bug:8409891
Snapshots frequently have their clip overwritten due to applying
deferred state - now, store tiling clip information in a separate
rect, outside of the snapshot so it isn't overwritten.
Change-Id: I21ca4c45dcd802eae99e8de86f11525196777ccb
Bug #7146141
When non-rectangular clipping occurs in a layer the render buffer
used as the stencil buffer is not cached. If this happens on a
View's hardware layer the render buffer will live for as long
as the layer is bound to the view. When a stencil buffer is
required because of a call to Canvas.saveLayer() it will be allocated
on every frame. A future change will address this problem.
If "show GPU overdraw" is enabled, non-rectangular clips are not
supported anymore and we fall back to rectangular clips instead.
This is a limitation imposed by OpenGL ES that cannot be worked
around at this time.
This change also improves the Matrix4 implementation to easily
detect when a rect remains a rect after transform.
Change-Id: I0e69fb901792d38bc0c4ca1bf9fdb02d7db415b9
Bug #7253839
1. Make sure we don't make GL calls while recording display lists
2. Disable an early and trivial clip optimization in font renderer
when a perspective transformation is used on the Canvas
Change-Id: I3f1052164239329346854f72d0a0d401fbfecf06
this introduced a dead lock in GradientCache's ctor.
This reverts commit dfe082f63e94cde9aee271c94d13de5e7217e036.
Bug: 7096001
Change-Id: I57b8bbab11fb7cb502fa58e3bbf5d19864db874f
Some views (such as ImageView and TextView) handle non-opaque alpha
values directly. This was originally an optimization, but we can handle it faster
in many cases without this optimization when DisplayList properties are enabled.
Basically, if a view has non-overlapping rendering, we set the alpha value directly
on the renderer (the equivalent of setting it on the Paint object) and draw each
primitive with that alpha value. Doing it this way avoids re-creating DisplayLists
while getting the same speedup that onSetAlpha() used to get pre-DisplayList properties.
Change-Id: I0f7827f075d3b35093a882d4adbb300a1063c288
This adds basic support for clip regions. It is currently disabled at compile
time. Enabling clip regions will require setting up a stencil buffer.
Change-Id: I638616a972276e38737f8ac0633692c3845eaa74
The Snapshot class is getting complicated enough that its implementation
should now live in a separate .cpp file. This will become particularly
useful when support for clip regions and paths will be added later on.
Change-Id: I050fac5683a9f7a0ff2f7a6beec3dd28aa5eb0d8
Bug #5650514
After invoking a GL functor, libhwui restores a few OpenGL states
including the current FBO. The renderer was however making the
wrong assumption that the FBO to restore to was the base layer
instead of the FBO associated with the current canvas state.
Change-Id: Ie565500832ebffd673f6a43b83422d6cc05470a0
This bug was affecting several third party applications. The bug was
introduced by an optimization that prevented the local clip to be
recomputed when needed.
Change-Id: I7b5c464db21b5ff1a7274af75a3a64d0e9d84772
With this change, the rendere keeps track of what regions are rendered into
and generates a mesh that matches these regions exactly. The mesh is used
to composite the layer on screen.
Change-Id: I1f342576b9134fb29caff7fb8f4c1da179fe956d
This optimization is currently disabled until Launcher is
modified to take advantage of it. The optimization can be
enabled by turning on RENDER_LAYERS_AS_REGIONS in the
OpenGLRenderer.h file.
Change-Id: I2fdf59d0f4dc690a3d7f712173ab8db3848b27b1
The local clip rect could be off by one pixel when a 3D transform
was applied to a text primitive. This would cause small rendering
glitches in Launcher in particular.
Change-Id: I1254f6a5838c26810cbe969b096679d40b8dd504
Prior to this change layers would clip their content incorrectly. They would
also not apply alpha properly.
Change-Id: Id7b3aaa7dbdc51de68fe050e64458f68e40503fd
This changes binds all textures to GL_TEXTURE0, this will have
to be changed when combining shader capabilities.
Change-Id: I02df4f5ba41e9b01ffa52fd7c26b41477c7ed18f
The LinearGradient class keeps a copy of the various parameters that
define the gradient. The copies are native arrays to avoid copying
Java arrays on every draw call. The gradient code path is implemented
until OpenGLRenderer::drawRect() (see TODO.) The actual gradient
implementation will be added in a latter change.
Change-Id: I9300d250ef5e2e9c2e097c3116ee71dfc9d752d8
This change also modifies the way the clip is stored. The clip is now
always stored in screen-space coordinates.
Change-Id: I96375784d82dfe975bc6477a159e6866e7052487
This change also cleans up the internal API a little bit by using mat4
everywhere instead of float[16] (for the ortho matrix for instance.)
Change-Id: I35924c7dc17bad17f30307118d5ed437c2ed37e0
The cache is used to draw layers so that a new
texture does not have to be recreated every time
a call to saveLayer() happens.
The FBO cache used a KeyedVector, which is a bad
idea. The cache should be able to store several
FBOs of the same size (this happens a lot during
scrolling with fading edges for instance.) This
will be changed in a future CL.
Change-Id: Ic316189e625f0dbcf0d273a71cc981a433d48726
The next step will be to add an FBO cache to avoid churning memory on every
frame we draw. This change also adds support for drawBitmap(Bitmap, Matrix, Paint).
Change-Id: I7825cdcf0cad9bffe6219e05d8328a53d4a6e583
This change also refactors the code base a bit by moving classes out of
OpenGLRenderer into separate headers/implementations. This makes the code
more manageable.
This change also adds documentation for implementation methods. The
undocumented methods are simply Skia's Canvas methods.
Change-Id: I54c68b443580a0129251dddc1a7ac95813d5289e
