{"id":"CVE-2026-97619","title":"In the Linux kernel, the following vulnerability has been resolved:\n\nio_uring/rw: end write accounting from ->ki_complete\n\nCommit b000145e9907 moved both the fsnotify calls and the write\naccounting out of the kiocb completion handler and…","summary":"In the Linux kernel, the following vulnerability has been resolved:\n\nio_uring/rw: end write accounting from ->ki_complete\n\nCommit b000145e9907 moved both the fsnotify calls and the write\naccounting out of the kiocb completion handler and…","severity":"none","vendor":"Linux","product":"Linux","affected":["Linux >= b000145e9907809406d8164c3b2b8861d95aecd1 < cc580cee4dfa2ec9099c30ecbd4d804cbb996432","Linux >= b000145e9907809406d8164c3b2b8861d95aecd1 < 055d43a1233edbd80e558889258105ce63051bcd","Linux >= b000145e9907809406d8164c3b2b8861d95aecd1 < 796aa0547557e63338657ed1c487906f9fac4c73","Linux ea2e6286e3e89a115ae554e20ba9aec2b2e1ddff","Linux 89a410dbd0f159ddd308f19d6eb682fc753e4771","Linux 2a853c206e553dd9c0a55c22858fd6a446d93e15","Linux >= 5.10.165 < 5.11","Linux >= 5.15.90 < 5.16","Linux >= 6.0.3 < 6.1","Linux 6.1"],"published":"2026-09-25","updated":"2026-09-25","sourceUpdated":"2026-09-25T11:17:16.117","source":"NVD","sourceUrl":"https://nvd.nist.gov/vuln/detail/CVE-2026-97619","references":[{"url":"https://git.kernel.org/stable/c/055d43a1233edbd80e558889258105ce63051bcd","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"},{"url":"https://git.kernel.org/stable/c/796aa0547557e63338657ed1c487906f9fac4c73","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"},{"url":"https://git.kernel.org/stable/c/cc580cee4dfa2ec9099c30ecbd4d804cbb996432","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"}],"tags":["nvd","cve.org"],"ingestedAt":"2026-09-25T11:06:38.897Z","slug":"CVE-2026-97619","body":"## Overview\n\nIn the Linux kernel, the following vulnerability has been resolved:\n\nio_uring/rw: end write accounting from ->ki_complete\n\nCommit b000145e9907 moved both the fsnotify calls and the write\naccounting out of the kiocb completion handler and into the\nio_req_rw_complete() task_work. However, only the fsnotify part actually\nneeded to move as it may sleep. Ending the write accounting is just a\npercpu_up_read() on the superblock writers sem.\n\nDeferring it is a problem, because it makes dropping SB_FREEZE_WRITE\nprotection depend on the ring owner getting to running task_work. But\nthe task may be blocked in freeze_super(), causing it to never get to\nthat:\n\n  task                             io-wq worker\n  --------------------------------------------------------------\n  io_write()\n    io_kiocb_start_write()         (takes sb_writers, hidden from\n                                    lockdep by __sb_writers_release)\n    write_iter() -> -EIOCBQUEUED\n  ioctl(FS_IOC_SHUTDOWN)\n    bdev_freeze()\n      freeze_super()\n        percpu_down_write()        <- waits for the reader above\n                                   io_write()\n                                     kiocb_start_write()\n                                       percpu_down_read()  <- queued\n                                                              behind the\n                                                              writer\n  <bio completes>\n    io_complete_rw()\n      queues io_req_rw_complete()  <- never runs, task is in D state\n\nEnd the write from io_complete_rw() instead, and leave only the fsnotify\ncalls in task_work.\n\n## Remediation\n\nRefer to the linked advisories for vendor-supplied fixes and affected version ranges.","depth":"sunlit","depthScore":3,"depthScoreParts":{"impact":2.8,"likelihood":0,"exploitation":0,"ransomware":0},"changes":[]}