net: rps: fix cpu unplug
softnet_data.input_pkt_queue is protected by a spinlock that
we must hold when transferring packets from victim queue to an active
one. This is because other cpus could still be trying to enqueue packets
into victim queue.
A second problem is that when we transfert the NAPI poll_list from
victim to current cpu, we absolutely need to special case the percpu
backlog, because we do not want to add complex locking to protect
process_queue : Only owner cpu is allowed to manipulate it, unless cpu
is offline.
Based on initial patch from Prasad Sodagudi & Subash Abhinov
Kasiviswanathan.
This version is better because we do not slow down packet processing,
only make migration safer.
Reported-by: Prasad Sodagudi <psodagud@codeaurora.org>
Reported-by: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/net/core/dev.c b/net/core/dev.c
index 683d493..171420e 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -7072,10 +7072,20 @@
oldsd->output_queue = NULL;
oldsd->output_queue_tailp = &oldsd->output_queue;
}
- /* Append NAPI poll list from offline CPU. */
- if (!list_empty(&oldsd->poll_list)) {
- list_splice_init(&oldsd->poll_list, &sd->poll_list);
- raise_softirq_irqoff(NET_RX_SOFTIRQ);
+ /* Append NAPI poll list from offline CPU, with one exception :
+ * process_backlog() must be called by cpu owning percpu backlog.
+ * We properly handle process_queue & input_pkt_queue later.
+ */
+ while (!list_empty(&oldsd->poll_list)) {
+ struct napi_struct *napi = list_first_entry(&oldsd->poll_list,
+ struct napi_struct,
+ poll_list);
+
+ list_del_init(&napi->poll_list);
+ if (napi->poll == process_backlog)
+ napi->state = 0;
+ else
+ ____napi_schedule(sd, napi);
}
raise_softirq_irqoff(NET_TX_SOFTIRQ);
@@ -7086,7 +7096,7 @@
netif_rx_internal(skb);
input_queue_head_incr(oldsd);
}
- while ((skb = __skb_dequeue(&oldsd->input_pkt_queue))) {
+ while ((skb = skb_dequeue(&oldsd->input_pkt_queue))) {
netif_rx_internal(skb);
input_queue_head_incr(oldsd);
}