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] PCI device name question
From: Brice Goglin (Brice.Goglin_at_[hidden])
Date: 2012-03-20 12:02:35


Le 20/03/2012 16:52, Jeffrey Squyres a écrit :
> On Mar 20, 2012, at 11:09 AM, Brice Goglin wrote:
>
>>> I looked at that. It's not immediately clear to me how to map that to "ib0" and "ib1", for example.
>> Also I see that /sys/class/infiniband/mlx4_0/node_guid contains
>> 0002:c903:000b:5348, that's ib0's end-of-address minus one, and ib1's
>> end-of-address minus two.
> Yep -- that's a carryover from IB. I don't remember offhand whether that GUID pattern is guaranteed by the device/port abstraction, or whether Mellanox equipment just happens to always do that.
>
> But I still don't know how to map that to ib0/ib1.

Actually, what we don't know is how to map that to port 1/2 (we have
ib0/ib1 mac addresses, those are = GUID+1/2 on my machine)

Brice