getmatrix: GPU => CPU

Macros
#define	magma_igetmatrix(m, n, dA_src, ldda, hB_dst, ldb,queue)
	Type-safe version of magma_getmatrix() for magma_int_t arrays.
#define	magma_igetmatrix_async(m, n, dA_src, ldda, hB_dst, ldb,queue)
	Type-safe version of magma_getmatrix_async() for magma_int_t arrays.
#define	magma_index_getmatrix(m, n, dA_src, ldda, hB_dst, ldb,queue)
	Type-safe version of magma_getmatrix() for magma_index_t arrays.
#define	magma_index_getmatrix_async(m, n, dA_src, ldda, hB_dst, ldb,queue)
	Type-safe version of magma_getmatrix_async() for magma_index_t arrays.
#define	magma_cgetmatrix(m, n, dA_src, ldda, hB_dst, ldb,queue)
	Type-safe version of magma_getmatrix() for magmaFloatComplex arrays.
#define	magma_cgetmatrix_async(m, n, dA_src, ldda, hB_dst, ldb, queue)
	Type-safe version of magma_getmatrix_async() for magmaFloatComplex arrays.
#define	magma_dgetmatrix(m, n, dA_src, ldda, hB_dst, ldb,queue)
	Type-safe version of magma_getmatrix() for double arrays.
#define	magma_dgetmatrix_async(m, n, dA_src, ldda, hB_dst, ldb, queue)
	Type-safe version of magma_getmatrix_async() for double arrays.
#define	magma_sgetmatrix(m, n, dA_src, ldda, hB_dst, ldb,queue)
	Type-safe version of magma_getmatrix() for float arrays.
#define	magma_sgetmatrix_async(m, n, dA_src, ldda, hB_dst, ldb, queue)
	Type-safe version of magma_getmatrix_async() for float arrays.
#define	magma_zgetmatrix(m, n, dA_src, ldda, hB_dst, ldb,queue)
	Type-safe version of magma_getmatrix() for magmaDoubleComplex arrays.
#define	magma_zgetmatrix_async(m, n, dA_src, ldda, hB_dst, ldb, queue)
	Type-safe version of magma_getmatrix_async() for magmaDoubleComplex arrays.
#define	magma_getmatrix(m, n, elemSize, dA_src, ldda, hB_dst, ldb,queue)
	Copy all or part of matrix dA_src on GPU device to hB_dst on CPU host. More...
#define	magma_getmatrix_async(m, n, elemSize, dA_src, ldda, hB_dst, ldb,queue)
	Copy all or part of matrix dA_src on GPU device to hB_dst on CPU host. More...

Detailed Description

Macro Definition Documentation

magma_getmatrix	(		m,
			n,
			elemSize,
			dA_src,
			ldda,
			hB_dst,
			ldb,
			queue
	)

Copy all or part of matrix dA_src on GPU device to hB_dst on CPU host.

Elements may be arbitrary size. Type-safe versions set elemSize appropriately.

This version synchronizes the queue after the transfer. See magma_getmatrix_async() for an asynchronous version.

Parameters

[in]	m	Number of rows of matrix A. m >= 0.
[in]	n	Number of columns of matrix A. n >= 0.
[in]	elemSize	Size of each element, e.g., sizeof(double).
[in]	dA_src	Source array of dimension (ldda,n), on GPU device.
[in]	ldda	Leading dimension of matrix A. ldda >= m.
[out]	hB_dst	Destination array of dimension (ldb,n), on CPU host.
[in]	ldb	Leading dimension of matrix B. ldb >= m.
[in]	queue	Queue to execute in.

magma_getmatrix_async	(		m,
			n,
			elemSize,
			dA_src,
			ldda,
			hB_dst,
			ldb,
			queue
	)

Copy all or part of matrix dA_src on GPU device to hB_dst on CPU host.

Elements may be arbitrary size. Type-safe versions set elemSize appropriately.

This version is asynchronous: it may return before the transfer finishes, if hB_dst is pinned CPU memory. See magma_getmatrix() for a synchronous version.

Parameters

[in]	m	Number of rows of matrix A. m >= 0.
[in]	n	Number of columns of matrix A. n >= 0.
[in]	elemSize	Size of each element, e.g., sizeof(double).
[in]	dA_src	Source array of dimension (ldda,n), on GPU device.
[in]	ldda	Leading dimension of matrix A. ldda >= m.
[out]	hB_dst	Destination array of dimension (ldb,n), on CPU host.
[in]	ldb	Leading dimension of matrix B. ldb >= m.
[in]	queue	Queue to execute in.