{
  "schema_version": "1.6.1",
  "id": "RPI-CVE-2026-89526",
  "modified": "2026-10-09T11:39:44Z",
  "upstream": [
    "CVE-2026-89526"
  ],
  "published": "2026-09-11T19:47:07Z",
  "summary": "svcrdma: Validate Read chunk positions before reconstruction",
  "details": "In the Linux kernel, the following vulnerability has been resolved:\n\nsvcrdma: Validate Read chunk positions before reconstruction\n\nThe RPC/RDMA Read chunk position field is supplied by the remote\nclient and stored verbatim in the parsed chunk list.\nxdr_count_read_segments() checks only 4-byte alignment; it never\ncompares the position against the received inline body length.\n\nIn the single-chunk path, svc_rdma_read_complete_one() splits the\nhead and tail kvecs at ch_position. A position past the inline\nbody underflows the tail length, exposing adjacent slab memory to\nthe upper XDR decoder.\n\nIn the multi-chunk path, svc_rdma_read_multiple_chunks() computes\ngap lengths between chunks as unsigned subtractions from\nch_position. Overlapping Read chunks cause these subtractions to\nunderflow. A final position past the inline body likewise\nunderflows the trailing gap length. svc_rdma_copy_inline_range()\nthen copies past the receive buffer into request pages that are\nreturned to the client through the Reply channel.\n\nBound inline-range copies in svc_rdma_copy_inline_range() against\nthe decoded inline RPC body saved in rc_saved_arg. Reject a\nsingle Read chunk positioned beyond that body, and reject\nmulti-chunk lists where accumulated read bytes exceed the next\nchunk's position. Apply the same position and overlap checks in\nthe call-chunk interleaving path.",
  "severity": [
    {
      "type": "CVSS_V3",
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/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-2026-89526"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/5ab3f6d882fe07ae5e61d0bcfeea00b9409155c2"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/f84ec84d8d4bc65f9ae23372570349687f66fa39"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/577097455d084610fc31e91e6a61c5793b6f04ba"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/3779b7b9e7d1c8ba4738f9d327de3b0288cefe9b"
    }
  ],
  "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": 9.8,
      "cvss_v3_severity": "Critical",
      "scenario": "AV:N - svcrdma is nfsd's RPC-over-RDMA transport; a remote peer reaches Read-chunk reconstruction on an enabled NFS/RDMA listener over routable iWARP or RoCE, consistent with prior kernel CNA scores for this subsystem.\nAC:L - The attacker fully controls the wire-supplied Read-chunk position (only 4-byte alignment is checked) and can deterministically underflow gap/tail lengths or copy past the receive buffer, with no race or uncontrollable layout required.\nPR:N - svc_rdma_process_read_list() and svc_rdma_read_complete_one() run in svc_rdma_recvfrom() while assembling the transport message, before svc_process() performs RPC or NFS authentication, and RDMA CM accept has no credential check.\nUI:N - Sending a crafted RPC-over-RDMA message with a malicious Read list is sufficient; no victim mount, click, or other user action is required.\nS:U - The vulnerable NFS/RDMA service and the affected kernel memory belong to the same host kernel security authority; this is not a VM, IOMMU, or sandbox escape.\nC:H - A position past the inline body underflows the tail kvec so the XDR decoder reads adjacent slab, and overlapping chunks make svc_rdma_copy_inline_range() memcpy past the receive buffer into request pages returned on the Reply channel, disclosing kernel memory.\nI:H - The reconstructed xdr_buf uses attacker-controlled head/tail lengths pointing into adjacent slab; XDR shift/copy helpers then memcpy through those kvecs, producing an out-of-bounds kernel write that can corrupt adjacent objects and hijack control flow.\nA:H - The unbounded inline copy and inflated kvec lengths walk unmapped memory or trip decoder/WARN paths, causing a kernel oops or panic that an unauthenticated peer can trigger repeatedly."
    }
  }
}