itaps-parallel FMDB libraries are updated on mesh machine

[Tim Tautges]

Hi all,
   I'll copy everything over to the other machine sometime tomorrow 
before noon.  I've already copied most things once (yesterday).  I'm 
only copying what's under /usr/local/itaps; if you want something else 
copied too, let me know.

- tim

txie at scorec.rpi.edu wrote:
> Hi all,
> 
> I just updated FMDB libraries on mesh machine (Tim's machine) based on the
> latest iMeshP.h version. The main modifications are name changes on some
> iMeshP functions. The paths to the libraries are:
> 
> /usr/local/itaps/FMDB_iMeshP/     (iMeshP enabled, for parallel)
> /usr/local/itaps/FMDB_iMesh/      (iMesh enabled, for serial)
> 
> Please see the following changes from Karen's last email, and make sure
> that   your program still works if it uses FMDB_iMeshP. Also let me know
> if you find any problems. Thanks.
> 
> Ting
> 
> 
> 
> 
>> Changes:
>>
>> -  Reformatted many, many, many comments for Doxygen.  (Anyone feel like
>> proofreading??)
>>
>> -  Changed function names for consistency with our conventions:
>>     iMeshP_Wait  ->  iMeshP_waitForRequest
>>     iMeshP_WaitAll  ->  iMeshP_waitForAllRequests
>>     iMeshP_WaitAny  ->  iMeshP_waitForAnyRequest
>>     iMeshP_WaitEnt  ->  iMeshP_waitForRequestEnt
>>     iMeshP_Test  ->  iMeshP_testRequest
>>     iMeshP_load  -> iMeshP_loadAll
>>     iMeshP_save  -> iMeshP_saveAll
>>     iMeshP_createGhostEnts  -> iMeshP_createGhostEntsAll
>>     iMeshP_deleteGhostEnts  ->  iMeshP_deleteGhostEntsAll
>>
>> -  Removed iMeshP_Status from the interface; it was defined only as a
>> handle, and we didn't give the apps a way to access any of its data.
>>
>> -  Removed some "const" keywords where I thought they didn't make sense.
>>
>> -  Removed redundant argument num_ghost_rules from iMeshP_getGhostEntInfo.
>>
>> -  In iMeshP_pushTags and its siblings, change the tag argument to a
>> source-tag argument; added a destination tag argument.
>> ====================================================================
>>
>> Out-standing questions and things to do:
>>
>> - In iMeshP_exchEntArrToPartsAll, what are implementations using for the
>> command_code?  Should I make this an enumerated type?   Or just "yes/no"
>> for
>> migration?  (Yes == migrate entity, no == create only a copy).
>>
>> - In iMeshP_exchEntArrToPartsAll, how are implementations interpreting the
>> update_ghost argument?
>>
>> - In iMeshP_exchEntArrToPartsAll, what adjacencies are being migrated with
>> the entities in the arguments?  Downward adjacencies only?  Is the
>> ownership
>> of the downward adjacencies automatically changed to something that "makes
>> sense"?  (I think Ting raised this issue before but I don't recall that we
>> resolved her questions.)
>>
>> -  How is the offset array in iMeshP_replaceOnPartBdry used?  I guessed
>> the
>> following; Onkar, let me know if it is correct.  If it is correct, do we
>> need offset_size?
>>       offset  (In)  Index into new_entities; old_entities[i]
>>                     is replaced by new_entities[offset[i]] to
>>                     new_entities[offset[i+1]-1].
>>
>> -  Resolve Jason's recent questions about creating ghost entities.
>>
>> -  Add functions to either pull or accumulate tag data as proposed by
>> Jason.
>>
>> -  Update iMeshP functions to reflect changes in iMesh, if any, to
>> iMesh_getAllVtxCoords and iMesh_getVtxCoordIndex as outlined by Jason.
>>
>>
>> Karen
>>
>>
>>
> 
> 
> 

-- 
================================================================
"You will keep in perfect peace him whose mind is
   steadfast, because he trusts in you."               Isaiah 26:3

              Tim Tautges            Argonne National Laboratory
          (tautges at mcs.anl.gov)      (telecommuting from UW-Madison)
          phone: (608) 263-8485      1500 Engineering Dr.
            fax: (608) 263-4499      Madison, WI 53706