OSDN Git Service
Dianne Hackborn [Tue, 5 Sep 2017 23:44:36 +0000 (16:44 -0700)]
Fix issue #
65359137: Apps with target API 26 can bypass background...
...limit introduced in Oreo by swiping away from the task list
There was an edge case where, when restarting a service, if the
process wasn't already running we would bring it up directly in
the idle state and never go through the mechanism of force stopping
services once an app goes idle.
The fix for now is pretty targeted, detecting when we have a new
uid that is immediately idle, and in that case doing the same
thing we do when an existing uid goes idle (stopping any services
in it). This means that in this flow you will get a glitch of
the service starting and then immediately stopping when idle is
applied at that point, but we do end up ultimately in the correct
final state.
Test: manual
Bug:
65359137
Change-Id: I64c29975ad75e7d7390512bf59673d423df16cf2
TreeHugger Robot [Tue, 5 Sep 2017 22:59:02 +0000 (22:59 +0000)]
Merge "audio: Fix for IPlayerShell object leak." into oc-mr1-dev
TreeHugger Robot [Tue, 5 Sep 2017 22:53:00 +0000 (22:53 +0000)]
Merge "Fix issue #
65298216: DeviceIdleController needs to watch..." into oc-mr1-dev
Hugo Benichi [Tue, 5 Sep 2017 22:35:45 +0000 (22:35 +0000)]
Merge "Nat464Xlat: correct racefree teardown" into oc-mr1-dev
TreeHugger Robot [Tue, 5 Sep 2017 22:07:48 +0000 (22:07 +0000)]
Merge "Allow translucent RenderNode->Bitmap captures" into oc-mr1-dev
TreeHugger Robot [Tue, 5 Sep 2017 21:44:30 +0000 (21:44 +0000)]
Merge "Always use SSID average for SpeedLabel." into oc-mr1-dev
Todd Kennedy [Tue, 5 Sep 2017 21:23:43 +0000 (21:23 +0000)]
Merge "resolve stage dir before validation" into oc-mr1-dev
Android Build Merger (Role) [Tue, 5 Sep 2017 21:13:29 +0000 (21:13 +0000)]
Merge "DO NOT MERGE: Dynamic doze check should be required before adjusting sensor rate. am:
1f3da1c8b9 -s ours am:
c7088d8bf8 -s ours am:
ef6e002b30 am:
44bdde7bd1" into oc-mr1-dev
TreeHugger Robot [Tue, 5 Sep 2017 21:08:53 +0000 (21:08 +0000)]
Merge "Double tap to expand PiP." into oc-mr1-dev
Julius D'souza [Tue, 5 Sep 2017 21:06:32 +0000 (21:06 +0000)]
DO NOT MERGE: Dynamic doze check should be required before adjusting sensor rate. am:
1f3da1c8b9 -s ours am:
c7088d8bf8 -s ours am:
ef6e002b30
am:
44bdde7bd1
Change-Id: Ie9ed360a7a1ab763310bcf743804bd0f3300f9a3
Julius D'souza [Tue, 5 Sep 2017 20:54:13 +0000 (20:54 +0000)]
DO NOT MERGE: Dynamic doze check should be required before adjusting sensor rate. am:
1f3da1c8b9 -s ours am:
c7088d8bf8 -s ours
am:
ef6e002b30
Change-Id: Ie9aabc371ae2bad03bfced07acce9502d6c897ee
Julius D'souza [Tue, 5 Sep 2017 20:43:31 +0000 (20:43 +0000)]
DO NOT MERGE: Dynamic doze check should be required before adjusting sensor rate. am:
1f3da1c8b9 -s ours
am:
c7088d8bf8 -s ours
Change-Id: I6cbd05dcb09b013e17f5c26c1fa24a656c7b8316
Tyler Gunn [Tue, 5 Sep 2017 20:41:03 +0000 (20:41 +0000)]
Merge "Prevent INTERACT_ACROSS_USERS exception in DefaultDialerManager." into oc-mr1-dev
Kai Wang [Tue, 5 Sep 2017 20:40:25 +0000 (20:40 +0000)]
Merge "MediaBrowserService: Clean up ConnectionRecord when browser is dead" into oc-mr1-dev
Pankaj Kanwar [Tue, 5 Sep 2017 20:34:32 +0000 (20:34 +0000)]
Merge "Remove config_tether_apndata for Telstra." into oc-dr1-dev
am:
e424d123ab
Change-Id: I58c3c189ef03bf4d054e69fae1c45852425f70d7
Pankaj Kanwar [Tue, 5 Sep 2017 20:25:36 +0000 (20:25 +0000)]
Merge "Remove config_tether_apndata for Telstra." into oc-dr1-dev
Julius D'souza [Tue, 5 Sep 2017 20:23:44 +0000 (20:23 +0000)]
DO NOT MERGE: Dynamic doze check should be required before adjusting sensor rate.
am:
1f3da1c8b9 -s ours
Change-Id: Ib94052397392808af53de5c6f265a0cd29950b4e
TreeHugger Robot [Tue, 5 Sep 2017 19:30:45 +0000 (19:30 +0000)]
Merge "Fix A2dpProfileTest#getLabelDeviceConnectedWithHighQualityCodec" into oc-mr1-dev
TreeHugger Robot [Tue, 5 Sep 2017 19:21:42 +0000 (19:21 +0000)]
Merge "Fix clipping on 3g icon" into oc-mr1-dev
Jean-Michel Trivi [Tue, 5 Sep 2017 19:13:05 +0000 (12:13 -0700)]
audio: Fix for IPlayerShell object leak.
IPlayerShell object is created for each MediaPlayer object and
this will register to linkToDeath, which will have global
reference. IPlayerShell release method is unregistering to Death
monitor. Here IPlayerShell object is not released at all, when
application releases mediaplayer object. As linkToDeath will have
global reference for each IPlayerShell object, GC will not be able
to clear this object.
Fix is to call IPlayerShell release when application releases
mediaplayer object.
authored-by: Divya Narayanan Poojary <dnaray@codeaurora.org>
Test: manual
Bug:
65374542
Change-Id: Icbd7632d462d44efdbb851c1c8a662bdef0e551c
Adam Lesinski [Tue, 5 Sep 2017 18:53:44 +0000 (18:53 +0000)]
Merge "AAPT2: Make sure to fail when XML linking fails" into oc-mr1-dev
Jason Monk [Tue, 5 Sep 2017 18:29:55 +0000 (18:29 +0000)]
Merge "Don't crash from null ServiceState in cdma" into oc-mr1-dev
John Reck [Tue, 5 Sep 2017 18:24:59 +0000 (11:24 -0700)]
Allow translucent RenderNode->Bitmap captures
Set translucent to true when doing createHardwareBitmapFromRenderNode.
This ensures that the buffer is initialized to transparent which
is a safer default for this path. Marginally less performant in some
cases, but as this is a rare capture safety is preferred for and
performance can be revisted if it turns out to be an issue.
Bug:
65164016
Test: builds
Change-Id: I6467c0ccf46e5e2a87c31cc4855ced8275236ca9
Sundeep Ghuman [Wed, 23 Aug 2017 22:53:00 +0000 (15:53 -0700)]
Always use SSID average for SpeedLabel.
When showing the SpeedLabel for the connected network, continue to use
the SSID based average rather than switching to the curve for the known
AP. This addresses changes in SpeedLabel from when the user selects a
network vs when it appears in the connected section.
Bug:
63073866
Test: runtest --path
frameworks/base/packages/SettingsLib/tests/integ/src/com/android/settingslib/wifi/AccessPointTest.java
Change-Id: I06edc1ca4b1b5c8e8615f48c000544ea22a6bbbd
Adrian Roos [Tue, 5 Sep 2017 18:01:57 +0000 (18:01 +0000)]
Merge "FalsingManager: Allow disabling the HIC via device config flag" into oc-mr1-dev
Todd Kennedy [Tue, 5 Sep 2017 17:59:03 +0000 (10:59 -0700)]
resolve stage dir before validation
We rely on the stage directory being correctly resolved
in order for validation to occur. This forcably happens
when commiting [after validation], reading or writing
to the session. For normal installs, this isn't an issue
since the session will be written to, thus, resolving
the stage directory.
Instead of waiting until the commit, resolve the stage
directory just prior to package validation.
Change-Id: I667b9d27749405d24e3a3c6666e9591ce3e883df
Fixes:
65369970
Test: bit FrameworksCoreTests:android.content.pm.PackageManagerTests\#testInstallNonexistentFile
Dianne Hackborn [Sat, 2 Sep 2017 00:07:04 +0000 (17:07 -0700)]
Fix issue #
65298216: DeviceIdleController needs to watch...
...interactivity, not display state
Now it does.
Test: manual
Bug:
65298216
Merged-In: Iec15bff87c6eef43f60b706fa71954c983a4a1e8
Change-Id: Iec15bff87c6eef43f60b706fa71954c983a4a1e8
Winson Chung [Wed, 16 Aug 2017 17:38:15 +0000 (10:38 -0700)]
Double tap to expand PiP.
- Also fix issue with dismiss scrim not clearing when quickly swiping in
the dismiss area and back, due to an incorrect check (menu state NONE
!= menu is not visible).
Bug:
62078042
Test: com.android.systemui.pip.phone.PipTouchStateTest
Test: Double tap PiP to expand it
Change-Id: I96d9ba169feadf0ea8315152fb3a5b2cfb69a2f0
Julius D'souza [Tue, 5 Sep 2017 17:17:06 +0000 (10:17 -0700)]
DO NOT MERGE: Dynamic doze check should be required before adjusting sensor rate.
Some devices have catastrophic firmware issues in switching the sensor rate;
for these devices, the sensor rate switch should be skipped entirely.
Change-Id: I04fc3d27c99f95b33f86f012a8d398f817f7713a
TreeHugger Robot [Tue, 5 Sep 2017 17:00:06 +0000 (17:00 +0000)]
Merge "Do not consider overlays when finding task to reuse." into oc-mr1-dev
TreeHugger Robot [Tue, 5 Sep 2017 16:28:13 +0000 (16:28 +0000)]
Merge "Add internal implementation for notifyPackageUse" into oc-mr1-dev
TreeHugger Robot [Tue, 5 Sep 2017 16:14:13 +0000 (16:14 +0000)]
Merge "Delete old TEXT_SELECTION_MENU_ITEM_ASSIST logs." into oc-mr1-dev
Mathieu Chartier [Tue, 5 Sep 2017 15:55:52 +0000 (15:55 +0000)]
Merge "Move to using a profile for boot image" into oc-mr1-dev
Adrian Roos [Tue, 5 Sep 2017 15:39:17 +0000 (15:39 +0000)]
Merge "AOD Policy: Fix pausing timeout to use minutes instead of seconds" into oc-mr1-dev
Jason Monk [Tue, 5 Sep 2017 15:20:30 +0000 (11:20 -0400)]
Don't crash from null ServiceState in cdma
Test: runtest systemui
Change-Id: Ib9813af4896d90878df970ef798f3f09e6522f3a
Fixes:
65224543
Evan Laird [Tue, 5 Sep 2017 15:06:05 +0000 (11:06 -0400)]
Fix clipping on 3g icon
Viewport was a _little_ too small.
Test: visual; adb shell am broadcast -a "com.android.systemui.demo" -e
command network -e mobile show -e datatype 3g
Fixes:
64725694
Change-Id: I7e09bb8bb2a93ce6dc6670c175284d3927b007d7
Adrian Roos [Tue, 5 Sep 2017 14:11:10 +0000 (16:11 +0200)]
FalsingManager: Allow disabling the HIC via device config flag
Bug:
65126931
Test: Verify HIC is still enabled on devices where it was not disabled
Change-Id: I30ac9e156dde693c34039d8c21e404a474dd86eb
Abodunrinwa Toki [Tue, 5 Sep 2017 13:14:56 +0000 (14:14 +0100)]
Delete old TEXT_SELECTION_MENU_ITEM_ASSIST logs.
This field is now being used for new smart selection logging.
See: Icc9d5b542140bdb01b2ad4211048bd83eae2ce83
Bug:
64914512
Test: bit FrameworksCoreTests:android.widget.TextViewActivityTest
Test: bit CtsWidgetTestCases:android.widget.cts.TextViewTest
Test: bit CtsAccessibilityServiceTestCases:android.accessibilityservice.cts.AccessibilityTextTraversalTest
Change-Id: If1ce8ca038b089500cd7adb9a146330abe909407
Hugo Benichi [Fri, 1 Sep 2017 01:23:32 +0000 (01:23 +0000)]
Nat464Xlat: correct racefree teardown
This patch relays the NetworkBaseObserver notifications about nat
464xlat stacked interfaces onto the ConnectivityService handler.
This allows to process interface up and down notifications in the
same thread context and eliminates several races:
- NPE risk due to race between fixupLinkProperties called on
ConnectivityService thread and interfaceRemoved called on
NetworkManagementService thread.
- stale LinkProperties pointer reads in both NetworkBaseObserver
callbacks not called on ConnectivityService handler.
- removes the race between stop() and interfaceRemoved().
- removes superfluous LinkProperties notifications when stop() is
called before the stacked interface goes up.
The teardown procedure logic common to stop() and interfaceRemoved() is
put into enterStoppedState() and enterIdleState().
This allows to distinguish and correctly handle the following teardown
scenarios:
- an IPv4 appears -> ConnectivityService calls Nat464Xlat#stop()
-> Nat464Xlat calls stopClatd
-> clatd stops
-> if the stacked interface was up, it is removed
-> Nat464Xlat#interfaceRemoved() is triggered and
a LinkProperties update is sent.
- network disconnects -> ConnectivityService calls Nat464Xlat#stop()
-> Nat464Xlat calls stopClatd
-> clatd stops
-> if the stacked interface was up, it is removed
-> Nat464Xlat#interfaceRemoved() is triggered and
a LinkProperties update is sent.
- clatd crashes or exit -> Nat464Xlat#interfaceRemoved() is triggered
-> Nat464Xlat unregisters itself as a network
observer
-> ConnectivityService is updated about the
stacked interface missing, and restarts
Nat464Xlat if needed.
Note that the first two scenarios have two cases: stop() can be called
before the notification for the stacked interface going up (STARTED), or
after (RUNNING). In the first case, Nat464Xlat must unregister
immediately as a network observer to avoid leaks.
This patch also:
- removes/simplifies comments related to the threading model which
are no obsolete.
- extract clatd management logic from ConnectivityService into
NetworkAgentInfo
- add new unit tests where there was none before.
Bug:
62918393
Bug:
62997041
Bug:
64571917
Bug:
65225023
Test: runtest frameworks-net
Merged-In: I27221a8a60fd9760b567ed322cc79228df877e56
Merged-In: I8f07dfbe5ea8259ff9f5793503f534945e67ad74
Merged-In: I8612db5e5050690db8cf41dd04944b4c22da340c
Merged-In: Icb2dc8229b5ea45e319233b588f2dbe39ea40d4c
Merged-In: Ibafea69224e832a6316c17dbb9b2d62a233088ac
(cherry picked from commit
ef502887ec58886e9347afb841aa06cb0d13acea)
Change-Id: I9d075048873b0e1c5ed45b5674ada3fb303c2bfb
TreeHugger Robot [Tue, 5 Sep 2017 10:33:18 +0000 (10:33 +0000)]
Merge "TextView text selection interaction logging." into oc-mr1-dev
Adrian Roos [Mon, 4 Sep 2017 15:39:50 +0000 (15:39 +0000)]
AOD Policy: Fix pausing timeout to use minutes instead of seconds
Bug:
65010918
Change-Id: I2a5f3a2109a38d7093fd0ea84fbb1026eb6c1215
Test: On AOD cover prox for 10s. Verify screen turns off.
TreeHugger Robot [Tue, 5 Sep 2017 07:57:00 +0000 (07:57 +0000)]
Merge "ConnectivityService: improve wakelock logging" into oc-mr1-dev
Bryce Lee [Fri, 1 Sep 2017 16:18:35 +0000 (09:18 -0700)]
Do not consider overlays when finding task to reuse.
An overlay should not be considered the top activity in a task when
considering a task when launching an activity. Doing so will alter
the behavior of launch modes, such as singleTop and singleTask. In
these cases, the developer has chosen such mode with the expectation
that their task will not have activities from other tasks placed on
top, which is the case for features such as locking a work profile.
This changelist addresses the issue by not considering an overlay to
be the top activity when finding a task based on ActivityRecord.
Fixes:
64839155
Test: bit FrameworksServicesTests:com.android.server.am.ActivityStackTests#testFindTaskWithOverlay
Change-Id: I2684baf6929e5af321404e2eef597f456ff87ee8
Hugo Benichi [Tue, 5 Sep 2017 04:25:07 +0000 (13:25 +0900)]
ConnectivityService: improve wakelock logging
This patch adds the following wakelock related counters to connectivity
service dumps included in bug reports:
- total number of wakelok acquisitions and releases
- total cumulative wakelock duration
- longest time the lock was held
Bug:
65085354
Test: runtest frameworks-net, also manually dumped connectivity service
and check new logging
Change-Id: I8f67750c2eea73abf3d44f7f6df484427a8ea3f9
Justin Klaassen [Tue, 5 Sep 2017 00:00:16 +0000 (17:00 -0700)]
Fix A2dpProfileTest#getLabelDeviceConnectedWithHighQualityCodec
Fixes:
63899732
Test: make RunSettingsLibRoboTests -j7
Change-Id: I60559a04a1c234acaca2919a67e8bbe126d2bb3a
Abodunrinwa Toki [Thu, 31 Aug 2017 11:34:36 +0000 (12:34 +0100)]
TextView text selection interaction logging.
TODO: Log typing over selection
Bug:
64914512
Test: bit FrameworksCoreTests:android.widget.TextViewActivityTest
Test: bit CtsWidgetTestCases:android.widget.cts.TextViewTest
Test: bit CtsAccessibilityServiceTestCases:android.accessibilityservice.cts.AccessibilityTextTraversalTest
Merged-In: Icc9d5b542140bdb01b2ad4211048bd83eae2ce83
Change-Id: Icc9d5b542140bdb01b2ad4211048bd83eae2ce83
Adrian Roos [Mon, 4 Sep 2017 15:37:36 +0000 (15:37 +0000)]
Merge "AOD: Keep brightness sensor registered while covered" into oc-mr1-dev
TreeHugger Robot [Mon, 4 Sep 2017 10:47:42 +0000 (10:47 +0000)]
Merge "Set flags on the correct Intent, please" into oc-mr1-dev
Erik Kline [Thu, 31 Aug 2017 12:09:45 +0000 (21:09 +0900)]
Record even more stats even more often
Like kale, one can never have enough stats. =)
Test: as follows
- built
- flashed
- booted
- runtest frameworks-net passes
Bug:
29337859
Bug:
32163131
Change-Id: Ieb47c3beed50f21c2c858fe57438afd48cfdc662
TreeHugger Robot [Sat, 2 Sep 2017 03:29:51 +0000 (03:29 +0000)]
Merge "Add a photos app filter and a metric for using it." into oc-mr1-dev
TreeHugger Robot [Sat, 2 Sep 2017 02:01:56 +0000 (02:01 +0000)]
Merge "Low ram devices now use 60% reduction for task snap shots" into oc-mr1-dev
TreeHugger Robot [Sat, 2 Sep 2017 01:55:29 +0000 (01:55 +0000)]
Merge "Fix issue #
65259946: APR: runtime restart: at com.android.server.am.ActivityManagerService" into oc-mr1-dev
TreeHugger Robot [Sat, 2 Sep 2017 01:54:58 +0000 (01:54 +0000)]
Merge "Gracefully handle integer overflows." into oc-mr1-dev
TreeHugger Robot [Sat, 2 Sep 2017 00:33:23 +0000 (00:33 +0000)]
Merge "Fix issue with PiP menu activity disallowing touches" into oc-mr1-dev
TreeHugger Robot [Sat, 2 Sep 2017 00:32:32 +0000 (00:32 +0000)]
Merge "Bluetooth: Add AVRCP 1.3 as a developer option for AVRCP version (1/2)" into oc-mr1-dev
Ajay Panicker [Fri, 1 Sep 2017 18:58:38 +0000 (11:58 -0700)]
Bluetooth: Add AVRCP 1.3 as a developer option for AVRCP version (1/2)
This is requied for the 2012 Mazda 3 carkit as the carkit refuses to send
AVRCP commands other than passthrough commands when the version is anything
else. AVRCP 1.3 is compatible with most carkits on the market and can be
used to get most carkits working at the cost of losing many features.
Bug:
37943083
Test: Set AVRCP 1.3 in developer options and see that SDP and the AVRCP
capabilites have updated to reflect this.
TestTracker: 105915/3975
Change-Id: I637b695e5cd8baa5f054ca0dda81372080a96269
TreeHugger Robot [Sat, 2 Sep 2017 00:24:16 +0000 (00:24 +0000)]
Merge "Add test for first draw blend behavior in HWUI pipeline" into oc-mr1-dev
TreeHugger Robot [Sat, 2 Sep 2017 00:16:48 +0000 (00:16 +0000)]
Merge "Fix NPE in ActivityView on surface destruction" into oc-mr1-dev
TreeHugger Robot [Fri, 1 Sep 2017 23:35:20 +0000 (23:35 +0000)]
Merge "use simOperator name in the notification" into oc-mr1-dev
Dianne Hackborn [Fri, 1 Sep 2017 23:18:15 +0000 (16:18 -0700)]
Fix issue #
65259946: APR: runtime restart: at com.android.server.am.ActivityManagerService
Catch security exception as well.
Bug:
65259946
Test: manual
Change-Id: Icffccb7d28ef2cf01955173f7dc5adbb81ed0caa
Mathieu Chartier [Tue, 29 Aug 2017 01:30:01 +0000 (18:30 -0700)]
Move to using a profile for boot image
Updated the system server profile, moved to using a profile for the
boot image.
Use cases for obtaining profiles:
Automated CUJ tests
Automated app launches
Manual usage of dialer, navigation, playstore, and play music.
Storage savings:
Boot art+vdex+oat size:
78305556->
61539326 (-17MB)
New system server profile has more methods (1.5MB -> 4MB oat) since it
has more coverage from use cases.
Ram savings (sample averages for CUJ test):
Average PSS of 3 runs (~100 samples total):
.Boot_vdex: 9359K -> 9902K
.oat_mmap: 21857K -> 13372K
.Heap: 29779K -> 29483K
.Boot_art: 15736K -> 18096K
.LinearAlloc: 9112K -> 8965K
Total PSS: 85846K -> 79819K (-6MB)
Average pages read from flash (512MB device) for CUJ tests (3 samples):
1508MB -> 1183MB
TODO: Investigate why .art goes up, maybe we are missing some classes
in the profiles.
Bug:
37966211
Bug:
63178181
Test: make and flash
Change-Id: I2031c52eab7b868c59d4c5a46cad50bfbc6cd714
Jeff Sharkey [Fri, 1 Sep 2017 18:51:51 +0000 (12:51 -0600)]
Gracefully handle integer overflows.
Try sticking with integer-based math as much as possible for speed,
but switch to double-based math if we detect that we'd end up
causing an overflow.
New tests to verify.
Test: bit FrameworksNetTests:com.android.server.net.NetworkStatsCollectionTest
Bug:
65257769
Change-Id: I1ae35599be134f81850c0a3d86928b057fba1eff
TreeHugger Robot [Fri, 1 Sep 2017 22:18:15 +0000 (22:18 +0000)]
Merge "Remove config_tether_apndata for Telstra." into oc-mr1-dev
Tyler Gunn [Fri, 1 Sep 2017 22:17:05 +0000 (15:17 -0700)]
Prevent INTERACT_ACROSS_USERS exception in DefaultDialerManager.
When TelecomManager methods perform a read phonestate check from a work
profile user, it is possible to get an INTERACT_ACROSS_USERS exception.
This is due to the fact that the filterByIntent method was not calling
queryIntentActivitiesAsUser.
Test: Manual
Bug:
31304557
Change-Id: I2a1869162e5a52aed986a99cc378a6c630af4a70
TreeHugger Robot [Fri, 1 Sep 2017 22:11:48 +0000 (22:11 +0000)]
Merge "Properly clean up broadcast-receiver ANR" into oc-mr1-dev
Chris Craik [Fri, 1 Sep 2017 21:39:18 +0000 (14:39 -0700)]
Add test for first draw blend behavior in HWUI pipeline
Bug:
65077146
Test: BakedOpRendererTests
Change-Id: Ia3139a01fe25344a19f2b947069bd686303b9fd5
TreeHugger Robot [Fri, 1 Sep 2017 21:34:22 +0000 (21:34 +0000)]
Merge "Don't over-acquire NPMS locks." into oc-mr1-dev
Andrii Kulian [Sat, 22 Jul 2017 21:59:51 +0000 (14:59 -0700)]
Fix NPE in ActivityView on surface destruction
Instance variable mSurface in ActivityView was not updated after surface
destroy/create cycle, which lead to NPE when we were trying to release
it again after second destroy callback.
Bug:
63338670
Test: Create app with activityView, cause activity restart.
Change-Id: I96447ebe77d175050565641ba94b748d2b633d99
TreeHugger Robot [Fri, 1 Sep 2017 21:29:41 +0000 (21:29 +0000)]
Merge "Change AccessPointPreference key generation to match AccessPoint." into oc-mr1-dev
Matthew Ng [Fri, 1 Sep 2017 21:20:24 +0000 (14:20 -0700)]
Low ram devices now use 60% reduction for task snap shots
On low ram devices, it will now use a reduce scale of 0.6f.
Test: manual - launch apps, launch recents see the bitmap on task
Change-Id: I016b896e9e43d4d0a549a47aa69e7b5541aa5dbc
Fixes:
65163822
Winson Chung [Fri, 1 Sep 2017 18:33:47 +0000 (11:33 -0700)]
Fix issue with PiP menu activity disallowing touches
- On devices where the expanded bounds are the same as the collapsed
bounds, there is no task stack animation and no subsequent animation
end callback, which means that touch state is not reset. Skip
disallowing touches in this state.
- Also fix race causing the PiP menu to ignore touches due to the pinned
animation end callback happening before the menu is first shown.
Bug:
65168767
Test: Tap menu on Ryu, rotate, and try and drag the PiP before the menu
times out
Change-Id: I0b264678aa3300fd4d4e9da5ef3d48a590232ae1
fionaxu [Fri, 1 Sep 2017 20:39:55 +0000 (13:39 -0700)]
use simOperator name in the notification
today we are using getNetworkOperator for carrier's name in the
notification which might be misleading in cases users romaing on
different network.
as today there is no reliable way to get carrier's name, we will use
getSimOperator and if that returns empty string, fallback to
networkOperatorName. In the long term, we can use better wording if
simOperatorName is unknown.
Bug:
65251050
Test: Manual
Change-Id: I6ffc37f8cc8e686d2e884283be2006cf704e8c64
Christopher Tate [Fri, 1 Sep 2017 20:37:22 +0000 (13:37 -0700)]
Set flags on the correct Intent, please
Bug:
65115426
Test: passes CTS backup suite
Change-Id: I497d077f563fef7b0800df626d15651c721d4ca7
Mehdi Alizadeh [Fri, 1 Sep 2017 20:14:55 +0000 (20:14 +0000)]
Merge "Adds PnoScanMetrics to wifi.proto" into oc-mr1-dev
TreeHugger Robot [Fri, 1 Sep 2017 20:07:59 +0000 (20:07 +0000)]
Merge "Add ability to include messages in oom reports." into oc-mr1-dev
Michael Plass [Fri, 1 Sep 2017 19:04:25 +0000 (19:04 +0000)]
DO NOT MERGE ANYWHERE: [AsyncChannel] Fix race in handling of sync result am:
02d802e71e -s ours am:
b5ee561702 -s ours am:
7391bedf87 -s ours
am:
8597268bc2 -s ours
Change-Id: I57d0fa2972f5f93ca0c12455873ce05356d6ebf4
TreeHugger Robot [Fri, 1 Sep 2017 18:54:49 +0000 (18:54 +0000)]
Merge "reloading the current page on Network lost->avail" into oc-mr1-dev
Michael Plass [Fri, 1 Sep 2017 18:50:10 +0000 (18:50 +0000)]
DO NOT MERGE ANYWHERE: [AsyncChannel] Fix race in handling of sync result am:
02d802e71e -s ours am:
b5ee561702 -s ours
am:
7391bedf87 -s ours
Change-Id: Ia4ed82872d97187158a8f958b44e68814b2c5356
Christopher Tate [Fri, 1 Sep 2017 18:41:39 +0000 (11:41 -0700)]
Properly clean up broadcast-receiver ANR
Specifically, if the only receiver component was disabled while the
broadcast was in flight *and* the app ANRed, we were failing to properly
abandon the broadcast and wound up stuck waiting for delivery completion
that would never happen.
Bug:
64854337
Test: manual
Change-Id: I9181830eca17981bf1ca403ac36f88c84c548360
Jeff Sharkey [Fri, 1 Sep 2017 17:27:13 +0000 (11:27 -0600)]
Don't over-acquire NPMS locks.
We only need to hold mNetworkPoliciesSecondLock when working with
subscription plans; before this CL we could end up acquiring the two
NPMS locks out of order, resulting in a deadlock.
Also annotate objects in NSS that require mStatsLock to be held.
Test: builds, boots
Bug:
65268076
Change-Id: I06497564424316ef895dc8dceba72ae784781dc3
Michael Plass [Fri, 1 Sep 2017 18:35:54 +0000 (18:35 +0000)]
DO NOT MERGE ANYWHERE: [AsyncChannel] Fix race in handling of sync result am:
02d802e71e -s ours
am:
b5ee561702 -s ours
Change-Id: I64e924948b9cd7934d41e2261af73e2232bbaf80
Michael Plass [Fri, 1 Sep 2017 18:11:19 +0000 (18:11 +0000)]
DO NOT MERGE ANYWHERE: [AsyncChannel] Fix race in handling of sync result
am:
02d802e71e -s ours
Change-Id: If1a7766ec46207280183c7eb08e036e526a12a34
kai [Tue, 29 Aug 2017 22:13:03 +0000 (15:13 -0700)]
MediaBrowserService: Clean up ConnectionRecord when browser is dead
This CL cleans up connectionRecords when binder dead and the connection
is lost.
Bug:
65132577
Test: check the number of ConnectionRecord, kill the media app, check
the number again.
Change-Id: I8b31745bcee220b9e70c3f43d7fc9d8dc3f6b83a
Daniel Nishi [Wed, 23 Aug 2017 17:27:08 +0000 (10:27 -0700)]
Add a photos app filter and a metric for using it.
Bug:
64147318
Test: SettingsLib integ tests
Change-Id: Id165a514aa65d2efdd4e6753c3e73762293e892e
Merged-In: I4cae384fa565ac0928177bd85804d7904cd6156c
Fyodor Kupolov [Fri, 1 Sep 2017 17:34:41 +0000 (17:34 +0000)]
Merge "Do not close idle connections by default" into oc-mr1-dev
TreeHugger Robot [Fri, 1 Sep 2017 17:09:37 +0000 (17:09 +0000)]
Merge "Fix home button size" into oc-mr1-dev
Michael Plass [Thu, 13 Jul 2017 17:09:07 +0000 (10:09 -0700)]
DO NOT MERGE ANYWHERE: [AsyncChannel] Fix race in handling of sync result
Bug:
62866191
Bug:
63074860
Bug:
65267749
Test: wifi unit tests
Change-Id: I1d59eb8d599de9d9041e0b9b7d731363675a40c9
(cherry picked from commit
56e46134d364f7f293158979765336721a6c752c)
(cherry picked from commit
9c1d56576e01060f40de74a4c0e35e95064351be)
Amit Mahajan [Thu, 31 Aug 2017 20:49:28 +0000 (13:49 -0700)]
Remove config_tether_apndata for Telstra.
Test: manual
Bug:
63564491
Change-Id: Iaca29bfd474b7df6489b1daeaeea182ebf8d81ae
Adrian Roos [Thu, 31 Aug 2017 13:17:17 +0000 (15:17 +0200)]
AOD: Keep brightness sensor registered while covered
Keeps the brightness sensor registered even if proximity is near.
This allows the brightness sensor to react faster after coming out
of a pocket.
Also wait with unpausing AOD until the brightness sensor reports
a valid brightness again.
Bug:
65010918
Test: Cover prox sensor for 10s, verify with 'adb shell dumpsys sensorservice' that the binned brightness sensor is still registered.
Test: runtest -x packages/SystemUI/tests/src/com/android/systemui/doze/DozeScreenBrightnessTest.java
Test: runtest -x packages/SystemUI/tests/src/com/android/systemui/doze/DozeTriggersTest.java
Change-Id: Ife1bb7399a7ad8a15b1f4d210bb38044cbe9d3ae
TreeHugger Robot [Fri, 1 Sep 2017 07:48:56 +0000 (07:48 +0000)]
Merge "Print correct timeout messages in JobServiceContext" into oc-mr1-dev
TreeHugger Robot [Fri, 1 Sep 2017 04:05:02 +0000 (04:05 +0000)]
Merge "Fix blending behavior with first draw op" into oc-mr1-dev
Hugo Benichi [Fri, 1 Sep 2017 02:49:17 +0000 (02:49 +0000)]
Merge "Revert "Nat464Xlat: interface notification handler on ConnectivityService"" into oc-mr1-dev
TreeHugger Robot [Fri, 1 Sep 2017 02:29:53 +0000 (02:29 +0000)]
Merge "Don't vibrate on un-tuned devices when opening notification shade." into oc-mr1-dev
TreeHugger Robot [Fri, 1 Sep 2017 02:18:10 +0000 (02:18 +0000)]
Merge "Darken car status bar." into oc-mr1-dev
TreeHugger Robot [Fri, 1 Sep 2017 01:53:14 +0000 (01:53 +0000)]
Merge "Track the global setting of BT A2DP on/off" into oc-mr1-dev
TreeHugger Robot [Fri, 1 Sep 2017 01:34:48 +0000 (01:34 +0000)]
Merge "Fixes TaskStackLayoutAlgorithm log errors when window rect is empty" into oc-mr1-dev
Lei Yu [Fri, 1 Sep 2017 00:54:01 +0000 (00:54 +0000)]
Merge "Hook up AOD knobs to Settings.Global" into oc-mr1-dev
Chris Craik [Fri, 1 Sep 2017 00:28:20 +0000 (17:28 -0700)]
Fix blending behavior with first draw op
Bug:
65077146
Test: Manual - uirendering tests don't allow test draw content
to be displayed first.
It's not always valid to disable blending on the first draw to the framebuffer,
since some blend modes affect the framebuffer in different ways. We now only
disable blending if the op is SRC_OVER to be safe.
For example:
canvas.drawColor(0xfeff0000, PorterDuff.Mode.CLEAR);
canvas.drawColor(Color.BLUE, PorterDuff.Mode.DST_OVER);
The BLUE should always be seen - the other draw should just clear the buffer.
Prior to this fix, the above code (put in a window background) would draw black.
In addition, this removes the disable behavior in drawRects(), since that should
never benefit from the optimization - that decoration is always drawn at the end
of a frame.
Change-Id: I34e8d9d62d6e1dfa00e9301f44c277475f2940a8
Jason Monk [Fri, 1 Sep 2017 00:13:36 +0000 (00:13 +0000)]
Merge "Hide battery percentage from accessibility." into oc-mr1-dev
TreeHugger Robot [Thu, 31 Aug 2017 23:56:42 +0000 (23:56 +0000)]
Merge "Fix issue #
63971747: Audio is not heard first time when played after a long pause" into oc-mr1-dev