xen-netfront: Rework the fix for Rx stall during OOM and network stress

Orabug: 25747721

The commit 90c311b0eeea ("xen-netfront: Fix Rx stall during network
stress and OOM") caused the refill timer to be triggerred almost on
all invocations of xennet_alloc_rx_buffers for certain workloads.
This reworks the fix by reverting to the old behaviour and taking into
consideration the skb allocation failure. Refill timer is now triggered
on insufficient requests or skb allocation failure.

Signed-off-by: Vineeth Remanan Pillai <vineethp@amazon.com>
Fixes: 90c311b0eeea (xen-netfront: Fix Rx stall during network stress and OOM)
Reported-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Backport from upstream 538d92912d3190a1dd809233a0d57277459f37b2

Signed-off-by: Dongli Zhang <dongli.zhang@oracle.com>
Acked-By: Joe Jin <joe.jin@oracle.com>

diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index 878ecd1330f25065c571ce4b5977ee98886fd2da..e442641e8b872a74e95c0c5233fb260a4f335997 100644 (file)
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -281,6 +281,7 @@ static void xennet_alloc_rx_buffers(struct netfront_queue *queue)
 {
        RING_IDX req_prod = queue->rx.req_prod_pvt;
        int notify;
+       int err = 0;
 
        if (unlikely(!netif_carrier_ok(queue->info->netdev)))
                return;
@@ -295,8 +296,10 @@ static void xennet_alloc_rx_buffers(struct netfront_queue *queue)
                struct xen_netif_rx_request *req;
 
                skb = xennet_alloc_one_rx_buffer(queue);
-               if (!skb)
+               if (!skb) {
+                       err = -ENOMEM;
                        break;
+               }
 
                id = xennet_rxidx(req_prod);
 
@@ -320,8 +323,13 @@ static void xennet_alloc_rx_buffers(struct netfront_queue *queue)
 
        queue->rx.req_prod_pvt = req_prod;
 
-       /* Not enough requests? Try again later. */
-       if (req_prod - queue->rx.sring->req_prod < NET_RX_SLOTS_MIN) {
+       /* Try again later if there are not enough requests or skb allocation
+        * failed.
+        * Enough requests is quantified as the sum of newly created slots and
+        * the unconsumed slots at the backend.
+        */
+       if (req_prod - queue->rx.rsp_cons < NET_RX_SLOTS_MIN ||
+           unlikely(err)) {
                mod_timer(&queue->rx_refill_timer, jiffies + (HZ/10));
                return;
        }