{"id":"CVE-2026-90342","title":"In the Linux kernel, the following vulnerability has been resolved:\n\nbpf: Fix mmap_lock deadlock on arena lock failure\n\nReported by the Sashiko AI review.\n\narena_vm_fault() returns VM_FAULT_RETRY when it can't take\narena->spinlock, but i…","summary":"In the Linux kernel, the following vulnerability has been resolved:\n\nbpf: Fix mmap_lock deadlock on arena lock failure\n\nReported by the Sashiko AI review.\n\narena_vm_fault() returns VM_FAULT_RETRY when it can't take\narena->spinlock, but i…","severity":"none","vendor":"Linux","product":"Linux","affected":["Linux >= b8467290edab4bafae352bf3f317055669a1a458 < af8087f16714f7620c4db3e732baaa1b2337948f","Linux >= b8467290edab4bafae352bf3f317055669a1a458 < 0b10b945479c954393d62ee3229d2f224a4ca91c","Linux 7.0"],"published":"2026-09-17","updated":"2026-09-17","sourceUpdated":"2026-09-17T17:17:32.877","source":"NVD","sourceUrl":"https://nvd.nist.gov/vuln/detail/CVE-2026-90342","references":[{"url":"https://git.kernel.org/stable/c/0b10b945479c954393d62ee3229d2f224a4ca91c","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"},{"url":"https://git.kernel.org/stable/c/af8087f16714f7620c4db3e732baaa1b2337948f","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"}],"tags":["nvd","cve.org"],"ingestedAt":"2026-09-17T16:21:47.817Z","epss":0.00155,"epssPercentile":0.05063,"slug":"CVE-2026-90342","body":"## Overview\n\nIn the Linux kernel, the following vulnerability has been resolved:\n\nbpf: Fix mmap_lock deadlock on arena lock failure\n\nReported by the Sashiko AI review.\n\narena_vm_fault() returns VM_FAULT_RETRY when it can't take\narena->spinlock, but it never took mmap_lock. The fault path assumes a\nVM_FAULT_RETRY handler already dropped mmap_lock and re-takes it on the\nretry, so mmap_lock gets taken twice and can deadlock:\n\n\tdo_user_addr_fault()\n\t{\n\t\tfault = handle_mm_fault(...);   // calls arena_vm_fault()\n\t\tif (fault & VM_FAULT_RETRY)\n\t\t\tgoto retry;   // re-locks mmap_lock\n\t\tmmap_read_unlock(mm);\n\t}\n\nReturn VM_FAULT_SIGBUS instead, for two reasons:\n\n1. We could keep VM_FAULT_RETRY, but then we'd have to drop the fault\n   lock first and cap the retry ourselves, the way __folio_lock_or_retry()\n   does.\n\n2. A failed raw_res_spin_lock_irqsave() already means a possible deadlock\n   was detected, so retrying just hits the same lock again.\n\nSo returning VM_FAULT_RETRY here is overkill.\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":[]}