Skip to content

Fix nvexec launch move-only callable forwarding - #2198

Open
fallintoplace wants to merge 2 commits into
NVIDIA:mainfrom
fallintoplace:fix/nvexec-launch-move-only-forwarding
Open

Fix nvexec launch move-only callable forwarding#2198
fallintoplace wants to merge 2 commits into
NVIDIA:mainfrom
fallintoplace:fix/nvexec-launch-move-only-forwarding

Conversation

@fallintoplace

Copy link
Copy Markdown
Contributor

What changed

  • Forward the stored launch callable with the sender's value category.
  • Add coverage for a trivially copyable callable that cannot be copied.

Why

nvexec::launch accepts movable callables, but connect() passed the stored callable as an lvalue. Consuming a sender with a move-only callable therefore tried to copy it.

Testing

  • git diff --check
  • Standalone C++ syntax check for the move-only callable
  • CUDA coverage is included for GPU CI

@copy-pr-bot

copy-pr-bot Bot commented Aug 15, 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.

@ericniebler

Copy link
Copy Markdown
Collaborator

/ok to test 567bc83

@fallintoplace

Copy link
Copy Markdown
Contributor Author

@ericniebler Hi, I just pushed the clang format fix. 6d12001

@ericniebler

Copy link
Copy Markdown
Collaborator

/ok to test 6d12001

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