Information Disclosure
Monthly
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7915: clear wcid mask under mutex after RCU pointer clear mt7915_remove_interface() cleared the wcid mask bit with no lock held and before clearing the RCU wcid pointer. The mask is a non-atomic RMW shared with the allocators, which all run under dev->mt76.mutex; on DBDC the two wiphys share one mt76_dev, so this raced add_interface/sta_add on the other band and could leak or double-hand-out a wcid. Clearing the bit before the RCU pointer also let a concurrent allocation reuse the index and publish its wcid, which the subsequent NULL assignment then wiped. Move the clear into the existing mutex section, after the RCU pointer is cleared.
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7996: bound TLV walk in mt7996_mcu_get_chip_config The response TLV loop advanced by tlv->len without a minimum, so a theoretical firmware response containing a zero-length TLV could spin forever, hanging the CPU during device probe. The u32 payload was also read without bounds checking. Reject a short fixed field, stop on a TLV whose length underruns the header or overruns the skb.
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7915: unwind state on add_interface failure When mt76_wcid_alloc() fails, mt7915_add_interface() returned without clearing the vif_mask/omac_mask bits it had already set, without removing the firmware dev info added earlier, and without clearing a monitor_vif pointer to the vif mac80211 is about to free. mac80211 does not call remove_interface() for a failed add, so the indices and firmware dev entry leaked permanently and testmode could dereference the stale monitor_vif. Add a proper error unwind.
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7996: reserve space for the CSA-abort countdown TLV When a CSA countdown is active, mt7996_mcu_beacon_cntdwn() emits two bss_bcn_cntdwn_tlv entries (the CSA countdown and the CCA-abort BCC), but MT7996_BEACON_UPDATE_SIZE only reserved one. With MBSSID enabled and a near-maximum beacon template the extra 8 bytes could push the offload command past MT7996_MAX_BSS_OFFLOAD_SIZE and trigger skb_over_panic(). Reserve room for both countdown TLVs.
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: cancel reset and rc work on device unregister Both drivers cancelled dump_work on unregister but left reset_work and rc_work to be flushed only by destroy_workqueue() in mt76_free_device(), which runs after the hw is unregistered and the hardware stopped. A reset_work that fires in that window calls ieee80211_restart_hw() and re-arms mac_work on an unregistered hw, and rc_work touches station state being torn down. Cancel both up front, alongside dump_work.
In the Linux kernel, the following vulnerability has been resolved: ACPI: processor: Unregister cpufreq notifier on init failure acpi_processor_driver_init() registers the cpufreq policy notifier before registering the ACPI processor driver and setting up CPU hotplug state. If driver_register() or cpuhp_setup_state() fails, the error path only unregisters the ACPI processor driver and the idle driver. The cpufreq notifier remains registered even though initialization failed. Mirror the module exit path on the init failure path and unregister the cpufreq notifier when it has been registered.
In the Linux kernel, the following vulnerability has been resolved: drm/msm/dsi: Drop dev_pm_opp_set_rate(0) dev_pm_opp_set_rate(0) removes the vote specified in required-opps but does not actually park the clock, making it run without the necessary power backing. Drop the explicit call to it. Every call site of ops->link_clk_disable() is followed by pm_runtime_put(), so the power vote will be rescinded if deemed safe. Patchwork: https://patchwork.freedesktop.org/patch/742783/
In the Linux kernel, the following vulnerability has been resolved: wifi: ath11k: fix leak in ath11k_service_ready_ext_event() Currently, during ath11k_service_ready_ext_event() processing, svc_rdy_ext.mac_phy_caps can be allocated during TLV parsing. This is a temporary allocation that is freed on the success path, but not on the error path. If parsing succeeds far enough to allocate mac_phy_caps and then fails on a later TLV, the allocation leaks. So free the allocation on the error path. Compile tested only.
In the Linux kernel, the following vulnerability has been resolved: regulator: core: use system_freezable_wq for init complete work schedule_delayed_work() uses system_wq, which is non-freezable, allowing regulator_init_complete_work to run concurrently with system suspend. This work fires ~30s after boot to disable unused regulators via I2C. When it races with PM suspend, the I2C adapter may already be suspended, triggering a -ESHUTDOWN warning in __i2c_transfer(): WARNING: ... at __i2c_transfer+0x36c/0x3c8 Call trace: __i2c_transfer i2c_transfer regmap_i2c_write _regmap_update_bits regulator_disable_regmap _regulator_do_disable regulator_late_cleanup regulator_init_complete_work_function process_one_work Switch to system_freezable_wq so the work is frozen before any device is suspended, eliminating the race.
In the Linux kernel, the following vulnerability has been resolved: bpf: Reject >8 byte return values on return-reading trampoline paths btf_distill_func_proto() builds the function model used for the fentry/fexit/fmod_ret/fsession trampolines and struct_ops. It has accepted a 16-byte __int128 return value since the trampoline was introduced: __get_type_size() returns the integer's type size, and the return-type check only rejected ret < 0. But the BPF trampoline preserves only 8 bytes of the return value (RAX on x86, i.e. R0). For an attach type that reads the target's return value the second half (RDX / R3) is neither saved nor restored, so a program attached to a function returning a 16-byte value corrupts the value seen by the real caller and itself observes only half of it. struct_ops trampolines have the same limitation. This affects the attach types that read the target's return value: fexit, fmod_ret and fsession (plus the _multi variants of fexit and fsession), and struct_ops. fentry/fentry_multi run before the target returns and are unaffected. Reject a >8 byte return value for these attach types in bpf_check_attach_target() and bpf_check_attach_btf_id_multi(), and for struct_ops in bpf_struct_ops_desc_init().
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7996: free vif links after clearing wcid entries on full reset mt7996_mac_reset_vif_iter() queues non-default vif links for kfree_rcu while dev->wcid[] still holds pointers to the wcid embedded in each freed link; mt76_reset_device() then dereferences those entries and runs mt76_wcid_cleanup() on them. If a grace period elapses in between, the cleanup operates on freed memory. Run mt76_reset_device() first, so the wcid entries are cleaned up and cleared while the links are still valid.
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7996: clear stale link state on full reset After a full chip reset, mac80211 reconfig replays interface, link and channel context setup. mt7996_vif_link_add() short-circuits when the link_id is still marked in mvif->valid_links, a state introduced for postponing link teardown to interface removal. The reset path frees the link structures without clearing those bits, so the replayed setup never re-creates dev_info/bss_info/STA records in the restarted firmware and never re-registers the link wcid, leaving the device inoperative. The reset path also leaks every allocated MLD index: per-link indices and the per-vif group/remap indices are re-allocated from scratch during reconfig, but the old bits stay set in the masks, so repeated full resets exhaust the index space. Clear valid_links in the reset vif iterator and reset the MLD index masks alongside the existing omac_mask clearing.
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7915: fix double hif2 init on the non-WED path mt7915_pci_init_hif2() was called unconditionally and again inside the WED-inactive branch. The helper increments the global hif_idx, writes the PCIe RECOG_ID register and takes a get_device() reference via mt7915_pci_get_hif2(), while removal only drops one reference. On non-WED dual-hif hardware this double-incremented hif_idx, wrote RECOG_ID twice and leaked a device reference. Only the call inside the WED-inactive branch is correct; drop the unconditional one. hif2 is already initialised to NULL.
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7915: release hif2 reference on probe IRQ failure The hif2 reference obtained by mt7915_pci_init_hif2() is only released on error paths that key off dev->hif2, which is not assigned until after the IRQ setup. If pci_alloc_irq_vectors() or the primary devm_request_irq() fails, the reference leaks. Drop it explicitly on those paths via mt7915_put_hif2().
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: reject out-of-range link ids in mt76_vif_link() mt76_vif_link() indexes mvif->link[] without validating link_id, but callers pass mvif->deflink_id / msta->deflink_id, which hold IEEE80211_LINK_UNSPECIFIED (0xf) until the first link has been added. Since IEEE80211_MLD_MAX_NUM_LINKS is 15, that reads one element past the end of the array, aliasing mt76_vif_data.offchannel_link. Reachable via mt7996_set_tsf()/mt7996_offset_tsf() and mt7996_net_fill_forward_path(). Bounds check link_id and return NULL, matching mt7996_sta_link() and mt7996_sta_link_protected().
In the Linux kernel, the following vulnerability has been resolved: wifi: nl80211: clean up color-change beacon data on errors nl80211_color_change() calls nl80211_parse_beacon() for the beacon_next template, which can allocate params.beacon_next.mbssid_ies and .rnr_ies. A parsing failure returned directly instead of using the out: cleanup, leaking any allocations completed before the error. Allocate the nested attribute table before parsing beacon_next. Its allocation failure can then return before beacon data exists, while a later parsing failure uses out: to release the parsed data.
In the Linux kernel, the following vulnerability has been resolved: wifi: brcmfmac: fix P2P action frame handling without device vif Some P2P action frame paths assume the P2P device vif is always available. That is not true when userspace sends non-P2P public action frames through the primary interface, or when action-frame abort runs after the P2P device vif has not been created. Fall back to the primary vif when aborting an action frame without a P2P device vif, and guard P2P device saved IE access before using it for peer channel search.
In the Linux kernel, the following vulnerability has been resolved: bpf: Fix mmap_lock deadlock on arena lock failure Reported by the Sashiko AI review. arena_vm_fault() returns VM_FAULT_RETRY when it can't take arena->spinlock, but it never took mmap_lock. The fault path assumes a VM_FAULT_RETRY handler already dropped mmap_lock and re-takes it on the retry, so mmap_lock gets taken twice and can deadlock: do_user_addr_fault() { fault = handle_mm_fault(...); // calls arena_vm_fault() if (fault & VM_FAULT_RETRY) goto retry; // re-locks mmap_lock mmap_read_unlock(mm); } Return VM_FAULT_SIGBUS instead, for two reasons: 1. We could keep VM_FAULT_RETRY, but then we'd have to drop the fault lock first and cap the retry ourselves, the way __folio_lock_or_retry() does. 2. A failed raw_res_spin_lock_irqsave() already means a possible deadlock was detected, so retrying just hits the same lock again. So returning VM_FAULT_RETRY here is overkill.
In the Linux kernel, the following vulnerability has been resolved: pinctrl: generic: free maps on pinctrl_generic_to_map() failure pinctrl_generic_to_map() parses DT configuration and allocates pinctrl maps via pinctrl_utils_reserve_map(). If subsequent steps (such as pinctrl_utils_add_map_mux(), pinctrl_generic_add_group(), pinconf_generic_parse_dt_config(), or pinctrl_utils_add_map_configs()) return an error, *maps may contain partially allocated map entries. Returning the error directly without freeing *maps leaks the allocated mapping memory across all drivers that rely on pinctrl_generic_to_map(). Fix this by calling pinctrl_utils_free_map() and resetting *maps, *num_maps, and *num_reserved_maps in the error path of pinctrl_generic_to_map().
In the Linux kernel, the following vulnerability has been resolved: serial: amba-pl011: keep console clock enabled for atomic writes pl011_console_write_atomic() runs from nbcon atomic context, where sleeping is not allowed. It calls clk_enable(), which takes the common-clk enable_lock. Under PREEMPT_RT that is a sleeping lock: clk_enable_lock() first tries spin_trylock_irqsave(), but on contention falls back to spin_lock_irqsave(). Therefore, an atomic-context printk on an RT kernel with a clk-backed pl011 can trip: BUG: sleeping function called from invalid context at spinlock_rt.c:48 __might_resched from rt_spin_lock rt_spin_lock from clk_enable_lock clk_enable_lock from clk_enable clk_enable from pl011_console_write_atomic ... from vprintk_emit This was found and reproduced on PREEMPT_RT. Arm32 and arm64 DT SoCs are affected; arm64 SBSA/ACPI has no clk, so clk_enable(NULL) short-circuits before the lock. In addition, write_atomic() may be invoked from NMI context and is documented to avoid locking. Removing clk_enable() from the callback also avoids a potentially unsafe NMI acquisition of the common-clock enable_lock. An nbcon atomic-capable console must be printable from any context, so the clock cannot be gated between writes. Enable the clock while the console is available for output: use clk_prepare_enable() in pl011_console_setup(), release it via clk_disable_unprepare() in the console .exit() callback, and drop the per-write clk_enable()/clk_disable() pairs from write_atomic() and write_thread(). When printk suspends consoles, drop the reference after uart_suspend_port() stops console access and restore it before uart_resume_port() -- but only if suspend actually marked the port suspended (a wake-capable tty stays running and must keep its clock), and keep it when console_suspend_enabled is false so no_console_suspend works. The active power cost of keeping the clock enabled is platform-dependent: none where the UART clock is a fixed always-on oscillator, real where it is a gateable clock branch, which then cannot be gated (nor possibly can its parent clocks) while the console is available for output. When serial core actually suspends the port, the reference is released so the clock provider can gate the clock tree.
In the Linux kernel, the following vulnerability has been resolved: serial: core: do fallible allocations before the console can be registered serial_core_add_one_port() allocates uport->tty_groups after uart_configure_port(), which may register the console. If the allocation fails, the driver unwinds the port while its console remains registered. The earlier uport->name allocation has a related failure path that leaves state->uart_port linked to a port being freed. Failslab reproduced a NULL dereference in PL011 console output and a KASAN use-after-free in i.MX console output after failed binds. Allocate the name and tty_groups before linking the port and configuring it. Reserve space for the optional driver attribute group because config_port() may populate uport->attr_group during configuration.
In the Linux kernel, the following vulnerability has been resolved: serial: core: clear freed pointers on uart_register_driver() failure uart_register_driver() leaves drv->state pointing to freed memory when tty_alloc_driver() fails. If tty_register_driver() fails, drv->tty_driver also retains a pointer after its reference is dropped. Drivers that use drv->state as an "already registered" flag can then skip registration on the next probe and pass the freed state to uart_add_one_port(). This issue was found with failslab on QEMU's raspi1ap board by failing registration and binding the PL011 port again. Clear both pointers on their failure paths, as uart_unregister_driver() already does.
In the Linux kernel, the following vulnerability has been resolved: tty: skip cdev_del() when no cdev is registered TTY device registration can fail before a cdev is allocated. Serial core keeps the port so setserial can still use it, and later removal passes the NULL cdev slot to cdev_del(), causing a NULL-pointer dereference. Only delete the cdev when the slot is not NULL.
In the Linux kernel, the following vulnerability has been resolved: tty: clear cdev pointer after cdev_add() failure tty_cdev_add() drops the cdev reference when cdev_add() fails, but leaves driver->cdevs[index] pointing to freed memory. tty_unregister_device() later passes that stale pointer to cdev_del(), causing a use-after-free. Clear the slot after dropping the reference.
In the Linux kernel, the following vulnerability has been resolved: dm-integrity: replace forgeable discard filler with a keyed sector marker The discard-block check in dm_integrity_rw_tag() treats a stored tag of all 0xf6 bytes (DISCARD_FILLER) as proof a block was discarded and skips HMAC verification. allow_discards is only accepted in dm-integrity's standalone mode. An attacker with raw write access to the backing device, but without the integrity key, can stamp any block with an all-0xf6 tag and have it served as authentic. Add a new "allow_discards_keyed" target argument that marks discarded blocks with a keyed checksum of (salt || sector) instead, computed by integrity_discard_checksum().
In the Linux kernel, the following vulnerability has been resolved: HID: asus: refactor the two workqueues and init sequence Multiple issues have been found within the hid-asus driver: - unchecked size in asus_raw_event() - unclean teardown of asus_probe on failure - possible use-after-free in asus_probe - multiple workqueue used for jobs where one was enough - sleeping calls in atomic context - packets of incorrect size being sent to the keyboard controller Join the two workqueues into one reusing the stopping mechanism of the brightness workqueue, use the joined workqueue to also move the asus_wmi_send_event() sleeping call away from atomic context and add a size check in asus_raw_event().
In the Linux kernel, the following vulnerability has been resolved: HID: logitech-hidpp: Fix FF device cleanup on init failure hidpp_ff_init() creates the input force-feedback device with input_ff_create(), then allocates the HID++ FF private data, effect ID array, and workqueue. If any of those allocations fail after input_ff_create() succeeds, the function returns an error without destroying the FF device. Add an unwind path that frees the private allocations made by hidpp_ff_init() and calls input_ff_destroy() for failures after input_ff_create() succeeds.
In the Linux kernel, the following vulnerability has been resolved: HID: steam: Reject short reads Steam Controller FEATURE reports encode the size of the message in the message itself. Previously we were trusting that the size reported matched the size we actually read, leading to a potential issue with short reads. Instead, we should actually verify the length of the read.
In the Linux kernel, the following vulnerability has been resolved: phonet: pep: do not write beyond optlen in getsockopt pep_getsockopt() clamps the reported length to the caller's buffer with min_t(), but then stores the value with put_user(val, (int __user *) optval), which always writes sizeof(int) bytes. A getsockopt() call with an optlen smaller than sizeof(int) thus reports the clamped length yet writes a full int, one to three bytes past the user buffer. Write the value with copy_to_user() bounded by len, so at most optlen bytes are copied, matching the length reported back to userspace.
In the Linux kernel, the following vulnerability has been resolved: ublk: validate auto buf reg before taking uring_cmd With UBLK_F_AUTO_BUF_REG, invalid sqe->addr can fail after ublk_fill_io_cmd() has set UBLK_IO_FLAG_ACTIVE. The uring_cmd is completed while the tag stays active, which can hang teardown. Split validation from buffer apply so the check has no side effects, then take the uring_cmd and store the already-validated buffer. Apply the same order in FETCH so io->buf is not written before __ublk_fetch() state checks.
In the Linux kernel, the following vulnerability has been resolved: rapidio: clear mport->net when rio_add_net() fails rio_alloc_net() stores the newly allocated rio_net in mport->net before rio_scan_alloc_net() registers the device. If rio_add_net() fails, rio_scan_alloc_net() drops the device reference with put_device(), which releases the rio_net through the device release callback. However, mport->net is left pointing at the freed object. A later mport unregister path can then dereference the dangling mport->net pointer and may try to free the same rio_net again. Clear mport->net in the rio_add_net() failure path, matching the cleanup done for the destID table allocation failure path.
In the Linux kernel, the following vulnerability has been resolved: fat: release buffer head after rebuilding parent fat_scan_logstart() leaves the matching directory entry's buffer head in sinfo.bh for the caller to release, just like fat_scan(). fat_rebuild_parent() uses the directory entry to rebuild the parent inode for the nostale_ro NFS export path, but does not release sinfo.bh after a successful scan. Release it once fat_build_inode() has consumed the directory entry data.
In the Linux kernel, the following vulnerability has been resolved: bpf, cgroup: Fix invalid storage access after __cgroup_bpf_attach failed A potential invalid storage access issue can occur after replacing a cgroup bpf prog. This occurs in the following scenario: 1. prog1 with storage is attached to a cgroup in multi-attach mode. 2. prog1 is replaced with prog2 using BPF_F_REPLACE in multi-attach mode, but fails midway (e.g. in bpf_trampoline_link_cgroup_shim or update_effective_progs). 3. A new prog3 is attached to the cgroup in multi-attach mode. The reason is that __cgroup_bpf_attach overwrites pl->storage with the new storage prior to attachment completion. When attachment fails midway, the cleanup path calls bpf_cgroup_storages_free(new_storage) to free the newly allocated storage, but fails to restore pl->storage back to old_storage. Consequently, the still-active prog1 holds invalid or dangling storage pointers, leading to an invalid memory access when prog1 executes and calls bpf_get_local_storage. Additionally, original pl->flags and cgrp->bpf.flags[atype] are left unrestored. Fix this by saving old_pl_flags, old_storage, and old_flags prior to the update, and properly restoring all of them in the cleanup path on error.
In the Linux kernel, the following vulnerability has been resolved: thermal: hwmon: Remove hwmon class device along with its parent The current code creates one hwmon device per thermal zone type and that device is registered under the first thermal zone of the given type. That turns out to be problematic when the thermal zone holding the hwmon device is removed. For example, say that there are two ACPI thermal zones on a system /sys/devices/virtual/thermal/thermal_zone0/ /sys/devices/virtual/thermal/thermal_zone1/ The current code registers a hwmon class device for thermal_zone0 only: /sys/devices/virtual/thermal/thermal_zone0/hwmon0/ because the type is "acpitz" for both of them, but it adds a sysfs attribute that belongs to thermal_zone1 under it: /sys/devices/virtual/thermal/thermal_zone0/hwmon0/temp2_input There is also /sys/devices/virtual/thermal/thermal_zone0/hwmon0/temp1_input which belongs to thermal_zone0. When thermal_zone0 is removed, say because the ACPI thermal driver is unbound from the underlying platform device, thermal_remove_hwmon_sysfs() skips the removal of hwmon0 because of the temp2_input attribute belonging to thermal_zone1 which effectively prevents thermal_zone0 removal from making progress. Address this by making thermal_remove_hwmon_sysfs() remove the entire hwmon class device interface for the given thermal zone type when the thermal zone device holding it is removed. To prevent races with thermal_add_hwmon_sysfs() that may interfere with this, carry out the entire addition and removal of hwmon sysfs interfaces for thermal zones under thermal_hwmon_list_lock. Also adjust the layout of the labels in thermal_add_hwmon_sysfs() to the current kernel coding style to align with the new "unlock" label.
In the Linux kernel, the following vulnerability has been resolved: RDMA/srp: fix heap information leak on a truncated SRP_CRED_REQ srp_recv_done() passes wc->byte_len to srp_process_rsp(). It passes nothing to srp_process_cred_req() and srp_process_aer_req(), which read fixed-size fields from the receive buffer without checking that those fields were received. The buffer size is max_ti_iu_len, which comes from the login response and is not validated. A target that advertises 8 and then sends an 8-byte SRP_CRED_REQ makes the initiator read req->tag from beyond the end of the buffer. req->tag is copied into the SRP_CRED_RSP and sent back, so those bytes reach the target. SRP_AER_REQ behaves the same way and also reads req->lun. The leak is 8 bytes per response. max_ti_iu_len also decides which slab cache the buffer comes from. With 8 the buffer is a kmalloc-8 object and the read is entirely outside it: BUG: KASAN: slab-out-of-bounds in srp_recv_done+0x172b/0x1aa0 Read of size 8 at addr ffff888104714da8 by task kworker/u8:3/50 which belongs to the cache kmalloc-8 of size 8 The buggy address is located 0 bytes to the right of allocated 8-byte region [ffff888104714da0, ffff888104714da8) Without KASAN the returned bytes are whatever is next in the slab. One run returned ".strtab". rsp->data[3] in srp_process_rsp() has the same problem: only resp_data_len is checked before it is read. Drop a request that is shorter than the structure being parsed, and check byte_len before the tsk_mgmt read.
In the Linux kernel, the following vulnerability has been resolved: ARM: 9483/1: select HAVE_POSIX_CPU_TIMERS_TASK_WORK Commit c6e61c06d606 ("ARM: 9463/1: Allow to enable RT") enabled PREEMPT_RT on ARM but did not select HAVE_POSIX_CPU_TIMERS_TASK_WORK. This leaves CONFIG_POSIX_CPU_TIMERS_TASK_WORK disabled, so CPU timers expire in hard IRQ context. On PREEMPT_RT this makes run_posix_cpu_timers() take the sleeping sighand->siglock: BUG: sleeping function called from invalid context at spinlock_rt.c:48 rt_spin_lock from lock_task_sighand lock_task_sighand from run_posix_cpu_timers run_posix_cpu_timers from update_process_times ARM handles TIF_NOTIFY_RESUME on all return-to-user paths, including v7-M. ARM32 KVM host support was removed by commit 541ad0150ca4 ("arm: Remove 32bit KVM host support"), so the select need not be conditional on KVM. Select it to defer POSIX CPU timer expiry to task context. Reproduced with setrlimit(RLIMIT_CPU, ...) and a busy loop. The same path is used by setitimer(ITIMER_PROF or ITIMER_VIRTUAL) and POSIX CPU timers created with timer_create().
In the Linux kernel, the following vulnerability has been resolved: ARM: 9484/1: enable interrupts when unhandled user faults are triggered PREEMPT_RT requires interrupts to be enabled when sending signals. When do_DataAbort()/do_PrefetchAbort() triggers unhandled user faults, that is `inf->fn()` return a non-zero value, and the interrupts are not enabled within the hook function, force_sig_fault() will be called with interrupts disabled. This can be triggered by user programs executing the bkpt instruction, with kernel config CONFIG_PERF_EVENTS=n. Enable interrupts in do_DataAbort()/do_PrefetchAbort() when unhandled user faults are triggered to fix the issue.
In the Linux kernel, the following vulnerability has been resolved: ARM: 9485/1: mm: acquire mmap write lock around show_pte() for user faults When CONFIG_DEBUG_USER=y, and cmdline "user_debug=31" is set, a user fault may trigger show_pte() without any lock. If another thread in the same process concurrently calls munmap(), the page table pages may be freed while show_pte() is still traversing them, causing a use-after-free in show_pte(). If CONFIG_ARM_LPAE=y, this may cause a kernel panic if the pages table of PMD are freed when show_pte() is running. Acquire mmap_write_lock() around show_pte() for user faults to fix the contention. For user faults, additionally restrict that show_pte() is called only when the addr is a user-space address (addr < TASK_SIZE). This is because the lock of tsk->mm only protects the virtual memory of user address space, furthermore, dumping the page tables of a kernel-space address for user faults is unnecessary and may have security implications. Keep everything unchanged for kernel faults, because the kernel is already in the "oops" state, acquiring a lock may risk a deadlock.
In the Linux kernel, the following vulnerability has been resolved: ocfs2: synchronize heartbeat callbacks with o2net teardown Patch series "ocfs2: harden heartbeat teardown races". This series fixes two OCFS2 heartbeat/o2net teardown races found by KASAN. This patch (of 2): Heartbeat callbacks stay registered while configfs local-node teardown enters o2net_stop_listening(). A node-down event can still run through o2net_disconnect_node() and o2net_set_nn_state() while teardown is destroying o2net_wq, so the later queue/flush operations can hit a dead workqueue. KASAN has caught this as a slab-use-after-free in __queue_work() with the call chain: KASAN slab-use-after-free in __queue_work+0x56/0xa90 Read of size 4 Call trace: dump_stack_lvl+0x66/0xa0 print_report+0xce/0x630 __queue_work+0x56/0xa90 srso_alias_return_thunk+0x5/0xfbef5 __virt_addr_valid+0x19f/0x330 kasan_report+0xe0/0x110 __queue_delayed_work+0x58/0x1e0 queue_delayed_work_on+0xb4/0xc0 o2net_set_nn_state+0x467/0x840 o2net_disconnect_node+0x7b/0xe0 o2net_hb_node_down_cb+0x54/0x60 o2hb_run_event_list+0x236/0x2d0 o2hb_check_slot+0xad4/0xbc0 lock_release+0xc8/0x290 o2hb_check_slot+0x9ea/0xbc0 trace_hardirqs_on+0x18/0x130 o2hb_do_disk_heartbeat+0x646/0xb30 (fs/ocfs2/cluster/heartbeat.c:1079) __lock_acquire+0x466/0x2260 lockdep_hardirqs_on_prepare+0xea/0x1a0 ktime_get_with_offset+0xe9/0x230 o2hb_thread+0x14e/0x770 kthread+0x1ad/0x1f0 ret_from_fork+0x3c9/0x540 __switch_to+0x2e9/0x730 ret_from_fork_asm+0x1a/0x30 Allocated by task stack: kasan_save_stack+0x33/0x60 kasan_save_track+0x14/0x30 __kasan_kmalloc+0xaa/0xb0 __kmalloc_noprof+0x292/0x760 __alloc_workqueue+0x736/0xc60 alloc_workqueue_noprof+0xb1/0x110 o2net_start_listening+0xe5/0x430 o2nm_node_local_store+0x184/0x310 configfs_write_iter+0x18a/0x210 vfs_write+0x469/0x810 ksys_write+0xd2/0x170 do_syscall_64+0x115/0x6a0 (arch/x86/entry/syscall_64.c:87) entry_SYSCALL_64_after_hwframe+0x77/0x7f Freed by task stack: kasan_save_stack+0x33/0x60 kasan_save_track+0x14/0x30 kasan_save_free_info+0x3b/0x60 __kasan_slab_free+0x5f/0x80 kfree+0x313/0x590 rcu_core+0x4f4/0x1320 handle_softirqs+0x156/0x660 queue_delayed_work_on o2net_set_nn_state o2net_disconnect_node o2net_hb_node_down_cb o2hb_run_event_list Keep heartbeat callbacks registered so quorum state still tracks node state, but stop them from driving o2net reconnect/disconnect work once local teardown starts. Mark the transport offline before destroying o2net_wq, wait for any in-flight heartbeat callback to finish, and delay bring-up replay until the new local node is published through o2nm_this_node(). The replay also has to stay serialized with heartbeat callback delivery. Otherwise a live-node snapshot can be copied, a real hb_down callback can install -ENOTCONN for a peer, and the stale replay can call o2net_hb_node_up() for that same peer and queue reconnect work even though heartbeat is already down. The buggy scenario involves two paths, with each column showing the order within that path: local-node teardown: heartbeat node-down callback: 1. configfs local-off enters 1. o2hb_run_event_list() invokes o2net_stop_listening(). o2net_hb_node_down_cb(). 2. teardown heads for 2. the callback reaches destroy_workqueue(o2net_wq). o2net_disconnect_node() and o2net_set_nn_state(). 3. teardown destroys and NULLs 3. the callback flushes or queues o2net_wq. work through o2net_wq.
In the Linux kernel, the following vulnerability has been resolved: bpf: Clear buf on error in __bpf_get_task_stack Both bpf_get_task_stack and bpf_get_task_stack_sleepable helpers that use __bpf_get_task_stack have buf defined as ARG_PTR_TO_UNINIT_MEM argument and we should initialize the buf on every return path. Adding missing buf memset for __bpf_get_task_stack fail paths. This provides deterministic buffer contents, which is useful when the buffer is used directly as a map key.
In the Linux kernel, the following vulnerability has been resolved: drm/sun4i: tcon: Drop TCON TOP device reference of_find_device_by_node() takes a device reference. Drop it after mux configuration succeeds.
In the Linux kernel, the following vulnerability has been resolved: drm/sun4i: crtc: Propagate layer initialization error sun4i_crtc_init() returns plain NULL when layer initialization fails, while all its other error paths return an error pointer. The only caller, sun4i_tcon_bind(), checks the result with IS_ERR() and happily continues with tcon->crtc set to NULL. sun4i_rgb_init() and sun4i_lvds_init() then dereference it in drm_crtc_mask(), which oopses. Return the error pointer instead.
In the Linux kernel, the following vulnerability has been resolved: cpufreq: imx6q: fix devres accumulation across driver rebind imx6_soc_volt is allocated with devm_kcalloc(cpu_dev, ...), where cpu_dev is the CPU device from get_cpu_device(0). That device is never unbound, so its devres list is never released, and imx6q_cpufreq_remove() does not free the array either. Every probe therefore adds an allocation that stays for the lifetime of the system. Allocate against the platform device instead. Its devres is released when the driver is unbound, which is exactly the lifetime the array wants: imx6q_set_target() reads it, and nothing may reach that after cpufreq_unregister_driver(). That makes the array actually go away on unbind, so also clear the file-scope pointer in remove and on the failed-probe path, rather than leave it pointing at memory devres is about to release. Tested by rebinding the driver on qemu's mcimx6ul-evk.
In the Linux kernel, the following vulnerability has been resolved: arm64: hibernate: Restore DAIF state on error Sashiko AI has reported that if swsusp_mte_save_tags() for some reason fails we return from swsusp_arch_suspend() with DAIF being masked - that is not what we'd expect. Restore the saved DAIF state before returning from the error path.
In the Linux kernel, the following vulnerability has been resolved: phy: sunplus: fix error handling in sp_uphy_init() Fix the error paths of sp_uphy_init() to undo exactly what each stage did: return directly if clk_prepare_enable() fails, release only the clock if reset_control_deassert() fails, and jump to err_reset if update_disc_vol() fails so the clock and reset are not leaked.
In the Linux kernel, the following vulnerability has been resolved: scsi: qla2xxx: Remove redundant VPD flash read in sysfs read path qla2x00_sysfs_read_vpd() called ha->isp_ops->read_optrom() a second time after releasing optrom_mutex. The repeated read is redundant and, unlike the first, runs without optrom_mutex held, exposing flash access to concurrent optrom operations. Drop the duplicate call.
In the Linux kernel, the following vulnerability has been resolved: firmware_loader: do not queue completed sysfs fallback requests fw_load_sysfs_fallback() calls device_add() before adding the fw_priv to pending_fw_head. device_add() publishes the fallback loading interface, so a userspace helper which discovers the device by scanning sysfs can write 0 to the loading attribute and complete the request before it is queued as pending. In that interleaving firmware_loading_store() calls fw_state_done() while pending_list still points to itself, so it cannot remove an entry from pending_fw_head. The subsequent unconditional list_add() then queues an already-completed fw_priv. Once the request is released, pending_fw_head can retain a pointer to freed memory and the next fallback request can fault while validating the list. Only in-flight fallback requests need suspend or reboot abort handling. If the request is already DONE after device_add(), return success from the fallback path without sending another uevent, waiting again, or queueing it as pending. This preserves the invariant that pending_fw_head contains only active fallback requests.
In the Linux kernel, the following vulnerability has been resolved: md/raid5: round bitmap stripes with sector division raid5_bitmap_sector_map() aligns the array range to full RAID5 stripe widths before converting it to component sectors. That width is chunk_sectors multiplied by the number of data disks, and it is not always a power of two. Reproduce with a 4-disk RAID5, 1024-sector chunks, and three data disks. The full-stripe width is 3072 sectors. For a one-sector write at array sector 3072, correct rounding gives array range [3072, 6144), which maps to component range [1024, 2048). The old round_down()/round_up() logic instead gives [1024, 4096), which maps to [0, 1024). Use sector_div() based arithmetic so the rounded range is aligned to the actual RAID5 stripe width. The deterministic mapper test now reports the fixed component range as [1024, 2048), while the old mask-based range was [0, 1024).
In the Linux kernel, the following vulnerability has been resolved: md: wait for behind writes before destroying bitmap __md_stop() destroyed the bitmap before calling mddev_detach(). That made mddev_detach() skip bitmap_ops->wait_behind_writes(), because the bitmap was already disconnected from mddev. This was still safe for the legacy bitmap because bitmap_destroy() waits for behind writes itself. llbitmap keeps that wait in its ->wait_behind_writes() operation instead, while ->destroy() tears down the llbitmap storage. With the old ordering, RAID1 behind-write completions could still run after llbitmap storage had been freed. Call mddev_detach() before md_bitmap_destroy() so the common detach path can wait for behind writes while the bitmap is still alive. Only destroy the bitmap after those users are gone.
In the Linux kernel, the following vulnerability has been resolved: md/md-llbitmap: prevent create failure bitmap UAF llbitmap_create() publishes mddev->bitmap before reading the bitmap superblock. This is needed because llbitmap_read_sb() can initialize a new bitmap and flush it through helpers that use mddev->bitmap. If llbitmap_read_sb() fails, the old cleanup dropped bitmap_info.mutex and freed llbitmap before clearing mddev->bitmap. Readers such as /proc/mdstat rely on bitmap_info.mutex to keep the bitmap pointer stable while collecting bitmap stats, so they could observe the stale pointer after the failed create path released the mutex. Clear mddev->bitmap while still holding bitmap_info.mutex, then free the failed llbitmap after dropping the mutex. This makes mutex-protected readers see either a live bitmap or no bitmap.
In the Linux kernel, the following vulnerability has been resolved: md/md-llbitmap: stop daemon timer rearm on destroy llbitmap_destroy() deletes pending_timer before flushing md_llbitmap_io_wq. However, daemon_work can still be queued or running after the timer has been deleted, and the daemon path can arm pending_timer again when it finds dirty chunks that are not ready to flush yet. If that happens during teardown, pending_timer can remain armed after llbitmap is freed and later dereference freed memory. Add a BITMAP_SHUTDOWN bit to llbitmap->flags, set it before deleting the timer, and make the timer and daemon paths stop queueing or rearming work once teardown starts. Cancel daemon_work before flushing the shared workqueue so no already queued daemon instance can race with the free. Use timer_shutdown_sync() so a daemon instance that passed the shutdown check before teardown cannot rearm the timer afterward. BITMAP_SHUTDOWN is a runtime-only state. Mask it out when reading and updating the llbitmap superblock so the shutdown state is never loaded from disk or persisted to disk.
In the Linux kernel, the following vulnerability has been resolved: md/raid1: don't set array_frozen in raid1_takeover() raid1_takeover() sets conf->array_frozen = 1 on the newly-allocated r1conf and nothing ever clears it, so every I/O to the array stalls permanently once _wait_barrier() sees it stuck at 1. This used to be harmless: level_store() called mddev_resume() right after pers->run(), which called raid1_quiesce(mddev, 0) and cleared array_frozen back to 0 regardless of what raid1_takeover() set. Commit b39f35ebe86d ("md: don't quiesce in mddev_suspend()") removed that quiesce(mddev, 0) call, so the pre-set now sticks. setup_conf() already zero-initializes the new r1conf via kzalloc, so just don't set array_frozen here. Same class of bug as commit 892da88d1cd9 ("md/raid10: fix a 'conf->barrier' leakage in raid10_takeover()"), also triggered by b39f35ebe86d.
In the Linux kernel, the following vulnerability has been resolved: coresight: etm4x: fix underflow for usage of (nrseqstate - 1) According to IHI006H Embedded Trace Macrocell Architecture Specification[0], TRCSEQEVR<n> is implemented only when TRCIDR5.NUMSEQSTATE is 0b100, in which case n ranges from 0 to 2; otherwise, TRCIDR5.NUMSEQSTATE is 0b000. IOW, the number of usage in the initialisation or setting TRCSEQEVR<n> with drvdata->nrseqstate - 1 in the loop could make underflow issue when TRCIDR5.NUMSEQSTATE is 0b000. Therefore, introduce nr_seq_ctrls field and untie it from nrseqstate. As part of this introduce ETM_MAX_SEQ_TRANSITIONS macro and apply nr_seq_ctrls and above macro to TRCSEQEVR<n> relevant fields setup.
In the Linux kernel, the following vulnerability has been resolved: coresight: etm4x: missing cscfg_csdev_disable_active_config() in perf enable In the perf enable path, there are missing cases where cscfg_csdev_disable_active_config() is not called: - Branch broadcast is selected but not supported by the hardware - etm4_enable_hw() fails This can lead to a leak of config_desc->active_cnt. Fix this by properly calling cscfg_csdev_disable_active_config() in these error paths.
In the Linux kernel, the following vulnerability has been resolved: perf: arm_pmuv3: Zero initialize hw_id branch stack field PERF_SAMPLE_BRANCH_HW_INDEX is supported by BRBE so hw_id is passed to userspace, but it's never set by the BRBE driver. Zero initialize it as it should be according to the docs: * For the architectures whose raw branch records are * already stored in age order, the hw_idx should be 0. It's probably too risky to remove PERF_SAMPLE_BRANCH_HW_INDEX from BRBE now in case anyone is setting it and reading the value, but zero initializing the whole struct also protects against the same issue with new fields that are added in the future.
In the Linux kernel, the following vulnerability has been resolved: bpf: Reject load-acquire from pointers requiring fault protection A BPF_LOAD_ACQ is not rewritten to a BPF_PROBE_MEM load by the verifier, unlike a regular BPF_LDX, so the JIT emits a plain load with no exception table entry and a fault panics the kernel instead of being handled. Reject the source pointer types that a BPF_LDX would have had that fault protection applied to, i.e. the ones bpf_convert_ctx_accesses() turns into BPF_PROBE_MEM: a bare PTR_TO_BTF_ID, PTR_TO_BTF_ID | PTR_UNTRUSTED, PTR_TO_BTF_ID | MEM_ALLOC | PTR_UNTRUSTED and PTR_TO_MEM | MEM_RDONLY | PTR_UNTRUSTED. This is reachable e.g. by loading ->mm out of a trusted task_struct yields an untrusted pointer to mm_struct, and it is NULL for a kernel thread: [...] SEC("tp_btf/sched_switch") int BPF_PROG(demo, bool preempt, struct task_struct *prev, struct task_struct *next) { struct mm_struct *mm = next->mm; /* untrusted */ out_ldx = (__u64)mm->pgd; /* BPF_LDX */ out_acq = load_acquire(&mm->pgd); /* BPF_LOAD_ACQ */ return 0; } [...] Both dereference the same pointer, but only the BPF_LDX is protected (x86-64 JIT, jump targets shown prog-relative): [...] ; out_ldx = (__u64)mm->pgd; 17: movq $-10485760, %r10 1e: movq %rsi, %r11 21: addq $184, %r11 28: subq %r10, %r11 2b: movabsq $140737498841088, %r10 35: cmpq %r10, %r11 38: ja 0x3e <-- kernel addr? 3a: xorl %edi, %edi <-- no: dst = 0, skip the load 3c: jmp 0x45 3e: movq 184(%rsi), %rdi <-- yes: load + extable entry [...] ; load_acquire(&mm->pgd) 53: movq %rsi, %rdi 56: movq 184(%rdi), %rax <-- no check, no extable entry [...] Note that BPF_PROBE_MEM is not visible in a bpftool xlated dump, as bpf_insn_prepare_dump() rewrites it back to BPF_MEM. A PTR_TRUSTED pointer is deliberately not on the list. Such a load is not converted either, but it does not need to be, since the pointer is guaranteed live, so load-acquire from it stays allowed. The check is gated on BPF_LOAD_ACQ so that atomic RMW and store-release error messages are unchanged; writes (RMW / store-release) to such pointers are already rejected elsewhere, so only load-acquire needs this.
In the Linux kernel, the following vulnerability has been resolved: scsi: sd: Fix special_vec mempool leak when scsi_alloc_sgtables() fails sd_set_special_bvec() allocates a special payload page for UNMAP and WRITE SAME commands. If scsi_alloc_sgtables() fails afterward in sd_setup_unmap_cmnd() or sd_setup_write_same{10,16}_cmnd(), the SCSI midlayer does not call uninit_command() because RQF_DONTPREP is not set yet, leaking the page. Call sd_uninit_command() on error, and clear RQF_SPECIAL_PAYLOAD after freeing the page.
In the Linux kernel, the following vulnerability has been resolved: btrfs: zoned: don't force read-only on transient -EAGAIN from reloc merge On a zoned FS, btrfs_delayed_refs_rsv_refill() returns -EAGAIN whenever the over-committed metadata plus the zone_unusable bytes exceeds the usable size in a metadata block-group to avoid heavy over-commit of metadata and early ENOSPC in one transaction. If this happens while doing reclaim, the transaction is getting aborted. Treat -EAGAIN as a soft, retryable condition in case of block-group reclaim.
In the Linux kernel, the following vulnerability has been resolved: btrfs: defrag: fix deadlock between defrag and delalloc space reservation While running fsstress with autodefrag and flushoncommit, hit a deadlock due to the fact that defrag reserves delalloc space while it's holding dirty and locked folios, besides the extent range lock. The stack traces are the following: [958.624] task:kworker/u50:3 state:D stack:0 pid:20365 tgid:20365 ppid:2 task_flags:0x4208060 flags:0x00080000 [958.626] Workqueue: events_unbound btrfs_async_reclaim_metadata_space [btrfs] [958.627] Call Trace: [958.628] <TASK> [958.628] __schedule+0x4be/0x10f0 [958.629] ? preempt_count_add+0x69/0xa0 [958.630] schedule+0x26/0xd0 [958.631] wait_current_trans+0x102/0x160 [btrfs] [958.632] ? __pfx_autoremove_wake_function+0x10/0x10 [958.633] start_transaction+0x374/0x900 [btrfs] [958.634] btrfs_commit_current_transaction+0x1d/0x70 [btrfs] [958.635] flush_space+0xca/0x5e0 [btrfs] [958.636] ? _raw_spin_unlock+0x15/0x30 [958.637] ? btrfs_reduce_alloc_profile+0x8c/0x190 [btrfs] [958.639] ? _raw_spin_unlock+0x15/0x30 [958.640] ? calc_available_free_space.isra.0+0x6f/0x110 [btrfs] [958.641] do_async_reclaim_metadata_space+0x84/0x190 [btrfs] [958.642] btrfs_async_reclaim_metadata_space+0x64/0x80 [btrfs] [958.644] process_one_work+0x19d/0x3a0 [958.644] worker_thread+0x1c4/0x330 [958.645] ? __pfx_worker_thread+0x10/0x10 [958.646] kthread+0xfc/0x130 [958.647] ? __pfx_kthread+0x10/0x10 [958.648] ret_from_fork+0x1f7/0x2c0 [958.648] ? __pfx_kthread+0x10/0x10 [958.649] ret_from_fork_asm+0x1a/0x30 [958.650] </TASK> [958.651] task:kworker/u49:7 state:D stack:0 pid:52990 tgid:52990 ppid:2 task_flags:0x4208060 flags:0x00080000 [958.653] Workqueue: writeback wb_workfn (flush-btrfs-334) [958.655] Call Trace: [958.655] <TASK> [958.656] __schedule+0x4be/0x10f0 [958.657] ? __blk_flush_plug+0xe9/0x140 [958.658] schedule+0x26/0xd0 [958.658] io_schedule+0x42/0x70 [958.659] folio_wait_bit_common+0x12b/0x330 [958.660] ? folio_wait_bit_common+0x100/0x330 [958.662] ? __pfx_wake_page_function+0x10/0x10 [958.663] extent_write_cache_pages+0x599/0x830 [btrfs] [958.664] ? acpi_fwnode_get_reference_args+0x1fa/0x270 [958.665] btrfs_writepages+0x77/0x130 [btrfs] [958.666] ? __pfx_end_bbio_data_write+0x10/0x10 [btrfs] [958.667] do_writepages+0xc6/0x160 [958.668] __writeback_single_inode+0x42/0x310 [958.669] writeback_sb_inodes+0x231/0x570 [958.670] wb_writeback+0x8a/0x340 [958.671] wb_workfn+0xbf/0x450 [958.672] ? finish_task_switch.isra.0+0xc1/0x350 [958.673] process_one_work+0x19d/0x3a0 [958.673] worker_thread+0x1c4/0x330 [958.674] ? __pfx_worker_thread+0x10/0x10 [958.675] kthread+0xfc/0x130 [958.676] ? __pfx_kthread+0x10/0x10 [958.676] ret_from_fork+0x1f7/0x2c0 [958.677] ? __pfx_kthread+0x10/0x10 [958.678] ret_from_fork_asm+0x1a/0x30 [958.679] </TASK> [958.679] task:btrfs-cleaner state:D stack:0 pid:296750 tgid:296750 ppid:2 task_flags:0x208040 flags:0x00080000 [958.681] Call Trace: [958.682] <TASK> [958.682] __schedule+0x4be/0x10f0 [958.683] schedule+0x26/0xd0 [958.684] handle_reserve_ticket+0x1b9/0x2c0 [btrfs] [958.685] ? __pfx_autoremove_wake_function+0x10/0x10 [958.686] reserve_bytes+0x283/0x4c0 [btrfs] [958.687] btrfs_reserve_metadata_bytes+0x18/0xb0 [btrfs] [958.688] btrfs_delalloc_reserve_metadata+0x121/0x320 [btrfs] [958.690] btrfs_delalloc_reserve_space+0x46/0xb0 [btrfs] [958.691] btrfs_defrag_file+0x903/0x1110 [btrfs] [958.692] btrfs_run_defrag_inodes+0x334/0x430 [btrfs] [958.694] cleaner_kthread+0x97/0x1c0 [btrfs] [958.694] ? __pfx_cleaner_kthread+0x10/0x10 [btrfs] [958.696] kthread+0xfc/0x130 [958.696] ? __pfx_kthread+0x10/0x10 [958.697] ret_ ---truncated---
In the Linux kernel, the following vulnerability has been resolved: btrfs: check if root is readonly when setting posix acl For a filesystem which has btrfs read-only property set to true, all write operations including acl and xattr should be denied. However, acl can still be set even if btrfs ro property is true. This happens because no function on the set_acl code path checks the root is readonly or not. It was checked in btrfs_setxattr_trans() but got removed in commit 353c2ea735e4 ("btrfs: remove redundant readonly root check in btrfs_setxattr_trans") That commit didn't check if all the callers properly check the root's read-only flag. A previous fix is commit b51111271b03 ("btrfs: check if root is readonly while setting security xattr"). Always check if the root is read-only before performing the set acl operation.
In the Linux kernel, the following vulnerability has been resolved: btrfs: retry verity reads for not-uptodate Merkle folios btrfs_read_merkle_tree_page() can find a folio in the mapping that is not uptodate. After taking the folio lock, the current code treats that state as a read error and returns -EIO. That can make a previous transient read failure sticky. If the failed read left a not-uptodate folio in the mapping, later callers find that folio and fail instead of retrying the read. Keep the existing page-cache insertion and locking order, but retry the Merkle item read when a not-uptodate folio is found in the mapping. Also unlock the folio when read_key_bytes() fails so that a later caller can lock it and retry the read.
In the Linux kernel, the following vulnerability has been resolved: btrfs: zoned: flush active metadata block group at btree_writepages() start btree_writepages() writes the btree inode's dirty metadata in ascending logical address order. On a zoned filesystem only one metadata and one system block group is active for writing at a time, and check_bg_is_active() (via btrfs_check_meta_write_pointer()) pivots the active block group as writeback moves from one block group to the next. If the active block group sits at a higher logical address than another block group that also holds dirty metadata, the ascending walk reaches the lower one first and, to write it, has to finish the active block group and activate the lower one. It cannot finish a block group that still has unsent IO, and during WB_SYNC_ALL && !for_sync (commit) writeback it deliberately refuses to wait for that IO under fs_info->zoned_meta_io_lock, as that can deadlock. The pivot thus cannot issue the submission itself either, so it gives up: btrfs_check_meta_write_pointer() returns -EAGAIN, which btrfs_write_and_wait_transaction() treats as fatal and aborts the transaction, forcing the filesystem read-only. This happens intermittently under metadata-heavy relocation (e.g. fstests btrfs/187). Flush the active metadata and system block groups at the start of btree_writepages(), under the fs_info->zoned_meta_io_lock it already holds, so they have no unsent IO left and the later pivot can finish them and make forward progress.
In the Linux kernel, the following vulnerability has been resolved: btrfs: qgroup: fix a wrong length calculation in qgroup_free_reserved_data() In that function, we round down the start position and round up the ending position. But during the calculation of @len, we use "round_up(start + len, sectorsize)", which is the rounded up end position, not the rounded up length. Which results a much larger length, and later we are still using "start + len", which is completely incorrect. Fix it by declaring a local @aligned_start and @aligned_len and use them instead.
In the Linux kernel, the following vulnerability has been resolved: pinctrl: airoha: add missed IRQ resource helpers Without hooking .irq_request_resources, gpiolib cannot set GPIOD_FLAG_USED_AS_IRQ. This breaks pin direction locking and can allow userspace or another driver to reconfigure an active IRQ pin as an output
In the Linux kernel, the following vulnerability has been resolved: Bluetooth: virtio_bt: avoid OOB read of build info string The virtbt_setup_zephyr() sends the Zephyr vendor command 0xfc08 (Read Build Information) and hands the response to bt_dev_info() and hci_set_fw_info() as a "%s" string starting at skb->data + 1, without checking the length. A backend that answers with status only leaves that pointer past the end of the received data, so the walk reads adjacent slab memory until it meets a NUL. Those bytes reach the kernel log and the firmware-info debugfs file. To fix this, print the string with a bounded "%.*s" limited to skb->len - 1. A short or unterminated response then prints as much as arrived instead of failing setup. This mirrors commit dd068ef04412 ("Bluetooth: bpa10x: avoid OOB read of revision string in bpa10x_setup()"), which fixed the identical pattern.
In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hci_sync: free the advertising instance on the failure and cancel paths adv_timeout_expire() hands a kmalloc()ed instance byte to hci_cmd_sync_queue() with a NULL destroy callback, and only adv_timeout_expire_sync() frees it. That leaks on two paths: - the return value is not checked, and hci_cmd_sync_queue() does not take ownership when it fails (-ENETDOWN, -ENODEV, -ENOMEM); - a cancelled entry is not released, as _hci_cmd_sync_cancel_entry() does not free entry->data when there is no destroy callback. hci_cmd_sync_clear() cancels every pending entry when the controller is unregistered. Free the buffer from a destroy callback, and in the caller when the entry could not be queued at all.
In the Linux kernel, the following vulnerability has been resolved: Bluetooth: MGMT: free the mesh send cancel command when it is cancelled mesh_send_cancel() queues the pending command with a NULL destroy callback, so it is only freed if send_cancel() runs. A cancelled entry is leaked, as _hci_cmd_sync_cancel_entry() does not release entry->data when there is no destroy callback, and hci_cmd_sync_clear() cancels every pending entry when the controller is unregistered. Nothing else reclaims it either: mgmt_pending_new() does not put the command on hdev->mgmt_pending. The leak also pins the socket reference taken by mgmt_pending_new(), so the mgmt socket is never released. Free the command from a destroy callback.
In the Linux kernel, the following vulnerability has been resolved: Bluetooth: MGMT: free the HCI command when it is cancelled mgmt_hci_cmd_sync() queues the pending command with a NULL destroy callback, so it is only freed if send_hci_cmd_sync() runs. A cancelled entry is leaked, as _hci_cmd_sync_cancel_entry() does not release entry->data when there is no destroy callback, and hci_cmd_sync_clear() cancels every pending entry when the controller is unregistered. Nothing else reclaims it either: mgmt_pending_new() does not put the command on hdev->mgmt_pending. The leak also pins the socket reference taken by mgmt_pending_new(), so the mgmt socket is never released. Free the command from a destroy callback. The now-empty done label is replaced by a direct return.
In the Linux kernel, the following vulnerability has been resolved: bpf, cgroup: Fix storage null-ptr-deref after replacing prog Syzkaller reported a storage null-ptr-deref issue after replacing prog. This occurs in the following scenario: 1. prog A, an empty prog, is attached to a cgrp. 2. prog B uses BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE and calls the bpf_get_local_storage helper. 3. link_update is called to replace prog A with prog B. The reason is that __cgroup_bpf_replace fails to alloc and assign the required cgrp storage for the incoming replacement prog. Consequently, the new prog inherits an uninit storage, leading to null-ptr-deref panic when kick the new prog. Fix this by rejecting a link update if new_prog's cgroup storage is incompatible with link->prog.
In the Linux kernel, the following vulnerability has been resolved: iio: light: gp2ap002: Fix unbalanced runtime PM on repeated event writes The IIO core does not filter duplicate writes to the event enable attribute, so writing the same value twice invokes write_event_config() twice. Enabling twice leaks a runtime PM reference, preventing the device from ever suspending again; disabling twice underflows the usage count and triggers a "Runtime PM usage count underflow" warning. Bail out early when the requested state matches the current state. While at it, switch to pm_runtime_resume_and_get() so a failed resume is propagated to userspace instead of silently marking the event enabled.
In the Linux kernel, the following vulnerability has been resolved: net/sched: cls_api: fix teardown of an adopted proto on insert-race loss In tc_new_tfilter() the create branch sets tp_created = 1 before calling tcf_chain_tp_insert_unique(). When the caller loses the race (another request inserted a proto at the same chain/prio first), insert_unique() destroys the caller's own tp_new and returns the winner's proto with an extra reference. tp_created was never cleared, so the loser's errout path treated the winner's live proto as its own and called tcf_chain_tp_delete_empty() on it, silently unlinking an active classifier that the winning request already advertised via RTM_NEWTFILTER. Track the outcome of the insert step in a single tri-state variable so each errout path reacts correctly: - TP_NOT_CREATED: no proto created; pursue the old path. - TP_CREATED: proto inserted successfully; same code path as before. - TP_NOT_OWNED: New - lost the insert race; tp is another request's proto (chain ref already released by tp_new's destroy) Both errout reactions are single expressions derived from the state. This fix is motivated by the Sashiko's automated review of Patch (net/sched: cls_api: Always acquire rtnl_lock when destroying locked classifiers) [1][2]. The review identified the silent-unlink behaviour of an adopted proto's teardown when a request loses the tcf_chain_tp_insert_unique() race. [1] https://sashiko.dev/#/patchset/20260801125632.360365-1-jhs%40mojatatu.com [2] https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20260801125632.360365-1-jhs%40mojatatu.com
In the Linux kernel, the following vulnerability has been resolved: bpf: Fix mmap_lock leak in irq_work path stack_map_get_build_id_offset() introduced a per-CPU irq_work to defer mmap_read_unlock() from NMI context, and bpf_find_vma() later reused the same mmap_unlock_work. Both callers only check whether the work is busy before taking mmap_lock, so a nested caller can reuse the slot before the first caller queues it. Two read locks may then be acquired while only one deferred unlock runs, leaking a read lock and blocking exit_mmap(). Reserve the per-CPU slot before mmap_read_trylock(). Use the same wrapper in stackmap and bpf_find_vma() so both callers release the reservation on trylock failure. Keep rejecting the slot while the irq_work remains busy. Release it after the irq_work callback unlocks the mm.
In the Linux kernel, the following vulnerability has been resolved: fbdev: kyro: Validate overlay viewport coordinates The overlay viewport end coordinates are computed from the viewport origin and dimensions using 32-bit unsigned arithmetic. Large input values can cause these calculations to wrap around before the resulting coordinates are passed to SetOverlayViewPort(). SetOverlayViewPort() packs the viewport coordinates into 16-bit register fields. The X coordinates are additionally adjusted by +2 and +1 before being written. Validate the coordinate calculations for 32-bit wraparound and ensure that the adjusted coordinates fit within their 16-bit register fields before calling SetOverlayViewPort(). Found by Linux Verification Center (linuxtesting.org) with SVACE.
In the Linux kernel, the following vulnerability has been resolved: iommu/vt-d: Fix iopf_refcount leak on RID domain replacement intel_iommu_attach_device() enables IOPF for the new domain but never disables it for the old one. device_block_translation(), called at the start of the function, tears down translation but does not touch any IOPF state; blocking_domain_attach_dev() has to call iopf_for_domain_remove() explicitly before invoking it for exactly this reason. identity_domain_attach_dev() has the same problem. Its comment claims that no PRI handling is needed because the device has been put in the blocking state, but the blocking state and the IOPF reference count are independent of each other. As a result, replacing a domain that has an iopf_handler with another domain at RID level leaks a reference in info->iopf_refcount. The count never drops back to zero, so iopf_queue_remove_device() is never called and iommu_disable_pci_pri() triggers its WARN_ON(info->iopf_refcount) when the device is released. The PASID paths already handle this correctly by way of iopf_for_domain_replace(); convert the two RID paths to do the same. Using the replace helper rather than a bare remove keeps the enable before the disable, so the reference count does not transiently reach zero and evict the device from the IOPF queue.
In the Linux kernel, the following vulnerability has been resolved: media: amd: isp4: release partial allocations in isp4if_alloc_fw_gpumem() isp4if_alloc_fw_gpumem() allocates several GPU memory pools in sequence. If one of them fails, it jumps to error_no_memory and returns -ENOMEM without releasing the pools that were already allocated, leaking them. Release the already-allocated pools before returning. isp4if_gpu_mem_free() is a no-op on pools that were not allocated, so calling isp4if_dealloc_fw_gpumem() here safely frees exactly the pools that succeeded. isp4if_gpu_mem_free() previously logged an error for a NULL entry, which is a normal case during partial-allocation cleanup, so make it silent.
In the Linux kernel, the following vulnerability has been resolved: media: amd: isp4: fix self-deadlock in isp4sd_pwron_and_init() error path isp4sd_pwron_and_init() holds ops_mutex via guard(mutex) and, on any init failure, jumps to err_deinit and calls isp4sd_pwroff_and_deinit(). That helper takes the same ops_mutex, re-acquiring a non-recursive mutex already held by the current thread, so any init failure deadlocks. Unwind the error path in stages instead, releasing only what each failure point acquired. This also avoids the issues that an unconditional teardown would hit at the earlier failures, such as a runtime-PM underflow from pm_runtime_resume_and_get() and MMIO access while the device is unpowered.
In the Linux kernel, the following vulnerability has been resolved: nvme-pci: release descriptor pools on probe failure The per-NUMA-node descriptor DMA pools are created lazily from nvme_init_hctx_common() once the admin tag set is allocated, but they are only destroyed in nvme_remove() via nvme_release_descriptor_pools(). Any probe failure after the admin tag set has been allocated unwinds through the out_disable label and nvme_pci_free_ctrl(), neither of which releases the pools, leaking the dma_pool objects. Release the descriptor pools in the out_disable error path. It must not be added to nvme_pci_free_ctrl(), as that would double-free against nvme_remove() on the normal teardown path.
In the Linux kernel, the following vulnerability has been resolved: amt: Don't support cross-netns setup. When a lower device is unregistered, amt_device_event() tries to unregister its upper AMT device, but it has two problems. 1. amt_lookup_upper_dev() looks up an upper device in the lower device's netns only 2. amt_device_event() unregisters a single upper device only If AMT device is created on a lower device in another netns, removing the lower device triggers the splat below and gets stuck until all upper devices are removed. [0] The cross-netns setup seems unintentional considering 1. and the following points: * amt_link_setup() sets dev->netns_immutable to true * skb_scrub_packet() is not called in the fast path * iproute2 binary fails to find cross-netns lower device via link-netns: # ip -n ns1 link add amt0 link-netns ns2 type amt dev veth1 Cannot find device "veth1" Instead of supporting it properly and preparing for per-netns netdev unreg, let's forbid cross-netns setup. Note that the problem 2. needs a separate fix. [0]: WARNING: net/core/dev.c:12518 at unregister_netdevice_many_notify+0x1cce/0x2250, CPU#48: ip/2031 Modules linked in: CPU: 48 UID: 0 PID: 2031 Comm: ip Not tainted 7.2.0-rc5+ #27 PREEMPT(full) Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.17.0-debian-1.17.0-1 04/01/2014 RIP: 0010:unregister_netdevice_many_notify (net/core/dev.c:12518) Code: 89 ef e8 d5 52 ae fe e9 d0 f4 ff ff 48 8d 3d f9 3b 9c 02 48 c7 c6 c0 0b 63 84 ba ab 1f 00 00 67 48 0f b9 3a e9 65 ff ff ff 90 <0f> 0b 90 eb 81 48 8d 3d f6 3b 9c 02 48 c7 c6 c0 0b 63 84 ba e2 1f RSP: 0018:ffffc90004abf160 EFLAGS: 00010212 RAX: ffff888104d38260 RBX: ffff88800b0911b8 RCX: dffffc0000000000 RDX: 0000000000000000 RSI: 0000000000000008 RDI: ffffffff85b9f880 RBP: ffffc90004abf2d0 R08: ffffffff85b9f887 R09: 1ffffffff0b73f10 R10: dffffc0000000000 R11: fffffbfff0b73f11 R12: ffff88800b091d08 R13: ffff88800b091178 R14: dffffc0000000000 R15: ffff88800b091000 FS: 00007f555b86c600(0000) GS:ffff8881942a0000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000562107d489c0 CR3: 0000000109a40002 CR4: 0000000000372ef0 Call Trace: <TASK> rtnl_dellink (net/core/rtnetlink.c:3632 net/core/rtnetlink.c:3674) rtnetlink_rcv_msg (net/core/rtnetlink.c:7112) netlink_rcv_skb (net/netlink/af_netlink.c:2556) netlink_unicast (net/netlink/af_netlink.c:1319) netlink_sendmsg (net/netlink/af_netlink.c:1900) ____sys_sendmsg (net/socket.c:775) __sys_sendmsg (net/socket.c:2738) do_syscall_64 (arch/x86/entry/syscall_64.c:63) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:121) ... unregister_netdevice: waiting for veth0 to become free. Usage count = 7 ref_tracker: netdev@ffff88800d7496d8 has 3/3 users at __netdev_adjacent_dev_insert (./include/linux/netdevice.h:4525 ./include/linux/netdevice.h:4554 net/core/dev.c:8791) __netdev_upper_dev_link (net/core/dev.c:8879 net/core/dev.c:8963) netdev_upper_dev_link (net/core/dev.c:9009) amt_newlink (drivers/net/amt.c:3321)
In the Linux kernel, the following vulnerability has been resolved: nfc: pn533: hold a reference to the request skb during send_frame __pn533_send_async() publishes the command and then calls dev->phy_ops->send_frame(). Once dev->cmd is set, an incoming frame can be matched to this command: the I2C threaded IRQ runs pn533_recv_frame(), which queues cmd_complete_work, and pn533_send_async_complete() frees cmd->req with consume_skb(). On the I2C transport, pn533_i2c_send_frame() still dereferences the same skb after i2c_master_send() returns, so a completion that races the send can free the skb while the transport is still using it. The request skb is owned by the command object and may be freed by command completion at any time after dev->cmd is published, so the transport send path must not assume it stays alive. Hold a temporary reference to the request skb across the send_frame() call so the transport always sees a live skb even if completion races the send. Add a pn533_send_cmd_frame() helper and use it from all three send paths.
In the Linux kernel, the following vulnerability has been resolved: ALSA: seq: Don't leak the extension cell pointer in the bounce payload The bounce_error_event() embeds the failed event in the bounce payload by pointing data.ext.ptr at it. When that event is a queued variable-length event, its own data.ext.ptr holds the address of its first extension cell, put there by snd_seq_event_dup(). The payload goes out verbatim through snd_seq_expand_var_event(), so the address reaches userspace. That is the same address commit 705dd6dcbc0e ("ALSA: seq: Clear variable event pointer on read") removed from the event header. The read path still clears it there, just above the call that expands the payload. Embed a sanitised copy instead, treated exactly as snd_seq_read() treats the header. A stack copy is enough because delivery is synchronous and snd_seq_event_dup() copies before returning. An unprivileged client reaches this by setting SNDRV_SEQ_FILTER_BOUNCE, queueing a variable-length event to a port that does not exist and reading the bounce back. Eight bytes on 64-bit, from its own pool.
In the Linux kernel, the following vulnerability has been resolved: RDMA/cxgb4: Free debugfs on registration failure c4iw_alloc() creates the per-device debugfs tree (dev->debugfs_root via setup_debugfs()), but it is removed only in c4iw_remove(), not in c4iw_dealloc(). When RDMA device registration fails, the registration worker's err_dealloc_ctx path calls c4iw_dealloc() directly, bypassing c4iw_remove(), so the debugfs dentries leak and outlive the freed c4iw_dev. Move debugfs_remove_recursive() into c4iw_dealloc() so every path that frees ctx->dev also removes its debugfs tree.
In the Linux kernel, the following vulnerability has been resolved: ubi: Fix rollback for explicit UBI device numbers ubi_init_attach() rolls back module initialization failures by scanning ubi_devices[0..i-1], where i is the mtd= parameter index. That assumes the parameter index matches the UBI device number. That assumption is not true when mtd= specifies an explicit ubi_num. A successfully attached device can be stored at a higher ubi_devices[] slot, and a later failure can miss it during rollback. Scan the full ubi_devices[] array and detach by the actual array index, matching the way UBI devices are stored.
In the Linux kernel, the following vulnerability has been resolved: mtd: ubi: Release device reference on busy detach ubi_detach_mtd_dev() obtains a device reference through ubi_get_device() before checking whether the UBI device is busy. The busy return path drops ubi->ref_count but leaves the device reference held, so the device object cannot be released after a later detach. Drop the device reference before returning -EBUSY.
In the Linux kernel, the following vulnerability has been resolved: ASoC: xilinx: formatter_pcm: fix stream_data leak on open error In xlnx_formatter_pcm_open(), stream_data is allocated and adata->play_stream or adata->capture_stream is assigned early. If a later step, such as snd_pcm_hw_constraint_step() or snd_pcm_hw_constraint_integer(), fails, the function returns the error immediately. ALSA does not call the close callback when open fails, so stream_data is leaked and the stream pointer is left dangling, pointing to a substream that ALSA frees. A later interrupt would then call snd_pcm_period_elapsed() on the freed substream. Free stream_data and clear the stream pointer on the error paths.
In the Linux kernel, the following vulnerability has been resolved: firewire: core: fix memory leak in error path of build_tree() In the error path of build_tree(), node instances can remain in the local linked list when the function returns. Whenever an invalid value is detected in the self ID sequence, each allocated node instance is either an entry in the linked list or an entry in the ports array of its parent node. Therefore, the allocate node instances can be safely released by traversing the linked list from its head. Release the remaining node instances with for_each_fw_node() before returning to the caller.
In the Linux kernel, the following vulnerability has been resolved: arm64/efi: Avoid voluntary preemption with efi_mm installed Gus reports a bad kernel memory access when using software PAN (CONFIG_ARM64_SW_TTBR0_PAN=y) on a machine with support for EFI runtime services: Unable to handle kernel access to user memory outside uaccess routines at virtual address 00000000f322ff30 Mem abort info: ESR = 0x0000000096000004 FSC = 0x04: level 0 translation fault Internal error: Oops: 0000000096000004 [#1] SMP Workqueue: efi_rts_wq efi_call_rts pstate: 80400005 (Nzcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : efi_call_rts+0xd8/0x288 Call trace: efi_call_rts+0xd8/0x288 (P) process_one_work+0x178/0x4f8 worker_thread+0x194/0x328 This is because the fpsimd context management code called from __efi_fpsimd_begin() can preempt voluntarily, returning later to the EFI code with an incorrect value for TTBR0_EL1 thanks to the deferred mm switching used by the software PAN implementation. Since EFI runtime services cannot preempt voluntarily and because the fpsimd switching code does not rely on the TTBR0_EL1 mappings, simply reorder the fpsimd switch so that it occurs before we change the page-table.
In the Linux kernel, the following vulnerability has been resolved: bpf, s390: Clear fetch destination on faulting arena atomic Same missing register clear as on riscv64. A RMW atomic on an arena pointer is converted to BPF_PROBE_ATOMIC and gets an exception table entry, but bpf_jit_probe_atomic_pre() only fills in the arena base and the probe offset, leaving probe->reg at the -1 that bpf_jit_probe_init() set, which bpf_jit_probe_post() writes into the entry and ex_handler_bpf() then reads back as "there is nothing to clear". That is right for a plain BPF_{ADD,AND,OR,XOR}, which only writes memory, but an RMW carrying BPF_FETCH also reads the old value into a register: src_reg for BPF_{ADD,AND,OR,XOR} | BPF_FETCH and BPF_XCHG, and r0 for BPF_CMPXCHG. So on a fault over an unmapped arena page the program resumes at the landing pad with whatever that register held before the atomic instead of the 0 that every other BPF_PROBE_* access delivers. Fill probe->reg in from bpf_atomic_load_reg(). Unlike x86-64 and arm64, s390x does not report arena violations from its exception handler, so there is no access direction to correct here, only the missing register clear.
In the Linux kernel, the following vulnerability has been resolved: s390/debug: Fix deadlock during unregister Unregistering an s390dbf debug area while one of the associated debugfs files is being written to can cause a deadlock: $ echo >.../vmur/level $ rmmod vmur =================================================== debugfs write debugfs_file_get() debug_unregister() mutex_lock(debug_mutex) debugfs_remove() wait for debugfs_file_put() debug_file_ops.write() debug_input() mutex_lock(debug_mutex) ==> DEADLOCK Fix this by splitting debug_unregister() into an s390dbf and debugfs part, and running only the s390dbf part with debug_mutex locked.
In the Linux kernel, the following vulnerability has been resolved: clocksource/drivers/samsung_pwm: Switch to raw_spinlock_t type Samsung PWM timer might be used as a clock source on some legacy systems. When PREEMPT_RT is enabled on ARM, regular spinlock is converted to a sleeping lock (mutex-based), which must not be used in atomic context such as hard interrupt handlers. Switch the samsung_pwm_lock to the raw_spinlock, which remains a true non-sleeping spinlock even under PREEMPT_RT.
In the Linux kernel, the following vulnerability has been resolved: net: page_pool: fix UAF in __page_pool_release_netmem_dma on xa_cmpxchg race This bug was discovered while testing the hns3 driver under channel reconfiguration (`ethtool -L` / `ethtool -G`) with iperf3 traffic on arm64. The race is intermittently triggered when page_pool_destroy() runs page_pool_scrub() concurrently with page return via page_pool_put_netmem() on a different CPU. A WARN in page_pool_clear_pp_info() surfaced the dangling DMA index bits left by the cmpxchg loser, which led to the investigation. page_pool_scrub() iterates pool->dma_mapped via xa_for_each() with no page ref held. __page_pool_release_netmem_dma() currently reads and writes netmem fields (dma_addr, DMA index bits in pp_magic) after xa_cmpxchg() returns. The unref path calls put_page() unconditionally regardless of the cmpxchg outcome; when it loses the cmpxchg, it still frees the page before the scrub winner finishes these netmem accesses, so scrub touches a freed page -- a Use-After-Free. Fix this by splitting the DMA release into two functions: 1. __page_pool_unmap_netmem_dma() caches dma_addr before xa_cmpxchg(), does the cmpxchg to remove the DMA mapping, and calls dma_unmap on the cached address. It never touches netmem fields after the cmpxchg, making it safe for the scrub path which holds no page ref. 2. __page_pool_release_netmem_dma() wraps the above and additionally clears dma_addr and DMA index bits in netmem fields. This is safe only when the caller holds a page ref, so it is used by the return path (page_pool_return_netmem). The scrub path calls __page_pool_unmap_netmem_dma() directly; the return path calls __page_pool_release_netmem_dma().
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7915: clear wcid mask under mutex after RCU pointer clear mt7915_remove_interface() cleared the wcid mask bit with no lock held and before clearing the RCU wcid pointer. The mask is a non-atomic RMW shared with the allocators, which all run under dev->mt76.mutex; on DBDC the two wiphys share one mt76_dev, so this raced add_interface/sta_add on the other band and could leak or double-hand-out a wcid. Clearing the bit before the RCU pointer also let a concurrent allocation reuse the index and publish its wcid, which the subsequent NULL assignment then wiped. Move the clear into the existing mutex section, after the RCU pointer is cleared.
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7996: bound TLV walk in mt7996_mcu_get_chip_config The response TLV loop advanced by tlv->len without a minimum, so a theoretical firmware response containing a zero-length TLV could spin forever, hanging the CPU during device probe. The u32 payload was also read without bounds checking. Reject a short fixed field, stop on a TLV whose length underruns the header or overruns the skb.
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7915: unwind state on add_interface failure When mt76_wcid_alloc() fails, mt7915_add_interface() returned without clearing the vif_mask/omac_mask bits it had already set, without removing the firmware dev info added earlier, and without clearing a monitor_vif pointer to the vif mac80211 is about to free. mac80211 does not call remove_interface() for a failed add, so the indices and firmware dev entry leaked permanently and testmode could dereference the stale monitor_vif. Add a proper error unwind.
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7996: reserve space for the CSA-abort countdown TLV When a CSA countdown is active, mt7996_mcu_beacon_cntdwn() emits two bss_bcn_cntdwn_tlv entries (the CSA countdown and the CCA-abort BCC), but MT7996_BEACON_UPDATE_SIZE only reserved one. With MBSSID enabled and a near-maximum beacon template the extra 8 bytes could push the offload command past MT7996_MAX_BSS_OFFLOAD_SIZE and trigger skb_over_panic(). Reserve room for both countdown TLVs.
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: cancel reset and rc work on device unregister Both drivers cancelled dump_work on unregister but left reset_work and rc_work to be flushed only by destroy_workqueue() in mt76_free_device(), which runs after the hw is unregistered and the hardware stopped. A reset_work that fires in that window calls ieee80211_restart_hw() and re-arms mac_work on an unregistered hw, and rc_work touches station state being torn down. Cancel both up front, alongside dump_work.
In the Linux kernel, the following vulnerability has been resolved: ACPI: processor: Unregister cpufreq notifier on init failure acpi_processor_driver_init() registers the cpufreq policy notifier before registering the ACPI processor driver and setting up CPU hotplug state. If driver_register() or cpuhp_setup_state() fails, the error path only unregisters the ACPI processor driver and the idle driver. The cpufreq notifier remains registered even though initialization failed. Mirror the module exit path on the init failure path and unregister the cpufreq notifier when it has been registered.
In the Linux kernel, the following vulnerability has been resolved: drm/msm/dsi: Drop dev_pm_opp_set_rate(0) dev_pm_opp_set_rate(0) removes the vote specified in required-opps but does not actually park the clock, making it run without the necessary power backing. Drop the explicit call to it. Every call site of ops->link_clk_disable() is followed by pm_runtime_put(), so the power vote will be rescinded if deemed safe. Patchwork: https://patchwork.freedesktop.org/patch/742783/
In the Linux kernel, the following vulnerability has been resolved: wifi: ath11k: fix leak in ath11k_service_ready_ext_event() Currently, during ath11k_service_ready_ext_event() processing, svc_rdy_ext.mac_phy_caps can be allocated during TLV parsing. This is a temporary allocation that is freed on the success path, but not on the error path. If parsing succeeds far enough to allocate mac_phy_caps and then fails on a later TLV, the allocation leaks. So free the allocation on the error path. Compile tested only.
In the Linux kernel, the following vulnerability has been resolved: regulator: core: use system_freezable_wq for init complete work schedule_delayed_work() uses system_wq, which is non-freezable, allowing regulator_init_complete_work to run concurrently with system suspend. This work fires ~30s after boot to disable unused regulators via I2C. When it races with PM suspend, the I2C adapter may already be suspended, triggering a -ESHUTDOWN warning in __i2c_transfer(): WARNING: ... at __i2c_transfer+0x36c/0x3c8 Call trace: __i2c_transfer i2c_transfer regmap_i2c_write _regmap_update_bits regulator_disable_regmap _regulator_do_disable regulator_late_cleanup regulator_init_complete_work_function process_one_work Switch to system_freezable_wq so the work is frozen before any device is suspended, eliminating the race.
In the Linux kernel, the following vulnerability has been resolved: bpf: Reject >8 byte return values on return-reading trampoline paths btf_distill_func_proto() builds the function model used for the fentry/fexit/fmod_ret/fsession trampolines and struct_ops. It has accepted a 16-byte __int128 return value since the trampoline was introduced: __get_type_size() returns the integer's type size, and the return-type check only rejected ret < 0. But the BPF trampoline preserves only 8 bytes of the return value (RAX on x86, i.e. R0). For an attach type that reads the target's return value the second half (RDX / R3) is neither saved nor restored, so a program attached to a function returning a 16-byte value corrupts the value seen by the real caller and itself observes only half of it. struct_ops trampolines have the same limitation. This affects the attach types that read the target's return value: fexit, fmod_ret and fsession (plus the _multi variants of fexit and fsession), and struct_ops. fentry/fentry_multi run before the target returns and are unaffected. Reject a >8 byte return value for these attach types in bpf_check_attach_target() and bpf_check_attach_btf_id_multi(), and for struct_ops in bpf_struct_ops_desc_init().
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7996: free vif links after clearing wcid entries on full reset mt7996_mac_reset_vif_iter() queues non-default vif links for kfree_rcu while dev->wcid[] still holds pointers to the wcid embedded in each freed link; mt76_reset_device() then dereferences those entries and runs mt76_wcid_cleanup() on them. If a grace period elapses in between, the cleanup operates on freed memory. Run mt76_reset_device() first, so the wcid entries are cleaned up and cleared while the links are still valid.
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7996: clear stale link state on full reset After a full chip reset, mac80211 reconfig replays interface, link and channel context setup. mt7996_vif_link_add() short-circuits when the link_id is still marked in mvif->valid_links, a state introduced for postponing link teardown to interface removal. The reset path frees the link structures without clearing those bits, so the replayed setup never re-creates dev_info/bss_info/STA records in the restarted firmware and never re-registers the link wcid, leaving the device inoperative. The reset path also leaks every allocated MLD index: per-link indices and the per-vif group/remap indices are re-allocated from scratch during reconfig, but the old bits stay set in the masks, so repeated full resets exhaust the index space. Clear valid_links in the reset vif iterator and reset the MLD index masks alongside the existing omac_mask clearing.
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7915: fix double hif2 init on the non-WED path mt7915_pci_init_hif2() was called unconditionally and again inside the WED-inactive branch. The helper increments the global hif_idx, writes the PCIe RECOG_ID register and takes a get_device() reference via mt7915_pci_get_hif2(), while removal only drops one reference. On non-WED dual-hif hardware this double-incremented hif_idx, wrote RECOG_ID twice and leaked a device reference. Only the call inside the WED-inactive branch is correct; drop the unconditional one. hif2 is already initialised to NULL.
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: mt7915: release hif2 reference on probe IRQ failure The hif2 reference obtained by mt7915_pci_init_hif2() is only released on error paths that key off dev->hif2, which is not assigned until after the IRQ setup. If pci_alloc_irq_vectors() or the primary devm_request_irq() fails, the reference leaks. Drop it explicitly on those paths via mt7915_put_hif2().
In the Linux kernel, the following vulnerability has been resolved: wifi: mt76: reject out-of-range link ids in mt76_vif_link() mt76_vif_link() indexes mvif->link[] without validating link_id, but callers pass mvif->deflink_id / msta->deflink_id, which hold IEEE80211_LINK_UNSPECIFIED (0xf) until the first link has been added. Since IEEE80211_MLD_MAX_NUM_LINKS is 15, that reads one element past the end of the array, aliasing mt76_vif_data.offchannel_link. Reachable via mt7996_set_tsf()/mt7996_offset_tsf() and mt7996_net_fill_forward_path(). Bounds check link_id and return NULL, matching mt7996_sta_link() and mt7996_sta_link_protected().
In the Linux kernel, the following vulnerability has been resolved: wifi: nl80211: clean up color-change beacon data on errors nl80211_color_change() calls nl80211_parse_beacon() for the beacon_next template, which can allocate params.beacon_next.mbssid_ies and .rnr_ies. A parsing failure returned directly instead of using the out: cleanup, leaking any allocations completed before the error. Allocate the nested attribute table before parsing beacon_next. Its allocation failure can then return before beacon data exists, while a later parsing failure uses out: to release the parsed data.
In the Linux kernel, the following vulnerability has been resolved: wifi: brcmfmac: fix P2P action frame handling without device vif Some P2P action frame paths assume the P2P device vif is always available. That is not true when userspace sends non-P2P public action frames through the primary interface, or when action-frame abort runs after the P2P device vif has not been created. Fall back to the primary vif when aborting an action frame without a P2P device vif, and guard P2P device saved IE access before using it for peer channel search.
In the Linux kernel, the following vulnerability has been resolved: bpf: Fix mmap_lock deadlock on arena lock failure Reported by the Sashiko AI review. arena_vm_fault() returns VM_FAULT_RETRY when it can't take arena->spinlock, but it never took mmap_lock. The fault path assumes a VM_FAULT_RETRY handler already dropped mmap_lock and re-takes it on the retry, so mmap_lock gets taken twice and can deadlock: do_user_addr_fault() { fault = handle_mm_fault(...); // calls arena_vm_fault() if (fault & VM_FAULT_RETRY) goto retry; // re-locks mmap_lock mmap_read_unlock(mm); } Return VM_FAULT_SIGBUS instead, for two reasons: 1. We could keep VM_FAULT_RETRY, but then we'd have to drop the fault lock first and cap the retry ourselves, the way __folio_lock_or_retry() does. 2. A failed raw_res_spin_lock_irqsave() already means a possible deadlock was detected, so retrying just hits the same lock again. So returning VM_FAULT_RETRY here is overkill.
In the Linux kernel, the following vulnerability has been resolved: pinctrl: generic: free maps on pinctrl_generic_to_map() failure pinctrl_generic_to_map() parses DT configuration and allocates pinctrl maps via pinctrl_utils_reserve_map(). If subsequent steps (such as pinctrl_utils_add_map_mux(), pinctrl_generic_add_group(), pinconf_generic_parse_dt_config(), or pinctrl_utils_add_map_configs()) return an error, *maps may contain partially allocated map entries. Returning the error directly without freeing *maps leaks the allocated mapping memory across all drivers that rely on pinctrl_generic_to_map(). Fix this by calling pinctrl_utils_free_map() and resetting *maps, *num_maps, and *num_reserved_maps in the error path of pinctrl_generic_to_map().
In the Linux kernel, the following vulnerability has been resolved: serial: amba-pl011: keep console clock enabled for atomic writes pl011_console_write_atomic() runs from nbcon atomic context, where sleeping is not allowed. It calls clk_enable(), which takes the common-clk enable_lock. Under PREEMPT_RT that is a sleeping lock: clk_enable_lock() first tries spin_trylock_irqsave(), but on contention falls back to spin_lock_irqsave(). Therefore, an atomic-context printk on an RT kernel with a clk-backed pl011 can trip: BUG: sleeping function called from invalid context at spinlock_rt.c:48 __might_resched from rt_spin_lock rt_spin_lock from clk_enable_lock clk_enable_lock from clk_enable clk_enable from pl011_console_write_atomic ... from vprintk_emit This was found and reproduced on PREEMPT_RT. Arm32 and arm64 DT SoCs are affected; arm64 SBSA/ACPI has no clk, so clk_enable(NULL) short-circuits before the lock. In addition, write_atomic() may be invoked from NMI context and is documented to avoid locking. Removing clk_enable() from the callback also avoids a potentially unsafe NMI acquisition of the common-clock enable_lock. An nbcon atomic-capable console must be printable from any context, so the clock cannot be gated between writes. Enable the clock while the console is available for output: use clk_prepare_enable() in pl011_console_setup(), release it via clk_disable_unprepare() in the console .exit() callback, and drop the per-write clk_enable()/clk_disable() pairs from write_atomic() and write_thread(). When printk suspends consoles, drop the reference after uart_suspend_port() stops console access and restore it before uart_resume_port() -- but only if suspend actually marked the port suspended (a wake-capable tty stays running and must keep its clock), and keep it when console_suspend_enabled is false so no_console_suspend works. The active power cost of keeping the clock enabled is platform-dependent: none where the UART clock is a fixed always-on oscillator, real where it is a gateable clock branch, which then cannot be gated (nor possibly can its parent clocks) while the console is available for output. When serial core actually suspends the port, the reference is released so the clock provider can gate the clock tree.
In the Linux kernel, the following vulnerability has been resolved: serial: core: do fallible allocations before the console can be registered serial_core_add_one_port() allocates uport->tty_groups after uart_configure_port(), which may register the console. If the allocation fails, the driver unwinds the port while its console remains registered. The earlier uport->name allocation has a related failure path that leaves state->uart_port linked to a port being freed. Failslab reproduced a NULL dereference in PL011 console output and a KASAN use-after-free in i.MX console output after failed binds. Allocate the name and tty_groups before linking the port and configuring it. Reserve space for the optional driver attribute group because config_port() may populate uport->attr_group during configuration.
In the Linux kernel, the following vulnerability has been resolved: serial: core: clear freed pointers on uart_register_driver() failure uart_register_driver() leaves drv->state pointing to freed memory when tty_alloc_driver() fails. If tty_register_driver() fails, drv->tty_driver also retains a pointer after its reference is dropped. Drivers that use drv->state as an "already registered" flag can then skip registration on the next probe and pass the freed state to uart_add_one_port(). This issue was found with failslab on QEMU's raspi1ap board by failing registration and binding the PL011 port again. Clear both pointers on their failure paths, as uart_unregister_driver() already does.
In the Linux kernel, the following vulnerability has been resolved: tty: skip cdev_del() when no cdev is registered TTY device registration can fail before a cdev is allocated. Serial core keeps the port so setserial can still use it, and later removal passes the NULL cdev slot to cdev_del(), causing a NULL-pointer dereference. Only delete the cdev when the slot is not NULL.
In the Linux kernel, the following vulnerability has been resolved: tty: clear cdev pointer after cdev_add() failure tty_cdev_add() drops the cdev reference when cdev_add() fails, but leaves driver->cdevs[index] pointing to freed memory. tty_unregister_device() later passes that stale pointer to cdev_del(), causing a use-after-free. Clear the slot after dropping the reference.
In the Linux kernel, the following vulnerability has been resolved: dm-integrity: replace forgeable discard filler with a keyed sector marker The discard-block check in dm_integrity_rw_tag() treats a stored tag of all 0xf6 bytes (DISCARD_FILLER) as proof a block was discarded and skips HMAC verification. allow_discards is only accepted in dm-integrity's standalone mode. An attacker with raw write access to the backing device, but without the integrity key, can stamp any block with an all-0xf6 tag and have it served as authentic. Add a new "allow_discards_keyed" target argument that marks discarded blocks with a keyed checksum of (salt || sector) instead, computed by integrity_discard_checksum().
In the Linux kernel, the following vulnerability has been resolved: HID: asus: refactor the two workqueues and init sequence Multiple issues have been found within the hid-asus driver: - unchecked size in asus_raw_event() - unclean teardown of asus_probe on failure - possible use-after-free in asus_probe - multiple workqueue used for jobs where one was enough - sleeping calls in atomic context - packets of incorrect size being sent to the keyboard controller Join the two workqueues into one reusing the stopping mechanism of the brightness workqueue, use the joined workqueue to also move the asus_wmi_send_event() sleeping call away from atomic context and add a size check in asus_raw_event().
In the Linux kernel, the following vulnerability has been resolved: HID: logitech-hidpp: Fix FF device cleanup on init failure hidpp_ff_init() creates the input force-feedback device with input_ff_create(), then allocates the HID++ FF private data, effect ID array, and workqueue. If any of those allocations fail after input_ff_create() succeeds, the function returns an error without destroying the FF device. Add an unwind path that frees the private allocations made by hidpp_ff_init() and calls input_ff_destroy() for failures after input_ff_create() succeeds.
In the Linux kernel, the following vulnerability has been resolved: HID: steam: Reject short reads Steam Controller FEATURE reports encode the size of the message in the message itself. Previously we were trusting that the size reported matched the size we actually read, leading to a potential issue with short reads. Instead, we should actually verify the length of the read.
In the Linux kernel, the following vulnerability has been resolved: phonet: pep: do not write beyond optlen in getsockopt pep_getsockopt() clamps the reported length to the caller's buffer with min_t(), but then stores the value with put_user(val, (int __user *) optval), which always writes sizeof(int) bytes. A getsockopt() call with an optlen smaller than sizeof(int) thus reports the clamped length yet writes a full int, one to three bytes past the user buffer. Write the value with copy_to_user() bounded by len, so at most optlen bytes are copied, matching the length reported back to userspace.
In the Linux kernel, the following vulnerability has been resolved: ublk: validate auto buf reg before taking uring_cmd With UBLK_F_AUTO_BUF_REG, invalid sqe->addr can fail after ublk_fill_io_cmd() has set UBLK_IO_FLAG_ACTIVE. The uring_cmd is completed while the tag stays active, which can hang teardown. Split validation from buffer apply so the check has no side effects, then take the uring_cmd and store the already-validated buffer. Apply the same order in FETCH so io->buf is not written before __ublk_fetch() state checks.
In the Linux kernel, the following vulnerability has been resolved: rapidio: clear mport->net when rio_add_net() fails rio_alloc_net() stores the newly allocated rio_net in mport->net before rio_scan_alloc_net() registers the device. If rio_add_net() fails, rio_scan_alloc_net() drops the device reference with put_device(), which releases the rio_net through the device release callback. However, mport->net is left pointing at the freed object. A later mport unregister path can then dereference the dangling mport->net pointer and may try to free the same rio_net again. Clear mport->net in the rio_add_net() failure path, matching the cleanup done for the destID table allocation failure path.
In the Linux kernel, the following vulnerability has been resolved: fat: release buffer head after rebuilding parent fat_scan_logstart() leaves the matching directory entry's buffer head in sinfo.bh for the caller to release, just like fat_scan(). fat_rebuild_parent() uses the directory entry to rebuild the parent inode for the nostale_ro NFS export path, but does not release sinfo.bh after a successful scan. Release it once fat_build_inode() has consumed the directory entry data.
In the Linux kernel, the following vulnerability has been resolved: bpf, cgroup: Fix invalid storage access after __cgroup_bpf_attach failed A potential invalid storage access issue can occur after replacing a cgroup bpf prog. This occurs in the following scenario: 1. prog1 with storage is attached to a cgroup in multi-attach mode. 2. prog1 is replaced with prog2 using BPF_F_REPLACE in multi-attach mode, but fails midway (e.g. in bpf_trampoline_link_cgroup_shim or update_effective_progs). 3. A new prog3 is attached to the cgroup in multi-attach mode. The reason is that __cgroup_bpf_attach overwrites pl->storage with the new storage prior to attachment completion. When attachment fails midway, the cleanup path calls bpf_cgroup_storages_free(new_storage) to free the newly allocated storage, but fails to restore pl->storage back to old_storage. Consequently, the still-active prog1 holds invalid or dangling storage pointers, leading to an invalid memory access when prog1 executes and calls bpf_get_local_storage. Additionally, original pl->flags and cgrp->bpf.flags[atype] are left unrestored. Fix this by saving old_pl_flags, old_storage, and old_flags prior to the update, and properly restoring all of them in the cleanup path on error.
In the Linux kernel, the following vulnerability has been resolved: thermal: hwmon: Remove hwmon class device along with its parent The current code creates one hwmon device per thermal zone type and that device is registered under the first thermal zone of the given type. That turns out to be problematic when the thermal zone holding the hwmon device is removed. For example, say that there are two ACPI thermal zones on a system /sys/devices/virtual/thermal/thermal_zone0/ /sys/devices/virtual/thermal/thermal_zone1/ The current code registers a hwmon class device for thermal_zone0 only: /sys/devices/virtual/thermal/thermal_zone0/hwmon0/ because the type is "acpitz" for both of them, but it adds a sysfs attribute that belongs to thermal_zone1 under it: /sys/devices/virtual/thermal/thermal_zone0/hwmon0/temp2_input There is also /sys/devices/virtual/thermal/thermal_zone0/hwmon0/temp1_input which belongs to thermal_zone0. When thermal_zone0 is removed, say because the ACPI thermal driver is unbound from the underlying platform device, thermal_remove_hwmon_sysfs() skips the removal of hwmon0 because of the temp2_input attribute belonging to thermal_zone1 which effectively prevents thermal_zone0 removal from making progress. Address this by making thermal_remove_hwmon_sysfs() remove the entire hwmon class device interface for the given thermal zone type when the thermal zone device holding it is removed. To prevent races with thermal_add_hwmon_sysfs() that may interfere with this, carry out the entire addition and removal of hwmon sysfs interfaces for thermal zones under thermal_hwmon_list_lock. Also adjust the layout of the labels in thermal_add_hwmon_sysfs() to the current kernel coding style to align with the new "unlock" label.
In the Linux kernel, the following vulnerability has been resolved: RDMA/srp: fix heap information leak on a truncated SRP_CRED_REQ srp_recv_done() passes wc->byte_len to srp_process_rsp(). It passes nothing to srp_process_cred_req() and srp_process_aer_req(), which read fixed-size fields from the receive buffer without checking that those fields were received. The buffer size is max_ti_iu_len, which comes from the login response and is not validated. A target that advertises 8 and then sends an 8-byte SRP_CRED_REQ makes the initiator read req->tag from beyond the end of the buffer. req->tag is copied into the SRP_CRED_RSP and sent back, so those bytes reach the target. SRP_AER_REQ behaves the same way and also reads req->lun. The leak is 8 bytes per response. max_ti_iu_len also decides which slab cache the buffer comes from. With 8 the buffer is a kmalloc-8 object and the read is entirely outside it: BUG: KASAN: slab-out-of-bounds in srp_recv_done+0x172b/0x1aa0 Read of size 8 at addr ffff888104714da8 by task kworker/u8:3/50 which belongs to the cache kmalloc-8 of size 8 The buggy address is located 0 bytes to the right of allocated 8-byte region [ffff888104714da0, ffff888104714da8) Without KASAN the returned bytes are whatever is next in the slab. One run returned ".strtab". rsp->data[3] in srp_process_rsp() has the same problem: only resp_data_len is checked before it is read. Drop a request that is shorter than the structure being parsed, and check byte_len before the tsk_mgmt read.
In the Linux kernel, the following vulnerability has been resolved: ARM: 9483/1: select HAVE_POSIX_CPU_TIMERS_TASK_WORK Commit c6e61c06d606 ("ARM: 9463/1: Allow to enable RT") enabled PREEMPT_RT on ARM but did not select HAVE_POSIX_CPU_TIMERS_TASK_WORK. This leaves CONFIG_POSIX_CPU_TIMERS_TASK_WORK disabled, so CPU timers expire in hard IRQ context. On PREEMPT_RT this makes run_posix_cpu_timers() take the sleeping sighand->siglock: BUG: sleeping function called from invalid context at spinlock_rt.c:48 rt_spin_lock from lock_task_sighand lock_task_sighand from run_posix_cpu_timers run_posix_cpu_timers from update_process_times ARM handles TIF_NOTIFY_RESUME on all return-to-user paths, including v7-M. ARM32 KVM host support was removed by commit 541ad0150ca4 ("arm: Remove 32bit KVM host support"), so the select need not be conditional on KVM. Select it to defer POSIX CPU timer expiry to task context. Reproduced with setrlimit(RLIMIT_CPU, ...) and a busy loop. The same path is used by setitimer(ITIMER_PROF or ITIMER_VIRTUAL) and POSIX CPU timers created with timer_create().
In the Linux kernel, the following vulnerability has been resolved: ARM: 9484/1: enable interrupts when unhandled user faults are triggered PREEMPT_RT requires interrupts to be enabled when sending signals. When do_DataAbort()/do_PrefetchAbort() triggers unhandled user faults, that is `inf->fn()` return a non-zero value, and the interrupts are not enabled within the hook function, force_sig_fault() will be called with interrupts disabled. This can be triggered by user programs executing the bkpt instruction, with kernel config CONFIG_PERF_EVENTS=n. Enable interrupts in do_DataAbort()/do_PrefetchAbort() when unhandled user faults are triggered to fix the issue.
In the Linux kernel, the following vulnerability has been resolved: ARM: 9485/1: mm: acquire mmap write lock around show_pte() for user faults When CONFIG_DEBUG_USER=y, and cmdline "user_debug=31" is set, a user fault may trigger show_pte() without any lock. If another thread in the same process concurrently calls munmap(), the page table pages may be freed while show_pte() is still traversing them, causing a use-after-free in show_pte(). If CONFIG_ARM_LPAE=y, this may cause a kernel panic if the pages table of PMD are freed when show_pte() is running. Acquire mmap_write_lock() around show_pte() for user faults to fix the contention. For user faults, additionally restrict that show_pte() is called only when the addr is a user-space address (addr < TASK_SIZE). This is because the lock of tsk->mm only protects the virtual memory of user address space, furthermore, dumping the page tables of a kernel-space address for user faults is unnecessary and may have security implications. Keep everything unchanged for kernel faults, because the kernel is already in the "oops" state, acquiring a lock may risk a deadlock.
In the Linux kernel, the following vulnerability has been resolved: ocfs2: synchronize heartbeat callbacks with o2net teardown Patch series "ocfs2: harden heartbeat teardown races". This series fixes two OCFS2 heartbeat/o2net teardown races found by KASAN. This patch (of 2): Heartbeat callbacks stay registered while configfs local-node teardown enters o2net_stop_listening(). A node-down event can still run through o2net_disconnect_node() and o2net_set_nn_state() while teardown is destroying o2net_wq, so the later queue/flush operations can hit a dead workqueue. KASAN has caught this as a slab-use-after-free in __queue_work() with the call chain: KASAN slab-use-after-free in __queue_work+0x56/0xa90 Read of size 4 Call trace: dump_stack_lvl+0x66/0xa0 print_report+0xce/0x630 __queue_work+0x56/0xa90 srso_alias_return_thunk+0x5/0xfbef5 __virt_addr_valid+0x19f/0x330 kasan_report+0xe0/0x110 __queue_delayed_work+0x58/0x1e0 queue_delayed_work_on+0xb4/0xc0 o2net_set_nn_state+0x467/0x840 o2net_disconnect_node+0x7b/0xe0 o2net_hb_node_down_cb+0x54/0x60 o2hb_run_event_list+0x236/0x2d0 o2hb_check_slot+0xad4/0xbc0 lock_release+0xc8/0x290 o2hb_check_slot+0x9ea/0xbc0 trace_hardirqs_on+0x18/0x130 o2hb_do_disk_heartbeat+0x646/0xb30 (fs/ocfs2/cluster/heartbeat.c:1079) __lock_acquire+0x466/0x2260 lockdep_hardirqs_on_prepare+0xea/0x1a0 ktime_get_with_offset+0xe9/0x230 o2hb_thread+0x14e/0x770 kthread+0x1ad/0x1f0 ret_from_fork+0x3c9/0x540 __switch_to+0x2e9/0x730 ret_from_fork_asm+0x1a/0x30 Allocated by task stack: kasan_save_stack+0x33/0x60 kasan_save_track+0x14/0x30 __kasan_kmalloc+0xaa/0xb0 __kmalloc_noprof+0x292/0x760 __alloc_workqueue+0x736/0xc60 alloc_workqueue_noprof+0xb1/0x110 o2net_start_listening+0xe5/0x430 o2nm_node_local_store+0x184/0x310 configfs_write_iter+0x18a/0x210 vfs_write+0x469/0x810 ksys_write+0xd2/0x170 do_syscall_64+0x115/0x6a0 (arch/x86/entry/syscall_64.c:87) entry_SYSCALL_64_after_hwframe+0x77/0x7f Freed by task stack: kasan_save_stack+0x33/0x60 kasan_save_track+0x14/0x30 kasan_save_free_info+0x3b/0x60 __kasan_slab_free+0x5f/0x80 kfree+0x313/0x590 rcu_core+0x4f4/0x1320 handle_softirqs+0x156/0x660 queue_delayed_work_on o2net_set_nn_state o2net_disconnect_node o2net_hb_node_down_cb o2hb_run_event_list Keep heartbeat callbacks registered so quorum state still tracks node state, but stop them from driving o2net reconnect/disconnect work once local teardown starts. Mark the transport offline before destroying o2net_wq, wait for any in-flight heartbeat callback to finish, and delay bring-up replay until the new local node is published through o2nm_this_node(). The replay also has to stay serialized with heartbeat callback delivery. Otherwise a live-node snapshot can be copied, a real hb_down callback can install -ENOTCONN for a peer, and the stale replay can call o2net_hb_node_up() for that same peer and queue reconnect work even though heartbeat is already down. The buggy scenario involves two paths, with each column showing the order within that path: local-node teardown: heartbeat node-down callback: 1. configfs local-off enters 1. o2hb_run_event_list() invokes o2net_stop_listening(). o2net_hb_node_down_cb(). 2. teardown heads for 2. the callback reaches destroy_workqueue(o2net_wq). o2net_disconnect_node() and o2net_set_nn_state(). 3. teardown destroys and NULLs 3. the callback flushes or queues o2net_wq. work through o2net_wq.
In the Linux kernel, the following vulnerability has been resolved: bpf: Clear buf on error in __bpf_get_task_stack Both bpf_get_task_stack and bpf_get_task_stack_sleepable helpers that use __bpf_get_task_stack have buf defined as ARG_PTR_TO_UNINIT_MEM argument and we should initialize the buf on every return path. Adding missing buf memset for __bpf_get_task_stack fail paths. This provides deterministic buffer contents, which is useful when the buffer is used directly as a map key.
In the Linux kernel, the following vulnerability has been resolved: drm/sun4i: tcon: Drop TCON TOP device reference of_find_device_by_node() takes a device reference. Drop it after mux configuration succeeds.
In the Linux kernel, the following vulnerability has been resolved: drm/sun4i: crtc: Propagate layer initialization error sun4i_crtc_init() returns plain NULL when layer initialization fails, while all its other error paths return an error pointer. The only caller, sun4i_tcon_bind(), checks the result with IS_ERR() and happily continues with tcon->crtc set to NULL. sun4i_rgb_init() and sun4i_lvds_init() then dereference it in drm_crtc_mask(), which oopses. Return the error pointer instead.
In the Linux kernel, the following vulnerability has been resolved: cpufreq: imx6q: fix devres accumulation across driver rebind imx6_soc_volt is allocated with devm_kcalloc(cpu_dev, ...), where cpu_dev is the CPU device from get_cpu_device(0). That device is never unbound, so its devres list is never released, and imx6q_cpufreq_remove() does not free the array either. Every probe therefore adds an allocation that stays for the lifetime of the system. Allocate against the platform device instead. Its devres is released when the driver is unbound, which is exactly the lifetime the array wants: imx6q_set_target() reads it, and nothing may reach that after cpufreq_unregister_driver(). That makes the array actually go away on unbind, so also clear the file-scope pointer in remove and on the failed-probe path, rather than leave it pointing at memory devres is about to release. Tested by rebinding the driver on qemu's mcimx6ul-evk.
In the Linux kernel, the following vulnerability has been resolved: arm64: hibernate: Restore DAIF state on error Sashiko AI has reported that if swsusp_mte_save_tags() for some reason fails we return from swsusp_arch_suspend() with DAIF being masked - that is not what we'd expect. Restore the saved DAIF state before returning from the error path.
In the Linux kernel, the following vulnerability has been resolved: phy: sunplus: fix error handling in sp_uphy_init() Fix the error paths of sp_uphy_init() to undo exactly what each stage did: return directly if clk_prepare_enable() fails, release only the clock if reset_control_deassert() fails, and jump to err_reset if update_disc_vol() fails so the clock and reset are not leaked.
In the Linux kernel, the following vulnerability has been resolved: scsi: qla2xxx: Remove redundant VPD flash read in sysfs read path qla2x00_sysfs_read_vpd() called ha->isp_ops->read_optrom() a second time after releasing optrom_mutex. The repeated read is redundant and, unlike the first, runs without optrom_mutex held, exposing flash access to concurrent optrom operations. Drop the duplicate call.
In the Linux kernel, the following vulnerability has been resolved: firmware_loader: do not queue completed sysfs fallback requests fw_load_sysfs_fallback() calls device_add() before adding the fw_priv to pending_fw_head. device_add() publishes the fallback loading interface, so a userspace helper which discovers the device by scanning sysfs can write 0 to the loading attribute and complete the request before it is queued as pending. In that interleaving firmware_loading_store() calls fw_state_done() while pending_list still points to itself, so it cannot remove an entry from pending_fw_head. The subsequent unconditional list_add() then queues an already-completed fw_priv. Once the request is released, pending_fw_head can retain a pointer to freed memory and the next fallback request can fault while validating the list. Only in-flight fallback requests need suspend or reboot abort handling. If the request is already DONE after device_add(), return success from the fallback path without sending another uevent, waiting again, or queueing it as pending. This preserves the invariant that pending_fw_head contains only active fallback requests.
In the Linux kernel, the following vulnerability has been resolved: md/raid5: round bitmap stripes with sector division raid5_bitmap_sector_map() aligns the array range to full RAID5 stripe widths before converting it to component sectors. That width is chunk_sectors multiplied by the number of data disks, and it is not always a power of two. Reproduce with a 4-disk RAID5, 1024-sector chunks, and three data disks. The full-stripe width is 3072 sectors. For a one-sector write at array sector 3072, correct rounding gives array range [3072, 6144), which maps to component range [1024, 2048). The old round_down()/round_up() logic instead gives [1024, 4096), which maps to [0, 1024). Use sector_div() based arithmetic so the rounded range is aligned to the actual RAID5 stripe width. The deterministic mapper test now reports the fixed component range as [1024, 2048), while the old mask-based range was [0, 1024).
In the Linux kernel, the following vulnerability has been resolved: md: wait for behind writes before destroying bitmap __md_stop() destroyed the bitmap before calling mddev_detach(). That made mddev_detach() skip bitmap_ops->wait_behind_writes(), because the bitmap was already disconnected from mddev. This was still safe for the legacy bitmap because bitmap_destroy() waits for behind writes itself. llbitmap keeps that wait in its ->wait_behind_writes() operation instead, while ->destroy() tears down the llbitmap storage. With the old ordering, RAID1 behind-write completions could still run after llbitmap storage had been freed. Call mddev_detach() before md_bitmap_destroy() so the common detach path can wait for behind writes while the bitmap is still alive. Only destroy the bitmap after those users are gone.
In the Linux kernel, the following vulnerability has been resolved: md/md-llbitmap: prevent create failure bitmap UAF llbitmap_create() publishes mddev->bitmap before reading the bitmap superblock. This is needed because llbitmap_read_sb() can initialize a new bitmap and flush it through helpers that use mddev->bitmap. If llbitmap_read_sb() fails, the old cleanup dropped bitmap_info.mutex and freed llbitmap before clearing mddev->bitmap. Readers such as /proc/mdstat rely on bitmap_info.mutex to keep the bitmap pointer stable while collecting bitmap stats, so they could observe the stale pointer after the failed create path released the mutex. Clear mddev->bitmap while still holding bitmap_info.mutex, then free the failed llbitmap after dropping the mutex. This makes mutex-protected readers see either a live bitmap or no bitmap.
In the Linux kernel, the following vulnerability has been resolved: md/md-llbitmap: stop daemon timer rearm on destroy llbitmap_destroy() deletes pending_timer before flushing md_llbitmap_io_wq. However, daemon_work can still be queued or running after the timer has been deleted, and the daemon path can arm pending_timer again when it finds dirty chunks that are not ready to flush yet. If that happens during teardown, pending_timer can remain armed after llbitmap is freed and later dereference freed memory. Add a BITMAP_SHUTDOWN bit to llbitmap->flags, set it before deleting the timer, and make the timer and daemon paths stop queueing or rearming work once teardown starts. Cancel daemon_work before flushing the shared workqueue so no already queued daemon instance can race with the free. Use timer_shutdown_sync() so a daemon instance that passed the shutdown check before teardown cannot rearm the timer afterward. BITMAP_SHUTDOWN is a runtime-only state. Mask it out when reading and updating the llbitmap superblock so the shutdown state is never loaded from disk or persisted to disk.
In the Linux kernel, the following vulnerability has been resolved: md/raid1: don't set array_frozen in raid1_takeover() raid1_takeover() sets conf->array_frozen = 1 on the newly-allocated r1conf and nothing ever clears it, so every I/O to the array stalls permanently once _wait_barrier() sees it stuck at 1. This used to be harmless: level_store() called mddev_resume() right after pers->run(), which called raid1_quiesce(mddev, 0) and cleared array_frozen back to 0 regardless of what raid1_takeover() set. Commit b39f35ebe86d ("md: don't quiesce in mddev_suspend()") removed that quiesce(mddev, 0) call, so the pre-set now sticks. setup_conf() already zero-initializes the new r1conf via kzalloc, so just don't set array_frozen here. Same class of bug as commit 892da88d1cd9 ("md/raid10: fix a 'conf->barrier' leakage in raid10_takeover()"), also triggered by b39f35ebe86d.
In the Linux kernel, the following vulnerability has been resolved: coresight: etm4x: fix underflow for usage of (nrseqstate - 1) According to IHI006H Embedded Trace Macrocell Architecture Specification[0], TRCSEQEVR<n> is implemented only when TRCIDR5.NUMSEQSTATE is 0b100, in which case n ranges from 0 to 2; otherwise, TRCIDR5.NUMSEQSTATE is 0b000. IOW, the number of usage in the initialisation or setting TRCSEQEVR<n> with drvdata->nrseqstate - 1 in the loop could make underflow issue when TRCIDR5.NUMSEQSTATE is 0b000. Therefore, introduce nr_seq_ctrls field and untie it from nrseqstate. As part of this introduce ETM_MAX_SEQ_TRANSITIONS macro and apply nr_seq_ctrls and above macro to TRCSEQEVR<n> relevant fields setup.
In the Linux kernel, the following vulnerability has been resolved: coresight: etm4x: missing cscfg_csdev_disable_active_config() in perf enable In the perf enable path, there are missing cases where cscfg_csdev_disable_active_config() is not called: - Branch broadcast is selected but not supported by the hardware - etm4_enable_hw() fails This can lead to a leak of config_desc->active_cnt. Fix this by properly calling cscfg_csdev_disable_active_config() in these error paths.
In the Linux kernel, the following vulnerability has been resolved: perf: arm_pmuv3: Zero initialize hw_id branch stack field PERF_SAMPLE_BRANCH_HW_INDEX is supported by BRBE so hw_id is passed to userspace, but it's never set by the BRBE driver. Zero initialize it as it should be according to the docs: * For the architectures whose raw branch records are * already stored in age order, the hw_idx should be 0. It's probably too risky to remove PERF_SAMPLE_BRANCH_HW_INDEX from BRBE now in case anyone is setting it and reading the value, but zero initializing the whole struct also protects against the same issue with new fields that are added in the future.
In the Linux kernel, the following vulnerability has been resolved: bpf: Reject load-acquire from pointers requiring fault protection A BPF_LOAD_ACQ is not rewritten to a BPF_PROBE_MEM load by the verifier, unlike a regular BPF_LDX, so the JIT emits a plain load with no exception table entry and a fault panics the kernel instead of being handled. Reject the source pointer types that a BPF_LDX would have had that fault protection applied to, i.e. the ones bpf_convert_ctx_accesses() turns into BPF_PROBE_MEM: a bare PTR_TO_BTF_ID, PTR_TO_BTF_ID | PTR_UNTRUSTED, PTR_TO_BTF_ID | MEM_ALLOC | PTR_UNTRUSTED and PTR_TO_MEM | MEM_RDONLY | PTR_UNTRUSTED. This is reachable e.g. by loading ->mm out of a trusted task_struct yields an untrusted pointer to mm_struct, and it is NULL for a kernel thread: [...] SEC("tp_btf/sched_switch") int BPF_PROG(demo, bool preempt, struct task_struct *prev, struct task_struct *next) { struct mm_struct *mm = next->mm; /* untrusted */ out_ldx = (__u64)mm->pgd; /* BPF_LDX */ out_acq = load_acquire(&mm->pgd); /* BPF_LOAD_ACQ */ return 0; } [...] Both dereference the same pointer, but only the BPF_LDX is protected (x86-64 JIT, jump targets shown prog-relative): [...] ; out_ldx = (__u64)mm->pgd; 17: movq $-10485760, %r10 1e: movq %rsi, %r11 21: addq $184, %r11 28: subq %r10, %r11 2b: movabsq $140737498841088, %r10 35: cmpq %r10, %r11 38: ja 0x3e <-- kernel addr? 3a: xorl %edi, %edi <-- no: dst = 0, skip the load 3c: jmp 0x45 3e: movq 184(%rsi), %rdi <-- yes: load + extable entry [...] ; load_acquire(&mm->pgd) 53: movq %rsi, %rdi 56: movq 184(%rdi), %rax <-- no check, no extable entry [...] Note that BPF_PROBE_MEM is not visible in a bpftool xlated dump, as bpf_insn_prepare_dump() rewrites it back to BPF_MEM. A PTR_TRUSTED pointer is deliberately not on the list. Such a load is not converted either, but it does not need to be, since the pointer is guaranteed live, so load-acquire from it stays allowed. The check is gated on BPF_LOAD_ACQ so that atomic RMW and store-release error messages are unchanged; writes (RMW / store-release) to such pointers are already rejected elsewhere, so only load-acquire needs this.
In the Linux kernel, the following vulnerability has been resolved: scsi: sd: Fix special_vec mempool leak when scsi_alloc_sgtables() fails sd_set_special_bvec() allocates a special payload page for UNMAP and WRITE SAME commands. If scsi_alloc_sgtables() fails afterward in sd_setup_unmap_cmnd() or sd_setup_write_same{10,16}_cmnd(), the SCSI midlayer does not call uninit_command() because RQF_DONTPREP is not set yet, leaking the page. Call sd_uninit_command() on error, and clear RQF_SPECIAL_PAYLOAD after freeing the page.
In the Linux kernel, the following vulnerability has been resolved: btrfs: zoned: don't force read-only on transient -EAGAIN from reloc merge On a zoned FS, btrfs_delayed_refs_rsv_refill() returns -EAGAIN whenever the over-committed metadata plus the zone_unusable bytes exceeds the usable size in a metadata block-group to avoid heavy over-commit of metadata and early ENOSPC in one transaction. If this happens while doing reclaim, the transaction is getting aborted. Treat -EAGAIN as a soft, retryable condition in case of block-group reclaim.
In the Linux kernel, the following vulnerability has been resolved: btrfs: defrag: fix deadlock between defrag and delalloc space reservation While running fsstress with autodefrag and flushoncommit, hit a deadlock due to the fact that defrag reserves delalloc space while it's holding dirty and locked folios, besides the extent range lock. The stack traces are the following: [958.624] task:kworker/u50:3 state:D stack:0 pid:20365 tgid:20365 ppid:2 task_flags:0x4208060 flags:0x00080000 [958.626] Workqueue: events_unbound btrfs_async_reclaim_metadata_space [btrfs] [958.627] Call Trace: [958.628] <TASK> [958.628] __schedule+0x4be/0x10f0 [958.629] ? preempt_count_add+0x69/0xa0 [958.630] schedule+0x26/0xd0 [958.631] wait_current_trans+0x102/0x160 [btrfs] [958.632] ? __pfx_autoremove_wake_function+0x10/0x10 [958.633] start_transaction+0x374/0x900 [btrfs] [958.634] btrfs_commit_current_transaction+0x1d/0x70 [btrfs] [958.635] flush_space+0xca/0x5e0 [btrfs] [958.636] ? _raw_spin_unlock+0x15/0x30 [958.637] ? btrfs_reduce_alloc_profile+0x8c/0x190 [btrfs] [958.639] ? _raw_spin_unlock+0x15/0x30 [958.640] ? calc_available_free_space.isra.0+0x6f/0x110 [btrfs] [958.641] do_async_reclaim_metadata_space+0x84/0x190 [btrfs] [958.642] btrfs_async_reclaim_metadata_space+0x64/0x80 [btrfs] [958.644] process_one_work+0x19d/0x3a0 [958.644] worker_thread+0x1c4/0x330 [958.645] ? __pfx_worker_thread+0x10/0x10 [958.646] kthread+0xfc/0x130 [958.647] ? __pfx_kthread+0x10/0x10 [958.648] ret_from_fork+0x1f7/0x2c0 [958.648] ? __pfx_kthread+0x10/0x10 [958.649] ret_from_fork_asm+0x1a/0x30 [958.650] </TASK> [958.651] task:kworker/u49:7 state:D stack:0 pid:52990 tgid:52990 ppid:2 task_flags:0x4208060 flags:0x00080000 [958.653] Workqueue: writeback wb_workfn (flush-btrfs-334) [958.655] Call Trace: [958.655] <TASK> [958.656] __schedule+0x4be/0x10f0 [958.657] ? __blk_flush_plug+0xe9/0x140 [958.658] schedule+0x26/0xd0 [958.658] io_schedule+0x42/0x70 [958.659] folio_wait_bit_common+0x12b/0x330 [958.660] ? folio_wait_bit_common+0x100/0x330 [958.662] ? __pfx_wake_page_function+0x10/0x10 [958.663] extent_write_cache_pages+0x599/0x830 [btrfs] [958.664] ? acpi_fwnode_get_reference_args+0x1fa/0x270 [958.665] btrfs_writepages+0x77/0x130 [btrfs] [958.666] ? __pfx_end_bbio_data_write+0x10/0x10 [btrfs] [958.667] do_writepages+0xc6/0x160 [958.668] __writeback_single_inode+0x42/0x310 [958.669] writeback_sb_inodes+0x231/0x570 [958.670] wb_writeback+0x8a/0x340 [958.671] wb_workfn+0xbf/0x450 [958.672] ? finish_task_switch.isra.0+0xc1/0x350 [958.673] process_one_work+0x19d/0x3a0 [958.673] worker_thread+0x1c4/0x330 [958.674] ? __pfx_worker_thread+0x10/0x10 [958.675] kthread+0xfc/0x130 [958.676] ? __pfx_kthread+0x10/0x10 [958.676] ret_from_fork+0x1f7/0x2c0 [958.677] ? __pfx_kthread+0x10/0x10 [958.678] ret_from_fork_asm+0x1a/0x30 [958.679] </TASK> [958.679] task:btrfs-cleaner state:D stack:0 pid:296750 tgid:296750 ppid:2 task_flags:0x208040 flags:0x00080000 [958.681] Call Trace: [958.682] <TASK> [958.682] __schedule+0x4be/0x10f0 [958.683] schedule+0x26/0xd0 [958.684] handle_reserve_ticket+0x1b9/0x2c0 [btrfs] [958.685] ? __pfx_autoremove_wake_function+0x10/0x10 [958.686] reserve_bytes+0x283/0x4c0 [btrfs] [958.687] btrfs_reserve_metadata_bytes+0x18/0xb0 [btrfs] [958.688] btrfs_delalloc_reserve_metadata+0x121/0x320 [btrfs] [958.690] btrfs_delalloc_reserve_space+0x46/0xb0 [btrfs] [958.691] btrfs_defrag_file+0x903/0x1110 [btrfs] [958.692] btrfs_run_defrag_inodes+0x334/0x430 [btrfs] [958.694] cleaner_kthread+0x97/0x1c0 [btrfs] [958.694] ? __pfx_cleaner_kthread+0x10/0x10 [btrfs] [958.696] kthread+0xfc/0x130 [958.696] ? __pfx_kthread+0x10/0x10 [958.697] ret_ ---truncated---
In the Linux kernel, the following vulnerability has been resolved: btrfs: check if root is readonly when setting posix acl For a filesystem which has btrfs read-only property set to true, all write operations including acl and xattr should be denied. However, acl can still be set even if btrfs ro property is true. This happens because no function on the set_acl code path checks the root is readonly or not. It was checked in btrfs_setxattr_trans() but got removed in commit 353c2ea735e4 ("btrfs: remove redundant readonly root check in btrfs_setxattr_trans") That commit didn't check if all the callers properly check the root's read-only flag. A previous fix is commit b51111271b03 ("btrfs: check if root is readonly while setting security xattr"). Always check if the root is read-only before performing the set acl operation.
In the Linux kernel, the following vulnerability has been resolved: btrfs: retry verity reads for not-uptodate Merkle folios btrfs_read_merkle_tree_page() can find a folio in the mapping that is not uptodate. After taking the folio lock, the current code treats that state as a read error and returns -EIO. That can make a previous transient read failure sticky. If the failed read left a not-uptodate folio in the mapping, later callers find that folio and fail instead of retrying the read. Keep the existing page-cache insertion and locking order, but retry the Merkle item read when a not-uptodate folio is found in the mapping. Also unlock the folio when read_key_bytes() fails so that a later caller can lock it and retry the read.
In the Linux kernel, the following vulnerability has been resolved: btrfs: zoned: flush active metadata block group at btree_writepages() start btree_writepages() writes the btree inode's dirty metadata in ascending logical address order. On a zoned filesystem only one metadata and one system block group is active for writing at a time, and check_bg_is_active() (via btrfs_check_meta_write_pointer()) pivots the active block group as writeback moves from one block group to the next. If the active block group sits at a higher logical address than another block group that also holds dirty metadata, the ascending walk reaches the lower one first and, to write it, has to finish the active block group and activate the lower one. It cannot finish a block group that still has unsent IO, and during WB_SYNC_ALL && !for_sync (commit) writeback it deliberately refuses to wait for that IO under fs_info->zoned_meta_io_lock, as that can deadlock. The pivot thus cannot issue the submission itself either, so it gives up: btrfs_check_meta_write_pointer() returns -EAGAIN, which btrfs_write_and_wait_transaction() treats as fatal and aborts the transaction, forcing the filesystem read-only. This happens intermittently under metadata-heavy relocation (e.g. fstests btrfs/187). Flush the active metadata and system block groups at the start of btree_writepages(), under the fs_info->zoned_meta_io_lock it already holds, so they have no unsent IO left and the later pivot can finish them and make forward progress.
In the Linux kernel, the following vulnerability has been resolved: btrfs: qgroup: fix a wrong length calculation in qgroup_free_reserved_data() In that function, we round down the start position and round up the ending position. But during the calculation of @len, we use "round_up(start + len, sectorsize)", which is the rounded up end position, not the rounded up length. Which results a much larger length, and later we are still using "start + len", which is completely incorrect. Fix it by declaring a local @aligned_start and @aligned_len and use them instead.
In the Linux kernel, the following vulnerability has been resolved: pinctrl: airoha: add missed IRQ resource helpers Without hooking .irq_request_resources, gpiolib cannot set GPIOD_FLAG_USED_AS_IRQ. This breaks pin direction locking and can allow userspace or another driver to reconfigure an active IRQ pin as an output
In the Linux kernel, the following vulnerability has been resolved: Bluetooth: virtio_bt: avoid OOB read of build info string The virtbt_setup_zephyr() sends the Zephyr vendor command 0xfc08 (Read Build Information) and hands the response to bt_dev_info() and hci_set_fw_info() as a "%s" string starting at skb->data + 1, without checking the length. A backend that answers with status only leaves that pointer past the end of the received data, so the walk reads adjacent slab memory until it meets a NUL. Those bytes reach the kernel log and the firmware-info debugfs file. To fix this, print the string with a bounded "%.*s" limited to skb->len - 1. A short or unterminated response then prints as much as arrived instead of failing setup. This mirrors commit dd068ef04412 ("Bluetooth: bpa10x: avoid OOB read of revision string in bpa10x_setup()"), which fixed the identical pattern.
In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hci_sync: free the advertising instance on the failure and cancel paths adv_timeout_expire() hands a kmalloc()ed instance byte to hci_cmd_sync_queue() with a NULL destroy callback, and only adv_timeout_expire_sync() frees it. That leaks on two paths: - the return value is not checked, and hci_cmd_sync_queue() does not take ownership when it fails (-ENETDOWN, -ENODEV, -ENOMEM); - a cancelled entry is not released, as _hci_cmd_sync_cancel_entry() does not free entry->data when there is no destroy callback. hci_cmd_sync_clear() cancels every pending entry when the controller is unregistered. Free the buffer from a destroy callback, and in the caller when the entry could not be queued at all.
In the Linux kernel, the following vulnerability has been resolved: Bluetooth: MGMT: free the mesh send cancel command when it is cancelled mesh_send_cancel() queues the pending command with a NULL destroy callback, so it is only freed if send_cancel() runs. A cancelled entry is leaked, as _hci_cmd_sync_cancel_entry() does not release entry->data when there is no destroy callback, and hci_cmd_sync_clear() cancels every pending entry when the controller is unregistered. Nothing else reclaims it either: mgmt_pending_new() does not put the command on hdev->mgmt_pending. The leak also pins the socket reference taken by mgmt_pending_new(), so the mgmt socket is never released. Free the command from a destroy callback.
In the Linux kernel, the following vulnerability has been resolved: Bluetooth: MGMT: free the HCI command when it is cancelled mgmt_hci_cmd_sync() queues the pending command with a NULL destroy callback, so it is only freed if send_hci_cmd_sync() runs. A cancelled entry is leaked, as _hci_cmd_sync_cancel_entry() does not release entry->data when there is no destroy callback, and hci_cmd_sync_clear() cancels every pending entry when the controller is unregistered. Nothing else reclaims it either: mgmt_pending_new() does not put the command on hdev->mgmt_pending. The leak also pins the socket reference taken by mgmt_pending_new(), so the mgmt socket is never released. Free the command from a destroy callback. The now-empty done label is replaced by a direct return.
In the Linux kernel, the following vulnerability has been resolved: bpf, cgroup: Fix storage null-ptr-deref after replacing prog Syzkaller reported a storage null-ptr-deref issue after replacing prog. This occurs in the following scenario: 1. prog A, an empty prog, is attached to a cgrp. 2. prog B uses BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE and calls the bpf_get_local_storage helper. 3. link_update is called to replace prog A with prog B. The reason is that __cgroup_bpf_replace fails to alloc and assign the required cgrp storage for the incoming replacement prog. Consequently, the new prog inherits an uninit storage, leading to null-ptr-deref panic when kick the new prog. Fix this by rejecting a link update if new_prog's cgroup storage is incompatible with link->prog.
In the Linux kernel, the following vulnerability has been resolved: iio: light: gp2ap002: Fix unbalanced runtime PM on repeated event writes The IIO core does not filter duplicate writes to the event enable attribute, so writing the same value twice invokes write_event_config() twice. Enabling twice leaks a runtime PM reference, preventing the device from ever suspending again; disabling twice underflows the usage count and triggers a "Runtime PM usage count underflow" warning. Bail out early when the requested state matches the current state. While at it, switch to pm_runtime_resume_and_get() so a failed resume is propagated to userspace instead of silently marking the event enabled.
In the Linux kernel, the following vulnerability has been resolved: net/sched: cls_api: fix teardown of an adopted proto on insert-race loss In tc_new_tfilter() the create branch sets tp_created = 1 before calling tcf_chain_tp_insert_unique(). When the caller loses the race (another request inserted a proto at the same chain/prio first), insert_unique() destroys the caller's own tp_new and returns the winner's proto with an extra reference. tp_created was never cleared, so the loser's errout path treated the winner's live proto as its own and called tcf_chain_tp_delete_empty() on it, silently unlinking an active classifier that the winning request already advertised via RTM_NEWTFILTER. Track the outcome of the insert step in a single tri-state variable so each errout path reacts correctly: - TP_NOT_CREATED: no proto created; pursue the old path. - TP_CREATED: proto inserted successfully; same code path as before. - TP_NOT_OWNED: New - lost the insert race; tp is another request's proto (chain ref already released by tp_new's destroy) Both errout reactions are single expressions derived from the state. This fix is motivated by the Sashiko's automated review of Patch (net/sched: cls_api: Always acquire rtnl_lock when destroying locked classifiers) [1][2]. The review identified the silent-unlink behaviour of an adopted proto's teardown when a request loses the tcf_chain_tp_insert_unique() race. [1] https://sashiko.dev/#/patchset/20260801125632.360365-1-jhs%40mojatatu.com [2] https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20260801125632.360365-1-jhs%40mojatatu.com
In the Linux kernel, the following vulnerability has been resolved: bpf: Fix mmap_lock leak in irq_work path stack_map_get_build_id_offset() introduced a per-CPU irq_work to defer mmap_read_unlock() from NMI context, and bpf_find_vma() later reused the same mmap_unlock_work. Both callers only check whether the work is busy before taking mmap_lock, so a nested caller can reuse the slot before the first caller queues it. Two read locks may then be acquired while only one deferred unlock runs, leaking a read lock and blocking exit_mmap(). Reserve the per-CPU slot before mmap_read_trylock(). Use the same wrapper in stackmap and bpf_find_vma() so both callers release the reservation on trylock failure. Keep rejecting the slot while the irq_work remains busy. Release it after the irq_work callback unlocks the mm.
In the Linux kernel, the following vulnerability has been resolved: fbdev: kyro: Validate overlay viewport coordinates The overlay viewport end coordinates are computed from the viewport origin and dimensions using 32-bit unsigned arithmetic. Large input values can cause these calculations to wrap around before the resulting coordinates are passed to SetOverlayViewPort(). SetOverlayViewPort() packs the viewport coordinates into 16-bit register fields. The X coordinates are additionally adjusted by +2 and +1 before being written. Validate the coordinate calculations for 32-bit wraparound and ensure that the adjusted coordinates fit within their 16-bit register fields before calling SetOverlayViewPort(). Found by Linux Verification Center (linuxtesting.org) with SVACE.
In the Linux kernel, the following vulnerability has been resolved: iommu/vt-d: Fix iopf_refcount leak on RID domain replacement intel_iommu_attach_device() enables IOPF for the new domain but never disables it for the old one. device_block_translation(), called at the start of the function, tears down translation but does not touch any IOPF state; blocking_domain_attach_dev() has to call iopf_for_domain_remove() explicitly before invoking it for exactly this reason. identity_domain_attach_dev() has the same problem. Its comment claims that no PRI handling is needed because the device has been put in the blocking state, but the blocking state and the IOPF reference count are independent of each other. As a result, replacing a domain that has an iopf_handler with another domain at RID level leaks a reference in info->iopf_refcount. The count never drops back to zero, so iopf_queue_remove_device() is never called and iommu_disable_pci_pri() triggers its WARN_ON(info->iopf_refcount) when the device is released. The PASID paths already handle this correctly by way of iopf_for_domain_replace(); convert the two RID paths to do the same. Using the replace helper rather than a bare remove keeps the enable before the disable, so the reference count does not transiently reach zero and evict the device from the IOPF queue.
In the Linux kernel, the following vulnerability has been resolved: media: amd: isp4: release partial allocations in isp4if_alloc_fw_gpumem() isp4if_alloc_fw_gpumem() allocates several GPU memory pools in sequence. If one of them fails, it jumps to error_no_memory and returns -ENOMEM without releasing the pools that were already allocated, leaking them. Release the already-allocated pools before returning. isp4if_gpu_mem_free() is a no-op on pools that were not allocated, so calling isp4if_dealloc_fw_gpumem() here safely frees exactly the pools that succeeded. isp4if_gpu_mem_free() previously logged an error for a NULL entry, which is a normal case during partial-allocation cleanup, so make it silent.
In the Linux kernel, the following vulnerability has been resolved: media: amd: isp4: fix self-deadlock in isp4sd_pwron_and_init() error path isp4sd_pwron_and_init() holds ops_mutex via guard(mutex) and, on any init failure, jumps to err_deinit and calls isp4sd_pwroff_and_deinit(). That helper takes the same ops_mutex, re-acquiring a non-recursive mutex already held by the current thread, so any init failure deadlocks. Unwind the error path in stages instead, releasing only what each failure point acquired. This also avoids the issues that an unconditional teardown would hit at the earlier failures, such as a runtime-PM underflow from pm_runtime_resume_and_get() and MMIO access while the device is unpowered.
In the Linux kernel, the following vulnerability has been resolved: nvme-pci: release descriptor pools on probe failure The per-NUMA-node descriptor DMA pools are created lazily from nvme_init_hctx_common() once the admin tag set is allocated, but they are only destroyed in nvme_remove() via nvme_release_descriptor_pools(). Any probe failure after the admin tag set has been allocated unwinds through the out_disable label and nvme_pci_free_ctrl(), neither of which releases the pools, leaking the dma_pool objects. Release the descriptor pools in the out_disable error path. It must not be added to nvme_pci_free_ctrl(), as that would double-free against nvme_remove() on the normal teardown path.
In the Linux kernel, the following vulnerability has been resolved: amt: Don't support cross-netns setup. When a lower device is unregistered, amt_device_event() tries to unregister its upper AMT device, but it has two problems. 1. amt_lookup_upper_dev() looks up an upper device in the lower device's netns only 2. amt_device_event() unregisters a single upper device only If AMT device is created on a lower device in another netns, removing the lower device triggers the splat below and gets stuck until all upper devices are removed. [0] The cross-netns setup seems unintentional considering 1. and the following points: * amt_link_setup() sets dev->netns_immutable to true * skb_scrub_packet() is not called in the fast path * iproute2 binary fails to find cross-netns lower device via link-netns: # ip -n ns1 link add amt0 link-netns ns2 type amt dev veth1 Cannot find device "veth1" Instead of supporting it properly and preparing for per-netns netdev unreg, let's forbid cross-netns setup. Note that the problem 2. needs a separate fix. [0]: WARNING: net/core/dev.c:12518 at unregister_netdevice_many_notify+0x1cce/0x2250, CPU#48: ip/2031 Modules linked in: CPU: 48 UID: 0 PID: 2031 Comm: ip Not tainted 7.2.0-rc5+ #27 PREEMPT(full) Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.17.0-debian-1.17.0-1 04/01/2014 RIP: 0010:unregister_netdevice_many_notify (net/core/dev.c:12518) Code: 89 ef e8 d5 52 ae fe e9 d0 f4 ff ff 48 8d 3d f9 3b 9c 02 48 c7 c6 c0 0b 63 84 ba ab 1f 00 00 67 48 0f b9 3a e9 65 ff ff ff 90 <0f> 0b 90 eb 81 48 8d 3d f6 3b 9c 02 48 c7 c6 c0 0b 63 84 ba e2 1f RSP: 0018:ffffc90004abf160 EFLAGS: 00010212 RAX: ffff888104d38260 RBX: ffff88800b0911b8 RCX: dffffc0000000000 RDX: 0000000000000000 RSI: 0000000000000008 RDI: ffffffff85b9f880 RBP: ffffc90004abf2d0 R08: ffffffff85b9f887 R09: 1ffffffff0b73f10 R10: dffffc0000000000 R11: fffffbfff0b73f11 R12: ffff88800b091d08 R13: ffff88800b091178 R14: dffffc0000000000 R15: ffff88800b091000 FS: 00007f555b86c600(0000) GS:ffff8881942a0000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000562107d489c0 CR3: 0000000109a40002 CR4: 0000000000372ef0 Call Trace: <TASK> rtnl_dellink (net/core/rtnetlink.c:3632 net/core/rtnetlink.c:3674) rtnetlink_rcv_msg (net/core/rtnetlink.c:7112) netlink_rcv_skb (net/netlink/af_netlink.c:2556) netlink_unicast (net/netlink/af_netlink.c:1319) netlink_sendmsg (net/netlink/af_netlink.c:1900) ____sys_sendmsg (net/socket.c:775) __sys_sendmsg (net/socket.c:2738) do_syscall_64 (arch/x86/entry/syscall_64.c:63) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:121) ... unregister_netdevice: waiting for veth0 to become free. Usage count = 7 ref_tracker: netdev@ffff88800d7496d8 has 3/3 users at __netdev_adjacent_dev_insert (./include/linux/netdevice.h:4525 ./include/linux/netdevice.h:4554 net/core/dev.c:8791) __netdev_upper_dev_link (net/core/dev.c:8879 net/core/dev.c:8963) netdev_upper_dev_link (net/core/dev.c:9009) amt_newlink (drivers/net/amt.c:3321)
In the Linux kernel, the following vulnerability has been resolved: nfc: pn533: hold a reference to the request skb during send_frame __pn533_send_async() publishes the command and then calls dev->phy_ops->send_frame(). Once dev->cmd is set, an incoming frame can be matched to this command: the I2C threaded IRQ runs pn533_recv_frame(), which queues cmd_complete_work, and pn533_send_async_complete() frees cmd->req with consume_skb(). On the I2C transport, pn533_i2c_send_frame() still dereferences the same skb after i2c_master_send() returns, so a completion that races the send can free the skb while the transport is still using it. The request skb is owned by the command object and may be freed by command completion at any time after dev->cmd is published, so the transport send path must not assume it stays alive. Hold a temporary reference to the request skb across the send_frame() call so the transport always sees a live skb even if completion races the send. Add a pn533_send_cmd_frame() helper and use it from all three send paths.
In the Linux kernel, the following vulnerability has been resolved: ALSA: seq: Don't leak the extension cell pointer in the bounce payload The bounce_error_event() embeds the failed event in the bounce payload by pointing data.ext.ptr at it. When that event is a queued variable-length event, its own data.ext.ptr holds the address of its first extension cell, put there by snd_seq_event_dup(). The payload goes out verbatim through snd_seq_expand_var_event(), so the address reaches userspace. That is the same address commit 705dd6dcbc0e ("ALSA: seq: Clear variable event pointer on read") removed from the event header. The read path still clears it there, just above the call that expands the payload. Embed a sanitised copy instead, treated exactly as snd_seq_read() treats the header. A stack copy is enough because delivery is synchronous and snd_seq_event_dup() copies before returning. An unprivileged client reaches this by setting SNDRV_SEQ_FILTER_BOUNCE, queueing a variable-length event to a port that does not exist and reading the bounce back. Eight bytes on 64-bit, from its own pool.
In the Linux kernel, the following vulnerability has been resolved: RDMA/cxgb4: Free debugfs on registration failure c4iw_alloc() creates the per-device debugfs tree (dev->debugfs_root via setup_debugfs()), but it is removed only in c4iw_remove(), not in c4iw_dealloc(). When RDMA device registration fails, the registration worker's err_dealloc_ctx path calls c4iw_dealloc() directly, bypassing c4iw_remove(), so the debugfs dentries leak and outlive the freed c4iw_dev. Move debugfs_remove_recursive() into c4iw_dealloc() so every path that frees ctx->dev also removes its debugfs tree.
In the Linux kernel, the following vulnerability has been resolved: ubi: Fix rollback for explicit UBI device numbers ubi_init_attach() rolls back module initialization failures by scanning ubi_devices[0..i-1], where i is the mtd= parameter index. That assumes the parameter index matches the UBI device number. That assumption is not true when mtd= specifies an explicit ubi_num. A successfully attached device can be stored at a higher ubi_devices[] slot, and a later failure can miss it during rollback. Scan the full ubi_devices[] array and detach by the actual array index, matching the way UBI devices are stored.
In the Linux kernel, the following vulnerability has been resolved: mtd: ubi: Release device reference on busy detach ubi_detach_mtd_dev() obtains a device reference through ubi_get_device() before checking whether the UBI device is busy. The busy return path drops ubi->ref_count but leaves the device reference held, so the device object cannot be released after a later detach. Drop the device reference before returning -EBUSY.
In the Linux kernel, the following vulnerability has been resolved: ASoC: xilinx: formatter_pcm: fix stream_data leak on open error In xlnx_formatter_pcm_open(), stream_data is allocated and adata->play_stream or adata->capture_stream is assigned early. If a later step, such as snd_pcm_hw_constraint_step() or snd_pcm_hw_constraint_integer(), fails, the function returns the error immediately. ALSA does not call the close callback when open fails, so stream_data is leaked and the stream pointer is left dangling, pointing to a substream that ALSA frees. A later interrupt would then call snd_pcm_period_elapsed() on the freed substream. Free stream_data and clear the stream pointer on the error paths.
In the Linux kernel, the following vulnerability has been resolved: firewire: core: fix memory leak in error path of build_tree() In the error path of build_tree(), node instances can remain in the local linked list when the function returns. Whenever an invalid value is detected in the self ID sequence, each allocated node instance is either an entry in the linked list or an entry in the ports array of its parent node. Therefore, the allocate node instances can be safely released by traversing the linked list from its head. Release the remaining node instances with for_each_fw_node() before returning to the caller.
In the Linux kernel, the following vulnerability has been resolved: arm64/efi: Avoid voluntary preemption with efi_mm installed Gus reports a bad kernel memory access when using software PAN (CONFIG_ARM64_SW_TTBR0_PAN=y) on a machine with support for EFI runtime services: Unable to handle kernel access to user memory outside uaccess routines at virtual address 00000000f322ff30 Mem abort info: ESR = 0x0000000096000004 FSC = 0x04: level 0 translation fault Internal error: Oops: 0000000096000004 [#1] SMP Workqueue: efi_rts_wq efi_call_rts pstate: 80400005 (Nzcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : efi_call_rts+0xd8/0x288 Call trace: efi_call_rts+0xd8/0x288 (P) process_one_work+0x178/0x4f8 worker_thread+0x194/0x328 This is because the fpsimd context management code called from __efi_fpsimd_begin() can preempt voluntarily, returning later to the EFI code with an incorrect value for TTBR0_EL1 thanks to the deferred mm switching used by the software PAN implementation. Since EFI runtime services cannot preempt voluntarily and because the fpsimd switching code does not rely on the TTBR0_EL1 mappings, simply reorder the fpsimd switch so that it occurs before we change the page-table.
In the Linux kernel, the following vulnerability has been resolved: bpf, s390: Clear fetch destination on faulting arena atomic Same missing register clear as on riscv64. A RMW atomic on an arena pointer is converted to BPF_PROBE_ATOMIC and gets an exception table entry, but bpf_jit_probe_atomic_pre() only fills in the arena base and the probe offset, leaving probe->reg at the -1 that bpf_jit_probe_init() set, which bpf_jit_probe_post() writes into the entry and ex_handler_bpf() then reads back as "there is nothing to clear". That is right for a plain BPF_{ADD,AND,OR,XOR}, which only writes memory, but an RMW carrying BPF_FETCH also reads the old value into a register: src_reg for BPF_{ADD,AND,OR,XOR} | BPF_FETCH and BPF_XCHG, and r0 for BPF_CMPXCHG. So on a fault over an unmapped arena page the program resumes at the landing pad with whatever that register held before the atomic instead of the 0 that every other BPF_PROBE_* access delivers. Fill probe->reg in from bpf_atomic_load_reg(). Unlike x86-64 and arm64, s390x does not report arena violations from its exception handler, so there is no access direction to correct here, only the missing register clear.
In the Linux kernel, the following vulnerability has been resolved: s390/debug: Fix deadlock during unregister Unregistering an s390dbf debug area while one of the associated debugfs files is being written to can cause a deadlock: $ echo >.../vmur/level $ rmmod vmur =================================================== debugfs write debugfs_file_get() debug_unregister() mutex_lock(debug_mutex) debugfs_remove() wait for debugfs_file_put() debug_file_ops.write() debug_input() mutex_lock(debug_mutex) ==> DEADLOCK Fix this by splitting debug_unregister() into an s390dbf and debugfs part, and running only the s390dbf part with debug_mutex locked.
In the Linux kernel, the following vulnerability has been resolved: clocksource/drivers/samsung_pwm: Switch to raw_spinlock_t type Samsung PWM timer might be used as a clock source on some legacy systems. When PREEMPT_RT is enabled on ARM, regular spinlock is converted to a sleeping lock (mutex-based), which must not be used in atomic context such as hard interrupt handlers. Switch the samsung_pwm_lock to the raw_spinlock, which remains a true non-sleeping spinlock even under PREEMPT_RT.
In the Linux kernel, the following vulnerability has been resolved: net: page_pool: fix UAF in __page_pool_release_netmem_dma on xa_cmpxchg race This bug was discovered while testing the hns3 driver under channel reconfiguration (`ethtool -L` / `ethtool -G`) with iperf3 traffic on arm64. The race is intermittently triggered when page_pool_destroy() runs page_pool_scrub() concurrently with page return via page_pool_put_netmem() on a different CPU. A WARN in page_pool_clear_pp_info() surfaced the dangling DMA index bits left by the cmpxchg loser, which led to the investigation. page_pool_scrub() iterates pool->dma_mapped via xa_for_each() with no page ref held. __page_pool_release_netmem_dma() currently reads and writes netmem fields (dma_addr, DMA index bits in pp_magic) after xa_cmpxchg() returns. The unref path calls put_page() unconditionally regardless of the cmpxchg outcome; when it loses the cmpxchg, it still frees the page before the scrub winner finishes these netmem accesses, so scrub touches a freed page -- a Use-After-Free. Fix this by splitting the DMA release into two functions: 1. __page_pool_unmap_netmem_dma() caches dma_addr before xa_cmpxchg(), does the cmpxchg to remove the DMA mapping, and calls dma_unmap on the cached address. It never touches netmem fields after the cmpxchg, making it safe for the scrub path which holds no page ref. 2. __page_pool_release_netmem_dma() wraps the above and additionally clears dma_addr and DMA index bits in netmem fields. This is safe only when the caller holds a page ref, so it is used by the return path (page_pool_return_netmem). The scrub path calls __page_pool_unmap_netmem_dma() directly; the return path calls __page_pool_release_netmem_dma().