Error in building CESM1.2 (physics_buffer.F90)

[g_vettoretti@utoronto_ca]

Hello, 

Has anyone come across this error in building CESM1.2 on an IBM power6 machine (like bluefire)mpxlf95_r -c -I. -I/scratch/p/peltier/guido/cesm12test/bld/pio -I/scinet/tcs/Libraries/netcdf-4.2.1.1/poe_xlc/include -I/scinet/tcs/Libraries/parallel-netcdf-1.3.1/include -I/scratch/p/peltier/guido/cesm12test/bld/mct/mct -I/scratch/p/peltier/guido/cesm12test/bld/mct/mpeu -I/scratch/p/peltier/guido/cesm12test/bld/pio -I/scratch/p/peltier/guido/cesm12test/bld/gptl -I/scratch/project/c/ccsm/cesm1_2_0/models/utils/pio -I/scratch/p/peltier/guido/cesm12test/bld/lib/include -I. -I/scratch/p/peltier/guido/cesm12test/SourceMods/src.cam -I/scratch/project/c/ccsm/cesm1_2_0/models/atm/cam/src/chemistry/pp_none -I/scratch/project/c/ccsm/cesm1_2_0/models/atm/cam/src/chemistry/mozart -I/scratch/project/c/ccsm/cesm1_2_0/models/atm/cam/src/chemistry/bulk_aero -I/scratch/project/c/ccsm/cesm1_2_0/models/atm/cam/src/chemistry/utils -I/scratch/project/c/ccsm/cesm1_2_0/models/atm/cam/src/physics/cam -I/scratch/project/c/ccsm/cesm1_2_0/models/atm/cam/src/dynamics/fv -I/scratch/project/c/ccsm/cesm1_2_0/models/atm/cam/src/utils/pilgrim -I/scratch/project/c/ccsm/cesm1_2_0/models/atm/cam/src/cpl_mct -I/scratch/project/c/ccsm/cesm1_2_0/models/atm/cam/src/cpl_share -I/scratch/project/c/ccsm/cesm1_2_0/models/atm/cam/src/control -I/scratch/project/c/ccsm/cesm1_2_0/models/atm/cam/src/utils -I/scratch/p/peltier/guido/cesm12test/bld/lib/include -g -qfullpath -qmaxmem=-1 -qarch=auto -qtune=auto -qcache=auto -qsclk=micro -O2 -qstrict -Q -qspill=6000 -WF,-DCO2A -WF,-DMAXPATCH_PFT=numpft+1 -WF,-DLSMLAT=1 -WF,-DLSMLON=1 -WF,-DPLON=288 -WF,-DPLAT=192 -WF,-DPLEV=26 -WF,-DPCNST=3 -WF,-DPCOLS=16 -WF,-DPSUBCOLS=1 -WF,-DPTRM=1 -WF,-DPTRN=1 -WF,-DPTRK=1 -WF,-DSTAGGERED -WF,-DSPMD -WF,-DAIX -WF,-DMCT_INTERFACE -WF,-DHAVE_MPI -WF,-DFORTRAN_SAME -WF,-DAIX -WF,-DCPRIBM -qsuffix=f=f90:cpp=F90 /scratch/project/c/ccsm/cesm1_2_0/models/atm/cam/src/physics/cam/physics_buffer.F901585-100 Internal compiler error -- please contact your service representative. For more information visit: http://www.ibm.com/support/docview.wss?ui
d=swg21110810
1501-511 Compilation failed for file physics_buffer.F90.
gmake: *** [physics_buffer.o] Error 255There really doesn't seem to be much other information.The modules that I am loading (from env_mach_specific) are:module load vacpp/12.1module load xlf/14.1
module load cmake/2.8.8
module load hdf5/1811-v18-poe-xlc
module load netcdf/4.2.1.1-poe-xlc
module load parallel-netcdf/1.3.1
#module load esmf/5.2.0rp1
module load lapack/3.4.2Thanks 

 

[jedwards]

Hi,
Here are two things you can try:1. A different level of the xlf compiler - use module avail to see what levels are available on your system and try another one.2. A different level of optimization - if that works you may be able to reduce optimization on just this file.