CVE-2026-93099

In the Linux kernel, the following vulnerability has been resolved:

fs/resctrl: Fix UAF from worker threads when domains are removed

The mbm_handle_overflow() and cqm_handle_limbo() workers read event counters and may sleep while doing so. They are scheduled via delayed_work embedded in struct rdt_l3_mon_domain. Architecture allocates and frees these domains from CPU hotplug callbacks under cpus_write_lock(), and the workers acquire cpus_read_lock() to keep the domain alive across their access.

A use-after-free can occur when a worker is blocked waiting for cpus_read_lock() while the hotplug core holds cpus_write_lock(): the architecture frees the rdt_l3_mon_domain that contains the worker's work_struct. When the worker unblocks, the container_of() it performs on the embedded work pointer dereferences freed memory.

Drop cpus_read_lock() from the workers and instead drain pending and in-flight work synchronously before the architecture can free the domain. Since architecture offlines the domain under cpus_write_lock() after it has been unlinked from the RCU list and a grace period has elapsed, no new work can be scheduled. The cancel only needs to wait out existing work. Drop rdtgroup_mutex during CPU offline around cancel_delayed_work_sync() so that a worker waiting on the mutex can complete before re-pinning the work on a different CPU.

When offlining a CPU the architecture may iterate over resources in any order. For example, the MBA control domain may be offlined before or after a corresponding L3 monitor domain. Ensure that resctrl fs cancels the workers no matter what order the architecture offlines the domains.

In the Linux kernel, the following vulnerability has been resolved:

fs/resctrl: Fix UAF from worker threads when domains are removed

The mbm_handle_overflow() and cqm_handle_limbo() workers read event counters and may sleep while doing so. They are scheduled via delayed_work embedded in struct rdt_l3_mon_domain. Architecture allocates and frees these domains from CPU hotplug callbacks under cpus_write_lock(), and the workers acquire cpus_read_lock() to keep the domain alive across their access.

A use-after-free can occur when a worker is blocked waiting for cpus_read_lock() while the hotplug core holds cpus_write_lock(): the architecture frees the rdt_l3_mon_domain that contains the worker's work_struct. When the worker unblocks, the container_of() it performs on the embedded work pointer dereferences freed memory.

Drop cpus_read_lock() from the workers and instead drain pending and in-flight work synchronously before the architecture can free the domain. Since architecture offlines the domain under cpus_write_lock() after it has been unlinked from the RCU list and a grace period has elapsed, no new work can be scheduled. The cancel only needs to wait out existing work. Drop rdtgroup_mutex during CPU offline around cancel_delayed_work_sync() so that a worker waiting on the mutex can complete before re-pinning the work on a different CPU.

When offlining a CPU the architecture may iterate over resources in any order. For example, the MBA control domain may be offlined before or after a corresponding L3 monitor domain. Ensure that resctrl fs cancels the workers no matter what order the architecture offlines the domains.

In the Linux kernel, the following vulnerability has been resolved:

fs/resctrl: Fix UAF from worker threads when domains are removed

The mbm_handle_overflow() and cqm_handle_limbo() workers read event counters and may sleep while doing so. They are scheduled via delayed_work embedded in struct rdt_l3_mon_domain. Architecture allocates and frees these domains from CPU hotplug callbacks under cpus_write_lock(), and the workers acquire cpus_read_lock() to keep the domain alive across their access.

A use-after-free can occur when a worker is blocked waiting for cpus_read_lock() while the hotplug core holds cpus_write_lock(): the architecture frees the rdt_l3_mon_domain that contains the worker's work_struct. When the worker unblocks, the container_of() it performs on the embedded work pointer dereferences freed memory.

Drop cpus_read_lock() from the workers and instead drain pending and in-flight work synchronously before the architecture can free the domain. Since architecture offlines the domain under cpus_write_lock() after it has been unlinked from the RCU list and a grace period has elapsed, no new work can be scheduled. The cancel only needs to wait out existing work. Drop rdtgroup_mutex during CPU offline around cancel_delayed_work_sync() so that a worker waiting on the mutex can complete before re-pinning the work on a different CPU.

When offlining a CPU the architecture may iterate over resources in any order. For example, the MBA control domain may be offlined before or after a corresponding L3 monitor domain. Ensure that resctrl fs cancels the workers no matter what order the architecture offlines the domains.

In the Linux kernel, the following vulnerability has been resolved:

fs/resctrl: Fix UAF from worker threads when domains are removed

The mbm_handle_overflow() and cqm_handle_limbo() workers read event counters and may sleep while doing so. They are scheduled via delayed_work embedded in struct rdt_l3_mon_domain. Architecture allocates and frees these domains from CPU hotplug callbacks under cpus_write_lock(), and the workers acquire cpus_read_lock() to keep the domain alive across their access.

A use-after-free can occur when a worker is blocked waiting for cpus_read_lock() while the hotplug core holds cpus_write_lock(): the architecture frees the rdt_l3_mon_domain that contains the worker's work_struct. When the worker unblocks, the container_of() it performs on the embedded work pointer dereferences freed memory.

Drop cpus_read_lock() from the workers and instead drain pending and in-flight work synchronously before the architecture can free the domain. Since architecture offlines the domain under cpus_write_lock() after it has been unlinked from the RCU list and a grace period has elapsed, no new work can be scheduled. The cancel only needs to wait out existing work. Drop rdtgroup_mutex during CPU offline around cancel_delayed_work_sync() so that a worker waiting on the mutex can complete before re-pinning the work on a different CPU.

When offlining a CPU the architecture may iterate over resources in any order. For example, the MBA control domain may be offlined before or after a corresponding L3 monitor domain. Ensure that resctrl fs cancels the workers no matter what order the architecture offlines the domains.

In the Linux kernel, the following vulnerability has been resolved:

fs/resctrl: Fix UAF from worker threads when domains are removed

The mbm_handle_overflow() and cqm_handle_limbo() workers read event counters and may sleep while doing so. They are scheduled via delayed_work embedded in struct rdt_l3_mon_domain. Architecture allocates and frees these domains from CPU hotplug callbacks under cpus_write_lock(), and the workers acquire cpus_read_lock() to keep the domain alive across their access.

A use-after-free can occur when a worker is blocked waiting for cpus_read_lock() while the hotplug core holds cpus_write_lock(): the architecture frees the rdt_l3_mon_domain that contains the worker's work_struct. When the worker unblocks, the container_of() it performs on the embedded work pointer dereferences freed memory.

Drop cpus_read_lock() from the workers and instead drain pending and in-flight work synchronously before the architecture can free the domain. Since architecture offlines the domain under cpus_write_lock() after it has been unlinked from the RCU list and a grace period has elapsed, no new work can be scheduled. The cancel only needs to wait out existing work. Drop rdtgroup_mutex during CPU offline around cancel_delayed_work_sync() so that a worker waiting on the mutex can complete before re-pinning the work on a different CPU.

When offlining a CPU the architecture may iterate over resources in any order. For example, the MBA control domain may be offlined before or after a corresponding L3 monitor domain. Ensure that resctrl fs cancels the workers no matter what order the architecture offlines the domains.

• net: bridge: fix vlan tunnel dst refcnt when egressing {CVE-2021-47222}
• net: bridge: fix vlan tunnel dst null pointer dereference {CVE-2021-47223}
• writeback: fix race between cgroup_writeback_umount() and inode_switch_wbs() {CVE-2026-64378}
• writeback, cgroup: increment isw_nr_in_flight before grabbing an inode {CVE-2026-64378}
• writeback, cgroup: add smp_mb() to cgroup_writeback_umount() {CVE-2026-64378}
• fs/writeback.c: use rcu_barrier() to wait for inflight wb switches going into workqueue when umount {CVE-2026-64378}
• libceph: validate banner payload length {CVE-2026-90067}
• drm/virtio: release the GEM object on virtio_gpu_vram_create() errors {CVE-2026-98606}
• procfs: fix missing RCU protection when reading real_parent in do_task_stat() {CVE-2026-46259}
• NFSv4/flexfiles: reject zero filehandle version count {CVE-2026-53392}
• smb: client: cancel reconnect work in clean_demultiplex_info() {CVE-2026-98175}
• smb: client: reject userspace cifs.idmap descriptions {CVE-2026-97564}
• smb: client: fix UAF and buffer leak in cifs_check_trans2() for malformed secondary T2 {CVE-2026-89637}
• smb: client: fix OOB read/write from unvalidated DataOffset in coalesce_t2() {CVE-2026-89633}
• net: usb: cx82310_eth: drop URB after 0xffff reboot sentinel to prevent partial_data heap overflow {CVE-2026-98025}
• RDMA/siw: Fix endpoint/socket association handling {CVE-2026-74345}
• RDMA/irdma: Fix out-of-bounds write in irdma_copy_user_pgaddrs {CVE-2026-74390}
• nfsd: fix clock domain mismatch in clients_still_reclaiming() {CVE-2026-89685}
• vdpa_sim_blk: reject out-of-range sector starts {CVE-2026-97991}
• Drivers: hv: vmbus: use generic driver_override infrastructure {CVE-2026-80676}
• HID: wacom: Fix OOB write in wacom_hid_set_device_mode() {CVE-2026-63916}
• nvme-tcp: do not accept C2HData based on blk_rq_payload_bytes() alone {CVE-2026-89482}
• nvmet-tcp: fix page fragment cache leak in error path {CVE-2026-74386}
• bpf: mark a NULL call argument precise {CVE-2026-98019}
• bpf: Don't predict JMP32 pointer vs zero comparisons {CVE-2026-98041}
• ASoC: SOF: ipc3-control: Fix heap overflow in bytes_ext put/get {CVE-2026-72262}
• ASoC: SOF: ipc3-topology: Move and correct size checks in sof_ipc3_control_load_bytes() {CVE-2026-72262}
• Bluetooth: L2CAP: fix chan mode for LE_CONN_REQ + EXT_FLOWCTL pchan {CVE-2026-98108}
• mm/page_reporting: use system_freezable_wq to fix UAF during suspend {CVE-2026-74481}
• netfilter: synproxy: refresh tcphdr after skb_ensure_writable {CVE-2026-64007}
• netfilter: nf_conntrack_sip: widen NAT rewrite delta to s32 in sip_help_tcp() {CVE-2026-74569}
• reboot: fix cad_pid use-after-free race {CVE-2026-97938}
• fs/resctrl: Fix UAF from worker threads when domains are removed {CVE-2026-93099}
• ext4: fix out-of-bounds read in ext4_read_inline_dir() {CVE-2026-89786}
• ext4: fix dirtyclusters double decrement on fs shutdown {CVE-2026-45920}
• openvswitch: Fix CT limit teardown use-after-free {CVE-2026-89488}
• sctp: fix NULL deref on untransmitted RECONF completion {CVE-2026-89477}
• netfilter: conntrack: remove sprintf usage {CVE-2026-53002}
• fuse: re-lock request before replacing page cache folio {CVE-2026-53388}
• fanotify: fix false positive on permission events {CVE-2026-46150}
• configfs_lookup(): don't leave ->s_dentry dangling on failure {CVE-2026-74359}
• sctp: add INIT verification after cookie unpacking {CVE-2026-72398}
• sctp: fix err_chunk memory leaks in INIT handling {CVE-2026-72413}
• sctp: validate embedded INIT chunk and address list lengths in cookie {CVE-2026-53224}
• sctp: stream: fully roll back denied add-stream state {CVE-2026-52929}
• sctp: validate cached peer INIT chunk length in COOKIE_ECHO processing {CVE-2026-53246}
• sctp: diag: reject stale associations in dump_one path {CVE-2026-52917}
• sctp: Hold sock lock while iterating over address list {CVE-2026-52917}
• sctp: revalidate list cursor after sctp_sendmsg_to_asoc() in SCTP_SENDALL {CVE-2026-46227}
• nfsd: gate nfs3…