RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-26 04:17:34 +00:00
Code Issues Activity Actions

LibGL: Implement GL_LINEAR texture filter

Browse source
This commit is contained in:
Stephan Unverwerth 2021-08-12 20:37:38 +02:00 • committed by Andreas Kling
parent 00603d9fd0
commit 00d527bc25
1 changed files with 50 additions and 43 deletions
Download patch file Download diff file Expand all files Collapse all files

93
Userland/Libraries/LibGL/Tex/Sampler2D.cpp
View file

@ -29,6 +29,27 @@ static constexpr float wrap_clamp(float value)
return clamp(value, 0.0f, 1.0f); return clamp(value, 0.0f, 1.0f);
} }
static constexpr float wrap(float value, GLint mode)
{
switch (mode) {
case GL_REPEAT:
return wrap_repeat(value);
// FIXME: These clamp modes actually have slightly different behaviour
case GL_CLAMP:
case GL_CLAMP_TO_BORDER:
case GL_CLAMP_TO_EDGE:
return wrap_clamp(value);
case GL_MIRRORED_REPEAT:
return wrap_mirrored_repeat(value);
break;
default:
VERIFY_NOT_REACHED();
}
}
FloatVector4 Sampler2D::sample(FloatVector2 const& uv) const FloatVector4 Sampler2D::sample(FloatVector2 const& uv) const
{ {
// FIXME: Calculate the correct mipmap level here, need to receive uv derivatives for that // FIXME: Calculate the correct mipmap level here, need to receive uv derivatives for that
@ -36,53 +57,39 @@ FloatVector4 Sampler2D::sample(FloatVector2 const& uv) const
MipMap const& mip = m_texture.mipmap(lod); MipMap const& mip = m_texture.mipmap(lod);
float x = uv.x(); float x = wrap(uv.x(), m_wrap_t_mode);
float y = uv.y(); float y = wrap(uv.y(), m_wrap_s_mode);
switch (m_wrap_s_mode) {
case GL_REPEAT:
x = wrap_repeat(x);
break;
// FIXME: These clamp modes actually have slightly different behaviour
case GL_CLAMP:
case GL_CLAMP_TO_BORDER:
case GL_CLAMP_TO_EDGE:
x = wrap_clamp(x);
break;
case GL_MIRRORED_REPEAT:
x = wrap_mirrored_repeat(x);
break;
default:
VERIFY_NOT_REACHED();
}
switch (m_wrap_t_mode) {
case GL_REPEAT:
y = wrap_repeat(y);
break;
// FIXME: These clamp modes actually have slightly different behaviour
case GL_CLAMP:
case GL_CLAMP_TO_BORDER:
case GL_CLAMP_TO_EDGE:
y = wrap_clamp(y);
break;
case GL_MIRRORED_REPEAT:
y = wrap_mirrored_repeat(y);
break;
default:
VERIFY_NOT_REACHED();
}
x *= mip.width() - 1; x *= mip.width() - 1;
y *= mip.height() - 1; y *= mip.height() - 1;
return mip.texel(static_cast<unsigned>(x), static_cast<unsigned>(y)); // Sampling implemented according to https://www.khronos.org/registry/OpenGL/specs/gl/glspec121.pdf Chapter 3.8
if (m_mag_filter == GL_NEAREST) {
return mip.texel(static_cast<unsigned>(x) % mip.width(), static_cast<unsigned>(y) % mip.height());
} else if (m_mag_filter == GL_LINEAR) {
// FIXME: Implement different sampling points for wrap modes other than GL_REPEAT
x -= 0.5f;
y -= 0.5f;
unsigned i0 = static_cast<unsigned>(x) % mip.width();
unsigned j0 = static_cast<unsigned>(y) % mip.height();
unsigned i1 = (i0 + 1) % mip.width();
unsigned j1 = (j0 + 1) % mip.height();
auto t0 = mip.texel(i0, j0);
auto t1 = mip.texel(i1, j0);
auto t2 = mip.texel(i0, j1);
auto t3 = mip.texel(i1, j1);
float frac_x = x - floorf(x);
float frac_y = y - floorf(y);
return t0 * (1 - frac_x) * (1 - frac_y) + t1 * frac_x * (1 - frac_y) + t2 * (1 - frac_x) * frac_y + t3 * frac_x * frac_y;
} else {
VERIFY_NOT_REACHED();
}
} }
} }