Skip to content

algebraic_* on f32x*/f64x* #539

Description

@Quant1um

Now that we have algebraic_* on floats on nightly, it would be also nice to have algebraic_* (and maybe some extra algebraic ops like algebraic_reduce_sum) on SIMD floats.

This would allow for faster SIMD code by allowing to opt-in into more relaxed floating-point semantics, allowing the compiler to reorder and fuse floating point operations.

Metadata

Metadata

Assignees

No one assigned

    Labels

    C-feature-requestCategory: a feature request, i.e. not implemented / a PR

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions