(j3.2006) [Fwd: Re: [MPI3 Fortran] Feedback from Fortran J3 meeting]

[Bill Long]

Somehow Craig managed to remove the "reply-to" J3 on his post.  
Forwarding the reply to the list.


-------- Original Message --------
Subject: 	Re: (j3.2006) [MPI3 Fortran] Feedback from Fortran J3 meeting
Date: 	Wed, 28 May 2008 12:54:21 -0500
From: 	Bill Long <longb at cray.com>
Reply-To: 	longb at cray.com
Organization: 	Cray Inc.
To: 	Craig Rasmussen <crasmussen at lanl.gov>
References: 	<0C6872D7-EE68-4D34-B7C5-553B0EC0BFD3 at lanl.gov> 
<200805271017.20443.donev1 at llnl.gov> 
<726E5F51-395D-4ED6-BCD2-799C1A32B1B3 at verizon.net> 
<200805271448.26076.donev1 at llnl.gov> 
<527CF641-CD8A-439D-B260-DF4363579A74 at verizon.net> 
<E42255E4-2302-490F-B565-4452EFAA3BE5 at lanl.gov> 
<ED5460F5-1F03-450A-BE19-6D738CEB4A50 at verizon.net> 
<483D8B6B.6070409 at cray.com> <776732AF-4528-4B2C-B66F-564E7FF282F7 at lanl.gov>



Craig Rasmussen wrote:
>
>
> But from information I received from MPI Forum, we have to do 
> something to prevent code motion to complete the picture.  What about:
>
>     MPI_Wait(req, ...)    ! completes MPI_Irecv on buf
>     buf(1) = 1

Compile error without the CALL in front of MPI....

If either Craig or Aleks can find any example of a current compiler that 
moves code involving references or definitions of variables with either 
the volatile or asynchronous attributes across ANY call to an external 
routine, then we might need to look further.  

Cheers,
Bill


-- 
Bill Long                                   longb at cray.com
Fortran Technical Support    &              voice: 651-605-9024
Bioinformatics Software Development         fax:   651-605-9142
Cray Inc., 1340 Mendota Heights Rd., Mendota Heights, MN, 55120

           



-- 
Bill Long                                   longb at cray.com
Fortran Technical Support    &              voice: 651-605-9024
Bioinformatics Software Development         fax:   651-605-9142
Cray Inc., 1340 Mendota Heights Rd., Mendota Heights, MN, 55120