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

float4 with better vectorization for adamw.cu #268

Open
wants to merge 5 commits into
base: master
Choose a base branch
from
Open

float4 with better vectorization for adamw.cu #268

wants to merge 5 commits into from

Conversation

lancerts
Copy link
Contributor

@lancerts lancerts commented Apr 27, 2024 •
edited
Loading

On 3070,
Kernel 2
time gpu 0.0799 ms
time cpu 0.0168 ms

Kernel 3
time gpu 0.0780 ms
time cpu 0.0166 ms

@lancerts lancerts changed the title Adamw kernel with float4 Adamw kernel with float4 for memory coalescing Apr 27, 2024
@lancerts lancerts marked this pull request as ready for review April 27, 2024 06:06
@lancerts lancerts changed the title Adamw kernel with float4 for memory coalescing adamw.cu with float4 for memory coalescing Apr 27, 2024
@lancerts lancerts changed the title adamw.cu with float4 for memory coalescing adamw.cu with float4 for vectorization Apr 28, 2024
@lancerts lancerts changed the title adamw.cu with float4 for vectorization adamw.cu with float4 for better vectorization Apr 28, 2024
@lancerts lancerts changed the title adamw.cu with float4 for better vectorization float4 with better vectorization for adamw.cu Apr 28, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@lancerts