feat(linux/vulkan-encode): default-on native NV12 capture + RGB true-extent
ci / web (push) Successful in 48s
ci / docs-site (push) Successful in 1m5s
apple / swift (push) Successful in 1m20s
decky / build-publish (push) Successful in 21s
docker / build-push (--build-arg FEDORA_VERSION=44, ci, ci/fedora-rpm.Dockerfile, punktfunk-fedora44-rpm) (push) Successful in 25s
docker / build-push (., web/Dockerfile, punktfunk-web) (push) Successful in 42s
ci / bench (push) Successful in 5m59s
apple / screenshots (push) Successful in 6m17s
docker / build-push (ci, ci/fedora-rpm.Dockerfile, punktfunk-fedora-rpm) (push) Successful in 6m29s
docker / build-push (ci, ci/rust-ci-noble.Dockerfile, punktfunk-rust-ci-noble) (push) Successful in 5m12s
docker / build-push (docs-site, docs-site/Dockerfile, punktfunk-docs) (push) Successful in 1m16s
deb / build-publish (push) Successful in 12m58s
deb / build-publish-host (push) Successful in 13m40s
windows-host / package (push) Successful in 10m46s
docker / build-push (ci, ci/rust-ci.Dockerfile, punktfunk-rust-ci) (push) Successful in 6m52s
android / android (push) Successful in 16m18s
arch / build-publish (push) Successful in 16m20s
docker / deploy-docs (push) Successful in 30s
ci / rust (push) Successful in 19m25s
rpm / build-publish (43, bazzite, punktfunk-fedora-rpm) (push) Successful in 14m19s
rpm / build-publish (44, fedora-44, punktfunk-fedora44-rpm) (push) Successful in 15m19s

Flip both zero-copy levers from opt-in to default:

- The capture negotiation now PREFERS gamescope's producer-side NV12 pod by
  default (PUNKTFUNK_PIPEWIRE_NV12=0 restores the packed-RGB negotiation).
  Codec-aware gating rides a new OutputFormat::nv12_native ->
  ZeroCopyPolicy::native_nv12_session edge resolved by the host facade from
  the session plan (pf_encode::linux_native_nv12_ok): only H265/AV1 sessions
  whose backend can open the raw Vulkan Video encoder ever see the NV12 pod
  -- an H264/Moonlight session (libav VAAPI, which would misread the
  two-plane buffer) keeps today's BGRx negotiation, as do the
  GameStream-resolve and portal-mirror paths, PyroWave, and NVENC prefs.
- RGB-direct's unaligned modes default to the true-extent direct import
  (PUNKTFUNK_VULKAN_RGB_TRUE_EXTENT=0 restores the padded-copy staging).
  Guarded-tested on Van Gogh with the kernel journal watched: clean, and at
  5.38 ms p50 the fastest 1080p encode path measured on that hardware. The
  EFC only exists on Mesa >= 26, where the codedExtent-driven session_init
  padding is guaranteed (verified back to Mesa 24.2).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
2026-07-21 13:42:03 +02:00
parent b4fbc94e36
commit 42198eb1b6
7 changed files with 91 additions and 29 deletions
+6
View File
@@ -259,6 +259,12 @@ pub struct ZeroCopyPolicy {
/// passthrough (like the VAAPI backend) instead of the EGL→CUDA import whose payloads only
/// NVENC can consume. Per-session (the codec is negotiated), unlike `backend_is_vaapi`.
pub pyrowave_session: bool,
/// THIS session's encoder can ingest a producer-native NV12 capture (the Linux raw Vulkan
/// Video backend on an H265/AV1 session — resolved by the host facade via
/// `pf_encode::linux_native_nv12_ok`). Gates whether the negotiation PREFERS gamescope's
/// producer-side NV12 pod: libav VAAPI (H264's backend) would misread the two-plane buffer,
/// so H264/GameStream/PyroWave sessions must never see NV12 frames.
pub native_nv12_session: bool,
/// The PyroWave encoder's Vulkan-importable dmabuf modifiers for the capture's packed-RGB fourcc,
/// resolved when the session encodes PyroWave (the passthrough advertises them so Mutter+NVIDIA,
/// which allocates tiled-only, still negotiates zero-copy). Empty otherwise.
+13 -9
View File
@@ -2086,22 +2086,26 @@ mod pipewire {
// PyroWave session (the wavelet encoder's own Vulkan device, any vendor) → hand the raw
// dmabuf straight to the encoder.
let vaapi_passthrough = zerocopy && !force_shm && importer.is_none() && raw_passthrough;
// Bring-up gate for producer-side NV12. Gamescope offers a one-fd LINEAR NV12 image when
// the consumer asks for it; keep this opt-in until the live Vulkan import path is proven.
// Raw passthrough is required because CUDA/PyroWave expect packed RGB, and 4:4:4/HDR must
// not be silently subsampled/downconverted.
let prefer_native_nv12 = std::env::var("PUNKTFUNK_PIPEWIRE_NV12").as_deref() == Ok("1")
// Producer-side NV12 (default-on; PUNKTFUNK_PIPEWIRE_NV12=0 escapes): gamescope offers a
// one-fd LINEAR NV12 image when the consumer asks — its compositor pass does the RGB→YUV,
// and the Vulkan Video encoder imports the buffer as its encode source directly (no host
// CSC at all). `native_nv12_session` restricts this to sessions whose encoder can ingest
// it (Linux vulkan-encode H265/AV1 — never H264/libav-VAAPI, GameStream-resolve, or
// PyroWave, whose Vulkan compute CSC ingests packed RGB only). Raw passthrough is
// required because the CUDA importer expects packed RGB, and 4:4:4/HDR must not be
// silently subsampled/downconverted. Non-NV12 compositors (KWin/GNOME) simply match the
// packed-RGB fallback pod.
let prefer_native_nv12 = std::env::var("PUNKTFUNK_PIPEWIRE_NV12").as_deref() != Ok("0")
&& policy.native_nv12_session
&& backend_is_vaapi
&& vaapi_passthrough
// PyroWave rides the same raw passthrough but its Vulkan compute CSC ingests packed
// RGB only — native NV12 would feed it an unreadable two-plane image.
&& !policy.pyrowave_session
&& !want_444
&& !want_hdr;
if prefer_native_nv12 {
tracing::info!(
"PUNKTFUNK_PIPEWIRE_NV12=1: preferring gamescope producer-side NV12 LINEAR \
DMA-BUF (no host RGB CSC)"
"zero-copy: preferring gamescope producer-side NV12 LINEAR DMA-BUF (no host \
RGB CSC; PUNKTFUNK_PIPEWIRE_NV12=0 restores the packed-RGB negotiation)"
);
}
// Modifiers our import stack handles for BGRx: the EGL-importable (tiled) set, plus LINEAR