RE: communication pattern causes corrupted data

"Joshua J. England" <[email protected]> 25 Mar 2003 14:12:44 -0800
Newsgroups gmane.network.myrinet.general
Message-ID <[email protected]>
I did a full recompile of the redhat 7.1 glibc explicitly setting
host=alphaev56-unknown-linux-gnu.  Just to be sure, I also recompiled
gm-1.6.3 and mpich-gm-1.2.4..8a.

The problem still occurs.

In doing some other tests, I have found that under gm-1.4/mpich-1.2..5
the problem does not exist under RH7.1.

Is gm-1.4+ allocating buffer space differently?  What else could be
going wrong here?

-JE.
-----------------------------------------------
Josh England
Sandia National Laboratory, Livermore, CA
High Performance Computing and Networking
email: [email protected]
phone: (925) 294-2076


On Mon, 2003-03-24 at 09:11, David B. Ritch wrote:
> I believe that we only recompiled glibc.  As a minimal change, you might
> even be able to get by with only recompiling a few routines, such as
> memcpy and/or bcopy, but I wouldn't really recommend that.
> 
> We observed this problem with one application, but when we updated
> glibc, several other undiagnosed problems went away.
> 
> dbr
> 
> On Mon, 2003-03-24 at 09:52, England, Joshua J wrote:
> > David, 
> > 
> > Thanks for the info.  Is it only glibc that needs recompiling, or should
> > other system libraries be done as well?
> > 
> > -JE
> 
> -- 
> David B. Ritch
> High Performance Technologies, Inc.