{
  "schema_version": "1.6.1",
  "id": "RPI-CVE-2026-80540",
  "modified": "2026-10-09T11:39:44Z",
  "upstream": [
    "CVE-2026-80540"
  ],
  "published": "2026-08-26T14:37:59Z",
  "summary": "drm/amdgpu: Fix UVD decode image min size calculation",
  "details": "In the Linux kernel, the following vulnerability has been resolved:\n\ndrm/amdgpu: Fix UVD decode image min size calculation\n\nThis needs to use pitch instead of width. Also reject pitch\nover 4096 to avoid overflow.\n\n(cherry picked from commit b41c8cb12e202b220353332ab87dc01a11f69304)",
  "severity": [
    {
      "type": "CVSS_V3",
      "score": "CVSS:3.1/AV:L/AC:L/PR:L/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": "6.1.184",
            "compiled_filter": false
          }
        }
      }
    }
  ],
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://www.cve.org/CVERecord?id=CVE-2026-80540"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/bc7397a033ac52f6d8c9bb6510d61694b2a3fce7"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/d058f7a6709441afe1784eecd8c0643dd84750bc"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/b7549e3f96c78921751c4b3e69af729662130d83"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/60539d517e8439621532d8c01091ac049c596b4b"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/271a7da84a6262a09de549912dcf6a749d169cb6"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/25ee120f3803ad9e416ef9f76f4c3234cc4d645b"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/5cbd8af02b0b9c8723fa30edcf6fccab5170af8d"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/b8bb9ba3f101a1b0011f785a577a4a0a38371174"
    }
  ],
  "database_specific": {
    "source": "https://git.kernel.org/pub/scm/linux/security/vulns.git",
    "upstream_modified": "2026-08-27T05:18:36Z",
    "severity": {
      "source": "kernel CNA",
      "cvss_v3_score": 7.8,
      "cvss_v3_severity": "High",
      "scenario": "AV:L - Reachable only via AMDGPU_CS ioctl on a local DRM render node (/dev/dri/renderD*), through amdgpu_cs_ioctl \u2192 amdgpu_cs_patch_ibs \u2192 amdgpu_uvd_ring_parse_cs \u2192 amdgpu_uvd_cs_pass2 \u2192 amdgpu_uvd_cs_msg \u2192 amdgpu_uvd_cs_msg_decode; not reachable from any network-facing kernel service.\nAC:L - An attacker with render-node access fully controls UVD decode message fields (width, height, pitch in msg[28], stream_type) in their own BO and can reliably set pitch > width (only width > pitch was rejected) to force an underestimated buf_sizes[0x2] check.\nPR:L - Exploitation requires only standard unprivileged local access to the AMDGPU DRM render node (DRM_AUTH|DRM_RENDER_ALLOW on AMDGPU_CS ioctl), not real root; user namespaces alone do not grant render-node access.\nUI:N - No victim interaction is required; a local attacker can directly craft and submit a UVD command stream with a decode message via AMDGPU_CS ioctl after creating a valid UVD session handle.\nS:U - Impact is confined to the kernel/GPU driver on the local host (GPU out-of-bounds access within the attacker's DRM client address space); it does not cross VM, IOMMU, or sandbox boundaries to a different security authority.\nC:H - Underestimating the decode image minimum size lets undersized buffers pass amdgpu_uvd_cs_pass2 validation while the UVD engine decodes using pitch stride, enabling out-of-bounds reads of adjacent GPU-mapped memory that may contain sensitive data.\nI:H - Using image_size (width-based) instead of (pitch*height)*3/2 for buf_sizes[0x2] allows the UVD hardware to perform large out-of-bounds writes past the validated buffer when pitch exceeds width, corrupting adjacent GPU-mapped memory.\nA:H - Large GPU out-of-bounds writes from the UVD decode engine can hang or reset the GPU driver and may trigger kernel oops/panic; unbounded pitch before the fix also risked integer overflow wrapping the minimum size to near-zero, amplifying corruption."
    }
  }
}