November 2021 - mpiwg-rma - mailman3.cels.anl.gov

Re: [mpiwg-rma] [MPIWG Fortran] "small count" variants of RMA functions
by Jeff Hammond 29 Nov '21

29 Nov '21

I don't think it's worth a plenary. It really comes down to whether the Fortran Wizards (Rolf, Bill, Hubert) accept this. The implementation of this change is easy. Jeff On Mon, Nov 29, 2021 at 4:03 PM Wesley Bland <work(a)wesbland.com> wrote: > As long as we’re having remote meetings, we’re not setting up WG time > during the meeting itself. So each WG is just having its usual con calls on > their own schedule (or in the case of Fortran, not). If you folks want to > throw together a plenary, we can still put that on the agenda though. > > Thanks, > Wes > > On Nov 29, 2021, at 3:45 AM, Jeff Hammond via mpiwg-fortran < > mpiwg-fortran(a)lists.mpi-forum.org> wrote: > > I created https://github.com/mpi-forum/mpi-issues/issues/528. > > Normally, we would have a Fortran WG meeting during the quarterly > meeting. Is there going to be a Fortran WG session at the December meeting? > > Jeff > > On Sat, Nov 27, 2021 at 9:14 PM William Gropp <wgropp(a)illinois.edu> wrote: > >> I agree with Jeff here. This is a good enhancement; I believe the only >> reason that we hadn’t done this before is that we’ve stuck closely to the >> original Fortran look-and-feel - which was pre-Fortran 90. We could have >> done this earlier, and now that there are explicit interfaces that are only >> available in Fortran08, there is no reason not to do this. >> >> Bill >> >> William Gropp >> Director, NCSA >> Thomas M. Siebel Chair in Computer Science >> University of Illinois Urbana-Champaign >> IEEE-CS President-Elect >> >> >> >> >> On Nov 25, 2021, at 4:42 AM, Jeff Hammond via mpiwg-fortran < >> mpiwg-fortran(a)lists.mpi-forum.org> wrote: >> >> RMA absolutely did the right thing with MPI_Aint size and displacement >> arguments. >> >> In C, it is no problem to pass 32b integers to these arguments when AINT >> is 64b, because C type promotion rules just work. >> >> In Fortran, I cannot do this, because an integer literal is not >> compatible with an AINT argument. I find it tedious to have to declare a >> variable just for this, or to explicitly cast with integer(100, >> MPI_ADDRESS_KIND). >> >> Since Fortran does polymorphism right and can have subroutine >> declarations for both default integer and AINT integer scalar arguments, is >> there a good reason not to add the former? >> >> This proposal applies not to just MPI_Win_allocate but any RMA function >> that uses AINT scalar arguments. >> >> Thanks, >> >> Jeff >> >> % mpifort y.F90 >> y.F90:20:60: >> >> 20 | call MPI_Win_allocate(100, 1, MPI_INFO_NULL, comm, XA, WA) >> | 1 >> Error: There is no specific subroutine for the generic 'mpi_win_allocate' >> at (1) >> >> % cat y.F90 >> ! USE mpi_f08 >> ! MPI_Win_allocate(size, disp_unit, info, comm, baseptr, win, ierror) >> ! USE, INTRINSIC :: ISO_C_BINDING, ONLY : C_PTR >> ! INTEGER(KIND=MPI_ADDRESS_KIND), INTENT(IN) :: size >> ! INTEGER, INTENT(IN) :: disp_unit >> ! TYPE(MPI_Info), INTENT(IN) :: info >> ! TYPE(MPI_Comm), INTENT(IN) :: comm >> ! TYPE(C_PTR), INTENT(OUT) :: baseptr >> ! TYPE(MPI_Win), INTENT(OUT) :: win >> ! INTEGER, OPTIONAL, INTENT(OUT) :: ierror >> program main >> use iso_fortran_env >> use mpi_f08 >> implicit none >> integer(kind=MPI_ADDRESS_KIND) :: as = 100 >> type(c_ptr) :: XA >> type(MPI_Win) :: WA >> TYPE(MPI_Comm) :: comm = MPI_COMM_WORLD >> call MPI_Win_allocate(as, 1, MPI_INFO_NULL, comm, XA, WA) >> call MPI_Win_allocate(100, 1, MPI_INFO_NULL, comm, XA, WA) >> end program main >> >> % mpicc y.c && echo SUCCESS >> SUCCESS >> >> % cat y.c >> #include <mpi.h> >> int main(void) >> { >> MPI_Aint as = 100; >> int * XA; >> MPI_Win WA; >> MPI_Win_allocate(as, 1, MPI_INFO_NULL, MPI_COMM_WORLD, XA, &WA); >> MPI_Win_allocate(100, 1, MPI_INFO_NULL, MPI_COMM_WORLD, XA, &WA); >> } >> >> >> -- >> Jeff Hammond >> jeff.science(a)gmail.com >> http://jeffhammond.github.io/ >> <https://urldefense.com/v3/__http://jeffhammond.github.io/__;!!DZ3fjg!rCl1MJ…> >> _______________________________________________ >> mpiwg-fortran mailing list >> mpiwg-fortran(a)lists.mpi-forum.org >> >> https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp… >> >> >> >> > > -- > Jeff Hammond > jeff.science(a)gmail.com > http://jeffhammond.github.io/ > _______________________________________________ > mpiwg-fortran mailing list > mpiwg-fortran(a)lists.mpi-forum.org > https://lists.mpi-forum.org/mailman/listinfo/mpiwg-fortran > > > -- Jeff Hammond jeff.science(a)gmail.com http://jeffhammond.github.io/

1 0

Re: [mpiwg-rma] [MPIWG Fortran] "small count" variants of RMA functions
by Wesley Bland 29 Nov '21

29 Nov '21

As long as we’re having remote meetings, we’re not setting up WG time during the meeting itself. So each WG is just having its usual con calls on their own schedule (or in the case of Fortran, not). If you folks want to throw together a plenary, we can still put that on the agenda though. Thanks, Wes > On Nov 29, 2021, at 3:45 AM, Jeff Hammond via mpiwg-fortran <mpiwg-fortran(a)lists.mpi-forum.org> wrote: > > I created https://github.com/mpi-forum/mpi-issues/issues/528 <https://github.com/mpi-forum/mpi-issues/issues/528>. > > Normally, we would have a Fortran WG meeting during the quarterly meeting. Is there going to be a Fortran WG session at the December meeting? > > Jeff > > On Sat, Nov 27, 2021 at 9:14 PM William Gropp <wgropp(a)illinois.edu <mailto:[email protected]>> wrote: > I agree with Jeff here. This is a good enhancement; I believe the only reason that we hadn’t done this before is that we’ve stuck closely to the original Fortran look-and-feel - which was pre-Fortran 90. We could have done this earlier, and now that there are explicit interfaces that are only available in Fortran08, there is no reason not to do this. > > Bill > > William Gropp > Director, NCSA > Thomas M. Siebel Chair in Computer Science > University of Illinois Urbana-Champaign > IEEE-CS President-Elect > > > > >> On Nov 25, 2021, at 4:42 AM, Jeff Hammond via mpiwg-fortran <mpiwg-fortran(a)lists.mpi-forum.org <mailto:[email protected]>> wrote: >> >> RMA absolutely did the right thing with MPI_Aint size and displacement arguments. >> >> In C, it is no problem to pass 32b integers to these arguments when AINT is 64b, because C type promotion rules just work. >> >> In Fortran, I cannot do this, because an integer literal is not compatible with an AINT argument. I find it tedious to have to declare a variable just for this, or to explicitly cast with integer(100,MPI_ADDRESS_KIND). >> >> Since Fortran does polymorphism right and can have subroutine declarations for both default integer and AINT integer scalar arguments, is there a good reason not to add the former? >> >> This proposal applies not to just MPI_Win_allocate but any RMA function that uses AINT scalar arguments. >> >> Thanks, >> >> Jeff >> >> % mpifort y.F90 >> y.F90:20:60: >> >> 20 | call MPI_Win_allocate(100, 1, MPI_INFO_NULL, comm, XA, WA) >> | 1 >> Error: There is no specific subroutine for the generic 'mpi_win_allocate' at (1) >> >> % cat y.F90 >> ! USE mpi_f08 >> ! MPI_Win_allocate(size, disp_unit, info, comm, baseptr, win, ierror) >> ! USE, INTRINSIC :: ISO_C_BINDING, ONLY : C_PTR >> ! INTEGER(KIND=MPI_ADDRESS_KIND), INTENT(IN) :: size >> ! INTEGER, INTENT(IN) :: disp_unit >> ! TYPE(MPI_Info), INTENT(IN) :: info >> ! TYPE(MPI_Comm), INTENT(IN) :: comm >> ! TYPE(C_PTR), INTENT(OUT) :: baseptr >> ! TYPE(MPI_Win), INTENT(OUT) :: win >> ! INTEGER, OPTIONAL, INTENT(OUT) :: ierror >> program main >> use iso_fortran_env >> use mpi_f08 >> implicit none >> integer(kind=MPI_ADDRESS_KIND) :: as = 100 >> type(c_ptr) :: XA >> type(MPI_Win) :: WA >> TYPE(MPI_Comm) :: comm = MPI_COMM_WORLD >> call MPI_Win_allocate(as, 1, MPI_INFO_NULL, comm, XA, WA) >> call MPI_Win_allocate(100, 1, MPI_INFO_NULL, comm, XA, WA) >> end program main >> >> % mpicc y.c && echo SUCCESS >> SUCCESS >> >> % cat y.c >> #include <mpi.h> >> int main(void) >> { >> MPI_Aint as = 100; >> int * XA; >> MPI_Win WA; >> MPI_Win_allocate(as, 1, MPI_INFO_NULL, MPI_COMM_WORLD, XA, &WA); >> MPI_Win_allocate(100, 1, MPI_INFO_NULL, MPI_COMM_WORLD, XA, &WA); >> } >> >> >> -- >> Jeff Hammond >> jeff.science(a)gmail.com <mailto:[email protected]> >> http://jeffhammond.github.io/ <https://urldefense.com/v3/__http://jeffhammond.github.io/__;!!DZ3fjg!rCl1MJ…>_______________________________________________ >> mpiwg-fortran mailing list >> mpiwg-fortran(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…> > > > > -- > Jeff Hammond > jeff.science(a)gmail.com <mailto:[email protected]> > http://jeffhammond.github.io/ <http://jeffhammond.github.io/>_______________________________________________ > mpiwg-fortran mailing list > mpiwg-fortran(a)lists.mpi-forum.org <mailto:[email protected]> > https://lists.mpi-forum.org/mailman/listinfo/mpiwg-fortran <https://lists.mpi-forum.org/mailman/listinfo/mpiwg-fortran>

1 0

Re: [mpiwg-rma] [MPIWG Fortran] "small count" variants of RMA functions
by Jeff Hammond 29 Nov '21

29 Nov '21

I created https://github.com/mpi-forum/mpi-issues/issues/528. Normally, we would have a Fortran WG meeting during the quarterly meeting. Is there going to be a Fortran WG session at the December meeting? Jeff On Sat, Nov 27, 2021 at 9:14 PM William Gropp <wgropp(a)illinois.edu> wrote: > I agree with Jeff here. This is a good enhancement; I believe the only > reason that we hadn’t done this before is that we’ve stuck closely to the > original Fortran look-and-feel - which was pre-Fortran 90. We could have > done this earlier, and now that there are explicit interfaces that are only > available in Fortran08, there is no reason not to do this. > > Bill > > William Gropp > Director, NCSA > Thomas M. Siebel Chair in Computer Science > University of Illinois Urbana-Champaign > IEEE-CS President-Elect > > > > > On Nov 25, 2021, at 4:42 AM, Jeff Hammond via mpiwg-fortran < > mpiwg-fortran(a)lists.mpi-forum.org> wrote: > > RMA absolutely did the right thing with MPI_Aint size and displacement > arguments. > > In C, it is no problem to pass 32b integers to these arguments when AINT > is 64b, because C type promotion rules just work. > > In Fortran, I cannot do this, because an integer literal is not compatible > with an AINT argument. I find it tedious to have to declare a variable > just for this, or to explicitly cast with integer(100,MPI_ADDRESS_KIND). > > Since Fortran does polymorphism right and can have subroutine declarations > for both default integer and AINT integer scalar arguments, is there a good > reason not to add the former? > > This proposal applies not to just MPI_Win_allocate but any RMA function > that uses AINT scalar arguments. > > Thanks, > > Jeff > > % mpifort y.F90 > y.F90:20:60: > > 20 | call MPI_Win_allocate(100, 1, MPI_INFO_NULL, comm, XA, WA) > | 1 > Error: There is no specific subroutine for the generic 'mpi_win_allocate' > at (1) > > % cat y.F90 > ! USE mpi_f08 > ! MPI_Win_allocate(size, disp_unit, info, comm, baseptr, win, ierror) > ! USE, INTRINSIC :: ISO_C_BINDING, ONLY : C_PTR > ! INTEGER(KIND=MPI_ADDRESS_KIND), INTENT(IN) :: size > ! INTEGER, INTENT(IN) :: disp_unit > ! TYPE(MPI_Info), INTENT(IN) :: info > ! TYPE(MPI_Comm), INTENT(IN) :: comm > ! TYPE(C_PTR), INTENT(OUT) :: baseptr > ! TYPE(MPI_Win), INTENT(OUT) :: win > ! INTEGER, OPTIONAL, INTENT(OUT) :: ierror > program main > use iso_fortran_env > use mpi_f08 > implicit none > integer(kind=MPI_ADDRESS_KIND) :: as = 100 > type(c_ptr) :: XA > type(MPI_Win) :: WA > TYPE(MPI_Comm) :: comm = MPI_COMM_WORLD > call MPI_Win_allocate(as, 1, MPI_INFO_NULL, comm, XA, WA) > call MPI_Win_allocate(100, 1, MPI_INFO_NULL, comm, XA, WA) > end program main > > % mpicc y.c && echo SUCCESS > SUCCESS > > % cat y.c > #include <mpi.h> > int main(void) > { > MPI_Aint as = 100; > int * XA; > MPI_Win WA; > MPI_Win_allocate(as, 1, MPI_INFO_NULL, MPI_COMM_WORLD, XA, &WA); > MPI_Win_allocate(100, 1, MPI_INFO_NULL, MPI_COMM_WORLD, XA, &WA); > } > > > -- > Jeff Hammond > jeff.science(a)gmail.com > http://jeffhammond.github.io/ > <https://urldefense.com/v3/__http://jeffhammond.github.io/__;!!DZ3fjg!rCl1MJ…> > _______________________________________________ > mpiwg-fortran mailing list > mpiwg-fortran(a)lists.mpi-forum.org > > https://urldefense.com/v3/__https://lists.mpi-forum.org/mailman/listinfo/mp… > > > -- Jeff Hammond jeff.science(a)gmail.com http://jeffhammond.github.io/

1 0

No meeting this week
by Joseph Schuchart 22 Nov '21

22 Nov '21

Dear all, As said at the last meeting, we won't meet this week because of Thanksgiving in the US. We will meet again on December 7. See you then, Joseph

1 0

Re: [mpiwg-rma] MPI RMA-WG Bi-Weekly Meeting
by Joseph Schuchart 09 Nov '21

09 Nov '21

We should meet again after the break, as usual at 12 central. Everyone should be on daylight saving time now so times are back to what they used to be. I'd like to continue the discussion on broader picture and potentially start the discussion on atomic operations. See you then Joseph On 9/28/21 11:08 AM, Joseph Schuchart wrote: > Quick reminder that the RMA WG is meeting today at 12 central. > > See you there, > Joseph > > On 9/13/21 7:31 PM, schuchart(a)icl.utk.edu wrote: >> >> Sorry for the late invite. I hope Zoom works for everyone (and I >> didn't mess up the setup) :) >> >> >> MPI RMA-WG Bi-Weekly Meeting >> >> /When/ >> Tue Sep 14, 2021 1pm – 2pm Eastern Time - New York >> /Where/ >> https://tennessee.zoom.us/j/99393831574(map >> <https://www.google.com/url?q=https%3A%2F%2Ftennessee.zoom.us%2Fj%2F99393831…>) >> >> /Who/ >> >> • >> schuchart(a)icl.utk.edu- organizer >> >> Hi there, >> >> Joseph Konstantin Schuchart is inviting you to a scheduled Zoom meeting. >> >> Join from PC, Mac, Linux, iOS or Android: >> https://tennessee.zoom.us/j/99393831574 >> <https://www.google.com/url?q=https%3A%2F%2Ftennessee.zoom.us%2Fj%2F99393831…> >> Password: 469555 >> >> Or iPhone one-tap (US Toll): +16468769923,99393831574# or >> +13017158592,99393831574# >> >> Or Telephone: Dial: +1 646 876 9923 (US Toll) +1 301 715 8592 (US >> Toll) +1 312 626 6799 (US Toll) +1 669 900 6833 (US Toll) +1 253 215 >> 8782 (US Toll) +1 346 248 7799 (US Toll) Meeting ID: 993 9383 1574 >> International numbers available: >> https://tennessee.zoom.us/u/aPxIWlo5F >> <https://www.google.com/url?q=https%3A%2F%2Ftennessee.zoom.us%2Fu%2FaPxIWlo5…> >> >

1 0