incfs 2 also creates a .incomplete folder, so we need to set correct
permissions on that folder too
Bug: 169084168
Test: With updated kernel, incfs_test works and adb install
--incremental works
Change-Id: Id7b9d017973777e37628664530dfed18c55c269f
This changes allow Incremental Service to directly report health status
to package manager service.
A health listener is created during package installation session to
monitor incremental storage health. After commit, a new listener is
created and will overwrite the old one.
The new listener will listen to incremental storage health and report
the status to package manager service, which will then send the status
to IncrementalStates, where the startability state and unstartable
reason might change, based on the health status code.
During reboot, for each incremental package, if it is not fully loaded,
the package manager service will register a health status listener to
continue monitor the health status of this package.
Test: unit test
Test: manual
BUG: 170435166
Change-Id: I220f230c523cfaf2c96019f9478554665e6af486
Based on go/incremental-states-design with basic
setter/getters.
Defines IncrementalStates class which handles state transitions.
New (internal) Intent actions: PACKAGE_FULLY_LOADED, PACKAGE_STARTABLE,
PACKAGE_UNSTARTABLE.
BUG: 168043976
Test: unit tests
Change-Id: I7b0ec2dd9f028ee620a9307a1e71ddf12ea5a9af
Incremental Serivce periodically polls loading progress and sends to
Package Manager Service. Package Manager provides APIs for other
interested parties to listen to the loading progress.
BUG: 165841827
Test: unit test
Change-Id: I44b9e17c2240b9efe53bc09fc728b6671f1f7dfe
This is to unblock Launcher's work on progress ring. Currently it uses
incfs getFilledBlocks(). Will switch to the new incfs progress reporting
API once it is ready.
Test: unit test
Test: adb shell dumpsys incremental
BUG: 165799231
Change-Id: Icd68124806454f888826294da36f109bca9771ac
AIDL's @nullable is represented as std::optional in C++ backend.
The previous aidl::nullable was a temporary solution to avoid
merge-conflicts between rvc-dev and master branches.
Bug: 151275535
Test: m
Change-Id: I8fe5bfc1763087e7ff364b4408f862436492d1db
The lib dir should always be at the same dir level as the APK file.
Test: manual with OTA
Test: atest CtsExtractNativeLibsHostTestCases
BUG: 160826412
Change-Id: I87d715c4739c78bb56b85d375838db08bafa1594
Only if the application is profileable.
Bug: 158238023
Fixes: 158238023
Test: atest PackageManagerShellCommandTest PackageManagerShellCommandIncrementalTest IncrementalServiceTest PackageParserTest
Change-Id: I8575830ec3f29850297fdbfbaa157072d6350a28
Merged-In: I8575830ec3f29850297fdbfbaa157072d6350a28
Part 2: continuous health checking, blocked and unhealthy states.
Bug: 153874006
Test: atest PackageManagerShellCommandTest PackageManagerShellCommandIncrementalTest IncrementalServiceTest
Change-Id: Ifdbff7dfa24b3bd9b96c534a40bb1226082749ca
Makes sure the behavior is consistent with legacy installs:
When the flag is on, the native libs will be extracted to subdirs under
lib/.
When the flag is off, the lib/ subdirs will be created but the native
libs are not extracted.
When the flag is off, check if the native libs are uncompressed and well
aligned.
Test: atest android.extractnativelibs.cts.CtsExtractNativeLibsHostTest
BUG: 157173358
Change-Id: Idb57fd7ca1115f787faf5cde3056c32ff3f60890
This saves system's resources.
Healthcheck will automatically start data loader on pending reads.
Bug: 153874006
Test: test PackageManagerShellCommandTest PackageManagerShellCommandIncrementalTest IncrementalServiceTest
Change-Id: I160153f1d3e7e653db5de06b59d6fbbc3667fdf2
Once pending read detected, try to start the dataloader.
Bug: 153874006
Test: test PackageManagerShellCommandTest PackageManagerShellCommandIncrementalTest IncrementalServiceTest
Change-Id: Ia8169ccbb0f710317715e6fddb9bc6a718543766
This is a temporary failure, does not fail the session, but requires
caller to re-commit. E.g. there are connectivity issues which can be
fixed later.
Bug: b/153874006
Test: atest PackageManagerShellCommandTest PackageManagerShellCommandIncrementalTest IncrementalServiceTest
Change-Id: I02791a2963130dbecb510c4a7cafcf04f6245761
Dispatch session destroy to be executed later.
We can't do anything blocking during one-way transaction processing.
This blocks all one-way transactions to this Binder, including other
status updates.
Bug: b/153874006
Test: atest PackageManagerShellCommandTest PackageManagerShellCommandIncrementalTest IncrementalServiceTest
Test: atest IncrementalInstallerTest#testIncrementalInstallFails_onPrepareImageReturnsFalse
Change-Id: Ie4385d603fca8b5205775e849e92846cff672f56
- crash on disappearing dataloaders,
- more robust binder callbacks processing,
- heavy unbind lifting moved to a separate thread to unblock Binder.
Bug: b/153874006
Test: atest PackageManagerShellCommandTest PackageManagerShellCommandIncrementalTest IncrementalServiceTest
Change-Id: Ie06c3c4dbecbdd552dd868e2092bf0ee48f8547a
This simplifies:
- resource management - no extra copies of controls,
- state management - all states in one place, no more hidden (bound but
not created) state.
Bug: b/153874006
Test: atest PackageManagerShellCommandTest PackageManagerShellCommandIncrementalTest IncrementalServiceTest
Change-Id: I3d16a099c7f42fcf14637c5a8e96bd6f99e073d1
This gets rid of annoying warnings in logcat about not being
able to create a directory and then directory already existing
Bug: 153704006
Test: atest IncrementalServiceTest
Test: adb install megacity.apk
Change-Id: Ib718960287f93cb383c06c9b9e3d0abf1ec42916
This is a big cleanup in IncrementalService that makes it behave
nicely on runtime restart, and more:
- fixed a bunch of threading issues in createStorage/bind
- made public functions correctly accept any path in any
bind mount and translate it to the proper root mount
- got rid of "using namespace" in headers, cleaned includes
- removed all unused functions
- set CLOEXEC bit on all duped FDs
Bug: 151241369
Test: atest PackageManagerShellCommandTest \
PackageManagerShellCommandIncrementalTest \
IncrementalServiceTest
Change-Id: Ided4415aabfbfca3925b5e71c91896055886ac4a
This is a pure refactoring.
Bug: b/153874006
Test: atest PackageManagerShellCommandTest PackageManagerShellCommandIncrementalTest IncrementalServiceTest
Change-Id: Ieda2be08d7359fa69b2d328c85b3606de6d02b3d
StorageID for an installation changes as we go from a staging
directory to the final one. That's why the only correct way
of waiting for the extraction to complete is by the whole
MountID, even if later it would cause the call to wait on all
instances of the app.
+ measure the waiting timing
Bug: 153513507
Test: adb install megacity.apk
Change-Id: I83558f155867ae5503719ecb63d591fc969c3995
Worker thread has to initialize JNI separately to be able
to call into managed binders implemented in the same
system_server process, e.g. DataLoaderManager
Bug: 153513507
Test: adb install megacity.nov4.apk; adb install megacity.v4.apk
Change-Id: I668e8664361cd2fb3353ec50efd689c7d613658f
IncrementalService can create the library files beforehand, but
delay filling in their data. As it takes quite a while in
general (over a second in cases when the phone is busy), it's
better to run the unzipping and filling in a separate thread
and only make sure it finishes before the whole installation
process is complete.
This speeds up the megacity.apk installation by ~250-300ms,
1000-1100ms -> 750-800ms
Bug: 153513507
Test: adb install megacity.apk
Change-Id: Ia44f7e45b9e0abaebdfb6fe5352f9dcf29ab4ece
Now it's unified with callback FS connector - we are passing the
callback pointer directly to dataloader. This restricts access only
to methods we want and only by someone we want.
Bug: b/153468113
Test: atest PackageManagerShellCommandTest PackageManagerShellCommandIncrementalTest IncrementalServiceTest
Change-Id: Ib557ebbe7c6c5ce92140eb20534a3626b3ac96d3
