Function: malloc

Applicability

Product	Supported
Atlas 350 Accelerator Card	√
Atlas A3 training product / Atlas A3 inference product	√
Atlas A2 training product / Atlas A2 inference product	√
Atlas training product	√
Atlas inference product	√
Atlas 200I/500 A2 inference product	√

Description

Allocates linear memory of the size on the device and returns the pointer to the allocated memory through dev_ptr. The start address of the memory is 64-byte aligned. The memory allocated by this API needs to be freed by calling acl.rt.free.

The allocation size is the input size rounded up to the nearest multiple of 32 bytes, plus 32 bytes. However, for huge pages with the memory allocation granularity of 1 GB, to save huge pages, this API only rounds up the size requested by users to an integer multiple of 32 bytes, but does not add an extra 32 bytes.

Prototype

C Prototype

        
             aclError aclrtMalloc(void **devPtr, size_t size, aclrtMemMallocPolicy policy)

Python Function

        
             dev_ptr, ret = acl.rt.malloc(size, policy)

Parameters

Parameter	Description
size	Int, size of the memory to be allocated, in bytes. The value cannot be 0.
policy	Int, memory allocation policy. If the configured memory allocation rule exceeds the value range of aclrtMemMallocPolicy and size is greater than or equal to 2 MB, memory is allocated based on huge pages. Otherwise, memory is allocated based on normal pages.

Parameter

Description

size

Int, size of the memory to be allocated, in bytes. The value cannot be 0.

policy

Int, memory allocation policy.

If the configured memory allocation rule exceeds the value range of aclrtMemMallocPolicy and size is greater than or equal to 2 MB, memory is allocated based on huge pages. Otherwise, memory is allocated based on normal pages.

Return Value

Return Value	Description
dev_ptr	Int, address of the pointer to the allocated device memory.
ret	Int, error code. 0 on success; else, failure.

Restrictions

Media data processing has higher requirements on the memory for storing the input and output data (for example, the start address of the memory must be 128-byte aligned). Therefore, the following dedicated memory allocation APIs are required:
- When the media data processing V1 API is called to perform operations such as cropping and resizing on images, the acl.media.dvpp_malloc API is called to allocate memory.
- When the media data processing V2 API is called to perform operations such as cropping and resizing on images, the acl.himpi.dvpp_malloc API is called to allocate memory.
The memory allocated by this API is not initialized. Call acl.rt.memset to initialize the memory and clear random numbers in the memory before using the memory.
This API does not perform implicit device synchronization or stream synchronization. The memory allocation result, either success or failure, is returned immediately.
Memory allocated by the acl.rt.malloc call needs to be freed by the acl.rt.free call.
Performance deterioration will be caused by the frequent calling of acl.rt.malloc to allocate memory and acl.rt.free to free memory. You are advised to allocate or manage memory in advance to avoid frequent memory allocation and deallocation.
The allocation size of acl.rt.malloc is the input size rounded up to the nearest multiple of 32 bytes, plus 32 bytes.
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 64-byte aligned (ALIGN_UP[m,64]).
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) x k.

Parent topic: Memory Management