SuperLU_DIST
4.0
superlu_dist on CPU and GPU clusters
|
Files | |
file | Cnames.h [code] |
Macro definitions. | |
file | comm.c |
Broadcast an array of dtype numbers. | |
file | cublas_utils.c |
file | cublas_utils.h [code] |
file | dcomplex.c |
Defines common arithmetic operations for complex type. | |
file | dcomplex.h [code] |
Header for dcomplex.c. | |
file | ddistribute.c |
Distribute the matrix onto the 2D process mesh. | |
file | dgsequ.c |
Computes row and column scalings. | |
file | dlamch.c |
Determines double precision machine parameters. | |
file | dlangs.c |
Returns the value of the one norm, the infinity norm, or the element of largest value. | |
file | dlaqgs.c |
Equilibrates a general sparse M by N matrix A. | |
file | dldperm.c |
Finds a row permutation so that the matrix has large entries on the diagonal. | |
file | dldperm_dist.c |
Finds a row permutation so that the matrix has large entries on the diagonal. | |
file | dlook_ahead_update.c |
Look-ahead update of the Schur complement. | |
file | dmemory.c |
Memory utilities. | |
file | dmyblas2.c |
Level 2 BLAS operations: solves and matvec, written in C. | |
file | dreadhb.c |
Read a DOUBLE PRECISION matrix stored in Harwell-Boeing format. | |
file | dreadMM.c |
Contributed by Francois-Henry Rouet. | |
file | dreadrb.c |
Read a matrix stored in Rutherford-Boeing format. | |
file | dreadtriple.c |
file | dreadtriple_noheader.c |
file | dscatter.c |
Scatter the computed blocks into LU destination. | |
file | dSchCompUdt-2Ddynamic.c |
THis file contains the main loop of pdgstrf which involves rank k update of the Schur complement. Uses 2D partitioning for the scatter phase. | |
file | dSchCompUdt-cuda.c |
This file contains the main loop of pdgstrf which involves rank k update of the Schur complement. Uses CUDA GPU. | |
file | dsp_blas2.c |
Sparse BLAS 2, using some dense BLAS 2 operations. | |
file | dsp_blas3.c |
Sparse BLAS3, using some dense BLAS3 operations. | |
file | dutil.c |
Several matrix utilities. | |
file | etree.c |
Elimination tree computation and layout routines. | |
file | get_perm_c.c |
Gets matrix permutation. | |
file | get_perm_c_parmetis.c |
Gets matrix permutation. | |
file | GetDiagU.c |
Extracts the main diagonal of matrix U. | |
file | html_mainpage.h [code] |
file | lsame.c |
Check if CA is the same letter as CB regardless of case. | |
file | machines.h [code] |
These macros define which machine will be used. | |
file | mc64ad_dist.c |
Permute large entries to the main diagonal. | |
file | memory.c |
Memory utilities. | |
file | mmd.c |
Implements the minimum degree algorithm. | |
file | old_colamd.c |
An approximate minimum degree column ordering algorithm. | |
file | old_colamd.h [code] |
colamd include file | |
file | pddistribute.c |
Re-distribute A on the 2D process mesh. | |
file | pdgsequ.c |
Computes row and column scalings. | |
file | pdgsmv.c |
Parallel sparse matrix-vector multiplication. | |
file | pdgsmv_AXglobal.c |
Performs sparse matrix-vector multiplication. | |
file | pdgsrfs.c |
Improves the computed solution to a system of linear equations and provides error bounds and backward error estimates. | |
file | pdgsrfs_ABXglobal.c |
Improves the computed solution and provies error bounds. | |
file | pdgssvx.c |
Solves a system of linear equations A*X=B. | |
file | pdgssvx_ABglobal.c |
Solves a system of linear equations A*X=B,. | |
file | pdgstrf.c |
Performs LU factorization in parallel. | |
file | pdgstrf2.c |
Performs panel LU factorization. | |
file | pdgstrf_irecv.c |
Performs LU factorization in parallel. | |
file | pdgstrf_sherry.c |
file | pdgstrf_X1.c |
Performs the LU factorization in parallel. | |
file | pdgstrs.c |
Solves a system of distributed linear equations A*X = B with a general N-by-N matrix A using the LU factors computed previously. | |
file | pdgstrs1.c |
Solves a system of distributed linear equations. | |
file | pdgstrs_Bglobal.c |
Solves a system of distributed linear equations A*X = B with a general N-by-N matrix A using the LU factorization. | |
file | pdgstrs_Bglobal_Bsend.c |
Solves a system of distributed linear equations. | |
file | pdgstrs_lsum.c |
Perform local block modifications: lsum[i] -= L_i,k * X[k]. | |
file | pdgstrsL.c |
Solves a lower triangular system L*X = B, with L being the lower triangular factor computed previously by PDGSTRF. | |
file | pdlangs.c |
Returns the value of the one norm, or the Frobenius norm, or the infinity norm, or the element of largest value. | |
file | pdlaqgs.c |
Equilibrates a general sparse M by N matrix. | |
file | pdsymbfact_distdata.c |
Redistribute the symbolic structure of L and U from the distribution. | |
file | pdutil.c |
Several matrix utilities. | |
file | psymbfact.c |
Implements parallel symbolic factorization. | |
file | psymbfact.h [code] |
Definitions for parallel symbolic factorization routine. | |
file | psymbfact_util.c |
Utilities for parallel symbolic factorization routine. | |
file | pxerbla.c |
file | pzdistribute.c |
Re-distribute A on the 2D process mesh. | |
file | pzgsequ.c |
Computes row and column scalings. | |
file | pzgsmv.c |
Parallel sparse matrix-vector multiplication. | |
file | pzgsmv_AXglobal.c |
Performs sparse matrix-vector multiplication. | |
file | pzgsrfs.c |
Improves the computed solution to a system of linear equations and provides error bounds and backward error estimates. | |
file | pzgsrfs_ABXglobal.c |
Improves the computed solution and provies error bounds. | |
file | pzgssvx.c |
Solves a system of linear equations A*X=B. | |
file | pzgssvx_ABglobal.c |
Solves a system of linear equations A*X=B,. | |
file | pzgstrf.c |
Performs LU factorization in parallel. | |
file | pzgstrf2.c |
Performs panel LU factorization. | |
file | pzgstrf_irecv.c |
Performs LU factorization in parallel. | |
file | pzgstrs.c |
Solves a system of distributed linear equations A*X = B with a general N-by-N matrix A using the LU factors computed previously. | |
file | pzgstrs1.c |
Solves a system of distributed linear equations. | |
file | pzgstrs_Bglobal.c |
Solves a system of distributed linear equations A*X = B with a general N-by-N matrix A using the LU factorization. | |
file | pzgstrs_lsum.c |
Perform local block modifications: lsum[i] -= L_i,k * X[k]. | |
file | pzlangs.c |
Returns the value of the one norm, or the Frobenius norm, or the infinity norm, or the element of largest value. | |
file | pzlaqgs.c |
Equilibrates a general sparse M by N matrix. | |
file | pzsymbfact_distdata.c |
Redistribute the symbolic structure of L and U from the distribution. | |
file | pzutil.c |
Several matrix utilities. | |
file | slamch.c |
Determines single precision machine parameters. | |
file | sp_colorder.c |
Permutes the columns of the original matrix. | |
file | sp_ienv.c |
Chooses machine-dependent parameters for the local environment. | |
file | static_schedule.c |
Performs static scheduling for the look-ahead factorization algorithm. | |
file | superlu_ddefs.h [code] |
Distributed SuperLU data types and function prototypes. | |
file | superlu_defs.h [code] |
Definitions which are precision-neutral. | |
file | superlu_enum_consts.h [code] |
enum constants header file | |
file | superlu_grid.c |
SuperLU grid utilities. | |
file | superlu_timer.c |
Returns the time in seconds used by the process. | |
file | superlu_zdefs.h [code] |
Distributed SuperLU data types and function prototypes. | |
file | supermatrix.h [code] |
Matrix type definitions. | |
file | symbfact.c |
Performs a symbolic factorization. | |
file | util.c |
Utilities functions. | |
file | util_dist.h [code] |
Header for utilities. | |
file | xerbla.c |
file | zdistribute.c |
Distribute the matrix onto the 2D process mesh. | |
file | zdistribute_mark.c |
Distribute the matrix onto the 2D process mesh. | |
file | zgsequ.c |
Computes row and column scalings. | |
file | zlangs.c |
Returns the one norm, or the Frobenius norm, or the infinity norm, or the element of largest value. | |
file | zlaqgs.c |
Equilibrates a general sparse M by N matrix A. | |
file | zldperm.c |
Finds a row permutation so that the matrix has large entries on the diagonal. | |
file | zldperm_dist.c |
Finds a row permutation so that the matrix has large entries on the diagonal. | |
file | zlook_ahead_update.c |
Look-ahead update of the Schur complement. | |
file | zmemory.c |
Memory utilities. | |
file | zmyblas2.c |
Level 2 BLAS operations: solves and matvec, written in C. | |
file | zreadhb.c |
Read a DOUBLE COMPLEX PRECISION matrix stored in Harwell-Boeing format. | |
file | zreadMM.c |
Contributed by Francois-Henry Rouet. | |
file | zreadrb.c |
Read a matrix stored in Rutherford-Boeing format. | |
file | zreadtriple.c |
file | zreadtriple_noheader.c |
file | zscatter.c |
Scatter the computed blocks into LU destination. | |
file | zSchCompUdt-2Ddynamic.c |
THis file contains the main loop of pdgstrf which involves rank k update of the Schur complement. Uses 2D partitioning for the scatter phase. | |
file | zSchCompUdt-cuda.c |
This file contains the main loop of pzgstrf which involves rank k update of the Schur complement. Uses CUDA GPU. | |
file | zsp_blas2.c |
Solves one of the systems of equations A*x = b, or A'*x = b. | |
file | zsp_blas3.c |
Sparse BLAS3, using some dense BLAS3 operations. | |
file | zutil.c |
Several matrix utilities. | |