search

enazoe / yolo-tensorrt

TensorRT8.Support Yolov5n,s,m,l,x .darknet -> tensorrt. Yolov4 Yolov3 use raw darknet *.weights and *.cfg fils. If the wrapper is useful to you,please Star it.
MIT License
1.18k stars 313 forks source link

unspecified launch failure #59

Open sdimantsd opened 3 years ago

sdimantsd commented 3 years ago

Hi! Thanks for you'r work!! I used you'r repo with tiny yolo, and sometimes i get the following error: Cuda failure: unspecified launch failure in file /home/ws/projects/yolo-trt/yolo.cpp at line 924

Im using Jetson Xavier with cuda 10.0 Thanks!

enazoe commented 3 years ago

which version of your jetpack and tensorrt？

sdimantsd commented 3 years ago 
$ python jetsonInfo.py
NVIDIA Jetson AGX Xavier [16GB]
 L4T 32.3.1 [ JetPack 4.3 ]
   Ubuntu 18.04.3 LTS
   Kernel Version: 4.9.140-tegra
nvdc: nvdcEventThread select failed: Bad file descriptor
 CUDA 10.0.326
   CUDA Architecture: 7.2
 OpenCV version: 3.4.0
   OpenCV Cuda: NO
 CUDNN: 7.6.3.28
 TensorRT: 6.0.1.10
 Vision Works: 1.6.0.500n
 VPI: 0.1.0

Thanks

enazoe commented 3 years ago

make sure have enough hard disk space when generate the engin file. And upgrade to jetpack4.4 ？I am not test it on jetpack4,3.

sdimantsd commented 3 years ago

OK, i will try on jetpack4.4. The engine created and it's running. but somtimes it's crush...

sdimantsd commented 3 years ago

Hi again :-) Now im running the file with jetpack4.4

$ python jetsonInfo.py NVIDIA Jetson Nano (Developer Kit Version) L4T 32.4.3 [ JetPack 4.4 ] Ubuntu 18.04.5 LTS Kernel Version: 4.9.140-tegra CUDA 10.2.89 CUDA Architecture: 5.3 OpenCV version: 3.4.0 OpenCV Cuda: NO CUDNN: 8.0.0.180 TensorRT: 7.1.3.0 Vision Works: 1.6.0.501 VPI: 4.4-b186

Somtimes m getting this error: ERROR: FAILED_EXECUTION: std::exception Cuda failure: unspecified launch failure in file /home/ws/projects/yolo-trt/yolo.cpp at line 931

sdimantsd commented 3 years ago

OK, it's probably by mistake... I used the GPU with many networks on the same time.