Ralph,

i made PR 782 for this
https://github.com/open-mpi/ompi/pull/782
and i am still double checking it.

i am still wondering whether we should add an MCA parameter for the api check
(no test, test and fail, test and warn) ...
or i am even wondering whether we should simply remove this test, since i am not 100% convinced
the test is bulletproof :
if ofed version is different between compilation and runtime, let openmpi crash
since this is a user-side problem.

Cheers,

Gilles

On 8/6/2015 8:43 AM, Ralph Castain wrote:
Yeah, I recall your earlier email on the subject. Sadly, I need someone from Mellanox to look at this as I don't have access to such equipment.

Josh? Mike? Gilles? Can someone please look at this?


On Wed, Aug 5, 2015 at 2:31 PM, Andy Wettstein <wettst...@uchicago.edu <mailto:wettst...@uchicago.edu>> wrote:

    Hi,

    I updated our OpenMPI install from 1.8.3 to 1.8.8 today and I'm
    getting
    this error:

    XRC error: bad XRC API (require XRC from OFED pre 3.12).

    This happens even using the exact same node to compile and run an
    example program. I saw a thread from a few weeks ago discussing this
    issue as well. I changed the dlsym if statement in btl_openib_xrc.c to
    this:

    if (NULL != dlsym(lib, "ibv_create_xrc_recv_qp@@IBVERBS_1.1")) {

    This seems to make the error message go away, so there must be
    something
    about that check that doesn't work right without the version in it.

    This is on a Scientific Linux 6.6 machine with MLNX_OFED 1.5.3-4.0.42.

    Andy

    --
    andy wettstein
    hpc system administrator
    research computing center
    university of chicago
    773.702.1104 <tel:773.702.1104>

    _______________________________________________
    users mailing list
    us...@open-mpi.org <mailto:us...@open-mpi.org>
    Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users
    Link to this post:
    http://www.open-mpi.org/community/lists/users/2015/08/27400.php




_______________________________________________
users mailing list
us...@open-mpi.org
Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users
Link to this post: 
http://www.open-mpi.org/community/lists/users/2015/08/27404.php

Reply via email to