Skip to content

Add Mfprelaxed to NVHPC builds#1651

Merged
sbryngelson merged 3 commits into
MFlowCode:masterfrom
wilfonba:NVIDIAPerfFlag
Jul 17, 2026
Merged

Add Mfprelaxed to NVHPC builds#1651
sbryngelson merged 3 commits into
MFlowCode:masterfrom
wilfonba:NVIDIAPerfFlag

Conversation

@wilfonba

Copy link
Copy Markdown
Contributor

Description

This PR adds the Mfprelaxes flag to NVHPC builds to fix a code-gen change in NVHPC 24.7 that caused a slowdown in the WENO and Riemann kernels. See https://forums.developer.nvidia.com/t/nvfortran-openacc-performance-regression-weno5-reconstruction-kernel-1-7x-slower-since-24-7-24-5-good-24-7-first-bad/376982 for details.

Type of change (delete unused ones)

  • Other (performance fix)

Testing

Relying on CI for testing. This change demonstrated the expected performance improvement with examples/3D_performance_test.

AI code reviews

Reviews are not retriggered automatically. To request a review, comment on the PR:

  • @claude full review — Claude full review (also triggers on PR open/reopen/ready)
  • Or add label claude-full-review — Claude full review via label

Copilot AI review requested due to automatic review settings July 16, 2026 22:02
@wilfonba
wilfonba requested a review from sbryngelson as a code owner July 16, 2026 22:02

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Pull request overview

Note

Copilot couldn't run its full agentic review because it didn't start before the timeout. Make sure your repository has a runner available, or add a copilot-code-review.yml file specifying one with the runs-on attribute. See the docs for more details.

Adds the NVHPC Fortran compiler flag -Mfprelaxed to mitigate a performance regression introduced in NVHPC 24.7 affecting WENO/Riemann kernels.

Changes:

  • Enable -Mfprelaxed for NVHPC Fortran compilation via CMake generator expressions.

Comment thread cmake/GPU.cmake
Comment thread cmake/GPU.cmake
sbryngelson
sbryngelson previously approved these changes Jul 17, 2026
@codecov

codecov Bot commented Jul 17, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 59.57%. Comparing base (e090d56) to head (e9632e2).
⚠️ Report is 7 commits behind head on master.

Additional details and impacted files
@@            Coverage Diff             @@
##           master    #1651      +/-   ##
==========================================
- Coverage   60.95%   59.57%   -1.39%     
==========================================
  Files          83       83              
  Lines       20003    21111    +1108     
  Branches     2983     3128     +145     
==========================================
+ Hits        12193    12576     +383     
- Misses       5782     6433     +651     
- Partials     2028     2102      +74     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

@sbryngelson
sbryngelson merged commit 1bf80cc into MFlowCode:master Jul 17, 2026
95 checks passed
@sbryngelson

Copy link
Copy Markdown
Member

good job

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Development

Successfully merging this pull request may close these issues.

3 participants