From 68cdd7728a7cc6715bbbff7decfe2b11c3bdf558 Mon Sep 17 00:00:00 2001 From: Open-IOV Date: Wed, 23 Sep 2026 05:44:08 -0400 Subject: Open-IOV: initial public release Open-IOV virtualizes NVIDIA GPUs so that unmodified guest drivers run with real GPU functionality - CUDA first, 3D and video to follow - on unmodified hosts, without a proprietary vGPU license and without patching any driver, firmware or hypervisor. Core principle: emulate the interface, pass through the work. kernel/ open-iov-vfio.ko - host VFIO mdev driver: virtual GPU device (PCI config space, BAR shadows, MSI-X), GSP boot emulation, GSP-RM message-queue/RPC emulation, memory model (GPU-VA to guest-PA discovery) and the semaphore-completion service that advances guest UVM progress. userspace/ libiov.so management API and examples. tools/ guest diagnostics, firmware analysis wrappers, RM probes. scripts/ stack manager, VM harness, build/load/test automation. config/ hardware/version profiles (the portability surface). docs/ documentation tree - start at docs/README.md. wiki/ long-form firmware and protocol deep dives. Status: initialization works end to end for a stock 570.144 Linux guest on Turing (TU104): driver probe, nvidia-smi, the full GSP-RM control battery, object creation and the early CUDA/UVM start-up path. GPU work forwarding - the step that makes CUDA kernels execute on real hardware - is the active work item; see docs/04-implementation/01-known-knowns-and-unknowns.md. License: BSD-3-Clause for the project; GPL-2.0 for kernel modules that link GPL-only kernel interfaces. See LICENSE and docs/09-reference/legal-and-licensing.md. --- patches/nvidia/README.md | 45 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 45 insertions(+) create mode 100644 patches/nvidia/README.md (limited to 'patches/nvidia/README.md') diff --git a/patches/nvidia/README.md b/patches/nvidia/README.md new file mode 100644 index 0000000..9c74f75 --- /dev/null +++ b/patches/nvidia/README.md @@ -0,0 +1,45 @@ +# NVIDIA open-gpu-kernel-modules Patches + +Open-IOV does **not** vendor the NVIDIA OpenRM source tree. Instead, the +installer clones upstream at a pinned tag and applies this patch series. + +## Pinned upstream + +- Repository: +- Tag: `610.57.04` +- Pristine-tree SHA256 of the delta base: tag `610.57.04` + +## Applying + +```bash +git clone --depth 1 --branch 610.57.04 \ + https://github.com/NVIDIA/open-gpu-kernel-modules.git +cd open-gpu-kernel-modules +git apply /path/to/open-iov/patches/nvidia/*.patch +``` + +(`install.sh` does this automatically.) + +## Regenerating the series + +If you modify the NVIDIA tree, regenerate the patches with: + +```bash +cd nvidia-open-gpu-kernel-modules +git format-patch 610.57.04..HEAD -o ../patches/nvidia/ +``` + +## What the series does + +`0001-Build-fixes-and-vGPU-VFIO-interface-integration.patch`: + +- Adds the vGPU VFIO interface headers (`nv-vgpu-vfio-interface.h`, + `nv-vgpu-vfio-interface-nvidia.h`) that `open-iov-vfio.ko` and + `nvidia.ko` share. +- Exports the vGPU VFIO symbols from `nvidia.ko` + (`nvidia_vgpu_vfio_probe/remove`, `nvidia_vgpu_vfio_get_ops/set_ops`, + `nv_vgpu_mdev_register_parent/unregister_parent`, ...). +- Reports version `570.144` to match the GSP firmware the guest driver + stack is validated against. +- Build fixes for kernel 6.8 out-of-tree builds and the + `Module.symvers.open-iov` reference list. -- cgit v1.2.3