-Id: Ie3883686891e7ee6fb45ceb01af1eb60b559d3a0
Eric Laurent [Tue, 7 Oct 2014 22:45:14 +0000 (15:45 -0700)]
AudioTrack: fix AudioTrackThread crash upon exit
In case of error during createTrack_l(), there is a possibility
that a newly created AudioTrackThread is resumed by requestExit()
just after entering the Thread class _threadLoop() but before
entering the class threadLoop(). In this case, processAudioBuffer()
is executed once with mCblk == 0 and we assert.
Bug:
17894033.
Change-Id: I93d0d0d4dbf14bcb88dad23e6ad49d2ecd47badd
Eric Laurent [Tue, 7 Oct 2014 20:46:02 +0000 (20:46 +0000)]
Merge "audio policy: fix build with USE_LEGACY_AUDIO_POLICY" into lmp-dev
Eric Laurent [Tue, 7 Oct 2014 20:45:51 +0000 (20:45 +0000)]
Merge "audio policy: fix lockup during mediaserver restart" into lmp-dev
Andy Hung [Tue, 7 Oct 2014 20:35:48 +0000 (20:35 +0000)]
Merge "NuPlayer: wait for renderer flush before decoder shutdown" into lmp-dev
Marco Nelissen [Tue, 7 Oct 2014 19:31:16 +0000 (19:31 +0000)]
Merge "Recover from errors" into lmp-dev
Eric Laurent [Tue, 7 Oct 2014 16:19:31 +0000 (09:19 -0700)]
audio policy: fix build with USE_LEGACY_AUDIO_POLICY
Bug:
17894512.
Change-Id: I6431c0b03ba02cf561413c34848a334caa0261fa
Eric Laurent [Tue, 7 Oct 2014 16:08:47 +0000 (09:08 -0700)]
audio policy: fix lockup during mediaserver restart
When pre or post processing effects are applied by audio policy service
automatically there is a potential lockup if the first call into
AudioPolicyService is made while creating one of those effects.
This is because effects are created with AudioPolicyService mutex held and
effect creation calls into methods (e.g registerClient()) which also acquire
the mutex.
The fix consists in adding a new mutex to AudioPolicyEffects class and not
hold the AudioPolicyService mutex when calling methods in of class.
Bug:
17830596.
Change-Id: Ie61c3671d3147f46e99ba208f39c1a3bf180779f
The Android Automerger [Tue, 7 Oct 2014 13:11:15 +0000 (06:11 -0700)]
merge in lmp-release history after reset to
584c14c370f141685b7e8963a13ed91584383ca1
Rachad [Tue, 7 Oct 2014 01:56:59 +0000 (01:56 +0000)]
Merge "stagefright: do not allocate output buffers for tunneled playback" into lmp-dev
Marco Nelissen [Sun, 5 Oct 2014 20:15:16 +0000 (13:15 -0700)]
Recover from errors
When the AAC decoder encounters an error, the lists of submitted
and consumed sizes need to be resynchronized, to prevent timestamp
drift.
Bug:
17638652
Change-Id: Ia6d5f48dc36be355803589e4a5796b01559942b8
Chong Zhang [Mon, 6 Oct 2014 18:39:27 +0000 (18:39 +0000)]
Merge "StageFrightRecorder: Enable meta-data mode for VP8 hardware encoder." into lmp-dev
Praveen Chavan [Mon, 6 Oct 2014 07:37:28 +0000 (00:37 -0700)]
StageFrightRecorder: Enable meta-data mode for VP8 hardware encoder.
Meta-data mode for input is preferred (rather required) for recording
with camera v1. This capability is queried and enabled per codec; enable
it for VP8 hardware encoder.
Bug:
17664960
Change-Id: I4735c61670ebe0a555002cc25ef3ed823ab02f1f
Andy Hung [Fri, 3 Oct 2014 16:53:53 +0000 (09:53 -0700)]
NuPlayer: wait for renderer flush before decoder shutdown
Bug:
17679341
Change-Id: Ie3883686891e7ee6fb45ceb01af1eb60b559d3a0
Ronghua Wu [Mon, 6 Oct 2014 15:38:17 +0000 (15:38 +0000)]
Merge "NuPlayer: add audioDecoderStillNeeded." into lmp-dev
The Android Automerger [Mon, 6 Oct 2014 13:12:47 +0000 (06:12 -0700)]
merge in lmp-release history after reset to
f0011d49c0f08753dc2dc254c5a70885a9986602
The Android Automerger [Sun, 5 Oct 2014 13:12:16 +0000 (06:12 -0700)]
merge in lmp-release history after reset to
f0011d49c0f08753dc2dc254c5a70885a9986602
Chong Zhang [Sat, 4 Oct 2014 01:35:42 +0000 (01:35 +0000)]
Merge "MediaRecorder: only dequeue available buffers from MediaCodec" into lmp-dev
Ronghua Wu [Fri, 3 Oct 2014 22:19:10 +0000 (15:19 -0700)]
NuPlayer: add audioDecoderStillNeeded.
Bug:
17752358
Change-Id: Ia3b3ee4df65e5564afd0723943d2d747f3a604af
Chong Zhang [Fri, 3 Oct 2014 23:12:37 +0000 (16:12 -0700)]
MediaRecorder: only dequeue available buffers from MediaCodec
Bug:
17514968
Change-Id: If232f92d163deb2440b927315c69e0c4d51ca290
Ronghua Wu [Fri, 3 Oct 2014 21:51:37 +0000 (21:51 +0000)]
Merge "NuPlayer: don't restart AudioSink when torn down due to pause timeout." into lmp-dev
Ronghua Wu [Thu, 2 Oct 2014 23:55:52 +0000 (16:55 -0700)]
NuPlayer: don't restart AudioSink when torn down due to pause timeout.
Bug:
17752358
Change-Id: I232d105480139bfc0b991bcd8e001930e577cf30
Eric Laurent [Fri, 3 Oct 2014 20:37:46 +0000 (20:37 +0000)]
Merge "audioflinger: fix pre processing transfer between record threads." into lmp-dev
Glenn Kasten [Sun, 28 Sep 2014 21:41:07 +0000 (14:41 -0700)]
Fix uncertainty of one normal mix buffer in AudioTrack::getTimestamp
The per-thread timestamp latch was not synchronized with the per-track
released frames. Now the value of each track's released frames is
latched along with the timestamp.
Bug:
17531839
Bug:
17669342
Change-Id: I9d50c8c6a5de55a3f4561ac40e20d497376c1257
Ronghua Wu [Fri, 3 Oct 2014 18:07:50 +0000 (18:07 +0000)]
Merge "stagefright: fix fake-stride handling for H263 SW decoder" into lmp-dev
Glenn Kasten [Fri, 3 Oct 2014 18:04:57 +0000 (18:04 +0000)]
Merge "Fix uncertainty of one normal mix buffer in AudioTrack::getTimestamp" into lmp-dev
Chong Zhang [Tue, 30 Sep 2014 20:12:22 +0000 (13:12 -0700)]
send available codec buffer count with codec notification
Bug:
14679336
Change-Id: Id927c96a9a14dd6ecd72540f0037d5841aa32154
Ronghua Wu [Thu, 2 Oct 2014 18:21:08 +0000 (11:21 -0700)]
stagefright: fix fake-stride handling for H263 SW decoder
- Use outputBufferWidth/Height to init H263 decoder.
- Handle stride change when using fake stride.
Bug:
17773094
Bug:
17326758
Change-Id: I66ce7b9c650383456923d75417c590e3a8a814e7
Eric Laurent [Fri, 3 Oct 2014 02:41:47 +0000 (19:41 -0700)]
audioflinger: fix pre processing transfer between record threads.
Fix two problems remaining with pre processing effects transfer from
one record thread to the next in case of tear down due to device connection:
1 - the enabled state of the effects was not communicated to the new HAL
input stream.
2 - the effects saved in orphan chains list were not transfered to the
new thread when a AudioRecord was created.
Bug:
17757378.
Change-Id: I0923c98470db3b51154dc89846157780a4c21e86