The nodeGetInfo code was always assuming that machine had a single NUMA node, which is not correct. The good news is that libnuma gives us this information pretty easily, so let's properly report it.

NOTE: With recent hardware starting to support CPU hot-add and hot-remove, both this code and the nodeCapsInitNUMA() code are quickly going to become obsolete. We'll have to think of a more dynamic solution for dealing with NUMA nodes and CPUs that can come and go at will.

Signed-off-by: Chris Lalancette <clalance(a)redhat.com&gt; --- src/nodeinfo.c | 6 +++++- 1 files changed, 5 insertions(+), 1 deletions(-) diff --git a/src/nodeinfo.c b/src/nodeinfo.c index 0748602..8d7e055 100644 --- a/src/nodeinfo.c +++ b/src/nodeinfo.c @@ -159,7 +159,11 @@ int linuxNodeInfoCPUPopulate(virConnectPtr conn, FILE *cpuinfo, nodeinfo->cpus = 0; nodeinfo->mhz = 0; - nodeinfo->nodes = nodeinfo->cores = 1; + nodeinfo->cores = 1; + if (numa_available() < 0) + nodeinfo->nodes = 1; + else + nodeinfo->nodes = numa_max_node() + 1; /* NB: It is impossible to fill our nodes, since cpuinfo * has no knowledge of NUMA nodes */

Hi Chris, On Thu, 11 Mar 2010 18:00:56 -0500 Chris Lalancette <clalance(a)redhat.com&gt; wrote: > The nodeGetInfo code was always assuming that machine had a > single NUMA node, which is not correct. The good news is that > libnuma gives us this information pretty easily, so let's > properly report it. > > NOTE: With recent hardware starting to support CPU hot-add > and hot-remove, both this code and the nodeCapsInitNUMA() > code are quickly going to become obsolete. We'll have to > think of a more dynamic solution for dealing with NUMA > nodes and CPUs that can come and go at will. > > Signed-off-by: Chris Lalancette <clalance(a)redhat.com&gt; > --- > src/nodeinfo.c | 6 +++++- > 1 files changed, 5 insertions(+), 1 deletions(-) > > diff --git a/src/nodeinfo.c b/src/nodeinfo.c > index 0748602..8d7e055 100644 > --- a/src/nodeinfo.c > +++ b/src/nodeinfo.c > @@ -159,7 +159,11 @@ int linuxNodeInfoCPUPopulate(virConnectPtr conn, FILE *cpuinfo, > > nodeinfo->cpus = 0; > nodeinfo->mhz = 0; > - nodeinfo->nodes = nodeinfo->cores = 1; > + nodeinfo->cores = 1; > + if (numa_available() < 0) > + nodeinfo->nodes = 1; > + else > + nodeinfo->nodes = numa_max_node() + 1; > > /* NB: It is impossible to fill our nodes, since cpuinfo > * has no knowledge of NUMA nodes */ This commit breaks the build for me with the following message: """ cc1: warnings being treated as errors nodeinfo.c: In function 'linuxNodeInfoCPUPopulate': nodeinfo.c:163: error: implicit declaration of function 'numa_available' nodeinfo.c:163: error: nested extern declaration of 'numa_available' nodeinfo.c:166: error: implicit declaration of function 'numa_max_node' nodeinfo.c:166: error: nested extern declaration of 'numa_max_node' make[3]: *** [libvirt_driver_la-nodeinfo.lo] Error 1 make[3]: *** Waiting for unfinished jobs.... ... """ The code in this file that uses the same functions seems to be protected by the HAVE_NUMACTL macro.