GHSA-4H26-3W83-PFH6
Vulnerability from github – Published: 2026-03-25 12:30 – Updated: 2026-04-02 15:31In the Linux kernel, the following vulnerability has been resolved:
bpf: Fix race in devmap on PREEMPT_RT
On PREEMPT_RT kernels, the per-CPU xdp_dev_bulk_queue (bq) can be accessed concurrently by multiple preemptible tasks on the same CPU.
The original code assumes bq_enqueue() and __dev_flush() run atomically with respect to each other on the same CPU, relying on local_bh_disable() to prevent preemption. However, on PREEMPT_RT, local_bh_disable() only calls migrate_disable() (when PREEMPT_RT_NEEDS_BH_LOCK is not set) and does not disable preemption, which allows CFS scheduling to preempt a task during bq_xmit_all(), enabling another task on the same CPU to enter bq_enqueue() and operate on the same per-CPU bq concurrently.
This leads to several races:
-
Double-free / use-after-free on bq->q[]: bq_xmit_all() snapshots cnt = bq->count, then iterates bq->q[0..cnt-1] to transmit frames. If preempted after the snapshot, a second task can call bq_enqueue() -> bq_xmit_all() on the same bq, transmitting (and freeing) the same frames. When the first task resumes, it operates on stale pointers in bq->q[], causing use-after-free.
-
bq->count and bq->q[] corruption: concurrent bq_enqueue() modifying bq->count and bq->q[] while bq_xmit_all() is reading them.
-
dev_rx/xdp_prog teardown race: __dev_flush() clears bq->dev_rx and bq->xdp_prog after bq_xmit_all(). If preempted between bq_xmit_all() return and bq->dev_rx = NULL, a preempting bq_enqueue() sees dev_rx still set (non-NULL), skips adding bq to the flush_list, and enqueues a frame. When __dev_flush() resumes, it clears dev_rx and removes bq from the flush_list, orphaning the newly enqueued frame.
-
__list_del_clearprev() on flush_node: similar to the cpumap race, both tasks can call __list_del_clearprev() on the same flush_node, the second dereferences the prev pointer already set to NULL.
The race between task A (__dev_flush -> bq_xmit_all) and task B (bq_enqueue -> bq_xmit_all) on the same CPU:
Task A (xdp_do_flush) Task B (ndo_xdp_xmit redirect) ---------------------- -------------------------------- __dev_flush(flush_list) bq_xmit_all(bq) cnt = bq->count /* e.g. 16 / / start iterating bq->q[] / <-- CFS preempts Task A --> bq_enqueue(dev, xdpf) bq->count == DEV_MAP_BULK_SIZE bq_xmit_all(bq, 0) cnt = bq->count / same 16! / ndo_xdp_xmit(bq->q[]) / frames freed by driver / bq->count = 0 <-- Task A resumes --> ndo_xdp_xmit(bq->q[]) / use-after-free: frames already freed! */
Fix this by adding a local_lock_t to xdp_dev_bulk_queue and acquiring it in bq_enqueue() and __dev_flush(). These paths already run under local_bh_disable(), so use local_lock_nested_bh() which on non-RT is a pure annotation with no overhead, and on PREEMPT_RT provides a per-CPU sleeping lock that serializes access to the bq.
{
"affected": [],
"aliases": [
"CVE-2026-23294"
],
"database_specific": {
"cwe_ids": [
"CWE-362"
],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2026-03-25T11:16:24Z",
"severity": "HIGH"
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\nbpf: Fix race in devmap on PREEMPT_RT\n\nOn PREEMPT_RT kernels, the per-CPU xdp_dev_bulk_queue (bq) can be\naccessed concurrently by multiple preemptible tasks on the same CPU.\n\nThe original code assumes bq_enqueue() and __dev_flush() run atomically\nwith respect to each other on the same CPU, relying on\nlocal_bh_disable() to prevent preemption. However, on PREEMPT_RT,\nlocal_bh_disable() only calls migrate_disable() (when\nPREEMPT_RT_NEEDS_BH_LOCK is not set) and does not disable\npreemption, which allows CFS scheduling to preempt a task during\nbq_xmit_all(), enabling another task on the same CPU to enter\nbq_enqueue() and operate on the same per-CPU bq concurrently.\n\nThis leads to several races:\n\n1. Double-free / use-after-free on bq-\u003eq[]: bq_xmit_all() snapshots\n cnt = bq-\u003ecount, then iterates bq-\u003eq[0..cnt-1] to transmit frames.\n If preempted after the snapshot, a second task can call bq_enqueue()\n -\u003e bq_xmit_all() on the same bq, transmitting (and freeing) the\n same frames. When the first task resumes, it operates on stale\n pointers in bq-\u003eq[], causing use-after-free.\n\n2. bq-\u003ecount and bq-\u003eq[] corruption: concurrent bq_enqueue() modifying\n bq-\u003ecount and bq-\u003eq[] while bq_xmit_all() is reading them.\n\n3. dev_rx/xdp_prog teardown race: __dev_flush() clears bq-\u003edev_rx and\n bq-\u003exdp_prog after bq_xmit_all(). If preempted between\n bq_xmit_all() return and bq-\u003edev_rx = NULL, a preempting\n bq_enqueue() sees dev_rx still set (non-NULL), skips adding bq to\n the flush_list, and enqueues a frame. When __dev_flush() resumes,\n it clears dev_rx and removes bq from the flush_list, orphaning the\n newly enqueued frame.\n\n4. __list_del_clearprev() on flush_node: similar to the cpumap race,\n both tasks can call __list_del_clearprev() on the same flush_node,\n the second dereferences the prev pointer already set to NULL.\n\nThe race between task A (__dev_flush -\u003e bq_xmit_all) and task B\n(bq_enqueue -\u003e bq_xmit_all) on the same CPU:\n\n Task A (xdp_do_flush) Task B (ndo_xdp_xmit redirect)\n ---------------------- --------------------------------\n __dev_flush(flush_list)\n bq_xmit_all(bq)\n cnt = bq-\u003ecount /* e.g. 16 */\n /* start iterating bq-\u003eq[] */\n \u003c-- CFS preempts Task A --\u003e\n bq_enqueue(dev, xdpf)\n bq-\u003ecount == DEV_MAP_BULK_SIZE\n bq_xmit_all(bq, 0)\n cnt = bq-\u003ecount /* same 16! */\n ndo_xdp_xmit(bq-\u003eq[])\n /* frames freed by driver */\n bq-\u003ecount = 0\n \u003c-- Task A resumes --\u003e\n ndo_xdp_xmit(bq-\u003eq[])\n /* use-after-free: frames already freed! */\n\nFix this by adding a local_lock_t to xdp_dev_bulk_queue and acquiring\nit in bq_enqueue() and __dev_flush(). These paths already run under\nlocal_bh_disable(), so use local_lock_nested_bh() which on non-RT is\na pure annotation with no overhead, and on PREEMPT_RT provides a\nper-CPU sleeping lock that serializes access to the bq.",
"id": "GHSA-4h26-3w83-pfh6",
"modified": "2026-04-02T15:31:36Z",
"published": "2026-03-25T12:30:22Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-23294"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/1872e75375c40add4a35990de3be77b5741c252c"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/6c10b019785dc282c5f45d21e4a3f468b8fd6476"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/ab1a56c9d99189aa5c6e03940d06e40ba6a28240"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H",
"type": "CVSS_V3"
}
]
}
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
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.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
How durations are calculated
Following the VTEM framework (Black Hat USA 2015), time to patch starts when the vendor or a coordinator was informed, total time to patch at discovery, and disclosure to patch at public disclosure; vendor response time runs from the report to its acknowledgement. Time to exploit and time unpatched start at the earliest discovery or report, falling back to public disclosure. Unpatched time ends at the first recorded patch, or the calculation time when no patch is recorded. Exploitation to patch shows the interval between those two events; a published proof of concept is a public exploit, not exploitation.
Each result shows its source dates. A negative interval means the end event came first; when date-only inputs leave the order open, the result says so. Date-only inputs are approximate. The earliest evidence wins: KEV report dates and sightings are labelled as such. Records describing another vulnerability are listed but never calculated from, and a PoC alone does not establish exploitation. Missing patch evidence does not prove that no patch exists. Filters below change the event list only.
Loading timeline…
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.