For feature type "embedded", launch IoTSystemService from
System Server.
Test: Wrote a IotSystemService and ensured that the service is
successfully started on an iot device.
Bug: 67641069
Change-Id: I5aa0408f35fb4a5b0787cf9b20454720d8ae130f
Cherry-picked and refactored from oc-iot-dev branch
Register services to be called as soon as a bug report is taken so relevant debug info can be captured.
dumpsys --priority CRITICAL times: adb shell dumpsys --priority CRITICAL | grep "was the duration"
0.005s was the duration of dumpsys SurfaceFlinger
0.002s was the duration of dumpsys cpuinfo
0.003s was the duration of dumpsys activity
0.025s was the duration of dumpsys window
Bug: 31774394
Test: adb shell setprop dumpstate.version "2.0-dev-priority-dumps" && \
adb bugreport ~/tmp_new.zip
Test: adb shell dumpsys --priority CRITICAL
Test: adb shell dumpsys --priority HIGH
Test: adb shell dumpsys --priority NORMAL
Change-Id: Ic656a05dd9a838f0b13043b4387dae7b5c691f7e
Test: reboot device - no crashes
Test: Ran privapp_permissions.py for fugu/marlin/taimen
Bug: 64693550
Change-Id: I9522ccaa75858b9a3d153c95c529eff70cd13994
Use the presence of FEATURE_WEBVIEW to determine whether a device is
intended to have a WebView implementation or not, instead of trying to
fall back to NullWebViewFactoryProvider after loading WebView fails.
This removes the need for nullwebview entirely, and eliminates a class
of possible bug where unexpected exceptions during loading cause the
fallback mechanism not to work reliably.
On devices which don't have the feature, don't start
WebViewUpdateService at all. Guard all the places which try to access
the service to return reasonable (empty/null) results when this is the
case, instead of throwing exceptions.
Change-Id: I839adaaf0abee28f4989e3fbc0c49c0732d0ec1c
Bug: 31849211
Fixes: 28529980
Test: on wear and non-wear, cts-tradefed run cts -m CtsWebkitTestCases
Add PackageManager API which allows apps to test whether the platform
officially supports the Wi-Fi RTT API.
Only start the RTT service if the device supports it.
Bug: 65014628
Test: integration tests pass
Change-Id: Ie88bd29ccd592e9a29ff24e8cc346afc360f03b0
Replace existing Wi-Fi RTT manager framework. Creating new framework
in parallel since there's code using the new framework - will be
switched over once new version ready.
New version is AIDL-based.
Bug: 65014552
Test: unit tests and integration tests
Change-Id: Id468c9b2a3c94eb30959f5ac5e4b1688fb8fc633
Keeping the legacy BackupManagerService for now, but we'll remove
it in a few weeks if there's no issues noticed.
In case of any problems, this CL can be cleanly reverted as a fallback.
Test: pretty much all we got in store
1) runtest -p com.android.server.backup frameworks-services
2) make gts -j40 && gts-tradefed run gts -m GtsBackupHostTestCases
3) make cts -j40 && cts-tradefed run cts -m CtsBackupTestCases
4) make pts -j40 && pts-tradefed run pts -m PtsBackupHostSideTestCases
5) manual: "adb shell bmgr backupnow --all" and then cloud restore,
D2D scenarios (both as source and target)
Bug: 65823538
Change-Id: I701764995c4fe2611a4941aecb45315aa457658e
Fix reliability triggering in the time zone update
feature. The reliability triggering is there to
resume an update check when things go wrong. For
example, if the device is turned off after a
time zone data app is updated but before the time
zone update process has finished pulling the data
out and installing it in /data. The reliability
trigger happens when the device is idle and will
complete the work.
The idle triggering currently doesn't work as it
was implemented incorrectly.
The code has been changed to use the
JobScheduler to trigger on idle and scheduling
has been made explicit.
Unit tests run with:
make -j30 FrameworksServicesTests
adb install -r -g \
"out/target/product/marlin/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 testing: adb shell am idle-maintenance, adb logcat/adb shell dumpsys timezone
Test: PTS: run pts -m PtsTimeZoneTestCases
Bug: 65657543
Change-Id: Ibb2c537955d09a5d538c50490dae94a518522f19
Merged-In: Ibb2c537955d09a5d538c50490dae94a518522f19
(cherry picked from commit cd1a109b87)
Fix reliability triggering in the time zone update
feature. The reliability triggering is there to
resume an update check when things go wrong. For
example, if the device is turned off after a
time zone data app is updated but before the time
zone update process has finished pulling the data
out and installing it in /data. The reliability
trigger happens when the device is idle and will
complete the work.
The idle triggering currently doesn't work as it
was implemented incorrectly.
The code has been changed to use the
JobScheduler to trigger on idle and scheduling
has been made explicit.
Unit tests run with:
make -j30 FrameworksServicesTests
adb install -r -g \
"out/target/product/marlin/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 testing: adb shell am idle-maintenance, adb logcat/adb shell dumpsys timezone
Test: PTS: run pts -m PtsTimeZoneTestCases
Bug: 65657543
Change-Id: Ibb2c537955d09a5d538c50490dae94a518522f19
Merged-In: Ibb2c537955d09a5d538c50490dae94a518522f19
(cherry picked from commit cd1a109b87)
Fix reliability triggering in the time zone update
feature. The reliability triggering is there to
resume an update check when things go wrong. For
example, if the device is turned off after a
time zone data app is updated but before the time
zone update process has finished pulling the data
out and installing it in /data. The reliability
trigger happens when the device is idle and will
complete the work.
The idle triggering currently doesn't work as it
was implemented incorrectly.
The code has been changed to use the
JobScheduler to trigger on idle and scheduling
has been made explicit.
Unit tests run with:
make -j30 FrameworksServicesTests
adb install -r -g \
"out/target/product/marlin/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 testing: adb shell am idle-maintenance, adb logcat/adb shell dumpsys timezone
Test: PTS: run pts -m PtsTimeZoneTestCases
Bug: 65657543
Change-Id: Ibb2c537955d09a5d538c50490dae94a518522f19
Introduces StatsCompanionService.java and sets it up as a system service.
This service is a helper for statsd. It will be responsible for setting
and cancelling alarms related to polling stats and anomaly detection.
It currently does not have selinux permission, so must be started
manually (or with selinux temporarily disabled).
Test: disable enforcing selinux, then set and cancel alarms and make
sure they work.
Change-Id: I3bd73acdd998ee424696cce40965134c14220d8f
...VOICE_RECOGNITION_SERVICE that is needed for SearchSpeechServices
System setup was incorrectly completely runing off
VoiceInteractionManagerService if the associated feature is not set,
but the service needs to always be running, and instead rely on
the existing logic of reducing functionality if it should not be
running full voice interaction services.
Test: manually booted and ran
Bug: 65055576
Change-Id: I9a83216d45689440c71d657ba2721faf0662b0ff
Create new BootTimingsTraceLog object for logging when starting HIDL
services from different thread, in order to avoid exception.
Bug: 64111691
Test: Manual - device boots, metrics are logged to system/event log
Change-Id: I05834b7fe8078fbf772f9b29f0dd3aa4288f8f23
Renamed BootTimingsTraceLog to TimingsTraceLog. It is now used for
boot and shutdown logging.
Added measurements for main stages of shutdown in the system server:
ShutdownTiming: SendShutdownBroadcast took to complete: 734ms
ShutdownTiming: ShutdownActivityManager took to complete: 203ms
ShutdownTiming: ShutdownPackageManager took to complete: 17ms
ShutdownTiming: ShutdownBt took to complete: 533ms
ShutdownTiming: ShutdownRadio took to complete: 534ms
ShutdownTiming: ShutdownNfc took to complete: 1536ms
ShutdownTiming: ShutdownRadios took to complete: 1538ms
ShutdownTiming: ShutdownStorageManager took to complete: 906ms
ShutdownTiming: SystemServerShutdown took to complete: 3918ms
Bug: 64569080
Test: shutdown/reboot and check logs
Change-Id: I636c045852cd1ed6be1c58af6608f70e95756389
some additional loggings
Test: manually flashed to device and test calling IncidentManager.java
functions
Change-Id: I69610414edde865b20ba632837f037fbe43f99db
This reverts commit 27c461ab3c.
I'm taking alternative approach, telephony code will treat "GMT" value as it would be unset. This is fine because GMT is a synthetic value that would never be set from other sources.
Change-Id: I889a844bfd75b7730a5d967065d27de03ef3ce2f
Test: flashed device, tested manually
Bug: 64056758
The tracing call was correct for aosp/master, but the tracing
code had been changed internally. The incorrect call meant it
wasn't actually tracking the points we're interested in and
it was reporting incorrectly (by ~1000 millis).
Bug: 64141572
Test: boot device / adb logcat
Change-Id: If11354d1d73cf90973ed4f43e885ae28ca1346ea
(cherry picked from commit a87cd5daa0)
Modify the condition used for starting RulesManagerService:
use mOnlyCore rather than disableNonCoreServices. mOnlyCore is
important for the secure lock screen and when mOnlyCore is true
the RulesManagerService cannot perform trust checks. Therefore,
disable the RulesManagerService when mOnlyCore == true.
disableNonCoreServices is apparently not used.
Bug: 64057583
Test: Boot device with time zone updates enabled and with a lock
pattern set.
Change-Id: Idd7cb6e11b1a91dfe7394f6c98f8fa3e0eec9dce
(cherry picked from commit 4fbaee8f49)
Telephony code depends on persist.sys.timezone not being set as a signal
to deduce a timezone from sim card information.
This reverts commit 20e020536e.
Bug: 64056758
Change-Id: I8db204bcffbdb960124b1ae0e02d53b27dd310a4
Modify the condition used for starting RulesManagerService:
use mOnlyCore rather than disableNonCoreServices. mOnlyCore is
important for the secure lock screen and when mOnlyCore is true
the RulesManagerService cannot perform trust checks. Therefore,
disable the RulesManagerService when mOnlyCore == true.
disableNonCoreServices is apparently not used.
Bug: 64057583
Test: Boot device with time zone updates enabled and with a lock
pattern set.
Merged-In: Idd7cb6e11b1a91dfe7394f6c98f8fa3e0eec9dce
Change-Id: Idd7cb6e11b1a91dfe7394f6c98f8fa3e0eec9dce
