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

Using CUB (planned) 2.1.0's batched memcpy #7329

Open
jakirkham opened this issue Jan 18, 2023 · 1 comment
Open

Using CUB (planned) 2.1.0's batched memcpy #7329

jakirkham opened this issue Jan 18, 2023 · 1 comment
Labels
cat:performance Performance in terms of speed or memory consumption prio:medium

Comments

@jakirkham
Copy link
Member

PR ( NVIDIA/cub#359 ) landed in CUB recently. Looks like this will be part of CUB 2.1.0. This would allow batching multiple memcpy's together.

Potentially would be useful in operations like concatenate, stack, block, etc. Anything involving multiple copies between arrays
@leofang
Copy link
Member

leofang commented Jan 19, 2023

I'd love to make #6700 as the first step.

@kmaehashi kmaehashi added cat:performance Performance in terms of speed or memory consumption prio:medium labels Jan 19, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
cat:performance Performance in terms of speed or memory consumption prio:medium
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@kmaehashi @jakirkham @leofang