videolan
/
vlc
mirror of https://gitea.goldendeliverer.com/videolan/vlc.git
1
0
Fork 0
Code Issues Projects Releases 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.
109462 Commits
7 Branches
132 Tags
718 MiB
Branch: master
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from 'master'
${ noResults }
vlc/modules/video_chroma/i420_rgb.h

373 lines
25 KiB
Raw Permalink Normal View History

* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
/*****************************************************************************
* i420_rgb.h : YUV to bitmap RGB conversion module for vlc
*****************************************************************************
LGPL Re-license almost all the playback modules to LGPLv2.1+ with authorization from their respective contributors (230+) This includes: - access, codec, packetizers, demux - audio filters, audio mixers, audio output - video filters, video chroma, video output - text renderers - XML parser - ARM NEON and SSE2 optimisations (mostly for chromas and filters) Some modules are not concerned: - BDA and DShow access modules because Manol Manolov is AWOL - Real RTSP, because it is derived from Xine - x264 and t140 because they are encoders only - DLL Loader, because it is derived from MPlayer - DTS packetizer, because Jon Lech Johansen is AWOL - Shine and WMAfixed, because they are derived from Rockbox - Real demuxer, as it is derived from MPlayer and Wang Bo is AWOL - MPC demuxer, as Yavor Doganov is AWOL - Tivo demuxer, because it is derived from an MPlayer fork - Playlist demuxer, (WPL and ZPL parts missing), because suheaven is AWOL - iOS audio output and video display, because author refuses the license change - Equalizer and compressor, because Ronald Wright is AWOL - Mono, Headphone and Dolby, because author refuses the license change - hqdn3d and yadif, because they are from MPlayer/libavfilter - remoteosd, because it derives from RealVNC code - MMX optimisations, because Ollie Lho, from SiS, is AWOL - Rotate, because it depends on GPL motion Nota Bene: - Some modules depend on GPL-only libraries, a LGPL module does not mean that the resulting binary module will be LGPL. Libraries affected would include liba52, libdvdcss, libdvdnav, libdvdread, faad2, libdca, libmad, libmpeg2, libpostproc, SRC, sid, zvbi and probably others.
14 years ago
* Copyright (C) 2000, 2004 VLC authors and VideoLAN
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
*
* Authors: Samuel Hocevar <sam@zoy.org>
*
LGPL Re-license almost all the playback modules to LGPLv2.1+ with authorization from their respective contributors (230+) This includes: - access, codec, packetizers, demux - audio filters, audio mixers, audio output - video filters, video chroma, video output - text renderers - XML parser - ARM NEON and SSE2 optimisations (mostly for chromas and filters) Some modules are not concerned: - BDA and DShow access modules because Manol Manolov is AWOL - Real RTSP, because it is derived from Xine - x264 and t140 because they are encoders only - DLL Loader, because it is derived from MPlayer - DTS packetizer, because Jon Lech Johansen is AWOL - Shine and WMAfixed, because they are derived from Rockbox - Real demuxer, as it is derived from MPlayer and Wang Bo is AWOL - MPC demuxer, as Yavor Doganov is AWOL - Tivo demuxer, because it is derived from an MPlayer fork - Playlist demuxer, (WPL and ZPL parts missing), because suheaven is AWOL - iOS audio output and video display, because author refuses the license change - Equalizer and compressor, because Ronald Wright is AWOL - Mono, Headphone and Dolby, because author refuses the license change - hqdn3d and yadif, because they are from MPlayer/libavfilter - remoteosd, because it derives from RealVNC code - MMX optimisations, because Ollie Lho, from SiS, is AWOL - Rotate, because it depends on GPL motion Nota Bene: - Some modules depend on GPL-only libraries, a LGPL module does not mean that the resulting binary module will be LGPL. Libraries affected would include liba52, libdvdcss, libdvdnav, libdvdread, faad2, libdca, libmad, libmpeg2, libpostproc, SRC, sid, zvbi and probably others.
14 years ago
* This program is free software; you can redistribute it and/or modify it
* under the terms of the GNU Lesser General Public License as published by
* the Free Software Foundation; either version 2.1 of the License, or
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
* (at your option) any later version.
Removes trailing spaces. Removes tabs.
19 years ago
*
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
LGPL Re-license almost all the playback modules to LGPLv2.1+ with authorization from their respective contributors (230+) This includes: - access, codec, packetizers, demux - audio filters, audio mixers, audio output - video filters, video chroma, video output - text renderers - XML parser - ARM NEON and SSE2 optimisations (mostly for chromas and filters) Some modules are not concerned: - BDA and DShow access modules because Manol Manolov is AWOL - Real RTSP, because it is derived from Xine - x264 and t140 because they are encoders only - DLL Loader, because it is derived from MPlayer - DTS packetizer, because Jon Lech Johansen is AWOL - Shine and WMAfixed, because they are derived from Rockbox - Real demuxer, as it is derived from MPlayer and Wang Bo is AWOL - MPC demuxer, as Yavor Doganov is AWOL - Tivo demuxer, because it is derived from an MPlayer fork - Playlist demuxer, (WPL and ZPL parts missing), because suheaven is AWOL - iOS audio output and video display, because author refuses the license change - Equalizer and compressor, because Ronald Wright is AWOL - Mono, Headphone and Dolby, because author refuses the license change - hqdn3d and yadif, because they are from MPlayer/libavfilter - remoteosd, because it derives from RealVNC code - MMX optimisations, because Ollie Lho, from SiS, is AWOL - Rotate, because it depends on GPL motion Nota Bene: - Some modules depend on GPL-only libraries, a LGPL module does not mean that the resulting binary module will be LGPL. Libraries affected would include liba52, libdvdcss, libdvdnav, libdvdread, faad2, libdca, libmad, libmpeg2, libpostproc, SRC, sid, zvbi and probably others.
14 years ago
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Lesser General Public License for more details.
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
*
LGPL Re-license almost all the playback modules to LGPLv2.1+ with authorization from their respective contributors (230+) This includes: - access, codec, packetizers, demux - audio filters, audio mixers, audio output - video filters, video chroma, video output - text renderers - XML parser - ARM NEON and SSE2 optimisations (mostly for chromas and filters) Some modules are not concerned: - BDA and DShow access modules because Manol Manolov is AWOL - Real RTSP, because it is derived from Xine - x264 and t140 because they are encoders only - DLL Loader, because it is derived from MPlayer - DTS packetizer, because Jon Lech Johansen is AWOL - Shine and WMAfixed, because they are derived from Rockbox - Real demuxer, as it is derived from MPlayer and Wang Bo is AWOL - MPC demuxer, as Yavor Doganov is AWOL - Tivo demuxer, because it is derived from an MPlayer fork - Playlist demuxer, (WPL and ZPL parts missing), because suheaven is AWOL - iOS audio output and video display, because author refuses the license change - Equalizer and compressor, because Ronald Wright is AWOL - Mono, Headphone and Dolby, because author refuses the license change - hqdn3d and yadif, because they are from MPlayer/libavfilter - remoteosd, because it derives from RealVNC code - MMX optimisations, because Ollie Lho, from SiS, is AWOL - Rotate, because it depends on GPL motion Nota Bene: - Some modules depend on GPL-only libraries, a LGPL module does not mean that the resulting binary module will be LGPL. Libraries affected would include liba52, libdvdcss, libdvdnav, libdvdread, faad2, libdca, libmad, libmpeg2, libpostproc, SRC, sid, zvbi and probably others.
14 years ago
* You should have received a copy of the GNU Lesser General Public License
* along with this program; if not, write to the Free Software Foundation,
* Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
*****************************************************************************/
video_converter: i420_rgb: lazy alloc conversion buffer Quote from A heap overflow world "4K ought to be enough for anybody"
8 years ago
#include <limits.h>
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
chroma: tidy plugin simd variant flags ...to use short labels like `PLUGIN_PLAIN` and `PLUGIN_SSE2` rather than `MODULE_NAME_IS_xxx_yyy[_zzz]`. this makes the code cleaner, easier to read, and improves consistency. it also helps remove a source of plugin vs. module confusion. trying to recognise that something like `MODULE_NAME_IS_i420_yuy2` refers to the 'plain' variant, compared to `MODULE_NAME_IS_i420_yuy2_sse2` for the SSE2 variant, was problematic. the `#else // defined(MODULE_NAME_IS_xxx_yyy_sse2)` instances have been changed to `#elif defined(PLUGIN_SSE2)` in the i420 plugin. the flag passed to the nv12 plugin was removed (rather than replaced with `PLUGIN_PLAIN`) because it is unused. the i420_rgb plugin used short `SSE2` type defines. these have been changed to `PLUGIN_SSE2` style for consistency.
5 years ago
#if !defined (PLUGIN_SSE2)
# define PLUGIN_PLAIN
i420_rgb: split files and clean up
13 years ago
#endif
render.c: RGB2 now gets the right color from the RGB color map. And a first cut at handling transparancy values properly. However to do this we merely needed to pixmap.[ch]: write our own routines for retrieving a close colormap index given a pixel value. Well, and also had to video_chroma/i420_rgb.[ch]: save the RGB colormap that is allocated. Modules.am: forgot new pixmap.c others: more misc abstraction/cleanup
22 years ago
/** Number of entries in RGB palette/colormap */
#define CMAP_RGB2_SIZE 256
/**
Chroma API change. Chromas are now normal video filters (almost).
18 years ago
* filter_sys_t: chroma method descriptor
render.c: RGB2 now gets the right color from the RGB color map. And a first cut at handling transparancy values properly. However to do this we merely needed to pixmap.[ch]: write our own routines for retrieving a close colormap index given a pixel value. Well, and also had to video_chroma/i420_rgb.[ch]: save the RGB colormap that is allocated. Modules.am: forgot new pixmap.c others: more misc abstraction/cleanup
22 years ago
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
* This structure is part of the chroma transformation descriptor, it
* describes the yuv2rgb specific properties.
render.c: RGB2 now gets the right color from the RGB color map. And a first cut at handling transparancy values properly. However to do this we merely needed to pixmap.[ch]: write our own routines for retrieving a close colormap index given a pixel value. Well, and also had to video_chroma/i420_rgb.[ch]: save the RGB colormap that is allocated. Modules.am: forgot new pixmap.c others: more misc abstraction/cleanup
22 years ago
*/
core: remove global *_sys_t typedefs Do not declare *_sys_t typedefs globally in vlc_common.h. Instead, declare them locally in each module that provides a definition. This paves the way to move C++ definitions into anonymous namespaces in order to respect C++ ODR. The picture_resource_t and sout_stream_id_sys_t typedefs will be handled separately, since they require specific additional changes. See #18033 Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
8 years ago
typedef struct
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
{
* all: fix some gcc warnings and removed all u8/u16/u32.
23 years ago
uint8_t *p_buffer;
video_converter: i420_rgb: lazy alloc conversion buffer Quote from A heap overflow world "4K ought to be enough for anybody"
8 years ago
size_t i_buffer_size;
uint8_t i_bytespp;
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
int *p_offset;
chroma: tidy plugin simd variant flags ...to use short labels like `PLUGIN_PLAIN` and `PLUGIN_SSE2` rather than `MODULE_NAME_IS_xxx_yyy[_zzz]`. this makes the code cleaner, easier to read, and improves consistency. it also helps remove a source of plugin vs. module confusion. trying to recognise that something like `MODULE_NAME_IS_i420_yuy2` refers to the 'plain' variant, compared to `MODULE_NAME_IS_i420_yuy2_sse2` for the SSE2 variant, was problematic. the `#else // defined(MODULE_NAME_IS_xxx_yyy_sse2)` instances have been changed to `#elif defined(PLUGIN_SSE2)` in the i420 plugin. the flag passed to the nv12 plugin was removed (rather than replaced with `PLUGIN_PLAIN`) because it is unused. the i420_rgb plugin used short `SSE2` type defines. these have been changed to `PLUGIN_SSE2` style for consistency.
5 years ago
#ifdef PLUGIN_PLAIN
render.c: RGB2 now gets the right color from the RGB color map. And a first cut at handling transparancy values properly. However to do this we merely needed to pixmap.[ch]: write our own routines for retrieving a close colormap index given a pixel value. Well, and also had to video_chroma/i420_rgb.[ch]: save the RGB colormap that is allocated. Modules.am: forgot new pixmap.c others: more misc abstraction/cleanup
22 years ago
/**< Pre-calculated conversion tables */
void *p_base; /**< base for all conversion tables */
uint8_t *p_rgb8; /**< RGB 8 bits table */
uint16_t *p_rgb16; /**< RGB 16 bits table */
uint32_t *p_rgb32; /**< RGB 32 bits table */
/**< To get RGB value for palette entry i, use (p_rgb_r[i], p_rgb_g[i],
p_rgb_b[i]). Note these are 16 bits per pixel. For 8bpp entries,
shift right 8 bits.
*/
uint16_t p_rgb_r[CMAP_RGB2_SIZE]; /**< Red values of palette */
uint16_t p_rgb_g[CMAP_RGB2_SIZE]; /**< Green values of palette */
uint16_t p_rgb_b[CMAP_RGB2_SIZE]; /**< Blue values of palette */
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
#endif
core: remove global *_sys_t typedefs Do not declare *_sys_t typedefs globally in vlc_common.h. Instead, declare them locally in each module that provides a definition. This paves the way to move C++ definitions into anonymous namespaces in order to respect C++ ODR. The picture_resource_t and sout_stream_id_sys_t typedefs will be handled separately, since they require specific additional changes. See #18033 Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
8 years ago
} filter_sys_t;
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
video_converter: i420_rgb: lazy alloc conversion buffer Quote from A heap overflow world "4K ought to be enough for anybody"
8 years ago
/*****************************************************************************
* Conversion buffer helper
*****************************************************************************/
static inline int AllocateOrGrow( uint8_t **pp_buffer, size_t *pi_buffer,
unsigned i_width, uint8_t bytespp )
{
if(UINT_MAX / bytespp < i_width)
return -1;
const size_t i_realloc = i_width * bytespp;
if(*pi_buffer >= i_realloc)
return 0;
uint8_t *p_realloc = realloc(*pp_buffer, i_realloc);
if(!p_realloc)
return -1;
*pp_buffer = p_realloc;
*pi_buffer = i_realloc;
return 0;
}
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
/*****************************************************************************
* Prototypes
*****************************************************************************/
chroma: tidy plugin simd variant flags ...to use short labels like `PLUGIN_PLAIN` and `PLUGIN_SSE2` rather than `MODULE_NAME_IS_xxx_yyy[_zzz]`. this makes the code cleaner, easier to read, and improves consistency. it also helps remove a source of plugin vs. module confusion. trying to recognise that something like `MODULE_NAME_IS_i420_yuy2` refers to the 'plain' variant, compared to `MODULE_NAME_IS_i420_yuy2_sse2` for the SSE2 variant, was problematic. the `#else // defined(MODULE_NAME_IS_xxx_yyy_sse2)` instances have been changed to `#elif defined(PLUGIN_SSE2)` in the i420 plugin. the flag passed to the nv12 plugin was removed (rather than replaced with `PLUGIN_PLAIN`) because it is unused. the i420_rgb plugin used short `SSE2` type defines. these have been changed to `PLUGIN_SSE2` style for consistency.
5 years ago
#ifdef PLUGIN_PLAIN
Chroma API change. Chromas are now normal video filters (almost).
18 years ago
void I420_RGB8 ( filter_t *, picture_t *, picture_t * );
void I420_RGB16 ( filter_t *, picture_t *, picture_t * );
void I420_RGB32 ( filter_t *, picture_t *, picture_t * );
i420_rgb: split files and clean up
13 years ago
#else
Chroma API change. Chromas are now normal video filters (almost).
18 years ago
void I420_R5G5B5 ( filter_t *, picture_t *, picture_t * );
void I420_R5G6B5 ( filter_t *, picture_t *, picture_t * );
void I420_A8R8G8B8 ( filter_t *, picture_t *, picture_t * );
void I420_R8G8B8A8 ( filter_t *, picture_t *, picture_t * );
void I420_B8G8R8A8 ( filter_t *, picture_t *, picture_t * );
void I420_A8B8G8R8 ( filter_t *, picture_t *, picture_t * );
- video_chromas: more SSE2 and MMX support and optimization, added SSE2 i420 -> RGB acceleration
19 years ago
#endif
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
/*****************************************************************************
* CONVERT_*_PIXEL: pixel conversion macros
*****************************************************************************
* These conversion routines are used by YUV conversion functions.
* conversion are made from p_y, p_u, p_v, which are modified, to p_buffer,
* which is also modified. CONVERT_4YUV_PIXEL is used for 8bpp dithering,
* CONVERT_4YUV_PIXEL_SCALE does the same but also scales the output.
*****************************************************************************/
#define CONVERT_Y_PIXEL( BPP ) \
/* Only Y sample is present */ \
p_ybase = p_yuv + *p_y++; \
*p_buffer++ = p_ybase[RED_OFFSET-((V_RED_COEF*128)>>SHIFT) + i_red] | \
p_ybase[GREEN_OFFSET-(((U_GREEN_COEF+V_GREEN_COEF)*128)>>SHIFT) \
+ i_green ] | p_ybase[BLUE_OFFSET-((U_BLUE_COEF*128)>>SHIFT) + i_blue];
#define CONVERT_YUV_PIXEL( BPP ) \
/* Y, U and V samples are present */ \
i_uval = *p_u++; \
i_vval = *p_v++; \
i_red = (V_RED_COEF * i_vval) >> SHIFT; \
i_green = (U_GREEN_COEF * i_uval + V_GREEN_COEF * i_vval) >> SHIFT; \
i_blue = (U_BLUE_COEF * i_uval) >> SHIFT; \
CONVERT_Y_PIXEL( BPP ) \
* ./modules/video_chroma/i420_rgb16.c: merged the RV15 and RV16 conversions because they're basically the same. Also, the MMX conversions now check that the bitmasks are the ones we expect.
24 years ago
#define CONVERT_Y_PIXEL_DITHER( BPP ) \
/* Only Y sample is present */ \
p_ybase = p_yuv + *p_y++; \
*p_buffer++ = p_ybase[RED_OFFSET-((V_RED_COEF*128+p_dither[i_real_y])>>SHIFT) + i_red] | \
p_ybase[GREEN_OFFSET-(((U_GREEN_COEF+V_GREEN_COEF)*128+p_dither[i_real_y])>>SHIFT) \
+ i_green ] | p_ybase[BLUE_OFFSET-((U_BLUE_COEF*128+p_dither[i_real_y])>>SHIFT) + i_blue];
#define CONVERT_YUV_PIXEL_DITHER( BPP ) \
/* Y, U and V samples are present */ \
i_uval = *p_u++; \
i_vval = *p_v++; \
i_red = (V_RED_COEF * i_vval) >> SHIFT; \
i_green = (U_GREEN_COEF * i_uval + V_GREEN_COEF * i_vval) >> SHIFT; \
i_blue = (U_BLUE_COEF * i_uval) >> SHIFT; \
CONVERT_Y_PIXEL_DITHER( BPP ) \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
#define CONVERT_4YUV_PIXEL( CHROMA ) \
*p_pic++ = p_lookup[ \
(((*p_y++ + dither10[i_real_y]) >> 4) << 7) \
+ ((*p_u + dither20[i_real_y]) >> 5) * 9 \
+ ((*p_v + dither20[i_real_y]) >> 5) ]; \
*p_pic++ = p_lookup[ \
(((*p_y++ + dither11[i_real_y]) >> 4) << 7) \
+ ((*p_u++ + dither21[i_real_y]) >> 5) * 9 \
+ ((*p_v++ + dither21[i_real_y]) >> 5) ]; \
*p_pic++ = p_lookup[ \
(((*p_y++ + dither12[i_real_y]) >> 4) << 7) \
+ ((*p_u + dither22[i_real_y]) >> 5) * 9 \
+ ((*p_v + dither22[i_real_y]) >> 5) ]; \
*p_pic++ = p_lookup[ \
(((*p_y++ + dither13[i_real_y]) >> 4) << 7) \
+ ((*p_u++ + dither23[i_real_y]) >> 5) * 9 \
+ ((*p_v++ + dither23[i_real_y]) >> 5) ]; \
#define CONVERT_4YUV_PIXEL_SCALE( CHROMA ) \
*p_pic++ = p_lookup[ \
( ((*p_y + dither10[i_real_y]) >> 4) << 7) \
+ ((*p_u + dither20[i_real_y]) >> 5) * 9 \
+ ((*p_v + dither20[i_real_y]) >> 5) ]; \
p_y += *p_offset++; \
p_u += *p_offset; \
p_v += *p_offset++; \
*p_pic++ = p_lookup[ \
( ((*p_y + dither11[i_real_y]) >> 4) << 7) \
+ ((*p_u + dither21[i_real_y]) >> 5) * 9 \
+ ((*p_v + dither21[i_real_y]) >> 5) ]; \
p_y += *p_offset++; \
p_u += *p_offset; \
p_v += *p_offset++; \
*p_pic++ = p_lookup[ \
( ((*p_y + dither12[i_real_y]) >> 4) << 7) \
+ ((*p_u + dither22[i_real_y]) >> 5) * 9 \
+ ((*p_v + dither22[i_real_y]) >> 5) ]; \
p_y += *p_offset++; \
p_u += *p_offset; \
p_v += *p_offset++; \
*p_pic++ = p_lookup[ \
( ((*p_y + dither13[i_real_y]) >> 4) << 7) \
+ ((*p_u + dither23[i_real_y]) >> 5) * 9 \
+ ((*p_v + dither23[i_real_y]) >> 5) ]; \
p_y += *p_offset++; \
p_u += *p_offset; \
p_v += *p_offset++; \
/*****************************************************************************
* SCALE_WIDTH: scale a line horizontally
*****************************************************************************
* This macro scales a line using rendering buffer and offset array. It works
* for 1, 2 and 4 Bpp.
*****************************************************************************/
#define SCALE_WIDTH \
if( b_hscale ) \
{ \
/* Horizontal scaling, conversion has been done to buffer. \
* Rewind buffer and offset, then copy and scale line */ \
p_buffer = p_buffer_start; \
p_offset = p_offset_start; \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
for( i_x = (p_filter->fmt_out.video.i_x_offset + p_filter->fmt_out.video.i_visible_width) / 16; i_x--; ) \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
{ \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
} \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
for( i_x = (p_filter->fmt_out.video.i_x_offset + p_filter->fmt_out.video.i_visible_width) & 15; i_x--; ) \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
{ \
*p_pic++ = *p_buffer; p_buffer += *p_offset++; \
} \
* *.h: shut up gcc (cast of lvalue).
22 years ago
p_pic = (void*)((uint8_t*)p_pic + i_right_margin ); \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
} \
else \
{ \
/* No scaling, conversion has been done directly in picture memory. \
* Increment of picture pointer to end of line is still needed */ \
* *.h: shut up gcc (cast of lvalue).
22 years ago
p_pic = (void*)((uint8_t*)p_pic + p_dest->p->i_pitch ); \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
} \
/*****************************************************************************
* SCALE_WIDTH_DITHER: scale a line horizontally for dithered 8 bpp
*****************************************************************************
* This macro scales a line using an offset array.
*****************************************************************************/
#define SCALE_WIDTH_DITHER( CHROMA ) \
if( b_hscale ) \
{ \
/* Horizontal scaling - we can't use a buffer due to dithering */ \
p_offset = p_offset_start; \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
for( i_x = (p_filter->fmt_out.video.i_x_offset + p_filter->fmt_out.video.i_visible_width) / 16; i_x--; ) \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
{ \
CONVERT_4YUV_PIXEL_SCALE( CHROMA ) \
CONVERT_4YUV_PIXEL_SCALE( CHROMA ) \
CONVERT_4YUV_PIXEL_SCALE( CHROMA ) \
CONVERT_4YUV_PIXEL_SCALE( CHROMA ) \
} \
} \
else \
{ \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
for( i_x = (p_filter->fmt_in.video.i_x_offset + p_filter->fmt_in.video.i_visible_width) / 16; i_x--; ) \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
{ \
CONVERT_4YUV_PIXEL( CHROMA ) \
CONVERT_4YUV_PIXEL( CHROMA ) \
CONVERT_4YUV_PIXEL( CHROMA ) \
CONVERT_4YUV_PIXEL( CHROMA ) \
} \
} \
/* Increment of picture pointer to end of line is still needed */ \
* *.h: shut up gcc (cast of lvalue).
22 years ago
p_pic = (void*)((uint8_t*)p_pic + i_right_margin ); \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
\
/* Increment the Y coordinate in the matrix, modulo 4 */ \
i_real_y = (i_real_y + 1) & 0x3; \
/*****************************************************************************
* SCALE_HEIGHT: handle vertical scaling
*****************************************************************************
* This macro handle vertical scaling for a picture. CHROMA may be 420, 422 or
* 444 for RGB conversion, or 400 for gray conversion. It works for 1, 2, 3
* and 4 Bpp.
*****************************************************************************/
#define SCALE_HEIGHT( CHROMA, BPP ) \
/* If line is odd, rewind 4:2:0 U and V samples */ \
if( ((CHROMA == 420) || (CHROMA == 422)) && !(i_y & 0x1) ) \
{ \
p_u -= i_chroma_width; \
p_v -= i_chroma_width; \
} \
\
/* \
* Handle vertical scaling. The current line can be copied or next one \
* can be ignored. \
*/ \
switch( i_vscale ) \
{ \
case -1: /* vertical scaling factor is < 1 */ \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
while( (i_scale_count -= (p_filter->fmt_out.video.i_y_offset + p_filter->fmt_out.video.i_visible_height)) > 0 ) \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
{ \
/* Height reduction: skip next source line */ \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
p_y += (p_filter->fmt_in.video.i_x_offset + p_filter->fmt_in.video.i_visible_width); \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
i_y++; \
if( (CHROMA == 420) || (CHROMA == 422) ) \
{ \
if( i_y & 0x1 ) \
{ \
p_u += i_chroma_width; \
p_v += i_chroma_width; \
} \
} \
else if( CHROMA == 444 ) \
{ \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
p_u += (p_filter->fmt_in.video.i_x_offset + p_filter->fmt_in.video.i_visible_width); \
p_v += (p_filter->fmt_in.video.i_x_offset + p_filter->fmt_in.video.i_visible_width); \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
} \
} \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
i_scale_count += (p_filter->fmt_in.video.i_y_offset + p_filter->fmt_in.video.i_visible_height); \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
break; \
case 1: /* vertical scaling factor is > 1 */ \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
while( (i_scale_count -= (p_filter->fmt_in.video.i_y_offset + p_filter->fmt_in.video.i_visible_height)) > 0 ) \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
{ \
/* Height increment: copy previous picture line */ \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
memcpy( p_pic, p_pic_start, (p_filter->fmt_out.video.i_x_offset + p_filter->fmt_out.video.i_visible_width) * BPP ); \
* *.h: shut up gcc (cast of lvalue).
22 years ago
p_pic = (void*)((uint8_t*)p_pic + p_dest->p->i_pitch ); \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
} \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
i_scale_count += (p_filter->fmt_out.video.i_y_offset + p_filter->fmt_out.video.i_visible_height); \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
break; \
} \
/*****************************************************************************
* SCALE_HEIGHT_DITHER: handle vertical scaling for dithered 8 bpp
*****************************************************************************
* This macro handles vertical scaling for a picture. CHROMA may be 420,
* 422 or 444 for RGB conversion, or 400 for gray conversion.
*****************************************************************************/
#define SCALE_HEIGHT_DITHER( CHROMA ) \
\
/* If line is odd, rewind 4:2:0 U and V samples */ \
if( ((CHROMA == 420) || (CHROMA == 422)) && !(i_y & 0x1) ) \
{ \
p_u -= i_chroma_width; \
p_v -= i_chroma_width; \
} \
\
/* \
* Handle vertical scaling. The current line can be copied or next one \
* can be ignored. \
*/ \
\
switch( i_vscale ) \
{ \
case -1: /* vertical scaling factor is < 1 */ \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
while( (i_scale_count -= (p_filter->fmt_out.video.i_y_offset + p_filter->fmt_out.video.i_visible_height)) > 0 ) \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
{ \
/* Height reduction: skip next source line */ \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
p_y += (p_filter->fmt_in.video.i_x_offset + p_filter->fmt_in.video.i_visible_width); \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
i_y++; \
if( (CHROMA == 420) || (CHROMA == 422) ) \
{ \
if( i_y & 0x1 ) \
{ \
p_u += i_chroma_width; \
p_v += i_chroma_width; \
} \
} \
else if( CHROMA == 444 ) \
{ \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
p_u += (p_filter->fmt_in.video.i_x_offset + p_filter->fmt_in.video.i_visible_width); \
p_v += (p_filter->fmt_in.video.i_x_offset + p_filter->fmt_in.video.i_visible_width); \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
} \
} \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
i_scale_count += (p_filter->fmt_in.video.i_y_offset + p_filter->fmt_in.video.i_visible_height); \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
break; \
case 1: /* vertical scaling factor is > 1 */ \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
while( (i_scale_count -= (p_filter->fmt_in.video.i_y_offset + p_filter->fmt_in.video.i_visible_height)) > 0 ) \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
{ \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
p_y -= (p_filter->fmt_in.video.i_x_offset + p_filter->fmt_in.video.i_visible_width); \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
p_u -= i_chroma_width; \
p_v -= i_chroma_width; \
SCALE_WIDTH_DITHER( CHROMA ); \
} \
i420_rgb: only handle visible pixels fixes a crash similar to #16686 when an RGB output is used instead of YUYV Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
10 years ago
i_scale_count += (p_filter->fmt_out.video.i_y_offset + p_filter->fmt_out.video.i_visible_height); \
* ./modules/*: moved plugins to the new tree. Yet untested builds include waveout, directx, qnx, beos, win32, macosx, and the AltiVec modules. * ALL: removed mention of AC3 in favour of A52. * ./configure.in, ./Makefile*: modules can now be built deeper than 1 directory. As a consequence, the build is even slower (but I'm fixing this) and make clean doesn't work anymore.
24 years ago
break; \
} \