ESPResSo
Extensible Simulation Package for Research on Soft Matter Systems
Loading...
Searching...
No Matches
ReactionKernelBulk_4_single_precision_CUDA.h
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 ReactionKernelBulk_4_single_precision_CUDA.h
17//! \\author pystencils
18//======================================================================================================================
19
20// kernel generated with pystencils v1.3.7+13.gdfd203a, lbmpy
21// v1.3.7+10.gd3f6236, sympy v1.12.1, lbmpy_walberla/pystencils_walberla from
22// waLBerla commit c69cb11d6a95d32b2280544d3d9abde1fe5fdbb5
23
24#pragma once
25#include "core/DataTypes.h"
26#include "core/logging/Logging.h"
27
28#include "gpu/GPUField.h"
29#include "gpu/GPUWrapper.h"
30
31#include "domain_decomposition/BlockDataID.h"
32#include "domain_decomposition/IBlock.h"
33#include "domain_decomposition/StructuredBlockStorage.h"
34#include "field/SwapableCompare.h"
35
36#include <functional>
37#include <unordered_map>
38
39#ifdef __GNUC__
40#define RESTRICT __restrict__
41#elif _MSC_VER
42#define RESTRICT __restrict
43#else
44#define RESTRICT
45#endif
46
47#if (defined WALBERLA_CXX_COMPILER_IS_GNU) || \
48 (defined WALBERLA_CXX_COMPILER_IS_CLANG)
49#pragma GCC diagnostic push
50#pragma GCC diagnostic ignored "-Wunused-parameter"
51#pragma GCC diagnostic ignored "-Wreorder"
52#endif
53
54namespace walberla {
55namespace pystencils {
56
58public:
60 BlockDataID rho_0ID_, BlockDataID rho_1ID_, BlockDataID rho_2ID_,
61 BlockDataID rho_3ID_, float order_0, float order_1, float order_2,
62 float order_3, float rate_coefficient, float stoech_0, float stoech_1,
63 float stoech_2, float stoech_3)
64 : rho_0ID(rho_0ID_), rho_1ID(rho_1ID_), rho_2ID(rho_2ID_),
65 rho_3ID(rho_3ID_), order_0_(order_0), order_1_(order_1),
66 order_2_(order_2), order_3_(order_3),
67 rate_coefficient_(rate_coefficient), stoech_0_(stoech_0),
68 stoech_1_(stoech_1), stoech_2_(stoech_2), stoech_3_(stoech_3) {}
69
70 void run(IBlock *block, gpuStream_t stream = nullptr);
71
72 void runOnCellInterval(const shared_ptr<StructuredBlockStorage> &blocks,
73 const CellInterval &globalCellInterval,
74 cell_idx_t ghostLayers, IBlock *block,
75 gpuStream_t stream = nullptr);
76
77 void operator()(IBlock *block, gpuStream_t stream = nullptr) {
79 }
80
81 static std::function<void(IBlock *)> getSweep(
82 const shared_ptr<ReactionKernelBulk_4_single_precision_CUDA> &kernel) {
83 return [kernel](IBlock *b) { kernel->run(b); };
84 }
85
86 static std::function<void(IBlock *, gpuStream_t)> getSweepOnCellInterval(
87 const shared_ptr<ReactionKernelBulk_4_single_precision_CUDA> &kernel,
88 const shared_ptr<StructuredBlockStorage> &blocks,
89 const CellInterval &globalCellInterval, cell_idx_t ghostLayers = 1) {
90 return [kernel, blocks, globalCellInterval,
91 ghostLayers](IBlock *b, gpuStream_t stream = nullptr) {
92 kernel->runOnCellInterval(blocks, globalCellInterval, ghostLayers, b,
93 stream);
94 };
95 }
96
97 std::function<void(IBlock *)> getSweep(gpuStream_t stream = nullptr) {
98 return [this, stream](IBlock *b) { this->run(b, stream); };
99 }
100
101 std::function<void(IBlock *)>
102 getSweepOnCellInterval(const shared_ptr<StructuredBlockStorage> &blocks,
103 const CellInterval &globalCellInterval,
104 cell_idx_t ghostLayers = 1,
105 gpuStream_t stream = nullptr) {
106 return [this, blocks, globalCellInterval, ghostLayers, stream](IBlock *b) {
107 this->runOnCellInterval(blocks, globalCellInterval, ghostLayers, b,
108 stream);
109 };
110 }
111
112 void configure(const shared_ptr<StructuredBlockStorage> & /*blocks*/,
113 IBlock * /*block*/) {}
114
115 inline float getOrder_0() const { return order_0_; }
116 inline float getOrder_1() const { return order_1_; }
117 inline float getOrder_2() const { return order_2_; }
118 inline float getOrder_3() const { return order_3_; }
119 inline float getRate_coefficient() const { return rate_coefficient_; }
120 inline float getStoech_0() const { return stoech_0_; }
121 inline float getStoech_1() const { return stoech_1_; }
122 inline float getStoech_2() const { return stoech_2_; }
123 inline float getStoech_3() const { return stoech_3_; }
124 inline void setOrder_0(const float value) { order_0_ = value; }
125 inline void setOrder_1(const float value) { order_1_ = value; }
126 inline void setOrder_2(const float value) { order_2_ = value; }
127 inline void setOrder_3(const float value) { order_3_ = value; }
128 inline void setRate_coefficient(const float value) {
129 rate_coefficient_ = value;
130 }
131 inline void setStoech_0(const float value) { stoech_0_ = value; }
132 inline void setStoech_1(const float value) { stoech_1_ = value; }
133 inline void setStoech_2(const float value) { stoech_2_ = value; }
134 inline void setStoech_3(const float value) { stoech_3_ = value; }
135
136private:
137 BlockDataID rho_0ID;
138 BlockDataID rho_1ID;
139 BlockDataID rho_2ID;
140 BlockDataID rho_3ID;
141 float order_0_;
142 float order_1_;
143 float order_2_;
144 float order_3_;
145 float rate_coefficient_;
146 float stoech_0_;
147 float stoech_1_;
148 float stoech_2_;
149 float stoech_3_;
150};
151
152} // namespace pystencils
153} // namespace walberla
154
155#if (defined WALBERLA_CXX_COMPILER_IS_GNU) || \
156 (defined WALBERLA_CXX_COMPILER_IS_CLANG)
157#pragma GCC diagnostic pop
158#endif
ReactionKernelBulk_4_single_precision_CUDA(BlockDataID rho_0ID_, BlockDataID rho_1ID_, BlockDataID rho_2ID_, BlockDataID rho_3ID_, float order_0, float order_1, float order_2, float order_3, float rate_coefficient, float stoech_0, float stoech_1, float stoech_2, float stoech_3)
std::function< void(IBlock *)> getSweepOnCellInterval(const shared_ptr< StructuredBlockStorage > &blocks, const CellInterval &globalCellInterval, cell_idx_t ghostLayers=1, gpuStream_t stream=nullptr)
static std::function< void(IBlock *, gpuStream_t)> getSweepOnCellInterval(const shared_ptr< ReactionKernelBulk_4_single_precision_CUDA > &kernel, const shared_ptr< StructuredBlockStorage > &blocks, const CellInterval &globalCellInterval, cell_idx_t ghostLayers=1)
static std::function< void(IBlock *)> getSweep(const shared_ptr< ReactionKernelBulk_4_single_precision_CUDA > &kernel)
void runOnCellInterval(const shared_ptr< StructuredBlockStorage > &blocks, const CellInterval &globalCellInterval, cell_idx_t ghostLayers, IBlock *block, gpuStream_t stream=nullptr)
cudaStream_t stream[1]
CUDA streams for parallel computing on CPU and GPU.
static double * block(double *p, std::size_t index, std::size_t size)
Definition elc.cpp:176
\file PackInfoPdfDoublePrecision.cpp \author pystencils