GHSA-F5R6-6W97-7R58
Vulnerability from github – Published: 2026-09-17 18:31 – Updated: 2026-09-17 18:31In the Linux kernel, the following vulnerability has been resolved:
smb: smbdirect: release pending child sockets outside the handler lock
smbdirect_socket_destroy() releases the listener's pending/ready child sockets while still holding the listener's handler lock, the &id_priv->handler_mutex taken via rdma_lock_handler(), not sc->listen.lock, and before the listener's own rdma_destroy_id(). That ordering has one real consequence and one cosmetic one.
The real one: smbdirect_socket_release() drops the child's last reference, which destroys the child's cm_id. Doing that before the listener's rdma_destroy_id() lets _cma_cancel_listens(), running from the listener's _destroy_id(), walk an already freed child id_priv, which KASAN catches as a slab-use-after-free during listener shutdown:
[ 4758.909130] BUG: KASAN: slab-use-after-free in __mutex_lock+0x1469/0x1560 [ 4758.911450] Read of size 1 at addr ffff88821c381db4 by task ksmbd.control/1652 [ 4758.913262] Call Trace: [ 4758.913267] [ 4758.913299] __mutex_lock+0x1469/0x1560 [ 4758.913408] _cma_cancel_listens+0x312/0x3b0 [ 4758.913413] _destroy_id+0x363/0xee0 [ 4758.913417] smbdirect_socket_destroy_sync+0x17d5/0x2440 [ 4758.913443] smbdirect_socket_release+0x124/0x230 [ 4758.913451] ksmbd_rdma_stop_listening+0x9f/0x190 [ 4758.913457] ksmbd_conn_transport_destroy+0x65/0x3c0 [ 4758.913463] kill_server_store+0x1fb/0x2b0 [ 4758.913501] kernfs_fop_write_iter+0x349/0x4d0 [ 4758.913507] vfs_write+0x5e7/0xc70 [ 4758.913528] ksys_write+0x12a/0x210 [ 4758.913541] do_syscall_64+0x135/0x460 [ 4758.913555] entry_SYSCALL_64_after_hwframe+0x77/0x7f
The cosmetic one: releasing a child recurses into smbdirect_socket_destroy(), which takes the child's own rdma_lock_handler() lock nested under the listener's. The listener's and the child's cm_id are always different instances, so this cannot deadlock for real; the CM core itself nests a new connection id's handler_mutex under the listening id's in cma_ib_req_handler(). But lockdep only sees one lock class, reports possible recursive locking, and then disables itself, hiding real locking bugs for the rest of the run:
[ 2424.579653] WARNING: possible recursive locking detected [ 2424.581180] 7.1.0-next-20260623+ #89 Not tainted [ 2424.582548] -------------------------------------------- [ 2424.584500] ksmbd.control/8854 is trying to acquire lock: [ 2424.586817] ffff888102303c20 (&id_priv->handler_mutex){+.+.}-{4:4}, at: smbdirect_socket_destroy_sync+0xc39/0x2440 [ 2424.590590] [ 2424.590590] but task is already holding lock: [ 2424.591601] ffff888102046c20 (&id_priv->handler_mutex){+.+.}-{4:4}, at: smbdirect_socket_destroy_sync+0xc39/0x2440 [ 2424.594178] [ 2424.594178] other info that might help us debug this: [ 2424.596634] Possible unsafe locking scenario: [ 2424.596634] [ 2424.598841] CPU0 [ 2424.599765] ---- [ 2424.600695] lock(&id_priv->handler_mutex); [ 2424.601836] lock(&id_priv->handler_mutex); [ 2424.602590] [ 2424.602590] *** DEADLOCK *** [ 2424.602590] [ 2424.604512] May be due to missing lock nesting notation
Splice the pending/ready children onto a local list under the listener's listen.lock, while the handler lock is held so a concurrent CM CONNECT_REQUEST cannot add more, but defer the actual smbdirect_socket_release() calls until after the listener's cm_id has been destroyed and its handler lock dropped. The children are independent sockets whose teardown needs neither the listener's handler lock nor its cm_id.
Found with ksmbdzzer [2], a KSMBD fuzzer that drives libFuzzer with a kcov-dataflow [1] coverage vector: it folds each instrumented comparison/argument's runtime operand value together with its PC (the default arm mixes them as pc⊕val) so that a new operand value at a known site counts as new coverage.
[1] https://lwn.net/Articles/1077606/ [2] https://github.com/yskzalloc/kcov-dataflow
{
"affected": [],
"aliases": [
"CVE-2026-90171"
],
"database_specific": {
"cwe_ids": [],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-09-17T17:17:10Z",
"severity": null
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\nsmb: smbdirect: release pending child sockets outside the handler lock\n\nsmbdirect_socket_destroy() releases the listener\u0027s pending/ready child\nsockets while still holding the listener\u0027s handler lock, the\n\u0026id_priv-\u003ehandler_mutex taken via rdma_lock_handler(), not\nsc-\u003elisten.lock, and before the listener\u0027s own rdma_destroy_id().\nThat ordering has one real consequence and one cosmetic one.\n\nThe real one: smbdirect_socket_release() drops the child\u0027s last\nreference, which destroys the child\u0027s cm_id. Doing that before the\nlistener\u0027s rdma_destroy_id() lets _cma_cancel_listens(), running from\nthe listener\u0027s _destroy_id(), walk an already freed child id_priv,\nwhich KASAN catches as a slab-use-after-free during listener shutdown:\n\n[ 4758.909130] BUG: KASAN: slab-use-after-free in __mutex_lock+0x1469/0x1560\n[ 4758.911450] Read of size 1 at addr ffff88821c381db4 by task ksmbd.control/1652\n[ 4758.913262] Call Trace:\n[ 4758.913267] \u003cTASK\u003e\n[ 4758.913299] __mutex_lock+0x1469/0x1560\n[ 4758.913408] _cma_cancel_listens+0x312/0x3b0\n[ 4758.913413] _destroy_id+0x363/0xee0\n[ 4758.913417] smbdirect_socket_destroy_sync+0x17d5/0x2440\n[ 4758.913443] smbdirect_socket_release+0x124/0x230\n[ 4758.913451] ksmbd_rdma_stop_listening+0x9f/0x190\n[ 4758.913457] ksmbd_conn_transport_destroy+0x65/0x3c0\n[ 4758.913463] kill_server_store+0x1fb/0x2b0\n[ 4758.913501] kernfs_fop_write_iter+0x349/0x4d0\n[ 4758.913507] vfs_write+0x5e7/0xc70\n[ 4758.913528] ksys_write+0x12a/0x210\n[ 4758.913541] do_syscall_64+0x135/0x460\n[ 4758.913555] entry_SYSCALL_64_after_hwframe+0x77/0x7f\n\nThe cosmetic one: releasing a child recurses into\nsmbdirect_socket_destroy(), which takes the child\u0027s own\nrdma_lock_handler() lock nested under the listener\u0027s. The listener\u0027s\nand the child\u0027s cm_id are always different instances, so this cannot\ndeadlock for real; the CM core itself nests a new connection id\u0027s\nhandler_mutex under the listening id\u0027s in cma_ib_req_handler(). But\nlockdep only sees one lock class, reports possible recursive locking,\nand then disables itself, hiding real locking bugs for the rest of the\nrun:\n\n[ 2424.579653] WARNING: possible recursive locking detected\n[ 2424.581180] 7.1.0-next-20260623+ #89 Not tainted\n[ 2424.582548] --------------------------------------------\n[ 2424.584500] ksmbd.control/8854 is trying to acquire lock:\n[ 2424.586817] ffff888102303c20 (\u0026id_priv-\u003ehandler_mutex){+.+.}-{4:4}, at: smbdirect_socket_destroy_sync+0xc39/0x2440\n[ 2424.590590]\n[ 2424.590590] but task is already holding lock:\n[ 2424.591601] ffff888102046c20 (\u0026id_priv-\u003ehandler_mutex){+.+.}-{4:4}, at: smbdirect_socket_destroy_sync+0xc39/0x2440\n[ 2424.594178]\n[ 2424.594178] other info that might help us debug this:\n[ 2424.596634] Possible unsafe locking scenario:\n[ 2424.596634]\n[ 2424.598841] CPU0\n[ 2424.599765] ----\n[ 2424.600695] lock(\u0026id_priv-\u003ehandler_mutex);\n[ 2424.601836] lock(\u0026id_priv-\u003ehandler_mutex);\n[ 2424.602590]\n[ 2424.602590] *** DEADLOCK ***\n[ 2424.602590]\n[ 2424.604512] May be due to missing lock nesting notation\n\nSplice the pending/ready children onto a local list under the\nlistener\u0027s listen.lock, while the handler lock is held so a concurrent\nCM CONNECT_REQUEST cannot add more, but defer the actual\nsmbdirect_socket_release() calls until after the listener\u0027s cm_id has\nbeen destroyed and its handler lock dropped. The children are\nindependent sockets whose teardown needs neither the listener\u0027s\nhandler lock nor its cm_id.\n\nFound with ksmbdzzer [2], a KSMBD fuzzer that drives libFuzzer with a\nkcov-dataflow [1] coverage vector: it folds each instrumented\ncomparison/argument\u0027s runtime operand value together with its PC (the\ndefault arm mixes them as pc\u2295val) so that a new operand value at a known\nsite counts as new coverage.\n\n[1] https://lwn.net/Articles/1077606/\n[2] https://github.com/yskzalloc/kcov-dataflow",
"id": "GHSA-f5r6-6w97-7r58",
"modified": "2026-09-17T18:31:54Z",
"published": "2026-09-17T18:31:53Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-90171"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/8ea83ef0c40ec4463695e22bc9ae43e29bcfede7"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/db82fbe4bb68e68e4aef00ef5b79f92991d8ef8e"
}
],
"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.
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.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.