Example output:
$ adb shell dumpsys attestation_verification
AttestationVerificationManagerService
    Event Log:
        Verification #2 [2024-04-25 22:09:52]
            Result: FAILURE(2)
            Certificate Chain Valid (inc. Trust Anchor): FAILURE
        Verification #1 [2024-04-25 22:08:37]
            Result: FAILURE(2)
            Certificate Chain Valid (inc. Trust Anchor): FAILURE

Seeing a result from dumpsys that doesn't fail will need to wait until
there is a user-signed build to test on a locked device. However, here's
an example output from a unit test:

04-25 15:23:23.954   408   467 D AVFTest : Verification #1 [2024-04-25 22:23:23]
04-25 15:23:23.954   408   467 D AVFTest :  Result: FAILURE(2)
04-25 15:23:23.954   408   467 D AVFTest :  Certificate Chain Valid (inc. Trust Anchor): OK
04-25 15:23:23.954   408   467 D AVFTest :  Local Binding: OK
04-25 15:23:23.954   408   467 D AVFTest :   Binding Type: 3
04-25 15:23:23.954   408   467 D AVFTest :  System Ownership: FAILURE
04-25 15:23:23.954   408   467 D AVFTest :  KeyStore Attestation Parameters
04-25 15:23:23.954   408   467 D AVFTest :   OS Version >= 10: OK
04-25 15:23:23.954   408   467 D AVFTest :   OS Patch Level in Range: OK
04-25 15:23:23.954   408   467 D AVFTest :   Attestation Version >= 3: OK
04-25 15:23:23.954   408   467 D AVFTest :   Keymaster Version >= 4: OK
04-25 15:23:23.954   408   467 D AVFTest :   Keymaster HW-Backed: OK
04-25 15:23:23.954   408   467 D AVFTest :   Key is HW Backed: OK
04-25 15:23:23.954   408   467 D AVFTest :   Boot State is VERIFIED: OK
04-25 15:23:23.954   408   467 D AVFTest :   Verified Boot is LOCKED: OK
04-25 15:23:23.954   408   467 D AVFTest :   Key Boot Level in Range: OK
04-25 15:23:23.954   408   467 D AVFTest :   Key Vendor Patch Level in Range: OK

Bug: 335498868
Test: atest AttestationVerificationTest
Change-Id: Iea0b19b3c3a25c5800ebd6fad85c0829a2cfa2b1
Flag: android.security.dump_attestation_verifications

Merge "[DeviceAwareAppOp] Fix the bug that misuses device Id in proxy and proxied attribution source" into main

In Android U, all receivers registering for non-system broadcasts
are required to include a flag indicating the receiver's exported
state. Apps registering for non-system broadcasts should use the
ContextCompat#registerReceiver APIs with flags.

Fixes: 339811510
Bug: 331804186
Test: lunch sdk_phone64_x86_64-trunk_staging-userdebug; time m -j90; \
emulator -wipe-data -no-snapshot -logcat *:V -logcat-output logcat.txt
Test: m RUN_ERROR_PRONE=true
Test: atest FaceInternalEnumerateClientTest
            FingerprintInternalEnumerateClientTest
	    BiometricDanglingReceiverTest
Flag: NONE
Change-Id: Ie4d438cc76b3bb2fbd39af999332fcd5e4a52427

Dropbox sends two DROPBOX_ENTY_ADDED broadcasts at the same time to
handle a CompatChange. The two broadcasts need to have different
DeliveryGroupMatchingKeys or one of them will get replaced by the other.

Fixes: 323450448
Test: atest CtsDropBoxManagerTestCases
Change-Id: Ibdf7439cf5eb86b0c267ebe7ddc6962912dcdfd6

Merge "[flexiglass] Ignore window insets in scene container root, and adjust Shade/QS layouts accordingly" into main

Merge "Merge "Add flag dump_verifications for attestation_verification" into main am: ad5e62b9 am: 16e3be89" into main

Merge "Add flag dump_verifications for attestation_verification" into main am: ad5e62b9 am: 16e3be89

Original change: https://android-review.googlesource.com/c/platform/frameworks/base/+/3081526



Change-Id: I89d06c0e6072f926bdd7d66b5a7f073544f4ff1d
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>

Original change: https://android-review.googlesource.com/c/platform/frameworks/base/+/3081526



Change-Id: Icdf18c6aa783a339fefb2210e5090c4c479971c0
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>

Adds FLAG_ONLY_ALERT_ONCE to the update we sent from system_server to
systemUI on direct reply, to transition the notification out of the
sending state and to the sent state.

This update should only be used to update the appearance of the
notification; it should not alert, because it's triggered by the user's
direct reply action, and the user already knows they direct replied
(they're the ones who did it). We already setPostSilently on the
notification record, but that just prevents beeping, buzzing, and
blinking; it doesn't prevent the notification from HUN-ing if, for
example, the enqueued update somehow gets re-triggered by the app (see
attached bug).

Modified unit tests to check we're adding this flag.

Bug: 335106917
Test: atest NotificationManagerServiceTest
Flag: ACONFIG android.app.lifetime_extension_refactor STAGING
Change-Id: I521d83eea27e7026e1050a27856ef86013fd659f

There is a small chance that sending a message might result in failure code 3 (other) for different reasons (e.g. driver is busy). We should disable SAM only if the AVR doesn't acknowledge the CEC message. This is similar to aosp/2619391.

Bug: 326821366
Change-Id: Iadaa65f91a2338a8cc2653fd9af6f17cb80ae727
Test: manual

Bug: 331795143
Change-Id: Icf36b1e0dcddf2c51e1349930d8ba7553918689d
Flag: EXEMPT bugfix
Test: Presubmits

This also add long clicks handling. Because the Expandable composable requires a shape and color, I had to split Modifier.tileModifier() slightly to avoid clipping the shape twice.

Flag: ACONFIG com.android.systemui.qs_ui_refactor DEVELOPMENT
Test: manually clicking on tiles
Bug: 331598956

Change-Id: I9acbe78d1a00afd48af7125fac95e65acb58a6d9

PageIndicator sometimes get in a state where the width isn't always updated when the number of pages change (possibly linked to if PageIndicator is visible at the time or not)
Adding this check forces a width recalculation if it doesn't match the current number of pages.

Fix: 322540792
Flag: none
Test: manually, opening QS
Change-Id: I12f584b2cd86db5a0c7e02fb88dcfaeb80893d42

We don't want to fall back to the last interactive brightness when the screen turns on because it might be brighter than the brightness applied shortly after, when a lux sample is available.

Bug: 336000148
Test: See that there's no bright flash when going from AOD to ON
Test: atest DisplayPowerControllerTest
Change-Id: I5fed20f6b4d2de3821f52ac06dc21c0dbe287aac

Bug: 335795282
Test: atest WMShellUnitTests:DragResizeWindowGeometryTests
Change-Id: I2eb0a9c6f01d9e0e7f640dec9f7af16cc46c1f4e

- Prior to calling onTransitionConsumed, we add another mixed keyguard
  transition handler to the list with the same transition, and
  onTransitionConsumed iterates the active transitions backwards and
  finds the newly added MixedTransition instead of the original

Bug: 329324086
Bug: 339038054
Test: Artificially fall into the keyguard path and verify it was finding
      the wrong mixed transition
Change-Id: I654b4b89448a9e15000c3e34103cf6f0359be02b