Normalize SIMD pipelines using TensorPrimitives_

[JimBobSquarePants]

Prerequisites

• I have written a descriptive pull-request title
• I have verified that there are no overlapping pull-requests open
• I have verified that I am following the existing coding patterns and practice as demonstrated in the repository. These follow strict Stylecop rules 👮.
• I have provided test coverage for my change (where applicable)

Description

This change introduces the internal TensorPrimitives_ compatibility type and uses the runtime's generic
operator-based SIMD model to normalize ImageSharp's span operations. The compatibility API follows
System.Numerics.Tensors.TensorPrimitives so it can be replaced by the runtime implementation when ImageSharp
moves to .NET 10.

The migration removes the replaced legacy SIMD implementations rather than retaining parallel pipelines. It also
uses portable Vector128, Vector256, and Vector512 APIs where the target framework exposes them, with
platform-specific intrinsics isolated behind compatibility helpers only where .NET 8 has no equivalent API.

The normalized implementation follows runtime main's widest-supported dispatch for every supported element type:
AVX-512 is selected when one complete Vector512<T> fits, and shorter inputs fall through to a width that fits.
SIMD layouts, masks, shuffles, overlap handling, scalar equivalence, and width-selection decisions are documented inline.

Performance

The tables below compare the implementation removed by this PR ("Before") with the retained normalized
implementation ("After"). Lower is better.

MemoryDiagnoser was enabled for every row containing an allocation result. 0 B / 0 B means BenchmarkDotNet
measured zero managed bytes allocated per operation for both the before and after implementations.

Tensor primitive migrations

Operation	Elements	Before	After	Ratio	Change
Clamp	processor workload	45.09 ns	38.47 ns	0.86	14.7% faster
ICC max	256	29.52 ns	13.26 ns	0.45	55.1% faster
ICC max	2,048	230.41 ns	96.90 ns	0.42	57.9% faster
ICC multiply	256	10.023 ns	6.549 ns	0.65	34.7% faster
ICC multiply	2,048	82.854 ns	63.982 ns	0.77	22.8% faster
JPEG multiply	256	9.853 ns	7.835 ns	0.80	20.5% faster
JPEG multiply	2,048	64.993 ns	45.332 ns	0.70	30.3% faster
Single add	256	12.31 ns	10.09 ns	0.82	18.0% faster
Single add	2,048	102.45 ns	58.72 ns	0.57	42.7% faster
UInt32 add	256	11.430 ns	9.954 ns	0.87	12.9% faster
UInt32 add	2,048	103.460 ns	73.943 ns	0.71	28.5% faster
Byte add	64	1.637 ns	2.165 ns	1.32	32.3% slower
Byte add	256	3.644 ns	3.727 ns	1.02	2.3% slower
Byte add	2,048	24.865 ns	24.601 ns	0.99	1.1% faster
Byte add	8,192	122.624 ns	103.183 ns	0.84	15.9% faster
Resize kernel normalization	7	2.807 ns	2.827 ns	1.01	0.7% slower
Resize kernel normalization	19	3.147 ns	5.003 ns	1.59	59.0% slower
Resize kernel normalization	32	3.066 ns	3.603 ns	1.18	17.5% slower
Resize kernel normalization	65	3.818 ns	5.327 ns	1.40	39.5% slower
Resize kernel normalization	161	9.283 ns	7.869 ns	0.85	15.2% faster
Resize kernel normalization	257	15.341 ns	9.234 ns	0.60	39.8% faster
Resize kernel normalization	2,048	125.141 ns	63.633 ns	0.51	49.2% faster

JPEG packing pipelines

Pipeline	Elements	Before	After	Ratio	Change	Managed allocation
Invert + normalize + interleave 4	128	103.39 ns	50.34 ns	0.49	51.3% faster	0 B / 0 B
Invert + normalize + interleave 4	1,024	862.64 ns	343.56 ns	0.40	60.2% faster	0 B / 0 B
Invert + normalize + interleave 4	4,096	3,561.36 ns	1,313.44 ns	0.37	63.1% faster	0 B / 0 B
Normalize + interleave 3	128	78.25 ns	43.30 ns	0.56	44.7% faster	0 B / 0 B
Normalize + interleave 3	1,024	603.88 ns	307.28 ns	0.51	49.1% faster	0 B / 0 B
Normalize + interleave 3	4,096	2,332.94 ns	1,219.93 ns	0.53	47.7% faster	0 B / 0 B
Normalize + interleave 4	128	98.87 ns	39.52 ns	0.40	60.0% faster	0 B / 0 B
Normalize + interleave 4	1,024	716.14 ns	283.87 ns	0.40	60.4% faster	0 B / 0 B
Normalize + interleave 4	4,096	2,959.24 ns	1,107.54 ns	0.38	62.6% faster	0 B / 0 B
Deinterleave 3	128	80.09 ns	40.41 ns	0.51	49.5% faster	0 B / 0 B
Deinterleave 3	1,024	592.62 ns	292.62 ns	0.50	50.6% faster	0 B / 0 B
Deinterleave 3	4,096	2,304.31 ns	1,128.42 ns	0.49	51.0% faster	0 B / 0 B

Additional migrations found during the library-wide audit

Operation	Elements	Before	After	Ratio	Change	Managed allocation
Histogram scalar add	256	97.631 ns	8.464 ns	0.09	91.3% faster (11.54x)	0 B / 0 B
Histogram scalar add	65,536	24.430 us	1.640 us	0.07	93.3% faster (14.90x)	0 B / 0 B
Sharpen-kernel negate	7	3.901 ns	2.489 ns	0.64	36.2% faster	0 B / 0 B
Sharpen-kernel negate	19	9.106 ns	2.895 ns	0.32	68.2% faster	0 B / 0 B
Sharpen-kernel negate	65	29.919 ns	4.502 ns	0.15	85.0% faster	0 B / 0 B
Sharpen-kernel negate	161	83.357 ns	4.721 ns	0.06	94.3% faster	0 B / 0 B
Gaussian normalization	7	3.677 ns	2.898 ns	0.79	21.2% faster	0 B / 0 B
Gaussian normalization	19	10.079 ns	5.222 ns	0.52	48.2% faster	0 B / 0 B
Gaussian normalization	65	32.780 ns	5.239 ns	0.16	84.0% faster	0 B / 0 B
Gaussian normalization	161	88.170 ns	7.029 ns	0.08	92.0% faster	0 B / 0 B
Bokeh complex-kernel scale	17	11.286 ns	8.420 ns	0.75	25.4% faster	0 B / 0 B
Bokeh complex-kernel scale	33	21.040 ns	3.732 ns	0.18	82.3% faster	0 B / 0 B
Bokeh complex-kernel scale	65	41.041 ns	4.577 ns	0.11	88.8% faster	0 B / 0 B
Bokeh complex-kernel scale	161	110.389 ns	9.590 ns	0.09	91.3% faster	0 B / 0 B
ICC byte LUT normalization	256	145.188 ns	9.997 ns	0.07	93.1% faster (14.52x)	0 B / 0 B
ICC UInt16 LUT normalization	256	143.157 ns	11.771 ns	0.08	91.8% faster (12.16x)	0 B / 0 B

Byte-add production callers

The PNG rows compare the exact AVX2 loop removed from UpFilter.Decode with the production
TensorPrimitives_.Add<byte> call in the same benchmark run. The runtime-shaped implementation has a
small-buffer dispatch cost, crosses over by 2,048 payload bytes, and benefits increasingly from AVX-512 on
larger scanlines.

Pipeline	Payload bytes	Before	After	Ratio	Change	Managed allocation
PNG Up-filter decode	256	3.939 ns	6.779 ns	1.72	72.1% slower	0 B / 0 B
PNG Up-filter decode	2,048	20.861 ns	18.692 ns	0.90	10.4% faster	0 B / 0 B
PNG Up-filter decode	8,192	93.117 ns	65.930 ns	0.71	29.2% faster	0 B / 0 B

Binary size

Release net8.0 artifacts were built from main and this PR using the same .NET SDK 10.0.302 installation and
project settings.

Artifact	main	This PR	Reduction
Managed assembly (SixLabors.ImageSharp.dll)	4,100,608 B	2,433,024 B	1,667,584 B (40.67%)
Portable symbols (SixLabors.ImageSharp.pdb)	2,033,448 B	1,047,496 B	985,952 B (48.49%)
Packed NuGet package	1,316,135 B	1,278,182 B	37,953 B (2.88%)

The package reduction is smaller than the assembly reduction because the removed generated IL is highly
repetitive and therefore compresses efficiently inside the NuGet package.

NativeAOT binary-size impact

Version	Native executable	Change
main	9,475,584 B	—
This PR	9,652,736 B	+177,152 B (1.87%)

Sizoscope attributes the remaining increase predominantly to statically specialized SIMD/operator code. Further material sharing would require runtime dispatch, reduced inlining, or eager initialization, potentially harming performance.

Both executables completed the representative processing and codec workload successfully with identical output.

[Copilot]

Pull request overview

This PR normalizes several SIMD-heavy pipelines around a new internal TensorPrimitives_ compatibility API (mirroring System.Numerics.Tensors.TensorPrimitives) and removes older, parallel SIMD implementations in favor of a single runtime-shaped dispatch model. It also adds targeted tests and benchmarks to validate correctness across scalar/SIMD width boundaries and to track performance impacts.

Changes:

• Introduce and adopt TensorPrimitives_-style span operations for common add/multiply/divide/negate/max workloads across codecs and processing helpers.
• Refactor/normalize affine Vector4 conversion traversal and pixel blender traversal using operator-based SIMD dispatch, and remove legacy JPEG converter SIMD class hierarchy.
• Add tests/benchmarks covering SIMD boundary behavior for PNG filters, ICC LUT normalization, pixel blending, convolution kernels, and Vector4 affine transforms.

Reviewed changes

Copilot reviewed 109 out of 111 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
tests/ImageSharp.Tests/Processing/Processors/Convolution/ConvolutionProcessorHelpersTest.cs	Adds regression test for Gaussian sharpen kernel correctness across sizes.
tests/ImageSharp.Tests/PixelFormats/Vector4ConvertersTests.cs	New tests validating affine Vector4 conversion traversal across hardware-width boundaries.
tests/ImageSharp.Tests/PixelFormats/PixelBlenderTests.cs	Strengthens blender tests to cover multiple bulk overloads, widths, and coverage/amount variants.
tests/ImageSharp.Tests/Metadata/Profiles/ICC/Various/IccLutTests.cs	New tests asserting bit-exact ICC LUT normalization for byte/ushort inputs (including exhaustive ushort).
tests/ImageSharp.Tests/Formats/Png/PngEncoderFilterTests.cs	Adds encoder/reference comparisons across SIMD boundaries and pixel strides; minor formatting tidy.
tests/ImageSharp.Benchmarks/ImageSharp.Benchmarks.csproj	Adds conditional net10 dependency on System.Numerics.Tensors for benchmarks.
tests/ImageSharp.Benchmarks/General/PixelConversion/Vector4AffineTransform.cs	New benchmark for operator-driven affine Vector4 transforms.
tests/ImageSharp.Benchmarks/Codecs/Png/PngFilterEncode.cs	New benchmark for shared PNG filter encode traversal.
tests/ImageSharp.Benchmarks/Codecs/Jpeg/ColorConversion/YccKColorConverter.cs	Updates benchmark to use adaptive converter selection rather than per-width implementations.
tests/ImageSharp.Benchmarks/Codecs/Jpeg/ColorConversion/YCbCrColorConversion.cs	Updates benchmark to use adaptive converter selection rather than per-width implementations.
tests/ImageSharp.Benchmarks/Codecs/Jpeg/ColorConversion/RgbColorConversion.cs	Updates benchmark to use adaptive converter selection rather than per-width implementations.
tests/ImageSharp.Benchmarks/Codecs/Jpeg/ColorConversion/JpegColorPackingScalar.cs	New preserved scalar JPEG packing loops for before/after comparison.
tests/ImageSharp.Benchmarks/Codecs/Jpeg/ColorConversion/JpegColorPacking.cs	New benchmark comparing scalar packing vs SIMD transpose implementations.
tests/ImageSharp.Benchmarks/Codecs/Jpeg/ColorConversion/GrayscaleColorConversion.cs	Updates benchmark to use adaptive converter selection rather than per-width implementations.
tests/ImageSharp.Benchmarks/Codecs/Jpeg/ColorConversion/ColorConversionBenchmark.cs	Fixes benchmark setup to actually populate allocated buffers.
tests/ImageSharp.Benchmarks/Codecs/Jpeg/ColorConversion/CmykColorConversion.cs	Updates benchmark to use adaptive converter selection rather than per-width implementations.
tests/ImageSharp.Benchmarks/Bulk/ShuffleByte4Channel.cs	Updates shuffle benchmark call to match new shuffle API shape.
tests/ImageSharp.Benchmarks/Bulk/Shuffle4Slice3Channel.cs	Updates shuffle benchmark to generic shuffle operators, removing control instance.
tests/ImageSharp.Benchmarks/Bulk/Shuffle3Channel.cs	Updates shuffle benchmark to generic shuffle operators, removing control instance.
tests/ImageSharp.Benchmarks/Bulk/Pad3Shuffle4Channel.cs	Updates pad+shuffle benchmark to generic shuffle operators, removing control instance.
src/ImageSharp/Processing/Processors/Normalization/HistogramEqualizationProcessor{TPixel}.cs	Replaces scalar histogram add loop with TensorPrimitives_ add.
src/ImageSharp/Processing/Processors/Convolution/Parameters/BokehBlurKernelDataProvider.cs	Normalizes complex-kernel scaling via TensorPrimitives_ over a flattened float span.
src/ImageSharp/Processing/Processors/Convolution/ConvolutionProcessorHelpers.cs	Replaces per-element kernel normalization/inversion loops with TensorPrimitives_ operations.
src/ImageSharp/PixelFormats/Utils/Vector4Converters.AffineOperators.cs	Adds stateful operator definitions for affine Vector4 transforms across register widths.
src/ImageSharp/PixelFormats/Utils/Vector4Converters.Affine.cs	Refactors affine traversal into a single generic operator-driven Apply loop and makes APIs public.
src/ImageSharp/PixelFormats/PixelBlenders/IPixelBlenderOperator.cs	New static-abstract operator contract for shared blender traversal across SIMD widths.
src/ImageSharp/PixelFormats/PixelBlenders/AssociatedAlphaPixelBlender{TPixel,TOperator}.cs	Refactors associated-alpha blender to use the new operator-based blender base and clamps amount.
src/ImageSharp/Metadata/Profiles/ICC/Various/IccLut.cs	Routes LUT normalization through a dedicated normalizer helper.
src/ImageSharp/Formats/Webp/Lossless/Vp8LHistogram.cs	Replaces AVX2-only add loop with TensorPrimitives_ add over sliced spans and removes old helper.
src/ImageSharp/Formats/Webp/AlphaDecoder.cs	Replaces SIMD/scalar branches with TensorPrimitives_ byte add for vertical unfiltering.
src/ImageSharp/Formats/Png/Filters/SubFilter.cs	Replaces bespoke encode loop with shared PngFilterEncoder traversal.
src/ImageSharp/Formats/Png/Filters/AverageFilter.cs	Replaces bespoke encode loop with shared PngFilterEncoder traversal.
src/ImageSharp/Formats/Jpeg/Components/Encoder/ComponentProcessor.cs	Replaces manual SIMD/scalar sum/multiply loops with TensorPrimitives_ add/multiply.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverterVector512.cs	Removes legacy Vector512 converter base class.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverterVector256.cs	Removes legacy Vector256 converter base class.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverterVector128.cs	Removes legacy Vector128 converter base class.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverterVector.cs	Removes legacy System.Numerics.Vector converter base class.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverterScalar.cs	Removes legacy scalar converter base class.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.YccKVector512.cs	Removes legacy YccK Vector512 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.YccKVector256.cs	Removes legacy YccK Vector256 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.YccKVector128.cs	Removes legacy YccK Vector128 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.YccKScalar.cs	Removes legacy YccK scalar implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.YCbCrVector512.cs	Removes legacy YCbCr Vector512 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.YCbCrVector256.cs	Removes legacy YCbCr Vector256 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.YCbCrVector128.cs	Removes legacy YCbCr Vector128 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.YCbCrScalar.cs	Removes legacy YCbCr scalar implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.TiffYccKVector256.cs	Removes legacy TIFF YccK Vector256 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.TiffYccKVector128.cs	Removes legacy TIFF YccK Vector128 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.TiffYccKScalar.cs	Removes legacy TIFF YccK scalar implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.TiffCmykVector512.cs	Removes legacy TIFF CMYK Vector512 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.TiffCmykVector256.cs	Removes legacy TIFF CMYK Vector256 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.TiffCmykVector128.cs	Removes legacy TIFF CMYK Vector128 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.TiffCmykScalar.cs	Removes legacy TIFF CMYK scalar implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.RgbVector512.cs	Removes legacy RGB Vector512 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.RgbVector256.cs	Removes legacy RGB Vector256 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.RgbVector128.cs	Removes legacy RGB Vector128 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.RgbScalar.cs	Removes legacy RGB scalar implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.RgbOperator.cs	Adds operator-based RGB converter implementation (scalar + SIMD lanes + ICC path).
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.GrayScaleVector512.cs	Removes legacy grayscale Vector512 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.GrayScaleVector256.cs	Removes legacy grayscale Vector256 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.GrayScaleVector128.cs	Removes legacy grayscale Vector128 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.GrayScaleScalar.cs	Removes legacy grayscale scalar implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.CmykVector512.cs	Removes legacy CMYK Vector512 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.CmykVector256.cs	Removes legacy CMYK Vector256 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.CmykVector128.cs	Removes legacy CMYK Vector128 implementation.
src/ImageSharp/Formats/Jpeg/Components/ColorConverters/JpegColorConverter.CmykScalar.cs	Removes legacy CMYK scalar implementation.
src/ImageSharp/Common/Helpers/Vector512Utilities.cs	Adds portable Vector512<byte> subtract-saturate helper mirroring runtime behavior.
src/ImageSharp/Common/Helpers/Vector256Utilities.cs	Aligns subtract-saturate fallback behavior/comments with runtime portable approach.
src/ImageSharp/Common/Helpers/Vector128Utilities.cs	Simplifies unsigned byte subtract-saturate fallback to a min/sub identity.
src/ImageSharp/Common/Helpers/TensorPrimitives_.Multiply.cs	Adds generic operator-based multiply traversal (scalar + 128/256/512).
src/ImageSharp/Common/Helpers/TensorPrimitives_.Divide.cs	Adds divide traversal with restricted vectorization rules to match runtime capabilities.
src/ImageSharp/Common/Helpers/TensorPrimitives_.Add.cs	Adds generic operator-based add traversal (span+span and span+scalar).
src/ImageSharp/Common/Helpers/Shuffle/IShuffle3.cs	Replaces stateful shuffle control with stateless operator (ZYXShuffle3) + intrinsic helper.
src/ImageSharp/Common/Helpers/Shuffle/IComponentShuffle.cs	Refactors shuffle contract to static-abstract scalar/vector operators.
src/ImageSharp/ColorProfiles/ColorProfileConverterExtensionsIcc.cs	Replaces ad-hoc SIMD/scalar loops with TensorPrimitives_ max/multiply over flattened Vector4 spans.

[Copilot]

Pull request overview

Copilot reviewed 109 out of 111 changed files in this pull request and generated no new comments.

Comments suppressed due to low confidence (1)

tests/ImageSharp.Benchmarks/Codecs/Png/PngFilterEncode.cs:87

• AverageFilter.Encode expects uint bytesPerPixel, but this benchmark passes BytesPerPixel as an int, which will not compile once the Encode signature is updated. Cast the constant to uint to match the API (consistent with other call sites in the PR).