Currently, if a network goes from CONNECTED to some other "live"
state (e.g., CONNECTING, because it's VERIFYING_POOR_LINK) and
back, ConnectivityService treats it as if a new network had
connected. This causes it to attempt to create the network
(which fails, since a network with that netid already exists), to
trigger verification, and if the verification succeeds, to tear
down the network because the request it's satisfying is already
satisfied by the network itself.
Instead, if creating the network fails, assume it's because the
network had already been created, and bail out.
Also, when validation completes, ignore NetworkRequests that were
being served by the same NetworkAgent as before.
Bug: 15244052
(cherry picked from commit cfff026ec47afc7e31f60f80e3deea7f4e2f9be5)
Change-Id: I52c2220e8f1d98fca765880be3040593e92722ed
Issues here:
"Reschedule" of an idle-mode task is not well-defined. In the
API I throw an error if you try to set a back-off policy on
an idle mode task.
Implementation-wise, i add a delay for a reschedule request of an
idle mode task. This means that if the phone's still in idle mode
after the delay they app will get a call back, but otherwise it'll
have to wait til the next one.
Implemented all API functions
New window layer that voice interaction service windows
go in to. Includes a new voice-specific content rectangle
that voice activities are placed in to.
Add specific animations for this layer, sliding down from
the top (though this can be customized by the voice interaction
service).
Also add the concept of activities running for voice interaction
services for purposes of adjusting the animation used for them,
again sliding from the top, but not (yet?) customizable by the
voice interaction service.
Change-Id: Ic9e0e8c843c2e2972d6abb4087dce0019326155d
This is making a HH device (and probably others) in a reboot loop
This reverts commit 0cac71e9bdee1e9e6b2faafec0f9f894effbcb67.
===
See:
W/dalvikvm( 2320): threadid=1: thread exiting with uncaught exception (group=0x952d3f28)
E/AndroidRuntime( 2320): *** FATAL EXCEPTION IN SYSTEM PROCESS: main
E/AndroidRuntime( 2320): java.lang.RuntimeException: Failed to create service com.android.server.am.ActivityManagerService$Lifecycle: service constructor threw an exception
E/AndroidRuntime( 2320): at com.android.server.SystemServiceManager.startService(SystemServiceManager.java:89)
E/AndroidRuntime( 2320): at com.android.server.SystemServer.startBootstrapServices(SystemServer.java:304)
E/AndroidRuntime( 2320): at com.android.server.SystemServer.run(SystemServer.java:244)
E/AndroidRuntime( 2320): at com.android.server.SystemServer.main(SystemServer.java:161)
E/AndroidRuntime( 2320): at java.lang.reflect.Method.invokeNative(Native Method)
E/AndroidRuntime( 2320): at java.lang.reflect.Method.invoke(Method.java:515)
E/AndroidRuntime( 2320): at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:836)
E/AndroidRuntime( 2320): at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:631)
E/AndroidRuntime( 2320): at dalvik.system.NativeStart.main(Native Method)
E/AndroidRuntime( 2320): Caused by: java.lang.reflect.InvocationTargetException
E/AndroidRuntime( 2320): at java.lang.reflect.Constructor.constructNative(Native Method)
E/AndroidRuntime( 2320): at java.lang.reflect.Constructor.newInstance(Constructor.java:423)
E/AndroidRuntime( 2320): at com.android.server.SystemServiceManager.startService(SystemServiceManager.java:78)
E/AndroidRuntime( 2320): ... 8 more
E/AndroidRuntime( 2320): Caused by: java.lang.NullPointerException
E/AndroidRuntime( 2320): at com.android.server.am.ActivityStackSupervisor.isLeanbackOnlyDevice(ActivityStackSupervisor.java:3508)
E/AndroidRuntime( 2320): at com.android.server.am.ActivityStackSupervisor.<init>(ActivityStackSupervisor.java:220)
E/AndroidRuntime( 2320): at com.android.server.am.ActivityManagerService.<init>(ActivityManagerService.java:2145)
E/AndroidRuntime( 2320): at com.android.server.am.ActivityManagerService$Lifecycle.<init>(ActivityManagerService.java:2073)
E/AndroidRuntime( 2320): ... 11 more
E/AndroidRuntime( 2320): Error reporting crash
E/AndroidRuntime( 2320): java.lang.NullPointerException
E/AndroidRuntime( 2320): at com.android.internal.os.RuntimeInit$UncaughtHandler.uncaughtException(RuntimeInit.java:89)
E/AndroidRuntime( 2320): at java.lang.ThreadGroup.uncaughtException(ThreadGroup.java:693)
E/AndroidRuntime( 2320): at java.lang.ThreadGroup.uncaughtException(ThreadGroup.java:690)
E/AndroidRuntime( 2320): at dalvik.system.NativeStart.main(Native Method)
I/Process ( 2320): Sending signal. PID: 2320 SIG: 9
I/ServiceManager( 174): service 'power' died
I/ServiceManager( 174): service 'sensorservice' died
Change-Id: Ib90ae4dff6542d67f5827b100a3ab6158ce88ae2
- For Leanback only devices we will force all activities to
live in the same app stack. This is a design decision for the
shy/gregarious changes we are planning to implement for
leanback devices.
Change-Id: I201f56541ba22356e9598f09419ad41e588c74dc
Introduce IWindowManager.keyguardGoingAway to notify that Keyguard
wants to dismiss it self. This method starts the state machine in
WindowAnimator which animates in the activity behind the keyguard.
Animating out the keyguard is done by the StatusBar/Keyguard
software when it receives the startKeyguardExitAnimation() callback.
Bug: 14118756
Change-Id: Id3b8f41189410bad808b4892fbec74245e59efce
ConnectivityService doesn't do this anymore.
bug:15077247
Change-Id: I3208c91b2c0369b594987f39ca29da7478435513
(cherry picked from commit 53013c87496980b534e447e717a32698fbd4bca0)
Adds events for when a TrustAgentService gets connected
or is stopped. Also explicitly revokes trust when a
trust agent gets disconnected, such that it shows up in
dumpsys.
Bug: 15281644
Change-Id: I5875a34da923345683279c1f755d43454ff6318d
New tasks were being started on ActivityViews because they
matched packages. This fix enforces a rule that new tasks
can only be started on ActivityViews if they are explicitly
targeted for that ActivityView.
Fixes bug 15162447.
Change-Id: I9ccb72171b5cda0897a0b9ffe4cbebfbb0d92c2c
Conflicts:
services/core/java/com/android/server/am/ActivityStackSupervisor.java
Bad files and OS updates can cause the restore to throw an NPE.
Rather than bring down the server process over this, just delete
the persistent file and keep going.
Fixes bug 15219594.
Change-Id: Id9cd39988ff93a26def036a05c46209364f2a4c0
Because ActivityManagerService.systemReady() is reentrant we could
restore tasks and start the TaskPersister more than one time. This
fix limits operations on TaskPersister to one time only.
Fixes bug 15256579.
Change-Id: I6bf2c26b37acdfd9b15a6f277966966b743d03b6
Listeners registered via INotificationMananger.registerListener()
must not be unbound by NoMan because it won't be able to re-bind
them later.
Bug: 15131411
Change-Id: Ic5088252c86e7c32c522ba1606f123cefde3720d
If you don't you're going to have a bad time. In particular we
did not act on the new flags and the documentLaunchMode setting
was ignored.
Fixes bug 15245852.
Change-Id: Ie1c435c4a821b9fc787e5e06e7b24aa98a242225
Tasks launched from the recent task list will now return to the list
when they are finished. Also tasks that are launched from the
notification panel and services will now return to the list,
provided that the launcher is not front and center when they are
launched.
Fixes bug 14464114.
Change-Id: Ic0d3731fc7248d1eaa80e5ee399753d80e80c979
Recent tasks that have the persistable flag set are
saved to /data/system/recent_tasks/ on shutdown and in the
background. Their thumbnails are saved to
/data/system/recent_images/.
Change-Id: Ifb820a01c412fe1f8c0f6e41aa655fafd89eaa8d
Invalidating screenshots when we resume the task that they were taken in. (Bug 13587139)
- Removing multiple calls to get the same thumbnail screenshot
(cherry picked from commit d4ce870e9ad24eff444443bd19ca2061f7c3099d)
Change-Id: Id1feea856a1374173c7f0d329d6f11482794df1a
If we rely on mNotificationList to be sorted, then we cannot allow
records to change without a corresponding call to sort. Currently
RankingFuture may modify records in a separate thread, while the sort
doesn't happen until later. This creates a window for race conditions.
Instead, RankingFuture should record operations to be performed on the
record that will replayed later, in a transaction along with a sort.
We can't simply overwrite the old record completely because another
future may be concurrently modifying a different aspect of the record.
Two futures that attempt to modify the same aspect will be serialized
and the second will overwrite eventually the first.
Change-Id: I9223cabdc60f72d8e37e6d8119bea1e0127185c0
(cherry picked from commit 77d3e0d0297caca5358879d36e8ba77710eb8e82)
