0

I installed the NVIDIA driver 367.44 from the Nvidia site. As a last step I was prompted to disable secure boot on re-boot. I did that, but I am not fully comfortable keeping it disabled. I want to re-enable it and do not know how to.

In anycase, I can not use the Nvidia card for rendering and it only works with the "--no-opengl-files" option. Anything else pushes Ubuntu into a log-in loop.

Edit: 

sudo mokutil --sb-state

shows: 

SecureBoot enabled

But while booting I get an annoying "Booting in insecure mode message"

EDIT: as I understand from the comments this might not pertain to UFEI secure boot, but the MOK secure boot which is needed for Ubuntu to use third party drivers. I am still using 14.04 and keeping the validation enabled did not cause me any issues with my nvidia drivers for CUDA.

I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:900] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
I tensorflow/core/common_runtime/gpu/gpu_init.cc:102] Found device 0 with properties: 
name: GeForce 940M
major: 5 minor: 0 memoryClockRate (GHz) 1.176
pciBusID 0000:04:00.0
Total memory: 1.96GiB
Free memory: 1.94GiB
I tensorflow/core/common_runtime/gpu/gpu_init.cc:126] DMA: 0 
I tensorflow/core/common_runtime/gpu/gpu_init.cc:136] 0:   Y 
I tensorflow/core/common_runtime/gpu/gpu_device.cc:755] Creating TensorFlow device (/gpu:0) -> (device: 0, name: GeForce 940M, pci bus id: 0000:04:00.0)
shaunakde
  • 111
  • 1
  • 1
  • 6

1 Answers1

1

I found a solution in this thread: Ubuntu: "Booting in insecure mode" with SecureBoot enabled

Basically this issue pertains to the validation in module MOK (Machine Owner Key).

To get rid of the message it is sufficient to type: 

sudo mokutil --enable-validation

There is a chance that in Ubuntu 16 the driver might stop working. But I can confirm that in ubuntu 14.04 - cuda functionality is unaffected. I was not using the driver for rendering anyway. 

>>> sess = tensorflow.Session()
I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:900] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
I tensorflow/core/common_runtime/gpu/gpu_init.cc:102] Found device 0 with properties: 
name: GeForce 940M
major: 5 minor: 0 memoryClockRate (GHz) 1.176
pciBusID 0000:04:00.0
Total memory: 1.96GiB
Free memory: 1.94GiB
I tensorflow/core/common_runtime/gpu/gpu_init.cc:126] DMA: 0 
I tensorflow/core/common_runtime/gpu/gpu_init.cc:136] 0:   Y 
I tensorflow/core/common_runtime/gpu/gpu_device.cc:755] Creating TensorFlow device (/gpu:0) -> (device: 0, name: GeForce 940M, pci bus id: 0000:04:00.0)
Community
  • 1
shaunakde
  • 111
  • 1
  • 1
  • 6