Skip to content

[FEA] Port the logic at page_data.cu:282 to use thread_groups and avoid the magic 32 multiples. #16235

Description

@mhaseeb123

We should probably port the whole logic at page_data.cu:282 and elsewhere where used to thread_groups and avoid the magic 32 multiples.

Originally posted by @vuule in #16109 (comment)

Metadata

Metadata

Assignees

No one assigned

    Labels

    0 - BacklogIn queue waiting for assignmentfeature requestNew feature or requestgood first issueGood for newcomerslibcudfAffects libcudf (C++/CUDA) code.

    Type

    No type

    Projects

    No projects

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions