[PATCH net v4] netrom: linearize and validate lengths in nr_rx_frame()

[PATCH net v4] netrom: linearize and validate lengths in nr_rx_frame()

[Stanislav Fort]

Linearize skb and add targeted length checks in nr_rx_frame() to
prevent out-of-bounds reads and potential use-after-free (UAF) when
processing malformed NET/ROM frames.

- Require skb to be linear and at least NR_NETWORK_LEN +
  NR_TRANSPORT_LEN (20 bytes) before reading
  network/transport fields.
- For existing sockets path, ensure NR_CONNACK includes the window
  byte (>= 21 bytes).
- For CONNREQ handling, ensure window (byte 20) and user address
  (bytes 21-27) are present (>= 28 bytes).
- Preserve existing BPQ extension handling:
  - NR_CONNACK len == 22 -> 1 extra byte (TTL)
  - NR_CONNREQ len == 37 -> 2 extra bytes (timeout)

These validations block malformed frames from triggering bounds
violations via short skbs. Because NET/ROM frames may originate from
remote peers, the issue is externally triggerable and therefore
security-relevant.

Suggested-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Cc: stable@vger.kernel.org
Signed-off-by: Stanislav Fort <disclosure@aisle.com>
---
Changes since v3:
- Wrap commit message at 74 columns
- Add Fixes tag and Cc: stable
- Drop Reported-by
- Add Reviewed-by from Eric Dumazet
---
 net/netrom/af_netrom.c | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c
index 3331669d8e33..f0660dd6d3b0 100644
--- a/net/netrom/af_netrom.c
+++ b/net/netrom/af_netrom.c
@@ -885,6 +885,11 @@ int nr_rx_frame(struct sk_buff *skb, struct net_device *dev)
 	/*
 	 *		skb->data points to the netrom frame start
 	 */

 +	if (skb_linearize(skb))
 +		return 0;
 +	if (skb->len < NR_NETWORK_LEN + NR_TRANSPORT_LEN)
 +		return 0;
 +
 	src 	= (ax25_address *)(skb->data + 0);
 	dest = (ax25_address *)(skb->data + 7);

@@ -927,6 +932,11 @@ int nr_rx_frame(struct sk_buff *skb, struct net_device *dev)
 	}

 	if (sk != NULL) {
 +		if (frametype == NR_CONNACK &&
 +		    skb->len < NR_NETWORK_LEN + NR_TRANSPORT_LEN + 1) {
 +			sock_put(sk);
 +			return 0;
 +		}
 		bh_lock_sock(sk);
 		skb_reset_transport_header(skb);

@@ -961,10 +971,14 @@ int nr_rx_frame(struct sk_buff *skb, struct net_device *dev)
 		return 0;
 	}

-	sk = nr_find_listener(dest);
 +	/* Need window (byte 20) and user address (bytes 21-27) */
 +	if (skb->len < NR_NETWORK_LEN + NR_TRANSPORT_LEN + 1 + AX25_ADDR_LEN)
 +		return 0;

 	user = (ax25_address *)(skb->data + 21);

 +	sk = nr_find_listener(dest);
 +
 	if (sk == NULL || sk_acceptq_is_full(sk) ||
 	    (make = nr_make_new(sk)) == NULL) {
 		nr_transmit_refusal(skb, 0);
-- 
2.39.3 (Apple Git-146)

Re: [PATCH net v4] netrom: linearize and validate lengths in nr_rx_frame()

[Jiri Kosina]

On Sun, 7 Sep 2025, Bernard Pidoux wrote:

> While applying netrom PATCH net v4
> patch says that 
> it is malformed on line 12.
> 
> However I cannot see any reason.

There is a superfluous leading space on each line of the diff. (wasn't the 
case with the previous patches, v4 is the first one to have this).

-- 
Jiri Kosina
SUSE Labs

Re: [PATCH net v4] netrom: linearize and validate lengths in nr_rx_frame()

[Jakub Kicinski]

On Sun, 7 Sep 2025 16:32:03 +0200 Bernard Pidoux wrote:
> While applying netrom PATCH net v4
> patch says that 
> it is malformed on line 12.

FWIW the version I received is completely mangled. There's a leading
space before each +. You can try B4 relay if your mail server is giving
you grief.

Re: [PATCH net v4] netrom: linearize and validate lengths in nr_rx_frame()

[Bernard f6bvp]

[-- Attachment #1: Type: text/plain, Size: 528 bytes --]

Hi,

There was also an additional tab in line 14...

Here is well formated patch I applied after fixing it for my own use.

NetRom is performing fine.

Thanks,

Bernard

Le 09/09/2025 à 02:43, Jakub Kicinski a écrit :
> On Sun, 7 Sep 2025 16:32:03 +0200 Bernard Pidoux wrote:
>> While applying netrom PATCH net v4
>> patch says that
>> it is malformed on line 12.
> 
> FWIW the version I received is completely mangled. There's a leading
> space before each +. You can try B4 relay if your mail server is giving
> you grief.

[-- Attachment #2: NetRom_patch_09-25_malformed.patch --]
[-- Type: text/plain, Size: 1299 bytes --]

diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c
index 3331669..f0660dd 100644
--- a/net/netrom/af_netrom.c
+++ b/net/netrom/af_netrom.c
@@ -885,6 +885,11 @@ int nr_rx_frame(struct sk_buff *skb, struct net_device *dev)
 	 *	skb->data points to the netrom frame start
 	 */
 
+	if (skb_linearize(skb))
+		return 0;
+	if (skb->len < NR_NETWORK_LEN + NR_TRANSPORT_LEN)
+		return 0;
+
 	src  = (ax25_address *)(skb->data + 0);
 	dest = (ax25_address *)(skb->data + 7);
 
@@ -927,6 +932,11 @@ int nr_rx_frame(struct sk_buff *skb, struct net_device *dev)
 	}
 
 	if (sk != NULL) {
+		if (frametype == NR_CONNACK &&
+		    skb->len < NR_NETWORK_LEN + NR_TRANSPORT_LEN + 1) {
+			sock_put(sk);
+			return 0;
+		}
 		bh_lock_sock(sk);
 		skb_reset_transport_header(skb);
 
@@ -961,10 +971,14 @@ int nr_rx_frame(struct sk_buff *skb, struct net_device *dev)
 		return 0;
 	}
 
-	sk = nr_find_listener(dest);
+	/* Need window (byte 20) and user address (bytes 21-27) */
+	if (skb->len < NR_NETWORK_LEN + NR_TRANSPORT_LEN + 1 + AX25_ADDR_LEN)
+		return 0;
 
 	user = (ax25_address *)(skb->data + 21);
 
+	sk = nr_find_listener(dest);
+
 	if (sk == NULL || sk_acceptq_is_full(sk) ||
 	    (make = nr_make_new(sk)) == NULL) {
 		nr_transmit_refusal(skb, 0);