Open MPI logo

Hardware Locality Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |  

This web mail archive is frozen.

This page is part of a frozen web archive of this mailing list.

You can still navigate around this archive, but know that no new mails have been added to it since July of 2016.

Click here to be taken to the new web archives of this list; it includes all the mails that are in this frozen archive plus all new mails that have been sent to the list since it was migrated to the new archives.

Subject: Re: [hwloc-devel] Hwloc perl binding
From: Samuel Thibault (samuel.thibault_at_[hidden])
Date: 2010-12-16 05:38:39

Guy Streeter, le Wed 15 Dec 2010 23:47:39 +0100, a écrit :
> I noticed that some functions (in or not in the ABI I don't recall right
> now, I'm not at my desk) need the topology parameter only because the
> topology structure holds pointers to platform-specific routines. That seems
> like a confusing generalization of the parameter's meaning/purpose.

In the case of platform-specific routines, that's needed because
depending on how the discovery was done (synthetic topology, loaded
topology of another machine, etc.) the platform-specific routines will
vary, and we depend on that. For instance, some programmers want to be
able to simulate the execution of a distribution algorithm on another
machine topology, in that case the cpu-binding functions should just
succeed without doing anything. It's thus necessary to have the
topology pointer along the cpuset, to be sure of what should be done.