Yes, the cpuid version works marvelously. The only thing it misses is 
memory/NUMA node (is it using libnuma?).

It also works for the dual Harpertown machine.

I will try HT now.

Thanks,
Dan

-----Original Message-----
From: hwloc-users-boun...@open-mpi.org 
[mailto:hwloc-users-boun...@open-mpi.org] On Behalf Of Samuel Thibault
Sent: Thursday, January 28, 2010 11:35 AM
To: Hardware locality user list
Subject: Re: [hwloc-users] No caches or hiearchy on RHEL 4.7 or 4.8

Dan Eaton, le Thu 28 Jan 2010 11:22:06 -0700, a écrit :
> I couldn't compile trunk (rather, autogen.sh failed because it requires 
> Autoconf > 2.63), so I just pulled the nightly build hwloc-1.0a1r1682 and 
> applied the patch there. Does that sound reasonable?

Yep, it should be fine.

> I've attached the new (much more verbose) result of lstopo.

Ok, so my cpuid backend indeed detects a dual quadcore.  I guess there
is something wrong in RHEL's kernel with APIC ID interpretation.
Maybe the kernel is disturbed by the fact that APIC IDs are not
contiguous. Could you try to enable the HT feature?

Samuel
_______________________________________________
hwloc-users mailing list
hwloc-us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-users

Reply via email to