Loading [MathJax]/extensions/TeX/AMSmath.js
ESPResSo
Extensible Simulation Package for Research on Soft Matter Systems
•All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Pages Concepts
StreamSweepSinglePrecision.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 StreamSweepSinglePrecision.cpp
17//! \\author pystencils
18//======================================================================================================================
19
20// kernel generated with pystencils v1.3.7, lbmpy v1.3.7, sympy v1.12.1, lbmpy_walberla/pystencils_walberla from waLBerla commit f36fa0a68bae59f0b516f6587ea8fa7c24a41141
21
22#include <cmath>
23
24#include "StreamSweepSinglePrecision.h"
25#include "core/DataTypes.h"
26#include "core/Macros.h"
27
28#define FUNC_PREFIX
29
30#if (defined WALBERLA_CXX_COMPILER_IS_GNU) || (defined WALBERLA_CXX_COMPILER_IS_CLANG)
31#pragma GCC diagnostic push
32#pragma GCC diagnostic ignored "-Wfloat-equal"
33#pragma GCC diagnostic ignored "-Wshadow"
34#pragma GCC diagnostic ignored "-Wconversion"
35#pragma GCC diagnostic ignored "-Wunused-variable"
36#endif
37
38#if (defined WALBERLA_CXX_COMPILER_IS_INTEL)
39#pragma warning push
40#pragma warning(disable : 1599)
41#endif
42
43using namespace std;
44
45namespace walberla {
46namespace pystencils {
47
48namespace internal_streamsweepsingleprecision_streamsweepsingleprecision {
49static FUNC_PREFIX void streamsweepsingleprecision_streamsweepsingleprecision(float *RESTRICT const _data_force, float *RESTRICT const _data_pdfs, float *RESTRICT _data_pdfs_tmp, float *RESTRICT _data_velocity, int64_t const _size_force_0, int64_t const _size_force_1, int64_t const _size_force_2, int64_t const _stride_force_0, int64_t const _stride_force_1, int64_t const _stride_force_2, int64_t const _stride_force_3, int64_t const _stride_pdfs_0, int64_t const _stride_pdfs_1, int64_t const _stride_pdfs_2, int64_t const _stride_pdfs_3, int64_t const _stride_pdfs_tmp_0, int64_t const _stride_pdfs_tmp_1, int64_t const _stride_pdfs_tmp_2, int64_t const _stride_pdfs_tmp_3, int64_t const _stride_velocity_0, int64_t const _stride_velocity_1, int64_t const _stride_velocity_2, int64_t const _stride_velocity_3) {
50 for (int64_t ctr_2 = 1; ctr_2 < _size_force_2 - 1; ctr_2 += 1) {
51 for (int64_t ctr_1 = 1; ctr_1 < _size_force_1 - 1; ctr_1 += 1) {
52 for (int64_t ctr_0 = 1; ctr_0 < _size_force_0 - 1; ctr_0 += 1) {
53 const float streamed_0 = _data_pdfs[_stride_pdfs_0 * ctr_0 + _stride_pdfs_1 * ctr_1 + _stride_pdfs_2 * ctr_2];
54 const float streamed_1 = _data_pdfs[_stride_pdfs_0 * ctr_0 + _stride_pdfs_1 * ctr_1 - _stride_pdfs_1 + _stride_pdfs_2 * ctr_2 + _stride_pdfs_3];
55 const float streamed_2 = _data_pdfs[_stride_pdfs_0 * ctr_0 + _stride_pdfs_1 * ctr_1 + _stride_pdfs_1 + _stride_pdfs_2 * ctr_2 + 2 * _stride_pdfs_3];
56 const float streamed_3 = _data_pdfs[_stride_pdfs_0 * ctr_0 + _stride_pdfs_0 + _stride_pdfs_1 * ctr_1 + _stride_pdfs_2 * ctr_2 + 3 * _stride_pdfs_3];
57 const float streamed_4 = _data_pdfs[_stride_pdfs_0 * ctr_0 - _stride_pdfs_0 + _stride_pdfs_1 * ctr_1 + _stride_pdfs_2 * ctr_2 + 4 * _stride_pdfs_3];
58 const float streamed_5 = _data_pdfs[_stride_pdfs_0 * ctr_0 + _stride_pdfs_1 * ctr_1 + _stride_pdfs_2 * ctr_2 - _stride_pdfs_2 + 5 * _stride_pdfs_3];
59 const float streamed_6 = _data_pdfs[_stride_pdfs_0 * ctr_0 + _stride_pdfs_1 * ctr_1 + _stride_pdfs_2 * ctr_2 + _stride_pdfs_2 + 6 * _stride_pdfs_3];
60 const float streamed_7 = _data_pdfs[_stride_pdfs_0 * ctr_0 + _stride_pdfs_0 + _stride_pdfs_1 * ctr_1 - _stride_pdfs_1 + _stride_pdfs_2 * ctr_2 + 7 * _stride_pdfs_3];
61 const float streamed_8 = _data_pdfs[_stride_pdfs_0 * ctr_0 - _stride_pdfs_0 + _stride_pdfs_1 * ctr_1 - _stride_pdfs_1 + _stride_pdfs_2 * ctr_2 + 8 * _stride_pdfs_3];
62 const float streamed_9 = _data_pdfs[_stride_pdfs_0 * ctr_0 + _stride_pdfs_0 + _stride_pdfs_1 * ctr_1 + _stride_pdfs_1 + _stride_pdfs_2 * ctr_2 + 9 * _stride_pdfs_3];
63 const float streamed_10 = _data_pdfs[_stride_pdfs_0 * ctr_0 - _stride_pdfs_0 + _stride_pdfs_1 * ctr_1 + _stride_pdfs_1 + _stride_pdfs_2 * ctr_2 + 10 * _stride_pdfs_3];
64 const float streamed_11 = _data_pdfs[_stride_pdfs_0 * ctr_0 + _stride_pdfs_1 * ctr_1 - _stride_pdfs_1 + _stride_pdfs_2 * ctr_2 - _stride_pdfs_2 + 11 * _stride_pdfs_3];
65 const float streamed_12 = _data_pdfs[_stride_pdfs_0 * ctr_0 + _stride_pdfs_1 * ctr_1 + _stride_pdfs_1 + _stride_pdfs_2 * ctr_2 - _stride_pdfs_2 + 12 * _stride_pdfs_3];
66 const float streamed_13 = _data_pdfs[_stride_pdfs_0 * ctr_0 + _stride_pdfs_0 + _stride_pdfs_1 * ctr_1 + _stride_pdfs_2 * ctr_2 - _stride_pdfs_2 + 13 * _stride_pdfs_3];
67 const float streamed_14 = _data_pdfs[_stride_pdfs_0 * ctr_0 - _stride_pdfs_0 + _stride_pdfs_1 * ctr_1 + _stride_pdfs_2 * ctr_2 - _stride_pdfs_2 + 14 * _stride_pdfs_3];
68 const float streamed_15 = _data_pdfs[_stride_pdfs_0 * ctr_0 + _stride_pdfs_1 * ctr_1 - _stride_pdfs_1 + _stride_pdfs_2 * ctr_2 + _stride_pdfs_2 + 15 * _stride_pdfs_3];
69 const float streamed_16 = _data_pdfs[_stride_pdfs_0 * ctr_0 + _stride_pdfs_1 * ctr_1 + _stride_pdfs_1 + _stride_pdfs_2 * ctr_2 + _stride_pdfs_2 + 16 * _stride_pdfs_3];
70 const float streamed_17 = _data_pdfs[_stride_pdfs_0 * ctr_0 + _stride_pdfs_0 + _stride_pdfs_1 * ctr_1 + _stride_pdfs_2 * ctr_2 + _stride_pdfs_2 + 17 * _stride_pdfs_3];
71 const float streamed_18 = _data_pdfs[_stride_pdfs_0 * ctr_0 - _stride_pdfs_0 + _stride_pdfs_1 * ctr_1 + _stride_pdfs_2 * ctr_2 + _stride_pdfs_2 + 18 * _stride_pdfs_3];
72 const float vel0Term = streamed_10 + streamed_14 + streamed_18 + streamed_4 + streamed_8;
73 const float momdensity_0 = -streamed_13 - streamed_17 - streamed_3 - streamed_7 - streamed_9 + vel0Term;
74 const float vel1Term = streamed_1 + streamed_11 + streamed_15 + streamed_7;
75 const float momdensity_1 = -streamed_10 - streamed_12 - streamed_16 - streamed_2 + streamed_8 - streamed_9 + vel1Term;
76 const float vel2Term = streamed_12 + streamed_13 + streamed_5;
77 const float rho = streamed_0 + streamed_16 + streamed_17 + streamed_2 + streamed_3 + streamed_6 + streamed_9 + vel0Term + vel1Term + vel2Term;
78 const float momdensity_2 = streamed_11 + streamed_14 - streamed_15 - streamed_16 - streamed_17 - streamed_18 - streamed_6 + vel2Term;
79 const float u_0 = momdensity_0 * ((1.0f) / (rho)) + 0.5f * ((1.0f) / (rho)) * _data_force[_stride_force_0 * ctr_0 + _stride_force_1 * ctr_1 + _stride_force_2 * ctr_2];
80 const float u_1 = momdensity_1 * ((1.0f) / (rho)) + 0.5f * ((1.0f) / (rho)) * _data_force[_stride_force_0 * ctr_0 + _stride_force_1 * ctr_1 + _stride_force_2 * ctr_2 + _stride_force_3];
81 const float u_2 = momdensity_2 * ((1.0f) / (rho)) + 0.5f * ((1.0f) / (rho)) * _data_force[_stride_force_0 * ctr_0 + _stride_force_1 * ctr_1 + _stride_force_2 * ctr_2 + 2 * _stride_force_3];
82 _data_velocity[_stride_velocity_0 * ctr_0 + _stride_velocity_1 * ctr_1 + _stride_velocity_2 * ctr_2] = u_0;
83 _data_velocity[_stride_velocity_0 * ctr_0 + _stride_velocity_1 * ctr_1 + _stride_velocity_2 * ctr_2 + _stride_velocity_3] = u_1;
84 _data_velocity[_stride_velocity_0 * ctr_0 + _stride_velocity_1 * ctr_1 + _stride_velocity_2 * ctr_2 + 2 * _stride_velocity_3] = u_2;
85 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2] = streamed_0;
86 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + _stride_pdfs_tmp_3] = streamed_1;
87 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 2 * _stride_pdfs_tmp_3] = streamed_2;
88 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 3 * _stride_pdfs_tmp_3] = streamed_3;
89 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 4 * _stride_pdfs_tmp_3] = streamed_4;
90 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 5 * _stride_pdfs_tmp_3] = streamed_5;
91 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 6 * _stride_pdfs_tmp_3] = streamed_6;
92 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 7 * _stride_pdfs_tmp_3] = streamed_7;
93 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 8 * _stride_pdfs_tmp_3] = streamed_8;
94 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 9 * _stride_pdfs_tmp_3] = streamed_9;
95 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 10 * _stride_pdfs_tmp_3] = streamed_10;
96 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 11 * _stride_pdfs_tmp_3] = streamed_11;
97 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 12 * _stride_pdfs_tmp_3] = streamed_12;
98 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 13 * _stride_pdfs_tmp_3] = streamed_13;
99 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 14 * _stride_pdfs_tmp_3] = streamed_14;
100 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 15 * _stride_pdfs_tmp_3] = streamed_15;
101 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 16 * _stride_pdfs_tmp_3] = streamed_16;
102 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 17 * _stride_pdfs_tmp_3] = streamed_17;
103 _data_pdfs_tmp[_stride_pdfs_tmp_0 * ctr_0 + _stride_pdfs_tmp_1 * ctr_1 + _stride_pdfs_tmp_2 * ctr_2 + 18 * _stride_pdfs_tmp_3] = streamed_18;
104 }
105 }
106 }
107}
108} // namespace internal_streamsweepsingleprecision_streamsweepsingleprecision
109
110void StreamSweepSinglePrecision::run(IBlock *block) {
111
112 auto velocity = block->getData<field::GhostLayerField<float, 3>>(velocityID);
113 auto pdfs = block->getData<field::GhostLayerField<float, 19>>(pdfsID);
114 auto force = block->getData<field::GhostLayerField<float, 3>>(forceID);
115 field::GhostLayerField<float, 19> *pdfs_tmp;
116 {
117 if (cache_pdfs_.find(block) == cache_pdfs_.end()) {
118 pdfs_tmp = pdfs->cloneUninitialized();
119 cache_pdfs_[block] = pdfs_tmp;
120 } else {
121 pdfs_tmp = cache_pdfs_[block];
122 }
123 }
124
125 WALBERLA_ASSERT_GREATER_EQUAL(-1, -int_c(force->nrOfGhostLayers()))
126 float *RESTRICT const _data_force = force->dataAt(-1, -1, -1, 0);
127 WALBERLA_ASSERT_EQUAL(force->layout(), field::fzyx)
128 WALBERLA_ASSERT_GREATER_EQUAL(-1, -int_c(pdfs->nrOfGhostLayers()))
129 float *RESTRICT const _data_pdfs = pdfs->dataAt(-1, -1, -1, 0);
130 WALBERLA_ASSERT_EQUAL(pdfs->layout(), field::fzyx)
131 WALBERLA_ASSERT_GREATER_EQUAL(-1, -int_c(pdfs_tmp->nrOfGhostLayers()))
132 float *RESTRICT _data_pdfs_tmp = pdfs_tmp->dataAt(-1, -1, -1, 0);
133 WALBERLA_ASSERT_EQUAL(pdfs_tmp->layout(), field::fzyx)
134 WALBERLA_ASSERT_GREATER_EQUAL(-1, -int_c(velocity->nrOfGhostLayers()))
135 float *RESTRICT _data_velocity = velocity->dataAt(-1, -1, -1, 0);
136 WALBERLA_ASSERT_EQUAL(velocity->layout(), field::fzyx)
137 WALBERLA_ASSERT_GREATER_EQUAL(force->xSizeWithGhostLayer(), int64_t(int64_c(force->xSize()) + 2))
138 const int64_t _size_force_0 = int64_t(int64_c(force->xSize()) + 2);
139 WALBERLA_ASSERT_EQUAL(force->layout(), field::fzyx)
140 WALBERLA_ASSERT_GREATER_EQUAL(force->ySizeWithGhostLayer(), int64_t(int64_c(force->ySize()) + 2))
141 const int64_t _size_force_1 = int64_t(int64_c(force->ySize()) + 2);
142 WALBERLA_ASSERT_EQUAL(force->layout(), field::fzyx)
143 WALBERLA_ASSERT_GREATER_EQUAL(force->zSizeWithGhostLayer(), int64_t(int64_c(force->zSize()) + 2))
144 const int64_t _size_force_2 = int64_t(int64_c(force->zSize()) + 2);
145 WALBERLA_ASSERT_EQUAL(force->layout(), field::fzyx)
146 const int64_t _stride_force_0 = int64_t(force->xStride());
147 const int64_t _stride_force_1 = int64_t(force->yStride());
148 const int64_t _stride_force_2 = int64_t(force->zStride());
149 const int64_t _stride_force_3 = int64_t(1 * int64_t(force->fStride()));
150 const int64_t _stride_pdfs_0 = int64_t(pdfs->xStride());
151 const int64_t _stride_pdfs_1 = int64_t(pdfs->yStride());
152 const int64_t _stride_pdfs_2 = int64_t(pdfs->zStride());
153 const int64_t _stride_pdfs_3 = int64_t(1 * int64_t(pdfs->fStride()));
154 const int64_t _stride_pdfs_tmp_0 = int64_t(pdfs_tmp->xStride());
155 const int64_t _stride_pdfs_tmp_1 = int64_t(pdfs_tmp->yStride());
156 const int64_t _stride_pdfs_tmp_2 = int64_t(pdfs_tmp->zStride());
157 const int64_t _stride_pdfs_tmp_3 = int64_t(1 * int64_t(pdfs_tmp->fStride()));
158 const int64_t _stride_velocity_0 = int64_t(velocity->xStride());
159 const int64_t _stride_velocity_1 = int64_t(velocity->yStride());
160 const int64_t _stride_velocity_2 = int64_t(velocity->zStride());
161 const int64_t _stride_velocity_3 = int64_t(1 * int64_t(velocity->fStride()));
162 internal_streamsweepsingleprecision_streamsweepsingleprecision::streamsweepsingleprecision_streamsweepsingleprecision(_data_force, _data_pdfs, _data_pdfs_tmp, _data_velocity, _size_force_0, _size_force_1, _size_force_2, _stride_force_0, _stride_force_1, _stride_force_2, _stride_force_3, _stride_pdfs_0, _stride_pdfs_1, _stride_pdfs_2, _stride_pdfs_3, _stride_pdfs_tmp_0, _stride_pdfs_tmp_1, _stride_pdfs_tmp_2, _stride_pdfs_tmp_3, _stride_velocity_0, _stride_velocity_1, _stride_velocity_2, _stride_velocity_3);
163 pdfs->swapDataPointers(pdfs_tmp);
164}
165
166void StreamSweepSinglePrecision::runOnCellInterval(const shared_ptr<StructuredBlockStorage> &blocks, const CellInterval &globalCellInterval, cell_idx_t ghostLayers, IBlock *block) {
167
168 CellInterval ci = globalCellInterval;
169 CellInterval blockBB = blocks->getBlockCellBB(*block);
170 blockBB.expand(ghostLayers);
171 ci.intersect(blockBB);
172 blocks->transformGlobalToBlockLocalCellInterval(ci, *block);
173 if (ci.empty())
174 return;
175
176 auto velocity = block->getData<field::GhostLayerField<float, 3>>(velocityID);
177 auto pdfs = block->getData<field::GhostLayerField<float, 19>>(pdfsID);
178 auto force = block->getData<field::GhostLayerField<float, 3>>(forceID);
179 field::GhostLayerField<float, 19> *pdfs_tmp;
180 {
181 if (cache_pdfs_.find(block) == cache_pdfs_.end()) {
182 pdfs_tmp = pdfs->cloneUninitialized();
183 cache_pdfs_[block] = pdfs_tmp;
184 } else {
185 pdfs_tmp = cache_pdfs_[block];
186 }
187 }
188
189 WALBERLA_ASSERT_GREATER_EQUAL(ci.xMin() - 1, -int_c(force->nrOfGhostLayers()))
190 WALBERLA_ASSERT_GREATER_EQUAL(ci.yMin() - 1, -int_c(force->nrOfGhostLayers()))
191 WALBERLA_ASSERT_GREATER_EQUAL(ci.zMin() - 1, -int_c(force->nrOfGhostLayers()))
192 float *RESTRICT const _data_force = force->dataAt(ci.xMin() - 1, ci.yMin() - 1, ci.zMin() - 1, 0);
193 WALBERLA_ASSERT_EQUAL(force->layout(), field::fzyx)
194 WALBERLA_ASSERT_GREATER_EQUAL(ci.xMin() - 1, -int_c(pdfs->nrOfGhostLayers()))
195 WALBERLA_ASSERT_GREATER_EQUAL(ci.yMin() - 1, -int_c(pdfs->nrOfGhostLayers()))
196 WALBERLA_ASSERT_GREATER_EQUAL(ci.zMin() - 1, -int_c(pdfs->nrOfGhostLayers()))
197 float *RESTRICT const _data_pdfs = pdfs->dataAt(ci.xMin() - 1, ci.yMin() - 1, ci.zMin() - 1, 0);
198 WALBERLA_ASSERT_EQUAL(pdfs->layout(), field::fzyx)
199 WALBERLA_ASSERT_GREATER_EQUAL(ci.xMin() - 1, -int_c(pdfs_tmp->nrOfGhostLayers()))
200 WALBERLA_ASSERT_GREATER_EQUAL(ci.yMin() - 1, -int_c(pdfs_tmp->nrOfGhostLayers()))
201 WALBERLA_ASSERT_GREATER_EQUAL(ci.zMin() - 1, -int_c(pdfs_tmp->nrOfGhostLayers()))
202 float *RESTRICT _data_pdfs_tmp = pdfs_tmp->dataAt(ci.xMin() - 1, ci.yMin() - 1, ci.zMin() - 1, 0);
203 WALBERLA_ASSERT_EQUAL(pdfs_tmp->layout(), field::fzyx)
204 WALBERLA_ASSERT_GREATER_EQUAL(ci.xMin() - 1, -int_c(velocity->nrOfGhostLayers()))
205 WALBERLA_ASSERT_GREATER_EQUAL(ci.yMin() - 1, -int_c(velocity->nrOfGhostLayers()))
206 WALBERLA_ASSERT_GREATER_EQUAL(ci.zMin() - 1, -int_c(velocity->nrOfGhostLayers()))
207 float *RESTRICT _data_velocity = velocity->dataAt(ci.xMin() - 1, ci.yMin() - 1, ci.zMin() - 1, 0);
208 WALBERLA_ASSERT_EQUAL(velocity->layout(), field::fzyx)
209 WALBERLA_ASSERT_GREATER_EQUAL(force->xSizeWithGhostLayer(), int64_t(int64_c(ci.xSize()) + 2))
210 const int64_t _size_force_0 = int64_t(int64_c(ci.xSize()) + 2);
211 WALBERLA_ASSERT_EQUAL(force->layout(), field::fzyx)
212 WALBERLA_ASSERT_GREATER_EQUAL(force->ySizeWithGhostLayer(), int64_t(int64_c(ci.ySize()) + 2))
213 const int64_t _size_force_1 = int64_t(int64_c(ci.ySize()) + 2);
214 WALBERLA_ASSERT_EQUAL(force->layout(), field::fzyx)
215 WALBERLA_ASSERT_GREATER_EQUAL(force->zSizeWithGhostLayer(), int64_t(int64_c(ci.zSize()) + 2))
216 const int64_t _size_force_2 = int64_t(int64_c(ci.zSize()) + 2);
217 WALBERLA_ASSERT_EQUAL(force->layout(), field::fzyx)
218 const int64_t _stride_force_0 = int64_t(force->xStride());
219 const int64_t _stride_force_1 = int64_t(force->yStride());
220 const int64_t _stride_force_2 = int64_t(force->zStride());
221 const int64_t _stride_force_3 = int64_t(1 * int64_t(force->fStride()));
222 const int64_t _stride_pdfs_0 = int64_t(pdfs->xStride());
223 const int64_t _stride_pdfs_1 = int64_t(pdfs->yStride());
224 const int64_t _stride_pdfs_2 = int64_t(pdfs->zStride());
225 const int64_t _stride_pdfs_3 = int64_t(1 * int64_t(pdfs->fStride()));
226 const int64_t _stride_pdfs_tmp_0 = int64_t(pdfs_tmp->xStride());
227 const int64_t _stride_pdfs_tmp_1 = int64_t(pdfs_tmp->yStride());
228 const int64_t _stride_pdfs_tmp_2 = int64_t(pdfs_tmp->zStride());
229 const int64_t _stride_pdfs_tmp_3 = int64_t(1 * int64_t(pdfs_tmp->fStride()));
230 const int64_t _stride_velocity_0 = int64_t(velocity->xStride());
231 const int64_t _stride_velocity_1 = int64_t(velocity->yStride());
232 const int64_t _stride_velocity_2 = int64_t(velocity->zStride());
233 const int64_t _stride_velocity_3 = int64_t(1 * int64_t(velocity->fStride()));
234 internal_streamsweepsingleprecision_streamsweepsingleprecision::streamsweepsingleprecision_streamsweepsingleprecision(_data_force, _data_pdfs, _data_pdfs_tmp, _data_velocity, _size_force_0, _size_force_1, _size_force_2, _stride_force_0, _stride_force_1, _stride_force_2, _stride_force_3, _stride_pdfs_0, _stride_pdfs_1, _stride_pdfs_2, _stride_pdfs_3, _stride_pdfs_tmp_0, _stride_pdfs_tmp_1, _stride_pdfs_tmp_2, _stride_pdfs_tmp_3, _stride_velocity_0, _stride_velocity_1, _stride_velocity_2, _stride_velocity_3);
235 pdfs->swapDataPointers(pdfs_tmp);
236}
237
238} // namespace pystencils
239} // namespace walberla
240
241#if (defined WALBERLA_CXX_COMPILER_IS_GNU) || (defined WALBERLA_CXX_COMPILER_IS_CLANG)
242#pragma GCC diagnostic pop
243#endif
244
245#if (defined WALBERLA_CXX_COMPILER_IS_INTEL)
246#pragma warning pop
247#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:42
walberla::pystencils::StreamSweepSinglePrecision::runOnCellInterval
void runOnCellInterval(const shared_ptr< StructuredBlockStorage > &blocks, const CellInterval &globalCellInterval, cell_idx_t ghostLayers, IBlock *block)
Definition StreamSweepSinglePrecision.cpp:166
block
static double * block(double *p, std::size_t index, std::size_t size)
Definition elc.cpp:172
walberla::pystencils::internal_streamsweepsingleprecision_streamsweepsingleprecision::streamsweepsingleprecision_streamsweepsingleprecision
static FUNC_PREFIX void streamsweepsingleprecision_streamsweepsingleprecision(float *RESTRICT const _data_force, float *RESTRICT const _data_pdfs, float *RESTRICT _data_pdfs_tmp, float *RESTRICT _data_velocity, int64_t const _size_force_0, int64_t const _size_force_1, int64_t const _size_force_2, int64_t const _stride_force_0, int64_t const _stride_force_1, int64_t const _stride_force_2, int64_t const _stride_force_3, int64_t const _stride_pdfs_0, int64_t const _stride_pdfs_1, int64_t const _stride_pdfs_2, int64_t const _stride_pdfs_3, int64_t const _stride_pdfs_tmp_0, int64_t const _stride_pdfs_tmp_1, int64_t const _stride_pdfs_tmp_2, int64_t const _stride_pdfs_tmp_3, int64_t const _stride_velocity_0, int64_t const _stride_velocity_1, int64_t const _stride_velocity_2, int64_t const _stride_velocity_3)
Definition StreamSweepSinglePrecision.cpp:49
walberla
\file PackInfoPdfDoublePrecision.cpp \author pystencils
Definition EKWalberla.hpp:36
velocity
static Utils::Vector3d velocity(Particle const &p_ref, Particle const &p_vs)
Velocity of the virtual site.
Definition relative.cpp:64