This should help finding unbalanced refcount put leading to negative
refcounts, unfreeable dst_entry-es and hanging network namespace
cleanup.

Previously we had messages like this in dmesg:

  dst_release: dst:ffff9d9686fef000 refcnt:-1

But without stack it's hard to guess the root cause.

https://virtuozzo.atlassian.net/browse/PSBM-125709
Signed-off-by: Pavel Tikhomirov <ptikhomi...@virtuozzo.com>
---
 net/core/dst.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/core/dst.c b/net/core/dst.c
index e3ddd2d50235..94891b43bddc 100644
--- a/net/core/dst.c
+++ b/net/core/dst.c
@@ -305,9 +305,9 @@ void dst_release(struct dst_entry *dst)
                unsigned short nocache = dst->flags & DST_NOCACHE;
 
                newrefcnt = atomic_dec_return(&dst->__refcnt);
-               if (unlikely(newrefcnt < 0))
-                       net_warn_ratelimited("%s: dst:%p refcnt:%d\n",
-                                            __func__, dst, newrefcnt);
+               WARN_RATELIMIT(unlikely(newrefcnt < 0),
+                              "%s: Refcount leak dst:%p refcnt:%d (please 
report to PSBM-125709)\n",
+                              __func__, dst, newrefcnt);
                if (!newrefcnt && unlikely(nocache))
                        call_rcu(&dst->rcu_head, dst_destroy_rcu);
        }
-- 
2.50.0

_______________________________________________
Devel mailing list
Devel@openvz.org
https://lists.openvz.org/mailman/listinfo/devel

Reply via email to