Catch an exception sent from native code looking for a specific known
bug and report via wtf which code path is hitting the bug. Revert this
change once bug has been resolved.
Bug: 37298089
Test: manual
Change-Id: Ieb98a8a82a9a2cffe4d0cfbbc8333f453b3e36d5
Signed-off-by: Michael Wachenschwanz <mwachens@google.com>
Make sure to not hold the wm lock when calling into power manager
service, because PWM will acquire a lock that might be contended.
Test: Make sure user activity timeout is still respected on
Keyguard
Test: Have activity with screenBrightness=1.0, make sure screen
is fully bright when opened
Bug: 37888898
Bug: 36631902
Change-Id: I4b5433dbaf8aa151465ae32232d3b3b8597715df
Previously, wakelock max/current/total durations were initialized to -1,
instead of 0. So if, e.g., an app held a full wakelock but not a partial
wakelock, then the partial wakelock would return -1 for these data. This
cl changes that to be 0. The value will only be -1 if the data is not
tracked (as is currently the case for full and window wakelocks), in
which case if the wakelock was held, -1 will be given to signify the
lack of tracking.
Test: manually confirm 0 instead of -1 in batterystats dump.
Change-Id: I934022294ba5adafb2e4d9a5be1dc20aab045cb0
Previously, partial wakelock times reported the blame apportioned to the
app (i.e. the times were pooled). This CL adds another field to provide
the actual time the partial wakelock was held (unpooled, irrespective of
other app's wakelock usage). Full and window wakelocks also now have the
equivalent field, but it always returns -1 (like their max and current
durations), since none of those are currently tracked. Kernel wakelocks
are unchanged.
Test: cts-tradefed run cts-dev -m CtsDumpsysHostTestCases -t android.dumpsys.cts.BatteryStatsDumpsysTest
Bug: 18998184
Change-Id: Ic4cc8c3abba9afa60902d0c8e17e11abeef52a96
This is the Java API for ::android::vintf::RuntimeInfo.
This will be used by CTS for device info report purposes.
Bug: 28656227
Test: cts-tradefed run cts -m CtsEdiHostTestCases --skip-preconditions
Change-Id: Id87c95a17e77d7ec1794a6f850de97edf9ae892d
Batterystats provides an API to count the number of BLE scan results.
Bug: 37720787
Test: runtest -x frameworks/base/core/tests/coretests/src/com/android/internal/os/BatteryStatsTests.java
Change-Id: Idcb7494b39e88dcbfbb3da1ebe90b8a2f8f4d55c
Move aggressive allocation to @SystemApi, which means we can hide
the "flags" API variants.
Remove UUID APIs, since we should use existing Serializable APIs.
Relax permission checks to allow apps to ask for their own stats.
Improve docs.
Test: cts-tradefed run commandAndExit cts-dev -m CtsAppSecurityHostTestCases -t android.appsecurity.cts.StorageHostTest
Bug: 37534687, 37534941, 37718184, 37738770
Change-Id: I6a763fb3ab3169c8d3329765bb31e1ee08d9ced7
When this restriction is enforced Bluetooth sharing option should not be
present when the user tries to share something. Previously this was handled
by explicitly disabling bluetooth sharing activity during managed
provisioning, now this code is to be removed (see topic CLs) and the same
behavior should be achieved by setting this restriction for profile owners
by default.
In Bluetooth:
1) Don't check restrictions on boot, it is invoked anyway through the
listener during boot.
2) Ignore when the restriction is "changed" from true to true - i think
it was the initial intent in that condition.
3) Disable the component for a particular user and not always the
system user. This is something that has to be fixed in O I think since
currently in secondary user the bluetooth itself gets disabled but the
sharing thing still shows up.
In DPMS:
1) Now ActiveAdmin for PO also contains a set of restrictions applied by
default.
2) Now all ActiveAdmins for POs are loaded quite early. That shouldn't
have huge impact though.
Bug: 36249732
Test: run cts -m CtsDevicePolicyManagerTestCases -t com.android.cts.devicepolicy.ManagedProfileTest#testBluetoothSharingRestriction
Test: run cts -m CtsDevicePolicyManagerTestCases -t com.android.cts.devicepolicy.DeviceOwnerTest#testBluetoothRestriction
Test: runtest --path frameworks/base/services/tests/servicestests/src/com/android/server/devicepolicy/DevicePolicyManagerServiceMigrationTest.java
Change-Id: I78c4ffbd503c4a10138e8c0862a9f206f24c5631
Merged-in: I78c4ffbd503c4a10138e8c0862a9f206f24c5631
(cherry picked from commit 7f4ad75218)
tldr; Have the phone check when it reboots if it last shut
down due to heat. If so, we should show the user an
notification and dialog giving them more details.
- Added hidden api to allow apps to query system_server
for the reason the phone last rebooted
- Added notification that is shown when the phone is
booted if the last shutdown was due to excessive
heat.
- Added dialog to provide more details about the
shutdown if the notification is tapped.
- Added tests to verify that file is read and
as expected.
- Updated code for high temperature warning that
is shown while the phone is running to avoid
mixups in the future.
Test: FrameworksServiceTests
Bug: 30994946
Change-Id: Ic25f42539911c89ba7f1834e206f7931d65c2865
Add missing API annotations for permissions and SdkConstants, and
invoke doclava with new "-android" flag.
Test: make -j32 offline-sdk-docs
Bug: 37526420
Change-Id: I970bb2655eb568fd25004636f134c794663a6c33
Update docs based on what new lint detector found. Add new @IntDef
to parameters or methods returning constants or flags, and add
@RequiresPermission to methods mentioning permissions.
Test: make -j32 offline-sdk-docs
Bug: 37526420
Change-Id: I7f640f7883fcb66b911a52ae93b83f77306571ec
Add support for AnyThread, CallSuper, and UiThread.
Another related CL started documenting @RequiresPermission, so remove
duplicated information in existing APIs.
Suppress auto-doc on a handful of classes that are already
well-documented.
Test: make -j32 offline-sdk-docs
Bug: 37526420
Change-Id: I791437dccec0f11d5349a23b982ba098cb551af8
DISALLOW_UNMUTE_MICROPHONE and DISALLOW_ADJUST_VOLUME can
be set by POs on secondary users as well. But they have no
effect on managed profiles.
(In the bug I erroneously mentionted DISALLOW_UNMUTE_DEVICE
which is @hidden instead of *_MICROPHONE).
Bug: 36024135
Test: IntelliJ quickdoc popup renders correctly.
Change-Id: I7a0e77ac993846cdc3a4ddf4d457e1282872107f
Make sure calls to readBuffer() and readEmbeddedBuffer()
get the correct size, parent and offset passed in, so
these can be validated by libhwbinder.
Modified HwBlob to take a length argument as well,
so it can be validated.
Bug: 30498700
Test: hidl_test, hidl_test_java, Youtube, Maps, Netflix, Camera
Change-Id: I28712db97ae29b46acfe952d3d92d1ce5f666a4d
When counting code size, don't include APKs baked into the system.
Settings already accounts external storage used by apps, so they
need a way to exclude that from the total space used by external
storage; give them new getAppSize() API.
Refine docs to explain that emulated storage might be included in
measured statistics.
Resolve symlinks as part of matching getUuidForPath().
Test: cts-tradefed run commandAndExit cts-dev -m CtsAppSecurityHostTestCases -t android.appsecurity.cts.StorageHostTest
Bug: 35812899, 35844919, 37193650
Change-Id: Iec3ce8b336b71dc98a7d25fdd30fa78e9ee826dc
Hook up to android.os.VintfObject.verify().
Bug: 36592877
Test: Flash on marlin; verify with package that has and doesn't have
compatibility.zip entry.
Change-Id: I4e73fa42f4e3fd2e1c5ffec2ffa6152538d62eee
This CL extends the existing API RecoverySystem.verifyPackage() to
verify the compatibility entry in an OTA package. It returns true
directly if such an entry doesn't exist. Otherwise it reads and passes
the contents to VintfObject to perform the verification.
This CL also adds a new system API
RecoverySystem.verifyPackageCompatibility() to verify a given
compatibility zip file. It extracts all the entries and passes them to
VintfObject.verify() to verify the package compatibility. This API is
for streaming A/B OTAs only, where we don't download the full package
but only needed entries.
Bug: 36592877
Bug: 36596980
Test: RecoverySystem.verifyPackage() returns the same result for
packages w/o a compatibility entry.
Change-Id: I038be672868a91820c045d1da100e8e33b23d442
Based on API council feedback, switch to using real UUID objects
instead of Strings. Since UUID is a general-purpose utility class
that will be passed around quite a bit, add it to Parcel and Bundle.
Define well-known namespaced UUID values for "default" and "primary
physical" storage devices, which will let us annotate a bunch of
things with @NonNull.
Define new extras for MANAGE_STORAGE intent that apps can use to
signal where and how much space they'd like the user to free up.
Test: cts-tradefed run commandAndExit cts-dev -m CtsAppSecurityHostTestCases -t android.appsecurity.cts.StorageHostTest
Bug: 37325923, 35812899, 35806020
Change-Id: I8421b126d680f69141a361c1e77223fe2bf4a325
Test: 1. Trigger the confitrmation dialog.
Ensure it looks exactly like the one from settings.
2. Call an API without associating the appa first
Ensure exception is thrown with a message mentioning the need to associate 1st
Change-Id: I94d4116e1988db869ed445ae3fd018c50590e3f4
android.os.VintfObject has two methods:
- report: return device info that can be reported to OTA server
- verify: verify that metadata for a given OTA package is
compatible.
Test: pass
Test: adb shell am instrument -w -e class android.os.VintfObjectTest \
com.android.frameworks.coretests/android.support.test.runner.AndroidJUnitRunner
Bug: 36814503
Change-Id: Iff8fae289eec8ae9cfc327d0d0d36a1cdd5e6800
