GitLab

Add a remote call addBluetoothDevice() using AIDL.
This was needed because onBind() is only called once.

Bug: 23219556
Bug: 23760886
Change-Id: Id7554ca55d596352d11dbd6ae3e403138a29c864
Signed-off-by: Phil Burk <philburk@google.com>
(cherry picked from commit 7cd06c0b)

The HashMap is being used but devices never added to the map.

Bug: 23429459
Change-Id: I1da5305a56f5bd48e1c5d9345e721ea8dd2eed4e
Signed-off-by: Phil Burk <philburk@google.com>
(cherry picked from commit e5f722e5)

Stop openBluetoothDevice() from calling its callback twice.

Bug: 23218338
Change-Id: I2e738b22a124c0336c46bef05680a82ee1f59e79
Signed-off-by: Phil Burk <philburk@google.com>
(cherry picked from commit 0ca998d7)

This will give an app an opportunity to learn whether an input port is busy
before the user tries to connect and then fails.

Bug: 22825043
Change-Id: Ifede60f166dfe66ea15453044fce06f4a8452b18
Signed-off-by: Phil Burk <philburk@google.com>
(cherry picked from commit b2355940)

Symptom:
The stack was deleted during cleanup, so the stack
list size has reduced by one and cause this issue.

IndexOutOfBoundsException: Invalid index 1, size is 1

Change-Id: Ic6722aed98a0ad789b331d00e89cb74bf3b36d4c

Reload the EndNowButton label when configuration
changes.

Change-Id: I042d31f0bb011996aa8219c3c412e7b71987244c

This reverts commit dc06bb07.

This should go into system/core.

Change-Id: Idf304ac5ccd4bcc86722f65688e3ee6284452287

The paths in libcore_to_document are now relative to
the build root, not libcore.

Bug: 22023363
Change-Id: I75c86d6332e5eaedc67dd0d3207491aa91c02a5f

Change-Id: Ibc2f98e4c26d0ac6ac7b4a73775c5213c4df4518

If there are 2 activties launched during screen off.
Both of them will be scheduled to pause immediately.
But if the first one does not complete pause in time,
and the second one is a non-fullscreen activity, the
first one will still be visible and no need to stop.

After the second one complete pause, mPausingActivity
is cleared, then the first one's complete pause is failed
and keep state at pausing. If it calls finish, it cannot
be destroyed because its state is pausing so expect it will
be paused later, but actually is has complete paused.

Sample and video:
https://code.google.com/p/android/issues/detail?id=190955

Solution:
A failed-to-pause activity should still change state to paused
if it is pausing. Then when the first activity calls finish,
it will satisfy the condition in ActivityStack.finishActivityLocked
"r.state != ActivityState.PAUSING" to continue the finish flow.

Change-Id: I2f211ddf5039b332b0c7d01ccd043aa18fe168f7

Bug: 25081604
Change-Id: Ie277fd964afcdc80b97432cbd611a8136a2059d7

An OTA package needs to be uncrypted before rebooting into recovery if
it sits on an encrypted /data partition. Once uncrypt gets started, we
cannot re-run it on the package again. Because the file may have been
fully or particially uncrypted and we may end up with a corrupt file
under recovery. Always clean up the package when the device boots into
the normal system to avoid that.

Bug: 24973532
Change-Id: I91682c103d1f2b603626c4bf8d818bced71e3674

With the original logic, if an app creates a system window, when the
user goes to home screen, the system window will be still there and
become unable to receive input events, because the system window will
be also changed to the stopped state with the app window, and the
current logic of ViewRootImpl forbid a stopped window receiving input
events.

This change prevents assigning the token of the app window to system
windows created by the app, so that when the app goes to the stopped
state, its system windows won't be affected (can still receive input
events).

This change is related to the following changes:
https://android.googlesource.com/platform/frameworks/base/+/a5d29971f8f4d463348e9de0f7af7e2e92a377a0
https://android.googlesource.com/platform/frameworks/base/+/15ed2754a308546a54ab1061b4e365d6fe9b898d
https://android.googlesource.com/platform/frameworks/base/+/c3672cd3f7e2bd87d6de9dada499de82b62fae84
https://android.googlesource.com/platform/frameworks/base/+/41725dedc33906aaafee36b2d6523596e2a8a52e

Bug:
https://code.google.com/p/android/issues/detail?id=189710

Change-Id: I515e47bafcf39a2b1bdf92f11f623bef8fb6263c

There are 4 components that all implement package parsing, they are:
1. sdcardd
2. libselinux
3. logd
4. runas

Create a library that can be used by all of them, and new ones as needed.

Change-Id: I87a406802f95d8e7bfd8ee85f723f80e9e6b6c0c
Signed-off-by: William Roberts <william.c.roberts@intel.com>

Get<TYPE>ArrayElements.

Bug: 20760800
Change-Id: I6ba56933a71721da88d31dbf74c48d42a924305a

Allow privileged apps to run in their own priv_app domain.

Motivation:
Untrusted_app is overprivileged due to the inclusion of privileged
apps like gmscore, play store and finsky. Moving these and other
privileged apps to their own domain reduces the permissions required
by untrusted_app.

A separate priv_app domain also protects priv-apps by further
isolating them from third party apps.

Bug: 22033466
Change-Id: I8e6ae5677c5a978301c453d0aa51ebed4459f5a0

Not to resume home activity in finishing stage as following case:

When only home task has 2 activities,
(ex: SetupWizard & DeviceOwnerProvisioningActivity),

If SetupWizard crashed during DeviceOwnerProvisioningActivity in
finishing stage,
makes DeviceOwnerProvisioningActivity cannot destroy immediately,
In resumeHomeStackTask, will resume DeviceOwnerProvisioningActivity
(returned by getHomeActivity),
But actually, DeviceOwnerProvisioningActivity cannot be resumed.
Caused ANR when back key since no more running activity can resume.

Change-Id: Icec49e092d2e7bee7f6a2d198913c1c5db7122f2

Change-Id: I5574ff201b857a29f30ebe8413dbe4128dcff3ae

The AppOpsService handles the 'root' pseudo-package as any other; it
gets no automatic allowances. This is reasonable, but it blocked me from
accessing the mms-sms provider through the 'content' command, even in a
root shell.

So I tried to change the rules:

$ adb root
$ adb shell appops set root WRITE_SMS allow
Error: No UID for root in user 0

This error occurs in the appops command because there isn't really a
package called root, so the UID lookup via PackageManager fails.

But we know that root is UID 0, so we can just skip the lookup.
(Also, AppOpsService handles the other way around in getOpsLocked method.)

Change-Id: Ie0cad67efa438a74a4d9921d29933610cfb13974