Bug: 343531585
Test: atest PermissionServicePerfTests

Change-Id: I8dd632b1d49574557bad9f69ade6cd29b66b5c0f

This reverts commit 796ea6e5.

Reason for revert: Possible cause for these spikes https://screenshot.googleplex.com/4BiHCKud3vo9dbY

Change-Id: I7f4ad73967845c7fdd0c95be5fe7f1aebd0c91a3

This reverts commit 1e48634c.

Reason for revert: Possible cause for these spikes https://screenshot.googleplex.com/4BiHCKud3vo9dbY

Change-Id: Id93aaa08df4eb0bdb44569eb4f148bee744447d6

A timeout was added to BenchmarkRunner.waitForBroadcastIdle method in
ag/27491356 I91579d8af35b4a7df76d6bc1bcfad328c107d7c2
and this caused the UserLifecycleTests results to be more spiky.
This CL reverts that change, and also increases the timeout in
waitForCpuIdle.

Bug: 342399089
Bug: 340812874
Test: atest UserLifecycleTests
Flag: None
Change-Id: I2686625530fc64552d4380ad63fd5235d3364f5e

Bug: 342399089
Bug: 340812874
Test: atest UserLifecycleTests
Flag: None
Change-Id: I91579d8af35b4a7df76d6bc1bcfad328c107d7c2

Bug: 316941721
Test: mmma .
Change-Id: Ifd928a3534c4c843fa0270370a646201b206c49b

1. Passing Parcelable in Bundle through AIDL can cause some unexpected
   exception (b/335444562)
2. Introduce WindowRelayoutResult so that if we need to add more optional
   fields in the future, we don't need to change the function signature.

Bug: 335601427
Bug: 335444562
Test: atest WmTests:WindowManagerServiceTests
Flag: com.android.window.flags.window_session_relayout_info
Change-Id: I291cdca2e1f730e5a1246409edc21d60f78e07ca

Bug: 334079478
Test: atest

Change-Id: Ieedd88fe5560f4add2c9a22cfe557dea3e3b4a16

The duplicate <test> tag causes the tests to run more than once. 

Bug: 333971355
Test:atest PackageManagerPerfTests
(cherry picked from https://googleplex-android-review.googlesource.com/q/commit:94be3681407ff9e3d9c305d221bf23fad5edd84f)
Merged-In: I1e3fa112edcac79cceaa64fee170a2633ee131b2
Change-Id: I1e3fa112edcac79cceaa64fee170a2633ee131b2

This CL restores the window background color back to white. The IME
color is  light as well. So when the IME is showing and hiding, the
gesture navigation bar won't need to change its color to adapt the
background, which might consume additional CPU resource.

Fix: 334024581
Flag: NA
Test: ImePerfTest#testShowImeWarm
Change-Id: I83bf308933d63cb138353d682a66cb8f6e1eb07a

The duplicate <test> tag causes the tests to run more than once. 

Bug: 333971355
Change-Id: I1e3fa112edcac79cceaa64fee170a2633ee131b2
Test:atest PackageManagerPerfTests

Since Android V, apps targeting SDK 35+ will be forced to go
edge-to-edge. This CL adds proper padding to the content view of
PerfTestActivity, so the EditText within it won't be obscured by system
bars. Otherwise, there would be unexpected scrolling while showing or
hiding IME.

This CL also removes the action bar from the activity, so that the
content won't be obscured by it.

Fix: 331025926
Bug: 309578419
Flag: com.android.window.flags.enforce_edge_to_edge
Test: ImePerfTest
Change-Id: I212ef4862b4249e288267fb8fb4448b9100107c1

The "force" parameter in stopUser is confusing and misleading. It sounds
like "force stopping" a user is a powerful, generic action. In reality,
this parameter is extremely narrow in its meaning: when false, it means
that a profile won't be stopped if it is a profile of the current or
system user. Since, hitherto, profiles always belong to the system user,
this effectively means that
stopUser(force==false) just means "stop the user but only if it isn't a
profile user", which is basically never desired (since then the caller
shouldn't be calling the method in the first place).
For full users (e.g. when called as part of a user switch), it is
completely irrelevant.

This confusion has propagated, since callers seem to think it means
something more. Moreover, due to HSUM, going forward, profiles may
belong to other users, in which case the effective action of this
parameter will change, so clarity is important.

Here, we try to clarify and wean usage off the "force" parameter. It is almost
always set to true, except in the very rare instances where false is actually
desired (when stopping a user due to there being too many running users,
and when stopping a background user due to a change in the corresponding
user restriction).

Unfortunately, we cannot simply get rid of the force-parameter version
of AMS.stopUser, but we can at least make it clear that it's not normal
to use.

Test: atest UserControllerTest

Test: CarSystemUI-tests:com.android.systemui.car.qc.QCLogoutButtonControllerTest
Test: CtsCarBuiltinApiTestCases
Test: CarServiceUnitTest

Test: atest SystemUIGoogleTests:com.google.android.systemui.communal.dock.callbacks.UserSwitcherTest

Bug: 324647580
Change-Id: I28f30d5ac68330850691141be93bcad05b4c3210

Fix kotlin nullable errors that were exposed by setting the retention
of android.annotation.NonNull and android.annotation.Nullable to
class retention.

Bug: 294110802
Test: builds
(cherry picked from https://googleplex-android-review.googlesource.com/q/commit:2522501d5b4b4eab8d20986aa1c4b6ed36640211)
Merged-In: Icfec5d0e20b534d7af9aeb1b539a711db76d5e5c
Merged-In: I79b5d963079237b70012b4db717620fb757bfa0d

Change-Id: Icfec5d0e20b534d7af9aeb1b539a711db76d5e5c

Bug: 325929237
Test: atest RubidiumPerfTests:JSScriptEnginePerfTests

Change-Id: I94c2cff412aed996d733a9317a50b35e54a35239

This CL is created as a best effort to migrate test targets
to the new android ownership model. If you find incorrect or unnecessary
attribution in this CL, please create a separate CL to fix that.

For more details please refer to the link below,
<add g3 doc link>

Bug: 304529413
Test: N/A
Change-Id: I6b6cf0abdd1fdb8fefa4184b3698f6951c192ac0

Allow Stylus handwriting to initiate handwriting without waiting for
InputConnection to be established.
When an editor view receives focus, it sends startInput binder call immidiately.
Focused editor (servedView) is then ready for any
startStylusHandwriting() calls.

This should help shave few millis from initiation latency.

Note: this intentionally doesn't optimize delegation since that is
migrating to newer APIs.

Bug: 308827131
Bug: 324670412

Test: atest HandwritingInitiatorTest StylusHandwritingTest

Change-Id: I0174e15fbbedf88dcd68ebe98333b6b07871985a

The TextViewSetTextMeasurePerfTest measures text with 0 pixel width
and height which is uncommon situation, so setting proper
width/height for measuring realistic numbers.

Bug: 323606812
Test: atest TextViewSetTextMeasurePerfTest
Change-Id: If864d9b3d11861da01da5881a746959baf3f160c

Mockito dominates the time to create MotionPredictor, and
it can be avoided.

The median time reduces from ~39614ns to ~683ns. (-98%).

Bug: 321913450
Test:  atest CorePerfTests:android.input.MotionPredictorBenchmark
Change-Id: Idea21b5627a25af8d5941bb8c7a1ff567eb2504d

Other than the startup latency, the latency of sending the events is
critical to content capture performance.

Test: ContentCapturePerfTests
Bug: 322344738
Change-Id: I1ecc90213d1aa9a48f1fc128b3b38ebf68343034

Unclear if this will help, but it won't hurt at least

Bug: 317869473
Test: speculative
Change-Id: Ib523d676ad0c69119e782ed7fb6116107e1a3b91

Add PermissionServicePerfTests. Added the test testInstallPackages() to
test the performacne of onPackageInstalled() calls when installing 30
packages. Use ManualBenchmark to fetch the trace and add it to the
benchmark duration.

Test: PermissionServicePerfTests
Change-Id: I5f6859ff57af2757ad2a7b9b0a9d89da30247973

The two multi-threaded read/write performance tests violate the
recommended Android database best practice: the writer loop creates a
new SQLiteDatabase on the test database file.  This leads to two
errors: the writer loop creates its database with an incompatible
journal mode in different from the actual database, and the test exits
without properly closing the database.

This change simply has the writer loop use the existing database for
its test.  The benchmark values for the multi-threaded read/write
tests change dramatically for the TRUNCATE journal mode and slightly
for the WAL journal mode; these metric changes are WAI.

Test: atest
 * SQLiteDatabasePerfTest

Bug: 318742652
Change-Id: Icb2534c6fbaf8daa3eb38e8b1b7f71331d571e9d

When porting SystemConfig, the APIs of Callback return value are incorrect.

BYPASS_INCLUSIVE_LANGUAGE_REASON=Legacy API

Bug: 309596860
Test: atest PackageManagerPerfTests:android.os.PackageParsingPerfTest
Change-Id: Ia02a02b7e2d20fa18442bc5bf400123b8a5996d4

Bug: 316972621
Test: atest android.graphics.perftests.CanvasPerfTest
Change-Id: I73ccfab4e9f325b1f864eb1d8f41c71a7c6ed7a5

Bug: 316941721
Test: mmma .
Change-Id: I19c6b2091da7e857db10ed1ddf9333c31b40e49e

Bug: 315116034
Test: HealthConnectReadWritePerfTest
Change-Id: Id44c76429ac4a587eae8f32837f62baaf9b58934

Bug: 315137410
Change-Id: I0a0d57717bec7608877a54c1c12977251c754cc4

1. Move implementations and related utils to internal
2. Make calling SystemConfig methods from ParsingPackageUtils.Callback
   to avoid calling from the client side.
3. Move isMatch and isEnabled from ComponentParseUtils to PackageInfoUtils
4. Move string from SELinuxUtil to SeinfoUtil
5. Move some methods from AndroidPackageUtils to AndroidPackageLegacyUtils
6. Copy some methods from PackageInfoUtils to AppInfoUtils
7. Use PackageParserException instead of PackageManagerException for
   validatePackageDexMetadata method

Bug: 309596860
Test: build pass and boot to home
Test: atest PackageManagerServiceServerTests
Test: atest PackageManagerComponentOverrideTests
Test: atest PermissionServiceMockingTests
Test: atest PackageManagerServiceUnitTests
Test: atest PackageManagerPerfTests
Change-Id: I3de48d0d8adf714447823408673e07ed379f27ab

Legacy transition -> shell transition:
 AppTransitionReady -> onTransactionReady
Global transaction is removed:
 open/closeSurfaceTransaction -> applyTransaction

Bug: 206872204
Test: atest InternalWindowOperationPerfTest
Change-Id: I9f619486395ce17e2d0d9db7958869f5c32e146a

The test has lost its functionality since [1].
Because the relayout invocation will be skipped if the
sequence number is old.

[1]: Ifb365789fa08103773fd3180e486ba1d92042fc5

Bug: 206872204
Test: atest RelayoutPerfTest
Change-Id: Ic8eb77a13a27a6c92279794cd7ed1981e1b03560

Add current perf team members shayba@ and jdduke@, and remove
former members jpakaravoor@, balejs@, and cfijalkovich@.

Test: N/A
Change-Id: If786fa38bc5f86d9e64b1672bdcf5e8cde178aec

Skip using the gainmap shader when drawing to a bitmap
with a known SDR colorspace as it'll be a no-op in
rendering results since W=1.0, but it'll be very expensive
in runtime due to CPU evaluation of the SKSL shader.

Before:
createScaledBitmapWithGainmap_median (ns): 179,455,037

After:
createScaledBitmapWithGainmap_median (ns): 15,643,742

Bug: 311085927
Test: benchmark in CL
Change-Id: I3f2506a2bd86bbe29c6e0837d31e4a231a7c182a

Removing the window will tear down the input connection for an embedded
window. To make it more generic and allow non windows to have
InputChannel, it's simpler to pass in an IBinder instead of creating a
fake IWindow.

Test: WindowStateTests
Bug: 278757236
Change-Id: Ie632429e3c890077d6c4788941c686977d1ba5c9

Skip using the gainmap shader when drawing to a bitmap
with a known SDR colorspace as it'll be a no-op in
rendering results since W=1.0, but it'll be very expensive
in runtime due to CPU evaluation of the SKSL shader.

Before:
createScaledBitmapWithGainmap_median (ns): 179,455,037

After:
createScaledBitmapWithGainmap_median (ns): 15,643,742

Bug: 311085927
Test: benchmark in CL
Change-Id: I3f2506a2bd86bbe29c6e0837d31e4a231a7c182a

Bug: 311064136
Test: m RUN_ERROR_PRONE=true javac-check
Change-Id: Ia6e4162a4244793f772dc880b73be0bc30f26d8b

Test: atest android.surfaceflinger.SurfaceFlingerPerfTest
Bug: 298240242
Change-Id: I87117178a296df9741def8e1c8e711c412d31761

Remove overheard by disabling features during the test including:
- perfetto tracing
- region sampling by hiding navbar
- transaction tracing

Reduce variations between each frame by
- forcing max frame rate (avoids any frame misses)
- consuming transform hint (avoids gpu comp)
- starting simpleperf after test setup to move test activity
launch outside of measurement window

Test: atest android.surfaceflinger.SurfaceFlingerPerfTest
Bug: 298240242
Change-Id: Ida7422c37b9afa323147949c9776c042ca97cd08

Created two simple benchmark tests to measure performance of
ViewConfiguration#get.

Bug: 299587011
Test: atest ViewConfigurationPerfTest

Change-Id: If502a4acb6bc6080af134675e610a76df59c538a

Fix kotlin nullable errors that were exposed by setting the retention
of android.annotation.NonNull and android.annotation.Nullable to
class retention.

Bug: 294110802
Test: builds
(cherry picked from https://googleplex-android-review.googlesource.com/q/commit:2522501d5b4b4eab8d20986aa1c4b6ed36640211)
Merged-In: Icfec5d0e20b534d7af9aeb1b539a711db76d5e5c
Change-Id: Icfec5d0e20b534d7af9aeb1b539a711db76d5e5c