Skip to content

Merge E3SM/master into Omega/develop#361

Merged
xylar merged 855 commits intodevelopfrom
master
Mar 18, 2026
Merged

Merge E3SM/master into Omega/develop#361
xylar merged 855 commits intodevelopfrom
master

Conversation

@xylar
Copy link
Copy Markdown

@xylar xylar commented Mar 13, 2026

This brings in module updates on Aurora and Perlmutter as well as SCORPIO (among others):
E3SM-Project#8109
E3SM-Project#8113
E3SM-Project#7996

Checklist

  • Testing
    • Add a comment to the PR titled Testing with the following:
      • Which machines CTest unit tests
        have been run on and indicate that are all passing.
      • The Polaris omega_pr test suite
        has passed, using the Polaris e3sm_submodules/Omega baseline
      • Document machine(s), compiler(s), and the build path(s) used for -p for both the baseline (Polaris e3sm_submodules/Omega) and the PR build
      • Indicate "All tests passed" or document failing tests

rgknox and others added 30 commits February 16, 2026 22:07
Update AGENTS.md to specify what compsets and resolutions
to possibly create to test code changes.  Also discourage
trying make a case or build if not on a supported machine.
Update development workflow information.
Remove old e3sm.org URL for supported machines.
Add examples of query_config commands to get supported
machine and compset info
Co-authored-by: ndkeen <5684727+ndkeen@users.noreply.github.com>
Co-authored-by: ndkeen <5684727+ndkeen@users.noreply.github.com>
Mam4xx has two constants: pcnst=40 and gas_pcnst=31. Array
was dimentioned by the second but indexed by the first.

[BFB]
Update the physical constants files with the latest changes from the PCD repository.
Change the list of github workflows used for CI/CD.
Explain how to get more info about test suites.
Add an AGENTS.md file to help provide context to AI programming tools.
Also make sure Claude Code, which is almost alone in ignoring it, uses it by using
a hidden CLAUDE.md file.
copilot will use AGENTS.md but will also try to make its own so have it also include
AGENTS.md so everyone has the same context.

Add a .fortls file for the Fortran Language Server

Closes #8086
[BFB]
Adding support for data compression using HDF5. Also,
adding support for initializing the MPI library with thread
support. This change also integrates a new version of
SCORPIO (1.9.2).

* Updating CIME build system to support HDF5 (with
   and without data compression support)
* Lossless data compression support using Blosc2 + Zstd
  and lossy compression support using ZFP
* Adding HDF5 compression libraries for Frontier,
  Perlmutter and ANL compute nodes
* Updating MPAS framework to support HDF5
* Updating to SCORPIO version 1.9.2

SCORPIO version 1.9.2 includes,
* Multiple fixes related to data compression
*  Misc bug fixes

[BFB]
The identification of the vertices that are common between two grid cells
is fixed by correcting the starting value in the second do-loop.
Additionally, the MOAB-related information is now created in ELM if the code
is compiled with -DHAVE_MOAB flag.

[BFB]
Remove `-fbacktrace` from GNU C/C++ debug flags
@xylar xylar mentioned this pull request Mar 13, 2026
8 tasks
bartgol and others added 2 commits March 13, 2026 08:57
It was using the wrong macro, causing a warning and failing
to add the tests to the desired labels
Add limited support for new grids with a few aliases. The new aliases were intended to support a PPE ensemble, so the combination of land and ocean grids will not be ideal for other purposes, but other users can build off these changes to add new aliases. Also, while these changes include new atmosphere initial conditions I have omitted new new drydep_srf_file entries, which are needed to run EAMv3 on these grids.

[BFB]
@xylar
Copy link
Copy Markdown
Author

xylar commented Mar 13, 2026

I'm in the process of testing this along with #360. I didn't realize there are non-BFB changes in #328 that just went in and that I shouldn't be including in my baseline for testing this merge.

@xylar xylar mentioned this pull request Mar 13, 2026
19 tasks
@xylar
Copy link
Copy Markdown
Author

xylar commented Mar 13, 2026

Testing

CTest unit tests:

  • Machine: chrysalis
  • Compiler: intel
  • Build type: Release
  • Result: All 39 tests passed
  • Log: /lcrc/group/e3sm/ac.xylar/polaris_0.10/chrysalis/test_20260313/omega-pr-merge-e3sm-master3/build/ctests.log

Polaris omega_pr suite

  • Baseline workdir: /lcrc/group/e3sm/ac.xylar/polaris_0.10/chrysalis/test_20260313/omega-pr-baseline
  • Baseline build: /lcrc/group/e3sm/ac.xylar/polaris_0.10/chrysalis/test_20260313/omega-pr-baseline/build
  • PR build: /lcrc/group/e3sm/ac.xylar/polaris_0.10/chrysalis/test_20260313/omega-pr-merge-e3sm-master3/build
  • PR workdir: /lcrc/group/e3sm/ac.xylar/polaris_0.10/chrysalis/test_20260313/omega-pr-merge-e3sm-master3
  • Machine: chrysalis
  • Partition: compute
  • Compiler: intel
  • Build type: <Debug|Release>
  • Log: /lcrc/group/e3sm/ac.xylar/polaris_0.10/chrysalis/test_20260313/omega-pr-merge-e3sm-master3/polaris_omega_pr.o1189731
  • Result: All tests passed

Note: the baseline was with the current develop, not the Polaris Omega submodule because of non-BFB changes from #328.

@xylar xylar requested a review from sbrus89 March 13, 2026 16:51
@xylar xylar self-assigned this Mar 13, 2026
@xylar
Copy link
Copy Markdown
Author

xylar commented Mar 13, 2026

@sbrus89, could you review this (and #360)? I just want to make sure you think it's the right time to do the merge.

jgfouca and others added 6 commits March 13, 2026 14:30
Use the right macro, avoiding a cmake warning, and ensuring the tests get the desired label.

[BFB]
…(PR #8162)

Port zm_conv_mcsp_calculate_shear

Ports zm_conv_mcsp_calculate_shear in zm_conv_mcsp.F90. I think this
is the only subroutine that needs to be ported from this file.

Also, removes the state_v calculations from
zm_conv_mcsp_calculate_shear. If you look at the subroutine, the only
output is happening here mcsp_shear(i) = storm_u_shear(i), so the _v
stuff was unused.

[BFB]

* jgfouca/port_zm_conv_mcsp_calculate_shear:
  Fix
  remove unused zmid
  Changes based on copilot review
  Remove unused var
  Fixes and remove state_v stuff that was never used
  Parallelize index search
  Progress, but diffs
  AI attempt
  Port zm_conv_mcsp_calculate_shear
Fix non bfb log calls in f90 entropy.

This should fix remaining CI issues once merged and baselines are regenerated.

[BFB]
@xylar
Copy link
Copy Markdown
Author

xylar commented Mar 17, 2026

Building with gnugpu on Perlmutter, I'm seeing the errors reported in #364. I'll need to follow up on that. I don't think it's an issue of wrong modules but I could be wrong.

jgfouca added 2 commits March 17, 2026 11:24
Deep in ZM, some f90 needed to use bfb math functions.

[BFB]
Fix some subtle non-bfbness with f90

Deep in ZM, some f90 needed to use bfb math functions.

Also, make bfb math stuff callable from pure/elemental fortran functions.

This should fix the last CI ZM problem.

[BFB]
@sbrus89
Copy link
Copy Markdown
Collaborator

sbrus89 commented Mar 17, 2026

I merged #360 and this PR into develop and tested on pm-cpu:

Polaris omega_pr suite

  • Baseline workdir: /pscratch/sd/s/sbrus/polaris_vert_advection_pr_suite_baseline/
  • Baseline build: /global/homes/s/sbrus/scratch/polaris_vert_advection_pr_suite_baseline/build
  • PR build: /global/homes/s/sbrus/scratch/polaris_merge_pr_suite/build
  • PR workdir: /global/homes/s/sbrus/scratch/polaris_merge_pr_suite
  • Machine: pm-cpu
  • Compiler: gnu
  • Build type: Release
  • Log: not found
  • Result: All tests passed

@xylar
Copy link
Copy Markdown
Author

xylar commented Mar 18, 2026

@sbrus89, I'm going to take that as tacit approval if this and #360

@xylar xylar merged commit d0b3482 into E3SM-Project:develop Mar 18, 2026
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.