June 2011 - mpiwg-tools - mailman3.cels.anl.gov

Re: [Mpi3-tools] MPI function symbol naming convention for tools
by Martin Schulz 20 Jun '11

20 Jun '11

Hi Craig, all, (I cc-ed the tools group on this as well - this is regarding the issue how the profiling interface will work with the proposed new Fortran bindings): On Jun 20, 2011, at 10:15 AM, Rasmussen, Craig E wrote: > I've talked with Jeff Squyres and we come up with naming convention that we'd like to propose for tools to have access to Fortran. > > I believe we need four different symbol naming conventions. Consider MPI_Recv(), the four functions are > > 1. mpi_recv_f > - This function implements (in C or Fortran) MPI_Recv with integer handles and the choice buffer passed by address. Does this mean the existing mpi_recv binding will go away? This would break any existing tool. Can we keep the old naming scheme for the old interface? Or is this intended as an additional symbol? > > 2. mpi_recv_f_nostatus > - This function implements (in C or Fortran) MPI_Recv with integer handles without a status variable and the choice buffer passed by address. Is this just for the new bindings? > > 3. mpi_recv_f08 > - This function implements (in C or Fortran) MPI_Recv with integer handles and the choice buffer passed via an array descriptor. > > 4. mpi_recv_f_nostatus > - This function implements (in C or Fortran) MPI_Recv with integer handles without a status variable and the choice buffer passed via an array descriptor. I assume you mean mpi_recv_f08_nostatus? > > Note that these functions only need to support integer handles because of the way we (thanks to Rolf) have chosen to define the handle types in the mpi_f08 module. > > Please comment on this naming convention. Martin > > -craig > > > > ________________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0

Re: [Mpi3-tools] Is there a call this AM?
by Dave Goodell 20 Jun '11

20 Jun '11

Here's the phone # info from the webex website if that's not coming up for you for some reason: Meeting number: 203 062 084 Meeting password: tools Audio conference: To receive a call back, provide your phone number when you join the meeting, or call the number below and enter the access code. Call-in toll-free number (US/Canada): +1-866-432-9903 Call-in toll number (US/Canada): +1-408-525-6800 Show toll-free dialing restrictions Access code: 203 062 084 -Dave On Jun 20, 2011, at 10:02 AM CDT, Mohror, Kathryn wrote: > Hi Marty, > > There's a link on the wiki page: https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/MPI3Tools > > Kathryn > >> -----Original Message----- >> From: mpi3-tools-bounces(a)lists.mpi-forum.org [mailto:mpi3-tools- >> bounces(a)lists.mpi-forum.org] On Behalf Of Marty Itzkowitz >> Sent: Monday, June 20, 2011 7:53 AM >> To: Martin Schulz >> Cc: mpitools >> Subject: [Mpi3-tools] Is there a call this AM? >> >> Please send the call info.... >> >> Thanks, >> Marty >> >> _______________________________________________ >> Mpi3-tools mailing list >> Mpi3-tools(a)lists.mpi-forum.org >> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools > > _______________________________________________ > Mpi3-tools mailing list > Mpi3-tools(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools

1 0

Re: [Mpi3-tools] Is there a call this AM?
by Mohror, Kathryn 20 Jun '11

20 Jun '11

Hi Marty, There's a link on the wiki page: https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/MPI3Tools Kathryn > -----Original Message----- > From: mpi3-tools-bounces(a)lists.mpi-forum.org [mailto:mpi3-tools- > bounces(a)lists.mpi-forum.org] On Behalf Of Marty Itzkowitz > Sent: Monday, June 20, 2011 7:53 AM > To: Martin Schulz > Cc: mpitools > Subject: [Mpi3-tools] Is there a call this AM? > > Please send the call info.... > > Thanks, > Marty > > _______________________________________________ > Mpi3-tools mailing list > Mpi3-tools(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools

1 0

[Mpi3-tools] Is there a call this AM?
by Marty Itzkowitz 20 Jun '11

20 Jun '11

Please send the call info.... Thanks, Marty

1 0

Re: [Mpi3-tools] New draft version of MPI_T
by Martin Schulz 17 Jun '11

17 Jun '11

Hi Jeff, On Jun 17, 2011, at 11:03 AM, Jeff Squyres wrote: > I'm sorry; I won't be able to review this before Monday. :-( That's fine - anybody who has time to look at it, that would be great, but we have the next two weeks to finish our internal review and to have the official copy for the reading ready. The deadline is 7/4. It would be good to get larger comments (if there are any, I hope not) throughout next week, though, so I have some time to integrate them. This Monday I would like to mainly cover the open questions listed in the email. These are the things I came across going through the various comments from last time and from editing the document. Thanks, Martin > > > On Jun 17, 2011, at 4:02 AM, Martin Schulz wrote: > >> Hi all, >> >> I have uploaded a new draft version of MPI_T, hopefully >> addressing all/most of the comments from last time, to the >> Wiki at: https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/MPI3Tools/mpit >> >> I have also attached the document. >> >> The color markings are now relative to the approved and >> current version of MPI. >> >> The following issues are still open or at least should be >> discussed by the group and it would be great if we could >> get through those next Monday during our TelCon: >> >> - Rolf requested to sort the arguments for each function to >> first IN then OUT >> - this causes problems with string returns (would split >> buffer and length) >> - unclear what the right label for Recv is >> >> - I added types to each prototype definition, but it was unclear >> to me what to put for the character array. >> >> - Should we make integers unsigned in the interface? >> >> - Should we ad const where appropriate? >> >> - Are we still OK with the text on the associating MPI objects >> with variables? I am not sure anymore what the result of the >> discussion was last time. >> >> - What other changes in the rest of the standard document >> do we still need to make - so far we only touched Init and >> Finalize, as well as the language appendix. Anything else? >> >> Also, it would be great if as many of you as possible could >> read the document for correctness and consistency. Feel >> free to send me corrections or, if you have too many, let >> me know and I'll send you the TeX file to edit. We have >> about two weeks now to get the document to a state so it >> can pass the reading and I hope this is achievable. >> >> If I missed any issue, please speak up. >> >> Thanks, >> >> Martin >> >> <mpi-report.pdf><Mail Attachment.txt><Mail Attachment.txt> > > > -- > Jeff Squyres > jsquyres(a)cisco.com > For corporate legal information go to: > http://www.cisco.com/web/about/doing_business/legal/cri/ > > _______________________________________________ > Mpi3-tools mailing list > Mpi3-tools(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools ________________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0

Re: [Mpi3-tools] MPI_T Status / Next Meeting
by Martin Schulz 17 Jun '11

17 Jun '11

On Jun 17, 2011, at 11:04 AM, Jeff Squyres wrote: > Martin -- > > Just to be sure, before I setup the webex -- you're asking for a webex on Monday, June 20, at 11am US Eastern time, right? Yes, that's right. Thanks, Martin > > > On Jun 16, 2011, at 4:45 AM, Martin Schulz wrote: > >> Hi all, >> >> I just wanted to send an update - I am currently rewriting the >> document based on the feedback from the last forum. This >> included a lot of cleaning of the TeX code, which took much >> longer than expected. The actual MPI_T text is basically >> done (and in SVN), but I still need to add the necessary >> appendices. I should have this done by tomorrow and I'll >> send this out then. >> >> As mentioned in previous emails (a while ago), we should >> meet again next Monday at the usual time to discuss the >> current state and to go over the remaining open issues. This >> also marks the four week mark before the forum, which then >> gives us, as discussed last time in San Jose, two weeks for >> an internal review before the official two week deadline for >> the reading at the July forum. >> >> I hope this works for most, >> >> Martin >> >> PS: Jeff, could you please create the meeting in WebEx? >> Thanks! >> >> ________________________________________________________________________ >> Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm >> CASC @ Lawrence Livermore National Laboratory, Livermore, USA >> >> >> >> _______________________________________________ >> Mpi3-tools mailing list >> Mpi3-tools(a)lists.mpi-forum.org >> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools > > > -- > Jeff Squyres > jsquyres(a)cisco.com > For corporate legal information go to: > http://www.cisco.com/web/about/doing_business/legal/cri/ > > > _______________________________________________ > Mpi3-tools mailing list > Mpi3-tools(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools ________________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0

Re: [Mpi3-tools] MPI_T Status / Next Meeting
by Jeff Squyres 17 Jun '11

17 Jun '11

Martin -- Just to be sure, before I setup the webex -- you're asking for a webex on Monday, June 20, at 11am US Eastern time, right? On Jun 16, 2011, at 4:45 AM, Martin Schulz wrote: > Hi all, > > I just wanted to send an update - I am currently rewriting the > document based on the feedback from the last forum. This > included a lot of cleaning of the TeX code, which took much > longer than expected. The actual MPI_T text is basically > done (and in SVN), but I still need to add the necessary > appendices. I should have this done by tomorrow and I'll > send this out then. > > As mentioned in previous emails (a while ago), we should > meet again next Monday at the usual time to discuss the > current state and to go over the remaining open issues. This > also marks the four week mark before the forum, which then > gives us, as discussed last time in San Jose, two weeks for > an internal review before the official two week deadline for > the reading at the July forum. > > I hope this works for most, > > Martin > > PS: Jeff, could you please create the meeting in WebEx? > Thanks! > > ________________________________________________________________________ > Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm > CASC @ Lawrence Livermore National Laboratory, Livermore, USA > > > > _______________________________________________ > Mpi3-tools mailing list > Mpi3-tools(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools -- Jeff Squyres jsquyres(a)cisco.com For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/

1 0

Re: [Mpi3-tools] New draft version of MPI_T
by Jeff Squyres 17 Jun '11

17 Jun '11

I'm sorry; I won't be able to review this before Monday. :-( On Jun 17, 2011, at 4:02 AM, Martin Schulz wrote: > Hi all, > > I have uploaded a new draft version of MPI_T, hopefully > addressing all/most of the comments from last time, to the > Wiki at: https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/MPI3Tools/mpit > > I have also attached the document. > > The color markings are now relative to the approved and > current version of MPI. > > The following issues are still open or at least should be > discussed by the group and it would be great if we could > get through those next Monday during our TelCon: > > - Rolf requested to sort the arguments for each function to > first IN then OUT > - this causes problems with string returns (would split > buffer and length) > - unclear what the right label for Recv is > > - I added types to each prototype definition, but it was unclear > to me what to put for the character array. > > - Should we make integers unsigned in the interface? > > - Should we ad const where appropriate? > > - Are we still OK with the text on the associating MPI objects > with variables? I am not sure anymore what the result of the > discussion was last time. > > - What other changes in the rest of the standard document > do we still need to make - so far we only touched Init and > Finalize, as well as the language appendix. Anything else? > > Also, it would be great if as many of you as possible could > read the document for correctness and consistency. Feel > free to send me corrections or, if you have too many, let > me know and I'll send you the TeX file to edit. We have > about two weeks now to get the document to a state so it > can pass the reading and I hope this is achievable. > > If I missed any issue, please speak up. > > Thanks, > > Martin > > <mpi-report.pdf><Mail Attachment.txt><Mail Attachment.txt> -- Jeff Squyres jsquyres(a)cisco.com For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/

1 0

[Mpi3-tools] New draft version of MPI_T
by Martin Schulz 17 Jun '11

17 Jun '11

Hi all, I have uploaded a new draft version of MPI_T, hopefully addressing all/most of the comments from last time, to the Wiki at: https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/MPI3Tools/mpit I have also attached the document. The color markings are now relative to the approved and current version of MPI. The following issues are still open or at least should be discussed by the group and it would be great if we could get through those next Monday during our TelCon: - Rolf requested to sort the arguments for each function to first IN then OUT - this causes problems with string returns (would split buffer and length) - unclear what the right label for Recv is - I added types to each prototype definition, but it was unclear to me what to put for the character array. - Should we make integers unsigned in the interface? - Should we ad const where appropriate? - Are we still OK with the text on the associating MPI objects with variables? I am not sure anymore what the result of the discussion was last time. - What other changes in the rest of the standard document do we still need to make - so far we only touched Init and Finalize, as well as the language appendix. Anything else? Also, it would be great if as many of you as possible could read the document for correctness and consistency. Feel free to send me corrections or, if you have too many, let me know and I'll send you the TeX file to edit. We have about two weeks now to get the document to a state so it can pass the reading and I hope this is achievable. If I missed any issue, please speak up. Thanks, Martin ________________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0

[Mpi3-tools] MPI_T Status / Next Meeting
by Martin Schulz 16 Jun '11

16 Jun '11

Hi all, I just wanted to send an update - I am currently rewriting the document based on the feedback from the last forum. This included a lot of cleaning of the TeX code, which took much longer than expected. The actual MPI_T text is basically done (and in SVN), but I still need to add the necessary appendices. I should have this done by tomorrow and I'll send this out then. As mentioned in previous emails (a while ago), we should meet again next Monday at the usual time to discuss the current state and to go over the remaining open issues. This also marks the four week mark before the forum, which then gives us, as discussed last time in San Jose, two weeks for an internal review before the official two week deadline for the reading at the July forum. I hope this works for most, Martin PS: Jeff, could you please create the meeting in WebEx? Thanks! ________________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0