Expand the "DynSystem" abbreviation to "DynamicSystem".
For simpler log grepping.
Bug: 165471299
Test: No
Change-Id: Ibec692ae45539dacd55e13ffaa903b58f3bf0417
DSU service would try to install a DSU scratch partition if host system
is debuggable.
If the scratch partition failed to install, then skip installing
scratch, as remount support should be optional.
Otherwise if the guest system is non-debuggable, then the scratch
partition would be ignored by guest, which is harmless.
Otherwise the guest system is debuggable, which means the guest system
would be able to do adb remount.
Relax partition allocation status polling rate from 10ms -> 100ms,
because there's no point polling faster than the screen framerate.
Bug: 165925766
Test: TH
Test: Install a DSU system on a debuggable host, reboot into the guest
system, and guest system can remount.
Change-Id: I9a8255483cc963ebcf7a2909e68ac69371cb369f
- buffer is in bytes, not bits (1MB vs 1Mb)
- this exception is actually a heuristic currently, pending b/28321379
Fixes: 149869862
Test: N/A
Change-Id: Id409a1ce7348666ec9ad6437acf67382be0a0a1f
With latest kernel changes, total GPU memory usage is reported and can
be obtained via a BPF program. Create JNI interface to query it and
report inside dumpsys meminfo output.
Bug: 171261987
Test: dumpsys meminfo
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Merged-In: I949a13836d5b5bc87fc43f60871b4fbf2add6480
Change-Id: I949a13836d5b5bc87fc43f60871b4fbf2add6480
In a previous change (https://r.android.com/1394342), we did a mass update of whitelist->allowlist
and blacklist->denylist in network policy related code. Updating
some usages of those (like allowlisted to allowed) to make them
sound natural.
Test: atest services/tests/servicestests/src/com/android/server/net/NetworkPolicyManagerServiceTest.java
Test: atest services/tests/servicestests/src/com/android/server/NetworkManagementInternalTest.java
Test: atest hostsidetests/net/src/com/android/cts/net/HostsideRestrictBackgroundNetworkTests.java
Change-Id: I6d34b0bd3cdb64d5872874fd9378bfc962a24f8d
Forward ownership of libbinder and libhwbinder related things to
respective libbinder and libhwbinder OWNERS sets, to support future
developments.
Bug: 149869862
Test: N/A
Change-Id: If14bfdd89719c7cb1e39594aecdb089d4d5c96e3
In order to support ConnectivityService mainline, the dependency
with NMS should be removed. The idle timer control API should be
replaced with calling INetd interface instead of using NMS hidden
API.
The state reporting will be triggered from updating idle timer
and netd unsolicited interface activity changed events.
BatteryStatsService should be able to get the netd event by
registering the listener directly. This commit exposes two system
APIs to report the radio power state for wifi and mobile for
supporting update idle timer from CS directly to INetd.
Bug: 170598012
Test: make update-api ; m
Change-Id: I286d6aa237ecb2653e1464e26a3581b8eeeb9e63
This API isn't aware of slot switch, and shouldn't be used by
phonesky. So we prefer to deprecate it in case new callers use
it accidentally. Also don't allow callers with REBOOT permission
to call it.
Bug: 170664917
Test: build
Change-Id: I540b87fad461cb2c900f3e9d177bb5147abc1a22
It helps remove it from the @CorePlatformApi
Bug: 154796679
Test: ArrayUtilsTest
Merged-In: I0c8f194a74a16b2cc46f9eea4571d5fb674fbc28
Change-Id: I0c8f194a74a16b2cc46f9eea4571d5fb674fbc28
Detailed design in http://go/multi-client-ror. We will reuse most of
the old APIs, and there are a few notable changes
1. Infer the package name from the existing context
2. Remove the unused update token
3. Add the REBOOT permission, so phonesky can hold it.
4. Add a new API for rebootAndApply, which takes a slotSwitch parameter
Bug: 170664917
Test: build
Change-Id: Icb012138bd1789ce03564aa255e5bd4e84374130
Detailed design in http://go/multi-client-ror. This cl adds internal
states in RecoverySystemService to keep track of multiple RoR
callers. The general principles works as:
1. Each client should call prepare before reboot
2. One client cannot clear the preparation of another client
3. If multiple clients have prepared the update, the subsequent
reboot will be first come, first served.
Bug: 170664917
Test: atest FrameworksServicesTests:RecoverySystemServiceTest
Change-Id: Ic8c7438f8a0300870ceed8857c46425fc0d006b9
Currently when debug layer apk is not installed, GraphicsEnvironment
logs the message and proceeds, which results in null pointer access
crash when later trying to access primaryAbi. Avoid null access by an
early return statement.
Bug: N/A
Test: specify not installed debug layer apk and no crash
Change-Id: I4f3b86b0a8cfff4fc608b3a59ebf0872db0465ff
Merged-In: I4f3b86b0a8cfff4fc608b3a59ebf0872db0465ff
(cherry picked from commit c5c17e66ba)
Currently when debug layer apk is not installed, GraphicsEnvironment
logs the message and proceeds, which results in null pointer access
crash when later trying to access primaryAbi. Avoid null access by an
early return statement.
Bug: b/175699044
Test: specify not installed debug layer apk and no crash
Change-Id: I4f3b86b0a8cfff4fc608b3a59ebf0872db0465ff
Merged-In: I4f3b86b0a8cfff4fc608b3a59ebf0872db0465ff
(cherry picked from commit c5c17e66ba)
The RecoverySystem mostly handles OTA update and interactions with
the recovery mode. Add the ota team as the owner.
Test: build
Change-Id: I81bae46d919ea70f04729d34b62d4a569541d914
The puller for the atom frequently fails. We believe it is due to the
size of the pulled atoms. No metric can be correctly build from this
atom.
To reduce the size, we apply transformations that we would likely
perform when analyzing this data. We combine times for all shared app
gids together and combine times for the same apps. We also exclude
isolated uids which are frequently recycled and removed from the
underlying data source.
If these optimization do not help, we will need to introduce different
atoms.
Bug: 157535126
Test: cmd stats pull-source 10010
Change-Id: I0d9a18c78f456e1b0f9d7ad322f15354ead9f02b
Merged-In: I0d9a18c78f456e1b0f9d7ad322f15354ead9f02b
(cherry picked from commit e69ade3545)
