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

GPU + MKL BLAS support #115

Merged
merged 1 commit into from
Mar 9, 2023
Merged

GPU + MKL BLAS support #115

merged 1 commit into from
Mar 9, 2023

Conversation

enricodetoma
Copy link
Contributor

#68 fixed GPU support, but was lost in a later commit.
With this pull request I introduced it again.

I also introduced MKL BLAS support with the flag WITH_MKL

@NeroBurner
Copy link
Collaborator

Thanks for the PR. Could you please split it into two PRs, one for MKL BLAS and one for the GULPU Support. I approve the MKL part right away, but I'm not knowledgeable enough with cuda to merge the GPU part.

Would those cuda changes be part of a newer updated upstream suitesparse release? If so I'd like to update to that instead of carrying patches to upstream code

@enricodetoma
Copy link
Contributor Author

enricodetoma commented Mar 8, 2023

I can split if you want, but I just tell you that, for CUDA, I simply cherry picked the only commit of #68 (097d746) which @jlblancoc already merged in 2018, but was probably lost when SuiteSparse was updated. I myself I'm totally not knowledgeable enough about CUDA.

Then I compiled both SuiteSparse and Ceres solver with CUDA and run all tests in Ceres solver (some of which depend on SuiteSparse) and they all succeeded.

I agree that this patch should be also part of an upstream SuiteSparse, to avoid "being lost" again when SuiteSparse is updated.

@jlblancoc jlblancoc enabled auto-merge March 9, 2023 10:16
@jlblancoc jlblancoc merged commit 5ee2eb4 into jlblancoc:master Mar 9, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants