wangyix [Tue, 4 Aug 2015 14:59:37 +0000 (07:59 -0700)]
Added registerChild; transforms, textures, glKey automatically handled.
BUG=skia:
Review URL: https://codereview.chromium.org/
1266633003
Stephen White [Tue, 4 Aug 2015 01:53:09 +0000 (21:53 -0400)]
Fix SkData leaks at GrResourceKey::setCustomData() call sites.
BUG=skia:
TBR=bsalomon@google.com
Review URL: https://codereview.chromium.org/
1270103003 .
mtklein [Tue, 4 Aug 2015 01:08:29 +0000 (18:08 -0700)]
Revert of SkCanvas::onDrawPicture() quick-reject (patchset #3 id:40001 of https://codereview.chromium.org/
1264133003/ )
Reason for revert:
webkit_unit_tests :(
Original issue's description:
> SkCanvas::onDrawPicture() quick-reject
>
> R=reed@google.com,mtklein@google.com
>
> Committed: https://skia.googlesource.com/skia/+/
48ed62b29d45e42b971aac8858da06781c93e6d7
TBR=reed@google.com,fmalita@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review URL: https://codereview.chromium.org/
1273433002
fmalita [Mon, 3 Aug 2015 22:30:35 +0000 (15:30 -0700)]
SkCanvas::onDrawPicture() quick-reject
R=reed@google.com,mtklein@google.com
Review URL: https://codereview.chromium.org/
1264133003
reed [Mon, 3 Aug 2015 21:44:32 +0000 (14:44 -0700)]
remove flags that control no code
BUG=skia:
Review URL: https://codereview.chromium.org/
1259653004
bsalomon [Mon, 3 Aug 2015 21:23:03 +0000 (14:23 -0700)]
Fix elliptical rrect clip shaders for large radii on devices with mediump
BUG=chromium:426217
Review URL: https://codereview.chromium.org/
1256353004
reed [Mon, 3 Aug 2015 21:18:04 +0000 (14:18 -0700)]
add bench for gpu-image -> cpu-surface
BUG=513695
TBR=
Review URL: https://codereview.chromium.org/
1261663004
Stephen White [Mon, 3 Aug 2015 20:30:23 +0000 (16:30 -0400)]
Fix resource cache test.
TBR=bsalomon@chromium.org
Review URL: https://codereview.chromium.org/
1257253007 .
senorblanco [Mon, 3 Aug 2015 20:04:03 +0000 (13:04 -0700)]
Implement caching of filled paths in the tessellated path renderer.
Paths are cached as tessellated triangle meshes in vertex buffers on the GPU. Stroked paths are not (yet) cached.
Paths containing no curved segments (linear paths) are reused at all scales. Paths containing curved segments are reused within a scale tolerance threshold.
In order to invalidate the cache when an SkPath is changed or deleted,
this required implementing genID change notification in SkPath. This is
modelled almost exactly on SkPixelRef::GenIDChangeListener.
However, It does not currently implement the check for unique genIDs,
so notifiers will fire when the first instance of an SkPathRef
using a given genID is destroyed.
Another caveat is that you cannot successfully add a change notifier
to an empty path, since it uses the "canonical" empty path which is
never modified or destroyed. For this reason, we prevent adding
listeners to it.
BUG=skia:4121,skia:4122, 497403
DOCS_PREVIEW= https://skia.org/?cl=
1114353004
Review URL: https://codereview.chromium.org/
1114353004
reed [Mon, 3 Aug 2015 18:45:34 +0000 (11:45 -0700)]
remove getdevice guard for android
BUG=skia:
NOTRY=True
Review URL: https://codereview.chromium.org/
1266363002
joshualitt [Mon, 3 Aug 2015 18:35:28 +0000 (11:35 -0700)]
Use new API everywhere for GrDefaultGeoProcFactory
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
801823b3d825d96baadeb9077c5ed1a8ada05cf2
Review URL: https://codereview.chromium.org/
1261083003
mtklein [Mon, 3 Aug 2015 17:20:28 +0000 (10:20 -0700)]
Missing overrides for Sinks in HWUI.
See crrev.com/
1263113002
BUG=skia:4138
Review URL: https://codereview.chromium.org/
1263143004
joshualitt [Mon, 3 Aug 2015 17:17:44 +0000 (10:17 -0700)]
Add abliity to set textblob cache budget to GrContext
BUG=skia:
Review URL: https://codereview.chromium.org/
1264283002
bsalomon [Mon, 3 Aug 2015 17:15:25 +0000 (10:15 -0700)]
Update assert to allow config conversion effect for all configs when not premul/unpremuling
TBR=robertphillips@google.com
Review URL: https://codereview.chromium.org/
1258763006
joshualitt [Mon, 3 Aug 2015 16:32:36 +0000 (09:32 -0700)]
Revert of Use new API everywhere for GrDefaultGeoProcFactory (patchset #5 id:80001 of https://codereview.chromium.org/
1261083003/)
Reason for revert:
breaking things
Original issue's description:
> Use new API everywhere for GrDefaultGeoProcFactory
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
801823b3d825d96baadeb9077c5ed1a8ada05cf2
TBR=robertphillips@google.com,bsalomon@google.com,joshualitt@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1264283004
joshualitt [Mon, 3 Aug 2015 16:03:33 +0000 (09:03 -0700)]
Use new API everywhere for GrDefaultGeoProcFactory
BUG=skia:
Review URL: https://codereview.chromium.org/
1261083003
Derek Sollenberger [Mon, 3 Aug 2015 15:41:53 +0000 (11:41 -0400)]
add missing comma
Review URL: https://codereview.chromium.org/
1268483003 .
bungeman [Mon, 3 Aug 2015 14:56:16 +0000 (07:56 -0700)]
Update NulCanvas so everything builds.
SkNulCanvas.h is out of date, so the 'all' target fails.
Review URL: https://codereview.chromium.org/
1270633004
halcanary [Mon, 3 Aug 2015 13:55:20 +0000 (06:55 -0700)]
Documentation: SkPathEffect
NOTRY=true
DOCS_PREVIEW= https://skia.org/user/api/skpaint?cl=
1269563008
Review URL: https://codereview.chromium.org/
1269563008
djsollen [Mon, 3 Aug 2015 13:03:06 +0000 (06:03 -0700)]
fix tests for android framework build
Review URL: https://codereview.chromium.org/
1263243003
joshualitt [Sat, 1 Aug 2015 17:33:40 +0000 (10:33 -0700)]
Add BW masks to random scaler context
TBR=bsalomon@google.com
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
a4a530614e9ba856aac3f16c29683e6a2a9153d9
Review URL: https://codereview.chromium.org/
1270793003
joshualitt [Sat, 1 Aug 2015 17:17:53 +0000 (10:17 -0700)]
Revert of Add BW masks to random scaler context (patchset #1 id:1 of https://codereview.chromium.org/
1270793003/)
Reason for revert:
breaks mac
Original issue's description:
> Add BW masks to random scaler context
>
> TBR=bsalomon@google.com
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
a4a530614e9ba856aac3f16c29683e6a2a9153d9
TBR=joshualitt@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1264843003
joshualitt [Sat, 1 Aug 2015 16:36:02 +0000 (09:36 -0700)]
Add BW masks to random scaler context
TBR=bsalomon@google.com
BUG=skia:
Review URL: https://codereview.chromium.org/
1270793003
joshualitt [Sat, 1 Aug 2015 14:33:41 +0000 (07:33 -0700)]
adding gm to use random scaler context
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
853336c532504b3436d7dcbf252419f00c79066d
Review URL: https://codereview.chromium.org/
1268853008
reed [Sat, 1 Aug 2015 14:03:20 +0000 (07:03 -0700)]
remove now-dead flags
BUG=skia:
Review URL: https://codereview.chromium.org/
1258963005
bungeman [Fri, 31 Jul 2015 22:46:03 +0000 (15:46 -0700)]
Fix no-gpu debugger.
Review URL: https://codereview.chromium.org/
1262723005
joshualitt [Fri, 31 Jul 2015 22:18:08 +0000 (15:18 -0700)]
Move strike to subrun in GrAtlasTextContext
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
77d89f7dd243a17452d3a5f16a98622993e6bdd9
Review URL: https://codereview.chromium.org/
1257253005
halcanary [Fri, 31 Jul 2015 22:16:23 +0000 (15:16 -0700)]
C API: remove dead code, simplify boilerplate
Review URL: https://codereview.chromium.org/
1261953006
joshualitt [Fri, 31 Jul 2015 22:10:31 +0000 (15:10 -0700)]
Revert of adding gm to use random scaler context (patchset #4 id:60001 of https://codereview.chromium.org/
1268853008/)
Reason for revert:
breaking bots
Original issue's description:
> adding gm to use random scaler context
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
853336c532504b3436d7dcbf252419f00c79066d
TBR=bsalomon@google.com,joshualitt@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1259033004
mtklein [Fri, 31 Jul 2015 21:47:25 +0000 (14:47 -0700)]
disable SkOpts on x86 iOS (simulator)
TBR=
BUG=skia:
Review URL: https://codereview.chromium.org/
1266443006
joshualitt [Fri, 31 Jul 2015 21:46:46 +0000 (14:46 -0700)]
adding gm to use random scaler context
BUG=skia:
Review URL: https://codereview.chromium.org/
1268853008
scroggo [Fri, 31 Jul 2015 21:29:26 +0000 (14:29 -0700)]
Fix straggling SK_VIRTUAL_CONSTRAINT_TYPEs
These were removed with crrev.com/
1239193002, but a few were missed
in files that are only build on Android framework.
Brings the Android build one step closer to building.
Review URL: https://codereview.chromium.org/
1268603003
joshualitt [Fri, 31 Jul 2015 21:07:50 +0000 (14:07 -0700)]
Revert of Move strike to subrun in GrAtlasTextContext (patchset #3 id:40001 of https://codereview.chromium.org/
1257253005/)
Reason for revert:
breaking bots
Original issue's description:
> Move strike to subrun in GrAtlasTextContext
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
77d89f7dd243a17452d3a5f16a98622993e6bdd9
TBR=bsalomon@google.com,robertphillips@google.com,jvanverth@google.com,joshualitt@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1266253002
mtklein [Fri, 31 Jul 2015 21:02:36 +0000 (14:02 -0700)]
Move SkOpts.h back to src/core.
The Chrome opts targets (sse2, ssse3, sse41, etc) don't have include/private on
their include path. This should unblock the roll.
TBR=reed@google.com
BUG=skia:4117
Review URL: https://codereview.chromium.org/
1268853007
bsalomon [Fri, 31 Jul 2015 20:59:30 +0000 (13:59 -0700)]
Some cleanup in GrTextureProvider and GrResourceProvider.
Review URL: https://codereview.chromium.org/
1261643004
joshualitt [Fri, 31 Jul 2015 20:58:27 +0000 (13:58 -0700)]
Move strike to subrun in GrAtlasTextContext
BUG=skia:
Review URL: https://codereview.chromium.org/
1257253005
halcanary [Fri, 31 Jul 2015 20:38:06 +0000 (13:38 -0700)]
C API: add sk_xfermode.h, impl, test
Review URL: https://codereview.chromium.org/
1242403005
bsalomon [Fri, 31 Jul 2015 19:15:50 +0000 (12:15 -0700)]
Speculative fix for http://crbug.com/515966
BUG=chromium:515966
Review URL: https://codereview.chromium.org/
1268723005
halcanary [Fri, 31 Jul 2015 19:02:18 +0000 (12:02 -0700)]
Documentation spelling error
BUG=skia:4157
NOTRY=true
TBR=
Review URL: https://codereview.chromium.org/
1267843002
halcanary [Fri, 31 Jul 2015 18:58:13 +0000 (11:58 -0700)]
Documentation: SkXfermode::Mode
BUG=skia:4157
NOTRY=true
DOCS_PREVIEW= https://skia.org/user/api/skpaint?cl=
1270643004
Review URL: https://codereview.chromium.org/
1270643004
joshualitt [Fri, 31 Jul 2015 18:55:29 +0000 (11:55 -0700)]
Another small fix to GrFontScaler
TBR=bsalomon@google.com
BUG=skia:
Review URL: https://codereview.chromium.org/
1261433006
bsalomon [Fri, 31 Jul 2015 18:53:11 +0000 (11:53 -0700)]
Remove unnecessary virtual destructor on SkTArray
Review URL: https://codereview.chromium.org/
1259143006
mtklein [Fri, 31 Jul 2015 18:50:27 +0000 (11:50 -0700)]
Port SkXfermode opts to SkOpts.h
Renames Sk4pxXfermode.h to SkXfermode_opts.h,
and refactors it a tiny bit internally.
This moves xfermode optimization from being "compile-time everywhere but NEON"
to simply "runtime everywhere". I don't anticipate any effect on perf or
correctness.
BUG=skia:4117
Review URL: https://codereview.chromium.org/
1264543006
reed [Fri, 31 Jul 2015 18:46:12 +0000 (11:46 -0700)]
use SkNextID::ImageID for android's stable id
BUG=skia:
Review URL: https://codereview.chromium.org/
1263463003
joshualitt [Fri, 31 Jul 2015 18:45:22 +0000 (11:45 -0700)]
Modifying TextBlobCacheTest to use SkRandomScalerContext
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
adcdca86ce425cf8c28bfad311cef028df756ee8
Review URL: https://codereview.chromium.org/
1266003002
bsalomon [Fri, 31 Jul 2015 18:20:16 +0000 (11:20 -0700)]
Make ANGLE perf decisions be runtime rather than compile time
R=robertphillips@google.com
Review URL: https://codereview.chromium.org/
1268953002
mtklein [Fri, 31 Jul 2015 17:46:50 +0000 (10:46 -0700)]
Port SkUtils opts to SkOpts.
With this new arrangement, the benefits of inlining sk_memset16/32 have changed.
On x86, they're not significantly different, except for small N<=10 where the inlined code is significantly slower.
On ARMv7 with NEON, our custom code is still significantly faster for N>10 (up to 2x faster). For small N<=10 inlining is still significantly faster.
On ARMv7 without NEON, our custom code is still ridiculously faster (up to 10x) than inlining for N>10, though for small N<=10 inlining is still a little faster.
We were not using the NEON memset16 and memset32 procs on ARMv8. At first blush, that seems to be an oversight, but if so it's an extremely lucky one. The ARMv8 code generation for our memset16/32 procs is total garbage, leaving those methods ~8x slower than just inlining the memset, using the compiler's autovectorization.
So, no need to inline any more on x86, and still inline for N<=10 on ARMv7. Always inline for ARMv8.
BUG=skia:4117
Review URL: https://codereview.chromium.org/
1270573002
djsollen [Fri, 31 Jul 2015 17:25:17 +0000 (10:25 -0700)]
Update Android Testing apps to support release mode
BUG=skia:4152
Review URL: https://codereview.chromium.org/
1259123003
bsalomon [Fri, 31 Jul 2015 16:34:24 +0000 (09:34 -0700)]
Restore read pixels perf on ANGLE
BUG=chromium:513797
Review URL: https://codereview.chromium.org/
1265003002
reed [Fri, 31 Jul 2015 15:49:11 +0000 (08:49 -0700)]
remove drawimagerect flags, as skia no longer respects them
BUG=skia:
TBR=
Review URL: https://codereview.chromium.org/
1262333006
joshualitt [Fri, 31 Jul 2015 15:25:07 +0000 (08:25 -0700)]
bump the size of the atlas id to 64 bits
BUG=skia:
Review URL: https://codereview.chromium.org/
1253003005
bsalomon [Fri, 31 Jul 2015 15:22:17 +0000 (08:22 -0700)]
Make SkIsPow2 templated
Review URL: https://codereview.chromium.org/
1271533002
joshualitt [Fri, 31 Jul 2015 14:55:09 +0000 (07:55 -0700)]
Revert of Modifying TextBlobCacheTest to use SkRandomScalerContext (patchset #3 id:40001 of https://codereview.chromium.org/
1266003002/)
Reason for revert:
breaking android
Original issue's description:
> Modifying TextBlobCacheTest to use SkRandomScalerContext
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
adcdca86ce425cf8c28bfad311cef028df756ee8
TBR=bungeman@google.com,bsalomon@google.com,joshualitt@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1267623004
joshualitt [Fri, 31 Jul 2015 14:44:11 +0000 (07:44 -0700)]
Modifying TextBlobCacheTest to use SkRandomScalerContext
BUG=skia:
Review URL: https://codereview.chromium.org/
1266003002
bsalomon [Fri, 31 Jul 2015 13:48:27 +0000 (06:48 -0700)]
Also provides the ResourceProvider to onDrawPath which allows the DF PR to no longer require access to GrContext.
Review URL: https://codereview.chromium.org/
1265763002
mtklein [Fri, 31 Jul 2015 13:43:04 +0000 (06:43 -0700)]
DM: track a direct/indirect bit for each Sink too.
The decoding tests can now veto indirect sinks like pipe-8888.
This moves Sink type detection from automatic to explicit; I can't think of any
way to automatically differentiate pipe-8888 from 8888 based only on the
output. (They should ideally be identical, after all.)
BUG=skia:4138
Review URL: https://codereview.chromium.org/
1263113002
joshualitt [Fri, 31 Jul 2015 13:37:50 +0000 (06:37 -0700)]
Remove unused field in micro cleanup
TBR=bsalomon@google.com
BUG=skia:
Review URL: https://codereview.chromium.org/
1268893002
bungeman [Fri, 31 Jul 2015 13:33:21 +0000 (06:33 -0700)]
Fix repeat builds with Android.
Currently the SampleApp and VisualBench builds for Android don't work
for rebuilds. The code is re-built, but the apks are not. This
results from the use of a directory as 'output' and the action to
build the apk not depending on the copied libraries.
Review URL: https://codereview.chromium.org/
1270643005
joshualitt [Fri, 31 Jul 2015 12:51:45 +0000 (05:51 -0700)]
Tweak the MaskFormat in one more place in GrBatchFontCache
BUG=510931
Review URL: https://codereview.chromium.org/
1268743002
robertphillips [Fri, 31 Jul 2015 12:17:24 +0000 (05:17 -0700)]
Address some SkLightingShader TODOs
This CL:
switches the light colors to be 3 scalars (SkColor3f)
adds some dox
Review URL: https://codereview.chromium.org/
1265983003
borenet [Fri, 31 Jul 2015 11:13:38 +0000 (04:13 -0700)]
Add android_run_app.go
Launches an app, pipes its output, and exits when the app exits.
Loosely based on the old buildbot code in python:
https://skia.googlesource.com/buildbot/+/
ac0663c599a443a4958c8cad5aefd25eb09eff58/slave/skia_slave_scripts/utils/android_utils.py
BUG=skia:4093
Review URL: https://codereview.chromium.org/
1256353002
kkinnunen [Fri, 31 Jul 2015 05:47:04 +0000 (22:47 -0700)]
Implement support for dual source blending in ES
Use EXT_blend_func_extended to implement dual source blending in OpenGL
ES. The extension is the ES version of ARB_blend_func_extended.
The extension provides gl_SecondaryFragColorEXT for ES 2.0 contexts.
The extension provides glBindFragDataLocationIndexed to bind a custom
fragment shader output to the secondary color for ES 3.0 contexts.
For ES 3.1 contexts, the extension would also give
"layout (location=0, index=1)" output varible layout modifier syntax,
but it is not used in this patch.
The extension needs #extension GL_EXT_blend_func_extended : require
directive for the variables to be available in ES 2.0. For ES 3.0, the
directive relaxes the rules for the amount of output variables without
layout location qualifiers.
OpenGL continues to use GL_ARB_blend_func_extended for dual source
blending.
Review URL: https://codereview.chromium.org/
1266773003
kkinnunen [Fri, 31 Jul 2015 05:43:52 +0000 (22:43 -0700)]
Make it possible to toggle MSAA for ES (EXT_multisample_compatibility)
Make it possible to turn MSAA on and off for OpenGL ES, too.
Use EXT_multisample_compatibility.
The extension adds token:
MULTISAMPLE_EXT 0x809D
The token can be supplied to glEnable/glDisable/glGetInteger.
This is needed in order to support mixed samples in OpenGL ES. This is
important for path rendering Chromium command buffer integration.
Review URL: https://codereview.chromium.org/
1270533002
reed [Fri, 31 Jul 2015 03:13:43 +0000 (20:13 -0700)]
lock pixels in image when bitmap is immutable and not-lazy
BUG=skia:
Review URL: https://codereview.chromium.org/
1266143003
reed [Fri, 31 Jul 2015 02:10:35 +0000 (19:10 -0700)]
fix empty image constructor
fBitmap was not fully constructed yet
BUG=skia:
TBR=
Review URL: https://codereview.chromium.org/
1264773004
reed [Fri, 31 Jul 2015 01:58:23 +0000 (18:58 -0700)]
unify pixelref and image ID space, so we can share IDs when we share pixels
I view this as a performance opportunity, not a feature or bug fix per-se.
BUG=skia:
Review URL: https://codereview.chromium.org/
1266883002
bsalomon [Thu, 30 Jul 2015 23:27:50 +0000 (16:27 -0700)]
Disable SRGB support on PowerVR Rogue due to SRGBReadWritePixels failure
TBR=jvanverth@google.com
BUG=skia:4148
Review URL: https://codereview.chromium.org/
1266933002
bsalomon [Thu, 30 Jul 2015 22:57:33 +0000 (15:57 -0700)]
Make TRecorder alloc_back return a void*
Speculative fix for this bug:
BUG=chromium:515679
TBR=robertphillips@google.com
Review URL: https://codereview.chromium.org/
1268493003
bsalomon [Thu, 30 Jul 2015 22:34:56 +0000 (15:34 -0700)]
SRGB read and write pixels working and unit test
Review URL: https://codereview.chromium.org/
1264003002
joshualitt [Thu, 30 Jul 2015 19:43:17 +0000 (12:43 -0700)]
find image whenever we generateGlyph. This should ensure the glyph is fully initialized before we start using the maskformat.
BUG=skia:
Review URL: https://codereview.chromium.org/
1270573003
caryclark [Thu, 30 Jul 2015 19:35:48 +0000 (12:35 -0700)]
linear gradient with stops discretized gm
TBR=reed@google.com
BUG=skia:517
Review URL: https://codereview.chromium.org/
1259983009
egdaniel [Thu, 30 Jul 2015 19:02:15 +0000 (12:02 -0700)]
Move some parts of onReadPixels up to GrGpu readPixels.
BUG=skia:
Review URL: https://codereview.chromium.org/
1270583003
msarett [Thu, 30 Jul 2015 18:36:40 +0000 (11:36 -0700)]
Rename SkCodec_libbmp to SkBmpCodec
BUG=skia:
Review URL: https://codereview.chromium.org/
1254963006
scroggo [Thu, 30 Jul 2015 18:33:04 +0000 (11:33 -0700)]
Call notifyPixelsChanged after modifying bitmap
In CodecSrc's scanline_subset test, we decode a subset of an image to a
bitmap, draw it to the canvas, and then repeat. This is fine for most
backends, but not for pipe. Pipe sees the same generation ID, so it
assumes it is the same bitmap it saw before, and just draws the
original one.
Call notifyPixelsChanged, so the bitmap will get a new generation ID,
fixing pipe.
BUG=skia:4138
Review URL: https://codereview.chromium.org/
1265983004
nodir [Thu, 30 Jul 2015 17:32:27 +0000 (10:32 -0700)]
Build buckets: tryjob access
Allowed project-skia-tryjob-access group to schedule builds on public
skia build buckets
R=borenet@chromium.org, rmistry@chromium.org
BUG=495680
Review URL: https://codereview.chromium.org/
1256333006
bungeman [Thu, 30 Jul 2015 17:31:57 +0000 (10:31 -0700)]
Revert character in typeface test on Android.
This is a partial revert of
f20488b4f2, particularly
https://skia.googlesource.com/skia/+/
f20488b4f2139e6ca09fee7e39b731dd8ab467db%5E%21/#F11
BUG=chromium:515366
TBR=mtklein@google.com
This is a (non-automatic) revert.
Review URL: https://codereview.chromium.org/
1265993003
robertphillips [Thu, 30 Jul 2015 17:11:42 +0000 (10:11 -0700)]
Minor cleanup of SkLightingShader
Fix:
PM component swizzling on Macs
Serialization
Review URL: https://codereview.chromium.org/
1270533003
mtklein [Thu, 30 Jul 2015 16:29:37 +0000 (09:29 -0700)]
Runtime CPU detection for rsqrt().
This enables the NEON sk_float_rsqrt() code for configurations that have NEON at run-time but not compile-time.
These devices will see about a 2x (1.26 -> 2.33) slowdown in sk_float_rsqrt(), but it should be more precise than our portable fallback.
(When inlined, the portable fallback and the NEON code are almost identical in speed. The only difference is precision. Going through a function pointer is causing all this slowdown. This is a good example of a place where Skia really benefits from compile-time NEON.)
BUG=skia:4117,skia:4114
No public API changes.
TBR=reed@google.com
Review URL: https://codereview.chromium.org/
1264893002
joshualitt [Thu, 30 Jul 2015 14:59:20 +0000 (07:59 -0700)]
Dont try to draw glyphs with unexpected mask formats
BUG=510931
Review URL: https://codereview.chromium.org/
1269743003
scroggo [Thu, 30 Jul 2015 14:47:45 +0000 (07:47 -0700)]
Remove SK_LEGACY_IMAGE_GENERATOR_ENUMS_AND_OPTIONS
Now that Chrome no longer depends on it, remove dead code.
Review URL: https://codereview.chromium.org/
1263013002
bsalomon [Thu, 30 Jul 2015 14:34:27 +0000 (07:34 -0700)]
Make GrGpu read/write pixels take GrSurface
Review URL: https://codereview.chromium.org/
1262473004
mtklein [Thu, 30 Jul 2015 14:30:16 +0000 (07:30 -0700)]
Lay groundwork for SkOpts.
This doesn't really do anything yet. It's just the CPU detection code, skeleton new .cpp files, and a few little .gyp tweaks.
BUG=skia:4117
Committed: https://skia.googlesource.com/skia/+/
ce2c5055cee5d5d3c9fc84c1b3eeed4b4d84a827
Review URL: https://codereview.chromium.org/
1255193002
borenet [Thu, 30 Jul 2015 14:09:20 +0000 (07:09 -0700)]
buildbot_spec: Add configuration, do_*_steps, upload_*_results, more env
NOTRY=true
BUG=skia:4132
Review URL: https://codereview.chromium.org/
1258193004
mtklein [Thu, 30 Jul 2015 14:03:20 +0000 (07:03 -0700)]
Revert of Optimize RGB16 blitH functions with NEON for ARM platform. (patchset #2 id:20001 of https://codereview.chromium.org/
1229673008/)
Reason for revert:
This doesn't draw correctly, e.g. our GM test named dashcubics.
Good: https://gold.skia.org/img/images/
0f7e8e226379afbad8a700e0a80fd8f1.png
Bad:
https://gold.skia.org/img/images/
56ce15fc67436065a3db4b8ee31f13ae.png
Original issue's description:
> Optimize RGB16 blitH functions with NEON for ARM platform.
>
> Here are some performance resultsi on Nexus 9:
> SkRGB16BlitterBlitH_neon:
> +--------+-----------+
> |height | C/NEON |
> +--------+-----------+
> |1 | 0.888531 |
> +--------+-----------+
> |8 | 1.231800 |
> +--------+-----------+
> |18 | 1.073327 |
> +--------+-----------+
> |32 | 1.136991 |
> +--------+-----------+
> |76 | 1.174638 |
> +--------+-----------+
> |85 | 1.188551 |
> +--------+-----------+
> |120 | 1.180261 |
> +--------+-----------+
> |128 | 1.183726 |
> +--------+-----------+
> |512 | 1.220806 |
> +--------+-----------+
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
6c72d5740231f47c664a8e765a8df05cd124c88c
TBR=djsollen@google.com,caryclark@google.com,reed@google.com,bero@linaro.com,yang.zhang@linaro.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1268513003
caryclark [Thu, 30 Jul 2015 13:41:39 +0000 (06:41 -0700)]
add radial gradient hard stop test
TBR=reed@google.com
BUG=skia:4140
Review URL: https://codereview.chromium.org/
1269723002
reed [Thu, 30 Jul 2015 12:46:05 +0000 (05:46 -0700)]
add helper to create RSXform w/ anchorPt
BUG=skia:
Review URL: https://codereview.chromium.org/
1269563006
yang.zhang [Thu, 30 Jul 2015 12:37:06 +0000 (05:37 -0700)]
Optimize RGB16 blitH functions with NEON for ARM platform.
Here are some performance resultsi on Nexus 9:
SkRGB16BlitterBlitH_neon:
+--------+-----------+
|height | C/NEON |
+--------+-----------+
|1 | 0.888531 |
+--------+-----------+
|8 | 1.231800 |
+--------+-----------+
|18 | 1.073327 |
+--------+-----------+
|32 | 1.136991 |
+--------+-----------+
|76 | 1.174638 |
+--------+-----------+
|85 | 1.188551 |
+--------+-----------+
|120 | 1.180261 |
+--------+-----------+
|128 | 1.183726 |
+--------+-----------+
|512 | 1.220806 |
+--------+-----------+
BUG=skia:
Review URL: https://codereview.chromium.org/
1229673008
robertphillips [Thu, 30 Jul 2015 12:09:17 +0000 (05:09 -0700)]
Update SkLightingShader to take a localMatrix
W/o this we can't draw lit objects anywhere but the origin.
Review URL: https://codereview.chromium.org/
1253223003
bungeman [Wed, 29 Jul 2015 21:46:12 +0000 (14:46 -0700)]
Remove WEBKIT_VERSION_MIN_REQUIRED and config.h.
WebKit style guide (https://www.webkit.org/coding/coding-style.html)
explicitly states:
> All implementation files must #include "config.h" first.
> Header files should never include "config.h".
Also, it isn't currently being used.
TBR=reed@google.com
This just removes unused bits.
Review URL: https://codereview.chromium.org/
1256133004
fmalita [Wed, 29 Jul 2015 21:40:06 +0000 (14:40 -0700)]
Double free in ~SkPictureData()
On subpicture parsing failures we clean up all fPictureRefs entries
*and* delete the array itself. But the destructor also deletes the
array => double free.
Alternatively, we can set fPictureCount to the number of successfully
parsed pictures such that the destructor handles all the cleanup.
BUG=515228
R=reed@google.com,mtklein@google.com
Review URL: https://codereview.chromium.org/
1264503011
bungeman [Wed, 29 Jul 2015 20:37:27 +0000 (13:37 -0700)]
Remove SK_BUILD_FOR SDL, BREW, and PALM.
SDL isn't an OS anyway, it's just a library views can use. Remaining
support for Brew was removed some time ago, and there are currently
no uses of SK_BUILD_FOR_PALM.
Review URL: https://codereview.chromium.org/
1268573002
mtklein [Wed, 29 Jul 2015 20:31:22 +0000 (13:31 -0700)]
Move the last headers. Cross your fingers.
If this doesn't work, one at a time...
BUG=skia:4126
No public API changes.
TBR=reed@google.com
Review URL: https://codereview.chromium.org/
1255373006
mtklein [Wed, 29 Jul 2015 20:14:05 +0000 (13:14 -0700)]
Move non-trivial constructors out-of-line.
There is more than one way to skin this SkPathPriv.h cat.
These constructors are large enough that they probably shouldn't have
been inlined like this anyway.
BUG=skia:4126
Review URL: https://codereview.chromium.org/
1253963004
reed [Wed, 29 Jul 2015 19:44:52 +0000 (12:44 -0700)]
remove (unnecessary?) call to notifyPixelsChanged
BUG=skia:
Review URL: https://codereview.chromium.org/
1265733002
bungeman [Wed, 29 Jul 2015 19:34:25 +0000 (12:34 -0700)]
Remove SK_BUILD_FOR_WINCE.
This hasn't been tested for years, and no one currently knows when it
last worked (if ever). It is doubtful that any of the remaining logic
would even make sense with a modern version of Embedded Compact 2013.
Review URL: https://codereview.chromium.org/
1260453008
robertphillips [Wed, 29 Jul 2015 19:28:04 +0000 (12:28 -0700)]
Move LightingShader to effects
Additionally this CL:
forces the light colors to be opaque
forces the light direction to be normalized
adds a raster implementation
adds a gm
Review URL: https://codereview.chromium.org/
1245883003
bungeman [Wed, 29 Jul 2015 18:49:40 +0000 (11:49 -0700)]
Clean up a few includes, introduce iwyu.
The current include-what-you-use with current clang is much less
noisy and more useful than it has been in the past. This change
introduces a few IWYU directives (which are helpful documentation for
humans as well) and fixes a few sets of includes.
Review URL: https://codereview.chromium.org/
1207893002
reed [Wed, 29 Jul 2015 18:44:52 +0000 (11:44 -0700)]
SkImage_Raster's pixels are always immutable.
To make this work, we tag their pixelrefs as temporarily immutable, allowing
ourselves to restore the pixels to mutability only when the image drops away.
This should allow us to wobble back and forth between writing to the Surface
and reading from the Image without a COW, with the Surface seeing mutable
pixels and the Image seeing immutable pixels.
The big idea is, Image doesn't need forever-immutable pixels, it just needs
pixels that are immutable as long as it's alive.
BUG=skia:
patch from issue
804523002 at patchset 40001 (http://crrev.com/
804523002#ps40001)
Review URL: https://codereview.chromium.org/
1254383006
borenet [Wed, 29 Jul 2015 18:43:07 +0000 (11:43 -0700)]
buildbot_spec.py: Include dm_flags and nanobench_flags
Eventually I'll want to merge them into this file, but not just yet.
NOTRY=true
BUG=skia:4132
Review URL: https://codereview.chromium.org/
1257933003
borenet [Wed, 29 Jul 2015 18:15:42 +0000 (11:15 -0700)]
buildbot_spec.py: Add Android/ChromeOS device info, add builder_cfg dict
NOTRY=true
BUG=skia:4132
Review URL: https://codereview.chromium.org/
1269543002