[PATCH AUTOSEL 5.14 09/40] netfilter: ipset: Fix oversized kvmalloc() calls

[PATCH AUTOSEL 5.14 09/40] netfilter: ipset: Fix oversized kvmalloc() calls

[Sasha Levin]

From: Jozsef Kadlecsik <kadlec@netfilter.org>

[ Upstream commit 7bbc3d385bd813077acaf0e6fdb2a86a901f5382 ]

The commit

commit 7661809d493b426e979f39ab512e3adf41fbcc69
Author: Linus Torvalds <torvalds@linux-foundation.org>
Date:   Wed Jul 14 09:45:49 2021 -0700

    mm: don't allow oversized kvmalloc() calls

limits the max allocatable memory via kvmalloc() to MAX_INT. Apply the
same limit in ipset.

Reported-by: syzbot+3493b1873fb3ea827986@syzkaller.appspotmail.com
Reported-by: syzbot+2b8443c35458a617c904@syzkaller.appspotmail.com
Reported-by: syzbot+ee5cb15f4a0e85e0d54e@syzkaller.appspotmail.com
Signed-off-by: Jozsef Kadlecsik <kadlec@netfilter.org>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/netfilter/ipset/ip_set_hash_gen.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/netfilter/ipset/ip_set_hash_gen.h b/net/netfilter/ipset/ip_set_hash_gen.h
index 6186358eac7c..6e391308431d 100644
--- a/net/netfilter/ipset/ip_set_hash_gen.h
+++ b/net/netfilter/ipset/ip_set_hash_gen.h
@@ -130,11 +130,11 @@ htable_size(u8 hbits)
 {
 	size_t hsize;
 
-	/* We must fit both into u32 in jhash and size_t */
+	/* We must fit both into u32 in jhash and INT_MAX in kvmalloc_node() */
 	if (hbits > 31)
 		return 0;
 	hsize = jhash_size(hbits);
-	if ((((size_t)-1) - sizeof(struct htable)) / sizeof(struct hbucket *)
+	if ((INT_MAX - sizeof(struct htable)) / sizeof(struct hbucket *)
 	    < hsize)
 		return 0;
 
-- 
2.33.0

[PATCH AUTOSEL 5.14 12/40] netfilter: ip6_tables: zero-initialize fragment offset

[Sasha Levin]

From: Jeremy Sowden <jeremy@azazel.net>

[ Upstream commit 310e2d43c3ad429c1fba4b175806cf1f55ed73a6 ]

ip6tables only sets the `IP6T_F_PROTO` flag on a rule if a protocol is
specified (`-p tcp`, for example).  However, if the flag is not set,
`ip6_packet_match` doesn't call `ipv6_find_hdr` for the skb, in which
case the fragment offset is left uninitialized and a garbage value is
passed to each matcher.

Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
Reviewed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/ipv6/netfilter/ip6_tables.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index de2cf3943b91..a579ea14a69b 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -273,6 +273,7 @@ ip6t_do_table(struct sk_buff *skb,
 	 * things we don't know, ie. tcp syn flag or ports).  If the
 	 * rule is also a fragment-specific rule, non-fragments won't
 	 * match it. */
+	acpar.fragoff = 0;
 	acpar.hotdrop = false;
 	acpar.state   = state;
 
-- 
2.33.0

[PATCH AUTOSEL 5.14 17/40] netfilter: nf_tables: Fix oversized kvmalloc() calls

[Sasha Levin]

From: Pablo Neira Ayuso <pablo@netfilter.org>

[ Upstream commit 45928afe94a094bcda9af858b96673d59bc4a0e9 ]

The commit 7661809d493b ("mm: don't allow oversized kvmalloc() calls")
limits the max allocatable memory via kvmalloc() to MAX_INT.

Reported-by: syzbot+cd43695a64bcd21b8596@syzkaller.appspotmail.com
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/netfilter/nf_tables_api.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c
index 081437dd75b7..4b6255c4b183 100644
--- a/net/netfilter/nf_tables_api.c
+++ b/net/netfilter/nf_tables_api.c
@@ -4336,7 +4336,7 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info,
 	if (ops->privsize != NULL)
 		size = ops->privsize(nla, &desc);
 	alloc_size = sizeof(*set) + size + udlen;
-	if (alloc_size < size)
+	if (alloc_size < size || alloc_size > INT_MAX)
 		return -ENOMEM;
 	set = kvzalloc(alloc_size, GFP_KERNEL);
 	if (!set)
-- 
2.33.0

[PATCH AUTOSEL 5.14 18/40] netfilter: nf_nat_masquerade: make async masq_inet6_event handling generic

[Sasha Levin]

From: Florian Westphal <fw@strlen.de>

[ Upstream commit 30db406923b9285a9bac06a6af5e74bd6d0f1d06 ]

masq_inet6_event is called asynchronously from system work queue,
because the inet6 notifier is atomic and nf_iterate_cleanup can sleep.

The ipv4 and device notifiers call nf_iterate_cleanup directly.

This is legal, but these notifiers are called with RTNL mutex held.
A large conntrack table with many devices coming and going will have severe
impact on the system usability, with 'ip a' blocking for several seconds.

This change places the defer code into a helper and makes it more
generic so ipv4 and ifdown notifiers can be converted to defer the
cleanup walk as well in a follow patch.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/netfilter/nf_nat_masquerade.c | 122 ++++++++++++++++++------------
 1 file changed, 75 insertions(+), 47 deletions(-)

diff --git a/net/netfilter/nf_nat_masquerade.c b/net/netfilter/nf_nat_masquerade.c
index 8e8a65d46345..415919a6ac1a 100644
--- a/net/netfilter/nf_nat_masquerade.c
+++ b/net/netfilter/nf_nat_masquerade.c
@@ -9,8 +9,19 @@
 
 #include <net/netfilter/nf_nat_masquerade.h>
 
+struct masq_dev_work {
+	struct work_struct work;
+	struct net *net;
+	union nf_inet_addr addr;
+	int ifindex;
+	int (*iter)(struct nf_conn *i, void *data);
+};
+
+#define MAX_MASQ_WORKER_COUNT	16
+
 static DEFINE_MUTEX(masq_mutex);
 static unsigned int masq_refcnt __read_mostly;
+static atomic_t masq_worker_count __read_mostly;
 
 unsigned int
 nf_nat_masquerade_ipv4(struct sk_buff *skb, unsigned int hooknum,
@@ -63,6 +74,63 @@ nf_nat_masquerade_ipv4(struct sk_buff *skb, unsigned int hooknum,
 }
 EXPORT_SYMBOL_GPL(nf_nat_masquerade_ipv4);
 
+static void iterate_cleanup_work(struct work_struct *work)
+{
+	struct masq_dev_work *w;
+
+	w = container_of(work, struct masq_dev_work, work);
+
+	nf_ct_iterate_cleanup_net(w->net, w->iter, (void *)w, 0, 0);
+
+	put_net(w->net);
+	kfree(w);
+	atomic_dec(&masq_worker_count);
+	module_put(THIS_MODULE);
+}
+
+/* Iterate conntrack table in the background and remove conntrack entries
+ * that use the device/address being removed.
+ *
+ * In case too many work items have been queued already or memory allocation
+ * fails iteration is skipped, conntrack entries will time out eventually.
+ */
+static void nf_nat_masq_schedule(struct net *net, union nf_inet_addr *addr,
+				 int ifindex,
+				 int (*iter)(struct nf_conn *i, void *data),
+				 gfp_t gfp_flags)
+{
+	struct masq_dev_work *w;
+
+	if (atomic_read(&masq_worker_count) > MAX_MASQ_WORKER_COUNT)
+		return;
+
+	net = maybe_get_net(net);
+	if (!net)
+		return;
+
+	if (!try_module_get(THIS_MODULE))
+		goto err_module;
+
+	w = kzalloc(sizeof(*w), gfp_flags);
+	if (w) {
+		/* We can overshoot MAX_MASQ_WORKER_COUNT, no big deal */
+		atomic_inc(&masq_worker_count);
+
+		INIT_WORK(&w->work, iterate_cleanup_work);
+		w->ifindex = ifindex;
+		w->net = net;
+		w->iter = iter;
+		if (addr)
+			w->addr = *addr;
+		schedule_work(&w->work);
+		return;
+	}
+
+	module_put(THIS_MODULE);
+ err_module:
+	put_net(net);
+}
+
 static int device_cmp(struct nf_conn *i, void *ifindex)
 {
 	const struct nf_conn_nat *nat = nfct_nat(i);
@@ -136,8 +204,6 @@ static struct notifier_block masq_inet_notifier = {
 };
 
 #if IS_ENABLED(CONFIG_IPV6)
-static atomic_t v6_worker_count __read_mostly;
-
 static int
 nat_ipv6_dev_get_saddr(struct net *net, const struct net_device *dev,
 		       const struct in6_addr *daddr, unsigned int srcprefs,
@@ -187,13 +253,6 @@ nf_nat_masquerade_ipv6(struct sk_buff *skb, const struct nf_nat_range2 *range,
 }
 EXPORT_SYMBOL_GPL(nf_nat_masquerade_ipv6);
 
-struct masq_dev_work {
-	struct work_struct work;
-	struct net *net;
-	struct in6_addr addr;
-	int ifindex;
-};
-
 static int inet6_cmp(struct nf_conn *ct, void *work)
 {
 	struct masq_dev_work *w = (struct masq_dev_work *)work;
@@ -204,21 +263,7 @@ static int inet6_cmp(struct nf_conn *ct, void *work)
 
 	tuple = &ct->tuplehash[IP_CT_DIR_REPLY].tuple;
 
-	return ipv6_addr_equal(&w->addr, &tuple->dst.u3.in6);
-}
-
-static void iterate_cleanup_work(struct work_struct *work)
-{
-	struct masq_dev_work *w;
-
-	w = container_of(work, struct masq_dev_work, work);
-
-	nf_ct_iterate_cleanup_net(w->net, inet6_cmp, (void *)w, 0, 0);
-
-	put_net(w->net);
-	kfree(w);
-	atomic_dec(&v6_worker_count);
-	module_put(THIS_MODULE);
+	return nf_inet_addr_cmp(&w->addr, &tuple->dst.u3);
 }
 
 /* atomic notifier; can't call nf_ct_iterate_cleanup_net (it can sleep).
@@ -233,36 +278,19 @@ static int masq_inet6_event(struct notifier_block *this,
 {
 	struct inet6_ifaddr *ifa = ptr;
 	const struct net_device *dev;
-	struct masq_dev_work *w;
-	struct net *net;
+	union nf_inet_addr addr;
 
-	if (event != NETDEV_DOWN || atomic_read(&v6_worker_count) >= 16)
+	if (event != NETDEV_DOWN)
 		return NOTIFY_DONE;
 
 	dev = ifa->idev->dev;
-	net = maybe_get_net(dev_net(dev));
-	if (!net)
-		return NOTIFY_DONE;
 
-	if (!try_module_get(THIS_MODULE))
-		goto err_module;
+	memset(&addr, 0, sizeof(addr));
 
-	w = kmalloc(sizeof(*w), GFP_ATOMIC);
-	if (w) {
-		atomic_inc(&v6_worker_count);
+	addr.in6 = ifa->addr;
 
-		INIT_WORK(&w->work, iterate_cleanup_work);
-		w->ifindex = dev->ifindex;
-		w->net = net;
-		w->addr = ifa->addr;
-		schedule_work(&w->work);
-
-		return NOTIFY_DONE;
-	}
-
-	module_put(THIS_MODULE);
- err_module:
-	put_net(net);
+	nf_nat_masq_schedule(dev_net(dev), &addr, dev->ifindex, inet6_cmp,
+			     GFP_ATOMIC);
 	return NOTIFY_DONE;
 }
 
-- 
2.33.0

[PATCH AUTOSEL 5.14 19/40] netfilter: nf_nat_masquerade: defer conntrack walk to work queue

[Sasha Levin]

From: Florian Westphal <fw@strlen.de>

[ Upstream commit 7970a19b71044bf4dc2c1becc200275bdf1884d4 ]

The ipv4 and device notifiers are called with RTNL mutex held.
The table walk can take some time, better not block other RTNL users.

'ip a' has been reported to block for up to 20 seconds when conntrack table
has many entries and device down events are frequent (e.g., PPP).

Reported-and-tested-by: Martin Zaharinov <micron10@gmail.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/netfilter/nf_nat_masquerade.c | 50 +++++++++++++++----------------
 1 file changed, 24 insertions(+), 26 deletions(-)

diff --git a/net/netfilter/nf_nat_masquerade.c b/net/netfilter/nf_nat_masquerade.c
index 415919a6ac1a..acd73f717a08 100644
--- a/net/netfilter/nf_nat_masquerade.c
+++ b/net/netfilter/nf_nat_masquerade.c
@@ -131,13 +131,14 @@ static void nf_nat_masq_schedule(struct net *net, union nf_inet_addr *addr,
 	put_net(net);
 }
 
-static int device_cmp(struct nf_conn *i, void *ifindex)
+static int device_cmp(struct nf_conn *i, void *arg)
 {
 	const struct nf_conn_nat *nat = nfct_nat(i);
+	const struct masq_dev_work *w = arg;
 
 	if (!nat)
 		return 0;
-	return nat->masq_index == (int)(long)ifindex;
+	return nat->masq_index == w->ifindex;
 }
 
 static int masq_device_event(struct notifier_block *this,
@@ -153,8 +154,8 @@ static int masq_device_event(struct notifier_block *this,
 		 * and forget them.
 		 */
 
-		nf_ct_iterate_cleanup_net(net, device_cmp,
-					  (void *)(long)dev->ifindex, 0, 0);
+		nf_nat_masq_schedule(net, NULL, dev->ifindex,
+				     device_cmp, GFP_KERNEL);
 	}
 
 	return NOTIFY_DONE;
@@ -162,35 +163,45 @@ static int masq_device_event(struct notifier_block *this,
 
 static int inet_cmp(struct nf_conn *ct, void *ptr)
 {
-	struct in_ifaddr *ifa = (struct in_ifaddr *)ptr;
-	struct net_device *dev = ifa->ifa_dev->dev;
 	struct nf_conntrack_tuple *tuple;
+	struct masq_dev_work *w = ptr;
 
-	if (!device_cmp(ct, (void *)(long)dev->ifindex))
+	if (!device_cmp(ct, ptr))
 		return 0;
 
 	tuple = &ct->tuplehash[IP_CT_DIR_REPLY].tuple;
 
-	return ifa->ifa_address == tuple->dst.u3.ip;
+	return nf_inet_addr_cmp(&w->addr, &tuple->dst.u3);
 }
 
 static int masq_inet_event(struct notifier_block *this,
 			   unsigned long event,
 			   void *ptr)
 {
-	struct in_device *idev = ((struct in_ifaddr *)ptr)->ifa_dev;
-	struct net *net = dev_net(idev->dev);
+	const struct in_ifaddr *ifa = ptr;
+	const struct in_device *idev;
+	const struct net_device *dev;
+	union nf_inet_addr addr;
+
+	if (event != NETDEV_DOWN)
+		return NOTIFY_DONE;
 
 	/* The masq_dev_notifier will catch the case of the device going
 	 * down.  So if the inetdev is dead and being destroyed we have
 	 * no work to do.  Otherwise this is an individual address removal
 	 * and we have to perform the flush.
 	 */
+	idev = ifa->ifa_dev;
 	if (idev->dead)
 		return NOTIFY_DONE;
 
-	if (event == NETDEV_DOWN)
-		nf_ct_iterate_cleanup_net(net, inet_cmp, ptr, 0, 0);
+	memset(&addr, 0, sizeof(addr));
+
+	addr.ip = ifa->ifa_address;
+
+	dev = idev->dev;
+	nf_nat_masq_schedule(dev_net(idev->dev), &addr, dev->ifindex,
+			     inet_cmp, GFP_KERNEL);
 
 	return NOTIFY_DONE;
 }
@@ -253,19 +264,6 @@ nf_nat_masquerade_ipv6(struct sk_buff *skb, const struct nf_nat_range2 *range,
 }
 EXPORT_SYMBOL_GPL(nf_nat_masquerade_ipv6);
 
-static int inet6_cmp(struct nf_conn *ct, void *work)
-{
-	struct masq_dev_work *w = (struct masq_dev_work *)work;
-	struct nf_conntrack_tuple *tuple;
-
-	if (!device_cmp(ct, (void *)(long)w->ifindex))
-		return 0;
-
-	tuple = &ct->tuplehash[IP_CT_DIR_REPLY].tuple;
-
-	return nf_inet_addr_cmp(&w->addr, &tuple->dst.u3);
-}
-
 /* atomic notifier; can't call nf_ct_iterate_cleanup_net (it can sleep).
  *
  * Defer it to the system workqueue.
@@ -289,7 +287,7 @@ static int masq_inet6_event(struct notifier_block *this,
 
 	addr.in6 = ifa->addr;
 
-	nf_nat_masq_schedule(dev_net(dev), &addr, dev->ifindex, inet6_cmp,
+	nf_nat_masq_schedule(dev_net(dev), &addr, dev->ifindex, inet_cmp,
 			     GFP_ATOMIC);
 	return NOTIFY_DONE;
 }
-- 
2.33.0

[PATCH AUTOSEL 5.14 20/40] netfilter: conntrack: serialize hash resizes and cleanups

[Sasha Levin]

From: Eric Dumazet <edumazet@google.com>

[ Upstream commit e9edc188fc76499b0b9bd60364084037f6d03773 ]

Syzbot was able to trigger the following warning [1]

No repro found by syzbot yet but I was able to trigger similar issue
by having 2 scripts running in parallel, changing conntrack hash sizes,
and:

for j in `seq 1 1000` ; do unshare -n /bin/true >/dev/null ; done

It would take more than 5 minutes for net_namespace structures
to be cleaned up.

This is because nf_ct_iterate_cleanup() has to restart everytime
a resize happened.

By adding a mutex, we can serialize hash resizes and cleanups
and also make get_next_corpse() faster by skipping over empty
buckets.

Even without resizes in the picture, this patch considerably
speeds up network namespace dismantles.

[1]
INFO: task syz-executor.0:8312 can't die for more than 144 seconds.
task:syz-executor.0  state:R  running task     stack:25672 pid: 8312 ppid:  6573 flags:0x00004006
Call Trace:
 context_switch kernel/sched/core.c:4955 [inline]
 __schedule+0x940/0x26f0 kernel/sched/core.c:6236
 preempt_schedule_common+0x45/0xc0 kernel/sched/core.c:6408
 preempt_schedule_thunk+0x16/0x18 arch/x86/entry/thunk_64.S:35
 __local_bh_enable_ip+0x109/0x120 kernel/softirq.c:390
 local_bh_enable include/linux/bottom_half.h:32 [inline]
 get_next_corpse net/netfilter/nf_conntrack_core.c:2252 [inline]
 nf_ct_iterate_cleanup+0x15a/0x450 net/netfilter/nf_conntrack_core.c:2275
 nf_conntrack_cleanup_net_list+0x14c/0x4f0 net/netfilter/nf_conntrack_core.c:2469
 ops_exit_list+0x10d/0x160 net/core/net_namespace.c:171
 setup_net+0x639/0xa30 net/core/net_namespace.c:349
 copy_net_ns+0x319/0x760 net/core/net_namespace.c:470
 create_new_namespaces+0x3f6/0xb20 kernel/nsproxy.c:110
 unshare_nsproxy_namespaces+0xc1/0x1f0 kernel/nsproxy.c:226
 ksys_unshare+0x445/0x920 kernel/fork.c:3128
 __do_sys_unshare kernel/fork.c:3202 [inline]
 __se_sys_unshare kernel/fork.c:3200 [inline]
 __x64_sys_unshare+0x2d/0x40 kernel/fork.c:3200
 do_syscall_x64 arch/x86/entry/common.c:50 [inline]
 do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
 entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x7f63da68e739
RSP: 002b:00007f63d7c05188 EFLAGS: 00000246 ORIG_RAX: 0000000000000110
RAX: ffffffffffffffda RBX: 00007f63da792f80 RCX: 00007f63da68e739
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000040000000
RBP: 00007f63da6e8cc4 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00007f63da792f80
R13: 00007fff50b75d3f R14: 00007f63d7c05300 R15: 0000000000022000

Showing all locks held in the system:
1 lock held by khungtaskd/27:
 #0: ffffffff8b980020 (rcu_read_lock){....}-{1:2}, at: debug_show_all_locks+0x53/0x260 kernel/locking/lockdep.c:6446
2 locks held by kworker/u4:2/153:
 #0: ffff888010c69138 ((wq_completion)events_unbound){+.+.}-{0:0}, at: arch_atomic64_set arch/x86/include/asm/atomic64_64.h:34 [inline]
 #0: ffff888010c69138 ((wq_completion)events_unbound){+.+.}-{0:0}, at: arch_atomic_long_set include/linux/atomic/atomic-long.h:41 [inline]
 #0: ffff888010c69138 ((wq_completion)events_unbound){+.+.}-{0:0}, at: atomic_long_set include/linux/atomic/atomic-instrumented.h:1198 [inline]
 #0: ffff888010c69138 ((wq_completion)events_unbound){+.+.}-{0:0}, at: set_work_data kernel/workqueue.c:634 [inline]
 #0: ffff888010c69138 ((wq_completion)events_unbound){+.+.}-{0:0}, at: set_work_pool_and_clear_pending kernel/workqueue.c:661 [inline]
 #0: ffff888010c69138 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work+0x896/0x1690 kernel/workqueue.c:2268
 #1: ffffc9000140fdb0 ((kfence_timer).work){+.+.}-{0:0}, at: process_one_work+0x8ca/0x1690 kernel/workqueue.c:2272
1 lock held by systemd-udevd/2970:
1 lock held by in:imklog/6258:
 #0: ffff88807f970ff0 (&f->f_pos_lock){+.+.}-{3:3}, at: __fdget_pos+0xe9/0x100 fs/file.c:990
3 locks held by kworker/1:6/8158:
1 lock held by syz-executor.0/8312:
2 locks held by kworker/u4:13/9320:
1 lock held by syz-executor.5/10178:
1 lock held by syz-executor.4/10217:

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot <syzkaller@googlegroups.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/netfilter/nf_conntrack_core.c | 70 ++++++++++++++++---------------
 1 file changed, 37 insertions(+), 33 deletions(-)

diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index d31dbccbe7bd..4f074d7653b8 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -75,6 +75,9 @@ static __read_mostly struct kmem_cache *nf_conntrack_cachep;
 static DEFINE_SPINLOCK(nf_conntrack_locks_all_lock);
 static __read_mostly bool nf_conntrack_locks_all;
 
+/* serialize hash resizes and nf_ct_iterate_cleanup */
+static DEFINE_MUTEX(nf_conntrack_mutex);
+
 #define GC_SCAN_INTERVAL	(120u * HZ)
 #define GC_SCAN_MAX_DURATION	msecs_to_jiffies(10)
 
@@ -2192,28 +2195,31 @@ get_next_corpse(int (*iter)(struct nf_conn *i, void *data),
 	spinlock_t *lockp;
 
 	for (; *bucket < nf_conntrack_htable_size; (*bucket)++) {
+		struct hlist_nulls_head *hslot = &nf_conntrack_hash[*bucket];
+
+		if (hlist_nulls_empty(hslot))
+			continue;
+
 		lockp = &nf_conntrack_locks[*bucket % CONNTRACK_LOCKS];
 		local_bh_disable();
 		nf_conntrack_lock(lockp);
-		if (*bucket < nf_conntrack_htable_size) {
-			hlist_nulls_for_each_entry(h, n, &nf_conntrack_hash[*bucket], hnnode) {
-				if (NF_CT_DIRECTION(h) != IP_CT_DIR_REPLY)
-					continue;
-				/* All nf_conn objects are added to hash table twice, one
-				 * for original direction tuple, once for the reply tuple.
-				 *
-				 * Exception: In the IPS_NAT_CLASH case, only the reply
-				 * tuple is added (the original tuple already existed for
-				 * a different object).
-				 *
-				 * We only need to call the iterator once for each
-				 * conntrack, so we just use the 'reply' direction
-				 * tuple while iterating.
-				 */
-				ct = nf_ct_tuplehash_to_ctrack(h);
-				if (iter(ct, data))
-					goto found;
-			}
+		hlist_nulls_for_each_entry(h, n, hslot, hnnode) {
+			if (NF_CT_DIRECTION(h) != IP_CT_DIR_REPLY)
+				continue;
+			/* All nf_conn objects are added to hash table twice, one
+			 * for original direction tuple, once for the reply tuple.
+			 *
+			 * Exception: In the IPS_NAT_CLASH case, only the reply
+			 * tuple is added (the original tuple already existed for
+			 * a different object).
+			 *
+			 * We only need to call the iterator once for each
+			 * conntrack, so we just use the 'reply' direction
+			 * tuple while iterating.
+			 */
+			ct = nf_ct_tuplehash_to_ctrack(h);
+			if (iter(ct, data))
+				goto found;
 		}
 		spin_unlock(lockp);
 		local_bh_enable();
@@ -2231,26 +2237,20 @@ get_next_corpse(int (*iter)(struct nf_conn *i, void *data),
 static void nf_ct_iterate_cleanup(int (*iter)(struct nf_conn *i, void *data),
 				  void *data, u32 portid, int report)
 {
-	unsigned int bucket = 0, sequence;
+	unsigned int bucket = 0;
 	struct nf_conn *ct;
 
 	might_sleep();
 
-	for (;;) {
-		sequence = read_seqcount_begin(&nf_conntrack_generation);
-
-		while ((ct = get_next_corpse(iter, data, &bucket)) != NULL) {
-			/* Time to push up daises... */
+	mutex_lock(&nf_conntrack_mutex);
+	while ((ct = get_next_corpse(iter, data, &bucket)) != NULL) {
+		/* Time to push up daises... */
 
-			nf_ct_delete(ct, portid, report);
-			nf_ct_put(ct);
-			cond_resched();
-		}
-
-		if (!read_seqcount_retry(&nf_conntrack_generation, sequence))
-			break;
-		bucket = 0;
+		nf_ct_delete(ct, portid, report);
+		nf_ct_put(ct);
+		cond_resched();
 	}
+	mutex_unlock(&nf_conntrack_mutex);
 }
 
 struct iter_data {
@@ -2486,8 +2486,10 @@ int nf_conntrack_hash_resize(unsigned int hashsize)
 	if (!hash)
 		return -ENOMEM;
 
+	mutex_lock(&nf_conntrack_mutex);
 	old_size = nf_conntrack_htable_size;
 	if (old_size == hashsize) {
+		mutex_unlock(&nf_conntrack_mutex);
 		kvfree(hash);
 		return 0;
 	}
@@ -2523,6 +2525,8 @@ int nf_conntrack_hash_resize(unsigned int hashsize)
 	nf_conntrack_all_unlock();
 	local_bh_enable();
 
+	mutex_unlock(&nf_conntrack_mutex);
+
 	synchronize_net();
 	kvfree(old_hash);
 	return 0;
-- 
2.33.0

[PATCH AUTOSEL 5.14 21/40] mac80211: Drop frames from invalid MAC address in ad-hoc mode

[Sasha Levin]

From: YueHaibing <yuehaibing@huawei.com>

[ Upstream commit a6555f844549cd190eb060daef595f94d3de1582 ]

WARNING: CPU: 1 PID: 9 at net/mac80211/sta_info.c:554
sta_info_insert_rcu+0x121/0x12a0
Modules linked in:
CPU: 1 PID: 9 Comm: kworker/u8:1 Not tainted 5.14.0-rc7+ #253
Workqueue: phy3 ieee80211_iface_work
RIP: 0010:sta_info_insert_rcu+0x121/0x12a0
...
Call Trace:
 ieee80211_ibss_finish_sta+0xbc/0x170
 ieee80211_ibss_work+0x13f/0x7d0
 ieee80211_iface_work+0x37a/0x500
 process_one_work+0x357/0x850
 worker_thread+0x41/0x4d0

If an Ad-Hoc node receives packets with invalid source MAC address,
it hits a WARN_ON in sta_info_insert_check(), this can spam the log.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Link: https://lore.kernel.org/r/20210827144230.39944-1-yuehaibing@huawei.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/mac80211/rx.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 2563473b5cf1..e023e307c0c3 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -4053,7 +4053,8 @@ static bool ieee80211_accept_frame(struct ieee80211_rx_data *rx)
 		if (!bssid)
 			return false;
 		if (ether_addr_equal(sdata->vif.addr, hdr->addr2) ||
-		    ether_addr_equal(sdata->u.ibss.bssid, hdr->addr2))
+		    ether_addr_equal(sdata->u.ibss.bssid, hdr->addr2) ||
+		    !is_valid_ether_addr(hdr->addr2))
 			return false;
 		if (ieee80211_is_beacon(hdr->frame_control))
 			return true;
-- 
2.33.0

[PATCH AUTOSEL 5.14 25/40] net: prevent user from passing illegal stab size

[Sasha Levin]

From: 王贇 <yun.wang@linux.alibaba.com>

[ Upstream commit b193e15ac69d56f35e1d8e2b5d16cbd47764d053 ]

We observed below report when playing with netlink sock:

  UBSAN: shift-out-of-bounds in net/sched/sch_api.c:580:10
  shift exponent 249 is too large for 32-bit type
  CPU: 0 PID: 685 Comm: a.out Not tainted
  Call Trace:
   dump_stack_lvl+0x8d/0xcf
   ubsan_epilogue+0xa/0x4e
   __ubsan_handle_shift_out_of_bounds+0x161/0x182
   __qdisc_calculate_pkt_len+0xf0/0x190
   __dev_queue_xmit+0x2ed/0x15b0

it seems like kernel won't check the stab log value passing from
user, and will use the insane value later to calculate pkt_len.

This patch just add a check on the size/cell_log to avoid insane
calculation.

Reported-by: Abaci <abaci@linux.alibaba.com>
Signed-off-by: Michael Wang <yun.wang@linux.alibaba.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 include/net/pkt_sched.h | 1 +
 net/sched/sch_api.c     | 6 ++++++
 2 files changed, 7 insertions(+)

diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index 6d7b12cba015..bf79f3a890af 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -11,6 +11,7 @@
 #include <uapi/linux/pkt_sched.h>
 
 #define DEFAULT_TX_QUEUE_LEN	1000
+#define STAB_SIZE_LOG_MAX	30
 
 struct qdisc_walker {
 	int	stop;
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index f87d07736a14..148edd0e71e3 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -513,6 +513,12 @@ static struct qdisc_size_table *qdisc_get_stab(struct nlattr *opt,
 		return stab;
 	}
 
+	if (s->size_log > STAB_SIZE_LOG_MAX ||
+	    s->cell_log > STAB_SIZE_LOG_MAX) {
+		NL_SET_ERR_MSG(extack, "Invalid logarithmic size of size table");
+		return ERR_PTR(-EINVAL);
+	}
+
 	stab = kmalloc(sizeof(*stab) + tsize * sizeof(u16), GFP_KERNEL);
 	if (!stab)
 		return ERR_PTR(-ENOMEM);
-- 
2.33.0

[PATCH AUTOSEL 5.14 26/40] mac80211: check return value of rhashtable_init

[Sasha Levin]

From: MichelleJin <shjy180909@gmail.com>

[ Upstream commit 111461d573741c17eafad029ac93474fa9adcce0 ]

When rhashtable_init() fails, it returns -EINVAL.
However, since error return value of rhashtable_init is not checked,
it can cause use of uninitialized pointers.
So, fix unhandled errors of rhashtable_init.

Signed-off-by: MichelleJin <shjy180909@gmail.com>
Link: https://lore.kernel.org/r/20210927033457.1020967-4-shjy180909@gmail.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/mac80211/mesh_pathtbl.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/net/mac80211/mesh_pathtbl.c b/net/mac80211/mesh_pathtbl.c
index efbefcbac3ac..7cab1cf09bf1 100644
--- a/net/mac80211/mesh_pathtbl.c
+++ b/net/mac80211/mesh_pathtbl.c
@@ -60,7 +60,10 @@ static struct mesh_table *mesh_table_alloc(void)
 	atomic_set(&newtbl->entries,  0);
 	spin_lock_init(&newtbl->gates_lock);
 	spin_lock_init(&newtbl->walk_lock);
-	rhashtable_init(&newtbl->rhead, &mesh_rht_params);
+	if (rhashtable_init(&newtbl->rhead, &mesh_rht_params)) {
+		kfree(newtbl);
+		return NULL;
+	}
 
 	return newtbl;
 }
-- 
2.33.0

[PATCH AUTOSEL 5.14 27/40] net: bgmac-platform: handle mac-address deferral

[Sasha Levin]

From: Matthew Hagan <mnhagan88@gmail.com>

[ Upstream commit 763716a55cb1f480ffe1a9702e6b5d9ea1a80a24 ]

This patch is a replication of Christian Lamparter's "net: bgmac-bcma:
handle deferred probe error due to mac-address" patch for the
bgmac-platform driver [1].

As is the case with the bgmac-bcma driver, this change is to cover the
scenario where the MAC address cannot yet be discovered due to reliance
on an nvmem provider which is yet to be instantiated, resulting in a
random address being assigned that has to be manually overridden.

[1] https://lore.kernel.org/netdev/20210919115725.29064-1-chunkeey@gmail.com

Signed-off-by: Matthew Hagan <mnhagan88@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/broadcom/bgmac-platform.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/broadcom/bgmac-platform.c b/drivers/net/ethernet/broadcom/bgmac-platform.c
index 4ab5bf64d353..df8ff839cc62 100644
--- a/drivers/net/ethernet/broadcom/bgmac-platform.c
+++ b/drivers/net/ethernet/broadcom/bgmac-platform.c
@@ -192,6 +192,9 @@ static int bgmac_probe(struct platform_device *pdev)
 	bgmac->dma_dev = &pdev->dev;
 
 	ret = of_get_mac_address(np, bgmac->net_dev->dev_addr);
+	if (ret == -EPROBE_DEFER)
+		return ret;
+
 	if (ret)
 		dev_warn(&pdev->dev,
 			 "MAC address not present in device tree\n");
-- 
2.33.0

[PATCH AUTOSEL 5.14 28/40] net: mdiobus: Fix memory leak in __mdiobus_register

[Sasha Levin]

From: Yanfei Xu <yanfei.xu@windriver.com>

[ Upstream commit ab609f25d19858513919369ff3d9a63c02cd9e2e ]

Once device_register() failed, we should call put_device() to
decrement reference count for cleanup. Or it will cause memory
leak.

BUG: memory leak
unreferenced object 0xffff888114032e00 (size 256):
  comm "kworker/1:3", pid 2960, jiffies 4294943572 (age 15.920s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 08 2e 03 14 81 88 ff ff  ................
    08 2e 03 14 81 88 ff ff 90 76 65 82 ff ff ff ff  .........ve.....
  backtrace:
    [<ffffffff8265cfab>] kmalloc include/linux/slab.h:591 [inline]
    [<ffffffff8265cfab>] kzalloc include/linux/slab.h:721 [inline]
    [<ffffffff8265cfab>] device_private_init drivers/base/core.c:3203 [inline]
    [<ffffffff8265cfab>] device_add+0x89b/0xdf0 drivers/base/core.c:3253
    [<ffffffff828dd643>] __mdiobus_register+0xc3/0x450 drivers/net/phy/mdio_bus.c:537
    [<ffffffff828cb835>] __devm_mdiobus_register+0x75/0xf0 drivers/net/phy/mdio_devres.c:87
    [<ffffffff82b92a00>] ax88772_init_mdio drivers/net/usb/asix_devices.c:676 [inline]
    [<ffffffff82b92a00>] ax88772_bind+0x330/0x480 drivers/net/usb/asix_devices.c:786
    [<ffffffff82baa33f>] usbnet_probe+0x3ff/0xdf0 drivers/net/usb/usbnet.c:1745
    [<ffffffff82c36e17>] usb_probe_interface+0x177/0x370 drivers/usb/core/driver.c:396
    [<ffffffff82661d17>] call_driver_probe drivers/base/dd.c:517 [inline]
    [<ffffffff82661d17>] really_probe.part.0+0xe7/0x380 drivers/base/dd.c:596
    [<ffffffff826620bc>] really_probe drivers/base/dd.c:558 [inline]
    [<ffffffff826620bc>] __driver_probe_device+0x10c/0x1e0 drivers/base/dd.c:751
    [<ffffffff826621ba>] driver_probe_device+0x2a/0x120 drivers/base/dd.c:781
    [<ffffffff82662a26>] __device_attach_driver+0xf6/0x140 drivers/base/dd.c:898
    [<ffffffff8265eca7>] bus_for_each_drv+0xb7/0x100 drivers/base/bus.c:427
    [<ffffffff826625a2>] __device_attach+0x122/0x260 drivers/base/dd.c:969
    [<ffffffff82660916>] bus_probe_device+0xc6/0xe0 drivers/base/bus.c:487
    [<ffffffff8265cd0b>] device_add+0x5fb/0xdf0 drivers/base/core.c:3359
    [<ffffffff82c343b9>] usb_set_configuration+0x9d9/0xb90 drivers/usb/core/message.c:2170
    [<ffffffff82c4473c>] usb_generic_driver_probe+0x8c/0xc0 drivers/usb/core/generic.c:238

BUG: memory leak
unreferenced object 0xffff888116f06900 (size 32):
  comm "kworker/0:2", pid 2670, jiffies 4294944448 (age 7.160s)
  hex dump (first 32 bytes):
    75 73 62 2d 30 30 31 3a 30 30 33 00 00 00 00 00  usb-001:003.....
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<ffffffff81484516>] kstrdup+0x36/0x70 mm/util.c:60
    [<ffffffff814845a3>] kstrdup_const+0x53/0x80 mm/util.c:83
    [<ffffffff82296ba2>] kvasprintf_const+0xc2/0x110 lib/kasprintf.c:48
    [<ffffffff82358d4b>] kobject_set_name_vargs+0x3b/0xe0 lib/kobject.c:289
    [<ffffffff826575f3>] dev_set_name+0x63/0x90 drivers/base/core.c:3147
    [<ffffffff828dd63b>] __mdiobus_register+0xbb/0x450 drivers/net/phy/mdio_bus.c:535
    [<ffffffff828cb835>] __devm_mdiobus_register+0x75/0xf0 drivers/net/phy/mdio_devres.c:87
    [<ffffffff82b92a00>] ax88772_init_mdio drivers/net/usb/asix_devices.c:676 [inline]
    [<ffffffff82b92a00>] ax88772_bind+0x330/0x480 drivers/net/usb/asix_devices.c:786
    [<ffffffff82baa33f>] usbnet_probe+0x3ff/0xdf0 drivers/net/usb/usbnet.c:1745
    [<ffffffff82c36e17>] usb_probe_interface+0x177/0x370 drivers/usb/core/driver.c:396
    [<ffffffff82661d17>] call_driver_probe drivers/base/dd.c:517 [inline]
    [<ffffffff82661d17>] really_probe.part.0+0xe7/0x380 drivers/base/dd.c:596
    [<ffffffff826620bc>] really_probe drivers/base/dd.c:558 [inline]
    [<ffffffff826620bc>] __driver_probe_device+0x10c/0x1e0 drivers/base/dd.c:751
    [<ffffffff826621ba>] driver_probe_device+0x2a/0x120 drivers/base/dd.c:781
    [<ffffffff82662a26>] __device_attach_driver+0xf6/0x140 drivers/base/dd.c:898
    [<ffffffff8265eca7>] bus_for_each_drv+0xb7/0x100 drivers/base/bus.c:427
    [<ffffffff826625a2>] __device_attach+0x122/0x260 drivers/base/dd.c:969

Reported-by: syzbot+398e7dc692ddbbb4cfec@syzkaller.appspotmail.com
Signed-off-by: Yanfei Xu <yanfei.xu@windriver.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/phy/mdio_bus.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/phy/mdio_bus.c b/drivers/net/phy/mdio_bus.c
index 53f034fc2ef7..6f4b4e5df639 100644
--- a/drivers/net/phy/mdio_bus.c
+++ b/drivers/net/phy/mdio_bus.c
@@ -537,6 +537,7 @@ int __mdiobus_register(struct mii_bus *bus, struct module *owner)
 	err = device_register(&bus->dev);
 	if (err) {
 		pr_err("mii_bus %s failed to register\n", bus->id);
+		put_device(&bus->dev);
 		return -EINVAL;
 	}
 
-- 
2.33.0

[PATCH AUTOSEL 5.14 31/40] net: sun: SUNVNET_COMMON should depend on INET

[Sasha Levin]

From: Randy Dunlap <rdunlap@infradead.org>

[ Upstream commit 103bde372f084206c6972be543ecc247ebbff9f3 ]

When CONFIG_INET is not set, there are failing references to IPv4
functions, so make this driver depend on INET.

Fixes these build errors:

sparc64-linux-ld: drivers/net/ethernet/sun/sunvnet_common.o: in function `sunvnet_start_xmit_common':
sunvnet_common.c:(.text+0x1a68): undefined reference to `__icmp_send'
sparc64-linux-ld: drivers/net/ethernet/sun/sunvnet_common.o: in function `sunvnet_poll_common':
sunvnet_common.c:(.text+0x358c): undefined reference to `ip_send_check'

Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jakub Kicinski <kuba@kernel.org>
Cc: Aaron Young <aaron.young@oracle.com>
Cc: Rashmi Narasimhan <rashmi.narasimhan@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/sun/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/sun/Kconfig b/drivers/net/ethernet/sun/Kconfig
index 309de38a7530..b0d3f9a2950c 100644
--- a/drivers/net/ethernet/sun/Kconfig
+++ b/drivers/net/ethernet/sun/Kconfig
@@ -73,6 +73,7 @@ config CASSINI
 config SUNVNET_COMMON
 	tristate "Common routines to support Sun Virtual Networking"
 	depends on SUN_LDOMS
+	depends on INET
 	default m
 
 config SUNVNET
-- 
2.33.0

Re: [PATCH AUTOSEL 5.14 28/40] net: mdiobus: Fix memory leak in __mdiobus_register

[Andrew Lunn]

On Tue, Oct 05, 2021 at 09:50:07AM -0400, Sasha Levin wrote:
> From: Yanfei Xu <yanfei.xu@windriver.com>
> 
> [ Upstream commit ab609f25d19858513919369ff3d9a63c02cd9e2e ]
> 
> Once device_register() failed, we should call put_device() to
> decrement reference count for cleanup. Or it will cause memory
> leak.

Hi Sasha

https://lkml.org/lkml/2021/10/4/1427

Please don't backport for any stable kernel.

	Andrew

Re: [PATCH AUTOSEL 5.14 28/40] net: mdiobus: Fix memory leak in __mdiobus_register

[Sasha Levin]

On Tue, Oct 05, 2021 at 04:02:23PM +0200, Andrew Lunn wrote:
>On Tue, Oct 05, 2021 at 09:50:07AM -0400, Sasha Levin wrote:
>> From: Yanfei Xu <yanfei.xu@windriver.com>
>>
>> [ Upstream commit ab609f25d19858513919369ff3d9a63c02cd9e2e ]
>>
>> Once device_register() failed, we should call put_device() to
>> decrement reference count for cleanup. Or it will cause memory
>> leak.
>
>Hi Sasha
>
>https://lkml.org/lkml/2021/10/4/1427
>
>Please don't backport for any stable kernel.

Dropped, thanks!

-- 
Thanks,
Sasha