April 2009 - mpiwg-tools - mailman3.cels.anl.gov

Re: [Mpi3-tools] MPI handles interface
by Bronis R. de Supinski 16 Apr '09

16 Apr '09

Jeff: Hmm. My expectation was there would be query functions for individual fields, a la MPI_Get_count. I guess I don't have a strong feeling about it -- when you want only one field, you will pay a somewhat significant cost with your approach but your approach will be an even bigger win when you want all of them. It may be harder for an implementation to add fields with your approach, I don't know... Bronis On Thu, 16 Apr 2009, Jeff Squyres wrote: > I took a first shot at removing the One Big Struct-ness (OBS) of the > MPI handles proposal: I removed the OBS from communicators and > replaced it with several "query" functions. A few structs still > remain -- those that are all pointers, for example, or those that will > likely otherwise not be problematic (e.g., not have holes). > > I only updated the communicators section in the header file; I did not > yet update the other 3 handle types that were prototyped (MPI_Status, > MPI_Request, MPI_Errhandler). I also did not yet update the > implementation to match, so it won't compile. I wanted to get some > feedback from the group before continuing. > > What about enums -- will those be problematic? I tended to prefer > enums in most places (vs. #define's) because a debugger can show those > value names (when debugging this software itself -- not necessarily > when in general use by a debugger for an MPI job). If enums are going > to be problematic, I can switch to #defines -- but it would be nice > not to, if possible. > > The communicator "query" functions that I created start around line 649: > > http:// www. open-mpi.org/hg/hgwebdir.cgi/jsquyres/mpi3-tools-handles/file/4c5c1e7c0ef1/… > > Could I get some comments on the group as to the general style of what > I did? > > If the web interface is too klunky for you, you can clone the repo > with Mercurial: > > hg clone http:// www. open-mpi.org/hg/hgwebdir.cgi/jsquyres/mpi3-tools-handles > > The individual file to look at is ompi/debuggers/mpihandles_interface.h. > > -- > Jeff Squyres > Cisco Systems > > _______________________________________________ > 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] MPI handles interface
by Jeff Squyres 16 Apr '09

16 Apr '09

I took a first shot at removing the One Big Struct-ness (OBS) of the MPI handles proposal: I removed the OBS from communicators and replaced it with several "query" functions. A few structs still remain -- those that are all pointers, for example, or those that will likely otherwise not be problematic (e.g., not have holes). I only updated the communicators section in the header file; I did not yet update the other 3 handle types that were prototyped (MPI_Status, MPI_Request, MPI_Errhandler). I also did not yet update the implementation to match, so it won't compile. I wanted to get some feedback from the group before continuing. What about enums -- will those be problematic? I tended to prefer enums in most places (vs. #define's) because a debugger can show those value names (when debugging this software itself -- not necessarily when in general use by a debugger for an MPI job). If enums are going to be problematic, I can switch to #defines -- but it would be nice not to, if possible. The communicator "query" functions that I created start around line 649: http://www.open-mpi.org/hg/hgwebdir.cgi/jsquyres/mpi3-tools-handles/file/4c… Could I get some comments on the group as to the general style of what I did? If the web interface is too klunky for you, you can clone the repo with Mercurial: hg clone http://www.open-mpi.org/hg/hgwebdir.cgi/jsquyres/mpi3-tools-handles The individual file to look at is ompi/debuggers/mpihandles_interface.h. -- Jeff Squyres Cisco Systems

1 0

[Mpi3-tools] Notes from MPI Forum WG Meeting
by Martin Schulz 09 Apr '09

09 Apr '09

Hi all, The notes from our WG discussions at the MPI Forum are now online on the Wiki. We focused exclusively on the debugging APIs and basically reached a state where we can move forward with writing each API up. Our next TelCon will be in two weeks on 4/20 at the usual time. The agenda will be first a quick report from this week's meeting with a chance for people who couldn't make it to comment, a short discussion on whether we should include the topic of PMI (Process Management Interface) in this WG, and (for a majority of the time) a discussion on the proposal for an MPI state/information interface. Also, please feel free to comment on the meeting minutes and the debugging APIs before the TelCon via email. Martin _______________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulz6 CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0

[Mpi3-tools] TelCon from MPI Forum Meeting soon
by Martin Schulz 07 Apr '09

07 Apr '09

Hi all, We are running a bit late at the MPI Forum and will dial in for the MPI tools WG as soon as possible (hopefully not more than 5-10 minutes). Martin _______________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulz6 CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0

Re: [Mpi3-tools] No TelCon / WG discussions at MPI-Forum Meeting
by John DelSignore 07 Apr '09

07 Apr '09

Hi Martin, I'm on, and so are a few others. We're waiting for you guys... Cheers, John D. Martin Schulz wrote: > Hi all, > > Just as a reminder: we will not have a MPI tools WG TelCon this > week due to the face to face meeting at the MPI Forum in Chicago. > The next regular call is scheduled for 4/20. > > However, there will a WG meeting on Tuesday afternoon at 2pm CT > to discuss the forum feedback on the proposed handle and message > queue APIs. We will have the ability to setup a speakerphone so > that people who can't make it to Chicago can call in. Here is > the dial-in information: > >> Toll Free Dial In Number for Offsite Participants: 866-914-3976 >> International Access/Caller Paid Dial In Number: 925-424-8105 >> Dial In Number for LLNL Onsite Participants: 4-8105 >> >> Access Security Code: 860609 # > > I hope many of you can make this. > > Martin > > > > _______________________________________________________________________ > Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulz6 > 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 > >

1 0

[Mpi3-tools] No TelCon / WG discussions at MPI-Forum Meeting
by Martin Schulz 05 Apr '09

05 Apr '09

Hi all, Just as a reminder: we will not have a MPI tools WG TelCon this week due to the face to face meeting at the MPI Forum in Chicago. The next regular call is scheduled for 4/20. However, there will a WG meeting on Tuesday afternoon at 2pm CT to discuss the forum feedback on the proposed handle and message queue APIs. We will have the ability to setup a speakerphone so that people who can't make it to Chicago can call in. Here is the dial-in information: > Toll Free Dial In Number for Offsite Participants: 866-914-3976 > International Access/Caller Paid Dial In Number: 925-424-8105 > Dial In Number for LLNL Onsite Participants: 4-8105 > > Access Security Code: 860609 # I hope many of you can make this. Martin _______________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulz6 CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0

[Mpi3-tools] Message Queue Display Slides for MPI forum meeting
by Chris Gottbrath 03 Apr '09

03 Apr '09

MPI3 Tools Working Group Folks, Here are the slides that I said I'd provide for Jeff to talk through at the MPI forum meeting next week. ********** This transmission contains confidential and/or legally privileged information from TotalView Technologies intended only for the use of the individual(s) to which it is addressed. If you are not the intended recipient, you are hereby notified that any disclosure, copying or distribution of this information or the taking of any action in reliance on the contents of this transmission is strictly prohibited. If you have received this transmission in error, please notify us immediately. ********** Cheers, Chris Chris Gottbrath, 508-652-7735 or 774-270-3155 Director of Product Management, TotalView Technologies chris.gottbrath(a)totalviewtech.com -- Learn how to radically simplify your debugging: http://www.totalviewtech.com/support/white_papers.html?id=163 >

1 0