enricobuehlerandClaude Opus 5
560e663aef
A LocalService principal could take over a virtual pad's shared input section and
forge HID input into the interactive desktop.
The host duplicates each pad's unnamed DATA section into the driver's WUDFHost, and
through gamepad proto v2 it learned that process from `driver_pid` in the named
bootstrap mailbox. That mailbox has to be LocalService-writable — that is what the
driver's own WUDFHost runs as — and the delivery gate, verify_is_wudfhost, only checks
that the target's IMAGE is %SystemRoot%\System32\WUDFHost.exe. That image is
world-executable. So anything running as LocalService — notably the deliberately
de-privileged plugin runner — could spawn its own WUDFHost (CREATE_SUSPENDED parks it
indefinitely with the right image path), publish that pid, and be handed
SECTION_MAP_READ|WRITE on a live section. For pf-mouse that section drives a real
absolute pointer, so it was desktop control; for the pads it was forged gamepad input
plus a read of the remote user's controller state.
The module docs claimed mailbox tampering "yields at worst a gamepad DoS, never a read
or an injection". That was wrong, and the reasoning behind it — that a LocalService
token is DACL-denied OpenProcess on a UMDF WUDFHost — only covers the REAL host, not
one the attacker spawned itself.
The pid now comes from the device stack (ChannelProof, proto 2 -> 3). The host asks the
devnode it SwDeviceCreate'd who is serving it, looked up by the instance id PnP handed
back, so a planted look-alike devnode is not a candidate and the kernel — not anything
the attacker supplies — does the routing. Only the driver PnP actually bound to that
device can answer. `driver_pid` survives as a liveness hint; a tamperer can still deny a
pad, which squatting the name always allowed, but can no longer choose the recipient.
Two rules keep the state machine honest around it: a delivery stands until its target
process EXITS (judged on a retained SYNCHRONIZE handle, so a recycled pid cannot fake
it, and UMDF's restart-after-driver-crash still re-attaches), and a pad with no
SwDeviceCreate devnode refuses to deliver rather than fall back — unless an operator
sets PUNKTFUNK_PAD_CHANNEL_TRUST_MAILBOX, which says so loudly.
Three transports, because Windows carries different things to different driver shapes,
and the obvious two did not survive contact with hidclass. Measured on .173 (Win11
26200): HidD_GetIndexedString is NOT forwarded to a UMDF HID minidriver at all — it
failed for every index including ones the driver demonstrably serves through the named
wrappers; and a private device interface registers and enumerates but cannot be OPENED
(ERROR_GEN_FAILURE), because hidclass owns IRP_MJ_CREATE on a devnode it is the FDO for.
That is exactly why pf-xusb was never affected: it is not a HID minidriver, so nothing
sits above it. What works:
* pf-xusb — a private IOCTL on its own GUID_DEVINTERFACE_XUSB.
* pf-mouse — the HID serial string. Verified: PFCP:3:0:7296, and 7296 was a genuine
service-spawned WUDFHost.exe. Safe here alone: nothing reads the virtual
mouse's serial, whereas a pad's is SDL/Steam dedup material.
* pf-gamepad — a HID feature report, and it cost NO report-descriptor change. The
captured descriptors already declare far more Feature ids than the driver
ever served: 0x85 is declared on DualSense, DualShock 4 and Edge alike and
used to fail with STATUS_INVALID_PARAMETER, so hidclass lets it through and
nothing can have depended on the old failure. The Deck's one feature report
is unnumbered and Steam drives it command->response, so its proof rides that
existing contract via a private two-byte command. Verified: feature 0x85
returned magic "PFCP", proto 3, pad_index 0, wudf_pid 18456 — and 18456 was
a WUDFHost — with the product string still 'DualSense Wireless Controller'.
Also renamed pf-dualsense -> pf-gamepad. One driver has always served four identities, so
the old name read as if the other three lived elsewhere. ONLY the package identity moved
(crate, INF/CAT/DLL, UMDF service, build script, CI lines, log file, env var). The four
HARDWARE IDS are deliberately unchanged — they bind every devnode the host creates and
every installed system — as are the Global\pfds-boot-<i> mailbox and PAD_MAGIC, which are
wire contract. `driver install --gamepad` now retires the pre-rename store package first,
matched on pf_dualsense.dll because that string appears only in the OLD inf; matching on
the hardware ids would delete what we are about to install. On .173 that separated 14
stale packages from the 1 new one with 0 ambiguous, and the renamed package binds the old
hwid (devgen root\pf_dualsense -> oem143.inf = pf_gamepad.inf).
The repo's own pre-commit/pre-push rustfmt hooks named the old crate, so they caught the
rename before the commit did — they now check pf-gamepad, and pf-mouse alongside it, which
they had been missing relative to the CI line.
Host and drivers MUST ship together: v2<->v3 fails closed in both directions by design,
with the existing "update host + drivers together" diagnostic.
The rename moved files that also carry the security change, so splitting this into two
commits would mean reconstructing an intermediate state that was never gated. It is one
commit on purpose.
Gated on the windows-amd64 runner with cargo clean first (the box's clock lags, so stale
artifacts would read as a vacuous green): clippy -D warnings clean for pf-inject,
pf-capture and pf-driver-proto, drivers workspace build + the CI clippy line clean,
cargo check --release -p punktfunk-host clean, 19 + 58 tests green. Also fixes pf-mouse
still writing its debug log to world-writable C:\Users\Public, which the 2026-07-17
review moved for the other three drivers and missed here.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Packaging punktfunk for Fedora / Bazzite
The punktfunk host links system FFmpeg (NVENC on NVIDIA, VAAPI on AMD/Intel, with a GPU-less software-H.264 fallback), PipeWire and Opus. This page covers packaging it for the Fedora Atomic / Bazzite world (rpm-ostree + bootc), where most of those deps are already present; the NVIDIA-specific notes below apply to the NVENC path.
👉 Ubuntu/Debian hosts install via
aptfrom Gitea's package registry — seedebian/README.md(apt update && apt upgradefor new builds).
👉 End-to-end Bazzite setup walkthrough (install → udev/group →
host.env→ service → firewall → verify → troubleshooting):bazzite/README.md. This file is the higher-level packaging rationale.
packaging/
rpm/punktfunk.spec # the RPM (builds punktfunk-host from source with cargo)
bazzite/host.env # gamescope-default config for a Bazzite appliance
bazzite/README.md # step-by-step Bazzite setup guide
bazzite/*sysext*.sh # the no-layering path: build/install/publish the systemd-sysext
bootc/Containerfile # bake punktfunk into a Bazzite-based atomic image
copr/ # COPR build-from-SCM settings
The other packaging targets have their own READMEs: debian/ (apt),
arch/ (pacman binary repo + PKGBUILD + SteamOS sysext),
flatpak/ (the client), windows/ (host installer +
drivers), plus kde/ and linux/ helpers. NixOS / Nix users get a flake (flake.nix at the
repo root) with reproducible host + client packages and a services.punktfunk NixOS module —
see nix/README.md.
What's needed beyond base Fedora
| Dependency | Where it comes from |
|---|---|
ffmpeg-libs with NVENC |
RPM Fusion nonfree (ffmpeg, not ffmpeg-free) |
NVIDIA driver (libnvidia-encode, libEGL_nvidia) |
Bazzite -nvidia images ship it; plain Fedora: akmod-nvidia + xorg-x11-drv-nvidia-cuda |
| gamescope, PipeWire, wireplumber | Bazzite ships these; plain Fedora: dnf install gamescope pipewire wireplumber |
opus, libei |
Fedora base / updates |
On Bazzite the only genuinely new runtime bits are ffmpeg-libs (RPM Fusion) + opus +
libei — the rest of the stack is already there. The default backend is gamescope
(packaging/bazzite/host.env), which the host spawns headless per session — no desktop login.
Option A — systemd-sysext (recommended; no layering, no reboot)
On Bazzite / Fedora Atomic the recommended install is the systemd-sysext image — rpm-ostree
layering is a last resort per the Bazzite docs (it slows every OS update and can block upgrades),
while a sysext overlays /usr at runtime, survives OS updates, and updates in one command with
no reboot. CI wraps the same RPMs below into the image, so content and channels are identical.
curl -fsSLO https://git.unom.io/unom/punktfunk/raw/branch/main/packaging/bazzite/punktfunk-sysext.sh
sudo bash punktfunk-sysext.sh install # then: sudo punktfunk-sysext update | status | remove
Full walkthrough (incl. the F43→F44 rebase behavior and migration off layering):
bazzite/README.md.
Option B — Gitea RPM registry (per-host, rpm-ostree layering)
The host's RPM is published to unom's self-hosted Gitea RPM registry (CI builds it on every
push), mirroring the Debian/apt setup. Add one repo file, install, and track
updates with rpm-ostree upgrade — no COPR account needed. Full guide: rpm/README.md.
# GPG-signed pkgs + Gitea-signed metadata → gpgcheck=1, repo_gpgcheck=1 (see rpm/README.md)
sudo tee /etc/yum.repos.d/punktfunk.repo >/dev/null <<'REPO'
[gitea-unom-bazzite]
name=punktfunk (unom, Bazzite)
baseurl=https://git.unom.io/api/packages/unom/rpm/bazzite
enabled=1
gpgcheck=1
repo_gpgcheck=1
gpgkey=https://git.unom.io/api/packages/unom/rpm/repository.key
https://git.unom.io/api/packages/unom/generic/punktfunk-keys/1/RPM-GPG-KEY-punktfunk
REPO
rpm-ostree install punktfunk && systemctl reboot
# updates: rpm-ostree upgrade && systemctl reboot
Option C — COPR (per-host, rpm-ostree install)
- Create a COPR project, enable build-from-SCM pointing at this repo, spec path
packaging/rpm/punktfunk.spec(seecopr/README.md). Under External Repositories add RPM Fusion nonfree soffmpeg-develresolves at build time. - On the Bazzite host:
# RPM Fusion (for the NVENC ffmpeg) — usually already enabled on Bazzite rpm-ostree install \ https://mirrors.rpmfusion.org/free/fedora/rpmfusion-free-release-$(rpm -E %fedora).noarch.rpm \ https://mirrors.rpmfusion.org/nonfree/fedora/rpmfusion-nonfree-release-$(rpm -E %fedora).noarch.rpm # enable the COPR + install punktfunk sudo wget -O /etc/yum.repos.d/_copr_punktfunk.repo \ https://copr.fedorainfracloud.org/coprs/enricobuehler/punktfunk/repo/fedora-$(rpm -E %fedora)/ rpm-ostree install punktfunk systemctl reboot
Option D — bootc (image-based, atomic)
Layer punktfunk into a Bazzite image once, then rebase any number of hosts onto it — no
per-host drift. See bootc/Containerfile:
podman build -t ghcr.io/<you>/bazzite-punktfunk -f packaging/bootc/Containerfile .
podman push ghcr.io/<you>/bazzite-punktfunk
# on the target:
sudo bootc switch ghcr.io/<you>/bazzite-punktfunk && systemctl reboot
First-run setup (all options)
ujust add-user-to-input-group # virtual gamepads need /dev/uinput (then re-login).
# On Bazzite use ujust, NOT `usermod -aG input` (atomic OS — it won't stick).
mkdir -p ~/.config/punktfunk
cp /usr/share/punktfunk/host.env.bazzite ~/.config/punktfunk/host.env # edit (gamescope app, etc.)
systemctl --user enable --now punktfunk-host
# Management web console (pairing + status) — pulled in by default (the host RPM Recommends it;
# `--no-install-recommends` / headless-only boxes can skip it). Enable it and read the login password:
systemctl --user enable --now punktfunk-web
journalctl --user -u punktfunk-web-init | sed -n 's/.*password generated: //p' # then open https://<host-ip>:47992
Pair a stock Moonlight client (mDNS-discovered), or connect the native punktfunk/1 client — via the
web console at https://<host-ip>:47992 or directly.
⚠️ COPR caveat: COPR's mock chroot has no
bun, so a COPR build produces onlypunktfunk+punktfunk-client— notpunktfunk-web. For the console on a COPR/bootc host, install from the Gitea RPM registry (Option B — it carriespunktfunk-web; the sysext image includes it too), which is also whybootc/Containerfileinstalls from there rather than COPR.
Why not Flatpak (for the HOST)?
The host needs unsandboxed access the zero-copy NVENC path, /dev/uinput, the PipeWire
graph and the compositor's privileged protocols — a Flatpak sandbox fights all of these.
An RPM (or the bootc layer) installs into the host system where those just work.
👉 The client is a different story — it IS shipped as a Flatpak (the only viable Steam Deck install path: SteamOS
/usris read-only and lackslibadwaita/libSDL3). Seeflatpak/README.md. The client sandbox only needs the GPU render node, Wayland, PipeWire audio, the network and hidraw — all expressible as finish-args.
Building the SRPM/RPM locally (Fedora only)
git archive --format=tar.gz --prefix=punktfunk-0.3.0/ -o ~/rpmbuild/SOURCES/punktfunk-0.3.0.tar.gz HEAD
rpmbuild -ba packaging/rpm/punktfunk.spec # needs the BuildRequires from the spec
# (0.3.0 = the spec's default %{pf_version}; the prefix and tarball name must match it)
(Not buildable on Debian/Ubuntu — use a Fedora toolbox/container or COPR.)