Bad Memory allocation for p2p Bandwidth and Latency Test

Accelerated Computing CUDA CUDA Programming and Performance
1

I’m currently executing CUDA 9.1 on ubuntu and supermicro server. When I execute p2p bandwidth tes, I’m getting following error.

[P2P (Peer-to-Peer) GPU Bandwidth Latency Test]

***NOTE: In case a device doesn't have P2P access to other one, it falls back to normal memcopy procedure.
So you can see lesser Bandwidth (GB/s) in those cases.

P2P Connectivity Matrix
     D\D
Unidirectional P2P=Disabled Bandwidth Matrix (GB/s)
terminate called after throwing an instance of 'std::bad_alloc'
  what():  std::bad_alloc
Aborted (core dumped)

Any help is appreciated