reed@google.com [Mon, 30 Jan 2012 17:09:45 +0000 (17:09 +0000)]
inject a 32767 bounded cliprect before using SuperSampler blitter, to avoid
crash/assert when our run-array is larger than int16_t. Better fix may be to
"tile" the drawing, so we never see a clip that's too wide, and perhaps this
technique can help us avoid disabling AA for large parths (not sure tho).
git-svn-id: http://skia.googlecode.com/svn/trunk@3104
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Mon, 30 Jan 2012 15:41:43 +0000 (15:41 +0000)]
extend fastbounds impls to include maskfilters and drawloopers. This allows
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
junov@chromium.org [Mon, 30 Jan 2012 14:53:22 +0000 (14:53 +0000)]
Adding a flush method to SkCanvas
BUG=http://code.google.com/p/skia/issues/detail?id=467
REVIEW=http://codereview.appspot.com/5600044/
TEST=gm uses the new code path on all its test cases
git-svn-id: http://skia.googlecode.com/svn/trunk@3102
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Mon, 30 Jan 2012 14:28:39 +0000 (14:28 +0000)]
Reland r3078 (original failures that led to revert were problems with the bot)
git-svn-id: http://skia.googlecode.com/svn/trunk@3101
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Mon, 30 Jan 2012 14:07:26 +0000 (14:07 +0000)]
fix compilation errors when VISUALIZE_COMPLEX_CLIP is set (debug assist code path)
Review URL: http://codereview.appspot.com/5596043/
git-svn-id: http://skia.googlecode.com/svn/trunk@3100
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Mon, 30 Jan 2012 14:02:44 +0000 (14:02 +0000)]
code cleanup for single_pass_path() and remove enum GrConvexHint
Review URL: code cleanup for single_pass_path() and remove enum GrConvexHint
git-svn-id: http://skia.googlecode.com/svn/trunk@3099
2bbb7eff-a529-9590-31e7-
b0007b416f81
vandebo@chromium.org [Sat, 28 Jan 2012 01:45:11 +0000 (01:45 +0000)]
Formatting cleanup: remove extra whitespace
Change by arthurhsu@chromium.org, original CL: http://codereview.appspot.com/5587049/
Review URL: https://codereview.appspot.com/5594043
git-svn-id: http://skia.googlecode.com/svn/trunk@3098
2bbb7eff-a529-9590-31e7-
b0007b416f81
twiz@google.com [Fri, 27 Jan 2012 19:12:00 +0000 (19:12 +0000)]
Expose functionality to purge the GPU texture cache while running SampleApp.
Press 'p' to purge the cache. A summary of the number of bytes of video memory
released is output to the command prompt.
Review URL: https://codereview.appspot.com/5587045
git-svn-id: http://skia.googlecode.com/svn/trunk@3097
2bbb7eff-a529-9590-31e7-
b0007b416f81
caryclark@google.com [Fri, 27 Jan 2012 16:52:38 +0000 (16:52 +0000)]
add formatting to suppress compiler warning
Review URL: https://codereview.appspot.com/5586043
git-svn-id: http://skia.googlecode.com/svn/trunk@3095
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Fri, 27 Jan 2012 00:47:24 +0000 (00:47 +0000)]
Rip out GrPlatformSurface (has been deprecated for some time, use GrPlatformTexture or GrPlatformRenderTarget instead)
Review URL: http://codereview.appspot.com/5576052/
git-svn-id: http://skia.googlecode.com/svn/trunk@3094
2bbb7eff-a529-9590-31e7-
b0007b416f81
digit@google.com [Thu, 26 Jan 2012 21:26:40 +0000 (21:26 +0000)]
posix: Avoid static initializers in static/global mutexes
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
bsalomon@google.com [Thu, 26 Jan 2012 20:33:19 +0000 (20:33 +0000)]
Change background color and add paths to convexpaths gm
git-svn-id: http://skia.googlecode.com/svn/trunk@3090
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Thu, 26 Jan 2012 04:13:31 +0000 (04:13 +0000)]
need to include SkColorFilter.h since we reference some methods on it
in our inline functions.
git-svn-id: http://skia.googlecode.com/svn/trunk@3089
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Wed, 25 Jan 2012 21:53:53 +0000 (21:53 +0000)]
Make SkRegion::operator== a member function, rather than a privately-declared
friend. Without this, calling code has access to operator==, but can fail to
link because the implementation is assumed to have static linkage.
http://codereview.appspot.com/5577047/
git-svn-id: http://skia.googlecode.com/svn/trunk@3088
2bbb7eff-a529-9590-31e7-
b0007b416f81
caryclark@google.com [Wed, 25 Jan 2012 18:57:23 +0000 (18:57 +0000)]
Intersection work in progress
Review URL: https://codereview.appspot.com/5576043
git-svn-id: http://skia.googlecode.com/svn/trunk@3087
2bbb7eff-a529-9590-31e7-
b0007b416f81
caryclark@google.com [Wed, 25 Jan 2012 16:47:04 +0000 (16:47 +0000)]
fix vertical bold text on mac
M src/ports/SkFontHost_mac_coretext.cpp
git-svn-id: http://skia.googlecode.com/svn/trunk@3086
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Wed, 25 Jan 2012 16:04:18 +0000 (16:04 +0000)]
disable SPEW_PURGE_STATUS
git-svn-id: http://skia.googlecode.com/svn/trunk@3085
2bbb7eff-a529-9590-31e7-
b0007b416f81
epoger@google.com [Tue, 24 Jan 2012 04:27:11 +0000 (04:27 +0000)]
whitespace change to trigger Skia buildbot
git-svn-id: http://skia.googlecode.com/svn/trunk@3082
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Tue, 24 Jan 2012 02:31:42 +0000 (02:31 +0000)]
Prefer a texture pixel ref to a render target pixel ref when creating an SkGpuDevice
Review URL: http://codereview.appspot.com/5570050/
git-svn-id: http://skia.googlecode.com/svn/trunk@3081
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Tue, 24 Jan 2012 02:23:07 +0000 (02:23 +0000)]
Whitespace commit to check if lion / win bots are hosed.
git-svn-id: http://skia.googlecode.com/svn/trunk@3080
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Sat, 21 Jan 2012 15:03:39 +0000 (15:03 +0000)]
revert 3078
git-svn-id: http://skia.googlecode.com/svn/trunk@3079
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Sat, 21 Jan 2012 14:48:36 +0000 (14:48 +0000)]
fix interior issues in convex path renderer
git-svn-id: http://skia.googlecode.com/svn/trunk@3078
2bbb7eff-a529-9590-31e7-
b0007b416f81
tomhudson@google.com [Fri, 20 Jan 2012 13:59:14 +0000 (13:59 +0000)]
Temporarily allow linear interpolation of 2-color vertical gradients to be
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
senorblanco@chromium.org [Thu, 19 Jan 2012 19:59:22 +0000 (19:59 +0000)]
Fix a warning in libtess in Release builds; define NDEBUG in Release on all
platforms.
BUG: http://code.google.com/p/skia/issues/detail?id=456
Review URL: http://codereview.appspot.com/5555054/
git-svn-id: http://skia.googlecode.com/svn/trunk@3076
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Thu, 19 Jan 2012 16:16:52 +0000 (16:16 +0000)]
Add flag to disable gl error checking when allocating gpu objects
Review URL: http://codereview.appspot.com/5558052/
git-svn-id: http://skia.googlecode.com/svn/trunk@3075
2bbb7eff-a529-9590-31e7-
b0007b416f81
schenney@chromium.org [Thu, 19 Jan 2012 15:31:01 +0000 (15:31 +0000)]
Cleanup of SkPath code to remove SK_OLD_EMPTY_PATH_BEHAVIOR and other
dead code. The patch also renames kAfterClose_SegmentState for the
SkPath::Iter class, to kEmptyContour_SegmentState which more
accurately reflects the meaning of the state. The Iter can be in that
state initially, before anything has been seen, let alone a close.
BUG=445
TEST=Covered by existing tests.
Review URL: https://codereview.appspot.com/5532091
git-svn-id: http://skia.googlecode.com/svn/trunk@3074
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Thu, 19 Jan 2012 14:23:24 +0000 (14:23 +0000)]
rebaseline with fix to drawAsAHairline
git-svn-id: http://skia.googlecode.com/svn/trunk@3073
2bbb7eff-a529-9590-31e7-
b0007b416f81
tomhudson@google.com [Wed, 18 Jan 2012 21:28:01 +0000 (21:28 +0000)]
Interpolate vertical linear gradients for improved quality.
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
bsalomon@google.com [Wed, 18 Jan 2012 21:00:54 +0000 (21:00 +0000)]
Rebaseline hairmodes GM at r3070
git-svn-id: http://skia.googlecode.com/svn/trunk@3071
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Wed, 18 Jan 2012 20:34:00 +0000 (20:34 +0000)]
Pull xfer mode test out of generic draw-as-hairline test. Use coverage rather than alpha to draw hairlines < 1pix wide in GPU.
Review URL: http://codereview.appspot.com/5528112/
git-svn-id: http://skia.googlecode.com/svn/trunk@3070
2bbb7eff-a529-9590-31e7-
b0007b416f81
epoger@google.com [Wed, 18 Jan 2012 20:29:56 +0000 (20:29 +0000)]
Fix mimetypes for all checked-in GM baselines (no image content changes)
git-svn-id: http://skia.googlecode.com/svn/trunk@3069
2bbb7eff-a529-9590-31e7-
b0007b416f81
epoger@google.com [Wed, 18 Jan 2012 20:11:13 +0000 (20:11 +0000)]
Stop writing out XPS baselines until we resolve skia issue 455
BUG=455
TBR=bungeman
Review URL: https://codereview.appspot.com/5554052
git-svn-id: http://skia.googlecode.com/svn/trunk@3068
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Wed, 18 Jan 2012 19:51:55 +0000 (19:51 +0000)]
Turn clipping back on in OSAA pass 1. Skip default cons on GrDrawState when saving off a GrDrawTarget's state.
Review URL: http://codereview.appspot.com/5553051/
git-svn-id: http://skia.googlecode.com/svn/trunk@3067
2bbb7eff-a529-9590-31e7-
b0007b416f81
epoger@google.com [Wed, 18 Jan 2012 19:20:39 +0000 (19:20 +0000)]
Remove apparent SkIntToScalar(SkScalar) call in SkScalerContext_FreeType
This change seems to fix the following assertion failure we have been seeing
in the Skia_Linux_Fixed_Debug build since http://code.google.com/p/skia/source/detail?r=3059 :
drawing... bitmapfilters [540 330]
../include/core/SkFixed.h:67: failed assertion "n >= -32768 && n <= 32767"
Review URL: https://codereview.appspot.com/5557049
git-svn-id: http://skia.googlecode.com/svn/trunk@3066
2bbb7eff-a529-9590-31e7-
b0007b416f81
junov@google.com [Wed, 18 Jan 2012 18:56:34 +0000 (18:56 +0000)]
Updating gm's usage message
git-svn-id: http://skia.googlecode.com/svn/trunk@3065
2bbb7eff-a529-9590-31e7-
b0007b416f81
schenney@chromium.org [Wed, 18 Jan 2012 18:06:10 +0000 (18:06 +0000)]
Somehow this change was missing from the previous commit.
git-svn-id: http://skia.googlecode.com/svn/trunk@3063
2bbb7eff-a529-9590-31e7-
b0007b416f81
schenney@chromium.org [Wed, 18 Jan 2012 18:02:10 +0000 (18:02 +0000)]
Fixing the behavior of SkPathMeasure to reflect changes in SkPath::Iter.
This implementation modifies SkPath::Iter extensively to avoid copying
the points when used to measure path length.
BUG=446
TEST=tests/PathMeasureTest.cpp
Review URL: https://codereview.appspot.com/5533074
git-svn-id: http://skia.googlecode.com/svn/trunk@3062
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Wed, 18 Jan 2012 17:06:35 +0000 (17:06 +0000)]
add GenA8FromLCD as a hack to force GDI to create the A8 mask from the LCD
results, rather than asking GDI directly for A8 (which it sometimes decides
to interpret as BW)
git-svn-id: http://skia.googlecode.com/svn/trunk@3061
2bbb7eff-a529-9590-31e7-
b0007b416f81
epoger@google.com [Wed, 18 Jan 2012 17:02:15 +0000 (17:02 +0000)]
Landing change from codereview.appspot.com/5540063/
git-svn-id: http://skia.googlecode.com/svn/trunk@3060
2bbb7eff-a529-9590-31e7-
b0007b416f81
junov@google.com [Wed, 18 Jan 2012 16:21:08 +0000 (16:21 +0000)]
Adding class SkDeferredCanvas for deferred rendering.
TEST=added a new pass to gm, so all gm tests are run through SkDeferredCanvas
REVIEW=http://codereview.appspot.com/5430058/
git-svn-id: http://skia.googlecode.com/svn/trunk@3059
2bbb7eff-a529-9590-31e7-
b0007b416f81
epoger@google.com [Tue, 17 Jan 2012 21:26:05 +0000 (21:26 +0000)]
Change download-baselines script to download images from skia-autogen SVN repo
I have copied the old version (which downloads the images from the buildbots
directly, but only works with our Mac buildbots) to download-baselines-old,
so we can use either version during a transition period.
Another difference: the new version sets the mimetype property of all image
files in the baseline_subdir, even those that have not changed.
BUG=386
http://code.google.com/p/skia/issues/detail?id=386 ('make buildbots write out RunGM image results to a browsable directory')
Review URL: https://codereview.appspot.com/5544056
git-svn-id: http://skia.googlecode.com/svn/trunk@3058
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Tue, 17 Jan 2012 21:03:05 +0000 (21:03 +0000)]
Attempt to reland 3054
git-svn-id: http://skia.googlecode.com/svn/trunk@3056
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Tue, 17 Jan 2012 20:14:58 +0000 (20:14 +0000)]
Revert r3054 for failing tests
git-svn-id: http://skia.googlecode.com/svn/trunk@3055
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Tue, 17 Jan 2012 19:58:34 +0000 (19:58 +0000)]
Add constant coverage to GrDrawState
Review URL: http://codereview.appspot.com/5543052/
git-svn-id: http://skia.googlecode.com/svn/trunk@3054
2bbb7eff-a529-9590-31e7-
b0007b416f81
epoger@google.com [Tue, 17 Jan 2012 19:20:02 +0000 (19:20 +0000)]
whitespace change to trigger Skia buildbot
git-svn-id: http://skia.googlecode.com/svn/trunk@3053
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Tue, 17 Jan 2012 18:26:38 +0000 (18:26 +0000)]
if the y-max has neighbors (prev and next) with the same Y values, then the
cross will be 0, but we can still know the direction by looking at the change
in X
Fixes the hypen in #7 of this bug http://code.google.com/p/chromium/issues/detail?id=109370
git-svn-id: http://skia.googlecode.com/svn/trunk@3049
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Tue, 17 Jan 2012 18:23:36 +0000 (18:23 +0000)]
rebaseline following change to subpixel positioning
git-svn-id: http://skia.googlecode.com/svn/trunk@3048
2bbb7eff-a529-9590-31e7-
b0007b416f81
epoger@google.com [Tue, 17 Jan 2012 17:23:22 +0000 (17:23 +0000)]
whitespace change to tri
gger Skia buildbot
git-svn-id: http://skia.googlecode.com/svn/trunk@3047
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Tue, 17 Jan 2012 16:28:34 +0000 (16:28 +0000)]
Incorporate changes related to Tom's post-hoc comments on the convex path renderer.
git-svn-id: http://skia.googlecode.com/svn/trunk@3045
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Tue, 17 Jan 2012 16:16:49 +0000 (16:16 +0000)]
fix assert
git-svn-id: http://skia.googlecode.com/svn/trunk@3044
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Tue, 17 Jan 2012 16:01:37 +0000 (16:01 +0000)]
Add GrDrawState reset
Review URL: http://codereview.appspot.com/5543059/
git-svn-id: http://skia.googlecode.com/svn/trunk@3043
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Tue, 17 Jan 2012 14:39:21 +0000 (14:39 +0000)]
second try to fix assert
git-svn-id: http://skia.googlecode.com/svn/trunk@3042
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Tue, 17 Jan 2012 14:31:23 +0000 (14:31 +0000)]
Change static asserts of scalar type to runtime asserts (re r3040)
git-svn-id: http://skia.googlecode.com/svn/trunk@3041
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Tue, 17 Jan 2012 14:25:10 +0000 (14:25 +0000)]
Add convex path renderer (disabled)
Review URL: http://codereview.appspot.com/5533061/
git-svn-id: http://skia.googlecode.com/svn/trunk@3040
2bbb7eff-a529-9590-31e7-
b0007b416f81
skyostil@google.com [Mon, 16 Jan 2012 10:45:53 +0000 (10:45 +0000)]
Modify SkBitmap::extractSubset() to respect volatility
The resulting subset bitmap will always have the same volatility flag as the
source bitmap.
BUG=452
Review URL: http://codereview.appspot.com/5544052
git-svn-id: http://skia.googlecode.com/svn/trunk@3039
2bbb7eff-a529-9590-31e7-
b0007b416f81
bungeman@google.com [Fri, 13 Jan 2012 16:07:11 +0000 (16:07 +0000)]
Rebaseline for 3037.
git-svn-id: http://skia.googlecode.com/svn/trunk@3038
2bbb7eff-a529-9590-31e7-
b0007b416f81
bungeman@google.com [Fri, 13 Jan 2012 15:02:58 +0000 (15:02 +0000)]
Subpixel text 3/8 of a pixel too far to the right.
http://codereview.appspot.com/5502097/
git-svn-id: http://skia.googlecode.com/svn/trunk@3037
2bbb7eff-a529-9590-31e7-
b0007b416f81
skyostil@google.com [Fri, 13 Jan 2012 14:56:51 +0000 (14:56 +0000)]
Modify SkBitmap::extractSubset() to respect opaqueness
The resulting subset bitmap will always have the same opaqueness flag as the
source bitmap.
BUG=439
Review URL: http://codereview.appspot.com/5534051
git-svn-id: http://skia.googlecode.com/svn/trunk@3036
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Thu, 12 Jan 2012 20:56:52 +0000 (20:56 +0000)]
Add gyp variable to disable OSAA
Review URL: http://codereview.appspot.com/5539047
git-svn-id: http://skia.googlecode.com/svn/trunk@3033
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Thu, 12 Jan 2012 20:38:57 +0000 (20:38 +0000)]
Fix vs2010 int to scalar warnings
git-svn-id: http://skia.googlecode.com/svn/trunk@3032
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Thu, 12 Jan 2012 20:19:56 +0000 (20:19 +0000)]
Add code path to SW rasterize paths and upload as a mask texture
Review URL: http://codereview.appspot.com/5542043/
git-svn-id: http://skia.googlecode.com/svn/trunk@3031
2bbb7eff-a529-9590-31e7-
b0007b416f81
tomhudson@google.com [Thu, 12 Jan 2012 20:13:49 +0000 (20:13 +0000)]
Create floating-point square root path for 16b mirrored radial gradients
(6x-16x speedup). Create specialized dy=0 path for clamped radial gradients
(25% speedup). Make 16b radial gradient proc signature similar to 32b,
passing in SkScalar instead of SkFixed and/or SkFloat.
git-svn-id: http://skia.googlecode.com/svn/trunk@3030
2bbb7eff-a529-9590-31e7-
b0007b416f81
schenney@chromium.org [Thu, 12 Jan 2012 20:04:06 +0000 (20:04 +0000)]
Adding new SkPathMeasure tests, defined out for now as they fail.
Review URL: http://codereview.appspot.com/5529078
git-svn-id: http://skia.googlecode.com/svn/trunk@3029
2bbb7eff-a529-9590-31e7-
b0007b416f81
senorblanco@chromium.org [Thu, 12 Jan 2012 19:10:35 +0000 (19:10 +0000)]
Add a wide stroked line path benchmark to bench. It uses the same stroke width
and endcap style as the stroked curved path.
Review URL: http://codereview.appspot.com/5529069/
git-svn-id: http://skia.googlecode.com/svn/trunk@3028
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Thu, 12 Jan 2012 18:17:11 +0000 (18:17 +0000)]
automatically inject a moveTo if we see a close followed by a line/quad/cubic
git-svn-id: http://skia.googlecode.com/svn/trunk@3027
2bbb7eff-a529-9590-31e7-
b0007b416f81
tomhudson@google.com [Thu, 12 Jan 2012 16:21:43 +0000 (16:21 +0000)]
Convert 32b radial gradient procs to take SkScalar instead of
SkFixed AND SkFloat.
http://codereview.appspot.com/5532083/
git-svn-id: http://skia.googlecode.com/svn/trunk@3026
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Thu, 12 Jan 2012 15:21:16 +0000 (15:21 +0000)]
remove unused bool param to SkMutex constructor
git-svn-id: http://skia.googlecode.com/svn/trunk@3025
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Thu, 12 Jan 2012 13:47:37 +0000 (13:47 +0000)]
Remove supports per-vertex coverage caps bit
Review URL: http://codereview.appspot.com/5529071/
git-svn-id: http://skia.googlecode.com/svn/trunk@3024
2bbb7eff-a529-9590-31e7-
b0007b416f81
tomhudson@google.com [Wed, 11 Jan 2012 21:54:26 +0000 (21:54 +0000)]
Remove obsolete SK_USE_FLOAT_SQRT
http://codereview.appspot.com/5535053/
git-svn-id: http://skia.googlecode.com/svn/trunk@3023
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Wed, 11 Jan 2012 21:04:50 +0000 (21:04 +0000)]
update baseline
git-svn-id: http://skia.googlecode.com/svn/trunk@3022
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Wed, 11 Jan 2012 21:03:05 +0000 (21:03 +0000)]
reenable direction test, fix handling of degenerate segments in the non-convex case
git-svn-id: http://skia.googlecode.com/svn/trunk@3021
2bbb7eff-a529-9590-31e7-
b0007b416f81
epoger@google.com [Wed, 11 Jan 2012 20:43:29 +0000 (20:43 +0000)]
Fix SkScalar values in ClipperTest.cpp; these were breaking one Windows buildbot
TBR=reed
Review URL: http://codereview.appspot.com/5535052
git-svn-id: http://skia.googlecode.com/svn/trunk@3020
2bbb7eff-a529-9590-31e7-
b0007b416f81
tomhudson@google.com [Wed, 11 Jan 2012 19:29:08 +0000 (19:29 +0000)]
Fix the string we're checking for in the #if defined(), so that loops actually do get unrolled on Windows (15% perf win claimed many months ago).
http://codereview.appspot.com/5528078
git-svn-id: http://skia.googlecode.com/svn/trunk@3019
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Wed, 11 Jan 2012 19:17:23 +0000 (19:17 +0000)]
disable test until I can fix it
git-svn-id: http://skia.googlecode.com/svn/trunk@3018
2bbb7eff-a529-9590-31e7-
b0007b416f81
bungeman@google.com [Wed, 11 Jan 2012 19:17:20 +0000 (19:17 +0000)]
Roll gyp to 1139.
http://codereview.appspot.com/5529070/
git-svn-id: http://skia.googlecode.com/svn/trunk@3017
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Wed, 11 Jan 2012 18:51:15 +0000 (18:51 +0000)]
Fix variable renaming in asserts related to r3015
git-svn-id: http://skia.googlecode.com/svn/trunk@3016
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Wed, 11 Jan 2012 18:45:39 +0000 (18:45 +0000)]
Remove shadersupport caps bit
Review URL: http://codereview.appspot.com/5528076/
git-svn-id: http://skia.googlecode.com/svn/trunk@3015
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Wed, 11 Jan 2012 18:16:39 +0000 (18:16 +0000)]
fix cheapComputeDirection() in the non-convex case to pivot on the y-max point
rather than pivoting on the next. Also remove the loop, as it is just logically
wrong to think we can use this trick only a non-extrema pt.
git-svn-id: http://skia.googlecode.com/svn/trunk@3014
2bbb7eff-a529-9590-31e7-
b0007b416f81
bungeman@google.com [Wed, 11 Jan 2012 17:54:25 +0000 (17:54 +0000)]
Rebaseline for 3010.
git-svn-id: http://skia.googlecode.com/svn/trunk@3013
2bbb7eff-a529-9590-31e7-
b0007b416f81
digit@google.com [Wed, 11 Jan 2012 17:44:41 +0000 (17:44 +0000)]
android: optimize atomics routines
This patch provides a slightly optimized implementation of
atomic increment/decrement functions by using static inlined
versions of the gcc intrinsics.
Review URL: http://codereview.appspot.com/5498069
git-svn-id: http://skia.googlecode.com/svn/trunk@3012
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Wed, 11 Jan 2012 16:41:26 +0000 (16:41 +0000)]
improve clamping logic when handling imprecise chopping of cubics
http://code.google.com/p/skia/issues/detail?id=444
git-svn-id: http://skia.googlecode.com/svn/trunk@3011
2bbb7eff-a529-9590-31e7-
b0007b416f81
bungeman@google.com [Wed, 11 Jan 2012 16:28:04 +0000 (16:28 +0000)]
Correct GPU glyph positioning.
http://codereview.appspot.com/5522050/
git-svn-id: http://skia.googlecode.com/svn/trunk@3010
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Wed, 11 Jan 2012 16:00:42 +0000 (16:00 +0000)]
fix compilation error in GrContext.cpp when PREFER_MSAA_OFFSCREEN_AA is enabled
http://codereview.appspot.com/5534068/
git-svn-id: http://skia.googlecode.com/svn/trunk@3009
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Wed, 11 Jan 2012 15:33:20 +0000 (15:33 +0000)]
fix comment
git-svn-id: http://skia.googlecode.com/svn/trunk@3008
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Wed, 11 Jan 2012 14:07:27 +0000 (14:07 +0000)]
detect (and reject) if radius ia NaN, by using !(radius > 0) instead of (radius <= 0)
git-svn-id: http://skia.googlecode.com/svn/trunk@3007
2bbb7eff-a529-9590-31e7-
b0007b416f81
senorblanco@chromium.org [Tue, 10 Jan 2012 21:56:40 +0000 (21:56 +0000)]
Export SkColorMatrixFilter via SK_API.
Review URL: http://codereview.appspot.com/5534064/
git-svn-id: http://skia.googlecode.com/svn/trunk@3006
2bbb7eff-a529-9590-31e7-
b0007b416f81
caryclark@google.com [Tue, 10 Jan 2012 21:46:10 +0000 (21:46 +0000)]
work in progress for shape operations
A experimental/Intersection
A experimental/Intersection/Intersections.h
A experimental/Intersection/DataTypes.cpp
A experimental/Intersection/QuadraticReduceOrder.cpp
A experimental/Intersection/IntersectionUtilities.cpp
A experimental/Intersection/CubicIntersection_Tests.h
A experimental/Intersection/LineParameteters_Test.cpp
A experimental/Intersection/ReduceOrder.cpp
A experimental/Intersection/QuadraticIntersection.cpp
A experimental/Intersection/Extrema.h
A experimental/Intersection/CubicIntersection_TestData.h
A experimental/Intersection/QuadraticParameterization_Test.cpp
A experimental/Intersection/TestUtilities.cpp
A experimental/Intersection/CubicRoots.cpp
A experimental/Intersection/QuadraticParameterization.cpp
A experimental/Intersection/QuadraticSubDivide.cpp
A experimental/Intersection/LineIntersection_Test.cpp
A experimental/Intersection/LineIntersection.cpp
A experimental/Intersection/CubicParameterizationCode.cpp
A experimental/Intersection/LineParameters.h
A experimental/Intersection/CubicIntersection.h
A experimental/Intersection/CubeRoot.cpp
A experimental/Intersection/SkAntiEdge.h
A experimental/Intersection/ConvexHull_Test.cpp
A experimental/Intersection/CubicBezierClip_Test.cpp
A experimental/Intersection/CubicIntersection_Tests.cpp
A experimental/Intersection/CubicBezierClip.cpp
A experimental/Intersection/CubicIntersectionT.cpp
A experimental/Intersection/Inline_Tests.cpp
A experimental/Intersection/ReduceOrder_Test.cpp
A experimental/Intersection/QuadraticIntersection_TestData.h
A experimental/Intersection/DataTypes.h
A experimental/Intersection/Extrema.cpp
A experimental/Intersection/EdgeApp.cpp
A experimental/Intersection/CubicIntersection_TestData.cpp
A experimental/Intersection/IntersectionUtilities.h
A experimental/Intersection/CubicReduceOrder.cpp
A experimental/Intersection/CubicCoincidence.cpp
A experimental/Intersection/CubicIntersection_Test.cpp
A experimental/Intersection/CubicIntersection.cpp
A experimental/Intersection/QuadraticUtilities.h
A experimental/Intersection/SkAntiEdge.cpp
A experimental/Intersection/TestUtilities.h
A experimental/Intersection/CubicParameterization_Test.cpp
A experimental/Intersection/LineIntersection.h
A experimental/Intersection/CubicSubDivide.cpp
A experimental/Intersection/CubicParameterization.cpp
A experimental/Intersection/QuadraticBezierClip_Test.cpp
A experimental/Intersection/QuadraticBezierClip.cpp
A experimental/Intersection/BezierClip_Test.cpp
A experimental/Intersection/ConvexHull.cpp
A experimental/Intersection/BezierClip.cpp
A experimental/Intersection/QuadraticIntersection_TestData.cpp
git-svn-id: http://skia.googlecode.com/svn/trunk@3005
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Tue, 10 Jan 2012 21:30:57 +0000 (21:30 +0000)]
stroking wants CW, so we reverse if the orig was CCW
git-svn-id: http://skia.googlecode.com/svn/trunk@3004
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Tue, 10 Jan 2012 19:28:01 +0000 (19:28 +0000)]
use SkIntToScalar
git-svn-id: http://skia.googlecode.com/svn/trunk@3003
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Tue, 10 Jan 2012 18:59:22 +0000 (18:59 +0000)]
use SkIntToScalar(10) instead of 10 for parameters (for scalar==fixed)
git-svn-id: http://skia.googlecode.com/svn/trunk@3002
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Tue, 10 Jan 2012 18:44:37 +0000 (18:44 +0000)]
add gm for reverseAddPath
git-svn-id: http://skia.googlecode.com/svn/trunk@3001
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Tue, 10 Jan 2012 18:40:03 +0000 (18:40 +0000)]
restore (debugged) cheapComputeDirection
git-svn-id: http://skia.googlecode.com/svn/trunk@3000
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Tue, 10 Jan 2012 18:13:23 +0000 (18:13 +0000)]
disable for now
git-svn-id: http://skia.googlecode.com/svn/trunk@2999
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Tue, 10 Jan 2012 18:02:09 +0000 (18:02 +0000)]
disable cheap check while its being debugged
git-svn-id: http://skia.googlecode.com/svn/trunk@2998
2bbb7eff-a529-9590-31e7-
b0007b416f81
bsalomon@google.com [Tue, 10 Jan 2012 18:01:01 +0000 (18:01 +0000)]
Remove accidentally checked in printf
git-svn-id: http://skia.googlecode.com/svn/trunk@2997
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Tue, 10 Jan 2012 18:00:10 +0000 (18:00 +0000)]
add SkPath::cheapComputeDirection() plus unittests
git-svn-id: http://skia.googlecode.com/svn/trunk@2996
2bbb7eff-a529-9590-31e7-
b0007b416f81
reed@google.com [Tue, 10 Jan 2012 15:33:12 +0000 (15:33 +0000)]
add SkPath::reverseAddPath()
git-svn-id: http://skia.googlecode.com/svn/trunk@2995
2bbb7eff-a529-9590-31e7-
b0007b416f81
epoger@google.com [Tue, 10 Jan 2012 14:10:34 +0000 (14:10 +0000)]
Add --add-new-files option to download-baselines.py script
This will allow the user to determine whether the script downloads just
updates for baselines that are already checked in, or all available baselines.
BUG=436
addresses http://code.google.com/p/skia/issues/detail?id=436 ('add "update" flag to download-baselines script')
Review URL: http://codereview.appspot.com/5523050
git-svn-id: http://skia.googlecode.com/svn/trunk@2993
2bbb7eff-a529-9590-31e7-
b0007b416f81
digit@google.com [Tue, 10 Jan 2012 10:28:30 +0000 (10:28 +0000)]
Android: disable static global initializers
Review URL: http://codereview.appspot.com/5503066
git-svn-id: http://skia.googlecode.com/svn/trunk@2992
2bbb7eff-a529-9590-31e7-
b0007b416f81
digit@google.com [Tue, 10 Jan 2012 10:00:59 +0000 (10:00 +0000)]
remove one un-needed static initializer
This patch removes the use of a static C++ object in the
implementation of SkMatrix44::isIdentity(). Instead, we rely
on direct comparison with a statically allocated array of
SkMScalar values, which is completely equivalent.
Review URL: http://codereview.appspot.com/5502067
git-svn-id: http://skia.googlecode.com/svn/trunk@2991
2bbb7eff-a529-9590-31e7-
b0007b416f81