Has any one had the MPI 3.2.2.006 mpi_file_write_all error?

Has any one had the MPI 3.2.2.006 mpi_file_write_all error?

Terrence Liao's picture

All,

I have a small testing code to test mpi_file_write_all and read all, it ran on PGI, but on Intel MPI, it fails.
Any clue? or parameters I need to set.

Thanks.

-- Terrence Liao

3 posts / 0 new
Last post
For more complete information about compiler optimizations, see our Optimization Notice.
Gergana Slavova (Intel)'s picture

Hi Terrence,

Welcome to the HPC forums!

It would be great if you could provide some additional information about the failure. What exactly is the error you're seeing? Could you note your compile and run commands here as well? Thanks.

The Intel MPI Library is not a compiler so I'm not fully sure what you mean when you say "it ran on PGI, but on Intel MPI, it fails." What MPI implementation are you using for your PGI compiler? If it's Intel MPI Library, we don't support PGI by default, but we can provide you with a binding kit you can apply to your current installation that will enable that support. Let me know and I can provide that to you.

If not PGI, when running with the Intel MPI Library, are you using the Intel Compilers then? Again, your compile/link line would help here.

Finally, it would be great if you could provide your test code as well. If you're not comfortable sharing that here, go ahead and send it to me via e-mail. That way I can try running it locally.

Thanks and I look forward to hearing from you.

Regards,
~Gergana

Gergana Slavova
Technical Consulting Engineer
Intel® Cluster Tools
E-mail: gergana.s.slavova_at_intel.com
Terrence Liao's picture

Problem gone after cluster reboot. -- Thanks.

Login to leave a comment.