Vulnerability-Lookup

GHSA-QW38-PPMM-FJPW

Vulnerability from github – Published: 2024-10-21 21:30 – Updated: 2024-10-25 21:31

Details

In the Linux kernel, the following vulnerability has been resolved:

perf: Fix perf_pending_task() UaF

Per syzbot it is possible for perf_pending_task() to run after the event is free()'d. There are two related but distinct cases:

the task_work was already queued before destroying the event;
destroying the event itself queues the task_work.

The first cannot be solved using task_work_cancel() since perf_release() itself might be called from a task_work (____fput), which means the current->task_works list is already empty and task_work_cancel() won't be able to find the perf_pending_task() entry.

The simplest alternative is extending the perf_event lifetime to cover the task_work.

The second is just silly, queueing a task_work while you know the event is going away makes no sense and is easily avoided by re-arranging how the event is marked STATE_DEAD and ensuring it goes through STATE_OFF on the way down.

Severity

7.8 (High)


                  
                    CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

Show details on source website

JSON

To clipboard

{
  "affected": [],
  "aliases": [
    "CVE-2022-48950"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-416"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2024-10-21T20:15:06Z",
    "severity": "HIGH"
  },
  "details": "In the Linux kernel, the following vulnerability has been resolved:\n\nperf: Fix perf_pending_task() UaF\n\nPer syzbot it is possible for perf_pending_task() to run after the\nevent is free()\u0027d. There are two related but distinct cases:\n\n - the task_work was already queued before destroying the event;\n - destroying the event itself queues the task_work.\n\nThe first cannot be solved using task_work_cancel() since\nperf_release() itself might be called from a task_work (____fput),\nwhich means the current-\u003etask_works list is already empty and\ntask_work_cancel() won\u0027t be able to find the perf_pending_task()\nentry.\n\nThe simplest alternative is extending the perf_event lifetime to cover\nthe task_work.\n\nThe second is just silly, queueing a task_work while you know the\nevent is going away makes no sense and is easily avoided by\nre-arranging how the event is marked STATE_DEAD and ensuring it goes\nthrough STATE_OFF on the way down.",
  "id": "GHSA-qw38-ppmm-fjpw",
  "modified": "2024-10-25T21:31:26Z",
  "published": "2024-10-21T21:30:50Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2022-48950"
    },
    {
      "type": "WEB",
      "url": "https://git.kernel.org/stable/c/517e6a301f34613bff24a8e35b5455884f2d83d8"
    },
    {
      "type": "WEB",
      "url": "https://git.kernel.org/stable/c/78e1317a174edbfd1182599bf76c092a2877672c"
    },
    {
      "type": "WEB",
      "url": "https://git.kernel.org/stable/c/8bffa95ac19ff27c8261904f89d36c7fcf215d59"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H",
      "type": "CVSS_V3"
    }
  ]
}

CVE-2022-48950 (GCVE-0-2022-48950)

Vulnerability from cvelistv5 – Published: 2024-10-21 20:05 – Updated: 2026-08-05 08:53

Title

perf: Fix perf_pending_task() UaF

Summary

In the Linux kernel, the following vulnerability has been resolved: perf: Fix perf_pending_task() UaF Per syzbot it is possible for perf_pending_task() to run after the event is free()'d. There are two related but distinct cases: - the task_work was already queued before destroying the event; - destroying the event itself queues the task_work. The first cannot be solved using task_work_cancel() since perf_release() itself might be called from a task_work (____fput), which means the current->task_works list is already empty and task_work_cancel() won't be able to find the perf_pending_task() entry. The simplest alternative is extending the perf_event lifetime to cover the task_work. The second is just silly, queueing a task_work while you know the event is going away makes no sense and is easily avoided by re-arranging how the event is marked STATE_DEAD and ensuring it goes through STATE_OFF on the way down.

Severity

7.8 (High)


                        
                          CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

SSVC

Exploitation: none Automatable: no Technical Impact: partial

CISA Coordinator · CISA-ADP (v2.0.3)

Decision recorded 2024-10-22 13:21 UTC

Assigner

Linux mitre root

References

3 references

URL	Tags
https://git.kernel.org/stable/c/8bffa95ac19ff27c8…
https://git.kernel.org/stable/c/78e1317a174edbfd1…
https://git.kernel.org/stable/c/517e6a301f34613bf…

Impacted products

2 products

Vendor	Product	Version
Linux	Linux	Affected: ca7b0a10287e2733bdafb01ef0d4038536625fe3 , < 8bffa95ac19ff27c8261904f89d36c7fcf215d59 (git) Affected: 078c12ccf1fb943cc18c84894c76113dc89e5975 , < 78e1317a174edbfd1182599bf76c092a2877672c (git) Affected: ca6c21327c6af02b7eec31ce4b9a740a18c6c13f , < 517e6a301f34613bff24a8e35b5455884f2d83d8 (git)
Linux	Linux	Affected: 5.15.77 , < 5.15.84 (semver) Affected: 6.0.7 , < 6.0.14 (semver)

Show details on NVD website

JSON

To clipboard

{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2022-48950",
                "options": [
                  {
                    "Exploitation": "none"
                  },
                  {
                    "Automatable": "no"
                  },
                  {
                    "Technical Impact": "partial"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2024-10-22T13:21:45.788376Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2024-10-22T13:28:40.788Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "defaultStatus": "unaffected",
          "product": "Linux",
          "programFiles": [
            "kernel/events/core.c"
          ],
          "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
          "vendor": "Linux",
          "versions": [
            {
              "lessThan": "8bffa95ac19ff27c8261904f89d36c7fcf215d59",
              "status": "affected",
              "version": "ca7b0a10287e2733bdafb01ef0d4038536625fe3",
              "versionType": "git"
            },
            {
              "lessThan": "78e1317a174edbfd1182599bf76c092a2877672c",
              "status": "affected",
              "version": "078c12ccf1fb943cc18c84894c76113dc89e5975",
              "versionType": "git"
            },
            {
              "lessThan": "517e6a301f34613bff24a8e35b5455884f2d83d8",
              "status": "affected",
              "version": "ca6c21327c6af02b7eec31ce4b9a740a18c6c13f",
              "versionType": "git"
            }
          ]
        },
        {
          "defaultStatus": "unaffected",
          "product": "Linux",
          "programFiles": [
            "kernel/events/core.c"
          ],
          "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
          "vendor": "Linux",
          "versions": [
            {
              "lessThan": "5.15.84",
              "status": "affected",
              "version": "5.15.77",
              "versionType": "semver"
            },
            {
              "lessThan": "6.0.14",
              "status": "affected",
              "version": "6.0.7",
              "versionType": "semver"
            }
          ]
        }
      ],
      "cpeApplicability": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
                  "versionEndExcluding": "5.15.84",
                  "versionStartIncluding": "5.15.77",
                  "vulnerable": true
                },
                {
                  "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
                  "versionEndExcluding": "6.0.14",
                  "versionStartIncluding": "6.0.7",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "In the Linux kernel, the following vulnerability has been resolved:\n\nperf: Fix perf_pending_task() UaF\n\nPer syzbot it is possible for perf_pending_task() to run after the\nevent is free()\u0027d. There are two related but distinct cases:\n\n - the task_work was already queued before destroying the event;\n - destroying the event itself queues the task_work.\n\nThe first cannot be solved using task_work_cancel() since\nperf_release() itself might be called from a task_work (____fput),\nwhich means the current-\u003etask_works list is already empty and\ntask_work_cancel() won\u0027t be able to find the perf_pending_task()\nentry.\n\nThe simplest alternative is extending the perf_event lifetime to cover\nthe task_work.\n\nThe second is just silly, queueing a task_work while you know the\nevent is going away makes no sense and is easily avoided by\nre-arranging how the event is marked STATE_DEAD and ensuring it goes\nthrough STATE_OFF on the way down."
        }
      ],
      "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 vulnerable path is reached via the local perf_event_open(2) syscall and subsequent event close/release; it is not reachable from the network stack or remote protocols.\nAC:L - An attacker fully controls event creation, overflow/sigtrap triggering, and close/destroy; destroying an event with pending_sigtrap itself queues perf_pending_task then frees the object, so the UAF does not depend on conditions outside attacker influence.\nPR:L - Unprivileged users can open sampling events with attr.sigtrap on their own tasks because ptrace_may_access succeeds for same_thread_group self-access, and default perf_event_paranoid still allows self-monitoring with exclude_kernel.\nUI:N - Exploitation requires only the attacker\u0027s own syscalls (perf_event_open, activity to overflow, close/exit); no victim user action is needed.\nS:U - Impact is confined to the local kernel privilege boundary (standard local privilege escalation); this is not a VM escape, IOMMU bypass, or other cross-authority boundary.\nC:H - This is a use-after-free of a large perf_event slab object (~1.4KB); after free, perf_pending_task/perf_sigtrap dereference attacker-sprayable memory, enabling arbitrary kernel read primitives.\nI:H - The same UAF provides a write/control-flow hijack surface via heap reuse of the freed perf_event (including callback_head and ctx pointer use), enabling arbitrary kernel write and code execution.\nA:H - The UAF reliably causes kernel oops/panic (demonstrated by syzbot KASAN crashes in task_work_run/perf_pending_task), fully impacting system availability."
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-08-05T08:53:05.541Z",
        "orgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
        "shortName": "Linux"
      },
      "references": [
        {
          "url": "https://git.kernel.org/stable/c/8bffa95ac19ff27c8261904f89d36c7fcf215d59"
        },
        {
          "url": "https://git.kernel.org/stable/c/78e1317a174edbfd1182599bf76c092a2877672c"
        },
        {
          "url": "https://git.kernel.org/stable/c/517e6a301f34613bff24a8e35b5455884f2d83d8"
        }
      ],
      "title": "perf: Fix perf_pending_task() UaF",
      "x_generator": {
        "engine": "bippy-1.2.0"
      }
    }
  },
  "cveMetadata": {
    "assignerOrgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
    "assignerShortName": "Linux",
    "cveId": "CVE-2022-48950",
    "datePublished": "2024-10-21T20:05:38.440Z",
    "dateReserved": "2024-08-22T01:27:53.625Z",
    "dateUpdated": "2026-08-05T08:53:05.541Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2"
}

Sightings

Author	Source	Type	Date	Other

Nomenclature

Seen: The vulnerability was mentioned, discussed, or observed by the user.
Confirmed: The vulnerability has been validated from an analyst's perspective.
Published Proof of Concept: A public proof of concept is available for this vulnerability.
Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
Not confirmed: The user expressed doubt about the validity of the vulnerability.
Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.

The MITRE ATT&CK techniques below are AI-generated suggestions, inferred from the description of the vulnerability by the CIRCL/vulnerability-attack-technique-classification-roberta-base model, served locally by ML-Gateway. They have not been verified by an analyst and are provided for guidance only.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.

Related by attack behaviour

Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.

Action not permitted

GHSA-QW38-PPMM-FJPW

CVE-2022-48950 (GCVE-0-2022-48950)

Tags

Sightings

Nomenclature

Related by attack behaviour