mirror of
https://github.com/kmc-jp/n64-emu.git
synced 2025-04-02 10:21:43 -04:00
149 lines
4.8 KiB
Text
149 lines
4.8 KiB
Text
#version 450
|
|
/* Copyright (c) 2020 Themaister
|
|
*
|
|
* 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.
|
|
*/
|
|
|
|
#if SUBGROUP
|
|
#extension GL_KHR_shader_subgroup_basic : require
|
|
#extension GL_KHR_shader_subgroup_vote : require
|
|
#extension GL_KHR_shader_subgroup_ballot : require
|
|
#extension GL_KHR_shader_subgroup_arithmetic : require
|
|
#endif
|
|
#include "small_types.h"
|
|
|
|
layout(local_size_x_id = 3, local_size_y_id = 4) in;
|
|
|
|
#include "noise.h"
|
|
#include "debug.h"
|
|
#include "data_structures_buffers.h"
|
|
#include "memory_interfacing.h"
|
|
|
|
layout(set = 0, binding = 3, std430) readonly buffer ColorBuffer
|
|
{
|
|
mem_u8x4 elems[];
|
|
} color;
|
|
|
|
layout(set = 0, binding = 3, std430) readonly buffer ColorRawBuffer
|
|
{
|
|
uint elems[];
|
|
} raw_color;
|
|
|
|
layout(set = 0, binding = 4, std430) readonly buffer DepthBuffer
|
|
{
|
|
int elems[];
|
|
} depth;
|
|
|
|
layout(set = 0, binding = 5, std430) readonly buffer ShadeAlpha
|
|
{
|
|
mem_u8 elems[];
|
|
} shade_alpha;
|
|
|
|
layout(set = 0, binding = 6, std430) readonly buffer Coverage
|
|
{
|
|
mem_i8 elems[];
|
|
} coverage;
|
|
|
|
layout(std430, set = 0, binding = 7) readonly buffer TileInstanceOffset
|
|
{
|
|
uint elems[];
|
|
} tile_instance_offsets;
|
|
|
|
layout(push_constant, std430) uniform Registers
|
|
{
|
|
uint fb_addr_index;
|
|
uint fb_depth_addr_index;
|
|
uint fb_width;
|
|
uint fb_height;
|
|
uint group_mask;
|
|
} registers;
|
|
|
|
layout(constant_id = 5) const int MAX_PRIMITIVES = 256;
|
|
layout(constant_id = 6) const int MAX_WIDTH = 1024;
|
|
|
|
const int TILE_BINNING_STRIDE = MAX_PRIMITIVES / 32;
|
|
const int MAX_TILES_X = MAX_WIDTH / int(gl_WorkGroupSize.x);
|
|
|
|
// Overall architecture of the tiling is from RetroWarp.
|
|
|
|
void main()
|
|
{
|
|
int x = int(gl_GlobalInvocationID.x);
|
|
int y = int(gl_GlobalInvocationID.y);
|
|
ivec2 tile = ivec2(gl_WorkGroupID.xy);
|
|
|
|
int linear_tile = tile.x + tile.y * MAX_TILES_X;
|
|
int linear_tile_base = linear_tile * TILE_BINNING_STRIDE;
|
|
|
|
uint coarse_binned = tile_binning_coarse.elems[linear_tile] & registers.group_mask;
|
|
if (coarse_binned == 0u)
|
|
return;
|
|
|
|
init_tile(gl_GlobalInvocationID.xy,
|
|
registers.fb_width, registers.fb_height,
|
|
registers.fb_addr_index, registers.fb_depth_addr_index);
|
|
|
|
while (coarse_binned != 0u)
|
|
{
|
|
int mask_index = findLSB(coarse_binned);
|
|
coarse_binned &= ~uint(1 << mask_index);
|
|
|
|
uint tile_instance = tile_instance_offsets.elems[linear_tile_base + mask_index];
|
|
uint binned = tile_binning.elems[linear_tile_base + mask_index];
|
|
|
|
while (binned != 0u)
|
|
{
|
|
int i = findLSB(binned);
|
|
binned &= ~uint(1 << i);
|
|
uint primitive_index = uint(i + 32 * mask_index);
|
|
|
|
uint index = tile_instance * (gl_WorkGroupSize.x * gl_WorkGroupSize.y) + gl_LocalInvocationIndex;
|
|
int coverage = int(coverage.elems[index]);
|
|
|
|
if (coverage >= 0)
|
|
{
|
|
if ((coverage & COVERAGE_FILL_BIT) != 0)
|
|
{
|
|
fill_color(derived_setup.elems[primitive_index].fill_color);
|
|
}
|
|
else if ((coverage & COVERAGE_COPY_BIT) != 0)
|
|
{
|
|
uint word = raw_color.elems[index];
|
|
copy_pipeline(word, primitive_index);
|
|
}
|
|
else
|
|
{
|
|
ShadedData shaded;
|
|
shaded.combined = u8x4(color.elems[index]);
|
|
shaded.z_dith = depth.elems[index];
|
|
shaded.shade_alpha = u8(shade_alpha.elems[index]);
|
|
shaded.coverage_count = u8(coverage);
|
|
depth_blend(x, y, primitive_index, shaded);
|
|
}
|
|
}
|
|
|
|
tile_instance++;
|
|
}
|
|
}
|
|
|
|
finish_tile(gl_GlobalInvocationID.xy,
|
|
registers.fb_width, registers.fb_height,
|
|
registers.fb_addr_index, registers.fb_depth_addr_index);
|
|
}
|