Skip to content

feat(ops): add linear_gluact_linear cuTile kernel & other updates - #194

Open
hannahli-nv wants to merge 11 commits into
mainfrom
tilegym_update
Open

feat(ops): add linear_gluact_linear cuTile kernel & other updates#194
hannahli-nv wants to merge 11 commits into
mainfrom
tilegym_update

Conversation

@hannahli-nv

@hannahli-nv hannahli-nv commented Sep 1, 2026 •
edited
Loading

Copy link
Copy Markdown
Collaborator

Description

Update codes.

This PR contains 11 new commit(s).

Commits included:

130e6cf Expand SwiGLU test_perf coverage with prefill shapes
e821978 tests: default --arch to sm100 when no GPU is available
e7655e3 cuTile: add narrow-K skinny matmul autotune tiles for sm120
30648e6 tilecpp: perf fixes for the activations and moe_align_block
637a3d0 Optimize cuTile modulated_rms_norm
d44fb27 Make the tilecpp optimization hints arch-agnostic
34f35a1 Tune cuTile gemm_alpha_beta for SM-throttled launches
ced08a8 feat(kernel-inventory): migrate runtime workloads to Kernel Factory JSONL
0142e48 perf(flashinfer/cutile): optimize MLA paged decode split-KV pipeline
a8c512b Optimize cuTile cross entropy for release qualification
152c784 feat(ops): add linear_gluact_linear cuTile kernel

CI Configuration

config:
  build: true
  # valid options are "ops" and "benchmark"
  test: []

Checklist

  • Code formatted and imports sorted via repo specifications (./format.sh)
  • Documentation updated (if needed)
  • CI configuration reviewed

nacgarg and others added 5 commits August 31, 2026 16:19
@nacgarg @xjmxyt
@qelk123 @hannahli-nv @0xtaruhi @anmolgupt @thschmid-nv
feat(ops): add linear_gluact_linear cuTile kernel
152c784 
Co-authored-by: Jinman Xie <jinmanx@nvidia.com>
Co-authored-by: Yinuo Liu <yinuol@nvidia.com>
Co-authored-by: Hannah Li <hanli@nvidia.com>
Co-authored-by: Zhengyi Zhang <zhengyiz@nvidia.com>
Co-authored-by: Anmol Gupta <anmolg@nvidia.com>
Co-authored-by: Thomas Schmid <thschmid@nvidia.com>
@xjmxyt
Optimize cuTile cross entropy for release qualification
a8c512b
@zfang-ai
perf(flashinfer/cutile): optimize MLA paged decode split-KV pipeline
0142e48
@lirundong
feat(kernel-inventory): migrate runtime workloads to Kernel Factory J…
ced08a8 
…SONL
@xjmxyt
Tune cuTile gemm_alpha_beta for SM-throttled launches
34f35a1
@copy-pr-bot

copy-pr-bot Bot commented Sep 1, 2026

Copy link
Copy Markdown

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@hannahli-nv

hannahli-nv commented Sep 2, 2026

Copy link
Copy Markdown
Collaborator Author

/ok to test 637a3d0

@hannahli-nv

hannahli-nv commented Sep 3, 2026

Copy link
Copy Markdown
Collaborator Author

/ok to test 9940d04

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

Reviewers

@xjmxyt xjmxyt xjmxyt approved these changes

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

6 participants

@hannahli-nv @xjmxyt @nacgarg @zfang-ai @lirundong @GordonWuCn