mul

Float4 result;
// Simple stuff
result.x = Dot(vector, matrix.x);
result.y = Dot(vector, matrix.y);
result.z = Dot(vector, matrix.z);
result.w = Dot(vector, matrix.w);
return result;

// COMPILES INTO //

mov         qword ptr [rcx+4],rax
mov         qword ptr [rcx],rax
mov         qword ptr [rcx+8],rax
mov         dword ptr [rcx+4],eax
mov         rax,rcx
movss       xmm2,dword ptr [r8+4]
movss       xmm0,dword ptr [rdx]
movss       xmm1,dword ptr [r8+8]
mulss       xmm0,dword ptr [r8]
mulss       xmm2,dword ptr [rdx+4]
mulss       xmm1,dword ptr [rdx+8]
addss       xmm2,xmm0
movss       xmm0,dword ptr [r8+0Ch]
addss       xmm2,xmm1
mulss       xmm0,dword ptr [rdx+0Ch]
addss       xmm2,xmm0
movss       dword ptr [rcx],xmm2

movss       xmm3,dword ptr [r8+14h]
movss       xmm0,dword ptr [r8+10h]
mulss       xmm0,dword ptr [rdx]
mulss       xmm3,dword ptr [rdx+4]
movss       xmm1,dword ptr [r8+18h]
addss       xmm3,xmm0
movss       xmm0,dword ptr [r8+1Ch]
mulss       xmm0,dword ptr [rdx+0Ch]
mulss       xmm1,dword ptr [rdx+8]

addss       xmm3,xmm1
addss       xmm3,xmm0
movss       dword ptr [rcx+4],xmm3

movss       xmm0,dword ptr [r8+20h]
movss       xmm1,dword ptr [r8+28h]
mulss       xmm0,dword ptr [rdx]
mulss       xmm1,dword ptr [rdx+8]
movss       xmm2,dword ptr [r8+24h]
mulss       xmm2,dword ptr [rdx+4]
addss       xmm2,xmm0
movss       xmm0,dword ptr [r8+2Ch]
mulss       xmm0,dword ptr [rdx+0Ch]
addss       xmm2,xmm1
addss       xmm2,xmm0
movss       dword ptr [rcx+8],xmm2

movss       xmm0,dword ptr [rdx]
movss       xmm3,dword ptr [r8+34h]
mulss       xmm3,dword ptr [rdx+4]
mulss       xmm0,dword ptr [r8+30h]
movss       xmm1,dword ptr [r8+38h]
addss       xmm3,xmm0
movss       xmm0,dword ptr [r8+3Ch]
mulss       xmm1,dword ptr [rdx+8]
mulss       xmm0,dword ptr [rdx+0Ch]
addss       xmm3,xmm1
addss       xmm3,xmm0
movss       dword ptr [rcx+0Ch],xmm3