When measuring program using ProgramMeasurer, I met with this error msg:
Exception ignored in: <function NDArrayBase.del at 0x7f05c632d260> Traceback (most recent call last): File “/home/tonylyc/tenset/python/tvm/_ffi/_ctypes/ndarray.py”, line 82, in del check_call(_LIB.TVMArrayFree(self.handle)) File “/home/tonylyc/tenset/python/tvm/_ffi/base.py”, line 346, in check_call raise get_last_ffi_error() tvm._ffi.base.TVMError: Traceback (most recent call last): [bt] (4) /home/tonylyc/tenset/build/libtvm.so(TVMArrayFree+0x30) [0x7f05c5c861d0] [bt] (3) /home/tonylyc/tenset/build/libtvm.so(tvm::runtime::NDArray::Internal::DefaultDeleter(tvm::runtime::Object*)+0x5b) [0x7f05c5c8c52b] [bt] (2) /home/tonylyc/tenset/build/libtvm.so(tvm::runtime::CUDADeviceAPI::FreeDataSpace(DLContext, void*)+0x8bf) [0x7f05c5d1972f] [bt] (1) /home/tonylyc/tenset/build/libtvm.so(tvm::runtime::detail::LogFatal::~LogFatal()+0x45) [0x7f05c4a5bd41] [bt] (0) /home/tonylyc/tenset/build/libtvm.so(tvm::runtime::Backtraceabi:cxx11+0x2c) [0x7f05c5c758ec] File “/home/tonylyc/tenset/src/runtime/cuda/cuda_device_api.cc”, line 127 TVMError: --------------------------------------------------------------- An internal invariant was violated during the execution of TVM. Please read TVM’s error reporting guidelines. More details can be found here: https://discuss.tvm.ai/t/error-reporting/7793.
Check failed: e == cudaSuccess || e == cudaErrorCudartUnloading == false: CUDA: misaligned address
Ubuntu20.04 RTX 4090 CUDA 11.8