Refactoring tracers butzin second try #576

JanStreffing · 2024-03-14T14:41:04Z

superseds: #572 build via patch, rather than rebase. Reduction of MR from +8,685 −150 to +700 −7 lines

…t_bc_surface', and remove the now obsolete transient tracer code from function 'bc_surface'.

mbutzin · 2024-03-19T15:43:41Z

superseds: #572 build via patch, rather than rebase. Reduction of MR from +8,685 −150 to +700 −7 lines

After my latest commits we now have version which, at least, can be successfully compiled. Whether the results will make sense is another question ..

JanStreffing · 2024-03-19T15:55:28Z

superseds: #572 build via patch, rather than rebase. Reduction of MR from +8,685 −150 to +700 −7 lines

After my latest commits we now have version which, at least, can be successfully compiled. Whether the results will make sense is another question ..

I guess you are compiling with intel. The GNU based test case still fails:


[ 99%] Building Fortran object src/CMakeFiles/fesom.dir/oce_setup_step.F90.o
cd /__w/fesom2/fesom2/build/src && /usr/bin/mpifort -DMETISRANDOMSEED=35243 -DMETIS_VERSION=5 -DPARMS -DPART_WEIGHTED -D__async_icebergs -I/usr/include -I/__w/fesom2/fesom2/lib/parms/src/../include -I/__w/fesom2/fesom2/src/async_threads_cpp -I/__w/fesom2/fesom2/build/src/async_threads_cpp  -g   -O2 -g -ffloat-store -finit-local-zero -finline-functions -fimplicit-none -fdefault-real-8 -ffree-line-length-none -c /__w/fesom2/fesom2/src/oce_setup_step.F90 -o CMakeFiles/fesom.dir/oce_setup_step.F90.o
/__w/fesom2/fesom2/src/oce_setup_step.F90:398:6:

  398 |       idlist((n_ic3d+1):(n_ic3d+3)) = (/14/)
      |      1
Error: Different shape for array assignment at (1) on dimension 1 (3 and 1)
/__w/fesom2/fesom2/src/oce_setup_step.F90:399:6:

  399 |       filelist((n_ic3d+1):(n_ic3d+3)) = (/'R14C.nc'/)
      |      1
Error: Different shape for array assignment at (1) on dimension 1 (3 and 1)
/__w/fesom2/fesom2/src/oce_setup_step.F90:400:6:

  400 |       varlist((n_ic3d+1):(n_ic3d+3))  = (/'R14C'/)
      |      1
Error: Different shape for array assignment at (1) on dimension 1 (3 and 1)
make[2]: *** [src/CMakeFiles/fesom.dir/build.make:1425: src/CMakeFiles/fesom.dir/oce_setup_step.F90.o] Error 1
make[2]: Leaving directory '/__w/fesom2/fesom2/build'
make[1]: *** [CMakeFiles/Makefile2:142: src/CMakeFiles/fesom.dir/all] Error 2
make[1]: Leaving directory '/__w/fesom2/fesom2/build'
make: *** [Makefile:133: all] Error 2
Error: Process completed with exit code 2.

mbutzin · 2024-03-19T16:00:44Z

superseds: #572 build via patch, rather than rebase. Reduction of MR from +8,685 −150 to +700 −7 lines

After my latest commits we now have version which, at least, can be successfully compiled. Whether the results will make sense is another question ..

I guess you are compiling with intel. The GNU based test case still fails:


[ 99%] Building Fortran object src/CMakeFiles/fesom.dir/oce_setup_step.F90.o
cd /__w/fesom2/fesom2/build/src && /usr/bin/mpifort -DMETISRANDOMSEED=35243 -DMETIS_VERSION=5 -DPARMS -DPART_WEIGHTED -D__async_icebergs -I/usr/include -I/__w/fesom2/fesom2/lib/parms/src/../include -I/__w/fesom2/fesom2/src/async_threads_cpp -I/__w/fesom2/fesom2/build/src/async_threads_cpp  -g   -O2 -g -ffloat-store -finit-local-zero -finline-functions -fimplicit-none -fdefault-real-8 -ffree-line-length-none -c /__w/fesom2/fesom2/src/oce_setup_step.F90 -o CMakeFiles/fesom.dir/oce_setup_step.F90.o
/__w/fesom2/fesom2/src/oce_setup_step.F90:398:6:

  398 |       idlist((n_ic3d+1):(n_ic3d+3)) = (/14/)
      |      1
Error: Different shape for array assignment at (1) on dimension 1 (3 and 1)
/__w/fesom2/fesom2/src/oce_setup_step.F90:399:6:

  399 |       filelist((n_ic3d+1):(n_ic3d+3)) = (/'R14C.nc'/)
      |      1
Error: Different shape for array assignment at (1) on dimension 1 (3 and 1)
/__w/fesom2/fesom2/src/oce_setup_step.F90:400:6:

  400 |       varlist((n_ic3d+1):(n_ic3d+3))  = (/'R14C'/)
      |      1
Error: Different shape for array assignment at (1) on dimension 1 (3 and 1)
make[2]: *** [src/CMakeFiles/fesom.dir/build.make:1425: src/CMakeFiles/fesom.dir/oce_setup_step.F90.o] Error 1
make[2]: Leaving directory '/__w/fesom2/fesom2/build'
make[1]: *** [CMakeFiles/Makefile2:142: src/CMakeFiles/fesom.dir/all] Error 2
make[1]: Leaving directory '/__w/fesom2/fesom2/build'
make: *** [Makefile:133: all] Error 2
Error: Process completed with exit code 2.

Correct, intel@albedo.

mbutzin · 2024-03-19T16:10:16Z

(n_ic3d+1):(n_ic3d+3)
I think this should read
(n_ic3d+1):(n_ic3d+1)
and pushed a corresponding commit ... What does GNU say?

JanStreffing · 2024-03-19T19:56:56Z

Good progress! The test case compiles, at runtime there is still an issue though. Btw, you can check the tests below this post. I've been copying and pasting the error from there.

  --> FESOM STARTS TIME LOOP                                 
 Updating SSS restoring data for month            1
  --check opt_visc--> Mean Ratio Resol/Rrossby =    5.0000000000000000     

Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

Backtrace for this error:

Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

Backtrace for this error:
#0  0xf20170b5d21 in ???
#1  0xf20170b4ef5 in ???
#2  0xf2016d0d08f in ???
#0  0x148098b00d21 in ???
#1  0x148098affef5 in ???
#2  0x14809875808f in ???
#3  0x5abd48913847 in diff_tracers_ale_
	at /__w/fesom2/fesom2/src/associate_mesh_ass.h:33
#4  0x5abd48912800 in solve_tracers_ale_
	at /__w/fesom2/fesom2/src/oce_ale_tracer.F90:251
#3  0x5d6016874847 in diff_tracers_ale_
	at /__w/fesom2/fesom2/src/associate_mesh_ass.h:33
#4  0x5d6016873800 in solve_tracers_ale_
	at /__w/fesom2/fesom2/src/oce_ale_tracer.F90:251
#5  0x5abd488f0448 in oce_timestep_ale_
	at /__w/fesom2/fesom2/src/oce_ale.F90:3248
#5  0x5d6016851448 in oce_timestep_ale_
	at /__w/fesom2/fesom2/src/oce_ale.F90:3248
#6  0x5abd487dff18 in __fesom_module_MOD_fesom_runloop
	at /__w/fesom2/fesom2/src/fesom_module.F90:511
#7  0x5abd4878fe2f in MAIN__
	at /__w/fesom2/fesom2/src/fesom_main.F90:15
#8  0x5abd4878fe2f in main
	at /__w/fesom2/fesom2/src/fesom_main.F90:10
#6  0x5d6016740f18 in __fesom_module_MOD_fesom_runloop
	at /__w/fesom2/fesom2/src/fesom_module.F90:511
#7  0x5d60166f0e2f in MAIN__
	at /__w/fesom2/fesom2/src/fesom_main.F90:15
#8  0x5d60166f0e2f in main
	at /__w/fesom2/fesom2/src/fesom_main.F90:10
--------------------------------------------------------------------------
Primary job  terminated normally, but 1 process returned
a non-zero exit code. Per user-direction, the job has been aborted.
--------------------------------------------------------------------------
--------------------------------------------------------------------------
mpirun noticed that process rank 1 with PID 0 on node d14426014abc exited on signal 11 (Segmentation fault).
--------------------------------------------------------------------------

Is this a good enough hint to figure out what happened? oce_ale_tracer.F90:251

This reverts commit f0aa180.

…ale'.

JanStreffing · 2024-03-20T12:41:47Z

Great the checks have passed.

JanStreffing · 2024-03-20T12:44:55Z

@patrickscholz Ready for review

JanStreffing added 3 commits March 14, 2024 15:18

WIP applying patch for transient tracers

3a71515

structural fixes

122ccb3

temporarly add rejects for info; delete later

0e1c857

JanStreffing added the help wanted Extra attention is needed label Mar 14, 2024

JanStreffing added this to the FESOM 2.6 milestone Mar 14, 2024

JanStreffing requested review from dsidoren and patrickscholz March 14, 2024 14:41

JanStreffing assigned patrickscholz Mar 14, 2024

JanStreffing mentioned this pull request Mar 14, 2024

Transient Tracers #572

Closed

mbutzin added 7 commits March 18, 2024 14:39

Include patches for rejected code. src/io_meandata.F90

f33560c

Include patches for rejected code.

ca3d5c0

Complete the update of transient tracer mean output.

961ebff

Code update of transient tracer switch message.

7b9f354

Include a 2nd bc_surface-type function extended for transient tracers.

48f02cf

Include sea ice concentration as input parameter for function 'transi…

ed42e93

…t_bc_surface', and remove the now obsolete transient tracer code from function 'bc_surface'.

Fix ill-defined transient tracer indices.

1520dc2

Change idlist index to initialize 14C from file(s).

79a49d2

mbutzin added 3 commits March 20, 2024 12:05

Bugfix: add missing argument to 'call diff_tracers_ale'.

f0aa180

Revert "Bugfix: add missing argument to 'call diff_tracers_ale'."

79f22f6

This reverts commit f0aa180.

Bugfix: add missing argument to call and interfaces of 'diff_tracers_…

9bb5ed1

…ale'.

JanStreffing added 2 commits March 20, 2024 13:43

Delete src/oce_ale_tracer.F90.rej

d698e7e

Delete src/oce_setup_step.F90.rej

19e0a10

JanStreffing mentioned this pull request Mar 20, 2024

Order of remaining 2.6 merge requests: #574

Closed

Merge branch 'refactoring' into refactoring_tracers_butzin_second_try

bba4ea9

JanStreffing mentioned this pull request Mar 25, 2024

2.1 awiesm 2.1 with transient tracers #570

Closed

patrickscholz approved these changes Apr 3, 2024

View reviewed changes

patrickscholz merged commit 35bcfe7 into refactoring Apr 3, 2024
4 checks passed

JanStreffing deleted the refactoring_tracers_butzin_second_try branch August 15, 2024 12:17

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Refactoring tracers butzin second try #576

Refactoring tracers butzin second try #576

JanStreffing commented Mar 14, 2024 •

edited

Loading

mbutzin commented Mar 19, 2024

JanStreffing commented Mar 19, 2024 •

edited

Loading

mbutzin commented Mar 19, 2024

mbutzin commented Mar 19, 2024

JanStreffing commented Mar 19, 2024

JanStreffing commented Mar 20, 2024

JanStreffing commented Mar 20, 2024

Refactoring tracers butzin second try #576

Refactoring tracers butzin second try #576

Conversation

JanStreffing commented Mar 14, 2024 • edited Loading

mbutzin commented Mar 19, 2024

JanStreffing commented Mar 19, 2024 • edited Loading

mbutzin commented Mar 19, 2024

mbutzin commented Mar 19, 2024

JanStreffing commented Mar 19, 2024

JanStreffing commented Mar 20, 2024

JanStreffing commented Mar 20, 2024

JanStreffing commented Mar 14, 2024 •

edited

Loading

JanStreffing commented Mar 19, 2024 •

edited

Loading