icmpv6_filter: correct minimum ICMPv6 message size
Subsystems (2)
cover icmpv6_filter: correct minimum ICMPv6 message size
These two patches correct the minimum ICMPv6 message size enforced by net/ipv6/raw.c:icmpv6_filter
The first patch corrects a type error. Because of the error, ICMPv6 raw sockets on 32 bit systems accepted ICMPv6 messages as small as 4 bytes, while 64 bit systems required at least 8 bytes.
The second patch reduces the amount of data we require from eight (i.e., the ICMPv6 header plus four bytes of message body) to four bytes. This is needed for protocols like RPL (RFC 6550) that use ICMPv6 messages with bodies smaller than four bytes.
Note that applications that assume that the kernel will not pass such short ICMPv6 messages on raw sockets may misbehave on 64 bit systems after applying these patches. However, even if such applications exist, they would already have that vulnerability on 32 bit systems.
- Werner
Werner Almesberger (2):
icmpv6_filter: fix "_hdr" incorrectly being a pointer
icmpv6_filter: allow ICMPv6 messages with bodies < 4 bytes
net/ipv6/raw.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)
Signature · 1 line
1.8.1.2
PATCH 1/2 icmpv6_filter: fix "_hdr" incorrectly being a pointer
"_hdr" should hold the ICMPv6 header while "hdr" is the pointer to it. This worked by accident.
--- net/ipv6/raw.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv6/raw.c +1 -1
@@ -108,7 +108,7 @@ found:*/ static int icmpv6_filter (const struct sock *sk,const struct sk_buff *skb) { -struct icmp6hdr *_hdr; +struct icmp6hdr _hdr;const struct icmp6hdr *hdr; hdr = skb_header_pointer(skb, skb_transport_offset(skb),
Signature · 1 line
1.8.1.2
Re: [PATCH net-next 1/2] icmpv6_filter: fix "_hdr" incorrectly being a pointer
From: Werner Almesberger <[email protected]>
Date: Fri, 2 Aug 2013 10:51:19 -0300
Quoted 4 lines
"
_hdr" should hold the ICMPv6 header while "hdr" is the pointer to it. This worked by accident.Signed-off-by Werner Almesberger <[email protected]>
Applied.
PATCH 2/2 icmpv6_filter: allow ICMPv6 messages with bodies < 4 bytes
In reply to Werner AlmesbergerBy using sizeof(_hdr), net/ipv6/raw.c:icmpv6_filter implicitly assumes that any valid ICMPv6 message is at least eight bytes long, i.e., that the message body is at least four bytes.
The DIS message of RPL (RFC 6550 section 6.2, from the 6LoWPAN world), has a minimum length of only six bytes, and is thus blocked by icmpv6_filter.
RFC 4443 seems to allow even a zero-sized body, making the minimum allowable message size four bytes.
--- net/ipv6/raw.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
net/ipv6/raw.c +6 -1
@@ -63,6 +63,8 @@#include <linux/seq_file.h> #include <linux/export.h> +#define ICMPV6_HDRLEN 4/* ICMPv6 header, RFC 4443 Section 2.1 */ +static struct raw_hashinfo raw_v6_hashinfo = { .lock =__RW_LOCK_UNLOCKED (raw_v6_hashinfo.lock), };
@@ -111,8 +113,11 @@ static int icmpv6_filter(const struct sock *sk, const struct sk_buff *skb)struct icmp6hdr _hdr;const struct icmp6hdr *hdr; +/* We require only the four bytes of the ICMPv6 header, not any +* additional bytes of message body in "struct icmp6hdr". +*/ hdr =skb_header_pointer (skb,skb_transport_offset (skb), -sizeof (_hdr), &_hdr); +ICMPV6_HDRLEN , &_hdr);if (hdr) {const __u32 *data = &raw6_sk (sk)->filter.data[0 ];unsigned int type = hdr->icmp6_type;
Signature · 1 line
1.8.1.2
Re: [PATCH net-next 2/2] icmpv6_filter: allow ICMPv6 messages with bodies < 4 bytes
From: Werner Almesberger <[email protected]>
Date: Fri, 2 Aug 2013 10:51:34 -0300
Quoted 4 lines
By using
sizeof(_hdr), net/ipv6/raw.c:icmpv6_filter implicitly assumes that any valid ICMPv6 message is at least eight bytes long, i.e., that the message body is at least four bytes.
Quoted 8 lines
The DIS message of RPL (RFC 6550 section 6.2, from the 6LoWPAN world), has a minimum length of only six bytes, and is thus blocked by
icmpv6_filter.RFC 4443 seems to allow even a zero-sized body, making the minimum allowable message size four bytes.
Signed-off-by Werner Almesberger <[email protected]>
Applied, thanks Werner.