search

marius-team / marius

Large scale graph learning on a single machine.
https://marius-project.org
Apache License 2.0
160 stars 45 forks source link

question : cmake with -DCMAKE_BUILD_TYPE=Debug #146

Open yichuan520030910320 opened 1 year ago

yichuan520030910320 commented 1 year ago

Why does running marius_train result in a core dump when I set -DCMAKE_BUILD_TYPE=Debug during pip install . or cmake, but not in release mode?

yichuan520030910320 commented 1 year ago

while use gdb it occurs problems here Thread 1 "marius_train" received signal SIGSEGV, Segmentation fault. 0x000000000045a9b4 in c10::intrusive_ptr<c10::TensorImpl, c10::UndefinedTensorImpl>::intrusive_ptr(c10::intrusive_ptr<c10::TensorImpl, c10::UndefinedTensorImpl> const&) () Missing separate debuginfos, use: yum debuginfo-install glibc-2.28-189.5.el8_6.x86_64 nvidia-driver-cuda-libs-535.54.03-1.el8.x86_64 (gdb) bt

0 0x000000000045a9b4 in c10::intrusive_ptr<c10::TensorImpl, c10::UndefinedTensorImpl>::intrusive_ptr(c10::intrusive_ptr<c10::TensorImpl, c10::UndefinedTensorImpl> const&) ()

1 0x00007ffff759add8 in c10::intrusive_ptr<c10::TensorImpl, c10::UndefinedTensorImpl>& c10::intrusive_ptr<c10::TensorImpl, c10::UndefinedTensorImpl>::operator=<c10::TensorImpl, c10::UndefinedTensorImpl>(c10::intrusive_ptr<c10::TensorImpl, c10::UndefinedTensorImpl> const&) & () from /home/yw8143/marius_artifact/build_debug/libmarius.so

2 0x00007ffff7598b3f in c10::intrusive_ptr<c10::TensorImpl, c10::UndefinedTensorImpl>::operator=(c10::intrusive_ptr<c10::TensorImpl, c10::UndefinedTensorImpl> const&) & ()

from /home/yw8143/marius_artifact/build_debug/libmarius.so

3 0x00007ffff7593468 in at::Tensor::operator=(at::TensorBase const&) & () from /home/yw8143/marius_artifact/build_debug/libmarius.so

4 0x00007ffff75934d9 in at::Tensor::operator=(at::Tensor const&) & () from /home/yw8143/marius_artifact/build_debug/libmarius.so

5 0x00007ffff7673bbe in LayeredNeighborSampler::getNeighbors(at::Tensor, int) () from /home/yw8143/marius_artifact/build_debug/libmarius.so

6 0x00007ffff75f3f45 in DataLoader::nodeClassificationSample(Batch*, int) () from /home/yw8143/marius_artifact/build_debug/libmarius.so

7 0x00007ffff75f2dd1 in DataLoader::getBatch(int) () from /home/yw8143/marius_artifact/build_debug/libmarius.so

8 0x00007ffff7771ef6 in SynchronousTrainer::train(int) () from /home/yw8143/marius_artifact/build_debug/libmarius.so

9 0x00000000004442b4 in marius(int, char**) ()

10 0x000000000044486d in main ()