AllocTensorMsgWithAlign
Applicability
Product |
Supported |
|---|---|
√ |
|
√ |
|
x |
|
x |
|
x |
Function Description
Allocates FlowMsg of the tensor type based on the shape, data type, and alignment size. The difference between this function and AllocTensorMsg is that AllocTensorMsg is 64-byte aligned by default. This function can be used to specify the alignment size, to facilitate performance optimization.
Prototype
1 | std::shared_ptr<FlowMsg> AllocTensorMsgWithAlign(const std::vector<int64_t> &shape, TensorDataType dataType, uint32_t align) |
Command-Line Options
Parameter |
Input/Output |
Description |
|---|---|---|
shape |
Input |
Shape of a tensor. |
dataType |
Input |
Data type of a tensor. |
align |
Input |
Alignment size for the allocated memory address. The value range is [32, 64, 128, 256, 512, 1024]. |
Returns
Allocated tensor pointer.
Troubleshooting
If no tensor pointer can be allocated, NULL is returned.
Constraints
None