This is a reland of e138730000
I fixed the tryjobs to move the compiled files into the right
location.
Original change's description:
> [pathkit] Reorganize to be more like CanvasKit
>
> Change-Id: I64226b57161f86277e0635592c2c11442888335a
> Bug: skia:8216
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/502817
> Reviewed-by: Nathaniel Nifong <nifong@google.com>
Bug: skia:8216
Change-Id: I535f906b43816d3143e95ff9350ba20b998d817d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503826
Reviewed-by: Nathaniel Nifong <nifong@google.com>
There are two approaches taken - naively using a bunch
of drawTexts and using drawGlyphs. The latter runs about
twice as fast (45 ms per frame vs 90 ms per frame) at the
expense of more complex code (need to layout individual
glyphs in the JS).
Change-Id: I5bc93048c0297f036215f86244b395cf812226a4
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/502369
Reviewed-by: Nathaniel Nifong <nifong@google.com>
Reviewed-by: Ben Wagner <bungeman@google.com>
The float/double versions of skstd::to_string produce better output than
the standard version, so let's keep those.
Other versions of to_string don't do anything at all now that
SkSL::String no longer exists. (Previously they existed to convert the
return type from std::string to SkSL::String.) We can just use the
std::to_string functions directly.
Change-Id: Ief513e474bd47ed97f1c13f4f64fb161f1654065
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503832
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Allows copying of multiple regions into a single texture, which supports
both filling mipLevels and rectangles in atlases.
Bug: skia:12845
Change-Id: I5fef9bda01125958a5491b4c5c8b3a66184357d2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/504056
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
This adds a Dehydrator.write(Program) to mirror the Rehydrator's
program() method and simplifies the API.
Change-Id: I1b6d6b722d0ce8e6a292132522f806e43d49ce85
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/502704
Reviewed-by: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
These Minis have a new CPU (I5-4278U) we've never cared about before, so
add that too. (We had some other Minis of the same model in the Golo,
but we didn't care about their CPUs because we were running only GPU
jobs on them.)
If you want to see a good diff of those, use FileMerge or something that
uses a similar algorithm; git makes a total hash of it.
Change-Id: Id7c4f6bc06f83d2ed4a0034e9174b9e69d3d15b2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/495424
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Erik Rose <erikrose@google.com>
You can still mutate program settings if you must, by going all the way
to ThreadContext::Context(), but unless your name is StartRuntimeShader,
you probably shouldn't.
Change-Id: Ie70a283970834f18cf8a30b6cd45dcdeb918ae0a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503829
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
We can use the native C++17 versions going forward, if we need them.
Change-Id: I53be7ed4ec72602b978c1dad252dff7be9e0f392
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503823
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Ben Wagner <bungeman@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
Have the AnalysisCanvas (also known as the diff canvas) convert
SkTextBlobs to GrSlugs and record the Strike differences. Make sure
these differences are tracked and recorded.
Bug: chromium:1278340
Change-Id: I2c8d62fa61511abd1e14d4bf595e6db1a0b5e26b
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503827
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Herb Derby <herb@google.com>
Change-Id: I8d4816f726196e6f4a5fbb0940b5dd1b2d7db7f5
Bug: skia:11209
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503821
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
This is a reland of d921f21fbc
Original change's description:
> Add SkSurface resolve function.
>
> This will insert a resolve msaa call into the stream of commands for
> the SkSurface. This is mostly useful for cases when a client wraps the
> resolve texture but has Skia draw with MSAA, and the client wants to
> make sure Skia resolves to their wrapped texture.
>
> Bug: chromium:1292418
> Change-Id: I6eddae967136716b9215fcd96e7d77a2457efdf2
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503340
> Reviewed-by: Robert Phillips <robertphillips@google.com>
> Commit-Queue: Greg Daniel <egdaniel@google.com>
Bug: chromium:1292418
Change-Id: I810b5618092c560f5bba900024b3b8c0c88baea9
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503717
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
This is a reland of 995d16fc91
Original change's description:
> For runtime shaders, deduce isOpaque automatically
>
> The forceOpaque parameter is now ignored. Instead, we do a conservative
> analysis of the shader's main function to determine if it always returns
> an opaque color. This is good enough to detect simple cases, including
> things like:
>
> return child.eval(p).rgb1
>
> Bug: skia:12643
> Bug: skia:12896
> Change-Id: I74b331aa12fadb1d0d1bb85f225dc7aa01ba2455
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503346
> Reviewed-by: John Stiles <johnstiles@google.com>
> Commit-Queue: Brian Osman <brianosman@google.com>
Bug: skia:12643
Bug: skia:12896
Change-Id: I4ae55cdd3f88c1b5a08bd59913df1c1cd48a4679
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503824
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Change-Id: Iaaaab94abd68a09b8194e267bca8f555c7a57a4e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503877
Commit-Queue: Robert Phillips <robertphillips@google.com>
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Reviewed-by: Eric Boren <borenet@google.com>
This is a reland of 805acda3f3
It fixes the #if SK_GL which was causing the Android roll
to fail.
This disables unit tests on Test-Ubuntu18-Clang-Golo-GPU-QuadroP400-x86_64-Release-All-TSAN_Vulkan
which were consistently crashing with OOM.
Original change's description:
> [fuzzer] Remove GL from (now-Vulkan) build
>
> The fuzzer runs against the Vulkan version of Swiftshader.
> There are no libGL.so (etc) on the fuzz runtime, so we
> want to avoid linking against those.
>
> The GL code that is #ifdef'd out is still necessary to
> avoid timeouts on TSAN with our NVIDIA jobs.
> https://skia-review.googlesource.com/c/skia/+/502638
>
> Procedure for testing this locally (and iterating):
> 1. In oss-fuzz checkout, run
> `python infra/helper.py shell skia`
> to pull up local interactive version of Docker
> fuzzer build image.
> 2. Run `compile` in fuzzer shell. Stop after
> the swiftshader compiles and is copied into /out
> with Ctrl + C.
> 3. Comment out the swiftshader compilation part [1]
> (no need to re-do this when modifying Skia code).
> `apt-get install nano -y`
> `nano ../build.sh`
> 4. Make change to Skia repo using normal methods.
> 5. Run the following in the Skia repo
> `git diff origin main > foo.patch`
> Copy the patch into the Docker shell using Ctrl+C
> and nano.
> 6. Apply the patch inside the Docker shell
> `git apply foo.patch`
> and re-compile (which should skip right to
> building the fuzzer libs)
> `compile`
> 7. Repeat 4-7 or make small changes directly in
> the Docker shell via nano.
> 8. When compilation and link succeeds, run
> `ldd /out/api_mock_gpu_canvas`
> to verify GL and friends were not dynamically linked.
>
> [1] https://github.com/google/oss-fuzz/pull/7214/files#diff-76f13875e33875cdd372f1f0933206be599cd87952f1bd1eaa57ca928ee9e3e1R49-R53
>
> Change-Id: Idf569820527c1304b0e5a68fd36295be89dfa2a0
> Bug: oss-fuzz:44132
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503016
> Reviewed-by: Brian Osman <brianosman@google.com>
> Commit-Queue: Kevin Lubick <kjlubick@google.com>
Bug: oss-fuzz:44132, skia:12900
Change-Id: Ia2eff9403b0035e7f86098f296d7d9b1bbfd4876
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503716
Reviewed-by: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Bug: skia:12701
Change-Id: I9af3c437512f2538066ddbbe576739c8873d0bc0
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503345
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Requires tweaking one inliner test to avoid an Intel driver bug (on
ANGLE).
Bug: chromium:709351
Cq-Do-Not-Cancel-Tryjobs: true
Change-Id: I08fac938396d6b90805ba9650c7a520af888bc12
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503819
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Bug: skia:12701
Change-Id: I33d24fc319acbfbb3a218a8dd916c5a64f15f028
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503342
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Bug: skia:12794
Change-Id: I445144f25e9573d0d9ec55d12c3a9d372772c3bd
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503348
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
Bug: skia:12701
Change-Id: Ia8189d75621ad2c6a15ed91c8a0b552322f40ade
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/502314
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
A glyph cache entry is made up of metrics, an image, and a path. If an
entry does not exist then nothing is known of the glyph yet. If there is
an entry it must have at least the metrics. The image and the path may
logically be added lazily as needed. Prior to this change it was not
possible to send both the image and the path of a single glyph. The
added test fails when a given strike is used for both the image and the
path. With this change the test passes, as the image is sent and the
path is sent, and both are merged into the remote glyph. The assertions
around not creating the glyph more than once are replaced with
assertions that no part of the glyph is set more than once.
Change-Id: I2fd4047ee15e0d584b68e1180c6fe15224889310
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503347
Reviewed-by: Herb Derby <herb@google.com>
Reviewed-by: Derek Sollenberger <djsollen@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
This reverts commit 995d16fc91.
Reason for revert: Breaks one SkSL test on ANGLE
Original change's description:
> For runtime shaders, deduce isOpaque automatically
>
> The forceOpaque parameter is now ignored. Instead, we do a conservative
> analysis of the shader's main function to determine if it always returns
> an opaque color. This is good enough to detect simple cases, including
> things like:
>
> return child.eval(p).rgb1
>
> Bug: skia:12643
> Bug: skia:12896
> Change-Id: I74b331aa12fadb1d0d1bb85f225dc7aa01ba2455
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503346
> Reviewed-by: John Stiles <johnstiles@google.com>
> Commit-Queue: Brian Osman <brianosman@google.com>
Bug: skia:12643
Bug: skia:12896
Change-Id: I7d86933f29ddb373222dff7f2a28a413fd777002
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503818
Auto-Submit: Brian Osman <brianosman@google.com>
Commit-Queue: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
It is up to the caller of SkCanvas::internalSaveLayer and
SkCanvas::internalRestore properly track SkCanvas::fSaveCount. The
AutoLayerForImageFilter class did not update these, so any attempt to
save a layer when it is active would potentially restore incorrectly.
Bug: chromium:1292419
Change-Id: Ic64c899d6da99c6c8e6848081379e7cbb4d37fa8
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503488
Reviewed-by: Herb Derby <herb@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
I have trouble remembering how to interpret the skip
strings, so I documented it more and required the use of
four named arguments. Regex helped transform the old calls
into the new ones.
Change-Id: I2b970053b95b9a4d3d90a718b5790b0b25ce8dae
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503736
Reviewed-by: Eric Boren <borenet@google.com>
These are guarded with $es3, so we can use them in SkSLTest ES3 tests,
but are not generally accessible in Runtime Effects yet.
(Metal will need a polyfill as well: skia:12898)
Change-Id: Ibc1e4b9dc230517163a8f6a4a3af0172d8275de1
Bug: skia:11209
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503483
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
The forceOpaque parameter is now ignored. Instead, we do a conservative
analysis of the shader's main function to determine if it always returns
an opaque color. This is good enough to detect simple cases, including
things like:
return child.eval(p).rgb1
Bug: skia:12643
Bug: skia:12896
Change-Id: I74b331aa12fadb1d0d1bb85f225dc7aa01ba2455
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503346
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
The previous CLs have removed the last significant differences between
SkSL::String and std::string. This CL removes SkSL::String entirely and
replaces it with std::string throughout the code.
Apologies for the very long CL, but I have done my best to make it as
simple and reviewable as possible. The vast majority of changes are
simple replacement of `SkSL::String` with `std::string`. In the rare
spots where code is moved from one place to another, it is logically
unchanged.
Change-Id: I39563d2db45da229f17f4504dfd63e00bde7a96e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503339
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
This reverts commit 805acda3f3.
Reason for revert: Looks to be breaking the Android roll. See https://android-build.googleplex.com/builds/pending/P29733268/aosp_bramble-userdebug/latest/view/logs/build_error.log
Original change's description:
> [fuzzer] Remove GL from (now-Vulkan) build
>
> The fuzzer runs against the Vulkan version of Swiftshader.
> There are no libGL.so (etc) on the fuzz runtime, so we
> want to avoid linking against those.
>
> The GL code that is #ifdef'd out is still necessary to
> avoid timeouts on TSAN with our NVIDIA jobs.
> https://skia-review.googlesource.com/c/skia/+/502638
>
> Procedure for testing this locally (and iterating):
> 1. In oss-fuzz checkout, run
> `python infra/helper.py shell skia`
> to pull up local interactive version of Docker
> fuzzer build image.
> 2. Run `compile` in fuzzer shell. Stop after
> the swiftshader compiles and is copied into /out
> with Ctrl + C.
> 3. Comment out the swiftshader compilation part [1]
> (no need to re-do this when modifying Skia code).
> `apt-get install nano -y`
> `nano ../build.sh`
> 4. Make change to Skia repo using normal methods.
> 5. Run the following in the Skia repo
> `git diff origin main > foo.patch`
> Copy the patch into the Docker shell using Ctrl+C
> and nano.
> 6. Apply the patch inside the Docker shell
> `git apply foo.patch`
> and re-compile (which should skip right to
> building the fuzzer libs)
> `compile`
> 7. Repeat 4-7 or make small changes directly in
> the Docker shell via nano.
> 8. When compilation and link succeeds, run
> `ldd /out/api_mock_gpu_canvas`
> to verify GL and friends were not dynamically linked.
>
> [1] https://github.com/google/oss-fuzz/pull/7214/files#diff-76f13875e33875cdd372f1f0933206be599cd87952f1bd1eaa57ca928ee9e3e1R49-R53
>
> Change-Id: Idf569820527c1304b0e5a68fd36295be89dfa2a0
> Bug: oss-fuzz:44132
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503016
> Reviewed-by: Brian Osman <brianosman@google.com>
> Commit-Queue: Kevin Lubick <kjlubick@google.com>
Bug: oss-fuzz:44132
Change-Id: I3832417c60ff425572717d37dc9609419922b18e
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503351
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Reviewed-by: Brian Osman <brianosman@google.com>
fe56532f50..b756b2c274
2022-02-03 jonahr@google.com Suppress flaky Mipmap/Multithreading tests on Win/SwS/ASAN
2022-02-03 kbr@chromium.org Metal: End the current render pass at glInvalidateFramebuffer.
2022-02-03 cnorthrop@google.com Vulkan: Update default FBO when fetch in use
2022-02-02 kbr@chromium.org Mac: Try to diagnose crashes inside Core Animation.
2022-02-02 ynovikov@chromium.org Revert "Vulkan: Enable ANGLE_texture_multisample"
2022-02-02 lexa.knyazev@gmail.com Metal: Remove emulatedInstanceID
2022-02-02 angle-autoroll@skia-public.iam.gserviceaccount.com Roll vulkan-deps from 9ee740d2e9ef to fe093f4789c2 (5 revisions)
2022-02-02 syoussefi@chromium.org Vulkan: Fix a few framebuffer fetch bugs
If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/angle-skia-autoroll
Please CC borenet@google.com on the revert to ensure that a human
is aware of the problem.
To file a bug in ANGLE: https://bugs.chromium.org/p/angleproject/issues/entry
To file a bug in Skia: https://bugs.chromium.org/p/skia/issues/entry
To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug
Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md
Cq-Include-Trybots: skia/skia.primary:Build-Debian10-Clang-x86_64-Release-ANGLE;skia/skia.primary:Test-Win10-Clang-AlphaR2-GPU-RadeonR9M470X-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-Golo-GPU-QuadroP400-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-NUC5i7RYH-GPU-IntelIris6100-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-NUC6i5SYK-GPU-IntelIris540-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-NUC8i5BEK-GPU-IntelIris655-x86_64-Debug-All-ANGLE;skia/skia.primary:Test-Win10-Clang-NUCD34010WYKH-GPU-IntelHD4400-x86_64-Debug-All-ANGLE
Tbr: borenet@google.com
Test: Test: FramebufferFetchES31.DefaultFramebufferMixedProgramsTest
Test: Test: FramebufferFetchES31.DefaultFramebufferTest
Change-Id: I8d1eab3d78f704f8276e48ad335fb0d3415add8d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503557
Commit-Queue: skia-autoroll <skia-autoroll@skia-public.iam.gserviceaccount.com>
Bot-Commit: skia-autoroll <skia-autoroll@skia-public.iam.gserviceaccount.com>
Instead of CubicPatch(writer) << control points ... code now writes
writer << Cubic(controlPoints). The main benefit of this is that the
control points are available to the PatchWriter. This allows it (in
follow up CLs) to automatically chop the curves and track the last join
control point.
Change-Id: I38b57a141261afd42e511cc195eb5e64a2382263
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/501837
Reviewed-by: Christopher Dalton <csmartdalton@google.com>
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
std::string unfortunately does not have a `string + string_view`
plus operator. (https://stackoverflow.com/q/44636549/291737)
This brings SkSL::String one step closer to matching std::string's API.
Change-Id: I44fd4538a9433442d2f34e63ebd120f3377fbb70
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503343
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
std::string is less permissive than SkSL::String about concatenation.
This change is a step towards eliminating SkSL::String.
(We couldn't continue using this technique for std::string, as it's
undefined behavior to add our own methods inside namespace std.)
Change-Id: I21e421182be23d3033f827758ea2b2c01fa99d26
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503341
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This reverts commit d921f21fbc.
Reason for revert: New test breaking GL android bots
Original change's description:
> Add SkSurface resolve function.
>
> This will insert a resolve msaa call into the stream of commands for
> the SkSurface. This is mostly useful for cases when a client wraps the
> resolve texture but has Skia draw with MSAA, and the client wants to
> make sure Skia resolves to their wrapped texture.
>
> Bug: chromium:1292418
> Change-Id: I6eddae967136716b9215fcd96e7d77a2457efdf2
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503340
> Reviewed-by: Robert Phillips <robertphillips@google.com>
> Commit-Queue: Greg Daniel <egdaniel@google.com>
Bug: chromium:1292418
Change-Id: I86e5f82f0e2a0921906c0caba964929750500965
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/503350
Auto-Submit: Greg Daniel <egdaniel@google.com>
Commit-Queue: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
