On Thu, Dec 13, 2012 at 07:46:00PM +0100, Anton Khirnov wrote: > --- a/libavcodec/ivi_common.c > +++ b/libavcodec/ivi_common.c > @@ -384,7 +384,8 @@ static int ivi_dec_tile_data_size(GetBitContext *gb) > * @param[in] tile pointer to the tile descriptor > * @return result code: 0 - OK, -1 = error (corrupted blocks data) > */ > -static int ivi_decode_blocks(GetBitContext *gb, IVIBandDesc *band, IVITile > *tile) > +static int ivi_decode_blocks(GetBitContext *gb, IVIBandDesc *band, IVITile > *tile, > + AVCodecContext *avctx)
The patch does LGTM. This is more of a general question than a review comment: Why always pass around these huge contexts instead of just the necessary AVClass? Of course it's just a pointer in either case, but often it feels like overkill... Diego _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
