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.

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

Updating matrix cores post for MI300X/A #16

Open
Epliz opened this issue May 1, 2024 · 1 comment
Open

Updating matrix cores post for MI300X/A #16

Epliz opened this issue May 1, 2024 · 1 comment

Comments

@Epliz
Copy link

Epliz commented May 1, 2024

Hi,

Would it possible for you to update the matrix cores post to describe the performance on MI300X/A as well as potentially describing the new functionalities of the architecture (fp8/bf8, sparsity support)?
Maybe in dedicated posts or just as updates to the original one?

That would be quite helpful, and interesting to many.
Best regards,
Epliz

@gsitaram
Copy link
Collaborator

gsitaram commented May 1, 2024

Hi @Epliz, thank you for reaching out to us. This update is something we have in mind as well. In the meantime, you are encouraged to use the AMD Matrix Instruction Calculator tool to get the details you need about the new architecture (gfx942) for CDNA3 GPUs (MI300A/MI300X), the expected throughput of instructions and the layout of the matrices in the input and output registers for different instructions, etc. This tool is kept up-to-date and can help answer most of your questions.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants