Mercurial > mplayer.hg

comparison libswscale/internal_bfin.S @ 27158:65b8334df960

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
spelling/grammar/wording overhaul
author diego
date Fri, 04 Jul 2008 13:49:45 +0000
parents 23f1738030fc
children
comparison
equal deleted inserted replaced
27157:e2797c291ba9 27158:65b8334df960
1 /* 1 /*
2 * Copyright (C) 2007 Marc Hoffman <marc.hoffman@analog.com> 2 * Copyright (C) 2007 Marc Hoffman <marc.hoffman@analog.com>
3 * April 20, 2007 3 * April 20, 2007
4 * 4 *
5 * Blackfin Video Color Space Converters Operations 5 * Blackfin video color space converter operations
6 * convert I420 YV12 to RGB in various formats, 6 * convert I420 YV12 to RGB in various formats
7 * 7 *
8 * This file is part of FFmpeg. 8 * This file is part of FFmpeg.
9 * 9 *
10 * FFmpeg is free software; you can redistribute it and/or 10 * FFmpeg is free software; you can redistribute it and/or
11 * modify it under the terms of the GNU Lesser General Public 11 * modify it under the terms of the GNU Lesser General Public
22 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA 22 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
23 */ 23 */
24 24
25 25
26 /* 26 /*
27 YUV420 to RGB565 conversion. This routine takes a YUV 420 planar macroblock 27 YUV420 to RGB565 conversion. This routine takes a YUV 420 planar macroblock
28 and converts it to RGB565. R:5 bits, G:6 bits, B:5 bits.. packed into shorts 28 and converts it to RGB565. R:5 bits, G:6 bits, B:5 bits.. packed into shorts.
29 29
30 30
31 The following calculation is used for the conversion: 31 The following calculation is used for the conversion:
32 32
33 r = clipz((y-oy)*cy + crv*(v-128)) 33 r = clipz((y-oy)*cy + crv*(v-128))
34 g = clipz((y-oy)*cy + cgv*(v-128) + cgu*(u-128)) 34 g = clipz((y-oy)*cy + cgv*(v-128) + cgu*(u-128))
35 b = clipz((y-oy)*cy + cbu*(u-128)) 35 b = clipz((y-oy)*cy + cbu*(u-128))
36 36
37 y,u,v are pre scaled by a factor of 4 i.e. left shifted to gain precision. 37 y,u,v are prescaled by a factor of 4 i.e. left-shifted to gain precision.
38 38
39 39
40 New factorization to eliminate the truncation error which was 40 New factorization to eliminate the truncation error which was
41 occuring due to the byteop3p. 41 occurring due to the byteop3p.
42 42
43 43
44 1) use the bytop16m to subtract quad bytes we use this in U8 this 44 1) Use the bytop16m to subtract quad bytes we use this in U8 this
45 then so the offsets need to be renormalized to 8bits. 45 then so the offsets need to be renormalized to 8bits.
46 46
47 2) scale operands up by a factor of 4 not 8 because Blackfin 47 2) Scale operands up by a factor of 4 not 8 because Blackfin
48 multiplies include a shift. 48 multiplies include a shift.
49 49
50 3) compute into the accumulators cy*yx0, cy*yx1 50 3) Compute into the accumulators cy*yx0, cy*yx1.
51 51
52 4) compute each of the linear equations 52 4) Compute each of the linear equations:
53 r = clipz((y - oy) * cy + crv * (v - 128)) 53 r = clipz((y - oy) * cy + crv * (v - 128))
54 54
55 g = clipz((y - oy) * cy + cgv * (v - 128) + cgu * (u - 128)) 55 g = clipz((y - oy) * cy + cgv * (v - 128) + cgu * (u - 128))
56 56
57 b = clipz((y - oy) * cy + cbu * (u - 128)) 57 b = clipz((y - oy) * cy + cbu * (u - 128))
58 58
59 reuse of the accumulators requires that we actually multiply 59 Reuse of the accumulators requires that we actually multiply
60 twice once with addition and the second time with a subtaction. 60 twice once with addition and the second time with a subtraction.
61 61
62 because of this we need to compute the equations in the order R B 62 Because of this we need to compute the equations in the order R B
63 then G saving the writes for B in the case of 24/32 bit color 63 then G saving the writes for B in the case of 24/32 bit color
64 formats. 64 formats.
65 65
66 api: yuv2rgb_kind (uint8_t *Y, uint8_t *U, uint8_t *V, int *out, 66 API: yuv2rgb_kind (uint8_t *Y, uint8_t *U, uint8_t *V, int *out,
67 int dW, uint32_t *coeffs); 67 int dW, uint32_t *coeffs);
68 68
69 A B 69 A B
70 --- --- 70 --- ---
71 i2 = cb i3 = cr 71 i2 = cb i3 = cr
75 75
76 uint32_t oy,oc,zero,cy,crv,rmask,cbu,bmask,cgu,cgv; 76 uint32_t oy,oc,zero,cy,crv,rmask,cbu,bmask,cgu,cgv;
77 77
78 coeffs is a pointer to oy. 78 coeffs is a pointer to oy.
79 79
80 the {rgb} masks are only utilized by the 565 packing algorithm. Note the data 80 The {rgb} masks are only utilized by the 565 packing algorithm. Note the data
81 replication is used to simplify the internal algorithms for the dual mac architecture 81 replication is used to simplify the internal algorithms for the dual Mac
82 of BlackFin. 82 architecture of BlackFin.
83 83
84 All routines are exported with _ff_bfin_ as a symbol prefix 84 All routines are exported with _ff_bfin_ as a symbol prefix.
85 85
86 rough performance gain compared against -O3: 86 Rough performance gain compared against -O3:
87 87
88 2779809/1484290 187.28% 88 2779809/1484290 187.28%
89 89
90 which translates to ~33c/pel to ~57c/pel for the reference vs 17.5 90 which translates to ~33c/pel to ~57c/pel for the reference vs 17.5
91 c/pel for the optimized implementations. Not sure why there is such a 91 c/pel for the optimized implementations. Not sure why there is such a