claude-noether/marfrit-packages
1
0
Fork 0
forked from marfrit/marfrit-packages
Code Pull Requests Activity

ffmpeg-v4l2-request-fourier: flip libavcodec daedalus ctx no_qpu → qpu-capable (0013)

Browse Source 
Patches 0003 (IDCT 4x4) and 0007 (qpel mc20) created the libavcodec.so
process-global daedalus_ctx via daedalus_ctx_create_no_qpu().  Rationale
at the time: cycle 6/9 had only CPU NEON paths, so a QPU-capable ctx
would have meant pointless Vulkan init in every host process (firefox-
fourier, mpv-fourier, daedalus_v4l2_daemon, ...).

Two things changed since:

  1. Every H.264 hot-path primitive now has a V3D7 compute shader.
     IDCT 4x4/8x8 + 8 deblock variants (luma+chroma × V+H × inter+intra)
     + 30 qpel positions (15 put_ + 15 avg_).  See daedalus-fourier
     PRs #28-#35.

  2. Dispatch overhead has been hammered down — buffer pool in
     v3d_runner (#160) plus persistent command buffer (#161).
     daedalus-fourier PR #36 bench measures the 1080p worst-case
     sum on hertz (Pi 5 V3D 7.1, 30 iters x 5 warmup):

       1080p worst-case sum (IDCT4 + deblock luma + qpel mc22):
         CPU NEON only:  5.57 ms
         QPU only:       1.30 ms   (CPU/QPU sum ratio = 4.30x)

PR #10's CPU-4x-faster-than-QPU verdict (which justified the original
no_qpu ctx choice) is now reversed by ~17x.

This commit adds 0013-h264-ctx-qpu-capable.patch which flips both H.264
TUs (h264_idct_daedalus.c, h264_qpel_daedalus.c) from
daedalus_ctx_create_no_qpu() to daedalus_ctx_create().

daedalus_ctx_create() probes for a usable Vulkan device and falls back
to no_qpu mode if unavailable, so this is safe on hosts without V3D
(x86 build runners, Debian aarch64 builders without renderD, etc.).
Hosts WITH V3D (Pi 5 deployment targets) now route the H.264 hot-path
through V3D compute instead of CPU NEON.

Wired into both arch PKGBUILD (source[] + prepare()) and debian
build-deb.sh; both pkgrel bumped 10 → 11.

Refs reauktion/daedalus-fourier!36.
This commit is contained in:
Claude (noether)
2026-05-25 21:09:31 +02:00
parent e323aa2316
commit 6047c04f7f
4 changed files with 177 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
arch/ffmpeg-v4l2-request-fourier/PKGBUILD
+5 -3
View File
@@ -24,7 +24,7 @@ _srcname=FFmpeg
_version='8.1'
_commit='b57fbbe50c9b2656fad86a1a7eeabfd2b2a50935' # v4l2-request-n8.1 tip 2026-04-24
pkgver=8.1.r123329.b57fbbe
pkgrel=10 # pkgrel=10 — H.264 luma qpel mc20 daedalus-fourier substitution (cycle 9, 2026-05-23)
pkgrel=11 # pkgrel=11 — libavcodec.so daedalus ctx flipped no_qpu → qpu-capable (PR #36 bench: QPU 4.30x, 2026-05-25)
epoch=2
# daedalus-fourier pin. 209a421 = PR #2 merge (Phase 8c — public API
@@ -99,8 +99,9 @@ source=("git+https://github.com/Kwiboo/FFmpeg.git#commit=${_commit}"
'0009-h264-deblock-chroma-daedalus-fourier.patch'
'0010-h264-deblock-luma-intra-daedalus-fourier.patch'
'0011-h264-chroma-dc-hadamard-daedalus-fourier.patch'
'0012-h264-qpel-rest-daedalus-fourier.patch')
sha256sums=('SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP')
'0012-h264-qpel-rest-daedalus-fourier.patch'
'0013-h264-ctx-qpu-capable.patch')
sha256sums=('SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP')
pkgver() {
cd "${_srcname}"
@@ -123,6 +124,7 @@ prepare() {
patch -Np1 -i "${srcdir}/0010-h264-deblock-luma-intra-daedalus-fourier.patch"
patch -Np1 -i "${srcdir}/0011-h264-chroma-dc-hadamard-daedalus-fourier.patch"
patch -Np1 -i "${srcdir}/0012-h264-qpel-rest-daedalus-fourier.patch"
patch -Np1 -i "${srcdir}/0013-h264-ctx-qpu-capable.patch"
}
build() {