- Oct 12, 2023
-
-
Victor Hsieh authored
-
Kweku Adams authored
-
Kweku Adams authored
If the persisted files have somehow accumulated duplicate versions of the same job (identified by uid-namespace-jobId), then drop the duplicates and only keep one. Bug: 289062813 Bug: 296890885 Test: atest JobStoreTest Change-Id: I3179261309031801f0506cf905206376e59af1fe (cherry picked from commit c8fbd937) Merged-In: I3179261309031801f0506cf905206376e59af1fe
-
Kweku Adams authored
-
Kweku Adams authored
The system may be busy when trying to bind to the app, so don't penalize it for slow binding. Now, we'll retain the job that was slow to bind and stop triggering an ANR. Bug: 292090044 Bug: 303533836 Test: atest CtsJobSchedulerTestCases:JobSchedulingTest Change-Id: Ida52d63c5e9710d9d590388d20fceb4fa8f2e0d5 (cherry picked from commit 71fdfd48) Merged-In: Ida52d63c5e9710d9d590388d20fceb4fa8f2e0d5
-
Harshit Mahajan authored
-
Jihoon Kang authored
-
Anton Hansson authored
-
Harshit Mahajan authored
This reverts commit 8d9ef830. Reason for revert: This is incorrectly disabling Rollback Bug: b/287568263 (cherry picked from https://googleplex-android-review.googlesource.com/q/commit:42783c9fbec0ac2f8aede663c74e3cf91ee2646b) (cherry picked from https://googleplex-android-review.googlesource.com/q/commit:dfb2bfeb59065b8464e70dbb103d45c70bc62e01) Merged-In: If5612bd68cd0ba7c3b9e9f68dac84be3f1482164 Change-Id: If5612bd68cd0ba7c3b9e9f68dac84be3f1482164
-
Anton Hansson authored
AudioVolumeGroup is not public API, so cannot be linked from public API javadoc. Bug: 303184203 Test: m docs Change-Id: I5026446e03921f7057b97806afadcd127f37e4fd
-
Treehugger Robot authored
-
Chris Göllner authored
Change-Id: Ia42f7a7b3f3eedd4fa784997686185125014a03c
-
Treehugger Robot authored
-
- Oct 11, 2023
-
-
Jihoon Kang authored
-
Eric Biggers authored
* changes: LockSettingsService: fix names of user lifecycle methods locksettings: Don't use AIDL V1 weaver Reconnect if weaver service connection dies Replace isWeaverAvailable() with getWeaverService()
-
Anton Hansson authored
-
T.J. Mercier authored
-
Treehugger Robot authored
-
Anton Hansson authored
-
Anton Hansson authored
-
Makoto Onuki authored
-
Shaquille Johnson authored
-
Anton Hansson authored
-
Jihoon Kang authored
Currently, hiddenapi_modular defines test api surface as a subset of module lib api surface, causing hiddenapi check failure in from-text stub build. In order to mitigate this and enable the hiddenapi check in from-text stub build, this change introduces a java_api_library module that generates the stub jar that contains both test and module lib api contributions. The module will not be available for consumption, but will only be used in hiddenapi check. Test: m android_test_module_lib_stubs_current.from-text --build-from-text-stub Bug: 191644675 Change-Id: I2b6c9f75d9ebcfd75fa5b509dbef14aa0f88ac97
-
Mark Punzalan authored
Like we currently do for @TestApi and @SystemApi, we look for "@FlaggedApi" and its parameters in the comment above the resource. Bug: 295328308 Test: Did the following: 1. Ran `atest aapt2_tests` 2. Ran `aapt2 link` on public-staging.xml with an added @FlaggedApi in the comment for the resource currently there. Verified the @FlaggedApi annotation is present in the generated R.java. 3. Ran `m framework-minus-apex` and it built successfully. 4. Modified the @FlaggedApi comment with an invalid value (e.g., `Flags.FOO` which does not exist). The build failed. Since @FlaggedApi is a source-only annotation (i.e., not present in the R.class file), we want to verify it was actually used in the build. (cherry picked from https://googleplex-android-review.googlesource.com/q/commit:6fa7a3c7c5213ad3a94406f46fec6d1219b3fa43) Merged-In: I0739f888d40269915c0acea5650c386da1a1782e Change-Id: I0739f888d40269915c0acea5650c386da1a1782e
-
Anton Hansson authored
- the *-check-docs target doesn't do anything the other targets don't - online-* are, I think, the old d.a.c targets but the ds-* targets are what's currently used - the hidden-* target isn't meaningfully different from others, either Test: presubmit Change-Id: Icab977d40917aa14c629629965c6126e20f4f3ee
-
Treehugger Robot authored
-
Anton Hansson authored
-
Anton Hansson authored
They don't seem to do anything. All the docs compile without them and I don't see any additional warnings/errors being printed. If they actually do something, the list is almost certainly incomplete anyway. Test: m docs Change-Id: I1bda7c4cc1d882d80cfe38ee734643c9eed030dc
-
Anton Hansson authored
This gives the docs build access to hidden symbols, which hides issues in the documentation. Baseline the newly discovered issues. Bug: 217734059 Test: m docs Merged-In: I357f5a341a170a37152d836861a330039a83886c Change-Id: I357f5a341a170a37152d836861a330039a83886c
-
Treehugger Robot authored
-
Anton Hansson authored
This reverts commit be284ebf. Reason for revert: OUT_CHANNEL_COUNT_MAX is not public API Change-Id: I9b5562136c17df2b0423c6e1b38927460429dd1c
-
Shaquille Johnson authored
Test: atest FrameworksCoreTests Change-Id: Idee0350fa218ceaaa1f08499e53225f6f99dc86c
-
Hansen Kurli authored
* changes: Coverage for testLegacyLockdownVpn() Create LockdownVpnTrackerTest
-
Hansen Kurli authored
Test coverage for behaviours that occur in ConnectivityServiceTest#testLegacyLockdownVpn(). Bug: 230548427 Test: atest LockdownVpnTrackerTest Change-Id: I4eaa714c1b06530cb33babf4f88688cd8201b5c5
-
- Oct 10, 2023
-
-
Kyle Zhang authored
-
Jack Wu authored
* changes: BatteryManager: make stateOfHealth a public api Add flag for state_of_health_public
-
Eric Biggers authored
Rename some methods to avoid confusion. No change in behavior. Test: presubmit Change-Id: I9885c934be4f99cf556aa6fcfadb6f65faf30e52 Merged-In: I9885c934be4f99cf556aa6fcfadb6f65faf30e52 (cherry picked from commit decb4ee3)
-
Yi-Yo Chiang authored
V1 and V2 reports IWeaver.read() errors differently. V1 throws ServiceSpecificException wrapping an error code. V2 returns the error code in the output parcel. The current client code expects the weaver HAL to be V2 or newer, however it doesn't check if the actual service version is V2 or not. If the service is V1, then IWeaver.read() errors would become unhandled exception, crashing the system_server. Since Weaver AIDL V1 is broken and should never be used, don't use Weaver AIDL V1 services. This fixes a regression in Android 14 where AIDL V1 started being used when available. Bug: 296984182 Bug: 296512452 Test: Boot on old (tm) vendor Change-Id: I32306fb8473c655e68d89d63a1e4f00c8bb5d61f Merged-In: I32306fb8473c655e68d89d63a1e4f00c8bb5d61f (cherry picked from commit 048aaf42)
-
Eric Biggers authored
For robustness reasons, reconnect to the Weaver service if it dies. This should not be relied on, but it's easy to do and is the same thing that LockSettingsService does for the Gatekeeper service. Bug: 287705522 Test: Treehugger Change-Id: Ic06c45353c6b7282e2f22eef13ea1c6e36fe5c58 Merged-In: Ic06c45353c6b7282e2f22eef13ea1c6e36fe5c58 (cherry picked from commit 5b21e30b)
-