crash_dump needs to acquire CAP_SYS_PTRACE to be able to ptrace
processes with capabilities. selinux should hopefully be sufficient for
restricting ptrace to processes that should do it.
Bug: http://b/34694637
Test: debuggerd `pidof system_server`
Change-Id: If46f0b9baa54529780f7767f309f76b102a42ab4
Test: Post notification with MessagingStyle.addHistoricMessage(), verify they show up while direct replying.
Change-Id: I26b801d05305a5251fe0d6fae1fe1c24064105a9
This fixes the inconsistency between no profile and a
removed profile.
Change-Id: I85d37ce42113bde8567be382372c5f4997601e85
Fixes: 34699518
Test: runtest -x
frameworks/base/services/tests/servicetests/src/com/android/server/pm/UserManagerTest.java
SINGLE_TOP is only guaranteed to work within the current task stack.
If multiple stacks are inactive in the background (eg. temporarily
switching from freeform mode to fullscreen mode by maximising one
activity) some tasks might get missed out and stay visible.
Test: manual -- start two work activities, one in fullscreen & one in freeform. start a personal
Test: activity in freeform too. switch to the fullscreen work app and lock the profile using
Test: `adb shell service call trust 6 i32 $USER_ID i32 1`
Test: __outcome after change__
Test: a) locking work profile should bring up the work password prompt in fullscreen
Test: b) skipping the prompt and switching to the personal freeform app should keep the freeform
Test: work app hidden by WorkLockActivity
Test: __outcome before change__
Test: Only (a) happened, for step (b) the work app was still visible
Bug: 31001762
Change-Id: I5979c5b12670a21c7bcd59e222fc518d395e3e29
Disable BluetoothOppLauncherActivity component if the
UserManager#BLUETOOTH_DISALLOWED user restriction is set.
Test: cts-tradefed run cts -m CtsDevicePolicyManagerTestCases --test
com.android.cts.devicepolicy.DeviceOwnerTest#testBluetoothRestriction
Bug: 32895060
Bug: 34239380
Change-Id: I3467c82881896f276f542583bb50807b5e27276b
Without this, the mocked version looks like it's calling
the default implementation (which immediately fails due to
a bunch of null fields) - with this it just returns null.
I would ultimately like to stop mocking out the controller
and instead use a mocked package manager (which there is
already some set up to do). But in the meanwhile, this at least
gets the test passing.
Test: Checked the test passes.
Change-Id: I92d1f2796cfce21bf55e927bd56690c582f41e91
This View is being used by several projects and so
I moved it to settingslib so there are not multiple
copies floating around.
Test: Manual & robotests for projects that use it
Bug: 34621372
Change-Id: I8d51a1b5383fad977754eeeb78dce545e506be04
- Check if the MB is not set for any record apart from the first one.
- If the record is chunked, first record must contain the type
Bug: 20722275
Test: Manual; write raw NDEF messages to a tag
Change-Id: I3e73fc5163bbcd8a1f23417e98f57b665f4f1fad
Signed-off-by: Ruchi Kandoi <kandoiruchi@google.com>
Added Policy to PasspointConfiguration and the corresponding parser
support in PPSMOParser.
While there, fix a typo in node name "CertSHA256Fingerprint" under
CertificateCredential.
Bug: 34198926
Test: frameworks/base/wifi/test/runtests.sh
Change-Id: Iabe27cd83b6658ed7d4f895d7fe2255fe2094ebb
Before
======
Enable log:
01-17 09:43:18 Enabled by com.android.systemui
01-18 07:13:10 Disabled by com.andorid.systemui
1 BLE Apps registered:
com.google.android.gms
After
=====
Enable log:
02-06 07:49:07 Enabled by system boot
02-06 07:49:35 Disabled by com.andorid.systemui
1 BLE App registered:
com.google.android.gms
Test: run bugreport, see differences in dumpsys
Change-Id: I6086f7f97c1fcbdb860d70d9516fd5bf630a9f5e
