On Friday 18 March 2005 21:16, Lilong Li wrote: > Actually this is related to npool. For example, for the scf run, i used > mpirun -np 16 pw.x -npool 8 < file.in > file.out > If I use the efg.x in the following way: > mpirun -np 16 efg.x -npool 8 < efg.in > efg.out > I will get 7 "error reading inputpp namelist" error.
because the code is trying ot read from 8 processors and finds just one input. It is actually a bug that wil be (hopefully) fixed in the next release. Thank you for pointing out this and other problems. > I also tried > mpirun -np 16 efg.x -npool 1 < efg.in > efg.out > This time there was no reading inputpp namelist error. But I still get > the davcio error. Probably with a different npool, efg.x reassigns the > calculation differently than the previous scf run. correct: presently you cannot read data from a previous parallel run unless - wavefunctions are collected into a single file, or - you use the same number of processes and pools Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050-509876, Fax:-563513 Piazza dei Cavalieri 7 I-56126 Pisa, Italy
