GROMACS development

04/23/2018

11:34 AM GROMACS Revision 049fdc53: Removed cu_realloc_buffered() in favor of reallocateDeviceBuffer()

Change-Id: I5f0d374f8e7e22505265f3cebc60fe29948ab6e9

11:22 AM GROMACS Revision 59ebd3a3: DeviceBuffer headers are added

For now they contain templated versions of GPU memory (re-)allocation
and host-to-device copy routines. The DeviceBuf...

04/20/2018

01:47 PM GROMACS Bug #2470: memory leak in OpenCL runs with

So I did a 1 rank/16 threads run of NB OpenCL, 96k water box on dev-haswell-gpu01 (with AMD GPU PRO drivers) for 4 ho...

04/12/2018

09:35 PM GROMACS Revision b4cab42a: Make in-source OpenCL kernel path a per-module parameter

Change-Id: I19cd66b441fec6ceba74eb648bea3409b6521917

05:35 PM GROMACS Revision e9e20e64: Remove CUDA_FUNC macros from many functions in PME GPU

As pme-gpu*.cpp files are not compiled during CPU builds anymore,
PME GPU functions that are not exposed to mdrun/tes...

04:50 PM GROMACS Revision 661842ba: Stop PME GPU code from compiling on non-CUDA builds

On non-CUDA builds: all entrypoints into PME GPU code are
marked as stubs; pme-gpu.cpp and pme-gpu-internal.cpp are
e...

04/10/2018

05:45 PM GROMACS Bug #2470 (Resolved): memory leak in OpenCL runs with

Applied in changeset commit:f2c9e7785cf77a6b6ede7a39654578185f4b285b.

05:24 PM GROMACS Revision f2c9e778: Prevent OpenCL timing memory leak

Fixes #2470
Change-Id: I4917de697bee7df98da0037e9165e52e660f83a0

03/27/2018

01:56 PM GROMACS Task #2453: PME OpenCL porting effort

Update: gather kernel working as well as spread on AMD/Intel/NVidia
https://github.com/yupinov/gromacs/tree/pme_op...

03/23/2018

04:59 PM GROMACS Task #2464 (New): GPU performance goals overview

This is an overview/place of dicussion of potential goals for GPU improvements in Gromacs.
We have 3 distinct cases,...