* [PATCH] netfilter: nf_conntrack_irc: fix parse_dcc() off-by-one OOB read
@ 2026-05-19 21:23 Muhammad Bilal
2026-05-19 21:29 ` Pablo Neira Ayuso
0 siblings, 1 reply; 3+ messages in thread
From: Muhammad Bilal @ 2026-05-19 21:23 UTC (permalink / raw)
To: Pablo Neira Ayuso
Cc: Florian Westphal, Phil Sutter, netfilter-devel, coreteam,
linux-kernel, stable, Muhammad Bilal
parse_dcc() treats data_end as an inclusive end pointer, but its only
caller passes data_limit = ib_ptr + datalen, which points one past the
last valid byte.
The newline search loop iterates while tmp <= data_end, so when no
newline is present, *tmp is read at tmp == data_end, one byte beyond
the region filled by skb_header_pointer().
irc_buffer is kmalloc'd as MAX_SEARCH_SIZE + 1 bytes and datalen is
capped at MAX_SEARCH_SIZE, so the stray read does not fault. The byte
is uninitialized or stale; if it contains an ASCII digit, simple_strtoul
will consume it and produce a wrong DCC IP or port in the conntrack
expectation. The extra allocation byte is also a fragile guard: if the
cap or allocation size changes, this becomes a real out-of-bounds read.
Change the loop and its post-loop check to use strict less-than,
consistent with the caller's exclusive-end convention. Update the
function comment accordingly.
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Cc: stable@vger.kernel.org
Signed-off-by: Muhammad Bilal <meatuni001@gmail.com>
---
net/netfilter/nf_conntrack_irc.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/net/netfilter/nf_conntrack_irc.c b/net/netfilter/nf_conntrack_irc.c
index 522183b9a..9a7b8f622 100644
--- a/net/netfilter/nf_conntrack_irc.c
+++ b/net/netfilter/nf_conntrack_irc.c
@@ -59,7 +59,7 @@ static const char *const dccprotos[] = {
/* tries to get the ip_addr and port out of a dcc command
* return value: -1 on failure, 0 on success
* data pointer to first byte of DCC command data
- * data_end pointer to last byte of dcc command data
+ * data_end one past end of data
* ip returns parsed ip of dcc command
* port returns parsed port of dcc command
* ad_beg_p returns pointer to first byte of addr data
@@ -77,10 +77,10 @@ static int parse_dcc(char *data, const char *data_end, __be32 *ip,
/* Make sure we have a newline character within the packet boundaries
* because simple_strtoul parses until the first invalid character. */
- for (tmp = data; tmp <= data_end; tmp++)
+ for (tmp = data; tmp < data_end; tmp++)
if (*tmp == '\n')
break;
- if (tmp > data_end || *tmp != '\n')
+ if (tmp >= data_end || *tmp != '\n')
return -1;
*ad_beg_p = data;
--
2.54.0
^ permalink raw reply related [flat|nested] 3+ messages in thread* Re: [PATCH] netfilter: nf_conntrack_irc: fix parse_dcc() off-by-one OOB read
2026-05-19 21:23 [PATCH] netfilter: nf_conntrack_irc: fix parse_dcc() off-by-one OOB read Muhammad Bilal
@ 2026-05-19 21:29 ` Pablo Neira Ayuso
2026-05-19 21:43 ` Muhammad Bilal
0 siblings, 1 reply; 3+ messages in thread
From: Pablo Neira Ayuso @ 2026-05-19 21:29 UTC (permalink / raw)
To: Muhammad Bilal
Cc: Florian Westphal, Phil Sutter, netfilter-devel, coreteam,
linux-kernel, stable
On Tue, May 19, 2026 at 05:23:28PM -0400, Muhammad Bilal wrote:
> parse_dcc() treats data_end as an inclusive end pointer, but its only
> caller passes data_limit = ib_ptr + datalen, which points one past the
> last valid byte.
>
> The newline search loop iterates while tmp <= data_end, so when no
> newline is present, *tmp is read at tmp == data_end, one byte beyond
> the region filled by skb_header_pointer().
>
> irc_buffer is kmalloc'd as MAX_SEARCH_SIZE + 1 bytes and datalen is
> capped at MAX_SEARCH_SIZE, so the stray read does not fault.
No crash here.
> The byte is uninitialized or stale; if it contains an ASCII digit,
> simple_strtoul will consume it and produce a wrong DCC IP or port in
> the conntrack expectation. The extra allocation byte is also a
> fragile guard: if the cap or allocation size changes, this becomes a
> real out-of-bounds read.
Other helpers replaced simple_stroul() already which is probably the
way to go.
This is nf-next material.
> Change the loop and its post-loop check to use strict less-than,
> consistent with the caller's exclusive-end convention. Update the
> function comment accordingly.
>
> Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
> Cc: stable@vger.kernel.org
> Signed-off-by: Muhammad Bilal <meatuni001@gmail.com>
> ---
> net/netfilter/nf_conntrack_irc.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/net/netfilter/nf_conntrack_irc.c b/net/netfilter/nf_conntrack_irc.c
> index 522183b9a..9a7b8f622 100644
> --- a/net/netfilter/nf_conntrack_irc.c
> +++ b/net/netfilter/nf_conntrack_irc.c
> @@ -59,7 +59,7 @@ static const char *const dccprotos[] = {
> /* tries to get the ip_addr and port out of a dcc command
> * return value: -1 on failure, 0 on success
> * data pointer to first byte of DCC command data
> - * data_end pointer to last byte of dcc command data
> + * data_end one past end of data
> * ip returns parsed ip of dcc command
> * port returns parsed port of dcc command
> * ad_beg_p returns pointer to first byte of addr data
> @@ -77,10 +77,10 @@ static int parse_dcc(char *data, const char *data_end, __be32 *ip,
>
> /* Make sure we have a newline character within the packet boundaries
> * because simple_strtoul parses until the first invalid character. */
> - for (tmp = data; tmp <= data_end; tmp++)
> + for (tmp = data; tmp < data_end; tmp++)
> if (*tmp == '\n')
> break;
> - if (tmp > data_end || *tmp != '\n')
> + if (tmp >= data_end || *tmp != '\n')
> return -1;
>
> *ad_beg_p = data;
> --
> 2.54.0
>
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [PATCH] netfilter: nf_conntrack_irc: fix parse_dcc() off-by-one OOB read
2026-05-19 21:29 ` Pablo Neira Ayuso
@ 2026-05-19 21:43 ` Muhammad Bilal
0 siblings, 0 replies; 3+ messages in thread
From: Muhammad Bilal @ 2026-05-19 21:43 UTC (permalink / raw)
To: Pablo Neira Ayuso
Cc: Florian Westphal, Phil Sutter, netfilter-devel, coreteam,
linux-kernel, stable, Muhammad Bilal
On Tue, May 19, 2026 at 05:29:54PM +0200, Pablo Neira Ayuso wrote:
> Other helpers replaced simple_stroul() already which is probably the
> way to go.
I can send a follow-up patch replacing simple_strtoul() in parse_dcc()
once this lands in nf-next, if that works for you.
> This is nf-next material.
Understood.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2026-05-19 21:45 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-05-19 21:23 [PATCH] netfilter: nf_conntrack_irc: fix parse_dcc() off-by-one OOB read Muhammad Bilal
2026-05-19 21:29 ` Pablo Neira Ayuso
2026-05-19 21:43 ` Muhammad Bilal
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox