hi_mpi_dvpp_malloc
Applicability
|
Product |
Supported |
|---|---|
|
Atlas 350 Accelerator Card |
√ |
|
|
√ |
|
|
√ |
|
|
√ |
|
|
√ |
|
|
☓ |
Function Description
Allocates device memory. The requested allocation must meet the media data processing requirements (for example,128-byte alignment of the start address). The allocation size is the input size rounded up to the nearest multiple of 32, plus 32 bytes. After this API is called to apply for memory, the lifecycle of the memory is managed by the user. If the memory is not used, the hi_mpi_dvpp_free API must be called to release the memory in a timely manner.
Parameters
|
Parameter |
Input/Output |
Description |
|---|---|---|
|
dev_id |
Input |
Device ID. This parameter is reserved and invalid. |
|
dev_ptr |
Output |
Pointer to the pointer to the allocated device memory |
|
size |
Input |
Allocated memory size, in bytes. |
Returns
- 0: success
- Other values: failure
Restrictions
- Memory allocated by this API call is accessible only to the calling process and cannot be shared among processes.
- Media data processing can access a maximum of 16 GB address space in each process.
This restriction applies to the following products:
Atlas A3 training product /Atlas A3 inference product Atlas A2 training product /Atlas A2 inference product Atlas inference product - When the output of media data processing is used as the input of model inference, if you use this API to allocate a large memory block and divide and manage the memory, each memory segment must meet the following requirements:
- The memory size is rounded up to the nearest multiple of 32 plus 32 bytes (m = ALIGN_UP[len,32] + 32 bytes).
- The memory start address must be 128-byte aligned (ALIGN_UP[m,128]).
len indicates the size of a memory segment. ALIGN_UP[len,k] indicates rounding up to a multiple of k bytes as in this formula: ((len – 1)/k + 1) × k.