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

Improve performance of transeq on OpenMP backend #132

Merged
merged 1 commit into from
Dec 11, 2024

Conversation

semi-h
Copy link
Member

@semi-h semi-h commented Dec 5, 2024

Closes #66, implemented the third option as described in the issue.

@semi-h semi-h requested a review from Nanoseb December 5, 2024 19:38
Copy link
Collaborator

@Nanoseb Nanoseb left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

looking good. Just a comment on the assumption made for the fused implementation. Thinking about it, maybe if the assumption isn't met we could revert back to the older implementation.

src/omp/exec_dist.f90 Outdated Show resolved Hide resolved
src/omp/kernels/distributed.f90 Show resolved Hide resolved
@semi-h
Copy link
Member Author

semi-h commented Dec 11, 2024

looking good. Just a comment on the assumption made for the fused implementation. Thinking about it, maybe if the assumption isn't met we could revert back to the older implementation.

So the assumption is not anything new in the fused implementation, it is because of the distributed algorithm and it would result in identical behaviour with non-fused basic implementation too. @Nanoseb

@Nanoseb
Copy link
Collaborator

Nanoseb commented Dec 11, 2024

So the assumption is not anything new in the fused implementation, it is because of the distributed algorithm and it would result in identical behaviour with non-fused basic implementation too. @Nanoseb

Ah ok, sure. Then it sounds good, specially with the issue you created.

@semi-h semi-h merged commit dc9a580 into xcompact3d:main Dec 11, 2024
2 checks passed
ia267 pushed a commit to ia267/x3d2 that referenced this pull request Feb 5, 2025
Improve performance of transeq on OpenMP backend
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.

Performance improvement in transeq in the OpenMP backend
2 participants