Some test cases in TrustManagerServiceTest are failing on internal main
because http://ag/23669695 changed how PackageMonitor works.  Sending an
ACTION_PACKAGE_CHANGED intent no longer causes
PackageMonitor#onPackageChanged to be called.  This CL fixes the test
failures by making TrustManagerServiceTest call the onPackageChanged
method of TrustManagerService's PackageMonitor directly.

Bug: 29385425
Test: atest TrustManagerServiceTest # on both AOSP and internal main
Change-Id: I0b4708e4223452430ff10e422a71b223656a3360

This android.bp file was moved, and the comment needs to be updated
to reflect that.

Test: Presubmits
Change-Id: I225690505ccbf0f2234e65cbc668ade9bfbd5e17

* changes:
  vintf: Disable kernel compat check at boot time.
  vintf: delete deprecated VintfObject.verify.

Before this change, if the kernel has a set of CONFIGs
that is not compatible with the system image, a dialog
is displayed for user / userdebug builds at boot time.

This check has been doing more harm than good because:

- This check is already enforced at build time and during
  VTS tests (See vts_treble_vintf_framework_test).
- The dialog blocks UI automation for tests. For these UI
  automation tests, they need to respond to the dialog.
- GKI has been enforced ecosystem-wide except for a few
  low-end devices of other verticals. For these non-GKI
  devices, the check enforced by VTS should guard this.
  Hence, the check does not give us any signal.
- During development, a kernel that corresponds to the latest
  release (android15 as of now) might not have valid kernel
  config requirements in userspace. Kernel development schedule
  is usually ahead of the userspace development schedule.
  It does not always carry the string "-mainline-", because
  it is not a mainline kernel. To unblock test automation on
  these latest, bleeding-edge kernels, this kernel check should
  go away.
- This is a small steps towards dropping the dependency on libvintf
  on libandroid_runtime. libvintf links to libselinux, which is
  huge. libandroid_runtime loads this, and the memory stays there
  forever. Ideally, we should disable the whole VINTF check at
  boot time, but let's do this one step at a time.

Bug: 272479887
Bug: 270169217
Test: TH
Change-Id: If24cdca9fb535b8f443c0d21f9a46c7ea25c1f9f

It is not used anywhere.

Test: TH
Bug: 270169217
Change-Id: Ifcc8412ed3629d2447908513faf1d6f5ed3f483c

Merge "Reset top activity's mWaitForEnteringPinnedMode when aborting an incomplete pip-entry" into main

The primary purpose of validate_framework_keymaps is to perform checks
on all *.kl, *.kcm, and *.idc files using the "validatekeymaps -q"
command. This action can be replaced by using a genrule.

Bug: 318785074
Test: m validate_framework_keymaps
Change-Id: I657466d7ee2cec29900beeb0610cb0d30f34f92e

Non-A/B has been deprecated for long. During non-A/B updates,
compatibility.zip is checked with VintfObject.verify(), which
is another deprecated function that always returns compatible
if a list of package VINTF XMLs are supplied. (see below.) Hence, the private
RecoverySystem.verifyPackageCompatibility was just useless code
that unconditionally returns true except for invalid ZIP file.
Remove it. Replace the public RecoverySystem.verifyPackageCompatibility
and make it return true unconditonally.

VintfObject.verify() with OTA XMLs are deprecated per
http://b/139300422 [VINTF] Delete OTA vintf checking code
in http://r.android.com/1194233 ("Delete VINTF compatibility checks
during OTA."), in 2019. We had decided that compatibility checks
during OTA should be removed, and moved to OTA generation time instead.
Using an old libvintf on the device to check against new libvintf
metadata required forward compatibility of libvintf, which cannot be
achieved. Instead, the device should verify the signature of the OTA to
verify its source.

Test: TH
Bug: 270169217
Bug: 139300422
Change-Id: I775d29e4cd1d165233e07cfb820d1fe343fa4757

Merge "Uptesting is moving folks to /external/robolectric from /external/robolectric-shadows" into main

This test is in neither presubmit nor postsubmit.  It should be in
presubmit, but add it to postsubmit as a first step.

Bug: 296464083
Test: N/A
Change-Id: I6453d92fea76efab6554e4315cf9f38d84e8e072

Copies the system version of the `api-versions.xml` file to a file
`api-versions_system.xml` in the top level dist directory.

Bug: 190665366
Test: m sdk dist
      # Check to make sure that the file exists.
Change-Id: Ie139f81a819eb5f389363ffdefa7bc42d36804e0

We maintain /external/robolectric (keep it in sync with github/google3) and are getting ready to delete /external/robolectric-shadows

Bug: 314757990
Test: atest same failing tests before and after in services/robotests.
Flag: NA

Change-Id: Ie3d2e902ffcba6b9d159b78592f7d6ce31288a2f

Add/update owners for SystemConfig related files
(See comment in: https://r.android.com/2775452)

c/j/c/a/s/SystemConfig.java has also moved so update the OWNERS
file for that too.

Change-Id: Ia773c7851d6aad4223ce5845d0993d527a5fd140

do not unmute when disconnect hdmi_out.

Bug: 297980843

Test: play some video, mute it, then unplug hdmitx, plug hdmitx

Change-Id: I1005550cbf7f8c2e586397ab7127a7fc087dde1d
Signed-off-by: hongchao.yin <hongchao.yin@amlogic.com>

Prevent the disabling of specified apps, avoiding cases
where disabling the app may result in an unusable system.

Bug: 200043113
Change-Id: I5b0ed8c34a40428e8d18072582e46854737c000a

The userSerial parameter is being removed from the vold methods that
don't use it.  Update the callers accordingly.

Note: later CLs will remove the serialNumber parameter from methods in
StorageManagerService too.  This CL simply updates the calls to vold.

Bug: 316035110
Test: presubmit
Flag: N/A, mechanical refactoring
Change-Id: Ide5397693d594e0b1fad81c031e3c9190fd0bdde

Test: m framework-all
Test: Compiled and applied OTA using settings app using stable interface
m Settings && adb install -r $ANDROID_PRODUCT_OUT/system_ext/priv-app/Settings/Settings.apk
Bug: 317914533

Change-Id: I393df5218c60303dbea60386c680b75b553ddac2

Change-Id: Icc8a7b899ccb54196ece9cccbfe3658547f0603c

When aborting an incomplete pip-entry, reset top activity mWaitForEnteringPinnedMode to prevent skipping it's window frame computation later.

Bug: 318553345
Test: Manual test

Change-Id: I9378f54957534c2fab7d877d03bd40d2573959da
Signed-off-by: buwenjing <buwenjing@xiaomi.com>