Thread (16 messages) 16 messages, 5 authors, 2016-09-10

Re: [PATCH] net_namespace: fixed net_device reference leak

flat view

From: Eric Dumazet <hidden>
Date: 2016-09-08 23:37:43

On Thu, 2016-09-08 at 16:16 -0700, Jojy Varghese wrote:
quoted hunk ↗ jump to hunk
The dst's do disappear but the net device does not. A better solution
is to only hold reference to dst's net device ONLY if it is not a
loopback device. In the case of the loopback device, we want to do
only a "put" on it and skip the "hold".

Updating the patch below:

---
 net/core/dst.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/net/core/dst.c b/net/core/dst.c
index 7e45593..f63027e 100644
--- a/net/core/dst.c
+++ b/net/core/dst.c
@@ -433,12 +433,11 @@ static void dst_ifdown(struct dst_entry *dst,
struct net_device *dev,
                dst->input = dst_discard;
                dst->output = dst_discard_out;
        } else {
-               if (dst->dev == dev_net(dst->dev)->loopback_dev)
On which tree are you working exactly ?
Current code is not looking like that.
quoted hunk ↗ jump to hunk
-                       dst->dev = init_net.loopback_dev;
-               else
+               if (dst->dev != dev_net(dst->dev)->loopback_dev) {
                        dst->dev = dev_net(dst->dev)->loopback_dev;
+                       dev_hold(dst->dev);
+               }

-               dev_hold(dst->dev);
                dev_put(dev);
        }
 }

I appreciate your desperate efforts, but I am telling you the bug is
elsewhere.

Do you have a reproducer of the bug, on latest David Miller tree
( https://git.kernel.org/cgit/linux/kernel/git/davem/net.git )

Thanks.
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help