Quantcast
Channel: Clusters and HPC Technology
Viewing all articles
Browse latest Browse all 930

Segmentation fault in MPIU_Handle_obj_alloc_unsafe

$
0
0

Dear Developers,

I am struggling with the following problem:

[2] [2] ERROR: LOCAL:EXIT:SIGNAL: fatal error
[2] [2] ERROR:    Fatal signal 11 (SIGSEGV) raised.
[2] [2] ERROR:    Signal was encountered at:
[2] [2] ERROR:       MPIU_Handle_obj_alloc_unsafe (/tmp/7b663e0dc22b2304e487307e376dc132.xtmpdir.nnlmpicl211.16412_32e/mpi4.32e.nnlmpibld05.20130522/dev/src/util/mem/x86_64/debug_dynamic/../../handlemem.c:353)
[2] [2] ERROR:    While processing:
[2] [2] ERROR:       MPI_Win_lock(lock_type=234, rank=2, assert=0, win=0xffffffffa0000002)
[0] [0] WARNING: starting premature shutdown

I got this diagnostic message by using the Intel ITAC. In our application we do a lot of 1sided passive "put" and "get" operations across working MPI processes. The memory attached to MPI's windows is allocated via 'malloc'. The program is running on the HP SL230s compute server equipped with two Intel E5-2660 (2.2 GHz Sandybridge) processors with 8 cores each (i.e. 16 cores per compute server).  The problem starts only when I start using more than 4 cores on a single node, regardless 'I_MPI_FABRICS' (=shm,dssm) setting and compiler's optimization leve used.  The software stack is listed below:

intel/14.0.1

impi/4.1.1.036

itac/8.1.3.037

I am wondering if there is a quick solution to this problem?

Thank you in advance!

Victor.


Viewing all articles
Browse latest Browse all 930

Trending Articles