These doubles are sometimes used as doubles and sometimes floats,
but it doesn't make sense to use values that can't be a float as one.
Just skip those in the test combinatorics.
Change-Id: Ibfdb699cac80b260258b164f95361110eb85c152
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309483
Reviewed-by: Mike Klein <mtklein@google.com>
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
These should work fine through SkVMBlitter,
and I'm not really sure why we didn't add them
to SkRasterPipeline.
Bug: chromium:1113777
Change-Id: Id89f82e6a53bd49d88a9562309acf0ab53ea5ec5
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309472
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
This adds an error variable that keeps track of the total distance from
the simplified line, and includes it when determining if we should
keep the next point. Using a sum of line distance is just a heuristic
but seems to address the current case of over simplification while
allowing us to keep a greedy strategy.
Bug: chromium:1086705
Change-Id: I29e21724db6b30495c2934e376a5e4d787c846a9
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309328
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
Profiling showed that repeated calls to `input_texel_color` were
actually a very expensive part of the test. Fortunately, we can expose
the texture's pixel buffer to the unit test method and easily reclaim
that performance.
Change-Id: I2c1cdd57a3e14dc859bdf03d8131137ca81364ff
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309437
Commit-Queue: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This matches the Vulkan type. This also fixes a warning when comparing
to another uint64_t.
Update GrVkSamplerYcbcrConversion::Key to match as well.
Change-Id: I6fc5d9456ee466dc3d98240acf935139a7808a02
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309325
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Leon Scroggins <scroggo@google.com>
Change-Id: I2102056847efe9e56ba6e8150e0d451bffc8f204
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309378
Reviewed-by: Ben Wagner <bungeman@google.com>
Reviewed-by: Kevin Lubick <kjlubick@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
The fuzzer has a found a pathological case where the maxRadius
needs to be reduce by 17 ulps.
Reduce the larger radius an ulp at a time until the sum fits.
Change-Id: I9cc4528667e7f9e902eff34d447fd4040a09742e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309417
Commit-Queue: Herb Derby <herb@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
There's no ordering between elements that we need to maintain, so
there's no reason to use a collection that maintains an ordering.
Change-Id: Ic5b71d84578eaf3dc3503a0fe5064146302120d5
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309416
Commit-Queue: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
In this demo the user may choose between one of three path rendering methods
1. SVG
2. Canvas2D Path2D API
3. CanvasKit
SVGs is animated using css transforms on the main thread, while Canvas2D and CanvasKit are animated
in a worker using OffscreenCanvas.
While the user views the result of the rendering, the demo collects framerate data and displays it
so the user may compare the performance of the three methods.
Change-Id: I8cd6e079bab8815614e09a276cfe78bee9557fda
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309327
Reviewed-by: Kevin Lubick <kjlubick@google.com>
This is a net reduction in code size because this idea was already
implemented separately in two places:
- dump_fragment_processor_tree (GrProcessorSet)
- describe_fp (ProcessorTest)
This consolidates the implementations. This CL also fixes a handful of
dumpInfo() methods that were not sufficiently descriptive--e.g. the FP
name was missing, or the implementation was just buggy.
Change-Id: If34ac46c97e9ae431c7c64b1247fc619703580b2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309324
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Instead of using SkPoint to store position information,
we can constrain the position information to be integers.
This will lead to two future improvements.
* Shrink position information to SkIPoint16
* Use integer arithmetic instead of floating point
in vertex calculation.
I'm interested in feedback on the comment which describes
the math behind this technique.
Change-Id: I8441bcbcad99d07e6d6d5e1788d1a47d87f22923
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/306948
Commit-Queue: Herb Derby <herb@google.com>
Reviewed-by: Ben Wagner <bungeman@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
This old workaround causes us to disable transfer-from support
which in turn hurts "async" read pixels perf by falling back
to synchronous reads.
This workaround was relevant to early Vulkan drivers circa 2016.
Bug: chromium:1113233
Change-Id: Ie4d78b483700144809f829373c10224f5ede85ca
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309316
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
These would just distract from the core of the upcoming review
Change-Id: Iaf9fa2d45ad0059a184491e8d66cc73672f1d368
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309284
Reviewed-by: Adlai Holler <adlai@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
The area calculation can cause overflow, though worst case
scenario is incorrect drawing.
This doesn't seem like an important optimization. Also it's
not correct for decal tile mode.
Bug: chromium:1112139
Change-Id: I4365c6b39f1190a9102cc76d3de1ef9ed0ece356
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309282
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
Auto-Submit: Brian Salomon <bsalomon@google.com>
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
This wasn't used anywhere yet, so the omission of `fmt` went unnoticed.
Change-Id: I6f5899c31b91ff04240de735fd3f9ef5b6da227e
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309179
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Change-Id: I30c4f0a1d13ac49247bab36cfb5dae1dbccbf348
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309283
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
* Fixes quality value
* Enables grabbing sample positions
* Enables centering samples for non-hardware AA
Change-Id: I5cabc0804b4a57347b7072b100c17310155c0ebc
Bug: skia:10476
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/308798
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
This isn't connected to anything yet and parallels what is done
for the text GrAtlasManager.
The rough pattern is that the GrContext hierarchy implements creating
the atlas but it is only exposed/made available via the flush state
(which derives from GrDrawMeshOp::Target).
TBR=bsalomon@google.com
Bug: 1108408
Change-Id: Iad7b5648877dbd20840e8d47c1d24131098ded94
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309128
Reviewed-by: Robert Phillips <robertphillips@google.com>
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
This reverts commit d169e1915c.
Reason for revert: broke chrome via code generator
Original change's description:
> Migrate GrSurfaceContext readPixels to take direct context
>
> After this lands we'll proceed up the stack and add the direct
> context requirement to the public API and SkImage.
>
> Bug: skia:104662
> Change-Id: I4b2d779a7fcd65eec68e631757821ac8e136ddba
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309044
> Commit-Queue: Adlai Holler <adlai@google.com>
> Reviewed-by: Robert Phillips <robertphillips@google.com>
TBR=robertphillips@google.com,adlai@google.com
Change-Id: I6126f2dca4bc902c903512ac486e22841cc472e5
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:104662
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309281
Reviewed-by: Adlai Holler <adlai@google.com>
Commit-Queue: Adlai Holler <adlai@google.com>
In an ideal world we wouldn't keep re-allocating the memory for the keys
Change-Id: I3b7954e131d483166d0bf0ce95510365869b134c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309123
Reviewed-by: Adlai Holler <adlai@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
48ba75ac68..23335ac01c
2020-08-11 ynovikov@chromium.org Roll chromium_revision ed7912a248..c8c2c64629 (794870:796654)
2020-08-11 ynovikov@chromium.org Fix clang revision regexp in chromium autoroller to match new format
2020-08-11 jmadill@chromium.org TestRunner: Fix test result handling.
2020-08-11 ynovikov@chromium.org Skip WebGL2CompatibilityTest.CopyMip1ToMip0 on Mac Intel GL
2020-08-11 angle-autoroll@skia-public.iam.gserviceaccount.com Roll glslang from f881f08358b0 to 2de6d657dde3 (2 revisions)
2020-08-11 ynovikov@chromium.org Suppress dEQP-GLES3 failures on D3D11 NVIDIA
2020-08-10 jmadill@chromium.org Vulkan: Remove TODO.
2020-08-10 jmadill@chromium.org Vulkan: Refactor image dependency commands.
2020-08-10 nguyenmh@google.com Invalid calls' pointer params are no longer captured
2020-08-10 jonahr@google.com Enable required extensions for Chromium to use the Metal backend.
2020-08-10 nguyenmh@google.com Implement param capture for glGetUniform calls and extensions
2020-08-10 lehoangq@gmail.com Metal: Implement CHROMIUM_copy_texture
2020-08-10 nguyenmh@google.com Update readme with how to run regression testing script
2020-08-10 nguyenmh@google.com Change test result types + add additional logging
2020-08-10 timvp@google.com Disabling failing dEQP-GLES2.functional.fbo.render.recreate_* tests
2020-08-10 jmadill@chromium.org Vulkan: Simpify check for started RenderPass + FB.
2020-08-10 nguyenmh@google.com Use striping when making test batches
2020-08-10 angle-autoroll@skia-public.iam.gserviceaccount.com Roll Vulkan-Loader from aba3ab0ce72f to 61bf3be73c1e (1 revision)
2020-08-10 angle-autoroll@skia-public.iam.gserviceaccount.com Roll Vulkan-Tools from a1c21580f50b to 1590d46aaaeb (3 revisions)
2020-08-10 angle-autoroll@skia-public.iam.gserviceaccount.com Roll Vulkan-ValidationLayers from 9ba1bc28d878 to b399fb08896a (2 revisions)
2020-08-10 angle-autoroll@skia-public.iam.gserviceaccount.com Roll Vulkan-Headers from a6d08c75d219 to 8446db62f194 (4 revisions)
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 fmalita@google.com on the revert to ensure that a human
is aware of the problem.
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/master/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
Bug: chromium:1104516,chromium:1112800
Tbr: fmalita@google.com
Test: Test: dEQP-GLES2.functional.fbo.render.recreate_*
Change-Id: Ia6845ea3dd10a390f133ffa113392fb2c309a587
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309257
Reviewed-by: skia-autoroll <skia-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: skia-autoroll <skia-autoroll@skia-public.iam.gserviceaccount.com>
Runtime effects previously allowed two kinds of global input variables:
'in' variables could be bool, int, or float. 'uniform' could be float,
vector, or matrix. Uniform variables worked like you'd expect, but 'in'
variables were baked into the program statically. There was a large
amount of machinery to make this work, and it meant that 'in' variables
needed to have values before we could make decisions about program
caching, and before we could catch some errors. It was also essentially
syntactic sugar over the client just inserting the value into their SkSL
as a string. Finally: No one was using the feature.
To simplify the mental model, and make the API much more predictable,
this CL removes 'in' variables entirely. We no longer need to
"specialize" runtime effect programs, which means we can catch more
errors up front (those not detected until optimization). All of the API
that referred to "inputs" (the previous term that unified 'in' and
'uniform') now just refers to "uniforms".
Bug: skia:10593
Change-Id: I971f620d868b259e652b3114f0b497c2620f4b0c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309050
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
After this lands we'll proceed up the stack and add the direct
context requirement to the public API and SkImage.
Bug: skia:104662
Change-Id: I4b2d779a7fcd65eec68e631757821ac8e136ddba
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309044
Commit-Queue: Adlai Holler <adlai@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
This is causing failures on the status dashboard.
Change-Id: I29e8fb5bef72282dbe6fafb5402607ed48aae707
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309136
Commit-Queue: John Stiles <johnstiles@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>
This fixes the asan failures on the bots
Change-Id: I1baaf5596714262f7e3ddf368bb38b91b1ffcfcd
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309054
Commit-Queue: Greg Daniel <egdaniel@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Auto-Submit: Greg Daniel <egdaniel@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
Reviewed-by: Florin Malita <fmalita@chromium.org>
C++17 introduced `inline constexpr` as a mechanism for declaring strings
in headers safely without macros. See https://abseil.io/tips/168
Change-Id: I68071fbadebf32753ca4a0586b96e32634083f6a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309047
Commit-Queue: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
This refactoring was requested during an earlier code review.
Change-Id: I93184e2bc936b3c8c69f84e475be2c234845ee81
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309120
Reviewed-by: Adlai Holler <adlai@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Opted the Galaxy S20 out of the test to allow Tryjobs to pass.
Change-Id: I8d4637a23f36edb012c96b7059b184c231c0436d
Bug: skia:10384, skia:10595
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309119
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Previously, MakeChildFP avoided infinite recursion by rejecting any FP
that took inputs. MakeChildFP now generates random inputs up to a
user-supplied tree depth.
The ProcessorOptimizationValidationTest test has been updated to
test up to a tree depth of 3. The ProcessorCloneTest has been left at
a tree depth of 1 due to a bug that only appears on Galaxy S20/Mali G77.
The Mali bug doesn't appear to be related to FP cloning, but probably
deserves further analysis. (It appears that on this device, these
processors hooked together in sequence render a tiny bit differently
each time: DitherEffect -> RectBlurEffect -> ImprovedPerlinNoise. By
visual inspection it looks like the dither varies on each draw.)
Change-Id: Ib8f619eb7a8a9c9254080303504c20065ff35453
Bug: skia:10384, skia:10595
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/308556
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
There are no intentional behavioral changes in this CL. It should
purely be cut, paste, re-shuffle & get re-compiling.
Bug: 1108408
Change-Id: Ic8457d0af76861f2ea881177f2b2c0fe0bf85146
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309040
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Bug: 1108408
Change-Id: Iad0b634a8a0d6dac64e3571a32e755d124401f66
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309077
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Previously, this test would synthesize a random FP, call clone(), render
both, and the compare the results and report differences.
Now, if a difference is discovered, this test will re-synthesize the
originally-created random FP from scratch without using clone().
Instead, we call TestCreate again using the same random seed.
Then we can render and compare that output against the original as well.
This will allow to better differentiate failures that were actually
caused by clone(), versus failures caused by other types of
inconsistency.
If the regenerated version still mismatches, there are a variety of
potential explanations:
- the FP's TestCreate() does not always generate the same FP from a
given seed
- the FP's Make() does not always generate the same FP when given the
same inputs
- the FP itself generates inconsistent pixels (shader UB?)
- the driver has a bug
Change-Id: I71701c0f3d33a08f6ee926313782620487d336bd
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/309076
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
This is essnetially the middle set of work to fully optimize dst reads
in vulkan. On the upper end we still need to get the knowledge of dst
reads from the GrProgramInfo. And on the lower side we need to actually
create the render pass and correctly use it vulkan. This change is
essentially adding the plumbing that will sit between those two parts.
Bug: skia:10409
Change-Id: I58ebdc6574a9bdeddc3658f56b0a7b43f28473e6
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/308765
Commit-Queue: Greg Daniel <egdaniel@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
The style guide only mentioned the `k` prefix for enums, but it is
also used throughout Skia for constant values.
Also removed guidance about using anonymous enums to declare an
integral constant. static constexprs should be preferred here.
Change-Id: I2f7d4d375cc914d4efb9fe60fbd90ddaaf8c1b69
No-Try: true
Docs-Preview: https://skia.org/?cl=308919
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/308919
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Salomon <bsalomon@google.com>
Reviewed-by: Mike Klein <mtklein@google.com>
After this CL the GrSmallPathAtlasMgr is almost all ready to be moved
to be a flush-time entity.
This is pulled out of the omnibus CL:
https://skia-review.googlesource.com/c/skia/+/307776 (Split the small path renderer into record-time and flush-time pieces)
Change-Id: Ie6a2196a640ce9be7a1e1c6d7bbdfd41413c42d0
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/308977
Reviewed-by: Jim Van Verth <jvanverth@google.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
The current limit of 78 is quite conservative, and we see no diffs via
gold when using a limit of 24. This also fixes a timeout uncovered by a
fuzzer.
Bug: oss-fuzz:23989
Change-Id: I0d7c33afca7e7e852bedfea52b104d6c7d8fcfd5
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/301938
Commit-Queue: Mike Reed <reed@google.com>
Reviewed-by: Mike Reed <reed@google.com>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Explanation: The sksl standalone compiler is used to convert the raw
(text) SkSL pre-includes into a "dehydrated" binary format. It also
(previously) depended on those files, as they were #included and used,
unless a special #define was changed. This created a dependency cycle
that we hid from GN (by lying about the outputs of the dehydrate step).
As a result, builds would never reach steady-state, because the compiler
would be rebuilt (due to the newer dehydrated files), and then the
dehydrated files would be rebuilt (due to the newer compiler).
This CL changes the logic so that the standalone compiler always uses
the textual pre-includes, and no longer depends on the dehydrated binary
files. Thus, to make any kind of change to the dehydrated files (whether
due to pre-include changes, or the encoding format itself), you just
need skia_compile_processors enabled. The dependencies are now honestly
communicated to GN, and we reach steady state after one build.
The NOTE above is because GN/ninja cache the dependencies of each
target, and will still think that the SkSLCompiler.obj linked into the
standalone compiler depends on the dehydrated files, at least until one
successful build, when it will realize that's no longer true.
Reland notes:
The bots originally rejected this CL, because SkSLCompiler was
hard-coded to load the text files from a relative path that assumed the
executable was in "<skia_checkout>/out/<some_dir>". That's not true for
bots, and it was fragile, even for users. Now, we use GN to directly
generate sksl_fp.sksl, and copy all of the other pre-includes to the
root out dir (working directory when running skslc). This means we
no longer need to generate the sksl_fp.sksl file into the src tree, and
the compiler can more safely assume that the files will be in the
working directory.
Bug: skia:10571
Change-Id: Id7837a9aba7ee0c3f7fa82eb84f7761e24b9c705
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/308896
Reviewed-by: Mike Klein <mtklein@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
A lot of this careful load-this-now logic is moot given the data flow
and scheduling that skvm has been doing for the last few months. We can
write all this in simple unconditional canonical order now and it'll be
reordered, dead code eliminated, strength reduced, etc. as possible.
One really big simplification is that we can always notionally load and
apply coverage and let skvm handle whether or not it is really needed:
when coverage is known to be splat(1.0f), the pre-blend multiplies drop
away, and now that lerp() is a little smarter, the post-blend lerp()s
drop away too.
Change-Id: I0534b484bf927ef315335c991704ed52bb6010d5
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/308917
Reviewed-by: Mike Reed <reed@google.com>
Commit-Queue: Mike Klein <mtklein@google.com>
