NVIDIA / NVIDIA/cccl

Publicly expose `cub::detail::ptx_compute_cap` and deprecate `cub::PtxVersion`

Open
#8,729 0 comments 0 reactions 1 assignee Claimed by @davebayer View on GitHub
Dominant language
C++
Stars
2.5k
Forks
486
Avg merge
2d 6h
Merged PRs (30d)
295

Description

#8728 is adding a new internal primitive `cub::detail::ptx_cc` that returns the compute capability of the PTX of the cubin that will be run on the current device when a kernel is launched. This function offers the same functionality as `cub::PtxVersion`, but is strongly typed. We should expose it at the public API as a replacement of `cub::PtxVersion`.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.