// 
// sub_180004520 from 0x180004520 to 0x180004fb9 (2713 bytes)
// 
// 5 xrefs:
// > sub_180003D30 @ 0x180003ea2
// > sub_180005520 @ 0x180005e6b
// > <???> @ 0x18017d694
// > <???> @ 0x18017d6f0
// > <???> @ 0x1801f627c
// 
// 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
// 

__int64 __fastcall sub_180004520(
        __int64 a1,
        float *a2,
        __int64 a3,
        _OWORD *a4,
        __int64 a5,
        __int64 a6,
        __int64 a7,
        int a8,
        int a9,
        int a10,
        int a11)
{
  _OWORD *v11; // rbp
  __int64 v12; // r12
  int v13; // r15d
  float *v14; // r11
  __int64 v15; // rbx
  __int64 result; // rax
  __int64 v17; // r10
  char v18; // r9
  int v19; // r13d
  int v20; // ecx
  __int64 v21; // r8
  int v22; // edi
  __m128 *v23; // r9
  __m128 *v24; // rcx
  __m128 v25; // xmm0
  __int64 v26; // rax
  __m128 v27; // xmm1
  __m128 v28; // xmm4
  __m128 v29; // xmm2
  __m128 v30; // xmm3
  __m128 v31; // xmm4
  __m128 v32; // xmm3
  __m128 v33; // xmm0
  __m128 v34; // xmm2
  int v35; // esi
  __m128i *v36; // rdi
  float *v37; // r9
  __int64 v38; // rax
  __m128i v39; // xmm2
  int i; // r9d
  int v41; // r8d
  unsigned int *v42; // rcx
  char v43; // r9
  int v44; // ecx
  int v45; // r8d
  int v46; // r13d
  __int64 v47; // r15
  __m128 v48; // xmm6
  __m128 v49; // xmm7
  __int64 v50; // rsi
  __m128 *v51; // r14
  __m128 *v52; // r9
  __m128 *v53; // rdi
  __m128 v54; // xmm0
  __m128 v55; // xmm1
  __m128 v56; // xmm5
  __m128 v57; // xmm2
  __m128 v58; // xmm1
  __m128 v59; // xmm0
  __m128 v60; // xmm2
  __m128 v61; // xmm5
  __m128 v62; // xmm4
  __m128 v63; // xmm3
  __m128 v64; // xmm5
  __m128 v65; // xmm1
  __m128 v66; // xmm1
  __m128 v67; // xmm2
  __m128 v68; // xmm4
  __m128 v69; // xmm0
  __m128 v70; // xmm5
  int v71; // r14d
  __m128 *v72; // r9
  __m128 *v73; // rdi
  float *v74; // r8
  __m128 v75; // xmm0
  __m128 v76; // xmm2
  __m128 v77; // xmm0
  __m128i v78; // xmm2
  int j; // r9d
  float v80; // xmm2_4
  float v81; // xmm3_4
  int v82; // r8d
  float *v83; // rcx
  float v84; // xmm1_4
  float v85; // xmm0_4
  char v86; // r9
  int v87; // r8d
  int v88; // r13d
  int v89; // eax
  __int64 v90; // r15
  __m128 v91; // xmm7
  __m128 v92; // xmm8
  __int64 v93; // r9
  int v94; // ecx
  __m128 v95; // xmm7
  __m128 *v96; // r8
  __m128 *v97; // rdi
  __m128 *v98; // rsi
  __m128 v99; // xmm0
  __m128 v100; // xmm1
  __m128 v101; // xmm2
  __m128 v102; // xmm5
  __m128 v103; // xmm6
  __m128 v104; // xmm0
  __m128 v105; // xmm3
  __m128 v106; // xmm1
  __m128 v107; // xmm4
  __m128 v108; // xmm2
  __m128 v109; // xmm1
  __m128 v110; // xmm6
  __m128 v111; // xmm0
  __m128 v112; // xmm4
  int v113; // r14d
  __m128 *v114; // rdi
  __m128 *v115; // rsi
  float *v116; // r8
  __m128 v117; // xmm0
  __int64 v118; // rax
  __m128 v119; // xmm1
  __m128i v120; // xmm2
  int v121; // r9d
  __int64 v122; // r12
  __int64 v123; // rbx
  float v124; // xmm3_4
  int v125; // r8d
  float v126; // xmm2_4
  float *v127; // rcx
  float v128; // xmm1_4
  float v129; // xmm0_4
  int v130; // ebp
  unsigned int v131; // r13d
  __int64 v132; // r12
  int v133; // r8d
  __int64 v134; // rdi
  __m128 *v135; // rcx
  __m128 v136; // xmm4
  __m128 v137; // xmm0
  __m128 v138; // xmm3
  __m128 v139; // xmm1
  __m128 v140; // xmm3
  __m128 v141; // xmm2
  __m128 v142; // xmm4
  __m128 v143; // xmm1
  __m128 v144; // xmm0
  __m128 v145; // xmm3
  int v146; // ebx
  float *v147; // r8
  __m128 *v148; // r9
  __m128 v149; // xmm2
  __m128i v150; // xmm2
  int v151; // esi
  int v152; // ecx
  float v153; // xmm0_4
  __int64 v154; // [rsp+60h] [rbp-68h]
  __int64 v155; // [rsp+60h] [rbp-68h]
  int v156; // [rsp+D0h] [rbp+8h]
  int v157; // [rsp+D0h] [rbp+8h]
  int v158; // [rsp+F0h] [rbp+28h]
  int v159; // [rsp+F0h] [rbp+28h]
  int v160; // [rsp+F8h] [rbp+30h]

  v11 = a4;
  v12 = a3;
  v13 = a7;
  v14 = a2;
  v15 = *(_QWORD *)(a1 + 16) + 4LL * (int)a5;
  result = *(_QWORD *)(a1 + 8);
  v17 = result + 4LL * (int)a6;
  if ( (_DWORD)a7 == 1 )
    return sub_180004FC0((_DWORD)a2, a3, (_DWORD)a4, v15, v17, a8, a9, a10, a11);
  if ( (_DWORD)a7 == 2 )
    return sub_180005220(
             (_DWORD)a2,
             a3,
             (_DWORD)a4,
             *(_DWORD *)(a1 + 16) + 4 * (int)a5,
             v15 + 2048,
             result + 4LL * (int)a6,
             v17 + 4096,
             a8,
             a9,
             a10,
             a11);
  v18 = a8;
  v19 = a8 >> 2;
  if ( a8 >> 2 )
  {
    v20 = a7 & 3;
    result = 16LL * (_DWORD)a7;
    do
    {
      v15 -= 16;
      v21 = 0;
      v22 = (int)a7 >> 2;
      if ( (int)a7 >> 2 )
      {
        v23 = (__m128 *)&v14[3 * (int)a7];
        v24 = (__m128 *)(v15 + 4096);
        do
        {
          v25 = v24[-128];
          v26 = v21 + (int)a7;
          v21 += 4;
          ++v23;
          v27 = v24[128];
          v28 = v24[-256];
          v29 = *v24;
          v24 += 512;
          v30 = v28;
          v31 = _mm_shuffle_ps(v28, v25, 238);
          v32 = _mm_shuffle_ps(v30, v25, 68);
          v33 = _mm_shuffle_ps(v29, v27, 68);
          v34 = _mm_shuffle_ps(v29, v27, 238);
          *(__m128 *)((char *)v23 - 12 * (int)a7 - 16) = _mm_shuffle_ps(v31, v34, 221);
          *(__m128 *)&v14[v26] = _mm_shuffle_ps(v31, v34, 136);
          *(__m128 *)((char *)v23 - 4 * (int)a7 - 16) = _mm_shuffle_ps(v32, v33, 221);
          v23[-1] = _mm_shuffle_ps(v32, v33, 136);
          --v22;
        }
        while ( v22 );
        v20 = a7 & 3;
        result = 16LL * (_DWORD)a7;
      }
      v35 = v20;
      if ( v20 )
      {
        v36 = (__m128i *)(v15 + (v21 << 11));
        v37 = &v14[2 * (int)a7 + (int)a7 + v21];
        do
        {
          v38 = v21 + (int)a7;
          ++v37;
          ++v21;
          v39 = *v36;
          v36 += 128;
          LODWORD(v37[-3 * (int)a7 - 1]) = _mm_srli_si128(v39, 12).m128i_u32[0];
          LODWORD(v14[v38]) = _mm_srli_si128(v39, 8).m128i_u32[0];
          LODWORD(v37[-(int)a7 - 1]) = _mm_srli_si128(v39, 4).m128i_u32[0];
          *((_DWORD *)v37 - 1) = _mm_srli_si128(v39, 0).m128i_u32[0];
          --v35;
        }
        while ( v35 );
        v20 = a7 & 3;
        result = 16LL * (_DWORD)a7;
      }
      v14 = (float *)((char *)v14 + result);
      --v19;
    }
    while ( v19 );
    v13 = a7;
    v18 = a8;
  }
  for ( i = v18 & 3; i; --i )
  {
    v15 -= 4;
    v41 = v13;
    if ( v13 )
    {
      v42 = (unsigned int *)v15;
      do
      {
        result = *v42;
        v42 += 512;
        *(_DWORD *)v14++ = result;
        --v41;
      }
      while ( v41 );
    }
  }
  v43 = a9;
  v44 = a9 >> 2;
  v158 = a9 >> 2;
  if ( a9 >> 2 )
  {
    result = (unsigned int)(4 * v13);
    v45 = v13 >> 2;
    v46 = v13 & 3;
    v47 = 4LL * (int)result;
    v154 = v47;
    v156 = v45;
    v160 = v46;
    do
    {
      v48 = *(__m128 *)(v12 - 16);
      v12 -= 16;
      v15 -= 16;
      v49 = *(__m128 *)v11;
      v50 = 0;
      if ( v45 )
      {
        v51 = (__m128 *)&v14[3 * (int)a7];
        v52 = (__m128 *)(v17 + 0x2000);
        v53 = (__m128 *)(v15 + 4096);
        do
        {
          v54 = v53[-256];
          result = v50 + (int)a7;
          v50 += 4;
          v55 = v53[-128];
          v56 = *v53;
          v57 = v53[128];
          v53 += 512;
          v58 = _mm_mul_ps(v55, v48);
          v59 = _mm_mul_ps(v54, v48);
          v60 = _mm_mul_ps(v57, v48);
          v61 = _mm_mul_ps(v56, v48);
          v62 = _mm_add_ps(_mm_mul_ps(v52[-512], v49), _mm_shuffle_ps(v59, v59, 27));
          v63 = _mm_add_ps(_mm_mul_ps(v52[-256], v49), _mm_shuffle_ps(v58, v58, 27));
          v64 = _mm_add_ps(_mm_shuffle_ps(v61, v61, 27), _mm_mul_ps(v49, *v52));
          v65 = v52[256];
          v52 += 1024;
          v66 = _mm_add_ps(_mm_mul_ps(v65, v49), _mm_shuffle_ps(v60, v60, 27));
          v67 = _mm_shuffle_ps(v62, v63, 68);
          v68 = _mm_shuffle_ps(v62, v63, 238);
          v69 = _mm_shuffle_ps(v64, v66, 68);
          v70 = _mm_shuffle_ps(v64, v66, 238);
          *(__m128 *)((char *)v51 - 12 * (int)a7) = _mm_shuffle_ps(v67, v69, 136);
          *(__m128 *)&v14[result] = _mm_shuffle_ps(v67, v69, 221);
          *(__m128 *)((char *)v51 - 4 * (int)a7) = _mm_shuffle_ps(v68, v70, 136);
          *v51++ = _mm_shuffle_ps(v68, v70, 221);
          --v45;
        }
        while ( v45 );
        v44 = v158;
        v47 = v154;
      }
      v71 = v46;
      if ( v46 )
      {
        v72 = (__m128 *)(v17 + (v50 << 12));
        v73 = (__m128 *)(v15 + (v50 << 11));
        v74 = &v14[2 * (int)a7 + (int)a7 + v50];
        do
        {
          v75 = *v73;
          result = v50 + (int)a7;
          ++v50;
          v73 += 128;
          v76 = _mm_mul_ps(v48, v75);
          v77 = *v72;
          v72 += 256;
          v78 = (__m128i)_mm_add_ps(_mm_shuffle_ps(v76, v76, 27), _mm_mul_ps(v77, v49));
          LODWORD(v74[-3 * (int)a7]) = _mm_srli_si128(v78, 0).m128i_u32[0];
          LODWORD(v14[result]) = _mm_srli_si128(v78, 4).m128i_u32[0];
          LODWORD(v74[-(int)a7]) = _mm_srli_si128(v78, 8).m128i_u32[0];
          *(_DWORD *)v74++ = _mm_srli_si128(v78, 12).m128i_u32[0];
          --v71;
        }
        while ( v71 );
        v44 = v158;
        v46 = v160;
      }
      v45 = v156;
      v14 = (float *)((char *)v14 + v47);
      v17 += 16;
      ++v11;
      v158 = --v44;
    }
    while ( v44 );
    v13 = a7;
    v43 = a9;
  }
  for ( j = v43 & 3; j; --j )
  {
    v80 = *(float *)(v12 - 4);
    v12 -= 4;
    v81 = *(float *)v11;
    v15 -= 4;
    v82 = v13;
    if ( v13 )
    {
      result = v17;
      v83 = (float *)v15;
      do
      {
        v84 = v81 * *(float *)result;
        result += 4096;
        v85 = v80 * *v83;
        v83 += 512;
        *v14++ = v84 + v85;
        --v82;
      }
      while ( v82 );
    }
    v17 += 4;
    v11 = (_OWORD *)((char *)v11 + 4);
  }
  v86 = a10;
  v87 = a10 >> 2;
  v159 = a10 >> 2;
  if ( a10 >> 2 )
  {
    v88 = v13 & 3;
    v89 = 4 * v13;
    v157 = v88;
    v90 = (unsigned int)(v13 >> 2);
    result = 4LL * v89;
    v155 = result;
    do
    {
      v91 = *(__m128 *)(v12 - 16);
      v12 -= 16;
      v17 -= 16;
      v92 = *(__m128 *)v11;
      v93 = 0;
      v94 = v90;
      v95 = _mm_shuffle_ps(v91, v91, 27);
      if ( (_DWORD)v90 )
      {
        v96 = (__m128 *)&v14[2 * (int)a7];
        v97 = (__m128 *)(v17 + 0x2000);
        v98 = (__m128 *)(v15 + 4096);
        v93 = 4 * v90;
        do
        {
          v99 = v97[-512];
          v100 = v97[-256];
          v101 = *v97;
          v102 = v97[256];
          v97 += 1024;
          v103 = _mm_sub_ps(_mm_mul_ps(v98[-256], v95), _mm_mul_ps(_mm_shuffle_ps(v99, v99, 27), v92));
          v104 = *v98;
          v105 = _mm_sub_ps(_mm_mul_ps(v98[-128], v95), _mm_mul_ps(_mm_shuffle_ps(v100, v100, 27), v92));
          v106 = v98[128];
          v98 += 512;
          v107 = _mm_sub_ps(_mm_mul_ps(v95, v104), _mm_mul_ps(_mm_shuffle_ps(v101, v101, 27), v92));
          v108 = _mm_shuffle_ps(v103, v105, 68);
          v109 = _mm_sub_ps(_mm_mul_ps(v106, v95), _mm_mul_ps(_mm_shuffle_ps(v102, v102, 27), v92));
          v110 = _mm_shuffle_ps(v103, v105, 238);
          v111 = _mm_shuffle_ps(v107, v109, 68);
          v112 = _mm_shuffle_ps(v107, v109, 238);
          *(__m128 *)((char *)v96 - 8 * (int)a7) = _mm_shuffle_ps(v108, v111, 136);
          *(__m128 *)((char *)v96 - 4 * (int)a7) = _mm_shuffle_ps(v108, v111, 221);
          *v96 = _mm_shuffle_ps(v110, v112, 136);
          *(__m128 *)((char *)v96++ + 4 * (int)a7) = _mm_shuffle_ps(v110, v112, 221);
          --v94;
        }
        while ( v94 );
        v87 = v159;
        result = v155;
      }
      v113 = v88;
      if ( v88 )
      {
        v114 = (__m128 *)(v17 + (v93 << 12));
        v115 = (__m128 *)(v15 + (v93 << 11));
        v116 = &v14[2 * (int)a7 + (int)a7 + v93];
        do
        {
          v117 = *v115;
          v118 = v93 + (int)a7;
          ++v93;
          v119 = *v114;
          v115 += 128;
          v114 += 256;
          v120 = (__m128i)_mm_sub_ps(_mm_mul_ps(v95, v117), _mm_mul_ps(_mm_shuffle_ps(v119, v119, 27), v92));
          LODWORD(v116[-3 * (int)a7]) = _mm_srli_si128(v120, 0).m128i_u32[0];
          LODWORD(v14[v118]) = _mm_srli_si128(v120, 4).m128i_u32[0];
          LODWORD(v116[-(int)a7]) = _mm_srli_si128(v120, 8).m128i_u32[0];
          *(_DWORD *)v116++ = _mm_srli_si128(v120, 12).m128i_u32[0];
          --v113;
        }
        while ( v113 );
        v87 = v159;
        v88 = v157;
        result = v155;
      }
      v14 = (float *)((char *)v14 + result);
      v15 += 16;
      ++v11;
      v159 = --v87;
    }
    while ( v87 );
    v13 = a7;
    v86 = a10;
  }
  v121 = v86 & 3;
  if ( v121 )
  {
    v122 = v12 - v17;
    v123 = v15 - (_QWORD)v11;
    do
    {
      v124 = *(float *)v11;
      result = v17;
      v17 -= 4;
      v125 = v13;
      v126 = *(float *)(result + v122 - 4);
      if ( v13 )
      {
        result = v17;
        v127 = (float *)((char *)v11 + v123);
        do
        {
          v128 = v126 * *v127;
          v127 += 512;
          v129 = v124 * *(float *)result;
          result += 4096;
          *v14++ = v128 - v129;
          --v125;
        }
        while ( v125 );
      }
      v11 = (_OWORD *)((char *)v11 + 4);
      --v121;
    }
    while ( v121 );
  }
  v130 = a11 >> 2;
  if ( a11 >> 2 )
  {
    result = (unsigned int)(4 * v13);
    v131 = v13 >> 2;
    v132 = 4LL * (int)result;
    do
    {
      v17 -= 16;
      v133 = v13 >> 2;
      v134 = 0;
      if ( v131 )
      {
        result = (__int64)&v14[2 * (int)a7];
        v135 = (__m128 *)(v17 + 0x2000);
        v134 = 4LL * v131;
        do
        {
          result += 16;
          v136 = _mm_mul_ps(v135[-512], (__m128)xmmword_180144650);
          v137 = _mm_mul_ps(v135[-256], (__m128)xmmword_180144650);
          v138 = *v135;
          v139 = v135[256];
          v135 += 1024;
          v140 = _mm_mul_ps(v138, (__m128)xmmword_180144650);
          v141 = _mm_shuffle_ps(v136, v137, 68);
          v142 = _mm_shuffle_ps(v136, v137, 238);
          v143 = _mm_mul_ps(v139, (__m128)xmmword_180144650);
          v144 = _mm_shuffle_ps(v140, v143, 68);
          v145 = _mm_shuffle_ps(v140, v143, 238);
          *(__m128 *)(result - 8LL * (int)a7 - 16) = _mm_shuffle_ps(v142, v145, 221);
          *(__m128 *)(result - 4LL * (int)a7 - 16) = _mm_shuffle_ps(v142, v145, 136);
          *(__m128 *)(result - 16) = _mm_shuffle_ps(v141, v144, 221);
          *(__m128 *)(result + 4LL * (int)a7 - 16) = _mm_shuffle_ps(v141, v144, 136);
          --v133;
        }
        while ( v133 );
      }
      v146 = v13 & 3;
      if ( (v13 & 3) != 0 )
      {
        v147 = &v14[2 * (int)a7 + v134];
        v148 = (__m128 *)(v17 + (v134 << 12));
        result = -(__int64)(int)a7;
        do
        {
          ++v147;
          v149 = *v148;
          v148 += 256;
          v150 = (__m128i)_mm_mul_ps(v149, (__m128)xmmword_180144650);
          LODWORD(v147[-2 * (int)a7 - 1]) = _mm_srli_si128(v150, 12).m128i_u32[0];
          LODWORD(v147[-(int)a7 - 1]) = _mm_srli_si128(v150, 8).m128i_u32[0];
          *((_DWORD *)v147 - 1) = _mm_srli_si128(v150, 4).m128i_u32[0];
          LODWORD(v147[(int)a7 - 1]) = _mm_srli_si128(v150, 0).m128i_u32[0];
          --v146;
        }
        while ( v146 );
      }
      v14 = (float *)((char *)v14 + v132);
      --v130;
    }
    while ( v130 );
  }
  v151 = a11 & 3;
  if ( (a11 & 3) != 0 )
  {
    do
    {
      v17 -= 4;
      v152 = v13;
      if ( v13 )
      {
        result = v17;
        do
        {
          v153 = *(float *)result;
          result += 4096;
          *v14++ = -v153;
          --v152;
        }
        while ( v152 );
      }
      --v151;
    }
    while ( v151 );
  }
  return result;
}