Changes between Version 9 and Version 10 of ticket/1851/General


Ignore:
Timestamp:
2017-02-23T11:44:26+01:00 (4 years ago)
Author:
frrh
Comment:

Legend:

Unmodified
Added
Removed
Modified
  • ticket/1851/General

    v9 v10  
    113113The only way I can do it is by introducing a check in scatter_global_ext (which is the variant issuing the error) to explicitly exclude negative J indices. That works OK but it seems a bit of a cludge.   
    114114 
     115 
     116 
     117== Coupled models == 
     118 
     119Repeating these tests with a coupled model set up. MS provides u-aj599 as the latest job. This features JP's MEDUSA_Stable branch at revision 7709 which contains '''SOME OF''' the fixes identified above.  
     120However it does not address the issue with the seemingly needless inclusion of unallocated arrays in namelists. 
     121 
     122So my WC is u-ak282debug. This immediately fails on compilation with another medusa related OOB issue in the dimensioning of ierr in oce.F90. However although this is related to MEDUSA the relevant code actually lives in the GO6 package branch. So I've got to create a bug fix branch for that as well as a new one for the up-to-date MEDUSA branch.  
     123 
     124So my branches are:  
     125   * branches/UKMO/dev_r5518_MEDUSA_stable_fixes_part2 
     126   * branches/UKMO/dev_r5518_GO6_package_fixes. 
     127 
     128 
     129 
     130