CVE-2026-31706
HighCVSS 8.8Exploitation Probability (EPSS)
Low risk40th percentile - higher than 40% of all known CVEs
Summary
In the Linux kernel, ksmbd's smb_inherit_dacl() trusted the num_aces value from the parent directory's DACL xattr without validating it against pdacl_size. An authenticated client could tamper with the security.NTACL attribute to set num_aces = 65535, causing a large (~8 MB) uninitialized allocation and potential overflow on 32-bit kernels. The fix validates num_aces, uses overflow-safe kmalloc_array, and tightens ACE size checks.
Risk Assessment
The vulnerability can be exploited by an authenticated SMB client to cause a kernel crash (denial of service) or potentially execute code due to unsafe memory allocation and overflow. It requires a tampered NTACL attribute on a parent directory, which may be less likely in typical setups but poses a real risk in environments with untrusted SMB clients.
Recommendation
Update the Linux kernel to a version containing the fix for CVE-2026-31706. If immediate patching is not possible, consider restricting access to ksmbd SMB shares or monitoring for unusual modifications of security attributes on filesystems.
Other vulnerabilities in Linux kernel (ksmbd)
See all- CVE-2026-89788Critical
In the Linux kernel, the ksmbd module contains a use-after-free vulnerability in smb2_tree_connect(). A newly created tree connection is published in the session and then used by the handler, while a concurrent session logoff can free the object before the handler finishes.
- CVE-2026-89635Critical
In the Linux kernel, the ksmbd_reopen_durable_fd() function rebinds every detached oplock on an inode to the reconnecting session instead of only the oplock owned by the reopened file. This allows one session to adopt another session's oplock, and when the adopting session is destroyed, a use-after-free occurs. The vulnerability is reachable with the default durable-handle and oplock configuration when two sessions open the same file with distinct AppInstanceIds.
- CVE-2026-80926Critical
A use-after-free vulnerability was fixed in the Linux kernel's ksmbd oplock break notification mechanism. The smb2_oplock_break_noti() function read the opinfo->conn pointer without locking and dereferenced it after sleeping allocations, allowing a race with durable handle teardown to access a freed connection. The flaw is reachable by any authenticated client holding a durable batch oplock.
- CVE-2026-43376Critical
A vulnerability has been identified in the Linux kernel within ksmbd, related to a use-after-free error due to immediate freeing of the oplock_info structure. This issue occurs when the pointer is nullified without a grace period delay in RCU, allowing access to already freed memory.
- CVE-2026-43185Critical
The ksmbd module in the Linux kernel has a signedness bug in smb_direct_prepare_negotiation(). An unsigned __u32 value from preferred_send_size is cast to a signed int before computing min_t(int, ...), so a malicious value of 0x80000000 is treated as smaller than max_recv_size and used to set the maximum allowed receive size. By sending a second message with a large value (>1420 bytes), the attacker can achieve a heap buffer overflow.
- CVE-2026-23428Critical
A use-after-free vulnerability exists in the Linux kernel ksmbd module involving the share_conf object in compound requests. smb2_get_ksmbd_tcon() reuses work->tcon without validating tcon->t_state, so after an earlier SMB2_TREE_DISCONNECT frees share_conf, subsequent commands dereference the freed memory.
- CVE-2023-54203Critical
In the Linux kernel, a vulnerability in the ksmbd module leads to a slab-out-of-bounds read in the init_smb2_rsp_hdr function. The issue occurs when an SMB1 mount attempt fails and the ksmbd server incorrectly handles SMB1 negotiation as an SMB2 operation, causing an out-of-bounds memory read.
- CVE-2023-38432Critical
An issue was discovered in the Linux kernel before 6.3.10 related to the validation of command payload size in ksmbd, leading to an out-of-bounds read.
- CVE-2023-38431Critical
An issue was discovered in the Linux kernel before 6.3.8 related to the lack of validation of the NetBIOS header's length in relation to SMB header sizes in ksmbd. This leads to an out-of-bounds read.
- CVE-2023-38427Critical
An issue was discovered in the Linux kernel before 6.3.8 related to integer underflow and out-of-bounds read in the deassemble_neg_contexts function in fs/smb/server/smb2pdu.c in ksmbd.
Original NVD description (English source)
In the Linux kernel, the following vulnerability has been resolved: ksmbd: validate num_aces and harden ACE walk in smb_inherit_dacl() smb_inherit_dacl() trusts the on-disk num_aces value from the parent directory's DACL xattr and uses it to size a heap allocation: aces_base = kmalloc(sizeof(struct smb_ace) * num_aces * 2, ...); num_aces is a u16 read from le16_to_cpu(parent_pdacl->num_aces) without checking that it is consistent with the declared pdacl_size. An authenticated client whose parent directory's security.NTACL is tampered (e.g. via offline xattr corruption or a concurrent path that bypasses parse_dacl()) can present num_aces = 65535 with minimal actual ACE data. This causes a ~8 MB allocation (not kzalloc, so uninitialized) that the subsequent loop only partially populates, and may also overflow the three-way size_t multiply on 32-bit kernels. Additionally, the ACE walk loop uses the weaker offsetof(struct smb_ace, access_req) minimum size check rather than the minimum valid on-wire ACE size, and does not reject ACEs whose declared size is below the minimum. Reproduced on UML + KASAN + LOCKDEP against the real ksmbd code path. A legitimate mount.cifs client creates a parent directory over SMB (ksmbd writes a valid security.NTACL xattr), then the NTACL blob on the backing filesystem is rewritten to set num_aces = 0xFFFF while keeping the posix_acl_hash bytes intact so ksmbd_vfs_get_sd_xattr()'s hash check still passes. A subsequent SMB2 CREATE of a child under that parent drives smb2_open() into smb_inherit_dacl() (share has "vfs objects = acl_xattr" set), which fails the page allocator: WARNING: mm/page_alloc.c:5226 at __alloc_frozen_pages_noprof+0x46c/0x9c0 Workqueue: ksmbd-io handle_ksmbd_work __alloc_frozen_pages_noprof+0x46c/0x9c0 ___kmalloc_large_node+0x68/0x130 __kmalloc_large_node_noprof+0x24/0x70 __kmalloc_noprof+0x4c9/0x690 smb_inherit_dacl+0x394/0x2430 smb2_open+0x595d/0xabe0 handle_ksmbd_work+0x3d3/0x1140 With the patch applied the added guard rejects the tampered value with -EINVAL before any large allocation runs, smb2_open() falls back to smb2_create_sd_buffer(), and the child is created with a default SD. No warning, no splat. Fix by: 1. Validating num_aces against pdacl_size using the same formula applied in parse_dacl(). 2. Replacing the raw kmalloc(sizeof * num_aces * 2) with kmalloc_array(num_aces * 2, sizeof(...)) for overflow-safe allocation. 3. Tightening the per-ACE loop guard to require the minimum valid ACE size (offsetof(smb_ace, sid) + CIFS_SID_BASE_SIZE) and rejecting under-sized ACEs, matching the hardening in smb_check_perm_dacl() and parse_dacl(). v1 -> v2: - Replace the synthetic test-module splat in the changelog with a real-path UML + KASAN reproduction driven through mount.cifs and SMB2 CREATE; Namjae flagged the kcifs3_test_inherit_dacl_old name in v1 since it does not exist in ksmbd. - Drop the commit-hash citation from the code comment per Namjae's review; keep the parse_dacl() pointer.
Vulnerability data from NVD (NIST) · CISA KEV · EPSS