Skip to content

Define ML::device_resource_ref over CCCL resource_ref type - #8715

Draft
bdice wants to merge 1 commit into
NVIDIA:mainfrom
bdice:cuml-resource-ref-aliases
Draft

bdice wants to merge 1 commit into
NVIDIA:mainfrom
bdice:cuml-resource-ref-aliases

Conversation

@bdice

@bdice bdice commented Sep 29, 2026

Copy link
Copy Markdown
Contributor

Add cuml/common/resource_ref.hpp, which defines ML::device_resource_ref directly over the CCCL type, and use it in place of rmm::device_async_resource_ref.

using device_resource_ref = cuda::mr::resource_ref<cuda::mr::device_accessible>;

<rmm/resource_ref.hpp> is no longer included from cuML. The alias names the same type as before, so the ABI is unchanged.

Part of rapidsai/build-planning#337.

@bdice bdice added improvement Improvement / enhancement to an existing function non-breaking Non-breaking change labels Sep 29, 2026
@copy-pr-bot

copy-pr-bot Bot commented Sep 29, 2026

Copy link
Copy Markdown

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.

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

Labels

CUDA/C++ improvement Improvement / enhancement to an existing function non-breaking Non-breaking change

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant