* [PATCH 6.1.y] net: dsa: fix netdev_priv() dereference before check on non-DSA netdevice events
@ 2025-02-07 7:06 jetlan9
2025-02-07 22:51 ` Sasha Levin
2025-02-08 2:18 ` Wenshan Lan
0 siblings, 2 replies; 3+ messages in thread
From: jetlan9 @ 2025-02-07 7:06 UTC (permalink / raw)
To: stable
Cc: Vladimir Oltean, syzbot+d81bcd883824180500c8, Florian Fainelli,
Eric Dumazet, Jakub Kicinski, Wenshan Lan
From: Vladimir Oltean <vladimir.oltean@nxp.com>
After the blamed commit, we started doing this dereference for every
NETDEV_CHANGEUPPER and NETDEV_PRECHANGEUPPER event in the system.
static inline struct dsa_port *dsa_user_to_port(const struct net_device *dev)
{
struct dsa_user_priv *p = netdev_priv(dev);
return p->dp;
}
Which is obviously bogus, because not all net_devices have a netdev_priv()
of type struct dsa_user_priv. But struct dsa_user_priv is fairly small,
and p->dp means dereferencing 8 bytes starting with offset 16. Most
drivers allocate that much private memory anyway, making our access not
fault, and we discard the bogus data quickly afterwards, so this wasn't
caught.
But the dummy interface is somewhat special in that it calls
alloc_netdev() with a priv size of 0. So every netdev_priv() dereference
is invalid, and we get this when we emit a NETDEV_PRECHANGEUPPER event
with a VLAN as its new upper:
$ ip link add dummy1 type dummy
$ ip link add link dummy1 name dummy1.100 type vlan id 100
[ 43.309174] ==================================================================
[ 43.316456] BUG: KASAN: slab-out-of-bounds in dsa_user_prechangeupper+0x30/0xe8
[ 43.323835] Read of size 8 at addr ffff3f86481d2990 by task ip/374
[ 43.330058]
[ 43.342436] Call trace:
[ 43.366542] dsa_user_prechangeupper+0x30/0xe8
[ 43.371024] dsa_user_netdevice_event+0xb38/0xee8
[ 43.375768] notifier_call_chain+0xa4/0x210
[ 43.379985] raw_notifier_call_chain+0x24/0x38
[ 43.384464] __netdev_upper_dev_link+0x3ec/0x5d8
[ 43.389120] netdev_upper_dev_link+0x70/0xa8
[ 43.393424] register_vlan_dev+0x1bc/0x310
[ 43.397554] vlan_newlink+0x210/0x248
[ 43.401247] rtnl_newlink+0x9fc/0xe30
[ 43.404942] rtnetlink_rcv_msg+0x378/0x580
Avoid the kernel oops by dereferencing after the type check, as customary.
Fixes: 4c3f80d22b2e ("net: dsa: walk through all changeupper notifier functions")
Reported-and-tested-by: syzbot+d81bcd883824180500c8@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/netdev/0000000000001d4255060e87545c@google.com/
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240110003354.2796778-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Wenshan Lan <jetlan9@163.com>
---
net/dsa/slave.c | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 5fe075bf479e..caeb7e75b287 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -2592,13 +2592,14 @@ EXPORT_SYMBOL_GPL(dsa_slave_dev_check);
static int dsa_slave_changeupper(struct net_device *dev,
struct netdev_notifier_changeupper_info *info)
{
- struct dsa_port *dp = dsa_slave_to_port(dev);
struct netlink_ext_ack *extack;
int err = NOTIFY_DONE;
+ struct dsa_port *dp;
if (!dsa_slave_dev_check(dev))
return err;
+ dp = dsa_slave_to_port(dev);
extack = netdev_notifier_info_to_extack(&info->info);
if (netif_is_bridge_master(info->upper_dev)) {
@@ -2652,11 +2653,13 @@ static int dsa_slave_changeupper(struct net_device *dev,
static int dsa_slave_prechangeupper(struct net_device *dev,
struct netdev_notifier_changeupper_info *info)
{
- struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_port *dp;
if (!dsa_slave_dev_check(dev))
return NOTIFY_DONE;
+ dp = dsa_slave_to_port(dev);
+
if (netif_is_bridge_master(info->upper_dev) && !info->linking)
dsa_port_pre_bridge_leave(dp, info->upper_dev);
else if (netif_is_lag_master(info->upper_dev) && !info->linking)
--
2.43.0
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH 6.1.y] net: dsa: fix netdev_priv() dereference before check on non-DSA netdevice events
2025-02-07 7:06 [PATCH 6.1.y] net: dsa: fix netdev_priv() dereference before check on non-DSA netdevice events jetlan9
@ 2025-02-07 22:51 ` Sasha Levin
2025-02-08 2:18 ` Wenshan Lan
1 sibling, 0 replies; 3+ messages in thread
From: Sasha Levin @ 2025-02-07 22:51 UTC (permalink / raw)
To: stable; +Cc: jetlan9, Sasha Levin
[ Sasha's backport helper bot ]
Hi,
Found matching upstream commit: 844f104790bd69c2e4dbb9ee3eba46fde1fcea7b
WARNING: Author mismatch between patch and found commit:
Backport author: jetlan9@163.com
Commit author: Vladimir Oltean<vladimir.oltean@nxp.com>
Status in newer kernel trees:
6.13.y | Present (exact SHA1)
6.12.y | Present (exact SHA1)
6.6.y | Present (different SHA1: 69a1e2d938db)
6.1.y | Not found
Note: The patch differs from the upstream commit:
---
1: 844f104790bd6 < -: ------------- net: dsa: fix netdev_priv() dereference before check on non-DSA netdevice events
-: ------------- > 1: 1c02698e9064e net: dsa: fix netdev_priv() dereference before check on non-DSA netdevice events
---
Results of testing on various branches:
| Branch | Patch Apply | Build Test |
|---------------------------|-------------|------------|
| stable/linux-6.1.y | Success | Failed |
Build Errors:
Build error for stable/linux-6.1.y:
ssh: connect to host 192.168.1.58 port 22: No route to host
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re:[PATCH 6.1.y] net: dsa: fix netdev_priv() dereference before check on non-DSA netdevice events
2025-02-07 7:06 [PATCH 6.1.y] net: dsa: fix netdev_priv() dereference before check on non-DSA netdevice events jetlan9
2025-02-07 22:51 ` Sasha Levin
@ 2025-02-08 2:18 ` Wenshan Lan
1 sibling, 0 replies; 3+ messages in thread
From: Wenshan Lan @ 2025-02-08 2:18 UTC (permalink / raw)
To: stable
Cc: Vladimir Oltean, syzbot+d81bcd883824180500c8, Florian Fainelli,
Eric Dumazet, Jakub Kicinski
Please ignore this patch for I forgot to add the upstream commit in this commit log, sorry for it.
At 2025-02-07 15:06:43, jetlan9@163.com wrote:
>From: Vladimir Oltean <vladimir.oltean@nxp.com>
>
>After the blamed commit, we started doing this dereference for every
>NETDEV_CHANGEUPPER and NETDEV_PRECHANGEUPPER event in the system.
>
>static inline struct dsa_port *dsa_user_to_port(const struct net_device *dev)
>{
> struct dsa_user_priv *p = netdev_priv(dev);
>
> return p->dp;
>}
>
>Which is obviously bogus, because not all net_devices have a netdev_priv()
>of type struct dsa_user_priv. But struct dsa_user_priv is fairly small,
>and p->dp means dereferencing 8 bytes starting with offset 16. Most
>drivers allocate that much private memory anyway, making our access not
>fault, and we discard the bogus data quickly afterwards, so this wasn't
>caught.
>
>But the dummy interface is somewhat special in that it calls
>alloc_netdev() with a priv size of 0. So every netdev_priv() dereference
>is invalid, and we get this when we emit a NETDEV_PRECHANGEUPPER event
>with a VLAN as its new upper:
>
>$ ip link add dummy1 type dummy
>$ ip link add link dummy1 name dummy1.100 type vlan id 100
>[ 43.309174] ==================================================================
>[ 43.316456] BUG: KASAN: slab-out-of-bounds in dsa_user_prechangeupper+0x30/0xe8
>[ 43.323835] Read of size 8 at addr ffff3f86481d2990 by task ip/374
>[ 43.330058]
>[ 43.342436] Call trace:
>[ 43.366542] dsa_user_prechangeupper+0x30/0xe8
>[ 43.371024] dsa_user_netdevice_event+0xb38/0xee8
>[ 43.375768] notifier_call_chain+0xa4/0x210
>[ 43.379985] raw_notifier_call_chain+0x24/0x38
>[ 43.384464] __netdev_upper_dev_link+0x3ec/0x5d8
>[ 43.389120] netdev_upper_dev_link+0x70/0xa8
>[ 43.393424] register_vlan_dev+0x1bc/0x310
>[ 43.397554] vlan_newlink+0x210/0x248
>[ 43.401247] rtnl_newlink+0x9fc/0xe30
>[ 43.404942] rtnetlink_rcv_msg+0x378/0x580
>
>Avoid the kernel oops by dereferencing after the type check, as customary.
>
>Fixes: 4c3f80d22b2e ("net: dsa: walk through all changeupper notifier functions")
>Reported-and-tested-by: syzbot+d81bcd883824180500c8@syzkaller.appspotmail.com
>Closes: https://lore.kernel.org/netdev/0000000000001d4255060e87545c@google.com/
>Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
>Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
>Reviewed-by: Eric Dumazet <edumazet@google.com>
>Link: https://lore.kernel.org/r/20240110003354.2796778-1-vladimir.oltean@nxp.com
>Signed-off-by: Jakub Kicinski <kuba@kernel.org>
>Signed-off-by: Wenshan Lan <jetlan9@163.com>
>---
> net/dsa/slave.c | 7 +++++--
> 1 file changed, 5 insertions(+), 2 deletions(-)
>
>diff --git a/net/dsa/slave.c b/net/dsa/slave.c
>index 5fe075bf479e..caeb7e75b287 100644
>--- a/net/dsa/slave.c
>+++ b/net/dsa/slave.c
>@@ -2592,13 +2592,14 @@ EXPORT_SYMBOL_GPL(dsa_slave_dev_check);
> static int dsa_slave_changeupper(struct net_device *dev,
> struct netdev_notifier_changeupper_info *info)
> {
>- struct dsa_port *dp = dsa_slave_to_port(dev);
> struct netlink_ext_ack *extack;
> int err = NOTIFY_DONE;
>+ struct dsa_port *dp;
>
> if (!dsa_slave_dev_check(dev))
> return err;
>
>+ dp = dsa_slave_to_port(dev);
> extack = netdev_notifier_info_to_extack(&info->info);
>
> if (netif_is_bridge_master(info->upper_dev)) {
>@@ -2652,11 +2653,13 @@ static int dsa_slave_changeupper(struct net_device *dev,
> static int dsa_slave_prechangeupper(struct net_device *dev,
> struct netdev_notifier_changeupper_info *info)
> {
>- struct dsa_port *dp = dsa_slave_to_port(dev);
>+ struct dsa_port *dp;
>
> if (!dsa_slave_dev_check(dev))
> return NOTIFY_DONE;
>
>+ dp = dsa_slave_to_port(dev);
>+
> if (netif_is_bridge_master(info->upper_dev) && !info->linking)
> dsa_port_pre_bridge_leave(dp, info->upper_dev);
> else if (netif_is_lag_master(info->upper_dev) && !info->linking)
>--
>2.43.0
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2025-02-08 2:20 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-02-07 7:06 [PATCH 6.1.y] net: dsa: fix netdev_priv() dereference before check on non-DSA netdevice events jetlan9
2025-02-07 22:51 ` Sasha Levin
2025-02-08 2:18 ` Wenshan Lan
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox