In the Linux kernel, the following vulnerability has been resolved: configfs: unhash the dentry...

zetlyn/cve-ghsa vulnerability ghsa GHSA-gmrw-x43x-r9qf cve CVE-2026-97949 known 2026-09-25

https://github.com/advisories/GHSA-gmrw-x43x-r9qf

Properties

severityunknown
receipt
Source
GitHub advisories
Its words
unknown
Read by
field:severity
Said since
2026-09-28 11:44 UTC
Last answered
2026-10-06 13:46 UTC
Original
open at the source
What the source handed over
{
  "credits": [],
  "cve_id": "CVE-2026-97949",
  "cvss": {
    "score": null,
    "vector_string": null
  },
  "cvss_severities": {
    "cvss_v3": {
      "score": 0.0,
      "vector_string": null
    },
    "cvss_v4": {
      "score": 0.0,
      "vector_string": null
    }
  },
  "cwes": [],
  "description": "In the Linux kernel, the following vulnerability has been resolved:\n\nconfigfs: unhash the dentry before dropping the item in rmdir\n\nconfigfs_get_config_item() treats a hashed dentry as proof that\nsd->s_element is a live config_item.  configfs_rmdir() breaks that:\nsimple_rmdir() leaves the dentry hashed, the last reference to the item is\ndropped right after, and the dentry is only unhashed by d_delete() once\n->rmdir() has returned.  configfs_symlink() resolves its target holding no\nlock on it, so get_target() can land in that window:\n\n  BUG: KASAN: slab-use-after-free in config_item_get+0x26/0x90\n   get_target fs/configfs/symlink.c:128 [inline]\n   configfs_symlink+0x4ab/0x1030 fs/configfs/symlink.c:185\n\nUnhash in configfs_remove_dir(), while the item is still guaranteed to be\nthere.  A reference obtained just before that stays harmless, as\ncreate_link() rechecks CONFIGFS_USET_DROPPING, already set by\nconfigfs_detach_prep().  Both configfs_unregister_subsystem() paths\nd_drop() after detaching, so this only makes rmdir match them.",
  "epss": {
    "percentage": 0.00166,
    "percentile": 0.05204
  },
  "ghsa_id": "GHSA-gmrw-x43x-r9qf",
  "github_reviewed_at": null,
  "html_url": "https://github.com/advisories/GHSA-gmrw-x43x-r9qf",
  "identifiers": [
    {
      "type": "GHSA",
      "value": "GHSA-gmrw-x43x-r9qf"
    },
    {
      "type": "CVE",
      "value": "CVE-2026-97949"
    }
  ],
  "nvd_published_at": "2026-09-25T11:17:22Z",
  "published_at": "2026-09-25T12:31:29Z",
  "references": [
    "https://nvd.nist.gov/vuln/detail/CVE-2026-97949",
    "https://git.kernel.org/stable/c/5d7cbfb65e47cc71d5d94d87d5d0d1679080f3eb",
    "https://git.kernel.org/stable/c/925d8564f1b2d9b46c5ab63b9bc942cea006da9b",
    "https://git.kernel.org/stable/c/f06c2d26d1999d37e93299db0ecead04ca7d0b9f",
    "https://github.com/advisories/GHSA-gmrw-x43x-r9qf"
  ],
  "repository_advisory_url": null,
  "severity": "unknown",
  "source_code_location": "",
  "summary": "In the Linux kernel, the following vulnerability has been resolved:\n\nconfigfs: unhash the dentry...",
  "type": "unreviewed",
  "updated_at": "2026-09-25T12:31:38Z",
  "url": "https://api.github.com/advisories/GHSA-gmrw-x43x-r9qf",
  "vulnerabilities": [],
  "withdrawn_at": null
}

Text

In the Linux kernel, the following vulnerability has been resolved: configfs: unhash the dentry... In the Linux kernel, the following vulnerability has been resolved: configfs: unhash the dentry before dropping the item in rmdir configfs_get_config_item() treats a hashed dentry as proof that sd->s_element is a live config_item. configfs_rmdir() breaks that: simple_rmdir() leaves the dentry hashed, the last reference to the item is dropped right after, and the dentry is only unhashed by d_delete() once ->rmdir() has returned. configfs_symlink() resolves its target holding no lock on it, so get_target() can land in that window: BUG: KASAN: slab-use-after-free in config_item_get+0x26/0x90 get_target fs/configfs/symlink.c:128 [inline] configfs_symlink+0x4ab/0x1030 fs/configfs/symlink.c:185 Unhash in configfs_remove_dir(), while the item is still guaranteed to be there. A reference obtained just before that stays harmless, as create_link() rechecks CONFIGFS_USET_DROPPING, already set by configfs_detach_prep(). Both configfs_unregister_subsystem() paths d_drop() after detaching, so this only makes rmdir match them.