1. Remove first initiation of dataspace in the constructor of
   RequestedLayerState;
2. snapshot->stretchEffect has been set in the approach of
   prepareBasicGeometryCompositionState(), so remove the second call.

Change-Id: I784223ddce80756dc88ec78e8f0deb042e6389db
Signed-off-by: jiayongqiang <jiayongqiang@xiaomi.com>

'mDrawingState.crop.makeInvalid()' has been called twice in the
approach of Layer::Layer(const LayerCreationArgs& args), so remove
the second call.

Test: atest LayerSnapshotTest, LayerTest

Change-Id: I4cc40d9ec697ff20a1b71fa7aa50f09cef6c7afc
Signed-off-by: jiayongqiang <jiayongqiang@xiaomi.com>

'Vector<DisplayState> displays' has been not used in the approach of
SurfaceFlinger::initializeDisplays(), so just remove it.

Test: atest SurfaceFlinger_InitializeDisplaysTest

Change-Id: I6bb063ccef487ab23bc97e72e342353bdcaea748
Signed-off-by: jiayongqiang <jiayongqiang@xiaomi.com>

This is an annoying artifact generated sometimes by working in rust. It
can interfere with git operations and we shouldn't let it be committed.

Other packages have similar rules, so this should be OK.

Test: n/a
Change-Id: I5f166dadb81d246c6d2a587a68299999db91cc11

* changes:
  nativewindow: Misc. improvements for AHardwareBuffer Rust wrapper
  nativewindow: Add more benchmarks to evaluate FFI costs
  nativewindow: Add C++/Rust benchmarks
  Add missing safety comments.
  libnativewindow: Add rust library and set up bindgen

Changes include:

  - Rename AHardwareBuffer to HardwareBuffer
  - Expose AHardwareBuffer as a raw pointer type
  - Making HardwareBuffer Send
  - HardwareBuffer now derives Debug, PartialEq and Eq
  - Use NonNull instead of a *mut pointer
  - Adding an into_raw function to match from_raw
  - Adding a Clone impl that acquires a ref

Bug: 296449936, 296100790
Test: atest libnativewindow_rs-internal_test
Merged-In: Iaf916fabe49190f47abd1a9ed34afdb76fd20e40
Change-Id: I5da6375582e98e8783b31ad8424f1e21c33467e5

This CL adds two new benchmarks, one for getting an AHardwareBuffer's ID
and another for getting its description.

These should (and do) ultimately take the same amount of time.

Test: just adding new tests
Merged-In: I5e09a2736ab829ca465aaa4073b439a605d49b5a
Change-Id: I7ee6dd4e989c85fe18221f51a998ae1c25221800

Test: atest nativewindow_buffer_benchmarks_rs nativewindow_buffer_benchmarks_cc
Merged-In: Ia2898439da46e57cf65e3d64030d98f052a694c6
Change-Id: Ib5aaa97f7130f1964824a2f6c8ca85d72d61ce4f

These will soon be required by a lint.

Bug: 290018030
Test: m rust
Merged-In: Iaa33bab93c458d963d45ec68daf243057b9f1c15
Change-Id: Ia822853e44cc74d0c27622c8d30ca7c404058663

For now we only expose AHardwareBuffer, but the exposed bindings will
make it easy to also expose ANativeWindow, ADataspace, and others in the
future.

NOTE: This makes the new presubmit post-submit to appease the SLO
checker

Bug: 291954749
Test: Added new unit tests for both bindings and the rust library.
Merged-In: I8ef24fc9111bb3fb72a4cdd941742d42d72fa776
Change-Id: I1dc87c9012fc1c1b805716cf7252fba2da11aec2

Bug: 302723053
Test: mma
Change-Id: I126772474763a55f43de0cd4a3d2605f7f84da3b

* changes:
  Add test to the BpBinder::getBinderProxyCount
  Move the binder proxy accounting to libbinder

When screen recording is enabled, a virtual screen will be generated,
and the cancel event will be incorrectly sent to the virtual screen.
The coordinates of the cancel event are the wrong logical coordinates
after being transformed by the transform of the windows of the virtual
screen.

Bug: 301020008
Test: atest inputflinger_tests
Change-Id: Icf1301e5174ac61eeedda326545f2b68ef7b8af0
Merged-In: Icf1301e5174ac61eeedda326545f2b68ef7b8af0

When a cancel event is synthesized for any reason for the drag window
when drag and drop is active, we should cancel the ongoing drag and drop
operation. Otherwise, the system may end up in an unexpected state
resulting in unwanted behavior.

Bug: 291181957
Test: atest inputflinger_tests

Change-Id: Ic5e328f0e445e3271cb1797725c4c00589a6b5e1
Merged-In: Ic5e328f0e445e3271cb1797725c4c00589a6b5e1
Signed-off-by: Linnan Li <lilinnan@xiaomi.corp-partner.google.com>

Bug: https://github.com/android/ndk/issues/1933
Test: treehugger
Change-Id: Ifbb4d08722aa953f3e86d770e8d6c9017ab9d680

Allow it to set up inet servers on different IP addresses.

Test: $ adb shell servicedispatcher -i 192.168.1.246 \
            android.hardware.vibrator.IVibrator/default
Bug: 293615467
Change-Id: I6684c96409425cb1ba397f6a76eafbbff9ae425b

Bug: 298314844
Test: atest BinderLibTest#BinderProxyCount
Merged-In: I97bbb5b7560f36294718ce22015079bc4f96ba22
Change-Id: I97bbb5b7560f36294718ce22015079bc4f96ba22

So the android.os.Debug#getBinderProxyObjectCount will include
the binder proxies from both of the Java and native layers.

Bug: 298314844
Test: dumpsys meminfo
Merged-In: I822fbbfd47bf1929f587a5147f124d99e7ba1059
Change-Id: I444392f59785c3b40b1b7b052ca6c10ec9a36bef

Test: Build with CMake. See aosp/2629369.
Bug: 285204695
Change-Id: I5a0396d1fba458e11f42266e806634aaa1e9206d

Bug: 300602767
Change-Id: Ie67bf2ef39faa8b679fde7831bd8f45d264a3350
Test: builds

Bug: 302724232
Test: build binder with GCC
Change-Id: I5bb7cff8999f6fb6f2411eb8697a79c979d12103

Bug: 298295554
Merged-In: I114a4cf83978f152959a6ec244d322603ff5257b
Change-Id: Ia35ae6e11052fcc3b2bdaabd6dc57d0d223de269

Bug: 300602767
Change-Id: Iccc941a3d44781703cf1dc82513e6354b1839ae7
Test: builds

Original change: https://android-review.googlesource.com/c/platform/frameworks/native/+/2773889



Change-Id: I2117a4a93b3c93e3a4189fea84709f47da0c9540
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>

Test: Build with CMake. See aosp/2629369.
Bug: 285204695
Change-Id: Ife316aa5c08ea0f365f52062cff52625ca877c46

Bug: 300602767
Change-Id: I114a4cf83978f152959a6ec244d322603ff5257b
Test: builds