Szilárd Páll





02:52 PM GROMACS Task #2490 (New): evaluate two-step communicators
We should-reevaluate whether/when do we need the two-step communicators implemented in mdrun (separate intra-/inter-n...
02:40 PM GROMACS Task #2304: Document and propose tracking mdrun heuristics
+ c_pullMaxNumLocalAtomsSingleThreaded in pull_internal.h (currently tuned with/for Haswell / gcc 5).
02:28 PM GROMACS Task #2488: use MPI non-blocking collectives to overlap pull comm
Having looked at the code, we will need to split @pull_calc_coms()@ and start the non-blocking collective early, prob...
02:00 PM GROMACS Task #2488 (New): use MPI non-blocking collectives to overlap pull comm
MPI non-blockign collectives should allow overlapping the costly collective COM pull communication with compute and i...


11:13 AM GROMACS Revision 1f866cc5: Fix icc warning in post-submit
icc refuses to understand GMX_UNUSED_VALUE.
Change-Id: Ieca8ad781e6f3bb7908f346a14367dc86ae8b15f


04:12 PM GROMACS Revision 4960eede: Add post-submit with ARM HPC Toolchain
Using gcc-7 and armclang-18.2.
Change-Id: I0cd310786598b7c980af970849ba0de95b91d2ad
04:12 PM GROMACS Revision 49ca05c7: Tweak PME run mode checks
These should allow compile-time evaluation when mdrun is compiled with
As the CPU-only build can now op...


02:49 PM GROMACS Revision ac331ffd: Minor fix for pre-submit matrix docs
Fixed comment to avoid confusion as CMake 3.8 did not introduce any new
CUDA feature that we actually do use (and can...
02:07 PM GROMACS Bug #2470: memory leak in OpenCL runs with
Aleksei Iupinov wrote:
> So I did a 1 rank/16 threads run of NB OpenCL, 96k water box on dev-haswell-gpu01 (with AMD...


11:37 AM GROMACS Revision 6543a8dd: Move the PME GPU clearing to the end of the step
This places the call to the end of the force-compute next to the nbnxn
output clearing also fixing the cycle counting...

Also available in: Atom