open-iov.git/.gitignore, branch dualOpen-IOV vGPU-IO: host-side GSP emulation for unmodified NVIDIA guest drivers
https://git.vgpu.io/open-iov.git/atom?h=dual2026-10-14T19:20:03+00:00gitignore: never track backup/editor leftovers (star.bak patterns)2026-10-14T19:20:03+00:00Arthur Hanson Rasmussonarthur@vgpu.io2026-10-14T19:20:03+00:00urn:sha1:17393a74824ce347ed0c4113a5e5f0c28b4ce441rename (cosmetic): green-eye -> risc-v-gsp-rm-debugger in user-facing names, docs, CLI, launcher and CI workflow; internal greeneye package/module identifiers unchanged (includes in-flight greeneye module edits present in the tree)2026-10-02T15:44:31+00:00Arthur Hanson Rasmussonarthur@vgpu.io2026-10-02T15:44:31+00:00urn:sha1:05797fd8cf884bb680540dfc188f36836a215c9fgitignore: keep the local evidence workspace and generated demo artifacts untracked; drop built RM tool binaries (source and Makefiles stay)2026-10-02T15:30:46+00:00Arthur Hanson Rasmussonarthur@vgpu.io2026-10-02T15:30:46+00:00urn:sha1:faabc48f6d9b2d531b5b7a7e1320ccd1e030e85cdocs: add complete SBOM, ignore local deployment resources, remove stale third-party references2026-10-02T06:50:38+00:00Arthur Hanson Rasmussonarthur@vgpu.io2026-10-02T06:50:38+00:00urn:sha1:44a870fa4325a027443595531fd75c1e786201cagitignore update2026-10-02T06:14:14+00:00Arthur Hanson Rasmussonarthur@vgpu.io2026-10-02T06:13:03+00:00urn:sha1:a233bf463a97e54c2f580d75a9a9c55b5e62e0bckernel: coherent GSP-RM replies for engine-context creation2026-09-23T15:46:33+00:00Arthur Hanson Rasmussonarthur@vgpu.io2026-09-23T15:46:33+00:00urn:sha1:5c47eacbe01f9353f2c21d96e1a3eb15a63fd991
- GSP_RM_ALLOC: echo the request payload back with status forced to NV_OK.
The kernel driver copies the reply params over the caller's params buffer
using the class-expected size, so a zeroed reply smashed libcuda's stack
(SIGSEGV at the engine-registration continuation); echoing makes the
copy-back idempotent.
- CE_GET_PHYSICAL_CAPS / CE_GET_ALL(_PHYSICAL)_CAPS: report sysmem, P2P and
BL copy caps and mark the GR-synchronous CE (GRCE, COPY0). Without a GRCE
the client's context walk ends with an empty secondary-engine record and
the engine RM alloc is issued with class 0 (rejected -> cuCtxCreate fails).
- TU104 engine table: CE0 shares the GR PBDMA ids (the driver's GR/GRCE
runqueue validation requires it), plus an entry for the async copy engine
(COPY14) advertised in the GSP static info engine caps.
- tools/guest-diag/ioctl_trace: decode the RM status fields of alloc,
control and free so the first failing driver call is visible in traces.
gitignore: keep the CUDA analysis scratch tree (incl. WORKING-HERE.md) local.
docs: public analysis describes components and behaviour only2026-09-23T11:36:38+00:00Arthur Hanson Rasmussonarthur@vgpu.io2026-09-23T11:36:38+00:00urn:sha1:6074b13d14c9c2436519e093145164bf433bc466
Move supporting detail to the adjacent local analysis notes (gitignored;
referenced by name from the public analysis) and ignore those notes tree-wide.
No behaviour or code changes.
Untrack legacy trees and purely legacy docs (kept locally)2026-09-23T10:06:26+00:00Arthur Hanson Rasmussonarthur@vgpu.io2026-09-23T10:06:26+00:00urn:sha1:333d6de1e764edc176b0003af095cee4888ecd2c
- patches/ -> legacy-patches/ (gitignored)
- legacy-driver-guest/ (old guest vGPU driver, gitignored)
- docs/old/ (superseded architecture write-up, gitignored)
- docs that only reference the legacy components stay tracked; their path
references now point at legacy-patches/
- install.sh PATCH_DIR updated
Open-IOV: initial public release2026-09-23T09:46:43+00:00Open-IOVdev@open-iov.org2026-09-23T09:44:08+00:00urn:sha1:2302fc09080bad51c73f293294e19fd4db464eb0
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.