Now that the per-capture worker subprocess makes an NVENC EGL/CUDA driver
fault survivable (design/zerocopy-worker-isolation.md), the reason the NVENC
zero-copy path stayed opt-in is gone. zerocopy::enabled() now defaults ON for
both GPU backends (was ON VAAPI / OFF NVENC). Fallbacks are intact: VAAPI's
one-shot CPU auto-downgrade (VAAPI-gated, never trips for NVENC) and NVENC's
per-capture fallback + worker-death latch.
Reframe the shipped host.env examples and setup guides to rely on the default
rather than force PUNKTFUNK_ZEROCOPY=1 (an explicit =1 skips the VAAPI
auto-downgrade).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Connecting reset an existing physical monitor's refresh (e.g. 120->60 Hz)
because the topology code read the physical's mode AFTER the virtual output
perturbed the compositor layout — by which point it had already been
downgraded. Read/preserve each physical's mode from a pre-connect snapshot.
- Mutter: build_primary_keeping_physicals takes the pre-virtual snapshot and
preserves each physical's real mode (pick_keep_mode, unit-tested)
- KWin: capture each output's mode when disabling for exclusive, re-assert it
on re-enable (a bare enable defaulted to ~60 Hz)
- Windows: skip the refresh-resetting SDC_TOPOLOGY_EXTEND when a physical is
already active
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Save named bundles of the display-management policy (the six behavior axes
plus the game-session axis) as custom presets, alongside the built-ins. A
custom preset is data — stored in <config>/display-presets.json — not a Preset
enum variant, so DisplayPolicy::effective() stays pure and the built-in set is
untouched; applying one writes a Custom policy via the existing PUT
/display/settings.
- policy.rs: CustomPreset/CustomPresetInput + load/add/update/delete store
- mgmt.rs: GET/POST /display/presets + PUT/DELETE /display/presets/{id},
surfaced on GET /display/settings
- web console: custom-preset cards with save-as / edit / delete + i18n
- regenerated api/openapi.json; docs
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The tiled EGL/GL→CUDA import crashed the whole host (SIGSEGV inside
libnvidia-eglcore via cuGraphicsMapResources) when the compositor
invalidated an imported dmabuf mid-map — reproduced on the Bazzite F44
Game→Desktop switch (design/zerocopy-hardening-handoff.md). A driver
SIGSEGV is uncatchable in-process, so the whole EglImporter (tiled
EGL/GL→CUDA and LINEAR Vulkan→CUDA) now runs in a per-capture
`zerocopy-worker` subprocess: dmabuf fds go over a SEQPACKET socketpair
(SCM_RIGHTS, sent once per buffer keyed by dmabuf st_ino; NeedFd resend
self-heals cache desync), frames come back as CUDA-IPC pooled device
buffers (still zero-copy, +one socket RTT/frame). Worker death poisons
the capturer so the existing capture-loss rebuild runs — the host
survives; 3 consecutive deaths latch the GPU import off (CPU/SHM path).
PUNKTFUNK_ZEROCOPY_INPROC=1 keeps the old in-process import for
debugging/A-B.
Also fixed along the way: a failed *tiled* import no longer falls
through to the CPU mmap de-pad (which scrambled tiled bytes; LINEAR
keeps the fallback); Nv12Blit dropped its GL textures while still
CUDA-registered (unregister now runs first); GlBlit had no Drop at all
(GL objects leaked per size change); VkBridge's per-fd src cache is now
invalidated on renegotiation/eviction instead of never.
Design: design/zerocopy-worker-isolation.md. Unit tests: 14 new
(protocol fd-passing, worker dispatch, client handshake/death/NeedFd,
death latch). On-glass validated on the RTX 5070 Ti/GNOME box (.21):
the worker path streams at p50 1.30 ms (NV12, 1800 frames 0-mismatched,
parity with the in-process path), and a kill -9 of the worker
mid-stream is survived by the host and recovered — poison -> capture
lost, rebuilding pipeline in place -> a fresh worker in ~185 ms ->
streaming resumes (2385 frames, 0 mismatched). A real KWin
compositor-crash repro is still pending (a worker kill -9 is strictly
harsher, so it corroborates).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The DXGI Desktop Duplication + WGC relay paths were removed; sealed
IDD-push (finished frames pushed straight into the host's own IddCx
driver, no screen-scraping) is now the sole Windows capture path. Fix the
stale "DXGI/WGC capture" claims in the root and punktfunk-host READMEs,
which also contradicted the push-based IDD description already present in
the root README.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Close the latency gap on the Android client with per-SoC decoder tuning, an
event-driven decode loop, and full system integration.
- Decoder selection: rank MediaCodecList decoders in Kotlin (hardware/vendor
preferred, software avoided, FEATURE_LowLatency probed) and create the chosen
one by name. Per-SoC low-latency keys gated on the codec-name prefix: Qualcomm
picture-order + low-latency, Exynos (also Google Tensor), Amlogic, HiSilicon;
MediaTek vdec-lowlatency set unconditionally. operating-rate = MAX (Qualcomm)
vs priority = 0 (else) are mutually exclusive. NVIDIA/Rockchip/Realtek have no
vendor key — covered by ranking + the standard low-latency key.
- Async decode loop: AMediaCodec async-notify replaces the poll loop, presenting a
decoded frame the instant it is ready instead of waiting out a poll interval.
Behind USE_ASYNC_DECODE with the synchronous loop kept for A/B during bring-up.
- System integration: Wi-Fi FULL_LOW_LATENCY lock and HDMI ALLM
(setPreferMinimalPostProcessing) for the stream's lifetime; game_mode_config.xml
opting out of OEM downscaling / FPS overrides.
- Pipeline: boost the data-plane pump + audio thread priorities, AAudio usage=Game,
DSCP marking on by default on Android, ADPF setPreferPowerEfficiency(false),
and setFrameRateWithChangeStrategy(ALWAYS) to force the HDMI mode switch on TV.
- lowLatencyMode master toggle (default on) as the escape hatch; the stats HUD now
shows the resolved decoder name.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The CI image's rustfmt reformats these files (multi-line assert!/tracing! macros,
match-arm and struct-variant wrapping) — pre-existing drift that the Format job caught.
Reformat to match. Pure formatting; no logic change. main.rs also gets a blank line
before a standalone comment so rustfmt stops mis-indenting it as a trailing-comment
continuation.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The >60 Hz virtual-monitor path (RecordVirtual "modes" with the client's exact WxH@Hz)
was gated behind PUNKTFUNK_MUTTER_VIRTUAL_REFRESH, default OFF, after a high-refresh
virtual CRTC SIGSEGV'd gnome-shell on session teardown. That crash was since fixed by
stopping the screencast before any monitor reconfig, so the gate is dead weight — and a
silent footgun: every non-headless GNOME client was capped at Mutter's PipeWire-derived
60 Hz unless they knew the hidden flag.
Make it the default: the custom-mode path now runs whenever mode.refresh_hz > 60 (≤60 Hz
stays byte-identical to before — Mutter's 60 Hz default is already correct), and the
virtual_refresh_enabled() env read is removed. Docs updated (configuration.md env table,
vrr-plan.md reference).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The branch's Windows-host code never ran through the fleet Windows clippy (kept off CI); the merge
to main exposed it. Fix the 4 -D warnings failures in cfg(windows) code:
- manager.rs: 3 unsafe blocks (isolate_displays_ccd / force_extend_topology / set_virtual_primary_ccd)
had one "both arms" SAFETY comment on the `match` line — clippy::undocumented_unsafe_blocks wants it
immediately before each block. Split into per-block SAFETY comments.
- win_display.rs: `.then(|| …)` on a POD u32 union read → `.then_some(…)` (eager is fine, discarded
when false) for clippy::unnecessary_lazy_evaluations.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The reconnect-preempt (e8531a9, preempt_same_identity) reads the process-global admission table
and signals same-identity live sessions. The three in-process-host tests each bind a fixed loopback
port and share that ONE table, so running them concurrently let one test's connection preempt +
close another's live session — an intermittent `next_au: Closed` in c_abi_connection_roundtrip
(surfaced under full-workspace load; a lucky pass hid it at e8531a9). Serialize them on a
poison-tolerant lock. Test-isolation only — in production a host is one process with unique client
certs, so same-identity preempt is correct. Full workspace `cargo test` now green (18 suites, 3× clean).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The old `..._and_forever_rejected` asserted a 400 for keep_alive=forever; now that it's accepted,
that PUT succeeded and WROTE gaming-rig into the process-global prefs, racing other tests. Rewrite
read-only: assert the surface (5 presets, effective, enforced axes) and read gaming-rig=forever off
the preset list — no write. Acceptance is covered on-glass (.116) + the pure policy tests.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Stage 8 polish. `GET /api/v1/local/summary` (the tray's loopback-only unauthenticated status
source) gains `kept_displays` — the count of lingering/pinned virtual displays (held with no live
session), over the already-validated `registry::snapshot()`. The tray shows it in the idle tooltip
("idle · 1 display kept"), so a user knows a display — and, under exclusive topology, their physical
monitors — is being held (e.g. a gaming-rig `forever` pin). Release stays via the console: a
state-changing release can't be an unauthenticated endpoint, and the non-elevated Windows tray
can't read the SYSTEM-DACL'd mgmt token, so a tray release button isn't cleanly cross-platform.
`#[serde(default)]` on the tray side keeps it compatible with an older host. Tray tests green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
`let (mut paths, mut modes)` — `modes` is only read (`modes.as_slice()`), never mutated. A
pre-existing unused_mut (from 8fa4757) that the Linux CI never caught because win_display.rs is
#[cfg(windows)]; surfaced by a manual .173 build. Would fail the release-gated Windows clippy.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Completes the last §6A-era preset. The Linux registry already resolved forever→Pinned (pure
lifecycle machine); the blockers were the Windows manager, the mgmt reject, and the console tag:
- Windows manager: new `MgrState::Pinned { mon }` — the last-released monitor under keep_alive=forever
is kept indefinitely (like Lingering but the linger timer never fires). A reconnect preempts +
recreates it (same as Lingering — a reused IddCx swap-chain is dead), snapshot reports "pinned",
and `force_release` (POST /display/release, the §8 escape hatch) frees a pinned monitor. release()
branches on the new `keep_alive_forever()`; all MgrState matches made exhaustive over Pinned.
- mgmt PUT /display/settings: stop rejecting keep_alive=forever (now honored on both platforms with a
release path). OpenAPI regenerated.
- web: un-disable the gaming-rig preset (DISABLED_PRESETS now empty) — one-click applies.
Linux paths + web/tsc/openapi green; 47 vdisplay tests pass. The Windows manager.rs is #[cfg(windows)]
(not compilable on the Linux dev box) — build-verified + on-glass validation on .173 to follow.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
On-glass testing (Test 2, KWin .116) surfaced that a reconnect within the QUIC idle-timeout
window (~8s) lands on a fresh SECOND display instead of reusing the kept one: the old session
was still Active (not yet Lingering), so the registry's keep-alive reuse (which only matches
Lingering) skipped it and the old session kept streaming to nobody. Three fixes:
#3 Same-client reconnect preempt (the real fix): admission::preempt_same_identity() lists a
reconnecting client's OWN still-live session(s) (same cert fingerprint); serve_session signals
their stop + waits the release grace BEFORE acquiring, so the zombie tears down → its display
lingers → the reconnect REUSES it instead of making a second. Implements the "preempts
downstream" the admission docs already promised. Independent of the mode_conflict policy; the
pure core (same_identity_stops) is unit-tested.
#2 Deliberate quit skips linger: a client that deliberately disconnects closes the QUIC connection
with QUIT_CLOSE_CODE (0x51, shared in core::quic); the host reads the ApplicationClosed reason
and tears the display down immediately (registry release() gained force_immediate →
Linger::Immediate; multi-session-safe via the pure lifecycle machine), while a bare disconnect
still lingers for reconnect. Threaded via a session quit flag → the DisplayLease.
NativeClient::disconnect_quit() + punktfunk-probe --quit drive it; GameStream (Quit App /
h_cancel) is a documented follow-up.
#1 Configurable disconnect-detection latency: the QUIC control-connection idle timeout
(stream_transport, 8s default) is host-tunable via --idle-timeout-ms / PUNKTFUNK_IDLE_TIMEOUT_MS,
clamped >=1s with a keep-alive that scales to it so a live session never false-closes. Default
unchanged (8s stays load-bearing for the Windows IDD-push reconnect flow).
Workspace check + 63 core / 215 host / 47 vdisplay tests green; clippy clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The native data plane used a random ephemeral UDP port (hole-punched), which a
strict firewall can't pre-open — so remote clients behind one couldn't connect.
Add an optional fixed data port:
- `Punktfunk1Options`/`NativeServe` gain `data_port`; `bind_data_socket` binds the
fixed port (→ direct, no hole-punch) or falls back to a random port + hole-punch
when unset or the fixed port is busy (a concurrent session already holds it).
- `UdpTransport::from_socket`/`from_socket_punch` adopt an already-bound socket, so
the host keeps the SAME data socket from handshake through streaming — no
drop-then-rebind window in which a concurrent session could steal a fixed port.
- `main.rs` wires the CLI flag through to `NativeServe`.
- Firewall docs updated (troubleshooting.md + apt/pacman/bazzite READMEs): control
plane is the fixed UDP 9777; the data plane is a separate random port that usually
needs no rule, with the fixed-port option for strict firewalls.
Unit-tested: default random+hole-punch, and fixed-port-then-fallback-when-busy.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
FEC/Reed-Solomon packetization ran inline on the encode loop (~3 ms/frame at 4K),
serializing behind encode and capping the GameStream frame rate below what the
encoder alone can sustain. Split it into a 3-stage pipeline, each stage on its own
thread joined by a depth-2 bounded queue:
encode loop → [raw AUs] → packetizer (FEC/RS) → [wire batch] → paced sender
- `spawn_packetizer`: turns each `RawFrame`'s access units into wire datagrams via
the stateful VideoPacketizer, off the encode loop. Above-normal priority (on the
per-frame critical path). Tallies goodput (bytes to the wire) for the stats window.
- Backpressure chains up: a slow sender blocks the packetizer, which fills the
encode→packetizer queue, which makes the encode loop drop the NEWEST frame — encode
itself never waits.
- A dropped frame now consumes no client-visible frameIndex (packetization is
downstream), so the host re-anchors the reference chain: a drop arms a keyframe on
the next iteration (`recover_after_drop`), routed through the same coalesce gate as
client IDR requests so a burst of drops (congestion) can't become an IDR storm.
- Perf/stats relabeled: `pkt` = AU drain, `send` = enqueue to the pipeline (both
should be near-zero now; nonzero = encode being stalled by pipeline backpressure).
Goodput read from the packetizer's atomic at the 1 s stats boundary.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Host-side completion of Stage 5 (§6A many-clients-as-monitors), all unit-tested;
two-session on-glass validation still pending (no GPU on the dev VM):
- Per-group topology restore (§6.1): the KWin `exclusive` restore no longer rides
the per-session StopGuard (which re-enabled the physical the moment the FIRST of
several exclusive sessions dropped, under a live sibling). KWin hands its restore
to the registry as a closure (new trait `take_topology_restore`); the registry
keeps it in the display group (`Entry.topology_restore`) and, on teardown, floats
it to a surviving same-group sibling (`hand_off_restore`) or runs it when the group
empties — outside the lock, before the last output's keepalive drops, so the
compositor never sees zero outputs. All three teardown paths (lease drop / linger
expiry / mgmt release) honor it. Single-display path byte-for-byte unchanged.
Unit-tested: float / run-on-last / non-carrier-first / never-cross-backend.
- Mutter group-aware (new trait `set_first_in_group`): the registry tells each
backend whether it's the first display of its group; a non-first Mutter session
EXTENDS into the already-exclusive desktop instead of re-applying a sole-monitor
ApplyMonitorsConfig that would disable the first session's virtual. (Mutter
connectors are un-nameable, so it can't build a keep-all-virtuals config; skipping
is the safe equivalent.) Single-session unchanged. Residual APPLY_TEMPORARY revert
documented.
- gamescope groups (§6.1): `registry::group_key` makes each gamescope spawn its own
group (independent nested session, no shared desktop) — never auto-rowed against or
restore-/topology-grouped with another gamescope. Applied in both the /display/state
assembly and the acquire-time position computation. Unit-tested.
Remaining Stage 5: the web console arrangement table, on-glass validation, and the
documented residuals (wlroots exclusive, Mutter APPLY_TEMPORARY). design doc updated.
cargo build/test (214)/clippy --all-targets/fmt green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
§6A layout, riding the Stages 1-3 registry with no protocol change:
- vdisplay/layout.rs: pure arrangement engine — auto-row (left-to-right in
acquire order, top-aligned) + manual (per-identity-slot offsets, auto-row
fallback for unpinned members). Unit-tested.
- Registry group model (Linux): group = backend (one desktop per compositor
session). /display/state groups entries, orders by acquire (gen), and computes
each member's position via the engine (pure `assemble_displays`, unit-tested).
DisplayInfo carries group/display_index/position/identity_slot/topology. The
backend reports its resolved slot via the new VirtualDisplay::last_identity_slot
(KWin only), so the arrangement + state key on per-client identity.
- Registry-driven position apply: new VirtualDisplay::apply_position(x,y) (default
no-op; KWin drives kscreen-doctor). Right after create the registry computes the
new display's position over its whole group (pure `position_for_new`, unit-tested)
and applies it — one seam for BOTH deterministic auto-row AND manual placement.
Guarded: the origin (0,0) is skipped, so a single-display / first-of-group session
(and every non-KWin backend) issues no positioning — the historical single-display
path is unchanged. On-glass-validation-pending.
- PUT /api/v1/display/layout: persists the console's manual arrangement via the pure
EffectivePolicy::with_manual_layout transform (locks current effective behavior
into explicit Custom fields + sets a manual layout, so arranging is orthogonal to
the other axes). OpenAPI regenerated.
- /display/settings `enforced` now lists all five axes (keep_alive, topology,
mode_conflict [Stage 4], identity [Stage 3], layout [Stage 5]) — was stale at
keep_alive+topology; the console reads it to know which controls are live.
Still Stage-5 TODO (design/display-management.md §11): Mutter/wlroots group-aware
analogues, per-group topology restore, the web arrangement table, gamescope decline.
cargo build/test/clippy/fmt green; OpenAPI in sync.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The critical latent bug Stage 3 introduced: per-slot output names mean a 2nd
exclusive session's other_enabled_outputs() (which disabled 'everything not named
Virtual-punktfunk') would black out the 1st session's Virtual-punktfunk-<id>
output. Fix: recognise the whole managed group by the shared Virtual-punktfunk
prefix — exclusive now disables only NON-managed outputs (bootstrap/physical),
never a group sibling. Plus first-slot-wins for the group primary
(a_managed_output_is_primary): a later session joins as a secondary monitor of the
shared desktop instead of stealing the shell off the first. Unit-tested.
Start of Stage 5 (§6A many-clients-one-desktop). Remaining: Mutter/wlroots
group-aware analogues, layout (auto-row/manual + /display/layout + console),
per-group topology restore, gamescope groups.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Two concurrent Windows sessions both drive the same pf-vdisplay monitor's
single-capturer IDD-push channel (newest-delivery-wins), which freezes the live
client and can wedge the driver (observed live: a concurrent-session test wedged
.173 → Moonlight 'no video'; needed a reboot). True multi-session capture is §6.6/
Stage 7. So on Windows 'separate' (incl. the unconfigured default) now resolves to
REJECT — a 2nd client gets a clean 503 and the live session is protected — instead
of join (which would freeze it). join/steal stay explicit opt-ins; Linux keeps
separate (real multi-view). Centralized as admission::effective_conflict(), shared
by the native handshake + GameStream h_launch.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Pull the GameStream mode-conflict decision out of h_launch into a pure
gamestream_admission(live, req_fp, policy) -> GsDecision so the 503/join/take-over
logic is unit-tested (no live session / same-client → Serve; different client →
Reject/Join/Serve per policy; anonymous requester treated as different) — the
GameStream path can't be driven without a Moonlight client, so this covers the logic.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Completes the mode-conflict admission surface deferred from the initial Stage 4:
- REJECT now delivers the reason to the client: punktfunk/1 closes the QUIC
connection with a distinct BUSY code (0x42) + the 'host busy: streaming WxH@Hz to
<client>' string, which the client reads from ApplicationClosed (validated on
loopback: the probe logs 'closed by peer: host busy … (code 66)').
- Windows default: separate (incl. the unconfigured default) resolves to JOIN — the
Windows native host admits a second client at the live mode instead of the old
silent last-wins reconfigure of the shared monitor (release-note behavior fix; the
reconfigure is now opt-in as steal). separate stays multi-view on Linux.
- GameStream 503: h_launch tracks the session owner fp (LaunchSession.owner_fp, kept
[u8;32] for Copy) and applies the policy when a DIFFERENT paired client launches —
reject → 503 (Moonlight 'host busy'), join → serve the live mode, steal/separate →
take over. Same-client re-launch is never a conflict.
Native reject-reason loopback-validated; Windows join-default pending .173 rebuild;
GameStream 503 pending a Moonlight client (can't drive /launch autonomously).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The mode_conflict policy is now enforced at ADMISSION, before the punktfunk/1
Welcome, when a DIFFERENT client connects while another client's session is live:
- separate (default, unconfigured → no change): each client its own display.
- join: admit at the live display's mode (honest-downgrade — the Welcome carries it).
- steal: signal the victim session(s)' stop flags, wait the release grace, serve.
- reject: refuse the handshake with a busy reason (live mode + client label).
New vdisplay/admission.rs: the pure decide() (unit-tested — same-client never
conflicts, anonymous clients each distinct, join targets the oldest session) + a
live-session registry (identity + mode + stop flag) sessions register in once up.
Wired into punktfunk1 serve_session: admit() before validate_dimensions, register
after the data plane binds. A same-client reconnect never conflicts.
Validated on loopback (two probes, distinct identities, differing modes) across all
four policies: separate→own mode, join→live mode, steal→victim interrupted,
reject→handshake refused.
Remaining Stage-4 surface (deferred): GameStream 503 path, Windows-specific
defaults (separate→join map, silent-reconfigure→steal), reject reason delivered to
the client as a typed message (currently host-side log + connection close).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Exclusive (topology=exclusive) was fire-and-forget — a field-reported bug had a
physical monitor STAY ACTIVE. isolate_displays_ccd now re-queries after each apply
and RETRIES (up to 4x) until count_other_active()==0, never trusting rc alone;
logs SOLE-active on success, an error if a display survives all attempts. Secure
desktop correctness depends on the lock screen not landing on a stray panel.
Primary: drop the temporary per-path diagnostic; pack the kept displays left-to-
right from the virtual's right edge instead of blindly shifting each by virt_width
(which left a dead gap when extend already placed them right).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Root cause: on a headless box the IDD auto-activates as the SOLE display, so
QueryDisplayConfig sees only the virtual — the physical is already deactivated
before set_virtual_primary_ccd runs (no physical to keep). Force EXTEND first to
reactivate every connected display alongside the virtual, then reposition to make
the virtual primary, keeping the physical active.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Temporary diagnostic — the physical monitor goes black in topology=primary
despite rc=0; the SSH/session-0 view can't see the real interactive-session
topology, so log the active paths the host actually operates on.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
modeInfoIdx lives in the Anonymous union (windows-rs), not directly on
sourceInfo — set_virtual_primary_ccd now reads .Anonymous.modeInfoIdx.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Implements the deferred Windows primary-only CCD (Stage 2). set_virtual_primary_ccd
repositions the virtual output's source to (0,0) = primary and shifts the physical
display(s) to its right, ALL kept active — one atomic CCD SetDisplayConfig (not GDI
CDS_SET_PRIMARY, which storms MODE_CHANGE_IN_PROGRESS with another display live).
The manager's should_isolate() becomes topology_action() (3-way): extend (skip),
primary (set_virtual_primary_ccd), exclusive (isolate_displays_ccd). Restore-on-teardown
covers both. Validates the user's two scenarios on a physical-monitor .173.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The KWin backend names its output Virtual-punktfunk-<id> from the client's
stable identity slot, so KWin persists per-output config (scale/mode) by name in
kwinoutputconfig.json and reapplies that client's scaling on reconnect — the KDE
scaling ask. Also fixes the latent clash where two concurrent sessions both used
Virtual-punktfunk (topology name-matching now uses the per-slot name).
- identity::global() + resolve_slot(fp, mode, default) — the shared persisted map
(Windows manager dropped its own field; both use the global — never same-process).
Default identity is per-platform: PerClient on Windows, Shared on Linux, so
unconfigured hosts keep today's behavior (Linux = single 'punktfunk' name).
- KwinDisplay carries the client fp (set_client_identity), computes the per-slot
name, threads it through the stream_virtual_output name + the topology helpers
(set_custom_refresh / apply_virtual_primary[_only] / other_enabled_outputs).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Generalize the Windows-only per-client stable-id map into vdisplay/identity.rs:
- DisplayIdentityMap keyed on a composable string (identity_key: fingerprint,
or fingerprint+resolution under per-client-mode); LRU at 15, persisted to
display-identity.json (migrated from the legacy pf-vdisplay-identity.json).
- Windows manager wired to it, picking the key from the identity policy.
- Foundation for KWin per-slot output naming (persistent KDE scaling) — the
KWin wiring is the next Stage-3 step (needs a KWin box).
- Unit-tested (stable, per-client-mode split, LRU, key composition).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The three topology levels become distinct behaviors (Stage 0 only did
extend-vs-exclusive, faking primary):
- vdisplay::effective_topology() -> the concrete level (console policy > legacy
*_VIRTUAL_PRIMARY env > Auto default). Backends read it directly at create
time; apply_session_env no longer writes the boolean env (one fewer connect-
path env mutation).
- Mutter: extend (no config), primary (virtual primary + physicals kept as
secondaries — build_primary_keeping_physicals), exclusive (sole, physicals
disabled). KWin: extend (no-op), primary (kscreen primary only), exclusive
(primary + disable others).
- Windows should_isolate treats primary as isolate (the primary-only CCD variant
is a follow-up); wlroots exclusive + the physical-keep effect need a
display-attached box (headless lab boxes can't observe primary vs exclusive).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The pooled entry's lifecycle transition was inside debug_assert_eq!, whose
arguments don't evaluate in release builds — so acquire() never ran, the entry
stayed Idle, and release saw Noop → immediate teardown (no keep-alive). Caught
on-glass on the CachyOS box.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The ownership split (design/display-management.md §3): the registry owns the
per-session virtual-display lifecycle on Linux, so a display can outlive its
session (keep-alive) and be reused on reconnect.
- registry.rs: a Linux pool driven by the pure lifecycle machine. acquire()
reuses a kept (lingering/pinned) display of the same backend+mode, else
creates one and keeps the backend's keepalive so the compositor output (and
its PipeWire node_id) survives the session. The session's capturer holds a
gen-stamped DisplayLease instead of the real keepalive; its drop drives
linger/teardown. Enabling fact: KWin/Mutter/gamescope put their node on the
DEFAULT PipeWire daemon (remote_fd=None) — reconnect re-attaches by node_id,
no fd re-open. wlroots (remote_fd=Some, xdpw portal) passes through unchanged
(teardown-on-drop) pending the fresh-portal-capture re-attach.
- Default (unconfigured) linger = Immediate → today's teardown-on-disconnect,
so no behavior change without a keep-alive policy; concurrent sessions still
each create their own output (reuse only matches LINGERING entries).
- Wired build_pipeline (punktfunk1) + gamestream through registry::acquire;
capture_virtual_output signature unchanged. Windows delegates to vd.create
(the manager already leases) — unchanged.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Covers the idle path (empty /display/state + released:0 /display/release) on a
unit-test host, exercising the wiring + auth without touching any global owner.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Stage 1 of design/display-management.md — the lifecycle core + the display
management surface:
- vdisplay/lifecycle.rs: pure per-slot state machine (Idle/Active{refs}/
Lingering{until}/Pinned) with acquire/release/expiry/force-release
transitions. No I/O, no OS types — the platform-neutral distillation of the
Windows manager's model. Unit + a 200k-iteration seeded property walk
(no leaks / double-frees / refcount underflow across arbitrary interleavings).
- vdisplay/registry.rs: neutral snapshot/release facade over the per-OS
lifecycle owners. Windows reads/controls the VirtualDisplayManager; Linux
keep-alive (a per-session pool) lands in a following increment (needs GPU-box
validation).
- windows/manager.rs: additive snapshot() + force_release() (no behavior change
to the on-glass-validated path).
- mgmt: GET /api/v1/display/state (live/kept displays) + POST /api/v1/display/release
(tear down lingering/pinned now; refuses active). OpenAPI regenerated.
- web console: Virtual displays card gains a live-display list (polled) with
per-row + release-all buttons and a linger countdown.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A user-configurable policy layer above the per-compositor VirtualDisplay
backends: keep-alive, topology, conflict, identity, layout, max-displays —
persisted to display-settings.json, editable from the web console, applied
per connect. Design: design/display-management.md.
Stage 0 stands up the surface and wires the two behaviors the existing code
can already express — the Windows monitor linger duration and the
"make the streamed output the sole desktop" topology — through it; every
other option is stored + echoed but not yet enforced (later stages). An
unconfigured host (no display-settings.json) keeps today's exact behavior.
- vdisplay/policy.rs: pure DisplayPolicy + 5 presets + JSON store (gpu-settings
pattern) + EffectivePolicy; 9 unit tests.
- vdisplay.rs: resolve_topology(Auto); apply_session_env drives *_VIRTUAL_PRIMARY
from the policy only when a settings file exists.
- windows/manager.rs: linger_ms() + should_isolate() read the policy when configured.
- mgmt: GET/PUT /api/v1/display/settings (bearer-only); PUT rejects keep_alive
forever until the lifecycle stage. OpenAPI regenerated.
- web console: Host → Virtual displays card (preset picker + custom fields); en+de.
- docs-site: virtual-displays.md + configuration.md cross-links.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
ABI_VERSION was doing double duty: the embeddable C surface AND the punktfunk/1
Hello/Welcome version that hosts equality-check. The WoL feature's v3 bump added
a client-local FFI function without changing a single wire byte — and every new
client started refusing against every deployed host ("ABI mismatch: client 3
host 2", observed live Deck → Bazzite). The wire now carries its own
WIRE_VERSION (still 2); ABI_VERSION stays 3 for the C header and the mgmt API's
informational field. Bump WIRE_VERSION only when the handshake/planes actually
change incompatibly.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The Wake-on-LAN batch landed with lints that fail `clippy -D warnings`
(doc continuation, char-array split, io::Error::other, redundant closure)
and an ungated `mod wol;` in the Windows client, which pulls windows-only
crates into the non-Windows stub build.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Add a runtime-free Wake-on-LAN sender in punktfunk-core (per-interface subnet-directed broadcast + 255.255.255.255 on ports 9/7, repeated, optional last-known-IP unicast) exposed both as a Rust fn and a punktfunk_wake_on_lan C-ABI (ABI v3), plus a parse_mac helper. The host enumerates its wake-capable NIC MAC(s) and advertises them in a new mDNS `mac` TXT record (routed NIC first), and best-effort detects & warns (never modifies) when the NIC isn't armed for WoL.
MAC delivery is via the unauthenticated mDNS TXT rather than the connection handshake by design: a spoofed MAC only makes a wake fail (the packet is inert; the cert fingerprint still gates the connection), and it avoids threading through the hot connect path.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The Steam-conflict gate scanned /sys/bus/hid/devices for non-virtual 28DE
devices, but the usbip/gadget virtual Decks present a REAL USB device (vhci
resolves through vhci_hcd, not /devices/virtual/) — so a just-ended session's
pad still detaching, or a concurrent session's live one, read as "physical
Steam controller attached" and degraded every back-to-back Deck session to
DualSense (observed live on Bazzite). Exclude our pads by their PFDK… serial
(HID_UNIQ), with the vhci_hcd path as belt and braces.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Video (Deck): the VAAPI zero-copy path renders corrupt/gray/washed-out on the
Deck — root-caused to Mesa >= 25.1 exporting radeonsi VCN decode surfaces TILED
(the Flatpak runtime's Mesa 26 drives both the decoder and GTK's GL, and GTK's
tiled-NV12 dmabuf import mishandles it; desktop Tier-1 validations ran distro
Mesa with linear export). `auto` now resolves to software on a Deck (clean,
correct-colour, easily handles 1280x800 HEVC); PUNKTFUNK_DECODER=vaapi still
forces the hw path, with the descriptor modifier dump + GSK_RENDERER as the
bisect levers. Also reserve extra_hw_frames=4 on the VAAPI decoder: the
presenter pins mapped surfaces past receive_frame, and the fixed pool recycling
a surface the renderer still samples is intermittent block corruption anywhere.
Input (Deck): with Steam Input ON for Punktfunk, SDL sees only Steam's virtual
X360 pad — the right trackpad arrives as a plain right stick and the left
trackpad/paddles/gyro not at all, silently. The client now checks once the
post-attach enumeration settles and raises a toast + warn naming the fix
(disable Steam Input for the shortcut). The host logs a one-shot warning when
InputPlumber is running (Bazzite default) since it can grab the virtual Deck
pad and re-emit it under a different identity.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Mac → Windows mic passthrough crackled heavily while the identical
stream was clean on the Linux host. Cause: clients push mic audio in
BURSTS on their own clock (the Mac input tap yields ~two 20 ms Opus
packets every ~42 ms) while the WASAPI render loop pulled a block every
~10 ms device period and greedily drained whatever was queued, padding
the rest with zeros — the queue sat near-empty and most periods
inserted mid-stream silence. The Linux backend has absorbed this since
day one with its priming jitter buffer; the WASAPI loop had none.
Port the same semantics: emit silence until ~48 ms is buffered (covers
the worst inter-burst gap), then play from the cushion (zero-filling
only a momentary shortfall), re-prime only after a genuine full drain
(client went quiet). Queue cap raised 80 → 120 ms for burst headroom;
steady-state added latency ≈ the 48 ms cushion.
Diagnosed live on .173: probe tone recording from CABLE Output proved
the endpoint wiring, then the burst-vs-period math explained the
crackle. Build-verified on Windows; on-glass listen pending.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The unified-stats equation's host+network stage was one opaque number
because the wire carried nothing but pts_ns. Now the host reports its own
share per frame: when the client's Hello sets VIDEO_CAP_HOST_TIMING (0x08),
the send thread emits a 13-byte 0xCF datagram — [tag][pts_ns u64][host_us
u32] — right after the AU's last packet leaves the socket, so host_us =
capture→fully-sent (capture read/convert, encode, FEC+seal, paced send)
against the same anchor the wire pts carries. Clients correlate by pts_ns
and derive network = (received + clock_offset − pts) − host_us; the two
terms tile per frame by construction.
Back-compat is free in all four combinations: old clients ignore unknown
datagram tags, old hosts ignore unknown cap bits (client keeps the combined
stage). The hardened data-plane format is untouched — this rides the
established QUIC side-plane pattern (0xC8…0xCE). NativeClient ORs the bit
in unconditionally and exposes next_host_timing(); the C ABI gains
PunktfunkHostTiming + punktfunk_connection_next_host_timing (additive).
The synthetic host emits 0xCF too, so pure-loopback protocol tests cover
the plane.
The probe reports the split (host_p50/p95_us · net_p50/p95_us) and is our
direct analogue of Sunshine's "host processing latency" — ours additionally
includes the paced send.
Validated on loopback (synthetic host + probe, debug build): 240/240 AUs
matched, host_p50 6.5 ms + net_p50 6.4 ms ≈ capture→received p50 13.0 ms.
Core suite + new 0xCF roundtrip/truncation test green; host+core+probe
clippy clean.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
One stat model everywhere (design/stats-unification.md): four measurement
points (capture/received/decoded/displayed), three stages that tile the
interval exactly, and a HUD that shows the addition explicitly —
end-to-end 14.2 ms p50 · 19.8 p95 · capture→on-glass
= host+network 9.8 + decode 2.1 + display 2.3
replacing each client's ad-hoc mix of overlapping absolutes (the Apple HUD's
three arrow lines that looked sequential but weren't), mean-vs-median decode
times (Windows/Linux), missing same-host-clock flags (Windows/Linux), and
three different names for the same capture→received measurement (probe's
"reassembled", Apple/Android's "client", Windows/Linux's post-decode "lat").
Per client: Apple threads receivedNs through the VT decode via the frame
refcon bit pattern so the decode stage exists at all (stage-1 fallback
honestly degrades to a capture→received headline); Windows carries
FrameTimes through the existing frame channel to the render thread and adds
e2e p50/p95 post-Present; Linux stamps received at AU pop and rides
decoded_ns on DecodedFrame to the paintable-set site; Android pairs receipt
stamps with MediaCodec output buffers via the codec's pts round-trip (JNI
stats array 14→16 doubles, indexes 0-13 unchanged). fps now uniformly counts
received AUs; lost/(received+lost) per window, hidden at zero.
docs-site gains "Understanding the Stats Overlay": what each line means, why
the equation only approximately sums (percentiles), and a line-by-line
Moonlight/Sunshine matrix — including that Moonlight has no end-to-end
number and its "network latency" is an ENet control RTT, so punktfunk's
headline must not be compared against any single Moonlight line.
Verified here: linux client + probe + core check/clippy/fmt green, android
native cargo-ndk arm64 check green. Pending: Windows CI + on-glass, swift
test on the mac, on-device Android.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Found by a live boot-order test (host started before the user session's
PipeWire): PwMicSource::open returned Ok before the daemon connection was
attempted, so a PipeWire that wasn't running surfaced as an instantly-dead
instance instead of an open failure — and the pump churned
open→die→reopen at heartbeat rate (1 Hz "virtual mic ready" log spam)
instead of backing off.
- PwMicSource::open now has a bring-up handshake (mirrors the Windows
backend): ready only after connect + stream connect succeed, so a
down daemon is an open ERROR and the pump's backoff engages.
- The pump triages deaths: an instance that lived >= 5 s (a one-off
daemon restart) reopens immediately with the backoff reset; one that
died right after opening counts as a failed open and backs off
(2 s → 60 s cap). New pump test rapid_death_backs_off.
Re-validated live: host started with PipeWire stopped → throttled
"unavailable" warns, zero churn; daemon started → mic node up on the
next retry; exactly one pump + one loop thread (no leak).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Mic passthrough silently died on real hosts. Root causes, all fixed:
- No liveness anywhere: a PipeWire restart (Linux) or any WASAPI device
error (Windows) killed the backend worker; push() fed the dead queue
for the rest of the host's life. VirtualMic now has a liveness
contract (push -> bool, alive(), discard()) and the new shared
audio::MicPump reopens with backoff, probing on an idle heartbeat so
the mic heals BETWEEN sessions too. Validated live: systemctl restart
pipewire -> node back in ~0.5 s, tone flows through the reopened
backend.
- Lazy creation: the mic device didn't exist until the first 0xCB
frame, but games bind their capture device at launch and never
re-follow. The pump opens eagerly at host start (node exists with
zero clients, elected default source).
- Windows headless dead-end: with VB-CABLE as the ONLY render endpoint
(exactly what the installer ships), the anti-echo guard rejected the
cable as the default render endpoint -> mic permanently dead. The new
wiring_plan (pure, unit-tested on every platform) assigns the mic its
endpoint FIRST (cable reserved for the mic), points the loopback at a
DIFFERENT endpoint, and the capture side now yields (explicit
endpoint or honest error) instead of the mic dying. Plan recomputed
per (re)open — endpoints churn at boot/logon/driver installs.
- Stale bursts: buffered audio from a previous session played into a
newly-attached recorder (observed live). Timestamped chunks + a
consumer-gap check in the process callback age everything past 1 s.
The Linux node mechanism stays the stream-based Audio/Source with
RT_PROCESS + priority.session: the canonical null-audio-sink adapter
recipe was tested on this box (PipeWire 1.6.2) and never gets a clock
(QUANT 0 -> pure silence), and WirePlumber reroutes a feeder targeting
it to the default sink (echo). Decision documented in the module docs.
Live-validated on this box (synthetic host + probe --mic-test,
pw-record): eager node, both attach orderings, PipeWire-restart
self-heal, post-session silence. Windows side compile/CI + on-glass
validation pending.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
enricobuehler