waytrue17 opened a new issue #20691: URL: https://github.com/apache/incubator-mxnet/issues/20691
## Description When a model contains `concat`, `slice_axis` and `broadcast_lesser_equal`, it fails the backward pass with segmentation fault. It only failed with non-hybridized model. Tested on MXNet 1.8.0 ### Error Message ``` Thread 1 "python" received signal SIGSEGV, Segmentation fault. 0x00007ffdd132e401 in mxnet::imperative::SetDependency(nnvm::NodeAttrs const&, mxnet::Context const&, std::vector<mxnet::NDArray*, std::allocator<mxnet::NDArray*> > const&, std::vector<mxnet::NDArray*, std::allocator<mxnet::NDArray*> > const&, std::vector<mxnet::engine::Var*, std::allocator<mxnet::engine::Var*> >*, std::vector<mxnet::engine::Var*, std::allocator<mxnet::engine::Var*> >*, std::vector<mxnet::Resource, std::allocator<mxnet::Resource> >*, std::vector<unsigned int, std::allocator<unsigned int> >*, mxnet::DispatchMode) () from /home/ubuntu/workspace/mxnet/python/mxnet/../../build/libmxnet.so (gdb) bt #0 0x00007ffdd132e401 in mxnet::imperative::SetDependency(nnvm::NodeAttrs const&, mxnet::Context const&, std::vector<mxnet::NDArray*, std::allocator<mxnet::NDArray*> > const&, std::vector<mxnet::NDArray*, std::allocator<mxnet::NDArray*> > const&, std::vector<mxnet::engine::Var*, std::allocator<mxnet::engine::Var*> >*, std::vector<mxnet::engine::Var*, std::allocator<mxnet::engine::Var*> >*, std::vector<mxnet::Resource, std::allocator<mxnet::Resource> >*, std::vector<unsigned int, std::allocator<unsigned int> >*, mxnet::DispatchMode) () at /home/ubuntu/workspace/mxnet/python/mxnet/../../build/libmxnet.so #1 0x00007ffdd1317050 in mxnet::Imperative::InvokeOp(mxnet::Context const&, nnvm::NodeAttrs const&, std::vector<mxnet::NDArray*, std::allocator<mxnet::NDArray*> > const&, std::vector<mxnet::NDArray*, std::allocator<mxnet::NDArray*> > const&, std::vector<mxnet::OpReqType, std::allocator<mxnet::OpReqType> > const&, mxnet::DispatchMode, mxnet::OpStatePtr) () at /home/ubuntu/workspace/mxnet/python/mxnet/../../build/libmxnet.so #2 0x00007ffdd132fae7 in std::_Function_handler<void (mxnet::OpStatePtr const&), mxnet::imperative::RunGraph(bool, nnvm::IndexedGraph const&, std::vector<mxnet::NDArray*, std::allocator<mxnet::NDArray*> > const&, unsigned long, unsigned long, std::vector<mxnet::OpReqType, std::allocator<mxnet::OpReqType> >&&, std::vector<unsigned int, std::allocator<unsigned int> >&&, std::vector<mxnet::OpStatePtr, std::allocator<mxnet::OpStatePtr> >*, std::vector<mxnet::DispatchMode, std::allocator<mxnet::DispatchMode> > const&, bool, std::vector<mxnet::TShape, std::allocator<mxnet::TShape> >*, std::function<void (char const*, char const*, void*)> const&, bool)::{lambda(mxnet::OpStatePtr const&)#1}>::_M_invoke(std::_Any_data const&, mxnet::OpStatePtr const&) () at /home/ubuntu/workspace/mxnet/python/mxnet/../../build/libmxnet.so #3 0x00007ffdd132fe62 in (anonymous namespace)::InvokeOperator(nnvm::IndexedGraph const&, int, bool, std::vector<mxnet::NDArray*, std::allocator<mxnet::NDArray*> > const&, mxnet::Context, std::vector<mxnet::OpStatePtr, std::allocator<mxnet::OpStatePtr> >*, std::vector<mxnet::NDArray*, std::allocator<mxnet::NDArray*> > const&, std::vector<mxnet::NDArray*, std::allocator<mxnet::NDArray*> > const&, std::vector<mxnet::OpReqType, std::allocator<mxnet::OpReqType> >*, std::vector<unsigned int, std::allocator<unsigned int> >*, std::function<void (mxnet::OpStatePtr const&)>) () at /home/ubuntu/workspace/mxnet/python/mxnet/../../build/libmxnet.so #4 0x00007ffdd1331227 in mxnet::imperative::RunGraph(bool, nnvm::IndexedGraph const&, std::vector<mxnet::NDArray*, std::allocator<mxnet::NDArray*> > const&, unsigned long, unsigned long, std::vector<mxnet::OpReqType, std::allocator<mxnet::OpReqType> >&&, std::vector<unsigned int, std::allocator<unsigned int> >&&, std::vector<mxnet::OpStatePtr, std::allocator<mxnet::OpStatePtr> >*, std::vector<mxnet::DispatchMode, std::allocator<mxnet::DispatchMode> > const&, bool, std::vector<mxnet::TShape, std::allocator<mxnet::TShape> >*, std::function<void (char const*, char const*, void*)> const&, bool) () at /home/ubuntu/workspace/mxnet/python/mxnet/../../build/libmxnet.so #5 0x00007ffdd131ebd3 in mxnet::Imperative::Backward(std::vector<mxnet::NDArray*, std::allocator<mxnet::NDArray*> > const&, std::vector<mxnet::NDArray*, std::allocator<mxnet::NDArray*> > const&, std::vector<mxnet::NDArray*, std::allocator<mxnet::NDArray*> > const&, bool, bool, bool) () at /home/ubuntu/workspace/mxnet/python/mxnet/../../build/libmxnet.so #6 0x00007ffdd11a1e3f in MXAutogradBackwardEx () at /home/ubuntu/workspace/mxnet/python/mxnet/../../build/libmxnet.so ``` ## To Reproduce ``` import mxnet as mx from mxnet.gluon import HybridBlock class _TestBlock(HybridBlock): def __init__(self): super(_TestBlock, self).__init__() def hybrid_forward(self, F, x, y): x = F.concat(x, x, dim=-1) x = F.slice_axis(x, axis=-1, begin=0, end=-1) return F.broadcast_lesser_equal(x, y) if __name__ == '__main__': block = _TestBlock() # block.hybridize() # hybridize works x = mx.nd.ones([32, 500, 4]) y = mx.nd.ones([32, 500, 1]) x.attach_grad() y.attach_grad() with mx.autograd.record(): result = block(x, y) # block.export('partitioned') result.backward() print(result) ``` ### Steps to reproduce (Paste the commands you ran that produced the error.) 1. pip3 install mxnet==1.8.0.post0 2. run the script ## What have you tried to solve it? ## Environment ***We recommend using our script for collecting the diagnostic information with the following command*** `curl --retry 10 -s https://raw.githubusercontent.com/apache/incubator-mxnet/master/tools/diagnose.py | python3` <details> <summary>Environment Information</summary> ``` # Paste the diagnose.py command output here ``` </details> -- 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: issues-unsubscr...@mxnet.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@mxnet.apache.org For additional commands, e-mail: issues-h...@mxnet.apache.org