Skip to content

Conversation

@jprhyne
Copy link
Contributor

@jprhyne jprhyne commented Oct 22, 2025

Description
When originally creating the new recursive xlarft behavior, the case of NX=1 was sufficient for performance, however after further investigation into the x{or,un}g subroutines, we found that some hardware benefited from a larger NX value, and as a result, wanted to add this behavior before finishing the other development.

This was accomplished by moving the old level 2 version of larft from the variants into the main SRC folder and renaming it to include level 2 in the name as well as adding a choice of NX based on brief testing on my Zen 4 laptop.

Checklist

  • The documentation has been updated.
  • If the PR solves a specific issue, it is set to be closed on merge.

Copy link
Contributor

@langou langou left a comment

Choose a reason for hiding this comment

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

Looks good. Thanks @jprhyne

@langou langou merged commit 7014148 into Reference-LAPACK:master Oct 23, 2025
12 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