1 /*****************************************************************************
2 * mc.h: motion compensation
3 *****************************************************************************
4 * Copyright (C) 2004-2013 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 *, intptr_t, pixel *,intptr_t, 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 x264_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, intptr_t i_dst, pixel **src, intptr_t i_src,
66 int mvx, int mvy, int i_width, int i_height, const x264_weight_t *weight );
68 /* may round up the dimensions if they're not a power of 2 */
69 pixel* (*get_ref)( pixel *dst, intptr_t *i_dst, pixel **src, intptr_t i_src,
70 int mvx, int mvy, int i_width, int i_height, const x264_weight_t *weight );
72 /* mc_chroma may write up to 2 bytes of garbage to the right of dst,
73 * so it must be run from left to right. */
74 void (*mc_chroma)( pixel *dstu, pixel *dstv, intptr_t i_dst, pixel *src, intptr_t i_src,
75 int mvx, int mvy, int i_width, int i_height );
77 void (*avg[12])( pixel *dst, intptr_t dst_stride, pixel *src1, intptr_t src1_stride,
78 pixel *src2, intptr_t src2_stride, int i_weight );
80 /* only 16x16, 8x8, and 4x4 defined */
81 void (*copy[7])( pixel *dst, intptr_t dst_stride, pixel *src, intptr_t src_stride, int i_height );
82 void (*copy_16x16_unaligned)( pixel *dst, intptr_t dst_stride, pixel *src, intptr_t src_stride, int i_height );
84 void (*store_interleave_chroma)( pixel *dst, intptr_t i_dst, pixel *srcu, pixel *srcv, int height );
85 void (*load_deinterleave_chroma_fenc)( pixel *dst, pixel *src, intptr_t i_src, int height );
86 void (*load_deinterleave_chroma_fdec)( pixel *dst, pixel *src, intptr_t i_src, int height );
88 void (*plane_copy)( pixel *dst, intptr_t i_dst, pixel *src, intptr_t i_src, int w, int h );
89 void (*plane_copy_interleave)( pixel *dst, intptr_t i_dst, pixel *srcu, intptr_t i_srcu,
90 pixel *srcv, intptr_t i_srcv, int w, int h );
91 /* may write up to 15 pixels off the end of each plane */
92 void (*plane_copy_deinterleave)( pixel *dstu, intptr_t i_dstu, pixel *dstv, intptr_t i_dstv,
93 pixel *src, intptr_t i_src, int w, int h );
94 void (*plane_copy_deinterleave_rgb)( pixel *dsta, intptr_t i_dsta, pixel *dstb, intptr_t i_dstb,
95 pixel *dstc, intptr_t i_dstc, pixel *src, intptr_t i_src, int pw, int w, int h );
96 void (*hpel_filter)( pixel *dsth, pixel *dstv, pixel *dstc, pixel *src,
97 intptr_t i_stride, int i_width, int i_height, int16_t *buf );
99 /* prefetch the next few macroblocks of fenc or fdec */
100 void (*prefetch_fenc) ( pixel *pix_y, intptr_t stride_y, pixel *pix_uv, intptr_t stride_uv, int mb_x );
101 void (*prefetch_fenc_420)( pixel *pix_y, intptr_t stride_y, pixel *pix_uv, intptr_t stride_uv, int mb_x );
102 void (*prefetch_fenc_422)( pixel *pix_y, intptr_t stride_y, pixel *pix_uv, intptr_t stride_uv, int mb_x );
103 /* prefetch the next few macroblocks of a hpel reference frame */
104 void (*prefetch_ref)( pixel *pix, intptr_t stride, int parity );
106 void *(*memcpy_aligned)( void *dst, const void *src, size_t n );
107 void (*memzero_aligned)( void *dst, size_t n );
109 /* successive elimination prefilter */
110 void (*integral_init4h)( uint16_t *sum, pixel *pix, intptr_t stride );
111 void (*integral_init8h)( uint16_t *sum, pixel *pix, intptr_t stride );
112 void (*integral_init4v)( uint16_t *sum8, uint16_t *sum4, intptr_t stride );
113 void (*integral_init8v)( uint16_t *sum8, intptr_t stride );
115 void (*frame_init_lowres_core)( pixel *src0, pixel *dst0, pixel *dsth, pixel *dstv, pixel *dstc,
116 intptr_t src_stride, intptr_t dst_stride, int width, int height );
118 weight_fn_t *offsetadd;
119 weight_fn_t *offsetsub;
120 void (*weight_cache)( x264_t *, x264_weight_t * );
122 void (*mbtree_propagate_cost)( int *dst, uint16_t *propagate_in, uint16_t *intra_costs,
123 uint16_t *inter_costs, uint16_t *inv_qscales, float *fps_factor, int len );
124 } x264_mc_functions_t;
126 void x264_mc_init( int cpu, x264_mc_functions_t *pf, int cpu_independent );