// 
// sub_180005220 from 0x180005220 to 0x180005515 (757 bytes)
// 
// 2 xrefs:
// > sub_180004520 @ 0x1800045fa
// > <???> @ 0x1801f62d0
// 
// This function is hookable on this platform!
// 
// Using BromaIDA 8.0.0 @ https://github.com/Stazzical/BromaIDA
// Using bindings at commit ba9f177b at Thu Jul 16 22:52:06 2026 from https://github.com/geode-sdk/bindings
// 

__m128 *__fastcall sub_180005220(
        __m128 *a1,
        __m128 *a2,
        _OWORD *a3,
        __m128 *a4,
        __m128 *a5,
        __m128 *a6,
        __m128 *a7,
        int a8,
        int a9,
        int a10,
        int a11)
{
  int i; // eax
  __m128 v14; // xmm2
  __m128 v15; // xmm2
  __m128 v16; // xmm1
  int v17; // edx
  __int32 v18; // eax
  __m128 *result; // rax
  int j; // edi
  __m128 v22; // xmm1
  __m128 v23; // xmm0
  __m128 v24; // xmm2
  __m128 v25; // xmm4
  __m128 v26; // xmm0
  __m128 v27; // xmm3
  int v28; // ebx
  float v29; // xmm0_4
  float v30; // xmm2_4
  float v31; // xmm2_4
  int k; // edi
  __m128 v33; // xmm2
  __m128 v34; // xmm0
  __m128 v35; // xmm5
  __m128 v36; // xmm4
  __m128 v37; // xmm1
  __m128 v38; // xmm1
  int v39; // ebx
  signed __int64 v40; // r9
  signed __int64 v41; // r10
  float v42; // xmm0_4
  float v43; // xmm2_4
  float v44; // xmm0_4
  int m; // r9d
  __m128 v46; // xmm2
  __m128 v47; // xmm2
  __m128 v48; // xmm1
  int v49; // r8d
  signed __int64 v50; // rdx
  float v51; // xmm0_4

  for ( i = a8 >> 2; i; --i )
  {
    v14 = a4[-1];
    --a4;
    --a5;
    v15 = _mm_shuffle_ps(v14, v14, 27);
    v16 = _mm_shuffle_ps(*a5, *a5, 27);
    *a1 = _mm_unpacklo_ps(v15, v16);
    a1[1] = _mm_unpackhi_ps(v15, v16);
    a1 += 2;
  }
  v17 = a8 & 3;
  if ( (a8 & 3) != 0 )
  {
    do
    {
      v18 = a4[-1].m128_i32[3];
      a4 = (__m128 *)((char *)a4 - 4);
      a5 = (__m128 *)((char *)a5 - 4);
      a1->m128_i32[0] = v18;
      a1->m128_i32[1] = a5->m128_i32[0];
      a1 = (__m128 *)((char *)a1 + 8);
      --v17;
    }
    while ( v17 );
  }
  result = a7;
  for ( j = a9 >> 2; j; --j )
  {
    v22 = *(__m128 *)a3;
    --a4;
    --a2;
    v23 = *a6;
    --a5;
    ++a3;
    ++a6;
    v24 = _mm_shuffle_ps(*a2, *a2, 27);
    v25 = _mm_add_ps(_mm_mul_ps(_mm_shuffle_ps(*a4, *a4, 27), v24), _mm_mul_ps(v23, v22));
    v26 = *result++;
    v27 = _mm_add_ps(_mm_mul_ps(_mm_shuffle_ps(*a5, *a5, 27), v24), _mm_mul_ps(v26, v22));
    *a1 = _mm_unpacklo_ps(v25, v27);
    a1[1] = _mm_unpackhi_ps(v25, v27);
    a1 += 2;
  }
  v28 = a9 & 3;
  if ( (a9 & 3) != 0 )
  {
    do
    {
      a2 = (__m128 *)((char *)a2 - 4);
      v29 = a6->m128_f32[0] * *(float *)a3;
      a4 = (__m128 *)((char *)a4 - 4);
      a5 = (__m128 *)((char *)a5 - 4);
      a6 = (__m128 *)((char *)a6 + 4);
      a1->m128_f32[0] = (float)(a2->m128_f32[0] * a4->m128_f32[0]) + v29;
      v30 = result->m128_f32[0];
      result = (__m128 *)((char *)result + 4);
      v31 = v30 * *(float *)a3;
      a3 = (_OWORD *)((char *)a3 + 4);
      a1->m128_f32[1] = v31 + (float)(a5->m128_f32[0] * a2->m128_f32[0]);
      a1 = (__m128 *)((char *)a1 + 8);
      --v28;
    }
    while ( v28 );
  }
  for ( k = a10 >> 2; k; --k )
  {
    v33 = *(__m128 *)a3;
    --a2;
    --a6;
    v34 = *a4;
    --result;
    ++a3;
    ++a4;
    v35 = _mm_shuffle_ps(*a2, *a2, 27);
    v36 = _mm_sub_ps(_mm_mul_ps(v35, v34), _mm_mul_ps(_mm_shuffle_ps(*a6, *a6, 27), v33));
    v37 = *a5++;
    v38 = _mm_sub_ps(_mm_mul_ps(v37, v35), _mm_mul_ps(_mm_shuffle_ps(*result, *result, 27), v33));
    *a1 = _mm_unpacklo_ps(v36, v38);
    a1[1] = _mm_unpackhi_ps(v36, v38);
    a1 += 2;
  }
  v39 = a10 & 3;
  if ( (a10 & 3) != 0 )
  {
    v40 = (char *)a4 - (char *)a3;
    v41 = (char *)a5 - (char *)a3;
    do
    {
      a2 = (__m128 *)((char *)a2 - 4);
      v42 = a6[-1].m128_f32[3] * *(float *)a3;
      a6 = (__m128 *)((char *)a6 - 4);
      result = (__m128 *)((char *)result - 4);
      a1->m128_f32[0] = (float)(*(float *)((char *)a3 + v40) * a2->m128_f32[0]) - v42;
      v43 = *(float *)((char *)a3 + v41);
      v44 = result->m128_f32[0] * *(float *)a3;
      a3 = (_OWORD *)((char *)a3 + 4);
      a1->m128_f32[1] = (float)(v43 * a2->m128_f32[0]) - v44;
      a1 = (__m128 *)((char *)a1 + 8);
      --v39;
    }
    while ( v39 );
  }
  for ( m = a11 >> 2; m; --m )
  {
    v46 = a6[-1];
    --a6;
    --result;
    v47 = (__m128)_mm_xor_si128((__m128i)_mm_shuffle_ps(v46, v46, 27), xmmword_180144630[0]);
    v48 = (__m128)_mm_xor_si128((__m128i)_mm_shuffle_ps(*result, *result, 27), xmmword_180144630[0]);
    *a1 = _mm_unpacklo_ps(v47, v48);
    a1[1] = _mm_unpackhi_ps(v47, v48);
    a1 += 2;
  }
  v49 = a11 & 3;
  if ( (a11 & 3) != 0 )
  {
    v50 = (char *)a6 - (char *)result;
    do
    {
      v51 = *(float *)((char *)&result->m128_f32[-1] + v50);
      result = (__m128 *)((char *)result - 4);
      a1 = (__m128 *)((char *)a1 + 8);
      a1[-1].m128_f32[2] = -v51;
      a1[-1].m128_f32[3] = -result->m128_f32[0];
      --v49;
    }
    while ( v49 );
  }
  return result;
}