On 2012年05月15日 01:41, Eric Blake wrote:
On 05/14/2012 07:12 AM, Osier Yang wrote:
> "Instead of developing one CPU with 12 cores, the Magny Cours is
> actually two 6 core “Bulldozer” CPUs combined in to one package"
>
> I.e, each package has two NUMA nodes, and the two numa nodes share
> the same core ID set (0-6), which means parsing the cores number
> from sysfs doesn't work in this case.
>
> +++ b/tests/nodeinfodata/linux-nodeinfo-sysfs-test-3-cpu-x86-output.txt
> @@ -1 +1 @@
> -CPUs: 48, MHz: 2100, Nodes: 1, Cores: 6
> +CPUs: 48, MHz: 2100, Nodes: 1, Cores: 12
I would like to tweak this test to output more information; instead of
just outputting nodes and cores, we should also add sockets and threads
into the test output, as well as VIR_NODEINFO_MAXCPUS(nodeinfo) (yes,
this means updating all three test files); doing so will help us ensure
that the numbers we are printing make sense.
But I also plan to add a test-4 on a machine with a cpu that was
hot-unplugged, to show an instance where CPUs is less than maxcpus, so I
can make those changes in my patch. Which means I'm fine with your
patch going in, as-is.
ACK.
Thanks, so I pushed it.
Regards,
Osier