1. 11 Jan, 2019 3 commits
  2. 10 Jan, 2019 1 commit
    • Samuel Poncé's avatar
      Various updates · bbb38902
      Samuel Poncé authored
      Replaced QPeig_read with iuqpeig in io_epw.f90, elphon_shuffle_wrap.f90, pw2wan2epw.f90
      
      Updated alignments and format  in epw.f90, epw_readin.f90, loadkmesh.f90, loadumat.f90,  kfold.f90
      
      allocate_epwq.f90
      -----------------------------------------
      1) - Updated based on PHonon/PH/allocate_phq.f90
      
      2) - dpqq and dpqq_so should be allocated if [okvan .and. (epsil .or. zue .or. l_head)]
      
      2) - Removed int3 and int3_nc allocation from allocate_epwq (they are only used for US)
         - It should be allocated in some place similar to PHonon/PH/phqscf.f90 or PHonon/PH/elphon.f90?
      
      3) - Moved allocation of nbnd_occ(nks) to epw_setup.f90 as in PHonon/PH/phq_setup.f90
      
      4) - Moved allocation of transp_temp(nstemp) to epw_setup.f90
      =========================================
      
      epw_setup.f90
      -----------------------------------------
      1) - Updated based on PHonon/PH/phq_setup.f90
      
      2) - Should we remove the lines for lgamma_gamma, alpha_mix, flmixdpot?
         - I commented out these lines for now and the result seams unchanged.
      
      3) - I think we can also remove the lines where npertx is defined.
      =========================================
      
      epw_summary.f90
      -----------------------------------------
      1) - Updated based on PHonon/PH/phq_summary.f90
      
      2) - Updated alignment
      =========================================
      
      openfilepw.f90
      -----------------------------------------
      1) - Updated based on PHonon/PH/openfilq.f90
      
      2) - Should we remove the lines where dvscf file used to be read?
         - These lines are commented out.
      ==========================================
      
      epw_init.f90
      ------------------------------------------
      1) - Updated based PHonon/PH/phq_init.f90
      
      2) - Removed calls to struc_fact, init_vloc, init_us_1 and gk_sort.
         - These subroutines are already called in epw_readin.f90
           through the call to read_file.
      
           CALL read_file (PW/src/read_file_new.f90)
             -> CALL read_xml_file (PW/src/read_file_new.f90)
                  -> CALL struc_fact (PW/src/struc_fact.f90)
                  -> CALL init_vloc (PW/src/init_vloc.f90)
      ==========================================
      
      dvanqq2.f90
      ------------------------------------------
      1) - Updated based on PHonon/PH/dvanqq.f90 and added documentation
      ==========================================
      
      elphon_shuffle_wrap
      ------------------------------------------
      1) - Updated call to createkmap_pw2
         - It seams that xq0 doesn't need to be given in
           createkmap_pw2(xk_all, nkstot, xq0) since
      
           xq0(:) = zero
           CALL createkmap_pw2(xk_all, nkstot, xq0)
      
           and createkmap_pw2 is only called here
      ==========================================
      
      createkmap.f90
      ------------------------------------------
      1) - Updated createkmap_pw2 based on ggen subroutine from Modules/recvec_subs.f90
         - Removed kmap writing in createkmap_pw2 since this is in fact written in createkmap
      
      2) - Added documentation
      
      3) - Removed the check on xk_cryst since this is already done on xk and the two arrays are the same.
      
      refold.f90
      ------------------------------------------
      1) - Added documentation
      
      2) - This subroutine could be moved at the end of createkmap.f90 file
           since it is only called once in createkmap_pw2
      ==========================================
      
      Update made by Roxana Margine.
      bbb38902
  3. 09 Jan, 2019 2 commits
  4. 31 Dec, 2018 1 commit
    • giannozz's avatar
      Pseudopotential cleanup: · a323a89b
      giannozz authored
      - variable oldvan deleted. I think everything should work as before,
        because both Q functions and the Herman-Skillman grid are correctly
        computed in this case as well, but cannot check it. The old Vanderbilt
        format is anyway really old, dating back to 20 years or so.
      - variable newpseudo merged into the upf% structure as upf%is_multiproj.
        I think I have reproduced exactly the previous behaviour but again
        cannot check it. I haven't checked the current logic (or lack thereof)
        of this variable, but a better logic would be to actually check whether
        the PP is multi-projector and in all cases (USPP and PAW as well as NCPP)
      a323a89b
  5. 13 Dec, 2018 1 commit
  6. 03 Dec, 2018 1 commit
  7. 27 Nov, 2018 2 commits
  8. 19 Nov, 2018 1 commit
  9. 17 Nov, 2018 1 commit
  10. 11 Nov, 2018 1 commit
  11. 10 Nov, 2018 2 commits
  12. 08 Nov, 2018 1 commit
  13. 29 Oct, 2018 2 commits
  14. 23 Oct, 2018 2 commits
  15. 19 Oct, 2018 1 commit
  16. 17 Oct, 2018 5 commits
  17. 16 Oct, 2018 3 commits
    • Samuel Poncé's avatar
      Bug in parallel raised by test-farm · 130204d6
      Samuel Poncé authored
      130204d6
    • Samuel Poncé's avatar
      Debug for serial case. · f5ff40b4
      Samuel Poncé authored
      IBTE must be compiled with MPI.
      f5ff40b4
    • Samuel Poncé's avatar
      Cleaning and optimization in EPW · dd817a57
      Samuel Poncé authored
      1) Complete re-writing of the IBTE (at least an order of magnitude faster).
      2) Supporting both k-point symmetry and no-symmetry
      3) Creating of a few new modules
      4) Addition of a epw_mob_ibte_sym for testing the IBTE using k-point symmetry.
      Both epw_mob_ibte_sym and epw_mob_ibte should give the same result but
      epw_mob_ibte_sym is much faster.
      dd817a57
  18. 11 Oct, 2018 5 commits
  19. 25 Sep, 2018 1 commit
  20. 24 Sep, 2018 1 commit
  21. 22 Sep, 2018 1 commit
    • Samuel Poncé's avatar
      Debug raised by test-farm · 25ed0d07
      Samuel Poncé authored
      Imposition of the phase (input variable lphase) is crucial to obtain
      consistent results between etf_mem 1 and 2.
      lphase has therefore been added to the input of the epw_polar test.
      25ed0d07
  22. 21 Sep, 2018 1 commit
    • Samuel Poncé's avatar
      New implementation of the WS vectors in EPW · e81850ca
      Samuel Poncé authored
      The new implementation relies on Wannier centers positions and is
      triggered with the input variable "use_ws == .true."
      When false, recover the old behavior.
      No test has been added yet for this.
      
      The update of the epw_polar test epw2 and epw3 is because of the
      removal in EPW of the exponential look-up table which slightly change
      numerical results.
      e81850ca
  23. 17 Sep, 2018 1 commit