piwalker/melonDS
1
0
Fork 0
mirror of https://github.com/melonDS-emu/melonDS.git synced 2025-07-25 15:19:53 -06:00
Code Issues Packages Projects Releases Wiki Activity

Compute shader renderer (#2041)

Browse Source 
* nothing works yet

* don't double buffer 3D framebuffers for the GL Renderer
looks like leftovers from when 3D+2D composition was done in the frontend

* oops

* it works!

* implement display capture for compute renderer
it's actually just all stolen from the regular OpenGL renderer

* fix bad indirect call

* handle cleanup properly

* add hires rendering to the compute shader renderer

* fix UB
also misc changes to use more unsigned multiplication
also fix framebuffer resize

* correct edge filling behaviour when AA is disabled

* fix full color textures

* fix edge marking (polygon id is 6-bit not 5)
also make the code a bit nicer

* take all edge cases into account for XMin/XMax calculation

* use hires coordinate again

* stop using fixed size buffers based on scale factor in shaders
this makes shader compile times tolerable on Wintel
- beginning of the shader cache
- increase size of tile idx in workdesc to 20 bits

* apparently & is not defined on bvec4
why does this even compile on Intel and Nvidia?

* put the texture cache into it's own file

* add compute shader renderer properly to the GUI
also add option to toggle using high resolution vertex coordinates

* unbind sampler object in compute shader renderer

* fix GetRangedBitMask for 64 bit aligned 64 bits
pretty embarassing

* convert NonStupidBitfield.h back to LF only new lines

* actually adapt to latest changes

* fix stupid merge

* actually make compute shader renderer work with newest changes

* show progress on shader compilation

* remove merge leftover
This commit is contained in:
RSDuck
2024-05-13 17:17:39 +02:00
committed by GitHub
parent c85a2103bb
commit 043244a56d
35 changed files with 4389 additions and 382 deletions
Download Patch File Download Diff File Expand all files Collapse all files

47
src/GPU_OpenGL.cpp
View File

@ -36,32 +36,27 @@ using namespace OpenGL;
std::optional<GLCompositor> GLCompositor::New() noexcept
{
assert(glBindAttribLocation != nullptr);
GLuint CompShader {};
std::array<GLuint, 3> CompShader {};
if (!OpenGL::BuildShaderProgram(kCompositorVS, kCompositorFS_Nearest, &CompShader[0], "CompositorShader"))
return std::nullopt;
glBindAttribLocation(CompShader[2], 0, "vPosition");
glBindAttribLocation(CompShader[2], 1, "vTexcoord");
glBindFragDataLocation(CompShader[2], 0, "oColor");
if (!OpenGL::LinkShaderProgram(CompShader.data()))
// OpenGL::LinkShaderProgram already deletes the shader program object
// if linking the shaders together failed.
if (!OpenGL::CompileVertexFragmentProgram(CompShader,
kCompositorVS, kCompositorFS_Nearest,
"CompositorShader",
{{"vPosition", 0}, {"vTexcoord", 1}},
{{"oColor", 0}}))
return std::nullopt;
return { GLCompositor(CompShader) };
}
GLCompositor::GLCompositor(std::array<GLuint, 3> compShader) noexcept : CompShader(compShader)
GLCompositor::GLCompositor(GLuint compShader) noexcept : CompShader(compShader)
{
CompScaleLoc = glGetUniformLocation(CompShader[2], "u3DScale");
Comp3DXPosLoc = glGetUniformLocation(CompShader[2], "u3DXPos");
CompScaleLoc = glGetUniformLocation(CompShader, "u3DScale");
Comp3DXPosLoc = glGetUniformLocation(CompShader, "u3DXPos");
glUseProgram(CompShader[2]);
GLuint screenTextureUniform = glGetUniformLocation(CompShader[2], "ScreenTex");
glUseProgram(CompShader);
GLuint screenTextureUniform = glGetUniformLocation(CompShader, "ScreenTex");
glUniform1i(screenTextureUniform, 0);
GLuint _3dTextureUniform = glGetUniformLocation(CompShader[2], "_3DTex");
GLuint _3dTextureUniform = glGetUniformLocation(CompShader, "_3DTex");
glUniform1i(_3dTextureUniform, 1);
// all this mess is to prevent bleeding
@ -136,7 +131,7 @@ GLCompositor::~GLCompositor()
glDeleteVertexArrays(1, &CompVertexArrayID);
glDeleteBuffers(1, &CompVertexBufferID);
OpenGL::DeleteShaderProgram(CompShader.data());
glDeleteProgram(CompShader);
}
@ -174,7 +169,7 @@ GLCompositor& GLCompositor::operator=(GLCompositor&& other) noexcept
CompVertices = other.CompVertices;
// Clean up these resources before overwriting them
OpenGL::DeleteShaderProgram(CompShader.data());
glDeleteProgram(CompShader);
CompShader = other.CompShader;
glDeleteBuffers(1, &CompVertexBufferID);
@ -244,11 +239,11 @@ void GLCompositor::Stop(const GPU& gpu) noexcept
glBindFramebuffer(GL_FRAMEBUFFER, 0);
}
void GLCompositor::RenderFrame(const GPU& gpu, GLRenderer& renderer) noexcept
void GLCompositor::RenderFrame(const GPU& gpu, Renderer3D& renderer) noexcept
{
int frontbuf = gpu.FrontBuffer;
int backbuf = gpu.FrontBuffer ^ 1;
glBindFramebuffer(GL_READ_FRAMEBUFFER, 0);
glBindFramebuffer(GL_DRAW_FRAMEBUFFER, CompScreenOutputFB[frontbuf]);
glBindFramebuffer(GL_DRAW_FRAMEBUFFER, CompScreenOutputFB[backbuf]);
glDisable(GL_DEPTH_TEST);
glDisable(GL_STENCIL_TEST);
@ -260,7 +255,7 @@ void GLCompositor::RenderFrame(const GPU& gpu, GLRenderer& renderer) noexcept
glClear(GL_COLOR_BUFFER_BIT);
// TODO: select more shaders (filtering, etc)
OpenGL::UseShaderProgram(CompShader.data());
glUseProgram(CompShader);
glUniform1ui(CompScaleLoc, Scale);
// TODO: support setting this midframe, if ever needed
@ -269,12 +264,12 @@ void GLCompositor::RenderFrame(const GPU& gpu, GLRenderer& renderer) noexcept
glActiveTexture(GL_TEXTURE0);
glBindTexture(GL_TEXTURE_2D, CompScreenInputTex);
if (gpu.Framebuffer[frontbuf][0] && gpu.Framebuffer[frontbuf][1])
if (gpu.Framebuffer[backbuf][0] && gpu.Framebuffer[backbuf][1])
{
glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, 256*3 + 1, 192, GL_RGBA_INTEGER,
GL_UNSIGNED_BYTE, gpu.Framebuffer[frontbuf][0].get());
GL_UNSIGNED_BYTE, gpu.Framebuffer[backbuf][0].get());
glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 192, 256*3 + 1, 192, GL_RGBA_INTEGER,
GL_UNSIGNED_BYTE, gpu.Framebuffer[frontbuf][1].get());
GL_UNSIGNED_BYTE, gpu.Framebuffer[backbuf][1].get());
}
glActiveTexture(GL_TEXTURE1);