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

Upgrade cupti_counters component #158

Open

jrmadsen opened this issue Mar 3, 2021 · 0 comments

Collaborator

jrmadsen commented Mar 3, 2021

The current implementation for the CUDA HW counters using the callback API forces a synchronize at the end of a marker
This doesn't necessarily have too much of an impact on performance since the callback API serializes the kernels anyway but it should be migrated to use the correlation ID and updated later since this scheme will be required with the new CUPTI profiling API

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