update Transmute for conv_gr_kernel_mod

[Benjamin Went (MetBenjaminWent)]

PR Summary

With the existing Transmute script for the conv_gr_kernel_mod, it seems we are missing coverage, which is affecting perfomance.

The script is also very complex, and it could be simpler.

The segment loop in this kernel is extremely important, and core of the existing transmute script. If this is not parallelised with OMP, the entire call stack looses OMP threading coverage. Before these changes, it was not getting OMP coverage, which would explain the issue.

By swapping to using the kernel local script, which puts OMP around all of the i loops, the coverage has improved, and therefore the OMP performance scaling.

Before (Red) and Post (Yellow) changes:

I expect the 1T is the OMP overhead, but this may need further investigations.

Sci/Tech Reviewer: Hacka Fett (@christophermaynard)
Code Reviewer: Oakley Brunt (@oakleybrunt)

PR: #583
Umbrella: #424

Code Quality Checklist

• I have performed a self-review of my own code
• My code follows the project's style guidelines
• Comments have been included that aid understanding and enhance the readability of the code
• My changes generate no new warnings
• All automated checks in the CI pipeline have completed successfully

Testing

• I have tested this change locally, using the LFRic Apps rose-stem suite
• If any tests fail (rose-stem or CI) the reason is understood and acceptable (e.g. kgo changes)
• I have added tests to cover new functionality as appropriate (e.g. system tests, unit tests, etc.)
• Any new tests have been assigned an appropriate amount of compute resource and have been allocated to an appropriate testing group (i.e. the developer tests are for jobs which use a small amount of compute resource and complete in a matter of minutes)

trac.log

Test Suite Results - lfric_apps - omp_kernels_grc/run5

Suite Information

Item	Value
Suite Name	omp_kernels_grc/run5
Suite User	benjamin.went
Workflow Start	2026-07-02T09:26:07
Groups Run	ex1a_omp_developer

Dependency	Reference	Main Like
casim	MetOffice/casim@2026.03.2	True
jules	MetOffice/jules@1ea67b4	True
lfric_apps	MetBenjaminWent/lfric_apps@omp_kernels_grc	False
lfric_core	MetOffice/lfric_core@b85ebad	True
moci	MetOffice/moci@2026.03.2	True
SimSys_Scripts	MetOffice/SimSys_Scripts@3e2998b	True
socrates	MetOffice/socrates@2026.03.2	True
socrates-spectral	MetOffice/socrates-spectral@2026.03.2	True
ukca	MetOffice/ukca@1cdb9c2	True

Task Information

✅ succeeded tasks - 51

Test Suite Results - lfric_apps - omp_kernels_grc/run6

Suite Information

Item	Value
Suite Name	omp_kernels_grc/run6
Suite User	benjamin.went
Workflow Start	2026-07-02T10:19:07
Groups Run	all

Dependency	Reference	Main Like
casim	MetOffice/casim@2026.03.2	True
jules	MetOffice/jules@1ea67b4	True
lfric_apps	MetBenjaminWent/lfric_apps@omp_kernels_grc	False
lfric_core	MetOffice/lfric_core@b85ebad	True
moci	MetOffice/moci@2026.03.2	True
SimSys_Scripts	MetOffice/SimSys_Scripts@3e2998b	True
socrates	MetOffice/socrates@2026.03.2	True
socrates-spectral	MetOffice/socrates-spectral@2026.03.2	True
ukca	MetOffice/ukca@1cdb9c2	True

Task Information

❌ failed tasks - 1

Task	State
run_shallow_water_galewsky_vi_ffsl-C96_ex1a_gnu_fast-debug-64bit	failed

✅ succeeded tasks - 1590

Test Suite Results - lfric_apps - omp_kernels_grc/run7

Suite Information

Item	Value
Suite Name	omp_kernels_grc/run7
Suite User	benjamin.went
Workflow Start	2026-07-02T12:12:38
Groups Run	ex1a_omp_C192_cce_production

Dependency	Reference	Main Like
casim	MetOffice/casim@2026.03.2	True
jules	MetOffice/jules@1ea67b4	True
lfric_apps	MetBenjaminWent/lfric_apps@omp_kernels_grc	False
lfric_core	MetOffice/lfric_core@b85ebad	True
moci	MetOffice/moci@2026.03.2	True
SimSys_Scripts	MetOffice/SimSys_Scripts@3e2998b	True
socrates	MetOffice/socrates@2026.03.2	True
socrates-spectral	MetOffice/socrates-spectral@2026.03.2	True
ukca	MetOffice/ukca@1cdb9c2	True

Task Information

✅ succeeded tasks - 11

Security Considerations

• I have reviewed my changes for potential security issues
• Sensitive data is properly handled (if applicable)
• Authentication and authorisation are properly implemented (if applicable)

Performance Impact

• Performance of the code has been considered and, if applicable, suitable performance measurements have been conducted

AI Assistance and Attribution

• Some of the content of this change has been produced with the assistance of Generative AI tool name (e.g., Met Office Github Copilot Enterprise, Github Copilot Personal, ChatGPT GPT-4, etc) and I have followed the Simulation Systems AI policy (including attribution labels)

Documentation

• Where appropriate I have updated documentation related to this change and confirmed that it builds correctly

PSyclone Approval

• If you have edited any PSyclone-related code (e.g. PSyKAl-lite, Kernel interface, optimisation scripts, LFRic data structure code) then please contact the TCD Team

Sci/Tech Review

• I understand this area of code and the changes being added
• The proposed changes correspond to the pull request description
• Documentation is sufficient (do documentation papers need updating)
• Sufficient testing has been completed

(Please alert the code reviewer via a tag when you have approved the SR)

Code Review

• All dependencies have been resolved
• Related Issues have been properly linked and addressed
• CLA compliance has been confirmed
• Code quality standards have been met
• Tests are adequate and have passed
• Documentation is complete and accurate
• Security considerations have been addressed
• Performance impact is acceptable

[Benjamin Went (MetBenjaminWent)]

Segments don't appear to need OMP across (existing script):

[Benjamin Went (MetBenjaminWent)]

Segments don't appear to need OMP across (existing script):

The loop here is very important, and core of the existing transmute script. If this is not parallelised with OMP, the entire call stack looses OMP threading coverage.......

[Oakley Brunt (oakleybrunt)]

Thanks Ben :D

[Hacka Fett (christophermaynard)]

The changes simplify the transmute scripts and crucially add OpenMP around the main
segmented loop over the domain. Previously, this was missing and so the main loop wasn't parallel. It is with these changes. Approved.

[Benjamin Went (MetBenjaminWent)]

Since move to Psyclone 3.3.1, we need fixes for this PR in #607 (which also contains this PR).

[Oakley Brunt (oakleybrunt)]

CR approved

[Benjamin Went (MetBenjaminWent)]

Note, work in #607 has highlighted that the options list no longer works for ignore_dependencies_for for PSyclone 3.3.1, so this script would continue to not effectively function!

I've pulled in the fixes from #607, so that this will now work with Psyclone 3.3.1