[Mpi-21] [mpi-21] problem with MPI_Get_count and MPI_Probe
treumann at [hidden]
Wed Mar 5 15:52:50 CST 2008
I did not notice this until I was going over the Ballot 4 slides. I would
prefer not to vote in the currently proposed wording
The correction is for page 22, not 222
MPI 1.1, page 22, line 48 reads
used after a call to MPI_PROBE. (End of rationale.)
but should read
used after a call to MPI_PROBE or MPI_IPROBE.
In this case the call to MPI_GET_COUNT should use the same
datatype as will be used in the MPI_RECV.
(End of rationale.)
(Advice to users.)
The buffer size required for the receive can be affected by
data conversions and by the stride of the receive
datatype. Calling MP_GET_COUNT with MPI_BYTE to calculate the
receive buffer size may not be portable unless the MPI_RECV
(End of advice to users.)
Dick Treumann - MPI Team/TCEM
IBM Systems & Technology Group
Dept 0lva / MS P963 -- 2455 South Road -- Poughkeepsie, NY 12601
Tele (845) 433-7846 Fax (845) 433-8363
mpi-21-bounces_at_[hidden] wrote on 01/18/2008 12:55:09 PM:
> This is a proposal for MPI 2.1, Ballot 4.
> This is a follow up to:
> Datatypes and MPI_PROBE
> in http://www.cs.uiuc.edu/homes/wgropp/projects/parallel/MPI/mpi-
> with mail discussion in
> Proposal for MPI 2.1, Ballot 4:
> MPI 1.1, page 222, line 48 reads
> used after a call to MPI_PROBE. (End of rationale.)
> but should read
> used after a call to MPI_PROBE or MPI_IPROBE.
> With a status returned from MPI_PROBE or MPI_IPROBE, the same
> dataypes are allowed as in a call to MPI_RECV to receive this message.
> (End of rationale.)
> Advice to users. To allocate the appropriate amount of memory as
> buffer, the same datatype as in the following receive call should be
> to determine the needed space. In portable programs due to possible
> conversions, it is not guaranteed that the count returned by
> with datatype MPI_BYTE is the correct amount of needed memory space in
> the receive buffer (although MPI_BYTE is matching every datatype).
> (End of advice to users.)
> Reason for the first part: The current MPI-1.1 text says "The
> datatype argument
> should match the argument provided by the receive call that set the
> variable." With MPI_PROBE, there isn't such a receive call.
> Reason for the advice to users: It helps to write portable code.
> Because malloc needs a byte count, users may write wrong programs
> by using MPI_BYTE.
> Discussion should be done through the new mailing list
> I have sent out this mail with CC through the old general list
> Best regards
> Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner_at_[hidden]
> 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: Allmandring 30)
> mpi-21 mailing list
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Mpi-21