1 /*****************************************************************************
2 * mc.h: motion compensation
3 *****************************************************************************
4 * Copyright (C) 2004-2011 x264 project
6 * Authors: Loren Merritt <lorenm@u.washington.edu>
8 * This program is free software; you can redistribute it and/or modify
9 * it under the terms of the GNU General Public License as published by
10 * the Free Software Foundation; either version 2 of the License, or
11 * (at your option) any later version.
13 * This program is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 * GNU General Public License for more details.
18 * You should have received a copy of the GNU General Public License
19 * along with this program; if not, write to the Free Software
20 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02111, USA.
22 * This program is also available under a commercial proprietary license.
23 * For more information, contact us at licensing@x264.com.
24 *****************************************************************************/
30 typedef void (* weight_fn_t)( pixel *, int, pixel *,int, const struct x264_weight_t *, int );
31 typedef struct x264_weight_t
33 /* aligning the first member is a gcc hack to force the struct to be
34 * 16 byte aligned, as well as force sizeof(struct) to be a multiple of 16 */
35 ALIGNED_16( int16_t cachea[8] );
40 weight_fn_t *weightfn;
41 } ALIGNED_16( x264_weight_t );
43 extern const x264_weight_t weight_none[3];
45 #define SET_WEIGHT( w, b, s, d, o )\
51 h->mc.weight_cache( h, &w );\
57 * XXX: Only width = 4, 8 or 16 are valid
58 * width == 4 -> height == 4 or 8
59 * width == 8 -> height == 4 or 8 or 16
60 * width == 16-> height == 8 or 16
65 void (*mc_luma)(pixel *dst, int i_dst, pixel **src, int i_src,
67 int i_width, int i_height, const x264_weight_t *weight );
69 /* may round up the dimensions if they're not a power of 2 */
70 pixel* (*get_ref)(pixel *dst, int *i_dst, pixel **src, int i_src,
72 int i_width, int i_height, const x264_weight_t *weight );
74 /* mc_chroma may write up to 2 bytes of garbage to the right of dst,
75 * so it must be run from left to right. */
76 void (*mc_chroma)(pixel *dstu, pixel *dstv, int i_dst, pixel *src, int i_src,
78 int i_width, int i_height );
80 void (*avg[10])( pixel *dst, int, pixel *src1, int, pixel *src2, int, int i_weight );
82 /* only 16x16, 8x8, and 4x4 defined */
83 void (*copy[7])( pixel *dst, int, pixel *src, int, int i_height );
84 void (*copy_16x16_unaligned)( pixel *dst, int, pixel *src, int, int i_height );
86 void (*store_interleave_8x8x2)( pixel *dst, int i_dst, pixel *srcu, pixel *srcv );
87 void (*load_deinterleave_8x8x2_fenc)( pixel *dst, pixel *src, int i_src );
88 void (*load_deinterleave_8x8x2_fdec)( pixel *dst, pixel *src, int i_src );
90 void (*plane_copy)( pixel *dst, int i_dst,
91 pixel *src, int i_src, int w, int h );
92 void (*plane_copy_interleave)( pixel *dst, int i_dst,
93 pixel *srcu, int i_srcu,
94 pixel *srcv, int i_srcv, int w, int h );
95 /* may write up to 15 pixels off the end of each plane */
96 void (*plane_copy_deinterleave)( pixel *dstu, int i_dstu,
97 pixel *dstv, int i_dstv,
98 pixel *src, int i_src, int w, int h );
100 void (*hpel_filter)( pixel *dsth, pixel *dstv, pixel *dstc, pixel *src,
101 int i_stride, int i_width, int i_height, int16_t *buf );
103 /* prefetch the next few macroblocks of fenc or fdec */
104 void (*prefetch_fenc)( pixel *pix_y, int stride_y,
105 pixel *pix_uv, int stride_uv, int mb_x );
106 /* prefetch the next few macroblocks of a hpel reference frame */
107 void (*prefetch_ref)( pixel *pix, int stride, int parity );
109 void *(*memcpy_aligned)( void *dst, const void *src, size_t n );
110 void (*memzero_aligned)( void *dst, int n );
112 /* successive elimination prefilter */
113 void (*integral_init4h)( uint16_t *sum, pixel *pix, int stride );
114 void (*integral_init8h)( uint16_t *sum, pixel *pix, int stride );
115 void (*integral_init4v)( uint16_t *sum8, uint16_t *sum4, int stride );
116 void (*integral_init8v)( uint16_t *sum8, int stride );
118 void (*frame_init_lowres_core)( pixel *src0, pixel *dst0, pixel *dsth, pixel *dstv, pixel *dstc,
119 int src_stride, int dst_stride, int width, int height );
121 weight_fn_t *offsetadd;
122 weight_fn_t *offsetsub;
123 void (*weight_cache)( x264_t *, x264_weight_t * );
125 void (*mbtree_propagate_cost)( int *dst, uint16_t *propagate_in, uint16_t *intra_costs,
126 uint16_t *inter_costs, uint16_t *inv_qscales, float *fps_factor, int len );
127 } x264_mc_functions_t;
129 void x264_mc_init( int cpu, x264_mc_functions_t *pf );