This class implements an ilu0-bicgstab solver on FPGA.
More...
#include <FPGASolverBackend.hpp>
|
| FpgaSolverBackend (std::string fpga_bitstream, int linear_solver_verbosity, int maxit, double tolerance, ILUReorder opencl_ilu_reorder) |
| Construct an fpgaSolver. More...
|
|
| ~FpgaSolverBackend () |
| Destroy an fpgaSolver, and free memory.
|
|
SolverStatus | solve_system (std::shared_ptr< BlockedMatrix > matrix, double *b, std::shared_ptr< BlockedMatrix > jacMatrix, WellContributions &wellContribs, BdaResult &res) override |
| Solve linear system, A*x = b, matrix A must be in blocked-CSR format. More...
|
|
void | get_result (double *x) override |
| Get result after linear solve, and peform postprocessing if necessary. More...
|
|
| BdaSolver (int linear_solver_verbosity, int max_it, double tolerance_) |
| Construct a BdaSolver, can be cusparseSolver, openclSolver, fpgaSolver. More...
|
|
| BdaSolver (int linear_solver_verbosity, int max_it, double tolerance_, unsigned int deviceID_) |
|
| BdaSolver (int linear_solver_verbosity, int max_it, double tolerance_, unsigned int platformID_, unsigned int deviceID_) |
|
| BdaSolver (std::string fpga_bitstream, int linear_solver_verbosity, int max_it, double tolerance_) |
|
virtual | ~BdaSolver () |
| Define virtual destructor, so that the derivedclass destructor will be called.
|
|
|
int | verbosity = 0 |
|
int | maxit = 200 |
|
double | tolerance = 1e-2 |
|
std::string | bitstream |
|
int | N |
|
int | Nb |
|
int | nnz |
|
int | nnzb |
|
unsigned int | platformID = 0 |
|
unsigned int | deviceID = 0 |
|
bool | initialized = false |
|
template<unsigned int block_size>
class Opm::Accelerator::FpgaSolverBackend< block_size >
This class implements an ilu0-bicgstab solver on FPGA.
◆ FpgaSolverBackend()
template<unsigned int block_size>
Construct an fpgaSolver.
- Parameters
-
[in] | fpga_bitstream | FPGA bitstream file name |
[in] | linear_solver_verbosity | verbosity of fpgaSolver |
[in] | maxit | maximum number of iterations for fpgaSolver |
[in] | tolerance | required relative tolerance for fpgaSolver |
[in] | opencl_ilu_reorder | select either level_scheduling or graph_coloring, see ILUReorder.hpp for explanation |
◆ get_result()
template<unsigned int block_size>
Get result after linear solve, and peform postprocessing if necessary.
- Parameters
-
[in,out] | x | resulting x vector, caller must guarantee that x points to a valid array |
Implements Opm::Accelerator::BdaSolver< block_size >.
◆ solve_system()
template<unsigned int block_size>
Solve linear system, A*x = b, matrix A must be in blocked-CSR format.
- Parameters
-
[in] | matrix | matrix A |
[in] | b | input vector, contains N values |
[in] | jacMatrix | matrix for preconditioner |
[in] | wellContribs | WellContributions, not used in FPGA solver because it requires them already added to matrix A |
[in,out] | res | summary of solver result |
- Returns
- status code
Implements Opm::Accelerator::BdaSolver< block_size >.
The documentation for this class was generated from the following files: