1) Texture offsets (fImageIncrement) have to be signed depending on
whether the texture is "right way up" (texture upload) or "upside down"
(render target), so the surface normals were coming out upside down.
2) Light normals have to y-negated on upload
These two bugs were cancelling each other out in SampleApp, (where we
were testing w/textures) but not in Chrome (where we were testing
w/render targets).
3) The extract-the-height-from-the-view-matrix hack I was using to
compare light positions vs. gl_FragCoord doesn't work in Chrome where we
compile with GR_STATIC_RECT_VB, and the view matrix contains more than
the viewport transform (to accomodate the canonical vertex buffer).
Fixed by passing the destination render target to
GrGLProgramStage::setData(), so it can flip the light positions in Y on
the CPU.
Review URL: http://codereview.appspot.com/6343109/
git-svn-id: http://skia.googlecode.com/svn/trunk@4605 2bbb7eff-a529-9590-31e7-b0007b416f81
This class tracks the texture that the object uses. A future commit will get rid of the
GrTexture pointer currenty stored in the GrDrawState, allowing us to have CustomStages
*without* textures.
Requires gyp change on next roll.
http://codereview.appspot.com/6306097/
git-svn-id: http://skia.googlecode.com/svn/trunk@4576 2bbb7eff-a529-9590-31e7-b0007b416f81
There are three light types for each: distant, point and spot, whose code
generation lives in a GrGLLight class hierarchy. This similar to the CPU
implementation, where each light type provides a function to compute the vector
from the surface plane to the light (surfaceToLight) and to compute the light
colour (emitLightColour). Instead of templated member functions, as in the CPU
implementation, these are virtual functions to emit the light-specific GLSL
code.
All of the code for the GPU path lives in the same file as that for the CPU
path, SkLightingImageFilter.cpp. In order to provide Ganesh a hook to access
it, SkImageFilter now has a asNewCustomStage() virtual, which allows an image
filter to return a GrCustomStage representing that filter.
Note that this patch does not handle the border conditions correctly (the
[top|bottom][Left|Right]Normal() functions in the CPU implementation). That
will come in a future patch.
Review URL: http://codereview.appspot.com/6345081/
git-svn-id: http://skia.googlecode.com/svn/trunk@4535 2bbb7eff-a529-9590-31e7-b0007b416f81
light, rather than the filter, so computation is localized. Changed a couple
of non-const refs in the image filter creation API const refs. Changed "const
SkColor&" to "SkColor", since it's just a uint32_t anyway.
Review URL: http://codereview.appspot.com/6352071/
git-svn-id: http://skia.googlecode.com/svn/trunk@4466 2bbb7eff-a529-9590-31e7-b0007b416f81
Now, SkComposeShaders, SkDiscreteMapper, SkCosineMapper,
Line2DPathEffect, and SkTransparentShader can be flattened
and unflattened. This makes more of our samples work when
using a pipe.
Review URL: https://codereview.appspot.com/6338064
git-svn-id: http://skia.googlecode.com/svn/trunk@4334 2bbb7eff-a529-9590-31e7-b0007b416f81
the caller instantiates a light (distant, point or spot), and an
SkDiffuseLightingFilter or SkSpecularLightingImageFilter with that light. A
Sobel edge detection filter is applied to the alpha of the incoming bitmap, and
the result is used as a height map for lighting calculations.
Review URL: http://codereview.appspot.com/6302101/
git-svn-id: http://skia.googlecode.com/svn/trunk@4314 2bbb7eff-a529-9590-31e7-b0007b416f81
called (for now at least) TwoPointConical, to distinguish it from the
existing one (TwoPointRadial), and to reflect its odd behavior where it draws
transparent outside the boundary of the "cone" between the two circles.
This impl is unsupported by GPU and Printing at the moment, and Chrome does not
yet invoke it.
Review URL: https://codereview.appspot.com/6299051
git-svn-id: http://skia.googlecode.com/svn/trunk@4197 2bbb7eff-a529-9590-31e7-b0007b416f81
Fix these class of warnings:
- unused functions
- unused locals
- sign mismatch
- missing function prototypes
- missing newline at end of file
- 64 to 32 bit truncation
The changes prefer to link in dead code in the debug build
with 'if (false)' than to comment it out, but trivial cases
are commented out or sometimes deleted if it appears to be
a copy/paste error.
Review URL: https://codereview.appspot.com/6299049
git-svn-id: http://skia.googlecode.com/svn/trunk@4181 2bbb7eff-a529-9590-31e7-b0007b416f81
- suppress PDF for twopointradial GM until we fix its assert/failure (vandebo)
- detect divide-by-zero and abort for now (reed)
git-svn-id: http://skia.googlecode.com/svn/trunk@4165 2bbb7eff-a529-9590-31e7-b0007b416f81
or stroke parameters for a path.
Today, the patheffect only sees if the caller was going to stroke or fill, and
if stroke, it just sees the width. With this change, the effect can see all of the
related parameters (e.g. cap/join/miter). No other change is intended at this
time.
After this change, I hope to use this additional data to allow SkDashPathEffect
to, at times, apply the stroke as part of its effect, which may be much more
efficient than first dashing, and then reading that and stroking it.
Most of these files changed just because of the new parameter to filterPath. The
key changes are in SkPathEffect.[h,cpp], SkPaint.cpp and SkScalerContext.cpp
Review URL: https://codereview.appspot.com/6250051
git-svn-id: http://skia.googlecode.com/svn/trunk@4048 2bbb7eff-a529-9590-31e7-b0007b416f81
or stroke parameters for a path.
Today, the patheffect only sees if the caller was going to stroke or fill, and if
stroke, it just sees the width. With this change, the effect can see all of the
related parameters (e.g. cap/join/miter). No other change is intended at this
time.
After this change, I hope to use this additional data to allow SkDashPathEffect
to, at times, apply the stroke as part of its effect, which may be much more
efficient than first dashing, and then reading that and stroking it.
Most of these files changed just because of the new parameter to filterPath. The
key changes are in SkPathEffect.[h,cpp], SkPaint.cpp and SkScalerContext.cpp
Review URL: https://codereview.appspot.com/6249050
git-svn-id: http://skia.googlecode.com/svn/trunk@4046 2bbb7eff-a529-9590-31e7-b0007b416f81
Implement SkGPipeCanvas::drawBitmap.
Fix a bug where needOpBytes did not necessarily provide as
many bytes as requested.
Fix a bug where needOpBytes would wipe old data without
calling notify.
Register SkEmbossMaskFilter so it can be flattened.
Override drawBitmapNine (currently unimplemented) so an
SkGPipeCanvas does not call its base class' method (which
would fail).
Review URL: https://codereview.appspot.com/6071045
git-svn-id: http://skia.googlecode.com/svn/trunk@3741 2bbb7eff-a529-9590-31e7-b0007b416f81
The current impl for SkFlattenable read/write buffers is
that they extend from SkReader32 and SkWriter32, but that
dependency must be abstract if we are to add any other
serialization format.
Review URL: https://codereview.appspot.com/5999045
git-svn-id: http://skia.googlecode.com/svn/trunk@3654 2bbb7eff-a529-9590-31e7-b0007b416f81
Changes serialization path for MorphologyImageFilter, handling of Windows
HRESULTS; otherwise just tweaks tests.
git-svn-id: http://skia.googlecode.com/svn/trunk@3642 2bbb7eff-a529-9590-31e7-b0007b416f81
The following changes were made by this CL:
1. Make flatten() a protected method as callers should use
SkFlattenableWriteBuffer to flatten an object
2. Make flatten a const method (including subclasses)
3. Mark subclass implementation of flatten with SK_OVERRIDE
4. Ensure overridden flatten impls call their parent
5. Remove no-op implementations of flatten from subclasses
Additionally, if necessary the unflattening constructor was
also moved to the protected section of the subclasses
header if it was not already there.
git-svn-id: http://skia.googlecode.com/svn/trunk@3540 2bbb7eff-a529-9590-31e7-b0007b416f81
us to perform quick-rejects when drawing objects with shadows (esp. text).
WebKit draws shadows w/ a looper (fg and shadow) and a maskfilter on the
shadow layer.
git-svn-id: http://skia.googlecode.com/svn/trunk@3103 2bbb7eff-a529-9590-31e7-b0007b416f81
This patch removes static initializers related to static and
global mutexes from the final library's machine code when
building on a pthread-capable system.
We use PTHREAD_MUTEX_INITIALIZER to perform POD-style
initialization. You need a line like the following to declare
a global mutex with it:
SkBaseMutex gMutex = { PTHREAD_MUTEX_INITIALIZER };
We introduce the SK_DECLARE_STATIC_MUTEX and SK_DECLARE_GLOBAL_MUTEX
macros to be able to declare static/global mutexes in the source tree
uniformly.
SkMutex is now defined as a sub-class of SkBaseMutex, with standard
construction/destruction semantics. This is useful if the mutex
object is a member of another C++ class, or allocated dynamically.
We also modify a few places to refer to SkBaseMutex instead of a
SkMutex, where it makes sense. Generally speaking, client code
should hold and use pointers to SkBaseMutex whenever they can
now.
We defined a new built-time macro named SK_USE_POSIX_THREADS
to indicate that we're using a pthread-based SkThread.h
interface. The macro will also be used in future patches
to implement other helper thread synchronization classes.
Finally, we inline the acquire() and release() functions in the
case of Posix to improve performance a bit.
Running: 'bench -repeat 10 -match mutex' on an Android device or
a 2.4GHz Xeon Linux desktop shows the following improvements:
Before After
Galaxy Nexus 1.64 1.45
Nexus S 1.47 1.16
Xoom 1.86 1.66
Xeon 0.36 0.31
This removes 5 static mutex initializers from the library
Review URL: https://codereview.appspot.com/5501066
git-svn-id: http://skia.googlecode.com/svn/trunk@3091 2bbb7eff-a529-9590-31e7-b0007b416f81
turned off. With a Chromium gyp change to add
SK_SIMPLE_TWOCOLOR_VERTICAL_GRADIENTS, this reduces the rebaselining from
r3073 from ~200 to ~20.
codereview.appspot.com/5558055/
git-svn-id: http://skia.googlecode.com/svn/trunk@3077 2bbb7eff-a529-9590-31e7-b0007b416f81
Consolidate interpolation functions, add new faster more accurate dithering
interpolator.
git-svn-id: http://skia.googlecode.com/svn/trunk@3072 2bbb7eff-a529-9590-31e7-b0007b416f81
of SkImageFilter. If that returns false, then the filter itself is invoked.
git-svn-id: http://skia.googlecode.com/svn/trunk@2977 2bbb7eff-a529-9590-31e7-b0007b416f81
matrix test slide. This was basically implemented in the same places where
the blending-based color filter was being done. The shader simply does a mat4
matrix multiply and a vec4 add.
Review URL: http://codereview.appspot.com/5489107/
git-svn-id: http://skia.googlecode.com/svn/trunk@2948 2bbb7eff-a529-9590-31e7-b0007b416f81
M include/effects/SkAvoidXfermode.h
M include/effects/SkDiscretePathEffect.h
M include/effects/Sk1DPathEffect.h
M include/effects/Sk2DPathEffect.h
M include/effects/SkBlurDrawLooper.h
M include/effects/SkPixelXorXfermode.h
M include/effects/SkDashPathEffect.h
M include/effects/SkColorMatrixFilter.h
M include/effects/SkEmbossMaskFilter.h
M include/effects/SkLayerDrawLooper.h
M include/effects/SkGroupShape.h
M include/effects/SkBlurImageFilter.h
M include/effects/SkRectShape.h
A include/effects/SkEffects.h
M include/effects/SkCornerPathEffect.h
M include/effects/SkGradientShader.h
M include/effects/SkBlurMaskFilter.h
M include/effects/SkLayerRasterizer.h
M include/core/SkMallocPixelRef.h
M include/core/SkFlattenable.h
M include/core/SkShape.h
M include/core/SkPixelRef.h
M include/core/SkGraphics.h
M include/core/SkPathEffect.h
M include/core/SkPostConfig.h
M include/core/SkXfermode.h
M include/core/SkColorFilter.h
M include/images/SkFlipPixelRef.h
M include/images/SkImageRef_GlobalPool.h
M src/effects/SkDashPathEffect.cpp
M src/effects/SkColorMatrixFilter.cpp
M src/effects/SkBlurImageFilter.cpp
M src/effects/SkGroupShape.cpp
M src/effects/SkCornerPathEffect.cpp
M src/effects/SkGradientShader.cpp
M src/effects/SkBlurMaskFilter.cpp
M src/effects/SkAvoidXfermode.cpp
M src/effects/Sk2DPathEffect.cpp
M src/effects/SkBlurDrawLooper.cpp
M src/effects/SkPixelXorXfermode.cpp
M src/effects/SkColorFilters.cpp
M src/effects/SkLayerDrawLooper.cpp
M src/effects/SkRectShape.cpp
A src/effects/SkEffects.cpp
M src/effects/SkLayerRasterizer.cpp
M src/effects/SkDiscretePathEffect.cpp
M src/effects/Sk1DPathEffect.cpp
A src/effects/SkEffects_none.cpp
M src/core/SkPixelRef.cpp
M src/core/SkGraphics.cpp
M src/core/SkFlattenable.cpp
M src/core/SkBitmapProcShader.h
M src/core/SkPathEffect.cpp
M src/core/SkShape.cpp
M src/core/SkXfermode.cpp
M src/core/SkMallocPixelRef.cpp
M src/core/SkBitmapProcShader.cpp
M src/images/SkFlipPixelRef.cpp
M src/images/SkImageRef_GlobalPool.cpp
A src/ports/SkGlobalInitialization_chromium.cpp
M src/ports/SkImageRef_ashmem.h
M src/ports/SkImageRef_ashmem.cpp
A src/ports/SkGlobalInitialization_default.cpp
M gyp/effects.gyp
M gyp/tools.gyp
M gyp/ports.gyp
git-svn-id: http://skia.googlecode.com/svn/trunk@2876 2bbb7eff-a529-9590-31e7-b0007b416f81
Specialize loops in apply_kernel and apply_kernel_interp,
improving blur benchmark performance by 25%.
git-svn-id: http://skia.googlecode.com/svn/trunk@2731 2bbb7eff-a529-9590-31e7-b0007b416f81
The imageblur test also needed to stop calling getDeviceSize(), since it was
returning 99999x99999 for the SkPicture case.
git-svn-id: http://skia.googlecode.com/svn/trunk@2699 2bbb7eff-a529-9590-31e7-b0007b416f81
SkBlurImageFilter, sets it on an SkPaint, passes that paint to saveLayer(),
draws the primitives which are to be blurred, then calls restore(), which
applies the blur. The blurs have separate sizes in the horizontal and vertical
direction. This feature is GPU-only for now.
NB: Due to the clipping change, there are slight pixel differences on the
blurs_gpu and shadows_gpu tests, so those will require rebaselining on all
platforms, as will some of the WebKit layout tests (TBD).
Review URL: http://codereview.appspot.com/5322068/
git-svn-id: http://skia.googlecode.com/svn/trunk@2643 2bbb7eff-a529-9590-31e7-b0007b416f81
we can, because an entire row is conservatively either outside or inside the
gradient.) Change the gradient benchmark to capture both cases, and add new
gm to detect errors in these optimized paths.
git-svn-id: http://skia.googlecode.com/svn/trunk@2327 2bbb7eff-a529-9590-31e7-b0007b416f81
I have manually examined all of these diffs and restored a few files that
seem to require manual adjustment.
The following files still need to be modified manually, in a separate CL:
android_sample/SampleApp/AndroidManifest.xml
android_sample/SampleApp/res/layout/layout.xml
android_sample/SampleApp/res/menu/sample.xml
android_sample/SampleApp/res/values/strings.xml
android_sample/SampleApp/src/com/skia/sampleapp/SampleApp.java
android_sample/SampleApp/src/com/skia/sampleapp/SampleView.java
experimental/CiCarbonSampleMain.c
experimental/CocoaDebugger/main.m
experimental/FileReaderApp/main.m
experimental/SimpleCocoaApp/main.m
experimental/iOSSampleApp/Shared/SkAlertPrompt.h
experimental/iOSSampleApp/Shared/SkAlertPrompt.m
experimental/iOSSampleApp/SkiOSSampleApp-Base.xcconfig
experimental/iOSSampleApp/SkiOSSampleApp-Debug.xcconfig
experimental/iOSSampleApp/SkiOSSampleApp-Release.xcconfig
gpu/src/android/GrGLDefaultInterface_android.cpp
gyp/common.gypi
gyp_skia
include/ports/SkHarfBuzzFont.h
include/views/SkOSWindow_wxwidgets.h
make.bat
make.py
src/opts/memset.arm.S
src/opts/memset16_neon.S
src/opts/memset32_neon.S
src/opts/opts_check_arm.cpp
src/ports/SkDebug_brew.cpp
src/ports/SkMemory_brew.cpp
src/ports/SkOSFile_brew.cpp
src/ports/SkXMLParser_empty.cpp
src/utils/ios/SkImageDecoder_iOS.mm
src/utils/ios/SkOSFile_iOS.mm
src/utils/ios/SkStream_NSData.mm
tests/FillPathTest.cpp
Review URL: http://codereview.appspot.com/4816058
git-svn-id: http://skia.googlecode.com/svn/trunk@1982 2bbb7eff-a529-9590-31e7-b0007b416f81
We could speed-up again if we...
- respected kDither and only built 1/2 of the table for non-dither requests
- output simple params to the gpu rather than always a texture
- detected that we have no alpha, and then can skip premul per-entry
git-svn-id: http://skia.googlecode.com/svn/trunk@1772 2bbb7eff-a529-9590-31e7-b0007b416f81
SkGPipe needs this, since it cannot (unlike SkPicture) see all of the factories
before it hands its data to the reader.
In this mode, the writer embedds the factory name the first time it sees it,
and then after that writes an index (referencing the fFactorySet). The reader
installs an empty array, and as it encounters names, appends them to that array
so that subsequent indices can be used to retrieve the previously named factory.
Some of the existing patheffects did not register their factory names, so those
changes are also part of this CL. Annoyingly, to register your factory using the
current scheme, it has to be in the public section of the class definition.
git-svn-id: http://skia.googlecode.com/svn/trunk@1663 2bbb7eff-a529-9590-31e7-b0007b416f81
calculation for Sk2DPathEffect currently evaluates all the bytes in the
embedded SkMatrix. This includes the type mask, which contains some
uninitialized padding. Changing it to use SkMatrix::flatten() and
SkMatrix::unflatten() (as SkGroupShape was doing) avoids the uninitialized
data errors.
Review URL: http://codereview.appspot.com/4529074/
git-svn-id: http://skia.googlecode.com/svn/trunk@1395 2bbb7eff-a529-9590-31e7-b0007b416f81
Both shader and gradient_shader write matrices to the flatten stream. However, they were
just calling write(&matrix, sizeof(SkMatrix)) and the matrix can contain lazily-computed
function ptrs as part of its internal cache. Thus two matrices that are logically the
same may write different bytes.
This is a problem because picture relies on flattening objects and then using the
flatten stream as a key into its cache. This matrix-write bug effectively kills the
effectiveness of the cache for shaders.
The fix is to write proper read/write functions for matrix (and region btw). These
call through to the existing low-level flatten routines (which just write into a
memory ptr).
git-svn-id: http://skia.googlecode.com/svn/trunk@1290 2bbb7eff-a529-9590-31e7-b0007b416f81
note: gradient caller doesn't so we can still draw wrong when the caller
converts its initial fx from float->fixed. Perhaps SkClampRange should offer
a float interface as well.
git-svn-id: http://skia.googlecode.com/svn/trunk@1149 2bbb7eff-a529-9590-31e7-b0007b416f81
promise that const SkPaint& stay const (so we don't have bugs if a const paint
is referenced from two threads in pictures)
git-svn-id: http://skia.googlecode.com/svn/trunk@1074 2bbb7eff-a529-9590-31e7-b0007b416f81
for another iter that runs back-to-front (B2FIter).
add unittest for Deque
git-svn-id: http://skia.googlecode.com/svn/trunk@821 2bbb7eff-a529-9590-31e7-b0007b416f81
Fixed rendering of shadows under bitmaps.
Added an option to SkBlurMask to produce higher quality blurs.
The HQ blur option is propagated through SkBlurDrawLooper so that it can be invoked by WebKit for drawing shadows.
Added a shadow rendering test to gm.
Bugs fixed: 146, 150
Related Chromium bugs: 11153, 42654
Code review: http://codereview.appspot.com/4174049
git-svn-id: http://skia.googlecode.com/svn/trunk@818 2bbb7eff-a529-9590-31e7-b0007b416f81
SkSafeRef() and SkSafeUnref().
This is basically a bug waiting to happen. An optimizing compiler can remove
checks for null on "this" if it chooses. However, SkRefCnt::safeRef() relies on
precisely this check...
void SkRefCnt::safeRef() {
if (this) {
this->ref();
}
}
Since a compiler might skip the if-clause, it breaks the intention of this
method, hence its removal.
static inline void SkSafeRef(SkRefCnt* obj) {
if (obj) {
obj->ref();
}
}
This form is not ignored by an optimizing compile, so we use it instead.
git-svn-id: http://skia.googlecode.com/svn/trunk@762 2bbb7eff-a529-9590-31e7-b0007b416f81
The regression was due to the fact that we were calling PlatformColorProc() for
every span (which in turns makes CPUID, a fairly expensive call). Since we draw
a lot of rects, and rects have 1-pixel wide spans for the vertical segments,
that's a lot of CPUID.
Fixed by cacheing the result of PlatformColorProc(), as is done for the other
platform-specific blitters.
Review URL: http://codereview.appspot.com/3669042/
git-svn-id: http://skia.googlecode.com/svn/trunk@636 2bbb7eff-a529-9590-31e7-b0007b416f81
obvious artifacts. We need a more comprehensive fix for all gradients (32bit
and 16bit) when we have a large number of colors (or they are spaced out very
non-linearly), to avoid banding.
git-svn-id: http://skia.googlecode.com/svn/trunk@458 2bbb7eff-a529-9590-31e7-b0007b416f81
call these new procs in (nearly) all the places we had inlined loops before.
In once instance (blitter_argb32::blitAntiH) we get different results by a
tiny bit. The new code is more accurate, and exactly inline with all of the
other like-minded blits, so I think the change is good going forward.
git-svn-id: http://skia.googlecode.com/svn/trunk@366 2bbb7eff-a529-9590-31e7-b0007b416f81
effectively chopping off the top row of pixels when we computed a blur.
git-svn-id: http://skia.googlecode.com/svn/trunk@358 2bbb7eff-a529-9590-31e7-b0007b416f81
(like gradients) to support predithering. If they do, then they would suppress
kConstInY16, since they no longer are const.
The blitters now check for each flag separately, so we don't have to give up
const-in-Y in the 32bit case, since in that mode we don't care about dithering.
git-svn-id: http://skia.googlecode.com/svn/trunk@339 2bbb7eff-a529-9590-31e7-b0007b416f81
has been computed, othewise we can't distinguish 0x00000000 from 0x00FF0000
Add fast case for index blit where we read 4 src pixels at a time
git-svn-id: http://skia.googlecode.com/svn/trunk@248 2bbb7eff-a529-9590-31e7-b0007b416f81
define constant for max value flatten/unflatten can return (so clients can put
the buffer on the stack)
git-svn-id: http://skia.googlecode.com/svn/trunk@243 2bbb7eff-a529-9590-31e7-b0007b416f81
this signals blitters that the shader will return the same results for a given
x value, independent of y. Useful inside blitRect(), where it can cache the
first call to shadeSpan() and reuse it on all subsequent scans. Works with
(non-rotated) linear-gradients, and Nx1 bitmaps.
git-svn-id: http://skia.googlecode.com/svn/trunk@214 2bbb7eff-a529-9590-31e7-b0007b416f81
add SkShape baseclass, in the hopes of having SkPicture inherit from that, and
also using shapes as the extension mechanism for things like animated-gif
git-svn-id: http://skia.googlecode.com/svn/trunk@174 2bbb7eff-a529-9590-31e7-b0007b416f81
Should not hurt features (we hope), but is an easy way to survive malicious
values on a small-memory machine like a handset.
git-svn-id: http://skia.googlecode.com/svn/trunk@123 2bbb7eff-a529-9590-31e7-b0007b416f81
