• Various security issues were addressed, for additional details on the med-high severity issues please review NVIDIA Product Security for more information
• Fixed an issue with power management that would result in the GPU falling off the bus (Kepler+ GPUs)
• Fixed an issue where nvidia-smi nvlink would not report any NVLink information
• Fixed a bug in unified memory for memory accesses between non-peer GPUs
• Fixed a crash in unified memory when running applications under MPS (on Pascal+ systems)
• Fixed an issue with P2P bandwidth performance under passthrough virtualization when pair of GPUs can support both PCIe and NVLink interconnects
• Fixed an issue where cudaDeviceReset would result in a segmentation fault when using CUDA IPC APIs
• Fixed an issue where CUDA-Z would hang AWS EC2 p3 instances in some cases
• Fixed a bug in the JIT compiler in some cases address computations involving immediate values may be incorrect

Maximus System
For Maximus systems (Quadro + Tesla in the same system), download the latest recommended Quadro driver. Quadro drivers are qualified for workstations and rendering applications, including Maximus configurations that use Quadro for visualization and Tesla GPUs for compute acceleration.

V-Series:

Tesla V100

P-Series:

Tesla P100, Tesla P40, Tesla P6, Tesla P4

K-Series:

Tesla K80, Tesla K40c, Tesla K40m, Tesla K40s, Tesla K40st, Tesla K40t, Tesla K20Xm, Tesla K20m, Tesla K20s, Tesla K20c, Tesla K10, Tesla K8

C-Class:

Tesla C2075, Tesla C2070, Tesla C2050

M-Class:

M60, M40, M6, M4, M2090, M2075, M2070, M2070-Q, M2050

X-Class:

Tesla X2070, Tesla X2090