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

[QST] Volta shared memory layout example #1158

Closed
ssiu opened this issue Oct 25, 2023 · 1 comment
Closed

[QST] Volta shared memory layout example #1158

ssiu opened this issue Oct 25, 2023 · 1 comment
Labels
? - Needs Triage question Question

Comments

@ssiu
Copy link

ssiu commented Oct 25, 2023 •
edited
Loading

Hi, I am currently studying these slides: https://developer.download.nvidia.com/video/gputechconf/gtc/2019/presentation/s9593-cutensor-high-performance-tensor-operations-in-cuda-v2.pdf

I was wondering if there's any sample code for loading data from global memory to shared memory as shown in page 23?

Can I also ask why there does not exist copy_sm70.hpp in cutlass/include/cute/arch and copy_traits_sm70.hpp in cutlass/include/cute/atom?

Thanks!
@thakkarV
Copy link
Collaborator

No special copy atoms are needed for Volta as all copies were thread copies with CUDA C++ exposure. The Default/Universal copy is sufficient pre Ampere

@ssiu ssiu closed this as completed Oct 26, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
? - Needs Triage question Question
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@thakkarV @ssiu