cccl
cccl copied to clipboard
CUDA Core Compute Libraries
Fetch and parse timing info from CI jobs.
Those types must be host only so we can safely SFINAE away our own free functions that would be ambiguous Fixes #1679
### Is this a duplicate? - [X] I confirmed there appear to be no [duplicate issues](https://github.com/NVIDIA/cccl/issues) for this bug and that I agree to the [Code of Conduct](CODE_OF_CONDUCT.md) ### Is...
This pull request adds hierarchy_dimensions type template that allows to describe a hierarchy of CUDA threads with a mix of static and dynamic information. It can be used to pass...
## Description closes: #1738 ## Checklist - [ ] New or existing tests cover these changes. - [ ] The documentation is up to date with these changes.
## Description Adds a reusable workflow to build RAPIDS repos from source with the devcontainer's CCCL mount. Currently triggered nightly and on PRs (but is not a required PR check)....
This mirrors our current escape hatch against `__nv_bfloat16` to also allow the user to disable half support Addresses: #1727
### Is this a duplicate? - [X] I confirmed there appear to be no [duplicate issues](https://github.com/NVIDIA/cccl/issues) for this bug and that I agree to the [Code of Conduct](CODE_OF_CONDUCT.md) ### Type...
CUB docs were updated to use the new rst/sphinx docs system. We should update Thrust/libcudacxx docs to do the same. ```[tasklist] ### Tasks - [ ] https://github.com/NVIDIA/cccl/issues/1742 - [ ]...