zhiics commented on issue #4511: [CONV] Asymmetric padding URL: https://github.com/apache/incubator-tvm/pull/4511#issuecomment-571435637 Yeah, I can confirm that it fails with the following error: ``` topi/tests/python/test_topi_depthwise_conv2d_back_weight.py::test_topi_depthwise_conv2d_backward_weight_nhwc Fatal Python error: Segmentation fault ``` when I ran `docker/bash.sh tvmai/ci-gpu:v0.56 ./tests/scripts/task_python_topi.sh` on the G4 instance. I can get many this type of errors when I just run `pytest -s -v topi/tests/python`: ``` topi/tests/python/test_topi_transform.py:168: in verify_concatenate check_device(device) topi/tests/python/test_topi_transform.py:159: in check_device foo = tvm.build(s, tensor_l + [out_tensor], device, name="concatenate") python/tvm/build_module.py:638: in build fhost, mdev = _build_for_device(flist, tar, target_host) python/tvm/build_module.py:504: in _build_for_device mdev = codegen.build_module(fdevice, str(target)) if fdevice else None python/tvm/codegen.py:36: in build_module return _Build(lowered_func, target) tvm/_ffi/_cython/./function.pxi:304: in tvm._ffi._cy3.core.FunctionBase.__call__ ??? tvm/_ffi/_cython/./function.pxi:239: in tvm._ffi._cy3.core.FuncCall ??? tvm/_ffi/_cython/./function.pxi:228: in tvm._ffi._cy3.core.FuncCall3 ??? _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ > ??? E tvm._ffi.base.TVMError: Traceback (most recent call last): E [bt] (5) /home/ubuntu/workspace/tvm/build/libtvm.so(TVMFuncCall+0x61) [0x7f22b56b8c01] E [bt] (4) /home/ubuntu/workspace/tvm/build/libtvm.so(+0xb592f5) [0x7f22b4ef42f5] E [bt] (3) /home/ubuntu/workspace/tvm/build/libtvm.so(tvm::codegen::Build(tvm::Array<tvm::LoweredFunc, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)+0xb85) [0x7f22b501a835] E [bt] (2) /home/ubuntu/workspace/tvm/build/libtvm.so(std::_Function_handler<void (tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*), tvm::runtime::TypedPackedFunc<tvm::runtime::Module (tvm::Array<tvm::LoweredFunc, void>)>::AssignTypedLambda<tvm::runtime::Module (*)(tvm::Array<tvm::LoweredFunc, void>)>(tvm::runtime::Module (*)(tvm::Array<tvm::LoweredFunc, void>))::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}>::_M_invoke(std::_Any_data const&, tvm::runtime::TVMArgs&&, tvm::runtime::TVMRetValue*&&)+0x4e) [0x7f22b5045f4e] E [bt] (1) /home/ubuntu/workspace/tvm/build/libtvm.so(tvm::codegen::BuildCUDA(tvm::Array<tvm::LoweredFunc, void>)+0x3c2) [0x7f22b564ec52] E [bt] (0) /home/ubuntu/workspace/tvm/build/libtvm.so(+0x131912b) [0x7f22b56b412b] E File "tvm/_ffi/_cython/./function.pxi", line 55, in tvm._ffi._cy3.core.tvm_callback E File "/home/ubuntu/workspace/tvm/python/tvm/autotvm/measure/measure_methods.py", line 591, in tvm_callback_cuda_compile E ptx = nvcc.compile_cuda(code, target=target, arch=AutotvmGlobalScope.current.cuda_target_arch) E File "/home/ubuntu/workspace/tvm/python/tvm/contrib/nvcc.py", line 94, in compile_cuda E cmd, stdout=subprocess.PIPE, stderr=subprocess.STDOUT) E File "/home/ubuntu/anaconda3/lib/python3.6/subprocess.py", line 709, in __init__ E restore_signals, start_new_session) E File "/home/ubuntu/anaconda3/lib/python3.6/subprocess.py", line 1275, in _execute_child E restore_signals, start_new_session, preexec_fn) E OSError: [Errno 12] Cannot allocate memory ```
---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
