CVE-2025-21709 (GCVE-0-2025-21709)
Vulnerability from cvelistv5
Published
2025-02-27 02:07
Modified
2026-08-05 11:53
Severity ?
7.8 (High) - CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
Summary
In the Linux kernel, the following vulnerability has been resolved: kernel: be more careful about dup_mmap() failures and uprobe registering If a memory allocation fails during dup_mmap(), the maple tree can be left in an unsafe state for other iterators besides the exit path. All the locks are dropped before the exit_mmap() call (in mm/mmap.c), but the incomplete mm_struct can be reached through (at least) the rmap finding the vmas which have a pointer back to the mm_struct. Up to this point, there have been no issues with being able to find an mm_struct that was only partially initialised. Syzbot was able to make the incomplete mm_struct fail with recent forking changes, so it has been proven unsafe to use the mm_struct that hasn't been initialised, as referenced in the link below. Although 8ac662f5da19f ("fork: avoid inappropriate uprobe access to invalid mm") fixed the uprobe access, it does not completely remove the race. This patch sets the MMF_OOM_SKIP to avoid the iteration of the vmas on the oom side (even though this is extremely unlikely to be selected as an oom victim in the race window), and sets MMF_UNSTABLE to avoid other potential users from using a partially initialised mm_struct. When registering vmas for uprobe, skip the vmas in an mm that is marked unstable. Modifying a vma in an unstable mm may cause issues if the mm isn't fully initialised.
References
Impacted products
Vendor Product Version
Linux Linux Version: d2406291483775ecddaee929231a39c70c08fda2
Version: d2406291483775ecddaee929231a39c70c08fda2
Version: d2406291483775ecddaee929231a39c70c08fda2
 Create a notification for this product.
Show details on NVD website

To clipboard 

{
  "containers": {
    "cna": {
      "affected": [
        {
          "defaultStatus": "unaffected",
          "product": "Linux",
          "programFiles": [
            "kernel/events/uprobes.c",
            "kernel/fork.c"
          ],
          "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
          "vendor": "Linux",
          "versions": [
            {
              "lessThan": "74c2471eb891a7dcb3874b21c106cda75f52be30",
              "status": "affected",
              "version": "d2406291483775ecddaee929231a39c70c08fda2",
              "versionType": "git"
            },
            {
              "lessThan": "da139948aeda677ac09cc0e7d837f8a314de7d55",
              "status": "affected",
              "version": "d2406291483775ecddaee929231a39c70c08fda2",
              "versionType": "git"
            },
            {
              "lessThan": "64c37e134b120fb462fb4a80694bfb8e7be77b14",
              "status": "affected",
              "version": "d2406291483775ecddaee929231a39c70c08fda2",
              "versionType": "git"
            }
          ]
        },
        {
          "defaultStatus": "affected",
          "product": "Linux",
          "programFiles": [
            "kernel/events/uprobes.c",
            "kernel/fork.c"
          ],
          "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
          "vendor": "Linux",
          "versions": [
            {
              "status": "affected",
              "version": "6.8"
            },
            {
              "lessThan": "6.8",
              "status": "unaffected",
              "version": "0",
              "versionType": "semver"
            },
            {
              "lessThanOrEqual": "6.12.*",
              "status": "unaffected",
              "version": "6.12.83",
              "versionType": "semver"
            },
            {
              "lessThanOrEqual": "6.13.*",
              "status": "unaffected",
              "version": "6.13.2",
              "versionType": "semver"
            },
            {
              "lessThanOrEqual": "*",
              "status": "unaffected",
              "version": "6.14",
              "versionType": "original_commit_for_fix"
            }
          ]
        }
      ],
      "cpeApplicability": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
                  "versionEndExcluding": "6.12.83",
                  "versionStartIncluding": "6.8",
                  "vulnerable": true
                },
                {
                  "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
                  "versionEndExcluding": "6.13.2",
                  "versionStartIncluding": "6.8",
                  "vulnerable": true
                },
                {
                  "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
                  "versionEndExcluding": "6.14",
                  "versionStartIncluding": "6.8",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "In the Linux kernel, the following vulnerability has been resolved:\n\nkernel: be more careful about dup_mmap() failures and uprobe registering\n\nIf a memory allocation fails during dup_mmap(), the maple tree can be left\nin an unsafe state for other iterators besides the exit path.  All the\nlocks are dropped before the exit_mmap() call (in mm/mmap.c), but the\nincomplete mm_struct can be reached through (at least) the rmap finding\nthe vmas which have a pointer back to the mm_struct.\n\nUp to this point, there have been no issues with being able to find an\nmm_struct that was only partially initialised.  Syzbot was able to make\nthe incomplete mm_struct fail with recent forking changes, so it has been\nproven unsafe to use the mm_struct that hasn\u0027t been initialised, as\nreferenced in the link below.\n\nAlthough 8ac662f5da19f (\"fork: avoid inappropriate uprobe access to\ninvalid mm\") fixed the uprobe access, it does not completely remove the\nrace.\n\nThis patch sets the MMF_OOM_SKIP to avoid the iteration of the vmas on the\noom side (even though this is extremely unlikely to be selected as an oom\nvictim in the race window), and sets MMF_UNSTABLE to avoid other potential\nusers from using a partially initialised mm_struct.\n\nWhen registering vmas for uprobe, skip the vmas in an mm that is marked\nunstable.  Modifying a vma in an unstable mm may cause issues if the mm\nisn\u0027t fully initialised."
        }
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "baseScore": 7.8,
            "baseSeverity": "HIGH",
            "vectorString": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H",
            "version": "3.1"
          },
          "scenarios": [
            {
              "lang": "en",
              "value": "AV:L - The corrupted mm_struct is produced by the fork()/clone() syscall path (dup_mmap()), which requires local execution on the target system. There is no remote or adjacent-network path to dup_mmap().\nAC:L - The attacker deterministically creates the unsafe mm by forking a process with many VMAs and delivering SIGKILL mid-dup_mmap() (the fatal_signal_pending() bail-out), or by forcing -ENOMEM under a memcg/RLIMIT_AS/overcommit limit, and can repeat this continuously to widen the window; the same attacker-generated memory pressure simultaneously activates the OOM killer/reaper that iterates the broken tree, so both sides of the race are attacker-driven.\nPR:L - Only the ability to call fork()/clone() and allocate memory is needed \u2014 any unprivileged local user or container process qualifies. No capability, namespace, or tracing privilege is required on the attacker\u0027s side.\nUI:N - The attacker triggers the failed fork and the resulting corrupted mm_struct entirely on their own; no victim action is required. The consuming iterators (uprobe registration, swapoff, OOM reaper) are background/system activity, not user interaction.\nS:U - The corruption and its consequences are confined to the kernel\u0027s own memory-management structures within the same security authority. No VM, IOMMU, or hypervisor boundary is crossed.\nC:H - The partially-initialised mm\u0027s maple tree hands out an internal XA_ZERO_ENTRY (0x406) as a struct vm_area_struct * and, past that point, live pointers to the parent process\u0027s VMAs, so iterators read through invalid pointers and operate with mm != vma-\u003evm_mm \u2014 a cross-address-space mismatch that lets page-table walks touch another process\u0027s memory; a sibling CLONE_VM thread munmap()ing during the unlocked window turns those entries into dangling freed-object pointers, giving a use-after-free read primitive.\nI:H - install_breakpoint()/uprobe_write_opcode() modifies VMAs and page tables of an mm that is not fully initialised (exactly what the fix\u0027s check_stable_address_space() gate prevents), and the OOM reaper path guarded by the new MMF_OOM_SKIP would unmap the parent\u0027s VMAs through the child\u0027s mmu_gather, flushing the wrong mm\u0027s TLB and leaving stale writable mappings to freed pages \u2014 memory corruption usable for arbitrary write.\nA:H - The demonstrated failure is a kernel OOPS dereferencing XA_ZERO_ENTRY at a low address (the sibling swapoff fix records a fault at 0x446), fatal with panic_on_oops; in register_for_each_vma() the oops occurs while holding mmap_write_lock(mm) and percpu_down_write(\u0026dup_mmap_sem), which are never released, hanging every subsequent fork() on the system."
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-08-05T11:53:48.428Z",
        "orgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
        "shortName": "Linux"
      },
      "references": [
        {
          "url": "https://git.kernel.org/stable/c/74c2471eb891a7dcb3874b21c106cda75f52be30"
        },
        {
          "url": "https://git.kernel.org/stable/c/da139948aeda677ac09cc0e7d837f8a314de7d55"
        },
        {
          "url": "https://git.kernel.org/stable/c/64c37e134b120fb462fb4a80694bfb8e7be77b14"
        }
      ],
      "title": "kernel: be more careful about dup_mmap() failures and uprobe registering",
      "x_generator": {
        "engine": "bippy-1.2.0"
      }
    }
  },
  "cveMetadata": {
    "assignerOrgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
    "assignerShortName": "Linux",
    "cveId": "CVE-2025-21709",
    "datePublished": "2025-02-27T02:07:22.452Z",
    "dateReserved": "2024-12-29T08:45:45.752Z",
    "dateUpdated": "2026-08-05T11:53:48.428Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2"
}


Log in or create an account to share your comment.




Tags
Taxonomy of the tags.




Sightings

Author Source Type Date

Nomenclature

  • Seen: The vulnerability was mentioned, discussed, or seen somewhere by the user.
  • Confirmed: The vulnerability is confirmed from an analyst perspective.
  • Published Proof of Concept: A public proof of concept is available for this vulnerability.
  • Exploited: This vulnerability was exploited and seen by the user reporting the sighting.
  • Patched: This vulnerability was successfully patched by the user reporting the sighting.
  • Not exploited: This vulnerability was not exploited or seen by the user reporting the sighting.
  • Not confirmed: The user expresses doubt about the veracity of the vulnerability.
  • Not patched: This vulnerability was not successfully patched by the user reporting the sighting.