[openib-general] [Bug 146] OFED-1.0 DAPL fails to build on SLES10 on IA64 with IA64_FETCHADD error

[John Partridge]

James Lentini wrote:
> This is included on IA64 and PPC systems. Since we have not done 
> testing on IA64 or PPC, I'm certain that this was a contribution for a 
> IA64 or PPC DAPL user. For that reason, I'm not certain why the 
> asm/system.h was included. It is certainly possible that it is 
> necessary on PPC but not IA64.
> 

James,

Thanks your patch has fixed the build errors. But I do suggest we leave
system.h included for PPC until we know we don't need it (see my patch I 
sent to you last week).

The resulting build from your last patch has been installed and we are
in the process of DAPL tests now. I do know that the libdat works with
Intel MPI (although we had to manually create a symlink from libdat.so.1
to libdat.so - should this not already exist?)

I do have one question about how the dapl RPM's are organized, we are
creating a DAPL interface for the ccNUMA xpmem on SGI Altix systems.
At the moment we have a libdat and libdapl(xpmem). It is our objective
to use the OFED-1.0 libdat, as libdat will be used for non-infiniband
interfaces I don't quite understand why libdat (dat.conf) are not a
separate RPM and the libdapl interfaces installed in a separate RPM ?
Would his not make more sense ?

Thanks
John


-- 
John Partridge

Silicon Graphics Inc
Tel:  651-683-3428
Vnet: 233-3428
E-Mail: johnip at sgi.com