scnc
/
quantum-espresso
mirror of https://gitlab.com/QEF/q-e.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

List of g^2 was incorrect in parallel case 

git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@2210 c92efa57-630b-4861-b058-cf58834340f0
This commit is contained in:
giannozz 2005-09-23 08:59:57 +00:00
parent 34f3df5d19
commit 34269e6856
1 changed files with 35 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

46
PP/pw_export.f90
View File

@ -380,7 +380,9 @@ subroutine write_export (pp_file,kunit,uspp_spsi, ascii, single_file, raw)
integer :: npool, nkbl, nkl, nkr, npwx_g integer :: npool, nkbl, nkl, nkr, npwx_g
integer :: ike, iks, npw_g, ispin, local_pw integer :: ike, iks, npw_g, ispin, local_pw
integer, allocatable :: ngk_g( : ) integer, allocatable :: ngk_g( : )
integer, allocatable :: itmp( :, : ) integer, allocatable :: itmp_g( :, : )
real(DP),allocatable :: rtmp_g( :, : )
real(DP),allocatable :: rtmp_gg( : )
integer, allocatable :: itmp1( : ) integer, allocatable :: itmp1( : )
integer, allocatable :: igwk( :, : ) integer, allocatable :: igwk( :, : )
integer, allocatable :: l2g_new( : ) integer, allocatable :: l2g_new( : )
@ -437,14 +439,33 @@ subroutine write_export (pp_file,kunit,uspp_spsi, ascii, single_file, raw)
end if end if
! collect all G vectors across processors within the pools ! collect all G vectors across processors within the pools
allocate( itmp( 3, ngm_g ) ) ! and compute their modules
itmp = 0 !
allocate( itmp_g( 3, ngm_g ) )
allocate( rtmp_g( 3, ngm_g ) )
allocate( rtmp_gg( ngm_g ) )
itmp_g = 0
do ig = 1, ngm do ig = 1, ngm
itmp( 1, ig_l2g( ig ) ) = ig1( ig ) itmp_g( 1, ig_l2g( ig ) ) = ig1( ig )
itmp( 2, ig_l2g( ig ) ) = ig2( ig ) itmp_g( 2, ig_l2g( ig ) ) = ig2( ig )
itmp( 3, ig_l2g( ig ) ) = ig3( ig ) itmp_g( 3, ig_l2g( ig ) ) = ig3( ig )
end do end do
call mp_sum( itmp , intra_pool_comm ) call mp_sum( itmp_g , intra_pool_comm )
!
! here we are in crystal units
rtmp_g(1:3,1:ngm_g) = REAL( itmp_g(1:3,1:ngm_g) )
!
! go to cartesian units (tpiba)
call cryst_to_cart( ngm_g, rtmp_g, bg , 1 )
!
! compute squared moduli
do ig = 1, ngm_g
rtmp_gg(ig) = rtmp_g(1,ig)**2 + rtmp_g(2,ig)**2 + rtmp_g(3,ig)**2
enddo
deallocate( rtmp_g )
! build the G+k array indexes ! build the G+k array indexes
allocate ( kisort( npwx ) ) allocate ( kisort( npwx ) )
@ -584,10 +605,13 @@ subroutine write_export (pp_file,kunit,uspp_spsi, ascii, single_file, raw)
if(.not.single_file) & if(.not.single_file) &
call iotk_link(50,"Main_grid","mgrid",create=.true.,binary=.not.ascii,raw=raw) call iotk_link(50,"Main_grid","mgrid",create=.true.,binary=.not.ascii,raw=raw)
call iotk_write_begin(50,"Main_grid",attr=attr) call iotk_write_begin(50,"Main_grid",attr=attr)
call iotk_write_dat(50,"g",itmp(1:3,1:ngm_g),fmt="(3i5)") call iotk_write_attr(attr,"units", "crystal",first=.true.)
call iotk_write_dat(50,"gg",gg(1:ngm_g)) call iotk_write_dat(50,"g",itmp_g(1:3,1:ngm_g),fmt="(3i5)", attr=attr)
call iotk_write_attr(attr,"units", "tpiba^2",first=.true.)
call iotk_write_dat(50,"gg",rtmp_gg(1:ngm_g),attr=attr)
call iotk_write_end(50,"Main_grid") call iotk_write_end(50,"Main_grid")
end if end if
deallocate( rtmp_gg )
! for each k point build and write the global G+k indexes array ! for each k point build and write the global G+k indexes array
allocate( igwk( npwx_g,nkstot ) ) allocate( igwk( npwx_g,nkstot ) )
@ -625,13 +649,13 @@ subroutine write_export (pp_file,kunit,uspp_spsi, ascii, single_file, raw)
call iotk_link(50,"Kpoint"//iotk_index(ik),"grid"//iotk_index(ik),create=.true.,binary=.not.ascii,raw=raw) call iotk_link(50,"Kpoint"//iotk_index(ik),"grid"//iotk_index(ik),create=.true.,binary=.not.ascii,raw=raw)
call iotk_write_begin(50,"Kpoint"//iotk_index(ik),attr) call iotk_write_begin(50,"Kpoint"//iotk_index(ik),attr)
call iotk_write_dat (50,"index",igwk(1:ngk_g(ik),ik)) call iotk_write_dat (50,"index",igwk(1:ngk_g(ik),ik))
call iotk_write_dat (50,"grid",itmp(1:3,igwk(1:ngk_g(ik),ik)),fmt="(3i5)") call iotk_write_dat (50,"grid",itmp_g(1:3,igwk(1:ngk_g(ik),ik)),fmt="(3i5)")
call iotk_write_end (50,"Kpoint"//iotk_index(ik)) call iotk_write_end (50,"Kpoint"//iotk_index(ik))
end if end if
end do end do
if(ionode) call iotk_write_end(50,"Wfc_grids") if(ionode) call iotk_write_end(50,"Wfc_grids")
deallocate( itmp ) deallocate( itmp_g )
#ifdef __PARA #ifdef __PARA