You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Enforce velocity field masking for NormalTransportVelocity - #638
cdb935bb5a Merge pull request #600 from xylar/omega/fix-pm-mpi-flags
b90c3dae8d Merge pull request #601 from andrewdnolan/omega/cxx20-for-coupled
9641c6cf9e Merge pull request #546 from xylar/omega/fix-cf-units
55c3de8c15 Merge pull request #566: e3sm/master into omega/develop
9aba4a8032 Merge pull request #597 from xylar/omega/fix-real-literal-nvcc
PR Omega:
878c46d266 Enforce velocity field masking for NormalTransportVelocity
044014eb4d Work
135755f734 Merge pull request #438 from mwarusz/omega/submesoscale-eddies
a341e65245 Merge pull request #562 from andrewdnolan/omega/guard-division-by-zero
46aa54cd68 Merge pull request #603 from vanroekel/vanroekel/omega/kpp-bug-fix
Notes
The diffs are identical to the diffs on pm-cpu in #628 (comment)
This fixes #588. On Chrysalis with Intel, a one-year QU240 run and the 30 one-day overflow runs used for #628 (SE-RK2 and RK4) are bit-for-bit identical to #628, since the horizontal FCT provisional thickness was the only code reading the transport velocity on inactive edge layers.
HorzTracerFluxLimiterMonotonicityCheckEnable finds no out-of-bounds values in 40 days of QU240 or 9 hours of overflow, where develop has 520,389 and 23,040 in the first two steps.
Both rows differ from the baseline in the same 7 tasks. These come from this PR: its AuxiliaryState.cpp change is the only Omega source change in the test merge. #622, also merged since the Polaris pin, changes only config_machines.xml and docs, and both builds link the same METIS. They are the 7 tasks mwarusz's pm-cpu Testing comment above lists, and the merry_go_round and seamount norms match it to every printed digit. That fits this PR changing the transport velocity the horizontal FCT reads on inactive edge layers.
On craygnu, overflow/.../smoke_test_horiz_adv_order_3 has a larger normalVelocity difference (linf 2.8e-5) than on craygnu-mphipcc or pm-cpu (about 1e-12). I did not look into it further. Testing #628 on Frontier found that this nonlinear FCT test can amplify round-off on craygnu.
Posted by Claude Code on @xylar's behalf. The testing, analysis and wording above are AI-authored; please check them accordingly.
The 7 baseline differences match the Chrysalis results for #628 (#628 (comment)) task for task and in every norm. That run tested against develop at the Polaris pin. The only merge into develop since then, #622, changes just cime_config/machines/config_machines.xml and a docs page.
Posted by Claude Code on @xylar's behalf. The testing, analysis and wording above are AI-authored; please check them accordingly.
On both rows, the 7 baseline differences match the results for #628 (#628 (comment)) task for task and in every norm. That run tested against develop at the Polaris pin. The only merge into develop since then, #622, changes just cime_config/machines/config_machines.xml and a docs page.
Posted by Claude Code on @xylar's behalf. The testing, analysis and wording above are AI-authored; please check them accordingly.
On both rows, the 7 baseline differences match the results for #628 (#628 (comment)) task for task and in every norm. That run tested against develop at the Polaris pin. The only merge into develop since then, #622, changes just cime_config/machines/config_machines.xml and a docs page.
Posted by Claude Code on @xylar's behalf. The testing, analysis and wording above are AI-authored; please check them accordingly.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
bugSomething isn't workingnon-BFBPR makes roundoff changes to answers.Omega
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
We enforce the following 3-zone masking for
NormalVelocity[MinLayerEdgeTop, MaxLayerEdgeBot]0outside of[MinLayerEdgeBot, MaxLayerEdgeTop]and inside[MinLayerEdgeTop, MaxLayerEdgeBot][MinLayerEdgeBot, MaxLayerEdgeTop]The same should be true for
NormalTransportVelocity, but I did not compute it in a way that preserves this masking. This PR fixes it.Checklist
Linting
Building
Testing
aurora, intel, mpich
aurora, intelgpu, mpich
chrysalis, intel, openmpi
ERS_Ld5.T62_oQU240.GOMEGA-IAFSMS_Ld3.ne4pg2_oQU240.WCYCL1850NS-OMEGAfrontier, craygnu, mpich
frontier, craygnu-mphipcc, mpich
ERS_Ld5.T62_oQU240.GOMEGA-IAFERS_Ld3.ne4pg2_oQU240.WCYCL1850NS-OMEGApm-cpu, gnu, mpich
pm-gpu, gnugpu, mpich
Provide relevant details in a comment to the PR titled
Testingwith the following:have been run on and indicate that are all passing.
has passed, using the Polaris
e3sm_submodules/Omegabaseline-pfor both the baseline (Polarise3sm_submodules/Omega) and the PR buildFixes #588
Closes #628