Files

T

2026-08-29 17:11:55 -05:00

17 KiB

Raw Blame History

video-output

A small live-video-to-display appliance for Fedora. It receives live video and puts it fullscreen on one chosen monitor. That is all it does.

Two jobs:

Command	Path
`video-output camera`	ATEM Mini Pro over USB → V4L2 → MJPEG decode → GPU texture → fullscreen HDMI
`video-output ndi`	ProPresenter → NDI → GPU texture → fullscreen HDMI → ATEM chroma key

No MPV, no GStreamer, no OBS, no browser, no MQTT, no audio, no recording, no playback controls. One process that a launcher starts and kills.

Status: both input backends are implemented. The camera path is verified on the production hardware. The installed NDI 6 runtime and source discovery have also been verified; live ProPresenter video still needs a sender to be broadcasting for end-to-end validation.

Why Rust, SDL3, libjpeg-turbo

Rust — the toolchain was already present, cargo test gives the test suite with no extra machinery, and a long-running unattended process that parses frames off a USB bus benefits from bounds checking. The whole dependency tree is two crates: sdl3-sys and libc. V4L2, TurboJPEG and NDI use small, hand-written FFI surfaces, so there is no binding crate to rot.
SDL3 — the one portable way to get a borderless fullscreen GPU window on a chosen Wayland output without writing raw xdg-shell code. Used through sdl3-sys (raw FFI) rather than a safe wrapper, for direct control over display selection and YUV texture formats.
libjpeg-turbo (TurboJPEG API) — SIMD MJPEG decode, and its tj3DecompressToYUVPlanes8 hands back raw YUV planes so the CPU never does a colour conversion.
NDI SDK — official discovery and receive APIs, loaded only when an NDI command runs. Audio and metadata output buffers are never requested.

The pixel path

The ATEM emits baseline 4:2:2 JPEG. Rather than decode to RGBA (4 bytes per pixel, CPU colour conversion), the planes are interleaved into packed YUY2 (2 bytes per pixel) and uploaded to an SDL YUY2 texture. SDL's selected renderer backend performs the YUV-to-output conversion. That avoids a CPU matrix conversion and halves upload bandwidth versus RGBA.

MJPEG → TurboJPEG → Y/U/V planes → interleave to YUY2 → SDL texture → GPU

4:2:0 camera sources skip even the interleave (planes go straight in as I420). Anything else (4:4:4, 4:4:0, 4:1:1, grayscale) falls back to packed BGRX.

For NDI, the receiver requests NDIlib_recv_color_format_fastest. On this platform that preserves the SDK's native packed 4:2:2 path: UYVY without alpha, or UYVA when a sender advertises alpha. UYVA begins with an ordinary UYVY plane, so the unused alpha plane is ignored and the same two-byte-per-pixel region is copied into an SDL UYVY texture. The ATEM keys the green background; alpha is not needed.

ProPresenter NDI → libndi fastest UYVY/UYVA → SDL UYVY texture → selected output

SDL accepts the packed UYVY texture and chooses how to convert it for the active renderer. The application does not assume that every SDL renderer implements that conversion in a particular shader or GPU stage.

Latency

Three places prevent stale data from accumulating, all on purpose:

Driver queue — each wakeup drains every queued V4L2 buffer and decodes only the newest, returning the rest unread.
NDI receive queue — after a blocking capture wakes, immediately available video frames are drained and freed; only the newest is copied for display.
Frame slot — a one-deep mailbox between the input thread and the renderer. Publishing a second frame throws the first away.

A backlog cannot grow in application memory. If capture, decoding, or display falls behind, stale frames are dropped instead of replayed later.

Fedora dependencies

sudo dnf install -y SDL3-devel turbojpeg-devel gcc

Package	Why
`SDL3-devel`	headers and the `libSDL3.so` link symlink (the `SDL3` runtime package alone is not enough to build)
`turbojpeg-devel`	`libturbojpeg.so` for the MJPEG decoder
`gcc`	the linker driver cargo invokes

Rust, if not present:

sudo dnf install -y cargo rust      # or: https://rustup.rs

You must be in the video group to open capture devices:

groups | grep -q video || sudo usermod -aG video "$USER"   # then log out and back in

NDI runtime

Download the NDI SDK for Linux from ndi.video and install its runtime library on the system loader path. On the production Fedora machine it is:

/usr/lib64/libndi.so.6

For an unpacked runtime elsewhere, set NDI_RUNTIME_DIR_V6 to the directory containing libndi.so.6. The SDK is not redistributable, so neither its headers nor runtime are vendored here.

The executable does not link against libndi. NDI commands load the runtime with dlopen/dlsym and resolve only the official find/receive entry points they use. Camera, display, device, help, and version commands therefore continue to work when NDI is absent. An NDI command instead exits with a clear NDI-specific loader error.

Build

cargo build --release
sudo install -m755 target/release/video-output /usr/local/bin/

The release binary remains small and does not contain or bundle the NDI SDK.

Discovering inputs and displays

Capture devices

video-output devices

/dev/video0
    card    : ATEM Mini Pro: Blackmagic Desig
    driver  : uvcvideo
    bus     : usb-0000:67:00.3-1
    usable  : yes, video capture + streaming
    [0] MJPG (Motion-JPEG)  [needs JPEG decode]
          1920x1080  @ 29.97 fps

/dev/video1
    card    : ATEM Mini Pro: Blackmagic Desig
    (not a video capture node -- nothing to stream from here)

This is the same information as v4l2-ctl --list-formats-ext, without needing v4l-utils installed.

On the ATEM Mini Pro specifically: /dev/video1 is a metadata node, not a second video node. It exposes no formats at all, so there is no uncompressed (YUYV/NV12) alternative on this hardware — MJPEG is the only option. The code is structured so an uncompressed device would only need a new branch in pixel_format_for_fourcc, not a restructure.

NDI sources

Ask the NDI finder to collect advertisements for five seconds:

video-output ndi-sources

The command prints each full, human-readable name exactly as NDI advertises it. Use that full name for production configuration:

1 source(s):
  [0] PROPRESENTER-PC (ProPresenter Lower Thirds)  ndi://...

Discovery time can be set from 1 to 300 seconds. Repeat --extra-ip when a sender is reachable by unicast but mDNS discovery does not cross the network boundary:

video-output ndi-sources --timeout 10 \
    --extra-ip 10.0.0.42 \
    --extra-ip 10.0.1.42

--source uses the same deterministic matching policy as displays: an exact case-insensitive name first, then a unique prefix, then a unique substring. Ambiguity is an error rather than a guess. Omitting --source selects the first visible source, which is convenient for a bench test but not recommended for a production profile.

Displays

video-output displays

Video driver: wayland
2 display(s):
  [0] Sceptre Tech Inc Sceptre F27 1920x1080+0+1080 @ 110.00 Hz  (primary)
  [1] HXA BMD HDMI   1920x1080+0+0 @ 29.97 Hz

Important, and probably not what you expect: on KDE Plasma / Wayland, SDL reports the monitor's model name, not the connector name. You will see HXA BMD HDMI, not HDMI-A-2. Always run video-output displays and use a name from that list.

--display accepts, in order of precedence:

an exact name (case-insensitive) — --display "HXA BMD HDMI"
a unique prefix — --display Sceptre
a unique substring — --display BMD
an index — --display 1 or --display #1
primary, the default

An ambiguous fragment is an error, never a guess: picking the wrong screen mid-service is the failure this is designed to avoid. Names are strongly preferred over indices, which shift when an output is hotplugged.

Running it

Simplest possible form:

video-output camera --device /dev/video0 --display "Sceptre Tech Inc Sceptre F27"
video-output ndi \
    --source "<exact name printed by video-output ndi-sources>" \
    --display "HXA BMD HDMI"

Full options: video-output --help.

Profiles

So the launcher only needs a stable word:

sudo mkdir -p /etc/video-output
sudo cp examples/multiview.toml /etc/video-output/
sudo cp examples/lower-thirds.toml /etc/video-output/
video-output multiview
video-output lower-thirds

A profile is a flat key = value file using the same names as the long options, plus mode = "camera" or mode = "ndi". Searched in order:

./<name>.toml
$XDG_CONFIG_HOME/video-output/<name>.toml
~/.config/video-output/<name>.toml
/etc/video-output/<name>.toml

Flags given after a profile name override it:

video-output multiview --display Sceptre

Unknown keys are rejected at startup rather than ignored — a typo should fail before the service, not during it.

From a launcher

It is an ordinary foreground process:

start it, and it opens fullscreen on the chosen output
SIGTERM (or SIGINT/SIGHUP) makes it exit cleanly
a missing or broken media source never closes the output window
fatal configuration, display-initialisation, or NDI-runtime errors return a nonzero status instead of retrying something that cannot recover

video-output multiview &
LAUNCHER_PID=$!
# ...
kill -TERM "$LAUNCHER_PID"

Nothing needs to be restarted because the camera was unplugged or ProPresenter stopped sending; it reconnects on its own. An optional systemd template is in packaging/[email protected].

Display behaviour

Borderless, fullscreen, no title bar, no controls, no OSD, cursor hidden, screensaver disabled. Fullscreen is fullscreen-desktop: it adopts the output's current resolution rather than forcing a mode change, so the HDMI link is never renegotiated on a live output.

Aspect ratio is preserved by SDL's letterbox logical presentation, so a 16:9 source fills a 16:9 output exactly and is never stretched.

Wayland gives a client no way to place a window, only to request fullscreen on a particular wl_output. The window is therefore created hidden, positioned on the target display, set fullscreen, and only then shown — and afterwards the program re-reads which display it actually landed on. If the compositor overrode the request you get a warning naming both displays, rather than a silently wrong monitor.

Failure behaviour

Situation	What happens
Camera missing at startup	one clear error, then quiet retries with backoff (250 ms → 5 s); the window stays up and black
Device busy	says so explicitly, keeps retrying
Permission denied	tells you to join the `video` group
Camera unplugged while running	`signal lost` logged once, black output, reopens automatically when it returns
Named NDI source absent at startup	logs the search condition once, stays black, and keeps discovering in the background
NDI source appears	creates the receiver automatically and begins displaying valid video
NDI source disappears	marks signal lost, clears to black, rebuilds discovery/receive state, and reconnects when it returns
Malformed or unsupported frame	dropped and counted; the receive process continues and stale output is not left on screen indefinitely
NDI runtime missing or incompatible	the NDI command exits nonzero with the attempted runtime path/symbol in its error; non-NDI commands are unaffected
Fatal decoder error	the JPEG handle is rebuilt so one bad frame cannot poison the ones after it
Backend panics	caught, logged, backend restarted; the display never goes down with it
No valid frame	black

Logging is edge-triggered — an input absent for twenty minutes produces a state transition, not a line every frame or discovery poll — plus one throughput line per minute (--stats <seconds>, 0 disables). Set VIDEO_OUTPUT_LOG=debug or --log debug for additional discovery and recovery diagnostics. Frame problems are counted and reported periodically instead of logged individually. Under systemd, journald severity prefixes are emitted automatically so journalctl -p warning filters correctly.

Testing

cargo test

The hardware-independent suite covers CLI and profile parsing, invalid configuration, deterministic display and NDI-source matching, reconnect policy, NDI and V4L2 ABI layouts, malformed NDI layouts, both stale-frame drop points, and the one-frame handoff. Process tests force an isolated missing NDI runtime and verify both the clear NDI error and that a non-NDI command still succeeds; on Linux, readelf also verifies that the executable has no libndi dynamic dependency.

Camera coverage remains intact: real JPEG fixtures exercise all four subsampling paths, with truncation and bit-flip fuzzing to prove that corrupt input cannot panic or wedge the decoder.

Regenerate the JPEG fixtures with python3 tests/data/generate.py (needs Pillow). They are solid green on purpose: green is what the ATEM keys on, so a wrong YUV matrix or range shows up in a test rather than on stage.

Manual camera validation on Fedora

Hardware paths that cannot be unit tested:

# 1. Devices and displays are discovered.
video-output devices
video-output displays

# 2. Windowed, so it does not take over a screen; expect ~29.97 fps and
#    "0 undecodable"
video-output camera --device /dev/video0 --display <name> --windowed --stats 3

# 3. Fullscreen on the intended output. Confirm the "Opened display:" line
#    names the monitor you meant, with no placement warning after it
video-output camera --device /dev/video0 --display <name> --stats 5

# 4. Recovery: unplug the ATEM's USB cable while it runs.
#    Expect "signal lost", black output, then "signal restored" on replug,
#    with no restart and no repeated log lines
# 5. Clean shutdown.
kill -TERM <pid>       # exits promptly, no error

Verified on this machine: Fedora 44, KDE Plasma on Wayland, ATEM Mini Pro on USB, 1920x1080 MJPEG @ 29.97 — sustained 30.0 fps decoded, 0 undecodable frames, 0 stale buffers, 0 frames dropped before display, ~33% of one CPU core and ~98 MB RSS.

Manual NDI validation on Fedora

First verify that the runtime loads and record the exact advertised name:

video-output ndi-sources --timeout 10

If the sender requires unicast discovery:

video-output ndi-sources --timeout 10 --extra-ip <ProPresenter-IP>

Then run the exact name on the production lower-thirds output:

video-output ndi \
    --source "<exact advertised ProPresenter source name>" \
    --display "HXA BMD HDMI" \
    --stats 5

Confirm all of the following:

The log names HXA BMD HDMI, the selected NDI source, and packed UYVY.
ProPresenter graphics appear fullscreen with the correct green background and no controls or OSD.
Stopping the ProPresenter NDI output logs signal loss once and turns the display black rather than freezing the last lower third.
Restarting the same output restores video without restarting video-output.
Leaving the source unavailable at startup keeps the black window alive and connects automatically when broadcasting begins.

At the last validation pass, the installed NDI 6.3.2 runtime loaded correctly but no source was advertising. Discovery/retry and the live-video path should be rechecked with ProPresenter broadcasting.

Troubleshooting

cannot find -lSDL3 or cannot find -lturbojpeg when building SDL3-devel / turbojpeg-devel are not installed. The runtime packages alone do not provide the link symlinks.

cannot open /dev/video0: Permission denied Join the video group and log out and back in.

device is busy Something else is capturing — often another copy of this program, or OBS.

no display matches "..." Run video-output displays. Remember the names are monitor models, not connector names, on KDE Wayland.

display "HDMI" is ambiguous Two outputs match that fragment. Use a longer one, or the exact name.

Warning that the compositor placed the window on the wrong display KWin overrode the fullscreen output request. Check for a KWin window rule matching app id video-output.

Low frame rate, many "stale buffers skipped" You are running a debug build. Use cargo build --release; the YUY2 interleave is several times slower without optimisation.

Log line says "decoded despite a libjpeg warning" Normal for the ATEM. It pads frames with a few bytes before the JPEG end-of-image marker, which makes libjpeg warn on an otherwise perfect image. These frames are complete and are kept. Truncated frames are caught separately by a structural check and dropped.

17 KiB Raw Blame History