SkSL does not support shrinking a vector via casting. Use a swizzle
instead.
Change-Id: Ieba78a05dad9c55f44c765924e28f0c7e1667a67
Bug: skia:12193
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/427198
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Boolean scalar-swizzling is currently not working.
Change-Id: Icd965e4b64a12311d098168f65622110d5fb3437
Bug: skia:12195
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/427038
Commit-Queue: John Stiles <johnstiles@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
The tests now check bool4-mat2 conversions, which fortunately do work,
and the vector-to-matrix tests include int and bool conversions as well.
Change-Id: I971271838a93081b9258deb7c1d13b7732fb2440
Bug: skia:12067
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/426757
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
The fuzzer quickly discovered that the newly introduced mat2-to-vec4
conversion code did not account for integer vectors. We now handle
`ivec4(mat2)` casts properly. This required some non-trivial
restructuring of the logic, but in the vast majority of cases, the types
will match and the end result will be identical.
Change-Id: If07c2fe4b4345bd767384b1802374910f65cd3f0
Bug: oss-fuzz:35998
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/426756
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
GLSL disallows mixing swizzle domains within a single swizzle:
http://screen/93eHNQDbx35hMdk
SkSL now disallows it as well.
Change-Id: Ied2e11ee04285b143a864e28cac30335f01aad0e
Bug: skia:10621
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/426458
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
GLSL supports casting vec4 into mat2 and vice versa, so SkSL should have
equivalent support. This CL allows the Compound constructor to take a
matrix as input, and fixes up backends to do the right thing when a
matrix shows up in the compound-constructor path.
Change-Id: I13289ad0a27ba59bddc3706093820594efebc693
Bug: skia:12067
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/426003
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Bug: skia:11296
Change-Id: I7d41614957d6fa535faadebbeca890b54b6977ac
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/425996
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
This reverts commit a89781215a.
Reason for revert: breakage on Windows
Original change's description:
> Add tests for matrix-vector conversions.
>
> GLSL supports casting vec4 into mat2 and vice versa, so SkSL should have
> equivalent support. Adding tests as a starting point.
>
> Change-Id: If8bcbf99afcec94d948d5da9e6205cb4a232af18
> Bug: skia:12067
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/425837
> Auto-Submit: John Stiles <johnstiles@google.com>
> Commit-Queue: Brian Osman <brianosman@google.com>
> Reviewed-by: Brian Osman <brianosman@google.com>
TBR=brianosman@google.com,ethannicholas@google.com,johnstiles@google.com
Change-Id: I2563041f538b1b20074385f1b61af5fc506ffad5
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:12067
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/426057
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
GLSL supports casting vec4 into mat2 and vice versa, so SkSL should have
equivalent support. Adding tests as a starting point.
Change-Id: If8bcbf99afcec94d948d5da9e6205cb4a232af18
Bug: skia:12067
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/425837
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Change-Id: Id2061ebe7873aa8b9480a2d8b0133c2fb79e79bb
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/424098
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Change-Id: I4066aafc5b6137bfaf38100ff237fd9833023f34
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/424097
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reland works around Adreno issue with this formulation of sk_Clockwise:
(sk_RTFlip.y < 0.0 ? !gl_FrontFacing : gl_FrontFacing)
and instead adds this to the top of the function:
bool sk_Clockwise = gl_FrontFacing;
if (sk_RTFlip.y < 0.0) {
sk_Clockwise = !sk_Clockwise;
}
Original description:
SkSL language features that are origin sensitive now use a uniform
to conditionally flip their result rather than generating different
code.
Previously we would insert a "rt height" uniform if sk_FragCoord needed
to be flipped. sk_FragCoord,y was implemented as "realFragCoord.y" or
"rtHeight - realFragCoord.y" depending on SkSL::ProgramSettings::fFlipY.
Now we instead use a two component vector rtFlip and sk_FragCoord.y is
always "rtFlip.x + rtFlip.y*realFragCoord.y". We configure rtFlip as
either (0, 1) or (rtHeight, -1). sk_Clockwise and dFdy simiarly use
rtFlip.y to emit code that always works with either origin.
Bug: skia:12037
Change-Id: I3a2ad6f5667eb4dcd823b939abd5698f89b58929
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/425178
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
This reverts commit 943108b0b2.
Reason for revert: clockwise GM bad on android.
Original change's description:
> Don't key progams/pipelines on origin.
>
> SkSL language features that are origin sensitive now use a uniform
> to conditionally flip their result rather than generating different
> code.
>
> Previously we would insert a "rt height" uniform if sk_FragCoord needed
> to be flipped. sk_FragCoord,y was implemented as "realFragCoord.y" or
> "rtHeight - realFragCoord.y" depending on SkSL::ProgramSettings::fFlipY.
>
> Now we instead use a two component vector rtFlip and sk_FragCoord.y is
> always "rtFlip.x + rtFlip.y*realFragCoord.y". We configure rtFlip as
> either (0, 1) or (rtHeight, -1). sk_Clockwise and dFdy simiarly use
> rtFlip.y to emit code that always works with either origin.
>
> Bug: skia:12037
>
> Change-Id: I7a09d0caac60a58d72b76645ff31bcabde4086b6
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/414796
> Commit-Queue: Brian Salomon <bsalomon@google.com>
> Reviewed-by: Greg Daniel <egdaniel@google.com>
> Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
TBR=egdaniel@google.com,bsalomon@google.com,ethannicholas@google.com
Change-Id: I91cc0d86be216f6c32e453a231de088c991be4b2
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:12037
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/425056
Reviewed-by: Brian Salomon <bsalomon@google.com>
Commit-Queue: Brian Salomon <bsalomon@google.com>
SkSL language features that are origin sensitive now use a uniform
to conditionally flip their result rather than generating different
code.
Previously we would insert a "rt height" uniform if sk_FragCoord needed
to be flipped. sk_FragCoord,y was implemented as "realFragCoord.y" or
"rtHeight - realFragCoord.y" depending on SkSL::ProgramSettings::fFlipY.
Now we instead use a two component vector rtFlip and sk_FragCoord.y is
always "rtFlip.x + rtFlip.y*realFragCoord.y". We configure rtFlip as
either (0, 1) or (rtHeight, -1). sk_Clockwise and dFdy simiarly use
rtFlip.y to emit code that always works with either origin.
Bug: skia:12037
Change-Id: I7a09d0caac60a58d72b76645ff31bcabde4086b6
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/414796
Commit-Queue: Brian Salomon <bsalomon@google.com>
Reviewed-by: Greg Daniel <egdaniel@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
This allows tests which draw this font on platforms which do not support
the COLRv1 table to draw something rather than nothing.
Change-Id: I024dd4809498d55f8c4c9f0de9e342fed5fb69ac
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/422757
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Ben Wagner <bungeman@google.com>
We've lived without this for a long time. It bloats sksl_gpu (inhibits
some inlining!), and if things go according to plan, we'll be removing
enum support from SkSL entirely soon.
Change-Id: If844bbe5fdae41df7930d5c8ea9b832f9dd1b922
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/419099
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Change-Id: I4280b5710dd8749ba766ba74d7a8886bc4e024bb
Bug: oss-fuzz:35124
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/417200
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Previously, when converting a program, the pipeline stage could assume
that an input to main() of type half4/float4 was always the input color.
This was a good assumption since the only possible inputs were
coordinates, or the input color.
This CL now recognizes that when a second float4 is passed to the main()
function, it should be a SK_DEST_COLOR_BUILTIN. This will let blend
functions pass in two colors.
ProgramToSkVM now takes a dest-color argument as well, but existing call
sites won't reference it (since they aren't for blend functions). I've
just passed the input-color a second time, since the value will never
actually be accessed.
Change-Id: I4214586bda605c6d287aa25b1b099e6ef5ba15a4
Bug: skia:12080
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/417261
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Runtime Blend effects always take two input colors--source and
destination--instead of one. This CL adds a new ProgramKind for blend
effects, a new program module (empty for now), and adds a test to
confirm that the signature for blend functions is checked. Currently
these are only accessible via skslc; there's no Runtime Effect API to
create one and the dest color isn't hooked up to anything.
Change-Id: I5272a811d2d76b878cfdf3429efa78c9c8b3fd97
Bug: skia:12080
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/416798
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
mix() has many overloads:
$genType mix($genType x, $genType y, $genType a);
$genType mix($genType x, $genType y, float a);
$genHType mix($genHType x, $genHType y, $genHType a);
$genHType mix($genHType x, $genHType y, half a);
$genType mix($genType x, $genType y, $genBType a);
$genHType mix($genHType x, $genHType y, $genBType a);
$genIType mix($genIType x, $genIType y, $genBType a);
$genBType mix($genBType x, $genBType y, $genBType a);
The top half were simple to implement via `evaluate_3_way_intrinsic`.
The bottom half--`x, y, $genBType`--required adding basic support into
`evaluate_n_way_intrinsic_of_type` for mixed argument types, since `x`
and `y` could be of any numeric type, but `a` is always boolean.
Fortunately, this didn't require major changes.
Change-Id: I015471f053c90d5a5c3ac67cc230d0f90950ff60
Bug: skia:12034
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/414443
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
$mat matrixCompMult($mat x, $mat y);
$hmat matrixCompMult($hmat x, $hmat y);
This required some minor changes to `evaluate_n_way_intrinsic_of_type`
to allow the inputs to be matrices instead of vectors. Fortunately, most
of the moving parts were already generic/flexible enough that this just
worked, but some explicit checks for `x.isVector` needed to become
`!x.isScalar()`, and `columns` needed to become `slotCount`.
Change-Id: I1e22ecad37a7e187a7171e1f590a720f07cf9832
Bug: skia:12034
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/414436
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
This was handled properly, but lacked a test.
Change-Id: I84adc7cb3d37ab85eef945c1e38fc43c6cd8aa01
Bug: skia:10932
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/414437
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Now that we know the minimum and maximum values of a given integer Type,
we can check for assigment statements or variable initial-values that
exceed those bounds and report it as an error. This check should work on
anything that can be optimized or folded down to an IntLiteral, but
isn't meant to be 100% exhaustive.
Change-Id: I4473b5b003e1b8e3385943ce60e303e95664e8ba
Bug: skia:10932
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/413437
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
We need to be careful to distinguish cross(vec3, vec3)--a real built-in
intrinsic--from cross(vec2, vec2)--an inline function in sksl_gpu, but
not actually a legitimate GLSL intrinsic.
Change-Id: I7e78c99dadfcbb637ae55a2503acfb7e591c932e
Bug: skia:12034
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/413440
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
The fuzzer found that it could overflow an int via a properly-crafted
constant-fold expression, leading to a UBSAN error. Constant-fold
expressions now guard against overflow (http://review.skia.org/413138)
and UBSAN is no longer triggered.
Change-Id: I07dba41e87bb9ceed37b84ec6b8922defbdc6550
Bug: oss-fuzz:32156, skia:12050
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/413836
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
This is one of the few compile-time constant functions with a variable
number of arguments, but it's otherwise pretty normal as intrinsics go.
$genType atan($genType y, $genType x);
$genHType atan($genHType y, $genHType x);
$genType atan($genType y_over_x);
$genHType atan($genHType y_over_x);
Change-Id: Ie852e10f37d73d53f69e806550872bc015f802d6
Bug: skia:12034
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/413439
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Expressions which would overflow/wrap the result type are now left
as-is.
Change-Id: I6a942f337e6e5761823f5c9dcd214fa58227a626
Bug: skia:10932, skia:12050
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/413138
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
$squareMat inverse($squareMat m);
$squareHMat inverse($squareHMat m);
Change-Id: I1a2b067dd276bb999107712c38d0124811b95e39
Bug: skia:12034
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/412937
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Change-Id: Ic6a3ce722c607d4d3c47d37e5749a01ae5fb193f
Bug: skia:12050
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/412668
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>
$genType refract($genType I, $genType N, float eta);
$genHType refract($genHType I, $genHType N, half eta);
The half form of refract was originally taking a `float eta` in our
headers, which seems wrong (and causes the DSL to break unless you add
casts). I've corrected the headers to use `half eta`.
Change-Id: I74b9ac330e0f7e99622d19cf7365aaa4cc910e57
Bug: skia:12034
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/412664
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Change-Id: I1630cf6ecab6a1a18a7318c247f8263d87e2cda9
Bug: skia:12034
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/412662
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
$genType reflect($genType I, $genType N);
$genHType reflect($genHType I, $genHType N);
Change-Id: I59889ad767829bf7e33838737d7b7f6d1cbc3ece
Bug: skia:12034
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/412777
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 implementation leans heavily into DSL. I also reworked the
`normalize` intrinsic to use more DSL to shrink the implementation.
$genType faceforward($genType N, $genType I, $genType Nref);
$genHType faceforward($genHType N, $genHType I, $genHType Nref);
Change-Id: I73ab11d3fe449d2f2c0ae0d745fc39824fc64771
Bug: skia:12034
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/412637
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 one is structured a bit differently; it gets to length() value,
then divides the input by its length using a bit of DSL. Since all the
inputs are constant, the constant-folder will do the right thing.
$genType normalize($genType x);
$genHType normalize($genHType x);
Change-Id: I51e5c65fa9e33738cbe253fcc97ee2160c48cfdd
Bug: skia:12034
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/412340
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
float length($genType x);
half length($genHType x);
Change-Id: I65b64fdba5f7bd53afba1d6f930217e3f1bd6f6e
Bug: skia:12034
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/412377
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Change-Id: I3427fbaf57787c3051db95ec5882c9292d7985cf
Bug: skia:12034
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/411312
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
In the majority of cases, a uniform is an equally good substitute, and
replacing `sqrt(N)` with `unknownInput` actually makes the test clearer.
Change-Id: I7bcb477571972d7aa2ce8c49b3674471f7310748
Bug: skia:12034
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/411306
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
This was originally designated for 2x2 matrices only, but this was not
right--all matrix comparisons actually need to be rewritten to fully
work around the bug.
Change-Id: I743d16a65bc55e93361a3dd8753653384583f063
Bug: skia:11308
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/411416
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 is allowed by GLSL, so we allow it too.
GLSL ES 1.0, Section 6.1: "The idiom “(void)” as a parameter list is
provided for convenience."
Change-Id: I551c505d3de518a75acd5e306f09f0f0767e43f2
Bug: skia:12025
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/411300
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This will allow us to rewrite `mat == mat` on Adreno 5xx/6xx GPUs when
running in GLSL.
Change-Id: I621e918a545a49b7ecb9c944ae59b1e7a7594bae
Bug: skia:11308
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/410996
Reviewed-by: Brian Osman <brianosman@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
