ESPResSo
Extensible Simulation Package for Research on Soft Matter Systems
Loading...
Searching...
No Matches
DiffusiveFluxKernelThermalized_single_precision.cpp
Go to the documentation of this file.
1//======================================================================================================================
2//
3// This file is part of waLBerla. waLBerla is free software: you can
4// redistribute it and/or modify it under the terms of the GNU General Public
5// License as published by the Free Software Foundation, either version 3 of
6// the License, or (at your option) any later version.
7//
8// waLBerla is distributed in the hope that it will be useful, but WITHOUT
9// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
10// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
11// for more details.
12//
13// You should have received a copy of the GNU General Public License along
14// with waLBerla (see COPYING.txt). If not, see <http://www.gnu.org/licenses/>.
15//
16//! \\file DiffusiveFluxKernelThermalized_single_precision.cpp
17//! \\author pystencils
18//======================================================================================================================
19
20// kernel generated with pystencils v1.4+1.ge851f4e, lbmpy v1.4+1.ge9efe34, sympy v1.12.1, lbmpy_walberla/pystencils_walberla from waLBerla commit 007e77e077ad9d22b5eed6f3d3118240993e553c
21
22#include <cmath>
23
24#include "DiffusiveFluxKernelThermalized_single_precision.h"
25#include "core/DataTypes.h"
26#include "core/Macros.h"
27
28#include "philox_rand.h"
29
30#define FUNC_PREFIX
31
32#if (defined WALBERLA_CXX_COMPILER_IS_GNU) || (defined WALBERLA_CXX_COMPILER_IS_CLANG)
33#pragma GCC diagnostic push
34#pragma GCC diagnostic ignored "-Wfloat-equal"
35#pragma GCC diagnostic ignored "-Wshadow"
36#pragma GCC diagnostic ignored "-Wconversion"
37#pragma GCC diagnostic ignored "-Wunused-variable"
38#endif
39
40#if (defined WALBERLA_CXX_COMPILER_IS_INTEL)
41#pragma warning push
42#pragma warning(disable : 1599)
43#endif
44
45using namespace std;
46
47namespace walberla {
48namespace pystencils {
49
50namespace internal_13067439141d91943f586adb1c937d5b {
51static FUNC_PREFIX void diffusivefluxkernelthermalized_single_precision_diffusivefluxkernelthermalized_single_precision(float D, float *RESTRICT const _data_j, float *RESTRICT const _data_rho, int64_t const _size_j_0, int64_t const _size_j_1, int64_t const _size_j_2, int64_t const _stride_j_0, int64_t const _stride_j_1, int64_t const _stride_j_2, int64_t const _stride_j_3, int64_t const _stride_rho_0, int64_t const _stride_rho_1, int64_t const _stride_rho_2, uint32_t block_offset_0, uint32_t block_offset_1, uint32_t block_offset_2, uint32_t field_size_0, uint32_t field_size_1, uint32_t field_size_2, uint32_t seed, uint32_t time_step) {
52 for (int64_t ctr_2 = 0; ctr_2 < _size_j_2; ctr_2 += 1) {
53 for (int64_t ctr_1 = 0; ctr_1 < _size_j_1; ctr_1 += 1) {
54 for (int64_t ctr_0 = 1; ctr_0 < _size_j_0; ctr_0 += 1) {
55 if (ctr_1 > 0 && ctr_2 > 0 && ctr_1 < _size_j_1 - 1 && ctr_2 < _size_j_2 - 1) {
56
57 float random_3_0;
58 float random_3_1;
59 float random_3_2;
60 float random_3_3;
61 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 3, seed, random_3_0, random_3_1, random_3_2, random_3_3);
62
63 float random_2_0;
64 float random_2_1;
65 float random_2_2;
66 float random_2_3;
67 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 2, seed, random_2_0, random_2_1, random_2_2, random_2_3);
68
69 float random_1_0;
70 float random_1_1;
71 float random_1_2;
72 float random_1_3;
73 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 1, seed, random_1_0, random_1_1, random_1_2, random_1_3);
74
75 float random_0_0;
76 float random_0_1;
77 float random_0_2;
78 float random_0_3;
79 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 0, seed, random_0_0, random_0_1, random_0_2, random_0_3);
80
81 _data_j[_stride_j_0 * ctr_0 + _stride_j_1 * ctr_1 + _stride_j_2 * ctr_2] = D * (_data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] - _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2]) * 0.16292407789368385f + (random_0_0 - 0.5f) * powf(D * (0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] + 0.5f * _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2]), 0.5f) * 1.977416969040271f;
82 }
83 if (ctr_1 > 0 && ctr_2 > 0 && ctr_0 < _size_j_0 - 1 && ctr_2 < _size_j_2 - 1) {
84
85 float random_3_0;
86 float random_3_1;
87 float random_3_2;
88 float random_3_3;
89 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 3, seed, random_3_0, random_3_1, random_3_2, random_3_3);
90
91 float random_2_0;
92 float random_2_1;
93 float random_2_2;
94 float random_2_3;
95 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 2, seed, random_2_0, random_2_1, random_2_2, random_2_3);
96
97 float random_1_0;
98 float random_1_1;
99 float random_1_2;
100 float random_1_3;
101 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 1, seed, random_1_0, random_1_1, random_1_2, random_1_3);
102
103 float random_0_0;
104 float random_0_1;
105 float random_0_2;
106 float random_0_3;
107 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 0, seed, random_0_0, random_0_1, random_0_2, random_0_3);
108
109 _data_j[_stride_j_0 * ctr_0 + _stride_j_1 * ctr_1 + _stride_j_2 * ctr_2 + _stride_j_3] = D * (_data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] - _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 - _stride_rho_1 + _stride_rho_2 * ctr_2]) * 0.16292407789368385f + (random_0_1 - 0.5f) * powf(D * (0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] + 0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 - _stride_rho_1 + _stride_rho_2 * ctr_2]), 0.5f) * 1.977416969040271f;
110 }
111 if (ctr_1 > 0 && ctr_2 > 0 && ctr_0 < _size_j_0 - 1 && ctr_1 < _size_j_1 - 1) {
112
113 float random_3_0;
114 float random_3_1;
115 float random_3_2;
116 float random_3_3;
117 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 3, seed, random_3_0, random_3_1, random_3_2, random_3_3);
118
119 float random_2_0;
120 float random_2_1;
121 float random_2_2;
122 float random_2_3;
123 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 2, seed, random_2_0, random_2_1, random_2_2, random_2_3);
124
125 float random_1_0;
126 float random_1_1;
127 float random_1_2;
128 float random_1_3;
129 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 1, seed, random_1_0, random_1_1, random_1_2, random_1_3);
130
131 float random_0_0;
132 float random_0_1;
133 float random_0_2;
134 float random_0_3;
135 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 0, seed, random_0_0, random_0_1, random_0_2, random_0_3);
136
137 _data_j[_stride_j_0 * ctr_0 + _stride_j_1 * ctr_1 + _stride_j_2 * ctr_2 + 2 * _stride_j_3] = D * (-_data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2 - _stride_rho_2] + _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2]) * 0.16292407789368385f + (random_0_2 - 0.5f) * powf(D * (0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2 - _stride_rho_2] + 0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2]), 0.5f) * 1.977416969040271f;
138 }
139 if (ctr_1 > 0 && ctr_2 > 0 && ctr_2 < _size_j_2 - 1) {
140
141 float random_3_0;
142 float random_3_1;
143 float random_3_2;
144 float random_3_3;
145 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 3, seed, random_3_0, random_3_1, random_3_2, random_3_3);
146
147 float random_2_0;
148 float random_2_1;
149 float random_2_2;
150 float random_2_3;
151 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 2, seed, random_2_0, random_2_1, random_2_2, random_2_3);
152
153 float random_1_0;
154 float random_1_1;
155 float random_1_2;
156 float random_1_3;
157 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 1, seed, random_1_0, random_1_1, random_1_2, random_1_3);
158
159 float random_0_0;
160 float random_0_1;
161 float random_0_2;
162 float random_0_3;
163 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 0, seed, random_0_0, random_0_1, random_0_2, random_0_3);
164
165 _data_j[_stride_j_0 * ctr_0 + _stride_j_1 * ctr_1 + _stride_j_2 * ctr_2 + 3 * _stride_j_3] = D * (_data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] - _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 - _stride_rho_1 + _stride_rho_2 * ctr_2]) * 0.11520472029718914f + (random_0_3 - 0.5f) * powf(D * (0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] + 0.5f * _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 - _stride_rho_1 + _stride_rho_2 * ctr_2]), 0.5f) * 1.6628028407278295f;
166 }
167 if (ctr_2 > 0 && ctr_1 < _size_j_1 - 1 && ctr_2 < _size_j_2 - 1) {
168
169 float random_3_0;
170 float random_3_1;
171 float random_3_2;
172 float random_3_3;
173 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 3, seed, random_3_0, random_3_1, random_3_2, random_3_3);
174
175 float random_2_0;
176 float random_2_1;
177 float random_2_2;
178 float random_2_3;
179 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 2, seed, random_2_0, random_2_1, random_2_2, random_2_3);
180
181 float random_1_0;
182 float random_1_1;
183 float random_1_2;
184 float random_1_3;
185 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 1, seed, random_1_0, random_1_1, random_1_2, random_1_3);
186
187 float random_0_0;
188 float random_0_1;
189 float random_0_2;
190 float random_0_3;
191 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 0, seed, random_0_0, random_0_1, random_0_2, random_0_3);
192
193 _data_j[_stride_j_0 * ctr_0 + _stride_j_1 * ctr_1 + _stride_j_2 * ctr_2 + 4 * _stride_j_3] = D * (_data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] - _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 + _stride_rho_1 + _stride_rho_2 * ctr_2]) * 0.11520472029718914f + (random_1_0 - 0.5f) * powf(D * (0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] + 0.5f * _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 + _stride_rho_1 + _stride_rho_2 * ctr_2]), 0.5f) * 1.6628028407278295f;
194 }
195 if (ctr_1 > 0 && ctr_2 > 0 && ctr_1 < _size_j_1 - 1) {
196
197 float random_3_0;
198 float random_3_1;
199 float random_3_2;
200 float random_3_3;
201 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 3, seed, random_3_0, random_3_1, random_3_2, random_3_3);
202
203 float random_2_0;
204 float random_2_1;
205 float random_2_2;
206 float random_2_3;
207 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 2, seed, random_2_0, random_2_1, random_2_2, random_2_3);
208
209 float random_1_0;
210 float random_1_1;
211 float random_1_2;
212 float random_1_3;
213 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 1, seed, random_1_0, random_1_1, random_1_2, random_1_3);
214
215 float random_0_0;
216 float random_0_1;
217 float random_0_2;
218 float random_0_3;
219 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 0, seed, random_0_0, random_0_1, random_0_2, random_0_3);
220
221 _data_j[_stride_j_0 * ctr_0 + _stride_j_1 * ctr_1 + _stride_j_2 * ctr_2 + 5 * _stride_j_3] = D * (_data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] - _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2 - _stride_rho_2]) * 0.11520472029718914f + (random_1_1 - 0.5f) * powf(D * (0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] + 0.5f * _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2 - _stride_rho_2]), 0.5f) * 1.6628028407278295f;
222 }
223 if (ctr_1 > 0 && ctr_1 < _size_j_1 - 1 && ctr_2 < _size_j_2 - 1) {
224
225 float random_3_0;
226 float random_3_1;
227 float random_3_2;
228 float random_3_3;
229 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 3, seed, random_3_0, random_3_1, random_3_2, random_3_3);
230
231 float random_2_0;
232 float random_2_1;
233 float random_2_2;
234 float random_2_3;
235 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 2, seed, random_2_0, random_2_1, random_2_2, random_2_3);
236
237 float random_1_0;
238 float random_1_1;
239 float random_1_2;
240 float random_1_3;
241 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 1, seed, random_1_0, random_1_1, random_1_2, random_1_3);
242
243 float random_0_0;
244 float random_0_1;
245 float random_0_2;
246 float random_0_3;
247 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 0, seed, random_0_0, random_0_1, random_0_2, random_0_3);
248
249 _data_j[_stride_j_0 * ctr_0 + _stride_j_1 * ctr_1 + _stride_j_2 * ctr_2 + 6 * _stride_j_3] = D * (_data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] - _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2 + _stride_rho_2]) * 0.11520472029718914f + (random_1_2 - 0.5f) * powf(D * (0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] + 0.5f * _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2 + _stride_rho_2]), 0.5f) * 1.6628028407278295f;
250 }
251 if (ctr_1 > 0 && ctr_2 > 0 && ctr_0 < _size_j_0 - 1) {
252
253 float random_3_0;
254 float random_3_1;
255 float random_3_2;
256 float random_3_3;
257 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 3, seed, random_3_0, random_3_1, random_3_2, random_3_3);
258
259 float random_2_0;
260 float random_2_1;
261 float random_2_2;
262 float random_2_3;
263 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 2, seed, random_2_0, random_2_1, random_2_2, random_2_3);
264
265 float random_1_0;
266 float random_1_1;
267 float random_1_2;
268 float random_1_3;
269 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 1, seed, random_1_0, random_1_1, random_1_2, random_1_3);
270
271 float random_0_0;
272 float random_0_1;
273 float random_0_2;
274 float random_0_3;
275 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 0, seed, random_0_0, random_0_1, random_0_2, random_0_3);
276
277 _data_j[_stride_j_0 * ctr_0 + _stride_j_1 * ctr_1 + _stride_j_2 * ctr_2 + 7 * _stride_j_3] = D * (_data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] - _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 - _stride_rho_1 + _stride_rho_2 * ctr_2 - _stride_rho_2]) * 0.11520472029718914f + (random_1_3 - 0.5f) * powf(D * (0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] + 0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 - _stride_rho_1 + _stride_rho_2 * ctr_2 - _stride_rho_2]), 0.5f) * 1.6628028407278295f;
278 }
279 if (ctr_1 > 0 && ctr_0 < _size_j_0 - 1 && ctr_2 < _size_j_2 - 1) {
280
281 float random_3_0;
282 float random_3_1;
283 float random_3_2;
284 float random_3_3;
285 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 3, seed, random_3_0, random_3_1, random_3_2, random_3_3);
286
287 float random_2_0;
288 float random_2_1;
289 float random_2_2;
290 float random_2_3;
291 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 2, seed, random_2_0, random_2_1, random_2_2, random_2_3);
292
293 float random_1_0;
294 float random_1_1;
295 float random_1_2;
296 float random_1_3;
297 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 1, seed, random_1_0, random_1_1, random_1_2, random_1_3);
298
299 float random_0_0;
300 float random_0_1;
301 float random_0_2;
302 float random_0_3;
303 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 0, seed, random_0_0, random_0_1, random_0_2, random_0_3);
304
305 _data_j[_stride_j_0 * ctr_0 + _stride_j_1 * ctr_1 + _stride_j_2 * ctr_2 + 8 * _stride_j_3] = D * (_data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] - _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 - _stride_rho_1 + _stride_rho_2 * ctr_2 + _stride_rho_2]) * 0.11520472029718914f + (random_2_0 - 0.5f) * powf(D * (0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] + 0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 - _stride_rho_1 + _stride_rho_2 * ctr_2 + _stride_rho_2]), 0.5f) * 1.6628028407278295f;
306 }
307 if (ctr_1 > 0 && ctr_2 > 0) {
308
309 float random_3_0;
310 float random_3_1;
311 float random_3_2;
312 float random_3_3;
313 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 3, seed, random_3_0, random_3_1, random_3_2, random_3_3);
314
315 float random_2_0;
316 float random_2_1;
317 float random_2_2;
318 float random_2_3;
319 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 2, seed, random_2_0, random_2_1, random_2_2, random_2_3);
320
321 float random_1_0;
322 float random_1_1;
323 float random_1_2;
324 float random_1_3;
325 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 1, seed, random_1_0, random_1_1, random_1_2, random_1_3);
326
327 float random_0_0;
328 float random_0_1;
329 float random_0_2;
330 float random_0_3;
331 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 0, seed, random_0_0, random_0_1, random_0_2, random_0_3);
332
333 _data_j[_stride_j_0 * ctr_0 + _stride_j_1 * ctr_1 + _stride_j_2 * ctr_2 + 9 * _stride_j_3] = D * (_data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] - _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 - _stride_rho_1 + _stride_rho_2 * ctr_2 - _stride_rho_2]) * 0.09406426022938992f + (random_2_1 - 0.5f) * powf(D * (0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] + 0.5f * _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 - _stride_rho_1 + _stride_rho_2 * ctr_2 - _stride_rho_2]), 0.5f) * 1.5025119784898082f;
334 }
335 if (ctr_1 > 0 && ctr_2 < _size_j_2 - 1) {
336
337 float random_3_0;
338 float random_3_1;
339 float random_3_2;
340 float random_3_3;
341 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 3, seed, random_3_0, random_3_1, random_3_2, random_3_3);
342
343 float random_2_0;
344 float random_2_1;
345 float random_2_2;
346 float random_2_3;
347 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 2, seed, random_2_0, random_2_1, random_2_2, random_2_3);
348
349 float random_1_0;
350 float random_1_1;
351 float random_1_2;
352 float random_1_3;
353 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 1, seed, random_1_0, random_1_1, random_1_2, random_1_3);
354
355 float random_0_0;
356 float random_0_1;
357 float random_0_2;
358 float random_0_3;
359 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 0, seed, random_0_0, random_0_1, random_0_2, random_0_3);
360
361 _data_j[_stride_j_0 * ctr_0 + _stride_j_1 * ctr_1 + _stride_j_2 * ctr_2 + 10 * _stride_j_3] = D * (_data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] - _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 - _stride_rho_1 + _stride_rho_2 * ctr_2 + _stride_rho_2]) * 0.09406426022938992f + (random_2_2 - 0.5f) * powf(D * (0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] + 0.5f * _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 - _stride_rho_1 + _stride_rho_2 * ctr_2 + _stride_rho_2]), 0.5f) * 1.5025119784898082f;
362 }
363 if (ctr_2 > 0 && ctr_1 < _size_j_1 - 1) {
364
365 float random_3_0;
366 float random_3_1;
367 float random_3_2;
368 float random_3_3;
369 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 3, seed, random_3_0, random_3_1, random_3_2, random_3_3);
370
371 float random_2_0;
372 float random_2_1;
373 float random_2_2;
374 float random_2_3;
375 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 2, seed, random_2_0, random_2_1, random_2_2, random_2_3);
376
377 float random_1_0;
378 float random_1_1;
379 float random_1_2;
380 float random_1_3;
381 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 1, seed, random_1_0, random_1_1, random_1_2, random_1_3);
382
383 float random_0_0;
384 float random_0_1;
385 float random_0_2;
386 float random_0_3;
387 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 0, seed, random_0_0, random_0_1, random_0_2, random_0_3);
388
389 _data_j[_stride_j_0 * ctr_0 + _stride_j_1 * ctr_1 + _stride_j_2 * ctr_2 + 11 * _stride_j_3] = D * (_data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] - _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 + _stride_rho_1 + _stride_rho_2 * ctr_2 - _stride_rho_2]) * 0.09406426022938992f + (random_2_3 - 0.5f) * powf(D * (0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] + 0.5f * _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 + _stride_rho_1 + _stride_rho_2 * ctr_2 - _stride_rho_2]), 0.5f) * 1.5025119784898082f;
390 }
391 if (ctr_1 < _size_j_1 - 1 && ctr_2 < _size_j_2 - 1) {
392
393 float random_3_0;
394 float random_3_1;
395 float random_3_2;
396 float random_3_3;
397 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 3, seed, random_3_0, random_3_1, random_3_2, random_3_3);
398
399 float random_2_0;
400 float random_2_1;
401 float random_2_2;
402 float random_2_3;
403 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 2, seed, random_2_0, random_2_1, random_2_2, random_2_3);
404
405 float random_1_0;
406 float random_1_1;
407 float random_1_2;
408 float random_1_3;
409 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 1, seed, random_1_0, random_1_1, random_1_2, random_1_3);
410
411 float random_0_0;
412 float random_0_1;
413 float random_0_2;
414 float random_0_3;
415 philox_float4(time_step, (block_offset_0 + ctr_0) % field_size_0, (block_offset_1 + ctr_1) % field_size_1, (block_offset_2 + ctr_2) % field_size_2, 0, seed, random_0_0, random_0_1, random_0_2, random_0_3);
416
417 _data_j[_stride_j_0 * ctr_0 + _stride_j_1 * ctr_1 + _stride_j_2 * ctr_2 + 12 * _stride_j_3] = D * (_data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] - _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 + _stride_rho_1 + _stride_rho_2 * ctr_2 + _stride_rho_2]) * 0.09406426022938992f + (random_3_0 - 0.5f) * powf(D * (0.5f * _data_rho[_stride_rho_0 * ctr_0 + _stride_rho_1 * ctr_1 + _stride_rho_2 * ctr_2] + 0.5f * _data_rho[_stride_rho_0 * ctr_0 - _stride_rho_0 + _stride_rho_1 * ctr_1 + _stride_rho_1 + _stride_rho_2 * ctr_2 + _stride_rho_2]), 0.5f) * 1.5025119784898082f;
418 }
419 }
420 }
421 }
422}
423} // namespace internal_13067439141d91943f586adb1c937d5b
424
425void DiffusiveFluxKernelThermalized_single_precision::run(IBlock *block) {
426 if (!this->configured_)
427 WALBERLA_ABORT("This Sweep contains a configure function that needs to be called manually")
428
429 auto j = block->getData<field::GhostLayerField<float, 13>>(jID);
430 auto rho = block->getData<field::GhostLayerField<float, 1>>(rhoID);
431
432 auto &field_size_1 = this->field_size_1_;
433 auto &seed = this->seed_;
434 auto &field_size_0 = this->field_size_0_;
435 auto &field_size_2 = this->field_size_2_;
436 auto &block_offset_2 = this->block_offset_2_;
437 auto &block_offset_1 = this->block_offset_1_;
438 auto &D = this->D_;
439 auto &time_step = this->time_step_;
440 auto &block_offset_0 = this->block_offset_0_;
441 WALBERLA_ASSERT_GREATER_EQUAL(-1, -int_c(j->nrOfGhostLayers()))
442 float *RESTRICT const _data_j = j->dataAt(-1, -1, -1, 0);
443 WALBERLA_ASSERT_GREATER_EQUAL(-1, -int_c(rho->nrOfGhostLayers()))
444 float *RESTRICT const _data_rho = rho->dataAt(-1, -1, -1, 0);
445 WALBERLA_ASSERT_GREATER_EQUAL(j->xSizeWithGhostLayer(), int64_t(int64_c(j->xSize()) + 2))
446 const int64_t _size_j_0 = int64_t(int64_c(j->xSize()) + 2);
447 WALBERLA_ASSERT_GREATER_EQUAL(j->ySizeWithGhostLayer(), int64_t(int64_c(j->ySize()) + 2))
448 const int64_t _size_j_1 = int64_t(int64_c(j->ySize()) + 2);
449 WALBERLA_ASSERT_GREATER_EQUAL(j->zSizeWithGhostLayer(), int64_t(int64_c(j->zSize()) + 2))
450 const int64_t _size_j_2 = int64_t(int64_c(j->zSize()) + 2);
451 const int64_t _stride_j_0 = int64_t(j->xStride());
452 const int64_t _stride_j_1 = int64_t(j->yStride());
453 const int64_t _stride_j_2 = int64_t(j->zStride());
454 const int64_t _stride_j_3 = int64_t(1 * int64_t(j->fStride()));
455 const int64_t _stride_rho_0 = int64_t(rho->xStride());
456 const int64_t _stride_rho_1 = int64_t(rho->yStride());
457 const int64_t _stride_rho_2 = int64_t(rho->zStride());
458 internal_13067439141d91943f586adb1c937d5b::diffusivefluxkernelthermalized_single_precision_diffusivefluxkernelthermalized_single_precision(D, _data_j, _data_rho, _size_j_0, _size_j_1, _size_j_2, _stride_j_0, _stride_j_1, _stride_j_2, _stride_j_3, _stride_rho_0, _stride_rho_1, _stride_rho_2, block_offset_0, block_offset_1, block_offset_2, field_size_0, field_size_1, field_size_2, seed, time_step);
459}
460
461void DiffusiveFluxKernelThermalized_single_precision::runOnCellInterval(const shared_ptr<StructuredBlockStorage> &blocks, const CellInterval &globalCellInterval, cell_idx_t ghostLayers, IBlock *block) {
462 if (!this->configured_)
463 WALBERLA_ABORT("This Sweep contains a configure function that needs to be called manually")
464
465 CellInterval ci = globalCellInterval;
466 CellInterval blockBB = blocks->getBlockCellBB(*block);
467 blockBB.expand(ghostLayers);
468 ci.intersect(blockBB);
469 blocks->transformGlobalToBlockLocalCellInterval(ci, *block);
470 if (ci.empty())
471 return;
472
473 auto j = block->getData<field::GhostLayerField<float, 13>>(jID);
474 auto rho = block->getData<field::GhostLayerField<float, 1>>(rhoID);
475
476 auto &field_size_1 = this->field_size_1_;
477 auto &seed = this->seed_;
478 auto &field_size_0 = this->field_size_0_;
479 auto &field_size_2 = this->field_size_2_;
480 auto &block_offset_2 = this->block_offset_2_;
481 auto &block_offset_1 = this->block_offset_1_;
482 auto &D = this->D_;
483 auto &time_step = this->time_step_;
484 auto &block_offset_0 = this->block_offset_0_;
485 WALBERLA_ASSERT_GREATER_EQUAL(ci.xMin() - 1, -int_c(j->nrOfGhostLayers()))
486 WALBERLA_ASSERT_GREATER_EQUAL(ci.yMin() - 1, -int_c(j->nrOfGhostLayers()))
487 WALBERLA_ASSERT_GREATER_EQUAL(ci.zMin() - 1, -int_c(j->nrOfGhostLayers()))
488 float *RESTRICT const _data_j = j->dataAt(ci.xMin() - 1, ci.yMin() - 1, ci.zMin() - 1, 0);
489 WALBERLA_ASSERT_GREATER_EQUAL(ci.xMin() - 1, -int_c(rho->nrOfGhostLayers()))
490 WALBERLA_ASSERT_GREATER_EQUAL(ci.yMin() - 1, -int_c(rho->nrOfGhostLayers()))
491 WALBERLA_ASSERT_GREATER_EQUAL(ci.zMin() - 1, -int_c(rho->nrOfGhostLayers()))
492 float *RESTRICT const _data_rho = rho->dataAt(ci.xMin() - 1, ci.yMin() - 1, ci.zMin() - 1, 0);
493 WALBERLA_ASSERT_GREATER_EQUAL(j->xSizeWithGhostLayer(), int64_t(int64_c(ci.xSize()) + 2))
494 const int64_t _size_j_0 = int64_t(int64_c(ci.xSize()) + 2);
495 WALBERLA_ASSERT_GREATER_EQUAL(j->ySizeWithGhostLayer(), int64_t(int64_c(ci.ySize()) + 2))
496 const int64_t _size_j_1 = int64_t(int64_c(ci.ySize()) + 2);
497 WALBERLA_ASSERT_GREATER_EQUAL(j->zSizeWithGhostLayer(), int64_t(int64_c(ci.zSize()) + 2))
498 const int64_t _size_j_2 = int64_t(int64_c(ci.zSize()) + 2);
499 const int64_t _stride_j_0 = int64_t(j->xStride());
500 const int64_t _stride_j_1 = int64_t(j->yStride());
501 const int64_t _stride_j_2 = int64_t(j->zStride());
502 const int64_t _stride_j_3 = int64_t(1 * int64_t(j->fStride()));
503 const int64_t _stride_rho_0 = int64_t(rho->xStride());
504 const int64_t _stride_rho_1 = int64_t(rho->yStride());
505 const int64_t _stride_rho_2 = int64_t(rho->zStride());
506 internal_13067439141d91943f586adb1c937d5b::diffusivefluxkernelthermalized_single_precision_diffusivefluxkernelthermalized_single_precision(D, _data_j, _data_rho, _size_j_0, _size_j_1, _size_j_2, _stride_j_0, _stride_j_1, _stride_j_2, _stride_j_3, _stride_rho_0, _stride_rho_1, _stride_rho_2, block_offset_0, block_offset_1, block_offset_2, field_size_0, field_size_1, field_size_2, seed, time_step);
507}
508
509} // namespace pystencils
510} // namespace walberla
511
512#if (defined WALBERLA_CXX_COMPILER_IS_GNU) || (defined WALBERLA_CXX_COMPILER_IS_CLANG)
513#pragma GCC diagnostic pop
514#endif
515
516#if (defined WALBERLA_CXX_COMPILER_IS_INTEL)
517#pragma warning pop
518#endif
FUNC_PREFIX
#define FUNC_PREFIX
\file AdvectiveFluxKernel_double_precision.cpp \author pystencils
Definition AdvectiveFluxKernel_double_precision.cpp:28
RESTRICT
#define RESTRICT
\file AdvectiveFluxKernel_double_precision.h \author pystencils
Definition AdvectiveFluxKernel_double_precision.h:40
DiffusiveFluxKernelThermalized_single_precision.h
walberla::pystencils::DiffusiveFluxKernelThermalized_single_precision::run
void run(IBlock *block)
Definition DiffusiveFluxKernelThermalized_single_precision.cpp:425
walberla::pystencils::DiffusiveFluxKernelThermalized_single_precision::runOnCellInterval
void runOnCellInterval(const shared_ptr< StructuredBlockStorage > &blocks, const CellInterval &globalCellInterval, cell_idx_t ghostLayers, IBlock *block)
Definition DiffusiveFluxKernelThermalized_single_precision.cpp:461
stream
cudaStream_t stream[1]
CUDA streams for parallel computing on CPU and GPU.
Definition common_cuda.cu:34
block
static double * block(double *p, std::size_t index, std::size_t size)
Definition elc.cpp:177
philox_float4
QUALIFIERS void philox_float4(uint32 ctr0, uint32 ctr1, uint32 ctr2, uint32 ctr3, uint32 key0, uint32 key1, float &rnd1, float &rnd2, float &rnd3, float &rnd4)
Definition electrokinetics/generated_kernels/philox_rand.h:218
philox_rand.h
Philox counter-based RNG from .
std
STL namespace.
walberla::pystencils::internal_13067439141d91943f586adb1c937d5b::diffusivefluxkernelthermalized_single_precision_diffusivefluxkernelthermalized_single_precision
static FUNC_PREFIX void diffusivefluxkernelthermalized_single_precision_diffusivefluxkernelthermalized_single_precision(float D, float *RESTRICT const _data_j, float *RESTRICT const _data_rho, int64_t const _size_j_0, int64_t const _size_j_1, int64_t const _size_j_2, int64_t const _stride_j_0, int64_t const _stride_j_1, int64_t const _stride_j_2, int64_t const _stride_j_3, int64_t const _stride_rho_0, int64_t const _stride_rho_1, int64_t const _stride_rho_2, uint32_t block_offset_0, uint32_t block_offset_1, uint32_t block_offset_2, uint32_t field_size_0, uint32_t field_size_1, uint32_t field_size_2, uint32_t seed, uint32_t time_step)
Definition DiffusiveFluxKernelThermalized_single_precision.cpp:51
walberla
\file PackInfoPdfDoublePrecision.cpp \author pystencils
Definition EKWalberla.hpp:38