Launcher swiping looks choppy. It's because we deliver the
motion events of the drags asynchronously, and sometimes we
may get an event to redraw before the motion event is posted, so we
end up drawing again without updating to the latest motion
information.
This fix makes input event processing more proactive. Every time
we run ViewRootImpl.performTraversals() (which is what
happens whenever we need to layout, measure, and/or draw), we
first process all pending input events, ensuring that we are
completely up-to-date with posted events prior to drawing, so that the
drawing we do is synchronous with the events we've received.
This eliminates the choppiness and means that we can now get the full
refresh rate on the screen with drag events.
The fix was done for Launcher, but it is pervasive in the system,
so this may fix other laggy drag behavior as well.
Change-Id: I8dbed6acadc2662f317f736e769f536f555701aa
- Don't try to create a thumbnail bitmap on the client side. This
wastes 64k, and isn't needed since we are doing screenshots.
- Optimize View to put all of the callback pointers out of line.
Added a couple new APIs so these don't need to be protected/public.
- Lazily create ViewGroup's cache paint.
- Change FrameworkPerf app to not use HW accel drawing, to give better
comparison with GB.
Change-Id: Iec56d02459820d74a4cc9c7ec9c1856563c82c7b
1. The window query API used to not-respect the root name space
while traversing the parent relation i.e. a client was able
to fetch the parent of a root name space node.
2. Children that are root name space were reported but their descendants
not. Actually such children should not be reported since they are
the root of a separate logical sub-tree. Such a tree is exposed by
its root allowing its traversal. The accessibility APIs should be
able to explore a virtual tree, i.e. one with a descendant which is
root name space, only if an accessibility event from there was received.
bug:5480096
Change-Id: I4c4d805aa2f6d4edba86eda213b5239bea83eed2
- IME service now switches between visible and perceptible depending on
whether it is being showm, allowing us to more aggressively free its
memory when not shown.
- The activity display time is no longer delayed by the activity
transition animation.
- New -R (repeat) option for launching activities with the am command.
- Improved some documentation on Loader to be clear about some methods
that apps should not normally call.
- FrameworkPerf test now allows you to select individual tests to run.
Change-Id: Id1f73de66dc93d63212183958a72119ad174318b
In order to completely mute the ringer (no vibrate), introduce an extra
state beyond mute, which mutes the vibrator as well, if it was enabled.
Bug: 5530217
Change-Id: Ib1f299ee6bbca56c1aa7e1100662591362d08307
Swiping the home screen causes the WindowManagerService to do
a bunch of work to keep the wallpapers in sync. First, it lays out
and places all windows. Also, it notifies the SystemUI process that
the wallpaper position has changed.
The layout/place operation is too much work - we only need to set
the position values for the wallpaper, not relayout the whole system.
The notification mechanism must exist, but should be optional. Most
wallpapers don't care (especially static ImageWallpapers). So we'll
give them a new API (WallpaperService.Engine.setWantsOffsets()) to
allow wallpapers to opt out of this process and avoid the performance
overhead.
Change-Id: I66c38375438937f14f6f5550565b28eb204b1e06
1. AccessibilityInteractionConnection was non-static inner
class, hence keeping a handle to the enclosing ViewRootImpl
resulting in leaking activities.
bug:5525412
Change-Id: Ie438861663d623d503995844125d9e15d677fc32
Fixed the documentation and comments for dips versus pixels.
Added a new DOUBLE_TAP_TOUCH_SLOP to replace the hard-coded and
*unscaled* value used in GestureDetector. Assuming this value was
ok for hdpi devices, set it to 20 / 1.5 ~= 16 in dips.
New method kept hidden to keep API unchanged.
Change-Id: I6d237faa2fdb7714e0e3558978af130df9a061a5
AccessibilityNodeInfo#getParent is checking whether the source view
accessibility id is specified instead that of the parent.
bug:5505799
Change-Id: I436fd2327def2762b4d307819ba0c23baa9a4e03
Added an interface that is the contract for a client to expose a virtual
view hierarchy to accessibility services. Clients impement this interface
and set it in the View that is the root of the virtual sub-tree. Adding
this finctionality via compostion as opposed to inheritance enables apps
to maintain backwards compatibility by setting the accessibility virtual
hierarchy provider on the View only if the API version is high enough.
bug:5382859
Change-Id: I7e3927b71a5517943c6cb071be2e87fba23132bf
This patch fixes the following warning:
W/Binder: The following Binder class should be static or
leaks might occur: android.view.inputmethod
.InputMethodManager.ControlledInputConnectionWrapper
...in hope of removing a possible memory leaks.
Change-Id: I081d10fd839f2985d7081492f4f5cf0b4da42fd2
Bugs: 5481376, 5461066
This patch fixes the following warning:
W/Binder: The following Binder class should be static or
leaks might occur: android.view.textservice
.SpellCheckerSession.InternalListener
...in hope of removing a possible memory leaks.
However in my tests it does nothing at all for the memory leak
in bug 5461066, so it was probably unrelated.
Change-Id: Id776665e9483121d22cc91e8d2d9f32e3413ec96
Bugs: 5481376,5461066
It is no longer sufficient to check the value of
internal.R.bool.config_showNavigationBar to determine if a
navigation bar (separate from the status bar) is shown on a
device, because the emulator needs to be able to override
this value (now possible by setting qemu.hw.mainkeys to "1"
or "0", for navbar or no navbar, respectively).
This logic is now contained in PhoneWindowManager, and any
clients wishing to know whether the system has a software
nav bar should consult the new hasNavigationBar() method.
Bug: 5404945
Change-Id: I119d32a8c84b88b2ef46f63244e7f11dc5de0359
A LayerScreenshot is a special type of layer that contains a screenshot of
the screen acquired when its created. It works just like LayerDim.
Make sure to call compositionComplete() after rendering into a FBO.
Bug: 5446982, 5467587, 5466259
Change-Id: I5d8a1b4c327f9973d950cd4f4c0bca7f62825cd4
Making the notfication delete-all animation smoother by carefully
choreographing the various parts of it. The problem with the previous
animation was that there was simply too much going on at the
same time, causing things like layout and recreating display-lists
in the middle of animations that became choppy as a result. This
approach swipes all items off quickly, then scrolls the shade up to the
top, making both sets of animations smoother as a result.
Fixes#5431207: Notification delete-all should be smoother
Change-Id: Iefe8ab5d661e05adcd10379dab85227d17904450
