Skip to content

Fix svs::Float16 issue #151

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

Merged
merged 2 commits into from
Jun 26, 2025
Merged

Fix svs::Float16 issue #151

merged 2 commits into from
Jun 26, 2025

Conversation

dian-lun-lin
Copy link
Member

@dian-lun-lin dian-lun-lin commented Jun 26, 2025

This PR addresses the issue where SVS uses generic arch to compile Float16 but still enable F16C. In this case, there are corner cases where large number becomes NaN.

@dian-lun-lin dian-lun-lin changed the title Fix base arch issue Fix svs::Float16 issue Jun 26, 2025
@dian-lun-lin dian-lun-lin enabled auto-merge (squash) June 26, 2025 04:50
Copy link
Member

@ibhati ibhati left a comment

Choose a reason for hiding this comment

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

I think this conversion should be fast enough without using AVX instructions as well. Should not impact much

@dian-lun-lin dian-lun-lin merged commit 9d1314c into intel:main Jun 26, 2025
11 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.

2 participants