Skip to content

[Triton-MLIR][Backend] Revert a bug of smem allocation for non-scratch cvt_layout#697

Closed
goostavz wants to merge 1 commit into
triton-lang:triton-mlirfrom
goostavz:goostavz/dev2
Closed

[Triton-MLIR][Backend] Revert a bug of smem allocation for non-scratch cvt_layout#697
goostavz wants to merge 1 commit into
triton-lang:triton-mlirfrom
goostavz:goostavz/dev2

Conversation

@goostavz
Copy link
Copy Markdown
Collaborator

No description provided.

@goostavz goostavz changed the title [Triton-MLIR][Backend] Revesmem allocation for non-scratch cvt_layout [Triton-MLIR][Backend] Revert smem allocation for non-scratch cvt_layout Sep 23, 2022
@goostavz goostavz changed the title [Triton-MLIR][Backend] Revert smem allocation for non-scratch cvt_layout [Triton-MLIR][Backend] Revert a bug of smem allocation for non-scratch cvt_layout Sep 23, 2022
@Jokeren
Copy link
Copy Markdown
Contributor

Jokeren commented Sep 23, 2022

@goostavz This PR has been merged and modified by #692

@goostavz goostavz closed this Sep 24, 2022
ZzEeKkAa pushed a commit to ZzEeKkAa/triton that referenced this pull request Aug 5, 2024
…m_warps. (triton-lang#697)

Fix the issue in `accumulatePartialReductions` when sizeInterWarps >
threads_per_warp.
Use the warp reduction multiple times to get final result when the
problem size > maximum reduce lane number.
brunomazzottiamd pushed a commit to brunomazzottiamd/triton that referenced this pull request Jan 29, 2025
* add layernorm backward pass

* format the layernorm.py script

* remove an unnecessary test shape

* fixed minor issues

* additional refinement
scxiao pushed a commit to scxiao/triton that referenced this pull request Apr 2, 2026
)

Summary:
TLX UT

Pull Request resolved: facebookexperimental/triton#697

Reviewed By: htyu

Differential Revision: D87833054

Pulled By: dshi7

fbshipit-source-id: f2e0f81734755cf8f45fd9f3e6802339fabc0aa1
AlexAUT added a commit to AlexAUT/triton that referenced this pull request Apr 20, 2026
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.

2 participants