s_spm_matrixvector.c File Reference

#include "common.h"
#include <lapacke.h>
#include <cblas.h>

Go to the source code of this file.

Functions
int	spm_sspmm (spm_side_t side, spm_trans_t transA, spm_trans_t transB, spm_int_t K, float alpha, const spmatrix_t *A, const float *B, spm_int_t ldb, float beta, float *C, spm_int_t ldc)
	Compute a matrix-matrix product.
int	spm_sspmv (spm_trans_t trans, float alpha, const spmatrix_t *A, const float *x, spm_int_t incx, float beta, float *y, spm_int_t incy)
	compute the matrix-vector product:
#define	LAPACKE_slascl_work(_dir_, _uplo_, _kl_, _ku_, _cfrom_, _cto_, _m_, _n_, _A_, _lda_)
	Alias if Lapacke zlscl is not available.
typedef struct __spm_smatvec_s	__spm_smatvec_t
	Typedef associated to structure.
typedef float(*	__conj_fct_t) (float)
	Typedef to define the op function applied to the element (id or ).
typedef int(*	__loop_fct_t) (const __spm_smatvec_t *)
	Typedef to the main loop function performing the matvec operation.
static float	__fct_id (float val)
	Identity function.
static void	s_spmm_build_Ctmp (int nrhs, const spmatrix_t *spm, const float *Cloc, spm_int_t ldcl, float **Cglb, spm_int_t *ldcg)
	Build a global C RHS, set to 0 for remote datas.
static void	s_spmm_build_Btmp (int nrhs, const spmatrix_t *spm, const float *Bloc, spm_int_t ldbl, float **Bglb, spm_int_t *ldbg)
	Build a global B vector by gathering datas from all nodes.

Detailed Description

SParse Matrix package matrix-vector multiplication routines.

Copyright

2016-2026 Bordeaux INP, CNRS (LaBRI UMR 5800), Inria, Univ. Bordeaux. All rights reserved.

Version

1.3.0

Author

Matthieu Kuhn

Mathieu Faverge

Tony Delarue

Alycia Lisito

Date

2024-07-26

Generated arithmetic file from /builds/fpruvost/spm/src/z_spm_matrixvector.c, normal z -> s, Fri Oct 23 13:05:00 2026

Definition in file s_spm_matrixvector.c.

Macro Definition Documentation

LAPACKE_slascl_work

#define LAPACKE_slascl_work	(		_dir_,
			_uplo_,
			_kl_,
			_ku_,
			_cfrom_,
			_cto_,
			_m_,
			_n_,
			_A_,
			_lda_ )

Value:

    __spm_slascl( (_cto_), (_m_), (_n_), (_A_), (_lda_) )

Alias if Lapacke zlscl is not available.

Parameters

[in]	_dir_	Unused parameter
[in]	_uplo_	Unused parameter
[in]	_kl_	Unused parameter
[in]	_ku_	Unused parameter
[in]	_cfrom_	Unused parameter
[in]	_cto_	The scaling factor of the matrix A
[in]	_m_	The number of rows of the matrix A
[in]	_n_	The number of columns of the matrix A
[in,out]	_A_	On entry the lda-by-n matrix to scale. On exit, the matrix is multiplied by cto.
[in]	_lda_	The leading dimension of the matrix A. lda >= max(1, m)

Definition at line 737 of file s_spm_matrixvector.c.

Referenced by spm_sspmm(), and spmScalMat().

Typedef Documentation

__spm_smatvec_t

typedef struct __spm_smatvec_s __spm_smatvec_t

Typedef associated to structure.

Definition at line 35 of file s_spm_matrixvector.c.

__conj_fct_t

typedef float(* __conj_fct_t) (float)

Typedef to define the op function applied to the element (id or ).

Definition at line 40 of file s_spm_matrixvector.c.

__loop_fct_t

typedef int(* __loop_fct_t) (const __spm_smatvec_t *)

Typedef to the main loop function performing the matvec operation.

Definition at line 45 of file s_spm_matrixvector.c.

Function Documentation

__fct_id()

float __fct_id ( float val )

static

Identity function.

Parameters

[in]

val

TODO

Returns

id(val)

Definition at line 62 of file s_spm_matrixvector.c.

s_spmm_build_Ctmp()

void s_spmm_build_Ctmp ( int nrhs, const spmatrix_t * spm, const float * Cloc, spm_int_t ldcl, float ** Cglb, spm_int_t * ldcg )

static

Build a global C RHS, set to 0 for remote datas.

Parameters

[in]	nrhs	The number of RHS.
[in]	spm	The pointer to the sparse matrix structure.
[in]	Cloc	The local matrix of size ldcl -by- nrhs
[in]	ldcl	The leading dimension of Cloc. ldcl >= max( 1, spm->nexp )
[out]	Cglb	On exit, contains the centralized version of the Cloc matrix.
[out]	ldcg	On exit, contains the leading dimension of the Cglb matrix. ldcg >= max(1, spm->gNexp )

Definition at line 948 of file s_spm_matrixvector.c.

References spmatrix_s::baseval, spmatrix_s::dof, spmatrix_s::dofs, spmatrix_s::gNexp, spmatrix_s::loc2glob, and spmatrix_s::n.

Referenced by spm_sspmm(), and spm_sspmv().

s_spmm_build_Btmp()

void s_spmm_build_Btmp ( int nrhs, const spmatrix_t * spm, const float * Bloc, spm_int_t ldbl, float ** Bglb, spm_int_t * ldbg )

static

Build a global B vector by gathering datas from all nodes.

Parameters

[in]	spm	The pointer to the sparse matrix structure.
[in]	nrhs	The number of RHS.
[in]	Bloc	The local B vector.
[in,out]	ldbl	The leading dimension of the local B vector.
[out]	Bglb	The global B vector.
[out]	ldbg	The leading dimension of the global B vector.

Definition at line 1012 of file s_spm_matrixvector.c.

References spmatrix_s::gNexp, and s_spmGatherRHS().

Referenced by spm_sspmm(), and spm_sspmv().