search

E3SM-Project / E3SM

Energy Exascale Earth System Model source code. NOTE: use "maint" branches for your work. Head of master is not validated.
https://docs.e3sm.org/E3SM
Other
341 stars 343 forks source link

HOMMEBFB_P24.f19_g16_rx1.A.pm-cpu failing #6478

Open rljacob opened 1 month ago

rljacob commented 1 month ago

HOMMEBFB_P24.f19_g16_rx1.A.pm-cpu is failing for both gnu and intel on perlmutter cpu.

Latest cdash output: intel https://my.cdash.org/tests/171993260 gnu https://my.cdash.org/tests/172059238

rljacob commented 1 month ago

Now the other test is timing out. @ndkeen do you know what's going on?

ndkeen commented 1 month ago

Quick inspections -- the other test, not the HOMMEBFB test -- looks like it completed.

Running manually, it seems ok HOMME_P24.f19_g16_rx1.A.pm-cpu_intel

mt5555 commented 4 weeks ago

The HOMMEBFB test is not expected to work - it needs some work creating a cmake config with the low optimization flags turned on to get C++ and Fortran to match. @oksanaguba has this on her todo list.

I think we should continue to run the suite (to ensure the HOMME_P24 case passes) because it tests the HOMME specific parts of the topo generation tool chain, which is needed on PM for generating new atmspheric grids.

If the failures for HOMMEBFB are distracting, could we modify jenkens to run just the HOMME_P24 test and not the HOMME integration suite?