Files
renderdoc/renderdoc/data/glsl/vk_ms2buffer.comp
T
2023-02-01 12:23:32 +00:00

123 lines
4.4 KiB
Plaintext

/******************************************************************************
* The MIT License (MIT)
*
* Copyright (c) 2020-2023 Baldur Karlsson
*
* 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 "glsl_globals.h"
#extension GL_EXT_samplerless_texture_functions : require
layout(local_size_x = MS_DISPATCH_LOCAL_SIZE, local_size_y = 1, local_size_z = 1) in;
layout(binding = 0) uniform utexture2DMSArray srcMS;
// binding = 1 used as stencil read in the depth-stencil copy fragment shaders
layout(binding = 2, std430) writeonly buffer dstBuf
{
uint result[];
};
layout(push_constant) uniform multisamplePush
{
int textureWidth;
int baseSlice;
int baseSample;
int byteSize;
int maxInvocationID;
int dispatchOffset;
}
mscopy;
#define textureWidth (mscopy.textureWidth)
#define baseSlice (mscopy.baseSlice)
#define baseSample (mscopy.baseSample)
#define byteSize (mscopy.byteSize)
#define maxInvocationID (mscopy.maxInvocationID)
#define dispatchOffset (mscopy.dispatchOffset)
void main()
{
int slice = baseSlice;
int sampleIdx = baseSample;
uint idx = gl_GlobalInvocationID.x;
if(int(idx) >= maxInvocationID)
{
return;
}
// for byteSize < 4, we sample multiple pixels to fill a single output buffer element
if(byteSize == 1)
{
int pxIdx = int(idx * 4);
int x0 = (pxIdx + 0) % textureWidth;
int y0 = (pxIdx + 0) / textureWidth;
int x1 = (pxIdx + 1) % textureWidth;
int y1 = (pxIdx + 1) / textureWidth;
int x2 = (pxIdx + 2) % textureWidth;
int y2 = (pxIdx + 2) / textureWidth;
int x3 = (pxIdx + 3) % textureWidth;
int y3 = (pxIdx + 3) / textureWidth;
uvec4 data = uvec4(texelFetch(srcMS, ivec3(x0, y0, slice), sampleIdx).x,
texelFetch(srcMS, ivec3(x1, y1, slice), sampleIdx).x,
texelFetch(srcMS, ivec3(x2, y2, slice), sampleIdx).x,
texelFetch(srcMS, ivec3(x3, y3, slice), sampleIdx).x);
result[dispatchOffset + idx] = (data.x << 0 | data.y << 8 | data.z << 16 | data.w << 24);
}
else if(byteSize == 2)
{
int pxIdx = int(idx * 2);
int x0 = (pxIdx + 0) % textureWidth;
int y0 = (pxIdx + 0) / textureWidth;
int x1 = (pxIdx + 1) % textureWidth;
int y1 = (pxIdx + 1) / textureWidth;
uvec2 data = uvec2(texelFetch(srcMS, ivec3(x0, y0, slice), sampleIdx).x,
texelFetch(srcMS, ivec3(x1, y1, slice), sampleIdx).x);
result[dispatchOffset + idx] = (data.x << 0 | data.y << 16);
}
else if(byteSize == 4)
{
int x0 = int(idx) % textureWidth;
int y0 = int(idx) / textureWidth;
uint data = texelFetch(srcMS, ivec3(x0, y0, slice), sampleIdx).x;
result[dispatchOffset + idx] = data;
}
else if(byteSize == 8)
{
int x0 = int(idx) % textureWidth;
int y0 = int(idx) / textureWidth;
uvec2 data = texelFetch(srcMS, ivec3(x0, y0, slice), sampleIdx).xy;
result[dispatchOffset + (idx * 2)] = data.x;
result[dispatchOffset + (idx * 2 + 1)] = data.y;
}
else if(byteSize == 16)
{
int x0 = int(idx) % textureWidth;
int y0 = int(idx) / textureWidth;
uvec4 data = texelFetch(srcMS, ivec3(x0, y0, slice), sampleIdx);
result[dispatchOffset + (idx * 4)] = data.x;
result[dispatchOffset + (idx * 4 + 1)] = data.y;
result[dispatchOffset + (idx * 4 + 2)] = data.z;
result[dispatchOffset + (idx * 4 + 3)] = data.w;
}
}