symmrb.h File Reference

---

Detailed Description

Implementation of the register blocking heuristic for symmetric matrices, as described in the paper by Lee, et al., in ICPP'04, but extended to the Symmetric or Hermitian cases and based on MBCSR format.

#include <oski/common.h>
#include <oski/modcommon.h>
#include <oski/heurexport.h>

Go to the source code of this file.

Data Structures
struct	oski_symmrb_t
	Stores the results of evaluating the register blocking heuristic. More...
struct	oski_profile_MBCSRSymmMatMult_t
	Stores a register blocking profile. More...
Name mangling.
#define	oski_symmrb_t   MANGLE_(oski_symmrb_t)
#define	oski_LoadProfileMBCSRSymmMatMult   MANGLE_(oski_LoadProfileMBCSRSymmMatMult)
#define	oski_UnloadProfileMBCSRSymmMatMult   MANGLE_(oski_UnloadProfileMBCSRSymmMatMult)
Defines
#define	INC_OSKI_HEUR_SYMMRB_H
	oski/heur/symmrb.h included.
#define	PROF_MBCSR_GET(P, r, c)   ((P)->perf[((r)-1)*(P)->max_c + (c)-1])
	Returns the performance (Mflop/s) at a block size , where and are 1-based.
#define	PROF_MBCSR_SET(P, r, c, v)   ((P)->perf[((r)-1)*(P)->max_c + (c)-1] = (v))
	Set the performance (Mflop/s) at a block size , where and are 1-based.
Functions
const oski_profile_MBCSRSymmMatMult_t *	oski_LoadProfileMBCSRSymmMatMult (const char *filename)
	Load the uniformly-aligned register blocking profile for sparse matrix-vector multiply.
void	oski_UnloadProfileMBCSRSymmMatMult (void)
	Unload the uniformly-aligned register blocking profile for sparse matrix-vector multiply.

---

Function Documentation

const oski_profile_MBCSRSymmMatMult_t* oski_LoadProfileMBCSRSymmMatMult (  const char *  filename  )

Load the uniformly-aligned register blocking profile for sparse matrix-vector multiply. This implementation

---