{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2025-38244","assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","state":"PUBLISHED","assignerShortName":"Linux","dateReserved":"2025-04-16T04:51:23.996Z","datePublished":"2025-07-09T10:42:26.622Z","dateUpdated":"2026-08-05T12:00:39.806Z"},"containers":{"cna":{"providerMetadata":{"orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux","dateUpdated":"2026-08-05T12:00:39.806Z"},"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nsmb: client: fix potential deadlock when reconnecting channels\n\nFix cifs_signal_cifsd_for_reconnect() to take the correct lock order\nand prevent the following deadlock from happening\n\n======================================================\nWARNING: possible circular locking dependency detected\n6.16.0-rc3-build2+ #1301 Tainted: G S      W\n------------------------------------------------------\ncifsd/6055 is trying to acquire lock:\nffff88810ad56038 (&tcp_ses->srv_lock){+.+.}-{3:3}, at: cifs_signal_cifsd_for_reconnect+0x134/0x200\n\nbut task is already holding lock:\nffff888119c64330 (&ret_buf->chan_lock){+.+.}-{3:3}, at: cifs_signal_cifsd_for_reconnect+0xcf/0x200\n\nwhich lock already depends on the new lock.\n\nthe existing dependency chain (in reverse order) is:\n\n-> #2 (&ret_buf->chan_lock){+.+.}-{3:3}:\n       validate_chain+0x1cf/0x270\n       __lock_acquire+0x60e/0x780\n       lock_acquire.part.0+0xb4/0x1f0\n       _raw_spin_lock+0x2f/0x40\n       cifs_setup_session+0x81/0x4b0\n       cifs_get_smb_ses+0x771/0x900\n       cifs_mount_get_session+0x7e/0x170\n       cifs_mount+0x92/0x2d0\n       cifs_smb3_do_mount+0x161/0x460\n       smb3_get_tree+0x55/0x90\n       vfs_get_tree+0x46/0x180\n       do_new_mount+0x1b0/0x2e0\n       path_mount+0x6ee/0x740\n       do_mount+0x98/0xe0\n       __do_sys_mount+0x148/0x180\n       do_syscall_64+0xa4/0x260\n       entry_SYSCALL_64_after_hwframe+0x76/0x7e\n\n-> #1 (&ret_buf->ses_lock){+.+.}-{3:3}:\n       validate_chain+0x1cf/0x270\n       __lock_acquire+0x60e/0x780\n       lock_acquire.part.0+0xb4/0x1f0\n       _raw_spin_lock+0x2f/0x40\n       cifs_match_super+0x101/0x320\n       sget+0xab/0x270\n       cifs_smb3_do_mount+0x1e0/0x460\n       smb3_get_tree+0x55/0x90\n       vfs_get_tree+0x46/0x180\n       do_new_mount+0x1b0/0x2e0\n       path_mount+0x6ee/0x740\n       do_mount+0x98/0xe0\n       __do_sys_mount+0x148/0x180\n       do_syscall_64+0xa4/0x260\n       entry_SYSCALL_64_after_hwframe+0x76/0x7e\n\n-> #0 (&tcp_ses->srv_lock){+.+.}-{3:3}:\n       check_noncircular+0x95/0xc0\n       check_prev_add+0x115/0x2f0\n       validate_chain+0x1cf/0x270\n       __lock_acquire+0x60e/0x780\n       lock_acquire.part.0+0xb4/0x1f0\n       _raw_spin_lock+0x2f/0x40\n       cifs_signal_cifsd_for_reconnect+0x134/0x200\n       __cifs_reconnect+0x8f/0x500\n       cifs_handle_standard+0x112/0x280\n       cifs_demultiplex_thread+0x64d/0xbc0\n       kthread+0x2f7/0x310\n       ret_from_fork+0x2a/0x230\n       ret_from_fork_asm+0x1a/0x30\n\nother info that might help us debug this:\n\nChain exists of:\n  &tcp_ses->srv_lock --> &ret_buf->ses_lock --> &ret_buf->chan_lock\n\n Possible unsafe locking scenario:\n\n       CPU0                    CPU1\n       ----                    ----\n  lock(&ret_buf->chan_lock);\n                               lock(&ret_buf->ses_lock);\n                               lock(&ret_buf->chan_lock);\n  lock(&tcp_ses->srv_lock);\n\n *** DEADLOCK ***\n\n3 locks held by cifsd/6055:\n #0: ffffffff857de398 (&cifs_tcp_ses_lock){+.+.}-{3:3}, at: cifs_signal_cifsd_for_reconnect+0x7b/0x200\n #1: ffff888119c64060 (&ret_buf->ses_lock){+.+.}-{3:3}, at: cifs_signal_cifsd_for_reconnect+0x9c/0x200\n #2: ffff888119c64330 (&ret_buf->chan_lock){+.+.}-{3:3}, at: cifs_signal_cifsd_for_reconnect+0xcf/0x200"}],"metrics":[{"cvssV3_1":{"version":"3.1","vectorString":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","baseScore":7.5,"baseSeverity":"HIGH"},"scenarios":[{"lang":"en","value":"AV:N - The vulnerable function is reached from `cifs_demultiplex_thread` while parsing data received from a remote SMB server (short/oversized RFC1002 PDUs, malformed compound `next_header`, `STATUS_NETWORK_SESSION_EXPIRED`), so a malicious/compromised SMB server or an attacker injecting into an unsigned SMB connection triggers it over the network. The DFS-referral and SMB-witness callers are likewise driven by remote server responses.\nAC:L - The attacker can force `cifs_reconnect(server, true)` into the inverted `chan_lock → srv_lock` sequence deterministically and at unbounded rate with a single malformed response, and the reconnect storms he causes themselves drive the opposing `ses_lock → chan_lock` session-setup path on other channel threads. No rare kernel config is needed — the loop starts at index 0, so even a single-channel mount takes the inverted order.\nPR:N - The attacker needs no account or privilege on the victim client; the reconnect triggers are evaluated on the received header before signature verification in `handle_mid()`, and the demux thread runs from `cifs_get_tcp_session()` onward, making the path reachable pre-authentication.\nUI:N - On an already-mounted share — the normal enterprise/cloud case — a compromised server or on-path attacker triggers the path with no victim action at all; no new mount, file open, or any other local operation is required.\nS:U - The deadlock is confined to the kernel's own locking of the CIFS client and does not cross into another security authority such as a hypervisor or IOMMU boundary.\nC:N - This is purely a lock-acquisition-order defect; no memory is read out of bounds, no freed object is dereferenced, and no kernel data is exposed to the attacker.\nI:N - No memory is written out of bounds or after free and no control-flow structure is corrupted, so the bug yields no data-modification or write primitive.\nA:H - The circular dependency is between raw `spin_lock()`s, so closing the cycle spins CPUs with preemption disabled into an unrecoverable hard lockup that hangs the cifsd thread, the mount path, and the whole system; on `panic_on_warn` kernels the lockdep WARNING alone is an immediate panic."}]}],"affected":[{"product":"Linux","vendor":"Linux","defaultStatus":"unaffected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["fs/smb/client/cifsglob.h","fs/smb/client/connect.c"],"versions":[{"version":"d7d7a66aacd6fd8ca57baf08a7bac5421282f6f8","lessThan":"c82c7041258d96e3286f6790ab700e4edd3cc9e3","status":"affected","versionType":"git"},{"version":"d7d7a66aacd6fd8ca57baf08a7bac5421282f6f8","lessThan":"7f3ead8ebc0ef65b6c89a13912b4e80218425629","status":"affected","versionType":"git"},{"version":"d7d7a66aacd6fd8ca57baf08a7bac5421282f6f8","lessThan":"fe035dc78aa6ca8f862857d45beaf7a0e03206ca","status":"affected","versionType":"git"},{"version":"d7d7a66aacd6fd8ca57baf08a7bac5421282f6f8","lessThan":"711741f94ac3cf9f4e3aa73aa171e76d188c0819","status":"affected","versionType":"git"}]},{"product":"Linux","vendor":"Linux","defaultStatus":"affected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["fs/smb/client/cifsglob.h","fs/smb/client/connect.c"],"versions":[{"version":"6.0","status":"affected"},{"version":"0","lessThan":"6.0","status":"unaffected","versionType":"semver"},{"version":"6.6.96","lessThanOrEqual":"6.6.*","status":"unaffected","versionType":"semver"},{"version":"6.12.36","lessThanOrEqual":"6.12.*","status":"unaffected","versionType":"semver"},{"version":"6.15.5","lessThanOrEqual":"6.15.*","status":"unaffected","versionType":"semver"},{"version":"6.16","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":"6.0","versionEndExcluding":"6.6.96"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"6.0","versionEndExcluding":"6.12.36"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"6.0","versionEndExcluding":"6.15.5"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"6.0","versionEndExcluding":"6.16"}]}]}],"references":[{"url":"https://git.kernel.org/stable/c/c82c7041258d96e3286f6790ab700e4edd3cc9e3"},{"url":"https://git.kernel.org/stable/c/7f3ead8ebc0ef65b6c89a13912b4e80218425629"},{"url":"https://git.kernel.org/stable/c/fe035dc78aa6ca8f862857d45beaf7a0e03206ca"},{"url":"https://git.kernel.org/stable/c/711741f94ac3cf9f4e3aa73aa171e76d188c0819"}],"title":"smb: client: fix potential deadlock when reconnecting channels","x_generator":{"engine":"bippy-1.2.0"}}}}