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

Fix tuning timing results #3460

Merged
merged 6 commits into from
Jan 11, 2025
Merged

Conversation

BrianHarrisonAMD
Copy link
Contributor

Tuning timing results for CK solvers were not timing the same as other solvers which lead to differences in results.
This is due to timing based off a scoped start/stop event which includes the CPU timing, and also includes the other start/stop events synchronizing which gives improper timing results (leading to higher overall runtime compared to other solvers).

There is also an additional issue where only the fastest solver within a family of solvers would record it's timing values.
The combination of these issues made it so the MISA solvers were being selected, and hiding the other potential CK solvers.

@BrianHarrisonAMD BrianHarrisonAMD marked this pull request as ready for review January 8, 2025 00:27
@BrianHarrisonAMD BrianHarrisonAMD requested a review from cderb January 8, 2025 16:57
@BradPepersAMD BradPepersAMD merged commit d94ef1f into develop Jan 11, 2025
2 of 124 checks passed
@BradPepersAMD BradPepersAMD deleted the bharriso/fix-tuning-timing-results branch January 11, 2025 22:00
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants