-
Notifications
You must be signed in to change notification settings - Fork 1.1k
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
[FEA] TRSM/POTRF example request #1210
Comments
we actually considered trsm several years ago, but we didn't have a good idea of how to reuse cutlass code to do it without big change. @ANIKET-SHIVAM as to potrf, @kerrmudgeon |
Is it still in planning? Do you have any discussion groups? Please add me to that group, if you do not mind. I am very happy to contribute to CUTLASS. |
we discussed it internally several years ago. we found it is not a low hang fruit as trmm or syrk etc. |
Hi @JieRen98, thanks for your interest in CUTLASS! At this time, we have no plans to add more BLAS3 ops to CUTLASS. We do welcome contributions! We can answer any questions you have through Github issues or discussions. Going to close for now |
Is your feature request related to a problem? Please describe.
I wish I could use CUTLASS to do TRSM/POTRF
Describe the solution you'd like
Competitively performance vs cublas/cusolver
Describe alternatives you've considered
cublas
Additional context
It is obvious that the latest cublas and cusolver are based on CUTLASS (can be verified using nsight-compute). So can you give some examples using cutlass to do trsm/potrf?
The text was updated successfully, but these errors were encountered: