Among the notable improvements, the driver introduces a new environment variable, CUDA_DISABLE_PERF_BOOST, allowing users to disable CUDA’s default behavior of automatically boosting GPU clock speeds to higher power states during compute workloads.

https://www.nvidia.com/en-us/drivers/details/257493/

  • tomalley8342@lemmy.world
    link
    fedilink
    arrow-up
    6
    ·
    3 days ago

    For virtualization users, the driver addresses a soft lockup issue involving the vfio-pci module, which could occur after powering off a virtual machine with a passed-through NVIDIA GPU. This fix improves reliability for users running GPU passthrough in environments such as KVM or QEMU.

    AMD friends watching from the sidelines

    • Karna@lemmy.mlOP
      link
      fedilink
      arrow-up
      3
      ·
      edit-2
      3 days ago

      The good news is Nvidia consumer grade GPUs don’t even support vGPU and can’t be passed though if Host OS is using it.

        • tomalley8342@lemmy.world
          link
          fedilink
          arrow-up
          1
          ·
          8 hours ago

          You can generally pass through modern Nvidia GPUs as the entire device, provided that you let go of it in the host. It not supporting vGPUs just means you can’t virtualize and split up the GPU workload across multiple VMs, which I believe is also the case for consumer AMD GPUs as well.