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

FP16 slower than BLAS_FP32 on Xeon Gold 6252 #2668

Open
lucasreis1 opened this issue Jun 3, 2024 · 0 comments
Open

FP16 slower than BLAS_FP32 on Xeon Gold 6252 #2668

lucasreis1 opened this issue Jun 3, 2024 · 0 comments

Comments

@lucasreis1
Copy link

Hey!

I'm benchmarking the library and looking to use its FP16 implementation on a modern AVX512 CPU and I'm seeing embarrassingly more performance on the standard FP32 BLAS GEMM. Is this normal?

For context, here are some measures I took with different matrix sizes:

plot_onlyfp

Note that I'm using the provided FP16Benchmark and comparing the output GFlops for each size.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@lucasreis1