OSDN Git Service
Andy McFadden [Thu, 5 Dec 2013 21:30:52 +0000 (21:30 +0000)]
Jeff Brown [Thu, 5 Dec 2013 21:29:06 +0000 (21:29 +0000)]
am
82278b2c: am
fe9611bd: Merge "Fix potential leak of audio input handle." into klp-dev
* commit '
82278b2ceeebbcf345ed81413eeffa82fa82e05b':
Andreas Huber [Thu, 5 Dec 2013 19:46:23 +0000 (19:46 +0000)]
am
4d9bc1a7: am
b8689b31: Support for "request-sync" parameter in ACodec
* commit '
4d9bc1a742b386db4985ad1f973091a50207fdc2':
Glenn Kasten [Thu, 5 Dec 2013 19:40:45 +0000 (19:40 +0000)]
am
60147776: (-s ours) am
7473e356: (-s ours) am
943917c4: (-s ours) Merge "Fix SoundPool.play() looping regression in 4.3 DO NOT MERGE"
* commit '
601477769cef254d3a95338e853a37ccf9aac63f':
Fix SoundPool.play() looping regression in 4.3 DO NOT MERGE
Lajos Molnar [Tue, 19 Nov 2013 01:28:28 +0000 (01:28 +0000)]
Merge "WA: Queue extra buffers on output port during reconfig if input EOS-ed" into klp-dev
Marco Nelissen [Fri, 15 Nov 2013 21:58:45 +0000 (21:58 +0000)]
Merge "Fix metadata access" into klp-dev
Marco Nelissen [Fri, 15 Nov 2013 21:49:58 +0000 (13:49 -0800)]
Fix metadata access
Metadata string pointers become invalid after setting more metadata,
so don't cache them.
b/
11692062
Change-Id: Iaf1afb24cf53f7fa36f49ce759355693494076e5
Eric Laurent [Fri, 15 Nov 2013 20:02:28 +0000 (12:02 -0800)]
audioflinger: do not use raw pointer for tracks
Commit
9da3d95 surfaced a problem caused by the use of a raw
pointer to a track in offload thread implementation.
Pointers to tracks should always be weak or strong pointers.
Bug:
11708529.
Change-Id: Ic48632532d186c9be8261f73cefdf824b9fbbd2b
Eric Laurent [Fri, 15 Nov 2013 15:34:11 +0000 (15:34 +0000)]
Merge "AwesomePlayer: correct stream type for offload" into klp-dev
Eric Laurent [Fri, 15 Nov 2013 01:28:47 +0000 (17:28 -0800)]
AwesomePlayer: correct stream type for offload
canOffloadStream() function in stagefright utils forces the
stream type to AUDIO_STREAM_MUSIC when querying the audio policy
manager if a particular track is offloadable or not.
This causes MP3 ringtones to be offloaded which is not a validated use case.
The fix consists in using the actual stream type read from the AudioSink.
Bug:
11410937.
Change-Id: I44b8e033a8e785a79cdc291b142f80b5580bdc4d
Lajos Molnar [Thu, 7 Nov 2013 05:04:34 +0000 (21:04 -0800)]
WA: Queue extra buffers on output port during reconfig if input EOS-ed
Some codecs may return input buffers before having them processed.
This causes a halt if we already signaled an EOS on the input
port. For now keep submitting output meta buffers one at a time if
the input EOS-ed, but the output not yet.
Normally, we submit an output buffer for each input buffer that is
with the component (waiting to be processed).
Change-Id: I8a1251bfb504f40f1e4085a1e220bf9a4d0b05d9
Signed-off-by: Lajos Molnar <lajos@google.com>
Bug:
11433909
Eino-Ville Talvala [Thu, 14 Nov 2013 18:32:13 +0000 (10:32 -0800)]
Camera2: Rework the FPS range vs. FPS single setting detection
Give up on current approach of writing out consistent FPS values
into parameters that will be read back by the app.
- Preserve app's latest set parameters exactly, and compare against
them when detecting if a new FPS range or single FPS value has been
selected.
- Since get() returns exactly what was set(), it doesn't matter if the
app calls getParameters() before its next setParameters(), in terms
of retriggering FPS selection logic. Before, the behavior varied
depending on whether the app re-read the parameters.
- As before, if app changes both range and single FPS in a single set
call, the range set wins. Otherwise, the value that has changed more
recently is used.
Bug:
11570973
Change-Id: I72b5e60c3f60e88d55127dd1bda87e26eaf929c6
Eric Laurent [Thu, 14 Nov 2013 16:43:19 +0000 (16:43 +0000)]
Merge "audioflinger: fix offload track transition" into klp-dev
Eric Laurent [Thu, 14 Nov 2013 16:42:30 +0000 (16:42 +0000)]
Merge "audioflinger: fix offload resume after drain" into klp-dev
Lajos Molnar [Wed, 13 Nov 2013 23:53:42 +0000 (23:53 +0000)]
Merge "stagefright: limit default max-input-size for AVC" into klp-dev
Lajos Molnar [Wed, 13 Nov 2013 23:52:37 +0000 (23:52 +0000)]
Merge "AwesomePlayer: Improve performance on high-fps clips" into klp-dev
Lajos Molnar [Wed, 13 Nov 2013 22:11:41 +0000 (14:11 -0800)]
stagefright: limit default max-input-size for AVC
AVC has a compression ratio of at least 2. Account for this, as
some codecs refuse exorbitant max-input-size requests.
Change-Id: Ib68b485879ba33a12a486f2ce18fe100be977699
Signed-off-by: Lajos Molnar <lajos@google.com>
Bug:
11599364
Marco Nelissen [Wed, 13 Nov 2013 18:11:54 +0000 (18:11 +0000)]
Merge "Only initialize sniffers once, don't hold mutex while sniffing" into klp-dev
Andreas Huber [Wed, 13 Nov 2013 18:06:01 +0000 (18:06 +0000)]
Merge "Ignore transport stream packets with the "transport_error_indicator" bit set" into klp-dev
Adam Hampson [Wed, 13 Nov 2013 17:25:33 +0000 (09:25 -0800)]
am
27ae268a: am
00cb884f: Merge "StageFright: Removes hacks for Nvidia MPEG2 decoder" into jb-mr2-dev
* commit '
27ae268af29b25d85f83aa34e92ac8b7f6c87c6f':
StageFright: Removes hacks for Nvidia MPEG2 decoder
Adam Hampson [Wed, 13 Nov 2013 17:21:33 +0000 (09:21 -0800)]
am
00cb884f: Merge "StageFright: Removes hacks for Nvidia MPEG2 decoder" into jb-mr2-dev
* commit '
00cb884fc870aba4052c8d187a45e4aec50a6b43':
StageFright: Removes hacks for Nvidia MPEG2 decoder
Adam Hampson [Wed, 13 Nov 2013 17:14:31 +0000 (17:14 +0000)]
Merge "StageFright: Removes hacks for Nvidia MPEG2 decoder" into jb-mr2-dev
Eric Laurent [Wed, 13 Nov 2013 03:25:43 +0000 (19:25 -0800)]
audioflinger: fix offload track transition
Make sure that when transitioning from one active
offloaded track to the next we flush DSP content
if both tracks are not on the same audio session.
This happens when switching between two applications
playing the same type of content (e.g MP3, stereo, 44.1).
In this case, we reuse the same output thread because the track
formats are compatible.
Bug:
11247103.
Change-Id: I2b9031591149adeb70766da5e0d21ff2933a37e8
Eric Laurent [Wed, 13 Nov 2013 03:10:17 +0000 (19:10 -0800)]
audioflinger: fix offload resume after drain
When pausing and resuming during the drain,
the drain sequence number shuold not be modified
otherwise the drain callback will be ignored.
This causes failure to notify end of stream to audio
player and transition to next song.
Bug:
11247103.
Change-Id: I2a35c5cc3fd6aa667cdd257f9e9cc8715cef5159
Lajos Molnar [Wed, 23 Oct 2013 22:19:47 +0000 (15:19 -0700)]
AwesomePlayer: Improve performance on high-fps clips
- Immediately retry rendering next frame after frame skip.
- Schedule next videoEvent based on the timestamp of the
next frame.
Change-Id: Ia106382c4c225321b682c1f7c2d126d7eab7d56d
Signed-off-by: Lajos Molnar <lajos@google.com>
Bug:
11159147
Andreas Huber [Mon, 11 Nov 2013 21:35:39 +0000 (13:35 -0800)]
Ignore transport stream packets with the "transport_error_indicator" bit set
which indicates that they contain at least one bitstream error that was
unrecoverable.
Change-Id: I10665fd190194fdd2568175fe00b6063d3fc10a5
related-to-bug:
11537029
Andreas Huber [Tue, 12 Nov 2013 18:37:46 +0000 (18:37 +0000)]
Merge "Back to the old way of making sure that no more buffers are submitted" into klp-dev
Eino-Ville Talvala [Thu, 7 Nov 2013 23:13:16 +0000 (15:13 -0800)]
Camera API2: Add support for partial result quirk
- Camera3Device: Include requestId in partial 3A results
- FrameProcessorBase: Add support for partial result listeners
- CameraDeviceClient: Listen and forward partial results to client
Bug:
11115603
Change-Id: If1cc6bbc32932742ac142c058688761ebf6a8024
Andreas Huber [Tue, 12 Nov 2013 00:21:51 +0000 (00:21 +0000)]
Merge "configure bitrate and bitrate mode at encoder initialization for VP8/9" into klp-dev
Marco Nelissen [Wed, 6 Nov 2013 18:50:17 +0000 (10:50 -0800)]
Only initialize sniffers once, don't hold mutex while sniffing
b/
11482896
Change-Id: Ie036554b4c9e7e8863cb07901d414ea457f1b2f1
(cherry picked from commit
377fce69297a0e5b7c3b266a7602b17146785635)
Lajos Molnar [Mon, 11 Nov 2013 16:16:46 +0000 (16:16 +0000)]
Merge "StageFrightRecorder: do not use meta-data mode for software encoders" into klp-dev
Eric Laurent [Mon, 11 Nov 2013 00:21:16 +0000 (00:21 +0000)]
Merge "TimedEventQueue: improve wakelock management" into klp-dev
Andreas Huber [Fri, 8 Nov 2013 18:57:34 +0000 (10:57 -0800)]
Back to the old way of making sure that no more buffers are submitted
to the component while in executing->idle state.
Change-Id: Ia207f0a112025ed0e2c4d62f404afbbfa694cbab
related-to-bug:
11557463,
11573543
Andreas Huber [Fri, 8 Nov 2013 16:19:35 +0000 (16:19 +0000)]
Merge "Properly handle executing->idle in GraphicBufferSource" into klp-dev
Eric Laurent [Tue, 5 Nov 2013 00:08:15 +0000 (16:08 -0800)]
TimedEventQueue: improve wakelock management
Do not acquire a wakelock unless the event is delayed by more
than a given time. This prevents from acquiring/releasing
wakelocks too often which has adverse effects on performance.
Bug:
11509471.
Change-Id: Id06803b393b40c4db5095a6e7ec02339fa581a38
Andy McFadden [Thu, 7 Nov 2013 22:39:03 +0000 (14:39 -0800)]
Replace assert with error message
The failure happens when vendor code sends EMPTY_BUFFER_DONE
twice on the same buffer. The second time through, we see a slot
that has already been emptied, and get very confused.
This should be handled with an error, not an assertion, so we now
log a stern complaint and carry on.
Bug
11557463
Change-Id: I76e53ff08924bca81afb6980df8e59a6012ff8fe
Andreas Huber [Thu, 7 Nov 2013 19:13:27 +0000 (11:13 -0800)]
Properly handle executing->idle in GraphicBufferSource
ACodec has initiated the transition from exectuing->idle and is waiting
for the component to return all buffers so they can be freed.
It's counter productive and with some vendors harmful to submit additional
buffers to the component from this point on.
Change-Id: I332be254a1faf94902c77b5792a373ab447e44d3
related-to-bug:
11573543
Eino-Ville Talvala [Fri, 11 Oct 2013 16:51:09 +0000 (09:51 -0700)]
Camera HAL3/API1: Add support for the partial result quirk.
- Camera2Client:
- Detect partial result quirk
- Camera3Device:
- Accumulate all partial results together
- Fire off 3A-only result once all 3A states are available
- FrameProcessorBase:
- Filter out partials, don't send to listeners
- FrameProcessor:
- Skip face detect on partials
- Make sure to only handle a 3A update for a given frame once
- Trigger AF notifications when AF mode or trigger changes, to
properly detect all AF transitions.
Bug:
11115603
Change-Id: Iea8aa73c568701562a46071f7ea100624251d10b
Eric Laurent [Wed, 6 Nov 2013 17:29:58 +0000 (17:29 +0000)]
Merge "audioflinger: fix direct output underrun" into klp-dev
Eric Laurent [Wed, 6 Nov 2013 02:22:08 +0000 (18:22 -0800)]
audioflinger: fix direct output underrun
Underruns on tracks handled by a direct or offloaded
threads were not properly reported to the applications
causing a failure to automatically recover.
Bug:
11535001.
Change-Id: I7a8696b4d646f78b16710addfe7c108d0dd0038d
Andreas Huber [Tue, 5 Nov 2013 20:36:17 +0000 (12:36 -0800)]
configure bitrate and bitrate mode at encoder initialization for VP8/9
Change-Id: I0389264db4102cabb0c6fcb4ca223d85cc9e86fb
related-to-bug:
11536167
Marco Nelissen [Mon, 4 Nov 2013 17:42:11 +0000 (17:42 +0000)]
Merge "Notify current video size if available" into klp-dev
Lajos Molnar [Fri, 1 Nov 2013 23:17:47 +0000 (23:17 +0000)]
Merge "Restore NuPlayer error and EOS handling" into klp-dev
Marco Nelissen [Fri, 1 Nov 2013 17:33:18 +0000 (10:33 -0700)]
Notify current video size if available
b/
11248101
Change-Id: I0fa3b5592efcf400c86692002267da6825e331ad
Eric Laurent [Thu, 31 Oct 2013 22:29:50 +0000 (22:29 +0000)]
Merge "AudioFlinger: fix duplicating thread standby." into klp-dev
Eric Laurent [Wed, 30 Oct 2013 23:57:06 +0000 (16:57 -0700)]
AudioFlinger: fix duplicating thread standby.
Commit
fd47797 fixing issue
11247103 introduced a regression
where the duplicating thread does not exit standby state as
expected. This in turn prevents the tracks sending audio to
other threads to complete their activity cycle normally.
The fix conists in clearing mStandby state also in the specialized
threadLoop_write() method of DuplicatingThread subclass.
Bug:
11451415.
Change-Id: Ie8a92c819a56a2834ea25fa70a8b7b2125721775
Igor Murashkin [Wed, 30 Oct 2013 21:09:49 +0000 (21:09 +0000)]
Merge "camera: Collect stack traces when bad things happen, print with dumpsys" into klp-dev
Lajos Molnar [Mon, 28 Oct 2013 21:59:21 +0000 (14:59 -0700)]
Restore NuPlayer error and EOS handling
This was erroneously removed by commit
a73c954
Change-Id: I4742339139e770f0181c99d427875897fd60b68e
Signed-off-by: Lajos Molnar <lajos@google.com>
Bug:
11413439
Lajos Molnar [Wed, 30 Oct 2013 20:23:17 +0000 (20:23 +0000)]
Merge "AwesomePlayer: ensure STARTED message works after reset" into klp-dev
Lajos Molnar [Wed, 30 Oct 2013 17:02:12 +0000 (10:02 -0700)]
StageFrightRecorder: do not use meta-data mode for software encoders
Camera HAL v1 uses vendor-specific camera-source meta-data mode, but
software encoders use generic gralloc-source mode, so they are not
compatible.
Change-Id: I0d1bab8eada26dcb70c97b1eec83f06d1f331239
Signed-off-by: Lajos Molnar <lajos@google.com>
Bug:
11446505
Eric Laurent [Wed, 30 Oct 2013 01:13:52 +0000 (01:13 +0000)]
Merge "audioflinger: enable effects after registration if needed" into klp-dev
Igor Murashkin [Wed, 23 Oct 2013 23:40:06 +0000 (16:40 -0700)]
camera: Collect stack traces when bad things happen, print with dumpsys
Bug:
11324229
Change-Id: Ic920e170a8f5a389fc139bbf5416c0e3186c7fb6
Lajos Molnar [Mon, 28 Oct 2013 22:01:15 +0000 (15:01 -0700)]
AwesomePlayer: ensure STARTED message works after reset
Change-Id: I06217c07c0b68476cb296fd6151f6a0d8d4e02d0
Signed-off-by: Lajos Molnar <lajos@google.com>
Bug:
11412881
Eric Laurent [Sat, 26 Oct 2013 01:10:40 +0000 (18:10 -0700)]
AudioFlinger: more fixes for offload audio lost
mStandby being false was not a reliable indication that
something had been written to audio HAL.
Considering the last track in mActiveTracks vector as the
latest activated track was wrong as this is a SortedVector.
Bug:
11247103.
Change-Id: I397d011c5afcdd779def3fe25aaae7669a472cfc
Marco Nelissen [Mon, 14 Jan 2013 22:12:05 +0000 (14:12 -0800)]
Assign blame for playback wakelocks.
Set a work source for the playback wakelock, so that playback is
counted against the requesting app instead of the media server.
Cherrypicked from master.
b/
9464621
Change-Id: I7329f88a288a95a582a78005a1c3d16a5a611e31
Andy McFadden [Thu, 24 Oct 2013 17:35:43 +0000 (17:35 +0000)]
Merge "Check for NULL buffer in repeat-latest" into klp-dev
Andy McFadden [Wed, 23 Oct 2013 20:39:43 +0000 (13:39 -0700)]
Check for NULL buffer in repeat-latest
Repeating the most-recent buffer appears to fail when the producer
disconnects. This adds a NULL check to avoid crashing.
Bug
10607920
Change-Id: Ied036004425f046fcf9c26303823681f3fddb99e
Vineeta Srivastava [Wed, 23 Oct 2013 19:58:15 +0000 (19:58 +0000)]
Merge "Fix kAutoRampDurationUs overflow issue" into klp-dev
Glenn Kasten [Fri, 18 Oct 2013 22:02:46 +0000 (22:02 +0000)]
Merge "Allow releaseBuffer after flush" into klp-dev
Vineeta Srivastava [Thu, 17 Oct 2013 21:23:00 +0000 (14:23 -0700)]
Fix kAutoRampDurationUs overflow issue
When kAutoRampDurationUs multiplies with mSampleRate, it overflows to int64.
Type cast it to int64 to make sure RampDurationUs properly.
BUg:
11162491
Change-Id: I4f93bc9acc8456e25623a9255ca7a5b206425009
Glenn Kasten [Fri, 18 Oct 2013 20:00:29 +0000 (13:00 -0700)]
Allow releaseBuffer after flush
After AudioTrack start checks for pending flush,
allow releaseBuffer on any previously obtained buffer.
For example, this can happen if the resampler has obtained
a buffer but not released the whole buffer yet.
Note that the resampler will be reading obsolete data.
Bug:
11285590
Change-Id: I0614fbb62e43604aac3089cce4b7797c87a306b5
Eric Laurent [Fri, 18 Oct 2013 01:59:26 +0000 (18:59 -0700)]
audioflinger: fix track terminated before playing
When a track is terminated before any audio was written,
we should not send a drain request to the HAL. The drain
will be ignored and we will wait the async callback for ever.
Also fix a potential deadlock by not deleting a track with the
thread mutex held: mPreviousTrack changed from a sp to a raw pointer.
Bug:
11247103.
Change-Id: I0a4b47b099204c7c326d45d700234ab8f72d8eae
Glenn Kasten [Thu, 17 Oct 2013 23:12:05 +0000 (23:12 +0000)]
Merge "Fix flush() followed by start()" into klp-dev
Andreas Huber [Thu, 17 Oct 2013 20:24:42 +0000 (13:24 -0700)]
Make sure we get to discover that the cache is sufficiently full again
by posting buffering events while in the CACHE_UNDERRUN state.
Change-Id: Ic60c8d57f0ae502aa2b6eec7160dc4224b989604
related-to-bug:
11262689
Glenn Kasten [Thu, 17 Oct 2013 18:32:39 +0000 (11:32 -0700)]
Fix flush() followed by start()
Bug:
11259166
Change-Id: I72526cdfd4e3ff4b06e6a317e577f57cc285205d
Glenn Kasten [Thu, 17 Oct 2013 00:59:33 +0000 (00:59 +0000)]
Merge "Fix race condition in AudioTrack::pause followed by start" into klp-dev
Ed Heyl [Thu, 17 Oct 2013 00:42:20 +0000 (17:42 -0700)]
merge in klp-release (no-op)
Glenn Kasten [Thu, 17 Oct 2013 00:02:13 +0000 (17:02 -0700)]
Fix race condition in AudioTrack::pause followed by start
Bug:
11148722
Change-Id: Iec88f00c8510363d4418e4b8d5b34feb06ecf04d
Zhijun He [Wed, 16 Oct 2013 22:47:47 +0000 (22:47 +0000)]
Merge "Camera: don't do hw support check for ZSL" into klp-dev
Eric Laurent [Wed, 16 Oct 2013 22:28:33 +0000 (22:28 +0000)]
Merge "AudioTrack: fix head position after restore" into klp-dev
Eric Laurent [Wed, 16 Oct 2013 22:12:32 +0000 (15:12 -0700)]
AudioTrack: fix head position after restore
The head position transfered to the new track
by restoreTrack_l() must take into account the frames that
are dropped from the old track to avoid a non recoverable
offset in the playback head position returned to applications.
Bug:
11230062.
Change-Id: I51143a08b95e8f264ed709ae2054360315f2b8b1
Zhijun He [Wed, 16 Oct 2013 22:09:17 +0000 (15:09 -0700)]
Camera: don't do hw support check for ZSL
Only use camera.disable_zsl_mode to control disable/enable ZSL.
Bug:
11258054
Change-Id: Ibf10b959d0913f7dedb59d89b571e9bf66fe978a
Eric Laurent [Tue, 15 Oct 2013 23:11:24 +0000 (16:11 -0700)]
AwesomePlayer: fix concurrent pause and teardown
There was a race condition if a pause request was received
while we were executing a teardown event. Although we hold
a mutex while processing the teardown event, there is a step
of the prepare sequence where we release the mutex if streaming.
In this case, the pause request is executed but ignored because
the player state is still preparing. At the end of the teardown
event processing we restore previous playback state and resume.
The fix consists in clearing the saved teardown playback state
when a pause request is received while processing a teardown event.
Bug:
11225491.
Change-Id: If0e61855ce5a336322f1ba8e5559bdc190beeb76
The Android Automerger [Tue, 15 Oct 2013 13:00:42 +0000 (06:00 -0700)]
merge in klp-release history after reset to klp-dev
Narayan Kamath [Mon, 14 Oct 2013 14:03:38 +0000 (15:03 +0100)]
Make Audio wakelock names less unique.
These wakelocks now show up in the client process, where they
might cause wakelock overflows if AudioFlinger threads die often.
The client process should be agnostic of audio flinger threading,
so don't make the wakelock names rely on audioflinger thread names.
Wakelock names now depend only on the audio flinger thread type, and
not the ID itself.
bug:
11200551
Change-Id: Ia1f71b569cb7502d617b77596abc245f461a6ec0
The Android Automerger [Mon, 14 Oct 2013 02:20:21 +0000 (19:20 -0700)]
merge in klp-release history after reset to klp-dev
The Android Automerger [Sun, 13 Oct 2013 13:01:45 +0000 (06:01 -0700)]
merge in klp-release history after reset to klp-dev
Eric Laurent [Sun, 13 Oct 2013 00:05:19 +0000 (17:05 -0700)]
SoundPool: handle new audio track event
If the AudioTrack is torn down, SoundPool will never
receive the buffer end event and the track will stay active
for ever.
The fix consists in stopping the AudioTrack when a new audiotrack
event is received.
Bug:
11193583.
Change-Id: I9876eb2a8f75c601368f669acd67b0accf6e2736
Eric Laurent [Sat, 12 Oct 2013 23:17:46 +0000 (16:17 -0700)]
audioflinger: enable effects after registration if needed
This fixes a bug where effects were not properly reenabled after
being moved from one playback thread to another. The effect
is enabled but the audio policy manager sees it as disabled.
Bug:
11181933.
Change-Id: I19cac7acbaf61c546e667fd85ed7d4eda1c716d8
Zhijun He [Fri, 11 Oct 2013 23:21:11 +0000 (16:21 -0700)]
Camera: Fix deadlock in Camera3OutputStream
process capture request thread, process capture result and setParameter
binder threads can run into circular locking situation when acquiring
StreamingProcessor lock, Camera3Stream lock, and bufferQueue lock.
Releasing the Camera3Stream lock briefly in process capture request
thread getbuffer call can break this deadlock.
Bug:
11016037
Change-Id: If08d4b134c26be26039b1d5363759e60f911bad6
Zhijun He [Fri, 11 Oct 2013 17:26:52 +0000 (10:26 -0700)]
Camera: Skip AE precapture when possible
Skip AE precapture when AE is already converged for still capture use case. This
could save still capture latency 6-7 frame time for AE already converged case.
Bug:
10712902
Change-Id: Ie5512618b76e5d87c62c57c3d96d2004c604e29e
Zhijun He [Fri, 11 Oct 2013 19:22:15 +0000 (19:22 +0000)]
Merge "Camera: Skip AE precapture when possible" into klp-dev
Eino-Ville Talvala [Fri, 11 Oct 2013 19:21:51 +0000 (19:21 +0000)]
Merge "Camera2/3: Add more tracing for API1" into klp-dev
Zhijun He [Fri, 11 Oct 2013 17:26:52 +0000 (10:26 -0700)]
Camera: Skip AE precapture when possible
Skip AE precapture when AE is already converged for still capture use case. This
could save still capture latency 6-7 frame time for AE already converged case.
Bug:
10712902
Change-Id: Ie5512618b76e5d87c62c57c3d96d2004c604e29e
Eino-Ville Talvala [Thu, 10 Oct 2013 22:13:09 +0000 (15:13 -0700)]
Camera2/3: Add more tracing for API1
- AF state
- Async autofocus completion
- Async takePicture completion
- Stages of takePicture
Bug:
10570497
Change-Id: Ida9fedd81aa4ee3ae3fb8dfada858a3bc3c213a3
The Android Automerger [Thu, 10 Oct 2013 13:01:02 +0000 (06:01 -0700)]
merge in klp-release history after reset to klp-dev
Zhijun He [Thu, 3 Oct 2013 23:12:43 +0000 (16:12 -0700)]
Camera: Save 1 frame time for takePicture call
Move the latest request signal call right before submitting request to HAL,
which avoids unnecessary 1 frame time wait.
Also fix request status constructor issue, and the log tag for Camera2Client.
Bug:
10570497
Change-Id: I0b1818658bee9983b117e49a5a00f1aab515c567
The Android Automerger [Wed, 9 Oct 2013 13:01:03 +0000 (06:01 -0700)]
merge in klp-release history after reset to klp-dev
Ed Heyl [Tue, 8 Oct 2013 22:59:50 +0000 (15:59 -0700)]
merge in KFS78N (no-op)
Igor Murashkin [Tue, 8 Oct 2013 22:57:32 +0000 (22:57 +0000)]
Merge "camera2: Don't race while dumping last frame metadata" into klp-dev
Ed Heyl [Tue, 8 Oct 2013 22:40:01 +0000 (15:40 -0700)]
merge in KQS81M
Eino-Ville Talvala [Tue, 8 Oct 2013 18:46:21 +0000 (18:46 +0000)]
Merge "Camera3: Improve resilience to stream failures" into klp-dev
Eino-Ville Talvala [Wed, 25 Sep 2013 01:04:19 +0000 (18:04 -0700)]
Camera3: Improve resilience to stream failures
- Handle binder connection drops (DEAD_OBJECT) for streams
- Just log an error, don't go to an error state when queue/dequeue fails
Bug:
10347526
Change-Id: Ib463ffe15d58db444cf5d8cad176a201c7d1facc
Eric Laurent [Tue, 8 Oct 2013 16:22:31 +0000 (09:22 -0700)]
AwesomePlayer: do not send events when paused.
When streaming audio and paused, AwesomePlayer should
stop sending BufferingEvents as they will keep a wake lock
for no reason.
TimedEventQueue should always acquire the wakelock with mediaserver
identity so that it is released with the same identity by the
event handler thread.
Bug:
11104408.
Change-Id: Ied0e03acd6ad2f5a4c0ec82d5c2aa4e1c6da772c
Igor Murashkin [Tue, 8 Oct 2013 01:01:46 +0000 (18:01 -0700)]
camera2: Don't race while dumping last frame metadata
Bug:
11095203
Change-Id: Icfb31e1719634b62004d6c15a95a9316e9642e4c
Glenn Kasten [Tue, 24 Sep 2013 18:52:37 +0000 (11:52 -0700)]
Cleanup openRecord error handling
Bug:
10888816
Change-Id: I84897dd7d30b370640b54e928f230604b873cb68
Glenn Kasten [Mon, 7 Oct 2013 16:16:06 +0000 (16:16 +0000)]
Merge "Cleanup openRecord error handling" into klp-dev
The Android Automerger [Sun, 6 Oct 2013 13:01:09 +0000 (06:01 -0700)]
merge in klp-release history after reset to klp-dev
Eric Laurent [Fri, 4 Oct 2013 23:23:48 +0000 (16:23 -0700)]
audioflinger: offload: fix pause/flush/resume
If a pause/flush/resume sequence is fast enough, resume is received while
we are still in PAUSING state in which case it is a NOP. If this happens,
flush is still forwarded to the audio HAL but is not preceeded by a pause
which can cause problems to the audio DSP.
It is necessary to preserve the flush as this sequence is typical to a seek.
The fix consists in forcing a pause/resume when a flush request must be
executed and the audio HAL has not been paused previously.
Bug:
11081559.
Change-Id: Ib84ed26d503a61c05933b923ec556b10cedfe140
Eric Laurent [Fri, 4 Oct 2013 00:54:26 +0000 (17:54 -0700)]
TimedEventQueue takes a wake lock
Take a wake lock when events present in a TimedEventQueue
to prevent from going idle before all delayed events are
processed.
Bug:
11057387.
Change-Id: I26a17df68068fde5e879a2fe7568dec439fc540f