| Message ID | tencent_56A9004EB8255CB305C85EB3E8F36D15C805@qq.com |
|---|---|
| State | New |
| Headers | show |
| Series | [FFmpeg-devel] avcodec/h264dec: add option to ignore in band parameter set | expand |
| Context | Check | Description |
|---|---|---|
| andriy/make_x86 | success | Make finished |
| andriy/make_fate_x86 | success | Make fate finished |
| andriy/make_ppc | success | Make finished |
| andriy/make_fate_ppc | success | Make fate finished |
| andriy/make_aarch64_jetson | success | Make finished |
| andriy/make_fate_aarch64_jetson | success | Make fate finished |
| andriy/make_armv7_RPi4 | success | Make finished |
| andriy/make_fate_armv7_RPi4 | success | Make fate finished |
On Sun, Feb 6, 2022 at 3:58 PM Zhao Zhili <quinkblack@foxmail.com> wrote: > > It works in three mode: > 0: don't ignore in-band ps > 1: ignore in-band ps > -1: if corrupted data is detected, then ignore in-band ps afterwards > > h264dec working hard to do error resilience, it doesn't drop a > whole packet when error is detected in a nalu. Then there is a > higher chance for fake sps/pps be found and used. This happened > in a mp4 file. h264dec failed to recovery after broken data due > to the fake pps, while other H.264 decoders have no such problem. This seems extremely hacky. I think you need to elaborate on the problem you are trying to fix, so that any approach to a solution can be judged properly. - Hendrik
> On Feb 6, 2022, at 11:31 PM, Hendrik Leppkes <h.leppkes@gmail.com> wrote: > > On Sun, Feb 6, 2022 at 3:58 PM Zhao Zhili <quinkblack@foxmail.com> wrote: >> >> It works in three mode: >> 0: don't ignore in-band ps >> 1: ignore in-band ps >> -1: if corrupted data is detected, then ignore in-band ps afterwards >> >> h264dec working hard to do error resilience, it doesn't drop a >> whole packet when error is detected in a nalu. Then there is a >> higher chance for fake sps/pps be found and used. This happened >> in a mp4 file. h264dec failed to recovery after broken data due >> to the fake pps, while other H.264 decoders have no such problem. > > This seems extremely hacky. I think you need to elaborate on the > problem you are trying to fix, so that any approach to a solution can > be judged properly. There are two problems here: Firstly, after split a packet into multiple NALUs, if part of these NALUs have broken data, drop the whole packet or not. For H.264 with avcc bitstream, the split itself is questionable if part of these NALUs are invalid. Keep going after broken H264_NAL_SLICE has no serious effect, but updating SPS/PPS in this case can be dangerous. Secondly, for some containers like mp4, updating SPS/PPS are supported by spec but don’t get widely support among implementations. FFmpeg can generate mp4 with SPS/PPS updating in-band, but demux and decoding only works without seek. On the whole, it make sense to have an option do ignore in-band parameter set for those container. > > - Hendrik > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
diff --git a/libavcodec/h264dec.c b/libavcodec/h264dec.c index a47caa95e8..1ee2b68459 100644 --- a/libavcodec/h264dec.c +++ b/libavcodec/h264dec.c @@ -681,6 +681,10 @@ static int decode_nal_units(H264Context *h, const uint8_t *buf, int buf_size) break; case H264_NAL_SPS: { GetBitContext tmp_gb = nal->gb; + if (h->ps.sps && h->ignore_in_band_ps == 1) { + av_log(h, AV_LOG_WARNING, "ignore in-band sps\n"); + break; + } if (avctx->hwaccel && avctx->hwaccel->decode_params) { ret = avctx->hwaccel->decode_params(avctx, nal->type, @@ -700,6 +704,10 @@ static int decode_nal_units(H264Context *h, const uint8_t *buf, int buf_size) break; } case H264_NAL_PPS: + if (h->ps.pps && h->ignore_in_band_ps == 1) { + av_log(h, AV_LOG_WARNING, "ignore in-band pps\n"); + break; + } if (avctx->hwaccel && avctx->hwaccel->decode_params) { ret = avctx->hwaccel->decode_params(avctx, nal->type, @@ -726,6 +734,8 @@ static int decode_nal_units(H264Context *h, const uint8_t *buf, int buf_size) } if (err < 0) { + if (h->ignore_in_band_ps == -1) + h->ignore_in_band_ps = 1; av_log(h->avctx, AV_LOG_ERROR, "decode_slice_header error\n"); } } @@ -1011,8 +1021,12 @@ static int h264_decode_frame(AVCodecContext *avctx, void *data, } buf_index = decode_nal_units(h, buf, buf_size); - if (buf_index < 0) + if (buf_index < 0) { + /* If there is corrupted data, then don't trust the in-band ps. */ + if (h->ignore_in_band_ps == -1) + h->ignore_in_band_ps = 1; return AVERROR_INVALIDDATA; + } if (!h->cur_pic_ptr && h->nal_unit_type == H264_NAL_END_SEQUENCE) { av_assert0(buf_index <= buf_size); @@ -1055,6 +1069,7 @@ static const AVOption h264_options[] = { { "nal_length_size", "nal_length_size", OFFSET(nal_length_size), AV_OPT_TYPE_INT, {.i64 = 0}, 0, 4, VDX }, { "enable_er", "Enable error resilience on damaged frames (unsafe)", OFFSET(enable_er), AV_OPT_TYPE_BOOL, { .i64 = -1 }, -1, 1, VD }, { "x264_build", "Assume this x264 version if no x264 version found in any SEI", OFFSET(x264_build), AV_OPT_TYPE_INT, {.i64 = -1}, -1, INT_MAX, VD }, + { "ignore_in_band_ps", "ignore in-band parameter set", OFFSET(ignore_in_band_ps), AV_OPT_TYPE_INT, {.i64 = 0}, -1, 1, VD }, { NULL }, }; diff --git a/libavcodec/h264dec.h b/libavcodec/h264dec.h index 8168c8e97b..760540bc9d 100644 --- a/libavcodec/h264dec.h +++ b/libavcodec/h264dec.h @@ -448,6 +448,7 @@ typedef struct H264Context { int bit_depth_luma; ///< luma bit depth from sps to detect changes int chroma_format_idc; ///< chroma format from sps to detect changes + int ignore_in_band_ps; H264ParamSets ps; uint16_t *slice_table_base;