[Mpi-forum] Draft standard - take II

Tue Oct 26 09:39:57 CDT 2010

> Additionally, there were frontmatter changes yesterday on the trunk
> last night that were made after the branch. Do they need to come over
> to the branch?

Which branch? What is going on?

Do we have a ticket to which we report our chapter PDF reviews?

> removed superfluous paragraph after review from Yutaka Ishikawa and
> Hidejuki Jitsumoto

Yes, it is okay - but it was complicated to make a review about 
Ticket 109, svn revision 628.

For all other chapters: My recommentation is to compare to rev 495
If all changes compared to rev 495 are okay (i.e., do not need any
Forum vote, and are clearly marked with Bill's macros) 
then the chapter is okay.

Exceptions, are 
  - Chapters 5, Biblio, and 
    Appendices A.2.3, A.3.3, A3.4.3 and B. Change-Log:
      here ticket 109 is included
  - Frontmatter: MPI-3.0 is introduced
  - all the rest: "no changes"

Test:

  cd chap-xxxx
  svn diff -r 495:629 | grep '^[++-]'

My reviews:

  0. chap-frontm - MANY CHANGES --> reviews are necessary
       Review of .tex and .pdf is needed
  1. chap-intro - SEVERAL CHANGES --> reviews are necessary
       All changes are clearly markuped, i.e.,
       make reviewdoc
       and reviewing colored mpi-report.pdf should be enough
  2. chap-terms - no changes
  3. chap-pt2pt - no changes (only a new Latex label)
  4. chap-datatypes - no changes
  5. chap-coll -->>>>> See ticket 109 for all review-discussions
  6. chap-context - no changes
  7. chap-topol - no changes
  8. chap-inquiry - no changes (only an unused macro is commented out)
  9. chap-misc - no changes
 10. chap-dynamic - no changes
 11. chap-one-side - no changes
 12. chap-ei - no changes
 13. chap-io - no changes
 14. chap-prof - one formatting change: MPI\_SEND --> \mpifunc{MPI\_SEND}
 15. chap-deprecated - no changes
 16. chap-binding - no changes
       Should be okay, because ticket 109 has nothing specific to this chapter
 A.1 chap-appLang - no changes
       Should be okay, because ticket 109 has nothing specific to this chapter
       ONE EXCEPTION: 3 times definition of MPI_User_function in A.1.3
                      see PDF reviews of Ticket 109
 B   chap-changes - Review is needed

Result of my (automated) reviews:

  Manual reviews are only needed for 
    - Frontmatter
    - 1 Intro
    - 5 Coll
    - 14 Profiling (controling this one change)
    - A.1.3 due to the reported issues in ticket 109
    - B Change-Log

Best regards
Rolf

----- Original Message -----
> From: "Jeff Squyres" <jsquyres at cisco.com>
> To: "Main MPI Forum mailing list" <mpi-forum at lists.mpi-forum.org>
> Sent: Tuesday, October 26, 2010 2:54:15 PM
> Subject: Re: [Mpi-forum] Draft standard - take II
> For example, these changes on the trunk are a little alarming. Who is
> the collectives chapter chair -- is it Torsten? It is not marked on
> the web page.
> 
> ------------------------------------------------------------------------
> r618 | htor | 2010-10-24 20:38:04 -0400 (Sun, 24 Oct 2010) | 1 line
> Changed paths:
> M /trunk/MPI-3.0/chap-coll/coll.tex
> 
> removed superfluous paragraph after review from Yutaka Ishikawa and
> Hidejuki Jitsumoto
> ------------------------------------------------------------------------
> 
> Whoa -- removed a whole paragraph? Wasn't this text voted on (multiple
> times)? Maybe removing that specific text is ok, but that log message
> gives me a little pause, for example...
> 
> Additionally, there were frontmatter changes yesterday on the trunk
> last night that were made after the branch. Do they need to come over
> to the branch?
> 
> 
> 
> On Oct 26, 2010, at 8:47 AM, Jeff Squyres wrote:
> 
> > On Oct 25, 2010, at 9:28 PM, Graham, Richard L. wrote:
> >
> >> Just a friendly reminder that we are trying to get the draft
> >> standard out before SC, and were aiming to have people review their
> >> sections by mid last week. By my records the only chapters that
> >> have passed WG reviews are:
> >> - Environment management
> >> - Profiling Interface
> >>
> >> I am equating 100% WG review == section reviewed.
> >
> > I'm not sure I understand. Are you looking for an affirmation from
> > every single member of every single WG? I thought that the committee
> > chair reporting back would be enough.
> >
> >> Please make your best effort to complete reviews ASAP, so we can
> >> get this out the door.
> >
> > Which are we reviewing, the trunk or the branch? (I don't think that
> > this has been conveyed to the Forum yet)
> >
> > Both changed as of yesterday.
> >
> > --
> > Jeff Squyres
> > jsquyres at cisco.com
> > For corporate legal information go to:
> > http://www.cisco.com/web/about/doing_business/legal/cri/
> >
> >
> > _______________________________________________
> > mpi-forum mailing list
> > mpi-forum at lists.mpi-forum.org
> > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum
> 
> 
> --
> Jeff Squyres
> jsquyres at cisco.com
> For corporate legal information go to:
> http://www.cisco.com/web/about/doing_business/legal/cri/
> 
> 
> _______________________________________________
> mpi-forum mailing list
> mpi-forum at lists.mpi-forum.org
> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum

-- 
Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner at hlrs.de
High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530
University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832
Head of Dpmt Parallel Computing . . . www.hlrs.de/people/rabenseifner
Nobelstr. 19, D-70550 Stuttgart, Germany . (Office: Allmandring 30)