To Whom It May Concern,
I’m trying to simulate a surface reaction on a slab with F2 flux (depo) using sevennet pot. But I continuously get the “Segmentation fault”. May I ask your advice on this error?
Error message:
Setting up Verlet run ...
Unit style : metal
Current step : 0
Time step : 0.001
[W TensorAdvancedIndexing.cpp:1615] Warning: scatter_reduce() is in beta and the API may change at any time. (function operator())
[lcas1314:2348818] *** Process received signal ***
[lcas1314:2348818] Signal: Segmentation fault (11)
[lcas1314:2348818] Signal code: Address not mapped (1)
[lcas1314:2348818] Failing at address: 0xaf00000047
[lcas1314:2348818] [ 0] /lib64/libpthread.so.0(+0x12ce0)[0x149ee0a6ace0]
[lcas1314:2348818] [ 1] /appl/LINUX/LAMMPS/bin/lmp_serial_02Aug2023_gpu_sevennet_240902.x[0x909f95]
[lcas1314:2348818] [ 2] /appl/LINUX/LAMMPS/bin/lmp_serial_02Aug2023_gpu_sevennet_240902.x[0x5c071f]
[lcas1314:2348818] [ 3] /appl/LINUX/LAMMPS/bin/lmp_serial_02Aug2023_gpu_sevennet_240902.x[0x55dae5]
[lcas1314:2348818] [ 4] /appl/LINUX/LAMMPS/bin/lmp_serial_02Aug2023_gpu_sevennet_240902.x[0x46ca6a]
[lcas1314:2348818] [ 5] /appl/LINUX/LAMMPS/bin/lmp_serial_02Aug2023_gpu_sevennet_240902.x[0x46cd7e]
[lcas1314:2348818] [ 6] /appl/LINUX/LAMMPS/bin/lmp_serial_02Aug2023_gpu_sevennet_240902.x[0x44cb6d]
[lcas1314:2348818] [ 7] /lib64/libc.so.6(__libc_start_main+0xf3)[0x149e885cacf3]
[lcas1314:2348818] [ 8] /appl/LINUX/LAMMPS/bin/lmp_serial_02Aug2023_gpu_sevennet_240902.x[0x44deae]
Input file:
########################################################################
# F2 molecular beam onto a Ge/Si slab �~@~S SevenNet e3gnn template
# Type map: 1 = Ge, 2 = Si, 3 = F
########################################################################
clear
units metal
atom_style atomic
atom_modify map yes
boundary p p f
# --- read slab & add vacuum ------------------------------------------
read_data SiGe_slab.data extra/atom/types 1
change_box all z delta 0.0 50.0 units box
# --- per-type masses --------------------------------------------------
mass 1 72.630 # Ge
mass 2 28.085 # Si
mass 3 18.998 # F
reset_atoms id sort yes # consecutive IDs for e3gnn
# --- SevenNet pair style ---------------------------------------------
pair_style e3gnn
pair_coeff * * deployed_serial.pt Ge Si F
# --- groups -----------------------------------------------------------
group slab type 1 2
group gas type 3
# --- molecule template ------------------------------------------------
molecule F2 F2.mol
neighbor 2.0 bin
neigh_modify every 1 delay 1 check yes
# --- dynamic deposition region ---------------------------------------
#variable zhi equal bound(slab,zmax)
#print "Topmost slab atom is at: ${zhi}"
#variable zlo equal $(v_zhi+5.0)
#print "Deposition region starts at: $(v_zhi+5.0)"
#variable zhi2 equal $(v_zhi+30.0)
#print "Deposition region ends at: $(v_zhi+30.0)"
#region dep block EDGE EDGE EDGE EDGE ${zlo} ${zhi2} units box side in
variable zhi_box equal zhi
variable zlo equal $(v_zhi_box - 15.0)
variable zhi2 equal $(v_zhi_box - 5.0)
region dep block EDGE EDGE EDGE EDGE ${zlo} ${zhi2} units box side in
# --- flux parameters --------------------------------------------------
fix ins gas deposit 1000 0 100 29494 region dep mol F2 near 2.0 orient 0 0 1 vz -5.0 -5.0 units box # 1 dimer per 100 steps
# --- dynamics ---------------------------------------------------------
timestep 0.001
velocity all create 300.0 87287 mom yes rot yes dist gaussian
fix slabT slab nvt temp 300.0 300.0 100.0
fix gasT gas nve
# --- I/O --------------------------------------------------------------
thermo 100
thermo_style custom step atoms temp pe ke etotal press
dump traj all custom 100 dump.F2-SiGe id type x y z
restart 500 restart.F2-SiGe
run 1000 # �~F~P quick test; change to 200000 for production
write_data final.F2-SiGe.data
F2.mol: placed two atoms in the distance without the defined bond as I expect the potential describe the interaction:
#F2 molecule - non bonded
2 atoms
Coords
1 0.000 0.000 0.000
2 1.412 0.000 0.000
Types
1 3
2 3