[Pw_forum] gipaw calculation with DFT+U missing prefix.hub files

Paolo Giannozzi p.giannozzi at gmail.com
Fri Sep 1 15:22:25 CEST 2017


Oh, I see: it happens when there is a single k-point per pool, because in
that case atomic wavefunctions are just kept in memory and never saved to a
buffer. Quick-and-dirty solution: in PW/src/orthoatwfc.f90, remove the line
    IF ( nks > 1 ) &
just before the line
   CALL save_buffer (wfcU, nwordwfcU, iunhub, ik)

Paolo

On Thu, Aug 31, 2017 at 6:52 PM, Jia Chen <jiachenchem at gmail.com> wrote:

> Great, thank you very much. I copy my input file below. It will be very
> helpful if you can tell anything is incorrect. Appreciate it.
> ===================================
> &control
> calculation         = 'scf',
> restart_mode        = 'from_scratch',
> nstep               = 5000,
> max_seconds         = 84000,
> wf_collect      = .TRUE.
> outdir          = './'
> pseudo_dir      = './'
> /
> &system
> ibrav               = 0,
> nat                 = 32,
> ntyp                = 4,
> ecutwfc             = 100.D0,   ! wave function cutoff
> occupations     ='smearing'
> degauss         =0.002
> nspin   = 1
> spline_ps = .true.
> lda_plus_u      = .TRUE.
> lda_plus_u_kind = 0
> Hubbard_U(1)    =5.0
> !Hubbard_J(1,1)   = 0.5
> starting_magnetization(1)=0.0
> /
>
> &electrons
> electron_maxstep        = 200
> conv_thr        = 1.D-8
> mixing_mode     = 'plain'
> mixing_beta     = 0.2D0
> /
> &ions
> /
>
> &cell
> !cell_dynamics = 'bfgs'
> !press = 0.00
> /
>
> CELL_PARAMETERS {angstrom}
>         7.0009999275         0.0000000000         0.0000000000
>         0.0000000000        10.3559999466         0.0000000000
>        -2.5006805806 <(500)%20680-5806>         0.0000000000         7.
> 9845650713 <(984)%20565-0713>
>
> ATOMIC_SPECIES
> Co     1.00    Co.pbe-sp-mt_gipaw.UPF
> K      1.00    K.pbe-tm-semi-gipaw-xy.UPF
> N      1.00    N.pbe-tm-new-gipaw-dc.UPF
> C      1.00    C.pbe-tm-new-gipaw-dc.UPF
>
>
> ATOMIC_POSITIONS {Angstrom}
> Co     0.000000000         5.177999973         0.000000000
> Co    -1.250340290         0.000000000         3.992282536
> K     3.203347102         7.930935236         1.002222654
> K     1.296972208         2.425064711         6.982342536
> K     2.547312498         2.752935263         2.990060000
> K     1.953006849         7.603064684         4.994505071
> K     0.000000000         0.000000000         0.000000000
> K    -1.250340290         5.177999973         3.992282536
> C     0.772793234         5.709262659         1.656797219
> C     3.727526180         4.646737288         6.327767971
> C     4.977866470         0.531262686         2.335485436
> C    -0.477547019         9.824737261         5.649079636
> C     5.430639095         4.620847312         0.905449658
> C    -0.930319681         5.735152634         7.079115532
> C     0.320020609         9.798847286         3.086832997
> C     4.180298842         0.557152661         4.897732075
> C     0.767450404         3.443370093         0.227560105
> C     3.732869202         6.912630162         7.757004803
> C     4.983209418         8.621369758         3.764722505
> C    -0.482889937         1.734629880         4.219842804
> N     1.250821945         6.012693810         2.634107989
> N     3.249497685         4.343306137         5.350456844
> N     4.499837901         0.834693837         1.358174546
> N     0.000481581         9.521306110         6.626390763
> N     4.491633074         4.281170320         1.449198582
> N     0.008686519         6.074829627         6.535366370
> N     1.259026809         9.459170293         2.543083834
> N     3.241292747         0.896829654         5.441481237
> N     1.260714400         2.488546784         0.387251417
> N     3.239605156         7.867453162         7.597313654
> N     4.489945446         7.666546758         3.605031118
> N     0.010374109         2.689453189         4.379533953
> K_POINTS automatic
> 1 1 1 0 0 0
>
>
> On Thu, Aug 31, 2017 at 8:50 AM, Davide Ceresoli <davide.ceresoli at cnr.it>
> wrote:
>
>> I will also do some tests, but I'm busy in the next few days.
>> D.
>>
>> On 08/31/2017 07:54 AM, Paolo Giannozzi wrote:
>> > Unfortunately there isn't much that can be done unless the problem you
>> mention
>> > is reproduced
>> >
>> > On Wed, Aug 30, 2017 at 11:42 PM, Jia Chen <jiachenchem at gmail.com
>> > <mailto:jiachenchem at gmail.com>> wrote:
>> >
>> >     what happened to me on four different machines when nspin = 1 is
>> that
>> >     prefix.hub files were created, but they are empty. One machine is
>> Cori of
>> >     nersc, and QE was the compiled by staffs there.
>> >
>> >     On Wed, Aug 30, 2017 at 4:15 PM, Paolo Giannozzi <
>> p.giannozzi at gmail.com
>> >     <mailto:p.giannozzi at gmail.com>> wrote:
>> >
>> >         It works for me on a simple test also for nspin=1
>> >
>> >         Paolo
>> >
>> >         On Wed, Aug 30, 2017 at 6:41 PM, Jia Chen <
>> jiachenchem at gmail.com
>> >         <mailto:jiachenchem at gmail.com>> wrote:
>> >
>> >             Thank you very much for this! The special something in my
>> >             calculation is npsin = 1, when it is 2, prefix.hub files
>> were written.
>> >
>> >             Cheers
>> >             Jia
>> >
>> >             On Wed, Aug 30, 2017 at 3:02 AM, Paolo Giannozzi
>> >             <p.giannozzi at gmail.com <mailto:p.giannozzi at gmail.com>>
>> wrote:
>> >
>> >                 I was actually confusing the *.hub files (containing
>> atomic
>> >                 wavefunctions) with the "occup" file, containing
>> occupancies of
>> >                 the Hubbard manifold. Anyway: files *hub are written at
>> the end
>> >                 of the run in normal conditions, so there must be
>> something
>> >                 special about your calculation
>> >
>> >                 Paolo
>> >
>> >                 On Wed, Aug 30, 2017 at 7:45 AM, Jia Chen <
>> jiachenchem at gmail.com
>> >                 <mailto:jiachenchem at gmail.com>> wrote:
>> >
>> >                     Dear Paolo Giannozzi,
>> >
>> >                     I used QE-6.1, and also tried QE-5.4. I checked
>> prefix.save
>> >                     directory, but didn't find prefix.hub files. Is
>> writing
>> >                     prefix.hub files an option can be activated? Thank
>> you very
>> >                     much?
>> >
>> >                     Cheers
>> >
>> >                     On Wed, Aug 30, 2017 at 1:32 AM, Paolo Giannozzi
>> >                     <p.giannozzi at gmail.com <mailto:
>> p.giannozzi at gmail.com>> wrote:
>> >
>> >                         It would help to know which version of the code
>> you
>> >                         used. In recent QE versions "prefix.hub" files
>> have been
>> >                         moved inside the "prefix.save/" directory.
>> Maybe GIPAW
>> >                         is trying to read them from the old location
>> and the
>> >                         empty files you see are just the result of a
>> fortran
>> >                         open of nonexistent files
>> >
>> >                         Paolo
>> >
>> >                         On Wed, Aug 30, 2017 at 5:24 AM, Jia Chen
>> >                         <jiachenchem at gmail.com <mailto:
>> jiachenchem at gmail.com>>
>> >                         wrote:
>> >
>> >                             Dear All,
>> >
>> >                             I am not sure if this is a good place to
>> ask this
>> >                             question, but since gipaw depends on QE, I
>> might try
>> >                             my luck here. I am trying to do nmr
>> calculation with
>> >                             gipaw based on DFT+U. The calculation ended
>> on a
>> >                             error message of missing prefix.hub files.
>> Pwscf
>> >                             actually created prefix.hub files with
>> DFT+U, but
>> >                             those file are empty. I hope gipaw experts
>> here can
>> >                             help me with this issue. Appreciate it.
>> >
>> >                             Cheers
>> >                             Jia Chen
>> >
>> >                             _____________________________
>> __________________
>> >                             Pw_forum mailing list
>> >                             Pw_forum at pwscf.org <mailto:
>> Pw_forum at pwscf.org>
>> >                             http://pwscf.org/mailman/listinfo/pw_forum
>> >                             <http://pwscf.org/mailman/listinfo/pw_forum
>> >
>> >
>> >
>> >
>> >
>> >                         --
>> >                         Paolo Giannozzi, Dip. Scienze Matematiche
>> Informatiche e
>> >                         Fisiche,
>> >                         Univ. Udine, via delle Scienze 208, 33100
>> Udine, Italy
>> >                         Phone +39-0432-558216
>> <tel:+39%200432%20558216>, fax
>> >                         +39-0432-558222 <tel:+39%200432%20558222>
>> >
>> >
>> >                         _______________________________________________
>> >                         Pw_forum mailing list
>> >                         Pw_forum at pwscf.org <mailto:Pw_forum at pwscf.org>
>> >                         http://pwscf.org/mailman/listinfo/pw_forum
>> >                         <http://pwscf.org/mailman/listinfo/pw_forum>
>> >
>> >
>> >
>> >                     _______________________________________________
>> >                     Pw_forum mailing list
>> >                     Pw_forum at pwscf.org <mailto:Pw_forum at pwscf.org>
>> >                     http://pwscf.org/mailman/listinfo/pw_forum
>> >                     <http://pwscf.org/mailman/listinfo/pw_forum>
>> >
>> >
>> >
>> >
>> >                 --
>> >                 Paolo Giannozzi, Dip. Scienze Matematiche Informatiche
>> e Fisiche,
>> >                 Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
>> >                 Phone +39-0432-558216 <tel:+39%200432%20558216>, fax
>> >                 +39-0432-558222 <tel:+39%200432%20558222>
>> >
>> >
>> >                 _______________________________________________
>> >                 Pw_forum mailing list
>> >                 Pw_forum at pwscf.org <mailto:Pw_forum at pwscf.org>
>> >                 http://pwscf.org/mailman/listinfo/pw_forum
>> >                 <http://pwscf.org/mailman/listinfo/pw_forum>
>> >
>> >
>> >
>> >             _______________________________________________
>> >             Pw_forum mailing list
>> >             Pw_forum at pwscf.org <mailto:Pw_forum at pwscf.org>
>> >             http://pwscf.org/mailman/listinfo/pw_forum
>> >             <http://pwscf.org/mailman/listinfo/pw_forum>
>> >
>> >
>> >
>> >
>> >         --
>> >         Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e
>> Fisiche,
>> >         Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
>> >         Phone +39-0432-558216 <tel:+39%200432%20558216>, fax
>> +39-0432-558222
>> >         <tel:+39%200432%20558222>
>> >
>> >
>> >         _______________________________________________
>> >         Pw_forum mailing list
>> >         Pw_forum at pwscf.org <mailto:Pw_forum at pwscf.org>
>> >         http://pwscf.org/mailman/listinfo/pw_forum
>> >         <http://pwscf.org/mailman/listinfo/pw_forum>
>> >
>> >
>> >
>> >     _______________________________________________
>> >     Pw_forum mailing list
>> >     Pw_forum at pwscf.org <mailto:Pw_forum at pwscf.org>
>> >     http://pwscf.org/mailman/listinfo/pw_forum
>> >     <http://pwscf.org/mailman/listinfo/pw_forum>
>> >
>> >
>> >
>> >
>> > --
>> > 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
>> >
>>
>> --
>> +--------------------------------------------------------------+
>>    Davide Ceresoli
>>    CNR Institute of Molecular Science and Technology (CNR-ISTM)
>>    c/o University of Milan, via Golgi 19, 20133 Milan, Italy
>>    Email: davide.ceresoli at cnr.it
>>    Phone: +39-02-50314276, +39-347-1001570 (mobile)
>>    Skype: dceresoli
>>    Website: http://sites.google.com/site/dceresoli/
>> +--------------------------------------------------------------+
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://pwscf.org/mailman/listinfo/pw_forum
>>
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>



-- 
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
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20170901/36b6eaed/attachment.html>


More information about the users mailing list