[production/RRFS.v1] Fix 'check all' using 0-sized arrays by grantfirl · Pull Request #2158 · ufs-community/ufs-weather-model

grantfirl · 2024-02-27T22:18:27Z

Commit Queue Requirements:

Fill out all sections of this template.
All sub component pull requests have been reviewed by their code managers.
Run the full Intel+GNU RT suite (compared to current baselines) on either Hera/Derecho/Hercules
[N/A] Commit 'test_changes.list' from previous step

Description:

This PR at least partially addresses #2023 (at least any runtime errors associated with the CCPP). The strategy is to simply make sure that every variable is allocated, although for variables that had been conditionally-allocated to save memory, they are now being allocated with zero size. This has a ripple effect in other repos where one needs to change from checking whether a variable is allocated to whether the size is > 0.

The change to Intel.cmake was used for testing only and I can remove this if we're not ready to permanently add it.

Commit Message:

* UFSWM - Add zero-size variable allocations to pass the `-check all` compiler option for CCPP-related code.
  * AQM - 
  * CDEPS - 
  * CICE - 
  * CMEPS - 
  * CMakeModules - 
  * FV3 - Add zero-size variable allocations to pass the `-check all` compiler option for CCPP-related code.
    * ccpp-physics - Add zero-size variable allocations to pass the `-check all` compiler option for CCPP-related code.
    * atmos_cubed_sphere - Add zero-size variable allocations to pass the `-check all` compiler option for CCPP-related code.
  * GOCART - 
  * HYCOM - 
  * MOM6 - 
  * NOAHMP - 
  * WW3 - 
  * stochastic_physics -

Priority:

High: Needed for RRFS.v1 release

Git Tracking

UFSWM:

Partially addresses Add -check all -check noarg_temp_created -ftrapuv to all DEBUG build flags of UFSWM and subcomponents #2023

Sub component Pull Requests:

FV3: [production/RRFS.v1] Fix 'check all' using 0-sized arrays NOAA-EMC/ufsatm#789
- ccpp-physics: [production/RRFS.v1] Fix 'check all' with 0-size allocations ccpp-physics#178
- ccpp-framework: Dummy-allocate inactive local variables NCAR/ccpp-framework#536
- atmos_cubed_sphere: [production/RRFS.v1] Fix 'check all' using 0-sized allocations NOAA-EMC/GFDL_atmos_cubed_sphere#86

UFSWM Blocking Dependencies:

None

Changes

Regression Test Changes (Please commit test_changes.list):

No Baseline Changes.

Input data Changes:

None.

Library Changes/Upgrades:

No Updates

Testing Log:

…all compilation flag for debug mode

grantfirl · 2024-02-27T22:19:41Z

Replaces #2102 and this is targeted at the RRFS release branch.

grantfirl · 2024-02-27T22:29:35Z

Retesting on Hera. Will post results when finished.

grantfirl · 2024-02-27T22:33:33Z

@jkbk2004 I kicked off the tests thinking that the baseline from where the RRFS branch was cut in the history would still be valid, but I see from #2147 that you're creating a new one. I'll retest using that baseline when it's ready.

grantfirl · 2024-02-28T16:53:23Z

@jkbk2004 My RTs were successful (no BL changes), but that was comparing against the saved baselines from the develop branch.

grantfirl · 2024-02-28T17:52:49Z

@jkbk2004 I'm re-running RTs pointing at the new RRFS baselines for 20240227. Just to confirm, those new baselines do not include the changes in #2147, right?

grantfirl · 2024-02-28T20:20:21Z

Throughput on Hera is slow for our account right now. It looks like it'll be a while.

Updates list of tests to run for RRFSv1 Adds wcoss2 regression test log

…ll_zero_alloc_RRFS

grantfirl · 2024-03-01T15:52:50Z

@jkbk2004 Hi re-running RTs again because the last time that I did, there were unexpected failures, probably because the RRFS baselines already contained #2147.

…oc_RRFS

…b.com/grantfirl/ufs-weather-model into feature/fix_check_all_zero_alloc_RRFS

jkbk2004 · 2024-03-02T18:02:01Z

@grantfirl I checked code base of production/rrfs.v1 branch. All looks good. On orion, baseline was reproduced ok. Somehow baseline files look screwed up on hera. I restored baselines and cleaned up to go only with rt.conf_rrfs. @MatthewPyle-NOAA hera/orion/hercules tests are done. This PR will be ready once you confirm on WCOSS2 test

MatthewPyle-NOAA · 2024-03-04T14:17:18Z

I had a lot of tests fail on WCOSS, which really didn't make sense to me. I'm recreating the baselines somewhat differently than I did the first time (using a -b rt.conf_rrfs option this time)

BrianCurtis-NOAA · 2024-03-04T14:19:50Z

I had a lot of tests fail on WCOSS, which really didn't make sense to me. I'm recreating the baselines somewhat differently than I did the first time (using a -b rt.conf_rrfs option this time)

You would want ./rt.sh -e -c -l rt.conf_rrfs

jkbk2004 · 2024-03-04T14:26:35Z

@MatthewPyle-NOAA I checked production/RRFS.v1 code base are ok during the merge process of #2147. So we are ok if you need to re-create baseline on wcoss. Sounds like similar things happened on hera.

grantfirl · 2024-03-04T15:21:22Z

@jkbk2004 @MatthewPyle-NOAA @BrianCurtis-NOAA I'm glad that tests are passing on Hera. This saves me from a "wild goose chase" of trying to figure out the failures that I was getting. My next course of action was to produce my own baseline from the top of production/RRFS.v1 and use that, but it sounds like everything is working OK now.

MatthewPyle-NOAA · 2024-03-04T16:30:22Z

I'm good with it now on WCOSS.

MatthewPyle-NOAA · 2024-03-04T19:25:37Z

@jkbk2004 Let me know if you need more from me at this stage. I do have a WCOSS log to add eventually.

jkbk2004 · 2024-03-04T19:42:42Z

@jkbk2004 Let me know if you need more from me at this stage. I do have a WCOSS log to add eventually.

@MatthewPyle-NOAA sure! go ahead to push wcoss log. then we can move on to merge/

MatthewPyle-NOAA · 2024-03-04T19:49:46Z

@jkbk2004 Let me know if you need more from me at this stage. I do have a WCOSS log to add eventually.

@MatthewPyle-NOAA sure! go ahead to push wcoss log. then we can move on to merge/

@jkbk2004 Where/how do I push it? I tested with Grant's feature branch, but don't have the permission to push to his space.

jkbk2004 · 2024-03-04T19:53:21Z

@jkbk2004 Let me know if you need more from me at this stage. I do have a WCOSS log to add eventually.

@MatthewPyle-NOAA sure! go ahead to push wcoss log. then we can move on to merge/

@jkbk2004 Where/how do I push it? I tested with Grant's feature branch, but don't have the permission to push to his space.

@grantfirl can you add Mathew for the permission to your repo?

grantfirl · 2024-03-04T20:01:39Z

@jkbk2004 Let me know if you need more from me at this stage. I do have a WCOSS log to add eventually.

@MatthewPyle-NOAA sure! go ahead to push wcoss log. then we can move on to merge/

@jkbk2004 Where/how do I push it? I tested with Grant's feature branch, but don't have the permission to push to his space.

@grantfirl can you add Mathew for the permission to your repo?

I've added @MatthewPyle-NOAA as a collaborator on my fork, so he should have permission to push now.

MatthewPyle-NOAA · 2024-03-04T20:03:11Z

Thanks @grantfirl WCOSS log file has been pushed now.

grantfirl · 2024-03-04T20:06:34Z

@jkbk2004 @MatthewPyle-NOAA Are we ready to start merging? I can do ccpp-physics and ccpp-framework, but nobody has reviewed NOAA-EMC/GFDL_atmos_cubed_sphere#86. NOAA-GFDL/GFDL_atmos_cubed_sphere#315 was approved, but I opened a new PR when I changed the target and rebased.

They both contain the same simple change, so I imagine that the new PR should have implicit approval.

grantfirl · 2024-03-04T22:06:51Z

OK, NOAA-EMC/GFDL_atmos_cubed_sphere#86 was approved. I think that we're ready to merge.

jkbk2004 · 2024-03-05T15:02:00Z

@MatthewPyle-NOAA It will be good idea to clean up commit messages when squash/merge. Do you want me to merge this pr?

MatthewPyle-NOAA · 2024-03-05T15:03:20Z

I can merge it after cleaning up the merge message.

haiqinli and others added 4 commits February 23, 2024 04:05

"point to the physics updates for RRFS.v1 code freeze"

c2388f5

"commit the regression test log file"

a3818c9

update FV3 submodule pointer, .gitmodules for testing, and add check …

ddff1cb

…all compilation flag for debug mode

Update rt.sh: RRFS.v1

f17e9de

grantfirl mentioned this pull request Feb 27, 2024

Fix -check all runtime errors (for CCPP) using zero-size allocations #2102

Closed

41 tasks

Update bl_date.conf

515721d

grantfirl marked this pull request as ready for review February 27, 2024 22:29

jkbk2004 added 3 commits February 27, 2024 19:34

Update rt.sh: DISKNM RRFS.v1

1e08ed0

add hera RT log: passed

3479b51

Update rt.sh: hercules DISKNM RRFS.v1

abd6382

jkbk2004 mentioned this pull request Feb 28, 2024

[production/RRFS.v1] physics updates for RRFS.v1 code freeze #2147

Merged

14 tasks

grantfirl changed the title ~~Fix 'check all' using 0-sized arrays~~ [production/RRFS.v1] Fix 'check all' using 0-sized arrays Feb 28, 2024

grantfirl mentioned this pull request Feb 28, 2024

Dummy-allocate inactive local variables NCAR/ccpp-framework#536

Merged

MatthewPyle-NOAA and others added 3 commits February 29, 2024 21:30

Update rt.sh for wcoss2 disk location

fcd292c

Updates list of tests to run for RRFSv1 Adds wcoss2 regression test log

add orion/hercules RT log: passed

22bb242

Merge branch 'production/RRFS.v1-codefreeze' into feature/fix_check_a…

5c1ded6

…ll_zero_alloc_RRFS

grantfirl changed the base branch from production/RRFS.v1 to develop March 1, 2024 15:41

grantfirl changed the base branch from develop to production/RRFS.v1 March 1, 2024 15:41

grantfirl and others added 5 commits March 1, 2024 11:15

Merge branch 'production/RRFS.v1' into feature/fix_check_all_zero_all…

1ea65e0

…oc_RRFS

fix typo in Intel.cmake"

d3a1f3a

Merge branch 'feature/fix_check_all_zero_alloc_RRFS' of https://githu…

11e0e60

…b.com/grantfirl/ufs-weather-model into feature/fix_check_all_zero_alloc_RRFS

Update RegressionTests_hera.log: passed

f7ee839

Update RegressionTests_orion.log: passed

94db8f6

Update RegressionTests_hercules.log: passed

1e7f596

Adds wcoss2 RT file

d1c4634

jkbk2004 mentioned this pull request Mar 4, 2024

[production/RRFS.v1] Fix 'check all' using 0-sized allocations NOAA-EMC/GFDL_atmos_cubed_sphere#86

Merged

6 tasks

jkbk2004 mentioned this pull request Mar 5, 2024

[production/RRFS.v1] Fix 'check all' with 0-size allocations ufs-community/ccpp-physics#178

Merged

update FV3 submodule pointer and revert .gitmodules

74b54c3

jkbk2004 requested a review from MatthewPyle-NOAA March 5, 2024 15:00

MatthewPyle-NOAA approved these changes Mar 5, 2024

View reviewed changes

MatthewPyle-NOAA merged commit d53f87a into ufs-community:production/RRFS.v1 Mar 5, 2024

jkbk2004 approved these changes Mar 5, 2024

View reviewed changes

grantfirl mentioned this pull request Oct 25, 2024

Second reconciliation PR from production/RRFS.v1 #2481

Closed

14 tasks

Conversation

grantfirl commented Feb 27, 2024 • edited by jkbk2004 Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Commit Queue Requirements:

Description:

Commit Message:

Priority:

Git Tracking

UFSWM:

Sub component Pull Requests:

UFSWM Blocking Dependencies:

Changes

Regression Test Changes (Please commit test_changes.list):

Input data Changes:

Library Changes/Upgrades:

Testing Log:

Uh oh!

grantfirl commented Feb 27, 2024

Uh oh!

grantfirl commented Feb 27, 2024

Uh oh!

grantfirl commented Feb 27, 2024

Uh oh!

grantfirl commented Feb 28, 2024

Uh oh!

grantfirl commented Feb 28, 2024

Uh oh!

grantfirl commented Feb 28, 2024

Uh oh!

grantfirl commented Mar 1, 2024

Uh oh!

jkbk2004 commented Mar 2, 2024

Uh oh!

MatthewPyle-NOAA commented Mar 4, 2024

Uh oh!

BrianCurtis-NOAA commented Mar 4, 2024

Uh oh!

jkbk2004 commented Mar 4, 2024

Uh oh!

grantfirl commented Mar 4, 2024

Uh oh!

MatthewPyle-NOAA commented Mar 4, 2024

Uh oh!

MatthewPyle-NOAA commented Mar 4, 2024

Uh oh!

jkbk2004 commented Mar 4, 2024

Uh oh!

MatthewPyle-NOAA commented Mar 4, 2024

Uh oh!

jkbk2004 commented Mar 4, 2024

Uh oh!

grantfirl commented Mar 4, 2024

Uh oh!

MatthewPyle-NOAA commented Mar 4, 2024

Uh oh!

grantfirl commented Mar 4, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

grantfirl commented Mar 4, 2024

Uh oh!

jkbk2004 commented Mar 5, 2024

Uh oh!

MatthewPyle-NOAA commented Mar 5, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

grantfirl commented Feb 27, 2024 •

edited by jkbk2004

Loading

grantfirl commented Mar 4, 2024 •

edited

Loading