{"id":"AZL-100298","summary":"CVE-2026-80947 affecting package kernel 6.6.150.1-1","details":"In the Linux kernel, the following vulnerability has been resolved:\n\nwifi: rtl8xxxu: fix use-after-free from rx_urb_wq on stop\n\nrtl8xxxu arms rx_urb_wq from the RX completion path:\nrtl8xxxu_rx_complete() hands the URB to rtl8xxxu_queue_rx_urb(), which\nqueues it on rx_urb_pending_list and, once the list grows past\nRTL8XXXU_RX_URB_PENDING_WATER, schedules rx_urb_wq.  The worker\nrtl8xxxu_rx_urb_work() drains rx_urb_pending_list, recovers priv through\ncontainer_of, and resubmits each URB through rtl8xxxu_submit_rx_urb(),\nwhich anchors it on rx_anchor and dereferences priv-\u003eudev.\n\nrtl8xxxu_stop() cancels the sibling work items (c2hcmd_work, ra_watchdog,\nupdate_beacon_work) but never cancels rx_urb_wq, so a worker armed during\nthe last burst of RX traffic can run rtl8xxxu_rx_urb_work() after\nrtl8xxxu_disconnect() has called ieee80211_free_hw(), which frees priv,\nproducing a use-after-free.  The window opens under active RX traffic\n(pending count above the watermark) followed by a disconnect.\n\nThere are two teardown races to close:\n\n  * rtl8xxxu_queue_rx_urb() decided whether to enqueue under rx_urb_lock\n    but called schedule_work() after dropping the lock.  A completion\n    that observed shutdown == false and released the lock could then call\n    schedule_work() after rtl8xxxu_stop() had set shutdown and\n    cancel_work_sync() had already returned, arming the worker to run\n    after the teardown.  Move schedule_work() under the same !shutdown\n    branch so the arming decision is atomic with the shutdown check.\n\n  * rtl8xxxu_rx_urb_work() anchors every URB it drained back onto\n    rx_anchor through rtl8xxxu_submit_rx_urb().  A worker still running\n    when usb_kill_anchored_urbs(&priv-\u003erx_anchor) returned would submit a\n    URB that escaped the kill.  In rtl8xxxu_stop(), call\n    cancel_work_sync(&priv-\u003erx_urb_wq) before the kill so the worker is\n    drained first.\n\nAfter priv-\u003eshutdown is set under rx_urb_lock, completions can no longer\nqueue rx_urb_wq. cancel_work_sync() then drains the last queued or running\nworker, and the following usb_kill_anchored_urbs() kills the URBs it may\nhave submitted.\n\nrtl8xxxu_disconnect() is covered because ieee80211_unregister_hw()\nguarantees .stop() runs for a live interface before ieee80211_free_hw()\nfrees priv.  The probe error path needs no cancel: rx_urb_wq is\nINIT_WORK()'d there but cannot have been scheduled, since no URB is\nsubmitted before ieee80211_register_hw() succeeds.\n\nThis bug was found by static analysis.","modified":"2026-09-13T06:06:36Z","published":"2026-09-11T20:19:00Z","upstream":["CVE-2026-80947"],"references":[{"type":"WEB","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-80947"}],"affected":[{"package":{"name":"kernel","ecosystem":"Azure Linux:3","purl":"pkg:rpm/azure-linux/kernel"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"last_affected":"6.6.150.1-1"}]}],"database_specific":{"source":"https://github.com/microsoft/AzureLinuxVulnerabilityData/blob/main/osv/AZL-100298.json"}}],"schema_version":"1.9.0"}