Hello,

I tried a 64 atoms supercell calculation with 8x8x8 k-point grid. There was the following error message :

excitingmpi(36114) malloc: * mmap(size=2405527552) failed (error code=12)
* error: can't allocate region
*** set a breakpoint in malloc_error_break to debug
At line 197 of file ../../src/init1.f90
Traceback: not available, compile with -ftrace=frame or -ftrace=full
Operating system error: Cannot allocate memory
Memory allocation failed

when I add either -g -ftrace=full or frame, I have the following error message :

excitingmpi(27612) malloc: * mmap(size=2405527552) failed (error code=12)
* error: can't allocate region
*** set a breakpoint in malloc_error_break to debug
At line 197 of file ../../src/init1.f90
Traceback: (Innermost first)
Called from line 45 of file ../../src/gndstate.f90
Called from line 31 of file ../../src/src_inputparser/tasklauncher.f90
Called from line 21 of file ../../src/mainxml/main.f90
Operating system error: Cannot allocate memory
Memory allocation failed

mpirun has exited due to process rank 5 with PID 27612 on
node supernano.physics.uiowa.edu exiting without calling "finalize". This may
have caused other processes in the application to be
terminated by signals sent by mpirun (as reported here).

The critical lines are :

The critical line is this:

init1.f90:197 allocate(sfacgk(ngkmax, natmtot, nspnfv, nkpt))

which is where the allocation failure occurs. From the mpi communication:

gndstate.f90: call MPI_Bcast(sfacgk, size(sfacgk), MPI_DOUBLE_PRECISION, 0, MPI_COMM_WORLD, ierr)

it appears that sfacgk is *not* split up among the different processes, but rather each process has a complete copy of sfacgk. Is it correct, and if it is a bug or was intentional?

Best,

Thomas