[mpich-discuss] Facing problem while using MPI_File_set_view

[Rob Latham]

On Mon, May 25, 2009 at 05:19:42PM -0400, Christina Patrick wrote:
> I am writing a program to read an array of 16384 x 16384 of type
> double ~ 2GB file size
..
> However, I do not want to use a buffer as big as 512 MB. So I use a
> smaller buffer (such as 8MB) and iterate over the file view of P0 (P1,
> P2, P3), my program starts throwing errors/get segmentation faults:
> [E 16:49:28.377861] Error: payload_progress: Bad address
> OR
> segmentation fault in ADIOI_Calc_my_off_len() at line: while
> (flat_file->type != fd->filetype) flat_file = flat_file->next;
> because flat_file becomes 0x0 in one of the processes.
> 
> If I use a column view and logically do the same thing, I do not face
> this problem. (I hope that I have been able to explain the problem. In
> case of doubt, please let me know.)

Always good to have a test program.  Thanks for sending one along.
What version of mpich are you using? 

When I try with MPICH2-1.1rc1, I get an error:

MPI_Type_create_subarray(161): Total size of array_of_sizes[] is 2147483648 which requires the MPI implementation to use a 64-bit MPI_Aint

Looks like you've run into a not-uncommon problem with
MPI_TYPE_CREATE_SUBARRAY and 32 bit platforms.  

Can you try again with 1.1rc1 ?

I think the problem with the UFS and PVFS cases might lie with some
value overflowing.  

When I reduce the size of the subarray (rows = 16k, cols = 8k), I
don't get any errors with a col_bufsize of 8MB on either UFS or PVFS2

Thanks
==rob

-- 
Rob Latham
Mathematics and Computer Science Division
Argonne National Lab, IL USA