Skip to content

feat(dispatcher): add has_backend_impl and expose op name on dispatch wrapper & other updates - #178

Open
hannahli-nv wants to merge 6 commits into
mainfrom
tilegym_update
Open

feat(dispatcher): add has_backend_impl and expose op name on dispatch wrapper & other updates#178
hannahli-nv wants to merge 6 commits into
mainfrom
tilegym_update

Conversation

@hannahli-nv

@hannahli-nv hannahli-nv commented Aug 5, 2026

Copy link
Copy Markdown
Collaborator

Description

Update codes.

This PR contains 6 new commit(s).

Commits included:

1f60eb1 fix(tilecpp): mask the FMHA forward tail store
7d9bc68 fix(cutile): MLA split-KV NaN on sm_80 for non-power-of-2 seqlen
91c81e1 cuTile ragged_bmm: tune both kernel variants instead of guessing swap_ab from the shape
1fc7474 fix(tilecpp): chunk_gated_delta_rule V-tile bounds and G/Beta dtype
4a57dd1 perf(cutile softmax): per-op approx exp/div in all forward kernels
ab152dd feat(dispatcher): add has_backend_impl and expose op name on dispatch wrapper

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

@copy-pr-bot

copy-pr-bot Bot commented Aug 5, 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

Copy link
Copy Markdown
Collaborator Author

/ok to test 8a72068

@hannahli-nv
hannahli-nv force-pushed the tilegym_update branch 4 times, most recently from 15b2869 to a61cd51 Compare August 7, 2026 13:05
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.

6 participants