marfrit/marfrit-packages
0
0
Fork 1
Code Issues 2 Pull Requests Actions Packages Projects Releases Wiki Activity

ffmpeg-v4l2-request-fourier: revert ctx flip — PR #36 was a measurement artifact (0015)

Browse Source 
Reverts the no_qpu → qpu-capable ctx flip that landed via patch 0014
(marfrit-packages PR #104).

PR #104 was justified by daedalus-fourier PR #36's "QPU 4.30x faster
than CPU NEON" bench result.  That number was a measurement artifact:
v3d_runner.read_spv() did a bare cwd-relative fopen() with no path
search, so when the bench was run from the source dir (as in PR #36),
the SPVs at $builddir/v3d_*.spv were not found, every QPU dispatch
returned -1 fast, and the bench loop timed the failure path.

daedalus-fourier PR #37 fixes the SPV search + bench preflight.
Corrected numbers on hertz (Pi 5 V3D 7.1):

  kernel             CPU ns/op  QPU ns/op  winner
  IDCT 4x4 luma          10.75     217.63  CPU 20.24x
  IDCT 8x8 luma          29.69     785.94  CPU 26.47x
  Deblock luma_v         17.63     467.42  CPU 26.51x
  Deblock luma_h         38.30     498.53  CPU 13.02x
  qpel mc20 (8x8)        30.17    1300.44  CPU 43.10x
  qpel mc02 (8x8)        17.69    1363.40  CPU 77.08x
  qpel mc22 (8x8)        71.60    1948.37  CPU 27.21x

  1080p sum: CPU 5.57 ms vs QPU 123.54 ms — QPU 22x SLOWER.

Until daedalus QPU dispatch overhead is actually competitive (separate
multi-task effort tracked on the daedalus-fourier side), libavcodec.so
substitution must stay on daedalus_ctx_create_no_qpu() so the host
processes (firefox-fourier RDD, mpv-fourier, daedalus_v4l2_daemon)
don't pessimize their H.264 decode path.

Adds 0015-h264-ctx-revert-to-no-qpu.patch (2-line revert of patch 0014)
to both arch PKGBUILD and debian build-deb.sh.  Both pkgrel bumped
11 → 12.  Refs reauktion/daedalus-fourier!37.
This commit is contained in:
Claude (noether)
2026-05-25 21:47:33 +02:00
parent 190f810843
commit f4047f3145
4 changed files with 153 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=11 # pkgrel=11 — libavcodec.so daedalus ctx flipped no_qpu → qpu-capable (PR #36 bench: QPU 4.30x on 1080p hot-path sum, 2026-05-25)
pkgrel=12 # pkgrel=12 — REVERT pkgrel=11 ctx flip; daedalus-fourier PR #36 4.30x headline was measurement artifact (PR #37 corrects: QPU 22x SLOWER than CPU)
epoch=2
# daedalus-fourier pin. 209a421 = PR #2 merge (Phase 8c — public API
@@ -101,8 +101,9 @@ source=("git+https://github.com/Kwiboo/FFmpeg.git#commit=${_commit}"
'0011-h264-chroma-dc-hadamard-daedalus-fourier.patch'
'0012-h264-qpel-rest-daedalus-fourier.patch'
'0013-h264-deblock-chroma-intra-daedalus-fourier.patch'
'0014-h264-ctx-qpu-capable.patch')
sha256sums=('SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP')
'0014-h264-ctx-qpu-capable.patch'
'0015-h264-ctx-revert-to-no-qpu.patch')
sha256sums=('SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP' 'SKIP')
pkgver() {
cd "${_srcname}"
@@ -127,6 +128,7 @@ prepare() {
patch -Np1 -i "${srcdir}/0012-h264-qpel-rest-daedalus-fourier.patch"
patch -Np1 -i "${srcdir}/0013-h264-deblock-chroma-intra-daedalus-fourier.patch"
patch -Np1 -i "${srcdir}/0014-h264-ctx-qpu-capable.patch"
patch -Np1 -i "${srcdir}/0015-h264-ctx-revert-to-no-qpu.patch"
}
build() {