{"id":"CVE-2026-92487","title":"In the Linux kernel, the following vulnerability has been resolved:\n\nexfat: fix valid_size extension over a shared writable mapping\n\nWhen a shared writable mapping has its valid_size extended by a buffered\nwrite or a page fault, exfat ze…","summary":"In the Linux kernel, the following vulnerability has been resolved:\n\nexfat: fix valid_size extension over a shared writable mapping\n\nWhen a shared writable mapping has its valid_size extended by a buffered\nwrite or a page fault, exfat ze…","severity":"none","vendor":"Linux","product":"Linux","affected":["Linux >= 82a81a7352bcf5f2756ac33d47ee0582737e9a85 < 24f20986b253c3e4eb13ebe3c2007d1d352408a6","Linux >= 82a81a7352bcf5f2756ac33d47ee0582737e9a85 < 1135704ed22f54873eb0498a232611d9eca30dd4","Linux 7.2"],"published":"2026-09-17","updated":"2026-09-17","sourceUpdated":"2026-09-17T17:17:50.800","source":"NVD","sourceUrl":"https://nvd.nist.gov/vuln/detail/CVE-2026-92487","references":[{"url":"https://git.kernel.org/stable/c/1135704ed22f54873eb0498a232611d9eca30dd4","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"},{"url":"https://git.kernel.org/stable/c/24f20986b253c3e4eb13ebe3c2007d1d352408a6","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"}],"tags":["nvd","cve.org"],"ingestedAt":"2026-09-17T16:21:47.786Z","epss":0.00189,"epssPercentile":0.08809,"slug":"CVE-2026-92487","body":"## Overview\n\nIn the Linux kernel, the following vulnerability has been resolved:\n\nexfat: fix valid_size extension over a shared writable mapping\n\nWhen a shared writable mapping has its valid_size extended by a buffered\nwrite or a page fault, exfat zeroes the page-cache gap below the new\nvalid_size. A store through the mapping can race with this zeroing and be\noverwritten.\n\nFix this by zeroing the gap lazily. Drop ->map_pages so that every first\nwrite fault goes through exfat_page_mkwrite(), which advances valid_size to\ncover the faulting page. With fault-around enabled, a store could install a\nwritable PTE, skip ->page_mkwrite(), and land past valid_size without\nadvancing it. Extending valid_size one faulting page at a time also leaves\nnever-written pages in a large mapping alone.\n\nThe gap is filled with block granularity, zeroing only the not-uptodate\nblocks and preserving blocks that may hold data stored through the mapping.\nOn the buffered-write path the invalidate lock is held and the gap is\nunmapped before zeroing, so a racing store re-faults and, under the inode\nlock, completes only after the gap has been zeroed and valid_size covers\nit.\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":[]}