* 1,4,8bpp support and context / deglobalize stuff
* by Michael Niedermayer (michaelni@gmx.at)
*
- * This file is part of FFmpeg.
+ * This file is part of Libav.
*
- * FFmpeg is free software; you can redistribute it and/or
+ * Libav 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 (at your option) any later version.
*
- * FFmpeg is distributed in the hope that it will be useful,
+ * Libav is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with FFmpeg; if not, write to the Free Software
+ * License along with Libav; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "rgb2rgb.h"
#include "swscale.h"
#include "swscale_internal.h"
-#include "libavutil/x86_cpu.h"
+#include "libavutil/cpu.h"
#include "libavutil/bswap.h"
extern const uint8_t dither_4x4_16[4][8];
dst[12*i+ 8] = dst[12*i+ 9] = g[Y]; \
dst[12*i+10] = dst[12*i+11] = b[Y];
+#define PUTBGR48(dst,src,i) \
+ Y = src[2*i]; \
+ dst[12*i+ 0] = dst[12*i+ 1] = b[Y]; \
+ dst[12*i+ 2] = dst[12*i+ 3] = g[Y]; \
+ dst[12*i+ 4] = dst[12*i+ 5] = r[Y]; \
+ Y = src[2*i+1]; \
+ dst[12*i+ 6] = dst[12*i+ 7] = b[Y]; \
+ dst[12*i+ 8] = dst[12*i+ 9] = g[Y]; \
+ dst[12*i+10] = dst[12*i+11] = r[Y];
+
#define YUV2RGBFUNC(func_name, dst_type, alpha) \
static int func_name(SwsContext *c, const uint8_t* src[], int srcStride[], int srcSliceY, \
int srcSliceH, uint8_t* dst[], int dstStride[]) \
PUTRGB48(dst_1,py_1,1);
ENDYUV2RGBFUNC()
+YUV2RGBFUNC(yuv2rgb_c_bgr48, uint8_t, 0)
+ LOADCHROMA(0);
+ PUTBGR48(dst_1,py_1,0);
+ PUTBGR48(dst_2,py_2,0);
+
+ LOADCHROMA(1);
+ PUTBGR48(dst_2,py_2,1);
+ PUTBGR48(dst_1,py_1,1);
+
+ LOADCHROMA(2);
+ PUTBGR48(dst_1,py_1,2);
+ PUTBGR48(dst_2,py_2,2);
+
+ LOADCHROMA(3);
+ PUTBGR48(dst_2,py_2,3);
+ PUTBGR48(dst_1,py_1,3);
+ENDYUV2RGBLINE(48)
+ LOADCHROMA(0);
+ PUTBGR48(dst_1,py_1,0);
+ PUTBGR48(dst_2,py_2,0);
+
+ LOADCHROMA(1);
+ PUTBGR48(dst_2,py_2,1);
+ PUTBGR48(dst_1,py_1,1);
+ENDYUV2RGBFUNC()
+
YUV2RGBFUNC(yuv2rgb_c_32, uint32_t, 0)
LOADCHROMA(0);
PUTRGB(dst_1,py_1,0);
PUTRGB(dst_1,py_1,3);
CLOSEYUV2RGBFUNC(8)
-#if 0 // Currently unused
-// This is exactly the same code as yuv2rgb_c_32 except for the types of
-// r, g, b, dst_1, dst_2
-YUV2RGBFUNC(yuv2rgb_c_8, uint8_t, 0)
- LOADCHROMA(0);
- PUTRGB(dst_1,py_1,0);
- PUTRGB(dst_2,py_2,0);
-
- LOADCHROMA(1);
- PUTRGB(dst_2,py_2,1);
- PUTRGB(dst_1,py_1,1);
-
- LOADCHROMA(2);
- PUTRGB(dst_1,py_1,2);
- PUTRGB(dst_2,py_2,2);
-
- LOADCHROMA(3);
- PUTRGB(dst_2,py_2,3);
- PUTRGB(dst_1,py_1,3);
-CLOSEYUV2RGBFUNC(8)
-#endif
-
// r, g, b, dst_1, dst_2
YUV2RGBFUNC(yuv2rgb_c_12_ordered_dither, uint16_t, 0)
const uint8_t *d16 = dither_4x4_16[y&3];
PUTRGB8(dst_1,py_1,3,6);
CLOSEYUV2RGBFUNC(8)
-#if 0 // Currently unused
-// This is exactly the same code as yuv2rgb_c_32 except for the types of
-// r, g, b, dst_1, dst_2
-YUV2RGBFUNC(yuv2rgb_c_4, uint8_t, 0)
- int acc;
-#define PUTRGB4(dst,src,i) \
- Y = src[2*i]; \
- acc = r[Y] + g[Y] + b[Y]; \
- Y = src[2*i+1]; \
- acc |= (r[Y] + g[Y] + b[Y])<<4; \
- dst[i] = acc;
-
- LOADCHROMA(0);
- PUTRGB4(dst_1,py_1,0);
- PUTRGB4(dst_2,py_2,0);
-
- LOADCHROMA(1);
- PUTRGB4(dst_2,py_2,1);
- PUTRGB4(dst_1,py_1,1);
-
- LOADCHROMA(2);
- PUTRGB4(dst_1,py_1,2);
- PUTRGB4(dst_2,py_2,2);
-
- LOADCHROMA(3);
- PUTRGB4(dst_2,py_2,3);
- PUTRGB4(dst_1,py_1,3);
-CLOSEYUV2RGBFUNC(4)
-#endif
-
YUV2RGBFUNC(yuv2rgb_c_4_ordered_dither, uint8_t, 0)
const uint8_t *d64 = dither_8x8_73[y&7];
const uint8_t *d128 = dither_8x8_220[y&7];
PUTRGB4D(dst_1,py_1,3,6);
CLOSEYUV2RGBFUNC(4)
-#if 0 // Currently unused
-// This is exactly the same code as yuv2rgb_c_32 except for the types of
-// r, g, b, dst_1, dst_2
-YUV2RGBFUNC(yuv2rgb_c_4b, uint8_t, 0)
- LOADCHROMA(0);
- PUTRGB(dst_1,py_1,0);
- PUTRGB(dst_2,py_2,0);
-
- LOADCHROMA(1);
- PUTRGB(dst_2,py_2,1);
- PUTRGB(dst_1,py_1,1);
-
- LOADCHROMA(2);
- PUTRGB(dst_1,py_1,2);
- PUTRGB(dst_2,py_2,2);
-
- LOADCHROMA(3);
- PUTRGB(dst_2,py_2,3);
- PUTRGB(dst_1,py_1,3);
-CLOSEYUV2RGBFUNC(8)
-#endif
-
YUV2RGBFUNC(yuv2rgb_c_4b_ordered_dither, uint8_t, 0)
const uint8_t *d64 = dither_8x8_73[y&7];
const uint8_t *d128 = dither_8x8_220[y&7];
SwsFunc ff_yuv2rgb_get_func_ptr(SwsContext *c)
{
SwsFunc t = NULL;
-#if (HAVE_MMX2 || HAVE_MMX) && CONFIG_GPL
- t = ff_yuv2rgb_init_mmx(c);
-#endif
-#if HAVE_VIS
- t = ff_yuv2rgb_init_vis(c);
-#endif
-#if CONFIG_MLIB
- t = ff_yuv2rgb_init_mlib(c);
-#endif
-#if HAVE_ALTIVEC
- if (c->flags & SWS_CPU_CAPS_ALTIVEC)
- t = ff_yuv2rgb_init_altivec(c);
-#endif
-#if ARCH_BFIN
- if (c->flags & SWS_CPU_CAPS_BFIN)
+ if (HAVE_MMX) {
+ t = ff_yuv2rgb_init_mmx(c);
+ } else if (HAVE_VIS) {
+ t = ff_yuv2rgb_init_vis(c);
+ } else if (CONFIG_MLIB) {
+ t = ff_yuv2rgb_init_mlib(c);
+ } else if (HAVE_ALTIVEC) {
+ t = ff_yuv2rgb_init_altivec(c);
+ } else if (ARCH_BFIN) {
t = ff_yuv2rgb_get_func_ptr_bfin(c);
-#endif
+ }
if (t)
return t;
av_log(c, AV_LOG_WARNING, "No accelerated colorspace conversion found from %s to %s.\n", sws_format_name(c->srcFormat), sws_format_name(c->dstFormat));
switch (c->dstFormat) {
+ case PIX_FMT_BGR48BE:
+ case PIX_FMT_BGR48LE: return yuv2rgb_c_bgr48;
case PIX_FMT_RGB48BE:
case PIX_FMT_RGB48LE: return yuv2rgb_c_48;
case PIX_FMT_ARGB:
return NULL;
}
-static void fill_table(uint8_t* table[256], const int elemsize, const int inc, uint8_t *y_table)
+static void fill_table(uint8_t* table[256], const int elemsize, const int inc, void *y_tab)
{
int i;
int64_t cb = 0;
+ uint8_t *y_table = y_tab;
y_table -= elemsize * (inc >> 9);
}
}
+static uint16_t roundToInt16(int64_t f)
+{
+ int r= (f + (1<<15))>>16;
+ if (r<-0x7FFF) return 0x8000;
+ else if (r> 0x7FFF) return 0x7FFF;
+ else return r;
+}
+
av_cold int ff_yuv2rgb_c_init_tables(SwsContext *c, const int inv_table[4], int fullRange,
int brightness, int contrast, int saturation)
{
|| c->dstFormat==PIX_FMT_RGB565LE
|| c->dstFormat==PIX_FMT_RGB555BE
|| c->dstFormat==PIX_FMT_RGB555LE
- || c->dstFormat==PIX_FMT_RGB444
+ || c->dstFormat==PIX_FMT_RGB444BE
+ || c->dstFormat==PIX_FMT_RGB444LE
|| c->dstFormat==PIX_FMT_RGB8
|| c->dstFormat==PIX_FMT_RGB4
|| c->dstFormat==PIX_FMT_RGB4_BYTE
|| c->dstFormat==PIX_FMT_MONOBLACK;
const int isNotNe = c->dstFormat==PIX_FMT_NE(RGB565LE,RGB565BE)
|| c->dstFormat==PIX_FMT_NE(RGB555LE,RGB555BE)
+ || c->dstFormat==PIX_FMT_NE(RGB444LE,RGB444BE)
|| c->dstFormat==PIX_FMT_NE(BGR565LE,BGR565BE)
- || c->dstFormat==PIX_FMT_NE(BGR555LE,BGR555BE);
+ || c->dstFormat==PIX_FMT_NE(BGR555LE,BGR555BE)
+ || c->dstFormat==PIX_FMT_NE(BGR444LE,BGR444BE);
const int bpp = c->dstFormatBpp;
uint8_t *y_table;
uint16_t *y_table16;
cgv = (cgv*contrast * saturation) >> 32;
oy -= 256*brightness;
+ c->uOffset= 0x0400040004000400LL;
+ c->vOffset= 0x0400040004000400LL;
+ c->yCoeff= roundToInt16(cy *8192) * 0x0001000100010001ULL;
+ c->vrCoeff= roundToInt16(crv*8192) * 0x0001000100010001ULL;
+ c->ubCoeff= roundToInt16(cbu*8192) * 0x0001000100010001ULL;
+ c->vgCoeff= roundToInt16(cgv*8192) * 0x0001000100010001ULL;
+ c->ugCoeff= roundToInt16(cgu*8192) * 0x0001000100010001ULL;
+ c->yOffset= roundToInt16(oy * 8) * 0x0001000100010001ULL;
+
+ c->yuv2rgb_y_coeff = (int16_t)roundToInt16(cy <<13);
+ c->yuv2rgb_y_offset = (int16_t)roundToInt16(oy << 9);
+ c->yuv2rgb_v2r_coeff= (int16_t)roundToInt16(crv<<13);
+ c->yuv2rgb_v2g_coeff= (int16_t)roundToInt16(cgv<<13);
+ c->yuv2rgb_u2g_coeff= (int16_t)roundToInt16(cgu<<13);
+ c->yuv2rgb_u2b_coeff= (int16_t)roundToInt16(cbu<<13);
+
//scale coefficients by cy
crv = ((crv << 16) + 0x8000) / cy;
cbu = ((cbu << 16) + 0x8000) / cy;
y_table16[i+2048] = (yval >> 4) << bbase;
yb += cy;
}
+ if (isNotNe)
+ for (i = 0; i < 1024*3; i++)
+ y_table16[i] = av_bswap16(y_table16[i]);
fill_table(c->table_rV, 2, crv, y_table16 + yoffs);
fill_table(c->table_gU, 2, cgu, y_table16 + yoffs + 1024);
fill_table(c->table_bU, 2, cbu, y_table16 + yoffs + 2048);
}
if(isNotNe)
for (i = 0; i < 1024*3; i++)
- y_table16[i] = bswap_16(y_table16[i]);
+ y_table16[i] = av_bswap16(y_table16[i]);
fill_table(c->table_rV, 2, crv, y_table16 + yoffs);
fill_table(c->table_gU, 2, cgu, y_table16 + yoffs + 1024);
fill_table(c->table_bU, 2, cbu, y_table16 + yoffs + 2048);
y_table32 = c->yuvTable;
yb = -(384<<16) - oy;
for (i = 0; i < 1024; i++) {
- uint8_t yval = av_clip_uint8((yb + 0x8000) >> 16);
- y_table32[i ] = (yval << rbase) + (needAlpha ? 0 : (255 << abase));
+ unsigned yval = av_clip_uint8((yb + 0x8000) >> 16);
+ y_table32[i ] = (yval << rbase) + (needAlpha ? 0 : (255u << abase));
y_table32[i+1024] = yval << gbase;
y_table32[i+2048] = yval << bbase;
yb += cy;
projects / ffmpeg / blobdiff