Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753321Ab3JUIuc (ORCPT ); Mon, 21 Oct 2013 04:50:32 -0400 Received: from shadbolt.e.decadent.org.uk ([88.96.1.126]:39318 "EHLO shadbolt.e.decadent.org.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753053Ab3JUIua (ORCPT ); Mon, 21 Oct 2013 04:50:30 -0400 Content-Type: text/plain; charset="UTF-8" Content-Disposition: inline Content-Transfer-Encoding: 8bit MIME-Version: 1.0 From: Ben Hutchings To: linux-kernel@vger.kernel.org, stable@vger.kernel.org CC: akpm@linux-foundation.org, "Steffen Klassert" , "David S. Miller" Date: Mon, 21 Oct 2013 09:46:28 +0100 Message-ID: X-Mailer: LinuxStableQueue (scripts by bwh) Subject: [PATCH 3.2 015/149] inetpeer: Invalidate the inetpeer tree along with the routing cache In-Reply-To: X-SA-Exim-Connect-IP: 212.20.242.100 X-SA-Exim-Mail-From: ben@decadent.org.uk X-SA-Exim-Scanned: No (on shadbolt.decadent.org.uk); SAEximRunCond expanded to false Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 5504 Lines: 186 3.2.52-rc1 review patch. If anyone has any objections, please let me know. ------------------ From: Steffen Klassert [ Upstream commit 5faa5df1fa2024bd750089ff21dcc4191798263d ] We initialize the routing metrics with the values cached on the inetpeer in rt_init_metrics(). So if we have the metrics cached on the inetpeer, we ignore the user configured fib_metrics. To fix this issue, we replace the old tree with a fresh initialized inet_peer_base. The old tree is removed later with a delayed work queue. Signed-off-by: Steffen Klassert Signed-off-by: David S. Miller Signed-off-by: Ben Hutchings --- include/net/inetpeer.h | 3 ++ net/ipv4/inetpeer.c | 80 +++++++++++++++++++++++++++++++++++++++++++++++++- net/ipv4/route.c | 1 + 3 files changed, 83 insertions(+), 1 deletion(-) diff --git a/include/net/inetpeer.h b/include/net/inetpeer.h index e9ff3fc..ac8a4ea 100644 --- a/include/net/inetpeer.h +++ b/include/net/inetpeer.h @@ -41,6 +41,7 @@ struct inet_peer { u32 pmtu_orig; u32 pmtu_learned; struct inetpeer_addr_base redirect_learned; + struct list_head gc_list; /* * Once inet_peer is queued for deletion (refcnt == -1), following fields * are not available: rid, ip_id_count, tcp_ts, tcp_ts_stamp @@ -96,6 +97,8 @@ static inline struct inet_peer *inet_getpeer_v6(const struct in6_addr *v6daddr, extern void inet_putpeer(struct inet_peer *p); extern bool inet_peer_xrlim_allow(struct inet_peer *peer, int timeout); +extern void inetpeer_invalidate_tree(int family); + /* * temporary check to make sure we dont access rid, ip_id_count, tcp_ts, * tcp_ts_stamp if no refcount is taken on inet_peer diff --git a/net/ipv4/inetpeer.c b/net/ipv4/inetpeer.c index 86f13c67..d77a7dd 100644 --- a/net/ipv4/inetpeer.c +++ b/net/ipv4/inetpeer.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -66,6 +67,11 @@ static struct kmem_cache *peer_cachep __read_mostly; +static LIST_HEAD(gc_list); +static const int gc_delay = 60 * HZ; +static struct delayed_work gc_work; +static DEFINE_SPINLOCK(gc_lock); + #define node_height(x) x->avl_height #define peer_avl_empty ((struct inet_peer *)&peer_fake_node) @@ -102,6 +108,50 @@ int inet_peer_threshold __read_mostly = 65536 + 128; /* start to throw entries m int inet_peer_minttl __read_mostly = 120 * HZ; /* TTL under high load: 120 sec */ int inet_peer_maxttl __read_mostly = 10 * 60 * HZ; /* usual time to live: 10 min */ +static void inetpeer_gc_worker(struct work_struct *work) +{ + struct inet_peer *p, *n; + LIST_HEAD(list); + + spin_lock_bh(&gc_lock); + list_replace_init(&gc_list, &list); + spin_unlock_bh(&gc_lock); + + if (list_empty(&list)) + return; + + list_for_each_entry_safe(p, n, &list, gc_list) { + + if(need_resched()) + cond_resched(); + + if (p->avl_left != peer_avl_empty) { + list_add_tail(&p->avl_left->gc_list, &list); + p->avl_left = peer_avl_empty; + } + + if (p->avl_right != peer_avl_empty) { + list_add_tail(&p->avl_right->gc_list, &list); + p->avl_right = peer_avl_empty; + } + + n = list_entry(p->gc_list.next, struct inet_peer, gc_list); + + if (!atomic_read(&p->refcnt)) { + list_del(&p->gc_list); + kmem_cache_free(peer_cachep, p); + } + } + + if (list_empty(&list)) + return; + + spin_lock_bh(&gc_lock); + list_splice(&list, &gc_list); + spin_unlock_bh(&gc_lock); + + schedule_delayed_work(&gc_work, gc_delay); +} /* Called from ip_output.c:ip_init */ void __init inet_initpeers(void) @@ -126,6 +176,7 @@ void __init inet_initpeers(void) 0, SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL); + INIT_DELAYED_WORK_DEFERRABLE(&gc_work, inetpeer_gc_worker); } static int addr_compare(const struct inetpeer_addr *a, @@ -448,7 +499,7 @@ relookup: p->pmtu_expires = 0; p->pmtu_orig = 0; memset(&p->redirect_learned, 0, sizeof(p->redirect_learned)); - + INIT_LIST_HEAD(&p->gc_list); /* Link the node. */ link_to_pool(p, base); @@ -508,3 +559,30 @@ bool inet_peer_xrlim_allow(struct inet_peer *peer, int timeout) return rc; } EXPORT_SYMBOL(inet_peer_xrlim_allow); + +void inetpeer_invalidate_tree(int family) +{ + struct inet_peer *old, *new, *prev; + struct inet_peer_base *base = family_to_base(family); + + write_seqlock_bh(&base->lock); + + old = base->root; + if (old == peer_avl_empty_rcu) + goto out; + + new = peer_avl_empty_rcu; + + prev = cmpxchg(&base->root, old, new); + if (prev == old) { + base->total = 0; + spin_lock(&gc_lock); + list_add_tail(&prev->gc_list, &gc_list); + spin_unlock(&gc_lock); + schedule_delayed_work(&gc_work, gc_delay); + } + +out: + write_sequnlock_bh(&base->lock); +} +EXPORT_SYMBOL(inetpeer_invalidate_tree); diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 94cdbc5..c45a155a3 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -939,6 +939,7 @@ static void rt_cache_invalidate(struct net *net) get_random_bytes(&shuffle, sizeof(shuffle)); atomic_add(shuffle + 1U, &net->ipv4.rt_genid); redirect_genid++; + inetpeer_invalidate_tree(AF_INET); } /* -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/