| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| On Xtensa SoCs built with CONFIG_XTENSA_MPU and CONFIG_USERSPACE, arch_buffer_validate() in arch/xtensa/core/mpu.c — the architecture hook that verifies a user-mode-supplied buffer is accessible to the calling user thread with the requested permission — defaulted its return value to 0 (access permitted) and only set a denial result inside its per-MPU-region probe loop. When the rounded extent of the buffer wraps the 32-bit address space (size + alignment offset near SIZE_MAX, or ROUND_UP(size + offset) overflowing to 0), the loop executes zero iterations and the function returns 0 = permitted without probing any MPU region.
The syscall-layer pre-checks (K_SYSCALL_MEMORY_SIZE_CHECK / Z_DETECT_POINTER_OVERFLOW) only catch a raw addr+size wrap and do not cover the ROUND_UP-induced wrap, and the string path (arch_user_string_nlen -> arch_buffer_validate) has no syscall-layer guard at all.
An unprivileged user-mode thread can therefore pass a crafted (addr, size) to any syscall that validates user buffers via k_usermode_from_copy/to_copy or k_usermode_string_copy and have validation succeed for memory it must not access; the kernel then reads from (disclosure) or, with write=1, writes to (corruption) attacker-chosen kernel or other-partition memory on the thread's behalf, enabling information disclosure, memory corruption, privilege escalation, and denial of service.
Affected from v3.7.0 (when Xtensa MPU userspace support was added) through v4.4.0. The fix changes the default to -EINVAL (deny by default), adds an explicit size_add_overflow check, and sets the success value only after the full range has been validated. |
| The CONFIG_USERSPACE verification handler for the k_thread_name_copy() system call (z_vrfy_k_thread_name_copy() in kernel/thread.c) calls k_object_find() on the caller-supplied thread pointer and then dereferences the returned struct k_object without checking it for NULL. k_object_find() returns NULL whenever the supplied pointer is not a registered (static or dynamic) kernel object.
The pre-fix guard tested thread == NULL instead of ko == NULL, so an unprivileged user-mode thread that invokes k_thread_name_copy() with any non-NULL but unregistered pointer (e.g. an arbitrary address) passes the NULL test, after which the verifier reads ko->type through a NULL pointer.
Because the syscall verifier runs in supervisor mode, this NULL dereference is a kernel-mode fault that halts or reboots the system, allowing untrusted user code to crash the kernel across the userspace security boundary (denial of service). The marshaller passes the thread argument to the verifier without any prior K_SYSCALL_OBJ validation, so the bad pointer reaches the defect directly.
The flaw affects builds with CONFIG_USERSPACE and CONFIG_THREAD_NAME enabled and has been present since the special-case lookup was introduced around v2.0.0; it is present in v4.4.0 and earlier. The fix changes the guard to check the k_object_find() return value (ko == NULL) before dereferencing it. |
| In Zephyr's kernel pipe implementation, the userspace syscall verifier z_vrfy_k_pipe_init() in kernel/pipe.c used K_SYSCALL_OBJ() (which requires the kernel object to already be initialized) instead of K_SYSCALL_OBJ_NEVER_INIT() (which rejects an already-initialized object). As a result, on CONFIG_USERSPACE builds an unprivileged user thread that has been granted access to a k_pipe object can invoke the k_pipe_init syscall to re-initialize a pipe that is already in use.
z_impl_k_pipe_init() unconditionally resets the ring buffer, sets pipe->waiting to 0, and re-initializes both wait queues (z_waitq_init on pipe->data and pipe->space) without waking or accounting for threads currently blocked on the pipe. Any thread already pended in k_pipe_read()/k_pipe_write() is left orphaned: still marked pending with pended_on pointing at the cleared wait queue and with stale qnode_dlist links into the (now re-initialized) embedded list head.
When such an orphaned waiter is later timed out or woken, the scheduler calls sys_dlist_remove() on its stale node, writing through dangling prev/next pointers into kernel wait-queue/scheduler structures, causing list corruption (an attacker-driven invalid kernel write), lost wakeups, indefinitely blocked threads, and silent data loss. The flaw lets a deprivileged user thread corrupt the state of a kernel object shared with other threads/partitions.
The fix switches the verifier to K_SYSCALL_OBJ_NEVER_INIT(), matching the existing k_msgq_init verifier, so a user thread can no longer re-initialize a live pipe. The vulnerable code shipped in v4.1.0 and remained through v4.4.0. |
| subsys/net/lib/lwm2m/lwm2m_pull_context.c copied the firmware-update Package URI into a fixed static buffer (context.uri, size CONFIG_LWM2M_SWMGMT_PACKAGE_URI_LEN, default 128) with memcpy(context.uri, uri, LWM2M_PACKAGE_URI_LEN), copying exactly the destination size with no length validation. The Firmware-Update object stores the server-supplied Package URI (/5/0/1) in a 255-byte buffer, so a LwM2M management server (or an on-path attacker on a session lacking strong DTLS) can WRITE a URI of 128-254 characters; only the first 128 bytes are then copied into context.uri with no NUL terminator. That buffer is subsequently consumed as a C string by http_parser_parse_url(context.uri, strlen(context.uri), ...), strlen-based CoAP URI-path/PROXY-URI option appends, and lwm2m_parse_peerinfo(), causing an out-of-bounds read of adjacent static memory. The over-read bytes are appended to outbound CoAP requests (information disclosure of adjacent device memory to the server/proxy) and can crash the device (denial of service). The vulnerable copy was introduced by the pull-context refactor (first released in v3.0.0) and is present through v4.4.0; the default-on CONFIG_LWM2M_FIRMWARE_UPDATE_PULL_SUPPORT path is affected. The fix adds a strlen(uri) >= sizeof(context.uri) check returning -ENOMEM and switches to strcpy(), guaranteeing a bounded, NUL-terminated buffer. |
| parse_ipv4() in subsys/net/ip/utils.c (reached via net_ipaddr_parse() for strings of the form "a.b.c.d:port") copies the port substring into a fixed 17-byte stack buffer (char ipaddr[NET_IPV4_ADDR_LEN + 1]) using a length of str_len - end - 1, where str_len is the full, unbounded input length and end is only the (<=15-byte) offset of the ':' delimiter. Because the destination size is never consulted, a crafted address string with a long suffix after the colon (e.g. "1.2.3.4:" followed by hundreds of bytes) causes an out-of-bounds stack write whose length and contents are fully attacker-controlled (memcpy of the suffix plus a trailing NUL), enabling memory corruption and at minimum a denial of service, and potentially control-flow hijack. The parser is reached from the standard socket API (zsock_getaddrinfo / literal-address resolution), DNS server-string configuration, and the eswifi Wi-Fi co-processor DNS-response path, so an application that resolves a network-influenced address string is exposed. The bug was introduced when the parser was added (Zephyr v1.9.0) and shipped in all releases through v4.4.0. The fix removes the unbounded copy and validates the port length before copying into a small dedicated buffer. Note: the equivalent IPv6 "[addr]:port" path in parse_ipv6() retains the same unbounded copy at this commit and remains a separate, still-reachable instance of the defect. |
| Zephyr's dynamic kernel-object tracking (kernel/userspace/userspace.c, formerly kernel/userspace.c) maintains a doubly-linked list (obj_list) of dynamically allocated kernel objects. Iteration over this list in k_object_wordlist_foreach() was performed under lists_lock using the SAFE iterator (which caches the next node), but list removal and freeing of nodes was performed under different, disjoint spinlocks: objfree_lock in k_object_free() and obj_lock in unref_check(). On an SMP system, while one CPU iterated obj_list under lists_lock, another CPU could unlink and k_free() the dyn_obj node that the iterator had cached as its next pointer, causing the iterator to dereference freed kernel memory (use-after-free / dangling list traversal). All of the racing operations are reachable from unprivileged user-mode threads via system calls: k_object_alloc/k_object_alloc_size and k_object_release drive removals through unref_check() (under obj_lock), while k_thread_abort and thread creation drive the iteration through k_thread_perms_all_clear()/k_thread_perms_inherit() (under lists_lock). A deprivileged user thread on a CONFIG_SMP + CONFIG_USERSPACE build can therefore corrupt the kernel's object-tracking structures across the userspace security boundary, yielding kernel memory corruption (potential privilege escalation) or a kernel crash (denial of service). The fix removes objfree_lock and serializes every obj_list modification under lists_lock, including holding it across find+remove in k_object_free() and around unref_check() in k_thread_perms_clear(). Affects CONFIG_SMP+CONFIG_USERSPACE+CONFIG_DYNAMIC_OBJECTS configurations; the defect dates to the 2019 spinlockification (commit 8a3d57b6cc6, first released in v1.14.0) and shipped through v4.4.0. |
| In Zephyr's WireGuard subsystem (subsys/net/lib/wireguard), wg_process_data_message() in wg_crypto.c linearizes an inbound transport-data payload into a fixed pool buffer of CONFIG_WIREGUARD_BUF_LEN bytes before decryption. The call net_buf_linearize(buf->data, data_len, pkt->buffer, ..., data_len) passed the attacker-derived data_len as both the destination capacity and the copy length, defeating the function's internal len = min(len, dst_len) bound. data_len is derived from the received UDP datagram length and is only lower-bounded by wg_ctrl_recv() (no upper bound). When data_len exceeds CONFIG_WIREGUARD_BUF_LEN — e.g. when the buffer length is lowered below the link MTU, on links with MTU above the buffer size, or via reassembled IPv4/IPv6 fragments that exceed it — the underlying memcpy writes past the end of the pool buffer, an out-of-bounds write (CWE-787). The overflow occurs before the Poly1305 authentication check, so it requires only a valid receiver session index rather than a valid authenticator, and is reachable by a malicious or compromised peer (or an on-path attacker driving an established session) over the network, yielding remote memory corruption and at minimum a reliable denial of service. The defect was present in the WireGuard implementation shipped in Zephyr 4.4.0. The fix adds an explicit data_len > CONFIG_WIREGUARD_BUF_LEN rejection and corrects the linearize call to pass net_buf_max_len(buf) as the destination capacity. |
| The nRF70 Wi-Fi driver's power-save event handler nrf_wifi_event_proc_get_power_save_info() in drivers/wifi/nrf_wifi/src/wifi_mgmt.c copied TWT (Target Wake Time) flow entries from an nrf_wifi_umac_event_power_save_info event into the fixed-size twt_flows[WIFI_MAX_TWT_FLOWS] (8-element) array of a caller-supplied struct wifi_ps_config, looping over event-provided num_twt_flows without validating it against WIFI_MAX_TWT_FLOWS or checking event_len. When num_twt_flows exceeds 8, the handler writes past the destination array (which is typically on the caller's stack, e.g. the wifi ps shell command) -- an out-of-bounds write of ~40-byte TWT entries -- and reads twt_flow_info[i] past the event buffer. The event is delivered by the nRF70 co-processor firmware in response to a host-initiated power-save GET, so reaching the overflow requires the firmware to emit a malformed or out-of-range event; the trust boundary is host-to-trusted-coprocessor rather than a direct remote-AP write, with over-the-air influence on the flow count being indirect and bounded by the 3-bit TWT flow-id space. Affected: builds with CONFIG_NRF70_STA_MODE on releases through v4.4.0. The fix rejects events with num_twt_flows > WIFI_MAX_TWT_FLOWS or with event_len shorter than the claimed entries, and adds a NULL check on the caller buffer. |
| In Zephyr's experimental USB host stack (CONFIG_USB_HOST_STACK), usbh_device_disconnect() (subsys/usb/host/usbh_device.c) freed the root usb_device slab object without clearing the cached pointer ctx->root. The bus removal handler dev_removed_handler() (subsys/usb/host/usbh_core.c) decides what to tear down solely from ctx->root, checking only that it is non-NULL.
Because UHC controller drivers (e.g. uhc_max3421e, uhc_mcux_common) synthesize UHC_EVT_DEV_REMOVED directly from physical bus line state with no debounce or state guard, an attacker with physical USB access (or a rogue device that bounces its connection) can deliver a second device-removed event after a root device disconnect. The handler then re-enters usbh_device_disconnect() with the dangling pointer, locking a mutex inside the freed object (use-after-free), removing the freed node from the device list, and calling k_mem_slab_free() on the already-freed block (double-free). If the slab block has been reissued to a newly attached device in between, this corrupts a live object.
Impact is denial of service (crash) and memory corruption; the attack vector is physical/local. The flaw was introduced in v4.4.0 by the connect/disconnect refactor and is fixed by clearing ctx->root in usbh_device_disconnect() before freeing. |
| The Nuvoton NuMaker HSUSBD USB device-controller driver (drivers/usb/udc/udc_numaker.c) armed the control Data IN stage unconditionally (base->CEPTXCNT = len in numaker_hsusbd_ep_trigger). Because the HSUSBD hardware cannot disarm a control Data IN already armed for a previous transfer, a USB host that cancels an in-flight control transfer (timeout) and then issues a new SETUP packet can drive the driver out of sync: stale data may be transmitted in the new transfer and the control endpoint can become permanently stuck NAK'ing every subsequent control transfer.
A malicious or buggy host (physical/adjacent attacker driving the bus) can repeatedly cancel-and-re-SETUP to wedge the device's USB control endpoint, denying service to the device's USB function (the device stops enumerating/responding on the control pipe) until a USB reset or re-plug. The flaw is an availability-only denial of service; the FIFO copy loops (bounded by net_buf length and the hardware BUFFULL flag) and the net_buf lifecycle are independent of the arming desync, so there is no out-of-bounds access, use-after-free, or information leak.
The fix monitors the IN-token and new-SETUP events (k_event) and only arms control Data IN when an IN token is present and no new SETUP has arrived, cancelling the current transfer on a new SETUP. Affects boards using the Nuvoton NuMaker HSUSBD controller (CONFIG_UDC_NUMAKER with DT_HAS_NUVOTON_NUMAKER_HSUSBD_ENABLED); shipped in v4.4.0. |
| The Bluetooth BAP Broadcast Assistant GATT client in subsys/bluetooth/audio/bap_broadcast_assistant.c reassembled remote Broadcast Receive State data into a single file-static net_buf_simple (att_buf, BT_ATT_MAX_ATTRIBUTE_LEN = 512 bytes) shared by all connection instances, while the BUSY flag, long-read handle, and reset/offset state were per-connection.
When the device acts as a Broadcast Assistant connected to multiple Scan Delegator peripherals, notification and long-read callbacks from different connections interleave on the shared buffer: the append in notify_handler (net_buf_simple_add_mem at the not-busy branch) performs no tailroom check, so receive-state notifications from two or more delegators accumulate on the same 512-byte buffer and, with a sufficiently large configured ATT MTU (BT_L2CAP_TX_MTU up to 2000) and two-to-three concurrent connections, write past the buffer into adjacent .bss (net_buf_simple_add only asserts in debug builds).
Even below the overflow threshold, one connection's net_buf_simple_reset zeroes the shared length while another connection's reassembly and GATT read offset are in flight, mixing one peer's data into another's parse. A malicious or compromised Scan Delegator (or two colluding peers) over BLE can trigger this, causing out-of-bounds writes (memory corruption / denial of service) and cross-connection data corruption.
The fix moves the buffer into the per-connection instance struct so each connection reassembles into its own buffer. Affects Zephyr releases shipping the Broadcast Assistant with the shared buffer, including v4.4.0 and earlier. |
| The Dhara flash translation layer disk driver (drivers/disk/ftl_dhara.c) implemented the dhara_nand_ callbacks so that, on a flash error, the error code was written unconditionally through the caller-supplied dhara_error_t err pointer (e.g. *err = DHARA_E_ECC in dhara_nand_read, and similar in dhara_nand_erase/prog/copy). The upstream Dhara library calls these callbacks with err == NULL along its journal-resume binary search: find_last_checkblock() invokes find_checkblock(j, mid, &found, NULL), which forwards the NULL pointer into dhara_nand_read(). This path runs during disk_ftl_access_init() -> dhara_map_resume() whenever the FTL disk is mounted/initialised. If a flash read error (uncorrectable ECC, bad block, controller error) occurs on one of the probed checkpoint pages, the driver dereferences and writes to NULL, faulting the kernel (denial of service). The trigger is conditioned on the NAND medium content/health, which can be influenced by media wear, induced faults, or a corrupted/crafted on-flash image. The fix routes all error assignments through the library's NULL-safe dhara_set_error() helper. Affects Zephyr v4.4.0, where the driver was introduced. |
| The MAX32xxx USB device controller driver (drivers/usb/udc/udc_max32.c, compatible adi_max32_usbhs) dereferenced an endpoint buffer in its OUT and IN transfer-completion handlers without checking it for NULL. udc_event_xfer_out_done() called net_buf_add(buf, ep_request->actlen) immediately after buf = udc_buf_get(ep_cfg), where udc_buf_get() returns NULL when the endpoint FIFO is empty. A transfer-completion event is queued from interrupt context and processed asynchronously by the driver thread; between queuing and processing, the endpoint FIFO can be drained by host-controlled control flow — in particular udc_setup_received() drains the EP0 OUT/IN FIFOs whenever a new SETUP packet arrives, and dequeue/disable/purge paths drain it likewise. A USB host that aborts an in-flight EP0 control transfer with a new SETUP packet (legal USB behavior) can therefore cause a stale XFER_OUT_DONE event to be processed against an empty FIFO, producing net_buf_add(NULL, ...), a near-NULL pointer dereference that faults and crashes the device. No authentication is required; the attacker is the USB host the device is connected to (physical bus access). Impact is denial of service (device crash). The defect was introduced when the MAX32 UDC driver was added and shipped in Zephyr v4.4.0. The fix adds NULL-buffer checks that return early with UDC_EVT_ERROR/-ENOBUFS in both the OUT-done and IN-done handlers. |
| Zephyr's DNS resolver detects mDNS (.local) queries in dns_resolve_name_internal() (subsys/net/lib/dns/resolve.c) with memcmp(strrchr(query, '.'), ".local", 7), which always reads a fixed 7 bytes from the suffix pointer. When the resolved hostname's final label is shorter than 7 bytes (e.g. names ending in .org, .com, .net, .io, or a trailing dot), the comparison reads 1-2 bytes past the string's NUL terminator. The hostname (query) is the caller-supplied name passed through the standard getaddrinfo()/dns_get_addr_info()/dns_resolve_name() path and is influenceable by operators or remote inputs (server names from configuration, parsed URLs, or app-facing interfaces). On a tightly-sized buffer with no slack (for example a userspace getaddrinfo call where the hostname is copied with k_usermode_string_alloc_copy to exactly strlen+1 bytes), the over-read crosses the allocation boundary; if that boundary is unmapped (guard page, memory-domain boundary under MPU, or an address sanitizer) the over-read faults, causing a denial of service. The over-read bytes are never returned, so there is no information disclosure. The flaw is compiled only when CONFIG_MDNS_RESOLVER is enabled, exists since v1.10.0, and is fixed by replacing the fixed-length memcmp with a NUL-safe strcmp(ptr, ".local"). |
| Zephyr's DNS resolver (subsys/net/lib/dns) parses resource records from DNS responses in dns_unpack_answer(), which validated only the fixed RR header (type, class, TTL, rdlength) and accepted any attacker-declared rdlength, including one extending past the end of the received datagram. The TXT and SRV consumers in dns_validate_record() (resolve.c) then read up to rdlength bytes (clamped only to a record-type maximum such as DNS_MAX_TEXT_SIZE, default 64, not to the packet) from the receive buffer via memcpy without their own bounds check, and pass the result to the application's resolve callback. A malicious or spoofed DNS server, an on-path attacker forging UDP DNS replies, or (with mDNS/LLMNR enabled) any LAN node can craft a truncated TXT or SRV response that causes an out-of-bounds read of adjacent receive-pool memory; the disclosed stale bytes (residual contents of prior DNS packets / uninitialized pool memory) are returned to the application as TXT/SRV record contents, an information leak, and may in some configurations cross the allocation boundary and fault, causing a denial of service. The read is bounded (~64 bytes for TXT, ~6 for SRV) and read-only (no write). The fix rejects any record whose declared rdata extends past dns_msg->msg_size at the single chokepoint in dns_unpack_answer(). Affected: v4.3.0 and v4.4.0. |
| The Zephyr Bluetooth controller ISO Adaptation Layer (subsys/bluetooth/controller/ll_sw/isoal.c) fails to validate the length field of a framed ISO PDU start segment. Per the Bluetooth specification a start segment (sc=0) always carries a 3-byte time_offset, so its segment-header len must be at least PDU_ISO_SEG_TIMEOFFSET_SIZE (3). isoal_check_seg_header() accepted start segments with len < 3 as valid, and isoal_rx_framed_consume() then computed length = seg_hdr->len - 3 in a uint8_t, underflowing to 253-255 when len is 0-2. That oversized length is passed to isoal_rx_append_to_sdu(), whose copy is clamped only against the destination SDU buffer size, not the source PDU length, so up to ~255 bytes of controller memory beyond the received PDU are copied (via sink_sdu_write_hci()/net_buf_add_mem) into an HCI ISO data packet and delivered to the host. The PDU and its segment headers are entirely attacker-controlled and arrive over the air, reachable through both the CIS and BIS-sync HCI data paths (hci_driver.c) and the vendor data path (ull_iso.c), so a remote CIS peer or a broadcaster the device is synced to can trigger an out-of-bounds read causing information disclosure to the host and potential denial of service (faults or malformed oversized HCI ISO packets). The flaw affects all Zephyr releases since framed ISO reception was introduced in v3.0.0. The fix rejects sc=0 segments with len < 3 in isoal_check_seg_header() and adds a guard before the subtraction in isoal_rx_framed_consume(). |
| The Zephyr net_buf library (lib/net_buf/buf.c) manipulated both of its reference counts -- the per-header buf->ref and the per-data-block ref_count at the start of each variable/heap data allocation -- with plain non-atomic C operators (buf->ref++, if (--buf->ref > 0), if (--(*ref_count))). The API is documented as self-synchronizing: callers may share one buffer across threads (e.g. via k_fifo) and each holder independently calls net_buf_unref() with no surrounding lock. Under true concurrency (SMP, or single-core preemption between the non-atomic load and store while another context unrefs the same buffer), two holders can both observe the same prior reference value and both conclude they are the last reference. For heap/variable-data pools (mem_pool_data_unref/heap_data_unref, used by zbus message subscribers, the IP stack RX/TX buffers when CONFIG_NET_BUF_FIXED_DATA_SIZE=n, capture, wireguard, ISO-TP and usbip) this produces a double k_heap_free()/k_free() of the same block -- heap-metadata corruption and a use-after-free on the heap-hardening poison pattern. For the per-header refcount the buffer is returned to the pool free LIFO twice for any pool type (including fixed-data pools used by Bluetooth and networking), corrupting the free list so a later allocation hands the same buffer to two owners. The fix converts both refcounts to atomic_inc/atomic_dec (overlaying buf->ref in an atomic_t-sized union and changing the data-block refcount from uint8_t to atomic_t). Impact is gated on genuine concurrency and on an application architecture that shares one buffer among multiple independent unref'ers; the trigger is a refcount/timing race rather than packet content, so an external attacker has at most weak indirect influence over the race window. Affects all Zephyr releases through v4.4.0. |
| A race condition in the Zephyr Bluetooth Classic RFCOMM host stack (subsys/bluetooth/host/classic/rfcomm.c) mishandles a simultaneous bidirectional session disconnect. When the local device has initiated a session teardown (state BT_RFCOMM_STATE_DISCONNECTING, DISC sent, RTX timer armed) and the connected peer concurrently sends its own DISC frame for dlci 0, rfcomm_handle_disc() invokes rfcomm_session_disconnected(), which unconditionally forced the session to BT_RFCOMM_STATE_DISCONNECTED without ever calling bt_l2cap_chan_disconnect().
Because the recovery timer was also cancelled and a later UA is ignored in the DISCONNECTED state, the session becomes permanently wedged: the underlying L2CAP channel is never released and the session slot in the fixed bt_rfcomm_pool[CONFIG_BT_MAX_CONN] array is never reclaimed (its conn pointer stays set).
Subsequent bt_rfcomm_dlc_connect() calls on that connection fail with -EINVAL due to the invalid session state, so RFCOMM service is denied for that peer, and repeated occurrences can exhaust the session pool. The DISC frame is peer-controlled over the air, but exploitation requires the peer's DISC to collide with a local-initiated disconnect (a high-complexity timing race). Impact is availability/resource-leak only; there is no memory-safety, confidentiality, or integrity consequence. The defect shipped in released versions (present in v4.4.0 and earlier).
The fix only transitions to DISCONNECTED when the session is not already in DISCONNECTING, preserving the proper L2CAP teardown path. |
| The asynchronous SNTP client in Zephyr (subsys/net/lib/sntp/sntp.c, sntp_close_async) closed the UDP socket file descriptor directly from the calling thread immediately after detaching it from the network socket service, without synchronizing with the socket-service poll thread.
The socket service thread polls each socket via zvfs_poll, which (in zsock_poll_prepare_ctx) registers a k_poll_event pointing into the socket's net_context (&ctx->recv_q) and then blocks in k_poll without holding a reference or lock. net_context objects are allocated from a fixed pool (contexts[CONFIG_NET_MAX_CONTEXTS]) and reused after close.
When sntp_close_async is invoked from a different thread than the poll thread (in the in-tree consumer subsys/net/lib/config/init_clock_sntp.c, the SNTP timeout handler runs on the system workqueue while the socket service thread is blocked in poll on the same fd), the close frees and may reuse the net_context while the poll thread still has a poller node linked into the freed object, resulting in a use-after-free / object confusion of kernel poll structures.
The SNTP timeout path is the normal no-response failure mode, so a network peer or off-path attacker who drops or delays the SNTP/NTP response can drive the racing close repeatedly (and periodically with NET_CONFIG_SNTP_INIT_RESYNC). The most likely consequence is a crash of the networking thread (denial of service), with potential memory corruption when the freed context slot is reallocated.
The fix defers the close to the socket service thread itself via net_socket_service_close (NET_SOCKET_SERVICE_CLOSE_SOCKETS), so the same thread that polls performs the close, eliminating the race. Affected releases: v4.2.0 through v4.4.0. |
| The USB CDC-NCM device class (subsys/usb/device_next/class/usbd_cdc_ncm.c) ignores the return value of usbd_ep_enqueue() in its ethernet transmit callback cdc_ncm_send(). When the enqueue fails, the function still calls k_sem_take(&data-sync_sem, K_FOREVER), blocking on a completion semaphore that is only ever signaled from the bulk-IN transfer-completion callback. Because nothing was enqueued, that callback never fires and the calling thread — a shared network traffic-class TX thread — deadlocks permanently while holding the interface TX lock, halting transmission until reboot (and leaking the transmit buffer).
The enqueue fails under conditions controlled by the attached USB host: usbd_ep_enqueue() returns -EPERM whenever the bus is suspended (a standard, persistent host operation), and the underlying udc_ep_enqueue() returns -EPERM/-ENODEV on disconnect, bus reset, or endpoint disable. The cdc_ncm_send() guard only checks the DATA_IFACE_ENABLED and IFACE_UP flags, not the suspended state, so a packet transmitted while the host holds the bus suspended reaches the failing enqueue and deadlocks the TX path.
The realistic trigger is a bus suspend that occurs while the exported network interface is active and has traffic to send — host sleep, USB selective/auto-suspend, or hub power management — after which any device-originated packet deadlocks the path, recoverable only by reboot. The impact is a persistent loss of the virtual network connection between the host's NCM interface and the Zephyr device; because the deadlocked thread is a shared traffic-class TX thread, egress on other network interfaces can stall as well. There is no memory corruption or information disclosure.
The defect was introduced with the CDC-NCM driver and shipped in releases through v4.4.0; it is fixed by checking the usbd_ep_enqueue() return value and freeing the buffer before the blocking wait. |
