Re: compiling charm++ with g++ 4.1

From: Leandro Martínez (leandromartinez98_at_gmail.com)
Date: Mon Nov 20 2006 - 14:27:43 CST

Thanks Brian,
Actually Gengbin Zheng answered right now and the problem is solved
in the latest cvs versions of charm++.
Leandro.

On 11/20/06, Brian Bennion <bennion1_at_llnl.gov> wrote:
>
> Hello Leandro,
> One possible fix is to delete the extra MsPacker term in the following
> line of MsgPacker.H
>
> Line 86: MsgPacker::MsgPacker(.......
>
> to
> Line 86: MsgPacker(.......
> This file is found in charm-5.9/src/ck-com
>
> Brian
>
> At 08:26 AM 11/20/2006, you wrote:
>
> We are having trouble compiling charm++ in amd64 machines running Fedora 5
> and 6.
> Apparently the error is:
>
> %g++ -fPIC -m64 -I../bin/../include -D__CHARMC__=1 -I. -c ComlibManager.C-o
> ComlibManager.o
> MsgPacker.h :86: error: extra qualification 'MsgPacker::' on member
> 'MsgPacker'
>
> This "extra qualification" error seems to be related to the use of g++ 4.1,
> which does
> not accept some things that the older versions of g++ do accept. I could
> not find
> were to modify the .C file in order to get over the error and I could not
> find any
> report of this problem for charm++ compilation (although I found
> descriptions of
> similar errors for other programs).
>
> Does anyone knows how to solve this problem?
>
> By the way, we are trying to compile charm++ from source because we seem
> to
> have a problem in load balancing in our simulations. What happens is that
> the simulations
> do not crash, but we get a single process running on a single processor
> and nothing
> happens. We ran the simulation with debugging options at the hang up
> occurs just
> after a load balancing point. We may have the problem reporte by Jim
> Phillips at:
>
> http://www.ks.uiuc.edu/Research/namd/wiki/index.cgi?NamdOnAMD64
>
> so, we are trying to compile charm++ from source with the suggest
> modifications. However, we are using the already the 2.6b2 version, for
> which he says that the problem "should be fixed".
>
> Is this problem actually fixed in this version?
> Thanks,
> Leandro.
>
>
>
>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:44:11 CST