Or the activity remained in PAUSING state and prevents the
activity to stop after screen off until stop timeout.
Bug: 202345179
Test: atest PinnedStackTests
Change-Id: Ibb8b3c48d483dcb96207ec2395a87b31d8eb107e
Merged-In: Ibb8b3c48d483dcb96207ec2395a87b31d8eb107e
(cherry picked from commit e84bd0d141)
- When swiping up to enter PIP w/ auto-enter, SysUI applies a transform
on the leash which is seamlessly applied to the task (to prevent jumps
when the leash is removed). This was intended to be applied in both
single and multiple activity PIP cases, but if that doesn't happen,
then when the transform is later cleared, it will inadvertently reset
the task's crop and matrix.
Bug: 201127015
Test: atest PinnedStackTests
Test: Open YT, enter PIP, turn screen off, turn screen back on
Change-Id: I0d6d46c7c52f7a43058902b50a26131f626ae92d
The advanced extension session request processor must not hold the
'mInterfaceLock' when calling the camera device implementation as it
could result in a deadlock with the camera device close that can
run at the same time and try to acquire the same locks in reverse order.
The request processor will only read from 'mCameraConfigMap' which is
initialized only once during session setup and will not change during
the session lifetime.
A sample stack trace will look like this:
Service thread:
1) android.hardware.camera2.impl.CameraCaptureSessionImpl.stopRepeating
2) android.hardware.camera2.impl.CameraAdvancedExtensionSessionImpl$RequestProcessor.stopRepeating
Client thread:
1) android.hardware.camera2.impl.CameraAdvancedExtensionSessionImpl.release
2) android.hardware.camera2.impl.CameraDeviceImpl.close
Bug: 202074988
Test: Manual using application,
Camera CTS
Change-Id: If992667a7d7e54be5bea1fcc4b1d71ec1b895ca0
CaptureFailure is created with a "dropped" parameter. But what's passed
in is mayHaveBuffers, which is opposite of "dropped".
Test: Camera CTS, GoogleCamera
Bug: 201967385
Change-Id: I42eb2daa78f116502cb1c5943015ea3f4af3f1e3
Under stress tests, it's possible for several messages to be sent before
CHRE has time to respond to the permission query which can cause several
permission queries to be sent and CHRE to be overwhelmed.
Fixes: 202201157
Test: Flash device and run stress test
Change-Id: I16871a116216cd288f0eba0e6c5572f5afa6b959
For unknown reasons these changes are causing a memory usage increase
affecting Go devices. This reverts commit
6eecc3ce91.
Bug: 201306859
Test: presubmits
Change-Id: Ifd820ead3099c61abf7403442219f3d1172b8505
This change resets the brightness short term model when RBC is turned
off or the strength is changed. We add a singular interaction when RBC is turned on, and add
interactions in RBC as normal.
Since users turning RBC in a bright environment could want any
brightness between the given one and MAX, but turning RBC off will often result in
a MAX interaction being logged, if the slider is already at MAX (given
RBC) which is not necessarily what we want.
In order to prevent unwanted changes to the brightness curve, we ignore
interactions that turn RBC off. Turning RBC on seems like a valid
indicator that the user wants the brightness darker, and at a specific
brightness, however turning RBC off is less of an indicator. Changing
the strength of RBC is also a good time to reset the short term model,
however after resetting, we still add an interaction after this.
Bug: 199187829
Test: manual
Change-Id: I6512d9642f2da634aa60af6b5573c792c0a116d5
It is possible the developer calls pauseTranslation() to show the
original text but it calls startTranslation() to show translated
text. Ideally the developer should call resumeTranslation but it
also make sense to call startTranslation() to show translated text.
When receiving translation response, we avoid showing transaltion
if the view already has the response and it's the same. But it is
good to also check if the view is showing translated text or not.
If the view is not translated text, it is possible developer calls
startTranslation() instead if calling pauseTranslation() to show
translated again, the fixing can resolve this case.
The issue case can be fixed by this change. But there is a deeper
problem about it's useless the caller call finishTranslation(). This
is planned to be fixed in next release.
Bug: 201238016
Test: manual. The issue case is fixed.
Test: manual. Test some chat apps, it still works fine.
Test: atest CtsTranslationTestCases
Change-Id: I699d0fa1d60ac96db094adcc6e17f4203df03214
This was broken by ag/15974634, but the presubmits did not catch that for some reason. That CL changed the brightness logic to be <= the dim value, so this failure (and fix) make sense.
Test: atest SystemUITests, since apparently treehugger won't do that and will +2 this anyway
Fixes: 202212576
Change-Id: Ibe18322fbe1e1fbae86c5274615ad0b6f727c8e2
