Open MPI logo

Open MPI User's Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Open MPI User's mailing list

Subject: Re: [OMPI users] error on malloc
From: Jeff Squyres (jsquyres_at_[hidden])
Date: 2011-09-12 10:24:44


This usually means that you have a memory error of some kind in your application.

Have you tried running your application through a memory-checking debugger, such as valgrind?

On Sep 5, 2011, at 3:48 AM, Jai Dayal wrote:

> Hi all,
> I've been beating my head on this for quite a while now. I don't have this problem when running with 1,2, or 3 procs, however, once I get to 4 or beyond, I have a problem.
>
> When I call malloc, I get this error:
>
> txserver:11055 terminated with signal 11 at PC=2b46886bc18a SP=7fff20f51030. Backtrace:
> /apps/x86_64/mpi/openmpi/gcc-4.3.4/openmpi-1.4.3_oobpr/lib/libopen-pal.so.0(opal_memory_ptmalloc2_int_malloc+0x54a)[0x2b46886bc18a]
> /apps/x86_64/mpi/openmpi/gcc-4.3.4/openmpi-1.4.3_oobpr/lib/libopen-pal.so.0[0x2b46886bd4f3]
> txserver[0x415769]
> txserver[0x40da8c]
> txserver[0x4344bb]
> txserver[0x4351cd]
> txserver[0x40e3d4]
> /lib64/libc.so.6(__libc_start_main+0xf4)[0x2b468a5af994]
> txserver(_ZNSt8ios_base4InitD1Ev+0x39)[0x40c889]
>
> However, only rank 0 calls this function. Which is strange. I can just put a dummy malloc in there (int * dummy = (int *)malloc(10);) for example, and it will still crash.
>
> Again, this does not happen with n < 4 procs. The crash happens on rank 0, as it's the only rank that calls this code...
>
> I'm perplexed.
>
> Thanks a lot,
> J.D.
>
> _______________________________________________
> users mailing list
> users_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/users

-- 
Jeff Squyres
jsquyres_at_[hidden]
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/