[Pw_forum] from 4.1 to 4.2 and 4.3.1 versions of QE in IBM sp6

giacsport at libero.it giacsport at libero.it
Fri Nov 11 04:39:53 CET 2011

Dear All,
              I am trying to plot my band for a Magnetic semiconductor, MnO2 
in some.
I am running the 4.1 version of QE at the CINECA IBM Sp6 machine.
I got this message.

     task #         0
     from bands : error #         1
     The bands code with constrained magnetization has not been tested

Ok. I looked in the forum and I saw the reason for it. 
Now, I was wondering  if more updated versions (4.2 and 4.3.1) could perform 
this kind of calculations.

I started testing it, and I simply replaced in my script for launching the 
job, the older version with the upgraded one (I tested both 4.2 and 4.3.1).
The same input which perfectly converged before, now gives me troubles and the 
message I got immediately is

     task #         1
     from  read_namelists  : error #        88
      reading namelist system

Where am i failing?  Ichjecked the forum once more and I think that it should 
not be mis-spelling (exactly the same Input). 
I checked the example file in the directory of the installed version of QE 4.2 
without finding macroscopic differences with my one (that I attach below)
A problem of parallel libraries? But this one too seems to me the real reason 
( I also tested the launching script in the same directory, getting the same 
message, and after loading the proper module xlf/12-1)

Maybe I am using some obsolete features....

May someone give me a hint to fix my doubt?
(well, if 4.2 and 4.3.1 can do band calculation for constrained magnetic 
systems, and also the reason of the difference for the old and new versions)
Many thanks,

  calculation = "scf"
  pseudo_dir = './'
  restart_mode= "from_scratch",
  prefix      = "MnO2",
  tprnfor     = .true.,
  tstress     = .true.,
    ibrav=  4, celldm(1) = 5.462171712, celldm(3) = 3.7856422, nat=  3, ntyp= 
    ecutwfc =100.0, nbnd=100, force_symmorphic=.true., nspin=2,
    starting_magnetization(1) = 1.,
    mixing_mode = 'plain'
    mixing_beta = 0.7
    conv_thr =  1.0d-8
K_POINTS (automatic)
9 9 1 0 0 0

 Mn  54.938  Mn_HSCV_LDA-1.0.UPF
  O  15.999  O.pz-mt.UPF

Mn  0.0000000000000000  0.0000000000000000  0.5000000000000000
O   0.3403727063099424  0.6596272936900647  0.5876622657201906
O   0.6596272936900647  0.3403727063099424  0.4123377342797880

Giacomo Giorgi, PhD.
The University of Tokyo, Todai, Japan

More information about the users mailing list