falkTX
/
FFmpeg
mirror of https://github.com/falkTX/FFmpeg.git
1
0
Fork 0
Code Issues 0 Releases 338 Wiki Activity
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
28447 Commits
32 Branches
283MB
Tree: dd561441b1
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'dd561441b1'
${ noResults }
FFmpeg/libavcodec/h264_loopfilter.c

653 lines
30KB
Raw Blame History 

  1. /*

  2.  * H.26L/H.264/AVC/JVT/14496-10/... loop filter

  3.  * Copyright (c) 2003 Michael Niedermayer <michaelni@gmx.at>

  4.  *

  5.  * This file is part of Libav.

  6.  *

  7.  * Libav is free software; you can redistribute it and/or

  8.  * modify it under the terms of the GNU Lesser General Public

  9.  * License as published by the Free Software Foundation; either

  10.  * version 2.1 of the License, or (at your option) any later version.

  11.  *

  12.  * Libav is distributed in the hope that it will be useful,

  13.  * but WITHOUT ANY WARRANTY; without even the implied warranty of

  14.  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU

  15.  * Lesser General Public License for more details.

  16.  *

  17.  * You should have received a copy of the GNU Lesser General Public

  18.  * License along with Libav; if not, write to the Free Software

  19.  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA

  20.  */

  21. 

  22. /**

  23.  * @file

  24.  * H.264 / AVC / MPEG4 part10 loop filter.

  25.  * @author Michael Niedermayer <michaelni@gmx.at>

  26.  */

  27. 

  28. #include "libavutil/intreadwrite.h"

  29. #include "internal.h"

  30. #include "dsputil.h"

  31. #include "avcodec.h"

  32. #include "mpegvideo.h"

  33. #include "h264.h"

  34. #include "mathops.h"

  35. #include "rectangle.h"

  36. 

  37. //#undef NDEBUG

  38. #include <assert.h>

  39. 

  40. /* Deblocking filter (p153) */

  41. static const uint8_t alpha_table[52*3] = {

  42.      0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,

  43.      0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,

  44.      0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,

  45.      0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,

  46.      0,  0,  0,  0,  0,  0,  0,  0,  0,  0,

  47.      0,  0,  0,  0,  0,  0,  4,  4,  5,  6,

  48.      7,  8,  9, 10, 12, 13, 15, 17, 20, 22,

  49.     25, 28, 32, 36, 40, 45, 50, 56, 63, 71,

  50.     80, 90,101,113,127,144,162,182,203,226,

  51.    255,255,

  52.    255,255,255,255,255,255,255,255,255,255,255,255,255,

  53.    255,255,255,255,255,255,255,255,255,255,255,255,255,

  54.    255,255,255,255,255,255,255,255,255,255,255,255,255,

  55.    255,255,255,255,255,255,255,255,255,255,255,255,255,

  56. };

  57. static const uint8_t beta_table[52*3] = {

  58.      0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,

  59.      0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,

  60.      0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,

  61.      0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,

  62.      0,  0,  0,  0,  0,  0,  0,  0,  0,  0,

  63.      0,  0,  0,  0,  0,  0,  2,  2,  2,  3,

  64.      3,  3,  3,  4,  4,  4,  6,  6,  7,  7,

  65.      8,  8,  9,  9, 10, 10, 11, 11, 12, 12,

  66.     13, 13, 14, 14, 15, 15, 16, 16, 17, 17,

  67.     18, 18,

  68.     18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,

  69.     18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,

  70.     18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,

  71.     18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,

  72. };

  73. static const uint8_t tc0_table[52*3][4] = {

  74.     {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 },

  75.     {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 },

  76.     {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 },

  77.     {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 },

  78.     {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 },

  79.     {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 },

  80.     {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 },

  81.     {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 },

  82.     {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 },

  83.     {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 },

  84.     {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 },

  85.     {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 0 }, {-1, 0, 0, 1 },

  86.     {-1, 0, 0, 1 }, {-1, 0, 0, 1 }, {-1, 0, 0, 1 }, {-1, 0, 1, 1 }, {-1, 0, 1, 1 }, {-1, 1, 1, 1 },

  87.     {-1, 1, 1, 1 }, {-1, 1, 1, 1 }, {-1, 1, 1, 1 }, {-1, 1, 1, 2 }, {-1, 1, 1, 2 }, {-1, 1, 1, 2 },

  88.     {-1, 1, 1, 2 }, {-1, 1, 2, 3 }, {-1, 1, 2, 3 }, {-1, 2, 2, 3 }, {-1, 2, 2, 4 }, {-1, 2, 3, 4 },

  89.     {-1, 2, 3, 4 }, {-1, 3, 3, 5 }, {-1, 3, 4, 6 }, {-1, 3, 4, 6 }, {-1, 4, 5, 7 }, {-1, 4, 5, 8 },

  90.     {-1, 4, 6, 9 }, {-1, 5, 7,10 }, {-1, 6, 8,11 }, {-1, 6, 8,13 }, {-1, 7,10,14 }, {-1, 8,11,16 },

  91.     {-1, 9,12,18 }, {-1,10,13,20 }, {-1,11,15,23 }, {-1,13,17,25 },

  92.     {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 },

  93.     {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 },

  94.     {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 },

  95.     {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 },

  96.     {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 },

  97.     {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 },

  98.     {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 },

  99.     {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 },

  100.     {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 }, {-1,13,17,25 },

  101. };

  102. 

  103. static void av_always_inline filter_mb_edgev( uint8_t *pix, int stride, int16_t bS[4], unsigned int qp, H264Context *h) {

  104.     const unsigned int index_a = qp + h->slice_alpha_c0_offset;

  105.     const int alpha = alpha_table[index_a];

  106.     const int beta  = beta_table[qp + h->slice_beta_offset];

  107.     if (alpha ==0 || beta == 0) return;

  108. 

  109.     if( bS[0] < 4 ) {

  110.         int8_t tc[4];

  111.         tc[0] = tc0_table[index_a][bS[0]];

  112.         tc[1] = tc0_table[index_a][bS[1]];

  113.         tc[2] = tc0_table[index_a][bS[2]];

  114.         tc[3] = tc0_table[index_a][bS[3]];

  115.         h->h264dsp.h264_h_loop_filter_luma(pix, stride, alpha, beta, tc);

  116.     } else {

  117.         h->h264dsp.h264_h_loop_filter_luma_intra(pix, stride, alpha, beta);

  118.     }

  119. }

  120. static void av_always_inline filter_mb_edgecv( uint8_t *pix, int stride, int16_t bS[4], unsigned int qp, H264Context *h ) {

  121.     const unsigned int index_a = qp + h->slice_alpha_c0_offset;

  122.     const int alpha = alpha_table[index_a];

  123.     const int beta  = beta_table[qp + h->slice_beta_offset];

  124.     if (alpha ==0 || beta == 0) return;

  125. 

  126.     if( bS[0] < 4 ) {

  127.         int8_t tc[4];

  128.         tc[0] = tc0_table[index_a][bS[0]]+1;

  129.         tc[1] = tc0_table[index_a][bS[1]]+1;

  130.         tc[2] = tc0_table[index_a][bS[2]]+1;

  131.         tc[3] = tc0_table[index_a][bS[3]]+1;

  132.         h->h264dsp.h264_h_loop_filter_chroma(pix, stride, alpha, beta, tc);

  133.     } else {

  134.         h->h264dsp.h264_h_loop_filter_chroma_intra(pix, stride, alpha, beta);

  135.     }

  136. }

  137. 

  138. static void filter_mb_mbaff_edgev( H264Context *h, uint8_t *pix, int stride, int16_t bS[7], int bsi, int qp ) {

  139.     int index_a = qp + h->slice_alpha_c0_offset;

  140.     int alpha = alpha_table[index_a];

  141.     int beta  = beta_table[qp + h->slice_beta_offset];

  142.     if (alpha ==0 || beta == 0) return;

  143. 

  144.     if( bS[0] < 4 ) {

  145.         int8_t tc[4];

  146.         tc[0] = tc0_table[index_a][bS[0*bsi]];

  147.         tc[1] = tc0_table[index_a][bS[1*bsi]];

  148.         tc[2] = tc0_table[index_a][bS[2*bsi]];

  149.         tc[3] = tc0_table[index_a][bS[3*bsi]];

  150.         h->h264dsp.h264_h_loop_filter_luma_mbaff(pix, stride, alpha, beta, tc);

  151.     } else {

  152.         h->h264dsp.h264_h_loop_filter_luma_mbaff_intra(pix, stride, alpha, beta);

  153.     }

  154. }

  155. static void filter_mb_mbaff_edgecv( H264Context *h, uint8_t *pix, int stride, int16_t bS[7], int bsi, int qp ) {

  156.     int index_a = qp + h->slice_alpha_c0_offset;

  157.     int alpha = alpha_table[index_a];

  158.     int beta  = beta_table[qp + h->slice_beta_offset];

  159.     if (alpha ==0 || beta == 0) return;

  160. 

  161.     if( bS[0] < 4 ) {

  162.         int8_t tc[4];

  163.         tc[0] = tc0_table[index_a][bS[0*bsi]] + 1;

  164.         tc[1] = tc0_table[index_a][bS[1*bsi]] + 1;

  165.         tc[2] = tc0_table[index_a][bS[2*bsi]] + 1;

  166.         tc[3] = tc0_table[index_a][bS[3*bsi]] + 1;

  167.         h->h264dsp.h264_h_loop_filter_chroma_mbaff(pix, stride, alpha, beta, tc);

  168.     } else {

  169.         h->h264dsp.h264_h_loop_filter_chroma_mbaff_intra(pix, stride, alpha, beta);

  170.     }

  171. }

  172. 

  173. static void av_always_inline filter_mb_edgeh( uint8_t *pix, int stride, int16_t bS[4], unsigned int qp, H264Context *h ) {

  174.     const unsigned int index_a = qp + h->slice_alpha_c0_offset;

  175.     const int alpha = alpha_table[index_a];

  176.     const int beta  = beta_table[qp + h->slice_beta_offset];

  177.     if (alpha ==0 || beta == 0) return;

  178. 

  179.     if( bS[0] < 4 ) {

  180.         int8_t tc[4];

  181.         tc[0] = tc0_table[index_a][bS[0]];

  182.         tc[1] = tc0_table[index_a][bS[1]];

  183.         tc[2] = tc0_table[index_a][bS[2]];

  184.         tc[3] = tc0_table[index_a][bS[3]];

  185.         h->h264dsp.h264_v_loop_filter_luma(pix, stride, alpha, beta, tc);

  186.     } else {

  187.         h->h264dsp.h264_v_loop_filter_luma_intra(pix, stride, alpha, beta);

  188.     }

  189. }

  190. 

  191. static void av_always_inline filter_mb_edgech( uint8_t *pix, int stride, int16_t bS[4], unsigned int qp, H264Context *h ) {

  192.     const unsigned int index_a = qp + h->slice_alpha_c0_offset;

  193.     const int alpha = alpha_table[index_a];

  194.     const int beta  = beta_table[qp + h->slice_beta_offset];

  195.     if (alpha ==0 || beta == 0) return;

  196. 

  197.     if( bS[0] < 4 ) {

  198.         int8_t tc[4];

  199.         tc[0] = tc0_table[index_a][bS[0]]+1;

  200.         tc[1] = tc0_table[index_a][bS[1]]+1;

  201.         tc[2] = tc0_table[index_a][bS[2]]+1;

  202.         tc[3] = tc0_table[index_a][bS[3]]+1;

  203.         h->h264dsp.h264_v_loop_filter_chroma(pix, stride, alpha, beta, tc);

  204.     } else {

  205.         h->h264dsp.h264_v_loop_filter_chroma_intra(pix, stride, alpha, beta);

  206.     }

  207. }

  208. 

  209. void ff_h264_filter_mb_fast( H264Context *h, int mb_x, int mb_y, uint8_t *img_y, uint8_t *img_cb, uint8_t *img_cr, unsigned int linesize, unsigned int uvlinesize) {

  210.     MpegEncContext * const s = &h->s;

  211.     int mb_xy;

  212.     int mb_type, left_type;

  213.     int qp, qp0, qp1, qpc, qpc0, qpc1, qp_thresh;

  214. 

  215.     mb_xy = h->mb_xy;

  216. 

  217.     if(!h->top_type || !h->h264dsp.h264_loop_filter_strength || h->pps.chroma_qp_diff) {

  218.         ff_h264_filter_mb(h, mb_x, mb_y, img_y, img_cb, img_cr, linesize, uvlinesize);

  219.         return;

  220.     }

  221.     assert(!FRAME_MBAFF);

  222.     left_type= h->left_type[0];

  223. 

  224.     mb_type = s->current_picture.mb_type[mb_xy];

  225.     qp = s->current_picture.qscale_table[mb_xy];

  226.     qp0 = s->current_picture.qscale_table[mb_xy-1];

  227.     qp1 = s->current_picture.qscale_table[h->top_mb_xy];

  228.     qpc = get_chroma_qp( h, 0, qp );

  229.     qpc0 = get_chroma_qp( h, 0, qp0 );

  230.     qpc1 = get_chroma_qp( h, 0, qp1 );

  231.     qp0 = (qp + qp0 + 1) >> 1;

  232.     qp1 = (qp + qp1 + 1) >> 1;

  233.     qpc0 = (qpc + qpc0 + 1) >> 1;

  234.     qpc1 = (qpc + qpc1 + 1) >> 1;

  235.     qp_thresh = 15+52 - h->slice_alpha_c0_offset;

  236.     if(qp <= qp_thresh && qp0 <= qp_thresh && qp1 <= qp_thresh &&

  237.        qpc <= qp_thresh && qpc0 <= qp_thresh && qpc1 <= qp_thresh)

  238.         return;

  239. 

  240.     if( IS_INTRA(mb_type) ) {

  241.         int16_t bS4[4] = {4,4,4,4};

  242.         int16_t bS3[4] = {3,3,3,3};

  243.         int16_t *bSH = FIELD_PICTURE ? bS3 : bS4;

  244.         if(left_type)

  245.             filter_mb_edgev( &img_y[4*0], linesize, bS4, qp0, h);

  246.         if( IS_8x8DCT(mb_type) ) {

  247.             filter_mb_edgev( &img_y[4*2], linesize, bS3, qp, h);

  248.             filter_mb_edgeh( &img_y[4*0*linesize], linesize, bSH, qp1, h);

  249.             filter_mb_edgeh( &img_y[4*2*linesize], linesize, bS3, qp, h);

  250.         } else {

  251.             filter_mb_edgev( &img_y[4*1], linesize, bS3, qp, h);

  252.             filter_mb_edgev( &img_y[4*2], linesize, bS3, qp, h);

  253.             filter_mb_edgev( &img_y[4*3], linesize, bS3, qp, h);

  254.             filter_mb_edgeh( &img_y[4*0*linesize], linesize, bSH, qp1, h);

  255.             filter_mb_edgeh( &img_y[4*1*linesize], linesize, bS3, qp, h);

  256.             filter_mb_edgeh( &img_y[4*2*linesize], linesize, bS3, qp, h);

  257.             filter_mb_edgeh( &img_y[4*3*linesize], linesize, bS3, qp, h);

  258.         }

  259.         if(left_type){

  260.             filter_mb_edgecv( &img_cb[2*0], uvlinesize, bS4, qpc0, h);

  261.             filter_mb_edgecv( &img_cr[2*0], uvlinesize, bS4, qpc0, h);

  262.         }

  263.         filter_mb_edgecv( &img_cb[2*2], uvlinesize, bS3, qpc, h);

  264.         filter_mb_edgecv( &img_cr[2*2], uvlinesize, bS3, qpc, h);

  265.         filter_mb_edgech( &img_cb[2*0*uvlinesize], uvlinesize, bSH, qpc1, h);

  266.         filter_mb_edgech( &img_cb[2*2*uvlinesize], uvlinesize, bS3, qpc, h);

  267.         filter_mb_edgech( &img_cr[2*0*uvlinesize], uvlinesize, bSH, qpc1, h);

  268.         filter_mb_edgech( &img_cr[2*2*uvlinesize], uvlinesize, bS3, qpc, h);

  269.         return;

  270.     } else {

  271.         LOCAL_ALIGNED_8(int16_t, bS, [2], [4][4]);

  272.         int edges;

  273.         if( IS_8x8DCT(mb_type) && (h->cbp&7) == 7 ) {

  274.             edges = 4;

  275.             AV_WN64A(bS[0][0], 0x0002000200020002ULL);

  276.             AV_WN64A(bS[0][2], 0x0002000200020002ULL);

  277.             AV_WN64A(bS[1][0], 0x0002000200020002ULL);

  278.             AV_WN64A(bS[1][2], 0x0002000200020002ULL);

  279.         } else {

  280.             int mask_edge1 = (3*(((5*mb_type)>>5)&1)) | (mb_type>>4); //(mb_type & (MB_TYPE_16x16 | MB_TYPE_8x16)) ? 3 : (mb_type & MB_TYPE_16x8) ? 1 : 0;

  281.             int mask_edge0 = 3*((mask_edge1>>1) & ((5*left_type)>>5)&1); // (mb_type & (MB_TYPE_16x16 | MB_TYPE_8x16)) && (h->left_type[0] & (MB_TYPE_16x16 | MB_TYPE_8x16)) ? 3 : 0;

  282.             int step =  1+(mb_type>>24); //IS_8x8DCT(mb_type) ? 2 : 1;

  283.             edges = 4 - 3*((mb_type>>3) & !(h->cbp & 15)); //(mb_type & MB_TYPE_16x16) && !(h->cbp & 15) ? 1 : 4;

  284.             h->h264dsp.h264_loop_filter_strength( bS, h->non_zero_count_cache, h->ref_cache, h->mv_cache,

  285.                                               h->list_count==2, edges, step, mask_edge0, mask_edge1, FIELD_PICTURE);

  286.         }

  287.         if( IS_INTRA(left_type) )

  288.             AV_WN64A(bS[0][0], 0x0004000400040004ULL);

  289.         if( IS_INTRA(h->top_type) )

  290.             AV_WN64A(bS[1][0], FIELD_PICTURE ? 0x0003000300030003ULL : 0x0004000400040004ULL);

  291. 

  292. #define FILTER(hv,dir,edge)\

  293.         if(AV_RN64A(bS[dir][edge])) {                                   \

  294.             filter_mb_edge##hv( &img_y[4*edge*(dir?linesize:1)], linesize, bS[dir][edge], edge ? qp : qp##dir, h );\

  295.             if(!(edge&1)) {\

  296.                 filter_mb_edgec##hv( &img_cb[2*edge*(dir?uvlinesize:1)], uvlinesize, bS[dir][edge], edge ? qpc : qpc##dir, h );\

  297.                 filter_mb_edgec##hv( &img_cr[2*edge*(dir?uvlinesize:1)], uvlinesize, bS[dir][edge], edge ? qpc : qpc##dir, h );\

  298.             }\

  299.         }

  300.         if(left_type)

  301.             FILTER(v,0,0);

  302.         if( edges == 1 ) {

  303.             FILTER(h,1,0);

  304.         } else if( IS_8x8DCT(mb_type) ) {

  305.             FILTER(v,0,2);

  306.             FILTER(h,1,0);

  307.             FILTER(h,1,2);

  308.         } else {

  309.             FILTER(v,0,1);

  310.             FILTER(v,0,2);

  311.             FILTER(v,0,3);

  312.             FILTER(h,1,0);

  313.             FILTER(h,1,1);

  314.             FILTER(h,1,2);

  315.             FILTER(h,1,3);

  316.         }

  317. #undef FILTER

  318.     }

  319. }

  320. 

  321. static int check_mv(H264Context *h, long b_idx, long bn_idx, int mvy_limit){

  322.     int v;

  323. 

  324.     v= h->ref_cache[0][b_idx] != h->ref_cache[0][bn_idx];

  325.     if(!v && h->ref_cache[0][b_idx]!=-1)

  326.         v= h->mv_cache[0][b_idx][0] - h->mv_cache[0][bn_idx][0] + 3 >= 7U |

  327.            FFABS( h->mv_cache[0][b_idx][1] - h->mv_cache[0][bn_idx][1] ) >= mvy_limit;

  328. 

  329.     if(h->list_count==2){

  330.         if(!v)

  331.             v = h->ref_cache[1][b_idx] != h->ref_cache[1][bn_idx] |

  332.                 h->mv_cache[1][b_idx][0] - h->mv_cache[1][bn_idx][0] + 3 >= 7U |

  333.                 FFABS( h->mv_cache[1][b_idx][1] - h->mv_cache[1][bn_idx][1] ) >= mvy_limit;

  334. 

  335.         if(v){

  336.             if(h->ref_cache[0][b_idx] != h->ref_cache[1][bn_idx] |

  337.                h->ref_cache[1][b_idx] != h->ref_cache[0][bn_idx])

  338.                 return 1;

  339.             return

  340.                 h->mv_cache[0][b_idx][0] - h->mv_cache[1][bn_idx][0] + 3 >= 7U |

  341.                 FFABS( h->mv_cache[0][b_idx][1] - h->mv_cache[1][bn_idx][1] ) >= mvy_limit |

  342.                 h->mv_cache[1][b_idx][0] - h->mv_cache[0][bn_idx][0] + 3 >= 7U |

  343.                 FFABS( h->mv_cache[1][b_idx][1] - h->mv_cache[0][bn_idx][1] ) >= mvy_limit;

  344.         }

  345.     }

  346. 

  347.     return v;

  348. }

  349. 

  350. static av_always_inline void filter_mb_dir(H264Context *h, int mb_x, int mb_y, uint8_t *img_y, uint8_t *img_cb, uint8_t *img_cr, unsigned int linesize, unsigned int uvlinesize, int mb_xy, int mb_type, int mvy_limit, int first_vertical_edge_done, int dir) {

  351.     MpegEncContext * const s = &h->s;

  352.     int edge;

  353.     const int mbm_xy = dir == 0 ? mb_xy -1 : h->top_mb_xy;

  354.     const int mbm_type = dir == 0 ? h->left_type[0] : h->top_type;

  355. 

  356.     // how often to recheck mv-based bS when iterating between edges

  357.     static const uint8_t mask_edge_tab[2][8]={{0,3,3,3,1,1,1,1},

  358.                                               {0,3,1,1,3,3,3,3}};

  359.     const int mask_edge = mask_edge_tab[dir][(mb_type>>3)&7];

  360.     const int edges = mask_edge== 3 && !(h->cbp&15) ? 1 : 4;

  361. 

  362.     // how often to recheck mv-based bS when iterating along each edge

  363.     const int mask_par0 = mb_type & (MB_TYPE_16x16 | (MB_TYPE_8x16 >> dir));

  364. 

  365.     if(mbm_type && !first_vertical_edge_done){

  366. 

  367.         if (FRAME_MBAFF && (dir == 1) && ((mb_y&1) == 0)

  368.             && IS_INTERLACED(mbm_type&~mb_type)

  369.             ) {

  370.             // This is a special case in the norm where the filtering must

  371.             // be done twice (one each of the field) even if we are in a

  372.             // frame macroblock.

  373.             //

  374.             unsigned int tmp_linesize   = 2 *   linesize;

  375.             unsigned int tmp_uvlinesize = 2 * uvlinesize;

  376.             int mbn_xy = mb_xy - 2 * s->mb_stride;

  377.             int j;

  378. 

  379.             for(j=0; j<2; j++, mbn_xy += s->mb_stride){

  380.                 DECLARE_ALIGNED(8, int16_t, bS)[4];

  381.                 int qp;

  382.                 if( IS_INTRA(mb_type|s->current_picture.mb_type[mbn_xy]) ) {

  383.                     AV_WN64A(bS, 0x0003000300030003ULL);

  384.                 } else {

  385.                     if(!CABAC && IS_8x8DCT(s->current_picture.mb_type[mbn_xy])){

  386.                         bS[0]= 1+((h->cbp_table[mbn_xy] & 4)||h->non_zero_count_cache[scan8[0]+0]);

  387.                         bS[1]= 1+((h->cbp_table[mbn_xy] & 4)||h->non_zero_count_cache[scan8[0]+1]);

  388.                         bS[2]= 1+((h->cbp_table[mbn_xy] & 8)||h->non_zero_count_cache[scan8[0]+2]);

  389.                         bS[3]= 1+((h->cbp_table[mbn_xy] & 8)||h->non_zero_count_cache[scan8[0]+3]);

  390.                     }else{

  391.                     const uint8_t *mbn_nnz = h->non_zero_count[mbn_xy] + 4+3*8;

  392.                     int i;

  393.                     for( i = 0; i < 4; i++ ) {

  394.                         bS[i] = 1 + !!(h->non_zero_count_cache[scan8[0]+i] | mbn_nnz[i]);

  395.                     }

  396.                     }

  397.                 }

  398.                 // Do not use s->qscale as luma quantizer because it has not the same

  399.                 // value in IPCM macroblocks.

  400.                 qp = ( s->current_picture.qscale_table[mb_xy] + s->current_picture.qscale_table[mbn_xy] + 1 ) >> 1;

  401.                 tprintf(s->avctx, "filter mb:%d/%d dir:%d edge:%d, QPy:%d ls:%d uvls:%d", mb_x, mb_y, dir, edge, qp, tmp_linesize, tmp_uvlinesize);

  402.                 { int i; for (i = 0; i < 4; i++) tprintf(s->avctx, " bS[%d]:%d", i, bS[i]); tprintf(s->avctx, "\n"); }

  403.                 filter_mb_edgeh( &img_y[j*linesize], tmp_linesize, bS, qp, h );

  404.                 filter_mb_edgech( &img_cb[j*uvlinesize], tmp_uvlinesize, bS,

  405.                                 ( h->chroma_qp[0] + get_chroma_qp( h, 0, s->current_picture.qscale_table[mbn_xy] ) + 1 ) >> 1, h);

  406.                 filter_mb_edgech( &img_cr[j*uvlinesize], tmp_uvlinesize, bS,

  407.                                 ( h->chroma_qp[1] + get_chroma_qp( h, 1, s->current_picture.qscale_table[mbn_xy] ) + 1 ) >> 1, h);

  408.             }

  409.         }else{

  410.             DECLARE_ALIGNED(8, int16_t, bS)[4];

  411.             int qp;

  412. 

  413.             if( IS_INTRA(mb_type|mbm_type)) {

  414.                 AV_WN64A(bS, 0x0003000300030003ULL);

  415.                 if (   (!IS_INTERLACED(mb_type|mbm_type))

  416.                     || ((FRAME_MBAFF || (s->picture_structure != PICT_FRAME)) && (dir == 0))

  417.                 )

  418.                     AV_WN64A(bS, 0x0004000400040004ULL);

  419.             } else {

  420.                 int i;

  421.                 int mv_done;

  422. 

  423.                 if( dir && FRAME_MBAFF && IS_INTERLACED(mb_type ^ mbm_type)) {

  424.                     AV_WN64A(bS, 0x0001000100010001ULL);

  425.                     mv_done = 1;

  426.                 }

  427.                 else if( mask_par0 && ((mbm_type & (MB_TYPE_16x16 | (MB_TYPE_8x16 >> dir)))) ) {

  428.                     int b_idx= 8 + 4;

  429.                     int bn_idx= b_idx - (dir ? 8:1);

  430. 

  431.                     bS[0] = bS[1] = bS[2] = bS[3] = check_mv(h, 8 + 4, bn_idx, mvy_limit);

  432.                     mv_done = 1;

  433.                 }

  434.                 else

  435.                     mv_done = 0;

  436. 

  437.                 for( i = 0; i < 4; i++ ) {

  438.                     int x = dir == 0 ? 0 : i;

  439.                     int y = dir == 0 ? i    : 0;

  440.                     int b_idx= 8 + 4 + x + 8*y;

  441.                     int bn_idx= b_idx - (dir ? 8:1);

  442. 

  443.                     if( h->non_zero_count_cache[b_idx] |

  444.                         h->non_zero_count_cache[bn_idx] ) {

  445.                         bS[i] = 2;

  446.                     }

  447.                     else if(!mv_done)

  448.                     {

  449.                         bS[i] = check_mv(h, b_idx, bn_idx, mvy_limit);

  450.                     }

  451.                 }

  452.             }

  453. 

  454.             /* Filter edge */

  455.             // Do not use s->qscale as luma quantizer because it has not the same

  456.             // value in IPCM macroblocks.

  457.             if(bS[0]+bS[1]+bS[2]+bS[3]){

  458.                 qp = ( s->current_picture.qscale_table[mb_xy] + s->current_picture.qscale_table[mbm_xy] + 1 ) >> 1;

  459.                 //tprintf(s->avctx, "filter mb:%d/%d dir:%d edge:%d, QPy:%d, QPc:%d, QPcn:%d\n", mb_x, mb_y, dir, edge, qp, h->chroma_qp[0], s->current_picture.qscale_table[mbn_xy]);

  460.                 tprintf(s->avctx, "filter mb:%d/%d dir:%d edge:%d, QPy:%d ls:%d uvls:%d", mb_x, mb_y, dir, edge, qp, linesize, uvlinesize);

  461.                 //{ int i; for (i = 0; i < 4; i++) tprintf(s->avctx, " bS[%d]:%d", i, bS[i]); tprintf(s->avctx, "\n"); }

  462.                 if( dir == 0 ) {

  463.                     filter_mb_edgev( &img_y[0], linesize, bS, qp, h );

  464.                     {

  465.                         int qp= ( h->chroma_qp[0] + get_chroma_qp( h, 0, s->current_picture.qscale_table[mbm_xy] ) + 1 ) >> 1;

  466.                         filter_mb_edgecv( &img_cb[0], uvlinesize, bS, qp, h);

  467.                         if(h->pps.chroma_qp_diff)

  468.                             qp= ( h->chroma_qp[1] + get_chroma_qp( h, 1, s->current_picture.qscale_table[mbm_xy] ) + 1 ) >> 1;

  469.                         filter_mb_edgecv( &img_cr[0], uvlinesize, bS, qp, h);

  470.                     }

  471.                 } else {

  472.                     filter_mb_edgeh( &img_y[0], linesize, bS, qp, h );

  473.                     {

  474.                         int qp= ( h->chroma_qp[0] + get_chroma_qp( h, 0, s->current_picture.qscale_table[mbm_xy] ) + 1 ) >> 1;

  475.                         filter_mb_edgech( &img_cb[0], uvlinesize, bS, qp, h);

  476.                         if(h->pps.chroma_qp_diff)

  477.                             qp= ( h->chroma_qp[1] + get_chroma_qp( h, 1, s->current_picture.qscale_table[mbm_xy] ) + 1 ) >> 1;

  478.                         filter_mb_edgech( &img_cr[0], uvlinesize, bS, qp, h);

  479.                     }

  480.                 }

  481.             }

  482.         }

  483.     }

  484. 

  485.     /* Calculate bS */

  486.     for( edge = 1; edge < edges; edge++ ) {

  487.         DECLARE_ALIGNED(8, int16_t, bS)[4];

  488.         int qp;

  489. 

  490.         if( IS_8x8DCT(mb_type & (edge<<24)) ) // (edge&1) && IS_8x8DCT(mb_type)

  491.             continue;

  492. 

  493.         if( IS_INTRA(mb_type)) {

  494.             AV_WN64A(bS, 0x0003000300030003ULL);

  495.         } else {

  496.             int i;

  497.             int mv_done;

  498. 

  499.             if( edge & mask_edge ) {

  500.                 AV_ZERO64(bS);

  501.                 mv_done = 1;

  502.             }

  503.             else if( mask_par0 ) {

  504.                 int b_idx= 8 + 4 + edge * (dir ? 8:1);

  505.                 int bn_idx= b_idx - (dir ? 8:1);

  506. 

  507.                 bS[0] = bS[1] = bS[2] = bS[3] = check_mv(h, b_idx, bn_idx, mvy_limit);

  508.                 mv_done = 1;

  509.             }

  510.             else

  511.                 mv_done = 0;

  512. 

  513.             for( i = 0; i < 4; i++ ) {

  514.                 int x = dir == 0 ? edge : i;

  515.                 int y = dir == 0 ? i    : edge;

  516.                 int b_idx= 8 + 4 + x + 8*y;

  517.                 int bn_idx= b_idx - (dir ? 8:1);

  518. 

  519.                 if( h->non_zero_count_cache[b_idx] |

  520.                     h->non_zero_count_cache[bn_idx] ) {

  521.                     bS[i] = 2;

  522.                 }

  523.                 else if(!mv_done)

  524.                 {

  525.                     bS[i] = check_mv(h, b_idx, bn_idx, mvy_limit);

  526.                 }

  527.             }

  528. 

  529.             if(bS[0]+bS[1]+bS[2]+bS[3] == 0)

  530.                 continue;

  531.         }

  532. 

  533.         /* Filter edge */

  534.         // Do not use s->qscale as luma quantizer because it has not the same

  535.         // value in IPCM macroblocks.

  536.         qp = s->current_picture.qscale_table[mb_xy];

  537.         //tprintf(s->avctx, "filter mb:%d/%d dir:%d edge:%d, QPy:%d, QPc:%d, QPcn:%d\n", mb_x, mb_y, dir, edge, qp, h->chroma_qp[0], s->current_picture.qscale_table[mbn_xy]);

  538.         tprintf(s->avctx, "filter mb:%d/%d dir:%d edge:%d, QPy:%d ls:%d uvls:%d", mb_x, mb_y, dir, edge, qp, linesize, uvlinesize);

  539.         //{ int i; for (i = 0; i < 4; i++) tprintf(s->avctx, " bS[%d]:%d", i, bS[i]); tprintf(s->avctx, "\n"); }

  540.         if( dir == 0 ) {

  541.             filter_mb_edgev( &img_y[4*edge], linesize, bS, qp, h );

  542.             if( (edge&1) == 0 ) {

  543.                 filter_mb_edgecv( &img_cb[2*edge], uvlinesize, bS, h->chroma_qp[0], h);

  544.                 filter_mb_edgecv( &img_cr[2*edge], uvlinesize, bS, h->chroma_qp[1], h);

  545.             }

  546.         } else {

  547.             filter_mb_edgeh( &img_y[4*edge*linesize], linesize, bS, qp, h );

  548.             if( (edge&1) == 0 ) {

  549.                 filter_mb_edgech( &img_cb[2*edge*uvlinesize], uvlinesize, bS, h->chroma_qp[0], h);

  550.                 filter_mb_edgech( &img_cr[2*edge*uvlinesize], uvlinesize, bS, h->chroma_qp[1], h);

  551.             }

  552.         }

  553.     }

  554. }

  555. 

  556. void ff_h264_filter_mb( H264Context *h, int mb_x, int mb_y, uint8_t *img_y, uint8_t *img_cb, uint8_t *img_cr, unsigned int linesize, unsigned int uvlinesize) {

  557.     MpegEncContext * const s = &h->s;

  558.     const int mb_xy= mb_x + mb_y*s->mb_stride;

  559.     const int mb_type = s->current_picture.mb_type[mb_xy];

  560.     const int mvy_limit = IS_INTERLACED(mb_type) ? 2 : 4;

  561.     int first_vertical_edge_done = 0;

  562.     av_unused int dir;

  563. 

  564.     if (FRAME_MBAFF

  565.             // and current and left pair do not have the same interlaced type

  566.             && IS_INTERLACED(mb_type^h->left_type[0])

  567.             // and left mb is in available to us

  568.             && h->left_type[0]) {

  569.         /* First vertical edge is different in MBAFF frames

  570.          * There are 8 different bS to compute and 2 different Qp

  571.          */

  572.         DECLARE_ALIGNED(8, int16_t, bS)[8];

  573.         int qp[2];

  574.         int bqp[2];

  575.         int rqp[2];

  576.         int mb_qp, mbn0_qp, mbn1_qp;

  577.         int i;

  578.         first_vertical_edge_done = 1;

  579. 

  580.         if( IS_INTRA(mb_type) ) {

  581.             AV_WN64A(&bS[0], 0x0004000400040004ULL);

  582.             AV_WN64A(&bS[4], 0x0004000400040004ULL);

  583.         } else {

  584.             static const uint8_t offset[2][2][8]={

  585.                 {

  586.                     {7+8*0, 7+8*0, 7+8*0, 7+8*0, 7+8*1, 7+8*1, 7+8*1, 7+8*1},

  587.                     {7+8*2, 7+8*2, 7+8*2, 7+8*2, 7+8*3, 7+8*3, 7+8*3, 7+8*3},

  588.                 },{

  589.                     {7+8*0, 7+8*1, 7+8*2, 7+8*3, 7+8*0, 7+8*1, 7+8*2, 7+8*3},

  590.                     {7+8*0, 7+8*1, 7+8*2, 7+8*3, 7+8*0, 7+8*1, 7+8*2, 7+8*3},

  591.                 }

  592.             };

  593.             const uint8_t *off= offset[MB_FIELD][mb_y&1];

  594.             for( i = 0; i < 8; i++ ) {

  595.                 int j= MB_FIELD ? i>>2 : i&1;

  596.                 int mbn_xy = h->left_mb_xy[j];

  597.                 int mbn_type= h->left_type[j];

  598. 

  599.                 if( IS_INTRA( mbn_type ) )

  600.                     bS[i] = 4;

  601.                 else{

  602.                     bS[i] = 1 + !!(h->non_zero_count_cache[12+8*(i>>1)] |

  603.                          ((!h->pps.cabac && IS_8x8DCT(mbn_type)) ?

  604.                             (h->cbp_table[mbn_xy] & ((MB_FIELD ? (i&2) : (mb_y&1)) ? 8 : 2))

  605.                                                                        :

  606.                             h->non_zero_count[mbn_xy][ off[i] ]));

  607.                 }

  608.             }

  609.         }

  610. 

  611.         mb_qp = s->current_picture.qscale_table[mb_xy];

  612.         mbn0_qp = s->current_picture.qscale_table[h->left_mb_xy[0]];

  613.         mbn1_qp = s->current_picture.qscale_table[h->left_mb_xy[1]];

  614.         qp[0] = ( mb_qp + mbn0_qp + 1 ) >> 1;

  615.         bqp[0] = ( get_chroma_qp( h, 0, mb_qp ) +

  616.                    get_chroma_qp( h, 0, mbn0_qp ) + 1 ) >> 1;

  617.         rqp[0] = ( get_chroma_qp( h, 1, mb_qp ) +

  618.                    get_chroma_qp( h, 1, mbn0_qp ) + 1 ) >> 1;

  619.         qp[1] = ( mb_qp + mbn1_qp + 1 ) >> 1;

  620.         bqp[1] = ( get_chroma_qp( h, 0, mb_qp ) +

  621.                    get_chroma_qp( h, 0, mbn1_qp ) + 1 ) >> 1;

  622.         rqp[1] = ( get_chroma_qp( h, 1, mb_qp ) +

  623.                    get_chroma_qp( h, 1, mbn1_qp ) + 1 ) >> 1;

  624. 

  625.         /* Filter edge */

  626.         tprintf(s->avctx, "filter mb:%d/%d MBAFF, QPy:%d/%d, QPb:%d/%d QPr:%d/%d ls:%d uvls:%d", mb_x, mb_y, qp[0], qp[1], bqp[0], bqp[1], rqp[0], rqp[1], linesize, uvlinesize);

  627.         { int i; for (i = 0; i < 8; i++) tprintf(s->avctx, " bS[%d]:%d", i, bS[i]); tprintf(s->avctx, "\n"); }

  628.         if(MB_FIELD){

  629.             filter_mb_mbaff_edgev ( h, img_y                ,   linesize, bS  , 1, qp [0] );

  630.             filter_mb_mbaff_edgev ( h, img_y  + 8*  linesize,   linesize, bS+4, 1, qp [1] );

  631.             filter_mb_mbaff_edgecv( h, img_cb,                uvlinesize, bS  , 1, bqp[0] );

  632.             filter_mb_mbaff_edgecv( h, img_cb + 4*uvlinesize, uvlinesize, bS+4, 1, bqp[1] );

  633.             filter_mb_mbaff_edgecv( h, img_cr,                uvlinesize, bS  , 1, rqp[0] );

  634.             filter_mb_mbaff_edgecv( h, img_cr + 4*uvlinesize, uvlinesize, bS+4, 1, rqp[1] );

  635.         }else{

  636.             filter_mb_mbaff_edgev ( h, img_y              , 2*  linesize, bS  , 2, qp [0] );

  637.             filter_mb_mbaff_edgev ( h, img_y  +   linesize, 2*  linesize, bS+1, 2, qp [1] );

  638.             filter_mb_mbaff_edgecv( h, img_cb,              2*uvlinesize, bS  , 2, bqp[0] );

  639.             filter_mb_mbaff_edgecv( h, img_cb + uvlinesize, 2*uvlinesize, bS+1, 2, bqp[1] );

  640.             filter_mb_mbaff_edgecv( h, img_cr,              2*uvlinesize, bS  , 2, rqp[0] );

  641.             filter_mb_mbaff_edgecv( h, img_cr + uvlinesize, 2*uvlinesize, bS+1, 2, rqp[1] );

  642.         }

  643.     }

  644. 

  645. #if CONFIG_SMALL

  646.     for( dir = 0; dir < 2; dir++ )

  647.         filter_mb_dir(h, mb_x, mb_y, img_y, img_cb, img_cr, linesize, uvlinesize, mb_xy, mb_type, mvy_limit, dir ? 0 : first_vertical_edge_done, dir);

  648. #else

  649.     filter_mb_dir(h, mb_x, mb_y, img_y, img_cb, img_cr, linesize, uvlinesize, mb_xy, mb_type, mvy_limit, first_vertical_edge_done, 0);

  650.     filter_mb_dir(h, mb_x, mb_y, img_y, img_cb, img_cr, linesize, uvlinesize, mb_xy, mb_type, mvy_limit, 0, 1);

  651. #endif

  652. }