While working on an unrelated test, I accidentally triggered a bug in
Metal code generation. Our struct-equality helper functions did not
properly handle vector fields. Wrapping each comparison in `all(...)`
fixes the problem. (all() on a scalar is allowed and does nothing.)
Our struct comparison tests now include a vector and a matrix.
Change-Id: I59061ae9c3c3ab2c2dbdcb5257bc23e2257152af
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470399
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
The callsites output these triangles now. This removes the final
direct Ganesh dependency from the tessellator code.
Bug: skia:12524
Change-Id: Id0d16839ed35a2bd9a812b8747ca8669fac5bed8
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470196
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>
RenderPassTask should only represent a single RenderPass with possible
subpasses (represented by DrawPasses). This reconstructs the RenderPass
setup to match this thinking. RenderPassDesc now stores TextureProxys for
the attachments, since we may not have instantiated the Textures by the
time we create the RenderPassTask.
Bug: skia:12466
Change-Id: I4f8cab77a297aaf76bea7dd1cc3afd00143b8103
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470460
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
When calling ProgramToSkVM, you can now pass in a pointer to an
SkVMDebugInfo struct. Passing in this pointer now enables trace opcodes,
removing the need for a dedicated `fSkVMDebugTrace` flag.
In this CL, the struct is empty, but in followup CLs it will contain
mapping tables that can be used to translate trace instructions into
human-readable results (e.g. slot numbers to variable names).
Change-Id: I6b38ec559723babed1f6d2efc4c5c2579c3b99db
Bug: skia:12614
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470398
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Previously, none of our `runtime` tests relied on the input coordinate
in any way, so all of the logic was hoisted above the main loop in every
test. This CL adds an artificial reliance on the input coordinate so
that we have at least some SkVM tests with real code in the main loop.
This lets us see debug trace instructions interleaved with real code.
The input coordinate is clamped against a known uniform value
(`colorGreen` always contains 0101) so that the final test output
remains consistent in practice.
Additionally, I noticed that this test was only enabled in ES3, but
it doesn't seem to have anything ES3-specific in it, so it's now
enabled across the board.
Change-Id: Ie82f40b1060edb6071e300040ac59fb7d27094b0
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470397
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
b5adbe272e..9c486b6752
2021-11-12 syoussefi@chromium.org Unoverride upstreamed dEQP fix
2021-11-12 syoussefi@chromium.org Vulkan: Write perf warnings in command buffer
2021-11-12 b.schade@samsung.com Vulkan: Reset mFramebuffer when mFramebufferCache is cleared
2021-11-12 syoussefi@chromium.org Vulkan: Make write-after-invalidate checks more precise
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 bsalomon@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: bsalomon@google.com
Test: Test: --deqp-surface-type=fbo --deqp-case=KHR-GLES31.*
Change-Id: Ia50994b00c16c0dcb7f96ad3e26a12b03dc4a367
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470595
Commit-Queue: skia-autoroll <skia-autoroll@skia-public.iam.gserviceaccount.com>
Bot-Commit: skia-autoroll <skia-autoroll@skia-public.iam.gserviceaccount.com>
- PathTessellator was accidentally not updating its own attribs when
kExplicitCurveType was needed.
- GrPathTessellationShader was asserting too broadly surrounding
kExplicitCurveType.
Bug: skia:12524
Change-Id: Iafe543cbc65c2225a9a99f14d7172182033e92c9
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470518
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Commit-Queue: Jorge Betancourt <jmbetancourt@google.com>
Auto-Submit: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Jorge Betancourt <jmbetancourt@google.com>
This is needed for enabling Skia's Vulkan backend for Flutter on Windows and iOS 9, as the standard libraries in both of these contexts don't have `std::shared_mutex`.
When disabled on Windows, VMA falls back to using Win32 SRWLock.
The `vulkan.h` import in `vk_mem_alloc.h` which would normally import `windows.h` is commented out for Skia, and so the relevant symbols need to be declared before importing `vk_mem_alloc.h`.
Relevant Flutter Engine PR: https://github.com/flutter/engine/pull/29520
Change-Id: I2b1a621417155071b21e5c6d3b5ccc375c92a622
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469207
Reviewed-by: Greg Daniel <egdaniel@google.com>
Auto-Submit: Brandon DeRosier <bdero@google.com>
Commit-Queue: Greg Daniel <egdaniel@google.com>
The tessellators are now agnostic of the shader class, which makes
them mostly usable by Graphite now. Also moves the code for writing
the middle-out vertex and index buffers into the PatchTessellators,
which makes sense since they also write the instance buffers.
Bug: skia:12524
Change-Id: I6c415645e389e056c0db1d93663b1b295d6b4535
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469631
Reviewed-by: Greg Daniel <egdaniel@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
This ports the third_party BUILD.gn files related to codecs
(with a best-effort on arm/SIMD stuff). This includes:
- libpng
- libjpeg-turbo
- libwebp
- wuffs (gif)
- libgifcodec
- dng_sdk and piex (raw codec)
This expands the string_flag_with_values macro to allow
multiple values to be set at once. This was added in Bazel 5.0.0,
however the latest pre-release version of that has a bug [1]
which slows down compilation dramatically. This was fixed at
ToT, but not released. As a result, I started using the Bazel
6 pre-release (via bazelisk).
The macro select_multi makes writing select() where multiple
elements could be on possible/easier.
One can try compiling certain codecs by running:
bazel build :skia-core --config clang --include_codec=raw_codec --include_codec=png_codec
Suggested Review Order:
- bazel/macros.bzl
- bazel/common_config_settings/defs.bzl and its BUILD.bazel
to see how the codec options are defined.
- BUILD.bazel to see how the codec settings are used.
- src/codec/BUILD.bazel to see the inclusion of Skia files to
deal with specific codecs.
- third_party/BUILD.bazel (while referencing the corresponding
BUILD.gn files, such as third_party/libwebp/BUILD.gn)
- Everything else.
Change-Id: I797375a35fa345d9835e7b2a2ab23371c45953c3
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469456
Reviewed-by: Leandro Lovisolo <lovisolo@google.com>
Reviewed-by: Ben Wagner <bungeman@google.com>
Reviewed-by: Leon Scroggins <scroggo@google.com>
This is a reland of 9613060bdf
Original change's description:
> Implement batching for convex tessellated paths
>
> Moves the view matrix transformation onto the CPU (when local coords
> are not used), and plumbs a color attrib through the tessellation
> patches.
>
> Bug: skia:12524
> Change-Id: Ic4740048b4fc85cb18e7235a7754d57762db3daf
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/468997
> Reviewed-by: Michael Ludwig <michaelludwig@google.com>
> Commit-Queue: Chris Dalton <csmartdalton@google.com>
Bug: skia:12524
Change-Id: I77cd079d8921b224925bd2f7364c564822886d61
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470436
Commit-Queue: Robert Phillips <robertphillips@google.com>
Auto-Submit: Chris Dalton <csmartdalton@google.com>
Reviewed-by: Robert Phillips <robertphillips@google.com>
Our inliner would ignore any functions with `inout` parameters, because
inlining them properly was more complex than just leaving the function
call. However, real-world code can sometimes contain helper functions
that have `inout` params that are never used at all (e.g. an uber-shader
with some features turned off).
We now read the ProgramUsage and check to see whether or not the
`inout`-qualified parameter is actually modified. If it's never changed,
the function now remains a candidate for inlining.
Change-Id: I92e494f94cc070801cb9aa28bd13faa689b806b6
Bug: skia:12636
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470299
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Yesterday's implementation was close but I realized later that it wasn't
quite ideal.
- Array index-folding was gated on `isCompileTimeConstant`, which is too
strict. The real limitation is `hasSideEffects`. If an array contains
a side-effecting expression, we should leave it alone. Otherwise it
is safe to pluck out an element from the array and toss the rest.
- Matrix index-folding was gated on `getConstantSubexpression` for the
extracted elements, but did not check the other elements at all. This
was too lenient; we now only proceed to the folding step if
`hasSideEffects` returns false.
I added some tests to verify the final behavior and also discovered a
small related issue. Diagonal matrices were not substituting literals
in for constant-values, which inhibited folding as well and would break
constant-expression evaluation. This is now fixed.
Change-Id: Idda32fd8643c1f32ba21475251cd4d4dd7cea94c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470396
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Functions which don't write to their out params should be safe to
inline, but we currently don't recognize this.
Change-Id: I753e48067c7be4473675ef6c95e61af17dc5ae41
Bug: skia:12636
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470298
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
We never used it internally, but the shaders used by Filament rely on
it. It doesn't exist in ES2 so this doesn't affect Runtime Effects.
Change-Id: Idb2afb15ff160b950ad02101bf6381a5d5c56468
Bug: skia:12635, skia:11209
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470156
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Change-Id: I7d441b5d4773b806be6cfc885b6ca13b6590dd46
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470456
Commit-Queue: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: Brian Salomon <bsalomon@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Replaced by Graphite.
Change-Id: Iac0ba212b078904a591677c9ce839a90562d0240
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470305
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
This appears unnecessary. SkGetJpegInfo() does not
need to be defined before implementation, right?
Change-Id: I07f6a689cd967c4c17de32e4429f2a8fa142128b
Bug: skia:12584
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470304
Commit-Queue: Leon Scroggins <scroggo@google.com>
Reviewed-by: Leon Scroggins <scroggo@google.com>
Bug: skia:12637
Change-Id: I2e2e5f14d5dac340a582e99f9df5ffed3c35c769
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470300
Reviewed-by: Ravi Mistry <rmistry@google.com>
Commit-Queue: Eric Boren <borenet@google.com>
Make CPU drawVertices and drawAtlas use SkVM if a SkRasterPipeline
blitter creation fails (e.g. due to runtime color filter).
Change-Id: Ib272f58cb729b7047e5c48eba866b435f4e075a3
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469903
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
Assert assumed that subtracting a smaller float from a larger float
and dividing by a positive float would produce a result greater than
zero. However, it may underflow and produce zero.
Bug: chromium:1268488
Change-Id: I1fac0db0801d105537b288c726869b22b836d24a
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469901
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
This reverts commit 9613060bdf.
Reason for revert: assertion failure on ANGLE
Original change's description:
> Implement batching for convex tessellated paths
>
> Moves the view matrix transformation onto the CPU (when local coords
> are not used), and plumbs a color attrib through the tessellation
> patches.
>
> Bug: skia:12524
> Change-Id: Ic4740048b4fc85cb18e7235a7754d57762db3daf
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/468997
> Reviewed-by: Michael Ludwig <michaelludwig@google.com>
> Commit-Queue: Chris Dalton <csmartdalton@google.com>
TBR=egdaniel@google.com,robertphillips@google.com,csmartdalton@google.com,michaelludwig@google.com,skcq-be@skia-corp.google.com.iam.gserviceaccount.com
Change-Id: Ie086376656777bb167075a9822d8c702cc7e41ec
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:12524
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/470296
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Commit-Queue: Robert Phillips <robertphillips@google.com>
Moves the view matrix transformation onto the CPU (when local coords
are not used), and plumbs a color attrib through the tessellation
patches.
Bug: skia:12524
Change-Id: Ic4740048b4fc85cb18e7235a7754d57762db3daf
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/468997
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Chris Dalton <csmartdalton@google.com>
Indexing into a constant matrix is a constant expression, so we are
obligated to support it for ES2 compatibility.
Change-Id: Ibe1e5bac39d9a88ce0222997a38e8b6952fdb336
Bug: skia:12472
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469819
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Change-Id: Ibeea56ebd5dce53af1252ca3ecf6cc6f010bd461
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469902
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
We should support constant-expressions involving matrices (GLSL ES2
does, WebGL does). We currently don't. We do properly report out-of-
range indexing, but we don't optimize away valid matrix index
expressions or allow matrices to be indexed in a constant-expression
context.
Change-Id: If58aa4c5f15abef421a412957072f3617b4176df
Bug: skia:12472
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469818
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
When optimizations are disabled, the compiler flags this function as
exiting without returning a value (since it doesn't convert the ifs from
maybe-taken into definitely-taken).
Change-Id: I226d0f6ba8cc664aecf1c5afaaf03e92038185ba
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469821
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>
Lots of potentially messy changes.
Also includes a major recipe roll.
Bug: chromium:1256037
Change-Id: Id05779802c5ca05921d93fa73c21930723793585
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/467980
Reviewed-by: Ravi Mistry <rmistry@google.com>
Commit-Queue: Eric Boren <borenet@google.com>
Previously, SkSL was unable to resolve the constant expression `x[y]`
for a constant-array `x` and a constant-integer-scalar `y`. Now, if `x`
and `y` are known, we can replace `x[y]` with the indexed array element.
Note that we need to be careful here, as it's not a valid optimization
to eliminate array elements that have side effects. We preserve side-
effecting expressions using the comma operator.
Change-Id: I5721337eb42b48c0b05f919c1cadfae19dd3b84f
Bug: skia:12472
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469839
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Previously, we didn't have tests which leveraged constant-evaluation of
array indexing (because we didn't support it), and our test files
commingled constant-indexing into vectors with constant-indexing into
arrays.
The test files now separate vector- and array-handling into separate
tests, and a ton of new cases have been added to ArrayFolding. The
ArrayFolding tests now require constant-evaluation of array indexing,
so they fail in this CL, but will be fixed in the followup CL.
Change-Id: I3b663e743d97d6db80627bc9b7808f88c99917a7
Bug: skia:12472
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469528
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Handle kColor_GradientType (which is returned by asGradient
for color shaders) as a straight color paint server.
Bug: skia:12622
Change-Id: I46fa21ed23a7824d67fc0460f92d649743e1b1f2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469900
Commit-Queue: Florin Malita <fmalita@chromium.org>
Commit-Queue: Florin Malita <fmalita@google.com>
Reviewed-by: Tyler Denniston <tdenniston@google.com>
This is unnecessary - the child FPs don't pay attention to the input
color anyway. Simplify OverrideInput while we're here - no one uses it
in the specialized form.
Change-Id: I07cd7ccafd3451e0436b61362a7f2cecfd5633ab
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469896
Reviewed-by: Greg Daniel <egdaniel@google.com>
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Bug: skia:12630
Change-Id: Ie4d4d9a62b4152b8ce3ebb5edebfe9aab88a2a25
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469858
Reviewed-by: Eric Boren <borenet@google.com>
Commit-Queue: Ravi Mistry <rmistry@google.com>
Bug: skia:12625
Change-Id: I8ad737855f25df83c85d619d69b006ccbd83658d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469897
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
This is generally not a property of individual FPs in the shader tree,
it's a property of the root level FP, and how it applies alpha to the
color generated by the shader FP. So, two improvements:
1) Remove all of the coverage-as-alpha inhibition happening in
individual shaders.
2) At the root of paint conversion, if there's a shader FP (and no
primitive color), we either: use ApplyPaintAlpha, which *is* safe
with coverage-as-alpha (updated flags to reflect this), or disable
coverage as alpha (because our FP is now just generating a color and
ignoring incoming alpha).
Note that we *could* choose to ALWAYS use ApplyPaintAlpha, even if the
paint color is opaque. That could result in extra (useless) work in some
situations, but it would enable the coverage-as-alpha optimization far
more often than we get it today.
Also: We don't change the behavior when there is a primitive color,
because that code path is already disabling coverage-as-alpha, thanks to
OverrideInput.
Change-Id: I1f040ee1217cf03e731ad1b8b497dc8e4df2de2b
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/469518
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
