Skip to content

Specify stream in kernel launches #582

@oliver-batchelor

Description

@oliver-batchelor

Is it possible to specify a stream for a kernel launch in Quadrants?

I have a bunch of mixed taichi-pytorch-cuda code, which would benefit from being able to specify a consistent stream to avoid race conditions through it. It mostly seems to work at the moment, but it seems like perhaps I'm just relying on luck that there aren't problems.

Maybe I just misunderstood how the Taichi/(now Quadrants) runtime is operating.

Cheers,
Oliver

Concisely describe the proposed feature
A clear and concise description of what you want. For example,

I would like to add an Apple Metal backend to the compiler so that my Macbook GPU can be utilized.

Describe the solution you'd like (if any)
A clear and concise description of what you want to achieve and implement.

Additional comments
Add any other context or screenshots about the feature request here.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions