2003-01-20 05:58:50 +08:00
|
|
|
|
2009-08-01 22:26:46 +08:00
|
|
|
! Copyright (C) 2002-2008 Quantum ESPRESSO group
|
2003-01-20 05:58:50 +08:00
|
|
|
! This file is distributed under the terms of the
|
|
|
|
|
! GNU General Public License. See the file `License'
|
|
|
|
|
! in the root directory of the present distribution,
|
|
|
|
|
! or http://www.gnu.org/copyleft/gpl.txt .
|
|
|
|
|
!
|
|
|
|
|
|
|
|
|
|
!=----------------------------------------------------------------------------=!
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
MODULE read_upf_v1_module
|
2003-01-20 05:58:50 +08:00
|
|
|
!=----------------------------------------------------------------------------=!
|
|
|
|
|
|
|
|
|
|
! this module handles the reading of pseudopotential data
|
|
|
|
|
|
|
|
|
|
! ... declare modules
|
2020-04-26 03:13:15 +08:00
|
|
|
USE upf_kinds, ONLY: DP
|
2018-12-30 05:09:38 +08:00
|
|
|
USE pseudo_types, ONLY : pseudo_upf
|
2020-04-26 03:13:15 +08:00
|
|
|
USE upf_utils, ONLY: matches
|
2018-12-30 05:09:38 +08:00
|
|
|
!
|
2003-01-20 05:58:50 +08:00
|
|
|
IMPLICIT NONE
|
|
|
|
|
SAVE
|
2006-03-07 21:26:52 +08:00
|
|
|
PRIVATE
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
PUBLIC :: read_upf_v1, scan_begin, scan_end
|
2003-01-20 05:58:50 +08:00
|
|
|
CONTAINS
|
|
|
|
|
!
|
|
|
|
|
!---------------------------------------------------------------------
|
2020-04-30 21:36:51 +08:00
|
|
|
SUBROUTINE read_upf_v1 ( file_pseudo, upf, ierr )
|
2003-01-20 05:58:50 +08:00
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
! read pseudopotential "upf" in the Unified Pseudopotential Format
|
2020-04-30 21:36:51 +08:00
|
|
|
! from file "file_pseudo" - File is closed on output.
|
|
|
|
|
! return error code in "ierr" (success: ierr=0)
|
2003-01-20 05:58:50 +08:00
|
|
|
!
|
2018-12-30 05:09:38 +08:00
|
|
|
IMPLICIT NONE
|
2003-01-20 05:58:50 +08:00
|
|
|
!
|
2020-04-30 21:36:51 +08:00
|
|
|
CHARACTER(LEN=*) :: file_pseudo
|
2007-03-19 03:24:56 +08:00
|
|
|
INTEGER, INTENT(OUT) :: ierr
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
TYPE (pseudo_upf), INTENT(INOUT) :: upf
|
2003-01-20 05:58:50 +08:00
|
|
|
!
|
|
|
|
|
! Local variables
|
|
|
|
|
!
|
2020-04-30 21:36:51 +08:00
|
|
|
INTEGER :: ios, iunps
|
2018-12-30 05:09:38 +08:00
|
|
|
CHARACTER (len=80) :: dummy
|
2003-01-20 05:58:50 +08:00
|
|
|
!
|
2020-04-30 21:36:51 +08:00
|
|
|
! Open the file
|
|
|
|
|
!
|
|
|
|
|
OPEN ( NEWUNIT = iunps, FILE = file_pseudo, STATUS = 'old', &
|
|
|
|
|
FORM = 'formatted', iostat=ios )
|
2020-05-01 16:19:29 +08:00
|
|
|
IF ( ios /= 0 ) GO TO 200
|
2004-04-30 21:16:03 +08:00
|
|
|
!
|
2020-05-02 21:58:25 +08:00
|
|
|
! First check if this pseudo-potential has spin-orbit or GIPAW information
|
2004-04-30 21:16:03 +08:00
|
|
|
!
|
2007-11-14 01:42:29 +08:00
|
|
|
upf%q_with_l=.false.
|
|
|
|
|
upf%has_so=.false.
|
2007-02-22 00:54:18 +08:00
|
|
|
upf%has_gipaw = .false.
|
2004-04-30 21:16:03 +08:00
|
|
|
addinfo_loop: do while (ios == 0)
|
|
|
|
|
read (iunps, *, iostat = ios, err = 200) dummy
|
2007-02-22 00:54:18 +08:00
|
|
|
if (matches ("<PP_ADDINFO>", dummy) ) then
|
2007-11-20 00:20:26 +08:00
|
|
|
upf%has_so=.true.
|
2007-02-22 00:54:18 +08:00
|
|
|
endif
|
|
|
|
|
if ( matches ( "<PP_GIPAW_RECONSTRUCTION_DATA>", dummy ) ) then
|
|
|
|
|
upf%has_gipaw = .true.
|
2004-04-30 21:16:03 +08:00
|
|
|
endif
|
2007-11-14 01:42:29 +08:00
|
|
|
if (matches ("<PP_QIJ_WITH_L>", dummy) ) then
|
|
|
|
|
upf%q_with_l=.true.
|
|
|
|
|
endif
|
2004-04-30 21:16:03 +08:00
|
|
|
enddo addinfo_loop
|
2007-02-22 00:54:18 +08:00
|
|
|
|
2003-01-20 05:58:50 +08:00
|
|
|
!------->Search for Header
|
|
|
|
|
! This version doesn't use the new routine scan_begin
|
|
|
|
|
! because this search must set extra flags for
|
|
|
|
|
! compatibility with other pp format reading
|
2020-04-30 21:36:51 +08:00
|
|
|
|
2003-01-20 05:58:50 +08:00
|
|
|
ios = 0
|
2020-04-30 21:36:51 +08:00
|
|
|
ierr= 1
|
2004-04-30 21:16:03 +08:00
|
|
|
rewind(iunps)
|
2003-01-20 05:58:50 +08:00
|
|
|
header_loop: do while (ios == 0)
|
|
|
|
|
read (iunps, *, iostat = ios, err = 200) dummy
|
|
|
|
|
if (matches ("<PP_HEADER>", dummy) ) then
|
|
|
|
|
ierr = 0
|
|
|
|
|
call read_pseudo_header (upf, iunps)
|
|
|
|
|
exit header_loop
|
|
|
|
|
endif
|
|
|
|
|
enddo header_loop
|
2020-05-01 16:19:29 +08:00
|
|
|
if (ierr .ne. 0) GO TO 200
|
2007-10-08 15:33:57 +08:00
|
|
|
!
|
|
|
|
|
! this should be read from the PP_INFO section
|
|
|
|
|
!
|
|
|
|
|
upf%generated='Generated by new atomic code, or converted to UPF format'
|
2003-01-20 05:58:50 +08:00
|
|
|
|
|
|
|
|
call scan_end (iunps, "HEADER")
|
|
|
|
|
|
2003-11-04 18:26:03 +08:00
|
|
|
! WRITE( stdout, * ) "Reading pseudopotential file in UPF format"
|
2020-05-02 21:58:25 +08:00
|
|
|
! Compatibility with later formats:
|
|
|
|
|
upf%has_wfc = .false.
|
2003-01-20 05:58:50 +08:00
|
|
|
|
|
|
|
|
!-------->Search for mesh information
|
|
|
|
|
call scan_begin (iunps, "MESH", .true.)
|
|
|
|
|
call read_pseudo_mesh (upf, iunps)
|
|
|
|
|
call scan_end (iunps, "MESH")
|
|
|
|
|
!-------->If present, search for nlcc
|
|
|
|
|
if ( upf%nlcc ) then
|
|
|
|
|
call scan_begin (iunps, "NLCC", .true.)
|
|
|
|
|
call read_pseudo_nlcc (upf, iunps)
|
|
|
|
|
call scan_end (iunps, "NLCC")
|
|
|
|
|
else
|
2007-10-08 15:33:57 +08:00
|
|
|
ALLOCATE( upf%rho_atc( upf%mesh ) )
|
2007-06-12 01:13:15 +08:00
|
|
|
upf%rho_atc = 0.0_DP
|
2003-01-20 05:58:50 +08:00
|
|
|
endif
|
2007-10-29 21:47:45 +08:00
|
|
|
!-------->Fake 1/r potential: do not read PP
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
if (.not. matches ("1/r", upf%typ) ) then
|
2003-01-20 05:58:50 +08:00
|
|
|
!-------->Search for Local potential
|
2007-10-29 21:47:45 +08:00
|
|
|
call scan_begin (iunps, "LOCAL", .true.)
|
|
|
|
|
call read_pseudo_local (upf, iunps)
|
|
|
|
|
call scan_end (iunps, "LOCAL")
|
2003-01-20 05:58:50 +08:00
|
|
|
!-------->Search for Nonlocal potential
|
2007-10-29 21:47:45 +08:00
|
|
|
call scan_begin (iunps, "NONLOCAL", .true.)
|
|
|
|
|
call read_pseudo_nl (upf, iunps)
|
|
|
|
|
call scan_end (iunps, "NONLOCAL")
|
|
|
|
|
!--------
|
2012-04-07 01:02:43 +08:00
|
|
|
else
|
|
|
|
|
call set_coulomb_nonlocal(upf)
|
2007-10-29 21:47:45 +08:00
|
|
|
end if
|
2003-01-20 05:58:50 +08:00
|
|
|
!-------->Search for atomic wavefunctions
|
|
|
|
|
call scan_begin (iunps, "PSWFC", .true.)
|
|
|
|
|
call read_pseudo_pswfc (upf, iunps)
|
|
|
|
|
call scan_end (iunps, "PSWFC")
|
|
|
|
|
!-------->Search for atomic charge
|
|
|
|
|
call scan_begin (iunps, "RHOATOM", .true.)
|
|
|
|
|
call read_pseudo_rhoatom (upf, iunps)
|
|
|
|
|
call scan_end (iunps, "RHOATOM")
|
2004-04-30 21:16:03 +08:00
|
|
|
!-------->Search for add_info
|
|
|
|
|
if (upf%has_so) then
|
|
|
|
|
call scan_begin (iunps, "ADDINFO", .true.)
|
|
|
|
|
call read_pseudo_addinfo (upf, iunps)
|
|
|
|
|
call scan_end (iunps, "ADDINFO")
|
|
|
|
|
endif
|
2007-11-26 21:24:42 +08:00
|
|
|
!-------->GIPAW data
|
|
|
|
|
IF ( upf%has_gipaw ) then
|
|
|
|
|
CALL scan_begin ( iunps, "GIPAW_RECONSTRUCTION_DATA", .false. )
|
|
|
|
|
CALL read_pseudo_gipaw ( upf, iunps )
|
|
|
|
|
CALL scan_end ( iunps, "GIPAW_RECONSTRUCTION_DATA" )
|
|
|
|
|
END IF
|
2007-11-21 22:41:30 +08:00
|
|
|
!--- Try to get the core radius if not present. Needed by the
|
|
|
|
|
! atomic code for old pseudo files
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
IF (upf%nbeta>0) THEN ! rcutus may be unallocated if nbeta=0
|
2020-04-30 21:36:51 +08:00
|
|
|
IF(upf%rcutus(1)<1.e-9_DP) THEN
|
|
|
|
|
call scan_begin (iunps, "INFO", .true.)
|
|
|
|
|
call read_pseudo_ppinfo (upf, iunps)
|
|
|
|
|
call scan_end (iunps, "INFO")
|
|
|
|
|
ENDIF
|
2007-11-21 22:41:30 +08:00
|
|
|
ENDIF
|
2003-01-20 05:58:50 +08:00
|
|
|
|
2020-04-30 21:36:51 +08:00
|
|
|
200 CLOSE (UNIT=iunps)
|
2003-01-20 05:58:50 +08:00
|
|
|
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
end subroutine read_upf_v1
|
2003-01-20 05:58:50 +08:00
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
subroutine scan_begin (iunps, string, rew)
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
implicit none
|
|
|
|
|
! Unit of the input file
|
|
|
|
|
integer :: iunps
|
|
|
|
|
! Label to be matched
|
|
|
|
|
character (len=*) :: string
|
|
|
|
|
! String read from file
|
|
|
|
|
character (len=75) :: rstring
|
|
|
|
|
! Flag if .true. rewind the file
|
|
|
|
|
logical :: rew
|
|
|
|
|
integer :: ios
|
|
|
|
|
|
|
|
|
|
ios = 0
|
|
|
|
|
if (rew) rewind (iunps)
|
2005-11-03 23:59:28 +08:00
|
|
|
do while (ios==0)
|
2003-01-20 05:58:50 +08:00
|
|
|
read (iunps, *, iostat = ios, err = 300) rstring
|
|
|
|
|
if (matches ("<PP_"//string//">", rstring) ) return
|
|
|
|
|
enddo
|
|
|
|
|
return
|
2020-04-26 03:13:15 +08:00
|
|
|
300 call upf_error ('scan_begin', 'No '//string//' block', abs (ios) )
|
2003-01-20 05:58:50 +08:00
|
|
|
end subroutine scan_begin
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
|
|
|
|
|
subroutine scan_end (iunps, string)
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
implicit none
|
|
|
|
|
! Unit of the input file
|
2005-11-03 23:59:28 +08:00
|
|
|
integer :: iunps
|
2003-01-20 05:58:50 +08:00
|
|
|
! Label to be matched
|
|
|
|
|
character (len=*) :: string
|
|
|
|
|
! String read from file
|
|
|
|
|
character (len=75) :: rstring
|
|
|
|
|
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, '(a)', end = 300, err = 300) rstring
|
2003-01-20 05:58:50 +08:00
|
|
|
if (matches ("</PP_"//string//">", rstring) ) return
|
|
|
|
|
return
|
2020-04-26 03:13:15 +08:00
|
|
|
300 call upf_error ('scan_end', &
|
2005-11-03 23:59:28 +08:00
|
|
|
'No '//string//' block end statement, possibly corrupted file', -1)
|
2003-01-20 05:58:50 +08:00
|
|
|
end subroutine scan_end
|
|
|
|
|
!
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
|
|
|
|
|
subroutine read_pseudo_header (upf, iunps)
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
implicit none
|
|
|
|
|
!
|
|
|
|
|
TYPE (pseudo_upf), INTENT(INOUT) :: upf
|
|
|
|
|
integer :: iunps
|
|
|
|
|
!
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
integer :: nw
|
|
|
|
|
character (len=80) :: dummy
|
2018-12-31 22:43:38 +08:00
|
|
|
! GTH analytical format: obviously not true in this case
|
|
|
|
|
upf%is_gth=.false.
|
|
|
|
|
! PP is assumed to be multi-projector
|
|
|
|
|
upf%is_multiproj = .true.
|
2003-01-20 05:58:50 +08:00
|
|
|
! Version number (presently ignored)
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, *, err = 100, end = 100) upf%nv , dummy
|
2003-01-20 05:58:50 +08:00
|
|
|
! Element label
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, *, err = 100, end = 100) upf%psd , dummy
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
! Type of pseudo (1/r cannot be read with default format!!!)
|
2008-03-28 01:18:51 +08:00
|
|
|
read (iunps, '(a80)', err = 100, end = 100) dummy
|
|
|
|
|
upf%typ=trim(adjustl(dummy))
|
|
|
|
|
!
|
|
|
|
|
if (matches ('US', upf%typ) ) then
|
2003-01-20 05:58:50 +08:00
|
|
|
upf%tvanp = .true.
|
2007-11-20 00:20:26 +08:00
|
|
|
upf%tpawp = .false.
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
upf%tcoulombp = .false.
|
2008-03-28 01:18:51 +08:00
|
|
|
else if (matches ('PAW', upf%typ) ) then
|
2007-11-20 00:20:26 +08:00
|
|
|
! Note: if tvanp is set to false the results are wrong!
|
|
|
|
|
upf%tvanp = .true.
|
|
|
|
|
upf%tpawp = .true.
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
upf%tcoulombp = .false.
|
2008-03-28 01:18:51 +08:00
|
|
|
else if (matches ('NC', upf%typ) ) then
|
2003-01-20 05:58:50 +08:00
|
|
|
upf%tvanp = .false.
|
2007-11-20 00:20:26 +08:00
|
|
|
upf%tpawp = .false.
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
upf%tcoulombp = .false.
|
2008-03-28 01:18:51 +08:00
|
|
|
else if (matches ('1/r', upf%typ) ) then
|
2007-10-29 21:47:45 +08:00
|
|
|
upf%tvanp = .false.
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
upf%tpawp = .false.
|
|
|
|
|
upf%tcoulombp = .true.
|
2003-01-20 05:58:50 +08:00
|
|
|
else
|
2020-04-26 03:13:15 +08:00
|
|
|
call upf_error ('read_pseudo_header', 'unknown pseudo type', 1)
|
2003-01-20 05:58:50 +08:00
|
|
|
endif
|
|
|
|
|
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, *, err = 100, end = 100) upf%nlcc , dummy
|
2003-01-20 05:58:50 +08:00
|
|
|
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, '(a20,t24,a)', err = 100, end = 100) upf%dft, dummy
|
2003-01-20 05:58:50 +08:00
|
|
|
|
|
|
|
|
read (iunps, * ) upf%zp , dummy
|
|
|
|
|
read (iunps, * ) upf%etotps, dummy
|
|
|
|
|
read (iunps, * ) upf%ecutwfc, upf%ecutrho
|
2006-04-28 00:19:24 +08:00
|
|
|
read (iunps, * ) upf%lmax , dummy
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
read (iunps, *, err = 100, end = 100) upf%mesh , dummy
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, *, err = 100, end = 100) upf%nwfc, upf%nbeta , dummy
|
|
|
|
|
read (iunps, '(a)', err = 100, end = 100) dummy
|
2020-04-27 21:03:58 +08:00
|
|
|
ALLOCATE( upf%els( upf%nwfc ), upf%lchi( upf%nwfc ), upf%oc( upf%nwfc ),&
|
|
|
|
|
upf%nchi( upf%nwfc ) )
|
2003-01-20 05:58:50 +08:00
|
|
|
do nw = 1, upf%nwfc
|
|
|
|
|
read (iunps, * ) upf%els (nw), upf%lchi (nw), upf%oc (nw)
|
2020-09-29 19:00:09 +08:00
|
|
|
upf%nchi (nw) = upf%lchi(nw)+1
|
2003-01-20 05:58:50 +08:00
|
|
|
enddo
|
2020-04-27 21:03:58 +08:00
|
|
|
! next lines for compatibility with upf v.2
|
|
|
|
|
ALLOCATE( upf%rcut_chi( upf%nwfc ), upf%rcutus_chi( upf%nwfc ), &
|
|
|
|
|
upf%epseu( upf%nwfc ) )
|
|
|
|
|
upf%rcut_chi = 0.0_dp; upf%rcutus_chi=0.0_dp; upf%epseu = 0.0_dp
|
2003-01-20 05:58:50 +08:00
|
|
|
|
|
|
|
|
return
|
|
|
|
|
|
2020-04-26 03:13:15 +08:00
|
|
|
100 call upf_error ('read_pseudo_header', 'Reading pseudo file', 1 )
|
2003-01-20 05:58:50 +08:00
|
|
|
end subroutine read_pseudo_header
|
|
|
|
|
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
|
|
|
|
|
subroutine read_pseudo_mesh (upf, iunps)
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
implicit none
|
|
|
|
|
!
|
|
|
|
|
integer :: iunps
|
|
|
|
|
TYPE (pseudo_upf), INTENT(INOUT) :: upf
|
|
|
|
|
!
|
2005-11-03 23:59:28 +08:00
|
|
|
integer :: ir
|
2003-01-20 05:58:50 +08:00
|
|
|
|
2020-04-28 16:15:50 +08:00
|
|
|
ALLOCATE( upf%r( upf%mesh ), upf%rab( upf%mesh ) )
|
2007-06-12 01:13:15 +08:00
|
|
|
upf%r = 0.0_DP
|
|
|
|
|
upf%rab = 0.0_DP
|
2003-01-20 05:58:50 +08:00
|
|
|
|
|
|
|
|
call scan_begin (iunps, "R", .false.)
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, *, err = 100, end = 100) (upf%r(ir), ir=1,upf%mesh )
|
2003-01-20 05:58:50 +08:00
|
|
|
call scan_end (iunps, "R")
|
|
|
|
|
call scan_begin (iunps, "RAB", .false.)
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, *, err = 101, end = 101) (upf%rab(ir), ir=1,upf%mesh )
|
2003-01-20 05:58:50 +08:00
|
|
|
call scan_end (iunps, "RAB")
|
2007-11-20 00:20:26 +08:00
|
|
|
|
2003-01-20 05:58:50 +08:00
|
|
|
return
|
|
|
|
|
|
2020-04-26 03:13:15 +08:00
|
|
|
100 call upf_error ('read_pseudo_mesh', 'Reading pseudo file (R) for '//upf%psd,1)
|
|
|
|
|
101 call upf_error ('read_pseudo_mesh', 'Reading pseudo file (RAB) for '//upf%psd,2)
|
2003-01-20 05:58:50 +08:00
|
|
|
end subroutine read_pseudo_mesh
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
subroutine read_pseudo_nlcc (upf, iunps)
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
implicit none
|
|
|
|
|
!
|
|
|
|
|
integer :: iunps
|
|
|
|
|
TYPE (pseudo_upf), INTENT(INOUT) :: upf
|
|
|
|
|
!
|
2005-11-03 23:59:28 +08:00
|
|
|
integer :: ir
|
2003-01-20 05:58:50 +08:00
|
|
|
!
|
2007-10-08 15:33:57 +08:00
|
|
|
ALLOCATE( upf%rho_atc( upf%mesh ) )
|
2007-06-12 01:13:15 +08:00
|
|
|
upf%rho_atc = 0.0_DP
|
2003-01-20 05:58:50 +08:00
|
|
|
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, *, err = 100, end = 100) (upf%rho_atc(ir), ir=1,upf%mesh )
|
2003-01-20 05:58:50 +08:00
|
|
|
!
|
|
|
|
|
return
|
|
|
|
|
|
2020-04-26 03:13:15 +08:00
|
|
|
100 call upf_error ('read_pseudo_nlcc', 'Reading pseudo file', 1)
|
2003-01-20 05:58:50 +08:00
|
|
|
return
|
|
|
|
|
end subroutine read_pseudo_nlcc
|
|
|
|
|
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
subroutine read_pseudo_local (upf, iunps)
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
implicit none
|
|
|
|
|
!
|
|
|
|
|
integer :: iunps
|
|
|
|
|
TYPE (pseudo_upf), INTENT(INOUT) :: upf
|
|
|
|
|
!
|
2005-11-03 23:59:28 +08:00
|
|
|
integer :: ir
|
2003-01-20 05:58:50 +08:00
|
|
|
!
|
2007-10-08 15:33:57 +08:00
|
|
|
ALLOCATE( upf%vloc( upf%mesh ) )
|
2007-06-12 01:13:15 +08:00
|
|
|
upf%vloc = 0.0_DP
|
2003-01-20 05:58:50 +08:00
|
|
|
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, *, err=100, end=100) (upf%vloc(ir) , ir=1,upf%mesh )
|
2003-01-20 05:58:50 +08:00
|
|
|
|
|
|
|
|
return
|
|
|
|
|
|
2020-04-26 03:13:15 +08:00
|
|
|
100 call upf_error ('read_pseudo_local','Reading pseudo file', 1)
|
2003-01-20 05:58:50 +08:00
|
|
|
return
|
|
|
|
|
end subroutine read_pseudo_local
|
|
|
|
|
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
|
|
|
|
|
subroutine read_pseudo_nl (upf, iunps)
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
implicit none
|
|
|
|
|
!
|
|
|
|
|
integer :: iunps
|
|
|
|
|
TYPE (pseudo_upf), INTENT(INOUT) :: upf
|
|
|
|
|
!
|
2008-04-04 17:12:30 +08:00
|
|
|
integer :: nb, mb, ijv, n, ir, ios, idum, ldum, icon, lp, i, ikk, l, l1,l2, nd
|
2003-01-20 05:58:50 +08:00
|
|
|
! counters
|
|
|
|
|
character (len=75) :: dummy
|
|
|
|
|
!
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
! Threshold for qfunc to be considered zero (inserted in version UPF v2)
|
|
|
|
|
upf%qqq_eps = -1._dp
|
|
|
|
|
!
|
2003-03-19 18:48:33 +08:00
|
|
|
if ( upf%nbeta == 0) then
|
|
|
|
|
upf%nqf = 0
|
|
|
|
|
upf%nqlc= 0
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
upf%kkbeta = 0
|
The following pseudopotential-related variables in module uspp_param:
zp, psd, dion, betar, jjj, qqq, qfunc, qfcoef, vloc_at, rinner,
nbeta, kkbeta, nqf, nqlc, lll, tvanp
have been replaced by the corresponding variables in structure 'upf'.
There shouldn't be any side effects, but who knows. There is still a
copy of the above variables that will be removed sooner or later.
Basically : variable([i,j,k,..,]n) => upf(n)%variable [(i,j,k,..)]
Note that upf%qfunc has for the time being three indices instead of two,
and that upf%kkbeta is the analogous of kkbeta and not what it used to be.
The logic of this operation will be clearer when it will be completed
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4308 c92efa57-630b-4861-b058-cf58834340f0
2007-10-05 17:26:23 +08:00
|
|
|
ALLOCATE( upf%kbeta( 1 ) )
|
2004-02-16 17:53:00 +08:00
|
|
|
ALLOCATE( upf%lll( 1 ) )
|
2007-10-08 15:33:57 +08:00
|
|
|
ALLOCATE( upf%beta( upf%mesh, 1 ) )
|
2004-02-16 17:53:00 +08:00
|
|
|
ALLOCATE( upf%dion( 1, 1 ) )
|
|
|
|
|
ALLOCATE( upf%rinner( 1 ) )
|
|
|
|
|
ALLOCATE( upf%qqq ( 1, 1 ) )
|
2007-10-08 15:33:57 +08:00
|
|
|
ALLOCATE( upf%qfunc ( upf%mesh, 1 ) )
|
2004-02-16 17:53:00 +08:00
|
|
|
ALLOCATE( upf%qfcoef( 1, 1, 1, 1 ) )
|
2005-02-13 01:08:42 +08:00
|
|
|
ALLOCATE( upf%rcut( 1 ) )
|
|
|
|
|
ALLOCATE( upf%rcutus( 1 ) )
|
|
|
|
|
ALLOCATE( upf%els_beta( 1 ) )
|
2003-03-19 18:48:33 +08:00
|
|
|
return
|
|
|
|
|
end if
|
The following pseudopotential-related variables in module uspp_param:
zp, psd, dion, betar, jjj, qqq, qfunc, qfcoef, vloc_at, rinner,
nbeta, kkbeta, nqf, nqlc, lll, tvanp
have been replaced by the corresponding variables in structure 'upf'.
There shouldn't be any side effects, but who knows. There is still a
copy of the above variables that will be removed sooner or later.
Basically : variable([i,j,k,..,]n) => upf(n)%variable [(i,j,k,..)]
Note that upf%qfunc has for the time being three indices instead of two,
and that upf%kkbeta is the analogous of kkbeta and not what it used to be.
The logic of this operation will be clearer when it will be completed
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4308 c92efa57-630b-4861-b058-cf58834340f0
2007-10-05 17:26:23 +08:00
|
|
|
ALLOCATE( upf%kbeta( upf%nbeta ) )
|
2003-01-20 05:58:50 +08:00
|
|
|
ALLOCATE( upf%lll( upf%nbeta ) )
|
2007-10-08 15:33:57 +08:00
|
|
|
ALLOCATE( upf%beta( upf%mesh, upf%nbeta ) )
|
2003-01-20 05:58:50 +08:00
|
|
|
ALLOCATE( upf%dion( upf%nbeta, upf%nbeta ) )
|
2005-02-13 01:08:42 +08:00
|
|
|
ALLOCATE( upf%rcut( upf%nbeta ) )
|
|
|
|
|
ALLOCATE( upf%rcutus( upf%nbeta ) )
|
|
|
|
|
ALLOCATE( upf%els_beta( upf%nbeta ) )
|
|
|
|
|
|
2003-01-20 05:58:50 +08:00
|
|
|
upf%kkbeta = 0
|
|
|
|
|
upf%lll = 0
|
2007-06-12 01:13:15 +08:00
|
|
|
upf%beta = 0.0_DP
|
|
|
|
|
upf%dion = 0.0_DP
|
|
|
|
|
upf%rcut = 0.0_DP
|
|
|
|
|
upf%rcutus = 0.0_DP
|
2005-02-13 01:08:42 +08:00
|
|
|
upf%els_beta = ' '
|
2003-01-20 05:58:50 +08:00
|
|
|
|
|
|
|
|
do nb = 1, upf%nbeta
|
|
|
|
|
call scan_begin (iunps, "BETA", .false.)
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, *, err = 100, end = 100) idum, upf%lll(nb), dummy
|
2006-01-19 00:15:26 +08:00
|
|
|
read (iunps, *, err = 100, end = 100) ikk
|
The following pseudopotential-related variables in module uspp_param:
zp, psd, dion, betar, jjj, qqq, qfunc, qfcoef, vloc_at, rinner,
nbeta, kkbeta, nqf, nqlc, lll, tvanp
have been replaced by the corresponding variables in structure 'upf'.
There shouldn't be any side effects, but who knows. There is still a
copy of the above variables that will be removed sooner or later.
Basically : variable([i,j,k,..,]n) => upf(n)%variable [(i,j,k,..)]
Note that upf%qfunc has for the time being three indices instead of two,
and that upf%kkbeta is the analogous of kkbeta and not what it used to be.
The logic of this operation will be clearer when it will be completed
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4308 c92efa57-630b-4861-b058-cf58834340f0
2007-10-05 17:26:23 +08:00
|
|
|
upf%kbeta(nb) = ikk
|
|
|
|
|
upf%kkbeta = MAX ( upf%kkbeta, upf%kbeta(nb) )
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, *, err = 100, end = 100) (upf%beta(ir,nb), ir=1,ikk)
|
2005-02-13 01:08:42 +08:00
|
|
|
|
2006-01-19 00:15:26 +08:00
|
|
|
read (iunps, *, err=200,iostat=ios) upf%rcut(nb), upf%rcutus(nb)
|
|
|
|
|
read (iunps, *, err=200,iostat=ios) upf%els_beta(nb)
|
2003-01-20 05:58:50 +08:00
|
|
|
call scan_end (iunps, "BETA")
|
2005-02-13 01:08:42 +08:00
|
|
|
200 continue
|
2003-01-20 05:58:50 +08:00
|
|
|
enddo
|
|
|
|
|
|
2004-02-16 17:53:00 +08:00
|
|
|
|
2003-01-20 05:58:50 +08:00
|
|
|
call scan_begin (iunps, "DIJ", .false.)
|
2008-04-04 17:12:30 +08:00
|
|
|
read (iunps, *, err = 101, end = 101) nd, dummy
|
|
|
|
|
do icon = 1, nd
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, *, err = 101, end = 101) nb, mb, upf%dion(nb,mb)
|
2003-01-20 05:58:50 +08:00
|
|
|
upf%dion (mb,nb) = upf%dion (nb,mb)
|
|
|
|
|
enddo
|
|
|
|
|
call scan_end (iunps, "DIJ")
|
|
|
|
|
|
2004-02-16 17:53:00 +08:00
|
|
|
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
if ( upf%tvanp .or. upf%tpawp) then
|
2003-01-20 05:58:50 +08:00
|
|
|
call scan_begin (iunps, "QIJ", .false.)
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, *, err = 102, end = 102) upf%nqf
|
2003-01-20 05:58:50 +08:00
|
|
|
upf%nqlc = 2 * upf%lmax + 1
|
|
|
|
|
ALLOCATE( upf%rinner( upf%nqlc ) )
|
|
|
|
|
ALLOCATE( upf%qqq ( upf%nbeta, upf%nbeta ) )
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
IF (upf%q_with_l .or. upf%tpawp) then
|
2007-11-14 01:42:29 +08:00
|
|
|
ALLOCATE( upf%qfuncl ( upf%mesh, upf%nbeta*(upf%nbeta+1)/2, 0:2*upf%lmax ) )
|
|
|
|
|
upf%qfuncl = 0.0_DP
|
|
|
|
|
ELSE
|
|
|
|
|
ALLOCATE( upf%qfunc ( upf%mesh, upf%nbeta*(upf%nbeta+1)/2 ) )
|
|
|
|
|
upf%qfunc = 0.0_DP
|
|
|
|
|
ENDIF
|
2020-04-28 16:15:50 +08:00
|
|
|
IF ( upf%nqf > 0 ) THEN
|
|
|
|
|
ALLOCATE( upf%qfcoef(upf%nqf, upf%nqlc, upf%nbeta, upf%nbeta ) )
|
|
|
|
|
ELSE
|
|
|
|
|
ALLOCATE( upf%qfcoef(1,1,1,1) )
|
|
|
|
|
ENDIF
|
2007-06-12 01:13:15 +08:00
|
|
|
upf%rinner = 0.0_DP
|
|
|
|
|
upf%qqq = 0.0_DP
|
|
|
|
|
upf%qfcoef = 0.0_DP
|
2020-04-28 16:15:50 +08:00
|
|
|
if ( upf%nqf > 0) then
|
2003-01-20 05:58:50 +08:00
|
|
|
call scan_begin (iunps, "RINNER", .false.)
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps,*,err=103,end=103) ( idum, upf%rinner(i), i=1,upf%nqlc )
|
2003-01-20 05:58:50 +08:00
|
|
|
call scan_end (iunps, "RINNER")
|
|
|
|
|
end if
|
|
|
|
|
do nb = 1, upf%nbeta
|
|
|
|
|
do mb = nb, upf%nbeta
|
|
|
|
|
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps,*,err=102,end=102) idum, idum, ldum, dummy
|
2003-01-20 05:58:50 +08:00
|
|
|
!" i j (l)"
|
|
|
|
|
if (ldum /= upf%lll(mb) ) then
|
2020-04-26 03:13:15 +08:00
|
|
|
call upf_error ('read_pseudo_nl','inconsistent angular momentum for Q_ij', 1)
|
2003-01-20 05:58:50 +08:00
|
|
|
end if
|
|
|
|
|
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps,*,err=104,end=104) upf%qqq(nb,mb), dummy
|
2003-01-20 05:58:50 +08:00
|
|
|
! "Q_int"
|
|
|
|
|
upf%qqq(mb,nb) = upf%qqq(nb,mb)
|
2007-10-06 16:23:39 +08:00
|
|
|
! ijv is the combined (nb,mb) index
|
|
|
|
|
ijv = mb * (mb-1) / 2 + nb
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
IF (upf%q_with_l .or. upf%tpawp) THEN
|
2007-11-14 01:42:29 +08:00
|
|
|
l1=upf%lll(nb)
|
|
|
|
|
l2=upf%lll(mb)
|
|
|
|
|
DO l=abs(l1-l2),l1+l2
|
|
|
|
|
read (iunps, *, err=105, end=105) (upf%qfuncl(n,ijv,l), &
|
|
|
|
|
n=1,upf%mesh)
|
|
|
|
|
END DO
|
|
|
|
|
ELSE
|
|
|
|
|
read (iunps, *, err=105, end=105) (upf%qfunc(n,ijv), n=1,upf%mesh)
|
|
|
|
|
ENDIF
|
2003-01-20 05:58:50 +08:00
|
|
|
|
|
|
|
|
if ( upf%nqf > 0 ) then
|
|
|
|
|
call scan_begin (iunps, "QFCOEF", .false.)
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps,*,err=106,end=106) &
|
2003-01-20 05:58:50 +08:00
|
|
|
( ( upf%qfcoef(i,lp,nb,mb), i=1,upf%nqf ), lp=1,upf%nqlc )
|
2006-04-18 15:33:11 +08:00
|
|
|
do i = 1, upf%nqf
|
|
|
|
|
do lp = 1, upf%nqlc
|
|
|
|
|
upf%qfcoef(i,lp,mb,nb) = upf%qfcoef(i,lp,nb,mb)
|
|
|
|
|
end do
|
|
|
|
|
end do
|
2003-01-20 05:58:50 +08:00
|
|
|
call scan_end (iunps, "QFCOEF")
|
|
|
|
|
end if
|
|
|
|
|
|
|
|
|
|
enddo
|
|
|
|
|
enddo
|
|
|
|
|
call scan_end (iunps, "QIJ")
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
else
|
2003-01-20 05:58:50 +08:00
|
|
|
upf%nqf = 1
|
|
|
|
|
upf%nqlc = 2 * upf%lmax + 1
|
|
|
|
|
ALLOCATE( upf%rinner( upf%nqlc ) )
|
|
|
|
|
ALLOCATE( upf%qqq ( upf%nbeta, upf%nbeta ) )
|
2007-10-08 15:33:57 +08:00
|
|
|
ALLOCATE( upf%qfunc ( upf%mesh, upf%nbeta*(upf%nbeta+1)/2 ) )
|
2003-01-20 05:58:50 +08:00
|
|
|
ALLOCATE( upf%qfcoef( upf%nqf, upf%nqlc, upf%nbeta, upf%nbeta ) )
|
2007-06-12 01:13:15 +08:00
|
|
|
upf%rinner = 0.0_DP
|
|
|
|
|
upf%qqq = 0.0_DP
|
|
|
|
|
upf%qfunc = 0.0_DP
|
|
|
|
|
upf%qfcoef = 0.0_DP
|
2003-01-20 05:58:50 +08:00
|
|
|
endif
|
|
|
|
|
|
2004-02-16 17:53:00 +08:00
|
|
|
|
2003-01-20 05:58:50 +08:00
|
|
|
return
|
|
|
|
|
|
2020-04-26 03:13:15 +08:00
|
|
|
100 call upf_error ('read_pseudo_nl', 'Reading pseudo file (BETA)', 1 )
|
|
|
|
|
101 call upf_error ('read_pseudo_nl', 'Reading pseudo file (DIJ)', 2 )
|
|
|
|
|
102 call upf_error ('read_pseudo_nl', 'Reading pseudo file (QIJ)', 3 )
|
|
|
|
|
103 call upf_error ('read_pseudo_nl', 'Reading pseudo file (RINNER)',4)
|
|
|
|
|
104 call upf_error ('read_pseudo_nl', 'Reading pseudo file (qqq)', 5 )
|
|
|
|
|
105 call upf_error ('read_pseudo_nl', 'Reading pseudo file (qfunc)',6 )
|
|
|
|
|
106 call upf_error ('read_pseudo_nl', 'Reading pseudo file (qfcoef)',7)
|
2003-01-20 05:58:50 +08:00
|
|
|
end subroutine read_pseudo_nl
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
subroutine read_pseudo_pswfc (upf, iunps)
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
implicit none
|
|
|
|
|
!
|
|
|
|
|
integer :: iunps
|
|
|
|
|
TYPE (pseudo_upf), INTENT(INOUT) :: upf
|
|
|
|
|
!
|
|
|
|
|
character (len=75) :: dummy
|
2005-11-03 23:59:28 +08:00
|
|
|
integer :: nb, ir
|
2003-01-20 05:58:50 +08:00
|
|
|
|
2007-10-08 15:33:57 +08:00
|
|
|
ALLOCATE( upf%chi( upf%mesh, MAX( upf%nwfc, 1 ) ) )
|
2007-06-12 01:13:15 +08:00
|
|
|
upf%chi = 0.0_DP
|
2003-01-20 05:58:50 +08:00
|
|
|
do nb = 1, upf%nwfc
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps, *, err=100, end=100) dummy !Wavefunction labels
|
|
|
|
|
read (iunps, *, err=100, end=100) ( upf%chi(ir,nb), ir=1,upf%mesh )
|
2003-01-20 05:58:50 +08:00
|
|
|
enddo
|
|
|
|
|
|
|
|
|
|
return
|
|
|
|
|
|
2020-04-26 03:13:15 +08:00
|
|
|
100 call upf_error ('read_pseudo_pswfc', 'Reading pseudo file', 1)
|
2003-01-20 05:58:50 +08:00
|
|
|
end subroutine read_pseudo_pswfc
|
|
|
|
|
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
subroutine read_pseudo_rhoatom (upf, iunps)
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
implicit none
|
|
|
|
|
!
|
|
|
|
|
integer :: iunps
|
|
|
|
|
TYPE (pseudo_upf), INTENT(INOUT) :: upf
|
|
|
|
|
!
|
2005-11-03 23:59:28 +08:00
|
|
|
integer :: ir
|
2003-01-20 05:58:50 +08:00
|
|
|
!
|
2007-10-08 15:33:57 +08:00
|
|
|
ALLOCATE( upf%rho_at( upf%mesh ) )
|
2007-06-12 01:13:15 +08:00
|
|
|
upf%rho_at = 0.0_DP
|
2005-11-03 23:59:28 +08:00
|
|
|
read (iunps,*,err=100,end=100) ( upf%rho_at(ir), ir=1,upf%mesh )
|
2003-01-20 05:58:50 +08:00
|
|
|
!
|
|
|
|
|
return
|
|
|
|
|
|
2020-04-26 03:13:15 +08:00
|
|
|
100 call upf_error ('read_pseudo_rhoatom','Reading pseudo file', 1)
|
2003-01-20 05:58:50 +08:00
|
|
|
end subroutine read_pseudo_rhoatom
|
2004-04-30 21:16:03 +08:00
|
|
|
!
|
|
|
|
|
!---------------------------------------------------------------------
|
2006-01-27 00:42:14 +08:00
|
|
|
subroutine read_pseudo_addinfo (upf, iunps)
|
2004-04-30 21:16:03 +08:00
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
! This routine reads from the new UPF file,
|
2006-01-27 00:42:14 +08:00
|
|
|
! and the total angular momentum jjj of the beta and jchi of the
|
2004-04-30 21:16:03 +08:00
|
|
|
! wave-functions.
|
|
|
|
|
!
|
2006-01-27 00:42:14 +08:00
|
|
|
implicit none
|
|
|
|
|
integer :: iunps
|
|
|
|
|
|
|
|
|
|
TYPE (pseudo_upf), INTENT(INOUT) :: upf
|
|
|
|
|
integer :: nb
|
|
|
|
|
|
|
|
|
|
ALLOCATE( upf%nn(upf%nwfc) )
|
2020-04-27 21:03:58 +08:00
|
|
|
ALLOCATE( upf%jchi(upf%nwfc) )
|
2006-01-27 00:42:14 +08:00
|
|
|
ALLOCATE( upf%jjj(upf%nbeta) )
|
|
|
|
|
|
|
|
|
|
upf%nn=0
|
2007-06-12 01:13:15 +08:00
|
|
|
upf%jchi=0.0_DP
|
2006-01-27 00:42:14 +08:00
|
|
|
do nb = 1, upf%nwfc
|
|
|
|
|
read (iunps, *,err=100,end=100) upf%els(nb), &
|
|
|
|
|
upf%nn(nb), upf%lchi(nb), upf%jchi(nb), upf%oc(nb)
|
2008-01-24 04:04:43 +08:00
|
|
|
if ( abs ( upf%jchi(nb)-upf%lchi(nb)-0.5_dp ) > 1.0d-7 .and. &
|
|
|
|
|
abs ( upf%jchi(nb)-upf%lchi(nb)+0.5_dp ) > 1.0d-7 ) then
|
2020-04-26 03:13:15 +08:00
|
|
|
call upf_error ( 'read_pseudo_upf', 'obsolete ADDINFO section ignored',-1)
|
2008-01-24 04:04:43 +08:00
|
|
|
upf%has_so = .false.
|
|
|
|
|
return
|
|
|
|
|
end if
|
2006-01-27 00:42:14 +08:00
|
|
|
enddo
|
|
|
|
|
|
2007-06-12 01:13:15 +08:00
|
|
|
upf%jjj=0.0_DP
|
2006-01-27 00:42:14 +08:00
|
|
|
do nb = 1, upf%nbeta
|
|
|
|
|
read (iunps, *, err=100,end=100) upf%lll(nb), upf%jjj(nb)
|
2008-01-24 04:04:43 +08:00
|
|
|
if ( abs ( upf%lll(nb)-upf%jjj(nb)-0.5_dp) > 1.0d-7 .and. &
|
|
|
|
|
abs ( upf%lll(nb)-upf%jjj(nb)+0.5_dp) > 1.0d-7 ) then
|
2020-04-26 03:13:15 +08:00
|
|
|
call upf_error ( 'read_pseudo_upf', 'obsolete ADDINFO section ignored',-1)
|
2008-01-24 04:04:43 +08:00
|
|
|
upf%has_so = .false.
|
|
|
|
|
return
|
|
|
|
|
end if
|
2006-01-27 00:42:14 +08:00
|
|
|
enddo
|
|
|
|
|
|
|
|
|
|
read(iunps, *) upf%xmin, upf%rmax, upf%zmesh, upf%dx
|
2005-02-13 01:08:42 +08:00
|
|
|
|
2006-01-27 00:42:14 +08:00
|
|
|
return
|
2020-04-26 03:13:15 +08:00
|
|
|
100 call upf_error ('read_pseudo_addinfo','Reading pseudo file', 1)
|
2004-04-30 21:16:03 +08:00
|
|
|
end subroutine read_pseudo_addinfo
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
!
|
2007-02-22 00:54:18 +08:00
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
SUBROUTINE read_pseudo_gipaw ( upf, iunps )
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
implicit none
|
|
|
|
|
!
|
|
|
|
|
INTEGER :: iunps
|
2013-11-16 04:49:02 +08:00
|
|
|
REAL (dp) :: version
|
2007-02-22 00:54:18 +08:00
|
|
|
TYPE ( pseudo_upf ), INTENT ( INOUT ) :: upf
|
|
|
|
|
!
|
|
|
|
|
CALL scan_begin ( iunps, "GIPAW_FORMAT_VERSION", .false. )
|
2013-11-16 04:49:02 +08:00
|
|
|
READ ( iunps, *, err=100, end=100 ) version
|
|
|
|
|
upf%gipaw_data_format = INT(version)
|
2007-02-22 00:54:18 +08:00
|
|
|
CALL scan_end ( iunps, "GIPAW_FORMAT_VERSION" )
|
|
|
|
|
|
2010-03-23 23:20:49 +08:00
|
|
|
IF ( upf%gipaw_data_format == 1 .or. upf%gipaw_data_format == 0 ) THEN
|
2007-02-22 00:54:18 +08:00
|
|
|
CALL read_pseudo_gipaw_core_orbitals ( upf, iunps )
|
|
|
|
|
CALL read_pseudo_gipaw_local ( upf, iunps )
|
|
|
|
|
CALL read_pseudo_gipaw_orbitals ( upf, iunps )
|
|
|
|
|
ELSE
|
2020-04-26 03:13:15 +08:00
|
|
|
CALL upf_error ( 'read_pseudo_gipaw', 'UPF/GIPAW in unknown format', 1 )
|
2007-02-22 00:54:18 +08:00
|
|
|
END IF
|
|
|
|
|
|
|
|
|
|
RETURN
|
|
|
|
|
|
2020-04-26 03:13:15 +08:00
|
|
|
100 CALL upf_error ( 'read_pseudo_gipaw', 'Reading pseudo file', 1 )
|
2007-02-22 00:54:18 +08:00
|
|
|
END SUBROUTINE read_pseudo_gipaw
|
|
|
|
|
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
SUBROUTINE read_pseudo_gipaw_core_orbitals ( upf, iunps )
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
IMPLICIT NONE
|
|
|
|
|
!
|
|
|
|
|
INTEGER :: iunps
|
|
|
|
|
TYPE ( pseudo_upf ), INTENT ( INOUT ) :: upf
|
|
|
|
|
!
|
|
|
|
|
CHARACTER ( LEN = 75 ) :: dummy1, dummy2
|
|
|
|
|
INTEGER :: nb, ir
|
|
|
|
|
|
|
|
|
|
CALL scan_begin ( iunps, "GIPAW_CORE_ORBITALS", .false. )
|
|
|
|
|
READ ( iunps, *, err=100, end=100 ) upf%gipaw_ncore_orbitals
|
|
|
|
|
|
|
|
|
|
ALLOCATE ( upf%gipaw_core_orbital_n(upf%gipaw_ncore_orbitals) )
|
|
|
|
|
ALLOCATE ( upf%gipaw_core_orbital_l(upf%gipaw_ncore_orbitals) )
|
|
|
|
|
ALLOCATE ( upf%gipaw_core_orbital_el(upf%gipaw_ncore_orbitals) )
|
|
|
|
|
ALLOCATE ( upf%gipaw_core_orbital(upf%mesh,upf%gipaw_ncore_orbitals) )
|
|
|
|
|
upf%gipaw_core_orbital = 0.0_dp
|
|
|
|
|
|
|
|
|
|
DO nb = 1, upf%gipaw_ncore_orbitals
|
|
|
|
|
CALL scan_begin ( iunps, "GIPAW_CORE_ORBITAL", .false. )
|
|
|
|
|
READ (iunps, *, err=100, end=100) &
|
|
|
|
|
upf%gipaw_core_orbital_n(nb), upf%gipaw_core_orbital_l(nb), &
|
|
|
|
|
dummy1, dummy2, upf%gipaw_core_orbital_el(nb)
|
|
|
|
|
READ ( iunps, *, err=100, end=100 ) &
|
|
|
|
|
( upf%gipaw_core_orbital(ir,nb), ir = 1, upf%mesh )
|
|
|
|
|
CALL scan_end ( iunps, "GIPAW_CORE_ORBITAL" )
|
|
|
|
|
END DO
|
|
|
|
|
|
|
|
|
|
CALL scan_end ( iunps, "GIPAW_CORE_ORBITALS" )
|
|
|
|
|
|
|
|
|
|
RETURN
|
|
|
|
|
|
2020-04-26 03:13:15 +08:00
|
|
|
100 CALL upf_error ( 'read_pseudo_gipaw_core_orbitals', 'Reading pseudo file', 1 )
|
2007-02-22 00:54:18 +08:00
|
|
|
END SUBROUTINE read_pseudo_gipaw_core_orbitals
|
|
|
|
|
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
SUBROUTINE read_pseudo_gipaw_local ( upf, iunps )
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
IMPLICIT NONE
|
|
|
|
|
!
|
|
|
|
|
INTEGER :: iunps
|
|
|
|
|
TYPE ( pseudo_upf ), INTENT ( INOUT ) :: upf
|
|
|
|
|
!
|
2009-02-26 00:51:30 +08:00
|
|
|
INTEGER :: ir
|
2007-02-22 00:54:18 +08:00
|
|
|
|
|
|
|
|
CALL scan_begin ( iunps, "GIPAW_LOCAL_DATA", .false. )
|
|
|
|
|
|
|
|
|
|
ALLOCATE ( upf%gipaw_vlocal_ae(upf%mesh) )
|
|
|
|
|
ALLOCATE ( upf%gipaw_vlocal_ps(upf%mesh) )
|
|
|
|
|
|
|
|
|
|
CALL scan_begin ( iunps, "GIPAW_VLOCAL_AE", .false. )
|
|
|
|
|
|
|
|
|
|
READ ( iunps, *, err=100, end=100 ) &
|
|
|
|
|
( upf%gipaw_vlocal_ae(ir), ir = 1, upf%mesh )
|
|
|
|
|
|
|
|
|
|
CALL scan_end ( iunps, "GIPAW_VLOCAL_AE" )
|
|
|
|
|
|
|
|
|
|
CALL scan_begin ( iunps, "GIPAW_VLOCAL_PS", .false. )
|
|
|
|
|
|
|
|
|
|
READ ( iunps, *, err=100, end=100 ) &
|
|
|
|
|
( upf%gipaw_vlocal_ps(ir), ir = 1, upf%mesh )
|
|
|
|
|
|
|
|
|
|
CALL scan_end ( iunps, "GIPAW_VLOCAL_PS" )
|
|
|
|
|
|
|
|
|
|
CALL scan_end ( iunps, "GIPAW_LOCAL_DATA" )
|
|
|
|
|
|
|
|
|
|
RETURN
|
|
|
|
|
|
2020-04-26 03:13:15 +08:00
|
|
|
100 CALL upf_error ( 'read_pseudo_gipaw_local', 'Reading pseudo file', 1 )
|
2007-02-22 00:54:18 +08:00
|
|
|
END SUBROUTINE read_pseudo_gipaw_local
|
|
|
|
|
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
SUBROUTINE read_pseudo_gipaw_orbitals ( upf, iunps )
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
IMPLICIT NONE
|
|
|
|
|
!
|
|
|
|
|
INTEGER :: iunps
|
|
|
|
|
TYPE ( pseudo_upf ), INTENT ( INOUT ) :: upf
|
|
|
|
|
!
|
|
|
|
|
CHARACTER ( LEN = 75 ) :: dummy
|
|
|
|
|
INTEGER :: nb, ir
|
|
|
|
|
|
|
|
|
|
CALL scan_begin ( iunps, "GIPAW_ORBITALS", .false. )
|
|
|
|
|
READ ( iunps, *, err=100, end=100 ) upf%gipaw_wfs_nchannels
|
|
|
|
|
|
|
|
|
|
ALLOCATE ( upf%gipaw_wfs_el(upf%gipaw_wfs_nchannels) )
|
|
|
|
|
ALLOCATE ( upf%gipaw_wfs_ll(upf%gipaw_wfs_nchannels) )
|
|
|
|
|
ALLOCATE ( upf%gipaw_wfs_rcut(upf%gipaw_wfs_nchannels) )
|
|
|
|
|
ALLOCATE ( upf%gipaw_wfs_rcutus(upf%gipaw_wfs_nchannels) )
|
|
|
|
|
ALLOCATE ( upf%gipaw_wfs_ae(upf%mesh,upf%gipaw_wfs_nchannels) )
|
|
|
|
|
ALLOCATE ( upf%gipaw_wfs_ps(upf%mesh,upf%gipaw_wfs_nchannels) )
|
|
|
|
|
|
|
|
|
|
inquire ( unit = iunps, name = dummy )
|
|
|
|
|
DO nb = 1, upf%gipaw_wfs_nchannels
|
|
|
|
|
CALL scan_begin ( iunps, "GIPAW_AE_ORBITAL", .false. )
|
|
|
|
|
READ (iunps, *, err=100, end=100) &
|
|
|
|
|
upf%gipaw_wfs_el(nb), upf%gipaw_wfs_ll(nb)
|
|
|
|
|
READ ( iunps, *, err=100, end=100 ) &
|
|
|
|
|
( upf%gipaw_wfs_ae(ir,nb), ir = 1, upf%mesh )
|
|
|
|
|
CALL scan_end ( iunps, "GIPAW_AE_ORBITAL" )
|
|
|
|
|
|
|
|
|
|
CALL scan_begin ( iunps, "GIPAW_PS_ORBITAL", .false. )
|
|
|
|
|
READ (iunps, *, err=100, end=100) &
|
|
|
|
|
upf%gipaw_wfs_rcut(nb), upf%gipaw_wfs_rcutus(nb)
|
|
|
|
|
READ ( iunps, *, err=100, end=100 ) &
|
|
|
|
|
( upf%gipaw_wfs_ps(ir,nb), ir = 1, upf%mesh )
|
|
|
|
|
CALL scan_end ( iunps, "GIPAW_PS_ORBITAL" )
|
|
|
|
|
END DO
|
|
|
|
|
|
|
|
|
|
CALL scan_end ( iunps, "GIPAW_ORBITALS" )
|
|
|
|
|
|
|
|
|
|
RETURN
|
|
|
|
|
|
2020-04-26 03:13:15 +08:00
|
|
|
100 CALL upf_error ( 'read_pseudo_gipaw_orbitals', 'Reading pseudo file', 1 )
|
2007-02-22 00:54:18 +08:00
|
|
|
END SUBROUTINE read_pseudo_gipaw_orbitals
|
|
|
|
|
!</apsi>
|
|
|
|
|
|
2007-11-21 22:41:30 +08:00
|
|
|
subroutine read_pseudo_ppinfo (upf, iunps)
|
|
|
|
|
!---------------------------------------------------------------------
|
|
|
|
|
!
|
|
|
|
|
implicit none
|
|
|
|
|
!
|
|
|
|
|
TYPE (pseudo_upf), INTENT(INOUT) :: upf
|
|
|
|
|
integer :: iunps
|
|
|
|
|
character (len=80) :: dummy
|
|
|
|
|
real(dp) :: rdummy
|
|
|
|
|
integer :: idummy, nb, ios
|
|
|
|
|
|
|
|
|
|
ios=0
|
|
|
|
|
DO while (ios==0)
|
|
|
|
|
READ (iunps, '(a)', err = 100, end = 100, iostat=ios) dummy
|
|
|
|
|
IF (matches ("Rcut", dummy) ) THEN
|
|
|
|
|
DO nb=1,upf%nbeta
|
|
|
|
|
READ (iunps, '(a2,2i3,f6.2,3f19.11)',err=100, end=100,iostat=ios) &
|
|
|
|
|
upf%els_beta(nb), idummy, &
|
|
|
|
|
idummy, rdummy, upf%rcut(nb), upf%rcutus (nb), rdummy
|
|
|
|
|
ENDDO
|
|
|
|
|
ios=100
|
2020-10-21 18:28:27 +08:00
|
|
|
ELSE
|
|
|
|
|
nb = 1
|
2007-11-21 22:41:30 +08:00
|
|
|
ENDIF
|
|
|
|
|
ENDDO
|
2020-10-19 17:55:46 +08:00
|
|
|
100 CONTINUE
|
|
|
|
|
! PP_INFO reports values only for bound valence states,
|
2020-10-21 18:28:27 +08:00
|
|
|
IF ( nb .LE. upf%nbeta ) THEN
|
|
|
|
|
IF (upf%els_beta(nb) == "</") THEN
|
|
|
|
|
upf%els_beta(nb:upf%nbeta)="__"
|
|
|
|
|
upf%rcut(nb:upf%nbeta)=-1.0
|
|
|
|
|
upf%rcutus(nb:upf%nbeta) = -1.0
|
|
|
|
|
END IF
|
2020-10-19 17:55:46 +08:00
|
|
|
END IF
|
|
|
|
|
RETURN
|
|
|
|
|
|
2007-11-21 22:41:30 +08:00
|
|
|
END SUBROUTINE read_pseudo_ppinfo
|
|
|
|
|
|
2012-04-07 01:02:43 +08:00
|
|
|
SUBROUTINE set_coulomb_nonlocal(upf)
|
|
|
|
|
IMPLICIT NONE
|
|
|
|
|
TYPE(pseudo_upf) :: upf
|
|
|
|
|
|
|
|
|
|
upf%nqf = 0
|
|
|
|
|
upf%nqlc= 0
|
|
|
|
|
upf%qqq_eps= -1._dp
|
|
|
|
|
upf%kkbeta = 0
|
|
|
|
|
ALLOCATE( upf%kbeta(1), &
|
|
|
|
|
upf%lll(1), &
|
|
|
|
|
upf%beta(upf%mesh,1), &
|
|
|
|
|
upf%dion(1,1), &
|
|
|
|
|
upf%rinner(1), &
|
|
|
|
|
upf%qqq(1,1), &
|
|
|
|
|
upf%qfunc(upf%mesh,1),&
|
|
|
|
|
upf%qfcoef(1,1,1,1), &
|
|
|
|
|
upf%rcut(1), &
|
|
|
|
|
upf%rcutus(1), &
|
|
|
|
|
upf%els_beta(1) )
|
|
|
|
|
RETURN
|
|
|
|
|
END SUBROUTINE set_coulomb_nonlocal
|
2020-04-26 03:13:15 +08:00
|
|
|
!
|
2003-01-20 05:58:50 +08:00
|
|
|
!=----------------------------------------------------------------------------=!
|
This is a quite complex check-in, but actually not very much is done. Changelog follows.
LP
UPF file format updated completely, UPFv2 introduced:
* ld1.x can still produce old format, with the switch upf_v1_format=.true. in inputp
this is disabled by default, but we can discuss if it should be the opposite.
* pw.x cp.x and all utilities should notice no difference
* some utilities in upftools still need to be updated, anyway conversion UPFv1 to UPFv2
is very easy, so this should be no big issue
* starting from now to produce an UPF file you need to fill the pseudo_upf derivedd type
and feed it to write_upf woutine in upf_module (Modules/upf.f90)
* extensive use of iotk
I have tried to make the new format as self contained as possible, e.g. there should be
minimal need for post-processing after the data is read, no more reconstruction of known
quantities, and no more odd syntax to save negligible quantity of space. Also the human
readable section is a bit richer, all the rest is more machine readable.
I hope this will not cause any throuble, and tried really hard to, all examples and all
tests works as fine as before and gives (what really looks like) the same results.
Other changes that I needed to make:
* radial grids are now allocatable, they management is a bit less of a hack too
* paw and uspp augmentation are stored in the same place
* paw print total all-electron energy if all atoms are paw, not very useful, but nice
* most of the pseudopotential-writing reading files have been renamed to some more
logical name, I spare you the list. E.g. read_oldpseudo -> read_pseudo_rrkj3
* paw_t derived type was only used in atomic, so I have put it there (as the pseudo_type
module take ages to recompile it was awkward to leave it there).
PAW tests inserted in test/ there are 6 of them, as a consequence I have also put 5 paw
pseudopotentials in the pseudo/ directory.
I will update the PAW scf examples soon, by deleting them (as running a pw with a PAW
pseudopotential requires no option at all). PAW generation examples should be updated.
A lot of small bugfixes here & there mostly uninitialized variables or unallocated
pointers used as subrotuine arguments.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@4769 c92efa57-630b-4861-b058-cf58834340f0
2008-04-03 23:50:43 +08:00
|
|
|
END MODULE read_upf_v1_module
|
2003-01-20 05:58:50 +08:00
|
|
|
!=----------------------------------------------------------------------------=!
|