[CP2K-user] [CP2K:20380] Compiling CP2K with GPU support doesn't utilize GPU

[Ray Schireman]

Hi all,

I compiled cp2k on a Slurm compute node with a V100 with the following 
command:

./install_cp2k_toolchain.sh --enable-cuda=yes --gpu-ver=V100 
--with-openmpi=install

No errors were reported, but when I go to run a benchmark (on the same 
node), the GPU is not being utilized. Here's the command I used:

mpirun -np 1 ../../exe/local/cp2k.popt -i H2O-2048.inp

In fact, the calculation results in an error after a few seconds:
Extrapolation method: initial_guess
[dg-gpunode04:20950:0:20950] Caught signal 11 (Segmentation fault: address 
not mapped to object at address 0x2b551b96b000)
==== backtrace (tid:  20950) ====
=================================

Program received signal SIGSEGV: Segmentation fault - invalid memory 
reference.

Backtrace for this error:
#0  0x2b46c0edf62f in ???
#1  0x2b46c08a40a7 in ???

I'm not really sure what's going on, my CPU version works just fine 
(besides poor performance on multiple nodes). I would appreciate any 
feedback.

Thank you for your time,
Ray


-- 
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/698dfe76-6c20-44c0-b2f2-01fe19d4f0cfn%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20240623/80388791/attachment.htm>