| Message ID | tencent_1A5A746EF0F810A9AFC4A3D70231D4069009@qq.com |
|---|---|
| State | New |
| Headers | show |
| Series | [FFmpeg-devel] lavc: fix h264 parser bug | expand |
| Context | Check | Description |
|---|---|---|
| yinshiyou/make_loongarch64 | success | Make finished |
| yinshiyou/make_fate_loongarch64 | fail | Make fate failed |
| andriy/make_x86 | success | Make finished |
| andriy/make_fate_x86 | fail | Make fate failed |
On Fri, 13 May 2022 00:22:02 +0800 Lingjiang Fang <vacingfang@foxmail.com> wrote: ping for review padding zero after pps may lead to decode crash in same ipads btw, I am working on update fate cases to fit this bugfix > will give a wrong nalu length(+1) when next nalu start with 00000001 > this bug will lead to an padding zero to pps in AVC Sequence > header(flv) or in avc1(mpr) > > --- > > libavcodec/h2645_parse.c | 4 ++++ > 1 file changed, 4 insertions(+) > > diff --git a/libavcodec/h2645_parse.c b/libavcodec/h2645_parse.c > index 03780680c6..2e941a2bd0 100644 > --- a/libavcodec/h2645_parse.c > +++ b/libavcodec/h2645_parse.c > @@ -134,6 +134,10 @@ int ff_h2645_extract_rbsp(const uint8_t *src, > int length, dst[di++] = src[si++]; > > nsc: > + if (src[si-1] == 0) { > + --di; > + --si; > + } > memset(dst + di, 0, AV_INPUT_BUFFER_PADDING_SIZE); > > nal->data = dst; Regards, Lingjiang Fang
diff --git a/libavcodec/h2645_parse.c b/libavcodec/h2645_parse.c index 03780680c6..2e941a2bd0 100644 --- a/libavcodec/h2645_parse.c +++ b/libavcodec/h2645_parse.c @@ -134,6 +134,10 @@ int ff_h2645_extract_rbsp(const uint8_t *src, int length, dst[di++] = src[si++]; nsc: + if (src[si-1] == 0) { + --di; + --si; + } memset(dst + di, 0, AV_INPUT_BUFFER_PADDING_SIZE); nal->data = dst;