[mpiwg-tools] Tools WG webex tomorrow!

[Jeff Squyres (jsquyres)]

I have a better suggestion (sorry... this is what happens at the last second...):

Current from Kathryn's PDF:

If the constant MPI_T_PVAR_ALL_HANDLES is passed in handle, the MPI implementation attempts to start all variables within the session identified by the parameter session for which handles have been allocated. In this case, the routine returns MPI_SUCCESS if all variables are started successfully, otherwise MPI_T_ERR_PVAR_NO_STARTSTOP is returned. MPI_SUCCESS is also returned if MPI_T_PVAR_ALL_HANDLES is passed in handles and if no handles are valid for session or all valid handles represent continuous variables. Continuous variables and variables that are already started are ignored when MPI_T_PVAR_ALL_HANDLES is specified.

Jeff's suggestion (delete the new sentence and add the text in ***):

If the constant MPI_T_PVAR_ALL_HANDLES is passed in handle, the MPI implementation attempts to start all variables within the session identified by the parameter session for which handles have been allocated. In this case, the routine returns MPI_SUCCESS if all variables are started successfully (even if there are no non-continuous variables to be started), otherwise MPI_T_ERR_PVAR_NO_STARTSTOP is returned.  Continuous variables and variables that are already started are ignored when MPI_T_PVAR_ALL_HANDLES is specified.




On Aug 27, 2013, at 12:09 PM, Kathryn Mohror <kathryn at llnl.gov> wrote:

> I made a ticket for this (#391) and uploaded a pdf with the proposed changes (also attached). 
> 
> Is anyone opposed to putting this up for a reading as errata?
> 
> Kathryn
> 
> 
> On Aug 27, 2013, at 5:14 AM, Jeff Squyres (jsquyres) <jsquyres at cisco.com> wrote:
> 
>> On Aug 26, 2013, at 7:23 PM, "Schulz, Martin" <schulz6 at llnl.gov> wrote:
>> 
>>>> I am now on the SUCCESS side of the fence. Does it need to be clarified as an errata? I would suggest (from 3.0 clean doc):
>> 
>> Just to be totally clear: I don't really care which way it goes.  I just saw this issue as an ambiguity that should be resolved.
>> 
>>>> page 580, line 41: 
>>>> In this case, the routine returns MPI_SUCCESS if all variables are started successfully, otherwise MPI_T_ERR_PVAR_NO_STARTSTOP is returned. MPI_SUCCESS is also returned if no handles are valid for session and MPI_T_PVAR_ALL_HANDLES is passed in handle. 
>>> 
>>> Hmm - I fear we would have to be even more explicit: MPI_SUCCESS is also returned if MPI_T_PVAR_ALL_HANDLES is passed in handles and if no handles are valid for session or all valid handles represent continuous (read only for reset) variables.
>> 
>> +1.
>> 
>> -- 
>> Jeff Squyres
>> jsquyres at cisco.com
>> For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/
>> 
>> _______________________________________________
>> mpiwg-tools mailing list
>> mpiwg-tools at lists.mpi-forum.org
>> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-tools
> 
> ______________________________________________________________
> Kathryn Mohror, kathryn at llnl.gov, http://people.llnl.gov/mohror1
> CASC @ Lawrence Livermore National Laboratory, Livermore, CA, USA
> 
> 
> 
> 
> <tools-3.pdf>


-- 
Jeff Squyres
jsquyres at cisco.com
For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/