search

ccsb-scripps / AutoDock-GPU

AutoDock for GPUs and other accelerators
https://ccsb.scripps.edu/autodock
GNU General Public License v2.0
391 stars 107 forks source link

void SetKernelsGpuData(GpuData*) error #180

Open alfredoq opened 2 years ago

alfredoq commented 2 years ago

Dear All,

I succesfully compiled AutoDock GPU in a Debian system, but when trying to run an example:

./bin/autodock_gpu_128wi --lfile ./input/1stp/derived/1stp_ligand.pdbqt --ffile ./input/1stp/derived/1stp_protein.maps.fld

I am getting the following error:

AutoDock-GPU version: v1.5.3

Running 1 docking calculation

Cuda device: NVIDIA A10 (#1 / 3) Available memory on device: 22506 MB (total: 22731 MB)

CUDA Setup time 0.252437s

Running Job #1 Using heuristics: (capped) number of evaluations set to 1132076 Local-search chosen method is: ADADELTA (ad) SetKernelsGpuData copy to cData failed invalid device symbol autodock_gpu_128wi: ./cuda/kernels.cu:130: void SetKernelsGpuData(GpuData*): Assertion `0' failed. Aborted

Is this related to the GPU I am trying to use (A10) and specific compilation requirements, or is it anything else?

Thank you very much in advance for the support,

regards

Alfredo

sisco0 commented 2 years ago

Kindly take a look at issue #172 where other user had the same problem. This problem is referred firstly to the platform where the kernels are compiled to.

Could you check if the #172 solution solves your problem? Also, there is an active PR on CMake that could lead to a solution, you could try this afterwards.

alfredoq commented 2 years ago

Dear Francisco, Thank you very much for your reply. The issue you pointed out solved my problem. Kind regards Alfredo

sisco0 commented 2 years ago

The current issue status could be changed to Closed, as a solution was already provided in a related issue. Also, the related solution has been marked to be working as expected for the current issue.