8000 how can the L2 arithmetic intensity be less than the HBM AI. · Issue #690 · ROCm/rocprofiler-compute · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

how can the L2 arithmetic intensity be less than the HBM AI. #690

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
Open
etiennemlb opened this issue Apr 30, 2025 · 2 comments
Open

how can the L2 arithmetic intensity be less than the HBM AI. #690

etiennemlb opened this issue Apr 30, 2025 · 2 comments
Labels
question Further information is requested Under Investigation

Comments

@etiennemlb
Copy link

Describe your question

I have the following roofline:

Image

This gives me a lower L2 ai than HBM ai. As all loads/stores that go through L2 should go through HBM we should have AI L1 >= AI L2 >= AI HBM right ?

Additional context

No response
@etiennemlb etiennemlb added the question Further information is requested label Apr 30, 2025
@etiennemlb
Copy link
Author

Additionally, how would one interpret the AI L1/L2 dots. Would that mean the farther left we are, the better the caches are used (we do more loads in caches/LDS). If the L1 dot is farther than L2 this means we use the L1, if the L2 is farther than the HBM that means we use it ?

@ppanchad-amd
Copy link

Hi @etiennemlb. Internal ticket has been created to assist with your issue. Thanks!

5163

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
question Further information is requested Under Investigation
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@etiennemlb @ppanchad-amd
0