| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
CameraHardwareInterfaceFlashControl class calls disconnectCameraDevice
when torch is disabled. This closes connection to the camera module,
but mDevice instance is kept and variable is non-NULL which will
prevent connection next time torch is going to be enabled.
Change-Id: Icb1ffb07f05256afd92821f0f4908cda5332c05b
|
|\ \ \
| | | |
| | | |
| | | |
| | | |
| | | | |
git://codeaurora.org/platform/frameworks/av into cm-13.0
Change-Id: Ia41df61c3ccfeb809572b63a4b1a8ca6bd85dfb2
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Synchronize public APIs of EffectDescriptorCollection
CRs-Fixed: 920103
Change-Id: I04ccac526c6f99e61e43288776653d6b7ff325c4
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Fixe race condition in AudioCommandThread::threadLoop() where a
command can be inserted in first position in the queue after the sleep
time has been calculated causing a longer delay than expected.
Also fix a failure to hold a wake lock while commands are still in the queue.
Bug: 22707905.
Change-Id: I813626986677bf00106acb37ee20d3dd75d5cf33
|
| |\ \ \ |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Synchronize access to APM when getDevicesForStream is called
on APM.
CRs-Fixed: 913227
Change-Id: I2ba6922341f035375270b02000ef5a7e078f6b5a
|
| |\ \ \ \
| | |/ / /
| |/| | | |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
--Added NULL check for mRemoteCallback
Change-Id: Ib88a128a52e81b8ec1052e3222b6d8b9e494afcc
|
| |\ \ \ \
| | |/ / /
| |/| | |
| | | | | |
connect"
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
If, within a binder Looper thread, a Binder object death notifier is
registered, and then a nested Binder transaction is begun, that
transaction may process a death notification callback for the
just-registered object, if it was dead at the time of registration.
This can lead to an unexpected nested call into the service, and cause
deadlock.
To avoid this, move the death notifier registration to the end of
handling the connect transaction.
Also remove one extra bit of disconnect logging.
Bug: 23525545
Change-Id: If01cbaf42704f55134118afefc9a8f7bdb014e09
|
| |/ / /
| | | |
| | | |
| | | |
| | | |
| | | | |
Add missing __FUNCTION__ arguments into log statements.
Change-Id: I0256bad626d490fb9360a73c104afaad7a012a97
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: I774f75b493c47386ca1eaf004d663432f1041a66
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
- Send USB connected/disconnected events to Audio HAL
only in case of USB AOA as these are used only in
case of USB AOA
Change-Id: I7db508f6eff5e8601d54f1e1bfa603358f9d6703
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
setPhoneState calls getDeviceForInputSource to check if
voice_tx device is available as part of Primary Audio
Module. getDeviceForInputSource always returns
DEVICE_IN_COMMUNICATION for SOURCE_VOICE_COMMUNICATION
on A Family and this is not added to Available devices
This is leading to creation of txPatch when it is not
needed
Fix is to use AUDIO_SOURCE_VOICE_CALL instead of
AUDIO_SOURCE_VOICE_COMMUNICATION
Change-Id: I9aa7dd2139b8374ec7f06676d06ef5addea4b2a9
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
- On A family, Only DEVICE_IN_COMMUNICATION is supported
by voip_tx profile. getDeviceForInputSource should
always return DEVICE_IN_COMMUNICATION to load
voip_tx profile
Change-Id: I3bb2fb70a89eadc35c2ced0c7f717ec15120bfe7
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
USB AOA is supported by primary HAL on A family through
customized connect/disconnect parameters.Added support
for that in AudioPolicyManager
Change-Id: I5212027b2711c24a9750629774f69d669a9d164f
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Added entry for DEVICE_IN_COMMUNICATION to support
VOIP usecase.
Change-Id: Ie7e5131b776d7b93c53ca0796dffa11316091805
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Currently, audio doesn't play on Brillo because the policy manager's
default value for all streams is set to 0. In Android, this value is
changed by calls from AudioService.java. If we set the default value to
1 in Brillo, then even if there are no calls, we will have audio.
BUG=none
TEST=manual
Change-Id: I1418c9ecfb4f1c81c9c0946eee11deed3c65b358
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Synchronize access to APM when getDevicesForStream is called
on APM.
Change-Id: I2ba6922341f035375270b02000ef5a7e078f6b5a
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This change prevents a crash in the camera service when the camera
HAL notifies the service about the absence of a removable camera and
there happens to be no client connected to the service. It checks
that the pointer returned from clientToDisconnect.get() is non-null
before trying to dereference it (as is done in existing code
immediately below this change).
Change-Id: I8055654bac980542e63ea7f52bf897eaafbc09bc
Signed-off-by: Tom Keel <thomas.keel@intel.com>
|
|\ \ \ \
| | |/ /
| |/| |
| | | |
| | | |
| | | | |
Android 6.0.0 release 26
Change-Id: I8a57007bf6efcd8b95c3cebf5e0444345bdd4cda
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Based on periods of the request thread and audio threads with
SCHED_FIFO policy, 1 is a more reasonable priority for HFR
request thread.
Bug: 24427480
Change-Id: I91f0066a0e114fc83abcc6a604ecbaa72c6a34e8
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Bookkeeping reprocess shutters separately so regular and reprocess
shutters together don't need to come in order.
Bug: 24497512
Change-Id: I4aaf22045131e9e2e26bf163f7df9ff4c5cd6259
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
- Move SchedulingPolicyService from audioservice to mediautils
- When starting up a high speed stream config, set request queue thread
to SCHED_FIFO using SchedulingPolicyService
Bug: 24227252
Change-Id: I224b59142bd111caf563779f55cddd62385b9bac
|
| |\ \ \ |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Because a preemtible capture session can preempt another one,
we end up in an endless loop situation were each session is allowed to restart
after being preempted, thus preempting the other one which restarts and so on.
To avoid this situation, we store which audio session was preempted when
a particular input started and prevent preemption of this active input by this session.
We also inherit sessions from the preempted input to avoid a 3 way preemption loop etc...
Bug: 24007220.
Change-Id: I0eab5299440ef3ab9e987635dc9a300cf42f2c79
|
| |/ / /
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Fixe race condition in AudioCommandThread::threadLoop() where a
command can be inserted in first position in the queue after the sleep
time has been calculated causing a longer delay than expected.
Also fix a failure to hold a wake lock while commands are still in the queue.
Bug: 22707905.
Change-Id: I813626986677bf00106acb37ee20d3dd75d5cf33
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
No routing operation should be allowed between setMode() and setPhoneState()
when starting a call as the audio HAL relies on a precise sequence of mode change
and routing change to select the initial audio device for the call.
Bug: 24083591.
Change-Id: I2d5ef62c11cf7aedc2ec7ca5e5fadd7ac875afbc
|
| |\ \ \ |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Only change the volume for AUDIO_STREAM_ACCESSIBILITY when
changing the media volume.
Bug 23366575
Change-Id: Ic42c726ec4d47615c20500c20e4d43cef159b3c6
|
| |\ \ \ \
| | |/ / /
| |/| | | |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Bug: 23992350
Change-Id: Ie86b24d9e5ea753c6dfc2e5526d371b339be96a1
|
| |\ \ \ \ |
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Always play Transmitted Through Speaker stream at max volume instead
of tracking media stream volume.
Bug: 23729461.
Change-Id: I61021f59617bb851a4cd7da80924e587029e96a3
|
| |\ \ \ \ \ |
|
| | | |/ / /
| | |/| | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Signal buffer returned even after it failed so the thread waiting
for it can wake up sooner.
Bug: 23981045
Change-Id: Iccbcc7ece2e0f6204da9c54f2bdd96ff6843a8f5
|
| |\ \ \ \ \ |
|
| | | |/ / /
| | |/| | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Bug: 23291988
Bug: 23614327
Bug: 23924081
Change-Id: Id1a519ed4bb2a6f4cb197da8450f7069b55c0d48
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Bug: 23999740
Change-Id: I3165404add455528b22951ba47b682a075b50087
|
| | |/ / /
| |/| | |
| | | | |
| | | | |
| | | | | |
Bug: 23940813
Change-Id: I0060dbaff5cd0c3d028ac5cd7aba7da3f0371150
|
| |/ / /
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
The audio HAL wakes up and configures the audio path when receiving
the first write() in standby state. This causes a certain amount of
process to take place in the mixer threads which is problematic for
fast mixer running at FIFO priority.
We now force a fake write() of 0 bytes from normal mixer to trigger
the audio path configuration before starting the fast mixer.
Bug: 23791972.
Change-Id: I54311b337fda956444846f5d2f53a3263d54e04b
|
| |\ \ \
| | | | |
| | | | |
| | | | | |
into mnc-dr-dev
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
This reverts commit cafe86a9cb6625bb1ec6383e16e28e4c9e455f87.
Bug: 23924093.
Change-Id: I186d1013b06a286eca93c30bb9b3545dc36695ff
|
| |/ / /
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Improve debugging of performance of concurrent calls to the service.
- Add ATRACE calls to camera service entry points
- Add ATRACE calls for camera HAL module calls
- Ensure that camera module init method is called before all others
Bug: 23718257
Bug: 23933447
Change-Id: I2a025866ab39eda3962877120cab13a063c7727e
|
| |\ \ \ |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Bug: 22938515
Change-Id: I1de653de169a3fbbaa693da6057897ea57772447
|
| |\ \ \ \
| | | | | |
| | | | | |
| | | | | | |
mnc-dr-dev
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
If, within a binder Looper thread, a Binder object death notifier is
registered, and then a nested Binder transaction is begun, that
transaction may process a death notification callback for the
just-registered object, if it was dead at the time of registration.
This can lead to an unexpected nested call into the service, and cause
deadlock.
To avoid this, move the death notifier registration to the end of
handling the connect transaction.
Also remove one extra bit of disconnect logging.
Bug: 23525545
Change-Id: If01cbaf42704f55134118afefc9a8f7bdb014e09
|
| |\ \ \ \ \
| | |/ / / /
| |/| | | | |
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Make the Vector of next requests a RequestThread member variable
to avoid memory allocation in every threadloop.
Bug: 23360060
Change-Id: I4f33e5c49f0f4deb1f9f45bada0909da748849e4
|