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

sw: Add DOTP benchmark #151

Merged
merged 2 commits into from
Jun 18, 2024
Merged

sw: Add DOTP benchmark #151

merged 2 commits into from
Jun 18, 2024

Conversation

mp-17
Copy link
Contributor

@mp-17 mp-17 commented Jun 11, 2024

And after all the noise around it, I share the implementation of DOTP that I am using.
Feel free to check it, fix it if needed, and add further optimizations.

Let me know if there are any mistakes.

NOTE: as of now, the kernel only supports a single cluster.

@colluca
Copy link
Collaborator

colluca commented Jun 11, 2024

Thanks @mp-17! I can take it on from here :)

Copy link
Contributor

@fischeti fischeti left a comment

Choose a reason for hiding this comment

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

Thanks Matteo! just some minor comments

sw/blas/dotp/scripts/datagen.py Outdated Show resolved Hide resolved
sw/blas/dotp/src/main.c Outdated Show resolved Hide resolved
@colluca colluca force-pushed the mperotti/dotp branch 2 times, most recently from 55b09e9 to a58acfe Compare June 12, 2024 14:07
Copy link
Contributor

@fischeti fischeti left a comment

Choose a reason for hiding this comment

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

LGTM. What would be nice is fp8, fp16, fp32 support, but we can leave this for later.

@colluca colluca merged commit dc4a6a6 into pulp-platform:main Jun 18, 2024
13 checks passed
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.

3 participants