GHSA-VW9X-RC5G-H57C
Vulnerability from github – Published: 2026-08-15 06:32 – Updated: 2026-08-17 06:33In the Linux kernel, the following vulnerability has been resolved:
tcp: defer md5sig_info kfree past RCU grace period in tcp_connect
The md5+ao reconciliation in tcp_connect() (net/ipv4/tcp_output.c) has two symmetric branches:
if (needs_md5) {
tcp_ao_destroy_sock(sk, false);
} else if (needs_ao) {
tcp_clear_md5_list(sk);
kfree(rcu_replace_pointer(tp->md5sig_info, NULL, ...));
}
Both branches free a per-socket auth-info object while the socket is in TCP_SYN_SENT and is already on the inet ehash (inserted by inet_hash_connect() in tcp_v4_connect()). Both branches are reachable by softirq RX-path readers that load the corresponding info pointer via implicit RCU before bh_lock_sock_nested() is taken.
The needs_md5 branch is fixed in the prior patch by re-introducing the call_rcu() free in tcp_ao_destroy_sock(): the equivalent per-key loop runs inside tcp_ao_info_free_rcu(), the RCU callback, so by the time it frees each tcp_ao_key all softirq readers that captured the container have already completed rcu_read_unlock().
The needs_ao branch is not symmetric in the same way. The container free can be deferred via kfree_rcu(md5sig, rcu) -- struct tcp_md5sig_info already has the required rcu member (include/net/tcp.h:1999-2002), and the rest of the tree already does this in the tcp_md5sig_info_add() rollback paths (net/ipv4/tcp_ipv4.c:1410, 1436). But the per-key teardown is done by tcp_clear_md5_list() in process context BEFORE the container's RCU grace period: it walks &md5sig->head and frees each tcp_md5sig_key with bare hlist_del + kfree. A concurrent softirq reader in __tcp_md5_do_lookup() / __tcp_md5_do_lookup_exact() (tcp_ipv4.c:1253, 1298) walks the same list via hlist_for_each_entry_rcu() and races with that bare kfree on the keys themselves -- a per-key slab use-after-free of the same class as the TCP-AO bug, on the same race window.
Fix this in two halves:
-
Convert the bare kfree() in tcp_connect() to kfree_rcu() so the md5sig_info container joins the rest of the md5sig lifecycle. The local-variable lift is mechanical and required because kfree_rcu() is a macro that expects an lvalue.
-
Make tcp_clear_md5_list() RCU-safe by replacing hlist_del + kfree(key) with hlist_del_rcu + kfree_rcu(key, rcu). struct tcp_md5sig_key already carries the rcu member (include/net/tcp.h:1995) and tcp_md5_do_del() (net/ipv4/tcp_ipv4.c:1456) already uses kfree_rcu, so this restores the lifecycle invariant the rest of the file follows rather than introducing a one-off.
The other caller of tcp_clear_md5_list() is tcp_md5_destruct_sock() (net/ipv4/tcp.c:412), which runs from the sock destructor when the socket is already unhashed and unreachable; the extra grace period there is unnecessary but harmless. Making the helper unconditionally RCU-safe is the cleaner contract.
The needs_ao branch is not reachable by the userns reproducer used to demonstrate the AO-side splat (the repro installs both keys but ends up in the needs_md5 branch because the connect peer matches the MD5 key, not the AO key); however the symmetric race exists and a maintainer touching this code should not have to think about which branch escapes RCU and which one does not.
[also credits to Qihang, who found that this races with tcp-diag]
{
"affected": [],
"aliases": [
"CVE-2026-72139"
],
"database_specific": {
"cwe_ids": [],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-08-15T06:21:31Z",
"severity": "CRITICAL"
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\ntcp: defer md5sig_info kfree past RCU grace period in tcp_connect\n\nThe md5+ao reconciliation in tcp_connect() (net/ipv4/tcp_output.c)\nhas two symmetric branches:\n\n\tif (needs_md5) {\n\t\ttcp_ao_destroy_sock(sk, false);\n\t} else if (needs_ao) {\n\t\ttcp_clear_md5_list(sk);\n\t\tkfree(rcu_replace_pointer(tp-\u003emd5sig_info, NULL, ...));\n\t}\n\nBoth branches free a per-socket auth-info object while the socket is\nin TCP_SYN_SENT and is already on the inet ehash (inserted by\ninet_hash_connect() in tcp_v4_connect()). Both branches are reachable\nby softirq RX-path readers that load the corresponding info pointer\nvia implicit RCU before bh_lock_sock_nested() is taken.\n\nThe needs_md5 branch is fixed in the prior patch by re-introducing\nthe call_rcu() free in tcp_ao_destroy_sock(): the equivalent per-key\nloop runs inside tcp_ao_info_free_rcu(), the RCU callback, so by the\ntime it frees each tcp_ao_key all softirq readers that captured the\ncontainer have already completed rcu_read_unlock().\n\nThe needs_ao branch is not symmetric in the same way. The container\nfree can be deferred via kfree_rcu(md5sig, rcu) -- struct\ntcp_md5sig_info already has the required rcu member\n(include/net/tcp.h:1999-2002), and the rest of the tree already does\nthis in the tcp_md5sig_info_add() rollback paths\n(net/ipv4/tcp_ipv4.c:1410, 1436). But the per-key teardown is done\nby tcp_clear_md5_list() in process context BEFORE the container\u0027s\nRCU grace period: it walks \u0026md5sig-\u003ehead and frees each\ntcp_md5sig_key with bare hlist_del + kfree. A concurrent softirq\nreader in __tcp_md5_do_lookup() / __tcp_md5_do_lookup_exact()\n(tcp_ipv4.c:1253, 1298) walks the same list via\nhlist_for_each_entry_rcu() and races with that bare kfree on the\nkeys themselves -- a per-key slab use-after-free of the same class\nas the TCP-AO bug, on the same race window.\n\nFix this in two halves:\n\n 1. Convert the bare kfree() in tcp_connect() to kfree_rcu() so the\n md5sig_info container joins the rest of the md5sig lifecycle.\n The local-variable lift is mechanical and required because\n kfree_rcu() is a macro that expects an lvalue.\n\n 2. Make tcp_clear_md5_list() RCU-safe by replacing hlist_del +\n kfree(key) with hlist_del_rcu + kfree_rcu(key, rcu). struct\n tcp_md5sig_key already carries the rcu member\n (include/net/tcp.h:1995) and tcp_md5_do_del()\n (net/ipv4/tcp_ipv4.c:1456) already uses kfree_rcu, so this\n restores the lifecycle invariant the rest of the file follows\n rather than introducing a one-off.\n\nThe other caller of tcp_clear_md5_list() is tcp_md5_destruct_sock()\n(net/ipv4/tcp.c:412), which runs from the sock destructor when the\nsocket is already unhashed and unreachable; the extra grace period\nthere is unnecessary but harmless. Making the helper unconditionally\nRCU-safe is the cleaner contract.\n\nThe needs_ao branch is not reachable by the userns reproducer used\nto demonstrate the AO-side splat (the repro installs both keys but\nends up in the needs_md5 branch because the connect peer matches\nthe MD5 key, not the AO key); however the symmetric race exists\nand a maintainer touching this code should not have to think about\nwhich branch escapes RCU and which one does not.\n\n[also credits to Qihang, who found that this races with tcp-diag]",
"id": "GHSA-vw9x-rc5g-h57c",
"modified": "2026-08-17T06:33:12Z",
"published": "2026-08-15T06:32:13Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-72139"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/33a1bee413628378fd036a4f2b17ba86b0bc560c"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/b74cd55038905d5e74c1de109ab78a30b2ea0e1f"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/da48b9bf1eb95a9cfd09d615ca58cfc2b03de369"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
"type": "CVSS_V3"
}
]
}
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.