MPI Forum Meetings logo

MPI Forum meeting: Votes and Formal Readings from the meeting

Meetings home |  Secretary notes |  March 2012 |  Votes

Meeting date: March 5-7, 2012
Meeting location: Chicago, CA, USA


Formal Readings:

NOTE: As of the end of the meeting, the owners of these tickets intend to bring them for vote at the next meeting.

  1. Ticket #187: For reductions: Grouping MPI_AINT, MPI_OFFSET, MPI_COUNT as Multi-language types
  2. Ticket #192: C++ binding for Dist_graph_neighbors_count is badly broken
  3. Ticket #194: Ensure MPI_Dims_create is really suitable for MPI_Cart_create
  4. Ticket #195: Topology awareness in MPI_Dims_create
  5. Ticket #217: MPI3 Hybrid Programming: Proposal for Helper Threads
  6. Ticket #256: MPI_PROC_NULL behavior for MPI_PROBE and MPI_IPROBE not directly defined
  7. Ticket #271: Functions to query MPI_Info object attached to windows and communicators
  8. Ticket #273: Add Immediate versions of nonblocking collective I/O routines (re-reading from last meeting)
  9. Ticket #278: Update examples to not use deprecated constructs (where possible)
  10. Ticket #281: Remove C++ bindings (re-reading from last meeting)
  11. Ticket #294: MPI_UNWEIGHTED should not be NULL (re-reading from last meeting)
  12. Ticket #300: Fix issue with definition of nonblocking in One Sided Chapter
  13. Ticket #303: Move MPI-2 deprecated functions to new "Removed interfaces" chapter
  14. Ticket #310: clarify MPI behavior when multiple MPI processes run in the same address space
  15. Ticket #311: provide a mechanism for migrating threads between MPI processes
  16. Ticket #313: MPI_INIT & MPI_FINALIZE
  17. Ticket #317: correct error related to MPI_REQUEST_FREE
  18. Ticket #323: New FT proposal
  19. Ticket #324: Clarify MPI_ERRORS_ARE_FATAL scope of abort (this was changed to ticket 0 -- SEE TICKET FOR MORE DETAILS)
  20. Ticket #326: MPI3 Fault Tolerance - Files
  21. Ticket #327: MPI3 Fault Tolerance - Dynamic process management
  22. Ticket #328: fix MPI_PROC_NULL behavior for mprobe/improbe/mrecv/imrecv


Votes:

NOTE: "Missed" counts the same as "abstain".

  1. Vote topic: Ticket #0 (usually small spelling and grammar fixes): "Big ticket 0 glob: #162, #182, #185, #207, #215, #219, #254, #255, #262, #263, #312, #322", Final 1st-and-only vote (ticket 0)

    YES: 17
    NO: 0
    ABSTAIN: 0
    MISSED: 4
    Result: Ballot passed


  2. Vote topic: Ticket #125: "Use of [] in output arrays ", 1st-and-only vote (ticket 0)

    YES: 18
    NO: 0
    ABSTAIN: 0
    MISSED: 3
    Result: Ballot passed


  3. Vote topic: Ticket #126: "Consistent use of [] for input arrays ", 1st-and-only vote (ticket 0)

    YES: 18
    NO: 0
    ABSTAIN: 0
    MISSED: 3
    Result: Ballot passed


  4. Vote topic: Ticket #293: "K&R style on 6.7.6 ", 1st-and-only vote (ticket 0)

    YES: 18
    NO: 0
    ABSTAIN: 0
    MISSED: 3
    Result: Ballot passed


  5. Vote topic: Ticket #321: "MPI_WAIT/TEST language unclear about non blocking I/O, Generalized, or collective requests. ", 1st-and-only vote (ticket 0)

    YES: 19
    NO: 0
    ABSTAIN: 0
    MISSED: 2
    Result: Ballot passed


  6. Vote topic: Ticket #158: "Clarification that MPI_Cart_map and MPI_Graph_map are local calls ", 1st-and-only vote (ticket 0)

    YES: 16
    NO: 2
    ABSTAIN: 1
    MISSED: 2
    Result: Ballot passed


  7. Vote topic: Ticket #259: "Text updates for MPI_DIST_GRAPH ", 1st-and-only vote (ticket 0)

    YES: 17
    NO: 1
    ABSTAIN: 0
    MISSED: 3
    Result: Ballot passed


  8. Vote topic: Ticket #308: "RMA - Remove sentence on offset in datatypes ", 1st vote

    YES: 17
    NO: 0
    ABSTAIN: 2
    MISSED: 2
    Result: Ballot passed


  9. Vote topic: Ticket #309: "RMA - Clarify usage of status for request-based RMA operations ", 1st vote

    YES: 19
    NO: 0
    ABSTAIN: 0
    MISSED: 2
    Result: Ballot passed


  10. Vote topic: Ticket #284: "Allocate a shared memory window ", 1st vote

    YES: 16
    NO: 0
    ABSTAIN: 3
    MISSED: 2
    Result: Ballot passed


  11. Vote topic: Ticket #280: "Add a new datatype creation routine - MPI_Type_create_hindexed_block ", 1st vote

    YES: 19
    NO: 0
    ABSTAIN: 1
    MISSED: 1
    Result: Ballot passed


  12. Vote topic: Ticket #168: "Nonblocking Communicator Duplication ", 1st vote

    YES: 17
    NO: 0
    ABSTAIN: 1
    MISSED: 3
    Result: Ballot passed


  13. Vote topic: Ticket #272: "Remove non-blocking collective C++ bindings ", 1st vote

    YES: 17
    NO: 0
    ABSTAIN: 2
    MISSED: 2
    Result: Ballot passed


  14. Vote topic: Ticket #286: "Noncollective Communicator Creation ", 1st vote

    YES: 19
    NO: 0
    ABSTAIN: 1
    MISSED: 1
    Result: Ballot passed


  15. Vote topic: Ticket #305: "Update MPI_Intercomm_create to use collective tag space ", 1st vote

    YES: 19
    NO: 0
    ABSTAIN: 1
    MISSED: 1
    Result: Ballot passed


  16. Vote topic: Ticket #298: "Adding clarification to MPI_WIN_LOCK_ALL ", 2nd vote

    YES: 20
    NO: 0
    ABSTAIN: 0
    MISSED: 1
    Result: Ballot passed


  17. Vote topic: Ticket #229: "New MPI-3 Fortran Support - #A: Overview over all related tickets ", 2nd vote

    YES: 19
    NO: 0
    ABSTAIN: 1
    MISSED: 1
    Result: Ballot passed


  18. Vote topic: Ticket #299: "Change displs arguments in neighborhood w collectives to MPI_Aint ", 2nd vote

    YES: 19
    NO: 0
    ABSTAIN: 1
    MISSED: 1
    Result: Ballot passed