MAGMA  2.3.0
Matrix Algebra for GPU and Multicore Architectures
 All Classes Files Functions Friends Groups Pages
lahef: Partial factorization; used by hetrf

Functions

magma_int_t magma_clahef_gpu (magma_uplo_t uplo, magma_int_t n, magma_int_t nb, magma_int_t *kb, magmaFloatComplex *hA, magma_int_t lda, magmaFloatComplex_ptr dA, magma_int_t ldda, magma_int_t *ipiv, magmaFloatComplex_ptr dW, magma_int_t lddw, magma_queue_t queues[], magma_event_t events[], magma_int_t *info)
 CLAHEF computes a partial factorization of a complex Hermitian matrix A using the Bunch-Kaufman diagonal pivoting method. More...
 
magma_int_t magma_dlasyf_gpu (magma_uplo_t uplo, magma_int_t n, magma_int_t nb, magma_int_t *kb, double *hA, magma_int_t lda, magmaDouble_ptr dA, magma_int_t ldda, magma_int_t *ipiv, magmaDouble_ptr dW, magma_int_t lddw, magma_queue_t queues[], magma_event_t events[], magma_int_t *info)
 DLASYF computes a partial factorization of a real symmetric matrix A using the Bunch-Kaufman diagonal pivoting method. More...
 
magma_int_t magma_slasyf_gpu (magma_uplo_t uplo, magma_int_t n, magma_int_t nb, magma_int_t *kb, float *hA, magma_int_t lda, magmaFloat_ptr dA, magma_int_t ldda, magma_int_t *ipiv, magmaFloat_ptr dW, magma_int_t lddw, magma_queue_t queues[], magma_event_t events[], magma_int_t *info)
 SLASYF computes a partial factorization of a real symmetric matrix A using the Bunch-Kaufman diagonal pivoting method. More...
 
magma_int_t magma_zlahef_gpu (magma_uplo_t uplo, magma_int_t n, magma_int_t nb, magma_int_t *kb, magmaDoubleComplex *hA, magma_int_t lda, magmaDoubleComplex_ptr dA, magma_int_t ldda, magma_int_t *ipiv, magmaDoubleComplex_ptr dW, magma_int_t lddw, magma_queue_t queues[], magma_event_t events[], magma_int_t *info)
 ZLAHEF computes a partial factorization of a complex Hermitian matrix A using the Bunch-Kaufman diagonal pivoting method. More...
 

Detailed Description

Function Documentation

magma_int_t magma_clahef_gpu ( magma_uplo_t  uplo,
magma_int_t  n,
magma_int_t  nb,
magma_int_t *  kb,
magmaFloatComplex *  hA,
magma_int_t  lda,
magmaFloatComplex_ptr  dA,
magma_int_t  ldda,
magma_int_t *  ipiv,
magmaFloatComplex_ptr  dW,
magma_int_t  lddw,
magma_queue_t  queues[],
magma_event_t  events[],
magma_int_t *  info 
)

CLAHEF computes a partial factorization of a complex Hermitian matrix A using the Bunch-Kaufman diagonal pivoting method.

The partial factorization has the form:

A = ( I U12 ) ( A11 0 ) ( I 0 ) if UPLO = MagmaUpper, or: ( 0 U22 ) ( 0 D ) ( U12' U22' )

A = ( L11 0 ) ( D 0 ) ( L11' L21' ) if UPLO = MagmaLower ( L21 I ) ( 0 A22 ) ( 0 I )

where the order of D is at most NB. The actual order is returned in the argument KB, and is either NB or NB-1, or N if N <= NB. Note that U' denotes the conjugate transpose of U.

CLAHEF is an auxiliary routine called by CHETRF. It uses blocked code (calling Level 3 BLAS) to update the submatrix A11 (if UPLO = MagmaUpper) or A22 (if UPLO = MagmaLower).

Parameters
[in]uploCHARACTER Specifies whether the upper or lower triangular part of the Hermitian matrix A is stored:
  • = MagmaUpper: Upper triangular
  • = MagmaLower: Lower triangular
[in]nINTEGER The order of the matrix A. N >= 0.
[in]nbINTEGER The maximum number of columns of the matrix A that should be factored. NB should be at least 2 to allow for 2-by-2 pivot blocks.
[out]kbINTEGER The number of columns of A that were actually factored. KB is either NB-1 or NB, or N if N <= NB.
[in,out]hACOMPLEX array, dimension (LDA,N) On entry, the Hermitian matrix A. If UPLO = MagmaUpper, the leading n-by-n upper triangular part of A contains the upper triangular part of the matrix A, and the strictly lower triangular part of A is not referenced. If UPLO = MagmaLower, the leading n-by-n lower triangular part of A contains the lower triangular part of the matrix A, and the strictly upper triangular part of A is not referenced. On exit, A contains details of the partial factorization.
[in]ldaINTEGER The leading dimension of the array A. LDA >= max(1,N).
[in,out]dACOMPLEX array on GPU, dimension (LDDA,N) On entry, the Hermitian matrix A. If UPLO = MagmaUpper, the leading n-by-n upper triangular part of A contains the upper triangular part of the matrix A, and the strictly lower triangular part of A is not referenced. If UPLO = MagmaLower, the leading n-by-n lower triangular part of A contains the lower triangular part of the matrix A, and the strictly upper triangular part of A is not referenced. On exit, A contains details of the partial factorization.
[in]lddaINTEGER The leading dimension of the array A. LDDA >= max(1,N).
[out]ipivINTEGER array, dimension (N) Details of the interchanges and the block structure of D. If UPLO = MagmaUpper, only the last KB elements of ipiv are set; if UPLO = MagmaLower, only the first KB elements are set.
If ipiv(k) > 0, then rows and columns k and ipiv(k) were interchanged and D(k,k) is a 1-by-1 diagonal block. If UPLO = MagmaUpper and ipiv(k) = ipiv(k-1) < 0, then rows and columns k-1 and -ipiv(k) were interchanged and D(k-1:k,k-1:k) is a 2-by-2 diagonal block. If UPLO = MagmaLower and ipiv(k) = ipiv(k+1) < 0, then rows and columns k+1 and -ipiv(k) were interchanged and D(k:k+1,k:k+1) is a 2-by-2 diagonal block.
[out]dW(workspace) COMPLEX array, dimension (LDDW,NB)
[in]lddwINTEGER The leading dimension of the array W. LDDW >= max(1,N).
[in]queuesmagma_queue_t queues contain the queues used for the partial factorization. Currently, only one queue is used.
[in]eventsmagma_event_t events contain the events used for the partial factorization. Currently, only one event is used.
[out]infoINTEGER
  • = 0: successful exit
  • > 0: if INFO = k, D(k,k) is exactly zero. The factorization has been completed, but the block diagonal matrix D is exactly singular.
magma_int_t magma_dlasyf_gpu ( magma_uplo_t  uplo,
magma_int_t  n,
magma_int_t  nb,
magma_int_t *  kb,
double *  hA,
magma_int_t  lda,
magmaDouble_ptr  dA,
magma_int_t  ldda,
magma_int_t *  ipiv,
magmaDouble_ptr  dW,
magma_int_t  lddw,
magma_queue_t  queues[],
magma_event_t  events[],
magma_int_t *  info 
)

DLASYF computes a partial factorization of a real symmetric matrix A using the Bunch-Kaufman diagonal pivoting method.

The partial factorization has the form:

A = ( I U12 ) ( A11 0 ) ( I 0 ) if UPLO = MagmaUpper, or: ( 0 U22 ) ( 0 D ) ( U12' U22' )

A = ( L11 0 ) ( D 0 ) ( L11' L21' ) if UPLO = MagmaLower ( L21 I ) ( 0 A22 ) ( 0 I )

where the order of D is at most NB. The actual order is returned in the argument KB, and is either NB or NB-1, or N if N <= NB. Note that U' denotes the conjugate transpose of U.

DLASYF is an auxiliary routine called by DSYTRF. It uses blocked code (calling Level 3 BLAS) to update the submatrix A11 (if UPLO = MagmaUpper) or A22 (if UPLO = MagmaLower).

Parameters
[in]uploCHARACTER Specifies whether the upper or lower triangular part of the symmetric matrix A is stored:
  • = MagmaUpper: Upper triangular
  • = MagmaLower: Lower triangular
[in]nINTEGER The order of the matrix A. N >= 0.
[in]nbINTEGER The maximum number of columns of the matrix A that should be factored. NB should be at least 2 to allow for 2-by-2 pivot blocks.
[out]kbINTEGER The number of columns of A that were actually factored. KB is either NB-1 or NB, or N if N <= NB.
[in,out]hADOUBLE PRECISION array, dimension (LDA,N) On entry, the symmetric matrix A. If UPLO = MagmaUpper, the leading n-by-n upper triangular part of A contains the upper triangular part of the matrix A, and the strictly lower triangular part of A is not referenced. If UPLO = MagmaLower, the leading n-by-n lower triangular part of A contains the lower triangular part of the matrix A, and the strictly upper triangular part of A is not referenced. On exit, A contains details of the partial factorization.
[in]ldaINTEGER The leading dimension of the array A. LDA >= max(1,N).
[in,out]dADOUBLE PRECISION array on GPU, dimension (LDDA,N) On entry, the symmetric matrix A. If UPLO = MagmaUpper, the leading n-by-n upper triangular part of A contains the upper triangular part of the matrix A, and the strictly lower triangular part of A is not referenced. If UPLO = MagmaLower, the leading n-by-n lower triangular part of A contains the lower triangular part of the matrix A, and the strictly upper triangular part of A is not referenced. On exit, A contains details of the partial factorization.
[in]lddaINTEGER The leading dimension of the array A. LDDA >= max(1,N).
[out]ipivINTEGER array, dimension (N) Details of the interchanges and the block structure of D. If UPLO = MagmaUpper, only the last KB elements of ipiv are set; if UPLO = MagmaLower, only the first KB elements are set.
If ipiv(k) > 0, then rows and columns k and ipiv(k) were interchanged and D(k,k) is a 1-by-1 diagonal block. If UPLO = MagmaUpper and ipiv(k) = ipiv(k-1) < 0, then rows and columns k-1 and -ipiv(k) were interchanged and D(k-1:k,k-1:k) is a 2-by-2 diagonal block. If UPLO = MagmaLower and ipiv(k) = ipiv(k+1) < 0, then rows and columns k+1 and -ipiv(k) were interchanged and D(k:k+1,k:k+1) is a 2-by-2 diagonal block.
[out]dW(workspace) DOUBLE PRECISION array, dimension (LDDW,NB)
[in]lddwINTEGER The leading dimension of the array W. LDDW >= max(1,N).
[in]queuesmagma_queue_t queues contain the queues used for the partial factorization. Currently, only one queue is used.
[in]eventsmagma_event_t events contain the events used for the partial factorization. Currently, only one event is used.
[out]infoINTEGER
  • = 0: successful exit
  • > 0: if INFO = k, D(k,k) is exactly zero. The factorization has been completed, but the block diagonal matrix D is exactly singular.
magma_int_t magma_slasyf_gpu ( magma_uplo_t  uplo,
magma_int_t  n,
magma_int_t  nb,
magma_int_t *  kb,
float *  hA,
magma_int_t  lda,
magmaFloat_ptr  dA,
magma_int_t  ldda,
magma_int_t *  ipiv,
magmaFloat_ptr  dW,
magma_int_t  lddw,
magma_queue_t  queues[],
magma_event_t  events[],
magma_int_t *  info 
)

SLASYF computes a partial factorization of a real symmetric matrix A using the Bunch-Kaufman diagonal pivoting method.

The partial factorization has the form:

A = ( I U12 ) ( A11 0 ) ( I 0 ) if UPLO = MagmaUpper, or: ( 0 U22 ) ( 0 D ) ( U12' U22' )

A = ( L11 0 ) ( D 0 ) ( L11' L21' ) if UPLO = MagmaLower ( L21 I ) ( 0 A22 ) ( 0 I )

where the order of D is at most NB. The actual order is returned in the argument KB, and is either NB or NB-1, or N if N <= NB. Note that U' denotes the conjugate transpose of U.

SLASYF is an auxiliary routine called by SSYTRF. It uses blocked code (calling Level 3 BLAS) to update the submatrix A11 (if UPLO = MagmaUpper) or A22 (if UPLO = MagmaLower).

Parameters
[in]uploCHARACTER Specifies whether the upper or lower triangular part of the symmetric matrix A is stored:
  • = MagmaUpper: Upper triangular
  • = MagmaLower: Lower triangular
[in]nINTEGER The order of the matrix A. N >= 0.
[in]nbINTEGER The maximum number of columns of the matrix A that should be factored. NB should be at least 2 to allow for 2-by-2 pivot blocks.
[out]kbINTEGER The number of columns of A that were actually factored. KB is either NB-1 or NB, or N if N <= NB.
[in,out]hAREAL array, dimension (LDA,N) On entry, the symmetric matrix A. If UPLO = MagmaUpper, the leading n-by-n upper triangular part of A contains the upper triangular part of the matrix A, and the strictly lower triangular part of A is not referenced. If UPLO = MagmaLower, the leading n-by-n lower triangular part of A contains the lower triangular part of the matrix A, and the strictly upper triangular part of A is not referenced. On exit, A contains details of the partial factorization.
[in]ldaINTEGER The leading dimension of the array A. LDA >= max(1,N).
[in,out]dAREAL array on GPU, dimension (LDDA,N) On entry, the symmetric matrix A. If UPLO = MagmaUpper, the leading n-by-n upper triangular part of A contains the upper triangular part of the matrix A, and the strictly lower triangular part of A is not referenced. If UPLO = MagmaLower, the leading n-by-n lower triangular part of A contains the lower triangular part of the matrix A, and the strictly upper triangular part of A is not referenced. On exit, A contains details of the partial factorization.
[in]lddaINTEGER The leading dimension of the array A. LDDA >= max(1,N).
[out]ipivINTEGER array, dimension (N) Details of the interchanges and the block structure of D. If UPLO = MagmaUpper, only the last KB elements of ipiv are set; if UPLO = MagmaLower, only the first KB elements are set.
If ipiv(k) > 0, then rows and columns k and ipiv(k) were interchanged and D(k,k) is a 1-by-1 diagonal block. If UPLO = MagmaUpper and ipiv(k) = ipiv(k-1) < 0, then rows and columns k-1 and -ipiv(k) were interchanged and D(k-1:k,k-1:k) is a 2-by-2 diagonal block. If UPLO = MagmaLower and ipiv(k) = ipiv(k+1) < 0, then rows and columns k+1 and -ipiv(k) were interchanged and D(k:k+1,k:k+1) is a 2-by-2 diagonal block.
[out]dW(workspace) REAL array, dimension (LDDW,NB)
[in]lddwINTEGER The leading dimension of the array W. LDDW >= max(1,N).
[in]queuesmagma_queue_t queues contain the queues used for the partial factorization. Currently, only one queue is used.
[in]eventsmagma_event_t events contain the events used for the partial factorization. Currently, only one event is used.
[out]infoINTEGER
  • = 0: successful exit
  • > 0: if INFO = k, D(k,k) is exactly zero. The factorization has been completed, but the block diagonal matrix D is exactly singular.
magma_int_t magma_zlahef_gpu ( magma_uplo_t  uplo,
magma_int_t  n,
magma_int_t  nb,
magma_int_t *  kb,
magmaDoubleComplex *  hA,
magma_int_t  lda,
magmaDoubleComplex_ptr  dA,
magma_int_t  ldda,
magma_int_t *  ipiv,
magmaDoubleComplex_ptr  dW,
magma_int_t  lddw,
magma_queue_t  queues[],
magma_event_t  events[],
magma_int_t *  info 
)

ZLAHEF computes a partial factorization of a complex Hermitian matrix A using the Bunch-Kaufman diagonal pivoting method.

The partial factorization has the form:

A = ( I U12 ) ( A11 0 ) ( I 0 ) if UPLO = MagmaUpper, or: ( 0 U22 ) ( 0 D ) ( U12' U22' )

A = ( L11 0 ) ( D 0 ) ( L11' L21' ) if UPLO = MagmaLower ( L21 I ) ( 0 A22 ) ( 0 I )

where the order of D is at most NB. The actual order is returned in the argument KB, and is either NB or NB-1, or N if N <= NB. Note that U' denotes the conjugate transpose of U.

ZLAHEF is an auxiliary routine called by ZHETRF. It uses blocked code (calling Level 3 BLAS) to update the submatrix A11 (if UPLO = MagmaUpper) or A22 (if UPLO = MagmaLower).

Parameters
[in]uploCHARACTER Specifies whether the upper or lower triangular part of the Hermitian matrix A is stored:
  • = MagmaUpper: Upper triangular
  • = MagmaLower: Lower triangular
[in]nINTEGER The order of the matrix A. N >= 0.
[in]nbINTEGER The maximum number of columns of the matrix A that should be factored. NB should be at least 2 to allow for 2-by-2 pivot blocks.
[out]kbINTEGER The number of columns of A that were actually factored. KB is either NB-1 or NB, or N if N <= NB.
[in,out]hACOMPLEX*16 array, dimension (LDA,N) On entry, the Hermitian matrix A. If UPLO = MagmaUpper, the leading n-by-n upper triangular part of A contains the upper triangular part of the matrix A, and the strictly lower triangular part of A is not referenced. If UPLO = MagmaLower, the leading n-by-n lower triangular part of A contains the lower triangular part of the matrix A, and the strictly upper triangular part of A is not referenced. On exit, A contains details of the partial factorization.
[in]ldaINTEGER The leading dimension of the array A. LDA >= max(1,N).
[in,out]dACOMPLEX*16 array on GPU, dimension (LDDA,N) On entry, the Hermitian matrix A. If UPLO = MagmaUpper, the leading n-by-n upper triangular part of A contains the upper triangular part of the matrix A, and the strictly lower triangular part of A is not referenced. If UPLO = MagmaLower, the leading n-by-n lower triangular part of A contains the lower triangular part of the matrix A, and the strictly upper triangular part of A is not referenced. On exit, A contains details of the partial factorization.
[in]lddaINTEGER The leading dimension of the array A. LDDA >= max(1,N).
[out]ipivINTEGER array, dimension (N) Details of the interchanges and the block structure of D. If UPLO = MagmaUpper, only the last KB elements of ipiv are set; if UPLO = MagmaLower, only the first KB elements are set.
If ipiv(k) > 0, then rows and columns k and ipiv(k) were interchanged and D(k,k) is a 1-by-1 diagonal block. If UPLO = MagmaUpper and ipiv(k) = ipiv(k-1) < 0, then rows and columns k-1 and -ipiv(k) were interchanged and D(k-1:k,k-1:k) is a 2-by-2 diagonal block. If UPLO = MagmaLower and ipiv(k) = ipiv(k+1) < 0, then rows and columns k+1 and -ipiv(k) were interchanged and D(k:k+1,k:k+1) is a 2-by-2 diagonal block.
[out]dW(workspace) COMPLEX*16 array, dimension (LDDW,NB)
[in]lddwINTEGER The leading dimension of the array W. LDDW >= max(1,N).
[in]queuesmagma_queue_t queues contain the queues used for the partial factorization. Currently, only one queue is used.
[in]eventsmagma_event_t events contain the events used for the partial factorization. Currently, only one event is used.
[out]infoINTEGER
  • = 0: successful exit
  • > 0: if INFO = k, D(k,k) is exactly zero. The factorization has been completed, but the block diagonal matrix D is exactly singular.