{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2026-99724","assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","state":"PUBLISHED","assignerShortName":"Linux","dateReserved":"2026-09-25T10:25:14.431Z","datePublished":"2026-11-20T05:51:00.658Z","dateUpdated":"2026-11-22T11:09:51.590Z"},"containers":{"cna":{"providerMetadata":{"orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux","dateUpdated":"2026-11-22T11:09:51.590Z"},"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nntfs: don't allow compression on volumes with clusters larger than 4K\n\nOn a volume made with \"mkntfs -c 8192\", a user can set\nFILE_ATTR_COMPRESSED on an empty file through system.ntfs_attrib.\nWriting to the file and reading it back after dropping the page cache\nthen crashes:\n\n  BUG: KASAN: slab-out-of-bounds in ntfs_read_compressed_block+0xa2c/0x2040\n  Read of size 8192 at addr ffff88804355a000 by task dd/106\n\nThe read path only handles clusters up to 4K. It copies each cluster\nout of a single page, and a compression block of 16 8K clusters does\nnot fit in the 64K ntfs_compression_buffer. Loading an inode already\nrejects compressed data on such volumes.\n\nReject the flag change the same way."}],"metrics":[{"cvssV3_1":{"version":"3.1","vectorString":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H","baseScore":7.8,"baseSeverity":"HIGH"},"scenarios":[{"lang":"en","value":"AV:L - FILE_ATTR_COMPRESSED reaches the patched ntfs_new_attr_flags() only through ntfs_setxattr() on system.ntfs_attrib, a local setxattr; the faulting copy is ntfs_read_compressed_block() called from ntfs_read_folio() on a later read(). Cluster size is on-volume metadata, and no remote protocol message carries the flag.\nAC:L - The attacker builds the volume with clusters above 4KiB (mkntfs -c 8192, which the boot-sector path accepts) and, once it is mounted writable, sets the flag on an empty file, writes, and drops that inode's folios. compression_unit is stored as 4 and the cluster_size memcpy_from_page() then runs on every such read, with no race and no spontaneous media error.\nPR:L - ntfs_setxattr() applies capable(CAP_SYS_ADMIN) to the $LXUID, $LXGID, $LXMOD, and $LXDEV names. system.ntfs_attrib skips that check, and xattr_permission() accepts system.* after may_write_xattr(). Default fmask 0 leaves i_mode 0777, so a local user who can create a regular file on the writable mount can set FILE_ATTR_COMPRESSED.\nUI:N - That local user creates the empty regular file, setxattrs system.ntfs_attrib into ntfs_new_attr_flags(), writes through ntfs_file_write_iter(), and reads through ntfs_file_read_iter() after invalidating the folio. ntfs_read_locked_inode() already rejects an on-disk compressed attribute when cluster_size is above 4096, so the missing check is reached by those calls alone.\nS:U - ntfs_read_compressed_block() overreads a block-device page and writes past the global ntfs_compression_buffer inside the host kernel that owns the mounted ntfs filesystem, the same security authority as a normal local privilege boundary.\nC:H - memcpy_from_page() in ntfs_read_compressed_block() copies vol->cluster_size bytes, 8192 in the reported case, from a single PAGE_SIZE bdev page, so the tail is the next direct-map page. On a short run that stays inside the 64KiB buffer, ntfs_decompress() can take those bytes as LZNT1 literals and copy them into the folio read() returns.\nI:H - ntfs_new_attr_flags() forces compression_unit 4, so the compression block is 16 clusters (128KiB when cluster_size is 8192) while ntfs_compression_buffer is vmalloc'd to NTFS_MAX_CB_SIZE (64KiB). The vcn loop copies each cluster to cb_pos and adds cluster_size, writing past that buffer.\nA:H - The copy is the reported KASAN slab-out-of-bounds read of 8192 bytes in ntfs_read_compressed_block(). A full 16-cluster block also writes past the 64KiB vmalloc region and faults on its guard page, oopsing the kernel."}]}],"affected":[{"product":"Linux","vendor":"Linux","defaultStatus":"unaffected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["fs/ntfs/ea.c"],"versions":[{"version":"fc053f05ca282a5e760b41f6560ac835c4e28037","lessThan":"df35f062d4f24e4c9d9711d6af24916d3784871c","status":"affected","versionType":"git"},{"version":"fc053f05ca282a5e760b41f6560ac835c4e28037","lessThan":"c77ab4eeeb3ed471bcc99a94673f0e0142c4b84a","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/ea.c"],"versions":[{"version":"7.1","status":"affected"},{"version":"0","lessThan":"7.1","status":"unaffected","versionType":"semver"},{"version":"7.3.5","lessThanOrEqual":"7.3.*","status":"unaffected","versionType":"semver"},{"version":"7.4-rc2","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":"7.3.5"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"7.1","versionEndExcluding":"7.4-rc2"}]}]}],"references":[{"url":"https://git.kernel.org/stable/c/df35f062d4f24e4c9d9711d6af24916d3784871c"},{"url":"https://git.kernel.org/stable/c/c77ab4eeeb3ed471bcc99a94673f0e0142c4b84a"}],"title":"ntfs: don't allow compression on volumes with clusters larger than 4K","x_generator":{"engine":"bippy-1.2.0"}}}}