Discussion related to cado-nfs

[Emmanuel Thomé <Emmanuel.Thome@inria.fr>]

Hi,

Please provide:
 - complete build log (both "make cmake" and "make bwc_full")
 - the full command line to bwc.pl
 - complete bwc.pl output
 - the output of "grep Cpus_allowed_list /proc/self/status" (or maybe
   replace "self" by the pid of a running bwc program).

There are many reasons which could cause symptoms similar to what you
observe.

E.

On Sun, Jan 06, 2019 at 10:24:32AM +0100, sara hendrix wrote:
> Hi all,
>  
> I compiled cado-nfs in MPI mode. Now, I am going to run linear algebra step
> (for RSA-512 bits) on multiple machines. I use ./bwc.pl in mpi mode running 
> on
> 1 server and 1 client each have 6 cores, 12 threads and connecting with 
> 10Gib
> ethernet cable. I set different parameters for mpi and threads based on
> readme file in bwc folder. For some parameters all threads on my 2 machines 
> are
> full load. However, when I run ./bwc.pl on one machine it is faster?
> Can anyone explain me the reason please?
>  
> Thanks,
> Sara

> _______________________________________________
> Cado-nfs-discuss mailing list
> Cado-nfs-discuss@lists.gforge.inria.fr
> https://lists.gforge.inria.fr/mailman/listinfo/cado-nfs-discuss