Move the projection surface to be a property of a DisplayList,
set to true for every background drawable.
Additionally, handle a projecting view background such that it doesn't
try to project onto itself (which is undesirable).
Change-Id: Ic70b17474bd87340e80767f8518f73b233419c7a
For now, ancestor views signal the acceptance of projections with a
save(0x20)/restore pair.
During the order traversal, each view with the save(0x20) code will
collect descendent views with mProjectToContainedVolume (which still
needs to be renamed) so that they can be drawn out of order later.
- *Temporary* sample code added to HwAccelerationTest.
- Note that a projected displaylist must not be clipped.
Change-Id: I45c493e845961535b958d59c53e8aff3f8891d9f
The eventual goal is for the StatefulBaseRenderer to serve as the
common base class between the DisplayListRenderer and OpenGLRenderer.
This will separate DisplayList recording, Snapshot stack management,
and the GL in OpenGLRenderer.
Additionally, avoid sp<> parameters, and use const parameters in
several places, with the intent of greatly reducing the surface area
where renderer subclasses can modify snapshot stack.
Next steps:
-move bulk of clipping logic into StatefulBaseRenderer
-disable direct snapshot access
Change-Id: Ibc3c6747134ec7daf8ea535866239fa73b874390
This will eventually serve as a base class to allow
DisplayListRenderer to split off from OpenGLRenderer, and could
eventually support other rendering approaches, such as an
SkCanvas/SkPicture.
This will also be the main source of (implementation-independent)
documentation of the canvas/renderer methods.
Change-Id: I52047f338f5cf86a3b0b3002af7154bff5c3c227
True 3d transformations are now supported by DisplayLists and the
renderer, initially with the translationZ property on view.
Renderer operations used directly by DisplayList (formerly,
clip/save/restore/saveLayer) are now more simply managed by allocating
them temporarily on the handler's allocator, which exists for a single
frame. This is much simpler than continuing to expand the pool of
pre-allocated DisplayListOps now that more operations are called
directly by DisplayList, especially with z ordered drawing.
Still TODO:
-APIs for camera positioning, shadows
-Make Z apis public, and expose through XML
-Make invalidation / input 3d aware
Change-Id: I95fe6fa03f9b6ddd34a7e0c6ec8dd9fe47c6c6eb
bug:4351353
quickReject and quickRejectNoScissor have been renamed and refactored:
- to make the scissor side effect clear and explicit
- dangerous methods no longer public
- to make the simple quick reject check logic const
- simple quick reject is now conservative
This CL also fixes several issues with line and point quickRejection -
sub-pixel and hairline lines are much less likely to be incorrectly
rejected, especially at small canvas scale.
Additionally, alpha modulation for AA points < 1px in size is now
correct, dumplicating SW behavior (similar to lines and stroked
shapes work).
Change-Id: Ibb0710c721b9fb415d05acf54dd3d2b4d602156a
bug:9969358
Instead of storing DeferredDisplayState within an op (thus forcing ops
to be tied to a single state instance), associate each op with a new
state at DeferredDisplayList insertion time.
Now, DisplayLists (and the ops within) can be reused in a single
DeferredDisplayList draw call, as ops will use different state
instances at different points in the frame.
Change-Id: I525ab2abe0c3883679f2fa00b219b293e9ec53d9
Bug #9589379
If the atlas is terminated/reinitialized and a view does not invalidate
in between it might end up using a stale AssetAtlas::Entry. This change
is similar to how 9patch meshes are cached in DrawPatchOp: we simply
track the generation ID of the cache to make sure we always use the
latest data.
Change-Id: Ib5abb3769d2ce0eabe9adc04e320ca27c422019e
This change adds refcounting of Res_png_9patch instances, the native
data structure used to represent 9-patches. The Dalvik NinePatch class
now holds a native pointer instead of a Dalvik byte[]. This pointer
is used whenever we need to draw the 9-patch (software or hardware.)
Since we are now tracking garbage collection of NinePatch objects
libhwui's PatchCache must keep a list of free blocks in the VBO
used to store the meshes.
This change also removes unnecessary instances tracking from
GLES20DisplayList. Bitmaps and 9-patches are refcounted at the
native level and do not need to be tracked by the Dalvik layer.
Change-Id: Ib8682d573a538aaf1945f8ec5a9bd5da5d16f74b
This change also fixes the way batched bitmaps were handled
inside a layer. The layer is now correctly dirtied to minimize
the amount of pixels to blend.
Fix alpha, mode and opaque computations for DrawPatchOp.
Change-Id: I1b6cd581c0f0db66c1002bb4fb1a9811e55bfa78
bug:8766924
Previously text bounds were calculated to be from 0 to totalAdvance in
the X, and from the font's top to bottom. These are incorrect,
especially in light of the font fallback mechanism.
Now, we calculate the bounds of the text as we layout each glyph.
Since these are much tighter bounds in the common case, this
significantly reduces the amount of clipping required (which in turn
enables more aggressive text merging).
Change-Id: I172e5466bf5975bf837af894a9964c41db538746
When the Android runtime starts, the system preloads a series of assets
in the Zygote process. These assets are shared across all processes.
Unfortunately, each one of these assets is later uploaded in its own
OpenGL texture, once per process. This wastes memory and generates
unnecessary OpenGL state changes.
This CL introduces an asset server that provides an atlas to all processes.
Note: bitmaps used by skia shaders are *not* sampled from the atlas.
It's an uncommon use case and would require extra texture transforms
in the GL shaders.
WHAT IS THE ASSETS ATLAS
The "assets atlas" is a single, shareable graphic buffer that contains
all the system's preloaded bitmap drawables (this includes 9-patches.)
The atlas is made of two distinct objects: the graphic buffer that
contains the actual pixels and the map which indicates where each
preloaded bitmap can be found in the atlas (essentially a pair of
x and y coordinates.)
HOW IS THE ASSETS ATLAS GENERATED
Because we need to support a wide variety of devices and because it
is easy to change the list of preloaded drawables, the atlas is
generated at runtime, during the startup phase of the system process.
There are several steps that lead to the atlas generation:
1. If the device is booting for the first time, or if the device was
updated, we need to find the best atlas configuration. To do so,
the atlas service tries a number of width, height and algorithm
variations that allows us to pack as many assets as possible while
using as little memory as possible. Once a best configuration is found,
it gets written to disk in /data/system/framework_atlas
2. Given a best configuration (algorithm variant, dimensions and
number of bitmaps that can be packed in the atlas), the atlas service
packs all the preloaded bitmaps into a single graphic buffer object.
3. The packing is done using Skia in a temporary native bitmap. The
Skia bitmap is then copied into the graphic buffer using OpenGL ES
to benefit from texture swizzling.
HOW PROCESSES USE THE ATLAS
Whenever a process' hardware renderer initializes its EGL context,
it queries the atlas service for the graphic buffer and the map.
It is important to remember that both the context and the map will
be valid for the lifetime of the hardware renderer (if the system
process goes down, all apps get killed as well.)
Every time the hardware renderer needs to render a bitmap, it first
checks whether the bitmap can be found in the assets atlas. When
the bitmap is part of the atlas, texture coordinates are remapped
appropriately before rendering.
Change-Id: I8eaecf53e7f6a33d90da3d0047c5ceec89ea3af0
Merge simple bitmap draw operations and text operations to avoid
issuing individual gl draws for each operation. Merging other ops to
be done eventually.
The methods are different - the bitmap merging generates a single
mesh for reused, unclipped images (esp. repeated images in a listview)
The text approach queries just defers the normal font rendering until
the last drawText in the sequence that can share the same shader.
Patches are sorted and merged, but don't yet have a multiDraw
implementation. For now, the pretending-to-merge gives better sorting
behavior by keeping similar patches together.
Change-Id: Ic300cdab0a53814cf7b09c58bf54b1bf0f58ccd6
Bug #8437401
A misplaced ref count decrement was causing a crash when attempting to
resize a layer to dimensions larger than the max texture size supported
by the GPU.
This change fixes the crash and clarifies the warnings to make it more
obvious what's happening.
Change-Id: I632dc1b90aaa2605969e10523491a81c4922d3dc
This change will greatly simplify the multi-threading of all
shape types.
This change also uses PathTessellator to render convex paths.
Change-Id: I4e65bc95c9d24ecae2183b72204de5c2dfb6ada4
Removes mMultipliedAlpha, using the snapshot alpha for all
non-overlapping display list alpha control.
Additionally, fixes opacity issues where children of
hasOverlappingRendering=false displaylists (both hw layer sublists and
other sublists with hasOverlappingRendering=false)
Change-Id: I6adc16da855835f9f518f8967628e5d0135c789b
bug:8037003
A recursive drawDisplayList call is now entirely deferred before
playing back to the screen and issuing GL commands. This way, the
entire stream can be inspected, optimized, and batch work (such as
uploading textures) before issuing commands.
Additionally, this fixes an issue where operations draw could move
across restores corresponding to saveLayer(alpha). Those and other
similar cases (such as complex clipping, requiring the stencil) are
now treated as batching barriers, with the operations that change
renderer state in a way that's difficult to defer are just re-issued
at flush time.
Change-Id: Ie7348166662a5ad89fb9b1e87558334fb826b01e
The deferred display lists model now allows us to precache glyphs
at their exact size on screen.
This change also removes debug markers when the renderer defers
and reorders display lists. It also adds a flush event marker.
Change-Id: I66ec5216dc12b93ecfdad52a7146b1cfb31fbeb4
If a perspective transform is set on the Canvas, drawText() should
not attempt to rasterize glyphs in screen space. This change uses
the old behavior instead (i.e. rasterize the glyphs at the native
font size and apply the transform on the resulting mesh.)
This change also adds an optimization: empty glyphs (spaces) do
not generate vertices anymore. This saves a lot of vertices in text
heavy applications such as Gmail.
Change-Id: Ib531384163f5165b5785501612a7b1474f3ff599
bug:8037003
The reordering enables similar operations to draw together, minimizing the
OpenGL state change operations that go inbetween draws. Eventually, multiple
complete canvas draw operations will be merged (into a single glDrawArrays call,
for example)
Reorders DisplayList draw operations when:
-They can move backwards in the command stream to be after similar
operations without violating draw ordering
-The OpenGLRenderer is in a simple, replayable state (no complex clip,
or filter/shadow etc)
Also adds two system properties to control the deferral/reordering:
"debug.hwui.disable_draw_defer"
"debug.hwui.disable_draw_reorder"
which can be set to "true" to control the display list manipulation
Change-Id: I5e89f3cb0ea2d2afd3e15c64d7f32b8406777a32
A layer will, by definition, clip its content already. The extra
clip added by the UI toolkit/renderer is wasteful and in some
situations very expensive. If a layered view is rotated, the clip
will be transformed into a non-rectangular region, thus forcing
the renderer to rely on the stencil buffer. The whole point of
using layers is to draw efficiently so let's avoid this case.
Change-Id: I813465f6468cc386b84a4d058c11a258fef8507a
bug:8037003
Changes the DisplayList from using stream read/write commands to use an array of
objects manually allocated on a linear buffer.
Depends on frameworks/native change https://googleplex-android-review.googlesource.com/#/c/257695/ which adds LinearAllocator
Also changes drawRects to use float count instead of rect count, to be more like drawLines/drawPoints
Change-Id: Ia2e4a11acd8f0a757042a05cbc9e7563cb73ee47
Region clipping, using Canvas.clipPath or Canvas.clipRegion, requires
a stencil buffer to be always present. In addition, extra wiring is
required in JNI and display lists.
This change only adds the necessary JNI/C++ APIs and some extra
plumbing to start the real work on properly supporting region
clipping.
A new debug define called DEBUG_CLIP_REGIONS can be used to draw
the current clip region. It is off by default, as is region
clipping.
The default implementation of clipPath() and clipRegion(), now
in native, mimics the previous Dalvik implementation to prevent
regressions.
Change-Id: I7903e7cfd7412b9b9b622566d4dbfce7bdcec00c
Fonts are now described by a transform matrix. This lead to switching
from a vector to a hashmap. This change therefore adds new comparators
and hash computations to Font.
Change-Id: I2daffa7d6287c18554c606b8bfa06640d28b4530
When profiling is enabled with debug.hwui.profile set to true,
setting debug.hwui.profile_visualizer to true will display the
profiling data directly on screen.
Change-Id: I3d5fe3f0347090815087b1cbfce66b8e76d9347b
A bug in software rendering caused animations on views that are offscreen
to not get drawn, therefore the animation doesn't continue (since old-style
animations depend on the logic in the drawing code to keep running). Fix is
to special case the isAnimating case in ViewRoot to go ahead and schedule
a traversal even if the dirty rect does not intersect with the visible region.
Issue #7396035 Animations starting offscreen don't draw run/end/draw properly (sw rendering only)
Change-Id: Iae25b3a424ddc5a16ba431ecd68cf42d5500db3f
Bug #7326824
When a layer is taken out of the cache and initialized it gets cleared
to prepare it for future rendering. This triggers the following sequence
of operations:
glBindFramebuffer(layer.fbo)
attach texture storage to FBO
glClear()
glBindFramebuffer(defaultFbo)
The clear forces a resolve on tilers which stalls the CPU for a little
while, thus producing jank during animations. This change moves the
clear to the next frame when we know we will have to execute a resolve
anyway.
Change-Id: Ic1939c25df20ed65a4c48dc81ee549b2cd8b6ec3
When memory gets low on a device, activities flush everything they can.
Hardware-accelerated activites, such as Launcher, flush GL resources and destroy
the GL context. However, some resources were still hanging around, due to deferred
destruction policies (we don't delete layers until the DisplayLists they are in
are finalized, to ensure we don't deref deleted objects). This meant that we were
referring to obsolete GL data in these objects. in particular, it meant that we might
come around later, after a new GL context was created, and delete a texture object
that was incorrect. We use the layer's "texture id" to refer to the texture underlying the
layer. But if there's a new GL context, then this texture ID is no longer valid, and
we may be deleting the texture that a different object (layer, icon, whatever) is referring
to, because the driver may return that same ID under the new GL context.
The fix is to more aggressively delete things that we know will not be used again
when the GL context is destroyed. In particular, we delete all resources being used
by all DisplayLists at GL context destruction time.
Issue #7195815 Textures corruption on all devices, in many apps
Change-Id: I52d2d208173690dbb794a83402d38f14ea4c6c22
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
Bug #7195815
We did not reclaim resources when reusing an existing DisplayList to
record a new empty list of commands. This would lead to various memory
leaks: bitmaps, paints, paths, matrices, etc.
This is not a common case but some apps run into this situation,
such as Launcher.
Change-Id: I2eb14ac86a212123f8edbe42b70a7c1c51fa4145
Launcher occasionally crashes with a stack trace indicating that the memory
of a Layer object is corrupt. It is possible for us to delete a Layer
structure and then, briefly, use it to draw a DisplayList again before
that DisplayList gets recreated (without the layer that got deleted).
When this happens, if the memory got corrupted, it's possible to crash.
The fix is to add Layer to the other objects which we currently refcount
(bitmaps, shaders, etc.). Then instead of deleting a Layer, we decrement the
refcount. We increment when creating it, then increment it again when it's
referenced from a DisplayList. Then we decrement the refcount instead of
deleting it, and decrement when we clear a DisplayList that refers to it.
Then when the refcount reaches 0, we delete it.
Issue #6994632 Native crash in launcher when trying to launch all apps screen
Change-Id: I0627be8d49bb2f9ba8d158a84b764bb4e7df934c
Previously, to draw a layered view with a changed Paint object for the
drawLayer operation, you'd have to invalidate the parent view, to get the
native DisplayList to pick up the new Paint properties. This change adds
API and functionality so that the developer can call setLayerPaint(), which
does the proper invalidation (lightweight, doesn't cause redrawing the view).
Issue #6923810 Make it easy to efficiently animate a layer's Paint
Change-Id: I7fea79788d50f6d9c86dd5e5b2a4490cb95142bb
There are two fixes here:
- precaching: instead of caching-then-drawing whenever there is a new
glyph, we cache at DisplayList record time. Then when we finally draw that
DisplayList, we just upload the affected texture(s) once, instead of once
per change. This is a huge savings in upload time, especially when there are
larger glyphs being used by the app.
- packing: Previously, glyphs would line up horizontally on each cache line, leaving
potentially tons of space vertically, especially when smaller glyphs got put into cache
lines intended for large glyphs (which can happen when an app uses lots of unique
glyphs, a common case with, for example, chinese/japanese/korean languages). The new
approach packs glyphs vertically as well as horizontally to use the space more efficiently
and provide space for more glyphs in these situations.
Change-Id: I84338aa25db208c7bf13f3f92b4d05ed40c33527
We recently changed the drawText implementation to accept more parameters
which were not taken into account during a dump operation.
Change-Id: Ia1dc682dc7bb2bb02a1f4a514ed5af5556f92a97
This implementation adds a drawGeneralText() method to the OpenGL
Renderer, which supports both a global x, y position, an array of
individual glyph positions, and also a length parameter (which enables
drawing of underline and strikethrough. It also adds the method to the
display list (with marshalling and unmarshalling).
With this change, the existing drawText() method is removed entirely, as
it's subsumed by the new method. It's easy enough to revert to the old
functionality if needed by passing in a NULL positions array.
Change-Id: I8c9e6ce4309fd51cc5511db85df99f6de8f4f6f5
External bug: http://code.google.com/p/android/issues/detail?id=34946
DisplayListRenderer::drawPath was not invoking quickReject() properly,
passing x,y,width,height instead of left,top,right,bottom. A path
could thus get rejected when it should be drawn instead.
While working on this change I found a similar issue with another
drawing command, drawBitmapData().
Change-Id: I56484e8c101768cde6a78625290872f7849dd5ee
