The compilation passed, but an error was reported in test phase

[hanhanhanhanhana]

What are the problems?(screenshots or detailed error messages)

Use ./build.sh -DPPLNN_USE_CUDA=ON to compile successfully, but use ./pplnn-build/tools/pplnn --use-cuda --onnx-model tests/testdata/conv.onnx to report an error Segmentation fault (core dumped )

What are the types of GPU/CPU you are using?

GPU

What's the operating system ppl.nn runs on?

Linux

What's the compiler and its version?

nvcc version:11.0

Which version(commit id or tag) of ppl.nn is used?

7d63ace

What are the commands used to build ppl.nn?

./build.sh -DPPLNN_USE_CUDA=ON

What are the execution commands?

./pplnn-build/tools/pplnn --use-cuda --onnx-model tests/testdata/conv.onnx

minimal code snippets for reproducing these problems(if necessary)

models and inputs for reproducing these problems (send them to [email protected] if necessary)

[hanhanhanhanhana]

The gpu type is p100

[jianfei-wangg]

The gpu type is p100

pplnn only support nv-gpu with sm > 7.5 now. P100 is sm_60, so segment fault occured. We now add the arch check and reminders for unsupported gpus with OpenPPL/ppl.kernel.cuda#25