bsalomon [Wed, 9 Mar 2016 17:55:55 +0000 (09:55 -0800)]
Reenable dev bounds clip reduction
TBR=joshualitt@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1781663002
Review URL: https://codereview.chromium.org/
1781663002
bsalomon [Wed, 9 Mar 2016 17:52:45 +0000 (09:52 -0800)]
Fix nullptr access in GrTextureProvider::createMipMappedTexture
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1777013002
Review URL: https://codereview.chromium.org/
1777013002
reed [Wed, 9 Mar 2016 17:50:50 +0000 (09:50 -0800)]
partial switch over to sp usage of shaders
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1776973003
Review URL: https://codereview.chromium.org/
1776973003
robertphillips [Wed, 9 Mar 2016 16:37:23 +0000 (08:37 -0800)]
Revert of don't create zero length intervals (patchset #1 id:1 of https://codereview.chromium.org/
1766243004/ )
Reason for revert:
This may, or may not, be blocking the DEPS roll with:
svg/W3C-SVG-1.1/painting-stroke-04-t.svg
- LayoutSVGPath {path} at (50,127) size 380x26 [stroke={[type=SOLID] [color=#000000] [stroke width=25.00] [dash offset=10.00] [dash array={10.00, 10.00}]}] [data="M 50 140 L 430 140"]
+ LayoutSVGPath {path} at (60,127) size 370x26 [stroke={[type=SOLID] [color=#000000] [stroke width=25.00] [dash offset=10.00] [dash array={10.00, 10.00}]}] [data="M 50 140 L 430 140"]
Original issue's description:
> don't create zero length intervals
>
> Dashing a pattern without zero-length intervals should
> not create them if the end of the on interval coincides
> with the beginning of the initial dash offset.
>
> R=reed@google.com
> BUG=591993
> GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1766243004
>
> Committed: https://skia.googlesource.com/skia/+/
18bbd00190623fb6cdb119df4a118ac3c1aed52a
TBR=reed@google.com,fmalita@chromium.org,caryclark@google.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=591993
Review URL: https://codereview.chromium.org/
1779803002
halcanary [Wed, 9 Mar 2016 15:52:09 +0000 (07:52 -0800)]
SkPDF: use sk_ref_sp
s/sk_sp<[^>]*>(SkRef(\([^)]*\)))/sk_ref_sp(\1)/
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1777923002
Review URL: https://codereview.chromium.org/
1777923002
bungeman [Wed, 9 Mar 2016 15:42:54 +0000 (07:42 -0800)]
Add SkDrawPosTextHCommand JSON, fix skiaserve build.
Adds SkDrawPosTextHCommand ::fromJSON and ::toJSON.
Both SkDrawPosTextCommand and SkDrawPosTextHCommand's
::toJSON write the correct number of positions, preventing reading
uninitialized memory.
The microhttpd build is now done in the build tree as opposed to in a
temporary directory.
The microhttpd build script uses os.path.join so that absolute paths
do not confuse the build. This allows compatibility with the cmake
gyp generator as CMake likes to pass absolute paths.
The microhttpd gyp target is now marked as 'none' since it is not
a 'static_library' target (which directs gyp to compile sources into
a static library). The dependencies to the action are updated to the
minimum required for sane re-building.
The everything gyp target now depends on the skiaserve gyp target.
This means that when using skia_build_server=1, building 'most' will
build skiaserve, but when skia_build_server is not defined the
skiaserve target will still be available if specified manually.
The old json.gyp is removed as it currently does not build anything.
All of the files currently referenced by it as sources no longer exist.
Review URL: https://codereview.chromium.org/
1775203002
msarett [Wed, 9 Mar 2016 15:14:41 +0000 (07:14 -0800)]
Deprecate SkImageDecoder
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1776153003
Review URL: https://codereview.chromium.org/
1776153003
bsalomon [Wed, 9 Mar 2016 14:25:15 +0000 (06:25 -0800)]
Don't allow nullptr in texels array params (unless using a transfer buffer).
Require all levels in writePixels to have a non-nullptr.
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1765633002
Committed: https://skia.googlesource.com/skia/+/
8ee78f31b2a29a5f76403755ea17bad9be74a3ec
Review URL: https://codereview.chromium.org/
1765633002
msarett [Wed, 9 Mar 2016 14:16:55 +0000 (06:16 -0800)]
Remove usage of SkImageDecoder in debugger
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1774313003
Review URL: https://codereview.chromium.org/
1774313003
caryclark [Wed, 9 Mar 2016 13:55:53 +0000 (05:55 -0800)]
don't create zero length intervals
Dashing a pattern without zero-length intervals should
not create them if the end of the on interval coincides
with the beginning of the initial dash offset.
R=reed@google.com
BUG=591993
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1766243004
Review URL: https://codereview.chromium.org/
1766243004
robertphillips [Wed, 9 Mar 2016 13:17:10 +0000 (05:17 -0800)]
Switch SkImageSource image filter over to new onFilterImage interface
This CL relies on: https://codereview.chromium.org/
1762013002/ (Swap over to using SkImageFilter::filterImage instead of filterImageDeprecated)
TBR=bsalomon@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1772933002
Review URL: https://codereview.chromium.org/
1772933002
reed [Wed, 9 Mar 2016 02:50:00 +0000 (18:50 -0800)]
use Make instead of Create to return a shared shader
Partially updated call sites. Undefine the flag in SkSHader.h to convert the remaining sites.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1772463002
Review URL: https://codereview.chromium.org/
1772463002
kulshin [Wed, 9 Mar 2016 01:30:05 +0000 (17:30 -0800)]
Initialize the font fallback when the direct write font
manager is created.
This ensures that the fallback is initialized prior to
sandbox startup in Chromium.
This enables the use of font fallback when combined with
Chromium change https://codereview.chromium.org/
1740593002/
Review URL: https://codereview.chromium.org/
1740533003
halcanary [Tue, 8 Mar 2016 23:10:16 +0000 (15:10 -0800)]
SkPDF: Add sk_sp setters; .release() becomes std::move()
Note to reviewers: Start with changes to SkPDFTypes.h
Many places that had a bare pointer owning a reference are refactored to
use a sk_sp.
There remain several places where a non-owning pointer `T*` should be
replaced with `const sk_sp<T>&` to eliminate the common pattern
`sk_sp<T>(SkRef(x))`.
Committed: https://skia.googlesource.com/skia/+/
9904c9212074279380e21f96575078734dbbd308
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1775043002
Review URL: https://codereview.chromium.org/
1775043002
mtklein [Tue, 8 Mar 2016 21:25:42 +0000 (13:25 -0800)]
Revert of nanobench: turn on f16 and sRGB on the bots (patchset #1 id:1 of https://codereview.chromium.org/
1769163006/ )
Reason for revert:
Debug bots failing at
Running writepix_RGBA_UPM f16
../../../src/core/SkBitmap.cpp:558: fatal error: ""false && \"Can't return addr for config\"""
Original issue's description:
> nanobench: turn on f16 and sRGB on the bots
>
> BUG=skia:
> GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1769163006
>
> Committed: https://skia.googlesource.com/skia/+/
6b5c5cc7b7bdffb4882706b915b544aba8714961
TBR=reed@google.com,mtklein@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1780463003
joshualitt [Tue, 8 Mar 2016 21:24:33 +0000 (13:24 -0800)]
Fix skiaserve in debug mode
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1774263002
Review URL: https://codereview.chromium.org/
1774263002
halcanary [Tue, 8 Mar 2016 21:03:55 +0000 (13:03 -0800)]
Revert of SkPDF: Add sk_sp setters; .release() becomes std::move() (patchset #2 id:20001 of https://codereview.chromium.org/
1775043002/ )
Reason for revert:
https://build.chromium.org/p/client.skia/builders/Linux%20Builder/builds/6405/steps/compile/logs/stdio
Original issue's description:
> SkPDF: Add sk_sp setters; .release() becomes std::move()
>
> Note to reviewers: Start with changes to SkPDFTypes.h
>
> Many places that had a bare pointer owning a reference are refactored to
> use a sk_sp.
>
> There remain several places where a non-owning pointer `T*` should be
> replaced with `const sk_sp<T>&` to eliminate the common pattern
> `sk_sp<T>(SkRef(x))`.
>
> Committed: https://skia.googlesource.com/skia/+/
9904c9212074279380e21f96575078734dbbd308
TBR=bungeman@google.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review URL: https://codereview.chromium.org/
1775143002
reed [Tue, 8 Mar 2016 20:54:48 +0000 (12:54 -0800)]
Add sk_ref_sp helper function.
Review URL: https://codereview.chromium.org/
1773943004
halcanary [Tue, 8 Mar 2016 20:38:22 +0000 (12:38 -0800)]
SkPDF: Add sk_sp setters; .release() becomes std::move()
Note to reviewers: Start with changes to SkPDFTypes.h
Many places that had a bare pointer owning a reference are refactored to
use a sk_sp.
There remain several places where a non-owning pointer `T*` should be
replaced with `const sk_sp<T>&` to eliminate the common pattern
`sk_sp<T>(SkRef(x))`.
Review URL: https://codereview.chromium.org/
1775043002
mtklein [Tue, 8 Mar 2016 20:34:11 +0000 (12:34 -0800)]
nanobench: turn on f16 and sRGB on the bots
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1769163006
Review URL: https://codereview.chromium.org/
1769163006
egdaniel [Tue, 8 Mar 2016 20:19:54 +0000 (12:19 -0800)]
Fill out more information in GrVkCaps
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1768083003
Review URL: https://codereview.chromium.org/
1768083003
brianosman [Tue, 8 Mar 2016 20:15:09 +0000 (12:15 -0800)]
Remove SkImageGeneratorUtils. None of this code is used.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1778613002
Review URL: https://codereview.chromium.org/
1778613002
jvanverth [Tue, 8 Mar 2016 20:09:27 +0000 (12:09 -0800)]
Buffer fix for VkProgram caching
When caching VkPrograms, we can end up with situations where we
overwrite uniform buffers while they're in use (or being prepared for
use) in the command buffer. This fix will address that. This also
addresses the rare but similar case of overwriting vertex or index
buffers when they're in flight.
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1774963003
Review URL: https://codereview.chromium.org/
1774963003
robertphillips [Tue, 8 Mar 2016 20:05:37 +0000 (12:05 -0800)]
Switch SkComposeImageFilter over to new onFilterImage interface
This CL depends on https://codereview.chromium.org/
1762013002/ (Swap over to using SkImageFilter::filterImage instead of filterImageDeprecated)
TBR=bsalomon@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1768283002
Review URL: https://codereview.chromium.org/
1768283002
bsalomon [Tue, 8 Mar 2016 19:53:31 +0000 (11:53 -0800)]
Expand coverage of vertices gm
Review URL: https://codereview.chromium.org/
1361173003
yujieqin [Tue, 8 Mar 2016 19:47:31 +0000 (11:47 -0800)]
Update DNG SDK version.
* Integrate integer overflow changes from security branch
BUG=b/
26535130
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1748363004
Committed: https://skia.googlesource.com/skia/+/
be2924ad17d47f4ed49666894d87571a28f89567
Committed: https://skia.googlesource.com/skia/+/
62c0f75159674087079140e09ef1396316577754
Review URL: https://codereview.chromium.org/
1748363004
mtklein [Tue, 8 Mar 2016 19:31:11 +0000 (11:31 -0800)]
add f16 and srgb --configs to nanobench
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1775093002
Review URL: https://codereview.chromium.org/
1775093002
aleksandar.stojiljkovic [Tue, 8 Mar 2016 19:18:21 +0000 (11:18 -0800)]
4377: drawVertices crashes when shader+colorfilter in the paint
Fix memory corruption that is result of static_casting SkFilterShaderContext*
to SkTriColorShaderContext* and then changing it's state using setup()
SkTriColorShader and it's Context could be wrapped by Compose, 3DShader and
SkFilterShader, but the link SkTriColorShaderContext -> SkTriColorShader
(fShader) always remain - so this way, data about triangle is propagated.
BUG=4377
Didn't add new unit test - it would be redundant since the GM vertices update here: https://codereview.chromium.org/
1361173003/ is covering the bug. With this patch it passes.
Review URL: https://codereview.chromium.org/
1431713003
joshualitt [Tue, 8 Mar 2016 18:56:41 +0000 (10:56 -0800)]
Create a small python driver for skiaserve
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1754213005
Review URL: https://codereview.chromium.org/
1754213005
robertphillips [Tue, 8 Mar 2016 18:45:32 +0000 (10:45 -0800)]
Rename SkSpecialImage::getBitmap to getBitmapDeprecated
Review URL: https://codereview.chromium.org/
1780433002
joshualitt [Tue, 8 Mar 2016 18:43:41 +0000 (10:43 -0800)]
Fix up GrAuditTrail to allow arbitrary reordering
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1765123002
Review URL: https://codereview.chromium.org/
1765123002
fmalita [Tue, 8 Mar 2016 18:20:49 +0000 (10:20 -0800)]
Fix 4f gradient swizzle post crrev.com/
1774523002
* store interval colors in pm4f-natural/RGBA order
* swizzle to dst order during interval advance
Also remove an unused Interval ctor.
R=reed@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1770153002
Review URL: https://codereview.chromium.org/
1770153002
reed [Tue, 8 Mar 2016 18:09:18 +0000 (10:09 -0800)]
make pm4f be RGBA always, not pmcolor order
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1774523002
Review URL: https://codereview.chromium.org/
1774523002
joshualitt [Tue, 8 Mar 2016 17:31:15 +0000 (09:31 -0800)]
Lazy init batch unique ID
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1772023003
Review URL: https://codereview.chromium.org/
1772023003
senorblanco [Tue, 8 Mar 2016 17:06:44 +0000 (09:06 -0800)]
Small GrTessellator refactor and cleanup.
Implement a VertexList, and use it.
Rename insert<> -> list_insert<>.
Rename remove<> -> list_remove<>.
Remove some spurious returns.
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1771373002
Review URL: https://codereview.chromium.org/
1771373002
reed [Tue, 8 Mar 2016 17:02:24 +0000 (09:02 -0800)]
work around leftshift for negative values
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1774963002
Review URL: https://codereview.chromium.org/
1774963002
mtklein [Tue, 8 Mar 2016 17:01:39 +0000 (09:01 -0800)]
DM: verbose and quiet logging
If you pass no flags, output is pretty much unchanged from today: calls to
info() go to the console, calls to vlog() are ignored.
If you pass --verbose and -w dir, all calls to info(), vlog() and all of stderr
are redirected to dir/verbose.log. info() calls also go to the console,
unless you pass --quiet, in which case nothing goes to the console at all.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1771143002
Review URL: https://codereview.chromium.org/
1771143002
robertphillips [Tue, 8 Mar 2016 16:47:23 +0000 (08:47 -0800)]
Switch SkOffsetImageFilter over to new onFilterImage interface
This CL relies on: https://codereview.chromium.org/
1762013002/ (Swap over to using SkImageFilter::filterImage instead of filterImageDeprecated)
TBR=bsalomon@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1766743003
Review URL: https://codereview.chromium.org/
1766743003
robertphillips [Tue, 8 Mar 2016 16:37:18 +0000 (08:37 -0800)]
Fix caching bug in new SkImageFilter::filterImage path
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1777573002
Review URL: https://codereview.chromium.org/
1777573002
bungeman [Tue, 8 Mar 2016 16:35:23 +0000 (08:35 -0800)]
Add element_type, swap, operators, fix reset on sk_sp.
The 'element_type' typedef is to play nice with std::pointer_traits.
The full complement of operators and swap to match unique_ptr so that
sk_sp can be properly compared to nullptr and used with standard
containers.
Update to 'reset' so that calling 'unref' is the last operation.
This also adds tests for these changes, and sets the fPtr to nullptr
in debug for easier bug finding.
Review URL: https://codereview.chromium.org/
1773453002
halcanary [Tue, 8 Mar 2016 16:32:12 +0000 (08:32 -0800)]
SkPDF: use sk_make_sp<T> when it makes sense.
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1773033002
Review URL: https://codereview.chromium.org/
1773033002
bsalomon [Tue, 8 Mar 2016 16:30:14 +0000 (08:30 -0800)]
Circle GP/batch housecleaning
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1764923003
Review URL: https://codereview.chromium.org/
1764923003
borenet [Tue, 8 Mar 2016 13:33:03 +0000 (05:33 -0800)]
Swarming bots: fix trybots
Requires a recipe-side change as well.
NOTRY=true
BUG=skia:4763
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1768333002
Review URL: https://codereview.chromium.org/
1768333002
benjaminwagner [Tue, 8 Mar 2016 01:58:03 +0000 (17:58 -0800)]
Hide new methods from non-Android.
Review URL: https://codereview.chromium.org/
1769933004
msarett [Tue, 8 Mar 2016 01:25:12 +0000 (17:25 -0800)]
Disable color space test for old versions of libpng
Should fix the Google3 build.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1769953002
Review URL: https://codereview.chromium.org/
1769953002
mtklein [Tue, 8 Mar 2016 01:20:26 +0000 (17:20 -0800)]
Archive SkRemote-related code. I want to start fresh.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1774013002
Review URL: https://codereview.chromium.org/
1774013002
cdalton [Mon, 7 Mar 2016 23:39:09 +0000 (15:39 -0800)]
Add support for 2x2 matrices
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1755483002
Review URL: https://codereview.chromium.org/
1755483002
halcanary [Mon, 7 Mar 2016 22:57:50 +0000 (14:57 -0800)]
SkPDF: s/SkAutoTUnref/sk_sp/g
Review URL: https://codereview.chromium.org/
1771073002
cdalton [Mon, 7 Mar 2016 21:58:26 +0000 (13:58 -0800)]
Add "sample locations" feature to GrProcessor
Adds a "sample locations" feature to GrProcessor. When enabled, this
allows a processor to know inside the shader where all the samples are
located. Also adds various infastructure to query, cache, and identify
multisample data.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1717393002
Review URL: https://codereview.chromium.org/
1717393002
mtklein [Mon, 7 Mar 2016 21:20:52 +0000 (13:20 -0800)]
dm: log log
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1771993002
Review URL: https://codereview.chromium.org/
1771993002
robertphillips [Mon, 7 Mar 2016 20:45:14 +0000 (12:45 -0800)]
Swap over to using SkImageFilter::filterImage instead of filterImageDeprecated
This CL relies on https://codereview.chromium.org/
1757983002/ (Add SkSpecialImage-based methods to SkImageFilter)
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1762013002
TBR=bsalomon@google.com
Review URL: https://codereview.chromium.org/
1762013002
benjaminwagner [Mon, 7 Mar 2016 20:41:49 +0000 (12:41 -0800)]
Add advance[XY]Fixed getters for Android.
This prepares for changing the type of these fields to float.
BUG=skia:4632
Review URL: https://codereview.chromium.org/
1772863002
halcanary [Mon, 7 Mar 2016 20:39:14 +0000 (12:39 -0800)]
SkPDF Create working move constructor for inner classes
BUG=chromium:592330,chromium:592702
Review URL: https://codereview.chromium.org/
1774633002
djsollen [Mon, 7 Mar 2016 20:30:47 +0000 (12:30 -0800)]
Update android_make to support both --gcc and --clang options.
This CL also fixes all the errors that resulted from compiling
with clang and ccache.
BUG=skia:4948
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1774503004
Review URL: https://codereview.chromium.org/
1774503004
mtklein [Mon, 7 Mar 2016 20:28:17 +0000 (12:28 -0800)]
add virtual ~Effect()
Today we make an sk_sp<Effect> from an sk_sp<EffectImpl>.
But when the sk_sp<Effect> dies, it calls ~Effect(), not ~EffectImpl().
Making ~Effect() virtual fixes this.
This should make our Google3 tests sized-delete clean, unblocking those folks.
Review URL: https://codereview.chromium.org/
1769093002
bsalomon [Mon, 7 Mar 2016 19:50:44 +0000 (11:50 -0800)]
Attempt to combine batches in forward direction before flush
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1763883005
Review URL: https://codereview.chromium.org/
1763883005
mtklein [Mon, 7 Mar 2016 19:05:15 +0000 (11:05 -0800)]
link with -rdynamic on Linux
This lets our tools get symbols when they backtrace themselves.
Here's an example improved stack trace.
Notice there are symbols for some Skia methods now.
Stack trace:
out/Release/dm() [0x5ec19b]
out/Release/dm() [0x5ec209]
/lib/x86_64-linux-gnu/libc.so.6(+0x36d40) [0x7f6ae0ce2d40]
/lib/x86_64-linux-gnu/libc.so.6(gsignal+0x39) [0x7f6ae0ce2cc9]
/lib/x86_64-linux-gnu/libc.so.6(abort+0x148) [0x7f6ae0ce60d8]
out/Release/dm() [0xcd19f9]
out/Release/dm() [0x8ca959]
out/Release/dm(_ZN11XfermodesGM16onOnceBeforeDrawEv+0x70) [0x8cb640]
out/Release/dm(_ZN6skiagm2GM4drawEP8SkCanvas+0x1c) [0x5fc81c]
out/Release/dm(_ZNK2DM5GMSrc4drawEP8SkCanvas+0x73) [0x5f3563]
out/Release/dm(_ZNK2DM10RasterSink4drawERKNS_3SrcEP8SkBitmapP9SkWStreamP8SkString+0xcc) [0x5f9eac]
out/Release/dm(_ZN4Task3RunERKS_+0xd5) [0x5f1685]
out/Release/dm() [0xa4d320]
out/Release/dm() [0xcb66ba]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x8182) [0x7f6ae2667182]
/lib/x86_64-linux-gnu/libc.so.6(clone+0x6d) [0x7f6ae0da647d]
Today the same crash looks more like this.
Stack trace:
out/Release/dm() [0x5ec19b]
out/Release/dm() [0x5ec209]
/lib/x86_64-linux-gnu/libc.so.6(+0x36d40) [0x7f6ae0ce2d40]
/lib/x86_64-linux-gnu/libc.so.6(gsignal+0x39) [0x7f6ae0ce2cc9]
/lib/x86_64-linux-gnu/libc.so.6(abort+0x148) [0x7f6ae0ce60d8]
out/Release/dm() [0xcd19f9]
out/Release/dm() [0x8ca959]
out/Release/dm() [0x8cb640]
out/Release/dm() [0x5fc81c]
out/Release/dm() [0x5f3563]
out/Release/dm() [0x5f9eac]
out/Release/dm() [0x5f1685]
out/Release/dm() [0xa4d320]
out/Release/dm() [0xcb66ba]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x8182) [0x7f6ae2667182]
/lib/x86_64-linux-gnu/libc.so.6(clone+0x6d) [0x7f6ae0da647d]
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1774593002
Review URL: https://codereview.chromium.org/
1774593002
borenet [Mon, 7 Mar 2016 18:57:26 +0000 (10:57 -0800)]
Swarming bots: add build products whitelist
Should reduce output isolate size (and therefore upload time) for
compile task by ~10x
NOTRY=true
BUG=skia:4763
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1768273002
Review URL: https://codereview.chromium.org/
1768273002
herb [Mon, 7 Mar 2016 17:17:47 +0000 (09:17 -0800)]
Rework files and add test cases.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1757193002
Review URL: https://codereview.chromium.org/
1757193002
msarett [Mon, 7 Mar 2016 17:16:52 +0000 (09:16 -0800)]
Revert of Update Skia's YUV API (patchset #5 id:160001 of https://codereview.chromium.org/
1716523002/ )
Reason for revert:
Blimp failures with manual roll in Chrome.
Original issue's description:
> Update Skia's YUV API
>
> We should match the recently designed API in SkCodec.
> https://codereview.chromium.org/
1549473003/
>
> This requires changes in Chromium as well.
>
> BUG=skia:
> GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1716523002
>
> Committed: https://skia.googlesource.com/skia/+/
095d31c8a0eeb5d491febf064bc3c8a44e22b94f
TBR=scroggo@google.com,reed@google.com,bsalomon@google.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1775493002
benjaminwagner [Mon, 7 Mar 2016 17:09:19 +0000 (09:09 -0800)]
Revert of Add BigRectGM. (patchset #2 id:20001 of https://codereview.chromium.org/
1758113005/ )
Reason for revert:
Causing ASAN failures. (I had expected float-cast ASAN failures, which are disabled right now, but we're also seeing integer overflows.)
Original issue's description:
> Add BigRectGM.
>
> BUG=skia:5060
> GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1758113005
>
> Committed: https://skia.googlesource.com/skia/+/
a9101eef5ffc5f329a5edc75fd67441d868b0e9b
TBR=mtklein@google.com
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:5060
Review URL: https://codereview.chromium.org/
1773593005
msarett [Mon, 7 Mar 2016 16:39:12 +0000 (08:39 -0800)]
Update Skia's YUV API
We should match the recently designed API in SkCodec.
https://codereview.chromium.org/
1549473003/
This requires changes in Chromium as well.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1716523002
Review URL: https://codereview.chromium.org/
1716523002
benjaminwagner [Mon, 7 Mar 2016 16:16:11 +0000 (08:16 -0800)]
Add BigRectGM.
BUG=skia:5060
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1758113005
Review URL: https://codereview.chromium.org/
1758113005
update-skps [Mon, 7 Mar 2016 15:28:08 +0000 (07:28 -0800)]
Update SKP version
Automatic commit by the RecreateSKPs bot.
TBR=
NO_MERGE_BUILDS
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1769023002
Review URL: https://codereview.chromium.org/
1769023002
msarett [Mon, 7 Mar 2016 15:09:03 +0000 (07:09 -0800)]
Use a smart pointer for SkColorSpace factories
This should fix master-skia in Android because we no longer need
to include src/core for SkCodec.h.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1766413002
Review URL: https://codereview.chromium.org/
1766413002
robertphillips [Mon, 7 Mar 2016 12:51:10 +0000 (04:51 -0800)]
Add SkSpecialImage-based methods to SkImageFilter
This is calved off of https://codereview.chromium.org/
1695823002/ (Get OffsetImageFilter really working with SkSpecialImages)
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1757983002
Review URL: https://codereview.chromium.org/
1757983002
reed [Mon, 7 Mar 2016 02:10:48 +0000 (18:10 -0800)]
Add variants of the setters on SkPaint which take a sk_sp<effect>. At the same time, change the internal storage to be sk_sp<effect>.
Follow-on CL might try to use = default for the constructors and assignment operators.
This reverts commit
992854d62e179a589aa7366e443246e3672c3248.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1770723002
Review URL: https://codereview.chromium.org/
1770723002
halcanary [Sun, 6 Mar 2016 22:47:10 +0000 (14:47 -0800)]
SkPDF: AutoTUnref<T> changes in use
::detach() -> ::release()
::operator T*() -> ::get()
This makes all use of AutoTUnref work the same as sk_sp.
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1772493002
Review URL: https://codereview.chromium.org/
1772493002
reed [Sun, 6 Mar 2016 21:54:00 +0000 (13:54 -0800)]
Fix behavior of sk_sp::reset(T*) and add unittest.
Previously, sk_sp::reset(T* t) did not release its own reference
if its internal pointer was the same as 't'. This leaks a reference.
Now always release the current reference when non-nullptr.
Review URL: https://codereview.chromium.org/
1767983002
reed [Sun, 6 Mar 2016 01:51:45 +0000 (17:51 -0800)]
Revert of add setter on SkPaint that takes sk_sp (patchset #1 id:1 of https://codereview.chromium.org/
1769803002/ )
Reason for revert:
investigate leak
Original issue's description:
> add setter on SkPaint that takes sk_sp
>
> BUG=skia:
> GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1769803002
>
> TBR=
>
> Committed: https://skia.googlesource.com/skia/+/
a917eba6ea8a6936f2c9271e487b14d14b99c98e
TBR=
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=skia:
Review URL: https://codereview.chromium.org/
1771613002
reed [Sun, 6 Mar 2016 00:39:25 +0000 (16:39 -0800)]
add setter on SkPaint that takes sk_sp
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1769803002
TBR=
Review URL: https://codereview.chromium.org/
1769803002
halcanary [Sat, 5 Mar 2016 16:30:28 +0000 (08:30 -0800)]
sk_sp: fix contravariant constructors
TBR=reed
This is obviously correct and needs to be fixed.
Review URL: https://codereview.chromium.org/
1771583002
reed [Sat, 5 Mar 2016 14:00:12 +0000 (06:00 -0800)]
remove unneeded flag for annotations
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1767783004
TBR=halcanary
Review URL: https://codereview.chromium.org/
1767783004
reed [Sat, 5 Mar 2016 00:36:20 +0000 (16:36 -0800)]
Move annotations to canvas virtual (patchset #8 id:140001 of https://codereview.chromium.org/
1744103002/ )"
Need to land chrome change first https://codereview.chromium.org/
1766723003/
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1758023003
TBR=
Review URL: https://codereview.chromium.org/
1758023003
jvanverth [Fri, 4 Mar 2016 23:16:25 +0000 (15:16 -0800)]
Fix stall in Vk clear
TBR=egdaniel@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1765863003
Review URL: https://codereview.chromium.org/
1765863003
msarett [Fri, 4 Mar 2016 22:20:49 +0000 (14:20 -0800)]
Check libpng version before reading color space
TBR=scroggo@google.com
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1767723003
Review URL: https://codereview.chromium.org/
1767723003
halcanary [Fri, 4 Mar 2016 21:53:22 +0000 (13:53 -0800)]
SkPDF: PDFDevice use SkTArray<T> rather than SkTDArray<T*>
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1767713002
Review URL: https://codereview.chromium.org/
1767713002
bsalomon [Fri, 4 Mar 2016 21:38:26 +0000 (13:38 -0800)]
Support building Vulkan on Linux.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1761163003
Review URL: https://codereview.chromium.org/
1761163003
msarett [Fri, 4 Mar 2016 21:27:35 +0000 (13:27 -0800)]
Set SkColorSpace object for PNGs and parse ICC profiles
Code for ICC profile parsing adapted from:
https://codereview.chromium.org/
1707033002/
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1726823002
Review URL: https://codereview.chromium.org/
1726823002
bsalomon [Fri, 4 Mar 2016 21:15:29 +0000 (13:15 -0800)]
Turn on gpu stats in debug builds.
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1769633002
Review URL: https://codereview.chromium.org/
1769633002
reed [Fri, 4 Mar 2016 20:56:51 +0000 (12:56 -0800)]
disable linear pipeline for now (so bots don't crash)
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1766813002
TBR=
Review URL: https://codereview.chromium.org/
1766813002
brianosman [Fri, 4 Mar 2016 19:59:27 +0000 (11:59 -0800)]
Ignore backspace when appending to the filter string in SampleApp. Fixes a bug on Windows where the initial Backspace gets appended, causing no slides to match.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1767683002
Review URL: https://codereview.chromium.org/
1767683002
reed [Fri, 4 Mar 2016 19:07:43 +0000 (11:07 -0800)]
Add support for new bitmapshader context (patchset #5 id:80001 of https://codereview.chromium.org/
1757993002/ )"
This reverts commit
cd660e1c07371d9cf97824245639b1c0b5ac92fc.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1760123003
Review URL: https://codereview.chromium.org/
1760123003
fmalita [Fri, 4 Mar 2016 19:01:24 +0000 (11:01 -0800)]
Relocate the specialized linear gradient interval builder
The specialized interval setup works really well for (4f) linear
gradients, but it seems unlikely to benefit other gradient subclasses.
Since it gets in the way of a general/fallback gradient impl, let's move
this code to Sk4fLinearGradient.
R=reed@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1764183002
Review URL: https://codereview.chromium.org/
1764183002
borenet [Fri, 4 Mar 2016 18:58:20 +0000 (10:58 -0800)]
Swarming tasks: print start and end time
This will help determine how long each component of the task takes.
NOTRY=true
BUG=skia:4763
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1769583002
Review URL: https://codereview.chromium.org/
1769583002
joshualitt [Fri, 4 Mar 2016 18:45:05 +0000 (10:45 -0800)]
Fix SkiaServe gpu JSON to work with any reordering algorithm
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1761003004
Review URL: https://codereview.chromium.org/
1761003004
herb [Fri, 4 Mar 2016 18:39:14 +0000 (10:39 -0800)]
break out the tile and matrix strategies
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1765953002
Review URL: https://codereview.chromium.org/
1765953002
halcanary [Fri, 4 Mar 2016 16:30:05 +0000 (08:30 -0800)]
Make Cmake work with debug build
Before this CL, the following failed to link:
cd .../skia
git fetch
git checkout origin/master
git clean -ffdx
SKIA="$PWD"
cd $(mktemp -d);
cmake "${SKIA}/cmake" -DCMAKE_BUILD_TYPE=Debug -G Ninja
ninja
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1757993006
Review URL: https://codereview.chromium.org/
1757993006
reed [Fri, 4 Mar 2016 15:38:11 +0000 (07:38 -0800)]
remove align16 calls in skhader context sizes. will handle this elsewhere as needed
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1768433003
TBR=
basically reverts previous CL, but keeps the create --> onCreate change
Review URL: https://codereview.chromium.org/
1768433003
egdaniel [Fri, 4 Mar 2016 15:35:10 +0000 (07:35 -0800)]
Add a cache of GrVkSamplers in GrVkResourceProvider.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1765523002
Review URL: https://codereview.chromium.org/
1765523002
bsalomon [Fri, 4 Mar 2016 15:06:43 +0000 (07:06 -0800)]
Revert of Don't allow nullptr in texels array params (unless using a transfer buffer). (patchset #3 id:60001 of https://codereview.chromium.org/
1765633002/ )
Reason for revert:
breaks vk build
Original issue's description:
> Don't allow nullptr in texels array params (unless using a transfer buffer).
>
> Require all levels in writePixels to have a non-nullptr.
> GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1765633002
>
> Committed: https://skia.googlesource.com/skia/+/
8ee78f31b2a29a5f76403755ea17bad9be74a3ec
TBR=jvanverth@google.com,cblume@google.com,cblume@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
Review URL: https://codereview.chromium.org/
1760343003
fmalita [Fri, 4 Mar 2016 14:29:35 +0000 (06:29 -0800)]
Fix SkRadialGradient center discontinuity
(AKA let's slow down radial gradients!)
Avoid discontinuities when r ~= 0.
BUG=skia:5052
R=reed@google.com,mtklein@google.com
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1753133003
Review URL: https://codereview.chromium.org/
1753133003
bsalomon [Fri, 4 Mar 2016 14:23:36 +0000 (06:23 -0800)]
Don't allow nullptr in texels array params (unless using a transfer buffer).
Require all levels in writePixels to have a non-nullptr.
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1765633002
Review URL: https://codereview.chromium.org/
1765633002
msarett [Fri, 4 Mar 2016 14:23:20 +0000 (06:23 -0800)]
Fix YUV horizontal and vertical sampling checks
The should fix crashes while decoding skps.
BUG=skia:4991
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1766523002
Review URL: https://codereview.chromium.org/
1766523002
borenet [Fri, 4 Mar 2016 12:55:26 +0000 (04:55 -0800)]
Swarming bots: setup for skipping download of build products
Turns out it's pretty easy to pass the compile outputs to the test task by just adding the hash to the "includes" list in the .isolated file. So the flow is:
1. Isolate skia repo
2. Run compile task, record hash of results
3. Isolate test inputs for DM. This writes a .isolated file
4. Edit the .isolated file from #3 to include the hash from #2
5. Upload the modified .isolated file to the isolate server
6. Trigger the swarming task for DM
7. Wait for DM task to finish, download results from isolate server
8. Upload results to GS as normal
I expect the swarming bots to break when this is committed due to the moved out directory. The associated recipe change will fix them.
NOTRY=true
BUG=skia:4763
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1759553003
Review URL: https://codereview.chromium.org/
1759553003
kkinnunen [Fri, 4 Mar 2016 08:12:33 +0000 (00:12 -0800)]
Avoid drawing NVPR DIF text when text size is 0
Avoid drawing text when using NVPR and device independent fonts.
The drawing calculations cause division by zero and produce NaNs
to transforms (due to fTextInverseRatio).
Reproed by top25desk_twitter.skp as well as the added testcase.
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1760163002
Review URL: https://codereview.chromium.org/
1760163002
reed [Fri, 4 Mar 2016 03:43:50 +0000 (19:43 -0800)]
disable assert for now
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1760373002
TBR=
Review URL: https://codereview.chromium.org/
1760373002
reed [Fri, 4 Mar 2016 02:18:25 +0000 (18:18 -0800)]
Enforce 16byte alignment in shader contexts (patchset #1 id:1 of https://codereview.chromium.org/
1759653004/ )"
This reverts commit
e38bcaf24b00066e167e03a5ac63cf828914d747.
BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1763973002
TBR=
Review URL: https://codereview.chromium.org/
1763973002
bsalomon [Fri, 4 Mar 2016 00:38:22 +0000 (16:38 -0800)]
Use installed vulkan SDK and remove checked in SDK
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1761643002
Review URL: https://codereview.chromium.org/
1761643002
bsalomon [Thu, 3 Mar 2016 23:43:03 +0000 (15:43 -0800)]
Pull and build shaderc rather than use checked in lib/header
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=
1760493003
Committed: https://skia.googlesource.com/skia/+/
7d4490ecfa4d974003f675ef004e77f85ecfadd2
Review URL: https://codereview.chromium.org/
1760493003