[mpiwg-tools] NULL args in pvar_get_info

Mohror, Kathryn mohror1 at llnl.gov
Mon Sep 23 18:20:43 CDT 2013


Hi Junchao,

This is a recent item we added to the errata. I don't believe it has been merged into the online errata yet.

The text for the errata is:

 page 568, line 31 add: Any output parameter to MPI_T_cvar_get_info can have a NULL value. In this case, the implementation will ignore the parameter and not return a value for that parameter.

page 576, line 46 add: Any output parameter to MPI_T_pvar_get_info can have a NULL value. In this case, the implementation will ignore the parameter and not return a value for that parameter.

page 586, line 22 add: Any output parameter to MPI_T_category_get_info can have a NULL value. In this case, the implementation will ignore the parameter and not return a value for that parameter.

You can see the evolution of this change here:

https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/378

Kathryn



On Sep 23, 2013, at 4:03 PM, Junchao Zhang <jczhang at mcs.anl.gov<mailto:jczhang at mcs.anl.gov>> wrote:

Hello,
  For function "int MPI_T_pvar_get_info(int pvar_index, char *name, int *name_len, int *verbosity, int *var_class, ...)", MPI_T defines semantics when name or name_len are NULL.
  But for other args, if they are NULL, is it a user program error or should we just ignore it? I prefer the latter. But the standard doesn't say it.

  Thanks
--Junchao Zhang
_______________________________________________
mpiwg-tools mailing list
mpiwg-tools at lists.mpi-forum.org<mailto:mpiwg-tools at lists.mpi-forum.org>
http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-tools

______________________________________________________________
Kathryn Mohror, kathryn at llnl.gov<mailto:kathryn at llnl.gov>, http://people.llnl.gov/mohror1
CASC @ Lawrence Livermore National Laboratory, Livermore, CA, USA




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mpi-forum.org/pipermail/mpiwg-tools/attachments/20130923/e0c76354/attachment-0001.html>


More information about the mpiwg-tools mailing list