diff --git a/src/basics/basics/writer/WbWriterVtkXmlImageBinary.cpp b/src/basics/basics/writer/WbWriterVtkXmlImageBinary.cpp
index 92434e18ba1a2d4b2bc2027c6697f2d1f8393cf5..798b55919df9e24dbc71ecfded5fb8a913cff8cf 100644
--- a/src/basics/basics/writer/WbWriterVtkXmlImageBinary.cpp
+++ b/src/basics/basics/writer/WbWriterVtkXmlImageBinary.cpp
@@ -40,7 +40,7 @@ using namespace std;
const std::string WbWriterVtkXmlImageBinary::pvdEndTag = " </Collection>\n</VTKFile>";
/*===============================================================================*/
string WbWriterVtkXmlImageBinary::writeCollection(const string &filename, const vector<string> &filenames,
- const double &timeStep, const bool &sepGroups)
+ const double &timeStep, const bool &sepGroups)
{
string vtkfilename = filename + ".pvd";
ofstream out(vtkfilename.c_str());
@@ -79,7 +79,7 @@ string WbWriterVtkXmlImageBinary::writeCollection(const string &filename, const
}
/*===============================================================================*/
string WbWriterVtkXmlImageBinary::addFilesToCollection(const string &filename, const vector<string> &filenames,
- const double &timeStep, const bool &sepGroups)
+ const double &timeStep, const bool &sepGroups)
{
string vtkfilename = filename;
fstream test(vtkfilename.c_str(), ios::in);
@@ -106,9 +106,10 @@ string WbWriterVtkXmlImageBinary::addFilesToCollection(const string &filename, c
return vtkfilename;
}
/*===============================================================================*/
-string WbWriterVtkXmlImageBinary::writeParallelFile(const string &filename, const UbTupleInt6 &wholeExtent, const UbTupleFloat3 &origin, const UbTupleFloat3 &spacing,
- vector<string> &pieceSources, vector<UbTupleInt6> &pieceExtents,
- vector<string> &pointDataNames, vector<string> &cellDataNames)
+string WbWriterVtkXmlImageBinary::writeParallelFile(const string &filename, const UbTupleInt6 &wholeExtent,
+ const UbTupleFloat3 &origin, const UbTupleFloat3 &spacing,
+ vector<string> &pieceSources, vector<UbTupleInt6> &pieceExtents,
+ vector<string> &pointDataNames, vector<string> &cellDataNames)
{
string vtkfilename = filename + ".pvti";
UBLOG(logDEBUG1, "WbWriterVtkXmlImageBinary::writeParallelFile to " << vtkfilename << " - start");
@@ -170,8 +171,8 @@ string WbWriterVtkXmlImageBinary::writeParallelFile(const string &filename, cons
}
/*===============================================================================*/
string WbWriterVtkXmlImageBinary::writeOctsWithCellData(const string &filename, vector<UbTupleFloat3> &nodes,
- vector<UbTupleInt8> &cells, vector<string> &datanames,
- vector<vector<double>> &celldata)
+ vector<UbTupleInt8> & /*cells*/, vector<string> &datanames,
+ vector<vector<double>> &celldata)
{
string vtkfilename = filename + getFileExtension();
UBLOG(logDEBUG1, "WbWriterVtkXmlImageBinary::writeOctsWithCellData to " << vtkfilename << " - start");
@@ -191,8 +192,8 @@ string WbWriterVtkXmlImageBinary::writeOctsWithCellData(const string &filename,
}
/*===============================================================================*/
string WbWriterVtkXmlImageBinary::writeOctsWithNodeData(const string &filename, vector<UbTupleFloat3> &nodes,
- vector<UbTupleUInt8> &cells, vector<string> &datanames,
- vector<vector<double>> &nodedata)
+ vector<UbTupleUInt8> & /*cells*/, vector<string> &datanames,
+ vector<vector<double>> &nodedata)
{
string vtkfilename = filename + getFileExtension();
UBLOG(logDEBUG1, "WbWriterVtkXmlImageBinary::writeOctsWithNodeData to " << vtkfilename << " - start");
@@ -213,8 +214,7 @@ string WbWriterVtkXmlImageBinary::writeOctsWithNodeData(const string &filename,
}
/*===============================================================================*/
string WbWriterVtkXmlImageBinary::writeNodesWithNodeData(const string &filename, vector<UbTupleFloat3> &nodes,
- vector<string> &datanames,
- vector<vector<double>> &nodedata)
+ vector<string> &datanames, vector<vector<double>> &nodedata)
{
string vtkfilename = filename + getFileExtension();
UBLOG(logDEBUG1, "WbWriterVtkXmlImageBinary::writeNodesWithNodeData to " << vtkfilename << " - start");
@@ -231,7 +231,8 @@ string WbWriterVtkXmlImageBinary::writeNodesWithNodeData(const string &filename,
return vtkfilename;
}
-void WbWriterVtkXmlImageBinary::getMetaDataOfImage(vector<UbTupleFloat3> &nodes, UbTupleFloat3& origin, UbTupleFloat3& spacing, UbTupleInt6& extent)
+void WbWriterVtkXmlImageBinary::getMetaDataOfImage(vector<UbTupleFloat3> &nodes, UbTupleFloat3 &origin,
+ UbTupleFloat3 &spacing, UbTupleInt6 &extent)
{
int nofNodes = (int)nodes.size();
val<1>(origin) = val<1>(nodes[0]);
@@ -247,17 +248,17 @@ void WbWriterVtkXmlImageBinary::getMetaDataOfImage(vector<UbTupleFloat3> &nodes,
int ny = (l_y) / val<2>(spacing);
val<3>(spacing) = val<3>(nodes[nx*ny])-val<3>(nodes[0]);
- val<1>(extent) = val<1>(origin)/val<1>(spacing); val<2>(extent) = val<1>(nodes[nofNodes-1])/val<1>(spacing);
- val<3>(extent) = val<2>(origin)/val<2>(spacing); val<4>(extent) = val<2>(nodes[nofNodes-1])/val<2>(spacing);
- val<5>(extent) = val<3>(origin)/val<3>(spacing); val<6>(extent) = val<3>(nodes[nofNodes-1])/val<3>(spacing);
+ val<1>(extent) = val<1>(origin) / val<1>(spacing); val<2>(extent) = val<1>(nodes[nofNodes - 1]) / val<1>(spacing);
+ val<3>(extent) = val<2>(origin) / val<2>(spacing); val<4>(extent) = val<2>(nodes[nofNodes - 1]) / val<2>(spacing);
+ val<5>(extent) = val<3>(origin) / val<3>(spacing); val<6>(extent) = val<3>(nodes[nofNodes - 1]) / val<3>(spacing);
}
-void WbWriterVtkXmlImageBinary::writeData(const string &vtkfilename,
- vector<string> &pointDataNames, vector<string> &cellDataNames,
- vector<vector<double>> &nodedata, vector<vector<double>> &celldata,
- UbTupleInt6& wholeExtent,
- UbTupleFloat3& origin, UbTupleFloat3& spacing, UbTupleInt6& extent, unsigned int precision)
+void WbWriterVtkXmlImageBinary::writeData(const string &vtkfilename, vector<string> &pointDataNames,
+ vector<string> &cellDataNames, vector<vector<double>> &nodedata,
+ vector<vector<double>> &celldata, UbTupleInt6 &wholeExtent,
+ UbTupleFloat3 &origin, UbTupleFloat3 &spacing, UbTupleInt6 &extent,
+ unsigned int precision)
{
ofstream out(vtkfilename.c_str(), ios::out | ios::binary);
out.precision(precision);
@@ -273,14 +274,14 @@ void WbWriterVtkXmlImageBinary::writeData(const string &vtkfilename,
throw UbException(UB_EXARGS, "couldn't open file " + vtkfilename);
}
- size_t nPoints = pointDataNames.size()>0 ? nodedata[0].size() : celldata[0].size();
+ size_t nPoints = pointDataNames.size() > 0 ? nodedata[0].size() : celldata[0].size();
- int bytesPerByteVal = 4; //==sizeof(int)
+ int bytesPerByteVal = 4; //==sizeof(int)
- int bytesScalarData = 1 /*scalar */ * (int)nPoints * sizeof(double);
+ int bytesScalarData = 1 /*scalar */ * (int)nPoints * sizeof(double);
int offset = 0;
-
+
// VTK FILE
out << "<?xml version=\"1.0\"?>\n";
out << "<VTKFile type=\"ImageData\" version=\"0.1\" byte_order=\"LittleEndian\" >"
@@ -307,23 +308,21 @@ void WbWriterVtkXmlImageBinary::writeData(const string &vtkfilename,
<< val<6>(extent) << "\">\n";
// DATA SECTION
- if (pointDataNames.size()>0)
- {
+ if (pointDataNames.size() > 0) {
out << " <PointData>\n";
for (size_t s = 0; s < pointDataNames.size(); ++s) {
- out << " <DataArray type=\"Float64\" Name=\"" << pointDataNames[s] << "\" format=\"appended\" offset=\""
- << offset << "\" /> \n";
+ out << " <DataArray type=\"Float64\" Name=\"" << pointDataNames[s]
+ << "\" format=\"appended\" offset=\"" << offset << "\" /> \n";
offset += (bytesPerByteVal + bytesScalarData);
}
out << " </PointData>\n";
}
- if (cellDataNames.size()>0)
- {
+ if (cellDataNames.size() > 0) {
out << " <CellData>\n";
for (size_t s = 0; s < cellDataNames.size(); ++s) {
- out << " <DataArray type=\"Float64\" Name=\"" << cellDataNames[s] << "\" format=\"appended\" offset=\""
- << offset << "\" /> \n";
+ out << " <DataArray type=\"Float64\" Name=\"" << cellDataNames[s]
+ << "\" format=\"appended\" offset=\"" << offset << "\" /> \n";
offset += (bytesPerByteVal + bytesScalarData);
}
out << " </CellData>\n";
@@ -336,7 +335,6 @@ void WbWriterVtkXmlImageBinary::writeData(const string &vtkfilename,
out << " <AppendedData encoding=\"raw\">\n";
out << "_";
-
// DATA SECTION
// pointData
for (size_t s = 0; s < pointDataNames.size(); ++s) {
@@ -359,4 +357,4 @@ void WbWriterVtkXmlImageBinary::writeData(const string &vtkfilename,
out << "</VTKFile>";
out << endl;
out.close();
-}
\ No newline at end of file
+}
diff --git a/src/gpu/GksGpu/BoundaryConditions/BoundaryCondition.h b/src/gpu/GksGpu/BoundaryConditions/BoundaryCondition.h
index fe4078af95904fa5e1580b54f3aa2edbb006bd3d..9c3bac9c3e2795fa99f339461c6a7f2d16448696 100644
--- a/src/gpu/GksGpu/BoundaryConditions/BoundaryCondition.h
+++ b/src/gpu/GksGpu/BoundaryConditions/BoundaryCondition.h
@@ -47,13 +47,13 @@ struct GKSGPU_EXPORT BoundaryCondition : virtual public BoundaryConditionStruct,
virtual bool isWall() = 0;
virtual bool isFluxBC();
-
+
virtual bool isInsulated();
virtual bool secondCellsNeeded();
virtual void runBoundaryConditionKernel( const SPtr<DataBase> dataBase,
- const Parameters parameters,
+ const Parameters parameters,
const uint level ) = 0;
BoundaryConditionStruct toStruct()
diff --git a/src/gpu/GridGenerator/TransientBCSetter/TransientBCSetter.cpp b/src/gpu/GridGenerator/TransientBCSetter/TransientBCSetter.cpp
index 3026458c988c4e7624e15304f144d01f40fb0cdf..5f3c4ad492b16c09b26acd00a624a54ad65dffda 100644
--- a/src/gpu/GridGenerator/TransientBCSetter/TransientBCSetter.cpp
+++ b/src/gpu/GridGenerator/TransientBCSetter/TransientBCSetter.cpp
@@ -156,7 +156,8 @@ void VTKFile::unloadFile()
this->loaded = false;
}
-void VTKFile::getData(real* data, uint numberOfNodes, const std::vector<uint>& readIndices, const std::vector<uint>& writeIndices, uint offsetRead, uint offsetWrite)
+void VTKFile::getData(real *data, uint numberOfNodes, const std::vector<uint> &readIndices,
+ const std::vector<uint> &writeIndices, uint offsetRead, uint offsetWrite)
{
if(!this->loaded) loadFile();
@@ -305,7 +306,10 @@ void VTKReader::fillArrays(std::vector<real>& coordsY, std::vector<real>& coords
this->planeNeighbor0PM.push_back(writeIdx);
this->planeNeighbor0MP.push_back(writeIdx);
this->planeNeighbor0MM.push_back(writeIdx);
- found0PP = true; found0PM = true; found0MM = true; found0MP = true;
+ found0PP = true;
+ found0PM = true;
+ found0MM = true;
+ found0MP = true;
}
else
{
@@ -325,40 +329,40 @@ void VTKReader::fillArrays(std::vector<real>& coordsY, std::vector<real>& coords
if(!found0PP) //NT in simulation is EN in precursor
{
- int idx = file.findNeighborPPM(posY, posZ, 0.f);
- if(idx!=-1)
+ int index = file.findNeighborPPM(posY, posZ, 0.f);
+ if(index!=-1)
{
found0PP = true;
- real dy = file.getX(idx)-posY;
- real dz = file.getY(idx)-posZ;
+ real dy = file.getX(index)-posY;
+ real dz = file.getY(index)-posZ;
this->weights0PP.emplace_back(1.f/(dy*dy+dz*dz+eps));
- this->planeNeighbor0PP.emplace_back(getWriteIndex(level, fileId, idx));
+ this->planeNeighbor0PP.emplace_back(getWriteIndex(level, fileId, index));
}
}
if(!found0PM) //NB in simulation is ES in precursor
{
- int idx = file.findNeighborPMM(posY, posZ, 0.f);
- if(idx!=-1)
+ int index = file.findNeighborPMM(posY, posZ, 0.f);
+ if(index!=-1)
{
found0PM = true;
- real dy = file.getX(idx)-posY;
- real dz = file.getY(idx)-posZ;
+ real dy = file.getX(index)-posY;
+ real dz = file.getY(index)-posZ;
this->weights0PM.emplace_back(1.f/(dy*dy+dz*dz+eps));
- this->planeNeighbor0PP.emplace_back(getWriteIndex(level, fileId, idx));
+ this->planeNeighbor0PP.emplace_back(getWriteIndex(level, fileId, index));
}
}
if(!found0MP) //ST in simulation is WN in precursor
{
- int idx = file.findNeighborMPM(posY, posZ, 0.f);
- if(idx!=-1)
+ int index = file.findNeighborMPM(posY, posZ, 0.f);
+ if(index!=-1)
{
found0MP = true;
- real dy = file.getX(idx)-posY;
- real dz = file.getY(idx)-posZ;
+ real dy = file.getX(index)-posY;
+ real dz = file.getY(index)-posZ;
this->weights0MP.emplace_back(1.f/(dy*dy+dz*dz+eps));
- this->planeNeighbor0MP.emplace_back(getWriteIndex(level, fileId, idx));
+ this->planeNeighbor0MP.emplace_back(getWriteIndex(level, fileId, index));
}
}
@@ -413,7 +417,7 @@ void VTKReader::getNextData(real* data, uint numberOfNodes, real time)
{
numberOfFiles++;
- printf("switching to precursor file no. %zd\n", numberOfFiles);
+ printf("switching to precursor file no. %zu\n", numberOfFiles);
if(numberOfFiles == this->fileCollection->files[level][id].size())
throw std::runtime_error("Not enough Precursor Files to read");
@@ -437,4 +441,4 @@ void VTKReader::getNextData(real* data, uint numberOfNodes, real time)
this->nFile[level][id] = numberOfFiles;
}
// }
-}
\ No newline at end of file
+}
diff --git a/src/gpu/GridGenerator/TransientBCSetter/TransientBCSetter.h b/src/gpu/GridGenerator/TransientBCSetter/TransientBCSetter.h
index 5bee61e194670c74ca8bd8da87f3881956fff466..1663a3ff37ba1bb062647847462d4e364baed93b 100644
--- a/src/gpu/GridGenerator/TransientBCSetter/TransientBCSetter.h
+++ b/src/gpu/GridGenerator/TransientBCSetter/TransientBCSetter.h
@@ -32,7 +32,7 @@ struct Quantity
class VTKFile
{
public:
- VTKFile(std::string _fileName):
+ explicit VTKFile(std::string _fileName):
fileName(_fileName)
{
readHeader();
@@ -100,9 +100,9 @@ public:
virtual ~FileCollection() = default;
- virtual size_t getNumberOfQuantities()=0;
+ virtual size_t getNumberOfQuantities() = 0;
- virtual FileType getFileType()=0;
+ virtual FileType getFileType() = 0;
protected:
std::string prefix;
@@ -118,8 +118,8 @@ public:
findFiles();
};
- FileType getFileType(){ return FileType::VTK; };
- size_t getNumberOfQuantities(){ return files[0][0][0].getNumberOfQuantities(); }
+ FileType getFileType() override{ return FileType::VTK; };
+ size_t getNumberOfQuantities() override{ return files[0][0][0].getNumberOfQuantities(); }
private:
@@ -198,4 +198,4 @@ private:
SPtr<FileCollection> createFileCollection(std::string prefix, FileType type);
SPtr<TransientBCInputFileReader> createReaderForCollection(SPtr<FileCollection> fileCollection, uint readLevel);
-#endif //TRANSIENTBCSETTER_H_
\ No newline at end of file
+#endif //TRANSIENTBCSETTER_H_
diff --git a/src/gpu/GridGenerator/grid/BoundaryConditions/Side.cpp b/src/gpu/GridGenerator/grid/BoundaryConditions/Side.cpp
index 0900c2d587ba9811c480427b833e7e083216cf10..5b191ee4e3fcdc0ec71633111085f70c5dc43479 100644
--- a/src/gpu/GridGenerator/grid/BoundaryConditions/Side.cpp
+++ b/src/gpu/GridGenerator/grid/BoundaryConditions/Side.cpp
@@ -1,28 +1,28 @@
//=======================================================================================
-// ____ ____ __ ______ __________ __ __ __ __
-// \ \ | | | | | _ \ |___ ___| | | | | / \ | |
-// \ \ | | | | | |_) | | | | | | | / \ | |
-// \ \ | | | | | _ / | | | | | | / /\ \ | |
-// \ \ | | | | | | \ \ | | | \__/ | / ____ \ | |____
-// \ \ | | |__| |__| \__\ |__| \________/ /__/ \__\ |_______|
-// \ \ | | ________________________________________________________________
-// \ \ | | | ______________________________________________________________|
-// \ \| | | | __ __ __ __ ______ _______
-// \ | | |_____ | | | | | | | | | _ \ / _____)
-// \ | | _____| | | | | | | | | | | \ \ \_______
+// ____ ____ __ ______ __________ __ __ __ __
+// \ \ | | | | | _ \ |___ ___| | | | | / \ | |
+// \ \ | | | | | |_) | | | | | | | / \ | |
+// \ \ | | | | | _ / | | | | | | / /\ \ | |
+// \ \ | | | | | | \ \ | | | \__/ | / ____ \ | |____
+// \ \ | | |__| |__| \__\ |__| \________/ /__/ \__\ |_______|
+// \ \ | | ________________________________________________________________
+// \ \ | | | ______________________________________________________________|
+// \ \| | | | __ __ __ __ ______ _______
+// \ | | |_____ | | | | | | | | | _ \ / _____)
+// \ | | _____| | | | | | | | | | | \ \ \_______
// \ | | | | |_____ | \_/ | | | | |_/ / _____ |
-// \ _____| |__| |________| \_______/ |__| |______/ (_______/
+// \ _____| |__| |________| \_______/ |__| |______/ (_______/
//
-// This file is part of VirtualFluids. VirtualFluids is free software: you can
+// This file is part of VirtualFluids. VirtualFluids is free software: you can
// redistribute it and/or modify it under the terms of the GNU General Public
-// License as published by the Free Software Foundation, either version 3 of
+// License as published by the Free Software Foundation, either version 3 of
// the License, or (at your option) any later version.
-//
-// VirtualFluids is distributed in the hope that it will be useful, but WITHOUT
-// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+//
+// VirtualFluids is distributed in the hope that it will be useful, but WITHOUT
+// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
// for more details.
-//
+//
// You should have received a copy of the GNU General Public License along
// with VirtualFluids (see COPYING.txt). If not, see <http://www.gnu.org/licenses/>.
//
@@ -65,14 +65,14 @@ void Side::addIndices(SPtr<Grid> grid, SPtr<BoundaryCondition> boundaryCondition
|| grid->getFieldEntry(index) == vf::gpu::FLUID_CFC
|| grid->getFieldEntry(index) == vf::gpu::FLUID_CFF
|| grid->getFieldEntry(index) == vf::gpu::FLUID_FCC
- || grid->getFieldEntry(index) == vf::gpu::FLUID_FCF
|| grid->getFieldEntry(index) == vf::gpu::FLUID_FCF
-
+ || grid->getFieldEntry(index) == vf::gpu::FLUID_FCF
+
//! Enforce overlap of BCs on edge nodes
|| grid->getFieldEntry(index) == vf::gpu::BC_PRESSURE
- || grid->getFieldEntry(index) == vf::gpu::BC_VELOCITY
- || grid->getFieldEntry(index) == vf::gpu::BC_NOSLIP
- || grid->getFieldEntry(index) == vf::gpu::BC_SLIP
+ || grid->getFieldEntry(index) == vf::gpu::BC_VELOCITY
+ || grid->getFieldEntry(index) == vf::gpu::BC_NOSLIP
+ || grid->getFieldEntry(index) == vf::gpu::BC_SLIP
|| grid->getFieldEntry(index) == vf::gpu::BC_STRESS ))
{
grid->setFieldEntry(index, boundaryCondition->getType());
@@ -175,7 +175,7 @@ void Side::setQs(SPtr<Grid> grid, SPtr<BoundaryCondition> boundaryCondition, uin
bool alignedWithNormal = (this->getNormal()[0]*grid->getDirection()[dir * DIMENSION + 0]+
this->getNormal()[1]*grid->getDirection()[dir * DIMENSION + 1]+
this->getNormal()[2]*grid->getDirection()[dir * DIMENSION + 2] ) > 0;
-
+
uint neighborIndex = grid->transCoordToIndex( neighborX, neighborY, neighborZ );
if((grid->getFieldEntry(neighborIndex) == vf::gpu::STOPPER_OUT_OF_GRID_BOUNDARY ||
grid->getFieldEntry(neighborIndex) == vf::gpu::STOPPER_OUT_OF_GRID ||
@@ -185,7 +185,7 @@ void Side::setQs(SPtr<Grid> grid, SPtr<BoundaryCondition> boundaryCondition, uin
else
qNode[dir] = -1.0;
}
-
+
boundaryCondition->qs.push_back(qNode);
}
@@ -214,7 +214,7 @@ void Geometry::addIndices(std::vector<SPtr<Grid> > grids, uint level, SPtr<Bound
for (int dir = 0; dir <= grids[level]->getEndDirection(); dir++)
{
- const real q = grids[level]->getQValue(index, dir);
+ const real q = grids[level]->getQValue(index, dir);
qNode[dir] = q;
@@ -284,7 +284,7 @@ void MY::addIndices(std::vector<SPtr<Grid> > grid, uint level, SPtr<BoundaryCond
real coordinateNormal = grid[level]->getStartY() + grid[level]->getDelta();
if( coordinateNormal > grid[0]->getStartY() + grid[0]->getDelta() ) return;
-
+
Side::addIndices(grid[level], boundaryCondition, "y", coordinateNormal, startInner, endInner, startOuter, endOuter);
}
@@ -331,6 +331,6 @@ void PZ::addIndices(std::vector<SPtr<Grid> > grid, uint level, SPtr<BoundaryCond
real coordinateNormal = grid[level]->getEndZ() - grid[level]->getDelta();
if( coordinateNormal < grid[0]->getEndZ() - grid[0]->getDelta() ) return;
-
+
Side::addIndices(grid[level], boundaryCondition, "z", coordinateNormal, startInner, endInner, startOuter, endOuter);
}
diff --git a/src/gpu/GridGenerator/grid/GridBuilder/GridBuilder.h b/src/gpu/GridGenerator/grid/GridBuilder/GridBuilder.h
index 065665d216e3cf7904530b94c8bb6480bb565c8a..f3d850384816f6690e5ffc158bbdc5e1df0ab328 100644
--- a/src/gpu/GridGenerator/grid/GridBuilder/GridBuilder.h
+++ b/src/gpu/GridGenerator/grid/GridBuilder/GridBuilder.h
@@ -163,4 +163,4 @@ public:
};
-#endif
\ No newline at end of file
+#endif
diff --git a/src/gpu/GridGenerator/grid/GridBuilder/LevelGridBuilder.cpp b/src/gpu/GridGenerator/grid/GridBuilder/LevelGridBuilder.cpp
index ff6f4913e4cd8a32c05272ef583f90a2cf226edc..003e6dcd223d2bf019c83f71349a9a7bec84efdc 100644
--- a/src/gpu/GridGenerator/grid/GridBuilder/LevelGridBuilder.cpp
+++ b/src/gpu/GridGenerator/grid/GridBuilder/LevelGridBuilder.cpp
@@ -1,28 +1,28 @@
//=======================================================================================
-// ____ ____ __ ______ __________ __ __ __ __
-// \ \ | | | | | _ \ |___ ___| | | | | / \ | |
-// \ \ | | | | | |_) | | | | | | | / \ | |
-// \ \ | | | | | _ / | | | | | | / /\ \ | |
-// \ \ | | | | | | \ \ | | | \__/ | / ____ \ | |____
-// \ \ | | |__| |__| \__\ |__| \________/ /__/ \__\ |_______|
-// \ \ | | ________________________________________________________________
-// \ \ | | | ______________________________________________________________|
-// \ \| | | | __ __ __ __ ______ _______
-// \ | | |_____ | | | | | | | | | _ \ / _____)
-// \ | | _____| | | | | | | | | | | \ \ \_______
+// ____ ____ __ ______ __________ __ __ __ __
+// \ \ | | | | | _ \ |___ ___| | | | | / \ | |
+// \ \ | | | | | |_) | | | | | | | / \ | |
+// \ \ | | | | | _ / | | | | | | / /\ \ | |
+// \ \ | | | | | | \ \ | | | \__/ | / ____ \ | |____
+// \ \ | | |__| |__| \__\ |__| \________/ /__/ \__\ |_______|
+// \ \ | | ________________________________________________________________
+// \ \ | | | ______________________________________________________________|
+// \ \| | | | __ __ __ __ ______ _______
+// \ | | |_____ | | | | | | | | | _ \ / _____)
+// \ | | _____| | | | | | | | | | | \ \ \_______
// \ | | | | |_____ | \_/ | | | | |_/ / _____ |
-// \ _____| |__| |________| \_______/ |__| |______/ (_______/
+// \ _____| |__| |________| \_______/ |__| |______/ (_______/
//
-// This file is part of VirtualFluids. VirtualFluids is free software: you can
+// This file is part of VirtualFluids. VirtualFluids is free software: you can
// redistribute it and/or modify it under the terms of the GNU General Public
-// License as published by the Free Software Foundation, either version 3 of
+// License as published by the Free Software Foundation, either version 3 of
// the License, or (at your option) any later version.
-//
-// VirtualFluids is distributed in the hope that it will be useful, but WITHOUT
-// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+//
+// VirtualFluids is distributed in the hope that it will be useful, but WITHOUT
+// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
// for more details.
-//
+//
// You should have received a copy of the GNU General Public License along
// with VirtualFluids (see COPYING.txt). If not, see <http://www.gnu.org/licenses/>.
//
@@ -105,17 +105,17 @@ void LevelGridBuilder::setSlipGeometryBoundaryCondition(real normalX, real norma
for (uint level = 0; level < getNumberOfGridLevels(); level++)
{
- if (boundaryConditions[level]->geometryBoundaryCondition != nullptr)
- {
- boundaryConditions[level]->geometryBoundaryCondition->normalX = normalX;
- boundaryConditions[level]->geometryBoundaryCondition->normalY = normalY;
- boundaryConditions[level]->geometryBoundaryCondition->normalZ = normalZ;
- boundaryConditions[level]->geometryBoundaryCondition->side->addIndices(grids, level, boundaryConditions[level]->geometryBoundaryCondition);
+ if (boundaryConditions[level]->geometryBoundaryCondition != nullptr)
+ {
+ boundaryConditions[level]->geometryBoundaryCondition->normalX = normalX;
+ boundaryConditions[level]->geometryBoundaryCondition->normalY = normalY;
+ boundaryConditions[level]->geometryBoundaryCondition->normalZ = normalZ;
+ boundaryConditions[level]->geometryBoundaryCondition->side->addIndices(grids, level, boundaryConditions[level]->geometryBoundaryCondition);
boundaryConditions[level]->geometryBoundaryCondition->fillSlipNormalLists();
*logging::out << logging::Logger::INFO_INTERMEDIATE << "Set Geometry Slip BC on level " << level << " with " << (int)boundaryConditions[level]->geometryBoundaryCondition->indices.size() <<"\n";
- }
+ }
}
}
@@ -123,10 +123,10 @@ void LevelGridBuilder::setSlipGeometryBoundaryCondition(real normalX, real norma
//! \brief Set stress boundary concdition using iMEM
//! \param samplingOffset number of grid points above boundary where velocity for wall model is sampled
//! \param z0 roughness length [m]
-//! \param dx dx of level 0 [m]
+//! \param dx dx of level 0 [m]
//!
-void LevelGridBuilder::setStressBoundaryCondition( SideType sideType,
- real nomalX, real normalY, real normalZ,
+void LevelGridBuilder::setStressBoundaryCondition( SideType sideType,
+ real nomalX, real normalY, real normalZ,
uint samplingOffset, real z0, real dx)
{
for (uint level = 0; level < getNumberOfGridLevels(); level++)
@@ -178,17 +178,17 @@ void LevelGridBuilder::setVelocityGeometryBoundaryCondition(real vx, real vy, re
for (uint level = 0; level < getNumberOfGridLevels(); level++)
{
- if (boundaryConditions[level]->geometryBoundaryCondition != nullptr)
- {
- boundaryConditions[level]->geometryBoundaryCondition->vx = vx;
- boundaryConditions[level]->geometryBoundaryCondition->vy = vy;
- boundaryConditions[level]->geometryBoundaryCondition->vz = vz;
- boundaryConditions[level]->geometryBoundaryCondition->side->addIndices(grids, level, boundaryConditions[level]->geometryBoundaryCondition);
+ if (boundaryConditions[level]->geometryBoundaryCondition != nullptr)
+ {
+ boundaryConditions[level]->geometryBoundaryCondition->vx = vx;
+ boundaryConditions[level]->geometryBoundaryCondition->vy = vy;
+ boundaryConditions[level]->geometryBoundaryCondition->vz = vz;
+ boundaryConditions[level]->geometryBoundaryCondition->side->addIndices(grids, level, boundaryConditions[level]->geometryBoundaryCondition);
boundaryConditions[level]->geometryBoundaryCondition->fillVelocityLists();
*logging::out << logging::Logger::INFO_INTERMEDIATE << "Set Geometry Velocity BC on level " << level << " with " << (int)boundaryConditions[level]->geometryBoundaryCondition->indices.size() <<"\n";
- }
+ }
}
}
@@ -230,7 +230,7 @@ void LevelGridBuilder::setNoSlipBoundaryCondition(SideType sideType)
noSlipBoundaryCondition->fillVelocityLists();
// now effectively just a wrapper for velocityBC with zero velocity. No distinction in Gridgenerator.
- boundaryConditions[level]->velocityBoundaryConditions.push_back(noSlipBoundaryCondition);
+ boundaryConditions[level]->velocityBoundaryConditions.push_back(noSlipBoundaryCondition);
}
}
}
@@ -241,23 +241,23 @@ void LevelGridBuilder::setNoSlipGeometryBoundaryCondition()
for (uint level = 0; level < getNumberOfGridLevels(); level++)
{
- if (boundaryConditions[level]->geometryBoundaryCondition != nullptr)
- {
- boundaryConditions[level]->geometryBoundaryCondition->side->addIndices(grids, level, boundaryConditions[level]->geometryBoundaryCondition);
+ if (boundaryConditions[level]->geometryBoundaryCondition != nullptr)
+ {
+ boundaryConditions[level]->geometryBoundaryCondition->side->addIndices(grids, level, boundaryConditions[level]->geometryBoundaryCondition);
*logging::out << logging::Logger::INFO_INTERMEDIATE << "Set Geometry No-Slip BC on level " << level << " with " << (int)boundaryConditions[level]->geometryBoundaryCondition->indices.size() <<"\n";
- }
+ }
}
}
-void LevelGridBuilder::setPrecursorBoundaryCondition(SideType sideType, SPtr<FileCollection> fileCollection, int timeStepsBetweenReads,
+void LevelGridBuilder::setPrecursorBoundaryCondition(SideType sideType, SPtr<FileCollection> fileCollection, int timeStepsBetweenReads,
real velocityX, real velocityY, real velocityZ, std::vector<uint> fileLevelToGridLevelMap)
{
- if(fileLevelToGridLevelMap.empty())
+ if(fileLevelToGridLevelMap.empty())
{
*logging::out << logging::Logger::INFO_INTERMEDIATE << "Mapping precursor file levels to the corresponding grid levels" << "\n";
- for (uint level = 0; level < getNumberOfGridLevels(); level++)
+ for (uint level = 0; level < getNumberOfGridLevels(); level++)
fileLevelToGridLevelMap.push_back(level);
}
else
@@ -413,9 +413,9 @@ std::shared_ptr<Grid> LevelGridBuilder::getGrid(int level, int box)
void LevelGridBuilder::checkLevel(int level)
{
if (level >= (int)grids.size())
- {
+ {
std::cout << "wrong level input... return to caller\n";
- return;
+ return;
}
}
@@ -426,16 +426,16 @@ void LevelGridBuilder::getDimensions(int &nx, int &ny, int &nz, const int level)
nz = grids[level]->getNumberOfNodesZ();
}
-void LevelGridBuilder::getNodeValues(real *xCoords, real *yCoords, real *zCoords,
- uint *neighborX, uint *neighborY, uint *neighborZ, uint *neighborNegative,
+void LevelGridBuilder::getNodeValues(real *xCoords, real *yCoords, real *zCoords,
+ uint *neighborX, uint *neighborY, uint *neighborZ, uint *neighborNegative,
uint *geo, const int level) const
{
grids[level]->getNodeValues(xCoords, yCoords, zCoords, neighborX, neighborY, neighborZ, neighborNegative, geo);
}
-GRIDGENERATOR_EXPORT void LevelGridBuilder::getFluidNodeIndices(uint *fluidNodeIndices, const int level) const
-{
+GRIDGENERATOR_EXPORT void LevelGridBuilder::getFluidNodeIndices(uint *fluidNodeIndices, const int level) const
+{
grids[level]->getFluidNodeIndices(fluidNodeIndices);
}
@@ -444,9 +444,9 @@ GRIDGENERATOR_EXPORT void LevelGridBuilder::getFluidNodeIndicesBorder(uint *flui
grids[level]->getFluidNodeIndicesBorder(fluidNodeIndices);
}
-uint LevelGridBuilder::getNumberOfFluidNodes(unsigned int level) const
+uint LevelGridBuilder::getNumberOfFluidNodes(unsigned int level) const
{
- return grids[level]->getNumberOfFluidNodes();
+ return grids[level]->getNumberOfFluidNodes();
}
GRIDGENERATOR_EXPORT uint LevelGridBuilder::getNumberOfFluidNodesBorder(unsigned int level) const
@@ -472,7 +472,7 @@ void LevelGridBuilder::getSlipValues(real* normalX, real* normalY, real* normalZ
for (uint index = 0; index < boundaryCondition->indices.size(); index++)
{
indices[allIndicesCounter] = grids[level]->getSparseIndex(boundaryCondition->indices[index]) + 1;
-
+
normalX[allIndicesCounter] = boundaryCondition->getNormalx(index);
normalY[allIndicesCounter] = boundaryCondition->getNormaly(index);
normalZ[allIndicesCounter] = boundaryCondition->getNormalz(index);
@@ -507,9 +507,9 @@ uint LevelGridBuilder::getStressSize(int level) const
return size;
}
-void LevelGridBuilder::getStressValues( real* normalX, real* normalY, real* normalZ,
- real* vx, real* vy, real* vz,
- real* vx1, real* vy1, real* vz1,
+void LevelGridBuilder::getStressValues( real* normalX, real* normalY, real* normalZ,
+ real* vx, real* vy, real* vz,
+ real* vx1, real* vy1, real* vz1,
int* indices, int* samplingIndices, int* samplingOffset, real* z0, int level) const
{
@@ -565,7 +565,7 @@ void LevelGridBuilder::getVelocityValues(real* vx, real* vy, real* vz, int* indi
{
for (uint i = 0; i < (uint)boundaryCondition->indices.size(); i++)
{
- indices[allIndicesCounter] = grids[level]->getSparseIndex(boundaryCondition->indices[i]) +1;
+ indices[allIndicesCounter] = grids[level]->getSparseIndex(boundaryCondition->indices[i]) +1;
vx[allIndicesCounter] = boundaryCondition->getVx(i);
vy[allIndicesCounter] = boundaryCondition->getVy(i);
@@ -644,17 +644,17 @@ uint LevelGridBuilder::getPrecursorSize(int level) const
return size;
}
-void LevelGridBuilder::getPrecursorValues( uint* neighbor0PP, uint* neighbor0PM, uint* neighbor0MP, uint* neighbor0MM,
+void LevelGridBuilder::getPrecursorValues( uint* neighbor0PP, uint* neighbor0PM, uint* neighbor0MP, uint* neighbor0MM,
real* weights0PP, real* weights0PM, real* weights0MP, real* weights0MM,
- int* indices, std::vector<SPtr<TransientBCInputFileReader>>& reader,
- int& numberOfPrecursorNodes, size_t& numberOfQuantities, uint& timeStepsBetweenReads,
+ int* indices, std::vector<SPtr<TransientBCInputFileReader>>& reader,
+ int& numberOfPrecursorNodes, size_t& numberOfQuantities, uint& timeStepsBetweenReads,
real& velocityX, real& velocityY, real& velocityZ, int level) const
{
int allIndicesCounter = 0;
int allNodesCounter = 0;
uint tmpTimeStepsBetweenReads = 0;
size_t tmpNumberOfQuantities = 0;
-
+
for (auto boundaryCondition : boundaryConditions[level]->precursorBoundaryConditions)
{
if( tmpTimeStepsBetweenReads == 0 )
@@ -680,7 +680,7 @@ void LevelGridBuilder::getPrecursorValues( uint* neighbor0PP, uint* neighbor0PM
BCreader->getWeights(weights0PP, weights0PM, weights0MP, weights0MM);
if(tmpNumberOfQuantities == 0)
tmpNumberOfQuantities = BCreader->getNumberOfQuantities();
- if(tmpNumberOfQuantities != BCreader->getNumberOfQuantities())
+ if(tmpNumberOfQuantities != BCreader->getNumberOfQuantities())
throw std::runtime_error("All precursor files must have the same quantities.");
allNodesCounter += BCreader->getNPointsRead();
velocityX = boundaryCondition->getVelocityX();
@@ -692,7 +692,7 @@ void LevelGridBuilder::getPrecursorValues( uint* neighbor0PP, uint* neighbor0PM
if (tmpTimeStepsBetweenReads == 0)
throw std::runtime_error("timeStepsBetweenReads of precursor needs to be larger than 0.");
timeStepsBetweenReads = tmpTimeStepsBetweenReads;
-
+
if (tmpNumberOfQuantities == 0)
throw std::runtime_error("Number of quantities in precursor needs to be larger than 0.");
numberOfQuantities = tmpNumberOfQuantities;
@@ -718,7 +718,7 @@ uint LevelGridBuilder::getGeometrySize(int level) const
{
if (boundaryConditions[level]->geometryBoundaryCondition)
return (uint)boundaryConditions[level]->geometryBoundaryCondition->indices.size();
-
+
return 0;
}
@@ -739,9 +739,9 @@ void LevelGridBuilder::getGeometryValues(real* vx, real* vy, real* vz, int level
{
for (uint i = 0; i < boundaryConditions[level]->geometryBoundaryCondition->indices.size(); i++)
{
- vx[i] = boundaryConditions[level]->geometryBoundaryCondition->getVx(i);
- vy[i] = boundaryConditions[level]->geometryBoundaryCondition->getVy(i);
- vz[i] = boundaryConditions[level]->geometryBoundaryCondition->getVz(i);
+ vx[i] = boundaryConditions[level]->geometryBoundaryCondition->getVx(i);
+ vy[i] = boundaryConditions[level]->geometryBoundaryCondition->getVy(i);
+ vz[i] = boundaryConditions[level]->geometryBoundaryCondition->getVz(i);
}
}
@@ -756,7 +756,7 @@ void LevelGridBuilder::getGeometryQs(real* qs[27], int level) const
}
}
-void LevelGridBuilder::writeArrows(std::string fileName) const
+void LevelGridBuilder::writeArrows(std::string fileName) const
{
QLineWriter::writeArrows(fileName, boundaryConditions[getNumberOfGridLevels() - 1]->geometryBoundaryCondition, grids[getNumberOfGridLevels() - 1]);
}
@@ -797,7 +797,7 @@ void LevelGridBuilder::findFluidNodes(bool splitDomain)
}
-void LevelGridBuilder::addFluidNodeIndicesMacroVars(const std::vector<uint>& fluidNodeIndicesMacroVars, uint level)
+void LevelGridBuilder::addFluidNodeIndicesMacroVars(const std::vector<uint>& fluidNodeIndicesMacroVars, uint level)
{
grids[level]->addFluidNodeIndicesMacroVars(fluidNodeIndicesMacroVars);
}
@@ -827,17 +827,17 @@ void LevelGridBuilder::sortFluidNodeIndicesAllFeatures(uint level)
grids[level]->sortFluidNodeIndicesAllFeatures();
}
-uint LevelGridBuilder::getNumberOfFluidNodesMacroVars(unsigned int level) const
+uint LevelGridBuilder::getNumberOfFluidNodesMacroVars(unsigned int level) const
{
return grids[level]->getNumberOfFluidNodeIndicesMacroVars();
}
-void LevelGridBuilder::getFluidNodeIndicesMacroVars(uint *fluidNodeIndicesMacroVars, const int level) const
+void LevelGridBuilder::getFluidNodeIndicesMacroVars(uint *fluidNodeIndicesMacroVars, const int level) const
{
grids[level]->getFluidNodeIndicesMacroVars(fluidNodeIndicesMacroVars);
}
-uint LevelGridBuilder::getNumberOfFluidNodesApplyBodyForce(unsigned int level) const
+uint LevelGridBuilder::getNumberOfFluidNodesApplyBodyForce(unsigned int level) const
{
return grids[level]->getNumberOfFluidNodeIndicesApplyBodyForce();
}
@@ -855,4 +855,4 @@ uint LevelGridBuilder::getNumberOfFluidNodesAllFeatures(unsigned int level) cons
void LevelGridBuilder::getFluidNodeIndicesAllFeatures(uint *fluidNodeIndicesAllFeatures, const int level) const
{
grids[level]->getFluidNodeIndicesAllFeatures(fluidNodeIndicesAllFeatures);
-}
\ No newline at end of file
+}
diff --git a/src/gpu/GridGenerator/grid/GridBuilder/LevelGridBuilder.h b/src/gpu/GridGenerator/grid/GridBuilder/LevelGridBuilder.h
index 56ae1e4fce6185591fba97f49ba504ced259aea5..2e0eaf13080c46260de2a0c845fbf784a2cc3e09 100644
--- a/src/gpu/GridGenerator/grid/GridBuilder/LevelGridBuilder.h
+++ b/src/gpu/GridGenerator/grid/GridBuilder/LevelGridBuilder.h
@@ -1,28 +1,28 @@
//=======================================================================================
-// ____ ____ __ ______ __________ __ __ __ __
-// \ \ | | | | | _ \ |___ ___| | | | | / \ | |
-// \ \ | | | | | |_) | | | | | | | / \ | |
-// \ \ | | | | | _ / | | | | | | / /\ \ | |
-// \ \ | | | | | | \ \ | | | \__/ | / ____ \ | |____
-// \ \ | | |__| |__| \__\ |__| \________/ /__/ \__\ |_______|
-// \ \ | | ________________________________________________________________
-// \ \ | | | ______________________________________________________________|
-// \ \| | | | __ __ __ __ ______ _______
-// \ | | |_____ | | | | | | | | | _ \ / _____)
-// \ | | _____| | | | | | | | | | | \ \ \_______
+// ____ ____ __ ______ __________ __ __ __ __
+// \ \ | | | | | _ \ |___ ___| | | | | / \ | |
+// \ \ | | | | | |_) | | | | | | | / \ | |
+// \ \ | | | | | _ / | | | | | | / /\ \ | |
+// \ \ | | | | | | \ \ | | | \__/ | / ____ \ | |____
+// \ \ | | |__| |__| \__\ |__| \________/ /__/ \__\ |_______|
+// \ \ | | ________________________________________________________________
+// \ \ | | | ______________________________________________________________|
+// \ \| | | | __ __ __ __ ______ _______
+// \ | | |_____ | | | | | | | | | _ \ / _____)
+// \ | | _____| | | | | | | | | | | \ \ \_______
// \ | | | | |_____ | \_/ | | | | |_/ / _____ |
-// \ _____| |__| |________| \_______/ |__| |______/ (_______/
+// \ _____| |__| |________| \_______/ |__| |______/ (_______/
//
-// This file is part of VirtualFluids. VirtualFluids is free software: you can
+// This file is part of VirtualFluids. VirtualFluids is free software: you can
// redistribute it and/or modify it under the terms of the GNU General Public
-// License as published by the Free Software Foundation, either version 3 of
+// License as published by the Free Software Foundation, either version 3 of
// the License, or (at your option) any later version.
-//
-// VirtualFluids is distributed in the hope that it will be useful, but WITHOUT
-// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+//
+// VirtualFluids is distributed in the hope that it will be useful, but WITHOUT
+// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
// for more details.
-//
+//
// You should have received a copy of the GNU General Public License along
// with VirtualFluids (see COPYING.txt). If not, see <http://www.gnu.org/licenses/>.
//
@@ -86,8 +86,8 @@ public:
GRIDGENERATOR_EXPORT void setPressureBoundaryCondition(SideType sideType, real rho);
GRIDGENERATOR_EXPORT void setPeriodicBoundaryCondition(bool periodic_X, bool periodic_Y, bool periodic_Z);
GRIDGENERATOR_EXPORT void setNoSlipBoundaryCondition(SideType sideType);
- GRIDGENERATOR_EXPORT void setPrecursorBoundaryCondition(SideType sideType, SPtr<FileCollection> fileCollection, int timeStepsBetweenReads,
- real velocityX=c0o1, real velocityY=c0o1, real velocityZ=c0o1,
+ GRIDGENERATOR_EXPORT void setPrecursorBoundaryCondition(SideType sideType, SPtr<FileCollection> fileCollection, int timeStepsBetweenReads,
+ real velocityX=c0o1, real velocityY=c0o1, real velocityZ=c0o1,
std::vector<uint> fileLevelToGridLevelMap = {});
GRIDGENERATOR_EXPORT void setEnableFixRefinementIntoTheWall(bool enableFixRefinementIntoTheWall);
@@ -106,7 +106,7 @@ public:
GRIDGENERATOR_EXPORT virtual void getFluidNodeIndicesBorder(uint *fluidNodeIndices, const int level) const override;
GRIDGENERATOR_EXPORT virtual void getNodeValues(real *xCoords, real *yCoords, real *zCoords,
- uint *neighborX, uint *neighborY, uint *neighborZ, uint *neighborNegative,
+ uint *neighborX, uint *neighborY, uint *neighborZ, uint *neighborNegative,
uint *geo, const int level) const override;
GRIDGENERATOR_EXPORT virtual void getDimensions(int &nx, int &ny, int &nz, const int level) const override;
@@ -116,12 +116,12 @@ public:
GRIDGENERATOR_EXPORT virtual void getSlipQs(real* qs[27], int level) const override;
GRIDGENERATOR_EXPORT uint getStressSize(int level) const override;
- GRIDGENERATOR_EXPORT virtual void getStressValues( real* normalX, real* normalY, real* normalZ,
- real* vx, real* vy, real* vz,
- real* vx1, real* vy1, real* vz1,
+ GRIDGENERATOR_EXPORT virtual void getStressValues( real* normalX, real* normalY, real* normalZ,
+ real* vx, real* vy, real* vz,
+ real* vx1, real* vy1, real* vz1,
int* indices, int* samplingIndices, int* samplingOffsets, real* z0, int level) const override;
GRIDGENERATOR_EXPORT virtual void getStressQs(real* qs[27], int level) const override;
-
+
GRIDGENERATOR_EXPORT uint getVelocitySize(int level) const override;
GRIDGENERATOR_EXPORT virtual void getVelocityValues(real* vx, real* vy, real* vz, int* indices, int level) const override;
GRIDGENERATOR_EXPORT virtual void getVelocityQs(real* qs[27], int level) const override;
@@ -131,9 +131,9 @@ public:
GRIDGENERATOR_EXPORT virtual void getPressureQs(real* qs[27], int level) const override;
GRIDGENERATOR_EXPORT uint getPrecursorSize(int level) const override;
- GRIDGENERATOR_EXPORT void getPrecursorValues( uint* neighbor0PP, uint* neighbor0PM, uint* neighbor0MP, uint* neighbor0MM,
- real* weights0PP, real* weights0PM, real* weights0MP, real* weights0MM,
- int* indices, std::vector<SPtr<TransientBCInputFileReader>>& reader,
+ GRIDGENERATOR_EXPORT void getPrecursorValues( uint* neighbor0PP, uint* neighbor0PM, uint* neighbor0MP, uint* neighbor0MM,
+ real* weights0PP, real* weights0PM, real* weights0MP, real* weights0MM,
+ int* indices, std::vector<SPtr<TransientBCInputFileReader>>& reader,
int& numberOfPrecursorNodes, size_t& numberOfQuantities, uint& timeStepsBetweenReads,
real& velocityX, real& velocityY, real& velocityZ, int level) const override;
GRIDGENERATOR_EXPORT virtual void getPrecursorQs(real* qs[27], int level) const override;
@@ -150,11 +150,11 @@ public:
GRIDGENERATOR_EXPORT SPtr<GeometryBoundaryCondition> getGeometryBoundaryCondition(uint level) const override;
protected:
-
+
struct BoundaryConditions
{
- BoundaryConditions() = default;
+ BoundaryConditions() = default;
std::vector<SPtr<SlipBoundaryCondition>> slipBoundaryConditions;
@@ -174,7 +174,7 @@ protected:
std::vector<std::shared_ptr<Grid> > grids;
std::vector<SPtr<BoundaryConditions> > boundaryConditions;
-
+
std::array<uint, 6> communicationProcesses;
void checkLevel(int level);
@@ -217,7 +217,7 @@ public:
void addFluidNodeIndicesMacroVars(const std::vector<uint>& fluidNodeIndicesMacroVars, uint level) override;
void addFluidNodeIndicesApplyBodyForce(const std::vector<uint>& fluidNodeIndicesApplyBodyForce, uint level) override;
void addFluidNodeIndicesAllFeatures(const std::vector<uint>& fluidNodeIndicesAllFeatures, uint level) override;
-
+
void sortFluidNodeIndicesMacroVars(uint level) override;
void sortFluidNodeIndicesApplyBodyForce(uint level) override;
void sortFluidNodeIndicesAllFeatures(uint level) override;
@@ -231,4 +231,3 @@ public:
};
#endif
-
diff --git a/src/gpu/GridGenerator/grid/GridImp.cpp b/src/gpu/GridGenerator/grid/GridImp.cpp
index 24dd169e70288b9ae18a29c23dc0c74b14246e3f..05c684410166e329ba63bbe3bdbf0c09e3a881ab 100644
--- a/src/gpu/GridGenerator/grid/GridImp.cpp
+++ b/src/gpu/GridGenerator/grid/GridImp.cpp
@@ -1,28 +1,28 @@
//=======================================================================================
-// ____ ____ __ ______ __________ __ __ __ __
-// \ \ | | | | | _ \ |___ ___| | | | | / \ | |
-// \ \ | | | | | |_) | | | | | | | / \ | |
-// \ \ | | | | | _ / | | | | | | / /\ \ | |
-// \ \ | | | | | | \ \ | | | \__/ | / ____ \ | |____
-// \ \ | | |__| |__| \__\ |__| \________/ /__/ \__\ |_______|
-// \ \ | | ________________________________________________________________
-// \ \ | | | ______________________________________________________________|
-// \ \| | | | __ __ __ __ ______ _______
-// \ | | |_____ | | | | | | | | | _ \ / _____)
-// \ | | _____| | | | | | | | | | | \ \ \_______
+// ____ ____ __ ______ __________ __ __ __ __
+// \ \ | | | | | _ \ |___ ___| | | | | / \ | |
+// \ \ | | | | | |_) | | | | | | | / \ | |
+// \ \ | | | | | _ / | | | | | | / /\ \ | |
+// \ \ | | | | | | \ \ | | | \__/ | / ____ \ | |____
+// \ \ | | |__| |__| \__\ |__| \________/ /__/ \__\ |_______|
+// \ \ | | ________________________________________________________________
+// \ \ | | | ______________________________________________________________|
+// \ \| | | | __ __ __ __ ______ _______
+// \ | | |_____ | | | | | | | | | _ \ / _____)
+// \ | | _____| | | | | | | | | | | \ \ \_______
// \ | | | | |_____ | \_/ | | | | |_/ / _____ |
-// \ _____| |__| |________| \_______/ |__| |______/ (_______/
+// \ _____| |__| |________| \_______/ |__| |______/ (_______/
//
-// This file is part of VirtualFluids. VirtualFluids is free software: you can
+// This file is part of VirtualFluids. VirtualFluids is free software: you can
// redistribute it and/or modify it under the terms of the GNU General Public
-// License as published by the Free Software Foundation, either version 3 of
+// License as published by the Free Software Foundation, either version 3 of
// the License, or (at your option) any later version.
-//
-// VirtualFluids is distributed in the hope that it will be useful, but WITHOUT
-// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+//
+// VirtualFluids is distributed in the hope that it will be useful, but WITHOUT
+// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
// for more details.
-//
+//
// You should have received a copy of the GNU General Public License along
// with VirtualFluids (see COPYING.txt). If not, see <http://www.gnu.org/licenses/>.
//
@@ -33,7 +33,6 @@
#include "GridImp.h"
#include <iostream>
-#include <omp.h>
#include <sstream>
# include <algorithm>
#include <cmath>
@@ -61,8 +60,8 @@ int DIRECTIONS[DIR_END_MAX][DIMENSION];
using namespace vf::gpu;
-GridImp::GridImp(Object* object, real startX, real startY, real startZ, real endX, real endY, real endZ, real delta, Distribution distribution, uint level)
- : object(object),
+GridImp::GridImp(Object* object, real startX, real startY, real startZ, real endX, real endY, real endZ, real delta, Distribution distribution, uint level)
+ : object(object),
startX(startX),
startY(startY),
startZ(startZ),
@@ -135,7 +134,7 @@ void GridImp::inital(const SPtr<Grid> fineGrid, uint numberOfLayers)
#pragma omp parallel for
for (int index = 0; index < (int)this->size; index++)
this->initalNodeToOutOfGrid(index);
-
+
if( this->innerRegionFromFinerGrid ){
*logging::out << logging::Logger::INFO_INTERMEDIATE << "Start setInnerBasedOnFinerGrid()\n";
this->setInnerBasedOnFinerGrid(fineGrid);
@@ -147,12 +146,12 @@ void GridImp::inital(const SPtr<Grid> fineGrid, uint numberOfLayers)
*logging::out << logging::Logger::INFO_INTERMEDIATE << "Start addOverlap()\n";
this->addOverlap();
-
+
*logging::out << logging::Logger::INFO_INTERMEDIATE << "Start fixOddCells()\n";
#pragma omp parallel for
for (int index = 0; index < (int)this->size; index++)
this->fixOddCell(index);
-
+
if( enableFixRefinementIntoTheWall )
{
*logging::out << logging::Logger::INFO_INTERMEDIATE << "Start fixRefinementIntoWall()\n";
@@ -180,12 +179,12 @@ void GridImp::inital(const SPtr<Grid> fineGrid, uint numberOfLayers)
}
}
}
-
+
*logging::out << logging::Logger::INFO_INTERMEDIATE << "Start findEndOfGridStopperNodes()\n";
#pragma omp parallel for
for (int index = 0; index < (int)this->size; index++)
this->findEndOfGridStopperNode(index);
-
+
*logging::out << logging::Logger::INFO_INTERMEDIATE
<< "Grid created: " << "from (" << this->startX << ", " << this->startY << ", " << this->startZ << ") to (" << this->endX << ", " << this->endY << ", " << this->endZ << ")\n"
<< "nodes: " << this->nx << " x " << this->ny << " x " << this->nz << " = " << this->size << "\n";
@@ -209,9 +208,9 @@ void GridImp::freeMemory()
if( this->neighborIndexZ != nullptr ) { delete[] this->neighborIndexZ; this->neighborIndexZ = nullptr; }
if( this->neighborIndexNegative != nullptr ) { delete[] this->neighborIndexNegative; this->neighborIndexNegative = nullptr; }
if( this->sparseIndices != nullptr ) { delete[] this->sparseIndices; this->sparseIndices = nullptr; }
- if( this->qIndices != nullptr ) { delete[] this->qIndices; this->qIndices = nullptr; }
- if( this->qValues != nullptr ) { delete[] this->qValues; this->qValues = nullptr; }
- if( this->qPatches != nullptr ) { delete[] this->qPatches; this->qPatches = nullptr; }
+ if( this->qIndices != nullptr ) { delete[] this->qIndices; this->qIndices = nullptr; }
+ if( this->qValues != nullptr ) { delete[] this->qValues; this->qValues = nullptr; }
+ if( this->qPatches != nullptr ) { delete[] this->qPatches; this->qPatches = nullptr; }
field.freeMemory();
}
@@ -254,7 +253,7 @@ void GridImp::discretize(Object* solidObject, char innerType, char outerType)
this->sparseIndices[index] = index;
if( this->getFieldEntry(index) == innerType ) continue;
-
+
real x, y, z;
this->transIndexToCoords(index, x, y, z);
@@ -279,7 +278,7 @@ bool GridImp::isInside(const Cell& cell) const
// | +-----+-----+-----+ | +-----+-----+-----+
// +---------+ +---------+
// 0 1 2 0 1 2
-// even even even
+// even even even
// odd odd odd
//
Cell GridImp::getOddCellFromIndex(uint index) const
@@ -349,7 +348,7 @@ void GridImp::addOverlap()
void GridImp::setOverlapTmp( uint index )
{
if( this->field.is( index, INVALID_OUT_OF_GRID ) ){
-
+
if( this->hasNeighborOfType(index, FLUID) ){
this->field.setFieldEntry( index, OVERLAP_TMP );
}
@@ -380,7 +379,7 @@ void GridImp::fixRefinementIntoWall(uint xIndex, uint yIndex, uint zIndex, int d
if( this->xOddStart && ( dir == 1 || dir == -1 ) && ( xIndex % 2 == 0 && xIndex != 0 ) ) return;
if( this->yOddStart && ( dir == 2 || dir == -2 ) && ( yIndex % 2 == 0 && yIndex != 0 ) ) return;
if( this->zOddStart && ( dir == 3 || dir == -3 ) && ( zIndex % 2 == 0 && zIndex != 0 ) ) return;
-
+
//////////////////////////////////////////////////////////////////////////
real dx{ 0.0 }, dy{ 0.0 }, dz{ 0.0 };
@@ -433,31 +432,31 @@ void GridImp::findStopperNode(uint index) // deprecated
void GridImp::findEndOfGridStopperNode(uint index)
{
- if (isValidEndOfGridStopper(index)){
+ if (isValidEndOfGridStopper(index)){
if( this->level != 0 )
- this->field.setFieldEntryToStopperOutOfGrid(index);
+ this->field.setFieldEntryToStopperOutOfGrid(index);
else
this->field.setFieldEntryToStopperOutOfGridBoundary(index);
}
-
- if (isValidEndOfGridBoundaryStopper(index))
- this->field.setFieldEntryToStopperOutOfGridBoundary(index);
+
+ if (isValidEndOfGridBoundaryStopper(index))
+ this->field.setFieldEntryToStopperOutOfGridBoundary(index);
}
void GridImp::findSolidStopperNode(uint index)
{
- if (isValidSolidStopper(index))
- this->field.setFieldEntry(index, STOPPER_SOLID);
+ if (isValidSolidStopper(index))
+ this->field.setFieldEntry(index, STOPPER_SOLID);
}
void GridImp::findBoundarySolidNode(uint index)
{
- if (shouldBeBoundarySolidNode(index))
- {
- this->field.setFieldEntry(index, BC_SOLID);
- this->qIndices[index] = this->numberOfSolidBoundaryNodes++;
- //grid->setNumberOfSolidBoundaryNodes(grid->getNumberOfSolidBoundaryNodes() + 1);
- }
+ if (shouldBeBoundarySolidNode(index))
+ {
+ this->field.setFieldEntry(index, BC_SOLID);
+ this->qIndices[index] = this->numberOfSolidBoundaryNodes++;
+ //grid->setNumberOfSolidBoundaryNodes(grid->getNumberOfSolidBoundaryNodes() + 1);
+ }
}
void GridImp::fixOddCell(uint index)
@@ -483,9 +482,9 @@ bool GridImp::isOutSideOfGrid(Cell &cell) const
bool GridImp::contains(Cell &cell, char type) const
{
for (const auto point : cell) {
- uint index = transCoordToIndex(point.x, point.y, point.z);
- if (index == INVALID_INDEX)
- continue;
+ uint index = transCoordToIndex(point.x, point.y, point.z);
+ if (index == INVALID_INDEX)
+ continue;
if (field.is(index, type))
return true;
}
@@ -495,8 +494,8 @@ bool GridImp::contains(Cell &cell, char type) const
bool GridImp::cellContainsOnly(Cell &cell, char type) const
{
for (const auto point : cell) {
- uint index = transCoordToIndex(point.x, point.y, point.z);
- if (index == INVALID_INDEX)
+ uint index = transCoordToIndex(point.x, point.y, point.z);
+ if (index == INVALID_INDEX)
return false;
if (!field.is(index, type))
return false;
@@ -507,8 +506,8 @@ bool GridImp::cellContainsOnly(Cell &cell, char type) const
bool GridImp::cellContainsOnly(Cell &cell, char typeA, char typeB) const
{
for (const auto point : cell) {
- uint index = transCoordToIndex(point.x, point.y, point.z);
- if (index == INVALID_INDEX)
+ uint index = transCoordToIndex(point.x, point.y, point.z);
+ if (index == INVALID_INDEX)
return false;
if (!field.is(index, typeA) && !field.is(index, typeB))
return false;
@@ -524,91 +523,91 @@ const Object * GridImp::getObject() const
void GridImp::setNodeTo(Cell &cell, char type)
{
for (const auto point : cell) {
- uint index = transCoordToIndex(point.x, point.y, point.z);
- if (index == INVALID_INDEX)
- continue;
- field.setFieldEntry(index, type);
+ uint index = transCoordToIndex(point.x, point.y, point.z);
+ if (index == INVALID_INDEX)
+ continue;
+ field.setFieldEntry(index, type);
}
}
void GridImp::setNodeTo(uint index, char type)
{
- if( index != INVALID_INDEX )
- field.setFieldEntry(index, type);
+ if( index != INVALID_INDEX )
+ field.setFieldEntry(index, type);
}
bool GridImp::isNode(uint index, char type) const
{
if( index != INVALID_INDEX )
- return field.is(index, type);
+ return field.is(index, type);
throw std::runtime_error("GridImp::isNode() -> index == INVALID_INDEX not supported.");
}
bool GridImp::isValidEndOfGridStopper(uint index) const
{
- // Lenz: also includes corner stopper nodes
- if (!this->field.is(index, INVALID_OUT_OF_GRID))
- return false;
+ // Lenz: also includes corner stopper nodes
+ if (!this->field.is(index, INVALID_OUT_OF_GRID))
+ return false;
- return hasNeighborOfType(index, FLUID);
+ return hasNeighborOfType(index, FLUID);
}
bool GridImp::isValidEndOfGridBoundaryStopper(uint index) const
{
- // Lenz: also includes corner stopper nodes
- if (!this->field.is(index, FLUID))
- return false;
+ // Lenz: also includes corner stopper nodes
+ if (!this->field.is(index, FLUID))
+ return false;
- return ! hasAllNeighbors(index);
+ return ! hasAllNeighbors(index);
}
bool GridImp::isValidSolidStopper(uint index) const
{
- // Lenz: also includes corner stopper nodes
- if (!this->field.is(index, INVALID_SOLID))
- return false;
+ // Lenz: also includes corner stopper nodes
+ if (!this->field.is(index, INVALID_SOLID))
+ return false;
- return hasNeighborOfType(index, FLUID);
+ return hasNeighborOfType(index, FLUID);
}
bool GridImp::shouldBeBoundarySolidNode(uint index) const
{
- if (!this->field.is(index, FLUID))
- return false;
+ if (!this->field.is(index, FLUID))
+ return false;
- return hasNeighborOfType(index, STOPPER_SOLID);
+ return hasNeighborOfType(index, STOPPER_SOLID);
}
bool GridImp::hasAllNeighbors(uint index) const
{
- // new version by Lenz, utilizes the range based for loop for all directions
- real x, y, z;
- this->transIndexToCoords(index, x, y, z);
- for (const auto dir : this->distribution) {
- const uint neighborIndex = this->transCoordToIndex(x + dir[0] * this->getDelta(), y + dir[1] * this->getDelta(), z + dir[2] * this->getDelta());
+ // new version by Lenz, utilizes the range based for loop for all directions
+ real x, y, z;
+ this->transIndexToCoords(index, x, y, z);
+ for (const auto dir : this->distribution) {
+ const uint neighborIndex = this->transCoordToIndex(x + dir[0] * this->getDelta(), y + dir[1] * this->getDelta(), z + dir[2] * this->getDelta());
- if (neighborIndex == INVALID_INDEX) return false;
- }
+ if (neighborIndex == INVALID_INDEX) return false;
+ }
- return true;
+ return true;
}
bool GridImp::hasNeighborOfType(uint index, char type) const
{
- // new version by Lenz, utilizes the range based for loop for all directions
- real x, y, z;
- this->transIndexToCoords(index, x, y, z);
- for (const auto dir : this->distribution) {
- const uint neighborIndex = this->transCoordToIndex(x + dir[0] * this->getDelta(), y + dir[1] * this->getDelta(), z + dir[2] * this->getDelta());
+ // new version by Lenz, utilizes the range based for loop for all directions
+ real x, y, z;
+ this->transIndexToCoords(index, x, y, z);
+ for (const auto dir : this->distribution) {
+ const uint neighborIndex = this->transCoordToIndex(x + dir[0] * this->getDelta(), y + dir[1] * this->getDelta(), z + dir[2] * this->getDelta());
- if (neighborIndex == INVALID_INDEX) continue;
+ if (neighborIndex == INVALID_INDEX) continue;
- if (this->field.is(neighborIndex, type))
- return true;
- }
+ if (this->field.is(neighborIndex, type))
+ return true;
+ }
- return false;
+ return false;
}
bool GridImp::nodeInNextCellIs(int index, char type) const
@@ -630,13 +629,13 @@ bool GridImp::nodeInNextCellIs(int index, char type) const
const uint indexXYZ = transCoordToIndex(neighborX, neighborY, neighborZ);
- const bool typeX = indexX == INVALID_INDEX ? false : this->field.is(indexX, type);
- const bool typeY = indexY == INVALID_INDEX ? false : this->field.is(indexY, type);
- const bool typeXY = indexXY == INVALID_INDEX ? false : this->field.is(indexXY, type);
- const bool typeZ = indexZ == INVALID_INDEX ? false : this->field.is(indexZ, type);
- const bool typeYZ = indexYZ == INVALID_INDEX ? false : this->field.is(indexYZ, type);
- const bool typeXZ = indexXZ == INVALID_INDEX ? false : this->field.is(indexXZ, type);
- const bool typeXYZ = indexXYZ == INVALID_INDEX ? false : this->field.is(indexXYZ, type);
+ const bool typeX = indexX == INVALID_INDEX ? false : this->field.is(indexX, type);
+ const bool typeY = indexY == INVALID_INDEX ? false : this->field.is(indexY, type);
+ const bool typeXY = indexXY == INVALID_INDEX ? false : this->field.is(indexXY, type);
+ const bool typeZ = indexZ == INVALID_INDEX ? false : this->field.is(indexZ, type);
+ const bool typeYZ = indexYZ == INVALID_INDEX ? false : this->field.is(indexYZ, type);
+ const bool typeXZ = indexXZ == INVALID_INDEX ? false : this->field.is(indexXZ, type);
+ const bool typeXYZ = indexXYZ == INVALID_INDEX ? false : this->field.is(indexXYZ, type);
return typeX || typeY || typeXY || typeZ || typeYZ
|| typeXZ || typeXYZ;
@@ -661,13 +660,13 @@ bool GridImp::nodeInPreviousCellIs(int index, char type) const
const uint indexXYZ = transCoordToIndex(neighborX, neighborY, neighborZ);
- const bool typeX = indexX == INVALID_INDEX ? false : this->field.is(indexX , type);
- const bool typeY = indexY == INVALID_INDEX ? false : this->field.is(indexY , type);
- const bool typeXY = indexXY == INVALID_INDEX ? false : this->field.is(indexXY , type);
- const bool typeZ = indexZ == INVALID_INDEX ? false : this->field.is(indexZ , type);
- const bool typeYZ = indexYZ == INVALID_INDEX ? false : this->field.is(indexYZ , type);
- const bool typeXZ = indexXZ == INVALID_INDEX ? false : this->field.is(indexXZ , type);
- const bool typeXYZ = indexXYZ == INVALID_INDEX ? false : this->field.is(indexXYZ, type);
+ const bool typeX = indexX == INVALID_INDEX ? false : this->field.is(indexX , type);
+ const bool typeY = indexY == INVALID_INDEX ? false : this->field.is(indexY , type);
+ const bool typeXY = indexXY == INVALID_INDEX ? false : this->field.is(indexXY , type);
+ const bool typeZ = indexZ == INVALID_INDEX ? false : this->field.is(indexZ , type);
+ const bool typeYZ = indexYZ == INVALID_INDEX ? false : this->field.is(indexYZ , type);
+ const bool typeXZ = indexXZ == INVALID_INDEX ? false : this->field.is(indexXZ , type);
+ const bool typeXYZ = indexXYZ == INVALID_INDEX ? false : this->field.is(indexXYZ, type);
return typeX || typeY || typeXY || typeZ || typeYZ
|| typeXZ || typeXYZ;
@@ -678,8 +677,8 @@ bool GridImp::nodeInCellIs(Cell& cell, char type) const
for (const auto node : cell)
{
const uint index = transCoordToIndex(node.x, node.y, node.z);
- if (index == INVALID_INDEX)
- continue;
+ if (index == INVALID_INDEX)
+ continue;
if (field.is(index, type))
return true;
}
@@ -696,9 +695,9 @@ void GridImp::setCellTo(uint index, char type)
for (const auto node : cell)
{
const uint nodeIndex = transCoordToIndex(node.x, node.y, node.z);
- if (nodeIndex == INVALID_INDEX)
- continue;
- this->field.setFieldEntry(nodeIndex, type);
+ if (nodeIndex == INVALID_INDEX)
+ continue;
+ this->field.setFieldEntry(nodeIndex, type);
}
}
@@ -712,10 +711,10 @@ void GridImp::setNonStopperOutOfGridCellTo(uint index, char type)
for (const auto node : cell)
{
const uint nodeIndex = transCoordToIndex(node.x, node.y, node.z);
- if (nodeIndex == INVALID_INDEX)
- continue;
+ if (nodeIndex == INVALID_INDEX)
+ continue;
- if( this->getFieldEntry( nodeIndex ) != STOPPER_OUT_OF_GRID &&
+ if( this->getFieldEntry( nodeIndex ) != STOPPER_OUT_OF_GRID &&
this->getFieldEntry( nodeIndex ) != STOPPER_OUT_OF_GRID_BOUNDARY )
this->field.setFieldEntry(nodeIndex, type);
}
@@ -770,7 +769,7 @@ uint GridImp::transCoordToIndex(const real &x, const real &y, const real &z) con
const uint yIndex = getYIndex(y);
const uint zIndex = getZIndex(z);
- if (xIndex >= nx || yIndex >= ny || zIndex >= nz)
+ if (xIndex >= nx || yIndex >= ny || zIndex >= nz)
return INVALID_INDEX;
return xIndex + nx * (yIndex + ny * zIndex);
@@ -819,20 +818,20 @@ TriangularMeshDiscretizationStrategy * GridImp::getTriangularMeshDiscretizationS
uint GridImp::getNumberOfSolidBoundaryNodes() const
{
- return this->numberOfSolidBoundaryNodes;
+ return this->numberOfSolidBoundaryNodes;
}
void GridImp::setNumberOfSolidBoundaryNodes(uint numberOfSolidBoundaryNodes)
{
- if (numberOfSolidBoundaryNodes < INVALID_INDEX)
- this->numberOfSolidBoundaryNodes = numberOfSolidBoundaryNodes;
+ if (numberOfSolidBoundaryNodes < INVALID_INDEX)
+ this->numberOfSolidBoundaryNodes = numberOfSolidBoundaryNodes;
}
real GridImp::getQValue(const uint index, const uint dir) const
{
- const int qIndex = dir * this->numberOfSolidBoundaryNodes + this->qIndices[index];
+ const int qIndex = dir * this->numberOfSolidBoundaryNodes + this->qIndices[index];
- return this->qValues[qIndex];
+ return this->qValues[qIndex];
}
uint GridImp::getQPatch(const uint index) const
@@ -858,7 +857,7 @@ void GridImp::findSparseIndices(SPtr<Grid> finerGrid)
{
*logging::out << logging::Logger::INFO_INTERMEDIATE << "Find sparse indices...";
auto fineGrid = std::static_pointer_cast<GridImp>(finerGrid);
-
+
this->updateSparseIndices();
#pragma omp parallel for
@@ -906,7 +905,7 @@ void GridImp::updateSparseIndices()
sparseSize = size - removedNodes;
}
-void GridImp::findFluidNodeIndices(bool splitDomain)
+void GridImp::findFluidNodeIndices(bool splitDomain)
{
// find sparse index of all fluid nodes
this->fluidNodeIndices.clear();
@@ -935,7 +934,7 @@ void GridImp::findFluidNodeIndicesBorder() {
// resize fluidNodeIndicesBorder (for better performance in copy operation)
size_t newSize = 0;
for (CommunicationIndices& ci : this->communicationIndices)
- newSize += ci.sendIndices.size();
+ newSize += ci.sendIndices.size();
this->fluidNodeIndicesBorder.reserve(newSize);
// copy all send indices to fluidNodeIndicesBorder
@@ -968,7 +967,7 @@ void GridImp::setNeighborIndices(uint index)
this->setStopperNeighborCoords(index);
return;
}
-
+
if (this->sparseIndices[index] == -1)
return;
@@ -1002,9 +1001,9 @@ void GridImp::setStopperNeighborCoords(uint index)
if (vf::Math::lessEqual(z + delta, endZ + (0.5 * delta)) && !this->field.isInvalidOutOfGrid(this->transCoordToIndex(x, y, z + delta)))
neighborIndexZ[index] = getSparseIndex(x, y, z + delta);
- if (vf::Math::greaterEqual(x - delta, endX) &&
- vf::Math::greaterEqual(y - delta, endY) &&
- vf::Math::greaterEqual(z - delta, endZ) &&
+ if (vf::Math::greaterEqual(x - delta, endX) &&
+ vf::Math::greaterEqual(y - delta, endY) &&
+ vf::Math::greaterEqual(z - delta, endZ) &&
!this->field.isInvalidOutOfGrid(this->transCoordToIndex(x - delta, y - delta, z - delta)))
{
neighborIndexNegative[index] = getSparseIndex(x - delta, y - delta, z - delta);
@@ -1035,7 +1034,7 @@ real GridImp::getNeighborCoord(bool periodicity, real startCoord, real coords[3]
return coords[direction] + delta;
}
-
+
return coords[direction] + delta;
}
@@ -1061,7 +1060,7 @@ real GridImp::getNegativeNeighborCoord(bool periodicity, real startCoord, real c
return getLastFluidNode(coords, direction, startCoord);
}
-
+
return coords[direction] - delta;
}
@@ -1154,15 +1153,15 @@ void GridImp::limitToSubDomain(SPtr<BoundingBox> subDomainBox, LbmOrGks lbmOrGks
// one layer for receive nodes and one for stoppers
if( lbmOrGks == LBM )
tmpSubDomainBox.extend(this->delta);
-
- if (!tmpSubDomainBox.isInside(x, y, z)
+
+ if (!tmpSubDomainBox.isInside(x, y, z)
&& ( this->getFieldEntry(index) == FLUID ||
this->getFieldEntry(index) == FLUID_CFC ||
this->getFieldEntry(index) == FLUID_CFF ||
this->getFieldEntry(index) == FLUID_FCC ||
this->getFieldEntry(index) == FLUID_FCF ||
this->getFieldEntry(index) == BC_SOLID ) )
- {
+ {
this->setFieldEntry(index, STOPPER_OUT_OF_GRID_BOUNDARY);
}
}
@@ -1184,13 +1183,13 @@ void GridImp::limitToSubDomain(SPtr<BoundingBox> subDomainBox, LbmOrGks lbmOrGks
void GridImp::findGridInterfaceCF(uint index, GridImp& finerGrid, LbmOrGks lbmOrGks)
{
- if (lbmOrGks == LBM)
- {
- gridInterface->findInterfaceCF (index, this, &finerGrid);
- gridInterface->findBoundaryGridInterfaceCF(index, this, &finerGrid);
- }
- else if (lbmOrGks == GKS)
- gridInterface->findInterfaceCF_GKS(index, this, &finerGrid);
+ if (lbmOrGks == LBM)
+ {
+ gridInterface->findInterfaceCF (index, this, &finerGrid);
+ gridInterface->findBoundaryGridInterfaceCF(index, this, &finerGrid);
+ }
+ else if (lbmOrGks == GKS)
+ gridInterface->findInterfaceCF_GKS(index, this, &finerGrid);
}
void GridImp::findGridInterfaceFC(uint index, GridImp& finerGrid)
@@ -1217,16 +1216,16 @@ void GridImp::mesh(Object* object)
if (triangularMesh)
triangularMeshDiscretizationStrategy->discretize(triangularMesh, this, INVALID_SOLID, FLUID);
else
- //new method for geometric primitives (not cell based) to be implemented
+ //new method for geometric primitives (not cell based) to be implemented
this->discretize(object, INVALID_SOLID, FLUID);
this->closeNeedleCells();
- #pragma omp parallel for
+ #pragma omp parallel for
for (int index = 0; index < (int)this->size; index++)
this->findSolidStopperNode(index);
- //#pragma omp parallel for
+ //#pragma omp parallel for
for (int index = 0; index < (int)this->size; index++) {
this->findBoundarySolidNode(index);
}
@@ -1359,7 +1358,7 @@ void GridImp::findQs(Object* object) //TODO: enable qs for primitive objects
findQsPrimitive(object);
}
-void GridImp::allocateQs()
+void GridImp::allocateQs()
{
this->qPatches = new uint[this->getNumberOfSolidBoundaryNodes()];
@@ -1379,8 +1378,8 @@ void GridImp::findQs(TriangularMesh &triangularMesh)
if( this->qComputationStage == qComputationStageType::ComputeQs )
allocateQs();
-
-
+
+
#pragma omp parallel for
for (int i = 0; i < triangularMesh.size; i++)
this->findQs(triangularMesh.triangles[i]);
@@ -1406,15 +1405,15 @@ void GridImp::findQs(Triangle &triangle)
//if (!field.isFluid(index))
// continue;
- if( index == INVALID_INDEX ) continue;
+ if( index == INVALID_INDEX ) continue;
const Vertex point(x, y, z);
if( this->qComputationStage == qComputationStageType::ComputeQs ){
if(this->field.is(index, BC_SOLID))
{
- calculateQs(index, point, triangle);
- }
+ calculateQs(index, point, triangle);
+ }
}
else if( this->qComputationStage == qComputationStageType::FindSolidBoundaryNodes )
{
@@ -1449,14 +1448,14 @@ void GridImp::findQsPrimitive(Object * object)
real x,y,z;
this->transIndexToCoords(index,x,y,z);
-
+
const Vertex point(x, y, z);
if( this->qComputationStage == qComputationStageType::ComputeQs ){
if(this->field.is(index, BC_SOLID))
{
- calculateQs(index, point, object);
- }
+ calculateQs(index, point, object);
+ }
}
else if( this->qComputationStage == qComputationStageType::FindSolidBoundaryNodes )
{
@@ -1477,66 +1476,66 @@ void GridImp::calculateQs(const uint index, const Vertex &point, Object* object)
{
Vertex pointOnTriangle, direction;
- real subdistance;
- int error;
- for (int i = distribution.dir_start; i <= distribution.dir_end; i++)
- {
- direction = Vertex( real(distribution.dirs[i * DIMENSION + 0]),
+ real subdistance;
+ int error;
+ for (int i = distribution.dir_start; i <= distribution.dir_end; i++)
+ {
+ direction = Vertex( real(distribution.dirs[i * DIMENSION + 0]),
real(distribution.dirs[i * DIMENSION + 1]),
- real(distribution.dirs[i * DIMENSION + 2]) );
+ real(distribution.dirs[i * DIMENSION + 2]) );
- uint neighborIndex = this->transCoordToIndex(point.x + direction.x * this->delta,
- point.y + direction.y * this->delta,
- point.z + direction.z * this->delta);
+ uint neighborIndex = this->transCoordToIndex(point.x + direction.x * this->delta,
+ point.y + direction.y * this->delta,
+ point.z + direction.z * this->delta);
- if (neighborIndex == INVALID_INDEX) continue;
+ if (neighborIndex == INVALID_INDEX) continue;
- error = object->getIntersection(point, direction, pointOnTriangle, subdistance);
+ error = object->getIntersection(point, direction, pointOnTriangle, subdistance);
- subdistance /= this->delta;
+ subdistance /= this->delta;
- if (error == 0 && vf::Math::lessEqual(subdistance, 1.0) && vf::Math::greaterEqual(subdistance, 0.0))
- {
- if ( -0.5 > this->qValues[i*this->numberOfSolidBoundaryNodes + this->qIndices[index]] ||
+ if (error == 0 && vf::Math::lessEqual(subdistance, 1.0) && vf::Math::greaterEqual(subdistance, 0.0))
+ {
+ if ( -0.5 > this->qValues[i*this->numberOfSolidBoundaryNodes + this->qIndices[index]] ||
subdistance < this->qValues[i*this->numberOfSolidBoundaryNodes + this->qIndices[index]] )
- {
+ {
+
+ this->qValues[i*this->numberOfSolidBoundaryNodes + this->qIndices[index]] = subdistance;
- this->qValues[i*this->numberOfSolidBoundaryNodes + this->qIndices[index]] = subdistance;
-
this->qPatches[ this->qIndices[index] ] = 0;
- }
- }
- }
+ }
+ }
+ }
}
bool GridImp::checkIfAtLeastOneValidQ(const uint index, const Vertex &point, Object* object) const
{
Vertex pointOnTriangle, direction;
- real subdistance;
- int error;
- for (int i = distribution.dir_start; i <= distribution.dir_end; i++)
- {
- direction = Vertex( real(distribution.dirs[i * DIMENSION + 0]),
+ real subdistance;
+ int error;
+ for (int i = distribution.dir_start; i <= distribution.dir_end; i++)
+ {
+ direction = Vertex( real(distribution.dirs[i * DIMENSION + 0]),
real(distribution.dirs[i * DIMENSION + 1]),
- real(distribution.dirs[i * DIMENSION + 2]) );
+ real(distribution.dirs[i * DIMENSION + 2]) );
- uint neighborIndex = this->transCoordToIndex(point.x + direction.x * this->delta,
- point.y + direction.y * this->delta,
- point.z + direction.z * this->delta);
+ uint neighborIndex = this->transCoordToIndex(point.x + direction.x * this->delta,
+ point.y + direction.y * this->delta,
+ point.z + direction.z * this->delta);
- if (neighborIndex == INVALID_INDEX) continue;
+ if (neighborIndex == INVALID_INDEX) continue;
- error = object->getIntersection(point, direction, pointOnTriangle, subdistance);
+ error = object->getIntersection(point, direction, pointOnTriangle, subdistance);
- subdistance /= this->delta;
+ subdistance /= this->delta;
- if (error == 0 && vf::Math::lessEqual(subdistance, 1.0) && vf::Math::greaterEqual(subdistance, 0.0))
- {
- return true;
- }
- }
+ if (error == 0 && vf::Math::lessEqual(subdistance, 1.0) && vf::Math::greaterEqual(subdistance, 0.0))
+ {
+ return true;
+ }
+ }
return false;
}
@@ -1565,7 +1564,7 @@ void GridImp::calculateQs(const Vertex &point, const Triangle &triangle) const
error = triangle.getTriangleIntersection(point, direction, pointOnTriangle, subdistance);
- subdistance /= this->delta;
+ subdistance /= this->delta;
if (error == 0 && subdistance < 1.0 && subdistance > 0.0)
{
@@ -1577,71 +1576,71 @@ void GridImp::calculateQs(const Vertex &point, const Triangle &triangle) const
void GridImp::calculateQs(const uint index, const Vertex &point, const Triangle &triangle) const
{
- Vertex pointOnTriangle, direction;
- real subdistance;
- int error;
- for (int i = distribution.dir_start; i <= distribution.dir_end; i++)
- {
+ Vertex pointOnTriangle, direction;
+ real subdistance;
+ int error;
+ for (int i = distribution.dir_start; i <= distribution.dir_end; i++)
+ {
#if defined(__CUDA_ARCH__)
- direction = Vertex(DIRECTIONS[i][0], DIRECTIONS[i][1], DIRECTIONS[i][2]);
+ direction = Vertex(DIRECTIONS[i][0], DIRECTIONS[i][1], DIRECTIONS[i][2]);
#else
- direction = Vertex( real(distribution.dirs[i * DIMENSION + 0]),
+ direction = Vertex( real(distribution.dirs[i * DIMENSION + 0]),
real(distribution.dirs[i * DIMENSION + 1]),
- real(distribution.dirs[i * DIMENSION + 2]) );
+ real(distribution.dirs[i * DIMENSION + 2]) );
#endif
- uint neighborIndex = this->transCoordToIndex(point.x + direction.x * this->delta,
- point.y + direction.y * this->delta,
- point.z + direction.z * this->delta);
+ uint neighborIndex = this->transCoordToIndex(point.x + direction.x * this->delta,
+ point.y + direction.y * this->delta,
+ point.z + direction.z * this->delta);
- if (neighborIndex == INVALID_INDEX) continue;
+ if (neighborIndex == INVALID_INDEX) continue;
- error = triangle.getTriangleIntersection(point, direction, pointOnTriangle, subdistance);
+ error = triangle.getTriangleIntersection(point, direction, pointOnTriangle, subdistance);
- subdistance /= this->delta;
+ subdistance /= this->delta;
- if (error == 0 && vf::Math::lessEqual(subdistance, 1.0) && vf::Math::greaterEqual(subdistance, 0.0))
- {
- if ( -0.5 > this->qValues[i*this->numberOfSolidBoundaryNodes + this->qIndices[index]] ||
+ if (error == 0 && vf::Math::lessEqual(subdistance, 1.0) && vf::Math::greaterEqual(subdistance, 0.0))
+ {
+ if ( -0.5 > this->qValues[i*this->numberOfSolidBoundaryNodes + this->qIndices[index]] ||
subdistance < this->qValues[i*this->numberOfSolidBoundaryNodes + this->qIndices[index]] )
- {
- this->qValues[i*this->numberOfSolidBoundaryNodes + this->qIndices[index]] = subdistance;
+ {
+ this->qValues[i*this->numberOfSolidBoundaryNodes + this->qIndices[index]] = subdistance;
this->qPatches[ this->qIndices[index] ] = triangle.patchIndex;
- }
- }
- }
+ }
+ }
+ }
}
bool GridImp::checkIfAtLeastOneValidQ(const uint index, const Vertex & point, const Triangle & triangle) const
{
- Vertex pointOnTriangle, direction;
- real subdistance;
- int error;
- for (int i = distribution.dir_start; i <= distribution.dir_end; i++)
- {
+ Vertex pointOnTriangle, direction;
+ real subdistance;
+ int error;
+ for (int i = distribution.dir_start; i <= distribution.dir_end; i++)
+ {
#if defined(__CUDA_ARCH__)
- direction = Vertex(DIRECTIONS[i][0], DIRECTIONS[i][1], DIRECTIONS[i][2]);
+ direction = Vertex(DIRECTIONS[i][0], DIRECTIONS[i][1], DIRECTIONS[i][2]);
#else
- direction = Vertex(real(distribution.dirs[i * DIMENSION + 0]),
+ direction = Vertex(real(distribution.dirs[i * DIMENSION + 0]),
real(distribution.dirs[i * DIMENSION + 1]),
- real(distribution.dirs[i * DIMENSION + 2]));
+ real(distribution.dirs[i * DIMENSION + 2]));
#endif
- uint neighborIndex = this->transCoordToIndex(point.x + direction.x * this->delta,
- point.y + direction.y * this->delta,
- point.z + direction.z * this->delta);
- if (neighborIndex == INVALID_INDEX) continue;
+ uint neighborIndex = this->transCoordToIndex(point.x + direction.x * this->delta,
+ point.y + direction.y * this->delta,
+ point.z + direction.z * this->delta);
+ if (neighborIndex == INVALID_INDEX) continue;
- error = triangle.getTriangleIntersection(point, direction, pointOnTriangle, subdistance);
+ error = triangle.getTriangleIntersection(point, direction, pointOnTriangle, subdistance);
- subdistance /= this->delta;
+ subdistance /= this->delta;
- if (error == 0 && vf::Math::lessEqual(subdistance, 1.0) && vf::Math::greaterEqual(subdistance, 0.0))
- {
- return true;
- }
- }
+ if (error == 0 && vf::Math::lessEqual(subdistance, 1.0) && vf::Math::greaterEqual(subdistance, 0.0))
+ {
+ return true;
+ }
+ }
return false;
}
@@ -1670,13 +1669,13 @@ void GridImp::findCommunicationIndices(int direction, SPtr<BoundingBox> subDomai
void GridImp::findCommunicationIndex( uint index, real coordinate, real limit, int direction ){
// negative direction get a negative sign
- real s = ( direction % 2 == 0 ) ? ( -1.0 ) : ( 1.0 );
+ real s = ( direction % 2 == 0 ) ? ( -1.0 ) : ( 1.0 );
- if (std::abs(coordinate - (limit + s * 0.5 * this->delta)) < 0.1 * this->delta)
- this->communicationIndices[direction].receiveIndices.push_back(index);
+ if (std::abs(coordinate - (limit + s * 0.5 * this->delta)) < 0.1 * this->delta)
+ this->communicationIndices[direction].receiveIndices.push_back(index);
- if (std::abs(coordinate - (limit - s * 0.5 * this->delta)) < 0.1 * this->delta)
- this->communicationIndices[direction].sendIndices.push_back(index);
+ if (std::abs(coordinate - (limit - s * 0.5 * this->delta)) < 0.1 * this->delta)
+ this->communicationIndices[direction].sendIndices.push_back(index);
}
bool GridImp::isSendNode(int index) const
@@ -1722,14 +1721,14 @@ uint GridImp::getReceiveIndex(int direction, uint index)
void GridImp::repairCommunicationIndices(int direction)
{
- this->communicationIndices[direction].sendIndices.insert( this->communicationIndices[direction].sendIndices.end(),
- this->communicationIndices[direction+1].sendIndices.begin(),
+ this->communicationIndices[direction].sendIndices.insert( this->communicationIndices[direction].sendIndices.end(),
+ this->communicationIndices[direction+1].sendIndices.begin(),
this->communicationIndices[direction+1].sendIndices.end() );
- this->communicationIndices[direction+1].receiveIndices.insert( this->communicationIndices[direction+1].receiveIndices.end(),
- this->communicationIndices[direction].receiveIndices.begin(),
+ this->communicationIndices[direction+1].receiveIndices.insert( this->communicationIndices[direction+1].receiveIndices.end(),
+ this->communicationIndices[direction].receiveIndices.begin(),
this->communicationIndices[direction].receiveIndices.end() );
this->communicationIndices[direction].receiveIndices = this->communicationIndices[direction+1].receiveIndices;
@@ -1834,19 +1833,19 @@ real GridImp::getMaximumOnNodes(const real &maxExact, const real &decimalStart,
return maxNode;
}
-uint GridImp::getXIndex(real x) const
-{
- return std::lround((x - startX) / delta);
+uint GridImp::getXIndex(real x) const
+{
+ return std::lround((x - startX) / delta);
}
uint GridImp::getYIndex(real y) const
-{
- return std::lround((y - startY) / delta);
+{
+ return std::lround((y - startY) / delta);
}
uint GridImp::getZIndex(real z) const
-{
- return std::lround((z - startZ) / delta);
+{
+ return std::lround((z - startZ) / delta);
}
real GridImp::getDelta() const
@@ -1861,11 +1860,11 @@ uint GridImp::getSize() const
uint GridImp::getSparseSize() const
{
- return this->sparseSize;
+ return this->sparseSize;
}
-uint GridImp::getNumberOfFluidNodes() const {
- return (uint)this->fluidNodeIndices.size();
+uint GridImp::getNumberOfFluidNodes() const {
+ return (uint)this->fluidNodeIndices.size();
}
Field GridImp::getField() const
@@ -2058,18 +2057,18 @@ void GridImp::getNodeValues(real *xCoords, real *yCoords, real *zCoords, uint *n
}
}
-void GridImp::getFluidNodeIndices(uint *fluidNodeIndices) const
-{
+void GridImp::getFluidNodeIndices(uint *fluidNodeIndices) const
+{
for (uint nodeNumber = 0; nodeNumber < (uint)this->fluidNodeIndices.size(); nodeNumber++)
fluidNodeIndices[nodeNumber] = this->fluidNodeIndices[nodeNumber];
}
-uint GridImp::getNumberOfFluidNodesBorder() const
-{
- return (uint)this->fluidNodeIndicesBorder.size();
+uint GridImp::getNumberOfFluidNodesBorder() const
+{
+ return (uint)this->fluidNodeIndicesBorder.size();
}
-void GridImp::getFluidNodeIndicesBorder(uint *fluidNodeIndicesBorder) const
+void GridImp::getFluidNodeIndicesBorder(uint *fluidNodeIndicesBorder) const
{
for (uint nodeNumber = 0; nodeNumber < (uint)this->fluidNodeIndicesBorder.size(); nodeNumber++)
fluidNodeIndicesBorder[nodeNumber] = this->fluidNodeIndicesBorder[nodeNumber];
@@ -2083,14 +2082,14 @@ void GridImp::addFluidNodeIndicesMacroVars(std::vector<uint> _fluidNodeIndicesMa
}
void GridImp::addFluidNodeIndicesApplyBodyForce(std::vector<uint> _fluidNodeIndicesApplyBodyForce)
-{
-
+{
+
size_t newSize = this->fluidNodeIndicesApplyBodyForce.size()+_fluidNodeIndicesApplyBodyForce.size();
this->fluidNodeIndicesApplyBodyForce.reserve(newSize);
std::copy(_fluidNodeIndicesApplyBodyForce.begin(), _fluidNodeIndicesApplyBodyForce.end(), std::back_inserter(this->fluidNodeIndicesApplyBodyForce));
}
-void GridImp::addFluidNodeIndicesAllFeatures(std::vector<uint> _fluidNodeIndicesAllFeatures)
+void GridImp::addFluidNodeIndicesAllFeatures(std::vector<uint> _fluidNodeIndicesAllFeatures)
{
size_t newSize = this->fluidNodeIndicesAllFeatures.size()+_fluidNodeIndicesAllFeatures.size();
@@ -2109,14 +2108,14 @@ void GridImp::sortFluidNodeIndicesMacroVars()
// Remove indices of fluidNodeIndicesAllFeatures from fluidNodeIndicesMacroVars
if(this->fluidNodeIndicesAllFeatures.size()>0)
{
- this->fluidNodeIndicesMacroVars.erase( std::remove_if( this->fluidNodeIndicesMacroVars.begin(), this->fluidNodeIndicesMacroVars.end(),
+ this->fluidNodeIndicesMacroVars.erase( std::remove_if( this->fluidNodeIndicesMacroVars.begin(), this->fluidNodeIndicesMacroVars.end(),
[&](auto x){return binary_search(fluidNodeIndicesAllFeatures.begin(),fluidNodeIndicesAllFeatures.end(),x);} ),
this->fluidNodeIndicesMacroVars.end()
);
}
// Remove indices of fluidNodeIndicesMacroVars from fluidNodeIndices
- this->fluidNodeIndices.erase( std::remove_if( this->fluidNodeIndices.begin(), this->fluidNodeIndices.end(),
+ this->fluidNodeIndices.erase( std::remove_if( this->fluidNodeIndices.begin(), this->fluidNodeIndices.end(),
[&](auto x){return binary_search(fluidNodeIndicesMacroVars.begin(),fluidNodeIndicesMacroVars.end(),x);} ),
this->fluidNodeIndices.end()
);
@@ -2134,14 +2133,14 @@ void GridImp::sortFluidNodeIndicesApplyBodyForce()
// Remove indices of fluidNodeIndicesAllFeatures from fluidNodeIndicesMacroVars
if(this->fluidNodeIndicesAllFeatures.size()>0)
{
- this->fluidNodeIndicesApplyBodyForce.erase( std::remove_if( this->fluidNodeIndicesApplyBodyForce.begin(), this->fluidNodeIndicesApplyBodyForce.end(),
+ this->fluidNodeIndicesApplyBodyForce.erase( std::remove_if( this->fluidNodeIndicesApplyBodyForce.begin(), this->fluidNodeIndicesApplyBodyForce.end(),
[&](auto x){return binary_search(fluidNodeIndicesAllFeatures.begin(),fluidNodeIndicesAllFeatures.end(),x);} ),
this->fluidNodeIndicesApplyBodyForce.end()
);
}
// Remove indices of fluidNodeIndicesMacroVars from fluidNodeIndices
- this->fluidNodeIndices.erase( std::remove_if( this->fluidNodeIndices.begin(), this->fluidNodeIndices.end(),
+ this->fluidNodeIndices.erase( std::remove_if( this->fluidNodeIndices.begin(), this->fluidNodeIndices.end(),
[&](auto x){return binary_search(fluidNodeIndicesApplyBodyForce.begin(),fluidNodeIndicesApplyBodyForce.end(),x);} ),
this->fluidNodeIndices.end()
);
@@ -2156,34 +2155,34 @@ void GridImp::sortFluidNodeIndicesAllFeatures()
// Remove duplicates
this->fluidNodeIndicesAllFeatures.erase( unique( this->fluidNodeIndicesAllFeatures.begin(), this->fluidNodeIndicesAllFeatures.end() ), this->fluidNodeIndicesAllFeatures.end() );
// Remove indices of fluidNodeIndicesMacroVars from fluidNodeIndices
- this->fluidNodeIndices.erase( std::remove_if( this->fluidNodeIndices.begin(), this->fluidNodeIndices.end(),
+ this->fluidNodeIndices.erase( std::remove_if( this->fluidNodeIndices.begin(), this->fluidNodeIndices.end(),
[&](auto x){return binary_search(fluidNodeIndicesAllFeatures.begin(),fluidNodeIndicesAllFeatures.end(),x);} ),
this->fluidNodeIndices.end()
);
}
}
-uint GridImp::getNumberOfFluidNodeIndicesMacroVars() const {
- return (uint)this->fluidNodeIndicesMacroVars.size();
+uint GridImp::getNumberOfFluidNodeIndicesMacroVars() const {
+ return (uint)this->fluidNodeIndicesMacroVars.size();
}
-uint GridImp::getNumberOfFluidNodeIndicesApplyBodyForce() const {
- return (uint)this->fluidNodeIndicesApplyBodyForce.size();
+uint GridImp::getNumberOfFluidNodeIndicesApplyBodyForce() const {
+ return (uint)this->fluidNodeIndicesApplyBodyForce.size();
}
-uint GridImp::getNumberOfFluidNodeIndicesAllFeatures() const {
- return (uint)this->fluidNodeIndicesAllFeatures.size();
+uint GridImp::getNumberOfFluidNodeIndicesAllFeatures() const {
+ return (uint)this->fluidNodeIndicesAllFeatures.size();
}
-void GridImp::getFluidNodeIndicesMacroVars(uint *_fluidNodeIndicesMacroVars) const
+void GridImp::getFluidNodeIndicesMacroVars(uint *_fluidNodeIndicesMacroVars) const
{
- std::copy(fluidNodeIndicesMacroVars.begin(), fluidNodeIndicesMacroVars.end(), _fluidNodeIndicesMacroVars);
+ std::copy(fluidNodeIndicesMacroVars.begin(), fluidNodeIndicesMacroVars.end(), _fluidNodeIndicesMacroVars);
}
-void GridImp::getFluidNodeIndicesApplyBodyForce(uint *_fluidNodeIndicesApplyBodyForce) const
+void GridImp::getFluidNodeIndicesApplyBodyForce(uint *_fluidNodeIndicesApplyBodyForce) const
{
std::copy(fluidNodeIndicesApplyBodyForce.begin(), fluidNodeIndicesApplyBodyForce.end(), _fluidNodeIndicesApplyBodyForce);
}
-void GridImp::getFluidNodeIndicesAllFeatures(uint *_fluidNodeIndicesAllFeatures) const
+void GridImp::getFluidNodeIndicesAllFeatures(uint *_fluidNodeIndicesAllFeatures) const
{
std::copy(fluidNodeIndicesAllFeatures.begin(), fluidNodeIndicesAllFeatures.end(), _fluidNodeIndicesAllFeatures);
}
diff --git a/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridProvider.h b/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridProvider.h
index 42ba7a464a6e76fd747cc6d11a01e3957c865765..4c50d458c60f04db14c247e16ef3dc44833cb8f7 100644
--- a/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridProvider.h
+++ b/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridProvider.h
@@ -30,7 +30,7 @@ public:
virtual void allocArrays_OffsetScale() = 0;
virtual void allocArrays_taggedFluidNodes() = 0;
- virtual void tagFluidNodeIndices(std::vector<uint> taggedFluidNodeIndices, CollisionTemplate tag, uint level) = 0;
+ virtual void tagFluidNodeIndices(const std::vector<uint>& taggedFluidNodeIndices, CollisionTemplate tag, uint level) = 0;
virtual void sortFluidNodeTags() = 0;
virtual void setDimensions() = 0;
diff --git a/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderFiles/GridReader.cpp b/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderFiles/GridReader.cpp
index 793400869dd29fe12be357ffcf87c0238ea70eb9..66af0d1c4603353148c6201de15d13d6243b0612 100644
--- a/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderFiles/GridReader.cpp
+++ b/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderFiles/GridReader.cpp
@@ -74,11 +74,11 @@ void GridReader::allocArrays_CoordNeighborGeo()
uint numberOfNodesGlobal = 0;
std::cout << "Number of Nodes: " << std::endl;
- for (uint level = 0; level <= maxLevel; level++)
- {
- int numberOfNodesPerLevel = coordX.getSize(level) + 1;
- numberOfNodesGlobal += numberOfNodesPerLevel;
- std::cout << "Level " << level << " = " << numberOfNodesPerLevel << " Nodes" << std::endl;
+ for (uint level = 0; level <= maxLevel; level++)
+ {
+ int numberOfNodesPerLevel = coordX.getSize(level) + 1;
+ numberOfNodesGlobal += numberOfNodesPerLevel;
+ std::cout << "Level " << level << " = " << numberOfNodesPerLevel << " Nodes" << std::endl;
setNumberOfNodes(numberOfNodesPerLevel, level);
@@ -130,9 +130,9 @@ void GridReader::allocArrays_BoundaryValues()
for (uint i = 0; i < channelBoundaryConditions.size(); i++)
{
- if ( this->channelBoundaryConditions[i] == "velocity") { fillVelocityVectors(i); }
- else if (this->channelBoundaryConditions[i] == "pressure") { setPressureValues(i); }
- else if (this->channelBoundaryConditions[i] == "outflow") { setOutflowValues(i); }
+ if ( this->channelBoundaryConditions[i] == "velocity") { fillVelocityVectors(i); }
+ else if (this->channelBoundaryConditions[i] == "pressure") { setPressureValues(i); }
+ else if (this->channelBoundaryConditions[i] == "outflow") { setOutflowValues(i); }
}
setVelocityValues();
@@ -223,13 +223,13 @@ void GridReader::allocArrays_taggedFluidNodes() {
// TODO
}
-void GridReader::tagFluidNodeIndices(std::vector<uint> taggedFluidNodeIndices, CollisionTemplate tag, uint level){
- std::cout << "GridReader::tagFluidNodeIndices not implemented" << std::endl;
+void GridReader::tagFluidNodeIndices(const std::vector<uint>& taggedFluidNodeIndices, CollisionTemplate tag, uint level){
+ std::cout << "GridReader::tagFluidNodeIndices not implemented" << std::endl;
// TODO
}
void GridReader::sortFluidNodeTags(){
- std::cout << "GridReader::sortFluidNodeTags not implemented" << std::endl;
+ std::cout << "GridReader::sortFluidNodeTags not implemented" << std::endl;
// TODO
}
@@ -285,23 +285,23 @@ void GridReader::fillVelocityVectors(int channelSide)
delete[] veloX_ValuesPerSide;
delete[] veloY_ValuesPerSide;
delete[] veloZ_ValuesPerSide;
- }
- }
+ }
+ }
}
-void GridReader::setVelocityValues() {
+void GridReader::setVelocityValues() {
for (int level = 0; level < (int)(velocityX_BCvalues.size()); level++) {
-
- int sizePerLevel = (int) velocityX_BCvalues[level].size();
+
+ int sizePerLevel = (int) velocityX_BCvalues[level].size();
std::cout << "complete size velocity level " << level << " : " << sizePerLevel << std::endl;
setVelocitySizePerLevel(level, sizePerLevel);
-
- if (sizePerLevel > 1) {
+
+ if (sizePerLevel > 1) {
cudaMemoryManager->cudaAllocVeloBC(level);
setVelocity(level, sizePerLevel);
- cudaMemoryManager->cudaCopyVeloBC(level);
+ cudaMemoryManager->cudaCopyVeloBC(level);
}
}
}
@@ -672,8 +672,8 @@ void GridReader::modifyQElement(std::shared_ptr<BoundaryQs> boundaryQ, unsigned
/*------------------------------------------------------------------------------------------------*/
/*---------------------------------------private q methods----------------------------------------*/
/*------------------------------------------------------------------------------------------------*/
-void GridReader::initalVectorForQStruct(std::vector<std::vector<std::vector<real>>> &Qs, std::vector<std::vector<int>> &index,
- std::shared_ptr<BoundaryQs> boundaryQ, unsigned int level) const
+void GridReader::initalVectorForQStruct(std::vector<std::vector<std::vector<real>>> &Qs, std::vector<std::vector<int>> &index,
+ std::shared_ptr<BoundaryQs> boundaryQ, unsigned int level) const
{
boundaryQ->setValuesInVector(Qs, level);
boundaryQ->setIndexInVector(index, level);
@@ -689,7 +689,7 @@ void GridReader::copyVectorsToQStruct(std::vector<std::vector<real>> &Qs,
for (int direction = 0; direction < para->getD3Qxx(); direction++) {
for (size_t indexQ = 0; indexQ < sizeOfValues; indexQ++) {
- qTemp.q27[direction][indexQ] = Qs[direction][indexQ];
+ qTemp.q27[direction][indexQ] = Qs[direction][indexQ];
}
}
@@ -851,46 +851,46 @@ void GridReader::setBoundingBox()
void GridReader::initPeriodicNeigh(std::vector<std::vector<std::vector<unsigned int> > > periodV, std::vector<std::vector<unsigned int> > periodIndex, std::string boundaryCondition)
{
- std::vector<unsigned int>neighVec;
- std::vector<unsigned int>indexVec;
-
- int counter = 0;
-
- for(unsigned int i=0; i<neighX->getLevel();i++) {
- if(boundaryCondition =="periodic_y"){
- neighVec = neighY->getVec(i);
- }
- else if(boundaryCondition =="periodic_x"){
- neighVec = neighX->getVec(i);
- }
- else if(boundaryCondition =="periodic_z"){
- neighVec = neighZ->getVec(i);
- }
- else {
- std::cout << "wrong String in periodicValue" << std::endl;
- exit(1);
- }
+ std::vector<unsigned int>neighVec;
+ std::vector<unsigned int>indexVec;
- for (std::vector<unsigned int>::iterator it = periodIndex[i].begin(); it != periodIndex[i].end(); it++) {
- if(periodV[i][0][counter] != 0) {
- neighVec[*it]=periodV[i][0][counter];
- }
+ int counter = 0;
- counter++;
- }
+ for(unsigned int i=0; i<neighX->getLevel();i++) {
+ if(boundaryCondition =="periodic_y"){
+ neighVec = neighY->getVec(i);
+ }
+ else if(boundaryCondition =="periodic_x"){
+ neighVec = neighX->getVec(i);
+ }
+ else if(boundaryCondition =="periodic_z"){
+ neighVec = neighZ->getVec(i);
+ }
+ else {
+ std::cout << "wrong String in periodicValue" << std::endl;
+ exit(1);
+ }
+ for (std::vector<unsigned int>::iterator it = periodIndex[i].begin(); it != periodIndex[i].end(); it++) {
+ if(periodV[i][0][counter] != 0) {
+ neighVec[*it]=periodV[i][0][counter];
+ }
- if(boundaryCondition =="periodic_y"){
- neighY->setVec(i, neighVec);
- }
- else if(boundaryCondition =="periodic_x"){
- neighX->setVec(i, neighVec);
- }
- else if(boundaryCondition =="periodic_z"){
- neighZ->setVec(i, neighVec);
- }
+ counter++;
+ }
- }
+
+ if(boundaryCondition =="periodic_y"){
+ neighY->setVec(i, neighVec);
+ }
+ else if(boundaryCondition =="periodic_x"){
+ neighX->setVec(i, neighVec);
+ }
+ else if(boundaryCondition =="periodic_z"){
+ neighZ->setVec(i, neighVec);
+ }
+
+ }
}
void GridReader::makeReader(std::shared_ptr<Parameter> para)
@@ -921,9 +921,9 @@ void GridReader::makeReader(std::vector<std::shared_ptr<BoundaryQs> > &BC_Qs, st
void GridReader::setChannelBoundaryCondition()
{
- for (std::size_t i = 0; i < channelDirections.size(); i++)
- {
- this->channelBoundaryConditions[i] = BC_Values[i]->getBoundaryCondition();
- std::cout << this->channelDirections[i] << " Boundary: " << channelBoundaryConditions[i] << std::endl;
- }
-}
\ No newline at end of file
+ for (std::size_t i = 0; i < channelDirections.size(); i++)
+ {
+ this->channelBoundaryConditions[i] = BC_Values[i]->getBoundaryCondition();
+ std::cout << this->channelDirections[i] << " Boundary: " << channelBoundaryConditions[i] << std::endl;
+ }
+}
diff --git a/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderFiles/GridReader.h b/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderFiles/GridReader.h
index 2c17c28022d88a65e922e23d1c89f5166f5a1716..041d2c3ce94592f792c5a850eebd14c07f4db1b4 100644
--- a/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderFiles/GridReader.h
+++ b/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderFiles/GridReader.h
@@ -3,9 +3,9 @@
#include "../GridProvider.h"
-#include <vector>
-#include <string>
#include <memory>
+#include <string>
+#include <vector>
#include "LBM/LB.h"
@@ -16,15 +16,14 @@ class BoundaryValues;
class BoundaryQs;
class CoordNeighborGeoV;
-class VIRTUALFLUIDS_GPU_EXPORT GridReader
- : public GridProvider
+class VIRTUALFLUIDS_GPU_EXPORT GridReader : public GridProvider
{
private:
- bool binaer;
- std::vector<std::string> channelDirections;
- std::vector<std::string> channelBoundaryConditions;
- std::shared_ptr<CoordNeighborGeoV> neighX, neighY, neighZ, neighWSB;
- std::vector<std::shared_ptr<BoundaryValues> > BC_Values;
+ bool binaer;
+ std::vector<std::string> channelDirections;
+ std::vector<std::string> channelBoundaryConditions;
+ std::shared_ptr<CoordNeighborGeoV> neighX, neighY, neighZ, neighWSB;
+ std::vector<std::shared_ptr<BoundaryValues>> BC_Values;
std::vector<std::vector<real>> velocityX_BCvalues, velocityY_BCvalues, velocityZ_BCvalues;
std::vector<std::vector<std::vector<real>>> velocityQs;
@@ -34,59 +33,62 @@ private:
std::vector<std::vector<real>> outflowBCvalues;
public:
- GridReader(FILEFORMAT format, std::shared_ptr<Parameter> para, std::shared_ptr<CudaMemoryManager> cudaMemoryManager);
- ~GridReader();
- void allocArrays_CoordNeighborGeo() override;
- void allocArrays_BoundaryValues() override;
+ GridReader(FILEFORMAT format, std::shared_ptr<Parameter> para,
+ std::shared_ptr<CudaMemoryManager> cudaMemoryManager);
+ ~GridReader() override;
+ void allocArrays_CoordNeighborGeo() override;
+ void allocArrays_BoundaryValues() override;
void allocArrays_OffsetScale() override;
void allocArrays_taggedFluidNodes() override;
- void tagFluidNodeIndices(std::vector<uint> taggedFluidNodeIndices, CollisionTemplate tag, uint level) override;
- void sortFluidNodeTags() override;
+ void tagFluidNodeIndices(const std::vector<uint> &taggedFluidNodeIndices, CollisionTemplate tag, uint level) override;
- void initalValuesDomainDecompostion(int level);
+ void sortFluidNodeTags() override;
- void setChannelBoundaryCondition();
+ void initalValuesDomainDecompostion(int level);
- void allocArrays_BoundaryQs() override;
- bool getBinaer();
- void setDimensions() override;
- void setBoundingBox() override;
- void initPeriodicNeigh(std::vector<std::vector<std::vector<unsigned int> > > periodV, std::vector<std::vector<unsigned int> > periodIndex, std::string way) override;
+ void setChannelBoundaryCondition();
+
+ void allocArrays_BoundaryQs() override;
+ bool getBinaer();
+ void setDimensions() override;
+ void setBoundingBox() override;
+ void initPeriodicNeigh(std::vector<std::vector<std::vector<unsigned int>>> periodV,
+ std::vector<std::vector<unsigned int>> periodIndex, std::string way) override;
private:
- void makeReader(std::shared_ptr<Parameter> para);
- void makeReader(std::vector<std::shared_ptr<BoundaryQs> > &BC_Qs, std::shared_ptr<Parameter> para);
+ void makeReader(std::shared_ptr<Parameter> para);
+ void makeReader(std::vector<std::shared_ptr<BoundaryQs>> &BC_Qs, std::shared_ptr<Parameter> para);
- void setPressureValues(int channelSide) const;
- void setPressRhoBC(int sizePerLevel, int level, int channelSide) const;
+ void setPressureValues(int channelSide) const;
+ void setPressRhoBC(int sizePerLevel, int level, int channelSide) const;
- void fillVelocityVectors(int channelSide);
+ void fillVelocityVectors(int channelSide);
void setVelocityValues();
- void setVelocity(int level, int sizePerLevel) const;
-
- void setOutflowValues(int channelSide) const;
- void setOutflow(int level, int sizePerLevel, int channelSide) const;
+ void setVelocity(int level, int sizePerLevel) const;
+ void setOutflowValues(int channelSide) const;
+ void setOutflow(int level, int sizePerLevel, int channelSide) const;
- //void fillVelocityQVectors(int channelSide);
+ // void fillVelocityQVectors(int channelSide);
void setPressQs(std::shared_ptr<BoundaryQs> boundaryQ) const;
- void setVelocityQs(std::shared_ptr<BoundaryQs> boundaryQ);
- void setOutflowQs(std::shared_ptr<BoundaryQs> boundaryQ) const;
- void setNoSlipQs(std::shared_ptr<BoundaryQs> boundaryQ) const;
- void setGeoQs(std::shared_ptr<BoundaryQs> boundaryQ) const;
- void modifyQElement(std::shared_ptr<BoundaryQs> boundaryQ, unsigned int level) const;
+ void setVelocityQs(std::shared_ptr<BoundaryQs> boundaryQ);
+ void setOutflowQs(std::shared_ptr<BoundaryQs> boundaryQ) const;
+ void setNoSlipQs(std::shared_ptr<BoundaryQs> boundaryQ) const;
+ void setGeoQs(std::shared_ptr<BoundaryQs> boundaryQ) const;
+ void modifyQElement(std::shared_ptr<BoundaryQs> boundaryQ, unsigned int level) const;
- void initalVectorForQStruct(std::vector<std::vector<std::vector<real>>> &Qs, std::vector<std::vector<int>> &index,
+ void initalVectorForQStruct(std::vector<std::vector<std::vector<real>>> &Qs, std::vector<std::vector<int>> &index,
std::shared_ptr<BoundaryQs> boundaryQ, unsigned int level) const;
void copyVectorsToQStruct(std::vector<std::vector<real>> &Qs, std::vector<int> &index,
QforBoundaryConditions &Q) const;
void initalQStruct(QforBoundaryConditions &Q, std::shared_ptr<BoundaryQs> boundaryQ, unsigned int level) const;
- void printQSize(std::string bc, std::shared_ptr<BoundaryQs> boundaryQ, unsigned int level) const;
- void setSizeNoSlip(std::shared_ptr<BoundaryQs> boundaryQ, unsigned int level) const;
- void setSizeGeoQs(std::shared_ptr<BoundaryQs> boundaryQ, unsigned int level) const;
- void setQ27Size(QforBoundaryConditions &Q, real* QQ, unsigned int sizeQ) const;
- bool hasQs(std::shared_ptr<BoundaryQs> boundaryQ, unsigned int level) const;
+ void printQSize(std::string bc, std::shared_ptr<BoundaryQs> boundaryQ, unsigned int level) const;
+ void setSizeNoSlip(std::shared_ptr<BoundaryQs> boundaryQ, unsigned int level) const;
+ void setSizeGeoQs(std::shared_ptr<BoundaryQs> boundaryQ, unsigned int level) const;
+ void setQ27Size(QforBoundaryConditions &Q, real *QQ, unsigned int sizeQ) const;
+ bool hasQs(std::shared_ptr<BoundaryQs> boundaryQ, unsigned int level) const;
+
public:
void initalGridInformations() override;
};
diff --git a/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderGenerator/GridGenerator.cpp b/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderGenerator/GridGenerator.cpp
index d89f4cd85985694b4529d338ab5e4c5a86cd3a38..d14973be77ab6766dc04ec95213efe663db3b873 100644
--- a/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderGenerator/GridGenerator.cpp
+++ b/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderGenerator/GridGenerator.cpp
@@ -20,13 +20,15 @@
using namespace vf::lbm::dir;
-GridGenerator::GridGenerator(std::shared_ptr<GridBuilder> builder, std::shared_ptr<Parameter> para, std::shared_ptr<CudaMemoryManager> cudaMemoryManager, vf::gpu::Communicator& communicator):
- mpiProcessID(communicator.getPID()), builder(builder)
+GridGenerator::GridGenerator(std::shared_ptr<GridBuilder> builder, std::shared_ptr<Parameter> para,
+ std::shared_ptr<CudaMemoryManager> cudaMemoryManager, vf::gpu::Communicator &communicator)
+ : mpiProcessID(communicator.getPID()), builder(builder)
{
this->para = para;
this->cudaMemoryManager = cudaMemoryManager;
this->indexRearrangement = std::make_unique<IndexRearrangementForStreams>(para, builder, communicator);
- this->interpolationGrouper = std::make_unique<InterpolationCellGrouper>(para->getParHallLevels(), para->getParDallLevels(), builder);
+ this->interpolationGrouper =
+ std::make_unique<InterpolationCellGrouper>(para->getParHallLevels(), para->getParDallLevels(), builder);
}
GridGenerator::~GridGenerator() = default;
@@ -59,15 +61,15 @@ void GridGenerator::allocArrays_CoordNeighborGeo()
std::cout << "Number of Level: " << numberOfLevels << std::endl;
int numberOfNodesGlobal = 0;
std::cout << "Number of Nodes: " << std::endl;
-
- for (uint level = 0; level < numberOfLevels; level++)
+
+ for (uint level = 0; level < numberOfLevels; level++)
{
const int numberOfNodesPerLevel = builder->getNumberOfNodes(level) + 1;
numberOfNodesGlobal += numberOfNodesPerLevel;
std::cout << "Level " << level << " = " << numberOfNodesPerLevel << " Nodes" << std::endl;
-
+
setNumberOfNodes(numberOfNodesPerLevel, level);
-
+
cudaMemoryManager->cudaAllocCoord(level);
cudaMemoryManager->cudaAllocSP(level);
//cudaMemoryManager->cudaAllocF3SP(level);
@@ -75,7 +77,7 @@ void GridGenerator::allocArrays_CoordNeighborGeo()
if(para->getUseTurbulentViscosity())
cudaMemoryManager->cudaAllocTurbulentViscosity(level);
-
+
if(para->getIsBodyForce())
cudaMemoryManager->cudaAllocBodyForce(level);
@@ -106,7 +108,7 @@ void GridGenerator::allocArrays_CoordNeighborGeo()
void GridGenerator::allocArrays_taggedFluidNodes() {
- for (uint level = 0; level < builder->getNumberOfGridLevels(); level++)
+ for (uint level = 0; level < builder->getNumberOfGridLevels(); level++)
{
for ( CollisionTemplate tag: all_CollisionTemplate )
{ //TODO: Need to add CollisionTemplate to GridBuilder to allow as argument and get rid of indivual get funtions for fluid node indices... and clean up this mess
@@ -155,16 +157,16 @@ void GridGenerator::allocArrays_taggedFluidNodes() {
}
}
VF_LOG_INFO("Number of tagged nodes on level {}:", level);
- VF_LOG_INFO("Default: {}, Border: {}, WriteMacroVars: {}, ApplyBodyForce: {}, AllFeatures: {}",
+ VF_LOG_INFO("Default: {}, Border: {}, WriteMacroVars: {}, ApplyBodyForce: {}, AllFeatures: {}",
para->getParH(level)->numberOfTaggedFluidNodes[CollisionTemplate::Default],
para->getParH(level)->numberOfTaggedFluidNodes[CollisionTemplate::SubDomainBorder],
para->getParH(level)->numberOfTaggedFluidNodes[CollisionTemplate::WriteMacroVars],
para->getParH(level)->numberOfTaggedFluidNodes[CollisionTemplate::ApplyBodyForce],
- para->getParH(level)->numberOfTaggedFluidNodes[CollisionTemplate::AllFeatures] );
+ para->getParH(level)->numberOfTaggedFluidNodes[CollisionTemplate::AllFeatures] );
}
}
-void GridGenerator::tagFluidNodeIndices(std::vector<uint> taggedFluidNodeIndices, CollisionTemplate tag, uint level) {
+void GridGenerator::tagFluidNodeIndices(const std::vector<uint>& taggedFluidNodeIndices, CollisionTemplate tag, uint level) {
switch(tag)
{
case CollisionTemplate::WriteMacroVars:
@@ -184,7 +186,7 @@ void GridGenerator::tagFluidNodeIndices(std::vector<uint> taggedFluidNodeIndices
break;
}
-
+
}
void GridGenerator::sortFluidNodeTags() {
@@ -201,7 +203,7 @@ void GridGenerator::sortFluidNodeTags() {
void GridGenerator::allocArrays_BoundaryValues()
{
std::cout << "------read BoundaryValues------" << std::endl;
- int blocks = 0;
+ int blocks;
for (uint level = 0; level < builder->getNumberOfGridLevels(); level++) {
const auto numberOfPressureValues = int(builder->getPressureSize(level));
@@ -229,12 +231,12 @@ void GridGenerator::allocArrays_BoundaryValues()
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
para->getParH(level)->slipBC.numberOfBCnodes = 0;
- if (numberOfSlipValues > 1)
- {
+ if (numberOfSlipValues > 1) {
blocks = (numberOfSlipValues / para->getParH(level)->numberofthreads) + 1;
para->getParH(level)->slipBC.numberOfBCnodes = blocks * para->getParH(level)->numberofthreads;
cudaMemoryManager->cudaAllocSlipBC(level);
- builder->getSlipValues(para->getParH(level)->slipBC.normalX, para->getParH(level)->slipBC.normalY, para->getParH(level)->slipBC.normalZ, para->getParH(level)->slipBC.k, level);
+ builder->getSlipValues(para->getParH(level)->slipBC.normalX, para->getParH(level)->slipBC.normalY,
+ para->getParH(level)->slipBC.normalZ, para->getParH(level)->slipBC.k, level);
cudaMemoryManager->cudaCopySlipBC(level);
}
para->getParD(level)->slipBC.numberOfBCnodes = para->getParH(level)->slipBC.numberOfBCnodes;
@@ -254,11 +256,11 @@ void GridGenerator::allocArrays_BoundaryValues()
para->getParH(level)->stressBC.numberOfBCnodes = blocks * para->getParH(level)->numberofthreads;
cudaMemoryManager->cudaAllocStressBC(level);
cudaMemoryManager->cudaAllocWallModel(level, para->getHasWallModelMonitor());
- builder->getStressValues( para->getParH(level)->stressBC.normalX, para->getParH(level)->stressBC.normalY, para->getParH(level)->stressBC.normalZ,
+ builder->getStressValues( para->getParH(level)->stressBC.normalX, para->getParH(level)->stressBC.normalY, para->getParH(level)->stressBC.normalZ,
para->getParH(level)->stressBC.Vx, para->getParH(level)->stressBC.Vy, para->getParH(level)->stressBC.Vz,
para->getParH(level)->stressBC.Vx1, para->getParH(level)->stressBC.Vy1, para->getParH(level)->stressBC.Vz1,
- para->getParH(level)->stressBC.k, para->getParH(level)->stressBC.kN,
- para->getParH(level)->wallModel.samplingOffset, para->getParH(level)->wallModel.z0,
+ para->getParH(level)->stressBC.k, para->getParH(level)->stressBC.kN,
+ para->getParH(level)->wallModel.samplingOffset, para->getParH(level)->wallModel.z0,
level);
cudaMemoryManager->cudaCopyStressBC(level);
@@ -268,7 +270,7 @@ void GridGenerator::allocArrays_BoundaryValues()
para->getParH(level)->numberOfStressBCnodesRead = para->getParH(level)->stressBC.numberOfBCnodes * para->getD3Qxx();
para->getParD(level)->numberOfStressBCnodesRead = para->getParH(level)->stressBC.numberOfBCnodes * para->getD3Qxx();
}
-
+
for (uint level = 0; level < builder->getNumberOfGridLevels(); level++) {
const auto numberOfVelocityValues = int(builder->getVelocitySize(level));
@@ -285,7 +287,8 @@ void GridGenerator::allocArrays_BoundaryValues()
cudaMemoryManager->cudaAllocVeloBC(level);
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
- builder->getVelocityValues(para->getParH(level)->velocityBC.Vx, para->getParH(level)->velocityBC.Vy, para->getParH(level)->velocityBC.Vz, para->getParH(level)->velocityBC.k, level);
+ builder->getVelocityValues(para->getParH(level)->velocityBC.Vx, para->getParH(level)->velocityBC.Vy,
+ para->getParH(level)->velocityBC.Vz, para->getParH(level)->velocityBC.k, level);
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
@@ -327,7 +330,7 @@ void GridGenerator::allocArrays_BoundaryValues()
const auto numberOfPrecursorValues = int(builder->getPrecursorSize(level));
*logging::out << logging::Logger::INFO_INTERMEDIATE << "size precursor level " << level << " : " << numberOfPrecursorValues << "\n";
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
- int blocks = (numberOfPrecursorValues / para->getParH(level)->numberofthreads) + 1;
+ blocks = (numberOfPrecursorValues / para->getParH(level)->numberofthreads) + 1;
para->getParH(level)->precursorBC.sizeQ = blocks * para->getParH(level)->numberofthreads;
para->getParD(level)->precursorBC.sizeQ = para->getParH(level)->precursorBC.sizeQ;
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
@@ -335,19 +338,19 @@ void GridGenerator::allocArrays_BoundaryValues()
para->getParD(level)->precursorBC.numberOfBCnodes = numberOfPrecursorValues;
para->getParH(level)->numberOfPrecursorBCnodesRead = numberOfPrecursorValues * para->getD3Qxx();
para->getParD(level)->numberOfPrecursorBCnodesRead = numberOfPrecursorValues * para->getD3Qxx();
-
+
if (numberOfPrecursorValues > 1)
{
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
cudaMemoryManager->cudaAllocPrecursorBC(level);
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
builder->getPrecursorValues(
- para->getParH(level)->precursorBC.planeNeighbor0PP, para->getParH(level)->precursorBC.planeNeighbor0PM,
- para->getParH(level)->precursorBC.planeNeighbor0MP, para->getParH(level)->precursorBC.planeNeighbor0MM,
- para->getParH(level)->precursorBC.weights0PP, para->getParH(level)->precursorBC.weights0PM,
- para->getParH(level)->precursorBC.weights0MP, para->getParH(level)->precursorBC.weights0MM,
- para->getParH(level)->precursorBC.k, para->getParH(level)->transientBCInputFileReader, para->getParH(level)->precursorBC.numberOfPrecursorNodes,
- para->getParH(level)->precursorBC.numberOfQuantities, para->getParH(level)->precursorBC.timeStepsBetweenReads,
+ para->getParH(level)->precursorBC.planeNeighbor0PP, para->getParH(level)->precursorBC.planeNeighbor0PM,
+ para->getParH(level)->precursorBC.planeNeighbor0MP, para->getParH(level)->precursorBC.planeNeighbor0MM,
+ para->getParH(level)->precursorBC.weights0PP, para->getParH(level)->precursorBC.weights0PM,
+ para->getParH(level)->precursorBC.weights0MP, para->getParH(level)->precursorBC.weights0MM,
+ para->getParH(level)->precursorBC.k, para->getParH(level)->transientBCInputFileReader, para->getParH(level)->precursorBC.numberOfPrecursorNodes,
+ para->getParH(level)->precursorBC.numberOfQuantities, para->getParH(level)->precursorBC.timeStepsBetweenReads,
para->getParH(level)->precursorBC.velocityX, para->getParH(level)->precursorBC.velocityY, para->getParH(level)->precursorBC.velocityZ,
level);
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
@@ -360,7 +363,9 @@ void GridGenerator::allocArrays_BoundaryValues()
for(auto reader : para->getParH(level)->transientBCInputFileReader)
{
- if(reader->getNumberOfQuantities() != para->getParD(level)->precursorBC.numberOfQuantities) throw std::runtime_error("Number of quantities in reader and number of quantities needed for precursor don't match!");
+ if(reader->getNumberOfQuantities() != para->getParD(level)->precursorBC.numberOfQuantities)
+ throw std::runtime_error(
+ "Number of quantities in reader and number of quantities needed for precursor don't match!");
}
cudaMemoryManager->cudaCopyPrecursorBC(level);
@@ -368,7 +373,7 @@ void GridGenerator::allocArrays_BoundaryValues()
// read first timestep of precursor into next and copy to next on device
for(auto reader : para->getParH(level)->transientBCInputFileReader)
- {
+ {
reader->getNextData(para->getParH(level)->precursorBC.next, para->getParH(level)->precursorBC.numberOfPrecursorNodes, 0);
}
@@ -382,7 +387,7 @@ void GridGenerator::allocArrays_BoundaryValues()
//read second timestep of precursor into next and copy next to device
real nextTime = para->getParD(level)->precursorBC.timeStepsBetweenReads*pow(2,-((real)level))*para->getTimeRatio();
for(auto reader : para->getParH(level)->transientBCInputFileReader)
- {
+ {
reader->getNextData(para->getParH(level)->precursorBC.next, para->getParH(level)->precursorBC.numberOfPrecursorNodes, nextTime);
}
@@ -398,7 +403,7 @@ void GridGenerator::allocArrays_BoundaryValues()
//start usual cycle of loading, i.e. read velocities of timestep after current and copy asynchronously to device
for(auto reader : para->getParH(level)->transientBCInputFileReader)
- {
+ {
reader->getNextData(para->getParH(level)->precursorBC.next, para->getParH(level)->precursorBC.numberOfPrecursorNodes, 2*nextTime);
}
@@ -476,7 +481,7 @@ void GridGenerator::initalValuesDomainDecompostion()
if (para->getNumprocs() < 2)
return;
if ((para->getNumprocs() > 1) /*&& (procNeighborsSendX.size() == procNeighborsRecvX.size())*/) {
-
+
// direction has to be changed in case of periodic BCs and multiple sub domains
std::vector<int> fillOrder = { 0, 1, 2, 3, 4, 5 };
@@ -556,7 +561,7 @@ void GridGenerator::initalValuesDomainDecompostion()
builder->getReceiveIndices(para->getParH(level)->recvProcessNeighborX[indexProcessNeighbor].index, direction,
level);
if (level != builder->getNumberOfGridLevels() - 1 && para->useReducedCommunicationAfterFtoC)
- indexRearrangement->initCommunicationArraysForCommAfterFinetoCoarseX(level, indexProcessNeighbor, direction);
+ indexRearrangement->initCommunicationArraysForCommAfterFinetoCoarseX(level, indexProcessNeighbor, direction);
////////////////////////////////////////////////////////////////////////////////////////
cudaMemoryManager->cudaCopyProcessNeighborXIndex(level, indexProcessNeighbor);
////////////////////////////////////////////////////////////////////////////////////////
@@ -619,7 +624,7 @@ void GridGenerator::initalValuesDomainDecompostion()
////////////////////////////////////////////////////////////////////////////////////////
// malloc on host and device
cudaMemoryManager->cudaAllocProcessNeighborY(level, indexProcessNeighbor);
- ////////////////////////////////////////////////////////////////////////////////////////
+ ////////////////////////////////////////////////////////////////////////////////////////
// init index arrays
builder->getSendIndices(para->getParH(level)->sendProcessNeighborY[indexProcessNeighbor].index, direction, level);
builder->getReceiveIndices(para->getParH(level)->recvProcessNeighborY[indexProcessNeighbor].index, direction,
@@ -638,7 +643,7 @@ void GridGenerator::initalValuesDomainDecompostion()
if (tempSend > 0) {
int indexProcessNeighbor = (int)para->getParH(level)->sendProcessNeighborZ.size();
-
+
para->getParH(level)->sendProcessNeighborZ.emplace_back();
para->getParD(level)->sendProcessNeighborZ.emplace_back();
para->getParH(level)->recvProcessNeighborZ.emplace_back();
@@ -930,7 +935,7 @@ void GridGenerator::allocArrays_BoundaryQs()
unsigned int sizeQ = para->getParH(i)->pressureBC.numberOfBCnodes;
QforBoundaryConditions Q;
getPointersToBoundaryConditions(Q, QQ, sizeQ);
-
+
builder->getPressureQs(Q.q27, i);
@@ -977,7 +982,7 @@ void GridGenerator::allocArrays_BoundaryQs()
unsigned int sizeQ = para->getParH(i)->slipBC.numberOfBCnodes;
QforBoundaryConditions Q;
getPointersToBoundaryConditions(Q, QQ, sizeQ);
-
+
builder->getSlipQs(Q.q27, i);
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
cudaMemoryManager->cudaCopySlipBC(i);
@@ -997,7 +1002,7 @@ void GridGenerator::allocArrays_BoundaryQs()
unsigned int sizeQ = para->getParH(i)->stressBC.numberOfBCnodes;
QforBoundaryConditions Q;
getPointersToBoundaryConditions(Q, QQ, sizeQ);
-
+
builder->getStressQs(Q.q27, i);
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
cudaMemoryManager->cudaCopyStressBC(i);
@@ -1165,7 +1170,7 @@ void GridGenerator::allocArrays_BoundaryQs()
void GridGenerator::allocArrays_OffsetScale()
{
- for (uint level = 0; level < builder->getNumberOfGridLevels() - 1; level++)
+ for (uint level = 0; level < builder->getNumberOfGridLevels() - 1; level++)
{
const uint numberOfNodesPerLevelCF = builder->getNumberOfNodesCF(level);
const uint numberOfNodesPerLevelFC = builder->getNumberOfNodesFC(level);
@@ -1204,7 +1209,7 @@ void GridGenerator::allocArrays_OffsetScale()
builder->getOffsetCF(para->getParH(level)->offCF.xOffCF, para->getParH(level)->offCF.yOffCF, para->getParH(level)->offCF.zOffCF, level);
builder->getOffsetFC(para->getParH(level)->offFC.xOffFC, para->getParH(level)->offFC.yOffFC, para->getParH(level)->offFC.zOffFC, level);
builder->getGridInterfaceIndices(para->getParH(level)->intCF.ICellCFC, para->getParH(level)->intCF.ICellCFF, para->getParH(level)->intFC.ICellFCC, para->getParH(level)->intFC.ICellFCF, level);
-
+
if (para->getUseStreams() || para->getNumprocs() > 1) {
// split fine-to-coarse indices into border and bulk
interpolationGrouper->splitFineToCoarseIntoBorderAndBulk(level);
@@ -1307,7 +1312,7 @@ std::string GridGenerator::verifyNeighborIndex(int level, int index , int &inval
//std::cout << para->getParH(level)->coordinateX[1] << ", " << para->getParH(level)->coordinateY[1] << ", " << para->getParH(level)->coordinateZ[1] << std::endl;
//std::cout << para->getParH(level)->coordinateX[para->getParH(level)->numberOfNodes - 1] << ", " << para->getParH(level)->coordinateY[para->getParH(level)->numberOfNodes - 1] << ", " << para->getParH(level)->coordinateZ[para->getParH(level)->numberOfNodes - 1] << std::endl;
-
+
real maxX = para->getParH(level)->coordinateX[para->getParH(level)->numberOfNodes - 1] - delta;
real maxY = para->getParH(level)->coordinateY[para->getParH(level)->numberOfNodes - 1] - delta;
real maxZ = para->getParH(level)->coordinateZ[para->getParH(level)->numberOfNodes - 1] - delta;
@@ -1348,8 +1353,8 @@ std::string GridGenerator::checkNeighbor(int level, real x, real y, real z, int
if (!neighborValid) {
oss << "NeighborX invalid from: (" << x << ", " << y << ", " << z << "), index: " << index << ", "
- << direction << " neighborIndex: " << neighborIndex <<
- ", actual neighborCoords : (" << neighborCoordX << ", " << neighborCoordY << ", " << neighborCoordZ <<
+ << direction << " neighborIndex: " << neighborIndex <<
+ ", actual neighborCoords : (" << neighborCoordX << ", " << neighborCoordY << ", " << neighborCoordZ <<
"), expected neighborCoords : (" << neighborX << ", " << neighborY << ", " << neighborZ << ")\n";
numberOfWrongNeihgbors++;
}
@@ -1384,4 +1389,4 @@ void GridGenerator::getPointersToBoundaryConditions(QforBoundaryConditions& boun
boundaryConditionStruct.q27[DIR_MMM] = &subgridDistances[DIR_MMM * numberOfBCnodes];
boundaryConditionStruct.q27[DIR_PMM] = &subgridDistances[DIR_PMM * numberOfBCnodes];
boundaryConditionStruct.q27[DIR_MPM] = &subgridDistances[DIR_MPM * numberOfBCnodes];
-}
\ No newline at end of file
+}
diff --git a/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderGenerator/GridGenerator.h b/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderGenerator/GridGenerator.h
index cee661a0c1092e7a3d334018133fd2f963ada991..c97ed02a64da1d5fafa18150c75d149f96484d44 100644
--- a/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderGenerator/GridGenerator.h
+++ b/src/gpu/VirtualFluids_GPU/DataStructureInitializer/GridReaderGenerator/GridGenerator.h
@@ -78,7 +78,7 @@ public:
void allocArrays_OffsetScale() override;
void allocArrays_taggedFluidNodes() override;
- void tagFluidNodeIndices(std::vector<uint> taggedFluidNodeIndices, CollisionTemplate tag, uint level) override;
+ void tagFluidNodeIndices(const std::vector<uint>& taggedFluidNodeIndices, CollisionTemplate tag, uint level) override;
void sortFluidNodeTags() override;
virtual void setDimensions() override;
diff --git a/src/gpu/VirtualFluids_GPU/GPU/GPU_Kernels.cuh b/src/gpu/VirtualFluids_GPU/GPU/GPU_Kernels.cuh
index a7b22c244da368b9591c5f528bb7bd5eaea91e50..567a4112e935e0a154288ec7beb1fc82d01c9d7a 100644
--- a/src/gpu/VirtualFluids_GPU/GPU/GPU_Kernels.cuh
+++ b/src/gpu/VirtualFluids_GPU/GPU/GPU_Kernels.cuh
@@ -53,130 +53,130 @@ __global__ void LB_Kernel_Casc_SP_MS_OHM_27( real s9,
bool EvenOrOdd);
__global__ void LB_Kernel_Kum_New_Comp_SRT_SP_27(
- real omega,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- int level,
- real* forces,
- bool EvenOrOdd);
+ real omega,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ int level,
+ real* forces,
+ bool EvenOrOdd);
__global__ void LB_Kernel_Cumulant_D3Q27All4(real omega,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- int level,
- real* forces,
- bool EvenOrOdd);
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ int level,
+ real* forces,
+ bool EvenOrOdd);
__global__ void LB_Kernel_Kum_AA2016_Comp_Bulk_SP_27(real omega,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- int level,
- real* forces,
- bool EvenOrOdd);
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ int level,
+ real* forces,
+ bool EvenOrOdd);
__global__ void LB_Kernel_Kum_1h_SP_27( real omega,
- real deltaPhi,
- real angularVelocity,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* coordX,
- real* coordY,
- real* coordZ,
- real* DDStart,
- int size_Mat,
- bool EvenOrOdd);
+ real deltaPhi,
+ real angularVelocity,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* coordX,
+ real* coordY,
+ real* coordZ,
+ real* DDStart,
+ int size_Mat,
+ bool EvenOrOdd);
__global__ void LB_Kernel_Cascade_SP_27( real s9,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- bool EvenOrOdd);
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ bool EvenOrOdd);
__global__ void LB_Kernel_Kum_New_SP_27( real s9,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- bool EvenOrOdd);
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ bool EvenOrOdd);
__global__ void LB_Kernel_Kum_IsoTest_SP_27( real omega,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- real* dxxUx,
- real* dyyUy,
- real* dzzUz,
- int size_Mat,
- bool EvenOrOdd);
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ real* dxxUx,
+ real* dyyUy,
+ real* dzzUz,
+ int size_Mat,
+ bool EvenOrOdd);
__global__ void LB_Kernel_Kum_Comp_SP_27(real s9,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- bool EvenOrOdd);
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ bool EvenOrOdd);
__global__ void Cumulant_One_preconditioned_errorDiffusion_chim_Comp_SP_27(
- real omega,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- int level,
- real* forces,
- bool EvenOrOdd);
+ real omega,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ int level,
+ real* forces,
+ bool EvenOrOdd);
__global__ void Cumulant_One_preconditioned_chim_Comp_SP_27(
- real omega,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- int level,
- real* forces,
- bool EvenOrOdd);
+ real omega,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ int level,
+ real* forces,
+ bool EvenOrOdd);
__global__ void Cumulant_One_chim_Comp_SP_27(
- real omega,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- int level,
- real* forces,
- bool EvenOrOdd);
+ real omega,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ int level,
+ real* forces,
+ bool EvenOrOdd);
inline __device__ void forwardChimeraWithK(real &mfa, real &mfb, real &mfc, real vv, real v2, real K);
@@ -189,57 +189,57 @@ inline __device__ void backwardChimeraWithK(real &mfa, real &mfb, real &mfc, rea
__global__ void LB_Kernel_WaleBySoniMalav_Cum_AA2016_Comp_SP_27(
- real omega_in,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int* neighborWSB,
- real* veloX,
- real* veloY,
- real* veloZ,
- real* DDStart,
- real* turbulentViscosity,
- int size_Mat,
- int level,
- real* forces,
- bool EvenOrOdd);
+ real omega_in,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int* neighborWSB,
+ real* veloX,
+ real* veloY,
+ real* veloZ,
+ real* DDStart,
+ real* turbulentViscosity,
+ int size_Mat,
+ int level,
+ real* forces,
+ bool EvenOrOdd);
__global__ void LB_Kernel_PM_Cum_One_Comp_SP_27( real omega,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- int level,
- real* forces,
- real porosity,
- real darcy,
- real forchheimer,
- unsigned int sizeOfPorousMedia,
- unsigned int* nodeIdsPorousMedia,
- bool EvenOrOdd);
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ int level,
+ real* forces,
+ real porosity,
+ real darcy,
+ real forchheimer,
+ unsigned int sizeOfPorousMedia,
+ unsigned int* nodeIdsPorousMedia,
+ bool EvenOrOdd);
__global__ void LB_Kernel_AD_Incomp_7( real diffusivity,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- real* DD7,
- int size_Mat,
- bool EvenOrOdd);
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ real* DD7,
+ int size_Mat,
+ bool EvenOrOdd);
__global__ void LB_Kernel_AD_Incomp_27( real diffusivity,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- real* DD27,
- int size_Mat,
- bool EvenOrOdd);
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ real* DD27,
+ int size_Mat,
+ bool EvenOrOdd);
__global__ void LBInit27( int myid,
int numprocs,
@@ -296,21 +296,21 @@ __global__ void InitAD27(unsigned int* neighborX,
bool EvenOrOdd);
__global__ void LB_PostProcessor_F3_2018_Fehlberg(
- real omega,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* rhoOut,
- real* vxOut,
- real* vyOut,
- real* vzOut,
- real* DDStart,
- real* G6,
- int size_Mat,
- int level,
- real* forces,
- bool EvenOrOdd);
+ real omega,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* rhoOut,
+ real* vxOut,
+ real* vyOut,
+ real* vzOut,
+ real* DDStart,
+ real* G6,
+ int size_Mat,
+ int level,
+ real* forces,
+ bool EvenOrOdd);
__global__ void LBCalcMac27( real* vxD,
real* vyD,
@@ -338,17 +338,17 @@ __global__ void LBCalcMacSP27( real* vxD,
bool isEvenTimestep);
__global__ void LBCalcMacCompSP27( real* vxD,
- real* vyD,
- real* vzD,
- real* rhoD,
- real* pressD,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- real* DD,
- bool isEvenTimestep);
+ real* vyD,
+ real* vzD,
+ real* rhoD,
+ real* pressD,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ real* DD,
+ bool isEvenTimestep);
__global__ void CalcConc7( real* Conc,
unsigned int* geoD,
@@ -360,26 +360,26 @@ __global__ void CalcConc7( real* Conc,
bool isEvenTimestep);
__global__ void GetPlaneConc7(real* Conc,
- int* kPC,
- unsigned int numberOfPointskPC,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- real* DD7,
- bool isEvenTimestep);
+ int* kPC,
+ unsigned int numberOfPointskPC,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ real* DD7,
+ bool isEvenTimestep);
__global__ void GetPlaneConc27(real* Conc,
- int* kPC,
- unsigned int numberOfPointskPC,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- real* DD27,
- bool isEvenTimestep);
+ int* kPC,
+ unsigned int numberOfPointskPC,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ real* DD27,
+ bool isEvenTimestep);
__global__ void CalcConc27(real* Conc,
unsigned int* geoD,
@@ -404,33 +404,33 @@ __global__ void LBCalcMedSP27( real* vxD,
bool isEvenTimestep);
__global__ void LBCalcMedCompSP27( real* vxD,
- real* vyD,
- real* vzD,
- real* rhoD,
- real* pressD,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- real* DD,
- bool isEvenTimestep);
+ real* vyD,
+ real* vzD,
+ real* rhoD,
+ real* pressD,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ real* DD,
+ bool isEvenTimestep);
__global__ void LBCalcMedCompAD27(
- real* vxD,
- real* vyD,
- real* vzD,
- real* rhoD,
- real* pressD,
- real* concD,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- real* DD,
- real* DD_AD,
- bool isEvenTimestep);
+ real* vxD,
+ real* vyD,
+ real* vzD,
+ real* rhoD,
+ real* pressD,
+ real* concD,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ real* DD,
+ real* DD_AD,
+ bool isEvenTimestep);
__global__ void LBCalcMacMedSP27( real* vxD,
real* vyD,
@@ -446,115 +446,115 @@ __global__ void LBCalcMacMedSP27( real* vxD,
bool isEvenTimestep);
__global__ void LBResetMedianValuesSP27(
- real* vxD,
- real* vyD,
- real* vzD,
- real* rhoD,
- real* pressD,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* vxD,
+ real* vyD,
+ real* vzD,
+ real* rhoD,
+ real* pressD,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void LBResetMedianValuesAD27(
- real* vxD,
- real* vyD,
- real* vzD,
- real* rhoD,
- real* pressD,
- real* concD,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* vxD,
+ real* vyD,
+ real* vzD,
+ real* rhoD,
+ real* pressD,
+ real* concD,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void LBCalc2ndMomentsIncompSP27( real* kxyFromfcNEQ,
- real* kyzFromfcNEQ,
- real* kxzFromfcNEQ,
- real* kxxMyyFromfcNEQ,
- real* kxxMzzFromfcNEQ,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- real* DD,
- bool isEvenTimestep);
+ real* kyzFromfcNEQ,
+ real* kxzFromfcNEQ,
+ real* kxxMyyFromfcNEQ,
+ real* kxxMzzFromfcNEQ,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ real* DD,
+ bool isEvenTimestep);
__global__ void LBCalc2ndMomentsCompSP27(real* kxyFromfcNEQ,
- real* kyzFromfcNEQ,
- real* kxzFromfcNEQ,
- real* kxxMyyFromfcNEQ,
- real* kxxMzzFromfcNEQ,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- real* DD,
- bool isEvenTimestep);
+ real* kyzFromfcNEQ,
+ real* kxzFromfcNEQ,
+ real* kxxMyyFromfcNEQ,
+ real* kxxMzzFromfcNEQ,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ real* DD,
+ bool isEvenTimestep);
__global__ void LBCalc3rdMomentsIncompSP27( real* CUMbbb,
- real* CUMabc,
- real* CUMbac,
- real* CUMbca,
- real* CUMcba,
- real* CUMacb,
- real* CUMcab,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- bool EvenOrOdd);
+ real* CUMabc,
+ real* CUMbac,
+ real* CUMbca,
+ real* CUMcba,
+ real* CUMacb,
+ real* CUMcab,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ bool EvenOrOdd);
__global__ void LBCalc3rdMomentsCompSP27(real* CUMbbb,
- real* CUMabc,
- real* CUMbac,
- real* CUMbca,
- real* CUMcba,
- real* CUMacb,
- real* CUMcab,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- bool EvenOrOdd);
+ real* CUMabc,
+ real* CUMbac,
+ real* CUMbca,
+ real* CUMcba,
+ real* CUMacb,
+ real* CUMcab,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ bool EvenOrOdd);
__global__ void LBCalcHigherMomentsIncompSP27( real* CUMcbb,
- real* CUMbcb,
- real* CUMbbc,
- real* CUMcca,
- real* CUMcac,
- real* CUMacc,
- real* CUMbcc,
- real* CUMcbc,
- real* CUMccb,
- real* CUMccc,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- bool EvenOrOdd);
+ real* CUMbcb,
+ real* CUMbbc,
+ real* CUMcca,
+ real* CUMcac,
+ real* CUMacc,
+ real* CUMbcc,
+ real* CUMcbc,
+ real* CUMccb,
+ real* CUMccc,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ bool EvenOrOdd);
__global__ void LBCalcHigherMomentsCompSP27( real* CUMcbb,
- real* CUMbcb,
- real* CUMbbc,
- real* CUMcca,
- real* CUMcac,
- real* CUMacc,
- real* CUMbcc,
- real* CUMcbc,
- real* CUMccb,
- real* CUMccc,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- bool EvenOrOdd);
+ real* CUMbcb,
+ real* CUMbbc,
+ real* CUMcca,
+ real* CUMcac,
+ real* CUMacc,
+ real* CUMbcc,
+ real* CUMcbc,
+ real* CUMccb,
+ real* CUMccc,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ bool EvenOrOdd);
__global__ void LBCalcMeasurePoints(real* vxMP,
real* vyMP,
@@ -612,60 +612,60 @@ __global__ void QDevice27(real* distributions,
bool isEvenTimestep);
__global__ void QDeviceComp27(
- real* distributions,
- int* subgridDistanceIndices,
- real* subgridDistances,
- unsigned int numberOfBCnodes,
- real omega,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int numberOfLBnodes,
- bool isEvenTimestep);
+ real* distributions,
+ int* subgridDistanceIndices,
+ real* subgridDistances,
+ unsigned int numberOfBCnodes,
+ real omega,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int numberOfLBnodes,
+ bool isEvenTimestep);
__global__ void QDeviceCompThinWallsPartOne27(real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
-
-__global__ void QDevice3rdMomentsComp27( real* distributions,
- int* subgridDistanceIndices,
- real* subgridDistances,
- unsigned int numberOfBCnodes,
- real omega,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int numberOfLBnodes,
- bool isEvenTimestep);
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
+
+__global__ void QDevice3rdMomentsComp27( real* distributions,
+ int* subgridDistanceIndices,
+ real* subgridDistances,
+ unsigned int numberOfBCnodes,
+ real omega,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int numberOfLBnodes,
+ bool isEvenTimestep);
__global__ void QDeviceIncompHighNu27(real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int numberOfLBnodes,
- bool isEvenTimestep);
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int numberOfLBnodes,
+ bool isEvenTimestep);
__global__ void QDeviceCompHighNu27( real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
//Velocity BCs
__global__ void QVelDevPlainBB27(
@@ -683,39 +683,39 @@ __global__ void QVelDevPlainBB27(
bool isEvenTimestep);
__global__ void QVelDevCouette27(real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QVelDev1h27( int inx,
- int iny,
- real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- real Phi,
- real angularVelocity,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* coordX,
- real* coordY,
- real* coordZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ int iny,
+ real* vx,
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ real Phi,
+ real angularVelocity,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* coordX,
+ real* coordY,
+ real* coordZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QVelDevice27(int inx,
int iny,
@@ -734,107 +734,107 @@ __global__ void QVelDevice27(int inx,
bool isEvenTimestep);
__global__ void QVelDeviceCompPlusSlip27(real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QVelDeviceComp27(real* velocityX,
- real* velocityY,
- real* velocityZ,
- real* distribution,
- int* subgridDistanceIndices,
- real* subgridDistances,
- unsigned int numberOfBCnodes,
- real omega,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int numberOfLBnodes,
- bool isEvenTimestep);
+ real* velocityY,
+ real* velocityZ,
+ real* distribution,
+ int* subgridDistanceIndices,
+ real* subgridDistances,
+ unsigned int numberOfBCnodes,
+ real omega,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int numberOfLBnodes,
+ bool isEvenTimestep);
__global__ void QVelDeviceCompThinWallsPartOne27(
- real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- uint numberOfBCnodes,
- real om1,
- uint* neighborX,
- uint* neighborY,
- uint* neighborZ,
- uint size_Mat,
- bool isEvenTimestep);
+ real* vx,
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ uint numberOfBCnodes,
+ real om1,
+ uint* neighborX,
+ uint* neighborY,
+ uint* neighborZ,
+ uint size_Mat,
+ bool isEvenTimestep);
__global__ void QThinWallsPartTwo27(
- real* DD,
- int* k_Q,
- real* QQ,
- uint numberOfBCnodes,
- uint* geom,
- uint* neighborX,
- uint* neighborY,
- uint* neighborZ,
- uint* neighborWSB,
- uint size_Mat,
- bool isEvenTimestep);
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ uint numberOfBCnodes,
+ uint* geom,
+ uint* neighborX,
+ uint* neighborY,
+ uint* neighborZ,
+ uint* neighborWSB,
+ uint size_Mat,
+ bool isEvenTimestep);
__global__ void QVelDeviceCompZeroPress27(
- real* velocityX,
- real* velocityY,
- real* velocityZ,
- real* distribution,
- int* subgridDistanceIndices,
- real* subgridDistances,
- unsigned int numberOfBCnodes,
- real omega,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int numberOfLBnodes,
- bool isEvenTimestep);
+ real* velocityX,
+ real* velocityY,
+ real* velocityZ,
+ real* distribution,
+ int* subgridDistanceIndices,
+ real* subgridDistances,
+ unsigned int numberOfBCnodes,
+ real omega,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int numberOfLBnodes,
+ bool isEvenTimestep);
__global__ void QVelDeviceIncompHighNu27(real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QVelDeviceCompHighNu27( real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QVeloDeviceEQ27(real* VeloX,
- real* VeloY,
- real* VeloZ,
+ real* VeloY,
+ real* VeloZ,
real* DD,
int* k_Q,
int numberOfBCnodes,
@@ -846,18 +846,18 @@ __global__ void QVeloDeviceEQ27(real* VeloX,
bool isEvenTimestep);
__global__ void QVeloStreetDeviceEQ27(
- real* veloXfraction,
- real* veloYfraction,
- int* naschVelo,
- real* DD,
- int* naschIndex,
- int numberOfStreetNodes,
- real velocityRatio,
- uint* neighborX,
- uint* neighborY,
- uint* neighborZ,
- uint size_Mat,
- bool isEvenTimestep);
+ real* veloXfraction,
+ real* veloYfraction,
+ int* naschVelo,
+ real* DD,
+ int* naschIndex,
+ int numberOfStreetNodes,
+ real velocityRatio,
+ uint* neighborX,
+ uint* neighborY,
+ uint* neighborZ,
+ uint size_Mat,
+ bool isEvenTimestep);
//Slip BCs
__global__ void QSlipDevice27(real* DD,
@@ -872,135 +872,135 @@ __global__ void QSlipDevice27(real* DD,
bool isEvenTimestep);
__global__ void QSlipDeviceComp27(real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QSlipDeviceComp27TurbViscosity(
- real* distributions,
- int* subgridDistanceIndices,
+ real* distributions,
+ int* subgridDistanceIndices,
real* subgridDistances,
unsigned int numberOfBCnodes,
- real omega,
+ real omega,
unsigned int* neighborX,
unsigned int* neighborY,
unsigned int* neighborZ,
real* turbViscosity,
- unsigned int numberOfLBnodes,
+ unsigned int numberOfLBnodes,
bool isEvenTimestep);
__global__ void QSlipPressureDeviceComp27TurbViscosity(
- real* distributions,
- int* subgridDistanceIndices,
+ real* distributions,
+ int* subgridDistanceIndices,
real* subgridDistances,
unsigned int numberOfBCnodes,
- real omega,
+ real omega,
unsigned int* neighborX,
unsigned int* neighborY,
unsigned int* neighborZ,
real* turbViscosity,
- unsigned int numberOfLBnodes,
+ unsigned int numberOfLBnodes,
bool isEvenTimestep);
__global__ void QSlipGeomDeviceComp27(real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- real* NormalX,
- real* NormalY,
- real* NormalZ,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ real* NormalX,
+ real* NormalY,
+ real* NormalZ,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QSlipNormDeviceComp27(real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- real* NormalX,
- real* NormalY,
- real* NormalZ,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ real* NormalX,
+ real* NormalY,
+ real* NormalZ,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
// Stress BCs (wall model)
__global__ void QStressDeviceComp27(real* DD,
- int* k_Q,
- int* k_N,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- real* turbViscosity,
- real* vx,
- real* vy,
- real* vz,
- real* normalX,
- real* normalY,
- real* normalZ,
- real* vx_bc,
- real* vy_bc,
- real* vz_bc,
- real* vx1,
- real* vy1,
- real* vz1,
- int* samplingOffset,
- real* z0,
- bool hasWallModelMonitor,
- real* u_star_monitor,
- real* Fx_monitor,
- real* Fy_monitor,
- real* Fz_monitor,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ int* k_Q,
+ int* k_N,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ real* turbViscosity,
+ real* vx,
+ real* vy,
+ real* vz,
+ real* normalX,
+ real* normalY,
+ real* normalZ,
+ real* vx_bc,
+ real* vy_bc,
+ real* vz_bc,
+ real* vx1,
+ real* vy1,
+ real* vz1,
+ int* samplingOffset,
+ real* z0,
+ bool hasWallModelMonitor,
+ real* u_star_monitor,
+ real* Fx_monitor,
+ real* Fy_monitor,
+ real* Fz_monitor,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void BBStressDevice27( real* DD,
- int* k_Q,
- int* k_N,
- real* QQ,
- unsigned int numberOfBCnodes,
- real* vx,
- real* vy,
- real* vz,
- real* normalX,
- real* normalY,
- real* normalZ,
- real* vx_bc,
- real* vy_bc,
- real* vz_bc,
- real* vx1,
- real* vy1,
- real* vz1,
- int* samplingOffset,
- real* z0,
- bool hasWallModelMonitor,
- real* u_star_monitor,
- real* Fx_monitor,
- real* Fy_monitor,
- real* Fz_monitor,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ int* k_Q,
+ int* k_N,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real* vx,
+ real* vy,
+ real* vz,
+ real* normalX,
+ real* normalY,
+ real* normalZ,
+ real* vx_bc,
+ real* vy_bc,
+ real* vz_bc,
+ real* vx1,
+ real* vy1,
+ real* vz1,
+ int* samplingOffset,
+ real* z0,
+ bool hasWallModelMonitor,
+ real* u_star_monitor,
+ real* Fx_monitor,
+ real* Fy_monitor,
+ real* Fz_monitor,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void BBStressPressureDevice27( real* DD,
- int* k_Q,
+ int* k_Q,
int* k_N,
real* QQ,
unsigned int numberOfBCnodes,
@@ -1043,19 +1043,19 @@ __global__ void QPressDevice27( real* rhoBC,
bool isEvenTimestep);
__global__ void QPressDeviceAntiBB27( real* rhoBC,
- real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* vx,
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QPressDeviceFixBackflow27( real* rhoBC,
real* DD,
@@ -1080,43 +1080,43 @@ __global__ void QPressDeviceDirDepBot27( real* rhoBC,
bool isEvenTimestep);
__global__ void QPressNoRhoDevice27( real* rhoBC,
- real* distributions,
- int* k_Q,
- int* k_N,
- int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int numberOfLBnodes,
- bool isEvenTimestep,
- int direction);
+ real* distributions,
+ int* k_Q,
+ int* k_N,
+ int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int numberOfLBnodes,
+ bool isEvenTimestep,
+ int direction);
__global__ void QPressZeroRhoOutflowDevice27( real* rhoBC,
- real* distributions,
- int* k_Q,
- int* k_N,
- int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int numberOfLBnodes,
- bool isEvenTimestep,
- int direction,
- real densityCorrectionFactor);
+ real* distributions,
+ int* k_Q,
+ int* k_N,
+ int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int numberOfLBnodes,
+ bool isEvenTimestep,
+ int direction,
+ real densityCorrectionFactor);
__global__ void QInflowScaleByPressDevice27( real* rhoBC,
- real* DD,
- int* k_Q,
- int* k_N,
- int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* DD,
+ int* k_Q,
+ int* k_N,
+ int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QPressDeviceOld27(real* rhoBC,
real* DD,
@@ -1131,16 +1131,16 @@ __global__ void QPressDeviceOld27(real* rhoBC,
bool isEvenTimestep);
__global__ void QPressDeviceIncompNEQ27( real* rhoBC,
- real* DD,
- int* k_Q,
- int* k_N,
- int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* DD,
+ int* k_Q,
+ int* k_N,
+ int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QPressDeviceNEQ27(real* rhoBC,
real* distribution,
@@ -1168,13 +1168,13 @@ __global__ void QPressDeviceEQZ27(real* rhoBC,
bool isEvenTimestep);
__global__ void QPressDeviceZero27( real* DD,
- int* k_Q,
- unsigned int numberOfBCnodes,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ int* k_Q,
+ unsigned int numberOfBCnodes,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QPressDeviceFake27(real* rhoBC,
real* DD,
@@ -1199,16 +1199,16 @@ __global__ void BBDevice27(real* distributions,
bool isEvenTimestep);
__global__ void QPressDevice27_IntBB(real* rho,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
// TODO: https://git.rz.tu-bs.de/irmb/VirtualFluids_dev/-/issues/29
//Schlaffer BCs
@@ -1250,18 +1250,18 @@ __global__ void QPrecursorDeviceCompZeroPress( int* subgridDistanceIndices,
real omega,
real* distributions,
real* subgridDistances,
- uint* neighborX,
- uint* neighborY,
+ uint* neighborX,
+ uint* neighborY,
uint* neighborZ,
- uint* neighborsNT,
+ uint* neighborsNT,
uint* neighborsNB,
uint* neighborsST,
uint* neighborsSB,
- real* weights0PP,
+ real* weights0PP,
real* weights0PM,
real* weights0MP,
real* weights0MM,
- real* vLast,
+ real* vLast,
real* vCurrent,
real velocityX,
real velocityY,
@@ -1276,18 +1276,18 @@ __global__ void PrecursorDeviceEQ27( int* subgridDistanceIndices,
int numberOfPrecursorNodes,
real omega,
real* distributions,
- uint* neighborX,
- uint* neighborY,
+ uint* neighborX,
+ uint* neighborY,
uint* neighborZ,
- uint* neighborsNT,
+ uint* neighborsNT,
uint* neighborsNB,
uint* neighborsST,
uint* neighborsSB,
- real* weights0PP,
+ real* weights0PP,
real* weights0PM,
real* weights0MP,
real* weights0MM,
- real* vLast,
+ real* vLast,
real* vCurrent,
real velocityX,
real velocityY,
@@ -1298,48 +1298,48 @@ __global__ void PrecursorDeviceEQ27( int* subgridDistanceIndices,
bool isEvenTimestep);
__global__ void PrecursorDeviceDistributions( int* subgridDistanceIndices,
- int numberOfBCNodes,
- int numberOfPrecursorNodes,
- real* distributions,
- uint* neighborX,
- uint* neighborY,
- uint* neighborZ,
- uint* neighborsNT,
- uint* neighborsNB,
- uint* neighborsST,
- uint* neighborsSB,
- real* weights0PP,
- real* weights0PM,
- real* weights0MP,
- real* weights0MM,
- real* fsLast,
- real* fsNext,
- real timeRatio,
- unsigned long long numberOfLBnodes,
- bool isEvenTimestep);
+ int numberOfBCNodes,
+ int numberOfPrecursorNodes,
+ real* distributions,
+ uint* neighborX,
+ uint* neighborY,
+ uint* neighborZ,
+ uint* neighborsNT,
+ uint* neighborsNB,
+ uint* neighborsST,
+ uint* neighborsSB,
+ real* weights0PP,
+ real* weights0PM,
+ real* weights0MP,
+ real* weights0MM,
+ real* fsLast,
+ real* fsNext,
+ real timeRatio,
+ unsigned long long numberOfLBnodes,
+ bool isEvenTimestep);
__global__ void QPrecursorDeviceDistributions( int* subgridDistanceIndices,
- real* subgridDistances,
- int sizeQ,
- int numberOfBCNodes,
- int numberOfPrecursorNodes,
- real* distributions,
- uint* neighborX,
- uint* neighborY,
- uint* neighborZ,
- uint* neighborsNT,
- uint* neighborsNB,
- uint* neighborsST,
- uint* neighborsSB,
- real* weights0PP,
- real* weights0PM,
- real* weights0MP,
- real* weights0MM,
- real* fsLast,
- real* fsNext,
- real timeRatio,
- unsigned long long numberOfLBnodes,
- bool isEvenTimestep);
-
+ real* subgridDistances,
+ int sizeQ,
+ int numberOfBCNodes,
+ int numberOfPrecursorNodes,
+ real* distributions,
+ uint* neighborX,
+ uint* neighborY,
+ uint* neighborZ,
+ uint* neighborsNT,
+ uint* neighborsNB,
+ uint* neighborsST,
+ uint* neighborsSB,
+ real* weights0PP,
+ real* weights0PM,
+ real* weights0MP,
+ real* weights0MM,
+ real* fsLast,
+ real* fsNext,
+ real timeRatio,
+ unsigned long long numberOfLBnodes,
+ bool isEvenTimestep);
+
//Advection / Diffusion BCs
__global__ void QAD7( real* DD,
real* DD7,
@@ -1358,62 +1358,62 @@ __global__ void QAD7( real* DD,
//////////////////////////////////////////////////////////////////////////
//! \brief \ref Advection_Diffusion_Device_Kernel : Factorized central moments for Advection Diffusion Equation
__global__ void Factorized_Central_Moments_Advection_Diffusion_Device_Kernel(
- real omegaDiffusivity,
- uint* typeOfGridNode,
- uint* neighborX,
- uint* neighborY,
- uint* neighborZ,
- real* distributions,
- real* distributionsAD,
- int size_Mat,
- real* forces,
- bool isEvenTimestep);
+ real omegaDiffusivity,
+ uint* typeOfGridNode,
+ uint* neighborX,
+ uint* neighborY,
+ uint* neighborZ,
+ real* distributions,
+ real* distributionsAD,
+ int size_Mat,
+ real* forces,
+ bool isEvenTimestep);
//////////////////////////////////////////////////////////////////////////
//! \brief \ref AD_SlipVelDeviceComp : device function for the slip-AD boundary condition
__global__ void AD_SlipVelDeviceComp(
- real * normalX,
- real * normalY,
- real * normalZ,
- real * distributions,
- real * distributionsAD,
- int* QindexArray,
- real * Qarrays,
- uint numberOfBCnodes,
- real omegaDiffusivity,
- uint * neighborX,
- uint * neighborY,
- uint * neighborZ,
- uint size_Mat,
- bool isEvenTimestep);
+ real * normalX,
+ real * normalY,
+ real * normalZ,
+ real * distributions,
+ real * distributionsAD,
+ int* QindexArray,
+ real * Qarrays,
+ uint numberOfBCnodes,
+ real omegaDiffusivity,
+ uint * neighborX,
+ uint * neighborY,
+ uint * neighborZ,
+ uint size_Mat,
+ bool isEvenTimestep);
__global__ void QADDirichlet27( real* DD,
- real* DD27,
- real* temp,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* DD27,
+ real* temp,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QADBB27( real* DD,
- real* DD27,
- real* temp,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* DD27,
+ real* temp,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QADVel7( real* DD,
real* DD7,
@@ -1476,105 +1476,105 @@ __global__ void QADPress27( real* DD,
bool isEvenTimestep);
__global__ void QADPressNEQNeighbor27(
- real* DD,
- real* DD27,
- int* k_Q,
- int* k_N,
- int numberOfBCnodes,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep
- );
+ real* DD,
+ real* DD27,
+ int* k_Q,
+ int* k_N,
+ int numberOfBCnodes,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep
+ );
__global__ void QNoSlipADincomp7( real* DD,
- real* DD7,
- real* temp,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* DD7,
+ real* temp,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QNoSlipADincomp27( real* DD,
- real* DD27,
- real* temp,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* DD27,
+ real* temp,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QADVeloIncomp7( real* DD,
- real* DD7,
- real* temp,
- real* velo,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* DD7,
+ real* temp,
+ real* velo,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QADVeloIncomp27( real* DD,
- real* DD27,
- real* temp,
- real* velo,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* DD27,
+ real* temp,
+ real* velo,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QADPressIncomp7(real* DD,
- real* DD7,
- real* temp,
- real* velo,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* DD7,
+ real* temp,
+ real* velo,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void QADPressIncomp27( real* DD,
- real* DD27,
- real* temp,
- real* velo,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* DD27,
+ real* temp,
+ real* velo,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
//Propeller BC
__global__ void PropellerBC(unsigned int* neighborX,
@@ -1585,7 +1585,7 @@ __global__ void PropellerBC(unsigned int* neighborX,
real* uy,
real* uz,
int* k_Q,
- unsigned int size_Prop,
+ unsigned int size_Prop,
unsigned int size_Mat,
unsigned int* bcMatD,
real* DD,
@@ -1602,19 +1602,19 @@ __global__ void scaleCF27(real* DC,
unsigned int* neighborFX,
unsigned int* neighborFY,
unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
unsigned int* posCSWB,
unsigned int* posFSWB,
unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF);
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF);
__global__ void scaleCFEff27(real* DC,
real* DF,
@@ -1624,18 +1624,18 @@ __global__ void scaleCFEff27(real* DC,
unsigned int* neighborFX,
unsigned int* neighborFY,
unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
unsigned int* posCSWB,
unsigned int* posFSWB,
unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
unsigned int nyF,
OffCF offCF);
@@ -1709,233 +1709,233 @@ __global__ void scaleCF_Fix_27(real* DC,
OffCF offCF);
__global__ void scaleCF_Fix_comp_27( real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffCF offCF);
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffCF offCF);
__global__ void scaleCF_0817_comp_27( real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffCF offCF);
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffCF offCF);
__global__ void scaleCF_comp_D3Q27F3_2018( real* DC,
- real* DF,
- real* G6,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffCF offCF);
+ real* DF,
+ real* G6,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffCF offCF);
__global__ void scaleCF_comp_D3Q27F3( real* DC,
- real* DF,
- real* G6,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffCF offCF);
+ real* DF,
+ real* G6,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffCF offCF);
__global__ void scaleCF_staggered_time_comp_27(real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffCF offCF);
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffCF offCF);
__global__ void scaleCF_RhoSq_comp_27( real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffCF offCF);
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffCF offCF);
__global__ void scaleCF_compressible(
- real* distributionsCoarse,
- real* distributionsFine,
+ real* distributionsCoarse,
+ real* distributionsFine,
unsigned int* neighborXcoarse,
unsigned int* neighborYcoarse,
unsigned int* neighborZcoarse,
unsigned int* neighborXfine,
unsigned int* neighborYfine,
unsigned int* neighborZfine,
- unsigned int numberOfLBnodesCoarse,
- unsigned int numberOfLBnodesFine,
+ unsigned int numberOfLBnodesCoarse,
+ unsigned int numberOfLBnodesFine,
bool isEvenTimestep,
- unsigned int* indicesCoarseMMM,
- unsigned int* indicesFineMMM,
- unsigned int numberOfInterfaceNodes,
- real omegaCoarse,
- real omegaFine,
+ unsigned int* indicesCoarseMMM,
+ unsigned int* indicesFineMMM,
+ unsigned int numberOfInterfaceNodes,
+ real omegaCoarse,
+ real omegaFine,
OffCF offsetCF);
__global__ void scaleCF_RhoSq_3rdMom_comp_27(real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffCF offCF);
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffCF offCF);
__global__ void scaleCF_AA2016_comp_27(real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffCF offCF);
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffCF offCF);
__global__ void scaleCF_NSPress_27(real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffCF offCF);
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffCF offCF);
__global__ void scaleCFThSMG7( real* DC,
real* DF,
@@ -1994,7 +1994,7 @@ __global__ void scaleCFThS27(real* DC,
unsigned int kCF,
real nu,
real diffusivity_fine,
- OffCF offCF);
+ OffCF offCF);
//fine to coarse
__global__ void scaleFC27(real* DC,
@@ -2005,18 +2005,18 @@ __global__ void scaleFC27(real* DC,
unsigned int* neighborFX,
unsigned int* neighborFY,
unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
unsigned int* posC,
unsigned int* posFSWB,
unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
unsigned int nyF);
__global__ void scaleFCEff27(real* DC,
@@ -2112,145 +2112,145 @@ __global__ void scaleFC_Fix_27( real* DC,
OffFC offFC);
__global__ void scaleFC_Fix_comp_27( real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffFC offFC);
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffFC offFC);
__global__ void scaleFC_0817_comp_27( real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffFC offFC);
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffFC offFC);
__global__ void scaleFC_comp_D3Q27F3_2018( real* DC,
- real* DF,
- real* G6,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffFC offFC);
+ real* DF,
+ real* G6,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffFC offFC);
__global__ void scaleFC_comp_D3Q27F3( real* DC,
- real* DF,
- real* G6,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffFC offFC);
+ real* DF,
+ real* G6,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffFC offFC);
__global__ void scaleFC_staggered_time_comp_27(real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffFC offFC);
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffFC offFC);
__global__ void scaleFC_RhoSq_comp_27( real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffFC offFC);
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffFC offFC);
__global__ void scaleFC_compressible(
real *distributionsCoarse,
@@ -2272,73 +2272,73 @@ __global__ void scaleFC_compressible(
OffFC offsetFC);
__global__ void scaleFC_RhoSq_3rdMom_comp_27(real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffFC offFC);
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffFC offFC);
__global__ void scaleFC_AA2016_comp_27(real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffFC offFC);
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffFC offFC);
__global__ void scaleFC_NSPress_27(real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- OffFC offFC);
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ OffFC offFC);
__global__ void scaleFCThSMG7( real* DC,
real* DF,
@@ -2397,47 +2397,47 @@ __global__ void scaleFCThS27( real* DC,
unsigned int kFC,
real nu,
real diffusivity_coarse,
- OffFC offFC);
+ OffFC offFC);
__global__ void DragLiftPost27( real* DD,
- int* k_Q,
- real* QQ,
- int numberOfBCnodes,
- double *DragX,
- double *DragY,
- double *DragZ,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ int* k_Q,
+ real* QQ,
+ int numberOfBCnodes,
+ double *DragX,
+ double *DragY,
+ double *DragZ,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void DragLiftPre27( real* DD,
- int* k_Q,
- real* QQ,
- int numberOfBCnodes,
- double *DragX,
- double *DragY,
- double *DragZ,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ int* k_Q,
+ real* QQ,
+ int numberOfBCnodes,
+ double *DragX,
+ double *DragY,
+ double *DragZ,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void CalcCP27(real* DD,
- int* cpIndex,
- int nonCp,
- double *cpPress,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ int* cpIndex,
+ int nonCp,
+ double *cpPress,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void getSendFsPre27(real* DD,
- real* bufferFs,
- int* sendIndex,
+ real* bufferFs,
+ int* sendIndex,
int buffmax,
unsigned int* neighborX,
unsigned int* neighborY,
@@ -2446,8 +2446,8 @@ __global__ void getSendFsPre27(real* DD,
bool isEvenTimestep);
__global__ void getSendFsPost27(real* DD,
- real* bufferFs,
- int* sendIndex,
+ real* bufferFs,
+ int* sendIndex,
int buffmax,
unsigned int* neighborX,
unsigned int* neighborY,
@@ -2456,8 +2456,8 @@ __global__ void getSendFsPost27(real* DD,
bool isEvenTimestep);
__global__ void setRecvFsPre27(real* DD,
- real* bufferFs,
- int* recvIndex,
+ real* bufferFs,
+ int* recvIndex,
int buffmax,
unsigned int* neighborX,
unsigned int* neighborY,
@@ -2466,8 +2466,8 @@ __global__ void setRecvFsPre27(real* DD,
bool isEvenTimestep);
__global__ void setRecvFsPost27(real* DD,
- real* bufferFs,
- int* recvIndex,
+ real* bufferFs,
+ int* recvIndex,
int buffmax,
unsigned int* neighborX,
unsigned int* neighborY,
@@ -2476,155 +2476,155 @@ __global__ void setRecvFsPost27(real* DD,
bool isEvenTimestep);
__global__ void getSendGsF3(
- real* G6,
- real* bufferGs,
- int* sendIndex,
- int buffmax,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* G6,
+ real* bufferGs,
+ int* sendIndex,
+ int buffmax,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void setRecvGsF3(
- real* G6,
- real* bufferGs,
- int* recvIndex,
- int buffmax,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* G6,
+ real* bufferGs,
+ int* recvIndex,
+ int buffmax,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void WallFunction27( real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void LBSetOutputWallVelocitySP27( real* vxD,
- real* vyD,
- real* vzD,
- real* vxWall,
- real* vyWall,
- real* vzWall,
- int numberOfWallNodes,
- int* kWallNodes,
- real* rhoD,
- real* pressD,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- real* DD,
- bool isEvenTimestep);
+ real* vyD,
+ real* vzD,
+ real* vxWall,
+ real* vyWall,
+ real* vzWall,
+ int numberOfWallNodes,
+ int* kWallNodes,
+ real* rhoD,
+ real* pressD,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ real* DD,
+ bool isEvenTimestep);
__global__ void GetVeloforForcing27( real* DD,
- int* bcIndex,
- int nonAtBC,
- real* Vx,
- real* Vy,
- real* Vz,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ int* bcIndex,
+ int nonAtBC,
+ real* Vx,
+ real* Vy,
+ real* Vz,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void InitParticles( real* coordX,
- real* coordY,
- real* coordZ,
- real* coordParticleXlocal,
- real* coordParticleYlocal,
- real* coordParticleZlocal,
- real* coordParticleXglobal,
- real* coordParticleYglobal,
- real* coordParticleZglobal,
- real* veloParticleX,
- real* veloParticleY,
- real* veloParticleZ,
- real* randArray,
- unsigned int* particleID,
- unsigned int* cellBaseID,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int* neighborWSB,
- int level,
- unsigned int numberOfParticles,
- unsigned int size_Mat);
+ real* coordY,
+ real* coordZ,
+ real* coordParticleXlocal,
+ real* coordParticleYlocal,
+ real* coordParticleZlocal,
+ real* coordParticleXglobal,
+ real* coordParticleYglobal,
+ real* coordParticleZglobal,
+ real* veloParticleX,
+ real* veloParticleY,
+ real* veloParticleZ,
+ real* randArray,
+ unsigned int* particleID,
+ unsigned int* cellBaseID,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int* neighborWSB,
+ int level,
+ unsigned int numberOfParticles,
+ unsigned int size_Mat);
__global__ void MoveParticles( real* coordX,
- real* coordY,
- real* coordZ,
- real* coordParticleXlocal,
- real* coordParticleYlocal,
- real* coordParticleZlocal,
- real* coordParticleXglobal,
- real* coordParticleYglobal,
- real* coordParticleZglobal,
- real* veloParticleX,
- real* veloParticleY,
- real* veloParticleZ,
- real* DD,
- real omega,
- unsigned int* particleID,
- unsigned int* cellBaseID,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int* neighborWSB,
- int level,
- unsigned int timestep,
- unsigned int numberOfTimesteps,
- unsigned int numberOfParticles,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* coordY,
+ real* coordZ,
+ real* coordParticleXlocal,
+ real* coordParticleYlocal,
+ real* coordParticleZlocal,
+ real* coordParticleXglobal,
+ real* coordParticleYglobal,
+ real* coordParticleZglobal,
+ real* veloParticleX,
+ real* veloParticleY,
+ real* veloParticleZ,
+ real* DD,
+ real omega,
+ unsigned int* particleID,
+ unsigned int* cellBaseID,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int* neighborWSB,
+ int level,
+ unsigned int timestep,
+ unsigned int numberOfTimesteps,
+ unsigned int numberOfParticles,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void MoveParticlesWithoutBCs( real* coordX,
- real* coordY,
- real* coordZ,
- real* coordParticleXlocal,
- real* coordParticleYlocal,
- real* coordParticleZlocal,
- real* coordParticleXglobal,
- real* coordParticleYglobal,
- real* coordParticleZglobal,
- real* veloParticleX,
- real* veloParticleY,
- real* veloParticleZ,
- real* DD,
- real omega,
- unsigned int* particleID,
- unsigned int* cellBaseID,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int* neighborWSB,
- int level,
- unsigned int timestep,
- unsigned int numberOfTimesteps,
- unsigned int numberOfParticles,
- unsigned int size_Mat,
- bool isEvenTimestep);
+ real* coordY,
+ real* coordZ,
+ real* coordParticleXlocal,
+ real* coordParticleYlocal,
+ real* coordParticleZlocal,
+ real* coordParticleXglobal,
+ real* coordParticleYglobal,
+ real* coordParticleZglobal,
+ real* veloParticleX,
+ real* veloParticleY,
+ real* veloParticleZ,
+ real* DD,
+ real omega,
+ unsigned int* particleID,
+ unsigned int* cellBaseID,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int* neighborWSB,
+ int level,
+ unsigned int timestep,
+ unsigned int numberOfTimesteps,
+ unsigned int numberOfParticles,
+ unsigned int size_Mat,
+ bool isEvenTimestep);
__global__ void initRandom(curandState* state);
__global__ void generateRandomValues(curandState* state,
- real* randArray);
+ real* randArray);
__global__ void CalcTurbulenceIntensity(
real* vxx,
diff --git a/src/gpu/VirtualFluids_GPU/GPU/LBMKernel.cu b/src/gpu/VirtualFluids_GPU/GPU/LBMKernel.cu
index 363fb76626b256205346a290a42a1da40e1c7445..ebab2b3e8a38bc221017ffb604d6f1c142afe1d9 100644
--- a/src/gpu/VirtualFluids_GPU/GPU/LBMKernel.cu
+++ b/src/gpu/VirtualFluids_GPU/GPU/LBMKernel.cu
@@ -114,429 +114,429 @@ void KernelCasSPMSOHM27( unsigned int numberOfThreads,
}
//////////////////////////////////////////////////////////////////////////
void KernelKumCompSRTSP27(
- unsigned int numberOfThreads,
- real omega,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DDStart,
- int size_Mat,
- int level,
- real* forces,
- bool EvenOrOdd)
+ unsigned int numberOfThreads,
+ real omega,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DDStart,
+ int size_Mat,
+ int level,
+ real* forces,
+ bool EvenOrOdd)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
LB_Kernel_Kum_New_Comp_SRT_SP_27 <<< grid.grid, grid.threads >>>(
- omega,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- DDStart,
- size_Mat,
- level,
- forces,
- EvenOrOdd);
+ omega,
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DDStart,
+ size_Mat,
+ level,
+ forces,
+ EvenOrOdd);
getLastCudaError("LB_Kernel_Kum_New_Comp_SRT_SP_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void KernelKum1hSP27( unsigned int numberOfThreads,
- real omega,
- real deltaPhi,
- real angularVelocity,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* coordX,
- real* coordY,
- real* coordZ,
- real* DDStart,
- int size_Mat,
- bool EvenOrOdd)
+ real omega,
+ real deltaPhi,
+ real angularVelocity,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* coordX,
+ real* coordY,
+ real* coordZ,
+ real* DDStart,
+ int size_Mat,
+ bool EvenOrOdd)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- LB_Kernel_Kum_1h_SP_27<<< grid.grid, grid.threads >>>(omega,
- deltaPhi,
- angularVelocity,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- coordX,
- coordY,
- coordZ,
- DDStart,
- size_Mat,
- EvenOrOdd);
- getLastCudaError("LB_Kernel_Kum_New_SP_27 execution failed");
+ LB_Kernel_Kum_1h_SP_27<<< grid.grid, grid.threads >>>(omega,
+ deltaPhi,
+ angularVelocity,
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ coordX,
+ coordY,
+ coordZ,
+ DDStart,
+ size_Mat,
+ EvenOrOdd);
+ getLastCudaError("LB_Kernel_Kum_New_SP_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void KernelCascadeSP27( unsigned int numberOfThreads,
- real s9,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DD,
- int size_Mat,
- bool EvenOrOdd)
+ real s9,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DD,
+ int size_Mat,
+ bool EvenOrOdd)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- LB_Kernel_Cascade_SP_27<<< grid.grid, grid.threads >>>(s9,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- DD,
- size_Mat,
- EvenOrOdd);
- getLastCudaError("LB_Kernel_Cascade_SP_27 execution failed");
+ LB_Kernel_Cascade_SP_27<<< grid.grid, grid.threads >>>(s9,
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DD,
+ size_Mat,
+ EvenOrOdd);
+ getLastCudaError("LB_Kernel_Cascade_SP_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void KernelKumNewSP27( unsigned int numberOfThreads,
- real s9,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DD,
- int size_Mat,
- bool EvenOrOdd)
+ real s9,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DD,
+ int size_Mat,
+ bool EvenOrOdd)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- LB_Kernel_Kum_New_SP_27<<< grid.grid, grid.threads >>>(s9,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- DD,
- size_Mat,
- EvenOrOdd);
- getLastCudaError("LB_Kernel_Kum_New_SP_27 execution failed");
+ LB_Kernel_Kum_New_SP_27<<< grid.grid, grid.threads >>>(s9,
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DD,
+ size_Mat,
+ EvenOrOdd);
+ getLastCudaError("LB_Kernel_Kum_New_SP_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void KernelKumNewCompSP27(unsigned int numberOfThreads,
- real s9,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DD,
- int size_Mat,
- int size_Array,
- int level,
- real* forces,
- bool EvenOrOdd)
-{
- //int Grid = size_Array / numberOfThreads;
- //dim3 grid(Grid, 1, 1);
- //dim3 threads(numberOfThreads, 1, 1 );
+ real s9,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DD,
+ int size_Mat,
+ int size_Array,
+ int level,
+ real* forces,
+ bool EvenOrOdd)
+{
+ //int Grid = size_Array / numberOfThreads;
+ //dim3 grid(Grid, 1, 1);
+ //dim3 threads(numberOfThreads, 1, 1 );
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- //LB_Kernel_Kum_New_Comp_SP_27<<< grid.grid, grid.threads >>>( s9,
- // bcMatD,
- // neighborX,
- // neighborY,
- // neighborZ,
- // DD,
- // size_Mat,
- // level,
- // forces,
- // EvenOrOdd);
- //getLastCudaError("LB_Kernel_Kum_New_Comp_SP_27 execution failed");
+ //LB_Kernel_Kum_New_Comp_SP_27<<< grid.grid, grid.threads >>>( s9,
+ // bcMatD,
+ // neighborX,
+ // neighborY,
+ // neighborZ,
+ // DD,
+ // size_Mat,
+ // level,
+ // forces,
+ // EvenOrOdd);
+ //getLastCudaError("LB_Kernel_Kum_New_Comp_SP_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CumulantOnePreconditionedErrorDiffusionChimCompSP27(unsigned int numberOfThreads,
- real s9,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DD,
- int size_Mat,
- int size_Array,
- int level,
- real* forces,
- bool EvenOrOdd)
-{
- //int Grid = size_Array / numberOfThreads;
- //dim3 grid(Grid, 1, 1);
- //dim3 threads(numberOfThreads, 1, 1 );
+ real s9,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DD,
+ int size_Mat,
+ int size_Array,
+ int level,
+ real* forces,
+ bool EvenOrOdd)
+{
+ //int Grid = size_Array / numberOfThreads;
+ //dim3 grid(Grid, 1, 1);
+ //dim3 threads(numberOfThreads, 1, 1 );
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- Cumulant_One_preconditioned_errorDiffusion_chim_Comp_SP_27 <<< grid.grid, grid.threads >>>( s9,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- DD,
- size_Mat,
- level,
- forces,
- EvenOrOdd);
- getLastCudaError("Cumulant_One_preconditioned_chim_Comp_SP_27 execution failed");
+ Cumulant_One_preconditioned_errorDiffusion_chim_Comp_SP_27 <<< grid.grid, grid.threads >>>( s9,
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DD,
+ size_Mat,
+ level,
+ forces,
+ EvenOrOdd);
+ getLastCudaError("Cumulant_One_preconditioned_chim_Comp_SP_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CumulantOnePreconditionedChimCompSP27( unsigned int numberOfThreads,
- real s9,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DD,
- int size_Mat,
- int size_Array,
- int level,
- real* forces,
- bool EvenOrOdd)
-{
- //int Grid = size_Array / numberOfThreads;
- //dim3 grid(Grid, 1, 1);
- //dim3 threads(numberOfThreads, 1, 1 );
+ real s9,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DD,
+ int size_Mat,
+ int size_Array,
+ int level,
+ real* forces,
+ bool EvenOrOdd)
+{
+ //int Grid = size_Array / numberOfThreads;
+ //dim3 grid(Grid, 1, 1);
+ //dim3 threads(numberOfThreads, 1, 1 );
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- Cumulant_One_preconditioned_chim_Comp_SP_27 <<< grid.grid, grid.threads >>>( s9,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- DD,
- size_Mat,
- level,
- forces,
- EvenOrOdd);
- getLastCudaError("Cumulant_One_preconditioned_chim_Comp_SP_27 execution failed");
+ Cumulant_One_preconditioned_chim_Comp_SP_27 <<< grid.grid, grid.threads >>>( s9,
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DD,
+ size_Mat,
+ level,
+ forces,
+ EvenOrOdd);
+ getLastCudaError("Cumulant_One_preconditioned_chim_Comp_SP_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CumulantOneChimCompSP27(unsigned int numberOfThreads,
- real s9,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DD,
- int size_Mat,
- int size_Array,
- int level,
- real* forces,
- bool EvenOrOdd)
-{
- //int Grid = size_Array / numberOfThreads;
- //dim3 grid(Grid, 1, 1);
- //dim3 threads(numberOfThreads, 1, 1 );
+ real s9,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DD,
+ int size_Mat,
+ int size_Array,
+ int level,
+ real* forces,
+ bool EvenOrOdd)
+{
+ //int Grid = size_Array / numberOfThreads;
+ //dim3 grid(Grid, 1, 1);
+ //dim3 threads(numberOfThreads, 1, 1 );
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- Cumulant_One_chim_Comp_SP_27 <<< grid.grid, grid.threads >>>( s9,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- DD,
- size_Mat,
- level,
- forces,
- EvenOrOdd);
- getLastCudaError("Cumulant_One_chim_Comp_SP_27 execution failed");
+ Cumulant_One_chim_Comp_SP_27 <<< grid.grid, grid.threads >>>( s9,
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DD,
+ size_Mat,
+ level,
+ forces,
+ EvenOrOdd);
+ getLastCudaError("Cumulant_One_chim_Comp_SP_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void KernelKumIsoTestSP27(unsigned int numberOfThreads,
- real s9,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DD,
- real* dxxUx,
- real* dyyUy,
- real* dzzUz,
- int size_Mat,
- bool EvenOrOdd)
+ real s9,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DD,
+ real* dxxUx,
+ real* dyyUy,
+ real* dzzUz,
+ int size_Mat,
+ bool EvenOrOdd)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- LB_Kernel_Kum_IsoTest_SP_27<<< grid.grid, grid.threads >>>(s9,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- DD,
- dxxUx,
- dyyUy,
- dzzUz,
- size_Mat,
- EvenOrOdd);
- getLastCudaError("LB_Kernel_Kum_IsoTest_SP_27 execution failed");
+ LB_Kernel_Kum_IsoTest_SP_27<<< grid.grid, grid.threads >>>(s9,
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DD,
+ dxxUx,
+ dyyUy,
+ dzzUz,
+ size_Mat,
+ EvenOrOdd);
+ getLastCudaError("LB_Kernel_Kum_IsoTest_SP_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void KernelKumCompSP27( unsigned int numberOfThreads,
- real s9,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DD,
- int size_Mat,
- bool EvenOrOdd)
+ real s9,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DD,
+ int size_Mat,
+ bool EvenOrOdd)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- LB_Kernel_Kum_Comp_SP_27<<< grid.grid, grid.threads >>>(s9,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- DD,
- size_Mat,
- EvenOrOdd);
- getLastCudaError("LB_Kernel_Kum_Comp_SP_27 execution failed");
+ LB_Kernel_Kum_Comp_SP_27<<< grid.grid, grid.threads >>>(s9,
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DD,
+ size_Mat,
+ EvenOrOdd);
+ getLastCudaError("LB_Kernel_Kum_Comp_SP_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void KernelPMCumOneCompSP27(unsigned int numberOfThreads,
- real omega,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DD,
- int size_Mat,
- int level,
- real* forces,
- real porosity,
- real darcy,
- real forchheimer,
- unsigned int sizeOfPorousMedia,
- unsigned int* nodeIdsPorousMedia,
- bool EvenOrOdd)
+ real omega,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DD,
+ int size_Mat,
+ int level,
+ real* forces,
+ real porosity,
+ real darcy,
+ real forchheimer,
+ unsigned int sizeOfPorousMedia,
+ unsigned int* nodeIdsPorousMedia,
+ bool EvenOrOdd)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- LB_Kernel_PM_Cum_One_Comp_SP_27 <<< grid.grid, grid.threads >>>(omega,
- neighborX,
- neighborY,
- neighborZ,
- DD,
- size_Mat,
- level,
- forces,
- porosity,
- darcy,
- forchheimer,
- sizeOfPorousMedia,
- nodeIdsPorousMedia,
- EvenOrOdd);
- getLastCudaError("LB_Kernel_PM_Cum_One_Comp_SP_27 execution failed");
+ LB_Kernel_PM_Cum_One_Comp_SP_27 <<< grid.grid, grid.threads >>>(omega,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DD,
+ size_Mat,
+ level,
+ forces,
+ porosity,
+ darcy,
+ forchheimer,
+ sizeOfPorousMedia,
+ nodeIdsPorousMedia,
+ EvenOrOdd);
+ getLastCudaError("LB_Kernel_PM_Cum_One_Comp_SP_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void KernelWaleBySoniMalavCumAA2016CompSP27(
- unsigned int numberOfThreads,
- real s9,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int* neighborWSB,
- real* veloX,
- real* veloY,
- real* veloZ,
- real* DD,
- real* turbulentViscosity,
- int size_Mat,
- int size_Array,
- int level,
- real* forces,
- bool EvenOrOdd)
-{
- //int Grid = size_Array / numberOfThreads;
- //dim3 grid(Grid, 1, 1);
- //dim3 threads(numberOfThreads, 1, 1 );
+ unsigned int numberOfThreads,
+ real s9,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int* neighborWSB,
+ real* veloX,
+ real* veloY,
+ real* veloZ,
+ real* DD,
+ real* turbulentViscosity,
+ int size_Mat,
+ int size_Array,
+ int level,
+ real* forces,
+ bool EvenOrOdd)
+{
+ //int Grid = size_Array / numberOfThreads;
+ //dim3 grid(Grid, 1, 1);
+ //dim3 threads(numberOfThreads, 1, 1 );
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- LB_Kernel_WaleBySoniMalav_Cum_AA2016_Comp_SP_27 << < grid.grid, grid.threads >> >(
- s9,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- neighborWSB,
- veloX,
- veloY,
- veloZ,
- DD,
- turbulentViscosity,
- size_Mat,
- level,
- forces,
- EvenOrOdd);
- getLastCudaError("LB_Kernel_WaleBySoniMalav_Cum_AA2016_Comp_SP_27 execution failed");
+ LB_Kernel_WaleBySoniMalav_Cum_AA2016_Comp_SP_27 << < grid.grid, grid.threads >> >(
+ s9,
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ neighborWSB,
+ veloX,
+ veloY,
+ veloZ,
+ DD,
+ turbulentViscosity,
+ size_Mat,
+ level,
+ forces,
+ EvenOrOdd);
+ getLastCudaError("LB_Kernel_WaleBySoniMalav_Cum_AA2016_Comp_SP_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void KernelADincomp7( unsigned int numberOfThreads,
- real diffusivity,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DD,
- real* DD7,
- int size_Mat,
- bool EvenOrOdd)
+ real diffusivity,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DD,
+ real* DD7,
+ int size_Mat,
+ bool EvenOrOdd)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
LB_Kernel_AD_Incomp_7<<< grid.grid, grid.threads >>>( diffusivity,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- DD,
- DD7,
- size_Mat,
- EvenOrOdd);
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DD,
+ DD7,
+ size_Mat,
+ EvenOrOdd);
getLastCudaError("LB_Kernel_AD_Incomp_7 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void KernelADincomp27( unsigned int numberOfThreads,
- real diffusivity,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* DD,
- real* DD27,
- int size_Mat,
- bool EvenOrOdd)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
-
- LB_Kernel_AD_Incomp_27<<< grid.grid, grid.threads >>>( diffusivity,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- DD,
- DD27,
- size_Mat,
- EvenOrOdd);
- getLastCudaError("LB_Kernel_AD_Incomp_27 execution failed");
+ real diffusivity,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* DD,
+ real* DD27,
+ int size_Mat,
+ bool EvenOrOdd)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+
+ LB_Kernel_AD_Incomp_27<<< grid.grid, grid.threads >>>( diffusivity,
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DD,
+ DD27,
+ size_Mat,
+ EvenOrOdd);
+ getLastCudaError("LB_Kernel_AD_Incomp_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void Init27( int myid,
@@ -558,7 +558,7 @@ void Init27( int myid,
dim3 threads ( grid_nx, 1, 1 );
dim3 grid ( grid_ny, grid_nz ); // Gitter fuer Kollision und Propagation
- LBInit27<<< grid, threads >>> ( myid,
+ LBInit27<<< grid, threads >>> ( myid,
numprocs,
u0,
geoD,
@@ -573,7 +573,7 @@ void Init27( int myid,
DD,
level,
maxlevel);
- getLastCudaError("LBInit27 execution failed");
+ getLastCudaError("LBInit27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void InitNonEqPartSP27( unsigned int numberOfThreads,
@@ -591,9 +591,9 @@ void InitNonEqPartSP27( unsigned int numberOfThreads,
real omega,
bool EvenOrOdd)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- LBInitNonEqPartSP27<<< grid.grid, grid.threads >>>( neighborX,
+ LBInitNonEqPartSP27<<< grid.grid, grid.threads >>>( neighborX,
neighborY,
neighborZ,
neighborWSB,
@@ -606,7 +606,7 @@ void InitNonEqPartSP27( unsigned int numberOfThreads,
DD,
omega,
EvenOrOdd);
- getLastCudaError("LBInitNonEqPartSP27 execution failed");
+ getLastCudaError("LBInitNonEqPartSP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void InitThS7( unsigned int numberOfThreads,
@@ -622,9 +622,9 @@ void InitThS7( unsigned int numberOfThreads,
real* DD7,
bool EvenOrOdd)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- InitAD7<<< grid.grid, grid.threads >>>( neighborX,
+ InitAD7<<< grid.grid, grid.threads >>>( neighborX,
neighborY,
neighborZ,
geoD,
@@ -635,7 +635,7 @@ void InitThS7( unsigned int numberOfThreads,
size_Mat,
DD7,
EvenOrOdd);
- getLastCudaError("InitAD7 execution failed");
+ getLastCudaError("InitAD7 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void InitADDev27( unsigned int numberOfThreads,
@@ -651,9 +651,9 @@ void InitADDev27( unsigned int numberOfThreads,
real* DD27,
bool EvenOrOdd)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- InitAD27<<< grid.grid, grid.threads >>>(neighborX,
+ InitAD27<<< grid.grid, grid.threads >>>(neighborX,
neighborY,
neighborZ,
geoD,
@@ -664,45 +664,45 @@ void InitADDev27( unsigned int numberOfThreads,
size_Mat,
DD27,
EvenOrOdd);
- getLastCudaError("InitAD27 execution failed");
+ getLastCudaError("InitAD27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void PostProcessorF3_2018Fehlberg(
- unsigned int numberOfThreads,
- real omega,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* rhoOut,
- real* vxOut,
- real* vyOut,
- real* vzOut,
- real* DDStart,
- real* G6,
- int size_Mat,
- int level,
- real* forces,
- bool EvenOrOdd)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
-
- LB_PostProcessor_F3_2018_Fehlberg <<< grid.grid, grid.threads >>> ( omega,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- rhoOut,
- vxOut,
- vyOut,
- vzOut,
- DDStart,
- G6,
- size_Mat,
- level,
- forces,
- EvenOrOdd);
- getLastCudaError("LB_PostProcessor_F3_2018_Fehlberg execution failed");
+ unsigned int numberOfThreads,
+ real omega,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* rhoOut,
+ real* vxOut,
+ real* vyOut,
+ real* vzOut,
+ real* DDStart,
+ real* G6,
+ int size_Mat,
+ int level,
+ real* forces,
+ bool EvenOrOdd)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+
+ LB_PostProcessor_F3_2018_Fehlberg <<< grid.grid, grid.threads >>> ( omega,
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ rhoOut,
+ vxOut,
+ vyOut,
+ vzOut,
+ DDStart,
+ G6,
+ size_Mat,
+ level,
+ forces,
+ EvenOrOdd);
+ getLastCudaError("LB_PostProcessor_F3_2018_Fehlberg execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CalcMac27( real* vxD,
@@ -723,7 +723,7 @@ void CalcMac27( real* vxD,
dim3 threads ( grid_nx, 1, 1 );
dim3 grid ( grid_ny, grid_nz );
- LBCalcMac27<<< grid, threads >>> ( vxD,
+ LBCalcMac27<<< grid, threads >>> ( vxD,
vyD,
vzD,
rhoD,
@@ -734,7 +734,7 @@ void CalcMac27( real* vxD,
size_Mat,
DD,
isEvenTimestep);
- getLastCudaError("LBCalcMac27 execution failed");
+ getLastCudaError("LBCalcMac27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CalcMacSP27( real* vxD,
@@ -751,9 +751,9 @@ void CalcMacSP27( real* vxD,
real* DD,
bool isEvenTimestep)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- LBCalcMacSP27<<< grid.grid, grid.threads >>> ( vxD,
+ LBCalcMacSP27<<< grid.grid, grid.threads >>> ( vxD,
vyD,
vzD,
rhoD,
@@ -765,38 +765,38 @@ void CalcMacSP27( real* vxD,
size_Mat,
DD,
isEvenTimestep);
- getLastCudaError("LBCalcMacSP27 execution failed");
+ getLastCudaError("LBCalcMacSP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CalcMacCompSP27( real* vxD,
- real* vyD,
- real* vzD,
- real* rhoD,
- real* pressD,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- unsigned int numberOfThreads,
- real* DD,
- bool isEvenTimestep)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
-
- LBCalcMacCompSP27<<< grid.grid, grid.threads >>> ( vxD,
- vyD,
- vzD,
- rhoD,
- pressD,
- geoD,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- DD,
- isEvenTimestep);
- getLastCudaError("LBCalcMacSP27 execution failed");
+ real* vyD,
+ real* vzD,
+ real* rhoD,
+ real* pressD,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ unsigned int numberOfThreads,
+ real* DD,
+ bool isEvenTimestep)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+
+ LBCalcMacCompSP27<<< grid.grid, grid.threads >>> ( vxD,
+ vyD,
+ vzD,
+ rhoD,
+ pressD,
+ geoD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ DD,
+ isEvenTimestep);
+ getLastCudaError("LBCalcMacSP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CalcMacThS7( real* Conc,
@@ -809,9 +809,9 @@ void CalcMacThS7( real* Conc,
real* DD7,
bool isEvenTimestep)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- CalcConc7<<< grid.grid, grid.threads >>> (Conc,
+ CalcConc7<<< grid.grid, grid.threads >>> (Conc,
geoD,
neighborX,
neighborY,
@@ -819,61 +819,61 @@ void CalcMacThS7( real* Conc,
size_Mat,
DD7,
isEvenTimestep);
- getLastCudaError("CalcConc7 execution failed");
+ getLastCudaError("CalcConc7 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void PlaneConcThS7(real* Conc,
- int* kPC,
- unsigned int numberOfPointskPC,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
+ int* kPC,
+ unsigned int numberOfPointskPC,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
unsigned int numberOfThreads,
- real* DD7,
- bool isEvenTimestep)
+ real* DD7,
+ bool isEvenTimestep)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfPointskPC);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfPointskPC);
- GetPlaneConc7<<< grid.grid, grid.threads >>> ( Conc,
- kPC,
- numberOfPointskPC,
- geoD,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- DD7,
- isEvenTimestep);
- getLastCudaError("GetPlaneConc7 execution failed");
+ GetPlaneConc7<<< grid.grid, grid.threads >>> ( Conc,
+ kPC,
+ numberOfPointskPC,
+ geoD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ DD7,
+ isEvenTimestep);
+ getLastCudaError("GetPlaneConc7 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void PlaneConcThS27(real* Conc,
- int* kPC,
- unsigned int numberOfPointskPC,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
+ int* kPC,
+ unsigned int numberOfPointskPC,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
unsigned int numberOfThreads,
- real* DD27,
- bool isEvenTimestep)
+ real* DD27,
+ bool isEvenTimestep)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfPointskPC);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfPointskPC);
- GetPlaneConc27<<< grid.grid, grid.threads >>> ( Conc,
- kPC,
- numberOfPointskPC,
- geoD,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- DD27,
- isEvenTimestep);
- getLastCudaError("GetPlaneConc27 execution failed");
+ GetPlaneConc27<<< grid.grid, grid.threads >>> ( Conc,
+ kPC,
+ numberOfPointskPC,
+ geoD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ DD27,
+ isEvenTimestep);
+ getLastCudaError("GetPlaneConc27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CalcConcentration27( unsigned int numberOfThreads,
@@ -886,9 +886,9 @@ void CalcConcentration27( unsigned int numberOfThreads,
real* DD27,
bool isEvenTimestep)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- CalcConc27<<< grid.grid, grid.threads >>> ( Conc,
+ CalcConc27<<< grid.grid, grid.threads >>> ( Conc,
geoD,
neighborX,
neighborY,
@@ -896,7 +896,7 @@ void CalcConcentration27( unsigned int numberOfThreads,
size_Mat,
DD27,
isEvenTimestep);
- getLastCudaError("CalcConc27 execution failed");
+ getLastCudaError("CalcConc27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CalcMedSP27( real* vxD,
@@ -913,9 +913,9 @@ void CalcMedSP27( real* vxD,
real* DD,
bool isEvenTimestep)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- LBCalcMedSP27<<< grid.grid, grid.threads >>> ( vxD,
+ LBCalcMedSP27<<< grid.grid, grid.threads >>> ( vxD,
vyD,
vzD,
rhoD,
@@ -927,75 +927,75 @@ void CalcMedSP27( real* vxD,
size_Mat,
DD,
isEvenTimestep);
- getLastCudaError("LBCalcMedSP27 execution failed");
+ getLastCudaError("LBCalcMedSP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CalcMedCompSP27( real* vxD,
- real* vyD,
- real* vzD,
- real* rhoD,
- real* pressD,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- unsigned int numberOfThreads,
- real* DD,
- bool isEvenTimestep)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
-
- LBCalcMedCompSP27<<< grid.grid, grid.threads >>> ( vxD,
- vyD,
- vzD,
- rhoD,
- pressD,
- geoD,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- DD,
- isEvenTimestep);
- getLastCudaError("LBCalcMedSP27 execution failed");
+ real* vyD,
+ real* vzD,
+ real* rhoD,
+ real* pressD,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ unsigned int numberOfThreads,
+ real* DD,
+ bool isEvenTimestep)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+
+ LBCalcMedCompSP27<<< grid.grid, grid.threads >>> ( vxD,
+ vyD,
+ vzD,
+ rhoD,
+ pressD,
+ geoD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ DD,
+ isEvenTimestep);
+ getLastCudaError("LBCalcMedSP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CalcMedCompAD27(
- real* vxD,
- real* vyD,
- real* vzD,
- real* rhoD,
- real* pressD,
- real* concD,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- unsigned int numberOfThreads,
- real* DD,
- real* DD_AD,
- bool isEvenTimestep)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
-
- LBCalcMedCompAD27 <<< grid.grid, grid.threads >>> (
- vxD,
- vyD,
- vzD,
- rhoD,
- pressD,
- concD,
- geoD,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- DD,
- DD_AD,
- isEvenTimestep);
- getLastCudaError("LBCalcMedAD27 execution failed");
+ real* vxD,
+ real* vyD,
+ real* vzD,
+ real* rhoD,
+ real* pressD,
+ real* concD,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ unsigned int numberOfThreads,
+ real* DD,
+ real* DD_AD,
+ bool isEvenTimestep)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+
+ LBCalcMedCompAD27 <<< grid.grid, grid.threads >>> (
+ vxD,
+ vyD,
+ vzD,
+ rhoD,
+ pressD,
+ concD,
+ geoD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ DD,
+ DD_AD,
+ isEvenTimestep);
+ getLastCudaError("LBCalcMedAD27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CalcMacMedSP27( real* vxD,
@@ -1012,9 +1012,9 @@ void CalcMacMedSP27( real* vxD,
unsigned int numberOfThreads,
bool isEvenTimestep)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- LBCalcMacMedSP27<<< grid.grid, grid.threads >>> ( vxD,
+ LBCalcMacMedSP27<<< grid.grid, grid.threads >>> ( vxD,
vyD,
vzD,
rhoD,
@@ -1026,270 +1026,270 @@ void CalcMacMedSP27( real* vxD,
tdiff,
size_Mat,
isEvenTimestep);
- getLastCudaError("LBCalcMacMedSP27 execution failed");
+ getLastCudaError("LBCalcMacMedSP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void ResetMedianValuesSP27(
- real* vxD,
- real* vyD,
- real* vzD,
- real* rhoD,
- real* pressD,
- unsigned int size_Mat,
- unsigned int numberOfThreads,
- bool isEvenTimestep)
+ real* vxD,
+ real* vyD,
+ real* vzD,
+ real* rhoD,
+ real* pressD,
+ unsigned int size_Mat,
+ unsigned int numberOfThreads,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
- LBResetMedianValuesSP27 << < grid.grid, grid.threads >> > (
- vxD,
- vyD,
- vzD,
- rhoD,
- pressD,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("LBResetMedianValuesSP27 execution failed");
+ LBResetMedianValuesSP27 << < grid.grid, grid.threads >> > (
+ vxD,
+ vyD,
+ vzD,
+ rhoD,
+ pressD,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("LBResetMedianValuesSP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void ResetMedianValuesAD27(
- real* vxD,
- real* vyD,
- real* vzD,
- real* rhoD,
- real* pressD,
- real* concD,
- unsigned int size_Mat,
- unsigned int numberOfThreads,
- bool isEvenTimestep)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
-
- LBResetMedianValuesAD27 << < grid.grid, grid.threads >> > (
- vxD,
- vyD,
- vzD,
- rhoD,
- pressD,
- concD,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("LBResetMedianValuesAD27 execution failed");
+ real* vxD,
+ real* vyD,
+ real* vzD,
+ real* rhoD,
+ real* pressD,
+ real* concD,
+ unsigned int size_Mat,
+ unsigned int numberOfThreads,
+ bool isEvenTimestep)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+
+ LBResetMedianValuesAD27 << < grid.grid, grid.threads >> > (
+ vxD,
+ vyD,
+ vzD,
+ rhoD,
+ pressD,
+ concD,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("LBResetMedianValuesAD27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void Calc2ndMomentsIncompSP27(real* kxyFromfcNEQ,
- real* kyzFromfcNEQ,
- real* kxzFromfcNEQ,
- real* kxxMyyFromfcNEQ,
- real* kxxMzzFromfcNEQ,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- unsigned int numberOfThreads,
- real* DD,
- bool isEvenTimestep)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
-
- LBCalc2ndMomentsIncompSP27<<< grid.grid, grid.threads >>> ( kxyFromfcNEQ,
- kyzFromfcNEQ,
- kxzFromfcNEQ,
- kxxMyyFromfcNEQ,
- kxxMzzFromfcNEQ,
- geoD,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- DD,
- isEvenTimestep);
- getLastCudaError("LBCalc2ndMomentsIncompSP27 execution failed");
+ real* kyzFromfcNEQ,
+ real* kxzFromfcNEQ,
+ real* kxxMyyFromfcNEQ,
+ real* kxxMzzFromfcNEQ,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ unsigned int numberOfThreads,
+ real* DD,
+ bool isEvenTimestep)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+
+ LBCalc2ndMomentsIncompSP27<<< grid.grid, grid.threads >>> ( kxyFromfcNEQ,
+ kyzFromfcNEQ,
+ kxzFromfcNEQ,
+ kxxMyyFromfcNEQ,
+ kxxMzzFromfcNEQ,
+ geoD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ DD,
+ isEvenTimestep);
+ getLastCudaError("LBCalc2ndMomentsIncompSP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void Calc2ndMomentsCompSP27( real* kxyFromfcNEQ,
- real* kyzFromfcNEQ,
- real* kxzFromfcNEQ,
- real* kxxMyyFromfcNEQ,
- real* kxxMzzFromfcNEQ,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- unsigned int numberOfThreads,
- real* DD,
- bool isEvenTimestep)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
-
- LBCalc2ndMomentsCompSP27<<< grid.grid, grid.threads >>> (kxyFromfcNEQ,
- kyzFromfcNEQ,
- kxzFromfcNEQ,
- kxxMyyFromfcNEQ,
- kxxMzzFromfcNEQ,
- geoD,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- DD,
- isEvenTimestep);
- getLastCudaError("LBCalc2ndMomentsCompSP27 execution failed");
+ real* kyzFromfcNEQ,
+ real* kxzFromfcNEQ,
+ real* kxxMyyFromfcNEQ,
+ real* kxxMzzFromfcNEQ,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ unsigned int numberOfThreads,
+ real* DD,
+ bool isEvenTimestep)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+
+ LBCalc2ndMomentsCompSP27<<< grid.grid, grid.threads >>> (kxyFromfcNEQ,
+ kyzFromfcNEQ,
+ kxzFromfcNEQ,
+ kxxMyyFromfcNEQ,
+ kxxMzzFromfcNEQ,
+ geoD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ DD,
+ isEvenTimestep);
+ getLastCudaError("LBCalc2ndMomentsCompSP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void Calc3rdMomentsIncompSP27(real* CUMbbb,
- real* CUMabc,
- real* CUMbac,
- real* CUMbca,
- real* CUMcba,
- real* CUMacb,
- real* CUMcab,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- unsigned int numberOfThreads,
- real* DD,
- bool isEvenTimestep)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
-
- LBCalc3rdMomentsIncompSP27<<< grid.grid, grid.threads >>> ( CUMbbb,
- CUMabc,
- CUMbac,
- CUMbca,
- CUMcba,
- CUMacb,
- CUMcab,
- geoD,
- neighborX,
- neighborY,
- neighborZ,
- DD,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("LBCalc3rdMomentsIncompSP27 execution failed");
+ real* CUMabc,
+ real* CUMbac,
+ real* CUMbca,
+ real* CUMcba,
+ real* CUMacb,
+ real* CUMcab,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ unsigned int numberOfThreads,
+ real* DD,
+ bool isEvenTimestep)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+
+ LBCalc3rdMomentsIncompSP27<<< grid.grid, grid.threads >>> ( CUMbbb,
+ CUMabc,
+ CUMbac,
+ CUMbca,
+ CUMcba,
+ CUMacb,
+ CUMcab,
+ geoD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DD,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("LBCalc3rdMomentsIncompSP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void Calc3rdMomentsCompSP27( real* CUMbbb,
- real* CUMabc,
- real* CUMbac,
- real* CUMbca,
- real* CUMcba,
- real* CUMacb,
- real* CUMcab,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- unsigned int numberOfThreads,
- real* DD,
- bool isEvenTimestep)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
-
- LBCalc3rdMomentsCompSP27<<< grid.grid, grid.threads >>> (CUMbbb,
- CUMabc,
- CUMbac,
- CUMbca,
- CUMcba,
- CUMacb,
- CUMcab,
- geoD,
- neighborX,
- neighborY,
- neighborZ,
- DD,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("LBCalc3rdMomentsCompSP27 execution failed");
+ real* CUMabc,
+ real* CUMbac,
+ real* CUMbca,
+ real* CUMcba,
+ real* CUMacb,
+ real* CUMcab,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ unsigned int numberOfThreads,
+ real* DD,
+ bool isEvenTimestep)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+
+ LBCalc3rdMomentsCompSP27<<< grid.grid, grid.threads >>> (CUMbbb,
+ CUMabc,
+ CUMbac,
+ CUMbca,
+ CUMcba,
+ CUMacb,
+ CUMcab,
+ geoD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DD,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("LBCalc3rdMomentsCompSP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CalcHigherMomentsIncompSP27(real* CUMcbb,
- real* CUMbcb,
- real* CUMbbc,
- real* CUMcca,
- real* CUMcac,
- real* CUMacc,
- real* CUMbcc,
- real* CUMcbc,
- real* CUMccb,
- real* CUMccc,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- unsigned int numberOfThreads,
- real* DD,
- bool isEvenTimestep)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
-
- LBCalcHigherMomentsIncompSP27<<< grid.grid, grid.threads >>> (CUMcbb,
- CUMbcb,
- CUMbbc,
- CUMcca,
- CUMcac,
- CUMacc,
- CUMbcc,
- CUMcbc,
- CUMccb,
- CUMccc,
- geoD,
- neighborX,
- neighborY,
- neighborZ,
- DD,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("LBCalcHigherMomentsIncompSP27 execution failed");
+ real* CUMbcb,
+ real* CUMbbc,
+ real* CUMcca,
+ real* CUMcac,
+ real* CUMacc,
+ real* CUMbcc,
+ real* CUMcbc,
+ real* CUMccb,
+ real* CUMccc,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ unsigned int numberOfThreads,
+ real* DD,
+ bool isEvenTimestep)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+
+ LBCalcHigherMomentsIncompSP27<<< grid.grid, grid.threads >>> (CUMcbb,
+ CUMbcb,
+ CUMbbc,
+ CUMcca,
+ CUMcac,
+ CUMacc,
+ CUMbcc,
+ CUMcbc,
+ CUMccb,
+ CUMccc,
+ geoD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DD,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("LBCalcHigherMomentsIncompSP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CalcHigherMomentsCompSP27( real* CUMcbb,
- real* CUMbcb,
- real* CUMbbc,
- real* CUMcca,
- real* CUMcac,
- real* CUMacc,
- real* CUMbcc,
- real* CUMcbc,
- real* CUMccb,
- real* CUMccc,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- unsigned int numberOfThreads,
- real* DD,
- bool isEvenTimestep)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
-
- LBCalcHigherMomentsCompSP27<<< grid.grid, grid.threads >>> ( CUMcbb,
- CUMbcb,
- CUMbbc,
- CUMcca,
- CUMcac,
- CUMacc,
- CUMbcc,
- CUMcbc,
- CUMccb,
- CUMccc,
- geoD,
- neighborX,
- neighborY,
- neighborZ,
- DD,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("LBCalcHigherMomentsCompSP27 execution failed");
+ real* CUMbcb,
+ real* CUMbbc,
+ real* CUMcca,
+ real* CUMcac,
+ real* CUMacc,
+ real* CUMbcc,
+ real* CUMcbc,
+ real* CUMccb,
+ real* CUMccc,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ unsigned int numberOfThreads,
+ real* DD,
+ bool isEvenTimestep)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
+
+ LBCalcHigherMomentsCompSP27<<< grid.grid, grid.threads >>> ( CUMcbb,
+ CUMbcb,
+ CUMbbc,
+ CUMcca,
+ CUMcac,
+ CUMacc,
+ CUMbcc,
+ CUMcbc,
+ CUMccb,
+ CUMccc,
+ geoD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ DD,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("LBCalcHigherMomentsCompSP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void LBCalcMeasurePoints27(real* vxMP,
@@ -1309,9 +1309,9 @@ void LBCalcMeasurePoints27(real* vxMP,
unsigned int numberOfThreads,
bool isEvenTimestep)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfPointskMP);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfPointskMP);
- LBCalcMeasurePoints<<< grid.grid, grid.threads >>> (vxMP,
+ LBCalcMeasurePoints<<< grid.grid, grid.threads >>> (vxMP,
vyMP,
vzMP,
rhoMP,
@@ -1326,7 +1326,7 @@ void LBCalcMeasurePoints27(real* vxMP,
size_Mat,
DD,
isEvenTimestep);
- getLastCudaError("LBCalcMeasurePoints execution failed");
+ getLastCudaError("LBCalcMeasurePoints execution failed");
}
//////////////////////////////////////////////////////////////////////////
void BcPress27( int nx,
@@ -1342,10 +1342,10 @@ void BcPress27( int nx,
unsigned int size_Mat,
bool isEvenTimestep)
{
- dim3 threads ( grid_nx, 1, 1 );
- dim3 grid ( grid_ny, 1 );
+ dim3 threads ( grid_nx, 1, 1 );
+ dim3 grid ( grid_ny, 1 );
- LB_BC_Press_East27<<< grid, threads >>> ( nx,
+ LB_BC_Press_East27<<< grid, threads >>> ( nx,
ny,
tz,
bcMatD,
@@ -1355,7 +1355,7 @@ void BcPress27( int nx,
DD,
size_Mat,
isEvenTimestep);
- getLastCudaError("LB_BC_Press_East27 execution failed");
+ getLastCudaError("LB_BC_Press_East27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void BcVel27(int nx,
@@ -1374,10 +1374,10 @@ void BcVel27(int nx,
real u0x,
real om)
{
- dim3 threads ( grid_nx, 1, 1 );
- dim3 grid ( grid_ny, 1 );
+ dim3 threads ( grid_nx, 1, 1 );
+ dim3 grid ( grid_ny, 1 );
- LB_BC_Vel_West_27<<< grid, threads >>> ( nx,
+ LB_BC_Vel_West_27<<< grid, threads >>> ( nx,
ny,
nz,
itz,
@@ -1392,7 +1392,7 @@ void BcVel27(int nx,
grid_nx,
grid_ny,
om);
- getLastCudaError("LB_BC_Vel_West_27 execution failed");
+ getLastCudaError("LB_BC_Vel_West_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QADPressDev7( unsigned int numberOfThreads,
@@ -1411,7 +1411,7 @@ void QADPressDev7( unsigned int numberOfThreads,
unsigned int size_Mat,
bool isEvenTimestep)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QADPress7<<< grid.grid, grid.threads >>>( DD,
DD7,
@@ -1427,7 +1427,7 @@ void QADPressDev7( unsigned int numberOfThreads,
neighborZ,
size_Mat,
isEvenTimestep);
- getLastCudaError("QADPress7 execution failed");
+ getLastCudaError("QADPress7 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QADPressDev27(unsigned int numberOfThreads,
@@ -1446,7 +1446,7 @@ void QADPressDev27(unsigned int numberOfThreads,
unsigned int size_Mat,
bool isEvenTimestep)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QADPress27<<< grid.grid, grid.threads >>>( DD,
DD27,
@@ -1462,39 +1462,39 @@ void QADPressDev27(unsigned int numberOfThreads,
neighborZ,
size_Mat,
isEvenTimestep);
- getLastCudaError("QADPress27 execution failed");
+ getLastCudaError("QADPress27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QADPressNEQNeighborDev27(
- unsigned int numberOfThreads,
- real* DD,
- real* DD27,
- int* k_Q,
- int* k_N,
- int numberOfBCnodes,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep
- )
-{
-
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
-
- QADPressNEQNeighbor27<<< grid.grid, grid.threads >>>(
- DD,
- DD27,
- k_Q,
- k_N,
- numberOfBCnodes,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep
- );
- getLastCudaError("QADPressNEQNeighbor27 execution failed");
+ unsigned int numberOfThreads,
+ real* DD,
+ real* DD27,
+ int* k_Q,
+ int* k_N,
+ int numberOfBCnodes,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep
+ )
+{
+
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
+
+ QADPressNEQNeighbor27<<< grid.grid, grid.threads >>>(
+ DD,
+ DD27,
+ k_Q,
+ k_N,
+ numberOfBCnodes,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep
+ );
+ getLastCudaError("QADPressNEQNeighbor27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QADVelDev7(unsigned int numberOfThreads,
@@ -1513,9 +1513,9 @@ void QADVelDev7(unsigned int numberOfThreads,
unsigned int size_Mat,
bool isEvenTimestep)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
- QADVel7<<< grid.grid, grid.threads >>> (
+ QADVel7<<< grid.grid, grid.threads >>> (
DD,
DD7,
temp,
@@ -1530,7 +1530,7 @@ void QADVelDev7(unsigned int numberOfThreads,
neighborZ,
size_Mat,
isEvenTimestep);
- getLastCudaError("QADVel7 execution failed");
+ getLastCudaError("QADVel7 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QADVelDev27( unsigned int numberOfThreads,
@@ -1583,7 +1583,7 @@ void QADDev7(unsigned int numberOfThreads,
unsigned int size_Mat,
bool isEvenTimestep)
{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QAD7<<< grid.grid, grid.threads >>> ( DD,
DD7,
@@ -1637,74 +1637,74 @@ void FactorizedCentralMomentsAdvectionDiffusionDeviceKernel(
//////////////////////////////////////////////////////////////////////////
void ADSlipVelDevComp(
- uint numberOfThreads,
- real * normalX,
- real * normalY,
- real * normalZ,
- real * distributions,
- real * distributionsAD,
- int* QindexArray,
- real * Qarrays,
- uint numberOfBCnodes,
- real omegaDiffusivity,
- uint * neighborX,
- uint * neighborY,
- uint * neighborZ,
- uint size_Mat,
- bool isEvenTimestep)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
-
- AD_SlipVelDeviceComp << < grid.grid, grid.threads >> > (
- normalX,
- normalY,
- normalZ,
- distributions,
- distributionsAD,
- QindexArray,
- Qarrays,
- numberOfBCnodes,
- omegaDiffusivity,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("AD_SlipVelDeviceComp execution failed");
+ uint numberOfThreads,
+ real * normalX,
+ real * normalY,
+ real * normalZ,
+ real * distributions,
+ real * distributionsAD,
+ int* QindexArray,
+ real * Qarrays,
+ uint numberOfBCnodes,
+ real omegaDiffusivity,
+ uint * neighborX,
+ uint * neighborY,
+ uint * neighborZ,
+ uint size_Mat,
+ bool isEvenTimestep)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
+
+ AD_SlipVelDeviceComp << < grid.grid, grid.threads >> > (
+ normalX,
+ normalY,
+ normalZ,
+ distributions,
+ distributionsAD,
+ QindexArray,
+ Qarrays,
+ numberOfBCnodes,
+ omegaDiffusivity,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("AD_SlipVelDeviceComp execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QADDirichletDev27( unsigned int numberOfThreads,
- real* DD,
- real* DD27,
- real* temp,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
+ real* DD,
+ real* DD27,
+ real* temp,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QADDirichlet27<<< grid.grid, grid.threads >>> (
- DD,
- DD27,
- temp,
- diffusivity,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ DD,
+ DD27,
+ temp,
+ diffusivity,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QADDirichletDev27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
@@ -1742,212 +1742,212 @@ void QADBBDev27(unsigned int numberOfThreads,
}
//////////////////////////////////////////////////////////////////////////
void QNoSlipADincompDev7(unsigned int numberOfThreads,
- real* DD,
- real* DD7,
- real* temp,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ real* DD7,
+ real* temp,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QNoSlipADincomp7<<< grid.grid, grid.threads >>> (
- DD,
- DD7,
- temp,
- diffusivity,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ DD,
+ DD7,
+ temp,
+ diffusivity,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QNoSlipADincomp7 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QNoSlipADincompDev27( unsigned int numberOfThreads,
- real* DD,
- real* DD27,
- real* temp,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ real* DD27,
+ real* temp,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QNoSlipADincomp27<<< grid.grid, grid.threads >>> (
- DD,
- DD27,
- temp,
- diffusivity,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ DD,
+ DD27,
+ temp,
+ diffusivity,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QNoSlipADincomp27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QADVeloIncompDev7( unsigned int numberOfThreads,
- real* DD,
- real* DD7,
- real* temp,
- real* velo,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ real* DD7,
+ real* temp,
+ real* velo,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QADVeloIncomp7<<< grid.grid, grid.threads >>> ( DD,
- DD7,
- temp,
- velo,
- diffusivity,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ DD7,
+ temp,
+ velo,
+ diffusivity,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QADVeloIncomp7 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QADVeloIncompDev27( unsigned int numberOfThreads,
- real* DD,
- real* DD27,
- real* temp,
- real* velo,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ real* DD27,
+ real* temp,
+ real* velo,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QADVeloIncomp27<<< grid.grid, grid.threads >>> (
- DD,
- DD27,
- temp,
- velo,
- diffusivity,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ DD,
+ DD27,
+ temp,
+ velo,
+ diffusivity,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QADVeloIncomp27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QADPressIncompDev7( unsigned int numberOfThreads,
- real* DD,
- real* DD7,
- real* temp,
- real* velo,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ real* DD7,
+ real* temp,
+ real* velo,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QADPressIncomp7<<< grid.grid, grid.threads >>>(
- DD,
- DD7,
- temp,
- velo,
- diffusivity,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ DD,
+ DD7,
+ temp,
+ velo,
+ diffusivity,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QADPressIncomp7 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QADPressIncompDev27( unsigned int numberOfThreads,
- real* DD,
- real* DD27,
- real* temp,
- real* velo,
- real diffusivity,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ real* DD27,
+ real* temp,
+ real* velo,
+ real diffusivity,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
- QADPressIncomp27<<< grid.grid, grid.threads >>>(DD,
- DD27,
- temp,
- velo,
- diffusivity,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ QADPressIncomp27<<< grid.grid, grid.threads >>>(DD,
+ DD27,
+ temp,
+ velo,
+ diffusivity,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QADPressIncomp27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
@@ -1991,44 +1991,44 @@ void QDevComp27(LBMSimulationParameter* parameterDevice, QforBoundaryConditions*
}
//////////////////////////////////////////////////////////////////////////
void QDevCompThinWalls27(unsigned int numberOfThreads,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* geom,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int* neighborWSB,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* geom,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int* neighborWSB,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QDeviceCompThinWallsPartOne27 <<< grid.grid, grid.threads >>> (DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QDeviceCompThinWallsPartOne27 execution failed");
QThinWallsPartTwo27 <<< grid.grid, grid.threads >>> ( DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- geom,
- neighborX,
- neighborY,
- neighborZ,
- neighborWSB,
- size_Mat,
- isEvenTimestep);
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ geom,
+ neighborX,
+ neighborY,
+ neighborZ,
+ neighborWSB,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QThinWallsPartTwo27 execution failed");
}
@@ -2053,58 +2053,58 @@ void QDev3rdMomentsComp27(LBMSimulationParameter* parameterDevice, QforBoundaryC
}
//////////////////////////////////////////////////////////////////////////
void QDevIncompHighNu27( unsigned int numberOfThreads,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QDeviceIncompHighNu27<<< grid.grid, grid.threads >>> (
- DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ DD,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QDeviceIncompHighNu27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QDevCompHighNu27( unsigned int numberOfThreads,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QDeviceCompHighNu27<<< grid.grid, grid.threads >>> (
- DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ DD,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QDevice27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
@@ -2130,59 +2130,59 @@ void QVelDevicePlainBB27(LBMSimulationParameter* parameterDevice, QforBoundaryCo
}
//////////////////////////////////////////////////////////////////////////
void QVelDeviceCouette27(unsigned int numberOfThreads,
- real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* vx,
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QVelDevCouette27<<< grid.grid, grid.threads >>> ( vx,
- vy,
- vz,
- DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ vy,
+ vz,
+ DD,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QVelDevicePlainBB27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QVelDevice1h27( unsigned int numberOfThreads,
- int nx,
- int ny,
- real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- real Phi,
- real angularVelocity,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- real* coordX,
- real* coordY,
- real* coordZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ int nx,
+ int ny,
+ real* vx,
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ real Phi,
+ real angularVelocity,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ real* coordX,
+ real* coordY,
+ real* coordZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
@@ -2196,14 +2196,14 @@ void QVelDevice1h27( unsigned int numberOfThreads,
QQ,
numberOfBCnodes,
om1,
- Phi,
- angularVelocity,
+ Phi,
+ angularVelocity,
neighborX,
neighborY,
neighborZ,
- coordX,
- coordY,
- coordZ,
+ coordX,
+ coordY,
+ coordZ,
size_Mat,
isEvenTimestep);
getLastCudaError("QVelDevice27 execution failed");
@@ -2234,36 +2234,36 @@ void QVelDev27(LBMSimulationParameter* parameterDevice, QforBoundaryConditions*
}
//////////////////////////////////////////////////////////////////////////
void QVelDevCompPlusSlip27(unsigned int numberOfThreads,
- real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* vx,
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QVelDeviceCompPlusSlip27<<< grid.grid, grid.threads >>> (
- vx,
- vy,
- vz,
- DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ vx,
+ vy,
+ vz,
+ DD,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QVelDeviceCompPlusSlip27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
@@ -2277,7 +2277,7 @@ void QVelDevComp27(LBMSimulationParameter* parameterDevice, QforBoundaryConditio
boundaryCondition->Vy,
boundaryCondition->Vz,
parameterDevice->distributions.f[0],
- boundaryCondition->k,
+ boundaryCondition->k,
boundaryCondition->q27[0],
boundaryCondition->numberOfBCnodes,
parameterDevice->omega,
@@ -2290,40 +2290,40 @@ void QVelDevComp27(LBMSimulationParameter* parameterDevice, QforBoundaryConditio
}
//////////////////////////////////////////////////////////////////////////
void QVelDevCompThinWalls27(unsigned int numberOfThreads,
- real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* geom,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int* neighborWSB,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* vx,
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* geom,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int* neighborWSB,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QVelDeviceCompThinWallsPartOne27<<< grid.grid, grid.threads >>> (vx,
- vy,
- vz,
- DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ vy,
+ vz,
+ DD,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QVelDeviceCompThinWallsPartOne27 execution failed");
- QThinWallsPartTwo27 <<< grid.grid, grid.threads >>> (
+ QThinWallsPartTwo27 <<< grid.grid, grid.threads >>> (
DD,
k_Q,
QQ,
@@ -2361,135 +2361,135 @@ void QVelDevCompZeroPress27(LBMSimulationParameter* parameterDevice, QforBoundar
}
//////////////////////////////////////////////////////////////////////////
void QVelDevIncompHighNu27(unsigned int numberOfThreads,
- real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* vx,
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QVelDeviceIncompHighNu27<<< grid.grid, grid.threads >>> (
- vx,
- vy,
- vz,
- DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ vx,
+ vy,
+ vz,
+ DD,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QVelDeviceIncompHighNu27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QVelDevCompHighNu27( unsigned int numberOfThreads,
- real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* vx,
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QVelDeviceCompHighNu27<<< grid.grid, grid.threads >>> (
- vx,
- vy,
- vz,
- DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ vx,
+ vy,
+ vz,
+ DD,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QVelDeviceComp27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QVeloDevEQ27(unsigned int numberOfThreads,
- real* VeloX,
- real* VeloY,
- real* VeloZ,
- real* DD,
- int* k_Q,
- int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* VeloX,
+ real* VeloY,
+ real* VeloZ,
+ real* DD,
+ int* k_Q,
+ int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QVeloDeviceEQ27<<< grid.grid, grid.threads >>> (VeloX,
- VeloY,
- VeloZ,
- DD,
- k_Q,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ VeloY,
+ VeloZ,
+ DD,
+ k_Q,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QVeloDeviceEQ27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QVeloStreetDevEQ27(
- uint numberOfThreads,
- real* veloXfraction,
- real* veloYfraction,
- int* naschVelo,
- real* DD,
- int* naschIndex,
- int numberOfStreetNodes,
- real velocityRatio,
- uint* neighborX,
- uint* neighborY,
- uint* neighborZ,
- uint size_Mat,
- bool isEvenTimestep)
+ uint numberOfThreads,
+ real* veloXfraction,
+ real* veloYfraction,
+ int* naschVelo,
+ real* DD,
+ int* naschIndex,
+ int numberOfStreetNodes,
+ real velocityRatio,
+ uint* neighborX,
+ uint* neighborY,
+ uint* neighborZ,
+ uint size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfStreetNodes);
- QVeloStreetDeviceEQ27 << < grid.grid, grid.threads >> > (
- veloXfraction,
- veloYfraction,
- naschVelo,
- DD,
- naschIndex,
- numberOfStreetNodes,
- velocityRatio,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("QVeloStreetDeviceEQ27 execution failed");
+ QVeloStreetDeviceEQ27 << < grid.grid, grid.threads >> > (
+ veloXfraction,
+ veloYfraction,
+ naschVelo,
+ DD,
+ naschIndex,
+ numberOfStreetNodes,
+ velocityRatio,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("QVeloStreetDeviceEQ27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QSlipDev27(LBMSimulationParameter* parameterDevice, QforBoundaryConditions* boundaryCondition)
@@ -2515,7 +2515,7 @@ void QSlipDevCompTurbulentViscosity27(LBMSimulationParameter* parameterDevice, Q
{
dim3 grid = vf::cuda::getCudaGrid( parameterDevice->numberofthreads, boundaryCondition->numberOfBCnodes);
dim3 threads(parameterDevice->numberofthreads, 1, 1 );
-
+
QSlipDeviceComp27TurbViscosity<<< grid, threads >>> (
parameterDevice->distributions.f[0],
boundaryCondition->k,
@@ -2555,7 +2555,7 @@ void QSlipDevComp27(LBMSimulationParameter* parameterDevice, QforBoundaryConditi
{
dim3 grid = vf::cuda::getCudaGrid( parameterDevice->numberofthreads, boundaryCondition->numberOfBCnodes);
dim3 threads(parameterDevice->numberofthreads, 1, 1 );
-
+
QSlipDeviceComp27<<< grid, threads >>> (
parameterDevice->distributions.f[0],
boundaryCondition->k,
@@ -2590,68 +2590,68 @@ void BBSlipDevComp27(LBMSimulationParameter* parameterDevice, QforBoundaryCondit
}
//////////////////////////////////////////////////////////////////////////
void QSlipGeomDevComp27(unsigned int numberOfThreads,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- real* NormalX,
- real* NormalY,
- real* NormalZ,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
-{
- vf::cuda::CudaGrid grid(numberOfThreads, numberOfBCnodes);
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ real* NormalX,
+ real* NormalY,
+ real* NormalZ,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
+{
+ vf::cuda::CudaGrid grid(numberOfThreads, numberOfBCnodes);
QSlipGeomDeviceComp27<<< grid.grid, grid.threads >>> (DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- NormalX,
- NormalY,
- NormalZ,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ NormalX,
+ NormalY,
+ NormalZ,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QSlipGeomDeviceComp27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QSlipNormDevComp27(unsigned int numberOfThreads,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- real* NormalX,
- real* NormalY,
- real* NormalZ,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ real* NormalX,
+ real* NormalY,
+ real* NormalZ,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QSlipNormDeviceComp27<<< grid.grid, grid.threads >>> (DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- NormalX,
- NormalY,
- NormalZ,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ NormalX,
+ NormalY,
+ NormalZ,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QSlipGeomDeviceComp27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
@@ -2796,36 +2796,36 @@ void QPressDev27(LBMSimulationParameter* parameterDevice, QforBoundaryConditions
//////////////////////////////////////////////////////////////////////////
void QPressDevAntiBB27( unsigned int numberOfThreads,
real* rhoBC,
- real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* vx,
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
QPressDeviceAntiBB27<<< grid.grid, grid.threads >>>( rhoBC,
- vx,
- vy,
- vz,
- DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ vx,
+ vy,
+ vz,
+ DD,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("QPressDeviceAntiBB27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
@@ -3108,32 +3108,32 @@ void BBDev27(LBMSimulationParameter* parameterDevice, QforBoundaryConditions* bo
}
//////////////////////////////////////////////////////////////////////////
void QPressDev27_IntBB( unsigned int numberOfThreads,
- real* rho,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* rho,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
- QPressDevice27_IntBB<<< grid.grid, grid.threads >>> (rho,
- DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("QPressDevice27_IntBB execution failed");
+ QPressDevice27_IntBB<<< grid.grid, grid.threads >>> (rho,
+ DD,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("QPressDevice27_IntBB execution failed");
}
// TODO: https://git.rz.tu-bs.de/irmb/VirtualFluids_dev/-/issues/29
//////////////////////////////////////////////////////////////////////////
@@ -3208,146 +3208,146 @@ void VelSchlaffer27( unsigned int numberOfThreads,
getLastCudaError("VelSchlaff27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
-void QPrecursorDevCompZeroPress(LBMSimulationParameter* parameterDevice,
- QforPrecursorBoundaryConditions* boundaryCondition,
- real timeRatio,
- real velocityRatio)
-{
-
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(parameterDevice->numberofthreads, boundaryCondition->numberOfBCnodes);
-
- QPrecursorDeviceCompZeroPress<<< grid.grid, grid.threads >>>(boundaryCondition->k,
- boundaryCondition->numberOfBCnodes,
- boundaryCondition->numberOfPrecursorNodes,
- boundaryCondition->sizeQ,
- parameterDevice->omega,
- parameterDevice->distributions.f[0],
- boundaryCondition->q27[0],
- parameterDevice->neighborX,
- parameterDevice->neighborY,
- parameterDevice->neighborZ,
- boundaryCondition->planeNeighbor0PP,
- boundaryCondition->planeNeighbor0PM,
- boundaryCondition->planeNeighbor0MP,
- boundaryCondition->planeNeighbor0MM,
- boundaryCondition->weights0PP,
- boundaryCondition->weights0PM,
- boundaryCondition->weights0MP,
- boundaryCondition->weights0MM,
- boundaryCondition->last,
- boundaryCondition->current,
- boundaryCondition->velocityX,
- boundaryCondition->velocityY,
- boundaryCondition->velocityZ,
- timeRatio,
- velocityRatio,
- parameterDevice->numberOfNodes,
- parameterDevice->isEvenTimestep);
- getLastCudaError("QPrecursorDeviceCompZeroPress execution failed");
+void QPrecursorDevCompZeroPress(LBMSimulationParameter* parameterDevice,
+ QforPrecursorBoundaryConditions* boundaryCondition,
+ real timeRatio,
+ real velocityRatio)
+{
+
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(parameterDevice->numberofthreads, boundaryCondition->numberOfBCnodes);
+
+ QPrecursorDeviceCompZeroPress<<< grid.grid, grid.threads >>>(boundaryCondition->k,
+ boundaryCondition->numberOfBCnodes,
+ boundaryCondition->numberOfPrecursorNodes,
+ boundaryCondition->sizeQ,
+ parameterDevice->omega,
+ parameterDevice->distributions.f[0],
+ boundaryCondition->q27[0],
+ parameterDevice->neighborX,
+ parameterDevice->neighborY,
+ parameterDevice->neighborZ,
+ boundaryCondition->planeNeighbor0PP,
+ boundaryCondition->planeNeighbor0PM,
+ boundaryCondition->planeNeighbor0MP,
+ boundaryCondition->planeNeighbor0MM,
+ boundaryCondition->weights0PP,
+ boundaryCondition->weights0PM,
+ boundaryCondition->weights0MP,
+ boundaryCondition->weights0MM,
+ boundaryCondition->last,
+ boundaryCondition->current,
+ boundaryCondition->velocityX,
+ boundaryCondition->velocityY,
+ boundaryCondition->velocityZ,
+ timeRatio,
+ velocityRatio,
+ parameterDevice->numberOfNodes,
+ parameterDevice->isEvenTimestep);
+ getLastCudaError("QPrecursorDeviceCompZeroPress execution failed");
}
//////////////////////////////////////////////////////////////////////////
void PrecursorDevEQ27( LBMSimulationParameter* parameterDevice,
- QforPrecursorBoundaryConditions* boundaryCondition,
- real timeRatio,
- real velocityRatio)
-{
-
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(parameterDevice->numberofthreads, boundaryCondition->numberOfBCnodes);
-
- PrecursorDeviceEQ27<<< grid.grid, grid.threads >>>(boundaryCondition->k,
- boundaryCondition->numberOfBCnodes,
- boundaryCondition->numberOfPrecursorNodes,
- parameterDevice->omega,
- parameterDevice->distributions.f[0],
- parameterDevice->neighborX,
- parameterDevice->neighborX,
- parameterDevice->neighborX,
- boundaryCondition->planeNeighbor0PP,
- boundaryCondition->planeNeighbor0PM,
- boundaryCondition->planeNeighbor0MP,
- boundaryCondition->planeNeighbor0MM,
- boundaryCondition->weights0PP,
- boundaryCondition->weights0PM,
- boundaryCondition->weights0MP,
- boundaryCondition->weights0MM,
- boundaryCondition->last,
- boundaryCondition->current,
- boundaryCondition->velocityX,
- boundaryCondition->velocityY,
- boundaryCondition->velocityZ,
- timeRatio,
- velocityRatio,
- parameterDevice->numberOfNodes,
- parameterDevice->isEvenTimestep);
- getLastCudaError("PrecursorDeviceEQ27 execution failed");
+ QforPrecursorBoundaryConditions* boundaryCondition,
+ real timeRatio,
+ real velocityRatio)
+{
+
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(parameterDevice->numberofthreads, boundaryCondition->numberOfBCnodes);
+
+ PrecursorDeviceEQ27<<< grid.grid, grid.threads >>>(boundaryCondition->k,
+ boundaryCondition->numberOfBCnodes,
+ boundaryCondition->numberOfPrecursorNodes,
+ parameterDevice->omega,
+ parameterDevice->distributions.f[0],
+ parameterDevice->neighborX,
+ parameterDevice->neighborX,
+ parameterDevice->neighborX,
+ boundaryCondition->planeNeighbor0PP,
+ boundaryCondition->planeNeighbor0PM,
+ boundaryCondition->planeNeighbor0MP,
+ boundaryCondition->planeNeighbor0MM,
+ boundaryCondition->weights0PP,
+ boundaryCondition->weights0PM,
+ boundaryCondition->weights0MP,
+ boundaryCondition->weights0MM,
+ boundaryCondition->last,
+ boundaryCondition->current,
+ boundaryCondition->velocityX,
+ boundaryCondition->velocityY,
+ boundaryCondition->velocityZ,
+ timeRatio,
+ velocityRatio,
+ parameterDevice->numberOfNodes,
+ parameterDevice->isEvenTimestep);
+ getLastCudaError("PrecursorDeviceEQ27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void PrecursorDevDistributions( LBMSimulationParameter* parameterDevice,
- QforPrecursorBoundaryConditions* boundaryCondition,
- real timeRatio,
- real velocityRatio)
-{
-
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(parameterDevice->numberofthreads, boundaryCondition->numberOfBCnodes);
-
- PrecursorDeviceDistributions<<< grid.grid, grid.threads >>>(boundaryCondition->k,
- boundaryCondition->numberOfBCnodes,
- boundaryCondition->numberOfPrecursorNodes,
- parameterDevice->distributions.f[0],
- parameterDevice->neighborX,
- parameterDevice->neighborY,
- parameterDevice->neighborZ,
- boundaryCondition->planeNeighbor0PP,
- boundaryCondition->planeNeighbor0PM,
- boundaryCondition->planeNeighbor0MP,
- boundaryCondition->planeNeighbor0MM,
- boundaryCondition->weights0PP,
- boundaryCondition->weights0PM,
- boundaryCondition->weights0MP,
- boundaryCondition->weights0MM,
- boundaryCondition->last,
- boundaryCondition->current,
- timeRatio,
- parameterDevice->numberOfNodes,
- parameterDevice->isEvenTimestep);
- getLastCudaError("QPrecursorDeviceCompZeroPress execution failed");
+ QforPrecursorBoundaryConditions* boundaryCondition,
+ real timeRatio,
+ real velocityRatio)
+{
+
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(parameterDevice->numberofthreads, boundaryCondition->numberOfBCnodes);
+
+ PrecursorDeviceDistributions<<< grid.grid, grid.threads >>>(boundaryCondition->k,
+ boundaryCondition->numberOfBCnodes,
+ boundaryCondition->numberOfPrecursorNodes,
+ parameterDevice->distributions.f[0],
+ parameterDevice->neighborX,
+ parameterDevice->neighborY,
+ parameterDevice->neighborZ,
+ boundaryCondition->planeNeighbor0PP,
+ boundaryCondition->planeNeighbor0PM,
+ boundaryCondition->planeNeighbor0MP,
+ boundaryCondition->planeNeighbor0MM,
+ boundaryCondition->weights0PP,
+ boundaryCondition->weights0PM,
+ boundaryCondition->weights0MP,
+ boundaryCondition->weights0MM,
+ boundaryCondition->last,
+ boundaryCondition->current,
+ timeRatio,
+ parameterDevice->numberOfNodes,
+ parameterDevice->isEvenTimestep);
+ getLastCudaError("QPrecursorDeviceCompZeroPress execution failed");
}
//////////////////////////////////////////////////////////////////////////
void QPrecursorDevDistributions( LBMSimulationParameter* parameterDevice,
- QforPrecursorBoundaryConditions* boundaryCondition,
- real timeRatio,
- real velocityRatio)
-{
-
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(parameterDevice->numberofthreads, boundaryCondition->numberOfBCnodes);
-
- QPrecursorDeviceDistributions<<< grid.grid, grid.threads >>>(boundaryCondition->k,
- boundaryCondition->q27[0],
- boundaryCondition->sizeQ,
- boundaryCondition->numberOfBCnodes,
- boundaryCondition->numberOfPrecursorNodes,
- parameterDevice->distributions.f[0],
- parameterDevice->neighborX,
- parameterDevice->neighborY,
- parameterDevice->neighborZ,
- boundaryCondition->planeNeighbor0PP,
- boundaryCondition->planeNeighbor0PM,
- boundaryCondition->planeNeighbor0MP,
- boundaryCondition->planeNeighbor0MM,
- boundaryCondition->weights0PP,
- boundaryCondition->weights0PM,
- boundaryCondition->weights0MP,
- boundaryCondition->weights0MM,
- boundaryCondition->last,
- boundaryCondition->current,
- timeRatio,
- parameterDevice->numberOfNodes,
- parameterDevice->isEvenTimestep);
- getLastCudaError("QPrecursorDeviceCompZeroPress execution failed");
+ QforPrecursorBoundaryConditions* boundaryCondition,
+ real timeRatio,
+ real velocityRatio)
+{
+
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(parameterDevice->numberofthreads, boundaryCondition->numberOfBCnodes);
+
+ QPrecursorDeviceDistributions<<< grid.grid, grid.threads >>>(boundaryCondition->k,
+ boundaryCondition->q27[0],
+ boundaryCondition->sizeQ,
+ boundaryCondition->numberOfBCnodes,
+ boundaryCondition->numberOfPrecursorNodes,
+ parameterDevice->distributions.f[0],
+ parameterDevice->neighborX,
+ parameterDevice->neighborY,
+ parameterDevice->neighborZ,
+ boundaryCondition->planeNeighbor0PP,
+ boundaryCondition->planeNeighbor0PM,
+ boundaryCondition->planeNeighbor0MP,
+ boundaryCondition->planeNeighbor0MM,
+ boundaryCondition->weights0PP,
+ boundaryCondition->weights0PM,
+ boundaryCondition->weights0MP,
+ boundaryCondition->weights0MM,
+ boundaryCondition->last,
+ boundaryCondition->current,
+ timeRatio,
+ parameterDevice->numberOfNodes,
+ parameterDevice->isEvenTimestep);
+ getLastCudaError("QPrecursorDeviceCompZeroPress execution failed");
}
//////////////////////////////////////////////////////////////////////////
@@ -3360,7 +3360,7 @@ extern "C" void PropVelo( unsigned int numberOfThreads,
real* uy,
real* uz,
int* k_Q,
- unsigned int size_Prop,
+ unsigned int size_Prop,
unsigned int size_Mat,
unsigned int* bcMatD,
real* DD,
@@ -3375,10 +3375,10 @@ extern "C" void PropVelo( unsigned int numberOfThreads,
ux,
uy,
uz,
- k_Q,
- size_Prop,
+ k_Q,
+ size_Prop,
size_Mat,
- bcMatD,
+ bcMatD,
DD,
EvenOrOdd);
getLastCudaError("PropellerBC execution failed");
@@ -3408,7 +3408,7 @@ void ScaleCF27( real* DC,
unsigned int numberOfThreads)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kCF);
-
+
scaleCF27<<< grid.grid, grid.threads >>> ( DC,
DF,
neighborCX,
@@ -3638,263 +3638,263 @@ void ScaleCF_Fix_27( real* DC,
}
//////////////////////////////////////////////////////////////////////////
void ScaleCF_Fix_comp_27( real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffCF offCF)
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffCF offCF)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kCF);
scaleCF_Fix_comp_27<<< grid.grid, grid.threads >>>( DC,
- DF,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posCSWB,
- posFSWB,
- kCF,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offCF);
+ DF,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posCSWB,
+ posFSWB,
+ kCF,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offCF);
getLastCudaError("scaleCF_Fix_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void ScaleCF_0817_comp_27(real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffCF offCF,
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffCF offCF,
CUstream_st *stream)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kCF);
scaleCF_0817_comp_27<<< grid.grid, grid.threads, 0, stream >>>( DC,
- DF,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posCSWB,
- posFSWB,
- kCF,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offCF);
+ DF,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posCSWB,
+ posFSWB,
+ kCF,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offCF);
getLastCudaError("scaleCF_0817_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void ScaleCF_comp_D3Q27F3_2018(real* DC,
- real* DF,
- real* G6,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffCF offCF)
+ real* DF,
+ real* G6,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffCF offCF)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kCF);
scaleCF_comp_D3Q27F3_2018 <<< grid.grid, grid.threads >>>(DC,
- DF,
- G6,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posCSWB,
- posFSWB,
- kCF,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offCF);
+ DF,
+ G6,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posCSWB,
+ posFSWB,
+ kCF,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offCF);
getLastCudaError("scaleCF_comp_D3Q27F3_2018 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void ScaleCF_comp_D3Q27F3(real* DC,
- real* DF,
- real* G6,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffCF offCF,
+ real* DF,
+ real* G6,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffCF offCF,
CUstream_st *stream)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kCF);
scaleCF_comp_D3Q27F3 <<< grid.grid, grid.threads, 0, stream >>>( DC,
- DF,
- G6,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posCSWB,
- posFSWB,
- kCF,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offCF);
+ DF,
+ G6,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posCSWB,
+ posFSWB,
+ kCF,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offCF);
getLastCudaError("scaleCF_comp_D3Q27F3 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void ScaleCF_staggered_time_comp_27( real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffCF offCF)
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffCF offCF)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kCF);
scaleCF_staggered_time_comp_27<<< grid.grid, grid.threads >>>( DC,
- DF,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posCSWB,
- posFSWB,
- kCF,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offCF);
+ DF,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posCSWB,
+ posFSWB,
+ kCF,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offCF);
getLastCudaError("scaleCF_Fix_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
@@ -3958,157 +3958,157 @@ void ScaleCF_compressible(LBMSimulationParameter * parameterDeviceC, LBMSimulati
//////////////////////////////////////////////////////////////////////////
void ScaleCF_RhoSq_3rdMom_comp_27(real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffCF offCF,
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffCF offCF,
CUstream_st *stream)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kCF);
scaleCF_RhoSq_3rdMom_comp_27<<< grid.grid, grid.threads, 0, stream >>>( DC,
- DF,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posCSWB,
- posFSWB,
- kCF,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offCF);
+ DF,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posCSWB,
+ posFSWB,
+ kCF,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offCF);
getLastCudaError("scaleCF_RhoSq_3rdMom_comp_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void ScaleCF_AA2016_comp_27(real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffCF offCF,
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffCF offCF,
CUstream_st *stream)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kCF);
scaleCF_AA2016_comp_27<<< grid.grid, grid.threads, 0, stream >>>(DC,
- DF,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posCSWB,
- posFSWB,
- kCF,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offCF);
+ DF,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posCSWB,
+ posFSWB,
+ kCF,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offCF);
getLastCudaError("scaleCF_AA2016_comp_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void ScaleCF_NSPress_27( real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posCSWB,
- unsigned int* posFSWB,
- unsigned int kCF,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffCF offCF)
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posCSWB,
+ unsigned int* posFSWB,
+ unsigned int kCF,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffCF offCF)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kCF);
scaleCF_NSPress_27<<< grid.grid, grid.threads >>>(DC,
- DF,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posCSWB,
- posFSWB,
- kCF,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offCF);
+ DF,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posCSWB,
+ posFSWB,
+ kCF,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offCF);
getLastCudaError("scaleCF_Fix_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
@@ -4219,7 +4219,7 @@ void ScaleCFThS27( real* DC,
real nu,
real diffusivity_fine,
unsigned int numberOfThreads,
- OffCF offCF)
+ OffCF offCF)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kCF);
@@ -4241,7 +4241,7 @@ void ScaleCFThS27( real* DC,
kCF,
nu,
diffusivity_fine,
- offCF);
+ offCF);
getLastCudaError("scaleCFThS27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
@@ -4268,7 +4268,7 @@ void ScaleFC27( real* DC,
unsigned int nyF,
unsigned int numberOfThreads)
{
-
+
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kFC);
scaleFC27<<< grid.grid, grid.threads >>> ( DC,
@@ -4500,263 +4500,263 @@ void ScaleFC_Fix_27(real* DC,
}
//////////////////////////////////////////////////////////////////////////
void ScaleFC_Fix_comp_27( real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffFC offFC)
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffFC offFC)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kFC);
scaleFC_Fix_comp_27<<< grid.grid, grid.threads >>> ( DC,
- DF,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posC,
- posFSWB,
- kFC,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offFC);
+ DF,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posC,
+ posFSWB,
+ kFC,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offFC);
getLastCudaError("scaleFC_Fix_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void ScaleFC_0817_comp_27( real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffFC offFC,
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffFC offFC,
CUstream_st *stream)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kFC);
scaleFC_0817_comp_27<<< grid.grid, grid.threads, 0, stream >>> (DC,
- DF,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posC,
- posFSWB,
- kFC,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offFC);
+ DF,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posC,
+ posFSWB,
+ kFC,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offFC);
getLastCudaError("scaleFC_0817_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void ScaleFC_comp_D3Q27F3_2018( real* DC,
- real* DF,
- real* G6,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffFC offFC)
+ real* DF,
+ real* G6,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffFC offFC)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kFC);
scaleFC_comp_D3Q27F3_2018 <<< grid.grid, grid.threads >>> (DC,
- DF,
- G6,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posC,
- posFSWB,
- kFC,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offFC);
+ DF,
+ G6,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posC,
+ posFSWB,
+ kFC,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offFC);
getLastCudaError("scaleFC_comp_D3Q27F3_2018 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void ScaleFC_comp_D3Q27F3( real* DC,
- real* DF,
- real* G6,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffFC offFC,
+ real* DF,
+ real* G6,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffFC offFC,
CUstream_st *stream)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kFC);
scaleFC_comp_D3Q27F3 <<< grid.grid, grid.threads, 0, stream >>> (DC,
- DF,
- G6,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posC,
- posFSWB,
- kFC,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offFC);
+ DF,
+ G6,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posC,
+ posFSWB,
+ kFC,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offFC);
getLastCudaError("scaleFC_0817_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void ScaleFC_staggered_time_comp_27( real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffFC offFC)
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffFC offFC)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kFC);
scaleFC_staggered_time_comp_27<<< grid.grid, grid.threads >>> ( DC,
- DF,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posC,
- posFSWB,
- kFC,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offFC);
+ DF,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posC,
+ posFSWB,
+ kFC,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offFC);
getLastCudaError("scaleFC_Fix_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
@@ -4818,157 +4818,157 @@ void ScaleFC_compressible(LBMSimulationParameter * parameterDeviceC, LBMSimulati
}
//////////////////////////////////////////////////////////////////////////
void ScaleFC_RhoSq_3rdMom_comp_27( real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffFC offFC,
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffFC offFC,
CUstream_st *stream)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kFC);
scaleFC_RhoSq_3rdMom_comp_27<<< grid.grid, grid.threads, 0, stream >>>(DC,
- DF,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posC,
- posFSWB,
- kFC,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offFC);
+ DF,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posC,
+ posFSWB,
+ kFC,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offFC);
getLastCudaError("scaleFC_RhoSq_3rdMom_comp_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void ScaleFC_AA2016_comp_27( real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffFC offFC,
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffFC offFC,
CUstream_st *stream)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kFC);
scaleFC_AA2016_comp_27<<< grid.grid, grid.threads, 0, stream >>>(DC,
- DF,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posC,
- posFSWB,
- kFC,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offFC);
+ DF,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posC,
+ posFSWB,
+ kFC,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offFC);
getLastCudaError("scaleFC_AA2016_comp_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void ScaleFC_NSPress_27(real* DC,
- real* DF,
- unsigned int* neighborCX,
- unsigned int* neighborCY,
- unsigned int* neighborCZ,
- unsigned int* neighborFX,
- unsigned int* neighborFY,
- unsigned int* neighborFZ,
- unsigned int size_MatC,
- unsigned int size_MatF,
- bool isEvenTimestep,
- unsigned int* posC,
- unsigned int* posFSWB,
- unsigned int kFC,
- real omCoarse,
- real omFine,
- real nu,
- unsigned int nxC,
- unsigned int nyC,
- unsigned int nxF,
- unsigned int nyF,
- unsigned int numberOfThreads,
- OffFC offFC)
+ real* DF,
+ unsigned int* neighborCX,
+ unsigned int* neighborCY,
+ unsigned int* neighborCZ,
+ unsigned int* neighborFX,
+ unsigned int* neighborFY,
+ unsigned int* neighborFZ,
+ unsigned int size_MatC,
+ unsigned int size_MatF,
+ bool isEvenTimestep,
+ unsigned int* posC,
+ unsigned int* posFSWB,
+ unsigned int kFC,
+ real omCoarse,
+ real omFine,
+ real nu,
+ unsigned int nxC,
+ unsigned int nyC,
+ unsigned int nxF,
+ unsigned int nyF,
+ unsigned int numberOfThreads,
+ OffFC offFC)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kFC);
scaleFC_NSPress_27<<< grid.grid, grid.threads >>> ( DC,
- DF,
- neighborCX,
- neighborCY,
- neighborCZ,
- neighborFX,
- neighborFY,
- neighborFZ,
- size_MatC,
- size_MatF,
- isEvenTimestep,
- posC,
- posFSWB,
- kFC,
- omCoarse,
- omFine,
- nu,
- nxC,
- nyC,
- nxF,
- nyF,
- offFC);
+ DF,
+ neighborCX,
+ neighborCY,
+ neighborCZ,
+ neighborFX,
+ neighborFY,
+ neighborFZ,
+ size_MatC,
+ size_MatF,
+ isEvenTimestep,
+ posC,
+ posFSWB,
+ kFC,
+ omCoarse,
+ omFine,
+ nu,
+ nxC,
+ nyC,
+ nxF,
+ nyF,
+ offFC);
getLastCudaError("scaleFC_Fix_27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
@@ -5079,7 +5079,7 @@ void ScaleFCThS27( real* DC,
real nu,
real diffusivity_coarse,
unsigned int numberOfThreads,
- OffFC offFC)
+ OffFC offFC)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, kFC);
@@ -5101,501 +5101,501 @@ void ScaleFCThS27( real* DC,
kFC,
nu,
diffusivity_coarse,
- offFC);
+ offFC);
getLastCudaError("scaleFCThS27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void DragLiftPostD27(real* DD,
- int* k_Q,
- real* QQ,
- int numberOfBCnodes,
- double *DragX,
- double *DragY,
- double *DragZ,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep,
- unsigned int numberOfThreads)
+ int* k_Q,
+ real* QQ,
+ int numberOfBCnodes,
+ double *DragX,
+ double *DragY,
+ double *DragZ,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep,
+ unsigned int numberOfThreads)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
- DragLiftPost27<<< grid.grid, grid.threads >>>(DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- DragX,
- DragY,
- DragZ,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("DragLift27 execution failed");
+ DragLiftPost27<<< grid.grid, grid.threads >>>(DD,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ DragX,
+ DragY,
+ DragZ,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("DragLift27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void DragLiftPreD27( real* DD,
- int* k_Q,
- real* QQ,
- int numberOfBCnodes,
- double *DragX,
- double *DragY,
- double *DragZ,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep,
- unsigned int numberOfThreads)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
-
- DragLiftPre27<<< grid.grid, grid.threads >>>( DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- DragX,
- DragY,
- DragZ,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("DragLift27 execution failed");
+ int* k_Q,
+ real* QQ,
+ int numberOfBCnodes,
+ double *DragX,
+ double *DragY,
+ double *DragZ,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep,
+ unsigned int numberOfThreads)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
+
+ DragLiftPre27<<< grid.grid, grid.threads >>>( DD,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ DragX,
+ DragY,
+ DragZ,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("DragLift27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CalcCPtop27(real* DD,
- int* cpIndex,
- int nonCp,
- double *cpPress,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep,
- unsigned int numberOfThreads)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, nonCp);
-
- CalcCP27<<< grid.grid, grid.threads >>>(DD,
- cpIndex,
- nonCp,
- cpPress,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("CalcCP27 execution failed");
+ int* cpIndex,
+ int nonCp,
+ double *cpPress,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep,
+ unsigned int numberOfThreads)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, nonCp);
+
+ CalcCP27<<< grid.grid, grid.threads >>>(DD,
+ cpIndex,
+ nonCp,
+ cpPress,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("CalcCP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void CalcCPbottom27( real* DD,
- int* cpIndex,
- int nonCp,
- double *cpPress,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep,
- unsigned int numberOfThreads)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, nonCp);
-
- CalcCP27<<< grid.grid, grid.threads >>>(DD,
- cpIndex,
- nonCp,
- cpPress,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("CalcCP27 execution failed");
+ int* cpIndex,
+ int nonCp,
+ double *cpPress,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep,
+ unsigned int numberOfThreads)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, nonCp);
+
+ CalcCP27<<< grid.grid, grid.threads >>>(DD,
+ cpIndex,
+ nonCp,
+ cpPress,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("CalcCP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void GetSendFsPreDev27(real* DD,
- real* bufferFs,
- int* sendIndex,
- int buffmax,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep,
- unsigned int numberOfThreads,
- cudaStream_t stream)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, buffmax);
-
- getSendFsPre27<<< grid.grid, grid.threads, 0, stream >>>(DD,
- bufferFs,
- sendIndex,
- buffmax,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("getSendFsPre27 execution failed");
+ real* bufferFs,
+ int* sendIndex,
+ int buffmax,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep,
+ unsigned int numberOfThreads,
+ cudaStream_t stream)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, buffmax);
+
+ getSendFsPre27<<< grid.grid, grid.threads, 0, stream >>>(DD,
+ bufferFs,
+ sendIndex,
+ buffmax,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("getSendFsPre27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void GetSendFsPostDev27(real* DD,
- real* bufferFs,
- int* sendIndex,
- int buffmax,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep,
- unsigned int numberOfThreads,
- cudaStream_t stream)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, buffmax);
-
- getSendFsPost27<<< grid.grid, grid.threads, 0, stream >>>(DD,
- bufferFs,
- sendIndex,
- buffmax,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("getSendFsPost27 execution failed");
+ real* bufferFs,
+ int* sendIndex,
+ int buffmax,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep,
+ unsigned int numberOfThreads,
+ cudaStream_t stream)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, buffmax);
+
+ getSendFsPost27<<< grid.grid, grid.threads, 0, stream >>>(DD,
+ bufferFs,
+ sendIndex,
+ buffmax,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("getSendFsPost27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void SetRecvFsPreDev27(real* DD,
- real* bufferFs,
- int* recvIndex,
- int buffmax,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep,
- unsigned int numberOfThreads,
- cudaStream_t stream)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, buffmax);
-
- setRecvFsPre27<<< grid.grid, grid.threads, 0, stream >>>(DD,
- bufferFs,
- recvIndex,
- buffmax,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("setRecvFsPre27 execution failed");
+ real* bufferFs,
+ int* recvIndex,
+ int buffmax,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep,
+ unsigned int numberOfThreads,
+ cudaStream_t stream)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, buffmax);
+
+ setRecvFsPre27<<< grid.grid, grid.threads, 0, stream >>>(DD,
+ bufferFs,
+ recvIndex,
+ buffmax,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("setRecvFsPre27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void SetRecvFsPostDev27(real* DD,
- real* bufferFs,
- int* recvIndex,
- int buffmax,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep,
- unsigned int numberOfThreads,
- cudaStream_t stream)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, buffmax);
-
- setRecvFsPost27<<< grid.grid, grid.threads, 0, stream >>>(DD,
- bufferFs,
- recvIndex,
- buffmax,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("setRecvFsPost27 execution failed");
+ real* bufferFs,
+ int* recvIndex,
+ int buffmax,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep,
+ unsigned int numberOfThreads,
+ cudaStream_t stream)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, buffmax);
+
+ setRecvFsPost27<<< grid.grid, grid.threads, 0, stream >>>(DD,
+ bufferFs,
+ recvIndex,
+ buffmax,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("setRecvFsPost27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void getSendGsDevF3(
- real* G6,
- real* bufferGs,
- int* sendIndex,
- int buffmax,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep,
- unsigned int numberOfThreads)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, buffmax);
-
- getSendGsF3 <<< grid.grid, grid.threads >>> (
- G6,
- bufferGs,
- sendIndex,
- buffmax,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("getSendGsF3 execution failed");
+ real* G6,
+ real* bufferGs,
+ int* sendIndex,
+ int buffmax,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep,
+ unsigned int numberOfThreads)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, buffmax);
+
+ getSendGsF3 <<< grid.grid, grid.threads >>> (
+ G6,
+ bufferGs,
+ sendIndex,
+ buffmax,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("getSendGsF3 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void setRecvGsDevF3(
- real* G6,
- real* bufferGs,
- int* recvIndex,
- int buffmax,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep,
- unsigned int numberOfThreads)
-{
- vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, buffmax);
-
- setRecvGsF3 <<< grid.grid, grid.threads >>> (
- G6,
- bufferGs,
- recvIndex,
- buffmax,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
- getLastCudaError("setRecvGsF3 execution failed");
+ real* G6,
+ real* bufferGs,
+ int* recvIndex,
+ int buffmax,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep,
+ unsigned int numberOfThreads)
+{
+ vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, buffmax);
+
+ setRecvGsF3 <<< grid.grid, grid.threads >>> (
+ G6,
+ bufferGs,
+ recvIndex,
+ buffmax,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
+ getLastCudaError("setRecvGsF3 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void WallFuncDev27(unsigned int numberOfThreads,
- real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* vx,
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfBCnodes);
WallFunction27<<< grid.grid, grid.threads >>> (
- vx,
- vy,
- vz,
- DD,
- k_Q,
- QQ,
- numberOfBCnodes,
- om1,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ vx,
+ vy,
+ vz,
+ DD,
+ k_Q,
+ QQ,
+ numberOfBCnodes,
+ om1,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("WallFunction27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void SetOutputWallVelocitySP27(unsigned int numberOfThreads,
- real* vxD,
- real* vyD,
- real* vzD,
- real* vxWall,
- real* vyWall,
- real* vzWall,
- int numberOfWallNodes,
- int* kWallNodes,
- real* rhoD,
- real* pressD,
- unsigned int* geoD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- real* DD,
- bool isEvenTimestep)
+ real* vxD,
+ real* vyD,
+ real* vzD,
+ real* vxWall,
+ real* vyWall,
+ real* vzWall,
+ int numberOfWallNodes,
+ int* kWallNodes,
+ real* rhoD,
+ real* pressD,
+ unsigned int* geoD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ real* DD,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfWallNodes);
LBSetOutputWallVelocitySP27<<< grid.grid, grid.threads >>> ( vxD,
- vyD,
- vzD,
- vxWall,
- vyWall,
- vzWall,
- numberOfWallNodes,
- kWallNodes,
- rhoD,
- pressD,
- geoD,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- DD,
- isEvenTimestep);
+ vyD,
+ vzD,
+ vxWall,
+ vyWall,
+ vzWall,
+ numberOfWallNodes,
+ kWallNodes,
+ rhoD,
+ pressD,
+ geoD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ DD,
+ isEvenTimestep);
getLastCudaError("LBSetOutputWallVelocitySP27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void GetVelotoForce27(unsigned int numberOfThreads,
- real* DD,
- int* bcIndex,
- int nonAtBC,
- real* Vx,
- real* Vy,
- real* Vz,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ int* bcIndex,
+ int nonAtBC,
+ real* Vx,
+ real* Vy,
+ real* Vz,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, nonAtBC);
GetVeloforForcing27<<< grid.grid, grid.threads >>> (DD,
- bcIndex,
- nonAtBC,
- Vx,
- Vy,
- Vz,
- neighborX,
- neighborY,
- neighborZ,
- size_Mat,
- isEvenTimestep);
+ bcIndex,
+ nonAtBC,
+ Vx,
+ Vy,
+ Vz,
+ neighborX,
+ neighborY,
+ neighborZ,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("GetVeloforForcing27 execution failed");
}
//////////////////////////////////////////////////////////////////////////
void InitParticlesDevice(real* coordX,
- real* coordY,
- real* coordZ,
- real* coordParticleXlocal,
- real* coordParticleYlocal,
- real* coordParticleZlocal,
- real* coordParticleXglobal,
- real* coordParticleYglobal,
- real* coordParticleZglobal,
- real* veloParticleX,
- real* veloParticleY,
- real* veloParticleZ,
- real* randArray,
- unsigned int* particleID,
- unsigned int* cellBaseID,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int* neighborWSB,
- int level,
- unsigned int numberOfParticles,
- unsigned int size_Mat,
- unsigned int numberOfThreads)
+ real* coordY,
+ real* coordZ,
+ real* coordParticleXlocal,
+ real* coordParticleYlocal,
+ real* coordParticleZlocal,
+ real* coordParticleXglobal,
+ real* coordParticleYglobal,
+ real* coordParticleZglobal,
+ real* veloParticleX,
+ real* veloParticleY,
+ real* veloParticleZ,
+ real* randArray,
+ unsigned int* particleID,
+ unsigned int* cellBaseID,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int* neighborWSB,
+ int level,
+ unsigned int numberOfParticles,
+ unsigned int size_Mat,
+ unsigned int numberOfThreads)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfParticles);
InitParticles<<< grid.grid, grid.threads >>> (coordX,
- coordY,
- coordZ,
- coordParticleXlocal,
- coordParticleYlocal,
- coordParticleZlocal,
- coordParticleXglobal,
- coordParticleYglobal,
- coordParticleZglobal,
- veloParticleX,
- veloParticleY,
- veloParticleZ,
- randArray,
- particleID,
- cellBaseID,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- neighborWSB,
- level,
- numberOfParticles,
- size_Mat);
+ coordY,
+ coordZ,
+ coordParticleXlocal,
+ coordParticleYlocal,
+ coordParticleZlocal,
+ coordParticleXglobal,
+ coordParticleYglobal,
+ coordParticleZglobal,
+ veloParticleX,
+ veloParticleY,
+ veloParticleZ,
+ randArray,
+ particleID,
+ cellBaseID,
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ neighborWSB,
+ level,
+ numberOfParticles,
+ size_Mat);
getLastCudaError("InitParticles execution failed");
}
//////////////////////////////////////////////////////////////////////////
void MoveParticlesDevice(real* coordX,
- real* coordY,
- real* coordZ,
- real* coordParticleXlocal,
- real* coordParticleYlocal,
- real* coordParticleZlocal,
- real* coordParticleXglobal,
- real* coordParticleYglobal,
- real* coordParticleZglobal,
- real* veloParticleX,
- real* veloParticleY,
- real* veloParticleZ,
- real* DD,
- real omega,
- unsigned int* particleID,
- unsigned int* cellBaseID,
- unsigned int* bcMatD,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int* neighborWSB,
- int level,
- unsigned int timestep,
- unsigned int numberOfTimesteps,
- unsigned int numberOfParticles,
- unsigned int size_Mat,
- unsigned int numberOfThreads,
- bool isEvenTimestep)
+ real* coordY,
+ real* coordZ,
+ real* coordParticleXlocal,
+ real* coordParticleYlocal,
+ real* coordParticleZlocal,
+ real* coordParticleXglobal,
+ real* coordParticleYglobal,
+ real* coordParticleZglobal,
+ real* veloParticleX,
+ real* veloParticleY,
+ real* veloParticleZ,
+ real* DD,
+ real omega,
+ unsigned int* particleID,
+ unsigned int* cellBaseID,
+ unsigned int* bcMatD,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int* neighborWSB,
+ int level,
+ unsigned int timestep,
+ unsigned int numberOfTimesteps,
+ unsigned int numberOfParticles,
+ unsigned int size_Mat,
+ unsigned int numberOfThreads,
+ bool isEvenTimestep)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, numberOfParticles);
MoveParticles<<< grid.grid, grid.threads >>> (coordX,
- coordY,
- coordZ,
- coordParticleXlocal,
- coordParticleYlocal,
- coordParticleZlocal,
- coordParticleXglobal,
- coordParticleYglobal,
- coordParticleZglobal,
- veloParticleX,
- veloParticleY,
- veloParticleZ,
- DD,
- omega,
- particleID,
- cellBaseID,
- bcMatD,
- neighborX,
- neighborY,
- neighborZ,
- neighborWSB,
- level,
- timestep,
- numberOfTimesteps,
- numberOfParticles,
- size_Mat,
- isEvenTimestep);
+ coordY,
+ coordZ,
+ coordParticleXlocal,
+ coordParticleYlocal,
+ coordParticleZlocal,
+ coordParticleXglobal,
+ coordParticleYglobal,
+ coordParticleZglobal,
+ veloParticleX,
+ veloParticleY,
+ veloParticleZ,
+ DD,
+ omega,
+ particleID,
+ cellBaseID,
+ bcMatD,
+ neighborX,
+ neighborY,
+ neighborZ,
+ neighborWSB,
+ level,
+ timestep,
+ numberOfTimesteps,
+ numberOfParticles,
+ size_Mat,
+ isEvenTimestep);
getLastCudaError("MoveParticles execution failed");
}
//////////////////////////////////////////////////////////////////////////
void initRandomDevice(curandState* state,
- unsigned int size_Mat,
- unsigned int numberOfThreads)
+ unsigned int size_Mat,
+ unsigned int numberOfThreads)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
initRandom<<< grid.grid, grid.threads >>> (state);
@@ -5603,9 +5603,9 @@ void initRandomDevice(curandState* state,
}
//////////////////////////////////////////////////////////////////////////
void generateRandomValuesDevice( curandState* state,
- unsigned int size_Mat,
- real* randArray,
- unsigned int numberOfThreads)
+ unsigned int size_Mat,
+ real* randArray,
+ unsigned int numberOfThreads)
{
vf::cuda::CudaGrid grid = vf::cuda::CudaGrid(numberOfThreads, size_Mat);
generateRandomValues<<< grid.grid, grid.threads >>> (state,randArray);
@@ -5636,7 +5636,7 @@ void CalcTurbulenceIntensityDevice(
vxx,
vyy,
vzz,
- vxy,
+ vxy,
vxz,
vyz,
vx_mean,
@@ -5652,16 +5652,3 @@ void CalcTurbulenceIntensityDevice(
getLastCudaError("CalcTurbulenceIntensity execution failed");
}
-
-
-
-
-
-
-
-
-
-
-
-
-
diff --git a/src/gpu/VirtualFluids_GPU/GPU/PrecursorBCs27.cu b/src/gpu/VirtualFluids_GPU/GPU/PrecursorBCs27.cu
index 78b190e37ebaa395c89aae3b47cd4cc4f3147306..b17ffefd13a8a3a6048dde69ffb1db6c5def23e1 100644
--- a/src/gpu/VirtualFluids_GPU/GPU/PrecursorBCs27.cu
+++ b/src/gpu/VirtualFluids_GPU/GPU/PrecursorBCs27.cu
@@ -1,4 +1,4 @@
-#include "LBM/LB.h"
+#include "LBM/LB.h"
#include <lbm/constants/NumericConstants.h>
#include <lbm/constants/D3Q27.h>
#include <lbm/MacroscopicQuantities.h>
@@ -16,18 +16,18 @@ __global__ void QPrecursorDeviceCompZeroPress( int* subgridDistanceIndices,
real omega,
real* distributions,
real* subgridDistances,
- uint* neighborX,
- uint* neighborY,
+ uint* neighborX,
+ uint* neighborY,
uint* neighborZ,
- uint* neighbors0PP,
+ uint* neighbors0PP,
uint* neighbors0PM,
uint* neighbors0MP,
uint* neighbors0MM,
- real* weights0PP,
+ real* weights0PP,
real* weights0PM,
real* weights0MP,
real* weights0MM,
- real* vLast,
+ real* vLast,
real* vCurrent,
real velocityX,
real velocityY,
@@ -43,8 +43,8 @@ __global__ void QPrecursorDeviceCompZeroPress( int* subgridDistanceIndices,
////////////////////////////////////////////////////////////////////////////////
// interpolation of velocity
- real vxLastInterpd, vyLastInterpd, vzLastInterpd;
- real vxNextInterpd, vyNextInterpd, vzNextInterpd;
+ real vxLastInterpd, vyLastInterpd, vzLastInterpd;
+ real vxNextInterpd, vyNextInterpd, vzNextInterpd;
uint kNeighbor0PP = neighbors0PP[k];
real d0PP = weights0PP[k];
@@ -90,7 +90,7 @@ __global__ void QPrecursorDeviceCompZeroPress( int* subgridDistanceIndices,
// if(k==16300)s printf("%f %f %f\n", vxLastInterpd, vyLastInterpd, vzLastInterpd);
real VeloX = (velocityX + (1.f-timeRatio)*vxLastInterpd + timeRatio*vxNextInterpd)/velocityRatio;
- real VeloY = (velocityY + (1.f-timeRatio)*vyLastInterpd + timeRatio*vyNextInterpd)/velocityRatio;
+ real VeloY = (velocityY + (1.f-timeRatio)*vyLastInterpd + timeRatio*vyNextInterpd)/velocityRatio;
real VeloZ = (velocityZ + (1.f-timeRatio)*vzLastInterpd + timeRatio*vzNextInterpd)/velocityRatio;
// From here on just a copy of QVelDeviceCompZeroPress
////////////////////////////////////////////////////////////////////////////////
@@ -156,29 +156,29 @@ __global__ void QPrecursorDeviceCompZeroPress( int* subgridDistanceIndices,
real f_PPP = (dist.f[DIR_MMM])[kMMM];
real f_MPP = (dist.f[DIR_PMM])[kPMM];
real f_PMP = (dist.f[DIR_MPM])[kMPM];
-
+
SubgridDistances27 subgridD;
getPointersToSubgridDistances(subgridD, subgridDistances, numberOfBCnodes);
-
+
////////////////////////////////////////////////////////////////////////////////
real drho = f_PMP + f_MPP + f_PPP + f_MMP + f_PMM + f_MPM + f_PPM + f_MMM +
- f_0PM + f_0PP + f_0MP + f_0MM + f_P0M + f_M0P + f_P0P + f_M0M + f_PM0 + f_MP0 + f_PP0 + f_MM0 +
- f_00P + f_00M + f_0P0 + f_0M0 + f_P00 + f_M00 + ((dist.f[DIR_000])[k000]);
+ f_0PM + f_0PP + f_0MP + f_0MM + f_P0M + f_M0P + f_P0P + f_M0M + f_PM0 + f_MP0 + f_PP0 + f_MM0 +
+ f_00P + f_00M + f_0P0 + f_0M0 + f_P00 + f_M00 + ((dist.f[DIR_000])[k000]);
real vx1 = (((f_PMP - f_MPM) - (f_MPP - f_PMM)) + ((f_PPP - f_MMM) - (f_MMP - f_PPM)) +
((f_P0M - f_M0P) + (f_P0P - f_M0M)) + ((f_PM0 - f_MP0) + (f_PP0 - f_MM0)) +
- (f_P00 - f_M00)) / (c1o1 + drho);
-
+ (f_P00 - f_M00)) / (c1o1 + drho);
+
real vx2 = ((-(f_PMP - f_MPM) + (f_MPP - f_PMM)) + ((f_PPP - f_MMM) - (f_MMP - f_PPM)) +
((f_0PM - f_0MP) + (f_0PP - f_0MM)) + (-(f_PM0 - f_MP0) + (f_PP0 - f_MM0)) +
- (f_0P0 - f_0M0)) / (c1o1 + drho);
+ (f_0P0 - f_0M0)) / (c1o1 + drho);
real vx3 = (((f_PMP - f_MPM) + (f_MPP - f_PMM)) + ((f_PPP - f_MMM) + (f_MMP - f_PPM)) +
(-(f_0PM - f_0MP) + (f_0PP - f_0MM)) + ((f_P0P - f_M0M) - (f_P0M - f_M0P)) +
- (f_00P - f_00M)) / (c1o1 + drho);
+ (f_00P - f_00M)) / (c1o1 + drho);
+
-
// if(k==16383 || k==0) printf("k %d kQ %d drho = %f u %f v %f w %f\n",k, KQK, drho, vx1, vx2, vx3);
real cu_sq=c3o2*(vx1*vx1+vx2*vx2+vx3*vx3) * (c1o1 + drho);
//////////////////////////////////////////////////////////////////////////
@@ -429,18 +429,18 @@ __global__ void PrecursorDeviceEQ27( int* subgridDistanceIndices,
int numberOfPrecursorNodes,
real omega,
real* distributions,
- uint* neighborX,
- uint* neighborY,
+ uint* neighborX,
+ uint* neighborY,
uint* neighborZ,
- uint* neighbors0PP,
+ uint* neighbors0PP,
uint* neighbors0PM,
uint* neighbors0MP,
uint* neighbors0MM,
- real* weights0PP,
+ real* weights0PP,
real* weights0PM,
real* weights0MP,
real* weights0MM,
- real* vLast,
+ real* vLast,
real* vCurrent,
real velocityX,
real velocityY,
@@ -456,8 +456,8 @@ __global__ void PrecursorDeviceEQ27( int* subgridDistanceIndices,
////////////////////////////////////////////////////////////////////////////////
// interpolation of velocity
- real vxLastInterpd, vyLastInterpd, vzLastInterpd;
- real vxNextInterpd, vyNextInterpd, vzNextInterpd;
+ real vxLastInterpd, vyLastInterpd, vzLastInterpd;
+ real vxNextInterpd, vyNextInterpd, vzNextInterpd;
uint kNeighbor0PP = neighbors0PP[k];
real d0PP = weights0PP[k];
@@ -503,7 +503,7 @@ __global__ void PrecursorDeviceEQ27( int* subgridDistanceIndices,
// if(k==16300) printf("%f %f %f\n", vxLastInterpd, vyLastInterpd, vzLastInterpd);
real VeloX = (velocityX + (1.f-timeRatio)*vxLastInterpd + timeRatio*vxNextInterpd)/velocityRatio;
- real VeloY = (velocityY + (1.f-timeRatio)*vyLastInterpd + timeRatio*vyNextInterpd)/velocityRatio;
+ real VeloY = (velocityY + (1.f-timeRatio)*vyLastInterpd + timeRatio*vyNextInterpd)/velocityRatio;
real VeloZ = (velocityZ + (1.f-timeRatio)*vzLastInterpd + timeRatio*vzNextInterpd)/velocityRatio;
// From here on just a copy of QVelDeviceCompZeroPress
////////////////////////////////////////////////////////////////////////////////
@@ -511,26 +511,26 @@ __global__ void PrecursorDeviceEQ27( int* subgridDistanceIndices,
Distributions27 dist;
getPointersToDistributions(dist, distributions, numberOfLBnodes, !isEvenTimestep);
- unsigned int KQK = subgridDistanceIndices[k]; //QK
+ unsigned int KQK = subgridDistanceIndices[k]; //QK
unsigned int k000 = KQK; //000
unsigned int kP00 = KQK; //P00
unsigned int kM00 = neighborX[KQK]; //M00
- unsigned int k0P0 = KQK; //n
- unsigned int k0M0 = neighborY[KQK]; //s
- unsigned int k00P = KQK; //t
- unsigned int k00M = neighborZ[KQK]; //b
- unsigned int kMM0 = neighborY[kM00]; //sw
- unsigned int kPP0 = KQK; //ne
- unsigned int kPM0 = k0M0; //se
- unsigned int kMP0 = kM00; //nw
- unsigned int kM0M = neighborZ[kM00]; //bw
- unsigned int kP0P = KQK; //te
- unsigned int kP0M = k00M; //be
- unsigned int k0PP = KQK; //tn
- unsigned int k0MM = neighborZ[k0M0]; //bs
- unsigned int kM0P = kM00; //tw
- unsigned int k0PM = k00M; //bn
- unsigned int k0MP = k0M0; //ts
+ unsigned int k0P0 = KQK; //n
+ unsigned int k0M0 = neighborY[KQK]; //s
+ unsigned int k00P = KQK; //t
+ unsigned int k00M = neighborZ[KQK]; //b
+ unsigned int kMM0 = neighborY[kM00]; //sw
+ unsigned int kPP0 = KQK; //ne
+ unsigned int kPM0 = k0M0; //se
+ unsigned int kMP0 = kM00; //nw
+ unsigned int kM0M = neighborZ[kM00]; //bw
+ unsigned int kP0P = KQK; //te
+ unsigned int kP0M = k00M; //be
+ unsigned int k0PP = KQK; //tn
+ unsigned int k0MM = neighborZ[k0M0]; //bs
+ unsigned int kM0P = kM00; //tw
+ unsigned int k0PM = k00M; //bn
+ unsigned int k0MP = k0M0; //ts
unsigned int kPMP = k0M0; //tse
unsigned int kMPM = kM0M; //bnw
unsigned int kMPP = kM00; //tnw
@@ -576,11 +576,11 @@ __global__ void PrecursorDeviceEQ27( int* subgridDistanceIndices,
//!
real drho = c0o1;
- real vx1 = VeloX;
+ real vx1 = VeloX;
- real vx2 = VeloY;
+ real vx2 = VeloY;
- real vx3 = VeloZ;
+ real vx3 = VeloZ;
real cusq = c3o2 * (vx1 * vx1 + vx2 * vx2 + vx3 * vx3);
@@ -625,7 +625,7 @@ __global__ void PrecursorDeviceEQ27( int* subgridDistanceIndices,
(dist.f[DIR_PPM])[kPPM] = f_MMP;
(dist.f[DIR_PPP])[kPPP] = f_MMM;
(dist.f[DIR_PMM])[kPMM] = f_MPP;
-
+
(dist.f[DIR_M00])[kM00] = f_P00;
(dist.f[DIR_MM0])[kMM0] = f_PP0;
(dist.f[DIR_M0M])[kM0M] = f_P0P;
@@ -650,25 +650,25 @@ __global__ void PrecursorDeviceEQ27( int* subgridDistanceIndices,
__global__ void PrecursorDeviceDistributions( int* subgridDistanceIndices,
- int numberOfBCnodes,
+ int numberOfBCnodes,
int numberOfPrecursorNodes,
- real* distributions,
- uint* neighborX,
- uint* neighborY,
- uint* neighborZ,
- uint* neighbors0PP,
- uint* neighbors0PM,
- uint* neighbors0MP,
- uint* neighbors0MM,
- real* weights0PP,
- real* weights0PM,
- real* weights0MP,
- real* weights0MM,
- real* fsLast,
- real* fsNext,
- real timeRatio,
- unsigned long long numberOfLBnodes,
- bool isEvenTimestep)
+ real* distributions,
+ uint* neighborX,
+ uint* neighborY,
+ uint* neighborZ,
+ uint* neighbors0PP,
+ uint* neighbors0PM,
+ uint* neighbors0MP,
+ uint* neighbors0MM,
+ real* weights0PP,
+ real* weights0PM,
+ real* weights0MP,
+ real* weights0MM,
+ real* fsLast,
+ real* fsNext,
+ real timeRatio,
+ unsigned long long numberOfLBnodes,
+ bool isEvenTimestep)
{
const unsigned k = vf::gpu::getNodeIndex();
@@ -715,31 +715,31 @@ __global__ void PrecursorDeviceDistributions( int* subgridDistanceIndices,
f0LastInterp = (f0Last[kNeighbor0PP]*d0PP + f0Last[kNeighbor0PM]*d0PM + f0Last[kNeighbor0MP]*d0MP + f0Last[kNeighbor0MM]*d0MM)*invWeightSum;
f0NextInterp = (f0Next[kNeighbor0PP]*d0PP + f0Next[kNeighbor0PM]*d0PM + f0Next[kNeighbor0MP]*d0MP + f0Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f1LastInterp = (f1Last[kNeighbor0PP]*d0PP + f1Last[kNeighbor0PM]*d0PM + f1Last[kNeighbor0MP]*d0MP + f1Last[kNeighbor0MM]*d0MM)*invWeightSum;
f1NextInterp = (f1Next[kNeighbor0PP]*d0PP + f1Next[kNeighbor0PM]*d0PM + f1Next[kNeighbor0MP]*d0MP + f1Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f2LastInterp = (f2Last[kNeighbor0PP]*d0PP + f2Last[kNeighbor0PM]*d0PM + f2Last[kNeighbor0MP]*d0MP + f2Last[kNeighbor0MM]*d0MM)*invWeightSum;
f2NextInterp = (f2Next[kNeighbor0PP]*d0PP + f2Next[kNeighbor0PM]*d0PM + f2Next[kNeighbor0MP]*d0MP + f2Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f3LastInterp = (f3Last[kNeighbor0PP]*d0PP + f3Last[kNeighbor0PM]*d0PM + f3Last[kNeighbor0MP]*d0MP + f3Last[kNeighbor0MM]*d0MM)*invWeightSum;
f3NextInterp = (f3Next[kNeighbor0PP]*d0PP + f3Next[kNeighbor0PM]*d0PM + f3Next[kNeighbor0MP]*d0MP + f3Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f4LastInterp = (f4Last[kNeighbor0PP]*d0PP + f4Last[kNeighbor0PM]*d0PM + f4Last[kNeighbor0MP]*d0MP + f4Last[kNeighbor0MM]*d0MM)*invWeightSum;
f4NextInterp = (f4Next[kNeighbor0PP]*d0PP + f4Next[kNeighbor0PM]*d0PM + f4Next[kNeighbor0MP]*d0MP + f4Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f5LastInterp = (f5Last[kNeighbor0PP]*d0PP + f5Last[kNeighbor0PM]*d0PM + f5Last[kNeighbor0MP]*d0MP + f5Last[kNeighbor0MM]*d0MM)*invWeightSum;
f5NextInterp = (f5Next[kNeighbor0PP]*d0PP + f5Next[kNeighbor0PM]*d0PM + f5Next[kNeighbor0MP]*d0MP + f5Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f6LastInterp = (f6Last[kNeighbor0PP]*d0PP + f6Last[kNeighbor0PM]*d0PM + f6Last[kNeighbor0MP]*d0MP + f6Last[kNeighbor0MM]*d0MM)*invWeightSum;
f6NextInterp = (f6Next[kNeighbor0PP]*d0PP + f6Next[kNeighbor0PM]*d0PM + f6Next[kNeighbor0MP]*d0MP + f6Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f7LastInterp = (f7Last[kNeighbor0PP]*d0PP + f7Last[kNeighbor0PM]*d0PM + f7Last[kNeighbor0MP]*d0MP + f7Last[kNeighbor0MM]*d0MM)*invWeightSum;
f7NextInterp = (f7Next[kNeighbor0PP]*d0PP + f7Next[kNeighbor0PM]*d0PM + f7Next[kNeighbor0MP]*d0MP + f7Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f8LastInterp = (f8Last[kNeighbor0PP]*d0PP + f8Last[kNeighbor0PM]*d0PM + f8Last[kNeighbor0MP]*d0MP + f8Last[kNeighbor0MM]*d0MM)*invWeightSum;
f8NextInterp = (f8Next[kNeighbor0PP]*d0PP + f8Next[kNeighbor0PM]*d0PM + f8Next[kNeighbor0MP]*d0MP + f8Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
} else {
f0LastInterp = f0Last[kNeighbor0PP];
f1LastInterp = f1Last[kNeighbor0PP];
@@ -808,25 +808,25 @@ __global__ void PrecursorDeviceDistributions( int* subgridDistanceIndices,
__global__ void QPrecursorDeviceDistributions( int* subgridDistanceIndices,
real* subgridDistances,
int sizeQ,
- int numberOfBCnodes,
+ int numberOfBCnodes,
int numberOfPrecursorNodes,
- real* distributions,
- uint* neighborX,
- uint* neighborY,
- uint* neighborZ,
- uint* neighbors0PP,
- uint* neighbors0PM,
- uint* neighbors0MP,
- uint* neighbors0MM,
- real* weights0PP,
- real* weights0PM,
- real* weights0MP,
- real* weights0MM,
- real* fsLast,
- real* fsNext,
- real timeRatio,
- unsigned long long numberOfLBnodes,
- bool isEvenTimestep)
+ real* distributions,
+ uint* neighborX,
+ uint* neighborY,
+ uint* neighborZ,
+ uint* neighbors0PP,
+ uint* neighbors0PM,
+ uint* neighbors0MP,
+ uint* neighbors0MM,
+ real* weights0PP,
+ real* weights0PM,
+ real* weights0MP,
+ real* weights0MM,
+ real* fsLast,
+ real* fsNext,
+ real timeRatio,
+ unsigned long long numberOfLBnodes,
+ bool isEvenTimestep)
{
const unsigned k = vf::gpu::getNodeIndex();
@@ -873,31 +873,31 @@ __global__ void QPrecursorDeviceDistributions( int* subgridDistanceIndices,
f0LastInterp = (f0Last[kNeighbor0PP]*d0PP + f0Last[kNeighbor0PM]*d0PM + f0Last[kNeighbor0MP]*d0MP + f0Last[kNeighbor0MM]*d0MM)*invWeightSum;
f0NextInterp = (f0Next[kNeighbor0PP]*d0PP + f0Next[kNeighbor0PM]*d0PM + f0Next[kNeighbor0MP]*d0MP + f0Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f1LastInterp = (f1Last[kNeighbor0PP]*d0PP + f1Last[kNeighbor0PM]*d0PM + f1Last[kNeighbor0MP]*d0MP + f1Last[kNeighbor0MM]*d0MM)*invWeightSum;
f1NextInterp = (f1Next[kNeighbor0PP]*d0PP + f1Next[kNeighbor0PM]*d0PM + f1Next[kNeighbor0MP]*d0MP + f1Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f2LastInterp = (f2Last[kNeighbor0PP]*d0PP + f2Last[kNeighbor0PM]*d0PM + f2Last[kNeighbor0MP]*d0MP + f2Last[kNeighbor0MM]*d0MM)*invWeightSum;
f2NextInterp = (f2Next[kNeighbor0PP]*d0PP + f2Next[kNeighbor0PM]*d0PM + f2Next[kNeighbor0MP]*d0MP + f2Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f3LastInterp = (f3Last[kNeighbor0PP]*d0PP + f3Last[kNeighbor0PM]*d0PM + f3Last[kNeighbor0MP]*d0MP + f3Last[kNeighbor0MM]*d0MM)*invWeightSum;
f3NextInterp = (f3Next[kNeighbor0PP]*d0PP + f3Next[kNeighbor0PM]*d0PM + f3Next[kNeighbor0MP]*d0MP + f3Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f4LastInterp = (f4Last[kNeighbor0PP]*d0PP + f4Last[kNeighbor0PM]*d0PM + f4Last[kNeighbor0MP]*d0MP + f4Last[kNeighbor0MM]*d0MM)*invWeightSum;
f4NextInterp = (f4Next[kNeighbor0PP]*d0PP + f4Next[kNeighbor0PM]*d0PM + f4Next[kNeighbor0MP]*d0MP + f4Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f5LastInterp = (f5Last[kNeighbor0PP]*d0PP + f5Last[kNeighbor0PM]*d0PM + f5Last[kNeighbor0MP]*d0MP + f5Last[kNeighbor0MM]*d0MM)*invWeightSum;
f5NextInterp = (f5Next[kNeighbor0PP]*d0PP + f5Next[kNeighbor0PM]*d0PM + f5Next[kNeighbor0MP]*d0MP + f5Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f6LastInterp = (f6Last[kNeighbor0PP]*d0PP + f6Last[kNeighbor0PM]*d0PM + f6Last[kNeighbor0MP]*d0MP + f6Last[kNeighbor0MM]*d0MM)*invWeightSum;
f6NextInterp = (f6Next[kNeighbor0PP]*d0PP + f6Next[kNeighbor0PM]*d0PM + f6Next[kNeighbor0MP]*d0MP + f6Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f7LastInterp = (f7Last[kNeighbor0PP]*d0PP + f7Last[kNeighbor0PM]*d0PM + f7Last[kNeighbor0MP]*d0MP + f7Last[kNeighbor0MM]*d0MM)*invWeightSum;
f7NextInterp = (f7Next[kNeighbor0PP]*d0PP + f7Next[kNeighbor0PM]*d0PM + f7Next[kNeighbor0MP]*d0MP + f7Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
f8LastInterp = (f8Last[kNeighbor0PP]*d0PP + f8Last[kNeighbor0PM]*d0PM + f8Last[kNeighbor0MP]*d0MP + f8Last[kNeighbor0MM]*d0MM)*invWeightSum;
f8NextInterp = (f8Next[kNeighbor0PP]*d0PP + f8Next[kNeighbor0PM]*d0PM + f8Next[kNeighbor0MP]*d0MP + f8Next[kNeighbor0MM]*d0MM)*invWeightSum;
-
+
} else {
f0LastInterp = f0Last[kNeighbor0PP];
f1LastInterp = f1Last[kNeighbor0PP];
diff --git a/src/gpu/VirtualFluids_GPU/GPU/PressBCs27.cu b/src/gpu/VirtualFluids_GPU/GPU/PressBCs27.cu
index 6a14cebd465a1e79acf14ea019abb34e66d9c85f..a72ebb1a385d16850554dad916aab7708235980b 100644
--- a/src/gpu/VirtualFluids_GPU/GPU/PressBCs27.cu
+++ b/src/gpu/VirtualFluids_GPU/GPU/PressBCs27.cu
@@ -1,5 +1,5 @@
/* Device code */
-#include "LBM/LB.h"
+#include "LBM/LB.h"
#include "lbm/constants/D3Q27.h"
#include "lbm/constants/NumericConstants.h"
#include "lbm/MacroscopicQuantities.h"
@@ -12,21 +12,21 @@ using namespace vf::lbm::dir;
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
__global__ void QInflowScaleByPressDevice27( real* rhoBC,
- real* DD,
- int* k_Q,
- int* k_N,
- int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ int* k_Q,
+ int* k_N,
+ int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
////////////////////////////////////////////////////////////////////////////////
- const unsigned x = threadIdx.x; // Globaler x-Index
- const unsigned y = blockIdx.x; // Globaler y-Index
- const unsigned z = blockIdx.y; // Globaler z-Index
+ const unsigned x = threadIdx.x; // Globaler x-Index
+ const unsigned y = blockIdx.x; // Globaler y-Index
+ const unsigned z = blockIdx.y; // Globaler z-Index
const unsigned nx = blockDim.x;
const unsigned ny = gridDim.x;
@@ -127,7 +127,7 @@ __global__ void QInflowScaleByPressDevice27( real* rhoBC,
D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
+ }
else
{
D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -218,23 +218,23 @@ __global__ void QInflowScaleByPressDevice27( real* rhoBC,
// real vx1, vx2, vx3;
real drho, drho1;
//////////////////////////////////////////////////////////////////////////
- //Dichte
+ //Dichte
drho1 = f1_TSE + f1_TNW + f1_TNE + f1_TSW + f1_BSE + f1_BNW + f1_BNE + f1_BSW +
- f1_BN + f1_TS + f1_TN + f1_BS + f1_BE + f1_TW + f1_TE + f1_BW + f1_SE + f1_NW + f1_NE + f1_SW +
- f1_T + f1_B + f1_N + f1_S + f1_E + f1_W + ((D.f[DIR_000])[k1zero]);
+ f1_BN + f1_TS + f1_TN + f1_BS + f1_BE + f1_TW + f1_TE + f1_BW + f1_SE + f1_NW + f1_NE + f1_SW +
+ f1_T + f1_B + f1_N + f1_S + f1_E + f1_W + ((D.f[DIR_000])[k1zero]);
drho = f_TSE + f_TNW + f_TNE + f_TSW + f_BSE + f_BNW + f_BNE + f_BSW +
- f_BN + f_TS + f_TN + f_BS + f_BE + f_TW + f_TE + f_BW + f_SE + f_NW + f_NE + f_SW +
- f_T + f_B + f_N + f_S + f_E + f_W + ((D.f[DIR_000])[kzero]);
+ f_BN + f_TS + f_TN + f_BS + f_BE + f_TW + f_TE + f_BW + f_SE + f_NW + f_NE + f_SW +
+ f_T + f_B + f_N + f_S + f_E + f_W + ((D.f[DIR_000])[kzero]);
//////////////////////////////////////////////////////////////////////////
- //Schallgeschwindigkeit
- real cs = c1o1 / sqrtf(c3o1);
+ //Schallgeschwindigkeit
+ real cs = c1o1 / sqrtf(c3o1);
//////////////////////////////////////////////////////////////////////////
- real rhoInterpol = drho1 * cs + (c1o1 - cs) * drho;
- //real diffRho = (rhoBC[k] + one) / (rhoInterpol + one);
- real diffRhoToAdd = rhoBC[k] - rhoInterpol;
- /////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
- //no velocity
- //////////////////////////////////////////
+ real rhoInterpol = drho1 * cs + (c1o1 - cs) * drho;
+ //real diffRho = (rhoBC[k] + one) / (rhoInterpol + one);
+ real diffRhoToAdd = rhoBC[k] - rhoInterpol;
+ /////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+ //no velocity
+ //////////////////////////////////////////
f_E = f1_E * cs + (c1o1 - cs) * f_E ;
f_W = f1_W * cs + (c1o1 - cs) * f_W ;
f_N = f1_N * cs + (c1o1 - cs) * f_N ;
@@ -261,16 +261,16 @@ __global__ void QInflowScaleByPressDevice27( real* rhoBC,
f_BSW = f1_BSW * cs + (c1o1 - cs) * f_BSW ;
f_BSE = f1_BSE * cs + (c1o1 - cs) * f_BSE ;
f_BNW = f1_BNW * cs + (c1o1 - cs) * f_BNW ;
- /////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
- //scale by press
- //////////////////////////////////////////
- //f_E = (f_E + c2over27 ) * diffRho - c2over27 ;
+ /////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+ //scale by press
+ //////////////////////////////////////////
+ //f_E = (f_E + c2over27 ) * diffRho - c2over27 ;
// f_W = (f_W + c2over27 ) * diffRho - c2over27 ;
// f_N = (f_N + c2over27 ) * diffRho - c2over27 ;
// f_S = (f_S + c2over27 ) * diffRho - c2over27 ;
// f_T = (f_T + c2over27 ) * diffRho - c2over27 ;
// f_B = (f_B + c2over27 ) * diffRho - c2over27 ;
- //f_NE = (f_NE + c1over54 ) * diffRho - c1over54 ;
+ //f_NE = (f_NE + c1over54 ) * diffRho - c1over54 ;
// f_SW = (f_SW + c1over54 ) * diffRho - c1over54 ;
// f_SE = (f_SE + c1over54 ) * diffRho - c1over54 ;
// f_NW = (f_NW + c1over54 ) * diffRho - c1over54 ;
@@ -290,16 +290,16 @@ __global__ void QInflowScaleByPressDevice27( real* rhoBC,
// f_BSW = (f_BSW + c1over216) * diffRho - c1over216;
// f_BSE = (f_BSE + c1over216) * diffRho - c1over216;
// f_BNW = (f_BNW + c1over216) * diffRho - c1over216;
- /////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
- // add press
- //////////////////////////////////////////
- f_E = (f_E + c2o27 * diffRhoToAdd);
+ /////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+ // add press
+ //////////////////////////////////////////
+ f_E = (f_E + c2o27 * diffRhoToAdd);
f_W = (f_W + c2o27 * diffRhoToAdd);
f_N = (f_N + c2o27 * diffRhoToAdd);
f_S = (f_S + c2o27 * diffRhoToAdd);
f_T = (f_T + c2o27 * diffRhoToAdd);
f_B = (f_B + c2o27 * diffRhoToAdd);
- f_NE = (f_NE + c1o54 * diffRhoToAdd);
+ f_NE = (f_NE + c1o54 * diffRhoToAdd);
f_SW = (f_SW + c1o54 * diffRhoToAdd);
f_SE = (f_SE + c1o54 * diffRhoToAdd);
f_NW = (f_NW + c1o54 * diffRhoToAdd);
@@ -319,9 +319,9 @@ __global__ void QInflowScaleByPressDevice27( real* rhoBC,
f_BSW = (f_BSW + c1o216 * diffRhoToAdd);
f_BSE = (f_BSE + c1o216 * diffRhoToAdd);
f_BNW = (f_BNW + c1o216 * diffRhoToAdd);
- /////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+ /////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
- //////////////////////////////////////////////////////////////////////////
+ //////////////////////////////////////////////////////////////////////////
if (isEvenTimestep==false)
{
D.f[DIR_P00 ] = &DD[DIR_P00 *size_Mat];
@@ -351,7 +351,7 @@ __global__ void QInflowScaleByPressDevice27( real* rhoBC,
D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
+ }
else
{
D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -384,46 +384,46 @@ __global__ void QInflowScaleByPressDevice27( real* rhoBC,
}
//////////////////////////////////////////////////////////////////////////
//__syncthreads();
- // -X
- //(D.f[DIR_P00 ])[ke ] = f_E ;
- //(D.f[DIR_PM0 ])[kse ] = f_SE ;
- //(D.f[DIR_PP0 ])[kne ] = f_NE ;
- //(D.f[DIR_P0M ])[kbe ] = f_BE ;
- //(D.f[DIR_P0P ])[kte ] = f_TE ;
- //(D.f[DIR_PMP ])[ktse ] = f_TSE ;
- //(D.f[DIR_PPP ])[ktne ] = f_TNE ;
- //(D.f[DIR_PMM ])[kbse ] = f_BSE ;
- //(D.f[DIR_PPM ])[kbne ] = f_BNE ;
- // X
- (D.f[DIR_M00 ])[kw ] = f_W ;
- (D.f[DIR_MM0 ])[ksw ] = f_SW ;
- (D.f[DIR_MP0 ])[knw ] = f_NW ;
- (D.f[DIR_M0M ])[kbw ] = f_BW ;
- (D.f[DIR_M0P ])[ktw ] = f_TW ;
- (D.f[DIR_MMP ])[ktsw ] = f_TSW ;
- (D.f[DIR_MPP ])[ktnw ] = f_TNW ;
- (D.f[DIR_MMM ])[kbsw ] = f_BSW ;
- (D.f[DIR_MPM ])[kbnw ] = f_BNW ;
- // Y
- //(D.f[DIR_0M0 ])[ks ] = f_S ;
- //(D.f[DIR_PM0 ])[kse ] = f_SE ;
- //(D.f[DIR_MM0 ])[ksw ] = f_SW ;
- //(D.f[DIR_0MP ])[kts ] = f_TS ;
- //(D.f[DIR_0MM ])[kbs ] = f_BS ;
- //(D.f[DIR_PMP ])[ktse ] = f_TSE ;
- //(D.f[DIR_MMP ])[ktsw ] = f_TSW ;
- //(D.f[DIR_PMM ])[kbse ] = f_BSE ;
- //(D.f[DIR_MMM ])[kbsw ] = f_BSW ;
- // Z
- //(D.f[DIR_00M ])[kb ] = f_B ;
- //(D.f[DIR_P0M ])[kbe ] = f_BE ;
- //(D.f[DIR_M0M ])[kbw ] = f_BW ;
- //(D.f[DIR_0PM ])[kbn ] = f_BN ;
- //(D.f[DIR_0MM ])[kbs ] = f_BS ;
- //(D.f[DIR_PPM ])[kbne ] = f_BNE ;
- //(D.f[DIR_MPM ])[kbnw ] = f_BNW ;
- //(D.f[DIR_PMM ])[kbse ] = f_BSE ;
- //(D.f[DIR_MMM ])[kbsw ] = f_BSW ;
+ // -X
+ //(D.f[DIR_P00 ])[ke ] = f_E ;
+ //(D.f[DIR_PM0 ])[kse ] = f_SE ;
+ //(D.f[DIR_PP0 ])[kne ] = f_NE ;
+ //(D.f[DIR_P0M ])[kbe ] = f_BE ;
+ //(D.f[DIR_P0P ])[kte ] = f_TE ;
+ //(D.f[DIR_PMP ])[ktse ] = f_TSE ;
+ //(D.f[DIR_PPP ])[ktne ] = f_TNE ;
+ //(D.f[DIR_PMM ])[kbse ] = f_BSE ;
+ //(D.f[DIR_PPM ])[kbne ] = f_BNE ;
+ // X
+ (D.f[DIR_M00 ])[kw ] = f_W ;
+ (D.f[DIR_MM0 ])[ksw ] = f_SW ;
+ (D.f[DIR_MP0 ])[knw ] = f_NW ;
+ (D.f[DIR_M0M ])[kbw ] = f_BW ;
+ (D.f[DIR_M0P ])[ktw ] = f_TW ;
+ (D.f[DIR_MMP ])[ktsw ] = f_TSW ;
+ (D.f[DIR_MPP ])[ktnw ] = f_TNW ;
+ (D.f[DIR_MMM ])[kbsw ] = f_BSW ;
+ (D.f[DIR_MPM ])[kbnw ] = f_BNW ;
+ // Y
+ //(D.f[DIR_0M0 ])[ks ] = f_S ;
+ //(D.f[DIR_PM0 ])[kse ] = f_SE ;
+ //(D.f[DIR_MM0 ])[ksw ] = f_SW ;
+ //(D.f[DIR_0MP ])[kts ] = f_TS ;
+ //(D.f[DIR_0MM ])[kbs ] = f_BS ;
+ //(D.f[DIR_PMP ])[ktse ] = f_TSE ;
+ //(D.f[DIR_MMP ])[ktsw ] = f_TSW ;
+ //(D.f[DIR_PMM ])[kbse ] = f_BSE ;
+ //(D.f[DIR_MMM ])[kbsw ] = f_BSW ;
+ // Z
+ //(D.f[DIR_00M ])[kb ] = f_B ;
+ //(D.f[DIR_P0M ])[kbe ] = f_BE ;
+ //(D.f[DIR_M0M ])[kbw ] = f_BW ;
+ //(D.f[DIR_0PM ])[kbn ] = f_BN ;
+ //(D.f[DIR_0MM ])[kbs ] = f_BS ;
+ //(D.f[DIR_PPM ])[kbne ] = f_BNE ;
+ //(D.f[DIR_MPM ])[kbnw ] = f_BNW ;
+ //(D.f[DIR_PMM ])[kbse ] = f_BSE ;
+ //(D.f[DIR_MMM ])[kbsw ] = f_BSW ;
//////////////////////////////////////////////////////////////////////////
}
}
@@ -469,21 +469,21 @@ __global__ void QInflowScaleByPressDevice27( real* rhoBC,
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
__global__ void QPressDeviceIncompNEQ27( real* rhoBC,
- real* DD,
- int* k_Q,
- int* k_N,
- int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ int* k_Q,
+ int* k_N,
+ int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
////////////////////////////////////////////////////////////////////////////////
- const unsigned x = threadIdx.x; // Globaler x-Index
- const unsigned y = blockIdx.x; // Globaler y-Index
- const unsigned z = blockIdx.y; // Globaler z-Index
+ const unsigned x = threadIdx.x; // Globaler x-Index
+ const unsigned y = blockIdx.x; // Globaler y-Index
+ const unsigned z = blockIdx.y; // Globaler z-Index
const unsigned nx = blockDim.x;
const unsigned ny = gridDim.x;
@@ -584,7 +584,7 @@ __global__ void QPressDeviceIncompNEQ27( real* rhoBC,
D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
+ }
else
{
D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -652,17 +652,17 @@ __global__ void QPressDeviceIncompNEQ27( real* rhoBC,
f1_TNE+f1_TSW+f1_TSE+f1_TNW+f1_BNE+f1_BSW+f1_BSE+f1_BNW;
real vx1 = ((f1_TSE - f1_BNW) - (f1_TNW - f1_BSE)) + ((f1_TNE - f1_BSW) - (f1_TSW - f1_BNE)) +
- ((f1_BE - f1_TW) + (f1_TE - f1_BW)) + ((f1_SE - f1_NW) + (f1_NE - f1_SW)) +
- (f1_E - f1_W);
+ ((f1_BE - f1_TW) + (f1_TE - f1_BW)) + ((f1_SE - f1_NW) + (f1_NE - f1_SW)) +
+ (f1_E - f1_W);
real vx2 = (-(f1_TSE - f1_BNW) + (f1_TNW - f1_BSE)) + ((f1_TNE - f1_BSW) - (f1_TSW - f1_BNE)) +
- ((f1_BN - f1_TS) + (f1_TN - f1_BS)) + (-(f1_SE - f1_NW) + (f1_NE - f1_SW)) +
- (f1_N - f1_S);
+ ((f1_BN - f1_TS) + (f1_TN - f1_BS)) + (-(f1_SE - f1_NW) + (f1_NE - f1_SW)) +
+ (f1_N - f1_S);
real vx3 = ((f1_TSE - f1_BNW) + (f1_TNW - f1_BSE)) + ((f1_TNE - f1_BSW) + (f1_TSW - f1_BNE)) +
- (-(f1_BN - f1_TS) + (f1_TN - f1_BS)) + ((f1_TE - f1_BW) - (f1_BE - f1_TW)) +
- (f1_T - f1_B);
+ (-(f1_BN - f1_TS) + (f1_TN - f1_BS)) + ((f1_TE - f1_BW) - (f1_BE - f1_TW)) +
+ (f1_T - f1_B);
real cusq=c3o2*(vx1*vx1+vx2*vx2+vx3*vx3);
@@ -693,15 +693,15 @@ __global__ void QPressDeviceIncompNEQ27( real* rhoBC,
f1_BNW -= c1o216*(drho1+(drho1+c1o1)*(c3o1*(-vx1+vx2-vx3)+c9o2*(-vx1+vx2-vx3)*(-vx1+vx2-vx3)-cusq));
f1_BSE -= c1o216*(drho1+(drho1+c1o1)*(c3o1*( vx1-vx2-vx3)+c9o2*( vx1-vx2-vx3)*( vx1-vx2-vx3)-cusq));
f1_TNW -= c1o216*(drho1+(drho1+c1o1)*(c3o1*(-vx1+vx2+vx3)+c9o2*(-vx1+vx2+vx3)*(-vx1+vx2+vx3)-cusq));
-
- drho1 = rhoBC[k];
- //if(vx1 < zero){
- // vx1 *= 0.9;
- //}
- //if(vx2 < zero){
- // vx2 *= c1o10;//0.9;
- //}
+ drho1 = rhoBC[k];
+
+ //if(vx1 < zero){
+ // vx1 *= 0.9;
+ //}
+ //if(vx2 < zero){
+ // vx2 *= c1o10;//0.9;
+ //}
f1_ZERO += c8o27* (drho1-(drho1+c1o1)*cusq);
f1_E += c2o27* (drho1+(drho1+c1o1)*(c3o1*( vx1 )+c9o2*( vx1 )*( vx1 )-cusq));
@@ -731,39 +731,39 @@ __global__ void QPressDeviceIncompNEQ27( real* rhoBC,
f1_BSE += c1o216*(drho1+(drho1+c1o1)*(c3o1*( vx1-vx2-vx3)+c9o2*( vx1-vx2-vx3)*( vx1-vx2-vx3)-cusq));
f1_TNW += c1o216*(drho1+(drho1+c1o1)*(c3o1*(-vx1+vx2+vx3)+c9o2*(-vx1+vx2+vx3)*(-vx1+vx2+vx3)-cusq));
- //drho1 = (drho1 + rhoBC[k])/2.f;
- //drho1 = drho1 - rhoBC[k];
+ //drho1 = (drho1 + rhoBC[k])/2.f;
+ //drho1 = drho1 - rhoBC[k];
//////////////////////////////////////////////////////////////////////////
__syncthreads();
- (D.f[DIR_P00 ])[ke ] = f1_W ;
- (D.f[DIR_M00 ])[kw ] = f1_E ;
- (D.f[DIR_0P0 ])[kn ] = f1_S ;
- (D.f[DIR_0M0 ])[ks ] = f1_N ;
- (D.f[DIR_00P ])[kt ] = f1_B ;
- (D.f[DIR_00M ])[kb ] = f1_T ;
- (D.f[DIR_PP0 ])[kne ] = f1_SW ;
- (D.f[DIR_MM0 ])[ksw ] = f1_NE ;
- (D.f[DIR_PM0 ])[kse ] = f1_NW ;
- (D.f[DIR_MP0 ])[knw ] = f1_SE ;
- (D.f[DIR_P0P ])[kte ] = f1_BW ;
- (D.f[DIR_M0M ])[kbw ] = f1_TE ;
- (D.f[DIR_P0M ])[kbe ] = f1_TW ;
- (D.f[DIR_M0P ])[ktw ] = f1_BE ;
- (D.f[DIR_0PP ])[ktn ] = f1_BS ;
- (D.f[DIR_0MM ])[kbs ] = f1_TN ;
- (D.f[DIR_0PM ])[kbn ] = f1_TS ;
- (D.f[DIR_0MP ])[kts ] = f1_BN ;
- (D.f[DIR_000])[kzero] = f1_ZERO;
- (D.f[DIR_PPP ])[ktne ] = f1_BSW ;
- (D.f[DIR_MMP ])[ktsw ] = f1_BNE ;
- (D.f[DIR_PMP ])[ktse ] = f1_BNW ;
- (D.f[DIR_MPP ])[ktnw ] = f1_BSE ;
- (D.f[DIR_PPM ])[kbne ] = f1_TSW ;
- (D.f[DIR_MMM ])[kbsw ] = f1_TNE ;
- (D.f[DIR_PMM ])[kbse ] = f1_TNW ;
- (D.f[DIR_MPM ])[kbnw ] = f1_TSE ;
+ (D.f[DIR_P00 ])[ke ] = f1_W ;
+ (D.f[DIR_M00 ])[kw ] = f1_E ;
+ (D.f[DIR_0P0 ])[kn ] = f1_S ;
+ (D.f[DIR_0M0 ])[ks ] = f1_N ;
+ (D.f[DIR_00P ])[kt ] = f1_B ;
+ (D.f[DIR_00M ])[kb ] = f1_T ;
+ (D.f[DIR_PP0 ])[kne ] = f1_SW ;
+ (D.f[DIR_MM0 ])[ksw ] = f1_NE ;
+ (D.f[DIR_PM0 ])[kse ] = f1_NW ;
+ (D.f[DIR_MP0 ])[knw ] = f1_SE ;
+ (D.f[DIR_P0P ])[kte ] = f1_BW ;
+ (D.f[DIR_M0M ])[kbw ] = f1_TE ;
+ (D.f[DIR_P0M ])[kbe ] = f1_TW ;
+ (D.f[DIR_M0P ])[ktw ] = f1_BE ;
+ (D.f[DIR_0PP ])[ktn ] = f1_BS ;
+ (D.f[DIR_0MM ])[kbs ] = f1_TN ;
+ (D.f[DIR_0PM ])[kbn ] = f1_TS ;
+ (D.f[DIR_0MP ])[kts ] = f1_BN ;
+ (D.f[DIR_000])[kzero] = f1_ZERO;
+ (D.f[DIR_PPP ])[ktne ] = f1_BSW ;
+ (D.f[DIR_MMP ])[ktsw ] = f1_BNE ;
+ (D.f[DIR_PMP ])[ktse ] = f1_BNW ;
+ (D.f[DIR_MPP ])[ktnw ] = f1_BSE ;
+ (D.f[DIR_PPM ])[kbne ] = f1_TSW ;
+ (D.f[DIR_MMM ])[kbsw ] = f1_TNE ;
+ (D.f[DIR_PMM ])[kbse ] = f1_TNW ;
+ (D.f[DIR_MPM ])[kbnw ] = f1_TSE ;
}
}
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
@@ -808,26 +808,26 @@ __global__ void QPressDeviceIncompNEQ27( real* rhoBC,
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
__global__ void QPressDeviceNEQ27(real* rhoBC,
- real* distribution,
+ real* distribution,
int* bcNodeIndices,
int* bcNeighborIndices,
int numberOfBCnodes,
- real omega1,
+ real omega1,
unsigned int* neighborX,
unsigned int* neighborY,
unsigned int* neighborZ,
- unsigned int numberOfLBnodes,
+ unsigned int numberOfLBnodes,
bool isEvenTimestep)
{
//////////////////////////////////////////////////////////////////////////
- //! The pressure boundary condition is executed in the following steps
- //!
- ////////////////////////////////////////////////////////////////////////////////
- //! - Get node index coordinates from threadIdx, blockIdx, blockDim and gridDim.
- //!
- const unsigned x = threadIdx.x; // global x-index
- const unsigned y = blockIdx.x; // global y-index
- const unsigned z = blockIdx.y; // global z-index
+ //! The pressure boundary condition is executed in the following steps
+ //!
+ ////////////////////////////////////////////////////////////////////////////////
+ //! - Get node index coordinates from threadIdx, blockIdx, blockDim and gridDim.
+ //!
+ const unsigned x = threadIdx.x; // global x-index
+ const unsigned y = blockIdx.x; // global y-index
+ const unsigned z = blockIdx.y; // global z-index
const unsigned nx = blockDim.x;
const unsigned ny = gridDim.x;
@@ -949,20 +949,20 @@ __global__ void QPressDeviceNEQ27(real* rhoBC,
//! - Calculate macroscopic quantities (for neighboring node)
//!
real drho1 = f1_TSE + f1_TNW + f1_TNE + f1_TSW + f1_BSE + f1_BNW + f1_BNE + f1_BSW +
- f1_BN + f1_TS + f1_TN + f1_BS + f1_BE + f1_TW + f1_TE + f1_BW + f1_SE + f1_NW + f1_NE + f1_SW +
- f1_T + f1_B + f1_N + f1_S + f1_E + f1_W + ((dist.f[DIR_000])[kzero]);
+ f1_BN + f1_TS + f1_TN + f1_BS + f1_BE + f1_TW + f1_TE + f1_BW + f1_SE + f1_NW + f1_NE + f1_SW +
+ f1_T + f1_B + f1_N + f1_S + f1_E + f1_W + ((dist.f[DIR_000])[kzero]);
real vx1 = (((f1_TSE - f1_BNW) - (f1_TNW - f1_BSE)) + ((f1_TNE - f1_BSW) - (f1_TSW - f1_BNE)) +
((f1_BE - f1_TW) + (f1_TE - f1_BW)) + ((f1_SE - f1_NW) + (f1_NE - f1_SW)) +
- (f1_E - f1_W)) / (c1o1 + drho1);
+ (f1_E - f1_W)) / (c1o1 + drho1);
real vx2 = ((-(f1_TSE - f1_BNW) + (f1_TNW - f1_BSE)) + ((f1_TNE - f1_BSW) - (f1_TSW - f1_BNE)) +
((f1_BN - f1_TS) + (f1_TN - f1_BS)) + (-(f1_SE - f1_NW) + (f1_NE - f1_SW)) +
- (f1_N - f1_S)) / (c1o1 + drho1);
+ (f1_N - f1_S)) / (c1o1 + drho1);
real vx3 = (((f1_TSE - f1_BNW) + (f1_TNW - f1_BSE)) + ((f1_TNE - f1_BSW) + (f1_TSW - f1_BNE)) +
(-(f1_BN - f1_TS) + (f1_TN - f1_BS)) + ((f1_TE - f1_BW) - (f1_BE - f1_TW)) +
- (f1_T - f1_B)) / (c1o1 + drho1);
+ (f1_T - f1_B)) / (c1o1 + drho1);
real cusq = c3o2 * (vx1 * vx1 + vx2 * vx2 + vx3 * vx3);
@@ -1110,16 +1110,16 @@ __global__ void QPressDeviceNEQ27(real* rhoBC,
////////////////////////////////////////////////////////////////////////////////
-__global__ void LB_BC_Press_East27( int nx,
- int ny,
- int tz,
- unsigned int* bcMatD,
+__global__ void LB_BC_Press_East27( int nx,
+ int ny,
+ int tz,
+ unsigned int* bcMatD,
unsigned int* neighborX,
unsigned int* neighborY,
unsigned int* neighborZ,
- real* DD,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
//thread-index
int ty = blockIdx.x;
@@ -1127,9 +1127,9 @@ __global__ void LB_BC_Press_East27( int nx,
int k, k1, nxny; // Zugriff auf arrays im device
- int x = tx + STARTOFFX; // Globaler x-Index
- int y = ty + STARTOFFY; // Globaler y-Index
- int z = tz + STARTOFFZ; // Globaler z-Index
+ int x = tx + STARTOFFX; // Globaler x-Index
+ int y = ty + STARTOFFY; // Globaler y-Index
+ int z = tz + STARTOFFZ; // Globaler z-Index
k = nx*(ny*z + y) + x;
nxny = nx*ny;
@@ -1167,7 +1167,7 @@ __global__ void LB_BC_Press_East27( int nx,
D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
+ }
else
{
D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -1374,10 +1374,10 @@ __global__ void LB_BC_Press_East27( int nx,
(D.f[DIR_PPM ])[kbne ] = f1_TSW -c1o216*drho1;
(D.f[DIR_MMM ])[kbsw ] = f1_TNE -c1o216*drho1;
(D.f[DIR_PMM ])[kbse ] = f1_TNW -c1o216*drho1;
- (D.f[DIR_MPM ])[kbnw ] = f1_TSE -c1o216*drho1;
+ (D.f[DIR_MPM ])[kbnw ] = f1_TSE -c1o216*drho1;
}
__syncthreads();
-}
+}
//////////////////////////////////////////////////////////////////////////////
@@ -1420,15 +1420,15 @@ __global__ void LB_BC_Press_East27( int nx,
//////////////////////////////////////////////////////////////////////////////
__global__ void QPressDevice27(real* rhoBC,
- real* DD,
- int* k_Q,
+ real* DD,
+ int* k_Q,
real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
+ unsigned int numberOfBCnodes,
+ real om1,
unsigned int* neighborX,
unsigned int* neighborY,
unsigned int* neighborZ,
- unsigned int size_Mat,
+ unsigned int size_Mat,
bool isEvenTimestep)
{
Distributions27 D;
@@ -1461,7 +1461,7 @@ __global__ void QPressDevice27(real* rhoBC,
D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
+ }
else
{
D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -1493,9 +1493,9 @@ __global__ void QPressDevice27(real* rhoBC,
D.f[DIR_MPM ] = &DD[DIR_PMP *size_Mat];
}
////////////////////////////////////////////////////////////////////////////////
- const unsigned x = threadIdx.x; // Globaler x-Index
- const unsigned y = blockIdx.x; // Globaler y-Index
- const unsigned z = blockIdx.y; // Globaler z-Index
+ const unsigned x = threadIdx.x; // Globaler x-Index
+ const unsigned y = blockIdx.x; // Globaler y-Index
+ const unsigned z = blockIdx.y; // Globaler z-Index
const unsigned nx = blockDim.x;
const unsigned ny = gridDim.x;
@@ -1505,11 +1505,11 @@ __global__ void QPressDevice27(real* rhoBC,
if(k<numberOfBCnodes)
{
- real *q_dirE, *q_dirW, *q_dirN, *q_dirS, *q_dirT, *q_dirB,
+ real *q_dirE, *q_dirW, *q_dirN, *q_dirS, *q_dirT, *q_dirB,
*q_dirNE, *q_dirSW, *q_dirSE, *q_dirNW, *q_dirTE, *q_dirBW,
*q_dirBE, *q_dirTW, *q_dirTN, *q_dirBS, *q_dirBN, *q_dirTS,
*q_dirTNE, *q_dirTSW, *q_dirTSE, *q_dirTNW, *q_dirBNE, *q_dirBSW,
- *q_dirBSE, *q_dirBNW;
+ *q_dirBSE, *q_dirBNW;
q_dirE = &QQ[DIR_P00 * numberOfBCnodes];
q_dirW = &QQ[DIR_M00 * numberOfBCnodes];
q_dirN = &QQ[DIR_0P0 * numberOfBCnodes];
@@ -1600,16 +1600,16 @@ __global__ void QPressDevice27(real* rhoBC,
real q, vx1, vx2, vx3, drho;
vx1 = ((f_TSE - f_BNW) - (f_TNW - f_BSE)) + ((f_TNE - f_BSW) - (f_TSW - f_BNE)) +
((f_BE - f_TW) + (f_TE - f_BW)) + ((f_SE - f_NW) + (f_NE - f_SW)) +
- (f_E - f_W);
+ (f_E - f_W);
vx2 = (-(f_TSE - f_BNW) + (f_TNW - f_BSE)) + ((f_TNE - f_BSW) - (f_TSW - f_BNE)) +
((f_BN - f_TS) + (f_TN - f_BS)) + (-(f_SE - f_NW) + (f_NE - f_SW)) +
- (f_N - f_S);
+ (f_N - f_S);
vx3 = ((f_TSE - f_BNW) + (f_TNW - f_BSE)) + ((f_TNE - f_BSW) + (f_TSW - f_BNE)) +
(-(f_BN - f_TS) + (f_TN - f_BS)) + ((f_TE - f_BW) - (f_BE - f_TW)) +
- (f_T - f_B);
+ (f_T - f_B);
real cu_sq=c3o2*(vx1*vx1+vx2*vx2+vx3*vx3);
//////////////////////////////////////////////////////////////////////////
@@ -1646,7 +1646,7 @@ __global__ void QPressDevice27(real* rhoBC,
D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
+ }
else
{
D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -1681,183 +1681,183 @@ __global__ void QPressDevice27(real* rhoBC,
q = q_dirE[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_M00])[kw]=c2o27* (drho+c3o1*(-vx1 )+c9o2*(-vx1 )*(-vx1 )-cu_sq);
- //(D.f[DIR_P00])[ke]=c2over27* (drho+three*( vx1 )+c9over2*( vx1 )*( vx1 )-cu_sq);
+ (D.f[DIR_M00])[kw]=c2o27* (drho+c3o1*(-vx1 )+c9o2*(-vx1 )*(-vx1 )-cu_sq);
+ //(D.f[DIR_P00])[ke]=c2over27* (drho+three*( vx1 )+c9over2*( vx1 )*( vx1 )-cu_sq);
}
q = q_dirW[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_P00])[ke]=c2o27* (drho+c3o1*( vx1 )+c9o2*( vx1 )*( vx1 )-cu_sq);
- //(D.f[DIR_M00])[kw]=c2over27* (drho+three*(-vx1 )+c9over2*(-vx1 )*(-vx1 )-cu_sq);
+ (D.f[DIR_P00])[ke]=c2o27* (drho+c3o1*( vx1 )+c9o2*( vx1 )*( vx1 )-cu_sq);
+ //(D.f[DIR_M00])[kw]=c2over27* (drho+three*(-vx1 )+c9over2*(-vx1 )*(-vx1 )-cu_sq);
}
q = q_dirN[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_0M0])[ks]=c2o27* (drho+c3o1*( -vx2 )+c9o2*( -vx2 )*( -vx2 )-cu_sq);
- //(D.f[DIR_0P0])[kn]=c2over27* (drho+three*( vx2 )+c9over2*( vx2 )*( vx2 )-cu_sq);
+ (D.f[DIR_0M0])[ks]=c2o27* (drho+c3o1*( -vx2 )+c9o2*( -vx2 )*( -vx2 )-cu_sq);
+ //(D.f[DIR_0P0])[kn]=c2over27* (drho+three*( vx2 )+c9over2*( vx2 )*( vx2 )-cu_sq);
}
q = q_dirS[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_0P0])[kn]=c2o27* (drho+c3o1*( vx2 )+c9o2*( vx2 )*( vx2 )-cu_sq);
- //(D.f[DIR_0M0])[ks]=c2over27* (drho+three*( -vx2 )+c9over2*( -vx2 )*( -vx2 )-cu_sq);
+ (D.f[DIR_0P0])[kn]=c2o27* (drho+c3o1*( vx2 )+c9o2*( vx2 )*( vx2 )-cu_sq);
+ //(D.f[DIR_0M0])[ks]=c2over27* (drho+three*( -vx2 )+c9over2*( -vx2 )*( -vx2 )-cu_sq);
}
q = q_dirT[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_00M])[kb]=c2o27* (drho+c3o1*( -vx3)+c9o2*( -vx3)*( -vx3)-cu_sq);
- //(D.f[DIR_00P])[kt]=c2over27* (drho+three*( vx3)+c9over2*( vx3)*( vx3)-cu_sq);
+ (D.f[DIR_00M])[kb]=c2o27* (drho+c3o1*( -vx3)+c9o2*( -vx3)*( -vx3)-cu_sq);
+ //(D.f[DIR_00P])[kt]=c2over27* (drho+three*( vx3)+c9over2*( vx3)*( vx3)-cu_sq);
}
q = q_dirB[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_00P])[kt]=c2o27* (drho+c3o1*( vx3)+c9o2*( vx3)*( vx3)-cu_sq);
- //(D.f[DIR_00M])[kb]=c2over27* (drho+three*( -vx3)+c9over2*( -vx3)*( -vx3)-cu_sq);
+ (D.f[DIR_00P])[kt]=c2o27* (drho+c3o1*( vx3)+c9o2*( vx3)*( vx3)-cu_sq);
+ //(D.f[DIR_00M])[kb]=c2over27* (drho+three*( -vx3)+c9over2*( -vx3)*( -vx3)-cu_sq);
}
q = q_dirNE[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_MM0])[ksw]=c1o54* (drho+c3o1*(-vx1-vx2 )+c9o2*(-vx1-vx2 )*(-vx1-vx2 )-cu_sq);
- //(D.f[DIR_PP0])[kne]=c1over54* (drho+three*( vx1+vx2 )+c9over2*( vx1+vx2 )*( vx1+vx2 )-cu_sq);
+ (D.f[DIR_MM0])[ksw]=c1o54* (drho+c3o1*(-vx1-vx2 )+c9o2*(-vx1-vx2 )*(-vx1-vx2 )-cu_sq);
+ //(D.f[DIR_PP0])[kne]=c1over54* (drho+three*( vx1+vx2 )+c9over2*( vx1+vx2 )*( vx1+vx2 )-cu_sq);
}
q = q_dirSW[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_PP0])[kne]=c1o54* (drho+c3o1*( vx1+vx2 )+c9o2*( vx1+vx2 )*( vx1+vx2 )-cu_sq);
- //(D.f[DIR_MM0])[ksw]=c1over54* (drho+three*(-vx1-vx2 )+c9over2*(-vx1-vx2 )*(-vx1-vx2 )-cu_sq);
+ (D.f[DIR_PP0])[kne]=c1o54* (drho+c3o1*( vx1+vx2 )+c9o2*( vx1+vx2 )*( vx1+vx2 )-cu_sq);
+ //(D.f[DIR_MM0])[ksw]=c1over54* (drho+three*(-vx1-vx2 )+c9over2*(-vx1-vx2 )*(-vx1-vx2 )-cu_sq);
}
q = q_dirSE[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_MP0])[knw]=c1o54* (drho+c3o1*(-vx1+vx2 )+c9o2*(-vx1+vx2 )*(-vx1+vx2 )-cu_sq);
- //(D.f[DIR_PM0])[kse]=c1over54* (drho+three*( vx1-vx2 )+c9over2*( vx1-vx2 )*( vx1-vx2 )-cu_sq);
+ (D.f[DIR_MP0])[knw]=c1o54* (drho+c3o1*(-vx1+vx2 )+c9o2*(-vx1+vx2 )*(-vx1+vx2 )-cu_sq);
+ //(D.f[DIR_PM0])[kse]=c1over54* (drho+three*( vx1-vx2 )+c9over2*( vx1-vx2 )*( vx1-vx2 )-cu_sq);
}
q = q_dirNW[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_PM0])[kse]=c1o54* (drho+c3o1*( vx1-vx2 )+c9o2*( vx1-vx2 )*( vx1-vx2 )-cu_sq);
- //(D.f[DIR_MP0])[knw]=c1over54* (drho+three*(-vx1+vx2 )+c9over2*(-vx1+vx2 )*(-vx1+vx2 )-cu_sq);
+ (D.f[DIR_PM0])[kse]=c1o54* (drho+c3o1*( vx1-vx2 )+c9o2*( vx1-vx2 )*( vx1-vx2 )-cu_sq);
+ //(D.f[DIR_MP0])[knw]=c1over54* (drho+three*(-vx1+vx2 )+c9over2*(-vx1+vx2 )*(-vx1+vx2 )-cu_sq);
}
q = q_dirTE[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_M0M])[kbw]=c1o54* (drho+c3o1*(-vx1 -vx3)+c9o2*(-vx1 -vx3)*(-vx1 -vx3)-cu_sq);
- //(D.f[DIR_P0P])[kte]=c1over54* (drho+three*( vx1 +vx3)+c9over2*( vx1 +vx3)*( vx1 +vx3)-cu_sq);
+ (D.f[DIR_M0M])[kbw]=c1o54* (drho+c3o1*(-vx1 -vx3)+c9o2*(-vx1 -vx3)*(-vx1 -vx3)-cu_sq);
+ //(D.f[DIR_P0P])[kte]=c1over54* (drho+three*( vx1 +vx3)+c9over2*( vx1 +vx3)*( vx1 +vx3)-cu_sq);
}
q = q_dirBW[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_P0P])[kte]=c1o54* (drho+c3o1*( vx1 +vx3)+c9o2*( vx1 +vx3)*( vx1 +vx3)-cu_sq);
- //(D.f[DIR_M0M])[kbw]=c1over54* (drho+three*(-vx1 -vx3)+c9over2*(-vx1 -vx3)*(-vx1 -vx3)-cu_sq);
+ (D.f[DIR_P0P])[kte]=c1o54* (drho+c3o1*( vx1 +vx3)+c9o2*( vx1 +vx3)*( vx1 +vx3)-cu_sq);
+ //(D.f[DIR_M0M])[kbw]=c1over54* (drho+three*(-vx1 -vx3)+c9over2*(-vx1 -vx3)*(-vx1 -vx3)-cu_sq);
}
q = q_dirBE[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_M0P])[ktw]=c1o54* (drho+c3o1*(-vx1 +vx3)+c9o2*(-vx1 +vx3)*(-vx1 +vx3)-cu_sq);
- //(D.f[DIR_P0M])[kbe]=c1over54* (drho+three*( vx1 -vx3)+c9over2*( vx1 -vx3)*( vx1 -vx3)-cu_sq);
+ (D.f[DIR_M0P])[ktw]=c1o54* (drho+c3o1*(-vx1 +vx3)+c9o2*(-vx1 +vx3)*(-vx1 +vx3)-cu_sq);
+ //(D.f[DIR_P0M])[kbe]=c1over54* (drho+three*( vx1 -vx3)+c9over2*( vx1 -vx3)*( vx1 -vx3)-cu_sq);
}
q = q_dirTW[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_P0M])[kbe]=c1o54* (drho+c3o1*( vx1 -vx3)+c9o2*( vx1 -vx3)*( vx1 -vx3)-cu_sq);
- //(D.f[DIR_M0P])[ktw]=c1over54* (drho+three*(-vx1 +vx3)+c9over2*(-vx1 +vx3)*(-vx1 +vx3)-cu_sq);
+ (D.f[DIR_P0M])[kbe]=c1o54* (drho+c3o1*( vx1 -vx3)+c9o2*( vx1 -vx3)*( vx1 -vx3)-cu_sq);
+ //(D.f[DIR_M0P])[ktw]=c1over54* (drho+three*(-vx1 +vx3)+c9over2*(-vx1 +vx3)*(-vx1 +vx3)-cu_sq);
}
q = q_dirTN[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_0MM])[kbs]=c1o54* (drho+c3o1*( -vx2-vx3)+c9o2*( -vx2-vx3)*( -vx2-vx3)-cu_sq);
- //(D.f[DIR_0PP])[ktn]=c1over54* (drho+three*( vx2+vx3)+c9over2*( vx2+vx3)*( vx2+vx3)-cu_sq);
+ (D.f[DIR_0MM])[kbs]=c1o54* (drho+c3o1*( -vx2-vx3)+c9o2*( -vx2-vx3)*( -vx2-vx3)-cu_sq);
+ //(D.f[DIR_0PP])[ktn]=c1over54* (drho+three*( vx2+vx3)+c9over2*( vx2+vx3)*( vx2+vx3)-cu_sq);
}
q = q_dirBS[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_0PP])[ktn]=c1o54* (drho+c3o1*( vx2+vx3)+c9o2*( vx2+vx3)*( vx2+vx3)-cu_sq);
- //(D.f[DIR_0MM])[kbs]=c1over54* (drho+three*( -vx2-vx3)+c9over2*( -vx2-vx3)*( -vx2-vx3)-cu_sq);
+ (D.f[DIR_0PP])[ktn]=c1o54* (drho+c3o1*( vx2+vx3)+c9o2*( vx2+vx3)*( vx2+vx3)-cu_sq);
+ //(D.f[DIR_0MM])[kbs]=c1over54* (drho+three*( -vx2-vx3)+c9over2*( -vx2-vx3)*( -vx2-vx3)-cu_sq);
}
q = q_dirBN[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_0MP])[kts]=c1o54* (drho+c3o1*( -vx2+vx3)+c9o2*( -vx2+vx3)*( -vx2+vx3)-cu_sq);
- //(D.f[DIR_0PM])[kbn]=c1over54* (drho+three*( vx2-vx3)+c9over2*( vx2-vx3)*( vx2-vx3)-cu_sq);
+ (D.f[DIR_0MP])[kts]=c1o54* (drho+c3o1*( -vx2+vx3)+c9o2*( -vx2+vx3)*( -vx2+vx3)-cu_sq);
+ //(D.f[DIR_0PM])[kbn]=c1over54* (drho+three*( vx2-vx3)+c9over2*( vx2-vx3)*( vx2-vx3)-cu_sq);
}
q = q_dirTS[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_0PM])[kbn]=c1o54* (drho+c3o1*( vx2-vx3)+c9o2*( vx2-vx3)*( vx2-vx3)-cu_sq);
- //(D.f[DIR_0MP])[kts]=c1over54* (drho+three*( -vx2+vx3)+c9over2*( -vx2+vx3)*( -vx2+vx3)-cu_sq);
+ (D.f[DIR_0PM])[kbn]=c1o54* (drho+c3o1*( vx2-vx3)+c9o2*( vx2-vx3)*( vx2-vx3)-cu_sq);
+ //(D.f[DIR_0MP])[kts]=c1over54* (drho+three*( -vx2+vx3)+c9over2*( -vx2+vx3)*( -vx2+vx3)-cu_sq);
}
q = q_dirTNE[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_MMM])[kbsw]=c1o216*(drho+c3o1*(-vx1-vx2-vx3)+c9o2*(-vx1-vx2-vx3)*(-vx1-vx2-vx3)-cu_sq);
- //(D.f[DIR_PPP])[ktne]=c1over216*(drho+three*( vx1+vx2+vx3)+c9over2*( vx1+vx2+vx3)*( vx1+vx2+vx3)-cu_sq);
+ (D.f[DIR_MMM])[kbsw]=c1o216*(drho+c3o1*(-vx1-vx2-vx3)+c9o2*(-vx1-vx2-vx3)*(-vx1-vx2-vx3)-cu_sq);
+ //(D.f[DIR_PPP])[ktne]=c1over216*(drho+three*( vx1+vx2+vx3)+c9over2*( vx1+vx2+vx3)*( vx1+vx2+vx3)-cu_sq);
}
q = q_dirBSW[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_PPP])[ktne]=c1o216*(drho+c3o1*( vx1+vx2+vx3)+c9o2*( vx1+vx2+vx3)*( vx1+vx2+vx3)-cu_sq);
- //(D.f[DIR_MMM])[kbsw]=c1over216*(drho+three*(-vx1-vx2-vx3)+c9over2*(-vx1-vx2-vx3)*(-vx1-vx2-vx3)-cu_sq);
+ (D.f[DIR_PPP])[ktne]=c1o216*(drho+c3o1*( vx1+vx2+vx3)+c9o2*( vx1+vx2+vx3)*( vx1+vx2+vx3)-cu_sq);
+ //(D.f[DIR_MMM])[kbsw]=c1over216*(drho+three*(-vx1-vx2-vx3)+c9over2*(-vx1-vx2-vx3)*(-vx1-vx2-vx3)-cu_sq);
}
q = q_dirBNE[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_MMP])[ktsw]=c1o216*(drho+c3o1*(-vx1-vx2+vx3)+c9o2*(-vx1-vx2+vx3)*(-vx1-vx2+vx3)-cu_sq);
- //(D.f[DIR_PPM])[kbne]=c1over216*(drho+three*( vx1+vx2-vx3)+c9over2*( vx1+vx2-vx3)*( vx1+vx2-vx3)-cu_sq);
+ (D.f[DIR_MMP])[ktsw]=c1o216*(drho+c3o1*(-vx1-vx2+vx3)+c9o2*(-vx1-vx2+vx3)*(-vx1-vx2+vx3)-cu_sq);
+ //(D.f[DIR_PPM])[kbne]=c1over216*(drho+three*( vx1+vx2-vx3)+c9over2*( vx1+vx2-vx3)*( vx1+vx2-vx3)-cu_sq);
}
q = q_dirTSW[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_PPM])[kbne]=c1o216*(drho+c3o1*( vx1+vx2-vx3)+c9o2*( vx1+vx2-vx3)*( vx1+vx2-vx3)-cu_sq);
- //(D.f[DIR_MMP])[ktsw]=c1over216*(drho+three*(-vx1-vx2+vx3)+c9over2*(-vx1-vx2+vx3)*(-vx1-vx2+vx3)-cu_sq);
+ (D.f[DIR_PPM])[kbne]=c1o216*(drho+c3o1*( vx1+vx2-vx3)+c9o2*( vx1+vx2-vx3)*( vx1+vx2-vx3)-cu_sq);
+ //(D.f[DIR_MMP])[ktsw]=c1over216*(drho+three*(-vx1-vx2+vx3)+c9over2*(-vx1-vx2+vx3)*(-vx1-vx2+vx3)-cu_sq);
}
q = q_dirTSE[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_MPM])[kbnw]=c1o216*(drho+c3o1*(-vx1+vx2-vx3)+c9o2*(-vx1+vx2-vx3)*(-vx1+vx2-vx3)-cu_sq);
- //(D.f[DIR_PMP])[ktse]=c1over216*(drho+three*( vx1-vx2+vx3)+c9over2*( vx1-vx2+vx3)*( vx1-vx2+vx3)-cu_sq);
+ (D.f[DIR_MPM])[kbnw]=c1o216*(drho+c3o1*(-vx1+vx2-vx3)+c9o2*(-vx1+vx2-vx3)*(-vx1+vx2-vx3)-cu_sq);
+ //(D.f[DIR_PMP])[ktse]=c1over216*(drho+three*( vx1-vx2+vx3)+c9over2*( vx1-vx2+vx3)*( vx1-vx2+vx3)-cu_sq);
}
q = q_dirBNW[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_PMP])[ktse]=c1o216*(drho+c3o1*( vx1-vx2+vx3)+c9o2*( vx1-vx2+vx3)*( vx1-vx2+vx3)-cu_sq);
- //(D.f[DIR_MPM])[kbnw]=c1over216*(drho+three*(-vx1+vx2-vx3)+c9over2*(-vx1+vx2-vx3)*(-vx1+vx2-vx3)-cu_sq);
+ (D.f[DIR_PMP])[ktse]=c1o216*(drho+c3o1*( vx1-vx2+vx3)+c9o2*( vx1-vx2+vx3)*( vx1-vx2+vx3)-cu_sq);
+ //(D.f[DIR_MPM])[kbnw]=c1over216*(drho+three*(-vx1+vx2-vx3)+c9over2*(-vx1+vx2-vx3)*(-vx1+vx2-vx3)-cu_sq);
}
q = q_dirBSE[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_MPP])[ktnw]=c1o216*(drho+c3o1*(-vx1+vx2+vx3)+c9o2*(-vx1+vx2+vx3)*(-vx1+vx2+vx3)-cu_sq);
- //(D.f[DIR_PMM])[kbse]=c1over216*(drho+three*( vx1-vx2-vx3)+c9over2*( vx1-vx2-vx3)*( vx1-vx2-vx3)-cu_sq);
+ (D.f[DIR_MPP])[ktnw]=c1o216*(drho+c3o1*(-vx1+vx2+vx3)+c9o2*(-vx1+vx2+vx3)*(-vx1+vx2+vx3)-cu_sq);
+ //(D.f[DIR_PMM])[kbse]=c1over216*(drho+three*( vx1-vx2-vx3)+c9over2*( vx1-vx2-vx3)*( vx1-vx2-vx3)-cu_sq);
}
q = q_dirTNW[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_PMM])[kbse]=c1o216*(drho+c3o1*( vx1-vx2-vx3)+c9o2*( vx1-vx2-vx3)*( vx1-vx2-vx3)-cu_sq);
- //(D.f[DIR_MPP])[ktnw]=c1over216*(drho+three*(-vx1+vx2+vx3)+c9over2*(-vx1+vx2+vx3)*(-vx1+vx2+vx3)-cu_sq);
+ (D.f[DIR_PMM])[kbse]=c1o216*(drho+c3o1*( vx1-vx2-vx3)+c9o2*( vx1-vx2-vx3)*( vx1-vx2-vx3)-cu_sq);
+ //(D.f[DIR_MPP])[ktnw]=c1over216*(drho+three*(-vx1+vx2+vx3)+c9over2*(-vx1+vx2+vx3)*(-vx1+vx2+vx3)-cu_sq);
}
}
}
@@ -1903,19 +1903,19 @@ __global__ void QPressDevice27(real* rhoBC,
//////////////////////////////////////////////////////////////////////////////
__global__ void QPressDeviceAntiBB27( real* rhoBC,
- real* vx,
- real* vy,
- real* vz,
- real* DD,
- int* k_Q,
- real* QQ,
- int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* vx,
+ real* vy,
+ real* vz,
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
Distributions27 D;
if (isEvenTimestep==true)
@@ -1947,7 +1947,7 @@ __global__ void QPressDeviceAntiBB27( real* rhoBC,
D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
+ }
else
{
D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -1979,9 +1979,9 @@ __global__ void QPressDeviceAntiBB27( real* rhoBC,
D.f[DIR_MPM ] = &DD[DIR_PMP *size_Mat];
}
////////////////////////////////////////////////////////////////////////////////
- const unsigned x = threadIdx.x; // Globaler x-Index
- const unsigned y = blockIdx.x; // Globaler y-Index
- const unsigned z = blockIdx.y; // Globaler z-Index
+ const unsigned x = threadIdx.x; // Globaler x-Index
+ const unsigned y = blockIdx.x; // Globaler y-Index
+ const unsigned z = blockIdx.y; // Globaler z-Index
const unsigned nx = blockDim.x;
const unsigned ny = gridDim.x;
@@ -1991,11 +1991,11 @@ __global__ void QPressDeviceAntiBB27( real* rhoBC,
if(k<numberOfBCnodes)
{
- real *q_dirE, *q_dirW, *q_dirN, *q_dirS, *q_dirT, *q_dirB,
+ real *q_dirE, *q_dirW, *q_dirN, *q_dirS, *q_dirT, *q_dirB,
*q_dirNE, *q_dirSW, *q_dirSE, *q_dirNW, *q_dirTE, *q_dirBW,
*q_dirBE, *q_dirTW, *q_dirTN, *q_dirBS, *q_dirBN, *q_dirTS,
*q_dirTNE, *q_dirTSW, *q_dirTSE, *q_dirTNW, *q_dirBNE, *q_dirBSW,
- *q_dirBSE, *q_dirBNW;
+ *q_dirBSE, *q_dirBNW;
q_dirE = &QQ[DIR_P00 *numberOfBCnodes];
q_dirW = &QQ[DIR_M00 *numberOfBCnodes];
q_dirN = &QQ[DIR_0P0 *numberOfBCnodes];
@@ -2087,25 +2087,25 @@ __global__ void QPressDeviceAntiBB27( real* rhoBC,
//real vx1, vx2, vx3, drho;
//vx1 = ((f_TSE - f_BNW) - (f_TNW - f_BSE)) + ((f_TNE - f_BSW) - (f_TSW - f_BNE)) +
// ((f_BE - f_TW) + (f_TE - f_BW)) + ((f_SE - f_NW) + (f_NE - f_SW)) +
- // (f_E - f_W);
+ // (f_E - f_W);
//vx2 = (-(f_TSE - f_BNW) + (f_TNW - f_BSE)) + ((f_TNE - f_BSW) - (f_TSW - f_BNE)) +
// ((f_BN - f_TS) + (f_TN - f_BS)) + (-(f_SE - f_NW) + (f_NE - f_SW)) +
- // (f_N - f_S);
+ // (f_N - f_S);
//vx3 = ((f_TSE - f_BNW) + (f_TNW - f_BSE)) + ((f_TNE - f_BSW) + (f_TSW - f_BNE)) +
// (-(f_BN - f_TS) + (f_TN - f_BS)) + ((f_TE - f_BW) - (f_BE - f_TW)) +
- // (f_T - f_B);
+ // (f_T - f_B);
//real cu_sq=c3o2*(vx1*vx1+vx2*vx2+vx3*vx3);
//////////////////////////////////////////////////////////////////////////
real drho = f_ZERO+f_E+f_W+f_N+f_S+f_T+f_B+f_NE+f_SW+f_SE+f_NW+f_TE+f_BW+f_BE+f_TW+f_TN+f_BS+f_BN+f_TS+
- f_TNE+f_TSW+f_TSE+f_TNW+f_BNE+f_BSW+f_BSE+f_BNW;
+ f_TNE+f_TSW+f_TSE+f_TNW+f_BNE+f_BSW+f_BSE+f_BNW;
drho = drho - rhoBC[k];
- drho *= 0.01f;
+ drho *= 0.01f;
////////////////////////////////////////////////////////////////////////////////
- real q;
+ real q;
//deltaRho = (rhoBC[k] + one) / (deltaRho + one);
////////////////////////////////////////////////////////////////////////////////
if (isEvenTimestep==false)
@@ -2137,7 +2137,7 @@ __global__ void QPressDeviceAntiBB27( real* rhoBC,
D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
+ }
else
{
D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -2172,7 +2172,7 @@ __global__ void QPressDeviceAntiBB27( real* rhoBC,
q = q_dirE[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_M00])[kw]=f_W-c2o27*drho;
+ (D.f[DIR_M00])[kw]=f_W-c2o27*drho;
}
q = q_dirW[k];
@@ -2184,19 +2184,19 @@ __global__ void QPressDeviceAntiBB27( real* rhoBC,
q = q_dirN[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_0M0])[ks]=f_S-c2o27*drho;
+ (D.f[DIR_0M0])[ks]=f_S-c2o27*drho;
}
q = q_dirS[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_0P0])[kn]=f_N-c2o27*drho;
+ (D.f[DIR_0P0])[kn]=f_N-c2o27*drho;
}
q = q_dirT[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_00M])[kb]=f_B-c2o27*drho;
+ (D.f[DIR_00M])[kb]=f_B-c2o27*drho;
}
q = q_dirB[k];
@@ -2232,13 +2232,13 @@ __global__ void QPressDeviceAntiBB27( real* rhoBC,
q = q_dirTE[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_M0M])[kbw]=f_BW-c1o54*drho;
+ (D.f[DIR_M0M])[kbw]=f_BW-c1o54*drho;
}
q = q_dirBW[k];
if (q>=c0o1 && q<=c1o1)
{
- (D.f[DIR_P0P])[kte]=f_TE-c1o54*drho;
+ (D.f[DIR_P0P])[kte]=f_TE-c1o54*drho;
}
q = q_dirBE[k];
@@ -2368,20 +2368,20 @@ __global__ void QPressDeviceAntiBB27( real* rhoBC,
//////////////////////////////////////////////////////////////////////////////
__global__ void QPressDeviceFixBackflow27( real* rhoBC,
- real* DD,
- int* k_Q,
- int numberOfBCnodes,
- real om1,
+ real* DD,
+ int* k_Q,
+ int numberOfBCnodes,
+ real om1,
unsigned int* neighborX,
unsigned int* neighborY,
unsigned int* neighborZ,
- unsigned int size_Mat,
+ unsigned int size_Mat,
bool isEvenTimestep)
{
////////////////////////////////////////////////////////////////////////////////
- const unsigned x = threadIdx.x; // Globaler x-Index
- const unsigned y = blockIdx.x; // Globaler y-Index
- const unsigned z = blockIdx.y; // Globaler z-Index
+ const unsigned x = threadIdx.x; // Globaler x-Index
+ const unsigned y = blockIdx.x; // Globaler y-Index
+ const unsigned z = blockIdx.y; // Globaler z-Index
const unsigned nx = blockDim.x;
const unsigned ny = gridDim.x;
@@ -2456,7 +2456,7 @@ __global__ void QPressDeviceFixBackflow27( real* rhoBC,
D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
+ }
else
{
D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -2559,20 +2559,20 @@ __global__ void QPressDeviceFixBackflow27( real* rhoBC,
//////////////////////////////////////////////////////////////////////////////
__global__ void QPressDeviceDirDepBot27( real* rhoBC,
- real* DD,
- int* k_Q,
- int numberOfBCnodes,
- real om1,
+ real* DD,
+ int* k_Q,
+ int numberOfBCnodes,
+ real om1,
unsigned int* neighborX,
unsigned int* neighborY,
unsigned int* neighborZ,
- unsigned int size_Mat,
+ unsigned int size_Mat,
bool isEvenTimestep)
{
////////////////////////////////////////////////////////////////////////////////
- const unsigned x = threadIdx.x; // Globaler x-Index
- const unsigned y = blockIdx.x; // Globaler y-Index
- const unsigned z = blockIdx.y; // Globaler z-Index
+ const unsigned x = threadIdx.x; // Globaler x-Index
+ const unsigned y = blockIdx.x; // Globaler y-Index
+ const unsigned z = blockIdx.y; // Globaler z-Index
const unsigned nx = blockDim.x;
const unsigned ny = gridDim.x;
@@ -2647,7 +2647,7 @@ __global__ void QPressDeviceDirDepBot27( real* rhoBC,
D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
+ }
else
{
D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -2803,16 +2803,16 @@ __host__ __device__ real computeOutflowDistribution(const real* const &f, const
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
__global__ void QPressNoRhoDevice27( real* rhoBC,
- real* distributions,
- int* k_Q,
- int* k_N,
- int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int numberOfLBnodes,
- bool isEvenTimestep,
+ real* distributions,
+ int* k_Q,
+ int* k_N,
+ int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int numberOfLBnodes,
+ bool isEvenTimestep,
int direction)
{
////////////////////////////////////////////////////////////////////////////////
@@ -2885,8 +2885,8 @@ __global__ void QPressNoRhoDevice27( real* rhoBC,
unsigned int k1bsw = neighborZ[k1sw];
////////////////////////////////////////////////////////////////////////////////
Distributions27 dist;
- getPointersToDistributions(dist, distributions, numberOfLBnodes, isEvenTimestep);
- real f[27], f1[27];
+ getPointersToDistributions(dist, distributions, numberOfLBnodes, isEvenTimestep);
+ real f[27], f1[27];
//////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
f1[DIR_P00] = (dist.f[DIR_P00])[k1e ];
f1[DIR_M00] = (dist.f[DIR_M00])[k1w ];
@@ -2986,9 +2986,9 @@ __global__ void QPressNoRhoDevice27( real* rhoBC,
(dist.f[DIR_MPP])[ktnw ] = computeOutflowDistribution(f, f1, DIR_MPP, cs);
(dist.f[DIR_PPM])[kbne ] = computeOutflowDistribution(f, f1, DIR_PPM, cs);
(dist.f[DIR_MPM])[kbnw ] = computeOutflowDistribution(f, f1, DIR_MPM, cs);
- break;
+ break;
- case ZPZ:
+ case ZPZ:
(dist.f[DIR_0M0])[ks ] = computeOutflowDistribution(f, f1, DIR_0M0, cs);
(dist.f[DIR_PM0])[kse ] = computeOutflowDistribution(f, f1, DIR_PM0, cs);
(dist.f[DIR_MM0])[ksw ] = computeOutflowDistribution(f, f1, DIR_MM0, cs);
@@ -3009,7 +3009,7 @@ __global__ void QPressNoRhoDevice27( real* rhoBC,
(dist.f[DIR_PPP])[ktne ] = computeOutflowDistribution(f, f1, DIR_PPP, cs);
(dist.f[DIR_MPP])[ktnw ] = computeOutflowDistribution(f, f1, DIR_MPP, cs);
(dist.f[DIR_PMP])[ktse ] = computeOutflowDistribution(f, f1, DIR_PMP, cs);
- (dist.f[DIR_MMP])[ktsw ] = computeOutflowDistribution(f, f1, DIR_MMP, cs);
+ (dist.f[DIR_MMP])[ktsw ] = computeOutflowDistribution(f, f1, DIR_MMP, cs);
break;
case ZZP:
@@ -3021,7 +3021,7 @@ __global__ void QPressNoRhoDevice27( real* rhoBC,
(dist.f[DIR_PPM])[kbne ] = computeOutflowDistribution(f, f1, DIR_PPM, cs);
(dist.f[DIR_MPM])[kbnw ] = computeOutflowDistribution(f, f1, DIR_MPM, cs);
(dist.f[DIR_PMM])[kbse ] = computeOutflowDistribution(f, f1, DIR_PMM, cs);
- (dist.f[DIR_MMM])[kbsw ] = computeOutflowDistribution(f, f1, DIR_MMM, cs);
+ (dist.f[DIR_MMM])[kbsw ] = computeOutflowDistribution(f, f1, DIR_MMM, cs);
break;
default:
break;
@@ -3036,22 +3036,22 @@ __host__ __device__ real computeOutflowDistribution(const real* const &f, const
}
__global__ void QPressZeroRhoOutflowDevice27( real* rhoBC,
- real* distributions,
- int* k_Q,
- int* k_N,
- int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int numberOfLBnodes,
- bool isEvenTimestep,
+ real* distributions,
+ int* k_Q,
+ int* k_N,
+ int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int numberOfLBnodes,
+ bool isEvenTimestep,
int direction,
real densityCorrectionFactor)
{
////////////////////////////////////////////////////////////////////////////////
const unsigned k = vf::gpu::getNodeIndex();
-
+
//////////////////////////////////////////////////////////////////////////
if(k>=numberOfBCnodes) return;
@@ -3079,8 +3079,8 @@ __global__ void QPressZeroRhoOutflowDevice27( real* rhoBC,
uint kN_MMM = neighborZ[k_MM0];
////////////////////////////////////////////////////////////////////////////////
Distributions27 dist;
- getPointersToDistributions(dist, distributions, numberOfLBnodes, isEvenTimestep);
- real f[27], fN[27];
+ getPointersToDistributions(dist, distributions, numberOfLBnodes, isEvenTimestep);
+ real f[27], fN[27];
//////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
f[DIR_000] = (dist.f[DIR_000])[k_000];
f[DIR_P00] = (dist.f[DIR_P00])[k_000];
@@ -3139,9 +3139,9 @@ __global__ void QPressZeroRhoOutflowDevice27( real* rhoBC,
fN[DIR_MMM] = (dist.f[DIR_MMM])[kN_MMM];
//////////////////////////////////////////////////////////////////////////
real drho = vf::lbm::getDensity(f);
-
+
real rhoCorrection = densityCorrectionFactor*drho;
-
+
real cs = c1o1 / sqrtf(c3o1);
getPointersToDistributions(dist, distributions, numberOfLBnodes, !isEvenTimestep);
@@ -3182,9 +3182,9 @@ __global__ void QPressZeroRhoOutflowDevice27( real* rhoBC,
(dist.f[DIR_MPP])[k_M00] = computeOutflowDistribution(f, fN, DIR_MPP, rhoCorrection, cs, c1o216);
(dist.f[DIR_PPM])[k_00M] = computeOutflowDistribution(f, fN, DIR_PPM, rhoCorrection, cs, c1o216);
(dist.f[DIR_MPM])[k_M0M] = computeOutflowDistribution(f, fN, DIR_MPM, rhoCorrection, cs, c1o216);
- break;
+ break;
- case ZPZ:
+ case ZPZ:
(dist.f[DIR_0M0])[k_0M0] =computeOutflowDistribution(f, fN, DIR_0M0, rhoCorrection, cs, c2o27);
(dist.f[DIR_PM0])[k_0M0] =computeOutflowDistribution(f, fN, DIR_PM0, rhoCorrection, cs, c1o54);
(dist.f[DIR_MM0])[k_MM0] =computeOutflowDistribution(f, fN, DIR_MM0, rhoCorrection, cs, c1o54);
@@ -3205,7 +3205,7 @@ __global__ void QPressZeroRhoOutflowDevice27( real* rhoBC,
(dist.f[DIR_PPP])[k_000] = computeOutflowDistribution(f, fN, DIR_PPP, rhoCorrection, cs, c1o216);
(dist.f[DIR_MPP])[k_M00] = computeOutflowDistribution(f, fN, DIR_MPP, rhoCorrection, cs, c1o216);
(dist.f[DIR_PMP])[k_0M0] = computeOutflowDistribution(f, fN, DIR_PMP, rhoCorrection, cs, c1o216);
- (dist.f[DIR_MMP])[k_MM0] = computeOutflowDistribution(f, fN, DIR_MMP, rhoCorrection, cs, c1o216);
+ (dist.f[DIR_MMP])[k_MM0] = computeOutflowDistribution(f, fN, DIR_MMP, rhoCorrection, cs, c1o216);
break;
case ZZP:
@@ -3217,7 +3217,7 @@ __global__ void QPressZeroRhoOutflowDevice27( real* rhoBC,
(dist.f[DIR_PPM])[k_00M] = computeOutflowDistribution(f, fN, DIR_PPM, rhoCorrection, cs, c1o216);
(dist.f[DIR_MPM])[k_M0M] = computeOutflowDistribution(f, fN, DIR_MPM, rhoCorrection, cs, c1o216);
(dist.f[DIR_PMM])[k_0MM] = computeOutflowDistribution(f, fN, DIR_PMM, rhoCorrection, cs, c1o216);
- (dist.f[DIR_MMM])[k_MMM] = computeOutflowDistribution(f, fN, DIR_MMM, rhoCorrection, cs, c1o216);
+ (dist.f[DIR_MMM])[k_MMM] = computeOutflowDistribution(f, fN, DIR_MMM, rhoCorrection, cs, c1o216);
break;
default:
break;
@@ -3256,21 +3256,21 @@ __global__ void QPressZeroRhoOutflowDevice27( real* rhoBC,
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
__global__ void QPressDeviceOld27(real* rhoBC,
- real* DD,
- int* k_Q,
- int* k_N,
- int numberOfBCnodes,
- real om1,
+ real* DD,
+ int* k_Q,
+ int* k_N,
+ int numberOfBCnodes,
+ real om1,
unsigned int* neighborX,
unsigned int* neighborY,
unsigned int* neighborZ,
- unsigned int size_Mat,
+ unsigned int size_Mat,
bool isEvenTimestep)
{
////////////////////////////////////////////////////////////////////////////////
- const unsigned x = threadIdx.x; // Globaler x-Index
- const unsigned y = blockIdx.x; // Globaler y-Index
- const unsigned z = blockIdx.y; // Globaler z-Index
+ const unsigned x = threadIdx.x; // Globaler x-Index
+ const unsigned y = blockIdx.x; // Globaler y-Index
+ const unsigned z = blockIdx.y; // Globaler z-Index
const unsigned nx = blockDim.x;
const unsigned ny = gridDim.x;
@@ -3371,7 +3371,7 @@ __global__ void QPressDeviceOld27(real* rhoBC,
D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
+ }
else
{
D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -3438,8 +3438,8 @@ __global__ void QPressDeviceOld27(real* rhoBC,
real drho1 = f1_ZERO+f1_E+f1_W+f1_N+f1_S+f1_T+f1_B+f1_NE+f1_SW+f1_SE+f1_NW+f1_TE+f1_BW+f1_BE+f1_TW+f1_TN+f1_BS+f1_BN+f1_TS+
f1_TNE+f1_TSW+f1_TSE+f1_TNW+f1_BNE+f1_BSW+f1_BSE+f1_BNW;
- //drho1 = (drho1 + rhoBC[k])/2.f;
- drho1 = drho1 - rhoBC[k];
+ //drho1 = (drho1 + rhoBC[k])/2.f;
+ drho1 = drho1 - rhoBC[k];
//////////////////////////////////////////////////////////////////////////
__syncthreads();
@@ -3470,7 +3470,7 @@ __global__ void QPressDeviceOld27(real* rhoBC,
(D.f[DIR_PPM ])[kbne ] = f1_TSW -c1o216*drho1; // c1o100; // zero; //
(D.f[DIR_MMM ])[kbsw ] = f1_TNE -c1o216*drho1; // c1o100; // zero; //
(D.f[DIR_PMM ])[kbse ] = f1_TNW -c1o216*drho1; // c1o100; // zero; //
- (D.f[DIR_MPM ])[kbnw ] = f1_TSE -c1o216*drho1; // c1o100; // zero; //
+ (D.f[DIR_MPM ])[kbnw ] = f1_TSE -c1o216*drho1; // c1o100; // zero; //
}
}
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
@@ -3515,22 +3515,22 @@ __global__ void QPressDeviceOld27(real* rhoBC,
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
__global__ void QPressDeviceEQZ27(real* rhoBC,
- real* DD,
- int* k_Q,
+ real* DD,
+ int* k_Q,
int* k_N,
- real* kTestRE,
- int numberOfBCnodes,
- real om1,
+ real* kTestRE,
+ int numberOfBCnodes,
+ real om1,
unsigned int* neighborX,
unsigned int* neighborY,
unsigned int* neighborZ,
- unsigned int size_Mat,
+ unsigned int size_Mat,
bool isEvenTimestep)
{
////////////////////////////////////////////////////////////////////////////////
- const unsigned x = threadIdx.x; // Globaler x-Index
- const unsigned y = blockIdx.x; // Globaler y-Index
- const unsigned z = blockIdx.y; // Globaler z-Index
+ const unsigned x = threadIdx.x; // Globaler x-Index
+ const unsigned y = blockIdx.x; // Globaler y-Index
+ const unsigned z = blockIdx.y; // Globaler z-Index
const unsigned nx = blockDim.x;
const unsigned ny = gridDim.x;
@@ -3631,7 +3631,7 @@ __global__ void QPressDeviceEQZ27(real* rhoBC,
D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
+ }
else
{
D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -3755,17 +3755,17 @@ __global__ void QPressDeviceEQZ27(real* rhoBC,
// //////////////////////////////////////////////////////////////////////////
// real drho1 = f1_ZERO+f1_E+f1_W+f1_N+f1_S+f1_T+f1_B+f1_NE+f1_SW+f1_SE+f1_NW+f1_TE+f1_BW+f1_BE+f1_TW+f1_TN+f1_BS+f1_BN+f1_TS+ f1_TNE+f1_TSW+f1_TSE+f1_TNW+f1_BNE+f1_BSW+f1_BSE+f1_BNW;
- //real vx1 = (((f1_TNE-f1_BSW)+(f1_BSE-f1_TNW)+(f1_BNE-f1_TSW)+(f1_TSE-f1_BNW)) + (((f1_NE-f1_SW)+(f1_TE-f1_BW))+((f1_SE-f1_NW)+(f1_BE-f1_TW))) + (f1_E-f1_W)) / (one + drho1);
- //real vx2 = (((f1_TNE-f1_BSW)+(f1_TNW-f1_BSE)+(f1_BNE-f1_TSW)+(f1_BNW-f1_TSE)) + (((f1_NE-f1_SW)+(f1_TN-f1_BS))+((f1_BN-f1_TS)+(f1_NW-f1_SE))) + (f1_N-f1_S)) / (one + drho1);
- //real vx3 = (((f1_TNE-f1_BSW)+(f1_TNW-f1_BSE)+(f1_TSW-f1_BNE)+(f1_TSE-f1_BNW)) + (((f1_TE-f1_BW)+(f1_TN-f1_BS))+((f1_TW-f1_BE)+(f1_TS-f1_BN))) + (f1_T-f1_B)) / (one + drho1);
+ //real vx1 = (((f1_TNE-f1_BSW)+(f1_BSE-f1_TNW)+(f1_BNE-f1_TSW)+(f1_TSE-f1_BNW)) + (((f1_NE-f1_SW)+(f1_TE-f1_BW))+((f1_SE-f1_NW)+(f1_BE-f1_TW))) + (f1_E-f1_W)) / (one + drho1);
+ //real vx2 = (((f1_TNE-f1_BSW)+(f1_TNW-f1_BSE)+(f1_BNE-f1_TSW)+(f1_BNW-f1_TSE)) + (((f1_NE-f1_SW)+(f1_TN-f1_BS))+((f1_BN-f1_TS)+(f1_NW-f1_SE))) + (f1_N-f1_S)) / (one + drho1);
+ //real vx3 = (((f1_TNE-f1_BSW)+(f1_TNW-f1_BSE)+(f1_TSW-f1_BNE)+(f1_TSE-f1_BNW)) + (((f1_TE-f1_BW)+(f1_TN-f1_BS))+((f1_TW-f1_BE)+(f1_TS-f1_BN))) + (f1_T-f1_B)) / (one + drho1);
// //////////////////////////////////////////////////////////////////////////
- ////real omega = om1;
+ ////real omega = om1;
// real cusq = c3o2*(vx1*vx1+vx2*vx2+vx3*vx3);
// //////////////////////////////////////////////////////////////////////////
- ////T�st MK
- ////if(vx1 < zero) vx1 = zero;
+ ////T�st MK
+ ////if(vx1 < zero) vx1 = zero;
// //////////////////////////////////////////////////////////////////////////
- ////becomes higher with neighbor source and lower with local source
+ ////becomes higher with neighbor source and lower with local source
// //real fZERO = c8over27* (rhoBC[k]-(one + rhoBC[k])*(cusq)) ;
// //real fE = c2over27* (rhoBC[k]+(one + rhoBC[k])*(three*( vx1 )+c9over2*( vx1 )*( vx1 )-cusq));
// //real fW = c2over27* (rhoBC[k]+(one + rhoBC[k])*(three*(-vx1 )+c9over2*(-vx1 )*(-vx1 )-cusq));
@@ -3794,7 +3794,7 @@ __global__ void QPressDeviceEQZ27(real* rhoBC,
// //real fBSE = c1over216* (rhoBC[k]+(one + rhoBC[k])*(three*( vx1-vx2-vx3)+c9over2*( vx1-vx2-vx3)*( vx1-vx2-vx3)-cusq));
// //real fTNW = c1over216* (rhoBC[k]+(one + rhoBC[k])*(three*(-vx1+vx2+vx3)+c9over2*(-vx1+vx2+vx3)*(-vx1+vx2+vx3)-cusq));
// //////////////////////////////////////////////////////////////////////////
- //// based on VirtualFluids (kucher + fard)
+ //// based on VirtualFluids (kucher + fard)
// real fZERO = c8over27 * rhoBC[k] * (one - cusq);
// real fE = c2over27 * rhoBC[k] * (one + three * ( vx1 ) + c9over2 * ( vx1 ) * ( vx1 ) - cusq);
// real fW = c2over27 * rhoBC[k] * (one + three * (-vx1 ) + c9over2 * (-vx1 ) * (-vx1 ) - cusq);
@@ -3823,7 +3823,7 @@ __global__ void QPressDeviceEQZ27(real* rhoBC,
// real fBSE = c1over216 * rhoBC[k] * (one + three * ( vx1-vx2-vx3) + c9over2 * ( vx1-vx2-vx3) * ( vx1-vx2-vx3) - cusq);
// real fTNW = c1over216 * rhoBC[k] * (one + three * (-vx1+vx2+vx3) + c9over2 * (-vx1+vx2+vx3) * (-vx1+vx2+vx3) - cusq);
//// ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
- //////test
+ //////test
//// real fZERO = c8over27 * ((drho1 + rhoBC[k]) / two) * (one - cusq);
//// real fE = c2over27 * ((drho1 + rhoBC[k]) / two) * (one + three * ( vx1 ) + c9over2 * ( vx1 ) * ( vx1 ) - cusq);
//// real fW = c2over27 * ((drho1 + rhoBC[k]) / two) * (one + three * (-vx1 ) + c9over2 * (-vx1 ) * (-vx1 ) - cusq);
@@ -3852,190 +3852,190 @@ __global__ void QPressDeviceEQZ27(real* rhoBC,
//// real fBSE = c1over216 * ((drho1 + rhoBC[k]) / two) * (one + three * ( vx1-vx2-vx3) + c9over2 * ( vx1-vx2-vx3) * ( vx1-vx2-vx3) - cusq);
//// real fTNW = c1over216 * ((drho1 + rhoBC[k]) / two) * (one + three * (-vx1+vx2+vx3) + c9over2 * (-vx1+vx2+vx3) * (-vx1+vx2+vx3) - cusq);
- //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+ //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
// based on BGK Plus Comp
- //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
- //double mfabb = (D.f[DIR_P00 ])[k1e ];
- //double mfcbb = (D.f[DIR_M00 ])[k1w ];
- //double mfbab = (D.f[DIR_0P0 ])[k1n ];
- //double mfbcb = (D.f[DIR_0M0 ])[k1s ];
- //double mfbba = (D.f[DIR_00P ])[k1t ];
- //double mfbbc = (D.f[DIR_00M ])[k1b ];
- //double mfaab = (D.f[DIR_PP0 ])[k1ne ];
- //double mfccb = (D.f[DIR_MM0 ])[k1sw ];
- //double mfacb = (D.f[DIR_PM0 ])[k1se ];
- //double mfcab = (D.f[DIR_MP0 ])[k1nw ];
- //double mfaba = (D.f[DIR_P0P ])[k1te ];
- //double mfcbc = (D.f[DIR_M0M ])[k1bw ];
- //double mfabc = (D.f[DIR_P0M ])[k1be ];
- //double mfcba = (D.f[DIR_M0P ])[k1tw ];
- //double mfbaa = (D.f[DIR_0PP ])[k1tn ];
- //double mfbcc = (D.f[DIR_0MM ])[k1bs ];
- //double mfbac = (D.f[DIR_0PM ])[k1bn ];
- //double mfbca = (D.f[DIR_0MP ])[k1ts ];
- //double mfbbb = (D.f[DIR_000])[k1zero];
- //double mfaaa = (D.f[DIR_PPP ])[k1tne ];
- //double mfcca = (D.f[DIR_MMP ])[k1tsw ];
- //double mfaca = (D.f[DIR_PMP ])[k1tse ];
- //double mfcaa = (D.f[DIR_MPP ])[k1tnw ];
- //double mfaac = (D.f[DIR_PPM ])[k1bne ];
- //double mfccc = (D.f[DIR_MMM ])[k1bsw ];
- //double mfacc = (D.f[DIR_PMM ])[k1bse ];
- //double mfcac = (D.f[DIR_MPM ])[k1bnw ];
- real mfabb = (D.f[DIR_P00 ])[k1e ];
- real mfcbb = (D.f[DIR_M00 ])[k1w ];
- real mfbab = (D.f[DIR_0P0 ])[k1n ];
- real mfbcb = (D.f[DIR_0M0 ])[k1s ];
- real mfbba = (D.f[DIR_00P ])[k1t ];
- real mfbbc = (D.f[DIR_00M ])[k1b ];
- real mfaab = (D.f[DIR_PP0 ])[k1ne ];
- real mfccb = (D.f[DIR_MM0 ])[k1sw ];
- real mfacb = (D.f[DIR_PM0 ])[k1se ];
- real mfcab = (D.f[DIR_MP0 ])[k1nw ];
- real mfaba = (D.f[DIR_P0P ])[k1te ];
- real mfcbc = (D.f[DIR_M0M ])[k1bw ];
- real mfabc = (D.f[DIR_P0M ])[k1be ];
- real mfcba = (D.f[DIR_M0P ])[k1tw ];
- real mfbaa = (D.f[DIR_0PP ])[k1tn ];
- real mfbcc = (D.f[DIR_0MM ])[k1bs ];
- real mfbac = (D.f[DIR_0PM ])[k1bn ];
- real mfbca = (D.f[DIR_0MP ])[k1ts ];
- real mfbbb = (D.f[DIR_000])[k1zero];
- real mfaaa = (D.f[DIR_PPP ])[k1tne ];
- real mfcca = (D.f[DIR_MMP ])[k1tsw ];
- real mfaca = (D.f[DIR_PMP ])[k1tse ];
- real mfcaa = (D.f[DIR_MPP ])[k1tnw ];
- real mfaac = (D.f[DIR_PPM ])[k1bne ];
- real mfccc = (D.f[DIR_MMM ])[k1bsw ];
- real mfacc = (D.f[DIR_PMM ])[k1bse ];
- real mfcac = (D.f[DIR_MPM ])[k1bnw ];
-
- //real mfcbb = (D.f[DIR_P00 ])[ke ];
- //real mfabb = (D.f[DIR_M00 ])[kw ];
- //real mfbcb = (D.f[DIR_0P0 ])[kn ];
- //real mfbab = (D.f[DIR_0M0 ])[ks ];
- //real mfbbc = (D.f[DIR_00P ])[kt ];
- //real mfbba = (D.f[DIR_00M ])[kb ];
- //real mfccb = (D.f[DIR_PP0 ])[kne ];
- //real mfaab = (D.f[DIR_MM0 ])[ksw ];
- //real mfcab = (D.f[DIR_PM0 ])[kse ];
- //real mfacb = (D.f[DIR_MP0 ])[knw ];
- //real mfcbc = (D.f[DIR_P0P ])[kte ];
- //real mfaba = (D.f[DIR_M0M ])[kbw ];
- //real mfcba = (D.f[DIR_P0M ])[kbe ];
- //real mfabc = (D.f[DIR_M0P ])[ktw ];
- //real mfbcc = (D.f[DIR_0PP ])[ktn ];
- //real mfbaa = (D.f[DIR_0MM ])[kbs ];
- //real mfbca = (D.f[DIR_0PM ])[kbn ];
- //real mfbac = (D.f[DIR_0MP ])[kts ];
- //real mfbbb = (D.f[DIR_000])[kzero];
- //real mfccc = (D.f[DIR_PPP ])[ktne ];
- //real mfaac = (D.f[DIR_MMP ])[ktsw ];
- //real mfcac = (D.f[DIR_PMP ])[ktse ];
- //real mfacc = (D.f[DIR_MPP ])[ktnw ];
- //real mfcca = (D.f[DIR_PPM ])[kbne ];
- //real mfaaa = (D.f[DIR_MMM ])[kbsw ];
- //real mfcaa = (D.f[DIR_PMM ])[kbse ];
- //real mfaca = (D.f[DIR_MPM ])[kbnw ];
- ////////////////////////////////////////////////////////////////////////////////////
- //real rho = (((((mfccc+mfaaa) + (mfaca+mfcac)) + ((mfacc+mfcaa) + (mfaac+mfcca))) +
- // (((mfbac+mfbca) + (mfbaa+mfbcc)) + ((mfabc+mfcba) + (mfaba+mfcbc)) + ((mfacb+mfcab) + (mfaab+mfccb))) +
- // ((mfabb+mfcbb) + (mfbab+mfbcb)) + (mfbba+mfbbc)) + mfbbb) + one;//!!!!Achtung + one
- ////////////////////////////////////////////////////////////////////////////////////
- real rho = rhoBC[k];
- ////////////////////////////////////////////////////////////////////////////////////
- real OoRho = c1o1 / (rho * 1.5f);
- ////////////////////////////////////////////////////////////////////////////////////
- real vvx = ((((mfccc-mfaaa) + (mfcac-mfaca)) + ((mfcaa-mfacc) + (mfcca-mfaac))) +
- (((mfcba-mfabc) + (mfcbc-mfaba)) + ((mfcab-mfacb) + (mfccb-mfaab))) +
- (mfcbb-mfabb)) * OoRho;
- real vvy =((((mfccc-mfaaa) + (mfaca-mfcac)) + ((mfacc-mfcaa) + (mfcca-mfaac))) +
- (((mfbca-mfbac) + (mfbcc-mfbaa)) + ((mfacb-mfcab) + (mfccb-mfaab))) +
- (mfbcb-mfbab)) * OoRho;
- real vvz =((((mfccc-mfaaa) + (mfcac-mfaca)) + ((mfacc-mfcaa) + (mfaac-mfcca))) +
- (((mfbac-mfbca) + (mfbcc-mfbaa)) + ((mfabc-mfcba) + (mfcbc-mfaba))) +
- (mfbbc-mfbba)) * OoRho;
- /////////////////////////
- //Test Values
- //double vvx = 0.016;
- //double vvy = zero;
- //double vvz = zero;
- ////////////////////////////////////////////////////////////////////////////////////////
- ////round off error test
- //if(vvx!=zero){
- // (kDistTest.f[DIR_P00 ])[k] = mfabb;
- // (kDistTest.f[DIR_M00 ])[k] = mfcbb;
- // (kDistTest.f[DIR_0P0 ])[k] = mfbab;
- // (kDistTest.f[DIR_0M0 ])[k] = mfbcb;
- // (kDistTest.f[DIR_00P ])[k] = mfbba;
- // (kDistTest.f[DIR_00M ])[k] = mfbbc;
- // (kDistTest.f[DIR_PP0 ])[k] = mfaab;
- // (kDistTest.f[DIR_MM0 ])[k] = mfccb;
- // (kDistTest.f[DIR_PM0 ])[k] = mfacb;
- // (kDistTest.f[DIR_MP0 ])[k] = mfcab;
- // (kDistTest.f[DIR_P0P ])[k] = mfaba;
- // (kDistTest.f[DIR_M0M ])[k] = mfcbc;
- // (kDistTest.f[DIR_P0M ])[k] = mfabc;
- // (kDistTest.f[DIR_M0P ])[k] = mfcba;
- // (kDistTest.f[DIR_0PP ])[k] = mfbaa;
- // (kDistTest.f[DIR_0MM ])[k] = mfbcc;
- // (kDistTest.f[DIR_0PM ])[k] = mfbac;
- // (kDistTest.f[DIR_0MP ])[k] = mfbca;
- // (kDistTest.f[DIR_000])[k] = KQK;
- // (kDistTest.f[DIR_PPP ])[k] = mfaaa;
- // (kDistTest.f[DIR_MMP ])[k] = mfcca;
- // (kDistTest.f[DIR_PMP ])[k] = mfaca;
- // (kDistTest.f[DIR_MPP ])[k] = mfcaa;
- // (kDistTest.f[DIR_PPM ])[k] = mfaac;
- // (kDistTest.f[DIR_MMM ])[k] = mfccc;
- // (kDistTest.f[DIR_PMM ])[k] = mfacc;
- // (kDistTest.f[DIR_MPM ])[k] = mfcac;
- //}else{
- // (kDistTest.f[DIR_P00 ])[k] = zero;
- // (kDistTest.f[DIR_M00 ])[k] = zero;
- // (kDistTest.f[DIR_0P0 ])[k] = zero;
- // (kDistTest.f[DIR_0M0 ])[k] = zero;
- // (kDistTest.f[DIR_00P ])[k] = zero;
- // (kDistTest.f[DIR_00M ])[k] = zero;
- // (kDistTest.f[DIR_PP0 ])[k] = zero;
- // (kDistTest.f[DIR_MM0 ])[k] = zero;
- // (kDistTest.f[DIR_PM0 ])[k] = zero;
- // (kDistTest.f[DIR_MP0 ])[k] = zero;
- // (kDistTest.f[DIR_P0P ])[k] = zero;
- // (kDistTest.f[DIR_M0M ])[k] = zero;
- // (kDistTest.f[DIR_P0M ])[k] = zero;
- // (kDistTest.f[DIR_M0P ])[k] = zero;
- // (kDistTest.f[DIR_0PP ])[k] = zero;
- // (kDistTest.f[DIR_0MM ])[k] = zero;
- // (kDistTest.f[DIR_0PM ])[k] = zero;
- // (kDistTest.f[DIR_0MP ])[k] = zero;
- // (kDistTest.f[DIR_000])[k] = zero;
- // (kDistTest.f[DIR_PPP ])[k] = zero;
- // (kDistTest.f[DIR_MMP ])[k] = zero;
- // (kDistTest.f[DIR_PMP ])[k] = zero;
- // (kDistTest.f[DIR_MPP ])[k] = zero;
- // (kDistTest.f[DIR_PPM ])[k] = zero;
- // (kDistTest.f[DIR_MMM ])[k] = zero;
- // (kDistTest.f[DIR_PMM ])[k] = zero;
- // (kDistTest.f[DIR_MPM ])[k] = zero;
- //}
-
- //////////////////////////////////////////////////////////////////////////////////////
- //// first bad fix for negative x velocity
- ////if(vvx > zero) vvx = zero;
- //////////////////////////////////////////////////////////////////////////////////////
- ////// second bad fix for negative x velocity
- ////if(vvx > zero){
- //// vvx = -vvx;
- //// vvy = -vvy;
- //// vvz = -vvz;
- ////}
- ////////////////////////////////////////////////////////////////////////////////////
- double vx2 = vvx * vvx;
- double vy2 = vvy * vvy;
- double vz2 = vvz * vvz;
- //////////////////////////////////////////////////////////////////////////////////
- //original
+ //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+ //double mfabb = (D.f[DIR_P00 ])[k1e ];
+ //double mfcbb = (D.f[DIR_M00 ])[k1w ];
+ //double mfbab = (D.f[DIR_0P0 ])[k1n ];
+ //double mfbcb = (D.f[DIR_0M0 ])[k1s ];
+ //double mfbba = (D.f[DIR_00P ])[k1t ];
+ //double mfbbc = (D.f[DIR_00M ])[k1b ];
+ //double mfaab = (D.f[DIR_PP0 ])[k1ne ];
+ //double mfccb = (D.f[DIR_MM0 ])[k1sw ];
+ //double mfacb = (D.f[DIR_PM0 ])[k1se ];
+ //double mfcab = (D.f[DIR_MP0 ])[k1nw ];
+ //double mfaba = (D.f[DIR_P0P ])[k1te ];
+ //double mfcbc = (D.f[DIR_M0M ])[k1bw ];
+ //double mfabc = (D.f[DIR_P0M ])[k1be ];
+ //double mfcba = (D.f[DIR_M0P ])[k1tw ];
+ //double mfbaa = (D.f[DIR_0PP ])[k1tn ];
+ //double mfbcc = (D.f[DIR_0MM ])[k1bs ];
+ //double mfbac = (D.f[DIR_0PM ])[k1bn ];
+ //double mfbca = (D.f[DIR_0MP ])[k1ts ];
+ //double mfbbb = (D.f[DIR_000])[k1zero];
+ //double mfaaa = (D.f[DIR_PPP ])[k1tne ];
+ //double mfcca = (D.f[DIR_MMP ])[k1tsw ];
+ //double mfaca = (D.f[DIR_PMP ])[k1tse ];
+ //double mfcaa = (D.f[DIR_MPP ])[k1tnw ];
+ //double mfaac = (D.f[DIR_PPM ])[k1bne ];
+ //double mfccc = (D.f[DIR_MMM ])[k1bsw ];
+ //double mfacc = (D.f[DIR_PMM ])[k1bse ];
+ //double mfcac = (D.f[DIR_MPM ])[k1bnw ];
+ real mfabb = (D.f[DIR_P00 ])[k1e ];
+ real mfcbb = (D.f[DIR_M00 ])[k1w ];
+ real mfbab = (D.f[DIR_0P0 ])[k1n ];
+ real mfbcb = (D.f[DIR_0M0 ])[k1s ];
+ real mfbba = (D.f[DIR_00P ])[k1t ];
+ real mfbbc = (D.f[DIR_00M ])[k1b ];
+ real mfaab = (D.f[DIR_PP0 ])[k1ne ];
+ real mfccb = (D.f[DIR_MM0 ])[k1sw ];
+ real mfacb = (D.f[DIR_PM0 ])[k1se ];
+ real mfcab = (D.f[DIR_MP0 ])[k1nw ];
+ real mfaba = (D.f[DIR_P0P ])[k1te ];
+ real mfcbc = (D.f[DIR_M0M ])[k1bw ];
+ real mfabc = (D.f[DIR_P0M ])[k1be ];
+ real mfcba = (D.f[DIR_M0P ])[k1tw ];
+ real mfbaa = (D.f[DIR_0PP ])[k1tn ];
+ real mfbcc = (D.f[DIR_0MM ])[k1bs ];
+ real mfbac = (D.f[DIR_0PM ])[k1bn ];
+ real mfbca = (D.f[DIR_0MP ])[k1ts ];
+ real mfbbb = (D.f[DIR_000])[k1zero];
+ real mfaaa = (D.f[DIR_PPP ])[k1tne ];
+ real mfcca = (D.f[DIR_MMP ])[k1tsw ];
+ real mfaca = (D.f[DIR_PMP ])[k1tse ];
+ real mfcaa = (D.f[DIR_MPP ])[k1tnw ];
+ real mfaac = (D.f[DIR_PPM ])[k1bne ];
+ real mfccc = (D.f[DIR_MMM ])[k1bsw ];
+ real mfacc = (D.f[DIR_PMM ])[k1bse ];
+ real mfcac = (D.f[DIR_MPM ])[k1bnw ];
+
+ //real mfcbb = (D.f[DIR_P00 ])[ke ];
+ //real mfabb = (D.f[DIR_M00 ])[kw ];
+ //real mfbcb = (D.f[DIR_0P0 ])[kn ];
+ //real mfbab = (D.f[DIR_0M0 ])[ks ];
+ //real mfbbc = (D.f[DIR_00P ])[kt ];
+ //real mfbba = (D.f[DIR_00M ])[kb ];
+ //real mfccb = (D.f[DIR_PP0 ])[kne ];
+ //real mfaab = (D.f[DIR_MM0 ])[ksw ];
+ //real mfcab = (D.f[DIR_PM0 ])[kse ];
+ //real mfacb = (D.f[DIR_MP0 ])[knw ];
+ //real mfcbc = (D.f[DIR_P0P ])[kte ];
+ //real mfaba = (D.f[DIR_M0M ])[kbw ];
+ //real mfcba = (D.f[DIR_P0M ])[kbe ];
+ //real mfabc = (D.f[DIR_M0P ])[ktw ];
+ //real mfbcc = (D.f[DIR_0PP ])[ktn ];
+ //real mfbaa = (D.f[DIR_0MM ])[kbs ];
+ //real mfbca = (D.f[DIR_0PM ])[kbn ];
+ //real mfbac = (D.f[DIR_0MP ])[kts ];
+ //real mfbbb = (D.f[DIR_000])[kzero];
+ //real mfccc = (D.f[DIR_PPP ])[ktne ];
+ //real mfaac = (D.f[DIR_MMP ])[ktsw ];
+ //real mfcac = (D.f[DIR_PMP ])[ktse ];
+ //real mfacc = (D.f[DIR_MPP ])[ktnw ];
+ //real mfcca = (D.f[DIR_PPM ])[kbne ];
+ //real mfaaa = (D.f[DIR_MMM ])[kbsw ];
+ //real mfcaa = (D.f[DIR_PMM ])[kbse ];
+ //real mfaca = (D.f[DIR_MPM ])[kbnw ];
+ ////////////////////////////////////////////////////////////////////////////////////
+ //real rho = (((((mfccc+mfaaa) + (mfaca+mfcac)) + ((mfacc+mfcaa) + (mfaac+mfcca))) +
+ // (((mfbac+mfbca) + (mfbaa+mfbcc)) + ((mfabc+mfcba) + (mfaba+mfcbc)) + ((mfacb+mfcab) + (mfaab+mfccb))) +
+ // ((mfabb+mfcbb) + (mfbab+mfbcb)) + (mfbba+mfbbc)) + mfbbb) + one;//!!!!Achtung + one
+ ////////////////////////////////////////////////////////////////////////////////////
+ real rho = rhoBC[k];
+ ////////////////////////////////////////////////////////////////////////////////////
+ real OoRho = c1o1 / (rho * 1.5f);
+ ////////////////////////////////////////////////////////////////////////////////////
+ real vvx = ((((mfccc-mfaaa) + (mfcac-mfaca)) + ((mfcaa-mfacc) + (mfcca-mfaac))) +
+ (((mfcba-mfabc) + (mfcbc-mfaba)) + ((mfcab-mfacb) + (mfccb-mfaab))) +
+ (mfcbb-mfabb)) * OoRho;
+ real vvy =((((mfccc-mfaaa) + (mfaca-mfcac)) + ((mfacc-mfcaa) + (mfcca-mfaac))) +
+ (((mfbca-mfbac) + (mfbcc-mfbaa)) + ((mfacb-mfcab) + (mfccb-mfaab))) +
+ (mfbcb-mfbab)) * OoRho;
+ real vvz =((((mfccc-mfaaa) + (mfcac-mfaca)) + ((mfacc-mfcaa) + (mfaac-mfcca))) +
+ (((mfbac-mfbca) + (mfbcc-mfbaa)) + ((mfabc-mfcba) + (mfcbc-mfaba))) +
+ (mfbbc-mfbba)) * OoRho;
+ /////////////////////////
+ //Test Values
+ //double vvx = 0.016;
+ //double vvy = zero;
+ //double vvz = zero;
+ ////////////////////////////////////////////////////////////////////////////////////////
+ ////round off error test
+ //if(vvx!=zero){
+ // (kDistTest.f[DIR_P00 ])[k] = mfabb;
+ // (kDistTest.f[DIR_M00 ])[k] = mfcbb;
+ // (kDistTest.f[DIR_0P0 ])[k] = mfbab;
+ // (kDistTest.f[DIR_0M0 ])[k] = mfbcb;
+ // (kDistTest.f[DIR_00P ])[k] = mfbba;
+ // (kDistTest.f[DIR_00M ])[k] = mfbbc;
+ // (kDistTest.f[DIR_PP0 ])[k] = mfaab;
+ // (kDistTest.f[DIR_MM0 ])[k] = mfccb;
+ // (kDistTest.f[DIR_PM0 ])[k] = mfacb;
+ // (kDistTest.f[DIR_MP0 ])[k] = mfcab;
+ // (kDistTest.f[DIR_P0P ])[k] = mfaba;
+ // (kDistTest.f[DIR_M0M ])[k] = mfcbc;
+ // (kDistTest.f[DIR_P0M ])[k] = mfabc;
+ // (kDistTest.f[DIR_M0P ])[k] = mfcba;
+ // (kDistTest.f[DIR_0PP ])[k] = mfbaa;
+ // (kDistTest.f[DIR_0MM ])[k] = mfbcc;
+ // (kDistTest.f[DIR_0PM ])[k] = mfbac;
+ // (kDistTest.f[DIR_0MP ])[k] = mfbca;
+ // (kDistTest.f[DIR_000])[k] = KQK;
+ // (kDistTest.f[DIR_PPP ])[k] = mfaaa;
+ // (kDistTest.f[DIR_MMP ])[k] = mfcca;
+ // (kDistTest.f[DIR_PMP ])[k] = mfaca;
+ // (kDistTest.f[DIR_MPP ])[k] = mfcaa;
+ // (kDistTest.f[DIR_PPM ])[k] = mfaac;
+ // (kDistTest.f[DIR_MMM ])[k] = mfccc;
+ // (kDistTest.f[DIR_PMM ])[k] = mfacc;
+ // (kDistTest.f[DIR_MPM ])[k] = mfcac;
+ //}else{
+ // (kDistTest.f[DIR_P00 ])[k] = zero;
+ // (kDistTest.f[DIR_M00 ])[k] = zero;
+ // (kDistTest.f[DIR_0P0 ])[k] = zero;
+ // (kDistTest.f[DIR_0M0 ])[k] = zero;
+ // (kDistTest.f[DIR_00P ])[k] = zero;
+ // (kDistTest.f[DIR_00M ])[k] = zero;
+ // (kDistTest.f[DIR_PP0 ])[k] = zero;
+ // (kDistTest.f[DIR_MM0 ])[k] = zero;
+ // (kDistTest.f[DIR_PM0 ])[k] = zero;
+ // (kDistTest.f[DIR_MP0 ])[k] = zero;
+ // (kDistTest.f[DIR_P0P ])[k] = zero;
+ // (kDistTest.f[DIR_M0M ])[k] = zero;
+ // (kDistTest.f[DIR_P0M ])[k] = zero;
+ // (kDistTest.f[DIR_M0P ])[k] = zero;
+ // (kDistTest.f[DIR_0PP ])[k] = zero;
+ // (kDistTest.f[DIR_0MM ])[k] = zero;
+ // (kDistTest.f[DIR_0PM ])[k] = zero;
+ // (kDistTest.f[DIR_0MP ])[k] = zero;
+ // (kDistTest.f[DIR_000])[k] = zero;
+ // (kDistTest.f[DIR_PPP ])[k] = zero;
+ // (kDistTest.f[DIR_MMP ])[k] = zero;
+ // (kDistTest.f[DIR_PMP ])[k] = zero;
+ // (kDistTest.f[DIR_MPP ])[k] = zero;
+ // (kDistTest.f[DIR_PPM ])[k] = zero;
+ // (kDistTest.f[DIR_MMM ])[k] = zero;
+ // (kDistTest.f[DIR_PMM ])[k] = zero;
+ // (kDistTest.f[DIR_MPM ])[k] = zero;
+ //}
+
+ //////////////////////////////////////////////////////////////////////////////////////
+ //// first bad fix for negative x velocity
+ ////if(vvx > zero) vvx = zero;
+ //////////////////////////////////////////////////////////////////////////////////////
+ ////// second bad fix for negative x velocity
+ ////if(vvx > zero){
+ //// vvx = -vvx;
+ //// vvy = -vvy;
+ //// vvz = -vvz;
+ ////}
+ ////////////////////////////////////////////////////////////////////////////////////
+ double vx2 = vvx * vvx;
+ double vy2 = vvy * vvy;
+ double vz2 = vvz * vvz;
+ //////////////////////////////////////////////////////////////////////////////////
+ //original
real XXb = -c2o3 + vx2;
real XXc = -c1o2 * (XXb + c1o1 + vvx);
real XXa = XXc + vvx;
@@ -4045,66 +4045,66 @@ __global__ void QPressDeviceEQZ27(real* rhoBC,
real ZZb = -c2o3 + vz2;
real ZZc = -c1o2 * (ZZb + c1o1 + vvz);
real ZZa = ZZc + vvz;
- //////////////////////////////////////////////////////////////////////////////////
- //unkonditioniert
- mfcbb = -(rhoBC[k] + c1o1) * XXc * YYb * ZZb - c2o27;
- mfabb = -(rhoBC[k] + c1o1) * XXa * YYb * ZZb - c2o27;
- mfbcb = -(rhoBC[k] + c1o1) * XXb * YYc * ZZb - c2o27;
- mfbab = -(rhoBC[k] + c1o1) * XXb * YYa * ZZb - c2o27;
- mfbbc = -(rhoBC[k] + c1o1) * XXb * YYb * ZZc - c2o27;
- mfbba = -(rhoBC[k] + c1o1) * XXb * YYb * ZZa - c2o27;
- mfccb = -(rhoBC[k] + c1o1) * XXc * YYc * ZZb - c1o54;
- mfaab = -(rhoBC[k] + c1o1) * XXa * YYa * ZZb - c1o54;
- mfcab = -(rhoBC[k] + c1o1) * XXc * YYa * ZZb - c1o54;
- mfacb = -(rhoBC[k] + c1o1) * XXa * YYc * ZZb - c1o54;
- mfcbc = -(rhoBC[k] + c1o1) * XXc * YYb * ZZc - c1o54;
- mfaba = -(rhoBC[k] + c1o1) * XXa * YYb * ZZa - c1o54;
- mfcba = -(rhoBC[k] + c1o1) * XXc * YYb * ZZa - c1o54;
- mfabc = -(rhoBC[k] + c1o1) * XXa * YYb * ZZc - c1o54;
- mfbcc = -(rhoBC[k] + c1o1) * XXb * YYc * ZZc - c1o54;
- mfbaa = -(rhoBC[k] + c1o1) * XXb * YYa * ZZa - c1o54;
- mfbca = -(rhoBC[k] + c1o1) * XXb * YYc * ZZa - c1o54;
- mfbac = -(rhoBC[k] + c1o1) * XXb * YYa * ZZc - c1o54;
- mfbbb = -(rhoBC[k] + c1o1) * XXb * YYb * ZZb - c8o27;
- mfccc = -(rhoBC[k] + c1o1) * XXc * YYc * ZZc - c1o216;
- mfaac = -(rhoBC[k] + c1o1) * XXa * YYa * ZZc - c1o216;
- mfcac = -(rhoBC[k] + c1o1) * XXc * YYa * ZZc - c1o216;
- mfacc = -(rhoBC[k] + c1o1) * XXa * YYc * ZZc - c1o216;
- mfcca = -(rhoBC[k] + c1o1) * XXc * YYc * ZZa - c1o216;
- mfaaa = -(rhoBC[k] + c1o1) * XXa * YYa * ZZa - c1o216;
- mfcaa = -(rhoBC[k] + c1o1) * XXc * YYa * ZZa - c1o216;
- mfaca = -(rhoBC[k] + c1o1) * XXa * YYc * ZZa - c1o216;
- //////////////////////////////////////////////////////////
- ////konditioniert
- //double OneOver216RhoPlusOne = c1over216*(rhoBC[k]+one);
- //double OnoOver216Rho = c1over216*rhoBC[k];
- //mfcbb = OnoOver216Rho*sixteen + OneOver216RhoPlusOne*twelve*(-(two*vy2) - two*vz2 + three*vy2*vz2 + vvx*(-two + three*vy2)*(-two + three*vz2) + vx2*(-two + three*vy2)*(-two + three*vz2));
- //mfabb = OnoOver216Rho*sixteen - OneOver216RhoPlusOne*twelve*(two*vy2 + two*vz2 - three*vy2*vz2 + vvx*(-two + three*vy2)*(-two + three*vz2) + vx2*(-four + six*vy2 + six*vz2 - nine*vy2*vz2));
- //mfbcb = four*(-(four*OneOver216RhoPlusOne) + four*OnoOver216Rho + OneOver216RhoPlusOne*(-two + three*vx2)*(one + three*vvy + three*vy2)*(-two + three*vz2));
- //mfbab = four*(four*OnoOver216Rho - OneOver216RhoPlusOne*three*(vvy*(-two + three*vx2)*(-two + three*vz2) - one*vx2*(one + three*vy2)*(-two + three*vz2) + two*(-(two*vy2) + vz2 + three*vy2*vz2)));
- //mfbbc = four*(-(four*OneOver216RhoPlusOne) + four*OnoOver216Rho + OneOver216RhoPlusOne*(-two + three*vx2)*(-two + three*vy2)*(one + three*vvz + three*vz2));
- //mfbba = four*(four*OnoOver216Rho - OneOver216RhoPlusOne*three*(vvz*(-two + three*vx2)*(-two + three*vy2) - one*vx2*(-two + three*vy2)*(one + three*vz2) + two*(vy2 - two*vz2 + three*vy2*vz2)));
- //mfccb = -(two*(-(OnoOver216Rho*two) + OneOver216RhoPlusOne*three*(-(two*vx2) - two*vy2 - six*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 + vvy*(one + three*vx2)*(-two + three*vz2) + vvx*(one + three*vvy + three*vy2)*(-two + three*vz2))));
- //mfaab = -(two*(-(OnoOver216Rho*two) + OneOver216RhoPlusOne*three*(-(two*vx2) - two*vy2 - six*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 - one*vvy*(one + three*vx2)*(-two + three*vz2) + vvx*(-one + three*vvy - three*vy2)*(-two + three*vz2))));
- //mfcab = two*(OnoOver216Rho*two + OneOver216RhoPlusOne*three*(two*vx2 + two*vy2 + six*vx2*vy2 - one*vz2 - three*vx2*vz2 - three*vy2*vz2 - nine*vx2*vy2*vz2 + vvy*(one + three*vx2)*(-two + three*vz2) + vvx*(-one + three*vvy - three*vy2)*(-two + three*vz2)));
- //mfacb = two*(OnoOver216Rho*two + OneOver216RhoPlusOne*three*(two*vx2 + two*vy2 + six*vx2*vy2 - one*vz2 - three*vx2*vz2 - three*vy2*vz2 - nine*vx2*vy2*vz2 - one*vvy*(one + three*vx2)*(-two + three*vz2) + vvx*(one + three*vvy + three*vy2)*(-two + three*vz2)));
- //mfcbc = -(two*(-(OnoOver216Rho*two) + OneOver216RhoPlusOne*three*(-(two*vx2) + vy2 + three*vx2*vy2 + vvz*(one + three*vx2)*(-two + three*vy2) - two*vz2 - six*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 + vvx*(-two + three*vy2)*(one + three*vvz + three*vz2))));
- //mfaba = -(two*(-(OnoOver216Rho*two) + OneOver216RhoPlusOne*three*(-(two*vx2) + vy2 + three*vx2*vy2 - one*vvz*(one + three*vx2)*(-two + three*vy2) - two*vz2 - six*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 + vvx*(-two + three*vy2)*(-one + three*vvz - three*vz2))));
- //mfcba = two*(OnoOver216Rho*two + OneOver216RhoPlusOne*three*(two*vx2 - one*vy2 - three*vx2*vy2 + vvz*(one + three*vx2)*(-two + three*vy2) + two*vz2 + six*vx2*vz2 - three*vy2*vz2 - nine*vx2*vy2*vz2 + vvx*(-two + three*vy2)*(-one + three*vvz - three*vz2)));
- //mfabc = two*(OnoOver216Rho*two + OneOver216RhoPlusOne*three*(two*vx2 - one*vy2 - three*vx2*vy2 - one*vvz*(one + three*vx2)*(-two + three*vy2) + two*vz2 + six*vx2*vz2 - three*vy2*vz2 - nine*vx2*vy2*vz2 + vvx*(-two + three*vy2)*(one + three*vvz + three*vz2)));
- //mfbcc = -(two*(-(OnoOver216Rho*two) + OneOver216RhoPlusOne*three*(vx2 - two*vy2 + three*vx2*vy2 + vvz*(-two + three*vx2)*(one + three*vy2) - two*vz2 + three*vx2*vz2 - six*vy2*vz2 + nine*vx2*vy2*vz2 + vvy*(-two + three*vx2)*(one + three*vvz + three*vz2))));
- //mfbaa = -(two*(-(OnoOver216Rho*two) + OneOver216RhoPlusOne*three*(vx2 - two*vy2 + three*vx2*vy2 - one*vvz*(-two + three*vx2)*(one + three*vy2) - two*vz2 + three*vx2*vz2 - six*vy2*vz2 + nine*vx2*vy2*vz2 + vvy*(-two + three*vx2)*(-one + three*vvz - three*vz2))));
- //mfbca = two*(OnoOver216Rho*two + OneOver216RhoPlusOne*three*(-(one*vx2) + two*vy2 - three*vx2*vy2 + vvz*(-two + three*vx2)*(one + three*vy2) + two*vz2 - three*vx2*vz2 + six*vy2*vz2 - nine*vx2*vy2*vz2 + vvy*(-two + three*vx2)*(-one + three*vvz - three*vz2)));
- //mfbac = two*(OnoOver216Rho*two + OneOver216RhoPlusOne*three*(-(one*vx2) + two*vy2 - three*vx2*vy2 - one*vvz*(-two + three*vx2)*(one + three*vy2) + two*vz2 - three*vx2*vz2 + six*vy2*vz2 - nine*vx2*vy2*vz2 + vvy*(-two + three*vx2)*(one + three*vvz + three*vz2)));
- //mfbbb = eight*(eight*OnoOver216Rho + OneOver216RhoPlusOne*three*(four*vy2 + four*vz2 - six*vy2*vz2 + vx2*(-two + three*vy2)*(-two + three*vz2)));
- //mfccc = OnoOver216Rho + OneOver216RhoPlusOne*three*(vvz + vx2 + three*vvz*vx2 + vy2 + three*vvz*vy2 + three*vx2*vy2 + nine*vvz*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 + vvy*(one + three*vx2)*(one + three*vvz + three*vz2) + vvx*(one + three*vvy + three*vy2)*(one + three*vvz + three*vz2));
- //mfaac = OnoOver216Rho + OneOver216RhoPlusOne*three*(vvz + vx2 + three*vvz*vx2 + vy2 + three*vvz*vy2 + three*vx2*vy2 + nine*vvz*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 - one*vvy*(one + three*vx2)*(one + three*vvz + three*vz2) + vvx*(-one + three*vvy - three*vy2)*(one + three*vvz + three*vz2));
- //mfcac = OnoOver216Rho + OneOver216RhoPlusOne*three*(vvz + vx2 + three*vvz*vx2 + vy2 + three*vvz*vy2 + three*vx2*vy2 + nine*vvz*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 - one*vvy*(one + three*vx2)*(one + three*vvz + three*vz2) - one*vvx*(-one + three*vvy - three*vy2)*(one + three*vvz + three*vz2));
- //mfacc = OnoOver216Rho + OneOver216RhoPlusOne*three*(vvz + vx2 + three*vvz*vx2 + vy2 + three*vvz*vy2 + three*vx2*vy2 + nine*vvz*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 + vvy*(one + three*vx2)*(one + three*vvz + three*vz2) - one*vvx*(one + three*vvy + three*vy2)*(one + three*vvz + three*vz2));
- //mfcca = OnoOver216Rho + OneOver216RhoPlusOne*three*(-(one*vvz) + vx2 - three*vvz*vx2 + vy2 - three*vvz*vy2 + three*vx2*vy2 - nine*vvz*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 - one*vvy*(one + three*vx2)*(-one + three*vvz - three*vz2) - one*vvx*(one + three*vvy + three*vy2)*(-one + three*vvz - three*vz2));
- //mfaaa = OnoOver216Rho - OneOver216RhoPlusOne*three*(vvz - one*vx2 + three*vvz*vx2 - one*vy2 + three*vvz*vy2 - three*vx2*vy2 + nine*vvz*vx2*vy2 - one*vz2 - three*vx2*vz2 - three*vy2*vz2 - nine*vx2*vy2*vz2 - one*vvy*(one + three*vx2)*(-one + three*vvz - three*vz2) + vvx*(-one + three*vvy - three*vy2)*(-one + three*vvz - three*vz2));
- //mfcaa = OnoOver216Rho + OneOver216RhoPlusOne*three*(-(one*vvz) + vx2 - three*vvz*vx2 + vy2 - three*vvz*vy2 + three*vx2*vy2 - nine*vvz*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 + vvy*(one + three*vx2)*(-one + three*vvz - three*vz2) + vvx*(-one + three*vvy - three*vy2)*(-one + three*vvz - three*vz2));
- //mfaca = OnoOver216Rho + OneOver216RhoPlusOne*three*(-(one*vvz) + vx2 - three*vvz*vx2 + vy2 - three*vvz*vy2 + three*vx2*vy2 - nine*vvz*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 - one*vvy*(one + three*vx2)*(-one + three*vvz - three*vz2) + vvx*(one + three*vvy + three*vy2)*(-one + three*vvz - three*vz2));
+ //////////////////////////////////////////////////////////////////////////////////
+ //unkonditioniert
+ mfcbb = -(rhoBC[k] + c1o1) * XXc * YYb * ZZb - c2o27;
+ mfabb = -(rhoBC[k] + c1o1) * XXa * YYb * ZZb - c2o27;
+ mfbcb = -(rhoBC[k] + c1o1) * XXb * YYc * ZZb - c2o27;
+ mfbab = -(rhoBC[k] + c1o1) * XXb * YYa * ZZb - c2o27;
+ mfbbc = -(rhoBC[k] + c1o1) * XXb * YYb * ZZc - c2o27;
+ mfbba = -(rhoBC[k] + c1o1) * XXb * YYb * ZZa - c2o27;
+ mfccb = -(rhoBC[k] + c1o1) * XXc * YYc * ZZb - c1o54;
+ mfaab = -(rhoBC[k] + c1o1) * XXa * YYa * ZZb - c1o54;
+ mfcab = -(rhoBC[k] + c1o1) * XXc * YYa * ZZb - c1o54;
+ mfacb = -(rhoBC[k] + c1o1) * XXa * YYc * ZZb - c1o54;
+ mfcbc = -(rhoBC[k] + c1o1) * XXc * YYb * ZZc - c1o54;
+ mfaba = -(rhoBC[k] + c1o1) * XXa * YYb * ZZa - c1o54;
+ mfcba = -(rhoBC[k] + c1o1) * XXc * YYb * ZZa - c1o54;
+ mfabc = -(rhoBC[k] + c1o1) * XXa * YYb * ZZc - c1o54;
+ mfbcc = -(rhoBC[k] + c1o1) * XXb * YYc * ZZc - c1o54;
+ mfbaa = -(rhoBC[k] + c1o1) * XXb * YYa * ZZa - c1o54;
+ mfbca = -(rhoBC[k] + c1o1) * XXb * YYc * ZZa - c1o54;
+ mfbac = -(rhoBC[k] + c1o1) * XXb * YYa * ZZc - c1o54;
+ mfbbb = -(rhoBC[k] + c1o1) * XXb * YYb * ZZb - c8o27;
+ mfccc = -(rhoBC[k] + c1o1) * XXc * YYc * ZZc - c1o216;
+ mfaac = -(rhoBC[k] + c1o1) * XXa * YYa * ZZc - c1o216;
+ mfcac = -(rhoBC[k] + c1o1) * XXc * YYa * ZZc - c1o216;
+ mfacc = -(rhoBC[k] + c1o1) * XXa * YYc * ZZc - c1o216;
+ mfcca = -(rhoBC[k] + c1o1) * XXc * YYc * ZZa - c1o216;
+ mfaaa = -(rhoBC[k] + c1o1) * XXa * YYa * ZZa - c1o216;
+ mfcaa = -(rhoBC[k] + c1o1) * XXc * YYa * ZZa - c1o216;
+ mfaca = -(rhoBC[k] + c1o1) * XXa * YYc * ZZa - c1o216;
+ //////////////////////////////////////////////////////////
+ ////konditioniert
+ //double OneOver216RhoPlusOne = c1over216*(rhoBC[k]+one);
+ //double OnoOver216Rho = c1over216*rhoBC[k];
+ //mfcbb = OnoOver216Rho*sixteen + OneOver216RhoPlusOne*twelve*(-(two*vy2) - two*vz2 + three*vy2*vz2 + vvx*(-two + three*vy2)*(-two + three*vz2) + vx2*(-two + three*vy2)*(-two + three*vz2));
+ //mfabb = OnoOver216Rho*sixteen - OneOver216RhoPlusOne*twelve*(two*vy2 + two*vz2 - three*vy2*vz2 + vvx*(-two + three*vy2)*(-two + three*vz2) + vx2*(-four + six*vy2 + six*vz2 - nine*vy2*vz2));
+ //mfbcb = four*(-(four*OneOver216RhoPlusOne) + four*OnoOver216Rho + OneOver216RhoPlusOne*(-two + three*vx2)*(one + three*vvy + three*vy2)*(-two + three*vz2));
+ //mfbab = four*(four*OnoOver216Rho - OneOver216RhoPlusOne*three*(vvy*(-two + three*vx2)*(-two + three*vz2) - one*vx2*(one + three*vy2)*(-two + three*vz2) + two*(-(two*vy2) + vz2 + three*vy2*vz2)));
+ //mfbbc = four*(-(four*OneOver216RhoPlusOne) + four*OnoOver216Rho + OneOver216RhoPlusOne*(-two + three*vx2)*(-two + three*vy2)*(one + three*vvz + three*vz2));
+ //mfbba = four*(four*OnoOver216Rho - OneOver216RhoPlusOne*three*(vvz*(-two + three*vx2)*(-two + three*vy2) - one*vx2*(-two + three*vy2)*(one + three*vz2) + two*(vy2 - two*vz2 + three*vy2*vz2)));
+ //mfccb = -(two*(-(OnoOver216Rho*two) + OneOver216RhoPlusOne*three*(-(two*vx2) - two*vy2 - six*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 + vvy*(one + three*vx2)*(-two + three*vz2) + vvx*(one + three*vvy + three*vy2)*(-two + three*vz2))));
+ //mfaab = -(two*(-(OnoOver216Rho*two) + OneOver216RhoPlusOne*three*(-(two*vx2) - two*vy2 - six*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 - one*vvy*(one + three*vx2)*(-two + three*vz2) + vvx*(-one + three*vvy - three*vy2)*(-two + three*vz2))));
+ //mfcab = two*(OnoOver216Rho*two + OneOver216RhoPlusOne*three*(two*vx2 + two*vy2 + six*vx2*vy2 - one*vz2 - three*vx2*vz2 - three*vy2*vz2 - nine*vx2*vy2*vz2 + vvy*(one + three*vx2)*(-two + three*vz2) + vvx*(-one + three*vvy - three*vy2)*(-two + three*vz2)));
+ //mfacb = two*(OnoOver216Rho*two + OneOver216RhoPlusOne*three*(two*vx2 + two*vy2 + six*vx2*vy2 - one*vz2 - three*vx2*vz2 - three*vy2*vz2 - nine*vx2*vy2*vz2 - one*vvy*(one + three*vx2)*(-two + three*vz2) + vvx*(one + three*vvy + three*vy2)*(-two + three*vz2)));
+ //mfcbc = -(two*(-(OnoOver216Rho*two) + OneOver216RhoPlusOne*three*(-(two*vx2) + vy2 + three*vx2*vy2 + vvz*(one + three*vx2)*(-two + three*vy2) - two*vz2 - six*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 + vvx*(-two + three*vy2)*(one + three*vvz + three*vz2))));
+ //mfaba = -(two*(-(OnoOver216Rho*two) + OneOver216RhoPlusOne*three*(-(two*vx2) + vy2 + three*vx2*vy2 - one*vvz*(one + three*vx2)*(-two + three*vy2) - two*vz2 - six*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 + vvx*(-two + three*vy2)*(-one + three*vvz - three*vz2))));
+ //mfcba = two*(OnoOver216Rho*two + OneOver216RhoPlusOne*three*(two*vx2 - one*vy2 - three*vx2*vy2 + vvz*(one + three*vx2)*(-two + three*vy2) + two*vz2 + six*vx2*vz2 - three*vy2*vz2 - nine*vx2*vy2*vz2 + vvx*(-two + three*vy2)*(-one + three*vvz - three*vz2)));
+ //mfabc = two*(OnoOver216Rho*two + OneOver216RhoPlusOne*three*(two*vx2 - one*vy2 - three*vx2*vy2 - one*vvz*(one + three*vx2)*(-two + three*vy2) + two*vz2 + six*vx2*vz2 - three*vy2*vz2 - nine*vx2*vy2*vz2 + vvx*(-two + three*vy2)*(one + three*vvz + three*vz2)));
+ //mfbcc = -(two*(-(OnoOver216Rho*two) + OneOver216RhoPlusOne*three*(vx2 - two*vy2 + three*vx2*vy2 + vvz*(-two + three*vx2)*(one + three*vy2) - two*vz2 + three*vx2*vz2 - six*vy2*vz2 + nine*vx2*vy2*vz2 + vvy*(-two + three*vx2)*(one + three*vvz + three*vz2))));
+ //mfbaa = -(two*(-(OnoOver216Rho*two) + OneOver216RhoPlusOne*three*(vx2 - two*vy2 + three*vx2*vy2 - one*vvz*(-two + three*vx2)*(one + three*vy2) - two*vz2 + three*vx2*vz2 - six*vy2*vz2 + nine*vx2*vy2*vz2 + vvy*(-two + three*vx2)*(-one + three*vvz - three*vz2))));
+ //mfbca = two*(OnoOver216Rho*two + OneOver216RhoPlusOne*three*(-(one*vx2) + two*vy2 - three*vx2*vy2 + vvz*(-two + three*vx2)*(one + three*vy2) + two*vz2 - three*vx2*vz2 + six*vy2*vz2 - nine*vx2*vy2*vz2 + vvy*(-two + three*vx2)*(-one + three*vvz - three*vz2)));
+ //mfbac = two*(OnoOver216Rho*two + OneOver216RhoPlusOne*three*(-(one*vx2) + two*vy2 - three*vx2*vy2 - one*vvz*(-two + three*vx2)*(one + three*vy2) + two*vz2 - three*vx2*vz2 + six*vy2*vz2 - nine*vx2*vy2*vz2 + vvy*(-two + three*vx2)*(one + three*vvz + three*vz2)));
+ //mfbbb = eight*(eight*OnoOver216Rho + OneOver216RhoPlusOne*three*(four*vy2 + four*vz2 - six*vy2*vz2 + vx2*(-two + three*vy2)*(-two + three*vz2)));
+ //mfccc = OnoOver216Rho + OneOver216RhoPlusOne*three*(vvz + vx2 + three*vvz*vx2 + vy2 + three*vvz*vy2 + three*vx2*vy2 + nine*vvz*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 + vvy*(one + three*vx2)*(one + three*vvz + three*vz2) + vvx*(one + three*vvy + three*vy2)*(one + three*vvz + three*vz2));
+ //mfaac = OnoOver216Rho + OneOver216RhoPlusOne*three*(vvz + vx2 + three*vvz*vx2 + vy2 + three*vvz*vy2 + three*vx2*vy2 + nine*vvz*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 - one*vvy*(one + three*vx2)*(one + three*vvz + three*vz2) + vvx*(-one + three*vvy - three*vy2)*(one + three*vvz + three*vz2));
+ //mfcac = OnoOver216Rho + OneOver216RhoPlusOne*three*(vvz + vx2 + three*vvz*vx2 + vy2 + three*vvz*vy2 + three*vx2*vy2 + nine*vvz*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 - one*vvy*(one + three*vx2)*(one + three*vvz + three*vz2) - one*vvx*(-one + three*vvy - three*vy2)*(one + three*vvz + three*vz2));
+ //mfacc = OnoOver216Rho + OneOver216RhoPlusOne*three*(vvz + vx2 + three*vvz*vx2 + vy2 + three*vvz*vy2 + three*vx2*vy2 + nine*vvz*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 + vvy*(one + three*vx2)*(one + three*vvz + three*vz2) - one*vvx*(one + three*vvy + three*vy2)*(one + three*vvz + three*vz2));
+ //mfcca = OnoOver216Rho + OneOver216RhoPlusOne*three*(-(one*vvz) + vx2 - three*vvz*vx2 + vy2 - three*vvz*vy2 + three*vx2*vy2 - nine*vvz*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 - one*vvy*(one + three*vx2)*(-one + three*vvz - three*vz2) - one*vvx*(one + three*vvy + three*vy2)*(-one + three*vvz - three*vz2));
+ //mfaaa = OnoOver216Rho - OneOver216RhoPlusOne*three*(vvz - one*vx2 + three*vvz*vx2 - one*vy2 + three*vvz*vy2 - three*vx2*vy2 + nine*vvz*vx2*vy2 - one*vz2 - three*vx2*vz2 - three*vy2*vz2 - nine*vx2*vy2*vz2 - one*vvy*(one + three*vx2)*(-one + three*vvz - three*vz2) + vvx*(-one + three*vvy - three*vy2)*(-one + three*vvz - three*vz2));
+ //mfcaa = OnoOver216Rho + OneOver216RhoPlusOne*three*(-(one*vvz) + vx2 - three*vvz*vx2 + vy2 - three*vvz*vy2 + three*vx2*vy2 - nine*vvz*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 + vvy*(one + three*vx2)*(-one + three*vvz - three*vz2) + vvx*(-one + three*vvy - three*vy2)*(-one + three*vvz - three*vz2));
+ //mfaca = OnoOver216Rho + OneOver216RhoPlusOne*three*(-(one*vvz) + vx2 - three*vvz*vx2 + vy2 - three*vvz*vy2 + three*vx2*vy2 - nine*vvz*vx2*vy2 + vz2 + three*vx2*vz2 + three*vy2*vz2 + nine*vx2*vy2*vz2 - one*vvy*(one + three*vx2)*(-one + three*vvz - three*vz2) + vvx*(one + three*vvy + three*vy2)*(-one + three*vvz - three*vz2));
//////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
//if (isEvenTimestep==true)
@@ -4136,7 +4136,7 @@ __global__ void QPressDeviceEQZ27(real* rhoBC,
// D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
// D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
// D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- //}
+ //}
//else
//{
// D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -4170,88 +4170,88 @@ __global__ void QPressDeviceEQZ27(real* rhoBC,
//////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
//__syncthreads();
- (D.f[DIR_P00 ])[ke ] = mfabb;//mfcbb;
- (D.f[DIR_M00 ])[kw ] = mfcbb;//mfabb;
- (D.f[DIR_0P0 ])[kn ] = mfbab;//mfbcb;
- (D.f[DIR_0M0 ])[ks ] = mfbcb;//mfbab;
- (D.f[DIR_00P ])[kt ] = mfbba;//mfbbc;
- (D.f[DIR_00M ])[kb ] = mfbbc;//mfbba;
- (D.f[DIR_PP0 ])[kne ] = mfaab;//mfccb;
- (D.f[DIR_MM0 ])[ksw ] = mfccb;//mfaab;
- (D.f[DIR_PM0 ])[kse ] = mfacb;//mfcab;
- (D.f[DIR_MP0 ])[knw ] = mfcab;//mfacb;
- (D.f[DIR_P0P ])[kte ] = mfaba;//mfcbc;
- (D.f[DIR_M0M ])[kbw ] = mfcbc;//mfaba;
- (D.f[DIR_P0M ])[kbe ] = mfabc;//mfcba;
- (D.f[DIR_M0P ])[ktw ] = mfcba;//mfabc;
- (D.f[DIR_0PP ])[ktn ] = mfbaa;//mfbcc;
- (D.f[DIR_0MM ])[kbs ] = mfbcc;//mfbaa;
- (D.f[DIR_0PM ])[kbn ] = mfbac;//mfbca;
- (D.f[DIR_0MP ])[kts ] = mfbca;//mfbac;
- (D.f[DIR_000])[kzero] = mfbbb;//mfbbb;
- (D.f[DIR_PPP ])[ktne ] = mfaaa;//mfccc;
- (D.f[DIR_MMP ])[ktsw ] = mfcca;//mfaac;
- (D.f[DIR_PMP ])[ktse ] = mfaca;//mfcac;
- (D.f[DIR_MPP ])[ktnw ] = mfcaa;//mfacc;
- (D.f[DIR_PPM ])[kbne ] = mfaac;//mfcca;
- (D.f[DIR_MMM ])[kbsw ] = mfccc;//mfaaa;
- (D.f[DIR_PMM ])[kbse ] = mfacc;//mfcaa;
- (D.f[DIR_MPM ])[kbnw ] = mfcac;//mfaca;
- //(D.f[DIR_P00 ])[ke ] = mfcbb;
- //(D.f[DIR_M00 ])[kw ] = mfabb;
- //(D.f[DIR_0P0 ])[kn ] = mfbcb;
- //(D.f[DIR_0M0 ])[ks ] = mfbab;
- //(D.f[DIR_00P ])[kt ] = mfbbc;
- //(D.f[DIR_00M ])[kb ] = mfbba;
- //(D.f[DIR_PP0 ])[kne ] = mfccb;
- //(D.f[DIR_MM0 ])[ksw ] = mfaab;
- //(D.f[DIR_PM0 ])[kse ] = mfcab;
- //(D.f[DIR_MP0 ])[knw ] = mfacb;
- //(D.f[DIR_P0P ])[kte ] = mfcbc;
- //(D.f[DIR_M0M ])[kbw ] = mfaba;
- //(D.f[DIR_P0M ])[kbe ] = mfcba;
- //(D.f[DIR_M0P ])[ktw ] = mfabc;
- //(D.f[DIR_0PP ])[ktn ] = mfbcc;
- //(D.f[DIR_0MM ])[kbs ] = mfbaa;
- //(D.f[DIR_0PM ])[kbn ] = mfbca;
- //(D.f[DIR_0MP ])[kts ] = mfbac;
- //(D.f[DIR_000])[kzero] = mfbbb;
- //(D.f[DIR_PPP ])[ktne ] = mfccc;
- //(D.f[DIR_MMP ])[ktsw ] = mfaac;
- //(D.f[DIR_PMP ])[ktse ] = mfcac;
- //(D.f[DIR_MPP ])[ktnw ] = mfacc;
- //(D.f[DIR_PPM ])[kbne ] = mfcca;
- //(D.f[DIR_MMM ])[kbsw ] = mfaaa;
- //(D.f[DIR_PMM ])[kbse ] = mfcaa;
- //(D.f[DIR_MPM ])[kbnw ] = mfaca;
-
- //(D.f[DIR_P00 ])[ke ] = fE ; //f1_E ; //fW; //fE ;
- //(D.f[DIR_M00 ])[kw ] = fW ; //f1_W ; //fE; //fW ;
- //(D.f[DIR_0P0 ])[kn ] = fN ; //f1_N ; //fS; //fN ;
- //(D.f[DIR_0M0 ])[ks ] = fS ; //f1_S ; //fN; //fS ;
- //(D.f[DIR_00P ])[kt ] = fT ; //f1_T ; //fB; //fT ;
- //(D.f[DIR_00M ])[kb ] = fB ; //f1_B ; //fT; //fB ;
- //(D.f[DIR_PP0 ])[kne ] = fNE; //f1_NE; //fSW; //fNE;
- //(D.f[DIR_MM0 ])[ksw ] = fSW; //f1_SW; //fNE; //fSW;
- //(D.f[DIR_PM0 ])[kse ] = fSE; //f1_SE; //fNW; //fSE;
- //(D.f[DIR_MP0 ])[knw ] = fNW; //f1_NW; //fSE; //fNW;
- //(D.f[DIR_P0P ])[kte ] = fTE; //f1_TE; //fBW; //fTE;
- //(D.f[DIR_M0M ])[kbw ] = fBW; //f1_BW; //fTE; //fBW;
- //(D.f[DIR_P0M ])[kbe ] = fBE; //f1_BE; //fTW; //fBE;
- //(D.f[DIR_M0P ])[ktw ] = fTW; //f1_TW; //fBE; //fTW;
- //(D.f[DIR_0PP ])[ktn ] = fTN; //f1_TN; //fBS; //fTN;
- //(D.f[DIR_0MM ])[kbs ] = fBS; //f1_BS; //fTN; //fBS;
- //(D.f[DIR_0PM ])[kbn ] = fBN; //f1_BN; //fTS; //fBN;
- //(D.f[DIR_0MP ])[kts ] = fTS; //f1_TS; //fBN; //fTS;
+ (D.f[DIR_P00 ])[ke ] = mfabb;//mfcbb;
+ (D.f[DIR_M00 ])[kw ] = mfcbb;//mfabb;
+ (D.f[DIR_0P0 ])[kn ] = mfbab;//mfbcb;
+ (D.f[DIR_0M0 ])[ks ] = mfbcb;//mfbab;
+ (D.f[DIR_00P ])[kt ] = mfbba;//mfbbc;
+ (D.f[DIR_00M ])[kb ] = mfbbc;//mfbba;
+ (D.f[DIR_PP0 ])[kne ] = mfaab;//mfccb;
+ (D.f[DIR_MM0 ])[ksw ] = mfccb;//mfaab;
+ (D.f[DIR_PM0 ])[kse ] = mfacb;//mfcab;
+ (D.f[DIR_MP0 ])[knw ] = mfcab;//mfacb;
+ (D.f[DIR_P0P ])[kte ] = mfaba;//mfcbc;
+ (D.f[DIR_M0M ])[kbw ] = mfcbc;//mfaba;
+ (D.f[DIR_P0M ])[kbe ] = mfabc;//mfcba;
+ (D.f[DIR_M0P ])[ktw ] = mfcba;//mfabc;
+ (D.f[DIR_0PP ])[ktn ] = mfbaa;//mfbcc;
+ (D.f[DIR_0MM ])[kbs ] = mfbcc;//mfbaa;
+ (D.f[DIR_0PM ])[kbn ] = mfbac;//mfbca;
+ (D.f[DIR_0MP ])[kts ] = mfbca;//mfbac;
+ (D.f[DIR_000])[kzero] = mfbbb;//mfbbb;
+ (D.f[DIR_PPP ])[ktne ] = mfaaa;//mfccc;
+ (D.f[DIR_MMP ])[ktsw ] = mfcca;//mfaac;
+ (D.f[DIR_PMP ])[ktse ] = mfaca;//mfcac;
+ (D.f[DIR_MPP ])[ktnw ] = mfcaa;//mfacc;
+ (D.f[DIR_PPM ])[kbne ] = mfaac;//mfcca;
+ (D.f[DIR_MMM ])[kbsw ] = mfccc;//mfaaa;
+ (D.f[DIR_PMM ])[kbse ] = mfacc;//mfcaa;
+ (D.f[DIR_MPM ])[kbnw ] = mfcac;//mfaca;
+ //(D.f[DIR_P00 ])[ke ] = mfcbb;
+ //(D.f[DIR_M00 ])[kw ] = mfabb;
+ //(D.f[DIR_0P0 ])[kn ] = mfbcb;
+ //(D.f[DIR_0M0 ])[ks ] = mfbab;
+ //(D.f[DIR_00P ])[kt ] = mfbbc;
+ //(D.f[DIR_00M ])[kb ] = mfbba;
+ //(D.f[DIR_PP0 ])[kne ] = mfccb;
+ //(D.f[DIR_MM0 ])[ksw ] = mfaab;
+ //(D.f[DIR_PM0 ])[kse ] = mfcab;
+ //(D.f[DIR_MP0 ])[knw ] = mfacb;
+ //(D.f[DIR_P0P ])[kte ] = mfcbc;
+ //(D.f[DIR_M0M ])[kbw ] = mfaba;
+ //(D.f[DIR_P0M ])[kbe ] = mfcba;
+ //(D.f[DIR_M0P ])[ktw ] = mfabc;
+ //(D.f[DIR_0PP ])[ktn ] = mfbcc;
+ //(D.f[DIR_0MM ])[kbs ] = mfbaa;
+ //(D.f[DIR_0PM ])[kbn ] = mfbca;
+ //(D.f[DIR_0MP ])[kts ] = mfbac;
+ //(D.f[DIR_000])[kzero] = mfbbb;
+ //(D.f[DIR_PPP ])[ktne ] = mfccc;
+ //(D.f[DIR_MMP ])[ktsw ] = mfaac;
+ //(D.f[DIR_PMP ])[ktse ] = mfcac;
+ //(D.f[DIR_MPP ])[ktnw ] = mfacc;
+ //(D.f[DIR_PPM ])[kbne ] = mfcca;
+ //(D.f[DIR_MMM ])[kbsw ] = mfaaa;
+ //(D.f[DIR_PMM ])[kbse ] = mfcaa;
+ //(D.f[DIR_MPM ])[kbnw ] = mfaca;
+
+ //(D.f[DIR_P00 ])[ke ] = fE ; //f1_E ; //fW; //fE ;
+ //(D.f[DIR_M00 ])[kw ] = fW ; //f1_W ; //fE; //fW ;
+ //(D.f[DIR_0P0 ])[kn ] = fN ; //f1_N ; //fS; //fN ;
+ //(D.f[DIR_0M0 ])[ks ] = fS ; //f1_S ; //fN; //fS ;
+ //(D.f[DIR_00P ])[kt ] = fT ; //f1_T ; //fB; //fT ;
+ //(D.f[DIR_00M ])[kb ] = fB ; //f1_B ; //fT; //fB ;
+ //(D.f[DIR_PP0 ])[kne ] = fNE; //f1_NE; //fSW; //fNE;
+ //(D.f[DIR_MM0 ])[ksw ] = fSW; //f1_SW; //fNE; //fSW;
+ //(D.f[DIR_PM0 ])[kse ] = fSE; //f1_SE; //fNW; //fSE;
+ //(D.f[DIR_MP0 ])[knw ] = fNW; //f1_NW; //fSE; //fNW;
+ //(D.f[DIR_P0P ])[kte ] = fTE; //f1_TE; //fBW; //fTE;
+ //(D.f[DIR_M0M ])[kbw ] = fBW; //f1_BW; //fTE; //fBW;
+ //(D.f[DIR_P0M ])[kbe ] = fBE; //f1_BE; //fTW; //fBE;
+ //(D.f[DIR_M0P ])[ktw ] = fTW; //f1_TW; //fBE; //fTW;
+ //(D.f[DIR_0PP ])[ktn ] = fTN; //f1_TN; //fBS; //fTN;
+ //(D.f[DIR_0MM ])[kbs ] = fBS; //f1_BS; //fTN; //fBS;
+ //(D.f[DIR_0PM ])[kbn ] = fBN; //f1_BN; //fTS; //fBN;
+ //(D.f[DIR_0MP ])[kts ] = fTS; //f1_TS; //fBN; //fTS;
//(D.f[DIR_000])[kzero] = fZERO;//f1_ZERO; //fZERO; //fZERO;
- //(D.f[DIR_PPP ])[ktne ] = fTNE; //f1_TNE; //fBSW; //fTNE;
- //(D.f[DIR_MMM ])[kbsw ] = fBSW; //f1_BSW; //fTNE; //fBSW;
- //(D.f[DIR_PPM ])[kbne ] = fBNE; //f1_BNE; //fTSW; //fBNE;
- //(D.f[DIR_MMP ])[ktsw ] = fTSW; //f1_TSW; //fBNE; //fTSW;
- //(D.f[DIR_PMP ])[ktse ] = fTSE; //f1_TSE; //fBNW; //fTSE;
- //(D.f[DIR_MPM ])[kbnw ] = fBNW; //f1_BNW; //fTSE; //fBNW;
- //(D.f[DIR_PMM ])[kbse ] = fBSE; //f1_BSE; //fTNW; //fBSE;
- //(D.f[DIR_MPP ])[ktnw ] = fTNW; //f1_TNW; //fBSE; //fTNW;
+ //(D.f[DIR_PPP ])[ktne ] = fTNE; //f1_TNE; //fBSW; //fTNE;
+ //(D.f[DIR_MMM ])[kbsw ] = fBSW; //f1_BSW; //fTNE; //fBSW;
+ //(D.f[DIR_PPM ])[kbne ] = fBNE; //f1_BNE; //fTSW; //fBNE;
+ //(D.f[DIR_MMP ])[ktsw ] = fTSW; //f1_TSW; //fBNE; //fTSW;
+ //(D.f[DIR_PMP ])[ktse ] = fTSE; //f1_TSE; //fBNW; //fTSE;
+ //(D.f[DIR_MPM ])[kbnw ] = fBNW; //f1_BNW; //fTSE; //fBNW;
+ //(D.f[DIR_PMM ])[kbse ] = fBSE; //f1_BSE; //fTNW; //fBSE;
+ //(D.f[DIR_MPP ])[ktnw ] = fTNW; //f1_TNW; //fBSE; //fTNW;
}
}
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
@@ -4295,19 +4295,19 @@ __global__ void QPressDeviceEQZ27(real* rhoBC,
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-__global__ void QPressDeviceZero27( real* DD,
- int* k_Q,
- unsigned int numberOfBCnodes,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+__global__ void QPressDeviceZero27( real* DD,
+ int* k_Q,
+ unsigned int numberOfBCnodes,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
////////////////////////////////////////////////////////////////////////////////
- const unsigned x = threadIdx.x; // Globaler x-Index
- const unsigned y = blockIdx.x; // Globaler y-Index
- const unsigned z = blockIdx.y; // Globaler z-Index
+ const unsigned x = threadIdx.x; // Globaler x-Index
+ const unsigned y = blockIdx.x; // Globaler y-Index
+ const unsigned z = blockIdx.y; // Globaler z-Index
const unsigned nx = blockDim.x;
const unsigned ny = gridDim.x;
@@ -4378,7 +4378,7 @@ __global__ void QPressDeviceZero27( real* DD,
D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
+ }
else
{
D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -4411,7 +4411,7 @@ __global__ void QPressDeviceZero27( real* DD,
}
//////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
//__syncthreads();
- //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+ //////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
(D.f[DIR_P00 ])[ke ] =c0o1;
(D.f[DIR_M00 ])[kw ] =c0o1;
(D.f[DIR_0P0 ])[kn ] =c0o1;
@@ -4483,21 +4483,21 @@ __global__ void QPressDeviceZero27( real* DD,
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
__global__ void QPressDeviceFake27( real* rhoBC,
- real* DD,
- int* k_Q,
- int* k_N,
- int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ int* k_Q,
+ int* k_N,
+ int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
////////////////////////////////////////////////////////////////////////////////
- const unsigned x = threadIdx.x; // Globaler x-Index
- const unsigned y = blockIdx.x; // Globaler y-Index
- const unsigned z = blockIdx.y; // Globaler z-Index
+ const unsigned x = threadIdx.x; // Globaler x-Index
+ const unsigned y = blockIdx.x; // Globaler y-Index
+ const unsigned z = blockIdx.y; // Globaler z-Index
const unsigned nx = blockDim.x;
const unsigned ny = gridDim.x;
@@ -4598,7 +4598,7 @@ __global__ void QPressDeviceFake27( real* rhoBC,
D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
+ }
else
{
D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
@@ -4665,24 +4665,24 @@ __global__ void QPressDeviceFake27( real* rhoBC,
real vx1, vx2, vx3;
vx1 = ((f1_TSE - f1_BNW) - (f1_TNW - f1_BSE)) + ((f1_TNE - f1_BSW) - (f1_TSW - f1_BNE)) +
((f1_BE - f1_TW) + (f1_TE - f1_BW)) + ((f1_SE - f1_NW) + (f1_NE - f1_SW)) +
- (f1_E - f1_W);
+ (f1_E - f1_W);
vx2 = (-(f1_TSE - f1_BNW) + (f1_TNW - f1_BSE)) + ((f1_TNE - f1_BSW) - (f1_TSW - f1_BNE)) +
((f1_BN - f1_TS) + (f1_TN - f1_BS)) + (-(f1_SE - f1_NW) + (f1_NE - f1_SW)) +
- (f1_N - f1_S);
+ (f1_N - f1_S);
vx3 = ((f1_TSE - f1_BNW) + (f1_TNW - f1_BSE)) + ((f1_TNE - f1_BSW) + (f1_TSW - f1_BNE)) +
(-(f1_BN - f1_TS) + (f1_TN - f1_BS)) + ((f1_TE - f1_BW) - (f1_BE - f1_TW)) +
- (f1_T - f1_B);
+ (f1_T - f1_B);
real cu_sq=c3o2*(vx1*vx1+vx2*vx2+vx3*vx3);
//////////////////////////////////////////////////////////////////////////
real drho1 = f1_ZERO+f1_E+f1_W+f1_N+f1_S+f1_T+f1_B+f1_NE+f1_SW+f1_SE+f1_NW+f1_TE+f1_BW+f1_BE+f1_TW+f1_TN+f1_BS+f1_BN+f1_TS+
f1_TNE+f1_TSW+f1_TSE+f1_TNW+f1_BNE+f1_BSW+f1_BSE+f1_BNW;
- //drho1 = (drho1 + rhoBC[k])/2.f;
- drho1 = drho1 - rhoBC[k];
+ //drho1 = (drho1 + rhoBC[k])/2.f;
+ drho1 = drho1 - rhoBC[k];
__syncthreads();
@@ -4712,7 +4712,7 @@ __global__ void QPressDeviceFake27( real* rhoBC,
(D.f[DIR_PPM ])[kbne ] = f1_TSW -c1o216*drho1; // c1o100; // zero; //
(D.f[DIR_MMM ])[kbsw ] = f1_TNE -c1o216*drho1; // c1o100; // zero; //
(D.f[DIR_PMM ])[kbse ] = f1_TNW -c1o216*drho1; // c1o100; // zero; //
- (D.f[DIR_MPM ])[kbnw ] = f1_TSE -c1o216*drho1; // c1o100; // zero; //
+ (D.f[DIR_MPM ])[kbnw ] = f1_TSE -c1o216*drho1; // c1o100; // zero; //
}
}
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
@@ -4757,460 +4757,460 @@ __global__ void QPressDeviceFake27( real* rhoBC,
//////////////////////////////////////////////////////////////////////////
__global__ void QPressDevice27_IntBB(real* rho,
- real* DD,
- int* k_Q,
- real* QQ,
- unsigned int numberOfBCnodes,
- real om1,
- unsigned int* neighborX,
- unsigned int* neighborY,
- unsigned int* neighborZ,
- unsigned int size_Mat,
- bool isEvenTimestep)
+ real* DD,
+ int* k_Q,
+ real* QQ,
+ unsigned int numberOfBCnodes,
+ real om1,
+ unsigned int* neighborX,
+ unsigned int* neighborY,
+ unsigned int* neighborZ,
+ unsigned int size_Mat,
+ bool isEvenTimestep)
{
- Distributions27 D;
- if (isEvenTimestep==true)
- {
- D.f[DIR_P00 ] = &DD[DIR_P00 *size_Mat];
- D.f[DIR_M00 ] = &DD[DIR_M00 *size_Mat];
- D.f[DIR_0P0 ] = &DD[DIR_0P0 *size_Mat];
- D.f[DIR_0M0 ] = &DD[DIR_0M0 *size_Mat];
- D.f[DIR_00P ] = &DD[DIR_00P *size_Mat];
- D.f[DIR_00M ] = &DD[DIR_00M *size_Mat];
- D.f[DIR_PP0 ] = &DD[DIR_PP0 *size_Mat];
- D.f[DIR_MM0 ] = &DD[DIR_MM0 *size_Mat];
- D.f[DIR_PM0 ] = &DD[DIR_PM0 *size_Mat];
- D.f[DIR_MP0 ] = &DD[DIR_MP0 *size_Mat];
- D.f[DIR_P0P ] = &DD[DIR_P0P *size_Mat];
- D.f[DIR_M0M ] = &DD[DIR_M0M *size_Mat];
- D.f[DIR_P0M ] = &DD[DIR_P0M *size_Mat];
- D.f[DIR_M0P ] = &DD[DIR_M0P *size_Mat];
- D.f[DIR_0PP ] = &DD[DIR_0PP *size_Mat];
- D.f[DIR_0MM ] = &DD[DIR_0MM *size_Mat];
- D.f[DIR_0PM ] = &DD[DIR_0PM *size_Mat];
- D.f[DIR_0MP ] = &DD[DIR_0MP *size_Mat];
- D.f[DIR_000] = &DD[DIR_000*size_Mat];
- D.f[DIR_PPP ] = &DD[DIR_PPP *size_Mat];
- D.f[DIR_MMP ] = &DD[DIR_MMP *size_Mat];
- D.f[DIR_PMP ] = &DD[DIR_PMP *size_Mat];
- D.f[DIR_MPP ] = &DD[DIR_MPP *size_Mat];
- D.f[DIR_PPM ] = &DD[DIR_PPM *size_Mat];
- D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
- D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
- D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
- else
- {
- D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
- D.f[DIR_P00 ] = &DD[DIR_M00 *size_Mat];
- D.f[DIR_0M0 ] = &DD[DIR_0P0 *size_Mat];
- D.f[DIR_0P0 ] = &DD[DIR_0M0 *size_Mat];
- D.f[DIR_00M ] = &DD[DIR_00P *size_Mat];
- D.f[DIR_00P ] = &DD[DIR_00M *size_Mat];
- D.f[DIR_MM0 ] = &DD[DIR_PP0 *size_Mat];
- D.f[DIR_PP0 ] = &DD[DIR_MM0 *size_Mat];
- D.f[DIR_MP0 ] = &DD[DIR_PM0 *size_Mat];
- D.f[DIR_PM0 ] = &DD[DIR_MP0 *size_Mat];
- D.f[DIR_M0M ] = &DD[DIR_P0P *size_Mat];
- D.f[DIR_P0P ] = &DD[DIR_M0M *size_Mat];
- D.f[DIR_M0P ] = &DD[DIR_P0M *size_Mat];
- D.f[DIR_P0M ] = &DD[DIR_M0P *size_Mat];
- D.f[DIR_0MM ] = &DD[DIR_0PP *size_Mat];
- D.f[DIR_0PP ] = &DD[DIR_0MM *size_Mat];
- D.f[DIR_0MP ] = &DD[DIR_0PM *size_Mat];
- D.f[DIR_0PM ] = &DD[DIR_0MP *size_Mat];
- D.f[DIR_000] = &DD[DIR_000*size_Mat];
- D.f[DIR_PPP ] = &DD[DIR_MMM *size_Mat];
- D.f[DIR_MMP ] = &DD[DIR_PPM *size_Mat];
- D.f[DIR_PMP ] = &DD[DIR_MPM *size_Mat];
- D.f[DIR_MPP ] = &DD[DIR_PMM *size_Mat];
- D.f[DIR_PPM ] = &DD[DIR_MMP *size_Mat];
- D.f[DIR_MMM ] = &DD[DIR_PPP *size_Mat];
- D.f[DIR_PMM ] = &DD[DIR_MPP *size_Mat];
- D.f[DIR_MPM ] = &DD[DIR_PMP *size_Mat];
- }
- ////////////////////////////////////////////////////////////////////////////////
- const unsigned x = threadIdx.x; // Globaler x-Index
- const unsigned y = blockIdx.x; // Globaler y-Index
- const unsigned z = blockIdx.y; // Globaler z-Index
-
- const unsigned nx = blockDim.x;
- const unsigned ny = gridDim.x;
-
- const unsigned k = nx*(ny*z + y) + x;
- //////////////////////////////////////////////////////////////////////////
-
- if(k < numberOfBCnodes)
- {
- ////////////////////////////////////////////////////////////////////////////////
- //real VeloX = vx[k];
- //real VeloY = vy[k];
- //real VeloZ = vz[k]; //(16.0*(u0*2.0)*bbx*bby*(grid_nx-bbx)*(grid_ny-bby))/(grid_nx*grid_nx*grid_ny*grid_ny)
- ////////////////////////////////////////////////////////////////////////////////
- real *q_dirE, *q_dirW, *q_dirN, *q_dirS, *q_dirT, *q_dirB,
- *q_dirNE, *q_dirSW, *q_dirSE, *q_dirNW, *q_dirTE, *q_dirBW,
- *q_dirBE, *q_dirTW, *q_dirTN, *q_dirBS, *q_dirBN, *q_dirTS,
- *q_dirTNE, *q_dirTSW, *q_dirTSE, *q_dirTNW, *q_dirBNE, *q_dirBSW,
- *q_dirBSE, *q_dirBNW;
- q_dirE = &QQ[DIR_P00 * numberOfBCnodes];
- q_dirW = &QQ[DIR_M00 * numberOfBCnodes];
- q_dirN = &QQ[DIR_0P0 * numberOfBCnodes];
- q_dirS = &QQ[DIR_0M0 * numberOfBCnodes];
- q_dirT = &QQ[DIR_00P * numberOfBCnodes];
- q_dirB = &QQ[DIR_00M * numberOfBCnodes];
- q_dirNE = &QQ[DIR_PP0 * numberOfBCnodes];
- q_dirSW = &QQ[DIR_MM0 * numberOfBCnodes];
- q_dirSE = &QQ[DIR_PM0 * numberOfBCnodes];
- q_dirNW = &QQ[DIR_MP0 * numberOfBCnodes];
- q_dirTE = &QQ[DIR_P0P * numberOfBCnodes];
- q_dirBW = &QQ[DIR_M0M * numberOfBCnodes];
- q_dirBE = &QQ[DIR_P0M * numberOfBCnodes];
- q_dirTW = &QQ[DIR_M0P * numberOfBCnodes];
- q_dirTN = &QQ[DIR_0PP * numberOfBCnodes];
- q_dirBS = &QQ[DIR_0MM * numberOfBCnodes];
- q_dirBN = &QQ[DIR_0PM * numberOfBCnodes];
- q_dirTS = &QQ[DIR_0MP * numberOfBCnodes];
- q_dirTNE = &QQ[DIR_PPP * numberOfBCnodes];
- q_dirTSW = &QQ[DIR_MMP * numberOfBCnodes];
- q_dirTSE = &QQ[DIR_PMP * numberOfBCnodes];
- q_dirTNW = &QQ[DIR_MPP * numberOfBCnodes];
- q_dirBNE = &QQ[DIR_PPM * numberOfBCnodes];
- q_dirBSW = &QQ[DIR_MMM * numberOfBCnodes];
- q_dirBSE = &QQ[DIR_PMM * numberOfBCnodes];
- q_dirBNW = &QQ[DIR_MPM * numberOfBCnodes];
- ////////////////////////////////////////////////////////////////////////////////
- //index
- unsigned int KQK = k_Q[k];
- unsigned int kzero= KQK;
- unsigned int ke = KQK;
- unsigned int kw = neighborX[KQK];
- unsigned int kn = KQK;
- unsigned int ks = neighborY[KQK];
- unsigned int kt = KQK;
- unsigned int kb = neighborZ[KQK];
- unsigned int ksw = neighborY[kw];
- unsigned int kne = KQK;
- unsigned int kse = ks;
- unsigned int knw = kw;
- unsigned int kbw = neighborZ[kw];
- unsigned int kte = KQK;
- unsigned int kbe = kb;
- unsigned int ktw = kw;
- unsigned int kbs = neighborZ[ks];
- unsigned int ktn = KQK;
- unsigned int kbn = kb;
- unsigned int kts = ks;
- unsigned int ktse = ks;
- unsigned int kbnw = kbw;
- unsigned int ktnw = kw;
- unsigned int kbse = kbs;
- unsigned int ktsw = ksw;
- unsigned int kbne = kb;
- unsigned int ktne = KQK;
- unsigned int kbsw = neighborZ[ksw];
- ////////////////////////////////////////////////////////////////////////////////
- real f_E, f_W, f_N, f_S, f_T, f_B, f_NE, f_SW, f_SE, f_NW, f_TE, f_BW, f_BE,
- f_TW, f_TN, f_BS, f_BN, f_TS, f_TNE, f_TSW, f_TSE, f_TNW, f_BNE, f_BSW, f_BSE, f_BNW;
-
- f_W = (D.f[DIR_P00 ])[ke ];
- f_E = (D.f[DIR_M00 ])[kw ];
- f_S = (D.f[DIR_0P0 ])[kn ];
- f_N = (D.f[DIR_0M0 ])[ks ];
- f_B = (D.f[DIR_00P ])[kt ];
- f_T = (D.f[DIR_00M ])[kb ];
- f_SW = (D.f[DIR_PP0 ])[kne ];
- f_NE = (D.f[DIR_MM0 ])[ksw ];
- f_NW = (D.f[DIR_PM0 ])[kse ];
- f_SE = (D.f[DIR_MP0 ])[knw ];
- f_BW = (D.f[DIR_P0P ])[kte ];
- f_TE = (D.f[DIR_M0M ])[kbw ];
- f_TW = (D.f[DIR_P0M ])[kbe ];
- f_BE = (D.f[DIR_M0P ])[ktw ];
- f_BS = (D.f[DIR_0PP ])[ktn ];
- f_TN = (D.f[DIR_0MM ])[kbs ];
- f_TS = (D.f[DIR_0PM ])[kbn ];
- f_BN = (D.f[DIR_0MP ])[kts ];
- f_BSW = (D.f[DIR_PPP ])[ktne ];
- f_BNE = (D.f[DIR_MMP ])[ktsw ];
- f_BNW = (D.f[DIR_PMP ])[ktse ];
- f_BSE = (D.f[DIR_MPP ])[ktnw ];
- f_TSW = (D.f[DIR_PPM ])[kbne ];
- f_TNE = (D.f[DIR_MMM ])[kbsw ];
- f_TNW = (D.f[DIR_PMM ])[kbse ];
- f_TSE = (D.f[DIR_MPM ])[kbnw ];
- ////////////////////////////////////////////////////////////////////////////////
- real vx1, vx2, vx3, drho, feq, q;
- drho = f_TSE + f_TNW + f_TNE + f_TSW + f_BSE + f_BNW + f_BNE + f_BSW +
- f_BN + f_TS + f_TN + f_BS + f_BE + f_TW + f_TE + f_BW + f_SE + f_NW + f_NE + f_SW +
- f_T + f_B + f_N + f_S + f_E + f_W + ((D.f[DIR_000])[kzero]);
-
- vx1 = (((f_TSE - f_BNW) - (f_TNW - f_BSE)) + ((f_TNE - f_BSW) - (f_TSW - f_BNE)) +
- ((f_BE - f_TW) + (f_TE - f_BW)) + ((f_SE - f_NW) + (f_NE - f_SW)) +
- (f_E - f_W))/(c1o1+drho);
-
-
- vx2 = ((-(f_TSE - f_BNW) + (f_TNW - f_BSE)) + ((f_TNE - f_BSW) - (f_TSW - f_BNE)) +
- ((f_BN - f_TS) + (f_TN - f_BS)) + (-(f_SE - f_NW) + (f_NE - f_SW)) +
- (f_N - f_S))/(c1o1+drho);
-
- vx3 = (((f_TSE - f_BNW) + (f_TNW - f_BSE)) + ((f_TNE - f_BSW) + (f_TSW - f_BNE)) +
- (-(f_BN - f_TS) + (f_TN - f_BS)) + ((f_TE - f_BW) - (f_BE - f_TW)) +
- (f_T - f_B))/(c1o1+drho);
-
- real cu_sq=c3o2*(vx1*vx1+vx2*vx2+vx3*vx3);
-
- //////////////////////////////////////////////////////////////////////////
- if (isEvenTimestep==false)
- {
- D.f[DIR_P00 ] = &DD[DIR_P00 *size_Mat];
- D.f[DIR_M00 ] = &DD[DIR_M00 *size_Mat];
- D.f[DIR_0P0 ] = &DD[DIR_0P0 *size_Mat];
- D.f[DIR_0M0 ] = &DD[DIR_0M0 *size_Mat];
- D.f[DIR_00P ] = &DD[DIR_00P *size_Mat];
- D.f[DIR_00M ] = &DD[DIR_00M *size_Mat];
- D.f[DIR_PP0 ] = &DD[DIR_PP0 *size_Mat];
- D.f[DIR_MM0 ] = &DD[DIR_MM0 *size_Mat];
- D.f[DIR_PM0 ] = &DD[DIR_PM0 *size_Mat];
- D.f[DIR_MP0 ] = &DD[DIR_MP0 *size_Mat];
- D.f[DIR_P0P ] = &DD[DIR_P0P *size_Mat];
- D.f[DIR_M0M ] = &DD[DIR_M0M *size_Mat];
- D.f[DIR_P0M ] = &DD[DIR_P0M *size_Mat];
- D.f[DIR_M0P ] = &DD[DIR_M0P *size_Mat];
- D.f[DIR_0PP ] = &DD[DIR_0PP *size_Mat];
- D.f[DIR_0MM ] = &DD[DIR_0MM *size_Mat];
- D.f[DIR_0PM ] = &DD[DIR_0PM *size_Mat];
- D.f[DIR_0MP ] = &DD[DIR_0MP *size_Mat];
- D.f[DIR_000] = &DD[DIR_000*size_Mat];
- D.f[DIR_PPP ] = &DD[DIR_PPP *size_Mat];
- D.f[DIR_MMP ] = &DD[DIR_MMP *size_Mat];
- D.f[DIR_PMP ] = &DD[DIR_PMP *size_Mat];
- D.f[DIR_MPP ] = &DD[DIR_MPP *size_Mat];
- D.f[DIR_PPM ] = &DD[DIR_PPM *size_Mat];
- D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
- D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
- D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
- }
- else
- {
- D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
- D.f[DIR_P00 ] = &DD[DIR_M00 *size_Mat];
- D.f[DIR_0M0 ] = &DD[DIR_0P0 *size_Mat];
- D.f[DIR_0P0 ] = &DD[DIR_0M0 *size_Mat];
- D.f[DIR_00M ] = &DD[DIR_00P *size_Mat];
- D.f[DIR_00P ] = &DD[DIR_00M *size_Mat];
- D.f[DIR_MM0 ] = &DD[DIR_PP0 *size_Mat];
- D.f[DIR_PP0 ] = &DD[DIR_MM0 *size_Mat];
- D.f[DIR_MP0 ] = &DD[DIR_PM0 *size_Mat];
- D.f[DIR_PM0 ] = &DD[DIR_MP0 *size_Mat];
- D.f[DIR_M0M ] = &DD[DIR_P0P *size_Mat];
- D.f[DIR_P0P ] = &DD[DIR_M0M *size_Mat];
- D.f[DIR_M0P ] = &DD[DIR_P0M *size_Mat];
- D.f[DIR_P0M ] = &DD[DIR_M0P *size_Mat];
- D.f[DIR_0MM ] = &DD[DIR_0PP *size_Mat];
- D.f[DIR_0PP ] = &DD[DIR_0MM *size_Mat];
- D.f[DIR_0MP ] = &DD[DIR_0PM *size_Mat];
- D.f[DIR_0PM ] = &DD[DIR_0MP *size_Mat];
- D.f[DIR_000] = &DD[DIR_000*size_Mat];
- D.f[DIR_PPP ] = &DD[DIR_MMM *size_Mat];
- D.f[DIR_MMP ] = &DD[DIR_PPM *size_Mat];
- D.f[DIR_PMP ] = &DD[DIR_MPM *size_Mat];
- D.f[DIR_MPP ] = &DD[DIR_PMM *size_Mat];
- D.f[DIR_PPM ] = &DD[DIR_MMP *size_Mat];
- D.f[DIR_MMM ] = &DD[DIR_PPP *size_Mat];
- D.f[DIR_PMM ] = &DD[DIR_MPP *size_Mat];
- D.f[DIR_MPM ] = &DD[DIR_PMP *size_Mat];
- }
- ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
- //Test
- //(D.f[DIR_000])[k]=c1o10;
- real rhoDiff = drho - rho[k];
- real VeloX = vx1;
- real VeloY = vx2;
- real VeloZ = vx3;
- ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
-
- q = q_dirE[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c2o27* (drho+c9o2*( vx1 )*( vx1 )-cu_sq);
- (D.f[DIR_M00])[kw]=(c1o1-q)/(c1o1+q)*(f_E-f_W+(f_E+f_W-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_E+f_W)-c2o27*(rhoDiff + c6o1*( VeloX )))/(c1o1+q);
- }
-
- q = q_dirW[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c2o27* (drho+c9o2*(-vx1 )*(-vx1 )-cu_sq);
- (D.f[DIR_P00])[ke]=(c1o1-q)/(c1o1+q)*(f_W-f_E+(f_W+f_E-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_W+f_E)-c2o27*(rhoDiff + c6o1*(-VeloX )))/(c1o1+q);
- }
-
- q = q_dirN[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c2o27* (drho+c9o2*( vx2 )*( vx2 )-cu_sq);
- (D.f[DIR_0M0])[ks]=(c1o1-q)/(c1o1+q)*(f_N-f_S+(f_N+f_S-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_N+f_S)-c2o27*(rhoDiff + c6o1*( VeloY )))/(c1o1+q);
- }
-
- q = q_dirS[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c2o27* (drho+c9o2*( -vx2 )*( -vx2 )-cu_sq);
- (D.f[DIR_0P0])[kn]=(c1o1-q)/(c1o1+q)*(f_S-f_N+(f_S+f_N-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_S+f_N)-c2o27*(rhoDiff + c6o1*(-VeloY )))/(c1o1+q);
- }
-
- q = q_dirT[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c2o27* (drho+c9o2*( vx3)*( vx3)-cu_sq);
- (D.f[DIR_00M])[kb]=(c1o1-q)/(c1o1+q)*(f_T-f_B+(f_T+f_B-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_T+f_B)-c2o27*(rhoDiff + c6o1*( VeloZ )))/(c1o1+q);
- }
-
- q = q_dirB[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c2o27* (drho+c9o2*( -vx3)*( -vx3)-cu_sq);
- (D.f[DIR_00P])[kt]=(c1o1-q)/(c1o1+q)*(f_B-f_T+(f_B+f_T-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_B+f_T)-c2o27*(rhoDiff + c6o1*(-VeloZ )))/(c1o1+q);
- }
-
- q = q_dirNE[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o54* (drho+c9o2*( vx1+vx2 )*( vx1+vx2 )-cu_sq);
- (D.f[DIR_MM0])[ksw]=(c1o1-q)/(c1o1+q)*(f_NE-f_SW+(f_NE+f_SW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_NE+f_SW)-c1o54*(rhoDiff + c6o1*(VeloX+VeloY)))/(c1o1+q);
- }
-
- q = q_dirSW[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o54* (drho+c9o2*(-vx1-vx2 )*(-vx1-vx2 )-cu_sq);
- (D.f[DIR_PP0])[kne]=(c1o1-q)/(c1o1+q)*(f_SW-f_NE+(f_SW+f_NE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_SW+f_NE)-c1o54*(rhoDiff + c6o1*(-VeloX-VeloY)))/(c1o1+q);
- }
-
- q = q_dirSE[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o54* (drho+c9o2*( vx1-vx2 )*( vx1-vx2 )-cu_sq);
- (D.f[DIR_MP0])[knw]=(c1o1-q)/(c1o1+q)*(f_SE-f_NW+(f_SE+f_NW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_SE+f_NW)-c1o54*(rhoDiff + c6o1*( VeloX-VeloY)))/(c1o1+q);
- }
-
- q = q_dirNW[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o54* (drho+c9o2*(-vx1+vx2 )*(-vx1+vx2 )-cu_sq);
- (D.f[DIR_PM0])[kse]=(c1o1-q)/(c1o1+q)*(f_NW-f_SE+(f_NW+f_SE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_NW+f_SE)-c1o54*(rhoDiff + c6o1*(-VeloX+VeloY)))/(c1o1+q);
- }
-
- q = q_dirTE[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o54* (drho+c9o2*( vx1 +vx3)*( vx1 +vx3)-cu_sq);
- (D.f[DIR_M0M])[kbw]=(c1o1-q)/(c1o1+q)*(f_TE-f_BW+(f_TE+f_BW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TE+f_BW)-c1o54*(rhoDiff + c6o1*( VeloX+VeloZ)))/(c1o1+q);
- }
-
- q = q_dirBW[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o54* (drho+c9o2*(-vx1 -vx3)*(-vx1 -vx3)-cu_sq);
- (D.f[DIR_P0P])[kte]=(c1o1-q)/(c1o1+q)*(f_BW-f_TE+(f_BW+f_TE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BW+f_TE)-c1o54*(rhoDiff + c6o1*(-VeloX-VeloZ)))/(c1o1+q);
- }
-
- q = q_dirBE[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o54* (drho+c9o2*( vx1 -vx3)*( vx1 -vx3)-cu_sq);
- (D.f[DIR_M0P])[ktw]=(c1o1-q)/(c1o1+q)*(f_BE-f_TW+(f_BE+f_TW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BE+f_TW)-c1o54*(rhoDiff + c6o1*( VeloX-VeloZ)))/(c1o1+q);
- }
-
- q = q_dirTW[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o54* (drho+c9o2*(-vx1 +vx3)*(-vx1 +vx3)-cu_sq);
- (D.f[DIR_P0M])[kbe]=(c1o1-q)/(c1o1+q)*(f_TW-f_BE+(f_TW+f_BE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TW+f_BE)-c1o54*(rhoDiff + c6o1*(-VeloX+VeloZ)))/(c1o1+q);
- }
-
- q = q_dirTN[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o54* (drho+c9o2*( vx2+vx3)*( vx2+vx3)-cu_sq);
- (D.f[DIR_0MM])[kbs]=(c1o1-q)/(c1o1+q)*(f_TN-f_BS+(f_TN+f_BS-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TN+f_BS)-c1o54*(rhoDiff + c6o1*( VeloY+VeloZ)))/(c1o1+q);
- }
-
- q = q_dirBS[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o54* (drho+c9o2*( -vx2-vx3)*( -vx2-vx3)-cu_sq);
- (D.f[DIR_0PP])[ktn]=(c1o1-q)/(c1o1+q)*(f_BS-f_TN+(f_BS+f_TN-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BS+f_TN)-c1o54*(rhoDiff + c6o1*( -VeloY-VeloZ)))/(c1o1+q);
- }
-
- q = q_dirBN[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o54* (drho+c9o2*( vx2-vx3)*( vx2-vx3)-cu_sq);
- (D.f[DIR_0MP])[kts]=(c1o1-q)/(c1o1+q)*(f_BN-f_TS+(f_BN+f_TS-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BN+f_TS)-c1o54*(rhoDiff + c6o1*( VeloY-VeloZ)))/(c1o1+q);
- }
-
- q = q_dirTS[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o54* (drho+c9o2*( -vx2+vx3)*( -vx2+vx3)-cu_sq);
- (D.f[DIR_0PM])[kbn]=(c1o1-q)/(c1o1+q)*(f_TS-f_BN+(f_TS+f_BN-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TS+f_BN)-c1o54*(rhoDiff + c6o1*( -VeloY+VeloZ)))/(c1o1+q);
- }
-
- q = q_dirTNE[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o216*(drho+c9o2*( vx1+vx2+vx3)*( vx1+vx2+vx3)-cu_sq);
- (D.f[DIR_MMM])[kbsw]=(c1o1-q)/(c1o1+q)*(f_TNE-f_BSW+(f_TNE+f_BSW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TNE+f_BSW)-c1o216*(rhoDiff + c6o1*( VeloX+VeloY+VeloZ)))/(c1o1+q);
- }
-
- q = q_dirBSW[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o216*(drho+c9o2*(-vx1-vx2-vx3)*(-vx1-vx2-vx3)-cu_sq);
- (D.f[DIR_PPP])[ktne]=(c1o1-q)/(c1o1+q)*(f_BSW-f_TNE+(f_BSW+f_TNE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BSW+f_TNE)-c1o216*(rhoDiff + c6o1*(-VeloX-VeloY-VeloZ)))/(c1o1+q);
- }
-
- q = q_dirBNE[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o216*(drho+c9o2*( vx1+vx2-vx3)*( vx1+vx2-vx3)-cu_sq);
- (D.f[DIR_MMP])[ktsw]=(c1o1-q)/(c1o1+q)*(f_BNE-f_TSW+(f_BNE+f_TSW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BNE+f_TSW)-c1o216*(rhoDiff + c6o1*( VeloX+VeloY-VeloZ)))/(c1o1+q);
- }
-
- q = q_dirTSW[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o216*(drho+c9o2*(-vx1-vx2+vx3)*(-vx1-vx2+vx3)-cu_sq);
- (D.f[DIR_PPM])[kbne]=(c1o1-q)/(c1o1+q)*(f_TSW-f_BNE+(f_TSW+f_BNE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TSW+f_BNE)-c1o216*(rhoDiff + c6o1*(-VeloX-VeloY+VeloZ)))/(c1o1+q);
- }
-
- q = q_dirTSE[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o216*(drho+c9o2*( vx1-vx2+vx3)*( vx1-vx2+vx3)-cu_sq);
- (D.f[DIR_MPM])[kbnw]=(c1o1-q)/(c1o1+q)*(f_TSE-f_BNW+(f_TSE+f_BNW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TSE+f_BNW)-c1o216*(rhoDiff + c6o1*( VeloX-VeloY+VeloZ)))/(c1o1+q);
- }
-
- q = q_dirBNW[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o216*(drho+c9o2*(-vx1+vx2-vx3)*(-vx1+vx2-vx3)-cu_sq);
- (D.f[DIR_PMP])[ktse]=(c1o1-q)/(c1o1+q)*(f_BNW-f_TSE+(f_BNW+f_TSE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BNW+f_TSE)-c1o216*(rhoDiff + c6o1*(-VeloX+VeloY-VeloZ)))/(c1o1+q);
- }
-
- q = q_dirBSE[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o216*(drho+c9o2*( vx1-vx2-vx3)*( vx1-vx2-vx3)-cu_sq);
- (D.f[DIR_MPP])[ktnw]=(c1o1-q)/(c1o1+q)*(f_BSE-f_TNW+(f_BSE+f_TNW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BSE+f_TNW)-c1o216*(rhoDiff + c6o1*( VeloX-VeloY-VeloZ)))/(c1o1+q);
- }
-
- q = q_dirTNW[k];
- if (q>=c0o1 && q<=c1o1)
- {
- feq=c1o216*(drho+c9o2*(-vx1+vx2+vx3)*(-vx1+vx2+vx3)-cu_sq);
- (D.f[DIR_PMM])[kbse]=(c1o1-q)/(c1o1+q)*(f_TNW-f_BSE+(f_TNW+f_BSE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TNW+f_BSE)-c1o216*(rhoDiff + c6o1*(-VeloX+VeloY+VeloZ)))/(c1o1+q);
- }
- }
+ Distributions27 D;
+ if (isEvenTimestep==true)
+ {
+ D.f[DIR_P00 ] = &DD[DIR_P00 *size_Mat];
+ D.f[DIR_M00 ] = &DD[DIR_M00 *size_Mat];
+ D.f[DIR_0P0 ] = &DD[DIR_0P0 *size_Mat];
+ D.f[DIR_0M0 ] = &DD[DIR_0M0 *size_Mat];
+ D.f[DIR_00P ] = &DD[DIR_00P *size_Mat];
+ D.f[DIR_00M ] = &DD[DIR_00M *size_Mat];
+ D.f[DIR_PP0 ] = &DD[DIR_PP0 *size_Mat];
+ D.f[DIR_MM0 ] = &DD[DIR_MM0 *size_Mat];
+ D.f[DIR_PM0 ] = &DD[DIR_PM0 *size_Mat];
+ D.f[DIR_MP0 ] = &DD[DIR_MP0 *size_Mat];
+ D.f[DIR_P0P ] = &DD[DIR_P0P *size_Mat];
+ D.f[DIR_M0M ] = &DD[DIR_M0M *size_Mat];
+ D.f[DIR_P0M ] = &DD[DIR_P0M *size_Mat];
+ D.f[DIR_M0P ] = &DD[DIR_M0P *size_Mat];
+ D.f[DIR_0PP ] = &DD[DIR_0PP *size_Mat];
+ D.f[DIR_0MM ] = &DD[DIR_0MM *size_Mat];
+ D.f[DIR_0PM ] = &DD[DIR_0PM *size_Mat];
+ D.f[DIR_0MP ] = &DD[DIR_0MP *size_Mat];
+ D.f[DIR_000] = &DD[DIR_000*size_Mat];
+ D.f[DIR_PPP ] = &DD[DIR_PPP *size_Mat];
+ D.f[DIR_MMP ] = &DD[DIR_MMP *size_Mat];
+ D.f[DIR_PMP ] = &DD[DIR_PMP *size_Mat];
+ D.f[DIR_MPP ] = &DD[DIR_MPP *size_Mat];
+ D.f[DIR_PPM ] = &DD[DIR_PPM *size_Mat];
+ D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
+ D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
+ D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
+ }
+ else
+ {
+ D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
+ D.f[DIR_P00 ] = &DD[DIR_M00 *size_Mat];
+ D.f[DIR_0M0 ] = &DD[DIR_0P0 *size_Mat];
+ D.f[DIR_0P0 ] = &DD[DIR_0M0 *size_Mat];
+ D.f[DIR_00M ] = &DD[DIR_00P *size_Mat];
+ D.f[DIR_00P ] = &DD[DIR_00M *size_Mat];
+ D.f[DIR_MM0 ] = &DD[DIR_PP0 *size_Mat];
+ D.f[DIR_PP0 ] = &DD[DIR_MM0 *size_Mat];
+ D.f[DIR_MP0 ] = &DD[DIR_PM0 *size_Mat];
+ D.f[DIR_PM0 ] = &DD[DIR_MP0 *size_Mat];
+ D.f[DIR_M0M ] = &DD[DIR_P0P *size_Mat];
+ D.f[DIR_P0P ] = &DD[DIR_M0M *size_Mat];
+ D.f[DIR_M0P ] = &DD[DIR_P0M *size_Mat];
+ D.f[DIR_P0M ] = &DD[DIR_M0P *size_Mat];
+ D.f[DIR_0MM ] = &DD[DIR_0PP *size_Mat];
+ D.f[DIR_0PP ] = &DD[DIR_0MM *size_Mat];
+ D.f[DIR_0MP ] = &DD[DIR_0PM *size_Mat];
+ D.f[DIR_0PM ] = &DD[DIR_0MP *size_Mat];
+ D.f[DIR_000] = &DD[DIR_000*size_Mat];
+ D.f[DIR_PPP ] = &DD[DIR_MMM *size_Mat];
+ D.f[DIR_MMP ] = &DD[DIR_PPM *size_Mat];
+ D.f[DIR_PMP ] = &DD[DIR_MPM *size_Mat];
+ D.f[DIR_MPP ] = &DD[DIR_PMM *size_Mat];
+ D.f[DIR_PPM ] = &DD[DIR_MMP *size_Mat];
+ D.f[DIR_MMM ] = &DD[DIR_PPP *size_Mat];
+ D.f[DIR_PMM ] = &DD[DIR_MPP *size_Mat];
+ D.f[DIR_MPM ] = &DD[DIR_PMP *size_Mat];
+ }
+ ////////////////////////////////////////////////////////////////////////////////
+ const unsigned x = threadIdx.x; // Globaler x-Index
+ const unsigned y = blockIdx.x; // Globaler y-Index
+ const unsigned z = blockIdx.y; // Globaler z-Index
+
+ const unsigned nx = blockDim.x;
+ const unsigned ny = gridDim.x;
+
+ const unsigned k = nx*(ny*z + y) + x;
+ //////////////////////////////////////////////////////////////////////////
+
+ if(k < numberOfBCnodes)
+ {
+ ////////////////////////////////////////////////////////////////////////////////
+ //real VeloX = vx[k];
+ //real VeloY = vy[k];
+ //real VeloZ = vz[k]; //(16.0*(u0*2.0)*bbx*bby*(grid_nx-bbx)*(grid_ny-bby))/(grid_nx*grid_nx*grid_ny*grid_ny)
+ ////////////////////////////////////////////////////////////////////////////////
+ real *q_dirE, *q_dirW, *q_dirN, *q_dirS, *q_dirT, *q_dirB,
+ *q_dirNE, *q_dirSW, *q_dirSE, *q_dirNW, *q_dirTE, *q_dirBW,
+ *q_dirBE, *q_dirTW, *q_dirTN, *q_dirBS, *q_dirBN, *q_dirTS,
+ *q_dirTNE, *q_dirTSW, *q_dirTSE, *q_dirTNW, *q_dirBNE, *q_dirBSW,
+ *q_dirBSE, *q_dirBNW;
+ q_dirE = &QQ[DIR_P00 * numberOfBCnodes];
+ q_dirW = &QQ[DIR_M00 * numberOfBCnodes];
+ q_dirN = &QQ[DIR_0P0 * numberOfBCnodes];
+ q_dirS = &QQ[DIR_0M0 * numberOfBCnodes];
+ q_dirT = &QQ[DIR_00P * numberOfBCnodes];
+ q_dirB = &QQ[DIR_00M * numberOfBCnodes];
+ q_dirNE = &QQ[DIR_PP0 * numberOfBCnodes];
+ q_dirSW = &QQ[DIR_MM0 * numberOfBCnodes];
+ q_dirSE = &QQ[DIR_PM0 * numberOfBCnodes];
+ q_dirNW = &QQ[DIR_MP0 * numberOfBCnodes];
+ q_dirTE = &QQ[DIR_P0P * numberOfBCnodes];
+ q_dirBW = &QQ[DIR_M0M * numberOfBCnodes];
+ q_dirBE = &QQ[DIR_P0M * numberOfBCnodes];
+ q_dirTW = &QQ[DIR_M0P * numberOfBCnodes];
+ q_dirTN = &QQ[DIR_0PP * numberOfBCnodes];
+ q_dirBS = &QQ[DIR_0MM * numberOfBCnodes];
+ q_dirBN = &QQ[DIR_0PM * numberOfBCnodes];
+ q_dirTS = &QQ[DIR_0MP * numberOfBCnodes];
+ q_dirTNE = &QQ[DIR_PPP * numberOfBCnodes];
+ q_dirTSW = &QQ[DIR_MMP * numberOfBCnodes];
+ q_dirTSE = &QQ[DIR_PMP * numberOfBCnodes];
+ q_dirTNW = &QQ[DIR_MPP * numberOfBCnodes];
+ q_dirBNE = &QQ[DIR_PPM * numberOfBCnodes];
+ q_dirBSW = &QQ[DIR_MMM * numberOfBCnodes];
+ q_dirBSE = &QQ[DIR_PMM * numberOfBCnodes];
+ q_dirBNW = &QQ[DIR_MPM * numberOfBCnodes];
+ ////////////////////////////////////////////////////////////////////////////////
+ //index
+ unsigned int KQK = k_Q[k];
+ unsigned int kzero= KQK;
+ unsigned int ke = KQK;
+ unsigned int kw = neighborX[KQK];
+ unsigned int kn = KQK;
+ unsigned int ks = neighborY[KQK];
+ unsigned int kt = KQK;
+ unsigned int kb = neighborZ[KQK];
+ unsigned int ksw = neighborY[kw];
+ unsigned int kne = KQK;
+ unsigned int kse = ks;
+ unsigned int knw = kw;
+ unsigned int kbw = neighborZ[kw];
+ unsigned int kte = KQK;
+ unsigned int kbe = kb;
+ unsigned int ktw = kw;
+ unsigned int kbs = neighborZ[ks];
+ unsigned int ktn = KQK;
+ unsigned int kbn = kb;
+ unsigned int kts = ks;
+ unsigned int ktse = ks;
+ unsigned int kbnw = kbw;
+ unsigned int ktnw = kw;
+ unsigned int kbse = kbs;
+ unsigned int ktsw = ksw;
+ unsigned int kbne = kb;
+ unsigned int ktne = KQK;
+ unsigned int kbsw = neighborZ[ksw];
+ ////////////////////////////////////////////////////////////////////////////////
+ real f_E, f_W, f_N, f_S, f_T, f_B, f_NE, f_SW, f_SE, f_NW, f_TE, f_BW, f_BE,
+ f_TW, f_TN, f_BS, f_BN, f_TS, f_TNE, f_TSW, f_TSE, f_TNW, f_BNE, f_BSW, f_BSE, f_BNW;
+
+ f_W = (D.f[DIR_P00 ])[ke ];
+ f_E = (D.f[DIR_M00 ])[kw ];
+ f_S = (D.f[DIR_0P0 ])[kn ];
+ f_N = (D.f[DIR_0M0 ])[ks ];
+ f_B = (D.f[DIR_00P ])[kt ];
+ f_T = (D.f[DIR_00M ])[kb ];
+ f_SW = (D.f[DIR_PP0 ])[kne ];
+ f_NE = (D.f[DIR_MM0 ])[ksw ];
+ f_NW = (D.f[DIR_PM0 ])[kse ];
+ f_SE = (D.f[DIR_MP0 ])[knw ];
+ f_BW = (D.f[DIR_P0P ])[kte ];
+ f_TE = (D.f[DIR_M0M ])[kbw ];
+ f_TW = (D.f[DIR_P0M ])[kbe ];
+ f_BE = (D.f[DIR_M0P ])[ktw ];
+ f_BS = (D.f[DIR_0PP ])[ktn ];
+ f_TN = (D.f[DIR_0MM ])[kbs ];
+ f_TS = (D.f[DIR_0PM ])[kbn ];
+ f_BN = (D.f[DIR_0MP ])[kts ];
+ f_BSW = (D.f[DIR_PPP ])[ktne ];
+ f_BNE = (D.f[DIR_MMP ])[ktsw ];
+ f_BNW = (D.f[DIR_PMP ])[ktse ];
+ f_BSE = (D.f[DIR_MPP ])[ktnw ];
+ f_TSW = (D.f[DIR_PPM ])[kbne ];
+ f_TNE = (D.f[DIR_MMM ])[kbsw ];
+ f_TNW = (D.f[DIR_PMM ])[kbse ];
+ f_TSE = (D.f[DIR_MPM ])[kbnw ];
+ ////////////////////////////////////////////////////////////////////////////////
+ real vx1, vx2, vx3, drho, feq, q;
+ drho = f_TSE + f_TNW + f_TNE + f_TSW + f_BSE + f_BNW + f_BNE + f_BSW +
+ f_BN + f_TS + f_TN + f_BS + f_BE + f_TW + f_TE + f_BW + f_SE + f_NW + f_NE + f_SW +
+ f_T + f_B + f_N + f_S + f_E + f_W + ((D.f[DIR_000])[kzero]);
+
+ vx1 = (((f_TSE - f_BNW) - (f_TNW - f_BSE)) + ((f_TNE - f_BSW) - (f_TSW - f_BNE)) +
+ ((f_BE - f_TW) + (f_TE - f_BW)) + ((f_SE - f_NW) + (f_NE - f_SW)) +
+ (f_E - f_W))/(c1o1+drho);
+
+
+ vx2 = ((-(f_TSE - f_BNW) + (f_TNW - f_BSE)) + ((f_TNE - f_BSW) - (f_TSW - f_BNE)) +
+ ((f_BN - f_TS) + (f_TN - f_BS)) + (-(f_SE - f_NW) + (f_NE - f_SW)) +
+ (f_N - f_S))/(c1o1+drho);
+
+ vx3 = (((f_TSE - f_BNW) + (f_TNW - f_BSE)) + ((f_TNE - f_BSW) + (f_TSW - f_BNE)) +
+ (-(f_BN - f_TS) + (f_TN - f_BS)) + ((f_TE - f_BW) - (f_BE - f_TW)) +
+ (f_T - f_B))/(c1o1+drho);
+
+ real cu_sq=c3o2*(vx1*vx1+vx2*vx2+vx3*vx3);
+
+ //////////////////////////////////////////////////////////////////////////
+ if (isEvenTimestep==false)
+ {
+ D.f[DIR_P00 ] = &DD[DIR_P00 *size_Mat];
+ D.f[DIR_M00 ] = &DD[DIR_M00 *size_Mat];
+ D.f[DIR_0P0 ] = &DD[DIR_0P0 *size_Mat];
+ D.f[DIR_0M0 ] = &DD[DIR_0M0 *size_Mat];
+ D.f[DIR_00P ] = &DD[DIR_00P *size_Mat];
+ D.f[DIR_00M ] = &DD[DIR_00M *size_Mat];
+ D.f[DIR_PP0 ] = &DD[DIR_PP0 *size_Mat];
+ D.f[DIR_MM0 ] = &DD[DIR_MM0 *size_Mat];
+ D.f[DIR_PM0 ] = &DD[DIR_PM0 *size_Mat];
+ D.f[DIR_MP0 ] = &DD[DIR_MP0 *size_Mat];
+ D.f[DIR_P0P ] = &DD[DIR_P0P *size_Mat];
+ D.f[DIR_M0M ] = &DD[DIR_M0M *size_Mat];
+ D.f[DIR_P0M ] = &DD[DIR_P0M *size_Mat];
+ D.f[DIR_M0P ] = &DD[DIR_M0P *size_Mat];
+ D.f[DIR_0PP ] = &DD[DIR_0PP *size_Mat];
+ D.f[DIR_0MM ] = &DD[DIR_0MM *size_Mat];
+ D.f[DIR_0PM ] = &DD[DIR_0PM *size_Mat];
+ D.f[DIR_0MP ] = &DD[DIR_0MP *size_Mat];
+ D.f[DIR_000] = &DD[DIR_000*size_Mat];
+ D.f[DIR_PPP ] = &DD[DIR_PPP *size_Mat];
+ D.f[DIR_MMP ] = &DD[DIR_MMP *size_Mat];
+ D.f[DIR_PMP ] = &DD[DIR_PMP *size_Mat];
+ D.f[DIR_MPP ] = &DD[DIR_MPP *size_Mat];
+ D.f[DIR_PPM ] = &DD[DIR_PPM *size_Mat];
+ D.f[DIR_MMM ] = &DD[DIR_MMM *size_Mat];
+ D.f[DIR_PMM ] = &DD[DIR_PMM *size_Mat];
+ D.f[DIR_MPM ] = &DD[DIR_MPM *size_Mat];
+ }
+ else
+ {
+ D.f[DIR_M00 ] = &DD[DIR_P00 *size_Mat];
+ D.f[DIR_P00 ] = &DD[DIR_M00 *size_Mat];
+ D.f[DIR_0M0 ] = &DD[DIR_0P0 *size_Mat];
+ D.f[DIR_0P0 ] = &DD[DIR_0M0 *size_Mat];
+ D.f[DIR_00M ] = &DD[DIR_00P *size_Mat];
+ D.f[DIR_00P ] = &DD[DIR_00M *size_Mat];
+ D.f[DIR_MM0 ] = &DD[DIR_PP0 *size_Mat];
+ D.f[DIR_PP0 ] = &DD[DIR_MM0 *size_Mat];
+ D.f[DIR_MP0 ] = &DD[DIR_PM0 *size_Mat];
+ D.f[DIR_PM0 ] = &DD[DIR_MP0 *size_Mat];
+ D.f[DIR_M0M ] = &DD[DIR_P0P *size_Mat];
+ D.f[DIR_P0P ] = &DD[DIR_M0M *size_Mat];
+ D.f[DIR_M0P ] = &DD[DIR_P0M *size_Mat];
+ D.f[DIR_P0M ] = &DD[DIR_M0P *size_Mat];
+ D.f[DIR_0MM ] = &DD[DIR_0PP *size_Mat];
+ D.f[DIR_0PP ] = &DD[DIR_0MM *size_Mat];
+ D.f[DIR_0MP ] = &DD[DIR_0PM *size_Mat];
+ D.f[DIR_0PM ] = &DD[DIR_0MP *size_Mat];
+ D.f[DIR_000] = &DD[DIR_000*size_Mat];
+ D.f[DIR_PPP ] = &DD[DIR_MMM *size_Mat];
+ D.f[DIR_MMP ] = &DD[DIR_PPM *size_Mat];
+ D.f[DIR_PMP ] = &DD[DIR_MPM *size_Mat];
+ D.f[DIR_MPP ] = &DD[DIR_PMM *size_Mat];
+ D.f[DIR_PPM ] = &DD[DIR_MMP *size_Mat];
+ D.f[DIR_MMM ] = &DD[DIR_PPP *size_Mat];
+ D.f[DIR_PMM ] = &DD[DIR_MPP *size_Mat];
+ D.f[DIR_MPM ] = &DD[DIR_PMP *size_Mat];
+ }
+ ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+ //Test
+ //(D.f[DIR_000])[k]=c1o10;
+ real rhoDiff = drho - rho[k];
+ real VeloX = vx1;
+ real VeloY = vx2;
+ real VeloZ = vx3;
+ ////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
+
+ q = q_dirE[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c2o27* (drho+c9o2*( vx1 )*( vx1 )-cu_sq);
+ (D.f[DIR_M00])[kw]=(c1o1-q)/(c1o1+q)*(f_E-f_W+(f_E+f_W-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_E+f_W)-c2o27*(rhoDiff + c6o1*( VeloX )))/(c1o1+q);
+ }
+
+ q = q_dirW[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c2o27* (drho+c9o2*(-vx1 )*(-vx1 )-cu_sq);
+ (D.f[DIR_P00])[ke]=(c1o1-q)/(c1o1+q)*(f_W-f_E+(f_W+f_E-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_W+f_E)-c2o27*(rhoDiff + c6o1*(-VeloX )))/(c1o1+q);
+ }
+
+ q = q_dirN[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c2o27* (drho+c9o2*( vx2 )*( vx2 )-cu_sq);
+ (D.f[DIR_0M0])[ks]=(c1o1-q)/(c1o1+q)*(f_N-f_S+(f_N+f_S-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_N+f_S)-c2o27*(rhoDiff + c6o1*( VeloY )))/(c1o1+q);
+ }
+
+ q = q_dirS[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c2o27* (drho+c9o2*( -vx2 )*( -vx2 )-cu_sq);
+ (D.f[DIR_0P0])[kn]=(c1o1-q)/(c1o1+q)*(f_S-f_N+(f_S+f_N-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_S+f_N)-c2o27*(rhoDiff + c6o1*(-VeloY )))/(c1o1+q);
+ }
+
+ q = q_dirT[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c2o27* (drho+c9o2*( vx3)*( vx3)-cu_sq);
+ (D.f[DIR_00M])[kb]=(c1o1-q)/(c1o1+q)*(f_T-f_B+(f_T+f_B-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_T+f_B)-c2o27*(rhoDiff + c6o1*( VeloZ )))/(c1o1+q);
+ }
+
+ q = q_dirB[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c2o27* (drho+c9o2*( -vx3)*( -vx3)-cu_sq);
+ (D.f[DIR_00P])[kt]=(c1o1-q)/(c1o1+q)*(f_B-f_T+(f_B+f_T-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_B+f_T)-c2o27*(rhoDiff + c6o1*(-VeloZ )))/(c1o1+q);
+ }
+
+ q = q_dirNE[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o54* (drho+c9o2*( vx1+vx2 )*( vx1+vx2 )-cu_sq);
+ (D.f[DIR_MM0])[ksw]=(c1o1-q)/(c1o1+q)*(f_NE-f_SW+(f_NE+f_SW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_NE+f_SW)-c1o54*(rhoDiff + c6o1*(VeloX+VeloY)))/(c1o1+q);
+ }
+
+ q = q_dirSW[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o54* (drho+c9o2*(-vx1-vx2 )*(-vx1-vx2 )-cu_sq);
+ (D.f[DIR_PP0])[kne]=(c1o1-q)/(c1o1+q)*(f_SW-f_NE+(f_SW+f_NE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_SW+f_NE)-c1o54*(rhoDiff + c6o1*(-VeloX-VeloY)))/(c1o1+q);
+ }
+
+ q = q_dirSE[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o54* (drho+c9o2*( vx1-vx2 )*( vx1-vx2 )-cu_sq);
+ (D.f[DIR_MP0])[knw]=(c1o1-q)/(c1o1+q)*(f_SE-f_NW+(f_SE+f_NW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_SE+f_NW)-c1o54*(rhoDiff + c6o1*( VeloX-VeloY)))/(c1o1+q);
+ }
+
+ q = q_dirNW[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o54* (drho+c9o2*(-vx1+vx2 )*(-vx1+vx2 )-cu_sq);
+ (D.f[DIR_PM0])[kse]=(c1o1-q)/(c1o1+q)*(f_NW-f_SE+(f_NW+f_SE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_NW+f_SE)-c1o54*(rhoDiff + c6o1*(-VeloX+VeloY)))/(c1o1+q);
+ }
+
+ q = q_dirTE[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o54* (drho+c9o2*( vx1 +vx3)*( vx1 +vx3)-cu_sq);
+ (D.f[DIR_M0M])[kbw]=(c1o1-q)/(c1o1+q)*(f_TE-f_BW+(f_TE+f_BW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TE+f_BW)-c1o54*(rhoDiff + c6o1*( VeloX+VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirBW[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o54* (drho+c9o2*(-vx1 -vx3)*(-vx1 -vx3)-cu_sq);
+ (D.f[DIR_P0P])[kte]=(c1o1-q)/(c1o1+q)*(f_BW-f_TE+(f_BW+f_TE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BW+f_TE)-c1o54*(rhoDiff + c6o1*(-VeloX-VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirBE[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o54* (drho+c9o2*( vx1 -vx3)*( vx1 -vx3)-cu_sq);
+ (D.f[DIR_M0P])[ktw]=(c1o1-q)/(c1o1+q)*(f_BE-f_TW+(f_BE+f_TW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BE+f_TW)-c1o54*(rhoDiff + c6o1*( VeloX-VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirTW[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o54* (drho+c9o2*(-vx1 +vx3)*(-vx1 +vx3)-cu_sq);
+ (D.f[DIR_P0M])[kbe]=(c1o1-q)/(c1o1+q)*(f_TW-f_BE+(f_TW+f_BE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TW+f_BE)-c1o54*(rhoDiff + c6o1*(-VeloX+VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirTN[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o54* (drho+c9o2*( vx2+vx3)*( vx2+vx3)-cu_sq);
+ (D.f[DIR_0MM])[kbs]=(c1o1-q)/(c1o1+q)*(f_TN-f_BS+(f_TN+f_BS-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TN+f_BS)-c1o54*(rhoDiff + c6o1*( VeloY+VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirBS[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o54* (drho+c9o2*( -vx2-vx3)*( -vx2-vx3)-cu_sq);
+ (D.f[DIR_0PP])[ktn]=(c1o1-q)/(c1o1+q)*(f_BS-f_TN+(f_BS+f_TN-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BS+f_TN)-c1o54*(rhoDiff + c6o1*( -VeloY-VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirBN[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o54* (drho+c9o2*( vx2-vx3)*( vx2-vx3)-cu_sq);
+ (D.f[DIR_0MP])[kts]=(c1o1-q)/(c1o1+q)*(f_BN-f_TS+(f_BN+f_TS-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BN+f_TS)-c1o54*(rhoDiff + c6o1*( VeloY-VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirTS[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o54* (drho+c9o2*( -vx2+vx3)*( -vx2+vx3)-cu_sq);
+ (D.f[DIR_0PM])[kbn]=(c1o1-q)/(c1o1+q)*(f_TS-f_BN+(f_TS+f_BN-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TS+f_BN)-c1o54*(rhoDiff + c6o1*( -VeloY+VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirTNE[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o216*(drho+c9o2*( vx1+vx2+vx3)*( vx1+vx2+vx3)-cu_sq);
+ (D.f[DIR_MMM])[kbsw]=(c1o1-q)/(c1o1+q)*(f_TNE-f_BSW+(f_TNE+f_BSW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TNE+f_BSW)-c1o216*(rhoDiff + c6o1*( VeloX+VeloY+VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirBSW[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o216*(drho+c9o2*(-vx1-vx2-vx3)*(-vx1-vx2-vx3)-cu_sq);
+ (D.f[DIR_PPP])[ktne]=(c1o1-q)/(c1o1+q)*(f_BSW-f_TNE+(f_BSW+f_TNE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BSW+f_TNE)-c1o216*(rhoDiff + c6o1*(-VeloX-VeloY-VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirBNE[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o216*(drho+c9o2*( vx1+vx2-vx3)*( vx1+vx2-vx3)-cu_sq);
+ (D.f[DIR_MMP])[ktsw]=(c1o1-q)/(c1o1+q)*(f_BNE-f_TSW+(f_BNE+f_TSW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BNE+f_TSW)-c1o216*(rhoDiff + c6o1*( VeloX+VeloY-VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirTSW[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o216*(drho+c9o2*(-vx1-vx2+vx3)*(-vx1-vx2+vx3)-cu_sq);
+ (D.f[DIR_PPM])[kbne]=(c1o1-q)/(c1o1+q)*(f_TSW-f_BNE+(f_TSW+f_BNE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TSW+f_BNE)-c1o216*(rhoDiff + c6o1*(-VeloX-VeloY+VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirTSE[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o216*(drho+c9o2*( vx1-vx2+vx3)*( vx1-vx2+vx3)-cu_sq);
+ (D.f[DIR_MPM])[kbnw]=(c1o1-q)/(c1o1+q)*(f_TSE-f_BNW+(f_TSE+f_BNW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TSE+f_BNW)-c1o216*(rhoDiff + c6o1*( VeloX-VeloY+VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirBNW[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o216*(drho+c9o2*(-vx1+vx2-vx3)*(-vx1+vx2-vx3)-cu_sq);
+ (D.f[DIR_PMP])[ktse]=(c1o1-q)/(c1o1+q)*(f_BNW-f_TSE+(f_BNW+f_TSE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BNW+f_TSE)-c1o216*(rhoDiff + c6o1*(-VeloX+VeloY-VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirBSE[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o216*(drho+c9o2*( vx1-vx2-vx3)*( vx1-vx2-vx3)-cu_sq);
+ (D.f[DIR_MPP])[ktnw]=(c1o1-q)/(c1o1+q)*(f_BSE-f_TNW+(f_BSE+f_TNW-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_BSE+f_TNW)-c1o216*(rhoDiff + c6o1*( VeloX-VeloY-VeloZ)))/(c1o1+q);
+ }
+
+ q = q_dirTNW[k];
+ if (q>=c0o1 && q<=c1o1)
+ {
+ feq=c1o216*(drho+c9o2*(-vx1+vx2+vx3)*(-vx1+vx2+vx3)-cu_sq);
+ (D.f[DIR_PMM])[kbse]=(c1o1-q)/(c1o1+q)*(f_TNW-f_BSE+(f_TNW+f_BSE-c2o1*feq*om1)/(c1o1-om1))*c1o2+(q*(f_TNW+f_BSE)-c1o216*(rhoDiff + c6o1*(-VeloX+VeloY+VeloZ)))/(c1o1+q);
+ }
+ }
}
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
diff --git a/src/gpu/VirtualFluids_GPU/Kernel/Kernels/BasicKernels/FluidFlow/Compressible/CumulantK17/CumulantK17.cu b/src/gpu/VirtualFluids_GPU/Kernel/Kernels/BasicKernels/FluidFlow/Compressible/CumulantK17/CumulantK17.cu
index 70b0c4352afee850a4e17243979268bd126b7b4a..477539348706de7410319045fa075a6cdf31d01c 100644
--- a/src/gpu/VirtualFluids_GPU/Kernel/Kernels/BasicKernels/FluidFlow/Compressible/CumulantK17/CumulantK17.cu
+++ b/src/gpu/VirtualFluids_GPU/Kernel/Kernels/BasicKernels/FluidFlow/Compressible/CumulantK17/CumulantK17.cu
@@ -6,132 +6,132 @@
#include <cuda.h>
-template<TurbulenceModel turbulenceModel>
+template<TurbulenceModel turbulenceModel>
std::shared_ptr< CumulantK17<turbulenceModel> > CumulantK17<turbulenceModel>::getNewInstance(std::shared_ptr<Parameter> para, int level)
{
- return std::shared_ptr<CumulantK17<turbulenceModel> >(new CumulantK17<turbulenceModel>(para,level));
+ return std::shared_ptr<CumulantK17<turbulenceModel> >(new CumulantK17<turbulenceModel>(para,level));
}
template<TurbulenceModel turbulenceModel>
void CumulantK17<turbulenceModel>::run()
{
- LB_Kernel_CumulantK17 < turbulenceModel, false, false > <<< cudaGrid.grid, cudaGrid.threads >>>( para->getParD(level)->omega,
- para->getParD(level)->neighborX, para->getParD(level)->neighborY, para->getParD(level)->neighborZ,
- para->getParD(level)->distributions.f[0],
- para->getParD(level)->rho,
- para->getParD(level)->velocityX, para->getParD(level)->velocityY, para->getParD(level)->velocityZ,
- para->getParD(level)->turbViscosity,
- para->getSGSConstant(),
- (unsigned long)para->getParD(level)->numberOfNodes,
- level,
- para->getForcesDev(),
- para->getParD(level)->forceX_SP, para->getParD(level)->forceY_SP, para->getParD(level)->forceZ_SP,
- para->getQuadricLimitersDev(),
- para->getParD(level)->isEvenTimestep,
- para->getParD(level)->taggedFluidNodeIndices[CollisionTemplate::Default],
- para->getParD(level)->numberOfTaggedFluidNodes[CollisionTemplate::Default]);
+ LB_Kernel_CumulantK17 < turbulenceModel, false, false > <<< cudaGrid.grid, cudaGrid.threads >>>( para->getParD(level)->omega,
+ para->getParD(level)->neighborX, para->getParD(level)->neighborY, para->getParD(level)->neighborZ,
+ para->getParD(level)->distributions.f[0],
+ para->getParD(level)->rho,
+ para->getParD(level)->velocityX, para->getParD(level)->velocityY, para->getParD(level)->velocityZ,
+ para->getParD(level)->turbViscosity,
+ para->getSGSConstant(),
+ (unsigned long)para->getParD(level)->numberOfNodes,
+ level,
+ para->getForcesDev(),
+ para->getParD(level)->forceX_SP, para->getParD(level)->forceY_SP, para->getParD(level)->forceZ_SP,
+ para->getQuadricLimitersDev(),
+ para->getParD(level)->isEvenTimestep,
+ para->getParD(level)->taggedFluidNodeIndices[CollisionTemplate::Default],
+ para->getParD(level)->numberOfTaggedFluidNodes[CollisionTemplate::Default]);
- getLastCudaError("LB_Kernel_CumulantK17 execution failed");
+ getLastCudaError("LB_Kernel_CumulantK17 execution failed");
}
template<TurbulenceModel turbulenceModel>
void CumulantK17<turbulenceModel>::runOnIndices( const unsigned int *indices, unsigned int size_indices, CollisionTemplate collisionTemplate, CudaStreamIndex streamIndex )
{
- cudaStream_t stream = para->getStreamManager()->getStream(streamIndex);
-
- switch (collisionTemplate)
- {
- case CollisionTemplate::Default:
- LB_Kernel_CumulantK17 < turbulenceModel, false, false > <<< cudaGrid.grid, cudaGrid.threads, 0, stream >>>(para->getParD(level)->omega,
- para->getParD(level)->neighborX, para->getParD(level)->neighborY, para->getParD(level)->neighborZ,
- para->getParD(level)->distributions.f[0],
- para->getParD(level)->rho,
- para->getParD(level)->velocityX, para->getParD(level)->velocityY, para->getParD(level)->velocityZ,
- para->getParD(level)->turbViscosity,
- para->getSGSConstant(),
- (unsigned long)para->getParD(level)->numberOfNodes,
- level,
- para->getForcesDev(),
- para->getParD(level)->forceX_SP, para->getParD(level)->forceY_SP, para->getParD(level)->forceZ_SP,
- para->getQuadricLimitersDev(),
- para->getParD(level)->isEvenTimestep,
- indices,
- size_indices);
- break;
-
- case CollisionTemplate::WriteMacroVars:
- LB_Kernel_CumulantK17 < turbulenceModel, true, false > <<< cudaGrid.grid, cudaGrid.threads, 0, stream >>>( para->getParD(level)->omega,
- para->getParD(level)->neighborX, para->getParD(level)->neighborY, para->getParD(level)->neighborZ,
- para->getParD(level)->distributions.f[0],
- para->getParD(level)->rho,
- para->getParD(level)->velocityX, para->getParD(level)->velocityY, para->getParD(level)->velocityZ,
- para->getParD(level)->turbViscosity,
- para->getSGSConstant(),
- (unsigned long)para->getParD(level)->numberOfNodes,
- level,
- para->getForcesDev(),
- para->getParD(level)->forceX_SP, para->getParD(level)->forceY_SP, para->getParD(level)->forceZ_SP,
- para->getQuadricLimitersDev(),
- para->getParD(level)->isEvenTimestep,
- indices,
- size_indices);
- break;
-
- case CollisionTemplate::SubDomainBorder:
- case CollisionTemplate::AllFeatures:
- LB_Kernel_CumulantK17 < turbulenceModel, true, true > <<< cudaGrid.grid, cudaGrid.threads, 0, stream >>>( para->getParD(level)->omega,
- para->getParD(level)->neighborX, para->getParD(level)->neighborY, para->getParD(level)->neighborZ,
- para->getParD(level)->distributions.f[0],
- para->getParD(level)->rho,
- para->getParD(level)->velocityX, para->getParD(level)->velocityY, para->getParD(level)->velocityZ,
- para->getParD(level)->turbViscosity,
- para->getSGSConstant(),
- (unsigned long)para->getParD(level)->numberOfNodes,
- level,
- para->getForcesDev(),
- para->getParD(level)->forceX_SP, para->getParD(level)->forceY_SP, para->getParD(level)->forceZ_SP,
- para->getQuadricLimitersDev(),
- para->getParD(level)->isEvenTimestep,
- indices,
- size_indices);
- break; case CollisionTemplate::ApplyBodyForce:
- LB_Kernel_CumulantK17 < turbulenceModel, false, true > <<< cudaGrid.grid, cudaGrid.threads, 0, stream >>>( para->getParD(level)->omega,
- para->getParD(level)->neighborX, para->getParD(level)->neighborY, para->getParD(level)->neighborZ,
- para->getParD(level)->distributions.f[0],
- para->getParD(level)->rho,
- para->getParD(level)->velocityX, para->getParD(level)->velocityY, para->getParD(level)->velocityZ,
- para->getParD(level)->turbViscosity,
- para->getSGSConstant(),
- (unsigned long)para->getParD(level)->numberOfNodes,
- level,
- para->getForcesDev(),
- para->getParD(level)->forceX_SP, para->getParD(level)->forceY_SP, para->getParD(level)->forceZ_SP,
- para->getQuadricLimitersDev(),
- para->getParD(level)->isEvenTimestep,
- indices,
- size_indices);
- break; default:
- throw std::runtime_error("Invalid CollisionTemplate in CumulantK17::runOnIndices()");
- break;
- }
+ cudaStream_t stream = para->getStreamManager()->getStream(streamIndex);
+
+ switch (collisionTemplate)
+ {
+ case CollisionTemplate::Default:
+ LB_Kernel_CumulantK17 < turbulenceModel, false, false > <<< cudaGrid.grid, cudaGrid.threads, 0, stream >>>(para->getParD(level)->omega,
+ para->getParD(level)->neighborX, para->getParD(level)->neighborY, para->getParD(level)->neighborZ,
+ para->getParD(level)->distributions.f[0],
+ para->getParD(level)->rho,
+ para->getParD(level)->velocityX, para->getParD(level)->velocityY, para->getParD(level)->velocityZ,
+ para->getParD(level)->turbViscosity,
+ para->getSGSConstant(),
+ (unsigned long)para->getParD(level)->numberOfNodes,
+ level,
+ para->getForcesDev(),
+ para->getParD(level)->forceX_SP, para->getParD(level)->forceY_SP, para->getParD(level)->forceZ_SP,
+ para->getQuadricLimitersDev(),
+ para->getParD(level)->isEvenTimestep,
+ indices,
+ size_indices);
+ break;
+
+ case CollisionTemplate::WriteMacroVars:
+ LB_Kernel_CumulantK17 < turbulenceModel, true, false > <<< cudaGrid.grid, cudaGrid.threads, 0, stream >>>( para->getParD(level)->omega,
+ para->getParD(level)->neighborX, para->getParD(level)->neighborY, para->getParD(level)->neighborZ,
+ para->getParD(level)->distributions.f[0],
+ para->getParD(level)->rho,
+ para->getParD(level)->velocityX, para->getParD(level)->velocityY, para->getParD(level)->velocityZ,
+ para->getParD(level)->turbViscosity,
+ para->getSGSConstant(),
+ (unsigned long)para->getParD(level)->numberOfNodes,
+ level,
+ para->getForcesDev(),
+ para->getParD(level)->forceX_SP, para->getParD(level)->forceY_SP, para->getParD(level)->forceZ_SP,
+ para->getQuadricLimitersDev(),
+ para->getParD(level)->isEvenTimestep,
+ indices,
+ size_indices);
+ break;
- getLastCudaError("LB_Kernel_CumulantK17 execution failed");
+ case CollisionTemplate::SubDomainBorder:
+ case CollisionTemplate::AllFeatures:
+ LB_Kernel_CumulantK17 < turbulenceModel, true, true > <<< cudaGrid.grid, cudaGrid.threads, 0, stream >>>( para->getParD(level)->omega,
+ para->getParD(level)->neighborX, para->getParD(level)->neighborY, para->getParD(level)->neighborZ,
+ para->getParD(level)->distributions.f[0],
+ para->getParD(level)->rho,
+ para->getParD(level)->velocityX, para->getParD(level)->velocityY, para->getParD(level)->velocityZ,
+ para->getParD(level)->turbViscosity,
+ para->getSGSConstant(),
+ (unsigned long)para->getParD(level)->numberOfNodes,
+ level,
+ para->getForcesDev(),
+ para->getParD(level)->forceX_SP, para->getParD(level)->forceY_SP, para->getParD(level)->forceZ_SP,
+ para->getQuadricLimitersDev(),
+ para->getParD(level)->isEvenTimestep,
+ indices,
+ size_indices);
+ break; case CollisionTemplate::ApplyBodyForce:
+ LB_Kernel_CumulantK17 < turbulenceModel, false, true > <<< cudaGrid.grid, cudaGrid.threads, 0, stream >>>( para->getParD(level)->omega,
+ para->getParD(level)->neighborX, para->getParD(level)->neighborY, para->getParD(level)->neighborZ,
+ para->getParD(level)->distributions.f[0],
+ para->getParD(level)->rho,
+ para->getParD(level)->velocityX, para->getParD(level)->velocityY, para->getParD(level)->velocityZ,
+ para->getParD(level)->turbViscosity,
+ para->getSGSConstant(),
+ (unsigned long)para->getParD(level)->numberOfNodes,
+ level,
+ para->getForcesDev(),
+ para->getParD(level)->forceX_SP, para->getParD(level)->forceY_SP, para->getParD(level)->forceZ_SP,
+ para->getQuadricLimitersDev(),
+ para->getParD(level)->isEvenTimestep,
+ indices,
+ size_indices);
+ break; default:
+ throw std::runtime_error("Invalid CollisionTemplate in CumulantK17::runOnIndices()");
+ break;
+ }
+
+ getLastCudaError("LB_Kernel_CumulantK17 execution failed");
}
template<TurbulenceModel turbulenceModel>
CumulantK17<turbulenceModel>::CumulantK17(std::shared_ptr<Parameter> para, int level)
{
- this->para = para;
- this->level = level;
+ this->para = para;
+ this->level = level;
+
+ myPreProcessorTypes.push_back(InitCompSP27);
- myPreProcessorTypes.push_back(InitCompSP27);
+ myKernelGroup = BasicKernel;
- myKernelGroup = BasicKernel;
+ this->cudaGrid = vf::cuda::CudaGrid(para->getParD(level)->numberofthreads, para->getParD(level)->numberOfNodes);
+ this->kernelUsesFluidNodeIndices = true;
- this->cudaGrid = vf::cuda::CudaGrid(para->getParD(level)->numberofthreads, para->getParD(level)->numberOfNodes);
- this->kernelUsesFluidNodeIndices = true;
-
- VF_LOG_INFO("Using turbulence model: {}", turbulenceModel);
+ VF_LOG_INFO("Using turbulence model: {}", turbulenceModel);
}
template class CumulantK17<TurbulenceModel::AMD>;
diff --git a/src/gpu/VirtualFluids_GPU/Kernel/Kernels/BasicKernels/FluidFlow/Compressible/CumulantK17/CumulantK17_Device.cu b/src/gpu/VirtualFluids_GPU/Kernel/Kernels/BasicKernels/FluidFlow/Compressible/CumulantK17/CumulantK17_Device.cu
index 34a444230019a1c6cfacb18e12fd73607fb2e09a..77e5172dae7b0ff6b51ed79a0a4356c7461801c5 100644
--- a/src/gpu/VirtualFluids_GPU/Kernel/Kernels/BasicKernels/FluidFlow/Compressible/CumulantK17/CumulantK17_Device.cu
+++ b/src/gpu/VirtualFluids_GPU/Kernel/Kernels/BasicKernels/FluidFlow/Compressible/CumulantK17/CumulantK17_Device.cu
@@ -1,28 +1,28 @@
-// ____ ____ __ ______ __________ __ __ __ __
-// \ \ | | | | | _ \ |___ ___| | | | | / \ | |
-// \ \ | | | | | |_) | | | | | | | / \ | |
-// \ \ | | | | | _ / | | | | | | / /\ \ | |
-// \ \ | | | | | | \ \ | | | \__/ | / ____ \ | |____
-// \ \ | | |__| |__| \__\ |__| \________/ /__/ \__\ |_______|
-// \ \ | | ________________________________________________________________
-// \ \ | | | ______________________________________________________________|
-// \ \| | | | __ __ __ __ ______ _______
-// \ | | |_____ | | | | | | | | | _ \ / _____)
-// \ | | _____| | | | | | | | | | | \ \ \_______
+// ____ ____ __ ______ __________ __ __ __ __
+// \ \ | | | | | _ \ |___ ___| | | | | / \ | |
+// \ \ | | | | | |_) | | | | | | | / \ | |
+// \ \ | | | | | _ / | | | | | | / /\ \ | |
+// \ \ | | | | | | \ \ | | | \__/ | / ____ \ | |____
+// \ \ | | |__| |__| \__\ |__| \________/ /__/ \__\ |_______|
+// \ \ | | ________________________________________________________________
+// \ \ | | | ______________________________________________________________|
+// \ \| | | | __ __ __ __ ______ _______
+// \ | | |_____ | | | | | | | | | _ \ / _____)
+// \ | | _____| | | | | | | | | | | \ \ \_______
// \ | | | | |_____ | \_/ | | | | |_/ / _____ |
-// \ _____| |__| |________| \_______/ |__| |______/ (_______/
+// \ _____| |__| |________| \_______/ |__| |______/ (_______/
//
-// This file is part of VirtualFluids. VirtualFluids is free software: you can
+// This file is part of VirtualFluids. VirtualFluids is free software: you can
// redistribute it and/or modify it under the terms of the GNU General Public
-// License as published by the Free Software Foundation, either version 3 of
+// License as published by the Free Software Foundation, either version 3 of
// the License, or (at your option) any later version.
-//
-// VirtualFluids is distributed in the hope that it will be useful, but WITHOUT
-// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+//
+// VirtualFluids is distributed in the hope that it will be useful, but WITHOUT
+// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
// for more details.
-//
+//
// You should have received a copy of the GNU General Public License along
// with VirtualFluids (see COPYING.txt). If not, see <http://www.gnu.org/licenses/>.
//
@@ -32,17 +32,17 @@
//! \brief Kernel for CumulantK17 including different turbulence models and options for local body forces and writing macroscopic variables
//!
//! CumulantK17 kernel using chimera transformations and quartic limiters as present in Geier et al. (2017). Additional options are three different
-//! eddy-viscosity turbulence models (Smagorinsky, AMD, QR) that can be set via the template parameter turbulenceModel (with default
-//! TurbulenceModel::None).
+//! eddy-viscosity turbulence models (Smagorinsky, AMD, QR) that can be set via the template parameter turbulenceModel (with default
+//! TurbulenceModel::None).
//! The kernel is executed separately for each subset of fluid node indices with a different tag CollisionTemplate. For each subset, only the locally
-//! required options are switched on ( \param writeMacroscopicVariables and/or \param applyBodyForce) in order to minimize memory accesses. The default
-//! refers to the plain cumlant kernel (CollisionTemplate::Default).
-//! Nodes are added to subsets (taggedFluidNodes) in Simulation::init using a corresponding tag with different values of CollisionTemplate. These subsets
+//! required options are switched on ( \param writeMacroscopicVariables and/or \param applyBodyForce) in order to minimize memory accesses. The default
+//! refers to the plain cumlant kernel (CollisionTemplate::Default).
+//! Nodes are added to subsets (taggedFluidNodes) in Simulation::init using a corresponding tag with different values of CollisionTemplate. These subsets
//! are provided by the utilized PostCollisionInteractiors depending on they specifc requirements (e.g. writeMacroscopicVariables for probes).
//=======================================================================================
/* Device code */
-#include "LBM/LB.h"
+#include "LBM/LB.h"
#include "lbm/constants/D3Q27.h"
#include <lbm/constants/NumericConstants.h>
#include "Kernel/Utilities/DistributionHelper.cuh"
@@ -57,25 +57,25 @@ using namespace vf::lbm::dir;
////////////////////////////////////////////////////////////////////////////////
template<TurbulenceModel turbulenceModel, bool writeMacroscopicVariables, bool applyBodyForce>
__global__ void LB_Kernel_CumulantK17(
- real omega_in,
- uint* neighborX,
- uint* neighborY,
- uint* neighborZ,
- real* distributions,
+ real omega_in,
+ uint* neighborX,
+ uint* neighborY,
+ uint* neighborZ,
+ real* distributions,
real* rho,
real* vx,
real* vy,
real* vz,
real* turbulentViscosity,
real SGSconstant,
- unsigned long numberOfLBnodes,
- int level,
- real* forces,
+ unsigned long numberOfLBnodes,
+ int level,
+ real* forces,
real* bodyForceX,
real* bodyForceY,
real* bodyForceZ,
- real* quadricLimiters,
- bool isEvenTimestep,
+ real* quadricLimiters,
+ bool isEvenTimestep,
const uint *fluidNodeIndices,
uint numberOfFluidNodes)
{
@@ -91,10 +91,10 @@ __global__ void LB_Kernel_CumulantK17(
//! - Get node index coordinates from threadIdx, blockIdx, blockDim and gridDim.
//!
const unsigned kThread = vf::gpu::getNodeIndex();
-
+
//////////////////////////////////////////////////////////////////////////
// run for all indices in size_Mat and fluid nodes
- if (kThread >= numberOfFluidNodes)
+ if (kThread >= numberOfFluidNodes)
return;
////////////////////////////////////////////////////////////////////////////////
//! - Get the node index from the array containing all indices of fluid nodes
@@ -202,7 +202,7 @@ __global__ void LB_Kernel_CumulantK17(
real vvz = ((((f_PPP - f_MMM) + (f_PMP - f_MPM)) + ((f_MPP - f_PMM) + (f_MMP - f_PPM))) +
(((f_0MP - f_0PM) + (f_0PP - f_0MM)) + ((f_M0P - f_P0M) + (f_P0P - f_M0M))) + (f_00P - f_00M)) *
oneOverRho;
-
+
////////////////////////////////////////////////////////////////////////////////////
//! - Add half of the acceleration (body force) to the velocity as in Eq. (42) \ref
//! <a href="https://doi.org/10.1016/j.camwa.2015.05.001"><b>[ M. Geier et al. (2015),
@@ -212,13 +212,13 @@ __global__ void LB_Kernel_CumulantK17(
for (size_t i = 1; i <= level; i++) {
factor *= c2o1;
}
-
+
real fx = forces[0];
real fy = forces[1];
real fz = forces[2];
if( applyBodyForce ){
- fx += bodyForceX[k_000];
+ fx += bodyForceX[k_000];
fy += bodyForceY[k_000];
fz += bodyForceZ[k_000];
@@ -232,7 +232,7 @@ __global__ void LB_Kernel_CumulantK17(
vvx += acc_x;
vvy += acc_y;
vvz += acc_z;
-
+
// Reset body force. To be used when not using round-off correction.
bodyForceX[k_000] = 0.0f;
bodyForceY[k_000] = 0.0f;
@@ -243,9 +243,9 @@ __global__ void LB_Kernel_CumulantK17(
//!
//!> Similar to Kahan summation algorithm (https://en.wikipedia.org/wiki/Kahan_summation_algorithm)
//!> Essentially computes the round-off error of the applied force and adds it in the next time step as a compensation.
- //!> Seems to be necesseary at very high Re boundary layers, where the forcing and velocity can
+ //!> Seems to be necesseary at very high Re boundary layers, where the forcing and velocity can
//!> differ by several orders of magnitude.
- //!> \note 16/05/2022: Testing, still ongoing!
+ //!> \note 16/05/2022: Testing, still ongoing!
//!
// bodyForceX[k_000] = (acc_x-(vvx-vx))*factor*c2o1;
// bodyForceY[k_000] = (acc_y-(vvy-vy))*factor*c2o1;
@@ -256,7 +256,7 @@ __global__ void LB_Kernel_CumulantK17(
vvy += fy * c1o2 / factor;
vvz += fz * c1o2 / factor;
}
-
+
////////////////////////////////////////////////////////////////////////////////////
// calculate the square of velocities for this lattice node
@@ -723,4 +723,4 @@ template __global__ void LB_Kernel_CumulantK17 < TurbulenceModel::Smagorinsky, f
template __global__ void LB_Kernel_CumulantK17 < TurbulenceModel::QR, false, false > ( real omega_in, uint* neighborX, uint* neighborY, uint* neighborZ, real* distributions, real* rho, real* vx, real* vy, real* vz, real* turbulentViscosity, real SGSconstant, unsigned long size_Mat, int level, real* forces, real* bodyForceX, real* bodyForceY, real* bodyForceZ, real* quadricLimiters, bool isEvenTimestep, const uint *fluidNodeIndices, uint numberOfFluidNodes);
-template __global__ void LB_Kernel_CumulantK17 < TurbulenceModel::None, false, false > ( real omega_in, uint* neighborX, uint* neighborY, uint* neighborZ, real* distributions, real* rho, real* vx, real* vy, real* vz, real* turbulentViscosity, real SGSconstant, unsigned long size_Mat, int level, real* forces, real* bodyForceX, real* bodyForceY, real* bodyForceZ, real* quadricLimiters, bool isEvenTimestep, const uint *fluidNodeIndices, uint numberOfFluidNodes);
\ No newline at end of file
+template __global__ void LB_Kernel_CumulantK17 < TurbulenceModel::None, false, false > ( real omega_in, uint* neighborX, uint* neighborY, uint* neighborZ, real* distributions, real* rho, real* vx, real* vy, real* vz, real* turbulentViscosity, real SGSconstant, unsigned long size_Mat, int level, real* forces, real* bodyForceX, real* bodyForceY, real* bodyForceZ, real* quadricLimiters, bool isEvenTimestep, const uint *fluidNodeIndices, uint numberOfFluidNodes);
diff --git a/src/gpu/VirtualFluids_GPU/Kernel/Kernels/BasicKernels/FluidFlow/Compressible/CumulantK17/CumulantK17_Device.cuh b/src/gpu/VirtualFluids_GPU/Kernel/Kernels/BasicKernels/FluidFlow/Compressible/CumulantK17/CumulantK17_Device.cuh
index b8cc9543e9b531c5aa90cb2961416a6cbc52377d..55c22def9c43ab2678fc808043859f43021270a5 100644
--- a/src/gpu/VirtualFluids_GPU/Kernel/Kernels/BasicKernels/FluidFlow/Compressible/CumulantK17/CumulantK17_Device.cuh
+++ b/src/gpu/VirtualFluids_GPU/Kernel/Kernels/BasicKernels/FluidFlow/Compressible/CumulantK17/CumulantK17_Device.cuh
@@ -5,25 +5,25 @@
#include <curand.h>
template< TurbulenceModel turbulenceModel, bool writeMacroscopicVariables, bool applyBodyForce > __global__ void LB_Kernel_CumulantK17(
- real omega_in,
- uint* neighborX,
- uint* neighborY,
- uint* neighborZ,
- real* distributions,
- real* rho,
- real* vx,
+ real omega_in,
+ uint* neighborX,
+ uint* neighborY,
+ uint* neighborZ,
+ real* distributions,
+ real* rho,
+ real* vx,
real* vy,
real* vz,
- real* turbulentViscosity,
- real SGSconstant,
- unsigned long numberOfLBnodes,
- int level,
- real* forces,
- real* bodyForceX,
- real* bodyForceY,
- real* bodyForceZ,
- real* quadricLimiters,
- bool isEvenTimestep,
- const uint *fluidNodeIndices,
+ real* turbulentViscosity,
+ real SGSconstant,
+ unsigned long numberOfLBnodes,
+ int level,
+ real* forces,
+ real* bodyForceX,
+ real* bodyForceY,
+ real* bodyForceZ,
+ real* quadricLimiters,
+ bool isEvenTimestep,
+ const uint *fluidNodeIndices,
uint numberOfFluidNodes);
#endif
diff --git a/src/gpu/VirtualFluids_GPU/KernelManager/BCKernelManager.cpp b/src/gpu/VirtualFluids_GPU/KernelManager/BCKernelManager.cpp
index 1310cf09e5ce05270e88cd3a53e4910816bd7628..e8fc3f318c920be36be7861a28659124a7b1e977 100644
--- a/src/gpu/VirtualFluids_GPU/KernelManager/BCKernelManager.cpp
+++ b/src/gpu/VirtualFluids_GPU/KernelManager/BCKernelManager.cpp
@@ -428,4 +428,4 @@ void BCKernelManager::runPrecursorBCKernelPost(int level, uint t, CudaMemoryMana
real tRatio = real(t_level-lastTime)/para->getParD(level)->precursorBC.timeStepsBetweenReads;
precursorBoundaryConditionPost(para->getParD(level).get(), ¶->getParD(level)->precursorBC, tRatio, para->getVelocityRatio());
-}
\ No newline at end of file
+}
diff --git a/src/gpu/VirtualFluids_GPU/LBM/LB.h b/src/gpu/VirtualFluids_GPU/LBM/LB.h
index a7517ca0f07b9c1379501aa3ff750637d230a7bd..cfdbbbae040a13f94e97d40d702b93d5a1e19c86 100644
--- a/src/gpu/VirtualFluids_GPU/LBM/LB.h
+++ b/src/gpu/VirtualFluids_GPU/LBM/LB.h
@@ -15,9 +15,9 @@
//////////////////////////
//porous media
-#define GEO_PM_0 5
-#define GEO_PM_1 6
-#define GEO_PM_2 7
+#define GEO_PM_0 5
+#define GEO_PM_1 6
+#define GEO_PM_2 7
//////////////////////////
#define GEO_SOLID 15
@@ -191,7 +191,7 @@ typedef struct OffFC{
// Distribution functions g 6
typedef struct Distri6 {
- real* g[6];
+ real* g[6];
} Distributions6;
// Distribution functions f 7
@@ -284,55 +284,53 @@ typedef struct WMparas{
//measurePoints
typedef struct MeasP{
- std::string name;
- uint k;
- std::vector<real> Vx;
- std::vector<real> Vy;
- std::vector<real> Vz;
- std::vector<real> Rho;
- //real* Vx;
- //real* Vy;
- //real* Vz;
- //real* Rho;
+ std::string name;
+ uint k;
+ std::vector<real> Vx;
+ std::vector<real> Vy;
+ std::vector<real> Vz;
+ std::vector<real> Rho;
+ //real* Vx;
+ //real* Vy;
+ //real* Vz;
+ //real* Rho;
}MeasurePoints;
//Process Neighbors
typedef struct PN27{
- real* f[27];
- uint memsizeFs;
- int* index;
- uint memsizeIndex;
- uint rankNeighbor;
- int numberOfNodes;
- int numberOfFs;
+ real* f[27];
+ uint memsizeFs;
+ int* index;
+ uint memsizeIndex;
+ uint rankNeighbor;
+ int numberOfNodes;
+ int numberOfFs;
}ProcessNeighbor27;
typedef struct PN_F3 {
- real* g[6];
- uint memsizeGs;
- int* index;
- uint memsizeIndex;
- uint rankNeighbor;
- int numberOfNodes;
- int numberOfGs;
+ real* g[6];
+ uint memsizeGs;
+ int* index;
+ uint memsizeIndex;
+ uint rankNeighbor;
+ int numberOfNodes;
+ int numberOfGs;
}ProcessNeighborF3;
//path line particles
typedef struct PLP{
- bool *stuck, *hot;
- real *coordXabsolut, *coordYabsolut, *coordZabsolut;
- real *coordXlocal, *coordYlocal, *coordZlocal;
- real *veloX, *veloY, *veloZ;
- real *randomLocationInit;
- uint *timestep;
- uint *ID;
- uint *cellBaseID;
- uint numberOfParticles, numberOfTimestepsParticles;
- uint memSizeID, memSizeTimestep, memSizerealAll, memSizereal, memSizeBool, memSizeBoolBC;
+ bool *stuck, *hot;
+ real *coordXabsolut, *coordYabsolut, *coordZabsolut;
+ real *coordXlocal, *coordYlocal, *coordZlocal;
+ real *veloX, *veloY, *veloZ;
+ real *randomLocationInit;
+ uint *timestep;
+ uint *ID;
+ uint *cellBaseID;
+ uint numberOfParticles, numberOfTimestepsParticles;
+ uint memSizeID, memSizeTimestep, memSizerealAll, memSizereal, memSizeBool, memSizeBoolBC;
}PathLineParticles;
-
-
//////////////////////////////////////////////////////////////////////////
inline int vectorPosition(int i, int j, int k, int Lx, int Ly )
{
@@ -341,7 +339,4 @@ inline int vectorPosition(int i, int j, int k, int Lx, int Ly )
}
//////////////////////////////////////////////////////////////////////////
-
#endif
-
-
diff --git a/src/gpu/VirtualFluids_GPU/Output/DistributionDebugInspector.h b/src/gpu/VirtualFluids_GPU/Output/DistributionDebugInspector.h
index 53e030c0d16116a4edef0135f9ab435c853fd66b..95fea46d4eba0c2f2ff0846d22ee5da4f6c357ea 100644
--- a/src/gpu/VirtualFluids_GPU/Output/DistributionDebugInspector.h
+++ b/src/gpu/VirtualFluids_GPU/Output/DistributionDebugInspector.h
@@ -1,28 +1,28 @@
//=======================================================================================
-// ____ ____ __ ______ __________ __ __ __ __
-// \ \ | | | | | _ \ |___ ___| | | | | / \ | |
-// \ \ | | | | | |_) | | | | | | | / \ | |
-// \ \ | | | | | _ / | | | | | | / /\ \ | |
-// \ \ | | | | | | \ \ | | | \__/ | / ____ \ | |____
-// \ \ | | |__| |__| \__\ |__| \________/ /__/ \__\ |_______|
-// \ \ | | ________________________________________________________________
-// \ \ | | | ______________________________________________________________|
-// \ \| | | | __ __ __ __ ______ _______
-// \ | | |_____ | | | | | | | | | _ \ / _____)
-// \ | | _____| | | | | | | | | | | \ \ \_______
+// ____ ____ __ ______ __________ __ __ __ __
+// \ \ | | | | | _ \ |___ ___| | | | | / \ | |
+// \ \ | | | | | |_) | | | | | | | / \ | |
+// \ \ | | | | | _ / | | | | | | / /\ \ | |
+// \ \ | | | | | | \ \ | | | \__/ | / ____ \ | |____
+// \ \ | | |__| |__| \__\ |__| \________/ /__/ \__\ |_______|
+// \ \ | | ________________________________________________________________
+// \ \ | | | ______________________________________________________________|
+// \ \| | | | __ __ __ __ ______ _______
+// \ | | |_____ | | | | | | | | | _ \ / _____)
+// \ | | _____| | | | | | | | | | | \ \ \_______
// \ | | | | |_____ | \_/ | | | | |_/ / _____ |
-// \ _____| |__| |________| \_______/ |__| |______/ (_______/
+// \ _____| |__| |________| \_______/ |__| |______/ (_______/
//
-// This file is part of VirtualFluids. VirtualFluids is free software: you can
+// This file is part of VirtualFluids. VirtualFluids is free software: you can
// redistribute it and/or modify it under the terms of the GNU General Public
-// License as published by the Free Software Foundation, either version 3 of
+// License as published by the Free Software Foundation, either version 3 of
// the License, or (at your option) any later version.
-//
-// VirtualFluids is distributed in the hope that it will be useful, but WITHOUT
-// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+//
+// VirtualFluids is distributed in the hope that it will be useful, but WITHOUT
+// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+// FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
// for more details.
-//
+//
// You should have received a copy of the GNU General Public License along
// with VirtualFluids (see COPYING.txt). If not, see <http://www.gnu.org/licenses/>.
//
@@ -45,7 +45,7 @@
class DistributionDebugInspector
{
public:
- DistributionDebugInspector(uint _inspectionLevel, real _minX, real _maxX, real _minY, real _maxY, real _minZ, real _maxZ, std::string _tag):
+ DistributionDebugInspector(uint _inspectionLevel, real _minX, real _maxX, real _minY, real _maxY, real _minZ, real _maxZ, std::string _tag):
inspectionLevel(_inspectionLevel),
minX(_minX),
maxX(_maxX),
@@ -55,8 +55,8 @@ public:
maxZ(_maxZ),
tag(_tag)
{};
-
- ~DistributionDebugInspector(){}
+
+ ~DistributionDebugInspector() = default;
void inspect(std::shared_ptr<Parameter> para, uint level, uint t);
@@ -73,4 +73,4 @@ std::string tag;
};
-#endif
\ No newline at end of file
+#endif
diff --git a/src/gpu/VirtualFluids_GPU/PreCollisionInteractor/ActuatorFarm.h b/src/gpu/VirtualFluids_GPU/PreCollisionInteractor/ActuatorFarm.h
index f771ce10cdfe81237ae1caa8a26f887488839aa4..8e21cdb6b21efd323f6723e21d6b28614109f1ec 100644
--- a/src/gpu/VirtualFluids_GPU/PreCollisionInteractor/ActuatorFarm.h
+++ b/src/gpu/VirtualFluids_GPU/PreCollisionInteractor/ActuatorFarm.h
@@ -44,10 +44,7 @@ public:
throw std::runtime_error("ActuatorFarm::ActuatorFarm: epsilon needs to be larger than dx!");
}
- virtual ~ActuatorFarm()
- {
-
- }
+ ~ActuatorFarm() override = default;
void addTurbine(real turbinePosX, real turbinePosY, real turbinePosZ, real diameter, real omega, real azimuth, real yaw, std::vector<real> bladeRadii);
void init(Parameter* para, GridProvider* gridProvider, CudaMemoryManager* cudaManager) override;
void interact(Parameter* para, CudaMemoryManager* cudaManager, int level, uint t) override;
@@ -197,4 +194,4 @@ private:
int streamIndex;
};
-#endif
\ No newline at end of file
+#endif
diff --git a/src/gpu/VirtualFluids_GPU/PreCollisionInteractor/PrecursorWriter.h b/src/gpu/VirtualFluids_GPU/PreCollisionInteractor/PrecursorWriter.h
index 3959fb2d8f5e719084d2d55e6f6ef84076383b45..3bae63a339255f3f72196e20096f6019cdd7748d 100644
--- a/src/gpu/VirtualFluids_GPU/PreCollisionInteractor/PrecursorWriter.h
+++ b/src/gpu/VirtualFluids_GPU/PreCollisionInteractor/PrecursorWriter.h
@@ -158,4 +158,4 @@ private:
uint writePrecision = 8;
};
-#endif //PRECURSORPROBE_H_
\ No newline at end of file
+#endif //PRECURSORPROBE_H_