Changeset 4c64400 in flexpart.git for src/mpi_mod.f90

Timestamp:

Nov 8, 2016, 4:42:27 PM (7 years ago)

Author:

Espen Sollum ATMOS <eso@…>

Branches:

master, 10.4.1_pesei, GFS_025, bugfixes+enhancements, dev, release-10, release-10.4.1, scaling-bug, univie

Children:

c8fc724

Parents:

16b61a5

Message:

Bugfix for double precision dry deposition calculation. Added (hardcoded) option to not use output kernel. Version/date string updated.

File:

• src/mpi_mod.f90 (modified) (3 diffs)

src/mpi_mod.f90

@@ -331 +331 @@
 ! Set maxpart per process
 ! eso 08/2016: Increase maxpart per process, in case of unbalanced distribution
-    maxpart_mpi=int(mp_maxpart_factor*maxpart/mp_partgroup_np)
+    maxpart_mpi=int(mp_maxpart_factor*real(maxpart)/real(mp_partgroup_np))
 
 ! Do not allocate particle data arrays for readwind process
@@ -832 +832 @@
       numpart=max(numpart,ipart)
 
-      deallocate(itra1_tmp,npoint_tmp,nclass_tmp,idt_tmp,itramem_tmp,itrasplit_tmp,&
-           & xtra1_tmp,ytra1_tmp,ztra1_tmp,xmass1_tmp)
+      deallocate(itra1_tmp,npoint_tmp,nclass_tmp,idt_tmp,itramem_tmp,&
+           &itrasplit_tmp,xtra1_tmp,ytra1_tmp,ztra1_tmp,xmass1_tmp)
 
       if (mp_dev_mode) then
@@ -2340 +2340 @@
 !
 ! TODO
-!   take into account nested wind fields
+!   NB: take into account nested wind fields by using a separate 
+!   subroutine mpif_gf_request_nest (and separate request arrays for the
+!   nested variables)
 !
 !*******************************************************************************