GROMACS development

01/21/2020

12:32 PM GROMACS Bug #3305: Case gives FPE with Debug build when GPU update is enabled

Artem, would it be possible just to fall back to CPU update & constraints on energy steps? Then it should work correc...

11:31 AM GROMACS Bug #3305: Case gives FPE with Debug build when GPU update is enabled

>@Alan, can you check if https://gerrit.gromacs.org/#/c/gromacs/+/15457/ fixes it?
Afraid not, we still get the FP...

01/20/2020

01:01 PM GROMACS Task #3156 (Fix uploaded): move ddUsesGpuDirectCommunication and related conditionals into the workload data structures

Implemented in https://gerrit.gromacs.org/c/gromacs/+/15437

01/17/2020

04:52 PM GROMACS Bug #3305: Case gives FPE with Debug build when GPU update is enabled

I've tracked this down - it looks like the GPU constraints is missing code to update the lincsd->rmsdData structure, ...

01/16/2020

06:04 PM GROMACS Bug #3305 (New): Case gives FPE with Debug build when GPU update is enabled

Running with GPU comm feaures + GPU update, this case reliably crashes on step 7200 (see below). It runs fine with GP...

01/15/2020

06:16 PM GROMACS Bug #3240: segv with GPU DD direct communication with GPU update and -dlb off

>I reproduced it on multiple of our dev machines with GV100 and GP100 cards.
Weird, I'm using (on the same commit)...

01:47 PM GROMACS Bug #3240: segv with GPU DD direct communication with GPU update and -dlb off

I still can't reproduce, it is running fine for me using latest release branch commit. Can you please let me know whi...

01/14/2020

06:44 PM GROMACS Bug #3240: segv with GPU DD direct communication with GPU update and -dlb off

Looks like the tpr has not uploaded properly (I can't see it), can you post it again? Thanks

01/06/2020

03:15 PM GROMACS Bug #3159: eliminate regression due to moving gmx_pme_send_coordinates()

>Yes, but this included feature work that should not have been targeted at 2020. I retargeted this at 2020.1 for just...

12/20/2019

01:05 PM GROMACS Bug #3159: eliminate regression due to moving gmx_pme_send_coordinates()

It is fixed in https://gerrit.gromacs.org/c/gromacs/+/14238 which I believe is now working correctly.