January 2021 - mpi-forum - mailman3.cels.anl.gov

Re: [Mpi-forum] Big Fortran hole in MPI-4.0 Embiggening
by Jeff Hammond 11 Jan '21

11 Jan '21

Given how many implementations claimed MPI 3.0 support when their RMA implementations were garbage piles of incorrectness or pathetic performance, it borders on the absurd to think that logical rigor about feature support will stop anyone from claiming MPI 4.0 support. Sent from my iPhone > On Jan 11, 2021, at 7:22 AM, Wesley Bland via mpi-forum <mpi-forum(a)lists.mpi-forum.org> wrote: > > I agree with Jeff that this is a non-issue. Bill’s assertion that adding Fortran support for large count once you have C support has been true in MPICH development at least and no vender would drop anything just to put MPI 4.0 compatibility on a marketing slide. They’d just put it on the slide anyway. Every implementation out there has some sort of exception, intentional or otherwise, that makes them technically non-compliant. In the end, their users determine 1) whether they care enough to request the feature and 2) whether they care enough to switch to a competing library. > > I think the best way forward is to do something like what had been proposed when the embiggening/pythonization started and separate bindings from specification and just say that an MPI implementation has MPI 4.0 C bindings and MPI 3.1 F08 bindings and MPI 5.0 Python bindings. That means each binding can be updated in whatever way makes sense for them. For 4.0, I don’t think this is an issue that needs to be resolved. Us telling the implementations what they have to do will have essentially no impact anyway. > > Thanks, > Wes > >> On Jan 11, 2021, at 6:26 AM, Rolf Rabenseifner via mpi-forum <mpi-forum(a)lists.mpi-forum.org> wrote: >> >> Dear Jeff and all, >> >>> I think their users would be very unhappy. >> >> Yes, users can be unhappy and compute centers may definitely recommend >> an other MPI library - at least for any software development. >> >> I want to remember that MPI-3.1 requires for the mpi module: >> >> "Provide explicit interfaces according to the Fortran routine interface specications. >> This module therefore guarantees compile-time argument checking and allows positional >> and keyword-based argument lists. If an implementation is paired with a >> compiler that either does not support TYPE(*), DIMENSION(..) from TS 29113, or >> is otherwise unable to ignore the types of choice buffers, then the implementation must >> provide explicit interfaces only for MPI routines with no choice buer arguments. See >> Section 17.1.6 for more details." >> >> Although there are no more such compilers that do not provide one of the two methods, >> there are more than 5 years after MPI-3.1 still MPI libraries that do not >> provide keyword-based argument lists with the mpi module. >> And those libraries provide such support for the mpi_f08 module. >> This means, they prove that they are not MPI-3.1 compliant :-) >> >> >> Sometimes implementors ignore the goals and the wording of the MPI standard. >> >> >> And there is also no request that users want to have a less-quality >> mpi module definition. >> >> Best regards >> Rolf >> >> >> ----- Original Message ----- >>> From: "Main MPI Forum mailing list" <mpi-forum(a)lists.mpi-forum.org> >>> To: "Main MPI Forum mailing list" <mpi-forum(a)lists.mpi-forum.org> >>> Cc: "Jeff Squyres" <jsquyres(a)cisco.com> >>> Sent: Sunday, January 10, 2021 5:09:55 PM >>> Subject: Re: [Mpi-forum] Big Fortran hole in MPI-4.0 Embiggening >> >>> I can't imagine a vendor would: >>> >>> - support mpi_f08 >>> - then stop supporting mpi_f08 just so that they can get an "MPI-4.0" release >>> out >>> - then support mpi_f08 again >>> >>> I think their users would be very unhappy. >>> >>> >>> >>> >>> >>> On Jan 10, 2021, at 10:55 AM, William Gropp via mpi-forum < [ >>> mailto:[email protected] | mpi-forum(a)lists.mpi-forum.org ] > wrote: >>> >>> I agree with Dan that this is a big change from the RCM. Further, the approach >>> in MPI has always been to encourage the users to make it clear to the vendors >>> what is acceptable in implementations, especially implementation schedules. >>> Nothing in the standard prohibits implementors from continuing to provide MPI >>> 3.x implementations while they work to provide a full MPI 4.0 implementation. >>> The MPI forum has no enforcement power on the implementors, and I believe this >>> text is unnecessary and will not provide the guarantee that Rolf wants. >>> Further, frankly once the C embiggened interface is implemented, creating the >>> mpi_f08 version is relatively straightforward. >>> >>> Bill >>> >>> William Gropp >>> Director, NCSA >>> Thomas M. Siebel Chair in Computer Science >>> University of Illinois Urbana-Champaign >>> IEEE-CS President-Elect >>> >>> >>> >>> >>> >>> >>> On Jan 10, 2021, at 7:44 AM, HOLMES Daniel via mpi-forum < [ >>> mailto:[email protected] | mpi-forum(a)lists.mpi-forum.org ] > wrote: >>> >>> Hi Rolf, >>> >>> This is a (somewhat contrived, arguably) reason for taking another tiny step >>> towards removing the “mpif.h” method of Fortran support and pushing users and >>> implementations towards preferring the Fortran 2008 interface, which is a >>> direction of travel that I fully support. >>> >>> I think this might be seen as quite a big change for implementers, especially if >>> it were to occur between the RCM and FRM. >>> >>> Cheers, >>> Dan. >>> — >>> Dr Daniel Holmes PhD >>> Architect (HPC Research) >>> [ mailto:[email protected] | d.holmes(a)epcc.ed.ac.uk ] >>> Phone: +44 (0) 131 651 3465 >>> Mobile: +44 (0) 7940 524 088 >>> Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT >>> — >>> The University of Edinburgh is a charitable body, registered in Scotland, with >>> registration number SC005336. >>> — >>> >>> >>> >>> >>> On 10 Jan 2021, at 12:22, Rolf Rabenseifner via mpi-forum < [ >>> mailto:[email protected] | mpi-forum(a)lists.mpi-forum.org ] > wrote: >>> >>> This email was sent to you by someone outside the University. >>> You should only click on links or attachments if you are certain that the email >>> is genuine and the content is safe. >>> >>> Dear MPI-Forum members, >>> >>> MPI-3.1 and MPI-4.0 includes the following rule on top of the 2nd page >>> of the Language Binding chapter: >>> >>> | MPI implementations providing a Fortran interface must provide >>> | one or both of the following: >>> | - The USE mpi_f08 Fortran support method. >>> | - The USE mpi and INCLUDE 'mpif.h' Fortran support methods. >>> >>> The embiggening was included only in the C and mpi_f08 bindings. >>> >>> Most implementors nowadays provide all three MPI Fortran support methods, >>> i.e., the mpi_f08 and the mpi module and mpif.h. >>> >>> For all MPI-3.1 libraries that provide all three MPI Fortran support methods, >>> the most easiest and fastest way to provide MPI-4.0 for C and Fortran is, >>> - to implement the embiggening for C, >>> - and to remove the mpi_f08 module, >>> - and maybe months or years later, to provide mpi_f08 again, now embiggened. >>> >>> This implementation path (by removing mpi_f08 from MPI-3.1 and to name >>> it MPI-4.0 without Fortran embiggening) was of course never intended, >>> when we decided to require the embiggened routines only for mpi_f08. >>> The goal was >>> - to not require additional work from the implementors for the old mpi module >>> - convince the users that it is a good idea to make a transition >>> to mpi_f08. >>> >>> The most simplest way to resolve this problem would be to require >>> mpi_f08 for MPI-4.0, i.e., to change the text to >>> >>> | MPI implementations providing a Fortran interface >>> | - must provide the USE mpi_f08 Fortran support method, >>> | - and additionally may provide both, >>> | the USE mpi and INCLUDE 'mpif.h' Fortran support methods. >>> >>> What is your opinion? >>> >>> I expect that we should discuss this next Wednesday at our MPI Forum telcon. >>> >>> Best regards >>> Rolf >>> >>> -- >>> Dr. Rolf Rabenseifner . . . . . . . . . .. [ mailto:[email protected] | 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 . . . [ >>> https://urldefense.com/v3/__http://www.hlrs.de/people/rabenseifner__;!!DZ3f… >>> | www.hlrs.de/people/rabenseifner ] . >>> Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room 1.307) . >>> _______________________________________________ >>> mpi-forum mailing list >>> [ mailto:[email protected] | mpi-forum(a)lists.mpi-forum.org ] >>> [ >>> https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp… >>> | https://lists.mpi-forum.org/mailman/listinfo/mpi-forum ] >>> >>> _______________________________________________ >>> mpi-forum mailing list >>> [ mailto:[email protected] | mpi-forum(a)lists.mpi-forum.org ] >>> [ >>> https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp… >>> | >>> https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp… >>> ] >>> >>> _______________________________________________ >>> mpi-forum mailing list >>> [ mailto:[email protected] | mpi-forum(a)lists.mpi-forum.org ] >>> https://lists.mpi-forum.org/mailman/listinfo/mpi-forum >>> >>> >>> _______________________________________________ >>> mpi-forum mailing list >>> mpi-forum(a)lists.mpi-forum.org >>> https://lists.mpi-forum.org/mailman/listinfo/mpi-forum >> >> -- >> 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) . >> _______________________________________________ >> mpi-forum mailing list >> mpi-forum(a)lists.mpi-forum.org >> https://lists.mpi-forum.org/mailman/listinfo/mpi-forum > > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org > https://lists.mpi-forum.org/mailman/listinfo/mpi-forum

1 0

Re: [Mpi-forum] Big Fortran hole in MPI-4.0 Embiggening
by Wesley Bland 11 Jan '21

11 Jan '21

I agree with Jeff that this is a non-issue. Bill’s assertion that adding Fortran support for large count once you have C support has been true in MPICH development at least and no vender would drop anything just to put MPI 4.0 compatibility on a marketing slide. They’d just put it on the slide anyway. Every implementation out there has some sort of exception, intentional or otherwise, that makes them technically non-compliant. In the end, their users determine 1) whether they care enough to request the feature and 2) whether they care enough to switch to a competing library. I think the best way forward is to do something like what had been proposed when the embiggening/pythonization started and separate bindings from specification and just say that an MPI implementation has MPI 4.0 C bindings and MPI 3.1 F08 bindings and MPI 5.0 Python bindings. That means each binding can be updated in whatever way makes sense for them. For 4.0, I don’t think this is an issue that needs to be resolved. Us telling the implementations what they have to do will have essentially no impact anyway. Thanks, Wes > On Jan 11, 2021, at 6:26 AM, Rolf Rabenseifner via mpi-forum <mpi-forum(a)lists.mpi-forum.org> wrote: > > Dear Jeff and all, > >> I think their users would be very unhappy. > > Yes, users can be unhappy and compute centers may definitely recommend > an other MPI library - at least for any software development. > > I want to remember that MPI-3.1 requires for the mpi module: > > "Provide explicit interfaces according to the Fortran routine interface specications. > This module therefore guarantees compile-time argument checking and allows positional > and keyword-based argument lists. If an implementation is paired with a > compiler that either does not support TYPE(*), DIMENSION(..) from TS 29113, or > is otherwise unable to ignore the types of choice buffers, then the implementation must > provide explicit interfaces only for MPI routines with no choice buer arguments. See > Section 17.1.6 for more details." > > Although there are no more such compilers that do not provide one of the two methods, > there are more than 5 years after MPI-3.1 still MPI libraries that do not > provide keyword-based argument lists with the mpi module. > And those libraries provide such support for the mpi_f08 module. > This means, they prove that they are not MPI-3.1 compliant :-) > > > Sometimes implementors ignore the goals and the wording of the MPI standard. > > > And there is also no request that users want to have a less-quality > mpi module definition. > > Best regards > Rolf > > > ----- Original Message ----- >> From: "Main MPI Forum mailing list" <mpi-forum(a)lists.mpi-forum.org> >> To: "Main MPI Forum mailing list" <mpi-forum(a)lists.mpi-forum.org> >> Cc: "Jeff Squyres" <jsquyres(a)cisco.com> >> Sent: Sunday, January 10, 2021 5:09:55 PM >> Subject: Re: [Mpi-forum] Big Fortran hole in MPI-4.0 Embiggening > >> I can't imagine a vendor would: >> >> - support mpi_f08 >> - then stop supporting mpi_f08 just so that they can get an "MPI-4.0" release >> out >> - then support mpi_f08 again >> >> I think their users would be very unhappy. >> >> >> >> >> >> On Jan 10, 2021, at 10:55 AM, William Gropp via mpi-forum < [ >> mailto:[email protected] <mailto:[email protected]> | mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]> ] > wrote: >> >> I agree with Dan that this is a big change from the RCM. Further, the approach >> in MPI has always been to encourage the users to make it clear to the vendors >> what is acceptable in implementations, especially implementation schedules. >> Nothing in the standard prohibits implementors from continuing to provide MPI >> 3.x implementations while they work to provide a full MPI 4.0 implementation. >> The MPI forum has no enforcement power on the implementors, and I believe this >> text is unnecessary and will not provide the guarantee that Rolf wants. >> Further, frankly once the C embiggened interface is implemented, creating the >> mpi_f08 version is relatively straightforward. >> >> Bill >> >> William Gropp >> Director, NCSA >> Thomas M. Siebel Chair in Computer Science >> University of Illinois Urbana-Champaign >> IEEE-CS President-Elect >> >> >> >> >> >> >> On Jan 10, 2021, at 7:44 AM, HOLMES Daniel via mpi-forum < [ >> mailto:[email protected] <mailto:[email protected]> | mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]> ] > wrote: >> >> Hi Rolf, >> >> This is a (somewhat contrived, arguably) reason for taking another tiny step >> towards removing the “mpif.h” method of Fortran support and pushing users and >> implementations towards preferring the Fortran 2008 interface, which is a >> direction of travel that I fully support. >> >> I think this might be seen as quite a big change for implementers, especially if >> it were to occur between the RCM and FRM. >> >> Cheers, >> Dan. >> — >> Dr Daniel Holmes PhD >> Architect (HPC Research) >> [ mailto:[email protected] <mailto:[email protected]> | d.holmes(a)epcc.ed.ac.uk <mailto:[email protected]> ] >> Phone: +44 (0) 131 651 3465 >> Mobile: +44 (0) 7940 524 088 >> Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT >> — >> The University of Edinburgh is a charitable body, registered in Scotland, with >> registration number SC005336. >> — >> >> >> >> >> On 10 Jan 2021, at 12:22, Rolf Rabenseifner via mpi-forum < [ >> mailto:[email protected] <mailto:[email protected]> | mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]> ] > wrote: >> >> This email was sent to you by someone outside the University. >> You should only click on links or attachments if you are certain that the email >> is genuine and the content is safe. >> >> Dear MPI-Forum members, >> >> MPI-3.1 and MPI-4.0 includes the following rule on top of the 2nd page >> of the Language Binding chapter: >> >> | MPI implementations providing a Fortran interface must provide >> | one or both of the following: >> | - The USE mpi_f08 Fortran support method. >> | - The USE mpi and INCLUDE 'mpif.h' Fortran support methods. >> >> The embiggening was included only in the C and mpi_f08 bindings. >> >> Most implementors nowadays provide all three MPI Fortran support methods, >> i.e., the mpi_f08 and the mpi module and mpif.h. >> >> For all MPI-3.1 libraries that provide all three MPI Fortran support methods, >> the most easiest and fastest way to provide MPI-4.0 for C and Fortran is, >> - to implement the embiggening for C, >> - and to remove the mpi_f08 module, >> - and maybe months or years later, to provide mpi_f08 again, now embiggened. >> >> This implementation path (by removing mpi_f08 from MPI-3.1 and to name >> it MPI-4.0 without Fortran embiggening) was of course never intended, >> when we decided to require the embiggened routines only for mpi_f08. >> The goal was >> - to not require additional work from the implementors for the old mpi module >> - convince the users that it is a good idea to make a transition >> to mpi_f08. >> >> The most simplest way to resolve this problem would be to require >> mpi_f08 for MPI-4.0, i.e., to change the text to >> >> | MPI implementations providing a Fortran interface >> | - must provide the USE mpi_f08 Fortran support method, >> | - and additionally may provide both, >> | the USE mpi and INCLUDE 'mpif.h' Fortran support methods. >> >> What is your opinion? >> >> I expect that we should discuss this next Wednesday at our MPI Forum telcon. >> >> Best regards >> Rolf >> >> -- >> Dr. Rolf Rabenseifner . . . . . . . . . .. [ mailto:[email protected] <mailto:[email protected]> | email >> rabenseifner(a)hlrs.de <mailto:[email protected]> ] . >> 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 . . . [ >> https://urldefense.com/v3/__http://www.hlrs.de/people/rabenseifner__;!!DZ3f… <https://urldefense.com/v3/__http://www.hlrs.de/people/rabenseifner__;!!DZ3f…> >> | www.hlrs.de/people/rabenseifner <http://www.hlrs.de/people/rabenseifner> ] . >> Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room 1.307) . >> _______________________________________________ >> mpi-forum mailing list >> [ mailto:[email protected] <mailto:[email protected]> | mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]> ] >> [ >> https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp… <https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp…> >> | https://lists.mpi-forum.org/mailman/listinfo/mpi-forum <https://lists.mpi-forum.org/mailman/listinfo/mpi-forum> ] >> >> _______________________________________________ >> mpi-forum mailing list >> [ mailto:[email protected] <mailto:[email protected]> | mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]> ] >> [ >> https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp… <https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp…> >> | >> https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp… <https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp…> >> ] >> >> _______________________________________________ >> mpi-forum mailing list >> [ mailto:[email protected] <mailto:[email protected]> | mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]> ] >> https://lists.mpi-forum.org/mailman/listinfo/mpi-forum <https://lists.mpi-forum.org/mailman/listinfo/mpi-forum> >> >> >> _______________________________________________ >> mpi-forum mailing list >> mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]> >> https://lists.mpi-forum.org/mailman/listinfo/mpi-forum <https://lists.mpi-forum.org/mailman/listinfo/mpi-forum> > > -- > Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner(a)hlrs.de <mailto:[email protected]> . > 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 <http://www.hlrs.de/people/rabenseifner> . > Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room 1.307) . > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]> > https://lists.mpi-forum.org/mailman/listinfo/mpi-forum <https://lists.mpi-forum.org/mailman/listinfo/mpi-forum>

1 0

Re: [Mpi-forum] Big Fortran hole in MPI-4.0 Embiggening
by Rolf Rabenseifner 11 Jan '21

11 Jan '21

Dear Jeff and all, > I think their users would be very unhappy. Yes, users can be unhappy and compute centers may definitely recommend an other MPI library - at least for any software development. I want to remember that MPI-3.1 requires for the mpi module: "Provide explicit interfaces according to the Fortran routine interface specications. This module therefore guarantees compile-time argument checking and allows positional and keyword-based argument lists. If an implementation is paired with a compiler that either does not support TYPE(*), DIMENSION(..) from TS 29113, or is otherwise unable to ignore the types of choice buffers, then the implementation must provide explicit interfaces only for MPI routines with no choice buer arguments. See Section 17.1.6 for more details." Although there are no more such compilers that do not provide one of the two methods, there are more than 5 years after MPI-3.1 still MPI libraries that do not provide keyword-based argument lists with the mpi module. And those libraries provide such support for the mpi_f08 module. This means, they prove that they are not MPI-3.1 compliant :-) Sometimes implementors ignore the goals and the wording of the MPI standard. And there is also no request that users want to have a less-quality mpi module definition. Best regards Rolf ----- Original Message ----- > From: "Main MPI Forum mailing list" <mpi-forum(a)lists.mpi-forum.org> > To: "Main MPI Forum mailing list" <mpi-forum(a)lists.mpi-forum.org> > Cc: "Jeff Squyres" <jsquyres(a)cisco.com> > Sent: Sunday, January 10, 2021 5:09:55 PM > Subject: Re: [Mpi-forum] Big Fortran hole in MPI-4.0 Embiggening > I can't imagine a vendor would: > > - support mpi_f08 > - then stop supporting mpi_f08 just so that they can get an "MPI-4.0" release > out > - then support mpi_f08 again > > I think their users would be very unhappy. > > > > > > On Jan 10, 2021, at 10:55 AM, William Gropp via mpi-forum < [ > mailto:[email protected] | mpi-forum(a)lists.mpi-forum.org ] > wrote: > > I agree with Dan that this is a big change from the RCM. Further, the approach > in MPI has always been to encourage the users to make it clear to the vendors > what is acceptable in implementations, especially implementation schedules. > Nothing in the standard prohibits implementors from continuing to provide MPI > 3.x implementations while they work to provide a full MPI 4.0 implementation. > The MPI forum has no enforcement power on the implementors, and I believe this > text is unnecessary and will not provide the guarantee that Rolf wants. > Further, frankly once the C embiggened interface is implemented, creating the > mpi_f08 version is relatively straightforward. > > Bill > > William Gropp > Director, NCSA > Thomas M. Siebel Chair in Computer Science > University of Illinois Urbana-Champaign > IEEE-CS President-Elect > > > > > > > On Jan 10, 2021, at 7:44 AM, HOLMES Daniel via mpi-forum < [ > mailto:[email protected] | mpi-forum(a)lists.mpi-forum.org ] > wrote: > > Hi Rolf, > > This is a (somewhat contrived, arguably) reason for taking another tiny step > towards removing the “mpif.h” method of Fortran support and pushing users and > implementations towards preferring the Fortran 2008 interface, which is a > direction of travel that I fully support. > > I think this might be seen as quite a big change for implementers, especially if > it were to occur between the RCM and FRM. > > Cheers, > Dan. > — > Dr Daniel Holmes PhD > Architect (HPC Research) > [ mailto:[email protected] | d.holmes(a)epcc.ed.ac.uk ] > Phone: +44 (0) 131 651 3465 > Mobile: +44 (0) 7940 524 088 > Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT > — > The University of Edinburgh is a charitable body, registered in Scotland, with > registration number SC005336. > — > > > > > On 10 Jan 2021, at 12:22, Rolf Rabenseifner via mpi-forum < [ > mailto:[email protected] | mpi-forum(a)lists.mpi-forum.org ] > wrote: > > This email was sent to you by someone outside the University. > You should only click on links or attachments if you are certain that the email > is genuine and the content is safe. > > Dear MPI-Forum members, > > MPI-3.1 and MPI-4.0 includes the following rule on top of the 2nd page > of the Language Binding chapter: > >| MPI implementations providing a Fortran interface must provide >| one or both of the following: >| - The USE mpi_f08 Fortran support method. >| - The USE mpi and INCLUDE 'mpif.h' Fortran support methods. > > The embiggening was included only in the C and mpi_f08 bindings. > > Most implementors nowadays provide all three MPI Fortran support methods, > i.e., the mpi_f08 and the mpi module and mpif.h. > > For all MPI-3.1 libraries that provide all three MPI Fortran support methods, > the most easiest and fastest way to provide MPI-4.0 for C and Fortran is, > - to implement the embiggening for C, > - and to remove the mpi_f08 module, > - and maybe months or years later, to provide mpi_f08 again, now embiggened. > > This implementation path (by removing mpi_f08 from MPI-3.1 and to name > it MPI-4.0 without Fortran embiggening) was of course never intended, > when we decided to require the embiggened routines only for mpi_f08. > The goal was > - to not require additional work from the implementors for the old mpi module > - convince the users that it is a good idea to make a transition > to mpi_f08. > > The most simplest way to resolve this problem would be to require > mpi_f08 for MPI-4.0, i.e., to change the text to > >| MPI implementations providing a Fortran interface >| - must provide the USE mpi_f08 Fortran support method, >| - and additionally may provide both, >| the USE mpi and INCLUDE 'mpif.h' Fortran support methods. > > What is your opinion? > > I expect that we should discuss this next Wednesday at our MPI Forum telcon. > > Best regards > Rolf > > -- > Dr. Rolf Rabenseifner . . . . . . . . . .. [ mailto:[email protected] | 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 . . . [ > https://urldefense.com/v3/__http://www.hlrs.de/people/rabenseifner__;!!DZ3f… > | www.hlrs.de/people/rabenseifner ] . > Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room 1.307) . > _______________________________________________ > mpi-forum mailing list > [ mailto:[email protected] | mpi-forum(a)lists.mpi-forum.org ] > [ > https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp… > | https://lists.mpi-forum.org/mailman/listinfo/mpi-forum ] > > _______________________________________________ > mpi-forum mailing list > [ mailto:[email protected] | mpi-forum(a)lists.mpi-forum.org ] > [ > https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp… > | > https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp… > ] > > _______________________________________________ > mpi-forum mailing list > [ mailto:[email protected] | mpi-forum(a)lists.mpi-forum.org ] > https://lists.mpi-forum.org/mailman/listinfo/mpi-forum > > > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org > https://lists.mpi-forum.org/mailman/listinfo/mpi-forum -- 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: [Mpi-forum] Big Fortran hole in MPI-4.0 Embiggening
by Jeff Squyres (jsquyres) 10 Jan '21

10 Jan '21

I can't imagine a vendor would: - support mpi_f08 - then stop supporting mpi_f08 just so that they can get an "MPI-4.0" release out - then support mpi_f08 again I think their users would be very unhappy. On Jan 10, 2021, at 10:55 AM, William Gropp via mpi-forum <mpi-forum(a)lists.mpi-forum.org<mailto:[email protected]>> wrote: I agree with Dan that this is a big change from the RCM. Further, the approach in MPI has always been to encourage the users to make it clear to the vendors what is acceptable in implementations, especially implementation schedules. Nothing in the standard prohibits implementors from continuing to provide MPI 3.x implementations while they work to provide a full MPI 4.0 implementation. The MPI forum has no enforcement power on the implementors, and I believe this text is unnecessary and will not provide the guarantee that Rolf wants. Further, frankly once the C embiggened interface is implemented, creating the mpi_f08 version is relatively straightforward. Bill William Gropp Director, NCSA Thomas M. Siebel Chair in Computer Science University of Illinois Urbana-Champaign IEEE-CS President-Elect On Jan 10, 2021, at 7:44 AM, HOLMES Daniel via mpi-forum <mpi-forum(a)lists.mpi-forum.org<mailto:[email protected]>> wrote: Hi Rolf, This is a (somewhat contrived, arguably) reason for taking another tiny step towards removing the “mpif.h” method of Fortran support and pushing users and implementations towards preferring the Fortran 2008 interface, which is a direction of travel that I fully support. I think this might be seen as quite a big change for implementers, especially if it were to occur between the RCM and FRM. Cheers, Dan. — Dr Daniel Holmes PhD Architect (HPC Research) d.holmes(a)epcc.ed.ac.uk<mailto:[email protected]> Phone: +44 (0) 131 651 3465 Mobile: +44 (0) 7940 524 088 Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT — The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. — On 10 Jan 2021, at 12:22, Rolf Rabenseifner via mpi-forum <mpi-forum(a)lists.mpi-forum.org<mailto:[email protected]>> wrote: This email was sent to you by someone outside the University. You should only click on links or attachments if you are certain that the email is genuine and the content is safe. Dear MPI-Forum members, MPI-3.1 and MPI-4.0 includes the following rule on top of the 2nd page of the Language Binding chapter: | MPI implementations providing a Fortran interface must provide | one or both of the following: | - The USE mpi_f08 Fortran support method. | - The USE mpi and INCLUDE 'mpif.h' Fortran support methods. The embiggening was included only in the C and mpi_f08 bindings. Most implementors nowadays provide all three MPI Fortran support methods, i.e., the mpi_f08 and the mpi module and mpif.h. For all MPI-3.1 libraries that provide all three MPI Fortran support methods, the most easiest and fastest way to provide MPI-4.0 for C and Fortran is, - to implement the embiggening for C, - and to remove the mpi_f08 module, - and maybe months or years later, to provide mpi_f08 again, now embiggened. This implementation path (by removing mpi_f08 from MPI-3.1 and to name it MPI-4.0 without Fortran embiggening) was of course never intended, when we decided to require the embiggened routines only for mpi_f08. The goal was - to not require additional work from the implementors for the old mpi module - convince the users that it is a good idea to make a transition to mpi_f08. The most simplest way to resolve this problem would be to require mpi_f08 for MPI-4.0, i.e., to change the text to | MPI implementations providing a Fortran interface | - must provide the USE mpi_f08 Fortran support method, | - and additionally may provide both, | the USE mpi and INCLUDE 'mpif.h' Fortran support methods. What is your opinion? I expect that we should discuss this next Wednesday at our MPI Forum telcon. Best regards Rolf -- Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner(a)hlrs.de<mailto:[email protected]> . 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<https://urldefense.com/v3/__http://www.hlrs.de/people/rabenseifner__;!!DZ3f…> . Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room 1.307) . _______________________________________________ mpi-forum mailing list mpi-forum(a)lists.mpi-forum.org<mailto:[email protected]> https://lists.mpi-forum.org/mailman/listinfo/mpi-forum<https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp…> _______________________________________________ mpi-forum mailing list mpi-forum(a)lists.mpi-forum.org<mailto:[email protected]> https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp… _______________________________________________ mpi-forum mailing list mpi-forum(a)lists.mpi-forum.org<mailto:[email protected]> https://lists.mpi-forum.org/mailman/listinfo/mpi-forum

1 0

Re: [Mpi-forum] Big Fortran hole in MPI-4.0 Embiggening
by William Gropp 10 Jan '21

10 Jan '21

I agree with Dan that this is a big change from the RCM. Further, the approach in MPI has always been to encourage the users to make it clear to the vendors what is acceptable in implementations, especially implementation schedules. Nothing in the standard prohibits implementors from continuing to provide MPI 3.x implementations while they work to provide a full MPI 4.0 implementation. The MPI forum has no enforcement power on the implementors, and I believe this text is unnecessary and will not provide the guarantee that Rolf wants. Further, frankly once the C embiggened interface is implemented, creating the mpi_f08 version is relatively straightforward. Bill William Gropp Director, NCSA Thomas M. Siebel Chair in Computer Science University of Illinois Urbana-Champaign IEEE-CS President-Elect > On Jan 10, 2021, at 7:44 AM, HOLMES Daniel via mpi-forum <mpi-forum(a)lists.mpi-forum.org> wrote: > > Hi Rolf, > > This is a (somewhat contrived, arguably) reason for taking another tiny step towards removing the “mpif.h” method of Fortran support and pushing users and implementations towards preferring the Fortran 2008 interface, which is a direction of travel that I fully support. > > I think this might be seen as quite a big change for implementers, especially if it were to occur between the RCM and FRM. > > Cheers, > Dan. > — > Dr Daniel Holmes PhD > Architect (HPC Research) > d.holmes(a)epcc.ed.ac.uk <mailto:[email protected]> > Phone: +44 (0) 131 651 3465 > Mobile: +44 (0) 7940 524 088 > Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT > — > The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. > — > >> On 10 Jan 2021, at 12:22, Rolf Rabenseifner via mpi-forum <mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]>> wrote: >> >> This email was sent to you by someone outside the University. >> You should only click on links or attachments if you are certain that the email is genuine and the content is safe. >> >> Dear MPI-Forum members, >> >> MPI-3.1 and MPI-4.0 includes the following rule on top of the 2nd page >> of the Language Binding chapter: >> >> | MPI implementations providing a Fortran interface must provide >> | one or both of the following: >> | - The USE mpi_f08 Fortran support method. >> | - The USE mpi and INCLUDE 'mpif.h' Fortran support methods. >> >> The embiggening was included only in the C and mpi_f08 bindings. >> >> Most implementors nowadays provide all three MPI Fortran support methods, >> i.e., the mpi_f08 and the mpi module and mpif.h. >> >> For all MPI-3.1 libraries that provide all three MPI Fortran support methods, >> the most easiest and fastest way to provide MPI-4.0 for C and Fortran is, >> - to implement the embiggening for C, >> - and to remove the mpi_f08 module, >> - and maybe months or years later, to provide mpi_f08 again, now embiggened. >> >> This implementation path (by removing mpi_f08 from MPI-3.1 and to name >> it MPI-4.0 without Fortran embiggening) was of course never intended, >> when we decided to require the embiggened routines only for mpi_f08. >> The goal was >> - to not require additional work from the implementors for the old mpi module >> - convince the users that it is a good idea to make a transition >> to mpi_f08. >> >> The most simplest way to resolve this problem would be to require >> mpi_f08 for MPI-4.0, i.e., to change the text to >> >> | MPI implementations providing a Fortran interface >> | - must provide the USE mpi_f08 Fortran support method, >> | - and additionally may provide both, >> | the USE mpi and INCLUDE 'mpif.h' Fortran support methods. >> >> What is your opinion? >> >> I expect that we should discuss this next Wednesday at our MPI Forum telcon. >> >> Best regards >> Rolf >> >> -- >> Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner(a)hlrs.de <mailto:[email protected]> . >> 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 <https://urldefense.com/v3/__http://www.hlrs.de/people/rabenseifner__;!!DZ3f…> . >> Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room 1.307) . >> _______________________________________________ >> mpi-forum mailing list >> mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]> >> https://lists.mpi-forum.org/mailman/listinfo/mpi-forum <https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp…> > > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]> > https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp… <https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp…>

1 0

Re: [Mpi-forum] Big Fortran hole in MPI-4.0 Embiggening
by HOLMES Daniel 10 Jan '21

10 Jan '21

Hi Rolf, This is a (somewhat contrived, arguably) reason for taking another tiny step towards removing the “mpif.h” method of Fortran support and pushing users and implementations towards preferring the Fortran 2008 interface, which is a direction of travel that I fully support. I think this might be seen as quite a big change for implementers, especially if it were to occur between the RCM and FRM. Cheers, Dan. — Dr Daniel Holmes PhD Architect (HPC Research) d.holmes(a)epcc.ed.ac.uk<mailto:[email protected]> Phone: +44 (0) 131 651 3465 Mobile: +44 (0) 7940 524 088 Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT — The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. — On 10 Jan 2021, at 12:22, Rolf Rabenseifner via mpi-forum <mpi-forum(a)lists.mpi-forum.org<mailto:[email protected]>> wrote: This email was sent to you by someone outside the University. You should only click on links or attachments if you are certain that the email is genuine and the content is safe. Dear MPI-Forum members, MPI-3.1 and MPI-4.0 includes the following rule on top of the 2nd page of the Language Binding chapter: | MPI implementations providing a Fortran interface must provide | one or both of the following: | - The USE mpi_f08 Fortran support method. | - The USE mpi and INCLUDE 'mpif.h' Fortran support methods. The embiggening was included only in the C and mpi_f08 bindings. Most implementors nowadays provide all three MPI Fortran support methods, i.e., the mpi_f08 and the mpi module and mpif.h. For all MPI-3.1 libraries that provide all three MPI Fortran support methods, the most easiest and fastest way to provide MPI-4.0 for C and Fortran is, - to implement the embiggening for C, - and to remove the mpi_f08 module, - and maybe months or years later, to provide mpi_f08 again, now embiggened. This implementation path (by removing mpi_f08 from MPI-3.1 and to name it MPI-4.0 without Fortran embiggening) was of course never intended, when we decided to require the embiggened routines only for mpi_f08. The goal was - to not require additional work from the implementors for the old mpi module - convince the users that it is a good idea to make a transition to mpi_f08. The most simplest way to resolve this problem would be to require mpi_f08 for MPI-4.0, i.e., to change the text to | MPI implementations providing a Fortran interface | - must provide the USE mpi_f08 Fortran support method, | - and additionally may provide both, | the USE mpi and INCLUDE 'mpif.h' Fortran support methods. What is your opinion? I expect that we should discuss this next Wednesday at our MPI Forum telcon. Best regards Rolf -- Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner(a)hlrs.de<mailto:[email protected]> . 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<http://www.hlrs.de/people/rabenseifner> . Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room 1.307) . _______________________________________________ mpi-forum mailing list mpi-forum(a)lists.mpi-forum.org<mailto:[email protected]> https://lists.mpi-forum.org/mailman/listinfo/mpi-forum

1 0

Big Fortran hole in MPI-4.0 Embiggening
by Rolf Rabenseifner 10 Jan '21

10 Jan '21

Dear MPI-Forum members, MPI-3.1 and MPI-4.0 includes the following rule on top of the 2nd page of the Language Binding chapter: | MPI implementations providing a Fortran interface must provide | one or both of the following: | - The USE mpi_f08 Fortran support method. | - The USE mpi and INCLUDE 'mpif.h' Fortran support methods. The embiggening was included only in the C and mpi_f08 bindings. Most implementors nowadays provide all three MPI Fortran support methods, i.e., the mpi_f08 and the mpi module and mpif.h. For all MPI-3.1 libraries that provide all three MPI Fortran support methods, the most easiest and fastest way to provide MPI-4.0 for C and Fortran is, - to implement the embiggening for C, - and to remove the mpi_f08 module, - and maybe months or years later, to provide mpi_f08 again, now embiggened. This implementation path (by removing mpi_f08 from MPI-3.1 and to name it MPI-4.0 without Fortran embiggening) was of course never intended, when we decided to require the embiggened routines only for mpi_f08. The goal was - to not require additional work from the implementors for the old mpi module - convince the users that it is a good idea to make a transition to mpi_f08. The most simplest way to resolve this problem would be to require mpi_f08 for MPI-4.0, i.e., to change the text to | MPI implementations providing a Fortran interface | - must provide the USE mpi_f08 Fortran support method, | - and additionally may provide both, | the USE mpi and INCLUDE 'mpif.h' Fortran support methods. What is your opinion? I expect that we should discuss this next Wednesday at our MPI Forum telcon. Best regards Rolf -- 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

Virtual MPI Forum Meeting next Wednesday (Jan. 13th) 10am US Central Time
by Martin Schulz 08 Jan '21

08 Jan '21

Hi all, Thanks for a productive meeting today - as discussed at the end, we'll meet next Wednesday (1/13) at the usual time (10am Central Time) to go over open issues and questions. In the meantime, please continue to take a look at the project board at https://github.com/mpi-forum/mpi-issues/projects/2 and push issues assigned to you forward. As discussed, we need all PRs for MPI 4.0 by Monday. Thanks! Martin -- Prof. Dr. Martin Schulz, Chair of Computer Architecture and Parallel Systems Department of Informatics, TU-Munich, Boltzmannstraße 3, D-85748 Garching Member of the Board of Directors at the Leibniz Supercomputing Centre (LRZ) Email: schulzm(a)in.tum.de

1 0

Reminder: Extra Virtual MPI Forum Meeting tomorrow (Friday) 9am US Central Time
by Martin Schulz 07 Jan '21

07 Jan '21

Hi all, Just as a reminder - we will have an out-of-sequence, extra virtual MPI forum meeting tomorrow (Friday) at 9am (!) US Central time. The link is, as always, available from the MPI Forum website. The plan is to go through issues that are still open for the completion of the final MPI 4.0 release candidate. The issues are all on the project board at: https://github.com/mpi-forum/mpi-issues/projects/2 Please take a look at it and see if you have any open issues still assigned or are needed as a reviewer. The deadline for all PRs is the 11th. Talk to you tomorrow, Thanks! Martin -- Prof. Dr. Martin Schulz, Chair of Computer Architecture and Parallel Systems Department of Informatics, TU-Munich, Boltzmannstraße 3, D-85748 Garching Member of the Board of Directors at the Leibniz Supercomputing Centre (LRZ) Email: schulzm(a)in.tum.de

1 0

MPI 4.0 Changes Reminders
by work＠wesbland.com 04 Jan '21

04 Jan '21

1 0