1 /*****************************************************************************
2 * i422_yuy2.c : Planar YUV 4:2:2 to Packed YUV conversion module for vlc
3 *****************************************************************************
4 * Copyright (C) 2000, 2001 the VideoLAN team
7 * Authors: Samuel Hocevar <sam@zoy.org>
8 * Damien Fouilleul <damienf@videolan.org>
10 * This program is free software; you can redistribute it and/or modify
11 * it under the terms of the GNU General Public License as published by
12 * the Free Software Foundation; either version 2 of the License, or
13 * (at your option) any later version.
15 * This program is distributed in the hope that it will be useful,
16 * but WITHOUT ANY WARRANTY; without even the implied warranty of
17 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18 * GNU General Public License for more details.
20 * You should have received a copy of the GNU General Public License
21 * along with this program; if not, write to the Free Software
22 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
23 *****************************************************************************/
25 /*****************************************************************************
27 *****************************************************************************/
33 #include <vlc_common.h>
34 #include <vlc_plugin.h>
35 #include <vlc_filter.h>
38 #include "i422_yuy2.h"
40 #define SRC_FOURCC "I422"
41 #if defined (MODULE_NAME_IS_i422_yuy2)
42 # define DEST_FOURCC "YUY2,YUNV,YVYU,UYVY,UYNV,Y422,IUYV,cyuv,Y211"
44 # define DEST_FOURCC "YUY2,YUNV,YVYU,UYVY,UYNV,Y422,IUYV,cyuv"
47 /*****************************************************************************
48 * Local and extern prototypes.
49 *****************************************************************************/
50 static int Activate ( vlc_object_t * );
52 static void I422_YUY2 ( filter_t *, picture_t *, picture_t * );
53 static void I422_YVYU ( filter_t *, picture_t *, picture_t * );
54 static void I422_UYVY ( filter_t *, picture_t *, picture_t * );
55 static void I422_IUYV ( filter_t *, picture_t *, picture_t * );
56 static void I422_cyuv ( filter_t *, picture_t *, picture_t * );
57 static picture_t *I422_YUY2_Filter ( filter_t *, picture_t * );
58 static picture_t *I422_YVYU_Filter ( filter_t *, picture_t * );
59 static picture_t *I422_UYVY_Filter ( filter_t *, picture_t * );
60 static picture_t *I422_IUYV_Filter ( filter_t *, picture_t * );
61 static picture_t *I422_cyuv_Filter ( filter_t *, picture_t * );
62 #if defined (MODULE_NAME_IS_i422_yuy2)
63 static void I422_Y211 ( filter_t *, picture_t *, picture_t * );
64 static picture_t *I422_Y211_Filter ( filter_t *, picture_t * );
67 /*****************************************************************************
69 *****************************************************************************/
71 #if defined (MODULE_NAME_IS_i422_yuy2)
72 set_description( N_("Conversions from " SRC_FOURCC " to " DEST_FOURCC) )
73 set_capability( "video filter2", 80 )
74 # define CPU_CAPABILITY 0
75 #elif defined (MODULE_NAME_IS_i422_yuy2_mmx)
76 set_description( N_("MMX conversions from " SRC_FOURCC " to " DEST_FOURCC) )
77 set_capability( "video filter2", 100 )
78 # define CPU_CAPABILITY CPU_CAPABILITY_MMX
79 #elif defined (MODULE_NAME_IS_i422_yuy2_sse2)
80 set_description( N_("SSE2 conversions from " SRC_FOURCC " to " DEST_FOURCC) )
81 set_capability( "video filter2", 120 )
82 # define CPU_CAPABILITY CPU_CAPABILITY_SSE2
84 set_callbacks( Activate, NULL )
87 /*****************************************************************************
88 * Activate: allocate a chroma function
89 *****************************************************************************
90 * This function allocates and initializes a chroma function
91 *****************************************************************************/
92 static int Activate( vlc_object_t *p_this )
94 filter_t *p_filter = (filter_t *)p_this;
97 if( !(vlc_CPU() & CPU_CAPABILITY) )
100 if( p_filter->fmt_in.video.i_width & 1
101 || p_filter->fmt_in.video.i_height & 1 )
106 switch( p_filter->fmt_in.video.i_chroma )
109 switch( p_filter->fmt_out.video.i_chroma )
112 p_filter->pf_video_filter = I422_YUY2_Filter;
116 p_filter->pf_video_filter = I422_YVYU_Filter;
120 p_filter->pf_video_filter = I422_UYVY_Filter;
123 case VLC_FOURCC('I','U','Y','V'):
124 p_filter->pf_video_filter = I422_IUYV_Filter;
128 p_filter->pf_video_filter = I422_cyuv_Filter;
131 #if defined (MODULE_NAME_IS_i422_yuy2)
133 p_filter->pf_video_filter = I422_Y211_Filter;
148 /* Following functions are local */
150 VIDEO_FILTER_WRAPPER( I422_YUY2 )
151 VIDEO_FILTER_WRAPPER( I422_YVYU )
152 VIDEO_FILTER_WRAPPER( I422_UYVY )
153 VIDEO_FILTER_WRAPPER( I422_IUYV )
154 VIDEO_FILTER_WRAPPER( I422_cyuv )
155 #if defined (MODULE_NAME_IS_i422_yuy2)
156 VIDEO_FILTER_WRAPPER( I422_Y211 )
159 /*****************************************************************************
160 * I422_YUY2: planar YUV 4:2:2 to packed YUY2 4:2:2
161 *****************************************************************************/
162 static void I422_YUY2( filter_t *p_filter, picture_t *p_source,
165 uint8_t *p_line = p_dest->p->p_pixels;
166 uint8_t *p_y = p_source->Y_PIXELS;
167 uint8_t *p_u = p_source->U_PIXELS;
168 uint8_t *p_v = p_source->V_PIXELS;
172 const int i_source_margin = p_source->p[0].i_pitch
173 - p_source->p[0].i_visible_pitch;
174 const int i_source_margin_c = p_source->p[1].i_pitch
175 - p_source->p[1].i_visible_pitch;
176 const int i_dest_margin = p_dest->p->i_pitch
177 - p_dest->p->i_visible_pitch;
179 #if defined (MODULE_NAME_IS_i422_yuy2_sse2)
181 if( 0 == (15 & (p_source->p[Y_PLANE].i_pitch|p_dest->p->i_pitch|
182 ((intptr_t)p_line|(intptr_t)p_y))) )
184 /* use faster SSE2 aligned fetch and store */
185 for( i_y = p_filter->fmt_in.video.i_height ; i_y-- ; )
187 for( i_x = p_filter->fmt_in.video.i_width / 16 ; i_x-- ; )
189 SSE2_CALL( SSE2_YUV422_YUYV_ALIGNED );
191 for( i_x = ( p_filter->fmt_in.video.i_width % 16 ) / 2; i_x-- ; )
193 C_YUV422_YUYV( p_line, p_y, p_u, p_v );
195 p_y += i_source_margin;
196 p_u += i_source_margin_c;
197 p_v += i_source_margin_c;
198 p_line += i_dest_margin;
202 /* use slower SSE2 unaligned fetch and store */
203 for( i_y = p_filter->fmt_in.video.i_height ; i_y-- ; )
205 for( i_x = p_filter->fmt_in.video.i_width / 16 ; i_x-- ; )
207 SSE2_CALL( SSE2_YUV422_YUYV_UNALIGNED );
209 for( i_x = ( p_filter->fmt_in.video.i_width % 16 ) / 2; i_x-- ; )
211 C_YUV422_YUYV( p_line, p_y, p_u, p_v );
213 p_y += i_source_margin;
214 p_u += i_source_margin_c;
215 p_v += i_source_margin_c;
216 p_line += i_dest_margin;
223 for( i_y = p_filter->fmt_in.video.i_height ; i_y-- ; )
225 for( i_x = p_filter->fmt_in.video.i_width / 8 ; i_x-- ; )
227 #if defined (MODULE_NAME_IS_i422_yuy2)
228 C_YUV422_YUYV( p_line, p_y, p_u, p_v );
229 C_YUV422_YUYV( p_line, p_y, p_u, p_v );
230 C_YUV422_YUYV( p_line, p_y, p_u, p_v );
231 C_YUV422_YUYV( p_line, p_y, p_u, p_v );
232 #elif defined (MODULE_NAME_IS_i422_yuy2_mmx)
233 MMX_CALL( MMX_YUV422_YUYV );
236 for( i_x = ( p_filter->fmt_in.video.i_width % 8 ) / 2; i_x-- ; )
238 C_YUV422_YUYV( p_line, p_y, p_u, p_v );
240 p_y += i_source_margin;
241 p_u += i_source_margin_c;
242 p_v += i_source_margin_c;
243 p_line += i_dest_margin;
245 #if defined (MODULE_NAME_IS_i422_yuy2_mmx)
252 /*****************************************************************************
253 * I422_YVYU: planar YUV 4:2:2 to packed YVYU 4:2:2
254 *****************************************************************************/
255 static void I422_YVYU( filter_t *p_filter, picture_t *p_source,
258 uint8_t *p_line = p_dest->p->p_pixels;
259 uint8_t *p_y = p_source->Y_PIXELS;
260 uint8_t *p_u = p_source->U_PIXELS;
261 uint8_t *p_v = p_source->V_PIXELS;
265 const int i_source_margin = p_source->p[0].i_pitch
266 - p_source->p[0].i_visible_pitch;
267 const int i_source_margin_c = p_source->p[1].i_pitch
268 - p_source->p[1].i_visible_pitch;
269 const int i_dest_margin = p_dest->p->i_pitch
270 - p_dest->p->i_visible_pitch;
272 #if defined (MODULE_NAME_IS_i422_yuy2_sse2)
274 if( 0 == (15 & (p_source->p[Y_PLANE].i_pitch|p_dest->p->i_pitch|
275 ((intptr_t)p_line|(intptr_t)p_y))) )
277 /* use faster SSE2 aligned fetch and store */
278 for( i_y = p_filter->fmt_in.video.i_height ; i_y-- ; )
280 for( i_x = p_filter->fmt_in.video.i_width / 16 ; i_x-- ; )
282 SSE2_CALL( SSE2_YUV422_YVYU_ALIGNED );
284 for( i_x = ( p_filter->fmt_in.video.i_width % 16 ) / 2; i_x-- ; )
286 C_YUV422_YVYU( p_line, p_y, p_u, p_v );
288 p_y += i_source_margin;
289 p_u += i_source_margin_c;
290 p_v += i_source_margin_c;
291 p_line += i_dest_margin;
295 /* use slower SSE2 unaligned fetch and store */
296 for( i_y = p_filter->fmt_in.video.i_height ; i_y-- ; )
298 for( i_x = p_filter->fmt_in.video.i_width / 16 ; i_x-- ; )
300 SSE2_CALL( SSE2_YUV422_YVYU_UNALIGNED );
302 for( i_x = ( p_filter->fmt_in.video.i_width % 16 ) / 2; i_x-- ; )
304 C_YUV422_YVYU( p_line, p_y, p_u, p_v );
306 p_y += i_source_margin;
307 p_u += i_source_margin_c;
308 p_v += i_source_margin_c;
309 p_line += i_dest_margin;
316 for( i_y = p_filter->fmt_in.video.i_height ; i_y-- ; )
318 for( i_x = p_filter->fmt_in.video.i_width / 8 ; i_x-- ; )
320 #if defined (MODULE_NAME_IS_i422_yuy2)
321 C_YUV422_YVYU( p_line, p_y, p_u, p_v );
322 C_YUV422_YVYU( p_line, p_y, p_u, p_v );
323 C_YUV422_YVYU( p_line, p_y, p_u, p_v );
324 C_YUV422_YVYU( p_line, p_y, p_u, p_v );
325 #elif defined (MODULE_NAME_IS_i422_yuy2_mmx)
326 MMX_CALL( MMX_YUV422_YVYU );
329 for( i_x = ( p_filter->fmt_in.video.i_width % 8 ) / 2; i_x-- ; )
331 C_YUV422_YVYU( p_line, p_y, p_u, p_v );
333 p_y += i_source_margin;
334 p_u += i_source_margin_c;
335 p_v += i_source_margin_c;
336 p_line += i_dest_margin;
338 #if defined (MODULE_NAME_IS_i422_yuy2_mmx)
345 /*****************************************************************************
346 * I422_UYVY: planar YUV 4:2:2 to packed UYVY 4:2:2
347 *****************************************************************************/
348 static void I422_UYVY( filter_t *p_filter, picture_t *p_source,
351 uint8_t *p_line = p_dest->p->p_pixels;
352 uint8_t *p_y = p_source->Y_PIXELS;
353 uint8_t *p_u = p_source->U_PIXELS;
354 uint8_t *p_v = p_source->V_PIXELS;
358 const int i_source_margin = p_source->p[0].i_pitch
359 - p_source->p[0].i_visible_pitch;
360 const int i_source_margin_c = p_source->p[1].i_pitch
361 - p_source->p[1].i_visible_pitch;
362 const int i_dest_margin = p_dest->p->i_pitch
363 - p_dest->p->i_visible_pitch;
365 #if defined (MODULE_NAME_IS_i422_yuy2_sse2)
367 if( 0 == (15 & (p_source->p[Y_PLANE].i_pitch|p_dest->p->i_pitch|
368 ((intptr_t)p_line|(intptr_t)p_y))) )
370 /* use faster SSE2 aligned fetch and store */
371 for( i_y = p_filter->fmt_in.video.i_height ; i_y-- ; )
373 for( i_x = p_filter->fmt_in.video.i_width / 16 ; i_x-- ; )
375 SSE2_CALL( SSE2_YUV422_UYVY_ALIGNED );
377 for( i_x = ( p_filter->fmt_in.video.i_width % 16 ) / 2; i_x-- ; )
379 C_YUV422_UYVY( p_line, p_y, p_u, p_v );
381 p_y += i_source_margin;
382 p_u += i_source_margin_c;
383 p_v += i_source_margin_c;
384 p_line += i_dest_margin;
388 /* use slower SSE2 unaligned fetch and store */
389 for( i_y = p_filter->fmt_in.video.i_height ; i_y-- ; )
391 for( i_x = p_filter->fmt_in.video.i_width / 16 ; i_x-- ; )
393 SSE2_CALL( SSE2_YUV422_UYVY_UNALIGNED );
395 for( i_x = ( p_filter->fmt_in.video.i_width % 16 ) / 2; i_x-- ; )
397 C_YUV422_UYVY( p_line, p_y, p_u, p_v );
399 p_y += i_source_margin;
400 p_u += i_source_margin_c;
401 p_v += i_source_margin_c;
402 p_line += i_dest_margin;
409 for( i_y = p_filter->fmt_in.video.i_height ; i_y-- ; )
411 for( i_x = p_filter->fmt_in.video.i_width / 8 ; i_x-- ; )
413 #if defined (MODULE_NAME_IS_i422_yuy2)
414 C_YUV422_UYVY( p_line, p_y, p_u, p_v );
415 C_YUV422_UYVY( p_line, p_y, p_u, p_v );
416 C_YUV422_UYVY( p_line, p_y, p_u, p_v );
417 C_YUV422_UYVY( p_line, p_y, p_u, p_v );
418 #elif defined (MODULE_NAME_IS_i422_yuy2_mmx)
419 MMX_CALL( MMX_YUV422_UYVY );
422 for( i_x = ( p_filter->fmt_in.video.i_width % 8 ) / 2; i_x-- ; )
424 C_YUV422_UYVY( p_line, p_y, p_u, p_v );
426 p_y += i_source_margin;
427 p_u += i_source_margin_c;
428 p_v += i_source_margin_c;
429 p_line += i_dest_margin;
431 #if defined (MODULE_NAME_IS_i422_yuy2_mmx)
438 /*****************************************************************************
439 * I422_IUYV: planar YUV 4:2:2 to interleaved packed IUYV 4:2:2
440 *****************************************************************************/
441 static void I422_IUYV( filter_t *p_filter, picture_t *p_source,
444 VLC_UNUSED(p_source); VLC_UNUSED(p_dest);
446 msg_Err( p_filter, "I422_IUYV unimplemented, please harass <sam@zoy.org>" );
449 /*****************************************************************************
450 * I422_cyuv: planar YUV 4:2:2 to upside-down packed UYVY 4:2:2
451 *****************************************************************************/
452 static void I422_cyuv( filter_t *p_filter, picture_t *p_source,
455 uint8_t *p_line = p_dest->p->p_pixels + p_dest->p->i_visible_lines * p_dest->p->i_pitch;
456 uint8_t *p_y = p_source->Y_PIXELS;
457 uint8_t *p_u = p_source->U_PIXELS;
458 uint8_t *p_v = p_source->V_PIXELS;
462 const int i_source_margin = p_source->p[0].i_pitch
463 - p_source->p[0].i_visible_pitch;
464 const int i_source_margin_c = p_source->p[1].i_pitch
465 - p_source->p[1].i_visible_pitch;
466 const int i_dest_margin = p_dest->p->i_pitch
467 - p_dest->p->i_visible_pitch;
469 #if defined (MODULE_NAME_IS_i422_yuy2_sse2)
471 if( 0 == (15 & (p_source->p[Y_PLANE].i_pitch|p_dest->p->i_pitch|
472 ((intptr_t)p_line|(intptr_t)p_y))) )
474 /* use faster SSE2 aligned fetch and store */
475 for( i_y = p_filter->fmt_in.video.i_height ; i_y-- ; )
477 p_line -= 2 * p_dest->p->i_pitch;
479 for( i_x = p_filter->fmt_in.video.i_width / 16 ; i_x-- ; )
481 SSE2_CALL( SSE2_YUV422_UYVY_ALIGNED );
483 for( i_x = ( p_filter->fmt_in.video.i_width % 16 ) / 2; i_x-- ; )
485 C_YUV422_UYVY( p_line, p_y, p_u, p_v );
487 p_y += i_source_margin;
488 p_u += i_source_margin_c;
489 p_v += i_source_margin_c;
490 p_line += i_dest_margin;
494 /* use slower SSE2 unaligned fetch and store */
495 for( i_y = p_filter->fmt_in.video.i_height ; i_y-- ; )
497 p_line -= 2 * p_dest->p->i_pitch;
499 for( i_x = p_filter->fmt_in.video.i_width / 16 ; i_x-- ; )
501 SSE2_CALL( SSE2_YUV422_UYVY_UNALIGNED );
503 for( i_x = ( p_filter->fmt_in.video.i_width % 16 ) / 2; i_x-- ; )
505 C_YUV422_UYVY( p_line, p_y, p_u, p_v );
507 p_y += i_source_margin;
508 p_u += i_source_margin_c;
509 p_v += i_source_margin_c;
510 p_line += i_dest_margin;
517 for( i_y = p_filter->fmt_in.video.i_height ; i_y-- ; )
519 for( i_x = p_filter->fmt_in.video.i_width / 8 ; i_x-- ; )
521 p_line -= 2 * p_dest->p->i_pitch;
523 #if defined (MODULE_NAME_IS_i422_yuy2)
524 C_YUV422_UYVY( p_line, p_y, p_u, p_v );
525 C_YUV422_UYVY( p_line, p_y, p_u, p_v );
526 C_YUV422_UYVY( p_line, p_y, p_u, p_v );
527 C_YUV422_UYVY( p_line, p_y, p_u, p_v );
528 #elif defined (MODULE_NAME_IS_i422_yuy2_mmx)
529 MMX_CALL( MMX_YUV422_UYVY );
532 p_y += i_source_margin;
533 p_u += i_source_margin_c;
534 p_v += i_source_margin_c;
535 p_line += i_dest_margin;
537 #if defined (MODULE_NAME_IS_i422_yuy2_mmx)
539 #elif defined (MODULE_NAME_IS_i422_yuy2_sse2)
546 /*****************************************************************************
547 * I422_Y211: planar YUV 4:2:2 to packed YUYV 2:1:1
548 *****************************************************************************/
549 #if defined (MODULE_NAME_IS_i422_yuy2)
550 static void I422_Y211( filter_t *p_filter, picture_t *p_source,
553 uint8_t *p_line = p_dest->p->p_pixels + p_dest->p->i_visible_lines * p_dest->p->i_pitch;
554 uint8_t *p_y = p_source->Y_PIXELS;
555 uint8_t *p_u = p_source->U_PIXELS;
556 uint8_t *p_v = p_source->V_PIXELS;
560 for( i_y = p_filter->fmt_in.video.i_height ; i_y-- ; )
562 for( i_x = p_filter->fmt_in.video.i_width / 8 ; i_x-- ; )
564 C_YUV422_Y211( p_line, p_y, p_u, p_v );
565 C_YUV422_Y211( p_line, p_y, p_u, p_v );