Thank you for reporting this. It was a rather funny bug due to the way character variables are read by FoX (with the old format there was no such problem). There is a workaround available in the development version. You may also try the attached patch that hopefully fixes the bug for good.
Paolo On Thu, Dec 13, 2018 at 9:52 AM Davide Sangalli <davide.sanga...@gmail.com> wrote: > Dear pw users, > I recently tried to run projwfc.x on three systems with SOC and > systematically run into the following error: > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > Error in routine d_matrix_so (7): > D_S (j=1/2) for this symmetry operation is not unitary > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > I reproduced the error onto three different systems: > MoS2, bulk iron, FeRh > with both the last git revision ( commit > 3e920635bff849e891bb3fcec99acc096c3bdcf6) > and qe-6.2.1 > > I had a look to the value of "capel" in the subroutine > "PP/src/d_matrix_so.f90" > for the case of MoS2 (inputs attached). > > In the loop over symmetries, "capel" becomes 2 for the symmetry number 9, > since > ZDOTU(2, d12_con(1,m), 1, dyj12(1,n,isym), 1) > is identically zero for all n and m. > > Kind regards, > D. > > > > > _______________________________________________ > users mailing list > users@lists.quantum-espresso.org > https://lists.quantum-espresso.org/mailman/listinfo/users -- Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche, Univ. Udine, via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222
diffs
Description: Binary data
_______________________________________________ users mailing list users@lists.quantum-espresso.org https://lists.quantum-espresso.org/mailman/listinfo/users