Xavier(cuda-10):error == cudaSuccess (48 vs. 0) no kernel image is available for execution on the d...

*** Check failure stack trace: ***
@ 0x7f84fd0128 google::LogMessage::Fail()
@ 0x7f84fd1f98 google::LogMessage::SendToLog()
@ 0x7f84fcfc90 google::LogMessage::Flush()
@ 0x7f84fd283c google::LogMessageFatal::~LogMessageFatal()
@ 0x7f85448f6c caffe::ReLULayer<>::Forward_gpu()
@ 0x7f8542284c caffe::CuDNNReLULayer<>::Forward_gpu()
@ 0x7f853988d4 caffe::Net<>::ForwardFromTo()
@ 0x7f85398a08 caffe::Net<>::Forward()
@ 0x7f8dd15018 caffe::Net<>::ForwardPrefilled()
@ 0x7f8dd137a4 VapdClassify::yolo_detect_for_fuseImg()

error in my caffe demo.
what should i do?

Hi,

Have you build the caffe library with the Xavier architecture.

For example, adding the 72 architecture into Makefile.config:
https://github.com/BVLC/caffe/blob/master/Makefile.config.example#L39

CUDA_ARCH := -gencode arch=compute_20,code=sm_20 \
		...
		-gencode arch=compute_61,code=sm_61 \
<b>		-gencode arch=compute_72,code=sm_72 \</b>
		-gencode arch=compute_61,code=compute_61

After updating, please recompile the library.
Thanks.

hi,this is my makefile.config,and i will try:
CUDA_ARCH := -gencode arch=compute_30,code=sm_30
-gencode arch=compute_35,code=sm_35
-gencode arch=compute_50,code=sm_50
-gencode arch=compute_52,code=sm_52
-gencode arch=compute_61,code=sm_61
-gencode arch=compute_70,code=sm_70

thank you!