/* * ProRes RAW decoder * * Copyright (c) 2025 Lynne * * This file is part of FFmpeg. * * FFmpeg 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, * 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 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA */ #define COMP_ID (gl_LocalInvocationID.z) #define BLOCK_ID (gl_LocalInvocationID.y) #define ROW_ID (gl_LocalInvocationID.x) shared float blocks[16][4*64]; const ivec2 scan[64] = { ivec2( 0, 0), ivec2( 4, 0), ivec2( 0, 2), ivec2( 4, 2), ivec2( 0, 8), ivec2( 4, 8), ivec2( 6, 8), ivec2( 2, 10), ivec2( 2, 0), ivec2( 6, 0), ivec2( 2, 2), ivec2( 6, 2), ivec2( 2, 8), ivec2( 8, 8), ivec2( 0, 10), ivec2( 4, 10), ivec2( 8, 0), ivec2(12, 0), ivec2( 8, 2), ivec2(12, 2), ivec2(10, 8), ivec2(14, 8), ivec2( 6, 10), ivec2( 2, 12), ivec2(10, 0), ivec2(14, 0), ivec2(10, 2), ivec2(14, 2), ivec2(12, 8), ivec2( 8, 10), ivec2( 0, 12), ivec2( 4, 12), ivec2( 0, 4), ivec2( 4, 4), ivec2( 6, 4), ivec2( 2, 6), ivec2(10, 10), ivec2(14, 10), ivec2( 6, 12), ivec2( 2, 14), ivec2( 2, 4), ivec2( 8, 4), ivec2( 0, 6), ivec2( 4, 6), ivec2(12, 10), ivec2( 8, 12), ivec2( 0, 14), ivec2( 4, 14), ivec2(10, 4), ivec2(14, 4), ivec2( 6, 6), ivec2(12, 6), ivec2(10, 12), ivec2(14, 12), ivec2( 6, 14), ivec2(12, 14), ivec2(12, 4), ivec2( 8, 6), ivec2(10, 6), ivec2(14, 6), ivec2(12, 12), ivec2( 8, 14), ivec2(10, 14), ivec2(14, 14), }; const float idct_scale[64] = { 0.1250000000000000, 0.1733799806652684, 0.1633203706095471, 0.1469844503024199, 0.1250000000000000, 0.0982118697983878, 0.0676495125182746, 0.0344874224103679, 0.1733799806652684, 0.2404849415639108, 0.2265318615882219, 0.2038732892122293, 0.1733799806652684, 0.1362237766939547, 0.0938325693794663, 0.0478354290456362, 0.1633203706095471, 0.2265318615882219, 0.2133883476483184, 0.1920444391778541, 0.1633203706095471, 0.1283199917898342, 0.0883883476483185, 0.0450599888754343, 0.1469844503024199, 0.2038732892122293, 0.1920444391778541, 0.1728354290456362, 0.1469844503024199, 0.1154849415639109, 0.0795474112858021, 0.0405529186026822, 0.1250000000000000, 0.1733799806652684, 0.1633203706095471, 0.1469844503024199, 0.1250000000000000, 0.0982118697983878, 0.0676495125182746, 0.0344874224103679, 0.0982118697983878, 0.1362237766939547, 0.1283199917898342, 0.1154849415639109, 0.0982118697983878, 0.0771645709543638, 0.0531518809229535, 0.0270965939155924, 0.0676495125182746, 0.0938325693794663, 0.0883883476483185, 0.0795474112858021, 0.0676495125182746, 0.0531518809229535, 0.0366116523516816, 0.0186644585125857, 0.0344874224103679, 0.0478354290456362, 0.0450599888754343, 0.0405529186026822, 0.0344874224103679, 0.0270965939155924, 0.0186644585125857, 0.0095150584360892, }; void idct8(uint block, uint offset, uint stride) { float t0, t1, t2, t3, t4, t5, t6, t7, u8; float u0, u1, u2, u3, u4, u5, u6, u7; /* Input */ t0 = blocks[block][0*stride + offset]; u4 = blocks[block][1*stride + offset]; t2 = blocks[block][2*stride + offset]; u6 = blocks[block][3*stride + offset]; t1 = blocks[block][4*stride + offset]; u5 = blocks[block][5*stride + offset]; t3 = blocks[block][6*stride + offset]; u7 = blocks[block][7*stride + offset]; /* Embedded scaled inverse 4-point Type-II DCT */ u0 = t0 + t1; u1 = t0 - t1; u3 = t2 + t3; u2 = (t2 - t3)*(1.4142135623730950488016887242097f) - u3; t0 = u0 + u3; t3 = u0 - u3; t1 = u1 + u2; t2 = u1 - u2; /* Embedded scaled inverse 4-point Type-IV DST */ t5 = u5 + u6; t6 = u5 - u6; t7 = u4 + u7; t4 = u4 - u7; u7 = t7 + t5; u5 = (t7 - t5)*(1.4142135623730950488016887242097f); u8 = (t4 + t6)*(1.8477590650225735122563663787936f); u4 = u8 - t4*(1.0823922002923939687994464107328f); u6 = u8 - t6*(2.6131259297527530557132863468544f); t7 = u7; t6 = t7 - u6; t5 = t6 + u5; t4 = t5 - u4; /* Butterflies */ u0 = t0 + t7; u7 = t0 - t7; u6 = t1 + t6; u1 = t1 - t6; u2 = t2 + t5; u5 = t2 - t5; u4 = t3 + t4; u3 = t3 - t4; /* Output */ blocks[block][0*stride + offset] = u0; blocks[block][1*stride + offset] = u1; blocks[block][2*stride + offset] = u2; blocks[block][3*stride + offset] = u3; blocks[block][4*stride + offset] = u4; blocks[block][5*stride + offset] = u5; blocks[block][6*stride + offset] = u6; blocks[block][7*stride + offset] = u7; } void main(void) { const uint tile_idx = gl_WorkGroupID.y*gl_NumWorkGroups.x + gl_WorkGroupID.x; TileData td = tile_data[tile_idx]; if (expectEXT(td.pos.x >= frame_size.x, false)) return; uint64_t pkt_offset = uint64_t(pkt_data) + td.offset; u8vec2buf hdr_data = u8vec2buf(pkt_offset); int qscale = pack16(hdr_data[0].v.yx); const ivec2 offs = td.pos + ivec2(COMP_ID & 1, COMP_ID >> 1); const uint w = min(tile_size.x, frame_size.x - td.pos.x) / 2; const uint nb_blocks = w / 8; /* We have to do non-uniform access, so copy it */ uint8_t qmat_buf[64] = qmat; [[unroll]] for (uint i = gl_LocalInvocationID.x; i < 64; i += gl_WorkGroupSize.x) { int v = int(imageLoad(dst, offs + 2*ivec2(BLOCK_ID*8, 0) + scan[i])[0]); float vf = float(sign_extend(v, 16)) / 32768.0; vf *= qmat_buf[i] * qscale; blocks[BLOCK_ID][COMP_ID*64 + i] = (vf / (64*4.56)) * idct_scale[i]; } barrier(); idct8(BLOCK_ID, COMP_ID*64 + ROW_ID*8, 1); blocks[BLOCK_ID][COMP_ID*64 + ROW_ID] += 0.5; barrier(); idct8(BLOCK_ID, COMP_ID*64 + ROW_ID, 8); barrier(); [[unroll]] for (uint i = gl_LocalInvocationID.x; i < 64; i += gl_WorkGroupSize.x) { int v = int(round(blocks[BLOCK_ID][COMP_ID*64 + i]*4095.0)); v = clamp(v, 0, 4095); v <<= 4; imageStore(dst, offs + 2*ivec2(BLOCK_ID*8 + (i & 7), i >> 3), ivec4(v)); } }