numa_complete_configuration() in hw/core/numa.c always adds a NUMA node
for the pSeries machine if none was specified, but without node distance
information for the single node created. This added node is also not
accounted for in numa_state->num_nodes, which returns zero.
NUMA FORM1 affinity code didn't rely on numa_state information to do its
job, but FORM2 does. As is now, this is the result of a pSeries guest
with NUMA FORM2 affinity when no NUMA nodes is specified:
$ numactl -H available: 1 nodes (0) node 0 cpus: 0 node 0 size: 16222 MB
node 0 free: 15681 MB No distance information available.
This can be amended in spapr_numa_FORM2_write_rtas_tables(). We're
always expecting at least one NUMA node, and we're going to enforce that
the local distance (the distance to the node to itself) is always 10.
This allows for the proper creation of the NUMA distance tables, fixing
the output of 'numactl -H' in the guest:
$ numactl -H available: 1 nodes (0) node 0 cpus: 0 node 0 size: 16222 MB
node 0 free: 15685 MB node distances: node 0
0: 10
CC: Igor Mammedov <imamm...@redhat.com>
Signed-off-by: Daniel Henrique Barboza <danielhb...@gmail.com>
---
Igor,
CCing you as a FYI. I wasn't sure whether there is a reason for
numa_complete_configuration() not adding distance info an update 'num_nodes'
for the auto-generated NUMA node, I decided to handle this case in
pseries side instead.
hw/ppc/spapr_numa.c | 13 ++++++++++++-
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/hw/ppc/spapr_numa.c b/hw/ppc/spapr_numa.c
index 659513b405..d8caf5f6bd 100644
--- a/hw/ppc/spapr_numa.c
+++ b/hw/ppc/spapr_numa.c
@@ -499,7 +499,7 @@ static void
spapr_numa_FORM2_write_rtas_tables(SpaprMachineState *spapr,
{
MachineState *ms = MACHINE(spapr);
NodeInfo *numa_info = ms->numa_state->nodes;
- int nb_numa_nodes = ms->numa_state->num_nodes;
+ int nb_numa_nodes = ms->numa_state->num_nodes ?: 1;
int distance_table_entries = nb_numa_nodes * nb_numa_nodes;
g_autofree uint32_t *lookup_index_table = NULL;
g_autofree uint32_t *distance_table = NULL;
@@ -539,6 +539,17 @@ static void
spapr_numa_FORM2_write_rtas_tables(SpaprMachineState *spapr,
for (src = 0; src < nb_numa_nodes; src++) {
for (dst = 0; dst < nb_numa_nodes; dst++) {
+ /*
+ * We need to be explicit with the local distance
+ * value to cover the case where the user didn't added any
+ * NUMA nodes, but QEMU adds the default NUMA node without
+ * adding the numa_info to retrieve the info from.
+ */
+ if (src == dst) {
+ node_distances[i++] = 10;
+ continue;
+ }
+
node_distances[i++] = numa_info[src].distance[dst];
}
}
--
2.31.1
