Skip to content

Unnecessary local mempcy in postAllReduce #5567

@wujingyue

Description

@wujingyue

doLocalCopy(output_tensor, input_tensor);

This can be fixed by generating host IR that reuses the same input and output buffer for allreduce.

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions