rps: NUMA flow limit allocations

Given we allocate memory for each cpu, we can do this
using NUMA affinities, instead of using NUMA policies
of the process changing flow_limit_cpu_bitmap value.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 2 insertions, 1 deletions

diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index cca444190907..cf9cd13509a7 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -122,7 +122,8 @@ static int flow_limit_cpu_sysctl(struct ctl_table *table, int write,
                                 synchronize_rcu();
                                 kfree(cur);
                         } else if (!cur && cpumask_test_cpu(i, mask)) {
-                                cur = kzalloc(len, GFP_KERNEL);
+                                cur = kzalloc_node(len, GFP_KERNEL,
+                                                   cpu_to_node(i));
                                 if (!cur) {
                                         /* not unwinding previous changes */
                                         ret = -ENOMEM;