gemv#
Computes a matrix-vector product using a general matrix.
Description#
The gemv routines compute a scalar-matrix-vector product and add the result to a scalar-vector product, with a general matrix.
The operation is defined as:
where:
op(
A) is one of op(A) =A, or op(A) =AT, or op(A) =AHalphaandbetaare scalarsAismxnmatrixxandyare vectors
gemv supports the following precisions:
T |
|---|
|
|
|
|
gemv (Buffer Version)#
Syntax#
namespace oneapi::mkl::blas::column_major {
void gemv(sycl::queue &queue,
oneapi::mkl::transpose trans,
std::int64_t m,
std::int64_t n,
T alpha,
sycl::buffer<T,1> &a,
std::int64_t lda,
sycl::buffer<T,1> &x,
std::int64_t incx,
T beta,
sycl::buffer<T,1> &y,
std::int64_t incy)
}
namespace oneapi::mkl::blas::row_major {
void gemv(sycl::queue &queue,
oneapi::mkl::transpose trans,
std::int64_t m,
std::int64_t n,
T alpha,
sycl::buffer<T,1> &a,
std::int64_t lda,
sycl::buffer<T,1> &x,
std::int64_t incx,
T beta,
sycl::buffer<T,1> &y,
std::int64_t incy)
}
Input Parameters#
- queue
The queue where the routine should be executed.
- trans
Specifies op(
A), the transposition operation applied to matrixA. See Data Types for more details.- m
Number of rows of matrix
A. Must be at least zero.- n
Number of columns of matrix
A. Must be at least zero.- alpha
Scaling factor for the matrix-vector product.
- a
Buffer holding input matrix
A. Size of the buffer must be at leastlda*nif column major layout is used, or at leastlda*mif row major layout is used. See Matrix Storage for more details.- lda
Leading dimension of matrix
A. Must be positive and at leastmif column major layout is used or at leastnif row major layout is used.- x
Buffer holding input vector
x. The lengthlenof vectorxisnifAis not transposed, andmifAis transposed. Size of the buffer must be at least (1 + (len- 1)*abs(incx)). See Matrix Storage for more details.- incx
Stride of vector
x. Must not be zero.- beta
Scaling factor for vector
y. Ifbeta= 0, vectorydoes not need to be initialized before callinggemv.- y
Buffer holding input/output vector
y. The lengthlenof vectoryism, ifAis not transposed, andnifAis transposed. Size of the buffer must be at least (1 + (len- 1)*abs(incy)). See Matrix Storage for more details.- incy
Stride of vector
y. Must not be zero.
Output Parameters#
- y
Buffer holding updated vector
y.
Examples#
An example of how to use gemv can be found in the oneMKL installation directory, under:
share/doc/mkl/examples/sycl/blas/source/gemv.cpp
gemv (USM Version)#
Syntax#
namespace oneapi::mkl::blas::column_major {
sycl::event gemv(sycl::queue &queue,
oneapi::mkl::transpose trans,
std::int64_t m,
std::int64_t n,
oneapi::mkl::value_or_pointer<T> alpha,
const T *a,
std::int64_t lda,
const T *x,
std::int64_t incx,
oneapi::mkl::value_or_pointer<T> beta,
T *y,
std::int64_t incy,
const std::vector<sycl::event> &dependencies = {})
}
namespace oneapi::mkl::blas::row_major {
sycl::event gemv(sycl::queue &queue,
oneapi::mkl::transpose trans,
std::int64_t m,
std::int64_t n,
oneapi::mkl::value_or_pointer<T> alpha,
const T *a,
std::int64_t lda,
const T *x,
std::int64_t incx,
oneapi::mkl::value_or_pointer<T> beta,
T *y,
std::int64_t incy,
const std::vector<sycl::event> &dependencies = {})
}
Input Parameters#
- queue
The queue where the routine should be executed.
- trans
Specifies op(
A), the transposition operation applied to matrixA. See Data Types for more details.- m
Number of rows of matrix
A. Must be at least zero.- n
Number of columns of matrix
A. Must be at least zero.- alpha
Scaling factor for the matrix-vector 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*nif column major layout is used, or at leastlda*mif row major layout is used.- lda
Leading dimension of matrix
A. Must be positive and at leastmif column major layout is used or at leastnif row major layout is used.- x
Pointer to the input vector
x. The lengthlenof vectorxisnifAis not transposed, andmifAis transposed. Size of the array holding vectorxmust be at least (1 + (len- 1)*abs(incx)). See Matrix Storage for more details.- incx
Stride of vector
x. Must not be zero.- beta
Scaling factor for vector
y. Ifbeta= 0, vectorydoes not need to be initialized before callinggemv. See Scalar Arguments for more information on thevalue_or_pointerdata type.- y
Pointer to input/output vector
y. The lengthlenof vectoryism, ifAis not transposed, andnifAis transposed. Size of the array holding input/output vectorymust be at least (1 + (len- 1)*abs(incy)). See Matrix Storage for more details.- incy
Stride of vector
y. Must not be zero.- dependencies
List of events to wait for before starting computation, if any. If omitted, defaults to no dependencies.
Output Parameters#
- y
Pointer to updated vector
y.
Return Values#
Output event to wait on to ensure computation is complete.