{
  "affected": [
    {
      "ranges": [
        {
          "events": [
            {
              "introduced": "bee25f97ad24641df55cb3887eb5bfcb2b9d8fbc"
            },
            {
              "fixed": "19b54740e2e1102c7d48553d9e0347c1f6af22b0"
            },
            {
              "fixed": "69cb2be898be6d5826cacd1a628f6945a24480b6"
            }
          ],
          "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git",
          "type": "GIT"
        }
      ]
    },
    {
      "package": {
        "ecosystem": "Linux",
        "name": "Kernel"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "7.2.0"
            },
            {
              "fixed": "7.2.6"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "database_specific": {
    "cna_assigner": "Linux",
    "osv_generated_from": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/90xxx/CVE-2026-90339.json"
  },
  "details": "In the Linux kernel, the following vulnerability has been resolved:\n\npowerpc/syscall: Fix syscall skip handling for seccomp and ptrace\n\nAfter enabling GENERIC_ENTRY on PowerPC, syscall_enter_from_user_mode()\nreturns -1 as a sentinel to signal that seccomp or ptrace has intercepted\nthe syscall and already set a return value via syscall_set_return_value().\nsystem_call_exception() was not handling this sentinel, and since -1UL\nis \u003e= NR_syscalls, the code fell into the out-of-range path and returned\n-ENOSYS, overwriting the errno already placed in regs-\u003egpr[3].\n\nThe naive fix of checking r0 == -1L before the NR_syscalls bounds check\nis ambiguous: a user legitimately calling syscall(-1) also produces r0 ==\n-1L, and a tracer intercepting such a call would have its injected return\nvalue silently discarded.\n\nFix this by introducing a thread flag that is set whenever\nsyscall_set_return_value() explicitly updates the return value. In\nsystem_call_exception(), check and clear this flag before dispatching\nthe syscall, and return the preset value directly when it is present.\nThis ensures that an explicitly supplied return value always suppresses\nsyscall execution, regardless of the syscall number.\n\nThis handles all seccomp actions correctly:\n\n  - SECCOMP_RET_ERRNO, SECCOMP_RET_TRACE (no tracer), SECCOMP_RET_USER_NOTIF:\n    all call syscall_set_return_value(), flag is set, injected value returned.\n  - SECCOMP_RET_TRAP, SECCOMP_RET_KILL: call syscall_rollback() and deliver\n    a signal; flag is not set, but the process is dying so the return value\n    is irrelevant.\n\nThe fix covers both ppc32 and ppc64 with no #ifdefs.",
  "id": "CVE-2026-90339",
  "modified": "2026-09-19T03:30:53.546699264Z",
  "published": "2026-09-17T16:08:50.990Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://git.kernel.org/stable/c/19b54740e2e1102c7d48553d9e0347c1f6af22b0"
    },
    {
      "type": "WEB",
      "url": "https://git.kernel.org/stable/c/69cb2be898be6d5826cacd1a628f6945a24480b6"
    },
    {
      "type": "ADVISORY",
      "url": "https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/90xxx/CVE-2026-90339.json"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-90339"
    },
    {
      "type": "PACKAGE",
      "url": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git"
    }
  ],
  "schema_version": "1.9.0",
  "summary": "powerpc/syscall: Fix syscall skip handling for seccomp and ptrace"
}