Charm++ parallel programming system

[Jim Phillips <jim AT ks.uiuc.edu>]

There actually never was native x86_64 support for MX because InfiniBand 
and x86_64 took over the market at about the same time MX came out. 
There were Itanium and 32-bit x86 Myrinet clusters but they used the older 
GM interface, and the only recent Myrinet clusters have been PowerPC-based 
(Indiana's Big Red, Mare Nostrum, and the UIUC Turing cluster).

Since it exists on other 64-bit platforms the port should be pretty easy, 
just copying conv-mach-mx.h and conv-mach-mx.sh to arch/net-linux_86_64 
from another platform and figuring out what the right #defines are.

-Jim

On Wed, 15 Jun 2011, Thomas Albers wrote:

> Hello!
>
> I am trying to build Charm-6.3.2 with Myrinet MX support on Linux,
> however the build fails:
>
> ta@porsche
>  ~/home/ta//NAMD_2.8_Source/charm-6.3.2 $ ./build charm++
> net-linux-x86_64  mx --with-production
> ...
> Fatal Error by charmc in directory
> /root/NAMD_2.8_oldsource/charm-6.3.2/net-linux-x86_64-mx/tmp
>   Command gcc -m64 -I/opt/open-mx/include -I../bin/../include
> -D__CHARMC__=1 -DFOR_CPLUS=1 -O -fno-stack-protector -c machine.c -o
> machine.o returned error code 1
>
> ta@porsche
>  ~/home/ta//NAMD_2.8_Source/charm-6.3.2 $ cd
> net-linux-x86_64-mx/tmp
> ta@porsche
> ~/home/ta//NAMD_2.8_Source/charm-6.3.2/net-linux-x86_64-mx/tmp $ gcc
> -m64 -I/opt/open-mx/include -I../bin/../include -D__CHARMC__=1
> -DFOR_CPLUS=1 -O -fno-stack-protector -c machine.c -o machine.o
> machine.c: In function 'InternalWriteToTerminal':
> machine.c:1482: warning: ignoring return value of 'write', declared with
> attribute warn_unused_result
> machine.c: At top level:
> machine.c:2425: error: redefinition of 'CmiBarrier'
> machine-mx.c:737: note: previous definition of 'CmiBarrier' was here
> machine.c:2455: error: redefinition of 'CmiBarrierZero'
> machine-mx.c:784: note: previous definition of 'CmiBarrierZero' was here
>
> What is happening here - has native MX support been withdrawn for this
> platform, and the recommended way is to use any MPI, is there something
> that I am missing, or is this a genuine bug?
>
> Regards,
> Thomas
> _______________________________________________
> charm mailing list
> charm AT cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/charm
> _______________________________________________
> ppl mailing list
> ppl AT cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/ppl