// set uniform to 4D vector
void Shader::SetUniform(const c8 * const name, const float4 &val)
{
PUSH_ACTIVE_SHADER(t);
Activate();
glUniform4fv(GetUniformLocation(name),1, val.GetVec());
POP_ACTIVE_SHADER(t);
};
// Multiply matrix and 4D vector together
float4 Mat44::Mult(const float4 &m) const
{
Mat44 tr = Transpose();
__m128 matcols[] =
{
_mm_load_ps(tr.mat),
_mm_load_ps(tr.mat+4),
_mm_load_ps(tr.mat+8),
_mm_load_ps(tr.mat+12)
};
__m128 v = _mm_load_ps(m.GetVec());
// Broadcast vector into SSE registers
__m128 xb = _mm_shuffle_ps(v,v,0x00);
__m128 yb = _mm_shuffle_ps(v,v,0x55);
__m128 zb = _mm_shuffle_ps(v,v,0xAA);
__m128 wb = _mm_shuffle_ps(v,v,0xFF);
// Perform multiplication by matrix columns
xb = _mm_mul_ps(xb, matcols[0]);
yb = _mm_mul_ps(yb, matcols[1]);
zb = _mm_mul_ps(zb, matcols[2]);
wb = _mm_mul_ps(wb, matcols[3]);
// Add results
__m128 r = _mm_add_ps(_mm_add_ps(xb, yb),_mm_add_ps(zb, wb));
float4 returnVec;
_mm_store_ps(returnVec.GetVec(), r);
return returnVec;
};
