Cuda runtime error =

[davide-grheco]

Describe the bug

Running the latest version of unidock from conda-forge, 1.1 is what is printed by version.

The error must be associated to one of the ligands, but I'm unsure as to which. I tried running a couple singularly and they did dock correctly on the same receptor.
I have run these calculations without problems several times on the same machine.

unidock_pipeline_ccibh.tar.gz
Fedora 40

unidock --receptor /tmp/unidock_6the843h/unidock_pipeline_ccibh/6cqz_stripped.pdbqt --gpu_batch $(find . -name "*.sdf") --dir /tmp/unidock_6the843h/unidock_pipeline_ccibh/docking_dir/docking_results --exhaustiveness 128 --max_step 20 --center_x 67.18 --center_y -33.17 --center_z 350.91 --size_x 22.5 --size_y 22.5 --size_z 22.5 --scoring vina --num_modes 3 --energy_range 3.0 --refine_step 3 --seed 181129 --verbosity 2 --keep_nonpolar_H

coring function : vina
Rigid receptor: /tmp/unidock_6the843h/unidock_pipeline_ccibh/6cqz_stripped.pdbqt
Grid center: X 67.18 Y -33.17 Z 350.91
Grid size  : X 22.5 Y 22.5 Z 22.5
Grid space : 0.375
Exhaustiveness: 128
CPU: 0
Verbosity: 2

Computing Vina grid ... done.
Total ligands: 89
Set GPU device id to 0
Available Memory = 3749MiB   Total Memory = 3895MiB
all_ligands.size():89
max_num_atoms105
max_num_torsions:11
max_num_rigids:3
max_num_lig_pairs:1089
Small Group:
Max num_atoms: 69 Max num_torsions: 7 Max num_rigids: 3 Max num_lig_pairs: 643
Group size: 22
Medium Group:
Max num_atoms: 81 Max num_torsions: 9 Max num_rigids: 2 Max num_lig_pairs: 730
Group size: 22
Large Group:
Max num_atoms: 86 Max num_torsions: 11 Max num_rigids: 3 Max num_lig_pairs: 805
Group size: 22
Extra Large Group:
Max num_atoms: 105 Max num_torsions: 11 Max num_rigids: 3 Max num_lig_pairs: 1089
Group size: 23
Small
Max num_atoms: 69 Max num_torsions: 7 Max num_rigids: 3 Max num_lig_pairs: 643
Group size: 22
Small Batch 1 size: 22
Performing docking (random seed: 181129) ... CUDA error at /home/conda/feedstock_root/build_artifacts/unidock_1710301842002/work/unidock/src/cuda/monte_carlo.cu:2854 code=700(cudaErrorIllegalAddress) "cudaDeviceSynchronize()" 


An error occurred: CUDA Runtime Error. 

Expected behavior

No response

To Reproduce

No response

Environment

No response

Additional Context

No response