Errata: Fortran "vector subscipts" and MPI_BOTTOM

[mpiforumbot]

Originally by RolfRabenseifner on 2014-08-12 10:32:10 -0500

---

Description

The changes are for consistency reasons.
Level is ticket-0.

History

It was overseen in MPI-3.0 that vector subscripts do not work together with
MPI_BOTTOM. Vector subscripts exists only in Fortran.

Extended Scope

None. (No need to add these changes to the erratas' document.)

Proposed Solution

-MPI-3.0 Section 17.1.13, page 629 lines 19-21 read*

Arrays with a vector subscript must not be used
as actual choice buffer arguments in any nonblocking
or split collective MPI operations.
They may, however, be used in blocking MPI operations.

-but should read*

Fortran
arrays with a vector subscript must not be used
as actual choice buffer arguments in any nonblocking
or split collective MPI operation, or together with MPI_BOTTOM.
They may, however, be used in blocking MPI operations.

Alternative Solutions

None.

Impact on Implementations

None required.

Impact on Applications / Users

None.

Entry for the Change Log

None.

[mpiforumbot]

Originally by RolfRabenseifner on 2014-08-13 02:54:10 -0500

---

This Fortran ticket contains a/some part/s of #427 that was split off into tickets #440 - #446.

It should be accepted by the Fortran/Language Binding chapter
committee / WG before the Sep. 2014 meeting in
Japan or voted there with a single errata vote.

[mpiforumbot]

Originally by schulzm on 2014-09-02 03:06:19 -0500

---

This doesn't look like a ticket 0 (i.e., just wording change to me)

[mpiforumbot]

Originally by RolfRabenseifner on 2014-09-05 11:12:42 -0500

---

The only important is, that the additions are done, because vector subscripts must not be used together with MPI_BOTTOM and this is missing in MPI-3.0. It must be an errata to MPI-3.0 because otherwise MPI_BOTTOM would not be implementable in MPI-3.0 if used in blocking routines.

[mpiforumbot]

Originally by jsquyres on 2014-12-05 09:15:48 -0600

---

I think the addition of the word "Fortran" is ticket 0. As chapter author, I accept it. :-)

The addition of MPI_BOTTOM isn't ticket 0, but I agree it's a good (trivial) errata.

[mpiforumbot]

Originally by jsquyres on 2014-12-09 16:26:18 -0600

---

Fortran WG: Upon further review, this ticket doesn't make sense. Withdrawn.