Skip to content

Conversation

@kavin-sai-krishna
Copy link
Contributor

This PR adds support for the slice_scatter op in exported program and fx graph translator torch frontend.

@Hzfengsy Hzfengsy assigned Hzfengsy and tlopex and unassigned Hzfengsy May 28, 2025
Copy link
Member

@tlopex tlopex left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM! Thanks!

@tlopex tlopex merged commit 29de9ab into apache:main May 28, 2025
11 of 12 checks passed
@tqchen
Copy link
Member

tqchen commented Jun 5, 2025

Some checklist for adding a new op.

  • C0: Can the operator be decomposed into smaller ops (if so, then it may not be necessary to introduce the high-level op). It is helpful to reuse existing ops when possible so we don't have to introduce C1/C2, otherwise we need C1/C2
  • C1: Introduce legalization rules so the op at least compiles on CPU
  • C2: Make sure the op compiles on CUDA

@tqchen
Copy link
Member

tqchen commented Jun 5, 2025

it would be good to followup to check the above check list

ShiboXing pushed a commit to ShiboXing/tvm that referenced this pull request Aug 10, 2025
* Implement slice_scatter e2e

* Fix lint issue
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants