On Tuesday 10 December 2013 16:02:39 Diego Biurrun wrote: > On Tue, Dec 10, 2013 at 03:42:28PM +0100, Guillaume Martres wrote: > > --- a/libavcodec/hevc_ps.c > > +++ b/libavcodec/hevc_ps.c > > @@ -328,12 +328,13 @@ int ff_hevc_decode_nal_vps(HEVCContext *s) > > > > GetBitContext *gb = &s->HEVClc.gb; > > int vps_id = 0; > > HEVCVPS *vps; > > > > + AVBufferRef *vps_buf = av_buffer_allocz(sizeof(*vps)); > > Shouldn't this be sizeof(*vps_buf)? No, we allocate an AVBufferRef which contains a member "data" with size "sizeof(*vps)", like the allocation for sps_buf. > > > - av_log(s->avctx, AV_LOG_DEBUG, "Decoding VPS\n"); > > + if (!vps_buf) > > + return AVERROR(ENOMEM); > > + vps = (HEVCVPS*)vps_buf->data; > > > > - vps = av_mallocz(sizeof(*vps)); > > - if (!vps) > > - return AVERROR(ENOMEM); > > + av_log(s->avctx, AV_LOG_DEBUG, "Decoding VPS\n"); > > Seems like it was copy-pasted from here. > > Moving the av_log to after VPS was decoded kind of changes the > semantics of the message. Whatever you prefer. This is consistent with what we do for the SPS.
_______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
