LeiWang1999 commented on PR #18319: URL: https://github.com/apache/tvm/pull/18319#issuecomment-3418237369
> the last failed case `tests/python/relax/test_codegen_cudnn.py::test_stacked_attention_split_offload[stacked_attention_size0] SKIPPED (require cudnn frontend) ` skipped on my local environment, and when I turn on `set(USE_CUDNN_FRONTEND ON)`, I came across: > > ```shell > CMake Error at cmake/modules/CUDA.cmake:89 (message): > Cannot find cudnn_frontend.h, please set USE_CUDNN_FRONTEND to the path of > the cuDNN frontend header > Call Stack (most recent call first): > CMakeLists.txt:457 (include) > ``` install https://github.com/NVIDIA/cudnn-frontend locally can fix this issue -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
