1 /*****************************************************************************
2 * i420_ymga.c : YUV to YUV conversion module for vlc
3 *****************************************************************************
4 * Copyright (C) 2000, 2001 the VideoLAN team
7 * Authors: Samuel Hocevar <sam@zoy.org>
9 * This program is free software; you can redistribute it and/or modify
10 * it under the terms of the GNU General Public License as published by
11 * the Free Software Foundation; either version 2 of the License, or
12 * (at your option) any later version.
14 * This program is distributed in the hope that it will be useful,
15 * but WITHOUT ANY WARRANTY; without even the implied warranty of
16 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
17 * GNU General Public License for more details.
19 * You should have received a copy of the GNU General Public License
20 * along with this program; if not, write to the Free Software
21 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
22 *****************************************************************************/
24 /*****************************************************************************
26 *****************************************************************************/
32 #include <errno.h> /* ENOMEM */
34 #include <vlc_common.h>
35 #include <vlc_plugin.h>
36 #include <vlc_filter.h>
38 #define SRC_FOURCC "I420,IYUV,YV12"
39 #define DEST_FOURCC "YMGA"
41 /*****************************************************************************
42 * Local and extern prototypes.
43 *****************************************************************************/
44 static int Activate ( vlc_object_t * );
45 static void I420_YMGA ( filter_t *, picture_t *, picture_t * );
46 static picture_t *I420_YMGA_Filter( filter_t *, picture_t * );
48 /*****************************************************************************
50 *****************************************************************************/
52 #if defined (MODULE_NAME_IS_i420_ymga)
53 set_description( N_("Conversions from " SRC_FOURCC " to " DEST_FOURCC) )
54 set_capability( "video filter2", 80 )
55 #elif defined (MODULE_NAME_IS_i420_ymga_mmx)
56 set_description( N_("MMX conversions from " SRC_FOURCC " to " DEST_FOURCC) )
57 set_capability( "video filter2", 100 )
58 add_requirement( MMX )
60 set_callbacks( Activate, NULL )
63 /*****************************************************************************
64 * Activate: allocate a chroma function
65 *****************************************************************************
66 * This function allocates and initializes a chroma function
67 *****************************************************************************/
68 static int Activate( vlc_object_t *p_this )
70 filter_t *p_filter = (filter_t *)p_this;
72 if( p_filter->fmt_in.video.i_width & 1
73 || p_filter->fmt_in.video.i_height & 1 )
78 if( p_filter->fmt_in.video.i_width != p_filter->fmt_out.video.i_width
79 || p_filter->fmt_in.video.i_height != p_filter->fmt_out.video.i_height )
82 switch( p_filter->fmt_in.video.i_chroma )
86 switch( p_filter->fmt_out.video.i_chroma )
89 p_filter->pf_video_filter = I420_YMGA_Filter;
104 /* Following functions are local */
106 VIDEO_FILTER_WRAPPER( I420_YMGA )
108 /*****************************************************************************
109 * I420_YMGA: planar YUV 4:2:0 to Matrox's planar/packed YUV 4:2:0
110 *****************************************************************************/
111 static void I420_YMGA( filter_t *p_filter, picture_t *p_source,
114 uint8_t *p_uv = p_dest->U_PIXELS;
115 uint8_t *p_u = p_source->U_PIXELS;
116 uint8_t *p_v = p_source->V_PIXELS;
120 /* Copy the Y part */
121 vlc_memcpy( p_dest->Y_PIXELS, p_source->Y_PIXELS,
122 p_dest->p[Y_PLANE].i_pitch * p_dest->p[Y_PLANE].i_visible_lines );
124 /* Copy the U:V part */
125 for( i_x = p_dest->p[U_PLANE].i_pitch * p_dest->p[U_PLANE].i_visible_lines / 64;
128 #if defined (MODULE_NAME_IS_i420_ymga)
129 *p_uv++ = *p_u++; *p_uv++ = *p_v++; *p_uv++ = *p_u++; *p_uv++ = *p_v++;
130 *p_uv++ = *p_u++; *p_uv++ = *p_v++; *p_uv++ = *p_u++; *p_uv++ = *p_v++;
131 *p_uv++ = *p_u++; *p_uv++ = *p_v++; *p_uv++ = *p_u++; *p_uv++ = *p_v++;
132 *p_uv++ = *p_u++; *p_uv++ = *p_v++; *p_uv++ = *p_u++; *p_uv++ = *p_v++;
133 *p_uv++ = *p_u++; *p_uv++ = *p_v++; *p_uv++ = *p_u++; *p_uv++ = *p_v++;
134 *p_uv++ = *p_u++; *p_uv++ = *p_v++; *p_uv++ = *p_u++; *p_uv++ = *p_v++;
135 *p_uv++ = *p_u++; *p_uv++ = *p_v++; *p_uv++ = *p_u++; *p_uv++ = *p_v++;
136 *p_uv++ = *p_u++; *p_uv++ = *p_v++; *p_uv++ = *p_u++; *p_uv++ = *p_v++;
138 __asm__( ".p2align 5 \n\
139 movd (%0), %%mm0 # Load 4 Cr 00 00 00 00 v3 v2 v1 v0 \n\
140 movd 4(%0), %%mm2 # Load 4 Cr 00 00 00 00 v3 v2 v1 v0 \n\
141 movd 8(%0), %%mm4 # Load 4 Cr 00 00 00 00 v3 v2 v1 v0 \n\
142 movd 12(%0), %%mm6 # Load 4 Cr 00 00 00 00 v3 v2 v1 v0 \n\
143 movd (%1), %%mm1 # Load 4 Cb 00 00 00 00 u3 u2 u1 u0 \n\
144 movd 4(%1), %%mm3 # Load 4 Cb 00 00 00 00 u3 u2 u1 u0 \n\
145 movd 8(%1), %%mm5 # Load 4 Cb 00 00 00 00 u3 u2 u1 u0 \n\
146 movd 12(%1), %%mm7 # Load 4 Cb 00 00 00 00 u3 u2 u1 u0 \n\
147 punpcklbw %%mm1, %%mm0 # u3 v3 u2 v2 u1 v1 u0 v0 \n\
148 punpcklbw %%mm3, %%mm2 # u3 v3 u2 v2 u1 v1 u0 v0 \n\
149 punpcklbw %%mm5, %%mm4 # u3 v3 u2 v2 u1 v1 u0 v0 \n\
150 punpcklbw %%mm7, %%mm6 # u3 v3 u2 v2 u1 v1 u0 v0 \n\
151 movq %%mm0, (%2) # Store CrCb \n\
152 movq %%mm2, 8(%2) # Store CrCb \n\
153 movq %%mm4, 16(%2) # Store CrCb \n\
154 movq %%mm6, 24(%2) # Store CrCb"
155 : : "r" (p_v), "r" (p_u), "r" (p_uv) );
157 p_v += 16; p_u += 16; p_uv += 32;