[f734e7d] | 1 | // GENERATED CODE --- DO NOT EDIT --- |
---|
| 2 | // Code is produced by sasmodels.gen from sasmodels/models/MODEL.c |
---|
| 3 | |
---|
| 4 | #ifdef __OPENCL_VERSION__ |
---|
| 5 | # define USE_OPENCL |
---|
| 6 | #endif |
---|
| 7 | |
---|
| 8 | // If opencl is not available, then we are compiling a C function |
---|
| 9 | // Note: if using a C++ compiler, then define kernel as extern "C" |
---|
| 10 | #ifndef USE_OPENCL |
---|
| 11 | # ifdef __cplusplus |
---|
[95e861b] | 12 | #include <cstdio> |
---|
[f734e7d] | 13 | #include <cmath> |
---|
[750ffa5] | 14 | using namespace std; |
---|
[f734e7d] | 15 | #if defined(_MSC_VER) |
---|
[750ffa5] | 16 | # define kernel extern "C" __declspec( dllexport ) |
---|
| 17 | inline double trunc(double x) { return x>=0?floor(x):-floor(-x); } |
---|
[6ee9d39] | 18 | inline double fmin(double x, double y) { return x>y ? y : x; } |
---|
| 19 | inline double fmax(double x, double y) { return x<y ? y : x; } |
---|
[f734e7d] | 20 | #else |
---|
[750ffa5] | 21 | # define kernel extern "C" |
---|
[f734e7d] | 22 | #endif |
---|
[750ffa5] | 23 | inline void SINCOS(double angle, double &svar, double &cvar) { svar=sin(angle); cvar=cos(angle); } |
---|
[f734e7d] | 24 | # else |
---|
[95e861b] | 25 | #include <stdio.h> |
---|
[750ffa5] | 26 | #include <tgmath.h> // C99 type-generic math, so sin(float) => sinf |
---|
| 27 | // MSVC doesn't support C99, so no need for dllexport on C99 branch |
---|
[f734e7d] | 28 | #define kernel |
---|
[750ffa5] | 29 | #define SINCOS(angle,svar,cvar) do {const double _t_=angle; svar=sin(_t_);cvar=cos(_t_);} while (0) |
---|
[f734e7d] | 30 | # endif |
---|
| 31 | # define global |
---|
| 32 | # define local |
---|
| 33 | # define constant const |
---|
[750ffa5] | 34 | // OpenCL powr(a,b) = C99 pow(a,b), b >= 0 |
---|
| 35 | // OpenCL pown(a,b) = C99 pow(a,b), b integer |
---|
[f734e7d] | 36 | # define powr(a,b) pow(a,b) |
---|
| 37 | # define pown(a,b) pow(a,b) |
---|
| 38 | #else |
---|
| 39 | # ifdef USE_SINCOS |
---|
| 40 | # define SINCOS(angle,svar,cvar) svar=sincos(angle,&cvar) |
---|
| 41 | # else |
---|
[750ffa5] | 42 | # define SINCOS(angle,svar,cvar) do {const double _t_=angle; svar=sin(_t_);cvar=cos(_t_);} while (0) |
---|
[f734e7d] | 43 | # endif |
---|
| 44 | #endif |
---|
| 45 | |
---|
| 46 | // Standard mathematical constants: |
---|
| 47 | // M_E, M_LOG2E, M_LOG10E, M_LN2, M_LN10, M_PI, M_PI_2=pi/2, M_PI_4=pi/4, |
---|
| 48 | // M_1_PI=1/pi, M_2_PI=2/pi, M_2_SQRTPI=2/sqrt(pi), SQRT2, SQRT1_2=sqrt(1/2) |
---|
| 49 | // OpenCL defines M_constant_F for float constants, and nothing if double |
---|
| 50 | // is not enabled on the card, which is why these constants may be missing |
---|
| 51 | #ifndef M_PI |
---|
| 52 | # define M_PI 3.141592653589793 |
---|
| 53 | #endif |
---|
| 54 | #ifndef M_PI_2 |
---|
| 55 | # define M_PI_2 1.570796326794897 |
---|
| 56 | #endif |
---|
| 57 | #ifndef M_PI_4 |
---|
| 58 | # define M_PI_4 0.7853981633974483 |
---|
| 59 | #endif |
---|
| 60 | |
---|
| 61 | // Non-standard pi/180, used for converting between degrees and radians |
---|
| 62 | #ifndef M_PI_180 |
---|
| 63 | # define M_PI_180 0.017453292519943295 |
---|
| 64 | #endif |
---|
| 65 | |
---|
| 66 | |
---|
| 67 | %(DEFINES)s |
---|
| 68 | |
---|
| 69 | %(SOURCES)s |
---|
| 70 | |
---|
| 71 | /* |
---|
| 72 | ########################################################## |
---|
| 73 | # # |
---|
| 74 | # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! # |
---|
| 75 | # !! !! # |
---|
| 76 | # !! KEEP THIS CODE CONSISTENT WITH KERNELPY.PY !! # |
---|
| 77 | # !! !! # |
---|
| 78 | # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! # |
---|
| 79 | # # |
---|
| 80 | ########################################################## |
---|
| 81 | */ |
---|
| 82 | |
---|
| 83 | #ifdef IQ_KERNEL_NAME |
---|
| 84 | kernel void IQ_KERNEL_NAME( |
---|
| 85 | global const double *q, |
---|
| 86 | global double *result, |
---|
| 87 | const int Nq, |
---|
| 88 | #ifdef IQ_OPEN_LOOPS |
---|
| 89 | #ifdef USE_OPENCL |
---|
| 90 | global double *loops_g, |
---|
| 91 | #endif |
---|
| 92 | local double *loops, |
---|
| 93 | const double cutoff, |
---|
| 94 | IQ_DISPERSION_LENGTH_DECLARATIONS, |
---|
| 95 | #endif |
---|
| 96 | IQ_FIXED_PARAMETER_DECLARATIONS |
---|
| 97 | ) |
---|
| 98 | { |
---|
| 99 | #ifdef USE_OPENCL |
---|
| 100 | #ifdef IQ_OPEN_LOOPS |
---|
| 101 | // copy loops info to local memory |
---|
| 102 | event_t e = async_work_group_copy(loops, loops_g, (IQ_DISPERSION_LENGTH_SUM)*2, 0); |
---|
| 103 | wait_group_events(1, &e); |
---|
| 104 | #endif |
---|
| 105 | |
---|
| 106 | int i = get_global_id(0); |
---|
| 107 | if (i < Nq) |
---|
| 108 | #else |
---|
| 109 | #pragma omp parallel for |
---|
| 110 | for (int i=0; i < Nq; i++) |
---|
| 111 | #endif |
---|
| 112 | { |
---|
| 113 | const double qi = q[i]; |
---|
| 114 | #ifdef IQ_OPEN_LOOPS |
---|
| 115 | double ret=0.0, norm=0.0; |
---|
| 116 | #ifdef VOLUME_PARAMETERS |
---|
| 117 | double vol=0.0, norm_vol=0.0; |
---|
| 118 | #endif |
---|
| 119 | IQ_OPEN_LOOPS |
---|
| 120 | //for (int radius_i=0; radius_i < Nradius; radius_i++) { |
---|
| 121 | // const double radius = loops[2*(radius_i)]; |
---|
| 122 | // const double radius_w = loops[2*(radius_i)+1]; |
---|
| 123 | |
---|
| 124 | const double weight = IQ_WEIGHT_PRODUCT; |
---|
| 125 | if (weight > cutoff) { |
---|
[750ffa5] | 126 | const double scattering = Iq(qi, IQ_PARAMETERS); |
---|
[062c56d] | 127 | //if (scattering >= 0.0) { // scattering cannot be negative |
---|
[750ffa5] | 128 | ret += weight*scattering; |
---|
[f734e7d] | 129 | norm += weight; |
---|
| 130 | #ifdef VOLUME_PARAMETERS |
---|
| 131 | const double vol_weight = VOLUME_WEIGHT_PRODUCT; |
---|
| 132 | vol += vol_weight*form_volume(VOLUME_PARAMETERS); |
---|
| 133 | norm_vol += vol_weight; |
---|
| 134 | #endif |
---|
[062c56d] | 135 | //} |
---|
[750ffa5] | 136 | //else { printf("exclude qx,qy,I:%%g,%%g,%%g\n",qi,scattering); } |
---|
[f734e7d] | 137 | } |
---|
| 138 | IQ_CLOSE_LOOPS |
---|
| 139 | #ifdef VOLUME_PARAMETERS |
---|
| 140 | if (vol*norm_vol != 0.0) { |
---|
| 141 | ret *= norm_vol/vol; |
---|
| 142 | } |
---|
| 143 | #endif |
---|
| 144 | result[i] = scale*ret/norm+background; |
---|
| 145 | #else |
---|
| 146 | result[i] = scale*Iq(qi, IQ_PARAMETERS) + background; |
---|
| 147 | #endif |
---|
| 148 | } |
---|
| 149 | } |
---|
| 150 | #endif |
---|
| 151 | |
---|
| 152 | |
---|
| 153 | #ifdef IQXY_KERNEL_NAME |
---|
| 154 | kernel void IQXY_KERNEL_NAME( |
---|
| 155 | global const double *qx, |
---|
| 156 | global const double *qy, |
---|
| 157 | global double *result, |
---|
| 158 | const int Nq, |
---|
| 159 | #ifdef IQXY_OPEN_LOOPS |
---|
| 160 | #ifdef USE_OPENCL |
---|
| 161 | global double *loops_g, |
---|
| 162 | #endif |
---|
| 163 | local double *loops, |
---|
| 164 | const double cutoff, |
---|
| 165 | IQXY_DISPERSION_LENGTH_DECLARATIONS, |
---|
| 166 | #endif |
---|
| 167 | IQXY_FIXED_PARAMETER_DECLARATIONS |
---|
| 168 | ) |
---|
| 169 | { |
---|
| 170 | #ifdef USE_OPENCL |
---|
| 171 | #ifdef IQXY_OPEN_LOOPS |
---|
| 172 | // copy loops info to local memory |
---|
| 173 | event_t e = async_work_group_copy(loops, loops_g, (IQXY_DISPERSION_LENGTH_SUM)*2, 0); |
---|
| 174 | wait_group_events(1, &e); |
---|
| 175 | #endif |
---|
| 176 | |
---|
| 177 | int i = get_global_id(0); |
---|
| 178 | if (i < Nq) |
---|
| 179 | #else |
---|
| 180 | #pragma omp parallel for |
---|
| 181 | for (int i=0; i < Nq; i++) |
---|
| 182 | #endif |
---|
| 183 | { |
---|
| 184 | const double qxi = qx[i]; |
---|
| 185 | const double qyi = qy[i]; |
---|
| 186 | #ifdef IQXY_OPEN_LOOPS |
---|
| 187 | double ret=0.0, norm=0.0; |
---|
| 188 | #ifdef VOLUME_PARAMETERS |
---|
| 189 | double vol=0.0, norm_vol=0.0; |
---|
| 190 | #endif |
---|
| 191 | IQXY_OPEN_LOOPS |
---|
| 192 | //for (int radius_i=0; radius_i < Nradius; radius_i++) { |
---|
| 193 | // const double radius = loops[2*(radius_i)]; |
---|
| 194 | // const double radius_w = loops[2*(radius_i)+1]; |
---|
| 195 | |
---|
| 196 | const double weight = IQXY_WEIGHT_PRODUCT; |
---|
| 197 | if (weight > cutoff) { |
---|
| 198 | |
---|
[750ffa5] | 199 | const double scattering = Iqxy(qxi, qyi, IQXY_PARAMETERS); |
---|
[062c56d] | 200 | //if (scattering >= 0.0) { // scattering cannot be negative |
---|
[f734e7d] | 201 | // TODO: use correct angle for spherical correction |
---|
| 202 | // Definition of theta and phi are probably reversed relative to the |
---|
| 203 | // equation which gave rise to this correction, leading to an |
---|
| 204 | // attenuation of the pattern as theta moves through pi/2. Either |
---|
| 205 | // reverse the meanings of phi and theta in the forms, or use phi |
---|
| 206 | // rather than theta in this correction. Current code uses cos(theta) |
---|
| 207 | // so that values match those of sasview. |
---|
| 208 | #ifdef IQXY_HAS_THETA |
---|
| 209 | const double spherical_correction |
---|
| 210 | = (Ntheta>1 ? fabs(cos(M_PI_180*theta))*M_PI_2:1.0); |
---|
[750ffa5] | 211 | ret += spherical_correction * weight * scattering; |
---|
[f734e7d] | 212 | #else |
---|
[750ffa5] | 213 | ret += weight * scattering; |
---|
[f734e7d] | 214 | #endif |
---|
| 215 | norm += weight; |
---|
| 216 | #ifdef VOLUME_PARAMETERS |
---|
| 217 | const double vol_weight = VOLUME_WEIGHT_PRODUCT; |
---|
| 218 | vol += vol_weight*form_volume(VOLUME_PARAMETERS); |
---|
| 219 | #endif |
---|
| 220 | norm_vol += vol_weight; |
---|
[062c56d] | 221 | //} |
---|
[750ffa5] | 222 | //else { printf("exclude qx,qy,I:%%g,%%g,%%g\n",qi,scattering); } |
---|
[f734e7d] | 223 | } |
---|
| 224 | IQXY_CLOSE_LOOPS |
---|
| 225 | #ifdef VOLUME_PARAMETERS |
---|
| 226 | if (vol*norm_vol != 0.0) { |
---|
| 227 | ret *= norm_vol/vol; |
---|
| 228 | } |
---|
| 229 | #endif |
---|
| 230 | result[i] = scale*ret/norm+background; |
---|
| 231 | #else |
---|
| 232 | result[i] = scale*Iqxy(qxi, qyi, IQXY_PARAMETERS) + background; |
---|
| 233 | #endif |
---|
| 234 | } |
---|
| 235 | } |
---|
| 236 | #endif |
---|