TensorRT 10.0.1
IGpuAsyncAllocator Class Reference

Application-implemented class for controlling asynchronous (stream ordered) memory allocation on the GPU. More...

#include <NvInferRuntime.h>

Detailed Description

Application-implemented class for controlling asynchronous (stream ordered) memory allocation on the GPU.

Warning
The lifetime of an IGpuAsyncAllocator object must exceed that of all objects that use it.

The advantage of deriving from IGpuAsyncAllocator instead of IGpuAllocator is that you only have to override two methods: allocateAsync() and deallocateAsync() to implement an allocator with asynchronous capability, whereas deriving from IGpuAllocator requires overriding four methods, including two deprecated methods.

See also
IGpuAllocator

The documentation for this class was generated from the following file:

  Copyright © 2024 NVIDIA Corporation
  Privacy Policy | Manage My Privacy | Do Not Sell or Share My Data | Terms of Service | Accessibility | Corporate Policies | Product Security | Contact