{
  "schema_version": "1.6.1",
  "id": "RPI-CVE-2025-21655",
  "modified": "2026-10-09T11:39:44Z",
  "upstream": [
    "CVE-2025-21655"
  ],
  "published": "2025-01-20T13:50:06Z",
  "summary": "io_uring/eventfd: ensure io_eventfd_signal() defers another RCU period",
  "details": "In the Linux kernel, the following vulnerability has been resolved:\n\nio_uring/eventfd: ensure io_eventfd_signal() defers another RCU period\n\nio_eventfd_do_signal() is invoked from an RCU callback, but when\ndropping the reference to the io_ev_fd, it calls io_eventfd_free()\ndirectly if the refcount drops to zero. This isn't correct, as any\npotential freeing of the io_ev_fd should be deferred another RCU grace\nperiod.\n\nJust call io_eventfd_put() rather than open-code the dec-and-test and\nfree, which will correctly defer it another RCU grace period.",
  "severity": [
    {
      "type": "CVSS_V3",
      "score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H"
    }
  ],
  "affected": [
    {
      "package": {
        "ecosystem": "Raspberry Pi OS:12",
        "name": "linux"
      },
      "versions": [
        "1:6.1.47-1+rpt4"
      ],
      "ecosystem_specific": {
        "binaries": [
          {
            "binary_name": "linux-image-6.1.0-rpi3-rpi-2712",
            "binary_version": "1:6.1.47-1+rpt4"
          },
          {
            "binary_name": "linux-image-6.1.0-rpi3-rpi-v6",
            "binary_version": "1:6.1.47-1+rpt4"
          },
          {
            "binary_name": "linux-image-6.1.0-rpi3-rpi-v7",
            "binary_version": "1:6.1.47-1+rpt4"
          },
          {
            "binary_name": "linux-image-6.1.0-rpi3-rpi-v7l",
            "binary_version": "1:6.1.47-1+rpt4"
          },
          {
            "binary_name": "linux-image-6.1.0-rpi3-rpi-v8",
            "binary_version": "1:6.1.47-1+rpt4"
          }
        ]
      },
      "database_specific": {
        "per_version": {
          "1:6.1.47-1+rpt4": {
            "linux_commit": "655fc658a15ae7a6f37103754adb39ba52a9a14e",
            "upstream_version": "6.1.47",
            "assessment": "commit ancestry (strak rule)",
            "upstream_fixed": "6.1.125",
            "compiled_filter": false
          }
        }
      }
    }
  ],
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://www.cve.org/CVERecord?id=CVE-2025-21655"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/6b63308c28987c6010b1180c72a6db4df6c68033"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/8efff2aa2d95dc437ab67c5b4a9f1d3f367baa10"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/a7085c3ae43b86d4b3d1b8275e6a67f14257e3b7"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/c9a40292a44e78f71258b8522655bffaf5753bdb"
    }
  ],
  "database_specific": {
    "source": "https://git.kernel.org/pub/scm/linux/security/vulns.git",
    "upstream_modified": "2026-08-05T12:19:11Z",
    "severity": {
      "source": "kernel CNA",
      "cvss_v3_score": 7.8,
      "cvss_v3_severity": "High",
      "scenario": "AV:L - The vulnerability is reached through the io_uring_setup/io_uring_register/io_uring_enter syscalls and an eventfd write() from a local process. No network-reachable component processes the affected data.\nAC:L - The attacker controls both sides of the race: it deterministically schedules the RCU callback via a write() to a polled eventfd, then issues UNREGISTER_EVENTFD/ring-close during the multi-millisecond grace period while another thread spins in io_eventfd_grab(), and can retry unboundedly by re-registering a fresh ev_fd.\nPR:L - Only an ordinary unprivileged local user account is needed \u2014 sysctl kernel.io_uring_disabled defaults to 0 (io_uring allowed for everyone) and IORING_REGISTER_EVENTFD/IORING_UNREGISTER_EVENTFD perform no capability check.\nUI:N - The attacker performs every step itself \u2014 ring setup, eventfd registration, poll arming, the triggering write(), and the unregister. No victim action is involved.\nS:U - The use-after-free corrupts kernel heap memory within the same security authority; there is no VM, IOMMU, or sandbox boundary crossed.\nC:H - The UAF lets a reallocated kmalloc-64 object supply the ->cq_ev_fd pointer, which is dereferenced as a struct eventfd_ctx *, giving attacker-directed kernel reads and, via typical UAF heap-spray techniques, disclosure of arbitrary kernel memory.\nI:H - The stale object yields an atomic RMW on freed memory, an arbitrary-address increment (ctx->count++), a call_rcu() on freed memory, and a wake_up_locked_poll() walking an attacker-controlled waitqueue whose func pointer is invoked \u2014 a control-flow hijack primitive enabling privilege escalation.\nA:H - Freeing io_ev_fd a grace period early reliably produces kernel oops/panic from the UAF read, the atomic RMW on freed memory, and RCU callback-list corruption."
    }
  }
}