NOTE: Linear blending is currently disabled in this CL as the
feature is still a work in progress
Android currently performs all blending (any kind of linear math
on colors really) on gamma-encoded colors. Since Android assumes
that the default color space is sRGB, all bitmaps and colors
are encoded with the sRGB Opto-Electronic Conversion Function
(OECF, which can be approximated with a power function). Since
the power curve is not linear, our linear math is incorrect.
The result is that we generate colors that tend to be too dark;
this affects blending but also anti-aliasing, gradients, blurs,
etc.
The solution is to convert gamma-encoded colors back to linear
space before doing any math on them, using the sRGB Electo-Optical
Conversion Function (EOCF). This is achieved in different
ways in different parts of the pipeline:
- Using hardware conversions when sampling from OpenGL textures
or writing into OpenGL frame buffers
- Using software conversion functions, to translate app-supplied
colors to and from sRGB
- Using Skia's color spaces
Any type of processing on colors must roughly ollow these steps:
[sRGB input]->EOCF->[linear data]->[processing]->OECF->[sRGB output]
For the sRGB color space, the conversion functions are defined as
follows:
OECF(linear) :=
linear <= 0.0031308 ? linear * 12.92 : (pow(linear, 1/2.4) * 1.055) - 0.055
EOCF(srgb) :=
srgb <= 0.04045 ? srgb / 12.92 : pow((srgb + 0.055) / 1.055, 2.4)
The EOCF is simply the reciprocal of the OECF.
While it is highly recommended to use the exact sRGB conversion
functions everywhere possible, it is sometimes useful or beneficial
to rely on approximations:
- pow(x,2.2) and pow(x,1/2.2)
- x^2 and sqrt(x)
The latter is particularly useful in fragment shaders (for instance
to apply dithering in sRGB space), especially if the sqrt() can be
replaced with an inversesqrt().
Here is a fairly exhaustive list of modifications implemented
in this CL:
- Set TARGET_ENABLE_LINEAR_BLENDING := false in BoardConfig.mk
to disable linear blending. This is only for GLES 2.0 GPUs
with no hardware sRGB support. This flag is currently assumed
to be false (see note above)
- sRGB writes are disabled when entering a functor (WebView).
This will need to be fixed at some point
- Skia bitmaps are created with the sRGB color space
- Bitmaps using a 565 config are expanded to 888
- Linear blending is disabled when entering a functor
- External textures are not properly sampled (see below)
- Gradients are interpolated in linear space
- Texture-based dithering was replaced with analytical dithering
- Dithering is done in the quantization color space, which is
why we must do EOCF(OECF(color)+dither)
- Text is now gamma corrected differently depending on the luminance
of the source pixel. The asumption is that a bright pixel will be
blended on a dark background and the other way around. The source
alpha is gamma corrected to thicken dark on bright and thin
bright on dark to match the intended design of fonts. This also
matches the behavior of popular design/drawing applications
- Removed the asset atlas. It did not contain anything useful and
could not be sampled in sRGB without a yet-to-be-defined GL
extension
- The last column of color matrices is converted to linear space
because its value are added to linear colors
Missing features:
- Resource qualifier?
- Regeneration of goldeng images for automated tests
- Handle alpha8/grey8 properly
- Disable sRGB write for layers with external textures
Test: Manual testing while work in progress
Bug: 29940137
Change-Id: I6a07b15ab49b554377cd33a36b6d9971a15e9a0b
bug:29127615
Primarily fixes case where 0 dimensioned layers could be
created/updated. Additionally, adds more logging in incomplete
framebuffer cases, if they still occur.
Change-Id: Ib90dbbafd6905aca3c8f46e64064e13a308f713d
Bug: 28667141
saveLayer clips the layer to the size it needs to
be and will translate content if necessary, but
the drawLayerOp that results from that was not
translated to handle the shifted draw content.
This fixes that
Change-Id: I3c9ffd5d0282fa1b958bced94c25e9744281e9be
Fixes: 28125010
Restructures 'scene defer', to implement window backdrop overdraw
avoidance in new render pipeline, and disable clipping to content draw
bounds.
Also restructures FrameBuilder's constructors, to separate out into
multiple defer methods.
Change-Id: I53facb904c1a4a4acc493d8a489921a79a50494e
bug:27873093
Adds the simple overdraw avoidance optimization to the new
pipeline. This means when LayerBuilder defers draws that are opaque over
the full area of the repaint region, it will discard all drawing content
beneth.
Also moves a lot of complexity out of BakedOpState's header.
Change-Id: Iffca6d8e1b170ef31a5d6c83d25592670e02323d
Fixes: 27825042
TextureView's matrix may not be set at record time - delay using it
until on RenderThread, when deferring the op.
Change-Id: Icf8b55d656e304ec049ca803b042dc2359482db2
bug:27810783
Previous drawColor->drawPaint conversion failed to preserve unbounded
nature of drawColor from old pipeline.
Change-Id: Ifd7a7b9d645f0887e252e48ca95d3195ee31615f
bug:26591194
Also, revert to using current clip bounds as drawColor/drawPaint bounds
for simplicity in new pipeline.
Change-Id: I1a6b3f9716b564b46df41d57dfe14475fdd24de0
bug:27275799
Skip applying clip/matrix properties from projected node op, since we
don't want to respect the clip, and matrix is already baked into
transformFromCompositedAncestor.
This skips op clips in the new pipeline, and fixed a double application
of scroll on ripple backgrounds.
Change-Id: I4f72448fe0463ab666564ca538d8b6bf525d98de
bug:27225580
bug:27281241
Empty unclipped savelayers (clipped at defer time, often by dirty rect)
were resulting in invalid layer clear rectangles. Simplify by just
rejecting these unclipped savelayers entirely at defer.
Also, use repaint rect as base clip for constructed ops within
LayerBuilder.
Change-Id: I5c466199e85201a2f68f5cdc60b29187c849961b
Bug: 27072626
Or at least fix this one and make it harder to leak
going forward. static_asserts are cool, use them liberally
Also makes allocation via LA (slightly) faster by fixing create*
variants to use rvalue references & std::forward, preventing
an accidental pass-by-value
Change-Id: I18fd696a93dd6ea49c8d8c587d4c8bee56dbfdf5
Bug: 27072626
Or at least fix this one and make it harder to leak
going forward. static_asserts are cool, use them liberally
Also makes allocation via LA (slightly) faster by fixing create*
variants to use rvalue references & std::forward, preventing
an accidental pass-by-value
Change-Id: I5e0e36e72c6dd93324194ebf9a95f8204f05f261
bug:26524690
Don't intersect the first clip with the viewport. Instead, the first
clip op should always be a replace op.
Additionally, only quick reject nodes that clip to bounds, since some
nodes (like ripples) draw outside their own bounds.
Change-Id: I96a52029f360328aba19af7349888cc0a026b5b1
This CL runs VectorDrawable animation on RenderThread. The changes in this CL
include:
- Convert all the animators in AnimatorSet for AVD into a set of RenderNodeAnimators.
- Hook up the new animators with RenderThread
- Add drawOp in RecordingCanvas for drawing VD so that during the animation
on RenderThread, all the property changes on VD can be reflected on the screen.
TODO:
- Implement reverse and reset for AVD.
Change-Id: I2df1d754f2db0ad098d9c15dde4bb2bdfafc2315
Skia's SkCanvas::SaveFlags are being deprecated. This CL introduces
the equivalent android::SaveFlags, converts all internal clients to
the new enum, and switches the saveLayer glue to the
SaveLayerRec-based API.
Change-Id: Icb1785f4e7c0f652b1f04b34a1e3ccb063c408f3
