Aborted (core dumped)

[debanjansen48]

Issue summary

terminate called after throwing an instance of 'std::runtime_error'
what(): CUDNN Error (out of memory?): CUDNN_STATUS_INTERNAL_ERROR
*** Aborted at 1707874291 (unix time) try "date -d @1707874291" if you are using GNU date ***
PC: @ 0x0 (unknown)
*** SIGABRT (@0x3e80000c63b) received by PID 50747 (TID 0x7f68affff000) from PID 50747; stack trace: ***
@ 0x7f6901831046 (unknown)
@ 0x7f6901038520 (unknown)
@ 0x7f690108c9fc pthread_kill
@ 0x7f6901038476 raise
@ 0x7f690101e7f3 abort
@ 0x7f69012e1bbe (unknown)
@ 0x7f69012ed24c (unknown)
@ 0x7f69012ed2b7 std::terminate()
@ 0x7f69012ed518 __cxa_throw
@ 0x7f69020cac5d caffe::CuDNNConvolutionLayer<>::Backward_gpu()
@ 0x7f690205929a caffe::Net<>::BackwardFromTo()
@ 0x7f69020593d9 caffe::Net<>::Backward()
@ 0x56254df4b18d CNNScorer::score()
@ 0x56254dfb7277 non_cache_cnn::eval_deriv()
@ 0x56254dffc886 model::eval_deriv()
@ 0x56254dfdf9f6 bfgs<>()
@ 0x56254dfe27a9 quasi_newton::operator()()
@ 0x56254ded4827 refine_structure()
@ 0x56254dedc1cf do_search()
@ 0x56254dedec67 main_procedure()
@ 0x56254dedf1c5 threads_at_work()
@ 0x56254df28260 boost::_bi::list6<>::operator()<>()
@ 0x56254df282e2 boost::detail::thread_data<>::run()
@ 0x7f6901c670cb (unknown)
@ 0x7f690108aac3 (unknown)
@ 0x7f690111c850 (unknown)
@ 0x0 (unknown)
Aborted (core dumped)

Steps to reproduce

gnina -r protein.pdb -l PT_1.sdf --autobox_ligand ligand.pdb --cnn_scoring refinement -o cnn_refined-PT.sdf.gz

Your system configuration

Operating system:LinuxMint
Compiler:GNU
CUDA version (if applicable):11.8
CUDNN version (if applicable):
Python version:3.9

[dkoes]

Build/run without cudnn for refinement to work.

[debanjansen48]

Respected sir are you suggesting to build gnina again and re install after removing cudNN?

[dkoes]

Or just use the pre-built binary.