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.

1 files changed, 86 insertions, 0 deletions

diff --git a/scripts/create_mdev.sh b/scripts/create_mdev.sh
new file mode 100755
index 0000000..a6ea91e
--- /dev/null
+++ b/scripts/create_mdev.sh
@@ -0,0 +1,86 @@
+#!/bin/bash
+# create_mdev.sh - Create NVIDIA vGPU mediated device for QEMU
+# Usage: ./create_mdev.sh [PCI_ADDRESS] [VGPU_TYPE]
+
+set -euo pipefail
+
+PCI_ADDR="${1:-}"
+VGPU_TYPE="${2:-pci-nvidia-vgpu-turing}"
+
+# Supported vGPU types (matching the mdev type names)
+VALID_TYPES=("pci-nvidia-vgpu-ampere" "pci-nvidia-vgpu-hopper" "pci-nvidia-vgpu-turing" "pci-nvidia-vgpu-pascal")
+
+# Find NVIDIA GPU if not specified
+if [[ -z "$PCI_ADDR" ]]; then
+    echo "Scanning for NVIDIA GPUs..."
+    mapfile -t GPUS < <(lspci -D -d 10de: | grep -E "VGA|3D controller" | awk '{print $1}')
+    if [[ ${#GPUS[@]} -eq 0 ]]; then
+        echo "ERROR: No NVIDIA GPU found"
+        exit 1
+    elif [[ ${#GPUS[@]} -eq 1 ]]; then
+        PCI_ADDR="${GPUS[0]}"
+        echo "Auto-selected GPU: $PCI_ADDR"
+    else
+        echo "Multiple GPUs found:"
+        for i in "${!GPUS[@]}"; do
+            echo "  [$i] ${GPUS[$i]}"
+        done
+        read -p "Select GPU [0-$((${#GPUS[@]}-1))]: " idx
+        PCI_ADDR="${GPUS[$idx]}"
+    fi
+fi
+
+# Validate PCI address format
+if [[ ! "$PCI_ADDR" =~ ^[0-9a-fA-F]{4}:[0-9a-fA-F]{2}:[0-9a-fA-F]{2}\.[0-9a-fA-F]$ ]]; then
+    echo "ERROR: Invalid PCI address format. Expected: dddd:bb:dd.f"
+    exit 1
+fi
+
+# Check if GPU exists
+if [[ ! -e "/sys/bus/pci/devices/$PCI_ADDR" ]]; then
+    echo "ERROR: PCI device $PCI_ADDR not found"
+    exit 1
+fi
+
+# Validate vGPU type
+if [[ ! " ${VALID_TYPES[*]} " =~ " ${VGPU_TYPE} " ]]; then
+    echo "ERROR: Invalid vGPU type. Valid types: ${VALID_TYPES[*]}"
+    exit 1
+fi
+
+MDEV_TYPE_PATH="/sys/bus/pci/devices/$PCI_ADDR/mdev_supported_types/$VGPU_TYPE"
+
+if [[ ! -d "$MDEV_TYPE_PATH" ]]; then
+    echo "ERROR: vGPU type $VGPU_TYPE not supported on $PCI_ADDR"
+    echo "Available types:"
+    ls "/sys/bus/pci/devices/$PCI_ADDR/mdev_supported_types/" 2>/dev/null || echo "  (none - check driver loaded)"
+    exit 1
+fi
+
+# Check available instances
+AVAILABLE=$(cat "$MDEV_TYPE_PATH/available_instances" 2>/dev/null || echo "0")
+if [[ "$AVAILABLE" -le 0 ]]; then
+    echo "ERROR: No available instances for $VGPU_TYPE on $PCI_ADDR"
+    exit 1
+fi
+
+echo "Creating $VGPU_TYPE on $PCI_ADDR..."
+
+# Generate UUID
+UUID=$(uuidgen)
+echo "$UUID" | sudo tee "$MDEV_TYPE_PATH/create" > /dev/null
+
+# Verify creation
+MDEV_PATH="/sys/bus/mdev/devices/$UUID"
+if [[ -d "$MDEV_PATH" ]]; then
+    echo "SUCCESS: Created mdev device $UUID"
+    echo "  Path: $MDEV_PATH"
+    echo "  PCI:  $PCI_ADDR"
+    echo "  Type: $VGPU_TYPE"
+    echo ""
+    echo "For QEMU, use:"
+    echo "  -device vfio-pci,sysfsdev=$MDEV_PATH,display=off"
+else
+    echo "ERROR: Failed to create mdev device"
+    exit 1
+fi
\ No newline at end of file