I missed that there is already android.media.audio.common.AudioMode.
However, due to use by the framework, it contains more values
(all of them are SYS_RESERVED), thus the implementation and VTS
have to be updated as well.
Bug: 205884982
Test: atest VtsHalAudioCoreTargetTest
Change-Id: If260413b1f774a69514de978792d5ad2eb46481a
Merged-In: If260413b1f774a69514de978792d5ad2eb46481a
This allows for more code reuse and composability when
implementing streams for a particular audio "backend."
The existing "stub" code has been moved to StreamStub* files.
Bug: 264712385
Test: atest VtsHalAudioCoreTargetTest
Change-Id: I97fd41f87eb6d01e1d57f0d70a86d3b2b3555837
The plumbing was missing from the CL which has added
the callback (aosp/2256190).
Bug: 205884982
Test: atest VtsHalAudioCoreTargetTest
Change-Id: Iab3c970b3efd0281e34e07f98864531acc4c519e
* changes:
audio: Allow going to 'IDLE' for synchronous drain
audio: Improve test coverage
audio VTS: Refactor support for non-deterministic SM behavior
For small buffers, the driver can perform draining
synhronously, returning control to the HAL only after
the buffer is empty. This makes going through
the 'DRAINING' state artificial. Thus, we allow going
to the 'IDLE' state directly.
In order to make sure that VTS handles both transitions:
to 'DRAINING' and to 'IDLE', correctly, add an "AOSP as
vendor" parameter "aosp.forceSynchronousDrain" to induce
this behavior in the default implementation.
Bug: 262402957
Test: atest VtsHalAudioCoreTargetTest
Change-Id: Ic8eaee53cb4596afb5317b4b905e004af3f112aa
Add a vendor-specific parameter "aosp.forceTransientBurst"
which helps to cover the case of non-full async writes.
This parameter is specific to the "AOSP as vendor" implementation,
other vendors are not required to have it—that's why it's
not in AIDL.
Fix minor issues in VTS found after revisiting the code, and by
looking at logs during testing.
Bug: 262402957
Test: atest VtsHalAudioCoreTargetTest
Change-Id: Ide961d91a8d9da24392561654f04eb8207b7b781
Use a DAG instead of a list for representing stream state machine
transitions. This allows accomodating multiple possible replies
from the stream state machine, and enables following different,
possibly converging test sequences depending on actual replies.
Update the async output state machine graph including possible
state transitions that are implemented but were missing from the
graph.
Bug: 262402957
Test: atest VtsHalAudioCoreTargetTest
Change-Id: Ie97f3ce9222eec812d4eb4dfbce1f678370bef26
This interface corresponds to BT-specific functions
of the IPrimaryDevice in the HIDL HAL.
Bug: 205884982
Test: atest VtsHalAudioCoreTargetTest
Change-Id: I5f52d65e06f42e1a5fb2e51d27b79e94fe423eaa
Fix the issue pointed out by gkaiser@ on aosp/2256190.
I also went over other usages the logic which checks for
methods support, and made the code more consistent.
Bug: 205884982
Test: atest VtsHalAudioCoreTargetTest
Change-Id: I1b06c1fd9dddcbdadfd6e04525493a049283f31f
Use 'gtid' as the command, returns the worker
thread tid.
Test: see bug
Bug: 253276925
Merged-In: Ia4fa129c62cf6df97821c34876b517944bc5322f
Change-Id: Ia4fa129c62cf6df97821c34876b517944bc5322f
This refactoring is fixing the hal_implementation_test workaround from
aosp/2363076. Adjusted the package name to match the aidl_interface and
declared the ISoundDose interface types only.
Test: VtsHalSoundDoseFactoryTargetTest, VtsHalAudioCoreTargetTest,
hal_implementation_test
Bug: 263388737
Change-Id: Ibce4cd95aaf600de2096179f137243939114ba6a
Merged-In: Ibce4cd95aaf600de2096179f137243939114ba6a
Use setMinSchedulerPolicy to enforce minimum thread priority
when accessing the audio HAL to ANDROID_PRIORITY_AUDIO
Test: instrumented
Bug: 261496726
Merged-In: I700b04d3060e9405b1445206882a6131cea0d045
Change-Id: I700b04d3060e9405b1445206882a6131cea0d045
