[FFmpeg-devel] [PATCH] lavc/hevc_parser: add 4 bytes startcode condition in hevc_find_frame_end

[Fu, Linjie]

> -----Original Message-----
> From: ffmpeg-devel [mailto:ffmpeg-devel-bounces at ffmpeg.org] On Behalf
> Of Michael Niedermayer
> Sent: Friday, November 30, 2018 07:01
> To: FFmpeg development discussions and patches <ffmpeg-
> devel at ffmpeg.org>
> Subject: Re: [FFmpeg-devel] [PATCH] lavc/hevc_parser: add 4 bytes
> startcode condition in hevc_find_frame_end
> 
> On Thu, Nov 29, 2018 at 04:14:29AM +0000, Fu, Linjie wrote:
> > > -----Original Message-----
> > > From: ffmpeg-devel [mailto:ffmpeg-devel-bounces at ffmpeg.org] On
> Behalf
> > > Of Michael Niedermayer
> > > Sent: Thursday, November 29, 2018 02:14
> > > To: FFmpeg development discussions and patches <ffmpeg-
> > > devel at ffmpeg.org>
> > > Subject: Re: [FFmpeg-devel] [PATCH] lavc/hevc_parser: add 4 bytes
> > > startcode condition in hevc_find_frame_end
> > >
> > > On Tue, Nov 27, 2018 at 08:16:55PM +0800, Linjie Fu wrote:
> > > > The startcode before VPS,SPS,PPS and the first NALU in an AU is 4 bytes.
> > > > Blindly taking the startcode as 3 bytes will leave 0x00 in last packet
> > > > and may lead to some warnings in parse_nal_units when s->flags is set
> to
> > > > PARSER_FLAG_COMPLETE_FRAMES.
> > > >
> > > > Add 4 bytes startcode condition in hevc_find_frame_end.
> > > > Modify the code to print the buf_size like in H264 and reduce the
> > > duplication.
> > > >
> > > > Signed-off-by: Linjie Fu <linjie.fu at intel.com>
> > > > ---
> > > >  libavcodec/hevc_parser.c | 15 ++++++++++-----
> > > >  1 file changed, 10 insertions(+), 5 deletions(-)
> > >
> > > breaks
> > >  make fate-hevc-bsf-mp4toannexb
> > > TEST    hevc-bsf-mp4toannexb
> > > --- -	2018-11-28 19:12:47.869732022 +0100
> > > +++ tests/data/fate/hevc-bsf-mp4toannexb	2018-11-28
> > > 19:12:47.864276885 +0100
> > > @@ -1 +1 @@
> > > -1873662a3af1848c37e4eb25722c8df9
> > > +73019329ed7f81c24f9af67c34c640c0
> > > Test hevc-bsf-mp4toannexb failed. Look at tests/data/fate/hevc-bsf-
> > > mp4toannexb.err for details.
> > > make: *** [fate-hevc-bsf-mp4toannexb] Error 1
> > >
> >
> > Investigate the root cause:
> >
> > In fate test:
> > Step 1. ffmpeg -i hevc-conformance/WPP_A_ericsson_MAIN10_2.bit -c
> copy -flags +bitexact hevc-mp4.mov
> > Step 2. ffmpeg -i hevc-mp4_after.mov -c:v copy -fflags +bitexact -f hevc
> hevc.out
> > Step3. md5sum and compared with the reference
> >
> > Compare the output file in Step 1 before and after the patch, and found it
> changed:
> > -rw-r--r-- 1 root     root     68460 11月 29 11:23 hevc-mp4_after.mov
> > -rw-r--r-- 1 root     root     68507 11月 29 11:31 hevc-mp4_before.mov
> >
> > Compare the original file with two output files:
> > SUFFIX_SEI bit information in  the first frame for example :
> >
> > WPP_A_ericsson_MAIN10_2.bit: { 50 01 84 31 00 19 39 77 d0 7b 3f bd 1e 09
> 38 9a 79 41 c0 16 11 da 18 aa 0a db 2b 19 fa 47 3f 0f 67 4a 91 9c a1 12 72 67 d6 f0
> 8f 66 ee 95 f9 e2 b9 ba 23 9a e8 80 }
> >
> > hevc-mp4_before.mov: { 50 01 84 31 00 19 39 77 d0 7b 3f bd 1e 09 38 9a 79
> 41 c0 16 11 da 18 aa 0a db 2b 19 fa 47 3f 0f 67 4a 91 9c a1 12 72 67 d6 f0 8f 66 ee
> 95 f9 e2 b9 ba 23 9a e8 80 00 }
> >
> > hevc-mp4_after.mov: { 50 01 84 31 00 19 39 77 d0 7b 3f bd 1e 09 38 9a 79 41
> c0 16 11 da 18 aa 0a db 2b 19 fa 47 3f 0f 67 4a 91 9c a1 12 72 67 d6 f0 8f 66 ee 95
> f9 e2 b9 ba 23 9a e8 80 }
> >
> > Before applying the patch, the output file has an extra 0x00 at the end of
> SUFFIX_SEI (because of blindly taking startcode as 3 bytes and left one byte).
> > After applying the patch, the output file has the same bit info like the
> original file.
> > So, I think this patch works.
> 
> if the patch break make fate then it doesnt work.
> If a change cause by a patch is correct it needs to update the
> test / its checksum
> This is also important so that anyone looking at the patch knows which tests
> change and anyone testing knows how to detect a test failure vs a passed
> test

Thanks, sent a [v2] patch with the updated FATE checksum.
---
Linjie

> [...]
> --
> Michael     GnuPG fingerprint:
> 9FF2128B147EF6730BADF133611EC787040B0FAB
> 
> Those who are best at talking, realize last or never when they are wrong.