halcanary [Tue, 26 May 2015 15:01:19 +0000 (08:01 -0700)]
sync-and-gyp: make smarter
This fixes two issues: If gyp is run with the CXX and CC environment
variables set, it will produce different ninja build files.
sync-and-gyp now tracks the values of those variables; if they change,
re-run gyp.
Secondly, we introduced find.py, which is used inside gyp files. If
the contents of certain directories changes, gyp should be re-run. So
now we track those directories too.
Review URL: https://codereview.chromium.org/
1153083002
stephana [Tue, 26 May 2015 13:19:11 +0000 (06:19 -0700)]
Disabling ~path_hairline in iOS nanobench
Temporary measure to make nanobench succeed.
TBR=borenet
BUG=skia:
Review URL: https://codereview.chromium.org/
1144173004
reed [Tue, 26 May 2015 04:21:27 +0000 (21:21 -0700)]
remove unneeded device:lockpixels
BUG=skia:
TBR=
Review URL: https://codereview.chromium.org/
1156073003
kkinnunen [Mon, 25 May 2015 05:12:21 +0000 (22:12 -0700)]
Avoid creating GrGLPath to the resource cache when creating GrGLPathRanges
Do not create a GrGLPath when loading glyphs with the driver and creating
GrGLPathRange based on those. The path would be inserted to the GPU
resource cache and then immediately be released.
Instead, just init the first path object of the path range as the
template.
Review URL: https://codereview.chromium.org/
1146983004
skia.buildbots [Sun, 24 May 2015 07:23:16 +0000 (00:23 -0700)]
Update SKP version
Automatic commit by the RecreateSKPs bot.
TBR=
NO_MERGE_BUILDS
Review URL: https://codereview.chromium.org/
1159443003
reed [Sun, 24 May 2015 02:14:51 +0000 (19:14 -0700)]
deprecate calling lockPixels
BUG=skia:
TBR=
Review URL: https://codereview.chromium.org/
1156003002
reed [Sat, 23 May 2015 20:21:06 +0000 (13:21 -0700)]
move readPixels from bitmap -> pixmap
BUG=skia:
TBR=
Review URL: https://codereview.chromium.org/
1153893003
reed [Sat, 23 May 2015 19:26:41 +0000 (12:26 -0700)]
remove bitmaps entirely from sprite blits (as source)
BUG=skia:
TBR=
Review URL: https://codereview.chromium.org/
1156713004
bungeman [Sat, 23 May 2015 00:55:26 +0000 (17:55 -0700)]
Revert of Implement Porter Duff XP with a blend table (patchset #12 id:220001 of https://codereview.chromium.org/
1124373002/)
Reason for revert:
Blocking DEPS roll into Chromium. Crashing virtual/gpu/fast/canvas/canvas-composite-*.html tests with the assert
../../third_party/skia/src/gpu/gl/builders/GrGLFragmentShaderBuilder.cpp:281: failed assertion "k110_GrGLSLGeneration != gpu->glslGeneration() || fOutputs.empty()"
Original issue's description:
> Implement Porter Duff XP with a blend table
>
> Removes the runtime logic used by PorterDuffXferProcessor to decide
> blend coeffs and shader outputs, and instead uses a compile-time
> constant table of pre-selected blend formulas.
>
> Introduces a new blend strategy for srcCoeff=0 that can apply coverage
> with a reverse subtract blend equation instead of dual source
> blending.
>
> Adds new macros in GrBlend.h to analyze blend formulas both runtime.
>
> Removes kSetCoverageDrawing_OptFlag and GrSimplifyBlend as they are no
> longer used.
>
> Adds a GM that verifies all xfermodes, including arithmetic, with the
> color/coverage invariants used by Porter Duff.
>
> Adds a unit test that verifies each Porter Duff formula with every
> color/coverage invariant.
>
> Major changes:
>
> * Uses a reverse subtract blend equation for coverage when srcCoeff=0
> (clear, dst-out [Sa=1], dst-in, modulate). Platforms that don't
> support dual source blending no longer require a dst copy for
> dst-in and modulate.
>
> * Sets BlendInfo::fWriteColor to false when the blend does not modify
> the dst. GrGLGpu will now use glColorMask instead of blending for
> these modes (dst, dst-in [Sa=1], modulate ignored for [Sc=1]).
>
> * Converts all SA blend coeffs to One for opaque inputs, and ISA to
> Zero if there is also no coverage. (We keep ISA around when there
> is coverage because we use it to tweak alpha for coverage.)
>
> * Abandons solid white optimizations for the sake of simplicity
> (screen was the only mode that previous had solid white opts).
>
> Minor differences:
>
> * Inconsequential differences in opt flags (e.g. we now return
> kCanTweakAlphaForCoverage_OptFlag even when there is no coverage).
>
> * Src coeffs when the shader outputs 0.
>
> * IS2C vs IS2A when the secondary output is scalar.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
9a70920db22b6309c671f8e5d519bb95570e4414
TBR=egdaniel@google.com,bsalomon@google.com,cdalton@nvidia.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1153993002
bungeman [Sat, 23 May 2015 00:54:44 +0000 (17:54 -0700)]
Revert of Fix LCD coverage regression in GrPorterDuffXPFactory (patchset #3 id:40001 of https://codereview.chromium.org/
1149883004/)
Reason for revert:
Reverting this fix to revert https://codereview.chromium.org/
1124373002
Original issue's description:
> Fix LCD coverage regression in GrPorterDuffXPFactory
>
> Fixes GrPorterDuffXPFactory to not use the blend table when getting
> info about the LCD coverage XP.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
3e2d59ef36e1b9a63e524cf1b27d5f7bcfb5ff0b
TBR=bsalomon@google.com,egdaniel@google.com,cdalton@nvidia.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1153803005
mtklein [Sat, 23 May 2015 00:18:21 +0000 (17:18 -0700)]
Everyone gets a namespace {}.
If we include Sk4px.h, SkPMFloat.h, or SkNx.h into files with different
SIMD flags, that could cause different definitions of the same method.
Normally that's moot, because all the code inlines, but in Debug it tends not
to. So in Debug, the linker picks one definition for us. That breaks _someone_.
Wrapping everything in a namespace {} keeps the definitions separate.
Tested locally, it fixes this bug.
BUG=skia:3861
This code is not yet enabled in Chrome, so shouldn't affect the roll.
NOTREECHECKS=true
Review URL: https://codereview.chromium.org/
1154523004
bungeman [Fri, 22 May 2015 22:22:48 +0000 (15:22 -0700)]
Revert of Move DstCoordTexture to GrXP, rename and remove the word "copy" from dstcopytexture names. (patchset #6 id:100001 of https://codereview.chromium.org/
1132093004/)
Reason for revert:
This is asserting on GTX660 bots with '!dst'. This may just be catching an existing issue.
https://build.chromium.org/p/client.skia/builders/Test-Ubuntu-GCC-ShuttleA-GPU-GTX660-x86_64-Debug/builds/622/steps/dm/logs/stdio
https://build.chromium.org/p/client.skia/builders/Test-Win7-MSVC-ShuttleA-GPU-HD2000-x86-Debug-ANGLE/builds/209/steps/dm/logs/stdio
https://build.chromium.org/p/client.skia/builders/Test-Win8-MSVC-ShuttleA-GPU-GTX660-x86_64-Debug/builds/318/steps/dm/logs/stdio
Original issue's description:
> Move DstCoordTexture to GrXP, rename and remove the word "copy" from dstcopytexture names.
>
> Committed: https://skia.googlesource.com/skia/+/
bb106b5f6ee18f7453b63f3a95f421b60a957f39
TBR=egdaniel@google.com,bsalomon@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review URL: https://codereview.chromium.org/
1158453004
bsalomon [Fri, 22 May 2015 21:32:10 +0000 (14:32 -0700)]
Move DstCoordTexture to GrXP, rename and remove the word "copy" from dstcopytexture names.
Review URL: https://codereview.chromium.org/
1132093004
cdalton [Fri, 22 May 2015 21:04:37 +0000 (14:04 -0700)]
Fix LCD coverage regression in GrPorterDuffXPFactory
Fixes GrPorterDuffXPFactory to not use the blend table when getting
info about the LCD coverage XP.
BUG=skia:
Review URL: https://codereview.chromium.org/
1149883004
bsalomon [Fri, 22 May 2015 21:01:46 +0000 (14:01 -0700)]
Store context options on caps.
Committed: https://skia.googlesource.com/skia/+/
f28cff71db2cbb1ff18a8fbf1e80ca761d1f69bc
Review URL: https://codereview.chromium.org/
1158433006
joshualitt [Fri, 22 May 2015 20:09:57 +0000 (13:09 -0700)]
fix for atlas is abandoned text corruption
BUG=skia:
Review URL: https://codereview.chromium.org/
1156693004
mtklein [Fri, 22 May 2015 19:43:41 +0000 (12:43 -0700)]
Simpler version of Plus w/ AA. ~25% faster too.
BUG=skia:3852
TBR=fmalita@chromium.org
Review URL: https://codereview.chromium.org/
1150693003
bsalomon [Fri, 22 May 2015 19:41:05 +0000 (12:41 -0700)]
Revert of Store context options on caps. (patchset #3 id:40001 of https://codereview.chromium.org/
1158433006/)
Reason for revert:
breaking chromeos build (???)
Original issue's description:
> Store context options on caps.
>
> Committed: https://skia.googlesource.com/skia/+/
f28cff71db2cbb1ff18a8fbf1e80ca761d1f69bc
TBR=joshualitt@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review URL: https://codereview.chromium.org/
1151603005
bsalomon [Fri, 22 May 2015 19:25:41 +0000 (12:25 -0700)]
Store context options on caps.
Review URL: https://codereview.chromium.org/
1158433006
reed [Fri, 22 May 2015 19:10:46 +0000 (12:10 -0700)]
use SkPixmap inside sprite blitters
BUG=skia:
Review URL: https://codereview.chromium.org/
1144923005
cdalton [Fri, 22 May 2015 18:36:57 +0000 (11:36 -0700)]
Implement Porter Duff XP with a blend table
Removes the runtime logic used by PorterDuffXferProcessor to decide
blend coeffs and shader outputs, and instead uses a compile-time
constant table of pre-selected blend formulas.
Introduces a new blend strategy for srcCoeff=0 that can apply coverage
with a reverse subtract blend equation instead of dual source
blending.
Adds new macros in GrBlend.h to analyze blend formulas both runtime.
Removes kSetCoverageDrawing_OptFlag and GrSimplifyBlend as they are no
longer used.
Adds a GM that verifies all xfermodes, including arithmetic, with the
color/coverage invariants used by Porter Duff.
Adds a unit test that verifies each Porter Duff formula with every
color/coverage invariant.
Major changes:
* Uses a reverse subtract blend equation for coverage when srcCoeff=0
(clear, dst-out [Sa=1], dst-in, modulate). Platforms that don't
support dual source blending no longer require a dst copy for
dst-in and modulate.
* Sets BlendInfo::fWriteColor to false when the blend does not modify
the dst. GrGLGpu will now use glColorMask instead of blending for
these modes (dst, dst-in [Sa=1], modulate ignored for [Sc=1]).
* Converts all SA blend coeffs to One for opaque inputs, and ISA to
Zero if there is also no coverage. (We keep ISA around when there
is coverage because we use it to tweak alpha for coverage.)
* Abandons solid white optimizations for the sake of simplicity
(screen was the only mode that previous had solid white opts).
Minor differences:
* Inconsequential differences in opt flags (e.g. we now return
kCanTweakAlphaForCoverage_OptFlag even when there is no coverage).
* Src coeffs when the shader outputs 0.
* IS2C vs IS2A when the secondary output is scalar.
BUG=skia:
Review URL: https://codereview.chromium.org/
1124373002
joshualitt [Fri, 22 May 2015 18:19:44 +0000 (11:19 -0700)]
Trivial patch to set a max lookback on reordering
BUG=skia:
Review URL: https://codereview.chromium.org/
1150253003
stephana [Fri, 22 May 2015 17:55:32 +0000 (10:55 -0700)]
BUG=skia:
NOTRY=true
TBR=borenet
Review URL: https://codereview.chromium.org/
1153783003
mtklein [Fri, 22 May 2015 17:54:39 +0000 (10:54 -0700)]
Move Sk4px Xfermode code to a header so we can use it twice.
- Once in SkXfermode as usual to pick up compile-time SSE and NEON
- Once in SkXfermode_arm_neon to pick up run-time NEON
This allows us to start cleaning up SkXfermode_arm_neon as we've done
for SkXfermode_SSE2. I'm saving this catharsis for a day when I need it.
The Sk4px xfermodes are generally faster than the existing NEON procs,
so this should also have the side effect of a perf win there.
This means our new Plus-AA code works for runtime NEON too.
BUG=skia:3852
Review URL: https://codereview.chromium.org/
1150313003
jcgregorio [Fri, 22 May 2015 17:42:13 +0000 (10:42 -0700)]
Add more detail on how DM checksums are done.
This will be linked to from Gold for image diffs that have
different digests but no pixel differences.
BUG=skia:
NOTRY=true
DOCS_PREVIEW= https://skia.org/?cl=
1151153005
Review URL: https://codereview.chromium.org/
1151153005
bsalomon [Fri, 22 May 2015 17:37:30 +0000 (10:37 -0700)]
Remove init from GrGLContextInfo and caps classes
Make Gr*Caps, GrGLContext* noncopyable
Review URL: https://codereview.chromium.org/
1153813002
reed [Fri, 22 May 2015 15:06:21 +0000 (08:06 -0700)]
add SkPixmap and external locking to bitmaps
BUG=skia:
Review URL: https://codereview.chromium.org/
1074983003
bsalomon [Fri, 22 May 2015 15:01:09 +0000 (08:01 -0700)]
rename GrDrawTargetCaps.h to GrCaps.h and move to include
Review URL: https://codereview.chromium.org/
1135113005
bsalomon [Fri, 22 May 2015 14:35:58 +0000 (07:35 -0700)]
Move GrPath(Range) creation to ResourceProvider.
Review URL: https://codereview.chromium.org/
1155593002
caryclark [Fri, 22 May 2015 13:26:52 +0000 (06:26 -0700)]
remove SK_LEGACY_STROKE_CURVES
The change in Chrome has been landed so this guard is no longer needed.
R=reed@google.com,fmalita@chromium.org
BUG=102411
Review URL: https://codereview.chromium.org/
1157623003
mtklein [Fri, 22 May 2015 13:08:29 +0000 (06:08 -0700)]
Re-proc SkBlitRow::Color32 for ARM.
This is a spiritual revert of http://crrev.com/
1104183004.
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
4e13a23d8f720e17660f26657b45b89fe4339004
Review URL: https://codereview.chromium.org/
1145283003
jvanverth [Fri, 22 May 2015 13:06:40 +0000 (06:06 -0700)]
Further work on sRGB support.
Fix some caps.
Add framebuffer enum.
BUG=skia:3856
Review URL: https://codereview.chromium.org/
1152563003
mtklein [Fri, 22 May 2015 12:51:37 +0000 (05:51 -0700)]
Revert of Re-proc SkBlitRow::Color32 for ARM. (patchset #3 id:40001 of https://codereview.chromium.org/
1145283003/)
Reason for revert:
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_rel_device_ninja/builds/70016/steps/compile%20%28with%20patch%29/logs/stdio
Original issue's description:
> Re-proc SkBlitRow::Color32 for ARM.
>
> This is a spiritual revert of http://crrev.com/
1104183004.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
4e13a23d8f720e17660f26657b45b89fe4339004
TBR=reed@google.com,mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1157633003
kkinnunen [Fri, 22 May 2015 05:45:56 +0000 (22:45 -0700)]
Mark SkPaths created by GrContext fallback draw logic as volatile
GrContext draw functions (drawRect, drawRRect,...) fall back to
GrContext::drawPath for complex cases. Mark the argument SkPath
as volatile, since it is.
Progressions for nvprmsaa:
tabl_androidpolice.skp_1 11.6ms -> 10.3ms 0.89x
desk_twitter.skp_1 9.21ms -> 8.08ms 0.88x
tabl_sahadan.skp_1 8.97ms -> 7.65ms 0.85x
desk_wikipedia.skp_1 5.63ms -> 4.5ms 0.8x
desk_booking.skp_1 11.5ms -> 8.87ms 0.77x
desk_ebay.skp_1 7ms -> 5.37ms 0.77x
Makes rrects and rects go to "simple path" path cache (cached by contents).
No change for msaa, gpu.
BUG=skia:
Review URL: https://codereview.chromium.org/
1150993002
mtklein [Fri, 22 May 2015 00:14:51 +0000 (17:14 -0700)]
Re-proc SkBlitRow::Color32 for ARM.
This is a spiritual revert of http://crrev.com/
1104183004.
BUG=skia:
Review URL: https://codereview.chromium.org/
1145283003
joshualitt [Thu, 21 May 2015 22:50:36 +0000 (15:50 -0700)]
Remove uniform view matrix from GrGLPrimitiveProcessor base class
BUG=skia:
Review URL: https://codereview.chromium.org/
1151693005
mtklein [Thu, 21 May 2015 22:47:40 +0000 (15:47 -0700)]
Fix Plus
This makes Plus with AA ~3% slower. Seems like a good deal.
GMs affected: mixed_xfermodes, the new one.
Based on https://codereview.chromium.org/
1150833003/
Still TODO: NEON. The new GM should show this.
BUG=skia:3852
Review URL: https://codereview.chromium.org/
1156453002
reed [Thu, 21 May 2015 22:46:20 +0000 (15:46 -0700)]
remove unused SkBitmap::scrollRect API
BUG=skia:
TBR=
Review URL: https://codereview.chromium.org/
1154623002
ochang [Thu, 21 May 2015 22:44:53 +0000 (15:44 -0700)]
Add mode validation to PtProcRec::init.
BUG=486946
Review URL: https://codereview.chromium.org/
1142563003
mtklein [Thu, 21 May 2015 21:08:14 +0000 (14:08 -0700)]
Clean up Sk4f xfermodes and covered _SSE2 xfermodes.
Before I get going on fixing Plus, it's nice to clear out the dead cruft.
BUG=skia:3852
Review URL: https://codereview.chromium.org/
1150833003
stephana [Thu, 21 May 2015 20:59:36 +0000 (13:59 -0700)]
Temporarily disable nanobench to see what else is in the way of the bots going green.
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1147333002
Eric Boren [Thu, 21 May 2015 20:19:22 +0000 (16:19 -0400)]
Whitespace change to trigger builds
Two major things have happened:
1. Android bots split between more machines. This caused lots of
congestion-related failures, leading to...
2. All buildslaves now use a single checkout for all of the builders
that they run. This should give us far fewer problems when we have
to re-download SKPs, toolchains, etc.
TBR=rmistry
BUG=skia:3835,3774
Review URL: https://codereview.chromium.org/
1154603002
mtklein [Thu, 21 May 2015 17:33:09 +0000 (10:33 -0700)]
Add PlusMergesAA GM
C.f. https://crrev.com/
1156453002/
BUG=skia:
Review URL: https://codereview.chromium.org/
1149103003
cdalton [Thu, 21 May 2015 15:27:44 +0000 (08:27 -0700)]
Make mixed samples contingent on auxiliary extensions
Moves the cap for mixed samples into GrShaderCaps and does not enable
it unless we have support for both dual source blending and
multisample disable.
Creates a dedicated cap for multisample disable.
Reconfigures the mixed samples cap to indicate the collective
capability of three different extensions:
GL_NV_framebuffer_mixed_samples
GL_NV_sample_mask_override_coverage
GL_EXT_raster_multisample
Imports tokens and procedures for GL_EXT_raster_multisample.
BUG=skia:
Review URL: https://codereview.chromium.org/
1151793002
halcanary [Thu, 21 May 2015 15:13:27 +0000 (08:13 -0700)]
SkBlockMemoryStream implements peek()
Review URL: https://codereview.chromium.org/
1146903004
jvanverth [Thu, 21 May 2015 15:12:27 +0000 (08:12 -0700)]
Add RGBA half float texture format.
BUG=skia:3829
Review URL: https://codereview.chromium.org/
1148243002
reed [Thu, 21 May 2015 13:29:05 +0000 (06:29 -0700)]
add assert that the lock-count on pixelref is balanced
BUG=skia:
TBR=
Review URL: https://codereview.chromium.org/
1147153004
kkinnunen [Thu, 21 May 2015 13:15:28 +0000 (06:15 -0700)]
Move font loading in gm tests and benches out of constructors
Constructing the gm tests and benches causes many calls to font loads.
This is visible as profiling samples in fontconfig and freetype on Linux
for all profiling runs of nanobench. This complicates analysis of
test-cases that are suspected of being slow due to font-related issues.
Move the font loading to GM::onOnceBeforeDraw and Benchmark::onPreDraw.
This way the code is not executed if the testcase does not match the
nanobench --match filter. This way the samples in font-related code are
more easy to identify as legitimate occurances caused by the testcase.
This should not cause differences in timings, because:
* Benchmark::preDraw / onPreDraw is defined to be run outside the timer
* GM::runAsBench is not enabled for any of the modified testcases. Also
nanobench untimed warmup round should run the onOnceBeforeDraw.
(and there are other GM::runAsBench gms already doing loading in
onOnceBeforeDraw).
Changes the behavior:
In TextBench:
Before, the test would report two different gms with the same name if
the color emoji font was not loaded successfully.
After, the test always reports all tests as individual names.
Generally:
The errors from loading fonts now print inbetween each testcase, as
opposed to printing during construction phase. Sample output:
( 143/145 MB 1872) 14.7ms 8888 gm quadclosepathResource /fonts/Funkster.ttf not a valid font.
( 160/160 MB 1831) 575µs 8888 gm surfacenewResource /fonts/Funkster.ttf not a valid font.
( 163/165 MB 1816) 12.5ms 8888 gm linepathResource /fonts/Funkster.ttf not a valid font.
( 263/411 MB 1493) 118ms 8888 gm typefacestyles_kerningResource /fonts/Funkster.ttf not a valid font.
( 374/411 MB 1231) 7.16ms 565 gm getpostextpathResource /fonts/Funkster.ttf not a valid font.
( 323/411 MB 1179) 4.92ms 565 gm stringartResource /fonts/Funkster.ttf not a valid font.
( 347/493 MB 917) 191ms 565 gm patch_gridResource /fonts/Funkster.ttf not a valid font.
( 375/493 MB 857) 23.9ms gpu gm clipdrawdrawCannot render path (0)
( 393/493 MB 706) 2.91ms unit test ParsePath------ png error IEND: CRC error
( 394/493 MB 584) 166ms gpu gm hairmodesResource /fonts/Funkster.ttf not a valid font.
Resource /fonts/Funkster.ttf not a valid font.
Resource /fonts/Funkster.ttf not a valid font.
...
Review URL: https://codereview.chromium.org/
1144023002
kkinnunen [Thu, 21 May 2015 07:37:30 +0000 (00:37 -0700)]
Improve caching of special case paths in GrStencilAndCoverPathRenderer
Cache lines and oval paths in their own cache domains. Skia has many
hard-to-replace codepaths that create volatile paths out of lines or
ovals.
Results for amd64:
desk_tigersvg.skp_1 3.06ms -> 3.07ms 1x
tabl_cnet.skp_1 2.3ms -> 2.3ms 1x
desk_baidu.skp_1 8.99ms -> 8.98ms 1x
desk_weather.skp_1 4.58ms -> 4.57ms 1x
desk_samoasvg.skp_1 12.3ms -> 12.3ms 1x
tabl_gamedeksiam.skp_1 15.8ms -> 15.7ms 1x
desk_chalkboard.skp_1 14.9ms -> 14.9ms 1x
desk_mapsvg.skp_1 6.57ms -> 6.54ms 1x
desk_wordpress.skp_1 2.2ms -> 2.19ms 1x
tabl_slashdot.skp_1 6.91ms -> 6.84ms 0.99x
desk_linkedin.skp_1 7.7ms -> 7.58ms 0.98x
desk_googlespreadsheet.skp_1 58.7ms -> 57.7ms 0.98x
tabl_ukwsj.skp_1 7.67ms -> 7.53ms 0.98x
tabl_engadget.skp_1 4.71ms -> 4.61ms 0.98x
desk_carsvg.skp_1 116ms -> 114ms 0.98x
tabl_nytimes.skp_1 3.71ms -> 3.61ms 0.97x
desk_googlespreadsheetdashed.skp_1 25.8ms -> 24.9ms 0.97x
tabl_hsfi.skp_1 5.3ms -> 5.1ms 0.96x
tabl_techmeme.skp_1 2.9ms -> 2.8ms 0.96x
tabl_cnn.skp_1 6.88ms -> 6.62ms 0.96x
desk_espn.skp_1 9.12ms -> 8.64ms 0.95x
desk_gws.skp_1 5.15ms -> 4.88ms 0.95x
tabl_gspro.skp_1 4.48ms -> 4.2ms 0.94x
desk_yahooanswers.skp_1 6.84ms -> 6.3ms 0.92x
tabl_pravda.skp_1 11.1ms -> 10.1ms 0.91x
tabl_sahadan.skp_1 11.8ms -> 10.7ms 0.91x
desk_wowwiki.skp_1 8.48ms -> 7.58ms 0.89x
tabl_googleblog.skp_1 6ms -> 5.34ms 0.89x
tabl_worldjournal.skp_1 6.08ms -> 5.35ms 0.88x
desk_booking.skp_1 15.3ms -> 13.4ms 0.88x
tabl_androidpolice.skp_1 15.5ms -> 13.5ms 0.87x
desk_twitter.skp_1 12.2ms -> 10.6ms 0.87x
tabl_nofolo.skp_1 5.49ms -> 4.76ms 0.87x
desk_mobilenews.skp_1 22ms -> 18.9ms 0.86x
desk_forecastio.skp_1 9.47ms -> 8.05ms 0.85x
tabl_culturalsolutions.skp_1 6.21ms -> 5.28ms 0.85x
desk_youtube.skp_1 16.1ms -> 13.5ms 0.84x
tabl_mlb.skp_1 9.75ms -> 8.01ms 0.82x
tabl_digg.skp_1 5.2ms -> 4.22ms 0.81x
desk_blogger.skp_1 10.2ms -> 8.24ms 0.81x
desk_gmailthread.skp_1 26.8ms -> 21.6ms 0.81x
desk_googleplus.skp_1 10.5ms -> 8.39ms 0.8x
tabl_frantzen.skp_1 4.55ms -> 3.58ms 0.79x
desk_pinterest.skp_1 8.85ms -> 6.88ms 0.78x
desk_ebay.skp_1 10.5ms -> 8.15ms 0.77x
tabl_transformice.skp_1 4.93ms -> 3.5ms 0.71x
Results for arm_v7_neon:
desk_samoasvg.skp_1 13.9ms -> 14.6ms 1.05x
desk_mapsvg.skp_1 8.31ms -> 8.75ms 1.05x
tabl_deviantart.skp_1 1.41ms -> 1.45ms 1.02x
desk_weather.skp_1 3.8ms -> 3.88ms 1.02x
desk_sfgate.skp_1 3.06ms -> 3.1ms 1.01x
desk_css3gradients.skp_1 2.78ms -> 2.79ms 1x
desk_espn.skp_1 6.52ms -> 6.43ms 0.99x
desk_gws.skp_1 4.16ms -> 4.09ms 0.98x
tabl_cnn.skp_1 4.66ms -> 4.58ms 0.98x
tabl_hsfi.skp_1 3.49ms -> 3.42ms 0.98x
tabl_cuteoverload.skp_1 2.41ms -> 2.35ms 0.98x
desk_yahooanswers.skp_1 5.28ms -> 5.14ms 0.97x
desk_carsvg.skp_1 90.8ms -> 87.9ms 0.97x
tabl_gspro.skp_1 2.81ms -> 2.71ms 0.96x
desk_wowwiki.skp_1 5.85ms -> 5.63ms 0.96x
tabl_pravda.skp_1 7.8ms -> 7.5ms 0.96x
desk_twitter.skp_1 8.14ms -> 7.8ms 0.96x
tabl_androidpolice.skp_1 10.4ms -> 9.96ms 0.96x
tabl_googleblog.skp_1 4.06ms -> 3.83ms 0.95x
desk_mobilenews.skp_1 15.2ms -> 14ms 0.93x
desk_booking.skp_1 9.89ms -> 9.08ms 0.92x
desk_forecastio.skp_1 6.16ms -> 5.65ms 0.92x
desk_blogger.skp_1 6.17ms -> 5.66ms 0.92x
tabl_digg.skp_1 3.73ms -> 3.41ms 0.91x
tabl_nofolo.skp_1 3.82ms -> 3.47ms 0.91x
tabl_worldjournal.skp_1 4.24ms -> 3.84ms 0.9x
desk_youtube.skp_1 10.5ms -> 9.39ms 0.9x
desk_googleplus.skp_1 7.01ms -> 6.19ms 0.88x
tabl_mlb.skp_1 5.91ms -> 5.22ms 0.88x
tabl_googlecalendar.skp_1 10.7ms -> 9.44ms 0.88x
desk_gmailthread.skp_1 19.2ms -> 16.8ms 0.88x
desk_ebay.skp_1 5.68ms -> 4.93ms 0.87x
desk_pinterest.skp_1 5.99ms -> 5.08ms 0.85x
desk_googlehome.skp_1 3.31ms -> 2.71ms 0.82x
tabl_transformice.skp_1 3.03ms -> 2.44ms 0.81x
desk_amazon.skp_1 6.05ms -> 4.84ms 0.8x
desk_facebook.skp_1 12.6ms -> 9.62ms 0.76x
Review URL: https://codereview.chromium.org/
1120023005
senorblanco [Wed, 20 May 2015 21:03:42 +0000 (14:03 -0700)]
Use high precision vertex attributes.
Specifying vertex attributes and the view matrix as mediump causes bad artifacts on Mali T6XX.
Covered by the bigmatrix GM in msaa4 mode.
BUG:skia=3545,3771
Review URL: https://codereview.chromium.org/
1145273002
reed [Wed, 20 May 2015 20:22:58 +0000 (13:22 -0700)]
purge imagefilter cache so we can better track leaks
BUG=skia:
Review URL: https://codereview.chromium.org/
1148973002
reed [Wed, 20 May 2015 20:06:00 +0000 (13:06 -0700)]
check for big sizes in NewData
BUG=486977
Review URL: https://codereview.chromium.org/
1148873004
mtklein [Wed, 20 May 2015 19:05:15 +0000 (12:05 -0700)]
Don't serialize SkPictures in SkPictureShaders when in untrusted mode.
This requires we "first" add a has-picture bool to SkPictureShader serialized format.
BUG=chromium:486947, billions and billions of others.
Review URL: https://codereview.chromium.org/
1151663002
robertphillips [Wed, 20 May 2015 18:49:55 +0000 (11:49 -0700)]
Fix for bisector computation bug in GrAAConvexTessellator
Review URL: https://codereview.chromium.org/
1151623002
mtklein [Wed, 20 May 2015 17:55:49 +0000 (10:55 -0700)]
Have SkPicture inherit from SkRefCnt.
This re-enables adoption tracking for SkPictures in Blink,
which should be green now that crrev.com/
1136123011 has landed.
BUG=skia:3847
Review URL: https://codereview.chromium.org/
1145153002
mtklein [Wed, 20 May 2015 17:32:22 +0000 (10:32 -0700)]
Make SkEmptyPicture a singleton.
In my confusion yesterday I accidentally left this as a non-singleton.
The issue in Blink was not related to this being a singleton,
and it should be safe to make it one.
This means recording an empty picture properly costs zero mallocs.
BUG=skia:
Review URL: https://codereview.chromium.org/
1147053002
mtklein [Wed, 20 May 2015 17:16:49 +0000 (10:16 -0700)]
More efficient SkRecorder::flushMiniRecorder()
This should involve no mallocs, frees, or memcpys, and less refcounting.
BUG=skia:
Review URL: https://codereview.chromium.org/
1144173002
bungeman [Wed, 20 May 2015 16:26:47 +0000 (09:26 -0700)]
Sort build files for consistent link order.
Prior to the introduction of find.py, GMs were liked in the order they
were listed in the gypi file, which was generally alphabetically. This
made it fairly easy to predict where slides would show up in SampleApp
and the order was consistent. This simply sorts the list of files in
find.py to restore the expectation that files should be listed in the
build in alphabetical order.
Review URL: https://codereview.chromium.org/
1144973003
bungeman [Wed, 20 May 2015 16:21:04 +0000 (09:21 -0700)]
Font variations.
Multiple Master and TrueType fonts support variation axes.
This implements back-end support for axes on platforms which
support it.
Committed: https://skia.googlesource.com/skia/+/
05773ed30920c0214d1433c07cf6360a05476c97
Committed: https://skia.googlesource.com/skia/+/
3489ee0f4fa34f124f9de090d12bdc2107d52aa9
Review URL: https://codereview.chromium.org/
1027373002
caryclark [Wed, 20 May 2015 15:22:45 +0000 (08:22 -0700)]
fix win 8 bot
Comment out assert (runtime documentation)
It fails on Win 8 because the result isn't bit identical.
TBR=bungeman@google.com
Review URL: https://codereview.chromium.org/
1148053003
stephana [Wed, 20 May 2015 14:31:32 +0000 (07:31 -0700)]
Unmount before app-install
TBR=borenet
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1145093005
fs [Wed, 20 May 2015 07:52:17 +0000 (00:52 -0700)]
Don't fail SkMergeImageFilter if one of the inputs are empty
If one of the inputs to a SkMergeImageFilter was clipped away or
otherwise caused the filterImage(...) invocation for it to return
false, the entire effect would be "failed" and return false --
regardless of if it had produced a result or not.
Instead of returning false directly if filterImage(...) for a source
returned false, consider all the inputs, and then only return false if
all of them do.
BUG=chromium:489046
Review URL: https://codereview.chromium.org/
1133523006
joshualitt [Tue, 19 May 2015 22:05:24 +0000 (15:05 -0700)]
small cleanup of GrAtlas
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
d706f11b6e3f4398ab93b23458a7599ee324be2c
Review URL: https://codereview.chromium.org/
1142263002
bungeman [Tue, 19 May 2015 21:45:11 +0000 (14:45 -0700)]
Revert of small cleanup of GrAtlas (patchset #2 id:20001 of https://codereview.chromium.org/
1142263002/)
Reason for revert:
Broke compile on Linux
Original issue's description:
> small cleanup of GrAtlas
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
d706f11b6e3f4398ab93b23458a7599ee324be2c
TBR=robertphillips@google.com,joshualitt@google.com,jvanverth@google.com,joshualitt@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1142273002
joshualitt [Tue, 19 May 2015 21:41:41 +0000 (14:41 -0700)]
small cleanup of GrAtlas
BUG=skia:
Review URL: https://codereview.chromium.org/
1142263002
joshualitt [Tue, 19 May 2015 21:28:04 +0000 (14:28 -0700)]
Adding immediate mode draw target for debug
BUG=skia:
Review URL: https://codereview.chromium.org/
1126043007
reed [Tue, 19 May 2015 21:13:31 +0000 (14:13 -0700)]
implement --quiet for DM
BUG=skia:
Review URL: https://codereview.chromium.org/
1139413003
kkinnunen [Tue, 19 May 2015 20:50:50 +0000 (13:50 -0700)]
Add a newline to the font load debug message
Add a newline to the font load debug message. Helps reading nanobench
results. Otherwise the message "Resource /fonts/Funkster.ttf not a valid
font." causes first result be hard to read or missing.
Review URL: https://codereview.chromium.org/
1142183002
mtklein [Tue, 19 May 2015 18:11:26 +0000 (11:11 -0700)]
Sketch splitting SkPicture into an interface and SkBigPicture.
Adds small pictures for drawRect(), drawTextBlob(), and drawPath().
These cover about 89% of draw calls from Blink SKPs,
and about 25% of draw calls from our GMs.
SkPicture handles:
- serialization and deserialization
- unique IDs
Everything else is left to the subclasses:
- playback(), cullRect()
- hasBitmap(), hasText(), suitableForGPU(), etc.
- LayerInfo / AccelData if applicable.
The time to record a 1-op picture improves a good chunk
(2 mallocs to 1), and the time to record a 0-op picture
greatly improves (2 mallocs to none):
picture_overhead_draw: 450ns -> 350ns
picture_overhead_nodraw: 300ns -> 90ns
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
c92c129ff85b05a714bd1bf921c02d5e14651f8b
Latest blink_linux_rel:
http://build.chromium.org/p/tryserver.blink/builders/linux_blink_rel/builds/61248
Committed: https://skia.googlesource.com/skia/+/
15877b6eae33a9282458bdb904a6d00440eca0ec
http://build.chromium.org/p/tryserver.blink/builders/linux_blink_rel/builds/62015
Review URL: https://codereview.chromium.org/
1112523006
caryclark [Tue, 19 May 2015 18:05:37 +0000 (11:05 -0700)]
handle large conic strokes better
A stroked conic computes the outset quad's control point by
computing the intersection of the quad's endpoints. If the
the denominator used to compute the scale factor for the
control point is small, check to see if the numerator is also
small so that the division stays bounded.
Also clean up error returns and internal function calls to
simplify the code.
Additionally, remove comic max curvature (unimplemented) and call
extrema functions instead to handle cases where the conic is degenerate
or is a line.
R=reed@google.com, fmalita@chromium.org
BUG=skia:3843
Review URL: https://codereview.chromium.org/
1144883003
halcanary [Tue, 19 May 2015 17:21:29 +0000 (10:21 -0700)]
Documentation: SkCanvas API
NOTRY=true
DOCS_PREVIEW= https://skia.org/?cl=
1127383010
DOCS_PREVIEW= https://skia.org/user/api?cl=
1127383010
DOCS_PREVIEW= https://skia.org/user/api/skcanvas?cl=
1127383010
DOCS_PREVIEW= https://skia.org/user/api/skpaint?cl=
1127383010
DOCS_PREVIEW= https://skia.org/user/api/skrect?cl=
1127383010
DOCS_PREVIEW= https://skia.org/user/api/skregion?cl=
1127383010
DOCS_PREVIEW= https://skia.org/user/api/skmatrix?cl=
1127383010
DOCS_PREVIEW= https://skia.org/user/api/canvas?cl=
1127383010
Review URL: https://codereview.chromium.org/
1127383010
bsalomon [Tue, 19 May 2015 16:29:46 +0000 (09:29 -0700)]
rename GrDrawTargetCaps to GrCaps
Review URL: https://codereview.chromium.org/
1133123009
joshualitt [Tue, 19 May 2015 15:23:30 +0000 (08:23 -0700)]
Preliminary attempt to remove batch tracker
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
cbfe91d82500f4ae8c3ff7bd74b3021a4b89fd84
Review URL: https://codereview.chromium.org/
1139723004
joshualitt [Tue, 19 May 2015 14:15:28 +0000 (07:15 -0700)]
Revert of Preliminary attempt to remove batch tracker (patchset #3 id:40001 of https://codereview.chromium.org/
1139723004/)
Reason for revert:
breaking bots
Original issue's description:
> Preliminary attempt to remove batch tracker
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
cbfe91d82500f4ae8c3ff7bd74b3021a4b89fd84
TBR=robertphillips@google.com,joshualitt@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1132323003
joshualitt [Tue, 19 May 2015 13:49:32 +0000 (06:49 -0700)]
Preliminary attempt to remove batch tracker
BUG=skia:
Review URL: https://codereview.chromium.org/
1139723004
stephana [Tue, 19 May 2015 13:27:09 +0000 (06:27 -0700)]
This allows to restart the device.
BUG=skia:
TBR=borenet
Review URL: https://codereview.chromium.org/
1137873007
robertphillips [Tue, 19 May 2015 11:35:38 +0000 (04:35 -0700)]
Fix GrAARectRenderer's handling of unsorted nested rects
What is going on here is that, after the mapPoints in fillAANestedRects, devInside was upside down so the isEmpty check was always firing. I don't see why we need to avoid having devInside sorted.
BUG=488103
Review URL: https://codereview.chromium.org/
1135753004
kkinnunen [Tue, 19 May 2015 06:02:07 +0000 (23:02 -0700)]
Improve caching of dashed paths in GrStencilAndCoverPathRenderer
Improve caching of dashed paths in GrStencilAndCoverPathRenderer.
Look up the (NVPR specific) GrGLPath based on GrStrokeInfo and
the original path.
Use unique keys for all GrPaths.
Dash the path with Skia dash stroker and use that path geometry for
NVPR path.
NVPR internal dashing stroke is not used, because the dashing
implementation of NVPR does not match Skia implementation.
Review URL: https://codereview.chromium.org/
1116123003
kkinnunen [Tue, 19 May 2015 05:47:33 +0000 (22:47 -0700)]
Make GrResourceCache perf less sensitive to key length change
Make GrResourceCache performance less sensitive to key length change.
The memcmp in GrResourceKey is called when SkTDynamicHash jumps the
slots to find the hash by a index. Avoid most of the memcmps by
comparing the hash first.
This is important because small changes in key data length can cause
big performance regressions. The theory is that key length change causes
different hash values. These hash values might trigger memcmps that
originally weren't there, causing the regression.
Adds few specialized benches to grresourcecache_add to test different
key lengths. The tests are run only on release, because on debug the
SkTDynamicHash validation takes too long, and adding many such delays
to development test runs would be unproductive. On release the tests
are quite fast.
Effect of this patch to the added tests on amd64:
grresourcecache_find_10 738us -> 768us 1.04x
grresourcecache_find_2 472us -> 476us 1.01x
grresourcecache_find_25 841us -> 845us 1x
grresourcecache_find_4 565us -> 531us 0.94x
grresourcecache_find_54 1.18ms -> 1.1ms 0.93x
grresourcecache_find_5 834us -> 749us 0.9x
grresourcecache_find_3 620us -> 542us 0.87x
grresourcecache_add_25 2.74ms -> 2.24ms 0.82x
grresourcecache_add_56 3.23ms -> 2.56ms 0.79x
grresourcecache_add_54 3.34ms -> 2.62ms 0.78x
grresourcecache_add_5 2.68ms -> 2.1ms 0.78x
grresourcecache_add_10 2.7ms -> 2.11ms 0.78x
grresourcecache_add_2 1.85ms -> 1.41ms 0.76x
grresourcecache_add 1.84ms -> 1.4ms 0.76x
grresourcecache_add_4 1.99ms -> 1.49ms 0.75x
grresourcecache_add_3 2.11ms -> 1.55ms 0.73x
grresourcecache_add_55 39ms -> 13.9ms 0.36x
grresourcecache_find_55 23.2ms -> 6.21ms 0.27x
On arm64 the results are similar.
On arm_v7_neon, the results lack the discontinuity at 55:
grresourcecache_add 4.06ms -> 4.26ms 1.05x
grresourcecache_add_2 4.05ms -> 4.23ms 1.05x
grresourcecache_find 1.28ms -> 1.3ms 1.02x
grresourcecache_find_56 3.35ms -> 3.32ms 0.99x
grresourcecache_find_2 1.31ms -> 1.29ms 0.99x
grresourcecache_find_54 3.28ms -> 3.24ms 0.99x
grresourcecache_add_5 6.38ms -> 6.26ms 0.98x
grresourcecache_add_55 8.44ms -> 8.24ms 0.98x
grresourcecache_add_25 7.03ms -> 6.86ms 0.98x
grresourcecache_find_25 2.7ms -> 2.59ms 0.96x
grresourcecache_find_4 1.45ms -> 1.38ms 0.95x
grresourcecache_find_10 2.52ms -> 2.39ms 0.95x
grresourcecache_find_55 3.54ms -> 3.33ms 0.94x
grresourcecache_find_5 2.5ms -> 2.32ms 0.93x
grresourcecache_find_3 1.57ms -> 1.43ms 0.91x
The extremely slow case, 55, is postulated to be due to the index jump
collisions running the memcmp. This is not visible on arm_v7_neon probably due
to hash function producing different results for 32 bit architectures.
This change is needed for extending path cache key in Gr
NV_path_rendering codepath. Extending is needed in order to add dashed
paths to the path cache.
Review URL: https://codereview.chromium.org/
1132723003
kkinnunen [Tue, 19 May 2015 05:25:47 +0000 (22:25 -0700)]
Improve NVPR stroke accuracy to fix thick strokes
Set the "path stroke error bound" path parameter to 0.02 for all paths.
This means that the stroked path area will be within 98% of the stroke
width in path space.
This should fix many cases where NVPR stroked paths were visibly different to
Skia stroked paths. One such path is in dashcubics gm.
This increases the amount of subdivisions the path object creation will
make for paths that need it. This in turn will increase gpu object space
requirements sligthly. Both of these effects should be unnoticeable.
GL_NV_path_rendering.txt:
"""
Every path object has a stroke approximation bound parameter
(PATH_STROKE_BOUND_NV) that is a floating-point value /sab/ clamped
between 0.0 and 1.0 and set and queried with the PATH_STROKE_BOUND_NV
path parameter. Exact determination of samples swept an orthogonal
centered line segment along cubic Bezier segments and rational
quadratic Bezier curves (so non-circular partial elliptical arcs) is
intractable for real-time rendering so an approximation is required;
/sab/ intuitively bounds the approximation error as a percentage of
the path object's stroke width. Specifically, this path parameter
requests the implementation to stencil any samples within /sweep/
object space units of the exact sweep of the path's cubic Bezier
segments or partial elliptical arcs to be sampled by the stroke where
sweep = ((1-sab)*sw)/2
where /sw/ is the path object's stroke width. The initial value
of /sab/ when a path is created is 0.2. In practical terms, this
initial value means the stencil sample positions coverage within 80%
(100%-20%) of the stroke width of cubic and rational quadratic stroke
segments should be sampled.
"""
BUG=skia:2049
Review URL: https://codereview.chromium.org/
1124423007
kkinnunen [Tue, 19 May 2015 05:23:54 +0000 (22:23 -0700)]
Make GrStrokeInfo inherit from SkStrokeRec
Make the code more readable by inheriting GrStrokeInfo from SkStrokeRec.
This should avoid the long .getStrokeRec() and .getStrokeRecPtr(). These
were a bit cumbersome especially in cases where an alias variable was
created for these, and then the reader had to keep track to which
StrokeInfo member the StrokeRec alias was pointing.
Removes SkStrokeRec::SkStrokeRec(const SkStrokeRec&). It was memcpying.
Try to play it safe wrt compiler using the possible padding of
superclass for subclass members. Instead, let the compiler generate
the copy constructor. Assignment operator was already
compiler-generated, so at least in that way this is consistent.
Renames GrStrokeInfo::applyDash to applyDashToPath for consistency
with superclass applyToPath.
Review URL: https://codereview.chromium.org/
1128113008
mtklein [Mon, 18 May 2015 21:53:43 +0000 (14:53 -0700)]
Revert of Sketch splitting SkPicture into an interface and SkBigPicture. (patchset #25 id:480001 of https://codereview.chromium.org/
1112523006/)
Reason for revert:
win_chromium_compile_dbg_ng
FAILED: ninja -t msvc -e environment.x86 -- E:\b\build\goma/gomacc "E:\b\depot_tools\win_toolchain\vs2013_files\VC\bin\amd64_x86\cl.exe" /nologo /showIncludes /FC @obj\third_party\skia\src\core\skia.SkBitmapHeap.obj.rsp /c ..\..\third_party\skia\src\core\SkBitmapHeap.cpp /Foobj\third_party\skia\src\core\skia.SkBitmapHeap.obj /Fdobj\skia\skia.cc.pdb
e:\b\build\slave\win\build\src\third_party\skia\include\core\skpicture.h(176) : error C2487: 'CURRENT_PICTURE_VERSION' : member of dll interface class may not be declared with dll interface
Original issue's description:
> Sketch splitting SkPicture into an interface and SkBigPicture.
>
> Adds small pictures for drawRect(), drawTextBlob(), and drawPath().
> These cover about 89% of draw calls from Blink SKPs,
> and about 25% of draw calls from our GMs.
>
> SkPicture handles:
> - serialization and deserialization
> - unique IDs
>
> Everything else is left to the subclasses:
> - playback(), cullRect()
> - hasBitmap(), hasText(), suitableForGPU(), etc.
> - LayerInfo / AccelData if applicable.
>
> The time to record a 1-op picture improves a good chunk
> (2 mallocs to 1), and the time to record a 0-op picture
> greatly improves (2 mallocs to none):
>
> picture_overhead_draw: 450ns -> 350ns
> picture_overhead_nodraw: 300ns -> 90ns
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
c92c129ff85b05a714bd1bf921c02d5e14651f8b
>
> Latest blink_linux_rel:
>
> http://build.chromium.org/p/tryserver.blink/builders/linux_blink_rel/builds/61248
>
> Committed: https://skia.googlesource.com/skia/+/
15877b6eae33a9282458bdb904a6d00440eca0ec
TBR=reed@google.com,robertphillips@google.com,fmalita@chromium.org,mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1130283004
stephana [Mon, 18 May 2015 20:50:05 +0000 (13:50 -0700)]
Making mount more robust
BUG=skia:
TBR=borenet
Review URL: https://codereview.chromium.org/
1131933007
mtklein [Mon, 18 May 2015 20:47:17 +0000 (13:47 -0700)]
Sketch splitting SkPicture into an interface and SkBigPicture.
Adds small pictures for drawRect(), drawTextBlob(), and drawPath().
These cover about 89% of draw calls from Blink SKPs,
and about 25% of draw calls from our GMs.
SkPicture handles:
- serialization and deserialization
- unique IDs
Everything else is left to the subclasses:
- playback(), cullRect()
- hasBitmap(), hasText(), suitableForGPU(), etc.
- LayerInfo / AccelData if applicable.
The time to record a 1-op picture improves a good chunk
(2 mallocs to 1), and the time to record a 0-op picture
greatly improves (2 mallocs to none):
picture_overhead_draw: 450ns -> 350ns
picture_overhead_nodraw: 300ns -> 90ns
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
c92c129ff85b05a714bd1bf921c02d5e14651f8b
Latest blink_linux_rel:
http://build.chromium.org/p/tryserver.blink/builders/linux_blink_rel/builds/61248
Review URL: https://codereview.chromium.org/
1112523006
mtklein [Mon, 18 May 2015 20:44:35 +0000 (13:44 -0700)]
Empty picture shaders should draw nothing.
BUG=skia:
Review URL: https://codereview.chromium.org/
1142053004
halcanary [Mon, 18 May 2015 20:15:56 +0000 (13:15 -0700)]
SkPDF: Add Sfntly to DEPS, gyp
Note: this can be disabled via:
GYP_DEFINES='skia_pdf_use_sfntly=0
Warning: dm is 34% slower and uses 9% more memory. This is
okay.
Motivation: We want to test this code path in DM, since it is
always used by Chromium and Android.
BUG=skia:3563
Committed: https://skia.googlesource.com/skia/+/
6a53b04e26749ea61f690ece408f2a1c0a5ad5bb
Review URL: https://codereview.chromium.org/
1134683006
djsollen [Mon, 18 May 2015 20:05:11 +0000 (13:05 -0700)]
Prevent integer wrap around for malloc size when creating a SkRegion
BUG=484998
Review URL: https://codereview.chromium.org/
1143603003
egdaniel [Mon, 18 May 2015 20:04:26 +0000 (13:04 -0700)]
Don't do discards when on Adreno4xx due to driver bug.
BUG=skia:
Review URL: https://codereview.chromium.org/
1126423006
reed [Mon, 18 May 2015 19:57:56 +0000 (12:57 -0700)]
add GM to test extractAlpha
BUG=skia:
TBR=
Review URL: https://codereview.chromium.org/
1142123003
caryclark [Mon, 18 May 2015 19:56:57 +0000 (12:56 -0700)]
fix builder winding again
Record the nesting level when finding the edge winding contribution
so that inner edges can be reversed as needed.
R=fmalita@chromium.org
BUG=skia:3838
Review URL: https://codereview.chromium.org/
1140383002
Florin Malita [Mon, 18 May 2015 19:14:08 +0000 (15:14 -0400)]
Remove SK_LEGACY_DRAWPICTURECALLBACK from skia_for_chromium_defines.gypi
The flag now lives in Chromium's SkUserConfig.h.
R=reed@google.com
Review URL: https://codereview.chromium.org/
1133243009
halcanary [Mon, 18 May 2015 17:27:16 +0000 (10:27 -0700)]
Revert of SkPDF: Add Sfntly to DEPS, gyp, SkUserConfig (patchset #7 id:570001 of https://codereview.chromium.org/
1134683006/)
Reason for revert:
break cros build
Original issue's description:
> SkPDF: Add Sfntly to DEPS, gyp
>
> Note: this can be disabled via:
> GYP_DEFINES='skia_pdf_use_sfntly=0
>
> Warning: dm is 34% slower and uses 9% more memory. This is
> okay.
>
> Motivation: We want to test this code path in DM, since it is
> always used by Chromium and Android.
>
> BUG=skia:3563
>
> Committed: https://skia.googlesource.com/skia/+/
6a53b04e26749ea61f690ece408f2a1c0a5ad5bb
TBR=reed@google.com,mtklein@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:3563
Review URL: https://codereview.chromium.org/
1128353004
halcanary [Mon, 18 May 2015 17:18:58 +0000 (10:18 -0700)]
SkPDF: Add Sfntly to DEPS, gyp
Note: this can be disabled via:
GYP_DEFINES='skia_pdf_use_sfntly=0
Warning: dm is 34% slower and uses 9% more memory. This is
okay.
Motivation: We want to test this code path in DM, since it is
always used by Chromium and Android.
BUG=skia:3563
Review URL: https://codereview.chromium.org/
1134683006
mtklein [Mon, 18 May 2015 16:29:10 +0000 (09:29 -0700)]
Revert of Move copy-surface-as-draw fallback to GrGLGpu. (patchset #12 id:220001 of https://codereview.chromium.org/
1144433002/)
Reason for revert:
Many GMs changed, some totally busted.
Original issue's description:
> Move copy-surface-as-draw fallback to GrGLGpu.
>
> Committed: https://skia.googlesource.com/skia/+/
5df6fee929823f50c55cc50f7c882a309c1b1de9
TBR=joshualitt@google.com,egdaniel@google.com,bsalomon@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review URL: https://codereview.chromium.org/
1142003003
mtklein [Mon, 18 May 2015 14:03:01 +0000 (07:03 -0700)]
sk4px the rest of the easy xfermodes.
Adds and uses fastMulDiv255Round() where possible,
which approximates x*y/255 as (x*y+x)/256. Seems like a sizeable
speedup, as seen below on Exclusion, Screen, and Modulate. The
existing NEON code uses this approximation for
{Src,Dst}x{In,Out,Over}, and without it we'd regress speed there.
This will require rebaselines whether or not we use this
approximation: the x86 bots change if we do, the ARM bots change
if we don't. None of the diffs are significant.
Desktop:
Xfermode_Screen_aa 5.82ms -> 5.54ms 0.95x
Xfermode_Modulate_aa 5.67ms -> 5.36ms 0.95x
Xfermode_Exclusion_aa 6.18ms -> 5.81ms 0.94x
Xfermode_Exclusion 5.03ms -> 4.24ms 0.84x
Xfermode_Screen 4.51ms -> 3.59ms 0.8x
Xfermode_Modulate 4.2ms -> 3.19ms 0.76x
Xfermode_DstOver 6.73ms -> 3.88ms 0.58x
Xfermode_SrcOut 6.47ms -> 3.48ms 0.54x
Xfermode_SrcIn 6.46ms -> 3.46ms 0.54x
Xfermode_DstOut 6.49ms -> 3.41ms 0.52x
Xfermode_DstIn 6.5ms -> 3.32ms 0.51x
Xfermode_Src_aa 9.53ms -> 4.75ms 0.5x
Xfermode_Clear_aa 9.65ms -> 4.8ms 0.5x
Xfermode_DstIn_aa 11.5ms -> 5.57ms 0.49x
Xfermode_DstOver_aa 11.6ms -> 5.63ms 0.49x
Xfermode_SrcOut_aa 11.6ms -> 5.5ms 0.47x
Xfermode_SrcIn_aa 11.7ms -> 5.51ms 0.47x
Xfermode_DstOut_aa 11.7ms -> 5.4ms 0.46x
N7 performance is close enough to 1x that I'm not sure whether
this is a net win, net loss, or truly neutral. I figure the bots will
show that.
I experimented with another approximation,
(x*(255-y))/255 ≈ (x*(256-y))/256. This was inconclusive, so I'm
leaving it out for now.
The remaining modes are the complicated conditional ones.
BUG=skia:
Review URL: https://codereview.chromium.org/
1141953004
scroggo [Mon, 18 May 2015 13:41:35 +0000 (06:41 -0700)]
Set OpenGLRenderer light center separately from initial setup
Cherry-pick from Android.
Fixes master-skia build.
BUG=b/
16523629
Review URL: https://codereview.chromium.org/
1134403003
bsalomon [Mon, 18 May 2015 13:26:15 +0000 (06:26 -0700)]
Move copy-surface-as-draw fallback to GrGLGpu.
Review URL: https://codereview.chromium.org/
1144433002
caryclark [Mon, 18 May 2015 12:12:56 +0000 (05:12 -0700)]
The path ops builder code needs to determine the winding of each contour added, and reverse windings if the contours are nested in other contours.
Cheap (one contour) paths can be evaluated and reversed as needed with a minimum of checking, but multi-contour paths invoke the regular path ops machinery to determine who is contained by whom.
More tests need to be added to verify that all corner cases are considered, but this fixes the cases in the bug thus far.
R=fmalita@chromium.org
TBR=reed@google.com
BUG=skia:3838
Review URL: https://codereview.chromium.org/
1129193006
halcanary [Sun, 17 May 2015 13:47:49 +0000 (06:47 -0700)]
Gyp: Sanity Part II (GM+Tests)
DOCS_PREVIEW= https://skia.org/dev/testing/tests?cl=
1133343006
Review URL: https://codereview.chromium.org/
1133343006
skia.buildbots [Sun, 17 May 2015 08:14:01 +0000 (01:14 -0700)]
Update SKP version
Automatic commit by the RecreateSKPs bot.
TBR=
NO_MERGE_BUILDS
Review URL: https://codereview.chromium.org/
1128353003