Procedures

Initialize axis

Add a new region to bc_set

Build an integer field, where cells=0 denotes interior cells, and cells=1 denotes boundary cells for input variable

Check whether there is BC for a given variable on a given region

Make sure bounds represent a plane

Resize array to accomodate a new element

Finalizes bc_set

Fetches a pointer to the val array describing the Dirichlet or Neumann BC of a given variable on a given region.

Returns the extents (lo and hi bounds) of a region.

Returns the index of a region, or -1 if not found.

Print to stdout information on bc_set, for debugging

Initializes bc_set

Read boundary conditions from file

Set boundary condition of a given type, for a given variable on a given region.

Imposes boundary conditions for a given variable.

Update ghostboundaries to enforce Dirichlet BC

Update ghostboundaries to enforce Neumann BC

Finds the intersection between block owned by this MPI rank, and the plane defining the region

Write bc_set to disk using HDF5. The file structure follows this convention: / (root) !-- Time !-- Iter !-- Region 1 !-- xlo !-- xhi !-- dir !-- side |-- Var 1 |-- Type |-- Values(:,:,:) !-- Var 2 . . !-- Region 2 . .

Finalize the block object

Returns the coordinates of the cell that contains the position p.

Print to stdout information about this block

Initialize block object

[DEPRECATED] Initialize block object

Return block ID and rank of the block where the point is located using a binary search alogirthm. Note that this function assumes that the point is within the domain, i.e., (pmin <= p <= pmax) and that any treatment for periodicity has been previously applied.

Partition a parent block into sub-blocks based on a given decomposition Nb(3) for parallel simulations. This will also define the partition axes, and update local bounds/extents, and global bounds/extents.

Read block data using HDF5

Associate the convenience pointers

Set block periodicity in each direction

Define MPI derived type for communicating ghostcells

Initialize a uniform grid on this block

Computes the bounds of the sub-block form the bounds of the parent block. Each sub-block gets about the same number of grid points in each direction.

Updates the dimensional extents of the block

Updates the ghostcell values of local grid owned by the current MPI rank. Note that each MPI rank stores only its portion of the grid, thus needs to have proper ghostcell values.

[DEPRECATED] Updates the ghostcell values of local grid owned by the current MPI rank. Note that each MPI rank stores only its portion of the grid, thus needs to have proper ghostcell values. Although SetupUniformGrid fills the ghostcells of x/y/z, it does it assuming fixed grid spacing, which may not be the correct if a non-uniform grid is used.

Update mid points

Update grid spacing arrays

Write block data using HDF5

Compute Momentum operator RHS for velocity component in x1-direction

Compute Momentum operator RHS for velocity component in x2-direction

Compute Momentum operator RHS for velocity component in x3-direction

Compute bodyforces based on old state

Perform corrector step: compute pressure at n+1 and divergence-free velocity at n+1.

Perform intermediate IB step

Perform predictor step: compute intermediate velocity.

Advance Resolved Particle centroids and markers to n+1

Initialize the solver

Initialize boundary conditions

Initialize and partition grid

Initialize differential operators on current grid.

Initialize fields used by this solver and read initial conditions.

Initialize monitors used by this solver.

Initialize differential operators on current grid.

Generate the divergence operator for the velocity field, and adjust at the boundaries.

Generate the gradient operator for the pressure field, and adjust at the boundaries.

Generate the pressure Laplacian operator

Generate the VF Laplacian operator

Generate the viscous Laplacian operator and adjust at the boundaries.

Initialize the visualization outputs

Compute collisions between resolved Particles and Immersed Boundaries

Get command line options

Initializes cblock to handle collisions. This extra block is expected to be coarser than the simulation block, but larger than the maximum object size. It is used to expedite neighbor searches.

Initializes cblock to handle collisions. This extra block is expected to be coarser than the simulation block, but larger than the maximum object size. It is used to expedite neighbor searches.

Get information about GPU

Addition

Addition

Update and add-up the ghostcells

Add up ghostcells in the x direction with non-blocking mpi directives

Add up ghostcells in the y direction with non-blocking mpi directives

Add up ghostcells in the z direction with non-blocking mpi directives

Allocate an Eulerian object

Assignment for Eulerian_obj

Assignment for Eulerian_obj

Assignment for Eulerian_obj

Deallocate an Eulerian object

Finalize the Eulerian object

Print info about this structure

Initialize an Eulerian field

Compute the mean of an Eulerian_obj

Multiplication

Multiplication

Compute norm2 of an Eulerian_obj

Addition

Addition

Update the ghostcells

Update the ghostcells in the x direction with non-blocking mpi directives

Update the ghostcells in the y direction with non-blocking mpi directives

Update the ghostcells in the z direction with non-blocking mpi directives

Add a new element to a collection of Eulerian objects

Finalize structure

Returns the index of an Eulerian_obj contained in this%fields given its name.

Return the base name of file to write

Return the base name of file to write

Print info about this collection of eulerian objects

Initialize a collection of Eulerian objects

Read Eulerian data

Read all Eulerian objects in file

Read Eulerian data using MPI binary file tools

Read one Eulerian object based on name

Set file overwritting

Set the base name of file to read

Set the base name of file to write

Write Eulerian data

Write Eulerian data using SILO

Write a single Eulerian object to file

Write a single Eulerian objects to file using SILO

Filtering kernel with support from -1.0 to 1.0 Box filter (step function)

Filtering kernel with support from -1.0 to 1.0 Parabolic filter

Filtering kernel with support from -1.0 to 1.0 Parabolic filter

Filtering kernel with support from -1.0 to 1.0 Parabolic filter

Filtering kernel with support from -1 to 1 Roma and Peskin's filter

Filtering kernel with support from -1.0 to 1.0 Triangular filter (linear interpolation)

Filtering kernel with support from -1.0 to 1.0 Triweight filter

Compute collision forces

Deallocate data

Initialize the solver

Initialize boundary conditions

Initialize and partition grid

Initialize collision components

Initialize fields used by this solver and read initial conditions.

Initialize monitors used by this solver.

Initialize differential operators on current grid.

Initialize output variables

Close hdf5 file with h5hut

Finalize structure

Flush step data to disk

Get number of fields in step

Get number of data points in step

Initialize structure

Jump to a specific time step

Get information about the last time step

Create a new time step and update attributes

Open a hdf5 file with h5hut

Read Lagrangian/1D data fom a hdf5 file with h5hut If no offset is provided, use default h5hut file view. Otherwise, set file view manually.

Read scalar attributes

Read an array of attributes

Read Eulerian/3D data from a hdf5 file with h5hut

Return number of time steps

Write Lagrangian/1D data to a hdf5 file with h5hut

Write scalar attributes

Write an array of attributes

Write the grid attributes

Write Eulerian/3D data to a hdf5 file with h5hut

Close hdf5 file

Close an HDF5 group

Create a group (analogous to directory) in an HDF5 file and update hash table

Finalize the hdf5 object

Function that will append and prepend '/' if missing

Returns the index of a group

Returns the HDF5 object id of the group

Initialize the hdf5 object

Open a hdf5 file

Open a group (analogous to directory) in an HDF5 file and updates hash table

Read a 1D dataset located under basegroup and given by name

Read a 3D dataset located under basegroup and given by name

Read a scalar attribute under a given group

Read a 1-D array of attributes under a given group

Read coordinates from HDF5 file. Only the root MPI rank does the reading, and then broadcasts to other MPI ranks.

Read the groups (i.e., directories) under a given base group in an HDF5 file.

Write an array/1D data to a HDF5 file.

Write Eulerian/3D data to a HDF5 file

Write a scalar attribute

Write an array of attributes

Write coordinates to HDF5 file. Only the root MPI rank does the writing.

Set the coefficients of the matrix

Define the entries of the matrix Ax=b one row at a time Finite difference/Finite volume 2nd order Laplacian: ddu/dxdx = -2 u(i,j,k)/dxdx + u(i-1,j,k)/dxdx + u(i+1,j,k)/dxdx ddu/dydy = -2 u(i,j,k)/dydy + u(i,j-1,k)/dydy + u(i,j+1,k)/dydy ddu/dzdz = -2 u(i,j,k)/dzdz + u(i,j,k-1)/dzdz + u(i,j,k+1)/dzdz

Destroy objects/pointers and clear data

Initialize the hypre object

Setup the hypre grid

Select one of the preconfigured solvers and get solver-specific parameters

Set the entries of the rhs vector, one element at a time

Set the entries of the rhs vector, one element at a time

Set the entries of the rhs vector, one element at a time

Setup the hypre objects in preparation for solves Note: Setting up HYPRE is an expensive operation so it's best to do this only once during a simulation as opposed to setting-up and destorying each time-step.

Setup the hypre grid

Setup matrix with IJ interface

Setup and build the matrix

Setup the rhs vector

Setup the rhs vector

Setup row indexing used with IJ interface

Setup the solution vector, and initialize it to zero

Setup solver with IJ interface

Setup the solution vector, and initialize it to zero

Setup the hypre solver

Setup the discretization stencil Each entry represents the relative offset (in index space) E.g.: a 2D 5-pt stencil would have the following geometry -- Offset { {0,0}, {-1,0}, {1,0}, {0,-1}, {0,1} } E.g.: a 3D 7-pt stencil would have the following geometry -- Offset { {0,0,0}, {-1,0,0}, {1,0,0}, {0,-1,0}, {0,1,0}, {0,0,-1}, {0,0,1} }

Solve the system Ax=b and return the solution

Solve the system Ax=b and return the solution, IJ interface

Solve the system Ax=b and return the solution, Struct interface

Integral of filtering kernel from 0 to r Box filter (step function) Here : r=x/l_f (non-dimensional position)

Integral of filtering kernel from 0 to r Cosine filter Here : r=x/l_f (non-dimensional position)

Integral of filtering kernel from 0 to r Cosine filter Here : r=x/l_f (non-dimensional position)

Integral of filtering kernel from 0 to r Parabolic filter Here : r=x/l_f (non-dimensional position)

Integral of filtering kernel from 0 to r Roma and Peskin filter Here : r=x/l_f (non-dimensional position)

Integral of filtering kernel from 0 to r Triangular filter (linear interpolation) Here : r=x/l_f (non-dimensional position)

Integral of filtering kernel from 0 to r Triweight filter Here : r=x/l_f (non-dimensional position)

Get a bump function centered on the lagrangian object

Routine to interpolate a field f defined on an Eulerian stencil to the location of a lagrangian object

Locate a Lagrangian object on an external grid. Returns the location of the cell containing the object.

Apply periodic boundary conditions

Communicate lagrangian objects across MPI_rank

Determines the size of the MPI derived type

Finalize the structure

Frees the MPI derived type

Returns the MPI rank that should own this lagrangian object based on which block it belongs to

Return the base name of file to write

Return the base name of file to write

Prints diagnostics information about the derived type

Initialize lagrangian objects related IO

Localize a Lagrangian object on the grid Returns the location of the closest collocated cell (staggering=0

Sorting routine: stacks active lagrangian objects at the beginning of the array then resizes

Changes the size of an array of Lagrangian objects

Reset the filter kerrnel Default is Triangle for interpolation and extrapolation

Adjust the size of the filter

Set file overwritting

Set the base name of file to read

Set the base name of file to write

Updates the total count of Lagrangian objects

Updates ghost objects Copies objects that lie "dist"-away from the block's boundaries to neighboring MPI-ranks and designate copies as Ghost Objects (id<0)

Update ghost objects in the idir direction Copies objects that lie "dist"-away from the block's boundaries in idir-direction to neighboring MPI-ranks. Copied-objects get a negative ID to designate them as ghost objects

Assignment

Print information about this marker

Add an IB plane

Add an IB sphere

Find the center of mass

Compute the solid volume fraction on the mesh

Compute a filtered quantity on the eulerian grid

Finalize the marker_set type. This subourtine replaces the inheritted lagrangian_final.

Compute the IB forcing. Interpolation is carried out by convolution Fails when grid spacing is greater than filter half width

Compute the IB forcing Interpolation is carried out by linear interpolation.

Initialize the marker_set type. This subourtine replaces the inheritted lagrangian_init.

Load markers from a binary STL. This is a serial routine.

Prepare marker_set for use with solvers

Read marker data from file in parallel

Set up parameters used when creating the MPI derived type Create the MPI structure

Set the sample type used in allocation of polymorphic variables

Set the chunk size.

Updates the Normals field

Updates the Surface Density Function

Write marker data to file in parallel

Finalize the monitor object

Define how to print numbers Format specifier for an integer Format specifier for a real Format specifier for a logical Format specifier for a full line

Intializes a single monitor

Prints to file or stdout content of the monitor

Finalize the monitor object

Creates a new monitor

Changes the size of an array of monitors

Returns the ID of a monitor identified by name

Print a number of diagnostics information

Initialize a set of monitors

Prints content of monitors

Set the label or value of a column of a certain monitor

Close file with MPI

Finalize structure

Read file attributes

Initialize structure

Open a hdf5 file with ngadata

Read Eulerian/3D data from a hdf5 file with h5hut

Create MPI types needed for IO

Set the file view

Close file with MPI

Define MPI Type and size for a NGA particle

Finalize structure

Read file attributes

Initialize structuree

Open raw NGA part file

Read NGA particles, convert and store them in LEAP particle structure

Apply Dirichlet boundary conditions to the RHS of a Laplacian equation

Build Laplacian operator using Morinishi's schemes

Compute d(U1 U1)/dx1

Compute d(U1 U2)/dx1

Compute d(U1 U1)/dx1

Compute d(U2 U1)/dx2

Compute d(U2 U2)/dx2

Compute d(U2 U1)/dx2

Compute d(U3 U1)/dx1

Compute d(U3 U1)/dx1

Compute d(U3 U1)/dx1

Compute the derivative in the x1-direction. Note: If input is face-centered (on x1), result is cell-centered (on x1m). If input is cell-centered (on x1m), result is face-centered (on x1).

Compute the derivative in the x2-direction. Note: If input is face-centered (on x2), result is cell-centered (on x2m). If input is cell-centered (on x2m), result is face-centered (on x2).

Compute the derivative in the x3-direction. Note: If input is face-centered (on x3), result is cell-centered (on x3m). If input is cell-centered (on x3m), result is face-centered (on x3).

Compute the divergence of a vector (in1,in2,in3) This function takes in1,in2,in3 cell-centered (stag=0) and returns the divergence on cell centers (stag=0)

Clear data

Compute the derivative in the x-direction (dir=1)

Compute the derivative in the y-direction (dir=2)

Compute the derivative in the z-direction (dir=3)

Compute the derivative in the x-direction (dir=1)

Compute the derivative in the z-direction (dir=2)

Compute the derivative in the z-direction (dir=3)

Interpolate in the x1-direction. Note: If input is face-centered (on x1), result is cell-centered (on x1m). If input is cell-centered (on x1m), result is face-centered (on x1).

Interpolate in the x2-direction. Note: If input is face-centered (on x2), result is cell-centered (on x2m). If input is cell-centered (on x2m), result is face-centered (on x2).

Interpolate in the x3-direction. Note: If input is face-centered (on x3), result is cell-centered (on x3m). If input is cell-centered (on x3m), result is face-centered (on x3).

Compute the divergence of a vector (in1,in2,in3) This function takes in1,in2,in3 face-centered (stag=1/2/3) and returns the divergence on cell centers (stag=0)

Compute the strain rate tensor from the velocity field. Result is on mid points (staggering=0). Tensor is stored as follows: S = 0.5*( grad(u) + grad(u)^T ) ( S(1) S(4) S(6) ) = ( S(4) S(2) S(5) ) ( S(6) S(5) S(3) )

Finalize MPI and the parallel environment

Initialize the parallel environement

Set the MPI file system hints

Subroutine to gracefully stop the execution with an optional error message.

Returns the elapsed time since an arbitrary origin. Note that different ranks return different WTIMEs

Builds a Cartesian topolgy with MPI

Resize entries array to add a new entry

Assing default to value: 0D version

Assing default to value: 1D version

Return ID of label in the array of entries Returns 0 if label not found.

Finalization routine

Initialization the praser

Check whether the field is defined

Read & parse the input file

Parse a line

Prints all variables found in the parsed file

Read value: 0D version

Read value: 1D version

Parse a line

Assignment

Assignment

Advance centers to next timestep

Change Particle Type. This delete all existing particle data and redclare the associated MPI type.

Create monitor file for Point Particles

Filter a quantity to the Eulerian grid

Prepare for use with solvers.

Read particle data from file in parallel

Set up parameters used when creating the MPI derived type

Set up parameters used when creating the MPI derived type

Set up parameters used when creating the MPI derived type

Set the sample type used in allocation of polymorphic variables

Store values from previous timestep

Update monitoring data

Write particle data to file in parallel

Write data to disk in Silo format

Adds a new variable to region

Resize array to accomodate a new element

Returns index of a variable in this region, or -1 if not found.

Initializes a region

Assignment

Advance centers to next timestep

Advance markers to next timestep

Create monitor file for Resolved Particles

Filter a quantity to the Eulerian grid

Finalize the ResPart_set type. This subourtine replaces the inheritted lagrangian_final.

Compute hydrodynamic force on particle

Compute the IB forcing

Return MPI rank of the lagrangian centroid owning this marker.

Compute hydrodynamic stresses on markers

Initialize the ResPart_set type. This subourtine replaces the inheritted lagrangian_init.

Prepare ResPart_set for use with solvers

Read ResPart data from file

Regroup markers with their respective centroids on the same MPI block

Filter a quantity to the Eulerian grid

Set up parameters used when creating the MPI derived type

description

Set the base name of file to write

Set the base name of file to write

Store values from previous timestep

Update lookup array -- The lookup array returns the local (MPI rank) index of a centroid when given the global ID of that centroid

Updates the Normals field

Updates the Surface Density Function

description

Write data to disk in Silo format

Set up silo groups for poor man's IO. Split MPI ranks into groups of size nproc_node. Each group will write squentially to its own file.

Finalize structure

Initialize structure

Create a new Silo virtual data base (VDB) for this timestep

Create silo files and their internal structure

Write the grid attributes

Write the grid attributes

Write Eulerian/3D data to a hdf5 file with silo

Write Eulerian/3D data to a hdf5 file with silo

Communicate markers contained in all solid_obj

Filter a quantity to the Eulerian grid

Finalize

Initialize a collection of solid_obj

Localize markers on the gril for all solid_obj

Read all solid_obj

Select interp/extrap kernels

Change filter size to desired value

Set file overwritting

Set the names of files to read

Set the names of files to write

Write all solid_obj

Add timing info

Determines whether simulation is over

Change run status to finished

Finalize

Updates timing info with elapsed time since beginning of current iteration

Initialize the timer

Move timer from n to n+1

Determine whether it is time to write visualization files

Determine whether it is time to write restart files

Updates timing info with elapsed time since beginning of current iteration