Skip to content

Use the new tuning API internally for detail::batched_topk::dispatch#9095

Draft
bernhardmgruber wants to merge 2 commits into
NVIDIA:mainfrom
bernhardmgruber:tune_batched_topk
Draft

Use the new tuning API internally for detail::batched_topk::dispatch#9095
bernhardmgruber wants to merge 2 commits into
NVIDIA:mainfrom
bernhardmgruber:tune_batched_topk

Conversation

@bernhardmgruber
Copy link
Copy Markdown
Contributor

  • SASS check

Fixes: #8482

@copy-pr-bot
Copy link
Copy Markdown
Contributor

copy-pr-bot Bot commented May 21, 2026

Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

@cccl-authenticator-app cccl-authenticator-app Bot moved this from Todo to In Progress in CCCL May 21, 2026
@bernhardmgruber
Copy link
Copy Markdown
Contributor Author

@elstehle Does the changeset proposed by this PR even make sense at this point? It seems there is no public API for batched topk yet, but I am going through benchmarks and move them from the two step allocation to the environment overload that does the allocation automatically.

Should I instead just leave a note in the benchmark to update them later?

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

Labels

None yet

Projects

Status: In Progress

Development

Successfully merging this pull request may close these issues.

Use the new tuning API internally for detail::batched_topk::dispatch

1 participant