OSDN Git Service

android-x86/frameworks-base.git
7 years agoFramework: Add libziparchive dependency
Andreas Gampe [Mon, 3 Jul 2017 20:03:29 +0000 (13:03 -0700)]
Framework: Add libziparchive dependency

Bug: 37342627
Test: m
Change-Id: I4bc85b87a6c4305f14586aba662bcccbd4b8b6e2

7 years agoMerge "ConnectivityManager: allow usage of TYPE_NONE"
Treehugger Robot [Wed, 5 Jul 2017 14:21:35 +0000 (14:21 +0000)]
Merge "ConnectivityManager: allow usage of TYPE_NONE"

7 years agoMerge "ConnectivityServiceTest: more informative assert failures"
Treehugger Robot [Wed, 5 Jul 2017 14:07:37 +0000 (14:07 +0000)]
Merge "ConnectivityServiceTest: more informative assert failures"

7 years agoConnectivityManager: allow usage of TYPE_NONE
Hugo Benichi [Tue, 20 Jun 2017 05:07:59 +0000 (14:07 +0900)]
ConnectivityManager: allow usage of TYPE_NONE

This patch allows to use TYPE_NONE for the legacy network type variable
of NetworkInfo. This usage is "safe" with respect to legacy APIs using
network types as most of them already returns null or do nothing for
TYPE_NONE.

Of the existing APIs in ConnectivityManager that accept a network type
argument, those which were already returning null or doing nothing for
TYPE_NONE are:
  getNetworkInfo(int)
  getNetworkForType(int)
  stopUsingNetworkFeature(int, String)
  networkCapabilitiesForType(int)
  requestRouteToHostAddress(int, InetAddress)
  reportInetCondition(int, int)
  isNetworkSupported(int)
  getLinkProperties(int)

Only setProvisioningNotificationVisible needs an additional guard
against TYPE_NONE.

Bug: 30088447
Bug: 62844794
Test: runtest frameworks-net
Change-Id: I112596fcd03d3c2cd42a2a84d265adb38e3944bb

7 years agoConnectivityServiceTest: more informative assert failures
Hugo Benichi [Mon, 26 Jun 2017 01:06:49 +0000 (10:06 +0900)]
ConnectivityServiceTest: more informative assert failures

Bug: 62918393
Test: runtest frameworks-net
Change-Id: If87315ef7cba6380596f48b92c7e0f6eeccc8368

7 years agoMerge "ConnectivityServiceTest: eliminate remaining sleep()"
Hugo Benichi [Wed, 5 Jul 2017 12:25:16 +0000 (12:25 +0000)]
Merge "ConnectivityServiceTest: eliminate remaining sleep()"

7 years agoConnectivityServiceTest: eliminate remaining sleep()
Hugo Benichi [Mon, 26 Jun 2017 01:20:00 +0000 (10:20 +0900)]
ConnectivityServiceTest: eliminate remaining sleep()

ConnectivityServiceTest was still using sleep() in a few places although
these were unnecessary:
  - in testSatisfiedThenLostNetworkRequestDoesNotTriggerOnAvailable(),
    expectNoCallback() and expectAvailableCallback() already include
    waitForIdleHandler calls that drain the message queues and make
    sleep no-ops.
  - in testTimedoutAfterUnregisterdNetworkRequest, the sleeps were
    introduced before unregisterNetworkCallback was changed to have a
    synchronous effect for callback unregistration, therefore the sleep
    becomes simply non-sensical. To reflect this the name of the method
    is also changed.

Bug: 62918393
Test: runtest frameworks-net
Change-Id: I7b701ecf5846a5e1890e86107b8d2544b419ce44

7 years agoMerge "Switch to USB_CONFIGURED as the primary USB readiness signal"
Treehugger Robot [Wed, 5 Jul 2017 05:56:29 +0000 (05:56 +0000)]
Merge "Switch to USB_CONFIGURED as the primary USB readiness signal"

7 years agoSwitch to USB_CONFIGURED as the primary USB readiness signal
Erik Kline [Tue, 4 Jul 2017 13:02:49 +0000 (22:02 +0900)]
Switch to USB_CONFIGURED as the primary USB readiness signal

Test: as follows
    - built
    - flashed
    - booted
    - "runtest frameworks-net" passes
    - USB tethering on and off works as expected
Bug: 32163131
Bug: 36216864
Bug: 62147658
Bug: 62552150

Merged-In: I3ae731c21f292a476a9e30e229d66ef6ab4a0545
Merged-In: I717fdd269b012f32548ebe0603999ae403b03100
Merged-Id: Ia8f7f3616f1358b0427386ce8aff26899e03ac07
Change-Id: Ia8f7f3616f1358b0427386ce8aff26899e03ac07
(cherry picked from commit c438e306cc69b9d4d62512030267e93011b29978)

7 years agoMerge "Revert "ConnectivityManager: allow usage of TYPE_NONE""
Roland Levillain [Tue, 4 Jul 2017 11:15:18 +0000 (11:15 +0000)]
Merge "Revert "ConnectivityManager: allow usage of TYPE_NONE""

7 years agoRevert "ConnectivityManager: allow usage of TYPE_NONE"
Roland Levillain [Tue, 4 Jul 2017 11:07:47 +0000 (11:07 +0000)]
Revert "ConnectivityManager: allow usage of TYPE_NONE"

This CL is breaking some internal builds.

This reverts commit 726931ebf4707795413578b57a1f033bf9a6e064.

Change-Id: Ie79214808d84c73f54a525f515b4c90a3fb23542

7 years agoMerge "Revert "ConnectivityServiceTest: more informative assert failures""
Roland Levillain [Tue, 4 Jul 2017 11:12:10 +0000 (11:12 +0000)]
Merge "Revert "ConnectivityServiceTest: more informative assert failures""

7 years agoRevert "ConnectivityServiceTest: more informative assert failures"
Roland Levillain [Tue, 4 Jul 2017 11:09:10 +0000 (11:09 +0000)]
Revert "ConnectivityServiceTest: more informative assert failures"

CL https://android-review.googlesource.com/#/c/420720/ is
breaking several internal builds, and CL
https://android-review.googlesource.com/#/c/422480/ must
be reverted beforehand.

This reverts commit a6981669c4e2707bcfb3db001e262b2da4bb9c17.

Change-Id: I5883409452d39e1834e183aa7b5ccdbf14291ca5

7 years agoMerge "Revert "ConnectivityServiceTest: eliminate remaining sleep()""
Roland Levillain [Tue, 4 Jul 2017 11:11:09 +0000 (11:11 +0000)]
Merge "Revert "ConnectivityServiceTest: eliminate remaining sleep()""

7 years agoRevert "ConnectivityServiceTest: eliminate remaining sleep()"
Roland Levillain [Tue, 4 Jul 2017 11:10:06 +0000 (11:10 +0000)]
Revert "ConnectivityServiceTest: eliminate remaining sleep()"

CL https://android-review.googlesource.com/#/c/420720/ is
breaking several internal builds, and CL
https://android-review.googlesource.com/#/c/422481/ must
be reverted beforehand.

This reverts commit 8afb9cfd000dadc730e52ebdeb4ad031b84b8e03.

Change-Id: If0b196b6860c4087a05b28770469d7850a76fd37

7 years agoMerge "ConnectivityServiceTest: eliminate remaining sleep()"
Treehugger Robot [Tue, 4 Jul 2017 09:25:53 +0000 (09:25 +0000)]
Merge "ConnectivityServiceTest: eliminate remaining sleep()"

7 years agoMerge changes I90c211dc,I4455f272
Treehugger Robot [Tue, 4 Jul 2017 09:22:07 +0000 (09:22 +0000)]
Merge changes I90c211dc,I4455f272

* changes:
  ConnectivityServiceTest: more informative assert failures
  ConnectivityManager: allow usage of TYPE_NONE

7 years agoConnectivityServiceTest: eliminate remaining sleep()
Hugo Benichi [Mon, 26 Jun 2017 01:20:00 +0000 (10:20 +0900)]
ConnectivityServiceTest: eliminate remaining sleep()

ConnectivityServiceTest was still using sleep() in a few places although
these were unnecessary:
  - in testSatisfiedThenLostNetworkRequestDoesNotTriggerOnAvailable(),
    expectNoCallback() and expectAvailableCallback() already include
    waitForIdleHandler calls that drain the message queues and make
    sleep no-ops.
  - in testTimedoutAfterUnregisterdNetworkRequest, the sleeps were
    introduced before unregisterNetworkCallback was changed to have a
    synchronous effect for callback unregistration, therefore the sleep
    becomes simply non-sensical. To reflect this the name of the method
    is also changed.

Bug: 62918393
Test: runtest frameworks-net
Change-Id: I78426665670f702304212753f417b3d5a8a2c107

7 years agoConnectivityServiceTest: more informative assert failures
Hugo Benichi [Mon, 26 Jun 2017 01:06:49 +0000 (10:06 +0900)]
ConnectivityServiceTest: more informative assert failures

Bug: 62918393
Test: runtest frameworks-net
Change-Id: I90c211dc6d6262475924ecabc2863c47aec5a0b9

7 years agoConnectivityManager: allow usage of TYPE_NONE
Hugo Benichi [Tue, 20 Jun 2017 05:07:59 +0000 (14:07 +0900)]
ConnectivityManager: allow usage of TYPE_NONE

This patch allows to use TYPE_NONE for the legacy network type variable
of NetworkInfo. This usage is "safe" with respect to legacy APIs using
network types as most of them already returns null or do nothing for
TYPE_NONE.

Of the existing APIs in ConnectivityManager that accept a network type
argument, those which were already returning null or doing nothing for
TYPE_NONE are:
  getNetworkInfo(int)
  getNetworkForType(int)
  stopUsingNetworkFeature(int, String)
  networkCapabilitiesForType(int)
  requestRouteToHostAddress(int, InetAddress)
  reportInetCondition(int, int)
  isNetworkSupported(int)
  getLinkProperties(int)

Only setProvisioningNotificationVisible needs an additional guard
against TYPE_NONE.

Bug: 30088447
Bug: 62844794
Test: runtest frameworks-net
Change-Id: I4455f2726d06406047086368628c1f253d854d8d

7 years agoMerge "Wakeup packet mark/mask configuration option"
Treehugger Robot [Tue, 4 Jul 2017 03:38:34 +0000 (03:38 +0000)]
Merge "Wakeup packet mark/mask configuration option"

7 years agoMerge "Add dumpsys support to RulesManagerService"
Neil Fuller [Mon, 3 Jul 2017 09:29:32 +0000 (09:29 +0000)]
Merge "Add dumpsys support to RulesManagerService"

7 years agoMerge "Networking unit tests: fix some flaky tests"
Treehugger Robot [Mon, 3 Jul 2017 09:23:26 +0000 (09:23 +0000)]
Merge "Networking unit tests: fix some flaky tests"

7 years agoMerge "Pass all offload-exempt prefixes into OffloadController"
Treehugger Robot [Mon, 3 Jul 2017 08:30:30 +0000 (08:30 +0000)]
Merge "Pass all offload-exempt prefixes into OffloadController"

7 years agoMerge "Switch IpManager and IpReachabilityMonitor to use SharedLog"
Treehugger Robot [Mon, 3 Jul 2017 08:09:48 +0000 (08:09 +0000)]
Merge "Switch IpManager and IpReachabilityMonitor to use SharedLog"

7 years agoNetworking unit tests: fix some flaky tests
Hugo Benichi [Thu, 29 Jun 2017 13:58:39 +0000 (22:58 +0900)]
Networking unit tests: fix some flaky tests

 - less strict regex for SharedLogTest: the subsecond part of the
   timestamp can have 0, 1, 2 or 3 digits.
 - refactor NetworkStatsServiceTest and NetworkStatsObserversTest to use
   waitForIdleHandler facility of ConnectivityServiceTest.
   NetworkStatsServiceTest was using a flaky custom version of
   waitForIdleHandler.

Bug: 62918393
Bug: 32561414
Test: runtest frameworks-net
Change-Id: I634acfb5f4fe1bd5267e3f14b9f645edc32d5d12

7 years agoSwitch IpManager and IpReachabilityMonitor to use SharedLog
Erik Kline [Thu, 29 Jun 2017 02:46:51 +0000 (11:46 +0900)]
Switch IpManager and IpReachabilityMonitor to use SharedLog

Test: as follows
    - built
    - flashed
    - booted
    - "runtest frameworks-net" passes
    - frameworks/opt/net/wifi/tests/wifitests/runtests.sh passes
    - frameworks/base/wifi/tests/runtests.sh passes
    - roamed from AP to AP watching IpRM logs accumulate
Bug: 62476366
Bug: 62162774

Merged-In: I4bef3b6b1975fa3c73cca0a0e95fad99708f061e
Merged-In: I265566158f3e5ebd21fb43e2eb0b8ba016fca76a
Change-Id: I155a26259420919eeed74166bacce3311caa49da
(cherry picked from commit 0fc10b2b1d7be649c80dda4ebdb2e31af216bfcc)

7 years agoMerge changes I6779a835,I4e3d8322
Lorenzo Colitti [Mon, 3 Jul 2017 01:39:32 +0000 (01:39 +0000)]
Merge changes I6779a835,I4e3d8322

* changes:
  Only connect to one address per address family.
  Update our user agent to something newer.

7 years agoMerge "[BUG] java.util.MissingFormatArgumentException: Format specifier '%d'"
Treehugger Robot [Mon, 3 Jul 2017 00:17:57 +0000 (00:17 +0000)]
Merge "[BUG] java.util.MissingFormatArgumentException: Format specifier '%d'"

7 years agoPass all offload-exempt prefixes into OffloadController
Erik Kline [Sat, 24 Jun 2017 10:29:10 +0000 (19:29 +0900)]
Pass all offload-exempt prefixes into OffloadController

Additionally:
    - eliminate useless TetherMasterUtilState

Test: as follows
    - built
    - flashed
    - booted
    - runtest frameworks passes
Bug: 29337859
Bug: 32163131

Change-Id: I0482ac9dd9e8ba558b253643dbc01875ce9475ef

7 years agoOnly connect to one address per address family.
Lorenzo Colitti [Wed, 28 Jun 2017 14:42:41 +0000 (23:42 +0900)]
Only connect to one address per address family.

This ensures probes don't take too long when a host has
multiple addresses. So for example:

$ adb shell settings put global captive_portal_http_url http://play.googleapis.com/generate_204
$ host play.googleapis.com
play.googleapis.com is an alias for googleapis.l.google.com.
googleapis.l.google.com has address 216.58.196.234
googleapis.l.google.com has address 216.58.197.10
googleapis.l.google.com has address 172.217.26.106
googleapis.l.google.com has address 172.217.27.170
googleapis.l.google.com has address 216.58.199.234
googleapis.l.google.com has address 216.58.221.170
googleapis.l.google.com has IPv6 address 2404:6800:400a:807::200a

results in:

06-29 21:38:09.695 - PROBE_DNS play.googleapis.com 270ms OK 2404:6800:400a:806::200a,172.217.26.106

Bug: 36532213
Test: as described in commit description
Change-Id: I6779a8356bac8d4a3912b0822379d45dfc662541

7 years agoMerge "Add boot profile support"
Treehugger Robot [Sat, 1 Jul 2017 03:24:12 +0000 (03:24 +0000)]
Merge "Add boot profile support"

7 years agoAdd boot profile support
Mathieu Chartier [Fri, 30 Jun 2017 18:47:38 +0000 (11:47 -0700)]
Add boot profile support

If the boot profile file exists, its passed to the runtime as an
image compiler option. Also pass speed-profile for this case.

Bug: 37966211
Test: make, flash, and look at logcat during boot

(cherry picked from commit f99e70e93408bd6dfda595866f13ba0068d4798e)

Change-Id: Ia09ff0200186382e45151b7cabb53016f8e0aa3f

7 years agoMerge "hidl getService throw correct exceptions."
Treehugger Robot [Fri, 30 Jun 2017 20:36:36 +0000 (20:36 +0000)]
Merge "hidl getService throw correct exceptions."

7 years agoAdd dumpsys support to RulesManagerService
Neil Fuller [Fri, 23 Jun 2017 15:43:45 +0000 (16:43 +0100)]
Add dumpsys support to RulesManagerService

Override the dump method so RuleManagerService can dump its state
into logs. Crude argument support has been added for dumping
specific fields in an easy to process way (for test scripts to use).

Tested with:
make -j30 FrameworksServicesTests
adb install -r -g \
  "out/target/product/angler/data/app/FrameworksServicesTests/FrameworksServicesTests.apk"
adb shell am instrument -e package com.android.server.timezone -w \
  com.android.frameworks.servicestests \
  "com.android.frameworks.servicestests/android.support.test.runner.AndroidJUnitRunner"

Test: See above.
Test: Manual; adb shell dumpsys timezone [-format_state piscotz]
Bug: 31008728
Change-Id: I0ad83aa245232ed0b983ceacd8accfb876824d6f

7 years agoMerge "Track move of distro installer code"
Neil Fuller [Fri, 30 Jun 2017 12:28:23 +0000 (12:28 +0000)]
Merge "Track move of distro installer code"

7 years agoWakeup packet mark/mask configuration option
Joel Scherpelz [Thu, 8 Jun 2017 06:35:21 +0000 (15:35 +0900)]
Wakeup packet mark/mask configuration option

Rules will only be installed if a device specific config.xml overrides
the default values of mark and mask.

This change also fixes a bug where wakeup rules added by the wakeup
controller would never be removed because updateInterfaces() is not
invoked on wifi network change or disable.

Test: as follows
    - built
    - flashed
    - booted
    - Manually verified that mask/mark is set correctly
Bug: 28806131

Change-Id: I2380b2383a593da35880d81cbade6eae4f79a776

7 years ago[BUG] java.util.MissingFormatArgumentException: Format specifier '%d'
liangweikang@xiaomi.com [Thu, 29 Jun 2017 06:36:30 +0000 (14:36 +0800)]
[BUG] java.util.MissingFormatArgumentException: Format specifier '%d'

Bug: https://issuetracker.google.com/issues/63121293

Test: null

Change-Id: I2745c415e21a7f26535605cc5bef7eaf2109ae8e
Signed-off-by: liangweikang@xiaomi.com <liangweikang@xiaomi.com>
7 years agoMerge "Implement cancelDownload and remove appName"
Hall Liu [Thu, 29 Jun 2017 21:33:11 +0000 (21:33 +0000)]
Merge "Implement cancelDownload and remove appName"

7 years agohidl getService throw correct exceptions.
Steven Moreland [Wed, 28 Jun 2017 17:21:00 +0000 (10:21 -0700)]
hidl getService throw correct exceptions.

NoSuchElementException needs to be thrown when getService is
called and the service is not available.

Test: hidl_test_java
Fixes: 63087967

Merged-In: I207098cfb732e888f91704f623e888e72e6ce7d9
Change-Id: I207098cfb732e888f91704f623e888e72e6ce7d9

7 years agoMerge "Support battery level reporting via Apple VSC (1/2)"
Treehugger Robot [Thu, 29 Jun 2017 17:44:10 +0000 (17:44 +0000)]
Merge "Support battery level reporting via Apple VSC (1/2)"

7 years agoSupport battery level reporting via Apple VSC (1/2)
Jack He [Wed, 21 Jun 2017 00:09:47 +0000 (17:09 -0700)]
Support battery level reporting via Apple VSC (1/2)

* Add related constants to BluetoothHeadset

Bug: 35874078
Test: make, test with supporting headsets
Change-Id: I9f49ea28efdf6de0e751a377d879fe13a47655d5

7 years agoMerge "Support battery level reporting via Plantronics XEVENT (1/2)"
Treehugger Robot [Thu, 29 Jun 2017 17:17:38 +0000 (17:17 +0000)]
Merge "Support battery level reporting via Plantronics XEVENT (1/2)"

7 years agoSupport battery level reporting via Plantronics XEVENT (1/2)
Jack He [Wed, 21 Jun 2017 00:07:40 +0000 (17:07 -0700)]
Support battery level reporting via Plantronics XEVENT (1/2)

* Add related constants to BluetoothHeadset

Bug: 35874078
Test: make, test with supporting headsets
Change-Id: Ied089a065bf9caa0d03f681950fefa2ea990bb3e

7 years agoMerge "No need to explicitly list ethernet upstream type"
Treehugger Robot [Thu, 29 Jun 2017 12:18:35 +0000 (12:18 +0000)]
Merge "No need to explicitly list ethernet upstream type"

7 years agoTrack move of distro installer code
Neil Fuller [Thu, 29 Jun 2017 11:39:49 +0000 (12:39 +0100)]
Track move of distro installer code

Track move of distro installer code; reflect the new
build target.

Tested with:
make -j30 FrameworksServicesTests
adb install -r -g \
  "out/target/product/angler/data/app/FrameworksServicesTests/FrameworksServicesTests.apk"
adb shell am instrument -e package com.android.server.timezone -w \
  com.android.frameworks.servicestests \
  "com.android.frameworks.servicestests/android.support.test.runner.AndroidJUnitRunner"

Bug: 31008728
Test: See above
Change-Id: Iad987d404af75af3694d50e25ed419889a9557f2

7 years agoMerge "Use sendBroadcastAsUser rather than sendBroadcast"
Treehugger Robot [Thu, 29 Jun 2017 11:29:37 +0000 (11:29 +0000)]
Merge "Use sendBroadcastAsUser rather than sendBroadcast"

7 years agoMerge "HFP: Modify comments for ACTION_HF_INDICATORS_VALUE_CHANGED (1/2)"
Andre Eisenbach [Thu, 29 Jun 2017 08:30:50 +0000 (08:30 +0000)]
Merge "HFP: Modify comments for ACTION_HF_INDICATORS_VALUE_CHANGED (1/2)"

7 years agoHFP: Modify comments for ACTION_HF_INDICATORS_VALUE_CHANGED (1/2)
Jack He [Thu, 22 Jun 2017 19:56:54 +0000 (12:56 -0700)]
HFP: Modify comments for ACTION_HF_INDICATORS_VALUE_CHANGED (1/2)

* Per HFP 1.7.1 spec page 102 of 144
  EXTRA_HF_INDICATORS_IND_ID should be int
  EXTRA_HF_INDICATORS_IND_VALUE should be int

Bug: 35874078
Test: make, PTS test, unit tests
runtest -c com.android.bluetooth.btservice.RemoteDevicesTest bluetooth

Change-Id: Idd316a82bab164b6ea1701bb261b1cc3dee196f1

7 years agoNo need to explicitly list ethernet upstream type
Erik Kline [Wed, 21 Jun 2017 14:27:55 +0000 (23:27 +0900)]
No need to explicitly list ethernet upstream type

Test: as follows
    - built
    - flashed
    - booted
    - runtest frameworks-net passes
Bug: 32163131
Bug: 36076442
Change-Id: I743c20d5b969e33c6b10d0b47ac06e6b8c968acf

7 years agoMerge "Use Bundle for inter-process message"
Treehugger Robot [Thu, 29 Jun 2017 01:51:04 +0000 (01:51 +0000)]
Merge "Use Bundle for inter-process message"

7 years agoImplement cancelDownload and remove appName
Hall Liu [Fri, 23 Jun 2017 23:32:30 +0000 (16:32 -0700)]
Implement cancelDownload and remove appName

Implements the cancelDownload call
Removes appName as an argument from all EMBMS-related apis
Implements the cross-filesystem move method in MbmsDownloadReceiver

Change-Id: Ie8ebfccca4fbb3d69289beb2ed77101fe2c7727f

7 years agoMerge changes from topics 'embms-cleanup-action', 'embms-download-token'
Hall Liu [Wed, 28 Jun 2017 20:26:15 +0000 (20:26 +0000)]
Merge changes from topics 'embms-cleanup-action', 'embms-download-token'

* changes:
  Implement the cleanup action in MbmsDownloadReceiver
  Add download request tokens and enforce destination clear

7 years agoUse Bundle for inter-process message
yinxu [Tue, 27 Jun 2017 21:05:17 +0000 (14:05 -0700)]
Use Bundle for inter-process message

The message sent from NetworkScanRequestTracker to TelephonyScanManager
might be across different process, so we have to wrap the List<CellInfo> into
Bundle because List<> is not parcelable.

Cherry-picked cleanly from:
https://googleplex-android-review.googlesource.com/#/c/2467157/

Test: Telephony sanity tests
Bug: 30954762

Merged-in: I77945b247530b85c0b82876d528549498a711550
Changg-Id: I77945b247530b85c0b82876d528549498a711550
(cherry picked from commit 56049437095d3cb8334f667ff73061f2926b5416)

Change-Id: Iec4043e611a8f58733a361ec9a52fc74e1c974d5

7 years agoMerge "Switch PackageStatusStorage to XML-based storage"
Neil Fuller [Wed, 28 Jun 2017 17:01:56 +0000 (17:01 +0000)]
Merge "Switch PackageStatusStorage to XML-based storage"

7 years agoUpdate our user agent to something newer.
Lorenzo Colitti [Wed, 28 Jun 2017 14:39:40 +0000 (23:39 +0900)]
Update our user agent to something newer.

Bug: 36532213
Test: aosp_bullhead-eng builds
Change-Id: I4e3d832248a746c9fef5302a1eafbe66ade8ed25

7 years agoMerge "Add APIs to get remote device's battery level (1/2)"
Treehugger Robot [Wed, 28 Jun 2017 08:26:13 +0000 (08:26 +0000)]
Merge "Add APIs to get remote device's battery level (1/2)"

7 years agoImplement the cleanup action in MbmsDownloadReceiver
Hall Liu [Thu, 22 Jun 2017 01:23:48 +0000 (18:23 -0700)]
Implement the cleanup action in MbmsDownloadReceiver

Implement ACTION_CLEANUP, add result codes for the ordered broadcasts,
and remove the download request extra for file descriptor requests

Change-Id: I5ee19ad0aefbdceacb636b3d17e197a6fb1bdaa6

7 years agoAdd APIs to get remote device's battery level (1/2)
Jack He [Sat, 17 Jun 2017 02:43:58 +0000 (19:43 -0700)]
Add APIs to get remote device's battery level (1/2)

* Add BluetoothDevice.getBatteryLevel() API to retreive battery level
  information of remote device
* Add BluetoothDevice.ACTION_BATTERY_LEVEL_CHANGED intent to notify user
  that remote device's battery level has changed

Bug: 35874078
Test: make, pair with devices and use them
Change-Id: I41051ee25383f5f3a1e505aef6f8c526385f58bd

7 years agoUse sendBroadcastAsUser rather than sendBroadcast
Neil Fuller [Tue, 27 Jun 2017 17:53:58 +0000 (18:53 +0100)]
Use sendBroadcastAsUser rather than sendBroadcast

According to messages in the logcat sendBroadcastAsUser
is the better option.

Test: Manual
Bug: 31008728
Change-Id: I06bcd3c53878fc2b1b817337ea8719fcc4975501

7 years agoMerge "IMS: Check for binderstate before calling endSession"
Brad Ebinger [Tue, 27 Jun 2017 16:46:57 +0000 (16:46 +0000)]
Merge "IMS: Check for binderstate before calling endSession"

7 years agoMerge "Switch to streaming data for time zone update"
Neil Fuller [Tue, 27 Jun 2017 11:56:53 +0000 (11:56 +0000)]
Merge "Switch to streaming data for time zone update"

7 years agoMerge "Clean up Predicate related code in android.test"
Paul Duffin [Tue, 27 Jun 2017 10:42:56 +0000 (10:42 +0000)]
Merge "Clean up Predicate related code in android.test"

7 years agoMerge "Clean up ClassPathPackageInfoSource"
Paul Duffin [Tue, 27 Jun 2017 10:42:51 +0000 (10:42 +0000)]
Merge "Clean up ClassPathPackageInfoSource"

7 years agoIMS: Check for binderstate before calling endSession
Benergy Meenan Ravuri [Tue, 20 Jun 2017 11:30:42 +0000 (17:00 +0530)]
IMS: Check for binderstate before calling endSession

ImsRegistrationListenerProxy listener, which is registered
as part of ImsManager.open, is not cleared even after moving
feature state to NOT_AVAILABLE because end session
is never called.

Check for binder state instead of feature state READY before
calling end session. End session will be triggered after
moving feature state to NOT_AVAILABLE.

Notify Ims state change only for the supported feature state
change to clients.

Test: Manual
Bug: 62816014
Change-Id: I1bb588ce2be9c181355ff744f43dad4e784e4ec5

7 years agoMerge "[Resources] house keeping in TypeList"
Treehugger Robot [Mon, 26 Jun 2017 22:50:23 +0000 (22:50 +0000)]
Merge "[Resources] house keeping in TypeList"

7 years agoMerge "Fix the inconsistency of scan status."
Ying Xu [Mon, 26 Jun 2017 17:05:57 +0000 (17:05 +0000)]
Merge "Fix the inconsistency of scan status."

7 years agoClean up ClassPathPackageInfoSource
Paul Duffin [Tue, 20 Jun 2017 15:18:07 +0000 (16:18 +0100)]
Clean up ClassPathPackageInfoSource

Part of the work of removing JUnit and dependent android.test classes
from the Android API involves providing a static library that developers
can include in their test applications to ease migration. That library
will be built directly from the source (as opposed to android.jar which
is built from stubs) and so developers will be able to see classes and
methods that are not present in the stubs. This change is one of a
number of similar changes that cleanup the existing non-API code in
order to minimize the additional methods and classes exposed externally.
The basic approach is to remove unused classes and methods, use least
visible access modifier possible and generally minimize the amount of
publicly visible code.

PackageInfoSources only provided a static field and accessor method so
they were moved into ClassPathPackageInfoSource and PackageInfoSources
was removed.

ClassPathPackageInfo was only used in ClassPathPackageInfoSource and in
TestGrouping. In the latter it was simply used as an intermediate value
between ClassPathPackageInfoSource.getPackage(String packageName) and
ClassPathPackageInfo.getTopLevelClassesRecursive(String packageName).
Moving that method into ClassPathPackageInfoSource allowed the
ClassPathPackageInfo to become an inner class of
ClassPathPackageInfoSource. As it is an inner class it no longer needed
an explicit reference to the containing ClassPathPackageInfoSource.

Bug: 30188076
Test: make checkbuild and ran FrameworkTestRunnerTests
Change-Id: Idb0b6a585030805b9cff8562abb93b7e5920c53a

7 years agoMerge "Carrier Config changes to support IMSI encryption."
Pankaj Kanwar [Mon, 26 Jun 2017 15:36:43 +0000 (15:36 +0000)]
Merge "Carrier Config changes to support IMSI encryption."

7 years agoClean up Predicate related code in android.test
Paul Duffin [Thu, 22 Jun 2017 09:47:25 +0000 (10:47 +0100)]
Clean up Predicate related code in android.test

Part of the work of removing JUnit and dependent android.test classes
from the Android API involves providing a static library that developers
can include in their test applications to ease migration. That library
will be built directly from the source (as opposed to android.jar which
is built from stubs) and so developers will be able to see classes and
methods that are not present in the stubs. This change is one of a
number of similar changes that cleanup the existing non-API code in
order to minimize the additional methods and classes exposed externally.
The basic approach is to remove unused classes and methods, use least
visible access modifier possible and generally minimize the amount of
publicly visible code.

The HasClassAnnotation and HasMethodAnnotation were never used
separately and only used by HasAnnotation. This merges the functionality
into a single nested class in TestPredicates, hidden behind the new
TestPredicates.hasAnnotation(). The HasAnnotationTest was renamed as
TestPredicatesTest and the HasClassAnnotationTest and
HasMethodAnnotationTest classes were removed as their tests provide no
additional coverage. The removal of the Has*Annotation.java files means
that the test-runner/src/android/test/suitebuilder/annotation/ directory
is empty apart from the package.html file so that was moved to
legacy-test to sit alongside the actual annotation classes.

The Predicates class, while part of the legacy-test module was only
ever used by the test-runner module and only its not() method was
actually used. So, the not() method and associated nested class were
moved to TestPredicates, the tests for not() were moved to
TestPredicatesTest and the Predicates* classes were removed.
That allowed for the removal of the legacy-android-tests as that is now
empty.

TestPredicates has a number of constants that were public. They were
hidden by moving them to the class that actually used them.

A minor generic issue was fixed in AssignableFrom.

Bug: 30188076
Test: make checkbuild and run FrameworkTestRunnerTests
Change-Id: I861da388a4146bb28e1e480d1b7ba9137b7b270e

7 years agoMerge "Avoid use of TimeZoneDistro.equals() method"
Treehugger Robot [Mon, 26 Jun 2017 14:40:28 +0000 (14:40 +0000)]
Merge "Avoid use of TimeZoneDistro.equals() method"

7 years agoMerge "Remove a.t.TestCase"
Paul Duffin [Mon, 26 Jun 2017 14:24:29 +0000 (14:24 +0000)]
Merge "Remove a.t.TestCase"

7 years agoSwitch PackageStatusStorage to XML-based storage
Neil Fuller [Wed, 17 May 2017 03:43:12 +0000 (04:43 +0100)]
Switch PackageStatusStorage to XML-based storage

Based on feedback from a framework developer: an XML file can be used
instead of a Sqlite DB.

To run tests:
make -j30 FrameworksServicesTests
adb install -r -g
"out/target/product/angler/data/app/FrameworksServicesTests/FrameworksServicesTests.apk"
adb shell am instrument -e package com.android.server.timezone -w
com.android.frameworks.servicestests \
    "com.android.frameworks.servicestests/android.support.test.runner.AndroidJUnitRunner"

Test: See above.
Bug: 31008728
Change-Id: I1e6614d26df0e37ccea4dff82867e0b6aba39ca6

7 years agoMerge "IpManager: always log immediate provisioning errors"
Hugo Benichi [Mon, 26 Jun 2017 11:43:11 +0000 (11:43 +0000)]
Merge "IpManager: always log immediate provisioning errors"

7 years agoMerge "Removed a.t.TestRunner and cleanup a.t.TestPrinter"
Paul Duffin [Mon, 26 Jun 2017 11:21:19 +0000 (11:21 +0000)]
Merge "Removed a.t.TestRunner and cleanup a.t.TestPrinter"

7 years agoSwitch to streaming data for time zone update
Neil Fuller [Thu, 22 Jun 2017 13:10:29 +0000 (14:10 +0100)]
Switch to streaming data for time zone update

Switch to streaming data for time zone update rather than
loading it into memory first.

Also make sure that the ParcelFileDescriptor passed to the
service is closed in all cases.

make -j30 FrameworksServicesTests
adb install -r -g \
    "out/target/product/angler/data/app/FrameworksServicesTests/FrameworksServicesTests.apk"
adb shell am instrument -e package com.android.server.timezone -w \
    com.android.frameworks.servicestests \
    "com.android.frameworks.servicestests/android.support.test.runner.AndroidJUnitRunner"

Test: See above
Test: Manual
Bug: 31008728
Change-Id: Ia1e27b204697caee62deb2a3d682800350bca800

7 years agoAvoid use of TimeZoneDistro.equals() method
Neil Fuller [Thu, 22 Jun 2017 13:10:29 +0000 (14:10 +0100)]
Avoid use of TimeZoneDistro.equals() method

The TimeZoneDistro method is going away.

make -j30 FrameworksServicesTests
adb install -r -g \
    "out/target/product/angler/data/app/FrameworksServicesTests/FrameworksServicesTests.apk"
adb shell am instrument -e package com.android.server.timezone -w \
    com.android.frameworks.servicestests \
    "com.android.frameworks.servicestests/android.support.test.runner.AndroidJUnitRunner"

Test: See above
Test: Manual
Bug: 31008728

Change-Id: Id90cf7ce534b888bf6df1bf1dc9d497608dcdbe9

7 years ago[Resources] house keeping in TypeList
Sean Lu [Mon, 26 Jun 2017 10:19:28 +0000 (18:19 +0800)]
[Resources] house keeping in TypeList

Symptom: Type was freed but still may be access
Root Cause: TypeList not cleared
Solution: clear the TypeList
Project: AOSP
Note: None
Test: Yes

Change-Id: I00225165a30d15a299b7a483ea393b4447d2a3fc

7 years agoMerge "Clean up TestCaseUtil"
Paul Duffin [Mon, 26 Jun 2017 09:15:05 +0000 (09:15 +0000)]
Merge "Clean up TestCaseUtil"

7 years agoCarrier Config changes to support IMSI encryption.
pkanwar [Wed, 31 May 2017 20:10:03 +0000 (13:10 -0700)]
Carrier Config changes to support IMSI encryption.

Adding 2 config parameters to CarrierConfigManager:
1. imsi_key_download_url.
2. imsi_key_expiration_days_time.

BUG: 35606429
Test: manual

Change-Id: I191dd0db5fa84a57faea88cb59c2c0b2604064b2

7 years agoIpManager: always log immediate provisioning errors
Hugo Benichi [Thu, 22 Jun 2017 06:19:35 +0000 (15:19 +0900)]
IpManager: always log immediate provisioning errors

And also migrate remaining Log.e usage to logError.

Bug: 62716267
Test: built and flashed
Change-Id: I61a6cec081d2bc0fe3d73e46fd52f1a42684e58c

7 years agoMerge "Relax SharedLogTest asserts"
Hugo Benichi [Sat, 24 Jun 2017 12:19:55 +0000 (12:19 +0000)]
Merge "Relax SharedLogTest asserts"

7 years agoMerge changes I193297e0,I43beb966
Hugo Benichi [Sat, 24 Jun 2017 00:01:22 +0000 (00:01 +0000)]
Merge changes I193297e0,I43beb966

* changes:
  NetworkCapabilities: Added support for TRANSPORT_LOWPAN
  Connectivity metrics: serialize networkId, transports, ifname

7 years agoAdd download request tokens and enforce destination clear
Hall Liu [Tue, 20 Jun 2017 00:20:10 +0000 (17:20 -0700)]
Add download request tokens and enforce destination clear

Start writing tokens for each download request after calling download(),
and check agains them when a completed-file intent is received with a
download request extra.
Also start checking that the download destination is clear.

Change-Id: I533fe2ba49ac56bed5eab44670fe8e9a273b1ffb

7 years agoFix the inconsistency of scan status.
yinxu [Fri, 23 Jun 2017 22:27:42 +0000 (15:27 -0700)]
Fix the inconsistency of scan status.

In types.hal and ril.h, the scan status is defined as:
    PARTIAL = 0x01
    COMPLETE = 0x02

But in NetworkScanResult.java, the scan status is defined as:
    public static final int SCAN_STATUS_PARTIAL = 0
    public static final int SCAN_STATUS_COMPLETE = 1

This CL makes changes to NetworkScanResult.java so that the scan status
is consistent both in Telephony and RIL.

Test: Telephony sanity tests
Bug: 30954762
Change-Id: Iadcd344ad3fe682da98a1a41c2964583f67dc792

7 years agoMerge "Replace MallocHelper with std::unique_ptr."
Elliott Hughes [Fri, 23 Jun 2017 17:51:45 +0000 (17:51 +0000)]
Merge "Replace MallocHelper with std::unique_ptr."

7 years agoMerge "Don't statically link libcrypto in frameworks."
Treehugger Robot [Fri, 23 Jun 2017 17:42:38 +0000 (17:42 +0000)]
Merge "Don't statically link libcrypto in frameworks."

7 years agoMerge "Rename Binder.destroy() to Binder.destroyBinder()."
Vladimir Marko [Fri, 23 Jun 2017 10:36:00 +0000 (10:36 +0000)]
Merge "Rename Binder.destroy() to Binder.destroyBinder()."

7 years agoNetworkCapabilities: Added support for TRANSPORT_LOWPAN
Robert Quattlebaum [Mon, 15 May 2017 22:53:29 +0000 (15:53 -0700)]
NetworkCapabilities: Added support for TRANSPORT_LOWPAN

This change simply adds a new constant, `TRANSPORT_LOWPAN`, for
identifying low-power wireless networks like Thread.

Bug: b/33073713
Merged-In: Ie4aa77496f8ff466fa1a5fbc556e9c029457a689

(cherry pick from commit 557e0c38b98f66d6147206f23d24d8d73b09e3e5)

Change-Id: I21f9b41b8b31c63ceeb1bc9c965f6da2614c356a

Test: runtest frameworks-net (not in original commit message)

7 years agoConnectivity metrics: serialize networkId, transports, ifname
Hugo Benichi [Wed, 15 Mar 2017 14:05:01 +0000 (23:05 +0900)]
Connectivity metrics: serialize networkId, transports, ifname

This patch adds translation from ConnectivityMetricsEvent to
IpConnectivityEvent of recently added fields:
 - top-level network id
 - top-level ifname
 - transports

Also adds inference of link layer from transports or ifname.

At the moment these new fields are not populated in
ConnectivityMetricsEvent. Follow-up patches will fill this gap for
the events of the android.net.metrics package.

Test: new unit tests, $ runtest frameworks-net passes
Bug: 34901696
Merged-In: I563a6a3183470bdfaabb7c781a1beaf6b1058bf0

(partial cherry pick from commit 73fd4d1fafd0208e843abb6a6ad8a46a6fb18143)

Change-Id: I6a00270e73a1bd07f23c367f2394d90a43ced47a

Test: runtest frameworks-net (not in original commit message)

7 years agoRelax SharedLogTest asserts
Hugo Benichi [Fri, 23 Jun 2017 01:06:39 +0000 (10:06 +0900)]
Relax SharedLogTest asserts

This patch makes the asserts of SharedLogTest looser so that it is less
likely to get invalidated by a date formatting change.

The timestamp format of LocalLog changed in commit 6229d990828 from
"mm-dd HH:MM:SS.xxx" to "yyy-mm-ddTHH:MM:SS.xxx".

Bug: 62918393
Test: runtest frameworks-net pass
Change-Id: I57ab3e05a0f3093f9c7b1760e596a3919d4f2ed7

7 years agoMerge "Bluetooth: minor documentation fix to de-confuse"
Marie Janssen [Fri, 23 Jun 2017 03:17:35 +0000 (03:17 +0000)]
Merge "Bluetooth: minor documentation fix to de-confuse"

7 years agoMerge "ASM Priority Boost Tool"
Treehugger Robot [Thu, 22 Jun 2017 22:10:56 +0000 (22:10 +0000)]
Merge "ASM Priority Boost Tool"

7 years agoMerge "Fix code and test related to rules version check"
Treehugger Robot [Thu, 22 Jun 2017 21:23:52 +0000 (21:23 +0000)]
Merge "Fix code and test related to rules version check"

7 years agoDon't statically link libcrypto in frameworks.
Robert Sloan [Wed, 21 Jun 2017 23:15:44 +0000 (16:15 -0700)]
Don't statically link libcrypto in frameworks.

That creates duplicate copies of libcrypto in memory, which confuses
conscrypt.

Test: CtsLibcoreTestCases
Change-Id: I3b03d37feb405f14f8495015be7ff768b4347737
Merged-In: I547bb838befa743d72d356c186ff3cd355736589

7 years agoASM Priority Boost Tool
Alan Leung [Sat, 13 May 2017 00:31:13 +0000 (17:31 -0700)]
ASM Priority Boost Tool

This tool is a replacement for the Jack plugin that allows injection
of static method calls before lock enter and exit.

A common use case would be to boost a thread's priority as soon as
it acquires a heavily contented lock and resetting the priority upon
release.

This tool is meant to be optionally invoked from /build/core/ rules
during the build process.

Test: JUnit Tests / resulting image on a bullhead.

Change-Id: If400414a0bf50f03768a1de2ebee42086a9d701f
Merged-In: If400414a0bf50f03768a1de2ebee42086a9d701f
(cherry picked from commit 93db63059ad5e9e1d59c8ce013559069d566ef4c)

7 years agoBluetooth: minor documentation fix to de-confuse
Marie Janssen [Fri, 13 Jan 2017 00:00:30 +0000 (16:00 -0800)]
Bluetooth: minor documentation fix to de-confuse

It was somewhat unclear which is the preferred method of retrieving a
BluetoothAdapter.  Make it clear that using BluetoothManager is
preferred, and remove references to the old method in BluetoothManager
docs since it is only avaialable in API 18 or higher.

Test: recompile, check that documentation is updated
Fix: 33355430
Change-Id: Ia20b4e45dca03bc2f13c2ab477799b89c5e14f45

7 years agoFix code and test related to rules version check
Neil Fuller [Thu, 22 Jun 2017 16:58:02 +0000 (17:58 +0100)]
Fix code and test related to rules version check

Fix code and test related to an incorrectly implemented
rules version check.

This meant that a device could not return the rules version
it shipped with.

Another unused (and also confusing) method and
associated test is being deleted here.

To run tests:
make -j30 FrameworksServicesTests
adb install -r -g \
  "out/target/product/angler/data/app/FrameworksServicesTests/FrameworksServicesTests.apk"
adb shell am instrument -e package com.android.server.timezone -w \
  com.android.frameworks.servicestests \
  "com.android.frameworks.servicestests/android.support.test.runner.AndroidJUnitRunner"

Test: See above.
Bug: 31008728
Change-Id: I31b55e88072dd73055cb3d8cf252be6eac920322

7 years agoMerge "Add a common call when wrapping an app."
Christopher Ferris [Thu, 22 Jun 2017 18:22:07 +0000 (18:22 +0000)]
Merge "Add a common call when wrapping an app."

7 years agoClean up TestCaseUtil
Paul Duffin [Thu, 22 Jun 2017 12:29:26 +0000 (13:29 +0100)]
Clean up TestCaseUtil

Part of the work of removing JUnit and dependent android.test classes
from the Android API involves providing a static library that developers
can include in their test applications to ease migration. That library
will be built directly from the source (as opposed to android.jar which
is built from stubs) and so developers will be able to see classes and
methods that are not present in the stubs. This change is one of a
number of similar changes that cleanup the existing non-API code in
order to minimize the additional methods and classes exposed externally.
The basic approach is to remove unused classes and methods, use least
visible access modifier possible and generally minimize the amount of
publicly visible code.

TestCaseUtil.getTestCaseNames() is only used by tests but its tests did
provide some coverage of the getTests() method so remove the method and
the tests the method was simply moved into TestCaseUtilTest and the
tests renamed to make it clearer that they are testing
TestCaseUtil.getTests().

Similarly, TestCaseUtil.createTestSuite() was only used by tests but its
tests did provide some coverage of the invokeSuiteMethodIfPossible()
method so the tests were modified and renamed to preserve that coverage.

TestCaseUtil.getTestAtIndex() was completely unused so was just removed.

Bug: 30188076
Test: make checkbuild and ran FrameworkTestRunnerTests
Change-Id: I62bbdbab428d7560f0c7df11f313fe60cfd31d13