Swap

Applicable Products

Function Description

API function

• asdBlasMakeSwapPlan: initializes the Swap operator configuration corresponding to the handle.
• asdBlasSswap: swaps two real-number vectors.

• asdBlasCswap: swaps two complex-number vectors.

Formula:

• asdBlasSswap formula:
  • Example:

    The input x is as follows:

    [1.0, 2.0]

    The input y is as follows:

    [3.0, 4.0]

    After the Sswap operator is called, the output x is as follows:

    [3.0, 4.0]

    The output y is as follows:

    [1.0, 2.0]

• asdBlasCswap formula:
  • Example:

    The input x is as follows:

    [1.0 + 1.0i, 2.0 + 2.0i]

    The input y is as follows:

    [3.0 + 3.0i, 4.0 + 4.0i]

    After the Cswap operator is called, the output x is as follows:

    [3.0 + 3.0i, 4.0 + 4.0i]

    The output y is as follows:

    [1.0 + 1.0i, 2.0 + 2.0i]

Function Prototype

• AspbStatus asdBlasMakeSwapPlan(asdBlasHandle handle)
• AspbStatus asdBlasSswap(asdBlasHandle handle, const int64_t n, aclTensor *x, const int64_t incx, aclTensor *y, const int64_t incy)
• AspbStatus asdBlasCswap(asdBlasHandle handle, const int64_t n, aclTensor *x, const int64_t incx, aclTensor *y,const int64_t incy)

Parameter Description

• asdBlasMakeSwapPlan

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

• asdBlasSswap & asdBlasCswap

  Parameter	Input/Output	Type	Description
  handle	Input	asdBlasHandle	Handle of the Swap operator.
  n	Input	const int64_t	Number of elements in vector x (or y)
  x	Input/Output	aclTensor *	x in the formula, which is a tensor on the device. The data format can be ND, and the shape is [n]. For Sswap, the supported data type is FLOAT32. For Cswap, the supported data type is COMPLEX64.
  incx	Input	const int64_t	Memory address increment between consecutive elements of x (currently restricted to 1).
  y	Input /Output	aclTensor *	y in the formula, which is a tensor on the device. The data format can be ND, and the shape is [n]. For Sswap, the supported data type is FLOAT32. For Cswap, the supported data type is COMPLEX64.
  incy	Input	const int64_t	Memory address increment between consecutive elements of y (currently restricted to 1).

Return Value Description

For details about the return values, see Return Value.

Constraints

• asdBlasMakeSwapPlan: none.
• asdBlasSswap
  • The input and output data formats, data types, and shapes must be consistent.
  • The input element count n currently supports the range [1, 6.71e+07].
  • The input shape of the operator is [n] and [n], and the output shape is [n] and [n].
  • The operator does not support the calculation for three or more dimensions.

• asdBlasCswap
  • The input and output data formats, data types, and shapes must be consistent.
  • The input element count n currently supports the range [1, 6.71e+07].
  • The input shape of the operator is [n] and [n], and the output shape is [n] and [n].
  • The operator does not support the calculation for three or more dimensions.

Calling Example

For details about the operator calling example, see Swap.