Pre-Compact Handoff — Session 2026-05-14 (FINAL post iter34)

Use this doc to resume the fresnel-fourier campaign after Claude context compaction. Campaign is at full close state: 5/5 codecs PASS.

TL;DR (read first)

Bug	Status	Fix
Bug 4 (H.264 keyframe-partial)	FIXED iter25 α-25	rkvdec image_fmt pre-seed via synthetic SPS at CreateContext
Bug 5 (HEVC libva all-zero CAPTURE)	FIXED iter25 + iter31	α-25 (image_fmt) + α-29 (slice_params.short_term_ref_pic_set_size from VAAPI st_rps_bits)
VP8 wrong output through libva	FIXED iter33 α-30	prepend VP8 uncompressed frame header (10 kf / 3 inter) to OUTPUT
MPEG-2 HW differs from SW	NOT A BUG	hantro IDCT precision (≤1 LSB / ~67 px); libva==kdirect bit-exact
Kernel diagnostic printks	CLEANED iter32 + iter34	7.0-14 ship

Codec	libva 10F sha	kdirect 10F sha	SW 10F sha	L==K	L==SW
H.264	dd4f5f2d552c07bc	same	same	✓	✓
HEVC	108f925bb6cbb6c9	same	same	✓	✓
VP9	cf35908ae0f9ab60	same	same	✓	✓
VP8	d3231e5b6c0ee10b	same	same	✓	✓
MPEG-2	95c5905890c937d4	same	933b744134e47ba4	✓	~

5/5 PASS the libva-vs-kdirect bit-exact contract.

Substrate state (where things live)

Component	Location	Tip
Campaign repo (this)	`/home/mfritsche/src/fresnel-fourier/`	`70ddbd6` on gitea master
Libva backend fork (noether)	`/home/mfritsche/src/libva-multiplanar/libva-v4l2-request-fourier/`	`7e0848d` on gitea master
Libva backend (fresnel deploy)	`/home/mfritsche/src/libva-v4l2-request-fourier/`	sync to gitea master, `ninja -C build`
Kernel source (boltzmann)	`~/src/kernel-agent-bootstrap/build/marfrit-packages/arch/linux-fresnel-fourier/`	pkgrel=14 clean (no diagnostic printks)
Kernel running on fresnel	`linux-fresnel-fourier 7.0-14`	clean shipping kernel
Test fixtures (fresnel)	`/home/mfritsche/fourier-test/bbb_*.{mp4,ts,webm}`	5 codecs at 720p10s or 1080p30
Anchors (fresnel)	`/tmp/final/{L,K,S}_<codec>.yuv`	10-frame YUV per codec per backend
Memory	`~/.claude/projects/-home-mfritsche-src-fresnel-fourier/memory/`	see entries below

Identity for gitea pushes

All git.reauktion.de interactions use claude-noether identity (per memory feedback_gitea_as_claude_noether.md). Backend remote URL: ssh://gitea@git.reauktion.de.claude-noether/marfrit/libva-v4l2-request-fourier.git.

Device map on 7.0-14 (REVERSED from 7.0-13)

/dev/video* and /dev/media* numbers SHIFT between kernel boots based on probe order. On 7.0-14 (current):

Driver	/dev/videoN	/dev/mediaN
rockchip-rga	video0	n/a
rk3399-vpu-enc	video1	(shared)
rk3399-vpu-dec (hantro)	video2	media0
rkvdec	video3	media1

Always re-probe via v4l2-ctl --info + media-ctl -p before hardcoding paths.

Backend commits delivered (chronological, this campaign day)

7e0848d iter33 α-30: prepend VP8 uncompressed frame header to OUTPUT buffer    ← VP8 fix
bf3e3d8 iter33: extend VP8 DIAG to dump VAAPI probability struct directly      (env-gated diag)
4b3c21b iter33 DIAG: env-gated dump of v4l2_ctrl_vp8_frame contents            (env-gated diag)
23eb1bd iter31 α-29: slice_params.short_term_ref_pic_set_size = picture->st_rps_bits  ← HEVC fix
68dbbdd iter30 DIAG: LIBVA_TS_SCALE env-gated timestamp multiplier             (env-gated diag)
0eca3ff iter29 DIAG: env-gated dump of HEVC slice_data trailing 80 bytes       (env-gated diag)
6646b16 Revert iter28b DIAG: trim=40 universal-trim broke IDR frame 1
cd286d9 iter28 α-28: bit_size = (slice_data_size - data_byte_offset) * 8 for HEVC
754be1d iter27 diag: env-gated VAAPI slice fields dump
719d813 iter27 α-27: populate slice_params.num_entry_point_offsets             (no-op)
66ef848 iter26 α-26: decode_params.short_term_ref_pic_set_size from VAAPI       (mis-routed cosmetic)
d062fec iter25 α-25 fix: FRAME_MBS_ONLY flag for H264 dummy SPS
db0b7f9 iter25 α-25: inject synthetic SPS before cap_pool_init to seed image_fmt  ← H264+HEVC frame 1 fix

The load-bearing commits are db0b7f9 + d062fec (α-25), 23eb1bd (α-29), 7e0848d (α-30). The DIAG commits are env-gated and inactive by default.

Campaign repo commits delivered (today's arc)

70ddbd6 iter34 close: kernel 7.0-14 CLEAN ship — 5/5 codecs PASS
cd2d077 iter33: MPEG-2 closed (libva==kdirect bit-exact) — 5/5 codecs PASS
51eee19 iter33 α-30 close: VP8 FIXED — 4/5 codecs PASS
acacf3d iter32 close: kernel substrate cleanup landed → 7.0-11 SHIPPING
85cc178 Update campaign session doc: full-day arc closes at 3/3 PASS
fde8a25 Update handoff doc: HEVC Bug 5 fully fixed (3/3 PASS)
c1f9738 iter31 α-29 close: HEVC Bug 5 remainder FIXED — 3/3 PASS
422ecaf Add pre-compact handoff doc for session resumption
…earlier in day:
c15fc6c, 8b17bf7, 02c4192, bf67900 (iter20-28 chain)

How to verify the current state

Run on fresnel (post-7.0-14 boot, devices: rkvdec /dev/video3+/dev/media1, hantro /dev/video2+/dev/media0):

for codec in h264:bbb_1080p30_h264.mp4:rk hevc:bbb_720p10s_hevc.mp4:rk vp9:bbb_720p10s_vp9.webm:rk vp8:bbb_720p10s_vp8.webm:ha mpeg2:bbb_720p10s_mpeg2.ts:ha; do
    name="${codec%%:*}"; rest="${codec#*:}"; fixture="${rest%:*}"; dev="${rest##*:}"
    if [ "$dev" = "rk" ]; then V=/dev/video3; M=/dev/media1
    else V=/dev/video2; M=/dev/media0; fi
    env LIBVA_DRIVER_NAME=v4l2_request \
        LIBVA_DRIVERS_PATH=/home/mfritsche/src/libva-v4l2-request-fourier/build/src \
        LIBVA_V4L2_REQUEST_VIDEO_PATH=$V LIBVA_V4L2_REQUEST_MEDIA_PATH=$M \
        ffmpeg -hide_banner -loglevel error -y \
        -hwaccel vaapi -hwaccel_output_format vaapi \
        -i "/home/mfritsche/fourier-test/$fixture" \
        -vf "hwdownload,format=nv12" -frames:v 10 \
        -f rawvideo -pix_fmt nv12 "/tmp/L_${name}.yuv"
    ffmpeg -hide_banner -loglevel error -y -hwaccel v4l2request -hwaccel_output_format drm_prime \
        -i "/home/mfritsche/fourier-test/$fixture" -vf "hwdownload,format=nv12" \
        -frames:v 10 -f rawvideo -pix_fmt nv12 "/tmp/K_${name}.yuv"
    L=$(sha256sum "/tmp/L_${name}.yuv" | cut -c1-16)
    K=$(sha256sum "/tmp/K_${name}.yuv" | cut -c1-16)
    [ "$L" = "$K" ] && echo "$name: PASS" || echo "$name: FAIL"
done

Expect: 5× PASS.

Root cause summary

Bug 4 + Bug 5 frame 1 (iter25 α-25): rkvdec_s_ctrl returns -EBUSY when first SPS triggers image_fmt reset on busy CAPTURE queue. libva pre-allocated 24 CAPTURE buffers at CreateContext (iter5b-β) before per-frame S_EXT_CTRLS. Fix: inject synthetic SPS at CreateContext, pre-cap_pool_init, while CAPTURE is empty.

Bug 5 frame 2+ (iter31 α-29): libva backend set slice_params->short_term_ref_pic_set_size = 0 (stale "VAAPI doesn't expose" comment). rkvdec's assemble_sw_rps (rkvdec-hevc.c:386-389) reads this; when zero with num_short_term_ref_pic_sets <= 1, falls back to 0 → entropy decoder consumes slice-header bits as long-term-RPS → garbage for every non-IDR slice. IDR is gated by !IDR_PIC so frame 1 was unaffected. Fix: slice_params->short_term_ref_pic_set_size = picture->st_rps_bits (VAAPI's field IS the slice-header bit count, per va_dec_hevc.h doc). α-26 had mis-routed this value into decode_params (same field name, different V4L2 semantics).

VP8 (iter33 α-30): ffmpeg-vaapi strips the VP8 uncompressed frame header (3 bytes interframe / 10 bytes keyframe) before submitting via VAAPI. ffmpeg-v4l2request keeps it. Hantro hard-codes first_part_offset = V4L2_VP8_FRAME_IS_KEY_FRAME(hdr) ? 10 : 3 and uses it for both mb_offset_bits and dct_part_offset. Without the prepended header in libva's OUTPUT, hantro's offset arithmetic lands inside the compressed bitstream and the entropy decoder produces garbage. Fix: in codec_store_buffer, prepend header_size zero bytes to OUTPUT for VP8 profile (hantro skips these bytes for actual parsing, uses ctrl-struct values).

Open items (low priority, no blockers)

Backend env-gated diagnostics cleanup — LIBVA_HEVC_DUMP_SLICE_TAIL (iter29), LIBVA_TS_SCALE (iter30), LIBVA_VP8_DUMP_FRAME (iter33) are env-gated and inactive by default. Leave for future regression debugging or clean up. Low priority.
α-26 cosmetic revert — decode_params->short_term_ref_pic_set_size = picture->st_rps_bits was mis-routed (rkvdec doesn't use that field). Could revert to 0. Cosmetic; no behavior change.
Libva multi-device probe — currently find_codec_device picks ONE device per session, requiring LIBVA_V4L2_REQUEST_VIDEO_PATH override to access both rkvdec (H264/HEVC/VP9) and hantro (VP8/MPEG-2) within one workflow. Architectural change in src/request.c::find_codec_device (~200-400 LOC). Design judgment from user welcome.

Memory entries (full set, this campaign)

feedback_rkvdec_image_fmt_pre_seed.md — α-25 (Bug 4 + Bug 5 frame 1)
feedback_va_st_rps_bits_is_slice_field.md — α-29 (Bug 5 frame 2+)
feedback_vaapi_strips_vp8_uncompressed_header.md — α-30 (VP8)
feedback_libva_byte_correct_kernel_bug.md — FULLY OVERTURNED (both Bug 4 + Bug 5 are libva-side fixes)
reference_fresnel_kernel_substrate.md — 7.0-14 clean, device-enumeration caveat noted
MEMORY.md index updated

Key commands quickreference

# Sync backend on fresnel + rebuild
ssh fresnel 'cd ~/src/libva-v4l2-request-fourier && git fetch && git reset --hard origin/master && ninja -C build'

# Identify which video device is rkvdec vs hantro after a fresh boot
ssh fresnel 'for v in /dev/video*; do v4l2-ctl -d $v --info 2>/dev/null | grep -E "^Card type" | head -1 | awk -v dev=$v "{print dev,\$0}"; done'

# 5-codec smoke (above script)

# Run libva HEVC (rkvdec is currently /dev/video3 on 7.0-14)
ssh fresnel 'env LIBVA_DRIVER_NAME=v4l2_request \
    LIBVA_DRIVERS_PATH=/home/mfritsche/src/libva-v4l2-request-fourier/build/src \
    LIBVA_V4L2_REQUEST_VIDEO_PATH=/dev/video3 LIBVA_V4L2_REQUEST_MEDIA_PATH=/dev/media1 \
    ffmpeg -hide_banner -loglevel error -y -hwaccel vaapi -hwaccel_output_format vaapi \
    -i /home/mfritsche/fourier-test/bbb_720p10s_hevc.mp4 \
    -vf "hwdownload,format=nv12" -frames:v 10 -f rawvideo -pix_fmt nv12 /tmp/x.yuv'

# kdirect reference (works for any codec; hwaccel auto-routes)
ssh fresnel 'ffmpeg -hide_banner -loglevel error -y -hwaccel v4l2request -hwaccel_output_format drm_prime \
    -i /home/mfritsche/fourier-test/bbb_720p10s_hevc.mp4 \
    -vf "hwdownload,format=nv12" -frames:v 10 -f rawvideo -pix_fmt nv12 /tmp/y.yuv'

# Reboot fresnel (sddm autologin reseats mfritsche)
ssh fresnel 'sudo systemctl reboot'; sleep 60

Safe vs needs-confirmation actions

Safe (no confirmation needed):

Read/grep on noether, boltzmann, fresnel
Push to gitea (claude-noether identity)
Reboot fresnel (sddm autologin restores session)
Build kernel on boltzmann via makepkg -ef --skipinteg --noconfirm
Deploy kernel via scp + sudo pacman -U
Run ffmpeg/cmp tests on fresnel

Needs confirmation:

Significant rebuild (~25-30 min CPU on boltzmann, e.g. ffmpeg full rebuild or fresh kernel build)
Architectural changes to libva multi-device probe (item 3 above) — affects backend design

11 KiB Raw Blame History Unescape Escape