The v2_apk_in_apex_upgrades APEX shim will be useful for testing fixes
for apk-in-apex caching bug.
Test: m com.android.apex.cts.shim.v2_apk_in_apex_upgrades
Bug: 225435110
Change-Id: I5d2270b8341c0405bf256c417acff4279c6b8754
As these interfaces get frozen, java services
implementing the interfaces are required to implement getInterfaceHash
and getInterfaceVersion [1]
[1]
https://source.android.com/devices/architecture/aidl/stable-aidl#new-meta-interface-methods
Bug: 190577319
Test: m
Change-Id: I00ced32a26fd1af7bc8cc23749d92ae90d5a5a0f
Merged-In: I00ced32a26fd1af7bc8cc23749d92ae90d5a5a0f
(cherry picked from commit 1b46333208)
UsbPortManager is made to support both AIDL & HIDL interface until
HIDL is deprecated eventually. UsbPortHal abstract class is defined
which the UsbPortManager grabs an instance of from UsbPortHalInstance
method to talk to the Hal implementation process. UsbPortHalInstance
queries the underlying hal implementation and instantiates UsbPortHal
with either UsbPortHidl or UsbPortAidl based on the IUsb interface
implemented by HAL implementation.
go/iusb-aidl-migrate.
Bug: 200993386
Bug: 199357330
Bug: 211677613
Bug: 213312081
Bug: 199358576
Test: Manually tested through UI.
CTS-Coverage-Bug: 215019881
Change-Id: I246fd6e5bee8138d3ad3b6994665a1d4788c8789
Merged-In: I246fd6e5bee8138d3ad3b6994665a1d4788c8789
(cherry picked from commit bbfa92e694)
InputDevices can be reconfigured when pointer capture changes, but the
app will be notified via InputManager when the updated device
information is available. This races with the onPointerCaptureChange
callback, so advise the use of InputDeviceListener in the pointer
capture documentation for developers to get around this.
Bug: 226425883
Test: None
Change-Id: I7aeb23722dc40b8fd37cb9667a4f23a336e8b5f1
Merged-In: I7aeb23722dc40b8fd37cb9667a4f23a336e8b5f1
During process startup, system creates a process group and places the
new process in it. If process group creation fails for some reason, the
new child process will stay in its parent's group. This poses danger
when the child is being frozen because the whole group is affected and
its parent is being frozen as well. Considering that the parent of an
app is Zygote, freezing it leads to the device becoming unusable.
Fix this by treating group creation failure as a fatal error which would
prevent the app from starting. When system_server fails to create a
process group, it will throw an exception that's not caught and will
self-destruct.
Bug: 227395690
Test: fake group creation failure and confirm app failure to start
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: Ib34b4a100c65a9a2705aaef8e7e32322fbedc531
Dark theme "TURN ON UNTIL SUNRISE"/MODE_NIGHT_AUTO is not kept after
reboot. The reason is that TwilightState will takes time to update
so lastState will be null in updateConfigurationLocked().
This lead to a reset of mComputedNightMode and mOverrideNightModeOn.
This is a special bug for "TURN ON UNTIL SUNRISE"/MODE_NIGHT_AUTO,
the other ones work fine and should not be touched, if we do not
treat this as a special case we break the other ones.
Test: Select "TURN ON UNTIL SUNRISE" in Dark Theme and reboot.
Bug: 229228293
Change-Id: Ie32a6ddbbf03fa43ee6dd5e35b3c35ec03e58e0e
* changes:
Modify javadoc for some extra values of VpnManager event
Add toString() for VpnProfileState
Move mSessionKey into Ikev2VpnRunner and make it to be a final var
The underlying network, NetworkCapabilities and LinkProperties
are not necessary to send to VPN app since the purpose of
CATEGORY_EVENT_DEACTIVATED_BY_USER and
CATEGORY_EVENT_ALWAYS_ON_STATE_CHANGED are to notify VPN
app that VPN is deactivated by the user or VPN always-on
status is changed.
Bug: 191413541
Test: Javadoc only changes.
Change-Id: I7fd4b6dad1a4363336f03c217e635fc45bcd0ab6
mSessionKey is initialized in the constructor of Ikev2VpnRunner,
move it into Ikev2VpnRunner and make it to be a final var can
make sure that mSessionKey is initialized before someone access
it.
Test: atest FrameworksNetTests:VpnTest
Change-Id: If7eb7e0ec8ab42b17a5636aeced65385e7849b6a
Trying to manually delete the result of this function as a local
ref is problematic.
Bug: 227719836
Test: N/A
Change-Id: Iea4509e08f935f268ee502825dab47caef8508a5
IsExternalStoragePath() in gsid only allows vfat
Test: m; Install DSU on a device using exFAT for external sdcard
Change-Id: I6073b268ead79d684422e4ef1fa3873713dd602e
This change ensures that carrier configs are saved as part of the
snapshot, allowing all entities in the VCN to query telephony
subscription information from the same source.
Test: atest FrameworksVcnTests
Change-Id: I31447d84fe98d6bd2cc78ead41746198728c400b
This change adds some integers to the AIDL interface in order to convey
status back to the caller of generateKey(). This will inform the caller
as to whether or not the errors that may occur during provisioning are
permanent, and if not, what to do with the transient error.
Bug: 227306369
Test: RemoteProvisionerUnitTests
Change-Id: I9202358a102b0fb0a104525632a005acb7355840
In current design, NPMS listens onThresholdReached() callback to
call updateNetworkRulesNL(), however, this callback mwill not be
called if data usage is not updated. Thus, it will cause to no
iptables rule for new metered interfaces.
Have NPMS listen onLinkProperitesChanged() callback and call
updateNetworkRulesNL() if interfaces has changed to add iptables
rules for new metered interfaces.
Bug: 225971066
Test: HostsideRestrictBackgroundNetworkTests
HostsideRestrictBackgroundNetworkTests# \
testDataAndBatterySaverModes_meteredNetwork \
--iterations 100
Change-Id: I49f999efc3fbe0a6a325ec138a3f060c7285e6ad
This CL is to convert CF presubmit tests to run in Test Mapping suite.
It has been verified that the test coverage is the same as the original
ATP config.
Bug: 145165552
Bug: 145574049
Test: TreeHugger
atest ManagedProvisioningTests
Merged-In: I8234cc5cadae2ed2162cb09ec8695c7262fcdb2e
Change-Id: I268448e682b7c45741a62fa3bc1157986f3827d1
