v1.0.1

v1.0.1 - 2026-08-12

Output latency cut from roughly 22 ms to 13 ms, a memory-ordering fix in the PCM
handoff, and recovery fixes for the sleep/wake path -- the last of these from a
report of no audio after a PSX game plus standby until the device was replugged.

Latency

• Capture blocks halved to 240 frames (5 ms). Latency is about two block
  periods: one to fill a block, and one more because the consumer has to start a
  whole block behind to survive the discrete arrival cadence. Block size is
  therefore the only lever that moves it. Confirmed on hardware that Sony's
  RAM-output submit accepts the smaller page.
• PCM staging widened from two slots to four, restoring the stall tolerance that
  smaller blocks would otherwise have halved. Slot count costs no latency -- it
  sets how far the feeder may fall behind, not where it starts reading.
• A third transport context, taking the feeder's slack from one packet period to
  two in exchange for 1 ms of delay.
• The feeder thread runs at priority 0x20 instead of 0x40 and is pinned to the
  capture worker's core, so a freshly published block is still in L1 when it is
  read back out rather than costing fifteen cache-line transfers per block.

Correctness

• Fixed memory ordering on the seqlock read side. The second guard read used an
  acquire load, which places its barrier after the read and leaves the data
  copy free to be reordered past the validation -- the classic way a torn read
  passes as valid. It now fences before a relaxed read, matching Linux's
  read_seqretry(). Same two instructions, correct order. The symptom would have
  been a rare, unreproducible click.
• The write side no longer uses a seq_cst store where a relaxed store plus a
  release fence suffice, removing one barrier per published block.

Recovery

• An attach that arrives while a teardown is still running is now remembered and
  retried once teardown completes. USBD offers a device exactly once, so refusing
  that attach previously lost it for good: the device stayed enumerated with no
  session behind it and only a replug recovered it. The window is widest coming
  out of sleep with audio streaming, which is when teardown is slowest.
• Releasing the AVConfig route no longer waits for Sony to acknowledge through
  the hooked device-stop wrapper. Capture is already stopped by that point, so
  Sony may reasonably decide there is nothing to stop and never call it. Observed
  both ways across sessions, which is exactly why the release must not depend on
  it; when it did not come, the wait burned its full one-second timeout.

Instrumentation

• Logging builds print the version and git revision as the first line of each
  session, and report per-session packet, starve, PCM-wait and resync counts plus
  the minimum producer lead, recording where each resync occurred. Release builds
  are byte-identical with or without these.

Measured on 3.65. Route release settles in a repeatable ~401 ms. Suspend retires
the session ahead of the detach that follows, so SET_INTERFACE(alt 0) is still
delivered and an external DAC returns to its internal clock.

USB completions were found to stall for ~250 ms at a time, several times per
session, independent of our thread priorities (verified by A/B). The producer
runs on through such a stall and ends up ~50 blocks ahead; the latest-wins
staging discards the backlog and resumes at current audio. A FIFO would instead
queue it and add that 250 ms to output latency permanently, on every stall.

v1.0

uac-pstv v1.0

USB audio output for the PlayStation TV. Plug in a USB sound card, DAC, or
headphone dongle and system audio comes out of it — games, PSX titles, LiveArea.
No app to launch, nothing to configure.

A different way of capturing audio

v0.1–v0.3 hooked individual SceAudio ports, which meant shipping a mixer and a
44.1→48 kHz resampler, and competing with the audio pipeline for timing — the
1.4% ring-buffer sample loss noted in v0.3.

v1.0 takes ownership of AVConfig's RAM-output route instead, the path Sony
already uses to feed Bluetooth. Audio is captured after Sony's own mixing and
rate conversion, so mixer.c and resampler_coeffs.h are gone — not dropped,
but no longer needed. Whatever the source rate, what arrives is already mixed
and already 48 kHz.

What's new

• 480-frame RAM-output A/B worker mirroring Sony's own DataSend worker, feeding
  a two-context isochronous transport with exactly one request in flight
• Firmware allowlist covering 3.60, 3.61, 3.63, 3.65, 3.67, 3.68, 3.71, 3.73 —
  SceAVConfig is byte-identical across all eight, so one build covers the range
• Session teardown moved off the USBD callback thread, fixing dropped events on
  rapid unplug/replug
• SET_INTERFACE(alt 0) on clean teardown, so external DACs release their clock
  and return to internal
• Suspend retires the session; resume re-asserts USB host mode
• Transport primed with silence so the device locks to a continuous stream
  instead of a ~20 ms gap
• Refuses to load on anything that isn't a PlayStation TV

Install

Copy uac_pstv.skprx to ur0:tai/ and add it under *KERNEL in
ur0:tai/config.txt. It must be listed above StorageMgr — USB drivers get
first refusal on a device in registration order.

*KERNEL
ur0:tai/uac_pstv.skprx
ur0:tai/storagemgr.skprx

Safety

With no USB audio device attached the plugin is inert — it registers a USB driver
and waits, never reading SceAVConfig and installing no hooks. module_start
dereferences no computed address, so it cannot stop the console booting. Nothing
is written to persistent storage; a reboot is a complete reset.

Known limitations

• Output only — microphones and headset input are unsupported
• UAC1 only, fixed 48 kHz stereo 16-bit; UAC2-only devices are declined at probe
• If releasing the audio route ever times out, system audio can stay silent until
  you reboot. Nothing is written anywhere, so a power cycle always restores it.

Testing status

Hardware-verified on 3.65 with a 0bda:4e27 adapter. The other seven
firmwares are verified by binary comparison of the modules, not by running on
each. No extended soak test yet.

v0.3

v0.3 - 2026-08-06

• Refactored uac1, mixer, stream class, and renamed audio to audio_hooks.

• fixed a bug where psx game with sample rate of 44148 have no in game audio.

• Area needs improvement: currently the isochronous usb transfer function
  with callback can not satisfy 48Khz audio, during on device testing we are
  seeing 1.4% of dropped sample from ring buffer.

v0.2

v0.2 - 2026-08-05

• Replaced the separate early-boot and audio modules with one self-contained
  uac_pstv.skprx loaded from taiHEN's *KERNEL section. The
  boot_config.txt helper is no longer required.
• Improved detection of adapters connected during boot when UAC PSTV is listed
  before StorageMgr. Hot-plug support remains available.
• Replaced the basic sample-rate conversion path with a compact 24-tap
  fixed-point polyphase FIR, reducing 44.1 kHz BGM interpolation hiss and
  aliasing while keeping native 48 kHz stereo bit-exact.
• Reduced capture capacity from eight simultaneous SceAudio ports to four,
  cutting static ring-buffer memory from 137,920 to 69,440 bytes while retaining
  one spare slot beyond the three-source pattern observed in testing.
• Simplified the USB streaming and mixer hot paths, retained saturating
  multi-source mixing, and kept inactive processing disabled when no supported
  USB adapter is attached.
• Expanded host-side tests with all-rate cadence, spectral and continuity
  coverage, deterministic drain/race checks, and resampler benchmarks.
• Simplified the build and release package to produce a single module and verify
  that release logging is compiled out.

v0.1

Changelog

v0.1 - 2026-08-05

• Split early USB transport and late SceAudio capture into separate kernel
  modules so a connected UAC device can enumerate during boot.
• Added fixed UAC1 stereo, signed 16-bit, 48 kHz isochronous output.
• Captured game, LiveArea/system-effect, and BGM audio ports.
• Added 44.1-to-48 kHz conversion, saturating multi-source mixing, hot-plug,
  callback draining, and inactive fast paths.
• Removed release-time file logging and USB power-reset fallbacks.
• Added host-side mixer correctness and deterministic drain/race tests.