MAGMA  2.3.0
Matrix Algebra for GPU and Multicore Architectures
 All Classes Files Functions Friends Groups Pages
Miscellaneous utilities

Modules

 make_lwork: Round lwork for float
 

Functions

void magma_version (magma_int_t *major, magma_int_t *minor, magma_int_t *micro)
 Returns version of MAGMA, as defined by MAGMA_VERSION_MAJOR, MAGMA_VERSION_MINOR, MAGMA_VERSION_MICRO constants. More...
 
magma_int_t magma_num_gpus (void)
 Determines the number of GPUs to use, based on $MAGMA_NUM_GPUS environment variable, and limited to actual number of GPUs available. More...
 
magma_int_t magma_is_devptr (const void *A)
 For debugging purposes, determines whether a pointer points to CPU or GPU memory. More...
 
magma_int_t magma_cKrylov_check (magma_solver_type solver)
 Checks whether a solver is among the list of Krylov solvers. More...
 
magma_int_t magma_dKrylov_check (magma_solver_type solver)
 Checks whether a solver is among the list of Krylov solvers. More...
 
magma_int_t magma_sKrylov_check (magma_solver_type solver)
 Checks whether a solver is among the list of Krylov solvers. More...
 
magma_int_t magma_zKrylov_check (magma_solver_type solver)
 Checks whether a solver is among the list of Krylov solvers. More...
 

Detailed Description

Function Documentation

void magma_version ( magma_int_t *  major,
magma_int_t *  minor,
magma_int_t *  micro 
)

Returns version of MAGMA, as defined by MAGMA_VERSION_MAJOR, MAGMA_VERSION_MINOR, MAGMA_VERSION_MICRO constants.

Parameters
[out]majorSet to major version number.
[out]minorSet to minor version number.
[out]microSet to micro version number.
magma_int_t magma_num_gpus ( void  )

Determines the number of GPUs to use, based on $MAGMA_NUM_GPUS environment variable, and limited to actual number of GPUs available.

If $MAGMA_NUM_GPUS is not set, uses 1.

Returns
Number of GPUs to use.
magma_int_t magma_is_devptr ( const void *  A)

For debugging purposes, determines whether a pointer points to CPU or GPU memory.

On CUDA architecture 2.0 cards with unified addressing, CUDA can tell if it is a device pointer or pinned host pointer. For malloc'd host pointers, cudaPointerGetAttributes returns error, implying it is a (non-pinned) host pointer.

On older cards, this cannot determine if it is CPU or GPU memory.

Parameters
[in]Apointer to test
Returns
1: if A is a device pointer (definitely),
0: if A is a host pointer (definitely or inferred from error),
-1: if unknown.
magma_int_t magma_cKrylov_check ( magma_solver_type  solver)

Checks whether a solver is among the list of Krylov solvers.

The result is passed in info: 1 yes - Krylov solver 0 no - no Krylov solver of the list.

Parameters
[in]solvermagma_solver_type
Returns
error
magma_int_t magma_dKrylov_check ( magma_solver_type  solver)

Checks whether a solver is among the list of Krylov solvers.

The result is passed in info: 1 yes - Krylov solver 0 no - no Krylov solver of the list.

Parameters
[in]solvermagma_solver_type
Returns
error
magma_int_t magma_sKrylov_check ( magma_solver_type  solver)

Checks whether a solver is among the list of Krylov solvers.

The result is passed in info: 1 yes - Krylov solver 0 no - no Krylov solver of the list.

Parameters
[in]solvermagma_solver_type
Returns
error
magma_int_t magma_zKrylov_check ( magma_solver_type  solver)

Checks whether a solver is among the list of Krylov solvers.

The result is passed in info: 1 yes - Krylov solver 0 no - no Krylov solver of the list.

Parameters
[in]solvermagma_solver_type
Returns
error