Skip to content

Conversation

viralbhadeshiya
Copy link

@viralbhadeshiya viralbhadeshiya commented Oct 14, 2025

Description

closes #6107

  • Replace all __popc occurrence with cuda::std::pop_count

Checklist

  • New or existing tests cover these changes.
  • The documentation is up to date with these changes.

Copy link
Contributor

copy-pr-bot bot commented Oct 14, 2025

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.

@cccl-authenticator-app cccl-authenticator-app bot moved this from Todo to In Review in CCCL Oct 14, 2025
@davebayer
Copy link
Contributor

/ok to test afd9ae6

@davebayer
Copy link
Contributor

Thanks for working on this :)

@github-project-automation github-project-automation bot moved this from In Review to In Progress in CCCL Oct 14, 2025

This comment has been minimized.

Copy link
Contributor

@davebayer davebayer left a comment

Choose a reason for hiding this comment

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

There is one more __popc(...) left in thrust/thrust/system/cuda/detail/set_operations.h

@viralbhadeshiya viralbhadeshiya requested a review from a team as a code owner October 14, 2025 15:42
@viralbhadeshiya
Copy link
Author

There is one more __popc(...) left in thrust/thrust/system/cuda/detail/set_operations.h

This is done.

@davebayer
Copy link
Contributor

/ok to test 00154d6

@davebayer
Copy link
Contributor

/ok to test 7f523b4

Copy link
Contributor

🥳 CI Workflow Results

🟩 Finished in 1h 27m: Pass: 100%/171 | Total: 1d 13h | Max: 1h 26m | Hits: 98%/353829

See results here.

@davebayer davebayer requested a review from miscco October 14, 2025 18:07
@miscco
Copy link
Contributor

miscco commented Oct 15, 2025

@elstehle can I bully you into checking SASS?

Copy link
Contributor

@elstehle elstehle left a comment

Choose a reason for hiding this comment

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

Confirmed that there are no sass changes.

Fyi, these are the steps I usually follow to check whether there are any sass changes:

  1. Identify the Device* algorithm(s) that may be affected by a change. This isn't always straight-forward. Sometimes, the change may only be affected on certain GPU architectures (this would require looking into the tuning policy of affected algorithms)
  2. Compile the benchmarks for the specific Device* algorithm(s) identified in (1) and dump the sass code. E.g., ninja cub.bench.radix_sort.keys.base && cuobjdump -sass ./bin/cub.bench.radix_sort.keys.base |c++filt > ./radix_sort.keys_after.sass.
  3. Checkout the main branch to compare against baseline sass code being generated: git checkout $(git merge-base HEAD upstream/main)
  4. Dump the sass code emitted on main branch. E.g., ninja cub.bench.radix_sort.keys.base && cuobjdump -sass ./bin/cub.bench.radix_sort.keys.base |c++filt > ./radix_sort.keys_before.sass.
  5. Check whether there are sass changes: git diff --text --no-index --word-diff radix_sort.keys_before.sass radix_sort.keys_after.sass

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.

[FEA]: Replaces uses of __popc with cuda::std::popcount

5 participants