GHSA-96HG-7P79-GGX2
Vulnerability from github – Published: 2026-05-28 12:30 – Updated: 2026-05-28 12:30In the Linux kernel, the following vulnerability has been resolved:
xfrm: defensively unhash xfrm_state lists in __xfrm_state_delete
KASAN reproduces a slab-use-after-free in __xfrm_state_delete()'s hlist_del_rcu calls under syzkaller load on linux-6.12.y stable (reproduced on 6.12.47, also reachable via the same code path on torvalds/master and on the ipsec tree). Nine unique signatures cluster in the xfrm_state lifecycle, the load-bearing one being:
BUG: KASAN: slab-use-after-free in __hlist_del include/linux/list.h:990 [inline] BUG: KASAN: slab-use-after-free in hlist_del_rcu include/linux/rculist.h:516 [inline] BUG: KASAN: slab-use-after-free in __xfrm_state_delete net/xfrm/xfrm_state.c Write of size 8 at addr ffff8881198bcb70 by task kworker/u8:9/435
Workqueue: netns cleanup_net Call Trace: __hlist_del / hlist_del_rcu __xfrm_state_delete xfrm_state_delete xfrm_state_flush xfrm_state_fini ops_exit_list cleanup_net
The other observed signatures hit the same slab object from __xfrm_state_lookup, xfrm_alloc_spi, __xfrm_state_insert and an OOB write variant of __xfrm_state_delete, all on the byseq/byspi hash chains.
__xfrm_state_delete() guards its byseq and byspi unhashes with value-based predicates:
if (x->km.seq)
hlist_del_rcu(&x->byseq);
if (x->id.spi)
hlist_del_rcu(&x->byspi);
while everywhere else in the file (e.g. state_cache, state_cache_input) the safer hlist_unhashed() check is used. xfrm_alloc_spi() sets x->id.spi = newspi inside xfrm_state_lock and then immediately inserts into byspi, but a path that observes x->id.spi != 0 outside of xfrm_state_lock can still skip-or-hit the byspi unhash inconsistently with whether x is actually on the list. The same holds for x->km.seq versus byseq, and the bydst/bysrc unhashes have no predicate at all, so a second __xfrm_state_delete() on the same object writes through LIST_POISON pprev.
The defensive change here:
- Use hlist_del_init_rcu() instead of hlist_del_rcu() on bydst, bysrc, byseq and byspi so a second deletion is a no-op rather than a write through LIST_POISON pprev. The byseq/byspi nodes are already initialised in xfrm_state_alloc().
- Test hlist_unhashed() rather than the value predicate for byseq/byspi, so the unhash decision tracks list state rather than mutable scalar fields.
Empirical verification: applied this patch on top of v6.12.47, rebuilt, and re-ran the same syzkaller harness for 1h16m on a previously-crashy configuration that produced ~100 hits each of slab-use-after-free Read in xfrm_alloc_spi / Read in __xfrm_state_lookup / Write in __xfrm_state_delete. After the patch, 7.1M execs across 32 VMs at ~1550 exec/sec produced zero xfrm_state UAF/OOB hits. /proc/slabinfo confirms the xfrm_state slab is actively allocated and freed during the run (~143 KiB resident), so the fuzzer is still exercising those code paths -- they just no longer crash.
Reproduction:
- Linux 6.12.47 x86_64 + KASAN_GENERIC + KASAN_INLINE + KCOV
- syzkaller @ 746545b8b1e4c3a128db8652b340d3df90ce61db
- 32 QEMU/KVM VMs x 2 vCPU on AWS c5.metal bare metal
- 9 unique signatures collected in ~9h, all within xfrm_state lifecycle
{
"affected": [],
"aliases": [
"CVE-2026-46116"
],
"database_specific": {
"cwe_ids": [],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-05-28T10:16:27Z",
"severity": null
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\nxfrm: defensively unhash xfrm_state lists in __xfrm_state_delete\n\nKASAN reproduces a slab-use-after-free in __xfrm_state_delete()\u0027s\nhlist_del_rcu calls under syzkaller load on linux-6.12.y stable\n(reproduced on 6.12.47, also reachable via the same code path on\ntorvalds/master and on the ipsec tree). Nine unique signatures cluster\nin the xfrm_state lifecycle, the load-bearing one being:\n\n BUG: KASAN: slab-use-after-free in __hlist_del include/linux/list.h:990 [inline]\n BUG: KASAN: slab-use-after-free in hlist_del_rcu include/linux/rculist.h:516 [inline]\n BUG: KASAN: slab-use-after-free in __xfrm_state_delete net/xfrm/xfrm_state.c\n Write of size 8 at addr ffff8881198bcb70 by task kworker/u8:9/435\n\n Workqueue: netns cleanup_net\n Call Trace:\n __hlist_del / hlist_del_rcu\n __xfrm_state_delete\n xfrm_state_delete\n xfrm_state_flush\n xfrm_state_fini\n ops_exit_list\n cleanup_net\n\nThe other observed signatures hit the same slab object from\n__xfrm_state_lookup, xfrm_alloc_spi, __xfrm_state_insert and an OOB\nwrite variant of __xfrm_state_delete, all on the byseq/byspi\nhash chains.\n\n__xfrm_state_delete() guards its byseq and byspi unhashes with\nvalue-based predicates:\n\n\tif (x-\u003ekm.seq)\n\t\thlist_del_rcu(\u0026x-\u003ebyseq);\n\tif (x-\u003eid.spi)\n\t\thlist_del_rcu(\u0026x-\u003ebyspi);\n\nwhile everywhere else in the file (e.g. state_cache, state_cache_input)\nthe safer hlist_unhashed() check is used. xfrm_alloc_spi() sets\nx-\u003eid.spi = newspi inside xfrm_state_lock and then immediately inserts\ninto byspi, but a path that observes x-\u003eid.spi != 0 outside of\nxfrm_state_lock can still skip-or-hit the byspi unhash inconsistently\nwith whether x is actually on the list. The same holds for x-\u003ekm.seq\nversus byseq, and the bydst/bysrc unhashes have no predicate at all,\nso a second __xfrm_state_delete() on the same object writes through\nLIST_POISON pprev.\n\nThe defensive change here:\n\n - Use hlist_del_init_rcu() instead of hlist_del_rcu() on bydst,\n bysrc, byseq and byspi so a second deletion is a no-op rather\n than a write through LIST_POISON pprev. The byseq/byspi nodes\n are already initialised in xfrm_state_alloc().\n - Test hlist_unhashed() rather than the value predicate for\n byseq/byspi, so the unhash decision tracks list state rather than\n mutable scalar fields.\n\nEmpirical verification: applied this patch on top of v6.12.47, rebuilt,\nand re-ran the same syzkaller harness for 1h16m on a previously-crashy\nconfiguration that produced ~100 hits each of slab-use-after-free\nRead in xfrm_alloc_spi / Read in __xfrm_state_lookup / Write in\n__xfrm_state_delete. After the patch, 7.1M execs across 32 VMs at\n~1550 exec/sec produced zero xfrm_state UAF/OOB hits. /proc/slabinfo\nconfirms the xfrm_state slab is actively allocated and freed during\nthe run (~143 KiB resident), so the fuzzer is still exercising those\ncode paths -- they just no longer crash.\n\nReproduction:\n\n - Linux 6.12.47 x86_64 + KASAN_GENERIC + KASAN_INLINE + KCOV\n - syzkaller @ 746545b8b1e4c3a128db8652b340d3df90ce61db\n - 32 QEMU/KVM VMs x 2 vCPU on AWS c5.metal bare metal\n - 9 unique signatures collected in ~9h, all within xfrm_state\n lifecycle",
"id": "GHSA-96hg-7p79-ggx2",
"modified": "2026-05-28T12:30:29Z",
"published": "2026-05-28T12:30:29Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-46116"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/14acf9652e5690de3c7486c6db5fb8dafd0a32a3"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/26edb0a3c99f9d958c212be68b21f1221614dcf0"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/4980162de555cb838f1a189ce7d2cbf5d2e7b050"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/a2e2d08fb070fab4947447171f1c4e3ca5a188e5"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/b4a53add2fa8f1b5aa17d4c5686c320785fab182"
}
],
"schema_version": "1.4.0",
"severity": []
}
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.