{ "dataType": "CVE_RECORD", "dataVersion": "5.2", "cveMetadata": { "cveId": "CVE-2025-39976", "assignerOrgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "state": "PUBLISHED", "assignerShortName": "Linux", "dateReserved": "2025-04-16T07:20:57.150Z", "datePublished": "2025-10-15T07:55:57.608Z", "dateUpdated": "2026-08-05T12:06:45.033Z" }, "containers": { "cna": { "providerMetadata": { "orgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "shortName": "Linux", "dateUpdated": "2026-08-05T12:06:45.033Z" }, "descriptions": [ { "lang": "en", "value": "In the Linux kernel, the following vulnerability has been resolved:\n\nfutex: Use correct exit on failure from futex_hash_allocate_default()\n\ncopy_process() uses the wrong error exit path from futex_hash_allocate_default().\nAfter exiting from futex_hash_allocate_default(), neither tasklist_lock\nnor siglock has been acquired. The exit label bad_fork_core_free unlocks\nboth of these locks which is wrong.\n\nThe next exit label, bad_fork_cancel_cgroup, is the correct exit.\nsched_cgroup_fork() did not allocate any resources that need to freed.\n\nUse bad_fork_cancel_cgroup on error exit from futex_hash_allocate_default()." } ], "metrics": [ { "cvssV3_1": { "version": "3.1", "vectorString": "CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H", "baseScore": 7.8, "baseSeverity": "HIGH" }, "scenarios": [ { "lang": "en", "value": "AV:L - The bug is reached through the clone()/clone3() syscall path (copy_process) combined with prctl(PR_FUTEX_HASH), both requiring local execution on the target system. There is no network-facing component in kernel/fork.c or kernel/futex/core.c.\nAC:L - The attacker fully controls both failure routes: forcing -ENOMEM on the GFP_KERNEL_ACCOUNT kvzalloc/alloc_percpu by sitting at a memcg limit is essentially deterministic, and the -EBUSY route is a race whose both sides (prctl PR_FUTEX_HASH_SET_SLOTS=0 and a clone(CLONE_VM) loop) are attacker-driven with unlimited retries. CONFIG_FUTEX_PRIVATE_HASH is default y on virtually all MMU builds, so no rare configuration is needed.\nPR:L - Any unprivileged local user can call pthread_create()/clone(CLONE_VM) and prctl(PR_FUTEX_HASH), neither of which has any capability check in kernel/sys.c or futex_hash_prctl(). No root, no namespace tricks, and no CAP_* are required.\nUI:N - The attacker triggers the faulty error path entirely from its own process via syscalls; no action by any other user or administrator is needed.\nS:U - The corrupted state (preempt_count, tasklist_lock, sighand->siglock, task/pid lists) is all within the kernel's own security authority on the same host. No hypervisor, IOMMU, or sandbox boundary is crossed.\nC:H - Spuriously releasing the global tasklist_lock and a shared sighand->siglock lets concurrent writers mutate the task list, thread list, pid hash, and signal structures simultaneously, producing list corruption and use-after-free of task_struct/sighand that can be groomed into a kernel memory read primitive. Per kernel scoring guidance, memory corruption leveragable for information disclosure is High.\nI:H - The stolen tasklist_lock/siglock permit racing list_add/list_del on global kernel structures, yielding dangling pointers and attacker-influenced writes, a classic path to control-flow hijack and privilege escalation. Additionally, preempt_count underflow leaves the CPU with corrupted context-tracking state that violates every atomicity assumption in the scheduler and RCU.\nA:H - Two unmatched preempt_enable() calls underflow the per-CPU preempt_count so the CPU permanently appears to be in hardirq/NMI context, breaking RCU, softirq processing, and preemption, which leads to RCU stalls, soft lockups, or panic; debug kernels BUG immediately on the bad unlock balance. Task-list corruption from the stolen tasklist_lock independently produces kernel oopses." } ] } ], "affected": [ { "product": "Linux", "vendor": "Linux", "defaultStatus": "unaffected", "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "programFiles": [ "kernel/fork.c" ], "versions": [ { "version": "7c4f75a21f636486d2969d9b6680403ea8483539", "lessThan": "f1635765cd0fdbf27b04d9a50be91a01b5adda13", "status": "affected", "versionType": "git" }, { "version": "7c4f75a21f636486d2969d9b6680403ea8483539", "lessThan": "4ec3c15462b9f44562f45723a92e2807746ba7d1", "status": "affected", "versionType": "git" } ] }, { "product": "Linux", "vendor": "Linux", "defaultStatus": "affected", "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "programFiles": [ "kernel/fork.c" ], "versions": [ { "version": "6.16", "status": "affected" }, { "version": "0", "lessThan": "6.16", "status": "unaffected", "versionType": "semver" }, { "version": "6.16.10", "lessThanOrEqual": "6.16.*", "status": "unaffected", "versionType": "semver" }, { "version": "6.17", "lessThanOrEqual": "*", "status": "unaffected", "versionType": "original_commit_for_fix" } ] } ], "cpeApplicability": [ { "nodes": [ { "operator": "OR", "negate": false, "cpeMatch": [ { "vulnerable": true, "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*", "versionStartIncluding": "6.16", "versionEndExcluding": "6.16.10" }, { "vulnerable": true, "criteria": "cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*", "versionStartIncluding": "6.16", "versionEndExcluding": "6.17" } ] } ] } ], "references": [ { "url": "https://git.kernel.org/stable/c/f1635765cd0fdbf27b04d9a50be91a01b5adda13" }, { "url": "https://git.kernel.org/stable/c/4ec3c15462b9f44562f45723a92e2807746ba7d1" } ], "title": "futex: Use correct exit on failure from futex_hash_allocate_default()", "x_generator": { "engine": "bippy-1.2.0" } } } }