{"id":"CVE-2026-99179","title":"watch_queue: Fix note pool publication race in watch_queue_set_size()","summary":"In the Linux kernel, the following vulnerability has been resolved:\n\nwatch_queue: Fix note pool publication race in watch_queue_set_size()\n\nThere is a race between watch_queue_set_size() and post_one_notification().\nwatch_queue_set_size(…","severity":"none","vendor":"Linux","product":"Linux","affected":["Linux >= c73be61cede5882f9605a852414db559c0ebedfd < 811fb6d0b07605e8e8d93bde8748be4b3028809a","Linux >= c73be61cede5882f9605a852414db559c0ebedfd < 9e73a6754457ffed8399b835552013b9363e51d1","Linux >= c73be61cede5882f9605a852414db559c0ebedfd < 5e531d62f1f8834faf7a4eed0d41105cf6570a8f","Linux 5.8"],"published":"2026-10-28","updated":"2026-10-28","sourceUpdated":"2026-10-28T10:35:32.734Z","source":"CVEORG","sourceUrl":"https://www.cve.org/CVERecord?id=CVE-2026-99179","references":[{"url":"https://git.kernel.org/stable/c/811fb6d0b07605e8e8d93bde8748be4b3028809a"},{"url":"https://git.kernel.org/stable/c/9e73a6754457ffed8399b835552013b9363e51d1"},{"url":"https://git.kernel.org/stable/c/5e531d62f1f8834faf7a4eed0d41105cf6570a8f"}],"tags":["cve.org"],"ingestedAt":"2026-10-28T11:09:40.381Z","slug":"CVE-2026-99179","body":"## Overview\n\nIn the Linux kernel, the following vulnerability has been resolved:\n\nwatch_queue: Fix note pool publication race in watch_queue_set_size()\n\nThere is a race between watch_queue_set_size() and post_one_notification().\nwatch_queue_set_size() publishes wqueue->notes, notes_bitmap, nr_pages and\nnr_notes with plain stores, and post_one_notification() reads them with\nplain loads under a different lock, so nothing orders the four stores\nagainst the poster's loads: depending on the build, notes can be stored\nlast (gcc 14 on x86-64 does this).  Since keyctl_watch_key() lets a watch\nattach to a queue before it is sized, post_one_notification() can already\nbe running on another CPU at that moment: it reads nr_notes > 0, passes\nits bounds check, and then reads notes while it is still NULL,\ndereferencing notes[0].\n\nAn unprivileged user triggers this reliably by watching one of its own\nkeys through a notification pipe and racing ioctl(IOC_WATCH_QUEUE_SET_SIZE)\non that pipe against keyctl(KEYCTL_SETPERM) on the key from another CPU,\ntriggering:\n\n  BUG: kernel NULL pointer dereference, address: 0000000000000000\n  RIP: 0010:post_one_notification.isra.0+0xa2/0x1e0\n  Call Trace:\n   __post_watch_notification+0x148/0x180\n   keyctl_setperm_key+0x119/0x130\n   do_syscall_64+0x108/0x4c0\n   entry_SYSCALL_64_after_hwframe+0x77/0x7f\n  Kernel panic - not syncing: Fatal exception in interrupt\n\nFix this by publishing nr_notes last with smp_store_release(), and by\nreading it once with smp_load_acquire() in post_one_notification() before\nthe bitmap and the page array are touched.  A poster that observes a\nnon-zero nr_notes then also observes the pool behind it; one that observes\nzero drops the notification as before.\n\n## Affected\n\n- `Linux >= c73be61cede5882f9605a852414db559c0ebedfd < 811fb6d0b07605e8e8d93bde8748be4b3028809a`\n- `Linux >= c73be61cede5882f9605a852414db559c0ebedfd < 9e73a6754457ffed8399b835552013b9363e51d1`\n- `Linux >= c73be61cede5882f9605a852414db559c0ebedfd < 5e531d62f1f8834faf7a4eed0d41105cf6570a8f`\n- `Linux 5.8`\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":[]}