Phase 8.5: full V4L2 m2m driver, VP9 decode via QBUF/DQBUF

Replaces the Phase 8.4 debugfs-triggered chardev path with a
real V4L2 m2m driver. Userspace clients now drive decoding the
standard way — S_FMT / REQBUFS / QBUF on the OUTPUT (bitstream)
queue, DQBUF on the CAPTURE (NV12M) queue. Kernel device_run
packs the bitstream into REQ_DECODE; daemon decodes via FFmpeg;
RESP_FRAME's inline NV12 pixel payload lands in the CAPTURE
buffer. Phase 8.6 swaps the inline payload for dmabuf so big
frames stop being capped at 64 KiB.

Kernel (daedalus_v4l2_main.c, rewritten + main.h added):
- Per-open struct daedalus_ctx: v4l2_fh, m2m_ctx, ctrl_handler,
  per-queue v4l2_pix_format_mplane.
- Two vb2_queues (vb2_vmalloc_memops for both — no DMA needed
  yet; 8.6 switches CAPTURE to dma_contig for dmabuf-export):
    OUTPUT  = V4L2_BUF_TYPE_VIDEO_OUTPUT_MPLANE,  VP9_FRAME
    CAPTURE = V4L2_BUF_TYPE_VIDEO_CAPTURE_MPLANE, NV12M
- Full v4l2_ioctl_ops table: querycap, enum_fmt, g/s/try_fmt
  for both queues, reqbufs/querybuf/qbuf/dqbuf/create_bufs/
  prepare_buf/expbuf/streamon/streamoff via v4l2_m2m_ioctl_*
  helpers.
- v4l2_m2m_ops.device_run: peeks next OUTPUT buf, builds
  REQ_DECODE inline with the bitstream bytes, enqueues with an
  auto-incrementing cookie, stores {ctx, src_buf, dst_buf} in
  a per-device inflight list. Job stays open until RESP_FRAME.
- daedalus_complete_resp_frame(): pops the inflight entry,
  memcpys inline NV12 pixels into the CAPTURE buffer (Y plane
  + interleaved CbCr), finishes via
  v4l2_m2m_buf_done_and_job_finish — NOT plain buf_done +
  job_finish, which leaves the src buf on the m2m queue and
  causes device_run to immediately re-run on the same input
  (caught on first run; second REQ_DECODE for same bitstream +
  eventual oops in stop_streaming on teardown).

Kernel (daedalus_v4l2_chardev.c):
- RESP_FRAME handler now hands inline pixel payload to
  daedalus_complete_resp_frame so it lands in the CAPTURE
  vb2 buffer. Existing PONG and debugfs test_decode paths still
  work; the latter produces a harmless ratelimited "unknown
  cookie" since it bypasses V4L2 m2m.

Daemon (decoder.c, decoder.h):
- daedalus_decoder_run_request signature extended with
  (nv12_out, nv12_cap, nv12_used). After the FNV-1a digest the
  decoder packs YUV420P into NV12 in the caller's buffer: Y
  plane line-by-line stripped of stride padding; Cb/Cr
  interleaved into a single chroma plane. Truncation silent —
  kernel only memcpys what fits in the CAPTURE plane.

Daemon (chardev_client.c):
- handle_req_decode allocates a response buffer sized for the
  full chardev payload, lets decoder fill the pixel area
  after the resp_frame struct, sends the full payload via the
  existing send_response.

Test client (tools/test_m2m_decode.c, new):
- Minimal V4L2 m2m client: S_FMT both queues, REQBUFS 1 each,
  mmap+fill OUTPUT, QBUF both, STREAMON, poll, DQBUF, dump
  CAPTURE planes to a raw NV12 file. ~250 LOC; verifies the
  whole flow without needing v4l2-ctl framing.

Roadmap update (docs/roadmap.md):
- Phase 8.4 retitled "daemon ↔ kernel decode round-trip"
  to reflect what actually shipped (vs. the original V4L2-
  ioctl-driven plan which moved here).
- Phase 8.5 retitled "full V4L2 m2m driver" with closure
  status.
- Phase 8.6 reshaped to two tracks: dmabuf + AV1/H.264/
  stateless controls + media controller. Adds the punch list
  of v4l2-compliance failures (DECODER_CMD, S_FMT colorspace)
  that 8.6 will fix.

Verification on hertz (Pi 5, 6.12.75+rpt-rpi-2712):

  Kernel + daemon build clean (-Wall -Wextra clean both sides).
  Test harness drives one VP9 keyframe end-to-end:
    OUTPUT REQBUFS -> 2
    CAPTURE REQBUFS -> 2
    QBUF OUTPUT[0] bytesused=1566
    QBUF CAPTURE[0]; STREAMON both
    poll revents=0x5
    DQBUF OUTPUT[0] flags=0x4001 (DONE)
    DQBUF CAPTURE[0] flags=0x4000 payloads=[12288, 6144]
    wrote 12288 Y + 6144 UV bytes to /tmp/out_m2m.nv12

  Pixel correctness vs reference:
    ffmpeg -i vp9_small.ivf -pix_fmt nv12 -f rawvideo -y ref.nv12
    cmp /tmp/out_m2m.nv12 /tmp/ref.nv12 → match ✓
  Byte-for-byte identical to FFmpeg's stock CPU decode.

  v4l2-compliance: detected as Stateless Decoder; most ioctls
  pass; two expected fails documented in closure doc
  (DECODER_CMD/media controller, S_FMT colorspace).

  Clean teardown: SIGTERM the daemon, rmmod the module, no
  oops/WARN in dmesg.

Per correctness-before-speed:
- Real V4L2 ioctl table (not stubs); uses v4l2-core helpers
  where they exist instead of reinventing.
- v4l2_m2m_buf_done_and_job_finish (not the manual sequence)
  to keep scheduler state consistent.
- Bit-exact reference comparison, not just "looks right."
- Documented every compliance failure with the planned fix.
- All resource paths (kmalloc/kfree, inflight list cleanup,
  src/dst buf removal in stop_streaming) handled on every
  error branch.

Phase 8.6 next: dmabuf-export for CAPTURE (removes 64 KiB
frame-size cap), add AV1+H.264 codecs, add V4L2 stateless
controls + media controller binding, fix the colorspace +
cookie-namespace compliance issues.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
2026-05-18 15:55:10 +00:00
parent 2a449632b9
commit 6f4b580f7c
10 changed files with 1465 additions and 121 deletions
+78 -3
View File
@@ -132,10 +132,67 @@ static int decoder_open_codec(struct daedalus_decoder *dec, uint32_t codec_id,
return 0;
}
/*
* Pack the decoded YUV planes into NV12 (Y followed by interleaved
* CbCr) in @out, advancing @out and tracking total bytes written
* in @used. Truncates silently at @cap. Returns 0 on success,
* -EINVAL if the source format isn't planar YUV 4:2:0.
*/
static int pack_nv12(struct AVFrame *fr, const AVPixFmtDescriptor *desc,
uint8_t *out, size_t cap, size_t *used)
{
int y, cw, ch, h, w;
size_t y_size, uv_size;
*used = 0;
if (!desc || !out || !cap)
return -EINVAL;
if (desc->nb_components < 3)
return -EINVAL;
w = fr->width;
h = fr->height;
cw = AV_CEIL_RSHIFT(w, desc->log2_chroma_w);
ch = AV_CEIL_RSHIFT(h, desc->log2_chroma_h);
y_size = (size_t) w * (size_t) h;
uv_size = (size_t) cw * (size_t) ch * 2u;
/* Y plane: w bytes per line stripped of stride padding. */
if (*used + y_size > cap)
return -EINVAL;
for (y = 0; y < h; y++)
memcpy(out + *used + (size_t) y * (size_t) w,
fr->data[0] + (size_t) y * (size_t) fr->linesize[0],
(size_t) w);
*used += y_size;
/* Interleave Cb and Cr into a single NV12 chroma plane. */
if (*used + uv_size > cap)
return -EINVAL;
{
uint8_t *uv = out + *used;
int x;
for (y = 0; y < ch; y++) {
const uint8_t *u = fr->data[1] +
(size_t) y * (size_t) fr->linesize[1];
const uint8_t *v = fr->data[2] +
(size_t) y * (size_t) fr->linesize[2];
for (x = 0; x < cw; x++) {
uv[(y * cw + x) * 2 + 0] = u[x];
uv[(y * cw + x) * 2 + 1] = v[x];
}
}
*used += uv_size;
}
return 0;
}
int daedalus_decoder_run_request(struct daedalus_decoder *dec,
const struct daedalus_req_decode *req,
const uint8_t *bitstream,
struct daedalus_resp_frame *resp)
struct daedalus_resp_frame *resp,
uint8_t *nv12_out, size_t nv12_cap,
size_t *nv12_used)
{
struct ffmpeg_loader *fm = dec->loader;
struct AVCodecContext *ctx = NULL;
@@ -143,6 +200,8 @@ int daedalus_decoder_run_request(struct daedalus_decoder *dec,
memset(resp, 0, sizeof(*resp));
resp->codec_id = req->codec_id;
if (nv12_used)
*nv12_used = 0;
rc = decoder_open_codec(dec, req->codec_id, &ctx);
if (rc == -ENOSYS) {
@@ -255,10 +314,26 @@ int daedalus_decoder_run_request(struct daedalus_decoder *dec,
resp->chroma_len = chroma_len;
resp->fnv1a_yuv = h;
log_info("decoder: OK %dx%d fmt=%d (%s) fnv1a=0x%08x luma=%u chroma=%u",
/*
* Pack pixels as NV12 into the caller's buffer if
* provided and big enough. Truncation is silent —
* the kernel will only copy as much as fits in the
* CAPTURE plane, so partial fills are fine for Phase
* 8.5. Phase 8.6 (dmabuf) eliminates the truncation.
*/
if (nv12_out && nv12_cap) {
int prc = pack_nv12(fr, desc, nv12_out, nv12_cap,
nv12_used);
if (prc < 0)
log_warn("decoder: NV12 pack failed (cap=%zu pix_fmt=%d) — kernel will see metadata only",
nv12_cap, fr->format);
}
log_info("decoder: OK %dx%d fmt=%d (%s) fnv1a=0x%08x luma=%u chroma=%u nv12=%zu",
fr->width, fr->height, fr->format,
desc ? desc->name : "?",
h, luma_len, chroma_len);
h, luma_len, chroma_len,
nv12_used ? *nv12_used : 0u);
}
fm->av_frame_unref(dec->frame);