acldvppToDtype

Applicability

Description

Converts the data type.

Prototype

Each operator has two-phase API calls. First, acldvppToDtypeGetWorkspaceSize is called to calculate the required workspace size based on the computation process. Then, acldvppToDtype is called to perform the computation. The two-phase APIs are as follows:

• First-phase API:

  1	acldvppStatus acldvppToDtypeGetWorkspaceSize(const aclTensor *self, aclDataType dtype, bool normalize, aclTensor *out, uint64_t *workspaceSize, aclOpExecutor **executor)

• Second-phase API:

  1	acldvppStatus acldvppToDtype(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, aclrtStream stream)

acldvppToDtypeGetWorkspaceSize

• Parameters:
  • self: input tensor of the operator. aclCreateTensor needs to be called to create data of the aclTensor type (stored on the device). The dataType of the input tensor can be UINT8 or FLOAT, and the format can be NCHW or NHWC. Discontinuous tensors are not supported, in addition, N can be 1 or empty, and C can be 1 or 3 (1 indicates the input GRAY image, and 3 indicates the input RGB image).
  • dtype: data type of the conversion destination. The value can be UINT8 or FLOAT.
  • normalize: whether to perform normalization based on the data type. If the value is false, only the data type changes and the value remains unchanged. If the value is true, normalization is performed based on the data type.
  • out: operator output tensor. Call aclCreateTensor to create data of the aclTensor type (stored on the device). The data type of the output tensor can be UINT8 or FLOAT, and the format can be NCHW or NHWC. Discontinuous tensors are not supported, in addition, N can be 1 or empty, and C can be 1 or 3 (1 indicates the input GRAY image, and 3 indicates the input RGB image). The data type of out must be the same as that of self. The C, H, and W values of out must be the same as those of self.
  • workspaceSize: size of the workspace to be allocated on the device.
  • executor: operator executor, containing the operator computation process.
• Returns:

  The status code acldvppStatus is returned. For details, see acldvpp Return Codes.

acldvppToDtype

• Parameters:
  • workspace: The aclrtMalloc API needs to be called to allocate the device memory. The memory size is workspaceSize. The memory address output by the aclrtMalloc API is passed here.
  • workspaceSize: The value must be the same as that obtained by calling acldvppToDtypeGetWorkspaceSize.
  • executor: op executor, which contains the operator computation process and is the same as the executor of acldvppToDtypeGetWorkspaceSize.
  • stream: stream for executing the task. You can reuse the created stream to save resources or call aclrtCreateStream to create a stream, and then pass the stream as an input parameter.
• Returns

  The status code acldvppStatus is returned. For details, see acldvpp Return Codes.

Constraints

The supported image resolution ranges from 6 x 4 to 4096 x 8192.