{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2026-72367","assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","state":"PUBLISHED","assignerShortName":"Linux","dateReserved":"2026-08-09T03:40:39.922Z","datePublished":"2026-08-15T05:56:04.669Z","dateUpdated":"2026-08-17T05:43:20.980Z"},"containers":{"cna":{"providerMetadata":{"orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux","dateUpdated":"2026-08-17T05:43:20.980Z"},"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\niomap: guard io_size EOF trim against concurrent truncate underflow\n\niomap: fix zero padding data issue in concurrent append writes\nchanged ioend accounting so that io_size tracks only valid data\nwithin EOF.  This trims io_size when a writeback range extends\npast end_pos:\n\n    ioend->io_size += map_len;\n    if (ioend->io_offset + ioend->io_size > end_pos)\n        ioend->io_size = end_pos - ioend->io_offset;\n\nHowever, if end_pos ends up below ioend->io_offset, the subtraction\nbecomes negative and is stored in size_t io_size, causing an unsigned\nwrap to a huge value.  This can happen when writeback continues past\nbyte-level EOF up to a block-aligned range, or when a concurrent\ntruncate shrinks the file after end_pos was sampled in\niomap_writeback_handle_eof().\n\nA wrapped io_size can mislead append detection and corrupt\ncompletion-time size handling, since filesystem end_io paths consume\nio_size for decisions such as on-disk EOF updates and unwritten/COW\ncompletion ranges.\n\nFix this by clamping io_size to zero when EOF has moved to or before\nthe ioend start offset.  This preserves the original intent of trimming\nio_size to valid in-EOF data while avoiding the underflow."}],"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 - On XFS/GFS2/NTFS backends exposed via nfsd or ksmbd, a remote client can trigger iomap buffered writeback by issuing concurrent WRITE and truncate/SETATTR operations over the network without local shell access.\nAC:L - The attacker controls both sides of the race by concurrently writing dirty pages and truncating the same file (local threads or parallel NFS/SMB requests); no uncontrollable memory layout or rare kernel configuration is required.\nPR:L - Exploitation requires write and truncate permission on the target file (authenticated NFS/SMB client or local unprivileged user with file write access); this is not a pre-authentication network code path.\nUI:N - No victim interaction is needed beyond the attacker's own write and truncate I/O; writeback is driven automatically by the VFS/page-cache path once pages are dirtied.\nS:U - Impact is confined to kernel/filesystem integrity within the same security authority (metadata corruption, incorrect extent completion, possible local privilege escalation), not a VM escape or cross-authority boundary.\nC:H - A wrapped io_size propagates into XFS completion paths (COW convert/end, unwritten conversion, append detection), potentially remapping or exposing unintended on-disk extents and data beyond the legitimate EOF-trimmed range.\nI:H - Huge bogus io_size drives xfs_reflink_convert_cow, xfs_reflink_end_cow, xfs_iomap_write_unwritten, and xfs_setfilesize with corrupted ranges, enabling incorrect on-disk metadata updates and file content modification.\nA:H - Corrupted ioend accounting can trigger pathological XFS extent walks, ASSERT failures, filesystem shutdown, or hung writeback during I/O completion, causing kernel oops or sustained denial of service."}]}],"affected":[{"product":"Linux","vendor":"Linux","defaultStatus":"unaffected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["fs/iomap/ioend.c"],"versions":[{"version":"51d20d1dacbec589d459e11fc88fbca419f84a99","lessThan":"1f38f65bf965fce9aa159d45c5347538f56c5973","status":"affected","versionType":"git"},{"version":"51d20d1dacbec589d459e11fc88fbca419f84a99","lessThan":"7f7780abb4c0fdc9a2603aea8e985ff14ee900e0","status":"affected","versionType":"git"},{"version":"51d20d1dacbec589d459e11fc88fbca419f84a99","lessThan":"55ec50d046c03b3724741957f7b007856e36dbe7","status":"affected","versionType":"git"},{"version":"82c59a86a247a8970d353d10f52a37e5564fb137","status":"affected","versionType":"git"},{"version":"6.12.10","lessThan":"6.13","status":"affected","versionType":"semver"}]},{"product":"Linux","vendor":"Linux","defaultStatus":"affected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["fs/iomap/ioend.c"],"versions":[{"version":"6.13","status":"affected"},{"version":"0","lessThan":"6.13","status":"unaffected","versionType":"semver"},{"version":"6.18.40","lessThanOrEqual":"6.18.*","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":"6.13","versionEndExcluding":"6.18.40"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"6.13","versionEndExcluding":"7.1.5"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"6.13","versionEndExcluding":"7.2"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"6.12.10"}]}]}],"references":[{"url":"https://git.kernel.org/stable/c/1f38f65bf965fce9aa159d45c5347538f56c5973"},{"url":"https://git.kernel.org/stable/c/7f7780abb4c0fdc9a2603aea8e985ff14ee900e0"},{"url":"https://git.kernel.org/stable/c/55ec50d046c03b3724741957f7b007856e36dbe7"}],"title":"iomap: guard io_size EOF trim against concurrent truncate underflow","x_generator":{"engine":"bippy-1.2.0"}}}}