This is fixed in Skia by passing the appropriate flag when the shader is
generated. The fix in HWUI is to reverse the premultiplication and
interpolation steps.
Test: bit CtsUiRenderingTestCases:.testclasses.ShaderTests
Bug: 34323783
Change-Id: I3417141949f62fcc696b6d8213a4b446d7d0cbf8
This change also fixes an issue with RGBA16F bitmaps when modulated
with a color (for instance by setting an alpha on the Paint object).
The color space conversion is currently done entirely in the shader,
by doing these operations in order:
1. Sample the texture
2. Un-premultiply alpha
3. Apply the EOTF
4. Multiply by the 3x3 color space matrix
5. Apply the OETF
6. Premultiply alpha
Optimizations:
- Steps 2 & 6 are skipped for opaque (common) bitmaps
- Step 3 is skipped when the color space's EOTF is close
to sRGB (Display P3 for instance). Instead, we use
a hardware sRGB fetch (when the GPU supports it)
- When step 3 is necessary, we use one of four standard
EOTF implementations, to save cycles when possible:
+ Linear (doesn't do anything)
+ Full parametric (ICC parametric curve type 4 as defined
in ICC.1:2004-10, section 10.15)
+ Limited parametric (ICC parametric curve type 3)
+ Gamma (ICC parametric curve type 0)
Color space conversion could be done using texture samplers
instead, for instance 3D LUTs, with or without transfer
functions baked in, or 1D LUTs for transfer functions. This
would result in dependent texture fetches which may or may
not be an advantage over an ALU based implementation. The
current solution favor the use of ALUs to save precious
bandwidth.
Test: CtsUiRenderingTests, CtsGraphicsTests
Bug: 32984164
Change-Id: I10bc3db515e13973b45220f129c66b23f0f7f8fe
Frankengradients (linearly interpolated RGB, gamma interpolated alpha) look
fantastic but unfortunately create sligh compatibility issues. For instance,
a gradient from 0xffea1030 to 0x00ea1030 (opaque to alpha, with a single
color) blended on top of 0xff101010 would not look the same as a single
opaque gradient from 0xffea1030 to 0xff101010. The difference is hardly
noticeable on simple gradients but it could cause confusion amongst app
developers. Their life is hard enough as it is, let's be good to them.
My crusade against the gamma world is not over and one day I shall
be the victor. I am patience.
Bug: 35485208
Test: UiRendering.ShaderTests, UiRendering.GradientTests, manual testing
Change-Id: I8204e60cdf0a6b12dfe22638d30ca9622687000e
RGBA16F bitmaps used as an SkBitmapShader would not set the hasLinearTexture
flag in the program description, causing a shader to be generated without
the proper opto-electronic transfer function.
Bug: 35482305
Test: bit -t CtsUiRenderingTestCases:android.uirendering.cts.testclasses.Rgba16fTests
Change-Id: I23354f8189a6b27b677eac9df82677e91282a31d
Alpha pre-multiplication must be done after applying the
opto-electronic transfer function when linear blending is
disabled. The correct way would be to pre-multiply before
gamma encoding but this leads to improper blending which
cannot be corrected without using sRGB frame buffers and
texture sampling.
Bug: 33010587
Test: cts-tradefed run singleCommand cts-dev --module CtsUiRenderingTestCases --test android.uirendering.cts.testclasses.GradientTests
Change-Id: I5f04bda4cb9f63674537aef5931621c14d601884
With linear blending turned off some textures were still
created as sRGB textures instead of linear textures.
Multi-stop gradients were not behaving properly on devices
with no support for float textures.
Gradients are now always interpolated in linear space
even if linear blending is off.
New functions to always force sRGB->linear->sRGB conversions.
Test: Manual testing
Bug: 29940137
Change-Id: Ie2f84ee2a65fd85570e88af813e841e0e625df6c
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
Also corrects some code under development behind the HWUI_NEW_OPS flags
to match the updated Skia API.
Include external/skia/include/private
use SrcConstraint for drawBitmapRect
clean up to allow removal of flags for SCALAR_DIV and IMAGEINFO_FIELDS
don't call DEPRECATED getDevice()
update to newer API for drawBitmapRect
asABitmap is deprecated, used isABitmap
previous-Change-Id: I12208855a95948897077b1c1549eb35416cc801e
previous-Change-Id: I5044f0f61315fe48c60d7af5e261a7d0ed574f56
previous-Change-Id: Ic34a3ba77b3f9e091fa7aaba75018a307abacdab
previous-Change-Id: I79f8dd779920565d1204f7fe67b3286b1bbf4e9b
previous-Change-Id: Ic04d1f8274f6a862ea00f8d241363cf31f5ec1ec
previous-Change-Id: I9e4ae257a1976c74302b6a73f17405174ae58cec
previous-Change-Id: I85de3462ad1e4877784df38edc4bcd0acbd24e5e
Change-Id: Ide8e2f669e91a13c32521af3a16efdaa085c81d0
Include external/skia/include/private
use SrcConstraint for drawBitmapRect
clean up to allow removal of flags for SCALAR_DIV and IMAGEINFO_FIELDS
don't call DEPRECATED getDevice()
update to newer API for drawBitmapRect
asABitmap is deprecated, used isABitmap
Change-Id: I519f54f97321a7a365ea81a3b78cb03b9bdca021
previous-Change-Id: I12208855a95948897077b1c1549eb35416cc801e
previous-Change-Id: I5044f0f61315fe48c60d7af5e261a7d0ed574f56
previous-Change-Id: Ic34a3ba77b3f9e091fa7aaba75018a307abacdab
previous-Change-Id: I79f8dd779920565d1204f7fe67b3286b1bbf4e9b
previous-Change-Id: Ic04d1f8274f6a862ea00f8d241363cf31f5ec1ec
previous-Change-Id: I9e4ae257a1976c74302b6a73f17405174ae58cec
bug:22390304
Fixes two issues:
1) The max texture unit wasn't large enough to handle the most complex
ComposeShader case (1 for draw primitive, 2 for gradient shader, 1 for
bitmap shader).
2) If a shader isn't supported by SkiaShader::store, the shader data
needs to be explicitly disabled, so we won't read uninitilized data
from it when trying to read shader information out.
Change-Id: I29ee7b7c1e07f67db88c1707bdc857daa305e713
bug:20063841
Restores old SkShader matrix behavior from before the Glop refactor.
Many drawing operations draw without sending the canvas transform to
the GL shader. In such cases, we need to adapt the matrix sent to the
SkShader logic to invert the canvas transform that's built into
the mesh.
Change-Id: I42b6f59df36ce46436322b95bf9ad2140795ee58
Adds remaining missing overrides and nullptr usages, missed due to
an extreme failure in tool usage.
Change-Id: I56abd72975a3999ad13330003c348db40f59aebf
Remove Clang cutout for unused parameters. Fix warnings.
Remove Clang cutout for deprecated Skia function usage. Has been
fixed in the L push.
Change-Id: I7ea073ff67127cc1e14e798b655e2c50615fe8e7
Narrow the use of #include directives in hwui, replacing with forward
declarations where straightforward. Speeds compiles; doesn't do any
restructuring of code.
Change-Id: Icac2baffb5896f55d8c6718e9bd9d4bfa02d3ca0
Instead of duplicating internal info about SkShader, inspect the
SkShader installed on the SkPaint.
core/java/android/view/GLES20Canvas.java:
Remove setupModifiers, nResetModifiers, and nSetupShader.
core/jni/android/graphics/Shader.cpp:
Remove calls to create/destroy the (previously) attached SkiaShader.
core/jni/android_view_GLES20Canvas.cpp:
Remove native code for setupShader and resetModifiers.
graphics/java/android/graphics/BitmapShader.java:
graphics/java/android/graphics/ComposeShader.java:
graphics/java/android/graphics/LinearGradient.java:
graphics/java/android/graphics/RadialGradient.java:
graphics/java/android/graphics/Shader.java:
graphics/java/android/graphics/SweepGradient.java:
Remove code keeping track of native SkiaShader.
libs/hwui/Caches.h:
Include Extensions.h.
libs/hwui/DeferredDisplayList.cpp:
Compare shaders on the paint, instead of on DrawModifiers.
libs/hwui/DisplayList.cpp:
libs/hwui/DisplayList.h:
Remove vector of SkiaShaders.
libs/hwui/DisplayListOp.h:
Access the SkShader on mPaint.
Remove SetupShaderOp and ResetShaderOp.
libs/hwui/DisplayListRenderer.cpp:
libs/hwui/DisplayListRenderer.h:
Remove resetShader, setupShader, refShader, and mShaderMap.
libs/hwui/FontRenderer.cpp:
Pass SkShader to setupDrawShader and setupDrawShaderUniforms.
libs/hwui/OpenGLRenderer.cpp:
libs/hwui/OpenGLRenderer.h:
Add LayerShader, a class inheriting from SkShader, to mimic the
behavior of SkiaLayerShader. Unlike SkiaLayerShader, it can be set on
the SkPaint so it can be inspected later.
Set a LayerShader instead of a SkiaLayerShader.
setupDrawShader and setupDrawShaderUniforms now inspect an SkShader
passed in.
Inspect SkShader instead of mDrawModifiers.mShader.
Remove resetShader and setupShader.
setupDrawColorUniforms now takes a boolean indicating whether there is
a shader.
Add an inline function for accessing the SkShader on an SkPaint.
In setupDrawBlending(Layer*, bool), do not check the shader (which will
never be set), but do check whether the color filter may change the
alpha (newly fixed behavior).
In setupDrawBlending(SkPaint, ...), check the SkShader and whether the
color filter affects alpha (the latter is new behavior).
libs/hwui/Renderer.h:
Remove pure virtual functions setupShader and resetShader.
libs/hwui/ResourceCache.cpp:
libs/hwui/ResourceCache.h:
Remove functions for refing/unrefing shaders.
libs/hwui/SkiaShader.cpp:
libs/hwui/SkiaShader.h:
Much of this code was redundant and has been removed.
Convert structs into class with nothing but static functions for
calling describe/setupProgram.
libs/hwui/TextureCache.cpp:
libs/hwui/TextureCache.h:
Use the SkPixelRef as the key to the bitmap Lru cache, since shader
inspection will provide a different SkBitmap pointer (though it will
hold the correct SkPixelRef with the correct generation ID).
tests/CanvasCompare/src/com/android/test/hwuicompare/DisplayModifier.java:
tests/CanvasCompare/src/com/android/test/hwuicompare/ResourceModifiers.java:
Update manual test to have more shaders: radial, sweep, compose,
invalid compose.
BUG:10650594
Change-Id: Iaa7189178bda1c55f96da044d2a9fa602ba36034
Projected RenderNodes are now wrapped with a ClipRect or masked
SaveLayer, so that they are clipped to the outline of the projection
receiver surface.
Change-Id: I1d4afc1bb5d638d650bc0b1dac51a498f216773e
This change sets textures filtering to GL_NEAREST by default. GL_LINEAR
filtering is only used when textures are transformed with a scale or
a rotation. This helps save a couple of fps on some GPUs.
Change-Id: I1efaa452c2c79905f00238e54d886a37203a2ac1
