platform/upstream/libSkiaSharp.git
9 years agoAdd simd.md to document Skia SIMD code / plans.
mtklein [Thu, 10 Sep 2015 21:21:12 +0000 (14:21 -0700)]
Add simd.md to document Skia SIMD code / plans.

BUG=skia:4117
NOTRY=true
DOCS_PREVIEW= https://skia.org/dev/contrib/simd?cl=1330083002

Review URL: https://codereview.chromium.org/1330083002

9 years agoCorrect a possible free after use.
herb [Thu, 10 Sep 2015 21:16:12 +0000 (14:16 -0700)]
Correct a possible free after use.

This shows up in TSAN runs as a use-after-free warning.

BUG=skia:

Review URL: https://codereview.chromium.org/1333003002

9 years agoSkNx_shuffle
mtklein [Thu, 10 Sep 2015 21:16:07 +0000 (14:16 -0700)]
SkNx_shuffle

This allows us to express shuffles more directly in code while also giving us a
convenient point to platform-specify particular shuffles for particular types.

No specializations yet.  Everyone just uses the (pretty good) default option.

BUG=skia:

Review URL: https://codereview.chromium.org/1301413006

9 years agofix missing clipmaskmanager
Brian Salomon [Thu, 10 Sep 2015 18:40:20 +0000 (14:40 -0400)]
fix missing clipmaskmanager

Review URL: https://codereview.chromium.org/1330213004 .

9 years agomake a shallow-copy so we don't get racy trying to lock/unlock our private bitmap
reed [Thu, 10 Sep 2015 18:38:58 +0000 (11:38 -0700)]
make a shallow-copy so we don't get racy trying to lock/unlock our private bitmap

BUG= 529995
NOTREECHECKS=True

Review URL: https://codereview.chromium.org/1327313002

9 years agoFix texture creation on stencil format test code
egdaniel [Thu, 10 Sep 2015 18:28:22 +0000 (11:28 -0700)]
Fix texture creation on stencil format test code

BUG=skia:

Review URL: https://codereview.chromium.org/1332853003

9 years agoPort SkMatrix opts to SkOpts.
mtklein [Thu, 10 Sep 2015 18:18:31 +0000 (11:18 -0700)]
Port SkMatrix opts to SkOpts.

No changes to the code, just moved around.

This will have the effect of enabling vectorized code on ARMv7.
Should be no effect on ARMv8 or x86, which would have been vectorized already.

nanobench --match mappoints changes on Nexus 5 (ARMv7):

_affine: 132 -> 95
_scale: 118 -> 47
_trans: 60 -> 37

A teaser:
We should next look at the ABCD->BADC shuffle we've noted that we need in _affine.  A quick hack showed doing that optimally is another ~35% speedup on x86.  Got to figure out how to do it best on ARM though: that same quick hack was a 2x slowdown there.  Good reason to resurrect that SkNx_shuffle() CL!

(I believe the answers are vrev64q_f32(v) and _mm_shuffle_ps(v,v, _MM_SHUFFLE(2,3,0,1), but we should probably find out in another CL.)

BUG=skia:4117

Review URL: https://codereview.chromium.org/1320673014

9 years agoRemove GrClipTarget
bsalomon [Thu, 10 Sep 2015 18:16:35 +0000 (11:16 -0700)]
Remove GrClipTarget

Review URL: https://codereview.chromium.org/1330353006

9 years agoSimplify installation of pipeling into GrDrawBatch in GrDrawTarget
bsalomon [Thu, 10 Sep 2015 18:10:50 +0000 (11:10 -0700)]
Simplify installation of pipeling into GrDrawBatch in GrDrawTarget

R=joshualitt@google.com

Review URL: https://codereview.chromium.org/1333943002

9 years agoLate creation of GrPathProcessor
joshualitt [Thu, 10 Sep 2015 18:00:51 +0000 (11:00 -0700)]
Late creation of GrPathProcessor

BUG=skia:

Review URL: https://codereview.chromium.org/1337513002

9 years agoCleanup GrDrawTarget now that all paths lead to GrBatch
bsalomon [Thu, 10 Sep 2015 17:42:55 +0000 (10:42 -0700)]
Cleanup GrDrawTarget now that all paths lead to GrBatch

Review URL: https://codereview.chromium.org/1315513008

9 years agoPort SkBlitRow::Color32 to SkOpts.
mtklein [Thu, 10 Sep 2015 17:38:02 +0000 (10:38 -0700)]
Port SkBlitRow::Color32 to SkOpts.

This was a pre-SkOpts attempt that we can bring under its wing now.

This should be a perf no-op, deo volente.

BUG=skia:4117

Review URL: https://codereview.chromium.org/1314863006

9 years agoDM: add --key and --properties to PNG description field.
mtklein [Thu, 10 Sep 2015 17:37:44 +0000 (10:37 -0700)]
DM: add --key and --properties to PNG description field.

This tags each image with the bot that created it.

BUG=skia:4322

Review URL: https://codereview.chromium.org/1332903002

9 years agoFix leak of path ranges
bsalomon [Thu, 10 Sep 2015 16:31:59 +0000 (09:31 -0700)]
Fix leak of path ranges

R=egdaniel@google.com

Review URL: https://codereview.chromium.org/1334753002

9 years agoCalculate pixel config and stencil fmt pairs once per pixel config.
egdaniel [Thu, 10 Sep 2015 15:37:20 +0000 (08:37 -0700)]
Calculate pixel config and stencil fmt pairs once per pixel config.

We use a temp FB and stencil buffer to test different stencil formats with
a given pixel config. We then keep a map from pixel config to desired stencil
format.

BUG=skia:

Review URL: https://codereview.chromium.org/1317443004

9 years agoAdd a mutex to GrContext::readSurfacePixels to protect against multiple CPU raster...
bsalomon [Thu, 10 Sep 2015 15:12:46 +0000 (08:12 -0700)]
Add a mutex to GrContext::readSurfacePixels to protect against multiple CPU raster threads accessing the same GrContext to read back GPU input data

BUG=chromium:524717

TBR=reed@google.com

Committed: https://skia.googlesource.com/skia/+/eb662bc407cec0585a821946fef123102cae64db

Review URL: https://codereview.chromium.org/1329313002

9 years agoThere's a set probability that a linear pipeline of random procs will be created...
wangyix [Thu, 10 Sep 2015 13:57:05 +0000 (06:57 -0700)]
There's a set probability that a linear pipeline of random procs will be created (old behavior), or a pipeline with a single proc tree (added behavior).

Had to move GrComposeEffect class definition from SkComposeShader.cpp to SkComposeShader.h so that GLProgramsTest can call GrComposeEffect::Create()

BUG=skia:4182

Review URL: https://codereview.chromium.org/1314923002

9 years agoInitialize subrun variables to make Valgrind happy
jvanverth [Thu, 10 Sep 2015 13:31:38 +0000 (06:31 -0700)]
Initialize subrun variables to make Valgrind happy

The check at line 1259 was failing in Valgrind, and in
the default constructor subrun.fMaskFormat is never inited.
Also set some other variables to avoid future problems.

Review URL: https://codereview.chromium.org/1315773005

9 years agoRevert of Add a mutex to GrContext::readSurfacePixels to protect against multiple...
bsalomon [Thu, 10 Sep 2015 01:57:49 +0000 (18:57 -0700)]
Revert of Add a mutex to GrContext::readSurfacePixels to protect against multiple CPU raster threads accessin… (patchset #1 id:1 of https://codereview.chromium.org/1329313002/ )

Reason for revert:
breaking the bots

Original issue's description:
> Add a mutex to GrContext::readSurfacePixels to protect against multiple CPU raster threads accessing the same GrContext to read back GPU input data
>
> BUG=chromium:524717
>
> TBR=reed@google.com
>
> Committed: https://skia.googlesource.com/skia/+/eb662bc407cec0585a821946fef123102cae64db

TBR=reed@google.com,reed@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=chromium:524717

Review URL: https://codereview.chromium.org/1334603002

9 years agoAdd a mutex to GrContext::readSurfacePixels to protect against multiple CPU raster...
bsalomon [Thu, 10 Sep 2015 01:05:03 +0000 (18:05 -0700)]
Add a mutex to GrContext::readSurfacePixels to protect against multiple CPU raster threads accessing the same GrContext to read back GPU input data

BUG=chromium:524717

TBR=reed@google.com

Review URL: https://codereview.chromium.org/1329313002

9 years agoSK_API on GrWrapTextureInBitmap
bsalomon [Wed, 9 Sep 2015 21:51:52 +0000 (14:51 -0700)]
SK_API on GrWrapTextureInBitmap

BUG=chromium:524717
TBR=reed@google.com

Review URL: https://codereview.chromium.org/1334563002

9 years agoRemove race.
herb [Wed, 9 Sep 2015 21:48:33 +0000 (14:48 -0700)]
Remove race.

BUG=skia:

Review URL: https://codereview.chromium.org/1327193003

9 years agofix leaked picture
reed [Wed, 9 Sep 2015 20:42:14 +0000 (13:42 -0700)]
fix leaked picture

BUG=skia:
TBR=

Review URL: https://codereview.chromium.org/1329143006

9 years agodon't peek w/o first acquiring the lock
reed [Wed, 9 Sep 2015 20:25:24 +0000 (13:25 -0700)]
don't peek w/o first acquiring the lock

BUG=skia:

Review URL: https://codereview.chromium.org/1324503005

9 years agoReland of more zero-length changes for svg compatibility (patchset #1 id:1 of https...
caryclark [Wed, 9 Sep 2015 20:20:49 +0000 (13:20 -0700)]
Reland of more zero-length changes for svg compatibility (patchset #1 id:1 of https://codereview.chromium.org/1334543002/ )

Reason for revert:
DEPS should be fixed now

Original issue's description:
> Revert of more zero-length changes for svg compatibility (patchset #6 id:100001 of https://codereview.chromium.org/1330623003/ )
>
> Reason for revert:
> breaks DEPS
>
> Original issue's description:
> > more zero-length changes for svg compatibility
> >
> > If a path contains a moveTo followed by a line or curve,
> > even if the line or curve has no length, SVG expects
> > the end caps to draw if the cap style is round or square.
> >
> > Fredrik Söderquist attached a patch to the chrome bug
> > (slightly modified here) that fixes layout test failures
> > resulting from deleting special-case code in SVG
> > dealing with zero-length path segments.
> >
> > R=reed@google.com,fs@opera.com
> > BUG=22974
> >
> > Committed: https://skia.googlesource.com/skia/+/62fb1ba1786863e545c89839b5706ad5151cec15
>
> TBR=fs@opera.com,reed@google.com
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
> BUG=22974
>
> Committed: https://skia.googlesource.com/skia/+/5ca4fa3846067a47e88d35ace895df3ebe3ec2a5

TBR=fs@opera.com,reed@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=22974

Review URL: https://codereview.chromium.org/1314833004

9 years agoBrian requested a new namespace for factories that create fragment processors that...
wangyix [Wed, 9 Sep 2015 19:58:32 +0000 (12:58 -0700)]
Brian requested a new namespace for factories that create fragment processors that combine 2 things with a xfermode.  This way, we can change the implementation of these factories in the future to not use GrComposeEffect if desired.

BUG=skia:4182

Review URL: https://codereview.chromium.org/1306163005

9 years agoParallel cache.
herb [Wed, 9 Sep 2015 19:16:09 +0000 (12:16 -0700)]
Parallel cache.

TBR=reed@google.com

BUG=skia:1330

Committed: https://skia.googlesource.com/skia/+/6f2a486040cb25465990196c229feb47e668e87f

Review URL: https://codereview.chromium.org/1264103003

9 years agoadd picture-image variant
reed [Wed, 9 Sep 2015 18:29:09 +0000 (11:29 -0700)]
add picture-image variant

BUG=skia:

Review URL: https://codereview.chromium.org/1329283002

9 years agoMimic Chrome better in our GYPs.
mtklein [Wed, 9 Sep 2015 17:46:20 +0000 (10:46 -0700)]
Mimic Chrome better in our GYPs.

To match Chrome, make sure we've disabled thread-safe statics, RTTI, and exceptions.  Linux needed -fno-threadsafe-statics, Mac needed all three.

Nothing important triggered this CL.  I just got confused when I saw exception handling (calls to delete, stack unwinding) in some generated code on my laptop.

BUG=skia:

Review URL: https://codereview.chromium.org/1323533003

9 years agoswitch over to using images
reed [Wed, 9 Sep 2015 17:45:36 +0000 (10:45 -0700)]
switch over to using images

BUG=skia:
TBR=

Review URL: https://codereview.chromium.org/1313543004

9 years agoSpecialize SkOncePtr<T[]>.
mtklein [Wed, 9 Sep 2015 17:00:22 +0000 (10:00 -0700)]
Specialize SkOncePtr<T[]>.

SkOncePtr<T[]> is identical to SkOncePtr<T> except we'll default to delete[]
for cleanup.

CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Debug-ASAN-Trybot

BUG=skia:

No public API changes.
TBR=reed@google.com

Review URL: https://codereview.chromium.org/1311893010

9 years agoMake GrProcessorDataManager a noop
joshualitt [Wed, 9 Sep 2015 17:00:12 +0000 (10:00 -0700)]
Make GrProcessorDataManager a noop

TBR=bsalomon@google.com
BUG=skia:

Review URL: https://codereview.chromium.org/1323963003

9 years agoGrPathRangeBatch
bsalomon [Wed, 9 Sep 2015 16:48:06 +0000 (09:48 -0700)]
GrPathRangeBatch

BUG=skia:

Review URL: https://codereview.chromium.org/1315563003

9 years agoRevert of more zero-length changes for svg compatibility (patchset #6 id:100001 of...
caryclark [Wed, 9 Sep 2015 16:34:22 +0000 (09:34 -0700)]
Revert of more zero-length changes for svg compatibility (patchset #6 id:100001 of https://codereview.chromium.org/1330623003/ )

Reason for revert:
breaks DEPS

Original issue's description:
> more zero-length changes for svg compatibility
>
> If a path contains a moveTo followed by a line or curve,
> even if the line or curve has no length, SVG expects
> the end caps to draw if the cap style is round or square.
>
> Fredrik Söderquist attached a patch to the chrome bug
> (slightly modified here) that fixes layout test failures
> resulting from deleting special-case code in SVG
> dealing with zero-length path segments.
>
> R=reed@google.com,fs@opera.com
> BUG=22974
>
> Committed: https://skia.googlesource.com/skia/+/62fb1ba1786863e545c89839b5706ad5151cec15

TBR=fs@opera.com,reed@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=22974

Review URL: https://codereview.chromium.org/1334543002

9 years agoPort uses of SkLazyPtr to SkOncePtr.
mtklein [Wed, 9 Sep 2015 16:09:53 +0000 (09:09 -0700)]
Port uses of SkLazyPtr to SkOncePtr.

This gives SkOncePtr a non-trivial destructor that uses std::default_delete
by default.  This is overrideable, as seen in SkColorTable.

SK_DECLARE_STATIC_ONCE_PTR still just leaves its pointers hanging at EOP.

BUG=skia:

No public API changes.
TBR=reed@google.com

Committed: https://skia.googlesource.com/skia/+/a1254acdb344174e761f5061c820559dab64a74c

Review URL: https://codereview.chromium.org/1322933005

9 years agowire up incNumDraws
joshualitt [Wed, 9 Sep 2015 15:29:47 +0000 (08:29 -0700)]
wire up incNumDraws

BUG=skia:

Review URL: https://codereview.chromium.org/1333623003

9 years agoGM: replace boilerplate with macros
halcanary [Wed, 9 Sep 2015 15:16:41 +0000 (08:16 -0700)]
GM: replace boilerplate with macros

I have verified locally that nothing draws differently.

Motivation:
*   SK_SIMPLE_GM makes it easier to write a GM.
*   Reducing 1100 lines of code makes maintenance easier.
*   Simple GMs are easy to convert to Fiddles.

Review URL: https://codereview.chromium.org/1333553002

9 years agoDetach cache in DF routines rather than passing it in.
jvanverth [Wed, 9 Sep 2015 15:05:12 +0000 (08:05 -0700)]
Detach cache in DF routines rather than passing it in.

This removes nesting of cache accesses, which avoids a possible
deadlock condition when the caches are the same.

Review URL: https://codereview.chromium.org/1310513008

9 years agoMake SkGraphics::Term a no-op, stop calling it.
mtklein [Wed, 9 Sep 2015 14:35:42 +0000 (07:35 -0700)]
Make SkGraphics::Term a no-op, stop calling it.

I'd remove it entirely but Android is calling it explicitly.

BUG=skia:4259

Committed: https://skia.googlesource.com/skia/+/925979f733fe8e70d84627147dee04d030423349

Review URL: https://codereview.chromium.org/1329853005

9 years agoRevert of Make SkGraphics::Term a no-op, stop calling it. (patchset #2 id:20001 of...
mtklein [Wed, 9 Sep 2015 14:22:09 +0000 (07:22 -0700)]
Revert of Make SkGraphics::Term a no-op, stop calling it. (patchset #2 id:20001 of https://codereview.chromium.org/1329853005/ )

Reason for revert:
SK_ATTR_DEPRECATED is meaningful to Android.  Don't use it.

Original issue's description:
> Make SkGraphics::Term a no-op, stop calling it.
>
> I'd remove it entirely but Android is calling it explicitly.
>
> BUG=skia:4259
>
> Committed: https://skia.googlesource.com/skia/+/925979f733fe8e70d84627147dee04d030423349

TBR=reed@google.com,scroggo@google.com,mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:4259

Review URL: https://codereview.chromium.org/1314483006

9 years agoRevert of Port uses of SkLazyPtr to SkOncePtr. (patchset #7 id:110001 of https:/...
mtklein [Wed, 9 Sep 2015 14:10:42 +0000 (07:10 -0700)]
Revert of Port uses of SkLazyPtr to SkOncePtr. (patchset #7 id:110001 of https://codereview.chromium.org/1322933005/ )

Reason for revert:
Breaks Chrome roll.

obj/skia/ext/skia_chrome.skia_memory_dump_provider.o
does not have -I include/private on its include path, but transitively includes SkMessageBus.h.

Original issue's description:
> Port uses of SkLazyPtr to SkOncePtr.
>
> This gives SkOncePtr a non-trivial destructor that uses std::default_delete
> by default.  This is overrideable, as seen in SkColorTable.
>
> SK_DECLARE_STATIC_ONCE_PTR still just leaves its pointers hanging at EOP.
>
> BUG=skia:
>
> No public API changes.
> TBR=reed@google.com
>
> Committed: https://skia.googlesource.com/skia/+/a1254acdb344174e761f5061c820559dab64a74c

TBR=herb@google.com,mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Review URL: https://codereview.chromium.org/1334523002

9 years agomore zero-length changes for svg compatibility
caryclark [Wed, 9 Sep 2015 14:04:33 +0000 (07:04 -0700)]
more zero-length changes for svg compatibility

If a path contains a moveTo followed by a line or curve,
even if the line or curve has no length, SVG expects
the end caps to draw if the cap style is round or square.

Fredrik Söderquist attached a patch to the chrome bug
(slightly modified here) that fixes layout test failures
resulting from deleting special-case code in SVG
dealing with zero-length path segments.

R=reed@google.com,fs@opera.com
BUG=22974

Review URL: https://codereview.chromium.org/1330623003

9 years agoPort uses of SkLazyPtr to SkOncePtr.
mtklein [Wed, 9 Sep 2015 13:48:29 +0000 (06:48 -0700)]
Port uses of SkLazyPtr to SkOncePtr.

This gives SkOncePtr a non-trivial destructor that uses std::default_delete
by default.  This is overrideable, as seen in SkColorTable.

SK_DECLARE_STATIC_ONCE_PTR still just leaves its pointers hanging at EOP.

BUG=skia:

No public API changes.
TBR=reed@google.com

Review URL: https://codereview.chromium.org/1322933005

9 years agoProvides various implementations of Android's SkBitmapRegionDecoder.
msarett [Tue, 8 Sep 2015 22:35:32 +0000 (15:35 -0700)]
Provides various implementations of Android's SkBitmapRegionDecoder.

Implements testing in DM for these implementations.

nanobench testing will follow after this.

TBR=scroggo
BUG=skia:

Committed: https://skia.googlesource.com/skia/+/76f755e6d54a32f9887ad254ce59a3a62f28bde4

Review URL: https://codereview.chromium.org/1288963002

9 years agoAdded TestCreate for SkComposeShader; will pick two random child procs that don...
wangyix [Tue, 8 Sep 2015 22:23:34 +0000 (15:23 -0700)]
Added TestCreate for SkComposeShader;  will pick two random child procs that don't have children of their own.  This prevents creating an arbitrarily large tree of procs.  Also, it will choose a random coefficient mode for the xfermode.

BUG=skia:4182

Review URL: https://codereview.chromium.org/1306163002

9 years agoskia: Initialize value passed to GetShaderiv in GLBench
hendrikw [Tue, 8 Sep 2015 22:18:56 +0000 (15:18 -0700)]
skia: Initialize value passed to GetShaderiv in GLBench

The command buffer's GetShaderiv and GetProgramiv code checks
that the success value passed in is either -1 or 0.

Review URL: https://codereview.chromium.org/1318143004

9 years agoMake SkGraphics::Term a no-op, stop calling it.
mtklein [Tue, 8 Sep 2015 22:18:21 +0000 (15:18 -0700)]
Make SkGraphics::Term a no-op, stop calling it.

I'd remove it entirely but Android is calling it explicitly.

BUG=skia:4259

Review URL: https://codereview.chromium.org/1329853005

9 years agoAdd number of draws to GrGpu stats
joshualitt [Tue, 8 Sep 2015 20:42:05 +0000 (13:42 -0700)]
Add number of draws to GrGpu stats

BUG=skia:

Review URL: https://codereview.chromium.org/1333543002

9 years agoUpdate documentation of colors on Status.
benjaminwagner [Tue, 8 Sep 2015 19:51:23 +0000 (12:51 -0700)]
Update documentation of colors on Status.

BUG=skia:
NOTRY=true
DOCS_PREVIEW= https://skia.org/?cl=1330063002

Review URL: https://codereview.chromium.org/1330063002

9 years agoRevert[2] of "switch to isABitmap, deprecate SK_SUPPORT_LEGACY_SHADERBITMAPTYPE"
reed [Tue, 8 Sep 2015 18:02:04 +0000 (11:02 -0700)]
Revert[2] of "switch to isABitmap, deprecate SK_SUPPORT_LEGACY_SHADERBITMAPTYPE"

master-skia has been updated to use isABitmap

This reverts commit ff390c9bdd852405d9dc0fd5e384b1f935d8df08.

BUG=skia:

Review URL: https://codereview.chromium.org/1310573008

9 years agoRe-enable nvpr text rendering in DM.
bsalomon [Tue, 8 Sep 2015 17:26:51 +0000 (10:26 -0700)]
Re-enable nvpr text rendering in DM.

Review URL: https://codereview.chromium.org/1306733007

9 years agoemitChild() used to generate a mangled outputColor based on the parent's outputColor...
wangyix [Tue, 8 Sep 2015 15:41:51 +0000 (08:41 -0700)]
emitChild() used to generate a mangled outputColor based on the parent's outputColor; now it just accepts an outputColor string.  It's now up to the programmer to declare outputColors if needed before emitting child code.

BUG=skia:4182

Review URL: https://codereview.chromium.org/1321253003

9 years agorefactor parts of SkGr.cpp for use by SkImages
reed [Tue, 8 Sep 2015 15:37:36 +0000 (08:37 -0700)]
refactor parts of SkGr.cpp for use by SkImages

BUG=skia:

Review URL: https://codereview.chromium.org/1315353006

9 years agoPort SkLazyFnPtr users to SkOncePtr.
mtklein [Tue, 8 Sep 2015 15:19:33 +0000 (08:19 -0700)]
Port SkLazyFnPtr users to SkOncePtr.

First baby step on road to replacing SkLazyFnPtr, SkLazyPtr, and SkOnce.

One of those very unusual times you get to use a function type,
not a function pointer type!

BUG=skia:

Review URL: https://codereview.chromium.org/1317263007

9 years agoComments: note existance of SkImage::newShader next to CreateBitmapShader
halcanary [Tue, 8 Sep 2015 14:12:25 +0000 (07:12 -0700)]
Comments: note existance of SkImage::newShader next to CreateBitmapShader

Review URL: https://codereview.chromium.org/1329083004

9 years agoCreate module system for VisualBench
joshualitt [Tue, 8 Sep 2015 14:08:11 +0000 (07:08 -0700)]
Create module system for VisualBench

The newly created VisualLightweightBenchModule is just the old VisualBench.cpp, but gutted to only include timing code.

Future CLs will harden this abstraction, but for this CL the module owns a backpointer to VisualBench.cpp for a couple of calls.

BUG=skia:

Review URL: https://codereview.chromium.org/1304083007

9 years agoRevert of Provides multiple implementations of Android's SkBitmapRegionDecoder (patch...
msarett [Tue, 8 Sep 2015 13:59:48 +0000 (06:59 -0700)]
Revert of Provides multiple implementations of Android's SkBitmapRegionDecoder (patchset #16 id:360001 of https://codereview.chromium.org/1288963002/ )

Reason for revert:
Breaking Android bots
Bad use of the utils folder

Original issue's description:
> Provides various implementations of Android's SkBitmapRegionDecoder.
>
> Implements testing in DM for these implementations.
>
> nanobench testing will follow after this.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/76f755e6d54a32f9887ad254ce59a3a62f28bde4

TBR=scroggo@google.com,reed@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Review URL: https://codereview.chromium.org/1322773004

9 years agoRevert of switch to isABitmap, deprecate SK_SUPPORT_LEGACY_SHADERBITMAPTYPE (patchset...
scroggo [Tue, 8 Sep 2015 13:24:08 +0000 (06:24 -0700)]
Revert of switch to isABitmap, deprecate SK_SUPPORT_LEGACY_SHADERBITMAPTYPE (patchset #3 id:40001 of https://codereview.chromium.org/1311963007/ )

Reason for revert:
Broke the Android canary:

  https://internal.skia.org/builders/crimson-cherry-474438/builds/124

Original issue's description:
> switch to isABitmap, deprecate SK_SUPPORT_LEGACY_SHADERBITMAPTYPE
>
> BUG=skia:
> TBR=
>
> Committed: https://skia.googlesource.com/skia/+/2d126b5c45e65a67a9945afa9294038a8eb3f2c8

TBR=reed@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Review URL: https://codereview.chromium.org/1309943004

9 years agoskia: use msaaConfigAttribList for SkOSWindow_win
hendrikw [Tue, 8 Sep 2015 12:53:23 +0000 (05:53 -0700)]
skia: use msaaConfigAttribList for SkOSWindow_win

eglChooseConfig was still being passed configAttribList, pass
msaaConfigAttribList instead

Review URL: https://codereview.chromium.org/1318663007

9 years agoAdd skstd::unique_ptr and use it.
bungeman [Mon, 7 Sep 2015 19:45:52 +0000 (12:45 -0700)]
Add skstd::unique_ptr and use it.

TBR=bsalomon@google.com
The one gpu include change is just to fix swap in implementation.

Review URL: https://codereview.chromium.org/1330503006

9 years agoswitch to isABitmap, deprecate SK_SUPPORT_LEGACY_SHADERBITMAPTYPE
reed [Mon, 7 Sep 2015 18:10:30 +0000 (11:10 -0700)]
switch to isABitmap, deprecate SK_SUPPORT_LEGACY_SHADERBITMAPTYPE

BUG=skia:
TBR=

Review URL: https://codereview.chromium.org/1311963007

9 years agoScanline decoding for gifs
msarett [Mon, 7 Sep 2015 15:54:01 +0000 (08:54 -0700)]
Scanline decoding for gifs

BUG=skia:

Committed: https://skia.googlesource.com/skia/+/e9c10b9121887e8c300bd41357461418e061984d

Review URL: https://codereview.chromium.org/1305123002

9 years agoUpdate SKP version
rmistry [Sun, 6 Sep 2015 07:20:11 +0000 (00:20 -0700)]
Update SKP version

Automatic commit by the RecreateSKPs bot.

TBR=
NO_MERGE_BUILDS

Review URL: https://codereview.chromium.org/1307693006

9 years agoRevert of Scanline decoding for gifs (patchset #15 id:380001 of https://codereview...
jcgregorio [Sat, 5 Sep 2015 17:16:28 +0000 (10:16 -0700)]
Revert of Scanline decoding for gifs (patchset #15 id:380001 of https://codereview.chromium.org/1305123002/ )

Reason for revert:
Breaks the build with SkScanlineDecoder.h not found:

http://build.chromium.org/p/client.skia/builders/Linux%20Builder/builds/3722/steps/compile/logs/stdio

../../third_party/skia/src/utils/SkBitmapRegionCanvas.h:10:10: fatal error: 'SkScanlineDecoder.h' file not found
#include "SkScanlineDecoder.h"

Original issue's description:
> Scanline decoding for gifs
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/e9c10b9121887e8c300bd41357461418e061984d

TBR=scroggo@google.com,djsollen@google.com,msarett@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Review URL: https://codereview.chromium.org/1320273004

9 years agoRevert of Parallel cache - preliminary (patchset #22 id:420001 of https://codereview...
herb [Fri, 4 Sep 2015 21:19:45 +0000 (14:19 -0700)]
Revert of Parallel cache - preliminary  (patchset #22 id:420001 of https://codereview.chromium.org/1264103003/ )

Reason for revert:
Seems to freeze android devices.

Original issue's description:
> Parallel cache.
>
> TBR=reed@google.com
>
> BUG=skia:1330
>
> Committed: https://skia.googlesource.com/skia/+/6f2a486040cb25465990196c229feb47e668e87f

TBR=reed@google.com,mtklein@google.com,mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:1330

Review URL: https://codereview.chromium.org/1327703003

9 years agoScanline decoding for gifs
msarett [Fri, 4 Sep 2015 20:28:25 +0000 (13:28 -0700)]
Scanline decoding for gifs

BUG=skia:

Review URL: https://codereview.chromium.org/1305123002

9 years agoBot fix
msarett [Fri, 4 Sep 2015 20:12:55 +0000 (13:12 -0700)]
Bot fix

https://codereview.chromium.org/1311503008
https://uberchromegw.corp.google.com/i/client.skia/builders/Test-Ubuntu-Clang-GCE-CPU-AVX2-x86_64-Coverage/builds/1083/steps/build%20dm/logs/stdio

BUG=skia:

Review URL: https://codereview.chromium.org/1331463004

9 years agoProvides various implementations of Android's SkBitmapRegionDecoder.
msarett [Fri, 4 Sep 2015 20:00:49 +0000 (13:00 -0700)]
Provides various implementations of Android's SkBitmapRegionDecoder.

Implements testing in DM for these implementations.

nanobench testing will follow after this.

BUG=skia:

Review URL: https://codereview.chromium.org/1288963002

9 years agoHandle zero-length encoded images gracefully during deserialization
fmalita [Fri, 4 Sep 2015 18:36:39 +0000 (11:36 -0700)]
Handle zero-length encoded images gracefully during deserialization

Image encoding may fail during serialization, resulting in zero-length
encoded data in the SKP.

Instead of invalidating the stream (and preventing deserialization of
the whole picture) we can instantiate placeholder images.

BUG=skia:4285
R=reed@google.com,robertphillips@google.com

Review URL: https://codereview.chromium.org/1308273011

9 years agoIco bug fix for opaque embedded images
msarett [Fri, 4 Sep 2015 17:53:53 +0000 (10:53 -0700)]
Ico bug fix for opaque embedded images

BUG=skia:

Review URL: https://codereview.chromium.org/1327963004

9 years agoParallel cache.
herb [Fri, 4 Sep 2015 17:38:58 +0000 (10:38 -0700)]
Parallel cache.

TBR=reed@google.com

BUG=skia:1330

Review URL: https://codereview.chromium.org/1264103003

9 years agoSkOncePtr
mtklein [Fri, 4 Sep 2015 17:26:27 +0000 (10:26 -0700)]
SkOncePtr

New abstraction to replace all of: SkOnce, SkLazyPtr, SkLazyFnPtr.

BUG=skia:

Review URL: https://codereview.chromium.org/1301323006

9 years agodraw 3 moveTos followed by various others [moves, lines, closes]
reed [Fri, 4 Sep 2015 17:03:26 +0000 (10:03 -0700)]
draw 3 moveTos followed by various others [moves, lines, closes]

BUG=skia:

Review URL: https://codereview.chromium.org/1311503008

9 years agoremove duplicated sample (already have a gm)
reed [Fri, 4 Sep 2015 17:03:22 +0000 (10:03 -0700)]
remove duplicated sample (already have a gm)

BUG=skia:

Review URL: https://codereview.chromium.org/1317693011

9 years agoFix SkOneShotDiscardablePixelRef to return discardable segment.
ssid [Fri, 4 Sep 2015 15:29:24 +0000 (08:29 -0700)]
Fix SkOneShotDiscardablePixelRef to return discardable segment.

The SkOneShotDiscardablePixelRef was not overriding
diagnostic_only_getDiscardable method of SkPixelRef. This caused the
memory to be acocunted under malloc instead of discardable and caused
wrong accounting.

BUG=528295

Review URL: https://codereview.chromium.org/1304083006

9 years agoMinor tweak to visual bench timing
joshualitt [Fri, 4 Sep 2015 14:48:41 +0000 (07:48 -0700)]
Minor tweak to visual bench timing

BUG=skia:

Review URL: https://codereview.chromium.org/1310153005

9 years agoupdate gm test size to include new elements
caryclark [Fri, 4 Sep 2015 13:19:48 +0000 (06:19 -0700)]
update gm test size to include new elements

TBR=
BUG=skia:562

Review URL: https://codereview.chromium.org/1305983006

9 years agochange for issue 562
caryclark [Fri, 4 Sep 2015 12:20:32 +0000 (05:20 -0700)]
change for issue 562

Add a current version of the patch described in issue 562.

Of interest: the reference row, which is described in the
bug as looking correct draws differently with Ganesh.

TBR=senorblanco
BUG=skia:562

Review URL: https://codereview.chromium.org/1324023006

9 years agoremove unused fields from SkOpSegment
caryclark [Fri, 4 Sep 2015 12:00:15 +0000 (05:00 -0700)]
remove unused fields from SkOpSegment

TBR=reed@google.com

Review URL: https://codereview.chromium.org/1322413002

9 years agoCMAKE + install command
halcanary [Thu, 3 Sep 2015 21:23:17 +0000 (14:23 -0700)]
CMAKE + install command

NOTRY=true

Review URL: https://codereview.chromium.org/1325903004

9 years agoMinor code cleanup (left over from prior CL)
robertphillips [Thu, 3 Sep 2015 20:32:33 +0000 (13:32 -0700)]
Minor code cleanup (left over from prior CL)

This was requested in https://codereview.chromium.org/1309543012/ (Add new GM to directly exercise perspective bitmap drawing)

Review URL: https://codereview.chromium.org/1320673011

9 years agoClean up some lint complaints
robertphillips [Thu, 3 Sep 2015 19:58:48 +0000 (12:58 -0700)]
Clean up some lint complaints

BUG=skia:4287

Review URL: https://codereview.chromium.org/1327943003

9 years agoAppurify bot: don't use flags that visualbench doesn't support
borenet [Thu, 3 Sep 2015 19:31:26 +0000 (12:31 -0700)]
Appurify bot: don't use flags that visualbench doesn't support

Depends on https://codereview.chromium.org/1320423005/
BUG=skia:4277

Review URL: https://codereview.chromium.org/1310903006

9 years agoFix memory leak in build example
mdempsky [Thu, 3 Sep 2015 19:19:23 +0000 (12:19 -0700)]
Fix memory leak in build example

NOTRY=true
DOCS_PREVIEW= https://skia.org/?cl=1319703004

Review URL: https://codereview.chromium.org/1319703004

9 years agoadb_wait_for_device: Skip `adb wait-for-device`
borenet [Thu, 3 Sep 2015 17:41:07 +0000 (10:41 -0700)]
adb_wait_for_device: Skip `adb wait-for-device`

This should still wait for the device to be attached and finished
booting, but will prevent no-output timeouts due to
`adb wait-for-device` never finishing.

The effect should be that, when an Android device is disconnected, the
"wait for device" step will run forever, or until the device
reconnects. This should prevent a disconnected device from spreading
purple all over the status page.

BUG=skia:

Review URL: https://codereview.chromium.org/1315753010

9 years agoAdd some useful information to trooper doc.
benjaminwagner [Thu, 3 Sep 2015 16:06:30 +0000 (09:06 -0700)]
Add some useful information to trooper doc.

BUG=skia:
NOTRY=true
DOCS_PREVIEW= https://skia.org/?cl=1308253005

Review URL: https://codereview.chromium.org/1308253005

9 years agoWill try composing a linear gradient with a color bitmap as well as an alpha8 bitmap...
wangyix [Thu, 3 Sep 2015 15:32:22 +0000 (08:32 -0700)]
Will try composing a linear gradient with a color bitmap as well as an alpha8 bitmap to make sure the paint's color is being used/ignored in the right places.

BUG=skia:4182

Review URL: https://codereview.chromium.org/1311043008

9 years agoAdd new GM to directly exercise perspective bitmap drawing
robertphillips [Thu, 3 Sep 2015 15:32:08 +0000 (08:32 -0700)]
Add new GM to directly exercise perspective bitmap drawing

BUG=skia:3326

Review URL: https://codereview.chromium.org/1309543012

9 years agoMissing EncodeData(SkImageInfo...) in SkImageEncoder_empty
fmalita [Thu, 3 Sep 2015 14:47:22 +0000 (07:47 -0700)]
Missing EncodeData(SkImageInfo...) in SkImageEncoder_empty

TBR=reed@google.com,msarett@google.com

Review URL: https://codereview.chromium.org/1304243007

9 years agoAdd a SkPixelSerializer SkImage encode variant
fmalita [Thu, 3 Sep 2015 14:17:25 +0000 (07:17 -0700)]
Add a SkPixelSerializer SkImage encode variant

R=reed@google.com
BUG=skia:4285

Review URL: https://codereview.chromium.org/1310633006

9 years agoadd interactive xfer sample
reed [Thu, 3 Sep 2015 02:05:10 +0000 (19:05 -0700)]
add interactive xfer sample

BUG=skia:
TBR=

Review URL: https://codereview.chromium.org/1323223004

9 years agoTesting CQ committer list
rmistry [Wed, 2 Sep 2015 23:25:17 +0000 (16:25 -0700)]
Testing CQ committer list

BUG=skia:
NOTRY=true
TBR=

Review URL: https://codereview.chromium.org/1308163007

9 years agoFix bug in ico decodes
msarett [Wed, 2 Sep 2015 23:23:42 +0000 (16:23 -0700)]
Fix bug in ico decodes

Some icos have been looking a bit strange in Gold.

Looks like the cause was that we were decoding to kUnpremul
accidentally.

BUG=skia:

Review URL: https://codereview.chromium.org/1321913003

9 years agoMoved committer list to chrome-infra-auth and deleted it from the repo
sergiyb [Wed, 2 Sep 2015 20:37:54 +0000 (13:37 -0700)]
Moved committer list to chrome-infra-auth and deleted it from the repo

R=rmistry@chromium.org
BUG=chromium:511311

Review URL: https://codereview.chromium.org/1314153006

9 years agoRemove unwanted images in Gold
msarett [Wed, 2 Sep 2015 20:20:52 +0000 (13:20 -0700)]
Remove unwanted images in Gold

These extra outputs were caused by recent changes to
push_codec_srcs.
https://codereview.chromium.org/1327433003/

*** First, I would argue that we do not want to test
"native" modes (ex: codec, scanline, etc) to scales
that require sampling (ex: 0.1, 0.2, etc).  Right now,
we are trying to scale jpegs to 0.1, settling for 0.125
as the closest option, and then trying to compare the
0.125 scaled image to the actual 0.1 scaled image in
Gold.

*** Second, I messed up and caused our test setup to
try to decode to kIndex8 and kGray8 "always" instead
of only when it is recommended.  The bad effect of this
happens because we can decode jpegs to kGray8 even if
they are color images.  Right now in Gold, we have a
bunch of untriaged gray versions of color images.

The second issue would have been caught if we signaled
a fatal failure for invalid conversions.  Maybe we should
look into this now that 565 is supported everywhere?

BUG=skia:

Review URL: https://codereview.chromium.org/1314163007

9 years agoGive reason while file open failed.
bungeman [Wed, 2 Sep 2015 19:41:35 +0000 (12:41 -0700)]
Give reason while file open failed.

Review URL: https://codereview.chromium.org/1326743004

9 years agoAdd new surface flag
jvanverth [Wed, 2 Sep 2015 19:26:10 +0000 (12:26 -0700)]
Add new surface flag

TBR=bsalomon@google.com

Review URL: https://codereview.chromium.org/1320533005

9 years agoRevert of Minor cleanup in GrTessellatingPathRenderer. (patchset #1 id:1 of https...
senorblanco [Wed, 2 Sep 2015 16:05:17 +0000 (09:05 -0700)]
Revert of Minor cleanup in GrTessellatingPathRenderer. (patchset #1 id:1 of https://codereview.chromium.org/1322023002/ )

Reason for revert:
Patch is incorrect -- floating point errors can cause zero-area monotone polys to be skipped, resulting in a smaller vertex count than estimated.

Reproduce as follows:

out/Debug/SampleApp --slide Fuzzer --msaa 4

switch to GPU mode. Result:

../../src/gpu/batches/GrTessellatingPathRenderer.cpp:1500: failed assertion "static_cast<int>(end - verts) == vertexCount"

Original issue's description:
> Minor cleanup in GrTessellatingPathRenderer.
>
> Vertex counts are always exact, so don't bother handling the case
> where they're different. Just assert.
> Rename variables to reflect.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/9389b871c3b0f06bb34626cf9bdbfe0c93779327

TBR=bsalomon@google.com,senorblanco@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:

Review URL: https://codereview.chromium.org/1321723006

9 years agoRemove GrAddPathRenderers_default
joshualitt [Wed, 2 Sep 2015 15:42:16 +0000 (08:42 -0700)]
Remove GrAddPathRenderers_default

BUG=skia:

Review URL: https://codereview.chromium.org/1311673014

9 years agoReplace font name table access with standard Skia function.
caryclark [Wed, 2 Sep 2015 15:08:32 +0000 (08:08 -0700)]
Replace font name table access with standard Skia function.

Remove partial name detection.

Fix typo in custom deserializer to use the sk_ truncated font name.

Review URL: https://codereview.chromium.org/1327683002