diff options
Diffstat (limited to 'src/shaders/film_grain_av1.c')
| -rw-r--r-- | src/shaders/film_grain_av1.c | 1001 |
1 files changed, 1001 insertions, 0 deletions
diff --git a/src/shaders/film_grain_av1.c b/src/shaders/film_grain_av1.c new file mode 100644 index 0000000..3b11ea3 --- /dev/null +++ b/src/shaders/film_grain_av1.c @@ -0,0 +1,1001 @@ +/* + * This file is part of libplacebo, which is normally licensed under the terms + * of the LGPL v2.1+. However, this file (film_grain_av1.c) is also available + * under the terms of the more permissive MIT license: + * + * Copyright (c) 2018-2019 Niklas Haas + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "shaders.h" +#include "shaders/film_grain.h" + +// Taken from the spec. Range is [-2048, 2047], mean is 0 and stddev is 512 +static const int16_t gaussian_sequence[2048] = { + 56, 568, -180, 172, 124, -84, 172, -64, -900, 24, 820, + 224, 1248, 996, 272, -8, -916, -388, -732, -104, -188, 800, + 112, -652, -320, -376, 140, -252, 492, -168, 44, -788, 588, + -584, 500, -228, 12, 680, 272, -476, 972, -100, 652, 368, + 432, -196, -720, -192, 1000, -332, 652, -136, -552, -604, -4, + 192, -220, -136, 1000, -52, 372, -96, -624, 124, -24, 396, + 540, -12, -104, 640, 464, 244, -208, -84, 368, -528, -740, + 248, -968, -848, 608, 376, -60, -292, -40, -156, 252, -292, + 248, 224, -280, 400, -244, 244, -60, 76, -80, 212, 532, + 340, 128, -36, 824, -352, -60, -264, -96, -612, 416, -704, + 220, -204, 640, -160, 1220, -408, 900, 336, 20, -336, -96, + -792, 304, 48, -28, -1232, -1172, -448, 104, -292, -520, 244, + 60, -948, 0, -708, 268, 108, 356, -548, 488, -344, -136, + 488, -196, -224, 656, -236, -1128, 60, 4, 140, 276, -676, + -376, 168, -108, 464, 8, 564, 64, 240, 308, -300, -400, + -456, -136, 56, 120, -408, -116, 436, 504, -232, 328, 844, + -164, -84, 784, -168, 232, -224, 348, -376, 128, 568, 96, + -1244, -288, 276, 848, 832, -360, 656, 464, -384, -332, -356, + 728, -388, 160, -192, 468, 296, 224, 140, -776, -100, 280, + 4, 196, 44, -36, -648, 932, 16, 1428, 28, 528, 808, + 772, 20, 268, 88, -332, -284, 124, -384, -448, 208, -228, + -1044, -328, 660, 380, -148, -300, 588, 240, 540, 28, 136, + -88, -436, 256, 296, -1000, 1400, 0, -48, 1056, -136, 264, + -528, -1108, 632, -484, -592, -344, 796, 124, -668, -768, 388, + 1296, -232, -188, -200, -288, -4, 308, 100, -168, 256, -500, + 204, -508, 648, -136, 372, -272, -120, -1004, -552, -548, -384, + 548, -296, 428, -108, -8, -912, -324, -224, -88, -112, -220, + -100, 996, -796, 548, 360, -216, 180, 428, -200, -212, 148, + 96, 148, 284, 216, -412, -320, 120, -300, -384, -604, -572, + -332, -8, -180, -176, 696, 116, -88, 628, 76, 44, -516, + 240, -208, -40, 100, -592, 344, -308, -452, -228, 20, 916, + -1752, -136, -340, -804, 140, 40, 512, 340, 248, 184, -492, + 896, -156, 932, -628, 328, -688, -448, -616, -752, -100, 560, + -1020, 180, -800, -64, 76, 576, 1068, 396, 660, 552, -108, + -28, 320, -628, 312, -92, -92, -472, 268, 16, 560, 516, + -672, -52, 492, -100, 260, 384, 284, 292, 304, -148, 88, + -152, 1012, 1064, -228, 164, -376, -684, 592, -392, 156, 196, + -524, -64, -884, 160, -176, 636, 648, 404, -396, -436, 864, + 424, -728, 988, -604, 904, -592, 296, -224, 536, -176, -920, + 436, -48, 1176, -884, 416, -776, -824, -884, 524, -548, -564, + -68, -164, -96, 692, 364, -692, -1012, -68, 260, -480, 876, + -1116, 452, -332, -352, 892, -1088, 1220, -676, 12, -292, 244, + 496, 372, -32, 280, 200, 112, -440, -96, 24, -644, -184, + 56, -432, 224, -980, 272, -260, 144, -436, 420, 356, 364, + -528, 76, 172, -744, -368, 404, -752, -416, 684, -688, 72, + 540, 416, 92, 444, 480, -72, -1416, 164, -1172, -68, 24, + 424, 264, 1040, 128, -912, -524, -356, 64, 876, -12, 4, + -88, 532, 272, -524, 320, 276, -508, 940, 24, -400, -120, + 756, 60, 236, -412, 100, 376, -484, 400, -100, -740, -108, + -260, 328, -268, 224, -200, -416, 184, -604, -564, -20, 296, + 60, 892, -888, 60, 164, 68, -760, 216, -296, 904, -336, + -28, 404, -356, -568, -208, -1480, -512, 296, 328, -360, -164, + -1560, -776, 1156, -428, 164, -504, -112, 120, -216, -148, -264, + 308, 32, 64, -72, 72, 116, 176, -64, -272, 460, -536, + -784, -280, 348, 108, -752, -132, 524, -540, -776, 116, -296, + -1196, -288, -560, 1040, -472, 116, -848, -1116, 116, 636, 696, + 284, -176, 1016, 204, -864, -648, -248, 356, 972, -584, -204, + 264, 880, 528, -24, -184, 116, 448, -144, 828, 524, 212, + -212, 52, 12, 200, 268, -488, -404, -880, 824, -672, -40, + 908, -248, 500, 716, -576, 492, -576, 16, 720, -108, 384, + 124, 344, 280, 576, -500, 252, 104, -308, 196, -188, -8, + 1268, 296, 1032, -1196, 436, 316, 372, -432, -200, -660, 704, + -224, 596, -132, 268, 32, -452, 884, 104, -1008, 424, -1348, + -280, 4, -1168, 368, 476, 696, 300, -8, 24, 180, -592, + -196, 388, 304, 500, 724, -160, 244, -84, 272, -256, -420, + 320, 208, -144, -156, 156, 364, 452, 28, 540, 316, 220, + -644, -248, 464, 72, 360, 32, -388, 496, -680, -48, 208, + -116, -408, 60, -604, -392, 548, -840, 784, -460, 656, -544, + -388, -264, 908, -800, -628, -612, -568, 572, -220, 164, 288, + -16, -308, 308, -112, -636, -760, 280, -668, 432, 364, 240, + -196, 604, 340, 384, 196, 592, -44, -500, 432, -580, -132, + 636, -76, 392, 4, -412, 540, 508, 328, -356, -36, 16, + -220, -64, -248, -60, 24, -192, 368, 1040, 92, -24, -1044, + -32, 40, 104, 148, 192, -136, -520, 56, -816, -224, 732, + 392, 356, 212, -80, -424, -1008, -324, 588, -1496, 576, 460, + -816, -848, 56, -580, -92, -1372, -112, -496, 200, 364, 52, + -140, 48, -48, -60, 84, 72, 40, 132, -356, -268, -104, + -284, -404, 732, -520, 164, -304, -540, 120, 328, -76, -460, + 756, 388, 588, 236, -436, -72, -176, -404, -316, -148, 716, + -604, 404, -72, -88, -888, -68, 944, 88, -220, -344, 960, + 472, 460, -232, 704, 120, 832, -228, 692, -508, 132, -476, + 844, -748, -364, -44, 1116, -1104, -1056, 76, 428, 552, -692, + 60, 356, 96, -384, -188, -612, -576, 736, 508, 892, 352, + -1132, 504, -24, -352, 324, 332, -600, -312, 292, 508, -144, + -8, 484, 48, 284, -260, -240, 256, -100, -292, -204, -44, + 472, -204, 908, -188, -1000, -256, 92, 1164, -392, 564, 356, + 652, -28, -884, 256, 484, -192, 760, -176, 376, -524, -452, + -436, 860, -736, 212, 124, 504, -476, 468, 76, -472, 552, + -692, -944, -620, 740, -240, 400, 132, 20, 192, -196, 264, + -668, -1012, -60, 296, -316, -828, 76, -156, 284, -768, -448, + -832, 148, 248, 652, 616, 1236, 288, -328, -400, -124, 588, + 220, 520, -696, 1032, 768, -740, -92, -272, 296, 448, -464, + 412, -200, 392, 440, -200, 264, -152, -260, 320, 1032, 216, + 320, -8, -64, 156, -1016, 1084, 1172, 536, 484, -432, 132, + 372, -52, -256, 84, 116, -352, 48, 116, 304, -384, 412, + 924, -300, 528, 628, 180, 648, 44, -980, -220, 1320, 48, + 332, 748, 524, -268, -720, 540, -276, 564, -344, -208, -196, + 436, 896, 88, -392, 132, 80, -964, -288, 568, 56, -48, + -456, 888, 8, 552, -156, -292, 948, 288, 128, -716, -292, + 1192, -152, 876, 352, -600, -260, -812, -468, -28, -120, -32, + -44, 1284, 496, 192, 464, 312, -76, -516, -380, -456, -1012, + -48, 308, -156, 36, 492, -156, -808, 188, 1652, 68, -120, + -116, 316, 160, -140, 352, 808, -416, 592, 316, -480, 56, + 528, -204, -568, 372, -232, 752, -344, 744, -4, 324, -416, + -600, 768, 268, -248, -88, -132, -420, -432, 80, -288, 404, + -316, -1216, -588, 520, -108, 92, -320, 368, -480, -216, -92, + 1688, -300, 180, 1020, -176, 820, -68, -228, -260, 436, -904, + 20, 40, -508, 440, -736, 312, 332, 204, 760, -372, 728, + 96, -20, -632, -520, -560, 336, 1076, -64, -532, 776, 584, + 192, 396, -728, -520, 276, -188, 80, -52, -612, -252, -48, + 648, 212, -688, 228, -52, -260, 428, -412, -272, -404, 180, + 816, -796, 48, 152, 484, -88, -216, 988, 696, 188, -528, + 648, -116, -180, 316, 476, 12, -564, 96, 476, -252, -364, + -376, -392, 556, -256, -576, 260, -352, 120, -16, -136, -260, + -492, 72, 556, 660, 580, 616, 772, 436, 424, -32, -324, + -1268, 416, -324, -80, 920, 160, 228, 724, 32, -516, 64, + 384, 68, -128, 136, 240, 248, -204, -68, 252, -932, -120, + -480, -628, -84, 192, 852, -404, -288, -132, 204, 100, 168, + -68, -196, -868, 460, 1080, 380, -80, 244, 0, 484, -888, + 64, 184, 352, 600, 460, 164, 604, -196, 320, -64, 588, + -184, 228, 12, 372, 48, -848, -344, 224, 208, -200, 484, + 128, -20, 272, -468, -840, 384, 256, -720, -520, -464, -580, + 112, -120, 644, -356, -208, -608, -528, 704, 560, -424, 392, + 828, 40, 84, 200, -152, 0, -144, 584, 280, -120, 80, + -556, -972, -196, -472, 724, 80, 168, -32, 88, 160, -688, + 0, 160, 356, 372, -776, 740, -128, 676, -248, -480, 4, + -364, 96, 544, 232, -1032, 956, 236, 356, 20, -40, 300, + 24, -676, -596, 132, 1120, -104, 532, -1096, 568, 648, 444, + 508, 380, 188, -376, -604, 1488, 424, 24, 756, -220, -192, + 716, 120, 920, 688, 168, 44, -460, 568, 284, 1144, 1160, + 600, 424, 888, 656, -356, -320, 220, 316, -176, -724, -188, + -816, -628, -348, -228, -380, 1012, -452, -660, 736, 928, 404, + -696, -72, -268, -892, 128, 184, -344, -780, 360, 336, 400, + 344, 428, 548, -112, 136, -228, -216, -820, -516, 340, 92, + -136, 116, -300, 376, -244, 100, -316, -520, -284, -12, 824, + 164, -548, -180, -128, 116, -924, -828, 268, -368, -580, 620, + 192, 160, 0, -1676, 1068, 424, -56, -360, 468, -156, 720, + 288, -528, 556, -364, 548, -148, 504, 316, 152, -648, -620, + -684, -24, -376, -384, -108, -920, -1032, 768, 180, -264, -508, + -1268, -260, -60, 300, -240, 988, 724, -376, -576, -212, -736, + 556, 192, 1092, -620, -880, 376, -56, -4, -216, -32, 836, + 268, 396, 1332, 864, -600, 100, 56, -412, -92, 356, 180, + 884, -468, -436, 292, -388, -804, -704, -840, 368, -348, 140, + -724, 1536, 940, 372, 112, -372, 436, -480, 1136, 296, -32, + -228, 132, -48, -220, 868, -1016, -60, -1044, -464, 328, 916, + 244, 12, -736, -296, 360, 468, -376, -108, -92, 788, 368, + -56, 544, 400, -672, -420, 728, 16, 320, 44, -284, -380, + -796, 488, 132, 204, -596, -372, 88, -152, -908, -636, -572, + -624, -116, -692, -200, -56, 276, -88, 484, -324, 948, 864, + 1000, -456, -184, -276, 292, -296, 156, 676, 320, 160, 908, + -84, -1236, -288, -116, 260, -372, -644, 732, -756, -96, 84, + 344, -520, 348, -688, 240, -84, 216, -1044, -136, -676, -396, + -1500, 960, -40, 176, 168, 1516, 420, -504, -344, -364, -360, + 1216, -940, -380, -212, 252, -660, -708, 484, -444, -152, 928, + -120, 1112, 476, -260, 560, -148, -344, 108, -196, 228, -288, + 504, 560, -328, -88, 288, -1008, 460, -228, 468, -836, -196, + 76, 388, 232, 412, -1168, -716, -644, 756, -172, -356, -504, + 116, 432, 528, 48, 476, -168, -608, 448, 160, -532, -272, + 28, -676, -12, 828, 980, 456, 520, 104, -104, 256, -344, + -4, -28, -368, -52, -524, -572, -556, -200, 768, 1124, -208, + -512, 176, 232, 248, -148, -888, 604, -600, -304, 804, -156, + -212, 488, -192, -804, -256, 368, -360, -916, -328, 228, -240, + -448, -472, 856, -556, -364, 572, -12, -156, -368, -340, 432, + 252, -752, -152, 288, 268, -580, -848, -592, 108, -76, 244, + 312, -716, 592, -80, 436, 360, 4, -248, 160, 516, 584, + 732, 44, -468, -280, -292, -156, -588, 28, 308, 912, 24, + 124, 156, 180, -252, 944, -924, -772, -520, -428, -624, 300, + -212, -1144, 32, -724, 800, -1128, -212, -1288, -848, 180, -416, + 440, 192, -576, -792, -76, -1080, 80, -532, -352, -132, 380, + -820, 148, 1112, 128, 164, 456, 700, -924, 144, -668, -384, + 648, -832, 508, 552, -52, -100, -656, 208, -568, 748, -88, + 680, 232, 300, 192, -408, -1012, -152, -252, -268, 272, -876, + -664, -648, -332, -136, 16, 12, 1152, -28, 332, -536, 320, + -672, -460, -316, 532, -260, 228, -40, 1052, -816, 180, 88, + -496, -556, -672, -368, 428, 92, 356, 404, -408, 252, 196, + -176, -556, 792, 268, 32, 372, 40, 96, -332, 328, 120, + 372, -900, -40, 472, -264, -592, 952, 128, 656, 112, 664, + -232, 420, 4, -344, -464, 556, 244, -416, -32, 252, 0, + -412, 188, -696, 508, -476, 324, -1096, 656, -312, 560, 264, + -136, 304, 160, -64, -580, 248, 336, -720, 560, -348, -288, + -276, -196, -500, 852, -544, -236, -1128, -992, -776, 116, 56, + 52, 860, 884, 212, -12, 168, 1020, 512, -552, 924, -148, + 716, 188, 164, -340, -520, -184, 880, -152, -680, -208, -1156, + -300, -528, -472, 364, 100, -744, -1056, -32, 540, 280, 144, + -676, -32, -232, -280, -224, 96, 568, -76, 172, 148, 148, + 104, 32, -296, -32, 788, -80, 32, -16, 280, 288, 944, + 428, -484 +}; + +static inline int get_random_number(int bits, uint16_t *state) +{ + int r = *state; + uint16_t bit = ((r >> 0) ^ (r >> 1) ^ (r >> 3) ^ (r >> 12)) & 1; + *state = (r >> 1) | (bit << 15); + + return (*state >> (16 - bits)) & ((1 << bits) - 1); +} + +static inline int round2(int x, int shift) +{ + if (!shift) + return x; + + return (x + (1 << (shift - 1))) >> shift; +} + +enum { + BLOCK_SIZE = 32, + SCALING_LUT_SIZE = 256, + + GRAIN_WIDTH = 82, + GRAIN_HEIGHT = 73, + // On the GPU we only need a subsection of this + GRAIN_WIDTH_LUT = 64, + GRAIN_HEIGHT_LUT = 64, + GRAIN_PAD_LUT = 9, + + // For subsampled grain textures + SUB_GRAIN_WIDTH = 44, + SUB_GRAIN_HEIGHT = 38, + SUB_GRAIN_WIDTH_LUT = GRAIN_WIDTH_LUT >> 1, + SUB_GRAIN_HEIGHT_LUT = GRAIN_HEIGHT_LUT >> 1, + SUB_GRAIN_PAD_LUT = 6, +}; + +// Contains the shift by which the offsets are indexed +enum offset { + OFFSET_TL = 24, + OFFSET_T = 16, + OFFSET_L = 8, + OFFSET_N = 0, +}; + +// Helper function to compute some common constants +struct grain_scale { + int grain_center; + int grain_min; + int grain_max; + float texture_scale; + float grain_scale; +}; + +static inline int bit_depth(const struct pl_color_repr *repr) +{ + int depth = PL_DEF(repr->bits.color_depth, + PL_DEF(repr->bits.sample_depth, 8)); + pl_assert(depth >= 8); + return PL_MIN(depth, 12); +} + +static struct grain_scale get_grain_scale(const struct pl_film_grain_params *params) +{ + int bits = bit_depth(params->repr); + struct grain_scale ret = { + .grain_center = 128 << (bits - 8), + }; + + ret.grain_min = -ret.grain_center; + ret.grain_max = (256 << (bits - 8)) - 1 - ret.grain_center; + + struct pl_color_repr repr = *params->repr; + ret.texture_scale = pl_color_repr_normalize(&repr); + + // Since our color samples are normalized to the range [0, 1], we need to + // scale down grain values from the scale [0, 2^b - 1] to this range. + ret.grain_scale = 1.0 / ((1 << bits) - 1); + + return ret; +} + +// Generates the basic grain table (LumaGrain in the spec). +static void generate_grain_y(float out[GRAIN_HEIGHT_LUT][GRAIN_WIDTH_LUT], + int16_t buf[GRAIN_HEIGHT][GRAIN_WIDTH], + const struct pl_film_grain_params *params) +{ + const struct pl_av1_grain_data *data = ¶ms->data.params.av1; + struct grain_scale scale = get_grain_scale(params); + uint16_t seed = (uint16_t) params->data.seed; + int bits = bit_depth(params->repr); + int shift = 12 - bits + data->grain_scale_shift; + pl_assert(shift >= 0); + + for (int y = 0; y < GRAIN_HEIGHT; y++) { + for (int x = 0; x < GRAIN_WIDTH; x++) { + int16_t value = gaussian_sequence[ get_random_number(11, &seed) ]; + buf[y][x] = round2(value, shift); + } + } + + const int ar_pad = 3; + int ar_lag = data->ar_coeff_lag; + + for (int y = ar_pad; y < GRAIN_HEIGHT; y++) { + for (int x = ar_pad; x < GRAIN_WIDTH - ar_pad; x++) { + const int8_t *coeff = data->ar_coeffs_y; + int sum = 0; + for (int dy = -ar_lag; dy <= 0; dy++) { + for (int dx = -ar_lag; dx <= ar_lag; dx++) { + if (!dx && !dy) + break; + sum += *(coeff++) * buf[y + dy][x + dx]; + } + } + + int16_t grain = buf[y][x] + round2(sum, data->ar_coeff_shift); + grain = PL_CLAMP(grain, scale.grain_min, scale.grain_max); + buf[y][x] = grain; + } + } + + for (int y = 0; y < GRAIN_HEIGHT_LUT; y++) { + for (int x = 0; x < GRAIN_WIDTH_LUT; x++) { + int16_t grain = buf[y + GRAIN_PAD_LUT][x + GRAIN_PAD_LUT]; + out[y][x] = grain * scale.grain_scale; + } + } +} + +static void generate_grain_uv(float *out, int16_t buf[GRAIN_HEIGHT][GRAIN_WIDTH], + const int16_t buf_y[GRAIN_HEIGHT][GRAIN_WIDTH], + enum pl_channel channel, int sub_x, int sub_y, + const struct pl_film_grain_params *params) +{ + const struct pl_av1_grain_data *data = ¶ms->data.params.av1; + struct grain_scale scale = get_grain_scale(params); + int bits = bit_depth(params->repr); + int shift = 12 - bits + data->grain_scale_shift; + pl_assert(shift >= 0); + + uint16_t seed = params->data.seed; + if (channel == PL_CHANNEL_CB) { + seed ^= 0xb524; + } else if (channel == PL_CHANNEL_CR) { + seed ^= 0x49d8; + } + + int chromaW = sub_x ? SUB_GRAIN_WIDTH : GRAIN_WIDTH; + int chromaH = sub_y ? SUB_GRAIN_HEIGHT : GRAIN_HEIGHT; + + const int8_t *coeffs[] = { + [PL_CHANNEL_CB] = data->ar_coeffs_uv[0], + [PL_CHANNEL_CR] = data->ar_coeffs_uv[1], + }; + + for (int y = 0; y < chromaH; y++) { + for (int x = 0; x < chromaW; x++) { + int16_t value = gaussian_sequence[ get_random_number(11, &seed) ]; + buf[y][x] = round2(value, shift); + } + } + + const int ar_pad = 3; + int ar_lag = data->ar_coeff_lag; + + for (int y = ar_pad; y < chromaH; y++) { + for (int x = ar_pad; x < chromaW - ar_pad; x++) { + const int8_t *coeff = coeffs[channel]; + pl_assert(coeff); + int sum = 0; + for (int dy = -ar_lag; dy <= 0; dy++) { + for (int dx = -ar_lag; dx <= ar_lag; dx++) { + // For the final (current) pixel, we need to add in the + // contribution from the luma grain texture + if (!dx && !dy) { + if (!data->num_points_y) + break; + int luma = 0; + int lumaX = ((x - ar_pad) << sub_x) + ar_pad; + int lumaY = ((y - ar_pad) << sub_y) + ar_pad; + for (int i = 0; i <= sub_y; i++) { + for (int j = 0; j <= sub_x; j++) { + luma += buf_y[lumaY + i][lumaX + j]; + } + } + luma = round2(luma, sub_x + sub_y); + sum += luma * (*coeff); + break; + } + + sum += *(coeff++) * buf[y + dy][x + dx]; + } + } + + int16_t grain = buf[y][x] + round2(sum, data->ar_coeff_shift); + grain = PL_CLAMP(grain, scale.grain_min, scale.grain_max); + buf[y][x] = grain; + } + } + + int lutW = GRAIN_WIDTH_LUT >> sub_x; + int lutH = GRAIN_HEIGHT_LUT >> sub_y; + int padX = sub_x ? SUB_GRAIN_PAD_LUT : GRAIN_PAD_LUT; + int padY = sub_y ? SUB_GRAIN_PAD_LUT : GRAIN_PAD_LUT; + + for (int y = 0; y < lutH; y++) { + for (int x = 0; x < lutW; x++) { + int16_t grain = buf[y + padY][x + padX]; + out[y * lutW + x] = grain * scale.grain_scale; + } + } +} + +static void generate_offsets(void *pbuf, const struct sh_lut_params *params) +{ + const struct pl_film_grain_data *data = params->priv; + unsigned int *buf = pbuf; + pl_static_assert(sizeof(unsigned int) >= sizeof(uint32_t)); + + for (int y = 0; y < params->height; y++) { + uint16_t state = data->seed; + state ^= ((y * 37 + 178) & 0xFF) << 8; + state ^= ((y * 173 + 105) & 0xFF); + + for (int x = 0; x < params->width; x++) { + unsigned int *offsets = &buf[y * params->width + x]; + + uint8_t val = get_random_number(8, &state); + uint8_t val_l = x ? (offsets - 1)[0] : 0; + uint8_t val_t = y ? (offsets - params->width)[0] : 0; + uint8_t val_tl = x && y ? (offsets - params->width - 1)[0] : 0; + + // Encode four offsets into a single 32-bit integer for the + // convenience of the GPU. That way only one LUT fetch is + // required for the entire block. + *offsets = ((uint32_t) val_tl << OFFSET_TL) + | ((uint32_t) val_t << OFFSET_T) + | ((uint32_t) val_l << OFFSET_L) + | ((uint32_t) val << OFFSET_N); + } + } +} + +static void generate_scaling(void *pdata, const struct sh_lut_params *params) +{ + assert(params->width == SCALING_LUT_SIZE && params->comps == 1); + float *data = pdata; + + struct { + int num; + uint8_t (*points)[2]; + const struct pl_av1_grain_data *data; + } *ctx = params->priv; + + float range = 1 << ctx->data->scaling_shift; + + // Fill up the preceding entries with the initial value + for (int i = 0; i < ctx->points[0][0]; i++) + data[i] = ctx->points[0][1] / range; + + // Linearly interpolate the values in the middle + for (int i = 0; i < ctx->num - 1; i++) { + int bx = ctx->points[i][0]; + int by = ctx->points[i][1]; + int dx = ctx->points[i + 1][0] - bx; + int dy = ctx->points[i + 1][1] - by; + int delta = dy * ((0x10000 + (dx >> 1)) / dx); + for (int x = 0; x < dx; x++) { + int v = by + ((x * delta + 0x8000) >> 16); + data[bx + x] = v / range; + } + } + + // Fill up the remaining entries with the final value + for (int i = ctx->points[ctx->num - 1][0]; i < SCALING_LUT_SIZE; i++) + data[i] = ctx->points[ctx->num - 1][1] / range; +} + +static void sample(pl_shader sh, enum offset off, ident_t lut, int idx, + int sub_x, int sub_y) +{ + int dx = (off & OFFSET_L) ? 1 : 0, + dy = (off & OFFSET_T) ? 1 : 0; + + static const char *index_strs[] = { + [0] = ".x", + [1] = ".y", + }; + + GLSL("offset = uvec2(%du, %du) * uvec2((data >> %d) & 0xFu, \n" + " (data >> %d) & 0xFu);\n" + "pos = offset + local_id.xy + uvec2(%d, %d); \n" + "val = "$"(pos)%s; \n", + sub_x ? 1 : 2, sub_y ? 1 : 2, off + 4, off, + (BLOCK_SIZE >> sub_x) * dx, + (BLOCK_SIZE >> sub_y) * dy, + lut, idx >= 0 ? index_strs[idx] : ""); +} + +struct grain_obj_av1 { + // LUT objects for the offsets, grain and scaling luts + pl_shader_obj lut_offsets; + pl_shader_obj lut_grain[2]; + pl_shader_obj lut_scaling[3]; + + // Previous parameters used to check reusability + struct pl_film_grain_data data; + struct pl_color_repr repr; + bool fg_has_y; + bool fg_has_u; + bool fg_has_v; + + // Space to store the temporary arrays, reused + uint32_t *offsets; + float grain[2][GRAIN_HEIGHT_LUT][GRAIN_WIDTH_LUT]; + int16_t grain_tmp_y[GRAIN_HEIGHT][GRAIN_WIDTH]; + int16_t grain_tmp_uv[GRAIN_HEIGHT][GRAIN_WIDTH]; +}; + +static void av1_grain_uninit(pl_gpu gpu, void *ptr) +{ + struct grain_obj_av1 *obj = ptr; + pl_shader_obj_destroy(&obj->lut_offsets); + for (int i = 0; i < PL_ARRAY_SIZE(obj->lut_grain); i++) + pl_shader_obj_destroy(&obj->lut_grain[i]); + for (int i = 0; i < PL_ARRAY_SIZE(obj->lut_scaling); i++) + pl_shader_obj_destroy(&obj->lut_scaling[i]); + *obj = (struct grain_obj_av1) {0}; +} + +bool pl_needs_fg_av1(const struct pl_film_grain_params *params) +{ + const struct pl_av1_grain_data *data = ¶ms->data.params.av1; + bool has_y = data->num_points_y > 0; + bool has_u = data->num_points_uv[0] > 0 || data->chroma_scaling_from_luma; + bool has_v = data->num_points_uv[1] > 0 || data->chroma_scaling_from_luma; + + for (int i = 0; i < 3; i++) { + enum pl_channel channel = channel_map(i, params); + if (channel == PL_CHANNEL_Y && has_y) + return true; + if (channel == PL_CHANNEL_CB && has_u) + return true; + if (channel == PL_CHANNEL_CR && has_v) + return true; + } + + return false; +} + +static inline bool av1_grain_data_eq(const struct pl_film_grain_data *da, + const struct pl_film_grain_data *db) +{ + const struct pl_av1_grain_data *a = &da->params.av1, *b = &db->params.av1; + + // Only check the fields that are relevant for grain LUT generation + return da->seed == db->seed && + a->chroma_scaling_from_luma == b->chroma_scaling_from_luma && + a->scaling_shift == b->scaling_shift && + a->ar_coeff_lag == b->ar_coeff_lag && + a->ar_coeff_shift == b->ar_coeff_shift && + a->grain_scale_shift == b->grain_scale_shift && + !memcmp(a->ar_coeffs_y, b->ar_coeffs_y, sizeof(a->ar_coeffs_y)) && + !memcmp(a->ar_coeffs_uv, b->ar_coeffs_uv, sizeof(a->ar_coeffs_uv)); +} + +static void fill_grain_lut(void *data, const struct sh_lut_params *params) +{ + struct grain_obj_av1 *obj = params->priv; + size_t entries = params->width * params->height * params->comps; + memcpy(data, obj->grain, entries * sizeof(float)); +} + +bool pl_shader_fg_av1(pl_shader sh, pl_shader_obj *grain_state, + const struct pl_film_grain_params *params) +{ + int sub_x = 0, sub_y = 0; + int tex_w = params->tex->params.w, + tex_h = params->tex->params.h; + + if (params->luma_tex) { + sub_x = params->luma_tex->params.w > tex_w; + sub_y = params->luma_tex->params.h > tex_h; + } + + const struct pl_av1_grain_data *data = ¶ms->data.params.av1; + bool fg_has_y = data->num_points_y > 0; + bool fg_has_u = data->num_points_uv[0] > 0 || data->chroma_scaling_from_luma; + bool fg_has_v = data->num_points_uv[1] > 0 || data->chroma_scaling_from_luma; + + bool tex_is_y = false, tex_is_cb = false, tex_is_cr = false; + for (int i = 0; i < 3; i++) { + switch (channel_map(i, params)) { + case PL_CHANNEL_Y: tex_is_y = true; break; + case PL_CHANNEL_CB: tex_is_cb = true; break; + case PL_CHANNEL_CR: tex_is_cr = true; break; + default: break; + }; + } + + if (tex_is_y && (sub_x || sub_y)) { + PL_WARN(sh, "pl_film_grain_params.channels includes PL_CHANNEL_Y but " + "plane is subsampled, this makes no sense. Continuing anyway " + "but output is likely incorrect."); + } + + if (!sh_require(sh, PL_SHADER_SIG_NONE, tex_w, tex_h)) + return false; + + pl_gpu gpu = SH_GPU(sh); + if (!gpu) { + PL_ERR(sh, "AV1 film grain synthesis requires a non-NULL pl_gpu!"); + return false; + } + + // Disable generation for unneeded component types + fg_has_y &= tex_is_y; + fg_has_u &= tex_is_cb; + fg_has_v &= tex_is_cr; + + int bw = BLOCK_SIZE >> sub_x; + int bh = BLOCK_SIZE >> sub_y; + bool is_compute = sh_try_compute(sh, bw, bh, false, sizeof(uint32_t)); + + struct grain_obj_av1 *obj; + obj = SH_OBJ(sh, grain_state, PL_SHADER_OBJ_AV1_GRAIN, + struct grain_obj_av1, av1_grain_uninit); + if (!obj) + return false; + + // Note: In theory we could check only the parameters related to luma or + // only related to chroma and skip updating for changes to irrelevant + // parts, but this is probably not worth it since the seed is expected to + // change per frame anyway. + bool needs_update = !av1_grain_data_eq(¶ms->data, &obj->data) || + !pl_color_repr_equal(params->repr, &obj->repr) || + fg_has_y != obj->fg_has_y || + fg_has_u != obj->fg_has_u || + fg_has_v != obj->fg_has_v; + + if (needs_update) { + // This is needed even for chroma, so statically generate it + generate_grain_y(obj->grain[0], obj->grain_tmp_y, params); + } + + ident_t lut[3]; + int idx[3] = {-1}; + + if (fg_has_y) { + lut[0] = sh_lut(sh, sh_lut_params( + .object = &obj->lut_grain[0], + .var_type = PL_VAR_FLOAT, + .lut_type = SH_LUT_TEXTURE, + .width = GRAIN_WIDTH_LUT, + .height = GRAIN_HEIGHT_LUT, + .comps = 1, + .update = needs_update, + .dynamic = true, + .fill = fill_grain_lut, + .priv = obj, + )); + + if (!lut[0]) { + SH_FAIL(sh, "Failed generating/uploading luma grain LUT!"); + return false; + } + } + + // Try merging the chroma LUTs into a single texture + int chroma_comps = 0; + if (fg_has_u) { + generate_grain_uv(&obj->grain[chroma_comps][0][0], obj->grain_tmp_uv, + obj->grain_tmp_y, PL_CHANNEL_CB, sub_x, sub_y, + params); + idx[1] = chroma_comps++; + } + if (fg_has_v) { + generate_grain_uv(&obj->grain[chroma_comps][0][0], obj->grain_tmp_uv, + obj->grain_tmp_y, PL_CHANNEL_CR, sub_x, sub_y, + params); + idx[2] = chroma_comps++; + } + + if (chroma_comps > 0) { + lut[1] = lut[2] = sh_lut(sh, sh_lut_params( + .object = &obj->lut_grain[1], + .var_type = PL_VAR_FLOAT, + .lut_type = SH_LUT_TEXTURE, + .width = GRAIN_WIDTH_LUT >> sub_x, + .height = GRAIN_HEIGHT_LUT >> sub_y, + .comps = chroma_comps, + .update = needs_update, + .dynamic = true, + .fill = fill_grain_lut, + .priv = obj, + )); + + if (!lut[1]) { + SH_FAIL(sh, "Failed generating/uploading chroma grain LUT!"); + return false; + } + + if (chroma_comps == 1) + idx[1] = idx[2] = -1; + } + + ident_t offsets = sh_lut(sh, sh_lut_params( + .object = &obj->lut_offsets, + .var_type = PL_VAR_UINT, + .lut_type = SH_LUT_AUTO, + .width = PL_ALIGN2(tex_w << sub_x, 128) / 32, + .height = PL_ALIGN2(tex_h << sub_y, 128) / 32, + .comps = 1, + .update = needs_update, + .dynamic = true, + .fill = generate_offsets, + .priv = (void *) ¶ms->data, + )); + + if (!offsets) { + SH_FAIL(sh, "Failed generating/uploading block offsets LUT!"); + return false; + } + + // For the scaling LUTs, we assume they'll be relatively constant + // throughout the video so doing some extra work to avoid reinitializing + // them constantly is probably worth it. Probably. + const struct pl_av1_grain_data *obj_data = &obj->data.params.av1; + bool scaling_changed = false; + if (fg_has_y || data->chroma_scaling_from_luma) { + scaling_changed |= data->num_points_y != obj_data->num_points_y; + scaling_changed |= memcmp(data->points_y, obj_data->points_y, + sizeof(data->points_y)); + } + + if (fg_has_u && !data->chroma_scaling_from_luma) { + scaling_changed |= data->num_points_uv[0] != obj_data->num_points_uv[0]; + scaling_changed |= memcmp(data->points_uv[0], + obj_data->points_uv[0], + sizeof(data->points_uv[0])); + } + + if (fg_has_v && !data->chroma_scaling_from_luma) { + scaling_changed |= data->num_points_uv[1] != obj_data->num_points_uv[1]; + scaling_changed |= memcmp(data->points_uv[1], + obj_data->points_uv[1], + sizeof(data->points_uv[1])); + } + + ident_t scaling[3] = {0}; + for (int i = 0; i < 3; i++) { + struct { + int num; + const uint8_t (*points)[2]; + const struct pl_av1_grain_data *data; + } priv; + + priv.data = data; + if (i == 0 || data->chroma_scaling_from_luma) { + priv.num = data->num_points_y; + priv.points = &data->points_y[0]; + } else { + priv.num = data->num_points_uv[i - 1]; + priv.points = &data->points_uv[i - 1][0]; + } + + // Skip scaling for unneeded channels + bool has_c[3] = { fg_has_y, fg_has_u, fg_has_v }; + if (has_c[i] && priv.num > 0) { + scaling[i] = sh_lut(sh, sh_lut_params( + .object = &obj->lut_scaling[i], + .var_type = PL_VAR_FLOAT, + .method = SH_LUT_LINEAR, + .width = SCALING_LUT_SIZE, + .comps = 1, + .update = scaling_changed, + .dynamic = true, + .fill = generate_scaling, + .priv = &priv, + )); + + if (!scaling[i]) { + SH_FAIL(sh, "Failed generating/uploading scaling LUTs!"); + return false; + } + } + } + + // Done updating LUTs + obj->data = params->data; + obj->repr = *params->repr; + obj->fg_has_y = fg_has_y; + obj->fg_has_u = fg_has_u; + obj->fg_has_v = fg_has_v; + + sh_describe(sh, "AV1 film grain"); + GLSL("vec4 color; \n" + "// pl_shader_film_grain (AV1) \n" + "{ \n" + "uvec2 offset; \n" + "uvec2 pos; \n" + "float val; \n" + "float grain; \n"); + + if (is_compute) { + GLSL("uvec2 block_id = gl_WorkGroupID.xy; \n" + "uvec2 local_id = gl_LocalInvocationID.xy; \n" + "uvec2 global_id = gl_GlobalInvocationID.xy; \n"); + } else { + GLSL("uvec2 global_id = uvec2(gl_FragCoord); \n" + "uvec2 block_id = global_id / uvec2(%d, %d); \n" + "uvec2 local_id = global_id - uvec2(%d, %d) * block_id; \n", + bw, bh, bw, bh); + } + + // Load the data vector which holds the offsets + if (is_compute) { + ident_t id = sh_fresh(sh, "data"); + GLSLH("shared uint "$"; \n", id); + GLSL("if (gl_LocalInvocationIndex == 0u) \n" + " "$" = uint("$"(block_id)); \n" + "barrier(); \n" + "uint data = "$"; \n", + id, offsets, id); + } else { + GLSL("uint data = uint("$"(block_id)); \n", offsets); + } + + struct grain_scale scale = get_grain_scale(params); + pl_color_repr_normalize(params->repr); + int bits = PL_DEF(params->repr->bits.color_depth, 8); + pl_assert(bits >= 8); + + ident_t minValue, maxLuma, maxChroma; + if (pl_color_levels_guess(params->repr) == PL_COLOR_LEVELS_LIMITED) { + float out_scale = (1 << bits) / ((1 << bits) - 1.0); + minValue = SH_FLOAT(16 / 256.0 * out_scale); + maxLuma = SH_FLOAT(235 / 256.0 * out_scale); + maxChroma = SH_FLOAT(240 / 256.0 * out_scale); + if (!pl_color_system_is_ycbcr_like(params->repr->sys)) + maxChroma = maxLuma; + } else { + minValue = SH_FLOAT(0.0); + maxLuma = SH_FLOAT(1.0); + maxChroma = SH_FLOAT(1.0); + } + + // Load the color value of the tex itself + ident_t tex = sh_desc(sh, (struct pl_shader_desc) { + .binding.object = params->tex, + .desc = (struct pl_desc) { + .name = "tex", + .type = PL_DESC_SAMPLED_TEX, + }, + }); + + ident_t tex_scale = SH_FLOAT(scale.texture_scale); + GLSL("color = vec4("$") * texelFetch("$", ivec2(global_id), 0); \n", + tex_scale, tex); + + // If we need access to the external luma plane, load it now + if (tex_is_cb || tex_is_cr) { + GLSL("float averageLuma; \n"); + if (tex_is_y) { + // We already have the luma channel as part of the pre-sampled color + for (int i = 0; i < 3; i++) { + if (channel_map(i, params) == PL_CHANNEL_Y) { + GLSL("averageLuma = color["$"]; \n", SH_INT(i)); + break; + } + } + } else { + // Luma channel not present in image, attach it separately + pl_assert(params->luma_tex); + ident_t luma = sh_desc(sh, (struct pl_shader_desc) { + .binding.object = params->luma_tex, + .desc = (struct pl_desc) { + .name = "luma", + .type = PL_DESC_SAMPLED_TEX, + }, + }); + + GLSL("pos = global_id * uvec2(%du, %du); \n" + "averageLuma = texelFetch("$", ivec2(pos), 0)["$"]; \n" + "averageLuma *= "$"; \n", + 1 << sub_x, 1 << sub_y, + luma, SH_INT(params->luma_comp), + tex_scale); + } + } + + ident_t grain_min = SH_FLOAT(scale.grain_min * scale.grain_scale); + ident_t grain_max = SH_FLOAT(scale.grain_max * scale.grain_scale); + + for (int i = 0; i < params->components; i++) { + enum pl_channel c = channel_map(i, params); + if (c == PL_CHANNEL_NONE) + continue; + if (!scaling[c]) + continue; + + sample(sh, OFFSET_N, lut[c], idx[c], sub_x, sub_y); + GLSL("grain = val; \n"); + + if (data->overlap) { + const char *weights[] = { "vec2(27.0, 17.0)", "vec2(23.0, 22.0)" }; + + // X-direction overlapping + GLSL("if (block_id.x > 0u && local_id.x < %du) { \n" + "vec2 w = %s / 32.0; \n" + "if (local_id.x == 1u) w.xy = w.yx; \n", + 2 >> sub_x, weights[sub_x]); + sample(sh, OFFSET_L, lut[c], idx[c], sub_x, sub_y); + GLSL("grain = dot(vec2(val, grain), w); \n" + "} \n"); + + // Y-direction overlapping + GLSL("if (block_id.y > 0u && local_id.y < %du) { \n" + "vec2 w = %s / 32.0; \n" + "if (local_id.y == 1u) w.xy = w.yx; \n", + 2 >> sub_y, weights[sub_y]); + + // We need to special-case the top left pixels since these need to + // pre-blend the top-left offset block before blending vertically + GLSL(" if (block_id.x > 0u && local_id.x < %du) {\n" + " vec2 w2 = %s / 32.0; \n" + " if (local_id.x == 1u) w2.xy = w2.yx; \n", + 2 >> sub_x, weights[sub_x]); + sample(sh, OFFSET_TL, lut[c], idx[c], sub_x, sub_y); + GLSL(" float tmp = val; \n"); + sample(sh, OFFSET_T, lut[c], idx[c], sub_x, sub_y); + GLSL(" val = dot(vec2(tmp, val), w2); \n" + " } else { \n"); + sample(sh, OFFSET_T, lut[c], idx[c], sub_x, sub_y); + GLSL(" } \n" + "grain = dot(vec2(val, grain), w); \n" + "} \n"); + + // Correctly clip the interpolated grain + GLSL("grain = clamp(grain, "$", "$"); \n", grain_min, grain_max); + } + + if (c == PL_CHANNEL_Y) { + GLSL("color[%d] += "$"(color[%d]) * grain; \n" + "color[%d] = clamp(color[%d], "$", "$"); \n", + i, scaling[c], i, + i, i, minValue, maxLuma); + } else { + GLSL("val = averageLuma; \n"); + if (!data->chroma_scaling_from_luma) { + // We need to load some extra variables for the mixing. Do this + // using sh_var instead of hard-coding them to avoid shader + // recompilation when these values change. + ident_t mult = sh_var(sh, (struct pl_shader_var) { + .var = pl_var_vec2("mult"), + .data = &(float[2]){ + data->uv_mult_luma[c - 1] / 64.0, + data->uv_mult[c - 1] / 64.0, + }, + }); + + int c_offset = (unsigned) data->uv_offset[c - 1] << (bits - 8); + ident_t offset = sh_var(sh, (struct pl_shader_var) { + .var = pl_var_float("offset"), + .data = &(float) { c_offset * scale.grain_scale }, + }); + + GLSL("val = dot(vec2(val, color[%d]), "$"); \n" + "val += "$"; \n", + i, mult, offset); + } + GLSL("color[%d] += "$"(val) * grain; \n" + "color[%d] = clamp(color[%d], "$", "$"); \n", + i, scaling[c], + i, i, minValue, maxChroma); + } + } + + GLSL("} \n"); + return true; +} |