LILAC fails at build phase because of some changes needed for the rpointer handling

[ekluzek]

Brief summary of bug

ctsm5.3.015 includes #2757 which includes changes to lnd_comp_nuopc.F90 that need to be moved to lnd_comp_esmf.F90 for LILAC.

https://github.com/ESCOMP/CTSM/pull/2757/files#diff-f2434c2fccfe6f177f22d26b605bd0f9698f51e85348e9cfdd8a13e4a66a09da

General bug information

CTSM version you are using: ctsm5.3.015 (to be made)
Does this bug cause significantly incorrect results in the model's science? No

Configurations affected: LILAC

Important details of your setup / configuration so we can reproduce the bug

This test:

LILACSMOKE_D_Ld2.f10_f10_mg37.I2000Ctsm50NwpSpAsRs.derecho_intel.clm-lilac

fails at the BUILD Step because of this.

[wwieder]

Thanks for creating this issue, @ekluzek. I'm curious what we're thinking about the prioritization of maintaining LILAC functionality vs. holding up the rpointer CESM tag? @billsacks, @dlawrenncar, @briandobbins do you have thought here?

Ultimately, I'm guessing it likely comes down to effort. Do you have an estimate for how big a lift is to to address this issue, Erik?

[briandobbins]

For the immediate future, we're going to get the rpointer tag in at the expense of LILAC - this is for two primary reasons. One, the CTSM tag is needed for the CAM tag, so things are piling up on their side, and two, we need to make another tag by the end of the year, though admittedly one with just a one-line change to input data. That's on a hard deadline due to an NCAR IT rule.

Other than that, I think it's entirely up to how things look for the LMWG and CESM3. I expect we'll have time, but you know better than I do here, Will. I'd even say if it's just two hours or so, we can do it now, I just want to make sure we're not in the 'days' territory right now, given how fast the holidays are coming up and that end-of-year deadline.

[dlawrenncar]

Agree that we shouldn’t slow down getting cam/CTSM tags completed to deal with this side effect on lilac.

…

On Wed, Dec 11, 2024 at 8:30 AM briandobbins ***@***.***> wrote: For the immediate future, we're going to get the rpointer tag in at the expense of LILAC - this is for two primary reasons. One, the CTSM tag is needed for the CAM tag, so things are piling up on their side, and two, we need to make *another* tag by the end of the year, though admittedly one with just a one-line change to input data. That's on a hard deadline due to an NCAR IT rule. Other than that, I think it's entirely up to how things look for the LMWG and CESM3. I expect we'll have time, but you know better than I do here, Will. I'd even say if it's just two hours or so, we can do it now, I just want to make sure we're not in the 'days' territory right now, given how fast the holidays are coming up and that end-of-year deadline. — Reply to this email directly, view it on GitHub <#2905 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AFABYVHR6ZJKBYYTHHMGU5D2FA5APAVCNFSM6AAAAABTMH4STKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKMZVHE4TOOJQGE> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[ekluzek]

This was fixed in #2912 and will come in with b4b-dev moving to master.