{"api_version":"1","generated_at":"2026-08-18T11:35:09+00:00","cve":"CVE-2026-72203","urls":{"html":"https://cve.report/CVE-2026-72203","api":"https://cve.report/api/cve/CVE-2026-72203.json","docs":"https://cve.report/api","cve_org":"https://www.cve.org/CVERecord?id=CVE-2026-72203","nvd":"https://nvd.nist.gov/vuln/detail/CVE-2026-72203"},"summary":{"title":"ntfs: skip extent mft records in writeback to prevent deadlock","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nntfs: skip extent mft records in writeback to prevent deadlock\n\nThis patch fixes the ABBA deadlock between extent_lock and extent\nmrec_lock triggered by xfstests generic/113, that occurs since the commit\n6994acf33bae (\"ntfs: use base mft_no when looking up base inode for\n\t\textent record\").\n\nPath A (inode writeback):\n  VFS writeback\n    -> ntfs_write_inode()\n      -> __ntfs_write_inode()\n        -> mutex_lock(&ni->extent_lock)\n        -> mutex_lock(&tni->mrec_lock)\n\nPath B (MFT folio writeback):\n  VFS writeback of $MFT dirty folios\n    -> ntfs_mft_writepages()\n      -> ntfs_write_mft_block()\n        -> ntfs_may_write_mft_record()\n          -> holds one extent mrec_lock from a previous iteration\n          -> tries to acquire another base inode extent_lock\n\nBy removing all extent_lock and extent mrec_lock acquisition from the MFT\nfolio writeback path, the ABBA lock ordering is eliminated:\n\nPath A: __ntfs_write_inode(): extent_lock -> mrec_lock\nPath B (removed): ntfs_write_mft_block(): mrec_lock -> extent_lock\n\nPath B is always redundant for extent records because:\n\n1. mark_mft_record_dirty(ext_ni) does NOT dirty the MFT folio.\n   It only sets NInoDirty(ext_ni) and marks the base VFS inode dirty\n   via __mark_inode_dirty(I_DIRTY_DATASYNC), which triggers Path A.\n   Therefore, normal extent modifications never create a situation where\n   the MFT folio is dirty and Path B is not scheduled.\n\n2. The MFT folio only gets dirtied via ntfs_mft_mark_dirty() inside\n   ntfs_mft_record_alloc(). But all identified callers in attrib.c\n   (ntfs_attr_add, ntfs_attr_record_move_away,\n   ntfs_attr_make_non_resident, ntfs_attr_record_resize) follow through\n   with mark_mft_record_dirty(), which triggers Path A to write the\n   complete record.\n\n3. ntfs_evict_big_inode() calls ntfs_commit_inode() before freeing extent\n   inodes, ensuring all dirty extents are flushed via Path A before the\n   base inode leaves the icache.","state":"PUBLISHED","assigner":"Linux","published_at":"2026-08-15 06:21:38","updated_at":"2026-08-18 07:16:53"},"problem_types":[],"metrics":[{"version":"3.1","source":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","type":"Secondary","score":"7.5","severity":"HIGH","vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","data":{"version":"3.1","vectorString":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","baseScore":7.5,"baseSeverity":"HIGH","attackVector":"NETWORK","attackComplexity":"LOW","privilegesRequired":"NONE","userInteraction":"NONE","scope":"UNCHANGED","confidentialityImpact":"NONE","integrityImpact":"NONE","availabilityImpact":"HIGH"}},{"version":"3.1","source":"CNA","type":"DECLARED","score":"7.5","severity":"HIGH","vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","data":{"baseScore":7.5,"baseSeverity":"HIGH","vectorString":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","version":"3.1"}}],"references":[{"url":"https://git.kernel.org/stable/c/f831ab09d521898bf1dd99bf5adfd630ea1428e3","name":"https://git.kernel.org/stable/c/f831ab09d521898bf1dd99bf5adfd630ea1428e3","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/76bc14c7097ff678b2b5dbfd4fa33b46897d87ce","name":"https://git.kernel.org/stable/c/76bc14c7097ff678b2b5dbfd4fa33b46897d87ce","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-72203","name":"CVE Program record","refsource":"CVE.ORG","tags":["canonical"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-72203","name":"NVD vulnerability detail","refsource":"NVD","tags":["canonical","analysis"]}],"affected":[{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 1da177e4c3f41524e886b7f1b8a0c1fc7321cac2 f831ab09d521898bf1dd99bf5adfd630ea1428e3 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 1da177e4c3f41524e886b7f1b8a0c1fc7321cac2 76bc14c7097ff678b2b5dbfd4fa33b46897d87ce git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 2.6.12","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 2.6.12 semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.1.5 7.1.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.2 * original_commit_for_fix","platforms":[]}],"timeline":[],"solutions":[],"workarounds":[],"exploits":[],"credits":[],"nvd_cpes":[],"vendor_comments":[],"enrichments":{"kev":null,"epss":{"cve_year":"2026","cve_id":"72203","cve":"CVE-2026-72203","epss":"0.004680000","percentile":"0.384800000","score_date":"2026-08-17","updated_at":"2026-08-18 00:11:47"},"legacy_qids":[]},"source_records":{"cve_program":{"containers":{"cna":{"affected":[{"defaultStatus":"unaffected","product":"Linux","programFiles":["fs/ntfs/mft.c"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"lessThan":"f831ab09d521898bf1dd99bf5adfd630ea1428e3","status":"affected","version":"1da177e4c3f41524e886b7f1b8a0c1fc7321cac2","versionType":"git"},{"lessThan":"76bc14c7097ff678b2b5dbfd4fa33b46897d87ce","status":"affected","version":"1da177e4c3f41524e886b7f1b8a0c1fc7321cac2","versionType":"git"}]},{"defaultStatus":"affected","product":"Linux","programFiles":["fs/ntfs/mft.c"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"status":"affected","version":"2.6.12"},{"lessThan":"2.6.12","status":"unaffected","version":"0","versionType":"semver"},{"lessThanOrEqual":"7.1.*","status":"unaffected","version":"7.1.5","versionType":"semver"},{"lessThanOrEqual":"*","status":"unaffected","version":"7.2","versionType":"original_commit_for_fix"}]}],"cpeApplicability":[{"nodes":[{"cpeMatch":[{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.1.5","versionStartIncluding":"2.6.12","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.2","versionStartIncluding":"2.6.12","vulnerable":true}],"negate":false,"operator":"OR"}]}],"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nntfs: skip extent mft records in writeback to prevent deadlock\n\nThis patch fixes the ABBA deadlock between extent_lock and extent\nmrec_lock triggered by xfstests generic/113, that occurs since the commit\n6994acf33bae (\"ntfs: use base mft_no when looking up base inode for\n\t\textent record\").\n\nPath A (inode writeback):\n  VFS writeback\n    -> ntfs_write_inode()\n      -> __ntfs_write_inode()\n        -> mutex_lock(&ni->extent_lock)\n        -> mutex_lock(&tni->mrec_lock)\n\nPath B (MFT folio writeback):\n  VFS writeback of $MFT dirty folios\n    -> ntfs_mft_writepages()\n      -> ntfs_write_mft_block()\n        -> ntfs_may_write_mft_record()\n          -> holds one extent mrec_lock from a previous iteration\n          -> tries to acquire another base inode extent_lock\n\nBy removing all extent_lock and extent mrec_lock acquisition from the MFT\nfolio writeback path, the ABBA lock ordering is eliminated:\n\nPath A: __ntfs_write_inode(): extent_lock -> mrec_lock\nPath B (removed): ntfs_write_mft_block(): mrec_lock -> extent_lock\n\nPath B is always redundant for extent records because:\n\n1. mark_mft_record_dirty(ext_ni) does NOT dirty the MFT folio.\n   It only sets NInoDirty(ext_ni) and marks the base VFS inode dirty\n   via __mark_inode_dirty(I_DIRTY_DATASYNC), which triggers Path A.\n   Therefore, normal extent modifications never create a situation where\n   the MFT folio is dirty and Path B is not scheduled.\n\n2. The MFT folio only gets dirtied via ntfs_mft_mark_dirty() inside\n   ntfs_mft_record_alloc(). But all identified callers in attrib.c\n   (ntfs_attr_add, ntfs_attr_record_move_away,\n   ntfs_attr_make_non_resident, ntfs_attr_record_resize) follow through\n   with mark_mft_record_dirty(), which triggers Path A to write the\n   complete record.\n\n3. ntfs_evict_big_inode() calls ntfs_commit_inode() before freeing extent\n   inodes, ensuring all dirty extents are flushed via Path A before the\n   base inode leaves the icache."}],"metrics":[{"cvssV3_1":{"baseScore":7.5,"baseSeverity":"HIGH","vectorString":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","version":"3.1"},"scenarios":[{"lang":"en","value":"AV:N - The ABBA deadlock is reachable through concurrent VFS writeback on an rw-mounted NTFS volume; ntfs registers ntfs_export_ops, so nfsd/ksmbd remote WRITE/CLOSE/fsync RPCs on an exported NTFS share hit __ntfs_write_inode() and ntfs_mft_writepages() the same as local aio/write syscalls.\nAC:L - xfstests generic/113 (aio-stress with concurrent O_DIRECT/O_SYNC writes) reliably reproduces the hang; an attacker controls both lock-order participants by issuing parallel file metadata writes and sync/writeback pressure without races or attacker-independent memory layout.\nPR:N - Exploitation needs only write access to an rw-mounted or NFS/SMB-exported NTFS volume to drive ntfs_mft_record_alloc and concurrent writeback; no CAP_SYS_ADMIN, root, or init-namespace privileges on the server are required to deadlock the writeback paths.\nUI:N - Once an NTFS volume is rw-mounted (dual-boot partition, removable media, loop mount, or network export), the attacker issues the concurrent writes itself; no additional victim clicks, opens, or cooperative actions are required beyond that standing configuration.\nS:U - The circular mutex wait between extent_lock and extent mrec_lock only hangs kernel writeback and I/O worker threads within the host kernel security authority; it does not cross VM, container, IOMMU, or other security-boundary scopes.\nC:N - This is a deterministic lock-order inversion with no out-of-bounds access, use-after-free, or information disclosure; blocked threads wait on mutexes without reading attacker-controlled freed memory or leaking kernel pointers.\nI:N - The defect prevents completion of metadata writeback by deadlocking mutex acquisition; it does not perform attacker-controlled memory writes, corrupt on-disk NTFS structures, or provide a metadata modification primitive beyond denying forward progress.\nA:H - The ABBA deadlock between inode writeback (extent_lock then extent mrec_lock) and $MFT folio writeback (held mrec_lock then extent_lock) hangs kworker writeback and sync/fsync callers indefinitely, causing system-wide I/O denial of service."}]}],"providerMetadata":{"dateUpdated":"2026-08-18T06:56:21.384Z","orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux"},"references":[{"url":"https://git.kernel.org/stable/c/f831ab09d521898bf1dd99bf5adfd630ea1428e3"},{"url":"https://git.kernel.org/stable/c/76bc14c7097ff678b2b5dbfd4fa33b46897d87ce"}],"title":"ntfs: skip extent mft records in writeback to prevent deadlock","x_generator":{"engine":"bippy-1.2.0"}}},"cveMetadata":{"assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","assignerShortName":"Linux","cveId":"CVE-2026-72203","datePublished":"2026-08-15T05:54:00.321Z","dateReserved":"2026-08-09T03:40:39.912Z","dateUpdated":"2026-08-18T06:56:21.384Z","state":"PUBLISHED"},"dataType":"CVE_RECORD","dataVersion":"5.2"},"nvd":{"publishedDate":"2026-08-15 06:21:38","lastModifiedDate":"2026-08-18 07:16:53","problem_types":[],"metrics":{"cvssMetricV31":[{"source":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","type":"Secondary","cvssData":{"version":"3.1","vectorString":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","baseScore":7.5,"baseSeverity":"HIGH","attackVector":"NETWORK","attackComplexity":"LOW","privilegesRequired":"NONE","userInteraction":"NONE","scope":"UNCHANGED","confidentialityImpact":"NONE","integrityImpact":"NONE","availabilityImpact":"HIGH"},"exploitabilityScore":3.9,"impactScore":3.6}]},"configurations":[]},"legacy_mitre":{"record":{"CveYear":"2026","CveId":"72203","Ordinal":"1","Title":"ntfs: skip extent mft records in writeback to prevent deadlock","CVE":"CVE-2026-72203","Year":"2026"},"notes":[{"CveYear":"2026","CveId":"72203","Ordinal":"1","NoteData":"In the Linux kernel, the following vulnerability has been resolved:\n\nntfs: skip extent mft records in writeback to prevent deadlock\n\nThis patch fixes the ABBA deadlock between extent_lock and extent\nmrec_lock triggered by xfstests generic/113, that occurs since the commit\n6994acf33bae (\"ntfs: use base mft_no when looking up base inode for\n\t\textent record\").\n\nPath A (inode writeback):\n  VFS writeback\n    -> ntfs_write_inode()\n      -> __ntfs_write_inode()\n        -> mutex_lock(&ni->extent_lock)\n        -> mutex_lock(&tni->mrec_lock)\n\nPath B (MFT folio writeback):\n  VFS writeback of $MFT dirty folios\n    -> ntfs_mft_writepages()\n      -> ntfs_write_mft_block()\n        -> ntfs_may_write_mft_record()\n          -> holds one extent mrec_lock from a previous iteration\n          -> tries to acquire another base inode extent_lock\n\nBy removing all extent_lock and extent mrec_lock acquisition from the MFT\nfolio writeback path, the ABBA lock ordering is eliminated:\n\nPath A: __ntfs_write_inode(): extent_lock -> mrec_lock\nPath B (removed): ntfs_write_mft_block(): mrec_lock -> extent_lock\n\nPath B is always redundant for extent records because:\n\n1. mark_mft_record_dirty(ext_ni) does NOT dirty the MFT folio.\n   It only sets NInoDirty(ext_ni) and marks the base VFS inode dirty\n   via __mark_inode_dirty(I_DIRTY_DATASYNC), which triggers Path A.\n   Therefore, normal extent modifications never create a situation where\n   the MFT folio is dirty and Path B is not scheduled.\n\n2. The MFT folio only gets dirtied via ntfs_mft_mark_dirty() inside\n   ntfs_mft_record_alloc(). But all identified callers in attrib.c\n   (ntfs_attr_add, ntfs_attr_record_move_away,\n   ntfs_attr_make_non_resident, ntfs_attr_record_resize) follow through\n   with mark_mft_record_dirty(), which triggers Path A to write the\n   complete record.\n\n3. ntfs_evict_big_inode() calls ntfs_commit_inode() before freeing extent\n   inodes, ensuring all dirty extents are flushed via Path A before the\n   base inode leaves the icache.","Type":"Description","Title":"ntfs: skip extent mft records in writeback to prevent deadlock"}]}}}