CUDA error at code=2(cudaErrorMemoryAllocation) "cudaEventCreate(&start)"?

my GPU-Card is Tesla:100,after successfully installed the nvidia driver-396.26(nvidia-smi print the gpu information).but when build the cuda9.2 with the proper procedure,the devicequery failed ,return no CUDA-capable device is detected; and nvidia-smi also not work with No devices were found returned.

also return an error called DBE(an uncorrectable double bit error ) has been detected on gpu int the framebuffer at patrition 5 ,subpatrition 1.

I have the same error. I am on Ubuntu 18.04 with Tesla V100, and CUDA 10.1. Any resolution or workaround for this error?

$ ./bandwidthTest
[CUDA Bandwidth Test] - Starting…
Running on…

Device 0: Tesla V100-PCIE-16GB
Quick Mode

CUDA error at code=2(cudaErrorMemoryAllocation) “cudaEventCreate(&start)”