hemm#
Computes a matrix-matrix product where one input matrix is hermitian and one is general.
Description#
The hemm routines compute a scalar-matrix-matrix product and add the result to a scalar-matrix product, where one of the matrices in the multiplication is hermitian. The argument left_right determines if the hermitian matrix, A, is on the left of the multiplication (left_right = side::left) or on the right (left_right = side::right). The operation is defined as:
If (left_right = side::left),
If (left_right = side::right),
where:
alphaandbetaare scalarsAis eithermxmornxnhermitian matrixBandCaremxnmatrices
hemm supports the following precisions:
T |
|---|
|
|
hemm (Buffer Version)#
Syntax#
namespace oneapi::mkl::blas::column_major {
void hemm(sycl::queue &queue,
oneapi::mkl::side left_right,
oneapi::mkl::uplo upper_lower,
std::int64_t m,
std::int64_t n,
T alpha,
sycl::buffer<T,1> &a,
std::int64_t lda,
sycl::buffer<T,1> &b,
std::int64_t ldb,
T beta,
sycl::buffer<T,1> &c,
std::int64_t ldc,
compute_mode mode = compute_mode::unset)
}
namespace oneapi::mkl::blas::row_major {
void hemm(sycl::queue &queue,
oneapi::mkl::side left_right,
oneapi::mkl::uplo upper_lower,
std::int64_t m,
std::int64_t n,
T alpha,
sycl::buffer<T,1> &a,
std::int64_t lda,
sycl::buffer<T,1> &b,
std::int64_t ldb,
T beta,
sycl::buffer<T,1> &c,
std::int64_t ldc,
compute_mode mode = compute_mode::unset)
}
Input Parameters#
- queue
The queue where the routine should be executed.
- left_right
Specifies whether matrix
Ais on the left side or right side of the multiplication. See Data Types for more details.- upper_lower
Specifies whether matrix
Ais upper or lower triangular. See Data Types for more details.- m
Number of rows of matrix
Band matrixC. Must be at least zero.- n
Number of columns of matrix
Band matrixC. Must be at least zero.- alpha
Scaling factor for matrix-matrix product.
- a
Buffer holding input matrix
A. Size of the buffer must be at leastlda*mifleft_right=side::leftorlda*nifleft_right=side::right. See Matrix Storage for more details.- lda
Leading dimension of matrix
A. Must be at leastmifleft_right=side::leftor at leastnifleft_right=side::right. Must be positive.- b
Buffer holding input matrix
B. Size of the buffer must be at leastldb*nif column major layout or at leastldb*mif row major layout is used. See Matrix Storage for more details.- ldb
Leading dimension of matrix
B. Must be at leastmif column major layout or at leastnif row major layout is used. Must be positive.- beta
Scaling factor for matrix
C.- c
Buffer holding input/output matrix
C. Size of the buffer must be at leastldc*nif column major layout or at leastldc*mif row major layout is used. See Matrix Storage for more details.- ldc
Leading dimension of matrix
C. Must be at leastmif column major layout or at leastnif row major layout is used. Must be positive.- mode
Optional. Compute mode settings. See Compute Modes for more details.
Output Parameters#
- c
Output buffer overwritten by
alpha*A*B+beta*Cifleft_right=side::leftoralpha*B*A+beta*Cifleft_right=side::right.
Note
If beta = 0, matrix C does not need to be initialized before calling hemm.
hemm (USM Version)#
Syntax#
namespace oneapi::mkl::blas::column_major {
sycl::event hemm(sycl::queue &queue,
oneapi::mkl::side left_right,
oneapi::mkl::uplo upper_lower,
std::int64_t m,
std::int64_t n,
oneapi::mkl::value_or_pointer<T> alpha,
const T *a,
std::int64_t lda,
const T *b,
std::int64_t ldb,
oneapi::mkl::value_or_pointer<T> beta,
T *c,
std::int64_t ldc,
compute_mode mode = compute_mode::unset,
const std::vector<sycl::event> &dependencies = {})
}
namespace oneapi::mkl::blas::row_major {
sycl::event hemm(sycl::queue &queue,
oneapi::mkl::side left_right,
oneapi::mkl::uplo upper_lower,
std::int64_t m,
std::int64_t n,
oneapi::mkl::value_or_pointer<T> alpha,
const T *a,
std::int64_t lda,
const T *b,
std::int64_t ldb,
oneapi::mkl::value_or_pointer<T> beta,
T *c,
std::int64_t ldc,
compute_mode mode = compute_mode::unset,
const std::vector<sycl::event> &dependencies = {})
}
Input Parameters#
- queue
The queue where the routine should be executed.
- left_right
Specifies whether matrix
Ais on the left side or right side of the multiplication. See Data Types for more details.- upper_lower
Specifies whether matrix
Ais upper or lower triangular. See Data Types for more details.- m
Number of rows of matrix
Band matrixC. Must be at least zero.- n
Number of columns of matrix
Band matrixC. Must be at least zero.- alpha
Scaling factor for matrix-matrix product. See Scalar Arguments for more information on the
value_or_pointerdata type.- a
Pointer to input matrix
A. Size of the array must be at leastlda*mifleft_right=side::leftorlda*nifleft_right=side::right. See Matrix Storage for more details.- lda
Leading dimension of matrix
A. Must be at leastmifleft_right=side::leftor at leastnifleft_right=side::right. Must be positive.- b
Pointer to input matrix
B. Size of the array must be at leastldb*nif column major layout or at leastldb*mif row major layout is used. See Matrix Storage for more details.- ldb
Leading dimension of matrix
B. Must be at leastmif column major layout or at leastnif row major layout is used. Must be positive.- beta
Scaling factor for matrix
C. See Scalar Arguments for more information on thevalue_or_pointerdata type.- c
Pointer to input/output matrix
C. Size of the array must be at leastldc*nif column major layout or at leastldc*mif row major layout is used. See Matrix Storage for more details.- ldc
Leading dimension of matrix
C. Must be at leastmif column major layout or at leastnif row major layout is used. Must be positive.- mode
Optional. Compute mode settings. See Compute Modes for more details.
- dependencies
Optional. List of events to wait for before starting computation, if any. If omitted, defaults to no dependencies.
modeanddependenciesmay be omitted independently; it is not necessary to specifymodein order to provide dependencies.
Output Parameters#
- c
Pointer to output matrix overwritten by
alpha*A*B+beta*Cifleft_right=side::leftoralpha*B*A+beta*Cifleft_right=side::right.
Note
If beta = 0, matrix C does not need to be initialized before calling hemm.
Return Values#
Output event to wait on to ensure computation is complete.