FC5PM compset fails to buil on Titan with PGI compiler

[acme-y9s]

I tried to run ACME with the prescribed aerosol mode on Titan, but failed to built it with PGI compiler.
The last several lines from atm bld log file are
PGF90-F-0004-Unable to open MODULE file modal_aero_convproc.mod (/lustre/atlas1/cli112/proj-shared/y9s/ACME/models/atm/cam/src/physics/cam/physpkg.F90: 1747)
PGF90/x86-64 Linux 14.10-0: compilation aborted
gmake: *** [physpkg.o] Error 2
gmake: *** Waiting for unfinished jobs....

Who has succeeded in running this compset on Titan or other platforms?

[worleyph]

This is the same bug as #227, and is being addressed as part of PR #243 . This is currently in the integration phase, so should be complete soon?

[singhbalwinder]

Hi @acme-y9s ,
This has been fixed in singhbalwinder/atm/polar-mods-refactoring branch. This branch is yet to be merged but you can download this code to see if it works for you.

[acme-y9s]

Thank you, I will try and let you know the result.

[acme-y9s]

I tried to checked out the branch, but failed by the below two command lines.
git branch singhbalwinder/atm/polar-mods-refactoring --track origin/singhbalwinder/atm/polar-mods-refactoring

git checkout singhbalwinder/atm/polar-mods-refactoring

No response for the second command line.

Maybe, I do not have the permission ?

[singhbalwinder]

Not sure what is going on here. I can checkout this branch just fine. @jedbrown , @douglasjacobsen , @jgfouca , do you guys know what might be causing this?

[douglasjacobsen]

@acme-y9s: What does git log -n 1 output?

[acme-y9s]

@singhbalwinder: I can successfully build both fc5pm compset and CAM4+CLM4.5 on Titan with your branch.
However, i failed to run the CAM4+CLM4.5 due to the urban net long wave radiation error ( no convergence). For the CAM5 + CLM4.5 case, it still on queue due to the computing nodes are not working today on Titan. Will let you know whether CAM5+CLM4.5 works or not.

[bishtgautam]

@acme-y9s : @worleyph had reported a similar issue ("urban net longwave radiation error: no convergence") with the PGI a while back (see #165). Does this case runs successfully with intel on Titan?

[bishtgautam]

@acme-y9s : After a little bit more digging, I realize that singhbalwinder/atm/polar-mods-refactoring branch was forked off before PR #178, which fixes the urban net longwave radiation, was merged into master. After PR #243 is merged in mater, CAM(4/5) + CLM4.5 should work correctly with PGI on Titan.

[acme-y9s]

Thanks Gautam!

---

Xiaoying Shi
Climate Change Science Institute
Environmental Sciences Division
Building 2040, Room E222, MS-6301
Oak Ridge National Laboratory
P.O. Box 2008
Oak Ridge, TN 37831-6301
Office: 865-241-9199 Mobile: 865-804-1900
shix@ornl.govmailto:maoj@ornl.gov 865-574-9501 (fax)
[cid:0C731E85-6831-431B-9BAD-F7270C06BF7F]

From: Gautam Bisht <notifications@github.commailto:notifications@github.com>
Reply-To: ACME-Climate/ACME <reply@reply.github.commailto:reply@reply.github.com>
Date: Tuesday, July 7, 2015 at 11:11 AM
To: ACME-Climate/ACME <ACME@noreply.github.commailto:ACME@noreply.github.com>
Cc: Xiaoying Shi <shix@ornl.govmailto:shix@ornl.gov>
Subject: Re: [ACME] FC5PM compset fails to buil on Titan with PGI compiler (#263)

@acme-y9shttps://github.com/acme-y9s : After a little bit more digging, I realize that singhbalwinder/atm/polar-mods-refactoring branch was forked off before PR #178#178, which fixes the urban net longwave radiation, was merged into master. After PR #243#243 is merged in mater, CAM+CLM4.5 should work correctly with PGI on Titan.

Reply to this email directly or view it on GitHubhttps://github.com//issues/263#issuecomment-119234513.