{
  "affected": [
    {
      "ranges": [
        {
          "events": [
            {
              "introduced": "b013840810c221f2b0cf641d01531526052dc1fb"
            },
            {
              "fixed": "0189dce07db2dc059ae0570e06758ec4233c6e11"
            },
            {
              "fixed": "550d00aa58193fb649a09a9c9e338c685adad784"
            },
            {
              "fixed": "992cc9f94ca924089a506ba9b327caa9af797529"
            }
          ],
          "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
          "type": "GIT"
        }
      ]
    },
    {
      "package": {
        "ecosystem": "Linux",
        "name": "Kernel"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "3.14.0"
            },
            {
              "fixed": "7.1.13"
            }
          ],
          "type": "ECOSYSTEM"
        },
        {
          "events": [
            {
              "introduced": "7.2.0"
            },
            {
              "fixed": "7.2.3"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "database_specific": {
    "cna_assigner": "Linux",
    "osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/80xxx/CVE-2026-80841.json"
  },
  "details": "In the Linux kernel, the following vulnerability has been resolved:\n\nnet/packet: defer vmalloc TX_RING free until skbs finish\n\nAF_PACKET TX_RING skbs keep a raw pointer to their ring frame. The skb\npage references preserve page-backed ring blocks after pg_vec is freed,\nbut they do not preserve a vmalloc mapping.\n\ntpacket_destruct_skb() currently drops the pending reference before\nwriting the timestamp and TP_STATUS_AVAILABLE to the frame. Move the\ndecrement after those stores. The smp_wmb() in __packet_set_status()\norders the frame stores before the decrement.\n\nAlso recheck pending TX frames under pg_vec_lock before non-closing\nring replacement, so a racing send cannot add a pending skb between\nthe initial check and the ring swap.\n\nRing allocation can produce a mixture of page-backed and vmalloc-backed\nblocks. Allocate deferred-work storage during TX ring setup when the\nfirst vmalloc-backed block is encountered, and keep its pointer in the\npg_vec allocation header. If allocation fails, return -ENOMEM from ring\nsetup. On socket close, a non-NULL pointer identifies a vmalloc-backed\nvector without a scan. If TX skbs remain, defer the whole vector to\nsystem_long_wq.\n\nAfter pg_vec is detached, a late destructor can skip the pending\ndecrement. Use socket write-memory accounting as the deferred lifetime\ngate instead: an skb remains charged through its final sock_wfree(),\nafter all ring-frame accesses. The delayed work retains a socket\nreference and reschedules itself until no TX skbs remain.\n\nMove pending_refcnt release to packet_sock_destruct() so late skb\ndestructors and deferred cleanup can safely use it after\npacket_release(). Page-backed teardown remains synchronous, and no lock\nis added to the TX completion hot path.",
  "id": "CVE-2026-80841",
  "modified": "2026-09-06T03:30:52.756054899Z",
  "published": "2026-09-04T15:54:51.459Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://git.kernel.org/stable/c/0189dce07db2dc059ae0570e06758ec4233c6e11"
    },
    {
      "type": "WEB",
      "url": "https://git.kernel.org/stable/c/550d00aa58193fb649a09a9c9e338c685adad784"
    },
    {
      "type": "WEB",
      "url": "https://git.kernel.org/stable/c/992cc9f94ca924089a506ba9b327caa9af797529"
    },
    {
      "type": "ADVISORY",
      "url": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/80xxx/CVE-2026-80841.json"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-80841"
    },
    {
      "type": "PACKAGE",
      "url": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git"
    }
  ],
  "schema_version": "1.9.0",
  "summary": "net/packet: defer vmalloc TX_RING free until skbs finish"
}