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, 47 insertions, 0 deletions

diff --git a/tools/module_ops/dry_run.sh b/tools/module_ops/dry_run.sh
new file mode 100755
index 0000000..e6bb6f2
--- /dev/null
+++ b/tools/module_ops/dry_run.sh
@@ -0,0 +1,47 @@
+#!/bin/bash
+# Run vGPU dry-run validation
+# Usage: sudo ./dry_run.sh
+
+set -euo pipefail
+
+VGPU_KERNEL="${OPEN_IOV_ROOT:-$(cd "$(dirname "${BASH_SOURCE[0]}")/../.." && pwd)}/kernel"
+
+echo "=== Open-IOV vGPU Dry-Run Validation ==="
+echo ""
+
+# Check if host nvidia.ko is loaded
+if ! lsmod | grep -q "^nvidia "; then
+    echo "ERROR: Host nvidia.ko not loaded. Load it first with:"
+    echo "  rzsh nvidia vgpu load"
+    exit 1
+fi
+
+echo "Host nvidia.ko: LOADED ✓"
+echo ""
+
+# Load vGPU module in dry-run mode
+echo "Loading nvidia-vgpu-vfio.ko in dry-run mode..."
+sudo insmod "$VGPU_KERNEL/nvidia-vgpu-vfio.ko" dyndbg='+p' dry_run=1
+
+# Check result
+sleep 2
+if dmesg -T | grep -q "DRY-RUN MODE: validating init without hardware changes"; then
+    echo ""
+    echo "✓ Dry-run validation PASSED"
+    echo "  Firmware, BAR layout, IRQ caps, IOMMU validated"
+else
+    echo ""
+    echo "✗ Dry-run validation FAILED"
+    echo "Check dmesg for errors:"
+    dmesg -T | grep -i -e nvidia-vgpu -e vgpu | tail -20
+fi
+
+# Unload
+echo ""
+echo "Unloading dry-run module..."
+sudo rmmod nvidia-vgpu-vfio
+
+echo ""
+echo "=== Dry-Run Complete ==="
+echo "If validation passed, load for real with:"
+echo "  sudo insmod $VGPU_KERNEL/nvidia-vgpu-vfio.ko dyndbg='+p'"
\ No newline at end of file