Merge hash from dev with io changes for cycle.f90

[ClaraDraper-NOAA]

DESCRIPTION OF CHANGES:

Merging hash from this PR in develop into v17 production branch, to speed up v17.
The main change is to reduce the number of file copies.

TESTS CONDUCTED:

• [ x] Compile branch on all Tier 1 machines using Intel (Orion, Jet, Ursa, Hercules and WCOSS2). DONE ON URSA AND WCOSS2 - all that is needed for this branch.
• [ x] Compile branch on Ursa using GNU.
• [ x] Compile branch in 'Debug' mode on WCOSS2.
• [x ] Compile with Doxygen on any machine with no errors.
• [ x ] Run unit tests locally on any Tier 1 machine. DONE ON URSA.
• [x ] Run relevant consistency tests locally on all Tier 1 machines. DONE ON URSA, WCOSS2.

DEPENDENCIES:

Needs to be merged before g_w PR

DOCUMENTATION:

All new and updated source code must be documented with Doxygen.

• [ x] Doxygen is updated.

ISSUE:

Resolves issue NOAA-EMC/global-workflow#4533 and NOAA-EMC/global-workflow#4655

[ClaraDraper-NOAA]

Draft, while I do testing.

[ClaraDraper-NOAA]

@BrianCurtis-NOAA Many of the ctests failed on ursa, I'm guessing because this isn't the develop branch. Which tests should I be doing to merge changes into this branch?

[BrianCurtis-NOAA]

I ran utests on the previous PR OK. They should still work.

[BrianCurtis-NOAA]

I ran utests on the previous PR OK. They should still work.

I should be more specific in that i ran it on the previous PR for this production branch.

[ClaraDraper-NOAA]

I ran utests on the previous PR OK. They should still work.

I should be more specific in that i ran it on the previous PR for this production branch.

Interesting. Let me re-run it to get the output again. We're talking about the ctests run from build_all.sh, right?

[BrianCurtis-NOAA]

I ran utests on the previous PR OK. They should still work.

I should be more specific in that i ran it on the previous PR for this production branch.

Interesting. Let me re-run it to get the output again. We're talking about the ctests run from build_all.sh, right?

Yes, i'll edit build_all.sh and uncomment the ctest line. then run BUILD_TESTING=ON ./build_all.sh

[ClaraDraper-NOAA]

I ran utests on the previous PR OK. They should still work.

I should be more specific in that i ran it on the previous PR for this production branch.

Interesting. Let me re-run it to get the output again. We're talking about the ctests run from build_all.sh, right?

Yes, i'll edit build_all.sh and uncomment the ctest line. then run BUILD_TESTING=ON ./build_all.sh

Nvmd. It's silly - the tests for chgres-cubs specify an account that I'm not a member of. Fixing it now.

[BrianCurtis-NOAA]

I ran utests on the previous PR OK. They should still work.

I should be more specific in that i ran it on the previous PR for this production branch.

Interesting. Let me re-run it to get the output again. We're talking about the ctests run from build_all.sh, right?

Yes, i'll edit build_all.sh and uncomment the ctest line. then run BUILD_TESTING=ON ./build_all.sh

sorry if this is one of those did you turn it off and back on again responses, but make sure to cd into fix and run the script to populate the fix dir. I think ./link_fixdirs emc ursa

[BrianCurtis-NOAA]

I ran utests on the previous PR OK. They should still work.

I should be more specific in that i ran it on the previous PR for this production branch.

Interesting. Let me re-run it to get the output again. We're talking about the ctests run from build_all.sh, right?

Yes, i'll edit build_all.sh and uncomment the ctest line. then run BUILD_TESTING=ON ./build_all.sh

Nvmd. It's silly - the tests for chgres-cubs specify an account that I'm not a member of. Fixing it now.

OK, whew. Did have me concerned. I've done that hundreds of times by now, probably.

[ClaraDraper-NOAA]

I ran utests on the previous PR OK. They should still work.

I should be more specific in that i ran it on the previous PR for this production branch.

Interesting. Let me re-run it to get the output again. We're talking about the ctests run from build_all.sh, right?

Yes, i'll edit build_all.sh and uncomment the ctest line. then run BUILD_TESTING=ON ./build_all.sh

Nvmd. It's silly - the tests for chgres-cubs specify an account that I'm not a member of. Fixing it now.

OK, whew. Did have me concerned. I've done that hundreds of times by now, probably.

For future reference, I had to change my account here:

cmake/mpiexec.ursa

All passed now.

[ClaraDraper-NOAA]

@BrianCurtis-NOAA I've done all the URSA tests, and am about to do the WCOSS ones. Can you please check compilation of jet, orion, and hercules?

[BrianCurtis-NOAA]

@BrianCurtis-NOAA I've done all the URSA tests, and am about to do the WCOSS ones. Can you please check compilation of jet, orion, and hercules?

Do you run production testing on jet/orion/hercules?

[ClaraDraper-NOAA]

@CoryMartin-NOAA and jiaruidong2017. Since you both reviewed the develop PR, could you please look at this one too?

[ClaraDraper-NOAA]

@BrianCurtis-NOAA I've done all the URSA tests, and am about to do the WCOSS ones. Can you please check compilation of jet, orion, and hercules?

Do you run production testing on jet/orion/hercules?

I missed that that needed to be done. I don't have access to those machines. I can do WCOSS though.

[BrianCurtis-NOAA]

@BrianCurtis-NOAA I've done all the URSA tests, and am about to do the WCOSS ones. Can you please check compilation of jet, orion, and hercules?

Do you run production testing on jet/orion/hercules?

I missed that that needed to be done. I don't have access to those machines. I can do WCOSS though.

That was a question to global-workflow. Are you using this production/GFS.v17.0.0 branch on jet or hercules or orion that I would want to test this there? I was under the impression that this is not the case.

[ClaraDraper-NOAA]

@BrianCurtis-NOAA I've done all the URSA tests, and am about to do the WCOSS ones. Can you please check compilation of jet, orion, and hercules?

Do you run production testing on jet/orion/hercules?

I missed that that needed to be done. I don't have access to those machines. I can do WCOSS though.

That was a question to global-workflow. Are you using this production/GFS.v17.0.0 branch on jet or hercules or orion that I would want to test this there? I was under the impression that this is not the case.

I have no idea! @CatherineThomas-NOAA ? Do we need to run production/GFS.v17.0.0 on jet/hercules/orion?

[ClaraDraper-NOAA]

all required tests have passed, ready for review.

@BrianCurtis-NOAA Jiarui (jiaruidong2017.) wasn't able to add his review. Do you need to invite him first?

[BrianCurtis-NOAA]

all required tests have passed, ready for review.

@BrianCurtis-NOAA Jiarui (jiaruidong2017.) wasn't able to add his review. Do you need to invite him first?

If Jiarui will be regularly reviewing work, then I can add them.

[ClaraDraper-NOAA]

all required tests have passed, ready for review.
@BrianCurtis-NOAA Jiarui (jiaruidong2017.) wasn't able to add his review. Do you need to invite him first?

If Jiarui will be regularly reviewing work, then I can add them.

He will, so that's probably a good idea.

[CatherineThomas-NOAA]

Do we need to run production/GFS.v17.0.0 on jet/hercules/orion?

No, just WCOSS2 and Gaea-C6.

[BrianCurtis-NOAA]

I will work on adding Gaea C6 support to develop branch soon, but I haven't tested anything on there yet. Once it comes back online i'll start, I think it's down today.

[ClaraDraper-NOAA]

All tests have passed. This is ready to merge, once we get the second review.

[BrianCurtis-NOAA]

Approving based on branch maintainers OK.