Just published.
The newest CVEs, straight from CVE.org and NVD.
-
CVE-2026-40660
A heap-based out-of-bounds (OOB) read vulnerability in the AMD Compressonator SDK could allow an attacker with local access to read out-of-bounds heap memory, potentially resulting in information disclosure or application crash.
-
CVE-2026-40661
An out-of-bounds write (OOW) condition in the AMD Compressonator SDK could allow an attacker with local access to supply a specially crafted DirectDraw Surface (DDS) file to a user or an automated pipeline, potentially resulting in heap corruption or denial of service (DoS).
-
CVE-2026-43603
A NULL pointer dereference in the Linux GPU driver could potentially lead to a kernel crash and result in denial of service.
-
CVE-2026-121508
Use-After-Free in µD3TN BPv6 Fragmenter Cleanup allows Unauthenticated Remote Attacker to Cause DoS on Allocation Failure
-
CVE-2026-121510
Remote Unauthenticated Heap Out-of-Bounds Write in µD3TN BPv6 Administrative Record Parser allows Unauthenticated Remote Attacker to Reliably Cause DoS
-
CVE-2026-121511
Heap Out-of-Bounds Read in µD3TN BPv6/BPv7 Fragmenter allows Local Attacker to Read Memory and Reliably Cause DoS
-
CVE-2026-121513
Off-by-one Heap Overflow in µD3TN v0.15.1 SPP Parser allows Unauthenticated Remote Attacker to Reliably Cause DoS
-
CVE-2026-121505
An out-of-bounds read in the NTS-KE client response parser (nts_client_process_response_core) in NTPsec before 1.2.4 allows a malicious NTS-KE server to read the client's stack memory, including pointers that defeat ASLR, which the client then sends in cleartext as NTS cookies, via an NTS-KE record whose declared length exceeds the data received.
-
CVE-2026-121506
An out-of-bounds read and write in NTS client extension field parsing (extens_client_recv) in NTPsec before 1.2.5 allows a remote attacker who can send NTP server-mode replies to an NTS client to corrupt heap memory and crash ntpd, and, combined with the NTS-KE client out-of-bounds read fixed in 1.2.4, to bypass NTS authentication and set the client's time, via an NTS Authenticator and Encrypted Extension Fields field whose nonce length exceeds the field.
-
CVE-2026-121507
Double Free in µD3TN BPv6 Fragmenter Block Duplication allows Unauthenticated Remote Attacker to Cause DoS on Allocation Failure
-
CVE-2026-121417
SurrealDB before 3.3.0 contains a use-after-free vulnerability in surrealdb-core KV backends that pass borrowed transaction state into affinitypool::spawn_local closures relying on future Drop for safety. Rust embedders that leak a dispatched future (e.g. via mem::forget) and drop the transaction cause pool workers to read and write freed memory, risking crashes or memory corruption.
-
CVE-2026-121408
FreeRDP before 3.32.2 contains a use-after-free in the urbdrc USB redirection channel because FindChannelById returns a channel pointer without a reference and the control-channel close path does not take the device manager's loading_lock. A malicious RDP server can send a control-channel CLOSE_REQUEST while the client announces a USB device, freeing the channel before the announcement thread reads channel->Write and crashing the client.
-
CVE-2026-121401
FreeRDP before 3.32.2 contains a denial of service vulnerability where rdg_process_packet dispatches a PKT_TYPE_EXTENDED_AUTH_MSG without verifying negotiated mode or auth context. A malicious or compromised RD Gateway can send the extended-auth packet before tunnel authentication, triggering a null-pointer dereference that crashes the client.
-
CVE-2026-121402
FreeRDP before 3.32.2 contains a NULL pointer dereference vulnerability that allows malicious RDP servers to crash xfreerdp clients by sending a zero-width RDPGFX ResetGraphics PDU. The failed resize in gdi_resize_ex leaves the primary display null without ending the session, so a later bitmap update crashes the client in gdi_Bitmap_Paint.
-
CVE-2026-121397
FreeRDP before 3.32.2 contains a heap buffer over-read vulnerability in save_png_to_buffer() that allows malicious RDP servers to leak client memory by sending 16 bpp DIB clipboard data. Attackers can offer a CF_DIB BI_BITFIELDS image that, when converted to PNG for a local application, embeds client heap bytes into the clipboard image.
-
CVE-2026-121398
FreeRDP before 3.32.2 contains a cross-thread use-after-free vulnerability in the freerdp-proxy dyn-channel-dump module during session teardown. A malicious RDP target can send dynamic channel data while the front client disconnects, letting the client thread read freed ChannelData and crash the proxy process.
-
CVE-2026-121387
FreeRDP 2.0.0 before 3.32.2 contains an out-of-bounds read vulnerability in freerdp_dsp_decode_gsm610 that allows malicious RDP servers to read past audio buffers by sending truncated GSM610 frames. Attackers can negotiate GSM610 on rdpsnd and send a WAVE2 message shorter than a 33- or 32-byte WAV49 frame to crash the client.
-
CVE-2026-100493
An integer overflow in the DDS RGB888 loader of AMD Compressonator SDK could allow an attacker to supply a crafted DDS file that causes an undersized heap allocation and a subsequent out-of-bounds read, potentially resulting in process termination or limited disclosure of heap memory.
-
CVE-2026-85034
The Bluetooth host's GATT client passed the value of a received Handle Value Notification, Handle Value Indication or Multiple Handle Value Notification to bt_gatt_subscribe_params::notify callbacks using only the length derived from the received ATT PDU. bt_att_recv() in subsys/bluetooth/host/att.c validates only a minimum PDU length and does not enforce the negotiated ATT MTU, and the fixed ATT L2CAP channel in subsys/bluetooth/host/l2cap.c applies no rx.mtu check either, so the only ceiling was the ACL reassembly buffer CONFIG_BT_BUF_ACL_RX_SIZE. Values longer than the 512-octet maximum attribute length defined by the Bluetooth Core Specification (Vol 3, Part F, Section 3.2.9) could therefore be handed to subscribers that size their buffers to that documented limit. A remote peer that a device has connected and subscribed to — a malicious, spoofed or compromised GATT server within radio range — triggers the flaw by sending a notification or indication whose value exceeds 512 octets on a subscribed handle. In the LE Audio unicast client this reaches a concrete out-of-bounds write: unicast_client_ep_notify() in subsys/bluetooth/audio/bap_unicast_client.c copies the notification into the 512-byte att_buf of struct unicast_client with net_buf_simple_add_mem() and no tailroom check whenever length equals bt_audio_get_max_ntf_size(), i.e. the negotiated ATT MTU minus three. notify_handler() in subsys/bluetooth/audio/bap_broadcast_assistant.c has the same shape. Reaching the overflow requires a build whose ATT MTU exceeds 515 octets (CONFIG_BT_L2CAP_TX_MTU and CONFIG_BT_BUF_ACL_RX_SIZE set for full-length attribute support); the shipped defaults are far smaller. The overflow writes peer-controlled bytes past att_buf directly over the adjacent struct net_buf_simple net_buf — its data, __buf, len and size fields — which the subsequent long-read accumulation then uses as its destination, turning a small linear overflow into a controlled write. net_buf_simple_add() only asserts on insufficient tailroom unless CONFIG_NET_BUF_HARDENING is enabled, and that option is disabled by default, so ordinary production builds perform the copy. Beyond the in-tree LE Audio clients, any application subscriber that sized its receive buffer to the specification's 512-octet limit was exposed to the same overflow. The fix rejects notification, indication and multiple-notification values longer than BT_ATT_MAX_ATTRIBUTE_LEN in call_notify_cb_and_maybe_unsubscribe() in subsys/bluetooth/host/gatt.c before any subscriber callback is invoked, so no subscriber can be handed more than the specified maximum. Remaining tuples of a multiple handle value notification are still delivered and oversized indications are still confirmed.
-
CVE-2026-85035
The can_add_rx_filter_msgq() system call installs a CAN RX filter whose callback (can_msgq_put() in drivers/can/can_common.c) is handed the caller's struct k_msgq * as user_data. The controller driver stores that raw pointer in its filter table and dereferences it from the RX path every time a matching frame arrives. The system call verifier z_vrfy_can_add_rx_filter_msgq() in drivers/can/can_handlers.c validated the queue only with K_SYSCALL_OBJ(), which is a point-in-time check of type, initialization and caller permission and says nothing about the object's lifetime. On a build with CONFIG_USERSPACE=y and CONFIG_DYNAMIC_OBJECTS=y, an unprivileged user-mode thread that has been granted access to a CAN device can obtain a message queue whose lifetime it fully controls: k_object_alloc(K_OBJ_MSGQ) followed by k_msgq_alloc_init(). The lifetime of such an object is driven solely by its permission bitmask, and a pointer retained by the kernel is not represented there. After registering the filter, the thread calls k_object_release() — or simply terminates — and unref_check() frees the struct k_msgq storage while the CAN driver still holds the pointer. The next matching CAN frame causes k_msgq_put() to run in supervisor context against freed memory that the attacker can have groomed with another kernel allocation: the frame payload is memcpy()d through the reused block's write_ptr, and the queue's wait_q (and poll_events under CONFIG_POLL) list nodes are manipulated. This yields kernel-heap corruption reachable from a user-mode thread, defeating the isolation CONFIG_USERSPACE is intended to provide; the same dangling pointer also arises benignly when an application thread exits without calling can_remove_rx_filter(). The fix validates the queue with K_SYSCALL_OBJ_STATIC(), which additionally rejects objects carrying K_OBJ_FLAG_ALLOC, so only statically defined queues (CAN_MSGQ_DEFINE()) may be registered from user mode. Supervisor-mode callers bypass the verifier and must themselves keep the queue valid until the filter is removed.
-
CVE-2026-85032
The MCTP over USB device class in subsys/pmci/mctp/mctp_usb.c splits RX handling across two threads: the system workqueue runs mctp_usb_class_out_work(), which parses host OUT data byte by byte and fills a packet buffer at usb->rx_pkt, while the USB device stack thread runs mctp_usb_class_disable(). The disable handler used k_work_cancel(), which removes a queued work item but returns immediately without waiting for a handler that is already running, and then called mctp_usb_reset_rx_state() to mctp_pktbuf_free() the packet buffer the worker may still be using. Both entry points into mctp_usb_class_disable() are driven by the attached USB host: a SET_CONFIGURATION request and a USB bus reset both reach it via usbd_config_reset() in subsys/usb/device_next/usbd_config.c. A host that streams partial MCTP frames on the bulk OUT endpoint and repeatedly forces a configuration change or bus reset therefore controls both the data being parsed and the moment the teardown runs, and can retry the race indefinitely without credentials or user interaction. If the free lands while the worker holds the pointer, the worker writes host-supplied bytes into freed memory of the dedicated MCTP heap (CONFIG_MCTP_HEAP_SIZE in subsys/pmci/mctp/mctp_memory.c), corrupting recycled buffers or sys_heap free-block metadata; this window requires genuine parallelism, i.e. CONFIG_SMP or a preemptible CONFIG_SYSTEM_WORKQUEUE_PRIORITY. With the default cooperative workqueue the worker can still yield inside mctp_bus_rx() (application callback or a control-message reply blocking in mctp_usb_tx()), after which it resumes and calls mctp_pktbuf_free(usb->rx_pkt) on a pointer the teardown has NULLed, dereferencing NULL and faulting on targets with null-pointer detection. The fix replaces the cancel with k_work_cancel_sync() plus a new struct k_work_sync out_work_sync, so disable waits for the RX worker to finish before draining the FIFO and freeing the packet buffer. The affected code is the MCTP subsystem, which is marked experimental, built only with CONFIG_MCTP_USB.
-
CVE-2026-120572
Incorrect boundary conditions in the Layout: Block and Inline component. This vulnerability was fixed in Firefox 159.0.1 and Thunderbird 160.
-
CVE-2026-99270
In the Linux kernel, the following vulnerability has been resolved: tcp: Fix potential UAF in reqsk_timer_handler(). When TCP socket migration fails at inet_ehash_insert() in reqsk_timer_handler(), we jump to the no_ownership: label and free the new reqsk immediately with __reqsk_free(). Thus, we must stop the new reqsk's timer before jumping to the label, but the timer might be missed since the cited commit, resulting in UAF. As we are in the original reqsk's timer context, we can safely call timer_delete_sync() for the new reqsk. Let's pass false to __inet_csk_reqsk_queue_drop() to stop the new reqsk's timer.
-
CVE-2026-99246
In the Linux kernel, the following vulnerability has been resolved: tipc: protect node reset trace dump with node lock The tipc_node_reset_links trace event asks tipc_node_dump() to walk the node's link entries. Unlike the other node events that request link data, this event runs without the node lock. This permits bearer teardown to free a link while the trace callback is dumping it: CPU 0 CPU 1 trace_tipc_node_reset_links() tipc_node_dump() l = n->links[0].link tipc_node_write_lock() kfree(l) n->links[0].link = NULL tipc_node_write_unlock() tipc_link_dump(l) tipc_link_dump() then dereferences the stale pointer. KASAN reported: BUG: KASAN: slab-use-after-free in tipc_link_dump Read of size 4 by task poc/115 Call Trace: tipc_link_dump+0x10cb/0x16b0 tipc_node_dump+0x4bb/0x740 trace_event_raw_event_tipc_node_class+0x258/0x360 tipc_node_reset_links+0x14d/0x1a0 tipc_rcv+0x13f5/0x3030 tipc_udp_recv+0x4e3/0x670 Allocated by task 0: tipc_link_create+0x1e1/0x1020 tipc_node_check_dest+0x7d2/0x11a0 tipc_disc_rcv+0xdbf/0x1430 Freed by task 89: kfree+0x131/0x3c0 tipc_node_link_down+0x267/0x4b0 tipc_node_delete_links+0xec/0x160 bearer_disable+0x107/0x260 Take the node write lock around the trace event. This serializes the dump against tipc_node_link_down(delete=true), which frees the link under the same write lock.
-
CVE-2026-118445
A use-after-free vulnerability in the pybddisasm Python bindings affects structure-valued members of INSTRUX objects. These members are exposed through Python objects that reference memory owned by the parent INSTRUX object. A retained reference to a member does not keep the parent allocation alive. Accessing the member after the parent INSTRUX object has been freed therefore accesses freed memory. Depending on how the member is used and whether the freed allocation has been reused, this may expose process memory or cause the Python process to crash. Exploitation requires an application to retain and subsequently access a member reference after the parent object has been freed. The pybddisasm package is no longer maintained.
-
CVE-2026-118446
A use-after-free vulnerability in the pybddisasm Python bindings affects array-valued members of INSTRUX objects, including ND_OPERAND arrays. These arrays are exposed as Python lists whose elements reference memory owned by the parent INSTRUX object. Retaining a list or an element does not keep the parent allocation alive. Accessing an element after the parent INSTRUX object has been freed therefore accesses freed memory. Depending on how the element is used and whether the freed allocation has been reused, this may expose process memory or cause the Python process to crash. Exploitation requires an application to retain and subsequently access an array element after the parent object has been freed. The pybddisasm package is no longer maintained.
-
CVE-2026-118447
A use of uninitialized memory vulnerability in the pybddisasm Python bindings may disclose process memory when instruction decoding fails. The nd_decode, nd_decode_ex, nd_decode_ex2, and nd_decode_with_context functions may return an INSTRUX object whose contents have not been initialized after an unsuccessful decoding operation, including failures caused by invalid parameters. Accessing members of the returned object may expose residual process memory. Exploitation requires an application to access and expose the returned object contents after a decoding failure. The pybddisasm package is no longer maintained.
-
CVE-2026-118448
A use of uninitialized memory vulnerability in the pybddisasm Python bindings affects nd_to_text. If the underlying text conversion fails, including failures caused by invalid parameters, the output buffer may remain uninitialized while still being converted into a Python return value. This may disclose residual process memory or cause the Python process to crash. Exploitation requires an application to invoke nd_to_text in a way that causes conversion to fail and to return or otherwise use the resulting output. The pybddisasm package is no longer maintained.
-
CVE-2026-109161
Insufficient bounds checking in the Digital I/O function may allow a remote attacker with high privileges to cause a denial of service.
-
CVE-2026-113421
In Eclipse ThreadX versions 6.0.1 through 6.5.2.202603, ThreadX Modules does not revoke a module's trace-buffer-full callback when the module is stopped and unloaded. With event tracing enabled and a kernel-owned trace buffer still active, a later buffer wrap invokes the callback from privileged thread context after the module's code has been released, so a user-mode module can have its code run with kernel privileges after unload, or, once the memory is reused, redirect control flow or cause a fault.
-
CVE-2026-118444
Out-of-bounds read vulnerabilities in the pybddisasm Python bindings may disclose adjacent process memory during instruction decoding. The nd_decode function may read up to 15 bytes without checking the actual length of the supplied Python buffer. The nd_decode_ex, nd_decode_ex2, and nd_decode_with_context functions pass the caller-supplied buffer size to the decoder without validating it against the actual length of the Python buffer. Data read beyond the buffer may be exposed through nd_get_instrux_bytes, which returns a fixed-length list of 16 bytes. Exploitation requires an application to invoke the affected APIs with a short buffer or an oversized length and to expose the resulting bytes to an attacker. The pybddisasm package is no longer maintained.
-
CVE-2026-103653
An out-of-bounds read vulnerability in the Qt CoAP message parsing (QCoapInternalReply::createFromFrame()) of the Qt CoAP module. Processing a malformed CoAP message can cause an application using Qt CoAP to crash (loss of availability) or to continue operating on incorrect data (potential loss of integrity).
-
CVE-2026-120831
A flaw was found in libslirp in if_output(). When moving a session from the fast queue to the batch queue, the code unlinked and reinserted an arbitrary packet via ifm->m_nextpkt instead of the session queue head, corrupting queue pointers. A guest generating burst traffic could crash the host networking process, causing a denial of service.
-
CVE-2026-76955
A buffer overflow vulnerability in the zdp_wtp binary in Zyxel NWA110AX firmware versions through 7.12(ABTG.0)C0 could allow a LAN-based attacker to escalate privileges on an affected device.
-
CVE-2026-107441
An out-of-bounds write vulnerability in the CGI program of Zyxel GS1900-24E firmware versions through 2.90(AAHK.2)C0 could allow an unauthenticated adjacent attacker to exploit the flaw and potentially execute OS commands via a crafted HTTP request.
-
CVE-2026-120830
A flaw was found in libslirp in icmp_forward_error(). When building an ICMP error response, the length of the quoted original IPv4 packet was taken from the IP header without clamping it to the actual buffer length. A truncated or malformed packet could cause an out-of-bounds heap read, leaking host memory into the ICMP error packet returned to a guest or container.
-
CVE-2026-120752
PyMuPDF 1.25.3 before 2.0 contains a stack-based buffer overflow vulnerability in JM_set_annot_callout_line() that copies a caller-supplied number of callout points into a fixed three-element stack array. Attackers who control the callout argument passed to Page.add_freetext_annot() can supply four or more points to overwrite the stack frame, crashing the process or executing code.
-
CVE-2026-120740
A flaw was found in hivex. By supplying a specially crafted Windows registry hive file, an attacker can trigger an out-of-bounds memory read when hivex parses registry values containing malformed large data block lists. This issue can cause the application processing the registry file to crash, resulting in a Denial of Service (DoS).
-
CVE-2026-58977
In is_valid_header of timed_cache_item.c, there is a possible out-of-bounds write due to a missing bounds check. This could lead to local escalation of privilege with System execution privileges needed. User interaction is not needed for exploitation.
-
CVE-2026-58925
The report details a heap Use-After-Free (UAF) vulnerability within the `android.hardware.automotive.evs-default` HAL service, stemming from unsynchronized access to an `std::unordered_map`. An unprivileged application with access to the EVS manager can trigger this UAF by making concurrent calls to `setExtendedInfo` and `getExtendedInfo`, potentially enabling code execution within the privileged HAL process.
-
CVE-2026-58926
A Use-After-Free (UAF) vulnerability exists in the EVS HAL, allowing a privileged client to free a `buffer_handle_t` while the producer thread is still using it. This race condition, triggered via `doneWithFrame` and `setMaxFramesInFlight`, enables an attacker to manipulate freed heap memory, potentially leading to the leakage of file descriptors from the EVS HAL process.
-
CVE-2026-58927
The vulnerability is a Use-After-Free (UAF) in the EVS AIDL HAL, arising from a data race on an unsynchronized `std::weak_ptr` within `EvsEnumerator::openCamera()`. Specifically, concurrent calls from a privileged UID (e.g., `AID_AUTOMOTIVE_EVS`) can cause a torn `weak_ptr` state, leading to a virtual call through a freed object's vtable at `cameraObj->getCameraInfo(&desc)`. This allows for potential arbitrary code execution within the `hal_evs_default` process.
-
CVE-2026-58924
The vulnerability is a heap out-of-bounds read in the EVS HAL's `getStreamList()` method. An attacker can craft a `CameraDesc.metadata` byte array, which is parsed without validation, leading to an out-of-bounds read at an attacker-controlled offset within the HAL process's heap. The read data is then exfiltrated back to the attacker, potentially leaking sensitive information like pointers and buffer handles.
-
CVE-2026-29508
ReadyMedia (formerly MiniDLNA) versions 1.3.0 through 1.3.3 contain a stack-based buffer overflow vulnerability that allows remote attackers to corrupt stack memory by crafting a malicious DFF audio file with an oversized PROP chunk size value that is used unvalidated to allocate a variable-length stack array and read file data into it. Attackers can place a crafted DFF file in a directory monitored by ReadyMedia to trigger automatic parsing via inotify, achieving remote code execution on unprotected builds or denial of service on hardened builds with stack protections.
-
CVE-2026-59048
In rb_cpu_init of trace.c, there is a possible out-of-bounds read due to an incorrect bounds check. This could lead to local information disclosure with no additional execution privileges needed. User interaction is not needed for exploitation.
-
CVE-2026-59042
In __pkvm_pages_to_ppages of mmu.c, there is a possible out-of-bounds write due to a race condition. This could lead to local escalation of privilege with no additional execution privileges needed. User interaction is not needed for exploitation.
-
CVE-2026-59043
In arm_lpae_split_blk_unmap of io-pgtable-arm-common.c, there is a possible memory corruption due to a race condition. This could lead to local escalation of privilege with no additional execution privileges needed. User interaction is not needed for exploitation.
-
CVE-2026-59044
In multiple locations, there is a possible use-after-free due to a logic error in the code. This could lead to local escalation of privilege with no additional execution privileges needed. User interaction is not needed for exploitation.
-
CVE-2026-59045
In multiple locations, there is a possible memory corruption due to a race condition. This could lead to local escalation of privilege with no additional execution privileges needed. User interaction is not needed for exploitation.
-
CVE-2026-59046
In __pkvm_guest_relinquish_to_host of mem_protect.c, there is a possible out-of-bounds write due to improper input validation. This could lead to local denial of service with no additional execution privileges needed. User interaction is not needed for exploitation.
The newest published CVE records from CVE.org and NVD, refreshed every few minutes, with severity, exploit odds and fix status.