[openib-general] MVAPICH failure on SGI Altix SLES10

[Tziporet Koren]

John Partridge wrote:
> I am trying to run the example from MPI_README.txt (and other MPI apps
> like pallas), but I keep getting a Couldn't modify SRQ limit error
> message :-
>
> mig129:~/OFED-1.0-pre1 # 
> /usr/local/ofed/mpi/gcc/mvapich-0.9.7-mlx2.1.0/bin/mpirun_rsh -rsh -np 2 
> -hostfile /root/cluster 
> /usr/local/ofed/mpi/gcc/mvapich-0.9.7-mlx2.1.0/tests/osutests-1.0/bw 1000 16
> [1] Abort: Couldn't modify SRQ limit
>   at line 995 in file viainit.c
> mpirun_rsh: Abort signaled from [1]
> [0] Abort: [mig125:0] Got completion with error, code=12
>   at line 2143 in file viacheck.c
> done.
>
> I am using OFED-1.0-pre1 (kernel modules are from OFED-1.0-pre1 also)
> OS is SLES10 SUSE Linux Enterprise Server 10 (ia64) VERSION = 10
>
> HW is SGI Altix ia64
>
> Can anyone help please ?
>
> Thanks
> John
>
>   
I guess you use older FW version. See in osu_mpi release notes:

- For users of Mellanox Technologies firmware fw-23108 or fw-25208 only:
  OSU MPI may fail in its default configuration if your HCA is burnt with an
  fw-23108 version that is earlier than 3.4.000, or with an fw-25208 version
  4.7.400 or earlier.
  Workaround:
  Option 1 - Update the firmware.
  Option 2 - In mvapich.conf, set VIADEV_SRQ_ENABLE=0


Tziporet