[QE-developers] QE-ESM-RISM stack overflow

[Kaňka Jiří]

Dear QE-ESM-RISM developers,

my model consists of 128 atoms (48 Au atoms in 4 layers simulating Au(111) surface + 4 adsorbed alkanethiols SC5H10COOH).
While smaller systems from the tutorial using RMM-DIIS were calculated smoothly with q-e-6.1 rism (q-e20180928),
the calculation of my larger system stopped during Laue-RISM calculation after 26 iterations with error message:

--------------------------------------------------------------------------
No OpenFabrics connection schemes reported that they were able to be
used on a specific port.  As such, the openib BTL (OpenFabrics
support) will be disabled for this port.

  Local host:           aman10
  Local device:         mlx4_0
  Local port:           2
  CPCs attempted:       rdmacm, udcm
--------------------------------------------------------------------------
[aman10.ics.muni.cz:48796] 31 more processes have sent help message help-mpi-btl-openib-cpc-base.txt / no cpcs for port
[aman10.ics.muni.cz:48796] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages
forrtl: severe (174): SIGSEGV, segmentation fault occurred

This has not happeed with the Davidson diagonalization.  When RISM is switched off the ESM calcualation of my system is correctly completed even with  RMM-DIIS.

Best regards,

Jiri Kanka
Institute of Photonics and Electronics ASCR
Prague

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/developers/attachments/20200803/eb8fa2d7/attachment.html>