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

ffmpeg-v4l2-request-fourier: substitute H.264 IDCT 4×4 → daedalus-fourier #76

Merged
marfrit merged 3 commits from claude-noether/marfrit-packages:noether/ffmpeg-fourier-idct4-daedalus into main 2026-05-21 19:57:32 +00:00
Conversation 0 Commits 3 Files Changed 6
+376 -5
marfrit commented 2026-05-21 19:45:24 +00:00
Owner
Copy Link
Copy Source

First cycle of the libavcodec.so substitution arc (reauktion/daedalus-v4l2#11 step 2). H264DSPContext.idct_add — called per 4×4 block from the intra-4×4 decode path in libavcodec/h264_mb.c — now dispatches through daedalus_recipe_dispatch_h264_idct4 instead of ff_h264_idct_add_neon.

What

  • Add 0003-h264-idct4-daedalus-fourier.patch (in both arch/ and debian/ ffmpeg-v4l2-request-fourier/). Creates libavcodec/aarch64/h264_idct_daedalus.c (ff_h264_idct_add_daedalus shim + lazy pthread_once context init via daedalus_ctx_create_no_qpu), patches libavcodec/aarch64/h264dsp_init_aarch64.c to wire c->idct_add to the shim, adds the new .o to libavcodec/aarch64/Makefile.
  • arch/PKGBUILD + debian/build-deb.sh: fetch + build daedalus-fourier (pinned at d87239d — lockstep with the daedalus-v4l2 daemon's inline build) with -DCMAKE_POSITION_INDEPENDENT_CODE=ON into a per-build temp prefix, then pass --extra-cflags=-I.../include --extra-ldflags=-L.../lib --extra-libs="-ldaedalus_core -lvulkan -lpthread" to FFmpeg configure. daedalus_core.a is static-linked into libavcodec.so.62.
  • debian/control Depends gains libvulkan1 (daedalus_core PUBLIC-links Vulkan::Vulkan for the queryable QPU substrate; the no-QPU constructor still works at runtime but the loader needs libvulkan.so.1 present to dlopen libavcodec.so.62).
  • arch/PKGBUILD depends gains vulkan-icd-loader, makedepends gains cmake / ninja / vulkan-headers.

Why

The recipe layer picks the substrate; for cycle 6 (H.264 IDCT 4×4) the recipe is CPU NEON, so this is effectively a NEON-to-NEON substitution with one extra dispatch call and recipe-table lookup. The point of this first cycle isn't perf wins — it's plumbing. Once the path is wired and stable, follow-up patches batch through the bulk paths (idct_add16 / idct_add16intra / idct_add8) and stack cycles 7/8/9 (IDCT 8×8, luma-v deblock, qpel mc20).

Bit-exact against ff_h264_idct_add_neon (daedalus-fourier cycle 6 green; FFmpeg's 4×4 block storage matches daedalus's column-major convention).

Scope NOT covered (deferred to follow-ups)

  • Bulk paths (idct_add16 / idct_add16intra / idct_add8) — most IDCT 4×4 calls in real H.264 streams go through these, not the per-block c->idct_add path; intra-4×4-only macroblocks are a minority. Batched substitution lands in a follow-up.
  • High-bit-depth (10-bit) path — not touched; 8-bit only.
  • Cycles 7/8/9 — separate PRs.

Risk

Dispatch overhead per single-block call (one library entry + recipe-table lookup) means a possible per-call regression compared to the in-tree NEON .S. Intra-4×4 is a minority of MBs in typical streams, so net daemon decode_us impact should be in noise. The daedalus-v4l2 daemon's decoder stats summary (added in r39+g3bc0da1) is the measurement instrument.

SONAME

Unchanged. libavcodec.so.62 / libavformat.so.62 / libavutil.so.60. No daedalus-v4l2-dkms or daedalus-v4l2 bump required.

Refs

  • reauktion/daedalus-v4l2 issue #11 (substitution arc): reauktion/daedalus-v4l2#11
  • marfrit/daedalus-fourier cycle 6 close (H.264 IDCT 4×4 NEON green)
First cycle of the libavcodec.so substitution arc (reauktion/daedalus-v4l2#11 step 2). H264DSPContext.idct_add — called per 4×4 block from the intra-4×4 decode path in libavcodec/h264_mb.c — now dispatches through `daedalus_recipe_dispatch_h264_idct4` instead of `ff_h264_idct_add_neon`. ## What - Add `0003-h264-idct4-daedalus-fourier.patch` (in both arch/ and debian/ ffmpeg-v4l2-request-fourier/). Creates `libavcodec/aarch64/h264_idct_daedalus.c` (`ff_h264_idct_add_daedalus` shim + lazy `pthread_once` context init via `daedalus_ctx_create_no_qpu`), patches `libavcodec/aarch64/h264dsp_init_aarch64.c` to wire `c->idct_add` to the shim, adds the new .o to `libavcodec/aarch64/Makefile`. - `arch/PKGBUILD` + `debian/build-deb.sh`: fetch + build daedalus-fourier (pinned at `d87239d` — lockstep with the daedalus-v4l2 daemon's inline build) with `-DCMAKE_POSITION_INDEPENDENT_CODE=ON` into a per-build temp prefix, then pass `--extra-cflags=-I.../include --extra-ldflags=-L.../lib --extra-libs="-ldaedalus_core -lvulkan -lpthread"` to FFmpeg configure. `daedalus_core.a` is static-linked into `libavcodec.so.62`. - `debian/control` Depends gains `libvulkan1` (daedalus_core PUBLIC-links Vulkan::Vulkan for the queryable QPU substrate; the no-QPU constructor still works at runtime but the loader needs `libvulkan.so.1` present to dlopen libavcodec.so.62). - `arch/PKGBUILD` depends gains `vulkan-icd-loader`, makedepends gains `cmake` / `ninja` / `vulkan-headers`. ## Why The recipe layer picks the substrate; for cycle 6 (H.264 IDCT 4×4) the recipe is CPU NEON, so this is effectively a NEON-to-NEON substitution with one extra dispatch call and recipe-table lookup. The point of this first cycle isn't perf wins — it's plumbing. Once the path is wired and stable, follow-up patches batch through the bulk paths (idct_add16 / idct_add16intra / idct_add8) and stack cycles 7/8/9 (IDCT 8×8, luma-v deblock, qpel mc20). Bit-exact against `ff_h264_idct_add_neon` (daedalus-fourier cycle 6 green; FFmpeg's 4×4 block storage matches daedalus's column-major convention). ## Scope NOT covered (deferred to follow-ups) - Bulk paths (idct_add16 / idct_add16intra / idct_add8) — most IDCT 4×4 calls in real H.264 streams go through these, not the per-block `c->idct_add` path; intra-4×4-only macroblocks are a minority. Batched substitution lands in a follow-up. - High-bit-depth (10-bit) path — not touched; 8-bit only. - Cycles 7/8/9 — separate PRs. ## Risk Dispatch overhead per single-block call (one library entry + recipe-table lookup) means a possible per-call regression compared to the in-tree NEON `.S`. Intra-4×4 is a minority of MBs in typical streams, so net daemon `decode_us` impact should be in noise. The daedalus-v4l2 daemon's `decoder stats` summary (added in r39+g3bc0da1) is the measurement instrument. ## SONAME Unchanged. libavcodec.so.62 / libavformat.so.62 / libavutil.so.60. No daedalus-v4l2-dkms or daedalus-v4l2 bump required. ## Refs - reauktion/daedalus-v4l2 issue #11 (substitution arc): https://git.reauktion.de/reauktion/daedalus-v4l2/issues/11 - marfrit/daedalus-fourier cycle 6 close (H.264 IDCT 4×4 NEON green)
marfrit added 1 commit 2026-05-21 19:45:24 +00:00
ffmpeg-v4l2-request-fourier: substitute H.264 IDCT 4×4 → daedalus-fourier e641d679d3 
First cycle of the libavcodec.so substitution arc (reauktion/daedalus-v4l2#11
step 2).  H264DSPContext.idct_add — called per 4×4 block from the
intra-4×4 decode path in libavcodec/h264_mb.c — now dispatches through
daedalus_recipe_dispatch_h264_idct4 instead of ff_h264_idct_add_neon.

## What

- Add 0003-h264-idct4-daedalus-fourier.patch (in both arch/ and
  debian/ ffmpeg-v4l2-request-fourier/).  Creates
  libavcodec/aarch64/h264_idct_daedalus.c (ff_h264_idct_add_daedalus
  shim + lazy pthread_once context init via
  daedalus_ctx_create_no_qpu), patches
  libavcodec/aarch64/h264dsp_init_aarch64.c to wire c->idct_add to
  the shim, adds the new .o to libavcodec/aarch64/Makefile.
- arch/PKGBUILD + debian/build-deb.sh: fetch + build
  daedalus-fourier (pinned at d87239d — lockstep with the
  daedalus-v4l2 daemon's inline build) with
  -DCMAKE_POSITION_INDEPENDENT_CODE=ON into a per-build temp prefix,
  then pass --extra-cflags=-I.../include --extra-ldflags=-L.../lib
  --extra-libs="-ldaedalus_core -lvulkan -lpthread" to FFmpeg
  configure.  daedalus_core.a is static-linked into libavcodec.so.62.
- debian/control Depends gains libvulkan1 (daedalus_core PUBLIC-links
  Vulkan::Vulkan for the queryable QPU substrate; the no-QPU
  constructor still works at runtime but the loader needs
  libvulkan.so.1 present to dlopen libavcodec.so.62).
- arch/PKGBUILD depends gains vulkan-icd-loader, makedepends gains
  cmake / ninja / vulkan-headers.

## Why

The recipe layer picks the substrate; for cycle 6 (H.264 IDCT 4×4)
the recipe is CPU NEON, so this is effectively a NEON-to-NEON
substitution with one extra dispatch call and recipe-table lookup.
The point of this first cycle isn't perf wins — it's plumbing.  Once
the path is wired and stable, follow-up patches batch through the
bulk paths (idct_add16 / idct_add16intra / idct_add8) and stack
cycles 7/8/9 (IDCT 8×8, luma-v deblock, qpel mc20).

Bit-exact against ff_h264_idct_add_neon (daedalus-fourier cycle 6
green; FFmpeg's 4×4 block storage matches daedalus's column-major
convention).

## Scope NOT covered

- Bulk paths (idct_add16 / idct_add16intra / idct_add8) — most IDCT
  4×4 calls in real H.264 streams go through these, not the per-
  block c->idct_add path; intra-4×4-only macroblocks are a minority.
  Batched substitution lands in a follow-up.
- High-bit-depth (10-bit) path — not touched; 8-bit only.
- Cycles 7/8/9 — separate PRs.

## SONAME

Unchanged.  libavcodec.so.62 / libavformat.so.62 / libavutil.so.60.
No daedalus-v4l2-dkms or daedalus-v4l2 bump required.

## Refs

- reauktion/daedalus-v4l2 issue #11 (substitution arc): reauktion/daedalus-v4l2#11
- marfrit/daedalus-fourier cycle 6 close (H.264 IDCT 4×4 NEON green)
claude-noether added 1 commit 2026-05-21 19:47:51 +00:00
CI: add cmake / ninja-build / libvulkan-dev / glslang-tools to ffmpeg-debian deps 88a65cb6d0 
The ffmpeg-v4l2-request-debian job now needs to build daedalus-fourier
into a temp prefix before configuring FFmpeg (substitution patch
0003-h264-idct4-daedalus-fourier.patch links libdaedalus_core.a into
libavcodec.so).  Mirror the build-deps the daedalus-v4l2-debian job
already declared for the same reason.

No-op on Arch — makepkg --syncdeps auto-installs cmake/ninja/
vulkan-headers from the PKGBUILD makedepends.
claude-noether added 1 commit 2026-05-21 19:56:50 +00:00
Merge remote-tracking branch 'origin/main' into noether/ffmpeg-fourier-idct4-daedalus a1dba5f630
marfrit merged commit a536e20218 into main 2026-05-21 19:57:32 +00:00
marfrit deleted branch noether/ffmpeg-fourier-idct4-daedalus 2026-05-21 19:57:32 +00:00
Sign in to join this conversation.
Reviewers
No Reviewers
Labels
No items
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
marfrit (Markus Fritsche)
No Assignees
2 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: marfrit/marfrit-packages#76
Delete Branch "claude-noether/marfrit-packages:noether/ffmpeg-fourier-idct4-daedalus"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?