Just published.
The newest CVEs, straight from CVE.org and NVD.
-
CVE-2026-118649
n8n before 2.43.5 and 2.44.x before 2.44.1 accepts a client-supplied workflow ID and caches each workflow's owner project in memory by that ID, so a crafted ID such as __proto__ makes the Data Table project check match every project. An authenticated member who can create and run workflows can list, read, modify, and delete Data Table rows in all projects of the instance.
-
CVE-2026-118677
n8n before 2.41.4 and 2.42.x before 2.42.1 contains a missing confirmation vulnerability in the Local Gateway desktop app that allows its deep-link connect handler to silently re-pair the app with a different n8n instance. Attackers can lure users running the Local Gateway into opening a crafted n8n-computer-use:// link to connect their computer to an attacker-controlled instance, inheriting previously granted permissions and acting on the computer through the gateway.
-
CVE-2026-118678
n8n before 2.41.4 and 2.42.x before 2.42.1 contains a prototype pollution vulnerability that allows low-privileged users to alter built-in object behaviour by saving workflows with specially named connection keys. Attackers can add keys such as __proto__ to workflow connections, which the periodic history compaction task processes, causing logins and writes to fail with server errors.
-
CVE-2026-118644
n8n before 2.41.5 and 2.42.x before 2.42.2 passes a failed workflow's full execution context, including the identity of the user who started the run, to the configured error workflow. A project editor can set a colleague's workflow settings.errorWorkflow to an error workflow they own, inherit the colleague's identity when that workflow fails, and resolve the colleague's per-user dynamic credential secrets.
-
CVE-2026-118265
Echo is a Go web framework. From 5.0.0 until 5.4.1, Echo v5's ExtractIPFromRealIPHeader validates whether the client-supplied X-Real-IP value belongs to a trusted range instead of validating the direct network peer. A client reaching Echo directly, or through a proxy that preserves the header, can supply a loopback or private address and have Context.RealIP return that spoofed identity. Applications using the extractor for IP allowlists, rate limiting, audit logs, or forwarding headers can consequently trust an attacker-controlled address; the default direct extractor and Echo v4 are unaffected. This issue is fixed in 5.4.1.
-
CVE-2026-118249
Highcharts JS is a JavaScript charting library based on SVG. From 13.0.0 until 13.1.0, Highcharts builds series data columns by using point option keys directly as column IDs. Own enumerable __proto__ and constructor keys from parsed point objects can therefore direct integer-indexed writes to Object.prototype or the Object constructor. The inherited values affect reads and for-in enumeration across the page and can disrupt rendering or alter consuming application behavior without user interaction beyond rendering the chart. This issue is fixed in 13.1.0.
-
CVE-2026-118250
Highcharts JS is a JavaScript charting library based on SVG. From 6.1.0 until 11.4.9, 12.6.1, and 3.1.0, Highcharts follows dot-separated series.keys paths without filtering each segment. An attacker who controls series.keys and matching point data can use a path through __proto__ to assign arbitrary named properties to Object.prototype. Those properties are visible to all JavaScript in the page, and no user interaction is required beyond rendering the chart. This issue is fixed in 11.4.9, 12.6.1, and 3.1.0.
-
CVE-2026-118243
Highcharts JS is a JavaScript charting library based on SVG. From 1.0.0 until 4.2.2, Highcharts Dashboards stores DataCursor listener and state maps in plain objects keyed by DataTable ID. A __proto__ table ID supplied through dataPool connector configuration resolves to Object.prototype, and sync registration creates an enumerable property there whose suffix can be influenced through the sync group option. The property is visible on every object, and lasting sync states can throw and stop features such as extremes synchronization. Exploitation requires untrusted dashboard configuration and an enabled component sync, but no user interaction beyond loading the board. This issue is fixed in 4.2.2.
-
CVE-2026-118247
Highcharts JS is a JavaScript charting library based on SVG. From 1.0.0 until Dashboards 4.2.2, and from 2.0.0 until Grid Pro 3.2.0, Highcharts DataTable single-cell writers use an untrusted column ID as a key without rejecting __proto__ or constructor. DataTable.setCell, including Dashboards connector write-back and Grid Pro editing paths, can therefore write integer-indexed values onto Object.prototype or the Object constructor. In Dashboards releases with column aliases, a literal __proto__ ID is redirected by the alias map, while an alias targeting __proto__ reaches the unsafe write; later releases without aliases accept the unsafe ID directly. Those inherited values affect array and object reads and for-in enumeration throughout the page, which can alter application control flow or disrupt rendering. Exploitation requires control of the table or component column ID and a cell, point, or API write. This issue is fixed in Dashboards 4.2.2 and Grid Pro 3.2.0.
-
CVE-2026-118248
Highcharts JS is a JavaScript charting library based on SVG. From 1.0.0 until 4.2.2, Highcharts Dashboards edit mode walks editableOptions propertyPath segments without rejecting prototype-related keys. An attacker who controls editableOptions can provide a path through __proto__ or constructor, and a user's edit writes the configured value onto Object.prototype. The property then becomes visible to JavaScript throughout the page, with consequences determined by the consuming application. This issue is fixed in 4.2.2.
-
CVE-2026-116266
MIME::Head versions from 5.518 before 5.521 for Perl decode a partly encoded multipart boundary without flagging the parse as ambiguous. The multipart_boundary() method decodes RFC 2047 encoded words anywhere in the boundary, but the ambiguity check only fires when the whole value is one encoded word. A boundary such as abc=?us-ascii?Q?def?= is used as abcdef, and ambiguous_content() returns false. Encoded words are not allowed in parameters, so other mail readers split the message on the literal value. A remote sender can hide an attachment from a filter that relies on ambiguous_content() while the recipient's mail client shows it.
-
CVE-2026-116267
MIME::Parser versions before 5.521 for Perl allow memory exhaustion via indented error messages from a truncated, deeply nested message. Each stored parser diagnostic is indented by three spaces per nesting level. When a message nested N levels deep ends without its closing boundaries, every level logs an error at its own depth, so the stored diagnostics grow with the square of N. The defaults place no limit on nesting depth.
-
CVE-2026-116269
MIME::Field::ParamVal versions before 5.521 for Perl turn an RFC 2231 filename with a hyphenated language tag into an empty recommended filename. A parameter such as filename*=utf-8'en-US'payload.exe is converted to an encoded word with the language tag appended to the charset. The decoder strips the language suffix only when it is a single word, so it cannot resolve the charset UTF-8*EN-US and returns an empty string. The recommended_filename() method returns that empty string rather than falling back to the name parameter. Other mail readers return payload.exe. A remote sender can get an attachment past a policy that checks file extensions on the recommended filename.
-
CVE-2026-116271
MIME::Parser::FileUnder versions before 5.521 for Perl reuse a pre-created output directory and follow symlinks in it, letting local users redirect attachment writes. Each message is written to a directory named msg-<time>-<pid>-<n> under the output base, and an existing directory of that name is used as is. A file name counts as unused when nothing exists at that path, which holds for a dangling symlink, and the body file is then opened in a way that follows the link. When the output base is writable by other users, as with output_under("/tmp"), a local user can pre-create the directory with a symlink named like an attachment. The parser then writes the attachment to the link target with the privileges of the parsing process.
-
CVE-2026-116272
MIME::Field::ParamVal versions before 5.521 for Perl do not flag a duplicate boundary parameter whose name contains a nested comment. A parameter written boundary((c))="A" or boundary(\))="A" is stored under that literal name rather than as boundary. A second boundary parameter is then not reported as a duplicate, and MIME::Parser splits the message on it. Readers that skip comments as RFC 5322 allows take the first boundary instead. A remote sender can hide an attachment from a filter that relies on ambiguous_content() while the recipient's mail client shows it.
-
CVE-2026-116132
HTTP::Body versions before 1.23_01 for Perl leave the temporary file of an incomplete multipart upload on disk. The multipart parser creates a temporary file as soon as a file part starts, but adds it to the upload list only when the part completes. Uploads are only ever deleted from that list, by the cleanup option or by the caller, so a body that ends inside a file part leaves its temporary file where nothing can find it. A remote attacker can fill the temporary directory by repeating such requests, leaving one file behind for each.
-
CVE-2026-116135
HTTP::Body versions before 1.23_01 for Perl drop mixed-case percent escapes from urlencoded form fields instead of decoding them. The decode table holds only the all-lowercase and all-uppercase spelling of each hex pair, while the escape pattern matches either case. An escape such as %aB finds no entry and is replaced with nothing, joining the bytes on either side. Hex digits in percent escapes are case-insensitive, and conforming decoders return the byte. A remote attacker can send a field name or value that a conforming decoder in front of the application reads differently, so that pass%aBword reaches the application as password.
-
CVE-2026-116264
MIME::Field::ParamVal versions before 5.521 for Perl allow one invalid Content-Type parameter to hide the parameters after it, including the multipart boundary. Parameter parsing stops at the first parameter whose name it cannot read, such as x:y=1 or a name containing an 8-bit byte, and drops every parameter after it. When the boundary is among them, MIME::Parser keeps the multipart body as a single undecoded part without marking the parse as ambiguous. Mail readers that skip the invalid parameter still find the boundary. A remote sender can hide an attachment from a filter that relies on ambiguous_content() while the recipient's mail client shows it.
-
CVE-2026-116265
MIME::Field::ParamVal versions before 5.521 for Perl do not flag an ambiguous multipart boundary written as RFC 2231 continuations. The boundary ambiguity checks look only at a plain boundary= parameter, and a boundary split into RFC 2231 sections (boundary*0=, boundary*1= and so on) is joined after they have run. A repeated section number is concatenated rather than reported as a duplicate. From version 5.518, an encoded word or trailing whitespace in such a boundary also goes unflagged, although multipart_boundary() decodes or strips it. A remote sender can make MIME::Parser split a message on a different boundary than other mail readers, hiding an attachment from a filter that relies on ambiguous_content() while the recipient's mail client shows it.
-
CVE-2026-110181
PostgreSQL Anonymizer contains a vulnerability related to anonymization functions defined using SQL-standard function bodies. PostgreSQL analyses such functions at declaration time, while Anonymizer only attempts to check them at execution time. This problem can be exploited by any user with the TEMPORARY privilege. The issue is fixed in PostgreSQL Anonymizer 3.3.0 and later versions.
-
CVE-2026-110183
PostgreSQL Anonymizer contains a vulnerability related to static masking and statistics persistence. Statistics pertaining to the original data remain visible in pg_stats until an ANALYZE operation is performed on the table, either manually or by autovacuum. This issue is fixed in PostgreSQL Anonymizer 3.3.0 and later versions.
-
CVE-2026-105615
PostgreSQL's optimized statistics on a table are visible as long as you have SELECT privilege on said table. A masked role can therefore read a limited subset of the data through the statistics view's histograms and mcvs. The vulnerability is resolved in PostgreSQL Anonymizer 3.3.0 and later versions.
-
CVE-2026-118510
A type-handling flaw in the relationship query building of this MongoDB integration library can cause a check for the absence of a related record to be compiled with an identifier of the wrong data type. In an application that relies on such a check to decide which documents a request may see or act on, a user with basic privileges may be served, may change, or may remove documents that the check was intended to withhold. The failure is silent and produces no error.
-
CVE-2026-118581
A BSON encoding component of the MongoDB Java Driver does not verify that an externally supplied raw document's self-declared length agrees with the extent of the byte range it is given before that range is copied into an outgoing database command. An application that wraps untrusted raw bytes in such a document and uses it as part of a command can therefore emit a structurally inconsistent message, causing the database to interpret attacker-supplied trailing content as additional fields of the surrounding command. No authentication to the driver or database is required beyond the ability to influence those bytes through the victim application, and no user interaction is needed. The result is unintended modification of the command actually executed, which can alter the data read or written under the application's credentials.
-
CVE-2026-117617
Jenkins 2.587 and earlier, LTS 2.580.1 and earlier checks for symbolic links in the path of an extracted file only before creating its parent directories, not immediately before writing it, and allows symbolic links in artifacts to replace existing files or empty directories, allowing attackers with Item/Configure permission to write files to arbitrary locations on the Jenkins controller file system.
-
CVE-2026-110952
A Kafka broker that authenticates clients with SASL/OAUTHBEARER using the bundled JWKS-backed validator always verifies the JWT signature, but two further checks are only performed when explicitly configured: - The issuer (iss) claim is verified only when sasl.oauthbearer.expected.issuer is set. When it is unset, a validly-signed token carrying the expected audience is accepted regardless of which issuer minted it. - An audience (aud) claim is required only when sasl.oauthbearer.expected.audience is set. When it is unset, a validly-signed token that omits the aud claim is accepted. (A token that carries an aud claim is still rejected when no expected audience is configured.) Both options default to unset and have been documented as optional (Importance LOW) since KIP-768 (3.1.0). Impact: The exposure is realizable mainly in deployments whose JWKS endpoint serves signing keys for more than one issuer (a multi-tenant identity provider, an aggregated/shared JWKS, or shared signing keys): a token minted for a different issuer, or a token lacking an audience, may be accepted to authenticate — potentially as a higher-privileged principal. In a single-issuer deployment where the provider controls the iss claim and issues audience-scoped tokens, exposure is limited. The signature trust anchor remains enforced in all cases. This is distinct from CVE-2026-33557 (acceptance of unsigned/unvalidated tokens); here the signature is validated and only the issuer/audience checks are not enforced by default. Mitigation (all affected versions, no upgrade required): Set both sasl.oauthbearer.expected.issuer (to your provider's exact iss value) and sasl.oauthbearer.expected.audience (to the audience(s) your brokers expect), per listener. Brokers that set these perform exact-match issuer/audience checks and are not affected. Both have been available since 3.1.0. Resolution: - From Apache Kafka 4.4.0: a broker configured with a JWKS endpoint fails to start unless both sasl.oauthbearer.expected.issuer and sasl.oauthbearer.expected.audience are set (secure by default). Deployments that intentionally do not restrict the issuer and/or audience may opt out per axis with sasl.oauthbearer.allow.unverified.issuer=true and/or sasl.oauthbearer.allow.unverified.audience=true.
-
CVE-2026-103404
An Incorrect Permission Assignment for Critical Resource vulnerability in RoutingManager in Juniper Networks Session Smart Conductor allows a local user to access sensitive routing credentials via triggering processes that write to the /tmp directory.
-
CVE-2026-103401
An Improper Link Resolution Before File Access ('Link Following') vulnerability in the unauthenticated web server handlers in Juniper Networks Session Smart Router and Session Smart Conductor allows a local unprivileged user to overwrite arbitrary files as root via unauthenticated API endpoints. This vulnerability can be exploited on Mist-managed routers and on unmanaged Routers that have not yet been onboarded to a Conductor. Routers already managed by Conductor are not affected.
-
CVE-2026-67187
Ninja Forms plugin for WordPress before 3.15.4 contains an incomplete server-side validation vulnerability that allows unauthenticated attackers to submit arbitrary values to Checkbox List, Multi-Select List, Image List, Country, and State fields because NF_Abstracts_List falls through to a validate() method that only checks required-field presence. Attackers can submit crafted formula payloads through the unauthenticated nf_ajax_submit endpoint that bypass CSV escaping applied only to Textbox and Textarea fields, resulting in CSV injection when an administrator downloads and opens the exported submissions file.
-
CVE-2026-118029
Unverified ownership of the default query result location in the Athena module in AWS SDK for pandas (awswrangler) before 3.16.1 might allow remote actors to read Athena query results and tamper with result sets returned to the caller via a pre-created S3 bucket. To remediate this issue, users should upgrade to version 3.16.1 or later. Users should also verify that the default Athena result bucket in each account and region is owned by their own account, and set an explicit result location or enforce a workgroup result configuration.
-
CVE-2026-21830
HCL BigFix AEX product is affected by a vulnerability in the aws-sdk package.. It may cause AWS API requests to be sent to unintended or non‐AWS endpoints.
-
CVE-2026-99216
In the Linux kernel, the following vulnerability has been resolved: Drivers: hv: Don't reset VMBus connection state on error/exit paths In two places, current code forces vmbus_connection.conn_state to DISCONNECTED prior to calling vmbus_disconnect(). But vmbus_disconnect() calls vmbus_initiate_unload(), which unloads the VMBus connection only if the conn_state is CONNECTED. Consequently, the connection remains and the Hyper-V host might continue to send messages or post VMBus interrupts. The problem was introduced in commit 74347a99e73ae ("x86/Hyper-V: Unload vmbus channel in hv panic callback"), in that vmbus_initiate_unload() started checking the current connection state before doing the unload. This commit removed some occurrences where conn_state was forced to DISCONNECTED, but these two remained, apparently due to an oversight at the time. Fix this by not forcing conn_state to DISCONNECTED during VMBus exit, or if a failure occurs when establishing the connection in the first place. Whatever value conn_state has at the time is then read by vmbus_initiate_unload() to determine if there is a connection to be unloaded.
-
CVE-2026-99207
In the Linux kernel, the following vulnerability has been resolved: scsi: target: Disable interrupts while holding delayed_cmd_lock target_do_delayed_work() takes dev->delayed_cmd_lock with a plain spin_lock(). It runs from system_percpu_wq, so interrupts stay enabled while the lock is held. Every other acquisition of that lock uses spin_lock_irqsave(): target_handle_task_attr() and transport_complete_ordered_sync() in target_core_transport.c, and target_non_ordered_release() in target_core_device.c. That was safe while every caller of transport_complete_task_attr() ran in process context: target_complete_ok_work() from target_completion_wq, transport_complete_qf() from dev->qf_work_queue, and transport_generic_request_failure() from the submit and execute paths. Since commit 06933066d88a ("scsi: target: Add support for completing commands from backend context") that is no longer true. With complete_type=1 on a fabric that sets direct_compl_supp, target_complete() runs target_complete_ok_work() inline in the backend's completion context, which target_complete_cmd_with_sense() documents as "May be called from interrupt context". For iblock that is the bio end_io. target_complete_ok_work() starts with transport_complete_task_attr(), which takes delayed_cmd_lock in two ways: - Every ORDERED command is put on dev->delayed_cmd_list by target_handle_task_attr() and dispatched from there by target_do_delayed_work(), which marks it SCF_TASK_ORDERED_SYNC. Its completion therefore goes through transport_complete_ordered_sync(). - A SIMPLE command that completes while an ORDERED command is pending drops the last reference to the killed dev->non_ordered, and percpu_ref_put() then runs target_non_ordered_release() in the completing context. Both take the lock and, if commands are waiting, call schedule_work(&dev->delayed_cmd_work), which queues on system_percpu_wq, i.e. on the CPU that took the completion. Further completions for the same queue are typically delivered there too, so one arriving while the work holds the lock spins on it forever. Reaching this needs complete_type=1, which is not the default (vhost-scsi, the only fabric with direct_compl_supp, defaults to TARGET_QUEUE_COMPL), a backend whose I/O completes from interrupt context, and ORDERED commands from the initiator. Linux virtio_scsi guests send only VIRTIO_SCSI_S_SIMPLE and never take the ordered path; any guest that uses the ORDERED task attribute, which vhost-scsi accepts, does. target_do_delayed_work() is a work item and always runs in process context, so spin_lock_irq() is sufficient. Found by inspection; no runtime report. A single ORDERED command goes through both acquisitions, so with CONFIG_PROVE_LOCKING the first one completed through direct completion should produce an inconsistent lock state report on delayed_cmd_lock (hardirq or softirq, depending on where the backend completes).
-
CVE-2026-99209
In the Linux kernel, the following vulnerability has been resolved: drm/xe: skip eviction of a BO with no resource xe_evict_flags() switches on tbo->resource->mem_type to choose an eviction placement. An xe BO can reach eviction with no ttm.resource, e.g. an imported dma-buf BO whose resource was freed during teardown while its attachment is still live: the exporter then invalidates it via xe_dma_buf_move_notify() -> xe_bo_evict(), and the switch dereferences a NULL resource and faults. A BO with no resource has nothing placed or mapped, so there is nothing to evict. Return an empty placement (num_placement = 0), mirroring the handling already used for non-xe scatter-gather BOs just above; ttm_bo_validate() then takes the pipeline-gutting path and returns 0. Without the fix the xe_dma_buf@xe_test_dmabuf_move_notify_no_resource dynamic live selftest will fail with below stack. BUG: kernel NULL pointer dereference, address: 0000000000000010 RIP: 0010:xe_evict_flags+0xf4/0x140 [xe] Call Trace: <TASK> xe_bo_evict+0x40/0xa0 [xe] ? ttm_bo_pipeline_gutting+0x71/0x180 [ttm] xe_test_dmabuf_move_notify_no_resource+0x182/0x440 [xe] xe_dma_buf_kunit+0x387/0x580 [xe] v2: Add Fixes tag and failure info in commit text (Matt Brost) (cherry picked from commit dbcfaac6331f1fcb0c2fe3ad3e9adb1e5d43ec0c)
-
CVE-2026-99210
In the Linux kernel, the following vulnerability has been resolved: accel/rocket: clear rdev on device init failure rocket_probe() stores the return value from rocket_device_init() in the file-scoped rdev pointer before checking whether it is an error pointer. If rocket_device_init() fails, rdev is left as an ERR_PTR. The rdev pointer is also used as the marker for whether the shared rocket device has already been initialized. A later core probe can therefore see a non-NULL rdev, skip rocket_device_init(), and dereference the error pointer when reading rdev->num_cores. Save the error code and clear rdev before returning from the rocket_device_init() failure path.
-
CVE-2026-99211
In the Linux kernel, the following vulnerability has been resolved: netlink: reset skb headers before unicast and multicast filtering Socket filters can use SKF_NET_OFF and SKF_LL_OFF to read relative to an skb's network and MAC headers. Netlink producers may reserve headroom without resetting those headers, allowing a receiving socket's filter to read data before the initialized message. For example, IWPM HELLO replies are allocated with dev_alloc_skb(), which reserves NET_SKB_PAD bytes. An unprivileged NETLINK_RDMA user can attach a classic socket filter and disclose this headroom through the length of the delivered reply. The resets in netlink_dump() do not help because IWPM sends a separately allocated skb through netlink_unicast(). Reset both headers before running the receiver's filter in unicast and broadcast delivery, as is already done for dump skbs. This also covers other netlink producers that reserve headroom without changing their allocation or message-building code.
-
CVE-2026-99212
In the Linux kernel, the following vulnerability has been resolved: net/rds: don't let the rdma_cm destroy an id RDS still owns on route failure rds_rdma_cm_event_handler_cmn() hands the return value of rdma_resolve_route() straight back to the rdma_cm on RDMA_CM_EVENT_ADDR_RESOLVED. A synchronous failure there - -ENOMEM from the route work allocation, an SA query that cannot be set up, a RoCE route with no usable device - makes addr_handler() destroy the id the event was delivered on. That id is ic->i_cm_id, and nothing clears the pointer: the connection sits in RDS_CONN_CONNECTING with a freed id until its shutdown calls rdma_disconnect() and rdma_destroy_id() on it. The ROUTE_RESOLVED case already avoids this: rds_ib_cm_initiate_connect() forces a zero return while ic->i_cm_id == cm_id, as the comment there explains. Do the same here - drop the connection and return 0 - so the id stays RDS's to destroy from the shutdown, and the reconnect gets a fresh one.
-
CVE-2026-99214
In the Linux kernel, the following vulnerability has been resolved: drm/vc4: hdmi: Move packet RAM enable to PM runtime Audio can be triggered by the HPD handler and start up before video has been restored, calling vc4_hdmi_write_infoframe() for the audio infoframe. vc4_hdmi_write_infoframe() needs Packet RAM to be on to store the infoframe, however, if the video hasn't been enabled, vc4_hdmi_encoder_post_crtc_enable() hasn't been called and therefore, the Packet RAM can be disabled by the time vc4_hdmi_write_infoframe() executes, causing the following splat: [ 201.570266] ------------[ cut here ]------------ [ 201.570293] Packet RAM has to be on to store the packet. [ 201.570367] WARNING: drivers/gpu/drm/vc4/vc4_hdmi.c:683 at vc4_hdmi_write_infoframe+0x404/0x494 [vc4], CPU#0: wireplumber/1606 [...] [ 201.658650] Hardware name: Raspberry Pi 4 Model B Rev 1.5 (DT) [...] [ 201.758499] Call trace: [ 201.760967] vc4_hdmi_write_infoframe+0x404/0x494 [vc4] (P) [ 201.766607] vc4_hdmi_write_audio_infoframe+0x1c/0x28 [vc4] [ 201.772247] write_infoframe+0xbc/0x140 [drm_display_helper] [ 201.777975] drm_atomic_helper_connector_hdmi_update_audio_infoframe+0x80/0xdc [drm_display_helper] [ 201.787140] vc4_hdmi_audio_prepare+0x4ac/0x570 [vc4] [ 201.792252] drm_connector_hdmi_audio_prepare+0x24/0x30 [drm_display_helper] [ 201.799390] hdmi_codec_prepare+0x100/0x160 [snd_soc_hdmi_codec] [ 201.805471] snd_soc_pcm_dai_prepare+0xd4/0x100 [snd_soc_core] [ 201.811375] __soc_pcm_prepare+0x60/0x12c [snd_soc_core] [ 201.816752] soc_pcm_prepare+0x34/0x58 [snd_soc_core] [ 201.821863] snd_pcm_do_prepare+0x70/0x8c [snd_pcm] [ 201.826798] snd_pcm_action_nonatomic+0xe4/0x174 [snd_pcm] [ 201.832350] snd_pcm_prepare+0xc8/0xd8 [snd_pcm] [ 201.837021] snd_pcm_common_ioctl+0x458/0x614 [snd_pcm] [ 201.842308] snd_pcm_ioctl+0x30/0x48 [snd_pcm] [ 201.846802] __arm64_sys_ioctl+0x88/0xd8 [ 201.850769] invoke_syscall+0x48/0xf4 [ 201.854470] el0_svc_common+0x80/0xd8 [ 201.858171] do_el0_svc+0x1c/0x28 [ 201.861520] el0_svc+0x3c/0xb0 [ 201.864605] el0t_64_sync_handler+0x68/0xdc [ 201.868834] el0t_64_sync+0x198/0x19c [ 201.872536] ---[ end trace 0000000000000000 ]--- Considering that Packet RAM enable is essentially a power management configuration, move its enabling/disabling to the Runtime PM hooks. Note that post_crtc_disable() still stops all infoframes, but keeps Packet RAM enabled, leaving it to be disabled by runtime suspend.
-
CVE-2026-99201
In the Linux kernel, the following vulnerability has been resolved: netlink: avoid hashing the network namespace pointer The netlink rhashtable key includes a raw struct net pointer and a user-controlled port ID. Both /proc/net/netlink and socket diagnostics expose the table's bucket order. By binding and rebinding chosen NETLINK_USERSOCK port IDs, an unprivileged reader can distinguish equal buckets and recover the low bits of the Jenkins hash. Its 32-bit seed and the limited set of kernel-image slides can then be searched offline to recover the address of init_net. Use the namespace's unique, non-address net_cookie in the comparison key instead. It is assigned before the per-net initializers run and remains unchanged for the namespace's lifetime. The lookup key and object hash are still built by netlink_compare_arg_init(), keeping lookup, insertion, removal and rehashing consistent while preserving namespace separation. Neither public table walker needs to change. Reading the socket's namespace cookie in netlink_compare() is safe under RCU, including during namespace teardown. netlink_release() removes the socket from the hash table and defers its final put with call_rcu(). cleanup_net() runs the per-net exit methods and waits for outstanding RCU callbacks with rcu_barrier() before freeing namespace storage. Unlike ns.ns_id, net_cookie is also available and initialized during setup_net() in older stable kernels, making the change straightforward to backport to v5.15 and later.
-
CVE-2026-99202
In the Linux kernel, the following vulnerability has been resolved: scsi: libsas: Fix SMP IO deadlock during HA resume smp_execute_task_sg() calls pm_runtime_get_sync() on the host before issuing an SMP command. When that command is itself issued from the HA resume path, the get_sync() deadlocks: it waits for the ongoing resume (the device is RPM_RESUMING), while the resume is blocked in sas_drain_work() waiting for that same SMP IO to complete. The deadlock needs an expander-attached SATA disk. During sas_resume_ha() -> sas_drain_work(), DISCE_RESUME -> sas_resume_sata() -> ata_sas_port_resume() requests ATA_EH_RESET, and the hard reset for such a disk is done via SMP PHY CONTROL (sas_ata_hard_reset() -> sas_phy_reset() -> sas_smp_phy_control() -> smp_execute_task_sg()). Direct-attached SATA resets through lldd_control_phy() and SSP devices use TMFs, so neither hits this. Replace the get_sync()/put_sync() pair with pm_runtime_get_noresume()/pm_runtime_put(). smp_execute_task_sg() only needs to hold off autosuspend while the SMP is in flight, and it must not try to resume the host: a sync resume issued from the HA resume path itself is what deadlocks, and by the time sas_resume_ha() runs, hw_init has already reinitialized the hardware, so the device is accessible without one. The usage reference is still required. Discovery work normally runs inside an event worker's PM reference, taken at sas_notify_port_event() notify time and held until the handler has flushed the disco queue. sas_rediscover_ex_phy() however requeues DISCE_REVALIDATE_DOMAIN from within the revalidation worker itself, and flush_workqueue() does not wait for work items queued during execution, so that chained revalidation runs with no outer PM reference - without the get_noresume(), its SMP could race autosuspend. For the BSG path, sas_smp_handler() is the only caller which may find the host autosuspended: expander SMP requests do not go through any SCSI device request queue, so nothing else in that path holds the host awake. Resume it there with pm_runtime_resume_and_get() and check the result.
-
CVE-2026-99205
In the Linux kernel, the following vulnerability has been resolved: sched/core: Call wakeup_preempt() for queued tasks during class change Paul reported recent RCU changes exposed a lost resched where CPU idles with a rcutorture thread waiting while being completely runnable. Dumping the rq state at the time of idling with runnable tasks showed the following: rq->next_class = &fair_sched_class rq->nr_running = 1 rq->cfs.h_nr_queued = 0 rq->rt.rt_nr_running = 1 Although the waiting task was an RT task, "rq->next_class" did not point to rt_sched_class and was stuck at the lower priority fair_sched_class. Inspecting sched_change_end() revealed the following race with __sched_setscheduler() for queued, but non-running tasks: CPU0 CPU1 ==== ==== __schedule() /* * rq->donor is a fair task that blocks. * rq->next_class = &fair_sched_class */ rq_lock(rq0) pick_next_task() /* nr_running = 0 */ pick_task_fair() sched_balance_newidle() raw_spin_rq_unlock(rq0) rq_lock(rq0) activate_task(rq0, p) /* p is a fair task */ rq0->nr_running++; rq0->cfs.h_nr_queued++; wakeup_preempt(rq, p) rq->next_class = &fair_sched_class resched_curr() rq_unlock(rq0) __sched_setscheduler(p) /* p is moved to SCHED_RR */ task_rq_lock(p) sched_change_begin() rq0->cfs.h_nr_queued--; rq0->rt.rt_nr_running++; sched_change_end() /* * p->on_rq = 1 (ctx->queued) * but p is not rq->donor * (!ctx->running) hence * resched_curr() is and * rq->next_class upgrade is * skipped. */ task_rq_unlock() raw_spin_rq_lock(rq0) /* * rq->nr_running = 1 * rq->cfs.h_nr_queued = 0 * rq->rt.rt_nr_running = 1 * rq->next_class = &fair_sched_class */ if (rq->cfs.h_nr_queued && ...) /* Above is false */ if (rq_modified_above(this_rq, &fair_sched_class)) /* Also false since rq->next_class is still fair */ /* * sched_balance_newidle() is considered to * have failed to pull any fair tasks and no * other wakeups were detected. */ return NULL; /* pick_task_fair() */ pick_task_idle() return rq->idle; next = rq->idle; /* Preemption indicators are lost */ clear_tsk_need_resched(prev); clear_preempt_need_resched(); rq->curr = rq->idle; !!! rq->idle switches in with rq->running == 1 but !need_resched() !!! In simple words, enqueue, followed by a class change of the newly queued task goes unnoticed in rq->next_class indicator and sched_balance_newidle() sees no fair task, nor any higher priority task enqueued. The above sequence will lead to CPU idling with a runnable task on the CPU. Call wakeup_preempt() for ctx->queued cases in sched_change_end() to re-evaluate rq->next_class after class change (ENQUEUE_CLASS) and set the NEED_RESCHED indicator accordingly. Newidle balance will correctly see an enqueue of a higher priority t ---truncated---
-
CVE-2026-99206
In the Linux kernel, the following vulnerability has been resolved: scsi: target: iscsi: Fix race in discovery auth updates Separate configfs opens can concurrently update the discovery AuthMethod parameter. Both callers may free the same value in iscsi_update_param_value(), causing a double-free. Protect the lookup and replacement with the discovery TPG access lock, as used by the other TPG configfs parameter updates.
-
CVE-2026-99191
In the Linux kernel, the following vulnerability has been resolved: calipso: keep BIG TCP payload_len in calipso_skbuff_setattr() calipso_skbuff_setattr() adjusts payload_len by the length of the option it inserts or resizes: payload = ntohs(ip6_hdr->payload_len); ip6_hdr->payload_len = htons(payload + len_delta); GRO can build BIG TCP packets larger than IPV6_MAXPLEN, and stores payload_len == 0 for them; ipv6_payload_len() then falls back to skb->len. For such a packet forwarded to a destination mapped to CALIPSO, the code above writes len_delta (or 65536 + len_delta when shrinking) into payload_len, and ipv6_payload_len() no longer falls back to skb->len. nf_tables, conntrack and sch_cake rely on it. The same arithmetic also wraps when a packet grows past IPV6_MAXPLEN, and a BIG TCP packet that shrinks below IPV6_MAXPLEN should get its real length back. Set payload_len from the skb length with ipv6_set_payload_len(), which stores 0 above IPV6_MAXPLEN and the real length otherwise, as ipv6_gro_complete() does. The network header was just reset to skb->data, so skb->len - sizeof(*ip6_hdr) is the payload length. Tested on one kernel (arm64, QEMU) with SELinux permissive, a minimal mdp-generated policy (network_peer_controls=1), "netlabelctl calipso add pass doi:16" and a NetLabel mapping of fd00:2::/64 to CALIPSO DOI 16. calipso_skbuff_setattr() was called through netlbl_skbuff_setattr() from a test-only debugfs hook (not part of this patch) adding a level-only label; payload_len after the call: case before fix after fix real length 1000, unlabeled 1016 1016 1016 70000 GSO, payload_len 0 16 0 70016 65530 GSO, grows past MAXPLEN 10 0 65546 65540 GSO with a 24 byte hop header, shrinks below MAXPLEN 65528 65532 65532 1000 with a 24 byte hop header 992 992 992 (0 means ipv6_payload_len() returns the real length.) On the forward path, with a client injecting unlabeled UDP through a router to a server, the router adds the CALIPSO option and the server delivers all datagrams (Udp6InDatagrams 5, Ip6InTruncatedPkts 0) with this patch.
-
CVE-2026-99192
In the Linux kernel, the following vulnerability has been resolved: net: bcmgenet: unshare the skb before writing the control block bcmgenet stores first_cb, last_cb and bytes_sent in the skb control block and reads them back on completion to free the skb and credit BQL. ether_setup() leaves IFF_TX_SKB_SHARING set, so the stack can resubmit a shared skb that is still in the ring. The next transmit overwrites last_cb, so the earlier slots complete without crediting their bytes and BQL stalls the queue until the watchdog resets it. Unshare the skb before writing the control block. Reproduce with pktgen in clone_skb mode, which hands the driver the same skb while earlier copies are still in the ring: modprobe pktgen pg=/proc/net/pktgen echo "add_device eth0" > $pg/kpktgend_0 echo "dst_mac AA:BB:CC:DD:EE:FF" > $pg/eth0 echo "dst 192.0.2.1" > $pg/eth0 echo "clone_skb 8" > $pg/eth0 echo "count 0" > $pg/eth0 echo start > $pg/pgctrl & sleep 5 echo stop > $pg/pgctrl wait cat /sys/class/net/eth0/queues/tx-0/byte_queue_limits/inflight inflight stays nonzero on the now idle queue. clone_skb 0 leaves it at zero.
-
CVE-2026-99183
In the Linux kernel, the following vulnerability has been resolved: accel/rocket: search every core slot when looking up a scheduler sched_to_core() walks rdev->cores[] up to rdev->num_cores, and rocket_remove() decrements num_cores for every core it removes. Unbind a core that is not the last one and the cores behind it fall outside the search, so sched_to_core() returns NULL for a core that is still bound and still running jobs. Neither caller checks the result: rocket_job_run(): rocket_fence_create(core), core->dev rocket_job_timedout(): dev_err(core->dev, "NPU job timed out") Unbinding the middle core of the three on an RK3588 while three clients are submitting to all of them faults twice, once from the surviving core's job queue and once from its reset work: KASAN: null-ptr-deref in range [0x0000000000000220-0x0000000000000227] Workqueue: fdad0000.npu drm_sched_run_job_work [gpu_sched] pc : rocket_job_run+0x234/0x838 [rocket] Call trace: rocket_job_run+0x234/0x838 [rocket] drm_sched_run_job_work+0x2cc/0xad8 [gpu_sched] process_one_work+0x640/0x14f0 KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007] Workqueue: rocket-reset-2 drm_sched_job_timedout [gpu_sched] pc : rocket_job_timedout+0xf0/0x1e0 [rocket] Call trace: rocket_job_timedout+0xf0/0x1e0 [rocket] drm_sched_job_timedout+0x188/0x6a0 [gpu_sched] Both are the third core: the workqueue names are its device and its core->index, and it was left at slot 2 while num_cores had dropped to 2. Search all the slots that were allocated, the way find_core_for_dev() now does. A core that is still bound is then found, and the two callers get the pointer they already assume they have. This does not make unbinding one core out of several safe. An open client keeps an entity pointing at the scheduler of the core that went away: drm_sched reports it as not ready for every job that lands on it, and the client waits in dma_fence_default_wait for a fence that will never signal. Stopping the NULL dereference is what belongs in a fix; the rest wants more thought.
-
CVE-2026-99184
In the Linux kernel, the following vulnerability has been resolved: accel/rocket: search every core slot when a core is removed rocket_remove() decrements rdev->num_cores for each core it removes, while find_core_for_dev() searches slots 0 to num_cores - 1. Unbinding the cores in the order they were bound therefore loses the last one: by the time it is removed the search range has already shrunk past its slot, so find_core_for_dev() returns -1 and rocket_remove() gives up without doing anything. num_cores never reaches zero, rocket_device_fini() never runs, and the file-scoped rdev keeps pointing at a device that is going away. Binding the cores again starts from that stale count, because rocket_probe() takes rdev->num_cores as the slot to fill. On an RK3588, which describes three cores, the second round lands on slots 1, 2 and 3 while rdev->cores was allocated with room for three: rocket fdab0000.npu: Rockchip NPU core 1 version: 1179210309 rocket fdac0000.npu: Rockchip NPU core 2 version: 1179210309 rocket fdad0000.npu: Rockchip NPU core 3 version: 1179210309 The write to rdev->cores[3] is past the end of the array. Nothing in tree reads the core array often enough to notice, so the overrun is silent today. It turned up while testing a devfreq series on top of this, where a worker walks every core a few times a second, and UBSAN caught the first bool it read out of the overrun entry: UBSAN: invalid-load in drivers/accel/rocket/rocket_devfreq.c:47:10 load of value 5 is not a valid value for type '_Bool' Workqueue: devfreq_wq devfreq_monitor Record how many slots were allocated and search all of them. Every core is then found on removal, num_cores reaches zero, the device is torn down and a later bind starts from a clean rdev. This does not make unbinding a single core out of several work. probe still takes num_cores as the slot to fill, so rebinding one core while its siblings stay bound would write over a slot that is already in use, and rocket_open() still reaches for cores[0] whether or not anything is there. Both of those want more thought than a fix should carry. Found by unbinding and rebinding all three cores on an Orange Pi 5 Plus. With this applied, 25 unbind/rebind rounds and 5 module unload/reload rounds run clean there: the cores land in slots 0, 1 and 2 every time, whichever order they are bound in, and the shared supply goes back to a single user after each round.
-
CVE-2026-99185
In the Linux kernel, the following vulnerability has been resolved: firewire: core: fix race condition in split transaction handling For a split transaction, either fw_core_handle_response() or split_transaction_timeout_callback() may manipulate the same transaction entry in the pending transaction list. The return value of timer_delete() is used to determine which function handles the entry. If the transaction is a split transaction and timer_delete() returns zero, the timeout callback may already be running. In this case, the response handler must leave the entry in the list for the timeout callback to remove. A previous code refactoring broke this behavior, introducing a race condition in which both functions could process the same transaction. This could lead to a use-after-free on the transaction initiator side. Fix the regression by ensuring that the response handler does not process the transaction when timer_delete() returns zero. The timeout callback is responsible for removing the entry from the pending transaction list.
-
CVE-2026-99187
In the Linux kernel, the following vulnerability has been resolved: perf: Fix perf_pmu_migrate_context() leaving events behind __perf_pmu_remove() walks the group leaders of a CPU and PMU with perf_event_groups_for_cpu_pmu(), removing each one as it goes. perf_remove_from_context() takes the leader out of the groups rbtree, after which rb_next() on that node returns NULL, prematurely ending the walk. Other groups stay behind in the source context. On CPU hotplug, they are then torn down with the dying CPU. They keep running until teardown, while the PMU's interrupt and migrated events are on a new CPU, which drivers rely on never happening. Walk ctx->event_list instead and pick out the PMU's events by their pmu_ctx.
-
CVE-2026-99189
In the Linux kernel, the following vulnerability has been resolved: gpio: mxc: fix race between chained IRQ handler install and probe completion mxc_update_irq_chained_handler() is called before the IRQ domain, the generic IRQ chip, and the port list entry are set up. If an interrupt arrives in that window: - mx3_gpio_irq_handler() calls generic_handle_domain_irq() with port->domain still NULL. - mx2_gpio_irq_handler() walks mxc_gpio_ports, but the port has not been added to the list yet. Additionally, if any of the subsequent probe steps (gpio_generic_chip_init(), devm_gpiochip_add_data(), irq_domain_create_legacy(), or mxc_gpio_init_gc()) fail, the error paths never unregister the chained handler, leaving a dangling handler that points at freed memory. Move the handler installation after all its dependencies are ready and after list_add_tail(), so the handler is never live while the data structures it touches are incomplete, and is never installed if probe fails.
-
CVE-2026-99174
In the Linux kernel, the following vulnerability has been resolved: ext4: free the sbi after a grace period ext4_get_link() follows a symlink that keeps its target in a data block in rcu pathwalk mode with ext4_getblk(EXT4_GET_BLOCKS_CACHED_NOWAIT). That reads s_mount_state in the ASSERT() of ext4_getblk() and again in ext4_es_lookup_extent(), and the lookup bumps the extent status hit or miss counter, all through sb->s_fs_info. ext4_put_super() destroys those counters, clears sb->s_fs_info and frees the sbi without waiting for a grace period. A pathwalk that is still in ext4_get_link() when the last reference to the mount goes away then dereferences a NULL s_fs_info, the freed sbi or the destroyed percpu counters. With a pathwalk held up in ext4_get_link() while the last reference goes away: KASAN: null-ptr-deref in range [0x00000000000000a8-0x00000000000000af] RIP: 0010:ext4_getblk+0xf0/0x640 Call Trace: <TASK> ext4_get_link+0x26c/0x410 pick_link+0xf96/0x1780 step_into_slowpath+0x9b8/0x1250 path_lookupat+0x484/0xb00 filename_lookup+0x1f1/0x5b0 vfs_statx+0xd7/0x3b0 vfs_fstatat+0x4d/0x90 __do_sys_newfstatat+0xa0/0x100 do_syscall_64+0x10e/0x5e0 entry_SYSCALL_64_after_hwframe+0x77/0x7f So far that has been hard to hit. umount(2) waits for a grace period in namespace_unlock() before it drops the mount, so only a mount whose last reference goes away elsewhere, like the close of a file or a chdir out of a lazily unmounted directory, can race a walker. Keep sb->s_fs_info and free the sbi together with the extent status counters after a grace period. Split the counters out of ext4_es_unregister_shrinker() for that. The shrinker still gets wasted in ext4_put_super(), the counters go with the sbi.
The newest published CVE records from CVE.org and NVD, refreshed every few minutes, with severity, exploit odds and fix status.