The Linux Kernel, the operating system core itself.

Security Fix(es):

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

wifi: rtw88: Use devm_kmemdup() in rtw_set_supported_band()

Simplify the code by using device managed memory allocations.

This also fixes a memory leak in rtw_register_hw(). The supported bands were not freed in the error path.

Copied from commit 145df52a8671 ("wifi: rtw89: Convert rtw89_core_set_supported_band to use devm_*").(CVE-2025-71273)

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

xfrm: hold dev ref until after transport_finish NF_HOOK

After async crypto completes, xfrm_input_resume() calls dev_put() immediately on re-entry before the skb reaches transport_finish. The skb->dev pointer is then used inside NF_HOOK and its okfn, which can race with device teardown.

Remove the dev_put from the async resumption entry and instead drop the reference after the NF_HOOK call in transport_finish, using a saved device pointer since NF_HOOK may consume the skb. This covers NF_DROP, NF_QUEUE and NF_STOLEN paths that skip the okfn.

For non-transport exits (decaps, gro, drop) and secondary async return points, release the reference inline when async is set.(CVE-2026-31663)

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

x86: shadow stacks: proper error handling for mmap lock

김영민 reports that shstk_pop_sigframe() doesn't check for errors from mmap_read_lock_killable(), which is a silly oversight, and also shows that we haven't marked those functions with "__must_check", which would have immediately caught it.

So let's fix both issues.(CVE-2026-43109)

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

srcu: Use irq_work to start GP in tiny SRCU

Tiny SRCU's srcu_gp_start_if_needed() directly calls schedule_work(), which acquires the workqueue pool->lock.

This causes a lockdep splat when call_srcu() is called with a scheduler lock held, due to:

call_srcu() [holding pi_lock] srcu_gp_start_if_needed() schedule_work() -> pool->lock

workqueue_init() / create_worker() [holding pool->lock] wake_up_process() -> try_to_wake_up() -> pi_lock

Also add irq_work_sync() to cleanup_srcu_struct() to prevent a use-after-free if a queued irq_work fires after cleanup begins.

Tested with rcutorture SRCU-T and no lockdep warnings.

Thanks to Boqun for similar fix in patch "rcu: Use an intermediate irq_work to start process_srcu()"

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

wifi: iwlwifi: fix 22000 series SMEM parsing

If the firmware were to report three LMACs (which doesn't exist in hardware) then using "fwrt->smem_cfg.lmac[2]" is an overrun of the array. Reject such and use IWL_FW_CHECK instead of WARN_ON in this function.(CVE-2026-43172)

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

bonding: fix type confusion in bond_setup_by_slave()

kernel BUG at net/core/skbuff.c:2306! Oops: invalid opcode: 0000 [#1] SMP KASAN NOPTI RIP: 0010:pskb_expand_head+0xa08/0xfe0 net/core/skbuff.c:2306 RSP: 0018:ffffc90004aff760 EFLAGS: 00010293 RAX: 0000000000000000 RBX: ffff88807e3c8780 RCX: ffffffff89593e0e RDX: ffff88807b7c4900 RSI: ffffffff89594747 RDI: ffff88807b7c4900 RBP: 0000000000000820 R08: 0000000000000005 R09: 0000000000000000 R10: 00000000961a63e0 R11: 0000000000000000 R12: ffff88807e3c8780 R13: 00000000961a6560 R14: dffffc0000000000 R15: 00000000961a63e0 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007fe1a0ed8df0 CR3: 000000002d816000 CR4: 00000000003526f0 Call Trace: <TASK> ipgre_header+0xdd/0x540 net/ipv4/ip_gre.c:900 dev_hard_header include/linux/netdevice.h:3439 [inline] packet_snd net/packet/af_packet.c:3028 [inline] packet_sendmsg+0x3ae5/0x53c0 net/packet/af_packet.c:3108 sock_sendmsg_nosec net/socket.c:727 [inline] __sock_sendmsg net/socket.c:742 [inline] ____sys_sendmsg+0xa54/0xc30 net/socket.c:2592 ___sys_sendmsg+0x190/0x1e0 net/socket.c:2646 __sys_sendmsg+0x170/0x220 net/socket.c:2678 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0x106/0xf80 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7fe1a0e6c1a9

When a non-Ethernet device (e.g. GRE tunnel) is enslaved to a bond, bond_setup_by_slave() directly copies the slave's header_ops to the bond device:

bond_dev->header_ops = slave_dev->header_ops;

This causes a type confusion when dev_hard_header() is later called on the bond device. Functions like ipgre_header(), ip6gre_header(),all use netdev_priv(dev) to access their device-specific private data. When called with the bond device, netdev_priv() returns the bond's private data (struct bonding) instead of the expected type (e.g. struct ip_tunnel), leading to garbage values being read and kernel crashes.

Fix this by introducing bond_header_ops with wrapper functions that delegate to the active slave's header_ops using the slave's own device. This ensures netdev_priv() in the slave's header functions always receives the correct device.

The fix is placed in the bonding driver rather than individual device drivers, as the root cause is bond blindly inheriting header_ops from the slave without considering that these callbacks expect a specific netdev_priv() layout.

The type confusion can be observed by adding a printk in ipgre_header() and running the following commands:

ip link add dummy0 type dummy
ip addr add 10.0.0.1/24 dev dummy0
ip link set dummy0 up
ip link add gre1 type gre local 10.0.0.1
ip link add bond1 type bond mode active-backup
ip link set gre1 master bond1
ip link set gre1 up
ip link set bond1 up
ip addr add fe80::1/64 dev bond1(CVE-2026-43456)

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

ata: libata-scsi: avoid Non-NCQ command starvation

When a non-NCQ command is issued while NCQ commands are being executed, ata_scsi_qc_issue() indicates to the SCSI layer that the command issuing should be deferred by returning SCSI_MLQUEUE_XXX_BUSY. This command deferring is correct and as mandated by the ACS specifications since NCQ and non-NCQ commands cannot be mixed.

However, in the case of a host adapter using multiple submission queues, when the target device is under a constant load of NCQ commands, there are no guarantees that requeueing the non-NCQ command will be executed later and it may be deferred again repeatedly as other submission queues can constantly issue NCQ commands from different CPUs ahead of the non-NCQ command. This can lead to very long delays for the execution of non-NCQ commands, and even complete starvation for these commands in the worst case scenario.

Since the block layer and the SCSI layer do not distinguish between queueable (NCQ) and non queueable (non-NCQ) commands, libata-scsi SAT implementation must ensure forward progress for non-NCQ commands in the presence of NCQ command traffic. This is similar to what SAS HBAs with a hardware/firmware based SAT implementation do.

Implement such forward progress guarantee by limiting requeueing of non-NCQ commands from ata_scsi_qc_issue(): when a non-NCQ command is received and NCQ commands are in-flight, do not force a requeue of the non-NCQ command by returning SCSI_MLQUEUE_XXX_BUSY and instead return 0 to indicate that the command was accepted but hold on to the qc using the new deferred_qc field of struct ata_port.

This deferred qc will be issued using the work item deferred_qc_work running the function ata_scsi_deferred_qc_work() once all in-flight commands complete, which is checked with the port qc_defer() callback return value indicating that no further delay is necessary. This check is done using the helper function ata_scsi_schedule_deferred_qc() which is called from ata_scsi_qc_complete(). This thus excludes this mechanism from all internal non-NCQ commands issued by ATA EH.

When a port deferred_qc is non NULL, that is, the port has a command waiting for the device queue to drain, the issuing of all incoming commands (both NCQ and non-NCQ) is deferred using the regular busy mechanism. This simplifies the code and also avoids potential denial of service problems if a user issues too many non-NCQ commands.

Finally, whenever ata EH is scheduled, regardless of the reason, a deferred qc is always requeued so that it can be retried once EH completes. This is done by calling the function ata_scsi_requeue_deferred_qc() from ata_eh_set_pending(). This avoids the need for any special processing for the deferred qc in case of NCQ error, link or device reset, or device timeout.(CVE-2026-45855)

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

ext4: drop extent cache when splitting extent fails

When the split extent fails, we might leave some extents still being processed and return an error directly, which will result in stale extent entries remaining in the extent status tree. So drop all of the remaining potentially stale extents if the splitting fails.(CVE-2026-45899)

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

selinux: fix overlayfs mmap() and mprotect() access checks

The existing SELinux security model for overlayfs is to allow access if the current task is able to access the top level file (the "user" file) and the mounter's credentials are sufficient to access the lower level file (the "backing" file). Unfortunately, the current code does not properly enforce these access controls for both mmap() and mprotect() operations on overlayfs filesystems.

This patch makes use of the newly created security_mmap_backing_file() LSM hook to provide the missing backing file enforcement for mmap() operations, and leverages the backing file API and new LSM blob to provide the necessary information to properly enforce the mprotect() access controls.(CVE-2026-46054)

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

mptcp: pm: ADD_ADDR rtx: always decrease sk refcount

When an ADD_ADDR is retransmitted, the sk is held in sk_reset_timer(). It should then be released in all cases at the end.

Some (unlikely) checks were returning directly instead of calling sock_put() to decrease the refcount. Jump to a new 'exit' label to call __sock_put() (which will become sock_put() in the next commit) to fix this potential leak.

While at it, drop the '!msk' check which cannot happen because it is never reset, and explicitly mark the remaining one as "unlikely".(CVE-2026-46158)

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

selinux: allow multiple opens of /sys/fs/selinux/policy

Currently there can only be a single open of /sys/fs/selinux/policy at any time. This allows any process to block any other process from reading the kernel policy. The original motivation seems to have been a mix of preventing an inconsistent view of the policy size and preventing userspace from allocating kernel memory without bound, but this is arguably equally bad. Eliminate the policy_opened flag and shrink the critical section that the policy mutex is held. While we are making changes here, drop a couple of extraneous BUG_ONs.(CVE-2026-46302)

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

ip6_vti: set netns_immutable on the fallback device.

john1988 and Noam Rathaus reported that vti6_init_net() does not set the netns_immutable flag on the per-netns fallback tunnel device (ip6_vti0).

Other similar tunnel drivers (like ip6_tunnel, sit, ip6_gre, and ip_tunnel) correctly set this flag during their fallback device initialization to prevent them from being moved to another network namespace.(CVE-2026-52909)

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

net: skbuff: fix missing zerocopy reference in pskb_carve helpers

pskb_carve_inside_header() and pskb_carve_inside_nonlinear() both copy the old skb_shared_info header into a new buffer via memcpy(), which includes the destructor_arg pointer (uarg) for MSG_ZEROCOPY skbs. Neither function calls net_zcopy_get() for the new shinfo, creating an unaccounted holder: every skb_shared_info with destructor_arg set will call skb_zcopy_clear() once when freed, but the corresponding net_zcopy_get() was never called for the new copy. Repeated calls drive uarg->refcnt to zero prematurely, freeing ubuf_info_msgzc while TX skbs still hold live destructor_arg pointers.

KASAN reports use-after-free on a freed ubuf_info_msgzc:

BUG: KASAN: slab-use-after-free in skb_release_data+0x77b/0x810 Read of size 8 at addr ffff88801574d3e8 by task poc/220

Call Trace: skb_release_data+0x77b/0x810 kfree_skb_list_reason+0x13e/0x610 skb_release_data+0x4cd/0x810 sk_skb_reason_drop+0xf3/0x340 skb_queue_purge_reason+0x282/0x440 rds_tcp_inc_free+0x1e/0x30 rds_recvmsg+0x354/0x1780 __sys_recvmsg+0xdf/0x180

Allocated by task 219: msg_zerocopy_realloc+0x157/0x7b0 tcp_sendmsg_locked+0x2892/0x3ba0

Freed by task 219: ip_recv_error+0x74a/0xb10 tcp_recvmsg+0x475/0x530

The skb consuming the late access still referenced the same uarg via shinfo->destructor_arg copied by pskb_carve_inside_nonlinear() without a refcount bump. This has been verified to be reliably exploitable: a working proof-of-concept achieves full root privilege escalation from an unprivileged local user on a default kernel configuration.

The fix follows the pattern of pskb_expand_head() which has the same memcpy/cloned structure. For pskb_carve_inside_header(), net_zcopy_get() is placed after skb_orphan_frags() succeeds, so the orphan error path needs no cleanup. For pskb_carve_inside_nonlinear(), net_zcopy_get() is placed after all failure points and just before skb_release_data(), so no error path needs cleanup at all -- matching pskb_expand_head() more closely and avoiding the need for a balancing net_zcopy_put().(CVE-2026-52943)

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

fs/fcntl: fix SOFTIRQ-unsafe lock order in fasync signaling

A SOFTIRQ-safe to SOFTIRQ-unsafe lock order deadlock can occur in send_sigio() and send_sigurg() when a process group receives a signal.

When FASYNC is configured for a process group (PIDTYPE_PGID), both functions use read_lock(&tasklist_lock) to traverse the task list. However, they are frequently called from softirq context:

• send_sigio() via input_inject_event -> kill_fasync
• send_sigurg() via tcp_check_urg -> sk_send_sigurg (NET_RX_SOFTIRQ)

The deadlock is caused by the rwlock writer fairness mechanism:

1. CPU 0 (process context) holds read_lock(&tasklist_lock) in do_wait().
2. CPU 1 (process context) attempts write_lock(&tasklist_lock) in fork() or exit() and spins, which blocks all new readers.
3. CPU 0 is interrupted by a softirq (e.g., TCP URG packet reception).
4. The softirq calls send_sigurg() and attempts to acquire read_lock(&tasklist_lock), deadlocking because CPU 1 is waiting.

Since PID hashing and do_each_pid_task() traversals are already RCU-protected, the read_lock on tasklist_lock is no longer strictly required for safe traversal. Fix this by replacing tasklist_lock with rcu_read_lock(), aligning the process group signaling path with the single-PID path. This also mitigates a potential remote denial of service vector via TCP URG packets.

Lockdep splat:

WARNING: SOFTIRQ-safe -> SOFTIRQ-unsafe lock order detected [...] Chain exists of: &dev->event_lock --> &f_owner->lock --> tasklist_lock

Possible interrupt unsafe locking scenario: CPU0 CPU1 ---- ---- lock(tasklist_lock); local_irq_disable(); lock(&dev->event_lock); lock(&f_owner->lock); <Interrupt> lock(&dev->event_lock);

*** DEADLOCK ***(CVE-2026-52946)

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

smb/client: fix possible infinite loop and oob read in symlink_data()

On 32-bit architectures, the infinite loop is as follows:

len = p->ErrorDataLength == 0xfffffff8 u8 *next = p->ErrorContextData + len next == p

On 32-bit architectures, the out-of-bounds read is as follows:

len = p->ErrorDataLength == 0xfffffff0 u8 *next = p->ErrorContextData + len next == (u8 *)p - 8(CVE-2026-52967)

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

netfilter: nf_tables: join hook list via splice_list_rcu() in commit phase

Publish new hooks in the list into the basechain/flowtable using splice_list_rcu() to ensure netlink dump list traversal via rcu is safe while concurrent ruleset update is going on.(CVE-2026-52988)

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

sched/psi: fix race between file release and pressure write

A potential race condition exists between pressure write and cgroup file release regarding the priv member of struct kernfs_open_file, which triggers the uaf reported in 1.

Consider the following scenario involving execution on two separate CPUs:

CPU0 CPU1 ==== ==== vfs_rmdir() kernfs_iop_rmdir() cgroup_rmdir() cgroup_kn_lock_live() cgroup_destroy_locked() cgroup_addrm_files() cgroup_rm_file() kernfs_remove_by_name() kernfs_remove_by_name_ns() vfs_write() __kernfs_remove() new_sync_write() kernfs_drain() kernfs_fop_write_iter() kernfs_drain_open_files() cgroup_file_write() kernfs_release_file() pressure_write() cgroup_file_release() ctx = of->priv; kfree(ctx); of->priv = NULL; cgroup_kn_unlock() cgroup_kn_lock_live() cgroup_get(cgrp) cgroup_kn_unlock() if (ctx->psi.trigger) // here, trigger uaf for ctx, that is of->priv

The cgroup_rmdir() is protected by the cgroup_mutex, it also safeguards the memory deallocation of of->priv performed within cgroup_file_release(). However, the operations involving of->priv executed within pressure_write() are not entirely covered by the protection of cgroup_mutex. Consequently, if the code in pressure_write(), specifically the section handling the ctx variable executes after cgroup_file_release() has completed, a uaf vulnerability involving of->priv is triggered.

Therefore, the issue can be resolved by extending the scope of the cgroup_mutex lock within pressure_write() to encompass all code paths involving of->priv, thereby properly synchronizing the race condition occurring between cgroup_file_release() and pressure_write().

And, if an live kn lock can be successfully acquired while executing the pressure write operation, it indicates that the cgroup deletion process has not yet reached its final stage; consequently, the priv pointer within open_file cannot be NULL. Therefore, the operation to retrieve the ctx value must be moved to a point after the live kn lock has been successfully acquired.

In another situation, specifically after entering cgroup_kn_lock_live() but before acquiring cgroup_mutex, there exists a different class of race condition:

CPU0: write memory.pressure CPU1: write cgroup.pressure=0 =========================== =============================

kernfs_fop_write_iter() kernfs_get_active_of(of) pressure_write() cgroup_kn_lock_live(memory.pressure) cgroup_tryget(cgrp) kernfs_break_active_protection(kn) ... blocks on cgroup_mutex

                                 	  cgroup_pressure_write()
                                 	  cgroup_kn_lock_live(cgroup.pressure)
                                 	  cgroup_file_show(memory.pressure, false)
                                 	    kernfs_show(false)
                                 	      kernfs_drain_open_files()
                                 	        cgroup_file_release(of)
                                 	          kfree(ctx)
                                 	            of->priv = NULL
                                 	  cgroup_kn_unlock()

... acquires cgroup_mutex ctx = of->priv; // may now be NULL if (ctx->psi.trigger) // NULL dereference

Consequently, there is a possibility that of->priv is NULL, the pressure write needs to check for this.

Now that the scope of the cgroup_mutex has been expanded, the original explicit cgroup_get/put operations are no longer necessary, this is because acquiring/releasing the live kn lock inherently executes a cgroup get/put operation.

1 BUG: KASAN: slab-use-after-free in pressure_write+0xa4/0x210 kernel/cgroup/cgroup.c:4011 Call Trace: pressure_write+0xa4/0x210 kernel/cgroup/cgroup.c:4011 cgroup_file_write+0x36f/0x790 kernel/cgroup/cgroup.c:43 ---truncated---(CVE-2026-52991)

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

netfilter: nat: use kfree_rcu to release ops

Florian Westphal says:

"Historically this is not an issue, even for normal base hooks: the data path doesn't use the original nf_hook_ops that are used to register the callbacks.

However, in v5.14 I added the ability to dump the active netfilter hooks from userspace.

This code will peek back into the nf_hook_ops that are available at the tail of the pointer-array blob used by the datapath.

The nat hooks are special, because they are called indirectly from the central nat dispatcher hook. They are currently invisible to the nfnl hook dump subsystem though.

But once that changes the nat ops structures have to be deferred too."

Update nf_nat_register_fn() to deal with partial exposition of the hooks from error path which can be also an issue for nfnetlink_hook.(CVE-2026-53000)

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

pppoe: drop PFC frames

RFC 2516 Section 7 states that Protocol Field Compression (PFC) is NOT RECOMMENDED for PPPoE. In practice, pppd does not support negotiating PFC for PPPoE sessions, and the current PPPoE driver assumes an uncompressed (2-byte) protocol field. However, the generic PPP layer function ppp_input() is not aware of the negotiation result, and still accepts PFC frames.

If a peer with a broken implementation or an attacker sends a frame with a compressed (1-byte) protocol field, the subsequent PPP payload is shifted by one byte. This causes the network header to be 4-byte misaligned, which may trigger unaligned access exceptions on some architectures.

To reduce the attack surface, drop PPPoE PFC frames. Introduce ppp_skb_is_compressed_proto() helper function to be used in both ppp_generic.c and pppoe.c to avoid open-coding.(CVE-2026-53003)

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

Bluetooth: l2cap: Add missing chan lock in l2cap_ecred_reconf_rsp

l2cap_ecred_reconf_rsp() calls l2cap_chan_del() without holding l2cap_chan_lock(). Every other l2cap_chan_del() caller in the file acquires the lock first. A remote BLE device can send a crafted L2CAP ECRED reconfiguration response to corrupt the channel list while another thread is iterating it.

Add l2cap_chan_hold() and l2cap_chan_lock() before l2cap_chan_del(), and l2cap_chan_unlock() and l2cap_chan_put() after, matching the pattern used in l2cap_ecred_conn_rsp() and l2cap_conn_del().(CVE-2026-53071)

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

Bluetooth: fix locking in hci_conn_request_evt() with HCI_PROTO_DEFER

When protocol sets HCI_PROTO_DEFER, hci_conn_request_evt() calls hci_connect_cfm(conn) without hdev->lock. Generally hci_connect_cfm() assumes it is held, and if conn is deleted concurrently -> UAF.

Only SCO and ISO set HCI_PROTO_DEFER and only for defer setup listen, and HCI_EV_CONN_REQUEST is not generated for ISO. In the non-deferred listening socket code paths, hci_connect_cfm(conn) is called with hdev->lock held.

Fix by holding the lock.(CVE-2026-53072)

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

bpf: Fix ld_{abs,ind} failure path analysis in subprogs

Usage of ld_{abs,ind} instructions got extended into subprogs some time ago via commit 09b28d76eac4 ("bpf: Add abnormal return checks."). These are only allowed in subprograms when the latter are BTF annotated and have scalar return types.

The code generator in bpf_gen_ld_abs() has an abnormal exit path (r0=0 + exit) from legacy cBPF times. While the enforcement is on scalar return types, the verifier must also simulate the path of abnormal exit if the packet data load via ld_{abs,ind} failed.

This is currently not the case. Fix it by having the verifier simulate both success and failure paths, and extend it in similar ways as we do for tail calls. The success path (r0=unknown, continue to next insn) is pushed onto stack for later validation and the r0=0 and return to the caller is done on the fall-through side.(CVE-2026-53090)

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

md: wake raid456 reshape waiters before suspend

During raid456 reshape, direct IO across the reshape position can sleep in raid5_make_request() waiting for reshape progress while still holding an active_io reference. If userspace then freezes reshape and writes md/suspend_lo or md/suspend_hi, mddev_suspend() kills active_io and waits for all in-flight IO to drain.

This can deadlock: the IO needs reshape progress to continue, but the reshape thread is already frozen, so the active_io reference is never dropped and suspend never completes.

raid5_prepare_suspend() already wakes wait_for_reshape for dm-raid. Do the same for normal md suspend when reshape is already interrupted, so waiting raid456 IO can abort, drop its reference, and let suspend finish.

The mdadm test tests/25raid456-reshape-deadlock reproduces the hang.(CVE-2026-53123)

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

netfilter: require Ethernet MAC header before using eth_hdr()

ip6t_eui64, xt_mac, the bitmap:ip,mac, hash:ip,mac, and hash:mac ipset types, and nf_log_syslog access eth_hdr(skb) after either assuming that the skb is associated with an Ethernet device or checking only that the ETH_HLEN bytes at skb_mac_header(skb) lie between skb->head and skb->data.

Make these paths first verify that the skb is associated with an Ethernet device, that the MAC header was set, and that it spans at least a full Ethernet header before accessing eth_hdr(skb).(CVE-2026-53131)

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

vsock/virtio: fix potential unbounded skb queue

virtio_transport_inc_rx_pkt() checks vvs->rx_bytes + len > vvs->buf_alloc.

virtio_transport_recv_enqueue() skips coalescing for packets with VIRTIO_VSOCK_SEQ_EOM.

If fed with packets with len == 0 and VIRTIO_VSOCK_SEQ_EOM, a very large number of packets can be queued because vvs->rx_bytes stays at 0.

Fix this by estimating the skb metadata size:

(Number of skbs in the queue) * SKB_TRUESIZE(0)(CVE-2026-53132)

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

netfilter: nft_fib: fix stale stack leak via the OIFNAME register

For NFT_FIB_RESULT_OIFNAME the destination register is declared with len = IFNAMSIZ (four 32-bit registers), but on the lookup-fail, RTN_LOCAL and oif-mismatch paths nft_fib{4,6}_eval() only writes one register via "*dest = 0". The remaining three registers are left as whatever was on the stack in nft_do_chain()'s struct nft_regs, and a downstream expression that loads the register span can leak that uninitialised kernel stack to userspace.

The NFTA_FIB_F_PRESENT existence check has the same shape: it is only meaningful for NFT_FIB_RESULT_OIF, yet it was accepted for any result type while the eval stores a single byte via nft_reg_store8(), leaving the rest of the declared span stale.

Fix both:

• replace the bare "*dest = 0" in the eval with nft_fib_store_result(), which strscpy_pad()s the whole IFNAMSIZ for OIFNAME (and is already used on the other early-return path), and

• restrict NFTA_FIB_F_PRESENT to NFT_FIB_RESULT_OIF and declare its destination as a single u8, so the marked span matches the one byte the eval writes.(CVE-2026-53134)

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

drm/amd/display: Clamp VBIOS HDMI retimer register count to array size

[Why & How] The VBIOS integrated info tables (v1_11 and v2_1) contain HdmiRegNum and Hdmi6GRegNum fields that are used as loop bounds when copying retimer I2C register settings into fixed-size arrays (dp*_ext_hdmi_reg_settings[9] and dp*_ext_hdmi_6g_reg_settings[3]). These u8 fields are not validated before use, so a malformed VBIOS can specify values up to 255, causing an out-of-bounds heap write during driver probe.

Clamp each register count to the destination array size using min_t() before the copy loops, in both get_integrated_info_v11() and get_integrated_info_v2_1().

(cherry picked from commit 5a7f0ef90195940c54b0f5bb85b87da55f038c69)(CVE-2026-53136)

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

drm/amd/display: Clamp HDMI HDCP2 rx_id_list read to buffer size

[Why & How] During HDCP 2.x repeater authentication over HDMI, the driver reads the sink's RxStatus register and extracts a 10-bit message size field (max value 1023). This value is used as the read length for the ReceiverID list without being clamped to the size of the destination buffer rx_id_list[177]. A malicious HDMI repeater could advertise a message size larger than the buffer, causing an out-of-bounds write during the I2C read.

Clamp the read length in mod_hdcp_read_rx_id_list() to the size of the rx_id_list buffer, matching the approach already used in the DP branch.

(cherry picked from commit 229212219e4247d9486f8ba41ef087358490be09)(CVE-2026-53137)

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

thunderbolt: Bound root directory content to block size

__tb_property_parse_dir() does not check that content_offset + content_len fits within block_len for the root directory case. When rootdir->length equals or exceeds block_len - 2, the entry loop reads past the allocated property block.

Add a bounds check after computing content_offset and content_len to reject directories whose content extends past the block.(CVE-2026-53149)

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

inet: frags: fix use-after-free caused by the fqdir_pre_exit() flush

On netns teardown, fqdir_pre_exit() walks the fqdir rhashtable and flushes every fragment queue that is not yet complete using inet_frag_queue_flush(). That helper frees all the skbs queued on the fragment queue but does not set INET_FRAG_COMPLETE, and leaves q->fragments_tail and q->last_run_head pointing at the freed skbs. The queue itself stays in the rhashtable.

fqdir_pre_exit() first lowers high_thresh to 0 to stop new queue lookups, but it cannot stop a fragment that already obtained the queue through inet_frag_find() earlier and stalled just before taking the queue lock. Once that fragment resumes after the flush and takes the queue lock, it passes the INET_FRAG_COMPLETE check and then dereferences the freed fragments_tail. inet_frag_queue_insert() reads FRAG_CB() and ->len of that pointer and, on the append path, writes ->next_frag, causing a slab use-after-free. IPv6, nf_conntrack_reasm6 and 6lowpan reassembly share the same flush path and are affected as well.

Reset rb_fragments, fragments_tail and last_run_head in inet_frag_queue_flush() so a flushed queue no longer points at the freed skbs. A fragment that resumes after the flush and takes the queue lock then finds an empty queue and starts a new run instead of dereferencing the freed fragments_tail. ip_frag_reinit() already performed this reset after its own flush, so drop the now duplicate code there.(CVE-2026-53175)

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

IB/isert: Reject login PDUs shorter than ISER_HEADERS_LEN

In drivers/infiniband/ulp/isert/ib_isert.c, isert_login_recv_done() computes the login request payload length as wc->byte_len minus ISER_HEADERS_LEN with no lower bound, and login_req_len is a signed int. A remote iSER initiator can post a login Send work request carrying fewer than ISER_HEADERS_LEN (76) bytes, so the subtraction underflows and login_req_len becomes negative.

isert_rx_login_req() then reads that negative length back into a signed int, takes size = min(rx_buflen, MAX_KEY_VALUE_PAIRS), and because the min() is signed it keeps the negative value; the value is then passed as the memcpy() length and sign-extended to a multi-gigabyte size_t. The copy into the 8192-byte login->req_buf runs far out of bounds and faults, crashing the target node. The login phase precedes iSCSI authentication, so no credentials are required to reach this path.

Reject any login PDU shorter than ISER_HEADERS_LEN before the subtraction, mirroring the existing early return on a failed work completion, so login_req_len can never go negative. The upper bound was already safe: a posted login buffer cannot deliver more than ISER_RX_PAYLOAD_SIZE, so the difference stays at or below MAX_KEY_VALUE_PAIRS and the existing min() clamps it; only the missing lower bound needs to be added.(CVE-2026-53176)

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

bnxt_en: Fix NULL pointer dereference

PCIe errors detected by a Root Port or Downstream Port cause error recovery services to run on all subordinate devices regardless of administrative state.

The .error_detected() callback, bnxt_io_error_detected(), disables and synchronizes IRQs via bnxt_disable_int_sync(), which calls bnxt_cp_num_to_irq_num() to map completion rings to IRQs using bp->bnapi.

Since bp->bnapi is allocated on NIC open and freed on NIC close, PCIe error recovery on a closed NIC can dereference a NULL pointer.

Check if bp->bnapi is NULL before disabling and synchronizing IRQs.(CVE-2026-53177)

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

udp: clear skb->dev before running a sockmap verdict

On the UDP receive path skb->dev is repurposed as dev_scratch (the truesize/state cache set by udp_set_dev_scratch()), through the union { struct net_device *dev; unsigned long dev_scratch; } in sk_buff.

When a UDP socket is in a sockmap, sk_data_ready is sk_psock_verdict_data_ready(), which calls udp_read_skb() -> recv_actor() (sk_psock_verdict_recv) to run the attached SK_SKB verdict program in softirq. If that program calls a socket-lookup helper (bpf_sk_lookup_tcp/udp, bpf_skc_lookup_tcp), bpf_skc_lookup() does:

if (skb->dev)
	caller_net = dev_net(skb->dev);

skb->dev still holds the dev_scratch value (a non-NULL integer), so dev_net() dereferences it as a struct net_device * and the kernel takes a general protection fault on a non-canonical address in softirq:

Oops: general protection fault, probably for non-canonical address 0x1010000800004a0 CPU: 1 UID: 0 PID: 1406 Comm: syz.2.19 Not tainted 7.1.0-rc6 #1 PREEMPT(full) RIP: 0010:bpf_skc_lookup net/core/filter.c:7033 [inline] RIP: 0010:bpf_sk_lookup+0x45/0x160 net/core/filter.c:7047 Call Trace: <IRQ> bpf_prog_4675cb904b7071f8+0x12e/0x14e bpf_prog_run_pin_on_cpu+0xc6/0x1f0 sk_psock_verdict_recv+0x1ba/0x350 udp_read_skb+0x31a/0x370 sk_psock_verdict_data_ready+0x2e3/0x600 __udp_enqueue_schedule_skb+0x4c8/0x650 udpv6_queue_rcv_one_skb+0x3ec/0x740 udp6_unicast_rcv_skb+0x11d/0x140 ip6_protocol_deliver_rcu+0x61e/0x950 ip6_input_finish+0xa9/0x150 NF_HOOK+0x286/0x2f0 ip6_input+0x117/0x220 NF_HOOK+0x286/0x2f0 __netif_receive_skb+0x85/0x200 process_backlog+0x374/0x9a0 __napi_poll+0x4f/0x1c0 net_rx_action+0x3b0/0x770 handle_softirqs+0x15a/0x460 do_softirq+0x57/0x80 </IRQ>

The rmem charge that dev_scratch accounted for is released by skb_recv_udp() on dequeue, just above, so the scratch is dead by the time recv_actor() runs. Clear skb->dev so bpf_skc_lookup() falls back to sock_net(skb->sk), which skb_set_owner_sk_safe() set just above.(CVE-2026-53184)

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

RDMA/srp: bound SRP_RSP sense copy by the received length

srp_process_rsp() copies sense data from rsp->data + resp_data_len, where resp_data_len is the full 32-bit value supplied by the SRP target and is never checked against the number of bytes actually received (wc->byte_len). The copy length is bounded to SCSI_SENSE_BUFFERSIZE, so at most 96 bytes are copied, but the source offset is not bounded.

A malicious or compromised SRP target on the InfiniBand/RoCE fabric that the initiator has logged into can return an SRP_RSP with SRP_RSP_FLAG_SNSVALID set and a large resp_data_len. The receive buffer is allocated at the target-chosen max_ti_iu_len, so the source of the sense copy lands past the bytes actually received; with resp_data_len near 0xFFFFFFFF it is gigabytes past the buffer and the read faults.

Copy the sense data only if it has not been truncated, that is, only if the response header, the response data, and the sense region fit within the bytes actually received; otherwise drop the sense and log. The in-tree iSER and NVMe-RDMA receive paths already bound their parse by wc->byte_len; this brings ib_srp into line with them.(CVE-2026-53186)

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

drm/virtio: fix dma_fence refcount leak on error in virtio_gpu_dma_fence_wait()

dma_fence_unwrap_for_each() internally calls dma_fence_unwrap_first() which does cursor->chain = dma_fence_get(head), taking an extra reference. On normal loop completion, dma_fence_unwrap_next() releases this via dma_fence_chain_walk() -> dma_fence_put().

When virtio_gpu_do_fence_wait() fails and the function returns early from inside the loop, the cursor->chain reference is never released. This is the only caller in the entire kernel that does an early return inside dma_fence_unwrap_for_each.

Add dma_fence_put(itr.chain) before the early return.(CVE-2026-53190)

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

mm/memory-failure: fix hugetlb_lock AA deadlock in get_huge_page_for_hwpoison

Two concurrent madvise(MADV_HWPOISON) calls on the same hugetlb page can trigger a recursive spinlock self-deadlock (AA deadlock) on hugetlb_lock when racing with a concurrent unmap:

thread#0 thread#1

madvise(folio, MADV_HWPOISON) -> poisons the folio successfully madvise(folio, MADV_HWPOISON) unmap(folio) try_memory_failure_hugetlb get_huge_page_for_hwpoison spin_lock_irq(&hugetlb_lock) <- held __get_huge_page_for_hwpoison hugetlb_update_hwpoison() -> MF_HUGETLB_FOLIO_PRE_POISONED goto out: folio_put() refcount: 1 -> 0 free_huge_folio() spin_lock_irqsave(&hugetlb_lock) -> AA DEADLOCK!

The out: path in __get_huge_page_for_hwpoison() calls folio_put() to drop the GUP reference while the hugetlb_lock is still held by the hugetlb.c wrapper get_huge_page_for_hwpoison(). If concurrent unmap has released the page table mapping reference, folio_put() drops the folio refcount to zero, triggering free_huge_folio() which attempts to re-acquire the non-recursive hugetlb_lock.

Fix this by moving hugetlb_lock acquisition from the hugetlb.c wrapper into get_huge_page_for_hwpoison(). Place spin_unlock_irq() before the folio_put() at the out: label so the folio is always released outside the lock.

[(CVE-2026-53207)

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

netfilter: nft_exthdr: fix register tracking for F_PRESENT flag

nft_exthdr_init() passes user-controlled priv->len to nft_parse_register_store(), which marks that many bytes in the register bitmap as initialized. However, when NFT_EXTHDR_F_PRESENT is set, the eval paths write only 1 byte (nft_reg_store8) or 4 bytes (*dest = 0 on TCP/DCCP error path). When len > 4, registers beyond the first are never written, retaining uninitialized stack data from nft_regs.

Bail out if userspace requests too much data when F_PRESENT is set.(CVE-2026-53218)

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

netfilter: x_tables: avoid leaking percpu counter pointers

The native and compat get-entries paths copy the fixed rule entry header from the kernelized rule blob to userspace before overwriting the entry's counter fields with a sanitized counter snapshot.

On SMP kernels, entry->counters.pcnt contains the percpu allocation address used by x_tables rule counters. A caller can provide a userspace buffer that faults during the initial fixed-header copy after pcnt has been copied but before the later sanitized counter copy runs. The syscall then returns -EFAULT while leaving the raw percpu pointer in userspace.

Copy only the fixed entry prefix before counters from the kernelized rule blob, then copy the sanitized counter snapshot into the counter field. Apply this ordering to the IPv4, IPv6, and ARP native and compat get-entries implementations so a fault cannot expose the internal percpu counter pointer.(CVE-2026-53219)

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

netfilter: revalidate bridge ports

ebt_redirect_tg() dereferences br_port_get_rcu() return without a NULL check, causing a kernel panic when the bridge port has been removed between the original hook invocation and an NFQUEUE reinject.

A mere NULL check isn't sufficient, however. As sashiko review points out userspace can not only remove the port from the bridge, it could also place the device in a different virtual device, e.g. macvlan.

If this happens, we must drop the packet, there is no way for us to reinject it into the bridge path.

Switch to _upper API, we don't need the bridge port structure. Also, this fix keeps another bug intact:

Both nfnetlink_log and nfnetlink_queue use CONFIG_BRIDGE_NETFILTER too aggressive, which prevents certain logging features when queueing in bridge family: NETFILTER_FAMILY_BRIDGE can be enabled while the old CONFIG_BRIDGE_NETFILTER cruft is off.

Fixes tag is a common ancestor, this was always broken.(CVE-2026-53220)

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

ip6_vti: fix incorrect tunnel matching in vti6_tnl_lookup()

In vti6_tnl_lookup(), when an exact match for a tunnel fails, the code falls back to searching for wildcard tunnels:

• Tunnels matching the packet's local address, with any remote address wildcard remote).

• Tunnels matching the packet's remote address, with any local address (wildcard local).

However, vti6 stores all these different types of tunnels in the same hash table (ip6n->tnls_r_l) prone to hash collisions.

The bug is that the fallback search loops in vti6_tnl_lookup() were missing checks to ensure that the candidate tunnel actually has a wildcard address.(CVE-2026-53221)

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

net: guard timestamp cmsgs to real error queue skbs

skb_is_err_queue() treats PACKET_OUTGOING as the sole marker for an skb from sk_error_queue. That assumption is not true for AF_PACKET sockets: outgoing packet taps are also delivered to packet sockets with skb->pkt_type == PACKET_OUTGOING, but their skb->cb is owned by AF_PACKET instead of struct sock_exterr_skb.

If such an skb is received with timestamping enabled, the generic timestamp cmsg path can read AF_PACKET control-buffer state as sock_exterr_skb::opt_stats. With SO_RXQ_OVFL enabled, the packet drop counter overlaps opt_stats. An odd drop count makes the path emit SCM_TIMESTAMPING_OPT_STATS with skb->len and skb->data. For non-linear skbs this copies past the linear head and can trigger hardened usercopy or disclose adjacent heap contents.

Keep skb_is_err_queue() local to net/socket.c, but make it verify that the PACKET_OUTGOING marker is paired with the sock_rmem_free destructor installed by sock_queue_err_skb(). AF_PACKET receive skbs use normal receive ownership and no longer pass as error-queue skbs, while legitimate sk_error_queue entries keep the PACKET_OUTGOING marker and sock_rmem_free ownership.(CVE-2026-53223)

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

net: openvswitch: fix possible kfree_skb of ERR_PTR

After the patch in the "Fixes" tag, the allocation of the "reply" skb can happen either before or after locking the ovs_mutex.

However, error cleanups still follow the classical reversed order, assuming "reply" is allocated before locking: it is freed after unlocking.

If "reply" allocation happens after locking the mutex and it fails, "reply" is left with an ERR_PTR, and execution jumps to the correspondent cleanup stage which will try to free an invalid pointer.

Fix this by setting the pointer to NULL after having saved its error value.(CVE-2026-53227)

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

ipv6: sit: reload inner IPv6 header after GSO offloads

ipip6_tunnel_xmit() caches the inner IPv6 header pointer at function entry and continues using it after iptunnel_handle_offloads().

For GSO skbs, iptunnel_handle_offloads() calls skb_header_unclone(). When the skb header is cloned, skb_header_unclone() can call pskb_expand_head(), which may move the skb head. The pskb_expand_head() contract requires pointers into the skb header to be reloaded after the call.

If the later skb_realloc_headroom() branch is not taken, SIT uses the stale iph6 pointer to read the inner hop limit and DS field. That can read from a freed skb head after the old head's remaining clone is released.

Reload iph6 after the offload helper succeeds and before subsequent reads from the inner IPv6 header. Keep the existing reload after skb_realloc_headroom(), since that branch can also replace the skb.(CVE-2026-53228)

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

net/mlx5e: xsk: Fix DMA and xdp_frame leak on XDP_TX xmit failure

In the XSK branch of mlx5e_xmit_xdp_buff(), when sq->xmit_xdp_frame() returns false (e.g. XDPSQ is full), the function returns without unmapping the DMA address or freeing the xdp_frame allocated by xdp_convert_zc_to_xdp_frame(). The xdpi_fifo push only happens on success, so the completion path cannot recover these entries.

With CONFIG_DMA_API_DEBUG=y, the leak surfaces on driver unbind:

DMA-API: pci 0000:08:00.0: device driver has pending DMA allocations while released from device [count=1116] One of leaked entries details: [device address=0x000000010ffd7028] [size=1534 bytes] [mapped with DMA_TO_DEVICE] [mapped as phy] WARNING: kernel/dma/debug.c:881 at dma_debug_device_change+0x127/0x180 ... DMA-API: Mapped at: debug_dma_map_phys+0x4b/0xd0 dma_map_phys+0xfd/0x2d0 mlx5e_xdp_handle+0x5ae/0xac0 [mlx5_core] mlx5e_xsk_skb_from_cqe_mpwrq_linear+0xc4/0x170 [mlx5_core] mlx5e_handle_rx_cqe_mpwrq+0xc1/0x290 [mlx5_core]

Add the missing unmap + xdp_return_frame, matching the cleanup already done in mlx5e_xdp_xmit(). has_frags is rejected earlier in this branch, so no per-frag unmap is needed.(CVE-2026-53229)

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

net/mlx5: Fix slab-out-of-bounds in mlx5_query_nic_vport_mac_list

mlx5_query_nic_vport_mac_list() sizes its firmware command buffer using the PF's log_max_current_uc/mc_list capabilities. When querying a VF vport with a larger configured max (via devlink), the firmware response can overflow this buffer:

BUG: KASAN: slab-out-of-bounds in mlx5_query_nic_vport_mac_list+0x453/0x4c0 [mlx5_core] Read of size 4 at addr ff1100013ffc8a12 by task kworker/u96:2/385

CPU: 12 UID: 0 PID: 385 Comm: kworker/u96:2 Not tainted 7.0.0-rc6+ #1 PREEMPT Hardware name: QEMU Standard PC (Q35 + ICH9, 2009) Workqueue: mlx5_esw_wq esw_vport_change_handler [mlx5_core] Call Trace: <TASK> dump_stack_lvl+0x69/0xa0 print_report+0x176/0x4e4 kasan_report+0xc8/0x100 mlx5_query_nic_vport_mac_list+0x453/0x4c0 [mlx5_core] esw_update_vport_addr_list+0x2e3/0xda0 [mlx5_core] esw_vport_change_handle_locked+0xa1f/0x1060 [mlx5_core] esw_vport_change_handler+0x6a/0x90 [mlx5_core] process_one_work+0x87f/0x15e0 worker_thread+0x62b/0x1020 kthread+0x375/0x490 ret_from_fork+0x4dc/0x810 ret_from_fork_asm+0x11/0x20 </TASK>

Fix by querying the vport's own HCA caps to size the buffer correctly. Refactor the function to allocate and return the MAC list internally, removing the caller's dependency on knowing the correct max.(CVE-2026-53230)

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

tcp: restrict SO_ATTACH_FILTER to priv users

This patch restricts the use of SO_ATTACH_FILTER (cBPF) on TCP sockets to users with CAP_NET_ADMIN capability.

This blocks potential side-channel attack where an unprivileged application attaches a filter to leak TCP sequence/acknowledgment numbers.(CVE-2026-53236)

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

netlabel: validate unlabeled address and mask attribute lengths

netlbl_unlabel_addrinfo_get() used the address attribute length to determine whether the attribute data could be read as an IPv4 or IPv6 address, but did not independently validate the corresponding mask attribute length. A crafted Generic Netlink request could therefore provide a valid IPv4/IPv6 address attribute with a shorter mask attribute, which would later be read as a full struct in_addr or struct in6_addr.

NLA_BINARY policy lengths are maximum lengths by default, so use NLA_POLICY_EXACT_LEN() for the unlabeled IPv4/IPv6 address and mask attributes. This rejects short attributes during policy validation and also exposes the exact length requirements through policy introspection.(CVE-2026-53238)

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

xfrm: policy: fix use-after-free on inexact bin in xfrm_policy_bysel_ctx()

Fix the race by pruning the bin while still holding xfrm_policy_lock, before dropping it. Use __xfrm_policy_inexact_prune_bin() directly since the lock is already held. The wrapper xfrm_policy_inexact_prune_bin() becomes unused and is removed.

Race:

CPU0 (XFRM_MSG_DELPOLICY) CPU1 (XFRM_MSG_NEWSPDINFO) ========================== ========================== xfrm_policy_bysel_ctx(): spin_lock_bh(xfrm_policy_lock) bin = xfrm_policy_inexact_lookup() __xfrm_policy_unlink(pol) spin_unlock_bh(xfrm_policy_lock) xfrm_policy_kill(ret) // wide window, lock not held xfrm_hash_rebuild(): spin_lock_bh(xfrm_policy_lock) __xfrm_policy_inexact_flush(): kfree_rcu(bin) // bin freed spin_unlock_bh(xfrm_policy_lock) xfrm_policy_inexact_prune_bin(bin) // UAF: bin is freed(CVE-2026-53239)

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

ALSA: PCM: Fix wait queue list corruption in snd_pcm_drain() on linked streams

snd_pcm_drain() uses init_waitqueue_entry which does not clear entry.prev/next, and add_wait_queue with a conditional remove_wait_queue that is skipped when to_check is no longer in the group after concurrent UNLINK. The orphaned wait entry remains on the unlinked substream sleep queue. On the next drain iteration, add_wait_queue adds the entry to a new queue while still linked on the old one, corrupting both lists. A subsequent wake_up dereferences NULL at the func pointer (mapped from the spinlock at offset 0 of the misinterpreted wait_queue_head_t), causing a kernel panic.

Replace init_waitqueue_entry/add_wait_queue/conditional remove_wait_queue with init_wait_entry/prepare_to_wait/ finish_wait. init_wait_entry clears prev/next via INIT_LIST_HEAD on each iteration and sets autoremove_wake_function which auto-removes the entry on wake-up. finish_wait safely handles both the already-removed and still-queued cases.(CVE-2026-53242)

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

net/802/mrp: fix vector attribute parsing in mrp_pdu_parse_vecattr

In mrp_pdu_parse_vecattr(), vector attribute events are encoded three per byte and valen tracks the number of events left to process.

The parser decrements valen after processing the first and second events from each event byte, but not after processing the third one. When valen is exactly a multiple of three, the loop continues after the last valid event and consumes the next byte as a new event byte, applying a spurious event to the MRP applicant state.

Additionally, when valen is zero the parser unconditionally consumes attrlen bytes as FirstValue and advances the offset, even though per IEEE 802.1ak a VectorAttribute with only a LeaveAllEvent has valen of zero and no FirstValue or Vector fields. This corrupts the offset for subsequent PDU parsing.

Also, when valen exceeds three the loop crosses byte boundaries but the attribute value is not incremented between the last event of one byte and the first event of the next. This causes the first event of the next byte to use the same attribute value as the third event rather than the next consecutive value.

Decrement valen after processing the third event, skip FirstValue consumption when valen is zero, and increment the attribute value at the end of each loop iteration.(CVE-2026-53245)

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

sctp: validate cached peer INIT chunk length in COOKIE_ECHO processing

When a listening SCTP server processes a COOKIE_ECHO chunk, the cached peer INIT chunk embedded after the cookie is parsed and its parameters are later walked by sctp_process_init() using sctp_walk_params().

However, the chunk header length of this cached INIT chunk was not validated against the remaining buffer in the COOKIE_ECHO payload. If the length field is inflated, the parameter walk can run beyond the actual received data, leading to out-of-bounds reads and potential memory corruption during later parameter handling (e.g. STATE_COOKIE processing and kmemdup() copies).

Add a bounds check in sctp_unpack_cookie() to ensure the cached INIT chunk length does not exceed the available data in the COOKIE_ECHO buffer before it is used.(CVE-2026-53246)

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

ipv4: restrict IPOPT_SSRR and IPOPT_LSRR options

This patch restricts setting Loose Source and Record Route (LSRR) and Strict Source and Record Route (SSRR) IP options to users with CAP_NET_RAW capability.

This prevents unprivileged applications from forcing packets to route through attacker-controlled nodes to leak TCP ISN and possibly other protocol information.

While LSRR and SSRR are commonly filtered in many network environments, they may still be supported and forwarded along some network paths.

RFC 7126 (Recommendations on Filtering of IPv4 Packets Containing IPv4 Options) recommend to drop these options in 4.3 and 4.4.(CVE-2026-53249)

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

Bluetooth: fix memory leak in error path of hci_alloc_dev()

Early failures in Bluetooth HCI UART configuration leak SRCU percpu memory.

When device initialization fails before hci_register_dev() completes, the HCI_UNREGISTER flag is never set. As a result, when the device reference count reaches zero, bt_host_release() evaluates this flag as false and falls back to a direct kfree(hdev).

Because hci_release_dev() is bypassed, the SRCU struct initialized early in hci_alloc_dev() is never cleaned up, resulting in a leak of percpu memory.

Fix the leak by explicitly calling cleanup_srcu_struct() in the fallback (unregistered) branch of bt_host_release() before freeing the device.(CVE-2026-53252)

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

Bluetooth: bnep: reject short frames before parsing

A BNEP peer can send a short BNEP SDU. bnep_rx_frame() reads the packet type byte immediately and, for control packets, reads the control opcode and setup UUID-size byte before proving that those bytes are present. bnep_rx_control() also dereferences the control opcode without rejecting an empty control payload.

Use skb_pull_data() for the fixed fields in bnep_rx_frame() so a NULL return gates each dereference. Split the control handler so the frame path can pass an opcode that has already been pulled, and keep the byte-buffer wrapper for extension control payloads.

For BNEP_SETUP_CONN_REQ, name the UUID-size byte before pulling the setup payload. struct bnep_setup_conn_req carries destination and source service UUIDs after that byte, each uuid_size bytes, so the parser now documents that tuple explicitly instead of leaving the pull length as an opaque multiplication.

Validation reproduced this kernel report: KASAN slab-out-of-bounds in bnep_rx_frame.isra.0+0x130c/0x1790 The buggy address belongs to the object at ffff88800c0f7908 which belongs to the cache kmalloc-8 of size 8 The buggy address is located 0 bytes to the right of allocated 1-byte region [ffff88800c0f7908, ffff88800c0f7909) Read of size 1 Call trace: dump_stack_lvl+0xb3/0x140 (?:?) print_address_description+0x57/0x3a0 (?:?) bnep_rx_frame+0x130c/0x1790 (net/bluetooth/bnep/core.c:306) print_report+0xb9/0x2b0 (?:?) __virt_addr_valid+0x1ba/0x3a0 (?:?) srso_alias_return_thunk+0x5/0xfbef5 (?:?) kasan_addr_to_slab+0x21/0x60 (?:?) kasan_report+0xe0/0x110 (?:?) process_one_work+0xfce/0x17e0 (kernel/workqueue.c:3200) worker_thread+0x65c/0xe40 (?:?) __kthread_parkme+0x184/0x230 (?:?) kthread+0x35e/0x470 (?:?) _raw_spin_unlock_irq+0x28/0x50 (?:?) ret_from_fork+0x586/0x870 (?:?) __switch_to+0x74f/0xdc0 (?:?) ret_from_fork_asm+0x1a/0x30 (?:?)(CVE-2026-53253)

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

Bluetooth: RFCOMM: validate skb length in MCC handlers

The RFCOMM MCC handlers cast skb->data to protocol-specific structs without validating skb->len first. A malicious remote device can send truncated MCC frames and trigger out-of-bounds reads in these handlers.

Fix this by using skb_pull_data() to validate and access the required data before dereferencing it.

rfcomm_recv_rpn() requires special handling since ETSI TS 07.10 allows 1-byte RPN requests. Handle this by validating only the DLCI byte first, and validating the full struct only when len > 1.(CVE-2026-53254)

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

Bluetooth: MGMT: validate advertising TLV before type checks

tlv_data_is_valid() reads each advertising data field length from data[i], then inspects data[i + 1] for managed EIR types before checking that the current field still fits inside the supplied buffer.

A malformed field whose length byte is the last byte of the buffer can therefore make the parser read one byte past the advertising data.

KASAN reported the following when a malformed MGMT_OP_ADD_ADVERTISING request reached that path:

BUG: KASAN: vmalloc-out-of-bounds in tlv_data_is_valid() Read of size 1 Call trace: tlv_data_is_valid() add_advertising() hci_mgmt_cmd() hci_sock_sendmsg()

Move the existing element-length check before any type-octet inspection so each non-empty element is proven to contain its type byte before the parser looks at data[i + 1].(CVE-2026-53255)

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

Bluetooth: RFCOMM: hold listener socket in rfcomm_connect_ind()

rfcomm_get_sock_by_channel() scans rfcomm_sk_list under the list lock, but returns the selected listener after dropping that lock without taking a reference. rfcomm_connect_ind() then locks the listener, queues a child socket on it, and may notify it after unlocking it.

The buggy scenario involves two paths, with each column showing the order within that path:

rfcomm_connect_ind(): listener close:

1. Find parent in 1. close() enters rfcomm_get_sock_by_channel() rfcomm_sock_release().
2. Drop rfcomm_sk_list.lock 2. rfcomm_sock_shutdown() without pinning parent. closes the listener.
3. Call lock_sock(parent) and 3. rfcomm_sock_kill() bt_accept_enqueue(parent, unlinks and puts parent. sk, true).
4. Read parent flags and may 4. parent can be freed. call sk_state_change().

If close wins the race, parent can be freed before rfcomm_connect_ind() reaches lock_sock(), bt_accept_enqueue(), or the deferred-setup callback.

Take a reference on the listener before leaving rfcomm_sk_list.lock. After lock_sock() succeeds, recheck that it is still in BT_LISTEN before queueing a child, cache the deferred-setup bit while the parent is locked, and drop the reference after the last parent use.

KASAN reported a slab-use-after-free in lock_sock_nested() from rfcomm_connect_ind(), with the freeing stack going through rfcomm_sock_kill() and rfcomm_sock_release().(CVE-2026-53256)

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

net/sched: act_api: use RCU with deferred freeing for action lifecycle

When NEWTFILTER and DELFILTER are run concurrently it is possible to create a race with an associated action.

Let's illustrate with CPU0 running NEWTFILTER and CPU1 running DELFILTER:

0: mutex_lock() <-- holds the idr lock 0: rcu_read_lock() 0: p = idr_find(idr, index) <-- action p is valid (RCU protects IDR) 0: mutex_unlock() <-- releases the idr lock 1: refcount_dec_and_mutex_lock() <-- refcnt 1->0, mutex held 1: idr_remove(idr, index) <-- Action removed from IDR 1: mutex_unlock() <-- mutex released allowing us to delete the action 1: tcf_action_cleanup(p); kfree(p) <-- Kfrees p immediately, no deferral 0: refcount_inc_not_zero(&p->tcfa_refcnt) <-- ouch, UAF p points to freed memory

This patch fixes the race condition between NEWTFILTER and DELFILTER by adding struct rcu_head to tc_action used in the deferral and introducing a call_rcu() in the delete path to defer the final kfree().

Note: this is a revert of commit d7fb60b9cafb ("net_sched: get rid of tcfa_rcu") but also modernization/simplification to directly use kfree_rcu().

Let's illustrate the new restored code path:

0: rcu_read_lock() 1: refcount_dec_and_mutex_lock() <-- refcnt 1->0, mutex held 1: idr_remove(idr, index) 1: mutex_unlock() 1: call_rcu(&p->tcfa_rcu, tcf_action_rcu_free) <-- defer kfree after grace period 0: p = idr_find(idr, index) 0: refcount_inc_not_zero(&p->tcfa_refcnt) <-- fails, refcnt already 0 1: rcu_read_unlock() <-- release so freeing can run after grace period

After CPU1 calls idr_remove(), the object is no longer reachable through the IDR. CPU0's subsequent idr_find() will return NULL, and even if it still held a stale pointer, the immediate kfree() is now deferred until after the RCU grace period, so no UAF can occur.(CVE-2026-53264)

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

netfilter: bridge: make ebt_snat ARP rewrite writable

The ebtables SNAT target keeps the Ethernet source address rewrite behind skb_ensure_writable(skb, 0). This is intentional: at the bridge ebtables hooks the Ethernet header is addressed through skb_mac_header()/eth_hdr(), while skb->data points at the Ethernet payload. Asking skb_ensure_writable() for ETH_HLEN bytes would check the payload, not the Ethernet header, and would reintroduce the small packet regression fixed by commit 63137bc5882a.

However, the optional ARP sender hardware address rewrite is different. It writes through skb_store_bits() at an offset relative to skb->data:

    skb_store_bits(skb, sizeof(struct arphdr), info->mac, ETH_ALEN)

skb_header_pointer() only safely reads the ARP header; it does not make the later sender hardware address range writable. If that range is still held in a nonlinear skb fragment backed by a splice-imported file page, skb_store_bits() maps the frag page and copies the new MAC address directly into it.

Ensure the ARP SHA range is writable before reading the ARP header and before calling skb_store_bits().(CVE-2026-53266)

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

netfilter: nft_ct: bail out on template ct in get eval

I noticed this issue while looking at a historic syzbot report 1.

A rule like the one below is enough to trigger the bug:

table ip t {
    chain pre {
        type filter hook prerouting priority raw;
        ct zone set 1
        ct original saddr 1.2.3.4 accept
    }
}

The first expression attaches a per-cpu template ct via nft_ct_set_zone_eval() (nf_ct_tmpl_alloc -> kzalloc, tuple is all zero, nf_ct_l3num(ct) == 0). The next expression then calls nft_ct_get_eval() on the same skb, treats the template as a real ct and hits the 16-byte memcpy path. With dreg at NFT_REG32_15 this overflows past struct nft_regs on the kernel stack; with smaller dreg values it silently clobbers adjacent registers.

Reject template ct at the eval entry and in nft_ct_get_fast_eval(), mirroring the check nft_ct_set_eval() already has. Additionally, bound the address copy in NFT_CT_SRC / NFT_CT_DST by priv->len instead of by nf_ct_l3num(ct): nf_ct_get_tuple() zeroes the tuple before pkt_to_tuple() fills in only the protocol-relevant leading bytes, so the trailing bytes of tuple->{src,dst}.u3.all are well-defined zero. priv->len is validated at rule load, so the copy size is now bounded by the destination register rather than by an untrusted field on the conntrack.

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

netfilter: synproxy: add mutex to guard hook reference counting

As the synproxy infrastructure register netfilter hooks on-demand when a user adds the first iptables target or nftables expression, if done concurrently they can race each other.

Introduce a mutex to serialize the refcount control blocks access from both frontends. While a per namespace mutex might be more efficient, it is not needed for target/expression like SYNPROXY.(CVE-2026-53269)

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

ipvs: clear the svc scheduler ptr early on edit

ip_vs_edit_service() while unbinding the old scheduler clears the svc->scheduler ptr after the scheduler module initiates RCU callbacks. This can cause packets to use the old scheduler at the time when svc->sched_data is already freed after RCU grace period.

Fix it by clearing the ptr early in ip_vs_unbind_scheduler(), before the done_service method schedules any RCU callbacks.

Also, if the new scheduler fails to initialize when replacing the old scheduler, try to restore the old scheduler while still returning the error code.(CVE-2026-53270)

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

net: bonding: fix NULL pointer dereference in bond_do_ioctl()

In bond_do_ioctl(), slave_dev is obtained via __dev_get_by_name() which can return NULL if the requested interface name does not exist. However, the subsequent slave_dbg() call is placed before the NULL check:

slave_dev = __dev_get_by_name(net, ifr->ifr_slave);
slave_dbg(bond_dev, slave_dev, "slave_dev=%p:\n", slave_dev); //here
if (!slave_dev)
    return -ENODEV;

The slave_dbg() macro expands to netdev_dbg(bond_dev, "(slave %s): " fmt, (slave_dev)->name, ...) which unconditionally dereferences slave_dev->name before the NULL check is performed. This results in a NULL pointer dereference kernel oops when a user calls bonding ioctl (e.g. SIOCBONDENSLAVE, SIOCBONDRELEASE, etc.) with a non-existent slave interface name.

This is reachable from userspace via the bonding ioctl interface with CAP_NET_ADMIN capability, making it a potential local denial-of-service vector.

Fix by moving the slave_dbg() call after the NULL check.(CVE-2026-53337)

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

netfilter: nf_conntrack: destroy stale expectfn expectations on unregister

NAT helpers such as nf_nat_h323 store a raw pointer to module text in exp->expectfn (e.g. ip_nat_q931_expect). nf_ct_helper_expectfn_unregister() only unlinks the callback descriptor and never walks the expectation table, so an expectation pending at module removal survives with a dangling exp->expectfn into freed module text.

When the expected connection arrives, init_conntrack() invokes exp->expectfn(), now a stale pointer into the unloaded module. Reproduced on a KASAN build by loading the H.323 helpers, creating a Q.931 expectation, unloading nf_nat_h323, then connecting to the expected port:

Oops: int3: 0000 [#1] SMP KASAN NOPTI RIP: 0010:0xffffffffa06102d1 init_conntrack.isra.0 (net/netfilter/nf_conntrack_core.c:1862) nf_conntrack_in (net/netfilter/nf_conntrack_core.c:2049) ipv4_conntrack_local (net/netfilter/nf_conntrack_proto.c:223) nf_hook_slow (net/netfilter/core.c:619) __ip_local_out (net/ipv4/ip_output.c:120) __tcp_transmit_skb (net/ipv4/tcp_output.c:1715) tcp_connect (net/ipv4/tcp_output.c:4374) tcp_v4_connect (net/ipv4/tcp_ipv4.c:345) __sys_connect (net/socket.c:2167) Modules linked in: nf_conntrack_h323 [last unloaded: nf_nat_h323]

Reaching the dangling state requires CAP_SYS_MODULE in the initial user namespace to remove a NAT helper that still has live expectations, so this is a robustness fix; leaving an expectation pointing at freed text is wrong regardless.

Add nf_ct_helper_expectfn_destroy(), which walks the expectation table and drops every expectation whose ->expectfn matches the descriptor being torn down. Call it from each NAT helper's exit path after the existing RCU grace period, so no expectation outlives the code it points at and no extra synchronize_rcu() is introduced. With the fix, the same reproducer runs to completion without the Oops.(CVE-2026-53349)

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

Bluetooth: fix UAF in l2cap_sock_cleanup_listen() vs l2cap_conn_del()

bt_accept_dequeue() unlinks a not-yet-accepted child from the parent accept queue and release_sock()s it before returning, so the returned sk has no caller reference and is unlocked.

l2cap_sock_cleanup_listen() walks these children on listening-socket close. A concurrent HCI disconnect drives hci_rx_work -> l2cap_conn_del() which runs l2cap_chan_del() + l2cap_sock_kill() and frees the child sk and its l2cap_chan; cleanup_listen() then uses both:

BUG: KASAN: slab-use-after-free in l2cap_sock_kill l2cap_sock_kill / l2cap_sock_cleanup_listen / __x64_sys_close Freed by: l2cap_conn_del -> l2cap_sock_close_cb -> l2cap_sock_kill

This is distinct from the two fixes already in this area: commit e83f5e24da741 ("Bluetooth: serialize accept_q access") serialises the accept_q list/poll and takes temporary refs inside bt_accept_dequeue(), and CVE-2025-39860 serialises the userspace close()/accept() race by calling cleanup_listen() under lock_sock() in l2cap_sock_release(). Neither covers l2cap_conn_del() running from hci_rx_work, so this UAF still reproduces on current bluetooth/master.

Take the reference at the source: bt_accept_dequeue() does sock_hold() while sk is still locked, before release_sock(); callers sock_put(). cleanup_listen() pins the chan with l2cap_chan_hold_unless_zero() under a brief child sk lock (serialising vs l2cap_sock_teardown_cb()), drops it before l2cap_chan_lock(), and skips a duplicate l2cap_sock_kill() on SOCK_DEAD. conn->lock is not taken here: cleanup_listen() runs under the parent sk lock and that would invert conn->lock -> chan->lock -> sk_lock (lockdep).

KASAN/SMP: an unprivileged listen/close vs HCI-disconnect race produced 12 use-after-free reports per run before this change; 0, and no lockdep report, over 1600+ raced iterations after it on bluetooth/master.(CVE-2026-53357)

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

Bluetooth: L2CAP: use chan timer to close channels in cleanup_listen()

l2cap_chan_close() removes the channel from conn->chan_l, which must be done under conn->lock. cleanup_listen() runs under the parent sk_lock, so acquiring conn->lock would invert the established conn->lock -> chan->lock -> sk_lock order.

Instead of calling l2cap_chan_close() directly, schedule l2cap_chan_timeout with delay 0 to close the channel asynchronously. The timeout handler already acquires conn->lock and chan->lock in the correct order.

The timer is only armed when chan->conn is still set: if it is already NULL, l2cap_conn_del() has already processed this channel (l2cap_chan_del + l2cap_sock_teardown_cb + l2cap_sock_close_cb), so there is nothing left to do. If l2cap_conn_del() races in after the timer is armed, __clear_chan_timer() inside l2cap_chan_del() cancels it; if the timer has already fired, the handler returns harmlessly because chan->conn was cleared.(CVE-2026-53358)

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

KVM: x86: Fix shadow paging use-after-free due to unexpected role

Commit 0cb2af2ea66ad ("KVM: x86: Fix shadow paging use-after-free due to unexpected GFN") fixed a shadow paging mismatch between stored and computed GFNs; the bug could be triggered by changing a PDE mapping from outside the guest, and then deleting a memslot. The rmap_remove() call would miss entries created after the PDE change because the GFN of the leaf SPTE does not match the GFN of the struct kvm_mmu_page.

A similar hole however remains if the modified PDE points to a non-leaf page. In this case the gfn can be made to match, but the role does not match: the original large 2MB page creates a kvm_mmu_page with direct=1, while the new 4KB needs a kvm_mmu_page with direct=0. However, kvm_mmu_get_child_sp() does not compare the role, and therefore reuses the page.

The next step is installing a leaf (4KB) SPTE on the new path which records an rmap entry under the gfn resolved by the walk. But when that child is zapped its parent kvm_mmu_page has direct=1 and kvm_mmu_page_get_gfn() computes the gfn for the 4KB page as sp->gfn + index instead of using sp->shadowed_translation[] (or sp->gfns[] in older kernels). It therefore fails to remove the recorded entry.

When the memslot is dropped the shadow page is freed but the rmap entry survives, as in the scenario that was already fixed. Code that later walks that gfn (dirty logging, MMU notifier invalidation, and so on) dereferences an sptep that lies in the freed page, causing the use-after-free.(CVE-2026-53359)

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

blk-cgroup: fix UAF in __blkcg_rstat_flush()

When multiple blkgs in the same blkcg are released concurrently, a use-after-free can occur. The race happens when one blkg's __blkcg_rstat_flush() removes another blkg's iostat entries via llist_del_all(). The second blkg sees an empty list and proceeds to free itself while the first is still iterating over its entries.

Move the flush from __blkg_release() (RCU callback) to blkg_release() (before call_rcu). This ensures the RCU grace period waits for any concurrent flush's rcu_read_lock() section to complete before freeing.(CVE-2026-63802)

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

Bluetooth: hci_sync: Set HCI_CMD_DRAIN_WORKQUEUE during device close

Since hci_dev_close_sync() can now be called during the reset path, we should also set HCI_CMD_DRAIN_WORKQUEUE. This avoids queuing timeouts while the hdev workqueue is being drained.(CVE-2026-63974)

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

Bluetooth: L2CAP: Fix possible crash on l2cap_ecred_conn_rsp

If dcid is received for an already-assigned destination CID the spec requires that both channels to be discarded, but calling l2cap_chan_del may invalidate the tmp cursor created by list_for_each_entry_safe and in fact it is the wrong procedure as the chan->dcid may be assigned previously it really needs to be disconnected.

Calling l2cap_chan_clone directly may still lead to l2cap_chan_del so instead schedule l2cap_chan_timeout with delay 0 to close the channel asynchronously.(CVE-2026-63975)

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

irq_work: Fix use-after-free in irq_work_single() on PREEMPT_RT

On PREEMPT_RT, non-HARD irq_work runs in per-CPU kthreads via run_irq_workd(), so irq_work_sync() uses rcuwait() to wait for BUSY==0.

After irq_work_single() clears BUSY via atomic_cmpxchg(), it still dereferences @work for irq_work_is_hard() and rcuwait_wake_up().

An irq_work_sync() caller on another CPU that enters after BUSY is cleared can observe BUSY==0 immediately, return, and free the work before those accesses complete — causing a use-after-free.

Fix this by wrapping run_irq_workd() in guard(rcu)() so that the entire irq_work_single() execution is within an RCU read-side critical section. Then add synchronize_rcu() in irq_work_sync() after rcuwait_wait_event() to ensure the caller waits for the RCU grace period before returning, preventing premature frees.(CVE-2026-64073)

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

rbd: eliminate a race in lock_dwork draining on unmap

Given how rbd_lock_add_request() and rbd_img_exclusive_lock() are written, lock_dwork may be (re)queued more than it's actually needed: for example in case a new I/O request comes in while we are in the middle of rbd_acquire_lock() on behalf of another I/O request. This is expected and with rbd_release_lock() preemptively canceling lock_dwork is benign under normal operation.

A more problematic example is maybe_kick_acquire():

if (have_requests || delayed_work_pending(&rbd_dev->lock_dwork)) {
        dout("%s rbd_dev %p kicking lock_dwork\n", __func__, rbd_dev);
        mod_delayed_work(rbd_dev->task_wq, &rbd_dev->lock_dwork, 0);
}

It's not unrealistic for lock_dwork to get canceled right after delayed_work_pending() returns true and for mod_delayed_work() to requeue it right there anyway. This is a classic TOCTOU race.

When it comes to unmapping the image, there is an implicit assumption of no self-initiated exclusive lock activity past the point of return from rbd_dev_image_unlock() which unlocks the lock if it happens to be held. This unlock is assumed to be final and lock_dwork (as well as all other exclusive lock tasks, really) isn't expected to get queued again. However, lock_dwork is canceled only in cancel_tasks_sync() (i.e. later in the unmap sequence) and on top of that the cancellation can get in effect nullified by maybe_kick_acquire(). This may result in rbd_acquire_lock() executing after rbd_dev_device_release() and rbd_dev_image_release() run and free and/or reset a bunch of things. One of the possible failure modes then is a violated

rbd_assert(rbd_image_format_valid(rbd_dev->image_format));

in rbd_dev_header_info() which is called via rbd_dev_refresh() from rbd_post_acquire_action().

Redo exclusive lock task draining to provide saner semantics and try to meet the assumptions around rbd_dev_image_unlock().(CVE-2026-64112)