On 08/11/2014 10:35 AM, Diego Biurrun wrote:
> On Mon, Aug 11, 2014 at 07:36:48AM -0700, John Stebbins wrote:
>> --- a/libavcodec/mpeg12dec.c
>> +++ b/libavcodec/mpeg12dec.c
>> @@ -1097,18 +1097,16 @@ static av_cold int mpeg_decode_init(AVCodecContext
>> *avctx)
>> {
>> Mpeg1Context *s = avctx->priv_data;
>> MpegEncContext *s2 = &s->mpeg_enc_ctx;
>> - int i;
>> -
>> - /* we need some permutation to store matrices,
>> - * until MPV_common_init() sets the real permutation. */
>> - for (i = 0; i < 64; i++)
>> - s2->idsp.idct_permutation[i] = i;
>>
>> ff_MPV_decode_defaults(s2);
>>
>> s->mpeg_enc_ctx.avctx = avctx;
>> s->mpeg_enc_ctx.flags = avctx->flags;
>> s->mpeg_enc_ctx.flags2 = avctx->flags2;
>> +
>> + /* we need some permutation to store matrices,
>> + * until ff_mpv_idct_init() sets the real permutation. */
>> + ff_mpv_idct_init(s2);
>> ff_mpeg12_common_init(&s->mpeg_enc_ctx);
>> ff_mpeg12_init_vlcs();
> This bit looks suspicious; at least the comment does no longer match
> the code.The code is correct. I moved it down a few lines because ff_mpv_idct_init references s2->avctx. But if you don't like useing ff_mpv_idct_init for dummy initialization, I can certainly revert this and just update the comment. I guess the comment is a bit misleading now. It means the *next* call to ff_mpv_idct_init that will happen in mpeg_decode_postinit. How about simplifying to: /* we need some permutation to store matrices until the decoder sets the real permutation */ -- John GnuPG fingerprint: D0EC B3DB C372 D1F1 0B01 83F0 49F1 D7B2 60D4 D0F7
signature.asc
Description: OpenPGP digital signature
_______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
