OSDN Git Service
Treehugger Robot [Thu, 22 Jun 2017 21:23:52 +0000 (21:23 +0000)]
Merge "Fix 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
Christopher Ferris [Thu, 22 Jun 2017 18:22:07 +0000 (18:22 +0000)]
Merge "Add a common call when wrapping an app."
Treehugger Robot [Thu, 22 Jun 2017 14:17:16 +0000 (14:17 +0000)]
Merge "Ensure that android.test tests pass"
Paul Duffin [Thu, 22 Jun 2017 11:52:18 +0000 (12:52 +0100)]
Ensure that android.test tests pass
It looks as though the tests in test-runner/tests have not actually been
run successfully for over 7 years. As a result they have degraded. This
change fixes the tests so that they will pass and provides instructions
on how to run them.
TestCaseUtilTest.testGetTestCaseNamesForTestSuiteWithSuiteMethod
This fails because it expected 2 names to be returned but only
returns 0. The reason for that is although TwoTestsInTestSuite has a
Test suite() method that does create a TestSuite with two tests the
TestCaseUtil method does not actually call suite(). Instead, because
TwoTestsInTestSuite is a TestSuite it just calls the tests() method
on it which returns an empty Enumeration because TwoTestsInTestSuite
is empty.
The support for "static Test suite() {}" is broken but fixing this
will affect the behavior of InstrumentationTestRunner which is used
in thousands of different places both in and outside Google and so
could cause untold problems.
TestSuiteBuilderTest.testIncludeAllPackagesUnderHere
Reformat the list, one per line and then add missing name
"testPublicConstructor".
ErrorTest/FailingTest
These tests are not meant to be run on their own, only as part of a
separate test. The RunAsPartOfSeparateTest annotation was added to
allow these to be excluded using notAnnotation as shown in the
instructions for running the tests.
Bug:
30188076
Test: followed new instructions in test-runner/tests/Android.mk
Change-Id: I60e7bee9cd08a9ab7777a2578fc58da772de5c1f
Treehugger Robot [Thu, 22 Jun 2017 07:46:34 +0000 (07:46 +0000)]
Merge "ConnectivityManager: fix style issue"
Joel Scherpelz [Thu, 22 Jun 2017 06:34:28 +0000 (06:34 +0000)]
Merge "Use RFC 7217 stable privacy addresses"
Joel Scherpelz [Wed, 7 Jun 2017 06:38:38 +0000 (15:38 +0900)]
Use RFC 7217 stable privacy addresses
Ask Netd to setup WiFi interfaces to use stable privacy addresses. This
should fail benignly on devices with kernels lacking RFC 7217 support.
Test: as follows
- built
- flashed
- booted
- Manually observed persistent stable secret
- Manually observed randomized IPv6 addresses (with flags = 800)
Bug:
17613910
Change-Id: I9465de16685b1eb0fd842446a530cf98d77a0c28
Christopher Ferris [Tue, 20 Jun 2017 23:13:40 +0000 (16:13 -0700)]
Add a common call when wrapping an app.
When an application is wrapped using setprop wrap.<application>, it
doesn't set the special zygote flag, or make the mallopt call. This
means that a wrapped application will behave differently, and not
be able to track zygote native allocations versus app native allocations.
This CL adds a nativePreApplicationInit call that is used after the
Zygote is forked, and when a wrapped application is called.
Bug:
62068500
Test: Booted sailfish system. Verified that the zygote has not called
Test: the new function. Verified the function is called when a wrapper
Test: is used. Verified the function is called when the Zygote forks
Test: a process.
Change-Id: I392e1b5429be77def63f7815b072d68e9408cc7f
(cherry picked from commit
2980de4a59db048851fbbac9cd847a96a12fa4ec)
Chen Xu [Wed, 21 Jun 2017 17:21:37 +0000 (17:21 +0000)]
Merge "migrate roaming operator configs to carrierconfig"
fionaxu [Fri, 2 Jun 2017 01:52:17 +0000 (18:52 -0700)]
migrate roaming operator configs to carrierconfig
Bug:
37885668
Test:runtest --path
packages/apps/CarrierConfig/tests/src/com/android/carrierconfig/CarrierConfigTest.java
Change-Id: Ic669581243a7a640f9ac1b5155954514168cd14b
Merged-In: Ic669581243a7a640f9ac1b5155954514168cd14b
Treehugger Robot [Wed, 21 Jun 2017 09:03:26 +0000 (09:03 +0000)]
Merge "Cleanup a.t.TestGrouping"
Paul Duffin [Wed, 21 Jun 2017 07:46:41 +0000 (07:46 +0000)]
Merge "Remove unused a.t.InstrumentationUtils"
Paul Duffin [Wed, 21 Jun 2017 07:46:04 +0000 (07:46 +0000)]
Merge "Cleanup a.t.InstrumentationTestRunner"
Paul Duffin [Wed, 21 Jun 2017 07:46:02 +0000 (07:46 +0000)]
Merge "Remove unused a.t.s.InstrumentationTestSuiteBuilder"
Paul Duffin [Wed, 21 Jun 2017 07:46:00 +0000 (07:46 +0000)]
Merge "Remove unused android.test.DatabaseTestUtils"
Treehugger Robot [Wed, 21 Jun 2017 07:06:47 +0000 (07:06 +0000)]
Merge "Always allow Ethernet to be an upstream"
Erik Kline [Thu, 9 Mar 2017 02:44:11 +0000 (11:44 +0900)]
Always allow Ethernet to be an upstream
If there are any upstream types defined at all, make sure that
either TYPE_ETHERNET is included somewhere within the sorted list
or force it to be at the front.
Test: as follows
- built
- flashed
- booted
- runtest frameworks-net passes
Bug:
32163131
Bug:
36076442
Change-Id: Ie61d1358f73d518de23f6ca48ca2765ca14a1067
Treehugger Robot [Tue, 20 Jun 2017 14:44:57 +0000 (14:44 +0000)]
Merge "Remove unused a.t.InstrumentationCoreTestRunner"
Paul Duffin [Tue, 20 Jun 2017 13:53:48 +0000 (14:53 +0100)]
Cleanup a.t.TestGrouping
TestGrouping is not part of the API but is used by classes that are and
so will need to be included in the static library we ship as part of the
SDK. As that library will be built directly from the source (as opposed
to android.jar which is built from stubs) developers will be able to see
classes and methods that are hidden and so not present in the stubs.
This change makes TestGrouping and its members as inaccessible as
possible. Unused methods and fields were removed and the TestGrouping
class was simplified by passing the ClassLoader into the constructor,
initializing the classLoader, making the classLoader field final,
and removing the setClassLoader(ClassLoader) method.
Bug:
30188076
Test: make checkbuild
Change-Id: Ie0545b1a61b74cdbc78667417f360b5197e2e128
Paul Duffin [Tue, 20 Jun 2017 13:04:27 +0000 (14:04 +0100)]
Remove unused a.t.InstrumentationUtils
Bug:
30188076
Test: make checkbuild
Change-Id: Ia06e4ae77f2c0ca341ef7c04f5f32bdd6a06d59c
Paul Duffin [Tue, 20 Jun 2017 13:41:20 +0000 (14:41 +0100)]
Cleanup a.t.InstrumentationTestRunner
ARGUMENT_TEST_CLASS, ARGUMENT_TEST_PACKAGE, ARGUMENT_TEST_SIZE_PREDICATE
and ARGUMENT_DELAY_MSEC were made package private because they are not
part of the API and are only used from within the same package.
The JavaDoc comment was changed to a multi-line comment because it is
not and should not be associated with any member.
Fixed a couple of minor JavaDoc issues.
Bug:
30188076
Test: make checkbuild
Change-Id: If28460aec2fd0080e0f5346a83abae393b7a9fb2
Paul Duffin [Tue, 20 Jun 2017 13:03:28 +0000 (14:03 +0100)]
Remove unused a.t.s.InstrumentationTestSuiteBuilder
Bug:
30188076
Test: make checkbuild
Change-Id: I30514123c077e6047227f888f5f1a20e71ee1988
Paul Duffin [Tue, 20 Jun 2017 10:05:25 +0000 (11:05 +0100)]
Remove unused a.t.InstrumentationCoreTestRunner
The only usage of this was in the unused
cts/tests/vm-tests-tf/AndroidManifest.xml which was deleted in
https://android-review.googlesource.com/#/c/418900/.
Bug:
30188076
Test: make checkbuild
Change-Id: I96f3586a7017f4cd14fa31207e0fff19c92f0de4
Paul Duffin [Tue, 20 Jun 2017 13:00:03 +0000 (14:00 +0100)]
Remove unused android.test.DatabaseTestUtils
Bug:
30188076
Test: make checkbuild
Change-Id: I11b1a3b90b9d17af9278ec8b28483eee01b75e81
Treehugger Robot [Tue, 20 Jun 2017 12:06:29 +0000 (12:06 +0000)]
Merge "Track TimeZoneDistroInstaller method changes"
Erik Kline [Tue, 20 Jun 2017 11:01:13 +0000 (11:01 +0000)]
Merge "Create a TetherInterfaceStateMachine when told the interface name."
Erik Kline [Tue, 20 Jun 2017 08:18:27 +0000 (17:18 +0900)]
Create a TetherInterfaceStateMachine when told the interface name.
Test: as follows
- built
- flashed
- booted
- runtest frameworks-net passes
Bug:
32163131
Bug:
62343300
Change-Id: I93f3bdcf20dc4f03cd621d7ab6066344642550e3
Hugo Benichi [Tue, 20 Jun 2017 05:10:14 +0000 (14:10 +0900)]
ConnectivityManager: fix style issue
Simple style fix flagged by linter.
Test: no functional change.
Change-Id: I2cb19912cba149eeaffdd157616d210259d42959
Hugo Benichi [Tue, 20 Jun 2017 05:04:01 +0000 (05:04 +0000)]
Merge "CaptivePortalLogin correctly unregisters callbacks"
Hugo Benichi [Tue, 20 Jun 2017 05:03:22 +0000 (05:03 +0000)]
Merge "Better errors from unregisterNetworkCallback"
Hugo Benichi [Tue, 20 Jun 2017 00:47:55 +0000 (00:47 +0000)]
Merge "DO NOT MERGE Instrument captive portal login activity"
Treehugger Robot [Mon, 19 Jun 2017 21:55:04 +0000 (21:55 +0000)]
Merge "Add an extra level of indirection to make sure that native IHwBinder"
Hall Liu [Mon, 19 Jun 2017 19:53:40 +0000 (19:53 +0000)]
Merge changes from topics 'embms-init-fix', 'embms-download-2'
* changes:
Add callback for initialization done
Embms download part 2
Treehugger Robot [Mon, 19 Jun 2017 19:17:38 +0000 (19:17 +0000)]
Merge "Remove legacy advertiser from map after it's no longer used"
Andreas Huber [Fri, 16 Jun 2017 21:56:12 +0000 (14:56 -0700)]
Add an extra level of indirection to make sure that native IHwBinder
objects, while alive, maintain a strong reference to their Java
counterpart.
Bug:
33042939
Test: hidl_test_java
Change-Id: I69179772b3218705e9ef46cbb10b21b6473280d6
Jakub Pawlowski [Wed, 14 Jun 2017 01:59:07 +0000 (18:59 -0700)]
Remove legacy advertiser from map after it's no longer used
Test: manual
Bug:
62597369
Change-Id: Ief1e1b054f29b59b1ed72514b138d4f209f9847a
Neil Fuller [Mon, 19 Jun 2017 11:56:08 +0000 (12:56 +0100)]
Track TimeZoneDistroInstaller method changes
The TimeZoneDistroInstaller now takes a TimeZoneDistro object
not bytes, which will help to hide whether the distro is in memory
or in storage. See the associated libcore change for details.
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: Ieead1bb87b9ae196abab93342ba5c5747ca45116
Hugo Benichi [Sat, 17 Jun 2017 03:47:33 +0000 (12:47 +0900)]
CaptivePortalLogin correctly unregisters callbacks
The NetworkCallback registered by the CaptivePortalLogin activity in
onCreate was unregistered in both onDestroy() and done(). In addition
done() can be called concurrently from different places (from the
webview, from the captive portal test probe, from the activity menu),
resulting in incorrectly unregistering the callback more than once.
This patch fixes the lifecycle management of the NetworkCallback
registered by the CaptivePortalLogin activity so that it is unregistered
once only in onDestroy.
In addition the done() method is made robust against multiple calls and
becomes a no-op after the first call. This avoids multiple calls
to CaptivePortal for the same captive portal.
Bug:
62497809
Test: tested manually with captive portal setup in the office
Change-Id: I77fbeb55cf91d3b44e91d2fecb800dae40279652
Hugo Benichi [Sat, 17 Jun 2017 04:14:12 +0000 (13:14 +0900)]
Better errors from unregisterNetworkCallback
This patch changes the validation of unregisterNetworkCallback in
ConnectivityManager so that the caller can better distinguish the case
of a callback that was never registered from the case of a callback that
has already been unregistered.
Bug:
62497809
Test: runtest frameworks-net passes
Change-Id: I58eda22725dd4e67dc4b64207e38cf482032df10
Hugo Benichi [Mon, 5 Jun 2017 05:52:24 +0000 (14:52 +0900)]
DO NOT MERGE Instrument captive portal login activity
Also add CAPTIVE_PORTAL_LOGIN_ACTIVITY_SSL_ERROR constants to
metrics_constants.proto for counting ssl error pages shown by the
captive portal login activity.
Bug:
36203355
Bug:
34901696
Bug:
36532213
Test: manually tested with captive portal testing setup while watching
event logs: adb logcat -b events | grep captiveportallogin
06-08 16:25:20.279 21558 21558 I sysui_action: [1004,com.android.captiveportallogin]
06-08 17:01:04.052 21558 21558 I sysui_action: [1007,com.android.captiveportallogin]
Merged-In: Ia05853506d424ad4cb2c11770ae7718fbef3f01c
(cherry picked from commit
c61dc92fc0619d101c4b789f6305012701cc47a0)
(cherry picked with DO NOT MERGE because oc-dev-plus-aosp already
contains a version of this commit compatible with proto2).
Change-Id: I55ea3c61842f16bf6a98e812fb6dc940f96f895c
Treehugger Robot [Mon, 19 Jun 2017 13:00:34 +0000 (13:00 +0000)]
Merge "Make interface IP serving code set LinkProperties"
Treehugger Robot [Mon, 19 Jun 2017 11:29:58 +0000 (11:29 +0000)]
Merge "Documentation: changed tid parameter description for getThreadPriority()"
Alessio Balsini [Wed, 7 Jun 2017 09:41:33 +0000 (11:41 +0200)]
Documentation: changed tid parameter description for getThreadPriority()
Wrong description of the "tid" parameter used by getThreadPriority():
the function does not "change" the priority of the thread associated to
"tid".
Defined also the meaning of setting tid parameter to 0.
Test: None
Change-Id: I8552f3b110eaab1aefa51477e82d3e295547be34
Signed-off-by: Alessio Balsini <alessio.balsini@gmail.com>
Erik Kline [Mon, 12 Jun 2017 09:20:08 +0000 (18:20 +0900)]
Make interface IP serving code set LinkProperties
Additionally, clean up awkward IPv6TetheringInterfaceServices
instantiation mechanics. In future, this class will be absorbed
by TetherInterfaceStateMachine (prior to its renaming to IpServer).
Test: as follows
- built
- flashed
- booted
- runtest frameworks-net passes
Bug:
29337859
Bug:
32163131
Change-Id: Ib620e3df182f9f8e2c019aae1cd8981eb0b29376
Hall Liu [Fri, 16 Jun 2017 21:59:57 +0000 (14:59 -0700)]
Add callback for initialization done
Add callback for initialization done in the framework, and listen to it
in the testapps. Make initialization asynchronous as well for both
download and streaming.
Change-Id: Iea7f803df9d2752401b2eca9f6c7375007cac35e
Treehugger Robot [Sat, 17 Jun 2017 01:25:12 +0000 (01:25 +0000)]
Merge "Simplify NfcCommand and avoid NullPointerExceptions"
Hall Liu [Wed, 7 Jun 2017 20:57:11 +0000 (13:57 -0700)]
Embms download part 2
Add support for multi-part file downloads.
Improves destination directory handling in the download process.
Change-Id: Ibad57bab8804530ce09305424790d5520cd02071
Treehugger Robot [Fri, 16 Jun 2017 23:28:23 +0000 (23:28 +0000)]
Merge "Add a svc command for enabling/disabling Bluetooth"
Myles Watson [Tue, 13 Jun 2017 18:42:36 +0000 (11:42 -0700)]
Add a svc command for enabling/disabling Bluetooth
Bug:
62067926
Test: svc bluetooth disable
svc bluetooth enable
Change-Id: I9a53d7ab1805df98d8cf1292cf1061eacc14655d
Neil Fuller [Fri, 16 Jun 2017 14:37:15 +0000 (14:37 +0000)]
Merge "Track movement of some libcore/tzdata files"
Treehugger Robot [Fri, 16 Jun 2017 13:49:56 +0000 (13:49 +0000)]
Merge "Fixes for tetheroffload crashes"
Erik Kline [Thu, 15 Jun 2017 09:06:34 +0000 (18:06 +0900)]
Fixes for tetheroffload crashes
Now that we can talk to the HALs (with some out of tree CLs and
"setenforce 0"), several crashes were encountered.
Fixes here include:
- avoid hidl_handle move semantics
- check HIDL method status return value (isOk())
- convert Java short port numbers to ints
- don't pass nulls to HIDL where Strings are required
(limitations in parceling)
Test: as follows
- built
- flashed
- booted
- runtest frameworks-net passes
- "setenforce 0" and start tethering
Bug:
29337859
Bug:
32163131
Merged-In: I5fa930be0c9eda491cf17bd4de9b55ab33672d25
Merged-In: Id1037d22826f4d426bccfa17dce0962c54518d64
Change-Id: I91314440c3a04e5f2502579b5f06dac9f25cf0cd
(cherry picked from commit
1185459c4169891bd7550655218e738c8bf1abe2)
Neil Fuller [Tue, 13 Jun 2017 14:12:17 +0000 (15:12 +0100)]
Track movement of some libcore/tzdata files
Some files in libcore/tzdata are moving to system/timezone
under the package com.android.timezone.distro.
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
Merged-In: I2600dd4bee22a571b9eb7303733591b60d1a8b2a
Change-Id: I2600dd4bee22a571b9eb7303733591b60d1a8b2a
Treehugger Robot [Thu, 15 Jun 2017 23:32:57 +0000 (23:32 +0000)]
Merge "When updating a split app, copy compiled files from base.apk only."
Myles Watson [Tue, 13 Jun 2017 23:07:32 +0000 (16:07 -0700)]
Simplify NfcCommand and avoid NullPointerExceptions
Test: adb root
adb shell
stop
svc nfc enable
start
svc nfc disable
Change-Id: Ie5170f0525f8830d1d80b5607191ce95e78a55ae
Ying Xu [Thu, 15 Jun 2017 16:50:09 +0000 (16:50 +0000)]
Merge "Use java.util.Objects.hashCode to get hashcode."
Jason Monk [Thu, 15 Jun 2017 15:21:36 +0000 (15:21 +0000)]
Merge "NfcTile: make sure NFC adapter is always available when needed"
Anas Karbila [Thu, 15 Jun 2017 13:36:53 +0000 (22:36 +0900)]
NfcTile: make sure NFC adapter is always available when needed
Test: enable NFC->reboot->directly open QS panel, NFC tile will show disabled state. This commit fixes it.
Change-Id: Iabbe4ec11a38614b9c65627d58c6ee9ec93435f3
yinxu [Wed, 14 Jun 2017 17:31:40 +0000 (10:31 -0700)]
Use java.util.Objects.hashCode to get hashcode.
Test: Telephony sanity tests
Change-Id: I8b845f2e5e44b242278b8955807d1b40e67e1414
Treehugger Robot [Wed, 14 Jun 2017 10:11:58 +0000 (10:11 +0000)]
Merge "Merge IPv6TetheringInterfaceServices into TetherInterfaceStateMachine"
Ying Xu [Wed, 14 Jun 2017 02:16:51 +0000 (02:16 +0000)]
Merge "Adds error codes and maximum RAN/Bands/Channels"
Treehugger Robot [Tue, 13 Jun 2017 21:49:46 +0000 (21:49 +0000)]
Merge "Do not CloseGuard KernelID or FieldID"
Robert Greenwalt [Tue, 13 Jun 2017 20:38:49 +0000 (20:38 +0000)]
Merge "Add updateMethod callback"
Treehugger Robot [Tue, 13 Jun 2017 20:30:37 +0000 (20:30 +0000)]
Merge "remove the lock-use for prunePrintService in case blocking main thread"
yinxu [Wed, 31 May 2017 01:04:19 +0000 (18:04 -0700)]
Adds error codes and maximum RAN/Bands/Channels
Added error codes which are mapped from the RadioError which is returned
from RIL, also added some errors which will be generated at telephony.
Those errors will be returned to the user with the onError() callback.
Added the maximun number of RAN/Bands/Channels in 1 scan request.
Test: Telephony sanity tests
Bug:
30954762
Change-Id: Ie7865ed2101d9564e26870a8ebbb0ceb7bd9bd1b
Pankaj Kanwar [Tue, 13 Jun 2017 20:22:22 +0000 (20:22 +0000)]
Merge "Radio Interface changes to pass the encryption object, including the key,to the modem."
Yang Ni [Fri, 28 Apr 2017 15:45:40 +0000 (08:45 -0700)]
Do not CloseGuard KernelID or FieldID
Bug:
28053584
Stop CloseGuarding for two reasons:
1) KernelID and FieldID objects are constructed in auto-generated
(RenderScript reflected) Java code. It would be impossible for a user to
explicitly call destroy() on them. Guarding them would leave a lot of
noisy warnings in logcat.
2) These KernelID and FieldID objects are not big compared to other
RenderScript objects, e.g. Allocations. They occupy almost no native
resources except for a native pointer. Leaving their destruction to Java
GC would be completely acceptable, since any delay in reclaiming them is
unlikely to cause memory pressure.
Test: CTS on x86_64 emulator
Change-Id: I587b5561a0b2bdbf0b2e95bf2995c20d5f5faf9d
Andreas Gampe [Tue, 13 Jun 2017 18:34:45 +0000 (18:34 +0000)]
Merge "Binder: Add more logging on Error"
Jeff Hao [Tue, 13 Jun 2017 18:09:10 +0000 (11:09 -0700)]
When updating a split app, copy compiled files from base.apk only.
This fixes issues with some splits failing to be recompiled if they
haven't changed, but the dex files they depend on have.
The real fix will be for frameworks to generate the new expected
classpath and check in DexFile.getDexOptNeeded. Then we can undo
this change and copy over all the compiled split files again.
Bug:
62269291
Test: cts-tradefed run singleCommand cts -d --module
CtsAppSecurityHostTestCases -t android.appsecurity.cts.SplitTests
Change-Id: Id8e6dc59b16d4f10a6c2d9d81027012204e0f26b
pkanwar [Mon, 22 May 2017 22:23:38 +0000 (15:23 -0700)]
Radio Interface changes to pass the encryption object, including the
key,to the modem.
We will now pass the ImsiEncryptionInfo object which includes mnc/mcc.
BUG:
35606429
Test: manual
Change-Id: Idf835d9b99a26cb0c0d73fb70ffa9875843ac62b
Treehugger Robot [Tue, 13 Jun 2017 15:27:09 +0000 (15:27 +0000)]
Merge changes I73ae8de1,I58260902
* changes:
Binder: Log details of pending exception.
Binder: Be forceful about a forceful exit.
Erik Kline [Tue, 13 Jun 2017 12:32:10 +0000 (21:32 +0900)]
Merge IPv6TetheringInterfaceServices into TetherInterfaceStateMachine
Ideally this would have been done immediately after IPv6 tethering code
was made part of an official named release. Getting this in now should
help with the relaying of LinkProperties that is required for offload.
Test: as follows
- built
- flashed
- booted
- runtest frameworks-net passes
- runtest frameworks-wifi passes
Bug:
29337859
Bug:
32163131
Change-Id: I015fa3cb05a15624f9e80b7aa5a871cefef431b7
Erik Kline [Tue, 13 Jun 2017 08:20:40 +0000 (08:20 +0000)]
Merge "make UpsreamNetworkMonitor the sole source of upstream network information"
Treehugger Robot [Tue, 13 Jun 2017 03:13:18 +0000 (03:13 +0000)]
Merge "Only disable Wi-Fi IP serving on specific interface, if available"
Erik Kline [Fri, 9 Jun 2017 08:08:52 +0000 (17:08 +0900)]
make UpsreamNetworkMonitor the sole source of upstream network information
Test: as follows
- built
- flashed
- booted
- runtest frameworks-net passes
Bug:
32163131
Change-Id: I66ce682cf9c31887264a095a4e4638d281d8ed39
Erik Kline [Fri, 9 Jun 2017 07:36:29 +0000 (16:36 +0900)]
Only disable Wi-Fi IP serving on specific interface, if available
Test: as follows
- built
- flashed
- booted
- runtest frameworks-net passes
Bug:
32163131
Change-Id: Ifd9131bad20810ee24c9436d8c0df7a81fae11d5
Andreas Gampe [Mon, 12 Jun 2017 17:43:05 +0000 (10:43 -0700)]
Binder: Add more logging on Error
Try to add the Error's message to the abort, so that it shows up
when the logcat is missing.
Sample message:
'jni_internal.cc:508] JNI FatalError called: java.lang.Error thrown during binder transaction: java.lang.LinkageError: This is a test.'
Bug:
62514767
Test: m
Change-Id: I12026bb7a8ec9438db493f135ed5d0177187f702
Pankaj Kanwar [Tue, 13 Jun 2017 00:28:42 +0000 (00:28 +0000)]
Merge "Address IMSI privacy issues."
pkanwar [Tue, 21 Feb 2017 03:23:55 +0000 (19:23 -0800)]
Address IMSI privacy issues.
Added getCarrierInfoForImsiEncryption and
setCarrierInfoForImsiEncryption methods.
CP from Master
Bug:
35606429
Test: runtest --path ..ImsiEncryptionInfoTest.java
Change-Id: I62bdc8ac5e57b749fc5675c28579ec2d9183643d
Narayan Kamath [Tue, 18 Apr 2017 16:48:39 +0000 (17:48 +0100)]
Binder: Log details of pending exception.
This was accidentally removed by change
d64abfcf93b59500a0d.
Test: manual; verify exception trace is printed in logs.
Bug:
36813403
Bug:
37435516
Change-Id: I73ae8de167c457d56fddb85943a8f08f8913f0f7
(cherry picked from commit
ccc9fad9e6abfa3cf082567669cc5d03ec7a3086)
Narayan Kamath [Wed, 12 Apr 2017 10:50:10 +0000 (11:50 +0100)]
Binder: Be forceful about a forceful exit.
We were previously using exit(1) when code servicing an IPC threw
any subclass of Error. That made it much harder to diagnose cases
where that happened because :
- exit runs global destructors, which might prove problematic (see
linked bug).
- such exits are often due to bugs in application code (things like
AssertionErrors being thrown) but aren't flagged as such by our
infrastructure, or by humans for that matter.
To address both issues, use FatalError() so that the runtime can dump
more useful information to the logs before it aborts.
Test: manual
Bug:
36813403
Change-Id: I5826090229109dc7cb19f0c3571c609f990cd36a
(cherry picked from commit
d64abfcf93b59500a0dba1626e73861848eb4407)
Marie Janssen [Mon, 12 Jun 2017 22:36:05 +0000 (22:36 +0000)]
Merge "Protect broadcast used in settings for bond cancel"
Marie Janssen [Fri, 9 Jun 2017 23:38:19 +0000 (16:38 -0700)]
Protect broadcast used in settings for bond cancel
Test: cancel pairing from notification tray
Bug:
62094630
Change-Id: Ia83a874c1ea9a1d9d5502609f81e94052bb5759c
yangbingqian [Mon, 12 Jun 2017 14:45:50 +0000 (22:45 +0800)]
remove the lock-use for prunePrintService in case blocking main thread
As in the change: I4f4cdaba65132dc2ef054877cbb097b499a723f6
the lock object is removed when calling RemotePrintSpooler
and so as it in the method prunePrintService(). But in the
constructor of UserState, the mLock is still held when calling
prunePrintService(), it may also block the main thread
It is better to take it out of the synchronized block
Test: null
Change-Id: I709b491d611dbcbf21a5fd493b879ed290dd0247
Signed-off-by: yangbingqian <yangbingqian@xiaomi.com>
Robert Greenwalt [Sat, 10 Jun 2017 00:03:56 +0000 (00:03 +0000)]
Merge "Changed Locale in ServiceInfo to a list."
Hall Liu [Fri, 9 Jun 2017 21:44:27 +0000 (21:44 +0000)]
Merge "Update documentation for streaming methods"
Jordan Liu [Fri, 9 Jun 2017 17:25:08 +0000 (17:25 +0000)]
Merge "Add IApnSourceService.aidl"
Hugo Benichi [Fri, 9 Jun 2017 11:31:21 +0000 (11:31 +0000)]
Merge "Revert "Instrument captive portal login activity""
Hugo Benichi [Fri, 9 Jun 2017 11:30:05 +0000 (11:30 +0000)]
Revert "Instrument captive portal login activity"
This reverts commit
f867da14eaa194cb53b0726b50e570268e70f1d7.
Change-Id: Ic900c91ed87c2d65c151f8386954bd7b74128348
Erik Kline [Fri, 9 Jun 2017 09:06:39 +0000 (09:06 +0000)]
Merge "Convert IControlsTethering from interface into callback"
Hugo Benichi [Fri, 9 Jun 2017 06:39:19 +0000 (06:39 +0000)]
Merge "Instrument captive portal login activity"
Hugo Benichi [Fri, 9 Jun 2017 06:38:09 +0000 (06:38 +0000)]
Merge "New Metrics constants for captive portal related UI"
Hugo Benichi [Mon, 5 Jun 2017 05:52:24 +0000 (14:52 +0900)]
Instrument captive portal login activity
Also add CAPTIVE_PORTAL_LOGIN_ACTIVITY_SSL_ERROR constants to
metrics_constants.proto for counting ssl error pages shown by the
captive portal login activity.
Bug:
36203355
Bug:
34901696
Bug:
36532213
Test: manually tested with captive portal testing setup while watching
event logs: adb logcat -b events | grep captiveportallogin
06-08 16:25:20.279 21558 21558 I sysui_action: [1004,com.android.captiveportallogin]
06-08 17:01:04.052 21558 21558 I sysui_action: [1007,com.android.captiveportallogin]
Merged-In: Ia05853506d424ad4cb2c11770ae7718fbef3f01c
(cherry picked from commit
c61dc92fc0619d101c4b789f6305012701cc47a0)
Change-Id: I9184f6263a003c2be28f41a550c6164c8d6ee907
Jordan Liu [Tue, 23 May 2017 20:07:07 +0000 (13:07 -0700)]
Add IApnSourceService.aidl
This service is bound to by TelephonyProvider on devices where we don't
use the apns-conf.xml file.
Bug:
37283190
Test: manual
Change-Id: I16da1c5acc5ba1fe0237cc0d37c73d831c2d1508
Merged-In: I16da1c5acc5ba1fe0237cc0d37c73d831c2d1508
Mathieu Chartier [Thu, 8 Jun 2017 21:44:06 +0000 (21:44 +0000)]
Merge "Add system property for hot startup method threshold"
Jakub Pawlowski [Thu, 8 Jun 2017 20:59:33 +0000 (20:59 +0000)]
Merge "Bluetooth: fix connectGatt overload invocations"
Jakub Pawlowski [Wed, 7 Jun 2017 20:45:33 +0000 (13:45 -0700)]
Bluetooth: fix connectGatt overload invocations
Test: sl4a GattOverBrEdrTest
Bug:
62387078
Change-Id: I5bfb87683e25f8862587e8000eb4c9bb6ff14f99
Jesse Hall [Thu, 8 Jun 2017 15:34:42 +0000 (15:34 +0000)]
Merge "Allow disabling GL preload even if gfxdriver isn't updatable"
Erik Kline [Tue, 6 Jun 2017 10:24:21 +0000 (19:24 +0900)]
Convert IControlsTethering from interface into callback
Additionally:
- add updateLinkProperties() method to new callback
- skeletally connect LinkProperties updates through
to OffloadController
TODOs liberally sprinkled through out.
Test: as follows
- built
- flashed
- booted
- runtest frameworks-net passes
Bug:
29337859
Bug:
32163131
Change-Id: I631d17b26be153534551a1615931fc98b598b953