Bug: 206676167
Test: m

Change-Id: I914366f6cef19df5f528bb0a43b51d17b50ca1ae

Revert submission 2970571-turn-on-network-restrictions

Reason for revert: b/326370901

Reverted changes: /q/submissionid:2970571-turn-on-network-restrictions

Change-Id: I7f996f370791c83d781e7321344279823603c439

Allow the package install confirmation dialog to appear properly after
the user turns on unknown sources. Also fix an incorrect session ID
of -1 being shown in related log warnings.

"Finish PIA if UnknownSourcesListener would like to start a PIA again"
is a change that caused this feature to fail for third-party app stores.
This change resolves that issue.

Issue: calyxos#2081
Test: atest CtsPackageInstallTestCases
Test: Manual: Install a third-party app store e.g. F-Droid. Try to
install an app. Tap Settings on the prompt to configure unknown sources.
Tap to turn on unknown sources. The install confirmation dialog appears.

BUG: 326612529
Change-Id: Idd6667cbd79945d849b74a62122162ae72d99210

ScriptC won't be supported on any new architecture (for example riscv64), so
throw an exception when it is used on an unsupported architecture.

Note that using `Build.SUPPORTED_ABIS` includes ABIs supported due to
dynamic binary translation, which do not support ScriptC. So we can't
use `Build.SUPPORTED_ABIS` to determine if ScriptC is supported or
not.

Instead we attempt loading `libRS.so`.
`libRS.so` is not present on riscv64 images, so this allows us to know
if we are on a riscv64 system.
A problem occurs on x86 images with binary translation for riscv64
(aka berberis): when running a riscv64 binary, the binary translation
system will fallback to the x86 version of the libraries via what is
called a "native bridge". This means that `libRS.so` would successfully
load on those system, which we do not want because even though it would load it wouldn't run properly.
This is why a separate CL in the binary translation codebase was landed that explicitly refuses to load `libRS.so`:
https://android-review.git.corp.google.com/c/platform/frameworks/libs/binary_translation/+/2971952

Bug: 206676167
Test: atest CtsRsBlasTestCases:android.cts.rsblas.IntrinsicBLAS#test_L3_SGEMM_API -- --abi x86_64

Change-Id: I2ed2e2ae531046d1d00e9a2d2f864375b0bb3570

It first needs the test to be in postsubmit to run SLO chacks. Once that
passes we can add it to presubmit in a later CL.

Test: config change
Bug: 291137901
Change-Id: Ie5f9d404a84c4a16e791cac5fc0a6b867561a67f

This shouldn't happen, but if it does we should just ignore it rather
than crashing system server.

Modify the tests to match.

Bug: 323665257
Bug: 326190135
Test: atest PackageDynamicCodeLoadingTests
(cherry picked from https://googleplex-android-review.googlesource.com/q/commit:1aea0e915748d661e2bdf251e34bdc32d22ce0e9)
Merged-In: I2ae7f506b15133f45f00c24e909f49d56fc23e66
Change-Id: I2ae7f506b15133f45f00c24e909f49d56fc23e66

Before collect new created task fragment in createTaskFragment(),
recomputeConfiguration() will set a crop in PendingTransaction,which
will cause starting window be cropped before staring window reparent
to root task in SyncTransaction.So addToSyncSet() before
recomputeConfiguration().

Test: pressure test cold-lauch Activity Embeded apps and no flash black.
Bug: b/324148725

Change-Id: I7ded9adb1c52d61b700edfdc83e7929cff67e529

Change-Id: I264b96ae2cc0e9999705b102422c7f1cf967a83d

Bug: 324993608
Test: TH
Change-Id: Ic3b7f9a7d8cf29d9f4cd51a39dd51488fbb62fe3

This reverts commit 951e4bd8 and commit
b6a3ff2d

Test: atest CtsHostsideNetworkTests
Test: atest resolv_integration_test
Test: atest FrameworksServicesTests:NetworkPolicyManagerServiceTest

Bug: 304347838
Change-Id: I0c8e0c713f0a7aae7f0ab99e42d97d5bf770bc47

Added high impact rollback as a last resort after low impact rollbacks
are tried. Also removed reference to automatic rollback deny list, since
we will be using impact level instead of the deny list going forward.

Bug: 291137901
Test: Unit tests
Change-Id: I4d75dd513b3f43c179baf93ece832f9007173ff4

There are two Thread flag files in frameworks/base and
packages/modules/Connectivity/thread. Both are currently using the same
"com.android.net.thread.flags" package name for their flags. While all
flags needs to be built in a aconfig_declarations target, two aconfig_declarations
can't have the same package name:

```
FAILED: out/soong/build.aosp_arm.ninja
cd "$(dirname "out/host/linux-x86/bin/soong_build")" && BUILDER="$PWD/$(basename "out/host/linux-x86/bin/soong_build")" && cd / && env -i  "$BUILDER"     --top "$TOP"     --soong_out "out/soong"     --out "out"     --soong_variables out/soong/soong.aosp_arm.variables -o out/soong/b
uild.aosp_arm.ninja --globListDir aosp_arm --globFile out/soong/globs-aosp_arm.ninja -l out/.module_paths/Android.bp.list --available_env out/soong/soong.environment.available --used_env out/soong/soong.environment.used.aosp_arm.build Android.bp
2 aconfig_declarations found for package com.android.net.thread.flags
```

To resolve this issue, this commit renames the platform flags to package
name "android.net.thread.platform.flags".

This commit is required by ag/26286387

Bug: 325886480
Merged-In: Icacf88f7c0aa499b4b29f4072ae58840f0b1761f
Change-Id: Icacf88f7c0aa499b4b29f4072ae58840f0b1761f

Bug: N/A
Test: manual
Change-Id: I64f066a696ec41d55716fd03fde395c50ec15067
Merged-In: I1b35078cbc5f581b6b13daa8afcb6eee1aa2a15f

Bug: none
Test: none
Flag: NONE
(cherry picked from https://googleplex-android-review.googlesource.com/q/commit:cc0dd126b74a56774d7887dcb4ccd655ad0d0e45)
Change-Id: I2c508cae9c2d499a0fc53b97698d4a0165f00c14

Some tests relying on the flag were left enabled by mistake.

Test: atest FrameworksServicesTests:com.android.server.net

Bug: 304347838
Bug: 325685265
Change-Id: I17a9fcaaa44f2c8b43c19e3445a329c8613ee3dd
Merged-In: I17a9fcaaa44f2c8b43c19e3445a329c8613ee3dd