n64-emu/third_party/parallel-rdp-standalone/parallel-rdp/shaders/ubershader.comp
2023-08-06 14:03:29 +09:00

103 lines
3.6 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.
*/
// RIP to any GPU which attempts to execute this monstrosity :)
#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 "debug.h"
#include "data_structures_buffers.h"
#include "noise.h"
#include "memory_interfacing.h"
#include "shading.h"
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);
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 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);
ShadedData shaded;
if (shade_pixel(x, y, primitive_index, shaded))
{
if ((shaded.coverage_count & COVERAGE_FILL_BIT) != 0)
fill_color(derived_setup.elems[primitive_index].fill_color);
else if ((shaded.coverage_count & COVERAGE_COPY_BIT) != 0)
copy_pipeline(shaded.z_dith, primitive_index);
else
depth_blend(x, y, primitive_index, shaded);
}
}
}
finish_tile(gl_GlobalInvocationID.xy,
registers.fb_width, registers.fb_height,
registers.fb_addr_index, registers.fb_depth_addr_index);
}