How can I mask floating point exception in Linux environment

I need to compile my Matlab codes and run in Linux servers. First of all, I can successfully compile my codes on my Window laptop by using "mcc -v -c main.m" in the Matlab command window. But when I do the same thing in a Matlab command window under the Linux enviroment (which I remotely login into a Linux server), I got the following error:

>> mcc -m -v main.m
Compiler version: 4.18 (R2012b)
Processing <MatlabRoot>/toolbox/matlab/mcc.enc
Processing <MatlabRoot>/toolbox/control/mcc.enc
Processing <MatlabRoot>/toolbox/comm/mcc.enc
Processing <MatlabRoot>/toolbox/shared/controllib/general/mcc.enc
Processing <MatlabRoot>/toolbox/shared/controllib/engine/mcc.enc
Processing <MatlabRoot>/toolbox/signal/mcc.enc
Processing <MatlabRoot>/ident/mcc.enc
Processing <MatlabRoot>/matlab/system/mcc.enc
Processing <MatlabRoot>/database/mcc.enc
Processing <MatlabRoot>/stats/mcc.enc
Processing <MatlabRoot>/fixedpoint/mcc.enc
Processing <MatlabRoot>/shared/statslib/mcc.enc
Processing <MatlabRoot>/optim/mcc.enc
Processing <MatlabRoot>/shared/optimlib/mcc.enc
Processing <MatlabRoot>/distcomp/mcc.enc
Floating exception
"<MatlabRoot>/bin/glnxa64/mcc" -E "/tmp/00fc2Tt3L8P_4034": Floating point exception
Error using mcc
Error executing mcc, return status = 136 (0x88).

I think one of the toolboxes may enable floating point exception. Anyway, I would like to mask the floating point exception for the Linux environment. But I don't know how to do it.