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

Copy trans fixl2 miss #1807

Open

coderfeli wants to merge 4 commits into 2.5_perf_fix from copy_trans_fixl2_miss

+131 −115

coderfeli commented Dec 25, 2024

optimize copy transpose op:
fix cacheline miss in L2
performance 2x for large shapes lastdim % 64 != 0
reach 4TB/s for typical cases

coderfeli added 4 commits

December 25, 2024 06:20


          opt copy transpose

d59db1b


          Copy.cu:fix copy for m % 8!=0

6d2c286


          rm error commit

f55a081


          revert submodules

dbe0ed3

coderfeli requested review from lcskrishna, pruthvistony, jerrymannil and fangche123

December 25, 2024 06:30

rocm-repo-management-api bot commented Dec 25, 2024 •

edited

Loading

Jenkins build for dbe0ed3480e7b79eff07337a19b68bba68e56371 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet