[dcmf] 32-bit ROMIO: What is the best solution?

[Rajeev Thakur]

I'd say this mailing list is better for it right now. Posting on the broader
mpich-discuss might just distract us.
 
Rajeev


  _____  

From: dcmf-bounces at lists.anl-external.org
[mailto:dcmf-bounces at lists.anl-external.org] On Behalf Of Bob Cernohous
Sent: Wednesday, February 27, 2008 4:19 PM
To: dcmf at lists.anl-external.org
Subject: Re: [dcmf] 32-bit ROMIO: What is the best solution?



bobc at us.ibm.com wrote on 02/08/2008 02:41:40 PM: 

>We all realize there are issues with MPI_Aint's in 32 bit implementations. 
>Using signed 32 bit addresses along with 64 bit offsets can result in 
>some pretty broken code.  I've reproduced several problems on 
>BGL/BGP/linux.   The problems are most obvious with romio files > 2G or 
>virtual addresses > 2G. 

Earlier this month I posted some comments to this mailing list about 
32-bit MPICH/ROMIO/MPI_Aint issues (see above and the archive).     

We've started to work on the problem.   We're (optionally) making 
MPI_Aint a 64 bit signed value even when integers/pointers are 32 bit 
values. 

We'd like to start sending fix patches to the community for comment. 

However, since we are (in parallel) contributing our BGP code back 
to MPICH and trying to converge on a common code base of 1.0.7, 
and because this problem applies to any 32-bit platform, 
we thought we should move the discussion to mpich-discuss to get 
the larger community's comments.   

It's unlikely, at this point, that our patches would apply to MPICH2, 
but that is our goal after the 1.0.7/BGP patch settles.   For now, we'd 
be sending these as RFC (request for comment) patches. 

Please let us know if you agree with discussing this on mpich-discuss, 
or if there is a better place. 

Bob Cernohous:  (T/L 553) 507-253-6093

BobC at us.ibm.com
IBM Rochester, Building 030-2(C335), Department 61L
3605 Hwy 52 North, Rochester,  MN 55901-7829

> Chaos reigns within.
> Reflect, repent, and reboot.
> Order shall return.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alcf.anl.gov/pipermail/dcmf/attachments/20080228/cb9a96c3/attachment.htm>