Also annotate the flags with @IntDef to make things clearer and safer
Add more debug logging
Revert to start/stop being synchronous since telephony and microphone will
need to be handled internally.
Bug: 16731586
Bug: 16514535
Bug: 16549061
Change-Id: I83695d52e9547269c95d443e4d921c9238b7401e
Keep track of requests as well as of networks that come and go.
This is necessary, for example, to ensure that we pretend that
HIPRI has gone down when the HIPRI request goes away, even
though the underlying cell network is actually completely
unaffected.
Also, ensure that when switching default networks we send
disconnect broadcasts (and do so *before* connect broadcasts, to
maintain the illusion).
Bug: 16610051
Change-Id: Ib3c831387124940156df05b312cc36bc0724373e
Instead of providing ways to override MmsConfig globally, this allows
the override per request, which is safer. Also added subId to the getter
of MessagingConfigurationManager.
b/16681649
Change-Id: Ia5ee391e2fa51fbdf87d111174d8df21f5f5e41f
When devices are configured with a doze component, the user
now has the option of disabling doze mode with a secure setting.
Bug:16703536
Change-Id: Ieab6ee8d2acf54580b86fb16e9d77b52080fb3b1
The method removeRecentTasksForUser() was being called without the
ActivityManagerService lock held. This change calles it while the
lock is held and renames it to indicate that it should be called
while holding the lock.
Fixes bug 16775111.
Change-Id: I002f9f96d407de458934da8bc728acd54f0c398c
- Send the notification light info up to SystemUI from NoMan.
- Doze mode will now periodically pulse the display (once)
if the notification light is active.
- Change "tease" terminology to "pulse", which is the singular.
Maintain the multi-pulse on buzz-beep-blink for now as extra
emphasis.
- Scrim controller now always takes number of pulses as an arg,
to support a single pulse (used for notification light, and
eventually pickup once available).
- Dial down the display brightness when pulsing.
Bug:15863249
Change-Id: Ifb208a27e82b66cff1d0c04e5b7f758098ea29cf
The 'log' table creation SQLite code had syntax errors
and left the table in a weird state.
This CL fixes the creation syntax and drops old tables
since they might be in an inconsistent state.
Bug: 16726344
Change-Id: I241d08c9e30cedf27ca37bfaf349975077de222d
The loop was initialising the boolean incorrectly causing the first
iteration to wait even when the condition was already true and have to
hit the timeout before loading continues.
Change-Id: I6a347cff2e4560b38c8aec3297e456b6f784a97e
Cherry-pick from aosp to lmp-dev.
When switching to a newly created user, the user may face this bug
when he opens the RecentTask screen.
A possible bug scenario is described as follows:
A user id of a removed user may be recycled when created a new user.
However, mRecentTasks is not correctly controlled so that old
information may still remain and be possibly mapped to wrong user.
This patch prevents this bug by explicitly removing old information
in mRecentTasks when removing existing user.
Change-Id: I1874dbd604598a5d740ae1e034981e21214c15c6
Signed-off-by: Eunae Kim <eunae.kim@lge.com>
Conflicts:
services/java/com/android/server/am/ActivityManagerService.java
Cherry-pick from aosp to lmp-dev
Sympton:
During testing, skip kill native crash process manually because it
will continue to die by default.
Root Cause:
Large process may take some time to do coredump.In auto test, crash
process will be killed immediately that results incomplete coredump
file.
Solution:
If the tester (IActivityController) will handle app crash event,
Do not kill native crashed process if the rom is debuggable.
Change-Id: Ia360af147d694125d440e5ba2f958c4759a50494
Conflicts:
services/java/com/android/server/am/ActivityManagerService.java
Cherry-picked from aosp into lmp-dev.
Sympton:
Next activity only adds to history but does not launch/resume then results ANR.
Root Cause:
In a rare timing, some windows are switched at the same time,
it will cause some finishing records on the top temporarily,
then set startIt to false that skip to resume the real top activty.
Solution:
If all activities in a task are finishing, do not use it to check.
The behavior/checking is the same concept as in JellyBean:
// If starting in an existing task, find where that is...
boolean startIt = true;
for (int i = NH-1; i >= 0; i--) {
ActivityRecord p = mHistory.get(i);
if (p.finishing) { // <--
continue;
}
Change-Id: I9d81a7b5182400c52e173da23eee61c74692beee
Conflicts:
services/java/com/android/server/am/ActivityStack.java
* changes:
Cherry pick Cleanup debug messages in WebViewFactory and WebViewUpdateService DO NOT MERGE
Cherry pick Move startIsolatedProcess implementation in ActivityManagerService DO NOT MERGE
CHerry pick Find WebView package name from a config resource. DO NOT MERGE
Cherry pick Remove dependency on WebView native library path. DO NOT MERGE
Cherry pick Make WebViewUpdateService a SystemService. DO NOT MERGE
Cherry pick Refactor ActivityManagerService and make WebViewFactory more pedantic. DO NOT MERGE
Cherry pick Refactor native library path logic in WebViewFactory DO NOT MERGE
Cherry pick Append WebView assets to AssetManager when loading the WebView. DO NOT MERGE
Cherry pick Introduce startIsolatedProcess private API in ActivityManager DO NOT MERGE
Cherry pick Load the WebView Java code from an APK. DO NOT MERGE
This is just a cleanup CL which fixes:
- Some erroneous debug messages about the relro creator process.
- The condition checked in WebViewUpdateService, to prevent the
WebView to be used from the SystemServer (it now looks at the
process id, previously it was erroneously looking at the uid).
- Adds a 5s. timeout to the waitForRelroCreationCompleted.
Original BUG:16403706
Original Change-Id: I43a953949050d7df5fe334cfa7257315ee6db071
Bug: 16723226
Change-Id: I2f40be3622b8e6c68b2b52cae7f4d3a95e148cbf
This CL addresses the comments in CL 512432, moving the implementation
of startIsolatedProcess in the outer ActivityManagerService class and
making the entry point in the ActivityManagerInternal sublcass just
a proxy method.
This change also addresses a potential NPE, due to startProcessLocked
returning null in some failure cases.
Original BUG:16403706
Original Change-Id: I21eff88c23221653f552cfc171647a839e42a802
Bug: 16723226
Change-Id: I4af9fd7eea10b11e06a5c37d8160c220376b3ed1
This CL removes the final use of the hardcoded WebView
library path from WebViewFactory. We now use a system
property to set the amount of address space to reserve
in the zygote rather than stat'ing the file.
See also I07ad5a774623c1899eb50474d999a2b1447783f0
Original Change-Id: Id3c1e3cc2e2a06c1021ad0260a37abcd07d78fd6
Bug: 16723226
Change-Id: I55acad96b6d9ee69f14b84d772abf7740ea96b94
Migrate WebViewUpdateService to the newer SystemService approach instead
of ServiceManager.addService.
Original Bug: 16403706
Original Change-Id: I21aa67a41c22c3c20ba9e82eb87e5d610fe130e8
Bug: 16723226
Change-Id: Id276b71ee547e683f0756bcee0f4978ce342c2af
This CL adds more robustness to the logic in WebViewFactory, checking
whether the isolated process did start at all and catching exceptions
in its java side.
Also, this addresses the refactor comments received in CL 509840.
Original BUG:16403706
Original Change-Id: Iaaea6d36142ece6d974c2438259edf421fce9f2e
Bug: 16723226
Change-Id: Id308f2ffde9b67a3eb4719c7b81b4f46421f0c2e
The new API spawns a isolated process, using a custom uid, entrypoint and
abi. Such API is used by the WebViewFactory to spawn its unpriviledged
but trusted process (hence the fixed uid) which rewrites the rerlo file
on boot / when an update occurs.
Since both the ActivityManager service and the WebViewUpdate service
live in the SystemServer their calls be dispatched locally and no
binder interface needs to be exposed for the new startIsolatedProcess API.
Original BUG:16403706
Original Change-Id: I327b59735c12698595e0dbcc4da5d759c9103b0a
Bug: 16723226
Change-Id: Iecb49888e11eec9d302d9712953fd498db5821af
Apps running on another profile of the same user were previously
considered to be running in the background, so their error dialogs
would not show unless ANR_SHOW_BACKGROUND was set.
@bug 15665024
Change-Id: I7e559c6b7f8befc1553bc0ec06e89e5d7906d198
This change addresses the following API council feedback.
- Remove all constants from TvContentRating. Instead this class becomes a
general parser for a tuple of "rating domain"+type+subtype(s).
- Instead of constants in the API, the rating definitions should come from
parsed XML meta-data tied to a TvInputService entry in the AndroidManifest.
This XML should define a ranked order of the rating constants along with
@string references for displaying in the Settings UI. Mention that the
"rating domain" should be scoped similar to a package name, for example
"com.youtube.ratings".
- For system-defined rating types (like US, Korea, etc) we should parse this
same XML format, but it may come from a hard-coded XML resource (since they
aren't tied to a single TvInputService.)
- Thoroughly document the built-in supported types in javadoc on TvContentRating.
Change-Id: I0a9526c73c8ca67fd0eeac63f3c63c05657a45e3
- Make sure input change listener is invoked
- The command from new device at the active path should
start routing control
Change-Id: Ic0ca52d42cc1738f9effced322b00f3bd1957f60
Added getActiveSource to provide Tv Input Service with the information
on HDMI active source/routing path. TIS side will be handled separately.
Change-Id: I3d7c12a6c9da9f96a7f22ba1f66ac2559928858d
Packages without their own backup agents can still have restorable
data to be delivered. Correct the logic for attempting a restore
at install time so that it no longer requires an app-defined backup
agent.
Bug 16688665
Change-Id: I7dcf7ef85ad6b5b393a2c97ce62f1b2eec9d3520
Removed all communication from wm to device policy manager.
Added initialization of cache in wm by dpms.
Change-Id: Ifa0b8bfcd625464b156d5cc0fb66d342deda1c27
This change addresses the following API council feedback:
TvTrackInfo:
-- Remove all KEY_/VALUE_ constants, and replace with explicit,
strongly-typed accessor methods.
-- Add additional Bundle getExtras() method (in addition to the explicit
types); most of the time this Bundle will be null.
Bug: 16542165
Change-Id: Ie48cb170b2bbf07d9460fdc8ed77d7db01799772
There are some bugs where getActiveNetworkInfo gives bad data (seemingly)
and this will give the backing data in logs.
bug:16610051
Change-Id: Iad867485ad78daeb3e88665dcd0fdb0af756a3bf
