1 /////////////////////////////////////////////////////////////////////////////////// 2 /// OpenGL Mathematics (glm.g-truc.net) 3 /// 4 /// Copyright (c) 2005 - 2014 G-Truc Creation (www.g-truc.net) 5 /// Permission is hereby granted, free of charge, to any person obtaining a copy 6 /// of this software and associated documentation files (the "Software"), to deal 7 /// in the Software without restriction, including without limitation the rights 8 /// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 9 /// copies of the Software, and to permit persons to whom the Software is 10 /// furnished to do so, subject to the following conditions: 11 /// 12 /// The above copyright notice and this permission notice shall be included in 13 /// all copies or substantial portions of the Software. 14 /// 15 /// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 /// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 /// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 18 /// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 /// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 20 /// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN 21 /// THE SOFTWARE. 22 /// 23 /// @ref core 24 /// @file glm/core/intrinsic_geometric.inl 25 /// @date 2009-05-08 / 2011-06-15 26 /// @author Christophe Riccio 27 /////////////////////////////////////////////////////////////////////////////////// 28 29 namespace glm{ 30 namespace detail{ 31 32 //length 33 GLM_FUNC_QUALIFIER __m128 sse_len_ps(__m128 x) 34 { 35 __m128 dot0 = sse_dot_ps(x, x); 36 __m128 sqt0 = _mm_sqrt_ps(dot0); 37 return sqt0; 38 } 39 40 //distance 41 GLM_FUNC_QUALIFIER __m128 sse_dst_ps(__m128 p0, __m128 p1) 42 { 43 __m128 sub0 = _mm_sub_ps(p0, p1); 44 __m128 len0 = sse_len_ps(sub0); 45 return len0; 46 } 47 48 //dot 49 GLM_FUNC_QUALIFIER __m128 sse_dot_ps(__m128 v1, __m128 v2) 50 { 51 # if((GLM_ARCH & GLM_ARCH_AVX) == GLM_ARCH_AVX) 52 return _mm_dp_ps(v1, v2, 0xff); 53 # else 54 __m128 mul0 = _mm_mul_ps(v1, v2); 55 __m128 swp0 = _mm_shuffle_ps(mul0, mul0, _MM_SHUFFLE(2, 3, 0, 1)); 56 __m128 add0 = _mm_add_ps(mul0, swp0); 57 __m128 swp1 = _mm_shuffle_ps(add0, add0, _MM_SHUFFLE(0, 1, 2, 3)); 58 __m128 add1 = _mm_add_ps(add0, swp1); 59 return add1; 60 # endif 61 } 62 63 // SSE1 64 GLM_FUNC_QUALIFIER __m128 sse_dot_ss(__m128 v1, __m128 v2) 65 { 66 __m128 mul0 = _mm_mul_ps(v1, v2); 67 __m128 mov0 = _mm_movehl_ps(mul0, mul0); 68 __m128 add0 = _mm_add_ps(mov0, mul0); 69 __m128 swp1 = _mm_shuffle_ps(add0, add0, 1); 70 __m128 add1 = _mm_add_ss(add0, swp1); 71 return add1; 72 } 73 74 //cross 75 GLM_FUNC_QUALIFIER __m128 sse_xpd_ps(__m128 v1, __m128 v2) 76 { 77 __m128 swp0 = _mm_shuffle_ps(v1, v1, _MM_SHUFFLE(3, 0, 2, 1)); 78 __m128 swp1 = _mm_shuffle_ps(v1, v1, _MM_SHUFFLE(3, 1, 0, 2)); 79 __m128 swp2 = _mm_shuffle_ps(v2, v2, _MM_SHUFFLE(3, 0, 2, 1)); 80 __m128 swp3 = _mm_shuffle_ps(v2, v2, _MM_SHUFFLE(3, 1, 0, 2)); 81 __m128 mul0 = _mm_mul_ps(swp0, swp3); 82 __m128 mul1 = _mm_mul_ps(swp1, swp2); 83 __m128 sub0 = _mm_sub_ps(mul0, mul1); 84 return sub0; 85 } 86 87 //normalize 88 GLM_FUNC_QUALIFIER __m128 sse_nrm_ps(__m128 v) 89 { 90 __m128 dot0 = sse_dot_ps(v, v); 91 __m128 isr0 = _mm_rsqrt_ps(dot0); 92 __m128 mul0 = _mm_mul_ps(v, isr0); 93 return mul0; 94 } 95 96 //faceforward 97 GLM_FUNC_QUALIFIER __m128 sse_ffd_ps(__m128 N, __m128 I, __m128 Nref) 98 { 99 //__m128 dot0 = _mm_dot_ps(v, v); 100 //__m128 neg0 = _mm_neg_ps(N); 101 //__m128 sgn0 = _mm_sgn_ps(dot0); 102 //__m128 mix0 = _mm_mix_ps(N, neg0, sgn0); 103 //return mix0; 104 105 __m128 dot0 = sse_dot_ps(Nref, I); 106 __m128 sgn0 = sse_sgn_ps(dot0); 107 __m128 mul0 = _mm_mul_ps(sgn0, glm::detail::minus_one); 108 __m128 mul1 = _mm_mul_ps(N, mul0); 109 return mul1; 110 } 111 112 //reflect 113 GLM_FUNC_QUALIFIER __m128 sse_rfe_ps(__m128 I, __m128 N) 114 { 115 __m128 dot0 = sse_dot_ps(N, I); 116 __m128 mul0 = _mm_mul_ps(N, dot0); 117 __m128 mul1 = _mm_mul_ps(mul0, glm::detail::two); 118 __m128 sub0 = _mm_sub_ps(I, mul1); 119 return sub0; 120 } 121 122 //refract 123 GLM_FUNC_QUALIFIER __m128 sse_rfa_ps(__m128 I, __m128 N, __m128 eta) 124 { 125 __m128 dot0 = sse_dot_ps(N, I); 126 __m128 mul0 = _mm_mul_ps(eta, eta); 127 __m128 mul1 = _mm_mul_ps(dot0, dot0); 128 __m128 sub0 = _mm_sub_ps(glm::detail::one, mul0); 129 __m128 sub1 = _mm_sub_ps(glm::detail::one, mul1); 130 __m128 mul2 = _mm_mul_ps(sub0, sub1); 131 132 if(_mm_movemask_ps(_mm_cmplt_ss(mul2, glm::detail::zero)) == 0) 133 return glm::detail::zero; 134 135 __m128 sqt0 = _mm_sqrt_ps(mul2); 136 __m128 mul3 = _mm_mul_ps(eta, dot0); 137 __m128 add0 = _mm_add_ps(mul3, sqt0); 138 __m128 mul4 = _mm_mul_ps(add0, N); 139 __m128 mul5 = _mm_mul_ps(eta, I); 140 __m128 sub2 = _mm_sub_ps(mul5, mul4); 141 142 return sub2; 143 } 144 145 }//namespace detail 146 }//namespace glm 147
