Vulnerability-Lookup

GHSA-437R-PPXQ-JXXH

Vulnerability from github – Published: 2026-08-15 15:30 – Updated: 2026-08-17 06:33

Details

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

afs: Fix UAF when sending a message

In afs_make_call(), there's a race with async call reception and destruction. If a call is dispatched that doesn't have call->write_iter set (used to specify the data content for FS.StoreData), then the first rxrpc_kernel_send_data() will not set MSG_MORE in the msghdr.

Once rxrpc_send_data() queues the last request packet, the response could come in at any time and cause the call to be completed and put. However, afs_make_call() will look at the call again to see it ->write_iter should be handled - something it's only allowed to do if it has its own ref on the call. Whilst this is the case for synchronous calls, it isn't true for async calls such as FS.FetchData.

There's also a potential UAF in afs_make_call() in the event that an asynchronous call is being sent, but the call fails in some way (e.g. it gets aborted from the server). The problem there is that afs_make_call() tries to abort a call if the rxrpc send fails, but the asynchronous notification from rxrpc may have caused the afs_call to be torn down.

generic/650 plays games with randomly taking CPUs offline, and can interject a significant delay such that the call is deallocated before afs_make_call() gets to check call->write_iter - and a UAF ensues (caught by KASAN).

BUG: KASAN: slab-use-after-free in afs_make_call+0x1c90/0x2210 [kafs] Read of size 8 at addr ffff888035e050e8 by task fsstress/1409

Fix this by making afs_make_op_call() give the op->call its own ref rather than transferring the caller's ref to it and then dropping the ref when afs_make_call() returns.

This also means that the afs_make_call() func never loses its ref on the call now.

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-2026-74506"
  ],
  "database_specific": {
    "cwe_ids": [],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2026-08-15T13:17:55Z",
    "severity": "HIGH"
  },
  "details": "In the Linux kernel, the following vulnerability has been resolved:\n\nafs: Fix UAF when sending a message\n\nIn afs_make_call(), there\u0027s a race with async call reception and\ndestruction.  If a call is dispatched that doesn\u0027t have call-\u003ewrite_iter\nset (used to specify the data content for FS.StoreData), then the first\nrxrpc_kernel_send_data() will not set MSG_MORE in the msghdr.\n\nOnce rxrpc_send_data() queues the last request packet, the response could\ncome in at any time and cause the call to be completed and put.  However,\nafs_make_call() will look at the call again to see it -\u003ewrite_iter should\nbe handled - something it\u0027s only allowed to do if it has its own ref on the\ncall.  Whilst this is the case for synchronous calls, it isn\u0027t true for\nasync calls such as FS.FetchData.\n\nThere\u0027s also a potential UAF in afs_make_call() in the event that an\nasynchronous call is being sent, but the call fails in some way (e.g. it\ngets aborted from the server).  The problem there is that afs_make_call()\ntries to abort a call if the rxrpc send fails, but the asynchronous\nnotification from rxrpc may have caused the afs_call to be torn down.\n\ngeneric/650 plays games with randomly taking CPUs offline, and can\ninterject a significant delay such that the call is deallocated before\nafs_make_call() gets to check call-\u003ewrite_iter - and a UAF ensues (caught\nby KASAN).\n\n   BUG: KASAN: slab-use-after-free in afs_make_call+0x1c90/0x2210 [kafs]\n   Read of size 8 at addr ffff888035e050e8 by task fsstress/1409\n\nFix this by making afs_make_op_call() give the op-\u003ecall its own ref rather\nthan transferring the caller\u0027s ref to it and then dropping the ref when\nafs_make_call() returns.\n\nThis also means that the afs_make_call() func never loses its ref on the\ncall now.",
  "id": "GHSA-437r-ppxq-jxxh",
  "modified": "2026-08-17T06:33:50Z",
  "published": "2026-08-15T15:30:34Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-74506"
    },
    {
      "type": "WEB",
      "url": "https://git.kernel.org/stable/c/4af1ec68d54b3871155914d584fb10669c41a861"
    },
    {
      "type": "WEB",
      "url": "https://git.kernel.org/stable/c/c0d3b81f703b2a9e37fe1347610a50cdf0078c27"
    },
    {
      "type": "WEB",
      "url": "https://git.kernel.org/stable/c/daaa726b14fc3026a6b328614d312b698f62f391"
    }
  ],
  "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-2026-74506 (GCVE-0-2026-74506)

Vulnerability from cvelistv5 – Published: 2026-08-15 12:27 – Updated: 2026-08-17 05:47

Title

afs: Fix UAF when sending a message

Summary

In the Linux kernel, the following vulnerability has been resolved: afs: Fix UAF when sending a message In afs_make_call(), there's a race with async call reception and destruction. If a call is dispatched that doesn't have call->write_iter set (used to specify the data content for FS.StoreData), then the first rxrpc_kernel_send_data() will not set MSG_MORE in the msghdr. Once rxrpc_send_data() queues the last request packet, the response could come in at any time and cause the call to be completed and put. However, afs_make_call() will look at the call again to see it ->write_iter should be handled - something it's only allowed to do if it has its own ref on the call. Whilst this is the case for synchronous calls, it isn't true for async calls such as FS.FetchData. There's also a potential UAF in afs_make_call() in the event that an asynchronous call is being sent, but the call fails in some way (e.g. it gets aborted from the server). The problem there is that afs_make_call() tries to abort a call if the rxrpc send fails, but the asynchronous notification from rxrpc may have caused the afs_call to be torn down. generic/650 plays games with randomly taking CPUs offline, and can interject a significant delay such that the call is deallocated before afs_make_call() gets to check call->write_iter - and a UAF ensues (caught by KASAN). BUG: KASAN: slab-use-after-free in afs_make_call+0x1c90/0x2210 [kafs] Read of size 8 at addr ffff888035e050e8 by task fsstress/1409 Fix this by making afs_make_op_call() give the op->call its own ref rather than transferring the caller's ref to it and then dropping the ref when afs_make_call() returns. This also means that the afs_make_call() func never loses its ref on the call now.

Severity

7.8 (High)


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

Assigner

Linux mitre root

References

3 references

URL	Tags
https://git.kernel.org/stable/c/c0d3b81f703b2a9e3…
https://git.kernel.org/stable/c/daaa726b14fc3026a…
https://git.kernel.org/stable/c/4af1ec68d54b38711…

Impacted products

2 products

Vendor	Product	Version
Linux	Linux	Affected: e49c7b2f6de7ff81ca34c56e4eeb4fa740c099f2 , < c0d3b81f703b2a9e37fe1347610a50cdf0078c27 (git) Affected: e49c7b2f6de7ff81ca34c56e4eeb4fa740c099f2 , < daaa726b14fc3026a6b328614d312b698f62f391 (git) Affected: e49c7b2f6de7ff81ca34c56e4eeb4fa740c099f2 , < 4af1ec68d54b3871155914d584fb10669c41a861 (git)
Linux	Linux	Affected: 5.8 Unaffected: 0 , < 5.8 (semver) Unaffected: 6.18.44 , ≤ 6.18.* (semver) Unaffected: 7.1.8 , ≤ 7.1.* (semver) Unaffected: 7.2 , ≤ * (original_commit_for_fix)

Show details on NVD website

JSON

To clipboard

{
  "containers": {
    "cna": {
      "affected": [
        {
          "defaultStatus": "unaffected",
          "product": "Linux",
          "programFiles": [
            "fs/afs/internal.h"
          ],
          "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
          "vendor": "Linux",
          "versions": [
            {
              "lessThan": "c0d3b81f703b2a9e37fe1347610a50cdf0078c27",
              "status": "affected",
              "version": "e49c7b2f6de7ff81ca34c56e4eeb4fa740c099f2",
              "versionType": "git"
            },
            {
              "lessThan": "daaa726b14fc3026a6b328614d312b698f62f391",
              "status": "affected",
              "version": "e49c7b2f6de7ff81ca34c56e4eeb4fa740c099f2",
              "versionType": "git"
            },
            {
              "lessThan": "4af1ec68d54b3871155914d584fb10669c41a861",
              "status": "affected",
              "version": "e49c7b2f6de7ff81ca34c56e4eeb4fa740c099f2",
              "versionType": "git"
            }
          ]
        },
        {
          "defaultStatus": "affected",
          "product": "Linux",
          "programFiles": [
            "fs/afs/internal.h"
          ],
          "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
          "vendor": "Linux",
          "versions": [
            {
              "status": "affected",
              "version": "5.8"
            },
            {
              "lessThan": "5.8",
              "status": "unaffected",
              "version": "0",
              "versionType": "semver"
            },
            {
              "lessThanOrEqual": "6.18.*",
              "status": "unaffected",
              "version": "6.18.44",
              "versionType": "semver"
            },
            {
              "lessThanOrEqual": "7.1.*",
              "status": "unaffected",
              "version": "7.1.8",
              "versionType": "semver"
            },
            {
              "lessThanOrEqual": "*",
              "status": "unaffected",
              "version": "7.2",
              "versionType": "original_commit_for_fix"
            }
          ]
        }
      ],
      "cpeApplicability": [
        {
          "nodes": [
            {
              "cpeMatch": [
                {
                  "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
                  "versionEndExcluding": "6.18.44",
                  "versionStartIncluding": "5.8",
                  "vulnerable": true
                },
                {
                  "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
                  "versionEndExcluding": "7.1.8",
                  "versionStartIncluding": "5.8",
                  "vulnerable": true
                },
                {
                  "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*",
                  "versionEndExcluding": "7.2",
                  "versionStartIncluding": "5.8",
                  "vulnerable": true
                }
              ],
              "negate": false,
              "operator": "OR"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "In the Linux kernel, the following vulnerability has been resolved:\n\nafs: Fix UAF when sending a message\n\nIn afs_make_call(), there\u0027s a race with async call reception and\ndestruction.  If a call is dispatched that doesn\u0027t have call-\u003ewrite_iter\nset (used to specify the data content for FS.StoreData), then the first\nrxrpc_kernel_send_data() will not set MSG_MORE in the msghdr.\n\nOnce rxrpc_send_data() queues the last request packet, the response could\ncome in at any time and cause the call to be completed and put.  However,\nafs_make_call() will look at the call again to see it -\u003ewrite_iter should\nbe handled - something it\u0027s only allowed to do if it has its own ref on the\ncall.  Whilst this is the case for synchronous calls, it isn\u0027t true for\nasync calls such as FS.FetchData.\n\nThere\u0027s also a potential UAF in afs_make_call() in the event that an\nasynchronous call is being sent, but the call fails in some way (e.g. it\ngets aborted from the server).  The problem there is that afs_make_call()\ntries to abort a call if the rxrpc send fails, but the asynchronous\nnotification from rxrpc may have caused the afs_call to be torn down.\n\ngeneric/650 plays games with randomly taking CPUs offline, and can\ninterject a significant delay such that the call is deallocated before\nafs_make_call() gets to check call-\u003ewrite_iter - and a UAF ensues (caught\nby KASAN).\n\n   BUG: KASAN: slab-use-after-free in afs_make_call+0x1c90/0x2210 [kafs]\n   Read of size 8 at addr ffff888035e050e8 by task fsstress/1409\n\nFix this by making afs_make_op_call() give the op-\u003ecall its own ref rather\nthan transferring the caller\u0027s ref to it and then dropping the ref when\nafs_make_call() returns.\n\nThis also means that the afs_make_call() func never loses its ref on the\ncall now."
        }
      ],
      "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 UAF is in kAFS client afs_make_call() during outbound FS.FetchData/YFS.FetchData RPCs, reached via local VFS read or readahead on an afs-mounted file (read_iter \u2192 afs_issue_read \u2192 afs_fs_fetch_data \u2192 afs_make_op_call), not from unsolicited remote packet processing alone.\nAC:L - The attacker controls both sides of the race: concurrent reads/readahead (as in xfstests generic/650 fsstress) while a fast or malicious AFS server reply completes and frees the async afs_call before afs_make_call() re-checks call-\u003ewrite_iter; no uncontrolled memory layout is required.\nPR:L - Any unprivileged local user with read permission on an already-mounted AFS volume can trigger async FetchData; mounting requires admin, but exercising the vulnerable read path needs only normal file access, matching the unprivileged fsstress reproducer.\nUI:N - No victim interaction is required; the attacker or local stress tool initiates the reads that dispatch the vulnerable async RPC, and AFS being mounted is an environmental precondition rather than interactive user action.\nS:U - The slab use-after-free in the kernel AFS client may enable local privilege escalation but does not cross a VM, container runtime, or IOMMU security boundary.\nC:H - KASAN reports slab-use-after-free with an 8-byte read of call-\u003ewrite_iter after async completion frees the afs_call; UAF over attacker-influenceable freed slab memory enables arbitrary kernel memory read primitives.\nI:H - The same use-after-free in afs_make_call() can be leveraged via heap grooming and spraying to obtain arbitrary kernel write or control-flow hijack, not merely a benign crash.\nA:H - The bug reliably triggers KASAN failures and kernel oops during normal async FetchData I/O, and UAF races can panic or wedge the system even before full exploit development."
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-08-17T05:47:55.797Z",
        "orgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
        "shortName": "Linux"
      },
      "references": [
        {
          "url": "https://git.kernel.org/stable/c/c0d3b81f703b2a9e37fe1347610a50cdf0078c27"
        },
        {
          "url": "https://git.kernel.org/stable/c/daaa726b14fc3026a6b328614d312b698f62f391"
        },
        {
          "url": "https://git.kernel.org/stable/c/4af1ec68d54b3871155914d584fb10669c41a861"
        }
      ],
      "title": "afs: Fix UAF when sending a message",
      "x_generator": {
        "engine": "bippy-1.2.0"
      }
    }
  },
  "cveMetadata": {
    "assignerOrgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67",
    "assignerShortName": "Linux",
    "cveId": "CVE-2026-74506",
    "datePublished": "2026-08-15T12:27:30.057Z",
    "dateReserved": "2026-08-15T05:44:03.908Z",
    "dateUpdated": "2026-08-17T05:47:55.797Z",
    "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.

Detection rules are retrieved from Rulezet.

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.

Action not permitted

GHSA-437R-PPXQ-JXXH

CVE-2026-74506 (GCVE-0-2026-74506)

Tags

Sightings

Nomenclature