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

Add support for CUDA 12.1 #1793

Merged
merged 4 commits into from
Mar 12, 2023
Merged

Add support for CUDA 12.1 #1793

merged 4 commits into from
Mar 12, 2023

Conversation

maleadt
Copy link
Member

@maleadt maleadt commented Mar 11, 2023

cc @amontoison, hopefully this goes smoother than the previous upgrade 🙂
I take it we should be able to remove the workarounds put in place for 12.0 now?

@maleadt maleadt added the installation CUDA is easy to install, right? label Mar 11, 2023
@amontoison
Copy link
Member

amontoison commented Mar 11, 2023

@maleadt We don't need to modify something for CUDA 12.1.0. If all tests passed that's fine 😃
I will just re-enable some tests with the version 12.1.1:

Don't forget to regenerate the wrappers.

@maleadt
Copy link
Member Author

maleadt commented Mar 11, 2023

I will just re-enable some tests with the version 12.1.1:

Let me try enabling those.

Don't forget to regenerate the wrappers.

That's not required? It's a minor version so should be backwards compatible; only if you want to use newer features you need updated headers. Ideally we'd have separate versions for each CUDA version, but that would mean we'd need to decide at precompilation time which toolkit to use, and we currently don't have that.

@amontoison
Copy link
Member

amontoison commented Mar 11, 2023

I will just re-enable some tests with the version 12.1.1:

Let me try enabling those.

Great! they solved the issue in this version 12.1.0.

Don't forget to regenerate the wrappers.

That's not required? It's a minor version so should be backwards compatible; only if you want to use newer features you need updated headers. Ideally we'd have separate versions for each CUDA version, but that would mean we'd need to decide at precompilation time which toolkit to use, and we currently don't have that.

No, it's not required but it's easier to keep track of when new features were added for the tests.
NVIDIA also deprecates some routines for more recent ones with minor version of CUDA and we can add the interfaces over time and not when we have a new breaking release.
The release notes of NVIDIA are not exhaustive.

@codecov
Copy link

codecov bot commented Mar 12, 2023

Codecov Report

Patch coverage has no change and project coverage change: -0.39 ⚠️

Comparison is base (aca936b) 60.23% compared to head (ea1d3de) 59.85%.

Additional details and impacted files
@@            Coverage Diff             @@
##           master    #1793      +/-   ##
==========================================
- Coverage   60.23%   59.85%   -0.39%     
==========================================
  Files         147      147              
  Lines       12086    12086              
==========================================
- Hits         7280     7234      -46     
- Misses       4806     4852      +46     

see 5 files with indirect coverage changes

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

@maleadt maleadt marked this pull request as ready for review March 12, 2023 12:44
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
installation CUDA is easy to install, right?
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants