Just in time for the Dublin meeting, I have posted several new MPI-2.2 proposals (it looks like a bunch of others have, too!) based on my items from the following URLs: http://www.cs.uiuc.edu/homes/wgropp/projects/parallel/MPI/mpi-errata/mpi2-2i... http://www.cs.uiuc.edu/homes/wgropp/projects/parallel/MPI/mpi-errata/mpi2-2i... http://www.cs.uiuc.edu/homes/wgropp/projects/parallel/MPI/mpi-errata/mpi2-2i... I numbered each of the proposals with the item number from the above URLs. Here's the resulting proposals on the wiki: https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/mpi22/CxxBindingsMissingCo... https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/mpi22/GrequestStartFnPtrAr... https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/mpi22/FnPtrTypedefNames https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/mpi22/TextUpdatesToLangBin... https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/mpi22/MPICancelWrongArgTyp... https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/mpi22/NamingConventionsPre... https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/mpi22/RemoveCxxDeprecatedS... Additionally, several of the suggestions that I made on the #squyresX URLs are now moot. Here's a listing by issue number, including brief descriptions of why they are moot: ====================================================================== 30.c ? p4: Two open source MPI implementations are cited that are no longer relevant. This entire sentence should be removed. Indeed, the language in the overall paragraph is forward-looking -- it should probably be re-worked to be in the present tense. Editor's comment: should be done by the Forum Editor (WDG): The historical context is important. Rather than remove this, it should be updated. --> now moot in the final 2.1 doc; the language was re-worked. ====================================================================== 30.o R2 p186.5: Descriptions for the C++ bindings need to be included here, since they are different than the C bindings. The text from 13.1.7 would seem to be sufficient. (See also 35.e) ---> seems moot now -- MPI-2.1 p195:41 says "Please see Section 16.1.7 on page 455 for further discussion about the C++ bindings for Dup() and Clone()." ====================================================================== 71.c ? - MPI-2.1 p571 Examples Index: there are still a bunch of repeated names, some in all caps, some in mixed case, etc. 71.d ? - MPI-2.1 p571 Examples Index: There are some examples listed just by MPI function name (e.g., MPI_SEND and MPI_Send) -- are we listing every MPI function in every MPI example? How was the selection for these example names given? Editor's comment: I tried to pick major routines in the examples (sometimes not all). Additionally, for some examples, I defined titles for the Index. 71.e ? - MPI-2.1 p574 MPI Constant and Predefined Handle Index: The first entry is still "MPI::*_NULL" 71.f ? - MPI-2.1 p574 MPI Constant and Predefined Handle Index: Are we listing the C++ constants and Fortran constants here, too? Or just the C constants? Or just language neutral? (I only see C++ predefined datatypes listed -- should we list all or none of them?) --> these still need to be cleaned up; but I don't think they need a proposal. ====================================================================== p10.35: due to the decision from last meeting (sort out the IN/OUT/INOUT mess in MPI-2.2), the language should be softened in this paragraph and the full paragraph following this one (because they contradict each other). Specifically, I propose changing: 10.35: Thus, in C++, IN aguments are either references... to Thus, in C++, IN arguments are usually either references... Editor's comment: Is there an exception from this rule? If yes, then the proposal is okay. --> now moot; MPI 2.1 p10:34 says "...usually..." Remember that the Forum was very careful to state that this is *not* a rule; it's a guideline. ====================================================================== 32.j' ? Which names should be visible in the Index for Dup and Clone? TODO --> MPI::Comm::Clone() and MPI::Comm::Dup() show up in MPI-2.1 Annex A. 4 (C++ bindings) and MPI_COMM_CLONE and MPI_COMM_DUP show up in the MPI Function Index. I think that this is sufficient. ====================================================================== 30.h MPI-2.1 p11.22: Fortran in this document refers to Fortran 90". For MPI-2.1, it is probably suitable to leave this, but we might want to make a statement (footnote or parenthetical) that it is expected to be updated in future MPI spec revisions. --> I no longer think that this is necessary. ====================================================================== 32.k R2 p445.36-38: Remove this entire paragraph ("Compilers that do not support..."). This feature has been a part of C++ since C+ +98, and exists in all modern C++ compilers. --> Moot: it's gone. (MPI-2.1 p443:46) ====================================================================== 33.n R2 p556.bottom: Missing prototypes for the MPI::Exceptions class --> Moot: now included in MPI-2.1 A.4.12. ====================================================================== 32.r 22 p463.17-18: Delete first sentence; delete "In MPI-2," --> Moot: done in MPI-2.2. ====================================================================== Enjoy. -- Jeff Squyres Cisco Systems