This reverts commit cf6e79f8

Updates:
    Prepare the shared pointers for the whole operation at once
    instead of re-locking them on each iteration.
    Still a regression of about 5% for changing theme's
    AssetManager object, vs the original 40%

    + change the log message to a warning as it doesn't break
      the app

Original comment:

    Use reference counted pointers for ApkAssets

    The primary reason for memory corruption is freed ApkAssets
    Java expected them to only be freed in the finalizers, but
    there are explicit close() calls now, destroying objects that
    are still in use in some AssetManager2 objects

    This CL makes sure those AssetManagers don't assume ApkAssets
    always exist, but instead tries to lock them in memory for any
    access

    It also adds logging in case of deleting an assets object with
    any weak pointers still existing. Those will get into the
    bugreports attached to related bugs to help with investigation.

    Benchmarks don't regress, and the device appears to be working.
    Given that the crashes used to be pretty rare, let's wait for
    any new reports or lack of those.

    + add a missing .clang-format file to the jni directory
    + enabled C++23 in the project that uses AssetManager headers

    Bug: 197260547
    Bug: 276922628
    Test: unit tests + boot + benchmarks
    Old change id: I495fd9e012fe370a1f725dbb0265b4ee1be8d805
(cherry picked from https://googleplex-android-review.googlesource.com/q/commit:b3455190124b41e2deb1774d9c05b396b73b41a2)
Merged-In: Id668fbcf07db17b09691a344c04e98df83006f97
Change-Id: Id668fbcf07db17b09691a344c04e98df83006f97

Ibe3679d810602456205e157fe2cb8cc510062215 restricted the argument to
PhysicalDisplayId, which prevents capture of virtual displays.

Fixes: 280785757
Test: screencap -d <id>
Change-Id: I41a8d1adaa51c6804c03f1b3084ac04ff6b4f1fa

Revert submission 22327091-br-cts-hsum

Reason for revert: DroidMonitor-triggered revert due to breakage https://android-build.googleplex.com/builds/quarterdeck?branch=git_udc-dev&target=aosp_bramble-userdebug&lkgb=10098913&lkbb=10101009&fkbb=10098962, bug 281753581.

Bug: 281753581

Reverted changes: /q/submissionid:22327091-br-cts-hsum

Change-Id: I4de169440f587d5a458300a5f360ba974239e530

Bug: 266703231

Test: Run CTS/GTS tests for backup
Change-Id: I7a8d09f9f1be83ff4d9f545439f93b2c5319f007

When desc.txt configuration interval is very long, it leads to too long sleep time, which affects the execution frequency of checkExit and leads to untimely exit of boot animation.

For example:
1080 2400 5
p 0 5 part0

Change-Id: I9806a934189642a31ceb72cbf35c3ff07fe6b7d8
Signed-off-by: niuwenchao <niuwenchao@xiaomi.com>

Bug: 272551810
Test: adb shell bmgr backupnow com.android.wallpaperbackup --monitor; adb shell bmgr restore 1 com.android.wallpaperbackup (http://go/paste/5502576853254144)
Change-Id: Idc3e288f2548448d9a39f69096e33def783e3853

This is a small binary that will load a GPU vendor provided shared
library called libgpudataproducer.so, look for the "start" symbol in
that library, and call it. This can be used to start the GPU hardware
counter Perfetto producer that will allow tracing of GPU profiling data
via Perfetto.

See go/gpu-profiling-arch for more details on the GPU profiling
architecture design.

Test: manual
Bug: b/273907878
Change-Id: Iacccb562d0ef154c105de034c8801642cab72532

Bug: 268463592
Test: manually
Change-Id: Ifa0064b8a79b6a302fdcf8f02a7feab582079dc8
Signed-off-by: Jack Wu <wjack@google.com>

According to [1] :
section 3.4 :
...
Creation of a client API context based on an EGLConfig will
fail unless the EGLConfig’s EGL_RENDERABLE_TYPE attribute
include the bit corresponding to that API and version.
...

section 3.7.1.6:
...
An EGL_BAD_MATCH error is generated if config does not support the
requested client API. This includes requesting creation of an OpenGL ES
1.x, 2.0, or 3.0 context when the EGL_RENDERABLE_TYPE attribute of config
does not contain EGL_OPENGL_ES_BIT, EGL_OPENGL_ES2_BIT, or EGL_-
OPENGL_ES3_BIT respectively.
...

Bootanimation uses EGL_CONTEXT_CLIENT_VERSION  2 for context creation,
so appropriate EGL_RENDERABLE_TYPE should be requested during eglChooseConfig.

1. https://registry.khronos.org/EGL/specs/eglspec.1.5.pdf



Change-Id: I5ebc18d8ace84197d069049decadb402576546b5
Signed-off-by: Andrii Chepurnyi <andrii_chepurnyi@epam.com>

* Previously, the height returned by Display#getSize was used to
  determine visible objects. Based on the API level and the presence of system decorations (e.g. notches), this can cause objects near the
  bottom of the screen to be considered out of bounds (i.e. only objects
  within y=0 and y=partial_height are considered even if the omitted
  decorations are at the top of the screen). cherry pick from aosp/2384832

Bug: 236669488
Test: manual
Merged-In: Iba21245c947cec0d0054f6d86e3585ba73233223
Change-Id: I50f143528d1f5a5b677eb2628efbcd658f4cc1b0
(cherry picked from commit afebeb7b)

This remerges ag/20838898 that was reverted due to breaking a
robolectric test in BackupFrameworksServicesRoboTests.

Test: 1. atest BackupFrameworksServicesRoboTests
      2. atest CtsBackupHostTestCases
Bug: 259953764
Change-Id: I11c7dbe9959f5b5d1acf53e7830b7256357768e2

This reverts commit 748c64d9.

Reason for revert: DroidMonitor: Potential culprit for b/265432205 - verifying through ABTD before revert submission. This is part of the standard investigation process, and does not mean your CL will be reverted

Bug: b/265432205
Change-Id: I11e21ad5048635716adb5724b7d78d5513fc60e9

Test: atest implemented and passing, to be sent in separate commit
Bug: 259953764
Change-Id: I33be1b5907a6c2a0dd08c31ba3401d28fac02b35

Implement the AIDL and HIDL coexist in framework

Note: This is a cherry-pick of ag/17905026
Bug: 218791946
Test: USB function switch success and AIDL service is running.
      atest VtsHalUsbGadgetV2_0HostTest
Signed-off-by: Ricky Niu <rickyniu@google.com>

Merged-In: I6d9bc861a4aceb753bf6f816daccd039e423994d
Change-Id: I6d9bc861a4aceb753bf6f816daccd039e423994d
(cherry picked from commit 74272e24)

Change the default display rotation value and boot animation orientation to the value specified in the sysprop ro.bootanim.set_orientation_<display_id>. Four values are supported: ORIENTATION_0, ORIENTATION_90, ORIENTATION_180 and ORIENTATION_270. If the value isn't specified or is ORIENTATION_0, nothing will be changed.

This is needed to support having default rotation behavior and boot animation in orientation different from the natural device orientation. For example, on tablets that may want to keep natural orientation
portrait for applications compatibility but have landscape orientation as a default choice from the UX perspective.

Bug: 260627934
Test: manual test with all 4 values and atest WmTests still pass
Change-Id: I322919999ef787ff2c149293d7e1c7e669796c9a

BUG: 259265292
Test: adb shell bootanimation
Change-Id: I6bd6ed78232cef56b3f99006f4d7780eeef8b881
Signed-off-by: yidong zhang <yidong.zhang@amlogic.com>

No need to copy the strings as we keep them in separate
containers anyway, may just reference them

Bug: 237583012
Test: build + boot + UTs

Change-Id: I2853cd3c992ca482ed6988e0c52a4037b158d999

Bug: 237583012
Test: build + boot + UTs

Change-Id: I849831f4466d3b9c7ec842b75256e7fcba77a0c0

Implement the AIDL and HIDL coexist in framework

Bug: 218791946
Test: USB function switch success and AIDL service is running.
      atest VtsHalUsbGadgetV2_0HostTest
Signed-off-by: Ricky Niu <rickyniu@google.com>
Change-Id: I6d9bc861a4aceb753bf6f816daccd039e423994d

The overlayable mechanism has defined which roles can overlay the
specified resources and which resources are overlayable.

In order to reduce the argument between platform and applications,
self-targeting applications should enforce the overlayable mechanism.
It makes the overlay range be defined in the overlayable list. It
also defines the limitation of what kind of package can overlay the
resources of the self-targeting application.

To fill the policy flag, OverlayManagerImpl.java gets the following
information and passes them as bool type to JNI OverlayManagerImpl.
* SYSTEM_PARTITION
* VENDOR_PARTITION
* PRODUCT_PARTITION
* ODM_PARTITION
* OEM_PARTITION
* SIGNATURE

Not support
* ACTOR_SIGNATURE
* CONFIG_SIGNATURE

JNI OverlayManagerImpl fills the policy flags according to the
information from the ApplicationInfo. 'enforce_overlayable'is true to
enforce checking overlayable before generating an idmap.

Fixes: 238061808
Bug: 205919743

Test: atest \
              OverlayHostTests \
              OverlayDeviceTests \
              SelfTargetingOverlayDeviceTests \
              OverlayRemountedTest \
              FrameworksServicesTests:com.android.server.om \
              CtsContentTestCases:android.content.om.cts \
              idmap2_tests

Change-Id: I749d1a90a00f0a91e7796bc0f9f668f9ecfdb8b0

Add OverlayManagerImpl to let applications create frro and idmap files
for overlaying itself's android resources.
* OverlayManagerImpl.java
  * check the FabricatedOverlayInternal.
* com_android_internal_content_om_OverlayManagerImpl.cpp
  * convert and read Java objects to native data.
  * call the APIs in libidmap2 to create frro and idmap files.

Add OverlayManagerImpl.cpp into AndroidRuntime by modifying Android.bp
* Add register_com_android_internal_content_om_OverlayManagerImpl
  register JNI methods
* add com_android_internal_content_om_OverlayManagerImpl.cpp
    * It's a adapter to delegate the tasks to SelfTargeting.cpp
    * dynamic link with libidmap2 by using dlopen and dlsym to call
      functions in libidmap2.

Add SelfTargeting.cpp into libidmap2
* createFrroFile
* createIdmapFile
* getFabricatedOverlayInfo

for interoperability between libandroid_runtime and libidmap2.
* move OverlayManifestInfo from libidmap2 to libandroidfw
* add FabricatedOverlayEntryParameters into libandroidfw

Bug: 205919743

Test: build
Test: atest \
                  OverlayHostTests \
                  OverlayDeviceTests \
                  SelfTargetingOverlayDeviceTests \
                  OverlayRemountedTest \
                  FrameworksServicesTests:com.android.server.om \
                  CtsContentTestCases:android.content.om.cts \
                  idmap2_tests

Change-Id: I5425f3229e9a3858e57427ef84e6abaf32e89b6e

Test: unit tests
Bug: 251283316
Change-Id: Iead53711c54596c3787eeba6dcf9ced129f94426

Previously, this was resulting in the output color roughtly as

  output_n = color_n * mask_n^2 * (mask_0 + mask_1 + mask_2 + mask_3)

where n is a color channel between 0 and 3.

This caused extra aliasing around edges, since the alpha fades too
quickly along an edge.

Bug: 239632466
Bug: 258246539
Test: verify pixel values for boot animation match mask on watch + phone
Test: modify uFade value, verify pixel colors match
Change-Id: I86742a181576e684aa698a190a8766f4223de799

* Previously, the height returned by Display#getSize was used to
  determine visible objects. Based on the API level and the presence of
  system decorations (e.g. notches), this can cause objects near the
  bottom of the screen to be considered out of bounds (i.e. only objects
  within y=0 and y=partial_height are considered even if the omitted
  decorations are at the top of the screen).

Bug: 193270758
Bug: 188156320
Bug: 27537269
Test: manual
Change-Id: Iba21245c947cec0d0054f6d86e3585ba73233223

The fix is taken from AOSP patch aosp/2154236.

This will make sure the UI Automator always has the entire display to
interact with. Otherwise, in multi-windowing mode or on devices with
task bar, it will only be able to interact with the current focused
window, the default launch bounds, or the display excluding task bar,
depends on the actual scenario.

Test: PlatformSenarioTests
Bug: 238981445
Bug: 238985243
Bug: 238581838
Bug: 233945217
Change-Id: I9223ef5552ddb081e75de1242670862f2813e56d
(cherry picked from commit d1e350c7)
Merged-In: I9223ef5552ddb081e75de1242670862f2813e56d

* Copied from the androidx.test.uiautomator OWNERS file
  (http://cs/aosp-androidx/test/uiautomator/OWNERS).
* This project is in maintenance mode, and this allows the uiautomator
  team to make occasional fixes as needed.

Test: N/A
Change-Id: If89ced39a77cb55a10cd0608fdc5745462a81ab7

When no display id is explicitly specified, the first display is
captured, and a warning message is printed out. This is to make
the utility to be backward compatible with previous versions so
tools depending on it continue to work.

Bug: 241285477
Test: manual, run screencap command to verify
Change-Id: I713e1e5c625fd80ecb5587e0b2313f58baaffd40

This change reinstates most dumpsys sections in incident, because at
least one of these sections are being used by ABT (ODPM info for Battery
Historian). The dumpsys sections were originally removed because they
were being collected elsewhere in dumpstate.

The most expensive dumpsys section (meminfo, ~6 seconds) remains
omitted, so most of the improvements of aosp/2156679 should be retained.

Test: - atest incidentd_test
      - Capture bug report on an internal build and verify that ABT's
	Battery Historian tab contains ODPM rails.
Bug: 252586237
Change-Id: I3e1e5ebea6d6829cb3b70ddb70721878053eb7ca

Add a lock over all frro iterator methods to make sure the
iterator isn't changed or removed from under the iteration
call while it's running.

+ Also introduce an ID ensuring the caller is the same for
  all the acquire/next/release methods

+ a bit of moves where they were supposed to be from the start

Bug: 245727875
Test: manual + presubmits
Merged-In: Ie585b6d113ebddee491c9ee433f070bed71a57cc

(cherry picked from commit de47d0b9)

Change-Id: Ib4870d680501426dcc0a62eebd003451120179a5

Sync with changes made in SurfaceFlinger and SurfaceComposerClient.
Use the system property, persist.boot.animation.displays, to determine
the displays to show the boot animation. If the system property is not
present, the first display from the display list is used.

Bug: 241285477
Test: manual, run bootanim command to verify
Change-Id: If29ecb12a68fc075daef2b8f674995d0f11cbc09

Add a lock over all frro iterator methods to make sure the
iterator isn't changed or removed from under the iteration
call while it's running.

+ Also introduce an ID ensuring the caller is the same for
  all the acquire/next/release methods

+ a bit of moves where they were supposed to be from the start

Bug: 245727875
Test: manual + presubmits
Change-Id: Ie585b6d113ebddee491c9ee433f070bed71a57cc

More moves and fewer allocations

Bug: 237583012
Test: unit tests
Change-Id: I5cf43c8af0743c0e4d96808f1e55ceb4f02d7021

Sync with changes made in SurfaceFlinger and SurfaceComposerClient.
In the case of multi-display devices, a display id must be specified
with "-d" option for screencap, since SurfaceFlinger is not guaranteed
to provide the correct default display.

Note this change breaks the default behavior of screencap, therefore
tools that rely on screencap will have to be modified to select a display to deal with multi-display devices.

Bug: 241285477
Test: manual, run screencap command to verify
Change-Id: Ibe3679d810602456205e157fe2cb8cc510062215

Without this change there are errors when running with the new version
of clang

Bug: 247585041
Test: verified that frros used for colors still works, automated tests

Change-Id: Iebbef8aaca866c6f34c381792c1e66bd6b2121f0

Bug: 247585041
Test: presubmit
Test: lunch aosp_cf_arm64_only_phone_hwasan-userdebug;
Test: make tidy-frameworks-base-cmds-idmap2_subset
Change-Id: I2cd78fddb330c3fda3bbb88c2470dfb8b51ea4cf

This also includes the change to use the configs to decide which (if
any) frros to use at runtime

Bug: 243066074
Test: Manual, updated and created automated tests
Change-Id: I3f1d23e2958ad170799880b9f5eb5bd8ceb1fa67

- Added support for SensorHeadTrackerTest. This does not affect other tests used by Hid Command tool.
- Added two new functions setGetReportResponse() and
  sendSetReportReply()

Bug: 228319672
Test: atest android.hardware.cts.SensorHeadTrackerTest
Change-Id: I83cfe1f71df57584dac7288ab092164a5c5dea3c

Bazel doesn't allow a module and file with the same name.

These are some extra modules I missed in aosp/2201115

Bug: 198619163
Test: m am, verified it was still installed to out/target/product/emulator64_x86_64/system/bin/am and not am.sh
Change-Id: I763a938918854541ec5b0bcf740b1b6cf0fce771

Bug: http://b/175635923
Test: treehugger
Change-Id: I8caf940f7b6e2fdd921516efea4aed14e0c2afb7

Once the client requests idmap2d to release a directory iterator,
idmap2d should follow the client request to free and release
frro_iter_.

It should not only check the value of frro_iter_ but also make
frro_iter_ to be nullopt in releaseFabricatedOverlayIterator.

Fixes: 244389702
Test: adb reboot; adb wait-for-device; adb logcat -s idmap2d
Change-Id: Ida1e7e7ae89f12407d26e66ab318a91e4f861df4