KEMBAR78
Cleanup autotune_fallback_to_aten post-deprecation · Issue #153298 · pytorch/pytorch · GitHub
Skip to content

Cleanup autotune_fallback_to_aten post-deprecation #153298

@henrylhtsang

Description

@henrylhtsang

🚀 The feature, motivation and pitch

This is part of #147479

Probably should divide it into 3 parts.

  1. Remove references of autotune_fallback_to_aten in tests, benchmarks, etc. If autotune_fallback_to_aten is set to False, we can safely remove it.
  2. Remove usage of autotune_fallback_to_aten inside inductor.
  3. Remove env setting of autotune_fallback_to_aten in config.py, and say it is deprecated.

Alternatives

No response

Additional context

No response

cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @chenyang78 @kadeng @muchulee8 @amjames @chauhang @aakhundov

Metadata

Metadata

Assignees

Labels

actionablegood first issuemodule: inductortriagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate module

Type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions