Open MPI logo

Open MPI User's Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Open MPI User's mailing list

From: Jeff Squyres \(jsquyres\) (jsquyres_at_[hidden])
Date: 2006-04-25 10:12:49


> -----Original Message-----
> From: users-bounces_at_[hidden]
> [mailto:users-bounces_at_[hidden]] On Behalf Of Michael Kluskens
> Sent: Tuesday, April 25, 2006 9:56 AM
> To: Open MPI Users
> Subject: [OMPI users] f90 module files compile a lot faster
>
> Strange thing, with the latest g95 and the last OpenMPI 1.1
> (a3r9704)
> [on OS X 10.4.6] there does not seem to be the compilation penalty
> for using "USE MPI" instead of "include mpi.h" that there used to be.
>
> My test programs compile almost instantly.

Blast. We changed the default F90 compile options but did not update
any of the documentation. Sigh. I'll fix that.

There are now 4 "sizes" of the F90 MPI bindings:

- trivial: includes a handful of MPI-2 F90-specific functions only
- small: trivial + all MPI functions that take no choice buffers
- medium: small + all MPI functions that take 1 choice buffer (this was
the only option in OMPI 1.0.x)
- large: medium + all MPI functions that take 2 choice buffers, but only
when both buffers are of the same type

The default is currently "small". So this is different than it was in
OMPI 1.0, but the compile time savings were judged to be worth it.

Sorry about the docs -- I can't believe I missed that. :-)

> However, I'm still seeing:
>
> [a.b.c.d:20221] [0,0,0] ORTE_ERROR_LOG: Not found in file base/
> soh_base_get_proc_soh.c at line 80
> [a.b.c.d:20221] [0,0,0] ORTE_ERROR_LOG: Not found in file base/
> oob_base_xcast.c at line 108
> [a.b.c.d:20221] [0,0,0] ORTE_ERROR_LOG: Not found in file base/
> rmgr_base_stage_gate.c at line 276
> [a.b.c.d:20221] [0,0,0] ORTE_ERROR_LOG: Not found in file base/
> soh_base_get_proc_soh.c at line 80
> [a.b.c.d:20221] [0,0,0] ORTE_ERROR_LOG: Not found in file base/
> oob_base_xcast.c at line 108
> [a.b.c.d:20221] [0,0,0] ORTE_ERROR_LOG: Not found in file base/
> rmgr_base_stage_gate.c at line 276

Ralph -- I'm going to throw this one to you. Can you poke around?

I think that this means that the SOH is saying that it found more than
one proc state / exit code in the GPR.

-- 
Jeff Squyres
Server Virtualization Business Unit
Cisco Systems