[JitDiff X64] [tannergooding] Update Matrix4x4 paths to use the new xplat in ...

[MihuBot]

Job completed in 21 minutes 46 seconds (remote runner delay: 1 minute 8 seconds).
dotnet/runtime#130192

Main commit: dotnet/runtime@58dfec7
PR commit: tannergooding/runtime@3149c66

Diffs

Diffs

Found 89 files with textual diffs.

Summary of Code Size diffs:
(Lower is better)

Total bytes of base: 158040784
Total bytes of diff: 158049953
Total bytes of delta: 9169 (0.01 % of base)
Total relative delta: 108.85
    diff is a regression.
    relative diff is a regression.


Total byte diff includes -1796 bytes from reconciling methods
	Base had    3 unique methods,     1937 unique bytes
	Diff had    1 unique methods,      141 unique bytes

Top file regressions (bytes):
        8732 : SixLabors.ImageSharp.dasm (0.39 % of base)
         437 : System.Private.CoreLib.dasm (0.01 % of base)

2 total files with Code Size differences (0 improved, 2 regressed), 352 unchanged.

Top method regressions (bytes):
         690 (5,307.69 % of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4+Impl:Invert(byref,byref):bool (FullOpts)
         677 (2,507.41 % of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4:Invert(System.Numerics.Matrix4x4,byref):bool (FullOpts)
         640 (129.29 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.ColorProfiles.Icc.Calculators.ColorTrcCalculator:.ctor(SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,bool):this (FullOpts)
         602 (132.60 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.ColorProfiles.ColorConversionOptions:.ctor():this (FullOpts)
         581 (124.95 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.ColorProfiles.Rgb:FromProfileConnectionSpace(SixLabors.ImageSharp.ColorProfiles.ColorConversionOptions,System.ReadOnlySpan`1[SixLabors.ImageSharp.ColorProfiles.CieXyz],System.Span`1[SixLabors.ImageSharp.ColorProfiles.Rgb]) (FullOpts)
         579 (293.91 % of base) : System.Private.CoreLib.dasm - System.Numerics.Plane:Transform(System.Numerics.Plane,System.Numerics.Matrix4x4):System.Numerics.Plane (FullOpts)
         574 (167.84 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.ColorProfiles.Rgb:FromProfileConnectingSpace(SixLabors.ImageSharp.ColorProfiles.ColorConversionOptions,byref):SixLabors.ImageSharp.ColorProfiles.Rgb (FullOpts)
         552 (110.40 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.ColorProfiles.Rgb:GetRgbToCieXyzMatrix(SixLabors.ImageSharp.ColorProfiles.WorkingSpaces.RgbWorkingSpace):System.Numerics.Matrix4x4 (FullOpts)
         534 (111.02 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.ColorProfiles.VonKriesChromaticAdaptation:Transform(byref,System.ValueTuple`2[SixLabors.ImageSharp.ColorProfiles.CieXyz,SixLabors.ImageSharp.ColorProfiles.CieXyz],System.Numerics.Matrix4x4):SixLabors.ImageSharp.ColorProfiles.CieXyz (FullOpts)
         510 (82.26 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.ColorProfiles.VonKriesChromaticAdaptation:Transform(System.ReadOnlySpan`1[SixLabors.ImageSharp.ColorProfiles.CieXyz],System.Span`1[SixLabors.ImageSharp.ColorProfiles.CieXyz],System.ValueTuple`2[SixLabors.ImageSharp.ColorProfiles.CieXyz,SixLabors.ImageSharp.ColorProfiles.CieXyz],System.Numerics.Matrix4x4) (FullOpts)
         214 (9.15 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.TransformExtensions:Transform(SixLabors.ImageSharp.Processing.IImageProcessingContext,SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder,SixLabors.ImageSharp.Processing.Processors.Transforms.IResampler):SixLabors.ImageSharp.Processing.IImageProcessingContext (FullOpts)
         211 (9.09 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.TransformExtensions:Transform(SixLabors.ImageSharp.Processing.IImageProcessingContext,SixLabors.ImageSharp.Rectangle,SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder,SixLabors.ImageSharp.Processing.Processors.Transforms.IResampler):SixLabors.ImageSharp.Processing.IImageProcessingContext (FullOpts)
         207 (57.02 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:CheckDegenerate(System.Numerics.Matrix4x4) (FullOpts)
         177 (31.55 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendMatrix(System.Numerics.Matrix4x4):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         177 (28.50 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendTranslation(SixLabors.ImageSharp.PointF):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         177 (29.26 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendTranslation(System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         177 (24.28 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependMatrix(System.Numerics.Matrix4x4):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         177 (22.18 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependTranslation(SixLabors.ImageSharp.PointF):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         177 (22.63 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependTranslation(System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (30.37 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.Processors.Transforms.ProjectiveTransformProcessor:.ctor(System.Numerics.Matrix4x4,SixLabors.ImageSharp.Processing.Processors.Transforms.IResampler,SixLabors.ImageSharp.Size):this (FullOpts)
         174 (22.86 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendRotationDegrees(float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (23.11 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendRotationRadians(float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (27.36 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendScale(float):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (25.97 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendScale(SixLabors.ImageSharp.SizeF):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (26.61 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendScale(System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (18.55 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependRotationDegrees(float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (18.71 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependRotationRadians(float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (21.22 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependScale(float):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (20.37 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependScale(SixLabors.ImageSharp.SizeF):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (20.76 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependScale(System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (17.77 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependSkewDegrees(float,float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (18.26 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependSkewRadians(float,float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (27.40 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.TransformExtensions:Transform(SixLabors.ImageSharp.Processing.IImageProcessingContext,SixLabors.ImageSharp.Rectangle,System.Numerics.Matrix4x4,SixLabors.ImageSharp.Size,SixLabors.ImageSharp.Processing.Processors.Transforms.IResampler):SixLabors.ImageSharp.Processing.IImageProcessingContext (FullOpts)
         172 (21.58 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendSkewDegrees(float,float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         172 (22.31 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendSkewRadians(float,float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         161 (1,238.46 % of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4:GetDeterminant():float:this (FullOpts)
         157 (64.08 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.Processors.Transforms.TransformUtilities:IsDegenerate(System.Numerics.Matrix4x4):bool (FullOpts)
         141 (Infinity of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4+Impl:Shuffle2(System.Runtime.Intrinsics.Vector128`1[float],System.Runtime.Intrinsics.Vector128`1[float],byte,byte,byte,byte):System.Runtime.Intrinsics.Vector128`1[float] (FullOpts) (0 base, 1 diff methods)
         126 (8.97 % of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4+Impl:Decompose(byref,byref,byref,byref):bool (FullOpts)
         108 (9.10 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:BuildMatrix(SixLabors.ImageSharp.Rectangle):System.Numerics.Matrix4x4:this (FullOpts)

Top method improvements (bytes):
       -1187 (-100.00 % of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4+Impl:<Invert>g__SoftwareFallback|53_1(byref,byref):bool (FullOpts) (1 base, 0 diff methods)
        -707 (-100.00 % of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4+Impl:<Invert>g__SseImpl|53_0(byref,byref):bool (FullOpts) (1 base, 0 diff methods)
         -43 (-100.00 % of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4+Impl:ThrowPlatformNotSupportedException() (FullOpts) (1 base, 0 diff methods)

Top method regressions (percentages):
         141 (Infinity of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4+Impl:Shuffle2(System.Runtime.Intrinsics.Vector128`1[float],System.Runtime.Intrinsics.Vector128`1[float],byte,byte,byte,byte):System.Runtime.Intrinsics.Vector128`1[float] (FullOpts) (0 base, 1 diff methods)
         690 (5,307.69 % of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4+Impl:Invert(byref,byref):bool (FullOpts)
         677 (2,507.41 % of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4:Invert(System.Numerics.Matrix4x4,byref):bool (FullOpts)
         161 (1,238.46 % of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4:GetDeterminant():float:this (FullOpts)
         579 (293.91 % of base) : System.Private.CoreLib.dasm - System.Numerics.Plane:Transform(System.Numerics.Plane,System.Numerics.Matrix4x4):System.Numerics.Plane (FullOpts)
         574 (167.84 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.ColorProfiles.Rgb:FromProfileConnectingSpace(SixLabors.ImageSharp.ColorProfiles.ColorConversionOptions,byref):SixLabors.ImageSharp.ColorProfiles.Rgb (FullOpts)
         602 (132.60 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.ColorProfiles.ColorConversionOptions:.ctor():this (FullOpts)
         640 (129.29 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.ColorProfiles.Icc.Calculators.ColorTrcCalculator:.ctor(SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,bool):this (FullOpts)
         581 (124.95 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.ColorProfiles.Rgb:FromProfileConnectionSpace(SixLabors.ImageSharp.ColorProfiles.ColorConversionOptions,System.ReadOnlySpan`1[SixLabors.ImageSharp.ColorProfiles.CieXyz],System.Span`1[SixLabors.ImageSharp.ColorProfiles.Rgb]) (FullOpts)
         534 (111.02 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.ColorProfiles.VonKriesChromaticAdaptation:Transform(byref,System.ValueTuple`2[SixLabors.ImageSharp.ColorProfiles.CieXyz,SixLabors.ImageSharp.ColorProfiles.CieXyz],System.Numerics.Matrix4x4):SixLabors.ImageSharp.ColorProfiles.CieXyz (FullOpts)
         552 (110.40 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.ColorProfiles.Rgb:GetRgbToCieXyzMatrix(SixLabors.ImageSharp.ColorProfiles.WorkingSpaces.RgbWorkingSpace):System.Numerics.Matrix4x4 (FullOpts)
         510 (82.26 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.ColorProfiles.VonKriesChromaticAdaptation:Transform(System.ReadOnlySpan`1[SixLabors.ImageSharp.ColorProfiles.CieXyz],System.Span`1[SixLabors.ImageSharp.ColorProfiles.CieXyz],System.ValueTuple`2[SixLabors.ImageSharp.ColorProfiles.CieXyz,SixLabors.ImageSharp.ColorProfiles.CieXyz],System.Numerics.Matrix4x4) (FullOpts)
         157 (64.08 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.Processors.Transforms.TransformUtilities:IsDegenerate(System.Numerics.Matrix4x4):bool (FullOpts)
         207 (57.02 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:CheckDegenerate(System.Numerics.Matrix4x4) (FullOpts)
         177 (31.55 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendMatrix(System.Numerics.Matrix4x4):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (30.37 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.Processors.Transforms.ProjectiveTransformProcessor:.ctor(System.Numerics.Matrix4x4,SixLabors.ImageSharp.Processing.Processors.Transforms.IResampler,SixLabors.ImageSharp.Size):this (FullOpts)
         177 (29.26 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendTranslation(System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         177 (28.50 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendTranslation(SixLabors.ImageSharp.PointF):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (27.40 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.TransformExtensions:Transform(SixLabors.ImageSharp.Processing.IImageProcessingContext,SixLabors.ImageSharp.Rectangle,System.Numerics.Matrix4x4,SixLabors.ImageSharp.Size,SixLabors.ImageSharp.Processing.Processors.Transforms.IResampler):SixLabors.ImageSharp.Processing.IImageProcessingContext (FullOpts)
         174 (27.36 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendScale(float):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (26.61 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendScale(System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (25.97 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendScale(SixLabors.ImageSharp.SizeF):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         177 (24.28 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependMatrix(System.Numerics.Matrix4x4):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (23.11 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendRotationRadians(float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (22.86 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendRotationDegrees(float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         177 (22.63 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependTranslation(System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         172 (22.31 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendSkewRadians(float,float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         177 (22.18 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependTranslation(SixLabors.ImageSharp.PointF):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         172 (21.58 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:AppendSkewDegrees(float,float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (21.22 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependScale(float):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (20.76 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependScale(System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (20.37 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependScale(SixLabors.ImageSharp.SizeF):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (18.71 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependRotationRadians(float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (18.55 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependRotationDegrees(float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (18.26 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependSkewRadians(float,float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         174 (17.77 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:PrependSkewDegrees(float,float,System.Numerics.Vector2):SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:this (FullOpts)
         214 (9.15 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.TransformExtensions:Transform(SixLabors.ImageSharp.Processing.IImageProcessingContext,SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder,SixLabors.ImageSharp.Processing.Processors.Transforms.IResampler):SixLabors.ImageSharp.Processing.IImageProcessingContext (FullOpts)
         108 (9.10 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder:BuildMatrix(SixLabors.ImageSharp.Rectangle):System.Numerics.Matrix4x4:this (FullOpts)
         211 (9.09 % of base) : SixLabors.ImageSharp.dasm - SixLabors.ImageSharp.Processing.TransformExtensions:Transform(SixLabors.ImageSharp.Processing.IImageProcessingContext,SixLabors.ImageSharp.Rectangle,SixLabors.ImageSharp.Processing.ProjectiveTransformBuilder,SixLabors.ImageSharp.Processing.Processors.Transforms.IResampler):SixLabors.ImageSharp.Processing.IImageProcessingContext (FullOpts)
         126 (8.97 % of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4+Impl:Decompose(byref,byref,byref,byref):bool (FullOpts)

Top method improvements (percentages):
       -1187 (-100.00 % of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4+Impl:<Invert>g__SoftwareFallback|53_1(byref,byref):bool (FullOpts) (1 base, 0 diff methods)
        -707 (-100.00 % of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4+Impl:<Invert>g__SseImpl|53_0(byref,byref):bool (FullOpts) (1 base, 0 diff methods)
         -43 (-100.00 % of base) : System.Private.CoreLib.dasm - System.Numerics.Matrix4x4+Impl:ThrowPlatformNotSupportedException() (FullOpts) (1 base, 0 diff methods)

43 total methods with Code Size differences (3 improved, 40 regressed), 815783 unchanged.

--------------------------------------------------------------------------------

Artifacts:

• PackageList.json (33 KB)
• diff-frameworks.txt (21 KB)
• LongDiffsRegressions.md (700 KB)
• ShortDiffsRegressions.md (55 KB)
• jit-diffs-pr.7z (141 MB)
• jit-diffs-main.7z (141 MB)

[MihuBot]

Top method regressions

690 (5,307.69 % of base) - System.Numerics.Matrix4x4+Impl:Invert(byref,byref):bool

 ; Assembly listing for method System.Numerics.Matrix4x4+Impl:Invert(byref,byref):bool (FullOpts)
 ; Emitting BLENDED_CODE for x64 + VEX + EVEX on Unix
 ; FullOpts code
 ; optimized code
-; rsp based frame
-; fully interruptible
+; rbp based frame
+; partially interruptible
 ; No PGO data
+; 0 inlinees with PGO data; 3 single block inlinees; 20 inlinees without PGO data
 ; Final local variable assignments
 ;
-;  V00 arg0         [V00,T00] (  3,  3   )   byref  ->  rdi         single-def
-;  V01 arg1         [V01,T01] (  3,  3   )   byref  ->  rsi         single-def
-;# V02 OutArgs      [V02    ] (  1,  1   )  struct ( 0) [rsp+0x00]  do-not-enreg[XS] addr-exposed "OutgoingArgSpace" <Empty>
+;  V00 arg0         [V00,T01] (  6,  6   )   byref  ->  rdi         single-def
+;  V01 arg1         [V01,T00] ( 10,  6   )   byref  ->  rsi         single-def
+;  V02 loc0         [V02,T06] ( 12, 12   )  simd16  ->  mm0         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V03 loc1         [V03,T09] ( 11, 11   )  simd16  ->  mm1         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V04 loc2         [V04,T10] ( 11, 11   )  simd16  ->  mm2         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V05 loc3         [V05,T11] ( 11, 11   )  simd16  ->  mm3         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V06 loc4         [V06,T25] (  3,  3   )  simd16  ->  mm4         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V07 loc5         [V07,T26] (  3,  3   )  simd16  ->  mm5         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V08 loc6         [V08,T27] (  3,  3   )  simd16  ->  mm3         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V09 loc7         [V09,T03] ( 13, 13   )  simd16  ->  mm4         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V10 loc8         [V10,T18] ( 10, 10   )  simd16  ->  registers   <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V11 loc9         [V11,T04] ( 13, 13   )  simd16  ->  registers   <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V12 loc10        [V12,T02] ( 16, 16   )  simd16  ->  registers   <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V13 loc11        [V13,T05] ( 13, 13   )  simd16  ->  registers   <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V14 loc12        [V14,T19] ( 10, 10   )  simd16  ->  registers   <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V15 loc13        [V15,T12] ( 11, 11   )  simd16  ->  mm5         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V16 loc14        [V16,T13] ( 11, 11   )  simd16  ->  mm7         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V17 loc15        [V17,T14] ( 11, 11   )  simd16  ->  mm9         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V18 loc16        [V18,T07] ( 12, 12   )  simd16  ->  registers   <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V19 loc17        [V19,T20] (  9,  9   )  simd16  ->  registers   <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V20 loc18        [V20,T08] ( 12, 11.50)  simd16  ->  mm16         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V21 loc19        [V21,T15] ( 11, 10.50)  simd16  ->  mm17         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V22 loc20        [V22,T16] ( 11, 10.50)  simd16  ->  mm18         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V23 loc21        [V23,T17] ( 11, 10.50)  simd16  ->  mm14         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V24 loc22        [V24,T31] (  2,  2   )  simd16  ->  mm1         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V25 loc23        [V25,T32] (  2,  2   )  simd16  ->  mm2         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V26 loc24        [V26,T33] (  2,  2   )  simd16  ->  mm3         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V27 loc25        [V27,T34] (  2,  2   )  simd16  ->  mm4         <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V28 loc26        [V28,T30] (  3,  2.50)   float  ->  mm0        
+;  V29 loc27        [V29,T28] (  5,  2.50)  simd16  ->  mm0         <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V30 loc28        [V30    ] (  0,  0   )  simd16  ->  zero-ref    <System.Numerics.Vector4>
+;# V31 OutArgs      [V31    ] (  1,  1   )  struct ( 0) [rsp+0x00]  do-not-enreg[XS] addr-exposed "OutgoingArgSpace" <Empty>
+;  V32 tmp1         [V32,T22] (  3,  6   )  simd16  ->  mm1         "dup spill"
+;* V33 tmp2         [V33    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V34 tmp3         [V34    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V35 tmp4         [V35    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V36 tmp5         [V36    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V37 tmp6         [V37    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V38 tmp7         [V38    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V39 tmp8         [V39    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V40 tmp9         [V40    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V41 tmp10        [V41    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V42 tmp11        [V42    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V43 tmp12        [V43    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V44 tmp13        [V44    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V45 tmp14        [V45    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V46 tmp15        [V46    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V47 tmp16        [V47    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V48 tmp17        [V48    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V49 tmp18        [V49    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V50 tmp19        [V50    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V51 tmp20        [V51    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V52 tmp21        [V52    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V53 cse0         [V53,T21] (  8,  8   )  simd16  ->  mm13         "CSE #01: moderate"
+;  V54 cse1         [V54,T29] (  5,  2.50)  simd16  ->  mm0         "CSE #02: moderate"
+;  V55 rat0         [V55,T23] (  3,  6   )  simd16  ->  mm0         "ReplaceWithLclVar is creating a new local variable"
+;  V56 rat1         [V56,T24] (  3,  6   )  simd16  ->  mm0         "ReplaceWithLclVar is creating a new local variable"
 ;
 ; Lcl frame size = 0
 
 G_M1478_IG01:
-						;; size=0 bbWeight=1 PerfScore 0.00
+       push     rbp
+       mov      rbp, rsp
+						;; size=4 bbWeight=1 PerfScore 1.25
 G_M1478_IG02:
-       mov      rax, 0xD1FFAB1E      ; code for System.Numerics.Matrix4x4+Impl:<Invert>g__SseImpl|53_0(byref,byref):bool
-						;; size=10 bbWeight=1 PerfScore 0.25
+       vmovups  xmm0, xmmword ptr [rdi]
+       vmovups  xmm1, xmmword ptr [rdi+0x10]
+       vmovups  xmm2, xmmword ptr [rdi+0x20]
+       vmovups  xmm3, xmmword ptr [rdi+0x30]
+       vmovhlps xmm4, xmm1, xmm0
+       vmovlhps xmm5, xmm2, xmm3
+       vmovhlps xmm3, xmm3, xmm2
+       vmovlhps xmm1, xmm0, xmm1
+       vshufps  xmm0, xmm1, xmm5, -120
+       vshufps  xmm1, xmm1, xmm5, -35
+       vshufps  xmm2, xmm4, xmm3, -120
+       vshufps  xmm3, xmm4, xmm3, -35
+       vpermilps xmm4, xmm2, 80
+       vpermilps xmm5, xmm3, -18
+       vpermilps xmm6, xmm0, 80
+       vpermilps xmm7, xmm1, -18
+       vshufps  xmm8, xmm2, xmm0, -120
+       vshufps  xmm9, xmm3, xmm1, -35
+       vmulps   xmm5, xmm4, xmm5
+       vmulps   xmm7, xmm7, xmm6
+       vmulps   xmm9, xmm8, xmm9
+       vpermilps xmm4, xmm2, -18
+       vpermilps xmm6, xmm3, 80
+       vpermilps xmm8, xmm0, -18
+       vpermilps xmm10, xmm1, 80
+       vshufps  xmm11, xmm2, xmm0, -35
+       vshufps  xmm12, xmm3, xmm1, -120
+       vbroadcastss xmm13, dword ptr [reloc @RWD00]
+       vxorps   xmm4, xmm13, xmm4
+       vfmadd231ps xmm5, xmm6, xmm4
+       vxorps   xmm6, xmm13, xmm8
+       vfmadd231ps xmm7, xmm10, xmm6
+       vxorps   xmm10, xmm13, xmm11
+       vfmadd231ps xmm9, xmm12, xmm10
+       vshufps  xmm10, xmm5, xmm9, 93
+       vpermilps xmm4, xmm1, 73
+       vshufps  xmm6, xmm10, xmm5, 50
+       vpermilps xmm8, xmm0, 18
+       vshufps  xmm10, xmm10, xmm5, -103
+       vshufps  xmm14, xmm7, xmm9, -3
+       vpermilps xmm11, xmm3, 73
+       vshufps  xmm12, xmm14, xmm7, 50
+       vpermilps xmm15, xmm2, 18
+       vshufps  xmm14, xmm14, xmm7, -103
+       vmulps   xmm16, xmm4, xmm6
+       vmulps   xmm17, xmm10, xmm8
+       vmulps   xmm18, xmm11, xmm12
+       vmulps   xmm14, xmm14, xmm15
+       vshufps  xmm10, xmm5, xmm9, 4
+       vpermilps xmm4, xmm1, -98
+						;; size=265 bbWeight=1 PerfScore 73.00
 G_M1478_IG03:
-       tail.jmp [rax]System.Numerics.Matrix4x4+Impl:<Invert>g__SseImpl|53_0(byref,byref):bool
-						;; size=3 bbWeight=1 PerfScore 2.00
+       vshufps  xmm6, xmm5, xmm10, -109
+       vpermilps xmm8, xmm0, 123
+       vshufps  xmm10, xmm5, xmm10, 38
+       vshufps  xmm11, xmm7, xmm9, -92
+       vpermilps xmm12, xmm3, -98
+       vshufps  xmm15, xmm7, xmm11, -109
+       vpermilps xmm19, xmm2, 123
+       vshufps  xmm11, xmm7, xmm11, 38
+       vxorps   xmm4, xmm13, xmm4
+       vfmadd231ps xmm16, xmm6, xmm4
+       vxorps   xmm6, xmm13, xmm8
+       vfmadd231ps xmm17, xmm10, xmm6
+       vxorps   xmm8, xmm13, xmm12
+       vfmadd231ps xmm18, xmm15, xmm8
+       vxorps   xmm10, xmm13, xmm19
+       vfmadd231ps xmm14, xmm11, xmm10
+       vpermilps xmm4, xmm1, 51
+       vshufps  xmm12, xmm5, xmm9, 74
+       vpermilps xmm6, xmm12, 44
+       vpermilps xmm8, xmm0, -115
+       vshufps  xmm15, xmm5, xmm9, 76
+       vpermilps xmm10, xmm15, -109
+       vpermilps xmm12, xmm3, 51
+       vshufps  xmm19, xmm7, xmm9, -22
+       vpermilps xmm15, xmm19, 44
+       vpermilps xmm19, xmm2, -115
+       vshufps  xmm11, xmm7, xmm9, -20
+       vpermilps xmm11, xmm11, -109
+       vmulps   xmm4, xmm4, xmm6
+       vmulps   xmm8, xmm10, xmm8
+       vmulps   xmm12, xmm12, xmm15
+       vmulps   xmm19, xmm11, xmm19
+       vsubps   xmm1, xmm16, xmm4
+       vaddps   xmm16, xmm16, xmm4
+       vaddps   xmm2, xmm8, xmm17
+       vsubps   xmm17, xmm17, xmm8
+       vsubps   xmm3, xmm18, xmm12
+       vaddps   xmm18, xmm12, xmm18
+       vaddps   xmm4, xmm14, xmm19
+       vsubps   xmm14, xmm14, xmm19
+       vshufps  xmm16, xmm16, xmm1, -40
+       vshufps  xmm17, xmm17, xmm2, -40
+       vshufps  xmm18, xmm18, xmm3, -40
+       vshufps  xmm14, xmm14, xmm4, -40
+       vpermilps xmm16, xmm16, -40
+       vpermilps xmm17, xmm17, -40
+       vpermilps xmm18, xmm18, -40
+       vpermilps xmm14, xmm14, -40
+       vmulps   xmm0, xmm16, xmm0
+						;; size=294 bbWeight=1 PerfScore 80.33
+G_M1478_IG04:
+       vpermilps xmm1, xmm0, -79
+       vaddps   xmm0, xmm1, xmm0
+       vpermilps xmm1, xmm0, 78
+       vaddps   xmm0, xmm1, xmm0
+       vandps   xmm1, xmm0, xmmword ptr [reloc @RWD04]
+       vmovss   xmm2, dword ptr [reloc @RWD20]
+       vucomiss xmm2, xmm1
+       ja       SHORT G_M1478_IG07
+						;; size=42 bbWeight=1 PerfScore 15.00
+G_M1478_IG05:
+       vbroadcastss xmm0, xmm0
+       vbroadcastss xmm1, dword ptr [reloc @RWD24]
+       vdivps   xmm0, xmm1, xmm0
+       vmulps   xmm1, xmm16, xmm0
+       vmovups  xmmword ptr [rsi], xmm1
+       vmulps   xmm1, xmm17, xmm0
+       vmovups  xmmword ptr [rsi+0x10], xmm1
+       vmulps   xmm1, xmm18, xmm0
+       vmovups  xmmword ptr [rsi+0x20], xmm1
+       vmulps   xmm0, xmm14, xmm0
+       vmovups  xmmword ptr [rsi+0x30], xmm0
+       mov      eax, 1
+						;; size=64 bbWeight=0.50 PerfScore 14.62
+G_M1478_IG06:
+       pop      rbp
+       ret      
+						;; size=2 bbWeight=0.50 PerfScore 0.75
+G_M1478_IG07:
+       vbroadcastss xmm0, dword ptr [reloc @RWD28]
+       vmovups  xmmword ptr [rsi], xmm0
+       vmovups  xmmword ptr [rsi+0x10], xmm0
+       vmovups  xmmword ptr [rsi+0x20], xmm0
+       vmovups  xmmword ptr [rsi+0x30], xmm0
+       xor      eax, eax
+						;; size=30 bbWeight=0.50 PerfScore 3.12
+G_M1478_IG08:
+       pop      rbp
+       ret      
+						;; size=2 bbWeight=0.50 PerfScore 0.75
+RWD00  	dd	80000000h		;        -0
+RWD04  	dq	7FFFFFFF7FFFFFFFh, 7FFFFFFF7FFFFFFFh
+RWD20  	dd	00000001h		; 1.4013e-45
+RWD24  	dd	3F800000h		;         1
+RWD28  	dd	FFC00000h		;      -nan
 
-; Total bytes of code 13, prolog size 0, PerfScore 2.25, instruction count 2, allocated bytes for code 13 (MethodHash=46b7fa39) for method System.Numerics.Matrix4x4+Impl:Invert(byref,byref):bool (FullOpts)
+
+; Total bytes of code 703, prolog size 4, PerfScore 188.83, instruction count 131, allocated bytes for code 703 (MethodHash=46b7fa39) for method System.Numerics.Matrix4x4+Impl:Invert(byref,byref):bool (FullOpts)
 ; ============================================================

677 (2,507.41 % of base) - System.Numerics.Matrix4x4:Invert(System.Numerics.Matrix4x4,byref):bool

 ; Assembly listing for method System.Numerics.Matrix4x4:Invert(System.Numerics.Matrix4x4,byref):bool (FullOpts)
 ; Emitting BLENDED_CODE for x64 + VEX + EVEX on Unix
 ; FullOpts code
 ; optimized code
-; rsp based frame
+; rbp based frame
 ; partially interruptible
 ; No PGO data
-; 0 inlinees with PGO data; 2 single block inlinees; 1 inlinees without PGO data
+; 0 inlinees with PGO data; 5 single block inlinees; 21 inlinees without PGO data
 ; Final local variable assignments
 ;
-;  V00 arg0         [V00    ] (  1,  1   )  struct (64) [rsp+0x10]  do-not-enreg[XS] addr-exposed ld-addr-op single-def <System.Numerics.Matrix4x4>
-;  V01 arg1         [V01,T00] (  3,  3   )   byref  ->  rsi         single-def
+;  V00 arg0         [V00,T01] (  4,  4   )  struct (64) [rbp+0x10]  do-not-enreg[SF] ld-addr-op single-def <System.Numerics.Matrix4x4>
+;  V01 arg1         [V01,T00] ( 10,  6   )   byref  ->  rdi         single-def
 ;# V02 OutArgs      [V02    ] (  1,  1   )  struct ( 0) [rsp+0x00]  do-not-enreg[XS] addr-exposed "OutgoingArgSpace" <Empty>
 ;* V03 tmp1         [V03    ] (  0,  0   )   byref  ->  zero-ref    "impAppendStmt"
-;* V04 tmp2         [V04    ] (  0,  0   )   ubyte  ->  zero-ref    "Inline return value spill temp"
+;  V04 tmp2         [V04,T02] (  3,  2   )   ubyte  ->  rax         "Inline return value spill temp"
+;  V05 tmp3         [V05,T07] ( 12, 12   )  simd16  ->  mm0         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V06 tmp4         [V06,T10] ( 11, 11   )  simd16  ->  mm1         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V07 tmp5         [V07,T11] ( 11, 11   )  simd16  ->  mm2         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V08 tmp6         [V08,T12] ( 11, 11   )  simd16  ->  mm3         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V09 tmp7         [V09,T26] (  3,  3   )  simd16  ->  mm4         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V10 tmp8         [V10,T27] (  3,  3   )  simd16  ->  mm5         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V11 tmp9         [V11,T28] (  3,  3   )  simd16  ->  mm3         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V12 tmp10        [V12,T23] (  3,  6   )  simd16  ->  mm1         "dup spill"
+;  V13 tmp11        [V13,T04] ( 13, 13   )  simd16  ->  mm4         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V14 tmp12        [V14,T19] ( 10, 10   )  simd16  ->  registers   "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V15 tmp13        [V15,T05] ( 13, 13   )  simd16  ->  registers   "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V16 tmp14        [V16,T03] ( 16, 16   )  simd16  ->  registers   "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V17 tmp15        [V17,T06] ( 13, 13   )  simd16  ->  registers   "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V18 tmp16        [V18,T20] ( 10, 10   )  simd16  ->  registers   "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V19 tmp17        [V19,T13] ( 11, 11   )  simd16  ->  mm5         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V20 tmp18        [V20,T14] ( 11, 11   )  simd16  ->  mm7         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V21 tmp19        [V21,T15] ( 11, 11   )  simd16  ->  mm9         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V22 tmp20        [V22,T08] ( 12, 12   )  simd16  ->  registers   "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V23 tmp21        [V23,T21] (  9,  9   )  simd16  ->  registers   "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V24 tmp22        [V24,T09] ( 12, 11.50)  simd16  ->  mm16         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V25 tmp23        [V25,T16] ( 11, 10.50)  simd16  ->  mm17         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V26 tmp24        [V26,T17] ( 11, 10.50)  simd16  ->  mm18         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V27 tmp25        [V27,T18] ( 11, 10.50)  simd16  ->  mm14         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V28 tmp26        [V28,T32] (  2,  2   )  simd16  ->  mm1         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V29 tmp27        [V29,T33] (  2,  2   )  simd16  ->  mm2         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V30 tmp28        [V30,T34] (  2,  2   )  simd16  ->  mm3         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V31 tmp29        [V31,T35] (  2,  2   )  simd16  ->  mm4         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V32 tmp30        [V32,T31] (  3,  2.50)   float  ->  mm0         "Inline stloc first use temp"
+;  V33 tmp31        [V33,T29] (  5,  2.50)  simd16  ->  mm0         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V34 tmp32        [V34    ] (  0,  0   )  simd16  ->  zero-ref    "Inline stloc first use temp" <System.Numerics.Vector4>
+;* V35 tmp33        [V35    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V36 tmp34        [V36    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V37 tmp35        [V37    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V38 tmp36        [V38    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V39 tmp37        [V39    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V40 tmp38        [V40    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V41 tmp39        [V41    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V42 tmp40        [V42    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V43 tmp41        [V43    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V44 tmp42        [V44    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V45 tmp43        [V45    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V46 tmp44        [V46    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V47 tmp45        [V47    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V48 tmp46        [V48    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V49 tmp47        [V49    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V50 tmp48        [V50    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V51 tmp49        [V51    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V52 tmp50        [V52    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V53 tmp51        [V53    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V54 tmp52        [V54    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V55 cse0         [V55,T22] (  8,  8   )  simd16  ->  mm13         "CSE #01: moderate"
+;  V56 cse1         [V56,T30] (  5,  2.50)  simd16  ->  mm0         "CSE #02: moderate"
+;  V57 rat0         [V57,T24] (  3,  6   )  simd16  ->  mm0         "ReplaceWithLclVar is creating a new local variable"
+;  V58 rat1         [V58,T25] (  3,  6   )  simd16  ->  mm0         "ReplaceWithLclVar is creating a new local variable"
 ;
-; Lcl frame size = 8
+; Lcl frame size = 0
 
 G_M25377_IG01:
-       push     rax
-       mov      rsi, rdi
+       push     rbp
+       mov      rbp, rsp
 						;; size=4 bbWeight=1 PerfScore 1.25
 G_M25377_IG02:
-       lea      rdi, [rsp+0x10]
-       mov      rax, 0xD1FFAB1E      ; code for System.Numerics.Matrix4x4+Impl:<Invert>g__SseImpl|53_0(byref,byref):bool
-       call     [rax]System.Numerics.Matrix4x4+Impl:<Invert>g__SseImpl|53_0(byref,byref):bool
-       nop      
-						;; size=18 bbWeight=1 PerfScore 4.00
+       vmovups  xmm0, xmmword ptr [rbp+0x10]
+       vmovups  xmm1, xmmword ptr [rbp+0x20]
+       vmovups  xmm2, xmmword ptr [rbp+0x30]
+       vmovups  xmm3, xmmword ptr [rbp+0x40]
+       vmovhlps xmm4, xmm1, xmm0
+       vmovlhps xmm5, xmm2, xmm3
+       vmovhlps xmm3, xmm3, xmm2
+       vmovlhps xmm1, xmm0, xmm1
+       vshufps  xmm0, xmm1, xmm5, -120
+       vshufps  xmm1, xmm1, xmm5, -35
+       vshufps  xmm2, xmm4, xmm3, -120
+       vshufps  xmm3, xmm4, xmm3, -35
+       vpermilps xmm4, xmm2, 80
+       vpermilps xmm5, xmm3, -18
+       vpermilps xmm6, xmm0, 80
+       vpermilps xmm7, xmm1, -18
+       vshufps  xmm8, xmm2, xmm0, -120
+       vshufps  xmm9, xmm3, xmm1, -35
+       vmulps   xmm5, xmm4, xmm5
+       vmulps   xmm7, xmm7, xmm6
+       vmulps   xmm9, xmm8, xmm9
+       vpermilps xmm4, xmm2, -18
+       vpermilps xmm6, xmm3, 80
+       vpermilps xmm8, xmm0, -18
+       vpermilps xmm10, xmm1, 80
+       vshufps  xmm11, xmm2, xmm0, -35
+       vshufps  xmm12, xmm3, xmm1, -120
+       vbroadcastss xmm13, dword ptr [reloc @RWD00]
+       vxorps   xmm4, xmm13, xmm4
+       vfmadd231ps xmm5, xmm6, xmm4
+       vxorps   xmm6, xmm13, xmm8
+       vfmadd231ps xmm7, xmm10, xmm6
+       vxorps   xmm10, xmm13, xmm11
+       vfmadd231ps xmm9, xmm12, xmm10
+       vshufps  xmm10, xmm5, xmm9, 93
+       vpermilps xmm4, xmm1, 73
+       vshufps  xmm6, xmm10, xmm5, 50
+       vpermilps xmm8, xmm0, 18
+       vshufps  xmm10, xmm10, xmm5, -103
+       vshufps  xmm14, xmm7, xmm9, -3
+       vpermilps xmm11, xmm3, 73
+       vshufps  xmm12, xmm14, xmm7, 50
+       vpermilps xmm15, xmm2, 18
+       vshufps  xmm14, xmm14, xmm7, -103
+       vmulps   xmm16, xmm4, xmm6
+       vmulps   xmm17, xmm10, xmm8
+       vmulps   xmm18, xmm11, xmm12
+       vmulps   xmm14, xmm14, xmm15
+       vshufps  xmm10, xmm5, xmm9, 4
+       vpermilps xmm4, xmm1, -98
+						;; size=266 bbWeight=1 PerfScore 69.00
 G_M25377_IG03:
-       add      rsp, 8
+       vshufps  xmm6, xmm5, xmm10, -109
+       vpermilps xmm8, xmm0, 123
+       vshufps  xmm10, xmm5, xmm10, 38
+       vshufps  xmm11, xmm7, xmm9, -92
+       vpermilps xmm12, xmm3, -98
+       vshufps  xmm15, xmm7, xmm11, -109
+       vpermilps xmm19, xmm2, 123
+       vshufps  xmm11, xmm7, xmm11, 38
+       vxorps   xmm4, xmm13, xmm4
+       vfmadd231ps xmm16, xmm6, xmm4
+       vxorps   xmm6, xmm13, xmm8
+       vfmadd231ps xmm17, xmm10, xmm6
+       vxorps   xmm8, xmm13, xmm12
+       vfmadd231ps xmm18, xmm15, xmm8
+       vxorps   xmm10, xmm13, xmm19
+       vfmadd231ps xmm14, xmm11, xmm10
+       vpermilps xmm4, xmm1, 51
+       vshufps  xmm12, xmm5, xmm9, 74
+       vpermilps xmm6, xmm12, 44
+       vpermilps xmm8, xmm0, -115
+       vshufps  xmm15, xmm5, xmm9, 76
+       vpermilps xmm10, xmm15, -109
+       vpermilps xmm12, xmm3, 51
+       vshufps  xmm19, xmm7, xmm9, -22
+       vpermilps xmm15, xmm19, 44
+       vpermilps xmm19, xmm2, -115
+       vshufps  xmm11, xmm7, xmm9, -20
+       vpermilps xmm11, xmm11, -109
+       vmulps   xmm4, xmm4, xmm6
+       vmulps   xmm8, xmm10, xmm8
+       vmulps   xmm12, xmm12, xmm15
+       vmulps   xmm19, xmm11, xmm19
+       vsubps   xmm1, xmm16, xmm4
+       vaddps   xmm16, xmm16, xmm4
+       vaddps   xmm2, xmm8, xmm17
+       vsubps   xmm17, xmm17, xmm8
+       vsubps   xmm3, xmm18, xmm12
+       vaddps   xmm18, xmm12, xmm18
+       vaddps   xmm4, xmm14, xmm19
+       vsubps   xmm14, xmm14, xmm19
+       vshufps  xmm16, xmm16, xmm1, -40
+       vshufps  xmm17, xmm17, xmm2, -40
+       vshufps  xmm18, xmm18, xmm3, -40
+       vshufps  xmm14, xmm14, xmm4, -40
+       vpermilps xmm16, xmm16, -40
+       vpermilps xmm17, xmm17, -40
+       vpermilps xmm18, xmm18, -40
+       vpermilps xmm14, xmm14, -40
+       vmulps   xmm0, xmm16, xmm0
+						;; size=294 bbWeight=1 PerfScore 80.33
+G_M25377_IG04:
+       vpermilps xmm1, xmm0, -79
+       vaddps   xmm0, xmm1, xmm0
+       vpermilps xmm1, xmm0, 78
+       vaddps   xmm0, xmm1, xmm0
+       vandps   xmm1, xmm0, xmmword ptr [reloc @RWD04]
+       vmovss   xmm2, dword ptr [reloc @RWD20]
+       vucomiss xmm2, xmm1
+       ja       SHORT G_M25377_IG06
+						;; size=42 bbWeight=1 PerfScore 15.00
+G_M25377_IG05:
+       vbroadcastss xmm0, xmm0
+       vbroadcastss xmm1, dword ptr [reloc @RWD24]
+       vdivps   xmm0, xmm1, xmm0
+       vmulps   xmm1, xmm16, xmm0
+       vmovups  xmmword ptr [rdi], xmm1
+       vmulps   xmm1, xmm17, xmm0
+       vmovups  xmmword ptr [rdi+0x10], xmm1
+       vmulps   xmm1, xmm18, xmm0
+       vmovups  xmmword ptr [rdi+0x20], xmm1
+       vmulps   xmm0, xmm14, xmm0
+       vmovups  xmmword ptr [rdi+0x30], xmm0
+       mov      eax, 1
+       jmp      SHORT G_M25377_IG07
+						;; size=66 bbWeight=0.50 PerfScore 15.62
+G_M25377_IG06:
+       vbroadcastss xmm0, dword ptr [reloc @RWD28]
+       vmovups  xmmword ptr [rdi], xmm0
+       vmovups  xmmword ptr [rdi+0x10], xmm0
+       vmovups  xmmword ptr [rdi+0x20], xmm0
+       vmovups  xmmword ptr [rdi+0x30], xmm0
+       xor      eax, eax
+						;; size=30 bbWeight=0.50 PerfScore 3.12
+G_M25377_IG07:
+       pop      rbp
        ret      
-						;; size=5 bbWeight=1 PerfScore 1.25
+						;; size=2 bbWeight=1 PerfScore 1.50
+RWD00  	dd	80000000h		;        -0
+RWD04  	dq	7FFFFFFF7FFFFFFFh, 7FFFFFFF7FFFFFFFh
+RWD20  	dd	00000001h		; 1.4013e-45
+RWD24  	dd	3F800000h		;         1
+RWD28  	dd	FFC00000h		;      -nan
 
-; Total bytes of code 27, prolog size 1, PerfScore 6.50, instruction count 8, allocated bytes for code 27 (MethodHash=b8199cde) for method System.Numerics.Matrix4x4:Invert(System.Numerics.Matrix4x4,byref):bool (FullOpts)
+
+; Total bytes of code 704, prolog size 4, PerfScore 185.83, instruction count 130, allocated bytes for code 704 (MethodHash=b8199cde) for method System.Numerics.Matrix4x4:Invert(System.Numerics.Matrix4x4,byref):bool (FullOpts)
 ; ============================================================

640 (129.29 % of base) - SixLabors.ImageSharp.ColorProfiles.Icc.Calculators.ColorTrcCalculator:.ctor(SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,bool):this

 ; Assembly listing for method SixLabors.ImageSharp.ColorProfiles.Icc.Calculators.ColorTrcCalculator:.ctor(SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,bool):this (FullOpts)
 ; Emitting BLENDED_CODE for x64 + VEX + EVEX on Unix
 ; FullOpts code
 ; optimized code
 ; rbp based frame
 ; partially interruptible
 ; No PGO data
-; 0 inlinees with PGO data; 7 single block inlinees; 1 inlinees without PGO data
+; 0 inlinees with PGO data; 10 single block inlinees; 21 inlinees without PGO data
 ; Final local variable assignments
 ;
-;  V00 this         [V00,T01] (  5,  5   )     ref  ->  rbx         this class-hnd single-def <SixLabors.ImageSharp.ColorProfiles.Icc.Calculators.ColorTrcCalculator>
-;  V01 arg1         [V01,T06] (  3,  3   )     ref  ->  r15         class-hnd single-def <SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry>
-;  V02 arg2         [V02,T07] (  3,  3   )     ref  ->  r14         class-hnd single-def <SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry>
-;  V03 arg3         [V03,T08] (  3,  3   )     ref  ->  r13         class-hnd single-def <SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry>
-;  V04 arg4         [V04,T09] (  3,  3   )     ref  ->  r12         class-hnd single-def <SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry>
-;  V05 arg5         [V05,T10] (  3,  3   )     ref  ->  [rbp-0xB0]  class-hnd single-def <SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry>
-;  V06 arg6         [V06,T14] (  1,  1   )     ref  ->  [rbp+0x10]  class-hnd single-def <SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry>
-;  V07 arg7         [V07,T12] (  3,  3   )   ubyte  ->  [rbp+0x18]  single-def
-;  V08 loc0         [V08,T15] (  4,  4   )  simd12  ->  mm0         single-def <System.Numerics.Vector3>
-;  V09 loc1         [V09,T16] (  4,  4   )  simd12  ->  mm4         single-def <System.Numerics.Vector3>
-;  V10 loc2         [V10,T17] (  4,  4   )  simd12  ->  mm1         single-def <System.Numerics.Vector3>
+;  V00 this         [V00,T02] (  5,  5   )     ref  ->  rbx         this class-hnd single-def <SixLabors.ImageSharp.ColorProfiles.Icc.Calculators.ColorTrcCalculator>
+;  V01 arg1         [V01,T08] (  3,  3   )     ref  ->  r15         class-hnd single-def <SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry>
+;  V02 arg2         [V02,T09] (  3,  3   )     ref  ->  r14         class-hnd single-def <SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry>
+;  V03 arg3         [V03,T10] (  3,  3   )     ref  ->  r13         class-hnd single-def <SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry>
+;  V04 arg4         [V04,T11] (  3,  3   )     ref  ->  r12         class-hnd single-def <SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry>
+;  V05 arg5         [V05,T12] (  3,  3   )     ref  ->  [rbp-0x70]  class-hnd single-def <SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry>
+;  V06 arg6         [V06,T15] (  1,  1   )     ref  ->  [rbp+0x10]  class-hnd single-def <SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry>
+;  V07 arg7         [V07,T14] (  3,  3   )   ubyte  ->  [rbp+0x18]  single-def
+;  V08 loc0         [V08,T36] (  4,  4   )  simd12  ->  mm0         single-def <System.Numerics.Vector3>
+;  V09 loc1         [V09,T37] (  4,  4   )  simd12  ->  mm4         single-def <System.Numerics.Vector3>
+;  V10 loc2         [V10,T38] (  4,  4   )  simd12  ->  mm1         single-def <System.Numerics.Vector3>
 ;  V11 OutArgs      [V11    ] (  1,  1   )  struct (64) [rsp+0x00]  do-not-enreg[XS] addr-exposed "OutgoingArgSpace" <UNNAMED>
-;  V12 tmp1         [V12,T00] (  5, 10   )     ref  ->  [rbp-0xB8]  class-hnd exact spill-single-def "NewArr temp" <SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry[]>
-;  V13 tmp2         [V13,T02] (  3,  6   )     ref  ->  [rbp-0xC0]  class-hnd exact spill-single-def "NewObj constructor temp" <SixLabors.ImageSharp.ColorProfiles.Icc.Calculators.TrcCalculator>
+;  V12 tmp1         [V12,T00] (  5, 10   )     ref  ->  [rbp-0x78]  class-hnd exact spill-single-def "NewArr temp" <SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry[]>
+;  V13 tmp2         [V13,T03] (  3,  6   )     ref  ->  [rbp-0x80]  class-hnd exact spill-single-def "NewObj constructor temp" <SixLabors.ImageSharp.ColorProfiles.Icc.Calculators.TrcCalculator>
 ;  V14 tmp3         [V14    ] (  3,  6   )  struct (64) [rbp-0x68]  do-not-enreg[XS] addr-exposed ld-addr-op "NewObj constructor temp" <System.Numerics.Matrix4x4>
-;  V15 tmp4         [V15,T13] (  2,  2   )   byref  ->  rsi         single-def "Inlining Arg"
-;  V16 tmp5         [V16    ] (  2,  2   )  struct (64) [rbp-0xA8]  do-not-enreg[XS] addr-exposed ld-addr-op "Inlining Arg" <System.Numerics.Matrix4x4>
+;  V15 tmp4         [V15,T01] (  9,  9   )   byref  ->  rbx         single-def "Inlining Arg"
+;* V16 tmp5         [V16    ] (  0,  0   )  struct (64) zero-ref    do-not-enreg[SF] ld-addr-op "Inlining Arg" <System.Numerics.Matrix4x4>
 ;* V17 tmp6         [V17    ] (  0,  0   )   byref  ->  zero-ref    "impAppendStmt"
 ;* V18 tmp7         [V18    ] (  0,  0   )   ubyte  ->  zero-ref    "Inline return value spill temp"
-;  V19 tmp8         [V19,T03] (  3,  6   )     ref  ->  rdi         single-def "arr expr"
-;  V20 tmp9         [V20,T04] (  3,  6   )     ref  ->  rdi         single-def "arr expr"
-;  V21 tmp10        [V21,T05] (  3,  6   )     ref  ->  rdi         single-def "arr expr"
-;  V22 cse0         [V22,T11] (  4,  3   )   byref  ->  rsi         "CSE #02: moderate"
+;  V19 tmp8         [V19,T30] (  9,  4.50)  simd16  ->  mm1         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V20 tmp9         [V20,T32] (  8,  4   )  simd16  ->  mm0         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V21 tmp10        [V21,T33] (  8,  4   )  simd16  ->  mm3         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V22 tmp11        [V22,T34] (  8,  4   )  simd16  ->  mm2         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V23 tmp12        [V23,T44] (  3,  1.50)  simd16  ->  mm4         single-def "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V24 tmp13        [V24,T45] (  3,  1.50)  simd16  ->  mm5         single-def "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V25 tmp14        [V25,T46] (  3,  1.50)  simd16  ->  mm2         single-def "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V26 tmp15        [V26,T39] (  3,  3   )  simd16  ->  mm0         "dup spill"
+;  V27 tmp16        [V27,T17] ( 13,  6.50)  simd16  ->  mm4         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V28 tmp17        [V28,T28] ( 10,  5   )  simd16  ->  registers   "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V29 tmp18        [V29,T18] ( 13,  6.50)  simd16  ->  registers   "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V30 tmp19        [V30,T16] ( 16,  8   )  simd16  ->  registers   "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V31 tmp20        [V31,T19] ( 13,  6.50)  simd16  ->  registers   "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V32 tmp21        [V32,T29] ( 10,  5   )  simd16  ->  registers   "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V33 tmp22        [V33,T22] ( 11,  5.50)  simd16  ->  mm5         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V34 tmp23        [V34,T23] ( 11,  5.50)  simd16  ->  mm7         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V35 tmp24        [V35,T24] ( 11,  5.50)  simd16  ->  mm9         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V36 tmp25        [V36,T20] ( 12,  6   )  simd16  ->  registers   "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V37 tmp26        [V37,T31] (  9,  4.50)  simd16  ->  registers   "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V38 tmp27        [V38,T21] ( 12,  6   )  simd16  ->  mm16         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V39 tmp28        [V39,T25] ( 11,  5.50)  simd16  ->  mm17         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V40 tmp29        [V40,T26] ( 11,  5.50)  simd16  ->  mm18         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V41 tmp30        [V41,T27] ( 11,  5.50)  simd16  ->  mm14         "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V42 tmp31        [V42,T52] (  2,  1   )  simd16  ->  mm0         single-def "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V43 tmp32        [V43,T53] (  2,  1   )  simd16  ->  mm2         single-def "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V44 tmp33        [V44,T54] (  2,  1   )  simd16  ->  mm3         single-def "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V45 tmp34        [V45,T55] (  2,  1   )  simd16  ->  mm4         single-def "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V46 tmp35        [V46,T47] (  3,  1.50)   float  ->  mm0         single-def "Inline stloc first use temp"
+;  V47 tmp36        [V47,T42] (  5,  2.50)  simd16  ->  mm0         single-def "Inline stloc first use temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V48 tmp37        [V48    ] (  0,  0   )  simd16  ->  zero-ref    single-def "Inline stloc first use temp" <System.Numerics.Vector4>
+;* V49 tmp38        [V49    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V50 tmp39        [V50    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V51 tmp40        [V51    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V52 tmp41        [V52    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V53 tmp42        [V53    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V54 tmp43        [V54    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V55 tmp44        [V55    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V56 tmp45        [V56    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V57 tmp46        [V57    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V58 tmp47        [V58    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V59 tmp48        [V59    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V60 tmp49        [V60    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V61 tmp50        [V61    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V62 tmp51        [V62    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V63 tmp52        [V63    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V64 tmp53        [V64    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V65 tmp54        [V65    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V66 tmp55        [V66    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V67 tmp56        [V67    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;* V68 tmp57        [V68    ] (  0,  0   )  simd16  ->  zero-ref    "Inline return value spill temp" <System.Runtime.Intrinsics.Vector128`1[float]>
+;  V69 tmp58        [V69,T48] (  3,  1.50)  simd16  ->  mm0         single-def "V16.[000..016)"
+;  V70 tmp59        [V70,T49] (  3,  1.50)  simd16  ->  mm1         single-def "V16.[016..032)"
+;  V71 tmp60        [V71,T50] (  3,  1.50)  simd16  ->  mm2         single-def "V16.[032..048)"
+;  V72 tmp61        [V72,T51] (  3,  1.50)  simd16  ->  mm3         single-def "V16.[048..064)"
+;  V73 tmp62        [V73,T07] (  5,  5   )   byref  ->  rax         single-def "Spilling address for field-by-field copy"
+;  V74 tmp63        [V74,T04] (  3,  6   )     ref  ->  rdi         single-def "arr expr"
+;  V75 tmp64        [V75,T05] (  3,  6   )     ref  ->  rdi         single-def "arr expr"
+;  V76 tmp65        [V76,T06] (  3,  6   )     ref  ->  rdi         single-def "arr expr"
+;  V77 cse0         [V77,T35] (  8,  4   )  simd16  ->  mm13         "CSE #03: moderate"
+;  V78 cse1         [V78,T43] (  5,  2.50)  simd16  ->  mm0         "CSE #04: moderate"
+;  V79 cse2         [V79,T13] (  4,  3   )   byref  ->  rbx         "CSE #02: moderate"
+;  V80 rat0         [V80,T40] (  3,  3   )  simd16  ->  mm0         "ReplaceWithLclVar is creating a new local variable"
+;  V81 rat1         [V81,T41] (  3,  3   )  simd16  ->  mm0         "ReplaceWithLclVar is creating a new local variable"
 ;
-; Lcl frame size = 216
+; Lcl frame size = 152
 
 G_M6329_IG01:
        push     rbp
        push     r15
        push     r14
        push     r13
        push     r12
        push     rbx
-       sub      rsp, 216
-       lea      rbp, [rsp+0x100]
-       mov      gword ptr [rbp-0xB0], r9
+       sub      rsp, 152
+       lea      rbp, [rsp+0xC0]
+       mov      gword ptr [rbp-0x70], r9
        mov      rbx, rdi
        mov      r15, rsi
        mov      r14, rdx
        mov      r13, rcx
        mov      r12, r8
        mov      ecx, dword ptr [rbp+0x18]
-						;; size=50 bbWeight=1 PerfScore 10.00
+						;; size=47 bbWeight=1 PerfScore 10.00
 G_M6329_IG02:
        mov      dword ptr [rbp+0x18], ecx
        mov      byte  ptr [rbx+0x10], cl
        mov      rdi, 0xD1FFAB1E      ; SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry[]
        mov      esi, 3
        call     CORINFO_HELP_NEWARR_1_PTR
-       mov      gword ptr [rbp-0xB8], rax
+       mov      gword ptr [rbp-0x78], rax
        lea      rdi, bword ptr [rax+0x10]
        mov      rsi, r12
        call     CORINFO_HELP_ASSIGN_REF
-       mov      r12, gword ptr [rbp-0xB8]
+       mov      r12, gword ptr [rbp-0x78]
        lea      rdi, bword ptr [r12+0x18]
-       mov      rsi, gword ptr [rbp-0xB0]
+       mov      rsi, gword ptr [rbp-0x70]
        call     CORINFO_HELP_ASSIGN_REF
        lea      rdi, bword ptr [r12+0x20]
        mov      rsi, gword ptr [rbp+0x10]
        call     CORINFO_HELP_ASSIGN_REF
        mov      rdi, 0xD1FFAB1E      ; SixLabors.ImageSharp.ColorProfiles.Icc.Calculators.TrcCalculator
        call     CORINFO_HELP_NEWSFAST
-       mov      gword ptr [rbp-0xC0], rax
+       mov      gword ptr [rbp-0x80], rax
        cmp      byte  ptr [rbp+0x18], 0
        sete     dl
        movzx    rdx, dl
        mov      rdi, rax
        mov      rsi, r12
        mov      r8, 0xD1FFAB1E      ; code for SixLabors.ImageSharp.ColorProfiles.Icc.Calculators.TrcCalculator:.ctor(SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry[],bool):this
        call     [r8]SixLabors.ImageSharp.ColorProfiles.Icc.Calculators.TrcCalculator:.ctor(SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry[],bool):this
        lea      rdi, bword ptr [rbx+0x08]
-       mov      rsi, gword ptr [rbp-0xC0]
+       mov      rsi, gword ptr [rbp-0x80]
        call     CORINFO_HELP_ASSIGN_REF
        mov      rdi, gword ptr [r15+0x10]
        cmp      dword ptr [rdi+0x08], 0
-       jbe      G_M6329_IG07
+       jbe      G_M6329_IG09
        vmovsd   xmm0, qword ptr [rdi+0x10]
        vinsertps xmm0, xmm0, dword ptr [rdi+0x18], 40
        mov      rdi, gword ptr [r14+0x10]
        cmp      dword ptr [rdi+0x08], 0
-       jbe      G_M6329_IG07
+       jbe      G_M6329_IG09
        vmovsd   xmm4, qword ptr [rdi+0x10]
        vinsertps xmm4, xmm4, dword ptr [rdi+0x18], 40
        mov      rdi, gword ptr [r13+0x10]
        cmp      dword ptr [rdi+0x08], 0
-						;; size=210 bbWeight=1 PerfScore 48.50
+						;; size=195 bbWeight=1 PerfScore 48.50
 G_M6329_IG03:
-       jbe      G_M6329_IG07
+       jbe      G_M6329_IG09
        vmovsd   xmm1, qword ptr [rdi+0x10]
        vinsertps xmm1, xmm1, dword ptr [rdi+0x18], 40
        vxorps   ymm2, ymm2, ymm2
        vmovdqu32 zmmword ptr [rbp-0x68], zmm2
        vmovaps  xmm2, xmm1
        vmovss   dword ptr [rsp], xmm2
        vmovshdup xmm2, xmm1
        vmovss   dword ptr [rsp+0x08], xmm2
        vunpckhps xmm1, xmm1, xmm1
        vmovss   dword ptr [rsp+0x10], xmm1
        vxorps   xmm1, xmm1, xmm1
        vmovss   dword ptr [rsp+0x18], xmm1
        vmovss   dword ptr [rsp+0x20], xmm1
        vmovss   dword ptr [rsp+0x28], xmm1
        vmovss   dword ptr [rsp+0x30], xmm1
        vmovss   xmm1, dword ptr [reloc @RWD00]
        vmovss   dword ptr [rsp+0x38], xmm1
        vmovshdup xmm1, xmm0
        vunpckhps xmm2, xmm0, xmm0
        vmovshdup xmm5, xmm4
        vunpckhps xmm6, xmm4, xmm4
        lea      rdi, [rbp-0x68]
        vxorps   xmm7, xmm7, xmm7
        vxorps   xmm3, xmm3, xmm3
        mov      rax, 0xD1FFAB1E      ; code for System.Numerics.Matrix4x4:.ctor(float,float,float,float,float,float,float,float,float,float,float,float,float,float,float,float):this
        call     [rax]System.Numerics.Matrix4x4:.ctor(float,float,float,float,float,float,float,float,float,float,float,float,float,float,float,float):this
        add      rbx, 20
-       mov      rsi, rbx
        vmovdqu32 zmm0, zmmword ptr [rbp-0x68]
-       vmovdqu32 zmmword ptr [rsi], zmm0
+       vmovdqu32 zmmword ptr [rbx], zmm0
        cmp      byte  ptr [rbp+0x18], 0
-       jne      SHORT G_M6329_IG05
-						;; size=172 bbWeight=1 PerfScore 36.83
+       jne      G_M6329_IG08
+						;; size=173 bbWeight=1 PerfScore 36.58
 G_M6329_IG04:
-       vmovdqu32 zmm0, zmmword ptr [rsi]
-       vmovdqu32 zmmword ptr [rbp-0xA8], zmm0
-       lea      rdi, [rbp-0xA8]
-       mov      rax, 0xD1FFAB1E      ; code for System.Numerics.Matrix4x4+Impl:<Invert>g__SseImpl|53_0(byref,byref):bool
-       call     [rax]System.Numerics.Matrix4x4+Impl:<Invert>g__SseImpl|53_0(byref,byref):bool
-						;; size=35 bbWeight=0.50 PerfScore 4.38
+       mov      rax, rbx
+       vmovups  xmm0, xmmword ptr [rax]
+       vmovups  xmm1, xmmword ptr [rax+0x10]
+       vmovups  xmm2, xmmword ptr [rax+0x20]
+       vmovups  xmm3, xmmword ptr [rax+0x30]
+       vmovhlps xmm4, xmm1, xmm0
+       vmovlhps xmm5, xmm2, xmm3
+       vmovhlps xmm2, xmm3, xmm2
+       vmovlhps xmm0, xmm0, xmm1
+       vshufps  xmm1, xmm0, xmm5, -120
+       vshufps  xmm0, xmm0, xmm5, -35
+       vshufps  xmm3, xmm4, xmm2, -120
+       vshufps  xmm2, xmm4, xmm2, -35
+       vpermilps xmm4, xmm3, 80
+       vpermilps xmm5, xmm2, -18
+       vpermilps xmm6, xmm1, 80
+       vpermilps xmm7, xmm0, -18
+       vshufps  xmm8, xmm3, xmm1, -120
+       vshufps  xmm9, xmm2, xmm0, -35
+       vmulps   xmm5, xmm4, xmm5
+       vmulps   xmm7, xmm7, xmm6
+       vmulps   xmm9, xmm8, xmm9
+       vpermilps xmm4, xmm3, -18
+       vpermilps xmm6, xmm2, 80
+       vpermilps xmm8, xmm1, -18
+       vpermilps xmm10, xmm0, 80
+       vshufps  xmm11, xmm3, xmm1, -35
+       vshufps  xmm12, xmm2, xmm0, -120
+       vbroadcastss xmm13, dword ptr [reloc @RWD04]
+       vxorps   xmm4, xmm13, xmm4
+       vfmadd231ps xmm5, xmm6, xmm4
+       vxorps   xmm6, xmm13, xmm8
+       vfmadd231ps xmm7, xmm10, xmm6
+       vxorps   xmm10, xmm13, xmm11
+       vfmadd231ps xmm9, xmm12, xmm10
+       vshufps  xmm10, xmm5, xmm9, 93
+       vpermilps xmm4, xmm0, 73
+       vshufps  xmm6, xmm10, xmm5, 50
+       vpermilps xmm8, xmm1, 18
+       vshufps  xmm10, xmm10, xmm5, -103
+       vshufps  xmm14, xmm7, xmm9, -3
+       vpermilps xmm11, xmm2, 73
+       vshufps  xmm12, xmm14, xmm7, 50
+       vpermilps xmm15, xmm3, 18
+       vshufps  xmm14, xmm14, xmm7, -103
+       vmulps   xmm16, xmm4, xmm6
+       vmulps   xmm17, xmm10, xmm8
+       vmulps   xmm18, xmm11, xmm12
+       vmulps   xmm14, xmm14, xmm15
+       vshufps  xmm10, xmm5, xmm9, 4
+						;; size=262 bbWeight=0.50 PerfScore 36.13
 G_M6329_IG05:
-       nop      
-						;; size=1 bbWeight=1 PerfScore 0.25
+       vpermilps xmm4, xmm0, -98
+       vshufps  xmm6, xmm5, xmm10, -109
+       vpermilps xmm8, xmm1, 123
+       vshufps  xmm10, xmm5, xmm10, 38
+       vshufps  xmm11, xmm7, xmm9, -92
+       vpermilps xmm12, xmm2, -98
+       vshufps  xmm15, xmm7, xmm11, -109
+       vpermilps xmm19, xmm3, 123
+       vshufps  xmm11, xmm7, xmm11, 38
+       vxorps   xmm4, xmm13, xmm4
+       vfmadd231ps xmm16, xmm6, xmm4
+       vxorps   xmm6, xmm13, xmm8
+       vfmadd231ps xmm17, xmm10, xmm6
+       vxorps   xmm8, xmm13, xmm12
+       vfmadd231ps xmm18, xmm15, xmm8
+       vxorps   xmm10, xmm13, xmm19
+       vfmadd231ps xmm14, xmm11, xmm10
+       vpermilps xmm4, xmm0, 51
+       vshufps  xmm12, xmm5, xmm9, 74
+       vpermilps xmm6, xmm12, 44
+       vpermilps xmm8, xmm1, -115
+       vshufps  xmm15, xmm5, xmm9, 76
+       vpermilps xmm10, xmm15, -109
+       vpermilps xmm12, xmm2, 51
+       vshufps  xmm19, xmm7, xmm9, -22
+       vpermilps xmm15, xmm19, 44
+       vpermilps xmm19, xmm3, -115
+       vshufps  xmm11, xmm7, xmm9, -20
+       vpermilps xmm11, xmm11, -109
+       vmulps   xmm4, xmm4, xmm6
+       vmulps   xmm8, xmm10, xmm8
+       vmulps   xmm12, xmm12, xmm15
+       vmulps   xmm19, xmm11, xmm19
+       vsubps   xmm0, xmm16, xmm4
+       vaddps   xmm16, xmm16, xmm4
+       vaddps   xmm2, xmm8, xmm17
+       vsubps   xmm17, xmm17, xmm8
+       vsubps   xmm3, xmm18, xmm12
+       vaddps   xmm18, xmm12, xmm18
+       vaddps   xmm4, xmm14, xmm19
+       vsubps   xmm14, xmm14, xmm19
+       vshufps  xmm16, xmm16, xmm0, -40
+       vshufps  xmm17, xmm17, xmm2, -40
+       vshufps  xmm18, xmm18, xmm3, -40
+       vshufps  xmm14, xmm14, xmm4, -40
+       vpermilps xmm16, xmm16, -40
+       vpermilps xmm17, xmm17, -40
+       vpermilps xmm18, xmm18, -40
+       vpermilps xmm14, xmm14, -40
+						;; size=294 bbWeight=0.50 PerfScore 39.17
 G_M6329_IG06:
+       vmulps   xmm0, xmm16, xmm1
+       vpermilps xmm1, xmm0, -79
+       vaddps   xmm0, xmm1, xmm0
+       vpermilps xmm1, xmm0, 78
+       vaddps   xmm0, xmm1, xmm0
+       vandps   xmm1, xmm0, xmmword ptr [reloc @RWD08]
+       vmovss   xmm2, dword ptr [reloc @RWD24]
+       vucomiss xmm2, xmm1
+       ja       SHORT G_M6329_IG07
+       vbroadcastss xmm0, xmm0
+       vbroadcastss xmm1, dword ptr [reloc @RWD00]
+       vdivps   xmm0, xmm1, xmm0
+       vmulps   xmm1, xmm16, xmm0
+       vmovups  xmmword ptr [rbx], xmm1
+       vmulps   xmm1, xmm17, xmm0
+       vmovups  xmmword ptr [rbx+0x10], xmm1
+       vmulps   xmm1, xmm18, xmm0
+       vmovups  xmmword ptr [rbx+0x20], xmm1
+       vmulps   xmm0, xmm14, xmm0
+       vmovups  xmmword ptr [rbx+0x30], xmm0
+       jmp      SHORT G_M6329_IG08
+						;; size=109 bbWeight=0.50 PerfScore 24.50
+G_M6329_IG07:
+       vbroadcastss xmm0, dword ptr [reloc @RWD28]
+       vmovups  xmmword ptr [rbx], xmm0
+       vmovups  xmmword ptr [rbx+0x10], xmm0
+       vmovups  xmmword ptr [rbx+0x20], xmm0
+       vmovups  xmmword ptr [rbx+0x30], xmm0
+						;; size=28 bbWeight=0.50 PerfScore 3.00
+G_M6329_IG08:
        vzeroupper 
-       add      rsp, 216
+       add      rsp, 152
        pop      rbx
        pop      r12
        pop      r13
        pop      r14
        pop      r15
        pop      rbp
        ret      
 						;; size=21 bbWeight=1 PerfScore 5.25
-G_M6329_IG07:
+G_M6329_IG09:
        call     CORINFO_HELP_RNGCHKFAIL
        int3     
 						;; size=6 bbWeight=0 PerfScore 0.00
 RWD00  	dd	3F800000h		;         1
+RWD04  	dd	80000000h		;        -0
+RWD08  	dq	7FFFFFFF7FFFFFFFh, 7FFFFFFF7FFFFFFFh
+RWD24  	dd	00000001h		; 1.4013e-45
+RWD28  	dd	FFC00000h		;      -nan
 
 
-; Total bytes of code 495, prolog size 25, PerfScore 105.21, instruction count 106, allocated bytes for code 495 (MethodHash=bfcbe746) for method SixLabors.ImageSharp.ColorProfiles.Icc.Calculators.ColorTrcCalculator:.ctor(SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,bool):this (FullOpts)
+; Total bytes of code 1135, prolog size 25, PerfScore 203.13, instruction count 224, allocated bytes for code 1135 (MethodHash=bfcbe746) for method SixLabors.ImageSharp.ColorProfiles.Icc.Calculators.ColorTrcCalculator:.ctor(SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccXyzTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,SixLabors.ImageSharp.Metadata.Profiles.Icc.IccTagDataEntry,bool):this (FullOpts)
 ; ============================================================

Note: some changes were skipped as they were too large to fit into a comment.

Larger list of diffs: https://gist.github.com/MihuBot/df15408bddc79e91feac4b9fb1c5e52a

[MihuBot]

@tannergooding