Compiling newstart doesn't work on Spirit

[abierjon]

When trying to compile newstart on Spirit with the following command line :

makelmdz_fcm -arch ifort_MESOIPSL -p mars -d 64x48x54 -j 8 newstart

the compilation crashes and returns the following error :

/home/abierjon/Modele_lastversion3/trunk/LMDZ.COMMON/libo/ifort_MESOIPSL_64x48x54_phymars_seq.e/.config/ppsrc/dyn_phys_sub/newstart.f(1784): error #6784: The number of actual arguments cannot be greater than the number of dummy arguments.   [PHYSDEM0]
      call physdem0("restartfi.nc",longitude,latitude,
-----------^
compilation aborted for /home/abierjon/Modele_lastversion3/trunk/LMDZ.COMMON/libo/ifort_MESOIPSL_64x48x54_phymars_seq.e/.config/ppsrc/dyn_phys_sub/newstart.f (code 1)
fcm_internal compile failed (256)
make: *** [/home/abierjon/Modele_lastversion3/trunk/LMDZ.COMMON/libo/ifort_MESOIPSL_64x48x54_phymars_seq.e/.config/bld/dyn_phys_sub.mk:108: newstart.o] Error 1
make: *** Waiting for unfinished jobs....
make -s -j 8 all failed (512) at /home/abierjon/FCM_V1.2/lib/Fcm/Build.pm line 1922.
->Make               : 10 seconds
->TOTAL              : 11 seconds
Build command failed on Wed Mar  1 17:49:52 2023.

Yangcheng also has this error on top of others.

[abierjon]

Actually, this is normal, true for all machines (not only spirit) and is explained by revision r2900 by Romain Vandemeulebrouck : "Utilitaries are not adapted yet." (to the subslope implementation in the model, which is done step by step through several commits)
I let the ticket open as long as the subslope implementation is not complete.

[emillour]

Fixed with r2909 and r2910