[331] | 1 | MODULE prtctl_trc |
---|
| 2 | !!============================================================================== |
---|
| 3 | !! *** MODULE prtctl *** |
---|
| 4 | !! Ocean system : print all SUM trends for each processor domain |
---|
| 5 | !!============================================================================== |
---|
| 6 | #if defined key_passivetrc |
---|
| 7 | |
---|
| 8 | USE par_trc_trp |
---|
| 9 | USE oce_trc ! ocean space and time domain variables |
---|
| 10 | USE in_out_manager ! I/O manager |
---|
| 11 | USE lib_mpp ! distributed memory computing |
---|
| 12 | |
---|
| 13 | IMPLICIT NONE |
---|
| 14 | PRIVATE |
---|
| 15 | |
---|
| 16 | !! * Module declaration |
---|
| 17 | INTEGER, DIMENSION(:), ALLOCATABLE, SAVE :: & !: |
---|
| 18 | nlditl , nldjtl , & !: first, last indoor index for each i-domain |
---|
| 19 | nleitl , nlejtl , & !: first, last indoor index for each j-domain |
---|
| 20 | nimpptl, njmpptl, & !: i-, j-indexes for each processor |
---|
| 21 | nlcitl , nlcjtl , & !: dimensions of every subdomain |
---|
| 22 | ibonitl, ibonjtl |
---|
| 23 | |
---|
| 24 | REAL(wp), DIMENSION(:,:), ALLOCATABLE :: & !: |
---|
| 25 | tra_ctl !: previous trend values |
---|
| 26 | |
---|
| 27 | !! * Routine accessibility |
---|
| 28 | PUBLIC prt_ctl_trc ! called by all subroutines |
---|
| 29 | PUBLIC prt_ctl_trc_info ! |
---|
| 30 | PUBLIC prt_ctl_trc_init ! called by opa.F90 |
---|
| 31 | !!---------------------------------------------------------------------- |
---|
| 32 | !! OPA 9.0 , LOCEAN-IPSL (2005) |
---|
| 33 | !! $Header$ |
---|
| 34 | !! This software is governed by the CeCILL licence see modipsl/doc/NEMO_CeCILL.txt |
---|
| 35 | !!---------------------------------------------------------------------- |
---|
| 36 | |
---|
| 37 | |
---|
| 38 | CONTAINS |
---|
| 39 | |
---|
| 40 | SUBROUTINE prt_ctl_trc (tab4d, mask, clinfo, ovlap, kdim, clinfo2) |
---|
| 41 | !!---------------------------------------------------------------------- |
---|
| 42 | !! *** ROUTINE prt_ctl *** |
---|
| 43 | !! |
---|
| 44 | !! ** Purpose : - print sum control 3D arrays over the same area |
---|
| 45 | !! in mono and mpp case. This way can be usefull when |
---|
| 46 | !! debugging a new parametrization in mono or mpp. |
---|
| 47 | !! |
---|
| 48 | !! ** Method : 2 possibilities exist when setting the ln_ctl parameter to |
---|
| 49 | !! .true. in the ocean namelist: |
---|
| 50 | !! - to debug a MPI run .vs. a mono-processor one; |
---|
| 51 | !! the control print will be done over each sub-domain. |
---|
| 52 | !! The nictl[se] and njctl[se] parameters in the namelist must |
---|
| 53 | !! be set to zero and [ij]splt to the corresponding splitted |
---|
| 54 | !! domain in MPI along respectively i-, j- directions. |
---|
| 55 | !! - to debug a mono-processor run over the whole domain/a specific area; |
---|
| 56 | !! in the first case the nictl[se] and njctl[se] parameters must be set |
---|
| 57 | !! to zero else to the indices of the area to be controled. In both cases |
---|
| 58 | !! isplt and jsplt must be set to 1. |
---|
| 59 | !! - All arguments of the above calling sequence are optional so their |
---|
| 60 | !! name must be explicitly typed if used. For instance if the mask |
---|
| 61 | !! array tmask(:,:,:) must be passed through the prt_ctl subroutine, |
---|
| 62 | !! it must looks like: CALL prt_ctl(mask=tmask). |
---|
| 63 | !! |
---|
| 64 | !! tab4d : 4D array |
---|
| 65 | !! mask : mask (3D) to apply to the tab4d array |
---|
| 66 | !! clinfo : information about the tab3d array |
---|
| 67 | !! ovlap : overlap value |
---|
| 68 | !! kdim : k- direction for 4D arrays |
---|
| 69 | !! |
---|
| 70 | !! History : |
---|
| 71 | !! 9.0 ! 05-07 (C. Talandier) original code |
---|
| 72 | !! ! 05-10 (C. Ethe ) adapted to passive tracer |
---|
| 73 | !!---------------------------------------------------------------------- |
---|
| 74 | !! * Arguments |
---|
| 75 | REAL(wp), DIMENSION(:,:,:,:), INTENT(in), OPTIONAL :: tab4d |
---|
| 76 | REAL(wp), DIMENSION(:,:,:), INTENT(in), OPTIONAL :: mask |
---|
| 77 | CHARACTER (len=*), DIMENSION(:), INTENT(in), OPTIONAL :: clinfo |
---|
| 78 | CHARACTER (len=*), INTENT(in), OPTIONAL :: clinfo2 |
---|
| 79 | INTEGER, INTENT(in), OPTIONAL :: ovlap |
---|
| 80 | INTEGER, INTENT(in), OPTIONAL :: kdim |
---|
| 81 | |
---|
| 82 | !! * Local declarations |
---|
| 83 | INTEGER :: overlap, numid, jn, js, sind, eind, kdir |
---|
| 84 | REAL(wp) :: zsum, zvctl |
---|
| 85 | REAL(wp), DIMENSION(jpi,jpj,jpk) :: zmask, ztab3d |
---|
| 86 | CHARACTER (len=20), DIMENSION(jptra) :: cl |
---|
| 87 | CHARACTER (len=10) :: cl2 |
---|
| 88 | !!---------------------------------------------------------------------- |
---|
| 89 | |
---|
| 90 | ! Arrays, scalars initialization |
---|
| 91 | overlap = 0 |
---|
| 92 | kdir = jpkm1 |
---|
| 93 | zsum = 0.e0 |
---|
| 94 | zvctl = 0.e0 |
---|
| 95 | cl(:) = '' |
---|
| 96 | cl2 = '' |
---|
| 97 | ztab3d(:,:,:) = 0.e0 |
---|
| 98 | zmask (:,:,:) = 1.e0 |
---|
| 99 | |
---|
| 100 | ! Control of optional arguments |
---|
| 101 | |
---|
| 102 | IF( PRESENT(ovlap) ) overlap = ovlap |
---|
| 103 | IF( PRESENT(kdim) ) kdir = kdim |
---|
| 104 | IF( PRESENT(clinfo ) ) cl(:) = clinfo(:) |
---|
| 105 | IF( PRESENT(clinfo2) ) cl2 = clinfo2 |
---|
| 106 | IF( PRESENT(mask) ) zmask (:,:,:) = mask(:,:,:) |
---|
| 107 | |
---|
| 108 | IF( lk_mpp ) THEN |
---|
| 109 | ! processor number |
---|
| 110 | sind = narea |
---|
| 111 | eind = narea |
---|
| 112 | ELSE |
---|
| 113 | ! processors total number |
---|
| 114 | sind = 1 |
---|
| 115 | eind = ijsplt |
---|
| 116 | ENDIF |
---|
| 117 | |
---|
| 118 | ! Loop over each sub-domain, i.e. the total number of processors ijsplt |
---|
| 119 | DO js = sind, eind |
---|
| 120 | |
---|
| 121 | numid = 90 + js |
---|
| 122 | |
---|
| 123 | ! Set indices for the SUM control |
---|
| 124 | IF( .NOT. lsp_area ) THEN |
---|
| 125 | IF (lk_mpp ) THEN |
---|
| 126 | nictls = MAX( 1, nlditl(js) - overlap ) |
---|
| 127 | nictle = nleitl(js) + overlap * MIN( 1, nlcitl(js) - nleitl(js)) |
---|
| 128 | njctls = MAX( 1, nldjtl(js) - overlap ) |
---|
| 129 | njctle = nlejtl(js) + overlap * MIN( 1, nlcjtl(js) - nlejtl(js)) |
---|
| 130 | ! Do not take into account the bound of the domain |
---|
[426] | 131 | IF( ibonitl(js) == -1 .OR. ibonitl(js) == 2 ) nictls = MAX(2, nictls) |
---|
| 132 | IF( ibonitl(js) == 1 .OR. ibonitl(js) == 2 ) nictle = MIN(nictle, nleitl(js) - 1) |
---|
| 133 | IF( ibonjtl(js) == -1 .OR. ibonjtl(js) == 2 ) njctls = MAX(2, njctls) |
---|
| 134 | IF( ibonjtl(js) == 1 .OR. ibonjtl(js) == 2 ) njctle = MIN(njctle, nlejtl(js) - 1) |
---|
[331] | 135 | ELSE |
---|
| 136 | nictls = MAX( 1, nimpptl(js) + nlditl(js) - 1 - overlap ) |
---|
| 137 | nictle = nimpptl(js) + nleitl(js) - 1 + overlap * MIN( 1, nlcitl(js) - nleitl(js) ) |
---|
| 138 | njctls = MAX( 1, njmpptl(js) + nldjtl(js) - 1 - overlap ) |
---|
| 139 | njctle = njmpptl(js) + nlejtl(js) - 1 + overlap * MIN( 1, nlcjtl(js) - nlejtl(js) ) |
---|
| 140 | ! Do not take into account the bound of the domain |
---|
[426] | 141 | IF( ibonitl(js) == -1 .OR. ibonitl(js) == 2 ) nictls = MAX(2, nictls) |
---|
| 142 | IF( ibonjtl(js) == -1 .OR. ibonjtl(js) == 2 ) njctls = MAX(2, njctls) |
---|
| 143 | IF( ibonitl(js) == 1 .OR. ibonitl(js) == 2 ) nictle = MIN(nictle, nimpptl(js) + nleitl(js) - 2) |
---|
| 144 | IF( ibonjtl(js) == 1 .OR. ibonjtl(js) == 2 ) njctle = MIN(njctle, njmpptl(js) + nlejtl(js) - 2) |
---|
[331] | 145 | ENDIF |
---|
| 146 | ENDIF |
---|
| 147 | |
---|
| 148 | IF( PRESENT(clinfo2) ) THEN |
---|
| 149 | DO jn = 1, jptra |
---|
| 150 | zvctl = tra_ctl(jn,js) |
---|
| 151 | ztab3d(:,:,:) = tab4d(:,:,:,jn) |
---|
| 152 | zsum = SUM( ztab3d(nictls:nictle,njctls:njctle,1:kdir) & |
---|
| 153 | & *zmask(nictls:nictle,njctls:njctle,1:kdir) ) |
---|
| 154 | WRITE(numid,FMT="(3x,a,' : ',D23.16)") cl(jn), zsum-zvctl |
---|
| 155 | tra_ctl(jn,js) = zsum |
---|
| 156 | ENDDO |
---|
| 157 | ELSE |
---|
| 158 | DO jn = 1, jptra |
---|
| 159 | ztab3d(:,:,:) = tab4d(:,:,:,jn) |
---|
| 160 | zsum = SUM( ztab3d(nictls:nictle,njctls:njctle,1:kdir) & |
---|
| 161 | & * zmask(nictls:nictle,njctls:njctle,1:kdir) ) |
---|
| 162 | WRITE(numid,FMT="(3x,a,' : ',D23.16)") cl(jn), zsum |
---|
| 163 | END DO |
---|
| 164 | ENDIF |
---|
| 165 | |
---|
| 166 | |
---|
| 167 | ENDDO |
---|
| 168 | |
---|
| 169 | END SUBROUTINE prt_ctl_trc |
---|
| 170 | |
---|
| 171 | SUBROUTINE prt_ctl_trc_info (clinfo) |
---|
| 172 | !!---------------------------------------------------------------------- |
---|
| 173 | !! *** ROUTINE prt_ctl_trc_info *** |
---|
| 174 | !! |
---|
| 175 | !! ** Purpose : - print information without any computation |
---|
| 176 | !! |
---|
| 177 | !! ** Action : - input arguments |
---|
| 178 | !! clinfo : information to print |
---|
| 179 | !! |
---|
| 180 | !! History : |
---|
| 181 | !! 9.0 ! 05-07 (C. Talandier) original code |
---|
| 182 | !!---------------------------------------------------------------------- |
---|
| 183 | !! * Arguments |
---|
| 184 | CHARACTER (len=*), INTENT(in) :: clinfo |
---|
| 185 | |
---|
| 186 | !! * Local declarations |
---|
| 187 | INTEGER :: numid, js, sind, eind |
---|
| 188 | !!---------------------------------------------------------------------- |
---|
| 189 | |
---|
| 190 | IF( lk_mpp ) THEN |
---|
| 191 | ! processor number |
---|
| 192 | sind = narea |
---|
| 193 | eind = narea |
---|
| 194 | ELSE |
---|
| 195 | ! total number of processors |
---|
| 196 | sind = 1 |
---|
| 197 | eind = ijsplt |
---|
| 198 | ENDIF |
---|
| 199 | |
---|
| 200 | ! Loop over each sub-domain, i.e. number of processors ijsplt |
---|
| 201 | DO js = sind, eind |
---|
| 202 | numid = 90 + js |
---|
| 203 | WRITE(numid,*)clinfo |
---|
| 204 | ENDDO |
---|
| 205 | |
---|
| 206 | |
---|
| 207 | END SUBROUTINE prt_ctl_trc_info |
---|
| 208 | |
---|
| 209 | SUBROUTINE prt_ctl_trc_init |
---|
| 210 | !!---------------------------------------------------------------------- |
---|
| 211 | !! *** ROUTINE prt_ctl_trc_init *** |
---|
| 212 | !! |
---|
| 213 | !! ** Purpose : open ASCII files & compute indices |
---|
| 214 | !! |
---|
| 215 | !! History : |
---|
| 216 | !! 9.0 ! 05-07 (C. Talandier) original code |
---|
| 217 | !! ! 05-10 (C. Ethe ) adapted to passive tracer |
---|
| 218 | !!---------------------------------------------------------------------- |
---|
| 219 | !! * Local declarations |
---|
| 220 | INTEGER :: js, numid, sind, eind |
---|
| 221 | CHARACTER (len=31) :: clfile_out |
---|
| 222 | CHARACTER (len=27) :: clb_name |
---|
| 223 | CHARACTER (len=19) :: cl_run |
---|
| 224 | !!---------------------------------------------------------------------- |
---|
| 225 | |
---|
| 226 | ! Allocate arrays |
---|
| 227 | ALLOCATE(nlditl (ijsplt)) |
---|
| 228 | ALLOCATE(nldjtl (ijsplt)) |
---|
| 229 | ALLOCATE(nleitl (ijsplt)) |
---|
| 230 | ALLOCATE(nlejtl (ijsplt)) |
---|
| 231 | ALLOCATE(nimpptl(ijsplt)) |
---|
| 232 | ALLOCATE(njmpptl(ijsplt)) |
---|
| 233 | ALLOCATE(nlcitl (ijsplt)) |
---|
| 234 | ALLOCATE(nlcjtl (ijsplt)) |
---|
| 235 | ALLOCATE(tra_ctl(jptra,ijsplt)) |
---|
| 236 | ALLOCATE(ibonitl(ijsplt)) |
---|
| 237 | ALLOCATE(ibonjtl(ijsplt)) |
---|
| 238 | |
---|
| 239 | ! Initialization |
---|
| 240 | tra_ctl (:,:)=0.e0 |
---|
| 241 | |
---|
| 242 | IF( lk_mpp ) THEN |
---|
| 243 | sind = narea |
---|
| 244 | eind = narea |
---|
| 245 | clb_name = "('mpp.top.output_',I3.3)" |
---|
| 246 | cl_run = 'MULTI processor run' |
---|
| 247 | ! use indices for each area computed by mpp_init subroutine |
---|
| 248 | nlditl(:) = nldit(:) |
---|
| 249 | nleitl(:) = nleit(:) |
---|
| 250 | nldjtl(:) = nldjt(:) |
---|
| 251 | nlejtl(:) = nlejt(:) |
---|
| 252 | ! |
---|
| 253 | nimpptl(:) = nimppt(:) |
---|
| 254 | njmpptl(:) = njmppt(:) |
---|
| 255 | ! |
---|
| 256 | nlcitl(:) = nlcit(:) |
---|
| 257 | nlcjtl(:) = nlcjt(:) |
---|
| 258 | ! |
---|
| 259 | ibonitl(:) = ibonit(:) |
---|
| 260 | ibonjtl(:) = ibonjt(:) |
---|
| 261 | ELSE |
---|
| 262 | sind = 1 |
---|
| 263 | eind = ijsplt |
---|
| 264 | clb_name = "('mono.top.output_',I3.3)" |
---|
| 265 | cl_run = 'MONO processor run ' |
---|
| 266 | ! compute indices for each area as done in mpp_init subroutine |
---|
| 267 | CALL sub_dom |
---|
| 268 | ENDIF |
---|
| 269 | |
---|
| 270 | DO js = sind, eind |
---|
| 271 | numid = 90 + js |
---|
| 272 | WRITE(clfile_out,FMT=clb_name) js-1 |
---|
| 273 | OPEN ( UNIT=numid, FILE=TRIM(clfile_out),FORM='FORMATTED' ) |
---|
| 274 | WRITE(numid,*) |
---|
| 275 | WRITE(numid,*) ' L O D Y C - I P S L' |
---|
| 276 | WRITE(numid,*) ' O P A model' |
---|
| 277 | WRITE(numid,*) ' Ocean General Circulation Model' |
---|
| 278 | WRITE(numid,*) ' version OPA 9.0 (2005) ' |
---|
| 279 | WRITE(numid,*) |
---|
| 280 | WRITE(numid,*) ' PROC number: ', js |
---|
| 281 | WRITE(numid,*) |
---|
| 282 | WRITE(numid,FMT="(19x,a20)")cl_run |
---|
| 283 | |
---|
| 284 | ! Print the SUM control indices |
---|
| 285 | IF( .NOT. lsp_area ) THEN |
---|
| 286 | IF ( lk_mpp ) THEN |
---|
| 287 | nictls = nlditl(js) |
---|
| 288 | nictle = nleitl(js) |
---|
| 289 | njctls = nldjtl(js) |
---|
| 290 | njctle = nlejtl(js) |
---|
| 291 | ELSE |
---|
| 292 | nictls = nimpptl(js) + nlditl(js) - 1 |
---|
| 293 | nictle = nimpptl(js) + nleitl(js) - 1 |
---|
| 294 | njctls = njmpptl(js) + nldjtl(js) - 1 |
---|
| 295 | njctle = njmpptl(js) + nlejtl(js) - 1 |
---|
| 296 | ENDIF |
---|
| 297 | ENDIF |
---|
| 298 | WRITE(numid,*) |
---|
| 299 | WRITE(numid,*) 'prt_tra_ctl : Sum control indices' |
---|
| 300 | WRITE(numid,*) '~~~~~~~' |
---|
| 301 | WRITE(numid,*) |
---|
| 302 | WRITE(numid,9000)' nlej = ', nlejtl(js), ' ' |
---|
| 303 | WRITE(numid,9000)' ------------- njctle = ', njctle, ' -------------' |
---|
| 304 | WRITE(numid,9001)' | |' |
---|
| 305 | WRITE(numid,9001)' | |' |
---|
| 306 | WRITE(numid,9001)' | |' |
---|
| 307 | WRITE(numid,9002)' nictls = ', nictls, ' nictle = ', nictle |
---|
| 308 | WRITE(numid,9002)' nldi = ', nlditl(js), ' nlei = ', nleitl(js) |
---|
| 309 | WRITE(numid,9001)' | |' |
---|
| 310 | WRITE(numid,9001)' | |' |
---|
| 311 | WRITE(numid,9001)' | |' |
---|
| 312 | WRITE(numid,9004)' njmpp = ',njmpptl(js),' ------------- njctls = ', njctls, ' -------------' |
---|
| 313 | WRITE(numid,9003)' nimpp = ', nimpptl(js), ' nldj = ', nldjtl(js), ' ' |
---|
| 314 | WRITE(numid,*) |
---|
| 315 | WRITE(numid,*) |
---|
| 316 | |
---|
| 317 | 9000 FORMAT(a41,i4.4,a14) |
---|
| 318 | 9001 FORMAT(a59) |
---|
| 319 | 9002 FORMAT(a20,i4.4,a36,i3.3) |
---|
| 320 | 9003 FORMAT(a20,i4.4,a17,i4.4) |
---|
| 321 | 9004 FORMAT(a11,i4.4,a26,i4.4,a14) |
---|
| 322 | ENDDO |
---|
| 323 | |
---|
| 324 | END SUBROUTINE prt_ctl_trc_init |
---|
| 325 | |
---|
| 326 | |
---|
| 327 | SUBROUTINE sub_dom |
---|
| 328 | !!---------------------------------------------------------------------- |
---|
| 329 | !! *** ROUTINE sub_dom *** |
---|
| 330 | !! |
---|
| 331 | !! ** Purpose : Lay out the global domain over processors. |
---|
| 332 | !! CAUTION: |
---|
| 333 | !! This part has been extracted from the mpp_init |
---|
| 334 | !! subroutine and names of variables/arrays have been |
---|
| 335 | !! slightly changed to avoid confusion but the computation |
---|
| 336 | !! is exactly the same. Any modification about indices of |
---|
| 337 | !! each sub-domain in the mppini.F90 module should be reported |
---|
| 338 | !! here. |
---|
| 339 | !! |
---|
| 340 | !! ** Method : Global domain is distributed in smaller local domains. |
---|
| 341 | !! Periodic condition is a function of the local domain position |
---|
| 342 | !! (global boundary or neighbouring domain) and of the global |
---|
| 343 | !! periodic |
---|
| 344 | !! Type : jperio global periodic condition |
---|
| 345 | !! nperio local periodic condition |
---|
| 346 | !! |
---|
| 347 | !! ** Action : - set domain parameters |
---|
| 348 | !! nimpp : longitudinal index |
---|
| 349 | !! njmpp : latitudinal index |
---|
| 350 | !! nperio : lateral condition type |
---|
| 351 | !! narea : number for local area |
---|
| 352 | !! nlcil : first dimension |
---|
| 353 | !! nlcjl : second dimension |
---|
| 354 | !! nbondil : mark for "east-west local boundary" |
---|
| 355 | !! nbondjl : mark for "north-south local boundary" |
---|
| 356 | !! |
---|
| 357 | !! History : |
---|
| 358 | !! ! 94-11 (M. Guyon) Original code |
---|
| 359 | !! ! 95-04 (J. Escobar, M. Imbard) |
---|
| 360 | !! ! 98-02 (M. Guyon) FETI method |
---|
| 361 | !! ! 98-05 (M. Imbard, J. Escobar, L. Colombet ) SHMEM and MPI versions |
---|
| 362 | !! 8.5 ! 02-08 (G. Madec) F90 : free form |
---|
| 363 | !!---------------------------------------------------------------------- |
---|
| 364 | !! * Local variables |
---|
| 365 | INTEGER :: ji, jj, js ! dummy loop indices |
---|
| 366 | INTEGER :: & |
---|
| 367 | ii, ij, & ! temporary integers |
---|
| 368 | irestil, irestjl, & ! " " |
---|
| 369 | ijpi , ijpj, nlcil, & ! temporary logical unit |
---|
| 370 | nlcjl , nbondil, nbondjl, & |
---|
| 371 | nrecil, nrecjl, nldil, nleil, nldjl, nlejl |
---|
| 372 | |
---|
| 373 | INTEGER, DIMENSION(:,:), ALLOCATABLE :: & |
---|
| 374 | iimpptl, ijmpptl, ilcitl, ilcjtl ! temporary workspace |
---|
| 375 | REAL(wp) :: zidom, zjdom ! temporary scalars |
---|
| 376 | !!---------------------------------------------------------------------- |
---|
| 377 | |
---|
| 378 | ! 1. Dimension arrays for subdomains |
---|
| 379 | ! ----------------------------------- |
---|
| 380 | ! Computation of local domain sizes ilcitl() ilcjtl() |
---|
| 381 | ! These dimensions depend on global sizes isplt,jsplt and jpiglo,jpjglo |
---|
| 382 | ! The subdomains are squares leeser than or equal to the global |
---|
| 383 | ! dimensions divided by the number of processors minus the overlap |
---|
| 384 | ! array (cf. par_oce.F90). |
---|
| 385 | |
---|
| 386 | ijpi = ( jpiglo-2*jpreci + (isplt-1) ) / isplt + 2*jpreci |
---|
| 387 | ijpj = ( jpjglo-2*jprecj + (jsplt-1) ) / jsplt + 2*jprecj |
---|
| 388 | |
---|
| 389 | ALLOCATE(ilcitl (isplt,jsplt)) |
---|
| 390 | ALLOCATE(ilcjtl (isplt,jsplt)) |
---|
| 391 | |
---|
| 392 | nrecil = 2 * jpreci |
---|
| 393 | nrecjl = 2 * jprecj |
---|
| 394 | irestil = MOD( jpiglo - nrecil , isplt ) |
---|
| 395 | irestjl = MOD( jpjglo - nrecjl , jsplt ) |
---|
| 396 | |
---|
| 397 | IF( irestil == 0 ) irestil = isplt |
---|
| 398 | DO jj = 1, jsplt |
---|
| 399 | DO ji = 1, irestil |
---|
| 400 | ilcitl(ji,jj) = ijpi |
---|
| 401 | END DO |
---|
| 402 | DO ji = irestil+1, isplt |
---|
| 403 | ilcitl(ji,jj) = ijpi -1 |
---|
| 404 | END DO |
---|
| 405 | END DO |
---|
| 406 | |
---|
| 407 | IF( irestjl == 0 ) irestjl = jsplt |
---|
| 408 | DO ji = 1, isplt |
---|
| 409 | DO jj = 1, irestjl |
---|
| 410 | ilcjtl(ji,jj) = ijpj |
---|
| 411 | END DO |
---|
| 412 | DO jj = irestjl+1, jsplt |
---|
| 413 | ilcjtl(ji,jj) = ijpj -1 |
---|
| 414 | END DO |
---|
| 415 | END DO |
---|
| 416 | |
---|
| 417 | zidom = nrecil |
---|
| 418 | DO ji = 1, isplt |
---|
| 419 | zidom = zidom + ilcitl(ji,1) - nrecil |
---|
| 420 | END DO |
---|
| 421 | |
---|
| 422 | zjdom = nrecjl |
---|
| 423 | DO jj = 1, jsplt |
---|
| 424 | zjdom = zjdom + ilcjtl(1,jj) - nrecjl |
---|
| 425 | END DO |
---|
| 426 | |
---|
| 427 | ! 2. Index arrays for subdomains |
---|
| 428 | ! ------------------------------- |
---|
| 429 | |
---|
| 430 | ALLOCATE(iimpptl(isplt,jsplt)) |
---|
| 431 | ALLOCATE(ijmpptl(isplt,jsplt)) |
---|
| 432 | |
---|
| 433 | iimpptl(:,:) = 1 |
---|
| 434 | ijmpptl(:,:) = 1 |
---|
| 435 | |
---|
| 436 | IF( isplt > 1 ) THEN |
---|
| 437 | DO jj = 1, jsplt |
---|
| 438 | DO ji = 2, isplt |
---|
| 439 | iimpptl(ji,jj) = iimpptl(ji-1,jj) + ilcitl(ji-1,jj) - nrecil |
---|
| 440 | END DO |
---|
| 441 | END DO |
---|
| 442 | ENDIF |
---|
| 443 | |
---|
| 444 | IF( jsplt > 1 ) THEN |
---|
| 445 | DO jj = 2, jsplt |
---|
| 446 | DO ji = 1, isplt |
---|
| 447 | ijmpptl(ji,jj) = ijmpptl(ji,jj-1)+ilcjtl(ji,jj-1)-nrecjl |
---|
| 448 | END DO |
---|
| 449 | END DO |
---|
| 450 | ENDIF |
---|
| 451 | |
---|
| 452 | ! 3. Subdomain description |
---|
| 453 | ! ------------------------ |
---|
| 454 | |
---|
| 455 | DO js = 1, ijsplt |
---|
| 456 | ii = 1 + MOD( js-1, isplt ) |
---|
| 457 | ij = 1 + (js-1) / isplt |
---|
| 458 | nimpptl(js) = iimpptl(ii,ij) |
---|
| 459 | njmpptl(js) = ijmpptl(ii,ij) |
---|
| 460 | nlcitl (js) = ilcitl (ii,ij) |
---|
| 461 | nlcil = nlcitl (js) |
---|
| 462 | nlcjtl (js) = ilcjtl (ii,ij) |
---|
| 463 | nlcjl = nlcjtl (js) |
---|
| 464 | nbondjl = -1 ! general case |
---|
| 465 | IF( js > isplt ) nbondjl = 0 ! first row of processor |
---|
| 466 | IF( js > (jsplt-1)*isplt ) nbondjl = 1 ! last row of processor |
---|
| 467 | IF( jsplt == 1 ) nbondjl = 2 ! one processor only in j-direction |
---|
| 468 | ibonjtl(js) = nbondjl |
---|
| 469 | |
---|
| 470 | nbondil = 0 ! |
---|
| 471 | IF( MOD( js, isplt ) == 1 ) nbondil = -1 ! |
---|
| 472 | IF( MOD( js, isplt ) == 0 ) nbondil = 1 ! |
---|
| 473 | IF( isplt == 1 ) nbondil = 2 ! one processor only in i-direction |
---|
| 474 | ibonitl(js) = nbondil |
---|
| 475 | |
---|
| 476 | nldil = 1 + jpreci |
---|
| 477 | nleil = nlcil - jpreci |
---|
| 478 | IF( nbondil == -1 .OR. nbondil == 2 ) nldil = 1 |
---|
| 479 | IF( nbondil == 1 .OR. nbondil == 2 ) nleil = nlcil |
---|
| 480 | nldjl = 1 + jprecj |
---|
| 481 | nlejl = nlcjl - jprecj |
---|
| 482 | IF( nbondjl == -1 .OR. nbondjl == 2 ) nldjl = 1 |
---|
| 483 | IF( nbondjl == 1 .OR. nbondjl == 2 ) nlejl = nlcjl |
---|
| 484 | nlditl(js) = nldil |
---|
| 485 | nleitl(js) = nleil |
---|
| 486 | nldjtl(js) = nldjl |
---|
| 487 | nlejtl(js) = nlejl |
---|
| 488 | END DO |
---|
| 489 | |
---|
| 490 | DEALLOCATE(iimpptl) |
---|
| 491 | DEALLOCATE(ijmpptl) |
---|
| 492 | DEALLOCATE(ilcitl) |
---|
| 493 | DEALLOCATE(ilcjtl) |
---|
| 494 | |
---|
| 495 | END SUBROUTINE sub_dom |
---|
| 496 | |
---|
| 497 | #else |
---|
| 498 | !!---------------------------------------------------------------------- |
---|
| 499 | !! Dummy module : NO passive tracer |
---|
| 500 | !!---------------------------------------------------------------------- |
---|
| 501 | #endif |
---|
| 502 | |
---|
| 503 | !!====================================================================== |
---|
| 504 | |
---|
| 505 | END MODULE prtctl_trc |
---|