{
  "schema_version": "1.6.1",
  "id": "RPI-CVE-2025-21780",
  "modified": "2026-10-09T11:39:44Z",
  "upstream": [
    "CVE-2025-21780"
  ],
  "published": "2025-02-27T02:18:18Z",
  "summary": "drm/amdgpu: avoid buffer overflow attach in smu_sys_set_pp_table()",
  "details": "In the Linux kernel, the following vulnerability has been resolved:\n\ndrm/amdgpu: avoid buffer overflow attach in smu_sys_set_pp_table()\n\nIt malicious user provides a small pptable through sysfs and then\na bigger pptable, it may cause buffer overflow attack in function\nsmu_sys_set_pp_table().",
  "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": "6.1.129",
            "compiled_filter": false
          }
        }
      }
    }
  ],
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://www.cve.org/CVERecord?id=CVE-2025-21780"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/3484ea33157bc7334f57e64826ec5a4bf992151a"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/e43a8b9c4d700ffec819c5043a48769b3e7d9cab"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/2498d2db1d35e88a2060ea191ae75dce853dd084"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/231075c5a8ea54f34b7c4794687baa980814e6de"
    },
    {
      "type": "FIX",
      "url": "https://git.kernel.org/stable/c/1abb2648698bf10783d2236a6b4a7ca5e8021699"
    }
  ],
  "database_specific": {
    "source": "https://git.kernel.org/pub/scm/linux/security/vulns.git",
    "upstream_modified": "2026-08-05T12:19:11Z",
    "severity": {
      "source": "kernel CNA",
      "cvss_v3_score": 7.8,
      "cvss_v3_severity": "High",
      "scenario": "AV:L - The vulnerability is reached exclusively by writing to the local sysfs attribute /sys/class/drm/cardN/device/pp_table; there is no remote or adjacent-network path to smu_sys_set_pp_table().\nAC:L - Triggering is fully deterministic \u2014 write a tiny pptable whose 16-bit usStructureSize matches the write length, then write a larger one; there is no race, no ASLR dependency, and the attacker chooses both the allocation size and the overflow length/content.\nPR:L - The attribute is 0644 by default, but on AMD desktop/handheld/gaming deployments the amdgpu pp_* power-tuning nodes are routinely made group-writable via udev rules so unprivileged GUI overclocking tools can use them, and the fix's own changelog describes the attacker as a \"malicious user\"; per the choose-higher rule this is scored as an unprivileged local user rather than init-namespace root.\nUI:N - The attacker performs two ordinary write() calls on their own; no victim action, mount, or file-open by another user is involved.\nS:U - The heap corruption occurs in the kernel's own slab allocator and is confined to the kernel security authority; no VM, IOMMU, or sandbox boundary is crossed.\nC:H - Beyond the write primitive, the stale power_play_table_size makes amdgpu_get_pp_table() memcpy up to PAGE_SIZE-1 bytes out of a 1-byte allocation straight back to userspace, directly disclosing ~4 KB of adjacent kernel heap; the overflow itself also enables arbitrary read via corrupted adjacent objects.\nI:H - This is an out-of-bounds heap write of up to ~4 KB with fully attacker-chosen contents starting from a kmalloc-8 object, smashing many adjacent slab objects \u2014 a classic primitive for function-pointer/credential overwrite and privilege escalation.\nA:H - Overwriting kilobytes of adjacent slab objects and freelist metadata reliably produces slab corruption panics/oopses, and smu_reset() is then invoked with the corrupted power-play table, hanging or wedging the GPU."
    }
  }
}