virtio_net: do not reschedule rx refill forever
authorMichael S. Tsirkin <mst@redhat.com>
Fri, 2 Jul 2010 16:32:55 +0000 (16:32 +0000)
committerDavid S. Miller <davem@davemloft.net>
Sat, 3 Jul 2010 05:27:25 +0000 (22:27 -0700)
We currently fill all of RX ring, then add_buf
returns ENOSPC, which gets mis-detected as an out of
memory condition and causes us to reschedule the work,
and so on forever. Fix this by oom = err == -ENOMEM;

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: stable@kernel.org # .34.x
Signed-off-by: David S. Miller <davem@davemloft.net>
drivers/net/virtio_net.c

index 1edb7a61983c05660aaec2624e596159ab2cd5c9..ee7571195b10d9137aa126ec1e66ecfba1765fa2 100644 (file)
@@ -415,7 +415,7 @@ static int add_recvbuf_mergeable(struct virtnet_info *vi, gfp_t gfp)
 static bool try_fill_recv(struct virtnet_info *vi, gfp_t gfp)
 {
        int err;
-       bool oom = false;
+       bool oom;
 
        do {
                if (vi->mergeable_rx_bufs)
@@ -425,10 +425,9 @@ static bool try_fill_recv(struct virtnet_info *vi, gfp_t gfp)
                else
                        err = add_recvbuf_small(vi, gfp);
 
-               if (err < 0) {
-                       oom = true;
+               oom = err == -ENOMEM;
+               if (err < 0)
                        break;
-               }
                ++vi->num;
        } while (err > 0);
        if (unlikely(vi->num > vi->max))