search

Zhou-sx / yolov5_Deepsort_rknn

Track vehicles and persons on rk3588 / rk3399pro.
315 stars 78 forks source link

3399pro开发板上遇到问题undefined reference to `log@GLIBC_2.29' #5

Open yexiayi opened 2 years ago

yexiayi commented 2 years ago

在3399pro开发板上尝试编译，sudo cmake ..后，sudo make -j2出现如下错误

fc004f1f28555a7dd728be4a8f8680a

目前尝试过将glibc从2.28升到2.29，未解决该问题。

ljdongysu commented 2 years ago

换一个librga.so进行尝试。 但是编译通过后运行报错： Bind NPU process on CPU 4 E RKNN: [07:01:31.700] failed to check rknpu hardware version: 0 E RKNN: [07:01:31.700] The rknn model does not match the RKNPU hardware version! E RKNN: [07:01:31.700] rknn_init, load model failed! rknn_init fail! ret=-1

yexiayi commented 2 years ago

换一个librga.so进行尝试。 但是编译通过后运行报错： Bind NPU process on CPU 4 E RKNN: [07:01:31.700] failed to check rknpu hardware version: 0 E RKNN: [07:01:31.700] The rknn model does not match the RKNPU hardware version! E RKNN: [07:01:31.700] rknn_init, load model failed! rknn_init fail! ret=-1

我换掉librga.so报了另一个错，不知道前面的问题解决没有。 image

Zhou-sx commented 2 years ago

换一个librga.so进行尝试。 但是编译通过后运行报错： Bind NPU process on CPU 4 E RKNN: [07:01:31.700] failed to check rknpu hardware version: 0 E RKNN: [07:01:31.700] The rknn model does not match the RKNPU hardware version! E RKNN: [07:01:31.700] rknn_init, load model failed! rknn_init fail! ret=-1

我怀疑是因为rk3399pro 不支持 rknn-toolkit2

guofenggitlearning commented 2 years ago

在现有环境下编译一下opencv，我也遇到过类似问题，与opencv有关。

johnjin404 commented 10 months ago

换个编译器，我在3588上同样的问题解决了，用官方的 https://gitlab.com/firefly-linux/prebuilts/gcc/linux-x86/aarch64/gcc-buildroot-9.3.0-2020.03-x86_64_aarch64-rockchip-linux-gnu