Use the improved vectorization algorithm for binary and ternary TensorPrimitives operations#93409

Merged

tannergooding merged 11 commits into

dotnet:mainfrom

tannergooding:vectorize-align-2

Oct 17, 2023

Commits on Oct 12, 2023

Update InvokeSpanSpanIntoSpan<TBinaryOperator> for TensorPrimitives to use the better SIMD algorithm
tannergooding
committed
Update InvokeSpanScalarIntoSpan<TTransformOperator, TBinaryOperator> for TensorPrimitives to use the better SIMD algorithm
tannergooding
committed
Update InvokeSpanSpanSpanIntoSpan<TTernaryOperator> for TensorPrimitives to use the better SIMD algorithm
tannergooding
committed
Update InvokeSpanSpanScalarIntoSpan<TTernaryOperator> for TensorPrimitives to use the better SIMD algorithm
tannergooding
committed
Update InvokeSpanScalarSpanIntoSpan<TTernaryOperator> for TensorPrimitives to use the better SIMD algorithm
tannergooding
committed
Improve codegen slightly by using case 0, rather than default
tannergooding
committed
Adjust the canAlign check to be latter, to reduce branch count for data under the threshold
tannergooding
committed
Add a comment explaining the NonTemporalByteThreshold
tannergooding
committed
Merge remote-tracking branch 'dotnet/main' into vectorize-align-2
tannergooding
committed

Commits on Oct 13, 2023

Merge remote-tracking branch 'dotnet/main' into vectorize-align-2
tannergooding
committed

Commits on Oct 16, 2023

Make sure xTransformOp.CanVectorize is checked on .NET Standard
tannergooding
committed