Nikolay Aleksandrov
2021-Jun-10 10:45 UTC
[Bridge] [PATCH net 0/2] net: bridge: vlan tunnel egress path fixes
From: Nikolay Aleksandrov <nikolay at nvidia.com>
Hi,
These two fixes take care of tunnel_dst problems in the vlan tunnel egress
path. Patch 01 fixes a null ptr deref due to the lockless use of tunnel_dst
pointer without checking it first, and patch 02 fixes a use-after-free
issue due to wrong dst refcounting (dst_clone() -> dst_hold_safe()).
Both fix the same commit and should be queued for stable backports:
11538d039ac6 ("bridge: vlan dst_metadata hooks in ingress and egress
paths")
Thanks,
Nik
Nikolay Aleksandrov (2):
net: bridge: fix vlan tunnel dst null pointer dereference
net: bridge: fix vlan tunnel dst refcnt when egressing
net/bridge/br_private.h | 4 ++--
net/bridge/br_vlan_tunnel.c | 38 +++++++++++++++++++++++--------------
2 files changed, 26 insertions(+), 16 deletions(-)
--
2.31.1
Nikolay Aleksandrov
2021-Jun-10 10:45 UTC
[Bridge] [PATCH net 1/2] net: bridge: fix vlan tunnel dst null pointer dereference
From: Nikolay Aleksandrov <nikolay at nvidia.com>
This patch fixes a tunnel_dst null pointer dereference due to lockless
access in the tunnel egress path. When deleting a vlan tunnel the
tunnel_dst pointer is set to NULL without waiting a grace period (i.e.
while it's still usable) and packets egressing are dereferencing it
without checking. Use READ/WRITE_ONCE to annotate the lockless use of
tunnel_id, use RCU for accessing tunnel_dst and make sure it is read
only once and checked in the egress path. The dst is already properly RCU
protected so we don't need to do anything fancy than to make sure
tunnel_id and tunnel_dst are read only once and checked in the egress path.
Fixes: 11538d039ac6 ("bridge: vlan dst_metadata hooks in ingress and egress
paths")
Signed-off-by: Nikolay Aleksandrov <nikolay at nvidia.com>
---
net/bridge/br_private.h | 4 ++--
net/bridge/br_vlan_tunnel.c | 38 +++++++++++++++++++++++--------------
2 files changed, 26 insertions(+), 16 deletions(-)
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index 7ce8a77cc6b6..e013d33f1c7c 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -90,8 +90,8 @@ struct bridge_mcast_stats {
#endif
struct br_tunnel_info {
- __be64 tunnel_id;
- struct metadata_dst *tunnel_dst;
+ __be64 tunnel_id;
+ struct metadata_dst __rcu *tunnel_dst;
};
/* private vlan flags */
diff --git a/net/bridge/br_vlan_tunnel.c b/net/bridge/br_vlan_tunnel.c
index 0d3a8c01552e..03de461a0d44 100644
--- a/net/bridge/br_vlan_tunnel.c
+++ b/net/bridge/br_vlan_tunnel.c
@@ -41,26 +41,33 @@ static struct net_bridge_vlan *br_vlan_tunnel_lookup(struct
rhashtable *tbl,
br_vlan_tunnel_rht_params);
}
+static void vlan_tunnel_info_release(struct net_bridge_vlan *vlan)
+{
+ struct metadata_dst *tdst = rtnl_dereference(vlan->tinfo.tunnel_dst);
+
+ WRITE_ONCE(vlan->tinfo.tunnel_id, 0);
+ RCU_INIT_POINTER(vlan->tinfo.tunnel_dst, NULL);
+ dst_release(&tdst->dst);
+}
+
void vlan_tunnel_info_del(struct net_bridge_vlan_group *vg,
struct net_bridge_vlan *vlan)
{
- if (!vlan->tinfo.tunnel_dst)
+ if (!rcu_access_pointer(vlan->tinfo.tunnel_dst))
return;
rhashtable_remove_fast(&vg->tunnel_hash, &vlan->tnode,
br_vlan_tunnel_rht_params);
- vlan->tinfo.tunnel_id = 0;
- dst_release(&vlan->tinfo.tunnel_dst->dst);
- vlan->tinfo.tunnel_dst = NULL;
+ vlan_tunnel_info_release(vlan);
}
static int __vlan_tunnel_info_add(struct net_bridge_vlan_group *vg,
struct net_bridge_vlan *vlan, u32 tun_id)
{
- struct metadata_dst *metadata = NULL;
+ struct metadata_dst *metadata = rtnl_dereference(vlan->tinfo.tunnel_dst);
__be64 key = key32_to_tunnel_id(cpu_to_be32(tun_id));
int err;
- if (vlan->tinfo.tunnel_dst)
+ if (metadata)
return -EEXIST;
metadata = __ip_tun_set_dst(0, 0, 0, 0, 0, TUNNEL_KEY,
@@ -69,8 +76,8 @@ static int __vlan_tunnel_info_add(struct net_bridge_vlan_group
*vg,
return -EINVAL;
metadata->u.tun_info.mode |= IP_TUNNEL_INFO_TX | IP_TUNNEL_INFO_BRIDGE;
- vlan->tinfo.tunnel_dst = metadata;
- vlan->tinfo.tunnel_id = key;
+ rcu_assign_pointer(vlan->tinfo.tunnel_dst, metadata);
+ WRITE_ONCE(vlan->tinfo.tunnel_id, key);
err = rhashtable_lookup_insert_fast(&vg->tunnel_hash,
&vlan->tnode,
br_vlan_tunnel_rht_params);
@@ -79,9 +86,7 @@ static int __vlan_tunnel_info_add(struct net_bridge_vlan_group
*vg,
return 0;
out:
- dst_release(&vlan->tinfo.tunnel_dst->dst);
- vlan->tinfo.tunnel_dst = NULL;
- vlan->tinfo.tunnel_id = 0;
+ vlan_tunnel_info_release(vlan);
return err;
}
@@ -182,12 +187,15 @@ int br_handle_ingress_vlan_tunnel(struct sk_buff *skb,
int br_handle_egress_vlan_tunnel(struct sk_buff *skb,
struct net_bridge_vlan *vlan)
{
+ struct metadata_dst *tunnel_dst;
+ __be64 tunnel_id;
int err;
- if (!vlan || !vlan->tinfo.tunnel_id)
+ if (!vlan)
return 0;
- if (unlikely(!skb_vlan_tag_present(skb)))
+ tunnel_id = READ_ONCE(vlan->tinfo.tunnel_id);
+ if (!tunnel_id || unlikely(!skb_vlan_tag_present(skb)))
return 0;
skb_dst_drop(skb);
@@ -195,7 +203,9 @@ int br_handle_egress_vlan_tunnel(struct sk_buff *skb,
if (err)
return err;
- skb_dst_set(skb, dst_clone(&vlan->tinfo.tunnel_dst->dst));
+ tunnel_dst = rcu_dereference(vlan->tinfo.tunnel_dst);
+ if (tunnel_dst)
+ skb_dst_set(skb, dst_clone(&tunnel_dst->dst));
return 0;
}
--
2.31.1
Nikolay Aleksandrov
2021-Jun-10 10:45 UTC
[Bridge] [PATCH net 2/2] net: bridge: fix vlan tunnel dst refcnt when egressing
From: Nikolay Aleksandrov <nikolay at nvidia.com>
The egress tunnel code uses dst_clone() and directly sets the result
which is wrong because the entry might have 0 refcnt or be already deleted,
causing number of problems. It also triggers the WARN_ON() in dst_hold()[1]
when a refcnt couldn't be taken. Fix it by using dst_hold_safe() and
checking if a reference was actually taken before setting the dst.
[1] dmesg WARN_ON log and following refcnt errors
WARNING: CPU: 5 PID: 38 at include/net/dst.h:230
br_handle_egress_vlan_tunnel+0x10b/0x134 [bridge]
Modules linked in: 8021q garp mrp bridge stp llc bonding ipv6 virtio_net
CPU: 5 PID: 38 Comm: ksoftirqd/5 Kdump: loaded Tainted: G W
5.13.0-rc3+ #360
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.14.0-1.fc33
04/01/2014
RIP: 0010:br_handle_egress_vlan_tunnel+0x10b/0x134 [bridge]
Code: e8 85 bc 01 e1 45 84 f6 74 90 45 31 f6 85 db 48 c7 c7 a0 02 19 a0 41 0f
94 c6 31 c9 31 d2 44 89 f6 e8 64 bc 01 e1 85 db 75 02 <0f> 0b 31 c9 31 d2
44 89 f6 48 c7 c7 70 02 19 a0 e8 4b bc 01 e1 49
RSP: 0018:ffff8881003d39e8 EFLAGS: 00010246
RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffffffffa01902a0
RBP: ffff8881040c6700 R08: 0000000000000000 R09: 0000000000000001
R10: 2ce93d0054fe0d00 R11: 54fe0d00000e0000 R12: ffff888109515000
R13: 0000000000000000 R14: 0000000000000001 R15: 0000000000000401
FS: 0000000000000000(0000) GS:ffff88822bf40000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f42ba70f030 CR3: 0000000109926000 CR4: 00000000000006e0
Call Trace:
br_handle_vlan+0xbc/0xca [bridge]
__br_forward+0x23/0x164 [bridge]
deliver_clone+0x41/0x48 [bridge]
br_handle_frame_finish+0x36f/0x3aa [bridge]
? skb_dst+0x2e/0x38 [bridge]
? br_handle_ingress_vlan_tunnel+0x3e/0x1c8 [bridge]
? br_handle_frame_finish+0x3aa/0x3aa [bridge]
br_handle_frame+0x2c3/0x377 [bridge]
? __skb_pull+0x33/0x51
? vlan_do_receive+0x4f/0x36a
? br_handle_frame_finish+0x3aa/0x3aa [bridge]
__netif_receive_skb_core+0x539/0x7c6
? __list_del_entry_valid+0x16e/0x1c2
__netif_receive_skb_list_core+0x6d/0xd6
netif_receive_skb_list_internal+0x1d9/0x1fa
gro_normal_list+0x22/0x3e
dev_gro_receive+0x55b/0x600
? detach_buf_split+0x58/0x140
napi_gro_receive+0x94/0x12e
virtnet_poll+0x15d/0x315 [virtio_net]
__napi_poll+0x2c/0x1c9
net_rx_action+0xe6/0x1fb
__do_softirq+0x115/0x2d8
run_ksoftirqd+0x18/0x20
smpboot_thread_fn+0x183/0x19c
? smpboot_unregister_percpu_thread+0x66/0x66
kthread+0x10a/0x10f
? kthread_mod_delayed_work+0xb6/0xb6
ret_from_fork+0x22/0x30
---[ end trace 49f61b07f775fd2b ]---
dst_release: dst:00000000c02d677a refcnt:-1
dst_release underflow
Fixes: 11538d039ac6 ("bridge: vlan dst_metadata hooks in ingress and egress
paths")
Signed-off-by: Nikolay Aleksandrov <nikolay at nvidia.com>
---
net/bridge/br_vlan_tunnel.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/net/bridge/br_vlan_tunnel.c b/net/bridge/br_vlan_tunnel.c
index 03de461a0d44..01017448ebde 100644
--- a/net/bridge/br_vlan_tunnel.c
+++ b/net/bridge/br_vlan_tunnel.c
@@ -204,8 +204,8 @@ int br_handle_egress_vlan_tunnel(struct sk_buff *skb,
return err;
tunnel_dst = rcu_dereference(vlan->tinfo.tunnel_dst);
- if (tunnel_dst)
- skb_dst_set(skb, dst_clone(&tunnel_dst->dst));
+ if (tunnel_dst && dst_hold_safe(&tunnel_dst->dst))
+ skb_dst_set(skb, &tunnel_dst->dst);
return 0;
}
--
2.31.1