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

WIP: [GPU] No usm host for dgpus when output is small #27558

Open
wants to merge 4 commits into
base: master
Choose a base branch
from
Open

WIP: [GPU] No usm host for dgpus when output is small #27558

wants to merge 4 commits into from
+19 −5

Commits on Nov 11, 2024

  1. [GPU] Do not use usm_host when network output tensor size is large 

    For dGPU, when network output size is large, performance is better with explicit copy then writing data to usm_host.
    @isanghao
    isanghao committed Nov 11, 2024
    Configuration menu
    Copy the full SHA
    be34b5e View commit details
    Browse the repository at this point in the history

  2. code fix

    @isanghao
    isanghao committed Nov 11, 2024
    Configuration menu
    Copy the full SHA
    ff3e3f8 View commit details
    Browse the repository at this point in the history

Commits on Nov 12, 2024

  1. cpplint fix

    @isanghao
    isanghao committed Nov 12, 2024
    Configuration menu
    Copy the full SHA
    f112bf6 View commit details
    Browse the repository at this point in the history

Commits on Nov 14, 2024

  1. test for more GPU

    @isanghao
    isanghao committed Nov 14, 2024
    Configuration menu
    Copy the full SHA
    abff1df View commit details
    Browse the repository at this point in the history