platform/upstream/libSkiaSharp.git
10 years agoFix ios build?
bsalomon@google.com [Tue, 21 Jan 2014 16:35:56 +0000 (16:35 +0000)]
Fix ios build?

Review URL: https://codereview.chromium.org/134083009

git-svn-id: http://skia.googlecode.com/svn/trunk@13131 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoMove GrGLInterface function pointers into a nested struct
commit-bot@chromium.org [Tue, 21 Jan 2014 16:09:18 +0000 (16:09 +0000)]
Move GrGLInterface function pointers into a nested struct

BUG=skia:2042
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/133073009

git-svn-id: http://skia.googlecode.com/svn/trunk@13130 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoScope our ANNOTATE_BENIGN_RACE.
commit-bot@chromium.org [Tue, 21 Jan 2014 15:11:02 +0000 (15:11 +0000)]
Scope our ANNOTATE_BENIGN_RACE.

This should unstick the DEPS roll.  When building with TSAN, Chromium's pulling
in the canonical one from dynamic_annotations.h, and we're trying to redefine it.

R=robertphillips@google.com, reed@google.com, bsalomon@google.com
TBR=reed@google.com
BUG=

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/144063002

git-svn-id: http://skia.googlecode.com/svn/trunk@13129 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoWhitepspace change to verify that the bots are working
borenet@google.com [Tue, 21 Jan 2014 14:25:35 +0000 (14:25 +0000)]
Whitepspace change to verify that the bots are working

This CL caused some buildslaves to fail to UpdateScripts:
https://skia.googlesource.com/buildbot/+/6707360ca6f76ae383ac7558015fb94f4d12577c

Review URL: https://codereview.chromium.org/144073002

git-svn-id: http://skia.googlecode.com/svn/trunk@13128 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoApply the CTM to filter parameters for SkBlurImageFilter, SkDropShadowImageFilter...
senorblanco@chromium.org [Mon, 20 Jan 2014 19:58:28 +0000 (19:58 +0000)]
Apply the CTM to filter parameters for SkBlurImageFilter, SkDropShadowImageFilter, SkDisplacementMapEffect and SkMorphologyImageFilter. I had resisted this (and may later put in an assert that the CTM contains no rotation, skew or perspective), but it just makes the filters play better with Skia generally.

This revealed that the displacement map was not handling clipping or upstream cropping at all well (the color would "swim" through the displacement at the edge of the clip rect). Fixed by passing through the correct offsets to the bitmap accesses in both raster and GPU paths. Same for morphology.

R=sugoi@google.com

Review URL: https://codereview.chromium.org/137053003

git-svn-id: http://skia.googlecode.com/svn/trunk@13127 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoAllow an override of MACOSX_DEPLOYMENT_TARGET, so the user can build against a differ...
senorblanco@chromium.org [Mon, 20 Jan 2014 16:14:02 +0000 (16:14 +0000)]
Allow an override of MACOSX_DEPLOYMENT_TARGET, so the user can build against a different SDK if desired.

R=mtklein@google.com, bungeman, mtklein

Review URL: https://codereview.chromium.org/138033018

git-svn-id: http://skia.googlecode.com/svn/trunk@13126 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoFix utils.gyp missing file
bsalomon@google.com [Fri, 17 Jan 2014 20:33:09 +0000 (20:33 +0000)]
Fix utils.gyp missing file

TBR=robertphillips@google.com

Review URL: https://codereview.chromium.org/142043003

git-svn-id: http://skia.googlecode.com/svn/trunk@13125 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoignore some verylargebitmap errors
commit-bot@chromium.org [Fri, 17 Jan 2014 20:16:05 +0000 (20:16 +0000)]
ignore some verylargebitmap errors

BUG=skia:1978
NOTREECHECK=True
NOTRY=True
R=senorblanco@google.com
TBR=senorblanco

Author: epoger@google.com

Review URL: https://codereview.chromium.org/140643003

git-svn-id: http://skia.googlecode.com/svn/trunk@13124 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoSpeed up GrResourceCache add and lookup by using TDynamicHash
commit-bot@chromium.org [Fri, 17 Jan 2014 17:56:21 +0000 (17:56 +0000)]
Speed up GrResourceCache add and lookup by using TDynamicHash

Speed up GrResourceCache add and lookup by using TDynamicHash instead
of GrTHashTable. GrTHashTable spends most of its time memmoving the
array elements while sorting after an add. Lookup is not particularly
fast either.

R=mtklein@google.com, bsalomon@google.com, reed@google.com

Author: kkinnunen@nvidia.com

Review URL: https://codereview.chromium.org/91453002

git-svn-id: http://skia.googlecode.com/svn/trunk@13122 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoTweak validate() to check less as the size of the hash grows.
commit-bot@chromium.org [Fri, 17 Jan 2014 17:55:14 +0000 (17:55 +0000)]
Tweak validate() to check less as the size of the hash grows.

This makes working with a non-trivial SkTDynamicHash less painful in Debug
builds.  validate() is skipped in Release, so this has no effect there.

BUG=
R=kkinnunen@nvidia.com, bsalomon@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/140753003

git-svn-id: http://skia.googlecode.com/svn/trunk@13121 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoMake leak counters thread-safe and turn them on by default for Debug
commit-bot@chromium.org [Fri, 17 Jan 2014 17:55:02 +0000 (17:55 +0000)]
Make leak counters thread-safe and turn them on by default for Debug

Make leak counters implemented with SK_DECLARE_INST_COUNT thread-safe.
Enable the leak counting for Debug builds when Skia is built as a
static library. Having SK_DECLARE_INST_COUNT without
SK_DEFINE_INST_COUNT relies on static variables in member functions
declared in the header files. These might be duplicated in the clients
of the library when Skia is built as a dynamic library, producing
incorrect operation.

Protect the instance counter initialization step (initStep) by
using SkOnce.

Makes SkOnce.h part of the public API, since SkInstCnt is public.

Protect the per-class child list shared variable with a per-class mutex.

Changes the behavior in the way that if the child list has been
"cleaned up", it will still try to create subsequent child lists.

BUG=skia:1219
R=robertphillips@google.com, mtklein@google.com, bsalomon@google.com, bungeman@google.com, djsollen@google.com

Author: kkinnunen@nvidia.com

Review URL: https://codereview.chromium.org/99483003

git-svn-id: http://skia.googlecode.com/svn/trunk@13120 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoAdding SK_API to SkGLContextHelper and SkNullGLContext for use in blink unit test...
commit-bot@chromium.org [Fri, 17 Jan 2014 16:59:03 +0000 (16:59 +0000)]
Adding SK_API to SkGLContextHelper and SkNullGLContext for use in blink unit test component build

BUG=skia:
R=bsalomon@chromium.org

Author: junov@chromium.org

Review URL: https://codereview.chromium.org/141813002

git-svn-id: http://skia.googlecode.com/svn/trunk@13119 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoMove GrGLExtensions from GrGLContextInfo to GrGLInterface
commit-bot@chromium.org [Fri, 17 Jan 2014 15:05:38 +0000 (15:05 +0000)]
Move GrGLExtensions from GrGLContextInfo to GrGLInterface

BUG=skia:2042
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/140823003

git-svn-id: http://skia.googlecode.com/svn/trunk@13118 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoAdd AUTHORS file for copyright tracking
commit-bot@chromium.org [Fri, 17 Jan 2014 15:00:14 +0000 (15:00 +0000)]
Add AUTHORS file for copyright tracking

BUG=skia:
R=jcgregorio@google.com, rmistry@google.com

Author: hcm@google.com

Review URL: https://codereview.chromium.org/138413004

git-svn-id: http://skia.googlecode.com/svn/trunk@13117 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoAdding SkPictureImageFilter to the fuzzer
commit-bot@chromium.org [Thu, 16 Jan 2014 22:54:41 +0000 (22:54 +0000)]
Adding SkPictureImageFilter to the fuzzer

BUG=skia:
R=senorblanco@google.com, senorblanco@chromium.org, sugoi@google.com

Author: sugoi@chromium.org

Review URL: https://codereview.chromium.org/140993004

git-svn-id: http://skia.googlecode.com/svn/trunk@13116 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRemove GR_GL_IGNORE_ES3_MSAA.s
commit-bot@chromium.org [Thu, 16 Jan 2014 20:49:46 +0000 (20:49 +0000)]
Remove GR_GL_IGNORE_ES3_MSAA.s

R=jvanverth@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/140933005

git-svn-id: http://skia.googlecode.com/svn/trunk@13115 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoClear the canvas before drawing the gears.
commit-bot@chromium.org [Thu, 16 Jan 2014 19:48:33 +0000 (19:48 +0000)]
Clear the canvas before drawing the gears.

BUG=skia:
R=robertphillips@google.com

Author: jcgregorio@google.com

Review URL: https://codereview.chromium.org/141113002

git-svn-id: http://skia.googlecode.com/svn/trunk@13114 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoMake tools/submit_try safer
commit-bot@chromium.org [Thu, 16 Jan 2014 19:48:19 +0000 (19:48 +0000)]
Make tools/submit_try safer

Using proc.wait() in combination with PIPE doesn't seem
to be safe for large patches according to
http://docs.python.org/2/library/subprocess.html.

Moving to proc.communicate().

Signed-off-by: Kévin PETIT <kevin.petit@arm.com>
BUG=skia:
R=djsollen@google.com, borenet@google.com

Author: kevin.petit.arm@gmail.com

Review URL: https://codereview.chromium.org/141023003

git-svn-id: http://skia.googlecode.com/svn/trunk@13113 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoAllow GrGLInterface factories in Chromium to continue to compile.
commit-bot@chromium.org [Thu, 16 Jan 2014 18:49:01 +0000 (18:49 +0000)]
Allow GrGLInterface factories in Chromium to continue to compile.

BUG=skia:2042
TBR=jvanverth@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/136573003

git-svn-id: http://skia.googlecode.com/svn/trunk@13112 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoMake GrGLContextInfo have private ptr to GrGLInterface
commit-bot@chromium.org [Thu, 16 Jan 2014 18:39:04 +0000 (18:39 +0000)]
Make GrGLContextInfo have private ptr to GrGLInterface

BUG=skia:2042
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/140843003

git-svn-id: http://skia.googlecode.com/svn/trunk@13111 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoMissing #include "SkTemplates.h" for SkAutoTMalloc.
commit-bot@chromium.org [Thu, 16 Jan 2014 18:16:37 +0000 (18:16 +0000)]
Missing #include "SkTemplates.h" for SkAutoTMalloc.

BUG=
R=halcanary@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/137783021

git-svn-id: http://skia.googlecode.com/svn/trunk@13110 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoquick fix for r13108
bsalomon@google.com [Thu, 16 Jan 2014 17:51:07 +0000 (17:51 +0000)]
quick fix for r13108

BUG=skia:2042
NOTREECHECKS
NOTRY
TBR=robertphillips@google.com

Review URL: https://codereview.chromium.org/135313004

git-svn-id: http://skia.googlecode.com/svn/trunk@13109 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRename GrGLBinding->GrGLStandard, no longer a bitfield
commit-bot@chromium.org [Thu, 16 Jan 2014 16:35:09 +0000 (16:35 +0000)]
Rename GrGLBinding->GrGLStandard, no longer a bitfield

BUG=skia:2042
R=jvanverth@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/133413003

git-svn-id: http://skia.googlecode.com/svn/trunk@13108 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoHide bitmap rowbyes assert when LEGACY_ONLOCKPIXELS
commit-bot@chromium.org [Thu, 16 Jan 2014 15:47:06 +0000 (15:47 +0000)]
Hide bitmap rowbyes assert when LEGACY_ONLOCKPIXELS

Motivation:  without this, we break Blink.

BUG=skia:

NOTREECHECKS=true
NOTRY=true
R=robertphillips@google.com

Author: halcanary@google.com

Review URL: https://codereview.chromium.org/140903002

git-svn-id: http://skia.googlecode.com/svn/trunk@13107 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoSK_API SkMallocPixelRef
commit-bot@chromium.org [Thu, 16 Jan 2014 14:46:16 +0000 (14:46 +0000)]
SK_API SkMallocPixelRef

Motivation: make https://codereview.chromium.org/138393002/ work.

R=reed@google.com, djsollen@google.com, bsalomon@google.com

Author: halcanary@google.com

Review URL: https://codereview.chromium.org/135853019

git-svn-id: http://skia.googlecode.com/svn/trunk@13106 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoSanitizing source files in Housekeeper-Nightly
skia.committer@gmail.com [Thu, 16 Jan 2014 07:01:48 +0000 (07:01 +0000)]
Sanitizing source files in Housekeeper-Nightly

git-svn-id: http://skia.googlecode.com/svn/trunk@13105 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoMove SkMessageBus::Get out of header, and retry crrev.com/106563002.
commit-bot@chromium.org [Wed, 15 Jan 2014 23:09:01 +0000 (23:09 +0000)]
Move SkMessageBus::Get out of header, and retry crrev.com/106563002.

BUG=
R=bsalomon@google.com, kkinnunen@nvidia.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/140053002

git-svn-id: http://skia.googlecode.com/svn/trunk@13104 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoFactor out UnPreMultiplyPreservingByteOrder
commit-bot@chromium.org [Wed, 15 Jan 2014 22:24:58 +0000 (22:24 +0000)]
Factor out UnPreMultiplyPreservingByteOrder

BUG=
R=reed@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/130303005

git-svn-id: http://skia.googlecode.com/svn/trunk@13103 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoUse MACOSX_DEPLOYMENT_TARGET instead of SDKROOT.
commit-bot@chromium.org [Wed, 15 Jan 2014 21:39:18 +0000 (21:39 +0000)]
Use MACOSX_DEPLOYMENT_TARGET instead of SDKROOT.

MACOSX_DEPLOYMENT_TARGET is the minimum unconditionally supported OS,
which should just be 10.6 for us until Chrome changes.  SDKROOT is the
maximum conditionally supported OS, which defaults sanely to whatever's
the latest SDK the machine has, so we don't need to mention it.

BUG=skia:
R=bungeman@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/137793010

git-svn-id: http://skia.googlecode.com/svn/trunk@13102 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoDM: don't change byte order when unpremultiplying to compare to PNGs.
commit-bot@chromium.org [Wed, 15 Jan 2014 21:28:25 +0000 (21:28 +0000)]
DM: don't change byte order when unpremultiplying to compare to PNGs.

We've already decoded the PNGs themselves into unpremultiplied bitmaps with native byte order.  SkColor is just not the right choice unless you get lucky.

dm -w /tmp/dm && dm -r /tmp/dm still works on Linux, and it's much closer to working on Mac:

0 tasks left, 2 failed
Failures:
  matrixconvolution_gpu
  colormatrix_gpu

BUG=skia:
R=reed@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/139943002

git-svn-id: http://skia.googlecode.com/svn/trunk@13101 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agorender_pictures: adjust unittests to show specific failures
commit-bot@chromium.org [Wed, 15 Jan 2014 20:35:54 +0000 (20:35 +0000)]
render_pictures: adjust unittests to show specific failures

BUG=skia:2043,skia:2044
R=rmistry@google.com

Author: epoger@google.com

Review URL: https://codereview.chromium.org/139383004

git-svn-id: http://skia.googlecode.com/svn/trunk@13100 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoChange to fuzzer for new constructor
commit-bot@chromium.org [Wed, 15 Jan 2014 20:20:24 +0000 (20:20 +0000)]
Change to fuzzer for new constructor

BUG=skia:
R=senorblanco@google.com, sugoi@google.com, senorblanco@chromium.org

Author: sugoi@chromium.org

Review URL: https://codereview.chromium.org/139613002

git-svn-id: http://skia.googlecode.com/svn/trunk@13099 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agono need to refcnt colortables, since pixelrefs own them
reed@google.com [Wed, 15 Jan 2014 19:59:45 +0000 (19:59 +0000)]
no need to refcnt colortables, since pixelrefs own them

BUG=skia:
R=halcanary@google.com, scroggo@google.com

Review URL: https://codereview.chromium.org/139463007

git-svn-id: http://skia.googlecode.com/svn/trunk@13098 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoMake GrGLInterface be context-specific on all platforms
commit-bot@chromium.org [Wed, 15 Jan 2014 19:32:03 +0000 (19:32 +0000)]
Make GrGLInterface be context-specific on all platforms

BUG=skia:2042
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/137753005

git-svn-id: http://skia.googlecode.com/svn/trunk@13097 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoAdd new baselines for bitmapsource, dropshadowimagefilter, pictureimagefilter and...
senorblanco@chromium.org [Wed, 15 Jan 2014 19:28:53 +0000 (19:28 +0000)]
Add new baselines for bitmapsource, dropshadowimagefilter, pictureimagefilter and resizeimagefilter GMs.

TBR=djsollen

Review URL: https://codereview.chromium.org/139513004

git-svn-id: http://skia.googlecode.com/svn/trunk@13096 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agodisable poplar/fontconfig by default on mac
commit-bot@chromium.org [Wed, 15 Jan 2014 19:28:27 +0000 (19:28 +0000)]
disable poplar/fontconfig by default on mac

BUG=skia:2045
NOTRY=true
R=vandebo@chromium.org, mtklein@google.com

Author: reed@google.com

Review URL: https://codereview.chromium.org/139863002

git-svn-id: http://skia.googlecode.com/svn/trunk@13095 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agorebaseline_server: ignore TSAN, ASAN, Valgrind bots
commit-bot@chromium.org [Wed, 15 Jan 2014 19:16:09 +0000 (19:16 +0000)]
rebaseline_server: ignore TSAN, ASAN, Valgrind bots

BUG=skia:2036
R=bsalomon@google.com

Author: epoger@google.com

Review URL: https://codereview.chromium.org/134643010

git-svn-id: http://skia.googlecode.com/svn/trunk@13094 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoadd asserts to validate
reed@google.com [Wed, 15 Jan 2014 19:15:23 +0000 (19:15 +0000)]
add asserts to validate

BUG=skia:
R=halcanary@google.com, scroggo@google.com

Review URL: https://codereview.chromium.org/139063003

git-svn-id: http://skia.googlecode.com/svn/trunk@13093 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agochange isRect to return true for 3-sided rectangular paths
commit-bot@chromium.org [Wed, 15 Jan 2014 18:00:57 +0000 (18:00 +0000)]
change isRect to return true for 3-sided rectangular paths

BUG=skia:
R=caryclark@google.com, yunchao.he@intel.com

Author: reed@google.com

Review URL: https://codereview.chromium.org/139483002

git-svn-id: http://skia.googlecode.com/svn/trunk@13092 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRoll fontconfig to 2.11.0 and reenable poppler on mac
commit-bot@chromium.org [Wed, 15 Jan 2014 17:24:22 +0000 (17:24 +0000)]
Roll fontconfig to 2.11.0 and reenable poppler on mac

R=mtklein@google.com, epoger@google.com

Author: vandebo@chromium.org

Review URL: https://codereview.chromium.org/138843002

git-svn-id: http://skia.googlecode.com/svn/trunk@13091 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoReland "Fix genID cloning bugs."
scroggo@google.com [Wed, 15 Jan 2014 16:56:52 +0000 (16:56 +0000)]
Reland "Fix genID cloning bugs."

SkBitmap.cpp:
When copyTo calls readPixels, only clone the genID if the resulting
SkPixelRef has the same dimensions as the original. This catches a
bug where copying an SkBitmap representing the subset of an SkPixelRef
(which implements onReadPixels) would result in the copy sharing the
genID. (Thanks to r6710, this case can only happen using setPixelRef,
so the updated GpuBitmapCopyTest checks for that.)

Move some unnecessary NULL checks to asserts.

When copyTo performs a memcpy, only clone the genID if the resulting
SkPixelRef has the same dimensions as the original. This catches a bug
where copying an extracted SkBitmap with the same width as its original
SkPixelRef would incorrectly have the same genID.

Add a comment and assert in deepCopyTo, when cloning the genID, since
that case correctly clones it.

BitmapCopyTest.cpp:
Pull redundant work out of the inner loop (setting up the source bitmaps
and testing extractSubset). Create a new inner loop for extractSubset, to
test copying the result to each different config.

Extract a subset that has the same width as the original, to catch the
bug mentioned above.

Remove the reporter assert which checks for the resulting rowbytes.

Add checks to ensure that copying the extracted subset changes the genID.

GpuBitmapCopyTest:
Create an SkBitmap that shares an existing SkPixelRef, but only represents
a subset. This is to test the first call to cloneGenID in SkBitmap::copyTo.
In this case, the genID should NOT be copied, since only a portion of the
SkPixelRef was copied.

Also test deepCopy on this subset.

TestIndividualCopy now takes a parameter stating whether the genID should
change in the copy. It also does a read back using the appropriate subset.
It no longer differentiates between copyTo and deepCopyTo, since that
distinction was only necessary for copying from/to configs other than 8888
(which are no longer being tested), where copyTo did a read back in 8888 and
then drew the result to the desired config (resulting in an imperfect copy).

BUG=skia:1742

Committed: http://code.google.com/p/skia/source/detail?r=13021

R=mtklein@google.com

Review URL: https://codereview.chromium.org/112113005

git-svn-id: http://skia.googlecode.com/svn/trunk@13090 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoIgnore out even if it's a symlink.
commit-bot@chromium.org [Wed, 15 Jan 2014 16:55:04 +0000 (16:55 +0000)]
Ignore out even if it's a symlink.

R=halcanary@google.com
BUG=

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/139473002

git-svn-id: http://skia.googlecode.com/svn/trunk@13089 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years ago[try 2] skpmaker: make the SKP more graphically interesting (add a border)
commit-bot@chromium.org [Wed, 15 Jan 2014 16:33:31 +0000 (16:33 +0000)]
[try 2] skpmaker: make the SKP more graphically interesting (add a border)

relanding https://codereview.chromium.org/137993008/

BUG=skia:1942
NOTRY=True
R=rmistry@google.com
TBR=rmistry

Author: epoger@google.com

Review URL: https://codereview.chromium.org/139583002

git-svn-id: http://skia.googlecode.com/svn/trunk@13088 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRevert of skpmaker: make the SKP slightly more interesting (add a border) (https...
commit-bot@chromium.org [Wed, 15 Jan 2014 16:04:45 +0000 (16:04 +0000)]
Revert of skpmaker: make the SKP slightly more interesting (add a border) (https://codereview.chromium.org/137993008/)

Reason for revert:
broke http://108.170.217.252:10117/builders/Build-Mac10.6-GCC-x86-Debug/builds/2837

Original issue's description:
> skpmaker: make the SKP slightly more interesting (add a border)
>
> This will help with render_pictures self-testing
> BUG=skia:1942
>
> Committed: http://code.google.com/p/skia/source/detail?r=13086

R=rmistry@google.com
TBR=rmistry@google.com
NOTREECHECKS=true
NOTRY=true
BUG=skia:1942

Author: epoger@google.com

Review URL: https://codereview.chromium.org/139553002

git-svn-id: http://skia.googlecode.com/svn/trunk@13087 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoskpmaker: make the SKP slightly more interesting (add a border)
commit-bot@chromium.org [Wed, 15 Jan 2014 15:40:13 +0000 (15:40 +0000)]
skpmaker: make the SKP slightly more interesting (add a border)

This will help with render_pictures self-testing
BUG=skia:1942
R=rmistry@google.com

Author: epoger@google.com

Review URL: https://codereview.chromium.org/137993008

git-svn-id: http://skia.googlecode.com/svn/trunk@13086 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRebaseline complexclip_aa_layer image for Nexus 4
robertphillips@google.com [Wed, 15 Jan 2014 14:52:03 +0000 (14:52 +0000)]
Rebaseline complexclip_aa_layer image for Nexus 4

https://codereview.chromium.org/138273008/

git-svn-id: http://skia.googlecode.com/svn/trunk@13085 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoWhen a picture lies about its size (i.e., claims to be 100x100 but only ever draws...
commit-bot@chromium.org [Wed, 15 Jan 2014 14:51:25 +0000 (14:51 +0000)]
When a picture lies about its size (i.e., claims to be 100x100 but only ever draws to 90x100) arbitrary pixel differences can appear in the undrawn portion. This is causing spurious differences to appear in the 8888 vs. gpu telemetry results.

R=bsalomon@google.com

Author: robertphillips@google.com

Review URL: https://codereview.chromium.org/136553004

git-svn-id: http://skia.googlecode.com/svn/trunk@13084 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoSanitizing source files in Housekeeper-Nightly
skia.committer@gmail.com [Wed, 15 Jan 2014 07:01:43 +0000 (07:01 +0000)]
Sanitizing source files in Housekeeper-Nightly

git-svn-id: http://skia.googlecode.com/svn/trunk@13083 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoImplement a resize image filter. This is needed for the "filterRes" feature in SVG...
senorblanco@chromium.org [Wed, 15 Jan 2014 04:49:18 +0000 (04:49 +0000)]
Implement a resize image filter. This is needed for the "filterRes" feature in SVG filter effects, which specifies the required size for intermediate processing buffers. In order to make this work, we need to render the primitive at the given resolution (doable at the callsite in Blink), and then to resize the result to the actual on-screen size. The latter is where this filter comes in.

It simply applies a scaling factor (and the current CTM) to its input, and draws its input bitmap at that size.

R=reed@google.com

Committed: https://code.google.com/p/skia/source/detail?r=13077
Reverted: https://code.google.com/p/skia/source/detail?r=13078

BUG=

Review URL: https://codereview.chromium.org/136863006

git-svn-id: http://skia.googlecode.com/svn/trunk@13082 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agouse some helper Make functions to initialize SkImageInfo
commit-bot@chromium.org [Wed, 15 Jan 2014 02:38:22 +0000 (02:38 +0000)]
use some helper Make functions to initialize SkImageInfo

BUG=
R=halcanary@google.com, scroggo@google.com

Author: reed@google.com

Review URL: https://codereview.chromium.org/137993012

git-svn-id: http://skia.googlecode.com/svn/trunk@13081 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agorebaseline blur related outputs
commit-bot@chromium.org [Wed, 15 Jan 2014 02:37:57 +0000 (02:37 +0000)]
rebaseline blur related outputs

rebaseline for the differences introduced by https://codereview.chromium.org/105893003/

BUG=
R=senorblanco@chromium.org, reed@google.com

Author: zheng.xu@arm.com

Review URL: https://codereview.chromium.org/136483002

git-svn-id: http://skia.googlecode.com/svn/trunk@13080 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoDeprecate two more functions using SK_ATTR_DEPRECATED macro.
tfarina@chromium.org [Wed, 15 Jan 2014 00:19:21 +0000 (00:19 +0000)]
Deprecate two more functions using SK_ATTR_DEPRECATED macro.

BUG=None
TEST=None
R=reed@google.com, scroggo@google.com

Review URL: https://codereview.chromium.org/136973007

git-svn-id: http://skia.googlecode.com/svn/trunk@13079 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRevert "Implement a resize image filter."
senorblanco@chromium.org [Tue, 14 Jan 2014 22:55:04 +0000 (22:55 +0000)]
Revert "Implement a resize image filter."

This reverts commit 3f996e31c7043929fc1553b9b5ecd6a82b5fe125 (r13077).

TBR=reed

Review URL: https://codereview.chromium.org/136793012

git-svn-id: http://skia.googlecode.com/svn/trunk@13078 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoImplement a resize image filter. This is needed for the "filterRes" feature in SVG...
senorblanco@chromium.org [Tue, 14 Jan 2014 22:20:48 +0000 (22:20 +0000)]
Implement a resize image filter. This is needed for the "filterRes" feature in SVG filter effects, which specifies the required size for intermediate processing buffers. In order to make this work, we need to render the primitive at the given resolution (doable at the callsite in Blink), and then to resize the result to the actual on-screen size. The latter is where this filter comes in.

It simply applies a scaling factor (and the current CTM) to its input, and draws its input bitmap at that size.

R=reed@google.com

Review URL: https://codereview.chromium.org/136863006

git-svn-id: http://skia.googlecode.com/svn/trunk@13077 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRevert "Make leak counters thread-safe and turn them on by default for Debug"
djsollen@google.com [Tue, 14 Jan 2014 21:54:44 +0000 (21:54 +0000)]
Revert "Make leak counters thread-safe and turn them on by default for Debug"

iThis CL is breaking the Android debug test bots by firing an assert.

BUG=skia:1219

Review URL: https://codereview.chromium.org/138683006

git-svn-id: http://skia.googlecode.com/svn/trunk@13076 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoAdd BUG_PREFIX to codereview.settings
commit-bot@chromium.org [Tue, 14 Jan 2014 21:19:21 +0000 (21:19 +0000)]
Add BUG_PREFIX to codereview.settings

NOTRY=True
NOTREECHECKS=True

(SkipBuildbotRuns)

BUG=skia:1983
R=borenet@google.com

Author: rmistry@google.com

Review URL: https://codereview.chromium.org/138863002

git-svn-id: http://skia.googlecode.com/svn/trunk@13075 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoMove macros from TestClassDef.h to Test.h
commit-bot@chromium.org [Tue, 14 Jan 2014 21:04:37 +0000 (21:04 +0000)]
Move macros from TestClassDef.h to Test.h

Motivation: those macros don't make any sense without the definitions
in Test.h.

BUG=
R=mtklein@google.com

Author: halcanary@google.com

Review URL: https://codereview.chromium.org/138563004

git-svn-id: http://skia.googlecode.com/svn/trunk@13074 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoConvert SkWriter32 to use an SkTDArray for its internal storage.
commit-bot@chromium.org [Tue, 14 Jan 2014 20:51:26 +0000 (20:51 +0000)]
Convert SkWriter32 to use an SkTDArray for its internal storage.

This reduces the allocation overhead of a null picture (create, beginRecording(), endRecording) from about 18K to about 1.9K.  (There's still lots more to prune.)

SkPictureFlat can exploit the fact that Writer32 is contiguous simplify its memory management.  The Writer32 itself becomes the scratch buffer.

Remove lots and lots of arbitrary magic numbers that were size guesses and minimum allocation sizes.  Keep your eyes open for the big obvious DUH why we save 16K per picture!  (Spoiler alert.  It's because that first save we issue in beginRecording() forces the old SkWriter32 to allocate 16K.)

Tests passing, DM passing.

bench --match writer: ~20% faster
null bench_record: ~30% faster
bench_record on buildbot .skps: ~3-6% slower, ranging 25% faster to 20% slower
bench_pictures on buildbot .skps: ~1-2% faster, ranging 13% faster to 28% slower

BUG=skia:1850
R=reed@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/137433003

git-svn-id: http://skia.googlecode.com/svn/trunk@13073 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoDefault Mac to 64-bit.
commit-bot@chromium.org [Tue, 14 Jan 2014 20:49:26 +0000 (20:49 +0000)]
Default Mac to 64-bit.

DM keeps exhausting 4G address space when it's really rolling.

BUG=
R=reed@google.com, borenet@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/135733004

git-svn-id: http://skia.googlecode.com/svn/trunk@13072 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agosubmit_try: Obtain the list of trybots from the checked-in slaves.cfg
commit-bot@chromium.org [Tue, 14 Jan 2014 19:18:45 +0000 (19:18 +0000)]
submit_try: Obtain the list of trybots from the checked-in slaves.cfg

This should drastically speed up the script.

BUG=
R=epoger@google.com, rmistry@google.com

Author: borenet@google.com

Review URL: https://codereview.chromium.org/136683006

git-svn-id: http://skia.googlecode.com/svn/trunk@13071 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoAdd default_msaa16 and nvpr configs to bench_pictures.cfg
commit-bot@chromium.org [Tue, 14 Jan 2014 18:47:07 +0000 (18:47 +0000)]
Add default_msaa16 and nvpr configs to bench_pictures.cfg

BUG=skia:2032
R=robertphillips@google.com, bsalomon@google.com

Author: borenet@google.com

Review URL: https://codereview.chromium.org/137313005

git-svn-id: http://skia.googlecode.com/svn/trunk@13070 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoInitialize written paths and strokerecs lazily during GPU drawPath
commit-bot@chromium.org [Tue, 14 Jan 2014 18:42:34 +0000 (18:42 +0000)]
Initialize written paths and strokerecs lazily during GPU drawPath

Initialize SkPaths and SkStrokeRecs lazily during GPU drawPath calls.
The constructors seem to appear in some profiler results on ARM (~1%).

R=bsalomon@google.com, robertphillips@google.com

Author: kkinnunen@nvidia.com

Review URL: https://codereview.chromium.org/118143002

git-svn-id: http://skia.googlecode.com/svn/trunk@13069 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoMake leak counters thread-safe and turn them on by default for Debug
commit-bot@chromium.org [Tue, 14 Jan 2014 18:42:26 +0000 (18:42 +0000)]
Make leak counters thread-safe and turn them on by default for Debug

Make leak counters implemented with SK_DECLARE_INST_COUNT thread-safe.
Enable the leak counting for Debug builds.

Protect the instance counter initialization step (initStep) by
using SkOnce.

Makes SkOnce.h part of the public API, since SkInstCnt is public.

Protect the per-class child list shared variable with a per-class mutex.

Changes the behavior in the way that if the child list has been
"cleaned up", it will still try to create subsequent child lists.

BUG=skia:1219

Committed: http://code.google.com/p/skia/source/detail?r=12635

R=robertphillips@google.com, mtklein@google.com, bsalomon@google.com, bungeman@google.com

Author: kkinnunen@nvidia.com

Review URL: https://codereview.chromium.org/99483003

git-svn-id: http://skia.googlecode.com/svn/trunk@13068 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRebaseline bleed and simpleblurroundrect due to GPU blur changes. Also, misc blur...
bsalomon@google.com [Tue, 14 Jan 2014 18:20:49 +0000 (18:20 +0000)]
Rebaseline bleed and simpleblurroundrect due to GPU blur changes. Also, misc blur GMs on N4.

BUG=skia:199s8

Review URL: https://codereview.chromium.org/138593002

git-svn-id: http://skia.googlecode.com/svn/trunk@13067 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoUse git_cl.Changelist.GetUpstreamBranch() for diff in submit_try
borenet@google.com [Tue, 14 Jan 2014 18:03:10 +0000 (18:03 +0000)]
Use git_cl.Changelist.GetUpstreamBranch() for diff in submit_try

BUG=skia:1862
R=epoger@google.com

Review URL: https://codereview.chromium.org/136823002

git-svn-id: http://skia.googlecode.com/svn/trunk@13066 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agodeprecate tools/test_rendering.py and tools/test_pictures.py
commit-bot@chromium.org [Tue, 14 Jan 2014 17:01:28 +0000 (17:01 +0000)]
deprecate tools/test_rendering.py and tools/test_pictures.py
they will be removed entirely on/after Feb 1 unless there is objection

BUG=skia:1943
R=borenet@google.com

Author: epoger@google.com

Review URL: https://codereview.chromium.org/135783003

git-svn-id: http://skia.googlecode.com/svn/trunk@13065 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRevert of Rebaselining bleed GMs due to recent GPU blur changes. (https://codereview...
commit-bot@chromium.org [Tue, 14 Jan 2014 16:21:31 +0000 (16:21 +0000)]
Revert of Rebaselining bleed GMs due to recent GPU blur changes. (https://codereview.chromium.org/138353002/)

Reason for revert:
the bleed_gpu GM is failing on all the test builders.

Original issue's description:
> Rebaselining bleed GMs due to recent GPU blur changes.
>
> BUG=skia:1998
>
> Committed: https://code.google.com/p/skia/source/detail?r=13063

R=bsalomon@google.com
TBR=bsalomon@google.com
NOTREECHECKS=true
NOTRY=true
BUG=skia:1998

Author: djsollen@google.com

Review URL: https://codereview.chromium.org/138133003

git-svn-id: http://skia.googlecode.com/svn/trunk@13064 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRebaselining bleed GMs due to recent GPU blur changes.
bsalomon@google.com [Tue, 14 Jan 2014 15:28:24 +0000 (15:28 +0000)]
Rebaselining bleed GMs due to recent GPU blur changes.

BUG=skia:1998

Review URL: https://codereview.chromium.org/138353002

git-svn-id: http://skia.googlecode.com/svn/trunk@13063 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoremove file added accidentally
commit-bot@chromium.org [Tue, 14 Jan 2014 13:51:14 +0000 (13:51 +0000)]
remove file added accidentally

TBR=reed@google.com
NOTREECHECKS=true
NOTRY=true

Author: caryclark@google.com

Review URL: https://codereview.chromium.org/131053006

git-svn-id: http://skia.googlecode.com/svn/trunk@13062 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoSanitizing source files in Housekeeper-Nightly
skia.committer@gmail.com [Tue, 14 Jan 2014 07:01:42 +0000 (07:01 +0000)]
Sanitizing source files in Housekeeper-Nightly

git-svn-id: http://skia.googlecode.com/svn/trunk@13061 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agomake ignored-tests.txt specify full test name, not partial name
commit-bot@chromium.org [Tue, 14 Jan 2014 02:54:11 +0000 (02:54 +0000)]
make ignored-tests.txt specify full test name, not partial name

BUG=skia:2022
R=bsalomon@google.com

Author: epoger@google.com

Review URL: https://codereview.chromium.org/136883006

git-svn-id: http://skia.googlecode.com/svn/trunk@13060 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRemove drawPosTextOnPath() from SkDevice API.
tfarina@chromium.org [Mon, 13 Jan 2014 21:42:25 +0000 (21:42 +0000)]
Remove drawPosTextOnPath() from SkDevice API.

All clients were updated including Chrome, so we can remove this now
from Skia.

This was committed on Chromium at
https://src.chromium.org/viewvc/chrome?view=rev&revision=244287

BUG=skia:1834
R=reed@google.com

Review URL: https://codereview.chromium.org/135583002

git-svn-id: http://skia.googlecode.com/svn/trunk@13059 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRevert of https://codereview.chromium.org/132643007/
commit-bot@chromium.org [Mon, 13 Jan 2014 20:39:14 +0000 (20:39 +0000)]
Revert of https://codereview.chromium.org/132643007/
Reason for revert: broken tests

TBR=
NOTREECHECKS=true
NOTRY=true
BUG=

Author: reed@google.com

Review URL: https://codereview.chromium.org/134843008

git-svn-id: http://skia.googlecode.com/svn/trunk@13058 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRevert "Revert of https://codereview.chromium.org/129423002/"
reed@google.com [Mon, 13 Jan 2014 20:28:50 +0000 (20:28 +0000)]
Revert "Revert of https://codereview.chromium.org/129423002/"

fixed new api references in unit tests

This reverts commit ffc0058e1fbcbd69617e1f41b2dce5b5765ff99e.

BUG=

Review URL: https://codereview.chromium.org/132643007

git-svn-id: http://skia.googlecode.com/svn/trunk@13057 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRevert of https://codereview.chromium.org/129423002/
commit-bot@chromium.org [Mon, 13 Jan 2014 20:23:48 +0000 (20:23 +0000)]
Revert of https://codereview.chromium.org/129423002/
Reason for revert: broke tests (compile)

R=scroggo@google.com, halcanary@google.com
TBR=halcanary@google.com, scroggo@google.com
NOTREECHECKS=true
NOTRY=true
BUG=

Author: reed@google.com

Review URL: https://codereview.chromium.org/137133003

git-svn-id: http://skia.googlecode.com/svn/trunk@13056 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoadd SkBitmap::installPixelRef()
reed@google.com [Mon, 13 Jan 2014 20:17:58 +0000 (20:17 +0000)]
add SkBitmap::installPixelRef()

BUG=
R=scroggo@google.com

Review URL: https://codereview.chromium.org/129423002

git-svn-id: http://skia.googlecode.com/svn/trunk@13055 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoAdd no-op NULL picture to bench_record.
commit-bot@chromium.org [Mon, 13 Jan 2014 19:09:36 +0000 (19:09 +0000)]
Add no-op NULL picture to bench_record.

Defaults of out/Release/bench_record are still fine; NULL's not too far off
from stable at 900 loops.  But for focus on NULL, I'd recommend
something like out/Release/bench_record --loops 100000 --skps.

Also: fail on failures, print 2 significant digits instead of fixed 4.2.

BUG=skia:1850
R=tomhudson@chromium.org

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/136873002

git-svn-id: http://skia.googlecode.com/svn/trunk@13054 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoMisc rebaselines related to recent GPU blur changes.
bsalomon@google.com [Mon, 13 Jan 2014 19:06:10 +0000 (19:06 +0000)]
Misc rebaselines related to recent GPU blur changes.

Review URL: https://codereview.chromium.org/130523003

git-svn-id: http://skia.googlecode.com/svn/trunk@13053 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoAdd handling alpha values in RRGGBBAA formatted styles.
commit-bot@chromium.org [Mon, 13 Jan 2014 18:28:21 +0000 (18:28 +0000)]
Add handling alpha values in RRGGBBAA formatted styles.

BUG=
R=robertphillips@google.com

Author: jcgregorio@google.com

Review URL: https://codereview.chromium.org/136553006

git-svn-id: http://skia.googlecode.com/svn/trunk@13052 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoAllocate memory in SkTDynamicHash on first use.
commit-bot@chromium.org [Mon, 13 Jan 2014 18:28:14 +0000 (18:28 +0000)]
Allocate memory in SkTDynamicHash on first use.

This eliminates any dynamic allocation for hash tables that are never used.
This helps SkPicture, where some tables (SkPaint) are almost always used, but
some rarely (SkMatrix) or never (SkRegion).

This also removes the (as yet unimportant) ability for the hash table to
shrink.  This makes resizing harder to reason about, so I'd like to leave it
out until we see a need.

BUG=skia:1850
R=tomhudson@chromium.org, reed@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/136403004

git-svn-id: http://skia.googlecode.com/svn/trunk@13051 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoGM: run mode comparisons for all raster configs.
commit-bot@chromium.org [Mon, 13 Jan 2014 18:27:59 +0000 (18:27 +0000)]
GM: run mode comparisons for all raster configs.

GM works like this today:
  - Test all enabled configs directly
  - Render into 8888
  - Test all enabled modes against that 8888.

This change makes it do this:
  - Test all enabled configs directly
    - for each config, if it's raster, test all enabled modes against its output.

Upshot is, we now check that picture draws identically to raster for 565 too, not just 8888.  optimizations-565 is correctly failing.

BUG=skia:1994
R=epoger@google.com, reed@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/129203002

git-svn-id: http://skia.googlecode.com/svn/trunk@13050 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoremove _gpu suffix from tests in ignored-tests.txt that are changed on all gpu config...
bsalomon@google.com [Mon, 13 Jan 2014 18:26:14 +0000 (18:26 +0000)]
remove _gpu suffix from tests in ignored-tests.txt that are changed on all gpu configs (msaa, ...).

Review URL: https://codereview.chromium.org/136713005

git-svn-id: http://skia.googlecode.com/svn/trunk@13049 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRebaseline NVPR roundrects images
robertphillips@google.com [Mon, 13 Jan 2014 17:15:33 +0000 (17:15 +0000)]
Rebaseline NVPR roundrects images

https://codereview.chromium.org/136753004/

git-svn-id: http://skia.googlecode.com/svn/trunk@13048 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoUse correct matrix to adjust blur radius for gpu
commit-bot@chromium.org [Mon, 13 Jan 2014 16:39:39 +0000 (16:39 +0000)]
Use correct matrix to adjust blur radius for gpu

BUG=skia:1998
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/134973002

git-svn-id: http://skia.googlecode.com/svn/trunk@13047 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRemove redundant auto rt object in skpaint2grpaintshader function.
commit-bot@chromium.org [Mon, 13 Jan 2014 15:57:05 +0000 (15:57 +0000)]
Remove redundant auto rt object in skpaint2grpaintshader function.

R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/133763006

git-svn-id: http://skia.googlecode.com/svn/trunk@13046 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoRemove rects GM from ignored-tests.txt and rebaseline
bsalomon@google.com [Mon, 13 Jan 2014 15:06:54 +0000 (15:06 +0000)]
Remove rects GM from ignored-tests.txt and rebaseline

BUG=skia:1882

Review URL: https://codereview.chromium.org/136803002

git-svn-id: http://skia.googlecode.com/svn/trunk@13045 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoAdd CONTRIBUTING file
borenet@google.com [Mon, 13 Jan 2014 15:06:26 +0000 (15:06 +0000)]
Add CONTRIBUTING file

This will help to direct those who stumble upon the new github mirror
to the proper channels for contributing code.

BUG=
R=jcgregorio@google.com

Review URL: https://codereview.chromium.org/134663003

git-svn-id: http://skia.googlecode.com/svn/trunk@13044 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoremove (now dead code) for SK_SUPPORT_LEGACY_COLORTYPE and SK_SUPPORT_LEGACY_ALPHATYPE
reed@google.com [Mon, 13 Jan 2014 15:04:50 +0000 (15:04 +0000)]
remove (now dead code) for SK_SUPPORT_LEGACY_COLORTYPE and SK_SUPPORT_LEGACY_ALPHATYPE

BUG=

Review URL: https://codereview.chromium.org/135033002

git-svn-id: http://skia.googlecode.com/svn/trunk@13043 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoremove remaining references to Sk64 (obsolete)
reed@google.com [Mon, 13 Jan 2014 14:53:55 +0000 (14:53 +0000)]
remove remaining references to Sk64 (obsolete)

BUG=

Review URL: https://codereview.chromium.org/136673002

git-svn-id: http://skia.googlecode.com/svn/trunk@13042 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoHandle drawBitmapRect src_rect->dst_rect mapping as a local matrix rather than view...
commit-bot@chromium.org [Mon, 13 Jan 2014 14:47:00 +0000 (14:47 +0000)]
Handle drawBitmapRect src_rect->dst_rect mapping as a local matrix rather than view matrix when there is a mask filter.

BUG=skia:1998
R=robertphillips@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/131323004

git-svn-id: http://skia.googlecode.com/svn/trunk@13041 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoYet another patch for r13038
robertphillips@google.com [Mon, 13 Jan 2014 14:13:46 +0000 (14:13 +0000)]
Yet another patch for r13038

git-svn-id: http://skia.googlecode.com/svn/trunk@13040 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoFix for r13038
robertphillips@google.com [Mon, 13 Jan 2014 13:46:45 +0000 (13:46 +0000)]
Fix for r13038

git-svn-id: http://skia.googlecode.com/svn/trunk@13039 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoPull in Chromium's version of GatherPixelRefs
robertphillips@google.com [Mon, 13 Jan 2014 13:33:26 +0000 (13:33 +0000)]
Pull in Chromium's version of GatherPixelRefs

https://codereview.chromium.org/134473002/

git-svn-id: http://skia.googlecode.com/svn/trunk@13038 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoAdd bench_record.
commit-bot@chromium.org [Mon, 13 Jan 2014 12:03:47 +0000 (12:03 +0000)]
Add bench_record.

I got fed up trying to figure out how best to measure recording cost with bench_pictures, render_pictures, etc, so I wrote a new tool.

Features welcome.

BUG=
R=tomhudson@google.com

Author: mtklein@google.com

Review URL: https://codereview.chromium.org/132573002

git-svn-id: http://skia.googlecode.com/svn/trunk@13037 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoNEON fast path for box blur
commit-bot@chromium.org [Mon, 13 Jan 2014 08:16:45 +0000 (08:16 +0000)]
NEON fast path for box blur

Calculate 8 channels in parallel by using 16-bits to store each channel. Due to the limitation of VQRDMULH, (int16 * int16 * 2 + 0x8000) >> 16, the fast path can only support kernelSize < 128.
8 significant bits are kept at least in each stage, the final error should less-equal than 1.

Pre-fetching memory for X-direction read. In fact pre-fetching memory doesn't help much for Y direction read, since it is a waste to load a cache line for only read 8 bytes.(I left it there to keep the symmetry. pre-fetch is cheap :) )

bench data on Nexus 10
before:
running bench [640 480]      blur_image_filter_large_10.00_10.00   8888:  cmsecs =  25081.48
running bench [640 480]      blur_image_filter_small_10.00_10.00   8888:  cmsecs =  25038.04
running bench [640 480]        blur_image_filter_large_1.00_1.00   8888:  cmsecs =  25209.04
running bench [640 480]        blur_image_filter_small_1.00_1.00   8888:  cmsecs =  24928.01
running bench [640 480]        blur_image_filter_large_0.00_1.00   8888:  cmsecs =  17160.98
running bench [640 480]       blur_image_filter_large_0.00_10.00   8888:  cmsecs =  17924.11
running bench [640 480]        blur_image_filter_large_1.00_0.00   8888:  cmsecs =  14609.19
running bench [640 480]       blur_image_filter_large_10.00_0.00   8888:  cmsecs =  14625.91

after:
running bench [640 480]      blur_image_filter_large_10.00_10.00   8888:  cmsecs =  14848.42
running bench [640 480]      blur_image_filter_small_10.00_10.00   8888:  cmsecs =  16037.29
running bench [640 480]        blur_image_filter_large_1.00_1.00   8888:  cmsecs =  14819.55
running bench [640 480]        blur_image_filter_small_1.00_1.00   8888:  cmsecs =  14563.69
running bench [640 480]        blur_image_filter_large_0.00_1.00   8888:  cmsecs =  11905.34
running bench [640 480]       blur_image_filter_large_0.00_10.00   8888:  cmsecs =  11883.85
running bench [640 480]        blur_image_filter_large_1.00_0.00   8888:  cmsecs =   9576.51
running bench [640 480]       blur_image_filter_large_10.00_0.00   8888:  cmsecs =   9793.84

BUG=
R=senorblanco@chromium.org, mtklein@google.com, reed@google.com, kevin.petit@arm.com, kevin.petit.arm@gmail.com

Author: zheng.xu@arm.com

Review URL: https://codereview.chromium.org/105893003

git-svn-id: http://skia.googlecode.com/svn/trunk@13036 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoCleanup: Remove unused includes from skia_unix.cpp
tfarina@chromium.org [Sat, 11 Jan 2014 16:42:10 +0000 (16:42 +0000)]
Cleanup: Remove unused includes from skia_unix.cpp

This also move gWindow into the main function.

BUG=None
TEST=SampleApp
R=mtklein@google.com

Review URL: https://codereview.chromium.org/133113002

git-svn-id: http://skia.googlecode.com/svn/trunk@13035 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agomerge multiple blurrect GMs into one GM and add versions
commit-bot@chromium.org [Fri, 10 Jan 2014 23:10:13 +0000 (23:10 +0000)]
merge multiple blurrect GMs into one GM and add versions

BUG=skia:1998
R=reed@google.com

Author: bsalomon@google.com

Review URL: https://codereview.chromium.org/134533006

git-svn-id: http://skia.googlecode.com/svn/trunk@13034 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoAdd a DEF_GPUTEST() macro.
tfarina@chromium.org [Fri, 10 Jan 2014 22:08:27 +0000 (22:08 +0000)]
Add a DEF_GPUTEST() macro.

This macro is similar to DEF_TEST() and simplifies the process of
setting up a GPU test.

BUG=skia:1952
TEST=tests
R=mtklein@google.com

Review URL: https://codereview.chromium.org/132293005

git-svn-id: http://skia.googlecode.com/svn/trunk@13033 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoremove (now unused) Sk64 -- just use int64_t
reed@google.com [Fri, 10 Jan 2014 22:05:41 +0000 (22:05 +0000)]
remove (now unused) Sk64 -- just use int64_t

BUG=

Review URL: https://codereview.chromium.org/134373006

git-svn-id: http://skia.googlecode.com/svn/trunk@13032 2bbb7eff-a529-9590-31e7-b0007b416f81

10 years agoJust use one version of the scripts in both the browser and in SkV8 by using feature...
commit-bot@chromium.org [Fri, 10 Jan 2014 21:33:01 +0000 (21:33 +0000)]
Just use one version of the scripts in both the browser and in SkV8 by using feature detection to determine if we are running in a browser, and if the platform supports the Path() object. Also add oval, console.log, and the snow example code.

BUG=
R=robertphillips@google.com

Author: jcgregorio@google.com

Review URL: https://codereview.chromium.org/132413002

git-svn-id: http://skia.googlecode.com/svn/trunk@13031 2bbb7eff-a529-9590-31e7-b0007b416f81