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

Add fast two-level mode N GEMV (#926) #939

Merged
merged 3 commits into from
Apr 27, 2021
Merged

Add fast two-level mode N GEMV (#926) #939

merged 3 commits into from
Apr 27, 2021

Commits on Apr 20, 2021

  1. Add fast two-level mode N GEMV 

    Very close to cublas in performance, with LayoutLeft or LayoutRight.
(~100 gflops for 1e5 * 1e5 double matrix on V100)
    @brian-kelley
    brian-kelley committed Apr 20, 2021
    Configuration menu
    Copy the full SHA
    a8e4ef5 View commit details
    Browse the repository at this point in the history

  2. Replace View[] with View(), count madd as 2 flops 

    Counting flops that way is consistent with the rest of Kokkos Kernels.
    @brian-kelley
    brian-kelley committed Apr 20, 2021
    Configuration menu
    Copy the full SHA
    48664f4 View commit details
    Browse the repository at this point in the history

Commits on Apr 26, 2021

  1. Increased gemv test tolerance slightly

    @brian-kelley
    brian-kelley committed Apr 26, 2021
    Configuration menu
    Copy the full SHA
    96430c0 View commit details
    Browse the repository at this point in the history