On Tue, May 09, 2017 at 12:29:55PM -0300, James Almer wrote: > The code was skipping the entire reported SEI message size regardless of > the amount of bits read. > While in theory safe for NALU where the picture timing SEI message is alone > or at the end as we're using the checked bitstream reader, it isn't in any > other situation, where every SEI message in the NALU after the picture > timing one would potentially fail to parse. > > Change the function name to one more in line with the rest of file, and > remove the bogus "Skipped SEI" debug message while at it. > > Signed-off-by: James Almer <jamr...@gmail.com> > --- > No test case, all the files i checked plus those in the FATE suite seem to > have one SEI message per NALU, or the Picture Timing SEI as the last one. > > libavcodec/hevc_sei.c | 15 +++++++-------- > 1 file changed, 7 insertions(+), 8 deletions(-)
I think the "skip to match the size" should be done in a single place for all SEI units and parsing should probably continue even if a SEI unit has a failure of some kind. the patch is fine too ... [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB The real ebay dictionary, page 1 "Used only once" - "Some unspecified defect prevented a second use" "In good condition" - "Can be repaird by experienced expert" "As is" - "You wouldnt want it even if you were payed for it, if you knew ..."
signature.asc
Description: Digital signature
_______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
