LAM/MPI logo

LAM/MPI General User's Mailing List Archives

  |   Home   |   Download   |   Documentation   |   FAQ   |   all just in this list

From: Nihar Sanghvi (nsanghvi_at_[hidden])
Date: 2004-03-17 10:51:57


Just a post for the mail archieves

Powered by LAM/MPI...
---------------------------------------
Nihar Sanghvi
LAM/MPI Team
Graduate Student (Indiana University)
http://www.lam-mpi.org
--------------------------------------

---------- Forwarded message ----------
Date: Wed, 17 Mar 2004 10:16:39 -0500
From: David Cronk <cronk_at_[hidden]>
To: nsanghvi_at_[hidden]
Subject: Re: LAM: LAM and Romio with derived datatypes

Thanks for the response. Now that you pointed out the section, I
remember reading this previously, but it evidently did not sink in. I
guess I will try to re-write my real example using type vector.

Thanks again,
Dave.

Nihar Sanghvi wrote:
> Dave --
>
> You are using subarray which has been defined as part of MPI2 standard.
>
> Current version of LAM does not have a full integration of ROMIO and its
> MPI2 datatypes cannot be used with ROMIO. You will be able to find
> documented under LAM user manual (section 5.2.6 and section 12.2)
>
> Thanks,
>
> Nihar
>
>
> On Tue, 16 Mar 2004, David Cronk wrote:
>
>
>>I am having some problems using Romio with LAM/MPI. In particular, with
>>derived datatypes. Attached is a simple example program that does not
>>work. This works fine with MPICH and Romio, but not LAM/MPI with Romio.
>> I have tried with 2 different LAM-7.0.4 installations, installed by
>>different people in different domains. These are both running on Linux
>>Clusters using Redhat. I did not do either install so I don't have
>>config output or anything like that.
>>
>>When I run my simple example I get the following error output:
>>
>>mpirun -np 2 a.out
>>[0] About to set view
>>Error: Unsupported datatype passed to ADIOI_Count_contiguous_blocks
>>[1] About to set view
>>Error: Unsupported datatype passed to ADIOI_Count_contiguous_blocks
>>-----------------------------------------------------------------------------
>>One of the processes started by mpirun has exited with a nonzero exit
>>code. This typically indicates that the process finished in error.
>>If your process did not finish in error, be sure to include a "return
>>0" or "exit(0)" in your C code before exiting the application.
>>
>>PID 14977 failed on node n1 (160.36.140.86) with exit status 1.
>>-----------------------------------------------------------------------------
>>
>>Note that I get a similar error when using some other (but not all)
>>derived datatypes as well. As I said, this simple program works fine
>>with MPICH. It also works fine on an Origin 3800 using the native MPI.
>> This convinces me there is nothing wrong with the code.
>>
>>Has anyone experienced a similar situation? Any advice is greatly
>>appreciated.
>>
>>Thanks,
>>Dave Cronk.
>>
>>--
>>Dr. David Cronk, Ph.D. phone: (865) 974-3735
>>Research Leader fax: (865) 974-8296
>>Innovative Computing Lab
>>http://www.cs.utk.edu/~cronk
>>University of Tennessee, Knoxville
>>
>
>
>
> Powered by LAM/MPI...
> ---------------------------------------
> Nihar Sanghvi
> LAM/MPI Team
> Graduate Student (Indiana University)
> http://www.lam-mpi.org
> --------------------------------------
>
> _______________________________________________
> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
>

-- 
Dr. David Cronk, Ph.D.                             phone: (865) 974-3735
Research Leader                                    fax: (865) 974-8296
Innovative Computing Lab
http://www.cs.utk.edu/~cronk
University of Tennessee, Knoxville