Privileged callers can now ask the transport for a string describing
its current state, and for an Intent that can be passed to startActivity()
in order to bring up its exported configuration UI. These will be used
in Settings in order to e.g. show the user the currently active account
being used for backup, and allow the user to choose an account.
The data being funnelled through IBackupManager here are the ones
already exposed by the transports in their implementation of the
IBackupTransport interface.
Bug: 2753632
Change-Id: I2227a2b111d8d0ddf221d63020e20c1316fff55b
Speeds up the decoding leaving more time for cpuidle (lp2),
thus lowering overall current drain.
Change-Id: Ia4e0636b1813810072fe0a4f94f6921adf2eab31
Signed-off-by: James Wylder <james.wylder@motorola.com>
* commit '1bec55bd244f72c65fa254118a72ad3ed10aa845':
Add more checks for bad values to protect from corrupt files.
Grant ACCESS_ALL_DOWNLOADS to DefaultContainerService
* commit 'd9c17cc1178bc09d30eead05bf5be7ec5a05cbe7':
Add more checks for bad values to protect from corrupt files.
Grant ACCESS_ALL_DOWNLOADS to DefaultContainerService
This fixes an issue where the overscroll feedback was rendering
several pixels above the bottom of the list on a flick.
Change-Id: I66160d1b72c04f0273f063910eff995e2b2569f0
* Allows an app to detect that it needs to have additional policies granted
* Add "refreshing" parameter to setActiveAdmin() to handle this case
* Minor cleanups to eliminate warnings (mostly for unused things)
Bug: 3253179
Change-Id: I4bf639bf560557130bf98e8cfb75f996fac416f1
Live streams seem to take a little while to initialize, so it's feasible
that we might get a play request before transitioned to the PREPARED
state. In that case track that we should start playing when onPrepare
completes.
Change-Id: I66ed0ef3a0905a1c5d4206fb2e7bbcc05a593fc7
Be compatible with the old DownloadProvider behavior of being able to
operate on content URIs returned from .insert()
Bug: 3242328
Change-Id: Ibd2ed553c87ea694d4b09a8265c7131fae588719
When a process that has registered one or several binder services with
the service manager crashes, that process will usually be restarted by
init (see init.rc). When the process comes back up, it will attempt
to re-register itself with the service manager. However, the binder
driver may not deliver the death notification to the service manager
before the the new service-registation request, in which case the
service manager will reject the request as a duplicate.
The same may occur if a process takes a while to actually exit. Since
the init language's restart command does not wait for a process to
exit before restarting it, there may be for a brief instant two
processes. The new one will attempt to register before the old
one has exited, causing the same problem.
This patch does not treat duplicate registrations as an error.
Instead, the service manager force-releases the old instance and
proceeds with registering the new instance of a service.
Change-Id: Ib70640ab13f5e22460fff6af2f9520b8a9ac9713
Signed-off-by: Iliyan Malchev <malchev@google.com>
