jetson 编译支持显卡的版本，是否需要加入tensorrt 支持

[captain0005]

我在使用jetson orin nx 的arm上使用，想将inspireface 编译一个支持显卡的版本，增加了全局cuda支持，但是实际使用时并没有使用到gpu，而且还提示缺少tensorrt。请问如果在该arm环境上使用gpu，还需要增加tensorrt编译选项么？

一下提示是我调用库输出的提示
== Pikachu v0.1.4 ==
[any_net.h][72]: You have forced the global use of MNN_CUDA as the neural network inference backend
[inference_helper_mnn.cpp][Initialize][168]: Enable CUDA
The device support i8sdot:1, support fp16:1, support i8mm: 0
Can't Find type=2 backend, use 0 instead
Can't Find type=2 backend, use 0 instead
[any_net.h][72]: You have forced the global use of MNN_CUDA as the neural network inference backend
[inference_helper_mnn.cpp][Initialize][168]: Enable CUDA
Can't Find type=2 backend, use 0 instead
Can't Find type=2 backend, use 0 instead
[any_net.h][72]: You have forced the global use of MNN_CUDA as the neural network inference backend
[inference_helper_mnn.cpp][Initialize][168]: Enable CUDA
Can't Find type=2 backend, use 0 instead
Can't Find type=2 backend, use 0 instead
[any_net.h][72]: You have forced the global use of MNN_CUDA as the neural network inference backend
[inference_helper_mnn.cpp][Initialize][168]: Enable CUDA
Can't Find type=2 backend, use 0 instead
Can't Find type=2 backend, use 0 instead