OSDN Git Service

android-x86/frameworks-base.git
6 years agoMerge "Preserve order of shared library files"
Paul Duffin [Tue, 24 Oct 2017 07:48:13 +0000 (07:48 +0000)]
Merge "Preserve order of shared library files"

6 years agoMerge changes from topic "bt_get_cod"
Treehugger Robot [Tue, 24 Oct 2017 06:14:37 +0000 (06:14 +0000)]
Merge changes from topic "bt_get_cod"

* changes:
  Add missing intents to protected broadcasts
  Get Bluetooth Class of Device

6 years agoAdd missing intents to protected broadcasts
Pulkit Bhuwalka [Tue, 26 Sep 2017 02:19:46 +0000 (19:19 -0700)]
Add missing intents to protected broadcasts

Intents broadcasted by the Bluetooth stack need to be added to protected
broadcasts else they complain with errors in the log since Bluetooth is
considered a system process. A few of these intents, especially AVRCP
intents were missing in the list.

Bug: None
Test: Verified errors don't show up after introducing the intents while
playing on a sink device.

Change-Id: Ie283c50243d9a1bac3fd937f58bc1ddc20867d2f

6 years agoGet Bluetooth Class of Device
Pulkit Bhuwalka [Wed, 20 Sep 2017 22:10:45 +0000 (15:10 -0700)]
Get Bluetooth Class of Device

Adds API to fetch Bluetooth CoD value of the stack. The API is hidden to
only be used by System code.

Bug: 36015415
Test: Verified fetching of COD from test app on flashed device, after modification
and after after reboot.

Change-Id: Ie35ecf141704c2aac46678da7cabdc7203a088f2

6 years agoMerge "Persist Bluetooth CoD config value across reboot"
Treehugger Robot [Tue, 24 Oct 2017 06:05:45 +0000 (06:05 +0000)]
Merge "Persist Bluetooth CoD config value across reboot"

6 years agoMerge "NetdEventListener: add rolling log of connect and dns stats"
Hugo Benichi [Mon, 23 Oct 2017 23:54:53 +0000 (23:54 +0000)]
Merge "NetdEventListener: add rolling log of connect and dns stats"

6 years agoMerge "Change Bluetooth HID Profile Name (1/6)"
Treehugger Robot [Mon, 23 Oct 2017 22:22:41 +0000 (22:22 +0000)]
Merge "Change Bluetooth HID Profile Name (1/6)"

6 years agoChange Bluetooth HID Profile Name (1/6)
Hansong Zhang [Fri, 20 Oct 2017 22:55:59 +0000 (15:55 -0700)]
Change Bluetooth HID Profile Name (1/6)

Make the Bluetooth HID profile name consistent with the Bluetooth HID service
name.

BluetoothInputHost → BluetoothHidDevice
BluetoothInputDevice → BluetoothHidHost
IBluetoothInputHost → IBluetoothHidDevice
IBluetoothInputDevice → IBluetoothHidHost
BluetoothProfile.INPUT_HOST → BluetoothProfile.HID_DEVICE
BluetoothProfile.INPUT_DEVICE → BluetoothProfile.HID_HOST

(Cherry-picked from commit c26c76c63d933f8057f795d05624f91b811c8c71)
Merged-In: Iadb890a54dd3d6868b87514472bbac6bb0c6179f
Bug: 68055651
Test: make
Change-Id: Iadb890a54dd3d6868b87514472bbac6bb0c6179f

6 years agoMerge "Add config to display VM number as default number of CF"
Jordan Liu [Mon, 23 Oct 2017 20:27:28 +0000 (20:27 +0000)]
Merge "Add config to display VM number as default number of CF"

6 years agoPreserve order of shared library files
Paul Duffin [Mon, 2 Oct 2017 09:23:25 +0000 (10:23 +0100)]
Preserve order of shared library files

Shared libraries are stored in a list so order is preserved. However,
when they are resolved to files, e.g. for use as a class path the
file names are added to an ArraySet which loses the order. Presumably
they are added to a Set to eliminate duplicates. This switches to a
LinkedHashSet which will preserve the order in which the files are
added while still avoiding duplicates.

It is possible that this could cause app compatibility issues as the
order in which shared libraries is being added is changing. Problems can
only arise if two libraries whose order changes have duplicate classes
and/or resources. In that case the app was only working by luck, as the
order provided by ArraySet is based on the numerical order of hash
codes.

This was found while investigating performance regressions in
GoogleDialer, unfortunately it does not fix the regressions.

Bug: 65552462
Test: flash -w and systrace GoogleDialer to ensure correct order
Change-Id: I0e94471cc481437712f7cf0dab63d88f50cf3b14
Merged-In: Ia01ce4821fa53e4785716b72c4f87a0b0ab4dcc8

6 years agoPersist Bluetooth CoD config value across reboot
Pulkit Bhuwalka [Tue, 26 Sep 2017 00:55:12 +0000 (17:55 -0700)]
Persist Bluetooth CoD config value across reboot

Create new Settings key for storage and update the API doc.

Bug: 36015415
Test: Verified by storing a value through a test app and restarting the
machine to ensure it's picked up.

Change-Id: I94aa054e525c4656bb3a824a29cae9c88f8904e0

6 years agoMerge "Use final release of ASM 6.0"
Colin Cross [Sun, 22 Oct 2017 01:37:11 +0000 (01:37 +0000)]
Merge "Use final release of ASM 6.0"

6 years agoNetdEventListener: add rolling log of connect and dns stats
Hugo Benichi [Wed, 27 Sep 2017 14:28:59 +0000 (23:28 +0900)]
NetdEventListener: add rolling log of connect and dns stats

This patch adds a rolling time order logs of basic statistics of
connect and dns, grouped by network id and including:
 - average error rate
 - average latency
 - max latency
 - total number of operations

The log is segmented in buckets of 5 minutes, and covers the last 4
hours.

Note that only blocking connect() calls latencies are recorded with the
current netd logging infrastructure.

Example of logs:
05:30:00.000: {netId=100, WIFI, dns avg=92ms max=525ms err=00.0% tot=37, connect avg=0ms max=23ms err=00.0% tot=25}
05:35:00.000: {netId=100, WIFI, dns avg=94ms max=537ms err=05.3% tot=57, connect avg=0ms max=0ms err=00.0% tot=69}
05:40:00.000: {netId=100, WIFI, dns avg=220ms max=350ms err=00.0% tot=4, connect avg=0ms max=0ms err=00.0% tot=2}
05:45:00.000: {netId=100, WIFI, dns avg=112ms max=113ms err=00.0% tot=1, connect avg=0ms max=0ms err=00.0% tot=1}
05:50:00.000: {netId=100, WIFI, dns avg=131ms max=269ms err=00.0% tot=2, connect avg=0ms max=0ms err=00.0% tot=1}

Bug: 65700460
Test: runtest frameworks-net
Change-Id: I54e76d18fbaaa92639fb675f93ea90b7615fd6bf

6 years agoMerge "Add Runnable mechanism to WakeupMessage"
Etan Cohen [Sat, 21 Oct 2017 02:04:50 +0000 (02:04 +0000)]
Merge "Add Runnable mechanism to WakeupMessage"

6 years agoMerge "Check success before registering DeathRecipient"
Hall Liu [Fri, 20 Oct 2017 23:10:09 +0000 (23:10 +0000)]
Merge "Check success before registering DeathRecipient"

6 years agoAdd Runnable mechanism to WakeupMessage
Etan Cohen [Fri, 20 Oct 2017 04:12:34 +0000 (21:12 -0700)]
Add Runnable mechanism to WakeupMessage

The WakeupMessage object provides a mechanism to create an Alarm
which will wake-up the device and deliver a message to a Handler.

Add a Runnable mechanism as an alternative to the message.

Bug: 67276378
Test: unit test
Change-Id: Icf5f03b2b25a340273f9d3fcd09f182216ea6238

6 years agoMerge "Remove @Presubmit from SettingsBackupTest.Java"
Treehugger Robot [Fri, 20 Oct 2017 20:14:43 +0000 (20:14 +0000)]
Merge "Remove @Presubmit from SettingsBackupTest.Java"

6 years agoRemove @Presubmit from SettingsBackupTest.Java
Ben Schwartz [Fri, 20 Oct 2017 17:41:12 +0000 (13:41 -0400)]
Remove @Presubmit from SettingsBackupTest.Java

It appears to be causing problems on some build bots.

Test: Problem can only be reproduced after merge...
Change-Id: I7c9c1c7a54f0e2981faecbf9793eef888f5c97d1

6 years agoUse final release of ASM 6.0
Colin Cross [Fri, 20 Oct 2017 19:57:56 +0000 (12:57 -0700)]
Use final release of ASM 6.0

Test: m checkbuild
Change-Id: I8b39c7887763a1a5efce2622421906f9b2e1994a

6 years agoMerge "Refactored area update info support"
Treehugger Robot [Fri, 20 Oct 2017 19:19:09 +0000 (19:19 +0000)]
Merge "Refactored area update info support"

6 years agoMerge "Add link_type for temp_layoutlib"
Dan Willemsen [Fri, 20 Oct 2017 18:14:56 +0000 (18:14 +0000)]
Merge "Add link_type for temp_layoutlib"

6 years agoMerge "Add a global setting to disable DNS over TLS"
Treehugger Robot [Fri, 20 Oct 2017 14:58:39 +0000 (14:58 +0000)]
Merge "Add a global setting to disable DNS over TLS"

6 years agoMerge "more O_CLOEXEC"
Treehugger Robot [Fri, 20 Oct 2017 14:44:26 +0000 (14:44 +0000)]
Merge "more O_CLOEXEC"

6 years agoMerge "Fix netlink group mismatch in fds used for offload"
Treehugger Robot [Fri, 20 Oct 2017 08:14:40 +0000 (08:14 +0000)]
Merge "Fix netlink group mismatch in fds used for offload"

6 years agoAdd config to display VM number as default number of CF
Qiong Liu [Mon, 16 Oct 2017 09:04:30 +0000 (17:04 +0800)]
Add config to display VM number as default number of CF

Add config to determine whether to display voicemail number when
call forwarding number is null. The default value is false.

Test: manual - checked the voicemail number shown as default
Bug: 67855679
Change-Id: Iffa7f9a31757e2df2535abf0269a768ff4e67e8c

6 years agoAdd link_type for temp_layoutlib
Dan Willemsen [Fri, 20 Oct 2017 05:48:53 +0000 (22:48 -0700)]
Add link_type for temp_layoutlib

So that it can be used by java libraries when we turn on dep
verification for host java libraries.

Test: m nothing
Change-Id: Iaa2acfb585db0a67b449e866883f71691395971e
Merged-In: I69a1a826d0bac8ede1f9a337c9c1d930bbcd04f3

6 years agoMerge "[framework] Fixed signal strength issue"
Pengquan Meng [Fri, 20 Oct 2017 05:44:42 +0000 (05:44 +0000)]
Merge "[framework] Fixed signal strength issue"

6 years agoFix netlink group mismatch in fds used for offload
Niranjan Pendharkar [Wed, 18 Oct 2017 23:58:25 +0000 (16:58 -0700)]
Fix netlink group mismatch in fds used for offload

Existing groups used for binding to netlink conntrack events don't
seem to work as expected. Fix this by using exact groups defined
in IOffloadConfig hal.

Test: as follows
      - Compiles and boots on walleye
      - Validated conntrack destroy events are received as expected.
      - Validated offload works as expected.

Bug:68018148
CRs-fixed:2126789

Change-Id: I76178fd7f109b82a4e4961347b8949937813d71d

6 years agoMerge "HwBinder: don't re-implement libhidl's getService"
Treehugger Robot [Fri, 20 Oct 2017 04:12:29 +0000 (04:12 +0000)]
Merge "HwBinder: don't re-implement libhidl's getService"

6 years agoMerge changes I3b56e75d,I36255a2c
Treehugger Robot [Fri, 20 Oct 2017 02:29:41 +0000 (02:29 +0000)]
Merge changes I3b56e75d,I36255a2c

* changes:
  Skeletal IpManager proxying to the new IpClient
  Rename IpManager to IpClient

6 years agoMerge "Remove "result2" in AccountManagerService"
Treehugger Robot [Fri, 20 Oct 2017 01:29:45 +0000 (01:29 +0000)]
Merge "Remove "result2" in AccountManagerService"

6 years agomore O_CLOEXEC
Nick Kralevich [Fri, 20 Oct 2017 00:51:50 +0000 (17:51 -0700)]
more O_CLOEXEC

Test: code compiles.
Change-Id: I4f76258cc607dee6754abbcab5ba5d4c339e69cd

6 years agoRefactored area update info support
Jack Yu [Thu, 22 Jun 2017 00:42:37 +0000 (17:42 -0700)]
Refactored area update info support

1. Refactored area update information support so it
   is not limited to channel 50 and Brazil.
2. Added channel 60 for India.

Test: Manual
bug: 37746109
Merged-In: Ic1c4683ecfe0556547ba15eb9b9e780644f17bff
Change-Id: Ic1c4683ecfe0556547ba15eb9b9e780644f17bff
(cherry picked from commit 93c42856235ba0a47b9eb1514d37fd7b654f0df0)

6 years agoMerge "Shard framework during Java compilation for target side."
Treehugger Robot [Thu, 19 Oct 2017 23:46:18 +0000 (23:46 +0000)]
Merge "Shard framework during Java compilation for target side."

6 years ago[framework] Fixed signal strength issue
Pengquan Meng [Thu, 19 Oct 2017 01:07:58 +0000 (18:07 -0700)]
[framework] Fixed signal strength issue

The flag GSM of signal strength maybe unknown under OOS, but the signal
strength changed has been notified. In this case, set the flag GSM based
on the signal strength data.

Test: manual
Bug: 62431424

Change-Id: I9e63c45ea8b7c6c37e6ecd2478808c310c3d69ae

6 years agoMerge "Add storageId to moveObjects"
Jerry Zhang [Thu, 19 Oct 2017 17:55:34 +0000 (17:55 +0000)]
Merge "Add storageId to moveObjects"

6 years agoMerge "Add support for AES-GCM-ESP as an IPSec algorithm"
Benedict Wong [Thu, 19 Oct 2017 17:06:30 +0000 (17:06 +0000)]
Merge "Add support for AES-GCM-ESP as an IPSec algorithm"

6 years agoMerge "Move most android.util.Mutable* classes to framework."
Tobias Thierer [Thu, 19 Oct 2017 16:08:07 +0000 (16:08 +0000)]
Merge "Move most android.util.Mutable* classes to framework."

6 years agoMerge "Framework: Prefer android.system.Os over libcore.io.Libcore.os"
Tobias Thierer [Thu, 19 Oct 2017 14:15:52 +0000 (14:15 +0000)]
Merge "Framework: Prefer android.system.Os over libcore.io.Libcore.os"

6 years agoHwBinder: don't re-implement libhidl's getService
Steven Moreland [Thu, 19 Oct 2017 06:19:43 +0000 (23:19 -0700)]
HwBinder: don't re-implement libhidl's getService

Rather than re-implementing getService in the JNI backend
for HIDL Java classes, we call the implementation in libhidl.

This does several good things:
- prevents frameworks/base/core from depending on PRODUCT_FULL_TREBLE
- prevents frameworks/base/core from having any user vs. userdebug differences
- ensures there are no subtle differences between C++ and Java getService
- removes essentially copy/pasted code
- unlocks the door to provide 'getService' 'tryGetService' semantics in Java
  (b/67981006)

Has the side effect of making passthrough mode kind work in Java, but
still with parcelling to transition between the ABIs:
- getRawServiceInternal returns passthrough service
- toBinder wraps that service in Bn*
- in-process binder calls are made in the same process

Test: device boots + hidl_test_java
Bug: 67974907
Change-Id: Ie459f95eb8a059f578f703d1f73ca42417cfeeb8

6 years agoMerge "Settings: Fix leak of BroadcastReceiver intents"
Treehugger Robot [Thu, 19 Oct 2017 05:53:23 +0000 (05:53 +0000)]
Merge "Settings: Fix leak of BroadcastReceiver intents"

6 years agoSkeletal IpManager proxying to the new IpClient
Erik Kline [Wed, 11 Oct 2017 13:00:11 +0000 (22:00 +0900)]
Skeletal IpManager proxying to the new IpClient

Test: as follows
    - built
    - flashed
    - booted
    - runtest frameworks-net passes
    - both wifi and ethernet connectivity works
Bug: 62476366

Change-Id: I3b56e75d545594b88849a621e4ae09f2e1d45c69

6 years agoRename IpManager to IpClient
Erik Kline [Wed, 11 Oct 2017 10:43:10 +0000 (19:43 +0900)]
Rename IpManager to IpClient

Test: as follows
    - built
    - flashed
    - booted
    - runtest frameworks-net passes
Bug: 62476366
Change-Id: I36255a2c8e1a2f52693716b445359cce0b4304af

6 years agoSettings: Fix leak of BroadcastReceiver intents
Manu Viswanadhan [Tue, 25 Oct 2016 15:08:23 +0000 (20:38 +0530)]
Settings: Fix leak of BroadcastReceiver intents

Use Case: Repeated BT ON/OFF

Failure: Broadcast receiver leak is observed with ON/OFF stress
test which leads to broadcast receiver delays.

Steps:
BT ON/OFF.

Root Cause: During every ON/OFF there are broadcast receiver leaks
caused by repeated registering of profile intent receivers.

Fix: Separate broadcast receiver created for profile intents, such
that broadcast receiver for adapter state intents is always
active. Unregister and re-register the profile receiver when BT
is turned OFF.

Test: BT ON/OFF

Bug: 35415158

Change-Id: Id1db747c61bb2dd33515ac45cdc2059844d3e4f5

6 years agoAdd support for AES-GCM-ESP as an IPSec algorithm
Benedict Wong [Wed, 23 Aug 2017 04:42:33 +0000 (21:42 -0700)]
Add support for AES-GCM-ESP as an IPSec algorithm

Allows native AES-GCM-ESP to be used as an IPSec transport/tunnel mode
algorithm with kernel support

Bug: 63589918
Test: IPsecService tests added, existing ones pass
Change-Id: Ie1a9a902be205f269aa37bf956198f2e5b177c21

6 years agoMerge "Allow ro. properties to have arbitrary lengths"
Tom Cherry [Wed, 18 Oct 2017 18:21:10 +0000 (18:21 +0000)]
Merge "Allow ro. properties to have arbitrary lengths"

6 years agoAllow ro. properties to have arbitrary lengths
Tom Cherry [Wed, 18 Oct 2017 16:25:17 +0000 (09:25 -0700)]
Allow ro. properties to have arbitrary lengths

Bug: 23102347
Bug: 34954705
Test: build
Change-Id: I99b074633c60e95cdd2284bf8a708b9187c1e310

6 years agoMove most android.util.Mutable* classes to framework.
Tobias Thierer [Wed, 18 Oct 2017 13:33:57 +0000 (14:33 +0100)]
Move most android.util.Mutable* classes to framework.

android.util is the only package shared between libcore and
framework, with only the Mutable* classes living in libcore.

This CL topic moves most of these classes to framework.
After this CL topic, only MutableInt and MutableLong remain
in libcore. This prevents future libcore dependencies on
android.util; it is a first step towards removing the package
overlap between libcore and framework.

Test: Treehugger
Bug: 67901714

Change-Id: Id466181cb0db747da17f38ddb0b99c3e522add16

6 years agoFramework: Prefer android.system.Os over libcore.io.Libcore.os
Tobias Thierer [Tue, 17 Oct 2017 19:26:20 +0000 (20:26 +0100)]
Framework: Prefer android.system.Os over libcore.io.Libcore.os

This is a pure refactoring with no a behavior change other than
that these calls now go through android.system.Os, which immediately
delegates to Libcore.os.

This is a first step towards separating framework (via
android.system.Os) vs. libcore (via Libcore.os) clients of these
low level APIs. Separating these is a prerequisite towards moving
the API parts of android.system into framework, and moving the
rest into a different package in libcore.

Test: Treehugger
Bug: 67901714

Change-Id: Ifd8349ec5416e5693f40dba48fdf2bef651b7d81
Merged-In: Ifd8349ec5416e5693f40dba48fdf2bef651b7d81

6 years agoMerge "Use soong_ui directly, instead of make"
Dan Willemsen [Wed, 18 Oct 2017 06:00:51 +0000 (06:00 +0000)]
Merge "Use soong_ui directly, instead of make"

6 years agoUse soong_ui directly, instead of make
Dan Willemsen [Wed, 18 Oct 2017 00:46:41 +0000 (17:46 -0700)]
Use soong_ui directly, instead of make

This will use ckati instead of make for dumping variables, for
consistent makefile parsing. It may also sandbox executions, etc.

Also skip the legacy make wrapper to build, and use soong_ui instead.
The command line arguments are equivalent, but -j is no longer
necessary, we'll default to a sane value.

Test: bit -b framework
Change-Id: Idc2c9d1abe84b6b783b8702925de5493d47a4509

6 years agoAdd storageId to moveObjects
Jerry Zhang [Tue, 17 Oct 2017 20:47:51 +0000 (13:47 -0700)]
Add storageId to moveObjects

When moving between different storages, the storageId
must be changed as well.

Bug: 67028892
Test: Move between storages with virtual sdcard.
Change-Id: Ie82aa8d53686bfffb7c26b46a49524a00f2f0214

6 years agoMerge "Add config to skip showing CF fail-to-disable dialog"
Jordan Liu [Tue, 17 Oct 2017 20:58:01 +0000 (20:58 +0000)]
Merge "Add config to skip showing CF fail-to-disable dialog"

6 years agoMerge "Frameworks: Remove warning for SystemProperties.get"
Tom Cherry [Tue, 17 Oct 2017 19:28:07 +0000 (19:28 +0000)]
Merge "Frameworks: Remove warning for SystemProperties.get"

6 years agoMerge "Batterystats handles nested unoptimized ble scans"
Treehugger Robot [Tue, 17 Oct 2017 19:03:52 +0000 (19:03 +0000)]
Merge "Batterystats handles nested unoptimized ble scans"

6 years agoAdd a global setting to disable DNS over TLS
Ben Schwartz [Mon, 2 Oct 2017 17:27:13 +0000 (13:27 -0400)]
Add a global setting to disable DNS over TLS

Bug: 63449024
Test: Manual.  Appears to work.
Change-Id: Ib3b58bfff5951240a2e02f18c362efa96635ef4f

6 years agoFrameworks: Remove warning for SystemProperties.get
Tom Cherry [Tue, 17 Oct 2017 16:20:01 +0000 (09:20 -0700)]
Frameworks: Remove warning for SystemProperties.get

We now disallow non-UTF8 property values directly in property_service,
so there is no need for this warning.

This partially reverts 33aea8d40e6cc99f7968a6c6605a81a4587b85e7.

Bug: 63177684
Test: make
Change-Id: I61f49f635c6095013f55b906e00925a8cbd1ea8c

6 years agoAdd config to skip showing CF fail-to-disable dialog
DongYuchen [Thu, 16 Feb 2017 10:22:36 +0000 (18:22 +0800)]
Add config to skip showing CF fail-to-disable dialog

Add config to skip showing CF fail-to-disable dialog. Because it is not
always correct to show fail-to-disable dialog. It depends on the way
of NW CF handling. The default value is set to false.

Test: manual - checked the call forwarding fail dialog was not shown
Bug: 35790386
Change-Id: I8b0e3d15736d95df8fcac205cb9a3ef2f8579988

6 years agoMerge "Enhance network scan request"
Sooraj Sasindran [Tue, 17 Oct 2017 06:03:31 +0000 (06:03 +0000)]
Merge "Enhance network scan request"

6 years agoBatterystats handles nested unoptimized ble scans
Bookatz [Tue, 11 Jul 2017 23:49:17 +0000 (16:49 -0700)]
Batterystats handles nested unoptimized ble scans

When a ble scan starts, it tells batterystats whether that scan is
unoptimized. When the scan stops, batterystats is not informed of
whether the stopped scan was unoptimized. Because the ble scan call
could not be nested (couldn't call start twice without stopping first),
this was fine, but now nesting is possible, so batterystats needs to
know whether the stopped ble scan is unoptimized.

Bug: 63456783
Test: runtest -x frameworks/base/core/tests/coretests/src/com/android/internal/os/BatteryStatsTests.java
Test: no new errors when run cts-dev -m CtsIncidentHostTestCases -t com.android.server.cts.BatteryStatsValidationTest#testUnoptimizedBleScans

Change-Id: Ia73f294cf1807ddaf20f1c0bcc28add001cac78c
Merged-In: Ia73f294cf1807ddaf20f1c0bcc28add001cac78c
(cherry picked from commit 94c5a313d7313a9a88abdf4267af43b87deb0dd2)

6 years agoMerge "Revert "When updating a split app, copy compiled files from base.apk only.""
Calin Juravle [Mon, 16 Oct 2017 22:26:06 +0000 (22:26 +0000)]
Merge "Revert "When updating a split app, copy compiled files from base.apk only.""

6 years agoCheck success before registering DeathRecipient
Hall Liu [Mon, 16 Oct 2017 22:25:31 +0000 (15:25 -0700)]
Check success before registering DeathRecipient

Before registering a DeathRecipient for EMBMS frontend apps, make sure
that whatever operation that they requested actually reported as
successful from the middleware.

Change-Id: I33605dc0a31d827248bed8c7ae6da22096732238
Fixes: 67864414
Test: manual, with testapps

6 years agoMerge "Add null checks to EMBMS AIDL impls"
Hall Liu [Mon, 16 Oct 2017 22:04:17 +0000 (22:04 +0000)]
Merge "Add null checks to EMBMS AIDL impls"

6 years agoMerge "Use -Werror in frameworks/base"
Chih-hung Hsieh [Mon, 16 Oct 2017 19:13:22 +0000 (19:13 +0000)]
Merge "Use -Werror in frameworks/base"

6 years agoMerge "Remove getNetworkType() from ServiceState"
nharold [Mon, 16 Oct 2017 18:30:01 +0000 (18:30 +0000)]
Merge "Remove getNetworkType() from ServiceState"

6 years agoRemove "result2" in AccountManagerService
tiansiming [Fri, 13 Oct 2017 02:57:25 +0000 (10:57 +0800)]
Remove "result2" in AccountManagerService

I think result2 here is to filter out other value returned from getAccountRemovalAllowed.
As the annotation shown in getAccountRemovalAllowed in AbstractAccountAuthenticator,
the result will only contain KEY_INTENT or KEY_BOOLEAN_RESULT or
KEY_ERROR_CODE and KEY_ERROR_MESSAGE. KEY_INTENT has been judged at the beginning of onResult,
KEY_ERROR_CODE and KEY_ERROR_MESSAGE seem extraneous when back to AccountManager,
so i believe that it's ok to use result directly here other than get one more Bundle object.
(tip: if result2 is nessesary, its name doesn't looks formal)

Bug:N/A
Test:N/A

Change-Id: I4e352baa1e74ac680398c328db0c4bc873ba528b
Signed-off-by: tiansiming <tiansiming@xiaomi.com>
6 years agoAdd null checks to EMBMS AIDL impls
Hall Liu [Fri, 13 Oct 2017 23:51:21 +0000 (16:51 -0700)]
Add null checks to EMBMS AIDL impls

Add null checks for things that the user passes in so that they don't
inadvertently cause NPEs for the middleware later on.

Change-Id: I09392a2ac9adec494fcbeeba889ce7fca6708323
Fixes: 67785040
Test: manual

6 years agoRevert "When updating a split app, copy compiled files from base.apk only."
Calin Juravle [Fri, 13 Oct 2017 23:46:26 +0000 (23:46 +0000)]
Revert "When updating a split app, copy compiled files from base.apk only."

This reverts commit d1235f54d4943a0a3a920013a5875b5193bd0490.

Reason for revert: We now include the classpath when asking the runtime
if we need to dexopt. This hack is no longer needed.

Bug: 62269291
Change-Id: I396778f6db0acbbe9017168b35e6023ff97f36fe

6 years agoUse -Werror in frameworks/base
Chih-Hung Hsieh [Tue, 3 Oct 2017 16:57:55 +0000 (09:57 -0700)]
Use -Werror in frameworks/base

* Fix unused variable and return value warnings.

Bug: 66996870
Test: build with WITH_TIDY=1
Change-Id: I890e65a20848d00559ba5a4f9691be1347b456af

6 years agoRemove getNetworkType() from ServiceState
Nathan Harold [Mon, 9 Oct 2017 18:27:40 +0000 (11:27 -0700)]
Remove getNetworkType() from ServiceState

This method was deprecated 4 years and 11 months ago
and set to be removed in 2013. It was replaced by
getVoiceNetworkType().

Bug: 67589287
Test: make checkbuild on goog/master
Change-Id: I5fefa755738ff2fb3f3babfcafd4dfb329314feb

6 years agoMerge "Update locked_region_code_injection tool to ASM 6.0-BETA."
Tobias Thierer [Fri, 13 Oct 2017 17:40:50 +0000 (17:40 +0000)]
Merge "Update locked_region_code_injection tool to ASM 6.0-BETA."

6 years agoEnhance network scan request
Sooraj Sasindran [Thu, 20 Jul 2017 09:43:58 +0000 (02:43 -0700)]
Enhance network scan request

BUG=63865942

Test: Verified by making changes in settings app to call the api
      and confirmed that the new information is delivered to vendor
      code
Merged-In: Ia17647bdedefed5230f4d113b7286d820f557e95
Change-Id: Ia17647bdedefed5230f4d113b7286d820f557e95

6 years agoMerge "Add change to show Invalid card when erroneous card is inserted."
Treehugger Robot [Fri, 13 Oct 2017 13:02:21 +0000 (13:02 +0000)]
Merge "Add change to show Invalid card when erroneous card is inserted."

6 years agoMerge "Fix ConnectivityServiceTest#testNetworkCallbackMaximum"
Hugo Benichi [Fri, 13 Oct 2017 01:44:36 +0000 (01:44 +0000)]
Merge "Fix ConnectivityServiceTest#testNetworkCallbackMaximum"

6 years agoMerge "Convert ext.jar to Android.bp"
Colin Cross [Fri, 13 Oct 2017 01:08:01 +0000 (01:08 +0000)]
Merge "Convert ext.jar to Android.bp"

6 years agoMerge "Add a new carrier config to suppress AoC indication"
Jordan Liu [Thu, 12 Oct 2017 23:42:17 +0000 (23:42 +0000)]
Merge "Add a new carrier config to suppress AoC indication"

6 years agoAdd a new carrier config to suppress AoC indication
Omata Shou [Tue, 22 Aug 2017 08:17:55 +0000 (17:17 +0900)]
Add a new carrier config to suppress AoC indication

Some carriers require to suppress the AoC(Advice of Charge) indication.
This carrier config is used to disable the popup dialog which warns the
user of data charges.

Test: Passed make build
Bug: 65036581
Merged-In: Ibb8d548573c9c747c790722ee526fcd626b3fa67
Change-Id: Ibb8d548573c9c747c790722ee526fcd626b3fa67

6 years agoMerge "Move to some properties users to libbase properties"
Tom Cherry [Thu, 12 Oct 2017 20:47:18 +0000 (20:47 +0000)]
Merge "Move to some properties users to libbase properties"

6 years agoUpdate locked_region_code_injection tool to ASM 6.0-BETA.
Tobias Thierer [Wed, 11 Oct 2017 15:46:05 +0000 (16:46 +0100)]
Update locked_region_code_injection tool to ASM 6.0-BETA.

ASM 6 is required to read version 53 (OpenJDK 9) class files.

Test: make EXPERIMENTAL_USE_OPENJDK9=true services.core
      (in the presence of other unsubmitted CLs that fix
      other issues)
Test: Copied all required dependencies into a subdirectory lib/,
      then ran the steps from TestMain.java, both for OpenJDK 8
      and OpenJDK 9 toolchains.
Bug: 67676752

Change-Id: If324d8ea28450d9a8426a2279190b2f4d9213ed5

6 years agoMerge "MSIM: Fix to show PUK view when two sims locked on PIN."
Pengquan Meng [Thu, 12 Oct 2017 18:10:37 +0000 (18:10 +0000)]
Merge "MSIM: Fix to show PUK view when two sims locked on PIN."

6 years agoMerge "DO NOT MERGE Update demo mode time to 8:00 for Android 8.0"
Treehugger Robot [Thu, 12 Oct 2017 18:00:06 +0000 (18:00 +0000)]
Merge "DO NOT MERGE Update demo mode time to 8:00 for Android 8.0"

6 years agoDO NOT MERGE Update demo mode time to 8:00 for Android 8.0
Shawn [Thu, 12 Oct 2017 02:55:43 +0000 (22:55 -0400)]
DO NOT MERGE Update demo mode time to 8:00 for Android 8.0

Change-Id: I6162889c3859d007629226be2a47a6276807aec0

6 years agoMove to some properties users to libbase properties
Tom Cherry [Wed, 11 Oct 2017 20:44:21 +0000 (13:44 -0700)]
Move to some properties users to libbase properties

The libcutils interface cannot read properties with value length > 92
characters, whereas the libbase one can.  ro.build.fingerprint may be
larger than this size in the future, so we move to libbase to prepare.

Bug: 23102347
Bug: 34954705
Test: build
Change-Id: Ic91aa63c7db1a7a01b9a1ee321b7854a9bcf6d12

6 years agoFix ConnectivityServiceTest#testNetworkCallbackMaximum
Hugo Benichi [Thu, 12 Oct 2017 14:09:25 +0000 (23:09 +0900)]
Fix ConnectivityServiceTest#testNetworkCallbackMaximum

Although commit 893a762c2f7 fixed some flakyness issues in
testNetworkCallbackMaximum so that it became stable when ran on its own,
it introduced a new source of random failures because instead of
registering callbacks after callbacks until a limit was reached, commit
893a762c2f7 changed the test logic to push the assertions right up to
the theoretical limit.

More precisely when registering and unregistering PendingIntents in a
loop, not introducing some delay for checking that previous
PendingIntents have been effectively unregistered can cause the test to
fail. This patch fixes this issue.

Bug: 32561414
Bug: 62918393
Test: runtest frameworks-net
      testNetworkCallbackMaximum now succeeds 100 in a row on sailfish

Change-Id: I086817a738ab99fd53ba76ca8faada6151f46472

6 years agoMerge "Import a few android.net unit tests into frameworks-net"
Hugo Benichi [Thu, 12 Oct 2017 11:58:32 +0000 (11:58 +0000)]
Merge "Import a few android.net unit tests into frameworks-net"

6 years agoMerge "Migrate most of core networking tests to junit4"
Treehugger Robot [Thu, 12 Oct 2017 09:06:24 +0000 (09:06 +0000)]
Merge "Migrate most of core networking tests to junit4"

6 years agoImport a few android.net unit tests into frameworks-net
Hugo Benichi [Thu, 12 Oct 2017 00:54:49 +0000 (09:54 +0900)]
Import a few android.net unit tests into frameworks-net

Bug: 62918393
Test: runtest frameworks-net
Change-Id: I3c7c60bcfeffec5c5ec287810312349237a2868a

6 years agoMigrate most of core networking tests to junit4
Hugo Benichi [Wed, 11 Oct 2017 02:26:25 +0000 (11:26 +0900)]
Migrate most of core networking tests to junit4

This patch is a batch of mechanical changes to test classes to migrate
away from AndroidTestCase and TestCase.

Bug: 62918393
Test: runtest frameworks-net
Change-Id: I74134609e511f22c4d9ecd65780e981f9ba7ae3f

6 years agoMSIM: Fix to show PUK view when two sims locked on PIN.
Pengquan Meng [Thu, 12 Oct 2017 00:43:33 +0000 (17:43 -0700)]
MSIM: Fix to show PUK view when two sims locked on PIN.

Currently when both sims are locked on PIN, upon exhausting
PIN attempts for SIM1 PIN, SIM1 PUK view is not displayed
and the user is not able to unlock the PUK.

This is due to while fetching the getSecurityMode in Key
-GuardSecurityModel, the ordering of checking the security
mode is PIN first and PUK later.So when one sim is locked
on PUK and the other is locked on PIN the ordering of checking
the PIN first is causing to return PIN as security mode in place
of PUK. This will lead to PUK lock screen not been shown to user.

Fix: In KeyGuardSecurityModel.java, while querying the security Mode
modify the order to check for security mode i.e PUK locked state first
and then PIN locked state.

In KeyGuardSimPinView and KeyGuardSimPukView, check for sim locked on
PIN and reset the flags to show the default message.

Test: manually test
Bug: 34796244

Change-Id: I157959d9eb313b9d3af84f86e1429448a0f7f065

6 years agoMerge "Fix URI unparceling and clarify an error condition"
Hall Liu [Wed, 11 Oct 2017 23:04:58 +0000 (23:04 +0000)]
Merge "Fix URI unparceling and clarify an error condition"

6 years agoMerge "Add new Download Error Codes"
Brad Ebinger [Wed, 11 Oct 2017 22:12:26 +0000 (22:12 +0000)]
Merge "Add new Download Error Codes"

6 years agoMerge "Add missing includes."
Treehugger Robot [Wed, 11 Oct 2017 21:37:28 +0000 (21:37 +0000)]
Merge "Add missing includes."

6 years agoAdd change to show Invalid card when erroneous card is inserted.
Wileen Chiu [Wed, 29 Oct 2014 17:18:21 +0000 (22:48 +0530)]
Add change to show Invalid card when erroneous card is inserted.

Add change to show "Invalid card" on UI on CARD_IO_ERROR when
erroneous SIM card is inserted

Bug: 23147806
Change-Id: Ibaacf06f2aad8f206095783cc431a87485f1c2ad

6 years agoAdd missing includes.
Dan Albert [Wed, 11 Oct 2017 19:41:26 +0000 (12:41 -0700)]
Add missing includes.

Test: mma
Bug: None
Change-Id: Iec8b0a7854091946a4bb6e804c5f4cce39a52b47

6 years agoMerge "Convert libjnigraphics to Android.bp"
Colin Cross [Wed, 11 Oct 2017 18:18:45 +0000 (18:18 +0000)]
Merge "Convert libjnigraphics to Android.bp"

6 years agoMerge "Add MM#6 string overlay"
Jordan Liu [Wed, 11 Oct 2017 16:15:39 +0000 (16:15 +0000)]
Merge "Add MM#6 string overlay"

6 years agoMerge "EventLogHelper: Clean up code"
Treehugger Robot [Wed, 11 Oct 2017 15:48:46 +0000 (15:48 +0000)]
Merge "EventLogHelper: Clean up code"

6 years agoMerge "IpManager: add display name for helpful logging, and Network"
Treehugger Robot [Wed, 11 Oct 2017 12:42:41 +0000 (12:42 +0000)]
Merge "IpManager: add display name for helpful logging, and Network"

6 years agoIpManager: add display name for helpful logging, and Network
Erik Kline [Wed, 4 Oct 2017 11:28:54 +0000 (20:28 +0900)]
IpManager: add display name for helpful logging, and Network

Test: as follows
    - built
    - flashed
    - booted
    - runtest frameworks-net passes
Bug: 62476366
Change-Id: I509506814fa704e54f3d67158203276312a24fa1

6 years agoMerge "Log upstream network suspend/resume callbacks"
Treehugger Robot [Wed, 11 Oct 2017 04:00:16 +0000 (04:00 +0000)]
Merge "Log upstream network suspend/resume callbacks"