General

Profile

Szilárd Páll

Issues

Projects

Activity

12/09/2019

07:13 PM GROMACS Bug #3231 (New): simulationWorkload incorrect for some of the propagators
As a side-effect of trying to improved the stateGpu management, I found that rerun runs have incorrect simulationWork...

12/04/2019

12:28 AM GROMACS Task #3205: address Debian experimental issues
Mark Abraham wrote:
> Nicholas, https://gerrit.gromacs.org/#/c/14624 would give us more insight into the Angles fail...
09:24 PM GROMACS Task #3216: improve CPU force reductions
Berk Hess wrote:
> What is code-path aware force clearing.
Not the best wording, I guess. On the GPU-offload code...
03:01 PM GROMACS Task #3205: address Debian experimental issues
Mark Abraham wrote:
> Szilárd Páll wrote:
> > The results of the beta1 show a couple of failures:
> > https://buil...
10:49 AM GROMACS Bug #2974: position-restraints regressiontest fails on intel GPU with OpenCL
It would be good to know what has changed, even if we don't fix it in 2019, so I'd prefer to not close this yet.
10:47 AM GROMACS Task #3189: implement heuristics for switching between different spread/gather kernel layouts
bumped because it requires more work

12/03/2019

12:40 AM GROMACS Bug #2974: position-restraints regressiontest fails on intel GPU with OpenCL
Something changed that seems to make the same test on the same hardware on the current 2020 branch whereas on 2019 it...
12:20 AM GROMACS Task #3189 (In Progress): implement heuristics for switching between different spread/gather kernel layouts
Paul Bauer wrote:
> not going to be in 2020
already merged, though uncertain about whether the current version is...
12:13 AM GROMACS Task #2699: Test POWER9
2020 beta3 builds and passed tests on our Raptor Talos II Power9 with + an NVIDIA GPU using gcc 8.3 and 9.2 + CUDA 10.1.
11:02 PM GROMACS Task #2699: Test POWER9
This issues seems to be more about xlc rather than Power, so the title seems is slightly misleading.
I've built an...

Also available in: Atom