Ssyr

Applicable Products

Function Description

API function

• asdBlasMakeSsyrPlan: initializes the Ssyr operator configuration corresponding to the handle.

• asdBlasSsyr: computes the outer product of single-precision vectors and adds the result to a matrix.

Formula:

• asdBlasSsyr formula:
• Example:

  The input x is as follows:

  [   [ 1 ], 
      [ 2 ] ]

  The input A is as follows:

  [   [ 1, 2 ], 
      [ 3, 4 ]  ]

  The input uplo is U.

  The input n is 2, the input lda is 2, and the input incx is 1.

  Input alpha: 2.0

  :

   [ 1, 2 ]

  :

   [ 2, 4 ]

  :

  [ [ 2 ], [ 4 ] ] × [ 1, 2 ] =  [ [ 2, 4 ], [ 4, 8 ] ]

  After the Ssyr operator is called, the upper triangular data is updated, and the output A is as follows:

  [   [ 3, 6 ], 
      [ 3, 12 ]  ].

Function Prototype

• AspbStatus asdBlasMakeSsyrPlan(asdBlasHandle handle)
• AspbStatus asdBlasSsyr(asdBlasHandle handle, asdBlasFillMode_t uplo, const int64_t n, const float *alpha, aclTensor *x, const int64_t incx, aclTensor *A, const int64_t lda)

Parameter Description

• asdBlasMakeSsyrPlan

  Parameter	Input/Output	Type	Description
  handle	Input	asdBlasHandle	Handle of the Ssyr operator.

• asdBlasSsyr

  Parameter	Input/Output	Type	Description
  handle	Input	asdBlasHandle	Handle of the Ssyr operator.
  uplo	Input	asdBlasFillMode_t	Triangular region of matrix A involved in the computation. ASDBLAS_FILL_MODE_LOWER // Lower triangle ASDBLAS_FILL_MODE_UPPER // Upper triangle
  n	Input	const int64_t	Number of elements in vector x, that is, the number of rows and columns of matrix A.
  alpha	Input	const float *	alpha, the scalar in the formula that scales the vector product.
  x	Input	aclTensor *	x in the formula, which is a tensor on the device. The data type can be FLOAT32, the data format can be ND, and the shape is [n].
  incx	Input	const int64_t	Memory address increment between consecutive elements of x (currently restricted to 1).
  A	Input/Output	aclTensor *	A in the formula, which is a tensor on the device. The data type can be FLOAT32, the data format can be ND, and the shape is [n, n].
  lda	Input	const int64_t	Storage stride of elements in each column of matrix A (currently constrained to n).

Return Value Description

For details about the return values, see Return Value.

Constraints

• asdBlasMakeSsyrPlan: none.

• asdBlasSsyr
  • The input element count n currently supports the range [1, 16384].
  • The input shape is [n] or [n, n], and the output shape is [n, n].
  • The operator does not support the calculation for three or more dimensions.

Calling Example

For details about the operator calling example, see Ssyr.