A long time ago, we had a concept of an "empty" process -- this was
a process that didn't have any interesting components in it, which
would be placed below everything else in the LRU list.
Empty processes didn't work out well, because you could get into
bad situations where you have filled your LRU list with things that
have hidden activities, pushing empty processes to the bottom and
being immediately killed as soon as they go into the list. So this
was removed.
This change brings the concept back, but in a slightly different
form, to address a more specific problem: for people who are switching
between N different applications, we would like to try to keep those
activities available in RAM in a consistent manner. Currently the
previous activities would be killed often quickly and suprisingly,
even on devices with lots of RAM. This is for two reasons:
(1) As you sit in one application, other things going on in the
background will go to the top of the LRU list, pushing down the
previous apps you have visited, even though you aren't aware at all
of these other things executing.
(2) There is a hard limit on the number of background processes
(currently 16) after which they are killed regardless of the amount
of available RAM. This is desireable because if there is lots of
RAM we can end up with tons and tons of processes sitting around,
not really serving any purpose, but using up resources.
To improve the situation, we have again a concept of "empty" processes
but now it means one with no activities. Processes that aren't empty
but in the background list are called hidden. We maintain these as
two parallel lists, each getting half of the process limit: so with
a 16 process limit, you can have at most 8 empty and 8 hidden processes.
This allows us to consistently keep up to 8 recent applications around
for fast app switching; we will also keep around 8 other processes to
make it more efficient for background work to execute again if it needs
to.
Change-Id: Iee06e45efc20787da6a1e50020e5421c28204bd7
Use notification manager as an indicator of when to stop the screensaver
due to immediately-needed user interaction.
Any notification with a fullScreenIntent will now exit the screensaver.
Bug: 6976796
Change-Id: Id91cd4c32f3fad018b56f4dfe24b37110bab32d8
If the incoming request is to notify of storage unmounted, don't mess
with apps that are in internal ASECs.
Bug: 6948035
Change-Id: I63ffb895c4d994ee03a5a9bd6bb23f69c88e2a87
Moved a bunch of methods from PackageManager to UserManager.
Fix launching of activities from recents to correct user.
Guest creation APIs
Change-Id: I0733405e6eb2829675665e225c759d6baa2b708f
Themes: Fused Location, Geofencing, LocationRequest.
API changes
o Fused location is always returned when asking for location by Criteria.
o Fused location is never returned as a LocationProvider object, nor returned
as a provider String. This wouldn't make sense because the current API
design assumes that LocationProvider's have fixed properties (accuracy, power
etc).
o The fused location engine will tune itself based on the criteria passed
by applications.
o Deprecate LocationProvider. Apps should use fused location (via Criteria
class), instead of enumerating through LocationProvider objects. It is
also over-engineered: designed for a world with a plethora of location
providers that never materialized.
o The Criteria class is also over-engineered, with many methods that aren't
currently used, but for now we won't deprecate them since they may have
value in the future. It is now used to tune the fused location engine.
o Deprecate getBestProvider() and getProvider().
o Add getLastKnownLocation(Criteria), so we can return last known
fused locations.
o Apps with only ACCESS_COARSE_LOCATION _can_ now use the GPS, but the location
they receive will be fudged to a 1km radius. They can also use NETWORK
and fused locatoins, which are fudged in the same way if necessary.
o Totally deprecate Criteria, in favor of LocationRequest.
Criteria was designed to map QOS to a location provider. What we
really need is to map QOS to _locations_.
The death knell was the conflicting ACCURACY_ constants on
Criteria, with values 1, 2, 3, 1, 2. Yes not a typo.
o Totally deprecate LocationProvider.
o Deprecate test/mock provider support. They require a named provider,
which is a concept we are moving away from. We do not yet have a
replacement, but I think its ok to deprecate since you also
need to have 'allow mock locations' checked in developer settings.
They will continue to work.
o Deprecate event codes associated with provider status. The fused
provider is _always_ available.
o Introduce Geofence data object to provide an easier path fowards
for polygons etc.
Implementation changes
o Fused implementation: incoming (GPS and NLP) location fixes are given
a weight, that exponentially decays with respect to age and accuracy.
The half-life of age is ~60 seconds, and the half-life of accuracy is
~20 meters. The fixes are weighted and combined to output a fused
location.
o Move Fused Location impl into
frameworks/base/packages/FusedLocation
o Refactor Fused Location behind the IProvider AIDL interface. This allow us
to distribute newer versions of Fused Location in a new APK, at run-time.
o Introduce ServiceWatcher.java, to refactor code used for run-time upgrades of
Fused Location, and the NLP.
o Fused Location is by default run in the system server (but can be moved to
any process or pacakge, even at run-time).
o Plumb the Criteria requirements through to the Fused Location provider via
ILocation.sendExtraCommand(). I re-used this interface to avoid modifying the
ILocation interface, which would have broken run-time upgradability of the
NLP.
o Switch the geofence manager to using fused location.
o Clean up 'adb shell dumpsys location' output.
o Introduce config_locationProviderPackageNames and
config_overlay_locationProviderPackageNames to configure the default
and overlay package names for Geocoder, NLP and FLP.
o Lots of misc cleanup.
o Improve location fudging. Apply random vector then quantize.
o Hide internal POJO's from clients of com.android.location.provider.jar
(NLP and FLP). Introduce wrappers ProviderRequestUnbundled and
ProviderPropertiesUnbundled.
o Introduce ProviderProperties to collapse all the provider accuracy/
bearing/altitude/power plumbing (that is deprecated anyway).
o DELETE lots of code: DummyLocationProvider,
o Rename the (internal) LocationProvider to LocationProviderBase.
o Plumb pid, uid and packageName throughout
LocationManagerService#Receiver to support future features.
TODO: The FLP and Geofencer have a lot of room to be more intelligent
TODO: Documentation
TODO: test test test
Change-Id: Iacefd2f176ed40ce1e23b090a164792aa8819c55
It turns out OFF_BECAUSE_OF_TIMEOUT is not the best indicator
of actual screen timeout.
For example, it is the reason passed down when acquiring a wake lock.
This was causing us to launch Dreams in the wrong situations, and
deadlocking on calls to WindowManager.
This fix simply adds an additional check ensuring the intention is to
turn the screen off.
Change-Id: If8adff446b5b1fcb19424b45878b75bfd0552b90
Added a new method to Context: startActivityAsUser() requiring the
INTERACT_ACROSS_USERS_FULL permission.
Show the correct Recents list, based on current user.
Added a getRecentTasksForUser() in ActivityManager. Hidden and requires
the INTERACT_ACROSS_USERS_FULL permission.
Change-Id: If5b56465efdd3ead36601a3b51ed4af157bbf35c
Change ConnectivityService to use a factory when creating
NetworkStateTrackers, which gives us a good place to inject mocks
for testing. Add initial tests to verify that network routes are
added and removed as networks changed.
Change-Id: I11cbc61a84c2ed4afa2670036295b1494eab26e1
At systemReady time, the window manager needs to obtain values from the lock
settings service (e.g. whether the lock screen is disabled). During the
initial boot of a wiped device, the window manager was getting the wrong
values because LockSettingsService.systemReady/migrateOldData had not yet been
called.
Change-Id: I60825d0e4fad53a5b4349dabf9e5e299b863d0d0
(cherry pick of 2458abaf6676d5015af733c1010ecd67a085e9e2 from jb-aah-dev)
- You can now use android:singleUser with receivers and providers.
- New API to send ordered broadcasts as a user.
- New Process.myUserHandle() API.
For now I am trying out "user handle" as the name for the numbers
representing users.
Change-Id: I754c713ab172494bb4251bc7a37a17324a2e235e
1.If a window is shown but never moved the window window
is never notified for its current location. Therefore,
accessibility nodes do not contain correct bounds in
screen coordinates.
bug:6926295
Change-Id: I7df18b095d33ecafffced75aba9e4f4693b0c393
Network type can be obtained through EXTRA_NETWORK_INFO, but offer
it as first-class extra since the returned NetworkInfo is deprecated.
Bug: 6936247
Change-Id: Ief59577afd2bd930f0f4c5650b413feef86bfbc3
Preloaded drawables now have a density associated with them, so we
can load the correct drawable if we are using a different density.
Window manager now formally keeps track of the density for each
screen, allowing it to be overridden like you can already do with
size, and relies on this density to drive itself internally and
the configurations it reports.
There are a new set of Bitmap constructors where you provide a
DisplayMetrics so they can be constructed with the correct density.
(This will be for when you can have different windows in the same
app running at different densities.)
ActivityThread now watches for density changes, and pushes them
to the DENSITY_DEVICE and Bitmap global density values for that
process.
A new am command allows you to change the density.
LocationProviders often rely on non-default networks, so pass the
network that actually changed, instead of the default.
Bug: 6929692
Change-Id: I31d9eec792e07259282aa1bb57ec66c01962df64
Enable feature in config. Expose Dream in public api for unbundled apps.
Unhide package. Add isDreaming() method to service.
Re-arrange the Dream api a bit. (use onStart as hook for subclasses).
Coordinate properly with power manager.
Replace old dock mode (don't fire old intent).
Change-Id: I1318d20cc1613e5d862f2913f2fcdc9719302cf7
Bug: 6921930
- Expose the existing Context.sendBroadcast() as
Context.sendBroadcastAsUser().
- Add new android:singleUser attribute for services.
- Add new INTERACT_ACROSS_USERS_FULL permission for full
system-level access to cross-user interface (allows
sendBroadcastAsUser() to send to any receiver).
- Add new INTERACT_ACROSS_USERS_FULL permission for
more restricted cross-user interaction: this is required
for android:singleUser, and allows you to use
sendBroadcastAsUser() but only to send to your own
receivers.
Change-Id: I0de88f6718e9505f4de72e3f45d29c0f503b76e9
