[3] | 1 | MODULE lib_mpp |
---|
[13] | 2 | !!====================================================================== |
---|
| 3 | !! *** MODULE lib_mpp *** |
---|
[1344] | 4 | !! Ocean numerics: massively parallel processing library |
---|
[13] | 5 | !!===================================================================== |
---|
[1344] | 6 | !! History : OPA ! 1994 (M. Guyon, J. Escobar, M. Imbard) Original code |
---|
| 7 | !! 7.0 ! 1997 (A.M. Treguier) SHMEM additions |
---|
| 8 | !! 8.0 ! 1998 (M. Imbard, J. Escobar, L. Colombet ) SHMEM and MPI |
---|
| 9 | !! ! 1998 (J.M. Molines) Open boundary conditions |
---|
| 10 | !! NEMO 1.0 ! 2003 (J.-M. Molines, G. Madec) F90, free form |
---|
| 11 | !! ! 2003 (J.M. Molines) add mpp_ini_north(_3d,_2d) |
---|
| 12 | !! - ! 2004 (R. Bourdalle Badie) isend option in mpi |
---|
| 13 | !! ! 2004 (J.M. Molines) minloc, maxloc |
---|
| 14 | !! - ! 2005 (G. Madec, S. Masson) npolj=5,6 F-point & ice cases |
---|
| 15 | !! - ! 2005 (R. Redler) Replacement of MPI_COMM_WORLD except for MPI_Abort |
---|
| 16 | !! - ! 2005 (R. Benshila, G. Madec) add extra halo case |
---|
| 17 | !! - ! 2008 (R. Benshila) add mpp_ini_ice |
---|
| 18 | !! 3.2 ! 2009 (R. Benshila) SHMEM suppression, north fold in lbc_nfd |
---|
[1345] | 19 | !! 3.2 ! 2009 (O. Marti) add mpp_ini_znl |
---|
[13] | 20 | !!---------------------------------------------------------------------- |
---|
[1344] | 21 | #if defined key_mpp_mpi |
---|
[13] | 22 | !!---------------------------------------------------------------------- |
---|
[1344] | 23 | !! 'key_mpp_mpi' MPI massively parallel processing library |
---|
| 24 | !!---------------------------------------------------------------------- |
---|
| 25 | !! mynode : indentify the processor unit |
---|
| 26 | !! mpp_lnk : interface (defined in lbclnk) for message passing of 2d or 3d arrays (mpp_lnk_2d, mpp_lnk_3d) |
---|
[473] | 27 | !! mpp_lnk_3d_gather : Message passing manadgement for two 3D arrays |
---|
[1344] | 28 | !! mpp_lnk_e : interface (defined in lbclnk) for message passing of 2d array with extra halo (mpp_lnk_2d_e) |
---|
| 29 | !! mpprecv : |
---|
[1345] | 30 | !! mppsend : SUBROUTINE mpp_ini_znl |
---|
[1344] | 31 | !! mppscatter : |
---|
| 32 | !! mppgather : |
---|
| 33 | !! mpp_min : generic interface for mppmin_int , mppmin_a_int , mppmin_real, mppmin_a_real |
---|
| 34 | !! mpp_max : generic interface for mppmax_int , mppmax_a_int , mppmax_real, mppmax_a_real |
---|
| 35 | !! mpp_sum : generic interface for mppsum_int , mppsum_a_int , mppsum_real, mppsum_a_real |
---|
| 36 | !! mpp_minloc : |
---|
| 37 | !! mpp_maxloc : |
---|
| 38 | !! mppsync : |
---|
| 39 | !! mppstop : |
---|
| 40 | !! mppobc : variant of mpp_lnk for open boundary condition |
---|
| 41 | !! mpp_ini_north : initialisation of north fold |
---|
| 42 | !! mpp_lbc_north : north fold processors gathering |
---|
| 43 | !! mpp_lbc_north_e : variant of mpp_lbc_north for extra outer halo |
---|
[13] | 44 | !!---------------------------------------------------------------------- |
---|
| 45 | !! History : |
---|
| 46 | !! ! 94 (M. Guyon, J. Escobar, M. Imbard) Original code |
---|
| 47 | !! ! 97 (A.M. Treguier) SHMEM additions |
---|
| 48 | !! ! 98 (M. Imbard, J. Escobar, L. Colombet ) SHMEM and MPI |
---|
| 49 | !! 9.0 ! 03 (J.-M. Molines, G. Madec) F90, free form |
---|
[233] | 50 | !! ! 04 (R. Bourdalle Badie) isend option in mpi |
---|
| 51 | !! ! 05 (G. Madec, S. Masson) npolj=5,6 F-point & ice cases |
---|
[532] | 52 | !! ! 05 (R. Redler) Replacement of MPI_COMM_WORLD except for MPI_Abort |
---|
[1344] | 53 | !! ! 09 (R. Benshila) SHMEM suppression, north fold in lbc_nfd |
---|
[13] | 54 | !!---------------------------------------------------------------------- |
---|
[247] | 55 | !! OPA 9.0 , LOCEAN-IPSL (2005) |
---|
[888] | 56 | !! $Id$ |
---|
[247] | 57 | !! This software is governed by the CeCILL licence see modipsl/doc/NEMO_CeCILL.txt |
---|
[13] | 58 | !!--------------------------------------------------------------------- |
---|
| 59 | !! * Modules used |
---|
[473] | 60 | USE dom_oce ! ocean space and time domain |
---|
| 61 | USE in_out_manager ! I/O manager |
---|
[1344] | 62 | USE lbcnfd ! north fold treatment |
---|
[3] | 63 | |
---|
[13] | 64 | IMPLICIT NONE |
---|
[415] | 65 | PRIVATE |
---|
[1344] | 66 | |
---|
| 67 | PUBLIC mynode, mppstop, mppsync, mpp_comm_free |
---|
| 68 | PUBLIC mpp_ini_north, mpp_lbc_north, mpp_lbc_north_e |
---|
| 69 | PUBLIC mpp_min, mpp_max, mpp_sum, mpp_minloc, mpp_maxloc |
---|
| 70 | PUBLIC mpp_lnk_3d, mpp_lnk_3d_gather, mpp_lnk_2d, mpp_lnk_2d_e |
---|
| 71 | PUBLIC mpprecv, mppsend, mppscatter, mppgather |
---|
[1528] | 72 | PUBLIC mppobc, mpp_ini_ice, mpp_ini_znl |
---|
[1344] | 73 | PUBLIC mppsize, mpprank |
---|
[415] | 74 | |
---|
[13] | 75 | !! * Interfaces |
---|
| 76 | !! define generic interface for these routine as they are called sometimes |
---|
[1344] | 77 | !! with scalar arguments instead of array arguments, which causes problems |
---|
| 78 | !! for the compilation on AIX system as well as NEC and SGI. Ok on COMPACQ |
---|
[13] | 79 | INTERFACE mpp_min |
---|
| 80 | MODULE PROCEDURE mppmin_a_int, mppmin_int, mppmin_a_real, mppmin_real |
---|
| 81 | END INTERFACE |
---|
| 82 | INTERFACE mpp_max |
---|
[681] | 83 | MODULE PROCEDURE mppmax_a_int, mppmax_int, mppmax_a_real, mppmax_real |
---|
[13] | 84 | END INTERFACE |
---|
| 85 | INTERFACE mpp_sum |
---|
| 86 | MODULE PROCEDURE mppsum_a_int, mppsum_int, mppsum_a_real, mppsum_real |
---|
| 87 | END INTERFACE |
---|
| 88 | INTERFACE mpp_lbc_north |
---|
| 89 | MODULE PROCEDURE mpp_lbc_north_3d, mpp_lbc_north_2d |
---|
| 90 | END INTERFACE |
---|
[1344] | 91 | INTERFACE mpp_minloc |
---|
| 92 | MODULE PROCEDURE mpp_minloc2d ,mpp_minloc3d |
---|
| 93 | END INTERFACE |
---|
| 94 | INTERFACE mpp_maxloc |
---|
| 95 | MODULE PROCEDURE mpp_maxloc2d ,mpp_maxloc3d |
---|
| 96 | END INTERFACE |
---|
[3] | 97 | |
---|
[181] | 98 | |
---|
[51] | 99 | !! ========================= !! |
---|
| 100 | !! MPI variable definition !! |
---|
| 101 | !! ========================= !! |
---|
[1629] | 102 | !$AGRIF_DO_NOT_TREAT |
---|
[2208] | 103 | INCLUDE 'mpif.h' |
---|
[1629] | 104 | !$AGRIF_END_DO_NOT_TREAT |
---|
[1344] | 105 | |
---|
| 106 | LOGICAL, PUBLIC, PARAMETER :: lk_mpp = .TRUE. !: mpp flag |
---|
[3] | 107 | |
---|
[1344] | 108 | INTEGER, PARAMETER :: nprocmax = 2**10 ! maximun dimension (required to be a power of 2) |
---|
| 109 | |
---|
| 110 | INTEGER :: mppsize ! number of process |
---|
| 111 | INTEGER :: mpprank ! process number [ 0 - size-1 ] |
---|
[1793] | 112 | !$AGRIF_DO_NOT_TREAT |
---|
| 113 | INTEGER, PUBLIC :: mpi_comm_opa ! opa local communicator |
---|
| 114 | !$AGRIF_END_DO_NOT_TREAT |
---|
[3] | 115 | |
---|
[869] | 116 | ! variables used in case of sea-ice |
---|
[1344] | 117 | INTEGER, PUBLIC :: ncomm_ice !: communicator made by the processors with sea-ice |
---|
[1345] | 118 | INTEGER :: ngrp_ice ! group ID for the ice processors (for rheology) |
---|
| 119 | INTEGER :: ndim_rank_ice ! number of 'ice' processors |
---|
| 120 | INTEGER :: n_ice_root ! number (in the comm_ice) of proc 0 in the ice comm |
---|
[1344] | 121 | INTEGER, DIMENSION(:), ALLOCATABLE :: nrank_ice ! dimension ndim_rank_ice |
---|
[1345] | 122 | |
---|
| 123 | ! variables used for zonal integration |
---|
| 124 | INTEGER, PUBLIC :: ncomm_znl !: communicator made by the processors on the same zonal average |
---|
| 125 | LOGICAL, PUBLIC :: l_znl_root ! True on the 'left'most processor on the same row |
---|
| 126 | INTEGER :: ngrp_znl ! group ID for the znl processors |
---|
| 127 | INTEGER :: ndim_rank_znl ! number of processors on the same zonal average |
---|
| 128 | INTEGER, DIMENSION(:), ALLOCATABLE :: nrank_znl ! dimension ndim_rank_znl, number of the procs into the same znl domain |
---|
[1344] | 129 | |
---|
| 130 | ! North fold condition in mpp_mpi with jpni > 1 |
---|
| 131 | INTEGER :: ngrp_world ! group ID for the world processors |
---|
[1345] | 132 | INTEGER :: ngrp_opa ! group ID for the opa processors |
---|
[1344] | 133 | INTEGER :: ngrp_north ! group ID for the northern processors (to be fold) |
---|
| 134 | INTEGER :: ncomm_north ! communicator made by the processors belonging to ngrp_north |
---|
| 135 | INTEGER :: ndim_rank_north ! number of 'sea' processor in the northern line (can be /= jpni !) |
---|
| 136 | INTEGER :: njmppmax ! value of njmpp for the processors of the northern line |
---|
| 137 | INTEGER :: north_root ! number (in the comm_opa) of proc 0 in the northern comm |
---|
| 138 | INTEGER, DIMENSION(:), ALLOCATABLE :: nrank_north ! dimension ndim_rank_north |
---|
[3] | 139 | |
---|
[1344] | 140 | ! Type of send : standard, buffered, immediate |
---|
[1601] | 141 | CHARACTER(len=1) :: cn_mpi_send = 'S' ! type od mpi send/recieve (S=standard, B=bsend, I=isend) |
---|
| 142 | LOGICAL :: l_isend = .FALSE. ! isend use indicator (T if cn_mpi_send='I') |
---|
[1344] | 143 | INTEGER :: nn_buffer = 0 ! size of the buffer in case of mpi_bsend |
---|
| 144 | |
---|
| 145 | REAL(wp), ALLOCATABLE, DIMENSION(:) :: tampon ! buffer in case of bsend |
---|
[3] | 146 | |
---|
[1344] | 147 | ! message passing arrays |
---|
| 148 | REAL(wp), DIMENSION(jpi,jprecj,jpk,2,2) :: t4ns, t4sn ! 2 x 3d for north-south & south-north |
---|
| 149 | REAL(wp), DIMENSION(jpj,jpreci,jpk,2,2) :: t4ew, t4we ! 2 x 3d for east-west & west-east |
---|
| 150 | REAL(wp), DIMENSION(jpi,jprecj,jpk,2,2) :: t4p1, t4p2 ! 2 x 3d for north fold |
---|
| 151 | REAL(wp), DIMENSION(jpi,jprecj,jpk,2) :: t3ns, t3sn ! 3d for north-south & south-north |
---|
| 152 | REAL(wp), DIMENSION(jpj,jpreci,jpk,2) :: t3ew, t3we ! 3d for east-west & west-east |
---|
| 153 | REAL(wp), DIMENSION(jpi,jprecj,jpk,2) :: t3p1, t3p2 ! 3d for north fold |
---|
| 154 | REAL(wp), DIMENSION(jpi,jprecj,2) :: t2ns, t2sn ! 2d for north-south & south-north |
---|
| 155 | REAL(wp), DIMENSION(jpj,jpreci,2) :: t2ew, t2we ! 2d for east-west & west-east |
---|
| 156 | REAL(wp), DIMENSION(jpi,jprecj,2) :: t2p1, t2p2 ! 2d for north fold |
---|
| 157 | REAL(wp), DIMENSION(1-jpr2di:jpi+jpr2di,jprecj+jpr2dj,2) :: tr2ns, tr2sn ! 2d for north-south & south-north + extra outer halo |
---|
| 158 | REAL(wp), DIMENSION(1-jpr2dj:jpj+jpr2dj,jpreci+jpr2di,2) :: tr2ew, tr2we ! 2d for east-west & west-east + extra outer halo |
---|
[51] | 159 | !!---------------------------------------------------------------------- |
---|
[1344] | 160 | !! NEMO/OPA 3.2 , LOCEAN-IPSL (2009) |
---|
[888] | 161 | !! $Id$ |
---|
[2208] | 162 | !! Software governed by the CeCILL licence (NEMOGCM/License_CeCILL.txt) |
---|
[1344] | 163 | !!---------------------------------------------------------------------- |
---|
[3] | 164 | |
---|
| 165 | CONTAINS |
---|
| 166 | |
---|
[1579] | 167 | FUNCTION mynode(ldtxt, localComm) |
---|
[51] | 168 | !!---------------------------------------------------------------------- |
---|
| 169 | !! *** routine mynode *** |
---|
| 170 | !! |
---|
| 171 | !! ** Purpose : Find processor unit |
---|
| 172 | !! |
---|
| 173 | !!---------------------------------------------------------------------- |
---|
[1579] | 174 | CHARACTER(len=*),DIMENSION(:), INTENT( out) :: ldtxt |
---|
| 175 | INTEGER, OPTIONAL , INTENT(in ) :: localComm |
---|
[532] | 176 | INTEGER :: mynode, ierr, code |
---|
| 177 | LOGICAL :: mpi_was_called |
---|
[1579] | 178 | |
---|
[1601] | 179 | NAMELIST/nammpp/ cn_mpi_send, nn_buffer |
---|
[51] | 180 | !!---------------------------------------------------------------------- |
---|
[1344] | 181 | ! |
---|
[1579] | 182 | WRITE(ldtxt(1),*) |
---|
| 183 | WRITE(ldtxt(2),*) 'mynode : mpi initialisation' |
---|
| 184 | WRITE(ldtxt(3),*) '~~~~~~ ' |
---|
[1344] | 185 | ! |
---|
| 186 | REWIND( numnam ) ! Namelist namrun : parameters of the run |
---|
[1601] | 187 | READ ( numnam, nammpp ) |
---|
[1344] | 188 | ! ! control print |
---|
[1601] | 189 | WRITE(ldtxt(4),*) ' Namelist nammpp' |
---|
| 190 | WRITE(ldtxt(5),*) ' mpi send type cn_mpi_send = ', cn_mpi_send |
---|
| 191 | WRITE(ldtxt(6),*) ' size in bytes of exported buffer nn_buffer = ', nn_buffer |
---|
[300] | 192 | |
---|
[1793] | 193 | CALL mpi_initialized ( mpi_was_called, code ) |
---|
| 194 | IF( code /= MPI_SUCCESS ) THEN |
---|
| 195 | WRITE(*, cform_err) |
---|
| 196 | WRITE(*, *) 'lib_mpp: Error in routine mpi_initialized' |
---|
| 197 | CALL mpi_abort( mpi_comm_world, code, ierr ) |
---|
| 198 | ENDIF |
---|
[415] | 199 | |
---|
[1793] | 200 | IF( mpi_was_called ) THEN |
---|
| 201 | ! |
---|
| 202 | SELECT CASE ( cn_mpi_send ) |
---|
| 203 | CASE ( 'S' ) ! Standard mpi send (blocking) |
---|
| 204 | WRITE(ldtxt(7),*) ' Standard blocking mpi send (send)' |
---|
| 205 | CASE ( 'B' ) ! Buffer mpi send (blocking) |
---|
| 206 | WRITE(ldtxt(7),*) ' Buffer blocking mpi send (bsend)' |
---|
| 207 | CALL mpi_init_opa( ierr ) |
---|
| 208 | CASE ( 'I' ) ! Immediate mpi send (non-blocking send) |
---|
| 209 | WRITE(ldtxt(7),*) ' Immediate non-blocking send (isend)' |
---|
| 210 | l_isend = .TRUE. |
---|
| 211 | CASE DEFAULT |
---|
| 212 | WRITE(ldtxt(7),cform_err) |
---|
| 213 | WRITE(ldtxt(8),*) ' bad value for cn_mpi_send = ', cn_mpi_send |
---|
[1579] | 214 | nstop = nstop + 1 |
---|
[1793] | 215 | END SELECT |
---|
| 216 | ELSE IF ( PRESENT(localComm) .and. .not. mpi_was_called ) THEN |
---|
| 217 | WRITE(ldtxt(7),*) ' lib_mpp: You cannot provide a local communicator ' |
---|
| 218 | WRITE(ldtxt(8),*) ' without calling MPI_Init before ! ' |
---|
| 219 | nstop = nstop + 1 |
---|
[532] | 220 | ELSE |
---|
[1601] | 221 | SELECT CASE ( cn_mpi_send ) |
---|
[524] | 222 | CASE ( 'S' ) ! Standard mpi send (blocking) |
---|
[1579] | 223 | WRITE(ldtxt(7),*) ' Standard blocking mpi send (send)' |
---|
[1793] | 224 | CALL mpi_init( ierr ) |
---|
[524] | 225 | CASE ( 'B' ) ! Buffer mpi send (blocking) |
---|
[1579] | 226 | WRITE(ldtxt(7),*) ' Buffer blocking mpi send (bsend)' |
---|
[1793] | 227 | CALL mpi_init_opa( ierr ) |
---|
[524] | 228 | CASE ( 'I' ) ! Immediate mpi send (non-blocking send) |
---|
[1579] | 229 | WRITE(ldtxt(7),*) ' Immediate non-blocking send (isend)' |
---|
[524] | 230 | l_isend = .TRUE. |
---|
[1793] | 231 | CALL mpi_init( ierr ) |
---|
[524] | 232 | CASE DEFAULT |
---|
[1579] | 233 | WRITE(ldtxt(7),cform_err) |
---|
[1601] | 234 | WRITE(ldtxt(8),*) ' bad value for cn_mpi_send = ', cn_mpi_send |
---|
[524] | 235 | nstop = nstop + 1 |
---|
| 236 | END SELECT |
---|
[1793] | 237 | ! |
---|
[415] | 238 | ENDIF |
---|
[570] | 239 | |
---|
[1793] | 240 | IF( PRESENT(localComm) ) THEN |
---|
| 241 | IF( Agrif_Root() ) THEN |
---|
| 242 | mpi_comm_opa = localComm |
---|
| 243 | ENDIF |
---|
| 244 | ELSE |
---|
| 245 | CALL mpi_comm_dup( mpi_comm_world, mpi_comm_opa, code) |
---|
| 246 | IF( code /= MPI_SUCCESS ) THEN |
---|
| 247 | WRITE(*, cform_err) |
---|
| 248 | WRITE(*, *) ' lib_mpp: Error in routine mpi_comm_dup' |
---|
| 249 | CALL mpi_abort( mpi_comm_world, code, ierr ) |
---|
| 250 | ENDIF |
---|
| 251 | ENDIF |
---|
| 252 | |
---|
[1344] | 253 | CALL mpi_comm_rank( mpi_comm_opa, mpprank, ierr ) |
---|
| 254 | CALL mpi_comm_size( mpi_comm_opa, mppsize, ierr ) |
---|
[629] | 255 | mynode = mpprank |
---|
[1344] | 256 | ! |
---|
[51] | 257 | END FUNCTION mynode |
---|
[3] | 258 | |
---|
| 259 | |
---|
[888] | 260 | SUBROUTINE mpp_lnk_3d( ptab, cd_type, psgn, cd_mpp, pval ) |
---|
[51] | 261 | !!---------------------------------------------------------------------- |
---|
| 262 | !! *** routine mpp_lnk_3d *** |
---|
| 263 | !! |
---|
| 264 | !! ** Purpose : Message passing manadgement |
---|
| 265 | !! |
---|
| 266 | !! ** Method : Use mppsend and mpprecv function for passing mask |
---|
| 267 | !! between processors following neighboring subdomains. |
---|
| 268 | !! domain parameters |
---|
| 269 | !! nlci : first dimension of the local subdomain |
---|
| 270 | !! nlcj : second dimension of the local subdomain |
---|
| 271 | !! nbondi : mark for "east-west local boundary" |
---|
| 272 | !! nbondj : mark for "north-south local boundary" |
---|
| 273 | !! noea : number for local neighboring processors |
---|
| 274 | !! nowe : number for local neighboring processors |
---|
| 275 | !! noso : number for local neighboring processors |
---|
| 276 | !! nono : number for local neighboring processors |
---|
| 277 | !! |
---|
| 278 | !! ** Action : ptab with update value at its periphery |
---|
| 279 | !! |
---|
| 280 | !!---------------------------------------------------------------------- |
---|
[1344] | 281 | REAL(wp), DIMENSION(jpi,jpj,jpk), INTENT(inout) :: ptab ! 3D array on which the boundary condition is applied |
---|
| 282 | CHARACTER(len=1) , INTENT(in ) :: cd_type ! define the nature of ptab array grid-points |
---|
| 283 | ! ! = T , U , V , F , W points |
---|
| 284 | REAL(wp) , INTENT(in ) :: psgn ! =-1 the sign change across the north fold boundary |
---|
| 285 | ! ! = 1. , the sign is kept |
---|
| 286 | CHARACTER(len=3), OPTIONAL , INTENT(in ) :: cd_mpp ! fill the overlap area only |
---|
| 287 | REAL(wp) , OPTIONAL , INTENT(in ) :: pval ! background value (used at closed boundaries) |
---|
| 288 | !! |
---|
[1718] | 289 | INTEGER :: ji, jj, jk, jl ! dummy loop indices |
---|
[1344] | 290 | INTEGER :: imigr, iihom, ijhom ! temporary integers |
---|
| 291 | INTEGER :: ml_req1, ml_req2, ml_err ! for key_mpi_isend |
---|
[888] | 292 | REAL(wp) :: zland |
---|
[1344] | 293 | INTEGER, DIMENSION(MPI_STATUS_SIZE) :: ml_stat ! for key_mpi_isend |
---|
[51] | 294 | !!---------------------------------------------------------------------- |
---|
[3] | 295 | |
---|
[1344] | 296 | IF( PRESENT( pval ) ) THEN ; zland = pval ! set land value |
---|
| 297 | ELSE ; zland = 0.e0 ! zero by default |
---|
| 298 | ENDIF |
---|
| 299 | |
---|
[51] | 300 | ! 1. standard boundary treatment |
---|
| 301 | ! ------------------------------ |
---|
[1718] | 302 | IF( PRESENT( cd_mpp ) ) THEN ! only fill added line/raw with existing values |
---|
[1344] | 303 | ! |
---|
[1718] | 304 | ! WARNING ptab is defined only between nld and nle |
---|
| 305 | DO jk = 1, jpk |
---|
| 306 | DO jj = nlcj+1, jpj ! added line(s) (inner only) |
---|
| 307 | ptab(nldi :nlei , jj ,jk) = ptab(nldi:nlei, nlej,jk) |
---|
| 308 | ptab(1 :nldi-1, jj ,jk) = ptab(nldi , nlej,jk) |
---|
| 309 | ptab(nlei+1:nlci , jj ,jk) = ptab( nlei, nlej,jk) |
---|
| 310 | END DO |
---|
| 311 | DO ji = nlci+1, jpi ! added column(s) (full) |
---|
| 312 | ptab(ji ,nldj :nlej ,jk) = ptab( nlei,nldj:nlej,jk) |
---|
| 313 | ptab(ji ,1 :nldj-1,jk) = ptab( nlei,nldj ,jk) |
---|
| 314 | ptab(ji ,nlej+1:jpj ,jk) = ptab( nlei, nlej,jk) |
---|
| 315 | END DO |
---|
[610] | 316 | END DO |
---|
[1344] | 317 | ! |
---|
| 318 | ELSE ! standard close or cyclic treatment |
---|
| 319 | ! |
---|
| 320 | ! ! East-West boundaries |
---|
| 321 | ! !* Cyclic east-west |
---|
| 322 | IF( nbondi == 2 .AND. (nperio == 1 .OR. nperio == 4 .OR. nperio == 6) ) THEN |
---|
[473] | 323 | ptab( 1 ,:,:) = ptab(jpim1,:,:) |
---|
| 324 | ptab(jpi,:,:) = ptab( 2 ,:,:) |
---|
[1344] | 325 | ELSE !* closed |
---|
| 326 | IF( .NOT. cd_type == 'F' ) ptab( 1 :jpreci,:,:) = zland ! south except F-point |
---|
| 327 | ptab(nlci-jpreci+1:jpi ,:,:) = zland ! north |
---|
[473] | 328 | ENDIF |
---|
[1344] | 329 | ! ! North-South boundaries (always closed) |
---|
| 330 | IF( .NOT. cd_type == 'F' ) ptab(:, 1 :jprecj,:) = zland ! south except F-point |
---|
| 331 | ptab(:,nlcj-jprecj+1:jpj ,:) = zland ! north |
---|
| 332 | ! |
---|
[51] | 333 | ENDIF |
---|
[3] | 334 | |
---|
[51] | 335 | ! 2. East and west directions exchange |
---|
| 336 | ! ------------------------------------ |
---|
[1344] | 337 | ! we play with the neigbours AND the row number because of the periodicity |
---|
| 338 | ! |
---|
| 339 | SELECT CASE ( nbondi ) ! Read Dirichlet lateral conditions |
---|
| 340 | CASE ( -1, 0, 1 ) ! all exept 2 (i.e. close case) |
---|
[51] | 341 | iihom = nlci-nreci |
---|
| 342 | DO jl = 1, jpreci |
---|
| 343 | t3ew(:,jl,:,1) = ptab(jpreci+jl,:,:) |
---|
| 344 | t3we(:,jl,:,1) = ptab(iihom +jl,:,:) |
---|
| 345 | END DO |
---|
[1345] | 346 | END SELECT |
---|
[1344] | 347 | ! |
---|
| 348 | ! ! Migrations |
---|
[51] | 349 | imigr = jpreci * jpj * jpk |
---|
[1344] | 350 | ! |
---|
[51] | 351 | SELECT CASE ( nbondi ) |
---|
| 352 | CASE ( -1 ) |
---|
[181] | 353 | CALL mppsend( 2, t3we(1,1,1,1), imigr, noea, ml_req1 ) |
---|
[51] | 354 | CALL mpprecv( 1, t3ew(1,1,1,2), imigr ) |
---|
[300] | 355 | IF(l_isend) CALL mpi_wait(ml_req1, ml_stat, ml_err) |
---|
[51] | 356 | CASE ( 0 ) |
---|
[181] | 357 | CALL mppsend( 1, t3ew(1,1,1,1), imigr, nowe, ml_req1 ) |
---|
| 358 | CALL mppsend( 2, t3we(1,1,1,1), imigr, noea, ml_req2 ) |
---|
[51] | 359 | CALL mpprecv( 1, t3ew(1,1,1,2), imigr ) |
---|
| 360 | CALL mpprecv( 2, t3we(1,1,1,2), imigr ) |
---|
[300] | 361 | IF(l_isend) CALL mpi_wait(ml_req1, ml_stat, ml_err) |
---|
| 362 | IF(l_isend) CALL mpi_wait(ml_req2, ml_stat, ml_err) |
---|
[51] | 363 | CASE ( 1 ) |
---|
[181] | 364 | CALL mppsend( 1, t3ew(1,1,1,1), imigr, nowe, ml_req1 ) |
---|
[51] | 365 | CALL mpprecv( 2, t3we(1,1,1,2), imigr ) |
---|
[300] | 366 | IF(l_isend) CALL mpi_wait(ml_req1, ml_stat, ml_err) |
---|
[51] | 367 | END SELECT |
---|
[1344] | 368 | ! |
---|
| 369 | ! ! Write Dirichlet lateral conditions |
---|
[51] | 370 | iihom = nlci-jpreci |
---|
[1344] | 371 | ! |
---|
[51] | 372 | SELECT CASE ( nbondi ) |
---|
| 373 | CASE ( -1 ) |
---|
| 374 | DO jl = 1, jpreci |
---|
| 375 | ptab(iihom+jl,:,:) = t3ew(:,jl,:,2) |
---|
| 376 | END DO |
---|
| 377 | CASE ( 0 ) |
---|
| 378 | DO jl = 1, jpreci |
---|
| 379 | ptab(jl ,:,:) = t3we(:,jl,:,2) |
---|
| 380 | ptab(iihom+jl,:,:) = t3ew(:,jl,:,2) |
---|
| 381 | END DO |
---|
| 382 | CASE ( 1 ) |
---|
| 383 | DO jl = 1, jpreci |
---|
| 384 | ptab(jl ,:,:) = t3we(:,jl,:,2) |
---|
| 385 | END DO |
---|
| 386 | END SELECT |
---|
[3] | 387 | |
---|
| 388 | |
---|
[51] | 389 | ! 3. North and south directions |
---|
| 390 | ! ----------------------------- |
---|
[1344] | 391 | ! always closed : we play only with the neigbours |
---|
| 392 | ! |
---|
| 393 | IF( nbondj /= 2 ) THEN ! Read Dirichlet lateral conditions |
---|
[51] | 394 | ijhom = nlcj-nrecj |
---|
| 395 | DO jl = 1, jprecj |
---|
| 396 | t3sn(:,jl,:,1) = ptab(:,ijhom +jl,:) |
---|
| 397 | t3ns(:,jl,:,1) = ptab(:,jprecj+jl,:) |
---|
| 398 | END DO |
---|
| 399 | ENDIF |
---|
[1344] | 400 | ! |
---|
| 401 | ! ! Migrations |
---|
[51] | 402 | imigr = jprecj * jpi * jpk |
---|
[1344] | 403 | ! |
---|
[51] | 404 | SELECT CASE ( nbondj ) |
---|
| 405 | CASE ( -1 ) |
---|
[181] | 406 | CALL mppsend( 4, t3sn(1,1,1,1), imigr, nono, ml_req1 ) |
---|
[51] | 407 | CALL mpprecv( 3, t3ns(1,1,1,2), imigr ) |
---|
[300] | 408 | IF(l_isend) CALL mpi_wait(ml_req1, ml_stat, ml_err) |
---|
[51] | 409 | CASE ( 0 ) |
---|
[181] | 410 | CALL mppsend( 3, t3ns(1,1,1,1), imigr, noso, ml_req1 ) |
---|
| 411 | CALL mppsend( 4, t3sn(1,1,1,1), imigr, nono, ml_req2 ) |
---|
[51] | 412 | CALL mpprecv( 3, t3ns(1,1,1,2), imigr ) |
---|
| 413 | CALL mpprecv( 4, t3sn(1,1,1,2), imigr ) |
---|
[300] | 414 | IF(l_isend) CALL mpi_wait(ml_req1, ml_stat, ml_err) |
---|
| 415 | IF(l_isend) CALL mpi_wait(ml_req2, ml_stat, ml_err) |
---|
[51] | 416 | CASE ( 1 ) |
---|
[181] | 417 | CALL mppsend( 3, t3ns(1,1,1,1), imigr, noso, ml_req1 ) |
---|
[51] | 418 | CALL mpprecv( 4, t3sn(1,1,1,2), imigr ) |
---|
[300] | 419 | IF(l_isend) CALL mpi_wait(ml_req1, ml_stat, ml_err) |
---|
[51] | 420 | END SELECT |
---|
[1344] | 421 | ! |
---|
| 422 | ! ! Write Dirichlet lateral conditions |
---|
[51] | 423 | ijhom = nlcj-jprecj |
---|
[1344] | 424 | ! |
---|
[51] | 425 | SELECT CASE ( nbondj ) |
---|
| 426 | CASE ( -1 ) |
---|
| 427 | DO jl = 1, jprecj |
---|
| 428 | ptab(:,ijhom+jl,:) = t3ns(:,jl,:,2) |
---|
| 429 | END DO |
---|
| 430 | CASE ( 0 ) |
---|
| 431 | DO jl = 1, jprecj |
---|
| 432 | ptab(:,jl ,:) = t3sn(:,jl,:,2) |
---|
| 433 | ptab(:,ijhom+jl,:) = t3ns(:,jl,:,2) |
---|
| 434 | END DO |
---|
| 435 | CASE ( 1 ) |
---|
| 436 | DO jl = 1, jprecj |
---|
| 437 | ptab(:,jl,:) = t3sn(:,jl,:,2) |
---|
| 438 | END DO |
---|
| 439 | END SELECT |
---|
[3] | 440 | |
---|
| 441 | |
---|
[51] | 442 | ! 4. north fold treatment |
---|
| 443 | ! ----------------------- |
---|
[1344] | 444 | ! |
---|
| 445 | IF( npolj /= 0 .AND. .NOT. PRESENT(cd_mpp) ) THEN |
---|
| 446 | ! |
---|
| 447 | SELECT CASE ( jpni ) |
---|
| 448 | CASE ( 1 ) ; CALL lbc_nfd ( ptab, cd_type, psgn ) ! only 1 northern proc, no mpp |
---|
| 449 | CASE DEFAULT ; CALL mpp_lbc_north( ptab, cd_type, psgn ) ! for all northern procs. |
---|
| 450 | END SELECT |
---|
| 451 | ! |
---|
[473] | 452 | ENDIF |
---|
[1344] | 453 | ! |
---|
[51] | 454 | END SUBROUTINE mpp_lnk_3d |
---|
[3] | 455 | |
---|
| 456 | |
---|
[888] | 457 | SUBROUTINE mpp_lnk_2d( pt2d, cd_type, psgn, cd_mpp, pval ) |
---|
[51] | 458 | !!---------------------------------------------------------------------- |
---|
| 459 | !! *** routine mpp_lnk_2d *** |
---|
| 460 | !! |
---|
| 461 | !! ** Purpose : Message passing manadgement for 2d array |
---|
| 462 | !! |
---|
| 463 | !! ** Method : Use mppsend and mpprecv function for passing mask |
---|
| 464 | !! between processors following neighboring subdomains. |
---|
| 465 | !! domain parameters |
---|
| 466 | !! nlci : first dimension of the local subdomain |
---|
| 467 | !! nlcj : second dimension of the local subdomain |
---|
| 468 | !! nbondi : mark for "east-west local boundary" |
---|
| 469 | !! nbondj : mark for "north-south local boundary" |
---|
| 470 | !! noea : number for local neighboring processors |
---|
| 471 | !! nowe : number for local neighboring processors |
---|
| 472 | !! noso : number for local neighboring processors |
---|
| 473 | !! nono : number for local neighboring processors |
---|
| 474 | !! |
---|
| 475 | !!---------------------------------------------------------------------- |
---|
[1344] | 476 | REAL(wp), DIMENSION(jpi,jpj), INTENT(inout) :: pt2d ! 2D array on which the boundary condition is applied |
---|
| 477 | CHARACTER(len=1) , INTENT(in ) :: cd_type ! define the nature of ptab array grid-points |
---|
| 478 | ! ! = T , U , V , F , W and I points |
---|
| 479 | REAL(wp) , INTENT(in ) :: psgn ! =-1 the sign change across the north fold boundary |
---|
| 480 | ! ! = 1. , the sign is kept |
---|
| 481 | CHARACTER(len=3), OPTIONAL , INTENT(in ) :: cd_mpp ! fill the overlap area only |
---|
| 482 | REAL(wp) , OPTIONAL , INTENT(in ) :: pval ! background value (used at closed boundaries) |
---|
| 483 | !! |
---|
| 484 | INTEGER :: ji, jj, jl ! dummy loop indices |
---|
| 485 | INTEGER :: imigr, iihom, ijhom ! temporary integers |
---|
| 486 | INTEGER :: ml_req1, ml_req2, ml_err ! for key_mpi_isend |
---|
[888] | 487 | REAL(wp) :: zland |
---|
[1344] | 488 | INTEGER, DIMENSION(MPI_STATUS_SIZE) :: ml_stat ! for key_mpi_isend |
---|
[51] | 489 | !!---------------------------------------------------------------------- |
---|
[3] | 490 | |
---|
[1344] | 491 | IF( PRESENT( pval ) ) THEN ; zland = pval ! set land value |
---|
| 492 | ELSE ; zland = 0.e0 ! zero by default |
---|
[888] | 493 | ENDIF |
---|
| 494 | |
---|
[51] | 495 | ! 1. standard boundary treatment |
---|
| 496 | ! ------------------------------ |
---|
[1344] | 497 | ! |
---|
[1718] | 498 | IF( PRESENT( cd_mpp ) ) THEN ! only fill added line/raw with existing values |
---|
[1344] | 499 | ! |
---|
[1718] | 500 | ! WARNING pt2d is defined only between nld and nle |
---|
| 501 | DO jj = nlcj+1, jpj ! added line(s) (inner only) |
---|
| 502 | pt2d(nldi :nlei , jj ) = pt2d(nldi:nlei, nlej) |
---|
| 503 | pt2d(1 :nldi-1, jj ) = pt2d(nldi , nlej) |
---|
| 504 | pt2d(nlei+1:nlci , jj ) = pt2d( nlei, nlej) |
---|
[610] | 505 | END DO |
---|
[1718] | 506 | DO ji = nlci+1, jpi ! added column(s) (full) |
---|
| 507 | pt2d(ji ,nldj :nlej ) = pt2d( nlei,nldj:nlej) |
---|
| 508 | pt2d(ji ,1 :nldj-1) = pt2d( nlei,nldj ) |
---|
| 509 | pt2d(ji ,nlej+1:jpj ) = pt2d( nlei, nlej) |
---|
[1344] | 510 | END DO |
---|
| 511 | ! |
---|
| 512 | ELSE ! standard close or cyclic treatment |
---|
| 513 | ! |
---|
| 514 | ! ! East-West boundaries |
---|
| 515 | IF( nbondi == 2 .AND. & ! Cyclic east-west |
---|
[473] | 516 | & (nperio == 1 .OR. nperio == 4 .OR. nperio == 6) ) THEN |
---|
[1344] | 517 | pt2d( 1 ,:) = pt2d(jpim1,:) ! west |
---|
| 518 | pt2d(jpi,:) = pt2d( 2 ,:) ! east |
---|
| 519 | ELSE ! closed |
---|
| 520 | IF( .NOT. cd_type == 'F' ) pt2d( 1 :jpreci,:) = zland ! south except F-point |
---|
| 521 | pt2d(nlci-jpreci+1:jpi ,:) = zland ! north |
---|
[473] | 522 | ENDIF |
---|
[1344] | 523 | ! ! North-South boundaries (always closed) |
---|
| 524 | IF( .NOT. cd_type == 'F' ) pt2d(:, 1 :jprecj) = zland !south except F-point |
---|
| 525 | pt2d(:,nlcj-jprecj+1:jpj ) = zland ! north |
---|
| 526 | ! |
---|
[51] | 527 | ENDIF |
---|
[3] | 528 | |
---|
[1344] | 529 | ! 2. East and west directions exchange |
---|
| 530 | ! ------------------------------------ |
---|
| 531 | ! we play with the neigbours AND the row number because of the periodicity |
---|
| 532 | ! |
---|
| 533 | SELECT CASE ( nbondi ) ! Read Dirichlet lateral conditions |
---|
| 534 | CASE ( -1, 0, 1 ) ! all exept 2 (i.e. close case) |
---|
[51] | 535 | iihom = nlci-nreci |
---|
| 536 | DO jl = 1, jpreci |
---|
| 537 | t2ew(:,jl,1) = pt2d(jpreci+jl,:) |
---|
| 538 | t2we(:,jl,1) = pt2d(iihom +jl,:) |
---|
| 539 | END DO |
---|
| 540 | END SELECT |
---|
[1344] | 541 | ! |
---|
| 542 | ! ! Migrations |
---|
[51] | 543 | imigr = jpreci * jpj |
---|
[1344] | 544 | ! |
---|
[51] | 545 | SELECT CASE ( nbondi ) |
---|
| 546 | CASE ( -1 ) |
---|
[181] | 547 | CALL mppsend( 2, t2we(1,1,1), imigr, noea, ml_req1 ) |
---|
[51] | 548 | CALL mpprecv( 1, t2ew(1,1,2), imigr ) |
---|
[300] | 549 | IF(l_isend) CALL mpi_wait(ml_req1,ml_stat,ml_err) |
---|
[51] | 550 | CASE ( 0 ) |
---|
[181] | 551 | CALL mppsend( 1, t2ew(1,1,1), imigr, nowe, ml_req1 ) |
---|
| 552 | CALL mppsend( 2, t2we(1,1,1), imigr, noea, ml_req2 ) |
---|
[51] | 553 | CALL mpprecv( 1, t2ew(1,1,2), imigr ) |
---|
| 554 | CALL mpprecv( 2, t2we(1,1,2), imigr ) |
---|
[300] | 555 | IF(l_isend) CALL mpi_wait(ml_req1,ml_stat,ml_err) |
---|
| 556 | IF(l_isend) CALL mpi_wait(ml_req2,ml_stat,ml_err) |
---|
[51] | 557 | CASE ( 1 ) |
---|
[181] | 558 | CALL mppsend( 1, t2ew(1,1,1), imigr, nowe, ml_req1 ) |
---|
[51] | 559 | CALL mpprecv( 2, t2we(1,1,2), imigr ) |
---|
[300] | 560 | IF(l_isend) CALL mpi_wait(ml_req1,ml_stat,ml_err) |
---|
[51] | 561 | END SELECT |
---|
[1344] | 562 | ! |
---|
| 563 | ! ! Write Dirichlet lateral conditions |
---|
[51] | 564 | iihom = nlci - jpreci |
---|
[1344] | 565 | ! |
---|
[51] | 566 | SELECT CASE ( nbondi ) |
---|
| 567 | CASE ( -1 ) |
---|
| 568 | DO jl = 1, jpreci |
---|
| 569 | pt2d(iihom+jl,:) = t2ew(:,jl,2) |
---|
| 570 | END DO |
---|
| 571 | CASE ( 0 ) |
---|
| 572 | DO jl = 1, jpreci |
---|
| 573 | pt2d(jl ,:) = t2we(:,jl,2) |
---|
| 574 | pt2d(iihom+jl,:) = t2ew(:,jl,2) |
---|
| 575 | END DO |
---|
| 576 | CASE ( 1 ) |
---|
| 577 | DO jl = 1, jpreci |
---|
| 578 | pt2d(jl ,:) = t2we(:,jl,2) |
---|
| 579 | END DO |
---|
| 580 | END SELECT |
---|
[3] | 581 | |
---|
| 582 | |
---|
[51] | 583 | ! 3. North and south directions |
---|
| 584 | ! ----------------------------- |
---|
[1344] | 585 | ! always closed : we play only with the neigbours |
---|
| 586 | ! |
---|
| 587 | IF( nbondj /= 2 ) THEN ! Read Dirichlet lateral conditions |
---|
[51] | 588 | ijhom = nlcj-nrecj |
---|
| 589 | DO jl = 1, jprecj |
---|
| 590 | t2sn(:,jl,1) = pt2d(:,ijhom +jl) |
---|
| 591 | t2ns(:,jl,1) = pt2d(:,jprecj+jl) |
---|
| 592 | END DO |
---|
| 593 | ENDIF |
---|
[1344] | 594 | ! |
---|
| 595 | ! ! Migrations |
---|
[51] | 596 | imigr = jprecj * jpi |
---|
[1344] | 597 | ! |
---|
[51] | 598 | SELECT CASE ( nbondj ) |
---|
| 599 | CASE ( -1 ) |
---|
[181] | 600 | CALL mppsend( 4, t2sn(1,1,1), imigr, nono, ml_req1 ) |
---|
[51] | 601 | CALL mpprecv( 3, t2ns(1,1,2), imigr ) |
---|
[300] | 602 | IF(l_isend) CALL mpi_wait(ml_req1,ml_stat,ml_err) |
---|
[51] | 603 | CASE ( 0 ) |
---|
[181] | 604 | CALL mppsend( 3, t2ns(1,1,1), imigr, noso, ml_req1 ) |
---|
| 605 | CALL mppsend( 4, t2sn(1,1,1), imigr, nono, ml_req2 ) |
---|
[51] | 606 | CALL mpprecv( 3, t2ns(1,1,2), imigr ) |
---|
| 607 | CALL mpprecv( 4, t2sn(1,1,2), imigr ) |
---|
[300] | 608 | IF(l_isend) CALL mpi_wait(ml_req1,ml_stat,ml_err) |
---|
| 609 | IF(l_isend) CALL mpi_wait(ml_req2,ml_stat,ml_err) |
---|
[51] | 610 | CASE ( 1 ) |
---|
[181] | 611 | CALL mppsend( 3, t2ns(1,1,1), imigr, noso, ml_req1 ) |
---|
[51] | 612 | CALL mpprecv( 4, t2sn(1,1,2), imigr ) |
---|
[300] | 613 | IF(l_isend) CALL mpi_wait(ml_req1,ml_stat,ml_err) |
---|
[51] | 614 | END SELECT |
---|
[1344] | 615 | ! |
---|
| 616 | ! ! Write Dirichlet lateral conditions |
---|
[51] | 617 | ijhom = nlcj - jprecj |
---|
[1344] | 618 | ! |
---|
[51] | 619 | SELECT CASE ( nbondj ) |
---|
| 620 | CASE ( -1 ) |
---|
| 621 | DO jl = 1, jprecj |
---|
| 622 | pt2d(:,ijhom+jl) = t2ns(:,jl,2) |
---|
| 623 | END DO |
---|
| 624 | CASE ( 0 ) |
---|
| 625 | DO jl = 1, jprecj |
---|
| 626 | pt2d(:,jl ) = t2sn(:,jl,2) |
---|
| 627 | pt2d(:,ijhom+jl) = t2ns(:,jl,2) |
---|
| 628 | END DO |
---|
| 629 | CASE ( 1 ) |
---|
| 630 | DO jl = 1, jprecj |
---|
| 631 | pt2d(:,jl ) = t2sn(:,jl,2) |
---|
| 632 | END DO |
---|
[1344] | 633 | END SELECT |
---|
[3] | 634 | |
---|
[1344] | 635 | |
---|
[51] | 636 | ! 4. north fold treatment |
---|
| 637 | ! ----------------------- |
---|
[1344] | 638 | ! |
---|
| 639 | IF( npolj /= 0 .AND. .NOT. PRESENT(cd_mpp) ) THEN |
---|
| 640 | ! |
---|
| 641 | SELECT CASE ( jpni ) |
---|
| 642 | CASE ( 1 ) ; CALL lbc_nfd ( pt2d, cd_type, psgn ) ! only 1 northern proc, no mpp |
---|
| 643 | CASE DEFAULT ; CALL mpp_lbc_north( pt2d, cd_type, psgn ) ! for all northern procs. |
---|
| 644 | END SELECT |
---|
| 645 | ! |
---|
[473] | 646 | ENDIF |
---|
[1344] | 647 | ! |
---|
[51] | 648 | END SUBROUTINE mpp_lnk_2d |
---|
[3] | 649 | |
---|
| 650 | |
---|
[473] | 651 | SUBROUTINE mpp_lnk_3d_gather( ptab1, cd_type1, ptab2, cd_type2, psgn ) |
---|
| 652 | !!---------------------------------------------------------------------- |
---|
| 653 | !! *** routine mpp_lnk_3d_gather *** |
---|
| 654 | !! |
---|
| 655 | !! ** Purpose : Message passing manadgement for two 3D arrays |
---|
| 656 | !! |
---|
| 657 | !! ** Method : Use mppsend and mpprecv function for passing mask |
---|
| 658 | !! between processors following neighboring subdomains. |
---|
| 659 | !! domain parameters |
---|
| 660 | !! nlci : first dimension of the local subdomain |
---|
| 661 | !! nlcj : second dimension of the local subdomain |
---|
| 662 | !! nbondi : mark for "east-west local boundary" |
---|
| 663 | !! nbondj : mark for "north-south local boundary" |
---|
| 664 | !! noea : number for local neighboring processors |
---|
| 665 | !! nowe : number for local neighboring processors |
---|
| 666 | !! noso : number for local neighboring processors |
---|
| 667 | !! nono : number for local neighboring processors |
---|
| 668 | !! |
---|
| 669 | !! ** Action : ptab1 and ptab2 with update value at its periphery |
---|
| 670 | !! |
---|
| 671 | !!---------------------------------------------------------------------- |
---|
[1344] | 672 | REAL(wp), DIMENSION(jpi,jpj,jpk), INTENT(inout) :: ptab1 ! first and second 3D array on which |
---|
| 673 | REAL(wp), DIMENSION(jpi,jpj,jpk), INTENT(inout) :: ptab2 ! the boundary condition is applied |
---|
| 674 | CHARACTER(len=1) , INTENT(in ) :: cd_type1 ! nature of ptab1 and ptab2 arrays |
---|
| 675 | CHARACTER(len=1) , INTENT(in ) :: cd_type2 ! i.e. grid-points = T , U , V , F or W points |
---|
| 676 | REAL(wp) , INTENT(in ) :: psgn ! =-1 the sign change across the north fold boundary |
---|
| 677 | !! ! = 1. , the sign is kept |
---|
| 678 | INTEGER :: jl ! dummy loop indices |
---|
| 679 | INTEGER :: imigr, iihom, ijhom ! temporary integers |
---|
| 680 | INTEGER :: ml_req1, ml_req2, ml_err ! for key_mpi_isend |
---|
| 681 | INTEGER, DIMENSION(MPI_STATUS_SIZE) :: ml_stat ! for key_mpi_isend |
---|
[473] | 682 | !!---------------------------------------------------------------------- |
---|
| 683 | |
---|
| 684 | ! 1. standard boundary treatment |
---|
| 685 | ! ------------------------------ |
---|
[1344] | 686 | ! ! East-West boundaries |
---|
| 687 | ! !* Cyclic east-west |
---|
| 688 | IF( nbondi == 2 .AND. (nperio == 1 .OR. nperio == 4 .OR. nperio == 6) ) THEN |
---|
[473] | 689 | ptab1( 1 ,:,:) = ptab1(jpim1,:,:) |
---|
| 690 | ptab1(jpi,:,:) = ptab1( 2 ,:,:) |
---|
| 691 | ptab2( 1 ,:,:) = ptab2(jpim1,:,:) |
---|
| 692 | ptab2(jpi,:,:) = ptab2( 2 ,:,:) |
---|
[1344] | 693 | ELSE !* closed |
---|
| 694 | IF( .NOT. cd_type1 == 'F' ) ptab1( 1 :jpreci,:,:) = 0.e0 ! south except at F-point |
---|
| 695 | IF( .NOT. cd_type2 == 'F' ) ptab2( 1 :jpreci,:,:) = 0.e0 |
---|
| 696 | ptab1(nlci-jpreci+1:jpi ,:,:) = 0.e0 ! north |
---|
| 697 | ptab2(nlci-jpreci+1:jpi ,:,:) = 0.e0 |
---|
[473] | 698 | ENDIF |
---|
| 699 | |
---|
[1344] | 700 | |
---|
| 701 | ! ! North-South boundaries |
---|
| 702 | IF( .NOT. cd_type1 == 'F' ) ptab1(:, 1 :jprecj,:) = 0.e0 ! south except at F-point |
---|
| 703 | IF( .NOT. cd_type2 == 'F' ) ptab2(:, 1 :jprecj,:) = 0.e0 |
---|
| 704 | ptab1(:,nlcj-jprecj+1:jpj ,:) = 0.e0 ! north |
---|
| 705 | ptab2(:,nlcj-jprecj+1:jpj ,:) = 0.e0 |
---|
[473] | 706 | |
---|
| 707 | |
---|
| 708 | ! 2. East and west directions exchange |
---|
| 709 | ! ------------------------------------ |
---|
[1344] | 710 | ! we play with the neigbours AND the row number because of the periodicity |
---|
| 711 | ! |
---|
| 712 | SELECT CASE ( nbondi ) ! Read Dirichlet lateral conditions |
---|
| 713 | CASE ( -1, 0, 1 ) ! all exept 2 (i.e. close case) |
---|
[473] | 714 | iihom = nlci-nreci |
---|
| 715 | DO jl = 1, jpreci |
---|
| 716 | t4ew(:,jl,:,1,1) = ptab1(jpreci+jl,:,:) |
---|
| 717 | t4we(:,jl,:,1,1) = ptab1(iihom +jl,:,:) |
---|
| 718 | t4ew(:,jl,:,2,1) = ptab2(jpreci+jl,:,:) |
---|
| 719 | t4we(:,jl,:,2,1) = ptab2(iihom +jl,:,:) |
---|
| 720 | END DO |
---|
| 721 | END SELECT |
---|
[1344] | 722 | ! |
---|
| 723 | ! ! Migrations |
---|
[473] | 724 | imigr = jpreci * jpj * jpk *2 |
---|
[1344] | 725 | ! |
---|
[473] | 726 | SELECT CASE ( nbondi ) |
---|
| 727 | CASE ( -1 ) |
---|
| 728 | CALL mppsend( 2, t4we(1,1,1,1,1), imigr, noea, ml_req1 ) |
---|
| 729 | CALL mpprecv( 1, t4ew(1,1,1,1,2), imigr ) |
---|
| 730 | IF(l_isend) CALL mpi_wait(ml_req1, ml_stat, ml_err) |
---|
| 731 | CASE ( 0 ) |
---|
| 732 | CALL mppsend( 1, t4ew(1,1,1,1,1), imigr, nowe, ml_req1 ) |
---|
| 733 | CALL mppsend( 2, t4we(1,1,1,1,1), imigr, noea, ml_req2 ) |
---|
| 734 | CALL mpprecv( 1, t4ew(1,1,1,1,2), imigr ) |
---|
| 735 | CALL mpprecv( 2, t4we(1,1,1,1,2), imigr ) |
---|
| 736 | IF(l_isend) CALL mpi_wait(ml_req1, ml_stat, ml_err) |
---|
| 737 | IF(l_isend) CALL mpi_wait(ml_req2, ml_stat, ml_err) |
---|
| 738 | CASE ( 1 ) |
---|
| 739 | CALL mppsend( 1, t4ew(1,1,1,1,1), imigr, nowe, ml_req1 ) |
---|
| 740 | CALL mpprecv( 2, t4we(1,1,1,1,2), imigr ) |
---|
| 741 | IF(l_isend) CALL mpi_wait(ml_req1, ml_stat, ml_err) |
---|
| 742 | END SELECT |
---|
[1344] | 743 | ! |
---|
| 744 | ! ! Write Dirichlet lateral conditions |
---|
| 745 | iihom = nlci - jpreci |
---|
| 746 | ! |
---|
[473] | 747 | SELECT CASE ( nbondi ) |
---|
| 748 | CASE ( -1 ) |
---|
| 749 | DO jl = 1, jpreci |
---|
| 750 | ptab1(iihom+jl,:,:) = t4ew(:,jl,:,1,2) |
---|
| 751 | ptab2(iihom+jl,:,:) = t4ew(:,jl,:,2,2) |
---|
| 752 | END DO |
---|
| 753 | CASE ( 0 ) |
---|
| 754 | DO jl = 1, jpreci |
---|
| 755 | ptab1(jl ,:,:) = t4we(:,jl,:,1,2) |
---|
| 756 | ptab1(iihom+jl,:,:) = t4ew(:,jl,:,1,2) |
---|
| 757 | ptab2(jl ,:,:) = t4we(:,jl,:,2,2) |
---|
| 758 | ptab2(iihom+jl,:,:) = t4ew(:,jl,:,2,2) |
---|
| 759 | END DO |
---|
| 760 | CASE ( 1 ) |
---|
| 761 | DO jl = 1, jpreci |
---|
| 762 | ptab1(jl ,:,:) = t4we(:,jl,:,1,2) |
---|
| 763 | ptab2(jl ,:,:) = t4we(:,jl,:,2,2) |
---|
| 764 | END DO |
---|
| 765 | END SELECT |
---|
| 766 | |
---|
| 767 | |
---|
| 768 | ! 3. North and south directions |
---|
| 769 | ! ----------------------------- |
---|
[1344] | 770 | ! always closed : we play only with the neigbours |
---|
| 771 | ! |
---|
| 772 | IF( nbondj /= 2 ) THEN ! Read Dirichlet lateral conditions |
---|
| 773 | ijhom = nlcj - nrecj |
---|
[473] | 774 | DO jl = 1, jprecj |
---|
| 775 | t4sn(:,jl,:,1,1) = ptab1(:,ijhom +jl,:) |
---|
| 776 | t4ns(:,jl,:,1,1) = ptab1(:,jprecj+jl,:) |
---|
| 777 | t4sn(:,jl,:,2,1) = ptab2(:,ijhom +jl,:) |
---|
| 778 | t4ns(:,jl,:,2,1) = ptab2(:,jprecj+jl,:) |
---|
| 779 | END DO |
---|
| 780 | ENDIF |
---|
[1344] | 781 | ! |
---|
| 782 | ! ! Migrations |
---|
[473] | 783 | imigr = jprecj * jpi * jpk * 2 |
---|
[1344] | 784 | ! |
---|
[473] | 785 | SELECT CASE ( nbondj ) |
---|
| 786 | CASE ( -1 ) |
---|
| 787 | CALL mppsend( 4, t4sn(1,1,1,1,1), imigr, nono, ml_req1 ) |
---|
| 788 | CALL mpprecv( 3, t4ns(1,1,1,1,2), imigr ) |
---|
| 789 | IF(l_isend) CALL mpi_wait(ml_req1, ml_stat, ml_err) |
---|
| 790 | CASE ( 0 ) |
---|
| 791 | CALL mppsend( 3, t4ns(1,1,1,1,1), imigr, noso, ml_req1 ) |
---|
| 792 | CALL mppsend( 4, t4sn(1,1,1,1,1), imigr, nono, ml_req2 ) |
---|
| 793 | CALL mpprecv( 3, t4ns(1,1,1,1,2), imigr ) |
---|
| 794 | CALL mpprecv( 4, t4sn(1,1,1,1,2), imigr ) |
---|
| 795 | IF(l_isend) CALL mpi_wait(ml_req1, ml_stat, ml_err) |
---|
| 796 | IF(l_isend) CALL mpi_wait(ml_req2, ml_stat, ml_err) |
---|
| 797 | CASE ( 1 ) |
---|
| 798 | CALL mppsend( 3, t4ns(1,1,1,1,1), imigr, noso, ml_req1 ) |
---|
| 799 | CALL mpprecv( 4, t4sn(1,1,1,1,2), imigr ) |
---|
| 800 | IF(l_isend) CALL mpi_wait(ml_req1, ml_stat, ml_err) |
---|
| 801 | END SELECT |
---|
[1344] | 802 | ! |
---|
| 803 | ! ! Write Dirichlet lateral conditions |
---|
| 804 | ijhom = nlcj - jprecj |
---|
| 805 | ! |
---|
[473] | 806 | SELECT CASE ( nbondj ) |
---|
| 807 | CASE ( -1 ) |
---|
| 808 | DO jl = 1, jprecj |
---|
| 809 | ptab1(:,ijhom+jl,:) = t4ns(:,jl,:,1,2) |
---|
| 810 | ptab2(:,ijhom+jl,:) = t4ns(:,jl,:,2,2) |
---|
| 811 | END DO |
---|
| 812 | CASE ( 0 ) |
---|
| 813 | DO jl = 1, jprecj |
---|
| 814 | ptab1(:,jl ,:) = t4sn(:,jl,:,1,2) |
---|
| 815 | ptab1(:,ijhom+jl,:) = t4ns(:,jl,:,1,2) |
---|
| 816 | ptab2(:,jl ,:) = t4sn(:,jl,:,2,2) |
---|
| 817 | ptab2(:,ijhom+jl,:) = t4ns(:,jl,:,2,2) |
---|
| 818 | END DO |
---|
| 819 | CASE ( 1 ) |
---|
| 820 | DO jl = 1, jprecj |
---|
| 821 | ptab1(:,jl,:) = t4sn(:,jl,:,1,2) |
---|
| 822 | ptab2(:,jl,:) = t4sn(:,jl,:,2,2) |
---|
| 823 | END DO |
---|
| 824 | END SELECT |
---|
| 825 | |
---|
| 826 | |
---|
| 827 | ! 4. north fold treatment |
---|
| 828 | ! ----------------------- |
---|
[1344] | 829 | IF( npolj /= 0 ) THEN |
---|
| 830 | ! |
---|
| 831 | SELECT CASE ( jpni ) |
---|
| 832 | CASE ( 1 ) |
---|
| 833 | CALL lbc_nfd ( ptab1, cd_type1, psgn ) ! only for northern procs. |
---|
| 834 | CALL lbc_nfd ( ptab2, cd_type2, psgn ) |
---|
| 835 | CASE DEFAULT |
---|
| 836 | CALL mpp_lbc_north( ptab1, cd_type1, psgn ) ! for all northern procs. |
---|
| 837 | CALL mpp_lbc_north (ptab2, cd_type2, psgn) |
---|
| 838 | END SELECT |
---|
| 839 | ! |
---|
| 840 | ENDIF |
---|
| 841 | ! |
---|
[473] | 842 | END SUBROUTINE mpp_lnk_3d_gather |
---|
| 843 | |
---|
| 844 | |
---|
[311] | 845 | SUBROUTINE mpp_lnk_2d_e( pt2d, cd_type, psgn ) |
---|
| 846 | !!---------------------------------------------------------------------- |
---|
| 847 | !! *** routine mpp_lnk_2d_e *** |
---|
| 848 | !! |
---|
| 849 | !! ** Purpose : Message passing manadgement for 2d array (with halo) |
---|
| 850 | !! |
---|
| 851 | !! ** Method : Use mppsend and mpprecv function for passing mask |
---|
| 852 | !! between processors following neighboring subdomains. |
---|
| 853 | !! domain parameters |
---|
| 854 | !! nlci : first dimension of the local subdomain |
---|
| 855 | !! nlcj : second dimension of the local subdomain |
---|
| 856 | !! jpr2di : number of rows for extra outer halo |
---|
| 857 | !! jpr2dj : number of columns for extra outer halo |
---|
| 858 | !! nbondi : mark for "east-west local boundary" |
---|
| 859 | !! nbondj : mark for "north-south local boundary" |
---|
| 860 | !! noea : number for local neighboring processors |
---|
| 861 | !! nowe : number for local neighboring processors |
---|
| 862 | !! noso : number for local neighboring processors |
---|
| 863 | !! nono : number for local neighboring processors |
---|
| 864 | !! |
---|
| 865 | !!---------------------------------------------------------------------- |
---|
[1344] | 866 | REAL(wp), DIMENSION(1-jpr2di:jpi+jpr2di,1-jpr2dj:jpj+jpr2dj), INTENT(inout) :: pt2d ! 2D array with extra halo |
---|
| 867 | CHARACTER(len=1) , INTENT(in ) :: cd_type ! nature of ptab array grid-points |
---|
| 868 | ! ! = T , U , V , F , W and I points |
---|
| 869 | REAL(wp) , INTENT(in ) :: psgn ! =-1 the sign change across the |
---|
| 870 | !! ! north boundary, = 1. otherwise |
---|
| 871 | INTEGER :: jl ! dummy loop indices |
---|
| 872 | INTEGER :: imigr, iihom, ijhom ! temporary integers |
---|
| 873 | INTEGER :: ipreci, iprecj ! temporary integers |
---|
| 874 | INTEGER :: ml_req1, ml_req2, ml_err ! for key_mpi_isend |
---|
| 875 | INTEGER, DIMENSION(MPI_STATUS_SIZE) :: ml_stat ! for key_mpi_isend |
---|
| 876 | !!---------------------------------------------------------------------- |
---|
[311] | 877 | |
---|
[1344] | 878 | ipreci = jpreci + jpr2di ! take into account outer extra 2D overlap area |
---|
[311] | 879 | iprecj = jprecj + jpr2dj |
---|
| 880 | |
---|
| 881 | |
---|
| 882 | ! 1. standard boundary treatment |
---|
| 883 | ! ------------------------------ |
---|
[1344] | 884 | ! Order matters Here !!!! |
---|
| 885 | ! |
---|
| 886 | ! !* North-South boundaries (always colsed) |
---|
| 887 | IF( .NOT. cd_type == 'F' ) pt2d(:, 1-jpr2dj : jprecj ) = 0.e0 ! south except at F-point |
---|
| 888 | pt2d(:,nlcj-jprecj+1:jpj+jpr2dj) = 0.e0 ! north |
---|
| 889 | |
---|
| 890 | ! ! East-West boundaries |
---|
| 891 | ! !* Cyclic east-west |
---|
| 892 | IF( nbondi == 2 .AND. (nperio == 1 .OR. nperio == 4 .OR. nperio == 6) ) THEN |
---|
| 893 | pt2d(1-jpr2di: 1 ,:) = pt2d(jpim1-jpr2di: jpim1 ,:) ! east |
---|
| 894 | pt2d( jpi :jpi+jpr2di,:) = pt2d( 2 :2+jpr2di,:) ! west |
---|
| 895 | ! |
---|
| 896 | ELSE !* closed |
---|
| 897 | IF( .NOT. cd_type == 'F' ) pt2d( 1-jpr2di :jpreci ,:) = 0.e0 ! south except at F-point |
---|
| 898 | pt2d(nlci-jpreci+1:jpi+jpr2di,:) = 0.e0 ! north |
---|
| 899 | ENDIF |
---|
| 900 | ! |
---|
[311] | 901 | |
---|
[1344] | 902 | ! north fold treatment |
---|
| 903 | ! ----------------------- |
---|
| 904 | IF( npolj /= 0 ) THEN |
---|
| 905 | ! |
---|
| 906 | SELECT CASE ( jpni ) |
---|
| 907 | CASE ( 1 ) ; CALL lbc_nfd ( pt2d(1:jpi,1:jpj+jpr2dj), cd_type, psgn, pr2dj=jpr2dj ) |
---|
| 908 | CASE DEFAULT ; CALL mpp_lbc_north_e( pt2d , cd_type, psgn ) |
---|
| 909 | END SELECT |
---|
| 910 | ! |
---|
[311] | 911 | ENDIF |
---|
| 912 | |
---|
[1344] | 913 | ! 2. East and west directions exchange |
---|
| 914 | ! ------------------------------------ |
---|
| 915 | ! we play with the neigbours AND the row number because of the periodicity |
---|
| 916 | ! |
---|
| 917 | SELECT CASE ( nbondi ) ! Read Dirichlet lateral conditions |
---|
| 918 | CASE ( -1, 0, 1 ) ! all exept 2 (i.e. close case) |
---|
[311] | 919 | iihom = nlci-nreci-jpr2di |
---|
| 920 | DO jl = 1, ipreci |
---|
| 921 | tr2ew(:,jl,1) = pt2d(jpreci+jl,:) |
---|
| 922 | tr2we(:,jl,1) = pt2d(iihom +jl,:) |
---|
| 923 | END DO |
---|
| 924 | END SELECT |
---|
[1344] | 925 | ! |
---|
| 926 | ! ! Migrations |
---|
[311] | 927 | imigr = ipreci * ( jpj + 2*jpr2dj) |
---|
[1344] | 928 | ! |
---|
[311] | 929 | SELECT CASE ( nbondi ) |
---|
| 930 | CASE ( -1 ) |
---|
| 931 | CALL mppsend( 2, tr2we(1-jpr2dj,1,1), imigr, noea, ml_req1 ) |
---|
| 932 | CALL mpprecv( 1, tr2ew(1-jpr2dj,1,2), imigr ) |
---|
| 933 | IF(l_isend) CALL mpi_wait(ml_req1,ml_stat,ml_err) |
---|
| 934 | CASE ( 0 ) |
---|
| 935 | CALL mppsend( 1, tr2ew(1-jpr2dj,1,1), imigr, nowe, ml_req1 ) |
---|
| 936 | CALL mppsend( 2, tr2we(1-jpr2dj,1,1), imigr, noea, ml_req2 ) |
---|
| 937 | CALL mpprecv( 1, tr2ew(1-jpr2dj,1,2), imigr ) |
---|
| 938 | CALL mpprecv( 2, tr2we(1-jpr2dj,1,2), imigr ) |
---|
| 939 | IF(l_isend) CALL mpi_wait(ml_req1,ml_stat,ml_err) |
---|
| 940 | IF(l_isend) CALL mpi_wait(ml_req2,ml_stat,ml_err) |
---|
| 941 | CASE ( 1 ) |
---|
| 942 | CALL mppsend( 1, tr2ew(1-jpr2dj,1,1), imigr, nowe, ml_req1 ) |
---|
| 943 | CALL mpprecv( 2, tr2we(1-jpr2dj,1,2), imigr ) |
---|
| 944 | IF(l_isend) CALL mpi_wait(ml_req1,ml_stat,ml_err) |
---|
| 945 | END SELECT |
---|
[1344] | 946 | ! |
---|
| 947 | ! ! Write Dirichlet lateral conditions |
---|
[311] | 948 | iihom = nlci - jpreci |
---|
[1344] | 949 | ! |
---|
[311] | 950 | SELECT CASE ( nbondi ) |
---|
| 951 | CASE ( -1 ) |
---|
| 952 | DO jl = 1, ipreci |
---|
| 953 | pt2d(iihom+jl,:) = tr2ew(:,jl,2) |
---|
| 954 | END DO |
---|
| 955 | CASE ( 0 ) |
---|
| 956 | DO jl = 1, ipreci |
---|
| 957 | pt2d(jl-jpr2di,:) = tr2we(:,jl,2) |
---|
| 958 | pt2d( iihom+jl,:) = tr2ew(:,jl,2) |
---|
| 959 | END DO |
---|
| 960 | CASE ( 1 ) |
---|
| 961 | DO jl = 1, ipreci |
---|
| 962 | pt2d(jl-jpr2di,:) = tr2we(:,jl,2) |
---|
| 963 | END DO |
---|
| 964 | END SELECT |
---|
| 965 | |
---|
| 966 | |
---|
| 967 | ! 3. North and south directions |
---|
| 968 | ! ----------------------------- |
---|
[1344] | 969 | ! always closed : we play only with the neigbours |
---|
| 970 | ! |
---|
| 971 | IF( nbondj /= 2 ) THEN ! Read Dirichlet lateral conditions |
---|
[311] | 972 | ijhom = nlcj-nrecj-jpr2dj |
---|
| 973 | DO jl = 1, iprecj |
---|
| 974 | tr2sn(:,jl,1) = pt2d(:,ijhom +jl) |
---|
| 975 | tr2ns(:,jl,1) = pt2d(:,jprecj+jl) |
---|
| 976 | END DO |
---|
| 977 | ENDIF |
---|
[1344] | 978 | ! |
---|
| 979 | ! ! Migrations |
---|
[311] | 980 | imigr = iprecj * ( jpi + 2*jpr2di ) |
---|
[1344] | 981 | ! |
---|
[311] | 982 | SELECT CASE ( nbondj ) |
---|
| 983 | CASE ( -1 ) |
---|
| 984 | CALL mppsend( 4, tr2sn(1-jpr2di,1,1), imigr, nono, ml_req1 ) |
---|
| 985 | CALL mpprecv( 3, tr2ns(1-jpr2di,1,2), imigr ) |
---|
| 986 | IF(l_isend) CALL mpi_wait(ml_req1,ml_stat,ml_err) |
---|
| 987 | CASE ( 0 ) |
---|
| 988 | CALL mppsend( 3, tr2ns(1-jpr2di,1,1), imigr, noso, ml_req1 ) |
---|
| 989 | CALL mppsend( 4, tr2sn(1-jpr2di,1,1), imigr, nono, ml_req2 ) |
---|
| 990 | CALL mpprecv( 3, tr2ns(1-jpr2di,1,2), imigr ) |
---|
| 991 | CALL mpprecv( 4, tr2sn(1-jpr2di,1,2), imigr ) |
---|
| 992 | IF(l_isend) CALL mpi_wait(ml_req1,ml_stat,ml_err) |
---|
| 993 | IF(l_isend) CALL mpi_wait(ml_req2,ml_stat,ml_err) |
---|
| 994 | CASE ( 1 ) |
---|
| 995 | CALL mppsend( 3, tr2ns(1-jpr2di,1,1), imigr, noso, ml_req1 ) |
---|
| 996 | CALL mpprecv( 4, tr2sn(1-jpr2di,1,2), imigr ) |
---|
| 997 | IF(l_isend) CALL mpi_wait(ml_req1,ml_stat,ml_err) |
---|
| 998 | END SELECT |
---|
[1344] | 999 | ! |
---|
| 1000 | ! ! Write Dirichlet lateral conditions |
---|
[311] | 1001 | ijhom = nlcj - jprecj |
---|
[1344] | 1002 | ! |
---|
[311] | 1003 | SELECT CASE ( nbondj ) |
---|
| 1004 | CASE ( -1 ) |
---|
| 1005 | DO jl = 1, iprecj |
---|
| 1006 | pt2d(:,ijhom+jl) = tr2ns(:,jl,2) |
---|
| 1007 | END DO |
---|
| 1008 | CASE ( 0 ) |
---|
| 1009 | DO jl = 1, iprecj |
---|
| 1010 | pt2d(:,jl-jpr2dj) = tr2sn(:,jl,2) |
---|
| 1011 | pt2d(:,ijhom+jl ) = tr2ns(:,jl,2) |
---|
| 1012 | END DO |
---|
| 1013 | CASE ( 1 ) |
---|
| 1014 | DO jl = 1, iprecj |
---|
| 1015 | pt2d(:,jl-jpr2dj) = tr2sn(:,jl,2) |
---|
| 1016 | END DO |
---|
[1344] | 1017 | END SELECT |
---|
[311] | 1018 | |
---|
| 1019 | END SUBROUTINE mpp_lnk_2d_e |
---|
| 1020 | |
---|
| 1021 | |
---|
[1344] | 1022 | SUBROUTINE mppsend( ktyp, pmess, kbytes, kdest, md_req ) |
---|
[51] | 1023 | !!---------------------------------------------------------------------- |
---|
| 1024 | !! *** routine mppsend *** |
---|
| 1025 | !! |
---|
| 1026 | !! ** Purpose : Send messag passing array |
---|
| 1027 | !! |
---|
| 1028 | !!---------------------------------------------------------------------- |
---|
[1344] | 1029 | REAL(wp), INTENT(inout) :: pmess(*) ! array of real |
---|
| 1030 | INTEGER , INTENT(in ) :: kbytes ! size of the array pmess |
---|
| 1031 | INTEGER , INTENT(in ) :: kdest ! receive process number |
---|
| 1032 | INTEGER , INTENT(in ) :: ktyp ! tag of the message |
---|
| 1033 | INTEGER , INTENT(in ) :: md_req ! argument for isend |
---|
| 1034 | !! |
---|
| 1035 | INTEGER :: iflag |
---|
[51] | 1036 | !!---------------------------------------------------------------------- |
---|
[1344] | 1037 | ! |
---|
[1601] | 1038 | SELECT CASE ( cn_mpi_send ) |
---|
[300] | 1039 | CASE ( 'S' ) ! Standard mpi send (blocking) |
---|
[1344] | 1040 | CALL mpi_send ( pmess, kbytes, mpi_double_precision, kdest , ktyp, mpi_comm_opa , iflag ) |
---|
[300] | 1041 | CASE ( 'B' ) ! Buffer mpi send (blocking) |
---|
[1344] | 1042 | CALL mpi_bsend( pmess, kbytes, mpi_double_precision, kdest , ktyp, mpi_comm_opa , iflag ) |
---|
[300] | 1043 | CASE ( 'I' ) ! Immediate mpi send (non-blocking send) |
---|
[1344] | 1044 | ! be carefull, one more argument here : the mpi request identifier.. |
---|
| 1045 | CALL mpi_isend( pmess, kbytes, mpi_double_precision, kdest , ktyp, mpi_comm_opa, md_req, iflag ) |
---|
[300] | 1046 | END SELECT |
---|
[1344] | 1047 | ! |
---|
[51] | 1048 | END SUBROUTINE mppsend |
---|
[3] | 1049 | |
---|
| 1050 | |
---|
[51] | 1051 | SUBROUTINE mpprecv( ktyp, pmess, kbytes ) |
---|
| 1052 | !!---------------------------------------------------------------------- |
---|
| 1053 | !! *** routine mpprecv *** |
---|
| 1054 | !! |
---|
| 1055 | !! ** Purpose : Receive messag passing array |
---|
| 1056 | !! |
---|
| 1057 | !!---------------------------------------------------------------------- |
---|
[1344] | 1058 | REAL(wp), INTENT(inout) :: pmess(*) ! array of real |
---|
| 1059 | INTEGER , INTENT(in ) :: kbytes ! suze of the array pmess |
---|
| 1060 | INTEGER , INTENT(in ) :: ktyp ! Tag of the recevied message |
---|
| 1061 | !! |
---|
[51] | 1062 | INTEGER :: istatus(mpi_status_size) |
---|
| 1063 | INTEGER :: iflag |
---|
[1344] | 1064 | !!---------------------------------------------------------------------- |
---|
| 1065 | ! |
---|
| 1066 | CALL mpi_recv( pmess, kbytes, mpi_double_precision, mpi_any_source, ktyp, mpi_comm_opa, istatus, iflag ) |
---|
| 1067 | ! |
---|
[51] | 1068 | END SUBROUTINE mpprecv |
---|
[3] | 1069 | |
---|
| 1070 | |
---|
[51] | 1071 | SUBROUTINE mppgather( ptab, kp, pio ) |
---|
| 1072 | !!---------------------------------------------------------------------- |
---|
| 1073 | !! *** routine mppgather *** |
---|
| 1074 | !! |
---|
| 1075 | !! ** Purpose : Transfert between a local subdomain array and a work |
---|
| 1076 | !! array which is distributed following the vertical level. |
---|
| 1077 | !! |
---|
[1344] | 1078 | !!---------------------------------------------------------------------- |
---|
| 1079 | REAL(wp), DIMENSION(jpi,jpj), INTENT(in ) :: ptab ! subdomain input array |
---|
| 1080 | INTEGER , INTENT(in ) :: kp ! record length |
---|
| 1081 | REAL(wp), DIMENSION(jpi,jpj,jpnij), INTENT( out) :: pio ! subdomain input array |
---|
[51] | 1082 | !! |
---|
[1344] | 1083 | INTEGER :: itaille, ierror ! temporary integer |
---|
[51] | 1084 | !!--------------------------------------------------------------------- |
---|
[1344] | 1085 | ! |
---|
| 1086 | itaille = jpi * jpj |
---|
| 1087 | CALL mpi_gather( ptab, itaille, mpi_double_precision, pio, itaille , & |
---|
[532] | 1088 | & mpi_double_precision, kp , mpi_comm_opa, ierror ) |
---|
[1344] | 1089 | ! |
---|
[51] | 1090 | END SUBROUTINE mppgather |
---|
[3] | 1091 | |
---|
| 1092 | |
---|
[51] | 1093 | SUBROUTINE mppscatter( pio, kp, ptab ) |
---|
| 1094 | !!---------------------------------------------------------------------- |
---|
| 1095 | !! *** routine mppscatter *** |
---|
| 1096 | !! |
---|
| 1097 | !! ** Purpose : Transfert between awork array which is distributed |
---|
| 1098 | !! following the vertical level and the local subdomain array. |
---|
| 1099 | !! |
---|
| 1100 | !!---------------------------------------------------------------------- |
---|
| 1101 | REAL(wp), DIMENSION(jpi,jpj,jpnij) :: pio ! output array |
---|
| 1102 | INTEGER :: kp ! Tag (not used with MPI |
---|
| 1103 | REAL(wp), DIMENSION(jpi,jpj) :: ptab ! subdomain array input |
---|
[1344] | 1104 | !! |
---|
| 1105 | INTEGER :: itaille, ierror ! temporary integer |
---|
[51] | 1106 | !!--------------------------------------------------------------------- |
---|
[1344] | 1107 | ! |
---|
[51] | 1108 | itaille=jpi*jpj |
---|
[1344] | 1109 | ! |
---|
| 1110 | CALL mpi_scatter( pio, itaille, mpi_double_precision, ptab, itaille , & |
---|
| 1111 | & mpi_double_precision, kp , mpi_comm_opa, ierror ) |
---|
| 1112 | ! |
---|
[51] | 1113 | END SUBROUTINE mppscatter |
---|
[3] | 1114 | |
---|
| 1115 | |
---|
[869] | 1116 | SUBROUTINE mppmax_a_int( ktab, kdim, kcom ) |
---|
[681] | 1117 | !!---------------------------------------------------------------------- |
---|
| 1118 | !! *** routine mppmax_a_int *** |
---|
| 1119 | !! |
---|
| 1120 | !! ** Purpose : Find maximum value in an integer layout array |
---|
| 1121 | !! |
---|
| 1122 | !!---------------------------------------------------------------------- |
---|
[1344] | 1123 | INTEGER , INTENT(in ) :: kdim ! size of array |
---|
| 1124 | INTEGER , INTENT(inout), DIMENSION(kdim) :: ktab ! input array |
---|
| 1125 | INTEGER , INTENT(in ), OPTIONAL :: kcom ! |
---|
| 1126 | !! |
---|
| 1127 | INTEGER :: ierror, localcomm ! temporary integer |
---|
[681] | 1128 | INTEGER, DIMENSION(kdim) :: iwork |
---|
[1344] | 1129 | !!---------------------------------------------------------------------- |
---|
| 1130 | ! |
---|
[869] | 1131 | localcomm = mpi_comm_opa |
---|
[1344] | 1132 | IF( PRESENT(kcom) ) localcomm = kcom |
---|
| 1133 | ! |
---|
| 1134 | CALL mpi_allreduce( ktab, iwork, kdim, mpi_integer, mpi_max, localcomm, ierror ) |
---|
| 1135 | ! |
---|
[681] | 1136 | ktab(:) = iwork(:) |
---|
[1344] | 1137 | ! |
---|
[681] | 1138 | END SUBROUTINE mppmax_a_int |
---|
| 1139 | |
---|
| 1140 | |
---|
[869] | 1141 | SUBROUTINE mppmax_int( ktab, kcom ) |
---|
[681] | 1142 | !!---------------------------------------------------------------------- |
---|
| 1143 | !! *** routine mppmax_int *** |
---|
| 1144 | !! |
---|
[1344] | 1145 | !! ** Purpose : Find maximum value in an integer layout array |
---|
[681] | 1146 | !! |
---|
| 1147 | !!---------------------------------------------------------------------- |
---|
[1344] | 1148 | INTEGER, INTENT(inout) :: ktab ! ??? |
---|
| 1149 | INTEGER, INTENT(in ), OPTIONAL :: kcom ! ??? |
---|
| 1150 | !! |
---|
| 1151 | INTEGER :: ierror, iwork, localcomm ! temporary integer |
---|
| 1152 | !!---------------------------------------------------------------------- |
---|
| 1153 | ! |
---|
[869] | 1154 | localcomm = mpi_comm_opa |
---|
[1344] | 1155 | IF( PRESENT(kcom) ) localcomm = kcom |
---|
| 1156 | ! |
---|
| 1157 | CALL mpi_allreduce( ktab, iwork, 1, mpi_integer, mpi_max, localcomm, ierror) |
---|
| 1158 | ! |
---|
[681] | 1159 | ktab = iwork |
---|
[1344] | 1160 | ! |
---|
[681] | 1161 | END SUBROUTINE mppmax_int |
---|
| 1162 | |
---|
| 1163 | |
---|
[869] | 1164 | SUBROUTINE mppmin_a_int( ktab, kdim, kcom ) |
---|
[51] | 1165 | !!---------------------------------------------------------------------- |
---|
| 1166 | !! *** routine mppmin_a_int *** |
---|
| 1167 | !! |
---|
| 1168 | !! ** Purpose : Find minimum value in an integer layout array |
---|
| 1169 | !! |
---|
| 1170 | !!---------------------------------------------------------------------- |
---|
| 1171 | INTEGER , INTENT( in ) :: kdim ! size of array |
---|
| 1172 | INTEGER , INTENT(inout), DIMENSION(kdim) :: ktab ! input array |
---|
[888] | 1173 | INTEGER , INTENT( in ), OPTIONAL :: kcom ! input array |
---|
[1344] | 1174 | !! |
---|
| 1175 | INTEGER :: ierror, localcomm ! temporary integer |
---|
[51] | 1176 | INTEGER, DIMENSION(kdim) :: iwork |
---|
[1344] | 1177 | !!---------------------------------------------------------------------- |
---|
| 1178 | ! |
---|
[869] | 1179 | localcomm = mpi_comm_opa |
---|
[1344] | 1180 | IF( PRESENT(kcom) ) localcomm = kcom |
---|
| 1181 | ! |
---|
| 1182 | CALL mpi_allreduce( ktab, iwork, kdim, mpi_integer, mpi_min, localcomm, ierror ) |
---|
| 1183 | ! |
---|
[51] | 1184 | ktab(:) = iwork(:) |
---|
[1344] | 1185 | ! |
---|
[51] | 1186 | END SUBROUTINE mppmin_a_int |
---|
[3] | 1187 | |
---|
[13] | 1188 | |
---|
[1345] | 1189 | SUBROUTINE mppmin_int( ktab, kcom ) |
---|
[51] | 1190 | !!---------------------------------------------------------------------- |
---|
| 1191 | !! *** routine mppmin_int *** |
---|
| 1192 | !! |
---|
[1344] | 1193 | !! ** Purpose : Find minimum value in an integer layout array |
---|
[51] | 1194 | !! |
---|
| 1195 | !!---------------------------------------------------------------------- |
---|
| 1196 | INTEGER, INTENT(inout) :: ktab ! ??? |
---|
[1345] | 1197 | INTEGER , INTENT( in ), OPTIONAL :: kcom ! input array |
---|
[1344] | 1198 | !! |
---|
[1345] | 1199 | INTEGER :: ierror, iwork, localcomm |
---|
[1344] | 1200 | !!---------------------------------------------------------------------- |
---|
| 1201 | ! |
---|
[1345] | 1202 | localcomm = mpi_comm_opa |
---|
| 1203 | IF( PRESENT(kcom) ) localcomm = kcom |
---|
[1344] | 1204 | ! |
---|
[1345] | 1205 | CALL mpi_allreduce( ktab, iwork, 1, mpi_integer, mpi_min, localcomm, ierror ) |
---|
| 1206 | ! |
---|
[51] | 1207 | ktab = iwork |
---|
[1344] | 1208 | ! |
---|
[51] | 1209 | END SUBROUTINE mppmin_int |
---|
[3] | 1210 | |
---|
[13] | 1211 | |
---|
[51] | 1212 | SUBROUTINE mppsum_a_int( ktab, kdim ) |
---|
| 1213 | !!---------------------------------------------------------------------- |
---|
| 1214 | !! *** routine mppsum_a_int *** |
---|
| 1215 | !! |
---|
[1344] | 1216 | !! ** Purpose : Global integer sum, 1D array case |
---|
[51] | 1217 | !! |
---|
| 1218 | !!---------------------------------------------------------------------- |
---|
[1344] | 1219 | INTEGER, INTENT(in ) :: kdim ! ??? |
---|
[51] | 1220 | INTEGER, INTENT(inout), DIMENSION (kdim) :: ktab ! ??? |
---|
[1344] | 1221 | !! |
---|
[51] | 1222 | INTEGER :: ierror |
---|
| 1223 | INTEGER, DIMENSION (kdim) :: iwork |
---|
[1344] | 1224 | !!---------------------------------------------------------------------- |
---|
| 1225 | ! |
---|
| 1226 | CALL mpi_allreduce( ktab, iwork, kdim, mpi_integer, mpi_sum, mpi_comm_opa, ierror ) |
---|
| 1227 | ! |
---|
[51] | 1228 | ktab(:) = iwork(:) |
---|
[1344] | 1229 | ! |
---|
[51] | 1230 | END SUBROUTINE mppsum_a_int |
---|
[3] | 1231 | |
---|
[13] | 1232 | |
---|
[1344] | 1233 | SUBROUTINE mppsum_int( ktab ) |
---|
| 1234 | !!---------------------------------------------------------------------- |
---|
| 1235 | !! *** routine mppsum_int *** |
---|
| 1236 | !! |
---|
| 1237 | !! ** Purpose : Global integer sum |
---|
| 1238 | !! |
---|
| 1239 | !!---------------------------------------------------------------------- |
---|
| 1240 | INTEGER, INTENT(inout) :: ktab |
---|
| 1241 | !! |
---|
| 1242 | INTEGER :: ierror, iwork |
---|
| 1243 | !!---------------------------------------------------------------------- |
---|
| 1244 | ! |
---|
| 1245 | CALL mpi_allreduce( ktab, iwork, 1, mpi_integer, mpi_sum, mpi_comm_opa, ierror ) |
---|
| 1246 | ! |
---|
| 1247 | ktab = iwork |
---|
| 1248 | ! |
---|
| 1249 | END SUBROUTINE mppsum_int |
---|
[3] | 1250 | |
---|
[13] | 1251 | |
---|
[1344] | 1252 | SUBROUTINE mppmax_a_real( ptab, kdim, kcom ) |
---|
| 1253 | !!---------------------------------------------------------------------- |
---|
| 1254 | !! *** routine mppmax_a_real *** |
---|
| 1255 | !! |
---|
| 1256 | !! ** Purpose : Maximum |
---|
| 1257 | !! |
---|
| 1258 | !!---------------------------------------------------------------------- |
---|
| 1259 | INTEGER , INTENT(in ) :: kdim |
---|
| 1260 | REAL(wp), INTENT(inout), DIMENSION(kdim) :: ptab |
---|
| 1261 | INTEGER , INTENT(in ), OPTIONAL :: kcom |
---|
| 1262 | !! |
---|
| 1263 | INTEGER :: ierror, localcomm |
---|
| 1264 | REAL(wp), DIMENSION(kdim) :: zwork |
---|
| 1265 | !!---------------------------------------------------------------------- |
---|
| 1266 | ! |
---|
| 1267 | localcomm = mpi_comm_opa |
---|
| 1268 | IF( PRESENT(kcom) ) localcomm = kcom |
---|
| 1269 | ! |
---|
| 1270 | CALL mpi_allreduce( ptab, zwork, kdim, mpi_double_precision, mpi_max, localcomm, ierror ) |
---|
| 1271 | ptab(:) = zwork(:) |
---|
| 1272 | ! |
---|
| 1273 | END SUBROUTINE mppmax_a_real |
---|
[3] | 1274 | |
---|
| 1275 | |
---|
[1344] | 1276 | SUBROUTINE mppmax_real( ptab, kcom ) |
---|
| 1277 | !!---------------------------------------------------------------------- |
---|
| 1278 | !! *** routine mppmax_real *** |
---|
| 1279 | !! |
---|
| 1280 | !! ** Purpose : Maximum |
---|
| 1281 | !! |
---|
| 1282 | !!---------------------------------------------------------------------- |
---|
| 1283 | REAL(wp), INTENT(inout) :: ptab ! ??? |
---|
| 1284 | INTEGER , INTENT(in ), OPTIONAL :: kcom ! ??? |
---|
| 1285 | !! |
---|
| 1286 | INTEGER :: ierror, localcomm |
---|
| 1287 | REAL(wp) :: zwork |
---|
| 1288 | !!---------------------------------------------------------------------- |
---|
| 1289 | ! |
---|
| 1290 | localcomm = mpi_comm_opa |
---|
| 1291 | IF( PRESENT(kcom) ) localcomm = kcom |
---|
| 1292 | ! |
---|
| 1293 | CALL mpi_allreduce( ptab, zwork, 1, mpi_double_precision, mpi_max, localcomm, ierror ) |
---|
| 1294 | ptab = zwork |
---|
| 1295 | ! |
---|
| 1296 | END SUBROUTINE mppmax_real |
---|
[13] | 1297 | |
---|
[3] | 1298 | |
---|
[1344] | 1299 | SUBROUTINE mppmin_a_real( ptab, kdim, kcom ) |
---|
| 1300 | !!---------------------------------------------------------------------- |
---|
| 1301 | !! *** routine mppmin_a_real *** |
---|
| 1302 | !! |
---|
| 1303 | !! ** Purpose : Minimum of REAL, array case |
---|
| 1304 | !! |
---|
| 1305 | !!----------------------------------------------------------------------- |
---|
| 1306 | INTEGER , INTENT(in ) :: kdim |
---|
| 1307 | REAL(wp), INTENT(inout), DIMENSION(kdim) :: ptab |
---|
| 1308 | INTEGER , INTENT(in ), OPTIONAL :: kcom |
---|
| 1309 | !! |
---|
| 1310 | INTEGER :: ierror, localcomm |
---|
| 1311 | REAL(wp), DIMENSION(kdim) :: zwork |
---|
| 1312 | !!----------------------------------------------------------------------- |
---|
| 1313 | ! |
---|
| 1314 | localcomm = mpi_comm_opa |
---|
| 1315 | IF( PRESENT(kcom) ) localcomm = kcom |
---|
| 1316 | ! |
---|
| 1317 | CALL mpi_allreduce( ptab, zwork, kdim, mpi_double_precision, mpi_min, localcomm, ierror ) |
---|
| 1318 | ptab(:) = zwork(:) |
---|
| 1319 | ! |
---|
| 1320 | END SUBROUTINE mppmin_a_real |
---|
[3] | 1321 | |
---|
| 1322 | |
---|
[1344] | 1323 | SUBROUTINE mppmin_real( ptab, kcom ) |
---|
| 1324 | !!---------------------------------------------------------------------- |
---|
| 1325 | !! *** routine mppmin_real *** |
---|
| 1326 | !! |
---|
| 1327 | !! ** Purpose : minimum of REAL, scalar case |
---|
| 1328 | !! |
---|
| 1329 | !!----------------------------------------------------------------------- |
---|
| 1330 | REAL(wp), INTENT(inout) :: ptab ! |
---|
| 1331 | INTEGER , INTENT(in ), OPTIONAL :: kcom |
---|
| 1332 | !! |
---|
| 1333 | INTEGER :: ierror |
---|
| 1334 | REAL(wp) :: zwork |
---|
| 1335 | INTEGER :: localcomm |
---|
| 1336 | !!----------------------------------------------------------------------- |
---|
| 1337 | ! |
---|
| 1338 | localcomm = mpi_comm_opa |
---|
| 1339 | IF( PRESENT(kcom) ) localcomm = kcom |
---|
| 1340 | ! |
---|
| 1341 | CALL mpi_allreduce( ptab, zwork, 1, mpi_double_precision, mpi_min, localcomm, ierror ) |
---|
| 1342 | ptab = zwork |
---|
| 1343 | ! |
---|
| 1344 | END SUBROUTINE mppmin_real |
---|
[13] | 1345 | |
---|
[3] | 1346 | |
---|
[1344] | 1347 | SUBROUTINE mppsum_a_real( ptab, kdim, kcom ) |
---|
| 1348 | !!---------------------------------------------------------------------- |
---|
| 1349 | !! *** routine mppsum_a_real *** |
---|
| 1350 | !! |
---|
| 1351 | !! ** Purpose : global sum, REAL ARRAY argument case |
---|
| 1352 | !! |
---|
| 1353 | !!----------------------------------------------------------------------- |
---|
| 1354 | INTEGER , INTENT( in ) :: kdim ! size of ptab |
---|
| 1355 | REAL(wp), DIMENSION(kdim), INTENT( inout ) :: ptab ! input array |
---|
| 1356 | INTEGER , INTENT( in ), OPTIONAL :: kcom |
---|
| 1357 | !! |
---|
| 1358 | INTEGER :: ierror ! temporary integer |
---|
| 1359 | INTEGER :: localcomm |
---|
| 1360 | REAL(wp), DIMENSION(kdim) :: zwork ! temporary workspace |
---|
| 1361 | !!----------------------------------------------------------------------- |
---|
| 1362 | ! |
---|
| 1363 | localcomm = mpi_comm_opa |
---|
| 1364 | IF( PRESENT(kcom) ) localcomm = kcom |
---|
| 1365 | ! |
---|
| 1366 | CALL mpi_allreduce( ptab, zwork, kdim, mpi_double_precision, mpi_sum, localcomm, ierror ) |
---|
| 1367 | ptab(:) = zwork(:) |
---|
| 1368 | ! |
---|
| 1369 | END SUBROUTINE mppsum_a_real |
---|
[869] | 1370 | |
---|
[3] | 1371 | |
---|
[1344] | 1372 | SUBROUTINE mppsum_real( ptab, kcom ) |
---|
| 1373 | !!---------------------------------------------------------------------- |
---|
| 1374 | !! *** routine mppsum_real *** |
---|
| 1375 | !! |
---|
| 1376 | !! ** Purpose : global sum, SCALAR argument case |
---|
| 1377 | !! |
---|
| 1378 | !!----------------------------------------------------------------------- |
---|
| 1379 | REAL(wp), INTENT(inout) :: ptab ! input scalar |
---|
| 1380 | INTEGER , INTENT(in ), OPTIONAL :: kcom |
---|
| 1381 | !! |
---|
| 1382 | INTEGER :: ierror, localcomm |
---|
| 1383 | REAL(wp) :: zwork |
---|
| 1384 | !!----------------------------------------------------------------------- |
---|
| 1385 | ! |
---|
| 1386 | localcomm = mpi_comm_opa |
---|
| 1387 | IF( PRESENT(kcom) ) localcomm = kcom |
---|
| 1388 | ! |
---|
| 1389 | CALL mpi_allreduce( ptab, zwork, 1, mpi_double_precision, mpi_sum, localcomm, ierror ) |
---|
| 1390 | ptab = zwork |
---|
| 1391 | ! |
---|
| 1392 | END SUBROUTINE mppsum_real |
---|
[3] | 1393 | |
---|
| 1394 | |
---|
[1344] | 1395 | SUBROUTINE mpp_minloc2d( ptab, pmask, pmin, ki,kj ) |
---|
| 1396 | !!------------------------------------------------------------------------ |
---|
| 1397 | !! *** routine mpp_minloc *** |
---|
| 1398 | !! |
---|
| 1399 | !! ** Purpose : Compute the global minimum of an array ptab |
---|
| 1400 | !! and also give its global position |
---|
| 1401 | !! |
---|
| 1402 | !! ** Method : Use MPI_ALLREDUCE with MPI_MINLOC |
---|
| 1403 | !! |
---|
| 1404 | !!-------------------------------------------------------------------------- |
---|
| 1405 | REAL(wp), DIMENSION (jpi,jpj), INTENT(in ) :: ptab ! Local 2D array |
---|
| 1406 | REAL(wp), DIMENSION (jpi,jpj), INTENT(in ) :: pmask ! Local mask |
---|
| 1407 | REAL(wp) , INTENT( out) :: pmin ! Global minimum of ptab |
---|
| 1408 | INTEGER , INTENT( out) :: ki, kj ! index of minimum in global frame |
---|
| 1409 | !! |
---|
| 1410 | INTEGER , DIMENSION(2) :: ilocs |
---|
| 1411 | INTEGER :: ierror |
---|
| 1412 | REAL(wp) :: zmin ! local minimum |
---|
| 1413 | REAL(wp), DIMENSION(2,1) :: zain, zaout |
---|
| 1414 | !!----------------------------------------------------------------------- |
---|
| 1415 | ! |
---|
| 1416 | zmin = MINVAL( ptab(:,:) , mask= pmask == 1.e0 ) |
---|
| 1417 | ilocs = MINLOC( ptab(:,:) , mask= pmask == 1.e0 ) |
---|
| 1418 | ! |
---|
| 1419 | ki = ilocs(1) + nimpp - 1 |
---|
| 1420 | kj = ilocs(2) + njmpp - 1 |
---|
| 1421 | ! |
---|
| 1422 | zain(1,:)=zmin |
---|
| 1423 | zain(2,:)=ki+10000.*kj |
---|
| 1424 | ! |
---|
| 1425 | CALL MPI_ALLREDUCE( zain,zaout, 1, MPI_2DOUBLE_PRECISION,MPI_MINLOC,MPI_COMM_OPA,ierror) |
---|
| 1426 | ! |
---|
| 1427 | pmin = zaout(1,1) |
---|
| 1428 | kj = INT(zaout(2,1)/10000.) |
---|
| 1429 | ki = INT(zaout(2,1) - 10000.*kj ) |
---|
| 1430 | ! |
---|
| 1431 | END SUBROUTINE mpp_minloc2d |
---|
[13] | 1432 | |
---|
[3] | 1433 | |
---|
[1344] | 1434 | SUBROUTINE mpp_minloc3d( ptab, pmask, pmin, ki, kj ,kk) |
---|
| 1435 | !!------------------------------------------------------------------------ |
---|
| 1436 | !! *** routine mpp_minloc *** |
---|
| 1437 | !! |
---|
| 1438 | !! ** Purpose : Compute the global minimum of an array ptab |
---|
| 1439 | !! and also give its global position |
---|
| 1440 | !! |
---|
| 1441 | !! ** Method : Use MPI_ALLREDUCE with MPI_MINLOC |
---|
| 1442 | !! |
---|
| 1443 | !!-------------------------------------------------------------------------- |
---|
| 1444 | REAL(wp), DIMENSION (jpi,jpj,jpk), INTENT(in ) :: ptab ! Local 2D array |
---|
| 1445 | REAL(wp), DIMENSION (jpi,jpj,jpk), INTENT(in ) :: pmask ! Local mask |
---|
| 1446 | REAL(wp) , INTENT( out) :: pmin ! Global minimum of ptab |
---|
| 1447 | INTEGER , INTENT( out) :: ki, kj, kk ! index of minimum in global frame |
---|
| 1448 | !! |
---|
| 1449 | INTEGER :: ierror |
---|
| 1450 | REAL(wp) :: zmin ! local minimum |
---|
| 1451 | INTEGER , DIMENSION(3) :: ilocs |
---|
| 1452 | REAL(wp), DIMENSION(2,1) :: zain, zaout |
---|
| 1453 | !!----------------------------------------------------------------------- |
---|
| 1454 | ! |
---|
| 1455 | zmin = MINVAL( ptab(:,:,:) , mask= pmask == 1.e0 ) |
---|
| 1456 | ilocs = MINLOC( ptab(:,:,:) , mask= pmask == 1.e0 ) |
---|
| 1457 | ! |
---|
| 1458 | ki = ilocs(1) + nimpp - 1 |
---|
| 1459 | kj = ilocs(2) + njmpp - 1 |
---|
| 1460 | kk = ilocs(3) |
---|
| 1461 | ! |
---|
| 1462 | zain(1,:)=zmin |
---|
| 1463 | zain(2,:)=ki+10000.*kj+100000000.*kk |
---|
| 1464 | ! |
---|
| 1465 | CALL MPI_ALLREDUCE( zain,zaout, 1, MPI_2DOUBLE_PRECISION,MPI_MINLOC,MPI_COMM_OPA,ierror) |
---|
| 1466 | ! |
---|
| 1467 | pmin = zaout(1,1) |
---|
| 1468 | kk = INT( zaout(2,1) / 100000000. ) |
---|
| 1469 | kj = INT( zaout(2,1) - kk * 100000000. ) / 10000 |
---|
| 1470 | ki = INT( zaout(2,1) - kk * 100000000. -kj * 10000. ) |
---|
| 1471 | ! |
---|
| 1472 | END SUBROUTINE mpp_minloc3d |
---|
[13] | 1473 | |
---|
[3] | 1474 | |
---|
[1344] | 1475 | SUBROUTINE mpp_maxloc2d( ptab, pmask, pmax, ki, kj ) |
---|
| 1476 | !!------------------------------------------------------------------------ |
---|
| 1477 | !! *** routine mpp_maxloc *** |
---|
| 1478 | !! |
---|
| 1479 | !! ** Purpose : Compute the global maximum of an array ptab |
---|
| 1480 | !! and also give its global position |
---|
| 1481 | !! |
---|
| 1482 | !! ** Method : Use MPI_ALLREDUCE with MPI_MINLOC |
---|
| 1483 | !! |
---|
| 1484 | !!-------------------------------------------------------------------------- |
---|
| 1485 | REAL(wp), DIMENSION (jpi,jpj), INTENT(in ) :: ptab ! Local 2D array |
---|
| 1486 | REAL(wp), DIMENSION (jpi,jpj), INTENT(in ) :: pmask ! Local mask |
---|
| 1487 | REAL(wp) , INTENT( out) :: pmax ! Global maximum of ptab |
---|
| 1488 | INTEGER , INTENT( out) :: ki, kj ! index of maximum in global frame |
---|
| 1489 | !! |
---|
| 1490 | INTEGER :: ierror |
---|
| 1491 | INTEGER, DIMENSION (2) :: ilocs |
---|
| 1492 | REAL(wp) :: zmax ! local maximum |
---|
| 1493 | REAL(wp), DIMENSION(2,1) :: zain, zaout |
---|
| 1494 | !!----------------------------------------------------------------------- |
---|
| 1495 | ! |
---|
| 1496 | zmax = MAXVAL( ptab(:,:) , mask= pmask == 1.e0 ) |
---|
| 1497 | ilocs = MAXLOC( ptab(:,:) , mask= pmask == 1.e0 ) |
---|
| 1498 | ! |
---|
| 1499 | ki = ilocs(1) + nimpp - 1 |
---|
| 1500 | kj = ilocs(2) + njmpp - 1 |
---|
| 1501 | ! |
---|
| 1502 | zain(1,:) = zmax |
---|
| 1503 | zain(2,:) = ki + 10000. * kj |
---|
| 1504 | ! |
---|
| 1505 | CALL MPI_ALLREDUCE( zain,zaout, 1, MPI_2DOUBLE_PRECISION,MPI_MAXLOC,MPI_COMM_OPA,ierror) |
---|
| 1506 | ! |
---|
| 1507 | pmax = zaout(1,1) |
---|
| 1508 | kj = INT( zaout(2,1) / 10000. ) |
---|
| 1509 | ki = INT( zaout(2,1) - 10000.* kj ) |
---|
| 1510 | ! |
---|
| 1511 | END SUBROUTINE mpp_maxloc2d |
---|
[3] | 1512 | |
---|
[13] | 1513 | |
---|
[1344] | 1514 | SUBROUTINE mpp_maxloc3d( ptab, pmask, pmax, ki, kj, kk ) |
---|
| 1515 | !!------------------------------------------------------------------------ |
---|
| 1516 | !! *** routine mpp_maxloc *** |
---|
| 1517 | !! |
---|
| 1518 | !! ** Purpose : Compute the global maximum of an array ptab |
---|
| 1519 | !! and also give its global position |
---|
| 1520 | !! |
---|
| 1521 | !! ** Method : Use MPI_ALLREDUCE with MPI_MINLOC |
---|
| 1522 | !! |
---|
| 1523 | !!-------------------------------------------------------------------------- |
---|
| 1524 | REAL(wp), DIMENSION (jpi,jpj,jpk), INTENT(in ) :: ptab ! Local 2D array |
---|
| 1525 | REAL(wp), DIMENSION (jpi,jpj,jpk), INTENT(in ) :: pmask ! Local mask |
---|
| 1526 | REAL(wp) , INTENT( out) :: pmax ! Global maximum of ptab |
---|
| 1527 | INTEGER , INTENT( out) :: ki, kj, kk ! index of maximum in global frame |
---|
| 1528 | !! |
---|
| 1529 | REAL(wp) :: zmax ! local maximum |
---|
| 1530 | REAL(wp), DIMENSION(2,1) :: zain, zaout |
---|
| 1531 | INTEGER , DIMENSION(3) :: ilocs |
---|
| 1532 | INTEGER :: ierror |
---|
| 1533 | !!----------------------------------------------------------------------- |
---|
| 1534 | ! |
---|
| 1535 | zmax = MAXVAL( ptab(:,:,:) , mask= pmask == 1.e0 ) |
---|
| 1536 | ilocs = MAXLOC( ptab(:,:,:) , mask= pmask == 1.e0 ) |
---|
| 1537 | ! |
---|
| 1538 | ki = ilocs(1) + nimpp - 1 |
---|
| 1539 | kj = ilocs(2) + njmpp - 1 |
---|
| 1540 | kk = ilocs(3) |
---|
| 1541 | ! |
---|
| 1542 | zain(1,:)=zmax |
---|
| 1543 | zain(2,:)=ki+10000.*kj+100000000.*kk |
---|
| 1544 | ! |
---|
| 1545 | CALL MPI_ALLREDUCE( zain,zaout, 1, MPI_2DOUBLE_PRECISION,MPI_MAXLOC,MPI_COMM_OPA,ierror) |
---|
| 1546 | ! |
---|
| 1547 | pmax = zaout(1,1) |
---|
| 1548 | kk = INT( zaout(2,1) / 100000000. ) |
---|
| 1549 | kj = INT( zaout(2,1) - kk * 100000000. ) / 10000 |
---|
| 1550 | ki = INT( zaout(2,1) - kk * 100000000. -kj * 10000. ) |
---|
| 1551 | ! |
---|
| 1552 | END SUBROUTINE mpp_maxloc3d |
---|
[3] | 1553 | |
---|
[869] | 1554 | |
---|
[1344] | 1555 | SUBROUTINE mppsync() |
---|
| 1556 | !!---------------------------------------------------------------------- |
---|
| 1557 | !! *** routine mppsync *** |
---|
| 1558 | !! |
---|
| 1559 | !! ** Purpose : Massively parallel processors, synchroneous |
---|
| 1560 | !! |
---|
| 1561 | !!----------------------------------------------------------------------- |
---|
| 1562 | INTEGER :: ierror |
---|
| 1563 | !!----------------------------------------------------------------------- |
---|
| 1564 | ! |
---|
| 1565 | CALL mpi_barrier( mpi_comm_opa, ierror ) |
---|
| 1566 | ! |
---|
| 1567 | END SUBROUTINE mppsync |
---|
[3] | 1568 | |
---|
| 1569 | |
---|
[1344] | 1570 | SUBROUTINE mppstop |
---|
| 1571 | !!---------------------------------------------------------------------- |
---|
| 1572 | !! *** routine mppstop *** |
---|
| 1573 | !! |
---|
| 1574 | !! ** purpose : Stop massilively parallel processors method |
---|
| 1575 | !! |
---|
| 1576 | !!---------------------------------------------------------------------- |
---|
| 1577 | INTEGER :: info |
---|
| 1578 | !!---------------------------------------------------------------------- |
---|
| 1579 | ! |
---|
| 1580 | CALL mppsync |
---|
| 1581 | CALL mpi_finalize( info ) |
---|
| 1582 | ! |
---|
| 1583 | END SUBROUTINE mppstop |
---|
[3] | 1584 | |
---|
| 1585 | |
---|
[1344] | 1586 | SUBROUTINE mppobc( ptab, kd1, kd2, kl, kk, ktype, kij ) |
---|
| 1587 | !!---------------------------------------------------------------------- |
---|
| 1588 | !! *** routine mppobc *** |
---|
| 1589 | !! |
---|
| 1590 | !! ** Purpose : Message passing manadgement for open boundary |
---|
| 1591 | !! conditions array |
---|
| 1592 | !! |
---|
| 1593 | !! ** Method : Use mppsend and mpprecv function for passing mask |
---|
| 1594 | !! between processors following neighboring subdomains. |
---|
| 1595 | !! domain parameters |
---|
| 1596 | !! nlci : first dimension of the local subdomain |
---|
| 1597 | !! nlcj : second dimension of the local subdomain |
---|
| 1598 | !! nbondi : mark for "east-west local boundary" |
---|
| 1599 | !! nbondj : mark for "north-south local boundary" |
---|
| 1600 | !! noea : number for local neighboring processors |
---|
| 1601 | !! nowe : number for local neighboring processors |
---|
| 1602 | !! noso : number for local neighboring processors |
---|
| 1603 | !! nono : number for local neighboring processors |
---|
| 1604 | !! |
---|
| 1605 | !!---------------------------------------------------------------------- |
---|
| 1606 | INTEGER , INTENT(in ) :: kd1, kd2 ! starting and ending indices |
---|
| 1607 | INTEGER , INTENT(in ) :: kl ! index of open boundary |
---|
| 1608 | INTEGER , INTENT(in ) :: kk ! vertical dimension |
---|
| 1609 | INTEGER , INTENT(in ) :: ktype ! define north/south or east/west cdt |
---|
| 1610 | ! ! = 1 north/south ; = 2 east/west |
---|
| 1611 | INTEGER , INTENT(in ) :: kij ! horizontal dimension |
---|
| 1612 | REAL(wp), INTENT(inout), DIMENSION(kij,kk) :: ptab ! variable array |
---|
| 1613 | !! |
---|
| 1614 | INTEGER :: ji, jj, jk, jl ! dummy loop indices |
---|
| 1615 | INTEGER :: iipt0, iipt1, ilpt1 ! temporary integers |
---|
| 1616 | INTEGER :: ijpt0, ijpt1 ! - - |
---|
| 1617 | INTEGER :: imigr, iihom, ijhom ! - - |
---|
| 1618 | INTEGER :: ml_req1, ml_req2, ml_err ! for key_mpi_isend |
---|
| 1619 | INTEGER :: ml_stat(MPI_STATUS_SIZE) ! for key_mpi_isend |
---|
| 1620 | REAL(wp), DIMENSION(jpi,jpj) :: ztab ! temporary workspace |
---|
| 1621 | !!---------------------------------------------------------------------- |
---|
[3] | 1622 | |
---|
[1344] | 1623 | ! boundary condition initialization |
---|
| 1624 | ! --------------------------------- |
---|
| 1625 | ztab(:,:) = 0.e0 |
---|
| 1626 | ! |
---|
| 1627 | IF( ktype==1 ) THEN ! north/south boundaries |
---|
| 1628 | iipt0 = MAX( 1, MIN(kd1 - nimpp+1, nlci ) ) |
---|
| 1629 | iipt1 = MAX( 0, MIN(kd2 - nimpp+1, nlci - 1 ) ) |
---|
| 1630 | ilpt1 = MAX( 1, MIN(kd2 - nimpp+1, nlci ) ) |
---|
| 1631 | ijpt0 = MAX( 1, MIN(kl - njmpp+1, nlcj ) ) |
---|
| 1632 | ijpt1 = MAX( 0, MIN(kl - njmpp+1, nlcj - 1 ) ) |
---|
| 1633 | ELSEIF( ktype==2 ) THEN ! east/west boundaries |
---|
| 1634 | iipt0 = MAX( 1, MIN(kl - nimpp+1, nlci ) ) |
---|
| 1635 | iipt1 = MAX( 0, MIN(kl - nimpp+1, nlci - 1 ) ) |
---|
| 1636 | ijpt0 = MAX( 1, MIN(kd1 - njmpp+1, nlcj ) ) |
---|
| 1637 | ijpt1 = MAX( 0, MIN(kd2 - njmpp+1, nlcj - 1 ) ) |
---|
| 1638 | ilpt1 = MAX( 1, MIN(kd2 - njmpp+1, nlcj ) ) |
---|
| 1639 | ELSE |
---|
| 1640 | CALL ctl_stop( 'mppobc: bad ktype' ) |
---|
| 1641 | ENDIF |
---|
| 1642 | |
---|
| 1643 | ! Communication level by level |
---|
| 1644 | ! ---------------------------- |
---|
| 1645 | !!gm Remark : this is very time consumming!!! |
---|
| 1646 | ! ! ------------------------ ! |
---|
| 1647 | DO jk = 1, kk ! Loop over the levels ! |
---|
| 1648 | ! ! ------------------------ ! |
---|
| 1649 | ! |
---|
| 1650 | IF( ktype == 1 ) THEN ! north/south boundaries |
---|
| 1651 | DO jj = ijpt0, ijpt1 |
---|
| 1652 | DO ji = iipt0, iipt1 |
---|
| 1653 | ztab(ji,jj) = ptab(ji,jk) |
---|
| 1654 | END DO |
---|
| 1655 | END DO |
---|
| 1656 | ELSEIF( ktype == 2 ) THEN ! east/west boundaries |
---|
| 1657 | DO jj = ijpt0, ijpt1 |
---|
| 1658 | DO ji = iipt0, iipt1 |
---|
| 1659 | ztab(ji,jj) = ptab(jj,jk) |
---|
| 1660 | END DO |
---|
| 1661 | END DO |
---|
| 1662 | ENDIF |
---|
[13] | 1663 | |
---|
[3] | 1664 | |
---|
[1344] | 1665 | ! 1. East and west directions |
---|
| 1666 | ! --------------------------- |
---|
| 1667 | ! |
---|
| 1668 | IF( nbondi /= 2 ) THEN ! Read Dirichlet lateral conditions |
---|
| 1669 | iihom = nlci-nreci |
---|
| 1670 | DO jl = 1, jpreci |
---|
| 1671 | t2ew(:,jl,1) = ztab(jpreci+jl,:) |
---|
| 1672 | t2we(:,jl,1) = ztab(iihom +jl,:) |
---|
| 1673 | END DO |
---|
| 1674 | ENDIF |
---|
| 1675 | ! |
---|
| 1676 | ! ! Migrations |
---|
| 1677 | imigr=jpreci*jpj |
---|
| 1678 | ! |
---|
| 1679 | IF( nbondi == -1 ) THEN |
---|
| 1680 | CALL mppsend( 2, t2we(1,1,1), imigr, noea, ml_req1 ) |
---|
| 1681 | CALL mpprecv( 1, t2ew(1,1,2), imigr ) |
---|
| 1682 | IF(l_isend) CALL mpi_wait( ml_req1, ml_stat, ml_err ) |
---|
| 1683 | ELSEIF( nbondi == 0 ) THEN |
---|
| 1684 | CALL mppsend( 1, t2ew(1,1,1), imigr, nowe, ml_req1 ) |
---|
| 1685 | CALL mppsend( 2, t2we(1,1,1), imigr, noea, ml_req2 ) |
---|
| 1686 | CALL mpprecv( 1, t2ew(1,1,2), imigr ) |
---|
| 1687 | CALL mpprecv( 2, t2we(1,1,2), imigr ) |
---|
| 1688 | IF(l_isend) CALL mpi_wait( ml_req1, ml_stat, ml_err ) |
---|
| 1689 | IF(l_isend) CALL mpi_wait( ml_req2, ml_stat, ml_err ) |
---|
| 1690 | ELSEIF( nbondi == 1 ) THEN |
---|
| 1691 | CALL mppsend( 1, t2ew(1,1,1), imigr, nowe, ml_req1 ) |
---|
| 1692 | CALL mpprecv( 2, t2we(1,1,2), imigr ) |
---|
| 1693 | IF(l_isend) CALL mpi_wait( ml_req1, ml_stat, ml_err ) |
---|
| 1694 | ENDIF |
---|
| 1695 | ! |
---|
| 1696 | ! ! Write Dirichlet lateral conditions |
---|
| 1697 | iihom = nlci-jpreci |
---|
| 1698 | ! |
---|
| 1699 | IF( nbondi == 0 .OR. nbondi == 1 ) THEN |
---|
| 1700 | DO jl = 1, jpreci |
---|
| 1701 | ztab(jl,:) = t2we(:,jl,2) |
---|
| 1702 | END DO |
---|
| 1703 | ENDIF |
---|
| 1704 | IF( nbondi == -1 .OR. nbondi == 0 ) THEN |
---|
| 1705 | DO jl = 1, jpreci |
---|
| 1706 | ztab(iihom+jl,:) = t2ew(:,jl,2) |
---|
| 1707 | END DO |
---|
| 1708 | ENDIF |
---|
[3] | 1709 | |
---|
| 1710 | |
---|
[1344] | 1711 | ! 2. North and south directions |
---|
| 1712 | ! ----------------------------- |
---|
| 1713 | ! |
---|
| 1714 | IF( nbondj /= 2 ) THEN ! Read Dirichlet lateral conditions |
---|
| 1715 | ijhom = nlcj-nrecj |
---|
| 1716 | DO jl = 1, jprecj |
---|
| 1717 | t2sn(:,jl,1) = ztab(:,ijhom +jl) |
---|
| 1718 | t2ns(:,jl,1) = ztab(:,jprecj+jl) |
---|
| 1719 | END DO |
---|
| 1720 | ENDIF |
---|
| 1721 | ! |
---|
| 1722 | ! ! Migrations |
---|
| 1723 | imigr = jprecj * jpi |
---|
| 1724 | ! |
---|
| 1725 | IF( nbondj == -1 ) THEN |
---|
| 1726 | CALL mppsend( 4, t2sn(1,1,1), imigr, nono, ml_req1 ) |
---|
| 1727 | CALL mpprecv( 3, t2ns(1,1,2), imigr ) |
---|
| 1728 | IF(l_isend) CALL mpi_wait( ml_req1, ml_stat, ml_err ) |
---|
| 1729 | ELSEIF( nbondj == 0 ) THEN |
---|
| 1730 | CALL mppsend( 3, t2ns(1,1,1), imigr, noso, ml_req1 ) |
---|
| 1731 | CALL mppsend( 4, t2sn(1,1,1), imigr, nono, ml_req2 ) |
---|
| 1732 | CALL mpprecv( 3, t2ns(1,1,2), imigr ) |
---|
| 1733 | CALL mpprecv( 4, t2sn(1,1,2), imigr ) |
---|
| 1734 | IF( l_isend ) CALL mpi_wait( ml_req1, ml_stat, ml_err ) |
---|
| 1735 | IF( l_isend ) CALL mpi_wait( ml_req2, ml_stat, ml_err ) |
---|
| 1736 | ELSEIF( nbondj == 1 ) THEN |
---|
| 1737 | CALL mppsend( 3, t2ns(1,1,1), imigr, noso, ml_req1 ) |
---|
| 1738 | CALL mpprecv( 4, t2sn(1,1,2), imigr) |
---|
| 1739 | IF( l_isend ) CALL mpi_wait( ml_req1, ml_stat, ml_err ) |
---|
| 1740 | ENDIF |
---|
| 1741 | ! |
---|
| 1742 | ! ! Write Dirichlet lateral conditions |
---|
| 1743 | ijhom = nlcj - jprecj |
---|
| 1744 | IF( nbondj == 0 .OR. nbondj == 1 ) THEN |
---|
| 1745 | DO jl = 1, jprecj |
---|
| 1746 | ztab(:,jl) = t2sn(:,jl,2) |
---|
| 1747 | END DO |
---|
| 1748 | ENDIF |
---|
| 1749 | IF( nbondj == 0 .OR. nbondj == -1 ) THEN |
---|
| 1750 | DO jl = 1, jprecj |
---|
| 1751 | ztab(:,ijhom+jl) = t2ns(:,jl,2) |
---|
| 1752 | END DO |
---|
| 1753 | ENDIF |
---|
| 1754 | IF( ktype==1 .AND. kd1 <= jpi+nimpp-1 .AND. nimpp <= kd2 ) THEN |
---|
| 1755 | DO jj = ijpt0, ijpt1 ! north/south boundaries |
---|
| 1756 | DO ji = iipt0,ilpt1 |
---|
| 1757 | ptab(ji,jk) = ztab(ji,jj) |
---|
| 1758 | END DO |
---|
| 1759 | END DO |
---|
| 1760 | ELSEIF( ktype==2 .AND. kd1 <= jpj+njmpp-1 .AND. njmpp <= kd2 ) THEN |
---|
| 1761 | DO jj = ijpt0, ilpt1 ! east/west boundaries |
---|
| 1762 | DO ji = iipt0,iipt1 |
---|
| 1763 | ptab(jj,jk) = ztab(ji,jj) |
---|
| 1764 | END DO |
---|
| 1765 | END DO |
---|
| 1766 | ENDIF |
---|
| 1767 | ! |
---|
| 1768 | END DO |
---|
| 1769 | ! |
---|
| 1770 | END SUBROUTINE mppobc |
---|
| 1771 | |
---|
[13] | 1772 | |
---|
[1344] | 1773 | SUBROUTINE mpp_comm_free( kcom ) |
---|
| 1774 | !!---------------------------------------------------------------------- |
---|
| 1775 | !!---------------------------------------------------------------------- |
---|
| 1776 | INTEGER, INTENT(in) :: kcom |
---|
| 1777 | !! |
---|
| 1778 | INTEGER :: ierr |
---|
| 1779 | !!---------------------------------------------------------------------- |
---|
| 1780 | ! |
---|
| 1781 | CALL MPI_COMM_FREE(kcom, ierr) |
---|
| 1782 | ! |
---|
| 1783 | END SUBROUTINE mpp_comm_free |
---|
[3] | 1784 | |
---|
[869] | 1785 | |
---|
[1344] | 1786 | SUBROUTINE mpp_ini_ice( pindic ) |
---|
| 1787 | !!---------------------------------------------------------------------- |
---|
| 1788 | !! *** routine mpp_ini_ice *** |
---|
| 1789 | !! |
---|
| 1790 | !! ** Purpose : Initialize special communicator for ice areas |
---|
| 1791 | !! condition together with global variables needed in the ddmpp folding |
---|
| 1792 | !! |
---|
| 1793 | !! ** Method : - Look for ice processors in ice routines |
---|
| 1794 | !! - Put their number in nrank_ice |
---|
| 1795 | !! - Create groups for the world processors and the ice processors |
---|
| 1796 | !! - Create a communicator for ice processors |
---|
| 1797 | !! |
---|
| 1798 | !! ** output |
---|
| 1799 | !! njmppmax = njmpp for northern procs |
---|
| 1800 | !! ndim_rank_ice = number of processors with ice |
---|
| 1801 | !! nrank_ice (ndim_rank_ice) = ice processors |
---|
| 1802 | !! ngrp_world = group ID for the world processors |
---|
| 1803 | !! ngrp_ice = group ID for the ice processors |
---|
| 1804 | !! ncomm_ice = communicator for the ice procs. |
---|
| 1805 | !! n_ice_root = number (in the world) of proc 0 in the ice comm. |
---|
| 1806 | !! |
---|
| 1807 | !!---------------------------------------------------------------------- |
---|
| 1808 | INTEGER, INTENT(in) :: pindic |
---|
| 1809 | !! |
---|
| 1810 | INTEGER :: ierr |
---|
| 1811 | INTEGER :: jjproc |
---|
| 1812 | INTEGER :: ii |
---|
| 1813 | INTEGER, DIMENSION(jpnij) :: kice |
---|
| 1814 | INTEGER, DIMENSION(jpnij) :: zwork |
---|
| 1815 | !!---------------------------------------------------------------------- |
---|
| 1816 | ! |
---|
| 1817 | ! Look for how many procs with sea-ice |
---|
| 1818 | ! |
---|
| 1819 | kice = 0 |
---|
| 1820 | DO jjproc = 1, jpnij |
---|
| 1821 | IF( jjproc == narea .AND. pindic .GT. 0 ) kice(jjproc) = 1 |
---|
| 1822 | END DO |
---|
| 1823 | ! |
---|
| 1824 | zwork = 0 |
---|
| 1825 | CALL MPI_ALLREDUCE( kice, zwork, jpnij, mpi_integer, mpi_sum, mpi_comm_opa, ierr ) |
---|
| 1826 | ndim_rank_ice = SUM( zwork ) |
---|
[3] | 1827 | |
---|
[1344] | 1828 | ! Allocate the right size to nrank_north |
---|
[1208] | 1829 | #if ! defined key_agrif |
---|
[1441] | 1830 | IF( ALLOCATED ( nrank_ice ) ) DEALLOCATE( nrank_ice ) |
---|
[1208] | 1831 | #else |
---|
[1441] | 1832 | IF( ASSOCIATED( nrank_ice ) ) DEALLOCATE( nrank_ice ) |
---|
[1208] | 1833 | #endif |
---|
[1344] | 1834 | ALLOCATE( nrank_ice(ndim_rank_ice) ) |
---|
| 1835 | ! |
---|
| 1836 | ii = 0 |
---|
| 1837 | nrank_ice = 0 |
---|
| 1838 | DO jjproc = 1, jpnij |
---|
| 1839 | IF( zwork(jjproc) == 1) THEN |
---|
| 1840 | ii = ii + 1 |
---|
| 1841 | nrank_ice(ii) = jjproc -1 |
---|
| 1842 | ENDIF |
---|
| 1843 | END DO |
---|
[1208] | 1844 | |
---|
[1344] | 1845 | ! Create the world group |
---|
| 1846 | CALL MPI_COMM_GROUP( mpi_comm_opa, ngrp_world, ierr ) |
---|
[869] | 1847 | |
---|
[1344] | 1848 | ! Create the ice group from the world group |
---|
| 1849 | CALL MPI_GROUP_INCL( ngrp_world, ndim_rank_ice, nrank_ice, ngrp_ice, ierr ) |
---|
[869] | 1850 | |
---|
[1344] | 1851 | ! Create the ice communicator , ie the pool of procs with sea-ice |
---|
| 1852 | CALL MPI_COMM_CREATE( mpi_comm_opa, ngrp_ice, ncomm_ice, ierr ) |
---|
[869] | 1853 | |
---|
[1344] | 1854 | ! Find proc number in the world of proc 0 in the north |
---|
| 1855 | ! The following line seems to be useless, we just comment & keep it as reminder |
---|
| 1856 | ! CALL MPI_GROUP_TRANSLATE_RANKS(ngrp_ice,1,0,ngrp_world,n_ice_root,ierr) |
---|
| 1857 | ! |
---|
| 1858 | END SUBROUTINE mpp_ini_ice |
---|
[869] | 1859 | |
---|
| 1860 | |
---|
[1345] | 1861 | SUBROUTINE mpp_ini_znl |
---|
| 1862 | !!---------------------------------------------------------------------- |
---|
| 1863 | !! *** routine mpp_ini_znl *** |
---|
| 1864 | !! |
---|
| 1865 | !! ** Purpose : Initialize special communicator for computing zonal sum |
---|
| 1866 | !! |
---|
| 1867 | !! ** Method : - Look for processors in the same row |
---|
| 1868 | !! - Put their number in nrank_znl |
---|
| 1869 | !! - Create group for the znl processors |
---|
| 1870 | !! - Create a communicator for znl processors |
---|
| 1871 | !! - Determine if processor should write znl files |
---|
| 1872 | !! |
---|
| 1873 | !! ** output |
---|
| 1874 | !! ndim_rank_znl = number of processors on the same row |
---|
| 1875 | !! ngrp_znl = group ID for the znl processors |
---|
| 1876 | !! ncomm_znl = communicator for the ice procs. |
---|
| 1877 | !! n_znl_root = number (in the world) of proc 0 in the ice comm. |
---|
| 1878 | !! |
---|
| 1879 | !!---------------------------------------------------------------------- |
---|
| 1880 | INTEGER :: ierr |
---|
| 1881 | INTEGER :: jproc |
---|
| 1882 | INTEGER :: ii |
---|
| 1883 | INTEGER, DIMENSION(jpnij) :: kwork |
---|
| 1884 | ! |
---|
| 1885 | !-$$ WRITE (numout,*) 'mpp_ini_znl ', nproc, ' - ngrp_world : ', ngrp_world |
---|
| 1886 | !-$$ WRITE (numout,*) 'mpp_ini_znl ', nproc, ' - mpi_comm_world : ', mpi_comm_world |
---|
| 1887 | !-$$ WRITE (numout,*) 'mpp_ini_znl ', nproc, ' - mpi_comm_opa : ', mpi_comm_opa |
---|
| 1888 | ! |
---|
| 1889 | IF ( jpnj == 1 ) THEN |
---|
| 1890 | ngrp_znl = ngrp_world |
---|
| 1891 | ncomm_znl = mpi_comm_opa |
---|
| 1892 | ELSE |
---|
| 1893 | ! |
---|
| 1894 | CALL MPI_ALLGATHER ( njmpp, 1, mpi_integer, kwork, 1, mpi_integer, mpi_comm_opa, ierr ) |
---|
| 1895 | !-$$ WRITE (numout,*) 'mpp_ini_znl ', nproc, ' - kwork pour njmpp : ', kwork |
---|
| 1896 | !-$$ CALL flush(numout) |
---|
| 1897 | ! |
---|
| 1898 | ! Count number of processors on the same row |
---|
| 1899 | ndim_rank_znl = 0 |
---|
| 1900 | DO jproc=1,jpnij |
---|
| 1901 | IF ( kwork(jproc) == njmpp ) THEN |
---|
| 1902 | ndim_rank_znl = ndim_rank_znl + 1 |
---|
| 1903 | ENDIF |
---|
| 1904 | END DO |
---|
| 1905 | !-$$ WRITE (numout,*) 'mpp_ini_znl ', nproc, ' - ndim_rank_znl : ', ndim_rank_znl |
---|
| 1906 | !-$$ CALL flush(numout) |
---|
| 1907 | ! Allocate the right size to nrank_znl |
---|
| 1908 | #if ! defined key_agrif |
---|
[1441] | 1909 | IF (ALLOCATED (nrank_znl)) DEALLOCATE(nrank_znl) |
---|
[1345] | 1910 | #else |
---|
[1441] | 1911 | IF (ASSOCIATED(nrank_znl)) DEALLOCATE(nrank_znl) |
---|
[1345] | 1912 | #endif |
---|
| 1913 | ALLOCATE(nrank_znl(ndim_rank_znl)) |
---|
| 1914 | ii = 0 |
---|
| 1915 | nrank_znl (:) = 0 |
---|
| 1916 | DO jproc=1,jpnij |
---|
| 1917 | IF ( kwork(jproc) == njmpp) THEN |
---|
| 1918 | ii = ii + 1 |
---|
| 1919 | nrank_znl(ii) = jproc -1 |
---|
| 1920 | ENDIF |
---|
| 1921 | END DO |
---|
| 1922 | !-$$ WRITE (numout,*) 'mpp_ini_znl ', nproc, ' - nrank_znl : ', nrank_znl |
---|
| 1923 | !-$$ CALL flush(numout) |
---|
| 1924 | |
---|
| 1925 | ! Create the opa group |
---|
| 1926 | CALL MPI_COMM_GROUP(mpi_comm_opa,ngrp_opa,ierr) |
---|
| 1927 | !-$$ WRITE (numout,*) 'mpp_ini_znl ', nproc, ' - ngrp_opa : ', ngrp_opa |
---|
| 1928 | !-$$ CALL flush(numout) |
---|
| 1929 | |
---|
| 1930 | ! Create the znl group from the opa group |
---|
| 1931 | CALL MPI_GROUP_INCL ( ngrp_opa, ndim_rank_znl, nrank_znl, ngrp_znl, ierr ) |
---|
| 1932 | !-$$ WRITE (numout,*) 'mpp_ini_znl ', nproc, ' - ngrp_znl ', ngrp_znl |
---|
| 1933 | !-$$ CALL flush(numout) |
---|
| 1934 | |
---|
| 1935 | ! Create the znl communicator from the opa communicator, ie the pool of procs in the same row |
---|
| 1936 | CALL MPI_COMM_CREATE ( mpi_comm_opa, ngrp_znl, ncomm_znl, ierr ) |
---|
| 1937 | !-$$ WRITE (numout,*) 'mpp_ini_znl ', nproc, ' - ncomm_znl ', ncomm_znl |
---|
| 1938 | !-$$ CALL flush(numout) |
---|
| 1939 | ! |
---|
| 1940 | END IF |
---|
| 1941 | |
---|
| 1942 | ! Determines if processor if the first (starting from i=1) on the row |
---|
| 1943 | IF ( jpni == 1 ) THEN |
---|
| 1944 | l_znl_root = .TRUE. |
---|
| 1945 | ELSE |
---|
| 1946 | l_znl_root = .FALSE. |
---|
| 1947 | kwork (1) = nimpp |
---|
| 1948 | CALL mpp_min ( kwork(1), kcom = ncomm_znl) |
---|
| 1949 | IF ( nimpp == kwork(1)) l_znl_root = .TRUE. |
---|
| 1950 | END IF |
---|
| 1951 | |
---|
| 1952 | END SUBROUTINE mpp_ini_znl |
---|
| 1953 | |
---|
| 1954 | |
---|
[1344] | 1955 | SUBROUTINE mpp_ini_north |
---|
| 1956 | !!---------------------------------------------------------------------- |
---|
| 1957 | !! *** routine mpp_ini_north *** |
---|
| 1958 | !! |
---|
| 1959 | !! ** Purpose : Initialize special communicator for north folding |
---|
| 1960 | !! condition together with global variables needed in the mpp folding |
---|
| 1961 | !! |
---|
| 1962 | !! ** Method : - Look for northern processors |
---|
| 1963 | !! - Put their number in nrank_north |
---|
| 1964 | !! - Create groups for the world processors and the north processors |
---|
| 1965 | !! - Create a communicator for northern processors |
---|
| 1966 | !! |
---|
| 1967 | !! ** output |
---|
| 1968 | !! njmppmax = njmpp for northern procs |
---|
| 1969 | !! ndim_rank_north = number of processors in the northern line |
---|
| 1970 | !! nrank_north (ndim_rank_north) = number of the northern procs. |
---|
| 1971 | !! ngrp_world = group ID for the world processors |
---|
| 1972 | !! ngrp_north = group ID for the northern processors |
---|
| 1973 | !! ncomm_north = communicator for the northern procs. |
---|
| 1974 | !! north_root = number (in the world) of proc 0 in the northern comm. |
---|
| 1975 | !! |
---|
| 1976 | !!---------------------------------------------------------------------- |
---|
| 1977 | INTEGER :: ierr |
---|
| 1978 | INTEGER :: jjproc |
---|
| 1979 | INTEGER :: ii, ji |
---|
| 1980 | !!---------------------------------------------------------------------- |
---|
| 1981 | ! |
---|
| 1982 | njmppmax = MAXVAL( njmppt ) |
---|
| 1983 | ! |
---|
| 1984 | ! Look for how many procs on the northern boundary |
---|
| 1985 | ndim_rank_north = 0 |
---|
| 1986 | DO jjproc = 1, jpnij |
---|
| 1987 | IF( njmppt(jjproc) == njmppmax ) ndim_rank_north = ndim_rank_north + 1 |
---|
| 1988 | END DO |
---|
| 1989 | ! |
---|
| 1990 | ! Allocate the right size to nrank_north |
---|
[1441] | 1991 | #if ! defined key_agrif |
---|
| 1992 | IF (ALLOCATED (nrank_north)) DEALLOCATE(nrank_north) |
---|
| 1993 | #else |
---|
| 1994 | IF (ASSOCIATED(nrank_north)) DEALLOCATE(nrank_north) |
---|
| 1995 | #endif |
---|
[1344] | 1996 | ALLOCATE( nrank_north(ndim_rank_north) ) |
---|
[869] | 1997 | |
---|
[1344] | 1998 | ! Fill the nrank_north array with proc. number of northern procs. |
---|
| 1999 | ! Note : the rank start at 0 in MPI |
---|
| 2000 | ii = 0 |
---|
| 2001 | DO ji = 1, jpnij |
---|
| 2002 | IF ( njmppt(ji) == njmppmax ) THEN |
---|
| 2003 | ii=ii+1 |
---|
| 2004 | nrank_north(ii)=ji-1 |
---|
| 2005 | END IF |
---|
| 2006 | END DO |
---|
| 2007 | ! |
---|
| 2008 | ! create the world group |
---|
| 2009 | CALL MPI_COMM_GROUP( mpi_comm_opa, ngrp_world, ierr ) |
---|
| 2010 | ! |
---|
| 2011 | ! Create the North group from the world group |
---|
| 2012 | CALL MPI_GROUP_INCL( ngrp_world, ndim_rank_north, nrank_north, ngrp_north, ierr ) |
---|
| 2013 | ! |
---|
| 2014 | ! Create the North communicator , ie the pool of procs in the north group |
---|
| 2015 | CALL MPI_COMM_CREATE( mpi_comm_opa, ngrp_north, ncomm_north, ierr ) |
---|
| 2016 | ! |
---|
| 2017 | END SUBROUTINE mpp_ini_north |
---|
[869] | 2018 | |
---|
| 2019 | |
---|
[1344] | 2020 | SUBROUTINE mpp_lbc_north_3d( pt3d, cd_type, psgn ) |
---|
[51] | 2021 | !!--------------------------------------------------------------------- |
---|
| 2022 | !! *** routine mpp_lbc_north_3d *** |
---|
| 2023 | !! |
---|
[1344] | 2024 | !! ** Purpose : Ensure proper north fold horizontal bondary condition |
---|
| 2025 | !! in mpp configuration in case of jpn1 > 1 |
---|
[51] | 2026 | !! |
---|
[1344] | 2027 | !! ** Method : North fold condition and mpp with more than one proc |
---|
| 2028 | !! in i-direction require a specific treatment. We gather |
---|
| 2029 | !! the 4 northern lines of the global domain on 1 processor |
---|
| 2030 | !! and apply lbc north-fold on this sub array. Then we |
---|
| 2031 | !! scatter the north fold array back to the processors. |
---|
[51] | 2032 | !! |
---|
| 2033 | !!---------------------------------------------------------------------- |
---|
[1344] | 2034 | REAL(wp), DIMENSION(jpi,jpj,jpk), INTENT(inout) :: pt3d ! 3D array on which the b.c. is applied |
---|
| 2035 | CHARACTER(len=1) , INTENT(in ) :: cd_type ! nature of pt3d grid-points |
---|
| 2036 | ! ! = T , U , V , F or W gridpoints |
---|
| 2037 | REAL(wp) , INTENT(in ) :: psgn ! = -1. the sign change across the north fold |
---|
| 2038 | !! ! = 1. , the sign is kept |
---|
| 2039 | INTEGER :: ji, jj, jr |
---|
| 2040 | INTEGER :: ierr, itaille, ildi, ilei, iilb |
---|
| 2041 | INTEGER :: ijpj, ijpjm1, ij, iproc |
---|
| 2042 | REAL(wp), DIMENSION(jpiglo,4,jpk) :: ztab |
---|
| 2043 | REAL(wp), DIMENSION(jpi ,4,jpk) :: znorthloc |
---|
| 2044 | REAL(wp), DIMENSION(jpi ,4,jpk,jpni) :: znorthgloio |
---|
[51] | 2045 | !!---------------------------------------------------------------------- |
---|
[1344] | 2046 | ! |
---|
| 2047 | ijpj = 4 |
---|
| 2048 | ijpjm1 = 3 |
---|
[1874] | 2049 | ztab(:,:,:) = 0.e0 |
---|
[1344] | 2050 | ! |
---|
| 2051 | DO jj = nlcj - ijpj +1, nlcj ! put in znorthloc the last 4 jlines of pt3d |
---|
| 2052 | ij = jj - nlcj + ijpj |
---|
| 2053 | znorthloc(:,ij,:) = pt3d(:,jj,:) |
---|
| 2054 | END DO |
---|
| 2055 | ! |
---|
| 2056 | ! ! Build in procs of ncomm_north the znorthgloio |
---|
| 2057 | itaille = jpi * jpk * ijpj |
---|
| 2058 | CALL MPI_ALLGATHER( znorthloc , itaille, MPI_DOUBLE_PRECISION, & |
---|
| 2059 | & znorthgloio, itaille, MPI_DOUBLE_PRECISION, ncomm_north, ierr ) |
---|
| 2060 | ! |
---|
| 2061 | ! ! recover the global north array |
---|
| 2062 | DO jr = 1, ndim_rank_north |
---|
| 2063 | iproc = nrank_north(jr) + 1 |
---|
| 2064 | ildi = nldit (iproc) |
---|
| 2065 | ilei = nleit (iproc) |
---|
| 2066 | iilb = nimppt(iproc) |
---|
| 2067 | DO jj = 1, 4 |
---|
| 2068 | DO ji = ildi, ilei |
---|
| 2069 | ztab(ji+iilb-1,jj,:) = znorthgloio(ji,jj,:,jr) |
---|
| 2070 | END DO |
---|
| 2071 | END DO |
---|
| 2072 | END DO |
---|
| 2073 | ! |
---|
| 2074 | CALL lbc_nfd( ztab, cd_type, psgn ) ! North fold boundary condition |
---|
| 2075 | ! |
---|
| 2076 | DO jj = nlcj-ijpj+1, nlcj ! Scatter back to pt3d |
---|
| 2077 | ij = jj - nlcj + ijpj |
---|
| 2078 | DO ji= 1, nlci |
---|
| 2079 | pt3d(ji,jj,:) = ztab(ji+nimpp-1,ij,:) |
---|
| 2080 | END DO |
---|
| 2081 | END DO |
---|
| 2082 | ! |
---|
| 2083 | END SUBROUTINE mpp_lbc_north_3d |
---|
[3] | 2084 | |
---|
| 2085 | |
---|
[1344] | 2086 | SUBROUTINE mpp_lbc_north_2d( pt2d, cd_type, psgn) |
---|
| 2087 | !!--------------------------------------------------------------------- |
---|
| 2088 | !! *** routine mpp_lbc_north_2d *** |
---|
| 2089 | !! |
---|
| 2090 | !! ** Purpose : Ensure proper north fold horizontal bondary condition |
---|
| 2091 | !! in mpp configuration in case of jpn1 > 1 (for 2d array ) |
---|
| 2092 | !! |
---|
| 2093 | !! ** Method : North fold condition and mpp with more than one proc |
---|
| 2094 | !! in i-direction require a specific treatment. We gather |
---|
| 2095 | !! the 4 northern lines of the global domain on 1 processor |
---|
| 2096 | !! and apply lbc north-fold on this sub array. Then we |
---|
| 2097 | !! scatter the north fold array back to the processors. |
---|
| 2098 | !! |
---|
| 2099 | !!---------------------------------------------------------------------- |
---|
| 2100 | REAL(wp), DIMENSION(jpi,jpj), INTENT(inout) :: pt2d ! 3D array on which the b.c. is applied |
---|
| 2101 | CHARACTER(len=1) , INTENT(in ) :: cd_type ! nature of pt3d grid-points |
---|
| 2102 | ! ! = T , U , V , F or W gridpoints |
---|
| 2103 | REAL(wp) , INTENT(in ) :: psgn ! = -1. the sign change across the north fold |
---|
| 2104 | !! ! = 1. , the sign is kept |
---|
| 2105 | INTEGER :: ji, jj, jr |
---|
| 2106 | INTEGER :: ierr, itaille, ildi, ilei, iilb |
---|
| 2107 | INTEGER :: ijpj, ijpjm1, ij, iproc |
---|
| 2108 | REAL(wp), DIMENSION(jpiglo,4) :: ztab |
---|
| 2109 | REAL(wp), DIMENSION(jpi ,4) :: znorthloc |
---|
| 2110 | REAL(wp), DIMENSION(jpi ,4,jpni) :: znorthgloio |
---|
| 2111 | !!---------------------------------------------------------------------- |
---|
| 2112 | ! |
---|
| 2113 | ijpj = 4 |
---|
| 2114 | ijpjm1 = 3 |
---|
[1874] | 2115 | ztab(:,:) = 0.e0 |
---|
[1344] | 2116 | ! |
---|
| 2117 | DO jj = nlcj-ijpj+1, nlcj ! put in znorthloc the last 4 jlines of pt2d |
---|
| 2118 | ij = jj - nlcj + ijpj |
---|
| 2119 | znorthloc(:,ij) = pt2d(:,jj) |
---|
| 2120 | END DO |
---|
[3] | 2121 | |
---|
[1344] | 2122 | ! ! Build in procs of ncomm_north the znorthgloio |
---|
| 2123 | itaille = jpi * ijpj |
---|
| 2124 | CALL MPI_ALLGATHER( znorthloc , itaille, MPI_DOUBLE_PRECISION, & |
---|
| 2125 | & znorthgloio, itaille, MPI_DOUBLE_PRECISION, ncomm_north, ierr ) |
---|
| 2126 | ! |
---|
| 2127 | DO jr = 1, ndim_rank_north ! recover the global north array |
---|
| 2128 | iproc = nrank_north(jr) + 1 |
---|
| 2129 | ildi=nldit (iproc) |
---|
| 2130 | ilei=nleit (iproc) |
---|
| 2131 | iilb=nimppt(iproc) |
---|
| 2132 | DO jj = 1, 4 |
---|
| 2133 | DO ji = ildi, ilei |
---|
| 2134 | ztab(ji+iilb-1,jj) = znorthgloio(ji,jj,jr) |
---|
[13] | 2135 | END DO |
---|
| 2136 | END DO |
---|
[1344] | 2137 | END DO |
---|
| 2138 | ! |
---|
| 2139 | CALL lbc_nfd( ztab, cd_type, psgn ) ! North fold boundary condition |
---|
| 2140 | ! |
---|
| 2141 | ! |
---|
| 2142 | DO jj = nlcj-ijpj+1, nlcj ! Scatter back to pt2d |
---|
[13] | 2143 | ij = jj - nlcj + ijpj |
---|
[1344] | 2144 | DO ji = 1, nlci |
---|
| 2145 | pt2d(ji,jj) = ztab(ji+nimpp-1,ij) |
---|
| 2146 | END DO |
---|
[13] | 2147 | END DO |
---|
[1344] | 2148 | ! |
---|
[13] | 2149 | END SUBROUTINE mpp_lbc_north_2d |
---|
[3] | 2150 | |
---|
| 2151 | |
---|
[1344] | 2152 | SUBROUTINE mpp_lbc_north_e( pt2d, cd_type, psgn) |
---|
| 2153 | !!--------------------------------------------------------------------- |
---|
| 2154 | !! *** routine mpp_lbc_north_2d *** |
---|
| 2155 | !! |
---|
| 2156 | !! ** Purpose : Ensure proper north fold horizontal bondary condition |
---|
| 2157 | !! in mpp configuration in case of jpn1 > 1 and for 2d |
---|
| 2158 | !! array with outer extra halo |
---|
| 2159 | !! |
---|
| 2160 | !! ** Method : North fold condition and mpp with more than one proc |
---|
| 2161 | !! in i-direction require a specific treatment. We gather |
---|
| 2162 | !! the 4+2*jpr2dj northern lines of the global domain on 1 |
---|
| 2163 | !! processor and apply lbc north-fold on this sub array. |
---|
| 2164 | !! Then we scatter the north fold array back to the processors. |
---|
| 2165 | !! |
---|
| 2166 | !!---------------------------------------------------------------------- |
---|
| 2167 | REAL(wp), DIMENSION(1-jpr2di:jpi+jpr2di,1-jpr2dj:jpj+jpr2dj), INTENT(inout) :: pt2d ! 2D array with extra halo |
---|
| 2168 | CHARACTER(len=1) , INTENT(in ) :: cd_type ! nature of pt3d grid-points |
---|
| 2169 | ! ! = T , U , V , F or W -points |
---|
| 2170 | REAL(wp) , INTENT(in ) :: psgn ! = -1. the sign change across the |
---|
| 2171 | !! ! north fold, = 1. otherwise |
---|
| 2172 | INTEGER :: ji, jj, jr |
---|
| 2173 | INTEGER :: ierr, itaille, ildi, ilei, iilb |
---|
| 2174 | INTEGER :: ijpj, ij, iproc |
---|
| 2175 | REAL(wp), DIMENSION(jpiglo,4+2*jpr2dj) :: ztab |
---|
| 2176 | REAL(wp), DIMENSION(jpi ,4+2*jpr2dj) :: znorthloc |
---|
| 2177 | REAL(wp), DIMENSION(jpi ,4+2*jpr2dj,jpni) :: znorthgloio |
---|
| 2178 | !!---------------------------------------------------------------------- |
---|
| 2179 | ! |
---|
| 2180 | ijpj=4 |
---|
[1874] | 2181 | ztab(:,:) = 0.e0 |
---|
[311] | 2182 | |
---|
[1344] | 2183 | ij=0 |
---|
| 2184 | ! put in znorthloc the last 4 jlines of pt2d |
---|
| 2185 | DO jj = nlcj - ijpj + 1 - jpr2dj, nlcj +jpr2dj |
---|
| 2186 | ij = ij + 1 |
---|
| 2187 | DO ji = 1, jpi |
---|
| 2188 | znorthloc(ji,ij)=pt2d(ji,jj) |
---|
| 2189 | END DO |
---|
| 2190 | END DO |
---|
| 2191 | ! |
---|
| 2192 | itaille = jpi * ( ijpj + 2 * jpr2dj ) |
---|
| 2193 | CALL MPI_ALLGATHER( znorthloc(1,1) , itaille, MPI_DOUBLE_PRECISION, & |
---|
| 2194 | & znorthgloio(1,1,1), itaille, MPI_DOUBLE_PRECISION, ncomm_north, ierr ) |
---|
| 2195 | ! |
---|
| 2196 | DO jr = 1, ndim_rank_north ! recover the global north array |
---|
| 2197 | iproc = nrank_north(jr) + 1 |
---|
| 2198 | ildi = nldit (iproc) |
---|
| 2199 | ilei = nleit (iproc) |
---|
| 2200 | iilb = nimppt(iproc) |
---|
| 2201 | DO jj = 1, ijpj+2*jpr2dj |
---|
| 2202 | DO ji = ildi, ilei |
---|
| 2203 | ztab(ji+iilb-1,jj) = znorthgloio(ji,jj,jr) |
---|
[311] | 2204 | END DO |
---|
[1344] | 2205 | END DO |
---|
| 2206 | END DO |
---|
[311] | 2207 | |
---|
| 2208 | |
---|
[1344] | 2209 | ! 2. North-Fold boundary conditions |
---|
| 2210 | ! ---------------------------------- |
---|
| 2211 | CALL lbc_nfd( ztab(:,:), cd_type, psgn, pr2dj = jpr2dj ) |
---|
[311] | 2212 | |
---|
[1344] | 2213 | ij = jpr2dj |
---|
| 2214 | !! Scatter back to pt2d |
---|
| 2215 | DO jj = nlcj - ijpj + 1 , nlcj +jpr2dj |
---|
| 2216 | ij = ij +1 |
---|
| 2217 | DO ji= 1, nlci |
---|
| 2218 | pt2d(ji,jj) = ztab(ji+nimpp-1,ij) |
---|
[311] | 2219 | END DO |
---|
| 2220 | END DO |
---|
[1344] | 2221 | ! |
---|
[311] | 2222 | END SUBROUTINE mpp_lbc_north_e |
---|
| 2223 | |
---|
[389] | 2224 | |
---|
[1344] | 2225 | SUBROUTINE mpi_init_opa( code ) |
---|
| 2226 | !!--------------------------------------------------------------------- |
---|
| 2227 | !! *** routine mpp_init.opa *** |
---|
| 2228 | !! |
---|
| 2229 | !! ** Purpose :: export and attach a MPI buffer for bsend |
---|
| 2230 | !! |
---|
| 2231 | !! ** Method :: define buffer size in namelist, if 0 no buffer attachment |
---|
| 2232 | !! but classical mpi_init |
---|
| 2233 | !! |
---|
| 2234 | !! History :: 01/11 :: IDRIS initial version for IBM only |
---|
| 2235 | !! 08/04 :: R. Benshila, generalisation |
---|
| 2236 | !!--------------------------------------------------------------------- |
---|
[1237] | 2237 | INTEGER :: code, ierr |
---|
[532] | 2238 | LOGICAL :: mpi_was_called |
---|
[1344] | 2239 | !!--------------------------------------------------------------------- |
---|
| 2240 | ! |
---|
| 2241 | CALL mpi_initialized( mpi_was_called, code ) ! MPI initialization |
---|
[532] | 2242 | IF ( code /= MPI_SUCCESS ) THEN |
---|
[1344] | 2243 | CALL ctl_stop( ' lib_mpp: Error in routine mpi_initialized' ) |
---|
| 2244 | CALL mpi_abort( mpi_comm_world, code, ierr ) |
---|
[532] | 2245 | ENDIF |
---|
[1344] | 2246 | ! |
---|
| 2247 | IF( .NOT. mpi_was_called ) THEN |
---|
| 2248 | CALL mpi_init( code ) |
---|
| 2249 | CALL mpi_comm_dup( mpi_comm_world, mpi_comm_opa, code ) |
---|
[532] | 2250 | IF ( code /= MPI_SUCCESS ) THEN |
---|
| 2251 | CALL ctl_stop( ' lib_mpp: Error in routine mpi_comm_dup' ) |
---|
| 2252 | CALL mpi_abort( mpi_comm_world, code, ierr ) |
---|
| 2253 | ENDIF |
---|
| 2254 | ENDIF |
---|
[1344] | 2255 | ! |
---|
[897] | 2256 | IF( nn_buffer > 0 ) THEN |
---|
| 2257 | IF ( lwp ) WRITE(numout,*) 'mpi_bsend, buffer allocation of : ', nn_buffer |
---|
| 2258 | ! Buffer allocation and attachment |
---|
[1344] | 2259 | ALLOCATE( tampon(nn_buffer) ) |
---|
| 2260 | CALL mpi_buffer_attach( tampon, nn_buffer,code ) |
---|
[897] | 2261 | ENDIF |
---|
[1344] | 2262 | ! |
---|
[13] | 2263 | END SUBROUTINE mpi_init_opa |
---|
[3] | 2264 | |
---|
[13] | 2265 | #else |
---|
| 2266 | !!---------------------------------------------------------------------- |
---|
| 2267 | !! Default case: Dummy module share memory computing |
---|
| 2268 | !!---------------------------------------------------------------------- |
---|
| 2269 | INTERFACE mpp_sum |
---|
| 2270 | MODULE PROCEDURE mpp_sum_a2s, mpp_sum_as, mpp_sum_ai, mpp_sum_s, mpp_sum_i |
---|
| 2271 | END INTERFACE |
---|
| 2272 | INTERFACE mpp_max |
---|
[681] | 2273 | MODULE PROCEDURE mppmax_a_int, mppmax_int, mppmax_a_real, mppmax_real |
---|
[13] | 2274 | END INTERFACE |
---|
| 2275 | INTERFACE mpp_min |
---|
| 2276 | MODULE PROCEDURE mppmin_a_int, mppmin_int, mppmin_a_real, mppmin_real |
---|
| 2277 | END INTERFACE |
---|
| 2278 | INTERFACE mppobc |
---|
| 2279 | MODULE PROCEDURE mppobc_1d, mppobc_2d, mppobc_3d, mppobc_4d |
---|
| 2280 | END INTERFACE |
---|
[1344] | 2281 | INTERFACE mpp_minloc |
---|
| 2282 | MODULE PROCEDURE mpp_minloc2d ,mpp_minloc3d |
---|
| 2283 | END INTERFACE |
---|
| 2284 | INTERFACE mpp_maxloc |
---|
| 2285 | MODULE PROCEDURE mpp_maxloc2d ,mpp_maxloc3d |
---|
| 2286 | END INTERFACE |
---|
[3] | 2287 | |
---|
[181] | 2288 | |
---|
[13] | 2289 | LOGICAL, PUBLIC, PARAMETER :: lk_mpp = .FALSE. !: mpp flag |
---|
[869] | 2290 | INTEGER :: ncomm_ice |
---|
[3] | 2291 | |
---|
[13] | 2292 | CONTAINS |
---|
[3] | 2293 | |
---|
[1579] | 2294 | FUNCTION mynode( ldtxt, localComm ) RESULT (function_value) |
---|
| 2295 | CHARACTER(len=*),DIMENSION(:), INTENT( out) :: ldtxt |
---|
| 2296 | INTEGER, OPTIONAL , INTENT(in ) :: localComm |
---|
[1559] | 2297 | IF( PRESENT( localComm ) .OR. .NOT.PRESENT( localComm ) ) function_value = 0 |
---|
[1579] | 2298 | IF( .FALSE. ) ldtxt(:) = 'never done' |
---|
[13] | 2299 | END FUNCTION mynode |
---|
[3] | 2300 | |
---|
[13] | 2301 | SUBROUTINE mppsync ! Dummy routine |
---|
| 2302 | END SUBROUTINE mppsync |
---|
[3] | 2303 | |
---|
[869] | 2304 | SUBROUTINE mpp_sum_as( parr, kdim, kcom ) ! Dummy routine |
---|
[13] | 2305 | REAL , DIMENSION(:) :: parr |
---|
| 2306 | INTEGER :: kdim |
---|
[869] | 2307 | INTEGER, OPTIONAL :: kcom |
---|
| 2308 | WRITE(*,*) 'mpp_sum_as: You should not have seen this print! error?', kdim, parr(1), kcom |
---|
[13] | 2309 | END SUBROUTINE mpp_sum_as |
---|
[3] | 2310 | |
---|
[869] | 2311 | SUBROUTINE mpp_sum_a2s( parr, kdim, kcom ) ! Dummy routine |
---|
[13] | 2312 | REAL , DIMENSION(:,:) :: parr |
---|
| 2313 | INTEGER :: kdim |
---|
[869] | 2314 | INTEGER, OPTIONAL :: kcom |
---|
| 2315 | WRITE(*,*) 'mpp_sum_a2s: You should not have seen this print! error?', kdim, parr(1,1), kcom |
---|
[13] | 2316 | END SUBROUTINE mpp_sum_a2s |
---|
[3] | 2317 | |
---|
[869] | 2318 | SUBROUTINE mpp_sum_ai( karr, kdim, kcom ) ! Dummy routine |
---|
[13] | 2319 | INTEGER, DIMENSION(:) :: karr |
---|
| 2320 | INTEGER :: kdim |
---|
[869] | 2321 | INTEGER, OPTIONAL :: kcom |
---|
| 2322 | WRITE(*,*) 'mpp_sum_ai: You should not have seen this print! error?', kdim, karr(1), kcom |
---|
[13] | 2323 | END SUBROUTINE mpp_sum_ai |
---|
[3] | 2324 | |
---|
[869] | 2325 | SUBROUTINE mpp_sum_s( psca, kcom ) ! Dummy routine |
---|
[13] | 2326 | REAL :: psca |
---|
[869] | 2327 | INTEGER, OPTIONAL :: kcom |
---|
| 2328 | WRITE(*,*) 'mpp_sum_s: You should not have seen this print! error?', psca, kcom |
---|
[13] | 2329 | END SUBROUTINE mpp_sum_s |
---|
| 2330 | |
---|
[869] | 2331 | SUBROUTINE mpp_sum_i( kint, kcom ) ! Dummy routine |
---|
[13] | 2332 | integer :: kint |
---|
[869] | 2333 | INTEGER, OPTIONAL :: kcom |
---|
| 2334 | WRITE(*,*) 'mpp_sum_i: You should not have seen this print! error?', kint, kcom |
---|
[13] | 2335 | END SUBROUTINE mpp_sum_i |
---|
| 2336 | |
---|
[869] | 2337 | SUBROUTINE mppmax_a_real( parr, kdim, kcom ) |
---|
[13] | 2338 | REAL , DIMENSION(:) :: parr |
---|
| 2339 | INTEGER :: kdim |
---|
[869] | 2340 | INTEGER, OPTIONAL :: kcom |
---|
| 2341 | WRITE(*,*) 'mppmax_a_real: You should not have seen this print! error?', kdim, parr(1), kcom |
---|
[13] | 2342 | END SUBROUTINE mppmax_a_real |
---|
| 2343 | |
---|
[869] | 2344 | SUBROUTINE mppmax_real( psca, kcom ) |
---|
[13] | 2345 | REAL :: psca |
---|
[869] | 2346 | INTEGER, OPTIONAL :: kcom |
---|
| 2347 | WRITE(*,*) 'mppmax_real: You should not have seen this print! error?', psca, kcom |
---|
[13] | 2348 | END SUBROUTINE mppmax_real |
---|
| 2349 | |
---|
[869] | 2350 | SUBROUTINE mppmin_a_real( parr, kdim, kcom ) |
---|
[13] | 2351 | REAL , DIMENSION(:) :: parr |
---|
| 2352 | INTEGER :: kdim |
---|
[869] | 2353 | INTEGER, OPTIONAL :: kcom |
---|
| 2354 | WRITE(*,*) 'mppmin_a_real: You should not have seen this print! error?', kdim, parr(1), kcom |
---|
[13] | 2355 | END SUBROUTINE mppmin_a_real |
---|
| 2356 | |
---|
[869] | 2357 | SUBROUTINE mppmin_real( psca, kcom ) |
---|
[13] | 2358 | REAL :: psca |
---|
[869] | 2359 | INTEGER, OPTIONAL :: kcom |
---|
| 2360 | WRITE(*,*) 'mppmin_real: You should not have seen this print! error?', psca, kcom |
---|
[13] | 2361 | END SUBROUTINE mppmin_real |
---|
| 2362 | |
---|
[869] | 2363 | SUBROUTINE mppmax_a_int( karr, kdim ,kcom) |
---|
[681] | 2364 | INTEGER, DIMENSION(:) :: karr |
---|
| 2365 | INTEGER :: kdim |
---|
[869] | 2366 | INTEGER, OPTIONAL :: kcom |
---|
[888] | 2367 | WRITE(*,*) 'mppmax_a_int: You should not have seen this print! error?', kdim, karr(1), kcom |
---|
[681] | 2368 | END SUBROUTINE mppmax_a_int |
---|
| 2369 | |
---|
[869] | 2370 | SUBROUTINE mppmax_int( kint, kcom) |
---|
[681] | 2371 | INTEGER :: kint |
---|
[869] | 2372 | INTEGER, OPTIONAL :: kcom |
---|
| 2373 | WRITE(*,*) 'mppmax_int: You should not have seen this print! error?', kint, kcom |
---|
[681] | 2374 | END SUBROUTINE mppmax_int |
---|
| 2375 | |
---|
[869] | 2376 | SUBROUTINE mppmin_a_int( karr, kdim, kcom ) |
---|
[13] | 2377 | INTEGER, DIMENSION(:) :: karr |
---|
| 2378 | INTEGER :: kdim |
---|
[869] | 2379 | INTEGER, OPTIONAL :: kcom |
---|
| 2380 | WRITE(*,*) 'mppmin_a_int: You should not have seen this print! error?', kdim, karr(1), kcom |
---|
[13] | 2381 | END SUBROUTINE mppmin_a_int |
---|
| 2382 | |
---|
[869] | 2383 | SUBROUTINE mppmin_int( kint, kcom ) |
---|
[13] | 2384 | INTEGER :: kint |
---|
[869] | 2385 | INTEGER, OPTIONAL :: kcom |
---|
| 2386 | WRITE(*,*) 'mppmin_int: You should not have seen this print! error?', kint, kcom |
---|
[13] | 2387 | END SUBROUTINE mppmin_int |
---|
| 2388 | |
---|
| 2389 | SUBROUTINE mppobc_1d( parr, kd1, kd2, kl, kk, ktype, kij ) |
---|
[1344] | 2390 | INTEGER :: kd1, kd2, kl , kk, ktype, kij |
---|
| 2391 | REAL, DIMENSION(:) :: parr ! variable array |
---|
| 2392 | WRITE(*,*) 'mppobc: You should not have seen this print! error?', parr(1), kd1, kd2, kl, kk, ktype, kij |
---|
[13] | 2393 | END SUBROUTINE mppobc_1d |
---|
| 2394 | |
---|
| 2395 | SUBROUTINE mppobc_2d( parr, kd1, kd2, kl, kk, ktype, kij ) |
---|
[1344] | 2396 | INTEGER :: kd1, kd2, kl , kk, ktype, kij |
---|
| 2397 | REAL, DIMENSION(:,:) :: parr ! variable array |
---|
| 2398 | WRITE(*,*) 'mppobc: You should not have seen this print! error?', parr(1,1), kd1, kd2, kl, kk, ktype, kij |
---|
[13] | 2399 | END SUBROUTINE mppobc_2d |
---|
| 2400 | |
---|
| 2401 | SUBROUTINE mppobc_3d( parr, kd1, kd2, kl, kk, ktype, kij ) |
---|
[1344] | 2402 | INTEGER :: kd1, kd2, kl , kk, ktype, kij |
---|
| 2403 | REAL, DIMENSION(:,:,:) :: parr ! variable array |
---|
| 2404 | WRITE(*,*) 'mppobc: You should not have seen this print! error?', parr(1,1,1), kd1, kd2, kl, kk, ktype, kij |
---|
[13] | 2405 | END SUBROUTINE mppobc_3d |
---|
| 2406 | |
---|
| 2407 | SUBROUTINE mppobc_4d( parr, kd1, kd2, kl, kk, ktype, kij ) |
---|
[1344] | 2408 | INTEGER :: kd1, kd2, kl , kk, ktype, kij |
---|
| 2409 | REAL, DIMENSION(:,:,:,:) :: parr ! variable array |
---|
| 2410 | WRITE(*,*) 'mppobc: You should not have seen this print! error?', parr(1,1,1,1), kd1, kd2, kl, kk, ktype, kij |
---|
[13] | 2411 | END SUBROUTINE mppobc_4d |
---|
| 2412 | |
---|
[1344] | 2413 | SUBROUTINE mpp_minloc2d( ptab, pmask, pmin, ki, kj ) |
---|
[181] | 2414 | REAL :: pmin |
---|
| 2415 | REAL , DIMENSION (:,:) :: ptab, pmask |
---|
| 2416 | INTEGER :: ki, kj |
---|
[1528] | 2417 | WRITE(*,*) 'mpp_minloc2d: You should not have seen this print! error?', pmin, ki, kj, ptab(1,1), pmask(1,1) |
---|
[181] | 2418 | END SUBROUTINE mpp_minloc2d |
---|
| 2419 | |
---|
[1344] | 2420 | SUBROUTINE mpp_minloc3d( ptab, pmask, pmin, ki, kj, kk ) |
---|
[181] | 2421 | REAL :: pmin |
---|
| 2422 | REAL , DIMENSION (:,:,:) :: ptab, pmask |
---|
| 2423 | INTEGER :: ki, kj, kk |
---|
[1528] | 2424 | WRITE(*,*) 'mpp_minloc3d: You should not have seen this print! error?', pmin, ki, kj, kk, ptab(1,1,1), pmask(1,1,1) |
---|
[181] | 2425 | END SUBROUTINE mpp_minloc3d |
---|
| 2426 | |
---|
[1344] | 2427 | SUBROUTINE mpp_maxloc2d( ptab, pmask, pmax, ki, kj ) |
---|
[181] | 2428 | REAL :: pmax |
---|
| 2429 | REAL , DIMENSION (:,:) :: ptab, pmask |
---|
| 2430 | INTEGER :: ki, kj |
---|
[1528] | 2431 | WRITE(*,*) 'mpp_maxloc2d: You should not have seen this print! error?', pmax, ki, kj, ptab(1,1), pmask(1,1) |
---|
[181] | 2432 | END SUBROUTINE mpp_maxloc2d |
---|
| 2433 | |
---|
[1344] | 2434 | SUBROUTINE mpp_maxloc3d( ptab, pmask, pmax, ki, kj, kk ) |
---|
[181] | 2435 | REAL :: pmax |
---|
| 2436 | REAL , DIMENSION (:,:,:) :: ptab, pmask |
---|
| 2437 | INTEGER :: ki, kj, kk |
---|
[1528] | 2438 | WRITE(*,*) 'mpp_maxloc3d: You should not have seen this print! error?', pmax, ki, kj, kk, ptab(1,1,1), pmask(1,1,1) |
---|
[181] | 2439 | END SUBROUTINE mpp_maxloc3d |
---|
| 2440 | |
---|
[51] | 2441 | SUBROUTINE mppstop |
---|
| 2442 | WRITE(*,*) 'mppstop: You should not have seen this print! error?' |
---|
| 2443 | END SUBROUTINE mppstop |
---|
| 2444 | |
---|
[1344] | 2445 | SUBROUTINE mpp_ini_ice( kcom ) |
---|
[888] | 2446 | INTEGER :: kcom |
---|
[1344] | 2447 | WRITE(*,*) 'mpp_ini_ice: You should not have seen this print! error?', kcom |
---|
[888] | 2448 | END SUBROUTINE mpp_ini_ice |
---|
[869] | 2449 | |
---|
[1345] | 2450 | SUBROUTINE mpp_ini_znl |
---|
| 2451 | WRITE(*,*) 'mpp_ini_znl: You should not have seen this print! error?' |
---|
| 2452 | END SUBROUTINE mpp_ini_znl |
---|
| 2453 | |
---|
[1344] | 2454 | SUBROUTINE mpp_comm_free( kcom ) |
---|
[869] | 2455 | INTEGER :: kcom |
---|
[1344] | 2456 | WRITE(*,*) 'mpp_comm_free: You should not have seen this print! error?', kcom |
---|
[869] | 2457 | END SUBROUTINE mpp_comm_free |
---|
| 2458 | |
---|
[3] | 2459 | #endif |
---|
[13] | 2460 | !!---------------------------------------------------------------------- |
---|
[3] | 2461 | END MODULE lib_mpp |
---|