pytorch / pytorch/executorch

Support better tiling algorithms for ANE

Open
#9,630 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

triaged
Dominant language
Python
Stars
5k
Forks
1.2k
Avg merge
2d 10h
Merged PRs (30d)
581

Description

🚀 The feature, motivation and pitch

Support better tiling algorithms for ANE in linear, SDPA, matmul, bmm, etc. We noticed that we can boost performance by explicitly splitting up ops in PyTorch, but ideally this would be done by the CoreML compiler.

Alternatives

No response

Additional context

No response

RFC (Optional)

No response

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Review the linked PyTorch PR #8715 and the listed targets—linear, SDPA, matmul, and bmm—to understand the current explicit splitting approach. The issue names no files or tests, and completion criteria for changing CoreML compiler tiling algorithms are not specified.

Written by the indexing model from the issue text.

Assessment

Domain
compilers, machine-learning, mobile-dev
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.