Skip to content

Fix regressed performance in BLS MUL precompiles #1053

Open
@chfast

Description

@chfast

If the scalar passed to BLST is not less than the subgroup order, BLST switches the multiplication algorithm to a slower one. Fix this by reducing the scalar if needed.

Metadata

Metadata

Assignees

No one assigned

    Labels

    PragueChanges for Prague upgradeprecompilesRelated to EVM precompiles

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions