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
InitialPDFsSetterSinglePrecisionCUDA.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 InitialPDFsSetterSinglePrecisionCUDA.h
17//! \\author pystencils
18//======================================================================================================================
19
20// kernel generated with pystencils v1.3.7, lbmpy v1.3.7, sympy v1.12.1,
21// lbmpy_walberla/pystencils_walberla from waLBerla commit
22// f36fa0a68bae59f0b516f6587ea8fa7c24a41141
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 pdfsID_,
61 BlockDataID velocityID_, float rho_0)
62 : forceID(forceID_), pdfsID(pdfsID_), velocityID(velocityID_),
63 rho_0_(rho_0) {}
64
65 void run(IBlock *block, gpuStream_t stream = nullptr);
66
67 void runOnCellInterval(const shared_ptr<StructuredBlockStorage> &blocks,
68 const CellInterval &globalCellInterval,
69 cell_idx_t ghostLayers, IBlock *block,
70 gpuStream_t stream = nullptr);
71
72 void operator()(IBlock *block, gpuStream_t stream = nullptr) {
74 }
75
76 static std::function<void(IBlock *)>
77 getSweep(const shared_ptr<InitialPDFsSetterSinglePrecisionCUDA> &kernel) {
78 return [kernel](IBlock *b) { kernel->run(b); };
79 }
80
81 static std::function<void(IBlock *, gpuStream_t)> getSweepOnCellInterval(
82 const shared_ptr<InitialPDFsSetterSinglePrecisionCUDA> &kernel,
83 const shared_ptr<StructuredBlockStorage> &blocks,
84 const CellInterval &globalCellInterval, cell_idx_t ghostLayers = 1) {
85 return [kernel, blocks, globalCellInterval,
86 ghostLayers](IBlock *b, gpuStream_t stream = nullptr) {
87 kernel->runOnCellInterval(blocks, globalCellInterval, ghostLayers, b,
88 stream);
89 };
90 }
91
92 std::function<void(IBlock *)> getSweep(gpuStream_t stream = nullptr) {
93 return [this, stream](IBlock *b) { this->run(b, stream); };
94 }
95
96 std::function<void(IBlock *)>
97 getSweepOnCellInterval(const shared_ptr<StructuredBlockStorage> &blocks,
98 const CellInterval &globalCellInterval,
99 cell_idx_t ghostLayers = 1,
100 gpuStream_t stream = nullptr) {
101 return [this, blocks, globalCellInterval, ghostLayers, stream](IBlock *b) {
102 this->runOnCellInterval(blocks, globalCellInterval, ghostLayers, b,
103 stream);
104 };
105 }
106
107 void configure(const shared_ptr<StructuredBlockStorage> & /*blocks*/,
108 IBlock * /*block*/) {}
109
110 inline float getRho_0() const { return rho_0_; }
111 inline void setRho_0(const float value) { rho_0_ = value; }
112
113private:
114 BlockDataID forceID;
115 BlockDataID pdfsID;
116 BlockDataID velocityID;
117 float rho_0_;
118};
119
120} // namespace pystencils
121} // namespace walberla
122
123#if (defined WALBERLA_CXX_COMPILER_IS_GNU) || \
124 (defined WALBERLA_CXX_COMPILER_IS_CLANG)
125#pragma GCC diagnostic pop
126#endif
std::function< void(IBlock *)> getSweepOnCellInterval(const shared_ptr< StructuredBlockStorage > &blocks, const CellInterval &globalCellInterval, cell_idx_t ghostLayers=1, gpuStream_t stream=nullptr)
void runOnCellInterval(const shared_ptr< StructuredBlockStorage > &blocks, const CellInterval &globalCellInterval, cell_idx_t ghostLayers, IBlock *block, gpuStream_t stream=nullptr)
static std::function< void(IBlock *)> getSweep(const shared_ptr< InitialPDFsSetterSinglePrecisionCUDA > &kernel)
std::function< void(IBlock *)> getSweep(gpuStream_t stream=nullptr)
void configure(const shared_ptr< StructuredBlockStorage > &, IBlock *)
static std::function< void(IBlock *, gpuStream_t)> getSweepOnCellInterval(const shared_ptr< InitialPDFsSetterSinglePrecisionCUDA > &kernel, const shared_ptr< StructuredBlockStorage > &blocks, const CellInterval &globalCellInterval, cell_idx_t ghostLayers=1)
InitialPDFsSetterSinglePrecisionCUDA(BlockDataID forceID_, BlockDataID pdfsID_, BlockDataID velocityID_, float rho_0)
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:172
\file PackInfoPdfDoublePrecision.cpp \author pystencils