<div dir="ltr">It came to my mind too. It should be easy for compilers to realize it is the only specific procedure of the generic.<div><div>But it looks gcc does not do so. </div></div></div><div class="gmail_extra"><br clear="all">
<div><div dir="ltr">--Junchao Zhang</div></div>
<br><br><div class="gmail_quote">On Thu, May 1, 2014 at 2:20 PM, William Gropp <span dir="ltr"><<a href="mailto:wgropp@illinois.edu" target="_blank">wgropp@illinois.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div style="word-wrap:break-word">The first problem is a quality of Fortran compiler implementation issue - there's no technical reason why the compiler can't tell you that "there is no subprogram MPI_SEND with a real for the second argument", for example. <div>
<br></div><div>Bill</div><div><span class="HOEnZb"><font color="#888888"><br><div>
<span style="border-collapse:separate;color:rgb(0,0,0);font-family:Helvetica;font-style:normal;font-variant:normal;font-weight:normal;letter-spacing:normal;line-height:normal;text-align:-webkit-auto;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;font-size:medium"><span style="border-collapse:separate;color:rgb(0,0,0);font-family:Helvetica;font-style:normal;font-variant:normal;font-weight:normal;letter-spacing:normal;line-height:normal;text-align:-webkit-auto;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;font-size:medium"><div style="word-wrap:break-word">
<span style="border-collapse:separate;color:rgb(0,0,0);font-family:Helvetica;font-style:normal;font-variant:normal;font-weight:normal;letter-spacing:normal;line-height:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;font-size:medium"><div style="word-wrap:break-word">
<div><div style="font-size:12px">William Gropp</div><div style="font-size:12px">Director, Parallel Computing Institute</div></div></div></span><span style="font-size:12px">Thomas M. Siebel Chair in Computer Science</span><span style="border-collapse:separate;color:rgb(0,0,0);font-family:Helvetica;font-style:normal;font-variant:normal;font-weight:normal;letter-spacing:normal;line-height:normal;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;font-size:medium"><div style="word-wrap:break-word">
<div><div style="font-size:12px">University of Illinois Urbana-Champaign</div></div><div><br></div></div></span><br></div></span><br></span><br>
</div>
<br></font></span><div><div><div class="h5"><div>On May 1, 2014, at 1:34 PM, Junchao Zhang wrote:</div><br></div></div><blockquote type="cite"><div><div class="h5"><div dir="ltr"><div>Hello,</div><div> I met a problem when implementing F08 binding in MPICH. I have a file mpi_f08.F90, which contains all interfaces</div>
<div><br></div><div><div>module mpi_f08</div><div>..</div><div>
<div>interface MPI_Send</div><div> subroutine MPI_Send_f08ts(buf, count, datatype, dest, tag, comm, ierror)</div><div> use :: mpi_f08_types, only : MPI_Datatype, MPI_Comm</div><div> implicit none</div><div>
type(*), dimension(..), intent(in) :: buf</div><div> integer, intent(in) :: count, dest, tag</div><div> type(MPI_Datatype), intent(in) :: datatype</div><div> type(MPI_Comm), intent(in) :: comm</div>
<div> integer, optional, intent(out) :: ierror</div><div> end subroutine MPI_Send_f08ts</div><div>end interface MPI_Send</div></div></div>..<div>end module mpi_f08</div><div><br></div><div>Then for each subroutine, I create a separate file which implement the wrapper function in Fortran. For example, send_f08ts.F90 has<br>
</div><div><br></div><div><div>subroutine MPI_Send_f08ts(buf, count, datatype, dest, tag, comm, ierror)</div><div> use :: mpi_f08, only : MPI_Datatype, MPI_Comm</div><div> use, intrinsic :: iso_c_binding, only : c_int</div>
<div> use :: mpi_c_interface, only : c_Datatype, c_Comm</div><div> use :: mpi_c_interface, only : MPIR_Send_cdesc</div><div><br></div><div> implicit none</div><div> ..</div><div>end subroutine MPI_Send_f08ts</div>
</div><div><br></div><div>Bill Long pointed out problems of the implementation. 1) When users supply a bad argument to MPI_Send(), the compiler complains "<span style="font-family:arial,sans-serif;font-size:13px">No specific match can be found for the generic subprogram call MPI_SEND</span>", instead of pointing out the offending arg. 2) The wrapper is not in a module, which <span style="font-family:arial,sans-serif;font-size:13px">eliminates many of the advantages of using modern Fortran.</span></div>
<div><span style="font-family:arial,sans-serif;font-size:13px">I met the first problem in my testing. I was wondering if it is possible to fix that, so I ask Fortran expert here for advices. </span></div><div><span style="font-family:arial,sans-serif;font-size:13px">Thank you very much.</span></div>
<div><br></div><div><div><div dir="ltr">--Junchao Zhang</div></div>
</div></div></div></div><div class="">
_______________________________________________<br>mpiwg-fortran mailing list<br><a href="mailto:mpiwg-fortran@lists.mpi-forum.org" target="_blank">mpiwg-fortran@lists.mpi-forum.org</a><br><a href="http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-fortran" target="_blank">http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-fortran</a></div>
</blockquote></div><br></div></div><br>_______________________________________________<br>
mpiwg-fortran mailing list<br>
<a href="mailto:mpiwg-fortran@lists.mpi-forum.org">mpiwg-fortran@lists.mpi-forum.org</a><br>
<a href="http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-fortran" target="_blank">http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-fortran</a><br></blockquote></div><br></div>