September 2009 - Mpi-22 - mailman3.cels.anl.gov

Re: [Mpi-22] MPI document review instructions for those at Helsinki meeting
by unknown＠example.com 02 Sep '09

02 Sep '09

There are only a few minor differences between 442 and 447 (I've looked at the full diff). It is mostly a few additional fixes to spacing and correcting macro usage. The one visible change is array indices in the graph topology examples; there was a clear off-by-one error there. (Note to everyone: much as I'd like to improve the document, further fixes should be postponed until MPI-3.) Bill On Sep 2, 2009, at 11:19 AM, Jesper-Larsson Träff wrote: > > the current version in the repository is 447? > > Jesper > > On Wed, Sep 02, 2009 at 09:14:40AM -0500, William Gropp wrote: >> Here are the two slides that I presented about the review >> responsibilities. For those not at the Helsinki meeting, please feel >> free to look at any part of the document. >> Bill >> > > >> >> >> William Gropp >> Deputy Director for Research >> Institute for Advanced Computing Applications and Technologies >> Paul and Cynthia Saylor Professor of Computer Science >> University of Illinois Urbana-Champaign >> >> >> >> > >> _______________________________________________ >> mpi-22 mailing list >> mpi-22_at_[hidden] >> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-22 > > <smime.p7s><ATT00001.txt> William Gropp Deputy Director for Research Institute for Advanced Computing Applications and Technologies Paul and Cynthia Saylor Professor of Computer Science University of Illinois Urbana-Champaign

1 0

Re: [Mpi-22] MPI document review instructions for those at Helsinki meeting
by unknown＠example.com 02 Sep '09

02 Sep '09

the current version in the repository is 447? Jesper On Wed, Sep 02, 2009 at 09:14:40AM -0500, William Gropp wrote: > Here are the two slides that I presented about the review > responsibilities. For those not at the Helsinki meeting, please feel > free to look at any part of the document. > Bill > > > > William Gropp > Deputy Director for Research > Institute for Advanced Computing Applications and Technologies > Paul and Cynthia Saylor Professor of Computer Science > University of Illinois Urbana-Champaign > > > > > _______________________________________________ > mpi-22 mailing list > mpi-22_at_[hidden] > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-22 *

1 0

MPI document review instructions for those at Helsinki meeting
by unknown＠example.com 02 Sep '09

02 Sep '09

Here are the two slides that I presented about the review responsibilities. For those not at the Helsinki meeting, please feel free to look at any part of the document. Bill William Gropp Deputy Director for Research Institute for Advanced Computing Applications and Technologies Paul and Cynthia Saylor Professor of Computer Science University of Illinois Urbana-Champaign *

1 0

Re: [Mpi-22] [MPI Forum] #7: Function pointer typedefs: "_function" vs. "_fn"
by unknown＠example.com 02 Sep '09

02 Sep '09

If you build the version with changes, all of the macros in the deleted or replaced code are still executed, which means that side- effects, such as index entries, are added. I checked the official version, and the only index entries for the _fn versions are from the deprecated discussion and the change log. Bill On Sep 2, 2009, at 5:27 AM, MPI Forum wrote: > #7: Function pointer typedefs: "_function" vs. "_fn" > --------------------------------------------------------- > +------------------ > Reporter: jsquyres > | Owner: jsquyres > Type: Enhancements to standard > | Status: assigned > Priority: Waiting for PDF reviews > | Milestone: 2009/07/27 Chicago > Version: MPI 2.2 > | Keywords: > Implementation: Completed | > Author_bill_gropp: 1 > Author_rich_graham: 0 | > Author_adam_moody: 0 > Author_torsten_hoefler: 0 | > Author_dick_treumann: 0 > Author_jesper_larsson_traeff: 0 | > Author_george_bosilca: 1 > Author_david_solt: 0 | > Author_bronis_de_supinski: 0 > Author_rajeev_thakur: 0 | > Author_jeff_squyres: 0 > Author_alexander_supalov: 1 | > Author_rolf_rabenseifner: 1 > --------------------------------------------------------- > +------------------ > > Comment(by bronis): > > Perhaps I am just confused. Why do the *_fn items still appear in the > index in ticket7_all_except_inquiry.pdf? Is it outdated? > > I have a low bandwidth connection and I would like to avoid pulling > down > the current version of the whole document... > > It appears impossible to check the index without doing that but it > also > seems unlikely looking at the comment history that the old entries > have > been removed. > > PDF review of everything other than the index is OK. > > -- > Ticket URL: <https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/7#comment:45 > > > MPI Forum <https://svn.mpi-forum.org/> > MPI Forum William Gropp Deputy Director for Research Institute for Advanced Computing Applications and Technologies Paul and Cynthia Saylor Professor of Computer Science University of Illinois Urbana-Champaign

1 0

Re: [Mpi-22] Fwd: Final state transition - was Re: [MPI Forum] #104: MPI_ARGV_NULL and MPI_ARGVS_NULL missing on p465:13-15.
by unknown＠example.com 01 Sep '09

01 Sep '09

I still disagree (the state diagrams you sent are exactly the same as the one I sent yesterday). What is the point of leaving the tickets open after their PDFs have been reviewed and the text has been successfully committed? You don't gain anything in terms of searchability of MPI 2.2 tickets, for example. All it does it make yet another step that has to be performed later. I repeat: it was definitely my intention that "Withdrawn", "Rejected", "Deferred", and "Text committed" were definitely *resolutions*, not *priorities*. The fact that there is a "Text committed" priority is a mistake. On Sep 2, 2009, at 12:59 AM, Rolf Rabenseifner wrote: > > As answer to Jeffs mail, again with attached state diagram. > > --- the forwarded message follows --- > > <MPI-2.2_ticket_states.pdf> > From: "William Gropp" <wgropp_at_[hidden]> > Date: August 26, 2009 8:01:26 AM GMT+03:00 > To: "MPI 2.2" <mpi-22_at_[hidden]> > Subject: Re: [Mpi-22] Final state transition - was Re: [MPI Forum] > #104:MPI_ARGV_NULL and MPI_ARGVS_NULL missing on p465:13-15. > Reply-To: "MPI 2.2" <mpi-22_at_[hidden]> > > > Thanks, Rolf. You are correct. > > Bill > > On Aug 25, 2009, at 2:53 AM, Rolf Rabenseifner wrote: > > > Hi all, > > > > based on our final state transition diagram, > > the last state transition we are currently doing is: > > > > Priority: Waiting for PDF reviews --> Ticket complete > > > > This should be done by the ticket author (blue arrow), > > and not by the chapter author (arrow is not black). > > > > We don't define to set the ticket to > > Status: closed > > Resolution: Text committed > > > > Therefore, I'll set my ticket back to > > Status: reopened > > > > All othe completed tickets are in the same state, > > i.e., Status "open" or "reopened", > > and Priority "ticket complete". > > > > I hope that this is okay. > > > > Best regards > > Rolf > > > >> From: Jeff Squyres <jsquyres_at_[hidden]> > >> Subject: Re: [Mpi-22] SVN committing MPI-2.2 latex > >> Date: Fri, 12 Jun 2009 08:47:48 -0700 > >> > >> Ok, try this diagram. > > ... > >> Attachment: MPI-2.2 ticket states.pdf (463Kbytes) > > > > > > On Tue, 25 Aug 2009 01:08:32 -0000 > > "MPI Forum" <mpi-22_at_[hidden]> wrote: > >> #104: MPI_ARGV_NULL and MPI_ARGVS_NULL missing on p465:13-15. > >> -------------------------------------------------- > >> +------------------------- > >> Reporter: RolfRabenseifner | > >> Owner: RolfRabenseifner > >> Type: Trivial text changes | > >> Status: closed > >> Priority: Ticket complete | > >> Milestone: 2009/06/08 California > >> Version: MPI 2.2 | > >> Resolution: Text committed > >> Keywords: | > >> Implementation: Unnecessary > >> Author_bill_gropp: 0 | > >> Author_rich_graham: 0 > >> Author_adam_moody: 0 | > >> Author_torsten_hoefler: 0 > >> Author_dick_treumann: 0 | > >> Author_jesper_larsson_traeff: 0 > >> Author_george_bosilca: 0 | > >> Author_david_solt: 0 > >> Author_bronis_de_supinski: 0 | > >> Author_rajeev_thakur: 0 > >> Author_jeff_squyres: 1 | > >> Author_alexander_supalov: 0 > >> Author_rolf_rabenseifner: 0 | > >> -------------------------------------------------- > >> +------------------------- > >> Changes (by jsquyres): > >> > >> * status: new => closed > >> * resolution: => Text committed > >> > >> > >> Comment: > >> > >> The changes Rolf made look good to me. I attached a new PDF, just > >> in case > >> anyone cares. > >> > >> Resolving this ticket to "text committed". > >> > >> -- > >> Ticket URL: > >> <https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/104#comment: > 31> > >> MPI Forum <https://svn.mpi-forum.org/> > >> MPI Forum > > > > Dr. Rolf Rabenseifner . . . . . . . . . .. email > rabenseifner_at_[hidden] > > 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) > > _______________________________________________ > > mpi-22 mailing list > > mpi-22_at_[hidden] > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-22 > > William Gropp > Deputy Director for Research > Institute for Advanced Computing Applications and Technologies > Paul and Cynthia Saylor Professor of Computer Science > University of Illinois Urbana-Champaign > > > > > _______________________________________________ > mpi-22 mailing list > mpi-22_at_[hidden] > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-22 > > > > <ATT15751052.txt> -- Jeff Squyres jsquyres_at_[hidden]

1 0

Fwd: Final state transition - was Re: [MPI Forum] #104: MPI_ARGV_NULL and MPI_ARGVS_NULL missing on p465:13-15.
by unknown＠example.com 01 Sep '09

01 Sep '09

As answer to Jeffs mail, again with attached state diagram. --- the forwarded message follows --- <strong>attached mail follows:</strong> Thanks, Rolf. You are correct. Bill On Aug 25, 2009, at 2:53 AM, Rolf Rabenseifner wrote: > Hi all, > > based on our final state transition diagram, > the last state transition we are currently doing is: > > Priority: Waiting for PDF reviews --> Ticket complete > > This should be done by the ticket author (blue arrow), > and not by the chapter author (arrow is not black). > > We don't define to set the ticket to > Status: closed > Resolution: Text committed > > Therefore, I'll set my ticket back to > Status: reopened > > All othe completed tickets are in the same state, > i.e., Status "open" or "reopened", > and Priority "ticket complete". > > I hope that this is okay. > > Best regards > Rolf > >> From: Jeff Squyres <jsquyres_at_[hidden]> >> Subject: Re: [Mpi-22] SVN committing MPI-2.2 latex >> Date: Fri, 12 Jun 2009 08:47:48 -0700 >> >> Ok, try this diagram. > ... >> Attachment: MPI-2.2 ticket states.pdf (463Kbytes) > > > On Tue, 25 Aug 2009 01:08:32 -0000 > "MPI Forum" <mpi-22_at_[hidden]> wrote: >> #104: MPI_ARGV_NULL and MPI_ARGVS_NULL missing on p465:13-15. >> -------------------------------------------------- >> +------------------------- >> Reporter: RolfRabenseifner | >> Owner: RolfRabenseifner >> Type: Trivial text changes | >> Status: closed >> Priority: Ticket complete | >> Milestone: 2009/06/08 California >> Version: MPI 2.2 | >> Resolution: Text committed >> Keywords: | >> Implementation: Unnecessary >> Author_bill_gropp: 0 | >> Author_rich_graham: 0 >> Author_adam_moody: 0 | >> Author_torsten_hoefler: 0 >> Author_dick_treumann: 0 | >> Author_jesper_larsson_traeff: 0 >> Author_george_bosilca: 0 | >> Author_david_solt: 0 >> Author_bronis_de_supinski: 0 | >> Author_rajeev_thakur: 0 >> Author_jeff_squyres: 1 | >> Author_alexander_supalov: 0 >> Author_rolf_rabenseifner: 0 | >> -------------------------------------------------- >> +------------------------- >> Changes (by jsquyres): >> >> * status: new => closed >> * resolution: => Text committed >> >> >> Comment: >> >> The changes Rolf made look good to me. I attached a new PDF, just >> in case >> anyone cares. >> >> Resolving this ticket to "text committed". >> >> -- >> Ticket URL: >> <https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/104#comment:31> >> MPI Forum <https://svn.mpi-forum.org/> >> MPI Forum > > Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner_at_[hidden] > 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) > _______________________________________________ > mpi-22 mailing list > mpi-22_at_[hidden] > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-22 William Gropp Deputy Director for Research Institute for Advanced Computing Applications and Technologies Paul and Cynthia Saylor Professor of Computer Science University of Illinois Urbana-Champaign _______________________________________________ mpi-22 mailing list mpi-22_at_[hidden] http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-22 *

1 0

Re: [Mpi-22] [MPI Forum] #146: Fix MPI_INIT description text
by unknown＠example.com 01 Sep '09

01 Sep '09

Jeff, I forward in the next mail Bill's answer to an earlier question on this topic. Currently, all 82 valid MPI-2.2 are kept as unclosed. They go from passed to wait pdf to complete. I expect that Bill will do the final step to close all 82 tickets when MPI-2.2 is done. Attached is your latest state diagram. Best regards Rolf On Tue, 1 Sep 2009 15:50:54 +0200 Jeff Squyres <jsquyres_at_[hidden]> wrote: > Rolf -- > > I don't understand. I *did* follow my own state diagram -- ticket > #146 is done done done, so I resolved it in the "Ticket Complete" > state. I attached the PDF for reference. > > You're saying that the ticket should be left open and just have its > priority changed to "Text Committed". I disagree. > > The state diagram has the ticket state go to "Text Committed", not > "Text Committed" followed by resolving it in "Text Committed" > *again*. It was definitely my intention when making this state > diagram that tickets should be *resolved* (i.e., closed) in with the > "Text Committed" state. If there is a "Text Committed" priority >(and I see that there is), I believe that to be a mistake. > > To be clear -- "Withdrawn", "Rejected", and "Deferred" are >resolution (closing) states. They are not priorities. "Text >Committed" was intended to be the same. > > Perhaps I'm just grouchy from jet lag. :-) > > > > > > On Sep 1, 2009, at 3:40 PM, MPI Forum wrote: > >> #146: Fix MPI_INIT description text >> ------------------------------------------------- >> +-------------------------- >> Reporter: jsquyres >> | Owner: jsquyres >> Type: Text (only) changes >> | Status: reopened >> Priority: Ticket complete >> | Milestone: 2009/07/27 Chicago >> Version: MPI 2.2 >> | Resolution: >> Keywords: | >> Implementation: Unnecessary >> Author_bill_gropp: 0 | >> Author_rich_graham: 0 >> Author_adam_moody: 0 | >> Author_torsten_hoefler: 0 >> Author_dick_treumann: 0 | >> Author_jesper_larsson_traeff: 0 >> Author_george_bosilca: 1 | >> Author_david_solt: 0 >> Author_bronis_de_supinski: 0 | >> Author_rajeev_thakur: 0 >> Author_jeff_squyres: 0 | >> Author_alexander_supalov: 0 >> Author_rolf_rabenseifner: 0 | >> ------------------------------------------------- >> +-------------------------- >> Changes (by RolfRabenseifner): >> >> * priority: Waiting for PDF reviews => Ticket complete >> * status: closed => reopened >> * resolution: Text committed => >> >> >> Comment: >> >> Sory Jeff, our rule is open + ticket complete. >> >> Please see your own diagram. >> >> -- >> Ticket URL: >><https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/146#comment:26 >> > >> MPI Forum <https://svn.mpi-forum.org/> >> MPI Forum >> > > > -- > Jeff Squyres > jsquyres_at_[hidden] Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner_at_[hidden] 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) *

1 0

Re: [Mpi-22] [MPI Forum] #146: Fix MPI_INIT description text
by unknown＠example.com 01 Sep '09

01 Sep '09

Rolf -- I don't understand. I *did* follow my own state diagram -- ticket #146 is done done done, so I resolved it in the "Ticket Complete" state. I attached the PDF for reference. You're saying that the ticket should be left open and just have its priority changed to "Text Committed". I disagree. The state diagram has the ticket state go to "Text Committed", not "Text Committed" followed by resolving it in "Text Committed" *again*. It was definitely my intention when making this state diagram that tickets should be *resolved* (i.e., closed) in with the "Text Committed" state. If there is a "Text Committed" priority (and I see that there is), I believe that to be a mistake. To be clear -- "Withdrawn", "Rejected", and "Deferred" are resolution (closing) states. They are not priorities. "Text Committed" was intended to be the same. Perhaps I'm just grouchy from jet lag. :-) On Sep 1, 2009, at 3:40 PM, MPI Forum wrote: > #146: Fix MPI_INIT description text > ------------------------------------------------- > +-------------------------- > Reporter: jsquyres > | Owner: jsquyres > Type: Text (only) changes > | Status: reopened > Priority: Ticket complete > | Milestone: 2009/07/27 Chicago > Version: MPI 2.2 > | Resolution: > Keywords: | > Implementation: Unnecessary > Author_bill_gropp: 0 | > Author_rich_graham: 0 > Author_adam_moody: 0 | > Author_torsten_hoefler: 0 > Author_dick_treumann: 0 | > Author_jesper_larsson_traeff: 0 > Author_george_bosilca: 1 | > Author_david_solt: 0 > Author_bronis_de_supinski: 0 | > Author_rajeev_thakur: 0 > Author_jeff_squyres: 0 | > Author_alexander_supalov: 0 > Author_rolf_rabenseifner: 0 | > ------------------------------------------------- > +-------------------------- > Changes (by RolfRabenseifner): > > * priority: Waiting for PDF reviews => Ticket complete > * status: closed => reopened > * resolution: Text committed => > > > Comment: > > Sory Jeff, our rule is open + ticket complete. > > Please see your own diagram. > > -- > Ticket URL: <https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/146#comment:26 > > > MPI Forum <https://svn.mpi-forum.org/> > MPI Forum > -- Jeff Squyres jsquyres_at_[hidden] *

1 0