LVS
lvs-devel
Google
 
Web LinuxVirtualServer.org

[PATCHv2 net-next 2/2] ipvs: avoid rcu_barrier during netns cleanup

To: Simon Horman <horms@xxxxxxxxxxxx>
Subject: [PATCHv2 net-next 2/2] ipvs: avoid rcu_barrier during netns cleanup
Cc: lvs-devel@xxxxxxxxxxxxxxx
From: Julian Anastasov <ja@xxxxxx>
Date: Thu, 12 Sep 2013 11:22:06 +0300
commit 578bc3ef1e473a ("ipvs: reorganize dest trash") added
rcu_barrier() on cleanup to wait dest users and schedulers
like LBLC and LBLCR to put their last dest reference.
Using rcu_barrier with many namespaces is problematic.

Trying to fix it by freeing dest with kfree_rcu is not
a solution, RCU callbacks can run in parallel and execution
order is random.

Fix it by creating new function ip_vs_dest_put_and_free()
which is heavier than ip_vs_dest_put(). We will use it just
for schedulers like LBLC, LBLCR that can delay their dest
release.

By default, dests reference is above 0 if they are present in
service and it is 0 when deleted but still in trash list.
Change the dest trash code to use ip_vs_dest_put_and_free(),
so that refcnt -1 can be used for freeing. As result,
such checks remain in slow path and the rcu_barrier() from
netns cleanup can be removed.

Signed-off-by: Julian Anastasov <ja@xxxxxx>
---
 include/net/ip_vs.h              |    6 ++++++
 net/netfilter/ipvs/ip_vs_ctl.c   |    6 +-----
 net/netfilter/ipvs/ip_vs_lblc.c  |    2 +-
 net/netfilter/ipvs/ip_vs_lblcr.c |    2 +-
 4 files changed, 9 insertions(+), 7 deletions(-)

diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h
index c32360d..7a33897 100644
--- a/include/net/ip_vs.h
+++ b/include/net/ip_vs.h
@@ -1446,6 +1446,12 @@ static inline void ip_vs_dest_put(struct ip_vs_dest 
*dest)
        atomic_dec(&dest->refcnt);
 }
 
+static inline void ip_vs_dest_put_and_free(struct ip_vs_dest *dest)
+{
+       if (atomic_dec_return(&dest->refcnt) < 0)
+               kfree(dest);
+}
+
 /*
  *      IPVS sync daemon data and function prototypes
  *      (from ip_vs_sync.c)
diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c
index a3df9bd..62786a4 100644
--- a/net/netfilter/ipvs/ip_vs_ctl.c
+++ b/net/netfilter/ipvs/ip_vs_ctl.c
@@ -704,7 +704,7 @@ static void ip_vs_dest_free(struct ip_vs_dest *dest)
        __ip_vs_dst_cache_reset(dest);
        __ip_vs_svc_put(svc, false);
        free_percpu(dest->stats.cpustats);
-       kfree(dest);
+       ip_vs_dest_put_and_free(dest);
 }
 
 /*
@@ -3820,10 +3820,6 @@ void __net_exit ip_vs_control_net_cleanup(struct net 
*net)
 {
        struct netns_ipvs *ipvs = net_ipvs(net);
 
-       /* Some dest can be in grace period even before cleanup, we have to
-        * defer ip_vs_trash_cleanup until ip_vs_dest_wait_readers is called.
-        */
-       rcu_barrier();
        ip_vs_trash_cleanup(net);
        ip_vs_stop_estimator(net, &ipvs->tot_stats);
        ip_vs_control_net_cleanup_sysctl(net);
diff --git a/net/netfilter/ipvs/ip_vs_lblc.c b/net/netfilter/ipvs/ip_vs_lblc.c
index b39b0e9..a81659d 100644
--- a/net/netfilter/ipvs/ip_vs_lblc.c
+++ b/net/netfilter/ipvs/ip_vs_lblc.c
@@ -136,7 +136,7 @@ static void ip_vs_lblc_rcu_free(struct rcu_head *head)
                                                   struct ip_vs_lblc_entry,
                                                   rcu_head);
 
-       ip_vs_dest_put(en->dest);
+       ip_vs_dest_put_and_free(en->dest);
        kfree(en);
 }
 
diff --git a/net/netfilter/ipvs/ip_vs_lblcr.c b/net/netfilter/ipvs/ip_vs_lblcr.c
index 688f613..b6a024b 100644
--- a/net/netfilter/ipvs/ip_vs_lblcr.c
+++ b/net/netfilter/ipvs/ip_vs_lblcr.c
@@ -130,7 +130,7 @@ static void ip_vs_lblcr_elem_rcu_free(struct rcu_head *head)
        struct ip_vs_dest_set_elem *e;
 
        e = container_of(head, struct ip_vs_dest_set_elem, rcu_head);
-       ip_vs_dest_put(e->dest);
+       ip_vs_dest_put_and_free(e->dest);
        kfree(e);
 }
 
-- 
1.7.3.4

--
To unsubscribe from this list: send the line "unsubscribe lvs-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html

<Prev in Thread] Current Thread [Next in Thread>