June 2013 - mpiwg-tools - mailman3.cels.anl.gov

Re: [Mpi3-tools] Tools WG webex: tomorrow!
by Rolf Rabenseifner 28 Jun '13

28 Jun '13

Jeff and Craig, does the Tools Fortran wrapper include calls to the C f2c (and c2f) function through a Fortran BIND(C) interface definition that makes these functions callable from the Fortran written tools wrappers? If yes, then we should continue to think about including these Fortran callable f2c (and c2f) functions as official Fortran Interfaces into the MPI library, because the BIND(C) Fortran interface for these functions seems to me not portable if these C functions are implemented as macros. And for portable programming of calling the f2c functions, we also Need types in Fortran that correspond to the C types MPI_Comm, etc., i.e., the concept of the standardized constants MPI_C_COMM_KIND, etc., to get the type definition INTEGER(KIND=MPI_C_COMM_KIND) in Fortran as equivalent to MPI_Comm in C. To minimize the tools' instrumentation overhead, such conversions should be possible as macros, especially if the conversion is the identity. Implementing such in Fortran functions inside of a Fortran module (i.e., with CONTAINS), allows that the Compiler is inlining the conversion and removing it in the case of the identity function. Best regards Rolf ----- Original Message ----- > From: "Jeff Squyres (jsquyres)" <jsquyres(a)cisco.com> > To: "MPI-3 WG" <mpi3-tools(a)lists.mpi-forum.org> > Sent: Thursday, June 27, 2013 8:23:30 PM > Subject: Re: [Mpi3-tools] Tools WG webex: tomorrow! > The notes from todays meeting are now posted on the wiki: > > https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/MPI3Tools/notes-2013-06-27 > > Enjoy! > > > On Jun 26, 2013, at 4:54 PM, Jeff Squyres <jsquyres(a)cisco.com> wrote: > > > This is your friendly neighborhood MPI-3 Tools Working Group > > teleconf reminder for tomorrow: > > > > Thursday, June 27, 2013, 11am US Eastern time > > Webex link: > > https://cisco.webex.com/ciscosales/j.php?ED=217124297&UID=0&PW=NMDEyN2UwZmM… > > Webex password: tools > > > > Wiki link: > > https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/MPI3Tools > > > > Agenda: > > > > MPI meeting results > > - Formal reading of MPI queue side document: went well > > https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/375 > > Anh to discuss details > > - Voted on MPI_T constants erroneously listed for Fortran: passed > > https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/354 > > Need to commit to latex and get PDF reviews > > - Fortran debacle: you've all been doing it wrong (for ~15 years) > > - Thoughts on new profiling interface for MPI-4 > > - ...anything else? > > > > -- > > Jeff Squyres > > jsquyres(a)cisco.com > > For corporate legal information go to: > > http://www.cisco.com/web/about/doing_business/legal/cri/ > > > > > -- > 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 -- Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner(a)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: Room 1.307)

1 0

Re: [Mpi3-tools] Tools WG webex: tomorrow!
by Kathryn Mohror 27 Jun '13

27 Jun '13

Thanks for the very comprehensive notes! On Jun 27, 2013, at 11:23 AM, Jeff Squyres (jsquyres) <jsquyres(a)cisco.com> wrote: > The notes from todays meeting are now posted on the wiki: > > https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/MPI3Tools/notes-2013-06-27 > > Enjoy! > > > On Jun 26, 2013, at 4:54 PM, Jeff Squyres <jsquyres(a)cisco.com> wrote: > >> This is your friendly neighborhood MPI-3 Tools Working Group teleconf reminder for tomorrow: >> >> Thursday, June 27, 2013, 11am US Eastern time >> Webex link: https://cisco.webex.com/ciscosales/j.php?ED=217124297&UID=0&PW=NMDEyN2UwZmM… >> Webex password: tools >> >> Wiki link: https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/MPI3Tools >> >> Agenda: >> >> MPI meeting results >> - Formal reading of MPI queue side document: went well >> https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/375 >> Anh to discuss details >> - Voted on MPI_T constants erroneously listed for Fortran: passed >> https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/354 >> Need to commit to latex and get PDF reviews >> - Fortran debacle: you've all been doing it wrong (for ~15 years) >> - Thoughts on new profiling interface for MPI-4 >> - ...anything else? >> >> -- >> Jeff Squyres >> jsquyres(a)cisco.com >> For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/ >> > > > -- > 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 ______________________________________________________________ Kathryn Mohror, kathryn(a)llnl.gov, http://people.llnl.gov/mohror1 CASC @ Lawrence Livermore National Laboratory, Livermore, CA, USA

1 0

Re: [Mpi3-tools] Tools WG webex: tomorrow!
by Jeff Squyres (jsquyres) 27 Jun '13

27 Jun '13

The notes from todays meeting are now posted on the wiki: https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/MPI3Tools/notes-2013-06-27 Enjoy! On Jun 26, 2013, at 4:54 PM, Jeff Squyres <jsquyres(a)cisco.com> wrote: > This is your friendly neighborhood MPI-3 Tools Working Group teleconf reminder for tomorrow: > > Thursday, June 27, 2013, 11am US Eastern time > Webex link: https://cisco.webex.com/ciscosales/j.php?ED=217124297&UID=0&PW=NMDEyN2UwZmM… > Webex password: tools > > Wiki link: https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/MPI3Tools > > Agenda: > > MPI meeting results > - Formal reading of MPI queue side document: went well > https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/375 > Anh to discuss details > - Voted on MPI_T constants erroneously listed for Fortran: passed > https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/354 > Need to commit to latex and get PDF reviews > - Fortran debacle: you've all been doing it wrong (for ~15 years) > - Thoughts on new profiling interface for MPI-4 > - ...anything else? > > -- > Jeff Squyres > jsquyres(a)cisco.com > For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/ > -- 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] Tools WG webex: tomorrow!
by Jeff Squyres (jsquyres) 26 Jun '13

26 Jun '13

This is your friendly neighborhood MPI-3 Tools Working Group teleconf reminder for tomorrow: Thursday, June 27, 2013, 11am US Eastern time Webex link: https://cisco.webex.com/ciscosales/j.php?ED=217124297&UID=0&PW=NMDEyN2UwZmM… Webex password: tools Wiki link: https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/MPI3Tools Agenda: MPI meeting results - Formal reading of MPI queue side document: went well https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/375 Anh to discuss details - Voted on MPI_T constants erroneously listed for Fortran: passed https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/354 Need to commit to latex and get PDF reviews - Fortran debacle: you've all been doing it wrong (for ~15 years) - Thoughts on new profiling interface for MPI-4 - ...anything else? -- 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] Updates from the Forum meeting
by Mohror, Kathryn 07 Jun '13

07 Jun '13

Hi all, A few things from the forum meeting: 1) Our next telecon will be on June 27. We elected to skip the phone call immediately following the forum. 2) I put some notes from the Forum meeting up on the wiki: https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/MPI3Tools Scroll down to the "Face to Face Meetings" section to find them. 3) I'll send out the results of our discussions on changes for the MPI_T interface that we worked on in the morning session on Wednesday likely next week. Kathryn ______________________________________________________________ Kathryn Mohror, kathryn(a)llnl.gov<mailto:[email protected]>, http://people.llnl.gov/mohror1 CASC @ Lawrence Livermore National Laboratory, Livermore, CA, USA

1 0

[Mpi3-tools] Fwd: CDesc
by Jeff Squyres (jsquyres) 05 Jun '13

05 Jun '13

Here's the descriptor struct that we discussed today in the meeting. Begin forwarded message: > From: Craig Rasmussen <rasmus(a)uoregon.edu> > Subject: CDesc > Date: June 5, 2013 2:37:12 PM PDT > To: Jeff Squyres <jsquyres(a)cisco.com> > > > Craig Rasmussen > CAS Scientific Programmer > rasmus(a)uoregon.edu > > > -- 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] Fortran Wrappers in Fortran?
by Marc-Andre Hermanns 04 Jun '13

04 Jun '13

Hi Rolf, comments inline. >> with the Fortran wrappers discussion taking off tomorrow, I wanted to >> explicitly ask about something that has spooked around in my head for >> some time, and I am unsure whether this is viable or not. >> >> It does not solve the name mangling is general, but maybe it is an >> easy approach in some cases, such as simple preinstrumented libraries. >> >> The new Fortran interface comes only with Fortran 2008+ compilers, >> because older compilers don't know all the new magic that is necessary >> to make it work, right? This would mean that the compiler also has the >> C interface capabilities of Fortran 2003. > > The new mpi_f08 module does not require Fortran 2008 + TS 29113. > Preliminary implementations (with MPI_SUBARRAYS_SUPPORTED=.FALSE.) > are possible with Fortran 2003. Same is valid for the new > requirements on the mpi module (e.g., compile time arument checking). Fortran 2003 would be sufficient for my idea, so that should be ok. >> Performance tools like Scalasca or Score-P, usually call >> straight-forward C routines in the wrappers to generate the >> measurement >> data and maybe do some simple computations in between. Usually, they >> just pass the initial parameters through. >> >> As most of our wrapper code is generated anyway. Would it be a viable >> approach to >> 1) generate a C interface definition for the parts of the measurement >> system in question to call the measurement routines from Fortran, and >> 2) generate Fortran wrappers that directly call the measurement system >> through the generated interface and then call the Fortran version of >> "PMPI_Func"? >> >> Would the Fortran compiler then take care of all the name mangling >> issues in this case? >> >> Using Fortran wrappers calling the Fortran PMPI routines would also >> make handle conversion (to call the C MPI wrapper) unnecessary and >> could save overhead. > > I'm not sure whether I understood your approach. > For me, it looks like that you want to go a similar way as Bill Long > who has Fortran-written wrappers that directly call the > MPI C routines. For this, he defined a module with the Interfaces > of the MPI C routines in Fortran 2003 syntax with BIND(C). > It is similar, but not quite the same. I also propose to have pure Fortran wrappers, but instead of calling the MPI C routines, I would call the Fortran PMPI functions. I would define a BIND(C) interface for my measurement system and call those routines in the wrapper. > He tries to get his Fortran MPI wrappers fully portable, > though the tools intercept only at the C Interface. My proposal would make tools intercept on both the C and the Fortran interface. As a the majority of the wrappers is generated, I think this could still be maintainable. > See also the Details in c20.comments.txt attached to a previous mail. I'll check for that (I may have to search the online archive, though). Cheers, Marc-Andre -- Marc-Andre Hermanns German Research School for Simulation Sciences GmbH c/o Laboratory for Parallel Programming 52062 Aachen | Germany Tel +49 241 80 99753 Fax +49 241 80 6 99753 Web www.grs-sim.de Members: Forschungszentrum Jülich GmbH | RWTH Aachen University Registered in the commercial register of the local court of Düren (Amtsgericht Düren) under registration number HRB 5268 Registered office: Jülich Executive board: Prof. Marek Behr, Ph.D | Prof. Dr. Sebastian M. Schmidt

1 0

Re: [Mpi3-tools] Fortran Wrappers in Fortran?
by Rolf Rabenseifner 04 Jun '13

04 Jun '13

Some answers inline. ----- Original Message ----- > From: "Marc-Andre Hermanns" <m.a.hermanns(a)grs-sim.de> > To: mpi3-tools(a)lists.mpi-forum.org > Sent: Tuesday, June 4, 2013 5:43:00 PM > Subject: [Mpi3-tools] Fortran Wrappers in Fortran? > Dear all, > > with the Fortran wrappers discussion taking off tomorrow, I wanted to > explicitly ask about something that has spooked around in my head for > some time, and I am unsure whether this is viable or not. > > It does not solve the name mangling is general, but maybe it is an > easy approach in some cases, such as simple preinstrumented libraries. > > The new Fortran interface comes only with Fortran 2008+ compilers, > because older compilers don't know all the new magic that is necessary > to make it work, right? This would mean that the compiler also has the > C interface capabilities of Fortran 2003. The new mpi_f08 module does not require Fortran 2008 + TS 29113. Preliminary implementations (with MPI_SUBARRAYS_SUPPORTED=.FALSE.) are possible with Fortran 2003. Same is valid for the new requirements on the mpi module (e.g., compile time arument checking). > Performance tools like Scalasca or Score-P, usually call > straight-forward C routines in the wrappers to generate the > measurement > data and maybe do some simple computations in between. Usually, they > just pass the initial parameters through. > > As most of our wrapper code is generated anyway. Would it be a viable > approach to > 1) generate a C interface definition for the parts of the measurement > system in question to call the measurement routines from Fortran, and > 2) generate Fortran wrappers that directly call the measurement system > through the generated interface and then call the Fortran version of > "PMPI_Func"? > > Would the Fortran compiler then take care of all the name mangling > issues in this case? > > Using Fortran wrappers calling the Fortran PMPI routines would also > make handle conversion (to call the C MPI wrapper) unnecessary and > could save overhead. I'm not sure whether I understood your approach. For me, it looks like that you want to go a similar way as Bill Long who has Fortran-written wrappers that directly call the MPI C routines. For this, he defined a module with the Interfaces of the MPI C routines in Fortran 2003 syntax with BIND(C). He tries to get his Fortran MPI wrappers fully portable, though the tools intercept only at the C Interface. See also the Details in c20.comments.txt attached to a previous mail. Best regards Rolf > Cheers, > Marc-Andre > -- > Marc-Andre Hermanns > German Research School for > Simulation Sciences GmbH > c/o Laboratory for Parallel Programming > 52062 Aachen | Germany > > Tel +49 241 80 99753 > Fax +49 241 80 6 99753 > Web www.grs-sim.de > > Members: Forschungszentrum Jülich GmbH | RWTH Aachen University > Registered in the commercial register of the local court of > Düren (Amtsgericht Düren) under registration number HRB 5268 > Registered office: Jülich > Executive board: Prof. Marek Behr, Ph.D | Prof. Dr. Sebastian M. > Schmidt > _______________________________________________ > Mpi3-tools mailing list > Mpi3-tools(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools -- Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner(a)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: Room 1.307)

1 0

[Mpi3-tools] Fortran Wrappers in Fortran?
by Marc-Andre Hermanns 04 Jun '13

04 Jun '13

Dear all, with the Fortran wrappers discussion taking off tomorrow, I wanted to explicitly ask about something that has spooked around in my head for some time, and I am unsure whether this is viable or not. It does not solve the name mangling is general, but maybe it is an easy approach in some cases, such as simple preinstrumented libraries. The new Fortran interface comes only with Fortran 2008+ compilers, because older compilers don't know all the new magic that is necessary to make it work, right? This would mean that the compiler also has the C interface capabilities of Fortran 2003. Performance tools like Scalasca or Score-P, usually call straight-forward C routines in the wrappers to generate the measurement data and maybe do some simple computations in between. Usually, they just pass the initial parameters through. As most of our wrapper code is generated anyway. Would it be a viable approach to 1) generate a C interface definition for the parts of the measurement system in question to call the measurement routines from Fortran, and 2) generate Fortran wrappers that directly call the measurement system through the generated interface and then call the Fortran version of "PMPI_Func"? Would the Fortran compiler then take care of all the name mangling issues in this case? Using Fortran wrappers calling the Fortran PMPI routines would also make handle conversion (to call the C MPI wrapper) unnecessary and could save overhead. Cheers, Marc-Andre -- Marc-Andre Hermanns German Research School for Simulation Sciences GmbH c/o Laboratory for Parallel Programming 52062 Aachen | Germany Tel +49 241 80 99753 Fax +49 241 80 6 99753 Web www.grs-sim.de Members: Forschungszentrum Jï¿½lich GmbH | RWTH Aachen University Registered in the commercial register of the local court of Dï¿½ren (Amtsgericht Dï¿½ren) under registration number HRB 5268 Registered office: Jï¿½lich Executive board: Prof. Marek Behr, Ph.D | Prof. Dr. Sebastian M. Schmidt

1 0

Re: [Mpi3-tools] [MPI3 Fortran] Fortran proposal w.r.t. BIND(C)/logical/linker symbols/etc. (take 2)
by Rolf Rabenseifner 03 Jun '13

03 Jun '13

Jeff, Bill, Craig, and all Fortran and tools people, I feel that c20.comments.txt is now a good basis for your discussions in San Jose. Best regards Rolf PS: c20 differs from c19 only in the note: - Note that the BIND(C) on the mpi_f08 handle types (MPI-3.0 p13:1-3) is kept as it is. -- Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner(a)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: Room 1.307)

1 0