Volume used to be managed per stream type.
VolumeGroup is a configurable way to manage volume affinity.
This CL allows to switch to VolumeGroup as new volume affinity management.
Bug: 124767636
Test: audio smoke tests
Change-Id: I71fc214f6db3158f0f05920cc3d700b29db1a4bc
Signed-off-by: François Gaffie <francois.gaffie@renault.com>
AudioInputDescriptor and AudioOutputDescriptor used to reference
AudioMix instances using a raw pointer. This isn't safe as AudioMix
was owned by AudioPolicyMix, which is not referenced by descriptors.
Change AudioMix* pointers in Audio{Input|Output}Descriptor to
wp<AudioPolicyMix> which reflects their relationship correctly.
To ensure that code does not operate on AudioMix instances
independently from AudioPolicyMix, and to avoid introducing
a lot of getter / setter methods into AudioPolicyMix, make
the latter to inherit AudioMix. This makes sense because
AudioPolicyMix is essentially a ref-counted version of AudioMix.
Bug: 124899895
Test: build and sanity check on crosshatch,
build crosshatch with USE_CONFIGURABLE_AUDIO_POLICY := 1
Change-Id: Ic508caedefe721ed7e7ba6ee3e9175ba9e8dc23a
The code for clearing software patch connections was missing
a check for a null track pointer.
Bug: 126789266
Bug: 128269566
Test: see repro steps in b/128269566
Change-Id: Ice6887622d5fa2fa3198ce15146bff3cb05f7451
This partially reverts ag/6327294.
Since libstagefright_xmlparser is no longer used by libmediandk,
double-loading can't happen with this lib.
Bug: 124268753
Test: m -j
Change-Id: I76c961b2ff03665b9211c3fc8cbdbb073daddbe1
These are `new`ed, but never passed to anything else. It looks like
simply stack allocating will do what we want here.
Caught by our static analyzer:
frameworks/av/media/libmediaplayer2/MediaPlayer2AudioOutput.cpp:524:5:
warning: Potential memory leak [clang-analyzer-cplusplus.NewDeleteLeaks]
frameworks/av/media/libmediaplayer2/MediaPlayer2AudioOutput.cpp:527:9:
warning: Potential memory leak [clang-analyzer-cplusplus.NewDeleteLeaks]
frameworks/av/media/libmediaplayer2/MediaPlayer2AudioOutput.cpp:529:9:
warning: Potential memory leak [clang-analyzer-cplusplus.NewDeleteLeaks]
Bug: None
Test: Ran the analyer again. It was happy.
Change-Id: I8f4cda4a5993f43ad50cf5a0d152ed7158d7c8e1
am: 14abfe53be -s ours
am skip reason: change_id I5c3abcd72e64b40e7cbac467646d99a1f5b362d0 with SHA1 38c13a703b is in history
Change-Id: I46e527a2690599830073d4861579af96cecfaf73
* changes:
codec2: VTS add support for Empty EOS in AudioDectest
codec2: VTS fix Master test for multiple IComponent services
codec2: VTS add support for nSamplesPerFrame in AudioEnctest
codec2: VTS add support for Empty EOS in VideoEnctest
codec2: VTS add support for Empty EOS in VideoDectest
codec2: VTS add support for Empty EOS in AudioEnctest
VTS: Fix DecodeTest For Audio Decoders.
1. support MPEG2 video codec
2. support codecs under V_MS/VFW/FOURCC which include H263, AVC, MPEG2,
MPEG4
3. handle some codec type with no codec private data
4. add codec type "MJPEG, XVID, DIVX, DIVX3" and support them in
matroskaextractor
Bug: 123594230
Test: test with file which has added codec type, and check if the codec
related track can be played normally
Change-Id: I4e0173b80576b1f80bc7a210a8ef029ee3a2af51
Fix the long (>1s) dequeueBuffer call when a stream is managed by
Camera3BufferManager and its consumer end discards free buffers.
Test: CTS, no more long dequeBuffer call in GCA mode switch
Bug: 126054873
Change-Id: I03d6526b076796bb44f15cc2c4a092ff3d04fc1d
Eino-Ville Talvala
Jayant Chowdhary
Eric Laurent
TreeHugger Robot
François Gaffie
Dichen Zhang
Mikhail Naganov
Xin Li
The Android Open Source Project
Jooyung Han
George Burgess IV
Yin-Chia Yeh
Sungtak Lee
Iris Chang
Robert Shih
Andy Hung
Marco Nelissen
android-build-team Robot
Lajos Molnar
Ray Essick
Chong Zhang