Artem Zhmurov





11:15 AM GROMACS Revision 9b682479: Disable GPU update/constraints when neither PME nor buffer ops are offloaded.
Using the GPU-version of update makes sense if forces are already on the GPU,
i.e. if at least:
1. PME is on the GPU ...


05:07 PM GROMACS Revision d74fb649: Change PaddedVector to PaddedHostVector for force CPU buffer
The difference between PaddedVector and PaddedHostVector is that
in the builds where CUDA is available, cudaMalloc is...


11:04 PM GROMACS Revision 62abb664: Fix for the post-submit failure
Commit 21abdb3 introduced several "unused variable" warning
in the post-submit matrix. This fixes the issue.
10:53 PM GROMACS Revision 873bf080: Decouple GPU force buffer management from buffer ops in NBNXM
When GPU-side buffer operations are used, the total forces on the
device are accumulated in NBNXM module in the local...
06:35 PM GROMACS Revision c69e061d: Decouple coordinates buffer management from buffer ops in NBNXM
To make transition to the new device-side state propagator data
manager easier, the copy and plain buffers management...
06:35 PM GROMACS Revision 21abdb3c: Reorganize PME code:
1. Split H2D copy and spread launch
2. Add getter for the padding, required in coordinates buffer
3. Add the getter f...


11:01 PM GROMACS Revision 22167aee: Making DeviceBuffer availible in non-GPU builds
Having DeviceBuffer availible in host-side code in all builds
allows to avoid passing the void-pointers for the devic...


01:47 PM GROMACS Task #2053: refine notation in GPU code
Mark Abraham wrote:
> Artem Zhmurov wrote:
> > Update for the naming convention for these variables for d_<bufferNa...


08:25 AM GROMACS Wiki edit: Integration_planning_for_GROMACS_2020_GPU_features (#7)
06:49 AM GROMACS Revision 7ddb7204: Add Parrinello-Rahman pressure coupling to CUDA Leap-Frog integrator
Parrinello-Rahman isotropic pressure coupling scheme is now
added to CUDA version of the Leap-Frog integrator. The

Also available in: Atom