Skip to content

workgroupUniformLoad: test pointer-to-atomic variant #4341

@dneto0

Description

@dneto0

See gpuweb/gpuweb#5141

This requires basic testing:

  • that all invocations get the same value from the atomic
  • that it passes uniformity checks (the load is treated as uniform)

Ideally we would also modify the memory model tests to stress test the case. (That would likely be on me, as I'm the maintainer of those tests)

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions