[mpich-discuss] MPI-IO of multidimensional arrays - issue reported by valgrind

[Rob Latham]

On Tue, Feb 23, 2010 at 02:20:51PM +0000, Ashley Pittman wrote:
 
> > We should probably be more careful with the 'conditional jump or move'
> > warnings, but they rarely if ever point to an actual defect. 
> 
> 
> That's a bold statement, it's true that they often appear to have no
> effect but if you are seeing an error in combination with a hang
> then I'd certainly look at it closely.

OK everyone! sorry for blaspheming the valgrind gods :>  

"hang in I/O", especially on GPFS on BlueGene, especially with
subarray file views, is almost always two issues: 

- a file locking problem, 
- or it's a failure to check the return code of an MPI routine. 

But I *will* spend some time to clean up the valgrind warnings. 

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