Netdev List
 help / color / mirror / Atom feed
From: Henry Martin <bsdhenrymartin@gmail.com>
To: lucien.xin@gmail.com, marcelo.leitner@gmail.com,
	davem@davemloft.net, kuba@kernel.org, pabeni@redhat.com,
	horms@kernel.org, edumazet@kernel.org
Cc: linux-sctp@vger.kernel.org, netdev@vger.kernel.org,
	Henry Martin <bsdhenrymartin@gmail.com>,
	stable@vger.kernel.org
Subject: [PATCH v2] sctp: fix skb_over_panic when adding addresses with a pending delete of another family
Date: Wed, 11 Nov 2026 16:23:13 +0800	[thread overview]
Message-ID: <20261111082313.1606366-1-bsdhenrymartin@gmail.com> (raw)
In-Reply-To: <CADvbK_daN6zZNPckkGh=r70ooNwTz59VViJy5jPp1w18SDPrzw@mail.gmail.com>

sctp_make_asconf_update_ip() accounts for a pending DEL_IP
parameter by reusing the length of the address being added. The
write-back side instead serializes the pending address itself,
with its own address family. If the pending delete is IPv6
(20-byte parameter) while the added addresses are IPv4 (8-byte),
the chunk is allocated 12 bytes short and sctp_addto_chunk()
writes past the end of the skb. E.g. on mainline (05441d697f7f):

    skbuff: skb_over_panic: text:ffffffff84859c77 len:392 put:20
        tail:0x188 end:0x180 dev:<NULL>
    kernel BUG at net/core/skbuff.c:215!
    CPU: 0 UID: 1000 PID: 379 Comm: poc_sctp
    Call Trace:
     skb_panic
     skb_put
     sctp_addto_chunk
     sctp_make_asconf_update_ip
     sctp_send_asconf_add_ip
     sctp_setsockopt_bindx
     __x64_sys_setsockopt

Requires addip enabled on the endpoint (net.sctp.addip_enable=1).

Account for the pending delete with its own address family, as the
write-back side already does.

This issue was discovered by Tencent CodeBuddy Security.

Cc: stable@vger.kernel.org
Fixes: 8a07eb0a50ae ("sctp: Add ASCONF operation on the single-homed host")
Signed-off-by: Henry Martin <bsdhenrymartin@gmail.com>
---
v2: move the pending-delete accounting out of the length loop and
    drop the del_pickup flag, which no longer guards anything.

 net/sctp/sm_make_chunk.c | 16 ++++++++++------
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index 90de272842512..b8e4db81636e2 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -2870,7 +2870,6 @@ struct sctp_chunk *sctp_make_asconf_update_ip(struct sctp_association *asoc,
 	int addr_param_len = 0;
 	union sctp_addr *addr;
 	int totallen = 0, i;
-	int del_pickup = 0;
 	struct sctp_af *af;
 	void *addr_buf;
 
@@ -2885,16 +2884,21 @@ struct sctp_chunk *sctp_make_asconf_update_ip(struct sctp_association *asoc,
 		totallen += addr_param_len;
 
 		addr_buf += af->sockaddr_len;
-		if (asoc->asconf_addr_del_pending && !del_pickup) {
-			/* reuse the parameter length from the same scope one */
-			totallen += paramlen;
-			totallen += addr_param_len;
-			del_pickup = 1;
-
-			pr_debug("%s: picked same-scope del_pending addr, "
-				 "totallen for all addresses is %d\n",
-				 __func__, totallen);
-		}
+	}
+
+	if (asoc->asconf_addr_del_pending) {
+		/* The DEL_IP parameter is serialized from the pending address
+		 * itself below; account for its own address family, which may
+		 * differ from the added ones.
+		 */
+		addr = asoc->asconf_addr_del_pending;
+		af = sctp_get_af_specific(addr->v4.sin_family);
+		totallen += paramlen;
+		totallen += af->to_addr_param(addr, &addr_param);
+
+		pr_debug("%s: picked del_pending addr, "
+			 "totallen for all addresses is %d\n",
+			 __func__, totallen);
 	}
 
 	/* Create an asconf chunk with the required length. */
@@ -2917,7 +2921,7 @@ struct sctp_chunk *sctp_make_asconf_update_ip(struct sctp_association *asoc,
 
 		addr_buf += af->sockaddr_len;
 	}
-	if (flags == SCTP_PARAM_ADD_IP && del_pickup) {
+	if (flags == SCTP_PARAM_ADD_IP && asoc->asconf_addr_del_pending) {
 		addr = asoc->asconf_addr_del_pending;
 		af = sctp_get_af_specific(addr->v4.sin_family);
 		addr_param_len = af->to_addr_param(addr, &addr_param);

      reply	other threads:[~2026-11-11  8:23 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-06 18:52 [PATCH] sctp: fix skb_over_panic when adding addresses with a pending delete of another family Henry Martin
2026-11-10 15:47 ` Xin Long
2026-11-11  8:23   ` Henry Martin [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20261111082313.1606366-1-bsdhenrymartin@gmail.com \
    --to=bsdhenrymartin@gmail.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@kernel.org \
    --cc=horms@kernel.org \
    --cc=kuba@kernel.org \
    --cc=linux-sctp@vger.kernel.org \
    --cc=lucien.xin@gmail.com \
    --cc=marcelo.leitner@gmail.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=stable@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox