Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Implement offseted mode for linearize_cache_indices CUDA kernel #2554

Closed
wants to merge 1 commit into from
Closed

Implement offseted mode for linearize_cache_indices CUDA kernel #2554

wants to merge 1 commit into from

Commits on May 2, 2024

  1. Implement offseted mode for linearize_cache_indices CUDA kernel 

    Summary:
This patch will allow base_offset as a parameter so all values in `offset` will be decresed by that amount. This is done in a way that no copy of `offset`.

The ultimate goal for this is to achieve multipass prefetch, which require calling this kernel on a segment of `indices` (rather than the whole). See unittest for its usage.

Differential Revision: D56863774
    @levythu @facebook-github-bot
    levythu authored and facebook-github-bot committed May 2, 2024
    Configuration menu
    Copy the full SHA
    2d9634a View commit details
    Browse the repository at this point in the history