FFmpeg's Vulkan Video decoder now runs on the PRESENTER's own VkDevice —
the decoded VkImage feeds the existing CICP CSC pass directly: zero
copy, no interop, and NVIDIA gets hardware decode for the first time
(its VAAPI is unusable by design). One decode architecture for every
vendor going forward; VAAPI-dmabuf and software remain the fallbacks
(auto: vulkan → vaapi → software; PUNKTFUNK_DECODER=vulkan pins it).
Presenter: instance 1.3; probes VK_KHR_video_queue/decode_queue + codec
extensions, a VIDEO_DECODE queue family (+ its codec caps via
QueueFamilyVideoPropertiesKHR), and the samplerYcbcrConversion/
timelineSemaphore/synchronization2 features — all enabled at device
creation when present, exported as a VulkanDecodeDevice handle bundle.
Decoder: AVVulkanDeviceContext built over those handles via pf-ffvk
(features chain, extension lists, deprecated queue indices + the qf[]
map); get_format supplies OUR frames context with MUTABLE_FORMAT so the
presenter's per-plane views are legal; output is DecodedImage::VkFrame
carrying AVVkFrame/frames-ctx pointers plus the lock fns.
Present: R8+R8G8 plane views over the multiplanar image, the live sync
state read under the AVVulkanFramesContext lock, a timeline-semaphore
wait(sem_value)/signal(sem_value+1) folded into the submit, layout/
queue-family/sem_value written back per FFmpeg's contract, and the frame
guard parked in the retire queue until the fence. CSC pass + video
framebuffer are now unconditional (NVIDIA has no dmabuf-import path).
Verified on the RTX 5070 Ti: device creates with decode_qf=3,
caps=DECODE_H264|H265|AV1|VP9; swapchain unaffected. Live stream
validation next.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
ffmpeg-sys-next binds a curated header list that omits every
hwcontext_*.h, so AVVulkanDeviceContext/AVVulkanFramesContext/AVVkFrame
— the surface that lets FFmpeg's Vulkan Video decoder run on the
presenter's own VkDevice and hand the decoded VkImages back — had no
Rust bindings at all. pf-ffvk generates them at build time from the
SYSTEM headers, which is the only ABI-safe option: the struct layout
depends on FF_API_* deprecation gates resolved in libavutil/version.h
(confirmed: FFmpeg 8.1 still carries the FIXED_QUEUES fields). Ships
ash<->vulkan.h handle conversions; opaque AVHW*/AVFrame types keep
ffmpeg-sys-next the owner of the core surface (pointer casts across).
Needs vulkan-headers (Arch) / libvulkan-dev (CI, added);
PF_FFVK_VULKAN_INCLUDE overrides for cross builds. Verified: bindgen
layout tests pass, av_vk_frame_alloc links and zero-initializes.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Under infinite GOP the pump only re-requested an IDR when the
reassembler's drop count climbed. A lost initial IDR (or a mid-GOP join)
delivers complete-but-undecodable delta frames instead — the reassembler
never drops, so recovery never fired and the stream froze on the last
good frame while libavcodec flooded stderr with missing-reference
errors. Reproduced at 4K@144 (large IDRs, higher loss); lower modes hid
it. Now a 3-frame no-output streak (~50 ms) forces a fresh IDR,
throttled and re-armed across the request→IDR round trip. Verified on
glass: 4K@144 recovers and holds. Also quiets libavcodec's raw stderr
(it bypassed tracing) to fatal-only, PUNKTFUNK_FFMPEG_LOG restores it.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The GTK client becomes a relm4 component tree: AppModel owns the window,
trust gate (rules 1-3), and the spawned session child's lifecycle as
typed messages; the hosts page is a child component with a
FactoryVecDeque of host cards — the HostsCallbacks Rc<dyn Fn> bag, the
busy Cell, and the Rc<RefCell<HostsUi>> cross-page pokes are gone.
Trust/settings/library dialogs stay plain GTK, invoked from update with
a ComponentSender.
Deleted with the in-process presenter: ui_stream.rs, video_gl.rs,
ui_gamepad_library.rs, launch.rs, the khronos-egl dep, and the
PUNKTFUNK_LEGACY_PRESENTER hatch. Every stream (and the console library)
now runs in punktfunk-session; the shell spawns it for card connects and
exec's it for --connect/--browse so the Decky wrapper keeps working. The
request-access flow gains --connect-timeout + a CancelHandle that kills
the child. Screenshot scenes re-pointed onto the components (verified:
hosts + library self-capture; the dialog scenes present correctly and
capture under a GL renderer); the gamepad-library scene is gone with the
page.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The run loop grows a browse mode: the console library idles between
streams in ONE window (no gamescope handoff), overlay actions launch
sessions via run_browse's callback, session end returns to the library.
The Overlay contract gains menu routing (MenuEvent → haptic pulse),
action draining, and session-phase edges.
pf-console-ui ports the GTK launcher wholesale: the coverflow's springs,
cursor arithmetic and recede/tilt constants move verbatim with their
tests (plus new projection tests — focused-card centering, the inner-
edge-recedes corridor); paint order is draw order (the gtk::Fixed
restack hack is gone); the aurora renders as an SkSL runtime shader at
full rate on every box (the 30 Hz CPU-upscale path and its frozen-on-
Deck fallback are deleted — the generated SkSL is compile-tested);
titles/scenes shape through textlayout (CJK-safe). Poster art streams
in as encoded bytes through the shared model and decodes renderer-side.
The session binary wires --browse host[:port] [--mgmt PORT]: KnownHosts
lookup (unpaired renders the pair-first scene), library + art fetch on
threads, PUNKTFUNK_FAKE_LIBRARY dev hook, and a session_params helper
shared with --connect. The minimal build refuses --browse cleanly.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The §6.1 presenter↔console-UI contract lands: pf-presenter exposes its
device (SharedDevice) and composites at most one premultiplied-alpha
quad per frame (new overlay.frag + LOAD render pass over the swapchain;
zero cost while the overlay returns None). pf-console-ui implements it
with skia-safe on the shared VkDevice: DirectContext via the ash
dispatch chain, a ring of two offscreen render targets (one-frame-in-
flight safe), damage-driven redraws — the OSD re-renders at 1 Hz, the
hint on capture toggles, nothing per-frame. Skia never touches the
swapchain. The session binary carries it behind the default ui feature:
4.9 MB stripped without, 10 MB with (measured).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Faults every dmabuf import so the failure-streak → force_software →
software-decode recovery is exercisable on healthy hardware — the plan's
§8 phase-2 acceptance requires demoting via a deliberately faulted
import rather than waiting for a broken driver.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
start_session_with routes desktop windowed connects to the spawned
Vulkan session binary (--connect --fp, --launch, --fullscreen from the
stream setting); spawn.rs bridges its stdout contract into the shell —
spinner until {"ready":true}, banner from the error/ended JSON line,
exit 3 + trust_rejected routed to the re-pair PIN ceremony, TOFU pins
the advert fingerprint only once the child proves it on a real connect.
The in-process GTK presenter stays for PUNKTFUNK_LEGACY_PRESENTER=1,
Gaming-Mode/--fullscreen and --browse launches (no second toplevel under
gamescope until phase 4 moves the console UI), the request-access flow,
and any spawn failure (silent fallback).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The decoder's NV12 dmabuf imports per-plane (R8 + GR88, explicit DRM
format modifier, dedicated dup'd-fd import, FOREIGN→graphics acquire)
and a fullscreen-triangle render pass converts it into the presenter's
video image with the CICP-driven coefficients ported from video_gl.rs
(same tests, plus a rows-vs-matrix agreement check). SPIR-V is committed
(shaders/build.sh regenerates) so builds and CI need no toolchain. The
import extension set is probed at device creation; unsupported boxes and
3-failure streaks demote the decoder to software via the existing
force_software contract. The session binary now honors the Settings
decoder preference instead of forcing software.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
punktfunk-session streams one --connect session in an SDL3 window: ash
swapchain with a transfer-only letterboxed blit of the software-decode
path (no graphics pipeline until the phase-2 dmabuf/CSC pass), the
ui_stream input-capture state machine on SDL events (scancode→VK table
cross-checked against the evdev one), gamepads via a new caller-pumped
GamepadService mode (SDL video owns the main thread here), and the
shell↔session stdout contract: {"ready":true}, per-window stats:
lines, JSON error + exit codes 0/2/3/4. Strict trust — no pin, no
connect. Design: punktfunk-planning linux-client-rearchitecture.md.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Session pump, FFmpeg decode, PipeWire audio, SDL3 gamepads, keymap, trust
store, mDNS discovery, library client and Wake-on-LAN move verbatim from
clients/linux into crates/pf-client-core, shared with the upcoming Vulkan
session binary (punktfunk-planning: linux-client-rearchitecture.md).
The GTK client re-exports them at the crate root so every existing
crate::-path keeps resolving; its manifest drops the moved-only deps.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Windows was missing two of the four stream shortcuts the GTK client has:
Ctrl+Alt+Shift+S (toggle the stats overlay live) and F11 (toggle fullscreen).
Add both to the low-level keyboard hook — S flips a HUD_VISIBLE atomic seeded
from Settings::show_hud at install (Settings is the default, the key overrides
it for the session, matching GTK), F11 drives a borderless-fullscreen toggle on
the window HWND and re-locks the pointer geometry for the new client rect. Both
are consumed locally, never sent on the wire.
Surface the full key set in two places, on both clients:
- in the UI: a read-only "In-stream keyboard shortcuts" reference card in the
Windows Settings > Input section (the counterpart of the GTK Shortcuts
window), plus the expanded HUD hint; the Linux keyboard hint gains F11.
- at stream start: a bottom-centre banner listing the shortcuts for the first
few seconds of every session, independent of the HUD setting. Linux gets the
matching start-flash of its capture hint (capture engages on map and hid it,
so the keys were never shown until the first release).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The APK was only reachable via the generic registry (or attached to a Gitea
Release on a vX.Y.Z tag) — a main-push canary or a PR/dispatch run surfaced no
downloadable APK on the run page itself. Add an upload-artifact step (v3, per
Gitea's GHES-identifying artifact backend, like apple.yml) that grabs whichever
APKs were built — the signed universal release APK on a main/tag push, else the
debug APK — so any run is a one-click sideload download.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The async decode loop (burst-feed + present-newest-per-vsync, the Apple client's
discipline) plus the per-SoC tuning was gated behind an experimental,
default-off toggle after the 5dc24a0 overhaul regressed on some phones. That
regression was the receive-side latency ratchet the async loop fed — a standing
queue that only grew — now fixed in the shared core (FrameChannel jumps to live
instead of accumulating it), so the fast pipeline is the default again.
Default the master toggle on via a fresh pref key (low_latency_mode_v2),
mirroring the migration da376b31 used to flip it off: a new key re-defaults every
install — including ones persisted off under the old key — to on, so the
promotion reaches users who had saved settings, not just fresh installs. Both
stale keys are abandoned unread. Toggle-off still restores the original
synchronous decode pipeline byte-for-byte as a per-device escape hatch.
Drop "(experimental)" from the settings labels and fix the now-stale default-off
wording in the native + Kotlin docs. No decode-path routing change — run_async is
reached simply because the toggle now defaults on.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Clients request a keyframe on every FEC-unrecoverable frame and keep asking
until the IDR actually arrives and decodes — a full round-trip on a link that is
already behind. The host answered every request with a full IDR and only
rate-limited under intra-refresh (opt-in, off by default), so on the default
path a Wi-Fi loss burst produced a 20-40x bitrate spike storm that deepened the
very loss it was recovering from — the second half of the periodic double-jolt.
Coalesce a request storm into at most one forced IDR per cooldown ALWAYS: serve
the first immediately (a genuinely wedged decoder still recovers at once), then
suppress for the window (2 s under intra-refresh's healing wave, 750 ms for a
full-IDR recovery — long enough to swallow one recovery event's round-trip echo,
short enough to re-issue a lost IDR promptly). Seed the cooldown at session open
and stamp it on both rebuild paths so the cold-open / post-rebuild storm
coalesces into the IDR the fresh encoder already emits.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The embedder-facing frame queue was a 16-deep sync_channel whose try_send
dropped the NEWEST access unit on overflow — backwards for a live stream (keeps
stale, discards fresh), a ~266 ms floor that could not self-drain (producer and
consumer both run at frame rate, so any depth a burst injects is conserved
forever — the latency ratchet), and a silent reference-chain break the loss
counters never saw. The clock-based flush meant to catch it was gated on the
skew handshake and never even drained that queue.
Replace it with a purpose-built FrameChannel (VecDeque + Condvar) exposing
depth() and clear(). Pre-decode AUs are reference-chained under the host's
infinite GOP, so they are never dropped mid-stream; instead, when the embedder
falls persistently behind, the pump JUMPS TO LIVE — flush_backlog() + clear the
queued AUs + request a keyframe — so decode re-anchors cleanly at an IDR.
Two cooldown-gated detectors, both suspended during a speed test:
- clock-based (existing): > FLUSH_LATENCY behind the skew-corrected clock for
FLUSH_AFTER_FRAMES straight; also catches kernel/reassembler backlog.
- clock-free (new): the hand-off queue sat >= QUEUE_HIGH without draining to
QUEUE_LOW for STANDING_FRAMES straight. Works on same-clock / no-handshake
sessions where the clock path is disarmed — the direct "the embedder can't
keep up" signal. A transient Wi-Fi clump drains in a few frames and never
trips it.
Bounded (90-frame hard cap, drop-oldest memory backstop) and diagnosable (each
jump logs queue_depth / flushed_datagrams / dropped_frames). next_frame's
external Timeout/Closed contract is unchanged, so every native client inherits
the fix. Adds 5 FrameChannel unit tests.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The mgmt REST API has bound 0.0.0.0:47990 by default since ae51276 so paired
clients can browse the game library over mTLS, but every packaged firewall
opener still excluded 47990 and the docs still claimed it was loopback-only.
On any host with an active firewall (ufw/firewalld) the LAN game-library
feature was silently broken.
Add 47990/tcp to the native firewall profiles (punktfunk.ufw [punktfunk-native]
+ punktfunk-native.xml) and correct the stale "loopback-only by default" text
across the debian/arch/bazzite READMEs and the docs site (incl. the factually
wrong --mgmt-bind default in host-cli.md, 127.0.0.1 -> 0.0.0.0). Opening the
port adds no admin exposure: off-loopback mgmt::require_auth serves only the
read-only status/library allowlist to a paired client cert; the bearer-token
admin surface stays loopback-only regardless of the bind.
Windows was already sound (shared parse_serve binds 0.0.0.0; service.rs already
firewall-opens 47990) — add a clarifying comment so the rule isn't mistaken for
accidental over-exposure.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The full git.unom.io package URL is painful to read/type on a Steam
Deck's on-screen keyboard; unom.io/pf-decky now redirects to it
(added in unom/infra's Caddyfile). Canary/pinned links stay long-form
since only /latest gets a short link.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Force patched undici/dompurify/postcss/esbuild/js-yaml pulled in
transitively via @unom/ui's payload/vite/storybook chains, and bump
vite 7.3.5 -> 7.3.6 so esbuild resolves consistently to 0.28.1
instead of splitting across 0.27.7/0.28.0. Closes all 26 bun audit
findings (3 high, 17 moderate, 6 low).
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Release 0.8.2: native AMF encoder reconnect reliability on Windows AMD — a
second connection no longer comes up black (the encoder now Flushes before
Terminate so a reconnect's overlapping teardown can't strand AMD's VCN encode
session, self-heals with a full context rebuild when a reconnect still wedges,
and logs a per-context bring-up number + first-AU line so a silent wedge is
visible); a native data-plane hardening pass that keeps the stream alive across
real Wi-Fi links; Android streaming wake/Wi-Fi locks that actually engage plus a
console-UI polish pass (per-controller glyphs, scrollable dialogs, animated
forms); and gamescope-takeover survival on Bazzite's SDDM session supervisor.
Also a Linux-client crash fix — the FlowBox activation cycle that stack-
overflowed on every host-card click.
The [workspace.package] version (inherited by every crate via version.workspace)
is the release being cut; refresh the 9 workspace entries in Cargo.lock to match
(CI builds --locked). Canary derives from the tag (scripts/ci/pf-version.sh), so
cutting v0.8.2 auto-advances canary to 0.9.0.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A second connection to a Windows AMD host came up black with nothing in the
logs. The native AMF encoder's teardown never Flushed before Terminate, so a
reconnect whose teardown overlapped the new session (a client may not signal an
explicit exit, so session 1 tears down late — on the reconnect preempt grace or
the QUIC idle timeout) left AMD's limited VCN encode-session slot occupied. The
new session's Init then opened onto a wedged session that returns AMF_OK but
never emits an AU. NVENC has no equivalent per-session cap, so NVIDIA never
showed it. Recovery couldn't help either: the stall watchdog re-Init'd the SAME
context, which can't clear a context/VCN-level fault, so it looped a dead
context until MAX_ENCODER_RESETS ended the session.
Reliability:
- Component::drop now Flushes before Terminate (mirrors reset() and the design
doc), releasing the VCN session cleanly so the next session's Init gets a free
slot.
- reset() escalates to a FULL context teardown once an in-place re-Init has run
without producing an AU (resets_without_output >= 2), so a wedged reconnect
self-heals via a fresh CreateContext+InitDX11 within the reset budget instead
of re-initing a dead context in a loop.
Logging (the failure was silent):
- Per-context bring-up sequence number (context #N) — distinguishes a first
connection from a reconnect's fresh context.
- A one-shot "AMF produced its first AU on this context" line; its absence after
a context #N bring-up is the smoking gun for a silent VCN wedge.
- Terminate result logged on drop for both the component and the context.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Diagnosed live on the .181 Bazzite F44 box (couldn't connect at all; field
reports of streams dying after 30 s-5 min):
Bazzite autologs into game mode via SDDM with Relogin=true, so the moment the
managed takeover stops gamescope-session-plus@<client>, SDDM logs back in and
restarts it within the same second. The resurrected autologin session then
fights our transient session-plus over the Steam single instance and the GPU
for the whole stream: its wrapper relaunches gamescope every ~7 s (each one
missing the wrapper's hard 5 s readiness window on a slow NVIDIA init), the
churn SIGSEGVs gamescopes, and eventually the streaming gamescope dies with it.
Meanwhile a client that gave up left the pipeline-rebuild retry loop SIGKILLing
and relaunching the box's Steam session for up to ~6 more minutes.
- stop_autologin_sessions: runtime-mask each autologin unit before the SIGKILL
stop, so no supervisor can restart it underneath the stream; match every
loaded instance (the unit flaps through activating/failed mid-churn). Every
restore path unmasks unconditionally (including the desktop-active early
return), and --runtime keeps the mask in tmpfs so a reboot clears it.
- launch_session: supervise the transient unit while polling for the node —
the session-plus wrapper kill -9s a gamescope that missed its 5 s readiness
handshake and exits 1, so relaunch it (after a short driver-settle cooldown)
instead of waiting the rest of the 45 s on a corpse.
- build_pipeline_with_retry: abort between attempts once the session's QUIC
connection is closed — no more minutes of Steam churn for a departed client.
Validated live on .181: cold-boot connect streams 2059 frames/45 s (p50
5.1 ms), zero SDDM resurrections while masked, TV session restored+unmasked on
disconnect, warm same-mode reconnect reuses the session (866 frames/15 s).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Cargo's debug profile is not "slower" for this library — it is unusable, and
it invalidated every on-device performance test to date: RustCrypto's AES-GCM
compiles to generic-array iterator closures with per-byte precondition checks
instead of ARMv8 hardware AES. Profiled live on a phone (simpleperf, 62k
samples): ~800 µs of user CPU per 1.4 KB packet — the receive pump pinned
above a full core yet only draining ~1,400 pkt/s of a 1,775 pkt/s (20 Mbps)
stream, 2.3 MB standing in the kernel socket buffer, the latency-bound flush
firing every 2 s forever. With release rust in the same debug APK: pump at
~12 % of a core, socket queue zero, no flushes, 2800x1260@120 streaming clean.
preDebugBuild now depends on cargoNdkRelease; `-PrustDebug` opts back into a
debug-profile native build for sessions that actually step through Rust.
Kotlin debuggability is unchanged.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
WifiLock.acquire() enforces the WAKE_LOCK permission, which the manifest never
declared — every acquisition since the first Wi-Fi lock shipped threw
SecurityException, silently swallowed by a bare runCatching. The phone's own
accounting proved it (dumpsys wifi: high_perf/low_latency active_time_ms = 0
across weeks of streams): every on-device session ran with Wi-Fi power save
fully active, whatever the code intended. Verified live after the fix: both
locks registered in WifiLockManager, mPowerSaveDisableRequests=2, ping RTT to
the streaming phone 3.8 ms avg. A failed acquire now logs loudly — this class
of failure must never be invisible again.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The baseline stream held only WIFI_MODE_FULL_HIGH_PERF, which is deprecated
AND non-functional on recent Android — so with the low-latency toggle off (the
default) Wi-Fi power save stayed fully active: downlink delivery clumped at
beacon intervals (a few hundred ms of latency mush, sawtoothing bitrate) and
the AP's power-save buffer periodically overflowed, killing whole frames every
few seconds (the host log's alternating loss_ppm=0/50000). Now every stream
holds FULL_LOW_LATENCY (API 29+, the only effective power-save disable;
foreground + screen-on, which a stream always is) AND FULL_HIGH_PERF (covers
older releases) — the same pair Moonlight holds. The experimental toggle no
longer selects the lock mode.
Also: declare tracing's "log" feature explicitly in the native crate (core
transport warnings → logcat must not hinge on quinn's default features), and
align the low-latency toggle's copy with its actual scope.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- Hint-bar glyphs now wear the driving controller's family (kit
Gamepad.styleFor by USB vendor id → MainActivity.lastPadStyle, kept live by
real input like lastPadIsGamepad): PlayStation pads get Canvas-drawn
cross/circle/square/triangle shapes in the classic colours, Nintendo pads
monochrome lettering, Xbox/Valve/unknown the coloured letter discs. Hint
chars stay semantic (KEYCODE_BUTTON names); only the rendering changes.
- The Options legend renders the pad's real Select-family button
(SelectButtonGlyph): Xbox View windows, PlayStation Create capsule,
Nintendo minus — instead of a bare capsule outline.
- GamepadDialog: body + action stack scroll together (title pinned) with
BringIntoViewRequester keeping the focused button visible — a 5-action host
options dialog compressed/clipped its last button in short landscape
windows because the pinned stack could not scroll.
- Console form polish: shared animateConsoleFocus (bg/border cross-fade +
spring scale) across settings rows / add-host fields / action rows;
ConsoleSwitch (spring knob, tinting track) replaces On/Off text on toggle
rows; choice values slide in the direction they were stepped
(AnimatedContent + SizeTransform) with chevrons that fade in place; the
focused row's detail unfolds via AnimatedVisibility; dialog buttons and
keyboard keycaps cross-fade (keycaps at 90 ms for hold-to-repeat).
- Console settings gain the "Low-latency mode" (Video) and "Auto-wake on
connect" (Interface) rows, round-tripping with the touch settings.
- Screenshot scene: StatsOverlay call updated to the 18-double layout + the
new decoderLabel parameter (fixes the android-screenshots CI compile).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Root-caused live on a phone at 100 Mbps (stream stuck seconds behind, then
oscillating): a stack of transport defects, each amplifying the next.
- MTU-safe shards: shard_payload 1452 overshot the IPv4/1500 budget (the old
math forgot the 40 B header + 24 B crypto ride inside the UDP payload and
counted IP+UDP as 8 B) — the kernel silently split EVERY video datagram into
two IP fragments, doubling per-datagram loss on Wi-Fi. New
config::mtu1500_shard_payload() = 1408 (1472 sealed = the exact ceiling),
negotiated in the Welcome, pinned by a unit test.
- Android batched I/O: recv/send batching was cfg(linux); Android is
target_os="android" and silently fell back to a syscall per datagram. The
libc crate binds neither recvmmsg/sendmmsg nor mmsghdr for Android, so a
local bionic extern binding provides them (API 21+, floor is 28); cbindgen
excludes them from the C header. The pump/runtime threads also get the
Apple-QoS analogue on Android: nice −8 (below the decode thread's −10).
- Latency-bounded receive: packets are consumed strictly in order at exactly
the arrival rate, so a standing queue (Wi-Fi stall, power-save clumping)
NEVER drains — observed as a stream permanently 6-7 s behind with both 32 MB
socket buffers full. The pump now flushes the entire backlog
(Session::flush_backlog: discard ring + kernel queue at memcpy speed, reset
the reassembler) and requests a keyframe when frames keep completing > 400 ms
behind the skew-corrected capture clock (30 consecutive, 2 s cooldown,
logged).
- Time-based loss window: the reassembler declared an incomplete frame lost a
fixed 4 INDICES behind the newest — 33 ms at 120 fps, inside normal Wi-Fi
retry/reorder timescales, so merely-late frames were pruned every few
seconds, each costing a recovery-IDR burst + an inflated loss report.
Now 120 ms of capture time (LOSS_WINDOW_NS), same fuse at every refresh
rate, with a 64-index hard cap bounding memory against hostile pts.
- Adaptive-FEC hysteresis: the controller was memoryless — one clean 750 ms
report dropped FEC from 8 % straight back to the 1 % floor, so periodic burst
loss (Wi-Fi scan / BT coexistence beats) always hit an unprotected stream and
ping-ponged 1↔8 % with a frozen frame per cycle (observed in the host log as
alternating loss_ppm=0/50000). Attack stays instant; decay is now one point
per clean report.
Verified: full core suite (incl. new flush + time-window tests) on macOS +
Linux, host release build, arm64 cargo-ndk build, and a 30 s wired probe run
at 2800x1260@120 — 3559/3559 frames, zero loss, capture→received p50 5.3 ms
(host 5.1 + network 0.3).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The Ctrl+Alt+Shift+Q/D handlers had no tracing, so a report of "the
disconnect shortcut doesn't work" was unverifiable from logs alone —
live tracing (added temporarily, then trimmed to these two lines)
showed the chord, `disconnect_quit()`, and the session teardown all
firing correctly and instantly every time; the confusion traced back
to the (now-fixed) FlowBox click crash having kept everyone from ever
reaching a live session to test the shortcut with in the first place.
Keep the two low-noise, deliberate-action log lines for the next time
this comes up; drop the per-keystroke debug trace used to diagnose it,
which would otherwise fire on every key during a stream.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
`child-activated` (fired by a pointer click) was bridged to `child.activate()`
so each card's own connect handler (wired on the child's `activate` signal)
would run. But `child.activate()` runs `GtkFlowBoxChild`'s default handler,
which re-emits `child-activated` on the FlowBox — bouncing straight back into
the same closure. Unguarded, that ping-pong recursed forever, overflowing the
stack on every single host-card click or Enter-key activation (confirmed live
via coredump/gdb: 43k+ stack frames of gobject signal emission, and the
`fatal runtime error: stack overflow, aborting` in the crash log).
A re-entrancy flag breaks the cycle after the one real activation. Added a
regression test that wires the identical FlowBox/FlowBoxChild signal cycle
against a real display and asserts it returns instead of recursing — it
reproduces the exact stack overflow against the old code.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The 5dc24a0 low-latency overhaul regressed badly on some phones. Every piece
of it — decoder ranking, per-SoC vendor keys, the async decode loop, pipeline
thread boosts, the ADPF max-performance bias, game-tagged AAudio, DSCP marking,
the Wi-Fi low-latency lock, HDMI ALLM and the forced TV mode switch — now rides
the "Low-latency mode (experimental)" toggle, default OFF. Off restores the
pre-overhaul pipeline byte-for-byte: the sync poll loop, the platform-default
decoder, and the original format keys (standard low-latency + blind Qualcomm
twin + priority=0 + operating-rate=MAX together).
- New pref key (low_latency_mode_experimental): the old key shipped default-ON,
so any install that ever saved settings persisted true — flipping the default
under the old key would leave exactly the regressed devices stuck on.
- DSCP is applied at socket creation, so the toggle reaches the transport via
NativeBridge.nativeSetLowLatencyMode → transport::set_dscp_default, called in
the connect choke point before nativeConnect; the core DSCP default reverts
to off everywhere.
- nativeStartAudio(handle, lowLatencyMode) gates AAudio usage=Game.
- VideoDecoders.pickDecoder now skips `.secure` decoder twins and decoders that
require FEATURE_SecurePlayback: they need a secure surface, and a secure twin
could out-score its plain sibling (only it advertising FEATURE_LowLatency),
which black-screens a clear stream.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
An mDNS discovery miss was forcing connects through the Wake-on-LAN wait
even for a host that's already up; add a Settings toggle ("Auto-wake on
connect") that skips the mDNS-liveness gate and dials straight through
when off.
Also default the console UI's button glyphs by form factor instead of
always starting in TV-remote style: a phone/tablet only ever enters the
console UI via a real controller, so it should show gamepad glyphs from
the first frame, not a remote's select/back glyphs. TV keeps the remote
default.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The 0.8.1 version bump (6c1e6ad) bumped Cargo.toml but not the generated
OpenAPI doc, leaving info.version stale at 0.8.0. Regenerated — version
string only, no API surface change.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The native AMF encoder (6f47aba) landed unformatted, failing CI's Format
step (and short-circuiting Clippy/Build/Test). Reformatted amf.rs with the
pinned rustfmt 1.96.0 — no functional change.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
`build_ui` (the GTK `activate` handler) started a fresh SDL3 gamepad
worker thread on every reactivation of the already-running singleton
(another --connect, the launcher clicked twice, ...). sdl3 only lets
the first thread ever to call sdl3::init() hold "main thread" status,
so every later activation's worker thread failed permanently with
"Cannot initialize `Sdl` from a thread other than the main thread",
silently disabling controller support for the rest of the process.
Start the GamepadService once in run() and clone it into build_ui
instead of starting a new one per activation.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Modifier keys arrive only as flagsChanged, and the direction was recovered
by diffing the device-dependent L/R bits (NX_DEVICE*KEYMASK) alone. Those
bits are undocumented and some keyboards omit them (only the class bit,
e.g. NX_CONTROLMASK, is set), so the diff saw no transition and the key
never reached the host — no Ctrl shortcuts. SDL/Moonlight key off the
event's keyCode for exactly this reason; do the same: keyCode names the
changed key, the class bit says up, the device bits (when present) pick
the side, and a tracked-held-state flip covers keyboards without them.
PUNKTFUNK_INPUT_DEBUG=1 now also logs every flagsChanged (keyCode + raw
flags) so a field report is diagnosable from client logs.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
apply_session_env unconditionally forced PUNKTFUNK_FORCE_SHM=1 for every
GNOME/Mutter session, added 2026-06-14 after a same-day stale-frame bug hunt
found Mutter has no implicit dmabuf fence on NVIDIA. That override silently
contradicted the documented "zero-copy is on by default for all Linux GPU
backends" behavior and left Mutter+NVIDIA hosts on the slower CPU/SHM path
unconditionally, with no way to opt back in.
Live retesting (192.168.1.21, RTX 5070 Ti, real client with cursor
movement/window drag/typing — the historical trigger) shows no visible
staleness with the override removed. Drop the automatic force; PUNKTFUNK_FORCE_SHM
stays as a manual escape hatch for anyone who does hit flashing/stale frames
on a Mutter+NVIDIA host.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Direct-SDK AMF encoder (encode/windows/amf.rs), the AMD analogue of the
direct-NVENC path, replacing the libavcodec *_amf dispatch. C-vtable FFI
pinned to AMF headers v1.4.36, runtime-loaded from the driver's amfrt64.dll
(no build feature, no new dependency) exactly as NVENC loads its DLL.
- AVC/HEVC (SDR NV12 + 10-bit HDR P010) and AV1 (RDNA3+, probed); a bounded
poll retires the libavcodec ~2-frame output hold; native in-place reset().
- Intra-refresh wave (PUNKTFUNK_INTRA_REFRESH), in-band HDR mastering/CLL
metadata (*InHDRMetadata -> HEVC SEI / AV1 OBU), and a native codec probe
feeding the GameStream advertisement (windows_backend_is_ffmpeg ->
windows_backend_is_probed).
- AMD dispatch / advertisement / 4:4:4 are native-only; the libavcodec AMF
fallback and the PUNKTFUNK_AMF_FFMPEG hatch are removed. FFmpeg serves QSV
only (its AMF path retained solely as the latency A/B comparator).
- Overload back-pressure: submit bounds in-flight surfaces below the input
ring, draining finished AUs (buffered for poll, FIFO-preserved) to free a
slot and retry on AMF_INPUT_FULL instead of tearing the encoder down and
forcing an IDR; this also closes a latent ring-overwrite corruption seen
under load on-glass.
Validated on the lab Ryzen iGPU (AMF runtime 1.4.37): HEVC/AVC across a
native reset, HEVC Main10 mastering+CLL SEIs byte-verified, intra-refresh
accepted, a backpressure burst FIFO-clean, and end-to-end via the macOS
client. Measured §5.2 latency A/B: native encode_us p50 ~5 ms (0.31 frame
periods) vs libavcodec ~17 ms (1.01). 4:4:4 stays unsupported (VCN hardware
limit). Live-gated tests skip cleanly on non-AMD boxes.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Like the Linux client, the Windows client had WOL send + MAC storage + a Wake
action + fire-and-forget auto-wake, but no poll-until-up + IP re-key. Add the
polished flow (mirrors Apple HostWaker + the request_access screen pattern):
- connect::wake_and_connect — send the magic packet, show a cancelable
Screen::Waking busy page, poll discovery::browse() until the host reappears
(re-sending every 6 s, 90 s budget), then dial; re-key the saved host
(KnownHosts::upsert) if it woke on a new IP.
- Screen::Waking + waking_page, routed in app/mod.rs (mirrors RequestAccess).
- the saved-host tile routes an offline-with-MAC tap to wake_and_connect;
MENU_WAKE stays a pure send-only button.
Reviewed against the request_access reference — DiscoveredHost/KnownHost/Target
types, the widgets, .call()/.lock(), and the initiate signature all match;
compile-verified by Windows CI (no local Windows toolchain).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The Linux client already had WOL send + MAC storage + a Wake action + auto-wake-
on-connect, but the auto-wake just fired a packet and did one dial to the stored
address — so a host that woke on a new DHCP lease failed, and there was no
"waiting" feedback. Add the polished flow (mirrors Apple/Android HostWaker):
- ui_trust::wake_and_connect — send the magic packet, poll mDNS until the host
reappears (re-sending every 6 s, 90 s budget) behind a cancelable "Waking…"
dialog, then connect; if it woke on a new IP, re-key the saved host first.
- trust::rekey_addr — no-churn addr/port update keyed by fingerprint.
- the hosts page routes an offline saved-host-with-MAC tap to on_wake_connect
(the new flow) instead of fire-and-forget wake + immediate dial.
Builds + clippy + fmt clean.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The core's deliberate-quit close (NativeClient::disconnect_quit → QUIT_CLOSE_CODE,
host skips the keep-alive linger) was implemented but never called by any client.
Wire it to each client's explicit user-disconnect action — NOT to a network drop /
host-ended / app-background (those keep the linger for a reconnect):
- core: new C-ABI punktfunk_connection_disconnect_quit(c) for the ABI clients
- Linux (direct-core): Ctrl+Alt+Shift+D + the controller escape chord
- Windows (direct-core): Ctrl+Alt+Shift+D
- Apple (C-ABI): PunktfunkConnection.disconnectQuit() + a `deliberate` flag on
SessionModel.disconnect() (sessionEnded passes false → keeps the linger)
- Android (JNI): new nativeDisconnectQuit export, called from the back gesture +
the Select+Start+L1+R1 chord (not the host-gone watchdog)
- probe already did this via --quit (77871d6)
Verified: core + Linux client + Android (cargo-ndk + gradle) build clean;
Windows/Apple compile-checked by CI.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
`cargo fmt --all --check` (a CI gate) failed on main: config.rs (the new
`zerocopy: val(...).map { !matches!(...) }` from 76bc7fe) and punktfunk1.rs
(the `reset_stalled_encoder` conditions from 167d590) were left unwrapped by
the pinned rustfmt. Pure reformat, no semantic change.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The library browser was browse-only — the A button (and a tap) did nothing.
Wire it to connect + boot straight into the selected title: thread a `launch`
id (the store-qualified library id `steam:<appid>` / `custom:<id>`) through
nativeConnect → NativeClient's Hello.launch (was hardcoded None), add a shared
connectToHost() the ConnectScreen and the library launcher both use, and have
LibraryScreen dial the host with launch=game.id on A / tap — with a launching
overlay + an "A Launch" hint. Verified: native compiles (cargo-ndk arm64),
app+kit Kotlin compiles (gradle, 3 ABIs).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
On-glass A/B on the Ryzen 7000 iGPU (1080p120 HDR P010, hevc_amf,
PUNKTFUNK_PERF stage split): the system-memory readback costs the encode
thread 2.7-2.9 ms p50 (6.6 ms p99) per frame in submit; the zero-copy D3D11
pool path does the same work in 0.26 ms p50 (0.5 ms p99) — and on an iGPU the
readback also burns the shared memory bandwidth the game needs. The docs-site
already promised "on by default ... D3D11 on Windows" since the Linux flip
(9814368 was Linux-only); the Windows code now delivers it.
PUNKTFUNK_ZEROCOPY becomes a tri-state override: unset defers to a per-vendor
default in zerocopy_enabled(vendor) — ON for AMF (validated above; open
failures still fall back to system-memory readback), OFF for QSV until it is
validated on Intel glass (the fallback only catches *setup* errors; a QSV
derive that opens but maps wrong would corrupt silently, so probe-never-assume
applies). Explicit values force either way: 0|false|off|no = readback,
anything else = zero-copy, so the old presence-style =1 keeps working.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Field reports: Windows AMD/Intel streams freeze after ~3-5 min regardless of
desktop activity. Root cause: the libavcodec AMF/QSV poll is non-blocking
(EAGAIN -> Ok(None)), and the encode loop's drain treated None as benign
without popping `inflight` — a wedged driver (QueryOutput stops producing)
meant frames kept being submitted, inflight grew unboundedly, no AU ever
reached the send thread, and nothing logged: a silent permanent freeze. The
input-side twin: once libavcodec's one-frame buffer fills, avcodec_send_frame
EAGAINs and the submit `?` killed the whole session.
Add `Encoder::reset()` (in-place encoder rebuild; implemented for AMF/QSV by
dropping the wedged libavcodec encoder so the next submit re-opens it on the
current device, forced IDR) and an encode-stall watchdog in the stream loop:
trip on a poll error, on no AU within max(2 s, 8 frame intervals) while frames
are owed, or on an owed backlog worth more than the window's frames (the
slow-leak latency-runaway form). Recovery is a bounded (5 consecutive, cleared
by any delivered AU) in-place rebuild + forced IDR — a logged ~one-second
hiccup instead of a dead stream; exhaustion or a reset-less backend still
fails the session with a clear error. Submit failures route through the same
bounded recovery. The three existing pipeline-rebuild paths (session switch,
mode switch, capture loss) now also clear the stale in-flight records that
pointed at the dropped encoder.
Backends whose poll blocks (direct NVENC sync, software) can't false-trip:
they never return Ok(None) mid-stream and drain inflight below depth each
tick. Validated: clippy -D warnings (nvenc,amf-qsv), 191 host tests, synthetic
E2E 300/300 frames, and an on-glass AMD iGPU session (1080p120 HDR hevc_amf).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
enricobuehler