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

Spot checking with CUDA 10.1 on White/RIDE #576

Closed
brian-kelley opened this issue Jan 22, 2020 · 2 comments
Closed

Spot checking with CUDA 10.1 on White/RIDE #576

brian-kelley opened this issue Jan 22, 2020 · 2 comments
Assignees
Milestone

Comments

@brian-kelley
Copy link
Contributor

brian-kelley commented Jan 22, 2020

White and RIDE support CUDA 10.1 now. Could I add that as one of the spot-check configurations? This way, all CUDA graph stuff (which I expect will be more than just Gauss-Seidel soon) will get tested in every PR.

@ndellingwood @kyungjoo-kim

@brian-kelley brian-kelley self-assigned this Jan 22, 2020
@ndellingwood
Copy link
Contributor

Yes, and please modify the xl spot check to only test Serial on white/ride.

@ndellingwood
Copy link
Contributor

I'll have a PR up soon to add tpl support, started working on it then got pulled into other things before finalizing.

brian-kelley added a commit to brian-kelley/kokkos-kernels that referenced this issue Jan 27, 2020
brian-kelley added a commit to brian-kelley/kokkos-kernels that referenced this issue Jan 27, 2020
-Spot check only does "Serial" on IBM as OpenMP takes way too long
-Add CUDA 10.1 as a build in both spot-check and full-check
  -CUDA 9.2 and 10.1 now done in spot check
  -CUDA 9.2, 10.0 and 10.1 now done in full check
ndellingwood added a commit that referenced this issue Jan 28, 2020
Update cm_test_all_sandia on white/ride (#576)
@jjwilke jjwilke added this to the 3.1 Release milestone Mar 10, 2020
@crtrott crtrott closed this as completed Apr 14, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

4 participants