{
  "schema_version": "1.6.1",
  "id": "RPI-CVE-2026-89601",
  "modified": "2026-10-09T11:39:44Z",
  "upstream": [
    "CVE-2026-89601"
  ],
  "published": "2026-09-11T19:47:07Z",
  "summary": "ext2: Fix lost inode updates for IS_SYNC inodes",
  "details": "In the Linux kernel, the following vulnerability has been resolved:\n\next2: Fix lost inode updates for IS_SYNC inodes\n\next2_setsize() and ext2_xattr_set2() had a construct like:\n\n\tif (IS_SYNC(inode)) {\n\t\tsync_inode_metadata(inode, 1);\n\t} else {\n\t\tmark_inode_dirty(inode);\n\t}\n\nwhich leads to lost inode updates for IS_SYNC inodes because\nsync_inode_metadata() does anything only if the inode is already dirty\nand hence inode updates may be simply lost. Fix the problem by\nunconditionally marking the inode dirty and *then* call\nsync_inode_metadata().",
  "severity": [
    {
      "type": "CVSS_V3",
      "score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H"
    }
  ],
  "affected": [
    {
      "package": {
        "ecosystem": "Raspberry Pi OS:12",
        "name": "linux"
      },
      "versions": [
        "1:6.1.47-1+rpt4"
      ],
      "ecosystem_specific": {
        "binaries": [
          {
            "binary_name": "linux-image-6.1.0-rpi3-rpi-2712",
            "binary_version": "1:6.1.47-1+rpt4"
          },
          {
            "binary_name": "linux-image-6.1.0-rpi3-rpi-v6",
            "binary_version": "1:6.1.47-1+rpt4"
          },
          {
            "binary_name": "linux-image-6.1.0-rpi3-rpi-v7",
            "binary_version": "1:6.1.47-1+rpt4"
          },
          {
            "binary_name": "linux-image-6.1.0-rpi3-rpi-v7l",
            "binary_version": "1:6.1.47-1+rpt4"
          },
          {
            "binary_name": "linux-image-6.1.0-rpi3-rpi-v8",
            "binary_version": "1:6.1.47-1+rpt4"
          }
        ]
      },
      "database_specific": {
        "per_version": {
          "1:6.1.47-1+rpt4": {
            "linux_commit": "655fc658a15ae7a6f37103754adb39ba52a9a14e",
            "upstream_version": "6.1.47",
            "assessment": "commit ancestry (strak rule)",
            "upstream_fixed": null,
            "compiled_filter": false
          }
        }
      }
    }
  ],
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://www.cve.org/CVERecord?id=CVE-2026-89601"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/5efb3350230fef831167740e69e6db2861101186"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/356984d1a5c32e94810cbb6c8dc7d8ff2d4d919a"
    }
  ],
  "database_specific": {
    "source": "https://git.kernel.org/pub/scm/linux/security/vulns.git",
    "upstream_modified": "2026-09-13T06:49:25Z",
    "severity": {
      "source": "kernel CNA",
      "cvss_v3_score": 8.8,
      "cvss_v3_severity": "High",
      "scenario": "AV:N - nfsd SETATTR/SETXATTR and ksmbd truncate/EA handling call notify_change()/vfs_setxattr() into ext2_setattr()/ext2_xattr_set2() when an ext2 volume is NFS/SMB-exported (NAS and embedded USB-storage appliances, including sync-mounted disks); a remote client with write access therefore reaches the lost IS_SYNC inode update without a local shell.\nAC:L - The attacker can put the target in the IS_SYNC state at will (owner-writable FS_IOC_SETFLAGS/chattr +S, or a sync-mounted export) and then deterministically truncate or set/remove xattrs; sync_inode_metadata() skips a clean inode every time, with no race or memory layout outside attacker control.\nPR:L - The path is gated only by ordinary VFS write checks (inode_permission MAY_WRITE, xattr_permission for user.* xattrs, or inode_owner_or_capable() for FS_IOC_SETFLAGS); an unprivileged local user or a low-privilege NFS/SMB client with write access to a file suffices, and no init-namespace capability is required.\nUI:N - The attacker performs every step themselves (setting the SYNC flag, truncating or changing xattrs/ACLs, and driving inode reclaim); no separate victim action such as mounting a crafted image is required.\nS:U - Impact is confined to filesystem metadata and file data on the host that mounted the ext2 volume; this is not a VM escape, IOMMU bypass, or other cross-authority boundary.\nC:H - Removing the last xattr on an IS_SYNC inode frees the EA block without persisting i_file_acl=0, leaving a dangling on-disk pointer; after that block is reused, ext2_xattr_get() memcpy()s its contents to userspace when the header looks valid, so the filesystem-level UAF discloses other files' data.\nI:H - The lost i_file_acl update leaves the on-disk inode pointing at a freed EA block, so a later in-place xattr write can corrupt whatever reused that block, including other users' files and metadata; intended-durable IS_SYNC size/ctime/ACL updates can also fail to land, corrupting filesystem state.\nA:H - Reloading the stale i_file_acl and reading an invalid reused EA block calls ext2_error(), which panics on errors=panic and remounts the volume read-only on errors=remount-ro; the resulting filesystem inconsistency can also render files and the volume unavailable."
    }
  }
}