1
/*****************************************************************************
2
* yuv_rgb.c : ARM NEONv1 YUV to RGB32 chroma conversion for VLC
3
*****************************************************************************
4
* Copyright (C) 2011 Sébastien Toque
7
* This program is free software; you can redistribute it and/or modify
8
* it under the terms of the GNU General Public License as published by
9
* the Free Software Foundation; either version 2 of the License, or
10
* (at your option) any later version.
12
* This program is distributed in the hope that it will be useful,
13
* but WITHOUT ANY WARRANTY; without even the implied warranty of
14
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15
* GNU General Public License for more details.
17
* You should have received a copy of the GNU General Public License
18
* along with this program; if not, write to the Free Software
19
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston MA 02110-1301, USA.
20
*****************************************************************************/
26
#include <vlc_common.h>
27
#include <vlc_plugin.h>
28
#include <vlc_filter.h>
30
#include "chroma_neon.h"
32
static int Open (vlc_object_t *);
35
set_description (N_("ARM NEON video chroma YUV->RGBA"))
36
set_capability ("video filter2", 250)
37
set_callbacks (Open, NULL)
41
static int CoefY[256];
42
static int CoefRV[256];
43
static int CoefGU[256];
44
static int CoefGV[256];
45
static int CoefBU[256];
47
// C reference version of the converter
48
static void I420_RGBA_C (filter_t *filter, picture_t *src, picture_t *dst)
50
const uint8_t *const out = dst->p->p_pixels;
51
const size_t width = src->p[Y_PLANE].i_visible_pitch;
52
const size_t height = src->p[Y_PLANE].i_visible_lines;
54
const int ypitch = src->p[Y_PLANE].i_pitch;
55
const int uvpitch = src->p[U_PLANE].i_pitch;
56
const int dpitch = dst->p->i_pitch / dst->p->i_pixel_pitch;
58
for (size_t j = 0; j < height; ++j)
60
const int y = j * ypitch;
61
const int u = (j>>1) * uvpitch;
62
const int d = j * dpitch;
64
for (size_t i = 0; i < width; ++i)
66
uint8_t Y = src->Y_PIXELS[y + i];
67
uint8_t U = src->U_PIXELS[u + (i>>1)];
68
uint8_t V = src->V_PIXELS[u + (i>>1)];
70
//coef = float * Precision + .5 (Precision=32768)
71
int R = CoefY[Y] + CoefRV[V];
72
int G = CoefY[Y] + CoefGU[U] + CoefGV[V];
73
int B = CoefY[Y] + CoefBU[U];
75
//rgb = (rgb+Precision/2) / Precision (Precision=32768)
80
if (unlikely(R < 0)) R = 0;
81
if (unlikely(G < 0)) G = 0;
82
if (unlikely(B < 0)) B = 0;
83
if (unlikely(R > 255)) R = 255;
84
if (unlikely(G > 255)) G = 255;
85
if (unlikely(B > 255)) B = 255;
87
((uint32_t*)out)[d + i] = R | (G<<8) | (B<<16) | (0xff<<24);
92
static void I420_RGBA (filter_t *filter, picture_t *src, picture_t *dst)
94
struct yuv_pack out = { dst->p->p_pixels, dst->p->i_pitch };
95
struct yuv_planes in = { src->Y_PIXELS, src->U_PIXELS, src->V_PIXELS, src->Y_PITCH };
96
i420_rgb_neon (&out, &in, filter->fmt_in.video.i_width, filter->fmt_in.video.i_height);
98
static void YV12_RGBA (filter_t *filter, picture_t *src, picture_t *dst)
100
struct yuv_pack out = { dst->p->p_pixels, dst->p->i_pitch };
101
struct yuv_planes in = { src->Y_PIXELS, src->V_PIXELS, src->U_PIXELS, src->Y_PITCH };
102
i420_rgb_neon (&out, &in, filter->fmt_in.video.i_width, filter->fmt_in.video.i_height);
105
static void NV21_RGBA (filter_t *filter, picture_t *src, picture_t *dst)
107
struct yuv_pack out = { dst->p->p_pixels, dst->p->i_pitch };
108
struct yuv_planes in = { src->Y_PIXELS, src->U_PIXELS, src->V_PIXELS, src->Y_PITCH };
109
nv21_rgb_neon (&out, &in, filter->fmt_in.video.i_width, filter->fmt_in.video.i_height);
112
static void NV12_RGBA (filter_t *filter, picture_t *src, picture_t *dst)
114
struct yuv_pack out = { dst->p->p_pixels, dst->p->i_pitch };
115
struct yuv_planes in = { src->Y_PIXELS, src->U_PIXELS, src->V_PIXELS, src->Y_PITCH };
116
nv12_rgb_neon (&out, &in, filter->fmt_in.video.i_width, filter->fmt_in.video.i_height);
119
VIDEO_FILTER_WRAPPER (I420_RGBA)
120
VIDEO_FILTER_WRAPPER (YV12_RGBA)
121
VIDEO_FILTER_WRAPPER (NV21_RGBA)
122
VIDEO_FILTER_WRAPPER (NV12_RGBA)
124
static int Open (vlc_object_t *obj)
126
filter_t *filter = (filter_t *)obj;
128
if (!(vlc_CPU() & CPU_CAPABILITY_NEON))
131
if (((filter->fmt_in.video.i_width | filter->fmt_in.video.i_height) & 1)
132
|| (filter->fmt_in.video.i_width != filter->fmt_out.video.i_width)
133
|| (filter->fmt_in.video.i_height != filter->fmt_out.video.i_height))
136
switch (filter->fmt_out.video.i_chroma)
138
case VLC_CODEC_RGB32:
139
if( filter->fmt_out.video.i_rmask != 0x000000ff
140
|| filter->fmt_out.video.i_gmask != 0x0000ff00
141
|| filter->fmt_out.video.i_bmask != 0x00ff0000 )
144
switch (filter->fmt_in.video.i_chroma)
147
filter->pf_video_filter = I420_RGBA_Filter;
150
filter->pf_video_filter = YV12_RGBA_Filter;
153
filter->pf_video_filter = NV21_RGBA_Filter;
156
filter->pf_video_filter = NV12_RGBA_Filter;
167
//precompute some values for the C version
168
/*const int coefY = (int)(1.164 * 32768 + 0.5);
169
const int coefRV = (int)(1.793 * 32768 + 0.5);
170
const int coefGU = (int)(0.213 * 32768 + 0.5);
171
const int coefGV = (int)(0.533 * 32768 + 0.5);
172
const int coefBU = (int)(2.113 * 32768 + 0.5);
173
for (int i=0; i<256; ++i)
175
CoefY[i] = coefY * (i-16) + 16384;
176
CoefRV[i] = coefRV*(i-128);
177
CoefGU[i] = -coefGU*(i-128);
178
CoefGV[i] = -coefGV*(i-128);
179
CoefBU[i] = coefBU*(i-128);
182
msg_Dbg(filter, "%4.4s(%dx%d) to %4.4s(%dx%d)",
183
(char*)&filter->fmt_in.video.i_chroma, filter->fmt_in.video.i_width, filter->fmt_in.video.i_height,
184
(char*)&filter->fmt_out.video.i_chroma, filter->fmt_out.video.i_width, filter->fmt_out.video.i_height);