Ok. Seems that there is no module loading issue, as all required module are loaded. You may run 'bash -x service nv_peer_mem start' or add more prints to startup scripts and check from the output what is the command leading the failure. Then concentrate on the command. Anything in dmesg?
↧