Hi Luca, On Sun, Jan 28, 2018 at 4:59 AM, Luca Barbato <[email protected]> wrote: > On 28/01/2018 03:24, Sean McGovern wrote: >> >> Bug-Id: 1101 >> Cc: [email protected] >> --- >> libavcodec/vc1_mc.c | 5 +++++ >> 1 file changed, 5 insertions(+) >> >> diff --git a/libavcodec/vc1_mc.c b/libavcodec/vc1_mc.c >> index f4632d6..fb5ca75 100644 >> --- a/libavcodec/vc1_mc.c >> +++ b/libavcodec/vc1_mc.c >> @@ -713,6 +713,11 @@ void ff_vc1_mc_4mv_chroma4(VC1Context *v, int dir, >> int dir2, int avg) >> lutuv = v->next_lutuv; >> use_ic = v->next_use_ic; >> } else { >> + if (!s->last_picture.f->data[1]) { >> + av_log(s->avctx, AV_LOG_WARNING, "Bad data in last picture >> frame.\n"); >> + return; >> + } >> + >> srcU = s->last_picture.f->data[1] + uvsrc_y * s->uvlinesize >> + uvsrc_x; >> srcV = s->last_picture.f->data[2] + uvsrc_y * s->uvlinesize >> + uvsrc_x; >> lutuv = v->last_lutuv; >> > > You should check for s->last_picture.f->data[0] outside the loop.
OK, will fix and post shortly. > > Look at ff_vc1_mc_4mv_chroma(), probably the same check does apply. It's already effectively done there, see around line 595. > > lu > _______________________________________________ > libav-devel mailing list > [email protected] > https://lists.libav.org/mailman/listinfo/libav-devel _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
