[mpich-discuss] MPI-IO and (potentially) unnecessary synchronization

[Rob Latham]

On Sat, May 07, 2011 at 01:45:53PM -0500, Rob Latham wrote:
> I'm reaching way back into the archives for this one.
> 
> I think this barrier is unneeded.  Yes, the file is opened with
> COMM_SELF but the drivers that support shared file pointers
> (ADIO_SHARED_FP) are the posix-compliant ones. 
> 
> POSIX semantics specify that deleting a file does not unlink a file.
> N processors have it open, it only gets removed after N closes. 

Rajeev wanted me to be careful around NFS volumes.  Fair enough: NFS
is dumber than a bag of hammers. 

r8541 fixes this in MPICH2 trunk and will probably make it into the
1.4 release. 

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