axmol/external/astc/astc_partition_tables.cpp

301 lines
7.7 KiB
C++
Raw Normal View History

2020-11-16 14:47:43 +08:00
// ----------------------------------------------------------------------------
// This confidential and proprietary software may be used only as authorised
// by a licensing agreement from Arm Limited.
// (C) COPYRIGHT 2011-2019 Arm Limited, ALL RIGHTS RESERVED
// The entire notice above must be reproduced on all authorised copies and
// copies may only be made to the extent permitted by a licensing agreement
// from Arm Limited.
// ----------------------------------------------------------------------------
/**
* @brief Functions for generating partition tables on demand.
*/
#include "astc_codec_internals.h"
static partition_info **partition_tables[4096];
/*
Produce a canonicalized representation of a partition pattern
The largest possible such representation is 432 bits, equal to 7 uint64_t values.
*/
static void gen_canonicalized_partition_table(int texel_count, const uint8_t * partition_table, uint64_t canonicalized[7])
{
int i;
for (i = 0; i < 7; i++)
canonicalized[i] = 0;
int mapped_index[4];
int map_weight_count = 0;
for (i = 0; i < 4; i++)
mapped_index[i] = -1;
for (i = 0; i < texel_count; i++)
{
int index = partition_table[i];
if (mapped_index[index] == -1)
mapped_index[index] = map_weight_count++;
uint64_t xlat_index = mapped_index[index];
canonicalized[i >> 5] |= xlat_index << (2 * (i & 0x1F));
}
}
static int compare_canonicalized_partition_tables(const uint64_t part1[7], const uint64_t part2[7])
{
if (part1[0] != part2[0])
return 0;
if (part1[1] != part2[1])
return 0;
if (part1[2] != part2[2])
return 0;
if (part1[3] != part2[3])
return 0;
if (part1[4] != part2[4])
return 0;
if (part1[5] != part2[5])
return 0;
if (part1[6] != part2[6])
return 0;
return 1;
}
/*
For a partition table, detect partitionss that are equivalent, then mark them as invalid. This reduces the number of partitions that the codec has to consider and thus improves encode
performance. */
static void partition_table_zap_equal_elements(int xdim, int ydim, int zdim, partition_info * pi)
{
int partition_tables_zapped = 0;
int texel_count = xdim * ydim * zdim;
int i, j;
uint64_t *canonicalizeds = new uint64_t[PARTITION_COUNT * 7];
for (i = 0; i < PARTITION_COUNT; i++)
{
gen_canonicalized_partition_table(texel_count, pi[i].partition_of_texel, canonicalizeds + i * 7);
}
for (i = 0; i < PARTITION_COUNT; i++)
{
for (j = 0; j < i; j++)
{
if (compare_canonicalized_partition_tables(canonicalizeds + 7 * i, canonicalizeds + 7 * j))
{
pi[i].partition_count = 0;
partition_tables_zapped++;
break;
}
}
}
delete[]canonicalizeds;
}
uint32_t hash52(uint32_t inp)
{
inp ^= inp >> 15;
inp *= 0xEEDE0891; // (2^4+1)*(2^7+1)*(2^17-1)
inp ^= inp >> 5;
inp += inp << 16;
inp ^= inp >> 7;
inp ^= inp >> 3;
inp ^= inp << 6;
inp ^= inp >> 17;
return inp;
}
int select_partition(int seed, int x, int y, int z, int partitioncount, int small_block)
{
if (small_block)
{
x <<= 1;
y <<= 1;
z <<= 1;
}
seed += (partitioncount - 1) * 1024;
uint32_t rnum = hash52(seed);
uint8_t seed1 = rnum & 0xF;
uint8_t seed2 = (rnum >> 4) & 0xF;
uint8_t seed3 = (rnum >> 8) & 0xF;
uint8_t seed4 = (rnum >> 12) & 0xF;
uint8_t seed5 = (rnum >> 16) & 0xF;
uint8_t seed6 = (rnum >> 20) & 0xF;
uint8_t seed7 = (rnum >> 24) & 0xF;
uint8_t seed8 = (rnum >> 28) & 0xF;
uint8_t seed9 = (rnum >> 18) & 0xF;
uint8_t seed10 = (rnum >> 22) & 0xF;
uint8_t seed11 = (rnum >> 26) & 0xF;
uint8_t seed12 = ((rnum >> 30) | (rnum << 2)) & 0xF;
// squaring all the seeds in order to bias their distribution
// towards lower values.
seed1 *= seed1;
seed2 *= seed2;
seed3 *= seed3;
seed4 *= seed4;
seed5 *= seed5;
seed6 *= seed6;
seed7 *= seed7;
seed8 *= seed8;
seed9 *= seed9;
seed10 *= seed10;
seed11 *= seed11;
seed12 *= seed12;
int sh1, sh2, sh3;
if (seed & 1)
{
sh1 = (seed & 2 ? 4 : 5);
sh2 = (partitioncount == 3 ? 6 : 5);
}
else
{
sh1 = (partitioncount == 3 ? 6 : 5);
sh2 = (seed & 2 ? 4 : 5);
}
sh3 = (seed & 0x10) ? sh1 : sh2;
seed1 >>= sh1;
seed2 >>= sh2;
seed3 >>= sh1;
seed4 >>= sh2;
seed5 >>= sh1;
seed6 >>= sh2;
seed7 >>= sh1;
seed8 >>= sh2;
seed9 >>= sh3;
seed10 >>= sh3;
seed11 >>= sh3;
seed12 >>= sh3;
int a = seed1 * x + seed2 * y + seed11 * z + (rnum >> 14);
int b = seed3 * x + seed4 * y + seed12 * z + (rnum >> 10);
int c = seed5 * x + seed6 * y + seed9 * z + (rnum >> 6);
int d = seed7 * x + seed8 * y + seed10 * z + (rnum >> 2);
// apply the saw
a &= 0x3F;
b &= 0x3F;
c &= 0x3F;
d &= 0x3F;
// remove some of the components if we are to output < 4 partitions.
if (partitioncount <= 3)
d = 0;
if (partitioncount <= 2)
c = 0;
if (partitioncount <= 1)
b = 0;
int partition;
if (a >= b && a >= c && a >= d)
partition = 0;
else if (b >= c && b >= d)
partition = 1;
else if (c >= d)
partition = 2;
else
partition = 3;
return partition;
}
void generate_one_partition_table(int xdim, int ydim, int zdim, int partition_count, int partition_index, partition_info * pt)
{
int small_block = (xdim * ydim * zdim) < 32;
uint8_t *partition_of_texel = pt->partition_of_texel;
int x, y, z, i;
for (z = 0; z < zdim; z++)
for (y = 0; y < ydim; y++)
for (x = 0; x < xdim; x++)
{
uint8_t part = select_partition(partition_index, x, y, z, partition_count, small_block);
*partition_of_texel++ = part;
}
int texels_per_block = xdim * ydim * zdim;
int counts[4];
for (i = 0; i < 4; i++)
counts[i] = 0;
for (i = 0; i < texels_per_block; i++)
{
int partition = pt->partition_of_texel[i];
pt->texels_of_partition[partition][counts[partition]++] = i;
}
for (i = 0; i < 4; i++)
pt->texels_per_partition[i] = counts[i];
if (counts[0] == 0)
pt->partition_count = 0;
else if (counts[1] == 0)
pt->partition_count = 1;
else if (counts[2] == 0)
pt->partition_count = 2;
else if (counts[3] == 0)
pt->partition_count = 3;
else
pt->partition_count = 4;
for (i = 0; i < 4; i++)
pt->coverage_bitmaps[i] = 0ULL;
const block_size_descriptor *bsd = get_block_size_descriptor(xdim, ydim, zdim);
int texels_to_process = bsd->texelcount_for_bitmap_partitioning;
for (i = 0; i < texels_to_process; i++)
{
int idx = bsd->texels_for_bitmap_partitioning[i];
pt->coverage_bitmaps[pt->partition_of_texel[idx]] |= 1ULL << i;
}
}
static void generate_partition_tables(int xdim, int ydim, int zdim)
{
int i;
partition_info *one_partition = new partition_info;
partition_info *two_partitions = new partition_info[1024];
partition_info *three_partitions = new partition_info[1024];
partition_info *four_partitions = new partition_info[1024];
partition_info **partition_table = new partition_info *[5];
partition_table[0] = NULL;
partition_table[1] = one_partition;
partition_table[2] = two_partitions;
partition_table[3] = three_partitions;
partition_table[4] = four_partitions;
generate_one_partition_table(xdim, ydim, zdim, 1, 0, one_partition);
for (i = 0; i < 1024; i++)
{
generate_one_partition_table(xdim, ydim, zdim, 2, i, two_partitions + i);
generate_one_partition_table(xdim, ydim, zdim, 3, i, three_partitions + i);
generate_one_partition_table(xdim, ydim, zdim, 4, i, four_partitions + i);
}
partition_table_zap_equal_elements(xdim, ydim, zdim, two_partitions);
partition_table_zap_equal_elements(xdim, ydim, zdim, three_partitions);
partition_table_zap_equal_elements(xdim, ydim, zdim, four_partitions);
partition_tables[xdim + 16 * ydim + 256 * zdim] = partition_table;
}
const partition_info *get_partition_table(int xdim, int ydim, int zdim, int partition_count)
{
int ptindex = xdim + 16 * ydim + 256 * zdim;
if (partition_tables[ptindex] == NULL)
generate_partition_tables(xdim, ydim, zdim);
return partition_tables[ptindex][partition_count];
}