Extracted the input system from the window manager service into
a new input manager service. This will make it easier to
offer new input-related features to applications.
Cleaned up the input manager service JNI layer somewhat to get rid
of all of the unnecessary checks for whether the input manager
had been initialized. Simplified the callback layer as well.
Change-Id: I3175d01307aed1420780d3c093d2694b41edf66e
The status bar and navigation bar are two completely separate
elements, with their own semantics. The system bar now classifies
itself as a navigation bar, since that is really how it behaves.
This required rewriting the HDMI resizing code, so that it is
all done by PhoneWindowManager since that is what is responsible
for the size of the navigation bar (and thus now system bar). This
actually gets rid of a fair amount of code, and means we can also
do the same thing for a pure navigation bar.
Likewise the system bar now has the navigation bar ability to be
hidden when requested by system UI flags. To get the behavior
we want on Xoom, we only allow the nav bar to be hidden when it
will help provide a better aspect ratio for showing widescreen
videos.
Finally the nav/system bar now animates when hidden and shown.
Change-Id: Ie927154b68376a0b61802f99171ff56b8da92e7a
- Isolate DimAnimator animation from the layout side.
- Isolate mWallpaperForceHidingChanged and mOrientationChangeComplete
from the animation side.
- Eliminate a redundant setting of mOrientationChangeComplete to true.
It was already true at that point.
- Synchronize changes to mWindows and mAppTokens on mAnimator. This is
a nop until we go to multiple threads.
- Synchronize AppWindowToken.freezingScreen on mAnimator.
- Modification to repeat layout debugging including temporary enabling
of spew on layout repeats.
Change-Id: Ic8d82b1c197144aaf6957caa5f71e175288220f2
Replace four booleans (mDrawPending, mCommitDrawPending, mReadyToShow
and mHasDrawn) with a single int that can take on the four states.
Move mLastHidden from WindowState to WindowStateAnimator.
Change-Id: Ieff319dfa19123bf5a6cdc98e9ab28fd432b8153
Replace references to mWinAnimator.mSurface with new member
mHasSurface.
Clean up odd looping structures.
Simplify logging.
Change-Id: I9cc52a657044220d7b92528928b11bb18a724aef
Moved drawPending and commitDrawPending and associated methods from
WindowState to WindowStateAnimator.
Created mechanism for passing results from WindowAnimator to
WindowManagerService. Initial results passed are mUpdateRotation and
mWallpaperMayChange.
Change-Id: Ib03d28f921580ac9426ea9233bea6eafc9ea964c
The mUpdateRotation variable was still in the WindowManagerService
mInnerFields object. This was masking the true mUpdateRotation found in
WindowAnimator.
Fixes Bug 6240025.
Change-Id: I6531002f870f30d22e19ba9af5cac86e1c7b9bcb
Migrated the bulk of Surface operations from WindowState to
WindowStateAnimator. There remain a multitude of cross-referencing
between the two classes and most of the other classes in the wm
package.
Change-Id: I4bfdfb84be31341371f3ef311aca8fc6a4966692
More refactoring. This time wallpaper animations were broken up from
WindowManagerService and the layout piece kept there while the
animation piece was moved into WindwoAnimator.
Also, applyAnimationLocked and applyEnterAnimationLocked were moved
from WindowManagerService to WindowState.
Change-Id: I05935023702ce05fdfdc804342ec14f719cdfea4
Remove the animation stepping from WindowState and move it into a new
class, WindowStateAnimator. Update all references to moved variables
in related files.
Change-Id: I7540d8f897b370c73975f3ffe450140861cb0cd1
In cases where a Surface does not go from hidden to shown, do not set
the perform layout flag. This keeps us out of repeated passes through
the layout code.
Fixes bug 6222487.
Change-Id: I22601bef5733d2f996a8cbdd50d6b89517bc3122
Introduction of the class WindowAnimator which takes care of all
animations stepping and Surface operations.
Change-Id: I78f1c269fa57df0616c08adbe156e3059709ae48
Removal of blur layer.
Deferral of Surface actions in BlackFrame from ctor to first use.
Combine common test into single method okToDisplay().
Remove redundant logic in DimAnimator.
Change-Id: I43af0415794a8f142803ce94d7e17539aafac67d
Surfaces were displaying animations in their entirety for a single
frame before the animation kicked in. This caused a flash on the
screen. By setting the animation to invisible (alpha=0) at their
start it makes no difference if they are displayed.
Fixed bug 6176540.
Removed redundant mDimAnimator.show call.
Change-Id: I47c1b0d38273b011d9115822a8476671d6a050fc
When stepAnimation returns false, do not return false immediately.
Instead carry out finish actions. Also, remove state machine that is no
longer necessary.
Fixes bug 6184070.
Change-Id: I530eb2b62b864bbce929f573d10b31b102152f1f
Further work to isolate layout from animation and surface operations.
Remove cruft and minor refactoring.
Change-Id: I6f910ed72c7c614996641c353870c2b2ab5e8bb4
(Dianne) pulled the animation steps out of the layout. Changes to
exposed layers cause repeated calls to layout code.
Combined animation steps into start and finish animation code.
Change-Id: I3602d1d6249d20987d102a54e3a67a7a39361b55
This performance an animation when, for example, a dialog window is
moved because the size of its content has changed.
Change-Id: I2d79a1a57f94e0f2f8ef706a473fca6c9cc637cf
Previous approximations weren't indicating completion and windows weren't being layered correctly as a result.
Change-Id: I08fcd278485bb87dc10bca257b9f8073108753f3
- Move mPolicy.startAnimationLw and mPolicy.finishAnimationLw into same method as mPolicy.animatingWindowLw.
- Fix first parameter of performLayoutLockedInner(initial, ...) to pass true on initial pass.
Change-Id: If1b47bb8a7e03cf427769c657e371abc0910b3e3
Removed the listeners and schedule animation / draw methods.
Instead all requests are posted as one-shot callbacks, which is a
better match for how clients actually use the Choreographer.
Bug: 5721047
Change-Id: I113180b2713a300e4444d0d987f52b8157b7ac15
This is between the two previous attempts. I returned the part from the
original that was breaking gallery, but have some new code to detect when
something about the window params has changed that would require a
layout pass to make sure we still do a layout then, even if the window is
not currently visible.
Change-Id: I07745e1f66022583e3076b84cc8bbe8bd2acd48f
Don't consider a window as a candidate for the top fullscreen window
if it is not going to be a candiate for layout.
This fix does not include the change to ignore app tokens that are
hidden. This causes problems in some dialogs that stay hidden until
their app is ready to display, but need to perform a series of relayouts
during that time to get to the right size. Dropping this part of
the change still (mostly?) seems to allow us to avoid the bad states.
Change-Id: Ic052cb1499d3287f47e9ffeac5cd2470ee5a308c
Don't consider a window as a candidate for the top fullscreen window
if it is not going to be a candiate for layout.
Also don't consider windows a candidate for layout if their app token
is hidden. This fixes a transient state where we are preparing to
unhide the window but have not done so yet.
Change-Id: Ife5299ffa003c1df1a4f787b7a2809cbf614ec16
We need to work more like before in determining whether the menu
key is needed -- in some cases look back in the window list to
determine this if we don't know the value from the current window.
This requires adding a new private flag indicating whether the
compat menu state is known for a window, which is set by
PhoneWindow as part of its existing process of computing the flag
for its own windows.
Now we can have a new API on WindowState to determine the value
of this flag for a window, which if needed walks back in the window list
to find a window the value is known for (or stops at what the policy
has determined is the top full-screen window, so we stop like we used
to at things like the lock screen or the bottom of an application).
Change-Id: I829de6d629b5af8bcb422cb85249ee4041c7205e
Make surface management between SurfaceView and the window manager
much more controlled, to ensure that SurfaceView always gets to report
the current surface is destroyed before the window manager actually
destroys it.
Also a small tweak to allow windows that have a wallpaper background
to still have a preview window. This makes launching home after it
has been killed feel much more responsive.
Change-Id: I0d22cf178a499601a770cb1dbadef7487e392d85
Introduce a new concept of "B" services. All running services are
classified as either A or B. B services are later in the LRU list.
Their oom_adj is after the home app. This allows us to better pick
services to kill based on how long they have running, and should
reduce the amount that we end up killing the home app.
This temporarly turns on a debug log when the oom_adj of a process
is changed. Sorry, I know it is noisy. This is needed to try to
track down why some processes are being killed.
Also add a flag to the SyncManager's service binding to allow the
syncing process to be more aggressively killed if it has done UI.
This is to address cases we have seen where sync is causing an 80MB
gmail process to be kept around, preventing other process from running.
Now what will happen is that the syncing process will aggressively be
killed by the system, and can then be restarted in a much lighter-weight
state.
Do a little tweak in the power manager to allow us to still do smooth
brightness changes even when the fancy TV off animation is in use.
And get rid of a debug log in the window manager that was accidentally
left in.
Change-Id: I64a8eeaaa1f096bab29c665fbff804c7f1d029e2
Fix a few places where we would unfreeze the screen too early.
Now that we are no longer relying on surface flinger freezing, we
can't depend on it keeping the screen frozen until surfaces get
drawn.
Change-Id: Icb03bf30c9599a5e2016817bfa5ca6458adc7249
The window manager now uses the app screen dimensions to compute
the various configuration properties, as it should.
This means that prime is official a "not long" device. Poor prime.
It probably feels inadequate now.
Because it is.
Oh and all that other stuff? Debugging logs. Turned off. And
why the heck not, debugging logs are great.
Change-Id: Iaaf8ef270d986d34fd046d699ef4c0ecea1981fc
This cleans up how ui flags are managed between the client and window manager.
It still reports the global UI mode state to the callback, but we now only clear
certain flags when the system goes out of a state (currently this just means the
hide nav bar mode), and don't corrupt other flags in the application when the
global state changes.
Also introduces a sequence number between the app and window manager, to avoid
using bad old data coming from the app during these transitions.
Change-Id: I40bbd12d9b7b69fc0ff1c7dc0cb58a933d4dfb23
This change ensures the wallpaper is rendered into an opaque surface
which avoids a glClear() in SurfaceFlinger. This should save quite
a bit of work on every frame when panning the workspace in launcher.
Change-Id: I9c1b8c324edf29826d5dbb1fb39d883e43375310
The key thing was to fix isVisibleOrBehindKeyguardLw() so that it
wouldn't count a window as not visible if it was just currently
in the process of drawing due to an orientation change.
Also improve logic in deciding when to turn screen on to better ensure
the screen is in a stable state, in particular treating screen off
as a frozen screen and not allowing it to turn on until the
update of the screen due to any config change is done.
Change-Id: If82199f3773270b2d07f9c7de9da2dad8c7b28d7
Bug: 5156144
Input channels could leak or simply live longer than they should
in some cases.
1. Monitor channels (used by the pointer location overlay) are never
unregistered, so they would leak.
Added code to handle failures in the receive callback by closing
the input channel.
2. The DragState held onto its input window and application handles
even after the input channel was disposed.
Added code to null these handles out when they are no longer needed.
3. Input channels previously used as input event targets would stick
around until the targets were cleared (usually on the next
event).
Added code to detect when the input dispatcher is in
an idle state and to proactively clear the targets then
to ensure that resources are released promptly.
4. Native input window handles held onto the input channel even
after the input window was removed from the input dispatcher.
Consequently, the input channel would not be disposed until
the input window handle itself was freed. Since the input
window handle is held from managed code, this meant that the
window's input channel could stick around until the next GC.
Refactored the input window handle to separate the properties
(info) and identify (handle) state into different objects.
Then modified the dispatcher to release the properties (info)
when no longer needed, including the input channel.
7. The pointer location overlay does not actually use its
standard input channel, only the monitor input channel.
Added INPUT_FEATURE_NO_INPUT_CHANNEL to allow windows to
request that they not be provided with an input channel
at all.
Improved some of the error handling logic to emit the status
code as part of the exception message.
Change-Id: I01988d4391a70c6678c8b0e936ca051af680b1a5
The input reader needs this information so that it knows how to
interpolate touches on an external touch screen.
Changed Display so that it asks the WindowManager what the real
display size is (as opposed to the raw display size). This means
it now takes into the forced display size set by
adb shell am display-size.
Replaced all calls to getRealWidth() / getRealHeight() /
getRealMetrics() in the WindowManager and replaced them with direct
usages of the mCurDisplayWidth / mCurDisplayHeight so that the WM
doesn't end up making a reentrant Binder call into itself.
Fixed the table status bar HeightReceiver so that it updates the
height on all configuration changes since it is possible that the
display size changed independently of an external HDMI display
being plugged / unplugged.
Improved the Display class documentation to make the distinctions
betweeen the various sizes clearer.
Change-Id: I3f75de559d3ebffed532ab46c4ae52c5e7f1da2b
This change moves the cached window and application input state
into the handle objects themselves. It simplifies the dispatcher
somewhat because it no longer needs to fix up references to
transient InputWindow objects each time the window list is updated.
This change will also make it easier to optimize setInputWindows
to avoid doing a lot of redundant data copying. In principle, only
the modified fields need to be updated. However, for now we
continue to update all fields in unison as before.
It turns out that the input dispatcher was inappropriately retaining
pointers to InputWindow objects within the mWindows InputWindow
vector. This vector is copy-on-write so it is possible and the
item pointers to change if an editing operation is performed on
the vector when it does not exclusively own the underlying
SharedBuffer. This bug was uncovered by a previous change that
replaced calls to clear() and appendVector() with a simple use
of operator= which caused the buffer to be shared. Consequently
after editItemAt was called (which it shouldn't have, actually)
the buffer was copied and the cached InputWindow pointers became
invalid. Oops. This change fixes the problem.
Change-Id: I0a259339a6015fcf9113dc4081a6875e047fd425
