[Mpi-21] Review of Chapter 14

Bronis R. de Supinski bronis at [hidden]
Tue Mar 11 09:45:24 CDT 2008



Rolf:

Overall, this chapter looks pretty good. I only saw two general issues.
One is simple: the use of "which" versus "that". Style guides instruct
that we use "that" for required clauses (if the clause were omitted the
sentence would not make sense, essentially). The other is that section
14.6 needs to be integrated into section 14.1 so that the document reads
as the MPI 2.1 standard rather than a straight merge of the MPI-1 standard
with related sections of MPI-2.

Specific edits for Chapter 14:

[477:18-20] Currently reads:

1. provide a mechanism through which all of the MPI defined functions may be accessed
   with a name shift. Thus all of the MPI functions (which normally start with start with the prefix
   "MPI_") should also be accessible with the prefix "PMPI_".

Integrate text from section 14.6; change to:

1. provide a mechanism through which all of the MPI defined functions
   (except those allowed as macros, see Section 2.6.5) may be accessed
   with a name shift. This requires, in C and Fortran, an alternate entry
   point name, with the prefix PMPI_ for each MPI function. The profiling
   interface in C++ is described in Section 13.1.10. For routines
   implemented as macros, it is still required that the PMPI_ version be
   supplied and work as expected, but it is not possible to replace at
   link time the MPI_ version with a user-defined version.

Related to this change, delete:

[482:27-35]

[477:33] Currently reads:

This is necessary

Antecedent is unclear; change to:

This separability is necessary

------------------

All other suggested changes are simple replacements of "which" by "that".
The following is a list of the relevant [page:line number] pairs:

[477:30], [478:1], [478:28], [479:8], [481:7], [481:12], [481:22], [481:26],
[481:29], [482:10], [482:11]

--------------------

That's it for that chapter. I'll send stuff for Chapter 9 next although
it might be easier for me to edit the tex file (with explanatory comments)
and then for you to do a diff. I'll look to see if it is available on the
web site...

Bronis



More information about the Mpi-21 mailing list