{ "dataType": "CVE_RECORD", "dataVersion": "5.2", "cveMetadata": { "cveId": "CVE-2026-72353", "assignerOrgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "state": "PUBLISHED", "assignerShortName": "Linux", "dateReserved": "2026-08-09T03:40:39.921Z", "datePublished": "2026-08-15T05:55:55.480Z", "dateUpdated": "2026-08-17T05:43:11.052Z" }, "containers": { "cna": { "providerMetadata": { "orgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "shortName": "Linux", "dateUpdated": "2026-08-17T05:43:11.052Z" }, "descriptions": [ { "lang": "en", "value": "In the Linux kernel, the following vulnerability has been resolved:\n\nntfs: avoid stale runlist element dereference in fallocate\n\nntfs_attr_fallocate() allocates holes and delayed allocations inside\ninitialized size by looking up the current runlist element under\nni->runlist.lock. The returned struct runlist_element is only a borrowed\npointer into ni->runlist.rl. A writer can replace and free that array\nafter the read lock is dropped, so later reads of rl->lcn, rl->length and\nrl->vcn can touch freed memory.\n\nThe buggy scenario involves two paths, with each column showing the order\nwithin that path:\n\nntfs_attr_fallocate():\n 1. Take ni->runlist.lock for read.\n 2. Get rl from ntfs_attr_find_vcn_nolock().\n 3. Drop ni->runlist.lock.\n 4. Read rl->lcn, rl->length and rl->vcn.\n\nmmap page_mkwrite:\n 1. Enter ntfs_filemap_page_mkwrite().\n 2. Reach __ntfs_write_iomap_begin() and ntfs_attr_map_cluster().\n 3. Merge allocation state with ntfs_runlists_merge().\n 4. Reallocate ni->runlist.rl in ntfs_rl_realloc(), freeing the old array.\n\nValidation reproduced this kernel report:\nBUG: KASAN: slab-use-after-free in ntfs_attr_fallocate+0xbb8/0xd00\n\nCall Trace:\n \n dump_stack_lvl+0x66/0xa0\n print_report+0xce/0x630\n ? ntfs_attr_fallocate+0xbb8/0xd00\n ? srso_alias_return_thunk+0x5/0xfbef5\n ? __virt_addr_valid+0x20d/0x410\n ? ntfs_attr_fallocate+0xbb8/0xd00\n kasan_report+0xe0/0x110\n ? ntfs_attr_fallocate+0xbb8/0xd00\n ntfs_attr_fallocate+0xbb8/0xd00\n ? lock_acquire+0x2b8/0x2f0\n ? __pfx_ntfs_attr_fallocate+0x10/0x10\n ? 0xffffffffc0000095\n ? down_write+0x10d/0x1e0\n ntfs_fallocate+0x5c9/0x1d00\n ? __pfx_ntfs_fallocate+0x10/0x10\n ? srso_alias_return_thunk+0x5/0xfbef5\n ? lock_acquire+0x2b8/0x2f0\n ? srso_alias_return_thunk+0x5/0xfbef5\n ? selinux_file_permission+0x3a7/0x510\n vfs_fallocate+0x29d/0xd30\n __x64_sys_fallocate+0xc7/0x150\n ? do_syscall_64+0x81/0x6a0\n do_syscall_64+0x115/0x6a0\n entry_SYSCALL_64_after_hwframe+0x77/0x7f\n\nAllocated by task 410:\n kasan_save_stack+0x33/0x60\n kasan_save_track+0x14/0x30\n __kasan_kmalloc+0xaa/0xb0\n __kvmalloc_node_noprof+0x353/0x920\n ntfs_rl_realloc+0x3f/0x110\n ntfs_runlists_merge+0xaa3/0x3010\n ntfs_attr_map_cluster+0x4e5/0xf80\n ntfs_attr_fallocate+0x53f/0xd00\n ntfs_fallocate+0x5c9/0x1d00\n vfs_fallocate+0x29d/0xd30\n __x64_sys_fallocate+0xc7/0x150\n do_syscall_64+0x115/0x6a0\n entry_SYSCALL_64_after_hwframe+0x77/0x7f\n\nFreed by task 424:\n kasan_save_stack+0x33/0x60\n kasan_save_track+0x14/0x30\n kasan_save_free_info+0x3b/0x60\n __kasan_slab_free+0x5f/0x80\n kfree+0x307/0x580\n ntfs_rl_realloc+0x6f/0x110\n ntfs_runlists_merge+0x7b1/0x3010\n ntfs_attr_map_cluster+0x4e5/0xf80\n __ntfs_write_iomap_begin+0x8cd/0x2280\n iomap_iter+0x6de/0x11e0\n iomap_page_mkwrite+0x391/0x650\n ntfs_filemap_page_mkwrite+0x1ac/0x400\n do_page_mkwrite+0x15c/0x280\n __handle_mm_fault+0xd6d/0x1ca0\n handle_mm_fault+0x19c/0x470\n do_user_addr_fault+0x23b/0x9c0\n exc_page_fault+0x5c/0xc0\n asm_exc_page_fault+0x26/0x30\n\nFix this by copying the needed runlist fields while the read lock is still\nheld and using only those scalar snapshots after unlocking.\n\nAfter the snapshot, ntfs_attr_map_cluster() can also find that the range\nis already mapped and return balloc=false. Only call ntfs_dio_zero_range()\nwhen new clusters were allocated, matching the write iomap path and\npreserving the zero-newly-allocated-holes behavior." } ], "metrics": [ { "cvssV3_1": { "version": "3.1", "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H", "baseScore": 8.8, "baseSeverity": "HIGH" }, "scenarios": [ { "lang": "en", "value": "AV:N - Remote clients can reach ntfs_attr_fallocate via ksmbd or nfsd vfs_fallocate on NTFS-backed shares, while concurrent SMB/NFS writes trigger ntfs_attr_map_cluster/ntfs_runlists_merge to free the runlist during fallocate.\nAC:L - The attacker controls both race sides with parallel fallocate and write/mmap requests on the same NTFS file; the fix commit documents a reproducible KASAN slab-use-after-free from this attacker-driven concurrency.\nPR:L - Only write permission on a regular file on a mounted NTFS volume is required (local unprivileged user or authenticated SMB/NFS client), not real root in the initial mount namespace.\nUI:N - No victim interaction is needed during exploitation; the attacker prepares sparse/hole file state and drives fallocate plus concurrent writes without requiring another user to mount or open files.\nS:U - Impact is kernel heap corruption and privilege escalation within the host; it does not cross VM, hypervisor, or IOMMU security boundaries.\nC:H - Slab use-after-free on runlist_element fields (rl->lcn, rl->length, rl->vcn) reads freed kernel memory, enabling standard UAF-based arbitrary read and kernel pointer disclosure primitives.\nI:H - Freed runlist metadata influences subsequent cluster allocation and ntfs_dio_zero_range operations, providing UAF-based heap manipulation and arbitrary kernel write exploitation potential.\nA:H - Reproduced as KASAN slab-use-after-free in ntfs_attr_fallocate; stale runlist dereference can cause kernel oops, BUG, or panic when the race succeeds." } ] } ], "affected": [ { "product": "Linux", "vendor": "Linux", "defaultStatus": "unaffected", "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "programFiles": [ "fs/ntfs/attrib.c" ], "versions": [ { "version": "495e90fa334828d4119061e2726af51d0a0fb4ed", "lessThan": "3dd3e43f17cda174009a51fe668046ed7afef46a", "status": "affected", "versionType": "git" }, { "version": "495e90fa334828d4119061e2726af51d0a0fb4ed", "lessThan": "88496c4ac5a6ade75619f4b1015706a8b924d50a", "status": "affected", "versionType": "git" } ] }, { "product": "Linux", "vendor": "Linux", "defaultStatus": "affected", "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "programFiles": [ "fs/ntfs/attrib.c" ], "versions": [ { "version": "7.1", "status": "affected" }, { "version": "0", "lessThan": "7.1", "status": "unaffected", "versionType": "semver" }, { "version": "7.1.5", "lessThanOrEqual": "7.1.*", "status": "unaffected", "versionType": "semver" }, { "version": "7.2", "lessThanOrEqual": "*", "status": "unaffected", "versionType": "original_commit_for_fix" } ] } ], "cpeApplicability": [ { "nodes": [ { "operator": "OR", "negate": false, "cpeMatch": [ { "vulnerable": true, "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*", "versionStartIncluding": "7.1", "versionEndExcluding": "7.1.5" }, { "vulnerable": true, "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*", "versionStartIncluding": "7.1", "versionEndExcluding": "7.2" } ] } ] } ], "references": [ { "url": "https://git.kernel.org/stable/c/3dd3e43f17cda174009a51fe668046ed7afef46a" }, { "url": "https://git.kernel.org/stable/c/88496c4ac5a6ade75619f4b1015706a8b924d50a" } ], "title": "ntfs: avoid stale runlist element dereference in fallocate", "x_generator": { "engine": "bippy-1.2.0" } } } }