{
  "schema_version": "1.6.1",
  "id": "RPI-CVE-2026-97478",
  "modified": "2026-10-09T11:39:44Z",
  "upstream": [
    "CVE-2026-97478"
  ],
  "published": "2026-09-24T16:08:25Z",
  "summary": "virt: acrn: Fix irqfd use-after-free during eventfd shutdown",
  "details": "In the Linux kernel, the following vulnerability has been resolved:\n\nvirt: acrn: Fix irqfd use-after-free during eventfd shutdown\n\nacrn_irqfd_deassign() and the eventfd EPOLLHUP wakeup can race and free\nthe same struct hsm_irqfd:\n\n  CPU0                                 CPU1\n  ----                                 ----\n  eventfd_release()\n    wake_up_poll(EPOLLHUP)\n      hsm_irqfd_wakeup()\n        queue_work(&irqfd->shutdown)\n                                       acrn_irqfd_deassign()\n                                         hsm_irqfd_shutdown()\n                                           list_del_init()\n                                           eventfd_ctx_remove_wait_queue()\n                                           eventfd_ctx_put()\n                                         kfree(irqfd)\n  hsm_irqfd_shutdown_work()\n    container_of(work, ..., shutdown)\n    irqfd->vm                  <-- use-after-free\n\nThe deassign path freed the irqfd while a shutdown work item was\nalready queued by EPOLLHUP (or vice versa), so the work item could\nresurrect a dangling pointer through container_of().\n\nSwitch to the lifetime model used by KVM irqfds:\n\n - Deassign/deinit only deactivate the irqfd: remove it from vm->irqfds\n   under irqfds_lock and queue the cleanup work.\n - hsm_irqfd_shutdown_work() becomes the sole owner that unhooks the\n   eventfd waitqueue entry, drops the eventfd reference and frees the\n   irqfd.\n - A new HSM_IRQFD_FLAG_SHUTDOWN bit guarded by test_and_set_bit()\n   ensures the cleanup work is queued at most once, no matter how many\n   of {EPOLLHUP, deassign, deinit} fire concurrently.  This is safe to\n   call from the waitqueue callback, which runs with wqh->lock held and\n   IRQs disabled and therefore cannot take irqfds_lock.\n - acrn_irqfd_deassign() flushes vm->irqfd_wq before returning so the\n   eventfd is fully detached on return.  acrn_irqfd_deinit() deactivates\n   every irqfd, flushes the workqueue and only then destroys it, so no\n   path can queue_work() onto a torn-down workqueue.\n - acrn_irqfd_assign() now installs the eventfd waitqueue entry and\n   publishes the irqfd to vm->irqfds under irqfds_lock, so the irqfd is\n   never visible to deassign/deinit before its waitqueue entry is in\n   place, and any EPOLLHUP that fires in the assign window queues\n   cleanup work that blocks on irqfds_lock until publication is done.",
  "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": null,
            "compiled_filter": false
          }
        }
      }
    }
  ],
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://www.cve.org/CVERecord?id=CVE-2026-97478"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/d8156aacfcabb81f52c20ef89094945434d08079"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/666c7f9e07925aa0863348f960e09bb89f8f05a3"
    }
  ],
  "database_specific": {
    "source": "https://git.kernel.org/pub/scm/linux/security/vulns.git",
    "upstream_modified": "2026-09-28T05:31:34Z",
    "severity": {
      "source": "kernel CNA",
      "cvss_v3_score": 7.8,
      "cvss_v3_severity": "High",
      "scenario": "AV:L - The UAF is reached from ioctl(ACRN_IOCTL_IRQFD) on /dev/acrn_hsm: acrn_dev_ioctl() copy_from_user()s struct acrn_irqfd and calls acrn_irqfd_config() \u2192 acrn_irqfd_assign()/acrn_irqfd_deassign(), while close() of that eventfd hits eventfd_release() \u2192 hsm_irqfd_wakeup(). No remote protocol carries the fd, flags, or EPOLLHUP.\nAC:L - The attacker owns both sides: one thread ioctl(ACRN_IOCTL_IRQFD) with ACRN_IRQFD_FLAG_DEASSIGN so acrn_irqfd_deassign() \u2192 hsm_irqfd_shutdown() kfree()s the hsm_irqfd, while another close()s a dup of the same eventfd so eventfd_release() \u2192 wake_up_poll(EPOLLHUP) \u2192 hsm_irqfd_wakeup() queues hsm_irqfd_shutdown_work(); the window is retryable.\nPR:L - acrn_dev_open(), acrn_dev_ioctl() and acrn_irqfd_config() have no capable()/CAP_* gate; only /dev/acrn_hsm DAC applies. ACRN Service VM deployments grant that node to the deprivileged acrn-dm device-model account (as with /dev/kvm), so a local user with that fd can CREATE_VM and IRQFD without init-namespace root.\nUI:N - The attacker opens /dev/acrn_hsm, ioctl(ACRN_IOCTL_CREATE_VM) then ACRN_IOCTL_IRQFD assign, and races close() of its own eventfd against deassign (or acrn_dev_release() \u2192 acrn_irqfd_deinit()); no other user must mount, open, or click.\nS:U - hsm_irqfd_shutdown_work() uses the freed struct hsm_irqfd in the Service VM kernel that runs acrn_dev_ioctl(). The attacker is already a Service VM process issuing HSM ioctls, not a User VM guest crossing into the ACRN hypervisor, so impact stays in one kernel authority.\nC:H - After acrn_irqfd_deassign() kfree()s the hsm_irqfd, hsm_irqfd_shutdown_work() container_of()s the queued work_struct and reads irqfd->vm (and later irqfd->eventfd) from the freed kmalloc object. Reclaiming that slot lets the attacker disclose reallocated kernel heap through those dangling loads.\nI:H - The same UAF lets hsm_irqfd_shutdown_work() mutex_lock(&irqfd->vm->irqfds_lock), list_del_init(&irqfd->list) and eventfd_ctx_put(irqfd->eventfd) on attacker-reclaimed hsm_irqfd fields, which is an arbitrary write / wait-queue function-pointer hijack, plus a second kfree of the sprayed object.\nA:H - hsm_irqfd_shutdown_work() loading irqfd->vm after hsm_irqfd_shutdown() kfree()s it oopses on a dangling pointer or corrupts the irqfds mutex/list; the EPOLLHUP work versus deassign kfree race in the commit is a Service VM kernel panic even without a successful spray."
    }
  }
}