robertphillips [Thu, 15 Oct 2015 15:01:48 +0000 (08:01 -0700)]
GrDrawContext now holds GrRenderTarget pointer
Review URL: https://codereview.chromium.org/
1404823005
scroggo [Thu, 15 Oct 2015 14:51:28 +0000 (07:51 -0700)]
Renable image benchmarking
- Remove --images '' to renable image benchmarking
- Add a flag to disable testing JPEG's buildTileIndex, since it also leaks memory
- Do not run images on GPU
- Do not run large interlaced images on 32 bit bots
- When buildTileIndex is not being used in the subset benches, do not use it for BRD
BUG=skia:3418
BUG=skia:4469
BUG=skia:4471
BUG=skia:4360
Review URL: https://codereview.chromium.org/
1396113002
joshualitt [Thu, 15 Oct 2015 14:18:29 +0000 (07:18 -0700)]
small tidy of benchmarkstream
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
691b6af907e55250a29a7a2a346b63c2026011c3
Review URL: https://codereview.chromium.org/
1395703002
msarett [Thu, 15 Oct 2015 13:41:06 +0000 (06:41 -0700)]
Reenable warnings in src/codec
BUG=skia:
Review URL: https://codereview.chromium.org/
1400343005
senorblanco [Thu, 15 Oct 2015 12:04:27 +0000 (05:04 -0700)]
Add a comment to explain clip intersection in crop rects.
BUG=skia:
R=robertphillips
Review URL: https://codereview.chromium.org/
1401973003
ericrk [Thu, 15 Oct 2015 00:33:29 +0000 (17:33 -0700)]
Fix floating point data loss in SkPictureShader
SkPictureShader had an invalid use of SkMax32 on two SkScalars which
ended up converting the result to an integer which caused us to lose
necessary floating point data.
BUG=473166
Review URL: https://codereview.chromium.org/
1401743004
msarett [Wed, 14 Oct 2015 22:21:11 +0000 (15:21 -0700)]
Revert of Supply separate flags for onBuildTileIndex (patchset #1 id:1 of https://codereview.chromium.org/
1402783008/ )
Reason for revert:
Missing an important line.
Original issue's description:
> Supply separate flags for onBuildTileIndex
>
> This is a follow up to:
> https://codereview.chromium.org/
1401283003/
>
> Supply separate flags for onBuildTileIndex
>
> Since png and jpeg's implementations of onBuildTileIndex rely on
> modifications to their underlying libraries, rather than whether we are
> running on Android, use separate flags that can be disabled
> independently.
>
> This will allow us to easily turn off the feature. It also is a step
> towards building and running on other platforms for testing (e.g.
> valgrind/ASAN to find memory leaks etc).
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
fc06e9c0e621744654e231ae6fa4460d88c0e27e
TBR=scroggo@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1394433005
bsalomon [Wed, 14 Oct 2015 22:01:50 +0000 (15:01 -0700)]
Don't abort analytic clipping if using MSAA until elements are examined
TBR=joshualitt@google.com
Review URL: https://codereview.chromium.org/
1399223005
joshualitt [Wed, 14 Oct 2015 21:45:07 +0000 (14:45 -0700)]
Revert of small tidy of benchmarkstream (patchset #2 id:20001 of https://codereview.chromium.org/
1395703002/ )
Reason for revert:
Breaks visualbench
Original issue's description:
> small tidy of benchmarkstream
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
691b6af907e55250a29a7a2a346b63c2026011c3
TBR=robertphillips@google.com,joshualitt@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1392833006
msarett [Wed, 14 Oct 2015 21:42:44 +0000 (14:42 -0700)]
Supply separate flags for onBuildTileIndex
This is a follow up to:
https://codereview.chromium.org/
1401283003/
Supply separate flags for onBuildTileIndex
Since png and jpeg's implementations of onBuildTileIndex rely on
modifications to their underlying libraries, rather than whether we are
running on Android, use separate flags that can be disabled
independently.
This will allow us to easily turn off the feature. It also is a step
towards building and running on other platforms for testing (e.g.
valgrind/ASAN to find memory leaks etc).
BUG=skia:
Review URL: https://codereview.chromium.org/
1402783008
bsalomon [Wed, 14 Oct 2015 19:54:52 +0000 (12:54 -0700)]
Store texture target on GrGLTexture. This is in preparation for using texture targets other than 2D.
Review URL: https://codereview.chromium.org/
1403123002
scroggo [Wed, 14 Oct 2015 16:00:35 +0000 (09:00 -0700)]
Remove unnecessary allocation in nanobench
Review URL: https://codereview.chromium.org/
1408593002
bsalomon [Wed, 14 Oct 2015 15:31:34 +0000 (08:31 -0700)]
Fix clipping when all analytic FP creations can be skipped
BUG=skia:
Review URL: https://codereview.chromium.org/
1400343004
fmalita [Wed, 14 Oct 2015 15:11:40 +0000 (08:11 -0700)]
Fix TextBlobTest valgrind error
When converting text to glyph IDs, the paint needs to reflect the actual
text encoding.
R=bungeman@google.com,herb@google.com
Review URL: https://codereview.chromium.org/
1404153002
joshualitt [Wed, 14 Oct 2015 15:04:22 +0000 (08:04 -0700)]
small tidy of benchmarkstream
BUG=skia:
Review URL: https://codereview.chromium.org/
1395703002
scroggo [Wed, 14 Oct 2015 14:54:34 +0000 (07:54 -0700)]
Supply separate flags for onBuildTileIndex
Since png and jpeg's implementations of onBuildTileIndex rely on
modifications to their underlying libraries, rather than whether we are
running on Android, use separate flags that can be disabled
independently.
This will allow us to easily turn off the feature. It also is a step
towards building and running on other platforms for testing (e.g.
valgrind/ASAN to find memory leaks etc).
Review URL: https://codereview.chromium.org/
1401283003
reed [Wed, 14 Oct 2015 14:49:35 +0000 (07:49 -0700)]
change back to method for localmatrix imagefilter
This method is different from MatrixFilter, in that MatrixFilter does not require a pre-existing
filter, but LocalM does. Also change the comment to be more general, as there is no promise that
we return a different subclass, and certainly not a specific subclass.
This pattern of obj->newWithModifiers() also more closely matches the pattern in SkImage (newSubset).
BUG=skia:
Review URL: https://codereview.chromium.org/
1402133002
joshualitt [Wed, 14 Oct 2015 14:49:28 +0000 (07:49 -0700)]
Create VisualStreamTimingModule
BUG=skia:
Review URL: https://codereview.chromium.org/
1385073002
senorblanco [Wed, 14 Oct 2015 11:53:31 +0000 (04:53 -0700)]
Implement SkLocalMatrixImageFilter.
At draw time, this filter simply concatenates the given matrix to the
CTM, and recurses on its input. The matrix is thus applied to any
upstream filter parameters and crop rects.
BUG=skia:
Review URL: https://codereview.chromium.org/
1392833005
scroggo [Tue, 13 Oct 2015 20:33:33 +0000 (13:33 -0700)]
Stop benching PNG buildTileIndex by default
It leaks memory and kills the bots. Supply an option to run it for
local comparison.
BUG=skia:4360
BUG=skia:3418
Review URL: https://codereview.chromium.org/
1396613007
kkinnunen [Tue, 13 Oct 2015 19:58:37 +0000 (12:58 -0700)]
Fix dm with Chromium command buffer driver
Fix a crash in dm when run with Chromium command buffer driver.
Also removes glCopyTextureCHROMIUM, it does not seem to be used.
This is removed on the grounds that it would need similar change,
if it was used. The variable fCopyTextureCHROMIUM does not seem to
be populated in the Chromium side, either.
BUG=skia:2992
Review URL: https://codereview.chromium.org/
1389213004
msarett [Tue, 13 Oct 2015 19:50:14 +0000 (12:50 -0700)]
Add subsetting to SkScanlineDecoder
This CL allows the SkScanlineDecoder to decode partial
scanlines.
This is a first step in efficiently implementing subsetting
in SkScaledCodec.
BUG=skia:4209
Review URL: https://codereview.chromium.org/
1390213002
herb [Tue, 13 Oct 2015 18:47:54 +0000 (11:47 -0700)]
Fix text axis alignment calculation
BUG=skia:
Review URL: https://codereview.chromium.org/
1405583002
fmalita [Tue, 13 Oct 2015 15:59:23 +0000 (08:59 -0700)]
[SkTextBlob] Remove incorrect builder assert
At the end of TightRunBounds, glyphPosX cannot exceed the start of the
next run. But glyphPosY is running ahead of glyphPosX (for
kFull_Positioning) => the glyphPosY assert is incorrect.
Since the two pointers advance in lock-step, there isn't much value in
the glyphPosY assert anyway - we might as well remove it.
BUG=chromium:542643
R=reed@google.com,bungeman@google.com
Review URL: https://codereview.chromium.org/
1405463004
rmistry [Tue, 13 Oct 2015 11:44:40 +0000 (04:44 -0700)]
Whitespace change to test CQ waiting for l-g-t-m
BUG=skia:
Review URL: https://codereview.chromium.org/
1401333002
reed [Mon, 12 Oct 2015 21:38:34 +0000 (14:38 -0700)]
adjust size of new apply-filter gm
BUG=skia:
TBR=
Review URL: https://codereview.chromium.org/
1395303003
msarett [Mon, 12 Oct 2015 20:33:18 +0000 (13:33 -0700)]
SkBitmapRegionCanvas bug fix
BUG=skia:
Review URL: https://codereview.chromium.org/
1393423004
reed [Mon, 12 Oct 2015 20:17:23 +0000 (13:17 -0700)]
post-commit fixes to applyFilter CL
BUG=skia:
Review URL: https://codereview.chromium.org/
1404483002
rmistry [Mon, 12 Oct 2015 20:07:28 +0000 (13:07 -0700)]
Whitespace change to turn Android bots green
BUG=skia:
TBR=borenet
Review URL: https://codereview.chromium.org/
1405433002
halcanary [Mon, 12 Oct 2015 20:05:04 +0000 (13:05 -0700)]
SkPDF: fall back on paths for unembeddable fonts.
Add GM, SkPDFFont::CanEmbedTypeface
BUG=skia:3866
Review URL: https://codereview.chromium.org/
1401763002
reed [Mon, 12 Oct 2015 18:30:02 +0000 (11:30 -0700)]
add applyFilter() to SkImage
Result:
- clients can get a filtered version of an image without having to setup a temp drawing environment
- for some cases, the process is more efficient even than (deprecated) drawSprite, since there is no need to draw/copy the result
Impl:
- made Proxy virtual so we don't need to have an existing device to use it
This, in conjunction with LocalMatrixImageFilter, should allow us to simplify and optimize ApplyImageFilter() in cc/output/gl_renderer.cc
BUG=skia:
Review URL: https://codereview.chromium.org/
1390913005
reed [Mon, 12 Oct 2015 18:10:10 +0000 (11:10 -0700)]
change more effects to only expose factories
BUG=skia:
TBR=
Review URL: https://codereview.chromium.org/
1395403002
fmalita [Mon, 12 Oct 2015 17:41:48 +0000 (10:41 -0700)]
Remove SK_SUPPORT_LEGACY_GRADIENT_DITHERING from Skia proper
Migrating the flag to embedder defines (Chromium already guarded).
Also augment gradient-focused GMs to generate both dithered/undithered
results.
BUG=skia:4436
R=reed@google.com,robertphillips@google.com
Review URL: https://codereview.chromium.org/
1400813006
bsalomon [Mon, 12 Oct 2015 17:39:46 +0000 (10:39 -0700)]
Remove image usage type enum. Use GrTextureParams instead.
BUG=skia:
Review URL: https://codereview.chromium.org/
1404433002
msarett [Mon, 12 Oct 2015 17:24:38 +0000 (10:24 -0700)]
SkBitmapRegionDecoder clean-up
Use SkCodecPrintf instead of SkDebugf.
Check if the conversion is possible rather than starting many decodes
that will certainly fail.
Small refactor to code that deals with subsets that fall outside
of the image.
BUG=skia:
Review URL: https://codereview.chromium.org/
1395383002
halcanary [Mon, 12 Oct 2015 17:11:18 +0000 (10:11 -0700)]
SkPDF: Optionally output PDF/A-2b archive format.
Note: this format does not yet pass validation tests.
Add skia_pdf_generate_pdfa GYP flag. Default to off for now.
PDF/A files are not reproducable, so they make correctness
testing harder.
Turn the Metadata struct into te SkPDFMetadata struct. This
splits out a lot of functionality around both kinds of metadata.
When PDF/A is used, add an ID entry to the trailer.
Add SkPDFObjNumMap::addObjectRecursively.
Test with
GYP_DEFINES=skia_pdf_generate_pdfa=1 bin/sync-and-gyp
ninja -C out/Release dm
out/Release/dm --config pdf --src skp gm -w /tmp/dm
With skia_pdf_generate_pdfa=0, all PDFs generated from GMs and
SKPs are identical. With skia_pdf_generate_pdfa=1, all PDFs
generated from GMs and SKPs render identically in Pdfium.
BUG=skia:3110
Review URL: https://codereview.chromium.org/
1394263003
msarett [Mon, 12 Oct 2015 15:14:40 +0000 (08:14 -0700)]
Enable more testing for scaled webps
BUG=skia:
Review URL: https://codereview.chromium.org/
1401093002
update-skps [Sun, 11 Oct 2015 07:20:57 +0000 (00:20 -0700)]
Update SKP version
Automatic commit by the RecreateSKPs bot.
TBR=
NO_MERGE_BUILDS
Review URL: https://codereview.chromium.org/
1401013002
fmalita [Fri, 9 Oct 2015 21:46:28 +0000 (14:46 -0700)]
[TextBlob] Fall back to TightRunBounds when the font bounds are empty
Empty font bounds are likely an indication of a font bug. As a best
effort, we can use TightRunBounds in this easily detectable case.
Since TightRunBounds only supports kDefault_Positioning currently, the
CL also reinstates handling of kFull_Positioning and
kHorizontal_Positioning (removed in http://crrev.com/
858153007).
BUG=507022
R=reed@google.com,bungeman@google.com
Review URL: https://codereview.chromium.org/
1399123002
Brian Salomon [Fri, 9 Oct 2015 20:54:48 +0000 (16:54 -0400)]
Fix SkWindow.cpp includes for non-gpu mode
TBR=bungeman@google.com
Review URL: https://codereview.chromium.org/
1400933002 .
bungeman [Fri, 9 Oct 2015 20:45:50 +0000 (13:45 -0700)]
Revert of SkPDF: Optionally output PDF/A-2b archive format. (patchset #5 id:80001 of https://codereview.chromium.org/
1394263003/ )
Reason for revert:
SkMD5 is not really part of the Skia library. This is breaking the roll by using it, since Chromium doesn't build it.
Original issue's description:
> SkPDF: Optionally output PDF/A-2b archive format.
>
> Note: this format does not yet pass validation tests.
>
> Add skia_pdf_generate_pdfa GYP flag. Default to off for now.
> PDF/A files are not reproducable, so they make correctness
> testing harder.
>
> Turn the Metadata struct into te SkPDFMetadata struct. This
> splits out a lot of functionality around both kinds of metadata.
>
> When PDF/A is used, add an ID entry to the trailer.
>
> Add SkPDFObjNumMap::addObjectRecursively.
>
> Test with
>
> GYP_DEFINES=skia_pdf_generate_pdfa=1 bin/sync-and-gyp
> ninja -C out/Release dm
> out/Release/dm --config pdf --src skp gm -w /tmp/dm
>
> With skia_pdf_generate_pdfa=0, all PDFs generated from GMs and
> SKPs are identical. With skia_pdf_generate_pdfa=1, all PDFs
> generated from GMs and SKPs render identically in Pdfium.
>
> BUG=skia:3110
>
> Committed: https://skia.googlesource.com/skia/+/
939c0fe51f157104758bcb268643c8b6d317a530
TBR=tomhudson@google.com,halcanary@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:3110
Review URL: https://codereview.chromium.org/
1398193002
bsalomon [Fri, 9 Oct 2015 20:36:42 +0000 (13:36 -0700)]
Move functions from SkGr to SkGrPriv.h
Review URL: https://codereview.chromium.org/
1397123002
jvanverth [Fri, 9 Oct 2015 20:27:33 +0000 (13:27 -0700)]
Add test with GLFW framework
BUG=skia:
Review URL: https://codereview.chromium.org/
1367403006
halcanary [Fri, 9 Oct 2015 20:09:58 +0000 (13:09 -0700)]
SkPDF: Optionally output PDF/A-2b archive format.
Note: this format does not yet pass validation tests.
Add skia_pdf_generate_pdfa GYP flag. Default to off for now.
PDF/A files are not reproducable, so they make correctness
testing harder.
Turn the Metadata struct into te SkPDFMetadata struct. This
splits out a lot of functionality around both kinds of metadata.
When PDF/A is used, add an ID entry to the trailer.
Add SkPDFObjNumMap::addObjectRecursively.
Test with
GYP_DEFINES=skia_pdf_generate_pdfa=1 bin/sync-and-gyp
ninja -C out/Release dm
out/Release/dm --config pdf --src skp gm -w /tmp/dm
With skia_pdf_generate_pdfa=0, all PDFs generated from GMs and
SKPs are identical. With skia_pdf_generate_pdfa=1, all PDFs
generated from GMs and SKPs render identically in Pdfium.
BUG=skia:3110
Review URL: https://codereview.chromium.org/
1394263003
scroggo [Fri, 9 Oct 2015 19:15:57 +0000 (12:15 -0700)]
Fix codec memory leaks in nanobench
SubsetTranslateBench.cpp:
Unref the color table, so it gets deleted.
SkBitmapRegionDecoderInterface.cpp:
Delete the stream if it is not used.
BUG=skia:3418
Review URL: https://codereview.chromium.org/
1396113003
msarett [Fri, 9 Oct 2015 18:07:34 +0000 (11:07 -0700)]
Fill incomplete images in SkCodec parent class
Rather than implementing some sort of "fill" in every
SkCodec subclass for incomplete images, let's make the
parent class handle this situation.
This includes an API change to SkCodec.h
SkCodec::getScanlines() now returns the number of lines it
read successfully, rather than an SkCodec::Result enum.
getScanlines() most often fails on an incomplete input, in
which case it is useful to know how many lines were
successfully decoded - this provides more information than
kIncomplete vs kSuccess. We do lose information when the
API is used improperly, as we are no longer able to return
kInvalidParameter or kScanlineNotStarted.
Known Issues:
Does not work for incomplete fFrameIsSubset gifs.
Does not work for incomplete icos.
BUG=skia:
Review URL: https://codereview.chromium.org/
1332053002
egdaniel [Fri, 9 Oct 2015 17:55:16 +0000 (10:55 -0700)]
Fix gpu drawBitmap to work when BM is A8 and we have a shader
BUG=skia:
Review URL: https://codereview.chromium.org/
1399763002
fmalita [Fri, 9 Oct 2015 17:22:46 +0000 (10:22 -0700)]
Optional gradient dithering
Update raster gradient impls to observe the paint dithering flag.
For now this new behavior is disabled pending internal/external
client updates to mitigate diffs.
BUG=skia:4436
R=reed@google.com,mtklein@google.com,tomhudson@google.com
Review URL: https://codereview.chromium.org/
1391303002
scroggo [Fri, 9 Oct 2015 15:40:59 +0000 (08:40 -0700)]
Focus SkScaledCodec on BitmapRegionDecoder
The primary goal of SkScaledCodec is to replace the current
implementation of BitmapRegionDecoder, which depends on modified
versions of libjpeg and libpng, with an implementation that uses
standard versions of the libaries. Since BitmapRegionDecoder only
supports PNG, WEBP and JPEG, limit SkScaledCodec to those classes.
We will focus on those three until we complete this primary goal.
Then we can continue to make SkScaledCodec work for other formats.
Fix some bugs in SkScaledCodec::NewFromStream:
- Handle a NULL input stream properly
- Ensure that the input stream is deleted as expected on bad data
Add tests for these error cases.
BUG=skia:4428
Review URL: https://codereview.chromium.org/
1389053002
scroggo [Thu, 8 Oct 2015 21:44:51 +0000 (14:44 -0700)]
Reland of Pass --images '' to nanobench to disable image benchmarking. (patchset #1 id:1 of https://codereview.chromium.org/
1400633002/ )
Reason for revert:
Still breaking. e.g.
http://build.chromium.org/p/client.skia/builders/Test-iOS-Clang-iPad4-GPU-SGX554-Arm7-Debug/builds/1466/steps/nanobench/logs/stdio
http://build.chromium.org/p/client.skia.android/builders/Perf-Android-GCC-Nexus7-GPU-Tegra3-Arm7-Release/builds/2280/steps/nanobench/logs/stdio
Original issue's description:
> Revert of Pass --images '' to nanobench to disable image benchmarking. (patchset #1 id:1 of https://codereview.chromium.org/
1381703003/ )
>
> Reason for revert:
> I have attempted to fix the problems that caused running nanobench with images to fail, so testing to see if they are in fact fixed.
>
> Original issue's description:
> > Pass --images '' to nanobench to disable image benchmarking.
> >
> > Enabling image benchmarking has caused most nanobench runs to fail,
> > both Debug and Release.
> >
> > BUG=skia:3418
> > NOTREECHECKS=true
> > TBR=scroggo@google.com
> >
> > Committed: https://skia.googlesource.com/skia/+/
1f7039c6c5f0f118b629994b0bac255345e5abd6
>
> BUG=skia:3418
>
> Committed: https://skia.googlesource.com/skia/+/
c2e59caa0a466040f4b4f08cb5604c86b2615efa
TBR=mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:3418
Review URL: https://codereview.chromium.org/
1396843003
scroggo [Thu, 8 Oct 2015 21:09:41 +0000 (14:09 -0700)]
Revert of Pass --images '' to nanobench to disable image benchmarking. (patchset #1 id:1 of https://codereview.chromium.org/
1381703003/ )
Reason for revert:
I have attempted to fix the problems that caused running nanobench with images to fail, so testing to see if they are in fact fixed.
Original issue's description:
> Pass --images '' to nanobench to disable image benchmarking.
>
> Enabling image benchmarking has caused most nanobench runs to fail,
> both Debug and Release.
>
> BUG=skia:3418
> NOTREECHECKS=true
> TBR=scroggo@google.com
>
> Committed: https://skia.googlesource.com/skia/+/
1f7039c6c5f0f118b629994b0bac255345e5abd6
BUG=skia:3418
Review URL: https://codereview.chromium.org/
1400633002
scroggo [Thu, 8 Oct 2015 19:01:39 +0000 (12:01 -0700)]
Update Subset benches to support interlacing and fix bugs
Instead of decoding one line at a time, if the ScanlineOrder is kNone,
decode all of the lines in one pass, and then copy the subset into the
output. This will allow us to more realistically test subset decodes
for interlaced png. It also makes running them not take forever.
Do *not* support other modes (besides kTopDown), since they are not
used by the big three we need to replace BitmapRegionDecoder
implementation (skbug.com/4428).
Fix a bug in SubsetTranslateBench and SubsetZoomBench:
When we decode another subset, we need to reset the scanline decode
first. This bug appears to have been present since the introduction of
these tests in crrev.com/
1160953002
BUG=skia:4205
BUG=skia:3418
Review URL: https://codereview.chromium.org/
1387233002
reed [Thu, 8 Oct 2015 16:22:02 +0000 (09:22 -0700)]
remove internalDrawBitmap, as it is no longer shared with any other caller, and can be folded into onDrawBitmap
BUG=skia:
Review URL: https://codereview.chromium.org/
1391333004
cdalton [Thu, 8 Oct 2015 15:04:09 +0000 (08:04 -0700)]
Updates nvpr text blobs to not store a direct reference to the
per-glyph GPU path object, but rather store a key for looking it up in
the resource cache. This allows the cache to purge glyphs when needed.
Also indirectly fixes a memory leak that was introduced with nvpr text
blobs.
BUG=skia:
Review URL: https://codereview.chromium.org/
1374853004
mtklein [Wed, 7 Oct 2015 19:46:43 +0000 (12:46 -0700)]
SkPath::fFirstDirection: seq-cst -> relaxed
We landed this originally with lazily-correct sequentially-consistent memory
order. It turns out that's regressed performance, we think particularly when
recording paths. We also think there's no need for anything but relaxed memory
order here.
We should see this chart go down if all goes well: https://perf.skia.org/#4329
There are also Chrome performance charts to watch in the linked bug.
BUG=chromium:537700
CQ_EXTRA_TRYBOTS=client.skia:Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Release-TSAN-Trybot,Test-Ubuntu-GCC-Golo-GPU-GT610-x86_64-Release-TSAN
No public API changes.
TBR=reed@google.com
Review URL: https://codereview.chromium.org/
1393833003
mtklein [Wed, 7 Oct 2015 18:49:00 +0000 (11:49 -0700)]
Automatically add TSAN bots to the CQ for changes affecting SkAtomics.h
BUG=skia:
Review URL: https://codereview.chromium.org/
1396523002
jvanverth [Wed, 7 Oct 2015 17:43:05 +0000 (10:43 -0700)]
Add DrawShipSim sample.
Adds a version of the DrawShip sample that uses drawRect with a bitmap.
For testing batching.
BUG=skia:
Review URL: https://codereview.chromium.org/
1395533002
bsalomon [Wed, 7 Oct 2015 16:30:05 +0000 (09:30 -0700)]
Remove ClipMaskType from GrCMM
Review URL: https://codereview.chromium.org/
1391653002
bsalomon [Wed, 7 Oct 2015 16:20:28 +0000 (09:20 -0700)]
Insert clip fragment processor outside GrCMM
Review URL: https://codereview.chromium.org/
1393553002
reed [Tue, 6 Oct 2015 21:05:32 +0000 (14:05 -0700)]
Revert "Revert of factories should return baseclass, allowing the impl to specialize (patchset #4 id:60001 of https://codereview.chromium.org/
1390523005/ )"
This reverts commit
95376a0dde3cdf414eb97a20cef3af19ed7e0151.
BUG=skia:
TBR=
Review URL: https://codereview.chromium.org/
1389083002
schenney [Tue, 6 Oct 2015 19:59:55 +0000 (12:59 -0700)]
Revert of factories should return baseclass, allowing the impl to specialize (patchset #4 id:60001 of https://codereview.chromium.org/
1390523005/ )
Reason for revert:
Breaks Chrome with this link error: ../../third_party/skia/include/effects/SkMorphologyImageFilter.h:75: error: undefined reference to 'SkMorphologyImageFilter::SkMorphologyImageFilter(int, int, SkImageFilter*, SkImageFilter::CropRect const*)'
../../third_party/skia/include/effects/SkMorphologyImageFilter.h:104: error: undefined reference to 'SkMorphologyImageFilter::SkMorphologyImageFilter(int, int, SkImageFilter*, SkImageFilter::CropRect const*)'
Presumably due to code in third_party/WebKit/Source/platform/graphics/filters/FEMorphology.cpp that contains:
#include "SkMorphologyImageFilter.h"
...
if (m_type == FEMORPHOLOGY_OPERATOR_DILATE)
return adoptRef(SkDilateImageFilter::Create(radiusX, radiusY, input.get(), &rect));
return adoptRef(SkErodeImageFilter::Create(radiusX, radiusY, input.get(), &rect));
Original issue's description:
> factories should return baseclass, allowing the impl to specialize
>
> waiting on https://codereview.chromium.org/
1386163002/# to land
>
> BUG=skia:4424
>
> Committed: https://skia.googlesource.com/skia/+/
80a6dcaa1b757826ed7414f64b035d512d9ccbf8
TBR=senorblanco@google.com,robertphillips@google.com,reed@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:4424
Review URL: https://codereview.chromium.org/
1389063002
egdaniel [Tue, 6 Oct 2015 18:20:09 +0000 (11:20 -0700)]
Blacklisting imagefilterscropexpand on s4 since it often causes the bot to hang and crash.
BUG=skia:4079
Review URL: https://codereview.chromium.org/
1373763005
reed [Tue, 6 Oct 2015 18:12:23 +0000 (11:12 -0700)]
factories should return baseclass, allowing the impl to specialize
waiting on https://codereview.chromium.org/
1386163002/# to land
BUG=skia:4424
Review URL: https://codereview.chromium.org/
1390523005
scroggo [Tue, 6 Oct 2015 17:47:52 +0000 (10:47 -0700)]
Allow SkIcoCodec to fail due to conversion
Even if the client chooses dimensions that were suggested by
SkIcoCodec, it is possible for the color conversion to be incorrect,
so it can still reach the exit case of kInvalidConversion.
For example, when running nanobench, we attempt to decode the image
to Alpha_8, which the codec does not support. This allows running
nanobench in debug mode without asserting.
BUG=skia:3418
Review URL: https://codereview.chromium.org/
1389943002
halcanary [Tue, 6 Oct 2015 16:41:47 +0000 (09:41 -0700)]
SkPDF: when drawing stroked path, draw using SVG rules for zero-length segments
The "zeroPath" and emptystroke GMs capture this issue.
This CL changes the following PDF GMs: emptystroke dashing4
lineclosepath dashing3 zeroPath linepath
complexclip3_complex complexclip3_simple roundrects
degeneratesegments filltypes strokerect pathfill
inverse_paths desk_chalkboard.skp
After this change, all PDF GMs look better (closer to 8888).
The dashing4, emptystroke, and zeroPath GMs still need a lot
of work to make them look right.
BUG=538726
Review URL: https://codereview.chromium.org/
1374383004
bsalomon [Tue, 6 Oct 2015 15:40:50 +0000 (08:40 -0700)]
Bye bye processor data manager.
Review URL: https://codereview.chromium.org/
1388113002
joshualitt [Tue, 6 Oct 2015 15:25:15 +0000 (08:25 -0700)]
Fix for visualbench loops forever
TBR=robertphillips@google.com
BUG=skia:
Review URL: https://codereview.chromium.org/
1387823004
joshualitt [Tue, 6 Oct 2015 15:05:10 +0000 (08:05 -0700)]
nvpr removal from VisualBench base context
TBR=bsalomon@google.com
BUG=skia:
Review URL: https://codereview.chromium.org/
1391633002
msarett [Tue, 6 Oct 2015 14:46:02 +0000 (07:46 -0700)]
Fix SkGifCodec to handle gifs where frameSize != imageSize
These are quite rare, causing us to miss a few bugs in how
we deal with these images.
Additionally, there is a behavior change. If the imageSize
is not large enough to contain the frame, we will "fix" the
image by increasing the image size.
SkScaledCodec is still buggy with regard to these gifs.
See skbug.com/4421. We will fix that after
1332053002
lands.
BUG=skia:
Review URL: https://codereview.chromium.org/
1386973002
herb [Tue, 6 Oct 2015 14:42:53 +0000 (07:42 -0700)]
Remove duplicate .cpp which should have been a .h
Review URL: https://codereview.chromium.org/
1389923002
robertphillips [Tue, 6 Oct 2015 14:38:23 +0000 (07:38 -0700)]
Dynamically allocate the GrDrawContexts
This CL moves the allocation and storage of the GrTextContexts into the DrawingManager. The GrDrawContexts now just get their GrTextContext from the DrawingManager.
Review URL: https://codereview.chromium.org/
1375153007
fmalita [Tue, 6 Oct 2015 14:24:03 +0000 (07:24 -0700)]
[SkDebugger] Handle path empty-bounds gracefully
Paths with empty bounds crash the debugger: mapping their bounds onto
the shapshot area gets the canvas into an unhappy state.
R=robertphillips@google.com
Review URL: https://codereview.chromium.org/
1387173003
mtklein [Tue, 6 Oct 2015 14:00:45 +0000 (07:00 -0700)]
Revert of Stop using SkScalerContext::getAdvance() in SkGlyphCache. (patchset #2 id:20001 of https://codereview.chromium.org/
1321243004/ )
Reason for revert:
Suspect this is the cause of regressions in crbug.com/527445. It's triggering on Windows and Linux, where getting advances does less than getting full metrics. It's not triggering on Mac, where this CL was a no-op.
Original issue's description:
> Stop using SkScalerContext::getAdvance() in SkGlyphCache.
>
> We think it'll simplify things to just always get the full metrics.
> On most platforms, it's no different, and we think the platforms that
> do differ (FreeType) will be nearly just as cheap.
>
> Removing this distinction helps us make SkGlyphCaches concurrent by
> removing a state (we-have-only-advances) from its logical state machine.
>
> We see no significant changes running SKPs before and after this CL.
> That makes sense, of course, because the SKPs bake some of this into drawPosText.
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
518a2923f11b819fa44ed5cff54155326959540f
TBR=reed@google.com,bungeman@google.com,herb@google.com,mtklein@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1383403003
cdalton [Mon, 5 Oct 2015 22:56:34 +0000 (15:56 -0700)]
Make path range loading explicit
Requires the caller to explicitly preload paths within a range before
calling drawPaths. This allows us to remove the implicit lazy load,
thereby eliminating a redundant check on every redraw of a text blob.
BUG=skia:
Review URL: https://codereview.chromium.org/
1382013002
cdalton [Mon, 5 Oct 2015 22:37:35 +0000 (15:37 -0700)]
Implement cached nvpr text blobs
BUG=skia:
Review URL: https://codereview.chromium.org/
1381073002
cdalton [Mon, 5 Oct 2015 22:28:20 +0000 (15:28 -0700)]
Use SkTextBlob for nvpr color bitmap fallbacks
BUG=skia:
Review URL: https://codereview.chromium.org/
1381873003
cdalton [Mon, 5 Oct 2015 21:57:20 +0000 (14:57 -0700)]
Add TextRun object to nvpr text
Adds a TextRun object that defines a resolution-independent, paint-
indepent draw for a string of nvpr text.
BUG=skia:
Review URL: https://codereview.chromium.org/
1375353004
msarett [Mon, 5 Oct 2015 21:20:27 +0000 (14:20 -0700)]
SkScaledCodec should implement onRewind()
This is a bug fix. I'm also adding a test.
BUG=skia:
Review URL: https://codereview.chromium.org/
1385703002
joshualitt [Mon, 5 Oct 2015 20:58:26 +0000 (13:58 -0700)]
Cleanup timing state machine
BUG=skia:
Review URL: https://codereview.chromium.org/
1386933002
herb [Mon, 5 Oct 2015 20:28:59 +0000 (13:28 -0700)]
Add more .h files to the .gypi
BUG=skia:
Review URL: https://codereview.chromium.org/
1386963002
joshualitt [Mon, 5 Oct 2015 20:24:55 +0000 (13:24 -0700)]
move visual bench interactive module to timing state machine
BUG=skia:
Review URL: https://codereview.chromium.org/
1382883003
fmalita [Mon, 5 Oct 2015 18:09:57 +0000 (11:09 -0700)]
Validate text blob runs after SkTextBlob construction.
This avoids tripping CFI when casting uninitialized SkTextBlob ptrs.
BUG=chromium:538754
R=bungeman@google.com,reed@google.com,krasin@google.com
Review URL: https://codereview.chromium.org/
1388543005
jvanverth [Mon, 5 Oct 2015 18:05:07 +0000 (11:05 -0700)]
Add ship sample
BUG=skia:
Review URL: https://codereview.chromium.org/
1359033005
scroggo [Mon, 5 Oct 2015 18:03:34 +0000 (11:03 -0700)]
Specify bench_type & source_type for image benches
This will allow us to use perf filtering for comparing SkImageDecoder
to SkCodec.
BUG=skia:3418
Review URL: https://codereview.chromium.org/
1387863002
mtklein [Mon, 5 Oct 2015 17:24:59 +0000 (10:24 -0700)]
Pass --images '' to nanobench to disable image benchmarking.
Enabling image benchmarking has caused most nanobench runs to fail,
both Debug and Release.
BUG=skia:3418
NOTREECHECKS=true
TBR=scroggo@google.com
Review URL: https://codereview.chromium.org/
1381703003
egdaniel [Mon, 5 Oct 2015 15:11:49 +0000 (08:11 -0700)]
Fix gpu dashing for case where all intervals are 0.
BUG=skia:4409
Review URL: https://codereview.chromium.org/
1381803005
mtklein [Mon, 5 Oct 2015 14:29:22 +0000 (07:29 -0700)]
Force cache SkTestScalerContext path bounds.
Looks like a race on this:
http://build.chromium.org/p/client.skia/builders/Test-Ubuntu-GCC-GCE-CPU-AVX2-x86_64-Debug-ASAN/builds/2608/steps/dm/logs/stdio
BUG=skia:
Review URL: https://codereview.chromium.org/
1373363007
joshualitt [Mon, 5 Oct 2015 14:23:30 +0000 (07:23 -0700)]
Factor out VisualBench timing code into a helper class
BUG=skia:
Review URL: https://codereview.chromium.org/
1375363003
robertphillips [Sun, 4 Oct 2015 19:21:33 +0000 (12:21 -0700)]
Revert of Have GrRectBlurEffect use Linear filtering (rather than Nearest Neighbor) (patchset #2 id:20001 of https://codereview.chromium.org/
1378023004/ )
Reason for revert:
Trial run confirmed layout test expectations - will reland after suppressions
Original issue's description:
> Have GrRectBlurEffect use Linear filtering (rather than Nearest Neighbor)
>
> Committed: https://skia.googlesource.com/skia/+/
2d70bcccc9ba8f3898a7ae506ba7410ed8e9c9db
>
> Committed: https://skia.googlesource.com/skia/+/
5175b9563055926b4969d57f06e4ae49b86e055b
TBR=bsalomon@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review URL: https://codereview.chromium.org/
1378813004
reed [Sun, 4 Oct 2015 18:28:36 +0000 (11:28 -0700)]
SkImage doesn't use props, so don't need to store it
BUG=skia:
TBR=bsalomon
Review URL: https://codereview.chromium.org/
1372153006
robertphillips [Sun, 4 Oct 2015 15:40:31 +0000 (08:40 -0700)]
Fix memory leak in ClipMaskManager
TBR=bsalomon@google.com
Review URL: https://codereview.chromium.org/
1385823002
robertphillips [Sun, 4 Oct 2015 15:36:51 +0000 (08:36 -0700)]
Have GrRectBlurEffect use Linear filtering (rather than Nearest Neighbor)
Committed: https://skia.googlesource.com/skia/+/
2d70bcccc9ba8f3898a7ae506ba7410ed8e9c9db
Review URL: https://codereview.chromium.org/
1378023004
update-skps [Sun, 4 Oct 2015 07:22:10 +0000 (00:22 -0700)]
Update SKP version
Automatic commit by the RecreateSKPs bot.
TBR=
NO_MERGE_BUILDS
Review URL: https://codereview.chromium.org/
1390433002
cdalton [Sat, 3 Oct 2015 20:28:35 +0000 (13:28 -0700)]
Simplify nvpr text
- Drops device-space glyphs in favor of perf/simplicity.
- Removes residual complexities that would flip glyphs vertically for
compatibility with nvpr glyph loading, which is no longer used.
- Drops hairline support since they required new paths for every draw
matrix, and could only be supported under certain circumstances.
- Quits checking for color bitmap fonts in canDrawText since the
normal color emoji fallback will handle them anyway.
BUG=skia:
Review URL: https://codereview.chromium.org/
1380973002
msarett [Fri, 2 Oct 2015 20:44:12 +0000 (13:44 -0700)]
Stop calling jpeg_finish_decompress()
jpeg_finish_decompress() does several things:
(1) Reads to the end of the image stream.
(2) Calls term_src(), a client provided function that
indicates we are done with the memory stream. Our current
implementation of term_src() does nothing.
(3) Calls jpeg_abort() which aborts the decode and cleans
up some memory.
I don't think we need to call this anymore.
(1) seems irrelevant.
It seems a little dangerous to get rid of (2), but it is
fine while our implementation of term_src() does nothing.
(3) We will call jpeg_destroy() on destruction of the
JpegDecoderManager. This should free all the memory,
making it unnecessary to call jpeg_abort() beforehand.
BUG=skia:4040
Review URL: https://codereview.chromium.org/
1370323004
scroggo [Fri, 2 Oct 2015 20:14:46 +0000 (13:14 -0700)]
Move all knowledge of X sampling into SkScaledCodec
Prior to this CL, each SkCodec subclass that allows sampling does an
extra check in onStartScanlineDecode to determine whether the X dimension
is supported for sampling. Remove this check, and provide a way for
SkScaledCodec to directly access the SkSwizzler, and update it to do
sampling. This way, the SkCodec knows nothing of sampling, but we can
still save the extra step of sampling afterwards.
FIXME: SkBmpRLECodec still calls SkScaledCodec::DimensionsSupported. It
seems like it could directly support sampling, rather than dealing with
SkScaledCodec (partially).
Add a new base class for SkSwizzler. It allows updating the swizzler
after it was created, which is done by SkScaledCodec. Modify SkSwizzler's
constructor/factory function to stop taking any info about sampling,
assume the sample size is one, and move modifying that into a virtual
function overridden from the base class.
BUG=skia:4284
Review URL: https://codereview.chromium.org/
1372973002
halcanary [Fri, 2 Oct 2015 19:15:55 +0000 (12:15 -0700)]
SkPDF: remove suppression
Review URL: https://codereview.chromium.org/
1379183005
joshualitt [Fri, 2 Oct 2015 18:27:14 +0000 (11:27 -0700)]
Incrementally flush GrDrawTarget
BUG=skia:
Committed: https://skia.googlesource.com/skia/+/
a7e878064509ef96b54d5507dab0b50def66dc13
Review URL: https://codereview.chromium.org/
1386463004
jvanverth [Fri, 2 Oct 2015 17:42:36 +0000 (10:42 -0700)]
Fix CrOS SampleApp
Add compile-time check for GPU build when using GPU flag
TBR=robertphillips@google.com
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review URL: https://codereview.chromium.org/
1376423004
joshualitt [Fri, 2 Oct 2015 17:16:03 +0000 (10:16 -0700)]
Revert of Incrementally flush GrDrawTarget (patchset #8 id:130001 of https://codereview.chromium.org/
1386463004/ )
Reason for revert:
breaking bots
Original issue's description:
> Incrementally flush GrDrawTarget
>
> BUG=skia:
>
> Committed: https://skia.googlesource.com/skia/+/
a7e878064509ef96b54d5507dab0b50def66dc13
TBR=robertphillips@google.com,bsalomon@google.com,joshualitt@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1386683002