<br><font size=2 face="sans-serif">Unfortunately, this MPI_Aint discussion
really has nothing to do with DCMF. The entrire problem and solution
is with MPICH2.</font>
<br>
<br><font size=2 face="sans-serif">Is there another, less broad, mpich
mailing list that we could use? Maybe a "mpich-development"
or something like that?</font>
<br>
<br><font size=2 face="sans-serif">Michael Blocksome<br>
Blue Gene Messaging Team Lead<br>
Advanced Systems SW Development<br>
blocksom@us.ibm.com<br>
</font>
<br>
<br><tt><font size=2>dcmf-bounces@lists.anl-external.org wrote on 02/28/2008
11:01:09 AM:<br>
<br>
> I'd say this mailing list is better for it right now. Posting on the<br>
> broader mpich-discuss might just distract us.</font></tt>
<br><tt><font size=2>> </font></tt>
<br><tt><font size=2>> Rajeev</font></tt>
<br><tt><font size=2>> <br>
> From: dcmf-bounces@lists.anl-external.org [mailto:dcmf-<br>
> bounces@lists.anl-external.org] On Behalf Of Bob Cernohous<br>
> Sent: Wednesday, February 27, 2008 4:19 PM<br>
> To: dcmf@lists.anl-external.org<br>
> Subject: Re: [dcmf] 32-bit ROMIO: What is the best solution?<br>
</font></tt>
<br><tt><font size=2>> <br>
> bobc@us.ibm.com wrote on 02/08/2008 02:41:40 PM: <br>
> <br>
> >We all realize there are issues with MPI_Aint's in 32 bit implementations.
<br>
> >Using signed 32 bit addresses along with 64 bit offsets can result
in <br>
> >some pretty broken code. I've reproduced several problems
on <br>
> >BGL/BGP/linux. The problems are most obvious with romio
files > 2G or <br>
> >virtual addresses > 2G. <br>
> <br>
> Earlier this month I posted some comments to this mailing list about
<br>
> 32-bit MPICH/ROMIO/MPI_Aint issues (see above and the archive).
<br>
> <br>
> We've started to work on the problem. We're (optionally) making
<br>
> MPI_Aint a 64 bit signed value even when integers/pointers are 32
bit <br>
> values. <br>
> <br>
> We'd like to start sending fix patches to the community for comment.
<br>
> <br>
> However, since we are (in parallel) contributing our BGP code back
<br>
> to MPICH and trying to converge on a common code base of 1.0.7, <br>
> and because this problem applies to any 32-bit platform, <br>
> we thought we should move the discussion to mpich-discuss to get <br>
> the larger community's comments. <br>
> <br>
> It's unlikely, at this point, that our patches would apply to MPICH2,
<br>
> but that is our goal after the 1.0.7/BGP patch settles. For
now, we'd <br>
> be sending these as RFC (request for comment) patches. <br>
> <br>
> Please let us know if you agree with discussing this on mpich-discuss,
<br>
> or if there is a better place. <br>
> <br>
> Bob Cernohous: (T/L 553) 507-253-6093<br>
> <br>
> BobC@us.ibm.com<br>
> IBM Rochester, Building 030-2(C335), Department 61L<br>
> 3605 Hwy 52 North, Rochester, MN 55901-7829<br>
> <br>
> > Chaos reigns within.<br>
> > Reflect, repent, and reboot.<br>
> > Order shall return._______________________________________________<br>
> dcmf mailing list<br>
> dcmf@lists.anl-external.org<br>
> http://lists.anl-external.org/cgi-bin/mailman/listinfo/dcmf<br>
> http://dcmf.anl-external.org/wiki</font></tt>