Netdev List
 help / color / mirror / Atom feed
* [PATCH] nfc: llcp: fix integer underflow and missing bounds checks in TLV parsing
@ 2026-05-25 20:21 Doruk Tan Ozturk
  0 siblings, 0 replies; 4+ messages in thread
From: Doruk Tan Ozturk @ 2026-05-25 20:21 UTC (permalink / raw)
  To: linux-nfc; +Cc: krzysztof.kozlowski, security, netdev, Doruk Tan Ozturk

Multiple out-of-bounds read vulnerabilities exist in the NFC LLCP TLV
parsers:

1. In nfc_llcp_recv_snl(), when an SDREQ TLV has length == 0,
   service_name_len = length - 1 underflows to SIZE_MAX (size_t is
   unsigned). The subsequent strncmp() and nfc_llcp_sock_from_sn()
   calls then read unbounded kernel heap memory.

2. All LLCP TLV parsing loops (nfc_llcp_recv_snl, nfc_llcp_connect_sn,
   nfc_llcp_parse_gb_tlv, nfc_llcp_parse_connection_tlv) read tlv[0]
   and tlv[1] without first verifying that at least 2 bytes remain in
   the buffer.

A nearby malicious NFC device can trigger these without authentication --
LLCP link activation happens automatically after NFC-DEP.

Fix by adding a minimum length check before the subtraction in the
SDREQ case, and adding bounds validation at the top of each TLV loop
iteration.

Found by pwnkit (https://github.com/0sec-labs/pwnkit), an automated
kernel source review tool by 0sec (https://0sec.ai).

Signed-off-by: Doruk Tan Ozturk <doruk@0sec.ai>
---
 net/nfc/llcp_core.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/net/nfc/llcp_core.c b/net/nfc/llcp_core.c
index XXXXXXX..YYYYYYY 100644
--- a/net/nfc/llcp_core.c
+++ b/net/nfc/llcp_core.c
@@ -1300,6 +1300,9 @@ static void nfc_llcp_recv_snl(struct nfc_llcp_local *local,
 
 	while (offset < tlv_len) {
+		if (offset + 2 > tlv_len)
+			break;
+
 		type = tlv[0];
 		length = tlv[1];
 
@@ -1307,6 +1310,9 @@ static void nfc_llcp_recv_snl(struct nfc_llcp_local *local,
 		switch (type) {
 		case LLCP_TLV_SDREQ:
+			if (length < 1)
+				break;
+
 			tid = tlv[2];
 			service_name = (char *) &tlv[3];
 			service_name_len = length - 1;

^ permalink raw reply	[flat|nested] 4+ messages in thread

* [PATCH] nfc: llcp: fix integer underflow and missing bounds checks in TLV parsing
@ 2026-05-25 20:24 Doruk Tan Ozturk
  2026-06-02 11:12 ` David Heidelberg
  2026-06-02 12:58 ` David Laight
  0 siblings, 2 replies; 4+ messages in thread
From: Doruk Tan Ozturk @ 2026-05-25 20:24 UTC (permalink / raw)
  To: oe-linux-nfc; +Cc: david+nfc, security, netdev, Doruk Tan Ozturk

Multiple out-of-bounds read vulnerabilities exist in the NFC LLCP TLV
parsers:

1. In nfc_llcp_recv_snl(), when an SDREQ TLV has length == 0,
   service_name_len = length - 1 underflows to SIZE_MAX (size_t is
   unsigned). The subsequent strncmp() and nfc_llcp_sock_from_sn()
   calls then read unbounded kernel heap memory.

2. All LLCP TLV parsing loops (nfc_llcp_recv_snl, nfc_llcp_connect_sn,
   nfc_llcp_parse_gb_tlv, nfc_llcp_parse_connection_tlv) read tlv[0]
   and tlv[1] without first verifying that at least 2 bytes remain in
   the buffer.

A nearby malicious NFC device can trigger these without authentication --
LLCP link activation happens automatically after NFC-DEP.

Fix by adding a minimum length check before the subtraction in the
SDREQ case, and adding bounds validation at the top of each TLV loop
iteration.

Found by pwnkit (https://github.com/0sec-labs/pwnkit), an automated
kernel source review tool by 0sec (https://0sec.ai).

Signed-off-by: Doruk Tan Ozturk <doruk@0sec.ai>
---
 net/nfc/llcp_core.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/net/nfc/llcp_core.c b/net/nfc/llcp_core.c
index XXXXXXX..YYYYYYY 100644
--- a/net/nfc/llcp_core.c
+++ b/net/nfc/llcp_core.c
@@ -1300,6 +1300,9 @@ static void nfc_llcp_recv_snl(struct nfc_llcp_local *local,
 
 	while (offset < tlv_len) {
+		if (offset + 2 > tlv_len)
+			break;
+
 		type = tlv[0];
 		length = tlv[1];
 
@@ -1307,6 +1310,9 @@ static void nfc_llcp_recv_snl(struct nfc_llcp_local *local,
 		switch (type) {
 		case LLCP_TLV_SDREQ:
+			if (length < 1)
+				break;
+
 			tid = tlv[2];
 			service_name = (char *) &tlv[3];
 			service_name_len = length - 1;

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] nfc: llcp: fix integer underflow and missing bounds checks in TLV parsing
  2026-05-25 20:24 [PATCH] nfc: llcp: fix integer underflow and missing bounds checks in TLV parsing Doruk Tan Ozturk
@ 2026-06-02 11:12 ` David Heidelberg
  2026-06-02 12:58 ` David Laight
  1 sibling, 0 replies; 4+ messages in thread
From: David Heidelberg @ 2026-06-02 11:12 UTC (permalink / raw)
  To: Doruk Tan Ozturk; +Cc: security, netdev, oe-linux-nfc

On 25/05/2026 22:24, Doruk Tan Ozturk wrote:
> Multiple out-of-bounds read vulnerabilities exist in the NFC LLCP TLV
> parsers:
> 
> 1. In nfc_llcp_recv_snl(), when an SDREQ TLV has length == 0,
>     service_name_len = length - 1 underflows to SIZE_MAX (size_t is
>     unsigned). The subsequent strncmp() and nfc_llcp_sock_from_sn()
>     calls then read unbounded kernel heap memory.
> 
> 2. All LLCP TLV parsing loops (nfc_llcp_recv_snl, nfc_llcp_connect_sn,
>     nfc_llcp_parse_gb_tlv, nfc_llcp_parse_connection_tlv) read tlv[0]
>     and tlv[1] without first verifying that at least 2 bytes remain in
>     the buffer.
> 
> A nearby malicious NFC device can trigger these without authentication --
> LLCP link activation happens automatically after NFC-DEP.
> 
> Fix by adding a minimum length check before the subtraction in the
> SDREQ case, and adding bounds validation at the top of each TLV loop
> iteration.
> 
> Found by pwnkit (https://github.com/0sec-labs/pwnkit), an automated
> kernel source review tool by 0sec (https://0sec.ai).
> 
> Signed-off-by: Doruk Tan Ozturk <doruk@0sec.ai>
> ---
>   net/nfc/llcp_core.c | 6 ++++++
>   1 file changed, 6 insertions(+)
> 
> diff --git a/net/nfc/llcp_core.c b/net/nfc/llcp_core.c
> index XXXXXXX..YYYYYYY 100644

Hello Doruk,

thank you for your patch.

The patch won't apply cleanly, could you please resent patch based against 
recent https://codeberg.org/linux-nfc/linux/commits/branch/for-next ?


> --- a/net/nfc/llcp_core.c
> +++ b/net/nfc/llcp_core.c
> @@ -1300,6 +1300,9 @@ static void nfc_llcp_recv_snl(struct nfc_llcp_local *local,
>   
>   	while (offset < tlv_len) {
> +		if (offset + 2 > tlv_len)

I think it would be more descriptive use define, thus

(offset + LLCP_HEADER_SIZE > tlv_len)

David


> +			break;
> +
>   		type = tlv[0];
>   		length = tlv[1];
>   
> @@ -1307,6 +1310,9 @@ static void nfc_llcp_recv_snl(struct nfc_llcp_local *local,
>   		switch (type) {
>   		case LLCP_TLV_SDREQ:
> +			if (length < 1)
> +				break;
> +
>   			tid = tlv[2];
>   			service_name = (char *) &tlv[3];
>   			service_name_len = length - 1;

-- 
David Heidelberg


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] nfc: llcp: fix integer underflow and missing bounds checks in TLV parsing
  2026-05-25 20:24 [PATCH] nfc: llcp: fix integer underflow and missing bounds checks in TLV parsing Doruk Tan Ozturk
  2026-06-02 11:12 ` David Heidelberg
@ 2026-06-02 12:58 ` David Laight
  1 sibling, 0 replies; 4+ messages in thread
From: David Laight @ 2026-06-02 12:58 UTC (permalink / raw)
  To: Doruk Tan Ozturk; +Cc: oe-linux-nfc, david+nfc, security, netdev

On Mon, 25 May 2026 22:24:27 +0200
Doruk Tan Ozturk <doruk@0sec.ai> wrote:

> Multiple out-of-bounds read vulnerabilities exist in the NFC LLCP TLV
> parsers:
> 
> 1. In nfc_llcp_recv_snl(), when an SDREQ TLV has length == 0,
>    service_name_len = length - 1 underflows to SIZE_MAX (size_t is
>    unsigned). The subsequent strncmp() and nfc_llcp_sock_from_sn()
>    calls then read unbounded kernel heap memory.
> 
> 2. All LLCP TLV parsing loops (nfc_llcp_recv_snl, nfc_llcp_connect_sn,
>    nfc_llcp_parse_gb_tlv, nfc_llcp_parse_connection_tlv) read tlv[0]
>    and tlv[1] without first verifying that at least 2 bytes remain in
>    the buffer.
> 
> A nearby malicious NFC device can trigger these without authentication --
> LLCP link activation happens automatically after NFC-DEP.
> 
> Fix by adding a minimum length check before the subtraction in the
> SDREQ case, and adding bounds validation at the top of each TLV loop
> iteration.

Why not fix it properly so that it doesn't read beyond the end of the skb.
1) Change offset and tlv_len to be 32bit - no point trying to do 16bit mathx.
2) Worry about non-linear skb - perhaps just process skb_header_len() bytes.
3) Allow for very short frames.
4) Don't read beyond the end of the skb data.

Something like:
	tlv = skb->data + LLCP_HEADER_SIZE;
	tlv_end = skb_tail_pointer(skb);
	for (; tlv + 2 < tlv_end; tlv += 2 + length) {
		type = tlv[0];
		length = tlv[1];
		if (tlv + 2 + length > tlv_end)
			break;

Then as well as the check for length >= 1 in SDREQ (even 1 doesn't make sense),
the SDRES path is missing a check for length >= 2.

-- David

> 
> Found by pwnkit (https://github.com/0sec-labs/pwnkit), an automated
> kernel source review tool by 0sec (https://0sec.ai).
> 
> Signed-off-by: Doruk Tan Ozturk <doruk@0sec.ai>
> ---
>  net/nfc/llcp_core.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/net/nfc/llcp_core.c b/net/nfc/llcp_core.c
> index XXXXXXX..YYYYYYY 100644
> --- a/net/nfc/llcp_core.c
> +++ b/net/nfc/llcp_core.c
> @@ -1300,6 +1300,9 @@ static void nfc_llcp_recv_snl(struct nfc_llcp_local *local,
>  
>  	while (offset < tlv_len) {
> +		if (offset + 2 > tlv_len)
> +			break;
> +
>  		type = tlv[0];
>  		length = tlv[1];
>  
> @@ -1307,6 +1310,9 @@ static void nfc_llcp_recv_snl(struct nfc_llcp_local *local,
>  		switch (type) {
>  		case LLCP_TLV_SDREQ:
> +			if (length < 1)
> +				break;
> +
>  			tid = tlv[2];
>  			service_name = (char *) &tlv[3];
>  			service_name_len = length - 1;
> 


^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2026-06-02 12:58 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-05-25 20:24 [PATCH] nfc: llcp: fix integer underflow and missing bounds checks in TLV parsing Doruk Tan Ozturk
2026-06-02 11:12 ` David Heidelberg
2026-06-02 12:58 ` David Laight
  -- strict thread matches above, loose matches on Subject: below --
2026-05-25 20:21 Doruk Tan Ozturk

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox