Commit Graph

6 Commits

Author SHA1 Message Date
enricobuehler ebd9967547 feat(pyrowave): Windows host encoder — separate-plane zero-copy D3D11→Vulkan
Wire PyroWave into the Windows host (design/pyrowave-windows-host-zerocopy.md).
Before this a macOS client + Windows host that both selected PyroWave silently ran
HEVC: the host never advertised CODEC_PYROWAVE and open_video_backend bailed.

Approach (zero-copy, no GPU→CPU→GPU): pyrowave owns its own Vulkan device
(create_device_by_compat, by render-GPU vendor/device-id — NOT LUID, invalid in
Session 0). The capturer runs a BGRA→YUV BT.709-limited CSC (matching rgb2yuv.comp)
into TWO SEPARATE shareable plane textures — full-res R8 Y + half-res R8G8 CbCr —
which the encoder imports into pyrowave's device. Separate single/two-component
textures import reliably on NVIDIA at any size; a single planar NV12 import does NOT
(the vendored interop test: "only very specific resource sizes" — confirmed on-glass:
1024² fine, 720p/1080p/1440p garbage). A shared D3D11 fence, signalled after the CSC,
is imported as a Vulkan timeline semaphore so the wavelet read is ordered after it.

- pf-encode: enc/windows/pyrowave.rs (Encoder impl, two-plane import + Linux-style
  plane views); host_wire_caps advertises CODEC_PYROWAVE on Windows when the backend
  isn't Software; open_video_backend routes a negotiated PyroWave session first;
  pyrowave-sys on the Windows target; interop confirmed at open → clean HEVC fallback.
- pf-encode: shared, unit-tested enc/pyrowave_wire.rs (single source of truth for the
  client-facing AU framing); Linux encoder uses it too.
- pf-capture: dxgi.rs BgraToYuvPlanes CSC; idd_push.rs pyrowave mode — forces the
  virtual display SDR (the VideoProcessor can't ingest the FP16 HDR ring), a
  two-plane shareable out-ring, a shared fence passed every frame (so a rebuilt
  encoder re-imports it). Threaded via OutputFormat::pyrowave.
- pf-frame: D3d11Frame::pyro carries the CbCr plane + fence; OutputFormat::pyrowave.

Verified on .173 (RTX 4090): full-host build + clippy -D warnings (nvenc,amf-qsv) +
fmt --all --check; pyrowave_wire unit tests; pyrowave_win_smoke GPU test round-trips
distinct Y/Cb/Cr (100/180/60) exactly at 1024²/720p/1080p/1440p; Stage-0 interop
validated in the real Session-0 service context on-glass. Deployed to the box.
Owed: final on-glass picture/latency confirmation.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-18 11:39:44 +02:00
enricobuehler df70bf00f9 test(qsv): live on-glass matrix — HEVC, Main10+HDR, AV1 10-bit, LTR-RFI, no-IDR retarget
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-17 19:55:43 +02:00
enricobuehler 9e6ab9b94d fix(qsv): enable D3D11 multithread protection before SetHandle (on-glass -16 on Arc)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-17 19:52:44 +02:00
enricobuehler 0a617e7779 fix(qsv): clippy — per-block SAFETY comments, div_ceil, clamp, vec_box allow
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-17 19:44:16 +02:00
enricobuehler 75d1322b8a fix(qsv): end the inner borrow before the retarget param rebuild
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-17 19:39:08 +02:00
enricobuehler 55e7f3fca9 feat(encode): native QSV backend — libvpl-sys + qsv.rs (Phases 0-3 of design/native-qsv-encoder.md)
Vendored MIT VPL dispatcher (static, trimmed tree, pin 674d015b/v2.17.0) built
via cmake+bindgen behind new feature 'qsv' (pf-encode + punktfunk-host forward).
qsv.rs: dispatcher session on the capture adapter (LUID-matched), SetHandle
D3D11, AsyncDepth=1/GopRefDist=1/VDEnc/CBR + HRD-off low-latency config,
GetSurfaceForEncode + GPU CopySubresourceRegion input (zero-copy, no readback
path), bounded sync-point poll, in-place reset with teardown escalation, no-IDR
bitrate retarget (Reset + StartNewSequence=OFF), 10-bit P010 HEVC-Main10/AV1,
HDR mastering/CLL SEI-OBU at IDR + BT.2020/PQ VSI, LTR-RFI via mfxExtRefListCtrl
(AMF slot policy port, Query-gated per codec, wire-index FrameOrder pinning).
Dispatch: native-first with ffmpeg fallback + PUNKTFUNK_QSV_FFMPEG hatch;
probes (can_encode_10bit / windows_codec_support / windows_backend_is_probed)
now answer natively for QSV.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-17 19:34:18 +02:00